The comparative anatomy of the nervous system of vertebrates including man

By

C. U. Ariens Kappers, M.D., Sc.D., Ll.D.

Director of The Central Institute Of Brain Research, Amsterdam, And Professor Op Comparative Neurology In The University Op Amsterdam

G. Carl Huber, M.D., Sc.D.

Late Dean Of The Qhaddate School, Director Op The Anatomical Laboratories, And Professor Of Anatomy In The University Op Michigan

And

Elizabeth Caroline Crosby, Ph.D.

Associate Professor Of Anatomy In The University Op Michigan

Volume One

Iiafner Publishing Company NEW YORK 19 6 7

Originally published 1936 in 2 volumes

Printed and published by HAFNER PUBLISHING COMPANY, INC. 31 East 10th Street New York, N.Y. 10003

Library of Congress Catalog Card Number 60-6766

Lithographed in the U.S.A. by Noblb Offset Peintfes, Inc. New York, N. Y. 10003

DEDICATED

TO

CHAELES JUDSON HERRICK,

GREAT PIONEER IN AMERICAN COMPARATIVE NEUROLOGT

Ann Arbor, October 4, 1959

PREFACE TO THE 1960 EDITION

Because of a considerable demand for copies of this reference book, consent has been given for its reproduction wdth the contents as originally published in 1936. In view of tlie recent rapid growth of our knowledge of the central nervous system, any book approximately a quarter of a centur)'^ old which deals widi the anatomy of the brain and the spinal cord is necessarily out of date. However, the early work does form a basis for later contributions. It is hoped, then, that the republication may be useful.

Elizabeth C. Crosby, Ph.D.

University of Michigan Medical Center

PREFACE

“ I do the very best I know how; the very best I can ; and I mean to keep doing so until the end. If the end brings me out all right, what is said against me won’t amount to anything. If the end brings me out wrong, ten angels swearing I was right would make no difference.”

Abraham Lincoln

The quotation given above hung over the desk of Dean G. Carl Huber, and was to him undoubtedly an inspiration for that intellectual independence which was so characteristic of him and which was so inspiring to those who collaborated with him in this book as well as in other fields. Without his wide range of knowledge in the sphere of neurological research, accumulated over many years from his early work on nerve repair in 1892 to the time of his death, and his great editorial skill, the result of many years of editorship of scientific journals and articles, it would have been quite impossible to prepare for publication and to have accepted, under present conditions, a text so technical and so large as is the present book. Through the years of preparation of this book it was his enthusiasm, his industry, his unfailing optimism, which carried the task on to completion. It is a matter of the greatest regret to both his collaborators and to the publishers that, while he saw the manuscript completed, he should not have lived to see, in published form, this work, which is so largely a product of his scholarship and inspiration.

This book is an outgrowth of “Die vergleichende Anatomie des Nervensystems,” which was written by C. U. Ariens Kappers in 1920-1921. When, in 1926, the present text was begun, the original plan on the part of both Dutch and American collaborators was to translate the German text into English and to make such revision as should be deemed necessary in the fight of later literature. However, so great additions had been made to the knowledge of the comparative anatomy of the nervous system during the years following its publication that it soon became evident that, in order to do justice to the situation, much of the text must be entirely rewritten, and additions made to the figures. The idea, then, of presenting a translation was abandoned. Thus, with the exception of certain portions of the text, such, for example, as the introduction and the pages on gyri and sulci, the present book offers a new presentation of the material available in comparative neurology, based on the available literature and on results of the research programs of the Institute of Brain Research at Amsterdam and the Laboratory of Comparative Neurology at the University of Michigan, both of which have collaborated in the preparation of the present text.

This collaboration has been so complete that it would be difficult to assign exclusive credit to either laboratory for any special portion. Mention should be made, however, of certain fields which have received attention from one laboratory or the other.

Foremost among the contributions from the Institute of Brain Research at .\msterdam are the framework of plan and material provided by the German edition and the thorough revision of the presentation of the theory of neurobiotaxis, written by the Dutch collaborator. Certain facts and figures with regard to the newer work on efferent cranial nerve nuclei and their roots were made available through the work of Dr. J. L. Addens of the Institute. The authors wsh here to express their thanks to Dr. Addens for his kindness in permitting the use of his figures and for suggestions with regard to this portion of the text. Some new figures have been added and cliches provided for the figures used from the German text by the laboratory at Amsterdam.

The chapters on diencephalon and telencephalon, particularly those portions dealing with these regions in reptiles, birds, and mammals, have received particular attention from the Laboratory of Comparative Neurology of the Anatomy Department at Michigan. The material presented concerning these regions includes the results of published and, to some extent, unpublished work along these lines, derived from the research programs of the American authors and of others associated in this laboratory, now or formerly, as staff or students. .'Vlso, from the work of staff members at Michigan, additions have been made to the accounts of the vestibular centers of reptiles, to trigeminal and facial centers of various vertebrates, and to the spinal cord of birds. The sources of any such information used, whether published or unpublished, are credited in the text, but especial mention is made here of the account of the marginal nuclei of birds, which was written by Dr. J. F. Huber. To these associates and colleagues who have so generously permitted the use of the results of their work, the authors wish to offer their most sincere thanks. Various figures, partly from published contributions but in many cases drawn especially for this text, have been contributed by the laboratory at Michigan. Other figures have been redrawn or relabeled to meet the requirements of American publishers, most of thus latter work ha\'ing been done by Dean Huber.

Very greatly appreciated permission to use figures which had appeared in WLtar publications was granted by the Wistar Institute. Through the kindne.'S of Dr. Woollard, Editor of the Journal of Anatomy, a similar permission was granted for the use of certain figures from this English journal. The authors arc also indebted to various investigators in other laboratories for the use of figures or, in a few ca-ses, of unpublished manuscript. It is not possible to mention here all of those to whom thanks arc due, but the sources of all such material u.-cd are credited in each case in the text. However, the authors wish to thank particularly Profe.'ssor C. Judson Herrick for the privilege of examining unpubli.'hed manuscript and figures and Dr. Jeannette Obenchain for certain timely sugge.^tions.

I o publi-h .O large a book under present conditions is very difficult. The rights of publication held by the Dutch firm of de Erven F. Bohn were mo.st kindly turned over by thi.s firm to the American publishers. It is quite impt. 'ible for the authors to exprc.-.s adequately their .sense of deep obligation to The Macmillan Company for the unbounded courtesy, help, and encouragement which have made possible the completion of this book under conditions which have been pecuUarly trying. No effort has been spared in any particular on the part of the representatives of this company to make this book all that it should be.

The correcting of the proof and the preparation of the subject index and author index were, of necessity, handled at the University of Michigan. In these fields the younger members of the Department offered greatly needed and greatly appreciated help. Dr. Tryphena Humphrey, Dr. Jean Weston, Dr. Russell Woodburne, and Dr. John Barnard, members of the staff of the Anatomy Department, and the secretaries. Miss Catherine Brook and Miss Elizabeth Switzer, all helped in making the prehminary drafts for the indexes. Dr. Weston, Dr. Humphrey, and Miss Switzer helped not only in typing the manuscript but in checking for error in the bibhographies and proof and in the final preparation of these indexes. Those who have attempted to edit manuscript on the scale of that presented in this book can appreciate how heartfelt the gratitude is to these younger people.

Almost ten years have passed since the book was begun, and now that the task is completed it represents only in small part the ideal that its authors were trying to attain. If, however, it in any way aids some of those who are working in this field to a better comprehension of the problems involved in the study of the nervous system and the possibilities offered by such study, the authors will feel that their labor has been well repaid.

C. U. Ariens ICappers Elizabeth C. Crosby

Introduction

The nervous system associates impulses arising within the body with those due to surface stimulations and prqvides appropriate effectory paths. It associates and integrates impulses from neuromuscular and neurotendinous endings and the semicircular canals (proprioceptive impulses of Sherrington, ’06) yet in such fashion that within certain limits the various components remain recognizable, each retaining its specific character. In similar ways it associates various stimuli from the outside world (exteroceptive impulses) and interrelates those received from various organs (interoceptive or visceral afferent impulses). Finally, it coordinates and integrates exteroceptive, proprioceptive, and interoceptive impulses into a dynamic, effective whole.

It is customary and correct to divide the constituents of the nervous system into nervous and non-nervous elements. To the former belong those elements, usually termed neurons, which serve the functions of reception (sensitivity), conduction, and integration of stimuli, all of which, but most particularly the last, give the nervous system its special significance. These various functions are characteristic of living substance in general and the protoplasm of cells, other than neurons, does not lack such properties. The protoplasm of unicellular forms has been shown to be capable not only of receiving stimuli, but also of conducting such stimuli to remote portions of the cell body and there instigating movements. In these unicellular animals, different impulses occurring at the same time may reenforce or inhibit each other. Also, in such simple multicellular forms as the sponges, which as yet have not developed a nervous system, conductivity is evident, in addition to the sensitivity of the body cells. This does not confine itself to the body of the stimulated cell but is intercellular, passing from one cell to another over protoplasmic intercellular bridges which make possible, not only the diffusion of single impulses but the correlation of various impulses occurring at the same time. Such intercellular conductivity, by means of intercellular protoplasmic bridges instead of through a nervous system, has been found in certain amphibian embryos (Schaper, ’98 ; Goldstein, ’04; Wintrebert, ’04) and is present also in smooth muscle tissue. It is only the enhancement of these previously mentioned functions which characterizes a neuron.

Another general characteristic of living substance {Hering, ’70; Butler, ’80; Laycock, ’75; Semon, ’ll), which is particularly outstanding in relation to the nervous system, is an internal power for setting up stimuli (engram formation). This is the ability to continue for a longer time than the impulse itself persists, the conditions produced by it. That such an ability is more characteristic of neurons than of other cells remains to be substantiated ; indeed the facts suggest the contrary to be true. This inherent power for the retention of the results of stimulation is merely more evident in the nerve cells because in them these results are frequently a part of conscious processes. This inherent ability for retention is the organic basis for memory (better ekphoria), becoming functional through the conduction and correlation of impulses. Under the term memory is implied that attribute by which sensations which are latent come to expression again through impulses which arouse the earlier sensation, either directly through new perceptions or indirectly through associations with previous sensations. This ability to remember is more particularly characteristic of the nervous system than of other parts of the body and its ekphoria has the further characteristic that it may disappear from consciousness without losing the engram.

Not only memory, but also attention and association, so characteristic of nervous functions, may be considered as characteristic of protoplasm as well. Attention is the tendency to concentrate on one function, a process evident in the specific development of other than nervous tissues, while association or correlation is observed in the functional relations of tissues such as tendon, bone, or muscle, or of organs and organ systems such as the gastrointestinal and respiratory systems.

In the nervous system, as in all living substance, there is an active striving, an inherent tendency, to supplement the activities and to elaborate the various impulses, which is recognizable chiefly through the end results, but which may be felt consciously as an indeterminate realization of the force of thought and will. The results of this entelechic* tendency (see p. 5a) are seen in the progressive development of the brain in accordance with a general plan, in the progressive differentiation and adjustment of its constituents, and in their mutual general relations. The entrance of more or less heterogeneous sensations and their correlation and adequate expression through efferent responses leads to a continually increasing and finely adapted consolidation in which the correlation of the various exteroceptive, proprioceptive, and even interoceptive impulses plays a large part. The development of the cortex in particular presents a continually increasing and finely adapted association and control of proprioceptive and exteroceptive impulses.

This development of a central nervous system is preceded by the formation of regions of particular sensitivity (the sense organs), which further the perception of those stimuli which relate the individual to the outside world. Among these, the perceptions of the so-called"" vital ” or protopathic impulses develop somewhat earlier than those of the more objective “gnostic” or epicritic sensations. This is to be expected, since the former are more .subjective and are largely protective, being concerned with impulses harmful or useful to the body, while the more objective gnostic or epicritic sensibility increases the knowledge of the outside world.

The impulses coming into the nervous system become effective through efferent nerve paths which provide an outlet for them. Yet not all afferent impulses are followed by motor responses. The interplay of the various incoming impulses may result merely in a condition of central equilibrium or correlation, so that there is no discharge over efferent paths, or a discharge may be initiated which is inhibitory in its effects, or merely gnostic in character in that it adds to the knowledge of the environment.

' The expression " cntclpphy ” i<! not used here as a property apart from the organi.sm, superimjKjM'd up<jii it. hut a« an inlierent organic property.

No teleological explanation of the development of the nervous system was intended in the reference to the inner urge or development and expression characteristic of living matter. The nervous system has not developed during phylogeny with the brain of man as its fixed pattern or goal. Evolution as a whole has occurred without molding itself to a form established in anticipation of the progressive development. The ape-like progenitor of man did not have in mind the characteristic human form as a pattern wliich he should imitate. This may be expressed by saying that the development here is entelechic rather than teleologic, for each stage is an end within itself — not a step toward a wellknown, predetermined goal.

CHAPTER I THE EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS

In the Protozoa no types of nervous organization are known, other than the fibrillar structures occurring in an excitomotor apparatus such as that present during the flagellate phase of the life history of the soil amoeba (Naegleria gruberi), described by Wilsori (’16), or those found in certain infusorians in connection with the ciliary apparatus. Examples of such fibrillar structures are the elaborate neuromotor mechanisms of Diplodinium ecaudatum {Sharp, ’14), Euplotes {Yocum, ’18 ; Taylor, ’20), and Trichonympha {Kofoid and Swezy, ’19; Hinshaw, ’26). Thus far such protoplasmic fibrillar differentiations do not appear to have been observed in sponges.^ However, Parker (’19) showed that in these latter animals, long muscle cells interconnected by protoplasmic bridges allow stimulations to spread over fairly great distances, and that different stimuli entering at the same time from different sources act as correlated or associated stimulations.

The first clearly demonstrable, structurally independent nerve cells are found in coelenterates, although the present knowledge of the nervous mechanisms in these forms is still incomplete. Over a decade ago Child (’21) substantiated a statement and its interpretation, made by Kleinenberg (’72), who described in the fresh water hydra certain cells which he regarded as neuromotor cells. These are cells in the epithelial layer and are often provided with hairs on the free surface which serve as aids in receiving impulses. The impulses, in turn, are elaborated by the muscular feet. Whether such cells may be regarded as nerve cells is still open to question. All epithelio-muscular cells receive and elaborate impulses. It is certain that neurons have not been observed to differentiate from such cells.

In addition to the cells described in the preceding paragraph, undoubted nerve cells make their appearance in coelenterates. These are of two types : (1) the neurosensory cells (sensory cells of Herrick, ’24a), and (2) more deeply situated, ganghon cells which form a subepithelial nerve plexus. The neurosensory cells may terminate directly on a muscle cell (as in the sea anemone) or the impulse may pass from the neurosensory to a primitive ganglion cell and be conveyed through the subepithelial net to the effector apparatus (as in the jellyfish). The Hertwigs (79 and ’80), Wolff (’04), 0. Hertwig (’18), Parker (’19), Droogleever Fortuyn (’20), Bozler (’27), Hanstrom (’29), and others have pointed out that almost always in these primitive nervous systems the entering impulses are received and sent on by neurosensory cells. These cells have a receptive as well as a conductive function. They are polarized cells always conducting from the surface to the effector apparatus or to the plexus in which they end. They precede the primitive ganglion cells in the sequence of activity of the nervous system and are probably more primitive than these latter cells. Furthermore, neurosensory cells are very numerous in

Coiiai«raii certain of the lower vertebrates and, presumably, are to be regarded not only . as phylogenetically older, but also, in part at least, as a

possible source of the ganglion cells.

Sponges, then, represent that stage in evolution in which a primitive type of muscle tissue

has made its appearance unaccompanied with nervous elements” {Parker, ’19). However, C. J. Herrick (’24) appears to have regarded the structures in the sponge, termed muscles by Parker, as neuromotor mechanisms, although he considered that the receptors and the effectors are not so highly developed in these forms as in certain of the Protozoa.

Further evidence of the greater age, phylogenetically speaking, of the neurosensory cells is to be found in the more primitive and less differentiated character of these cells as compared with the primitive ganglion cells. Thus the cell bodies of the neurosensory cells do not have characteristic dendritic offshoots and do not contain tigroid substance. Their neuraxes, like those of certain ganglion cells, are unmedullated. These cells are much more numerous in the invertebrates than in the vertebrates. The visual cells of Amphioxus, the neurosensory cells of the parietal and lateral eyes, and the infundibular cells found in fishes are examples of neurosensory cells in the lower vertebrates. The peripheral olfactory neurons represent the most typical examples of such cells in the mammals.

The cell bodies of neurosensory cells usually lie in the surface epithelium and frequently their free borders carry one or more hairs. These sense hairs may be pointed or they may have knobs at the end, as in the infundibular organ of the fishes (fig. 2B). According to van der Siricht (’09), the centrosome lies at the peripheral protrusion of the olfactory cells. In cases where the main portion of the cell body lies below the covering epithelium, a prolongation of that cell body may reach the surface, as in che olfactory cells in figure 1 and figure 2C. In certain places, such as in the saccus vasculosus or the infundibular organ of fishes, the neurosensory cells lie in the surface epithelium of an internal organ. In the saccus vasculosus, the sense hairs extend into the ventricle, in which is the cerebrospinal fluid (fig. 2B). Sometimes neither the neurosensory cells nor their offshoots reach a surface, but are embedded in other tissue. Examples of such a condition are to be found in the visual cells of the compound eyes of insects and of worms. The spinal light-perceiving cells of Amphioxus are neurosensory cells found within the central nervous system (refer to fig. 61). Cells thus embedded are not provided with sense hairs but sometimes present a striated border.

Fig. 1. Neurosensory cells in the olfactory epithelium of a mammal and their connections with other neurons. V. Gehuchten.

Neurosensory cells have distinct fibrils in their protoplasm (figs. 2B and 62). These fibrils, which continue into the efferent process or neuraxis, sometimes show a special arrangement, as is the case in the light-perceiving cells of Amphioxus.

The efferent offshoot or the neuraxis often divides dichotomously (particularly in the lower vertebrates) and nearly always has synaptic relations with some other nerve cell (a neuron or, in lower animals, a subepithelial plexus of ganglion cells). As was stated previously, the neuraxes of neurosensory cells do not possess a myelin sheath as do the neuraxes of many neurons.

Their rate of conduction is slow.

Thus it is only 6-20 cm. per second in the fila olfactoria of the pike{Nicolai, ’01), whereas it may be as high as 120 m. per second in the nervus ischiadicus of mammals.

This indicates that the rate in the latter is at least 600 times that in the former.

The primitive character of the neurosensory cells is evident in those relatively rare cases in which such a cell carries on all the functions of a nervous path, including the reception and conduction of the stimuli and the direct innervation of the muscle fibers. Examples of this occur occasionally in coelenterates, such as the sea anemone and certain polyps, and in the tentacles of mollusks (Samassa, ’93).

Typical neurosensory cells of the invertebrates have long neuraxes and distinct neurofibrils. The rods and cones of the vertebrate retina, although in reality neurosensory cells, have very short neuraxes (see fig. 30) and the neurofibrils are inevident in their cell bodies. Schneider (’02) and Bernard and Cantab (’03) considered that these cells possess fibrillae, while Kolmer (’04) questioned their presence.^ However, in the neuraxes of these cells neurofibrils are evident {Ariens Kappers). In the neurosensory cells of the retina, the diplosomes lie in that part of the cell body which receives the stimulus (figs. 3A and 3B). In adults such diplosomes are found at the border of the inner and outer limb and are connected by a strong “outer thread” which is turned toward the top {Kolmer, ’04; Held, ’05 ; Retzius, ’05) and, less constantly, by a much finer “inner thread.” In the region of the rods and cones such special structures as the ellipsoid bodies and the visual purple may develop. For the relation of these to color vision, reference should be made to the special treatises on this subject. Rods and cones show a phototropic reaction toward light {Engelmann, ’85; van Genderen Start, ’87). Rods show a positive and cones a negative phototropism, but the reaction of the latter is slight, especially in higher animals {Garten, ’07).

According to Amhronn and Held (’90), the fila olfactoria of the pike have a lecithin-like substance within the fibers, as the neurites of young neurons may have before the formation of the

myelin sheaths. This probably offers a resistance to conduction.

Fig. 2. Different forms of neurosensory cells : A, of a coelenterate (Metridium dianthus, Havet ) ; B, of the saccus vasculosus of a fish (Trulta iridea, Dammerman) ;

C, of the olfactory epithelium of a rabbit (i». Gehuchten) ;

D, of a moliusk (Limax, Veratti).

These facts show clearly that nervous elements are able to show tropic reactions of two sorts ; those of a stimulo-petal as well as those of a stimulo-fugal type.

The differences between neurosensory cells and sense cells or neuroepithelial cells must be borne in mind. The latter are of more recent development — occurring first in arthropods — and are more nearly related to the sheath cells (lemnoblasts) than to the nerve cells. They have no neuraxes and do not present neurofibrils, although fibrils of sensory nerves may terminate on them or, according to certain observers {Heringa, ’17b, and others), may terminate both on and in them. Typical examples of sense cells are the hair cells situated between the Deiters’ cells in the organ of Corti, the hair cells found in the ampullary portions of the semicircular canals, and the sense cells present in the lateral line organs and in the taste buds (fig. 139). The sense cells often originate from ordinary epithelial cells. If they retain their position in an epithelial layer, they may be provided with hairs (for example, the sense cells in the taste buds and the auditory cells of the organ of Corti). However, they do not invariably reach the surface of the epithehum, but may be covered over by other epithelial cells as in the case of Merkel’s tactile disks (fig. 26A). Sometimes they are embedded in connective tissue, in which case they do not possess sense hairs. It is probable that, in certain cases, sense cells lower the threshold of perception. Their functional activities are in accord with the lack of formation of true neurofibrils within the cells. They may improve perception but are not conductors, having no neuraxes. Those fibrillae which are found on or, according to some writers, in them are to be regarded as the telodendria of peripheral neurons (London, ’05 ; Kohner, ’05 ; Boeke, ’08 and elsewhere). In the auditory cells (Held, ’04) and the taste cells of Botezat (’08), the centrosome lies at that side of the cell which is turned toward the stimulus. Sometimes it is connected with the sense hair in a manner similar to the relations of the diplosomes of the hairs of the neurosensory cells.

The fibrillar ellipsoidal portion of the rods is not to be considered as a neurofibrillar

formation.

Fig. 3. Neurosensory cells (rods and cones) of the retina.

A. In a salmon embryo after Furst. The position of the diplosomes is noted in the future peripheral part of the cells.

B. Adult rods (left) and cones (right) from the eye of a bony fish (Blennius), according to Kolmer. The position of the diplosomes and their connections with the outer (A./.) and inner threads (/./.) are to be noted.

The primitive ganglion

cells (see page 1) are found

as far down in the animal

scale as coelenterates. In

these animals they occur

in the subepithelial plexus.

In the medusae, a more

concentrated portion of S^aicytial arrangement of primitive

^ intestine of Pontobdella. Apathy,

this nerve net or plexus

forms a nerve ring around the margin of the umbrella. The primitive ganglion cells show a higher degree of differentiation than do the neurosensory cells, for not only are they provided with neurofibrils, but they also contain tigroid or Nissl substance (McClure, ’96; Wolff, ’04; Smallwood and Rogers, ’09). Such tigroid substance has never been found in the neurosensory cells. However, in comparison with the neurons of the higher invertebrates and the vertebrates, these ganglion cells are still very primitive. Their processes show no differentiation into neuraxes and dendrites. The same process may conduct in either direction, that is, cellulofugally or cellulopetally. Thus the nervous impulse in the umbrella of a medusa may run both clockwise and counterclockwise over the same nerve ring. The plexus thus formed by these primitive ganglion cells is spoken of as an asynaptic or nonsynaptic network. Recently Bozler (’27) has expressed doubt with regard to the syncytial character of the primitive ganghon cells in coelenterates, although most observers are inclined to regard this network as syncytial in character, the result of incomplete cell division, believing that the connecting bands of protoplasm are relatively broad.

Primitive ganglion cells do not have processes surrounded by myelin sheaths. Conduction is still very slow in such a plexus. Parker ('18) found that the impulse in Metridium traveled at a rate of 12-14 cm. per second at 21° C. Harvey (’22) found the rate to be 46 cm. per second in jellyfish. The average of these figures (that is, 24 cm. per second) indicates that the rapidity of conduction is little more in these primitive ganglion cells of coelenterates than in the

neurosensory cells (see the previous account). It is even less in the cells of Metridium than in the previously described neurosensory elements.

The types of cells mentioned above — (1) the neurosensory cells and (2) the primitive ganglion cells — are the only constituents of the primitive nervous system of the coelenterates. In the progress of higher evolution, nerve cells of these types become less numerous and are replaced by the synaptic polarized nerve cells or neurons typical of higher forms. In man, neurosensory cells occur only in the olfactory organ and in the retina. In all probability, primitive ganglion cells are represented in the amacrine cells of the retina

Fia. 5. Abdominal nerve cord of earthworm. Reizius, Above, two unipolar neurons ; medially placed, a primitive multipolar neuron.

{Ramon y Cajal, '93 and elsewhere). Somewhat similar cells are said to be found in the tectum of lower fishes {Tretjakoff, ’09) and in the bulbar formation of certain urodeles {Herrick, ’24).

The term neuron {Waldeyer, ’91) implies a polarized nerve cell with its cell body, its processes, and their terminations. Primitive neurons, which in the course of evolution are added to the neurosensory cells and to the primitive ganglion cells, have made their appearance in the flatworms. They are characterized by the presence of two types of processes, dendrites and neuraxes or axons. The dendrites are those processes which carry impulses toward the cell body (cellulopetally) ; the neuraxis or axon is that process which carries impulses away from the cell body (cellulofugally). In their simplest form these neurons resemble the neurosensory cells, differing only through the presence of one or two cytoplasmic prolongations which extend to the surface covering of the body. Such peripheral processes are dendrites. This specialization in the direction of conduction is not dependent upon a monoconductive character of the process itself, since it has been shown experimentally that either the neuraxis or the dendrite may carry stimuli in either direction. The action current demonstrates that when a fiber is stimulated in its course, the nervous current passes in both directions along the fiber from the point of stimulation. Not the process itself, then, but its topographic relations with other neurons and its position with regard to receptor or effector determine the direction in which impulses will pass along it. Since such relations are relatively constant, it is possible to predict with reasonable certainty that the direction of conduction over particular fibers will remain the same."* Similar polarization is found in the neurosensory cells, since the location of such a cell permits it to receive stimulation only at its cell body or sense hair, while its neuraxis must always carry impulses cellulofugally. Polarization, then, does not occur first in the neurons, although it finds there a morphologic expression in the differentiated structure of the receptive dendrites and the effective neuraxes. The cytological differentiation is primarily the result of the polarized transmission and not its cause. Once established, however, the specialized cytological structiue undoubtedly contributes to the smoothness of polarized transmission.

A dendrite is to be regarded solely as a prolongation of the cell body, which was the original receptor and is still the only one in a neurosensory cell. In every respect, dendrites retain the characteristics of the protoplasm of saji the cell body. They share in its trophic functions, as is nerve of an embryo of indicated by the presence of tigroid substances and cobaya. v. Gehuch oxydases within them (see below). The neuraxis differs transition stage from from the dendrite in structure, in having no tigroid sub- bipolar ganglion cell (a) to stance and no oxydases; probably it has more readily “o“opo>®ganglioncell(c). ionizable potassium salts. Consequently, the principles of structure remain very much the same in the neuron as in the neurosensory cell, the receptive portion being generally cell protoplasm and only the cellulofugal offshoot showing specialization. It is interesting in this connection, that the developing neuron or neuroblast shows resemblance to a neurosensory cell in that it has only a neuraxis at first. The dendrites arise later.

The differentiation of dendrites and neuraxes is not the only indication of the polarized character of the nemons. Although inherently both types of processes appear to be capable of conducting in either way, the impulse may be transmitted from one neuron to another only in one direction. The region where the stimulation passes from one neuron to the next is spoken of as a synapse, and synaptic conduction is strictly polarized. It is evident from the above facts that this is not due to a monoconductive character of the nerve processes involved, and the precise causes underlying this polarization at the S 3 mapse are still matters of considerable dispute. A further discussion of the synapse is to be found on page 23.

Only with the axon reflexes in the autonomic cells is conduction cellulopetal in the neuraxis.

Genetically, the sensory or afferent neuron is very closely allied to the neurosensory cells. Two indications of this have already been given and are as follows : (1) the neurosensory cells, because of their special position, have a polarization suggestive of the polarized condition which corresponds to that found in developing neurons, the neuroblasts ; (2) the fully differentiated neurosensory cells have only sensory functions. Further evidence in favor of the relation between neurosensory cells and sensory neurons is furnished by the fact that in certain invertebrates sensory roots consist partly of the processes of neurosensory cells and partly of those of true neurons. Thus it is sometimes difficult to state exactly where the division lies between those neurosensory cells with one or two peripheral offshoots and the typical sensory neurons. This relation between the two is so intimate that neurosensory cells are frequently designated merely as sensory neurons.

Fig. 7. A. Position of the centrosome in the dendritic process of an embryonic bipolar ganglion cell, van der Slricht.

B Position of the centrosome in a monopolar spinal ganglion cell of an adult rat. Hatai.

Motor neurons, also, are probably related genetically to neurosensory cells, for these latter cells, as the previous account has shown (see references to Samassa, ’93 ; Parker, ’19 ; Droogleever Fortuyn, ’20), are not only receptive but may also be directly connected with body musculature. When, in the more highly differentiated animals, a division of labor takes place, the neurosensory cell becomes the receptor or sensory side of the arc and an efferent mechanism (represented at first by the primitive ganglion cells) appears. This efferent mechanism is primarily without permanent polarization, but with temporary polarization impressed upon it at the instant of the passage of an impulse from the polarized neurosensory or primitive sensory cell.

Neurons of higher forms differ from neurosensory cells in having tigroid substance and dendrites, which, unlike the short and usually single receptive processes occasionally seen on neurosensory cells (Veratti, ’00a), are usually numerous and of considerable length. These dendrites permit the reception of stimulations by the neuron from different directions and from different sources, so that the impulse discharging over the neuraxis may be the resultant of a number of different stimulations.® A further difference between typical sensory neurons and neurosensory cells is to be foimd in the presence, at least frequently, of a myehn sheath around the neuraxis and, in certain cases, around the dendrites (particularly in the peripheral sensory nerves). A myelin sheath has been recognized as far down in phylogeny as the arthropods, although its relations are somewhat different there than in higher vertebrates (Retzius, ’90, and Nageotte ’16). The development of the myelin sheath increases the rapidity of conduction so that it may rise to 125 meters per second. It influences conduction through the elimination of fatty substances, which are poor conductors, from the neuraxis, and through the increase of the current due to the sheath itself {Gdthlin, ’13 and ’17), and serves as an insulator by preventing such lateral radiation from the neuraxis as does occur where these processes are entirely naked (see the account of the parallel fibers of the cerebellum, p. 700).

Fig. 8. A spinal ganglion cell with fenestrated border from Orthagoriscus mola. G. Levi.

Although neurons are found in the flatworms, they become much more abundant in higher invertebrates and are practically the only nerve cells present in the central nervous system of vertebrates. They vary in form, size, position, and specialized function, as well as in the number of their processes, and these differences have been made the basis of special classifications.

One of the most usual classifications is that into unipolar, bipolar, and multipolar types. The most primitive nemons are bipolar in form, having a single dendritic process and a neuraxis. They resemble most closely the neurosensory cells. They occur in the animal scale at least as low down as the flatworms. A variation of this type is the unipolar or monopolar neuron in which the dendrite and neuraxis arise from a single stem. According to Hanstrom (’29), this is a secondary condition due to a (trophic) migration of the cell body toward the periphery of the fiber tracts. Such monopolar types are found very frequently in annelids and arthropods and in the spinal and most of the cranial ganglia of vertebrates.

In the nematode, Distomum, Havet (’00) described rather deeply situated cells which he

termed nerve cells. Such cells have one or two processes running to the periphery, comparable in appearance to those of the neurosensory cells but showing side branches. These cells appear intermediate in type between the neurosensory cells and the typical sensory neurons. Similar cells having two or more receptive processes were described by Veralti (’00a) in mollusks, where they are particularly evident in the tentacles.

But while the cells of the ganglia are unipolar or bipolar in type, cells of the central nervous system are, with a few exceptions (such as certain neurons of the mesencephalic nucleus of the trigeminal), bipolar or multipolar in type. Many of the Purkinje cells of the cerebellum (fig. 48) afford typical examples of bipolar cells, with the neuraxis arising at one pole and a large dendrite, which soon divides, at the other pole. Peculiar bipolar cells occur rarely, in which two dendrites are given off from the cell body and the neuraxis arises as a branch from one of these dendrites. The horizontal cells of Ramon y Cajal (’91 and ’ll) and Retzius (’93a and ’94), present in the cortex of very young mammalian and human foetuses, are the only pluriaxonal neurons known. These cells produce two (and occasionally even three) neuraxes. Such neuraxes (fig. 9) do not leave the cell from neighboring areas on the cell body but arise from dendrites and in positions relatively far apart. They may give off collaterals.

Fig. 9. Horizontal cell from the molecular zone of the cerebral cortex of a rabbit eight days old. Ramdn y Cajal.

./I, neuraxis; D, dendrites; C, collaterals.

The majority of neurons in the vertebrate central nervous system are multipolar cells with several dendrites and one neuraxis for each cell. Typical examples of such types are found in the pyramidal cells of the cerebral cortex (fig. 10), the motor cells of the ventral horn of the cord (fig. 96), and the mitral cells of the olfactory bulb (fig. 1). But regardless of various classifications, all neurons have certain characteristics in common, which now will be summarized briefly.

The Structure of Neurons and Their Processes

As a liv’ing cell, the neuron has a protoplasmic structure comparable to that of other cells of the body. Usually this structure is regarded as colloidal in character and as having a great resemblance to an emulsion type {Greeley, ’04 ; Ralph Lillie, ’24). The cell body contains a large, spheroidal nucleus with a nuclear membrane inclosing a relatively small but variable amount of chromatin. One or more nucleoli are present ; rather large, spheroidal masses, from which, in certain cells, a so-called fiber of Roncoroni {Mend, '06; Schafer, ’12) extends into the cytoplasm. Certain observ’ers {Hatai, '01 ; Page May and Walker, ’07, and othens) believe that the nucleolus may migrate through the nuclear membrane ; other investigators regard the substance thus described as chrornidial ia character. In Lophius, Holmgren (’99, ’00) showed a depression of the nuclear membrane in the region of the nucleolus and a radiation of nuclear substance to form the tigroid substance of the cytoplasm.

A consideration of the cytoplasm (perikaryon) of the neuron shows it to be highly differentiated. A centrosome is sometimes evident. According to Ariens Kappers, the earliest account of a centrosome in a neuron was given for the frog ; this led to its recognition in the neurons of many animal types (for example, von Lenhossek, ’95, and Dehler, ’95, in the frog ; Lewis, ’96, in certain annelids ; Ramdn y Cajal, ’09, in batrachians, reptiles, and mammals) and to the demonstration of its presence in the large pyramidal cells of the postcentral gyrus of a thirty-year-old man. The position of the centrosome is somewhat variable. Often, but not always, it lies near the nucleus (Hatai, ’01 ; see also fig. lOD), and in the multipolar neurons is near the point of entrance of the principal dendrite (van der Slricht, ’09; see figs. 7 and lOD). According to Del Rio Hortega (’16c), the centrosome often acquires a bacillar form, particularly in older individuals. Then frequently alterations of a filamentous character appear, and part of the filament may extend out into the dendrite. It is possible that the process may be analogous to the formation of the outer thread in the rods and cones of the retina. Since it is questionable whether cell division in neurons ever occurs after the differentiation of the specific cytoplasmic constituents, such mitosis is certainly exceedingly rare, if ever present, in older individuals. Consequently the function of the centrosome in such neurons is unknown. It may be associated with the reception of impulses (Ariens Kappers, ’20).

In this connection, one is reminded that the granules (basal corpuscles) at the base of the hairs in the neurosensory cells and the diplosomes of the rods and cones — both derivatives of the centrosomes — are located in the cells near the point of entrance of the impulse. It likewise agrees with the fact discovered by Held (’09a), that the position of the centrosome coincides with that of the fibrillogenous zone. This zone is evidently the first center of stimulation (or its embryologic equivalent). Since the stimulofugal process, the neuraxis, is formed first (fig. 47), the centrosome lies first near its point of exit, but later shifts its position with the appearance of the principal dendrite. Such facts are of great interest in the study of the underlying principles of tissue differentiation and as applied to the conceptions advocated by Rabl (’89)j Driesch (’94), and Ariens Kappers (’20).

The neurofibrillar constituents of the neuron (figs. lOB, HA, and IIB) are present in the cell body, dendrites, and neuraxis. The presence of fibrils within the neuraxis had been commented upon earlier (Barker, ’99), but they were described first in cell bodies and dendrites by SchuUze (’71) and apparently are invariably present in properly fixed material. Precisely what the nature of these fibrils may be is still in controversy. According to SchuUze (’71), Ranvier (’74), Flemming (’82, ’95), Dogiel (’95), Lugaro (’97), Bethe (’03, see also ’08), Wolff (’05), Marui (’18), and others, they are fibrils extending as discrete structures throughout dendrites, cell body, and neuraxis to their ultimate destination. Apathy (’97, ’98, ’07, ’08) found a network of neurofibrils which constituted intercellular bridges in the leech. According to his theory, they took origin from so-called nerve cells and passed through a series of ganglion cells to their termination. In invertebrates and certain vertebrates, he found neurofibrils entering by the dendrites, ramifying and anastomosing within the cell body, and reuniting to enter the neuraxis. Other observers {Frommann, ’64 ; Ramon y Cajal, ’09 ; Bielschowsky , ’28, and many others) have regarded them as anastomosing, the major fibrils forming the heavier threads of a network ; still others have considered that they show indication of being tubular in character {Schafer, ’12, p. 234, fig. 368). Mann (’98) stated that the neurofibrils are the only portions of the nerve fiber continued across the nodes of Ranvier, but this statement is difficult of proof and lacks confirmation. Repeatedly the question as to the possibility of their being fixation artifacts has been raised. In some of the tissue culture work on developing motor and sympathetic nerves, the most careful examination of a living, developing fiber failed to reveal any indication of neurofibrils, although after fixation typical fibrils were present (Matsumato, ’20 ; Lewis and Lewis, ’24). However, Schultze (’71) had claimed that he was able to differentiate such fibrils in fresh cells which had been prepared without the use of fixatives or stains. By the use of polarized light, neurofibrils were seen by Grant Smith (’06) in both living and fixed material and Howard (’08) demonstrated neurofibrils both in the retinal cells of various vertebrates and in the nerve fibers as well. Bozler (’27) was able to find neurofibrils in the living nerve cells of Rhizostoma and Tiegs (’27 ; see also ’31) found them in teased fragments of living fetal rabbit cord. Recently de Renyi (’29) demonstrated the presence of neurofibrils in the giant nerve fibers of the lobster, although he had not demonstrated them in the frog (’29a). Moreover, intravitam stains demonstrate a neurofibrillar structure in the neiwe cells, but the objection that such stains are toxic has been raised by certain experimenters. Schafer (’12) thought that the ease with which neurofibrils show varicosities suggests that they are se mifl uid in character and referred to Carlson’s {Carlson, ’04, ’05 ; Jenkins and Carlson, ’03, ’04) experiments on the necessity for a fluid-conducting substance in the nerve as supporting evidence for this view {Schafer, ’12, p. 235). Jenkins and Carlson (see also Leydig, ’97) ascribed the most important role in nervous conduction to this more fluid perifibrillar substance, as did Parker (’29) and Ariens Kappers (’32), according to whom the neurofibrils are coagulation products, caused by frequent impulse currents.

Three points of view, then, regarding conductivity along the neuron have been presented ; (1) that it occurs along the neurofibrils — that these are the specific conducting elements ; (2) that it occurs in the more fluid neuroplasma between the more stable neurofibrils ; (3) that it is due to an interaction between the neurofibrils and the neuroplasm. Apathy (’97 and elsewhere) considered that in the primitive ganglion cells of invertebrates the neurofibrils pass from one cell body to the other, through the syncytial network. Schultze (’71), Bethe (’03, ’04, ’06, ’07, ’08, etc.), Wolff (’03), Bielschowsky (’05), Oudendal (’12), Marui (’18), and others believed that at times they are continuous across the synapse in the vertebrate nervous system, although this is not in accord with the views of radical supporters of the neuron theory {Retzius, ’08 ; Ramon y Cajal, ’06, ’09, etc., and many others). The fact that the nervous impulse may irradiate sidewise from the neuron shows that these fibrils are not a conditio sine qua non for conduction.

Fig. 10. A. Neuroblasts from the cortex of an embryo of Cavia cobaya, 23 cm. in length. Held. Note the plasmodeses between the first stage neuroblasts in the middle.

B. Unipolar neuroblasts of second stage of a duck embryo of the third day of incubation. Held. The first development of the neurofibrils is noted in the region of the process and here also are to be seen the diplosomes.

C. Neuroblast in vitro from a lymph culture of Rana palestris ; the neuroblast on the left after 52 hours of culture, and that on the right after 56^ hours of culture. Harrison.

D. In the upper figure, the cell body of a Purkinje cell of an adult rat, showing the location of the centrosomes opposite the dendrite. In the lower figure, the cell body of a pyramidal cell, showing the centrosome opposite the main dendrite. Halai.

The form and size of the neurofibrils vary under varying conditions. Thus in invertebrates, at least, they appear to be finer in the sensory than in the motor neurons {Apathy, ’97 ; see also ’98). They become less numerous and thicker

during cold and hunger and during hibernation {Ramon y Cajal, ’04, ’09 ; Tello

/ / ’04 ; Marinesco, ’06 ; Donaggio,

y ’06; and others), and in general show an increase in number and are finer during functional activity (hg. 11 ■ see also Dustin, ’06).

^ \ Various pathologic conditions also

' affect the neurofibrils and may lead I to their disappearance. For such changes, texts on neuropathology A should be consulted.

^ Nissl granules {Nissl, ’84) —

y' ^ chromidial or tigroid substance {von

n Lenhossek, ’97) — consist of nucleo ” proteids which contain iron {Scott,

0 ’99 ; Macallum, ’05 ; Nicholson, ^ J ’23). These disappear on treat ment with ammonia and, in accordance with their acid character, are stained by basic dyes. They are

^ present in the larger nerve cells, but

\ have not been demonstrated satis VuS B * factorily in some of the smaller cells

^ % * (granular types) of the cerebro Fio. 11. The form of the neurofibrils in a lizard; A spinal nervoUS system. The gran and a, during hibernation ; B and b, in the awakened ules are present in the Cell body animal after it had been warmed for several hours at j i j -i n j. j. • ., 0 ,,

30’ c. Ramdn y Cajal. dendrites but not m the neu raxis nor at its place of attachment to the nerve cell (the axon hillock or implantation cone, fig. 12). They tend to form clumps or masses and vary in number, appearance, and size in the different cells. Using these morphologic differences, Nissl (’03) worked out an elaborate classification of this substance, for the details of which, reference is made to his work. It may be stated here that he subdivided nerve cells into two main groups, those in which the cell body was well developed and distinctly outlined, the somatochrome nerve cells, and those in which the cytoplasm was either very poorly developed or contained little stainable substance. This second group fell into two classes : the granules or cytochrome cells, with small nuclei approximating in size those of neuroglia cells and the caryochrome cells with larger nuclei approaching in size that of the usual neuron. The majority of nerve cells, of course, belong to the first group, the somatochrome cells of Nissl. These somatochrome cells were thrown into various subgroups, these being based on the types and arrangement of the Nissl or tigroid substance. It is now known that, to some extent, the size and character of the Nissl bodies in a given cell are dependent upon the fixation employed {Cowdry, ’24, and others), but in similarly fixed material or in cells side by side in the same material and con^quently subjected to precisely the same conditions, differences in the size-nisaftHS

Fig. 12. To the left, a normal pyramidal cell from the motor corte.x (area gigantopyramidalis) of an adult man ; to the right, a tigroly tic cell with scalloping of the nuclear membrane.

appearance are evident. That these morphologic differences apparently have a basis in functional differences in the cells has been shown by the studies of Jflcobsohn (’08) and of Malone (’13, ’23). This latter worker has shown that differences in size of cell and in general characteristics of the Nissl substance could be demonstrated in cells supplying voluntary, heart, and smooth muscle. His work also indicated that the functions of various portions of the nervous system could be predicated from their cell types.

In spite of their characteristic appearance in properly stained material, formed Nissl granules have not been demonstrated satisfactorily in living cells {Cowdry, ’24, and others). There is every probability that the form which they take in the fixed material is the result of a coagulation of the protoplasm due to the use of fixation fluids. The work of MoU (’15) appeared to indicate that the substance is quite different in dead cells than in the living neuron. H. C. Voorhoeve ('26) explained its non-appearance in the living cells as due to its having a light refracting index in ordinary or ultra-violet light not unlike that of the rest of the cytoplasm. However, the so-called axon reaction, the chromatolysis or tigrolysis (fig. 12, to the right) of the Nissl substance after injury to the neuraxis, the disappearance of this substance wholly or in part after fatigue or after toxins either administered or produced by disease, and the reappearance, under favorable conditions, of these granules within the cell appear to establish the existence of a substance within the living cell vital to its effective functional activity and indicated — although not directly represented — by the stainable Nissl substance of ffxed material. Macallmn (’05) regarded this substance (a nucleoproteid) as the greatest iron-containing protein \vithin the cell. iVIicrochemical tests (Scott, '99 ; Macallum, '05 ; Nicholson, ’23) indicate the stainable substance of Nissl to be an iron-containing protein. Nicholson (’23) showed further that an actual loss of iron and consequently a real chemical change can be demonstrated where chromatolysis has occurred following an injury to the neura-xis.

The origin of the Nissl substance is still in question. The nucleoproteids of these cells are supposed to originate in the nucleus ® (Holmgren, ’99, ’99a, '00, ’00a; Scott, ’99; Cameron, ’06, and others). In a nerve cell of Lophius, Holmgren (’99) figured a wide gap in the nuclear membrane opposite the centrosome and the apparent formation of Nissl granules in the region. Various authors have favored the theory of such a migration of nucleoproteins from the chromatin material and used it as a basis for explaining the small amount of basichromatin present in the nucleus of the neuron (fig. 12, on left). It is true, certainly, that more chromatin is present in the nuclei of the caryochrome cells of Nissl (those which do not contain tigroid bodies) than in other neurons. Also, both chromatin and Nissl substance are fermented by nuclease (van Herwerden, ’13, ’13a) and both react positively to the nuclear reaction of Feulgen and Rossenbeck (Redenz, ’25, and Voorhoeve, ’26). Consequently, it seems to be established reasonably well that nucleoproteids migrating from the nucleus play a part in the formation of the Nissl substance of the nerve cell. That they form all of this substance is at least open to question, for the amount of Nissl substance appears to be too great to have been formed entirely from such migratory material. The fact that Nissl granules appear first at the periphery of the cell (van Biervliet, ’00 ; Marcora, ’ll) favors the idea that the cytoplasm may take part in their formation. Emil Holmgren (’00) believed that lipoid substance contributes to this. Muhlmarin (’12), working with young embryos, reached the conclusion that the substance contributed must be a neuroglobuline, since the Nissl granules, when treated with a methyl-green pyronine mixture, stained only with the pyronine.

The function of the Nissl substance is not known with certainty. Evidently it is related to the metabolism of the cell. There is a considerable body of evidence suggesting that its importance in metabolic activities is due to the ironholding properties of its nucleoproteid. Presumably the iron serves as a catalytic agent in accelerating oxidation within the cells {Warburg, T4 ; see also ’28). A. P. Mathews (’24) says that “iron catalyzes reactions in which hydrogen peroxide or ozonic acid takes part ; in the system, oxygen-water and its action in the cell are no doubt connected with this property.”

Holmgren (’99, ’00) and Page May and Walker (’07) speak of nucleoli migrating through

the nuclear membrane. It is possible they may be chromatic substance.

Thus it would seem that the Nissl substance is related directly to the passage of the nervous impulse.

The facts mentioned before — that it disappears after injury to the neuraxis and exposure to toxins and upon fatigue — support this interpretation.

Mitochondria are present in neurons as well as in other cells of higher forms. They are regarded as phospholipins, possibly in combination with a small amount of protein {Regaud, ’08 ; FaureFremiet, ’10 ; Cowdry, '18, ’24, and others). Microchemical tests show that, unlike Nissl granules, they contain practically no iron. Morphologically they may be either granular or filamentous in type, but usually they have the latter form in the nerve cells. The numerical variation is great, but mitochondria appear to be more abundant in younger, more active cells, and to decrease in senescence. They are more numerous in cells that have little fat and fewer where the amount of fat is increased appreciably. All these facts suggest that there is a direct relation

between the rate of oxidation within the cell and the amount of mitochondria . Attempts have been made, by means of a special apparatus, to count the number of mitochondria in nerve cells. The material used was the cranial nerve cells of white mice. From a study of such cells, Thurlow (’17) arrived at the conclusion that, per unit volume, a reasonably constant number of mitochondria is present.

It is not possible to enter here into a discussion of the various functions attributed to mitochondria. The literature which has accumulated on the subject is very great. Those interested may find very suggestive bibliographies in the papers of Duesberg (’12) and Cowdry (’18 and ’24). This latter author has collected a list of about eighty substances which are regarded as being affected in some way by mitochondria during their formation. The work of Kingsbury (’12) and Mayer and Schaeffer (’13) suggests that mitochondria may play some part in protoplasmic respiration.

The so-called Golgi apparatus (fig. 14) was first described in nerve cells by Golgi (’98 ; see also ’01) and Held (’02). Since that time, through the work of many experimenters, it has been shown to be present in all normal hving cells, although evidence is lacking that it is either structurally or functionally the same in all cells. It is regarded as lipin hnked with protein material, being similar in constitution to mitochondria {Gatenby, ’20). It hes around the nucleus and only enters the dendritic processes of a nerve cell on rare occasions {Sanchez, '16). It never enters the neuraxis. Penfield (’21) beheved that it is displaced to the periphery of the cell on section of the neuraxis (retispersion) and then soon disappears (retisolution). Students of invertebrate material (Hyman, ’23, for example) have conceived of it as consisting of two constituents, differentiable through their affinity for stain. It has not been observed clearly in unstained hving cells of vertebrates, nor has it been stained differentially through the use of vital dyes (Cowdry, ’24). The microdissection methods of Kite and Chambers (see Chambers ’24, for bibhography, quoted also by Cowdry, ’24) and the centrifugation experiments of Cowdry appear to indicate that, in the hving cell of vertebrates at least, the Golgi apparatus does not differ greatly in density from that of the remainder of the cytoplasm. In fixed material stained by silver-impregnation methods, the Golgi apparatus has the appearance of a reticulum, consisting of more or less densely arranged black threads. There is a certain amount of evidence — as yet not particularly clear cut or convincing — which indicates that the Golgi apparatus is affected by certain pathological conditions.^

Fig. 14. Golgi net in the spinal ganglion cell of a dog. Golgi.

m, mitochondria; n, Nissl granules.

Note the accumulation of the mitochondria around the axon hillock.

Holmgren ( 04, 04a, ’14, ’15, and elsewhere) described a canalicular system in cells (including neurons) which consists of so-called lymph-canaliculi into which processes of surrounding cells (trophozyten) extend, forming a network which he termed the trophospongium (fig. 15). That the canalicular system is present seems to be well established, but whether or not it is to be regarded as identical with the Golgi apparatus is still in dispute. Thus Ramon y Cajal (’09 and ’15) appears to have regarded it as identical, and used the term “appareil tubuleux du protoplasma” or Golgi-Holmgren canals. Duesberg (’14) regarded the systems described by Golgi and Holmgren as the same in all neurons and in some (but not all) non-nervous^cells, while Misch (’03), Bergen (’04), Sjdvall (’06), and Penfield (21 , see also 20) considered them quite independent structures. Cowdry ( 24, p. 342) stated that “observations are not lacking that there is often a close coriespondence between systems of clear canals and blackened networks in normal nerve cells. ”

Pigment containing lecithin is present in some nerve cells (Muhlmann, ’01 ; Obersteiner, ’03, and many others). It is usually found in a mass near the nucleus and may be in sufficient amounts to give color to the cell mass, as in the substantia nigra. It is more plentiful in older than in younger cells and in human than in lower forms.

Fig. 15. Spinal ganglion cell of a rabbit showing trophospongium and trophocytea extending into the cell. Holmgren.

'•eticular nets over certain neurons. According to ^ f e ^ these nets are sometimes regarded as terminal

pr<.ce.ssc.s of the neuraxes of other neurons, fonning thus a pericellular synapse, and sometimes arc considered to l>c of neurogliar origin. j i >

prGCC.*v>C.S \jt viit.. «tui44i4At;o UJ ULUUX iiuun

arc considered to l>c of neurogliar origin.

Oxidizing enzymes, occurring as very small granules within the cell, have been described by Katsanuma (’15), Marinesco (’19 and ’19a), and Ariens Kappers (’20). The presence of these oxydases (fig. 16) or oxydones (von Gierke) in the cell protoplasm (perikaryon) and in the dendrites is further evidence that these parts of the neuron play the more important role in the metabolic and particularly

Fig. 16. Reaction to a naphthol and paraphenylin diamine to show the oxydases within ■ three nerve cells (A and B from the telencephalon, and C from the mesencephalon of a bony fish). Ariens Kappers.

the anabolic functions of the cell. Still further confirmation is given by the work of Unna (’16), who demonstrated, through the use of nongalit and potassium permanganate staining, that a larger amount of oxygen is present in the dendrites and cell body than in the Nissl substance. The morphologic effects of the anabolic function of the dendrites are seen in such parts of a nervous system as have no entering blood vessels. Examples of these are found in the horizontal cells of the developing cortex and in the motor cells of the spinal cord of Ammocoetes (Tretjakoff, ’08). The dendrites of such cells have a tendency to approach the vascularized surface, ending there in thickenings.

The foregoing account has indicated repeatedly that the dendritic processes of the neuron usually have the cytoplasmic structure of the cell body and that they are vitally concerned in its metabolic- functions. Their direction of conduction of impulses is afferent with respect to the cell body. Dendrites within the central nervous system usually are relatively short and branched and do not carry myelin sheaths. The peripheral process of a spinal or cranial sensory neuron is long and has the structural characteristics of a neuraxis. Such a process is regarded as a dendrite because it conducts toward the cell body (see the account of spinal ganglion neurons).

The neuraxis, that process which conducts impulses away from the cell body, is specialized for conduction. That it also has metabolism is indicated by the production of CO 2 by a fiber even at rest, with an increase in the CO 2 when the impulse passes (Tashiro, T3, ’13a, ’15, ’17, ’22). Nissl substance is absent from it and from the axon hillock, but neurofibrils are continued through it into its finest branches. It varies in length from a millimeter or so to nearly a meter. It may or may not have a myelin sheath or medullary sheath about it ; outside, but not inside of the central nervous system, it will carry a neurolemma sheath.

In medullated peripheral nerves (whether the neuraxes of efferent or the dendrites of afferent nerves) the structure of the fiber is as follows :• The central portion, or axis cyfinder, consists of neurofibrils embedded in a fluid material, the neuroplasma. Around this portion is the myelin, separated from the neuraxis, according to some observers, by a thin, structureless membrane, the axolemma (sheath of Mauthner). The myelin sheath itself is composed of lecithin ( a phospholipin) , cholesterin, kephalin, and protagon. Its morphologic structure, as seen in fixed material, appears to vary with the fixing reagents used. Apparently on treatment with certain reagents, a keratin-like, coagulable substance separates out from the sheath and is laid down as a neurokeratin net. Funnels of Golgi probably are precipitated substances. Oblique interruptions of the myelin sheath, which Bruno (’31) regarded as of the same character as the neurolemma but which numerous observers have regarded as artifacts, are usually described under the name of Schmidt-Lantermann segments, although this is a misnomer according to Diamare (’32), who stated that they were described first by the Italian, Wladimiro Zawerthal. Perhaps the myelin, which is known to have a crystalline character, consists of oblique crystals, responsible for the appearance of the above-mentioned structures. Since the pattern of the net varies with the fixing fluid used, it appears probable that the structural patterns described for myeUn are largely artifacts (Huber, ’27). Nageotte (’09a), however, believed the neurokeratin net to be formed of mitachondria. Doinikow (’ll) regarded the myelin substance as deposited in a protoplasmic meshwork, produced by the cytoplasm of the neurolemma cells (the plasma cells of Schwann) . He thought that the more dense portion of the cell, together with the nucleus, remained on the outside as the sheath cell ; Nemiloff (’10) considered that the protoplasmic network associated with the neurolemma extends into the myelin sheath. At intervals along the fiber, the myelin sheath becomes discontinuous, forming the so-called nodes of Ranvier. At such nodes the neuraxis and the neurolemma sheaths remain continuous.® Such nodes occur at intervals of about 1 mm. on fibers of 10^ diameter. The branching of medullated fibers occurs at the

Sometimes (ATemtio#, 10; see figures and accounts in Bailey’s Histology, Bailey, Strong, am Mwyn, 25) the sheath cells are regarded as terminating at the nodes in such a way that the axolemma and the so-called neurolemma become continuous with each other. Thus the axolemma IS regarded as the inner boundary of the cell, the neurolemma sheath as the outer boundary, and the neurokeratm net as a fibrillar network within the cell. Observers, who take this point of view regarding the relations of the myelin to the sheath cells, consider certain of the neuroglia cells of t^ central nervous system as directly comparable with the neurolemma cells of the peripheral nodes in higher forms, but in some invertebrates, such as crustaceans, the division may take place within the internodal segment (Nageotte, T6). One sheath cell nucleus is found between every two nodes. In vertebrates such a nucleus lies external to the myelin sheath, but in arthropods the nucleus is inside of it.

The varying points of view regarding the structure of the myelin raise the question as to its embryological origin. Appearing, as it does in development, after the formation of the neuraxis and its enveloping membrane, the neurolemma sheath, and between these two ectodermal structures, the possibility of its mesodermal origin (which was taught earlier) appears to be definitely excluded. It may be safely concluded that it is ectodermal in origin, but the question as to whether or not it is derived from the neuraxis or from the neurolemma sheath

Fig. 17. A. Nerve fiber from the popliteal nerve of a fetal calf of 45 cm. length, showing the formation of the medullary sheath.

B. Preparations from the ventral white funiculus of the spinal cord of a fetal calf of 32 cm. length with glia cells with myelin droplets.

is less easily decided. Ranvier (’78) regarded the neurolemma as forming the myelin sheath and this conclusion has had considerable support ; Bardeen (’03) and von Kdlliker (’04) believed that the myelin is formed by the neuraxis. In his 1927 paper, Huber (pp. 1095-96) stated that evidence, though at present inconclusive, is at hand to support the interpretation that the myelin sheath is a part of the neuron.

It is certain that the myelin first appears during development (figs. 17A and 17B) as a thin, continuous layer around the nerve fiber, with no nodes of Ranvier; also, in regeneration of cut or injured nerves, no internodes and nodal segments are present at first. This would appear to indicate that it is related to the axis cylinder rather than to the neurolemma sheath. Also, myelin sheaths are present in the central nervous system where there are no neurolemma sheath cells and where only very rarely are neuroglia cells with myelin droplets found (fig. 17B). Furthermore, Ambronn and Held (’95) showed that within the fila olfactoria of the pike, a doubly refracting substance, probably a sort of myelin or lecithin, is demonstrable. Therefore, the present evidence appears to favor the theory that the myelin is a differentiation of the peripheral portion of the neuraxis (cf. Apathy, ’97, and Gothlin, ’17). Particularly favorable is the recent work of Speidel (’33). He studied in great detail the process of myelinization and myelin adjustment in living nerve cells over a period of several months, noting the selective myelinization which occurred under the influence of sheath cells in non-myelin emergent and myelin emergent fibers, and concluding myelin is “ an adjunct of the axis cylinder,” not part of the sheath cell. (For myeliniza^ tion within the central nervous system, see pages 1660 to 1662.

With regard to the function of this sheath, one theory is that the medullary sheath serves as an insulator, making possible more efficient conduction of the nervous impulses. In support of this contention are the facts that the myelin sheath is a poor conductor, that the passage of the nervous impulse is much more rapid in medullated than in non-meduUated fibers, that the majority, at least, of peripheral nerves which have highly specialized terminations and many of the longer tracts of the central nervous system have medullary sheaths, and that these sheaths develop, ortogenetically and phylogenetically, with the differentiation of the various fiber systems. Moreover, myelin is lacking in regions where insulation is undesirable ; it is not present at the synapse where the impulse passes from one neuron to another nor on the dendrites (with the exception of those of peripheral sensory neurons) nor on the telodendria of the neuraxis. Usually it is not found on the cell body, although there are a few exceptions to this, as, for example, around the bipolar cells of the spiral ganglion and sometimes around spinal ganglion cells in sharks. However, all these facts which seem to indicate, in one way or another, that the myelin has a protective and insulating function do not solve the question as to how the accumulation is effected. So far as its constituents may arise from the nerve fiber itself, it is probable that this secretion by the fiber is due to the passage of the nervous impulse. The researches of Ambronn and Held (’95), referred to previously, show that myelin formation is greatly affected by the functioning of the tracts, apparently influenced by the impulses passing through them. Lecithin, which, as was stated previously, forms the chief component of the myelin sheath, in physiological salt solution shows striking kataphoretic properties. This has been shown by Hermann (’97), who described its passage to the anode as one of the most astonishing microscopic phenomena he had ever witnessed. Placing a part of a peripheral nerve of a frog in line with the electrodes (but not touching them), and passing a constant current, he saw a great outflowing of the nerve contents — particularly, but not exclusively, the myelin — toward the anodal pole of the nerve. Here the myelin heaped up as curl-like particles, but, on reversal of the current, was reabsorbed by the nerve, while the myelin at the opposite pole (now anodic) collected in heaps at the end of the nerve. Such an experiment shows clearly the tendency of the myelin to be carried in the direction of the anode. _ In the living nerve, then, the action current, which implies considerable galvanic potentials at the surface of the nerve, might cause the myelin constituents (in so far as they originate from the neuraxis or from the sheath cells) to gather at the surface, thus forming a sheath about the fiber. The above facts offer a reasonable solution of the problem regarding the production of myelin and Its appearance at the periphery of the fiber {Ariens Kappers). However, these explanations do not make clear the reasons why the telodendria are unmyelinated.

In connection with the accumulation of myelin about the neuraxis must be mentioned a fact which comes to the attention repeatedly in the study of certain fiber tracts of lower animals {Ariens Kappers, ’20). In the conunissura superior habenularum of plagiostomes, medullated fibers are frequently arranged at the periphery of non-meduUated ones.

The same is evident in the fasciculus retroflexus. It suggests that, through induction, the periphery of the bundle is strongly anodic, and so is responsible for the formation of myelinated fibers. Whether this is to be regarded as analogous to the peripheral accumulations of myelin around the individual neuraxes is problematical but at least is worthy of consideration.

Synaptic Relations between Neurons

Telodendria of the neuraxes of neurosensory cells are generally in s3Tiaptic connection with the primitive ganglion cells. Sense cells, or neuroepithelia, support the terminal fibrils of sensory fibers but are not in themselves neurons. Primitive ganglion cells may be discrete (Bozler, ’27) or may appear to be connected by processes, the cells forming a non-polarized, or at least a transiently polarized, syncytium in which

the neurofibrils pass from one part to another {Apathy, '97). A distinct barrier exists between neurons of the higher nervous systems, a region which is spoken of as the synapse, which is demonstrable physiologically and, in some cases, anatomically.

According to certain observers, actual protoplasmic continuity is often, if not always, present in the region of the synapse. Apathy (’97, '98, ’07, and elsewhere), Bethe (’01, ’03, ’04, and ’08), Wolff (’03), Prentiss (’04), Dogiel (’05), Bielschowsky (’05, ’28), London (’05), Schultze (’05), Oudendal (’12), Held (’09a, ’29), and others considered that neurofibrils may pass from one cell to another cell without interruption (fig. 19). Beccari (’07) suggested a similar condition where the terminal fibers of the crossed neuraxis of the Mauthner cell in teleosts synapsed with the cells of the ventral horn. Marui (T8) claimed a similar continuity of neurofibrils in his study of the Mauthner cells of teleosts. He thought that the neurofibrils of one neuron may be continuous with those of another but that the perifibrillar protoplasm is very poor near the synapse, the fibrils being surrounded closely by glia. Tiegs (’31) believed himself to have demonstrated neurofibrillar continuity across the synapse in various nerve cells, including Mauthner’s cells. Bartelmez (’15, ’20, and, with Hoerr, ’33), working with the Mauthner cells of fishes, believed that his material demonstrated clearly that in these cells actual cell membranes intervene between the regions in synaptic relation with each other (fig. 20). Direct evidence of an anatomical barrier at the synapse has been presented only in rare cases. However, a considerable body of observers regard the synapse as a region where the cell processes are in contact with each other but are not continuous — “ contiguity without continuity.” Based on the embryological work of His (’90) — who showed that the nerve processes are outgrowths of the nerve cell and who regarded the neurons as an entity — and on the confirmation and acceptance of this observer’s work by various embryologists (see Streeter, ’12), the idea of contiguity but not continuity has found strong support in the histologic and cytologic studies of Golgi (’82, ’83, ’85, 06, 07, etc.), Ramon y Cajal (’06, ’07, ’09, etc.), and many others. Indirect but supporting evidence for the integrity of the neuron is to be found in the work of Harrison (’06, ’07, ’10, and elsewhere), of Lewis and Lemis (’12), of evi { 17), and of their associates, and in some of the experimental and neurosurgical work on nerve degeneration and regeneration (Waller, ’52 ; Ramrier, ’78 ; Howell and Huber, ’92; Stroebe, ’93; Huber, ’95, ’00, ’16-’17, ’19, ’20, ’27; Perronnto ’07; Poscharissky , ’07 ; Ramon y Cajal, ’08, ’08a, and ’28; Ranson, 12; Boeke, ’16 and ’17; Dustin, ’17 ; Huber and Lewis, '20, and others), since the results of their work in general favor the development of the neuraxis as a do^vngrowth of the nerve cell.

Fig. 19. The continuation of the terminal fibrils of the neuraxes of the basket cells into the intracellular fibrils of the Purkinje cells. Oudendal.

Fig. 18. Cells from the nucleus trapezoideus of the cat. Veratti. Different types of end baskets (from the Edtnger textbook, ’08) .

Fig. 20. Mauthner cell of a bony fish. Barlelmez.

Note the axon cap surrounding the axon hillock and the end feet of Auerbach along the lateral dendrite.

Toxins are particularly effective at the S 3 mapse and the susceptibility to fatigue is greatly increased. There is considerable difference of opinion as to the way in which a synaptic membrane (or membranes) might function. It has been regarded by some observers as semipermeable in character and many physiologists favor the theory that at the passage of a nerve impulse, some substances, possibly ions thus dissociated, are able to pass in one direction and in one direction only.

Whether or not there is a continuity of nerve substance at the S 3 mapse, there is strong evidence of physiological differences in the region, which indicates the presence of a physiologic if not an anatomic barrier. Whether or not neurofibrillar continuity may occur across the synapse (and it is conceivable that this may vary in different animals, or even in different regions in the same animal), it is probable from the developmental history and from the work on nerve degeneration and regeneration that each neuron is fundamentally a morphologic and trophic unit.

A study of the sequence of the phylogenetic and ontogenetic development of the nerve processes and the direction of their growth affords some suggestions regarding the factors underlying the polarization of the S 3 mapse. Thus neurosensory cells have no dendrites. Likewise neuroblasts, from which the true neuron arises, form their neuraxes first by an outgrowth of the cell protoplasm, this process growing in the direction along which the current is passing {Bok, ’15). Its growth, then, is stimulo-concurrent and begins at the side of the neuroblast opposite the receptive pole, as it does also in the neurosensory cell. Considerably later, at about the time that the tigroid substance appears in the cell body, the ordinary protoplasm of the cell shows a tendency to form processes toward the source of its stimulation (stimulopetally, Ariens Kappers, ’07, and elsewhere). “ The difference in development of these outgrowths is very striking, since the neuraxis grows away from its source of stimulation, while the dendrite tends to develop in the direction from which it receives its stimuli.’" The outgrowth of these processes in opposite directions has a suggestive resemblance to the polarized growth in the electric field, and indeed may have its basis in polarized bioelectric conditions (see the discussion of neurobiotaxis). It seems quite possible, then, that where two different bioelectric poles lie in close relationship to each other, as at the synapse, a double surface membrane (in the sense of Helmholtz and Quincke) may be formed, through which only such a current may pass as agrees with the bioelectric polarization of both surfaces ; that is, with the stimulo-concurrent bioelectric character of the neuraxis and the stimulo-petal character of the dendrite. Thus, although the neuraxis or the dendrite stimulated along its course may conduct in either direction — because such a fiber is homogeneous in itself — yet where two different polarized surfaces lie in contact, the transmission of the impulse can take place only in a direction in accordance with the polarizations in the region. The above explanation of the condition at the synapse, on the basis of the oppositely polarized character of the offshoots, does not necessarily exclude the presence of some substance such as the so-called junctional substance or the synaptic membrane of Sherrington (’06). Such a substance, however, if it is not a product of the glia, which is highly improbable, may be produced by, or be a part of, the neuron itself. Its formation at this region may be due to the meeting of the two differently polarized layers.

It is to be noted that the cell body of the neuroblast itself, up to the time that it has fully

developed Nissl substance, tends to keep its original position and so to behave like a neuraxis. After the development of the Nissl substance, however, the cell body tends to shift along the dendrites in the direction from which it receives its more important stimulation.

It is evident that the stimulo-concurrent character of the neuraxis and the stimulo-petal character of the dendrite and cell body involve their approaching at the synapse (see fig. 46), and each transmission of conduction tends to increase the proximity of their mutual approach.

Fig. 21 The arch of Frommann, showing intramedullary relations on the right and extramedullary relations on the left.

It is well known that at the synapse the rate of conductivity lowers, and also that this place is more susceptible to fatigue and to drugs. This is due, perhaps, to the fact that the most peripheral fibrils of the telodendria are closely surrounded by a glial reticulum {Marui, ’18), the glia being very susceptible to metabolic changes and to drugs. Textbooks of physiology should be consulted for further accounts of the characteristics of the synapse. An excellent discussion of the problem of nerve conduction and of the synapse, from the physico-chemical standpoint, is to be found in the work of R. S. Lillie (’18, ’19, ’23).

Individual neurons are joined into larger units which are termed cell complexes (coelenterates), nerve strands (worms), ganglia (worms, arthropods, and mollusks, as well as vertebrates), and central and autonomic nervous systems (vertebrates). Cell complexes, formed of single primitive ganglion cells, are found, in lower invertebrates. Those plexuses which are not formed of single ganglion cells and their offshoots, but the trabeculae of which contain bundles of nerve fibers and many nerve cells, as is the case with most invertebrates, are termed cell strands or cords. These strands may contain primitive ganglion cells as well as neurons and processes of neurosensory cells. A somewhat similar arrangement is seen in a sympathetic plexus — such as the coeliac — except that the cellular elements of such plexuses are typical neurons.

By ganglia aie understood accumulations of nerve cells. They are present in invertebrates as far down as worms. Spinal ganglia are examples of them in the vertebrates. Such ganglia may consist of cell bodies, but they become more complicated with the increase in neuropil, which is composed of the processes of the cells and especially the finer branches which weave backward and forward among the cell bodies of the neurons. The term neuropil is synonymous with the intercellular gray of Nissl and refers to a diffuse meshwork of dendrites and

Fig 22 Neuron of a sympathetic chain ganglion surrounded by its capsule Ramdti y Cajal A, small subcapsular dendrite ; a, postgangbomc neuraxis; 6, a dendrite surrounded by the spiral fibrils of a preganglionic neuraxis; c, capsule cells.

neuraxes. Such a meshwork has only the semblance of being diffuse. Actually it consists of fibers which establish precise connections, since only such an arrangement could explain the sharply defined conduction which occurs within it. These ganglia tend to join into larger groups — to centralize. This is particularly evident in the ganglia of arthropods, mollusks, and insects, and leads to the formation of a type of central nervous system.

The term central nervous system is generally applied where not only anatomic and histologic, but physiologic relations as well, indicate a high degree of centralization of nervous function. This centralization is present during the entire ontogenetic development in vertebrates, where the nervous system is a large unit from the beginning. Centralization may arise gradually during ontogeny in invertebrates, through the development of separate primordia. The term central nervous system is usually confined to that of vertebrates. Neuropil, somewhat more evenly distributed, is present in the vertebrate central nervous system.

The autonomic nervous system is considered on pages 233 to 242, and will receive only brief mention here. It consists of chain ganglia (Grenzstrang) situated on either side of the vertebral column, of large sympathetic plexuses called collateral or prevertebral ganglia, and of small, scattered cell masses on.

28 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

in, or near the organs receiving innervation from them, and termed peripheral ganglia. All such ganglia consist of the cell bodies of postganglionic or sympathetic neurons (figs. 22 and 23), the neuraxes of which, frequently, but not invariably unmedullated, supply heart muscle, smooth muscle, oi glands. All such ganglia, wherever situated, are related to the central nervous system through preganglionic neurons, the cell bodies of which lie in brain or spinal cord and the medullated neuraxes of which terminate in pericellular, subcapsular synapses around the cell bodies of postganglionic neurons. There would appear to be some question in the minds of certain observers with regard to the type

of synapse between pre- and postganglionic neurons. Pericellular, subcapsular endings have been demonstrated by Dogiel (’99a), Huber (’99a), and Ramon y Cajal (’09). The prevalence of this type of neuron synapse in man has been questioned. Stohr (’27, ’28) believed them to be of common occurrence in the superior cervical ganglion, but to be rare in other ganglia in man. The evidence appears to indicate that in frogs only pericellular terminations are found.

The autonomic system, in higher vertebrates at least, falls into a craniosacral or parasympathetic portion, with preganglionics through cranial (III, VII, IX, X, XI) and sacral nerves, and a thoraco-lumbar or sympathetic division, with preganglionics from the thoracic and lumbar levels of the spinal cord. Most organs are known to receive innervation from both systems, the results of such innervation producing different, and sometimes diametrically opposite, effects. Thus in the innervation of the heart, the two neuron, cranial chain — the preganglionics of which run in the vagus to synapse on the heart wall with the postganglionic neurons — produces inhibition, while the thoraco-lumbar

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEjVIENTS 29

innervation — with preganglionics from the uppermost thoracic segments synapsing with postganglionic neurons, the cell bodies of which lie in cervical sympathetic and stellate ganglia — produces acceleration.

The chain ganglia are connected with the ventral nerve roots by so-called white rami communicantes, through which the neuraxes of the preganglionic neurons and the visceral sensory fibers pass. The strand by which fibers leave a chain ganglion for other sympathetic ganglia is called a gray ramus, while the strand which connects it with the nerve root is termed the gray ramus communicantis. Gray rami communicantes are present on all of the spinal nerves of man.

The sympathetic system itself is an efferent system, but visceral sensory fibers, with cell bodies in spinal and cranial ganglia, accompany the pre- and postganglionic neurons. Such visceral sensory neurons have their cell bodies located in spinal or cranial ganglia and are comparable to other sensory neurons except for their peripheral and central distributions.

The Connections of Nerves WITH Other Tissues

SENSORY endings

The most primitive sensory

apparatus is found in the neuro sensory cells described in the sensory nerve endings m the skin of Petro preceding pages. These cells are myzon marinus, Retzius. c., cutis ; ep , epidermis, numerous in invertebrates, particularly in the lower types, where they are concerned with general tactile sensibility. In man, neurosensory cells are found in the olfactory epithelium and in the retina. Bipolar sensory cells, with free sensory endings in the skin, are present also in invertebrates and are numerous in vertebrates.

The peripheral endings of sensory nerves, as seen in higher vertebrates, may be either free or encapsulated, and the encapsulated fall into three main subdivisions : those ivith thin capsules, those with thick capsules, and those in which the capsule contains tissue elements. This classification of sensory nerve terminations follows that of Huber (T9).

Of such terminations, the free sensory endings appear^ to be phylogenetically the oldest. In such a free sensory termination, the fiber, if medullated, branches repeatedly at the nodes of Ranvier. The smaller branches lose their medullary

30 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

sheaths and the finer telodendria distribute throughout the area, which may be relatively large. The figures presented are from Retzius (’92) and Huber (’00) (see figs. 24, 25). In the preparation from which the latter figure was made, the

Fig. 25. Free sensory ending in the mucosa and epithelium of the urethra of a cat. G. Carl Huber.

peripheral area of distribution was approximately a millimeter square. According to Vitali (’06) and Stefanelli (’15 and ’16), these are not really endings but form an anastomosing, unmedullated network. However, other observers do not agree with this interpretation. Heringa (’18) often found annular endings between epithelial cells. In the lowest vertebrates, these simple sensory endings are the only endings in the skin, as Retzius (’92) demonstrated for the lamprey.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 31

Johnston (’09) found similar endings in the intermuscular septa in fishes. They are found in various regions in higher animals, as for example, in the cornea, meninges, gingiva, adventitia of the blood vessels, and intestines, and in general on surfaces which are only sensitive to great differences in temperature or to painful stimuli, or show general tactile sensibility. Thus free sensory endings are found particularly on those fibers directly concerned in the conduction of such primitive perceptions as deal directly with the preservation of the life of the animal and which are termed protopathic {Head, Rivers, and Sherren, ’05) or vital perceptions {Fabrilius, ’10 ; Ariens Kappers, ’20). These stimuli are often disagreeable ones, being concerned with an appreciation of stimuli more or less injurious in character; consequently the reflexes produced by them are often negative reflexes. The free endings {Sherrington, ’06) act principally as noxireceptive endings {nocere — to injure). However, it would be an exaggeration of this function to consider them only as noxireceptive endings, for lower animals, as well as higher ones, receive stimuli which are agreeable to them and which lead to positive reactions. Thus lower animals often exhibit positive reflexes to stimulations from skin surfaces having free sensory endings ; such terminations must, therefore, be, in part, gratoreceptive.^^ Generally speaking, however, such endings carry impulses in which the element of personal welfare (euphoria) is an important factor ; thus the term protopathic *- or vital sensibility may be applied to all of this group. The work of Ranson (’15, ’30 ; but see Sheehan, Anat. Rec., vol. 55, ’33) and others has indicated that free sensory endings, at least in so far as they receive painful stimuli from body surfaces, are, for the most part, endings of unmyelinated nerve fibers. The work of Windle (’26) and others indicates that endings of this type distributed to the head may be terminations of myelinated nerve fibers. In addition to painful stimuli, a poorly localized (dyscritic, Parsons, ’27) sense of touch is transferred by such fibers and probably also the appreciation of very high or very low temperatures, which, from their importance for the welfare of the body, fall into the category of vital perceptions.

Beginning at least as far down as amphibians and reptiles, more complicated sensory terminations make their appearance. This greater complexity of structure goes hand in hand with a finer differentiation of perception. It implies the termination of nerves in relation to other structures, such as special epithelial cells, connective tissue or other tissue elements, specifically arranged to further in some way the specialized endings of the nerves and thus assist perception.

In forms below mammals, the most highly specialized terminations are found in the tactile end-organs of birds, which are known as the corpuscles of Grandry {Grandry, ’69) or Grandry-Merkel {Merkel, ’75), the corpuscles of Herbst (’48), and those of Key-Retzius (’75-’76). The terminations of Grandry, which are present in the skin of the bill of such birds as the goose or the duck, are composed of two or more cells, surrounded by a connective tissue capsule. The cells are

"This word, introduced by Ariens Kappers (’20), is derived from gratus, meaning useful,

'^’"““From protos, meaning that which appears first, and pathos, meaning that which affects. This nomenclature originated with Head, Rivers, and Holmes.

32 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

regarded by some observers {Izquierdo, 79) as of epithelial origin ; by others {Szymonowicz, ’97) as derived from the connective tissue. Such cells are placed side by side within a capsule and have the opposed surfaces hollowed out so that a space intervenes, in which is the so-called tactile disk, where the nerve fiber comes into relation with the cells. Such cells measure loy. X 50jii according to Slohr (’28), and have a fibrillar structure which has been studied by various observers {Nowick, ’10, and others). The nucleus of each cell lies at the inner border near the tactile disk. To these disks, trabeculae of connective tissue extend from the capsule. Dogiel (’91, ’04), Dogiel and Willanen (’00), Heringa (’17a), and Boeke (’25a) believed that the terminal unmedullated fibers break up into finer fibrils in the disks and penetrate the cells. These become continuous with the intracytoplasmic fibrils of the tactile cells {Heringa, ’17a; Boeke, ’25a; Lawrenijew, ’26, and others) . Other workers considered that a fine, unmedullated fiber forms a pericellular plexus inside the capsule {Dogiel, ’04; Botezat, ’06). Degeneration involves both cells and fibers {Schafer, ’12). The corpuscles of Herbst (’48), similar to the Vater-Pacinian corpuscles {Valer, 1741 ; Pacini, 1836 and 1840) of mammals, are found in the skin of the bill region and in the tongue and palate of water birds (see fig. 24C). Such corpuscles have a thick lamellar capsule of connective tissue and a central core of cubic cells. The fiber passes through the central part of the core to end in an enlargement at its outer end. It may or may not show branching. The Key-Retzius corpuscle {Key-Retzius, ’76), intermediate in type between the Herbst and the Vater-Pacinian, is found in the bill of certain water birds.

In mammals there is a great variety of sensory nerve terminations and only a brief account of certain more generally recognized types can be given here. A primitive type of connection in relation with the epithelium is found in the tactile menisci of the mouse (fig. 26A) and the pig. Three well-differentiated terminations {Dogiel, ’92, ’93) bear great structural resemblances to each other, the differences being mainly in their distribution and size. These are the genital corpuscles, the spherical end bulbs of Krause, and the tactile corpuscles of Meissner (’53; see fig. 26B). All three of these terminations have a thin lamellar capsule into which the nerve fibers enter. Within this capsule, at least in Meissner’s corpuscle, lie cells {Krause, ’82) which are oriented at right angles to the length of the corpuscle and which are termed either tactile cells or “Kolbenzellen ” by various authors (see Stohr, ’28). The number of nerve fibers varies with the size of the corpuscle ; thus the smaller spherical end bulbs of Krause and Meissner’s corpuscles receive only one fiber, larger corpuscles of these two types and the smaller genital ones receive two or three such fibers, and the larger genital corpuscles may ha.ve eight or ten medullated fibers entering them. There is difficulty in determining positively the extent of division of a nerve fiber before entering a large genital corpuscle ; the numbers here given, therefore, are not above question. The medullated sheaths disappear either before they enter or at their entrance into the spherical end bulbs of Krause, but are retained by fibers passing to tactile and genital corpuscles until after their entrance into the corpuscle. In all three types the fiber (or fibers, as the case may be) forms two or three spiral turns within

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 33

the corpuscle and then breaks up into numerous fine branches which certainly interlace and possibly anastomose. Dogiel (’91, ’92, ’93) appears to have shown that terminal branches of the nerve fibers form a periterminal network which is continuous with the intracytoplasmic fibrils. The genital corpuscles {Dogiel, ’93 ; Huber, ’19) are the largest of the three, but vary in size from .04 to .1 mm. in breadth and .06 to .4 mm. in length. They are distributed to the external geni

Fig. 26, A. Tactile disks and tactile cells associated with a tactile hair of a mouse,

B. Meissner’s tactile corpuscle from a human finger tip,

C. Herbst corpuscle from the bill of a duck,

D. Karobshaped, richly looped terminations of the neurofibrils in a Pacinian corpuscle from the mesentery of a cat, E. van der Velde.

talia, such as the glans penis, prepuce, and clitoris. They are sometimes lobulated. The tactile corpuscles of Meissner (Wagner, '52 ; Meissner, '53 ; Dogiel, ’92 ; von Kolliker, ’89-’02; Szymonowicz, ’95; Ruffini, ’02; van der Velde, ’09; Ramon y Cajal, ’09; Botezat, ’12; Schafer, ’12; Heringa, ’17, ’17b, ’18, ’20; Htiber, ’19; Boeke, ’25a, and others) are an irregular peanut shape, varying from .02 mm. to .03 mm. in breadth and averaging approximately .08 mm. in length. They are found in the connective tissue papillae under the epidermis of the skin, being particularly plentiful in the skin of the hands and feet, especially that of the index finger. They also occur at the tip of the tongue, around the lips and eyelids, on the outer surface of the forearm and in the nipple region. The end bulbs of

34 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Krause also are found in the lip region and in the oral cavity, in the conjunctiva of the eyes, around the cornea, in the region of the external genitalia, and possibly elsewhere. They vary from .02 to . 1 mm. in length ; an average width is .02 mm., although some of them may attain four times that diameter. Larger fibers from the network inside of the genital corpuscles, and from the corpuscles of Grandry and Meissner as well (Dogiel, ’92), may occasionally leave one corpuscle. Such a fiber may terminate in other corpuscles (at least with the genital terminations) and in the skin (fibers of Dogiel). Meissner’s tactile corpuscles, and perhaps other encapsulated nerve terminations, may show an unmyelinated nerve fiber (fiber of Timofeew) as well as the usual myelinated nerve fiber. Dubrenil (Ariens Kappers) advocated the theory that the fibers of Timofeew have to do with painful overstimulations of such corpuscles. It is supposed that their overstimulation in cases of causalgia (Weir Mitchell) causes excessive pain. As this pain disappears after periarterial sympathectomy, it would appear that they carry general visceral sensibility (pain) and are not concerned in higher perception (Le Riche). This, however, has not been confirmed as yet. In the human finger tip, Perez (’31) found peculiar, complicated endings with fibrils to the basal stratum of the skin near a tactile meniscus. It is probable that Meissner’s tactile corpuscles serve for finer discrimination and thus are concerned in epicritic or gnostic sensibility (eucritic sensibility. Parsons, ’27), since they are most abundant in places where such discrimination is best developed.

Examples of sensory terminations with thick capsules are the corpuscles of Herbst and Key-Retzius and the Vater-Pacinian corpuscles (see fig. 26), all of which are more deeply situated than the ones just described. A brief accoimt of the Herbst and Key-Retzius corpuscles has been given earlier. The Vater-Pacinian corpuscle (Vater, 1741 ; Pacini, 1836 and 1840) has a very thick lamellar capsule, the lamellae numbering between twenty and sixty and consisting of connective tissue fibers arranged in concentric layers with a small amount of lymph between each two lamellae. Schumaker (’ll) was of the opinion that an actual capsule was formed around the fluid and that the lines indicating the lamellae were, in ordinary preparations, really double walls where two capsules came together. Plate-like cells, sometimes regarded as connective tissue cells (Huber, ’19) and sometimes as endothelium (Schwalbe, ’87), lie in interlamellar spaces, in which lymph is found also. P acini showed that the outer capsule layers represent a continuation of the connective tissue sheath of the nerve fiber. The inner lamellae have been added later. A nerve 6ber — large and medullated and surrounded by a little endoneurium — enters each corpuscle and retains its medullary sheath until it reaches the core. Here the sheath is lost and the fiber runs through the core, which consists of a dark staining protoplasmic mass, and, after branching several times at the end, forms terminal enlargements consisting of a dense neurofibrillar network and end disks. Sometimes, in passing through the core, the fiber divides into large branches and a modification of the capsule occurs. In preparations from the pancreas of the cat, Relzius (’98b) described small side branches on the main fiber of the Pacinian corpuscle as it passed through the core, each branch terminating in an enlargement. Sola (’99) studied these corpuscles in the mes

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 35

entery of the cat and found that there the fiber breaks up into branches which ascend and show enlargements near the peripheral pole of the corpuscle. Great variations are foimd in the final termination of the fiber within the core. At the end of the main fiber may be a simple enlargement or it may break up into ramifying branches (Sala) ; also, it may continue through one corpuscle, pass out, and enter a second. Dogiel (TO), van der Velde (’07), and others were able, by the use of various special methods, to demonstrate neurofibrils within the main fiber and its branches. Numerous observers (Eohner, ’05 ; vander Velde, ’09 ; Dogiel, ’10) Boeke, ’25a, and others) described fine fibrils leaving these enlargements and entering into the cytoplasm of the core. A satellite fiber (Sokolow, ’99), quite independent of the main fiber, forms a plexus about the core. A rich capillary plexus lies in close relation to the nerve fiber at its place of entrance into the corpuscle. According to Stohr (’28) there are no capillaries within the core. The Pacinian corpuscles are visible to the naked eye and were first described from gross dissection {Valer, 1741). They are widely distributed, being found in the subcutaneous fat, in the deeper parts of the dermis (particularly of the hands and feet), aroimd bones and joints, in the septa between muscles, and even in muscles and tendons. They sometimes occur on nerve plexuses such as the solar plexus, and they are very plentiful along the nerves in the abdominal mesentery and peritoneum (for the course of these fibers to the spinal ganglion see Sheehan, ’32). They have been demonstrated on the nerves to the clitoris, penis, urethra, mammary glands, and nipples, on the intercostal and sacral plexus nerves, and cutaneous nerves to the upper extremities. These corpuscles probably are concerned with the perception of pressure and tension and possibly with stereognostic sense.

Muscles and tendons have (1) free sensory endings, (2) muscle spindles or neuro-muscular terminations, and (3) neurotendinous terminations. Simple inter-muscular endings are present in petromyzonts {Johnston, '08) and comphcated end-trees and nets have been described in the muscular coats of the trachea, stomach, and intestine (Ploschko, ’97 ; Carpenter, ’18). Basket-like sensory endings on muscle were seen by Retzius (’91) in Myxine and by Giacomini (’98, ’98a) in sharks, bony fishes, and amphibians (“terminaison en panier”) ; still more complicated brush-like endings occur in certain fishes. Most workers have not seen muscle spindles in forms below tailless amphibians, although Allen (’17) found them in the caudal heart of cyclostomes. Ranidn y Cajal (’88), Sihler (’95), von Kdlliker (’96), Huber and DeWitt (’98), Dogiel (’02), and Hines (’30) have described muscle spindles in tailless amphibians. Such spindles are typically encapsulated endings containing intrafusal muscle fibers and receive a medullated nerve fiber (or rarely 2) breaking up in the usual fashion and also an unmedullated fiber {Huber and DeWitt, Hines, and others) which is sometimes thought to be motor and sometimes sympathetic in function. Hines (’30) stated that in the sartorius muscle of the frog “the non-medullated fiber is found lacking only after the removal of the abdominal sympathetic chain. She found a motor type of polar ending in certain spindles of the frog pectoralis abdominis. Ranvier ( 78), Sihler (’95), Perrondto (’01), Huber and DeWitt (98), Kulchitsky (24), Boeke

36 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

(’22, ’27, ’30), and Hines (’30) described muscle spindles in reptiles. The dense, two-layered, Wous tissue capsule has from 1 (python, Kulchitsky; boa, Hines), 2 to 5 (alligator, Hines), or 2 to 8 (tortoise, Huber and DeWiit) intrafusal muscle fibers, striated throughout, although with increased nuclei in the equatorial region. A heavily medullated proprioceptive fiber, after entrance to the capsule, divides into two branches which coil around the fiber. A second thinner fiber, unmedullated or thinly medullated, forms a polar ending within the spindle. This latter has been regarded as sympathetic by Kulchitsky (’24) and motor by Perroncilo (’01), Hines (’30), and others. Boeke (’27) believed that the intrafusal fiber of certain reptilian muscle spindles receives both motor and sympathetic innervation. The muscle spindles of birds have been studied by Giacomini (’98a), Huber and DeWitt (’98), and others.

Kerschner (’88), Ramon y Cajal (’88), Ruffini (’92, ’93, ’98, ’98a), Huber and DeWitt (’98, ’00), Baum (’00), Dogiel (’02, ’06), Tello (’22), Langworthy (’24, bibliography. Chap. V), Hines (’27, ’30, ’30a), and Hines and Tower (’28) have described mammalian muscle spindles. Each consists of a small bundle of intrafusal muscle fibers (3 or 4 to 20) surrounded by a connective tissue sheath. In man this ending measures from .08 to .25 mm. in width and from 2 to 10 mm. in length (Baum). These spindles are said to be more numerous in the extremities than elsewhere and to be lacking in certain muscles (for distribution see Hines, ’30). Proximal polar, equatorial, and distal polar regions have been recognized. In the polar regions the muscle is striated and in the distal polar region the fibers appear smaller and more numerous — possibly because of division of the fiber (Schafer, ’ 12) . In the equatorial region the striations are less clear and the nuclei more numerous and the fibers resemble more nearly embryonic muscle. The axial sheath which surrounds the intrafusal fibers is separated from the main capsule by a periaxial lymph space. The main capsule, formed of several concentric layers of fibrous connective tissue, is continuous with the connective tissue over the muscle fibers and fasciculi. These spindles receive an unmedullated fiber (arising from a medullated fiber) which terminates in the polar region as a “terminaison en grappe” (1 to 4 to a spindle). The Henle sheath around such a medullated fiber becomes continuous with the capsule, the myelin sheath extending to the inner axial sheath. The fibers divide several times in their course and terminate in spirals or annulo-spirals or in disks or flower-like endings, the latter representing terminal branches of annulo-spiral fibers. According to Kerschner (’93), Huber and DeWitt (’98), and Huber (’19) they are the “ terminaisons en forme de fleur” of Ruffini (’98). The sensory function of these neuromuscular endings was proved experimentally by Sherrington (’94), who demonstrated that such fibers did not show Wallerian degeneration when the ventral roots, carrying motor nerves, were cut in the monkey and cat. The “terminaison en grappe” (Dogiel, ’02), sometimes described as sensory, sometimes as motor, and again as sympathetic, is considered later (p. 42).

Neurotendinous endings (fig. 28), like neuromuscular endings, contain tissue elements within the capsule. Golgi (’80) first described them, and the terminations are often called the organs of Golgi. According to von Kolliker (’89-’02)

Fig. 27. Neuromuscular terminations. r> j.

^1. Very simple neuromuscular spindle in the cat. oc *c. r j e * i

B. LuromuLuIar nerve end organ from the intrinsic plantar muscles of the dog, from teased

reparation of tissue stained in methylene blue. Hu an

38 NERVOUS SYSTEMS OP VERTEBRATES AND OF MAN

such terminations in man are from 1.28 to 1.42 mm. in length, while their width varies from .17 mm. to .25 mm. That they may be even larger is indicated by

Ciaccio (’90, quoted by Huber, ’19), who

Fig. 28 . Compound neurotendinous sensory end organ from the fascia of the back muscles of the albino rat. Huber and DeWill.

described such a termination with a length of between 2 and 3 mm. The capsule, consisting of several connective tissue lamellae, is continuous at its poles with the connective tissue sheaths around the tendon fasciculi. As in neuromuscular terminations, an axial sheath and a periaxial lymph space are present. The inclosed tendon fasciculi are smaller than elsewhere in the tendon, number from 8 to 15 {Huber and DeWitt, ’00 ; Huber, ’19), and appear embryonic in staining characteristics and the arrangement and number of the nuclei. The medullated fibers branch before and after entering the capsule. After they lose their myelin sheath, they divide still further and then spread out over the inclosed tendon fasciculi and terminate in end disks. In their course over the tendon fibers, short branches are given off which may partly surround the fasciculi. Golgi rSO), Cattaneo(’88), von Kdlliker (’89), Huber and DeWitt (’00), Dogiel (’06), Cilimbaris f’lO), and many others have added to our knowledge of these terminations. It appears probable that the neurotendinous terminations maybe phylogenetically younger than the neuromuscular endings, since tendons, as such, appear to be of more recent evolutionary origin than are muscles. According to Pansini (’89), however, neurotendinous terminar tions are not entirely absent in sela

chians and teleosts, and similar end organs (the end organs of Sachs, ’75, and Rolletl, ’76) have been observed in amphibians and reptiles. Their structure is shown in fig. 28, It is evident that neurotendinous and neuromuscular endings are of great significance as organs for the stereognostic sense ; they make possible a determination of the position of the body and its extremities.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 39

Other sensory terminations in relation to muscle and tendon have been described by Giacomini (’98), Ceccherelli (’04), Dogiel (’06). Such consist of fine branches to the region near the tips of the muscle fibers and to the tendons near their attachment to the muscle. Huber (’99) described a special type of sensory nerve-ending in the extrinsic eye muscles of the rabbit. Huber (’00a) also described a special type of sensory nerve-ending in the tendons of the extrinsic eye muscles of the cat. Small Pacinian corpuscles, approaching in form and structure the cylindrical end bulbs of ICrause; are found in the intermuscular connective tissue septa and joint capsules.

MOTOR AND OTHER EFFECTORY ENDINGS

The motor termination on skeletal muscle is termed the motor plate or soleplate ending. Medullated neuraxes of ventral horn neurons pass out from the central nervous system to the muscle, and, after repeated branchings, terminate as unmedullated fibers within the muscle substance under the sarcolemma, the ectodermal neurolemma of the nerve fiber giving the appearance of being continuous with the mesodermal sarcolemma. The region of penetration, frequently termed Doyere’s elevation, or the sole plate, is characterized by a sarcoplasma containing many nuclei (sole nuclei), which are large and clear and contain one or more nucleoli. Small granular nuclei can also be seen along the terminal fibers (Schafer, ’12). Henle’s sheath appears continuous over the ending and has been called the telolemma (Kuhne, ’86). A single sole-plate ending is found customarily on each fiber. Under the sarcolemma the nerve fiber may break up into smaller branches, and the neurofibrillar elements (first demonstrated by Ram6n y Cajal, ’ll) of the entering fiber may spread out. The amount of branching varies. Around the neurofibrils is the periterminal network described by Boeke (’ll, '21, ’26, ’27, etc.) and also seen by Stefanelli ('ll), Erlacher f'15), Agduhr (’16), Murray (’24), and Iwanaga (’26). Boeke (’26) and Heringa (’31) regarded this substance as transitional or receptive substance (in the sense of Langley) between the nervous and muscular substance. According to Boeke (’26) the periterminal fibrils approach the anisotropic disk of the muscle; however, Heringa (’31) believed he had shown that it is connected with the Y-granules of the isotropic disk. Neither Roefte nor Heringa regarded the periterminal fibrils as neurofibrils (though they stain differentially) but as products of the sarcoplasma.^^ According to Wilkinson (’30a), the periterminal network of Boeke is probably an artifact “due to chemical alteration of the cytoplasm of the sole plate, produced by prolonged fixation in formalin.’’ He presented as evidence studies of these endings by various techniques and WeiVs results (’29) on the effects of formalin fixation on lipoids of the central nervous system. Occasionally a delicate network is formed around a muscle fiber by the terminal branching of a delicate unmedullated nerve fiber which has separated from the sole-plate ending. Such nearlying little nerve fibers are called ultraterminal fibers. Sometimes the network

“This might be considered as additional proof that fibrillar structures may arise in other than nerve cells, as is shown by the neurophanes of ciliates and fiagellates by the Kofoid school {Ariens Kappers).

40 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

may be very rich, as in snakes {Boeke, ’22), but it is found very seldom in man {Boeke, ’ll, and Iwanaga, ’26). Ultraterminal fibers have been seen by Perroncito

(’01), Ruffini (’00), Crevatin (’01), and Cilimbaris (’10).

Bremer (’82 and ’83) traced unmedullated nerve fibers into the sole-plate endings of frogs and lizards. Grabower (’02) noted the entrance of thin, unmedullated nerve fibers into the soleplate ending, but showed that the fibers which he was describing were branches of motor nerve fibers. Perroncito (’02 and ’03) traced immedullated fibers which entered the motor ending and broke up into a number of branches. He was not able to discover their final termination. They had approached the sole-plate ending within the Henle’s sheath of the motor fibers. Perroncito (’02) regarded them as sensory, but later suggested for them a sympathetic character ('03, quoted from Boeke, not available). Gemelli (’05) believed he could trace these fibers of Perroncito (’02) to their ultimate termination where they became continuous with branches of the motor fibers. Botezat (’06) described both medullated and unmedullated fibers to the voluntary striated muscle fibers of birds. In his series of studies on the innervation

«. rectua muscle of a caf t

motor ending (m/) and accessory nerve fil>er (a/.). Lower (*09, '10a, '10b, H;

Iigure from the tongue muscle of a mouse. Boeke, *12, '13, '15, '16, '17, '21, '22,

Motor nerve terminations in striated voluntary

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 41

’20, ’27, ’30) described accessory nerve fibers, unmedullated and with small hypolemmal nerve endings. These he regarded as sympathetic fibers. He did not regard them as homologous to the Perroncito-Gemelli fibers (Boeke, ’13). Boeke’s results are based on studies of a considerable range of material, including that from mammals and forms below mammals and from both embryos and adults. In his work he has employed both normal and experimental methods. Among his attempts to establish the sympathetic character of the accessory fibers by physiologic experimentation, are Boeke' s (’13) operations on eye muscle nerves in the eat. In normal preparations of these eye muscles, Boeke identified sole-plate endings, epilemmal sensory endings, and hypolemmal terminations of his unmedullated accessory fibers, which he regarded as sympathetic terminations. Three and a half to four days after cutting the eye muscle nerves, these latter endings remained intact, although the motor and sensory endings had degenerated. Three weeks after a similar operation, not only had the motor and sensory endings disappeared, but also most of the terminations of the accessory fibers as well as their fibers, although some unmedullated fibers remained. Removal of the superior cervical ganglion, with the eye muscle nerves intact, appears to have led to a decrease, but not to a total disappearance, of the accessory fibers. Soleplate and sensory terminations, of course, were intact in this second type of experiment. These experiments upon the extrinsic eye muscles of the cat are somewhat inconclusive, for when sufficient time had elapsed after resection of the oculomotor, both medullated and unmedullated fibers suffered degeneration, while removal of the superior cervical ganglion, in his opinion, appears to have decreased but did not eliminate the accessory fibers. With regard to the work on the oculomotor, the results of Hines, reported at the 1930 meeting of the American Association of Anatomists, are of interest. She found, in a study of the extraocular muscles of the rabbit, that “when the oculomotor nerve is severed at the base of the brain, only nerves supplying blood vessels remain.” Thus the extravascular innervation of these muscles apparently is from the oculomotor alone. This is substantiated further by the fact that in the rabbit, at least, removal of superior and middle cervical ganglia affected no terminations in these muscles except certain ones on the blood vessels. Boeke and Dusser de Barenne (’19) showed intact accessory fibers to the intercostals along with degenerated motor and sensory fibers, following extirpation of the roots and spinal ganglia of the sixth to ninth thoracic segments on the right side of a cat. Corroborative evidence was offered by Agduhr (’19a) and by Kuntz and Kerper (’24). These latter observers used dogs and studied the eighth intercostal muscle four weeks after the seventh to ninth thoracic nerves were sectioned proximal to the rami commimicantes. They found the motor and sensory fibers degenerated while the unmedullated fibers (accessory fibers) and their terminations remained intact. Various other observers {Aoyagi, ’12 ; Terni, ’22 ; Iwanaga, ’26 (in a few cases) ; Kuntz, '27 ; Hines, in skeletal muscle according to Weed, '27, and others) have described accessory fibers to voluntary striated muscle.

But while a considerable number of workers favor Boeke s idea of a sympathetic innervation of striated muscle fibers by means of his accessory fiber and

42 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

its teniiinations, an almost equal number are opposed to such an interpretation, or at best are unwilling to accept it as proven, Wilkinson (’29) questioned Boeke’s results, basing his criticisms, to a considerable extent, on a study of the latter observer’s preparations. He (’30 ; see also ’30a) followed this by later experimental work on the innervation of the eye muscles of the cat, using in all some twent 3 '-one animals, and varying the time permitted for nerve degeneration from 3 to 55 days. He found that the motor endings degenerated in about three days, while the sensory endings, which were more resistant, persisted for at least five

days and four hours after the operation. Wilkinson interpreted certain small endings described by Boeke (’27) as present on the eye muscles after the cutting of the nerve, as remains of terminations of proprioceptive fibers. Wilkinson (’29) stated that, in his opinion, sympathetic innervation of voluntary muscle is only for the supply of the blood vessels of that muscle. Boeke (’30) rephed to 11 ilkinson s 1929 paper, emphasizing again the presence of accessory fibers. Certain other observers have failed to verify the presence of the accessory nerve fiber of Boeke. Neither Slefanelli (’12) nor Tello (’17) were able to confirm the presence of these fibers. Ramon y Cajal (’25) could not demonstrate such an inner\ation and emphasized particularly the difficulty of distinguishing an unmedullated branch of a motor neuraxis away from its origin from the unmedullated sjmpathctic fibers. Woollard (’27), who studied a great range of normal, pathologies , and operated material by various methods, obtained negative results with regard to a dual innervation of voluntary striated muscle, except in the case of eye muscles ; there he believed he was able to demonstrate unmedullated sympathetic 1 ers o snia n^usc e fibers, while the large muscle fibers were innervated by solep ate endings Thus his results are unfavorable to Boeke’s idea of a sympathetic ° striated muscle. They favor, rather, the idea of Kul snindles of° fho t t i' Tower (’28), in their study of the muscle

to the intnf / to support a sympathetic innervation

Itriat dfinH the innervation of

the blood vessel Postganglionic fibers from the chain ganglia to never sel , : %K but stated that “the author has

muscle fibers A. furth ^ P^oxus to end hypolemmally in skeletal

votonrv Itriateil f f ^^^®tion of the dual innervation of

bttsRd panieuhTv '/i! 241 to 242. Those readers

should refer to the mn “^ question of the existence of the accessory fibers to their illustrations. ^ above, to their bibliographies and, particularly,

“terminaison en^^grapp" ”^has°h sole-plate ending, another termination

’79; //aitrand DeWilt ’’OS- n ^^"7 ^^^‘^"bed by various observers (Tschiriew, tain workers (A'ZSa ’02, and others). By cer

tain workers (A'ldc/iite/-,, 'OA n 1 ’ 1 / ’ 02, and others). By cer rogarded as terminations’ of «’ ’25, “terminaison en grappe’’ were

them is certainly inconclusive^^Otll^^’'^ neurons, but the evidence offered by motor endings and D/W.,; 7>r\o\ ’ have regarded these terminations as

forms of the motor end nlst f’29) believed them to be immature

as. ndmgs, at least very similar or possibly iden

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 43

tical with these, have been described in certain muscles — as for example, in the extrinsic eye muscles — as sensory {Huber, ’99 ; Wilkinson, ’29). It is not impossible that in certain muscles such terminations may be sensory and in others motor (see Hines, ’27) ; if sensory, their position is epilamellar, if motor, hypolamellar.

The above account of the motor or sole-plate ending is based almost entirely on conditions as found in mammals. However, such terminations are present not only in vertebrates below mammals but even in invertebrates. In fact, motor terminations were observed first in the muscles of insects by Doyere (’40), and it was not until about twenty years later that they were described in amphibians by Kiihne (’62) and in reptiles, birds, and mammals by Rouget (’62). A considerable amount of study has been devoted to these endings in invertebrates and lower vertebrates, of which only the briefest mention can be made here.

In crustaceans, usually the unmyelinated fibers branch and terminate in free endings, but may have more highly differentiated terminations suggestive of those in mammals (Rctefus, ’90). In worms, the fibers terminate in very simple, small plates on the muscle fiber, each terminal nerve fiber supplying a large number of muscle cells. Hansen (’81), Heymans (’89), and Burger (’91) have contributed to our knowledge of these terminations. Methods employed in the study of motor endings in mammalian muscles have not been used successfully, for the most part, in the study of insect material and, although perimuscular plexuses (R. Monti, ’91) have been stained by Ehrlich’s methylene blue method, certain of the finer details of the terminations are as yet unknown. In the leg muscles of insects, endings similar to those in mammals have been described ; the innervation of the wing muscles shows marked peculiarities, the branches of the nerves forming a latticework about the whole muscle cell.

The terminations in the voluntary muscle of lower vertebrates have been studied in representative forms by various observers. Retzius (’91 and ’98a) found them to be relatively very simple in Amphioxus lanceolatus. Here the thorny, unmyelinated nerve fibers, often without division, pass along the muscle fibers, bifurcating once or twice near their terminations, and then each fiber supplies a large number of muscle fasciculi. In lower fishes in general, the motor terminations approach in type those found in higher vertebrates, although there is considerable variation in type even in the same animal, as Retzius ( 91) showed in his work on Mjrxine glutinosa. In this cyclostome, he found an elongated type of termination in the more central portions of the muscle bundle, true terminal motor plates toward the ends of the bundle. Cartilaginous and bony fishes have true motor end plates.

Sillier (’95), Ramdn y Cajal (’88, '09, and elsewhere), Cuccati (’88), Dogiel (’90), Retzius (’92a), Huber and DeWitt (’98), Garven (’25), Hines (’27), and Lawrentjew (’28) have studied the motor end plates in amphibians. Ramon y Cajal and Dogiel both described terminations, not only at the ends of the fibers but following the bifurcation of unmedullated collaterals given off the main fiber during its course. The former observer (’09) regarded the peculiar terminations described by Cuccati as fixation artifacts and the terminal arborizations described by him

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 45

knobs (as is the case with the depressor nerves which in part reach the auricle, Woollard, ’26), or may merely divide and subdivide until they disappear from view{WooUard, ’26). Single fibers may twist and turn to orm curious snarls or ball like terminations {Dogiel, ’98), and encapsulated endings have also been demonstrated (Michailow, ’08, and others). Similarly, in the endocardium and in the valves of the heart, exceedingly richly branched, free sensory endings have been seen by Smiriiow (’95), Dogiel (’98), Michailow (’08), and Woollard (’26). Encapsulated endings may be found (Michailow, ’08) in the endocardium.

The sympathetic innervation to smooth muscle is by way of lateral branches from an unmedullated fiber, which branches terminate in small granules (Huber and DeWitt, ’98; Stohr, ’28, and others). Secretory fibers break up into very fine terminal fibrils which often show distinct varicosities and which come into very intimate relations with the gland cells. There is still difference of opinion among workers in this field as to whether or not the ultimate terminations are intraprotoplasmic. It is entirely beyond the scope of the present account to attempt a discussion of the details of their distribution in the various glands. Those desiring such details are referred to the review by Stohr (’28), in Mollendorf’s “Handbuch der Mikroskopischen Anatomie des Menschen,” where brief accounts of the innervation of various glands and a helpful bibliography are to be found. Reference is made here to the work of Anderson (’92) and Rhinehart (’12) on the thyroid, to the accounts of Retzius (’92b), Korolkow (’92), and Huber (’96) for the salivary glands and that of Dogiel (’93a) for lachrymal glands, and to the descriptions of the terminations in the pancreas as given by Muller (’92), Peiisa (’05), and de Castro (’22), and of the endings in the kidney as figured and described by Sjnirnow (’Ola) and Stohr (’28). Pines (’31) reviewed the histologic details of endocrine gland innervation.

The Ectodermal Supporting Tissue of the Central Nervous System

The central nervous system contains, in addition to neurons, certain elements, largely of ectodermal origin, which are not conductive but are supportive and secretory in character and possibly have other functions as well. These nonnervous elements are more important in vertebrates than in invertebrates, and are better developed in higher than in lower vertebrates. They are distinct from the connective tissue, which is mesodermal in origin. For an account of the nonnervous elements in the invertebrate nervous system, reference is made to the recent excellent book by Hanstrom (’29).

The non-nervous tissue within the central nervous system differentiates along three lines, leading to the formation of three types of structural elements which will be considered in the order named. These are the ependyme, the choroid plexuses, and the neuroglia.

In the fetus, certain ectodermal supporting elements of the central nervous system send processes from cell bodies near the ventricle to the periphery of the cord. These processes branch near the periphery and their terminations anastomose or interlace to form the external limiting membrane. These cells are termed spongioblasts, and they are the anlagen for both ependyme and neuroglia. In

46 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

lower vertebrates, such as fishes and amphibians (Huber, ’03, and others), the supporting elements of the nervous system retain this radial arrangement, but in higher forms, during development, certain of the spongioblasts lose their connections with both the ventricular surface and the external limiting membrane and become neuroglia. Others of the cells retain their position along the ventricular wall. These are the forerunners of the ependyme. At first the processes of such cells extend to the periphery but later this connection, for the most part, is lost. The ependymal cells have one or more cilia on the side towards the ventricle.

Fig. 30. To the right of the figure, ependyme; to the left, glia, from the spinal cord of a human fetus of 14 cm. length, von Lenhossek.

The base of the cell is branched but such branches can be traced only a short distance and are soon lost in the gray of the nervous system. In the region of the posterior median septum and the anterior median fissure the processes still reach the surface (Ranson, ’30) in the adult. According to Ramon (’30), the cilia are lost in the adult human. It appears quite certain that the ependyme may have a secretory function, at least in certain regions. Thus, along the ventricular wall of the thalamus in certain fishes, reptiles (fig. 31), birds, and even mammals, there are patches of tall ependymal cells, overlying a rich capillary plexus and often giving evidence at their surfaces of an albuminous deposit (Ariens Rappers, ’21 ; Rendahl, ’24 ; Holmgren and van der Horst, ’25 ; Huber and Crosby, ’26 ; Hocke Hoogenboom, ’28; Charlton, '28, and others). Granules

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 47

can also be demonstrated in ependymal cells, particularly in developing animals. Wislocki and Putnam (’21) have showed that in the area postrema fluids will penetrate from the blood vessels into the ventricle. Frederikse has demonstrated intercellular substance between the ependymal cells similar to that between choroid cells. All of these characteristics show the close relationship between the ependyma and its specialized derivatives, among which may be included the choroid epithelium, which will now receive consideration.

Certain parts of the brain wall — the roof of the fourth and third ventricles and a part of the walls of the lateral ventricles — contain neither nerve fibers nor nuclear masses and only consist of a single layer of epithelium.

This layer, with the vascular pia, constitutes the tela choroidea, from which infolded highly vascular tufts project into the ventricles and form the so-called choroid plexuses. The choroid plexuses formed by these vascularized tufts of choroid epithelium are relatively larger during later embryonic development (Loeper,

’04) than during adult life. This is in conformity with the fact that choroid plexuses never attain as great a size in higher animals as they do in certain lower forms.

The cells of the choroid epithelium are simple cuboidal in type and are held together by a homogeneous intercellular substance (see fig. 32,

Hexanchus). In embryonic or young stages, they are provided, as are embryonic ependymal cells, with active cilia {Studnicka, '00 ; Brookover, ’10, and others; Kramer, ’ll). In older individuals these cilia become less evident and often disappear {Ariens Kappers, ’20). The ventricular part of the cell often is hyalin in character, and the secretory function of the cell is evidenced by the appearance of striations in the cytoplasm and by the frequent appearance of a layer of precipitated albumin on its surface. The protoplasm of the cell shows fine granulations, which first appear around the nucleus. According to Galeolti (’97), these arise from the nucleus but pass secondarily into the cytoplasm, where they are surrounded by a membrane. These basophilic granulations have been termed globuloblasts by Schnopfhagen (’82), and are said to have a delicate lipoid membrane. Besides these granules, acidophile granules {Goldmann, ’13 and ’13a) are present, containing oxydases which are very abundant in choroid cells {Pighini, ’12).

Fig. 31. A highly vascularized region of the ependyme in the diencephalon of a reptile; probably secretory.

48 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The capillary plexus overlying the choroid epithelium is very rich ; the capillary vessels are rather large, and their endothelium, particularly at the top of the villi, often lies close against the choroid cells (see fig. 32 of Acanthias and Aeipenser). This latter condition is illustrated when the vessels retract through shrinkage, leaving only an empty space between the endothelial lining and the choroid epithelium. In some cases a small amount of mesenchyrnatous tissue from the

Acanthus

Fig. 32. Choroidal epithelium from Acanthias, from the roof of the fourth ventricle of Acipenser, and the roof of the third ventricle of Hexanchus. Note the intimate relation of the capillary walls to the choroidal .epithelium in Acanthias and Acipenser and the lightly stained upper portion of the choroidal cells of Acipenser.

overlying membrane may accompany the capillaries (see Hexanchus). In the meningeal tissue overlying the choroid villi, or even on the ventricular side, are large mononuclear cells (Kobner, ’21). According to Sundwall (’17) these constitute a type of “mast” cell. Their positive staining reaction to pyrrol blue led Ooldmann (’13a ; see also ’13) to term them pyrrol cells. They show a similar positive reaction toward Weigert stains (Ariens Kappers, ’20). They are very clear in lower animals (Ceratodus). Asa Chandler (’ll) found many of these cells, with picric stainable granules, in Lepidosteus. In Amia, Acipenser, Polyodon, and ganoids in general, large masses of these granular cells, together with

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 49

certain pigment ceils, form a gland-like mass in the region of the calamus scriptorius and behind it {van der Horst, ’25, and Tilney, ’27). This glandula myelencephalica {van der Horst, '25) is well provided with blood vessels and is probably concerned with some sort of interchange between the ventricles and the blood.

The cuboidal cells of the choroid epithelium (separated by cement substance) appear to be secretory in character, and many observers regard their function as that of secretion of the cerebrospinal fluid. Moreover, the choroid epithelium may serve as a semipermeable membrane. It is probable that the so-called foramen of Magendie {Magendie, 1825 and 1842), which was once supposed to permit the flow of the cerebrospinal fluid into the subarachnoid space, is merely an artifact,*'* although Rogers and IFesZ (’31) described it in man as a complete defect of the lower part of the ventricular roof. It is lacking in lower vertebrates and recent evidence seems fairly conclusive that normally it is not present in higher mammals and probably not in man {Bland-Sutton, ’23 ; Huber and Huber, unpublished ; and Ariens Kappers, ’29). Whether or not the foramina of Luschka, situated one on each side in the region of the lateral recess of the fourth ventricle, are present is still questioned by some observers. The work of Strong, Greene, and Oliveira (’26) and Rasmussen (’27) appears to favor their presence in mammals. Thus it is probable that the choroid plexus permits the cerebrospinal fluid, and possibly other substances, to pass through it into the subarachnoid spaces and so ultimately to reach the vascular system. It is thus a part of the mechanism for regulating intraventricular and intracranial pressure.

The choroid plexus plays an important part as a protective mechanism, not permitting the passage into the ventricles of certain substances which may be present in the blood, such as some antitoxins (for tetanus and diphtheria), gall pigments, and certain medicinal substances {Meyer and Ranson, ’13). In this respect choroid tissue bears some resemblance to the placenta and has been termed “placenta cerebralis. ’’ Loeper (’04) and Goldmann (’13) have shown that the choroid epithelium, particularly in the fetus, is well supplied with glycogen, which is known as a reserve food.

The greater extent of the choroid plexuses in fishes and in the fetus probably is due to the fact that the ventricular fluid develops earlier in phylogenetic and ontogenetic history than do the subarachnoid spaces. Consequently the secretion of the fluid into the ventricles precedes the transudation of this fluid through the choroid plexuses into the subarachnoid spaces. The glandular character of the choroid is proved by the action of such drugs as pilocarpine upon its secretion. This action is indirect, being brought about by the effect of the drug upon the sympathetic innervation. Where the choroid plexuses have been removed experimentally, as in the frog, the animal soon dies under spastic conditions {Pellizzi, ’ll). Thus it would seem that the choroid plexus is a selective and secretory

“ The choroid ple.xus in the region wherein the foramen of Magendie is thought to occur, situated in front of the calamus scriptorius, is covered by the cerebellum which is attached to the choroid plexus by the pia and arachnoid. On lifting up the cerebellum so as to see the foramen of Magendie, the choroid plexus at this place is easily torn and a foramen is made. In three human brains, carefully examined in this way, the torn-out part of the choroid plexus could be seen attached to the meninges (Ariens Kappers).

50 NERVOUS SYSTEMS OF VERTEBRATES AND OP MAN

membrane, interpolated between the liquor cerebrospinalis internus on the one hand and the blood capillaries of the pia and the extracerebral fluid (liquor cerebrospinalis externus) on the other hand.

Mestrezat (Tl- 12) produced a fluid with a high chloride and a small protein content, similar, at least, to the cerebrospinal fluid. He showed that this fluid placed in a collodion sac showed no change when the sac was immersed in plasma. Weed and McKihhen (T9) found that an increase in the pressure of the cerebrospinal fluid followed intravenous injection of hypotonic solutions while a decrease of pressure followed the use of hypertonic solutions. Fremont-Smith (’27), commenting on the experiments of Mestrezat, emphasized the fact that if the choroid plexus had properties similar to the collodion walls of the sac, it too might serve as a semipermeable membrane, and showed further that variations in the chloride concentration of the cerebrospinal fluid is correlated with variations in the plasma proteins. Moreover, he pointed out that “only methods which affect the capillary pressure in the plexus or the osmotic pressure in the plasma affect the amount produced or the pressure of the cerebrospinal fluid.” Wislocki (’28, p. 1076) also reached the conclusion that the cells of the choroid plexus formed a semipermeable membrane “which serves as a mechanism interposed between blood stream and ventricles.” Very recently Clark (’28), in reviewing the question in connection with his work on the innervation of the choroid plexus, reached the conclusion that “from an unbiased view of the literature, one can safely say that it has not been proven that the epithelial cells of the plexus actively secrete the cerebrospinal fluid. On the contrary, the evidence thus far agrees as well, if not better, with the hypothesis that the choroid plexus is only an elaborately formed semipermeable membrane” (p. 13). Foley (’23) and Forbes, Fremont-Smith, and Woljf (’28) were able to reverse the direction of flow of the fluid through the plexus, although the attempts of Wislocki and Putnam (’21) and of others to do this had been unsuccessful.

In the cat, Clark (’28; compare Slohr, ’28 and earlier) described nerve fibers with sensory terminations to the choroid plexus, arising directly from the dorsolateral part of the medulla. These are the fibers described earlier by Benediki (’73) and are sometimes known as Benedikt’s thirteenth nerve. They supply the lateral tufts of the plexus, according to Clark, while the medial tufts receive fibers from the medulla by way of the taenia. These latter fibers, while sensory to the choroidal epithelium, differ in type of termination from the sensory fibers supplying the lateral tufts, and there is marked difference in the terminations of the adult as compared with those found in new-born animals. All the terminations demonstrated by Clark in the plexus, with the exception of sympathetic endings on the smooth muscle of blood vessels, were sensory in type.

Spongioblasts, as has been stated, form not only the aniagen of ependymal cells but also of another of the central nervous system supporting tissues, the neuroglia cells (fig. 33), under which term are included both astrocytes and oligodendroglia. Those cells which are to become astrocytes lose their cilia. Their cell bodies shift to a greater distance from the ventricular surface, keeping in contact with it only by means of a centrally directed process. A pial ex

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 51

pansiou is present, and consequently bipolar elements are formed which, according to de Castro (’20), may produce other bipolar elements by mitotic division. The centrosome and the Golgi net, originally located toward the central canal, lie opposite the central process {de Castro, '20). When the connection with the central canal is lost, the cell is termed a glioblast or an astroblast {von Lenhossek, ’95a) ; this is the displaced epithelial cell of Ramdn y Cajal (’09). A loss of its connection with the pial surface converts the astroblast into an astrocyte. Penfield (’2S) called attention to the fact that all the various developmental stages of the spongioblasts into adult neuroglia and ependyma are to be found in the adult human nervous system. Thus the primitive ependymal cells or spongioblasts are found in the region of the ventral sulcus of the cord and the median raphd of the medulla oblongata, while spongioblasts with pial attachments occur in the cerebellar astrocytes (the process being termed the fiber of Bergman) and in the subpial neuroglia {Penfield, ’28).

Germinal cells (or medulloblasts), which have shifted away from the ventricular surface during development, may form not only neuroblasts but spongioblasts as T^ell. They may remain for a long time as indifferent cells and during later development become either of these cellular elements. According to Bailey and Cushing (’26), they may retain their indifferent form in the adult. Thus astrocytes may arise from more deeply situated medulloblastic elements. These astrocytes are usually unipolar in type. During development the cells undergo further transformation. The process (sometimes more than one) grows out in the direction of the capillaries and forms trumpet-like endings or sucker-like feet around the vessels. The developing neuroglial cells now acquire smaller processes which pass out in all directions from the cell bodies. As they develop into adult cells some of the astrocytes retain their protoplasmic processes, others develop neuroglial fibrils and still others have both protoplasmic processes and fibrils. The first are called a protoplasmic type and the second a fibrillar type {von Kdlliker, ’89-’02, and Andriezen, ’93). The third type is said to be mixed, since fibrils are found in the more superficial expansions, the deeper expansions remaining typically protoplasmic. The neuroglial fibrils are fine, straight, and unbranched, and are always somewhere in relation to a cell {Huber, ’03, and others) and probably covered with cytoplasm under normal conditions. Retjnolds and Slater (’28) have pointed out that this covering is exceedingly thin “little more than a prolongation of the cell membrane.’’ Such fibers do not pass from the cytoplasm of one cell to that of another. According to da Fano (’06) and Penfield (’28), under exceptional conditions the fibrils may be free from the cell. The fibrillar type is found chiefly in the white matter of the brain and cord ; the protoplasmic type in the gray. There are exceptions to this, however, Spielmeyer (’22, quoted from Penfield, ’28) having demonstrated fibrous astrocytes in the thalamus. . Certain of this type which are partly wrapped around neurons belong to the satellite cells. Large cells with oval nuclei showing little chromatin occur among the protoplasmic type and are called giant cells. Perivascular astrocytes and marginal glia cells represent specialized forms of the fibrous type, while the astroblast, as was mentioned before, finds representation

52 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

in the subpial astrocyte of the adult cerebellum. Both protoplasmic and fibrillar types are said to multiply by amitotic division.

The adult astrocyte, or so-called common neuroglia cell, is an irregularly shaped cell with many branching processes (hence the term spider cell). Its cytoplasm is reticulated, the neuroglial fibrils, according to Del Rio Hortega (T6, T6a), being formed from the reticulum. Granules or gliosomes (see AMcarro, T3, and Del Rio Hortega, T6b) are present in the cell cytoplasm. These respond to certain special stains for mitachondria and so are believed by Nageotle (’20) to be of that character. Eisath (’06) regarded them as responsible for the formation of the fibrils, an opinion with which Fieandt (’10) was in agreement. Pigment granules, due to cell degeneration {Penfield, ’28), are found in astrocytes as in nerve cells. Ramon y Cajal (’13) described the Golgi apparatus as present in young mammals and suggested its probable persistence in a perinuclear position in older individuals, where by present methods it is unstainable. Practically invariably a centrosome can be demonstrated in astrocytes. The nucleus, which may be vesicular {Huher, ’03), has scattered chromatin granules and no nucleoli {Penfield, ’28).

Before closing this very brief summary of the structural relations of the astrocyte, the perivascular and subpial types must be considered somewhat more fully. The terminal enlargements of the astrocytes where they lie up against the pia were beheved by Held (’04) to form a “membrana limitans gliosa superficialis,” and he thought that the sucker-like feet of the neuroglial cells (astrocytes) formed a similar “membrana limitans gliosa vascularis” around the vessels (figs. 33A and B ; see also AriSns Kappers, ’29). These are regarded as shutting off the ectodermal nervous system from the mesodermal supporting and vascular tissue, and the layers between the brain parenchjona and the perivascular spaces form the pia-glial membrane of Schaltenbrand and Bailey (’27-’28). Between this latter membrane and the vessels, a narrow, perivascular lymph space remains — the space of Virchow-Robin. This is the only perivascular space in the central nervous system, except where larger vessels carrying a considerable amount of adventitia have, in addition, periadventitial lymph spaces associated with the vessels. Particularly at the line between the gray and the white matter, the spaces of Virchow-Robin show some enlargement ; these are known as the accessory pial spaces of Held (’04).

Held based his discussion of these membranes on the supposition that the neuroglial tissue forms a syncytium {Held, ’04 and ’09). A somewhat different interpretation of the relations is to be found in the work of Alzheimer (’10) and Ramon y Cajal (’13), where the evidence indicates an independence of the astrocytes. Such cells, according to the latter observer, merely show terminal expansions along the capillary walls and remain independent.

The astrocytes, both protoplasmic and fibrous (for cytogenesis, see Jones, ’32)

Near the membrana limitans gliosa, the neuroglia tissue often has the appearance of containing spaces, and since the nervous tissue is more scarce here, an impression is often given of the presence of fluid spaces. These are probably due to shrinkage but have been termed lacunae marginales and lacunae perivasculares.

’c.'MaUo“of processes of neuroglia cells to blood vessels. Bou,mn.

54 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

are of ectodermal origin and represent the chief supporting tissue within the central nervous system. Since astrocytes do lie in such intimate relation to the vascular supply, the suggestion of Held (’04) that they may be of importance in furthering the nutritive exchange of the nerve cell is pertinent. Reynolds and Slater (’28) have suggested that the protoplasmic type may be more efficient in facilitating the nourishment of the nerve cells and the fibrillar type afford better supporting tissue. Astrocytes have been regarded as serving as insulators. Mawas (’10), Achucarro (’ll), NageoUe (’20), and others have regarded neuroglia cells of this type as providing internally secreting tissue. There is a slowly accumulating body of evidence (see Del Rio Hortega, ’17 ; Penfield, ’28, for example) which would apparently favor such conclusions, but the evidence is far from complete as yet. It has been suggested that'such a secretion may exercise an influence over the general nervous morale.

A special type of interstitial cell is the oligodendroglia, which formed a part of the third element of the nervous system as the term was applied by Ramdn y Cajal (’09). Del Rio Hortega (’21) separated this third element of Ramdn y Cajal into two types, the oligodendroglia and the microglia. Frequently now {Penfield, ’28) the oligodendroglia are placed as a specialized type of neuroglia, and the term “third element” is limited to the microglia. In development, oligodendroglia develop from migratory spongioblasts which do not take on recognizable form until fairly late in embryonic life, although, according to Del Rio Hortega (’21), they are present before microglia appear and are found in both gray and white matter at birth. (For cytogenesis, see also Jones, ’32.)

Oligodendroglia are termed interfascicular where they lie between bundles of fibers over which their processes form a plexiform arrangement, while those that lie close to the cell bodies of neurons are perineuronal satellites, which belong to the general group of satellite cells. According to Penfield (’28), the position of the interfascicular oligodendroglia is in rows between the nerve fibers ; these may be regarded as homologous with the neurolemmal sheaths of peripheral nerve fibers and as concerned in the production of myelin within the central nervous system {Del Rio Hortega, ’22). Oligodendroglial cells, while forming the majority of interstitial cells of the nervous system, according to Reynolds and Slater (’28) are particularly numerous and most granular during the time when myelinization of the nerve tracts is at its maximum. Penta (’31) noted the close relation of oligodendroglia and myelin sheaths.

Oligodendroglial cells are not so large as astrocytes and they have but few processes (none of which terminate in vascular feet, Penfield, ’32) and no fibers. The chromatin within the small nuclei is densely packed. For further details, the papers of Del Rio Hortega (’21, ’22), Penfield (’24, ’25a, ’28, ’32), and Reynolds and Slater (’28) should be consulted.

Oligodendroglia have been regarded as concerned with the formation of the myelin sheaths within the central nervous system and their preservation. Reference has been made in this review to Del Rio Hortega’ s suggestion that the oligodendroglia of the central nervous system are homologous to the neurolemma of the peripheral nerves.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 55

Other specialized cells belonging to the general group of interstitial cells of the nervous system are the microglia or mesoglia cells of Del Rio Hortega, often termed “Hortega cells” since they were first stained completely by his method (Del Rio Hortega, '20). They are small cells, soraefmes oval but sometimes elongated or irregular in outline, the form depending upon the position of the cell. They have small, irregular, darkly staining nuclei, a granular cytoplasm and thin processes which run in all directions and which usually have perpendicular spiny e.xcrescences. The processes do not end in terminal e.xpansions. The cells have no centrosomes and no Golgi apparatus. Although they may contain granules of various sorts, these do not seem to be gliosomes and the cells contain no fibrils. Microglia appear to be migratory elements which occur as satellite cells around capillaries or neurons. They have been referred to as “tercer” elements by Del Rio Hortega, and are supposed to be derived, not from spongioblasts, but from fibroblasts which have migrated into the central nervous system from the pia (Del Rio Hortega, '21 ; Marinescu and Tupa, '25 ; Del Rio Hortega and Penfield, '26 ; Penfield, '28 ; Reynolds and Slater, '28 ; see also Zand, '30 ; Ishikawa, '32), and thus to be mesodermal in origin. Metz and Spatz ('24), however, held them to be ectodermal. In embryonic development the microglia cells do not occur first near the central canal or the ventricles as they might be e.xpected to do if they were of spongioblastic origin. Instead of this, they make their first appearance at the periphery of the central nervous system, particularly in those regions where masses of white substance lie in relation to the pia mater — in the brain, in the regions of the tela choroidea of the third and of the fourth ventricle and the pes pedunculi, and in the cord, beneath the pia in the dorsal and ventral sulci (Del Rio Hortega, '21 ; see also '21a). The microglia enter the white substance first, and first reach the gray substance a few days after birth ; twenty days after birth they are already more abundant in the gray matter than in the white matter (Strong, '25, in Bailey, Strong and Elwyn). Such evidence is regarded as indicating a pial origin for the microglia. Bergman (’27), who has made an e.xtensive study of the microglia, did not express himself with certainty as regards their origin. Metz and Spatz (’24), working with pathologic material, believed them to be ectodermal in origin.

Microglia are phagocytic in function, at least in the amoeboid form (Penfield, '25 and '28 ; Reynolds and Slater, '28 ; Zand, ’30). “Whether or not the ramified form of mesoglia in the normal brain serves any function is purely a matter of conjecture” (Peiifield, ’28).

Amphioxus, which possesses no blood vessels within its central nervous system and has no myelinated fibers, has only ependymal cells and fibers within its cord. In Petromyzon, which likewise lacks intraspinal vessels and myelin sheaths, similar conditions exist, for although forerunners of glia cells without connection with the central canal are present, the peripheral processes extend to the surface to form the limiting membrane of the cord. Real autonomic neuroglia cells do not make their appearance until the plagiostomes, where the cord is richly provided with vessels and with myelin sheaths. Above the plagiostome level, neuroglia elements become progressively better differentiated.

56 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN The Mbsodebmal Supporting Tissue of the Central Nervous System

THE MENINGES

The meninges in lower vertebrates are very different from those in mammals, including man. Although formerly certain observers, misled by superficial resemblances, supposed a dura mater, arachnoid, and pia mater to exist in cyclostomes and plagiostomes, later work has shown this to be incorrect. In 1884 Sagemehl pointed out that a real arachnoid does not occur in fishes, and Sterzi (’00-01) showed clearly that in cyclostomes and plagiostomes only one undifferentiated meninx is to be found, which he termed meninx primitiva. He considered this to furnish the source of the pia, arachnoid, and dura of higher animals. It appears more probable, however, that the real dural membrane of higher animals develops from the mesenchymatous blastema, immediately adjacent to the meninx primitiva. A brief summary of the meninges in the various classes of vertebrates follows.

Among cyclostomes, the relations in Petromyzon fluviatilis are essentially those described by Sterzi (’OO-’Ol). The cord is surrounded by tissue, which, by this observer, was called the meninx primitiva. This is continuous with the sheath over the nerve roots. From this membrane strong lateral ligaments extend laterally into the perimeningeal tissue. The membrane does not send septa into the spinal cord and consequently is easily detached. With this absence of intramedullary connective tissue septa is associated an absence of intramedullary blood vessels, and the food material for the cord must reach it through the superficial glial layer (the limitans gliosa superficialis). Between the endochondral layer of the vertebrae and the meninx primitiva is a broad layer of round or oval mucoid cells which form the perimeningeal tissue.

In the selachians, meningeal septa of connective tissue with accompanying blood vessels grow into the spinal cord from the surrounding tissue, but separated from the nervous elements by the development of a membrana limitans vascularis (membrana gliosa limitans). Thus a real penetration of the meningeal tissue into the nervous system does not occur although the nervous tissue and vascular system are brought into closer relation. The gray substance has a richer blood supply than does the white substance.

“ The external or periosteal dural membrane originates from the endosteal (or endochondral) connective tissue, which, in lower vertebrates, usually lies at a considerable distance from the meninx primitiva and separated from the origin of the real dural membrane by the peridural or perimeningeal tissue. In our opinion, it is better not to consider the so-called external or periosteal membrane of the spinal dura (which follows all the sinuosities of the bone) a part of the dura proper, since, although it fuses with it in the cranial cavity of the adult, in the embryonic condition the two are differentiable from each other (see Gegenbaur, '96 ; Poirier and Charpy, '01 ; Testut, ’ll , Sterzi, '00, and Rauber, ’03). The separation into periosteal and internal layers of the dura spinalis only leads to confusion. This confusion disappears if the so-called periosteal dura is left, where it belongs from nature and origin, with the connective tissue of the endochondrium or endosteum, while the term dura is restricted to the internal dural membrane of other authors.

As a matter of fact, spaces occupied by these septa are comparable to the fissures in the forebrain, with this difference — the former are much smaller and are filled up in greater part by the penetrating tis.sue, while the arachnoid cavities are larger and penetrate more deeply into the brain fissures. A further resemblance is to be found in the fact that in higher animals tlic dura remains outside of the brain fissures and the septa of the spinal cord.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 57

A single sheet of tissue — the meninx primitiva (Sterzi, ’OO-’Ol) — is present ; this is illustrated in Scy Ilium (fig. 34B), in which only the one meninx (d) was found, without differentiation into layers. The four spinal hgaments, already

a b

/

I

h

B

Fig. 34. A. Spinal cord of Petromyzon in situ; xx, space between meninx primitiva and cord, caused by shrinkage j a, perimeningeal tissue j 6, meninx primitiva ; c, lateral ligament.

B. Spinal cord of Scyllium cam’cula, tn siCu; a, lateral ligament; b, endochondrium; c, perimeningeal tissue; d, menin.x primitiva; c, perimeningeal vein; /, radix posterior; g, vertebra; A, arteria ventralis anterior.

described by Sterzi, are very evident in the shark. These consist of two strongly developed lateral ligaments (fig. 34B, a) and a very poorly developed dorsal ligament, as well as a somewhat better developed ventral one. The last two ligaments are merely thickenings of the meninx primitiva ; the lateral ligaments extend out for some distance. Outside of this meninx, between it and the

58 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

endoi’hachis or perichondrium, is a large amount of perimeningeal tissue (c) which, on the whole, has rather looser meshes than in cyclostomes, although it becomes somewhat more dense toward the endorhachis. It is mucoid in character.

The relations in ganoids (Acipenser and Polyodon) are similar to those just described for selachians. They will not receive further consideration here.

Such accounts of the meninges in bony fishes as are to be found in the literature are not in accord. In part, this may be due to actual differences within the various highly specialized groups of fishes. That different descriptions are not due entirely to differences in animals is illustrated by the varying accounts of Sterzi and Sagemehl, which are based in part on the same material, Bachus. According to Sterzi, the membranes in teleosts are similar in character to those in other fishes. There is a single meninx primitiva (which includes the leptomeninx and the dura in an undifferentiated state). This single meninx is separated from the endorhachis by a loose perimeningeal tissue of fatty character.'® Often this primitive meninx can be divided further into two leaves, an outer more or less pigmented portion formed of flattened cells and an inner part, A thickened bundle, the ligamentum ventralis, connects the meninx with the endorhachis at the level of the fifth to the eighth intervertebral disks, while the ligamentum laterale, lying between the anterior and posterior roots, passes from the meninx to the endorhachis through the perimeningeal space {Sterzi). The meninx primitiva has numerous capillaries which run into the cord, especially into the gray substance, and which are accompanied in course by fibrous connective tissue strands.

Sagemehl, unlike Sterzi, distinguished two membranes around the cord — (1) a dural membrane and (2) an underlying tissue — separated by a fissure, which he regarded as the anlage of pia and arachnoid. This latter is a meninx secundaria (Gefasshaut of Sagemehl). It is differentiated into inner and outer portions only in certain places and this differentiation is not comparable to that of the arachnoid and pial membranes of higher forms.

Ariens Kappers (’25 and ’26) called attention to the fact that the relations in teleosts vary with the animals studied. He compared the relations in the small teleost, Girardinius, with those in Lophius piscatorius and found them to be very different in the two forms. In Girardinius the meningeal tissue around the cord and in the medulla oblongata region and the cranium shows no differentiation into two layers and consequently a meninx primitiva, comparable to that of Sterzi’s account, was recognized here. In the lateral regions of the cord he found this membrane continuous with the periosteum with scarcely any intervening space. Dorsally a wider space occurs, filled in by an exceedingly thin and exceedingly wide-meshed perimeningeal tissue carrying large veins particularly on its dorsal side. The relations of the meningeal membrane are essentially the

Sterzi (and also Sagetnehl) regarded the perimeningeal tissue in elasmobranchs and ganoids as mucous in character but as adipose in teleosts. This is not always correct. An Acipenser sturio in the Amsterdam collection has, for instance, a large quantity of perimeningeal fat but mucous tissue has also been seen by us in several teleosts. Both of these tissues are fitted to serve as a buffer substance in a movable inclosure and they may replace each other as such (Ariens Kappers).

c

a 5 c

Fig. 35. A. The cervical cord of Lophius surrounded by meninges, a, dural layer; b, perimeningeal tissue; c, supramedullary spinal ganglion cells; d, internal leptomeningeal layer of the meninx; e, external leptomeningeal layer; /, leptomeninx (meninx secunda).

B. Enlarged photograph of the meninges in Lophius. a, spinal cord; b, internal layer of leptomeninx; c, external layer of leptomeninx; d, dural layer; e, fissure.

59

60 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

same in the medulla oblongata region and in the cranium of Girardinius, except that the relatively large cranial cavity permits the differentiation of a much larger amount of perimeningeal tissue.

Other relations exist in a large specimen of Lophius piscatorius (figs. 35A and B), which was available for study. Here there are large quantities of widemeshed perimeningeal tissue, but the structure of the meningeal tissue lying under it is very different from that in Girardinius. The tissue immediately under this perimeningeal layer (6) forms a fibrous layer (a) which is better developed in some places than in others, but which is always recognizable as a distinct layer. If this layer were separated from the underlying meningeal tissue by a continuous cavity, one might speak correctly of a well-differentiated dura mater. However, such a continuous separation as that described by Sagemehl as a "pericerebral lymphraum” and considered analogous to the subdural cavity of mammals is not demonstrable in Lophius. The relations in this animal are similar to those described by van Gelderen (’24) for early human embryos. This latter author found that the inner layer or the ectomeninx consists, in human embryos of 19.6 mm. length, of a tissue which, because of its greater compactness, contrasts distinctly with the overlying leptomeningeal tissue, although not separated from it as yet by a distinct space. Such a separation has not occurred in 25 to 30 mm. embryos. It is making its appearance as localized separations in embryos 35 to 40 mm. in length. Conditions similar to this latter stage are found in Lophius. In Lophius, the wide-meshed leptomeningeal tissue, although surrounding the whole spinal cord on all sides, is especially evident at the lateral and ventral sides of the medulla oblongata (a fact in agreement with the results of Weed (’16a), who found that the meningeal differentiation occurs first in the basal regions in ontogenetic development). In this animal, the tissue is not a receptacle for the liquor cerebrospinalis extemus as it is in higher animals, but probably is only supporting in character.

The presence of this wide-meshed leptomeningeal differentiation in Lophius and not in Girardinius may be associated, in part at least, with the occurrence of a much larger vertebral canal in the latter animal. In larger fishes there is a greater development of the skull and of the vertebral canal than of the nervous system itself, and a consequent increase in the tissues intervening between the brain and the bony wall. This increase, which is more evident in the perimeningeal than in the leptomeningeal tissue of Lophius — although present in both — indicates differentiation to a stage of development just preceding the appearance of an arachnoidal layer. A large quantity of perimeningeal mucoid or adipose tissue is present, this serving as a buffer tissue, thus permitting great flexibility of movement on the part of these animals. The thin perimeningeal adipose tissue which is present in man, in the space between the actual dural membrane and the endosteum of the vertebral column, may be a phylogenetic remnant of this buffer tissue. Such fatty tissue is lacking in the cranial cavity, which changes relatively less in form relations {Poirier and Charpy, ’01).

The fibrous membrane lying on the leptomeninx is regarded as dural tissue, since the fibrous condensation proves that there is development in the direction

EVOLUTION AND MOKPHOLOGY OF NERVOUS ELEMENTS 61

of a strongly fibrous dura mater. Van Gelderen (’24a) found that this dural lamella develops in Lophius in connection with (and so probably from) the meningeal tissue itself. The open-meshed character of the leptomeningeal part of the meninx is very striking (figs. 35A and B). In many places it is possible to distinguish an external layer (e) from an internal layer (a) (figs. 35A and B). In the former, the cells are almost perpendicular — in palisade arrangement — to the external layer of flat mesenchymal epithelial cells with which it is covered (figs. 35A and B). The meshes of the inner part (a) are less regular than those of the outer (e). The inner layer extends into the septa. It has more blood

Fio. 36. The fourth ventricle with its high choroidal roof in Petromyzon

fluviatilis.

vessels than does the outer part, particularly where it lies directly upon the cord. Although the tissue is fairly wide meshed, particularly in the region showing palisade arrangement, nevertheless it does not appear to be comparable to the trabecular tissue of the arachnoid. True arachnoid trabeculae, that is, narrow, connective tissue bands covered with mesenchymal epithelial cells, are not present. The pseudotrabeculae are branchings of single cells and consequently might well be designated as monocellular trabeculae. They are somewhat similar to the reticular connective tissue of lymph glands. In typical arachnoid tissue, the meshes are wider and the trabeculae less numerous.

There is no doubt that actual arachnoidal cavities are lacking in the lowest vertebrates such as cyclostomes, plagiostomes, ganoids, and teleosts. Consequently no liquor cerebrospinalis externus, such as fills the mammalian subarachnoidal cavities and which in man considerably surpasses in volume that of the liquor cerebrospinalis internus, is present ; but, although this former fluid is absent, it is a noteworthy fact that the liquor cerebrospinalis internus — that is, the ventricular fluid — is very frequently of considerable volume in lower fishes. The amount of this latter fluid is evidenced in plagiostomes (especially sharks)

62 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and cyclostomes not only by the wide ventricles but also by the great expansion of the choroid plexuses. This choroid membrane may bulge out very considerably over the fourth ventricle in the roof of the midbrain and rhombencephalon, as observed in Petromyzon (fig. 36). Similarly protruding choroid membranes are found in Ceratodus {Bing and Burckhardl, ’04 and ’05 ; N. Holmgren and van der Horst, ’25), in Lepidosteus and in Amia (see Ariens Kappers, ’26) ; the choroid roof of the third ventricle (the so-called parencephalon) evaginates to such a degree that the choroidal sacs, filled with a liquor cerebrospinalis internus,

extend far frontalward as well as caudalward along the outer side of the brain wall (fig. 37). In higher an’mals, especially in mammals in which the arachnoidal cavities and the liquor cerebrospinalis externus are markedly developed, the choroidal membranes no longer have the form of outwardly protuding sacs but, with few exceptions, are folded into the ventricle. Thus a large volume of liquor cerebrospinalis internus may be present in lower animals where the arachnoidal cavities and the external fluid are lacking. This relationship in lower forms is not accidental. The liquor cerebrospinalis externus does not originate in the arachnoidal spaces but from the ventricular fluid, which diffuses secondarily through the choroid membranes into the arachnoidal spaces. Consequently the appearance of a liquor cerebrospinalis externus and of arachnoidal and subarachnoidal cavities is due to the greater development of the functioning of the choroid plexus as a semipermeable membrane in these forms, while its relatively less permeability in lower forms explains the strong protrusion of the choroidal sacs in lower fishes, as well as the lack of proper arachnoidal cavities in these animalsThis phylogenetic process is repeated in the embryological development as has been shown by Weed (’16), who proved that the liquor cerebrospinalis externus is found in the arachnoidal cavities of the pig for the first time in an embryo of 14 mm., although there is fluid within the ventricle in the first stages of ventricular development. Moreover, Loeper (’04) called attention to the fact that the choroidal sacs are larger in the fetus than in the adult.

The meninges are more highly developed in amphibians than in fishes, at least two membranes being demonstrable. The inner of these membranes is

Fig. 37. Cross section through the frontal portion of the thalamus, showing dorsal, lateral, and ventral recesses of the third ventricle in Lepidosteus osseus.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 63

pigmented. Slerzi called them the dura mater and the meninx secundaria. Such a division is only indicated in the tailed Amphibia, but the two are well differentiated in the frog. The dura mater contains lacunae, the further development of which appears to be dependent upon an increased metabolism. The membranes are better developed in the cephalic regions ; caudally in urodeles, only one membrane is found, and this is formed by the union of the two more cephalic layers, being held closely to the endorhachis (periosteum of the vertebrae) by means of connective tissue trabeculae. A large perimeningeal space is found outside of the meninges. This space is not filled with mucus nor with fat, as in fishes, but consists of a series of peculiar little tubes filled with a white substance. They are supposed to be continuations of the saccus endolymphaticus (see Chapter IV) which enters the cranial cavity and reaches the vertebral canal through the foramen magnum, extending as far as the eleventh spinal nerve.

At the level of each spinal nerve, little sacs arise from the major sac and surround the spinal ganglia. The contents of the organ depend upon the condition of nutrition of the animal. The epithelial lining consists of cuboidal cells which may become greatly flattened when the organ is distended. An analogous structure is found in Dipnoi and in teleosts frontally near the fourth ventricle. The significance of the arrangement in any of the forms is not understood thoroughly

Fio. 38. The well-developed dura mater and subdural spaces m Athena (the owl). A, subdural space; B, leptomeninx : C, dura mater.

as yet. , , , . ,

In reptiles the dura is fairly well separated from the underlying leptomemnx or meninx secundaria of Sierzi. This latter does not show, as yet, differentiation into pia and arachnoid and consequently no arachnoidal cavities containing liquor cerebrospinalis externus are present. A wide perimemngeal or peridural space occurs in reptiles. This space contains a large amount of tissue in which are large epimeningeal veins with endothelial walls. The denticulate ligaments are well developed in snakes (Slerzi). Smaller dorsal and ventral ligaments are

^'^ThLura in the avian brain is more differentiated (fig. 38) and the subdural space is more evident than in lower forms. A dura and a secondary meninx are present, and in the latter there is an indication of the beginning of arachnmdal spaces ;hich, although still small, allow injected flmd to spread over a considerable area of the cervical and into the thoracic region, as has been shown by Hansen Pruss (’23). Slerzi regarded the secondary meninx as consisting of an outer endothelial (or mesenchymal epithelial) covering, a middle v^cular, and an

inner portion lying close to » LvVedtVsVnd a few trabecuke"

64 NERVOUS SYSTEMS OF VERTEBRATES AND OF ^N

and a ventral ligament are present. In the lumbar regions there is a fibrous thickening between the two denticulate ligaments and the ventral ligament which is called the ponticuli interligamentarii and which lies in the fissures between the eminentiae ventrales of the anterior horns.

The meninges of the cord show a distinct advance in differentiation in mammals as compared with birds, since the meninx secunda first differentiates definitely into arachnoid and pia in the mammalian forms (marsupials and placentals). The pia lies directly upon the spinal cord and consists of fibrous connective tissue containing blood vessels but not a capillary net {Weed, T4). Pigment cells occur in certain regions of this layer in manamals, including man. Frequently in older individuals, concretions of lime are found. The inner layer of the pia, the membrana intima piae of Held (’09), consists of large endothelial (or mesenchymal epithelial) cells. It lies immediately upon the membrana gliosa perivascularis.

There is no general agreement as to the exact place where the line between pia and arachnoid should be drawn. According to certain observers, the arachnoid may be compared to a spongy sac, the walls of which are interconnected by trabeculae of fibrous connective tissue. Walls and trabeculae alike are covered by a thin layer of simple pavement epithelium, which is variously termed either endothelium, mesenchymal epithelium, or mesothelium {Weed, ’16a). The inner wall lies against the pia, the outer wall against the subdural space, and the spaces between the trabeculae are termed the intraarachnoid spaces. Like the pia, the innermost layer of the arachnoid and some of the larger trabeculae carry vessels which run to the brain and cord.

Other observers {Weed, ’23, for example) regard the arachnoid as consisting of merely the outer layer of the above account and the arachnoid trabeculae, while the inner layer is counted with the pia. In such a case, the openings between the trabeculae and above the inner layer are usually termed the subarachnoid spaces and the arachnoid is considered a non-vascular membrane.

The intraarachnoid or subarachnoid chambers are continuous with the perivascular spaces of Virchow-Robin in the cord, so that there is communication between the perivascular lymph spaces and the cerebrospinal fluid. Consequently, substances injected into the subarachnoid spaces may pass into the perivascular spaces and reach the spinal cord from there {Goldmann, ’13), if they are able to penetrate the membrana intima piae and membrana limitans gliosa. Weed (’23) found it possible to inject the entire perivascular system with hypertonic solutions tlirough the subarachnoid space. This space in the cord is continuous frontally with that at the base of the brain, which is also connected with similar spaces above the cerebellum and the cerebrum, although this latter connection is less easy to demonstrate {Goldmann, ’13). Key and Retzius (’75 “ .V ca-^.c came under obi-crvation where the pia was entirely covered with sucli concretions, yet tlie man, who died of pneumonia, had never complained of headaclic or of pain from the cord region and liad Ki'’en no intimation that he was aware of any .stimulation of the meninges which might have ari'-en from ^uch particle.s (Ariens Kappers). According to mo.st ob-'crver-s, thc.'-e concretioms ari.'c through tlie calcification of cartilaginous plates (.see Obcrslcvier’n Arliciten dcr Neurologic, ’03).

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 65

’76) were the first to demonstrate definitely this continuity of these spaces of the cord and brain.

A considerable number of observers agree that the subdural and subarachnoid spaces do not interconnect under normal conditions. Substances introduced into the latter space do not appear in the subdural spaces [Key and Retzius, ’75-76; TFeed, ’14; Cushing, ^ »

’14). However, Quirwke (’72)

the subdural space entered

Quincke ^ were correct, it would

tion it may be stated that

Leary and Edwards (’33) re garded the subdural space as formed by the apposition of two structures which were developmentally unlike, the lining being formed by ectodermal arachnoidal and mesodermalfibroelastic dural tissue.

In so far as it has been possible to judge, the consensus of opinion appears to be that the subdural space, in reality a

potential space, is anatomi cally closed, and that it is not essential to the outflow of the cerebrospinal fluid.

The dura mater is formed B

of fibrous connective tissue Fia. 39. A. The spinal cord of a newborn cat in siCu.

•ii. XT. nu A .4 dura mater: B, perimeningeal adipose tissue; C, ligamen Wlth both white fibrous and fj^denticulatum; P, arachnoid region,

yellow elastic fibrils. Usually B. The spinal cord of a newborn cat in situ. A, epimenin it is regarded as consisting of 5 ^™”;*' g'rSSUSra 5 .“ VS.?”'"" two layers, but of these only ... , . • i t.

the inner is true dura mater. The outer, which lies in close relation with the

Fig. 39. A. The spinal cord of a newborn cat in situ. A, dura mater; B, pcrimeningeal adipose tissue; C, ligamentu’m denticulatum; D, arachnoid region.

B. The spinal cord of a newborn cat in situ. A, epimeningeal veins; B, perimeningeal adipose tissue; C, ligamentum denticulatum; O, subdural space ; F, spinal veins.

the inner is true dura mater. The outer, which lies in close relation with the skull, is to be regarded as periosteum. In the vertebral canal the so-called intradural space occurs between the two layers. This space, also lined with mesenchymal epithelium, is really a remnant of the perimeningeal tissue and contains many fat cells and the dural sinuses. Arteries and veins are found in the dura, in addition to the great sinuses of the so-called intradural space. The dura

66 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

passes out on the roots, becoming continuous with the perineurium of the nerve trunks.

The fluid of the subarachnoid or intraarachnoid spaces is called the liquor cerebrospinalis externus ; that inside the ventricles of the brain and cord, the liquor cerebrospinalis intemus. The chemical difference between the two is slight (perhaps a difference in the percentage of dextrose). Both fluids, according to Halliburton (’16) and others, contain but little protein, small quantities of salt, and traces of dextrose. They are clear, have a low specific gravity, and contain few cells. They differ from the l 3 nnph of the body in having a smaller percentage of lipoids and fewer cells.

The source of the cerebrospinal fluid has been a matter of much dispute. Early workers {Magendie, 1825, 1842) believed that it was formed by the leptomeninges. Faivre (’53) suggested first that the cells of the choroid plexuses of the ventricles were concerned with its formation. Since then, many observers have brought fonvard evidence favoring these plexuses as a source of the cerebrospinal fluid. Among such workers are Cappelletli (’00), Pettit and Gerard (’02), Meek (’07), Mott (’10), Weed (’14b, ’17a, and elsewhere), and others. The work of Weed and of others suggests that perivascular tissue within the nervous system may also contribute to the formation of the fluid. There is evidence that ventricular ependymal cells in regions other than those of the choroid plexuses may secrete cerebrospinal fluid. This question of the probable secretory function of the choroidal epithelial cells has been discussed earlier (see p. 49) and does not require further discussion here (see pp. 47 to 50). The cerebrospinal fluid which enters the subarachnoid or intraarachnoid spaces collects particularly in the cisterna cerebellomedullaris. The flow is relatively rapid in the region of the cord and to the ventral side of the brain, but slower and less efficient to the upper portions of the cerebral hemispheres (Weed, ’23) as seen in injection experiments.

There has not been general agreement as to the course of this fluid from the intraarachnoid or subarachnoid spaces into the general circulation. In an effort to solve this problem. Key and Retzius (’75-’76) injected a gelatin mass, colored by Berlin blue, under low continuous pressure into the subarachnoid or intraarachnoid spaces of a cadaver. Their results led them to believe that the fluid reached the cerebral sinuses by way of the Pacchionian bodies (arachnoid villi), although a limited amount escaped into the lymphatics associated with the nerve roots and the intervertebral ganglia. Difficulties in accepting this theory arose from the fact that in lower animals and in infants the large Pacchionian granulations are lacking. However, the results of Key and Retzius have been corroborated by Weed, who found that the major drainage occurs along the Pacchionian bodies (arachnoid villi), while a slower drainage into the lymphatic system takes place along certain emergent nerves. In the brain, where there are large venous sinuses and Pacchionian bodies or arachnoid villi, according to the experiments of Quincke (’72), Reiner and Schnitzler (’94), Hill (’90), Spina (’00, ’00a), Lewandowsky (’00), and Weed(’\i, ’14a, ’17a, ’32), the major escape of the cerebrospinal fluid is directly into the venous sinuses of the dura. Small quantities undoubtedly pass along the cranial nerves, particularly the olfactory

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 67

and optic (TFeed, T4a, T7, etc.). In relation to the olfactory the fluid collects first in a perineural cul-de-sac on the lamina cribrosa and then finds an outlet into the lymph vessels of the nose and those associated with other regions of nerve distribution. In the spinal cord region, the cerebrospinal fluid escapes chiefly along the nerve roots into the lymphatic system (Weed, ’14a).

Weed (’17, ’32) and van Gelderen (’24, ’24a) have studied the development of the meninges carefully and reached the conclusion that they are mesodermal derivatives. Harvey and Burr (’26)-® considered that they are partly of ectodermal origin. Leary and Edwards (’33) considered that the subdural space is bounded by ectodermal arachnoid and mesodermal fibroblastic tissue. On the whole the evidence for a mesodermal origin for these membranes seems more conclusive at present, particularly in view of the more recent work of Flexner (’29), in the Contributions to Embryology (110, Publ. 394) of the Carnegie Institute. The first meninx to develop is the so-called entomeninx or leptomeninx which corresponds to meninx secundaria (pia and arachnoid) of Sterzi (’00). Later the ectomeninx differentiates from the surrounding connective tissue, and eventually becomes separated from the endomeninx by the appearance of a subdural space. Fatty tissue and dural sinuses then separate the dura from the periosteum. Finally the entomeninx differentiates into pia and arachnoid. For details of this development the papers of Weed (’17, ’32) and van Gelderen (’24, ’24a) should be consulted. The matter is reviewed in the 1932 account of Weed to be found in “Cytology and cellular pathology of the nervous system” edited by Penfield (Hoeber, New York).

Supporting Elements of the Peripheral Nervous System

THE ectodermal SUPPORTING TISSUE OF THE PERIPHERAL NERVES

The supporting elements of the peripheral nervous system fall into two main groups — one of ectodermal, the other of mesodermal origin. The former group will be considered first. It includes the neurolemma sheath cells or lemnocytes and the capsules of the spinal and sympathetic ganglion cells. To these must be added the myelin sheaths of peripheral medullated fibers, particularly if they are to be regarded either as derivatives of the neurolemma sheath cells or as parts of these cells, as certain observers suppose to be the case. The structure of the myelin and the problems concerned with its development and function have been discussed earlier, and reference is made here to the previous account (pp. 20-23).

The neurolemma sheath (nucleated sheath of Schwann) is usually described as a thin and apparently homogeneous membrane surrounding the myelin sheath of the peripheral medullated fiber and lying next to the axolemma of the unmedullated fiber. Huber (’16, ’17) suggested a delicate fibrillar structure for the neurolemma ; Bruno (’31) described it as singly refractive, isotropic, and having

Harvey and Burr (’26) believed that in amphibians the entomeninx (or leptomeninx) arises from the neural crest cells. Flexner (’29), studying amphibian material, could not confirm their results but derived this meninx from mesoderm.

68 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

some elasticity. One (and in some lower vertebrates more than one) oval or somewhat flattened nucleus may be found in each internodal segment. This nucleus is surrounded by a small amount of granular cytoplasm which shows a Golgi net. As was stated previously, certain observers regard the neurokeratin net, and even the axolemma, as a part of the sheath cell, the latter forming the inner boundary and the neurolemma sheath the outer boundary of the cell. To such workers, the node represents the place of fusion of adjacent neurolemma cells. (A more extended account of this is given on page 20, footnote.)

There has been considerable difference of opinion as to the embryologic origin of the neurolemma sheath cells. Carpenter and Main (’07) and Held (’09a) believed that usually they arise from the neural tube, but certain observers are of the opinion that, in certain cases at least, they may arise from peripheral epithelium such as that of the mucous membrane surrounding the fila olfactoria {Disse, ’97, and Held), or the supporting cells of sensory corpuscles (as, for example, Elmer’s organ, Boeke and de Groot, ’08). Indeed, according to Heringa (’17), they may sometimes be of mesenchymal origin. However, the mesenchyme thus giving origin to sheath cells is probably developed from ectomesoderm, derived from certain regions of the neural crest, and thus in reality of ectodermal origin. Bailey and Cushing (’26) regarded them as derived, at least in part, from the neural crest; Kuniz (’22, ’29) thought them to be partly of neural crest derivation. Harrison (’06) found that when the neural crest is ablated at an early stage in the developing tadpole, development proceeds and the ventral horn neurons send out their neuraxes into the near-lying somites. Such neuraxes are naked fibers, lacking the neurolemma sheaths present in the control. Therefore, Harrison concluded that here, at least, the neurolemma sheaths are from the neural crest.

In normal development the sheath cells grow out with the developing nerve fibers. At first they are not arranged in a single layer along the axis cylinders, but entwine among the bundles of fibers (fig. 40A) in a more or less irregular way. Then they penetrate the bundles (fig. 40B) and finally become arranged around the single nerve fibers (fig. 40C). Only after such an arrangement of the neurolemma cells has occurred does the finer connective tissue, the endoneurium, extend into the funiculus among the nerve fibers. Ml peripheral nerve fibers including those of the sympathetic system — are surrounded by neuroleimna sheaths until such fibers approach their terminations. In the region of spinal and sympathetic ganglia the neurolemma sheaths become continuous with the nucleated capsules over these cells. Such capsules arc composed of a single layer of flattened cells, often termed amphicytes or trophocytes (fig. 22). These are ectodermal in origin. iVs the ncurolemma sheaths approach the central nervous .system, they terminate in such a way that their central ends form a curved Hue along each root, the so-called arc of Frommann (see fig. 21) — not to be confused with Frominaim's lines or crosses, obtained by silver nitrate treatment of peripheral nerves. In the lumbar region, this arc lies at some distance from the cord, while in the cervical region it lies more or less inside of the cord {Rcdlich, ’97 ; i.evi, ’06). 'I'he region of the roots between the termination of the sheaths and their entrance to the cord is relatively less well protected. This fact has been

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 69

used by Redlich as an explanation of the greater predisposition of the lumbosacral cord for developing tabes. According to Michailow (’09a), there is considerable individual variation in the position of the arc of Frommann in man, a fact which

may have clinical significance.

Of late years there has been special interest in the problems associated with the

degeneration and regeneration of peripheral nerves. In connection with this, there has been much interest as to the function of the sheath cells or lemnocytes. Practically all observers would grant that they were a supporting tissue, but there has been, and still is, much difference of opinion as to whether or not they have any further function.

The monophyletic theory of nerve fiber formation regards the nerve process (neuraxis or dendrite, as the case may be) as the outgrowth of a single cell ; the polyphyletic theory considers that such a fiber is produced by the end-to-end arrangement and fusion of a number of cells. In its extreme form, it regards them as autonomous links of a chain, but in its more modified form it regards them, during development, as under the influence of

Fig. 40. Series of drawings illustrating the development of the neurolemma sheath.

A. Intercostal nerve from a 30 mm. sheep embryo, showing perifunicular sheath cells.

B. Sciatic nerve of a 70 mm. sheep embryo, showing wandering of sheath cells into the nerve funiculi.

C. Sciatic nerve of a 24 cm. sheep embryo with intrafunicular sheath cells. To the left, sheath cells amongst the primitive nerve fibers devoid of myelin sheaths; to the right, nerve fibers with myelin sheaths.

those more centrally placed

neurons of which the nerve fibers are to constitute the dendrites or neuraxes in

the fully developed peripheral nerve. Certain observers, who favor the polyphyletic theory, regard the neurofibrils as derived from the protoplasm of the sheath cells or the “Bandfasern” and speak of peripheral autoregeneration {von Biingner, ’91 ; Galeotti and Levi, '95 ; Kennedy, ’97 ; Bethe, ’03, and others). In its extreme form, however, the polyphyletic, or chain theory, is regarded by most observers as disproved, but there are still those who regard the influence

70 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

from the more centrally placed neurons as purely functional in character. Thus it has been thought that under the stimulative influence of central cells (which may send processes into them for a short distance), the sheath cells or lemnocytes may form neurofibrils and thus cause growth of a nerve fiber from other sources than the tip of the neuroblast. Held ('09a ; see also ’29) believed that he could demonstrate that neurofibrils arise in the central neuroblasts and that primarily all neurofibrils arise from these cells and become embedded secondarily in the syncytial cell mass of neurolemma cells. However, he believed it possible that fibrils of this origin, once embedded in the protoplasm of the sheath cells, might show independent regeneration for a short time and for a short distance, even if severed from the central neuron, and that by such local degeneration a nerve might heal per primam, if the ends are inamediately sutured. Held emphasized the importance of the neuro blasts, and unlike Beihe (’03) and Apathy (’08), did not regard the function of the sheath cells as that of neuroformation.

Heringa (’17) believed that the neurofibrils occur in the protoplasm itself of the neurolemma cells, and that the axoplasm or neuroplasm around the neurofibrils is continuous with the protoplasm of these cells. In cross section, this neuroplasm has a reticular aspect, which to Heringa’ s mind indicated an alveolar or foam-like structure, with the more fluid protoplasm filling the minute alveoli. Whether such a structure actually exists in life or whether it is the result of fixation is still a disputed question. Many observers are inclined to regard the neuroplasm as a colloidal mass. In this are more or less dense, gel-like portions which form an irregular anastomosing reticulum, similar to that demonstrated for colloids, rather than a strictly alveolar or foam-like arrangement such as Butschli predicated as the structure of protoplasm. Heringa accepted the idea that the axoplasm is in the form of a syncytium, in the sense of Sedgwick (’95) and Rohde (’16), with fibrils derived from different cells running through this syncytium. The neurofibrils occur in the denser part of the neuroplasm. Traced peripheralward, this latter decreases in amount. After the loss of the myelin sheath, some observers regard the neuroplasm as directly continuous with the protoplasm of the sheath cells, and this is used as evidence that the neurofibrils are formed in and from these cells. The results of Spielmeyer (’17) on regeneration of nerves appear to indicate the local formation of neurofibrils in the protoplasm of the neurolemma sheath cells. Likewise the researches of various workers on nerve terminations (cells of Grandry, Meissner’s tactile corpuscles, sole-plate endings, etc.) suggest that neurofibrils may penetrate other cells and that the neuroplasm of the nerve process may become continuous with the protoplasm of these cells.

In contrast to the views of the polyphyleticists are those of the observers who believe the nerve fiber to be an outgrowth of the neuron only, and the neurolemma to be incapable of forming neurofibrils. Many prominent scientists have favored this view since the time of His (’90). Among the later contributions in support of this view must be mentioned particularly the experimental work of Harrison (’06, ’07, ’07a, ’08, ’10, ’10a, ’24, etc.) and his school and some of the work on nerve regeneration and nerve transplantation carried on during the World War.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 71

Harrison’s work has already been mentioned in connection with the origin of neurolemma. His results also indicated that, in the absence of neurolemma, the nerve process grows out to its termination in the somite, and that it is the direct outgrowth from a single nerve cell. The work on nerve degeneration and regeneration is so important in confirming the monophyletic theory of neuron development that a brief review is included.

The very brief review given here of the process of nerve regeneration is based chiefly on the extended account of Huber (’27). The tips of the central neuraxes show degenerative changes within 24 hours after the injury, as indicated by silver staining {H uber, ’27) . Such changes resemble those found in the degenerating central stmnp. Within this time also, as indicated by the work of Perroncito (’07), Ranson (’12), and others, evidences of regeneration appear in the form of deUcate branches on the sides of both medullated and unmedullated neuraxes. For the most part, these branches disappear secondarily and are to be regarded chiefly as abortive attempts at regeneration. The effective regeneration of myelinated fibers has not gone far before the end of the first week following the injury, at which time many new branches and subdivisions of the neuraxes — as high as fifty per medullated fiber (Ranson) — have made their appearance. The unmyelinated fibers (Ranson, Huber) begin regeneration within about two weeks. 'These new neuraxes terminate in end disks or incremental cones, the size of which varies, perhaps with the amount of resistance encountered by the tip during its growth (Htiber). Such neuraxes may follow a relatively straight course through the old neurolemma sheaths or may form elaborate spirals, the spirals of Perroncito (’07). Through the wound region the fibers are not parallel ; they have an irregular zigzag course, but beyond the wound, if they reach the distal degenerating stump of the nerve, they may assume again a regular parallel course and, if the degeneration of the severed portion is not complete, the downgrowing neuraxes may penetrate the old neurolemma sheaths, several of such fibers sometimes lying within a single sheath (Huber, ’27). Others of these downgrowing fibers may be found in the endoneurial, perineurial, and even in the epineurial sheaths of the nerve.

While these processes have been going on in that portion of the nerve fiber retaining its attachment to the cell body, active degenerative processes are occurring in the severed stump. Several days (2 to 3 in human) after section of a peripheral nerve, structural changes appear in the severed portion, first in the form of granulations or swellings of the neurofibrils (Monckeberg axid Bethe, ’99) or in silver impregnation material, as staining irregularities (Ranson). Then changes occur in the myelin sheath leading to fragmentation of it and then of the neuraxis. The sheath cells increase in size and multiplication of their nuclei (by mitosis) occurs (von Biingner, ’91 ; Huber, ’92). Gradually the sheath cells acquire a phagocytic action toward the broken-down myelin. These sheath cells form the syncytial protoplasmic bands termed “Bandfasern by the German writers (von Biingner). Although they have been variously regarded as of mesodermal, of ectodermal, or of mixed mesodermal and ectodermal origin, the results of modem experimental and embryological investigations appear to

72 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

indicate their ectodermal derivation. Huber stated that their protoplasm is undifferentiated and possibly may bfe embryonic in character. There has been much controversy with regard to the part which these “Bandfasern" play in the regeneration of fibers and the question is not settled as yet to the satisfaction of all. For the details of this dispute and for a discussion of the various theories, reference is made to the works on nerve degeneration and regeneration listed in the bibliography at the end of this chapter. Only a few of the more recent views can be mentioned here. Boeke (’16, ’17) and Ranson (’12) regarded the neurofibrillar strands as always intraprotoplasraic, lying within the “Bandfasem,” and in fact the tendency at present is to regard the growing neuraxes within neurolemma sheaths as having such an intraprotoplasmic position. By certain observers {Kirk and Lewis, ’17), they were regarded as serving as conduits' for the growing nerve fibers, although their presence is not regarded as necessary for such growth. In his monograph on nerve regeneration, Huber said of these fibers; “Faint, longitudinal striation is now and again seen in the ‘Bandfasem’ in silver preparations, but transition stages between the syncytial protoplasmic bands and developed neuraxes have not been demonstrated.’’ Preparations obtained from experiments on nerve transplantation, in which the sheath cells of the transplanted fibers do not proliferate, and where regenerating nerve fibers may grow for very considerable distances through regions from which sheath cells are absent, has led Huber to confirm Harrison’s results — that the presence of sheath cells is not necessary to the growth of the nerve fiber. The weight of evidence at present seems to us to be in favor of regarding the nerve fiber as the outgrowth always of a neuron and the sheath cell as concerned in protective and supporting functions, rather than in the production of conductive elements. Although not directly concerned with the problems here under discussion, the topic of nerve degeneration and regeneration cannot be dropped without some tribute to the great significance of the work carried on in this field both in America and abroad in the period preceding and during the World War. As products of the experimentation and observations of that time have come improved methods of preventing amputation neuromas and valuable contributions to the surgical treatment of nerves severed with loss of substance. The contributors have been many, of whom may be mentioned : Perrondto (’07), Poscharissky (’07), Ramon y Cajal (’08, ’08a, ’08b), Ranson (’12), Huber f’OOb, ’16-’17, ’19, ’20, ’27, ’27a), Boeke (’16, ’17), Dustin (’18), Ingehrigsten (’16, ’16a, ’18), Kirk and Lewis (’17), Elsberg (’19), Huber and Lewis (’20), and others. For a later presentation of his views on nerve degeneration and regeneration, the reader should consult the 1925 contribution of Ram6n y Cajal.

THE MESODERMAL SUPPORTING TISSUE OF THE PERIPHERAL NERVES

The peripheral nerves, like the central nervous system, are surrounded by mesodermal tissue. The bundles of nerves, or funiculi as they are called technically, are surrounded by a dense fibrous connective tissue sheath, continuous centrally with the dura and with the capsules of the ganglia. This perineurium

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 73

forms the most definite sheath and it is through this that the stitches are taken in suturing severed nerves.

The perineurium is continued into the funiculus and between the various nerve fibers as a thin connective tissue lamella, spoken of as endoneurium. The endoneurium also enters the spinal ganglia, forming a binding and supporting tissue for the cells and fibers. Its lymph spaces are said to be continuous with subarachnoid spaces, and substances in the lymph stream may filter into subarachnoid spaces which extend out for some distance with the roots. Various toxic agents may thus enter subarachnoid spaces. In some regions these endoneurial lymph channels, the spaces inside the “guaina sussidiaria” oi Ruffini (’05) or the endoneurial lymph sheath of Retzius (’98), appear to correspond with relatively definite regions of subarachnoid spaces without being in open communication. Recently Kazue Yuien (’28) showed that the lymph current in these endoneural lymph spaces is in a direction opposite to that of the nerve current, running centrally with motor roots and peripherally with sensory roots.

Between and over the funiculi, holding them together and binding them into a nerve trunk, is a looser areolar connective tissue, carrying smaller blood vessels and lymph vessels and containing scattered fat cells — the so-called epineurium. This looser connective tissue relates the nerve trunk to the surrounding structures.

The number of fibers in the funiculi decreases peripheralward until single fibers can be traced. These are surrounded by the neurolemma and a single thin connective tissue sheath with flat cells, forming the sheath of Henle. In terminations, such as the cells of Grandry, where the sensory fiber enters a cell, the sheath of Henle terminates shortly before the point of entrance. Where the sensory fiber enters the core of a sensory corpuscle, this sheath frequently becomes continuous with the capsule of the corpuscle. The Henle sheath of certain sympathetic neurons is believed to become continuous at times with the basement membrane of glandular cells.

Structural Laws of the Nervous System

THE PRINCIPLES OP NEUROBIOTAXIS

Before dealing with the comparative anatomy of the nervous system in vertebrates, we wish to discuss briefly certain principles which appear to deterniine the arrangement of the nervous elements. For many years the factors which have determined the structures of the nervous system — the form of its neurons and their positions and connections — have puzzled the minds of neurologists. Embryologists first gave the matter consideration, but were interested chiefly in the connections between the central nervous system and the periphery, that is, in the factors determining the course of nerve roots. Thus Hensen (’03) assumed that “all nerves originated by an insufficient separation in the primary connections between ganglion cells and their peripheral organs during evolution.” This assumption was the forerunner of the conception that a real separation of the cellular constituents of the body never occurs but that

74 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

its elements remain connected with each other. A similar conception of syndesmism was defended in England by Sedgwick (’95), who stated that “nerves are developments of the reticulum, elongated strands of the pale substance composing this reticulum with some of their nuclei.” At a later time reference will be made to this conception which may contain some truth, .although it certainly does not explain the peculiar selectivity of neuronal connections.

In contrast with the views just quoted are those of Balfour (’76) and His (’89), who showed that the connections of nerves with their end-organs are secondarily acquired. His (’89 ; see also ’87), who in general regarded mechanical factors as the most important ones in embryonic development, considered that the underlying principles determining growth and arrangement of the nervous elements presented a purely mechanical problem. He attempted to solve this problem by assuming that the direction of growth of the nervous processes is determined by paths of least resistance. Dustin (’10) called this the hodogenetic principle. He pointed out that the presence of such preformed and thus guiding paths explained why peripheral nerves, after severance, grow out along the degenerated nerve sheaths of the peripheral stump, while fibers of the central nervous system, where distinct neurolemma sheaths are lacking, practically never reach their functional end station and consequently do not show functional regeneration. However, the presence later of well-established pathways does not necessarily offer an explanation of their formation.

If it be assumed that inherently the developing nerve cell is capable of sending out processes but that the direction of these processes is dependent upon previously existing paths in the surrounding tissue, then the question remains as to what establishes the typical arrangement in this tissue so that functionally correct nervous impulses — and no others — result. Furthermore, the precise character of the functional connections within the central nervous system, where neurolemma sheaths do not occur, remains entirely unexplained by this theory. Another embryologist. Held (’09a; see also ’06 and ’07), proceeding on the assumption already formulated by Hensen and Sedgwick that the nerve cells are ab origine connected by intracellular protoplasmic bridges, believed that he was: able to demonstrate that the fibrils grow into such bridges. As Held himself: recognized, even such a condition would not explain why nerve fibrils grow into, certain of such protoplasmic bridges and not into others. He thought that the] position of the cells (1. c., p. 47) and the directions of their neuraxes (1. c., p. 68)i might be determining factors but not of sufficient importance to explain the^ typical and precise selectivity evinced in the formation of nerve tracts. HelCi\ stated (1. c., p. 270) that the determining principle had, as yet, escaped scientifi<5-] research. A similar remark was made by Harrison (’10), to whom we owe th(;j first researches on the outgrowth of nerve fibers in tissue cultures. This observe;];, stated that “there is nothing in this work which throws any light upon process by which the final connection of a nerve fiber is established.” Harriso^

Though Penfield (’28) has shown that the oligodendroglia surrounds the nerve fibers witlii'ii; the central nervous system, these cells have by no means such a regular arrangement as the sheal'i^ cells of peripheral nerves.

EVOLUTION AND MORPHOLOGY OP NERVOUS ELEMENTS 75

thought that it must be a form of specific reaction between each kind of nerve fiber and its particular end-organ.

Ramdn y Cajal (’93) was the first to hypothecate that tropistic influences play a part in determining the direction of the outgrowth of fibers. He also believed that the nervous elements of the central nervous system e.xert reciprocal tropistic influences on each other. He thought that the connections of the nervous elements are determined by the secretion of attracting and repelling substances and by the sensitivity of the nerve cells to such substances. The m.aterial thus secreted appears in different parts of the nervous system at different periods of their embryonic development and can be formed by ependymal cells ” as well as by neurons. Ramon y Cajal stated that the stage of attraction “coincides with the evolution of the cell.” However, of necessity, such a statement leads to further difficulty, for Ramon y Cajal did not make clear

A

Fia. tl. The neuraxw of iieuroblasts adjacent to a fiber tract grow out in the direct ion of the por[>endicuIariy irradiatuig currents of tlie latter, as indicated by the arrows in the figure. The arrow drawn with the solid line indicates the direction of growth of the activating bundle. Boh.

what factors underlie the evolution of the cells and consequently the secretion of these specific substances. Thus his theory does not offer an explanation of the causes of the local and temporary selectivity of this chemical process nor the factors that produce the different characters of the axons and dendrites (the so-called dynamic polarization of the neuron). Ramon y Cajal himself remarked (in his work on the retina), “Cette th^orie presuppose des conditions pr6alables chimiques et morphologiques tout h fait inexplicables : on peut dire que cette theoric eioigne la difficulte sans cependant parvenir a la r&oudre.”

More important than Ramon y Cajal’ s chemotactic theory — of greater importance even than he himself realized — is a remark of this author concerning the shift of nerve cells during the embryonic development of the nervous system. This remark (unknown to Ariens Kappers when he began his researches) is in perfect harmony with the latter’s observations, and he is more than glad to be able to confirm Ram6n y Cajal’s statements on this point, since he arrived at a similar conclusion by an entirely different method, that is, not by embryonic but by comparative phylogenetic studies. The remark made by Ramon y Cajal runs as follows; “If, during embryonic development, new axons pass to some

Although we are not inclined to attribute any great part in this tropic phenomenon to the central spongioblasts, it is a fact that certain non-nervou.s elements (chief cells, muscular cells, epithelium, and even cicatrices of connective tissue) are able, at certain stages of their dewlnnment, to affect the direction of outgrowing nervous elements.

76 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

region of the central nervous system, ganglion cells may approach these axons in two different ways, either by sending forth long dendrites, or by a migration of the cell body itself” (translated from his textbook, p. 560). Ramon y Cajal mentioned, as an example of the shift of cells, the movement of the superficial layer of granular cells, which at an early period of development cover the surface of the cerebellum but which later on shift into its depths. He also called attention to the fact that the cell bodies composing the spinal ganglia, which originally lie very close to the neural tube, later on move a short distance peripheralward away from the spinal cord. Comparative researches on the medulla oblongata and the mesencephalon {Ariens Kappers, ’07, ’08a, ’20) indicate that changes in position of the cell bodies are determined by a process of taxis or tropism, due to the stimulation of such cells and the bioelectric consequences of such stimulation which determine the selectivity of the neuronal connections and the differences between the dendrites and neuraxes, the so-called dynamic polarization of the neuron (compare with p. 79). By what means these processes are reproduced engrammatically under embryologic conditions, it is not possible to state at present. A similar statement applies to the entire ontogenetic development. Thus the formation of the extremities for walking and grasping can be explained only by use of engrammatic factors, the specific characters of which are unknown as yet. That the electrical potentials arising during evolution {Child, ’21), the sequence of which may be determined by engrammatic factors, may play a part in this process is possible.

Topographic differences are observed in homologous cell groups in the central nervous system of vertebrates. Such differences are most evident in the nuclei of the motor roots of the oblongata {Black, ’17, ’22; van der Horst, ’18, and Addens, ’28 ; for these references see Chapter V), but may also be seen in cells of the spinal cord, the midbrain, and the forebrain {Herrick, ’10 ; Elliot Smith, ’10 ; Dart, ’20 ; for these references see Chapter IX). These differences appear to be caused by differences in the nervous stimulations which reach the nuclei, the topography of the cell groups depending upon the direction from which the greater number or the more dominant impulses reach the cells. Thus we have to do with the process of taxis or tropism, to which the term “neurobiotaxis” {Ariens Kappers, ’08) has been applied, since it occurs in the nervous system during life. A striking example of this process of neurobiotaxis is seen in fig. 42, where the dorsal position of the abducens nucleus associated with the large medial longitudinal fasciculus (vestibular and optic reflex tract in Acanthias) contrasts strongly with the ventral position of the same nucleus in the bony fish, where the medial longitudinal bundle is small but where the ventral optic reflex tracts that influence this cell group develop to a much greater size (the tractus tecto bulbaris : fig. 43). Cases of this type make it clear that an increase in the amount of stimulation of a certain tract does not lead to the approach of all nuclear groups of the medulla oblongata to this tract but only of such definite groups as have a functional relationship with it. Thus such nuclear groups as the cells of the facial nucleus that innervate the gills do not migrate in the direction of this strengthened reflex field since their function is

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 77

not related to that of these optic reflex tracts. There is, then, a selection, and obviously definite relationships are required to determine the location of the nuclear groups. Further, this relation is a functional one and consists in a simultaneous excitation of the enlarged stimulating tract and the motor cells. If the motor cells and the enlarged center or tract have a stimulative interrelation — that is, if the motor cells are in action at the same moment at which a nervous current passes through the enlarged tract (as is the case with the eye muscle nuclei and the optic reflex tracts) — then the cells are attracted by those

Fig. 42. The medulla oblongata of Acanthias vulgaris, van der Horst Note the dorsal position of the abducens nucleus m association with the marked development of the medial longitudmal fasciculus (f I V-) tracts and not otherwise. Anatomically this is expressed by the fact that the nucleus of the abducens nerve shifts from a position near one path for visual reflexes (the medial longitudinal bundle) to a position near another path for visual reflexes (the tractus tecto-bulbaris ventralis) with an increase in the size of the latter bundle. However, the increased number of taste fibers in the medulla oblongata of certain fishes does not have the slightest influence on the eye muscle nuclei, since a functional relation does not exist between the two, but this increase does cause migration of the nuclei of nerves supplying the jaws and gills which make effectual those impulses carried by the taste buds. Thus the positions and relations of the dendrites and cell bodies of neurons of the central nervous system are regulated in conformity with that law of psychology which has long been known as the law of association. According to this law.

78 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

simultaneousness of excitations or their successive occurrence is the leading factor. The early work {Ariens Kappers, ’07) on motor cells and their dendrites led to a more complete study of the course of the neuraxes, and it was supposed (’08, ’08a) that electric conditions (potential differences) existing between the regions where the fibers begin and where they terminate may explain this law,

Fig. 43. The medulla oblongata of a bony fish (Mugil chelo). van der Horst. Note the ventral position of the abducens nucleus in association with the marked development of the ventral optic reflex path (tractus tecto bulbaris,

Ir.t.b.).

namely, that the potential relationship between two regions always underlies the establishment of eonnections between such regions.

The above discussion shows that associated stimulation was found to be the determining factor in both dendritic and neuraxonic outgrowths. In the following chapters, many examples of this phenomenon will be mentioned.

It has been possible (Ariens Kappers) to formulate the following laws :

(1) If several centers of stimulation are present in the nervous system, the outgrowth of the chief dendrites and eventually the shifting of cells takes place in the direction whence the greatest number of stimulations reach the cell.

(2) This outgrowth or shifting, however, only takes place between stimulatively correlated centers.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 79

(3) Temporarily correlated excitation plays a part also in establishing the connections of the neuraxes.

Fig. 44 shows the outgrowth of the dendrite and the final shifting of the cell body in a stimulo-petal direction and the stimulo-concurrent course of the neurite.

The comparative anatomy of the fiber tracts of the central nervous system gives much evidence in support of these laws as leading principles in the formation of tracts. However, the question arises: If associated excitation causes the formation of dendrites and the shift of ganglion cells, as well as the formation of neuraxes, how, then, are to be explained the differences m character of the dendrites and neuraxes ; that is, the dynamic polarization of the neuron?

This is no easy problem. The migration of the dendrites and cell bodies of the neurons toward the center of excitation is a stimulo-petal tropism. The case is different and more difficult with the neuraxis, since this process apparently does not grow toward the exciting center but in the same direction as the nerve current that irradiates from that center and consequently away from the exciting center itself (fig. 46).

That the neuraxis does grow in the same direction as the exciting current and that this current plays an important part in its formation was proved by Bok (’15). Bok formd that when a bundle of unmedullated nerve fibers within the central nervous system grows out and passes neuroblasts in Fig. 44. Changes in the position of the facial

its course, these neuroblasts become nucleus with shortening of the dendrites and length . ^ , ,, „ , enmg of the axons. A, shark ; B, lizard; C, mouse,

activated by the fiber bundle. Such a

neuroblast sends out a neuraxis in a direction perpendicular to the bundle and this neuraxis apparently grows in the direction of the current that irradiates sidewise from the growing nerve fibers (see fig. 47).

Further proof of this is found in the fact that neuroblasts lying in the region of the growing bundle are only activated if the bundle has reached their level. Thus the outgrowth of the processes is like that of the dendrites, where the shift of the cells is determined by the passage of the nervous current. However, the neuraxis grows away from the region of stimulation, its direction of growth

80 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

being stimulo-fugal, or better, stimulo-concurrent (i.e. with the current), while the growth of the dendrite indicates a stimulo-petal tropism. Attention should

be called here to the fact that both the neuroblasts and the neurosensory cells which have no dendrites may shift with the nervous current away from the center of excitation. This is contrary to the shift which occurs later with the adult ganglion cells.

It is clear, however, that the functional connection of the outgrowing neuraxis cannot be determined by this process of irradiation alone. Boh also reaUzed this and came to the conclusion that the final connection or end point of the neuraxis is determined by the chief law of neurobiotaxis, namely, by a

Fig. 45. Cross section of the spinal cord of a larva of stimulative relationship between Acanthius of 3 cm. length. Reiaius. c, commissural ceU; the

m motor cell. •'

neuraxis receives its activation and the neuron with which the new neuraxis connects. The irradiation of side currents from naked neuraxes is unquestionable. An example of it is found in

Fig. 46. The growth of the dendrites (upper figure) and the shifting of the cell body (lower figure) in the direction of the impulses. The course of the neuraxis corresponds to the direction of the nerve current (centrifugal or stimulo-concurrent).

the relation e.xisting between the dendrites of the Purkinje cell and the parallel fibers of the cerebellum (see next page).

The question of selectivity on the part of the termination of the neura.xis will

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEIVIENTS 81

be considered later. When the stimulo-concurrent character of the neuraxis and the stunulo-petal character of the dendrite and adult ganglion cell were once established, the difficulty then arose of finding the physicochemical basis for these phenomena. It may be asked in the first place how it is possible for two opposed tropisms to occur in one and the same cell. Ariens Kappers is inclined to believe that bioelectric influences of the nature of galvano-tropisms have here the chief influence. That these bioelectric factors are carried by chemical substances in the form of ions is evident. Thus the process has an important chemical aspect, as is always the case where electrical phenomena are associated with organic substances. However, these bioelectric influences are not to be regarded as the equivalent of chemotropisms (in the sense in which that word is generally used), since the electrical property of the ions is the determining factor rather than the chemical properties and chemical conditions which accompany these special electrical conditions. The dissociated or ionized stages play the most important part in these growth phenomena just as they do in stimulation and conduction (see also Lillie). Pure chemical or hormonic conditions are not the determining factor because a formation of chemical substances generally leads to more permanent chemical conditions while physiological changes in stimulation and conduction are temporary and of short duration. Moreover, introspective observations of the activities of our own brains compel the conviction that they are primarily dependent upon processes of short duration and that more constant relations are made possible only by frequent repetition.

It is possible, of course, that during embryonic development the bioelectrical conditions are of longer duration than they are in later life. This seems possible (1) because of certain electrical phenomena well known in nerve physiology,

(2) because of the exactly opposite and antipolar character of the tropisms, and (3) because certain collaterals of neuraxes in the beginning very often run exactly perpendicularly to the motor neuraxis. An illustration of this last point is found in mitral cell neuraxes, and likewise in collaterals of peripheral nerves {Speidel, ’32) gro%ving out perpendicularly at the nodes of Ranvier.

The perpendicular position of the dendrites of Purkinje cells with respect to the parallel fibers within the cerebellum (fig. 48) is highly suggestive of bioelectrical fields in connection with stimulative activities, the more so since

Fig. 47. The activation of the neuroblasfs (at the right) through the irradiation of an unmedullated but growing nerve bundle (at the left) is indieated. (See Boh.) Note the horizontal course of the newly formed neuraxes, which grow in a direction corresponding to the irradiation of the impulse. The most proximal neuroblasts have grown the most in correspondence with the growth of the activating bundle. The vertical arrow indicates the direction of the irradiation.

82 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Addison (’ll) showed that the perpendicular orientation of the Purkinje dendrites, originally diversely spread out, is only completed 10 days after birth in the rat. Other examples may be found in the spinal cord of the lamprey, where the dendrites of motor cells are orientated at right angles to the unmedullated longitudinal tracts (Treljakoff, ’08). The fact that certain of these dendrites reach the periphery of the cord, where they end in thickenings, suggested to Treljakoff that their arrangement is due to metabolic influences, since the spinal

A B

c

Fia. IS. niiistration of the dendrites of the Purkinje cells. From the cerebclhar cortex of the cat. A, section parallel to the fissure; D, section perpendicular to the fissure; C, cerebellar cortex of young bird. Golgi. Note the peripheral thickening of the dendrites and their relation to the pia mater.

cord of the lamprey does not contain blood ve.ssel.s and food materials for the nerve substances mu.st come from the pcrimedullary vessels. Doubtles.s their clo.se relation to the surface of the cord may be explained in this way (see p. 152). It does not .‘■cem probable, however, that the perpendicuhir position with respect to the axis of the cord can be attributed to the cfTcct of the peripheral blood ve,-^‘ls alone, since contact with the surface might ejuite as well be obtained if the dendrites should extend out at different angles. In fact, these dendrites might most ea-ily reach tlu; dr)rsal and ventral surfaces if they were to spread out in a fan-like manner parallel to the longitudinal axis of the cord, .since the dor.s<>- ventral dimensioas of the cord are very .small in this animal. In thi.s

84 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

of this fact in the formation of the nervous system has been confirmed by Child (’21), who also is a believer in the theory of bioelectric potentials as leading factors in the formation of the neuron. Nerve cells situated in the vicinity of the electro-negative excitation center first produce (probably on account of change in surface tension as Marinesco, ’13, supposed) a positive process corresponding to the direction of irradiation of the nervous current from the excitation center. This first offshoot is the neuraxis. As Child has already emphasized, the region of origin of the neuraxis is electrically determined in each cell by the electrical currents of its environment. It seems probable to us that positive ions, such as potassium, play an iruportant part in the growth of the neuraxis, just as different concentrations of electrolytes are of importance in the passage of the nervous current itself, as has been demonstrated by Macdonald (’02), Sherrington (’06), Lillie (’ll), and others. The neuraxis, more than any other part of the neuron, contains at its periphery large quantities of potassium compounds, as Macdonald (’02), Macallum (’05), and Macallum and Menten (’05) — worldng independently of each other and in different ways — have shown. This high percentage of potassium must favor the stimuloconcurrent growth of the neuraxis, the more so on account of its relaxing or lowering influence on surface tension (Macallum) and its membrane-loosening influence (Hober, ’22 ; Kunio Sato, ’29). Just as Hdber (’22) is inclined to consider the action current of the nerve affected by cationic transport at the seat of action, so we consider the outgrowth of the neuraxes as being greatly influenced by such a cationic transport, the outgrowth of the neuraxis being apparently influenced by the action current. This may explain the favorable influence of slight stimulations on regeneration. This is not so strange after all, for it is generally recognized that the function of an organ and its growth are two different aspects of the same or allied processes. The electric galvano-tropic character of the outgrowth of the neuraxis has been confirmed experimentally in serum cultures of embryonic tissue in Harrison’s laboratory by Ingvar (’19, ’20), who found that such an outgrowth may be determined by a constant galvanic current of the strength of two to four billionths of an ampere, density approximately yoW to WotT) nonpolarized electrodes. In this connection mention should be made of the interesting experiments performed on neuroblasts in vitro by Peterfi _a.nd Kapel (’28). These observers noted that if very young neuroblasts were touched by a microsurgical needle an outgrowth of the neuroblast from the side opposite the injury resulted, thus making it probable that an injury current running away from the injury causes this outgrowth. Further, Peterfi (’32) showed that in somewhat older stages dendrites may grow out and the cell may shift its cathodic direction.

Turning now to a consideration of the dendrites, we find that they arise at a much later stage. Somewhat later still, the cell body itself begins to shift in a direction opposite to that of the growth of the neuraxis, that is, toward the center of excitation (stimulo-petal shifting). The stimulo-petal cathodic tropism of the dendrites and of the perinuclear cell protoplasm coincides in appearance with that of the tigroid substance and consequently only takes place when the

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 85

neuron has reached a much more advanced stage of development. This tropism of the dendrites and of the cell itself is in perfect harmony with the phenomena of excitation or contraction at the cathode as stated in Pfluger’s law and is found in any protoplasm. Thus, for the amoeba, Loeb and Maxwell (’96) demonstrated that the animal tends to move toward the field of excitation. In connection herewith, it is .an interesting fact that a muscle, subjected to a slight galvanic current, shows an excitation and contraction at the cathode and a slight rela.\ation at the anode, the latter reminding us of the lengthening of the neura,xis at the anode. That the galvano-tropism of the dendrite is actually opposite to that of the neuraxis- is evident, and the consideration of this outgrowth of the dendrites and shifting of the adult cell body as a tropism in the direction of the negative excitation field raises the question as to what conditions in cell or dendrites favor such a tropism. In seeking an e.xplanation, it becomes a matter of interest that the cathodic shift only occurs after the appearance of the Nissl substance and in those parts of the cell which contain these Nissl granules, that is, in the dendrites and cell body but not in the neuraxis. It may even be said that the shifting of the cells is most evident in those cells that contain the greatest amount of Nissl substance — the motor cells, in which this process was first realized and most studied. The Nissl or tigroid substance, then, is probably associated with the appearance of this stimulo-petal cathodic tropism. The young neurobUist, which contains no Nissl substance, not only fails to show a cathodic tropism, but may even exliibit, to some slight degree, a tropism of the opposite type as was experimentally confirmed by Peterfi (’32). What, then, is the influence of the chromidial substance on this process? It may be that kataphoretic phenomena {Hardy, ’99) and certain facts of colloidal chemistry, which according to Greeley (’04) and von Herwerden ('13a) may be applied to protophismic colloidal suspensions, furnish the solution of this problem. It is known that the Nissl substance during life {Cowdry, ’16, ’24), particularly during the development of the cells {van Bicrvliet, ’00), is in a more or less fluid condition and that it surrounds the fibrillar substance. This chromidial substance is a very complex substance, but it is certain that it contains acid derivatives (compounds of nucleic acids with iron ; see above). Furthermore, excitation involving oxidation increases the acid content. According to Hardy’s researches, acids favor the shifting toward the cathode of such colloidal matter as is suspended in them. The late appearance of the chromidial substance in the cell body, in which it appears only when the neuraxis has already attained considerable growth, would offer an explanation for the late formation of the dendrites. Thus it appears that the developmental character of the neuron may be explained by bioelectric forces and is in perfect harmony with the bioelectric phenomena known in nerve physiology. The growth of the neuraxis depends on the action current, whereas the formation and contraction of the dendrites is homologous to the cathodic outgrowth and contraction of pseudopodia in a galvanic current and is favored by metabolic cell processes.

The part played by the dendrites and the cytoplasm in the assimilation of oxygen, a part already assumed by Golgi, is confirmed by researches on the

86 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

localization of oxidizing enzymes. These enzymes only occur in the dendrites and the cytoplasm of the cell body — not in the neuraxis. Unna’s (’ 16) experiments “ with “rongalit white” likewise show the presence of a large amount of stored oxygen in the dendrites and cytoplasm and so confirm this view. Consequently it is not strange that the dendrites show a nutritive tropism and that this early nutritive tropism of the dendrites and their early tendency to pass toward nutritive surfaces {Tretjakoff, ’08) can later change into a stimulative tropism. This is easily explained by the fact that stimulation increases the assimilation of oxygen, as has been experimentally proved by L. Hill (’00) and as is in perfect harmony with Hering’s thesis that the autonomic (that is, the unstimulated nutritive) absorption of oxygen is enhanced by an “allonomic” (that is, stimulative) absorption of oxygen at the pole where the stimulus enters the neuron. Thus the nutritive and stimulative tropisms of the dendrites are not opposed to each other. On the contrary, the one t 3 q)e of tropism may be transformed into the other type because of their fundamental relation to each other.

MONOAXONISM AND POLYDENDRITISM

Certain of the characteristics of the nerve cells have yet to be considered. First among these is the fact that only one neuraxis (monoaxonism) leaves the cell body, whereas a large number of dendrites (polydendritism) may grow out from the cell body in any direction toward the several centers of stimulation. To explain monoaxonism, return must be made to the theory of the stimulo-concurrent direction of growth of the neuraxes of neuroblasts (fig. 40). Such neuraxes are perpendicular to the activating tract or center. Also, the collaterals of the axis cylinders attain a perpendicular position and both of these facts point to a condition of perfect polarity. If, now, two (or more) different excitation centers simultaneously activate one cell, it is to be expected that only one neuraxis will grow along the resultant line of the two forces, that is, of the two currents, since it is only in this line that an equal influence on both sides of the growing neuraxis is possible. But what will happen if two or more activating centers exert their influence, not simultaneously but successively, upon the same cell body? One of these activating centers will of necessity be first and will cause the formation of the axon hillock. When, however, such an axon hillock has arisen — that is, when a small efferent pole has been formed on the cell body — it is to be expected that the greater opportunity for conductivity offered by this zone will tend to favor the passage of any new bioelectric current set up in the cell body to this region of the cell rather than tend toward the formation of a new neuraxis at some other part of the neuron. Thus the nerve impulse will follow the path where the conductivity is best. In doing this, it will tend to pass in the direction already laid down by the original bioelectric currents. This is the more possible because the original irradiating character of the nervous currents favors their reaching all

“ Speaking of the formation of T-shaped ganglion cells from bipolar cells, Hanslrom’s (’29) conclusion should be mentioned. He concluded that this is caused by a trophic or nutritive migration of those ceils to the periphery of tracts.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 87

parts of the cell, and such currents necessarily reach the axon hillock.^^ The first zone of efferent fibrils formed at the anlage of the axon hillock thus becomes the point of origin of the neuraxis, which, by its unique conductive properties, is to carry the stimuli away from the cell body. So although stimulations coming from other directions might change the course of the axonic outgrowth, it does not affect its origin, if once established, nor create a new a.xon hillock.

The conditions underlying the formation of the dendrites are altogether different from those producing neuraxis formation, since the whole cytoplasm is sensitive to stimulation. Thus when a stimulus appears in the vicinity of a cell, the part of the protoplasm nearest to such a point of stimulation will receive the impulse and may shift in the direction of the stimulus. Since the protoplasm is much alike throughout the celt body and the Nissl substance is present in the whole cell (with the exception of the axon hillock) this process of shifting is, of course, not restricted to any one field, and is peculiarly favored in no place except the area nearest the stimulation. Centers of excitation situated elsewhere may, and even must, cause new protoplasmic outgrowths in their own directions. In connection with the supposed r61e of the Nissl substance in the stimulo-petal tropism of the cytoplasm and dendrites, it is interesting to note that at places where dendrites bifurcate, a considerably larger amount of Nissl substance is usually found (cone chromidial de bifurcation) than elsewhere in the dendrite.

Thus monoaxonism is found to be a result of the special character and the polar localization of the anodal tropic part of the cell, whereas polydendritism is based on the fact that the cell protoplasm contains no places peculiarly favored for receiving stimulations and everywhere is able to give a cathodic tropism.

SELECTIVITY IN TUB CONNECTIONS OF NEURONS

It has already been shown that the anatomical relations occurring within the nervous system and the factors underlying cell shifting point clearly to the fact that the relationships which determine connections are synchronic or immediately successive functional activities. This fundamental law of neurobiotaxis not only shows that the well-known law of association in psychology is a neurobiotactic law, but it also indicates how wonderfully polarized is the whole process of tract formation and how well this process fits into the class of bioelectric phenomena. In explaining selectivity, the following points must be emphasized. It may be assumed that a state of excitation once set up in a budding neuraxis proceeds rapidly, and that a strong current of internal positive (external negative) potential reaches the budding cone of this neuraxis. That this is more than mere supposition has been indicated by the researches of Scaffidi (’10),“ who proved that if a This place also appears to contain diplosoines. As it does not seem reasonable to suppose that the position of the diplosome (which position is extremely variable) can determine the “working" point of the current, one must conclude that the place of the centrosome is bioelectrically defined by the influence of the radiating current. Such a conclusion is of importance from an erabryological standpoint.

“L. 0 ., p. 345: “Die Schnittflachen der Nerven, welche nach der Durchtrennung m dlu bleiben, sowohl nach wcnigen Minuten als nach verschiedenen Monaten fast immer positivereagicren, u.s.w. Die positive Ladung der Schnittflachen verschwindet rascher nach der einfachen Durchtrennung, d.h. wann Versehmelzung folgt.”

88 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

nerve which has been sectioned starts to regenerate, its budding cone shows a positive potential current which may be called the regeneration current or growth current. That nerve growth and nerve function are closely interrelated has been pointed out by Coghill (’29, and elsewhere). If it may now be assumed that a number of nerve cells are situated in the neighborhood of the budding cone, one of which cells is already functioning while the others are not, it will be evident that the cell which is functioning is the only one that presents a selective point to the budding neuraxis. The potential which runs along the budding neuraxis can find its natural counterpart only in the already ionized cell and not in the cells that are not stimulated and consequently are indifferent objects, not presenting any predilection for stimulation. The raised surface potential, which appears near the cell shortly after a stimulation and which makes its appearance inunediately following the stimulation, probably plays the principal part in establishing the connection between the growing neuraxis and the cell or dendrite with which it is to come into relation (see fig. 46 left side of the cell body). In other words, the growth current of the budding neuraxis will find a favored direction for its growth to a neuron which has been set in action just previously. In the physiology of nervous currents there are several facts which favor the supposition that an immediately prior excitation exercises a direct influence upon the course of other excitations occurring in the nervous system. Thus Sherrington (’06) emphasized “that the threshold of a reflex is lowered by the excitation just preceding its own.” This fact also explains why a connection just established between two neurons (a synapse) “is an apparatus for coordination and may introduce a common path” — that is, of several excitations (1. c., pp. 184-351). This principle of selectivity, based on the influence of a just preceding functional state as a center of attraction for other nervous currents or budding neuraxes, is of great use in making clear the connections of the nervous system with the muscles. Bok (’17) emphasized that the connection of certain muscles with widely separated parts of the central nervous system has to be explained by the fact that the tonic change of the muscle (since the muscle formation precedes the formation of the nerve roots) acts tropistically upon the central nerve fibers. A contraction (or an analogous embryonic condition ; for example, the embryonic growth of a muscle “®) produced by some external non-nervous condition may attract action currents which chance to be present at that moment in the central nervous system. This attraction may lead to the outgrowth of processes toward that muscle. In accordance with this conception is the fact that the first motor roots, as they occur in invertebrates and as Coghill (see bibliography for Chap. II) found them in larvae of amphibians, originate as collaterals of longitudinal tracts within the spinal cord. Only later do the special neuroblasts which lie near such longitudinal tracts become activated and form neuraxes which replace these collaterals and constitute the real motor roots. It follows that originally the myotomes attract action currents, or analogous currents, present in the central nervous system.

As mentioned before, cations, such as potassium, probably play an important

That rapidly proliferating tissue attracts axons or collaterals was shown by Ramdn y Cajal (’00) in proliferating granulation connective tissue.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 89

part in the growth of the neuraxis, just as they do in the action current. In this connection must be emphasized the importance of the researches of Howell (’06), and particularly of Zwaardemaker (’19), regarding the r61e of potassium in the transmission of stimuli and in growth in general {Lewis and Lewis, ’12), It is probable (see also Detwiler, ’26, and Hoadley, ’25) that, in the embryo, proliferation of muscle tissue exerts an influence similar to the functioning of that tissue. Proliferating muscle tissue attracts the irradiations of the nervous currents and thus draws toward itself the outgrowths from the spinal cord. Proliferating connective tissue attracts collaterals of the roots and even of the lateral funiculi (experiments of Ramon y Cajal, ’00, ’ll). Child (’21) and Hoadley (’25) have called attention to the probability that, during evolution of tissue, electrical conditions occur which are very similar to those occurring on stimulation. However, the sequence in which this embryonic muscle proliferation takes place is a problem, the solution of which must be left to the myologists.

It is evident, of course, that if a nervous connection is once established, nervous stimulation must have a great influence over the further development of the limbs and other organs. That such is the case has been proved experimentally by Hamburger (’02, ’04), W eiss (’26), and others. This effect of the establishment of a nervous connection is not confined to motor neurons alone, but is equally true for the sensory cells, as has been illustrated in connection with taste buds and Meissner’s tactile corpuscles {Olmsted, ’20 ; May, ’25). Burr (’16, ’16a, and ’20), Detwiler (’23, ’23a, ’24, etc.), and Stone (’24) showed that nervous tissue, such as placodes, may be transplanted to various regions near the central nervous system of embryos and will establish new connections with centers with which they were formerly entirely unrelated. A transplanted part of the olfactory placode grew out into the dorsal thalamus in one of Burr's experiments. May and Detwiler (’25) have obtained similar results in transplantation of the eye. Thus, in one of these experiments, an eye was transplanted near the medulla oblongata and became connected with the medulla oblongata center. These experiments are of great importance, although they are not to be regarded as leading to the conclusion that under normal circumstances selectivity depends only on proximity. Such experiments prove chiefly that any state of embryonic growth or regeneration may activate and direct the growth of nervous tissue. This power of activation and growth is to be regarded as a general tendency of growing tissues. According to Burr the influence exerted is analogous to that of mitogenic rays. The special problems of brain anatomy, the special relations underlying the distinct selectivity that give it its functional value, are approached more closely in the experiments of Detwiler (’20, ’22, ’23, ’23b, ’24, ’26, etc.) and Hooker (’15, ’16, ’17, ’23, and ’25). Detvnler transplanted a limb a considerable distance (4 or 5 segments) caudal to its natural insertion and even then saw the root fibers to it arise from the normal segments. Detwiler (’23b) and Wieman (’25) cut out a part of the embryonic cord, reset it in the cord in the reversed way or at an angle (sometimes with other tissue in between), and observed the behavior of the formerly descending and ascending tracts or root fibers. The first thing observed here, as in the previously mentioned researches of Burr, Detwiler, and May, was

90 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

that the outgrowth of a bundle was primary to neuroblastic proliferations, a result which appears to confirm Bok’s theory of the stimulating influence of the fibers on neuroblasts. Hooker found further that there was a marked tendency on the part of the regenerating fibers to avoid entering the opposite (that is, the

reverse) wound surface in a reset piece, and Wieman found that the development of the ascending tracts depended on a prior formation of descending ones through the operated segment. Such experiments as these are certain, to show new aspects of the problems of nerve growth.

That funetional fasciculation relationship is an important factor in determining the course of neuraxes and that simultaneous fimction plays a large part in the process of “fasciculation” (joining of neuraxes in one bundle) is most evident from the course of the optic nerve fibers in higher mammals and man. Whereas in fishes and all vertebrates below the mammals there is a total decussation of the optic nerves, in primates there is a hemidecussation, the temporal fibers of one eye running together with the nasal fibers of the other. This change in the course of

Fig. so. Total decussation of the optic fibers in a fish with little more than panoramic vision. Note the convergence in the optic tectum of those optic tract fibers, the visual fields of which join or overlap.

Fig. 51. Semidecussation of the optic fibers in man with binocular, convergent vision. Note that those fibers of the visual fields that overlap join in their course to the lateral geniculate nucleus (Gen.).

the optic nerve corresponds exactly with the collaboration of the nasal visual field of one eye with the temporal visual field of the other in producing convergent binocular vision (see fig. 51), which only occurs in mammals. A more simple example of the influence of simultaneous stimulation on the central arrangement of nerve fibers is seen in the fact that the cutaneous sensory fibers of the vagus.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEIVIENTS 91

after entering the bulb, immediately join the cutaneous sensory fibers of the descending trigeminal, whereas the taste fibers of the vagus join the taste fibers of the glossopharyngeal in their final course. So it appears that neurobiotaxis is a very active principle also in the collective arrangement of the nervous pathways. It may be of interest to discuss . —

briefly the cellular migrations oc- ^

curring in the sympathetic system. po.g. /

In the ontogenetic development /

of this system various migrations

of cells are observed, for which it 0 is not easy to furnish explanation. ^

According to the recent researches

of van Campenhout (’30, ’30a) and

Terni (’31) on the embryonic de- sp-g. ^ — v.

velopment of the sympathetic /

chain and its ganglia in the chick, ^ /

the cells which probably arise ^ /

from the neural crest (although l—^ /

according to Kuntz, ’29, certain

cells migrate along the ventral Fig. 52. Scheme of the development of the postganroot) first form what is known glionic neurons of the sympathetic nervous system (po.y.)

. , . ., , j 1 from the same primordium as the spinal ganglion cells

as a primary chain situated along (sp.g.) and their neurobiotatic migration as a consequence

cells migrate along the ventral Fig. 52. Scheme of the development of the postganroot) first form what is known glionic neurons of the sympathetic nervous system (po.y.)

. , . ., , j 1 from the same primordium as the spinal ganglion cells

as a primary chain situated along (sp.g.) and their neurobiotatic migration as a consequence the aorta, which location may of f**® postganglionic a.xon reflex (fully drawn arrow). _ .j . , , . , . The central reflex arc over the spinal ganglion cells and

evidence an interaction between preganglionic neurons (pr.g.), indicated by the dotted these cells and the arterial blood, arrow, is used less. The peripheral dendrites of the spinal Thn Ir. ganglion cclls are not drawn.

^ 1 • . , • V X xiic uru uver iiic auniui kuhkiiou cciia anu

evidence an interaction between preganglionic neurons (pr.g.), indicated by the dotted these cells and the arterial blood, arrow, is used less. The peripheral dendrites of the spinal The next step in the development

consists in the formation of the secondary or the final sympathetic chain, initiated by the budding in a central direction of cell groups from the primary chain. According to van Campenhout and Terni, this budding in a central direction coincides with a considerable outgrowth of preganglionic fibers that link up with these buds ; according to van Campenhout this shifting is a neurobiotactic process, the cells growing out into the direction of the impulse that reaches them through the preganglionic fibers. What is left of the primary sjmipathetic chain forms the juxtaneural and other peripheral ganglia of the sympathetic system. These cells do not shift in the direction of the effectors. However, as nerve cells shift in the direction of impulses, this seems to show that the majority of impulses acting on the juxtaneural postganglionic cells come from the effectors with which these postganglionic cells are connected. However, since these connections are axonic connections, this implies, if the explanation be correct, that

the majority of the reflexes acting on the more peripheral ganglia of the sympathetic system are axonic reflexes of the character described by Langley and Anderson (’95). The dominance in the viscera of axonic reflexes over central reflexes would seem to be in harmony with the scarcity of visceral afferent fibers in comparison with somatic afferent fibers, approximately only 5 to 8 per cent of the nerve fibers to the intestine being sensory (afferent) fibers (Ranson, ’31).

92

NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Aiiothor fcaturo on vvliicih ncurobiotiictie relations per!iai)s may j'ivc some light is the synaptic coiKlilion of the interneuronal eonneetions. In this relation, consideration should he given to the fact that impulses may pass the interneuronal connections in one direction oidy, that is, from axon to cell or dendrite and not from cell or dendrite to axon. In order to explain this f.ninjlcy and Andcrmni (’92) and Shcrrintjlon (’OG) accepted the pre.sence of a synaptic membrane between these two parts of the connection, a membrane that could be piussed in one direction only by nerve impulses. However, the jjre.sence of an irreversible membrane has not been demonstrated in the central nervous system. It would seem that the irreversible character of the synapse may be explained by the opposite neurobiotactic characters of the axon and the cell dendrites. .Since the growth of the axon is stimulo-concurrent and the growth of the dendrites and cell body stimulo-petal, the normal transmission of the impulse must neces.sarily bring both nearer to eacli other, while a reversetl transmission (if possible) apparently would draw them apart and finally l)reak the connection (see fig. .o2). .Although the consideration of these facts may not suHice to explain the irreversible character of the synapse, it at least s1k)ws that the irreversibility of the synaps,e is in perfect harmony witli the opposite stimulative tropism of a.xon and dendrite.

UKSUMK CONCI.USIO.NS

The phylogenetic differences in position of the cells of the motor nuclei suggest that the positions of the dendrites and the cell body are detennined by the impulses which reach them. Further researches show that the determining influence is evident only in such cells as have a previous or indirect afiinity for these impulses or lie in a region where these impulses accumulate. Tliis afiinity consists of simultaneous or successive states of action (stimulative correlation). Consequently a law governing nervous arrangement can be laid down. This law has long since been acknowledged to be one of the major laws governing the development of our mental capacities, that is, the law of mental association.

The acknowledgment of correlative function as the fundamental fact determining the arrangement of tlie cells and their dendrites raises the question as to whether the same law may be in effect in determining the final courses and connections of the neuraxes. A careful comparison of the regions where such paths begin and terminate shows that here, also, an associative affinity can be pointed out and that this affinity determines the place that the neuraxis will end. Under this fundamental law — that neurobiotactic processes occur between correlated systems — the tropisms of the dendrites and cell body take place in a direction opposite to that of the nerve current, that is, toward the center of stimulation. They are thus stimulo-petal, while the course of the impulse over the neuraxis is in the same direction as the direction of the axon current, that is, stimulo-fugal, or more correctly, stimulo-concurrent. That the development of the neura.xis is nevertheless a consequence of the stimulus has been proved by Bok (’15a), who showed that irradiating impulses activate neuroblasts in such a way as to form a neuraxis, so that here also a stimulogenous formation occurs. This correlation of stimuli thus plays a fundamental role in all processes of neurobiotaxis, although

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 93

the dendrites and cell bodies grow toward the stimulus while the neuraxis grows away from the stimulating center but under the influence of irradiating impulses from it. How, then, is it possible that a single nerve imit, the neuron, should show such clearly opposite polar differences, that one part of its protoplasm approaches the source of stimulation (the stimulo-petal dendrites and cell body) while the other grows in the direction of the stimulus or irradiation proceeding from it (stimulo-concurrent neuraxis) ?

The solution of this problem is aided by a consideration of other tropisms in nature which are more accessible to experimentation. A study of galvano-tropism reveals phenomena which are forcible reminders of the conditions just described for nervous tissue. Thus the root tips of plants grow toward the electro-negative pole and monocellular animals move in the same direction. This process, however, is reversible. By placing the amoeba, or the root tips of the growing plants, in a stronger solution of potassium or sodium chloride (which increases its conductivity), the tropisms become reversed and the movement or growth, as the case may be, is toward the positive pole. Lecithin also shifts its position in a galvanic current (kataphoresis) , but both it and albumen, under ordinary conditions (that is, under those conditions in which they occur in the animal body), move toward the positive pole. Addition of potassium enhances the anodic character of this process. Similar reversible kataphoretic phenomena are described with albuminoids, bacteria, and yeast cells. There is considerable evidence that the results of these galvanotropic and kataphoretic experiments may be used to explain the formation of the nervous system through the stimuli which act upon it. It is known that the surface of a nervous tract which is stimulated forms a negative pole, a cathode, with respect to the surrounding regions, which regions then represent an anodic field as compared with the center of stimulation. A neuroblast about this electro-negative center forms first an anodic offshoot or neuraxis, which will grow in the direction of the radiating current from the center of stimulation because of the anodo-tropic character of the cell protoplasm. The process thus formed probably derives its chemical and tropic characteristics from the potassium present within it. The potassium enhances the anodo-tropic, in these cases stimulo-concurrent, character of the neuraxis and besides increases its conductivity. In addition, potassium ions have a membrane loosening and surface tension lowering influence on protoplasmic membranes and thus favor its outgrowth.

The stimulo-petal, cathodic tropisms of the dendrites and of the perinuclear protoplasm are probably favored by the appearance of the Nissl bodies, as they do not take place until the neuron has developed its tigroid substance. This cathodic tropism, followed by a gradual shortening (contraction) of the dendrite and a displacing of the cell itself (as in cathodic tropisms with amoeba), occurs in accordance with the phenomena of cathodic stimulation, as stated in Pfliiger’s laws {Loeb and Maxwell, ’96 ; Boruttau, ’08) and as seen in such animal protoplasm as is susceptible to stimulation (e.g. the amoeba under normal conditions).

At the moment that the galvanic potential makes itself felt within the nervous system, there is probably a facilitation of stimulus-transition in the receptive

94 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

part of the cell and enhanced sensitivity at the cathode, as is well known in experimental neurology. Thus the first development and lengthening of the stimulo-concurrent neuraxes ” is a consequence of the increased anodo-tropic character of the tissue and is strengthened by potassium salts. The formation, and much later contraction, of the dendrites and the displacement of the perinuclear protoplasm toward the cathode is a special case of Pfliiger’s laws.

It is evident that this explanation of the dynamic polarization of the neuron gives no clew to factors determining the final connections of the neuraxis. This final connection always takes place in a territory or with a cell which has correlated activity, that is, exhibits just previous or simultaneous electric phenomena. Non-stimulated centers are all equally indifferent to it.

Monoaxonism is the resultant of all the forces (or stimulations) acting on a place of predilection (axon hillock). Polydendritism is not only possible but usual because the perinuclear and dendritic protoplasm are ever 3 rsvhere equally sensitive to cathodic influences and may respond at several different places to stimuli of as many different origins, each stimulus affecting the region nearest it. The trophic or metabolic tropism occasionally shown by the cell body or dendrites is not necessarily antagonistic to their stimulative tropism ; stimulation enhancing the metabolism of the cell, they may go hand in hand.

The ability of the neuron to receive at a given time different stimuli over as many distinct dendritic processes and to discharge the resultant of these over a single neuraxis may be considered the material expression of the property of the neuraxis of serving as a final common path.®

BIBLIOGRAPHY

AcHtjcARBO, N. 1911. Neuroglia y elementos intersticiales patologicos del cerebro, impregnados por los raetodos de reduccion de la plata o por sus modificasionis. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 9, p. 161.

. 1913. Notas sobre la estruetura y funciones de la neuroglia y en particular de

la neuroglia de la corteza cerebral humana. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 11, p. 187.

. 1915. Contribucion al estudio gliotectonico de la corteza cerebral. El asta

de Ammon y la fascia dentata. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 12, p. 229.

. 1915, De revolution de la nevroglie, et specialement de ses relations avec

I’appareil vasculaire. Trab, d. lab. de invest, biol. Univ. de Madrid, vol. 13, p. 169Adamkiewicz, A. 1881, 1882. Die Blutgefasse des menschlichen Riickenmarkes. I Theil: Die Gefasse der Riickenmarks-Substanz. II Theil: Die Gefasse der Riickenmarksoberflache. Sitzungsb. d. K, Akad. d. Wissensch. zu Wien, Math.nat. Cl,, Bd. 84, Abt. 3, S. 469 ; Bd. 85, Abt. 3, S. 101.

^Curiously, also, muscle tissue, subjected to a galvanic current, shows a shortening at the cathode and a slight relaxation or lengthening at the anode.

^ Granted that these conceptions are correct, it seems hardly necessary to emphasize that we do not believe that psychic contents or conscious realization ever could be explained by such considerations. They may, at best, give us an idea of some physico-chemical processes that accompany its evolution and explain the form in which our nervous elements appear. Although the laws of neurobiotaxis are probably parallel to the laws of psychological processes, the conscious revelations of life remain self-imposing. They are revealed by various processes but not explained by any phenomena whatever.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 95

Addison, W. 1911. The development of the Purkinje cells and of their cortical layers in the cerebellum of the albino rat. J. Comp. Neurol., vol. 21, p. 459.

Agduhr, E. 1916. Morphologischer Beweis der doppelten (plurisegmentalen) motorischen Innervation der Einzelnen quergestreiften Muskelfasern bei den Saugetieren. Anat. Anz., Bd. 49, S. 1.

. 1917. Triiningen inverkan pa den morfologiska bilden ov det motoriska nerven systemet. Hygiea.

. 1919. Is the post-embryonal growth of the nervous system due only to an

increase in size or also to an increase in number of the neurones? Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 2 I 2 , p. 944.

. 1919a. Are the cross striated muscle fibers of the extremities also innervated

sympathetically? Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 2 I 2 , p. 1231.

Aguerre, J. A. 1900. Untersuchungen fiber die menschliche Neuroglia. Arch. f. mikr. Anat., Bd. 56, S. 509.

Alcock, N. H., and Lynch, G. R. 1911. On the relation between the physical, chemical, and electrical properties of the nerves. Part IV. Potassium, chlorine, and potassium chloride. J. Physiol., vol. 42, p. 107.

Allen, W. F. 1917. Distribution of the spinal nerves in Polistotrema and some special studies on the development of spinal nerves. J. Comp. Neurol., vol. 28, p. 137.

Alzheimer, A. 1910. Beitrage zur Kenntnis der pathologischen Neuroglia und ihrer Beziehungen zu den Abbauvorgangen im Nervengewebe. Nissl-Alzheimer, histologische und histopathologische Arbeiten fiber die Grosshirnrinde, Jena, Bd. 3, S. 401.

Ambronn, H., and Held, H. 1895. Ueber Entwicklung und Bedeutung des Nen^enmarks. Sitzungsb. der kon. sachs. Gesellsch. d. Wissensch., Bd. 47, S, 38.

. 1896. Beitrage zur Kenntnis des Nervenmarks. I. Ueber Entwicklung und

Bedeutung des Nervenmarks. Arch. f. Anat. u. Physiol., Anat. Abt., H. 3 u. 4, S. 202.

II. Ueber Beobachtung an lebenden und frischen Nervenfasern und die Sichtbarkeit ihrer doppelten Konturierung. Ibid. S. 214.

Anderson, 0. 1891-92. Die Nerven der Schilddrfise. Biol. Foren. Forhandl. Verhandl. d. biol. Ver. in Stockholm, Bd. 4, S. 98.

Andriezen, W. 1893. The neuroglia elements in the human brain. Brit. Med. J., vol. 2, p. 227.

VON Anrep, B. 1880. Ueber die physiologische Wirkung des Cocain. Arch. f. d. ges. Physiol. (Pflfiger’s), Bd. 93, S. 134.

Antoni, N. 1908. Die Frage von einer neurofibrillaren Kontinuitat im Zentralnervensystem der Wirbeltiere. Folia neuro-biol., Bd. 2, S. 34.

Aoyagi, T. 1912. Zur Histologie des N. phrenicus, des Zweichfells und der motorischen Nervenendigung in demselben. Mitt. med. Fakult. k. Univ. Tokyo, vol. 10, p. 233.

VON Apathy, S. 1897. Das leitende Element des Nervensystems. Mitt. a. d. Zool. Stat. in Neapel, Bd. 12, S. 495.

. 1898. Ueber Neurofibrillen. Proc. Internal. Cong. Zool. Cambridge.

. 1907. Bemerkungen zu den Ergebnissen Ramon y Cajal’s hinsichtlich der

feineren Beschaffenheit des Nervensystems. Anat. Anz., Bd. 31, S. 481 ; S. 523.

. 1908. Der Vergleich der Neurofibrillen mit Protoplasmastromen oder Proto plasmafaden. Folia neuro-biol., Bd. 1, S. 289.

Ariens Kappers, C. U. 1904. Recherches sur le d5veloppement des gaines dans le tube nerveux. Petrus Camper. Nederl. bijdr. t. de anat., vol. 2, p. 223.

. 1907. Phylogenetische Verlagerungen der motorischen Oblongatakerne, ihre

Ursache und Bedeutung. Neurol. Centralbl., Bd. 26, S. 834. Rapport du Congrfis international de Psychiatrie et de Neurologie. Amsterdam, 1907.

96 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Akiens Kappers, C. U. 1908. The structure of the autonomic nervous system compared with its functional activity. J. Physiol., vol. 37, p. 139.

. 1908a. Weitere Mitteilungen iiber Neurobiotaxis. Die Selektivitat der Zellen wanderung. Die Bedeutung synchronischer Reizverwandtschaft. Verlauf und Endigung der zentralen sogenannten motorischen Bahnen. Folia neuro-bioL, Bd. 1, S. 507.

. 1908b. Ueber die Bildung von Faserverbindungen auf Grund von simultanen

und sukzessiven Reizen. Bericht des dritten Congresses fiir experimentelle Physiologie in Frankfurt a/Main.

. 1908c. Weitere Mitteilungen iiber Neurobio taxis. II. Die phylogenetische

Entwickelung des horizontalen Schenkels des Facialiswurzelknies. Folia neurobiol., Bd. 2, S. 255.

. 1909. Weitere Mitteilungen iiber Neurobiotaxis. III. Ueber den Einfluss

der Neurone der Geschmackskerne auf den motorischen Facialis- und Glossopharyngeuskern und ihr Verhalten zur Radix descendens Nervi quinti. Folia neuro-biol., Bd. 3, S. 259.

. 1910. Weitere Mitteilungen iiber Neurobiotaxis. IV. The migrations of the

abducens-nucleus and the concomitating changes of its root-6bers. Psychiat. en Neurol. BL, Amsterdam, Bd. 14, S. 271.

. 1910a. Weitere Mitteilungen iiber Neurobiotaxis. V. The migrations of the

motor cells of the bulbar Trigeminus, Abducens, and Facialis in the series of Vertebrates, and the differences in the course of their root-fibers. Verhandel. d. kon. Akad. V. Wetensch. te Amsterdam, Tweede Sectie, Deel 16, No. 4.

. 1911. Weitere Mitteilungen fiber Neurobiotaxis. VI. The migrations of the

motor rootrcells of the vagus group, and the phylogenetic differentiation of the hypoglossus nucleus from the spino-occipital-system. Psychiat. en Neurol. Bl., Amsterdam, Bd. 15, S. 408.

. 1912. Weitere Mitteilungen fiber Neurobiotaxis. VII. Die phylogenetische

Entwicklung der motorischen Wurzelkerne in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-Erganzungs-Heft, S, 1.

. 1913. Phenomena of neurobiotaxis in the central nervous system. Sect. Anat.

and Embryol. 17th Intemat. Cong. Med., London.

. 1914. Weitere Mitteilungen fiber Neurobiotaxis. VIII. Ueber die motorischen

Glossopharyngeus und Facialis bei niedem Vertebraten. Folia neuro-biol., Bd. 9, S. 383.

. 1917. Further contributions on Neurobiotaxis. IX. An attempt to compare

the phenomena of neurobiotaxis with other phenomena of taxis and tropism. The dynamic polarization of the neuron. Psychiat. en Neurol. Bl., Amsterdam, Bd. 20, S. 422; also J. Comp, Neurol,, vol, 27, p. 261.

. 1917a. Verschijnselen van Neurobiotaxis in het optische Stelsel. Psychiat. en

Neurol. BL, Amsterdam, Bd. 21, S. 320.

. 1917b. Versuch einer Erklarung des Verhaltens an der Synaps. Psychiat. en

Neurol. BL, Amsterdam, Bd. 21, S. 440.

. 1919. The logetic character of growth. J. Comp. Neurol., vol. 31, p. 51.

. 1919a. Phenomena of neurobiotaxis as demonstrated by the position of the

motor nuclei of the oblongata. J. Nerv. and Ment. Dis., vol. 50, p. 1.

. 1919b. Die Bildung von Bogenfasem als primare Reflexwege der vitalen (proto pathischen) Empfindungen. Feestbundel Dr. Kerbert (in Bijdragen tot de Dierkunde uitgegeven door het Zool. Genootschap, Amsterdam).

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) Bohn, Haarlem.

. 1925. The meninges in cyclostomes, selachians and teleosts compared with those

in man. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 28, no. 1, p. 72.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 97

ARiiiNs I^APPEUS, C. U. 1926. The meninges in lower vertebrates compared with those in mammals. Mch. Neurol, and Psychiat., vol. 15, p. 281.

. 1926. Sur la possibilitd d’une localisation ou specialisation intracellulaire de

quelques fonctions psychiques. Rapport du Congr^s international de Psychology. Groningen.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

. 1932. Principles of development of the nervous system (neurobio taxis).

Section II in Cytology and cellular pathology of the nervous system, edited by W. Penfield. P. Hoeber, New York.

. 1933. La correlation entre le prociis de conduction et la croissance du nerf.

Volume JuhUaire for Marinesco, p. 353.

Aspluxd, M. 1890. Zur Kenntnis der Verbindung des Ruckenmarkes mit der Pia mater. Nord. med. Ark., vol. 22, no. 4, p. 1.

Athi.vs, M. 1905. Anatomia da Cellula nervosa. Lisboa.

Aueubach, L. 1898. Nervenendigung in den Centralorganen. Neurol. Centralbl., Bd. 17, S. 445.

. 1904. Extra- sowie intracellulare Netze nervoser Natur in den Centralorganen

von Wirbeltieren. Anat. Anz., Bd. 25, S. 47.

Ayeiis, H. 1896. The origin and growth of brain cells in the adult body. J. Comp. Neurol, vol. 6, p. 211.

Bailey, P., and Cushing, H. W. 1926. A classification of the tumors of the glioma group on a histogenetic basis. J. B. Lippincott Co., Philadelphia and London.

Bailey, P., and Schaltenbiiand, G. 1927. Die mukose Degeneration der Oligodendroglia. Zetischr. f. Ncrvenheilkunde, Bd. 97, S. 231.

Bailey, F. R., STnoNO, 0., and Elwyn, A. 1925. Textbook of histology. Wm. Wood and Co., New York.

Balfour, F. M. 1876. On the development of the spinal nerves in the elasmobranch fishes. Proc. Roy. Soc., London.

Bancroft, F. W. 1904. Note on the galvanotropic reactions of the medusa Polyorchis penicillata. J. Exper. Zool, vol. 1, p. 289.

Bardee.n, C. R. 1903. The growth and histogenesis of the cerebrospinal nerves in mammals. Am. J. Anat., vol. 2, p. 231.

. 1910. The development of the skeleton and of the connective tissues. Keibel

and Mall, Manual of human embryology. J. B. Lippincott Co., Philadelphia and London, vol. 1, p. 292.

Barker, L. F. 1899. The nervous system and its constituent neurones. Appleton and Co., New York.

Bartelxiez, G. W. 1915. Mauthner’s cell and the nucleus motorius tegmenti.

J. Comp. Neurol, vol. 25, p. 87.

. 1920. The morphology of the synapse in vertebrates. Arch. Neurol, and

Psychiat., vol. 4, p. 122.

Bartelxiez, G. W., and Hoerr, N. L. The vestibular club endings in Ameiurus.

J. Comp. Neurol, vol. 57, p. 401.

Baum, J. 1900. Beitrage zur Kenntnis der Muskelspindeln. Anat. Hefte, vol. 13,

S. 249.

Beccari, N. 1907. Ricerche sulle cellule e fibre del Mauthner e sulle loro connessioni in pesci ed anfibii. (Salmo fario, S. irideus e Salamandrina perspicillata). Arch, ital. di anat. e di embriol, vol. 6, p. 660.

. 1915. Contribution A I’^tude des fonctions des cathions Na, K et Ca dans le

tissu musculaire. Arch. ital. de biol, vol. 63, p. 293.

. 1930. Diffeienze di grandezzaedi forma in rapporto von I’eta nello terminazioni

a coppa del nucleo del nervo oculomotore nei Pisci, Teleostei. Monitore zool. ital, vol. 5, p. 132.

98 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Becht, F. C., and Gunnar, H. 1921. Studies on 'the cerebrospinal fluid. VII. A study of the volume changes of the cerebrospinal fluid after adrenalin, pituitrin, pilocarpin, and atropin. Am. J. Physiol., vol. 56, p. 231.

Behr, C. 1917. Ueber die Ernahrung des sehnerven in physiologischer Beziehung und als Ursache der Unheilbarkeit der tabisehen Sehnervenatrophie. Klin. MonatsbI. f. Augenh., Bd. 59, S. 241.

Benedikt, M. 1873. Ueber die Innervation des Plexus chorioideus inferior. Virchow's Arch. f. path. Anat., Bd. 59, S. 395. According to Clark also in Schmidt's Jahrbuchers, 1874, Bd. 163, S. 124.

Bergen, F. 1904. Zur Kenntnis gewisser Strukturbilder (“Netzapparate," “Saftkanalchen," “Trophospongien") im Protoplasma verschiedener Zellenarten. Arch, f. mikr. Anat., Bd. 64, S. 498.

Bergman. 1927. Die cellen van Hortega. Dissertatie, Utrecht.

Berkeley, H. J. 1894. The cerebellar cortex of the dog. Johns Hopkins Hospital Reports, vol. 3, p. 195.

Bernard, H. M., and Cantab, M. A. 1903. Studies on the retina. Part VI. The continuity of the nerves through the vertebrate retina. Quart. J. Micr. Sc., vol. 47, p. 303.

Bethb, a. 1901. Ueber die regeneration peripherischen Nerven. Neurol. Centralbl., Bd. 20, S. 720.

. 1903. Allgemeine Anatomic und Physiologic des Nervensystems. Thieme,

Leipzig.

. 1904. Der heutige Stand der Neurontheorie. Deutsche med. Wochenschr.,

Bd. 30, S. 1201.

. 1906. Bemerkungen zur Zellkettentheorie. Anat. Anz., Bd. 28, S. 604.

. 1907. Ein neuer Beweis fur die leitende Funktion der Neurofibrillen. Folia

neuro-biol., Bd. 1, S. 100.

. 1908. Die Nervenregeneration und die Verheilung durchschnittener Nerven.

Folia neuro-biol., Bd. 1, S. 63.

Bielschowsky, M. 1905. Die histologische Seite der Neuronlehre. J. f. Psychol, u. Neurol., Bd. 5, S. 128.

. 1907. Ueber sensible Nervenendigungen in der Haut zweier Insectivoren (Talpa

europaea und centetes ecaudatus). Anat. Anz., Bd. 31, S. 187.

. 1928. G. von Mollendorf's Handbuch der mikroskopischen Anatomie des

Menschen. I. Nervengewebe, S. 1. J. Springer, Berlin.

VAN Biervliet, j. 1900. La substance chromophile pendant le cours du d^veloppement de la cellule nerve use (chromolyse physiologique et chromolyse experimentale). Le Nevraxe, vol. 1, p. 31.

Bing, R., and Burckhardt, R. 1904. Das Centralnervensystem von Ceratodus Forsteri. Anat. Anz., Bd. 25, S. 588. 1905 in Semon. Zool. Forsch. in Janaische Denkschr., Bd. 4.

Binzwanger, 0., AND Berger, H. 1898. Beitrage zur Kenntniss der Lymphcirculation in des Grosshimrinde. Virchow's Arch. f. path. Anat., Bd. 152, S. 525.

Black, Davidson. 1916, 1922. See bibliography for Chap. V.

Bland-Sutton, j. 1922. On the choroid plexuses and ventricles of the brain as a secreting organ. Lancet, vol. 204, p. 1143.

Boeke, j. 1902. On the structure of the light-percepting cells in the spinal cord, on the neurofibrillae in the ganglion cells and on the innervation of the striped muscles in Amphioxus lanceolatus. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 5, p. 350.

_ 1907. On the structure of the nerve-cells in the central nervous system of

Branchiostoma lanceolata. First communication. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 10‘, p. 86,

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 99

Boeke, J. 1908. On the structure of the ganglion-cells in the central nervous system of Branchiostoma lanceolata. Second communication. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 11*, p. 53.

. 1909. Over vorm en ontwikkeling van de motorische eindplaten bij de hoogere

vertebraten. Verslag kon. Akad. v. Wetensch. te Amsterdam, vol. 17-, p. 1008.

. 1910. Die motorische Endplatte bei den hoheren Vertebraten, ihre Entwick lung. Form und Zusammenhang mit der Muskelfaser. Anat. Anz., Bd. 35, S. 193.

. 1910a. Ueber eine aus marklosen Fasem hervorgehende zweite Art von hypo lemmalen Nervenendplatten bei den quergestreiften Muskelfasern der Vertebraten. Anat. Anz., Bd. 35, S. 481.

. 1910b. Demonstration von Praparaten mit periterminalem Netzwerk und

akzessorischen Fasern. Briisseler Anat. Kongress. Verhandl. d. anat. Gesellsch. (Ergiinzungsheft zum Bd. 37, Anat. Anz.), Bd. 24, S. 178.

. 1911. Beitriige zur Kenntniss der motorischen Nervenendigungen. I. Die

Form und Struktur der motorischen Endplatte der quergestreiften Muskelfasern bei den hoheren Vertebraten. II. Die akzessorischen Fasern und Endplattchen. Internal. Monatschr. f. Anat. u. Physiol., Bd. 28, S. 377.

. 1912. Ueber De- und Regeneration der motorischen Endplatten und die doppelte

Innervation der quergestreiften Muskelfasern bei den Saugetieren. Verhandl. d. anat. Gesellsch. (Ergiinzungsheft zum Bd. 41, Anat. Anz.), Bd. 26, S. 149.

. 1913. Die doppelte (motorische imd sympathische) efferente Innervation der

quergestreiften Muskelfasern. Anat. Anz., Bd. 44, S. 343.

. 1915. On the termination of the efferent nerves in plain muscle-cells, and its

bearing on the sympathetic (accessory) innervation of the striated muscle-fiber. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 17^, p. 982.

. 1916. Studien zur Nervenregeneration. I. Die Regeneration der motorischen

Nervenelemente und die Regeneration der Nerven der Muskelspindeln. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, Deel 18, no. 6.

. 1917. Studien zur Nervenregeneration. II. Die Regeneration nach Vereini gung vergleichartiger Nerv'enstiicke (heterogene Regeneration), und die Funktion der Augenmuskel und Zungennerven. Die allgemeinen Gesetze der Nervenregeneration. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, Deel 19, no. 5.

. 1921. The innervation of striped muscle fibers and Langley’s receptive substance.

Brain, vol. 44, p. 1.

. 1922. Zur Innervation der quergestreiften Muskeln bei den Ophidiern. “ Termi nazioni a Grappolo.” Sympathische Innervation. Libro en Honor de S. Ramon y Cajal, vol. 1, p. 113.

. 1924. The innervation of the muscle fibers of the myocardium and of the atrioventricular bundle of this in the heart of the tortoise. Proc. Roy. Acad. Sc., Amsterdam, Bd. 28, S. 32.

. 1925. Die Innervation des Herzbeutels und Hisschen Biindels beim Schild krotenherz. Verslag. kon. Akad. v. Wetensch. te Amsterdam, vol. 33, no. 9, p. 933.

. 1925a. Die intrazellulare Lage der Nervenendigungen im Epithel und ihre

Beziehungen zum Zellkern. Jahrb. L Morphol. u. mikr. Anat., Abt. 2, Zeitsehr. f. mikr.-anat. Forschung, Bd. 2, S. 341.

. 1926. Die Beziehungen der Nervenfasem zu den Bindegewebselementen und

Tastzellen. Das periterminale Netzwerk der motorischen und sensiblen Nervenendigungen, seine morphologische und physiologisehe Bedeutung, Entwicklung und Regeneration. Jahrb. f. Morphol. u. mikr. Anat., Abt. 2, Zeitsehr. f. mikr.anat. Forschung, Bd. 4, S. 448.

. 1927. Die morphologische Grundlage der sympathischen Innervation der quergestreiften Muskelfasern. Jahrb. f. Morphol. u. mikr. Anat., Abt. 2, Zeitsehr. f. mikr.-anat. Forschung, Bd. 8, S. 561.

100 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Boeke, J. 1930. Some remarks on the papers by H, J. Wilkinson on the innervation of striped-muscle fibers. J. Comp. Neurol., vol. 51, p. 299.

. 1932. Some observations on the structure and innervation of smooth muscle

fibers (arrectores spinarum of the hedgehog) and blood vessels. J. Comp. Neurol., vol. 56, p. 27.

Boeke, J., and Dusseu de Babenne, J. G. 1919. The sympathetic innervation of the cross-striated muscle fibers of vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 2P, p. 1227.

Boeke, J., and de Groot, G. J. 1908. Physiological regeneration of neurofibrillar endnets (tactile discs) in the organ of Eimer in the mole. Kon. Akad v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 10^, p. 452.

DE Boer, S. 1913. Die quergestreiften Muskeln erhalten ihre tonische Innervation mittels der Verbindungsaste des S 3 unpathicus (thoracales autonomes System). Folia neuro-biol., Bd. 7, S. 378.

. 1913. Ueber den Skelettmuskeltonus, 2*® Mitteilung. Die tonische Innervation

der quergestreiften Muskeln bei Warmblutem. Folia neuro-biol., Bd. 7, S. 837. Bok, S. T. 1915. Die Entwicklung der Himnerven und ihrer zentralen Bahnen. Die stimulogene Fibrillation. Folia neuro-biol., Bd. 9, S. 475.

. 1915a. Stimulogeneous fibrillation as the cause of the structure of the nervous

system. Psychiat. en neurol. BL, Amsterdam, Bd. 19, S. 393.

. 1917. The development of reflexes and reflex tracts. I. The reflex-circle.

Psychiat. en neurol. BL, Amsterdam, Bd. 21, S. 281.

Boll, F. 1877. Ueber Zersetzungsbilder der markhaltigen Nervenfaser. Arch. f. Anat. u. Physiol., Anat. Abt., 1877, S. 288.

Boruttau, H. 1908. Taschenbuch der Physiologie. In 2 Teil. Klinkhardt, Leipzig. Botezat, E. 1897. Die Nervenendigungen an den Tasthaaren von Saugethieren. Arch. f. mikr. Anat., Bd. 50, S. 142.

. 1906. Die Nervenendapparate in den Mundteilen der Vogel und die einheitliche

Endigungsweise der peripheren Nerven bei den Wirbeltieren. Zeitschr. f. wissensch. ZooL, Bd. 84, S. 205.

. 1907. Die fibrillare Struktur von Nervenendapparaten in Hautgebilden. Anat.

Anz., Bd. 30, S. 321.

. 1907. Beitrage zur Kenntnis der Nervenenden in der Mundschleimhaut. Anat.

Anz., Bd. 31, S. 575.

. 1908. Die sensiblen Apparate und die Geschmacksknospen der Vogel. Vers.

d. deutschen Nat. u. Arzte.

. 1910. Fa.sern und Endplatten von Nerven zweiter Art an den gestreiften

Muskeln der Vogel. Anat. Anz., Bd. 35, S. 396.

. 1912. Die Apparate der Gefuhlsinnes der nackten und behaarten Saugetierhaut,

mit Berucksichtigung des Menschen. Anat. Anz., Bd. 42, S. 193 ; S. 273.

Bozler, E. 1927. Unter.suchungen uber das Nerven.system der Coelenteraten. I. Kontinuitat oder Kontakt zwischen der Nervenzellen. Zeitschr. f. wissensch. Biol., Abt. B. : Zeitschr. f. Zellforsch. u. mikr. Anat., Bd. 5, S. 244.

Brads, H. 1911. Die Entstehung der Nervenbahnen. F. C. W. Vogel, Leipzig. Bremer, L. 1882. Ueber die Endigungen der markhaltigen und marklo.sen Nerven im quergestreiften Muskel. Arch. f. mikr. Anat., Bd. 21, S. 105,

. 1883. Ueber die Mu.«kelspindeln nebst Bemerkungen uber Structur, Neubildung

und Innervation der quergestreiften Muskelfascr. Arch. f. mikr. Anat., Bd. 22, S. 318.

Brookover, C. 1910. The olfactory nerve, the nervus terininalis and the prcoptic •syrnpatlietic .system in Arnia calva. J. Comp. Neurol., vol. 20, p. 49 (.sec p. 96). BrOc'KE, E. T. 1918. Neucre An.schauungen uber den Muskeltonus. Deutsclie med. Wochenschr., Bd. 44-, S. 121 u. 152,

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 101

Bruno, G. 1931. Sulla struttura e su alcune proprieta fisiche della fibra nervosa midollata. Arch. Ital. Anat. e Embriol., vol. 29, p. 1.

BtiHLER, A. 1895. Protoplasma-Structur in Vorderhirnzellen der Eidechse. Wurzburg.

. 1898. Untersuchungen uber den Bau der Nervenzellen. Wurzburger med.

Zeitschr., N. F., Bd. 31, S. 285.

VON BiiNGNER, 0. 1891. Ueber die Degenerations und Regenerations-vorgange am Nerven nach Verletzungen. Beitr. z. path. Anat. u. z. allg. Path. (Ziegler’s Beitr ) Bd. 10, no. 4, S. 321.

Burger, 0. 1891. Beitrage zur Kenntniss der Nervensystems der Wirbellosen : Neue Untersuchungen iiber das Nervensystem der Nemertinen. Mitt. a. d. zool. Station zu Neapel, Leipzig, Bd. 10, S. 206.

Burr, H. S. 1916. Regeneration in the brain of Amblystoma. I. The forebrain. J. Comp. Neurol., vol. 36, p. 203.

. 1916a. The effect of the removal of the nasal placodes on Amblystoma embryos.

J. Exper. Zool., vol. 20, p. 27.

. 1920. The transplantation of the cerebral hemispheres in Amblystoma.

J. Exper. Zool., vol. 30, p. 159.

. 1923. Experimental hyperplasia of the cerebral hemispheres in Amblystoma.

Anat. Rec., vol. 25.

. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol.,

vol. 45, p. 33.

Burrows, M. T. 1911. The growth of tissues of the chick embryo outside the animal body, with special reference to the nervous system. J. Exper. Zool., vol. 10, p. 63.

Butler, S. 1880. Unconscious memory. D. Bogue, London.

Cameron, J. 1906. The histogenesis of nerve fibers, a cytological study of the embryonic cell-nucleus. J. Anat., vol. 40, p. 3.

VAN Ca.mpenhout, E. 1930. Historical survey of the development of the sjunpathetic nervous system. Quart. Rev. Biol. vol. 5, pp. 23 and 217.

. 1930a. Contribution to the problem of the development of the sympathetic

nervous system. J. Exper. Zool., vol. 56, p. 295.

Cappelletti, L. 1900. L’efflusso del Uquido cerebro-spinale dalla fistola cefalorachidiana in condizioni normali e sotto I’influenza di alcuni farmaci. Atti d. Accad. d. sc. med. e. nat. in Ferrara, vol. 74, p. 85.

. 1901. L’ecoulement du liquide cerebro-spinal par la fistule cephalo-rachidienne

en conditions normales et sous I’influence de quelques medicaments, .^irch. ital. de biol., vol. 36, p. 299.

Carlson, A. J. 1904. Contributions to the physiology of the ventral nerve cord of Myriapods (Centipedes and Millipedes). J. Exper. Zool., vol. 1, p. 269.

. 1905. Further evidence of the fluidity of the conducting substance in nerve.

Am. J. Physiol., vol. 13, p. 351.

Carlson, A. J., and Braafladt, L. H. 1915. Contributions to the physiology of the stomach. XVIII. On the sensibility of the gastric mucosa. Am. J. Physiol., vol. 36, p. 153.

Carpenter, F. W. 1918. Nerve endings of sensory type in the muscular coat of the stomach and small intestine. J. Comp. Neurol., vol. 29, p. 553.

C.arpenter, F. W., and Conel, J. L. 1914. A study of ganglion cells in the sympathetic nervous system, with special reference to intrinsic sensory neurones. J. Comp. Neurol., vol. 24, p. 269.

Carpenter, F. W., and Main, R. C. 1907. The migration of medullary cells into the ventral nerve-roots of pig embryos. Anat. Anz., Bd. 31, S. 303.

de Castro, F. 1916. Nota sobre la disposici6n del aparato reticular do Golgi en los botones gustativos. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 14, p. 107.

102 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

De Castro, F. 1920. Estudios sobre la neuroglia de la corteza cerebral del hombre y de los animales. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 18, p. 1,

. 1922. Contribucidn al conocimiento de la innervacidn del pancreas. Libro en

honor de S, Ramon y Cajal, Bd. 1.

Cathelin, F. 1912. La circulation du liquids c4phalo-rachidien. J. B. Bailli^re, Paris.

Cattaneo, A. 1888. Organes nerveux terminaux musculo-tendineux, leurs conditions normales et leur manifere de se comporter apr6s la section des racines nerveuses et des nerfs spinaux. Arch. ital. de biol., vol. 10, p. 337,

Cecchbrelli, G. 1902. Sulle piastre motrice e sulle fibrille ultraterminali nei muscoli della lingua di Rana esculenta. Monitors zool. ital., vol. 13.

. 1904. Sulle “terminazioni nervose a paniere” del Giacomini, nei muscoli dorsali

degli Anfibi anuri adulti. Anat. Anz., Bd. 24, S. 428.

. 1908. Gontributo alia conoscenza delle espansioni nervose di senso nella mucosa

del cavo orale e della lingua dell’uomo. Internal. Monatschr. f. Anat. u. Physiol., Bd. 15, S. 273.

Chambers, R. 1924. The physical structure of protoplasm as determined by microdissection and injection. General Cytology, edited by Cowdry, sect. 5, p. 237. Univ. of Chicago Press, Chicago.

Chandler, A. C. 1911. On a lymphoid structure lying over the myelencephalon of Lepidosteus. Univ. California Publ,, Zool., vol. 9, p. 85.

Charlton, H. H. 1928. A gland-like ependymal structure in the brain. Kon. Akad. V. Wetensch. te Amsterdam, Proc. sect, sc., vol. 3P, p. 1.

Chase, M. R., and Ranson, S. W. 1914, The structure of the roots, trunk, and branches of the vagus nerve. J. Comp. Neurol., vol. 24, p, 31.

Child, C. M. 1921. The origin and development of the nervous system. Univ. of Chicago Press, Chicago.

. 1924. Physiological foundations of behavior. Henry Holt and Co., New York.

Chvostek, F. 1880. Weitere Beobachtungen von sehr zahlreichen und sehr grossen Kalkplattchen in der Arachnoidea spinalis. Wien. med. Presse, Bd. 21, S. 1625; S. 1661.

CiAccio, G. V. 1890. Intorno alle piastre nervose finali nei tendini dei vertebrati. Mem. r. Accad. d. sc. d. 1st. di Bologna, Serie 4, vol. 10, p. 301.

CiLiMBARis, A. 1910. Histologische Untersuchungen iiber die Muskelnspindeln der Augenmuskeln. Arch. f. mikr, Anat., Bd. 75, S. 362.

Clark, S. L. 1928. Nerve endings in the choroid plexus of the fourth ventricle. J. Comp. Neurol., vol. 47, no. 1, p. 1.

CoEHN, A., AND Barratt, W. 1905. Ueber Galvanotaxis vom Standpunkte der physikalischen Chemie. Zeitschr. f. allg. Physiol., Bd. 5, S. 1.

CoGGi, A. 1902. Sullo sviluppo del sistema nervoso periferico dei Vertebrati. Monitore zool. ital.

CoGHiLL, G. 1929. Anatomy and the problem of behavior. Cambridge Univ. Press.

Cowdry, E. V. 1912. Mitochondria and other cytoplasmic constituents of the spinal ganglion cells of the pigeon. Anat. Rec., vol. 6, p. 33.

. 1913. The relations of mitochondria and other cytoplasmic constituents in spinal

ganglion cells of the pigeon. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 29, S. 473.

. 1913a. Les mitochondries dans les cellules des ganglions spinaux trait^s par la

m6thode de Bensley. Eibliog. anat., vol. 23, p. 311.

. 1914. The comparative distribution of mitochondria in spinal ganglion cells of

vertebrates. Am. J. Anat., vol. 17, p. 1.

1914a. The development of the cytoplasmic constituents of the nerve cells of

the chick. I. Mitochondria and neurofibrils. Am. J, Anat., vol. 15, p. 389.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 103

CowDRY, E. V. 1916. The structure of chromophile cells of the nervous system. Contrib. Embryol., Publ. Carnegie Inst. Washington, vol. 4, no. 11, p. 27.

. 1918. The mitochondrial constituents of protoplasm. Contrib. Embryol.,

Publ. Carnegie Inst. Washington, vol. 8, p. 39. ’

. 1924. Cytological constituents — mitochondria, Golgi apparatus, and chromidial

substance. General Ci^ology (edited by Cowdry), sect. VI, Univ. of Chicago Press.

Cox, W. H. 1898. Experimenteele bijdragen tot de pathologische histologie en physiologic der gangliencellen. I. De spinaal-gangliencel. Psychiat. en Neurol. BL, Amsterdam, Bd. 2, S. 189.

. 1898. De zelfstandigheid van der zenuw-fibrillen in het neuron. Eene studie

over het granulanet en de fibrillen der spinaalgangliencel. Psychiat. en Neurol. BL, Amsterdam, Bd. 2, S. 371.

Crevatin, F. 1901. Sulle fibre nervose ultraterminali. Arch. d. sc. d. 1st. di Bologna. Feb. 10.

CuccATi, G. 1888. Delle terminazione nervose nei muscoli abdominali della rana temporaria e della Rana esculenta. Intemat. Monatschr. f. Anat. u. Physiol., Bd 5, S. 337.

Cushing, H. 1902. Physiologische und anatomische Beobachtungen fiber den Einfluss von Himkompression auf den intracraniellen Kreislauf und fiber einige hiermit verwandte Erscheinungen. Mitt. a. d. Grenzgeb. d. Med. u. Chir., Jena, Bd. 9.

. 1902. Some experimental and clinical observations concerning states of increased

intracranial tension. Am. J. Med. Sc., vol. 124, p. 375.

. 1914. Studies on the cerebrospinal fluid. I. Introduction. J. Med. Research,

vol. 31 (new series vol. 26), p. 1.

Dandy, W. E., and Blackfan, K. D. 1913. An experimental and clinical study of internal hydrocephalus. J. A. M. A., vol. 61, p. 2216.

, 1914. Internal hydrocephalus : an experimental, clinical, and pathological study.

Am. J. Dis. Children, vol. 8, p. 406.

Dehler, A. 1895. Beitrag zur Kenntnis vom feineren Bau der sympathischen GanglienzeUe des Frosches. Arch. f. mikr. Anat., Bd. 46, S. 724.

Deineka, D. 1908. ITnfluence de la temperature ambiante sur la regeneration des fibres nerveuses. Folia neuro-bioL, Bd. 2, S. 13.

Del Rio Hortega, P. 1916. Contribution A 1’ etude de I’histopathologie de la nevroglie. Les variations dans le ramollissement cerebral. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 14, p. 1.

. 1916a. Estructura fibrilar del protoplasma neuroglico y origine de las gliofi brillas. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 14, p. 269.

■ . 1916b. Gliosomas y gliofibrillas. Bol. d. 1. Soc. Esp. d. Biol.

. 1916c. Estudios sobre el centrosoma de las ceiulas nerviosas y neuroglicas de

los vertebrados, en sus formas normal y anormales. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 14, p. 117.

. 1917. Alteraciones de la neuroglia en la intoxicacion por pilocarpine. Lab.

Rev. de Ciene. Biol, y Med. Esp. Barcelona.

. 1920. La microglia y su transformacion en ceiulas en bastonuto y cuerpos

granulo-adiposos. Arch, de neurobiol . ; also Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 18, p. 37.

. 1921. El “Tercer Elemento” de los nendosos Histogenesis y evolucifin normal,

e.xodo y distribucion regional de la microglia. Mem. de la real. Soc. espagn. de hist, natur., vol. 2, p. 1.

. 1921a. Histogenesis y evolucion normal ; e.xodo y distribucifin regional de la

microglia. Mem. Real. Soc. Esp. d. Hist. Nat., vol. 11, p. 213.

■ . 1922. Son homologables la glia de es casas radiciones (oligodendroglia). Bol.

de la Real. Soc. Esp. d. Hist. Nat., January. (Quoted from Penfield ’28.)

104 NERVOUS SYSTEMS OF VERTEBRATES AND OF AIAN

Del Rio Hortega, P. 1925. Condrioma y granulaciones espdcificas de la celulas Deuroglicas. Bol. Real. Soc, Esp. d. Hist. Nat., vol. 25, p. 34.

. 1928. Tercera aportacion al conocimento morpholdgico e interpretacion functional de la oligodendroglia. Mem. Real. Soc. Esp. d. Hist. Nat., vol. 14, p. 5.

Del Rio Hortega, P., and Penfield, W. 1926. Cerebral cicatri.\'. The reaction of neuroglia to brain wounds. (Quoted from Penfield ’28.)

Detweiler, S. R. 1920. Experiments on the transplantation of limbs in Amblystoma. The formation of nerve plexuses and the function of the limbs. J. Exper. Zool., vol. 31, p. 117.

. 1922. Experiments on the transplantation of limbs in Amblystoma. Further

observations on peripheral nerve connections. J. Exper. Zool., vol. 35, p. 115.

. 1923. Experiments on the transplantation of the spinal cord in Amblystoma,

and their bearing upon the stimuli involved in the differentiation of nerve cells. J. Exper. Zool., vol. 37, p. 339.

. 1923a. Studies on the retina. An expierimental study of the gecko retina.

J. Comp. Neurol., vol. 36, p. 125.

. 1923b. E.xperiments on the transplantation of the spinal cord in Amblystoma

and their bearing upon the stimuli involved in the differentiation of nerve cells. J. Exper. Zool., vol. 37, p. 339.

. 1924. Further observation on proliferation of nerve cells in grafted units of

spinal cord. Anat. Rec., vol. 27, p. 87.

. 1926. The effect of the reduction of skin and muscle on the development of

spinal ganglia. J. Exper. Zool., vol. 45, p. 399.

Dewey, K. W. 1918. A contribution to the study of the pathways of the cerebrospinal fluid and the choroid ple.xus. Anat. Rec., vol. 15, p. 1.

Diamare, V. 1932. Una revendicazione Italiana riguardo all incisure di Schmid tLantermann della fibra nerviosa midollata. Rend. Accad. Sci. Fis. Mat. (Cl. Soc. R. Napoli), vol. 1, p. 82.

Disse, j. 1897. Die erste Entwickclung dcs Riechnerven. Anat. Hefte, Bd. 9, S. 255.

Dogiel, a. S. 1890. Mcthylnblautinction der motorischen Nerv’cnendigungen in den Muskeln der Amphibien und Reptilien. Arch. f. mikr. Anat., Bd. 35, S. 305.

. 1891. Die Nervenkorperchen (Endkolben v. Krause) in der Cornea und Conjunctiva bulbi des Mcnschen. Arch. f. mikr. Anat., Bd. 37, S. 602.

. 1892. Die Nervenendigungen in Meissnerschen Tastkorperchen. internat.

Monatschr. f. Anat. u. Physiol., Bd. 9, S. 76.

. 1893. Die Nervenendigungen in der Haut der iiusseren Genitalorgano dcs

Mcnschen. Arch. f. mikr. Anat., Bd. 41, S. 585.

. 1893a. Die Nervenendigungen in der Thriinendrusc dor Saugethiere. Arch. f.

mikr. Anat., Bd. 42, S. 632.

. 1895. Die Structur der Nervenzcllen der Retina. Arch. f. mikr. Anat., Bd. 46,

S. 391.

. 1895a. Zur Fragc viber den fcincrcn Bau dcs sympathischen Ncrven.systems bci

den Siiugethieren. Arch. f. mikr. Anat., Bd. 46, S. 305.

. 1890. Zwei Arten sympathEcher Nervenzcllen. Anat. Anz., Bd. 1 1, S. 079.

. 1898. Die scnsiblen Nervenendigungen iin Herzen und in den BIutgef;u-.H'n der

Saugethiere. Arch. f. mikr. Anat., Bd. 52, S. 41.

. 1899. Zur Frage libcr den Bau der Herbat’.schen Korperchen und die Methylen hlaufi.xirung nacli Bethe. Zeil.'-chr. f. wis.'-en.sch. Zool., Bd. 00, S. 358.

. 1899.1. Zur Fragc uher den feineren Bau dor Herzganglien dcs .Men.-chen und

der S.iugethierc. Arch. f. mikr. Anat., Bd. 5:1, .8. 237.

. 1902. Die Nervenendigungen iin Bauchfell, in den Selinen, den .Mu.kelpindeln

und dem Centrum tendinemn des Diaphragmiis heim Men.-chen und hel .S.iugethieren. Arch. f. mikr. Anat., Bd. 59, S. 1.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 105

Dogiel, a. S. 1904. Ueber die Nervenendigungen in den Grandryschen und Herbstschen Korperchen im Zusammenhange mit der Frage der Neuronentheorie. Anat. Anz., Bd. 25, S. 558.

. 1905. Der fibrillare Ban der Nervenendapparate in der Haut des Menschen und

der Saugetiere und die Neuronentheorie. Anat. Anz., Bd. 27, S. 97.

. 1906. Zur Frage uber den fibrilliiren Bau der Sehnenspindeln oder der Golgi’schen

Korperchen. Arch. f. mikr. Anat., Bd. 67, S. 638.

. 1908. Der Bau der Spinalganglien des Menschen und der Saugethiere.

G. Fischer, Jena.

. 1910. Zur Frage iiber den Bau der Kapseln der Vater-Pacinischen und HerbsG

schen Korperchen und iiber das Verhalten der Blutgefiisse zu denselben. Folia neuro-biol., Bd. 4, S. 218.

Dogiel, A. S., and Willanen, K. 1900. Die Beziehungen der Nerven zu den Grandryschen Korperchen. Zeitschr. f. wissensch. ZooL, Bd. 67, S. 349.

VON Doinikow, B. 1911. Beitrage zur Histologic und Histopathologie der peripheren Nerven. Nissl-Alzheimer, histologische und histopathologische Arbeiten iiber die Grosshimrinde, Jena. Bd. 4, S, 445.

Donaggio, a. 1901. Sur les appareils fibrillaires endocellulaires de conduction dans les centres nerveux des vert^br^s sup6rieurs. Congrfes international de Physiologie de Turin.

. 1903. Su speciali apparati fibrillari in dementi cellulari nervosi di alcuni centri

dell’acustico. (Ganglio ventrale, nucleo del corpo trapezoide.) Bibliog. anat., vol. 12, p. 89.

. 1904. II reticolo fibrillare endocellulare e il cilindrasse della cellula nervosa dei

vertebrati e metodi vari di colorazione elettiva del reticolo endocellulare e del reticolo periferico basati sull’azione della piridina sul tessuto nervoso. Riv. sper. di freniat., vol. 30, p. 397.

. 1904. Anatomia e fisiologia delle vie di conduzione endocellulari. XII Con gresso della societii freniatrica italiana, Genova.

. 1906. Effetti dell’azione combinata del digiuno e del freddo sui centri nervosi

di mammiferi adulti. Riv. sper. di freniat., vol. 32, p. 373.

DoybRE. 1840. Memoire sur les Tardigrades. Annales des sciences naturelles, vol. 14, p. 346 (Quoted from Hines ’27).

Driesch, Hans. 1894. Analytische Theorie der organischer Entwicklung. Leipzig.

Droogleever Fortuyn, A. B. 1917. Eenige beschouwingen over de evolutie van het zenuwstelsel. Handb. v. h. XVI Ned. Nat. en Geneesk. Congres, den Haag.

. 1920. Die Leitungsbahnen im Nervensystem der Wirbellosen. E. F. Bohn,

Haarlem.

Ducceschi, V. 1912. I. Ueber die Anwesenheit der Ruffinischen Korperchen in der Zunge der Vogel. II. Ueber die Funktion der Ruffinischen Korperchen. Folia neuro-biol., Bd. 6, S. 579.

Duesberg, J. 1912. Plastosomen, "apparato reticolare interno,” und Chromidialapparat. Zeitschr. f. d. ges. Anat., Abt. 3 (Ergeb. d. Anat. u. Entwicklungsgesch.), Bd. 20, p. 567.

. 1914. Trophospongien und Golgischer Binnenapparat. Verhandl. d. anat.

Gesellsch. (Erganzungsheft zum Bd. 46, Anat. Anz.), Bd. 28, S. 11.

■ . 1920. Cytoplasmic structures in the seminal epithelium of the opossum. Con trib. Embryol., Publ. Carnegie Inst. Washington, vol. 9 (28), p. 49.

Dusser de Barenne, J. G. 1916. Ueber die Innervation und den Tonus der quergestreiften Muskeln. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 166, S. 145.

Dustin, A. P. 1906. Contribution k I’dtude de I’influence de I’age et de I’activitie fonctionelle sur le neurone. Bruxelles.

106 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Dustin, A. P. 1917. Le R61e des tropismes et de I’odogen^se dans la Regeneration du systeme nerveux. Arch, de biol., vol. 25, p. 269. Quoted from Huber (’27).

. 1918. La fasciculation des nerfs, son importance dans le diagnostic, le prognostic

et le traitement des lesions nerveuses. Travaux de ambulance de I’ocean, Masson et Cie, Paris, vol. II.

Eberth, C. J., and Bunge, R. 1895. Die Nerven der Chromatophoren bei Fischen. Arch. f. mikr. Anat., Bd. 46, S. 370.

Eisath, G. 1906. Ueber normale und pathologischo Histologic der menschlichen Neuroglia. Monatschr. f. Psychiat. u. Neurol., Bd. 20, pp. 1, 139, and 140.

Elsberg, C. a. 1919. Technic of nerve suture and nerve grafting. J. A. M. A., vol. 73, p. 1422.

Elwyn and Strong. 1920. A textbook of histology by F. R. Bailey. 6th edition, Wm. Wood and Co., New York.

Engelmann, T. W. 1885. Ueber Bewegungen der Zapfen und Pigmentzellen' der Netzhaut unter dem Einfluss des Lichtes und des Nervensystems. Arch. f. d. ges. Physiol, (pfliiger’s), Bd. 35, S. 498.

Erlacher, P. 1915. Ueber die motorischen Nervenendigungen. Zeitschr. f. orthop. Chirurg., Bd. 34, S. 561.

Essick, C. R. 1920. Formation of macrophages by the cells lining the subarachnoid cavity in response to the stimulus of particulate matter. Contrib. to Embryol. no. 42, Publ. 272. Carnegie Inst., Washington, p. 377.

Eurich, F. W. 1898. Contributions to the comparative anatomy of the neuroglia. Anat., vol. 32, p. 688.

Fabritius, H. 1910. Versuch elner Psychophysiologie des Gefiihles. Monatschr. f. Psychiat. u. Neurol., Bd. 28, S. 400.

Faivre. 1853. Des granulationes meningiennes. Th6se de Paris.

DA Fano, G. 1906. Osservazioni sulla fine struttura della nevroglia. Ricerche n. lab. di anat. norm. d. r. Univ. di Roma, vol. 12, p. 1.

Farrar, C. B. 1907. The embryonic pia. Am. J. Insanity, vol. 63, p. 295.

Faure;-Fremiet, E. 1910. Etude sur les mitochondries, des protozoaires et des cellules sexuelles. Arch, d’anat. micr., vol. 11, p. 457.

Ferraro, A, 1925. Lo stato odierno delle nostre conoscenze sulla struttura e funzione dei plassi coroidei. I Cervello, vol. 4, p. 3.

Fieandt, H. 1910. Eine neue Methode zur Darstellung des Gliagewebes nebst Beitragen zur Kenntnis des Baues und der Anordnung der Neuroglia des Hundeshirns. Arch. f. mikr. Anat., Bd. 76, S. 125.

Findlay, J. W. 1899. The choroid plexuses of the lateral ventricles of the brain, their histology, normal and pathological (in relation especially to insanity). Brain, vol. 22, p. 161.

Flemming, W. 1882. Vom Bau der Spinalganglienzellen. Beitriige zur Anat. u. Embryol. als Festgabe f. J. Henle. S. 12.

. 1895. Ueber den Bau der Spinalganglienzellen bei Siiugethieren, und Bemer kungen fiber den centralen Zellen. Arch. f. mikr. Anat., Bd. 46, S. 379. '

Foley, F. E. B. 1923. Alterations in the currents and absorption of cerebrospinal fluid following salt administration. Arch. Surg., vol. 6, p. 587.

Forbes, H. S., Fremont-Smith, F., and Wolff, H. S. 192^ Resorption of the cerebrospinal fluid through the choroid plexus. Arch. Neurol, and Psychiat., vol. 19, p. 73.

Forssman, j. 1900. Zur Kenntniss des Neurotropismus. Beitr. z. path. Anat. u. z. allg. Path. (Ziegler’s Beitr.), Bd. 27, S. 407.

Forster, L. 1894. Zur Kenntniss der Muskelspindeln. Virchow’s Arch. f. path. Anat., Bd. 137, S. 121.

Frazier, C. H. 1915. The cerebrospinal fluid in health and disease. J. A. M. A., : vol. 64, p. 1119.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 107

Fremont-Ssiith, F. 1927. The nature of the cerebrospinal fluid. Arch. Neurol, and Psychiat., vol. 17, p. 317.

Fromxiann, C. 1864. Zur Silberfarbung der Axencylinder. Virchow’s Arch. f. path. Anat., Bd. 31, S. 151.

Fukutake, K. 1925. Beitriige zur Histologic und Entwicklungsgeschichte des Herznervensystems. Zeitschr. f. d. ges. Anat., Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 76, S. 592.

Galeotti, G. 1897. Studio morfologico e citologico della volta del diencefalo in alcuni vertebrati. Riv. di pat. nerv., vol. 2, p. 481.

Galeotti, G., and Levi, G. 1895. Ueber die Neubildung des nervosen Elemente in dem wiedererzengten Muskelgewebe. Beitr. z. path. Anat. u. z. allg. Path (Ziegler’s Beitr.), Bd. 17, no. 2, S. 369.

Garten. 1907. Ein Deutungsversuch der Bewegungsvorgange in der Netzhaut. Verhand. des VII Physiol. Congr. Heidelberg.

Garven, H. S. D. 1925. The nerve endings in the panniculus carnosus of the hedgehog, with special reference to the sympathetic innervation of striatedonuscle. Brain, vol. 48, p. 380.

Gassner, G. 1906. Der Galvanotropismus der Wurzeln. Botanische Zeitung. Bd. 64, S. 149.

Gatenby, J. B. 1920. On the relationship between the formation of yolk and mitochondria and Golgi apparatus during oogenesis. J. Roy. Micr. Soo., vol. 151, p. 129.

Gegenbaor, C. Lehrbuch der Anatomie des Menschen. 6*® Auflage, Bd. 2, S. 441. W. Engelmann, Leipzig.

. 1898. Vergleichende Anatomie der Wirbeltiere mit Beriicksichtigung der Wirbel losen. Teil II, W. Engelmann, Leipzig.

VAN Gehuchten, a. 1892. Contributions b, I’dtude de I’innervation des poils. Anat. Anz., Bd. 7, S. 341.

. 1905. I’Etat actuel de la doctrine des neurones. Nederl. tijdschr. v. geneesk.,

year 41, vol. 1, p. 1812.

. 1907. Anatomie du syst^me nerveux de I’homme. A. Uyspruyst-Dieudonne,

Louvain, 4‘^"'® edition.

VAN Gelderen, C. 1924. De ontwikkeling der Sinus durae matris bij den mensch. Nederl. tijdschr. v. geneesk., vol. 68, p. 2850.

. 1924a. Zur vergleichenden Anatomie der Sinus durae matris. Anat. Anz., Bd.

58, S. 472.

. 1924-26. Die Morphologic der Sinus durae matris. Mitteil. I-IV. Zeitschr.

f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 73, S. 541 ; Bd. 78, S. 339.

Gemelli, a. 1905. Sur la structure des plaques motrices chez les reptiles.

Le Nevraxe, vol. 7, p. 105.

VAN Genderen Stort, a. G. H. 1887. Ueber Form- und Ortsveranderungen der Netzhautelemente unter Einfluss von Licht und Dunkel. Arch. f. Ophth. (.von Graefe’s), Bd. 33, S. 229.

Giacomini, E. 1898. Sulla maniera onde i nervi si terminano nei tendini e nelle estremitd, delle fibre muscular! degli arti negli Anfibii urodeli. Monitore zool. ital., vol. 9, no. 5, p. 105.

. 1898a. Sulla maniera onde i nervi si termano nei miocommi e nella estremita

delle fibre muscular! dei miomeri negli Anfibii. Monitore zool. ital., vol. 9, no. 4, p. 92.

Goldmann E E. 1909, 1912. Die aussere und innere Sekretion des gesunden und kranken Organismus im Lichte der “vitalen Farbung.” Beitr. z. klin. Chir., Teil I., Bd. 64, S. 192; Teil II., Bd. 78, S. 1.

108 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Goldmann, E. E. 1913. Experimentelle Untersuchungen iiber die Funktion der Plexus chorioidea und der Hirnhaute. Arch. f. klin. Chir., Bd. 101, S. 735.

. 1913a. Vitalfarbung am Zentralnervensystem. Beitrag zur Physio-pathologie

des Plexus chorioideus und der Hirnhaute. G. Reimer, Berlin. Also in Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin. Phys.-math. Classe, S. 1.

Goldstein, K. 1904. Kritische und experimentelle Beitrage zur Frage nach dem Einfluss des Zentralnervensystems auf die embryonale Entwicklung imd die Regeneration. Arch. f. Entwicklungsmech., Bd. 18, S. 57.

. 1904a. Abhangigkeit der Muskulatur vom Zentralnervensystem wahrend der

Embryonalzeit. Arch. f. Entwicklungsmech., Bd. 18, S. 584.

Golgi, C. 1873. Quoted from Krause, 1926. Enzyklopadie der mikroskopischen Technik. Dritte Auflage. Bd. 2, S. 17. (First adequate statement of Golgi method.)

. 1880. Sui nervi nei tendini dell’uomo e di altri vertebrati e di un nuovo organo

nervoso terminale musoulo-tendineo. Mem. d. R. Accad. d. Sc. di Torino, Ser. IP, vol. 32, p. 359.

. 1882, 1883, 1885. Sulla fina anatomia degli organ! central! del sistema nervoso.

Riv. sper. di freniat., vol. 9, pp. 1, 161, 165, 210, 361, and 385 ; vol. 11, pp. 72 and 193.

. 1886. Sulla fina anatomia degli organ! central! del sistema nervoso. U. Hoepli,

Milano.

. 1898. Intorno allei struttura delli cellule nervose. Boll. d. Soc. med.-chir.,

Pavia. F. 1, 16 pp. See also Verhandl. d. anat. Gesellsch. (Erganzungsheft zum Bd. 17, Anat. Anz.), S. 205, 1901.

. 190A La doctrine du neurone. Conference Nobel faite Stockholm. Dec.

1906.

. 1907. La dottrina del neurone. Teoria e fatti. Arch, di fisiol., vol. 4, p. 187.

Goronowitsch, a. 1888. Das Gehirn und die Cranialnerven von Acipenser ruthenus. Morphol. Jahrb., Bd. 13, S. 515.

Gothlin, G. F. 1913. Die doppelbrechenden Eigenschaften des Nervengewebes, ihre Ursachen und ihre biologischen Konsequenzen. Kimgl. Svenska Vetenskapsakademiens Handlung., Bd., 51, no. 1.

. 1917. Relation entre le fonctionnement et la structure des elements nerveux.

Conference faite devant la faculte de medecine d’Upsal, 1917. Upsala Liikare forenings Forhandlingar, Bd. 13.

Gourewitch, M. 1906. Contribution a I’etude de la resistance du reseau fibrillaire des cellules nerveuses de la moelle epiniere des lapins adultes. Riv. sper. di freniat., vol. 32, p. 926.

Grabower, B. 1902. Ueber Nervenendungen im menschlichen Muskel. Arch. f. mikr. Anat., Bd. 60, S. 1.

Grandry, M. 1868. Recherches sur la structure du cylindre-a.xe et des cellules nerveuses. Bull, de I’Acad. de Bruxelles, 2'^™® Serie, vol. 28.

. 1869. Recherches sur les corpuscles de Pacini. J. de I’anat. et physiol., vol. 6,

p. 390.

. lS69a. De la structure intime du cylindre de l’a.\'e et des cellules nerveuses.

J. de I’anat. et physiol., vol. 6, p. 289.

Greeley, A. W. 1904. Experiments on the physical structure of the protoplasm of Paramoccium and its relation to the reactions of the organism to thermal, chemical, and electrical stimuli. Biol. Bull., Marine Biol. Lab., Woods Hole, Mass., vol. 7, p. 3.

Gurewitsch, M. j. 1908. Zur Morphologic der fibrillaren Apparates der Nervenzellen im normalen und pathologischen Zustande. Folia neuro-biol., Bd. 2, S. 197.

Gurwitscii, a. 1904. Morphologic und Biologic der Zelle. G. Fischer, Jena.

H,\.GGQUisT, G. 1914. Histophysiologische Studien fiber die Teraperatursinnc der Haut des Menschen. ^Vnat. Anz., Bd. 45, S. 46.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 109

Haggquist, G. 1914. Von Zellen nervoser Art in der Epidermis des Menschen. Anat. Anz., Bd. 47, S. 285.

. 1915. Studien iiber die Teraperatursinne der Haut des Menschen. Stockholm.

Halliburton, W. D. 1908. Biochemistry of nervous tissues. Folia neuro-biol., Bd. 1, S. 38.

. 1916. The possible functions of the cerebrospinal fluid. Proc. Roy. Soc. Med.,

vol. 10, p. 1.

Hamakeh, J. I. 1898. The nervous system of Nereis virens Sars. A study in comparative neurology. Bull. Mus. Comp. Zool., Harvard, vol. 32, no. 6, p. 89. Hamburger, H. J. 1892. Ueber den Einfluss der Athmung auf die Permeabilitat der Blutkorperchen. Zeitschr. f. Biol., Bd. 28 (N. F. 10), S. 405.

. 1902-04. Osmotischer Druck und lonenlehre in den medicinischen Wissen schaften. Zugleich Lehrbuch physikalisch-chemischer Methoden. J. F. Bergmann, Wiesbaden.

Hamburger, H. J., and van Lier, G. A. 1902. Die Durchlassigkeit der rothen Blutkorperchen fiir die Anionen von Natriumsalzen. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 492.

Hamilton, A. 1901. The division of differentiated cells in the central nervous system of the white rat. J. Comp. Neurol., vol. 11, p. 297.

Hansen, H. 1881. Arch, de biol., vol. 2, p. 342. (Quoted from Ramon y Cajal '09, p. 378.)

Hansen-Pruss, 0. C. 1923. Meninges of birds, with a consideration of the sinus rhomboidalis. J. Comp. Neurol., vol. 36, p. 193.

Hanstrom, B. 1929. Die vergleichende Anatomie des Nervensystems der Wirbellosen. Julius Springer, Berlin.

Hardy, W. B. 1899. On the coagulation of proteid by electricity. J. Physiol., vol. 24,

p. 288.

. 1928. Molecular orientation in living matter. J. Gen. Physiol., vol. 8, p. 642.

Harrison, R. G. 1901. Ueber die Histogenese des peripheren Nervensystems bei Salmo salar. Arch. f. mikr. Anat., Bd. 57, S. 354.

. 1904. An experimental study of the relation of the nervous system to the developing musculature in the embryo of the frog. Am. J. Anat., vol. 3, p. 197.

. 1906. Further experiments on the development of peripheral nerves. Am. J.

Amat., vol. 5, p. 121.

. 1907. Observations on the living developing nerve fiber. Anat. Rec., vol. 1,

no. 5, p. 116.

. 1907a. Experiments in transplanting limbs and their bearing upon the problems

of the development of nerves. J. Exper. Zool., vol. 4, p. 239.

. 1908. Embryonic transplantations and the development of the nervous system.

Anat. Rec., vol. 2, p. 385.

. 1910. The development of peripheral nerve fibers in altered surroundings.

Zeitschr. f. wissensch. Biol., Abt. D., Arch. f. Entwicklungsmechn. d. Organ., Bd. 30, Zweiter Teil, S. 15.

. 1910a. The outgrowth of the nerve fiber as a mode of protoplasmic movement.

J. Exper. Zool., vol. 9, p. 787.

. 1913. The life of tissues outside the organism from the embryological standpoint.

Tr. Cong. Am. Physicians and Surgeons, vol. 9, p. 63. Washington, May 6 and 7.

. 1924. Neuroblast versus sheath cell in the development of peripheral nerves.

J. Comp. Neurol., vol. 37, p. 123.

Harvey, E. N. 1922. Some recent experiments on the nature of the nervous impulse. J. Nerv. and Ment. Dis., vol. 55, p. 503.

Harvey, S. C., and Burr, H. S. 1926. The development of the meninges. Arch. Neurol, and Psychiat., vol. 15, p. 545.

110 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Hatai, S. 1901. On the presence of the centrosome in certain ncr\'e cells of the white rat. J. Comp. Neurol., vol. 11, p. 25.

. 1901a. On the mitosis in the nerve cells of the cerebellar corte.K of foetal cats.

J. Comp. Neurol., vol. 11, p. 277.

Havet, j. 1900. Contributions it l’6tudc du systeme nerveux dcs Trcmatodes. La Cellule, vol. 17, p. 353.

. 1922. La structure du systiimc nerveux dcs actinies, leur mticanisme neuro musculaire. Libro en honor dc D. S. Ramdn y Cajal, vol. 1, p. 477.

Hay, j. 1901. Red and white muscle. Liverpool Med.-Chir. J., Sept., p, 431.

Head, H., Rivers, W. H. R., and Siierren, J. 1905. The afferent nervous system from a new aspect. Brain, vol. 28, p. 99.

Head, H., and Sherren, J. 1905. The conscciuenccs of injury to the peripheral nerves in man. Brain, vol. 28, p. 116.

Heidenhain, M. 1911. Plasma und Zelle. T. 11. G. Fischer, Jena.

Held, H. 1895, 1897. Beitriige zur Structur der Nervenzellen und ihrer Fortsatze. Arch. f. Anat. u. Physiol., Anat. Abt., Abt. I in Bd. for 1895, S. 396 ; Abt. II in Bd. for 1897, S. 204.

. 1902. Ueber den Bau der grauen und weissen Substanz. Arch. f. Anat. u.

Physiol., Anat. Abt., p. 189.

. 1904. Ueber den Bau der Neuroglia und iiber die Wand der Lymphgefiissc in

Haut und Schleimhaut. Abhandl. d. math.-phys. Klasse d. kon. siichs. Gesellsch. d. Wissensch., Bd. 28, no. 4, S. 199.

. 1904a. Untersuchungen iiber den feineren Bau des Gehororgans der Wirbeltiere.

I. Zur Kenntnis des Cortischen Organs und der (ibrigen Sinnesapparate des Labyrinthes bei Saugetieren. Abhandl. d. math.-phys. Klasse d. kon. sachs. Gesellsch. d. Wissensch., Bd. 28, no. 1, S. 1.

. 1905. Zur weiteren Kenntnis der Nervenendfiisse und zur Struktur der

Sehzellen. Abhandl. d. math.-phys. Klasse d. kon. siichs. Gesellsch. d. Wissensch., Bd. 29, no. 2, S. 143.

. 1906. Zur Histogenese der Nervenleitung. Verhandl. d. anat. Gesellsch.

(Erganzungsheft zum Bd. 29, Anat. Anz.), Bd. 20, S. 185.

. 1907. Kritische Bemerkungen zu der Verteidigung der Neuroblasten- und

Neuronentheorie durch R. Cajal. Anat. Anz., Bd. 30, S. 369.

. 1909. Ueber die Neuroglia marginalis der menschlichen Grosshirnrinde.

Monatschr. f. Psychiat. u. Neurol., Bd. 26, S. 360.

. 1909a. Die Entwicklung des Nervengewebes bei den Wirbeltieren. Barth,

Leipzig.

. 1929. Die Lehre von den Neuronen und vom Neurocytium und ihre heutiger

Stand. Fortschritte d. Naturwissensch. Forschung.

Hensen, V. 1864. Ueber die Entwickelung des Gewebes und der Nerven im Schwanze der Froschlarve. Virchow’s Arch. f. path. Anat., Bd. 31, S. 51.

. 1868. Ueber die Nerven im Schwanz der Froschlarven. Arch. f. mikr. Anat.,

Bd. 4, S. 111.

. 1903. Die Entwicklungsmechanik der Nervenbahnen im Embryo der Saugetiere.

Ein Probeversuch. Lipsius u. Tischer, Kiel u. Leipzig.

Hbrbst. 1848. Gottinger Nachrichten. Reference taken from Ram6n y Cajal, ’ll Hering, H. E. 1870. Ueber das Gedachtnis als eine allgemeine Function der organisierten Materie. Wien.

Heringa, G. C. 1917. The intraprotoplasmic position of the neurofibrils in the axon and in the endorgan. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., Deel. 25. (Quoted from Aliens Kappers ’20; see also ’17b).

. 1917a. Le d^veloppement des corpuscles de Grandry et de Herbst. Arch.

n^erl. d. sc. exactes et nat.. Series III B., vol. 3, p. 235.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 111

Heringa, G. C. 1917b. De intraprotoplasmatische ligging der neurofibrillen in axon en endorgaan. Handl. 16 ned. nat. u. gen. congres den Haag, p. 343.

. 1920. Untersuchungen iiber den Ban und die Entwicklung des peripheren,

sensiblen peripheren Nervensystem. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweedie Sectie, vol. 21, p. 1.

. 1931. Das Boekische periterminale Netzwerk der quergestreiften Muskelfasern

und diesen Verbindung mit der kontraktilen Substanz. Zeitschr. f. wissensch. Biol. ; Abt. B., Zeitschr. f. Zellforsch. u. mikr. Anat., Bd. 23, S. 505.

Hermann, L. 1897. Eine physikalische Erscheinung am Nerven. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 67, S. 240.

Herrick, C. J. 1924. The amphibian forebrain. II. The olfactory bulb of Amblystoma. J. Comp. Neurol., vol. 37, p. 373.

. 1924a. Neurological foundations of behavior. Henry Holt and Co., New

York.

. 1931. An introduction to neurology. W. B. Saunders Co., Philadelphia and

London. 5th edition.

Hertwig, 0. AND R. 1879-80. Die Actinien anatomisch und histologisch mit besonderer Beriicksichtigung des Nervenmuskelsystems untersucht. Jenaische Ztschr. f. Naturw., Bd. 13, S. 519; Bd. 14, S. 39.

Hertwig, 0. 1918. Das Werden der Organismen. Zur Wiederlegung von Darwin’s

Zufallstheorie durch das Gesetz in der Entwicklung. 2 ed., G. Fischer, Jena.

VAN Herwerden, M. A. 1913. Ueber die Nucleasewirkung auf tierische Zellen. Ein Beitrag zur Chromidienfrage. Arch. f. Zellforsch., Bd. 10, S. 431.

. 1913a. Ueber die chemische Zusammensetzung der Nissl’schen Korner der

Ganglienzellen. Berliner klin. Wochenschr., Bd. 50, S. 1820.

Hesse, R. 1904. Ueber den feineren Bau der Stabchen und Zapfen einiger Wirbelthiere. Zool. Jahrb., Supplement-Band 7, S. 471.

Heymans, R. 1889. Beitrag zur Kenntnis des Epithels und der Driisen des menschlichen Kehikopfes im gesunden und im kranken Zustande. Virchow’s Arch. f. path. Anat. u. Physiol., Bd. 118, S. 320.

Hill, L. 1896. The physiology and pathology of the cerebral circulation. An experimental research. J. and A. Churchill, London.

. 1900. On cerebral anaemia and the effects which follow ligation of the cerebral

arteries’ tonus. Phil. Tr. Roy. Soc., London, Series B., vol. 193, p. 69.

Hines, M. 1927. Nerve and muscle. Quart. Rev. Biol., vol. 2, no. 2, p. 149.

. 1930. Studies in the innervation of skeletal muscle. HI. The extra-ocular

muscles in the rabbit. Anat. Rec., vol. 45, p. 220.

. 1930a. The innervation of the muscle spindle. Proc. A. Research in Nerv. and

Ment. Dis., vol. 9, p. 124.

. 1933. Studies in the innervation of skeletal muscle. IV. Of certain muscles of

the boa constrictor. J. Comp. Neurol., vol. 56, p. 105.

Hines, M., and Tower, S. 1928, Studies on the innervation of skeletal muscle. II. Of muscle spindles in certain muscles of the kitten. Bull. Johns Hopkins Hospital, vol. 42, p. 264.

HmsEY, J. C. 1930. Three types of fibers going to skeletal muscle and its blood vessels. Proc. A. Research Nerv. and Ment. Dis., vol. 9, p. 153.

Hinshaw, H. C. 1926. On the morphology and mitosis of Trichomonas buccalis (Goodey) Kofoid. Univ. California Publ., Zool., Univ. California Press, vol. 29, no. 7, p. 159.

His, W. 1865. Ueber ein perivasculares Canalsystem in den nervosen Centralorganen. W. Engelmann, Leipzig.

. 1887. Die Entwickelung der ersten Nervenbahnen beim menschlichen Embryo.

Uebersichtliche Darstellung. Arch. f. Anat. u. Physiol., Anat. Abt., S. 368.

112 NERVOUS SYSTEMS OF VERTEBRATES AND OF IvL\N

His, W. 1889. Die Neuroblasten und deren Entstehung im embrj'onalen Mark. S. Hirzel, Leipzig.

. 1890. Histogenese und Zusammenhang der Nervenelemente. Arch. f. Anat.

u. Physiol., Anat. Abt., Supplement-Band, 1890, S. 95.

. 1891. Jun. Die Entwickelung des Herznervensystems bei den Wirbelthieren.

Abbandl. d. math.-phys. Klasse d. kon. sachs. Gesellsch. d. Wissensch., Bd. 18, no. 1, S. 1.

. 1903. Die Haute und Hohlen des Korpers. Wiederabdruck eines akademischen

Programmes vom Jahre 1865. Arch. f. Anat. u. Physiol., Anat. Abt., S. 368.

. 1904. Die Entwickelung des menschlichen Gehirns wahrend der ersten Monate.

Untersuchungsergebnisse. S. Hirzel, Leipzig.

Ho.\dley, L. 1925. The differentiation of the isolated chick primordia in chorioallantoic grafts. III. On the specificity of nerve processes arising from the mesencephalon in grafts. J. E.xper. Zool., vol. 42, p. 163.

Hober, R. 1902. Physikalische Chemie der Zelle und der Gewebe. W. Engelmann, Leipzig, 5*^ Aufl.

Hocke Hoogenboom, K. J. 1928. De hersenen von Polyodon folium, Lacep. Dissertation, Amsterdam. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 311, 1929.

Holmgren, E. 1899. Zur Kenntnis der Spinalganglienzellen von Lophius piscatorius. i\nat. Hefte, Bd. 12, S. 71.

. lS99a. Weitere Mitteilungen iiber den Bau der Nervenzellen. Anat. Anz.,

Bd. 16, S. 388.

. 1900. Studien in der feineren Anatomic der Nervenzellen. Anat. Hefte, Bd.

15, S. 1.

. 1900a. Noch weitere Mitteilungen fiber den Bau der Nervenzellen verschicdener

Tiere. .•\jiat. Anz., Bd. 17, S. 113.

, 1904. Ucber die Trophospongien centraler Nervenzellen. Arch. f. Anat. u.

Physiol., .Anat. Abt., S. 15.

. 1904a. Ueber die Trophospongien der Nervenzellen. Anat. Anz., Bd. 24,

S. 225.

. 1904b. Beitrage zur Morphologic der Zelle. 11. Verschiedene Zellarten.

iAnat. Hefte, Bd. 25, S. 97.

. 1912. Weitere Untersuchungen fiber die morphologisch nachw-eisbaren stofilichcn

Veriinderungen der Muskelfusem. Kungl. Svenska Vetenskaps. Akademiens Handlingar, Bd. 49, no. 2.

. 1914. Trophospongium und Apparato reticolare der spinalen Ganglienzellen.

Anat. Anz., Bd. 46, S. 127.

. 1915. Die Trophospongien spinaler Ganglienzellen. Arkiv for Zoologi, Bd. 9,

no. 15, S. I.

Hol.\igre.v, N., and van der Horst, C. J. 1925. Contribution to the morphology of the brain of Ceratodu.s. .Acta Zool., Bd. 6, S. 59.

Hooker, D. 1911. The development und function of voluntary and cardiac muscle in embryos without ncrvc.s. J. E.xper. Zool., vol. 11, p. 159.

. 1915. Studie.s on regeneration in the spinal cord. I. An analy.sis of the procc.-s

leading to it.s reunion after it ha.s D-cn completely severed in frog embiyos at the stage <)f clo.'cd neund fuld.s. .1. Comj). Neurol., vol. 25, p. 409.

. 1910. Some resull.s from rever.-ing a portion of the spinal cord end for end in

frog embiyos. Anat. Bee., vol. 10, p. 198 (Proe. .Am. A. Anat.), .'jurnmary only.

. 1917. .Studies on ^generation in the .spinal cord. 11. The effect of rcver.-al

of a fsirtion of the spin.d cord at the .■•tage t)f clo.M-d neural fold.H on the healing of tlie cord wound.-’, on tiie polarity of t he element.s of the cord and on the behavior of frog enibry(>.'. J. Comp. Neurol., vol. 27, p. 421.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 113

Hooker, D. 1923. The nature of the divisions of the neuroblastic cells in the regenerating spinal cords of amphibian larvae. Anat. Rec., vol. 24, p. 377.

. 1925. Studies on regeneration in the spinal cord. III. Reestablishment of

anatomical and physiological continuity after transection in frog tadpoles. J. Comp. Neurol., vol. 38, p. 315.

VAN DER Horst, C. J. 1925. The myelencephalic gland of Polyodon, Acipenser, and Amia. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 28, p. 43^

Howard, A. D. 1908. The visual cells in vertebrates, chiefly in Necturus maculosus. J. Morphol., vol. 19, p. 561.

Howell, W. H. 1906. Vagus inhibition of the heart in its relation to the inorganic salts of the blood. Am. J. Physiol., vol. 15, p. 280.

Howell, W. H., and Huber, G. C. 1892. A physiological, histological, and clinical study of the degeneration and regeneration in the peripheral nerve fibers after severance of their connections with the nerve centers. J. Physiol., vol. 13, p. 335.

Huber, G. Carl. 1892. Ueber das Verhalten der Kerne der Schwann’schen Scheide bei Nervendegenerationen. Arch. f. mikr. Anat., Bd. 40, S. 409.

. 1895. A study of the operative treatment for loss of nerve substance in peripheral nerves. J. Morphol., vol. 11, p. 629.

. 1896. Observations on the innervation of the sublingual and submaxillary glands.

J. Exper. Med., vol. 1, p. 281.

. 1899. A note on sensory nerve endings in the extrinsic eye muscles of the rabbit.

Atypical motor endings of Retzius. Anat. Anz., Bd. 15, S. 335.

. 1899a. A contribution on the minute anatomy of the sympathetic ganglia of

the different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1900. Observations on sensory nerve fibers in visceral nerves, and on their modes

of terminating. J. Comp. Neurol., vol. 10, p. 135.

. 1900a. Sensory nerve terminations in the tendons of the extrinsic eye-muscles

of the cat. J. Comp. Neurol., vol. 10, p. 152.

. 1900b. Observations on the degeneration and regeneration of motor and sensory

nerve endings in voluntary muscles. Am. J. Physiol., vol. 3, no. 7, p. 339.

. 1901. Studies on the neuroglia. Am. J. Anat., vol. 1, p. 45.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ.

of Michigan. Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

. 1913. The morphology of the sympathetic system. Report 17th Internet.

Cong, of Med., London. See also Folia neuro-biol., vol. 7, p. 616.

. 1916-17. Operative treatment of peripheral nerves after severance, more particularly after loss of substance ; A critical review. J. Lab. and Clin. Med., vol. 2, p. 837. (Prepared for the Committee on Medicine and Hygiene, of the National Research Council.)

. 1919. Textbook of histology. J. B. Saunders and Co., New York. (Printed

in 1904 ; reprinted several times including 1919.)

. 1919. Transplantation of peripheral nerves. Arch. Neurol, and Psychiat.,

vol. 2, p. 266.

. 1920. Repair of peripheral nerve injuries. Surg., Gynec. and Obst., vol. 30,

no. 5, p. 464.

. 1927. Experimental observations on peripheral nerve repair. Excerpt from

the Medical Department of the United States Army in the World War. United States Government Printing Office, Washington.

. 1927a. A study of the operative treatment for loss of nerve substance in peripheral nerves. J. Morphol., vol. 11, no. 3, p. 629.

Huber, G. Carl, and Crosby, E. C. 1926. On thalamic and tectal nuclei and fiber paths in reptiles. J. Comp. Neurol., vol. 40, p. 97.

114 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Huber, G. Carl, and Crosby, E. C. 1929. The nuclei and fiber paths of the avian diencephalon, with consideration of telencephalic and certain mesencephalic centers and connections. J. Comp. Neurol., vol. 48, p. 1. (See especially figs. 13 and 15, darkened area along ventricle.)

Huber, G. Carl, and Lewis, D. 1920. Amputation neuromas, their development and prevention. Arch. Surg., vol. 1, no. 1, p. 85.

Huber, G. Carl, and DeWitt, L. 1898. A contribution on the motor-nerve-endings and on the nerve-endings in the muscle spindles. J. Comp. Neurol., vol. 7, p. 169.

. 1900. A contribution on the nerve terminations in neuro-tendinous end-organs.

J. Comp. Neurol., vol. 10, p. 159.

Hunter, J. I. 1925. The sympathetic innervation of striated muscles. Brit. Med. J., vol. 1, pp. 197, 251, and 350.

Hworostuchin, W. 1911. Zur Frage fiber den Bau des Plexus chorioideus. Arch. f. mikr. Anat., Bd. 77, S. 232.

Hyman, 0. W. 1923. Spermic dimorphism in Fasciolana tulipa. J. Morphol., vol. 37, p. 307.

Imamura, S. 1902. Beitriige zur Histologic des Plexus chorioideus des Menschen. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner's), Bd. 8, S. 272.

Ingebrigtsen, R. 1916. Sur la transplantation des nerfs. Lyon chirurgical, vol. 8, no. 5, p. 828.

. 1916a. A contribution to the biology of peripheral nerves in transplantation.

II. Life of peripheral nerves of mammals in plasma. J. E.xper. Med., vol. 23, p. 251.

. 1918. Am Nerventransplantation. Christiania.

Ingvar, S. 1919-20. Reaction of cells to the galvanic current in tissue cultures. Proc. Soc. Exper. Biol, and Med., vol. 17, p. 198.

IsHiKAWA, K. 1932. On the genesis of Hortega cells. Folia Anat. Japonica, vol. 10, p. 229.

IwANAGA, 1. 1926. Studien fiber die motorischen Nervenendigungen : I, 11, III.

Mitt. fi. allg. u. path. Anat., Unv., Sendai, Japan, vol. 2, S. 257.

IzQUiERDO, V. 1879. Beitrage zur Kenntniss der Endigung der sensiblen Nerven. Inaugural Dissertation, Strassburg.

Jaburek, L. 1930. Ueber die Struktur der Nervenfasern. Arb. a. d. neurol. Inst, a. d. Wien. Univ. (Obersteiner’s), Bd. 32, p. 1.

Jacobsohn, L. 1908. Ueber die Kerne des menschlichen Rfickenmarks. Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin, p. 389.

Jaderhol.m, G. a. 1906. Endozellulare Netze oder durchlaufende Fibrillen in den Ganglienzellen? Arch. f. mikr. Anat,, Bd. 67, S. 103.

Jenkins, O. P., and Carlson, A. 1903. The rate of the nervous impulse in the ventral nerve-cord of certain worms. J. Comp. Neurol., vol. 13, p. 259.

. 1904. Physiological evidence of the fluidity of the conducting substance in the

pedal nerves of the slug — Ariolimax columbianus. J. Comp. Neurol., vol. 15, p. 85.

Johnston, J. B. 1908. Additional notes on the cranial nerves of petromyzonts. J. Comp. Neurol., vol. 18, p. 569.

. 1908. On the .significance of the caliber of the parts of the neuron in vertebrates.

J. Comp. Neurol., vol. 18, p. 601.

. 1909. The nerv'ous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

Jones, 0. W. 1932. Cytogenesis of oligodendroglia and astrocytes. Arch. Neurol, and Psychiat., vol. 28, p. 1030.

Joins, H. 1909. Le.s voies conductrices ncurofibrillaires. J. do Neurol., vol. 14.

Kadvi, H. 1889. Ueber die Blutgeftisse des menschlichen Ruckenmarkes. Gubrynowicz and Schmidt, Lemberg.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELE^'IENTS 115

Kallius, E. 1S94. Untersuchungen uber die Netzhaut der Saugetiere. Anat. Hefte, Bd. 3, S. 527.

Kat6, H. 1909. Zur Netzstruktur der Neurofibrillen. Folia neuro-biol., Bd. 3, S. 21.

K.vTS.vN’UiiA, S. 1915. Zur Frage der Naphtolblauo.xydasereaction des Nervensystems. Beitr. z. path. Anat., u. z. allg. Path. (Ziegler’s), Bd. 60, S. 150.

Kennedy, R. 1S97. On the regeneration of nerves. Phil. Tr. Roy. Soc., London, vol. 9, p. 472.

Kehschner, L. 18S8. Bemerkungen iiber eln besonderes Muskelsystem im willkurlichen Muskel. Anat. Anz., Bd. 3, S. 126.

. 1893. Ucber die Fortschritte in der Erkenntnis der Muskelspindeln. Anat.

Anz., Bd. S, S. 449.

Key, E. a. H., and Retzius, G. 1875-76. Studien in der Anatomie des Nervensystems und des Bindegewebes. Samson and Wallin, Stockholm.

Kingsbury, B. F. 1912. Cytoplasmic fixation. Anat. Rec., vol. 6, p. 39.

Kirk, E. G., AND Lewis, D. D. 1917. Regeneration in peripheral nerves : an experimental study. Bull. Johns Hopkins Hospital, vol. 28, no. 312, p. 117.

Kleinenberg, N. 1872. Hydra. Einc anatomisch-entwicklungsgeschichtliche Untersuchung. W. Engelmann, Leipzig.

IvLESTADT, B. 1915. Experimcntelle Untersuchungen iiber die resorptive Funktion des Epithels der Plexus chorioideus und des Ependyms der Seitenventrikel. Centralbl. f. allg. Path. u. path. Anat., Bd. 26, S. 161.

Kofoid, C. a., and Swezy, 0. 1919. Studies on the parasites of termites I-IV. Univ. California Pubi., Zool., vol. 20, nos. 1-4, p. 1.

VON Kolliker, a. 1862. Ueber die letze Endigung in den Muskeln des Frosches. Wurzburger naturwisscnsch. Zeitschr., Bd. 3, H. 1, S. 1.

. 1879. Entwicklungsgeschichte des Menschen und der hoheren Thiere. 2nd

edition. W. Engelmann, Leipzig.

. 1880. Grundriss der Entwicklungsgeschichte des Menschen und der hoheren

Thiere. W. Engelmann, Leipzig.

. 1889-02. Handbuch dcr Gewebelehre des Menschen. W. Engelmann, Leipzig.

. 1904. Ueber die Entwickelung der Nervenfasern. Verhandl. d. anat. Gesellsch.,

Bd. 18, S. 7.

Kolmer, W. 1904. Ueber ein Strukturelement der Stiibchen und Zapfen der Froschretina. Anat. Anz., Bd. 25, S. 102.

. 1904a. Ueber Kristalle in Ganglienzellen. Anat. Anz., Bd. 25, S. 618.

. 1905. Zur Kenntnis des Verhaltens der Neurofibrillen an der Peripherie. Anat.

Anz., Bd. 27, S. 416.

. 1914. Zur Histologic der Augenhiiute. Anat. Anz., Bd. 47, S. 417.

. 1921. Ueber eine eigenartige Beziehung von Wanderzellen zu den Chorioideal plexus des Gehims der Wirbeltiere. Anat. Anz., Bd. 54, S. 15.

Kolster, R. 1900. Ueber das Vorkommen von Centralkorpem in den Nervenzellen von Cottus scorpius. Anat. Anz., Bd. 17, S. 172.

. 1901. Ueber Centralgebilde in Vorderhomzellen der Wirbeltiere. Anat. Hefte,

Bd. 16, S. 151.

Kooy, F. H. 1918. Concerning the extension of sarcoma piae matris. Psychiat. en neurol. Bl., Amsterdam, vol. 22, p. 447.

Korolkow, P. 1892. Die Nervenendigung in den Speicheldrusen. Anat. Anz., Bd.

7, S. 109.

Kraus, F., and Blumenthal, F. 1923. • Ueber die Reizempfindlichkeit des Centrosoms und ihre Bedeutung fiir die Organoplastik und fur die anorganoplastischen Gewebswucherung. Zeitschr. f. Krebsforsch., Bd. 20, S. 211.

Krause, W. 1882. Die Nervenendigungen in den Tatskorperchen. Arch. f. mikr. Anat., Bd. 20, S. 212.

116 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Krebs. P. 1905. Die Nervenendigungen im Musculus stapedius mit besonderer Beriicksichtigung der bei der Farbung angewandten Technik. Arch. f. mikr, Anat., Bd. 65, S. 704.

Kbeibich, C. 1916. Zur Anatomie des Tigroids. Anat. Anz., Bd. 49. S. 56.

Kkoemer. 1878. Die Knochenneubildungen in der Arachnoidea. Allg. Zeitschr. f. Psychiat., Bd. 34, S. 602.

Kuhne, W. 1862. Ueber die peripherischen Endorgane der motorischen Nen^en. W. Engelmann, Leipzig. (Quoted from Hines ’27.)

. 1886. Neue Untersuchungen iiber motorische Nervenendigungen. Zeitschr. f.

Biol., Bd. 23, S. 1.

Kulchitsky, N, 1924. Nerve endings in muscles. J. Anat., vol. 58, p. 152.

. 1924-5. Nerve endings in the muscles of the frog. J. Anat., vol. 59, p. 1.

Kuntz, a. 1922. Experimental studies on the histogenesis of the sympathetic nervous system. J. Comp. Neurol., vol. 34, p. 1.

. 1927. On the occurrence of sympathetic nerve fibers in muscles of the extremities following experimental degeneration of the spinal nerves. J. Comp. Neurol., vol. 43, p. 357.

. 1929. The autonomic nervous system. Lea and Febiger, Philadelphia.

Kuntz, A., and Kerper, A. H. 1924. The sympathetic innervation of voluntary muscle. Proc. Soc. Exper. Biol, and Med., vol. 22, p. 23.

Landacre, F. L. 1921. The fate of the neural crest in the head of the urodeles. J. Comp. Neurol., vol. 33, p. 1.

Langerhans, P. 1868. Ueber die Nerven der menschlichen Haut. Virchow's Arch, f. path. Anat., Bd. 44, S. 325.

Langley, J. N. 1891. On the course and connections of the secretory fibers supplying the sweat glands of the feet of the cat. J. Physiol, vol. 12, p. 347.

. 1904. On the question of commissural fibers between nerve-cells having the same

function and situated in the same sympathetic ganglion, and on the function of postganglionic nerve plexuses. J. Physiol., vol. 31, p. 244.

. 1905. The autonomic nerves. Lecture read in the scientific sections of the

Amsterdam University.

Langley, J. N., and Anderson, H. K. 1894. On reflex action from sympathetic ganglia. J. Physiol., vol. 16, p. 410. For 1892, see Chap. II.

. 1895. The innervation of the pelvic and adjoining viscera. Part II. The

bladder. Part III. The external generative organs. Part IV. The internal generative organs. Part V. Position of the nerve cells on the course of the efferent nerve fibers. J. Physiol., vol. 19, p. 71.

Langley, J. N., and Magnus, R. 1905. Some observations of the movements of the intestine before and after degenerative section of the mesenteric nerves. J. Physiol., vol. 33, p. 34.

Larsell, 0. 1923. Some aspects of the innervation of the lung. Northwest Med.,

vol. 22, no. 9, p. 311.

Lawrentjew, B. j. 1926. Ueber das Chondriom in den Grandryschen Korperchen. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 6,

S. 41.

. 1926a. Ueber die Verbreitung der nervosen Elemente (einschliesslich der "inter stitiellen Zcllen” Cajal’s) in der glatten Muskulatur, ihre Endigungsweise in den glatten Muskelzellen. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.anat. Forschung, Bd. 6, S. 467.

. 1928. Studien fiber den feineren Bau dor Nervenendigungen in der quer ge.streiften .Mu.skulaturdcsFrosche.s. 11. Ueber die "plun.'^egmentelle Innervation" der Mu.skclfa.“cm. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.anat. For.-^chiing, Bd. 14, S. 511,

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 117

Laycock, T. 1875. A chapter oh some organic laws of personal and ancestral memory. J. Ment. Sc., vol. 21, p. 155.

Leahy, T., and Edwards, E. A. 1933. The subdural space and its linings. Arch. Neurol, and Psychiat., vol. 29, p. 691.

Lefebure. 1909. Sur les corpuscules de Wagner-Meissner on corpuscules detact. Rev. gen. d’histol.

Legendre, R. 1906. Sur divers aspects de neurofibrilles intracellulaires obtenus par la m6thode de Bielschowsky. Anat. Anz., Bd. 29, S. 361.

VON Lenhossek, M. 1891. Zur Kenntnis der Neuroglia des menschlichen Riickenmarkes. Verhandl. d. anat. Gesellsch., Bd. 5, S. 193.

. 1895. Centrosom und Sphare in den Spinalganglienzellen des Frosches. Arch.

f. mikr. Anat., Bd. 46, S. 345.

. 1895. Der feinere Bau des Nervensystems im Lichte neuester Forschungen.

2“= Aufl., Fischer, Berlin.

. 1897. Ueber den Bau der Spinalganglienzellen des Menschen. Arch. f. Psychiat., Bd. 29, S. 345.

. 1898. Bemerkungen iiber den Bau der Spinalganglienzellen. Neurol. Centralbl.,

Bd. 17, S. 577.

. 1898a. Nervensystem. Ergeb. d. Anat. u. Entwicklungsgesch. (Merkel u.

Bonnet), Bd. 7, S. 110.

. 1910. Ueber die physiologische Bedeutung der Neurofibrillen. Anat. Anz.,

Bd. 36, S. 257 und 321.

Leontowitsch, a. W. 1901. Die Innervation der menschlichen Haut. Internat. Monatschr. f. Anat. u. Physiol., Bd. 18, S. 142.

. 1929. Zur Frage iiber die Nervenendkorperchen im Herzen des Frosches.

Zeitschr. f. wissensch. Biol. ; Abt. B., Zeitschr. f. Zellforsch. u. mikr. Anat., Bd. 9, S. 277.

Levi, G. 1896. Contributo alia fisiologia della cellula nervosa. Riv. di pat. nerv., vol. 1, p. 169.

. 1906. Studien zur normalen und pathologischen Anatomic der hinteren Riicken marks. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 13, S. 62.

. 1908. I gangli cerebrospinali. Studi di Istologia comparata e di Istogenesi.

Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., 1912, Bd. 6, S. 188.

. 1916. Sull’origine delle reti nervose nelle colture de tessuti. Atti d. r. Accad.

d. Lincei, Rendic. cl. sc., Serie V, vol. 25*, p. 663.

. 1917. Connessioni e struttura degli elementi nervosi sviluppati fuori dell’ organismo. Atti. d. r. Accad. d. Lincei, Serie V, vol. 12, fasc. 4, p. 142.

Lewandowsky, M. 1900. Zur Lehre von der Cerebrospinalfliissigkeit. Zeitschr. f. klin. Med., Bd. 40, S. 480.

Lewis, M. 1896. Centrosome and sphere in certain of the nerve cells of an invertebrate. Anat. Anz., Bd. 12, S. 241.

. 1898. Studies on the central and peripheral nervous systems of two polychaete

annelids. Proc. Am. Acad. Arts and Sc., vol. 33, no. 14, p. 222.

Lewis, W. H., and Lewis, M. R. 1912. The cultivation of sympathetic nerves from the intestine of chick embryos in saline solutions. Anat. Rec., vol. 6, p. 7.

. 1924. Behavior of cells in tissue cultures. Sect. 7, General Cytology (edited

by E. V. Cowdry), Univ. of Chicago Press.

Lewy, F. H. 1914. Beitrag zur Kenntnis der Lymphwege des Gehims. (Der Transport in der Lymphe loslicher Substanzen.) Arch. f. Anat. u. Physiol., Anat. Abt., S. 143.

. 1921. Die Lymphraume des Gehirns, ihr Bau und ihre Geschwulste. Virchow’s

Arch. f. path. Anat., Bd. 232, S. 400.

118 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Leydig, F. 1897. Arch. f. Anat. u. Physiol., Anat.’Abt. (Quoted from Schafer, ’12).

Lillie, R. S. 1911. The relation of stimulation and conduction in irritable tissues to changes in the permeability of the limiting membranes. Am. J. Physiol., vol. 28, p. 197.

. 1917. The formation of structures resembling organic growths by means of

electrolytic local action in metals and the physiological significance and control of this type of action, Biol. Bull., Marine Biol, Lab,, Woods Hole, Mass., vol. 33, p. 135.

. 1918. Transmission of activation in passive metals as a model of the protoplasmic or nerv’ous type of transmission. Science, N. S., vol. 48, p. 51.

. 1919. Nervous and other forms of protoplasmic transmission. Scient. Monthly,

vol. 8, pp. 456 and 552.

. 1923. Protoplasmic action and nervous action. Univ. of Chicago Press,

. 1924. Reactivity of the cell. General Cytology (edited by E. V. Cowdry),

chapter V. Univ. of Chicago Press.

Loeb, J. 1896. Zur Theorie des Galvanotropismus. III. Ueber die polare Erregung der Hautdrussen von Amblystoma durch den constanten Strom. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 65, S. 308.

. 1906. Vorlesungen fiber die Dynamik der Lebenserscheinungen. J. A. Barth,

Leipzig.

Loeb, J., and Budgett, S. P. 1897. Zur Theorie des Galvanotropismus. IV. Mittheilung. Ueber die Ausscheidung electropositiver lonen an der iiusseren /Ijiodenflache protoplasmatischer Gebilde als Ursache der Abweichungen vom Pflugcr’schen Erregungsgesetz. Arch. f. d. ges. Physiol. (Pflfiger’s), Bd. 65, S. 518.

Loeb, J., and Geuhy, W. E. 1896. Zur Theorie des Galvanotropismus. IL Mittheilung. Vcrsuche an Wirbelthieren. Arch. f. d. ges. Physiol. (Pfifiger’s), Bd, 65, S. 41.

Loeb, J., and Maxwell, S. S. 1896. Zur Theorie des Galvanotropismus. Arch, f. d. ges. Physiol. (Pflfiger’s), Bd. 63, S. 121.

Loepeb, M. 1904. Sur quelques points de I’histologie normale et pathologiquc des pie.\'us chorioides de I’liomme. Arch, de m6d. exp6r. et d'anat. path., vol. 16, p. 473.

London, E. J. 1905. Zur Lehre von dem feineren Bau des Nervensystcms. Arch. f. mikr. Anat., Bd. GO, S. 111.

London, E. S., and Pesker, D. J. 1906. Ueber die Entwicklung des periphcren Nervensystcms bei Siiugetieren (weissen Mtiusen). Arch. f. mikr. Anat., Bd. 67, S. 303.

Lugauo, L. 1897. Riv. di pat. ncrv., vol. 2, (Quoted from ILimdn y Cajal, ’ll.)

. 1901. Allgemeine pathologischc Anatomic der Neuroglia. In Ilandbuch der

pathologischen Anatomic des Nervensystcms, by Flatau, E.. Jacob.sohn, L., and Minor. L. S. Karger, Berlin.

. 1906. Sul neurotropi.smo e sui trapianti dei nervi. Riv. di p:it. nerv., vol. 11,

p. 320.

- . 1909. Una prova dell'esistcnza delle neurofibrillc nel vivcntc. Riv. di pat.

nerv., vol. 14, p. 1.

VON Lusciik,\, II. 1855. Die Adergeflcchte de.s men.«chlichen Gehirn.s. G. Reiiner, Berlin.

M.ic.\llu.\i, a. B. 1905. On the di.stribution of pota.s.-ium in anim.'d and vegetable cel!.. J. Phy.-iol., vol. 32, p. 95.

. 1912. Surface ten.-ion and vital phenomena Toronto University Studies.

Phy-i'ilogical Si'rie.J, no. 8,

.MAf.vLLi’ii, A. B., AND .Ml.nte.n, M. L. 1905. Dl.-trihulion of chloride.s in nerve cell’* and filires. Froc. Roy. .‘'oc., lA>ndon, Series B, vol. 77, p. 105.

.MacDonald, .1. .S. 1902. The injury current of nerves : The key to its phy.-ical .struc ture. Rej><jrt of Thoin.-on-Vates Ddxjratory, vol. 1.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 119

MacDonald, J. S. 1905. The structure and function of nerve fibres. Proc. Roy. Soc., London, Series B, vol. 76, p. 322.

McClure, C. F. W. 1896. On the presence of centrosome and attraction spheres in the ganglion cells of Helix pomatia. Bull., Princeton College.

Maccabruni, F. 1912. Zur feineren Struktur der Nervenfasem. Folia neuro-biol Bd. 6, S. 17.

Magendie, F. 1825. Recherches sur le liquide cdphalo-rachidien. Paris.

. 1842. Recherches physiologiques et cliniques sur le liquide cdphalo-rachidien

ou c^r^bro-spinal. Mequianon-Marvis, fils, Paris.

Mall, F. P. 1904. The development of the blood vessels of the brain of the human embryo. Am. J. Anat., vol. 4, p. 1.

Malone, E. F. 1913. The nucleus cardiacus nervi vagi and the three distinct types of nerve cells which innervate the three different types of muscle. Am. J. Anat., vol. 15, p. 121.

. 1923. The cell structure of the superior olive in man. J. Comp. Neurol., vol. 35,

p. 205.

. 1929. Functional significance of histologic character in preganglionic visceral

neurons. Arch. Neurol, and Psychiat., vol. 22, no. 2, p. 295.

Mann, G. 1894. Histological changes induced in sympathetic, motor, and sensory cells by functional activity. J. Anat., vol. 29, p. 100. Scottish Microscopical Society, May 18, 1894.

. 1898. Die fibrillare Struktur der Nervenzellen. Verhandl. d. anat. Gesellsch.,

Bd. 12, S. 39.

Mansfeld, G. 1915. Untersuchungen fiber den chemischen Muskeltonus. II. Mitteilung. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 161, S. 478.

Marcora, F. 1911. Ueber die Histogenese des Zentralnervensystems mit besonderer Rficksicht auf die innere Struktur der Nervenelemente. Folia neuro-biol., Bd. 5, S. 928.

Marinesco, M. G. 1900. Du r61e de la n4vroglie dans l’4volution des inflammations et des tumeurs. Rev. neurol., vol. 8, p. 886 and p. 1146.

. 1906. Quelques recherches sur la morphologie normale et pathologique des

cellules des ganglions spinaux et sympathiques de I’homme. Le N6vraxe, vol. 8, fasc. 1, p. 7.

. 19lk Sur la structure colloidale des cellules nerveuses et ses variations

I’etat normal et pathologique. 3“ Congris intemat. de Neurol, et de Psychiat. Aofit.

. 1919, 1919a. Recherches histologiques sur les oxydases et etudes histologiques

sur les oxydases et les pero.xydases. Compt. rend. Soc. de biol., vol. 82, pp. 98 and 258.

Marinescu, G., and Tufa. 1925. Einige Untersuchungen fiber die Mikroglia in normalen und pathologischen Zustand. Zentralbl. f. d. ges. Neurol, u. Psychiat., Bd. 41, S. 137.

Martynopp, tv. 1914. Nervenendapparate in der Brustwarze der Frau und von Saugetierweibchen. Folia neuro-biol., Bd. 8, S. 249.

Marui, K. 1918. On the finer structure of the synapse of the Mauthner cell with especial consideration of the “Golgi net” of Bethe, nervous terminal feet and the nervous pericellular terminal net of Held. J. Comp. Neurol., vol. 30, p. 127.

Mathews, A. P. 1924. Some general aspects of the chemistry of cells. General Cytology, edited by E. V. Cowdry. Sect. 2, p. 15. Univ. of Chicago Press.

Matsdmoto, T. 1920. The granules, vacuoles, and mitachondria in the S 3 Tnpathetic nerve fibers cultivated in vitro. Bull. Johns Hopkins Hospital, vol. 31, p. 91.

Mattauschek, E. 1909. Ein Beitrag zur Kenntnis der Arachnoidea spinalis. Arb. a. d. neurol. Inst. a. d. TVien. Univ. (Obersteiner’s), Bd. 17, S. 150.

120 NERVOUS SYSTEMS OF VERTEBRATES AND OF AIAN

Mawas, J. 1910. Note sur la structure et la signification glandulairc probable des cellules nevrogliques du systeme nerveux central des vertebres. Compt. rend. Soc. de biol., vol. 69, no. 24, p. 45.

,Mav, P., and Walker, C. E. 1907. Note on the multiplication and migration of nucleoli in nerve cells of mammals. Quart. J. E.xper. Physiol., vol. 1, p. 202.

May, R. M. 1925. The relation of nerves to degenerating and regenerating taste buds. J. Exper. Zool., vol. 42, p. 371.

May, R. M., and Detweiler, S. R. 1925. The relation of transplanted eyes to developing nerve-centers. J. Exper. Zool., vol. 43, p. 83.

Mayer, A., and Schaeffer, G. 1913. Une hypothese de travail sur le role physiologique des mitochondries. Compt. rend. Soc. de biol., vol. 74, p. 1384.

Mays, G. 1893. Ueber die Entwicklung der motorischen Nervenendigung. Zeitschr, f. Biol., Bd. 29 (N. F. Bd. 11), S. 41.

Meek, W. J. 1907. A study of the choroid ple.xus. J. Comp. Neurol., vol. 17, p. 2S6.

Meissner, G. 1853. Beitriige zur Anatomie und Physiologie der Haut. L. Voss, Leipzig.

Me.ncl, E. 1906. Einige Beobachtungen fiber die Roncoronischen Fibrillen der Nervenzellenkerne. Arch. f. mikr. Anat., Bd. 68, S. 527.

Merkel, F. 1875. Tastzellen und Tastkorperchen bei den Haustieren und beim Menschen. Arch. f. mikr. Anat., Bd. 11, S. 636.

. 1875a. Ueber die Endigung der sensiblcn Nerven in der Kaut. Nachr. v. d. k.

Gesellsch. d. Wissensch. u. d. Georg.-Aug. Univ., Gottingen, S. 123. (Quoted from Barker ’99.)

. 1880. Ueber die Endigungen der sensiblen Nerven in der Haut der Wirbelthiere.

H. Schmidt, Rostock.

Merzdaciier, L. 1909. Gibt cs praformierte pcriccllulare Lymphraumc? Neurol. Centralbl., Bd. 28, S. 975.

Mestrezat, W. 1911-12. Le liquide cephalo-rachidien normal et palhologique. Theses Universitt; do Montpellier, no. 17. (Quoted from Clark '29 ; also FreinonlSmith ’27.)

Metz, A,, and Spatz, H. 1924. Die Hortcgaschen Zcllcn (= das sogenannte "drittc Element”) und fiber ihre funktionelle Bedeutung. Zeitschr. f. d. gcs. Neurol, u. Psychiat., Bd. 89, S. 138.

Meyer, A. 1898. Critical review of the data and general methods and deductions of modern neurology, J. Comp. Neurol., v'ol. 8, p. 249.

Meyer, IL, a.nd Ranso.m, F. 1903. Untensuchungen fiber den Tetanus. Arch. f. exper. Patli. u. Pharmakol., Bd. 49, p. 309 (listed as 1913, p. 49).

.Michailow, S. 1908. Das inlrakardiale Nervensystem des Fro.schcs und die Mctluuic von Ramon y Cajal. Internal. Monat.sclir. f. Anat. u. Physiol., Bd. 25, S. 351.

. 1908a. Die Ncurofibrillcn der .sympathischen GanglicnzcIIcn bei Saiigelicrcn.

Folia neuro-biol., Bd. 1, S. 037.

— . 1909. Die Structur der typischen Vatcr-P.'icini.schcn Korperchen und ihre

physiologi.schc Bedeutung. Folia neuro-biol., Bd. 2, S. 003.

. 1909a. Experiincntclle Untersuchung filx;r die Verbindung des peripheren

i-ensiblen Neurons mit dem sympatlii.schcn Nervensystem. Folia neuro-biol., Bd. 3, S. 152.

.Mi.scir, J. 1903. Da.s Binnennetz der .spinalen Ganglienzellen bei ver.'ichiedencn \Virl>ellieren. Internal. .Monat.sclir. f. Anal. u. Phy.siol., Bd. 20, S. 329,

vo.N .Mo.nakow, C., and .Moroue, R. 192,8. Introduction bioiogique i\ I’etude de la neurologic cl de la psychopat hologie. Alcan. Paris.

.Munckeiieik;, G., and IIetiie, A. 1899. Die Degeneration der markhaltigen Nervenfa."vrn der WirU'Ithiere imter hatipt.-achlichcr Beriick-sichtigung des Verhaltens der Primitivfibrillen. Arch. f. rnikr, Anat., Bd. 51, S, 135.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 121

jMonti, R. 1891. Ricerche microscopiche sul sistema nervoso degli insetti. Rendic. d. R. 1st. Lombardo, ser. 2, vol. 25.

Mott, F. W. 1910. The cerebrospinal fluid. Lancet, vol. 2, pp. 1 and 79.

. 1915. ]\Iicroscopic e.xamination of the central nervous system in three cases of

spontaneous hypothyroidism in relation to a type of insanity. Proc. Roy. Soc. Med. (Sect. Psychiat.), vol. 8, p. 58.

IMuhljlvnn, M. 1901. Die Veriinderungen der Nervenzellen in verschiedenem Alter beim IVIeerschweinchen. Anat. Anz., Bd. 19, S. 377.

. 1912. iMikrochemische Untersuchungen an der wachsender Nervenzelle.

Arch. f. mikr. Anat., Bd. 79, S. 175.

Muller, E. 1892. Zur Kenntnis der Ausbreitung und Endigungsweise der MagenDarm-, und Pancreasnerven. Arch. f. mikr. Anat., Bd. 40, S. 390. ’

. 1899. Studien iiber Neuroglia. Arch. f. mikr. Anat., Bd. 55, S. 11.

. 1908. Zur Kenntnis der Ausbreitung und Endigungsweise der Magen-, Darm-,

und Pancreasnerven. Arch. f. mikr. Anat., Bd. 72, S. 554.

Muller-Hettlinge.v, J. 1883. Ueber galvanische Erscheinungen an keimenden Samen. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 31, S. 193.

Murray, P. D. F. 1924. The motor nerve endings of the limb muscles of the frog (Rana temporaria) and of the muscles of the pectoral fin of the dogfish (squalus acanthias). Proc. Lin. Soc. New South Wales, vol. 49, p. 371.

Nageotte, J. 1907. Etude sur la greffe des ganglions rachidiens; variations et tropismes du neurone sensitif. Anat. Anz., Bd. 31, S. 225.

. 1909. Granulations lipoides du tissu nerveux. Compt. rend. Soc. de biol.,

vol. 66, p. 24,

. 1909a. Mitochondries et grains spumeux dans les cellules nerveuses. Compt.

rend. Soc. de biol., vol. 67, p. 130.

. 1910. Phdnomfines de s^cr^tion dans le protoplasma des cellules nivrogliques

de la substance grise. Compt. rend. Soc. de biol., vol. 68, p. 1068.

. 1910, Incisures de Schmidt-Lanterman et protoplasma des cellules de Schwann.

Compt. rend. Soc. de biol., vol. 68, p. 39.

. 1911. Betrachtungen fiber den tatsachlichen Bau und die kfinstlich hervor gerufenen Deformationen der markhaltigen Nervenfaser. Arch. f. mikr. Anat., Bd. 77, S. 245.

. 1913. Quelques considerations sur la fibre nerveuse 3, myeline, A propos du

travail de F. Maccabruni. Folia neuro-biol., Bd. 7, S. 611.

. 1916. Note sur les fibres 5 myeline et sur les etranglements de Ranvier chez

certains Crustaces. Compt. rend. Soc. de biol., vol. 79, p. 259.

. 1920. Phenomlnes de secretion dans le protoplasma des cellules nevroglique

de la substance grisse. Compt. rend. Soc. de biol., vol. 68, p. 1068.

Nexiiloff, a. 1908. Beobachtungen fiber die Nervenelemente bei Ganoiden und Knochenfischen. Tell I. : Der Bau der Nervenzellen. Arch. f. mikr. Anat., Bd. 72, S. 1.

. 1910. Ueber die Beziehung der sog. “Zellen der Schwannschen Scheide,” zur

Myelin in der Nervenfasern von Saugetieren. Arch, f . mikr. Anat., Bd. 76, no. 2, S. 329.

Nicholson, F. M. 1923. The changes in amount and distribution of the ironcontaining proteins of nerve cells following injury to their axones. J. Comp. Neurol., vol. 36, p. 37.

Nicolai, G. F. 1901. Ueber die Leitungsgeschwindigkeit im Riechnerven des Hechtes. Arch. f. d. ges. Physiol. (Pflfiger’s), Bd. 85, S. 65.

Nissl, F. 1884. NaturforscherversammI z Strassburg.

. 1903. Die Neuronenlehre und ihre Anhanger. G. Fischer, Jena.

Nonne, M. 1902. Ueber diffuse Sarkomatose der Pia mater der ganzen Centralnervensystems. Deutsche Zeitschr. f. Nervenh., Bd. 21, S. 396.

122 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Nowick, N. 1910. Zur Frage von dem Bau der Tastzellen in den Grandryschen Korperchen. Anat. Anz., Bd. 36, S. 217.

Nusbaum, J. 1913. Ueber den sogenannten inneren Golgischen Netzapparat und sein Verhaltnis zu den Mitochondrien, Chromidien und anderen Zellstrukturen in Tierreich. Zusammenfassendes Sammelreferat. Arch. f. Zellforsch., Bd. 10, S. 359.

Obersteiner, H. 1900. Zur Histologie der Gliazellen in der Molecularsehichte der Grosshirnrinde. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 7, S. 301.

. 1903. Arbeiten der Neurologic.

. 1912. Anleitung beim Studium des Baues der nervosen Zentralorgane im

gesunden und kranken Zustande. Aufl., F. Deuticke, Leipzig und Wien.

. 1913. Bemerkungen zur Bedeutung der wechseinden Grosse von Nervenzellen.

Festschrift f. Bianchi.

Olmsted, J. M. D. 1920. The nerve as a formative influence in the development of tastebuds. J. Comp. Neurol., vol. 31, p. 465.

OuDENDAL, A. J. F. 1912. Ueber den Zusammenhang der Auslaufer der Korbzellen mit den Zellen von Purkinje in der Rinde des Kleinhirns. Psychiat. en neurol. Bl., Amsterdam, Bd. 16, S. 10.

Pacini, F. 1836. Nuovo giornale dei letterati, vol. 32, Pisa. (Quoted from Ram6n y Cajal ’ll.)

. 1840. Nuovi organi scoperti nel corpo umano. Pistoja.

Paladino, G. 1892. De la continuation de la ndvroglie dans le squelette my41inique des fibres nerveuses et de la constitution pluricellulaire du cylindera.xe. Rendic. d. r. Accad. d. sc. fis. e mat. di Napoli, Ser. 2“, vol. 6, p. 153. See also Arch. ital. de biol., 1893, vol. 19, p. 26.

Pansini, S. 1889. Des terminaisions des nerfs sur les tendons des vertebres. Arch, ital. de biol., vol. 11, p. 225.

Parker, G. H. 1909. The origin of the nervous system and its appropriation of effectors. Pop. Sc. Month., vol. 80, pp. 56, 137, 253, and 338.

. 1911. The origin and significance of the primitive nervous system. Proc. Am.

Philos. Soc., vol. 50, p. 217.

, 1914. The origin and evolution of the nervous system. Pop. Sc. Month.,

vol. 84, p. 118.

. 1918. The rate of transmission in the nerve net of the coelenterates. J. Gen.

Physiol., vol. 1, p. 231.

. 1919. The elementary nervous system. J. B. Lippincott Co., Philadelphia and

London.

. 1929. The neurofibril hypothesis. Quart. Rev. Biol., vol. 4, p. 155.

Parsons, J. H. 1927. Introduction to the theory of perception. Cambridge Psychological Library.

Pastori, G. 1929. Di un methodo di dofpia impregnazione adatto alio studio della fine strut tura delle cellule gangliari simpatiche. Monitore zool. ital., vol. 40, p. 182.

Pellizzi, G. B. 1896. Sulla struttura e sull’origine delle granulazioni ependimali (Contributo all’istologia e patologia della nevroglia). Riv. sper. di freniat., vol. 22, p. 466.

. 1911. Experinientelle histologische Untersuchungen fiber die Plexus chorioidei

(.\dergeflechte). Folia neuro-biol., Bd, 5, S. 305.

Penkield, W. G. 1920. Alterations of the Golgi apparatus in nerve cells. Brain, vol. 43, p. 290.

. 1921. The Golgi apparatus and its relationship to Holmgren’s trophospongiurn

in nerve cells. Comparison during retispersion. Anat. Rec., vol. 22, p. 57.

. 1924. Microglie et son rapport avec la d6g6D5ration ndvrogliale dans un gliome.

Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 22, p. 277.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 123

Penfield, W. G. 1924. Oligodendroglia and its relation to classical neuroglia. Brain, vol. 47, p. 430.

. 1925. Microglia and the process of phagocytosis in gliomas. Am. J. Path

vol. 1, p. 77.

. 1925a. Phagocytic activity of microglia in the central nervous system. Proc.

N. y. Path. Soc., N. S., vol. 25, nos. 1 and 5.

. 1928. Neuroglia and microglia. The interstitial tissue of the central nervous

system. Special Cytology (edited by E. V. Cowdry), sect. 30. Hoeber, New York.

. 1932. Neuroglia : normal and pathological. Vol. 2, sect. 9, p. 423 in Cytology

and cellular pathology of the nervous system, edited by W. Penfield. P. Hoeber, New York.

Pensa, a. 1905. Osservazione nella distribuzione dei nervi sanguini e dei nervi nel pancreas. Internal. Monatschr. f. Anat. u. Physiol. Bd. 22, S. 90.

Penta, P. 1931. Studi sulla neuroglia quattro tipi di oligodendrocili secondo Del Rio Hortega. Boll. Soc. Ital. Biol. Sperim., vol. 6, p. 15.

Pehez, R. M. 1931. Una importante modalidad de terminacion nerviosa en la piel humana. Arch. Neurobiol., vol. 11, p. 23.

Perroncito, a. 1901. Sur la terminaisons des nerfs dans les fibres musculaires strides. Arch. ital. de biol., vol. 36, p. 245.

. 1902. Etudes ultdrieures sur la terminaison des nerfs dans les muscles a fibres

strides. Arch. ital. de biol., vol. 38, p. 393.

. 1907. Die Regeneration der Nerven. Beitr. z. path. Anat. u. z. allg. Path.

(Ziegler’s), Bd. 42, no. 2, p. 355.

Peterfi, T. 1932. De invloed van electrische prikkeling op de zenuwcel. Genostschap ter Bevordering van Natuur — , Genees — en Heelkun de te Amsterdam. Biol. Abdeeli, p. 1.

Peterfi, T., and ILapel, O. 1928. Die Wirkung der Austechens auf das Protoplasma der in vitro gezuchteten Gewekszellen. IV Die Pigmentzellen. Arch. f. exper. Zellforschung. Vol. 5. p. 341.

Pettit, A., and Girard, J. 1902. Sur la morphologic des plexus choroldes du systdme nerveux central. Corapt. rend. Soc. de biol., vol. 54, p. 698. Also in Arch, d’anat. micr., vol. 5, p. 213.

PiGHiNi, G. 1912. Chemische und biochemische Untersuchungen fiber das Nervensystem unter normalen und pathologischen Bedingungen. I. Mitteilung. Ueber die Indophenol-Oxydase im Zentralnervensystem, in der Tela chorioidea und in der Cerebrospinalflfissigkeit. Biochem. Zeitschr., Bd. 42, S. 124.

Pines, L. 1931. Allgemeine Ergebnisse unserer Untersuchungen fiber die Innervation der innersekretorischen Organe. Arch. f. ges. Physiologie (Pflfiger’s), vol. 228, S. 373.

Plaut, F., Rehm, 0., AND ScHOTTMULLEE, H. 1913. Leitfaden zur Untersuchung der Zerebrospinalflfissigkeit. G. Fischer, Jena.

Ploschko, a. (mitgeteilt v. Arnstein). 1897. Die Nervenendigungen und Ganglien der Respirationsorgane. Anat. Anz., Bd. 13, S. 12.

Poirier, P., and Charpy, A. 1901. Traite d’anatomie humaine. I‘" fasc., Masson

and Co., Paris. -dj oo c? fr.*

PoNZio, F. 1906. Le terminazioni nervose nel polmone. Anat. Anz., Bd. 28, b. 74.

PoPA, T. 1924. Structure fonctionelle de la dure-mere craniens avec considerations generates sur les facteurs mecaniques craniens chez les vertebres en general et chez I’homme en particulier. Ann. sc. de I’Univ. de Jassy, vol. 13, p. 119.

POSCHARISSKY J 1907. Ueber die histologische Vorgange an den penphenschen Nerven Trh Kontinuitatstrennung. Beitr. z. path. Anat. u. z. allg. Path. (Zeigler’s), Bd. 41, no. 1, S. 52.

124 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Prenant, a. 1911. ProblSmes cytologiques gdndraux soulev6s par I’^tude des cellules musculaires. J. de I’anat. et physiol., vol. 47, p. 601.

Prentiss, C. W. 1904. The nervous structures of the palate of the frog : The peripheral networks and the nature of their cells and fibers. J. Comp. Neurol., vol. 14, p. 93.

Pruijs, W. M. 1927. Ueber die Mikroglia, ihre Herkunft, Funktion und ihre Verhaltniss zu anderen Glia Elementen. Zeitschr. f. d. ges. Neurol, u. Psychiat., Bd. 108, S. 298.

Putnam, T. J., and Schaltenbrand, G. 1927. Untersuchungen zum Kreislauf des Liquor cerebrospinalis mit Hilfe intravenoser Fluoreszineinspretzungen. Deutsches Zeitschr. f. Nervenheilk., vol. 96, S. 123.

Quincke, H. 1872. Physiologic der Cerebrospinalfliissigkeit. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 153.

Rabl, K. 1889. Ueber die Prinzipien der Histologic. Verhandl. d. anat. Gesellsch., Bd. 3, S. 39.

Raffaele, F. 1900. Per la genesi dei nervi de catene cellulari. Anat. Anz., Bd. 18, S. 337.

Ram6n y Cajae, S. 1881. Observaciones microc6picas sobre las terminaciones nerviosas en los mdsculos voluntaries (avec trois planches en lithographie). Zaragoza.

. 1888. Terminaciones en los husos musculares de la Rana. Rev. trim, de histol.

norm, y patol., vol. 1, fasc. 1.

. 1891. Sur la structure de I’^corce c4r6brale de quelques mamraif^res. La

Cellule, vol. 7, p. 123.

. 1893. La ratine des vert4br4s. La Cellule, vol. 9, p. 119.

. 1900. Textura del sistema nervioso del hombre y de los vertebrados, vol. 1.

N. Moya, Madrid. 1899-1904.

. 1904. Variaciones morfoldgicas normales y patol6gicas del reticulo neurofibrilar.

Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 3, fasc. 1, p. 9.

. 1905. Genesis de los fibres nerviosos en el embrion y observaciones contraries

a la teoria catenaria. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 4.

. 1906. Structure et connections des neurones. Conference Nobel faite a Stockholm. December.

. 1907. Nouvelles observations sur revolution des neuroblastes avec quelques

remarques sur I’hypothese neurogenetique de Hensen — Held. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 5, p. 169.

. 1907a. Note sur la degenerescence traumatique des fibres nerveuses du cervelet

et du cerveau. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 5, p. 105.

. 1908. Influencia de las condiciones mecdnicas sobre la regeneracidn de los

nervios. Bol. Soc. espan. de biol.

. 1908a. Regeneracion de los nervos. La clinica moderna. Zaragoza, vol. 7,

no. 75, p. 185.

. 1908b. Nervenregeneration. J. A. Barth, Leipzig.

. 1909, 1911. Histologic du systeme nerveux de I’homme et des vertebres. Vol.

I, 1909; vol. 11, 1911. A. Maloine, Paris.

. 1910. Algunas observaciones favorables d. la hip6tesis neurotr6pica. Trab. d.

lab. de invest, biol. Univ. de Madrid, vol. 8, p. 63.

. 1913. Sobre un nuevo proceder de impregnacion de la neuroglia y sus resultados

en los centros nerviosos del hombre y animales. Trab. d. lab. do invest, biol. Univ. de Madrid, vol. 11, p. 219.

. 1913a. Contribuci6n al conocimiento de la neuroglia del cerebro humane.

Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 11, p. 255.

. 1915. Algunas variaciones fisiologicas y patoldgicas del aparato reticular de

Golgi. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 12, p. 127.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 125

R\m6n' y Cajal, S. 1925. Quelques remarqucs sur les plaques motrices de la langue (les inaminifilrcs. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 23, p. 245.

. 1928. Degeneration and regeneration of the nervous system. Translated by

May. O.xford Univ. Press, London.

ILvnson, S. W. 1912. Degenenition and regeneration of nerve fibers. J. Comp. Neurol., vol. 22, p. 487.

. 1912a. The structure of the spinal ganglia and of the spinal nerves. J. Comp.

Neurol., vol. 22, p. 159.

. 1915. Unmyelinated nerve-fibres as conductors of protopathic sensation.

Brain, vol. 38, p. 381.

. 1930. The anatomy of the nervous system. 4th edition. W. B. Saunders Co.,

Philadelphia and London.

Ranso.v, S. W., and Billingslky, P. R. 1918. The superior cervical ganglion and the cervical portion of the sympathetic trunk. J. Comp. Neurol., vol. 29, p. 313. R.\nvier, M. L. 1874 (also 1889). Traitd technique d’histologie. F. Savy, Paris.

. 1878. Lemons sur I’histologic du systdme ncrveu.Y. F. Savy, Paris. Vols. I-II.

. 1888. Quoted from Slohr (’28), p. 317.

ILvs.musse.n, a. T. 1927. Additional evidence favoring the normal e.\-istence of the lateral apertures of the fourth ventricle of man. Anat. Rec., vol. 33, p. 179. R-vuuEn, A. A. 1903. Lchrbuch dcr Anatomic des Mcnschen. O'” Aufl., Bd. 2. G. Thiemc, Leipzig.

Redenz, E. 1925. Mikrochemischer Nachweis von Nuklcin.saurc in den Nisslschollen der motorischen Ganglienzcllcn. Wiirzburgcr med. Zeitschr., N. F., Bd. 49, S. 220. Redlicu, E. 1897. Die Pathologic der tabischen Hinterstrangserkrankung. G. Fischer, Jena.

Reqaud, C. 1908. Sur les mitochondries di lY'pithdlium sdminal. IV. Faits et hypotheses rclatifs h Icur constitution. Compt. rend. Soc. de biol., vol. 65, p. 718. Reinek, M. 1895. Zur Lehrc vom Hirndruck. Wien. klin. Wochenschr., Bd. 8, S. 371. Reiner, M., and Schnitzler, J. 1894. Ueber die Abflusswege des Liquor cerebrospinalis. Zentralbl. f. Physiol., Bd. 8, S. 684.

Rendahl, H. 1924. Embryologische und morphologischc Studien iiber das Zwischenhim beim Huhn. Acta Zool., Bd. 5, S. 241.

DE Rfi.NYi, G. S. 1929. The structure of cells in tissues as revealed by microdissection. II. The physical properties of the living axis cylinder in the myelinated nerve fiber of the frog. J. Comp. Neurol., vol. 47, p. 405.

. 1929a. Idem. III. Observations on the sheaths of myelinated nerve fibers of

the frog. J. Comp. Neurol., vol. 48, p. 295.

. 1929b. Idem. IV. Observations on neurofibrils in the living nervous tissue

of the lobster (Homarus amcricanus). J. Comp. Neurol., vol. 48, p. 441.

Retzius, G. 1890. Zur Kenntnis des Nervensystems der Krustaceens. Biol. Untersuch., N. F., Bd. 1, S. 1.

. 1891. Zur Kenntnis des centralen Nervensystems von Amphioxus lanceolatus.

Biol. Untersuch., N. F., Bd. 2, S. 29.

. 1891a. Zur Kenntnis des centralen Nervensystems von Myxine glutinosa.

Biol. Untersuch., Bd. 2, S. 47.

. 1892. Die sensibilen Nervenendigungen in der Haut des Petromyzon. Biol.

Untersuch., N. F., Bd. 3, S. 37.

. 1892a. Zur Kenntnis der motorischen Nervenendigungen. Biol. Untersuch.,

N. F., Bd. 3, S. 41. , ^ J

. 1892b Zur Kenntnis der Nervenendigung m den Drusen und Zhanen. Biol.

Untersuch., N. F., Bd. 4, S. 64. . ^ ^ r- , tt . k

. 1893. Weiteres fiber die Endigungsweise des Gehomerven. Biol, untersuch.,

N. F., Bd. 5, S. 35.

126 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Retzius, G. 1893. Die Cajal’schen Zellen der Grosshirnrinde beim Menschen und bei Saugethieren. Biol. XJntersuch., N. F., Bd. 5, S. 1.

. 1893. Studien liber Ependym und Neuroglia. Biol. Untersuch.,N. F., Bd.5, S.9.

. 1894. Weitere Beitrage zur Kenntniss der Cajal’schen Zellen der Grosshirnrinde

des Menschen. Biol. Untersuch., N. F,, Bd. 6, S. 29.

. 1894a. Die Neuroglia des Gehims beim Menschen und bei Saugethieren. Biol.

Untersuch., N. F., Bd. 6, S. 1.

. 1898. Was ist die Henle’sche Scheide der Nervenfasern ? Anat. Anz., Bd. 15,

S. 140.

. 1898a. Zur Frage von der Endigungsweise der peripherischen sensiblen Nerven.

Biol. Untersuch., N. F., Bd. 8, S. 114.

. 1898b. Die Methylenenblaufarbung bei den lebenden Amphioxus. Biol. Untersuch., N. F., Bd. 8, S. 118.

. 1905. Zur Kenntnis vom Bau der Selachier-Retina. Biol. Untersuch., N. F.,

Bd. 12, S. 55.

. 1908. The principles of the minute structure of the nervous system as revealed

by recent investigations. Croonian Lecture. Proc. Roy. Soc., London, Series B, vol. 80, p. 414.

Reynolds, F., and Slater, J. 1928. A study of the structure and function of the interstitial tissue of the central nervous system. Edinburgh Med. J., p. 49. Rhinehabt, D. a. 1912. The nerves of the thyreoid and parathyreoid bodies. Am. J. Anat., vol. 13, p. 91.

Ritter, C. 1891. Zur Histologie der Zapfen der Fischretina. Internat. Monatschr. f. Anat. u. Physiol., vol. 8, S. 128.

RrvA, E. 1905, 1906. Lesioni del reticolo neurofibrillare della cellula nervosa nell’inanizione sperimentale studiate con i metodi del Donaggio. Riv. sper. di freniat., vol. 31, p. 245, and vol. 32, p. 400.

Rivers, H. R., and Head, H. 1908. A human experiment in nerve division. Brain, vol. 31, p. 323.

Rogers, L., and West, C. M. 1931. The Foramen of Magendie. J. Anat., vol. 65, p. 457.

Robertson, W. F. 1900. A textbook of pathology in relation to mental diseases. W. F. Clay, Edinburgh.

Rohde, E. 1916. Die Entstehung des Nervensystems aus einem vielkernigen Plasmodium und die sekundare Verbindung von Ganglienzelle und Nervenfa.ser. Zeitschr. f. wissensch. ZooL, Bd. 116, S. 627.

Rollett, a. 1876. Ueber einen Nervenplexus und Nervenendigungen in einer Sehnc.

Sitzungsb. d. Akad. de wissensch. zu Wien, Math.-nat. Cl., Bd. 73, Abt. 3, S. 34. Rossi, O. 1908. Process! rigenerativi e degenerativi conseguenti a ferite asettichc del sistema nervoso centrale. Midollo spinalle e ner vo ottico. Riv. di pat. nerv., vol. 13, p. 481.

Rossi, U. 1908. Per la rigenerazione dei neuroni. Trab. d. lab. de invest, biol. Univ'.

de Madrid, vol. 6, p. 227.

Rothert, W. 1907. Die neuen Untersuchungen iiber den Galvanotropismus der Pflanzenwurzeln. Zeitschr. f. allg. Physiol., Bd. 7, S. 142.

Rouget, Ch. 1862. Note sur la terminaison des nerfs moteurs dans les muscles chez les reptiles, les oiseaux et les mammifCres. Compt. rend, hebdom. d. stances d. I'acad. d. sc., vol, 55, p. 548.

Ruffini, a. 1892. Di una particulare reticella nervosa e di ulcuni corpuscoli dei Pacini che si trovano in conne.ssione cogli organ! muculo-tcndinci del gatto. Rendic. d. R. Accad. d. Lincei, vol. 1, p. 31.

. lS92a. Sur la termination nerveu.x dans les faisceaux mu.sculaires ct .sur lour

signification physiologique. Arch, ital. de biol,, vol. 18, p. 106.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 127

RuFFiNr, A. 1893. Sur un reticule special, et sur quelques corpuscles de Pacini qui se trouvent en conne.xion avcc les organnes musculo-tendineux du chat. Arch. ital. de biol., vol. IS, p. 101.

. 1898. Sulla fina anatomia dei fusi neuro-muscolari del gatto. Sienna.

. lS9Sa. On the minute anatomy of the neuromuscular spindles of the cat, and

their physiological significance. J. Physiol., vol. 23, p. 190.

. 1900. Sulle fibrille nervose ultraterminali nelle piastre motrici dell’uomo. Riv.

di pat. nerv., vol. 5, p. 433.

. 1902. Sull’apparato nervoso di Timofeew od apparato ultra-terminale nei

corpuscoli del Aleissner della cute umano. Bibliogr. anat., Bd. 11, p. 267.

. 1905. Sur les e.xpansions nerveuses de la peau chez I’homme et quelques autres

mammiferes. Rev. gen. d’histol., vol. 1, fasc. 3, p. 421.

Sabin, F. R. 1912. The development of the lymphatic system. Manual of human embryology, vol. 2. Keibel, F. and Mall, F. P., J. B. Lippincott and Co., Philadelphia and London.

Sachs, C. 1875. Die Nerven der Sehnen. Arch. f. Anat., Physiol, u. wissensch. Med., no. 4, S. 402.

Sagemehl, hi. 1884. Beitriigc zur verglcichenden Anatomie der Fische. II. Einige Bemerkungen iiber die Hirnhiiute der Knochenfische. Morphol. Jahrb., Bd. 9, S. 457.

Sala, G. 1899. Untersuchungen iiber die Struktur der Pacinis’chen Korperchen. Anat. Anz., Bd. 16, S. 193.

Salvi, G. 1898. Histogen5se et structure des mdninges. Thfese de Paris.

Samassa, P. 1893. Ueber die Nerven des augentragenden Fuhlers von Helix pomatia. Zool. Jahrb., Abt. f. Anat., Bd. 7, S. 593.

Sanchez, M. 1916. Recherches sur le rdseau endocellulaire de Golgi dans les cellules de I’dcorce du cervelet. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 14, p. 87.

Sato, K. 1929. Ueber den Wasser- und N-Haushalt der Gewebe nach Glukoseinfusion : Wasserhauskalt der Gewebe nach Glukoseinfusion. Tohuku J. Exper. Med., vol. 13, p. 545.

. 1929a. Ueber den Wasser und N-Hauskalt der Gewebe nach Glukoseinfusion ;

N-Hauskalt der Gewebe nach Infusion hypertonischer Glukoselosung. Tohuku J. E.xper. Med., vol. 13, p. 553.

ScAFFiDi, F. V. 1910. Sulle correnti di demarcasione dei nervi duronte la degenerazione AValleriana e la regenerazione. Zeitschr. f. allg. Physiol., Bd. 11, S. 339.

Schafer, E. A. 1912. Quain’s textbook of microscopic anatomy, vol. 2, pt. 1. Longmans, Green and Co., London, New York, Bombay, and Calcutta.

Schaffer, J. 1894. Die oberfliichliche Gliahiille und das Stiitzgeriist des weissen Riickenmarksmantels. Anat. Anz., Bd. 9, S. 262.

. 1896. Ueber einem neuen Befund von Centrosomen in Ganglien- und Knorpel zellen. Sitzungsb. d. K. Akad. d. Wissensch. zu Wien, Math. -nat. CL, Bd. 105,

Abt. 3, S. 21. . .

SCHALTENBRAND, G., AND Baibey, P. 1927. Die perivaskulare Piaghamembran des Gehims. J. f. Psychol, u. Neurol., vol. 35, S. 199.

Schafer, A. 1897. Die fruhesten Differenzirungsvorgange im Centralnervensystem. Krit’isclie Studie und Versuch einer Geschichte der Entwickelung nervoser Substanz. Zeitschr. f. wissensch. Biol., Abt. D, Arch. f. Entwicklungsmechn. d. Organ., Bd 5 S 81

. 1898 Experimentelle Studien an Amphibienlarven. Erste Mittheilung. Ha ben kiinstlich anceleate Defekte des Centralnervensystems oder die vollstandige E^miTtion derselben einen nachweisbaren Einflusz auf die Entmckelung der Gesammtorganismus junger Froschlarven ? Zeitschr. f. wissensch. Biol., Abt. D, Arch. f. Entwicklungsmechn. d. Organ., Bd. 6, S. 151.

128 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ScHELLENBERG, H. C. 1906. UntersuchuBgen iiber den Einflus der Salze auf die Wachstumsrichtung der Wurzeln, zunachst an der Erbsenwurzel. Flora, Bd. 96, S. 474.

ScHiEFFERDEGKER, P. 1906. Ueber das Verhalten der Fibrillen des Achsenzylinders an den Ranvierschen Einschntirungen der markhaltigen Nervenfasern. Arch. f. mikr. Anat., Bd. 67, S. 783.

ScHLAPFER, V. 1905. Ueber den Ban und die Funktion der Epithelzellen des Plexus chorioideus in Beziehung zur Granulalehre und mit besonderer Beriicksichtigung der vitalen Farbungsmethoden. Beitr. z. path. Anat. u. z. allg. Path. (Ziegler's), Supplementband 7, S. 101.

Schneider, K. C. 1902. Lehrbuch der vergleichenden Histologic der Tiere. G. Fischer, Jena.

ScHNOPFHAGEN, F. 1882. Das Ependyma der Gehirnventrikel und die an demselben bemerkbaren Granulationen. Jahrb. f. Psychiat. u. Neurol., Bd. 3, S. 1.

ScHROEDER VAN DER Kolk, J. L. C. 1859. Bau und Functionen der Medulla spinalis und oblongata. Aus dem Hollandischen iibertragen von Dr. F. W. Thiele. F. Vieweg u. Sohn, Braunschweig.

ScHULTZE, M. 1871. Allgemeines iiber die Strukturelemente des Nervensystems. Strieker’s Handbuch der Lehre von den Geweben des Menschen und der Thiere. W. Engelmann, Leipzig.

ScHULTZE, 0. 1905. Beitriige zur Histogenese des Nervensystems. I. Ueber die

multizellulare Entstehung der peripheren sensiblen Nervenfaser und das Vorhandensein eines allgemeinen Endnetzes sensibler Neuroblasten bei Amphibienlarven. Arch. f. mikr. Anat., Bd. 66, S. 41.

VON ScHUMAKER, S. 1911. Bcitrage zur Kenntnis des Baues und der Funktion der Lamellenkorperchen. Arch. f. mikr. Anat., Bd. 77, S. 157.

Schwalbe, G. 1887. Lehrbuch der Anatomie der Sinnesorgane. Hoffmann’s Lehrbuch der Anatomie des Menschen, Bd. 2, Abt. 3, S. 1. E. Besold, Erlangen.

Scott, F. H. 1899. On the structure, micro-chemistry and development of nerve cells, with special reference to their nuclein compounds. Tr. Roy. Canad. Inst., vol. 6, p. 405.

Sedgwick, A. 1895. On the inadequacy of the cellular theory of development of nerves, particularly of the third nerve, and of the sympathetic in Elasmobranchii. Quart. J. Micr. Sc., N. S., vol. 37, p. 87.

Semon, R. 1911. Die Mneme als erhaltendes Prinzip in Wechsel des organischen Geschehens. W. Engelmann, Leipzig.

Sfameni, P. 1900. Di una particolare reticella nervosa amielinica existente intorno ai corpuscoli di Grandry. Ann. di freniat. e Sc. aff. d. R. Manicomio di Torino.

. 1901. Sur un r^seau nerveaux amy61inique e.xistant autour des corpuscules

de Grandry. Arch. ital. de biol., vol. 35, p. 198.

. 1902. Sal modo di terminare dei nervi nei genitali esterni della femina. Mom tore zool. ital., vol. 13.

. 1904. Sulle terminazioni nervose nei genitali femminili esterni e sul loro sig nificato morfologico e funzionale. Arch, di fisiol., vol. 1, p. 345.

Sharp, R. G. 1914. Diplodinium ecaudatum, with an account of its neuromotor apparatus. Univ. California Publ., Zool., vol. 13, p. 43.

Sheehan, D. 1932. The cell station of the Vater-Pacinian corpuscle in retroperitoneal tissue. Brain, vol. 55, p. 493.

Sherrington, C. S. 1894. The nerves of skeletal muscles. J. Physiol., vol. 17, p. 211.

. 1906. The integrative action of the nervous system. C. Scribner’s Sons, New

York.

SiHLER, C. 1895. Ueber Muskelspindeln und intramuskuliire Nervenendigungen bei Schlangen und Froschen. Arch. f. mikr. Anat., Bd. 46, S. 709.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 129

Sjovall, E. 190G. Ueber Spinalganglienzellen und Markscheiden. Zugleich ein Versuch die Wirkungsweiso dcr Osiniumsaure zu analysieren. Anat. Hcfte, Bd. 30. S. 250.

Smallwood, W . M., an’d Rogeus, C. G. 1909. Studies on nerve cells. II. The comparative cytology and physiology of some of the metabolic bodies in the cytopko-m of invertebnite nerve cells. Folia neuro-biol., Bd. 3, S. 11.

Smiii.n’ow, a. 1895. Ueber die sensiblcn Ncrvcncndigungcn im Herzen bei Amphibien und Siiugetieren. Anat. Anz., Bd. 10, S. 737.

. 1901. Zur Frage von dcr Endigung dcr motorischen Nerven in den Herzmuskeln

der Wirbcltiere. Anat. Anz., Bd. 18, S. 105.

. 1901a. Ueber die Nervenendigungen in den Niercn der Siiugetiere. iVnat.

.Anz., Bd. 19, S. 3-17.

S.\iiTH, G. 1900. The eye.s of certain pulnionatc gastcropods, with special reference to the neurofibrillae of Limax m:iximu.s. Bull. Mus. Comp. Zool., Harvard, vol. 48, p. 233.

SoKOLow, A. 1899. Zur Frage iiber die Endigung der Nerven in den Vater-Pacini ’schen Korix-'rchcn. .Anat. .Anz., Bd. 10, S. 421.

Si'EiDEL, C. 1932. Studies of living nerves. I. The movements of individual sheath cells and nerve sprouts correlated with the process of myelin-sheath formation in amphibian larvae. J. E.\p. Zool., vol. 01, p. 279.

. 1933. Idem. II. Activities of amoeboid growth cones, sheath cells, and

myelin segments, as revealed by prolonged observations of individual nerve fibers in frog tadpoles. Am. J. Anat., vol. 52, p. 1.

SpiELMEVEii, \V. 1917. Ueber Regeneration peripherischer Nerven. Zeitschr. f. d.

ges. Neurol, u. P.sycliiat., Bd. 30, S. 421.

, 1922. Histopathologic des Nervensystems. J. Springer, Berlin.

Spina, A. 1898. Experimentellcr Beitrag zur Kenntniss der Hyperamie des Gehirns. Wien. mcd. Bl., Bd. 21, S. 247.

. 1899. Kxiicrimcnteile Untcrsuchungen fiber die Bildung des Liquor cerebro spinalis. .Arch, f. d. ges. Phy.siol. (Pfliiger’s), Bd. 70, S. 204.

. 1900. UelxT den Einfiuss des hohen Blutdruckes auf die Neubildung der Cere brospinalflubiigkeit. Arch. f. d. ges. Physiol. (PflUger’s), Bd. 80, S. 370.

. 1900a, 1901. Untersuchungen fiber die Resorption des Liquors bei normalem

und erhohtem intrucraniellcm Druckc. Arch. f. d. ges. Physiol. (Pflfiger’s), I. .Mittheilung, Bd. 83, S. 120; II. iMittheilung, Bd. S3, S. 415.

Steiwnelli, a. 1911. Contributo alia pifi intima conoscenzadei rapporti tra le piastre motrici. Monitore zool. ital., vol. 22, p. 101.

. 1912. La piastra motricc sccondo Ic vccchic e le nuovo veduti. Ann. di neurol.,

vol. 30, p. 101.

. 1915. Sui dispositivi microscopic! della sensibility cutanea e nella mucosa orale

dei Rettili. Internat. Monatschr. f. .(Anat. u. Physiol., Bd. 32, S. 8.

. 1910. Nuovo contributo alia conoscenza delle espansioni sensitive dei Rettili,

e considerazioni sulla tessitura del sistema nervoso periferico. Internat. Monatschr. f. Anat. u. Phy.siol., Bd. 32, S. 22.

Steuzi, G. 1900, 1901. Ricerche intorno all’ anatomia comparata ed all’ ontogenesi delle meningi, e considerazioni sulla filogencsi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 00, pt. 2, p. 1101.

. 1902. Intorno alia divisione della dura madre dall’ endocranio. Monitore

zool. ital., vol. 13.

. 1902a. Recherches sur I’anatomie comparde et sur I’ontogdnese des mdninges.

Arch. ital. de biol., vol. 37, p. 257.

Sroim, P., Ju. 1920. Ueber die Innervation derHarnblase und der Samenblase beim Menschen. Zugleich ein Beitrag fiber die Beziehungen zwischen nerv und glatte

130 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Muskulatur. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 78, S. 555.

Stohr, P., Jr. 1927. Beobachtungen und Bemerkungen iiber den Aufbau des sympathischen Grenzstranges. Zeitschr. f. wissensch. Biol . ; Abt. B., Zeitschr. f. Zellforsch. u, mikr. Anat., Bd. 5, S. 117.

. 1928. Das peripherischen Nervensystem. Das Handbuch der mikroscopischen

Ainatomie des Menschen, edited by von Mollendorf, G. Fischer, Jena.

Stone, L. S. 1924. Experiments on the transplantation of placodes of the cranial ganglia in the amphibian embryo. I. Heterotropic transplantations of the ophthalmic placode upon the head of Amblystoma punctata. J. Comp. Neurol., vol. 38, p. 73.

. 1928. Idem. III. Preauditory and postauditory placodal materials interchanged. J. Comp. Neurol., vol. 47, p. 117.

Strasseh, H. 1892. Alte und neue Probleme der ent-wickelimgsgeschichten Forschung auf dem Gebiete des Nervensystems. Ergeb. d. Anat. u. Entwicklungsgesch. (Merkel u. Bonnet), Bd. 1, S. 721.

Streeter, G. L. 1904. The structm-e of the spinal cord of the ostrich. Am. J. Anat., vol. 3, p. 1.

. 1912. The development of the nervous system. Manual of human embryology,

Keibel, F., and Mall, F. P. J. B. Lippincott Co., Philadelphia and London.

VAN DER Stricht, 0. 1909. Le neuroepith41ium olfactif et sa membrane limitante

interne. M^m. Acad. roy. de m4d. de Belgique, vol. 20, no. 2.

Stroebe, H. 1893. Experimentalle Untersuchungen iiber Degeneration und Regeneration peripherer Nerven nach Verletzungen. Beitr. z. path. Anat. u. z. allg. Path, (Ziegler’s), Bd. 13, no. 2, S. 160.

Strong, R. M., Greene, L. D., and Oliveira, J. V. 1926. The lateral aperture of the fourth ventricle in man. Anat. Rec., vol. 32, p. 223.

Strongman, B. T. 1917. A preliminary experimental study on the relation between mitochondria and discharge of nervous activity. Anat. Rec.. vol. 12, p. 167.

Studnicka, F. K. 1900. Untersuchungen iiber den Bau des Ependyms der nervosen Centralorgane. Anat. Hefte, Bd. 15, S. 301.

SuNDWALL, J. 1917. The choroid plexus with special reference to interstitial granular cells. Anat. Rec., vol. 12, p. 221.

SzYMONOWicz, L. 1895. Beitrage zur Kenntnis der Nervenendigungen in Hautgebilden. Arch. f. mikr. Anat., Bd. 45, S. 624.

. 1897. Ueber den Bau und die Entwicklung der Nervenendigungen im Enten schnabel. Arch. f. mikr. Anat., Bd. 48, S. 329.

Tajiura, T. 1927. Studies on the neuroglia, carried out mainly in the normal dog brain. Orienta Bui. Neuro-biol., vol. 1, p. 57.

Tashiro, S. 1913. Carbon dioxide production from nerve fibers when resting and when stimulated ; a contribution to the chemical basis of irritability. Am. J. Physiol., vol. 32, p. 107.

. 1913a. A new method and apparatus for the estimation of e.xceedingly minute

quantities of carbon dioxide. Am. J. Physiol., vol. 32, p. 137.

. 1915. The metabolism of resting nerve and its correlation with the direction

and rate of nerve impulse. Proc. Am. Physiol. Soc., Am. J. Physiol., vol. 36, p. 368.

. 1917. A chemical sign of life. Univ. of Chicago Press, Chicago.

. 1922. Studies on alkaligenesis. I. Ammonia production in the nerve fiber

during excitation. Am. J. Physiol., vol. 60, p. 519.

Taylor, C. V. 1920. Demonstration of the function of the neuromotor apparatus in Euplotes by the method of microdissection. Univ. California Publ., Zool., vol. 19 , p. 403.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 131

Tello, J. F. 1904. Las neurofibrillas en los vertebrados inferiores. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 3, p. 113.

. 1905. Terminaciones en los museulos estriados. Trab. d. lab. de invest, biol.

Univ. de Madrid, vol. 4.

. 1907. Degeneration et regeneration des plaques motrices apres la section des

nerfs. Trab. d. lab. de invest, biol. de Madrid, vol. 5, p. 117.

. 1917. Genesis de las terminaciones nerviosas motrices y sensitivas. I. En el

sistema locomotor de los vertebrados superiores. Histogenesis muscular. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 15, p. 103.

. 1922. Die Enstehungen der motorischen und sensiblen Nervenendigungen.

Zeitschr. f. d. ges. Anat., Bd. 64, S. 348.

Teeni, T. 1922. Sulla partecipazione del sistema nervoso autonomo alia innervazione dei muscoli volontari. Schweiz. Arch. f. Neurol, u. Psychiat., Bd. 11, S. 60. Abstract in J. Neurol, and Psychopath., vol. 5, p. 76, 1924-5.

. 1931. II simpatico cervicale degli Amnioti (ricerche dimorfologia comparativa).

Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. und Entwicklungsgesch., Bd. 96, S. 289.

Testut, L. 1911. Traite d’anatomie humaine. 0. Doin, Paris. Later edition, 192123.

Thanhopfer, L. 1892. Ueber die Nervenendigung der quergestreiften Muskelfasern und uber Re- und Degeneration derselben im lebenden Korper. Anat. Anz., Bd. 7, S. 635.

Thurlow, M. DeG. 1917. Quantitative studies on mitachondria in nerve cells. Contrib. Embryol., Publ. Carnegie Inst. Washington, vol. 6, no. 16, p. 35.

Tiegs, 0. W. 1927. Australian J. Exper. Biol, and Med. Sc., vol. 4, p. 25. (Quoted from Tiegs ’31.)

. 1931. A study of the neurofibril structure of the nerve cell. J. Comp. Neurol.,

vol. 52, p. 189.

Tilney, F. 1927. A glandular outgrowth from the roof of the oblongata in Amia calva. J. Comp. Neurol., vol. 43, p. 433.

Tretjakofp, D. 1902. Zur Frage der Nerven der Haut. Zeitschr. f. wissensch. Zool., Bd. 71, S. 625.

. 1908. Das Nervensystem von Ammocoetes, I. Das Ruckenmark. Arch.

f. mikr. Anat., Bd. 73, S. 607.

. 1909. Das Nervensystem von Ammocoetes. II. Gehirn. Arch. f. mikr.

Anat., Bd. 75, S. 636.

Tschiriew, S. 1879. Sur les terminaisons nerveuses dans les muscles strips. Arch, de Physiol., 2“' S5rie, vol. 6, p. 89.

. 1879a. Etudes sur la physiologie des nerfs des muscles stries. Arch, de Physiol.,

2““ S^rie, vol. 6, p. 295.

Turner, J. 1906. A study of the minute structure of the olfactory lobe and cornu ammonis, as revealed by the pseudovital method. (With remarks on the plan of nervous structure of vertebrates in general.) Brain, vol. 29, pp. 113 and 57.

Unna, P. G. 1916. Ueber Sauerstoffsorte und Reduktionsorte. Eine histochemische Studie. Arch. f. mikr. Anat., Bd. 87, S. 96.

Vater. 1741 Dissertatio de consensu partium corporis humani. Vitemburgae.

VAN DER Velde, E. 1907. Die fibrillare Struktur in den Nervenendorganen der Vogel und der Saugetiere. Anat. Anz., Bd. 31, S. 621.

. 1909. Die fibrillare Struktur der Nervenendorganen. Internat. Monatschr. f.

Anat. u. Physiol., Bd. 26, S. 225. ...

Veratti, E. 1900. Su alcune particolarita di struttura dei centri acustici nei mam miferi. Pavia.

132 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Veraiti, E. 1900a. Recherche sul sistema nervoso del Limax. Mem. d. R. 1st. Lombardo di sc. e lett., vol. 18.

Verworn, M. 1889. Die polare Erregung der Protisten durch den galvanischen Strom. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 45, S. 1.

. 1890. Idem. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 46, S. 267.

. 1896. Untersuchungen iiber die polare Erregung der lebendigen Substanz durch

den constanten Strom. III. Mitteilung. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 62, S. 415. IV. Mitteilung. Ibid., Bd. 65, S. 47.

. 1907. Die Vorgange in den Elementen des Nervensystems. Zeitschr. f. allg.

Physiol., Bd. 6, Sammelreferate, S. 11.

. 1908. Bemerkungen zum heutigen Stand der Neuronlehre. Med. Klin., Bd. 4,

S. 111.

ViGNAL, W. 1889. Ddveloppement des dldments du systSme nerveu.x cdrdbro-spinal. G. Masson, Paris.

ViTALi, G. 1906. Contributo alio studio istologico dell’unghia. Le espansioni nervose nel derma sottoungueale dell’uomo. Internat. Monatschr. f. Anat. u. Physiol., Bd. 23, S. 239.

VooRHOEVE, H. C. 1926. De nuclealreadie van Feulgen an Rossenbeck en haar waarde voor de histologie. Dissertatie, Amsterdam.

Wagner, W. 1852. Ueber das Vorhandensein bisher unbekannter eigenthiimlicher Tastkorperchen, usw. Nachr. v. d. k. Gesellsch. d. Wissensch. u. d. Georg.-Aug. Univ., Gottingen.

Waldeyer, W. 1891. Ueber einige neuere Forschungen im Gebiete der Anatomic des Centralnervensystems. Deutsche med. Wochenschr., Bd. 17, S. 1213.

Waller, A. 1852. Sur la reproduction des nerfs et sur la structure et les fonctions des ganglions spinaux. Arch. f. Anat., Physiol, u. wissensch. Med. (Muller’s), S. 392. Warburg, 0. 1914. Ueber die Rolle des Eisens in der Atmung des Seeigcleis nebst

Bemerkung iiber einiges durch Eisenbeschleunegte O.xydationen, Zeitschr. 1. physiol. Chem., Bd. 92, S. 231.

. 1928. Ueber die katalytischen Wirkungen der lebenden Substanz. J. Springer,

Berlin.

Weed, L. H. 1914. Studies on cerebro-spinal fluid. No. II. The theories of drainage of cerebro-spinal fluid with an analysis of the methods of investigation. J. Med. Research, vol, 31 (N. S., vol. 26), p. 21.

. 1914a. Idem. No. III. The pathways of escape from the subarachnoid spaces

with particular reference to the arachnoid villi. J. Med. Rc.search, vol. 31 (N. S., vol. 26), p. 51.

. 1914b. Idem. No. IV. The dual source of cerebro-spinal fluid. J. M^-'d.

Research, vol. 31 (N. S., vol. 26), p. 93.

. 1916. The establishmeni of the circulation of the cerebro-spinal fluid. Anat.

Rec., vol. 10, p. 256.

. 1916a. The formation of the cranial subarachnoid spaces. Anat. Rec., vol. 10,

p. 475.

. 1917. The development of the cerebro-spinal spaces in pig and man. Conlrib.

Kmbryol., Publ. Carnegie Inst., Washington, vol. 5, no. 14, p. 1.

. 1917a. .\n anatomical con.sideration of the cerebro-spinal fluid. .\nat. Rec.,

vol. 22, (). 161.

1920. The experimental production of an internal hydroccplialu.s. Contrib. Kmbryol., Publ. Carnegie Inst., Washington, vol. 9, no. 27, p. 425.

. 1923. The ab.-orption of cerebrospinal fluid into the venous system. Ain. J.

.\nat . vol 31, p. 191.

— . 1927. Problems of the relation of muscle and nerve. Bull. John.'j Ilopkios

llo'pital, vol. 10, p. 117.

EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS 133

Weed, L. H., and McKibben, P. S. 1919. Pressure changes in the cerebrospinal fluid following intravenous injections of solutions of various concentrations. Am. J. Physiol., vol. 48, p. 512.

Wegefabth, P. 1914. Studies on cerebro-spinal fluid. No. V. The drainage of intra-ocular fluids. J. Med. Research, vol. 31 (N. S., vol. 26), p. 119.

. 1914. Idem. No. VI. The establishment of drainage of intra-ocular and

intracranial fluids into the venous system. J. Med. Research, vol. 31 (N. S., vol. 26), p. 149.

Wegefabth, P., and Weed, L. H. 1914. Idem. No. VII. The analogous processes of the cerebral and ocular fluids. J. Med. Research, vol. 31 (N. S., vol. 26), p. 167.

Weigert, Carl. 1895. Beitrage zur Kenntnis der normalen menschlichen Neuroglia. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main.

Weil, A. 1929. The influence of formalin fixation on the lipoids of the central nervous system. J. Biol. Chem., vol. 83, no. 3, p. 601.

Weiss, P. 1924. Die Funktion transplantierter Amphibien e.xtremitiiten. Aufstellung einer Resonanztheorie der motorischen Nerventatigkeit auf Grand abgestimmter Endorgane. Arch. f. mikr. Anat., Bd. 102, S. 635.

. 1926. The relation between central and peripheral coordination. J. Comp.

Neurol., vol. 40, p. 241.

WiEJiAN, H. L. 1925. Heteroplastic grafts of the anterior limb of the cord in Amblystoma embryos. Science, vol. 61, p. 422.

. 1926. The effect of heteroplastic grafts of the spinal cord on the development of

the limb of Amblystoma. J. Exper. Zool., vol. 45, p. 335.

Wilkinson, H. J. 1929. The innervation of striated muscle. Australian Med. J., vol. 2, p. 768.

. 1930. Experimental studies on the innervation of striated muscle. J. Comp.

Neurol., vol. 51, p. 129.

. 1930a. Observations on the sole plates of motor endorgans. J. Comp. Neurol.,

vol. 50, p. 133.

Wilkinson, H. J., AND Burkitt, A. 1926. Nerve endings in adipose tissue. Australian Med. J., p. 3.

Wilson, C. W. 1916. On the life history of a soil Amoeba. Univ. California Publ., Zool., vol. 16, p. 241.

WiNDLE, W. F. 1926. Non-bifurcating nerve fibers of the trigeminal nerve. J. Comp. Neurol., vol. 40, p. 229.

WiNKLER-JuNiHS, E. 1917. The development of the protoplasm in the reactive neuroglia cell. Psychiat. en neurol. BL, Amsterdam, Bd. 21, S. 93.

. 1918. Neurogliaveranderingen bij carcinoommetastasen in de hersenschors.

Psychiat. en neurol. Bl., Amsterdam, Bd. 22, S. 175.

■ . 1919. Het epicerebrale, perivasculaire lymphstelsel van His ten opzichte van

de marginale en perivasculaire neuroglia. Psychiat. en neurol. BL, Amsterdam,

Bd. 23, S. 172. ,

Winslow, C.-E. A., AND Fleeson, E. H. 1926. The influence of electrolytes upon the electrophoretic migration of bacteria and of yeast cells. J. Gen. Physiol., vo . ,

P- 195- . . . . . •

WiNTREBERT, M. P. 1904. Sur I’existence d’une irritabihte excitomotrice pnmitive, ind^pendante des voies nerveuses chez les embryons cili^s de Batraciens. omp

rend. Soc. de biol., vol. 57, p. 645. _

Wislocki, G. B. 1928. The cytology of the cerebrospinal pathway.

sect. 31 (edited by E. V. Cowdry), Hoeber, New York , . , . _ Wislocki, G. B., and Putnam, T. J. 1921. Absorption from the ventricles in experimentally produced internal hydrocephalus. Am. J. Anat., vo . , p.

Special cytology.

134 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Wlassak, R. 1898. Die Herkunft des Myelins. Ein Beitrag zur Physiologie des nervosen Stutzgewebes. Zeitschr. f. wissensch. Biol. ; Abt. D., Arch. f. Entwicklungsmechn. d. Organ., Bd. 6, S. 453.

Wolff, M. 1903. Ueber die Kontinuitat des perifibrillaren Neuroplasmas (Hyaloplasma, Leydig-Nansen). Anat. Anz., Bd. 23, S. 20.

. 1904. Das Nervensystem der polypoiden Hydrozoa und Scyphozoa. Zeitschr. f.

allg. Physiol., Bd. 3, S. 191.

. 1905. Neue Beitrage zur Kenntnis des Neurons. Biol. Zentralbl., Bd. 25,

S. 691 u. 729.

WooLLARD, H. H. 1926. The innervation of the heart. J. Anat., vol. 55, p. 345.

. 1926a. The innervation of blood vessels. Heart, vol. 13, p. 319.

. 1927. The innervation of voluntary muscle. Compt. rend. A. de anat., vol. 22,

p. 256.

Weeden, J. 1905. Die Nervenendungen in der harten Himhaut des Riickenmarks von Saugetieren. Arch. f. mikr. Anat., Bd. 66, S. 128.

Yocum, H. B. 1918. The neuromotor apparatus of Euplotes patella. TJniv. California Publ., Zool., vol. 18, p. 337.

Yoshimura, K. 1910. Das histochemische Verhalten des menschlichen Plexus chorioideus, zugleich ein Beitrag zur Frage der Plexus-Sekretion. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 18, S. 1.

Yuien, K. 1928. On the direction of the lymphatic stream in the nerve. Folia anat. japon., Bd. 6, S. 301.

Zand, N. 1930. La microglia et les histiocytes. Ann. Anat. Path, e Anat. Normale Med.-Chir., vol. 7, p. 565.

Zanda, L. 1889. Ueber die Entwicklung der Osteome der Arachnoidea spinalis. Beitr. z. path. Anat. u. z. allg. Path. (Ziegler's), Bd. 5, S. 391.

Ziegler, P. 1896. Ueber die Mechanik des normalen und pathologischen Hirndruckes. Arch. f. klin. Chir., Bd. 53, S. 74.

ZwAARDEMAKER, H. J. 1919. Ueber die Bedeutung des Kaliums im Organismus. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 173, S. 28.

Chapter II The Comparative Anatomy Of The Spinal Cord

The following pages contain a survey of the structure of the central nervous system of vertebrates including man. In order to make as clear as possible the principal homologies and the different variations and to give an insight into the progressive evolution of each part, the material has been arranged under the major subdivisions of the nervous system and a comparative description of each subdivision has been presented separately. For various reasons the spinal cord is considered first. It is the only subdivision which keeps a segmental character. In most animals, to a noteworthy degree, it is relatively independent of the other parts of the nervous system and has not been modified so extensively as have they in the course of evolution.

Foreshadowings of a spinal cord system are to be found in some invertebrate types. Thus in Enteropneusta a part of the central nervous system arises as a medullary tube from the dorsal body ectoderm, and in certain larvae of tunicates (Ascidia) a large part of the nervous system, lying on a chorda dorsalis, resembles a spinal cord. However, in its structure it is suggestive of only the most primitive stages of the embryonic cord of mammals, although, in Appendicularia, it is said to give off metameric roots. The interrelation existing between vertebrate and higher invertebrate types is still a subject of controversy. The matter cannot be considered here ; those interested in this phase of the subject should refer to the original papers or to the short compendium “Evolution of the Nervous System in Invertebrates, Vertebrates, and Man” {Ariens Kappers, ’29; Bohn, Haarlem), where the relations in invertebrates are sketched briefly. The present account opens with a description of the brain and spinal cord of Amphioxus, since this animal, classed as a chordate and not a true vertebrate, may be considered as introducing the relations present in vertebrates.

The Brain and Spinal Cord op Ajiphioxus

A short review of the whole central nervous system of this partly primitive and partly regressive animal precedes the more detailed account of the cord. The central nervous system in Amphioxus arises as an ectodermal medullary fold. The elosure of this fold proceeds in a frontal direction, the connection with the frontal ectoderm remaining an open cleft, the neuroporus anterior. In this central nervous system, a frontal part with a rather large ventricle (fig. 53) may be distinguished from a caudal part, the lumen of which is considerably narrower and which has the form of a vertical cleft. Generally, the frontal part is considered the brain (archencephalon, vo?i Kupffer, ’06) and the caudal part the

135

136 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

spinal cord of the animal. The ventricle of the brain is continuous through the neuropore with a frontal groove, Kdlliker’s olfactory groove (von Kdlliker, ’43 ; Edinger, ’06), from which was believed to arise an unpaired nerve (complete monorrhiny).^ However, recently the olfactory character of this groove and the presence of an olfactory nerve have been denied by Franz (’23).^

From the ventral side of the cephalic end of the brain arises a paired nerve which is known by various names, as, for example, nervus terminalis (fig. 53),

'n.ierm. -p-igm. inf. or^ cenfr.

Fig. S3. Sagittal section to one side of the midline through the anterior end of the central nervous system of Amphioxus laneeolatus.

nervus apicis (van Wijhe, ’14), nervus septalis I, nervus spinalis I, or nervus I (various authors). At the anterior end of the brain, dorsal to the nerve just mentioned, are pigment granules which form a pigment spot to which the older Retzius appears to have called attention first in a letter (Muller, 1839, quoted from Franz, ’23). This pigment spot may be doubled (Muller and others so regarded it) but it is variable, and Franz (’23) stated that usually it is unpaired. Earlier observers regarded this as an eye spot but later work has disproved this interpretation. Franz believed himself to have demonstrated that the nerve fibers traced to it by Edinger (’06) are bundles of the first nerve (nervus terminalis, mentioned above) and that they can be traced forward. He arrived at the conclusion, then, that the pigment spot is not a sense organ (as have also Slieda, ’73 ; Niisslin, ’77 ; Rohde, ’88, and others).

Caudal to the entrance of the first nerve, a groove is seen on the ventral brain wall. This groove is provided with a ciliated epithelium, sensory in type, which has been described with great minuteness by Boeke (’08 ; see also ’02, ’02b), who regarded it as showing great similarity to the sensory epithelium occurring in the

' The term “ complete monorrhiny ” is used where one olfactory groove and one olfactory nerve are present; incomplete monorrhiny implies the presence of one olfactory groove giving rise to two olfactory nerves (adult Petromyzon). The groove in the adult lamprey arises embryologically from two independent grooves to which a third, unpaired one is added (Ariens Kappers).

2 “Cleiszelgrube am Neuroporusrest kein Sinnesorgan. Kein Riechnerv (mit Edinger)."

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 137

infundibular region (saccus vasculosus) of fishes. Thus considered, it offers an excellent “point de repere” for the homologizing of parts lying in front of and behind it. Presumably its presence would indicate diencephalic regions with the telencephahc region represented by the area immediately in front of it, including the place of entrance of the nervus terminalis. The region caudad to the infundibular organ {Boeke, ’02, ’08, ’13) would be, in a general way at least, the homologue of the midbrain base, the medulla oblongata, and the spinal cord, although there might be some question about the use of the term midbrain in the strict sense of the word. The difficulty in recognizing this last region is due to the absence of those structures which usually are regarded as criteria of the area, for Amphioxus does not have true eyes comparable to those found in craniotes, and consequently optic, trochlear, and oculomotor nerves are lacking.

However, not all observers are agreed as to the homologizing of these regions. Delsman (’13 ; see also ’13a and ’18) regarded the whole front end of the nervous system of Amphioxus as representative of the vertebrate midbrain and hindbrain and beheved that no homologues of thalamus and forebrain are present. He based his conclusions on the extent of the notochord, which in this animal extends beyond the cephalic end of the brain tube. Franz (’23), in a more recent paper, offered a slightly different set of hypotheses in his suggestion that the clumps of ganglion cells along the first two nerves (often termed the nodules of Quatrefrage) together with scattered ganglion cells of the region are probable representatives of the vertebrate forebrain. Franz was unable to see a great similarity between the cihated epithelium of the ventral groove in Amphioxus and the epithehum of the saccus vasculosus {Boeke, ’13) and stated that the vesicle at the anterior end of the spinal cord of Amphioxus is representative of vertebrate diencephalic and mesencephalic regions. The evidence for the inclusion of mesencephalon does not appear altogether satisfactory. Franz relegated the regions immediately caudal, where the large dorsal cells are found, to the medulla oblongata.

Tur nin g; now to a consideration of the nervous system caudal to the infimdibular region, one of the most evident changes is the almost immediate enlargement of the ventricle. This has been termed the fossa rhomboidalis (or ventriculus quartus) by Hatschek (’92), but until it has been definitely decided as to whether or not the cephalic end of this region is midbrain it appears wiser to use some less specific name — such as ventriculus posterior — for this portion of the ventricular system. Belonging to the region immediately caudal to the ventral groove is the second nerve (the first dorsal root and regarded by earlier observers as the first nerve). This nerve, which, according to Franz (’23) and others, arises close to the tip of the first myotome, has almost symmetrical relations with its fellow of the opposite side. Very frequently it is doubled, that is, it consists of a cephalic and a caudal part, from both of which (but particularly from the cephalic part) dorsal rami are given off. The chief distribution of this nerve is to the rostral fin, which the first nerve also supplies and with which nerve the second nerve anastomoses {Franz). If the region from which this nerve arises be conceded to represent the vertebrate midbrain, then the second nerve of

138 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Atnphioxus is the probable homologue of the dorsal root of the N. opthahmcus profundus V of the embryonic midbrain.

Any attempt to compare the medulla oblongata with that of true vertebrate types presents problems which cannot be solved entirely as yet. In craniotes the peripheral coimections of the medulla are provided for by the branchial nerves (trigeminal, facial, glossopharyngeal, and vago-accessory complex), by the vestibular nerves, and, through aquatic amphibians, by those for the lateral lines. The lateral line nerves and organs of equilibrium innervated by vestibular nerves are not found in Amphioxus, not even a trace of them being present during embryonic life, according to van Wijhe (’13).

This observer (’89, ’93, ’02, ’07, ’14) regarded the left auricular branchial cleft as forming the mouth, and consequently believed that jaws are lacking. His conclusions do not agree with those of Ayers (’21), who did not regard the jaw apparatus as related in any way to the gills, but stated that jaws are present in Amphioxus and that with these are associated muscles and tendons which tie the bases of the jaws together and, at least in the case of the muscles, extend forward to the tips of the bars and on to the tentacular projections. Ayers wrote : "The chondral jaw bars which frame the buccal opening in Amphioxus are the simplest expression among living forms of the vertebrate mandible. But they include much more than the definite mandible, e.g. of cartilaginous fishes. ’’ (For details of the anatomy of this region and of the way in which the apparatus functions, Ayers’ paper should be consulted.) The five nerves lying back of N. II (N. Ill to N. VII) Ayers regarded as associated with the jaw musculature, although he found certain variations ; these then should represent, in general, parts of the trigeminal nerve of craniotes. These five nerves (N. Ill to N, VII) exhibit asymmetrical relations on the two sides, the left nerve innervating the greater part of the velum {Ayers, ’21 ; Franz, ’23) and preoral tentacles of the two sides and the walls of the preoral or buccal cavity {Franz) including the organs connected with it. There is considerable individual variation, the absence of certain visceral branches or the presence of extradorsal roots not being unusual {Ayers, ’21). The details of their distribution cannot be entered into here ; those interested are referred to the papers of Fusari (’89), Dogiel (’03), Kulchin (’13), Ayers (’21), and Franz (’23). It is sufficient to state that on their emergence from behind the respective myotomes, the ventral branches arise almost directly ventralward, where they enter into the formation of inner and outer jaw bar plexuses and a tentacular plexus, which in turn give rise to terminal nerves. In association with these plexuses are larger nerve fibers coursing the length of the jaw bars close to the base of the tentacles and giving off branches to these tentacles.

If these five nerves (N. Ill to N. VII) be regarded as representative of parts of the trigeminal, presumably the cephalic limit of the medulla oblongata will lie in the region between the exits of N. II and N. Ill (counting the nervus terminalis as N. I). The caudal limit is even less easy to determine. In this the external morphology of the area does not prove of much assistance. The posterior ventricle loses its thin roof at the level of entrance of the third dorsal

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 139

root (N. rV). This root, together with the preceding and more than half of the succeeding dorsal roots, belongs to buccovelar and peribranchial regions of the body, and the secondary groupings of these nerves may produce the medulla oblongata nerves of higher forms. In such a case it is self-evident that the opening of the ventricle, the so-called calamus scriptorius, must be shifted considerably caudalward in craniotes. Branchial arches of a sort are present in Amphioxus and the giU clefts lead into a wide chamber between the pharynx and the body wall, which is termed the atrium. Breathing appears to be regulated in large part by the atrial (pterygeal) musculature (van Wijhe, ’14), which is beUeved to receive its innervation from branches of the dorsal roots. This relation suggests, of course, the mi xed ner\'es of the medulla oblongata of higher forms, but if all such nerves are to be regarded as cranial nerves,^ then the lower limit for the medulla oblongata must be placed very far caudalward, since the munber of such nerves is relatively large.

From the foregoing account it is evident that as yet the definite recognition and delimitation of a medulla oblongata segment in Amphioxus is not practicable.

The differentiation within the central nervous system does not afford much aid unless the contention of Rohde (’88) be accepted that the disappearance of the dorsal gangUon cells of Joseph marks the termination of the brain and the beginning of the cord, which latter area is indicated further by the appearance of the most cephalic of the colossal cells (see pp. 143). Consequently, in the following account, which deals with the central nervous system of Amphioxus from the caudal end of the ventral groove (infundibular region) to the caudal end of the cord, no further attempt at differentiation into medulla oblongata and cord will be made.

Behind the posterior ventricle, for most of the remainder of its extent, the central nervous system of Amphioxus has a triangular outline, with the apex pointing dorsalward. The base of the triangle is more or less concave, a consequence of the round form of the chorda dorsalis which presses upon this region of the cord as in cyclostomes (fig. 55). Ventrally, the central canal has an oval lumen which is connected with a smaller enlargement just under the dorsal surface by means of a narrow slit. The outhne of the cord (and probably part of the medulla oblongata) is similar to that found in the embryos of higher animals. In the most caudal regions of the spinal cord no nervous elements occur ; only an ependymal tube is present with large cells so oriented that their inner surfaces are directed backward. In most cases the tube ends in an ampullar enlargement (fig. 54).

’ It has been difficult to obtain exact figures for the number of nerves in Amphioxus. Apparently there are about 64 in all, of which those from the third to the seventh, inclusive (counting nervus terminalis as N. I), belong to the buccovelar region, those from the eighth to about the thirty-ninth to the peribranchial region, and the remainder to the postbranchial region.

Fig. 54. Caudal ampullar enlargement of the end of the spinal cord of Amphioxus. Retzius.

140

nervous systems of vertebrates and of man

showlrpisL:! n. m

are due to secondary changes Whprp«« • ^ character while others

ventral roots on hoi siTs pass Tat the

and ventral roots alternate. This is a primitiw ^ptioxus the dorsal

of a sensory root correspondinff orioiTin^l f entrance

which it has traveled from tlie skin wS th septum through

the middle of the myotor^e whichTt supplies T^T? sity, alternate, mereaa tte .t dfrs'a “n‘'

prtmrtrve feature of this autaal, the alternation between th^TS SdoL”

and between the right and left ventral rnnta • i

Amphioxus, is a secondary feature due to a shifting of the lefSdetfTh respect to the right side. Since this secondary shifting cotem lest ^ myotome, a left ventral root is found but rarelv in the^D J! ? ^alf of a dorsal root although indications of such a condition may occur (fig.ter ""

Ayers ( 21a) studied carefully the motor roots in Amphioxus Wp in i f la composed of from 100 to 250 fibers, the number deeding upon^P fT myotomes supplied. Four somatic efferent branch^ are preLnt n t t

teltnyntfolhtl

.?thtva1*r:“t™.““*’“‘®^-^^^ AffithbranchisCrE:;?

The roots proceed from alternate right and left thickenings of the central nervous system m which regions, according to Retzius (’91), highly interwoven fibrillar networks occur, due in part to the tendency of the roots to curve outward Their cells of origin are not known ivith certainty at the present time, although It IS bought that medium-sized cells on the upper or lateral border of the cen n canal may give rise to them (Wolff, ’07; Johnston, ’05; Stendell, ’14a) Such cells send their dendrites into the dorsal funiculus, but in many cases it has not

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 141

Epidermis Myotomo

Med. spin.

been possible to trace their neuraxes into the ventral root, although Wolff (’07 ; fig. 2, p. 197) showed such a relation. It is possible that in Amphioxus, as in cyclostomes, the roots emerge at a level other than that of their cells of origin, hence the difficulty in tracing them. Certainly the observations of Franz (’23) indicate this.

However, Wolff (’07 ; fig. 2, p. 197) was of the opinion that certain cells lying below the central canal also send their neuraxes into ventral roots and, in each case, a large dendrite into the dorsal root of the other side, thus forming a most primitive, crossed reflex arc. Schneider f’OS) and Franz (’23) believed that they could trace the motor fibers from stellate cells lying underneath the central canal.'*

That the dorsal roots emerge alternately on the right and left sides is e-vddent from the longitudinal section in figure 56, where also the alternation of the myotomes can be seen.

Two features of the dorsal roots must be emphasized particularly. The first is that they are not connected with the ventral roots ; the second is that, while they do not possess definite cell aggregates in the shape of spinal ganglia, both extra- and intramedullary ganglion cells are present. The first point is easily confirmed ; the second has been the subject of much controversy. In the spinal cord itself, Relzius (’91) found cells, the peripheral offshoots of which run into the dorsal roots. He thought that they must be

Fig. 56. Horizontal section through the spinal cord of Amphioxus to demonstrate the alternating and intermyotomic course of the dorsal roots.

Boeke (’08a) described both motor and neuromuscular terminations in the myotomic muscle of Amphioxus. Van Wijhe (Verslag der Ned. Dierk. Vereeniging 1893, quoted from Boeke, ’08a) believed that endings of higher forms, similar to neurotendinous endings, can be demonstrated in the ligamentum intermusculare of Amphioxus. Dogiel (’03) thought that certain fine fibers in the ventral root might be sensory fibers and Boeke believed that the endings described by him might be the terminations of such fibers {Boeke, ’08a, pp. 286 and 288). That ventral roots may cariy such sensory fibers for muscular sensibility has not been confirmed generally but in principle is quite possible, since — as we shall see in Chapter V — sensory fibers in motor nerves are known to occur in mammals in the eye muscle nerves, in the motor root of the trigeminal, and probably in the facial and hypoglossal nerves. Ayers (’21a) observed fibers of the dorsal root which run in the muscle septa and which apparently carry muscle sense.

142 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

considered as spinal ganglion cells which, instead of lying outside the spinal cord, remain within it. Heymans and van der Siricht (’98) confirmed Retzius’ suggestion that dorsal root fibers arise from intramedullary cells. J ohnston (’05) also found such cells within the central nervous system but emphasized the presence of similar cells in the course of the dorsal nerve roots themselves. Rohde (’88), Dogiel (’03), and others have described extramedullary neurons which they regarded as forerunners of sensory ganglion cells. The evidence at present indicates that both types are present. Kutchin (’13) quoted Johnston’s observations, apparently agreeing with them. These intramedullary spinal (and cranial) ganglion cells ^ are arranged around the fibers and converge toward them in such a way as to suggest that the cells are migrating into the roots. The outward lateral curve of the longitudinal fibers described hy Johnston and others gives

— - further support to this view.

However, among later observers there are still those who ques^ certain of these observations.

Franz (’23) recognized gan glion cells along the course of the j I j j first two nerves but was unable to

J/\ 1 identify such cells along the other

' nerves. He stated (p. 453) that

Fia. 57. Dorsal roots with the intramedullary root cells mistook glia cells for ganglion

m Amphioxus. Johnston. ,, , ,, , , , ? . ,

cells and that a study of a rich

collection of material did not document Dogiel’s findings. It is evident, then, that the spinal ganglia, which in higher vertebrates arise from the extramedullary neural crests, in Amphioxus still have their origin in the medullary tube itself. Moreover, in the adult animal, these cells retain in part this primitive position and in part migrate toward the periphery under the influence of neurobiotaxis.

The neuraxes of the dorsal root neurons either ascend or descend, or do both, after a dichotomous division. Some cross the midline. A great number of fibers end at the level of entrance of the root. The dorsal root contains both coarse and fine fibers, the former possibly somatic afferent and the latter visceral afferent. Such a conclusion is in accord with their position in the cord, since J ohnston (’05) found that the latter form a bundle in the very center of the spinal cord, medioventral to the former, an arrangement characteristic of these components in higher forms. Another difference between the two bundles is to be found in the fact that visceral afferent fibers do not divide dichotomously nor do they send a larger branch in an ascending direction. In addition to the afferent or sensory elements, the dorsal roots also contain efferent fibers for the viscera

‘ In teleosts, sensory spinal ganglion cells are found in the cord. In vertebrates, even in man, a part of the sensory trigeminal roots preserves this peculiarity, the ganglion cells of the mesencephalic root of the trigeminal remaining in the roof of the midbrain. The presence of intramedullary sensory ganglion cells in Amphioxus is not an aberrant but a primitive condition, which has been altered secondarily through neurobiotactic influences (which lead to the migration of the cell in the direction of the stimulus).

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 143

(visceral efferent fibers), the cells of origin for which are located lateral to the ventral part of the central canal.

After reaching the body wall, each dorsal root divides into two branches, a smaller dorsal and a larger ventral branch. Both enter the skin of the animal, where they end in free intradermal terminations. In addition to these, a twig from the ventral branch runs to the deeper parts of the body carrying the visceral afferent and visceral efferent components ; this twig is lacking in the most caudal nerves. According to a number of observers (Fusari, ’89 ; ran. Wijhe, ’93 ; Dogiel, ’03 ; Kutchin, ’13 ; Boeke, ’35, Quart. J. Microsc. Sc., vol. 77) the visceral efferent fibers ® form considerable plexuses, e.g. in the region of the atrium. In these plexuses multipolar cells occur, so that it is very probable that, in addition to the direct visceral motor fibers for the striated muscles of the atrium and the M. M. transversi, preganglionic and postganglionic fibers are present.

Turning now to a consideration of the cells within the central nervous system, the so-called cells of Joseph will be described first. These cells, described in detail by Joseph (’04), had been seen by earlier observers {Owsjannikow, ’68; Rohde, ’88). Franz (’23) termed them “Gehirnzellen” and found them distributed from slightly in front of a transverse plane through N. II to a similar plane through the left dorsal sixth root, the greatest number of the cells being found in planes between the ventral groove (or infundibular region) and the fifth dorsal root. In arrangement, these cells show a tendency toward symmetry and have been described as resembling very closely the unpigmented sensory cells of the eye cups described by Hesse (’98), although they have a dorsal rather than the lateral or ventral position characteristic of the latter structures. On this ground, various observers have regarded them as capable of perceiving light, but the experiments of Parker (’08) have not tended to confirm this view. Moreover, Franz (’23) has stated that the structure of these cells does not indicate such a light-perceiving function. He found them difficult to stain, but in his preparations they appeared as multipolar or, in some cases, bipolar cells with deeply stained nuclei and somewhat striated protoplasm foUorving certain fixatives. He was unable to distinguish the neuraxis from the dendrites. He found that, in general, the processes tend to extend longitudinally, some passing cephalad and others caudad. He discussed the possibility of their contributing fibers to the first and second dorsal nerves or of their functioning as association fibers, but was unable to reach any definite conclusions. Boeke (’08a) regarded these cells as concerned with equilibrium. They have no known homologue in craniotes.

Prominent among the secondary neurons of the nervous system are the giant or colossal cells of Rohde (’88 ; Boeke, ’08b ; figs. 58, 59, and 60). These cells ’’ occur chiefly in the midline above the ventral enlargement of the central canal,

® Johnston ('05) described their cells of origin as forming a more or less continuous column in the spinal cord, thus resembling the visceral efferent cells of fishes. Franz (’23) believed that they lie dorsal to his cells of origin of the ventral roots.

^They have nothing whatever to do with the large, so-called light-perceiving cells of Joseph (’04) which are found in the roof of the posterior ventricle. The giant ganglion cells lie behind the posterior limit of the cells of Joseph (’04).

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 145

first in the direction of the entering impulse, with the base of the cell directed toward the entering root fibers. Its initial course thus corresponds with the direction of the stimulus that enters it, a fact which is in perfect harmony with the doctrine of the stimulogenous formation of the neuraxis (stimulogenous fibrillation of Bok, '15 and ’15a, see Chapter I). Having arrived at the periphery of the spinal cord, the neuraxis bends underneath the central canal, thus reaching the other side, and then courses in a longitudinal direction (figs. 59 and 60).

During the first part of its course, the fiber gives off collaterals. These may divide dichotomously, the larger branch running in the same direction as the primary neuraxis. That the neuraxes of the anterior group of giant cells send their processes caudalward is in accord with the fact that most reflexes in Amphioxus run aborally. The numerous frontally coursing neuraxes are explainable on the ground that skin sensibiUty decreases from the head to the tail region and many tail reflexes are elaborated orally.® These neurons have been interpreted in various ways. It appears probable that they should be compared with the commissural cells of higher forms, their processes being comparable to the arcuate fibers (Bogenfasern) of His (see also Wolff, ’07). As will be seen later, secondary neurons appear relatively very early in vertebrate embryonic development; the cell bodies of these lie chiefly in the dorsal regions of the cord while their neuraxes cross in the ventral commissure below the central canal. Not only are they among the first elements of the spinal cord to make their appearance, but they resemble the giant cells of Rohde (’88) further in that the caudally located cells send their neuraxes forward while the cephalically situated neurons send their processes caudalward {Bok, ’15, bibliography under Chapter I).

® That responses to stimuli from the tail are made possible by frontal conduction is in accordance with the experiments of Parker (’08), who proved that the motor effect of stimuli applied to a severed hind part is much less than that obtained when the severed forepart or the hindpart of the uninjured animal is stimulated. It follows that the strongest effectory center has a frontal position. This center e.xerts a strong influence even on caudal stimuli. In accordance with these physiological experiments is the fact that each somatic afferent root divides centrally into an ascending and a descending branch, the former being the larger and longer.

Fig. 60. Entrance of sensory root and the radiation of its fibers about a giant cell. The origin of the neuraxis of the giant cell is shown and its course in a stimulo-concurrent direction. Relzius.

146 NERVOUS SYSTEMS OF VERTEBRATES AND OF ALAN

In addition to these crossed reflex paths, there are bipolar cells which send processes caudalward and cephalicward ; also there are short intercalated neurons and other neuron types, the functions of none of which are clearly definable as yet (Retzius, ’91). Between the ependymal cells of the central canal sensory cells have been described by Edinger (’06) and by Siendell (’14a). However, Franz

(’23) was unable to confirm the statements of Edinger and Siendell. The neuraxes of such sensory cells converge, forming, at least in certain cases, a single neuraxis and showing, in general, a striking similarity to similarly located cells in cyclostomes (Tretjakoff, ’09, see bibliography for cyclostomes). Such a fusion into a single process also occurs in the giant nerve fibers (neurochords) of annehds and in Alauthner’s fiber in Ceratodus {Holmgren and van der Horst, '25 ; see bibliography for ganoids and teleosts, Chapter V).

Finally, the peculiar light-perceiving cells of this animal merit a brief description. A number of pigmented ocelli lie ventral to the central canal on both sides and are arranged in a segmental way. The most anterior are at the level of the third muscle segment and consist of two eye cups on each side. Between the fourth segment and the middle of the body the number increases considerably, consisting of groups of twenty-five on a side ; after the middle of the body has been reached there is a considerable decrease, but often there is only a single ocellus per segment or even none toward the tail region (see fig. G3). The above account of the distribution of the pigment cells is based on the account of Hesse (’98). Studying young pelagic larvae, van Wijhe (’93) found two groups of such cells, a cephalic and a caudal group, separated by a region in which are many less pigment cells per segment. The two groups fuse later, however, so that then, on the- whole, the arrangement corresponds to that given by Hesse. The segmental groups are separated by intersegmental spaces ; sometimes in the young animals, the ocelli within a single segment fall into two secondary groups. In their segmental arrangement, the ocelli show an alternation corresponding to that of the segments of the body (the right muscle segment, according to Hesse, being about half a segment caudal to the corresponding left muscle segment). As a result, the anterior right ocellus approaches the position of the second left ocellus. Hesse (’98) .-showed that these ocelli are made up of two cells, a cap-like pigment cell in the concavity of which a large, unpigmented sensory cell is found, the light

I'lo. Gl. Sagittal bcction through the spinal coni of Amphioxim : A, left of the central canal; B, ventral to the central canal. The dificrencci m orientation of the light cells in the difTercnl regions is to bo noted particularly.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 147

perceiving cell (fig. 62). According to Boeke (’02a), occasionally two pigment cells may be present around one eye cup in the pelagic larvae. The eye cups are arranged in such a way that a cup under the central canal and at the right of it has its pigment cap turned ventrally, while cups situated on the canal have the concavities turned dorsally (fig. 61). Those eye cups, particularly the outer

most, situated at the left side of the central canal, look to the right and dorsally, while the eye cups (and particularly the innermost) lying at the right of the central canal are directed toward the right and ventralward. Most of the cells.

consequently, receive their stimulation from the ventral side of the animal, a few from the dorsal side. However, the animal lies mostly on its side, so that the ocelli are directed sidewise in a horizontal plane, having the same direction as does the fimbrial apparatus of the mouth. The orientation of the ocelli thus corresponds to the life habits of the animal. The portions of the light cells adjacent to the pigment cells show a fringe of small rods, similar to that seen in the

light-perceiving cells of worms. Boeke (’02a) Fig. 62. Light cell and its fibrils in found, in that part of the eye cell which lies Amphioxus. Boeke. a and 6, outer

inside the pigment cup, a bean-shaped vitreous emergence of the nerve process, body, and between this body and the nucleus,

a second bean-shaped vitreous structure which he believed to be associated also with visual functions. The neurofibrils described by this observer form a

plexus around the nucleus, then surround the adjacent vitreous body and form a secondary plexus in the region between it and the pigment cup in the region of the second vitreous body (fig. 62). A nerve process is described as emerging from these cells, frequently, but not invariably, on the side opposite the pigment cup. No one, as yet, has been able to trace the course of this process.

The light-perceiving functions of these cells have been proved experimentally by several observers. Parker (’08) showed not only that the spinal cord is sensitive to light throughout the extent of these cells but also that this sensitivity increases or decreases with the increase or decrease of the ocelli. Thus the part

of the spinal cord containing the least number of these cells is the least sensitive to light. Since the skin shows the same structure everywhere, the differences in sensitivity can be explained only on the basis of differences in distribution of the light-perceiving cells.® It is evident that the optic function in these animals is of a very primitive character and that it is restricted to vital functions, the eye cups serving only as regulators of the position of the body with respect to light (that is, they have a photostatic function).

Though discussions involving the physiology of the sense organs do not lie within the scope of the present book, yet the mention of a few of the outstanding results obtained by Parker through physiological methods may not be amiss.

® The skin of this animal is probably not sensitive to light since, according to Parker (’08), only fresh-water animals exhibit this characteristic.

148 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

particularly since his experiments on sensations of temperature, on tactile functions, and on the chemical sense aid in recognizing which sensations are the most primitive. Amphioxus is negatively thermotropic to considerable changes of temperature ; cold only stimulates but heat has both a stimulating and a directive effect. A general tactile sense is present and it is obvious that this sense is more developed on the head (especially on the cirri) than on the tail and is least well developed in the middle regions of the body.’“ Amphioxus is sensitive to acids, to alkalies, and to bitter substances, but not to sweet ones. This chemical sense is best developed on the cirri and tentacles, which, according to Dogiel (’03) and Kutchin (’13), are provided with special sensory cells. Franz (’23) found special sense cells in the tentacles but did not find them continued into nerve processes. Nerve fibers were found to terminate on the sense cells as on the other epithelial cells. The tail ranks second to the head in its sensitivity to chemical agents, while the middle part of the body is the least sensitive to these substances. Similar conditions are found in fishes. In general, the responses of Amphioxus are negative in character, the peripheral nerve endings being mostly of a nocireceptive type. Structurally these endings consist of free arborizations (or networks?) which are not only sensitive to the stimuli mentioned in the foregoing paragraph, but also to pain, and possibly have a gratoreceptive character as well.

The non-nervous substance of the spinal cord apparently consists of large ependymal cells, the peripheral offshoots of which {Nansen, ’85 ; Rohde, ’88 ; von Lenhossek, and von Kolliker) reach the surface of the spinal membrane. Although Franz (’23) occasionally found nuclei in such ependymal cells in their middle portion, true astrocytes and glial ground substance are not present in Amphioxus with the possible exception of certain cells described by Erik Muller (’99), which, according to that observer, lie near the emergence of the roots and send their unstained processes both into these roots and into the cord. It is evident, then, that we are dealing here with a very primitive condition, such as can be found in higher animals only at a very early stage in embryonic life. Muller (’99) showed that these ependymal fibers cover the nerve processes without forming a real autonomous glial tissue independent from the cells lining the central canal.

Intramedullary septa of connective tissue and intramedullary blood vessels do not occur. Meninges are not differentiated ; only a loose tissue is present which may possibly be regarded as no more than the meninx primitiva, but which also includes the endorhachis (see cyclostomes).

The Spinal Cord of CycLOSTO.MES

Compared with that of Amphio.xus, the nervous system of cyclostomes shows a much higher degree of organization. Within this class, the two divisions (the petromyzonts and the myxinoids) have spinal cords which are in most respects identical but brains which show considerable differences in structure.

Probably ciliated celts occurring in the hkin perceive vabrationa of the water wliich in certain higher animals arc perceived by the lateral line organs {Aricna Kappera).

\"*>-’ ■

Fig 63 Distribution of the pigment eyes in the spinal cord of Amphioxus. The specimen, which

“.CT%3£!tr£.ta%:s.rK^^^^

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 151

these levels." Consequently the ventral roots appear between the levels of the dorsal roots (fig. 66). This alternation of the roots, which is a consequence of the intermyotomic course of dorsal sensory roots and the intramyotomic ending of the motor or ventral roots, becomes less marked in the caudal parts of the spinal cord. A further example of the primitive conditions to be found in cyclostomes is seen in the lack of union of the dorsal and ventral roots in petromyzonts ; only in myxinoids do these roots unite to form a mixed spinal nerve.

The conditions found in myxinoids may be illustrated from the accounts given of Polistotrema (Bdellostoma), which has been studied by a number of observers, among whom may be mentioned Ranson and Thompson (’86), Worthington (’05), Allen (’17), and others. Cole (’07, see also ’96 and ’98) has figured the distribution of the nerves in a typical abdominal segment. For the details of the distribution of these spinal cord nerves, the original papers must be consulted. The results may be briefly summarized as follows. According to Ranson and Thompson, in Bdellostoma, at certain levels of the cord, one

dorsal and two ventral roots are given off. From the spinal ganglion situated on the dorsal root, a dorsal sensory ramus and a ventral sensory ramus arise. The two ventral roots, after uniting, send a ventral motor ramus to join the ventral sensory ramus, thus forming a mixed ventral bundle. Branches of the ventral roots unite to form a dorsal motor ramus which distributes independently of the dorsal sensory ramus. The account of the first spinal nerve in Bdellostoma given by Worthington substantiates that just quoted but adds certain details, among which may be mentioned her description of a mixed branch given off soon after the formation of the common ventral trunk, which branch runs dorsalward to enter the dorsally situated trunk musculature. Still further details, such as a division into terminal muscular

“ This point needs control since in animals such as Petroinyzon, where the dorsal and ventral roots do not unite, it is difficult to decide which belong together.

Fig. 66. Alternate emergence of the ventral and dorsal roots in Petromyzon. Johnston (’06).

Fio. 65. The posterior ampullar enlargement of the posterior end of the spinal cord of Petromyzon. Relzius.

152 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and cutaneous branches, are indicated in Cole’s diagram. This last observer united the dorsal motor and sensory rami, a point in which his results differ not only from those previously quoted but also from those of Allen, who has reviewed the question most thoroughly and pointed out that the dorsal motor ramus, which is the more cephalic, terminates before the dorsal sensory ramus has reached its level. Allen has shown further that for some time during embryonic life the motor and sensory roots have separate courses in Polistotrema, and that in the adult “all of the dorsal and aU of the most caudal ventral motor and sensory spinal nerves were distributed in separate motor and sensory rami, while the ventral motor and sensory components of all but the extreme caudal spinal nerves united in forming mixed rami ventrales. ” It will be seen, then, that in this myxinoid there is an interesting transition between the conditions in Petromyzon and those in higher forms.

The two frontal pairs of nerves, which innervate the most frontal division of the ventral trunk muscles of Petromyzon, cannot be homologized with the most frontal spino-occipital nerves of higher fishes {Furbringer, ’97), since they innervate prebranchial myomeres which are lacking in the latter forms {Biitschli, ’12). Following these are the occipito-spinal nerves, i.e. such nerves as are taken up into the skull and assimilated by the hypoglossal nerve in higher vertebrates (see Chapter V). After these follow the true spinal nerves. However, all of these nerves emerge outside of the skull (the palaeo-cranium) in Petromyzon, and all show practically the same structure.

The cells of origin of a ventral root scarcely ever are found at its level of emergence. Before they leave the cord, the neuraxes of such neurons course for a considerable distance, usually in a caudal but sometimes in a cranial direction, giving off in either case intramedullary collaterals. However, Tretjakoff (’09) found in Ammocoetes that customarily a motor fiber gives off no collaterals if it traverses only a short course from the cell to the root but may give off collaterals ending as branched processes if it is long. It may be that the motor roots are primitively collaterals of intramedullary fibers (cf. fig. 83, the larvae of amphibians), the cells of origin of which lie mostly at frontal levels. Another factor determining the course of the motor fibers is to be found in the fact that the sensory roots, stimuli over which govern the topography of the motor cells, do not lie at the same level as do the motor roots.

The motor cells, wliich in Amphioxus seem to lie near the central canal only, have migrated laterally in cyclostomes, where distinct wing-like cornua of the gray substance occur. However, it is not possible to make a division between ventral and dorsal horns, only one horn being present on each side (figs. 64 and OS). The motor cells are found in the lateral parts of these wings and vary somewliat in general configuration. Frequently the motor neurons send their processes at right angles to the longitudinal axis of the cord, and such processes tend to break up in a marginal plexus {Rclzius, '91 ; Tretjakoff, ’09** ; see also figs. 67A and B). The motor neurons were divided by Tretjakoff into three types, depend TrcljahoiJ’ (’09) Il.^crilK■d thi.s ending of the dendrites at the j)eriphcry of the .spinal cord to conditiun-' of nutrition which are beat at the atirface, aince the spinal cord of thi.i animal contaiiw

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 153

ing upon the spread of the dendrites. Type I has dendrites which spread out almost exclusively at the level of the cell body, type II, dendrites which extend short distances longitudinally before breaking up, while type III is a neuron having dendrites which may extend for considerable distances before breaking up into marginal plexuses. A part of these dendrites are in contact with the longitudinal fibers of the spinal cord. Many of them, however, split up in relation with the external limiting membrane.

Of those dendrites associated with fiber bimdles, the greatest munber, according to Tretjakoff, pass out into the dorsolateral and ventrolateral columns of the cord on the same side on which the cell body is located. Others break up around Muller’s fibers which arise in the vestibular region of the medulla oblongata. From other motor cells, dendrites pass to the opposite side of the cord, spreading out there, in general in the same transverse level. Usually these last are branches of a medial dendrite and cross either in the commissura grisea ventralis or the commissura dorsalis ; occasionally the decussating branch is from a lateral dendrite. The transverse position of the dendrites is correlated with the longitudinal course of the unmedullated fiber bundles among which they run. Their close connection with the external limiting membrane is for nutritive pmposes. The number of dendrites varies with the mass of the cell body.

Certain points with regard to the sympathetic innervation of cyclostomes appear to be still matters of dispute. The position within the cord of the cells of origin of the preganglionic fibers has not been determined as yet. It is probable that, in addition to the somatic efferent fibers for skeletal muscle, the ventral roots contain some visceral efferent fibers, although preganglionics are found also, as in Amphioxus, in the dorsal roots (Freud, ’77 ; Tretjakoff, ’09). There appears

no blood vessels. However, this does not explain the transverse position of the dendrites, although it may contribute to their peripheral ending. Furthermore, in Myxine, where intraspinal blood vessels do occur, these dendrites have a perpendicular position. The Purkinje cells in higher animals offer an example of a similar condition, for they branch in one plane, the dendrites being perpendicular to the longitudinally-running, parallel fibers of the molecular layer. In this case, no pial blood vessels exert an influence on the orientation of the processes, since the molecular layer itself contains many blood vessels, and the outgrowth and peculiar topographic arrangement of the dendrites of the Purkinje cells only occur 10 days after birth (Addison, T1 ; see bibliography for mammalian cerebellum) and apparently for fimctional reasons.

Fig. 67. Superficial ramifications of the dendrites of the large spinal cord cells of Ammocoetes, after Tretjakoff.

A. Longitudinal section through the spinal cord.

B. Cross section through the spinal cord; the marginal dendritic plexus and the network of dendrites around Mailer’s fibers (at right) are to be noted particularly.

154 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

to be general accord that no ganglionated sympathetic chain such as occurs in higher vertebrates is present in cyclostomes. Johnston (’02), however, described a bundle of fibers situated in the gill region as a sympathetic trunk, and thought that they supplied blood vessels. Subcutaneous ganglion cells had been described in these forms by Relzius (’90), and Johnston found many such ganglion cells, particularly along cranial nerves, such as the facial. He suggested that these might be sympathetic ganglion cells. In this connection, it is of interest that Allen (’17) regarded the cells scattered along the course of certain of the dorsal and ventral mixed rami — from their bipolar character and their preponderance in the dorsal rather than in the ventral rami — as outwandering dorsal root or

Dorsoi ftenftory cells Dors (umculufl

Fig 68 Cross section through the spinal cord of Petromyzon,

sensory rather than sympathetic ganglion cells. Allen suggested that the sympathetic system may arise in higher vertebrates either as an independent development or possibly as some type of modification of this early migratory process. Apparently in line with Allen’s suggestion is the work of Brandt (’22), who believed that Myxine has no sympathetic innervation other than that associated with the vagus. In contrast with such findings is the early published work of Juhn (’87), who found, associated with the heart and the intestinal tract, plexuses of fibers which contained peripheral ganglion cells (ganglions S3Tnpathiques profonds). Both vagus and spinal nerves were in relation with these plexuses, but the latter nerves reached them through paired collateral ganglia (ganglions superficiel of Julin, ’87) situated at either side of the aorta. He found these ganglia related to both dorsal and ventral spinal nerves and suggested this as evidence that both afferent (visceral sensory of present terminology) and efferent fibers were represented. Obviously the sympathetic innervation in cyclostomes offers a field for further investigation.

The dorsal roots and their cells of origin have been objects of much controversy. Their sensory fibers, the peripheral branches of which end freely in the skin {Relzius, ’92), originate for the greater part from extramedullary cells which

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 155

(unlike those of Amphioxus) form true ganglia. These ganglia consist of spindleshaped enlargements on the dorsal roots (fig. 66), formed mostly of bipolar cells. There is considerable question as to whether or not intramedullary spinal ganglion cells are present. Beccari (’09) was of the opinion that such cells are represented by the dorsal cells or “ Hinterzellen” of the cord. In a single preparation, Kolmer (’05) was able to trace a process of a dorsal cell into the dorsal root. He did not regard such a relation as the rule and stated that the homology of this multipolar cell with the spinal ganglion cell is very doubtful. Tretjakoff (’09) denied completely that these cells are sensory root cells. Beccari’ s opinion that such cells are intramedullary ganglion cells, comparable to those found in Amphi

Fia. 69. Paramedian sagittal section through the cervical cord of Petromyzon. S. D. C., dorsal sensory cells.

oxus, receives further support from the resemblance of these cells to intramedullary and supramedullary root cells of higher fishes (see figs. 68 and 81).*^ In the spinal cord of Petromyzon the dorsal cells are arranged in two symmetrical rows (see figs. 68 and 69) ; their processes form the dorsomedial bundle of the cord. Their dendrites, which lie in the dorsal root, constitute about one-fifth of that bundle.

Most of the dorsal root fibers, with the exception of the visceral efferents, arise as processes of extramedullary ganglion cells and enter the dorsal and dorsolateral funiculi of the cord. The coarser fibers dichotomize {Nansen, ’85) ; the finer fibers timn in a longitudinal direction without dividing. The former may be somatic afferent and the latter visceral afferent fibers. The dichotomizing fibers take up a position close to the dorsomedial bundle arising from the intramedullary ganglion cells and, according to Tretjakoff (’09), send collaterals to the periphery of the cord, where they come into synaptic relations with the dendrites of motor and intercalary cells. The non-dichotomizing fibers run in the dorsolateral

“ It is hardly necessary to say that these sensory intramedullary cells have nothing to do with the so-called cells of von Lfenhossfik which occur in the cervical region in birds and reptiles and which send neuraxes out through the posterior roots ; the latter are large and multipolar motor cells with the cell bodies situated in the ventral hom (see fig. 96).

156 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

funiculi ; they have, in general, less collaterals, but come, as does the other type, into synaptic relations near the periphery of the cord with motor and coordinating neurons. Thus an intricate plexus is found near the surface of the cord, concerned in bringing the incoming sensory stimuli to the dendrites of effectory cells. A direct synaptic relation of sensory fibers with the bodies of motor cells is the exception. Intercalated neurons, having both crossed and uncrossed processes, frequently interrelate the incoming stimuli with the effectory path.

Intercalated neurons are very abundant in cyclostomes. To this group belong certain large cells which are probably the homologues of the giant cells of Amphioxus. The cell bodies of these neurons — unhke those of the commissural neurons of higher vertebrates which are usually situated in the dorsal horn region of the cord — for the most part have a somewhat ventral position {Tretjakoff, ’09). This may be associated with the flattening of the cord in these animals {Ariens Kappers, ’20). The dendrites of these cells extend in all directions from the ceU body, some of them even reaching the other side of the cord by way of the commissura protoplasmatica posterior. The neuraxes cross the ventral raph6, form T-like divisions, and run in caudal and frontal directions in the ventrolateral funiculus ; ultimately they terminate in the gray substance of the spinal cord or in the peripheral dendritic plexus. The frontal limits of the tract have not been ascertained for cyclostomes ; it is quite probable that most of the fibers do not pass beyond the cord, but there is the possibility that some of them may reach the medulla oblongata to synapse there with the large reticular cells found at the level of the vagus. These fibers, spoken of as the ventral arcuate system, probably are present as neuraxes of giant cells in Amphioxus, and appear in all vertebrates as one of the first constituents of the spinal cord. They form the most primitive secondary sensory tract of the cord and carry from the skin the so-called vital or protopathic sensations — general tactile, pain, and temperature. They also carry some proprioceptive impulses from the muscles, since primitive intramedullary ganglion cells also send sensory offshoots to the muscles (see fig. 83B) . In addition to the crossed fibers, uncrossed fibers are present in the cord of Petromyzon ; these latter appear to enter the dorsal and dorsolateral funiculi and are to be regarded as processes of intersegmental or intercalary neurons. Both crossed and uncrossed fibers belong to this intrinsic apparatus of the cord.

The spinal cord of Petromyzon also contains elements which carry impulses from the medulla oblongata regions to the cord. These are Muller’s fibers, which are conspicuous on accoimt of their size (figs. 64, 68, and 69). The cells of origin of these fibers lie in the midbrain and the medulla oblongata (see fig. 316). The fibers, which number from 6 to 8 on a side, can be traced to the caudal part of the cord. For the most part they do not cross but synapse with the motor cells of the same side, the neuraxes of which pass in turn to the only organ of locomotion possessed by the animal (Petromyzon has no fins), the tail. No collaterals ard given off from Muller’s fibers ; nevertheless, impulses imdoubtedly are sent out from them as they pass through the various levels of the cord, for the dendrites of the motor and the intercalary cells lie in very intimate relation with them throughout their course (fig. 67B). This apparatus is new in cyclostomes. All bodies of

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 157

the dorsomedial cells of Rohde in Amphioxus are not comparable to the cells of origin of Muller’s fibers ; they do not lie in the same region of the brain, they are not under the influence of optic and vestibular impulses, and they have a contralateral course rather than the homolateral one taken by Muller’s fibers. Allen (’17) discussed the possibility that processes of the giant Muller cells enter the ventral root directly. He was unable to trace them into it, but he found giant fibers in the peripheral nerves.

Cephalization has gone much farther in cyclostomes than in Amphioxus. The aboral refle.xes are developed much more highly in the former animals, due to the greater development of the brain and sense organs. Physiologically this is seen in the greater contrast between the movements of the severed hindparts and those of severed foreparts in cyclostomes than in Amphioxus.

In addition to their sensitivity to touch, pain, and temperature changes and their primitive muscular sensibility, petromyzonts respond to certain chemical stimuli and, although intramedullary light cells are absent, react to light independently of eyes. Parker (’12) proved that the skin of these animals is more sensitive on the head than on the trunk. This same observer demonstrated that the larval form of Petromyzon (Ammocoetes) can be stimulated by sour, alkaline, salty, and bitter solutions but not by sweet solutions. However, chemical stimulations are most effective on the skin of the head, less on that of the tail, and least effective on the skin of the middle of the body. Since the thresholds for light and chemical sensibility vary and the region of highest sensitivity for the one is not the most sensitive area for the other, it is probable that they are subserved by separate nerves. Special sensory end-organs for these senses have not been demonstrated in cyclostomes. Retzius (’92) and Stefanelli (’32) found only free intra-epidermal nerve endings (fig. 24) . The special sense terminations in the skin are related to taste and lateral line nerves from the brain and not to spinal nerves.

Tretjakof (’13) described “Sinnesorgane” in the ependymal layer around the central canal ; these cells suggest those mentioned by Edinger (’06) and Stendell (’14) but denied by Fra7iz (’23 ; see p. 146) for Amphioxus. A further examination for such cells should yield both interest and profit to the observer.

Compared with the supporting tissue in the spinal cord of Amphioxus, that of Petromyzon shows a higher degree of development, since it consists not only of the radiating processes of ependymal cells, but also contains certain distinct elements which may be termed neuroglia cells {Retzius, ’91). The cell bodies of these glia cells lie at the line between the fiber layer and the gray substance ; the processes run toward the periphery, where they participate in the formation of the membrana limitans externa. The course of these processes, like that of the motor nerve cells, is probably governed by the more favorable nutritive conditions existent at the periphery in those forms which have no (Petromyzon) or few (Myxinoids) intramedullary blood vessels. It is possible that certain mechanical factors play a part in determining their orientation.

In the central canal of the cord a peculiar homogeneous fiber is found, which is generally known as Reissner’s fiber (Reissner, ’60 ; see bibliography for teleosts).

158 NERVOUS SYSTEMS OF VERTEBRATES AND OF AIAN

This structure was regarded by earlier observers either as an artifact or as a nerve fiber. Later work has shown it to be neither. It is distinctly non-nervous in character and is associated with the ependymal cells below the commissura posterior (the subcommissural organ of Dendy). According to Sargent (see bibliography for teleosts), it probably develops from processes of the ependymal cells in this region and possibly from other cells as well. According to Dendy and Nicholls (TO, see bibliography for mammalian cord) and Nicholls (’12, T2a, see also ’17, under bibliography for plagiostomes), it has to do with the regulation of body flexion and also is related to the pose of the body at rest. Dendy suggested that, although itself non-nervous, it may, through tension, secondarily affect nerve cells and so ultimately lead to body responses. Whether or not one may attribute a secondary part to it in the transmission of stimuli seems extremely doubtful from the evidence at hand. Jordan (’25, see bibliography for teleosts)

studied this structure in fishes (trout) and interpreted it as affording sup

port for the inside of the neural tube. A discussion of this question is to _ be found in the papers of Dendy and Nicholls, Nicholls, and Jordan referred to above.

The Spinal Cokd of Plagiostomes The study of the spinal cord of

Fio. 70. Posterior end of the spinal eord of a shark plagiostomes is important because

here, for the first time, the structure of the cord may be considered the prototype of that of higher vertebrates. In form it is not flat, as in cyclostomes, but round (Rajidae) or oval (Selachii), and it extends beyond the vertebral canal in the latter (fig. 70, postcaudal cord; Slerzi, ’12), projecting as an ependymal tube into the tail fin. At its termination it is surrounded by lymphoid tissue. In tliis caudal region the cord contains polynuclear glandular cells {Dahlgrcn, ’14 ; Speidel, ’19 and ’22 ; see also bibliography for teleosts). A difference in the level of emergence of the dorsal and ventral roots is still fairly frequent (figs. 72A and B). In Rajidae the alternation of the roots is less con

spicuous and all the nerves are more compressed. Outside the vertebral canal, the roots unite and continue as mi.xed roots (fig. 71), a contrast to the arrangement in Amphioxus and Petromyzon, where such a junction does not appear. Preganglionic branches are given off which, on uniting, form ple.xuses, but a chain of ganglia on each side is not evident.**

The size of the spinal nerves decreases considerably in the oral direction, as may be seen in figure 72 A and B. These oral nerves are called spino-occipital

"Tliere i.s need for furtlior study of the sjanp.ithetic 8y.stctn of plagiostome.s. Witliin our knowledge, tlierc i.s no thorough treatment of the subject. .Vn intcrc-sting account of the ciliary ganglion and the ganglia on the branchial nerves is given by Norris and Ilwjhes (’20) for the dogfLh.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 159

nerves. On a study of figure 72, it will be seen that in the most frontal nerves the dorsal roots are entirely lacking, these frontal nerves thus differing in plagiostomes from the typical spinal nerves. Two groups of nerves may be distinguished, the occipital nerves and the occipitospinal nerves. The former are those nerves which first change into cranial nerves (hypoglossus,

Furbringer, ’97). Whereas the palaeocranium of cyclostomes ends caudally at the region of the labyrinth and does not include any spinal nerves, selachians have a caudal enlargement of the skull (protometameric assimilation) embracing the most frontal group — the occipital nerves. In a second enlargement (au.ximeric assimilation) of the cranium, which only occurs in higher animals, a second group of spinal nerves is included in the skull (the group of occipitospinal nerves, Furbringer, ’97).

The number of the occipital nerves is variable. Most commonly there are two on each side, but there may be one or more additional nerves, which are usually

more or less rudimentary. The major nerves have dorsal branches for the dorsal trunk musculature and ventral branches which unite with the ventral branch of the first spinal nerve (this carries both somatic afferent and somatic efferent fibers). Later these branches are joined by the ventral ramus of the second spinal nerve and the corresponding ramus of the third spinal nerve runs in intimate relation with the fiber mass but not incorporated with the other components. The whole bundle is called the hypobranchial nerve This nerve divides into three major branches and supplies efferent fibers to the brachial plexus and particularly to a group of muscles lying in front of the shoulder girdle. It also sends afferent fibers to the skin on the ventral and ventro-lateral side in the branchial and post-branchial regions. The

IX

X XI

. N. ocap.

IX

X

XI

Vent.

rooU

Fig. 72. A. Anterior roots at the transition to the medulla oblongata in Carcharias glaucus. Ventral view. , ,, , B. Posterior roots at the transition to the medulla oblongata m Carcharias glaucus. The reduction of these roots and their asymmetrical arrangement should be noted. Dorsal view.

SleTzi.

160 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

details of the distribution of these nerves are to be found in the communication on dogfish by Norris and Hughes (’20), from which the above account is taken. Dorsal roots are lacking in the occipital nerves. The loss of these dorsal roots no doubt is related to the fact that the sensory areas which might logically be expected to be supplied by them he so near the head that they are provided for by the sensory branchial nerves. Occasionally the most caudal occipital root has a ganglion (Notidanidae) which is often demonstrable embryologically.

The real spinal nerves are very much alike in structure. The ventral root fibers, particularly in the shark, run obliquely caudad within the spinal cord, since in many cases their cells of origin lie frontal to the emergence of their roots.

All efferent fibers remain imcrossed (fig. 74).

The arrangement of the gray substance is markedly different from that of cyclostomes and resembles that of higher vertebrates ; for while, with cyclostomes, there is no evidence as yet of a division into dorsal and ventral horns, such a division does exist in plagiostomes (fig. 73). However, the spinal gray substance of this latter form shows a marked difference in arrangement from that of higher vertebrates in that the dorsal horns lie so close together that there is hardly any white substance between them. This gives the gray substance the shape of an inverted 7 ix), this peculiarity being especially noticeable in sharks. The cause for this will be considered shortly.

The cells of the ventral horn are large, frequently with oval cell bodies. The outlines of the lateral cells often show a convexity parallel to the border of the gray substance (fig. 74). Their conspicuous net of dendrites stretches into the white substance of the cord and reaches the periphery, as is the case with cyclostomes. Three types of dendrites can be distinguished (fig. 74). To the smallest group belong those which stretch beyond the ventral raph6 into the ventral funiculus of the other side, thereby forming an anterior or ventral protoplasmic commissure (Com. prot. ant. ; fig. 74) which is also present in other vertebrates.*® The dendrites, which extend into the dorsal horns in the direction of the entrance of the dorsal root, are much more numerous (dendrites for sensory-motor reflexes ; fig. 74). These dendrites, as in teleosts, constitute a part of the simple reflex arc

In higher vertebrates a decided change in the distribution of the dendrites occurs, for the dendrites of the ventral horn neurons which extend into the dorsal horn are greatly decreased in number, while there is a considerable increase in the collaterals of dorsal root fibers which extend as far as the ventral horn.

Dora fume.

Fia. 73. Cross section of the upper cervical region of the spinal cord of Galeus canis. Note the well-developed gelatinous substance of the dorsal horns and the small size of the dorsal funiculi.

THE COMPAEATIVE ANATOIVIY OF THE SPINAL CORD 161

formed by the dorsal and ventral horn neurons. The greatest number of the dendrites of the ventral horn cells branch out into the lateral fimiculus and form a ple.\us at the periphery of the cord, the marginal dendritic plexus, which has already been described for cyclostomes. With plagiostomes this plexus lies chiefly in the region of the lateral funiculus and in the dorsal part of the ventral funiculus (iMarg. dendritic net, in figs. 74 and 76). In sections counterstained with paracarmine this marginal dendritic area is noticeable in the fully developed spinal cord as a reddish stained field (see fig. 77). This field fonns the so-called

Dorsal

Fiq. 74. Demonstration of the dorsal root, ventral horn cell reflex in a ray. The dendrites of the ventral horn cells branch in the gray substance of the dorsal horns, von Lenhossik.

campus triangularis funicuh laterahs of Borcherl (’03) and is weU developed, particularly in the upper third of the spinal cord. This marginal dendritic plexus comes in contact with many collaterals of fiber bundles near the periphery. The ventral root cells appear to be distributed fairly regularly throughout the cord ; it has been impossible as yet to show either a segmental pattern or an arrangement dependent upon muscle grouping. It is even questionable whether or not there is any variation or increase in number in the regions where the motor fibers for the fins originate.*® Certainly one cannot speak of intumescentiae cervicahs and lumbalis in these animals. Some visceral efferent fibers may be present in the

“ The intercalary ligaments and the neurapophysis of the vertebrae may cause certain impressions on the spinal cord (Slerzi, 73). These give an appearance of segmentation in cases ■where they exist, but are caused by non-nervous factors and have no intrinsic value.

162 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ventral roots of plagiostomes but these components emerge chiefly through the dorsal roots as in lower forms.

In contrast to the conditions found in Amphioxus and in cyclostomes, all cells of origin for the sensory fibers in fully grown plagiostomes lie in extramedullary spinal ganglia. In embryonic stages, however, intramedullary ganglion cells have been demonstrated. They were described by Beard (’92, ’96), under the name of “transient ganglion cells.’’ He found them near the mid-dorsal line, close to the region of attachment of the neural crest. Thus far, they have been identified in the larvae of Raja (Beard) and of Pristiurus (von Lenhossek, ’92). Besides a sensory branch to the skin, the neurons send one to the muscles for proprioceptive sensibility, a condition found also in larval amphibians (fig. 83). At the end of larval life these cells disappear ; they have never been seen to join

the intervertebral ganglion cells. Levi (’08) and von Lenhossek (’95) have shown that, in the latter ganglia, intermediate t 3 rpes occur between the bipolar cell and the typical unipolar cell with its T-shaped division. There has been a great increase in the unipolar type in plagiostomes as compared with cyclostomes, but they are still sparse and, according to von Lenhossek, are confined in Pristiurus to the proximal part of the ganglion. Usually the centrally directed neuraxis is thinner than the peripherally directed dendrite. The dendrites have free sensory endings at various levels of the skin but no specialized nerve terminations have been described. There is no reason to suppose that general sensibility in sharks is more highly organized than in cyclostomes. The animal (Parker, ’12) has a sense of touch but little ability to localize the tactile stimuli ; it has a chemical sense and appreciates pain and temperature ; proprioceptive impulses are probably carried into the central nervous system from the muscles. The occasionally occurring taste buds and the lateral line organs of the trunk are innervated by cranial nerves — respectively, by the facial and the lateral line nerves. The sensory nerves to the skin of the trunk are arranged segmentally but in such a manner that each nerve, in its distribution, overlaps one-half of the area of distribution of adjacent nerves on the dorsal side and a similar but somewhat larger field on the ventral side, where the segments themselves are broader (van Rijnberk, ’17 ; also, ’05 and ’05a). The larger size of the ventral trunk dermatomes — which here cannot be explained as in birds (fig. 97) by the larger circumference of the ventral part of the animal — probably is associated with the greater irritation to which the ventral part is exposed (de Boer, ’18; see bibliography for mammalian cord) ; this would also explain the greater overlapping of the ventral branches (neurobiotaxis).

Centrally the fibers of the dorsal root run mostly within the gray substance of the dorsal horns. There are no compact dorsal funiculi but only scattered

Fig. 75. Segmental innervation of the akin of Scyllium. van Rijnberk. The white field between the two dotted fields indicates a segment isolated between two other segments and represents an analgesic skin area produced by the cutting through of three serial posterior roots.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 163

bundles of fibers intermingled with the dorsal gray, since the two horns are joined by bridges of gray substance (fig. 77) which in sharks almost entirely connect them. It is only in higher vertebrates, as Brouwer (’15), who worked with a series of animals, has pointed out, that well-developed dorsal fimiculi and clearly separated dorsal horns are to be found. This observer has estimated that in fishes the dorsal funiculi make up only about 6 per cent of the white substance, while in man they form nearly 39 per cent of the white substance. Von Lenhossek (’92) has shown further that, at least in sharks, the larger part of the dorsal funiculus consists of descending neuraxes of dorsal fimicular cells. It is obvious, therefore, that only a very small percentage of the ascending fibers of the dorsal root is to be found in these funiculi. Nevertheless, the ascending branches of the dorsal roots are greater in number in plagiostomes than in either Amphioxus or Petromyzon, but the frontal accumulation characteristic of higher forms has not appeared and the fiber bundles are scattered. On their entrance into the cord, the dorsal root fibers spread out fan-like in the gray substance, some ending in the dorsal and some in the ventral portions of the dorsal horn. The most dorsal fibers have a more or less local ending (fig. 74, possibly visceral afferent fibers are present in this bundle). The more ventral part, made up of coarser fibers, forms relatively long ascending and descending tracts. It is the shorter fibers of this ventral division that come into synaptic relation with the dorsally directed dendrites of the ventral root cells (see previous account ; also fig. 74). The fibers of this division are somatic afferent. The central ending of the visceral afferent fibers is not sufficiently well known.

The short course of the ascending fibers of the dorsal roots in plagiostomes compared with that of homologous fibers in mammals explains the lack of a frontal accumulation of fibers. Associated with the lack of such bundles is the absence of the dorsal funicular nuclei of Goll and Burdach, or nucleus gracilis and nucleus cuneatus {Brouwer, ’15). Likewise a medial lemniscus, in the strict sense of the word, is lacking. Walleriberg (’07) was of the opinion that the vestige of the dorsal gray found medial to the nucleus of the spinal trigeminal tract at the level of the calamus scriptorius is to be regarded as a beginning of the dorsal funicular nuclei. Of necessity, it would represent a very primitive vestige of these nuclear masses, since Wallenberg himself could demonstrate no medial lemniscus fibers running forward to thalamic regions.

The cells of origin of the visceral efferent fibers have not been demonstrated with certainty. Von Lenhossek (’95) was able to follow fibers through the spinal ganglia which have no relation to its ganglion cells. They can be traced out through the dorsal roots and, peripheral to the ganglia, they separate from the spinal nerves and pass to the intestine. Gray rami and a characteristic sympathetic ganglionated chain do not exist here {Young, ’33), but the fibers in question form plexuses containing postganglionic cells, the neuraxes of which pass to the walls of large vessels and to the intestinal tract. The visceral efferent fibers are accompanied by dendrites of visceral afferent neurons as they leave the spinal ganglia. Erik Muller (’20) found longitudinal connections between the visceral roots of the different segments. This author, who made a very elaborate study

164 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

of the sympathetic system of these animals, found the relations of this system much like those in cyclostomes, there being dorsal as well as ventral visceral efferent roots and, in addition, peripheral nerve cells on the intestine, which, however, Muller considered at least partly sensory cells.

There are two types of cells giving rise to fibers endogenous to the cord. These are the commissural cells and the funicular cells. The commissural cells (fig, 76) are scattered throughout the whole gray substance, even being interpolated between the motor cells of the ventral horn (as in cyclostomes). Their dendrites extend throughout the full breadth of the spinal cord, passing partly to the contralateral side through the commissura protoplasmica posterior or dorsalis. They have been beautifully illustrated for Pristiurus embryo by von Lenhossek (’92 and ’95). Their neuraxes, which constitute the arcuate fibers of His, pass through the frontal commissure, which in selachians consists of two divisions, one part which lies directly below the central canal, comparable to the position of a similar commissure in mammals, and another portion which is situated more ventrally. This latter is called the commissura accessoria or the commissure of Mauthner. It is not equally well developed in all plagiostomes and is some Fig. 76 . Embryo of Spinax. von Lenhossek. times absent. The arcuate fibers, which Cells of origin of arcuate fibers and ventral horn , , ..1,1 1 i i

neurons, with peripheral dendritic branches. homologous With those already de scribed for Amphioxus and for cyclostomes, branch after crossing in the commissure and their frontally running fibers form the phylogenetically oldest secondary sensory tract from the cord to the medulla oblongata. This is the first sensory projection tract of the cord and is probably analogous to the tract of Edinger (spino-tectal tract) as described for mammals. Some of these fibers, arising in cervical regions of the cord and accompanied by secondary fibers from the spinal trigeminal nucleus, may even reach the roof of the midbrain, thus forming a true spinomesencephalic system {Wallenberg; see account in Chapter VIII).

The non-crossed correlation tracts which originate from funicular cells are of two kinds. The first has cells of origin in the ventral horns, which send their neuraxes into the ventral funiculi of the same side (anterior or ventral fundamental funiculi) . The second sort arises from cell bodies arranged in dorsolateral and dorsomedial groups in the posterior horn. The dorsolateral group sends its neuraxes into the lateral funiculus ; the dorsomedial group — consisting of larger cells — discharges its neuraxes into the most ventral part of the dorsal funiculus. These latter, larger cells send their dendrites into the commissura protoplasmica posterior or dorsalis.

165

166 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Spino-cerebellar fibers are present — at least in sharks. This is to be e.xpected in animals where the body movements are so well developed. The fibers originate mainly homolaterally in the upper cervical cord, maintain a lateral position in their course through the medulla oblongata, and enter the more central part of the cerebellum (corpus cerebelli; fig. 366). They are suggestive of the dorsal spinocerebellar tract of higher vertebrates (the direct cerebellar tract or cerebellar tract of Flechsig). A ventral spino-cerebellar tract (Gowers’ cerebellar tract) cannot be demonstrated with certainty. If present, it is probably small (see the chapter on the cerebellum). There is some indication of a spino-olivary tract, situated at the periphery of the cord in the upper part of the cervical region. The position of the cells of origin for this fiber system is uncertain, though they probably lie in the dorsal horn.

The descending tracts are larger and, in consequence, there is an increased dominance of the higher centers over body movements (cephalization) in plagiostomes as compared with cyclostomes. The descending fibers originate mainly in the medulla oblongata and the midbrain. Fibers from the medulla oblongata have several nuclei of origin. One of the larger bundles is the tractus octavomotorius cruciatus, which arises from the acoustico-lateral nuclear region (Ariens Kappers, ’06) and, accompanied by fibers from the cerebellum (tractus cerebellomotorius cruciatus et rectus), enters the cord, where the combined bundles form a considerable part of the ventral funiculus. Other descending elements of the spinal cord are neuraxes of the large reticular cells of the medulla oblongata. The tractus vestibulo-spinalis medialis (tractus octavo-spinalis medialis of Wallenberg, ’07 ; fasciculus medianus of Slieda, ’73) arises from the ventral vestibular nucleus of Ariens Kappers (’06) and descends uncrossed into the medial part of the lateral funiculus. This carries some direct root fibers. A crossed tract from the area of the acoustic nerve reaches the ventrolateral funiculus ; the tractus octavo-spinalis lateralis cruciatus of Wallenberg (’07). (For further details consult Chapter IV.)

To summarize, the motor centers of the cord are under the influence of impulses from the lateral line, vestibular, and cerebellar centers and thus possess abundant static innervation, as is to be expected in animals e.xhibiting such rapid movements and as was foreshadowed by the Mullerian fibers of cyclostomes. These long descending paths are supplemented by short descending paths, intrinsic to the cord, which are concerned in spinal refle.xes. The presence of direct tccto-spinal fibers is uncertain, as is also any direct connection from the inferior lobes of the diencephalon (through which responses of the body to olfactory impulses might be mediated; see Chapter VIII). Although direct tracts are wanting, impulses from these centers are able to reach the cord readily by way of the reticular cells of the medulla oblongata with which they are known to be in synaptic relation.

The supporting substance of the plagiostome cord is better developed than that in cyclostomes but is primitive as compared with that in mammals. Tims proces.ses of the ependymal cells extend through the cord to its periphery, where they contribute to the formation of the membrana limitans externa. The

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 167

ependymal cells of the obliterated dorsal part of the central canal appear as real glia cells in the region of the posterior septum and are numerous close to the intraspinal blood vessels. True autonomous glia cells occur in the gray as well as in the white substance {Erik Muller, ’99), but the cells show at least one primitive character in having large peripheral offshoots which extend near or into the peripheral plexois, thus showing clearly their origin from ependymal cells. It is only in the immediate neighborhood of the cell body that it is possible to see secondary, beard-like processes which are suggestive of spin cells {von LenhossSk, '95) or astrocytes. The fiber of Reissner has been frequently identified in plagiostomes, and Nicholls (’12, ’12a, ’17) has examined it very thoroughly. He, like Bendy (’02; see also Bendy and Nicholls, ’10, bibliography for mammalian cord), believed that the fiber is concerned in the automatic control of the flexure and pose of the body. He regarded it as related mainly to the cells of the subcommissural organ of Bendy, but also in part to other ependymal cells. He did not credit it with any direct nervous conductivity. Jordan (’25) regarded it as supporting in character (see bibliography for teleosts).

The Spinal Cord of Ganoids and Teleosts

In the following account of the spinal cords of ganoids and teleosts, in order to avoid repetition, attention will be devoted largely to certain details in which these forms differ from plagiostomes. Of the two, the ganoids resemble the plagiostomes the more closely, while the teleosts show considerable differences in structure.

Mention has been made of the fact (p. 158) that in sharks (although not in rays) a group of nerves is found between the branchial nerves and the first spinal nerve, to which the names of occipital and occipito-spinal nerves have been applied by various observers {Fiirbringer, ’97 ; van der Horst, ’18, and many others as well). In the figures by Addens (see Chapter V) only the term “ occipital nerve ” has been used. The dipnoan, Ceratodus, and the holocephalan, Chimaera, have occipital nerves (fig. 239, Chapter V), and so do the ganoids, Acipenser, Amia, and Calamoichthys (fig. 240, Chapter V). Many of the teleosts appear to lack such nerves (either occipital or occipito-spinal, in the sense of Fiirbringer). The conditions in such teleosts are illustrated in the charts of Megalops, Lophius, and Orthagoriscus to be found in Chapter V. However, two nerves of this type, one emerging from the skull (an occipital nerve or Nerve z of Fiirbringer) and a second immediately caudal to the skull (an occipito-spinal nerve or Nerve a of Fiirbringer), have been described in the trout by Beccari (’22), who showed that the assimilation of a single vertebra occurred during the formation of the skull in this fish.

The more anterior of the two nerves is represented almost entirely by a ventral ramus (with only a rudimentary and often independent dorsal ganglion present). In the trout this occipital ramus, together with the ventral ramus of the following nerve (the spino-occipital nerve of Beccari) and the first two spinal nerves, forms a cervico-brachial plexus by way of which fibers from the cervical trunk pass to the coracohyoid muscle. Other fibers of these ventral rami unite

168 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

with similar rami of the next two spinal nerves (the 3rd and 4th of Beccari and the 4th and 5th of Furhringer) to supply innervation to the pectoral fins. Beccari

A c

Fig. 78. A. Trigla hirundo.

B. Brain and spinal cord of Orthagoriscus after removal of the cauda equina. Bela Haller. Note shortness of the spinal cord.

C. Brain and spinal cord of Orthagoriscus mola with cauda equina in position. Preparation at the Amsterdam Institute for Brain Research.

raised the question of a possible relationship between the coracohyoid musculature of these fishes and the tongue musculature of Amniotes. This matter will be referred to again under the account of the h 3 rpoglossal in Chapter V.

In teleosts such as Megalops, Lophius, and Orthagoriscus, where such additional nerves as the occipital (and occipito-spinal) appear to be lacking, immediately behind the branchial nerves (and in part at their level) are pure spinal nerves with complete and even with large dorsal roots. The ventral roots of these teleostean nerves innervate the more cephalic dorsal and ventral longitudinal musculature and in part the muscles of the pectoral girdle, while the musculature innervated by the spino-occipital nerves in plagiostomes has disappeared. The reduction of the cephalic part of the spinal cord of teleosts, which was brought to the attention by the work of Furhringer (’97) and which has been studied by van der Horst and other observers (Chapter V), is the result of a shortening of the cord in the region between the skull and the vertebral column. Here the most frontal vertebrae and the corresponding musculature and nerves have been lost. The first nerves of the spinal cord emerge through the skull due to auximeric assimilation by the cranium.

At the caudal end of the spinal cord in most teleosts are found the large glandular cells first described by Dahlgren (’14; see bibliography for plagiostomes) and Speidel (’19) for the skate. Speidel (’22) found them in twenty-six of the thirty forms examined by him. In addition to these cells, a peculiar ventral enlargement of the cord is to

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 169

be seen at about the level of the last vertebra. This was first observed by Arsaky (1813), and studied further by Weber (1827), Zagorsky (1833), and Girgensohn (1846). It was studied in detail by Yerne (’14) and especially by Favaro (’25) under the name of hypophysis caudalis. Apart from glandular cells, this hypophysis caudalis contains septa of connective tissue and blood vessels of meningeal origin. The structure is lacking in Murenoids and Lophiobranchia and is best developed in Clupea, Atherina, and Lophius, where it may be divided into two lobes, each of which curves upward around the cord (for further details see the work of Favaro, ’25).

Vent, root^

Fia. 79. Cross section through the posterior end of the spinal cord and cauda equina of Lophius piscatorius. Note the smaller circumference of the dorsal roots as compared with the ventral roots.

In ganoids and primitive teleosts (see fig. 78) the spinal cord is very long and extends throughout the whole vertebral canal. In plectognathes there is a considerable discrepancy in the length of the cord and that of the canal ; particularly clear examples of this are to be found in Orthagoriscus and in Lophius.

In certain of these animals, such as Lophius, on a most cursory examination the cord appears to extend the entire length of the canal, as is the case with most other teleosts. However, study of the preparations reveals that the spinal cord proper forms only a small part of this substance and that the greater part of the vertebral colinnn is filled with roots running longitudinally, thus forming a relatively large cauda equina, in the center of which the filum terminale can be traced as a fine fiber (fig. 79, Lophius). Since the union between the dorsal and', ventral nerve roots occurs outside of the vertebral column and since the ganglia are also situated outside, the elongated portions of the dorsal roots consisFmly ! of the central connections between ganglia and cord.

170 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Burr (’28) studied the brain and cord of Orthagoriscus mola. In this fish the cord is greatly reduced in length and there is a corresponding reduction in the number of spinal nerves, of which he recognized twenty-one pairs. The first two of these are dorsal components of the first spinal nerve, which Burr regarded as the possible homologue of the hypoglossal nerve. Three others of the rootlets become associated with other nerves, leaving sixteen spinal nerves (motor and sensory) on a side. This great reduction is associated with the almost total lack of tail musculature, as had been pointed out earlier (Ariens Kappers, ’20) . Of this fish. Burr stated : ‘ ‘ Orthagoriscus is really nothing but an enormous swimming head with a short trunk and no tail, correlated with which is a marked shortening of the cord. ”

Two factors at least must receive consideration in discussing the differences in length of the spinal cord and spinal canal in mammalian forms (see p. 220). These are (1) the atrophy of the sacral segments of the cord due to the reduction or entire loss of the caudal portion of the body and (2) the continuance of growth of the vertebral column (in connection with the development of the pelvic girdle) after the spinal cord has attained its greatest length. Unequal growth of the spinal cord and spinal column are not found in the plectognathe fishes, so that the second factor, probably the more important one in mammals, does not operate here. The first causal factor listed above is evidently the effective one, for in plectognathes the caudal region of the body shows considerable reduction and consequently is much less developed than the cephalic region. Whether other factors may play a r6le here also is uncertain. It is interesting that the sensory roots in particular are much thinner in the caudal region, as is evident in figure 79, while the cephalic sensory roots are hypertrophied.

In teleosts the fusion of dorsal and ventral roots to form a mixed nerve is constant. Frequently, however, these roots do not emerge at quite the same level, although this condition is less evident than in lower vertebrates. The cells of origin of the ventral roots of the cord, as a rule, lie more ventrally in teleosts than in plagiostomes. This may be due to a greater neurobiotactic influence of the ventral fiber systems, as is clearly the determining factor in the position of the somatic efferent nuclei of the medulla oblongata (see Chapter V) in these animals. A comparison of figure 77 with figures 80 and 81 shows plainly this difference in position in the cord. The relations of the gray matter of the cord foreshadow the conditions found in higher forms ; ventral horns are clearly indicated though not so sharply defined as they become phylogenetically later, while short projections joined by a wide band of gray, the corpus commune posterius of Keenan (’28), represent the dorsal horns. According to Keenan, it is probable that this corpus commune posterius contains within it the representation of the body of the dorsal horn, while th e bilat eral enlarg em ents of t he fis h represent the substantia gelatinosa Rdlandi of th^mammal.

^ 'IhVnany teleosts two groups of efferent cells are clearly distinguishable in the gray matter of the upper cervical cord : a so-called dorsomedial motor and a ventral or ventrolateral motor group (figs. 80 and 81). The former group is presumably concerned with the innervation of the cervical fins ; the latter group.

THE COIVrPARATIVE ANATOMY OF THE SPINAL CORD 171

which extends throughout the body and which frequently consists of somewhat smaller cells, is probably for motor fibers to trunk musculature. The marginal plexus is not so well developed in teleosts as in plagiostomes. The majority of the dendrites branch in the white substance, some of them forming a commissura protoplasmica anterior or ventralis extending to the other side {Martin, ’95; van Gehuchten, ’95), while others are directed toward the region of entrance of the dorsal roots and are probably concerned in sensory-motor reflexes. Usually the neuraxes of the cells run a short distance in a caudal direction in the ventrolateral funiculus, where they give off collaterals.

Dors. mot. cerv. nucl

Fig. 80. The spinal cord of Salmo fario at the level of entrance of the first spinal nerve, van der Horsl.

Collaterals arising from the neuraxes of vertebrate motor cells in the earlier part of their course were described first by Golgi (’83, mammalian bibliography). Von Lenhossek (’95) regarded such collaterals as present in mammals but as absent in sub-mammahan forms, including fishes. However, van Gehuchten (’95) and Martin (’95) described such collaterals in fishes and Ramon y Cajal (’09) believed that such collaterals were present in fishes, amphibians, and reptiles, although less numerous than in mammals. In mammals such collaterals turn back into the gray, where they terminate in free arborizations. There is a difference of opinion with regard to their relations in fishes. By certain observers these relations are regarded as essentially similar to those in mammals. Other observers have considered that the collaterals terminate in

172 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the perimedullary protoplasmic (or the marginal) plexus in these and related forms {Sola, ’92; Ramdn y Cajal, ’09).

The electric cells of Malapterurus and Gymnotus electricus are a special type of motor cell. In plagiostomes (Torpedo) the electric organ is differentiated from gill musculature and accordingly is innervated by branchial nerves, but in the teleosts mentioned above, the organ is a derivative of the body musculature or skin. It is innervated by spinal nerves. In Malapterurus electricus on each side there is only one large electric cell, which is situated in the upper part of the cord. From this cell a coarse fiber, with heavy myelinated sheath, runs to the electric apparatus in the skin {Bilharz, ’57 ; Ballowitz, ’99). The electric eel, Gymnotus electricus, however, has several such cells. Nearly all the gray substance at the level of the nerves supplying these organs, particularly in the portion dorsolateral to the central canal, is filled with large multipolar cells. In Mormyrus, which is also described as an electric fish, the fibers innervating the electric organ join the ventral root of the third spinal nerve {Stendell, ’15), the fibers possessing an enormous myelin sheath. They innervate derivatives of myotomic musculature, consequently their cells of origin are situated in the ventral horn of the cord in the above-mentioned segments.

The present knowledge regarding the sympathetic system in teleosts is relatively very meager, especially in regard to the finer details. Stannius (’54) and Huher (’99) divided the 'sympathetic system into cephalic, trunk, and caudal (post-anal) portions. These portions consist of a series of ganglia connected by intervening fibers into two cords which lie on either side of the vertebral column. The cephalic portions are associated particularly with the branchial nerves. The chain ganglia of trunk and post-anal regions are connected by rami communicantes (Young, ’31). A large collateral sympathetic ganglion — the splanchnic — is found at the cephalic end of the trunk where the two cords approach each other. In teleosts visceral efferent fibers are said to course partly in the ventral and partly in the dorsal roots. Neuraxes of visceral efferent neurons, as stained in methylene-blue preparations, were found to form well-developed intracapsular plexuses surrounding the cell bodies of sympathetic neurons (Huber, ’99). Neuraxes of cells within the cord certainly pass directly through the spinal ganglia without synapse, and in part are probably of this type. The cells of origin for the preganglionic fibers have not been identified clearly, though probably they lie near the base of the dorsal horn. Peripheral to the spinal ganglia, where the dorsal and ventral roots unite, the preganglionic fibers leave the spinal nerves and enter as white rami communicantes into the formation of the plexus which shows the characteristics of a ganglionated chain, comparable to that of higher forms. Kuntz (’ll) studied the development of sympathetic ganglia in fishes.

The posterior root ganglia or spinal gang lia of teleosts have bipolar and unipolar neurons and intermediate types between these (Martin, ’95). Somatic afferent fibers of the dorsal root may have their origin in these ganglia, but they arise in part from cells situated within the gray matter of the cord (as in cyclostomes) and in part from sensory cells lying directly over the cord (fig. 81) or

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 173

above the caudal part of the medulla oblongata (Burr, ’28, fig. 34). Such primitive intra- and supramedullary ganglion cells were first described in Lophius piscatorius (fig. 81) by Fritsch (’86), and have been considered by various other observers. The supramedullary neurons ” extend over nearly the whole spinal cord but accumulate chiefly in the cervical region. Here they usually lie above the spinal cord (underneath the dura) but are found also dorsal to the motor nucleus of the vagus. Spinal ganglion cells often attain considerable size in teleosts and these supramedullary cells are very large. In Orthagoriscus (Letd, ’08 ; Burr, ’28) they are multipolar and are surrounded by a sort of fenestrated membrane through which capillaries pass in order to form a rich plexus around the cell bodies. In Lophius small endocellular canals (trophospongium of Holmgren, ’03) have been described as arising from the capsule and these may carry a capillary supply within the cell (Siudnicka, ’03). In Antennarius and in Tetrodon the central, unmedullated processes of these intra- and supramedullary cells ascend in a dorsolateral bundle to the spinal trigeminal nucleus. Neuraxes from dorsal nerves of the cervical cord also reach this region (Ariens Kappers, ’20, German edition of this text).

” The following list gives the fishes in which these cells have been found, with the names of those observers who found them. In those animals that are marked with an asterisk (as in Plagiostomes), the cells occur only in larval stages. This table was prepared by van der Horst.

Ordbk I. Malacoplerygii.

Fam. Salmonidae : Salmo salar,* Salmo faris* (Rohon, van Gehuchten, Harrison) ; Coregonus adspergus, Coregonus albus (Johnston).

Order II. Ostariophysii.

fam. Cyprinidae : Rhodeus* (Sfudntcto) ; Catostomus (/o/wsfon).

Order III. Symbranchidae.

Fam. Symbranchidae : Monopterus, Symbranchus.

Order IV. Heteromi.

Fam. Fierosperidae : Fierosper horaei.

Order V. Catosteomi.

Fam. Syngnathidae : Syngnathus, Hippocampus.

Order VI. Anacanthini.

Fam. Gadidae ; Ixitta.

Order VII. Acanthopterygii.

Fam. Percidae : Perea fiuviatilis (Koester) ; Labran.*

Fam. Sciaenidae; Corvina nigra (Tagliani).

Fam. Labridae : Ctenolabrus pavo (Johnstdh) ; Ctenolabrus adspersus (Sargent).

Fam. Scorpaenidae : Scorpaena (Tagliani).

Fam. Cotlidae : Hemitripterus (Tagliani) ; Cottus.

Fam. Triglidae : Trigla ; Peristedion (Tagliani).

Fam. Dactylopteridae ; Dactylopterus (Tagliani).

Fam. Callionymidae : Callionymus lyra.

Fam. Batrachidae: Batrachus (Tagliani).

Fam. Zoarcidae : Zoarces vivipara.

Order VIII. Pediculali.

Fam. Lophiidae : Lophius piscatorius (Fritsch) ; and all other Pediculati (Sargent).

Fam. Antennariidae : Antennarius histrio.

Order IX. Plectognathi.

fam. Ballistidae ; BalisUis (Tagliani) ; Gymnodontes.

Fam. Tetrodonbidae : Tetrodon (Haller).

Fam. Molidae : Orthagoriscus mola (Haller, Tagliani, Levi, Burr).

Besides these, cells have been demonstrated in larval stages of Ganoids (Acipenser, von Kupffer) and in Lepodosteus (Beard), as well as in Dipnoi (Protopterus, Burckhardl).

174 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

In his work on Orthagoriscus mola, Burr (’28) traced the unmedullated neuraxes of these neurons from the ventral side of the cell Jbj 2 dy along the medial side of the dorsal or sensory funiculus until they reached a point slightly lateral to the dorsal commissure. Here either the neuraxes divided or in part turned cephalad and in part caudad, for fascicles which gradually decreased in size could be followed both forward and backward from this region. Small cells form a nucleus on the medial side of this bundle. Burr was of the opinion that the neuraxes forming this longitudinal bundle are in synaptic relation with “the

Supramed gang cells

Meninx

pnm Intramed gang cells

Cerv ecus lobe-.

Dors lat fasc and first dors — root bun die

Dors med group vent horn cells

Crossed lat vest sp tr

Dors root

Vent root

Crossed refiex fasc

Ventrolat group vent horn cells

Cent long fasc Med vest sp tr

Fig 81. Cross section through the upper cervical cord of Lophius piscatorius.

sensory mechanism” but was unable to trace any of them definitely to their termination.

Fritsch (’86) stated (according to Burr, ’28, confirmed by Addens) that certain fibers of these supramedullary cells, situated above the vagal area in Lophius, after entrance send their large processes forward along the spinal trigeminal tract until the levels of vagal and trigeminal roots are reached. Certain of the fibers are supposed to leave the central nervous system with these nerves, but their peripheral termination is uncertain, although Fritsch was able to trace them for a considerable distance among the peripheral fibers of the trigeminal nerve because of their unmedullated character and great size. This observer believed that they are in some way associated with the sensory distribution in Lophius which, because of the habits of the animal, is extremely rich. The possibility of the proprioceptive character of these fibers and their

176 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

trigeminal tract, which cell mass is more intimately associated with the dorsal horn proper. In the cod {Herrick, ’07a) both of these nuclei receive fibers from the spinal trigeminal tract, from the dorsolateral and the dorsal funiculi of the cord and from the gustatory centers. In the sea robin {Herrick, ’07) the lateral funicular nucleus, occasionally termed the inferior gustatory nucleus, is farther cephalad than the medial funicular nucleus and has the appearance of an enlargement of the latter gray mass and an extension of it over the spinal nucleus of the trigeminal nerve. This lateral funicular nucleus receives cutaneous fibers from the vagus, fibers from the dorsolateral funiculus of the cord, and a definite tract from the gustatory lobe of the facial nerve (Gadus, Herrick, ’07a). In Lophius and Trigla, where there is a great hypertrophy of the first sensory cervical nerves, there is an amazing enlargement of these frontal portions of the dorsal horns,** so that they appear as distinct swellings on the

Fig. 82. A preparation of the brain and cervical cord of Trigla hirundo in which the tactile organ (fin) on the left side is lacking. The absence of a left posterior cervical eminence should be noted. Preparation from the Anatomical Museum at Amsterdam.

surface. In Lophius there is a sensory lobe on each side (fig. 81) which is a center of termination for the first cervical and descending trigeminal roots. The former provides for the greatly increased sensitivity of the cephalic fins, the latter for the cutaneous impulses from the enormous head of the animal. In Trigla hirundo three pairs of enlargements are visible in the upper part of the cervical cord. These have a purely local significance. They correspond to the upper three cervical nerves which innervate the three fin rays that have become modified to form organs of touch. That these lobes are actually the centers of termination of these tactile fibers from the fins is clearly indicated in the specimen of Trigla hirundo figured (fig. 82), for in th s particular animal, the most caudal of the three fin rays and the most caudal enlargement on the same side are both lacking (compare fig. 82). In such animals the long-ascending fibers of the roots running through the dorsolateral fasciculus are particularly large and so lead to a well-developed lateral as well as a medial funicular nucleus. The funicular nuclei, it should be emphasized, are not directly homologous with the nucleus gracilis and nucleus cuneatus of higher forms. They are concerned, as has been shown, with fibers from cervical regions rather than with ascending fibers from the whole cord and they do not give rise to long-ascending systems

Caudal sensory roots are thinner than motor roots ; cephalic senbory roots are hypertrophied.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 177

such as the lemniscus of higher forms but, to a great extent, to short connections with the gray near their level. Thus thej’’ provide primarily for richer connections from the neck (and head) regions to spinal cord motor centers.

The relations in the American species of Trigla (Prionotus carolinus) have been very minutely studied by Herrick (’07). Like the above-mentioned European species, this animal has three fin raj’^s which are innervated by the first three sensory nerves. It does not have three sensory lobes on each side of the cervical cord, but six, which may be classified into three groups. The first lobe receives the descending trigeminal root, somatic root fibers from the vagus, and the first cervical nerve. The next two lobes {i.e. 2 and 3), which belong together, receive the second cervical nerve. The last three lobes (4, 5, and 6) receive the third sensory cervical nerve. Those lobes which belong together are united by fasciculi proprii and, in addition, produce commissural systems which run frontalward into the somatic part of the commissura infima. In this animal, the dorsal funiculus of the spmal cord consists mainly of secondary descending fibers of these lobes. The ventrolateral and particularly the dorsolateral funiculi are considerably strengthened by descending fibers of the region — especially of the most caudal enlargements. In addition to root fibers, the lobes receive fibers from the dorsal horns themselves.

The secondary tracts in the spinal cord of teleosts are found to be both contralateral and homolateral. The cells of origin of the contralateral tracts are the commissural cells, phylogenetically the oldest secondary neurons of the cord. The neuraxes of these neurons, after crossing the midline through the commissura anterior or ventralis, enter the ventral and ventrolateral funiculi of the other side, mainly that portion of the fasciculus longitudinalis medialis^® which lies between the commissura anterior (ventralis) proper and the commissura anterior (ventralis) accessoria. Many of these fibers divide dichotomously, their ascending and descending branches being mainly intrinsic to the cord. It is doubtful whether any fibers from the cord reach the midbrain centers directly by way of the medial longitudinal fasciculus.

Neuraxes of other commissural cells, after decussating, enter the ventral funiculus, where they divide dichotomously. The branches turn into the lateral funiculus and ascend as far as the medulla oblongata (I'OJi Gehuchten, ’95), thus forming a tractus spino-bulbaris cruciatus. Some of them may reach the midbrain (tractus spino-mesencephalicus ; consult fig. SI). This is apparently the secondary sensory path of Edinger (’08) which carries primitive perceptions of pain, temperature, and ordinary, non-discrmunative tactile sensibility. The cells of origin of the homolateral endogenous fibers of the spinal cord, known as the funicular cells, are spread throughout the dorsal horn of the cord throughout its length, including the hypertrophied cervical portions. Their dendrites receive stimuli from contralateral roots through the commissura protoplasmica anterior or ventralis, as well as impulses through root fibers of the same

‘’This fasciculus becomes continuous with a bundle similarly designated (or sometimes termed fasciculus longitudinalis posterior) which occupies a position below the ventricular floor of the medulla oblongata and midbrain.

178 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

side. (This applies also to the commissural cells.) The axons of the funicular cells enter the ventral, the dorsal, and particularly the lateral funiculi of the cord. Cells similar to the commissural and funicular cells of the dorsal horn of the cord constitute the lateral funicular nucleus, which is to be regarded, then, as an enlargement of the medial funicular nucleus (see previous discussion).

Crossed paths to higher centers (probably to the midbrain) and shorter bundles to the reticular gray of the medulla oblongata and to near-lying ventral horns are present from the hypertrophied dorsal horns and from the lateral funicular nucleus. The numbers of fibers ascending to higher centers, as these are represented by midbrain, are not very great, suggesting, as Herrick (’07) pointed out, that the hypertrophy is associated with a local reflex pattern. However, Burr (’28) believed that a spino-thalamic path is present in Qrthagoriscus. A fairly large tract connecting the spinal cord and cerebellum has been described for Prionotus by Herrick, and confirmed for Gadus by Brickner (’29). In Prionotus this consists of heavily myelinated fibers which are readily distinguished from the fibers of the spinal trigeminal tract with which the bundle is associated. Herrick followed it to the vagal lobe region as a separate tract, but behind that level it could not be differentiated so sharply from other bundles of the dorsolateral funiculus. Herrick believed that probably it is continued caudally in Prionotus. The direction of conduction is not known, but it appears probable that it is the homologue of the spino-cerebellar system of higher forms and that, at least, it includes fibers homologous to the ma mm alian dorsal spinocerebellar tract.

In addition to this tract, fibers have been observed running at the ventrolateral periphery of the medulla oblongata and entering the cerebellum by the anterior medullary velum ; Ariens Kap'pers (’06) and Wallenberg (’07) proved their cerebello-petal character. They are probably homologous to fibers of the ventral spino-cerebellar fasciculus of higher vertebrates. In addition, external arcuate fibers to the inferior olive are observed in Lophius {Ariens Kappers, ’06). Burr (’28) described a ventral spino-cerebellar tract to the cerebellum from the cord and inferior olive in Orthagoriscus. According to his account it ascends from cord levels to a superficial position on either side of the ventromedian sulcus and can be traced forward to the level of the vagus and inferior olive, where it lies ventral to the latter. Here, according to Burr, a decussation of both coarse and fine fibers takes place ; the coarse representing those from the cord, the fine those from the olive. Cephalad to the decussation Burr followed the tract forward to the level of entrance of the acoustic nerve, where the bundles turned dorsalward and entered the eminentia granularis. To the writers’ {Huber and Crosby) knowledge, such a medullar decussation of a ventral spino-cerebellar system has not been described elsewhere.

In general, the descending paths resemble those described for selachians. However, there are some noteworthy differences in the degree of development of certain paths. The number of descending fibers is greater in the plagiostomes than in the teleosts. However, the latter group shows a greater development of the veitrolateral funiculus, where the tractus octavo- (vestibulo-)spinalis cruciatus

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 179

lateralis of Wallenberg (probably the tecto-spinal of Burr, ’28) is situated. A similar, uncrossed tract lies in the ventral funiculus, tractus octavo- (vestibulo-) spinalis medialis. It is the ventral position of these tracts, particularly of the large lateral one, which is regarded as responsible for the more ventral position of the ventral horn cells (fig. 81). These tracts are present in plagiostomes but are small in contrast to the medial longitudinal fasciculus of those fishes.

In teleosts many fibers from the reticular cells of the oblongata enter the spinal cord. According to Bartehnez (’15) they run caudalward, crossed and uncrossed, and a part of them accompany the large fibers of Mauthner (’59; also see Chapter IV) as far as the termination of the latter around the motor nuclei for the tail region. In Orthagoriscus Burr described a medial longitudinal fasciculus through the cord (cent. long, fasc., fig. 81). Reticulospinal tracts arise, according to Burr (’28), from the anlage of the inferior olive. These can be followed as medial and lateral bundles into the cord. The lateral bundle can be followed throughout the cord, the medial bundle apparently extending through upper levels only. By means of these reticular cells, optic and static stimuli, as well as perceptions of cutaneous sensibility from the head and impulses brought in through olfactory centers, are transmitted to the motor centers innervating the body and tail musculature.

An outstanding difference between plagiostomes and teleosts is the presence in the latter of descending gustatory tracts. Those members of the group which possess many taste buds on the body (Gadidae) have taste buds on the breast fins {Herrick, ’05, ’06). Here secondary and tertiary gustatory tracts run back from the region of the facial lobe into the spinal cord, at first closely joined to the spinal trigeminal tract and at some levels practically surrounding it, except laterally. Before the level of the hypertrophied dorsal funicular region is reached, the gustatory path has become a part of the dorsolateral fasciculus and continues back in the cord in that white mass. This descending tract ends in part in the medial and lateral funicular nuclei and in part extends farther caudalward. Further details will be discussed in a later chapter.

The researches on the supporting tissue of the spinal cord, made by Retzius (’93a), Marlin (’95), and van Gehuchlen (’95), were based on embryonic material ; they show the typical radially directed fibers of the ependymal cells, the cell bodies of which lie near the central canal, while their trumpet-shaped processes extend toward the periphery.

Autonomous glia cells or astrocytes (that is, cells the bodies of which have migrated away from the central canal) are not present in these early stages. The researches of Roister (’98), Erik Muller (’99), and Marenghi (’08) leave no doubt that adult animals show a higher degree of development of the glia cells, similar to that of the sharks. Burr (’28) found, in an adult teleost (Orthagoriscus), ependymal cells the processes of which form the external and internal limiting membranes of the cord. Small fibers from the columnar ependymal cells lining the central canal of the spinal cord extend dorsalward, interdigitate with fibers of the dorsal commissure, and terminate in flat, plate-like endings which lie in close relation to the inner surface of the meninges.

180 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The Spinae Cord op Amphibians

Teleosts are a collateral branch of the vertebrate phylum. A consideration of the phylogenetic development of the spinal cord reveals, as von Lenhoss4Ic has pointed out, that, in many respects, amphibians resemble plagiostomes rather than teleosts. Similar relations exist for the forebrain and medulla oblongata {Ariens Kappers and Hammar, ’18). The resemblance of these animals — particularly of the tailed amphibians — to the plagiostomes is so conspicuous that only a brief account of the tailed forms is necessary. A knowledge of the very primitive conditions present in larvae of tailed amphibians is of general biological importance and throws a clearer light on certain structural features. The following account opens with a brief review of the conditions in these larval forms.

In amphibians, as in plagiostomes — and in contrast with cyclostomes and teleosts — the spinal cord originates as a medullary plate (fig. 220), divided on each side by the sulcus limitans which runs longitudinally along the future venti’icular wall and which separates a dorsal, sensory region from a ventral, motor region. However, such a metamerism is not visible after the closing of the tube and consequent formation of a central canal. In both larval and adult forms of tailed amphibians, the number of roots is very large, but there are outstanding differences in the composition of these roots at the different ages. CoghilVs work (’14, ’24, ’24a, ’26, ’29)^“ on the first sensory neurons of larval Amblystoma makes it very clear that such cells are homologous to the giant ganglion cells or Rohon-Beard cells, which are transitory structures in plagiostomes and permanent and important structures in cyclostomes and teleosts. In amphibians the cell bodies of these neurons are situated in the lateral and dorsolateral regions of the cord and extend forward into the medulla oblongata. Their dendrites, which come in from the periphery, have a branch from the skin and one from the muscles, so that obviously they carry both exteroceptive and proprioceptive stimuli (fig. 83). The neuraxes of the cells (R. B., fig. 83) run cephalad in the sensory tract and end on commissural cells. In the youngest stage in which a reflex may occur, the cells are found only on the border between the spinal cord and the medulla oblongata (fig. 83), but later appear both in front of and behind this level.

The commissural cells (or arcuate fiber cells, fig. 83), the cell bodies of which, according to Coghill (’24), lie in the ventral floor plate, transmit the impulses to longitudinally conducting neurons (fig. 83). The longitudinally directed stem fiber of a descending neuron is intrinsic to the cord, but a collateral of it leaves

“ The work of Coghill affords a picture of an early behavior pattern substantiated by anatomic and physiologic firidings, gives a clue to the stage of differentiation a neuron must attain before it can function, and indicates, since the neurons go on differentiating after becoming functional, that “Growth and function, therefore, may go on pari passu in nerve cells” (Coghill, ’24, p. 26). Ramdn y Cajal (’09) and others supposed that functional activity is an important factor in activating such growth, but Coghill (’26) believed that an inherent potentiality of the cell produces such growth and “that nervous mechanisms while growing, acquire their specificity in behavior through the primary correlation of their growth processes with receptor and effector functions.” In this connection reference is made to the innervation of transplanted limbs by cranial nerves as indicated by the work of Nicholas (J. Comp. Neurol., vol. 57, p. 253) and others.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 181

the cord as the primitive motor root for a myotome (a condition found also in Crustacea). The stem fiber synapses with a neuron the neuraxis of which shows the same structural characteristics, i.e. the stem fiber remains in the cord as a longitudinally running process, while a branch passes to a myotome (see fig. 83).

Longitud) Dally CODductlD^ cclla \

Fig. S3. A. The branching of a peripheral (sensory) process of a Rohon-Beard ganglion cell into the skin and muscle of a Triton larva. Herrick and Coghill.

A.B., ascending bundle.

B. Transmission of the impulse of the Rohon-Beard cells {R.B.) through an arcuate cell to the longitudinally conducting cell with collaterals which form ventral root fibers. Triton larva. Herrick and Coghill. M., myotome.

Obviously this reflex path differs markedly from those of the full-grown cord. The following peculiarities in structure of the early larval cord, brought out by Herrick and Coghill (’15), may be emphasized :

1. A single sensory neuron (Rohon-Beard cell) may receive exteroceptive stimuli from the skin as well as proprioceptive stimuli from sensory terminations in muscle, the latter arising when the muscle contracts. It is evident, then, that not only generalized touch, temperature, and pain sensations, but also a primitive muscle sense belong to the more primitive type of body sensibility, and further, that all these primitive sensations pass centrally over a common secondary path.

2. The primitive motor neurons are not differentiated individually but are collaterals of descending fibers.

3. The reflex path which must be followed by a sensory impulse before it becomes effective is a very long one, particularly in the youngest stage where commissural cells are found only in the most frontal regions of the cervical cord.

4. Many synapses occur, the individual neurons being relatively short.

182 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

This early reflex path gradually becomes modified {Herrick and Coghill, ’15). The ventral commissure, composed of processes of the commissural cells, extends backward and forward from its primitive position until it may be found at all levels of the nervous system behind the optic chiasma. Later the transitory ganglion cells disappear and spinal ganglia of the adult type develop from the neural crest. In these ganglia there are separate representatives for the exteroceptive and proprioceptive cells.

On the efferent side, typical motor neurons appear, probably by modifications of the embryonic cells. The details of the transformation are not known. It is brought about, probably, through an atrophy of the main neuraxis distal to the place of origin of the motor branch, so that what was originally a collateral of the stem fiber becomes the major process. Apart from these changes, new ventral horn cells, new commissural cells, and new funicular cells are formed. At the same time dendrites of the commissural and motor cells develop enormously over the entire width of the spinal cord, as in cyclostomes and plagiostomes, making possible the reception of stimuli from many different systems.

In the fully developed state, in certain ways, tailless Amphibia are regressive when compared with the tailed forms and with plagiostomes ; in other particulars they show a higher development. More than twenty pairs of spinal nerves are present in tailed amphibians. In the sharks the number is even greater and is further augmented by a number of intracranial occipital and extracranial occipito-spinal roots. In the tailless amphibians, as represented by frogs, there are no occipital nerves. The occipito-spinal roots are represented by one nerve only, a part of the ventral branch of the second spinal nerve which innervates the tongue (the hypoglossal, Gaupp, ’01). The first spinal nerve is lacking and the remaining spinal nerves number only ten or eleven. The decrease in number of spinal nerves is due chiefly to the fact that the more caudal ones innervate the tail Fig 84 Spinal embryo and that these disappear in the adult with the

cord and filum ter- atrophy of that organ. In consequence of this disappearance of minale of Rana mu- the more caudal nerves, the cord of the full-grown frog has a very long filum terminale.^' This filum terminale does not contain any root cells but only occasional, and frequently varicose, branches of funicular and commissural cells (fig. 85). It consists principally of glious elements, which are the continuation of the substantia gliosa centralis. In the cephalic half of the filum there are a munber of descending fibers of the dorsal lumbo-sacral roots. These lie in the dorsal part, close to the midline. In the ventrolateral funiculus, descending neuraxes of secondary neurons are evident

It will be remembered that a filum terminale is produced for similar reasons in certain teleosts.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 183

(see fig. 85, F. a. 1.). In the caudal half of the filum these fiber bundles have almost entirely disappeared. The central canal has shifted ventrally and large, roimd openings are seen in the glious substance (fig. 85B). In very rare cases the shortening of the spinal cord may go still further.

Smallwood (’16) described the spinal cord of a toad (Bufo) at the posterior end of which the seventh to the tenth nerves were pressed into a single fiber mass.

Similar relations were found in Pipa by Tensen (’27), who also stated that fibers of one segment joined those of the ne.vt before the root passed through the foramen intervertebrale.

An outstanding difference between the spinal cords of amphibians and fishes is seen in the presence of an evident intumescentia cervicalis and an intumescentia lumbalis in the former animals. These enlargements are due to the development of fairly large extremities.

The ventral and dorsal nerve roots of the larvae (having intramyotomic and intermyotomic courses, respectively) alternate, but those of the adult frog emerge at practically the same level (see fig. 86).

The ventral nerve trunks consist of munerous small bundles so that the level of emergence of a ventral root occupies a greater longitu dinal area than does that of the corresponding dorsal root, although the fibers of the latter are mostly coarser and their number greater than that of the motor root.

Dorsal

cleft

Central canal

B

Fig 85 A Cross section through the upper half of the filum

terminale of Rana mugiens , . ,1. .

B Cross section through the posterior half of the filum termmale of Rana mugiens

184 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The ventral bxindles contain only uncrossed fibers, the cells of origin of which, as with plagiostomes, exhibit a tendency to accumulate at the lateral border of the gray substance. In the region of the cervical and lumbar enlargements, the gray substance of the ventral horns increases (fig. 86). Two cell colunms can be distinguished in the ventral horn, a medioventral one in general, which extends through the entire spinal cord, and a lateral (or ventrolateral) group, which, as a rule, appears only in the regions of the two enlargements. The regions of distribution of the lateral group suggest that its neurons are concerned with the innervation of the musculature of the extremities. This has been confirmed by experimental work {Sana, ’05). They terminate in motor endings which have been described by numerous observers {Huber and DeWitt, ’98, and, among the latest contributors, Garven, ’25, and Hines, ’27). The question of the plurisegmental innervation of the frog muscles has been studied with apparently contradictory results. Cattell and Stiles (’24) reached the conclusion that a plurisegmental innervation occurs in 70 per cent of the skeletal muscles, while Kaiz (’25) thought that none of the muscles in the frog have a double segmental innervation. De Boer (’26 and ’27), working on the leg muscles of the frog, found only monosegmental innervation of the muscle fibers. A recent advocate of the plurisegmental innervation of the striated muscle fibers of the frog is Lawrentjew (’31).

As with sharks, the dendrites of the ventral horn cells of the frog spread over the entire circumference of the white substance as far as the periphery of the cord and there form a perimedullary or marginal dendritic net. This net is particularly conspicuous in the ventrolateral (and lateral) funiculus (compare fig. 88 with fig. 74), where it enters into synaptic relations with the collaterals of this white column. Certain of the dendrites decussate in the commissura anterior or ventrahs (commissura protoplasmica anterior; fig. 88). So-called sensitivomotor collaterals of dendrites {Ramon y Cajal, ’09, vol. 1, p. 503 ; les collaterales sensitivo-motrices) can be traced to the gray substance of the dorsal horns but are seldom present in the adult frog (where the dorsal funiculi send sensitivo-motor collaterals to the ventral horn cells).

In addition to the somatic efferent fibers, the ventral roots carry visceral efferent fibers as well. Opinion differs as to whether all the preganglionics are to be found in this root ; Steinach (’98) believed that they also occur in the dorsal roots, basing his statements on physiological experimentation. Horton-Smith (’97) questioned his conclusions. The cells of origin for the preganglionics probably lie at the base of the dorsal horn, lateral to the central canal. The visceral efferent fibers terminate in intracapsular, pericellular plexuses surrounding the cell bodies of the unipolar sympathetic neurons, the visceral efferent fibers winding about the neuraxes of the postganglionic sympathetic neurons in several spiral turns before forming the pericellular plexuses {Huber, ’99). The development of the sympathetic system of Amphibia has been described by Kuntz (’ll, ’22, ’29).

After the disappearance of the giant ganglion or Rohon-Beard intramedullary cells, which were present during development, only the extramedullary ganglion cells remain. These spinal ganglion cells are mostly monopolar and rarely bipolar

Med. divis.

185

186 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

in character {Levi, ’08). Multipolar cells have been described in tadpoles {Disse, ’93). Many of the unipolar neurons have a greatly coiled process (or glomerulus) which ends in the T- or Y-shaped division characteristic of such cells. Certain of the larger of these neurons present special differentiations {Huber, ’96) in the form of several fine branches which are given off from the greatly coiled process while it is still within the capsule surrounding the ganglion cell. Such fine branches may have a coiled or spiral course but they extend to the cell body, where they have flat, disk-like terminations. There is considerable variation in the number and arrangement of these secondary branches ; usually they are found near the surface of the ganglion but they may be deeper. The end disks are known to contain a granular cytoplasm. The above description is based on that of Huber (’96), which is the first account of these end disks in relation to spinal ganglion cells. The results have been documented by later work on mammals.

The sensory nerves terminate chiefly in free sensory endings, although a few Pacinian corpuscles are present. The sensory impulses in the frog are practically all of the grade of protopathic or vital sensibility. Thus they are general tactile, temperature, and pain impulses, with impulses of proprioceptive type, also, from terminations within muscles {Ramon y Cajal, ’88 ; von Kolliker, ’89 ; Dogiel, ’90 ; Huber and DeWitt, ’98; Hines, ’27, and others). The dendrites of the spinal ganglion cells as they pass to the skin are distributed in a segmental way {Eckhard, ’49 ; Sherrington, ’93). Their arrangement over the large posterior extremities of this animal indicates that the extension of the skin has had an influence on the distribution and relations of these peripheral branches. The nerve overlapping in the trunk is very great. Whether it is more pronounced than that in the extremities is as yet uncertain. In any case the overlapping in the extremities is very considerable (see fig. 87 ; Sherrington, ’93).

The probable presence of preganglionic fibers in the dorsal roots has been referred to earlier in this account. The neuraxes of the spinal ganglion cells enter the spinal cord near the so-called zona marginalis of the dorsal horns and there arrange into two bundles, a lateral one which primarily remains in the zona marginalis (see fig. 86) and a larger medial one which enters the dorsal funiculus. The zona marginalis is differentiated clearly for the first time phylogenetically in these forms {Keenan, ’29). Even here (Rana catesbyana) its fascicles are somewhat scattered but still recognizable as a tract. The substantia gelatinosa, which this zone borders on its lateral side and with which it is in synaptic relation, is not highly developed in the frog although present throughout the cord {Keenan, ’29). In the tadpole this medial bundle synapses with dendrites of motor cells which extend into the dorsal horn. As metamorphosis goes on, these dendrites are lost, and longer dendritic branches (sensitivo-motor collaterals of Ramdn y Cajal, ’09) extend from the fibers of this medial bundle into the ventral horns. After giving off collaterals at the level of entrance, the fibers of the medial bundle divide dichotomously and ascend and descend. The increase in fibers in the dorsal funiculi of the amphibians as compared with those in fishes has led to a wider separation of the dorsal horns in the former animals. Consequently the gray has

Fig. 87. A. The first research for the purpose of establishing the segmental innervation of the skin in the extremities of the frog. Eckhard, 1847.

B. The segmental innervation of the skin of the posterior extremities of the frog. Sherrington. Note the distribution. On the right extremity, segment VIII is isolated; on the left it is completely overlapped by segments VII, IX, and X.

187

188 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the form of an H in the amphibian cord in contrast to the inverted Y (a) which is its characteristic outline in the teleostean cord.

In fishes the dorsal funiculi comprise only 5 per cent to 6 per cent while in certain frogs they make up 13 per cent of the total white substance of the cord in the cervical region. In the bullfrog even a larger percentage (20 per cent) of the total white substance is to be found in the dorsal funiculi in the cervical region, although in the lumbar region they form only 13 per cent. This of course raises the question as to the causes underlying these differences. Sandmeyer (’92) found degenerated fibers near the dorsal midline above the level of the lesion, some even extending into the lower end of the medulla oblongata.^^ Yet it is doubtful whether the frontal accumulation, which the presence of these fibers would suggest, is nearly large enough to account for the increase in size of the dorsal funiculi at the anterior end of the cord. That other factors influence this increase was clearly indicated by Wallenberg (’07), who pointed out that in the frog the spinal trigeminal root and also a part of the vestibular and vagus roots descend for a considerable distance in the dorsal funiculi of the spinal cord. The vagus (and glossopharyngeal) roots reach as far as the second or third spinal segment, the vestibular reaches to the sixth spinal segment, while the trigeminal, running in the lateral part of the dorsal funicular area and gradually decreasing in size as it passes caudad, descends to the beginning of the lumbar enlargement. Consequently the dorsal funiculi do not consist wholly of dorsal root fibers. The greater proportional size of these funiculi in the cervical cord of the frog does not imply any great frontal accumulation of spinal root fibers from the various levels of the cord, such as is typical for mammals, but rather depends, for the most part, upon the presence in upper cord regions of descending bulbar fibers which decrease in amount caudally and lead to a corresponding decrease in size of the dorsal funiculi. If such descending fibers could be eliminated from the cervical cord, the dorsal funiculi would probably show but small increase in their cephalic as compared with their more caudal portions. It will be seen that even in reptiles but little frontal accumulation of dorsal root fibers occurs.

It is not possible at this time to enter into a detailed account of the connections of the secondary neurons of the cord in the frog. The well-controlled observations agree with conditions characteristic of the shark. There are a number of observations, less carefully controlled or extremely difficult of control — in part rather fantastic in character — which do not fit in with the present knowledge of the nervous system and apparently need further investigation. These will not be discussed at present.

The commissural cells, belonging to the earliest neurons of the spinal cord, appear to lie principally in the ventral horns, although certain of them are found dorsally near the midline (fig. 88). Neuraxes of these cells give rise to so-called arcuate fibers which decussate in the anterior or ventral commissure and then reach the ventrolateral area of the other side (fig. 88), where they turn, for the greater part, in a frontal direction. They form the coarse bundles which are indicated in figure 86. Sandmeyer (’92) has shown experimentally that these » Koppen (’88) may be consulted in regard to this point, also.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 189

contain both shorter and longer fibers. Through collaterals and stem fibers of this system, impulses reach the motor tegmentum of the medulla oblongata, the tectum, and possibly the caudal end of the diencephalon. As was stated previously, no true dorsal funicular nuclei — and consequently no medial lemniscus — are present in amphibians, but the ventrolateral bundle is supplemented by fibers from the dorsal horns of the opposite side throughout the levels of distribution of the spinal trigeminal tract. The “spinal lemniscus” system described

Intercalated Dors arcuate ScQB -mot. cells or com celts

Fig. 88. Position of motor neurons, arcuate fibers, and commissural neurons in an older Bufo larva. Sala.

by Herrick ('14) for Axolotl belongs to this ventrolateral fiber complex.^^ Collaterals and possibly stem fibers of this arcuate system turn caudalward after crossing and so transmit aboral reflexes (F. a. 1., fig. 85A). Collaterals from uncrossed fibers in the ventral, lateral, and dorsal funiculi synapse with neurons of the ventral and dorsal horns of the same side and, after a decussation through the commissura dorsalis, with those of the opposite side as well.

A spino-cerebellar path is present in these animals, although it is not so large as in the fishes because of the relatively small size of the cerebellum in these forms. These homolateral fibers arise in the frog from the mediodorsal region of the dorsal horn, pass to the dorsolateral funiculus of the same side, and ascend. They are

There is a difference of opinion as to the use of the word “lemniscus” for any part of this system, many workers preferring to reserve the name for fibers arising from nucleus gracilis and nucleus cuneatus (medial lemniscus), from secondary centers associated with the cochlear Vlllth (lateral lemniscus), and from the sensory Vth nucleus (trigeminal lemniscus).

190 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

joined in the medulla oblongata, as in Axolotl {Herrick), at the level of the frontal sensory trigeminal nucleus by fibers from that nucleus. The tract

decussates in part in the decussatio veli and ends chiefly (if not exclusively) in the corpus cerebelli.

From the preceding description it will be seen that the phylogenetically older tracts are well developed in amphibians, while the nuclei of the dorsal funiculi and the medial lemniscus are lacking. Consequently there can be little, if any, projection of finer sensibility from the body upon the thalamus. This fits well the conception that the sensations of pain, temperature, and general tactile sensibility (possibly a primitive muscle sense) which are to be found in the ventrolateral and lateral funiculi in mammals, as in amphibians, are phylogenetically old. The paths for tactile discrimination and for highly developed proprioceptive impulses (running in mammals through the dorsal funicular nuclei to the medial lemniscus and by way of that path to the thalamus) are phylogenetically much younger and consequently are not as yet organized into definite paths in amphibians.

The Spinal Cord of Reptiles

Due to the persistence of the tail in all reptiles and the retention, in contrast to the mammalian tail, of its primitive metamerism {Gegenbaur, ’98), the reptilian spinal cord extends throughout the entire vertebral canal, and the formation of a cauda equina or of a considerable filum terminale does not occur. This condition appears to be constant for all reptiles and, in these respects, the spinal cord of the reptile shows a more primitive character than does that of the frog. In other respects there is much variation in the form of the spinal cord within the reptilian class. Reptiles fall into three main groups ; the saurians and hydrosaurians, which are provided with limb

Fig 89 A The brain and spinal and trunk musculature ; the serpents, which

‘■"b brain a^d’'” pinat corTof a musculature ; and the turtles,

turtle Bojanxe (Coirparetvith fig 90 ) which lack trunk muscles but have well

TIIE COMP.-VRATIVE ANATOMY OF THE SPINAL CORD 191

developed muscles for the head, neck, tail; and extremities. The influence of these variations in body form upon the nervous system is visible macroscopically. The spinal cords of hydrosaurians (fig. 91B) show considerable enlargements in the cervical and lumbar regions. In certain large fossil dinosaurs, having enormously developed posterior extremities, the cavity occupied by the lumbar enlargement within the vertebral column surpasses in volume the skull cavity. In serpents the cervical and lumbar enlargements are lacking (fig. 91 A), as is to be expected in the absence of limbs. In turtles there is a remarkable decrease in the size of the cord between the two enlargements, due not only to the absence of body musculature but also to the relatively small sensory supply to the shell, so that both dorsal and ventral roots of the nerves, as well as dorsal and ventral horns within the gray of the cord, are considerably reduced. It is to be emphasized, however, that the shell of the turtle has a sensory innervation, although this is scanty. Consequently the dorsal roots are less reduced than the ventral, which in this region carry only preganglionics in the turtle (see fig. 90).

There is no alternation of dorsal and ventral roots in reptiles. Both emerge at the same transverse level (fig. 90). The cells of origin of the uncrossed ventral roots, which have a position similar to the homologous cells in amphibians, form groups which are somewhat removed from the central canal and which border more or less on the white substance. Certain of the dendrites run through the entire white substance and form a marginal plexus which, however, is not so large as in amphibians, being confined to the lateral and ventral parts of the cord (see Ramon y Cajal, '91, and Banchi, ’03, and fig. 93). The cells which contribute to this marginal or perimeduUury plexus are of various sorts : ventral horn cells, funicular cells, anterior or ventral commissure cells, cells of von Lenhoss^k in the cervical cord, and possibly other neurons also. The arrangement of the ventral horn neurons varies with the animal and with the level under consideration. Terni (’20) has called attention to the possibility (as had Ayers for Amphioxus and Bardeen for selachians) that the ventral roots may not be limited to one myotome only, but may give some fibers to adjacent myotomes. In the thoracic cord in turtles, where cells of origin for somatic efferent fibers are absent due to the lack of trunk musculature, the ventral horns are very narrow and such cells as do occur are commissural cells, various intrinsic cells, and possibly cells of origin for preganglionic fibers. In the remaining regions of the turtle cord two major subdivisions of motor nuclei are found — a more medial group, consisting of the cell bodies of neurons for neck and tail musculature, and more lateral groups, particularly well developed in this form, which are present in cervical and lumbar enlargements and are concerned with the motor innervation of the limbs.

In the serpents the gray substance has a very regular arrangement, the ventral horns showing some similarity to those of the shark (compare fig. 91 A with fig. 77). These motor cells are probably comparable to the medial group described for turtles and crocodiles.

In crocodiles a medial group for the innervation of trunk musculature is present throughout the length of the cord, as in the serpents, while lateral groups occur in the lumbar and cervical enlargements, as in the turtles. The spinal cord of the

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 193

crocodile presents several peculiarities. In the first place the eccentric frontal position of the central canal is striking. Figure 91B shows that the gray substance of the ventral horns lies far behind this canal. The cause of this is unknown. A second peculiarity is found in the presence of the nuclei marginales (n. m.), which were described first by Gaskell (’85) in the alligator. They have been demonstrated in reptilian spinal cords by a number of observers, among whom may be mentioned von Kolliker-^ (’02),

Slerzi {’Oi), Shimada (’12), and Terni (’26). They have been seen in a wide variety of reptiles, such as turtles {Aliens Kappers,

’20), lizards {von Kolliker,

’02; Terni, ’21a and ’26, and others), and snakes {Shitnada, ’12). The most satisfactory account to date appears to be that of Terni (’26), who studied them in a late embryo of Gongylus ocellatus. Here they appear as groups of cells segmentally arranged in the spinal cord, throughout both trunk and tail.

They are situated near the surface in the ventrolateral 91 Cross section through the cervical cord of Python

portion of the lateral white reticulatus

, B Cross section through the cervical gord of Crocodilus norosus

funiculus but do not appear to form macroscopic

eminences such as are seen in avian forms. The neurons constituting the cell group (gruppi cellular! periferici. Term) are spindle shaped and have long dendrites, many of which extend longitudinally, parallel to the fiber bundles of the lateral white funiculus. Term was able to trace the neuraxes of certain of these cells into the homolaterai lateral white funiculus but could follow them for only a short distance. Other neuraxes appear to cross as ventral

^ Von Kolliker christened these cell masses the nuclei of Hofmann, after the technician who first drew his attention to them in birds It would seem more correct to designate them the nuclei of Gaskell, since this latter observer had discovered and described them many years previously.

194 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

white commissure fibers, probably to enter the contralateral white matter of the cord.

These marginal nuclei are situated in the region where the marginal dendritic plexus is particularly well developed in reptiles. It is probable that they are formed of neurons the cell bodies of which have migrated from the spinal (possibly from ventral horn) gray along their dendrites toward the source of their stimulation, according to the principles of neurobiotaxis. This does not necessarily imply that they are cells of origin for motor fibers — in fact, the evidence at present is against such an interpretation, since they have never been seen to give rise to motor fibers and their cell outlines are dissimilar to those of the typical multipolar motor neurons. Possibly they are cells of origin for ascending secondary paths. These cells are considered again and much more fully under the account of the spinal cord of birds, where the marginal nuclei are much better developed (see page 206).

That the ventral roots of reptiles contain preganglionic as well as somatic motor fibers is generally accepted on the basis of experimental work, some of the early studies in this field having been carried out by Gaskell (’86 and elsewhere) on the alligator and other reptiles. An excellent illustration of their presence from the anatomic point of view is to be found in the turtle, where the absence of myotomic muscle in the thoracic region limits the ventral root to fibers of the preganglionic type. So far as is known, preganglionic fibers, such as occur in the dorsal roots of lower vertebrates, have never been demonstrated for reptiles. Likewise, the cells of origin for the preganglionic fibers have not been identified with certainty for all reptiles. In the turtle they must occupy a fairly central position, for here only the central portions of the gray substance are preserved in the thoracic region. In Pseudopus pall. Takahashi (’13) found a group of cells which he regarded as the nuclei of origin for the preganglionics in the lower cervical cord. They lie between the dorsal and ventral horns and, in the thoracic region, form a slight lateral protrusion suggestive of the so-called lateral horn described for mammals. With the use of the methylene-blue method, it has been demonstrated that in turtles (Chrysemys picta and Chelhydra serpentina) the preganglionic visceral efferent fibers terminate in relatively complex intracapsular, pericellular plexuses surrounding the cell bodies of the multipolar sympathetic neurons {Huber, ’99). Kuntz (’ll, ’29) described the development of the reptilian sympathetic system.

An interesting series of dissections of the reptilian sympathetic system has been given by Hirl (’21), These dissections were made on specially prepared material of Lacerta agilis and Lacerta ocellata, Chamaeleo vulgaris, Hatteria punctata, and Varanus nicolitus. (For an account of the vagus nerve and cervical sympathetic in embryonic and adult Testudo graeca and Emys europaea see Francescon, ’31.) Hirl's findings agree with the earlier account of Fischer for these regions. The following account of the gross anatomy of the reptilian sympathetic system is based directly on the descriptions given by Hirt except where otherwise stated. The original paper must be consulted for the details and the variations in the different lizards. It must be emphasized that the scheme built up by Hirt

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 195

is based on gross dissections and does not take into account the finer anatomy of the system.

Following the earlier work of Fischer, Hirt recognized a superficial and a deep portion of the head sympathetics. The superficial portion is represented by a strand of fibers which passes from the second trigeminal branch to the facial (with which it may or may not be connected) and then to the petrosal ganglion of the glossopharyngeal. After either entering into relation with this ganglion or passing ventral to it, the superficial portion extends caudalward to unite with the deep portion and to become continuous with the superficial portion of the cervical sympathetic.

The superficial portion of the cervical sympathetic usually extends caudalward to the first, but occasionally to the second thoracic chain ganglion ; it may have an anastomosis with the ganglion on the trunk of the vagus. The deep cervical portion of the sympathetic extends between the vagus and hypoglossal, on the one hand, and the first thoracic chain ganglion, on the other hand. Its course is through the deep neck musculature and fine twigs connect it with the cervical nerves and with the deep portion of the head sympathetics (at least in the Chameleon).

The first thoracic chain ganglion gives off a considerable number of branches to the brachial plexus, to the first cervical nerve, and to the brachial and carotid arteries. Extending caudalward from the first thoracic ganglion are a series of ganglia numbering from 17 to 27, the exact number depending on the reptile studied, interconnected and with most of the ganglia of the chain demonstrably connected to spinal nerves by a single ramus or double rami communicantes (although white and gray rami cannot be distinguished macroscopically). However, certain ganglia, particularly in the lower thoracic and abdominal region, belong in the chain, but in saurians, at least, appear to lack direct connection with the spinal nerves. Hirt called attention to the fact that an unusually thin band of fibers connects the fourth and fifth chain ganglia and interpreted this as indicating that most of the fibers pass cephalad above that region and caudalward below it. He noted, also, the presence of almost microscopic clumps of sympathetic cells within the rami communicantes. Such cells apparently had not migrated far enough to reach the sympathetic ganglia. Their position is interesting in view of the results obtained by Riquier (T4) in his study of the thoracic sympathetics in turtles, particularly Thalassochelys caretta. Earlier work had appeared to indicate that turtles lacked thoracic chain ganglia, but Riquier believed that such ganglia are present but fused more or less completely with the spinal ganglia (the second to the seventh completely, the eighth to the tenth partly). Hirt gave in some detail the distribution of the sympathetic fibers from the trunk ganglia as these are revealed by gross dissection. For such details the original paper must be consulted. It may be stated that the thoracic chain ganglia send fibers to the heart, the stomach, and the intercostal arteries. Those to the heart arise from the upper thoracic chain ganglia and pass partly directly to the heart and partly form a component of the superficial cervical sympathetic to the vagus region, where they accompany bundles of that nerve

196 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

to the heart. The abdominal chain ganglia (from the fourteenth to the eighteenth) send out fibers which form a plexus and terminate in the gut. The abdominal ganglia also supply the reproductive organs. Spanner (’29) studied the abdominal system in Anguis fragilis with particular emphasis on the innervation of the kidneys in this form. A special pelvic ganglion (in the Chameleon, for example) supplies branches to the bladder and to the cloaca or fibers to these regions may pass directly from the chain ganglia (the twentieth to the twenty-fifth).

Caudally, the chains of the two sides unite to form a ganglion impar, from which two very fine branches, with small enlargements apparent along their extent, continue caudalward. These two branches represent the tail sympathetic chain as described by Term (’22), and the small enlargements are probably the segmental ganglia recognized by him. This latter observer (’21 and ’22) believed that from this caudal portion of the system fibers are distributed to the striated muscles of the tail. In his study of embryonic material of Gongylus he found such fibers within the voluntary muscle at a time when no vasomotor fibers were demonstrable.

Throughout the whole cervical cord of the lizard very coarse efferent fibers can be seen leaving the dorsal roots as far as the eighth cervical segment. Their cells of origin were described by Beccari (’13, ’14) as lying in the medial part of the ventral horn. They are multipolar in character, are termed the cells and fibers of von Lenhoss4k (see bibliography for birds), and as yet have never been described in forms other than reptiles and birds. Apparently the fibers are not present in mammals, but since the spinal portion of the accessory nerve arises as far caudad as the seventh cervical segment and its fibers pass cephalad through the vertebral canal outside of the cord to join the cranial portion of the nerve, Beccari suggested that the fibers of von Lenhossek are the reptilian and avian equivalents of the mammalian accessory nerve. However, the peripheral terminations of fibers of von Lenhossek have not been traced in either birds or reptiles and a musculature extending as far as the level of the eighth cervical nerve, which is in any way related to the trapezius, has not been observed in these forms. Banchi (’03) believed that a similar cell group may be found in the lumbar cord in turtles. The functional significance of the cells and fibers of von Lenhossek in lizards or birds cannot be regarded as entirely settled at present. This matter is discussed again in Chapter V.

The remaining components of the dorsal roots are somatic afferent and visceral afferent in character. In the full-grown reptile, the cells of origin of both types of fibers are situated in the spinal ganglia. Intramedullary dorsal root cells (cells of Rohon-Beard) have been pointed out in the embryo of Trophidonotus, but they persist for only a short time. The spinal ganglion cells are chiefly unipolar in character, but are often lobulated and may be very irregular in form (Levi, ’08). Both ganglia and ganglion cells decrease in size toward the caudal end of the animal (Terni, ’22). This last-mentioned author has pointed out that in regeneration of the tail in Gongylus, the volume of such spinal ganglion cells and their peripheral processes increase greatly and this increase is associated with new and richer peripheral connections.

Fig. 92. Skin segments in the lizard (dorsal view at left,

ventral view at right).

van Trigt.

197

198 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The sensory nerves in reptiles have chiefly free sensory, neuromuscular, and tactile endings. The only highly developed corpuscles are those of Pacini, and Rollett and Sachs. Van Trigt (’17), who studied the finer peripheral distribution of the dorsal root nerves, showed that important overlappings — up to two-thirds of a segment — occur in the skin. The shape of the trunk dermatome is peculiar, since it is broader on the ventral than on the dorsal side, a reversal of the condition found in most vertebrates. S. de Boer (’18) regarded this as due to the fact that the ventral side of the body carries scales which reduce its susceptibility to stimuli so that in reptiles, in contrast to most animals, the dorsal side can be stimulated more readily and so receives the greater increase in nerve terminations. On the extremities the dermatomes show considerable displacement and less overlap (fig. 92). (For the de Boer reference see bibliography for the mammalian spinal cord.)

The substantia gelatinosa caps the body of the dorsal horn as in mammals and, in crocodiles at least, extends along the horn almost as far as the medial septum. It increases, as does the marginal zone of Lissauer, in the sacral region. The deeper part is more compact than more superficial portions of the substantia gelatinosa, since the lateral region is broken into by coarse fascicles which in certain regions separate it from the marginal zone (Lissauer’s tract). The above account is based on that given by Keenan (’29).

There appears to be some variation in the relations of the dorsal root fibers after their entrance into the cord. In crocodiles the relations are very similar to those in amphibians, the dorsal root splitting up into (1) a coarser fibered dorsomedial bundle, from which collaterals are given off to the ventral horn (sensitivo-motor collaterals, Ramdn y Cajal, ’09) and also to cells of the dorsal horn, and (2) a finer, partly unmedullated fiber path which runs in the marginal region over the dorsal horn and terminates in relation with the underlying gray. Presumably the dorsomedial bundle is homologous with the dorsal or posterior funiculus of higher forms, while the lateral is probably comparable to Lissauer’s tract. In certain reptiles somewhat different relations have been described (Tropidonotus natrix, Retzius, ’94 — fig. 93, and Gongylus ocellatus, Terni). Term (’22a) found that upon penetration of the cord, the dorsal root fibers divide into two bundles, one dorsomedial and the other ventrolateral. The former enters the dorsal funiculus ; here the fibers show T-shaped divisions, the branches ascending and descending. Short collaterals enter the dorsal horn and some fibers synapse with the large dorsal commissural cells but more pass to the ventral horn. The ventrolateral bundle, consisting of coarse fibers, gives collaterals for sensory-motor reflexes and synapses with the large ventral commissure cells which give origin to fibers of the medial longitudinal fasciculus in the cord.*^ A considerable part of the dorsal root fibers turns into the lateral funiculus in turtles {Banchi, ’03 ; de Lange, ’17 ; see also fig. 79), and the bundle is even larger in

“ It may be that the peculiar method of locomotion of serpents and of Angius which, as in fish, necessitates the movement of the whole body is responsible for the more lateral position of this group of dorsal root fibers which supply collaterals to the motor neurons. Such an explanation, however, would not make clear the lateral position of these fibers in turtles.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 199

serpents (Ramon ij Cajal, ’91 ; Rcizius, '94 iind ’98; van Gehnchlcn, ’97, and dc Lange, '17). Tlie caliber of these lateral fibers and the fact that the collaterals to the ventral horn arise from them (at least in lizards and serpents, see fig. 93) suggest strongly that this bundle of laterally placed dorsal root fibers, as it occurs in turtles, lizards, and serpents, is probably not a lateral extension of the finer lateral or marginal bundle of amphibians, crocodiles, and mammals but a displaced portion of the dorsal funiculus such as was described for certain teleosts (sea robin, Herrick, ’07 ; see bibliography for ganoids and teleosts).

Fio. 03. The root fibers (R. post.), in the lateral funiculus (R. (at.), and in tlic dorsal funiculus (Fuse, ilors.) in Tropidonotus nalrix. Rdsius. C.C., cominissuro cells.

The position of dorsal root fibers in the lateral funiculus must be taken into account in all measurements which attempt to establish a comparison between the number of such ascending fibers in the cords of other animals and that in serpents, turtles, and at least certain lizards. It is self-evident, then, that a comparison of the dorsal funiculi of serpents or turtles on the one hand with those of frogs on the other can lead to no valid conclusions in the matter. The error will be increased considerably by the presence in the dorsal funiculi of the frog of a considerable number of descending tracts (see the account of the amphibian spinal cord). Thus it is not surprising that such a comparison should show relatively larger dorsal funiculi in the bullfrog (20 per cent) than in the turtle (Dammonia, 16.2 per cent), snake (11.9 per cent), and Chameleon (12 per cent).=“ However, a measurement of the homologous columns in the tailed amphibians (Siren), where the bulbar fibers in the dorsal funiculi are much less numerous, gives a higher percentage of fibers for reptiles.

For the first time phylogenetically, an accumulation frontalward of fibers from lower levels of the cord is clearly evident in reptiles, so that a readily

“ Animals of approximately equal size were chosen ; this is a necessaiy condition in establishing such comparisons (see page 212).

200 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

definable fasciculus gracilis and fasciculus cuneatus may be said to be formed, although there is the beginning of such an accumulation in amphibians. In the former animals, too, for the first time appear dorsal funicular nuclear masses comparable to the mammalian nucleus gracilis (nucleus of Goll) and nucleus cuneatus (nucleus of Burdach) . Such nuclei have been recognized by Christensen (’17), Zeehandelaar (’21), Huber and Crosby (’26), Shanklin (’30), and others. These nuclear masses are regarded generally as the nuclei of termination for muscle sensibihty and finer tactile discrimination and the nuclei of origin for the pathway (medial lemniscus) by which these impulses reach thalamic and, after a synapse, cortical centers. The neothalamic areas and somatic telencephalic centers show a tremendous increase in development in reptiles as compared with that of homologous regions in amphibians, and the appearance of a pathway between these bulbar nuclei and the appropriate thalamic centers is to be expected. Indications of a medial lemniscus in reptiles have been found by several observers, but a complete and satisfactory account is yet to be given. Christensen (’17) stated that, in Sphenodon, arcuate fibers arising from nucleus gracilis and nucleus cuneatus cross the midhne and accumulate in a ventrolateral position, which they occupy into the midbrain regions. Christensen believed they pass to the thalamus, but neither his figures nor description makes clear their course beyond the midbrain or their nuclei of termination. Huber and Crosby (’26) were able to trace crossed fibers arising from the nucleus gracilis and nucleus cuneatus, and accumulating into a bundle which appeared to turn forward, but they were unable to trace it to the thalamus. In Chameleon, Shanklin (’30) described such ascending fibers as far as the midbrain, but he also could not follow them to thalamic centers. The midbrain relations of the bundle thus described by Shanklin certainly suggest the relations of a medial lemniscus system as these are known for mammalian forms. It appears probable that such a fiber system is present in reptiles and that its complete demonstration only awaits material cut in an exceptionally favorable plane. From the evolutionary standpoint, one of the most important advances in the reptilian cord is this accumulation of dorsal root fibers in the dorsal funiculi and their termination in associated medullary nuclei. Correlated with this great advance in the central connections mediating finer tactile discrimination is the development peripherally of special sensory corpuscles.

A large spino-cerebellar tract has been described for the alligator (Alligator sclerops, de Lange, ’17 ; Alligator mississippiensis, Huber and Crosby, ’26). This tract lies in the cord just under the dorsal horn, enters the medulla oblongata in relation with the spinal tract and nucleus of the trigeminal, then swings dorsalward, lateral to this nucleus and its associated fibers, and enters the cerebellum. It carries crossed and uncrossed fibers. The cells of origin for what appears to be a homologous tract were figured by Banchi (’03) for Emys and by Terni (’20) for Gongylus ocellatus. Both workers designated them as Clarke’s nucleus and Terni, in particular, stated that they are probably the reptilian equivalent of the mammalian nucleus of that name. If so, this tract should probably be regarded as the dorsal spino-cerebellar tract. According to Christensen (’17) the ventral

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 201

spino-cerebellar tract enters the anterior part of the cerebellum, the dorsal spinocerebellar tract the posterior part, in Sphenodon. In Alligator mississippiensis a small bundle of spino-cerebellar fibers, a part at least of the ventral spino-cerebellar system, accojnpanicsthc spino-mesencephalic tract into medullar regions and then swings dorsalward to enter the cerebellum. Larsell (’20) described a ventral spino-cerebellar tract as the largest tract connected wdth the cerebellum in certain lizards (.Vnniella, Sceloporus and Phrynosoma, Gerrhonotus and Thamnophis). This runs with the spino-tectal tract in its course through the bulb, then arches around the trigeminal roots and, swinging dorsally, distributes to the granular layer of the cerebellum. There is more or less decussation to the oppo•site side, varying apparently with the animal under consideration. This same observer described a dorsal spino-cerebellar system but was not entirely certain tis to the origin of any of its components from the cord. In Chameleon, Shanklm (’30) found a dorsal spino-cerebellar sj'stem containing crossed and uncrossed components but with the cmphiisis on the crossed fibers, and a ventral spinocerebellar tract which, from its general relations, appeared to be comparable to the spino-cerebelLar component accompanying the spino-tectal or spino-mesencephalic tract in the Alligator. A similar tract has been figured by Frederikse (’31) in his atlas of the lizard’s brain. For other accounts of these systems by Hindenach (’31), Larsdl (’33), and others the text and bibliography on the reptilian cerebellum should be consulted (Chapter VII ; also figs. 165 and 202).

The secondary sensory tract of Edinger (’OS) — or the spino-mesencephalic (spino-tectal) tract — which carries pain, temperature, and a generalized sense of touch, shows considerable incrca.se in size over the homologous tract in amphibians but has es.sentially the same characteristics. It terminates in the tectum (and tegmentum) of the midbrain ; its direct connection with thalamic centers has not been established definitely. With a few exceptions, the connections and relations of the reptilian arcuate, funicular, and commissural fibers show no pronounced differences from the connections and relations of homologous fibers of the amphibian cord ; consequently a detailed account of them for reptiles will be omitted. However, attention is drawn to the presence in these latter animals of an accessory commissure of iilauthner in addition to the usual anterior or ventral commissure of the cord. In this particular, reptiles resemble fishes (in which a similar accessory commi-ssure is found) rather than amphibians in which such a commissure is lacking. The reptilian accessory commissure carries neuraxes of commissural and collaterals of funicular cells, as well as decussating dendrites of motor neurons.

.-Vccording to Terni (’22a) the longitudinal fibers which lie dorsal to the accessory commissure form the true fasciculus longitudinalis medialis of the cord ; this fasciculus carries the oldest spinal refle.x paths. He described large ascending arcuate fibers, arising from large cells in the dorsal horn (3-4 in each segment). These cells (grand! cellule commissurali dorsali of Terni), which develop early in embryologic life, are distributed symmetrically in the dorsal gray throughout the length of the spinal cord. Their major dendrites distribute as a dorsal group to the region of the entering dorsal root, as ventral groups to the ventral horn and

202 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the lateral white funiculus, and as a medial group to the lateral funiculus of the opposite side. The neuraxes of the cells swing ventromedialward and ultimately cross without bifurcation and enter the fasciculus of the opposite side. The fibers are probably comparable to the arcuate fibers of Amphioxus. In addition to this component, Terni described crossed and uncrossed neuraxes of ventral funicular cells, the cell bodies of which lie dorsal and medial to the ventral horn cells. The dendrites of such cells receive impulses from the dorsal root fibers. The neuraxes may enter the fasciculus longitudinalis medialis of the same or opposite side, there to bifurcate and descend, or such a fiber may cross and, after bifurcation near the midline, send a descending fiber into the anterior white funiculus and an ascending fiber to the medial longitudinal fasciculus (Terni). Terni believed also that fibers emerging with the ventral root may penetrate, after their origin from ventral horn cells, the medial longitudinal fasciculus and run in it a short distance before their emergence from the cord.

Another peculiarity of the reptilian cord is found in the frequent presence of two dorsal commissures, which not only carry collaterals of dorsal funicular fibers, but also contain dendrites and even cells associated with the commissural components. The two bundles in certain reptiles are united to form a single dorsal commissure. Particularly in Emys (Banchi, ’03), but also in Lacerta, the cell bodies of certain of the commissural cells have a distinctly dorsomedial position, some of them lying practically in the midline. This is a reminder of the dorsal position of many of the commissural cells of teleosts and of the dorsomedial cells of Amphioxus.

The major descending paths to the cord arise in the medulla oblongata and, of these, the vestibulo-spinal pathos are the best developed (for details see Chapter IV) . They lie in the ventral and the ventrolateral funiculi. A rich system of tecto-bulbar fibers carries impulses from the optic tectum to the medulla oblongata. The fibers synapse with the reticular cells and then reach the cord by way of reticulo-spinal paths. Direct tecto-spinal paths have not been demonstrated satisfactorily for reptiles. Olfactory impulses are conveyed, after passing several synapses, to the reticular cells of medulla oblongata and midbrain and thus indirectly to the cord. Secondary neurons from the chief sensory nucleus of the trigeminal also synapse with reticular cells.

The supporting tissue in reptiles resembles greatly that of mammals. Distinct neuroglia fibers appear here first. Deep connective tissue septa, carrying rich capillary blood plexuses, bring a rich blood supply to the inner portions of the cord.

The Spinal Cord of Birds

The avian spinal cord, like the reptilian, extends throughout the whole length of the vertebral canal and has neither a cauda equina nor a pronounced filum terminale. The cord is absent only in a small part of the tail region, where remmants of atrophied vertebrae are found. The spinal cord of birds differs from that of the majority of reptiles in having a greater number of cervical and of lumbosacral segments and a lesser number of thoracic segments. For example.

Tm-: COMPARATIVE ANATOiMY OF TtlE SPINAL CORD 203

nerves, according to

the ostrich has v'il pairs of nerves, of wliich 15 are cervical, 8 are thoracic, 19 are luinbosacral, and 9 are coccygeal (Slrccler, ’04). (For the pigeon see fig.’ 94.)

Two distinct enlargements are present in all birds (even the penguin), and of these the cervical one is usually the larger. In the ostrich the cervical enlargement is situated between the first and fourth thoracic Slrcctcr, a statement which almost agrees with that of Furbrinijcr (’97), who stated that the brachial plexus in this bird arises between the seventeenth and twenty-first segments of the cord. The lumbosacral enlargement, although usually smaller than the cervical, is occasionally greater in such birds as the ostrich, where there is a marked development of the musculature of the posterior extremities. In this bird it extends from the third to the fourteenth lumbosacral segment and here the cord has a conspicuously large transverse diameter. Dorsall}', in this region, is the so-called sinus rhomboidalis sacralis, which is present in all birds.

This sinus (fig. 94B) might be termed more appropriately the sinus lumbosacralis, since it is found chiefly in the lumbar region. In the ostrich it lies between the seventh and the twelfth lumbosacral segments {Slrccler,

’04), forming a secondary gap between the two dorsal funicular regions of the cord.

It appears to be formed under the influence of the unusually large sensory roots of the sciatic, the fibers entering on both sides of its upper edge (fig. 94B). Similar causes assist in producing the fourth ventricle of the medulla oblongata, but the sinus rhomboidalis, or lumbosacralis, cannot be considered as a sacral ventricle, for, as has been clearly shown, the central canal continues through the region {Hanson-Pruss, ’23 ; Ariens Kappers, ’24, J. F. Huber, Dis

22

23

24

Fia. 91. yl. Drawing to scale of (lie brain and bpinal cord of the pigeon, showing the number of the nerve roots, tlie formation of the plexuses, aud ,the nerve roots involved in these plexuses. John F. ll uber.

B. Drawing showing the position of the lumbo-sacral sinus in the pigeon in relation to the nerve roots. John F. Huber.

204 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

sertation ’33). The sinus lumbosacralis is filled with a semitransparent substance composed of vacuolated cells containing glycogen {Terni, ’26). Some observers regard these cells as derivatives of neuroglia tissue from the posterior septum {Streeter, ’04 ; Terni, ’24) ; others (Hanson-Pruss, ’23 ; Ariens Rappers, ’24) consider them to be of leptomeningeal origin. The central canal passes through this tissue (fig. 95C) and the region dorsal to this canal is traversed by myelinated commissural fibers extending between the dorsal horns of the two sides.

A further peculiarity of this region is the enlargements of the ventral horns, visible on the surface as the eminentiae ventrales. The motor roots run at the lateral side of these eminences. Dorsolateral to these protuberances smaller elevations occur in the ostrich, which are caused by the nuclei marginales {Gaskell, ’85 ; von Kdlliker, ’01, ’02, ’02a). These will be considered later. In the cassowary analogous conditions are found, and the elevations produced by the marginal nuclei are present in certain other birds, although they may vary somewhat in size. The spinal cord decreases rapidly in diameter behind the lumbar enlargement.

The ventral and dorsal roots have the arrangement usual for higher vertebrates, the ventral leaving the cord in many minute bundles, the dorsal in several larger fascicles. Centrally there is a striking contrast between the size of the dorsal and the ventral gray horns, the latter being very much the larger. A reversal of this condition is sometimes found in the upper cervical cord (fig. 99), due to the development of the spinal trigeminal nucleus in that region.

The cell groups within the spinal cord are more clearly defined than in reptiles. These groups have been studied carefully by Streeter (’04) for the ostrich. He divided the nuclear masses into five groups — a lateral, a central, a peripheral, a commissural, and a dorsal group. These groups will be considered with the discussion of the appropriate regions of the cord beginning with the ventral horn regions.

The lateral group consists of ventrolateral, dorsolateral, and lateral cells, but in the ostrich (according to Streeter) the differentiation of sub-groups is not sufficiently clear to recognize these cells as distinct nuclear masses. Streeter stated that in sections through the lumbosacral enlargement, the median nuclei or columns of mammalian spinal cord (which supply trunk musculature) are represented by the ventromedian cells, while the remaining more laterally situated cells fall into anterior, middle, and posterior nuclei or columns, in all probability corresponding to the ventrolateral, dorsolateral, and retrodorsolateral columns of higher forms, which supply extremity musculature. The lateral and more particularly the dorsolateral cells are especially numerous in the cervical and lumbar enlargements (figs. 95A, B, and C) . To the lateral nucleus of Streeter probably belong the cells of origin for the motor fibers to the pectoralis major muscle. These cells constitute the “grand noyau central’’ described by Sana (’14) in his work on pigeon and chicken. Such a cell group naturally would be less developed in runners than in flyers. There is reason to believe that in many birds the nuclear differentiation of the ventral horn gray shows a closer approach to mammalian conditions than appears to be the case in the ostrich.

Fig. 95. A. Section through the spinal cord of the pigeon at level of emergence of the 12th nerve, showing the development of the lateral column of motor cells through the cervical enlargement and the dorsal magnocellular column.

B. Section through the thoracic cord of the pigeon between the levels of emergence of the 16th and the 17th nerves. Note the position of the preganglionic column described by Term, also the great width of the ventral as compared with the dorsal funiculus at this level.

C. Section through the lumbo-sacral enlargement in the pigeon cord between the level of emergence of the 24th and 25th nerves, showing especially the rhomboidal or lumbo-sacral sinus, the accessory lobes of Lachi, and the lateral column of motor cells. John F. Huber.

205

206 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

All of the neurons of the ventral horn group are large (.02 to .04 mm. in diameter in the ostrich) and multipolar in character. Various illustrations in the Uterature indicate that these cells are comparable in general appearance to the homologous cells of mammals (for example, see the illustrations of Martin, ’95 ; Streeter, ’04; Ramon y Cajal, ’09, and Term, ’23, ’26). The dendrites of the ventral horn cells form a marginal plexus in avian embryos, particularly along the ventrolateral margin, but in full-grown animals little or nothing is left of this plexus.

The central group as described by Streeter for the ostrich is not comparable to the central group of manunals but consists of cells of var3dng sizes (but for the most part smaller than the ventral horn neurons) which are scattered through the region of junction of the dorsal and ventral horns. Giant cells, .03 to .09 mm. in diameter, with round or four-sided cell bodies, containing rather fine, lightly stained Nissl substance, are found in the central group in the ostrich. Such neurons are most plentiful in the lumbosacral enlargement, particularly in its upper portion, according to Streeter, but are also found immediately above the cervical enlargement.

Marginal nuclei were considered briefly in the discussion of the reptilian cord. Marginal nuclei composed of paragriseal nerve cells are particularly well developed in the avian cord. The account here given of the marginal nuclei of the avian cord is taken largely from unpublished observations made by J. F. Huher (Laboratory of Comparative Neurology, University of Michigan) and, as prepared by him, is related to observations recorded in the literature. The expression “paragriseal nerve cells ’’ as here used is meant to include all nerve-cell bodies definitely lying outside of the gray matter of the spinal cord. Such outlying cells were noted, according to Sherrington (’90), by Beisso as early as 1873 in the o.x, in the cord of which he saw cells in the white substance near the gray matter. Gaskell (’85) was the first worker to observe cells at the periphery of the cord, and he described a discontinuous peripheral group of ganglion cells in a ventrolateral position in the spinal cord of the alligator. Four years later he (Gaskell, ’89) mentioned the same group of cells in crocodiles and birds and figured them in a section of the cord of an eight-day chick.

Such cells were later more fully discussed by Lachi (’89), who described five to eight pairs of “Lobi Accessori” projecting near the attachment of the denticulate Ugament and corresponding to the limits of the metameres in the lumbar enlargement of the bird cord. He spoke of these masses as being made up of a gelatinous substance similar to that in the rhomboid sinus, with cells resembling ventral horn cells in the substance. He also observed similar cells scattered at the union of the ventral and lateral funiculi at the levels between the accessory lobes, which impressed him as being links between the ventral horn and the accessory lobes.

Von Lenhossck (’94, ’95) described cells in the boundary zone between the ventral and the lateral funiculus in silver-impregnated sections of a nine-day chick cord, which showed the processes of these cells to pass through the anterior or ventral commi.ssure into the ventral funiculus of the opposite side. These cells he believed to be the same as fusiform cells observed in the external zone of the

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 207

lateral funiculus of a sixteen-day chick cervical cord and referred to by Ramon y Cajal (’93). He divided them into several types, but considered them all as dislocated commissural cells, a view concurred in by Ramon y Cajal (’99 and ’09).

A very satisfactory account of the outlying cells in the spinal cord of reptiles and of birds was given by von Kolliker (’02), who named the cells he described “Hofmann’s Nuclei” after his technician, P. Hofmann, who called his attention to these cell groups. Von Kolliker made observations on quail, dove, and embryonic and adult chicken preparations. He described the Hofmann nuclei as superficial, just beneath the pia and just dorsal to the attachment of the denticulate ligament, and as having a definite segmental arrangement, and divided them into two groups, the “ Hofmann’sche Grosskerne and Kleinkerne” or the major and minor nuclei of Hofmann. The major nuclei are found in the lumbo-sacral region of the cord, bulge out from the cord, and present a relatively loose glial web in which are embedded multipolar cells (10-16-27At) similar to the motor cells of the ventral horn (20-30-45ai). The minor nuclei are found in the lateral portion of the cervical, thoracic, and lower sacral regions, do not bulge from the cord but are groups of cells, some very similar to and some smaller than those of the major nuclei. He believed the Hofmann nuclei to come ontogenetically from the ventral horn mass but suggested three possibilities as to their function : (1) That they are cells of origin of preganglionic fibers ; (2) that they are cells of origin for motor fibers; (3) that they are ventral commissural cells. This last view is supported by the work of von Lenhossek and Ramdn y Cajal. Von Kolliker also mentioned cells near the origin of the ventral root. Sterzi (’04) described peripheral cells in reptiles, considering them as motor elements.

Streeter (’04) gave a discussion of the outlying cells as present in the ostrich. He quite definitely followed von Kolliker’s description and grouped the cells into major marginal nuclei, minor marginal nuclei, and scattered cells. There are sLx pairs of major nuclei seen macroscopically, one pair at the level of each sulcus transversus from the thirtieth to the thirty-sixth segment, and just dorsal to the ligamentum denticulatum. The major nuclei cells, in a glial framework, are like those of the lateral group of the ventral horn except that they are only J to ^ as large. His description of the minor nuclei was similar to von Kolliker' s concerning their location, but he said that the cells were small and not definitely multipolar. Streeter described one or two cells like ventral horn cells in size and shape in most sections through the lumbosacral region, lying among the fibers at the periphery of the ventrolateral column near the major nuclei. He did not discuss the possible function of any of the outlying cells.

More recent mention is made of paragriseal cells by Krause (’22), who briefly described the Hofmann nuclei of von Kolliker in birds and reptiles, and by Poljak (’24), who studied outlying nuclei in Chiroptera, as well as by Terni (’26), who described these cell groups in reptiles and birds. Terni found ten major marginal groups or nuclei (now and again eleven) in chick embryos, segraentally arranged and extending from the first lumbar segment to the fifth from the last sacral segment (fourth from the last sacral segment if eleven major nuclei). In the pigeon he found nine or ten major marginal nuclei, and in the swallow, goose.

208 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and duck, eight of such nuclei. He ascribed several hundred cells to the larger major nuclei, the neuraxes of which could be traced only through the anterior or ventral commissure to enter the most medial part of the opposite ventral funiculus. However, certain of the neuraxes were observed coursing longitudinally for a distance before crossing over. No stem fiber endings nor endings of collaterals could be found terminating in major nuclei, which led Temi to postulate that the cells of the major nuclei may have a spontaneous origin of impulses. He also recorded the fact that there are to be found many glycogen granules in the supporting cells and also in the nerve cells.^^

The following observations were made by J. F. Huber on serial longitudinal and cross sections of adult dove cord, stained in toluidin blue, and forming a part of the extensive series of avian central nervous systems in the Laboratory of Comparative Neurology, Department of Anatomy of the University of Michigan.

Peripheral paragriseal cells are noted throughout the entire length of the cord, forming a colvunn of cells on either side just under the surface of the lateral fimiculus near its junction with the ventral funiculus. It is practically a complete column but has very definite segmental thickenings, which are evidenced in such a way that the cells are much more numerous in the cephalic half of the space between the levels of emergence of consecutive nerves than in the caudal half of this space. There is some variation in the picture at different levels of the cord, but the above description can be applied to the cervical, thoracic, and coccygeal portions. As one passes from the upper cervical to the coccygeal portion (excluding the lumbosacral portion) there is a marked decrease, as above indicated, in the number of cells at the emergence of the rootlets.

In the lumbosacral region, the peripheral paragriseal cells are embedded in a loose glial framework forming protrusions from the cord, one of which completely occupies the distance between the emergence of consecutive nerves (these being much closer together here than in the other portions of the cord). These protrusions are undoubtedly the structures described by Lachi as “Lobi Accessori,” by von Kolliker as “ Hofmann’sche Grosskerne,” and by Streeter as “major marginal nuclei,” and the parts of the coliunn in the cervical, thoracic, and coccygeal regions are made up of the “Hofmann’sche Kleinkerne” of von Kolliker, or the “minor marginal nuclei” of Streeter.

In previous discussions there has been a described or implied difference between the peripheral paragriseal cells in the lumbo-sacral region and those in the cervical, thoracic, and sacro-coccygeal regions, but it seems probable that the fact that the lumbo-sacral region cells are in a loose glial framework instead of “lined up” by the fibers of the lateral funiculus, as the others are, can to a large extent

” Glycogen is an oxygen .storer occurring especially in those regions where the restoration of nonnal oxygen content to the blood cannot take place rapidly. The stimulation of the respiratory center of the medulla oblongata, according to physiologists and especially according to the recent work of Gcscll (’26), is due largely, if not solely, to the metabolic condition of either the blood or of the cells (Gcscll) of the center. Moreover, the so-called gelatinous cells of the sinus lumbo-sacrali.s contain considenible oxygen, according to Temi ('2-1), and the extent of thc.so cells corresponds very nearly with that of the nuclei marginalcs majores. The possibility of incoming impulses from fibers entering this nucleus has not been exhausted, however (Ariirv Kappcri).

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 209

account for the impression of difference. The cells as seen in the cervical, thoracic, and coccygeal regions are ovoid, about 20-30/i long and 7-1 Iju wide, and are quite regularly longitudinally disposed. They have a spherical nucleus occupying most of the width of the cell and often about in the middle of the length. Processes are seen coming off from each end of the cell. In the lumbo-sacral region there are some cells which correspond exactly to the above description and many cells wliich appear to be this type of cell cut at different angles. There are some that appear as irregular multipolar cells, somewhat similar to ventral horn cells but less than half the size and lighter staining (see fig. 95).

The structures which have previously been separated into major and minor nuclei are here considered as making up a continuous column of cells, for three main reasons. First, it is beheved that the cell type is the same throughout the whole length of the cord. Secondly, there is a gradual transition from one situation to the other at both the upper and lower ends of the lumbo-sacral region and not an abrupt, distinct change. That is, between the twentieth spinal nerve and the twenty-first, the peripheral cells, with a very small amount of glial framework, form a slight protrusion from the cord. Between the twenty-first and twentysecond there is a more marked protrusion. Then between the twenty-second and twenty-tliird, twenty-third and twenty-fourth, and so on to the twentyseventh and twenty-eighth nerves, there are the full-sized protrusions already described. Again, between the twenty-eighth and the twenty-ninth, and the twenty-ninth and thirtieth, there is a decrease in the amount of bulging, and between the thirtieth and thirty-first there is none. The third reason for describing these structures as a column of cells with segmental increase is that in graphic reconstructions made from the longitudinal sections of the cord of the adult dove, they appear as such.

In addition to the column of peripheral paragriseal cells here described, there are scattered paragriseal cells which are distributed in the white matter, lateral and ventrolateral to the ventral horn in the lower half of the thoracic and the lumbo-sacral regions. They have no apparent, regular arrangement but are scattered irregularly from the ventral horn out almost to the surface of the cord. These cells begin to come in about the middle of the thoracic region, increase in number to the lumbo-sacral region where they are the most numerous, and disappear quite abruptly near the caudal termination of the lumbo-sacral sinus.

The scattered paragriseal cells are much like the ventral horn multipolar cells, the only apparent difference being that they seem to present a slightly smaller diameter. However, there are to be seen cells of this order which are as large as the largest ventral horn cells, so that it seems logical to consider them as dislocated ventral horn cells, while other cells belong to the peripheral paragriseal column (see fig. 95).

It thus would appear that the nerve cells lying outside of the central gray in the dove spinal cord can be grouped into (1) peripheral paragriseal cells, which form a column with segmental increases throughout the cord, just under the surface of the lateral funiculus near its junction with the anterior funiculus, and which are probably displaced anterior commissural cells; and (2) scattered

210 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

paragriseal cells found in the lower thoracic and lumbo-sacral regions which are probably displaced ventral horn cells in part, and in part of the type of peripheral paragriseal cells.

The presence of efferent fibers — the so-called fibers of von LenhossSk (’90) — in. the dorsal roots has been frequently confirmed {Ramdn y Cajal, '90 ; va7i Gehuchten, ’93 ; and Martin, ’95). They have been pointed out in the cervical cord of birds. They arise from the dorsal part of the ventral horn region and have been regarded as motor (von Lenhossek, Ramon y Cajal, van Gehuchlen) and, by certain observers, as the homologue of the mammalian spinal accessory (Beccari, ’13 ; see bibliography for reptiles). Their specific peripheral termination is not known and their significance is not fully understood as yet.

Fibers of voa Lcnhossdk

Finer visceral efferent (preganglionic) components of the dorsal root have been described for the avian cervical, thoracic, and lumbar cord. The majority of preganglionic fibers in birds, as in reptiles, leave the cord through the ventral roots. In preparations treated by the methylene-blue method, visceral efferent preganglionic fibers were found to form delicate intracapsular, pericellular ple.xuses surrounding the multipolar cell bodies of the postganglionic sympathetic neurons, very similar to such synaptic relations found in the sympathetic system of mammals {Huber, ’99). Their cells of origin have not been localized with

Beccari (’13 and '14 ; see biblioRrapliy for reptiles) regarded the homologous fibers in reptile.s as the forerunners of the spinal root of the maininalian accessory nerve. This is certainly the most illuminating suggestion olTcrcd thus far regarding them. Two obvious difficulties show the need of further confirmatory work before its final acceptance. The.se are : (1) the lack of knowledge as to whether or not the distribution is comparable to that of the spinal accc.s.sory nerve and (2) the difficulty of explaining the apparently similar cells described by Banchi (’03 ; sec bibliography for reptiles) as present in the turtle cord in the lumbar region (sec account of reptilian cord, page 190).

THE COMPAEATIVE ANATOMY OF THE SPINAL CORD 211

certainty for all birds, but in Grus, Buceros, Columba, and Anser, Takahashi (T3) found cells in the intermediate portion of the lower cervical and thoracic cord, which in the latter region agreed topographically with the distribution of the cells of origin of the mammalian intermediolateral column. By homology these avian cells were regarded as the avian center of origin for preganglionic fibers. These cells were covered by a small amount of fibers and cells, similar in appearance to the substantia gelatinosa, which

represented, probably, the nucleus of termination of the incoming visceral sensory fibers.

A somewhat similar distribution of the avian preganglionic centers had been given by Langley (’04) in his physiological research concerning the movements of the feathers, for he considered that the pregangUonic colunm e.xtended from the last cervical to the third lumbar segment. Terni (’23) , in his study of chick embryos, traced the intermediolateral column (see also fig. 95) from the cephalic end of the first thoracic to the caudal end of the second lumbar segment. Term’s figures of the preganglionic cells in the chick suggest the comnussural group of Streeter (’04), which consists, in ostrich, of neurons scattered through the thoracic portions of the ventral gray commissure. Streeter suggested that these might be visceral in function.

The cells of origin for both visceral and somatic sensory fibers are in the spinal ganglia. The dendrites of the

Fig. 97. The e.xtent of the thoracic segments in the dove. Deelman (*19).

In A, two adjacent thoracic roots were cut. The cross-hatched zone is anaesthetic. In B,

former distribute peripherally with the sympathetic system, carrying affer

a root oral and one caudal to the root under consideration were cut. The cross-hatched zones are anaesthetic.

ent impulses from visceral surfaces.

The distribution of the somatic afferents to the skin has been studied by Sparvoli (’07), and particularly by Deelman (’19) and Miss Kaiser (’24). The dermatome has the shape of a trapezium or a triangle with the base lying centralward. The shape is due, according to Deelman, to the greater size of the ventral part of the body.-® There is a pronounced reciprocal overlapping of the peripheral nerves amounting to about half of a dermatome.

In figure 97 the anaesthetic zones are indicated by stripes. In 97A two adjoining roots were cut ; in 97B two adjoining roots caudal and two oral to the root studied were cut ; the remaining dorsal root is evidently unable to maintain the normal sensitivity of the skin.

212 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The neuraxes of spinal ganglion cells, on entrance to the cord, pass to the dorsal funiculus, occupying its most lateral part (zone of entrance), where they divide into ascending and descending branches, as in saurians and mammals. It is chiefly from the region of this zone of entrance that the sensory-motor collaterals arise.^® There are no dorsal root fibers in the lateral funiculus, as was the case with ophidians and Chelonia, but finely medullated and non-medullated fibers from the root cut the dorsal gray substance and terminate in synaptic relation with its cells. Although the avian dorsal funiculi contain the dorsal root fibers, which in snakes and turtles are located largely in the lateral funiculus, nevertheless the avian dorsal funiculi are smaller than those of reptiles. Brouwer (’15) found, that in the birds studied by him, these funiculi totaled only 7.01 per cent to 8.7 per cent of the white substance of the cord, and confirmatory results (7.7 per cent) were obtained in the chicken (Ariens Kappers). A comparison of these figures with the 10.41 per cent obtained in similar measurements carried out on the small lizard (Brouwer), or the 16.2 per cent computed for Danunonia (Ariens Kappers ; see p. 199), makes it evident that the percentage of dorsal funicular fibers in birds is considerably lower than that in reptiles.” Biouwer believed that this decrease is apparent rather than real ; that is, it is due to the presence of highly developed tecto-spinal and vestibulo-spinal tracts which increase the ventrolateral funiculus, rather than due to an actual decrease in the posterior funiculi. In order to gain a more accurate idea of the changes which have occurred, another method of comparison was adopted which did not involve the ventrolateral funiculi. This was a comparison of the size of the dorsal funiculi with that of the gray substance at any selected level. The following results were obtained :

Rana mugiens : the area of the dorsal funiculi equals 40 per cent of the area of the gray substance.

Dammonia subtr. : the area of the dorsal funiculi equals 75 per cent of the area of the gray substance.

Callus domesticus ; the area of the dorsal funiculi equals 22 per cent of the area of the gray substance.

Didelphis; the area of the dorsal funiculi equals 40 per cent of the area of the gray substance.

Putorius putorius : the area of the dorsal funiculi equals 46 per cent of the area of the gray substance.

Oedipomidas : the area of the dorsal funiculi equals 50 per cent of the area of the gray substance.

Callitrix : the area of the dorsal funiculi equals 62.5 per cent of the area of the gray substance.

Conip-arc witii thi.s figure 71, wliere the sensorj'-motor fibers are collaterals of dendrites of motor neurons.

In making .such comparhons, it must be kept in mind that other conditions being equal, the larger the animal, the greater the cross section of the white matter of the cord as compared with that of the gray, .os Hoi ij (’13 : see bibliography, for mammals) has pointed out. This is particularly applicable to the po.~terior funiculi. Thus the largest of the amphibians has been compared in thus table with a relatively small bird and small reptile, so that there is no great difference in body .Mzc. With .Mich a largo bird .a,s the o..,trich the percentage amount of dorsal funicular fibers is greater (Utrcclcr, ’01), but such a compari.Mm is unreliable.

214

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 215

ments and the termination of incoming sensory fibers within a few segments of their entrance in birds are correlated with differences in the kind of locomotion carried on by the two animal groups. In birds the movements are restricted to a few segments, the anterior extremities are specialized for flying, the posterior extremities for walking. In contrast to this condition, in the snake there are

Inf olivary nucl.

A

t

i

B

Fig. 99. A. Nucleus of the dorsal funiculi {H.S.K.} and the spinal trigeminal nucleus (Nud.spin.V) of the cassowary.

B. The medial and the lateral nuclei of the dorsal funiculi in Cacatua roseicapita.

no localized organs of locomotion, so that the whole trunk takes part in the movements of progression. Even in quadruped saurians and in lower mammals, in walking or running a regular and exact coordination takes place between the movements of the front and of the hind legs. Anyone who has watched a bird attempting to escape by flapping its wings and running at the same time will scarcely have failed to notice the obvious lack of coordination between these movements.

216 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Of course not all of the fibers of the dorsal funiculus terminate in the cord ; a small portion of them ascend as far as the lower part of the medulla oblongata. Friedlander (’98) demonstrated, by means of degeneration experiments, these long ascending bundles in the pigeon and was able to show that certain fibers of

Arcaate cciia the lumbar cord run into the upper ' / cervical cord next to the dorsal sep / turn, while fibers from the upper

levels of the cord are situated lateral to these. Thus medial and lateral divisions of each dorsal funiculus, comparable respectively to fasciculus gracilis and fasciculus cuneatus of mammals, are present, although small in birds. Small dorsal funicular nuclei are also demonstrable (fig. 99). As is to be expected, the number of fibers running forward from these nuclei to the thalamus (bulbo-thalamic or the medial lemniscus) is necessarily small, although such are present according to Wallenberg (’04).

Next to these coarser fibers of the dorsal root is a smaller bundle made up of finely medullated and, in part, unmedullated dorsal root fibers. These form a cap over the

Fig. 100. Cross section through the spinal cord of a 4-day chick embryo. Ram6n. y Cajal (’90, fig. 1', p. 610). W, the enlarged growing tips of the axons; K, ventral root axon.

dorsal horn, the fibers dividing in the marginal zone of Lissauer and terminating, soon after entrance, in synaptic relations with substantia gelatinosa, which is well developed in birds and in general comparable to that in reptiles, particularly in snakes (Keenan, ’29). The cell bodies of the commissural neurons (which are cells of origin for the arcuate fibers of His) are found principally in the dorsal part of the gray substance (fig. 100) in avian embryos. In adults the greater number of these lie in the ventral hom.^^ Others are found in the marginal nuclei. Their neuraxes decussate in the anterior or ventral commissure and send ascending and descending fibers into the ventral and lateral funiculi ; collaterals pass to the ventral horn of the same side. These cells belong to the oldest elements of the spinal cord. They are already present as the dorsomedial cells of Amphioxus, the most frontally situated cells sending their neuraxes caudalward, while the caudal cells send them in a frontal direction (see page 145). In this connection it is interesting that in the ehick embryo (where they develop

“This shifting of reticular elements of the spinal cord (for these migrating cells must be regarded as such) from the sensory to the motor region is also observable in the medulla oblongata. These neurons, mainly at least, give rise to the spino-bulbar and spino-tectal paths of the cord, as will be discussed in a later account.

THE COiVIPARATIVE ANATOMY OF THE SPINAL CORD 217

before the motor root cells) a similar pattern is found ; for here, too, the frontal commissural cells send their neuraxes backward, while those of the caudal cells run frontalward {Bok, T5). Apparently, then, cervical stimuli run aborally while stimuli from the tail run orally. At a later stage the fibers dichotomize, each sending out an ascending and a descending branch so that there is a compensation for the original differences.

Streeter’s table (fig. 98) shows that the ventrolateral funiculus increases at the level of the cervical and particularly of the lumbar enlargement. That this is mainly an increase due to an additional number of short fibers running in these columns to connect neighboring segments, and is not brought about by the accumulation of long ascending paths, is clearly demonstrated by the almost immediate decrease of the ventrolateral columns in the region cephalad to the enlargements.

Funicular cells appear in the embryologic development after the formation of the commissural cells. They lie chiefly in the medial regions of the gray Fio. lOl. Spinal cord of a 5-day chick embryo. Ramin substance and send many den- ■EandC.commissm-e cells; and Z), homolateral

. funicular cells.

drites into the region of the

dorsal horn. Their neuraxes run particularly in the lateral funiculus, although some enter the dorsal and ventral funiculi as well. As a rule, the neuraxes are small and extend only for short distances, in their course giving off collaterals to the gray substance. However, there are a certain number of longer endogenous fibers.

A part of the commissural cells give rise to longer ascending tracts which rim in the ventral and lateral funiculi. These are the forerunners of the mammalian ventral spino-thalamic (for general tactile sensibility) and lateral spino-thalamic (for pain and temperature) tracts. They are the avian representatives of the spino-bulbar and spino-mesencephalic or spino-tectal fibers of Edinger (’08), phylogenetically the oldest secondary sensory tract which establishes relations frontally with optic and static centers.

That large, homolateral tracts are present in the cord is evident from the accompanying figures (fig. 102). B shows the ascending degenerations which are evident in the intumescentia cervicalis after a hemisection of the cord in the

218 NERVOUS SYSTEMS OF VERTEBRATES AND OP MAN

lumbar region {Friedldnder, ’98). Among these paths is the spino-cerebellar tract, which was represented in lower vertebrates. In fishes the major portion of this fiber system arises within the cervical cord ; in birds there is a great increase in the system, the tract arising from all levels of the cord and receiving a considerable component from the lumbar region.^^ Friedldnder (’98) stated, and

HS.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 219

liition of fibers, so that the bundle grows considerably thicker as the higher cord levels are reached. In the medulla oblongata the fibers pass through the corpus restifornie to the cerebellum ; consequently, in general relations, if not precisely in nuclear origin, the tract is comparable to the mammalian dorsal spino-cerebellar tract. A smaller bundle, probably similarly a homologue of the ventral spinocerebellar .system, runs frontally and passes the velum medullare anterius, then turns upward and backward to reach the cerebellum of the same and, after decussation, the opposite side (Ariens Kappers). The pre.sence of the.se tracts has been confirmed in the recent work on the brain stern by Sanders (’29 ; fig. 393).

There has been considerable question lus to the amount of spino-cerebellar decussation in the cord. Kuhn and Trendelenburg ('ll) believed that fibers, at least in the lumbar cord, from their origin in the dorsal horn region, decussated and then took up a po.sition at the periphery of the cord and ran forward to the cerebellum. In the levels containing the centers for the wing, .spino-cerebellar fibers ran forward on the same .side of the cord. According to these observers, both contra- and homolatcral paths enter the cerebellum by way of the inferior cerebellar peduncle, where they are situated partly lateral and partly medial to the nucleus mcdialis cerebelli and cross in large numbers. However, Groebbels (’27) found in the cord only a spino-cerebellar system which was uncrossed.

Accompanying the spino-cerebellar tract on its medial side is a system of descending cerebello-spinal fibers which can be traced caudalward into the lumbar cord (Frenkel, ’09). Groebbels (’27) also found such a system descending with the vcstibulo-spinal tract. Presumably these are the homologues of the uncinate fasciculus, Russel’s tract or the “Hakenbiindel” of mammals, which carries fibers of contralateral and homolateral origin from the roof nuclei of the cerebellum to bulbar and spinal centers. The great development of the uncinate fasciculus in birds shows the importance of the cerebellum in these animals.

A bundle of ascending and descending homolateral fibers is located in the ventromedial portion of the cord, near the ventral median fissure (fig. 102). Its function and its mammalian homologue, if such e.xists, are as yet unknown.

Descending pathways are present in birds from the nuclei of the vestibular nerve and direct vestibular root fibers are believed to enter the cord also. Groebbels (’27) carried out a series of experimental lesions on the spinal cord and medulla oblongata of the pigeon, following these by a study of certain reflexes by means of physiologic methods, checked by proper histological observations. He concluded that centers in the metathalamus and the tegmentum of the midbrain involved in certain rotation movements discharge through the medial longitudinal fasciculus, and that impulses for centers concerned in the elevation of the wings and in tail movements, after entering the medulla oblongata over the vestibular roots, reach the appropriate wing and tail motor centers in the cord directly through the contralateral fasciculus longitudinalis medialis, or indirectly by this fasciculus after a synapse in Deiter’s vestibular nucleus.^^

“ The distribution of fibers arising in Deiter’s nucleus to both the homolateral and contralateral medial longitudinal fasciculi in birds has been documented by Wallenberg ('98; see also p. 1025), Groebbels (’27), and others.

220 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Apparently the transmission of impulses leading to a turning of the neck may travel by another path as well as by the medial longitudinal fasciculus. There are large crossed and uncrossed dorsal and ventral tecto-bulbar paths in the bird {Edinger, Wallenberg, and Holmes, '03 ; Craigie, ’28 ; Papez, ’29 ; Sanders, ’29). Their course has been checked in degeneration material by Miinzer and Wiener (’98), and Edinger, Wallenberg, and Holmes (’03). Impulses carried by these tracts reach the cord after a sjmapse on the reticular cells of the medulla oblongata. Bundles of the crossed dorsal tecto-bulbar tract can be traced to the caudal end of the medulla oblongata and may even enter the cord (Edinger, Wallenberg, and Holmes, ’03). Likewise the uncrossed components of this dorsal tract appear to be accompanied by tecto-spinal fibers. Both of these run in the ventral funiculus of the cord, in relation with the ventral ground bundles (the continuation of the medial longitudinal fasciculus of the brain stem).

The well-developed avian red nucleus probably sends fibers to the cord, homologous to those forming the rubro-spinal tract of mammals. Presumably these lie in the medial part of the lateral funiculus but further research is necessary for their establishment. Sandmetjer (’92) mentioned the presence of a pyramidal tract in pigeons, but other observers have been unable to confirm his statements in this particular.

From the above discussion it is evident that the cord in birds contains, in addition to a highly developed system of endogenous tracts, a series of longer conduction pathways which carry impulses from the cerebellum and vestibular apparatus, and, directly and indirectlj’-, from the optic tectum. This development of the longer descending pathways is undoubtedly associated with the great part which motor activity plays in the behavior of these animals.

It is of interest that studies on the growth of the central neiwous system in the chicken, carried out by Latimer (’25), indicate that the spinal cord increases in weight 19 times from hatching until maturity. It makes up .51 per cent of the body weight at hatching and about .12 per cent of that weight at maturity.

The supporting substance in the avian cord is highly developed. It shows, in addition to the radially arranged ependymal fibers, astrocytes associated with longer and shorter neuroglia fibers. These latter fibers are particularly rich along the capillaries and near the margin of the cord. Microglia cells are found also in birds. For details of the cell structure of the supporting tissue, the work of Lachi (’89), Relzius (’93), von Lenhosseh (’94), Ihiber (’03), Ramdn y Cajal (’09), and others should be consulted.

The Spinal Cord op Majlmals

EXTENT .\.ND GROSS MORPHOLOGY OF THE SPINAL CORD

Unlike the spinal cord of birds and reptiles, that of adult mammals .seldom extends through the entire length of the vertebral canal. The caudal part of the canal is occupied by the filum terminale and the cauda ecpiina, a condition re.'^embling that found in some teleosts and in Rana. In early embryonic life, however, the cord occupies the whole extent of the canal. Kunilomo (’18) and

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 221

Streeter (’19) have shown for man, where the tail is atrophied, that the caudal region becomes secondarily dedifferentiated. {Harmeier, '33, found elements characteristic of human cord in the filum.) The last three coccygeal ganglia atrophy and disappear in the first half of embryonic life and the cephalic part of the dedifferentiated area remains as the walls of the ventriculus terminalis’” while the caudal part of the cord thins down and redifferentiates into a fibrous strand, the filum tcrminale. At the caudal end of this strand a small vesicle (the coccygeal medullary vestige) persists at least until late in embryonic life. It is possible that this may be compared with the terminal region of the cord in fishes, where the large, clear, polynuclear cells of Dahlgrcn and Speidel (bibliography for plagiostomes) are to be found. However, it must not be confused with the glandula coccygea or glomus coccygeum, which lies ventral to the coccyx in close relation to the arteria sacralis media and consists of a vascular plexus rather than chromaffin tissue (Stoerk, ’OG. .Vreh. f. mikr. .tVnat., Bd. 69, S. 322), as was formerly supposed.

One condition which appears to influence the discrepancy in length of the spinal cord and spinal canal is the atrophy of certain myotomes producing tail musculature, for it is to lie remembered that the tail musculature in mammals — no matter how large that organ may grow — has no longer the metameric character found in reptiles but is formed by differentiation and transformation of the proximal tail musculature, while the more caudal tail myotomes (Gegenbauer, ’98) are lost. Thus in miunmals other than man — monkeys, carnivores, ungulates, and cetaceans (Phocaena among them) — there is a lack of correspondence between the length of the vertebral canal and of the spinal cord. This is less considerable in monkeys than hi man ; in carnivores the conus temunalis reaches almost to the end of the lumbar column, while in ungulates it extends even into the middle of the sacral region. An interesting difference in the extent of the spinal cord as compared with the length of the vertebral canal is to be found in the monotremes, Echidna and Ornithorhyncus. In the former, the spinal cord ends midway in its course through the canal ; in the latter, the cord extends into ihe sacral region. Both animals have a tail of about one-quarter of the total body length, but in Ornithorhyncus it is a strong, muscular structure much used in swimming, while in Echidna it is a much thinner, little used organ. A comparison of certain mammals which are almost tailless, such as certain rodents (Lepus), chiropteres, and the insectivores (Erinaceus), reveals that whereas in the rabbit the spinal cord extends into the sacral region, in the other two forms it is very short {Gegenbauer, '98). Presumably this difference in extent of the cord is associated with the greater development of the posterior extremities in the rabbit and of the anterior extremities in the chiropteres and the insectivores.

“At the end of the conus terminalis, which contains but little nerve substance, the central canal shows a pronounced distention known as the ventricle of Krause but not to be confused with the ventriculus terminalis of cyclostomes and sharks. The ventricle extends dorsalward here, and, in the horse, a dorsal opening has been pointed out by Kemeufen ( 16 ) and others. It is probable that this is secondarily acquired and b not to be regarded as a posterior neuropore {Anens Kappers).

222 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

While size or use of the extremities or the tail may play a part in deciding the relative extent of the spinal cord, certain facts make it evident that other factors

Fig. 103. Topographic relations of the spinal cord to the vertebral canal in man. Gower.

play the chief r61e, particularly that of an unequal growth of the vertebral column and the cord. A careful study of figure 103 reveals that, for man at least, any particular cord segment lies somewhat higher than its homologous vertebra, and that the discrepancy increases caudalward. An exact knowledge of this condition is of great interest to anatomists, pathologists, and clinicians. For example, it is often of prime importance for the surgeon to know the exact relation of the cord segment to the spinous process of the nearest vertebra. The explanation of this difference in the length of the cord and the canal is to be found mainly in the continued growth of the vertebral column after the cord has reached its definite length. As the caudal end of the vertebral column keeps on growing in connection with the development of the pelvic girdle, the most obvious discrepancies are seen in the caudal regions. This is well shown for man in figure 103. The lower end of the spinal cord proper, which because of its conical shape is called the conus terminalis, is found at the level of the intervertebral disk between the first and second lumbar vertebrae, and that part of the canal formed by the last four lumbar and the entire sacral column contains only a filum terminale and a cauda equina.” Consequently, lumbar punctures are made below the third lumbar vertebra, thus insuring against the danger of damage to the cord.

Another peculiarity in the position of the spinal cord with reference to the vertebral canal, which is seen in certain animals, is the remarkable width of the canal in comparison with the diameter of the cord. This is due to the greater growth of the vertebrae than of the cord. An excellent example of this is to be found in the dugong and whale, where the diameter of the cervical canal may be twelve times the size of the cervical cord {Dexler and Eger, ’ll).

Another noteworthy feature is the eccentric position of the cord described by Hochstetter (’97) and de Burlet (’ll) for Choloepus, Brachypus, and Cetacea. Since it is eccentric in this last form {de Burlet), the hanging positions of the first two groups cannot be the causal factor. Its dislocation to the side

a study of 234 human subjects, McCotler (’16) found no direct relation between the length of the cord and the height of the subject, although short individuals usually had short cords and tall individuals long cords. A

similar condition existed with reference to short and long vertebral columns. This observer found '

the average length of the cord in the negro to be 43.4 cm. ; in the white male, 44.79 cm. ; in the female, 4 1 .8 cm. The lowest point of the cord varied in different individuals from the inferior border

of the second lumbar to a level through the middle of the body of the twelfth thoracic vertebra.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 223

is due to the presence of a large vein (fig. 104). The very large epidural blood vessels, with the character of sinuses, which were so conspicuous in fishes, are still found in the elephant and in carnivores.

A luetanieric arrangement of the human spinal cord substance has been described {Bolk, ’06), but whether or not it is a constant feature of such a fourweeks-old human embryo needs confirmation. Bolk found that the sensory area lying dorsal to the sulcus limitans presents lateral segmental swellings which bulge out into the central canal ; these cones point toward the regions of entrance of the root fibers and are separated by intei segmental pouch-like extension of the

central canal. During further development the alar plate increases thioughout its entile length and loses its metameric character. The sulcus limitans and the intersegmental pouches disappear and the upper half of the central canal is obliterated.

The full-grown cord shows outivardly only the usual enlargements in the cervical and lumbar regions caused by the development of the extremities. The lumbar enlargement is almost or entirely lacking in those

ITig 104 Vertebral canal with spinal cord and great vein in Choloepus dc ButUI (*11)

animals in which the posterior , . , ,

extremities are wanting, such as the dugong {Dexler Eger, H) aiid the extrenuues are wanuug, ^^..rhek ’96 and ’96a; Hephurn and Water common whale {Gumerg, 85 , ^

’04.’^ 'However. Cunningham {I I )

• j n 1 ViJn t-Viprp is a small intumescentia lumbo-sacrahs with a

m Phocaena and Delphm there is a of

corresponding jJ^'orSiR part which the tails of these animals play in

enlargement is due to the common whale or the dugong, as

then: activity, for they are all other mammals both enlarge their popular name JPrt;^o enlargements appears to be associated

meats occur ; ^rior and posterior extremities. Thus in the

with the comparative size of the anter developed and the forekangaroo, where the hind limbs and th pprviral in size (Power,

limbs are smaU, the lumbar enlargement ^^X'Vroato tE

’04). However, in most cases the ^f the forelimbs surpasses

particularly strilang ^ p^terior extremities are small. In higher

the size of the body itself and the p enlargement is the greater be mammals, particularly m primates, t cephalic regions (compare

cause of the increase of white ^'^bstance m the ^

figs.l05andl06). This increase is due to thence ^

cephalic direction in tracts ascending distribution of descending paths,

in a cephalo-caudal direction because

S%.lv ^ ■') Hr}?

ff^J>'^ 'i

p ^ ;^' 'K ’V%^^ii,/ H'i

t.r'kh' “'

Fig. 105. Cross section through the C tt n irrr . « , the i r segment of the orang outang Matpr. V^k’ ^ segments and through the slight development of the (lateral) finger center in C VI T

that of rnan (see fig. 106). *II and C VIII in comparison with

224

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 225

The fissura mediana anterior or ventralis cuts in deeper in mammals than in non-mammals in consequence of the increased development of the ventral white funiculi ; likewise the dorsolateral sulcus, corresponding with the line of entrance of the dorsal roots, is plainer in mammals than in such lower animals in which it is present, because of the greater development in mammals of the adjoining white funiculi. The relation of gray and white substance has, on the whole, increased in favor of the white substance in mammals. Numerous measurements have been taken to ascertain the relation of the total white substance, or certain of its subdivisions, to the gray substance. In such measurements, the size of the animal must be taken into account since Hovy (’13) found, in comparing such measurements made on the cords of larger and smaller animals, that the increase in size of the cords of the former is due, to a considerably greater degree, to the increase of white matter than to the increase of gray (see also Lassek, ’35, J. Comp. Neurol., vol. 62) . In large animals this increase is in accordance with the formula, worked out by E. de Vries, which states that the cross-sectional area of the white substance increases as the cube and that of the gray substance as the square.

DORSAL AND VENTRAL ROOT FIBERS

The numerical ratio between dorsal and ventral root fibers is generally in favor of the former, but the greater size of many of the ventral root fibers does not make this fact evident macroscopically. In some animals the ratio is reversed. For example, in cetaceans (Cunningham, ’77 ; Hatschek, ’96, ’96a) the ventral roots are larger and contain more fibers than do the dorsal roots. The smaller number of fibers in the latter roots is probably correlated with the slight development of the sensibility of the skin, the hair, usually closely associated with sensory terminations, being absent. Perhaps another contributing factor is the great amount of muscle fiber to be supplied in the tail region (where the condition is most marked) as compared with the skin area to be innervated (Cunningham). According to Cunningham, the differences between the two roots are so great that the sensory roots have only half the size of the motor roots in the Cauda equina.

Hatai (’03) found the ratio between the dorsal and ventral roots in three pairs of spinal nerves of the rat to be 1 ; 2 : 3 (in favor of the dorsal root). This observer also stated that there is a great increase in the number of medullated fibers in the ventral roots of the spinal nerves of the white rat during growth, the increase being more rapid in the ventral than in the dorsal roots. Interesting counts of the number of medullated fibers in the dorsal and ventral roots of adult man were made by Ingbert (’03, ’03a, ’04). He found 653,627 fibers in the dorsal roots and 203,700 fibers in the ventral roots of each side. Taking into consideration Birge’s (’82) results on frogs, in which he obtained a ratio of 1 to 1.2 in favor of the dorsal roots, Ingbert concluded that in general there is a greater relative increase phylogenetieally in the number of medullated fibers in the dorsal than in the ventral roots.

Two sorts of medullated fibers are distinguishable in the ventral roots — coarse fibers and fine ones. The coarse fibers (each of which may give off a col

226 NERVOUS SYSTEMS OP VERTEBRATES AND OP MAN

lateral which, before its emergence, runs back into the somatic motor or somatic efferent root,^®

gray substance) form

Fig 106. Cross section through the C // CVII anAT v r,

of the ventral horn in C V// (finger centers) in man wUh tlT?n b expansion

n witn that in the orang outang (.see fig. 105).

their aim the comparison of separate park orttetXte ru'd *" measurements Iiaving as

to later. Recently Ic/e (’29) studied the postnatal growtlHn ten*”^^^^"^' i albino rat and found, as iVang had done for the onL th.,t median and sciatic nerves of the cmale than in the male animal and that 1^ nerves greater in the

tlie animal, with slower growth during later life but active^Sl^p to 25oX?““

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 227

SOMATIC EFFERENT CENTERS : THEIR ARRANGEMENT AND THE DISTRIBUTION OF THEIR FIBERS

The cells of origin of the somatic efferent fibers lie in the ventral horns of the same side. They are large, multipolar cells with coarse tigroid granules (in fixed preparations) and with neuraxes which acquire a medullary sheath soon after they arise from the cell bodies. Neurofibrils and other structures characteristic of nerve cells are present. The dendrites of these cells are branched and are frequently very long, and they extend in all directions from the cell body with which

Fia, lOG. — {Conlinucd)

they are connected. Many years ago (’90) RamOn ij Cajal called attention to the fact that they, together with dendrites of other spinal cord neurons, form a protoplasmic commissure through which they course to the other side of the cord. According to this Spanish observer, other processes extend toward the ventral surface of the cord of the fetal or new-born mammal, but in adult mammals they pass for only a short distance or not at all into the white matter. A reference to their relations in lower vertebrates indicates that the embryonic relations of these ventral horn dendrites in mammals is a recapitulation of their phylogenetic development. Other dendrites extend dorsally and laterally, or m directions cephalic and caudal to the cell body. From the neura.xis, near its level of emergence from the gray of the cord, arises a collateral branch which turns back into the gray substance. Such “collatdrales initiales” were de cribed first by Golgi (’S3), and their presence was confirmed later by Ramon y Cajal (’90), von Kollikcr ('!)0), van Gehuchten (’91), and others. The main neuraxes cut through the white matter in small bundles and continue into the ventral roots. The number of

228 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ventral horn neurons appears to be relatively constant for a given kind of animal. Thus the average cell coimt for seven normal cords, as obtained by Murray ('28) for the white mouse, showed totals of 6841 motor neurons on the left side and 6910 on the right side. In making these counts the whole cord was divided, by means of a ruled grating, into 2 mm. segments and the number of cells on each side recorded for each segment. The following table shows how close is the agreement per segment for the two sides {Murray, '28; Table I, p. 572).

TABLE I

Average op Seven Normal Cords

Left cervical

2567

28

293

431

597

751

467

32

286

489

606

771

467

Right cervical

2651

Thoracic

247

232

Thoracic

201

213

195

197

201

208

236

234

273

266

1642

289

299

1649

Lumbo-sacral

333

321

Lumbo-sacral

391

361

469

449

507

530

2087

387

420

2081

Coccygeal

201

Coccygeal

156

1

111

55

71

1

18

28

545

4

1

529

Total

6841

6910

1

Descriptions of arrangements of the spinal cord neurons in various mammals are in terms of nuclear groups. Such descriptions are based largely on a study of cross sections. A study of longitudinal sections shows that the cells are arranged in long columns more or less parallel to each other. There is a somewhat common pattern for all mammals, but differences in the degree of development or interpretation of the limits of the various nuclear groups or columns and wide differences in nomenclature make exact comparisons in any two mammals difficult and sometimes impossible. It is of interest in this connection that Angulo y Gonzales (’27) found that there is no segmental arrangement of the motor neurons of the spinal cord in the albino rat at birth. The neurons, although in well-defined groups in cross sections of the cord, formed practically continuous columns in longitudinal sections. He found that the so-called ventrolateral and dorsolateral groups are capable of division into several subgroups. This author regarded

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 229

the “progressive morpholo^cal differentiation of the musculature as associated with or perhaps dependent upon the anatomical differentiation of these columns.” Craw ('28) demonstrated that there is no increase in the number of motor neurons in the spinal cord of man from birth up to three years of age. Her results have been substantiated by Ngowyang (’30), who found no evidence of mitosis among the motor neurons of the albino mouse cord during the twenty days following birth.

Balk (’10), by dissection methods, followed the motor fibers in their course between cord and muscle and on this basis built up a scheme of the segmental innervation of the different muscles. His results imply that the motor cells have undergone no secondary shifting in the longitudinal direction but lie in their original segments. In most cases this assumption appears to be justified, but that such a secondary sliifting sometimes occurs is evidenced by Berkelbach van der SprenkeVs (’15 ; see Chapter V) observation that in Erinaceus the cells of origin for the omohyoid muscle lie in the 12th nucleus in front of the 1st spinal, root while their neuraxes leave the cord in the 2nd root. The longitudinal localization, based on Balk’s, results, is given in the accompanying table (p. 230), study of which reveals certain interesting facts. Aside from the Mm. lumbricales, interossei, subcostales, and transversus thoracicus (triangularis), the innervations of which are rather variable and complex, there are 153 myotomic muscles, of which 48 are mono-, 65 bi-, 27 tri-, and 12 pluri-segmental. Thus apparently only 48 of the muscles are derived from a single myotome while 65 are derived from two myotomes, so that about 45 per cent of the myotomic muscles of the body belong to the bisegmental group (see also Cattell and Stiles, ’24). Agduhr (’16 and ’19), working on the cat, and Hines (’27), working on amphibians, reached the conclusion that individual muscle fibers in certain muscles have a plurisegmental innervation, particularly in such muscles as comply with many fimctional demands (as for example the flexor digitorum in carnivores), thus providing for an overlapping of the motor areas just as there is a sensory overlap. Agduhr’ s work suggested that this motor overlap is acquired secondarily, or may at least be increased by training. However, it is only fair to state that several observers, among whom are Katz (’25) and de Boer (’26), both working on the frog (see bibUography for amphibians), and Wilkinson (’30, using mammahan material), question the presence of a plurisegmental innervation in the material available to them for study.

^Numerous observers have analyzed by microscopic study the major nuclear masses in the spinal cords of various mammals, the work of Cxmningham (’77), Hatschek (’96, ’96a), Hepburn and Waterston (’04), Voris (’28), Pressley and Cobh (’29), and many others meriting attention. Particular reference is made to the fundamental work of Bruce (’01 ; also ’06) on the spinal cord, which affords the basis for the present-day analysis of the human cord. From the accompanying figure 109, it is seen that the anteromedial or ventromedial group is present practically throughout the cord, except in the last lumbar and first sacral segments where it is either extremely small or entirely lacking. Apparently this is supplemented in the upper cervical and in the thoracic segments

SEGMENTAL TOPOGRAPHY OP THE SPINAL CORD INNERVATION OF THE VOLUNTARY MUSCULATURE IN MAN; AFTER BOLK

RectU3 capitis anticus et lateralis . . . . .

Rectus cap. post, minor . Rectus cap. post, rnaior . Obliquus capitis sup. . .

Obliquus capitis inV. . .

Longus colli

Scalenus anticus . . .

Scalenus medius . . .

Scalenus posticus . . .

Thyreo-hyoi'deus . . .

Sterno-hyoideus . . .

Omo-hyoldeus .... Sterno-thyrcoldeus . .

Trapezius

Sterno-cleido-mastoidcus Levator scapulae . . .

Rhomboides

Serratus anticus . . .

Supraspinatus .... Infraspinatus ....

Teres minor

Teres maior

Latissimus dorai . . .

Subscapularis ....

Subclavius

Pectoralis maior . . .

Pectoralis minor , . . Deltoldcs ......

Coraco-brachialis . . .

Biceps brachii ....

Brachialis

Anconaeus longus . . .

Anconaeus internus . . Anconaeus oxternus . .

Anconaeus quartus . .

Pronator teres .... Pronator quadratus . .

Flexor carpi radialis . .

Palmaris longus . . . Flexor carpi ulnaris . .

Flexor pollicis longus . .

Flexor digitorum sublimus Flexor digitorum profundus Palmaris brevis .... Abductor pollicis brevis . Opponens pollicis . . .

Flexor pollicis brevis . .

Adductor pollicis . . .

Flexor brevis dig. V et Opponens dig. V . .

Adductor dig. V . . .

Lumbricales

Interossei

Brachio-radialis . . .

Ext. carpi rad. longus Ext. carpi rad. brevis

Supinator

Abductor pollicis longus . Extensor pollicis longus . Extensor dig. comm. . .

Extensor pollicis brevis . Extensor indicis proprius Extensor dig. V et ext.

carpi ulnaris .... Serratus posticus sup.

Cl.

C 1.

C2.

Cl.

C2.

C 1.

2.

3. 4. 5.

C3.

4,

5.

C2.

3.

4. 5. 6. 7. 8.

C 5.

G.

7. 8.

C 1.

2.

Cl.

2.

3.

C 1.

2.

3.

C 1.

2.

3.

C2.

3.

Acccssorus

C2.

3.

Accessorus **

C3.

4.

C5.

G.

CS.

G.

7.

C4.

5.

C5.

G.

C5.

G.

C6.

7.

CG.

7.

8.

C5.

0.

C5.

6.

7. 8.

C7.

8.

C5.

G.

C6.

7.

C5.

6.

C5.

6.

CG.

7.

8.

C7.

8.

CG.

7.

C7.

8.

CG.

C6.

7.

8. Th. 1.

C G.

7.

C7.

8.

Th. 1.

C8.

Th. 1.

C G.

7.

C7.

8.

Th. 1.

C7.

8.

Th. 1.

C8.

CG.

7.

C G.

7.

CG.

7.

C7.

8.

Th. 1.

C8.

Th. 1.

C7.

8.

Th. 1

C8.

Th. 1.

C5.

G.

CG.

7.

CG.

7.

C5.

6.

7.

CG.

7.

C7.

8.

G 7.

8.

CG.

7.

C7.

8.

C7.

8.

Th. 1. 2. 3. 4. 5.

Serratus posticus inf. . . Intcrcostales . . . .

Diaphragma

Transversus thoracis . . Quadratus lumborum . .

Obliquus abdominis ext. Obliquus abdominis int. Transversus abdominis . Rectus abdominis . . .

Pyramidalis

Cremaster

Psoas

Iliacus

Glutacus maximus . .

Glutaeus medius . . .

Glutacus minimus . . .

Tensor fu.sciae latac . .

Piriformis

Obturator internus + Gemellus sup

Quadratus feraoris -f Gemellus inf

Sartorius

Vastus externus . . .

Vastus medius .... Vastus internus . . . .

Rectus femoris ....

Pectineus

Adductor longus . . .

Adductor brevis . . .

Adductor magnus . . .

Obturator externus . .

Gracilis

Semimembranosus . . . Semitendinosus .... Bjeeps fern, caput longum Biceps fem. caput breve . Gastrocnemius ....

Soleus

Plantaris

Flexor digit, long. . . .

Flexor hallucis long. . .

Tibialis posticus . . .

Poplitcus

Tibialis anticus .... Extensor hallucis long. . Extensor digit, long. . .

Peronaeus longus . . . Peronaeus brevis . . . Extensor digit, brevis -1extensor hallucis brevis Abductor hallucis . . . Flexor hallucis brevis . .

Adductor hallucis . . . Flexor digit, brevis Flexor brevis dig. V. . . Opponens dig. V. . . .

Abductor dig. V. . _ . . Interossei . . . . ’ . . Lumbricales I und II . . Lumbricales III und IV .

Levator ani

Sphincter ani .... Bulbo-cavernosus sive Constrictor cunni . . Ischio-cavernosus . .

Th. 9. 10. 11. 12. Th. 1. bis 11.

C‘l. 5.

Th. 3. 4. 5. 9.

LI. 2.

Th.7.8.9.10.U.12.H. Th.8.9.10.11.12.Ll. Th.8.9. 10.11.12.L1. Th.6.7.8.9. 10.11.12. Th. 12.

L 1.

L2. 3.

L3. 4.

L S. SI. 2.

L4. 5. SI.

L4. 5.

SI. 2.

L4. 5. SI. 2.

L4. 5. SI.

L2. 3.

L3. 4.

L2. 3.

L3. 4.

L2. 3.

L3. 4.

L4. 5.

L4. 5. SI.

S 1. 2. 3.

L5. SI.

SI. 2.

L5. SI. 2.

L 4. 5. S 1.

LS. SI.

L 5. SI. 2.

L5. SI.

L4. 5. SI.

L 4. 5. S 1.

L4. 5. S 1.

L 5. SI.

L5. S 1.

L 4. 5. S 1L5. SI.

L 5. SI. 2.

S 1. 2.

LS. SI.

SI. 2.

S 1. 2.

SI. 2.

L5. SI.

Si. 2.

S4. 5.

S3. 4.

The trapezius and sterno-cleido-masfoid muscles are innervated in primary instance through the spinal accessory nerve, originally a branchial nerve (see Chapter V). In man the nucleus of the spinal accessory nerve extends from the first cervical segment to the fifth or seventh cervical segment. Only in the second and third cervical segments are ventral root fibers associated with it.

230

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 231

by a dorsomedial group. So far as can be judged from recent evidence, both groups are to be regarded as nuclei for trunk musculature and consequently remain intact after amputations of the extremities. In Phocaena naturally no limb centers can be demonstrated in the lumbar cord, but centers for trunk musculature are iiighly developed and are divisible, according to Hepburn and Waierston (’04), into secondary groups which they have termed anteromesial and postero-external groups and which occupy the body of the ventral horn (see fig. 107).

The cells of origin for motor fibers to the muscles of the pectoral and pelvic girdles and of the extremities are localized more laterally in the ventral horn in the region of the cervical and ^

lumbar enlargements. This is ' ’

comparable to the position of t.. ’ • .

similar groups in amphibians, ^ -‘.("j; \

reptiles, and birds. Their absence may be noted in figure 108, which illustrates conditions after

the amputation of an arm. Taft ' • " _ \

(’20), however, found no decrease

in the number of cells after an ' t . ■ . . .» /

amputation. In an abranchus, ' /

Curtis and Helmholz (’ll) found the medial groups of ventral horn cells apparently as in normal individuals but the lateral groups replaced by scattered cells. Essentially similar results had been obtained earlier by Elders (’10,

’14), who found the lateral portion of the ventral horn and its appropriate motor neuron groups greatly reduced or absent in a man born without a forearm. The great lateral projections characteristic of most mammals are wanting in the lumbar region in Phocaena, as might be expected {Hepburn and Tl^atersfoa, ’04, and Pressley and Cobb, ’29). These latter observers noted that cell counts indicated a marked cervical enlargement for the flippers, a slight increase for the innervation of the tail musculature, but no enlargement in the lumbar region. Their results indicated a correlation between the cross diameter of the cord, the number and develop

/

-/

Fig. 107. Hypertrophy of the ventromedial (tail) nuclear group in the 11th lumbo-sacral segment of Phocaena communis. Hepburn and )Valerslon (’04).

ment of the muscles innervated at a particular level, and the number of neurons present at the level.

vThe nuclei of the lateral part of the ventral horn fall into four major cell columns; ventrolateral or anterolateral (4^8C, 2L-2S), dorsolateral or posterolateral (3-8C, 2-5L, 1-3S), retrodorsolateral or retroposterolateral (8C, IT, 1— 3S), and central (3 or 4C— 6 or 7C, 2Lr-2S). These are based on data frorn Bruce (’01, Quain’s Anatomy), Jacobsohn (’08), Jacobsohn and Kalinowski (’08a), Elders (’10), and Huber and Crosby (1930 edition of Piersol’s Anatomy,

232 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

from which figure 109 is taken). The finer localization is such that the cells of origin for the most distal muscles (those of hands and feet) occupy the most

dorsal position {Bok, ’28). These muscles are derived from the most ventral part of the perivisceral division of the myotome, the ventral division thus being supplied by more dorsal neurons. Thus the cell bodies, the neuraxes of which supply pectoralis major (a ventral myotome derivative), lie dorsolateral to those for latissimus dorsi, while the cells of origin for fibers to the arm and hand lie dorsolateral to those for pectoralis major. The location of the neurons (fig. 109) supplying the fingers in the nueleus retrodorsolateralis (8C-1T) explains the striking lateral extension of the ventral horn in the 8th cervical segment in primates. This extension is more marked in human than in anthro

Fiq. 108. Cross section tiirough the spinai cord at the level of the cervical enlargement in a man without a right arm. (The right aide of the cord is shown on the right side of the figure.) Elders.

Fia. 109. Conventional transverse section of the human spinal cord. All of the cell columns are represented as though found at one cross-section level. The relative position of the respective columns is given, with the accessory shifted dorsomedialward out of anterior horn for convenience of indicating. The figure legends record the extent of each cell column. Huber and Crosby.

poids, while in the latter the nucleus for the pectoralis major muscle is larger. The localization for the lower extremities is similar to that for the upper (compare fig. 109). Braeutigam (’92) and Sakai (’13) have pointed out that the ven

THE COMPARATIVE ANATOMY OF THE SPINAL C'ORD S

tral horns in the 3rd and 4th sacral segments, where the centers for the sexual organs and perineum are found, are somewhat wider in women than in men, but that this difference disappears with old age.

As yet no satisfactory explanation has been given of the causes underlying the reversal, in the dorsoventral plane, of the position of the cells of origin and the myotomic distribution of their neuraxes, so that the more dorsal motor cells supply the more 'ventral myotomic muscle and vice versa. It is imcertain whether the condition is to be regarded as primary or whether it is acquired secondarily through neurobiotaxis or the influence of some other biologic factor. In view of the neurobiotactic shifting of cells during the development of the pattern of the motor nuclei of the medulla oblongata, it is reasonable to suppose that secondary changes in position of spinal cord cells may be due to similar stimulative influences. Thus the nuclei for trunk and tail musculature, which are most closely associated with body-statics, are situated ventromedially in order to lie in closer relation with the caudal continuation of the medial longitudinal bundle and with the vestibulo-spinal fibers, while the centers concerned chiefly with highly organized activities of the hand and foot are situated dorsally, as close as possible to the great descending paths from higher centers, particularly the rubro-spinal tract."'® Special nuclear groups for the phrenic (C3 or C4-C7) and for the spinal accessory (C1-C4 or C6) are found in the cervical cord.

Intermed lat col

THE VISCERAL EFFERENT CENTERS. THE SYMPATHETIC SYSTEM

a. The Central Location of Preganglionic Neurons.

The finer fibers of the ventral roots are preganglionic fibers passing to sympathetic ganglia, which in mammals are connected with the central nervous system only through the ventral roots (Gafiri, ’96 ; also manyothers). Lugaro (’06) believed that an occasional preganglionic fiber is to be found in the dorsal root of mammals, but his work needs confirmation (see p. 236). The cells of origin for such preganglionic fibers, according to the work of Onuf and Collins (’98), Huber (’99a, ’13),

Jacobsohn (’08), Jacobsohn and Kalinowski (’08), Elders (’10), Bok (’22),

Gagel (’28), and Greving (’28) are to be found in the intermediolateral (and intermediomedial, Bok; see also Gagel) column of the spinal cord, throughout the thoracic and upper lumbar regions. Here they form an extension of the gray substance, which is generally present in mammals (described recently by Voris, ’28, for the opossum cord), but which is particularly

« That the lateral cortico-spinal tract is not the initiating factor here seems probable, since a similar position is occupied by the corresponding cells in birds, although the above mentioned tract is not present.

Fig. 110.

L. 1

Section through the fourth thoracic segment of Simia satyrus.

234 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

clear for monkey and man, and which for decades has been designated the lateral \xoxn {Slilling, 1842 — “ Seitenhorngruppe ” ; see Slilling, 1859). A group of cells of similar form and relations [Greving, ’28) is found in the second, third, and fourth sacral segments (sacral component of the cranio-sacral sympathetic system). According to Bok (’22), the location of these cells is determined to a very large extent by a tract which courses at the side of them and apparently is in synaptic relation with these cells. This tract was designated by Bok the intermedio-longitudinal tract and was regarded by him as an important pathway for the mediation of impulses involved in sympathetic reflexes.

h. The Relations and Certain Functions of the Sympathetic System. Muscle Tonus.

The histological structure of the peripheral sympathetic system has been considered in the first chapter. There appears here only a short review of the major anatomical and topographical relations, summarized after the researches and interpretations of Langley (’90, ’91, ’91a, ’92, ’92a, ’94, ’95, ’96, ’03, and elsewhere), Langley and Anderson (’95 and ’04), Huber (’99, ’99a, '99b, ’13), Langley and Magnus (’05), van der Broek (’07, ’08), Kuntz (’10, '13, ’20, ’27, ’29; for other forms see bibliographies), Ranson and Billingsley (’18, ’18a, ’18b), Ranson (’18, ’31), Muller {’2A), Kuntz and Kerper (’24, etc.), Gagel (’28), Kuntz and Farnsworth (’28), Lawrentjew (’29, ’31), Trostanetsky (’29), Pines (’31), also van Campenhout (’30, ’30a) and Terni (’31); see bibliography. Chapter I, for the last two observers.

The sympathetic nervous system is an efferent system, although its fibers may be and usually are accompanied by visceral sensory fibers (interoceptive fibers), the cells of origin of which, however, lie in cranial and dorsal root ganglia. Langley and his school have shown clearly that the system as a whole, which they term the autonomic system, may be divided into two major subdivisions : 1, a sympathetic proper, or thoracico-lumbar division, and 2, a parasympathetic, or cranio-sacral division. The former receives its preganglionic fibers from the thoracic and upper lumbar cord, 1T-3L {Greving, 8C-2L), the latter from the brain stem by way of cranial nerves and from the sacral cord, 2-4S; consequently, by reason of these facts, the terms thoracico-lumbar (sympathetic proper) and cranio-sacral (parasympathetic) sympathetics, indicating functional activities as weU as morphologic relations, are gradually receiving wider use. Both subdivisions, by whatever names they are known, have certain characteristics in common. They consist in both subdivisions of two neurons in a chain made up of a preganglionic and a postganglionic neuron. The preganglionic neurons, if from the thoracico-lumbar portion of the cord, have their cells of origin in the intermediolateral column or lateral column (and possibly in the cells of the associated intermediate zone, Gagel, ’28). This column has been recognized by practically all students of the histology of the cord {Bruce, ’06 ; Jacobsohn, ’08; Poljak, ’24, and many others). A preganglionic center for the sacral portion of the craniosacral system also has been recognized repeatedly {Jacobsohn, ’08 ; Takahashi, ’13 ; Gagel, ’28, and many others). For the cranial portion of the sympathetic, the preganglionic neurons lie in the Edinger-Westphal

THE COMPAHATIVE ANATOMY OF THE SPINAL CORD 235

nucleus of the oculomotor nerve, the superior salivatory nucleus of the facial nerve, the inferior salivatory nucleus of the glossopharyngeal nerve, and the dorsal efferent nucleus of the vagus and of the bulbar portion of the accessory.

Fig. 111. Scheme of the human autonomic or sympathetic nervous system, mainly after L. R. Miiller, the vagus after Iwama, from figure of C. U. Ariens Kappers, redrawn in black.

From whichever preganglionic center they arise, the preganglionic fibers pass out as medullated fibers to end in sjmaptic relation with postganglionic neurons in either chain, prevertebral, or peripheral sympathetic ganglia. The preganglionic fiber is spoken of often as a white ramus fiber because, if spinal in origin, it leaves the ventral root by way of the white ramus communicans, but

236 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the term “preganglionic” is preferable. The postgangUonic or sympathetic neurons lie in one of the above-mentioned ganglia and send their unmedullated or medullated neuraxes to heart muscle, to smooth muscle, or to glandular tissue. The chain sympathetic ganglia are found on either side of the vertebral colunrn, attached by rami conununicantes to the ventral roots of the thoracic and upper lumbar spinal nerves and connected with each other by communicating fiber bundles. In the cervical region, due to a secondary fusion of the cell masses, there are only three ganglia on a side — the superior, middle, and inferior cervical sympathetic ganglia. These cervical sympathetic ganglia receive their preganglionic fibers from the spinal cord through the upper four or five thoracic nerves, and in man no white rami communicantes occur on the cervical nerves. The cervical nerves, therefore, have only gray rami. It should be noted, however, that Glaser (’24, extracardiac nerves ; Muller, ’24, p. 151) stated that the accelerator fibers to the heart arise from the last cervical and the upper thoracic segments. Chain sympathetic ganglia are wanting at the lower lumbar and upper sacral levels, that is, the region between the thoracico-lumbar and sacral subdivisions.

The name of prevertebral or collateral ganglia is applied to the larger masses of sympathetic ganglion cells found ventral to the vertebral column (such as the coeUac and mesenteric plexuses), while peripheral ganglia consist of small masses of sympathetic cell bodies, or sometimes just scattered sympathetic neurons found near, or on, or even inside of the organ supplied by them (juxtamural and intramural ganglia). It may be emphasized that the preganglionic fibers of the thoracico-lumbar division of the sympathetic system form synaptic relations in the chain and prevertebral or collateral ganglia, while the preganglionic fibers of the cranio-sacral subdivision synapse in peripheral ganglia.

Certain workers describe a spinal parasympathetic system {Kure, ’27-’28, ’30, ’31, ’34 ; Selle, ’31, Saegusa, ’31, and others). Fibers arise from parasympathetic centers at various levels {Kawagizi, ’31), pass to spinal ganglion neurons by the dorsal root, and synapse on ganglion cells (see p. 244) which distribute through thinly myelinated or unmyelinated fibers (see Sheehan, ’33) for vasomotor, secretory, and trophic functions. Such preganglionics need further proof, but efferent dorsal root fibers occur {Okelberry, ’35, J. Comp. Neurol., vol. 62, and others).

Both thoracico-lumbar and cranio-sacral systems pass to most organs (except the glands and the smooth muscle and blood vessels of the skin). It is possible that occasionally the two innervations reenforce each other, but in general when stimulated separately they produce (hfferent and sometimes diametrically opposite results. This does not imply necessarily that they are antagonistic to each other in the normal functioning of the organs, but that working together they produce a balanced action. Thus the innervation of the sphincter of the iris is by way of a two-neuron chain, of which the preganglionics arise from the EdingerWestphal nucleus of the oculomotor nerve and synapse with the cell bodies of the postganglionic neurons in the ciliary ganglion, from whence fibers pass to the circular muscle of the iris, while dilation of the pupil is brought about by a two-neuron chain, of which the preganglionic neurons have their cells of origin

THE COMPAEATIVE ANATOMY OF THE SPINAL CORD 237

in the intermediolateral column of the first four thoracic segments while their neuraxes pass by way of the sympathetic chain to form pericellular, subcapsular synapses with postganglionic neurons with cell bodies situated in the superior cervical gangUon and neuraxes passing to the dilator apparatus of the eye {Langley and Anderson, ’04). Likewise the inhibitors of the heart pass as preganglionic fibers of the vagus to sympathetic ganglia located on the atrial wall of the heart, where postganglionic neuraxes in turn form the typical “clover-leaf” terminations on heart muscle {Huber and DeWitt, ’98, see Chapter I), while the accelerating impulses for the organ are by way of preganglionics from upper thoracic to superior, middle, and inferior cervical (probably also IT) sympathetic ganglia and postganglionics from there to the heart wall (where their finer terminations are not as yet fully determined). In Woollard’s study (’26) of the innervation of the heart in such mammals as rabbit, cat, and guinea pig, he found that all the ganglion cells of the intracardiac ganglia belong to the paras 3 Tnpathetic (or cranio-sacral) sympathetic system, although two types of synapses — pericapsular and pericellular — suggest some difference in function. Woollard obtained evidence which points toward a double innervation (from both thoracico-lumbar or sympathetic and cranio-sacral or parasympathetic systems) of the atrial musculature and that of the atrioventricular bundle, but only a thoracico-lumbar (or sympathetic) innervation of the ventricles.

The innervations of certain glands are examples of a double innervation which causes the production of different kinds of secretions. Thus the stimulation of the preganglionic chorda tympani fibers passing to the sympathetic ganglia on the hilum of the submaxillary gland {Langley, ’90) produces a serous secretion while the stimulation of the thoracico-lumbar chain (1-4T to superior cervical and postganglionics by way of carotid plexus) leads to a thick mucoid type of secretion (see Langley, ’90; Hitzker, '14; Kuntz, ’29).

Smooth muscle and glands of the digestive tract to near the region of the descending colon receive preganglionic fibers through the vagus, with which are associated spinal accessory fibers, which synapse with ganglion cells in peripheral sympathetic ganglia situated between the muscle layers (myenteric or Auerbach’s plexus) or in the submucosa (submucal, enteric, or Meissner’s plexus) . According to many texts of human anatomy, the vagus fibers extend into the descending colon, beyond t.bis the preganglionic cranio-sacral fibers arise from the sacral cord. At least for certain mammals, an exactly comparable distribution of cranial and sacral components has not been demonstrated by all observers ; for example, in dogs and cats Schmidt (’30) carried the vagus to the large intestine at least as far as the distal portion of the transverse colon while he traced the sacral nerves to all of the large intestine. A secondary sympathetic innervation to the whole gastro-intestinal tract is supplied by the thoracico-lumbar division, the synapse occurring in ganglia outside the organ. This was documented in an excellent study of the enteric plexuses by Hill (’27), who found that the vagus synapses in the ganglia within the digestive tract wall, while the synapse of the thoracico-lumbar fibers is in visceral or collateral ganglia, from which fibers course with those of the myenteric plexus in the small intestine, but form

238 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

separate fasciculi in the stomach wall, internal to the myenteric plexus. The details of the finer structure of the enteric plexuses have been studied intensively of recent years. Reference is made here to the bibliographies and accounts of Hill (’27), Stohr (’30), Lawrentjew (’31), and Solcolowa (’31). Many others might be quoted. Hill described neurons of type I and of type II in the myenteric plexus, and type II neurons in the submucal plexus. Type I neurons were regarded as association neurons and type II as forming synaptic relations with preganglionic fibers. She asserted that no distinct local reflex mechanism is provided by these plexuses.^ In an elaborate study of the enteric plexuses of the cat and rabbit and of man Stohr (’30) described reticula with associated ganglion cells in Meissner’s plexus, or the submucal plexus proper, in the plexus entericus internus, and in Auerbach’s or the myenteric plexus. Type I and type II neurons were found ; between those of type I anastomotic, plasmatic connections were observed. Stohr believed that Schwann’s syncytial conducting tissue, with such fibers embedded in it as those from the fine tertiary reticulum of Meissner’s plexus, is present, and regarded this tissue and the embedded fibers as constituting a nervous terminal plasmodium. Unquestionable demonstration of the intraprotoplasmic endings of fibers in smooth muscle does not appear to have been given. Lawrentjew (’31) described two types of neurons in the myenteric plexus of the dog, with one of which the vagus and the nervus erigens were in synaptic relation. He assumed that the other type had some relation to spontaneous movements of the digestive tract.

Many more examples of this double innervation could be quoted if space permitted. It would be found that such examples show a repetition of similar patterns.^^ It may be said that in general cranio-sacral preganglionics .synapse on or near the organ while those from the thoracico-lumbar system synapse in chain or collateral ganglia. The sympathetic system supplying the viscera is sometimes called the vegetative system.'*’

At the surface of the body and in the extremities no sympathetic ganglia are found. The chain ganglia of the thoracico-lumbar system send postganglionic fibers to sudoriferous and sebaceous glands, to the erector pili muscle, and to the smooth muscle of peripheral blood vessels. These postganglionic fibers pass back on the nerve roots by way of the gray rami communicantes and distribute with the peripheral nerves in a segmental manner {Langley, ’91, ’91a, ’94, ’95, ’96, etc. ; van Rijnberk, ’07, ’08). One of the first clear demonstrations of this

'■ Troslanelzhy (’28) described certain so-callcd glia cells (connective tissue cells of Johnson,

’25) in the myenteric plexus which Lawrenljew (’29) believed were concerned with the formation, within the gut wall, of a primitive nerve plexus. A study of preparations, differentially stained, led J ohmon and P armor (’31) to the conclusion that “ there are no glia cells and no primitive nerve net in the gut wall of the frog, cat, and dog.”

” Attention is directed here to the work on the innervation of endocrine glands, certain phases of which will be considered in Chapter VIII, pp. 1185 and 1180. Special reference is made here to the excellent review of this matter by Pines (’29), to his bibliography, and to Blohr (’28).

■“ Much of the present knowledge of the location of the .synapse is due to the use of drugs such as nicotine, which was employed first and .so ably by Langley and his coworkers. Apparently nicotine checks conduction over the synapse while permitting it over the fibers, making it pos.siblo to determine in which of several ganglia through which it pa.sse3 a fiber will synap.‘-c.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 239

distribution of the preganglionic fibers to the periphery of the body was made by Langley (’95) in illustrating the sympathetic innervation of the hair in the cat. Langley concluded that, while in general the distribution of the sympathetics and the sensory to a given region of the skin is the same, this does not necessarily imply that the terminal distribution falls within the same limits. Occasionally it does, as this observer showed to be the case with fibers of the 1st, 3rd, and 5th lumbar in cat, but apparently on the whole the postganglionic pilomotor fibers have less terminal overlap in the skin than do the sensory fibers. Langley considered the “spread” of the terminal fibers a matter of secondary importance. The cutaneous territories of spinal and sympathetic ganglia have been regarded as coinciding in extent for pilomotors (van Rijnberk, ’07) and for vasoconstrictors (Phillips and Woollard, ’32, J. Anat., vol. 67) but possibly not for vasodilators nor vasoconstrictors to deep vessels (Telford and Stopford, ’32, J. Anat., vol. 67).

A review of the literature on the innervation of blood vessels is to be found in the relatively recent accounts of Stohr (’28 and ’30) and in the work of Hinsey (’28; see also ’30), and such a review does not require repetition here. It is sufficient to state that blood vessels at or near the periphery of the body, including those within the voluntary striated muscle, receive an innervation from the thoracico-lumbar sympathetic system, with the cell body of the postganglionic neuron within the chain ganglia. Lumbar ganglionectomy (Brown and Adson, ’30) produces marked vasodilation of the arteries of the lower extremities (feet and lower legs). Such dilation is said still to be present at the end of three years. Efferent fibers from cells of spinal gangha (spinal parasympathetic system) for vasomotor functions have been described by Kure (’27-28, ’30, ’31), andHirl (’29). Sensory innervation of blood vessels appears to be supplied by visceral sensory fibers, with cells of origin in the appropriate spinal ganglia.

Much attention has been centered within recent years upon the possible relation of the sympathetic nervous system to muscle tonus. It is generally conceded that muscular movement — contraction in response to external stimuli — is dependent upon impulses along the neuraxes of ventral horn neurons, but by what means the other function of muscle, that of maintenance of reflex postural contraction (due to the properties of contractiUty and plasticity of skeletal muscle), is carried on is understood less clearly. The most generally accepted view for the last twenty years has been that advocated by Sherrington (’94, ’97, ’97a, ’98, ’15), that this tonic contracture is due to the reflex arc made possible by the presence of a sensory (proprioceptive) innervation of striated volimtary muscle (Brondgeest’s tonus, so-called after the man who first described it).

A clear-cut statement of Sherrington’s point of view with regard to the importance of the proprioceptive components of the nerv’e in the maintenance of muscle tonus and the significance of tonus with relation to the postural refle.xes is given in his 1915 paper, which should be consulted by those interested in the matter. Two quotations may serve to illustrate his point of view in the matter : “The observations on the tonus of skeletal muscle in the mammal, therefore.

240 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

go to show that the phenomenon is in skeletal muscle nothing more nor less than postural contraction” (p. 202) ; later (p. 233), “Reflex maintenance and adjustment of posture is a chief portion of the reflex work of the proprioceptive system, just as sensation of and perception of posture is a chief portion of the psychical output of that system.” Sherrington recognized that postural reflexes may be affected and even initiated by other than proprioceptive impulses from the muscles involved, and mentioned in this connection the work of Ewald (’91) on pigeons, the various experiments of Magnus and de Kleijn (’12), see Magnus (’24), showing postural reflexes following stimulations of the labyrinth, and Weed’s study (’14) of decerebrate rigidity. Such studies as those of Magmis and de Kleijn, Sherrington, and Weed indicate that the higher centers exert a marked influence over the efferent centers of the cord, such that their elimination produces decerebrate rigidity.

Langelaan (’15, ’22) and Ranson (’28) suggested that possibly posture is maintained by a physical change in the condition of the protoplasm following tonic innervation, and the latter observer suggested that tonic innervation may pass from the cord to the muscle by way of the dorsal roots (a suggestion previously made by Frank). Bottazzi (’97) predicted that the maintenance of contraction in a muscle fiber was a function of its sarcoplasm. In his 1930 review of the question of muscle tonus, Ranson stated : “The importance of the dorsal roots for muscle tonus has been overstated.” He pointed out that when the fibers of the ventral roots are uninjured they are capable of producing tonic contractions in de-afferented muscle. Such muscle can show tonic crossed extensor reflexes, labyrinthine tonic reflexes, and conditions of decerebrate rigidity, according to Ranson. De-afferented muscle, then, not only recovers part of its tonus, but, according to this same observer, may show hypertonicity.

Over a half century ago, Ranvier (’74) pointed out the thin red and thick white fibers in skeletal muscle and showed that the former required a less number of stimulations per second to produce continuous contraction and had a much longer latent period (four times as great) than the latter. Griitzner (’83),

A very considerable amount of work has been done on the histologic and physiologic differences in these types of muscle and on the significance of those differences. KUhne (’05), after careful removal of all blood from the vessels in the material studied, was able to prepare various derivatives of hemoglobin from water extracts of finely divided red muscle and Lankester (’71) thought that a pigment isolated from the pharyngeal muscles of Linnaeus and Paladina was identical with hemoglobin in spectroscopic tests. Keilin's (’25) results are in agreement in cs.sentials with those of Lankester, for although by spectroscopic examination he could detect a slight difference in the hemoglobin of the muscle as compared with that of the blood, this difference Is less than that existing between the hemoglobins of different species. Histologically, the red muscle fiber of mammals is distinguishable from the white not only through its pigment but because of its smaller size, its relatively greater amount of cytoplasm, and the fact that its nuclei, like those of cmbrj’onic muscle, frequently lie away from the sarcolcmma out in the muscle substance. Sucli muscle fibers are sometimes spoken of as protoplasm-poor and protoplasm-rich fibers (Knoll, ’91, “ protoplasmarmc und protoplasmarciche Musculatur”). In amphibians and reptiles where pigment cannot be used as an indicator of a particular type of fiber, since the great majority of the fibers are white in color, there is found in addition to the intrafusal fibers of the neuromuscular spindles, medium-sized muscle fibers, supposedly di-stinguisliablo by granules with special affinity for the gold chloride of the Ilanvicr technique, and clear, large fibers (Hines, ’27, for the frog). .•Vlthough certain ob.servers have inferred such to be the case, evidence is lacking (as Hines pointed

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 241

Hay (’01), Cobb and Fulton (’25), Adrian (’25), and others have substantiated the existence of a longer latent period and a slower and more prolonged contraction of the red muscle fibers.

The presence of these two types of fibers with such different physiologic characteristics aroused the interest of various observers and raised the question as to whether or not they were innervated alike. It will be remembered that Boeke had described an unmedullated accessory fiber, in addition to the usual medullated motor fiber with its typical ending, and that he regarded this accessory fiber as sympathetic. Mosso (’04) had suggested that the maintenance of muscle tonus was a fimction of the sympathetic. De Boer (’13, ’13a, ’14) and Langelaan (’22) obtained loss of muscle tone on the homolateral side of frogs and cats from which the abdominal sympathetic chain had been excised. However, the work of Cobb (’18) and of van Rijnberk (’18, ’18a) did not support the idea of any tonic effect of the sympathetic upon skeletal musculature. Influenced by a paper of Kulchitsky (’24), in which this latter observer described in the snake (python) a sympathetic innervation for the granular muscle fibers and a somatic motor innervation for the clear fibers, Hunter (’24 and ’25) and Hunter and Latham (’25) predicted a similar innervation for the avian and ma mmali an red and white muscle fibers, the former fibers, in their opinion, receiving the sympathetic which provided thus for the maintenance of plastic tone, the latter the somatic motor innervation, concerned with body movements. They also attempted to show the application for these notions to the surgical treatment of cases showing hypertonicity. Hunter himself used goats, sea gulls, and fowls in preparing experimental evidence for his claims, and his work on fowls appears to have been corroborated by the work of Kuntz and Kerper (’26) on these birds and on pigeons, in which they obtained a loss of muscle tonus in the wing muscles after cutting the dorsal roots and then the sympathetic trunks. Unfortunately, the early death of the brilliant Dr. Hunter prevented his offering a satisfactory verification of his ideas, but investigators in anatomical, physiological, and surgical fields have carried on investigations with a view of solving the problem if possible.

For the most part, satisfactory anatomical and physiological evidence for any fundamental difference in the innervation of red and white fibers is lacking. Latham’s technique, on which part of Hunter’s evidence was based, appears to have been inadequate {Hunter and Latham, ’25 ; see also Hines, ’27). In the kitten and in the young guinea pig, Cushing Smith (quoted from Hines, ’27) found non-medullated nerves supplying single small fibers and medullated nerves with sole-plate endings terminating on the larger muscle fibers in the field. Certain other observers have agreed with the earlier work of Hay (’01) and have found

out) that these granular fibers of reptiles and amphibians are the homologues of the red muscle fibers — and the clear, of the white muscle fibers — of mammals.

There is no general agreement as to the distribution of white and red muscle fibers in mammals. While certain skeletal muscles in such animals as the rabbit {Meyer, ’75 ; Ranvier, ’80) appear to be composed entirely of red fibers and others entirely of white, the majority, and possibly all of the skeletal muscles of the body, in the highest mammals, and particularly in man {Roberts, ’16 ; Starling, ’20 ; Hines, ’27), contain an admixture of both red and white muscle fibers.

242 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

that both types of muscle fibers have typical motor endings, which even may be branches of a single axis cylinder (Garven, ’25 ; Hines, ’27) in the frog.

Tower (’26) suggested that the sympathetic may indirectly influence muscle tonus through its innervation of the blood vessels to the region and hence its influence on the blood supply of the muscle. Among those who have been unable to get satisfactory evidence that the sympathetic innervation of muscle fibers influences muscle tone may be mentioned Cohh (’18), Newton (’24), Tower (’26), Coman (’26), and Ranson (’30). Finally, the surgical application of Hunter’s ideas as represented in an attempt to relieve hypertonicity has not met sufficient success to warrant its use, in the opinion of many surgeons {Crothers, ’25 ; Kanavel, Pollock and Davis, ’25).

Some authors believe that the sympathetic fibers only influence the sarcoplasm and that their destruction causes earlier fatigue of the muscle. The question is obviously unsettled, although at present the accumulated evidence is rather against such a function of the sympathetic than in favor of it. It is entirely beyond the scope of the present text to discuss all the literature on this subject (see, however, pp. 40 to 42). An excellent review of the literature up to the time of publication of his paper on the dual innervation of muscle is to be found in the article by Wilson (’21), and a r4sum4 of much of the literature dealing with the relation of the sympathetic to tonic innervation of muscle is available in the papers of Riesser (’25), Cobb (’25), Ranson (’26), Ranson and Hinsey (’26), Hines (’27), Forbes (’30), and others.

The influence of higher centers upon the sympathetic system is considered in Chapters VIII and X, and need not be entered into here. Reference is made to pages 1185 and 1186.

THE PRIMARY NEURONS OF THE VISCERAL AFFERENT SYSTEM

As was stated previously, visceral sensory fibers from visceral surfaces, such as the mucous membranes of the digestive tract, accompany post- and preganglionic fibers toward the cord. They often pass in their course through sympathetic ganglia, but their cells of origin lie in cranial or dorsal root ganglia, and, if in spinal nerves, their neuraxes enter the spinal cord through the dorsal roots. Their nuclei of termination within the cord are not certainly known. Takahashi (’13), however, has described an area near the intermediolateral cell column, which shows a resemblance in general characteristics to the substantia gelatinosa and which he regarded as visceral sensory. This is certainly the region in which a visceral sensory center might be expected to lie.

According to Dogiel (’96, ’97), cells within the sympathetic ganglia send processes to arborize around spinal ganglion cells and by certain neurologists are regarded as providing the mechanism for referred pain, i.e. affording a means by which sensations arising in the viscera may be referred to the surface of the body. As yet, observers in general have not confirmed Dogiel’s findings in regard to these cells, although Hirt (’28) figured such neurons.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 243

THE PRIMARY NEURONS OF THE SOMATIC AFFERENT SYSTEM

a. The Spinal Ganglia (figs. 6 to 8, IS to IS).

The structure of the mammalian spinal ganglion cells (Dogiel, ’96, ’97, ’08 ; Levi, ’08 ; Marinesco, ’06 ; Ramon y Cajal, ’07 ; Bielschowsky, ’08 ; Huber and Guild, ’13 ; Hirt, ’28) has received frequent consideration. They have an origin similar to those of birds and reptiles. The spinal ganglion neurons of mammalian embryos are chiefly bipolar cells which acquire monopolarity secondarily. However, multipolar types have been found in mammals (von Lenhoss6k, ’98 ; Dogiel, ’08 ; Ramon y Cajal, ’09). The cell bodies vary in shape from spherical or flask shaped to a mushroom type. Such cells contain tigroid granules, largely concentrated around the nucleus (von Lenhossek, ’95). The tigroid granules are much finer than those found in the ventral horn neurons. A centrosome has been described (Buhler, ’98 ; Hatai, '01 ; van der Stricht, ’06), although its function is uncertain. Neurofibrils, a Golgi apparatus (Golgi, ’98) and a canalicular system of Holmgren (Holmgren, ’00, ’14, etc. ; Sjovall, ’06, see bibliography for Chapter I) are present. In other words, a spinal ganglion cell has the structural characteristics of a nerve cell. Surrounding each spinal ganglion cell there is found a pericellular capsule composed of pavement or low cubic cells of ectodermal origin, often termed endothelial cells by earlier writers, which has been described by many observers (Dogiel, ’96, ’97, ’08; Ramdn y Cajal, ’09). According to Ramdn y Cajal (’09), fine collagenous fibrils form a plexus wliich completes this capsule on the outside of the cells. Inside of this capsule, Ramdn y Cajal and Oloriz (’97) have described certain stellate or fusiform stellate cells which send out processes to the cell body of the inclosed spinal ganglion cell. Passing off from certain of the smaller spherical ganglion cells is a single process which passes, without coiling, out of the pericellular capsule, which is continued over it as the neurolemma sheath, and into the fiber bundles in the central portion of the spinal ganglion, where the process bifurcates in a T- or Y-shaped division, one process passing out toward the periphery as a dendrite and one entering the central nervous system as a neuraxis. It is probable that these smaller spinal ganglion cells ^ve rise to thinly medullated or unmedullated nerve fibers. Certain large spinal ganglion cells with uncoiled processes which pass down in a similar manner to T- and Y-shaped divisions have been described by Ramdn y Cajal (’07). The majority of the mammalian spinal ganglion cells have a single process which, almost immediately on leaving the cell body, forms a system of loops or coils which surround the cell body at one pole, lying between it and the pericellular capsule, and is ultimately continued into a more or less straight process (surrounded by myelin sheath and neurolemma) which forms the usual T- or Y-shaped division Mthin the fiber region of the ganglion. The coiled portion of the process was termed the initial glomerulus by Ramdn y Cajal (see ’07), and although the term in implication is a misnomer, it has been retained, the name “ glomerular process being used rather generally. A wide range of variation in the details of the glomerular structuie is seen in the figures of Dogiel (’96, ’97, ’08), Ramdn y Cajal (’09), Huber and Guild (’13), and

244 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

others. Frequently a number of processes may arise from a special ganglion cell which ultimately unite to form a single coiled process. In certain of the spinal ganglion cells, the cell body shows curious cytoplasmic loops and projections which give it a fenestrated appearance. In many of the cells, fine branches are given off from the processes which turn back to terminate on the cell body in end disks. These were first described by Huber (’96) in amphibians, and later for mammals by Dogiel (’08), Ramdn y Cajal (’09), and Huber and Guild (’13). Ramdn y Cajal designated these processes as the dendrites of the “type de Huber,’’ and Huber himself regarded these processes as of dendritic significance. Dogiel’s (’08) comprehensive monograph on the spinal ganglia of man and mammals, which is extensively illustrated with figures based on methyleneblue preparations, should be consulted for detail as to the forms and varieties of these processes of spinal ganglion cells. Other processes of similar appearance may terminate outside the pericellular capsule and may take origin from the single process outside the capsule {Ramdn y Cajal, ’09; Huber and Guild, ’13). Windle (’31) and O'Donnell and Windle (’33) described cells in the ventral roots of marmnals which they regarded as spinal ganglion cell types. Haiai (’02) enumerated the large and small cells in the spinal ganglia of the adult white rat and found that in three selected regions, approximately 60 per cent of the cells belonged to the so-called small type of ganglion cells, which are in an undeveloped condition. Moreover, Haiai (’02) found that the number of cells in a spinal ganglion is always more than twice the number of medullated fibers of the corresponding root. Warrington and Griffith (’04), working on the second cervical nerve of cat, found that almost 70 per cent belonged to the small “obscure’’ cells. Roughly speaking, about two-thirds are small ganglion cells. A number of observers {Hodge, ’89 ; Gaule and Lewin, ’96 ; Biihler, ’98 ; Haiai, ’02 ; Ranson, ’12 ; and Hardesty, ’05 — for frogs) have shown that the number of medullated fibers in a spinal root is much less than the total number of ganglion cells of that root, there being, according to Ranson (’12), 32 cells per one medullated fiber. It is supposed, then, that the small cells are associated with the unmedullated fibers ; this would imply that there are more unmedullated than medullated fibers in at least certain of the spinal nerves. ’® Intrinsic cells {Dogiel, ’08 ; Hirl, ’28) and efferent (“ sympathetic,” Kiss, ’32) neurons (see pp. 236 and 239) have been described in spinal ganglia. Fisher and Ranson (’33) have justly criticized the results of Kiss.

b. The Distribution of the Spinal Ganglion Cell Dendrites to the Skin.

Much attention has been given to examining and plotting the somatic sensory areas of the skin and the work has been carried on by various methods. Bolk

“ Davida (’80), Railione (’84), Slreeicr (’09), and Nichols ('ll) h.ave spoken of double spinal Kiinglia jus-sociated with certain of the spinal nerves. Nichols found that, in man, in tlio lumbar and upper sacral regions there is a marked tendency for tlie ganglia to double with or without complete separation. This is particularly noticeable in the 3rd to 5th lumbar region (where Nichols found an e.vtra ganglion on an average in every third case). An acccssorj’ ganglion sometimes found in the cer\’icat region is called the ganglion of Ilyrtl. The irosition of sacral and caudal spinal ganglia has been considered by Holmdahl (’00).

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 245

(’95, ’97, ’98, ’98a, ’99, etc.) examined them morphologically in the human body, tracing every root from the cord to the sensory terminations in the skin. In this way the skin areas of all the spinal nerves were established (with the exception of the first cervical, which has no sensory root). By this means he found,

Fig. 112. The dermatomea of the human body, accordmg to Bolk. The segments are numbered throughout with Arabic numerals. The 2nd segment, not present m the figures, is on the head.

on comparing the tnmk segments with the spinous processes of the two vertebrae between which a nerve left the cord, that the trunk segments had shifted caudalward. From figure 112 it is clear that the spinous processes of vertebrae are caudal to the corresponding segments of the cord. Balk’s work indicates that the skin areas are likewise caudal to the spinous process of the vertebra corresponding in number to the nerve. Thus there is evidently a great shifting

“ The first cervical nerve has no sensory root (only a motor root) This is a reduction which appeared in the upper spinal cord of fishes (compare Carcharias, fig 72) The areas innervated by the dorsal rami of the sensory roots are smaller in length and breadth than those innervated by the ventral rami. As a result, the entire dorsal field is smaller.

246 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

caudalward of the position of the dermatomes as compared with the location of the cell bodies of the neurons innervating them. Moreover, not all roots participate in the task of supplying the skin areas, since dorsal rami for the trunk region are lacking for the 7th and 8th cervical. Consequently the dermatomes do not follow each other along the back in a perfect series ; for example, the first thoracic segment lies directly behind the 6th cervical segment. A similar gap occurs in the lumbar region on the dorsal side of the trunk where the 3rd lumbar is followed immediately by the 1st sacral. The border between the dorsal and ventral skin areas is termed the dorsolateral line (see fig. 112). This commences at the head and extends to the upper side of the os coccyx. An examination of the areas dorsal and ventral to the line indicates'that the dorsal and ventral branches of an equivalent root do not always innervate adjoining areas. In the first place, a mutual displacement occurs in a dorsoventral direction (see left, fig. 112). Secondly, the dorsal and ventral divisions of the 6th cervical nerve and the 1st thoracic lie far apart and are separated by skin regions of other roots. In the sacral region, the 1st sacral dermatome (dermatome 26) lies partly on the sacrum and partly on the lower leg, with the result that disease of a single spinal cord segment may cause pain in two entirely different places. The absence of the dorsal dermatomes mentioned above is probably associated with the formation of the segments for the extremities from the lower cervical and from the lumbosacral regions. Undoubtedly the segmentation of the trunk, as presented hy.Bolk, has great clinical significance (see fig. 113).

In the formation of the upper extremities, first the 7th and 8th cervical segments shift out from the trunk into the extremity anlage. Then an extension of the 6th cervical occurs (partly followed by the 6th) and of the 1st thoracic. In conformity with this pattern, the 7th and 8th cervical segments in the adult innervate that part (the hand) which is removed farthest from the trunk, and are followed toward the trunk (figs. 114-115) by the other segments. A similar process occurs with the lower extremities, only that here it is complicated by the torsion of these extremities, this torsion being made particularly clear by the arrangement of the dermatomes (figs. 116-117).

The overlapping of the segments, which is so conspicuous in lower animals, also occurs in mammals. Sherrington (’93), Winkler and van Rijnberk (’01, '02, ’03) and their collaborators, also Dusser de Barenne (’11a), S. de Boer (’18), and Klessens (’13 and ’13a) have studied these overlappings in various ways. In the cat this actually amounts to two-thirds, in the monkey (Macacus) to one-half, and in the dog to one-third of the adjoining segment. Consequently loss of sensation is sometimes not evident after the section of one dorsal root.

Many observers have adopted Langley’s method of determining the residual sensibility by cutting roots (usually two above and two below) on either side of the nerve to be studied and then determining the distribution of the uninjured nerve. A different method, used by de Boer, depends upon rendering the emerging dorsal root hyperalgesic by the application of strychnine, according to the method of Dusser de Barenne, and then determining the hyperalgesic skin zone.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 247

This has an advantage in that the less sensitive border zone of the segment {Winkler and van Rijnberk) appears more distinctly.

A study of certain pathologic cases suggests that in man the overlap is not so great as with cat, dog, or monkey, and that the dermatomes, as indicated by Bolk, are clinically reliable if a small correction is made for the overlap. Vari

Fiq. 113. Transection of the human spinal (segments 18 and 19). Brouwer.

cord at the level of the 10th and 11th thoracic The shaded portion was anaesthetic.

ous attempts at determining this overlap in man have been made {Thorbum, '93, and others), but to Brouwer ('14, '15a) belongs great credit for his painstaking and accurate study of the matter. This observer has shown that the overlap is not so great in man as in smaller animals. Bolk and Sherrington had called attention to the fact that one must have regard for the outer form of the body in considerations of the breadth and extent of the myotome. Brouwer (’14, ’15a) expanded this theory to explain the differences in extent of the dermatomes in man and lower animals. He was the first to point out that if a body or a part of a body enlarges, the root areas are drawn apart, a conception which has also

250 NERVOUS SYSTEMS OF VERTEBRATES AND OF IVIAN

protopathic of Head’s and his associates’ terminology — are the first to reappear, while those for finer perceptions and discrimination — the gnostic or epicritic — return later, just as they appeared later in phylogeny. Consequently, there exists an analogy between the phylogenetic and regenerative sequence of skin perceptions.

This work of Head and his associates {Head, Rivers and Sherren, ’05 ; Rivers and Head, '08 ; Head, ’20) relates itself to the work on nerve overlap. It was found that whenever a peripheral nerve was divided in the forearm or in the

hand, protopathic sensibility was lost over an area of smaller extent than was epicritic. The nearer the lesion lay to the central nervous system — particularly if it cut a dorsal root — the more definite and extensive was the loss of protopathic sensibihty. Head regarded the unit of the protopathic system as situated in one or more dorsal roots, while the unit for the epicritic system was allocated to one of the larger peripheral nerves. The protopathic fibers overlap to some extent, but the amount of overlap of the protopathic fibers carried by any one dorsal root is considerably less than that of the epicritic. This wider spread of the more delicate epicritic {Head) or gnostic {Ariens Kappers; Introduction, p. iv) sensations agrees with de Boer’s conception that neurobiotactic factors determine the outgrowth of the fibers, since the more sensitive epicritic or gnostic nerve fibers must also be more sensitive to neurotrophic influences.

Not all neurologists have agreed with the interpretations of their results offered by Head and his colleagues. For example. Pollock (’20), in a clinical study of 500 war cases, arrived at the conclusion that the early return of pain is due to nerve overlap and that this sensation does not reappear early where the anatomic distribution of the neighboring nerves does not permit them to assume the function of the injured fibers.

THE SECONDARY CENTERS WITHIN THE CORD

At their entrance into the spinal cord, two bundles can be distinguished in the dorsal roots : (1) a lateral bundle of finer fibers which branch in the marginal layer and in the substantia gelatinosa, and (2) a larger, medial bundle of coarse fibers, concerned particularly, but not exclusively, with the formation of the dorsal funiculus. It is obvious that these two dorsal root bundles must be

Fig. 118. At the left, Macacus, according to SAem'nsIon; at the right, man, according to Bolk. The figures show the effect of the extension of the neck on the arrangement of the segments. Brouwer.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 251

credited with different functions, but before entering upon this question, there must be further consideration of the structure of the dorsal horns themselves. Three divisions occur in the dorsal horns of mammals (fig. 119 ; Ranson, T4, ’31 ; Keenan, ’29 ; and others) ; first, a marginal zone or zona marginalis, which is also present in lower animals, varying with the development of sensibility ; secondly, the substantia gelatinosa Rolandi, present in all vertebrates from fishes on ; thirdly, the body of the dorsal horn which forms the principal part of the horn in mammals and which is present in lower vertebrates, although in var 3 dng forms

Z M.

Fig. 119. Section through the second cervical segment of Hippotragus niger. Dors.r., dorsal root; N.XI, nervus accessorius; Z.M., zona margmalis; S.G.R , substantia gelatinosa Rolandi; H.h.K., body of posterior horn ; Proc ret., processus reticularis.

(fused in fishes into the corpus commune, probably ; Keenan, ’29). The marginal zone of Waldeyer or Lissauer is the most dorsolateral of the three divisions and usually extends from the substantia gelatinosa Rolandi to the surface of the cord but may extend into the lateral funiculus also {LeszUnyi, ’12 ; Sana, ’14 ; Keenan, ’29). The zone becomes narrow and deeper in those regions in which the substantia gelatinosa lies farthest from the surface {Keenan). In Weigert and silver preparations it appears to consist mainly of a great number of fine, partly medullated and partly non-medullated fibers (fig. 119). In part, at least, it is the path through which Lissauer’s tract passes to the substantia gelatinosa. Its gray substance contains some neurons, the dendrites of which form a border layer between the gray substance of the horn and the adjoining white substance. There they enter into synapse with the collaterals of the dorsal roots. Less frequently a dendrite of these cells enters the depth of the gray substance of the dorsal horns {Ramon y Cajal, ’09). These cells are apparently funicular cells which in lower animals lie embedded in the body of the dorsal horn itself but here are partly displaced toward the periphery of the cord.

252 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ramon y Cajal supposed that the displacement of these cells results from the lack of space, believing that the number of funicular cells has so increased in mammalia that there is not sufficient place for them within the area occupied by the body proper of the dorsal horn. It is, however, clear that this does not explain why the displacement should occur toward the dorsal surface instead of frontally or laterally. Probably displacement is in the direction of the incoming stimuli from the dorsal roots, due to neurobiotaxis (Ariens Kappers).

Fig. 120. Section through the sacral cord of Gazella dorcas. Bianchi. Figure demonstrates the great development and the convolutions of the substantia gelatinosa.

In the mammals studied thus far — with the apparent exception of Cetacea {Ariens Kappers) — the substantia gelatinosa is present throughout the extent of the cord although it varies in amount at different levels. According to the tables of Sana (’09) it is greatest in amount at the level of the lumbo-sacral enlargement, shows another though relatively smaller increase at the cervical enlargement, and has its least development in the mid-thoracic region. Keenan’s results (’29) appear to be confirmatory. The substantia gelatinosa varies in its degree of development in different mammals and there is reason to suppose that this variation is influenced to some extent by the development of the peripheral sensibility of the organs and possibly particularly those associated with the hairs {Ariens Kappers). Proof for this statement is found in the absence of this gray mass in Cetacea except as a spinal trigeminal nucleus (Hatschek, ’96a), since these animals have poorly developed sensibility of the skin and lack hair. However, the work of Ranson (’14) and Keenan (’29) indicated that there is no direct relation between the size of the marginal or Lissauer’s zone and that of the substantia gelatinosa in a given animal.

The substantia gelatinosa Rolandi forms a cap over the body of the dorsal horn, intervening between this latter and the marginal zone. In some mammals — such as ungtflates — it is very well developed (see fig. 120). This is par

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 253

ticularly the case with Gazella dorcas {Biach, ’07), where its mass is further increased through the formation of convolutions. In the dog and in Simia Satyrus (fig. 110 ; also Keenan, ’29) the Rolandic layer of one side may become continuous with that of the other side so that dorsally the entire gray substance is covered by a cortex-like, so-called gelatinous layer. It will be seen in the following chapters that such surface extensions of gray substance are typical of highly organized sensory nuclei with pronounced local functions {Ariens Kappers, ’14) ; therefore, their appearance here is to be expected.

The structure of the cells in this region is in agreement with their tendency to layer formation. The dendrites of its cells, which for the most part are small, tend to branch in a single plane which is parallel to the dorsolateral surface of the cord. As seen in transverse sections, these cells appear relatively flat, so that in such sections their dendrites are not traced readily into the gray matter of the cord (Ramon y Cajal, ’09). The dendritic branches tend to spread over the gray of the body of the dorsal horn so that the gray has a cortex-like arrange- ficial arrangement of the cells and dendrites of the ment. The gelatinous appearance resuits from the presence of many un medullated or finely meduUated fibers and the presence of many dendrites, which form a rich protoplasmic net. The neuraxes of these cells are not long. They distribute either to the fasciculus proprius of the lateral funiculus or send collaterals to other cells of the gelatinous substance (Ramon y Cajal, ’09). Functionally considered, this substantia gelatinosa Rolandi has to do with vital sensibility. It may be concerned also with vasomotor and pilomotor reflexes (Sana, ’09). Its relation to the descending root of the trigeminal is important in this respect.

In the body of the dorsal horn, too, there are found several nuclear masses which in many cases have not been analyzed sufficiently but constitute nuclear groups which vary somewhat at different levels. Of the nerve cells of the dorsal horn, one well-known group deserves special consideration and is here described as the nucleus dorsalis of Clarke (’68). It consists of a clump of nerve cells situated at the base of the horn on its medial side, from the last cervical to the first (sometimes the third) lumbar segment. It is essentially, then, a nucleus characteristic of the thoracic region (fig. 109). In cell type it consists cluefly of large cells, somewhat round in outline in cross section, fusiform in longitudinal (Hvher, ’28 ; fig. 123), and with short dendrites largely confined to the column itself (Ramon y Cajal, ’09 ; fig. 124). They are rich in chromopffilic substance. According to some observers these cells frequently show pigmentation. They are

254 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

interspersed with smaller cells with relatively few, small, chromophilic granules. Only the large cells show chromatolysis after section of the dorsal spino-cerebellar fasciculus (van Gehuchten, ’01), so presumably it is this cellular mass which furnishes the cells of origin for the neuraxes constituting the fasciculus in question. In the cervical cord, the cervical nucleus of Stilling replaces, at least as regards position, the dorsal nucleus of Clarke (’68).

Collaterals of more medial dorsal root fibers pass to dorsal and possibly ventral horn gray. Some cross. Dorsal horn neurons send neuraxes to ventral horn cells and with intrinsic and commissural cells of the cord distribute to higher and lower adjacent segments for purpose of intersegmental reflexes by means of propius bundles (see fig. 109). Lateral dorsal root fibers end in the gelatinous substance.

SECONDARY ascending PATHS

a. Paths within the Cord for Exteroceptive Pain and Temperature

It is possible to turn now to a consideration of the main ascending paths of the spinal cord (see fig. 109). So far as these paths are known, they fall into two

major exteroceptive groups and two proprioceptive groups. The ascending paths for the interoceptive impulses have not been described, but it has been suggested that they may be represented by a series of neurons in chain formation, and that the bundle is probably unmyelinated (Davis, ’22 ; Davis, Hart and Crain, ’29 ; Huber and Crosby, ’30). An

F, 0 . 122 . Sectionthrougrthr^hthoracicsegment. Hun:an. ^xterocept^e group will receive consideration first.

Incoming impulses of pain and temperature enter the spinal cord over the processes of the smaller spinal ganglion neurons, those for pain over unmyelinated or finely myelinated and those for temperature over finely myelinated fibers (Ranson, ’13, ’14, ’14a, ’15, ’31; Sheehan, ’32). The fibers entering from the lateral division of the dorsal root occupy a position within the cord at the peak of the dorsal horn, where they divide dichotomously in order to form the dorsolateral fasciculus or Lissauer’s tract. The fibers of this fasciculus consist of the neuraxes which ascend or descend for only a short distance (since the ascending impulse crosses within a segment of its entrance ; see Forster, ’27), terminating then within the gray of the dorsal horn. After one or more synapses in the substantia gelatinosa, larger cells of this latter gray mass give rise to axons which collectively form the lateral spino-thalamic fasciculus. This fasciculus and the ventrolateral or ventral spino-thalamic fasciculus together, in mammals, are the mammalian representatives of the spino

THE COMPARATIVE ANATOMY OP THE SPINAL CORD 255

mesencephalic and spino-tectal systems (secondary ascending tract of Edinger) of lower forms.

Hie icscaichcs of Pclrcn (’02, '10, 'll) indicate that, after decussating in the anterior or ventral white commissure, the neuraxes forming the lateral spinothalamic libel’s run for a distance of four or five segments in the ventral funiculus near the gray matter of the ventral horn ; then they gradually shift lateralward into the ventrolateral funiculus, mid ascend, the more caudal fibers being the more superficial (Forster, ’27). According to PcKje M(uj (’00, Kohnstamm and Quensd, ’07, and others), certain of the fibers, after decussation, pass directly to the thalamus and tectum.

Some few lioniolateral bundles are present also. Others in this lateral spinothalamic bundle terminate in gray matter of the cord during tlieir course, the tract being joined by neura.\es of neurons of the next order. Thus a chain of several links is built up which ultimately terminates in the thalamus.

R(inso)i and Billingsley ('10, 'IGa; see also Ranson and von Hess, ’15) have shown that these short chains are much more prominent in the cat than in man.

In this carnivore pain is conducted chiefly through such a chain and has bilateral paths. The recovery of an appreciation of pain after complete hemisection of the cord is believed to be an indication that, even in man, a homolateral chain of neurons may replace in function the destroyed (Head,

’05, ’20) contralateral path. However, the temperature sensations are believed to follow crossed paths entirely or almost entirely. The spino-thalamic fibers are accompanied by spino-tectal fascicles which terminate in the superior and inferior colliculi. Mott (’92) has described and Ariens Kappers confirmed

in the human brain, the termination of fibers of this system not only in the roof of the midbrain but in the medial geniculate body as well, thus establishing a

Fio. 123. A portion of a longitudinal section of the human spinal cord taken at the level of the bases of the dorsal gray horns from the TA. VlII and Th. IX segments. The prominent cell column placed along the mesial side of each gray column represents the nucleus dorsalis or the column of Clarke. X approximately 5. G. Carl Huber.

256 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

relation between optic and auditory impulses and pain impulses from the skin and probably from muscle {Head and Thompson, ’06).^’ (Likewise tactile impulses brought from the cord by way of the ventral spino-thalamic fasciculus undoubtedly reach the tectum in part,)

The evidence for the conveyance of pain and temperature sensations in the lateral spino-thalamic fasciculus rests on experimental and clinical evidence {Bechterew a,r\dHolzinger,-’Q4i; Berlholet, ’OQ ; see also Page May, ’06; Rothmann, '06; Spiegel and Bernis, ’25 ; Spiegel, ’32). It is probable that the temperature

component may be separated spatially

from that of pain. This path for pain carries impulses evoked either through stimulation of epicritic or protopathic sensibility {Head, ’05, ’20) or visceral sensibility.

The almost immediate crossing of the pain impulses after their entrance to the cord is made use of in various ways ; thus it is used in the diagnosis of the condition of syringomyelia (a progressive affection of the ependymal region of the cord), Frazier and Spiller (’23) and Peel take advantage of the position of the lateral spino-thalamic tract close to the periphery of the cord by cutting the fasciculus to relieve pain in cases of cancer, particularly of the pelvic or near-lying regions, the results indicating that it carries visceral as well as somatic painful impulses (see p. 268). The same method has been used now and again to relieve the pain of tabetic gastric crisis (however, according to Davis, Hart, and Crain, '29, not always with marked success).

In their course cephalad through the spinal cord, medulla oblongata and pons, the lateral spino-thalamic and spino-tectal systems lie in intimate relation with the ventral spino-cerebellar tract and it is believed by certain observers that they may contribute a few bundles to this lateral fiber system which provide a means for the conveyance of painful sensations to the cerebellum {Kohnslamm, '02 ; see also Page May, ’06) . The main fiber bundle joins the other lemnisci systems at the upper levels of the pons, proceeding forward to the midbrain and diencephalon (fig. 521).

■ = 7

Fig. 124. Nerve cells in the body of the dorsal horn. Ramdn y Cajal. A, Clarke’s column; B, medio-central nucleus ; B, central canal ; P, dorsal funiculus.

b. Paths vnthin the Cord for Tcuitile Impulses and Proprioceptive Paths of the Dorsal Funiculi.

As was stated previously, the dorsal roots on entering the cord divide into a thin unmyelinated and faintly myelinated lateral ramus and a medial ramus

" D6j6rine is inclined to believe that painful impulses from muscle may travel in the dorsal funiculus, but this is improbable.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 257

consisting of coarser medullated fibers. The fibers of the latter arise from the large cells of the spinal ganglia and, soon after their entrance into the dorsal funiculus, divide dichotomously, forming a long ascending and a short descending branch. The descending branches form, in the lowest third of the cord, the triangular field of Philippe and Gombault, through the midregions of the cord, the oval field of Flechsig, and, in the cervical regions and the beginning of the thoracic cord, the comma field of Schultze. The triangular field and the oval field together are frequently termed the septomarginal field or fasciculus, while the comma field is called the fasciculus interfascicularis. From both the shorter descending and longer ascending fibers terminal branches are given off at all levels of the spinal cord which end around the appropriate neurons of the spinal cord gray. The fibers here described may be divided into two broad groups on the basis of the types of impulses carried, for, in part, the spinal ganglion neurons from which they arise are connected peripherally with terminations stimulated by tactile and general sensibility impulses and, in part, they are in relation wth neuromuscular and neurotendinous terminations and Pacinian corpuscles. Thus the fibers are partly conveyors of exteroceptive and partly of proprioceptive impulses. Those carrying exteroceptive impulses will be considered first.

Tactile impulses brought into the spinal cord through this medial division of the dorsal root are distributed to many levels of the cord and from the upper part of the body and upper extremities, at least, may even reach the medulla oblongata through the long-ascending branches which form the dorsal funiculi of the spinal cord (fasciculus gracilis and fasciculus cuneatus)."** At all levels of the cord, neurons of the dorsal horn gray, around which these primary tactile fibers terminate, send neuraxes to the contralateral side of the cord through the anterior or ventral white commissure. Such neuraxes turn forward sfightly after crossing, but soon accumulate in the ventrolateral portion of the cord ventral to the ventral horn gray, forming the ventral or ventrolateral spino-thalamic fasciculus in which they ascend throughout the cord. This fasciculus, as was noted for the lateral spino-thalamic fasciculus, is believed to carry a few homolateral fibers. Added to at all levels, this tract ascends in this position throughout the cord. As the upper levels of the cord are reached, it enters the meduUa

« Rasmussen’s (’32) diagrams and explanations indicate that the majority remain uncrossed until the medulla oblongata is reached.

Fio. 125.

Vent, horn cella

Reflex collaterals of the dorsal funiculus in the mouse, von Lenhossek.

258 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

oblongata with other bundles of the ventral funiculus and lies in the latter brain segment near the midline, dorsal {Dejerine, ’01) or slightly dorsolateral {May, ’06, and others) to the medial lemniscus and ventral or slightly ventrolateral to the medial tecto-spinal tract. Other observers {Rasmussen, ’32, and others) have regarded it as retaining a lateral position through the medulla oblongata. In its further course it accompanies the medial lemniscus, swinging lateral and then dorsolateral with the latter bundle in order to reach the diencephalon, where it distributes to the lateral part of the dorsal thalamus, according to some observers to the ventral and, according to others, to the lateral nucleus (see fig. 521).

In addition to the ventral spino-thalamic path, a spino-tectal path of similar origin is associated with the lateral tecto-spinal path, which it accompanies to the tectum. It also is a part of the primitive spino-mesencephalic system of Edinger (’08) and others.

Clinical and experimental work {Petren, ’02; Fahritius, ’07a, ’10, ’12; Brouwer, ’15a) has shown that finer perceptions involving finer tactile discrimination or two-point discrimination pass over fibers which do not terminate in the gray of the cord at or near their levels of entrance, but which run frontally in the dorsal funiculi for long distances and (except for the lowest levels of the cord) enter the medulla oblongata, where they terminate around the cell bodies of nucleus gracilis and particularly nucleus cuneatus. They may give off collaterals in their course. From these nuclei of the medulla oblongata, the impulse is carried forward over the medial lemniscus to the lateral division of the thalamus. This specialized tactile sensibility is increased greatly in manunals and provision for its course through the dorsal funiculus to higher centers is associated with increased dorsal funiculi (see p. 261).

With regard to the path for light touch, as against discriminative tactile and general tactile sensibiUty, there is still some question. Its persistence after the disappearance of two-point discrimination, in many clinical cases, suggests that it has a distinct course. This has been emphasized recently by Rasmussen (’32), who has suggested that it may ascend in the dorsal proprius region (the lateral portion of the dorsal funiculus) and cross with the sensory decussation in the medulla oblongata, although the relations with nucleus gracilis and nucleus cuneatus are unclear. The course of these fibers requires further demonstration.

The “stepping up” of the tactile impulses through the dorsal funicular regions of the spinal cord is of very considerable significance. Pain impulses have practically all crossed within about a segment {Forster, ’27) above their point of entrance. Consequently a hemisection of the cord is followed by a loss of pain sensibihty on the opposite side of the body, beginning with about one segment below the point of the lesion, while tactile sensibility remains. This loss of pain with the retention of tactile sensibility but with a motor paralysis on the same side of the lesion and certain proprioceptive disturbances is known as the BrownSequard syndrome.

Consideration must now be given to the bundles which carry proprioceptive impulses within the central nervous system. Such proprioceptive impulses pass from the various proprioceptive terminations over medullated sensory nerves

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 259

to certain of the larger neurons of the spinal ganglia. The neuraxes of these gangUon cells, as did those carrying tactile impulses, enter the cord over the medial division of the dorsal roots, of which they form the greater per cent, and so reach the dorsal funicular region. Here the fibers divide dichotomously, the shorter divisions descending in one of the descending bundles of the region (the septomarginal fasciculus or the fasciculus interfascicularis), while the main fibers enter the fasciculus gracilis, for regions below the waist line, or the fasciculus cuneatus for regions above. The fibers of these fasciculi are in order, from sacral region most dorsomedial to uppermost cervical region lateral and ventrolateral, and with certain fiber bundles of fasciculus cuneatus ventral to fasciculus gracihs {Winkler, T8 ; Boh, ’28). At all levels the fibers give off collaterals to the spinal cord gray. The fasciculus gracilis is found throughout the extent of the cord, the fasciculus cuneatus for all levels above about the waist line. It is obvious, from the foregoing accounts, that these bundles carry proprioceptive impulses from neuromuscular, neurotendinous end-organs, cylindrical end bulbs, and Pacinian corpuscles and, as stated, also impulses of the tactile discriminative type.

A study of the manunalian spinal cord indicates that there is a marked frontal accumulation of fibers in the dorsal funiculi of the spinal cord which is to be regarded as characteristic of these forms {Brouwer, ’15a). If a comparison of the mammalian with the reptilian spinal cord of the cervical region is made, it is evident that there is a relatively much larger dorsal funicular region in the former than in the latter animals. This relative increase in mammals is a progressive increase from the lower levels of the spinal cord to the higher levels, and it is this progressive factor which is significant rather than its relative size at any particular level in comparison with the gray or white matter of that level. In reptiles the relative accumulation between the lumbar and the upper cervical cord is at most 5 per cent, while in man it is certainly more than 100 per cent. A comparison of the relative increase in the dorsal funiculus of the cervical cord as compared with lower spinal cord levels is easily demonstrated by gross inspection if the cords of a series of lower and higher mammals and man are available for study.

In table I (prepared by Arims Kappers) the relative sizes of the dorsal funiculi of the cervical cords of a series of mammals are compared with the sizes of the gray matter of corresponding levels. Table II, prepared by Brouwer, shows the percentage relations of the dorsal funicular regions to the total white substance of the cervical cord in a smular series of animals. The tables indicate that in higher, as compared with lower, mammals, there is a relatively remarkable increase in the percentage amount of the dorsal funiculi, whether compared with the gray or the white substance of the cord.

The increase in the dorsal funiculi in the larger members of a class is shown in Part B, particularly in the first table, where the percentage relations of the tracts to the gray are compared for the cervical region. Here large and small marsupials, carnivores, and monkeys are used. The table is completed by a comparison of two animals, mouse and elephant, of different classes but representing extremes in size.

260 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

TABLE I

Percentage Relations op the Dorsal Funiculi to the Gray Substance

Part A. Lower and Higher Animals

Cent

Macropus 103

Ursus malayanus 106

Cebus fatellus 112

Homo sapiens 185

Part B. Smaller and Larger Animals Marsupials

Small — Diclelphys marsupialis 41

Large — Macropus robustus 103

Carnivora

Small — Putorius putorius 46

Large — Ursus malayanus 106

Monkeys

Small — Oedipomidas oedipus 50

Large — Cebus fatuellus 112

Small — Mus musculus 22

Large — Elephas indicus 292

The causes underlying this relatively greater size of the dorsal funiculi in the larger animals are quite obvious, since it is a measure, in a sense, of the greater amount of body sensibility represented there.

TABLE II

The Percentage Relations of the Dorsal Funiculi to the Total White

Substance

op the Cervical Cord (after Brouwer)

Part A

Part B

Lower and Higher Orders of Animals Per

Smaller and Larger Animals

Didelphys marsupialis . . .

. . 16.64

Marsupial

Cent

Lepus cuniculus

. . 21.06

Small — Didelphys marsupialis , .

16.64

Bos taurus

. . 22.02

Large — Macropus robustus . . .

21.82

Ursus malayanus

. . 23.86

Carnivore

Cebus fatuellus

. . 38.91

Small — Putorius putorius ....

18.27

Large — Ursus malayanus ....

23.86

Monkey

Small — Oedipomidas oedipus . . .

19.74

Large — Cebus fatuellus ....

26.43

Small — Sorex vulgaris

14.25

Large — Elephas indicus

32.25

Not only the dorsal funiculi but other parts of the white substance as well show an increase {Hovy, ’13), for it appears that in corresponding levels of the cord in different animals, the increase of the gray substance is as the square while the white substance (because its fibers are related to lower and higher levels as well as to the level under consideration) increases as the cube. Here comparisons should be made between the mouse, the agouti, and the elephant (fig. 126). There is less difference in the gray substance in the first than in the second list.

THE COMPARATIVE ANATOMY OE THE SPINAL CORD 261

It is still somewhat difficult to explain why there should be so remarkable an increase in the relative size of the dorsal funiculi in higher animals and particularly in primates, including man, although the underlying causes have been studied thoroughly by Brouwer (’15a). Muscle, tendon, joint sense — deep sensibility, in other words — reach thalamic and ultimately cortical centers by way of the dorsal funiculi to the funicular nuclei of the medulla and, after synapse, to the thalamus via the medial lemniscus. In addition to these proprioceptive impulses, tactile discrimination also passes forward over the fibers of the dorsal funiculi. Deep sensibility and tactile discrimination are greatly increased both as regards their peripheral terminations and their central connections, and this is reflected in the marked development of the dorsal funiculi of higher mammals and of man. Yet by

no means is the matter settled. Various observers {Peiren, ’02, ’10; Page May, ’06; and others) have pointed out the necessity for a finther analysis of the various components included under the term muscle sense and tendon sense. These include.

certainly, sensations of weight, of passive position, and of both active and passive movement, together with those involved in postural reflexes, and probably sensations such as the pull on the skin over a flexed

Fig. 126. A, Mouse; B, Agouti; C, Elephant. These figures illustrate the relative increase of the amount of white substance as compared with the amount of the gray substance in comparing the spinal cords of smaller and of larger mammals, de Vries.

joint and the like. Peripherally they arise from neuromuscular and neuroten

dinous terminations, from cylindrical end bulbs and Pacinian corpuscles, and from free sensory endings. Centrally, much remains to be known concerning the details of their distribution. In addition to their course in the dorsal

funiculi of the cord, various observers have recognized their probable course in the ventral funicular region as well. Thus Head and his associates (’05, ’20), basing their arguments on experimental and clinical work, have pointed out that while sensations of deep pressure reach the central nervous system over different fibers than those carrying tactile sensibility, yet once within the cord both types of sensibility run forward in the ventral spino-thalamic path as well as in other paths. Petre'n (’10), also, as a result of clinical studies, recognized a path forward for muscle sensibility in the ventral funiculus of the cord, although he carried ascending tactile impulses over bundles of the lateral rather than the ventral funiculus. It is possible that joint sense, unlike other proprioceptive impulses of the cord, is confined to the dorsal funiculus, in association with the late development phylogenetically of the extremities and the absence of Pacinian corpuscles in the lowest vertebrates (Ariens Kappers).

The intimate relation existing between the highly developed sense of discrimination and the impulses from joints, tendons, and muscles that is, between gnostic or epicritic and deep sensibility — all of which ascend through the dorsal

262 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

funiculus, permits the acquisition of a better judgment of spatial relations. The combination of these two senses forms a stereognostic sense, which is of the highest importance for an accurate knowledge of our surroundings. It is self-evident that the further development of this sense is connected with an increase of the dorsal funiculi. These dorsal funiculi phylogenetically show higher development only in terrestrial animals. This is not surprising if we consider of how much greater importance to life on land is a knowledge of various surrounding objects, and how much greater variability in movement, and greater discrimination, is required for adequate reactions to the surroundings under terrestrial than under aquatic conditions. It is evident that the separation and distribution of the impulses of pain, temperature, touch, and deep or proprioceptive sensibility — so masterfully built up by Pelren, Head and his co-workers, Sherrmgton, Ranson, Fabritius, Brouwer, and others (see references through preceding pages) for higher manunals and man — are intimately interrelated with the phylogenetic development of these systems, and that the greater development of the dorsal funiculi, particularly their frontal accumulation in higher animals — and especially in man — , is the logical outcome of the phylogenetic history.

The Brown-Sequard syndrome, obtained after hemisection of the cord, gives a striking illustration of the disposition of the proprioceptive and exteroceptive sensory pathways in the cord. There is a loss of perception of pain and temperature caudal to the hemisection but on the opposite side of the body and beginning about one segment below the level of the lesion. There is no disturbance in the appreciation of pain and temperature on the same side unless a peripheral nerve has been cut. Tactile sensibility remains on both sides, but careful examination for threshold indicates a loss of finer discrimination on the same side as the lesion. Great disturbance of proprioceptive impulses is evident on the same side below the injury ; homolateral conscious motor control is nearly (depending upon the amount of ventral cortico-spinal fibers) if not completely lost below the plane of the lesion. The above clinical results indicate that pain and temperature impulses passing to higher centers are all carried to the opposite side of the cord soon after their entrance ; that tactile impulses have both crossed and uncrossed ascending paths; that apparently those concerned with finer discriminations pass up on the same side ; and that at least the finer proprioceptive impulses, in their course by way of thalamus to cortex, ascend on the side of the cord at which they enter.

c. The Bulbar Terminalions of the Tactile and Proprioceptive Paths of the Dorsal Funiculi.

The nuclei of termination for the dorsal funiculi are nucleus gracilis and nucleus cuneatus, which lie at appro.ximately the same level in the lower medulla oblongata, the former cell mass beginning and ending slightly farther caudal than the latter but both of them extending cephalad to the level of the calamus scriptorius. It is logical to consider these nuclear columns, the nuclei of termination for many of the axons constituting the dorsal funiculi, in connection with the di-scussion of fasciculus gracilis and fasciculus cuneatus here described in the chapter

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 263

on the spinal cord. Nucleus gracilis and nucleus cuneatus as they relate more definitely to the medulla oblongata will receive consideration in the following chapter (p. 341). The eminences on the dorsal surface of the medulla oblongata due to these gray masses are the clava (nucleus gracilis) and the cuneate tubercle (nucleus cuneatus). Nucleus cuneatus receives the terminations of the fasciculus cuneatus, composed of ascending root fibers from the cervical and thoracic levels of the cord in general; the nucleus gracilis receives terminations of the ascending root fibers from the remainder of the cord by way of the fasciculus gracilis.

In marsupials (fig. 127) these nuclei are still very small, the nucleus cuneatus (nucleus of Burdach) showing a somewhat broader base than the nucleus gracilis (the nucleus of Goll). In other mammals various relations are observed. Thus in cetaceans the nucleus gracilis is very small, as is to be expected in a form lacking posterior extremities and with such poorly developed skin sensibility.

Nucleus cuneatus, although not large, is better developed in these animals, a fact associated with the presence of anterior extremities.

Frontally and laterally this latter nucleus gradually goes over into the nucleus magnocellularis of von Monakow (nucleus cuneatus externus), which receives descending vestibular fibers and sends fibers into the vermis of the cerebellmn (Yoshida ’24) and into von Monakow’s bundle (the rubro-spinal), a descending tract of the spinal cord (Anens Kappers). In the seal the nuclei of the dorsal funiculi are more highly developed than in cetaceans. Even a certain lamination occurs, such as is frequently characteristic of sensory centers, as has been previously mentioned in the description of the substantia gelatinosa of ungulates. Nucleus gracilis and nucleus cuneatus have a laminated structure in carnivores. In edentates (particularly in Myrmecophaga) both nuclei are highly developed. The more frontal part of the nucleus gracilis in these latter animals shows a well-marked dorsal extension. It forms a cover of gray substance along the margin of the brain stem, which, extending lateralward beyond the cuneate nucleus, sometimes comes into relation with the peripheral part of this latter nucleus or with its lateral differentiation, the nucleus cuneatus externus. From the lateral wing of the nucleus gracilis fibers arch down toward the inferior olive {Ariens Kappers). Similar relations of the nuclear masses are found in the chimpanzee. In Macacus Ferraro and Barrera (’35, J. Comp. Neur., vol. 62) described lamination. The greatest development of the nuclei of the dorsal funiculi is to be found in Atelidae and Cebidae (fig. 128) ; it is difficult, because of their size, to separate them at certain levels and they show a striking lameUation, the lamellae often being broken mto segments.

KadrfEscMf Kadpte. KsdosL

scending root of Vtb of Didelphys virgmiana Huber and Crosby

264 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The large size of the nucleus gracilis in these monkeys is associated with the great role the tail plays in their arboreal life.

In man both nuclei are relatively well developed. Although nucleus gracilis extends to the first cervical segment, it is smaller than is the same nucleus in the monkey. A dorsolateral extension of this nuclear mass is present and from this fibers extend ventrolaterally. Since they lie in relation with the trigeminal, they have been termed fibrae praetrigeminalis {Mingazzini, ’28). They are supposed to join the fiber layer over the inferior olive {Ziehen, ’99, ’03, ’13), but

their termination and significance are not understood at present. Possibly they may have connection with the inferior olive. In lower vertebrates (where no dorsal funicular nuclei occur) there is some indication that external arcuate fibers run into the inferior olive from the upper end of the cord {Ariens Kappers).

As with the alligator and certain birds, so also in some mammals either the nucleus gracilis of one side joins that of the other or a separate accumulation of cells occurs in the midline; this is known as the nucleus of Bischoff (’00). Since it receives the most medial and consequently the most caudal fibers Fig. 128. Lamination in nuclei gracilis of fasciculus gracilis, it is often considered as and cuneatus of Ccbus hypoieucus. concerned in the reception of ascending neu raxes of sensory cells innervating the tail. Its occurrence in rats, shrewmice, kangaroos, great anteaters, and some monkeys, and to some extent in cetaceans, appears to support this conception. It is not invariably present in all mammals with a well-developed tail, as Ziehen (’99, ’03, ’13) has pointed out. For e.xample, in Atcles and Cebus, where the hypertrophy of the most caudal sensory regions, particularly those of the tail, is so evident, no separate nucleus of Bischolf occurs. The factors producing the nucleus of Bischoff are not yet sufficiently understood ; the possibility of its acting merely as a nucleus of correlation for the nuclei graciles must receive consideration (see fig. 127).

Nucleus cuneatus does not consist of a single mass of gray matter but of several masses which, particularly dorsolaterally, are more or less separated by fiber bands. The nucleus cuneatus externus, to which reference has been made previously (nucleus of von Monakow or of Blumenbach, who first described it), receives not only descending vestibular fibers, but sends out also homolatcral descending fibers into the cord {von Monakow, ’83) and gives a.scending fibers to the vermis of the cerebellum {Yoshida, ’24). In the vicinity of the nucleus cuneatus, isolated bits of the substantia gelatinosa trigemini may be seen, sometimes embedded in the funiculus cuneatus. These are easily di.vtinguishable from the nucleus cuneatus externus through the small size of their cells {Karpins, ’04) and through their connection with descending trigeminal fibers.

THE COMPAEATIVE ANATOMY OF THE SPINAL CORD 265

The majority of impulses passing out of nucleus gracilis and nucleus cuneatus travel by way of the dorsal and ventral superficial arcuate and the deep arcuate fibers (internal arcuates). These latter arise as neuraxes of cells situated in the nuclei of the dorsal funicuh, pass ventromediahvard, and decussate in the sensory decussation. After crossing, the majority of them turn directly forward and form the main part of the medial lemniscus. This lenmiscus terminates chiefly in the ventral nucleus of the thalamus (its course and further terminations will be considered in a later chapter; see fig. 129).

Impulses arising from the nucleus gracilis and the nucleus cuneatus may pass to the cerebellum by way of the dorsal superficial arcuates which constitute one of the components of the inferior cerebellar peduncle (or corpus restiforme). Such a connection with the lateral part of nucleus cuneatus (von Monakow) was found by Yoshida*^

(’24) by Nissl degeneration after lesion of the cerebellum. This observation was confirmed by the account of de Kelyz and Hotldelink (’30), who saw disturbances of postural reflexes of the body in the rabbit, after lesion or stimulation of the nuclei gracilis and cuneatus. This is largely a homolateral tract. Other fibers pass ventralward and medialward as a part of the internal arcuate system of the medulla oblongata. Part of these, after crossing in the sensory decussation, form the medial lemniscus which continues forward to terminate in the ventral nucleus of the thalamus (as was stated above), but accompanying the fibers which cross to enter the medial lemniscus are fascicles which, after decussation, swing ventralward and forward around the pyramids, where they come into relation with scattered gray masses situated on the ventral side of the medulla oblongata, external and sometimes slightly lateral to the pyramidal tracts. These masses of gray matter are called the arcuate nuclei. Either with or without a synapse in them, the crossed fibers from the nuclei of the dorsal funiculi continue dorsolateralward as ventral superficial arcuates until they reach the inferior cerebellar peduncle or corpus restiforme and, by way of this peduncle, the cerebellum.

d. The Proprioceptive Paths in the Lateral Funiculus of the Cord.

Thus far particular attention has been devoted to those proprioceptive impulses which ascend through the dorsal funicuh of the spinal cord However,

« Yoshida did not believe that nucleus gracilis (nucleus of GoU) and the medial part of nucleus cuneatus (nucleus of Burdach) have such connections.

Fio 129. Cross section of the medulla oblongata of man, showmg the nuclei of the dorsal fumcuh, the internal or deep arcuate fibers, and the begmning of the sensory decussation and the superficial ventral arcuate fibers

266 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

not all proprioceptive impulses entering over spinal nerves reach the higher centers through the fasciculus gracilis and the fasciculus cuneatus. Other and, for the most part, phylogenetically older paths also serve such functions. Prominent among these is the dorsal spino-cerebellar tract (see fig. 134).

The neuraxes of the neurons in nucleus dorsalis or Clarke’s column form the dorsal spino-cerebellar fasciculus (direct cerebellar tract of Flechsig, 76). According to many observers this is entirely or mainly uncrossed, but MacNally and Horsley (’09) and Ramon y Cajal (’09) believed that the uncrossed bundle is joined, by way of the anterior or ventral white commissure, by fibers from the nucleus of the opposite side.

After their origin from nucleus dorsalis the fibers pass to a position at the periphery of the cord, just ventral to the dorsal horn, and there turn forward. They are so arranged that those from the more caudal parts of the cord lie farthest lateral while those from higher levels of the cord are medial {Sherrington and Laslett, ’03, quoted by Page May, ’06, p. 763, as following Ramon y Cajal and von Lenhossek) . As the lower end of the medulla oblongata is reached the tract turns dorsofrontalward, overriding the nucleus of the descending root of the trigeminal. It enters the cerebellum by way of the inferior cerebellar peduncle (corpus restiforme), distributing by way of mossy fibers to the homolateral and in part the contralateral granular layer (for the details of this distribution see Chapter VII, p. 802), In its course through the upper part of the cord and the medulla oblongata it sends collaterals and stem fibers to the gray of the various levels and particularly to the lateral reticular nucleus of the medulla oblongata {Thomas, ’97 ; MacNalty and Horsley, ’09 ; and others). Accompanying the main fibers are long internuncial spinal cord fibers {Sherrington and Laslett, ’03).

Many writers have contributed to our knowledge of the ventral spino-cerebellar tract (sometimes called the indirect spino-cerebellar tract or Gowers’ (’79) cerebellar tract, although this observer regarded it as probably carrying painful sensations from the opposite side of the body) and of its nuclei of origin ; among these may be mentioned Marburg (’03), Page May (’06), Dejerine (’95), Jacobsohn (’08), Pirie (’08), and Ingvar (’19). The path arises from cells situated in the body of the dorsal horn. The.se cells lie somewhat less medial than do those of Clarke’s column. The neura.xes, chiefly after decussation in the anterior or ventral white commis-sure — although some arise from the same side {Ljubuschin, ’02 ; Laslett and Warrington, ’99) — collect at the periphery of the cord just ventral to the dorsal spino-cerebellar tract and pass for^vard, in company with lateral spinothalamic fibers, through cord, medulla oblongata, and pons. In the medulla oblongata these fibers lie at the surface of the brain stem, dorsolateral to the inferior olive and closely associated with the rubro-.spinal, lateral tecto-spinal, and lateral spino-thalamic fasciculi. They occupy a somewhat deeper position in the pons, being situated ventrolateral to the facial nucleus, ventral and at certain levels slightly ventromedial to the nucleus of the descending root of the trigeminal nerve, and lateral to the superior olive. They are crossed by trapezoid fibers which .separate them into small fascicles. According to certain observers {van Gehuchlcn, '01 ; Dydijnski, '03 ; Collier and Buzzard, ’03, and others ; see also

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 267

Page May, ’06), a few bundles separate from the tract here and enter the cerebellum by way of the middle cerebellar peduncle, the brachium pontis. Near the beginning of the mesencephalon they swing dorsalward and enter the cerebellum at the side of the superior cerebellar peduncle (brachium conjunctivum).®'’ Their distribution in the cerebellum is discussed in Chapter VII. However, it is of interest to consider, with regard to the amount of crossing of the spino-cerebellar systems in the cord, that the proportion between crossed and uncrossed terminations in the cerebellum shows a reverse condition, that is, the number of crossed to uncrossed of the dorsal spino-cerebellar tract is as 2 : 1, while for the ventral it is as 4 : 1 {MacNalty and Horsley, ’09). (For these tracts see fig. 134.)

The relative development of these two systems in mammals is a matter of some importance for an understanding of their functions. In a study of the proprioceptive paths of mammals Weil (’27) found that planimetric indices showed a marked preponderance of dorsal and ventral spino-cerebellar tracts in ungulate brains. He suggested that this preponderance is a reflection of the “great exactness in muscle synergy” required in animals which have so great a speed of movement. With the exception of the ungulates, he found relatively great constancy in the spino-cerebellar system of mammals, although the spinal components of the afferent and efferent paths to cortical centers increase from lower to higher manamals.

In the lower lumbar cord, in the medial gray matter of the dorsal horn, there is a group of smaller cells, the neuraxes of which form the cornu-commissural bundle. This bundle, which Ues ventral to the dorsal funicuh against the commissura posterior or dorsalis, becomes particularly visible when the dorsal funiculi are degenerated (for example, in tabes dorsalis).

In the ventrolateral region of the cervical cord, near the periphery, is the triangular tract of Helweg, more commonly called the ohvo-spinal path. Most observers believe this to be a descending tract, accompanied on its medial side by ascending proprioceptive spino-olivary fibers. The ohvo-spinal tract is only a small bundle, irregularly triangular or crescent shaped {Obersteiner, ’00, ’01), and is probably concerned with movements of the neck. It is sometimes termed bulbo-spinal rather than olivo-spinal, being regarded as carrying fibers not only from the inferior olive but from the reticular gray of the medulla oblongata as well. By some authors (among others, Kaplan, ’16; Ariens Kappers) the main tract is regarded as ascending and is termed tractus spino-olivaris (see fig. 134).

e. Paths within the Cord for Visceral Afferent Impulses.

The course of visceral afferent impulses within the cord is still a matter of dispute and uncertainty. It is generally recognized that certain neurons of the spinal ganglia, usually distributing with the sympathetic system peripherally, receive sensations from the viscera and that such sensations are carried into the

“ Certain observers have regarded this tract as wholly contralateral in origin. This is probably not wholly correct ; though a portion of the fibers is crossed another portion appears not to be crossed. The relative number in the crossed component is difficult to determine by reason of the fact that these fibers run near the crossed lateral spino-thalamic fasciculus, vnth which they have been confused.

268 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

central nervous system over the neuraxes of such spinal ganglion cells and terminate in the gray of the cord, probably in the region near the base of the dorsal horn (see page 242). Successful elimination of pain by removal of the dorsal root ganglia (and rami communicantes) offers evidence of this course (Gasa, '24 ; Leriche, '25; Archibald, '28; and Davis, Hart, and Crain, '29, in dogs). The elimination of pain in certain cases of cancer (such as cancer of the prostate) through ablation of the lateral spino-thalamic tract suggests that certain secondary visceral afferent paths travel with this system. Spiegel and Bemis ('25) found that only by destruction of the ventrolateral white columns of the cord were they able to do away with reactions following stimulation of the splanchnic nerve. The region destroyed by them was large enough, however, to reach a considerable amount of underlying gray, as Davis, Hart, and Crain ('29) have pointed out. These latter workers, working with dogs, found that no lesions of the spinal cord short of a complete tran-section of that portion of the nervous system served to abolish the pain of forceful dilation of the cystic duct and biliary passages unless the operation involved a very considerable portion of the gray matter of the cord. They then reached the conclusion that short fibers, forming a chain system of neurons close to the spinal gray, form the main ascending path for visceral impulses. Huber and Crosby ('29, '30) were of the opinion that visceral afferent impulses reach the higher centers through such a chain series of neurons. It appears probable that such short fibers may run partly with the lateral spino-thalamic tract and partly deeper, in close relation to the spinal gray.

That still other regions of the cord may have visceral afferents for mediating reflexes is possible. Ranson and Hess ('15) stated that nociceptive reflexes are conducted to various levels by the Lissauer tract at the apex of the dorsal horn, although pathways to consciousness of painful impulses are provided in higher forms by spino-thalamic tracts. Ranson and Hess ('15), and Ranson and Billingsley ('16) believed that a path in the lateral funiculus of the cord (“not the same as the spino-thalamic tract although it occupies the same position in the cord, Ranson and Billingsley, p. 14) is concerned in the conduction of depressor reflexes of the vascular system mediated by the sciatic nerves. Ranson and Hess ('15) localized the path for pressor reflexes at the peak of the dorsal horn, regarding it either as chiefly homolateral or as equally bilateral. Thus two paths for vasomotor reflexes appear to be present in the cord. The whole matter of the spinal visceral reflex and of ascending visceral pathways needs further attention.

PATHS DESCENDING INTO THE COED

The descending systems show marked increase in mammals as compared with those in lower forms. Among the oldest of these are the vestibular tracts, which were already well developed in fishes. The vestibulo-spinal fibers run in two bundles, a medial and a lateral. The medial bundle consists of crossed fibers from certain of the vestibular nuclei (figs. 134, 506), entering the cord in the ventromedial region with other fibers of the medial longitudinal fasciculus. According to Gray ('26), contralateral fibers from the medial and inferior vestibular nuclei enter the medial longitudinal fasciculus. In this way they reach the

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 269

upper part of the cord. The lateral (or ventrolateral) vestibulo-spinal path contains some direct root fibers and fibers from the lateral vestibular nucleus {Gray). The bundle runs in the ventral funiculus throughout the greater extent of the cord, its fibers terminating at all levels around ventral horn neurons. This is one of the chief paths concerned in the maintenance of equilibrium on the reflex plane.

The reticulo-spinal paths are also very old phylogenetically. They have recently been described in detail for cat by Papez (’26), on whose account the following description of the bundles is based. Van Gehuchten (’91), Probst (’02), Lewandowsky (’04), de Lange (’07), and Ramdn y Cajal (’09) have all contributed to our knowledge of the system. Papez confirmed the results of Cajal, Probst, and van Gehuchten, all of whom found that reticular cells send processes into the median longitudinal fasciculus. Papez found the medial tract in the cat to be descending largely, and to lie in the lateral part of the median longitudinal fasciculus with such an arrangement of fibers that the more dorsal are of medullar origin and the more ventral of pontine origin. Papez suggested that the bundle is the same as the tractus pontis descendens of Lewandowsky (’04).

The lateral reticulo-spinal fasciculus arises in part at least from cells just medial to the motor nucleus of the trigeminus. It is a crossed path which runs caudalward to enter the cord between the crossed pyramidal and rubro-spinal paths. The ventral reticulo-spinal fasciculus, probably the homologue of the tecto-oUvary of Lewandowsky (’04), arises near the superior cerebellar peduncle, possibly in the ventrolateral tegmental nucleus, and descends into the cord, lying internal to the vestibulo-spinal tract (see fig. 506).

Recently Allen (’27) carried out a series of experiments on rabbits, which, he believed, indicates that fibers which conduct impulses to the respiratory motor nuclei of the cord from the cortex and superior colliculus and those bundles which mediate impulses arising in the superior colliculus and terminating around motor centers for hind limbs are located in the ventral columns or funiculi and the ventral part of the lateral funiculi. He believed that his experiments eliminate entirely tecto-spinal and practically cortico- and rubro-spinal tracts for the conduction of such impulses. He appears, however, to have taken into accoimt only the medial (crossed) tecto-spinal path present in the anterior funiculus and not to have considered the possible presence of a lateral tecto-spinal tract. Allen reached the conclusion, at all events, that the main conduction paths for cerebral and superior collicular connection with the motor centers for respiration in the cord and for the discharge of the superior colliculus to the motor neurons supplying the hind legs are to be found in the lateral and ventral reticulo-spinal tracts in the rabbit.

From the visceral sensory centers of the medulla oblongata, associated with the afferent fibers of the facial, glossopharyngeal, and particularly the vagus, with synaptic relations in the nucleus of the fasciculus solitarius, impulses reach the spinal cord for the completion of reflex arcs associated with the heart beat, coughing, vomiting, and respiration. By exactly what path or paths such central connections are estabUshed is still somewhat in dispute. A solitario-spinal path {Kosaka and Yagita, ’05 ', Ramdn y Cajal, ’09 ; Hirose, 16, and others) from the

270 NERVOUS SYSTEMS OF VERTEBRATES AND OF JVIAN

nucleus of the solitary fasciculus to preganglionic and other appropriate motor centers of the cord is postulated. For many of these reflexes there is at present no definite proof available, and where specific reflexes have been studied the results are conflicting and not entirely confirmatory. Thus various observers, including Gad and Marinesco (’92) and Porter (’95), believed that impulses passing from the medulla oblongata to the cord for the mediation of respiratory reflexes are situated in the lateral funiculus, while Rothmann (’02) found that approximately the dorsal half of the lateral funiculus had nothing to do with respiration while diaphragmatic respiration was stopped and thoracic respiration weakened by cutting the ventral part of the lateral funiculus and that costal respiration was affected most markedly by injury to the region immediately ventral to the ventral horn. Rothmann’ s observations were made on the cat ; more recent work by Allen (’27), carried out on the guinea pig as well as the cat, is confirmatory of the results of the former investigator. Allen concluded that in building the reflex center to the cord, sensory impulses reaching the nucleus of the fasciculus solitarius are relayed through the reticular formation of the medulla oblongata to reticular cells, such reticular cells giving rise to reticulo-spinal fibers (visceral bulbo-spinal fasciculus of Allen) which S 3 mapse about the appropriate efferent neurons in the spinal cord.

The rubro-spinal fasciculus (Monakow’s aberrant lateral tract), which has been found in submammalian types (birds), is relatively large in mammals with the corresponding increase in the red nucleus (fig. 134).* Directly after its origin the tract crosses to the opposite side of the cord in the ventral tegmental decussation and then, swinging directly lateralward, runs caudalward just dorsal to the lateral tecto-spinal and enters the cord just ventral to the lateral corticospinal path. The rubro-spinal fasciculus extends throughout the cord, diminishing progressively in cephalo-caudad direction, since at all levels its fibers terminate around ventral horn neurons. This path constitutes one of the main discharge paths for the cerebellar impulses from the dentate nucleus, which reach the red nucleus by way of the superior cerebellar peduncle. It is also a discharge path for the cortex and striatum since these are connected with the red nucleus by striorubral and cortico-rubral paths. A discussion of the probable functions of this nucleus and hence of its discharge path is to be found in Chapter VIII, p. 1085.

Two descending paths from the tectum are present in the cord, a medial and a lateral tecto-spinal fasciculus. The medial tecto-spinal fasciculus arises in the tectum, swings ventralward immediately to cross in the dorsal tegmental decussation, and runs caudad just ventral to the medial longitudinal fasciculus and sometimes is regarded as a part of that bundle. It is now and again termed the fasciculus predorsalis. It is believed to terminate in part around the cells of the intermediolateral column of the human cord. The lateral tecto-spinal fasciculus passes ventrolateralward as it leaves the tectum and takes a position just ventral to the rubro-spinal tract, which position it maintains throughout the remainder of its course. It terminates around ventral horn neurons. Both these tracts arc

“ The forerunner of the red nucleus in the form of large reticular elements is to be found in fi-'lies.

272 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

A considerable number of observers have studied the cortico-spinal paths in ungulates. Ziehen (’00) found crossed fibers passing to both dorsal and lateral funiculi ; /. L. King (’ll) described crossed and uncrossed fibers passing into the lateral column but could follow none beyond the first cervical ; Dexler (’07), studying sheep and goat, found an incomplete decussation with crossed fibers into the lateral and uncrossed fibers into the ventral funiculus.

The more dorsal tract runs exclusively in the dorsal funiculus in rodents such as the squirrel (Goldstein, ’04) , guinea pig (Bechterew, ’90 ; Wallenberg, ’98 ; Goldstein, ’04 ; Beveley, ’15, and others), the rat (Goldstein, ’04 ; King, ’10 ; Ranson, ’13, ’14 ; and others), and Arctomys (compare figs. 130 and 131). In the guinea pig

Nuc!. gracU.

Nucl.desc. rootV

I

Fig. 131. Cebus fatuellus. Passage of the pyramidal tracts into the lateral funiculi.

Beveley was able to trace the cortico-spinal tract into the lumbar cord (fig. 132). Simpson (’14) traced cortico-spinal fibers in the red squirrel and the chipmunk into the dorsal funiculus as far down as sacral cord. Considerable loss of fibers was seen in the cervical cord, however. According to Simpson, notable exceptions among rodents are found in rabbits and hares, where cortico-spinal fibers — partly crossed and partly homolateral — pass into the lateral funiculus. In the Canadian porcupine this same observer (Simpson, ’12, ’14a) was able to demonstrate that the cortico-spinal tract breaks up into four bundles on entering the cord. Two of these are crossed, the larger passing to the dorsal column and a smaller to the lateral column ; two bundles are uncrossed and enter the dorsal and ventral columns of the same side.

Higher mammals, such as the cat and various primates, have a crossed cortico-spinal fasciculus confined to the lateral funiculus of the spinal cord. It reaches its greatest development in primates, where it usually extends to the last sacral segment. With such a crossed or lateral cortico-spinal path there are said to be certain uncrossed bundles. The ventral cortico-spinal path.

274 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Cervical

Thoracic

These higher figures for primates (including man) indicate clearly the great significance of this tract in these forms.

The greater physiologic influence of the cortex on the movements in higher animals is both quantitative and qualitative. Experimental studies of the cortex, associated with the studies of lesions resulting, indicate that there are great differences in the amount of cortical control. Thus in many animals the movements of only the anterior extremities are directly under cortical control, as is indicated in the studies of cortical stimulations. In dogs n which the motor area of the left cerebral hemisphere was destroyed, Lassek, Dowd, and Weil (’30) showed that the pyramidal tract distributed 55 per cent of its fibers to the cervical segments of the cord, 20 per cent to the thoracic, and 25 per cent to lumbar and sacral regions. According to these observers, this distribution is comparable in general to that in man except for the fact that with relatively less thoracic musculature in man, relatively fewer fibers distribute to the thoracic region. The grade of influence is also different, for while in many of the lower animals, shortly following a removal of the cortical regions to even a large degree, complete motor control of the extremities and trunk reappears, such a destruction in higher mammals is followed by much more severe disturbances and in primates, and particularly in man, by disablement. Figure 134 illustrates the positions of these fiber tracts in the human spinal cord.

Lumbar

Sacral

Lower eacral

SIZE, GBOWTH, AND CHEMICAL DIFFERENTIATION OF THE SPINAL. CORD

Fig. 133. and ventral Erb.

Degeneration of the lateral pyramidal tracts in man.

Before closing this account of the vertebrate spinal cord, some mention is demanded of the very fundamental work relating to size, growth, and chemical differentiation of this and other parts of the nervous system during growth and other changing conditions. This work has been carried out very largely on white rats, material particularly suited to the problems considered because of the relatively short span of life and the ability of these animals to produce many offspring.

An adequate statement of this important type of work is impossible within the limits of the present text. The original papers should be consulted for details. Only certain general results can be stated here.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 275

Formulas for calrulafiiig tlic weight of the brain in the albino rat have been deviled [Ilalai, ’09). Standard tables allowing for the rat the relations existing between the weight and length of the body and the weight of the brain and cord ns compared with the length of the body have been computed {Donaldson, ’09) and a compaii^-on made between the growth of the human central nervous system and that of the rat {Dotmldson, 'OS and '09). Tables showing the percentage of water content in the brain and coid of the albino rat and a discussion of certain factor.^ involved in '•uch variations .as occur are presented {Donaldson, ’10, ’11b, ’15, ’IG) ; also a table is given in both the 1915 and 1916 papers of

Tnl fyUl ir gJ fwc.

Tif, ].■}) Can% ml mil'll ilnprim of n rro'is tmlion of the cord of mon, in llic whitp Fiibxtoncc of \\hich nrr (IiaprammilinlK outlined the nsccndinp nnd dcwndiiipfilxr potliiorfT'nouli, m re^icclne rehtivc position Asccndinp fasciculi remain white, deseendinK fa-'ciruli are stippled Tlic fasciculus proprius, which praetiealU ruiroiiiids the pn% matter, is coaered wath black lines Certain primary nnd 'ccondnrj neurons arc shown llulxr and Croshy

Donald'fon giving the percentage amount of water picsent in cord and brain for the first year of life (305 days). Comparisons between Norway and albino rats with respect to cord, brain, and body weight and to the percentage w-ater content of the central nervous system have been considered and discussed {Donaldson and Hatai, ’11a; see also Donaldson, '12). Moreover, Donaldson (’IG) by greater refinement in methods of calculation was able to show that a certain relation exists at a given age betw een the « eight of the ncrxmus system and the percentage of water present, since the “heavier brain or cord usually shows the smaller percentage of water.’’ It was found that exercise leaves unchanged the weight of the cord in relation to the body weight, although it increases slightly the w'eight of the brain (2.4-2.7 per cent), even though afforded only after the greater part of the grow'th has occurred {Donaldson, ’ll) Rats bearing young have distinctly heavier spinal cords which showi a greater water content {Watson, 05). Severe underfeeding alone decreases very slightly the w'ater content of the brain (Halai, '04; Donaldson, ’16). Severe underfeeding for tw'enty-one days, followed by a normal diet for two hundred days, produces no change in the relative weight of

276 NERVOUS SYSTEMS OF ^TSRTEBRATES AND OF ]\E\.N

brain or cord, although there is an increase in the water content {Hatai, ’07). The same results were obtained -nith reference to cord weight when severe underfeeding during the first three to six months of life produced stunting of the animal. Underfeeding {Halai, ’04, and Donaldson, ’ll) stops body growth and retards the growth of the central nenmus system but does not greatly modify the percentage of water in the brain. Diseases such as pneumonia {King, ’ll) arrest brain development, decrease the water content, and, of course, decrease the body weight. Castration, which does not affect the bod 3 " growth in the white rat {Slotscnlnirg, ’09), also is not associated vdth anj' change in the percentage of water in the central nervous sj'stem, according to Donaldson and Halai fll). These observers did note that the brain and cord weight was slightlj’’ less in the castrated animals. Later experiments by Halai (’13) showed a lesser, almost negligible decrease in the brain weight of castrated and also of spayed and semispayed animals, but a sliglit increase in the weight of the cord. These differences are so slight that removal of the sex glands {Halai, ’13, and others) is regarded as an e.xpcrimental condition (as is inbreeding) w'hich is indifferent in its effect on brain growth. Among unfavorable conditions, that of domestication holds first place, according to these observers, while exercise taken voluntarily and highlj varied diets afford particularly favorable conditions. Donaldson (’15) believed that the variation in water content in the central nervous sj'stem is dependent especially on the metabolic activity of the bodj’’ as a whole, increasing with the increase of anabolic processes and decreasing with increased catabolism.

Review of the Ouganization and PnocnEssms Developaient of the

Spinae Cord

AMPHIOXUS

In .-Ymphioxus, the dorsal and ventral roots emerge alternatel} from the spinal cord and are entirely separate in their peripheral distribution, for the ventral have an intramyotomic and the dors.al an intermj’otomic course. The cells of origin of the somatic and visceral efferent fibers have not been definitely localized and probaljl}' do not lie at the levels of the emergent roots. It is possible that they arise ns collaterals of longitudinal fibens.

Three types of fillers arc present in the dorsal roots: somatic afferents witli free scnsor\' endings for the skin and for mu.sclc sense, Ansceral afferents with free terminations for mucous membranes, and visceral efferents, motor to unstriated mti.'-clc such as that of the atrium. The cells of origin of the afferent fillers arc lii}iolar and lie partlj' within the cord near the region of the dorsal roots and ]iarlly along the pcriphcr.'d course of the fibers. Thc.se colls arc never grouped into spinal ganglia. The coarse neuraxes of the somatic afferent neuron.s divide dichotomously and .ascend and dc.sccnd, occupj'ing a position lateral to the finer, un<livided visceral afferent fibers.

.Most conspicuous among the secondary cells of the cord are the dorsoincdi.'d giant ganglion cells, 'I'lie cel! bodie.s of the.se are situated near the entrance of the afferent root.s and, like the roots, occur alternately on the right and left sidc.s.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 277

Their neuraxes cross the midline, forming thus a ventral arcuate system. After decussating, the fibers from the. posterior part of the body run cephalad, alternately on the right and left sides, while those from the cephalic regions nm caudad in a smular fashion. These neuraxes form the earlier secondary sensory or afferent tract of the cord for protopathic or vital sensibility. In Amphioxus they are largely for mediating protective reflexes. Smaller funicular cells, with neuraxes having a homolateral distribution, are also present.

The ciliated neuro-sensory cells of the tentacles and the ocelli (spinal eyes), particularly numerous in the upper and lower segments of the cord, are specialized structures occurring only in Amphioxus.

CYCLOSTOMES

In cyclostomes the dorsal and ventral roots alternate as in Amphioxus. In Petromyzon these roots do not join to form a mixed nerve, but in myxinoids they unite into a common trunk. The central gray substance has differentiated sufiiciently so that a horn of gray (with sensory or afferent and motor or efferent elements) extends laterally on either side of the central canal.

The cells of origin for the somatic efferent fibers lie in the horn of either side but rarely at the level where the root fibers emerge. Their dendrites spread out chiefly in a plane at right angles to the longitudinal tracts of the cord, due, to some degree, to the irradiation of impulses from these longitudinal fibers which are as yet totally immedullated. Some of the dendrites terminate about Muller’s fibers, but the majority of them form a marginal dendritic net, to the development of which trophic conditions contribute (particularly in Petromyzon, where intramedullary blood vessels are lacking). Part of the visceral efferent fibers run in the ventral roots.

The dorsal roots carry somatic and visceral afferent and visceral efferent components. The somatic afferents, as in Amphioxus, have free sensory endings. About one-fifth of the fibers arise from intramedullary cells, the remainder from extramedullary and mostly bipolar ganglion cells. The central dendrites of these bipolar ganglion cells dichotomize in the dorsolateral region of the cord and give off relatively brief ascending and descending branches. Collaterals enter the gray of the horn and also the marginal dendritic plexus. The visceral afferent fibers terminate in the gray medial to the termination of the somatic afferents. The location of the cell bodies which give rise to visceral efferent fibers is unknown.

The secondary systems of the cord are much better developed in cyclostomes than in Amphioxus. First among these systems are the ventral arcuate fibers, which arise in the gray of one side, decussate, and ascend in the ventrolateral region on the side opposite that of their origin. These crossed fibers form a secondary ascending tract. They are much more numerous than the neuraxes of the giant ganglion cells of Amphioxus but nevertheless are to be regarded as an homologous system. In the main this secondarj’’ ascending system is intrinsic to the cord in cyclostomes, but a few of the bundles are believed to reach the medulla oblongata. Cells of the spinal gray send homolateral fibers into the dorsal and

278 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

lateral regions of the white matter of the cord. These fibers usually run for only short distances.

The cord of cyclostomes is almost entirely for local reflexes. Muller’s fibers and fibers from the reticular cells of midbrain and medulla oblongata provide the only pathways by which impulses from higher centers reach the levels of the cord. Muller’s fibers originate from cells of the midbrain and medulla oblongata and some of them extend through the cord to its most caudal segments. Dendrites of the motor cells of the spinal cord lie in intimate relation with the fibers. Smaller reticulo-spinal paths reach various levels of the cord. A group of polynuclear glandular cells occurs at the caudal end of the cord.

PLAGIOSTOMES

The myelinization of the fiber bundles, the union of the dorsal and ventral roots to form a common nerve trunk, and the arrangement of the gray are the most striking ways in which the plagiostomes furnish a prototype for the spinal cord of higher vertebrates. Sympathetic chain ganglia are not present, but, at the region of junction of the dorsal and ventral roots, a small visceral efferent branch is given off which runs to the intestine and to the walls of the blood vessels.

In the more frontal regions of the cord there is a striking difference in the relative size of the dorsal and ventral roots, the latter being very small and, in the most frontal regions, entirely lacking. These roots form the so-called occipitospinal nerves. Cephalad to them are the occipital nerves, which in plagiostomes are included within the skull (the neocranium) through the protometameric assimilation of the spinal roots. In cyclostomes, where there is only a palaeocranium, the occipital nerves are represented by the first spinal roots and lie entirely outside the skull.

No lumbar or cervical enlargements occur in plagiostomes. Consequently the various levels show marked resemblance to each other. The spinal cord gray in plagiostomes shows, for the first time phylogenetically, a differentiation into dorsal and ventral horns, due to neurobiotactic migration of the gray matter. The shape of the gray is more or less that of an inverted Y (x) because the fibers of the dorsal funiculi do not separate sharply from the two dorsal horns, but are grouped into separate bundles which are scattered among the gray masses of the horns.

The cell bodies of the somatic motor or efferent neurons lie on the same side of the cord as do the roots to which they contribute fibers, but generally in front of the emergence of these roots. The neuraxes have, therefore, a caudal and somewhat oblique course before leaving the cord. The dendrites of the motor cells, as in cyclostomes, form a marginal plexus and this plexus is particularly rich near the lateral surface in plagiostomes. In addition to somatic efferent fibers, visceral efferent fibers are probably present in the ventral roots. The dorsal horns, joining over some distance, consist for the greater part of gray matter, comparable to the substantia gelatinosa of higher forms, and active in receiving protopathic or vital impulses.

The dorsal roots carry somatic and visceral afferent and visceral efferent components. In the embryo the afferent fibers arise from the transient intra

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 279

medullary cells, but in fully developed animals they have origin only from extramedullary or spinal ganglion cells. The neuraxes of certain of the sensory ganglion cells, on entrance to the cord, end directly in the gray of the dorsal horn ; other fibers divide dichotomously and form ascending and descending bundles in the dorsal part of the cord, more or less intermingled with the gray of the dorsal horn. These ascending bundles of the dorsal funiculi consist mainly of fibers which run for relatively short distances before terminating in the gray. In these animals there is no accumulation of dorsal root fibers for medulla oblongata centers such as is characteristic of the region in higher vertebrates. In order to complete the reflex arcs, the motor neurons send large dendrites into the dorsal horn, where a synapse occurs either directly with incoming sensory fibers or with short collaterals from them. This is spoken of as the primitive sensory-motor reflex path and represents a primitive condition, the reverse of relations found in higher vertebrates, where the sensory dendrites terminate near or, in some cases probably, around the cell bodies of motor neurons.

Spino-bulbar and spino-mesencephalic tracts are present in plagiostomes. Like the secondary sensory paths of cyclostomes, to which they are phylogenetically related, they arise from dorsal horn cells the neuraxes of which (ventral arcuate fibers) decussate in the ventral or anterior commissure and enter the ventrolateral region of the cord. In this latter region the fibers divide, forming ascending and descending branches. AH of the descending and most of the ascending fibers from the more posterior parts of the body are intrinsic to the cord, but the ascending ones from the more frontal regions collect into a bundle which distributes to bulbar and midbrain regions as well as to spinal centers. Homolateral funicular fibers, which have increased greatly in plagiostomes as compared with cyclostomes, form bundles which enter the ventral, lateral, and dorsal funiculi. They are largely descending, a fact associated with the relatively great importance of aboral reflexes in these lower vertebrates. Among the homolateral paths is an ascending one to the cerebellvun — tractus spino-cerebellaris.

A comparison with the spinal cord of cyclostomes indicates that there has been an increase in the systems which enter the plagiostome cord from higher centers. Such descending tracts arise peripherally from vestibular and lateral fine centers and from the reticular nuclei of the medulla oblongata. Direct fiber tracts have not been demonstrated with certainty from the optic tectum, the inferior lobes, and the base of the midbrain, but these areas are known to be connected by fiber bundles with the reticular nuclei. Reticulo-spinal tracts from these latter nuclei furnish then a final common path (Sherrington, ’09) for the impulses from higher centers (particularly the midbrain). Indirectly, by way of vestibular and lateral line centers (probably also through reticular cells), the cerebellum finds discharge paths to the cord.

TELEOSTS

A reduction of the frontal end of the spinal cord occurs in all teleosts. The occipital and occipito-spinal nerves of plagiostomes often are lacking in teleosts (consult p. 167), and spinal nerves with fully developed dorsal roots follow im

280 NERVOUS SYSTEMS OF ’^^RTEBRATES AND OF MAN

mediately the branchial nerves of the medulla oblongata, occurring even partly within the skull. In certain teleosts there is further a marked reduction at the caudal extremity of the cord, where a considerable portion of the vertebral canal contains onl}”- longitudinal root fibers. This caudal reduction, which is caused by an atrophj”^ of the posterior body regions of the animals, does not occur in all teleosts. However, it is very distinct in Plectognathes, especially in Orthagoriscus and Lophius.

In all teleosts, peripheral to the junction of the dorsal and ventral roots and outside of the vertebral canal, the rami communicantes of the sympathetic system form a true ganglionated sympathetic chain. It is the first appearance of this chain in phylogenj'. The cell bodies of the somatic motor neurons lie farther ventralward in teleosts than in plagiostomes, due to the greater neurobiotactic influence of the ventral bundles in the former group. These motor cells oftch are arranged in dorsomedial and in ventrolateral groups. Their dendrites form, as in all fishes, marginal plexuses which are particularly well developed on the lateral side of the cord. Preganglionics or general ■idsceral efferent fibers are likewise present in the ventral roots.

In certain teleosts (Malopterurus and Gymnotus, probably also Mormj'rus) some of the motor cells have become cells of origin for the so-called electric nerves. In Gymnotus these are somatic motor cells, the electric organs arising from muscles derived from sonutes. In Malopterurus the neurons are of the visceral efferent type, since the electric organs arc derived from the smooth muscles of the skin.

The dorsal horns are in close contiguit}'^ since the dorsal funiculi show relatively little development. The horns present a structure which is comparable to the substantia gelalinosa of higher forms.

The dorsal roots in bony fishes carry somatic and visceral afferent fibers and probably a few visceral efferents. In contrast to those of adult plagiostomes, the somatic sensory fibers of many teleosts arise partly in intramedullar}’’ ganglion cells. These intramedullary cells may be present either throughout the entire spinal cord or only in the ccrv-ical region. The extramedullary cells of origin of the afferent fibers fall, in many teleosts, into two main subdivisions, tboso situated above the cord (and medulla oblongata) and known as supramcdullary ganglion cells, and the usual spinal ganglion cells of vertebrates which arc found on either side of the cord at every spinal segment.

The central off.shoots of the spinal ganglion colls generally enter lateral to the dorsal horns and there divide into ascending and descending branches. Apparently in some bony fishes (cod) these fibers run in the dorsal or posterior funiculi, but in others (sea robin) many of them, particularly from the upper part of the cord, enter its dorsolateral region and ascend there for some distance Ijefore terminating in the gray of the donsai horns. As a rc.sult of this course of the dorsal root fibers, the dorsal funiculi are very small in teleosts and consist particularly of descending secondary neurons transmitting aboral reflexes.

.Secondary neurons from the dorsal horn region send their neuraxes across the midline to the ventral funicular region of the opposite side. These soon turn

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 281

slightly lateralward and become part of the ventrolateral fasciculus of the cord. Some of the fibers of this fasciculus, chiefly those from the cervical cord, pass from the cord to higher centers, thus forming spino-bulbar and spino-mesencephalic tracts. Homolateral fibers from dorsal horn cells are present in teleosts in essentially the same relations as those described for similar bundles in plagiostomes. Particular attention is called to the presence of spino-cerebellar tracts.

In certain teleosts (for example, Trigla and Lophius) the dorsal horns of the cervical cord are greatl 3 ’^ enlarged in consequence of a h^^pertrophy of the tactile fibers in their corresponding peripheral nerves. This increased mass of dorsal graj' substance in the cervical region is not comparable to the nuclei of the dorsal funiculi in mammals. It has only local importance.

The descending paths from the frontal regions of the brain to the cord are similar in teleosts and in plagiostomes. Nervous impulses brought in by optic, trigeminal, and olfactory nerves are ultimately discharged, after one or more synapses, into the reticular centers of the brain stem and then pass to the motor centers of the cord by way of reticulo-spinal paths. Vestibulo-spinal paths are also present. The most conspicuous cells of the medulla oblongata are the giant cells of Mauthner. These cells send neuraxes caudalward throughout the extent of the cord. They are concerned particularly with mediating tail reflexes in response to vestibular (saccular) stimuli and afford excellent illustrations of the final common path of Sherrington. Indirectly they afford a discharge path for optic and trigeminal centers. Secondary descending visceral pathways from taste centers of the medulla oblongata are present in teleosts, extending as far as the cervical cord. Such paths have not, as yet, been demonstrated in plagiostomes.

AMPHIBIA

From the standpoint of gross structure the spinal cord of amphibians is greatly reduced when compared with that of plagiostomes. Histologically there is a greater resemblance between the two divisions. The numerous occipital and occipito-spinal roots of the plagiostomes are represented by only one nerve on a side in amphibians. This nerve joins the second spinal nerve, the first spinal being absent, and forms the so-called hypoglossal of some authors.

In such plagiostomes as the adult shark there may be as many as a hundred pairs of spinal roots, but in adult tailless amphibians, such as the frog or toad, there are onlj'^ ten or eleven pairs, and occasionally less. In the larval amphibians and in the tailed forms there are many more roots than in the adult frog, the roots disappearing secondarily with the loss of the tail and the caudal end of the spinal cord changing into an atrophic filum terminale.

Lumbar and cervical enlargements first make their appearance in amphibians, as a consequence of the development of the extremities. The somatic efferent fibers, which in the embryo arise as collaterals of longitudinally coursing fibers, in the adult are formed in the usual way by the main neuraxes of the motor cells. The motor neurons fall into ventromedial and ventrolateral groups, the former extending throughout the cord, the latter more or less confined to the region of the two enlargements where the motor fibers for the upper and lower extremities

282 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

are given off. The marginal plexus is formed largely by the dendrites of these motor cells but receives fibers also from other cells of the ventral horn. It extends over nearly the whole periphery of the cord but is best developed, as in the shark, in the ventrolateral and lateral regions.

Intramedullary neurons, as cells of origin for sensory fibers, occur in amphibians as in plagiostomes (shark) only during development. The dendrite of such an intramedullary neuron is formed peripherally by the union of two branches, one from the skin (exteroceptive) and the other from muscle (proprioceptive). In the adult such intramedullary cells have disappeared and only extramedullary cells, arranged as dorsal root ganglia, are present.

The neuraxes of the dorsal root neurons, on entering the cord, form a smaller lateral and a larger medial bundle. The lateral one, composed of finer fibers, lies at the margin of the dorsal horn, the individual neuraxes terminating soon after entrance in the gray of the horn. The gray matter of the amphibian cord has a definite butterfly-shape, due to the greater width of the dorsal funiculi and the separation of the dorsal gray into two horns. In the frog this separation of these dorsal horns is due to an increase of the fibers between the horns, amongst which are descending trigeminus root fibers which extend caudalward as far as the lumbar region. On entering the cord the fibers of the coarser medial bundle enter the dorsal funiculus. From this position, in tadpoles, collaterals are given off through a bundle which turns into the dorsal gray and there synapse with long dendrites of the somatic motor neurons. In adult frogs the relations are slightly different. For the first time in phylogeny, in the adult frog, the collaterals of this medial bundle are long and extend almost into the ventral horn. After giving off these collaterals, the fibers dichotomize and ascend and descend in the dorsal funiculi. Although the number of ascending fibers may be somewhat greater than the number of descending fibers in amphibian forms, still there is no evidence of a marked accumulation of dorsal root fibers in the upper levels of the cord. It is true that in the cervical region of the frog the dorsal funiculi have a greater size than at more caudal levels, but this fact is due rather to the presence of descending trigeminal, vagal, and vestibular fibers in that region than to an increased number of dorsal root fibers. True dorsal root nuclei are not present in amphibians and a medial lemniscus is likewise lacking.

Secondary sensory fibers, arising from dorsal horns of the cord (and particularly of the cervical cord), after decussating in the ventral commissure, ascend as spino-bulbar and spino-mesencephalic fibers. They carrj’- pain, temperature, and primitive tactile sensibility (according to some observers, primitive muscle sense as well) to the medulla oblongata and to midbrain centers. These tracts are the homologues of the primitive secondary sensory path of Edingcr and the forerunners of the spino-thalamic sj’stem of higher forms. An uncrossed spinocerebellar tract, homologous with that of plagiostomes and of bony fishes, is present in amphibians, but is smaller in frogs, for example, than in higher fishes.

The descending paths of the cord are augmented by root fibers of the trigeminal, vestibular, and vagus nerves. In frogs, where this has been particularly studied, the functional acti\nties of the spinal cord are influenced strongly by the

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 283

impulses carried over these cranial nerves, and especially by those passing over the trigeminal, since descending fibers of this root extend even into the lumbar cord.

REPTILES

The relations existing in the reptilian spinal cord may be considered as a direct introduction to the organization of this structure in mammals. In all orders of reptiles, however, the cord extends throughout the whole length of the vertebral canal and does not end in an atrophic Slum terminale (as is the case in some fishes, tailless amphibians, and mammals). Its great extent in reptilian for ms is due to the persistence here of a metameric tail musculature. In other respects the various orders of this class show great differences. Macroscopically, three t 3 TDes of cord can be distinguished : (1) that of sphenodons, lizards, and crocodiles, which have both trunk and extremity musculature^^; (2) that of snakes, where extremities and consequently extremity muscles are lacking; and (3) that of turtles, which have no thoracic muscles but possess musculature for neck, tail, and extremities. In the first group there is a well-developed cord with cervical and lumbar enlargements ; in the second group no cervical and lumbar enlargements are present ; in the third group the cord is strikingly thin in the thoracic region.

In the ventral horn, cells of von Lenhossek are present in addition to the usual somatic and general visceral efferent cells. These cells of von Lenhoss4k have efferent fibers which emerge through the dorsal roots and which are regarded by certain observers as the precursors of the spinal accessory nerve of mammals.

The marginal dendritic net is much smaller in full-grown reptiles than in amphibians, but is considerably developed in the embryo. The marginal nuclei, which are particularly well developed in crocodiles, occur along the lateral border of the cord where the marginal plexus of dendrites is most pronounced in lower animals. The function of these cells is as yet unknown. It is possible that they are commissural (arcuate) cells which have migrated secondarily from their original position in the gray.

Intramedullary dorsal root cells are present in embryonic forms, but they disappear relatively quickly and only the extramedullary cells remain. The neuraxes of these extramedullary cells show somewhat different central relations in the different classes of reptiles. In lizards they have a course similar to that in ampiiibians. The lateral finer bundle runs along the margin of the horn and terminates in the dorsal gray soon after entrance to the cord. The fibers of the coarser medial bundle enter the dorsal funiculus, give off collaterals to the spinal gray of the level, and then break up into ascending and descending branches. In tortoises and snakes a condition reminiscent of that in teleosts is found, for the major bundles of the dorsal root fibers run in the dorsal part of the lateral funiculus, sending collaterals to the cells of the dorsal horn, as does the medial bundle of lizards. The number of ascending dorsal root fibers has increased

“ In certain of the Lacertilia (Anguis) the extremities are wanting ; in Sheltopusik only posterior extremities are present.

284 NERVOUS SYSTEMS OF VERTEBRATES AND OF IMAN

considerably in aU reptiles as compared wdth amphibians, and true nuclei of the dorsal funiculi (nucleus gracilis and nucleus cuneatus) are present in reptilian forms, the first clear representatives, phylogenetically, of these nuclei. It is probable that fibers representative of the medial lemniscus are present, but they have never been clearly demonstrated for the reptiles studied. Such a lemniscus system would make possible the direct projection of higher sensory qualities (stereognostic sensibility) upon the thalamus. Associated with tliis advance centrally is the development peripherally of complicated sensory terminations (Pacinian corpuscles as well as other endings). The higher development manifested thus in reptiles is regarded as due to the greater importance of the skin sensibility and the finer adjustments of movements necessary to meet the demands of terrestrial life.

There is no great difference in the crossed spino-mesencephalic tract and in the spino-cerebellar paths of frogs and reptiles, although the systems (particularly the spino-cerebeUar) are considerably larger in the latter animals. Descending pathways from the medulla oblongata reach the cord directly, but these are particularly from the reticular gray. Descending root fibers from vestibular and trigeminal nerves do not extend far down in the dorsal funiculi, as was the case with the frog. Tecto-bulbar fibers transmit impulses to reticular cells of the medulla oblongata, which in turn relay them to the cord. It is possible that tecto-spinal fibers may enter the upper part of the cord, but they have not been demonstrated. A medial longitudinal fasciculus, continuous with the ventral ground bundle of the cord, is present. There is no direct forebrain connection with the cord in reptiles.

BIRDS

The avian cord differs macroscopically from that of the reptile in having, in general, a longer cervical region and in possessing a lumbo-sacral sinus which lies between the dorsal funiculi of the lumbo-sacral cord. This sinus is filled with a clear, semi-transparent tissue and is formed, apparently, by the lateral migrations of the dorsal horns, a consequence of the great number of dorsal root fibers at these levels.

In general, the ventral roots in birds are slightly larger than the dorsal, and the ventral horns are broader than the dorsal, particularly in Cursores such as the ostrich. Cervical and lumbar enlargements are always present. In the ostrich the lumbar enlargement is the greater, but in birds which are good fliers the cervical enlargement surpasses the lumbar in size. In the ventral horn there is a v^ery distinct medial or v-entromedial group of cells for the innervation of body musculature and a lateral or v'entrolateral group for the innervmtion of the e.xtremitj' muscles. The v'cntrolatcral group is particularly large in the intumesccntia lumbalis of the ostrich, where the cell bodies of the motor neurons for the legs are found. In most birds an especially well-developed group of cells is present in the cervical region. This nuclear mass is often termed the fljing center since it gives rise to the fibers supplying the main muscle of the wings, the pectoralis major.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 285

A marginal dendritic plexus is present only in the embryo. As in reptiles, nuclei marginales appear near the ventrolateral surface of the avian cord. They are best developed in lumbar and cervical regions. Ontogenetically their cells are derived from the gray of the ventral horn. The suggestion is made {Ariens Kappers) that they migrate toward the surface under the influence of neurobiotactic stimuli from the ventrolateral regions of the cord. Probably these are cells which contributed to the formation of the marginal plexus in embryos. They are embedded in a gUous tissue, richly provided with glycogen. Their function is not known, but there is evidence that their neuraxes contribute to the formation of arcuate fibers which cross in the anterior commissure and then turn forward.

The spino-mesencephahc tract, carrying pain, temperature, and general tactile impulses, which lies near the surface in the ventrolateral portion of the cord, also mediates protopathic or vital sensibility. Consequently the necessary functional and structural conditions for the neurobiotactic displacement of the marginal nuclei toward the periphery are furnished by the spino-mesencephalic system.

The central processes of the spinal ganglion cells run in the dorsal funiculi and in the marginal zone. A comparison of the dorsal funiculi with either the remaining white substance or the gray substance of the cord shows they are relatively much smaller in birds than in reptiles. The smaller size of the avian funiculi is due in part to the less development of skin sensibility in animals having feathers and in part to the short course of the ascending fibers. As a result of this latter condition, there is no great accumulation cephalad of root fibers in the dorsal funicular region and the associated nuclei (nucleus gracilis and nucleus cuneatus) are small. There is probably the beginning of a medial lemniscus system in birds, but it is very small and as yet has not been demonstrated clearly.

The endogenous fibers of the cord appear to be more numerous than in reptiles, particularly the shorter intersegmental ones. Likewise, spino-bulbar and spinomesencephahc tracts are well developed in the upper region of the cord. The homolateral, ascending spino-cerebellar system is very large in birds, for it arises from the whole extent of the cord and occupies its entire lateral periphery, at least in the upper levels of the cord. This tract is accompanied on its medial side by a descending cerebello-spinal tract. Descending vestibulo-spinal paths, both crossed and uncrossed, reach the cord. Tecto-spinal paths are present in both the ventral and ventrolateral funiculi. A rubro-spinal path may be present but it has never been demonstrated clearly. From the above r6sum6 it is evident that the avian cord is influenced much more by descending paths from vestibular nuclei, from cerebellum, and from optic tectum than is the case with reptiles, amphibians, and fishes. The increase in the descending paths from these bmn centers is undoubtedly associated with the development of highly speciahzed motility in birds. No direct paths from forebrain to the cord have been demonstrated satisfactorily.

MAMMALS

The cord in mammals differs from that in birds and reptiles in never occi^iying the whole length of the vertebral canal. There is nearly always a great differ

286 NERVOUS SYSTEMS OF VERTEBRATES AND OP MAN

ence in the length of the spinal cord and of the spinal column. This lack of correspondence of the segments of the cord and the column is not confined to the sacral region — although it is most striking there — but is found at other levels as well. This discrepancy in length is due largely to the continued growth of the spinal column after the spinal cord has attained its full length. As a consequence, the spinal-cord segments always lie higher than the corresponding vertebrae. In many mammals the tail has partly or wholly disappeared. Even in those mammals where a tail is present, it has lost its metaineric character and consists only of the muscles of certain segments while muscles of other primitive segments have disappeared. This loss of certain muscles at the posterior end of the body has resulted in the atrophy of caudal portions of the cord so that the discrepancy between segments of the cord and of the column is greatest at the caudal regions of the body. In man the conus terminalis usually terminates about the level of the second lumbar vertebra. The extent of the cord varies in different mammals and there may be considerable difference in length in animals belonging to the same order (for example, in the morvotremes Echidna and Ornithorhynchus).

Occasionally there is a great difference between the diameter of the cord and of that portion of the vertebral canal in which it lies. Thus in the cervical region in dugong and whale, the lumen of the vertebral canal is about twelve times the transverse diameter of the cord.

Cervical enlargements are present in all mammals. Lmnbar enlargements are usually found but are lacking in certain mammals, such as dugong and whale, which have no posterior extremities. In Phocaena and Delphinus the lumbar intumescentia is somewhat farther caudal than in other mammals, due to the great development of the tail musculature. The lumbo-sacral enlargement is larger than the cervical in the kangaroo, where the musculature of the posterior extremities and of the tail is much more highly developed than that of the anterior extremities. However, in general the cervical is larger, due to the greater size of the long ascending and long descending paths toward the upper levels of the cord. Any comparison of the two enlargements in most mammals must take into consideration the fact that the cervical enlargement, while it shows some increase in gray, is dependent particularly for its size upon the white matter of the region, while the lumbar enlargement is due to a great augmentation of the gray of the lumbar region. However, in some animals such as the bat, the gray of the cervical region is greater than that of the lumbar.

The ventral roots, in most mammals, have less fibers than the dorsal roots but are a little thicker because of the greater diameter of many of the ventral root fibers. However, in Cetacea, due to the poor development of skin sensibility, including pain, the dorsal roots carry the smaller number of fibers.

Both coarse and fine fibers are present in the ventral roots. The fine fibers, in general, are preganglionic fibers with cells of origin in the intermediolateral column of the cord and fibers which pass out to chain or collateral ganglia in order to form pericellular, subcapsular synapses around postganglionic neurons. The coarse fibers supply the skeletal musculature. Their cells, within the ventral horn of the spinal cord, are arranged in two major columns, medial

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 287

and lateral. The medial column extends practically throughout the cord, the neuraxes of its cells supplying motor fibers to trunk musculature. The lateral splits up into several nuclear masses which are concerned with extremity muscles, the major divisions at present recognized being the ventrolateral, the dorsolateral, and the retrodorsolateral (see previous account). Within this lateral column apparently there is a functional topography such that the muscles which lie near the trunk (those of the hip and shoulder) are innervated by the ventrolateral group, the muscles of the extremities proper by the dorsolateral nuclear masses, and the muscles farthest away from the body (those of the fingers and toes) by the retrodorsolateral group. It is to be expected that this last nucleus will be most clearly developed in man, where the fingers have the greatest independence and the most exactness of movement. The explanation for these relations is yet to be given. It is suggested that simple mechanical relations during the development of the nerves and respective muscle groups may offer a solution ; it is possible that neurobiotactic factors may play a part.

The gray substance of the dorsal horns shows a distinct division into three parts: the zona marginalis, the substantia gelatinosa of Rolando, and the main mass of the dorsal horn. The zona marginahs does not differ greatly from that in lower forms. The substantia gelatinosa is well developed in most mammals, especially so in ungulates but much less so in cetaceans. The substantia gelatinosa spreads like a cap over the top of the dorsal horn and may join its fellow of the other side (carnivores and monkeys) or may form marked convolutions (ungulates). Its unmyelinated, gelatinous appearance arises from the abundance of relatively small, spindle-shaped cells, the dendrites of which spread out into the gray substance of the substantia gelatinosa. The neuraxes of these cells, at least in many cases, are not long but enter the neighboring funiculi. Collaterals of these neuraxes pass up and down to reach the gray of higher and lower levels. Opinions differ as to whether any of the neuraxes pass directly into the lateral spino-thalamic tract.

Several nuclear groups are distinguishable in the main mass of the dorsal horn. Distinctly differentiated is the dorsal nucleus or nucleus of Clarke, a collection of relatively large cells in the medial region of the horn. It extends from the seventh cervical to the first or sometimes the second lumbar, and gives origin to the dorsal spino-cerebellar fasciculus. This nucleus and its cephalic continuation into the cervical region were first described by Stilling, and the cervical portion still carries his name. In general, the central and lateral regions of the dorsal horn are less differentiated. The cells of origin for the ventral spino-cerebellar fasciculus lie in this region but are not easily delimited as a special nuclear mass. Here, likewise, are the cells of origin of the spino-thalamic paths and probably of other long ascending systems such as the spino-oUvary, spino-vestibular, and spino-tectal paths. Neurons are present also, the neuraxes of which form the phylogenetically very old, descending, endogenous bundles of the posterior funiculi, the cornu-commissural fibers (ventral field or dorsal ground bundle).

288 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

On entering the cord the dorsal root fibers split into a lateral and a medial bundle. The lateral bundle, supposedly carrying temperature and pain impulses, consists of fine medullated and unmedullated fibers which enter Lissauer’s fasciculus at the tip of the dorsal horn. Here the fibers divide dichotomously, but the resulting branches are short and enter the substantia gelatinosa Rolandi about a segment or so above and below the level of entrance. Here, either directly or through interpolated neurons, they synapse with the cells of origin of the lateral spino-thalamic tract. This tract consists of crossed fibers from all levels of the cord which, after decussation in the anterior or ventral white commissure, run forward to the brain in company with the ventral spino-cerebellar tract. When this latter tract turns toward the cerebellum, at the upper border of the pons, the lateral spino-thalamic tract continues forward with the medial lemniscus to the thalamus. In addition to these long fibers, the tract is believed to carry also shorter fibers which turn into the gray at various levels of the brain and cord. Here they synapse with cells which in turn send their neuraxes into the lateral spino-thalamic bundle. Thus this fiber system consists in part of long neuraxes and in part of a series or chain of neurons with cell bodies situated at various levels. There is evidence that it carries some homolateral fibers.

The coarse fibers of the dorsal roots (the medial bundle) carry tactile sensibility, deep pressure, and muscle sense. Collaterals, and probably stem fibers as well, terminate around the dorsal horn neurons, which in turn send their neuraxes across the midline into the ventral portion of the white matter of the cord and then forward to higher centers as the ventral spino-thalamic tract. The stem fibers carrying tactile sensibility ascend in fasciculus gracilis and fasciculus cuneatus, sending collaterals into the gray at all levels. Some of these stem fibers reach the nuclei of the posterior funiculi — nucleus gracilis and nucleus cuneatus — within the medulla oblongata. It is believed that the ventral spino-thalamic tract is concerned with general tactile sensibility and deep pressure, while impulses involving tactile discrimination pass over the fasciculus gracilis and fasciculus cuneatus.

Impulses brought in from proprioceptive terminations, such as neuromuscular and neurotendinous endings, are carried also by fibers of the medial bundle. Such fibers, after sending collaterals to the spinal gray, divide dichotomously in the region of the dorsal funiculus. The collaterals to the gray synapse on the same and probably on the opposite side with the cells of Clarke’s nucleus and with the nuclei of origin of the ventral spino-cerebellar and probably of the spino-vestibular and spino-olivary (in cervical regions) fasciculi. They also come into direct or indirect synaptic relations with motor neurons. After the division of the stem fibers, the longer branches ascend. These branches give off collaterals at all levels and, in lower mammals, many of the main fibers terminate within the cord. Some bundles in all mammals — an increasingly large proportion in the higher mammals — reach the upper levels of the cord and terminate in nucleus gracilis and nucleus cuneatus of the medulla oblongata. The arrangement of the fibers of the dorsal funiculi is such that those from the lower extremities and from the lower part of the body He near the posterior median septum, while those from the

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 289

upper parts of the body and the upper limbs are added on the ventral and on the lateral sides. Descending collaterals of the dorsal funicular fibers are found in the triangular, oval, and conama fields of the cord (see diagram).

To recapitulate : pain, temperature, and general tactile sensibility (according to certain observers, some general muscle sense as well) brought in over the dorsal roots of the spinal nerves are relayed in dorsal horn gray and then carried by arcuate fibers through the anterior or ventral white commissure to the lateral and ventral spino-thalamic fasciculi of the cord, in which tracts they ascend. These spino-thalamic tracts are in general comparable to the spino-mesencephalic and spino-bulbar tracts of submammalian forms. They are descendants of the secondary sensory tract of lower vertebrates, which is regarded as phylogenetically the oldest secondary ascending fasciculus of the cord. Even in man the spino-thalamic fascicuU (particularly the lateral spino-thalamic fasciculus) are supposed to consist, in part at least, of chains of neurons. In addition to these fasciculi, in mammals there is a great accumulation of ascending fibers in the dorsal funiculi and well-developed nuclei associated with their terminations in the medulla oblongata. This great development of the dorsal funiculi and their associated gray had begun to appear in reptiles and birds and attains a progressively greater development in passing from lower to higher mammals. Thus the higher the animals rank in the series of mammals the greater are the dorsal funiculi. This appears to be due to the increased development of the stereognostic sense. Clinically it can be demonstrated that in man the ascending dorsal funicular fibers carry impulses for finer tactile discrimination and for deep joint and muscle sense, these latter permitting an appreciation of the position of the parts of the body. Anatomically, this greater development is associated with an increase in the number and the differentiation of the nerve terminations in the skin, tendons, muscles, and joints. The progressive increase in the percentage of the total white substance of the cervical spinal cord occupied by the dorsal funicular fibers is illustrated by the following figures. In the cat these funiculi make up 22 per cent, in monkeys 26 per cent, and in man 39 per cent of the total white matter of the cord. The nucleus gracilis (nucleus of Goll), situated the more medially, receives ascending fibers by way of the fasciculus gracihs from the lumbar region and sacral region. The nucleus cuneatus, lying lateralward, is the end station for fibers from the thoracic and cervical regions by way of fasciculus cuneatus. The reasons underlying the differentiation into these two nuclear masses are not clearly understood. The fact that the nuclei are less clearly separated in mammals other than primates suggests that the separation may be associated with the great independence of the upper from the lower extremities in these highest mammals and particularly in man. There is considerable variation in the degree of development of the nuclei in the different forms. This is especially true of the nucleus gracilis. In Cetacea, where the posterior extremities are lacking, this nucleus is noticeably small. In some forms (edentates, chimpanzee) it is so large that it arches dorsolaterally oyer the nucleus cuneatus. Both nucleus gracilis and nucleus cuneatus, but particularly the former, show evident lamination in certain animals having large tails, such as

290 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Cebus and Ateles. Between the nucleus gracilis and its fellow of the other side is an unpaired nucleus — the nucleus of Bischoff — which is present in many but not in all tailed animals and is sometimes present in forms where tails are not particularly well developed. The specific function of this nucleus is not well understood at present. The lateral part of the nucleus cuneatus (nucleus cuneatus externus or nucleus Burdach-Monakow) has ascending fiber connections with the cerebellum of the same side and forms also a descending homolateral spinal fasciculus.

Neuraxes from cell bodies of nucleus gracilis and nucleus cuneatus swing ventromedialward as internal arcuate fibers, cross in the sensory decussation, and turn forward as the major constituents of the medial lemniscus, through which tract they reach the ventral part of the lateral division of the thalamus. This lemniscus is present in all mammals but increases in size with the increase in number of the posterior funicular fibers and the development of their associated nuclear masses. It is the pathway by which proprioceptive impulses and finer discriminative tactile sensibihty are carried to the thalamus. Other secondary sensory pathways from the cord are the spino-tectal, spino-vestibular, and spinoohvary fasciculi.

A great development of the descending fascicuU or paths from higher centers to the cord occurs in mammals as compared with submammalian forms. In addition to the vestibulo-spinal fibers which enter the cord with the medial longitudinal fasciculus there is a vestibulo-spinal path in the ventrolateral region of either side of the cord. Reticulo-spinal paths are well developed in many mammals (cat, for example) and probably in man. Descending pathways, carrying impulses directly or indirectly from respiratory centers and other visceral centers of the medulla, enter the upper levels of the cord. A cerebellospinal tract, the fasciculus of Russell, reaches the upper cord. Descending impulses from the tectum are carried by both medial and lateral tecto-spinal paths. Of these the lateral tecto-spinal is of particular interest, since it distributes to the cells of origin of somatic motor fibers. The medial tecto-spinal tract terminates in the intermediolateral column that gives rise to preganglionics for the cervical sympathetic gangUa. Thus this medial tecto-spinal tract forms a part of the pathway traveled by impulses which produce dilation of the pupil (see discussion, p. 1074). Olivo-spinal fibers from the inferior olive to the cervical spinal cord are concerned in conveying nervous impulses to motor neurons supplying neck musculature. These muscles determine the position of the head.

The rubro-spinal tract is situated in the medial part of the lateral funiculus of the cord, between the lateral cortico-spinal and the lateral tecto-spinal tracts. It is much larger in most mammals than in birds and reptiles and extends farther caudad. While it is possible to distinguish in these descending tracts old systems which have become enlarged to meet changed and more complex conditions of mammalian behavior, the projection of impulses from the forebrain directly upon the spinal cord takes place through a newly acquired tract in mammals. The cortico-spinal path arises from the giant pyramidal cells of the precentral

THE COMPARATIVE ANATOMY OP THE SPINAL CORD 291

gyrus (also from ‘premotor’ cortex, p. 1064), descends through the posterior limb of the internal capsule and the middle part of the cerebral peduncle, interdigitates with the bundles of the pons, and collects into a compact fiber mass on the ventral surface of the medulla oblongata, where it is known as the pyramid. From here on the course varies somewhat in different mammals. In man about three-fourths of the tract decussates at the line between medulla oblongata and cord, in the so-called pyramidal or motor decussation. These crossed fibers form the lateral pyramidal or lateral cortico-spinal tract. According to some observers, this tract also carries some uncrossed fibers. It extends usually throughout the cord, stem fibers and collaterals terminating around somatic motor cells at all levels. The remaining one-fourth forms a ventral corticospinal path of varying length, which runs in the ventral white column of the cord and sends its fibem through the anterior or ventral white commissure to terminate around somatic motor neurons on the opposite side of the cord. It is supposed to be concerned particularly with the innervation of trunk musculature.

The prevailing course of the homologue of the human lateral cortico-spinal tract in monotremes, marsupials, ungulates, and rodents is in the dorsal funiculi. This suggests that the tract grew into the cord under the neurobiotactic influence of sensory fibers of the region. In marsupials- and ungulates the lateral cortico-spinal path descends only so far as the cervical region ; in carnivores and rodents it reaches the lumbar cord. In primates it extends into the lumbar cord and probably runs the whole length of the cord in man. The ventral cortico-spinal path, which is smaller than the lateral tract, has been traced in rodents into the lumbar cord. It has been carried that far in several higher mammals.

A progressive enlargement of the cortico-spinal system is found in passing from lower to higher mammals. In dogs these paths form about 10 per cent of the total white substance of the cord ; in monkeys about 20 per cent, and in man about 30 per cent. These figures furnish anatomical evidence for the well-recognized fact that the higher the position of the animal in the evolutionary scale the greater the dominance of its cortical centers over the centers of brain stem and cord, and consequently the greater its influence over body movements and over behavior. It follows that the possibifity of independent functioning of subcortical centers is much diminished in mammals and particularly in higher mammals and in man. Consequently, lesions of the motor cortex in the lowest mammals cause little if any permanent disturbance to motility, while similar lesions in human cortex may well disable a man for life. It is of interest that in unilateral lesions of the human motor cortex there is a tendency for the legs to show somewhat more improvement and a quicker approach to recovery than the arms. This is believed to be due to the fact that local cord reflexes play a larger part in leg than in arm movements. It is possible that it indicates, likewise, a bilateral cortical innervation for the lower extremities.

292 NERVOUS SYSTEMS OF ^^RTEBRATES AND OF MAN

BIBLIOGRAPHY

AMPHIOXUS

{Includes literature on Amphioxits for Chapters VIII and IX)

Ariens ICappers, C. U. 1926. The meninges in lower vertebrates compared Mth those in mammals. Arch. Neurol, and Psychiat., vol. 15, p. 281.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

Ayers, H. 1921. Vertebrate cephalogenesis. V. Origin of jaw apparatus and trigeminus complex — Amphioxus, Ammocoetes, Bdellostoma, Callorhynchus. J. Comp. Neurol., vol. 33, p. 339.

. 1921a. Ventral spinal nerves in Amphioxus. J. Comp. Neurol., vol. 33, p. 155.

Boeke, j. 1902. On the infundibular region of the brain of Amphioxus lanccolatus. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 4, p. 695.

. 1902a. On the structure of the light-percepting cells in the spinal cord, on the

neurofibrillae in the ganglion cells and on the innervation of the striped muscles in Amphioxus lanceolatus. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 5, p. 350.

. 1902b. Ueber das Homologon des Infundibularorganes bei Amphioxus lanceolatus. Anat. Anz., Bd. 21, S. 411.

. 1907. On the structure of the nerve-cells in the central nervous system of

Branchiostoma lanceolatum. First communication. Kon. Akad. v. Wetensch. to Amsterdam, Proc. sect, sc., vol. 10', p. 86.

. 1908. Das Infundibularorgan im Gehime des Amphioxus. Anat. Anz., Bd. 32,

S. 473.

. 1908a. Die Innervierung der Muskelsegmente des Amphioxus (Branchiostoma

lanceolatum) nebst einigen Bemerkungen fiber die Endigungsweise der motorisclien Nerven bei den Vertebraten. Anat. Anz., Bd. 33, S. 273.

. 1908b. On the structure of the ganglion-cells in the central nervous system of

Branchiostoma lanceolata. Second communication. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 11', p. 53.

. 1913. Neue Beobachtungen iiber das Infundibularorgan im Gehirn des Am phio.xus und das homologe Organ des Craniotengehirnes. Anat. Anz., Bd. 44, S. 460.

Dels.man, H. C. 1913. 1st das Hirnbliischen des Amphioxus dem Gehirn dor Kranioten homolog? Anat. Anz., Bd. 44, S. 481.

. 1913a. Der Ursprung der Vertebraten. Eine neue Theorie. Mitt. a. d. zool.

Station zu Neapcl, Leipzig, Bd. 20, S. 647.

. 1918. Short history of the head of vertebrates. Kon. Akad. v. Wetcnscli. tc

Amsterdam, Proc. sect, sc., vol. 20-, p. 1005.

Dogiel, a. S. 1903. Das periphere Ners'-ensystem des Amphioxus (Branchiostoma lanceolatum). Anat. Heftc, Bd. 21, S. 145.

Droogleever Fortuyn, a. B. 1919. On two ncr\’es of vertebrates agreeing in structure with nerves of invertebrates. Kon. Akad. v. Wetensch. tc Amsterdam, Proc. sect, sc., vol. 2P, p. 956.

Edin’ger, L. 1900. Einiges vom “Gehirn ” des Amphioxus. Anat. Anz., Bd. 28, S. 417.

Fr.^nz, U. 1923. Haut, Sinnc.sorgane und Nervensystem der Akranicr. (Fauna ct Analomia ccylanica. Nr. 13.) Jcnaische Zcitschr. f. Naturw., Bd. 59, S. 401.

Fu.s.\ri, II. 1889. Beitrag zum Studium des peripheren Nerv'cnsj'slems von Amphioxus lanccolatus. internat. Monat.schr. f. Anat. u. Physiol., Bd. 6, S. 125.

Gegeniuur, C. 1898. Vcrglcichendc Anatomic der Wirbclthierc mit Berilcksichtigung der Wirljcllopen, Toil II, W. Engelmanii, I.eipzig.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 293

Hatschek, B. 1881. Studien uber die Entwicklung von Amphioxus. Arb. a. d. zool. Inst. d. Univ. Wien.

. 1892. Die Metamerie des Amphioxus und des Ammocoetes. Verhand. d. anat.

Gesellsch., Bd. 6, S. 136.

Hesse, R. 1898. Untersuchungen iiber die Organe der Lichtempfindung bei niederen Thieren. IV. Die Sehorgane des Amphioxus. Zeitschr. f. wissensch. Zool., Bd. 63, S. 456.

Heymans, J. F., and van der Stricht, O. 1898. Sur le systeme nerveux de 1’ Amphioxus et en particulier sur la constitution de la gfoese des racines sensibles. Mem. cour. et d. sav. Strang, de I’Akad. R. Bruxelles, vol. 16.

Johnston, J. B. 1905. The cranial and spinal ganglia and the viscero-motor roots in Amphioxus. Biol. Bull., Marine Biol. Lab., Woods Hole, Mass., vol. 9, p. 112.

Joseph, H. 1904. Ueber eigentumliche Zellstrukturen im Zentralnervensystem von Amphioxus. Verhandl. d. anat. Gesellsch., Bd. 18, S. 16.

VON Kolliker, a. 1843. Ueber das Geruchsorgan von Amphioxus. Arch. f. Anat., Physiol, u. wissensch. Med. (Muller’s), S. 32.

Krause, W. 1897. Die Farbenempfindung des Amphioxus. Zool. Anz., Bd. 20, S. 513.

. 1898. Die Lichtempfindung des Amphioxus. Anat. Anz., Bd. 14, S. 470.

VON Kupffer, K. 1906. Die Morphogenie des Centralnervensystems. Handbuch der vergleichenden und experimentellen Entwickelungslehre der Wirbeltiere, 0. Hertwig, Bd. 2, Teil 3, S. 1. G. Fischer, Jena.

Kutchin, H. L. 1913. Studies on the peripheral nen'ous system of Amphioxus. Contrib. Zool. Lab., Harvard, no. 240. Also Proc. Am. Acad. Arts and Sc., vol. 49, p. 571.

Langerhans, P. 1876. Zur Anatomie des Amphioxus lanceolatus. Arch. f. mikr. Anat., Bd. 12, S. 290.

VON Lenhoss^k, M. 1895. Der feinere Bau des Nervensystems im Lichte neuester Forschungen. 2*® Aufl., Fischer, Berlin.

Muller, E. 1899. Studien iiber Neuroglia. Arch. f. mikr. Anat., Bd. 55, S. 11.

Nansen, F. 1885. The structure and combination of the histological elements of the central nervous system. Bergens Museum Aarsberetung.

Nusslin. 1877. Zur IGitik des Amphioxusauges. Inaug. Diss., Tubingen naturwiss. Fakultat (quoted from Franz, ’23).

Owsjannikow. 1868. Ueber das Zentralnervensystem des Amphioxus. Bull. Acad. Sc. St. Petersbourg, vol. 12.

Parker, G. H. 1908. The sensory reactions of Amphioxus. Proc. Am. Acad. Arts and Sc., vol. 43, no. 16, p. 415.

Platt, J. B. 1892. Fibres connecting the central nervous system and chorda in Amphioxus. Anat. Anz., Bd. 7, S. 282.

Retzius, G. 1891. Zur Kenntniss des centralen Nervensystems von Amphioxus lanceolatus. Biol. Untersuch., N. F., Bd. 2, S. 29.

. 1898. Die Methylenblaufarbung bei dem lebenden Amphioxus. Biol. Untersuch., N. F., Bd. 8, S. 118.

Rohde, E. 1888. Histologische Untersuchungen iiber das Nervensystem von Amphioxus. Zool. Beitr., Bd. 2, Hefte 2.

Schneider, K. C. 1902. Lehrbuch der vergleichenden Histologie der Tiere. G. Fischer, Jena.

. 1908. Histologisches Praktikum der Tiere. Jena.

Stendell, W. 1914. Die Hypophysis cerebri. Bd. 8 of Lehrbuch der vergleichende mikroskopischen Anatomie der Wirbeltiere by A. Oppel. G. Fischer, Jena.

. 1914. Zur Histologie des Ruckenmarkes von Amphioxus. Anat. Anz., Bd. 46,

R 258.

294 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Stieda, L 1873. Studien iiber Amphioxus lanceolatus. Mdm. Acad. Sc. St. P(5tersbourg. S^r. 7, vol. 19.

VAN WiJHE, J. W. 1884. Ueber den vorderen Neuroporus und die phylogenetische Function des Canalis neurentericus der Wirbelthiere. Zool. Anz., Bd. 7, S. 683.

. 1889. Die Kopfregion der Cranioten beim Amphioxus, nebst Bemerkungen iiber

die Wirbeltheorie des Schadels. Anat. Anz., Bd. 4, S. 558.

. 1893. Ueber Amphioxus. Anat. Anz., Bd. 8, S. 152.

. 1894. Over de herzenzenuwen der Cranioten bij Amphioxus. Verslag. kon.

Akad. V. Wetensch. te Amsterdam, vol. 3, p. 108.

. 1902. Beitrage zur Anatomie der Kopfregion des Amphioxus lanceolatus.

Petrus Camper. Nederl. bijdr. t. de anat., vol. 1, p. 109.

. 1907. Die Homologisirung des Mundes des Amphioxus und die primitive Leibes gliederung der Wirbelthiere. Petrus Camper, Nederl. bijdr. t. de anat., vol. 4, p. 61.

. 1913. On the metamorphosis of Amphioxus lanceolatus. Kon. Akad. v.

Wetensch. te Amsterdam, Proc. sect, sc., vol. 16^, p. 574.

. 1914. Studien iiber Amphioxus. I. Mund und Darmkanal wahrend der

Metamorphose. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, vol. 18, p. 3.

Wolff, M. 1907. Bemerkungen zur Morphologie und zur Genese des AmphioxusRiickenmarkes. Biol. Zentralbl., Bd. 27, S. 186, 196 u. 223.

CVCLOSTOMES

Allen, W. F. 1916. Studies on the spinal cord and medulla of cyclostomes with special reference to the formation and expansion of the roof plate and the flattening of the spinal cord. J. Comp. Neurol., vol. 26, p. 9.

. 1917. Distribution of the spinal nerves in Polistotrema and some special studies

on the development of spinal nerves. J. Comp. Neurol., vol. 28, p. 137.

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

. 1925. The meninges in Cyclostomes, Selachians, and Teleosts compared with

those in man. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 28, p. 72.

Ayers, H. 1908. The ventricular fibres in the brain of myxinoids. Anat. Anz., Bd. 32, S. 445.

Beccari, N. 1909. Le cellule dorsali o posteriori dei ciclostomi. Monitore zool. ital., vol. 20, no. 11.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny. A study of the phenomena of neurobiotaxis. Part I. Cyclostomi and Pisces. J. Comp. Neurol., vol. 27, p, 467.

Brandt, W. 1922. Das Darmnervensystem von Myxine glutinosa. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 65, S. 284.

Butschli, 0. 1912. Vorlesungen fiber vergleichende Anatomie. W. Engelmann,

Leipzig.

Cole, F. J. 1896. On the cranial nerves of Chimaera monstrosa. Tr. Roy. Soc., Edinburgh, vol. 38, p. 631.

. 1898. The peripheral distribution of the cranial nerves of Ammocoetes. Anal.

Anz., Bd. 15, S. 195.

. 1907. A monograph on the general morphology of the myxinoid fi.shcs, ba.scd on

a study of Myxine. Part II. The anatomy of the muscles. Tr. Roy. Soc., Edinburgh, vol. 45, p. 683.

Dendy, a. 1902. On a pair of ciliated grooves in the brain of the Ammocoetc apparently serving to promote the circulation of the fluid in the brain-ca\'ity. Proc. Roy. Soc., London, vol. 69, p. 485.

THE C0iMPARATm5 ANATOMY OF THE SPINAL CORD 295

EDJNGKit, L. 190S. VorlcsunRcn fiber den Biiu der nervosen Centralorgane des ?iIcn?chon und dor Thierc. 7" Aufl., F. C. W. Vogel, Leipzig.

Foettixgki!. 1S7G. Kcchcrches sur la structure de I’epiderme des Cyclostomes. Bull, de I’Acad. roy. d. sc., d. let. et d. beaiLx-arts de Belgique, 2' Serie, vol. 14.

Freud, S. 1877. Ueber den Ursprung der hintcren Nen’cnwurzeln im Ruckenmark Yom Ammocoetes (Petromj-zon plancri). Sitzungsb. d. K. Akad. d. Wissensch. zu Wien, iMath.-nat. Cl., Bd. 75, Abt. 3, S. 15.

. ^ 1878. Ueber Spinalganglien und RQckenmark des Petromyzon. Sitzungsb. d.

K. Akad. d. Wissensch. zu Wien, Math.-nat. Cl., Bd. 78, Abt. 3, S. 81.

FtlnnniNGER, M. 1897. Ueber die spino-occipitalen Ncrv’cn der Selachier und Holoccphalen und ihre vergleichende ^Iorphologic. Fest.schr. f. Carl Gegcnbaur, Bd. 3, S. 349. W. Engclmann, I.z;ipzig.

Fusari, R. 190G. Contributo alio studio dei ner\-i cutanei c delle terminazioni nella cute c nella mucosa, omlc dell’ Ammococte.s branchialis. Arch, per Ic sc. med., Torino, vol. 30, p. GOO.

Gegen'rauu, C. 1898. Vergleichende Anatomie der Wirbclthierc mit Berucksichtigung dcr AVirbelloscn. W. Engclmann, lAiipzig.

Ja.vse.v, J. 1930. The brain of Mj-xinc glutinosa. J. Comp. Neurol., vol. 49, p. 359.

JorrxsTO.v, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

. 190G. The nen-ous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

. 1908. Note on (he pre.sence or absence of the glossopharjTigeal nen-e in myxi noids. .<Vnat. Rcc., vol. 2, p. 233.

■. 190Sa. Additional notes on the cranial nen’es of Petromyzonts. J. Comp.

Neurol., vol. 18. p. 5G9.

JuLi.v, C. 18S7. Lc systeme ncn-cu.x grand .sympathique de I’Ammocoetes. Anat. Anz., Bd. 2, S. 192.

IvoLMEn, W. 1905. Zur Kenntn's des Ruckenmarks von Ammocoetes. Anat. Hefte, Bd. 29, S. 1G3.

VON' KurFFER, C. 1890. Die Entwicklung von Petromyzon planeri. Arch. f. mikr. Anat., Bd. 35, S. 4G9.

Levi, G. 1908. I gangli ccrcbrospinali. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di cmbriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. G, S. 188. 1912.

lilAHEN'Gfii, G. 1903. Alcunc particolarifa di struttura e di innenmzione della cute dcir Ammocoetes branchialis. Zcitschr. f. wissensch. Zook, Bd. 75, S. 421.

MOller, E. 1899. Studicn uber Neuroglia. Arch. f. mikr. Anat., Bd. 55, S. 11.

Nansen", F. 1885. The structure and combination of the histological elements of the central nervous .system. Bergens Museum Aarsberetung.

Nicholes, G. E. 19*12. An c.xperimental investigation on the function of Reissner’s fiber. Anat. Anz., Bd. 40, S. 409.

■ . 1912a. The structure and development of Reissner’s fiber and the subcom missural organ. Part I. Quart. J. Micr. Sc., vol. 58, p. 1.

Parker, G. H. 1912. The relation of smell, taste, and the common chemical sense in vertebrates. J. Acad. Nat. Sc., Philadelphia, vol. 15, p. 219, 2nd series.

Ranson, W. B., and Thompson, D’A. W. 1886. On the spinal and visceral nerves of Cyclostomata. Zool. Anz., no. 227. (Quoted from Allen, ’17.)

Reissner, E. 1860. Bcitrage zur Kenntniss von Bau des Riickenmarkes von Petromyzon fl’uviatilis. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 545.

Retzius, G. 1890. Ueber die Ganglienzellen der Cercbrospinalganglien und fiber subcutane Ganglienzellen bei Myxine glutinosa. Biol. Untersuch., N. F. Bd. 1, S. 97.

. 1891. Zur Kenntniss des centralen Nervensystems von Myxine glutinosa.

Biol. Untersuch., N. F., Bd. 2, S. 47.

296 NERVOUS SYSTEMS OF VERTEBRATES AND OF mN

Retzius, G. 1892. Die sensiblen Nen-enendigungen in der Haut des Pctromj'zon. Biol. Untersuch., N. F., Bd. 3, S. 41.

. 1892a. Das hintere Ende des Riickenmarks von Petromyzon. Biol. Foren.

Forhandl. Verhandl. d. biol. Ver. in Stockholm.

Rothig, P. 1913. Beitrage zur Lehre der Neurobiotaxis. Die Verschiebung des motorischen Kerne in der Oblongata bei MjTcine glutinosa und bei einigen Amphibien (Necturus mac., CrjTJtobranchus japonicus, Bufo and Rana). Verslag. kon. Akad. V. Wetcnschappen te Amsterdam, vol. 22, p. 315.

Stefanelli, a. 1932. Terminazioni ner\'ose di noto e di senso nci pctromvzonti. Arch. zool. Ital., vol. 17, p. 85.

Steuzi, G. 1900-01. Ricerche intomo all’anatomia comparata ed all’ontogencsi dellc meningi. Atti d. r. Inst. Veneto di sc., lett., ed arti, vol. 60, part 2, p. 1101.

. 1904. Die Blutgefiisse des Riickenmarks. Untersuchungcn fiber ihrc vcrglci chende Anatomic und Entwicklungsgeschichte. Anat. Hefte, Bd. 24, Abt. 1, S. 1.

. 1909. II sistema nen’oso centrale dei vertebrati ; ricerche anatomiche ed cmbrio logischc. Vol. I, Ciclostomi. A. Draghi, Padova.

Stcdnicka, F. K. 1913. Epidermoidalc Sinneszcllen bei jungcn Ammocoeten (Proammocoeten). Anat. Anz., Bd. 44, S. 102.

Tretjakoff, D. 1909. Das Nen'ensystem von Ammococtcs. I. Das Rfickenmark. Arch. f. mikr. Anat., Bd. 73, S. 607.

. 1913. Die zentralen Sinnesorgane bei Petromyzon. Arch. f. mikr. Anat., Bd.

83, S. 68.

WonxiTiNGTON, J. 1905. The descriptive anatomy of the brain and cranial ncn'cs of Bdcllostoma dombcyi. Quart. J. Micr. Sc., vol. 49, part 1, p. 137.

PLAGIOSTOMES

AnikNs KAPPEn.s, C. U. 1906. The structure of the teleostean and selachian brain. J. Comp. Neurol., vol. 16, p. 1.

. 1911. Wcitere Mitteilungcn fiber Ncurobiotaxis. VI. The migration of the

motor root cells of the vagus group and the phylogenetic dilTcrentiation of the hypoglos.«us nucleus from the spino-occipital system. Psychiat. en ncurol. Bl., Amsterdam, Bd. 15, S. 408.

. 1912. Idem. VII. Die phylogcnetische Entwicklung der motorischen IVurzcI keme in Oblongata und Alittclhim. Folia neuro-biol., Bd. 6, Sommer-ErgiinzungsHoft, S. I.

. 1920-21. Vcrglcichcndc Anatomic des Ncr\’cnsystcms. (German edition of

prc=cnt text.) E. F. Bohn, Haarlem.

Ba.vchi, a. 1903. Sullc vie di conne.^ionc del ccrvellctto. Ricerche anatoino-companitivc c sperimentali. Arch. ital. di anat. c di cmbriol., vol. 2, p. 426.

Bearu, j. 1892. Tlic transient ganglion cells and their nerves in Raja batis. Anat. Anz., Bd. 7, S. 191.

. 1896, The hi‘-torj' of a transient nen’ous apparatus in certain Ichthyop'-ida.

An account of the development and degeneration of ganglion-cells and nervc-fil>ers. Part I, ILaja batis. Zf)ol. Jahrb., Abt. f. Anat., Bd. 9, S. 319.

Bop.riit.iiT, M. 1903. Zur l\enntni.« de.s ZentralnervensyRtems von Torjtedo. 1”' .Mitteilung, Neurol. Arb. herau'-gegelten von 0. Vogt, G. Fischer, Jena.

. 1900. Idem. Zweite Mitteilung. Morphol. Jahrb., Bd. 30, S. 52.

Bitofwfi!, B. 1915. Die biologi'-che Bedeutung der Dermatornerie. Beitrag mr Kenntni'- der Seginentalanatomie und der Sensibilitat^Ieitung irn Rtickenrnark und in der .Merlulla ribloncata. Folia neuro-biol., Bd. 9, S. 225. l)AUi.oiii S’, \’. 191-}. On the electric motor nerve centers in the skates (Hajid.ae) Science, vol. 10, ii'). 1011, p. 802.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 297

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7** Aufl., F. C. W. Vogel, Leipzig.

Furbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologie. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

Johnston, J. B. 1906. The nervous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

Kalberlah, F. 1900. Ueber das Ruckenmark der Plagiostomen. Ein Beitrag zur vergleichenden Anatomie des Centralnervensystems. Zeitschr. f. Naturwissensch , Bd. 73, S. 1.

Keenan, E. 1928. The phylogenetic development of the substantia gelatinosa Rolandi. Part I. Fishes. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 31, no. 8, p. 1.

VON Kupfeer, K. 1906. Die Morphogenie des Centralnervensystems. Handbuch der vergleichenden und experimentellen Entwickelungslehre der Wirbeltiere, O. Hertwig, Bd. 2, Teil 3, S. 1. G. Fischer, Jena.

VON Lenhoss^k, M. 1892. Beobachtungen an den Spinalganglien und dem Riickenmark von Pristiuruscmbryonen. Anat. Anz., Bd. 7, S. 519.

. 1895. Der feinere Bau des Nervensystems im Lichte neuester Forschungen.

2" Aufl., Fischer, Berlin.

Levi, G. 1908. I gangli cerebrospinal!. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

MOller, E. 1899. Studien iiber Neuroglia. Arch. f. mikr. Anat., Bd. 55, S. 11.

. 1920. Beitrage zur Kenntnis der autonomen Nervensystem. I. Ueber die

Entwicklung des SjTupathicus und des Vagus bei den Selachiem. Arch. f. mikr. Anat., Bd. 94, S. 208.

Nicholes, G. E. 1912. An experimental investigation on the function of Reissner’s fiber and the subcommissural organ. Anat. Anz., Bd. 40, S. 409.

. 1912a. The structure and development of Reissner’s fiber and the subcommissural organ. Part I. Quart. J. Micr. Sc., vol. 58, p. 1.

. 1917. Some experiments on the nature and function of Reissner’s fiber.

J. Comp. Neurol., vol. 27, p. 117.

Norris, H. W., and Hughes, S. P. 1920. The cranial, occipital, and anterior spinal nerves of the dogfish, Squalus acanthias. J. Comp. Neurol., vol. 31, p. 293.

Parker, G. H. 1912. The relation of smell, taste, and the common chemical sense in vertebrates. J. Acad. Nat. Sc., Philadelphia, vol. 15, p. 219. 2nd series.

Ram6n t Cajal, S. 1909-11. Histologic du syst^me nerveux de I’homme et des vert4br4s. 2 vol., A. Maloine, Paris.

Retzius, G. 1895. Ueber den Bau des Riickenmarkes der Selachier. Biol. Untersuch., N. F., Bd. 7, S. 34.

VAN Rijnberk, G. a. 1905. Beobachtungen fiber die Pigmentation der Haut bei Scyllium catulus und canicula, und ihre Zuordnung zu der segmentalen Hautinnen^ation dieser Thiere. Petrus Camper, Nederl. bijdr. t. de anat., vol. 3, p. 137.

. 1905a. I disegni cutanei dei vertebrati in rapporto alia dottrina segmentale.

Arch, di fisiol., vol. 3, p. 1.

. 1917. Sur I’innervation segmentale de la peau de la nageoire thoracique chez

lerequin (Scyllium catalus). Folia neuro-biol., Bd. 10, S. 423.

Sargent, P. E. 1900. Reissner’s fiber in the canalis centralis of vertebrates. Anat. Anz., Bd. 17, S. 33.

298 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Speidel, C. C. 1919. Gland cells of internal secretion in the spinal cord of the skates. Publ. 281, Carnegie Inst., Washington. Also Papers Dep. Marine Biol. Carnegie Inst., Washington, vol. 13, p. 1.

. 1922. Further comparative studies in other fishes of cells that are homologous

to the large irregular glandular cells in the spinal cord of the skates. J. Comp. Neurol., vol. 34, p. 303.

Sterzi, G. 1900-01. Ricerche intomo all’anatomia comparata ed all’ontogenesi delle meningi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 60, part 2, p. 1101.

. 1912. II sistema nervoso centrale dei vertebrati; ricerche anatomiche ed

embriologiche. Vol. 2, Pesci. Libro I. Selaci. A. Draghi, Padova.

Stieda, L. 1873. Ueber den Bau des Riickenmarkes der Rochen und Haie. Zeitschr. f. \vissensch. Zool., Bd. 23, S. 435.

Studnicka, F. K. 1900. Untersuchungen fiber den Bau des Ependyms der ncrvosen Centralorgane. Anat. Hefte, Bd. 15, S. 301.

Wallenberg, A. 1907. Beitriige zur Kenntnis des Gehirns der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

Young, J. 1933. The autonomic nervous system of Selachians. Quart. J. Micr. Sc., vol. 75, p. 571.

GANOIDS AND TELEOSTS

Ariens ICappers, C. U. 1906. The structure of the teleostean and selachian hrain. J. Comp. Neurol., vol. 16, p. 1.

. 1911. Weltere Mitteilungen fiber Neurobiotaxis. VI. The migrations of the

motor root-cells of the vagus group, and the phylogenetic differentiation of the hypoglossus nucleus from the spino-occipital-system. Psychiat. en Neurol. Bl., Amsterdam, Bd. 15, S. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wur zelkeme in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommcr-Ergilnzungs-Heft, S. 1.

. 1920-21. Vergleichende Anatomic des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

. 1925. The meninges in cyclostomcs, selachians, and teleosts compared with those

in man. Kon. Akad. v. Wctensch. te Amsterdam, Proc. sect, sc., vol. 28, p. 72.

. 1926. The meninges in lower vertebrates compared with those in mammals.

Arch. Neurol, and Psychiat., vol. 15, p. 281.

Arsaky. 1813. De piscium cerebro ct medullo spinali. Dissertatio inauguralis, Hatai.

Ballowitz, E. 1899. Das elektrische Organ des afrikanischen Zittcnvclscs (Maloptcrurus electricus Lacdpiide). G. Fischer, Jena.

Bartle.mez, G. W. 191.5. Mauthner’s cell and the nucleus motorius tegmenti. J. Comp. Neurol., vol. 25, p. 87.

Brccari, N. 1907. Ricerche sulle cellule e fibre del Mauthner e sullc loro connessioni in pcsci ed anfibi (Salmo fario, S. iridcus e Salamandrinn pcrspicillata). Arch. itnl. di anat. c di embriol., vol. 6, p. 660.

. 1918-19. Peculiari modalitii ncllc connc.ssioni di alcuni ncuroni del sistema

ncrv’oso centrale dei Pcsci. Arch. ital. di anat. c di embriol., vol. 17, p. 239.

. 1922. Studi comparative sulla struttura del Rliombenccfalo. I. Ncnd spino occipitali c nervo ipoglo.sso. II. Centri tegmentali. Arch. ital. di anat. c di embriol., vol. 19, p. 123.

Biluarz, T. 1857. Das elcclrische Organ des Ziltcrwclses. Anatomisch bc.schricbcn. W. Engclmann, I.x’ipzig.

Bolk, h. 1908. Ueber die scgmcntale Anordnung dor Mclanoblnstcn bci jungon Tclcosticm. Vcrhandl. d. anat. Gcsellsch., Bd. 22, S. 135.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 299

Bolk, L. 1910. Beobachtungen iiber Entwicklung und Lagerung von Pigmentzellen bei Knochenfischembryonen. Arch. f. mikr. Anat., Bd. 75, S. 414.

Brickner, R. M. 1929. A description and interpretation of certain parts of the teleostean midbrain and thalamus. J. Comp. Neurol., vol. 47, p. 225.

Brouwer, B. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der Segmentalanatomie und der Sensibilitatsleitung im Riickenmark und in der Medulla oblongata. Folia neuro-bioL, Bd. 9, S. 225.

Burr, H. S. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol., vol. 45, p. 33.

Dahlgren, U. 1897. The giant ganglion cells in the spinal cord of the order Heterosomata Cope (Anacanthini pleuronectoidei Guenther). Anat. Anz., Bd. 13, S. 281.

. 1898. The giant ganglion cell apparatus. J. Comp. Neurol., vol. 8, p. 177.

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 2“=' Teil, 7*“ Aufl., P. C. W. Vogel, Leipzig.

Favaro, G. 1925. Contributi alio studio morfologico dell’ipofisi caudale rigon fiamento caudale della midolla spinale bei teleostei. Atti d. r. Accad. d. Lincei, Mem. cl. sc., Series 6, vol. 1, p. 29.

Fritsch, G. 1884. Ueber den Angelapparat des Lophius piscatorius. Sitzungsb. d. kon. preuss. Akad. d. Wissensch., Berlin, Bd. 2, S. 1145.

. 1884. Die Elektrischen Fische im Lichte der Descendenzlehre. Berlin.

. 1886. Ueber einige bemerkenswerthe Elemente des Centralnervensystems von

Lophius piscatorius. Arch. f. mikr. Anat., Bd. 27, S. 13. .

FOrbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologie. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

VAN Gehuchten, A. 1895. La moelle 4piniere de la truite. La Cellule, vol. 11, p. 111.

Gihgensohn. 1846. Anatomie und Physiologie des Fischnervensystems. M^m. Acad. Imperiale St. Petersbourg.

Haller, B. 1891. Ueber das Centralnervensystem, insbesondere iiber das Riickenmark von Orthagoriscus mola. Morphol. Jahrb., Bd. 17, S. 198.

. 1895. Untersuchungen iiber das Riickenmark der Teleostier. Morphol. Jahrb.,

Bd. 23, S. 21.

Herrick, C. J. 1899. The cranial and first spinal nerves of Menidia : A contribution upon the nerve components of the bony fishes. J. Comp. Neurol., vol. 15, p. 375.

. 1905. The central gustatory paths in the brains of bony fishes. J. Comp.

Neurol., vol. 15, p. 375.

. 1906. On the centers for taste and touch in the medulla oblongata of fishes.

J. Comp. Neurol., vol. 16, p. 403.

■ . 1907. The tactile centers in the spinal cord and brain of the sea robin (Prionotus

carolinus L.). J. Comp. Neurol., vol. 17, p. 307.

. 1907a. A study of the vagal lobes and funicular nuclei of the brain of the codfish. J. Comp. Neurol., vol. 17, p. 67.

. 1931. An introduction t6 neurology. 5“* edition, W. B. Saunders Co., Philadelphia and London.

Holmgren, E. 1903. Ueber die sogenannten “intracellularen Faden” der Nen-enzellen von Lophius piscatorius. Anat. Anz., Bd. 23, S. 37.

VAN DER Horst, C. J. 1918. Die motorischen Kerne und Bahnen in dem Gehim der Fische. Ihr taxonomischer Wert und neurobiotaktische Bedeutung. Tijdrschr. d. Ned. Dierk., Ver. 16.

Huber, G. Carl. 1899. A contribution on the minute anatomy of the sympathetic ganglia of the different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ. of

Michigan, Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

300 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Johnston, J. B. 1900. The giant ganglion cells in Catostomus and Coregonus. J. Comp. Neurol., vol. 10, p. 375.

. 1901. The brain of Acipenser. A contribution to the morphology of the vertebrate brain. Zool. Jahrb., Abt. f. Anat., Bd. 15, S. 59.

. 1906. The nervous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

JoEBAN, H. 1925. The structure and staining reactions of the Reissner’s fiber apparatus, particularly the subcommissural organ. Am. J. Anat., vol. 34, p. 427.

Keenan, E. 1928. The phylogenetic development of the substantia gelatinosa Rolandi. Part I. Fishes. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 31, no. 8, p. 1.

VON Kolliker, A. 1889-02. Handbuch der Gewebelehre des Menschen. Bd. 2. W. Engelmann, Leipzig.

Kolster, R. 1898. Studien fiber das centrale Nervensystem. 1. TJeber das Rfickenmark einiger Teleostier. A. Hirschwald, Berlin.

Kuntz, A. 1911. The development of the sympathetic nervous system in certain fishes. J. Comp. Neurol., vol. 21, p. 177.

von LenhossAk, M. 1892. Beobachtxmgen an den Spinalganglien und dem Rfickenmark von Pristiurus embryonen. Anat. Anz., Bd. 7, S. 519.

. 1895. Der feinere Bau des Nervensystems im Lichte neuester Forschungen.

2*® Aufl., Fischer, Berlin.

Levi, G. 1908. I gangli cerebrospinali. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

Marenghi, G. 1908. La Neuroglia nei pesci Teleostei. Referred to in Folia neurobiol., Bd. 1.

Martin, I. 1895. Contribution A I’Atude de la structure interne de la moelle 4pinRre chez le poulet et chez la truite. La Cellule, vol. 11, p. 55.

Marui, K. 1918. On the finer structure of the synapse of the Mauthner cell with especial consideration of the "Golgi-net” of Bethe, nervous terminal feet and the “nervous pericellular terminal net” of Held. J. Comp. Neurol., vol. 30, p. 127.

Mauthner, L. 1859. Untersuchungen fiber den Bau des Rfickenmarkes der Fische. Sitzungsb. d. K. Akad. d. Wissensch. zu Wien, Math.-nat. Cl., Bd. 34, S. 31.

Muller, E. 1899. Studien fiber Neuroglia. Arch. f. mikr. Anat., Bd. 55, S. 11.

Norri« H. W., and Hughes, S. P. 1920. The cranial, occipital, and anterior spinal nerves of the dogfish, Squalus acanthias. J. Comp. Neurol., vol. 31, p. 293.

Ram6n y Cajal, S. 1909-11. Histologie du syst4me nerveux de I’homme et des vert4br4s. 2 vols. A. Maloine, Paris.

Retzius, G. 1893. Die nervosen Elemente im Rfickenmarke der Knochenfischc. Biol. Untersuch., N. F., Bd. 5, S. 27.

. 1893. Studien fiber Ependym und Neuroglia. Biol. Untersuch., N. F., Bd. 5,

S. 9.

VAN Rijnberk, G. 1904. Beobachtungen fiber die Pigmentation der Haut bei Scillium catulus und canicula, und ihre Zuordnung zu der segmentalen Hautinnervation dieser Tiere. Petrus Camper, Nederl. bijdr. t. de anat., vol. 3, p. 137.

. 1906. Sulla metameria nel sistema nervoso simpatico. I. L’lnnervazione

pigmentomotrice. Arch, di fisiol., vol. 3, p. 601.

. 1906. Ueber den durch Chromatophoren bedingten Farbenwechsel der Tiere

(sogenannten chromatische Hautfunction). Ergeb. d. Physiol. (Asher u. Spiro), Bd. 5, Abt. 2, S. 347.

Sagemehl, M. 1884. Beitrage zur vergleichenden Anatomie der Fische. II. Einige Bemerkungen fiber die Himhaute der Knochenfischc. Morphol. Jahrb., Bd. 9, S. 457.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 301

Sala, L. 1892. Estructura de la mddula espinal de los batracios. Barcelona.

Sargent, P. E. 1898. The giant ganglion cells in the spinal cord of Ctenolabrus adspersus (Walb.-Goode). J. Comp. Neurol., vol. 8, p. 183.

. 1899. The giant ganglion cells in the spinal cord of Ctenolabrus coeruleus.

Anat. Anz., Bd. 15, S. 212.

Schmidt, W. 1921. Ueber die Umwandlung von Schleimgewebe in Fettgewebe in der Himhaut der ICnochenfische. Arch. f. mikr. Anat., Bd. 95, S. 414.

Schreiner, K. E. 1902. Einige Ergebnisse iiber den Bau und die Entwicklung der Occipitalregion von Amia und Lepidosteus. Zeitschr. f. wissensch. ZooL, Bd 72 S. 467.

Speidel, C. C. 1919. Gland-cells of internal secretion in the spinal cord of skates. Publ. Carnegie Inst. Washington no. 281. Also Papers Dep. Marine Biol. Carnegie Inst. Washington, vol. 13, p. 1.

. 1922. Further comparative studies in other fishes of cells that are homologous

to the large irregular glandular cells in the spinal cord of the skates. J. Comp. Neurol., vol. 34, p. 303.

Stannids, H. 1854. Handbuch der Anatomie der Wirbelthiere. 2 Aufl., Veit u. Co., Berlin.

Stendell, W. 1915. Das Nervus electricus von Mormyrus. Zool. Anz., Bd. 45, S. 438.

Sterzi, G. 1900-01. Ricerche intorno all’anatomia comparata ed all’ontogenesi delle meningi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 60, part 2, p. 1101.

Studnicka, F. K. 1903. Ueber die pericellulare und endocellulare Blutcapillairen der Riesenzellen von Lophius piscatorius. Sitzungsb. d. K. Akad. d. Wissensch. zu Wien, Math.-nat Cl.

Tagliani, G. 1899. Ueber die Riesennervenzellen im Rilckenmarke von Solea impar. Anat. Anz., Bd. 15, S. 234.

Verne, J. 1914. Contribution a I’dtude des cellules neurogliques, specialement au point de leur activity formatrice. Arch, d’anat. micr., vol. 16, p. 149.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehirns der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

Weber, E. H. 1827. Knoten und unpaarer Faden, mit dem sich das Riickenmaik bei einigen Fischen endigt, namentlich beim Cyprinus carpio. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 316.

Young, J. 1931. On the autonomic nervous system of the teleostean fish Uranoscopus scaber. Quart. J. Micr. Sc., vol. 74, p. 491.

Zagorsky. 1833. De systemate nerves piscium considerationes. Dissertatio inauguralis. Dorpat.

AMPHIBIANS

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Ariens Kappers, C. U., and Hammer, E. 1918. Das Zentralnervensystem des Ochsenfrosches (Rana catesbyana). Psychiat. en neurol. Bl., Amsterdam, vol. 22, p. 368.

Bhige, E. a. 1882. Die Zahl der Nervenfasem und der motorischen Ganglienzellen im Riickenmark des Frosches. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 435.

DE Boer, S. 1926. The monosegmental innervation of the muscle fibers of the M. gastrocnemius of the frog. Am. J. Physiol., vol. 75, p. 285.

. 1927. Die segmentelle Innervation einiger Froschmuskeln. Zeitschr. f. Biol.,

Bd. 85, no. 5, S. 471.

302 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Brouwer, B. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der Segmentalanatomie und der Sensibilitatsleitung im Riickenmark und in der Medulla oblongata. Folia neuro-biol., Bd. 9, S. 225.

Burckhardt, K. R. 1889. Histologische Untersuchungen am Riickenmark der Tritonen. Arch. f. mikr. Anat., Bd. 34, S. 131.

Cattell, J. McK., and Stiles, P. G. 1924. The place of fatigue in striated muscle ; evidence for failure of excitation apart from contraction. Am, J. Physiol., vol. 69, p. 645.

CoGHiLL, G. E. 1908. The development of the swimming movement in amphibian embryos. Anat. Rec., vol. 2, p. 148.

. 1909. The reaction to tactile stimuli and the development of the swimming

movement in embryos of Diemyctylus torosus, Eschscholtz. J. Comp. Neurol., vol. 19, p. 83.

. 1913. The primary ventral roots and somatic motor column of Amblystoma.

J. Comp. Neurol., vol. 23, p. 121.

. 1913a. The correlation of structural development and function in the growth

of the vertebrate nervous system. Science, vol. 37.

. 1914. Correlated anatomical and physiological studies of the growth of the

nervous system of Amphibia. I. The afferent system of the trunk of Amblystoma. J. Comp Neurol., vol. 24, p. 161.

. 1924. Idem. III. The floor plate of Amblystoma. J. Comp. Neurol., vol. 37,

p. 37.

. 1924a. Idem. IV. Rates of proliferation and differentiation in the central

nervous system of Amblystoma. J. Comp. Neurol., vol. 37, p. 71.

. 1926. Idem. VII. The growth of the pattern of the association mechanism of

the rhombencephalon and spinal cord of Amblystoma punctatum. J. Comp. Neurol., vol. 42, p. 1.

. 1929. The early development of behavior in Amblystoma and in man. Arch.

Neurol, and Psychiat., vol. 21, p. 989.

Disse, j. 1893. Ueber die Spinalganglien der Amphibien. Verhandl. d. anat. Gesellsch., Verslag. 7, S. 201.

Dogiel, a. S. 1890. Methylenblautinktion der motorischen Nervenendigungen in den Muskeln der Amphibien und Reptilien. Arch. f. mikr. Anat., Bd. 35, S. 305.

Dunn, E. H. 1914. The presence of medullated nerve fibers passing from the spinal ganglion to the ventral root in the frog, Rana pipiens. J. Comp. Neurol., vol. 24, p. 429.

Eckhard, C. 1849. Ueber Reflexbewegungen der vier letzten Nervenpaare des Frosches. Zeitschr. f. rat. Med., Bd. 7, S. 281.

Edinger, L. 1908. Vorlesungen fiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 2'®^ Teil, 7*" Aufi., F. C. W. Vogel, Leipzig.

Gad, j. 1884. Einiges fiber Centren und Leitungsbahnen im Riickenmark des Frosches mit einem Excurs fiber Leitungsbahnen im Riickenmark von Kaninchen und Katze. Stahel, Wfirzburg.

Garven, H. 1925. The nerve endings in the panniculus carnosus of the hedgehog, with special reference to the sympathetic innervation of striated muscle. Brain, vol. 48, p. 380.

Gaupp, E. 1901. Anatomic der Frosches. Teil 2, 3“* Aufi., F. Vieweg u. Sohn, Braunschweig.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Berficksichtigung der Wirbellosen. W. Engelmann, Leipzig.

VAN Gehuchten, a. 1898. La moelle 6piniere des larves des Batraciens (Salamandra maculosa). Arch, de biol., vol. 15, p. 599.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 303

Hardesty, I. 1899. The number and arrangement of the fibers forming the spinal nerves in the frog (Rana virescens). J. Comp. Neurol., vol. 9, p. 64.

. 1900. Further obser\'ations on the conditions determining the number and

arrangement of the fibers forming the spinal nerves of the frog (Rana virescens). J. Comp. Neurol., vol. 10, p. 323.

Herrick, C. J. 1914. The medulla oblongata of larval Amblystoma. J. Comp. Neurol., vol. 24, p. 343.

Herrick, C. J., and Coghill, G. E. 1915. The development of reflex mechanisms in Amblystoma. J. Comp. Neurol., vol. 25, p. 65.

Hines, M. 1927. Nen'e and muscle. Quart. Rev. Biol., vol. 2, no. 2, p. 149. Hooker, D. 1925. Studies on regeneration in the spinal cord. III. Reestablishment of anatomical and physiological continuity after transection in frog tadpoles. J. Comp. Neurol., vol. 38, p. 315.

Horton-Smith, R. j. 1897. On efferent fibres in the posterior roots of the frog. J. Physiol., vol. 21, p. 101.

Huber, G. Carl. 1896. The spinal ganglia of Amphibia. Anat. Anz., Bd. 12, S. 417.

. 1899. A contribution on the minute anatomy of the sympathetic ganglia of

different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ.

of Michigan, Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

Huber, G. Carl, and DeWitt, L. 1898. A contribution on the motor nerve-endings and on the nerve-endings in the muscle spindles. J. Comp. Neurol., vol. 7, p. 169.

Katz, L. N. 1925. The supposed plurisegmental innervation of muscle fibers. Proc. Roy. Soc., London, Ser. B., vol. 99, p. 1.

Keenan, E. 1929. The phylogenetic development of the substantia gelatinosa Rolandi. Part II. Amphibians, reptiles, and birds. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 29, p. 229.

VON Kolliker, a. 1899. See bibliography for Chapter I.

Koppen, M. 1888. Zur Anatomie des Froschgehims. Arch. f. Anat. u. Physiol., Anat. Abt., S. 1.

Kuntz, a. 1911. The development of the sympathetic nervous system in the Amphibia. J. Comp. Neurol., vol. 21, p. 297.

. 1922. Experimental studies on the histogenesis of the sympathetic nervous

system. J. Comp. Neurol., vol. 34, p. 1.

. 1929. The autonomic nervous system. Lea and Febiger, Philadelphia.

Lavdov'sky, M. 1891. Vom Aufbau des Riickenmarks. Histologisches fiber die Neuroglia und die Nen^ensubstanz. Arch. f. mikr. Anat., Bd. 38, S. 264. Laivrentjew, B. j. For references see bibliography for Chapter I.

Levi, G. 1908. I gangli cerebrospinali. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

Ram6n y Cajal, S. 1888. Terminaciones nendosas en los husos musculares de la rana. Rev. trim, histol. norm, y patol., vol. 1, fasc. 1.

. 1909-11. Histologie du systtoe nerveux de I’homme et des vert^brds. 2 vols.

A. Maloine, Paris.

Retzius, G. 1893. Studien fiber Ependym und Neuroglia. Biol. Untersuch., N. F., Bd. 5, S. 9.

Sala y Pons. Estructura de la medula espinal de los batracios. Barcelona, Febrero (Quoted from Ram6n y Cajal, ’09).

Sandmeyer, W. 1892. Sekundare Degeneration nach Rxstirpation motorischer Centra. Zeitschr. f. Biol., Bd. 28 (N. F. Bd. 10), S. 177.

304 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Sano, F. 1905. Beitrage zur Kenntnis der motorischen Keme im Riickenmark der Wirbeltiere. Anat. Anz., Erganzungsch. zu Bd. 27, S. 9.

ScLAVUNOs, G. 1892. Beitrage zur feineren Anatomie des Riickenmarkes der Amphibien. Festschr. f. A. von Kolliker, Leipzig. S. 95.

Sherrington, C. S. 1893. Experiments in examination of the peripheral distribution of the fibers of the posterior roots of some spinal nerves. Phil. Tr. Roy. Soc., London, Ser. B., vol. 184, p. 641.

Smaelwoob, W. M. 1916. A short spinal cord in the toad. Anat. Rec., vol. 10, p. 515.

Steinach, E. 1893. Ueber die motorische Innervation des Darmtractus durch die hinteren Spinalnervenwurzeln. Lotos, Bd. 14.

. 1898. Ueber die viszero-motorischen Functionen der Hinterwurzeln und fiber

die tonische Hemmungswirkung der Medulla oblongata auf den Darm des Frosches. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 71, S. 523.

Sterzi, G. 1900-01. Ricerche intemo all’anatomia comparata ed all’ontogenesi delle meningi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 60, part 2, p. 1101.

Tensen, J. 1927. Einige Bemerkungen fiber das Nervensystem von Pipa pipa. Acta ZooL, Bd. 8, S. 151.

Turkheim. 1902. Ueber das Riickenmark von Cryptobran chos japonicus. Inaugural Dissertation, Wien.

Wallenberg, A. 1907. Die kaudale Endigung der bulbo-spinalen Wurzeln des Trigeminus, Vestibularis und Vagus beim Frosche. Anat. Anz., Bd. 30, S. 564.

REPTILES

AriEns Kappers, C. U. 1911. Weitere Mitteilungen fiber Neurobiotaxis. VI. The migrations of the motor root-cells of the vagus group, and the phylogenetic differentiation of the hypoglossus nucleus from the spino -occipital system. Psychiat. en Neurol. BL, Amsterdam, Bd. 15, S. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wurzel keme in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-ErganzungsHeft, S. 1.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

Banchi, a. 1903. La minuta struttura della midolla spinale dei Chelonii (Emys europaea). Arch. ital. di anat. e di embriol., vol. 2, p. 291.

Beccari, N. 1913. Sulla spettanza delle fibre del Lenhoss^k al sistema del nervo accessoria e contribute alia morfologia di questo nervo. (Osservazioni in Lacerta muralis.) Arch. ital. di anat. e di embriol., vol. 11, p. 299.

. 1914. II IX, N) XI e XII pajo di nervi cranici e i nervi cervicali negli embrioni

di Lacerta muralis. (Contribuzioni alio studio del significato morfologico dei nervi della testa.) Arch. ital. di anat. e di embriol., vol. 13, p. 1.

Brottwer, B. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der Segmentalanatomie und der Sensibilitatsleitung im Riickenmark und in der Medulla oblongata. Folia neuro-biol., Bd. 9, S. 225.

Christensen, K. 1917. The morphology of the brain of Sphenedon. Univ. of Iowa Studies in Nat. Hist., vol. 12, no. 1, p. 3. Published by the University, Iowa City, Iowa.

Edinger, L. 1908. Vorlesungen fiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 2‘ Teil, 7“* Aufl., F. C. W. Vogel, Leipzig.

Francescon, a. 1931. Ricerche sul nervo vago e sul simpatico dei cheloni. Arch. Zool. ital., vol. 16, p. 695.

Frederikse, a. 1931. The lizard’s brain. An investigation of the histological structure of the brain of Lacerta vivipara. Dissertatie, Amsterdam.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 305

Gaskell, W. H. 1885. On a segmental group of ganglion cells in the spinal cord of the alligator. Proc. Physiol. Soc., J. Physiol., vol. 7, Proe. p. 19.

. 1886. On the structure, distribution, and function of the nerves which innervate

the visceral and vascular systems. J. Physiol., vol. 7, p. 1.

. 1889. On the relation between the structure, function, distribution, and origin

of the cranial nerves ; together with a theory of the origin of the nervous system of vertebrata. J. Physiol., vol. 10, p. 153.

Gegenbatjr, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. Bd. 1. W. Engelmann, Leipzig.

VAN Gehuchten, A. 1897. Contribution ^ I’dtude de la moelle dpini5re chez les vert4brds (Tropidonotus natrix). La Cellule, vol. 12, p. 115.

Gitjlliani. 1893. Sulla struttura della midolla spinale della Lacerta virida. Ricerche n. lab. di anat. norm. d. r. Univ. di Roma, vol. 11.

Grimm, J. 1864. Ein Beitrag zur Kenntnis vom Bau des Ruckenmarkes von Vipera berus Lin. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 502.

Higgins, G. M. 1923. Development of the primitive reptilian vertebral column, as shown by a study of Alligator mississippiensis. Am. J. Anat., vol. 31, p. 373.

HmT, A. 1921. Der Grenzstrang des Sympathicus bei einigen Sauriern. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 62, S. 536.

Huber, G. Carl. 1899. A contribution on the minute anatomy of the sympathetic ganglia of different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ.

of Michigan. Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

Huber, G. Carl, and Crosby, E. C. 1926. On thalamic and tectal nuclei and fiber paths in the brain of the American alligator. J. Comp. Neurol., vol. 40, p. 97.

Keenan, E. 1929. The phylogenetic development of the substantia gelatinosa Rolandi. Part II. Amphibians, reptiles, and birds. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 32‘, p. 299.

mv Kollucss, A. 1902. Veber die oberRachhchen Nervenkerne im Marke der Vogel und Reptilien. Zeitschr. f. wissensch. 2!ool., Bd. 72, S. 126.

Kuntz, a. 1911. The development of the sympathetic nervous system in turtles. Am. J. Anat., vol. 11, p. 279.

. 1929. The autonomic nervous system. Lea and Febiger, Philadelphia.

DE Lange, S. J. 1917. Das Hinterhim, das Nachhim und das Riickenmark der Reptilien. Folia neuro-biol., Bd. 10, S. 385.

Larsell, 0. 1926. The cerebellum of reptiles ; lizards and snakes. J. Comp. Neurol., vol. 41, p. 59.

Levi, G. 1908. I ganglt cerebrospinal!. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

Muller, E. 1920. Beitrage zur Kenntniss des autonomen Nervensystems. I. Ueber die Entwicklung des Sympathicus und des Vagus bei den Selachiern. Arch. f. mikr. Anat., Bd. 94, S. 208.

Muller, E., and Liljestrand. 1918. Anatomische und experimen telle Untersuchungen tiber das autonome Nervensystem der Elasmobranchier nebst Bemerkungen uber die Darmnerven bei den Amphibien und Saugetieren. Arch. f. Anat. u. Physiol., Anat. Abt., S. 137.

Ram6n t Cajal, S. 1891. La medulla espinal de los reptiles. Barcelona.

. 1899-1904. Textura del sistema nervioso del hombre y de los vertebrados.

Vol. 1. N. Moya, Madrid.

. 1909-11. Histologie du syst^me nerveux de I’homme et des vert6br4s. Vol. 1.

A. Maloine, Paris.

306 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Eawitz, B. 1915. Das zentrale Nervensystem einiger Madagassicher Reptilien. Teil I. Das Riickenmark. In Voeltzkow Raise in Ort-Afrika, Bd. 4.

Retzitjs, G. 1894. Die embryonale Entwicklung der Riickenmarkselemente bei den Ophidiem. Biol. Untersuch., N. F., Bd. 6, S. 41.

. 1898. Weiteres iiber die embryonale Entwicklung der Riickenmarkselemente

der Ophidier. Biol. Untersuch., N. F., Bd. 8, S. 105.

. 1898a. Zur Kenntniss der Entwicklung der Elemente des Riickenmarks von

Anguis fragilis. Biol. Untersuch., N. F., Bd. 8, S. 109.

Riqxjier, C. 1914. Studio macro- e microscopico sul simpatico dei cheloni con osservazioni intomo alle anastomosi cellulari. Riv. di pat. nerv., vol. 19, p. 422.

Schaffer, K. 1891. Vergleichend-anatomische Untersuchungen iiber Riickenmarksfaserung. Arch. f. mikr. Anat., Bd. 38, S. 157.

Shanklin, W. M. 1930. The central nervous system of Chameleon vulgaris. Acta ZooL, Bd. 11, S. 425.

Shimada, K. 1912. Ueber die Segmentierung des eigentiimlichen Riickenmarksbandes und die “Hoffmann’schen Kerne” (Kolliker) des Riickenmarkes von einigen Schlangen (Trigonocephalus ; Tropidonotus tigrinus). Anat. Anz., Bd. 42, S. 417.

Spanner, R. 1929. Der Bauchsympathicus der Blindschleiche und seine Beziehungen zur Innervation der Niere. Zeitschr. f. wissensch. Biol. ; Abt. B., Zeitschr. f. Zellforsch. u. mikr. Anat., Bd. 8, S. 740.

Stefanelli, a. 1916. Nuovo contribute alia conoscenza delle espansioni sensitive dei Rettili, e considerazioni sulla tessitura del sistema nervoso periferico. Internal. Monatschr. f. Anat. u. Physiol., Bd. 32, S. 22.

Sterzi, a. I. 1904. I gruppi cellulari periferici della midolla spinale dei Rettili. Atti d. Soc. tosc. di sc. nat. Mem., Pisa, vol. 20.

Sterzi, G. 1900-01. Ricerche intorno all’anatomia comparata ed all’ontogenesi delle meningi. E considerazioni sulla filogenesi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 60, part 2, p. 1101.

Takahashi,,D. 1913. Zur vergleichenden Anatomic des Seitenhoms im Riickenmark der Vertebraten. Arb. a. d. neurol. In.st. a. d. Wien. Univ. (Obersteiner’s), Bd. 20, S. 62.

Terni, T. 1914. Sulla correlazione fra ampiezza del territorio di innervazione e volume

delle cellule gangliari : 1. Ricerche sui ganglii spinali della coda nei Chelonii. Anat. Anz., Bd. 47, S. 369.

. 1920. Sulla correlazione fra ampiezza del territorio di innervazioni e grandezza

delle cellule gangliari : 2. Ricerche sui gangli spinali che innervano la coda rigenerata, nei Sauri (Gongylus ocellatus). Arch. ital. di anat. e di embriol., vol. 17, p. 507.

. 1921. Sulla presenza dei tronco del simpatico nella coda di rettili e sulla distribu tione di rami di esso efferenti a muscoli volontari. Gior. d. r. Accad. di med. di Torino, vol. 84.

. 1922. Ricerche istologische sul midollo spinale dei rettilli. Arch. ital. di anat.

e di embriol., vol. 18, Suppl. p. 183.

. 1922. Sulla partacipariosa die sistema nervoso autonomo alia innervaziona dei

muscolo sul simpatico caudale dei Rettili. Schweiz. Arch. f. Neurol, u. Psychiat., vol. 11, p. 60.

. 1922a. Ricerche istologiche sui midollo spinale dei rettili, con particolarc

riguardo ai componenti spinali del fascicolo longitudinale mediale. (Osscrvazioni in Gongylus ocellatus Wagl.) Arch. ital. di anat. e di embriol., vol. 18, Suppl., p. 183.

. 1926. Sui nuclei marginali del midollo spinale dei Sauropsidi. Arch. ital. di

anat. e di embriol., vol. 23, p. 610.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 307

VAN Trigt, H. 1917. La dermatom^rie du Ldzard. Arch, ne^rl. de physiol., S4r. IIIc, vol. 2, p. 51.

Zeehandelaar, I. 1921. Ontogenese und phylogenese der Hinterstrangkeme in Verband mit der Sensibilitat. Folia neuro-biol., Bd. 12, S. 1.

Brans

AriSns Kappers, C. U. 1911. Weitere Mitteilungen liber Neurobiotaxis. VT. The migrations of the motor root-cells of the vagus group, and the phylogenetic differentiation of the hypoglossus nucleus from the spino-occipital-system. Psychiat. en Neurol. Bl., Amsterdam, Bd. 15, S. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wurzel keme in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-ErgangungsHeft, S. 1.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

. 1924. The lumbo-sacral sinus in the spinal cord of birds and its histological constituents. Psychiat. en Neurol. Bl., Amsterdam, Bd. 28, S. 405. See also annual Contributions of the Peking Union Medical College, Peking, 1924, p. 87.

. 1926. The meninges in lower vertebrates compared with those in mammals.

Arch. Neurol, and Psychiat., vol. 15, p. 281.

Bok, S. T. 1915. Die Entwicklung der Hirnner\'en und ihrer zentralen Bahnen. Die stimulogene Fibrillation. Folia neuro-biol., Bd. 9, S. 475.

Brouwer, B. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der Segmentalanatomie und der Sensibilitatsleitung im Riickenmark und in der Medulla oblongata. Folia neuro-biol., Bd. 9, S. 225.

Craigie, E. H. 1928. Observations on the brain of the humming bird (Chrysolampis mosquitus Linn, and Chlorostilbon caribaeus Lawr.). J. Comp. Neurol., vol. 45, p. 377.

Deelman, H. T. 1919. Bijdrage tot de kennis van de dermatomerie bij de vogels (duif). Verslag. kon. Akad. v. Wetensch. te Amsterdam, vol. 28*, no. 5, p. 557.

Duval, M. 1877. Recherches sur le sinus rhomboidal des oiseaux sur son d4veloppement et sur la n4vroglie p4ri4pendymaire. J. de I’anat. et physiol., vol. 13,

p. 1.

Edinger, L. 1908. Vorlesungen fiber den Bau der nenmsen Centralorgane des Menschen und der Thiere. Teil 2, 7*® Aufl., F. C. W. Vogel, Leipzig.

Edinger, L., Wallenberg, A., and Holmes, G. 1903. Untersuchungen iiber die vergleichendende Anatomie des Gehirns. 5. Das Vorderhirn der Vogel. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main, Bd. 20, no. 4, S. 343.

Frenkel, B. 1909. Die Kleinhimbahnen der Taube. Bull, intemat. Acad d. sc. de Cracovie, vol. 2, p. 123.

Friedlander, a. 1898. Untersuchungen fiber das Riickenmark und das IGeinhim der Vogel. Neurol. Centralbl., Bd. 17, S. 351 und 397.

Furbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologic. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

Gaskell, W. H. 1885. On a segmental group of ganglion cells in the spinal cord of the alligator. Proc. Physiol. Soc., J. Physiol., vol. 7, Proc. p. 19.

. 1889. On the relation between the structure, function, distribution, and origin

of the cranial nerves ; together with a theory of the origin of the nervous sj'stem of Vertebrata. J. Physiol., vol. 10, p. 153.

VAN Gehuchten, a. 1893. Les dl6ments nerveux moteurs des racines post<Srieurcs. Ainat. Anz., Bd. 8, S. 215.

308 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

VAN Gelderen, C. 1924. Die Morphologie der Sinus durae matris. Zweiter Teil. Die vergleichende Ontogenie der neurokraniellen Venen der Vogel und Siiugetiere. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 73, S. 541.

Gesell, R. 1926. Another outlook on the chemical regulation of respiration. Science, vol. 63, p. 58.

Golgi, C. 1894. Beitrage zur feineren Anatomie der Centralorgane des Nervensystems. Untersuchungen tiber den feineren Bau des centralen und peripherischen Nervensystems. G. Fischer, Jena.

Groebbels, F. 1927. Die Lage- und Bewegungsreflexe der Vogel. VII. Die Lageund Bewegungsreflexe der Haustaube nach Liisionen des Riickenmarks und der Oblongata. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 218, S. 198.

Hansen-Pruss, 0. C. 1923. Meninges of birds, with a consideration of the sinus rhomboidalis. J. Comp. Neurol., vol. 36, p. 193.

Hardesty, I. 1905. Observations on the spinal cord of the emu and its segmentation. J. Comp. Neurol., vol. 15, p. 81.

Hovy, a. j. 1913. On the relation between the quantity of white and gray substance in the central nervous system. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 16‘, p. 311.

Huber, G. Carl. 1899. A contribution on the minute anatomy of the sympathetic ganglia of the different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ.

of Michigan, Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

Huber, J. F. 1933. Nerve roots and nuclear groups in the spinal cord of the pigeon. Dissertation.

Imhof, G. 1904. Anatomie und Entwicklungsgeschichte des Lumbalmarkes bei den Vogeln. Arch. f. mikr. Anat., Bd. 65, S. 498.

Ingvar, S. 1919. Zur Phylo- und Ontogenese des Kleinhirns. Folia neuro-biol., Bd. 11, S. 205.

Kaiser, L. 1923, 1924. De segmentale innervatie van de huid bij dc duif (Columba livia var. domestica). Nieuwe Verhandl. van het Batafsch Genootschop te Rotterdam, 2'*° Reeks, 9'*'’ Deel, 1‘” Stuk, p. 1. Dissertatie, Amsterdam, 1924.

Keenan, E. 1929. The phylogenetic development of the substantia gelatinosa Rolnndi. Part II. Amphibians, reptiles, and birds. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 32', p. 299.

VON Kolliker, a. 1901. Ueber einen noch unbekannten Nervenzellenkern im Riickenmark der Vogel. Anz. d. K. Akad. d. Wissensch., no. 15.

. 1902. Weitere Beobachtungen iiber die Hofmann ’schen Kerne am Mark der

Vogel. Anat. Anz., Bd. 21, S. 81.

. 1902a. Ueber die oberflachlichen Nervenkerne im Marke der Vogel und Rep tilien. Zeitschr. f. wissensch. Zool., Bd. 72, S. 126.

Krause, R. 1922. Mikroskopische Anatomic der Wirbclticre in Einzcldarstellungcn. Bd. 2. W. de Gruyter and Co., Berlin and Leipzig, 1921-3.

KffiiN, A., AND Trendelenburg, W. 1911. Die exogenen und endogenen Bahncn des Riickenmarks der Taube mit der Degenerationsmethode unter.sucht. Arch. f. Anat. u. Physiol., Anat. Abt., S. 35.

Kuntz, a. 1911. The evolution of the sympathetic nervous system in vertebrates.

J. Comp. Neurol., vol. 21, p. 215.

. 1922. Experimental studie.s on the histogenesis of the sympathetic nervous

.system. J. Comp. Neurol., vol. 34, p. 1.

. 1929. The autonomic nervous .“y.stcin. lAia and Fcbiger, Philadelphia.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 309

Lachi, P. 1889. Alcune particolarita anatomiche del rigonfiamento sacrale nel midollo degli uccelli. Lobi accessori. Atti d. Soc. tosc. di sc. nat. Mem., Pisa, vol 10

p. 280.

Langley, J. N. 1904. On the sympathetic system of birds, and on the muscles which move the feathers. J. Physiol., vol. 30, p. 221.

Latimer, H. 1925. The postnatal growth of the central nervous system of the chicken. J. Comp. Neurol., vol. 38, p. 251.

VON Lenhoss£k, M. 1890. Ueber Nervenfasern in den hinteren Wurzein, welche aus dem Vorderhorn entspringen. Anat. Anz., Bd. 5, S. 360.

. 1894. Ueber oberflachliche Nervenzellen im Riickenmarke des Hiinchens. In

Beitrage zur Histologie des Nervensystems und der Sinnesorgane, p. 79. J. F. Bergmann, Wiesbaden.

. 1895. Der feinere Bau des Nervensystems im Lichte neuester Forschungen.

2*® Aufl., Fischer, Berlin.

Levi, G. 1908. I gangli cerebrospinali. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e di embriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

Martin, I. 1895. Contribution S, l’4tude de la structure interne de la moelle dpinRre chez le poulet et chez la truite. La Cellule, vol. 11, p. 55.

MiiNZER, E., AND Wiener, H. 1898. Beitrage zur Anatomic und Physiologic des Centralnervensystems der Taube. Monatschr. f. Psychiat. u. Neurol., Bd. 3, S. 379.

Papez, j. W. 1929. Comparative neurology. Thos. Y. Crowell Co., New York. PoLJAK, S. 1924. Die Struktur eigentiimlichkeiten des Riickenmarks bei den Chiropteren. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 74, S. 509.

Ram6n y Cajal, S. 1890. A quelle 4poque apparaissent les expansions des cellules nerveuses de la moelle ^pinidre du poulet? Anat. Anz., Bd. 5, S. 609 und 631.

■ . 1890. Sur I’origine et les ramifications des fibres nerveuses de la moelle embryon naire. Anat. Anz., Bd. 5, S. 85 und 111.

. 1893. Neuvo concepto de la histologia de los centres nerviosas. Rev. de

Ciencias Med. de Barcelona, vol. 18, 1892.

. 1899-1904. Textura del sistema nervioso del hombre y de los vertebrados.

Vol. 1. N. Moya, Madrid.

. 1909-11. Histologie du syst^me nerveux de I’homme et des vert6brds.

A. Maloine, Paris.

Retzius, G. 1893. Studien fiber Ependym und Neuroglia. Biol. Untersuch., N. F., Bd. 5, S. 9.

. 1902. Zur Kenntniss der oberflachlichen ventralen Nervenzellen im Lendenmark

der Vogel. Biol. Untersuch., N. F., Bd. 10, S. 21.

Sanders, E. B. 1929. A consideration of certain bulbar, midbrain, and cerebellar centers and fiber tracts in birds. J. Comp. Neurol., vol. 49, p. 155.

Sandmeybr, W. 1892. Secundare Degeneration nach Extirpation motorischer Centra.

Zeitschr. f. Biol., Bd. 28 (N. F. Bd. 10), S. 177.

Sano, F. 1914. Experiences de localisation chez la grenouille, le pigeon, la poule.

Annexe au Rapport presents an XIV CongrSs des Medicins alienistes. Pau, 1914. Sherrington, C. S. 1890. On outlying nerve-cells in the mammalian spinal cord.

Phil. Tr. Roy. Soc., London, Ser. B., vol. 181, p. 33.

Shdiiazono, j. 1902. Das Kleinhirn der Vogel. Arch. f. mikr. Anat., Bd. 80, S. 397. Sparvoli, R. 1907. Sull’innervazione segmentale della cute negli uccelli. Arch. ital. di anat. e di embriol., vol. 7, p. 469.

Sterzi, a. I. 1904. I gruppi cellulari periferici della midolla spinale dei Rettili. Atti d. Soc. tosc. di sc. nat. Mem., Pisa, vol. 20.

310 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Sterzi, G. 1900-01. Ricerche intomo all’anatomie comparata ed aU’ontogenesi delle meningi. E considerazioni sulla filogenesi. Atti d. r. 1st. Veneto di sc., lett. ed arti, vol. 60, part 2, p. 1101.

Stieda, L. 1869. Studien iiber das centrale Nervensystem der Vogel und Saugethiere. Zeitschr. f. wissensch. Zool., Bd. 19, S. 1.

Streeter, G. L. 1904. The structure of the spinal cord of the ostrich. Am. J. Anat., vol. 3, p. 1.

Takahashi, D. 1913. Zur vergleichenden Anatomic des Seitenhorns im Ruckenmark der Vertebraten. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 20, S. 62.

Terni, T. 1920. Sulla correlazione fra ampiezza del territorio di innervazione e grandezza della cellule gangliari. 2. Ricerche sui gangli spinali che innervano la coda rigenerata, nei Sauri (Gongylus ocellatus). Arch. ital. di anat. e di embriol., vol. 17, p. 607.

. 1923. Ricerche anatomiche sul sistema nervoso autonomo degli Ucelli. I. II

sistema pregangliare spinale. Arch. ital. di anat. e di embriol., vol. 20, p. 433.

. 1926. Sul nuclei marginali del midollo spinale dei Sauropsidi. Arch. ital. di

anat. e di embriol., vol. 23, p. 610.

Waeeenberg, a. 1898. Die secundare Acusticusbahn der Taube. Anat. Anz., Bd. 14, S. 353.

. 1904. Neue Untersuchungen iiber den Himstamm der Taube. Anat. Anz., Bd.

24, S. 357.

Zeehandelaar, I. 1921. Ontogenese und phylogenese der Hinterstrangkerne in Verband mit der Sensibilitat. Folia neuro-biol., Bd. 12, S. 1.

MAMMALS

Adrian, E. D. 1925. Interpretation of the electromyogram. Lancet, vol. 208, p. 1229.

Agduhr, E. 1915. Anatomische, statistische und experimentelle Untersuchungen iiber den N. medianus und N. ulnaris, besonders deren motorisches Innervationsgebiet im Vorderarm von Equidae, Cervidae, Bovidae, Ovidae, Suidae, Canidae, und Felidae, speziell von Haustieren, nebst einigen Bemerkungen iiber die Muskulatur desselben Gebietes und iiber N. musculocutaneus. Anat. Hefte, Bd. 52, S. 497.

. 1916. Morphologischer Beweis der doppelten (plurisegmentalen) motorischen

Innervation der einzelnen quergestreiften Muskelfasem bei den Saugctieren. Anat. Anz., Bd. 49, S. 1.

. 1919. Are the cross striated muscle fibers of the extremities also innervated

sympathetically? Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 2P, p. 1231.

Alezais et Peyron. 1920. L’6volution des vestiges mddullaires coccygiens chez I’embryon, et le foetus humain jusqu’au sixidme mois. Compt. rend. Soc. de bioL, vol. 83, p. 230. Also Rdunion biologique de Marseille, 1920.

Allen, W. F. 1927. Experimental-anatomical studies of the visceral bulbospinal pathway in the cat and guinea pig. J. Comp. Neurol., vol. 42, p. 393.

. 1927. Location in the spinal cord of pathways which conduct impulses from the

cerebrum and the superior colliculus, affecting respiration. J. Comp. Neurol., vol. 43, p. 451.

Angulo y Gonzalez, A. W. 1927. The motor nuclei in the cervical cord of the albino rat at birth. J. Comp. Neurol., vol. 43, p. 115.

Archibald, E. 1928. Effect of .sympathectomy upon the pain of organic disease of arteries of the lower limbs and for obscure abdominal pain. Ann. Surg., vol. 68, p. 499.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 311

Aegutinskt, P. 1897. Ueber eine regelmassige Gliedening in der grauen Substanz des Ruckenmarks beim Neugeborenen und uber die Mittelzellen. Arch. f. mUcr. Anat., Bd. 48, S. 496.

Ahiens Kappers, C. U. 1908. Weitere Mitteilungen iiber Neurobiotaxis. Die Selektivitat der Zellenwanderung. Die Bedeutung synchronischer Reizerwandtsehaft. Verlauf und Endigung der zentralen sogenannten motorischen Bahnen. Folia neuro-bfol., Bd. 1, S. 507.

. 1914. Ueber das Rindenproblem und die Tendenz innerer Himteile sich durch

Oberflachen-Vermehrung, statt Volumzunahme zu vergrosseren. Folia neuro-biol Bd. 8, S. 507.

. 1926. The meninges in lower vertebrates compared with those in mammals.

Arch. Neurol, and Psychiat., vol. 15, p. 281.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

Athanassio-Benistt. 1918. Clinical forms of nerve lesions. Univ. of London Press, London.

Auerbach, L. 1890. Zur Anatomic der aufsteigend degenerierenden Systeme des Ruckenmarks. Anat. Anz., Bd. 5, S. 214.

Barker, L. F. 1899. The nervous system and its constituent neurones. Appleton and Co., New York.

Basler, A. 1913. Ueber die Beteiligung des Muskelsinnes am absichtlichen Tasten. Arch. f. d. ges. Physiol. fPfluger’s), Bd. 153, S. 353.

VON Bechterew, W. 1890. Ueber die verschiedenen Lagen und Dimensionen der Pyramidenbahnen beim Menschen und den Thieren und fiber das Vorkonunen von Fasem in denselben, welche sich durch eine frfihere Entwickelung anszeichnen. Neurol. Centralbl., Bd. 9, S. 738.

. 1899. Die Leitungsbahnen im Gehim und Rfickenmark. 2°'* edition, Leipzig.

. 1925. Ueber ein besonderes Fasembfindel in der Lendenkreuzgegend des Rficken marks. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesoh., Bd. 76, S. 779.

VON Bechterew, W., and Holzinger, F. 1894. Die sensiblen Bahnen im Rfickenmark. Neurol. Centralbl., Bd. 13, S. 642.

Bernhardt, M. 1902, Allgemeine Pathologie der peripherischen Nerven. Nothnagel’s Handbuch.

Bertholet, E. 1906. Les voies de la sensibility dolorifique et calorifique dans la moelle. Le N^vraxe, vol. 7, p. 283.

Biach, P. 1906. Vergleichend-anatomische Untersuchungen fiber den Bau der Zentralkanales bei den Saugetieren. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 13, S. 399.

. 1907. Das Rfickenmark der Ungulaten. Arb. a. d. neurol. Inst. a. d. Wien.

Univ. (Obersteiner’s), Bd. 16, S. 487.

Bielschowsky, M. 1908. Ueber den Bau der Spinalganglien unter normalen imd pathologischen Verhaltnissen. J. f. Psychol, u. Neurol., Bd. 11, S. 188.

Bing, R. 1904. Beitrag zur Kenntniss der endogenen Rfickenmarksfasem beim Menschen. Arch, f, Psychiat., Bd. 39, S. 74.

. 1906. Experimentelles zur Physiologic des Tractus spinocerebellares. Arch. f.

Anat. u. Physiol., Physiol. Abt., S. 250.

Benswanger, 0., AND MoELi, C. 1883. Zur Frage der secundaren Degeneration. Neurol. Centralbl., Bd. 2, S. 9.

Bischopf, E. 1900. Beitrag zur Anatomic des Igelgehimes. Anat. Anz., Bd. 18,

S. 348.

Blix, M. 1884. Experimentelle Beitrage zur Losung der Frage fiber die specifcche Energie der Hautnerven. Zeitschr. f. Biol,, Bd. 20 (N. F. Bd. 2), S. 141; also Bd. 21 (N. F. Bd. 3). S. 145.

312 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Boeke, J. 1910. Ueber eine aus marklosen Fasern hervorgehende zweite Art von hypolemmalen Nervenendplatten bei den quergestreiften Muskelfasem der Vertebraten. Anat. Anz., Bd. 35, S. 481.

. 1913. Die doppelte (motorische und sympathische) efferente Innervation der

quergestreiften Muskelfasem. Anat. Anz., Bd. 44, S. 343.

. 1930. Some remarks on the papers of H. J. Wilkinson on the innervation of

striped muscle fiber. J. Comp. Neurol., vol. 51, p. 299.

Boeke, J., and Dusser de Barenne, J. G. 1919. The sympathetic innervation of the cross striated muscle fibers of vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 21, pt. 2, p. 1227.

DE Boer, S. 1913. Die quergestreiften Muskeln erhalten ihre tonische Innervation mittels der Verbindungsaste des Sympathicus (thoracales autonomes system). Folia neuro-biol., Bd. 7, S. 378.

. 1913a. Ueber den Skelettmuskeltonus, 2** Mitteilung. Die tonische Innervation der quergestreiften Muskeln bei Warmbliitem. Folia neuro-biol., Bd. 7, S. 837.

. 1914. Die autonome tonische Innervation der Skelettmuskulatur. Erwiderung

auf die Kritik Beritoffs. Folia neuro-biol., Bd. 8, S. 429.

. 1915. Ueber die autonome tonische Innervation der Skelettmuskulatur.

Erwiderung zur Berichtigung und Erganzung der Kritik Beritoffs, Folia neurobiol., Bd. 9, S. 119.

. 1918. Ueber die Form und den Bau der Dermatome und ihre Verbreitung fiber

den tierischen Korper. Psychiat. en neurol. Bl., Amsterdam, Bd. 22, S. 94.

Bok, S. T. 1922. Die Entwicklung von Reflexen und Reflexbahnen. II. Die Ontogenese des Rfickenmarkreflexapparates mit den zentralen Verhaltnissen des Nervus sympathicus. Inaugural Dissertation, Amsterdam. Psychiat. en neurol. Bl., Amsterdam, Bd. 26, S. 174.

. 1928. Das Rfickenmark. G. von Mollendorf’s Handbuch der mikroskopische

Anatomie des Menschen. Nervensystem, Bd. 4/1, S. 478. J. Springer, Berlin.

Bolk, L. 1894. Beziehungen zwischen Skelet, Muskulatur und Nerven der Extremitaten dargelegt am Beckengfirtel, an dessen Muskulatur sowie am Plexus lumbosacralis. Morphol. Jahrb., Bd. 21, S. 241.

. 1895. Rekonstruktion der Segmentirung der Gliedmassenmuskulatur, dargelegt

an den Muskeln des Oberschenkels und des Schultergfirtels. Morphol. Jahrb., Bd. 22, S. 357.

. 1897. Een en ander uit de segmentaal-anatomie van het menschelijk lichaam.

Nederl. tijdschr. v. geneesk., vol. 33, part 2, p. 365.

. 1898. Die Segmentaldifferenzierung des menschlichen Rumpfes und seiner

Extremitaten. I. Einleitendes. Historische Uebersicht. Das erste bis vierte Segment. Morphol. Jahrb., Bd. 25, S. 465.

. 1898a. Idem. II. Die Segmente der oberen Extremitat. Erster Theil.

Morphol. Jahrb., Bd. 26, S. 91.

. 1899. Idem. Beitrage zur Anatomie und Morphogenese des menschlichen

Korpers. III. Die Segmente der oberen Extremitat. Zweiter Theil. Die Sklerozonie des Skelettes der oberen Extremitat. Morphol. Jahrb., Bd. 27, S. 630.

. 1899a. Idem. IV. Die Segmente der oberen Extremitat. Dritter Theil.

Beziehungen zwischen Metamerie und Muskeltopographie. Morphol. Jahrb., Bd. 28, S. 105.

. 1906. Ueber die Neuromerie des embryonalen menschlichen Rfickenmarkes.

Anat. Anz., Bd. 28, S. 204.

. 1910. De segmentale innervatie van romp en ledematen bij den mensch.

E. F. Bohn, Haarlem.

Borchert, M. 1902. Experimentelle Untersuchungen an den Hinterstrangen des Rfickenmarks. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 389.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 313

Bottazzi, P. 1897. The oscillations of the auricular tonus in the batrachian heart, with a theory on the function of sarcoplasma in muscular tissues. J. Physiol., vol. 21, p. 1.

Braeutigam, H. 1892. Vergleichend-anatomische Untersuchungen iiber den Conus medullaris. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 1, S. 111.

Bregmann, L. E. 1915. Neue Untersuchungen zur Kenntnis der Pyramidenbahn.

I. Der Anted der Pyramide am Riickenmarksquerschnitt bei verschiedenen Tieren und seine Entwickelung beim Menschen. Anat. Anz., Bd. 48, S. 75.

VAN DER Broek, A. J. P. 1907. Untersuchungen iiber den Bau des sympathischen Nervensystems der Saugetiere. I. Ted. Der Halssympathicus. Morphol. Jahrb., Bd. 37, S. 202.

. 1908. Idem. II. Der Rumpf- und Beckensympathicus. Morphol. Jahrb., Bd.

38, S. 532.

Brouwer, B. 1912. Die Bedeutung der Bulbarlasion bei Syringomyelie fur die sensiblen Ausfallerscheinungen. Monatschr. f. Psychiat. u. Neurol., Bd. 32, S. 301.

. 1914. Over Dermatomerie. Bijdrage tot de kennis der segmentaalanatomie

en der sensibele geleiding bij den mensch. Nederl. tijdschr. v. geneesk., vol. 58, part 2, p. 657.

. 1915. Ueber Querlasion des Riickenmarks beim Menschen und das Bastiaansche

Gesetz. Psychiat. en neurol. BL, Amsterdam, Bd. 19, S. 377.

. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der

Segmentalanatomie und der Sensibilitatsleitung im Riickenmark und in der Medulla oblongata. Folia neuro-biol, Bd. 9, S. 225.

. 1916. Over een verlamming van Brown-Sequard. Nederl. tijdschr. v. geneesk.,

vol. 60, part 2, p. 1979.

. 1917. Ueber den Verlauf der Beinfasem in der Pyramidenbahn. Psychiat. en

neurol. BL, Bd. 21, S. 99.

Brouwer, B., and Blauwkuip, H. J. J. 1915. Ueber das Zentralnervensystem bei pemizioser Anamie. Monatschr. f. Psychiat. u. Neurol., Bd. 38, S. 287.

Brown, G. E., and Adson, A. W. 1930. Physiologic effects of thoracic and lumbar ganglionectomy or trunk section. Pub. for Research in Nervous and mental disease, vol. 9, p. 721.

Brown-S^quard. 1855. Experimental and clinical researches on the physiology and pathology of the spinal cord. Richmond. Earlier papers are to be found in Compt. rend, hebdom. d. seances d. I’acad. d. sc., 1847 and the Gas. med. de Paris for 1850. (Quoted from Page May ’09.)

Bruce, A. 1901. A topographical atlas of the spinal cord. Williams and Norgate, London.

. 1906. Distribution of the cells in the intermedio-lateral tract of the spinal cord.

Tr. Roy. Soc., Edinburgh, vol. 45, p. 105.

VON Brucke, E. T. 1918. Neuere Anschauungen iiber den Muskeltonus. Deutsche med. Wochenschr., Bd. 44°, S. 121 und S. 152.

Buhler, a. 1898. Untersuchungen iiber den Bau der Nervenzellen. Wurzburger med. Zeitschr., Bd. 31, S. 285.

Burdach, K. Fr. 1819-26. Vom Baue und Leben des Gehims. Dyk, Leipzig. 2Bks.

de Buruet, H. M. 1911. Het aderlijke stelsel der Bradypodidae. Handl. XIII^' Ned. Nat. en Gen. Congres, Groningen. Tweede Sectie, p. 268.

Carpenter, F. W., and Conel, J. L. 1914. A study of ganglion cells in the sympathetic neri'ous system, with special reference to intrinsic sensory neurones.

J. Comp. Neurol., vol. 24, p. 269.

Cassirer, R. 1912. Die vasomotorisch-tropischen Neurosen. S. Karger, Berlin.

. 1923. Erkrankungen der Sympathicus. Lehrbuch der Nervenkrankheiten by

H. Oppenheim. 7**' edition, S. Karger, Berlin.

314 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Cattell, J. McK., and Stiles, P. G. 1924. The place of fatigue in striated muscle : evidence for failure of excitation apart from contraction. Am. J. Physiol., vol. 49, p. 645.

Clarke, J. L. 1868. Researches on the intimate structure of the brain. Phil. Tr. Roy. Soc., London, vol. 158, p. 263.

Claude, H., and Chauvet, S. 1911. S6meiologie r^elle des sections totales des nerfs mixtes peripheriques. A, Maloine, Paris.

Cobb, S. 1918. A note on the supposed relation of the sympathetic nerves to decerebrate rigidity, muscle tone, and tendon reflexes. Am. J. Physiol., vol. 46, p. 478.

. 1925. Review on the tonus of skeletal muscle. Physiol. Rev., vol. 5, p. 518.

Cobb, W., and Fulton, J. F. 1925. Quotation from tonus of skeletal muscle. Physiol. Rev., vol. 5, p. 520.

Collier, J., and Buzzard, E. F. 1903. Degenerations resulting from lesions of posterior nerve roots and from transverse lesions of the spinal cord in man. Brain, vol. 26, p. 559.

Collins, J. 1894. A contribution to the arrangement and functions of the cells of the cervical spinal cord. New York Med. J., vol. 59, pp. 40 and 98.

Coman, F. D. 1926. Observations on the relation of the sympathetic nervous system to skeletal muscle tonus. Bull. Johns Hopkins Hospital, vol. 38, p. 163.

Craw, C. H. 1928. The distribution of the nerve cells in the ventral columns of the spinal cord. J. Comp. Neurol., vol. 45, p. 283.

Crothers, B. 1925. Cases treated by S 3 unpathetic ramisection by Dr. N. D. Royle, Arch. Neurol, and Psychiat., vol. 13, p. 638.

CuNfio, B. 1911. Maladies des nerfs. Nouveau traits de chirurgie. (Le Dentu et P. Delbet), J. B. Bailli^re, Paris.

Cunningham, D. J. 1877. The spinal nervous system of the porpoise and dolphin. J. Anat., vol. 11, p. 209.

Curtis, A. H., and Helmholz, H. F. 1911. A study of the anterior horn cells of an abrachius and their relations to the development of the extremities. J. Comp. Neurol., vol. 21, p. 323.

Dana, C. L. 1906. The question of protopathic and epicritic sensibility and the distribution of the trigeminus nerve (third branch). J. Nerv. and Ment. Dis., vol. 33, p. 577.

Davida, L. 1880. Ueber die Multiplicitat der Lumbal- und Sacral-Spinalganglien. Centralbl. f. d. med. Wissensch., Bd. 18, S. 465.

Davis, L. E. 1922. Pathway for visceral afferent impulses within the spinal cord. Am. J. Physiol., vol. 59, p. 381.

Davis, L., Hart, J. T., and Crain, R. C. 1929. The pathway for visceral afferent impulses within the spinal cord. II. Experimental dilatation of the biliary ducts. Surg., Gynec. and Obst., vol. 48, p. 647.

Dejerine, j. 1895, 1901. Anatomic des centres nerveux. Rueff et Cie, Paris.

. 1914. Sdmiologie des affections du systSme nerveux. Masson et Cie, Paris.

Dejerine, J., and Sottas, J. 1895. Sur la distribution des fibres endogtoes dans le cordon postdrieur de la moelle et sur la constitution du cordon de Goll. Compt. rend. Soc. de biol., Ser. 10, vol. 2, p. 465.

Dejerine, J., and Spiller, W. G. 1895. Contribution k I’dtude de la textura des cordons postdrieurs de la moelle 4pini^re; du trajet intramedullaire des racines postdrieures sacrdes et lombaires infdrieures. Compt. rend. Soc. de biol., Ser. 10, vol. 2, p. 622.

Dejerin-e, j., and Tho.mas, a. 1896. Contribution A I’dtude du trajet intra-mddullaire des racines postdrieures dans la rdgion cervicale et dorsale supdrieure de la moelle dplnidre. Compt. rend. Soc. de biol., Ser. 10, vol. 3, p. 675.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 315

Dendy, a. 1912. Reissner’s 6ber and the subcommissural organ in the vertebrate brain. Proc. British A. Adv. Sc., 1912.

Dendy, A., and Nicholes, G. E. 1910. On the occurrence of a mesocoelic recess in the human brain and its relation to the sub-commissural organ of lower vertebrates; with special reference to the distribution of Reissner’s fibre in the vertebrate series and its possible function. Proc. Roy. Soc., London, Ser B vol. 82, p. 515. ■’

Dexler, H. 1907. Zur Anatomic des Zentralnervensystems von Elephas indicus.

Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 15, S. 137.

Dexler, H., and Eger, 0. 1911. Beitrage zur Anatomic des Saugerriickenmarkes.

I. Halicore dugong Erxl. Morphol. Jahrb., Bd. 43, S. 107.

Dogiel, A. S. 1895. Zur Frage fiber den feineren Bau des sympathischen Nervensystems bei den Saugethieren. Arch. f. mikr. Anat., Bd. 46, S. 305.

. 1896. Zwei Arten sympathischer Nervenzellen. Anat. Anz., Bd. 11, S. 679.

. 1897. Zur Frage fiber den feineren Bau der Spinalganglien und deren Zellen

bei Saugetieren. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 14, S. 73.

. 1908. Der Bau der Spinalganglien des Menschen und der Saugetiere.

G. Fischer, Jena.

Donaldson, H. H. 1895. The growth of the brain ; a study of the nervous system in relation to education. The Contemporary Science Series, London.

. 1908. A comparison of the albino rat with man in respect to the growth of the

brain and of the spinal cord. J. Comp. Neurol., vol. 18, p. 345.

. 1909. On the relation of the body length to the body weight and to the weight

of the brain and of the spinal cord in the albino rat (Mus non'egicus var. albus).

J. Comp. Neurol., vol. 19, p. 155.

. 1910. On the percentage of water in the brain and in the spinal cord of the

albino rat. J. Comp. Neurol., vol. 20, p. 119.

. 1911. On the influence of exercise on the weight of the central nervous system

of the albino rat. J. Comp. Neurol., vol. 21, p. 129.

. 1911a. The effect of underfeeding on the percentage of water, on the ether alcohol extract and on medullation in the central nervous system of the albino rat. J. Comp. Neurol., vol. 21, p. 139.

• . 1911b. An interpretation of some differences in the percentage of water found

in the central nervous system of the albino rat and due to conditions other than age.

J. Comp. Neurol., vol. 21, p. 161.

. 1912. A comparison of the European, Norway, and albino rats (Mus norvegicus

and Mus norvegicus albinus) with those of North America in respect to the weight of the central nervous system and to cranial capacity. J. Comp. Neurol., vol. 22,

p. 71.

-. 1915. The rat. Reference tables and data for the albino rat (Mus norvegicus albinus) and the Norway rat (Mus norvegicus). Mem. Wistar Inst. Anat. and

Biol., no. 6. , , J . , . , J

1916. A revision of the percentage of water in the brain and m the spinal cord

of the albino rat. J. Comp. Neurol., vol. 27, p. 77.

Donaldson H. H., and Davis, D. J. 1903. A description of charts showing the areas of the cross sections of the spinal cord at the level of each spinal nerve. J . Comp.

Neurol., vol. 13, p. 19. , . n c i xi,

Donaldson, H. H., and Hatai, S. 1911. Note on the influence of castration on the

weight of the brain and spinal cord in the albino rat and on the percentage of water

in them. J. Comp. Neurol., vol. 21, p. 155. . , . x . , j

1911 A comparison of the Norway rat with the albino rat in respect to body

length, brain weight, spinal cord weight, and the percentage of water in both the brain and the spinal cord. J. Comp. Neurol., vol. 21, p. 417.

316 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Donaldson', H. H., Hatai, S., and King, H. D. 1915. Postnatal Rrowth of the brain under several e.vperimental conditions. Studies on the albino rat. J. Ncrv. and Ment. Dis., vol. 42, p. 797.

Drabeke, J. 1903. Zur mikroskopisclicn Kenntnis dcr Pyramidenkrcuzunp der Chiropteren. Anat. Anz., Bd. 23, S. 449.

. 1903. Ueber einen bisher nicht beobachtetcn Ncia-enkem (Hofmann-Kollikcr)

im Rtickenmark von Chiropteren. Anat. Anz., Bd. 23, S. 571.

. 1904. Zur Kenntnis des Ruckcnmarks und der Pyramidenbahncn von Talpa

europea. Monatschr. f. Psychiat. u. Neurol., Bd. 15, S. 401.

Dufour, H. 1896. Sur Ic groupement des fibres endog6ncs dc la moellc dans Ics cordons post^rieurs. Arch, de Neurol., vol. 2, p. 81.

Dubser de Baren'Ne, j. G. 1910. Die Strycbninwirkung auf das Zcntralncrvcnsj'stcni. I. Die Segmentare, strong lokalisicrtc Strychninvcrgiftung der dorsalcn Riickcnmarksmechanismen ; cin Beitrag zur Dcrmatomerie der hintcrcn E.\treniitfit des Hundcs. Zcntralbl. f. Physiol., Bd. 24, S. 1100.

. 1911. Idem. II. Zur Wirkung des Strychnins bei loknler Applikation auf das

Rtickenmark. Folia neuro-biol., Bd. 5, S. 42.

. 1911. Idem. III. Die segmentare Strychninvcrgiftung der dorsalcn Riicken marksmechanismen ; cin Beitrag zur Dcrmatomerie der hintcrcn Extremitiit des Hundes. Folia ncuro-biol., Bd. 5, S. 342.

. 1912. Idem. IV. Theorctische Betrachtungen. Folia ncuro-biol., Bd. 6,

S. 277.

. 1913. Idem. V. Mittcilung. Folia ncuro-biol., Bd. 7, S. 549.

. 1913a. Zur Kenntnis dcr Alloasthcsic. Monatschr. f. Psychiat. u. Neurol., Bd.

34, S. 523.

. 1913b. Ueber die Enthirnungsstarre (Decerebrate rigidity Sherrington’s) in

ihrer Bozichung zur cffcrcntcn Innervation dcr quergestreiften Muskulatur. Folia ncuro-biol., Bd. 7, S. G51.

. 1914. Ueber Allonosthcsic. Expcrimcntcller Beitrag zur Kenntnis dic.TS

Sj'mptoms. P.sychiat. on ncurol. Bl., Amsterdam, Bd. 18, S. ICO.

. 1917. Uclx;r die Innervation und den Tonus der quergestreiften Muskoln.

Arch. f. d. ge.s. Physiol. (Pfltiger’s), Bd. ICG, S. 145.

voN Dvdynski, Ij. 1903. Ein IBeitrag zum Studium des Vorlaufs ciniger Riiekcnmarksstriinge. Ncurol. Ccntralbl., Bd. 22, S. 898.

Edinger, L. 1882. Kiickenmark und Gchirn in cincm Fallc von angeboronem Mangel cinc.s Vorderanns. Virchow’s Arch. f. path. Anat., Bd. 89, S. 40.

. 1885. Zur Kenntnis dcr Vorlnufes dcr Ilinterstrangfasern in der Medulla

oblongata und im unteren Kleinhomschcnkcl. Ncurol. Ccntralbl., Bd. 3, S. 73.

. 1889. Vorgleichend-entwicklung.«gcschicbllieho und anatomisehe Studien ini

Bcreiche de.s Centralnervensystcms. Anal. Anz., Bd. 4, S. 121.

. 1889. Die Forlscfzung der hintcrcn Rllckcnrnnrkswurzeln zum Gehirn. Anat.

Anz., Bd. 4, S. 480.

. 1890. ICinige.s vom ^'erlnuf der Geftihlsbahnen im centnilen Nervensy.«lenie,

Deutsebe ined, Wocheiwehr., Bd. 10, S. 421.

. 1908. \'orie.“ungen 0l)er den Ban iler nen'o^en Ccntralorgane des Meii'-cben und

der Thiere. 7" Aull., Ersler Toil, F. C. W. Vogel, I/'ipzig. Aho H*’’ edition, 1911.

Et.nr.u.s, C;. 1910. Die motorisrhen Ontren und die Fonn cies Vorderhnrns in den

ftinf letzten Segnienteri de« C'ervikalinnrkes und detn ersfen Dor.-.'ihegtnenlc eines Manne«, der ohne liiiken Vorderann gelmren {•'t. Monatsrhr. f. IVvehiat. u, Neurol., Bd. 28, .8. 491.

. 1914. I'ely’r die dern Vordernnne und der Hand nigehririgen Bahnen ef'ter

Drtinung und die Bahnen zaeiter Ordnung einea .Mannea, dor ohne linken Vorderarm gelxiren i‘-t .Mon.at'-chr. f. P-ychial. u. Neurol., Bd, 3,';, .S. J77.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 317

Ewald, J. Rich. 1891. N. octavus. Weisbaden. (Quoted from Muskens ’30, see p. 510.)

Fabritids, H. 1907. Ueber die Gruppierung der motorischen Bahnen innerhalb der Pyramidenstrange beim Menschen. Arb. a. d. path. Inst. d. Univ. Helsingfors, Bd. 2, S. 199.

. 1907a. Studien iiber die sensible Leitung im menschlichen Riickenmark, auf

Grand klinischer und pathologisch-anatomischer Tatsachen. Arb. a. d. path. Inst. d. Univ. Helsingfors, Bd. 2, S. 91.

. 1910. Versuch einer Psychophysiologie des Gefiihles. Monatschr. f. Psychiat.

u. Neurol., Bd. 28, S. 400.

. 1912. Zur Frage nach der sensiblen Leitung im menschlichen Riickenmark.

Monatschr. f. Psychiat. u. Neurol., Bd. 31, S. 103, 279, 376, 463 und 546.

Fabritius, H., and von Bermann, E. 1913. Zur Kenntnis der Haut- und Tiefensensibilitat untersucht mittels der Abschniiningsmethode. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 151, S. 125.

Fedeli, F. 1915. Recherches histologiques sur la dure-mere. Arch. ital. de biol., vol. 63, p. 220.

Figueiredo-Rodrigdes, J. a. 1901. Das Riickenmark des Orang-Utan. Arch. f. mikr. Anat., Bd. 59, S. 417.

Fisher, C., and Ranson, S. W. 1933. On the so-called sympathetic cells in the spinal ganglia. Am. J. Anat., vol. 68, p. 1.

Flechsig, P. E. 1876. Die Leitungsbahnen im Gehim und Riickenmark des Menschen. W. Engelmann, Leipzig.

Forbes, R. D. 1930. Some recent surgical advances. Northwest Med., vol. 29, p. 199.

Forster, 0. 1917. Die Topik der Sensibilitatsstorangen bei Unterbrechung der sensiblen Leitungsbahnen. Deutsche Zeitschr. f. Nervenh., Bd. 56, S. 185.

. 1927. tlber die Vorderseitenstrangdurchschneidung. Arch. f. Psychiat., Bd. 81,

S. 707.

Franck, F., and Pitres, A. 1880. Des d6g6n6rations secondaires de la moelle dpiniire consficutives a I’ablation du gyrus slgmoide chez la chien. Gaz. med. de Paris, vol. 51, p. 152.

Fransen, j. W. P. 1907. Le systeme vasculaire abdominal et pelvien des primates. Anatomie descriptive et relations segmentales. Petrus Camper, Nederl. bijdr. t. de anat., vol. 4, p. 215.

Franz, S. I. 1910. Sensations following nerve division. Government hospital for the insane, Washington. Bull. no. 2, p. 33.

Frazier, C. H. 1920. Section of the anterolateral columns of the spinal cord for relief of pain ; a report of six cases. Arch. Neurol, and Psychiat., vol. 4, p. 137.

Frazier, C. H., and Spilder, W. G. 1923. Section of the anterolateral columns of the spinal cord (chordotomy). Discussion of the physiologic effects and clinical results in a series of eight cases. Arch. Neurol, and Psychiat., vol. 9, p. 1.

VON Frey, M. 1896. Untersuchungen iiber die Sinnesfunctionen der menschlichen Haut. Abhandl. d. math.-phys. Klasse d. kon. sachs. Gesellsch. d. Wissensch., Bd. 23, no. 3, S. 169.

. 1914. Neue Untersuchungen iiber die Sinnesleistungen der menschlichen Haut.

Fortschr. d. Psychol., Bd. 2, S. 207.

. 1914a. Beobachtungen an Hautflaehen mit geschadigter Innervation. Zeitschr.

f. Biol., Bd. 63 (N. F. Bd. 45), S. 335.

Gabri, G. 1896. A propos des cellules radiculaires post6rieures de v. Lenhoss6k et Ram6n y Cajal. Arch. ital. de biol., vol. 26, p. 115.

Gad, j. 1893. On the respiratory center in the medulla oblongata. Tr. 1'‘ Panamerican Med. Congress, Washington.

318 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Gad, J., and Mahinesco, G. 1892. Recherches experimentales sur le centre respiratoire bulbaire. Compt. rend, hebdom. d. seances d. I’acad. d. sc., vol. 115, p. 444.

Gaged, 0. 1928. Zur Histologie und Topographie der vegetativen Zentren im Riickenmark. Zeitschr. f. d. ges. Anat . ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 85, S. 213.

Gans, a. 1916, Die P 3 Tainidenbahn der Phocaena. Anat. Anz., Bd. 49, S. 281.

VON Gasa, W. 1924. Resection of paravertebral nerves and isolated severing of ramus communicans in treatment of vegetative neuroses of internal organs. Arch. f. klin. Chir., vol. 133, p. 479.

Gaskeld, W. H. 1916. The involuntary nervous system. Longmans, Green and Co., London and New York.

Gaule, J., and Lewin, T. 1896. Ueber die Zahlen der Nervenfasem und Ganglienzellen in den Spinalganglien des Kaninchen. Zentralbl. f. Physiol., Bd. 10, no. 15 u. 16, S. 437 u. 464.

Gegenbaub, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

VAN Gehuchten, a. 1891. La structure des eentres nerveux. La moelle 4pini&re et le cervelet. La Cellule, vol. 7, p. 81.

. 1901. Les voies ascendantes du cordon lateral de la moelle dpinito et leurs

rapports avec le faisceau rubrospinal. Le Nevr'axe, vol. 3, p. 157.

. 1903. La d4g4n4rescence dite retrograde ou degenerescence Walierienne in directe. Le Nevraxe, vol. 5, p. 1.

. 1920. Les maladies nerveuses. A. Dystpruyst, Louvain.

VAN Gehuchten, A., and de Neeff, C. 1900. Les noyaux moteurs de la moelle lombo-sacree chez I’homme. Le Ndvraxe, vol. 1, p. 201.

VAN Gelderen, C. 1924. Die Morphologie der Sinus durae matris. Zweiter Teil. Die vergleichende Ontogenie der neurokraniellen Venen der Vogel und Saugctiere. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 73, S. 541.

Glaser. 1924. See L. A. Muller’s Die Lebensnerven, p. 151. J. Springer, Berlin.

Goldscheider. 1911. Ueber Henry Head’s Lehre vom Temperatursinn der Haut. Med. Klin., Bd. 7 (February), S. 293.

Goldstein, K. 1904. Zur vergleichenden Anatomie der Pyramidenbahn. Anat. Anz., Bd. 24, S. 451.

. 1910. Ueber die aufsteigende Degeneration nach Querschnittsunterbrechung

des Riickenmarks (Tractus spino-cerebellaris posterior, tractus spino-olivaris, tractus spino-thalamicus). Neurol. Centralbl., Bd. 29, S. 898.

Golgi, C. 1883. Recherches sur I’histologie des centres nerveux. Arch. ital. de biol., vol. 3, p. 285, and vol. 4, p. 92.

. 1890. Ueber den feineren Bau des Ruckenmarkes. Anat. Anz., Bd. 5, S. 372 u.

423.

Gombault, a. and Philippe, 1894. Contribution I’dtude des Idsions systematisdcs dans les cordons Wanes de la moelle dpiniere. Arch, de med. exper. et d’anat, path., pp. 365 and 538.

Gowers, W. R. 1879. The diagnosis of diseases of the spinal cord. J. and A. Churchill, London.

Gray, L. P. 1926. Some experimental evidence on the connections of the vestibular mechanism in the cat. J. Comp. Neurol., vol. 41, p. 319.

Greving, R. 1928. F. Die zentralen Antcile des vegatativen Nervensystems. G. von Mollcndorf’s Handbuch der mikro.skopi.schen Anatomie des Menschen, Nerv'cnsystem, S. 917. J. Springer, Berlin.

GRtiTZNER. 1883. Quoted from Hines ’27.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 319

Guldberg. 1885. Ueber das Zentralnervensystem der Bartenwale. Videnskabs, Selskabs Forthandlinger, Christiana. ’

Hacker, F. 1913. Beobachtungen an einer Hautstelle mit dissoziertpr Empfindigungslahmung. Zeitschr. f. Biol., Bd. 61 (N. F., Bd. 43), S. 231.

Haggquist, G. 1914. Histophysiologische Studien iiber die Temperatursinne der Haut des Menschen. Anat. Anz., Bd. 45, S. 46,

. 1915. Studien iiber die Temperatursinne der Haut des Menschen. Handl.

Kungl. Svenska Vetenskapsakad.

Hardesty, I. 1902. Observations on the medulla spinalis of the elephant with some comparative studies of the intumescentia cervicalis and the neurones of the columns anterior. J. Comp. Neurol., vol. 12, p. 125.

. 1905. On the number and relations of the ganglion cells and medullated nerve

fibers in the spinal nerves of frogs of different ages. J. Comp. Neurol., vol. 15, p. 17.

Hatai, S. 1901. On the presence of the centrosomes in certain nerve cells of the white rat. J. Comp. Neurol., vol. 11, p. 25.

. 1902. Number and size of the spinal ganglion cells and dorsal root fibers in the

white rat at different ages. J. Comp. Neurol., vol. 12, p. 107.

. 1903. On the increase in the number of medullated nerve fibers in the ventral

roots of the spinal nerves of the growing white rat. J. Comp. Neurol., vol. 13, p. 177.

. 1904. The effect of partial starvation on the brain of the white rat. Am. J.

Physiol., vol. 12, p. 116.

. 1904a. A note on the significance of the form and contents of the nucleus in the

spinal ganglion cells of the foetal rat. J. Comp. Neurol., vol. 14, p. 27.

. 1907. Effect of partial starvation followed by a return to normal diet, on the

growth of the body and central nervous system of albino rats. Am. J. Physiol., vol. 18, p. 309.

. 1908. Preliminary note on the size and condition of the central nervous system

in albino rats experimentally stunted. J. Comp. Neurol., vol. 18, p. 151.

. 1909. Note on the formulas used for calculating the weight of the brain in the

albino rats. J. Comp. Neurol., vol. 19, p. 169.

. 1913. The effect of castration, spaying, or semispaying on the weight of the

central nervous system and of the hypophysis of the albino rat ; also the effect of semispaying on the remaining ovary. J. Exper. Zool., vol. 15, p. 296.

Hatschek, R. 1896. Ueber das Riickenmark des Delphins (Delphinus delphis). Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 4, S. 286.

. 1896a. Ueber das Ruckenmark des Seehundes (Phoca vitulina) im Vergleiche

mit dem Hundes. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 4,

S. 313.

Hay, j. 1901. Red and white muscle. Liverpool Med.-Chir. J., Sept., p. 431.

Head, H. 1905. Marshall Hall address.' Roy. Med. Chir. Soc., May 23. (Quoted from Page May ’06 ; see also Head ’20.)

. 1920. Studies in neurology. Oxford Medical Publications, Oxford Univ. Press.

Head, H., and Holmes, G. 1911. Sensory disturbances from cerebral lesions. Brain, vol. 34, p. 102.

Head, H., and Rivers, W. H. R. 1908. A human expenment in nerve division. Brain, vol. 31, p. 323.

Head, H., Rivers, W. H. R., and Shehren, J. 1905. The afferent nervous system from a new aspect. Brain, vol. 28, p. 99.

Head, H., and Sherren, J. 1905. The consequences of injury to the peripheral ner^res

in man. Brain, vol. 28, p. 116. . , . ,

Head, H., and Thompson, T. 1906. The grouping of afferent impulses within the

spinal cord. Brain, vol. 29, p. 537.

320 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Helweg. 1888. Ueber den centralen Verlauf der vasomotorischen Nervenbahnen. Arch. f. Psychiat., Bd. 19, S. 104.

Henneberg, B. 1901. Ueber den centralen Verlauf des Gowers’schen Biindels beim Menschen. Neurol. Centralbl., Bd. 20, S. 334.

Hepburn, D., and Waterston, D. 1904. A comparative study of the gray and white matter, of the motor-cell groups, and of the spinal accessory nerve, in the spinal cord of the porpoise. J. Anat., vol. 38, p. 105.

Herbst. 1900. Le sympathique cervical. Etude anatomique et chirurgicale. These, Paris.

Herrick, C. J. 1931. An introduction to neurology. 5*'* edition. W. B. Saunders Co., Philadelphia and London.

Hill, C. J. 1927. A contribution to our knowledge of the enteric plexuses. Phil. Tr. Roy. Soc., London, Ser. B, vol. 215, p. 355.

Hines, M. 1927. Nerve and muscle. Quart. Rec. Biol., vol. 2, no. 2, p. 149.

. 1930. Studies in the innervation of skeletal muscle. III. The extraocular

muscles in the rabbit. Anat. Rec., vol. 45, p. 220. For further references see bibliography of Chapt. I.

Hines, M., and Tower, S. 1928. Studies on the innervation of skeletal muscle. II. Of muscle spindles in certain muscles of the kitten. Bull. Johns Hopkins Hospital, vol. 42, p. 264.

Hinsey, j. C. 1928. Observations on the innervation of the blood vessels in skeletal muscle. J. Comp. Neurol., vol. 47, p. 23.

. 1930. Three types of fibers going to skeletal muscle and its blood vessels.

Proc. A. Research Nerv. and Ment. Dis., vol. 9, p. 153.

Hirose, K. 1916. Ueber eine bulbospinale Bahn. Folia neuro-biol., Bd. 10, S. 371.

Hirt, a. 1928. Ueber den Aufbau des Spinalganglions und seine Beziehungen zum Sympathicus. Zeitschr. ges. Anat. I. Abt. Zeitschr. f. Anat. u. Entwicklungsgeschichte, vol. 87, S. 275.

His, W. 1897. Ueber die Entwickelung des Bauchsympathicus beim Huhnchen und Menschen. Arch. f. Anat. u. Physiol., Anat. Abt., S. 137.

Hitzkek, H. 1914. Ueber den Einfluss der Nervenleitungen auf das mikroskopische Bild der Glandula submaxillaris des Hundes. Arch. f. ges. Physiol., Bd. 159, S. 487.

Hochstetter, F. 1897. Ein Beitrag zur vergleichenden Anatomic des Venensystems der Edentaten. Morphol. Jahrb., Bd. 25, S. 362.

Hodge, C. F. 1889. Some effects of electrically stimulating ganglion cells. Am. J. Physiol., vol. 2, p. 376.

Holmdahl, D. E. 1932. Die Lage der sakralen und kaudalen Spinalganglien. Anat. Anz., vol. 68, p. 242.

Holmes, G., and May, W. P. 1909. On the exact origin of the pyramidal tracts in man and other mammals. Brain, vol. 32, p. 1.

Horsley, V. 1908. Note on the existence of Reissner’s fibre in higher vertebrates. Brain, vol. 31, p. 147.

Hoskins, E. R. 1914. On the vascularization of the spinal cord of the pig. Anat. Rec., vol. 8, p. 371.

Hovy, a. j. 1913. On the relation between the quantity of white and gray substance in the central nervous system. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 16', p. 311.

Huber, G. Carl. 1899. A contribution on the minute anatomy of the sympathetic ganglia of different classes of vertebrates. J. Morphol., vol. 16, p. 27.

. 1899a. A brief account of the minute anatomy of the sympathetic nervous

system and its relation to the cerebrospinal nervous system. Tr. Mich. State Med. Soc., vol. 23, p. 398. May, 1899. See also J. Morphol., vol. 16.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 321

Huder, G. Carl. lS99b. Observations on the innervation of the intracranial vessels. J. Comp. Neurol., vol. 9, p. 1.

. 1903. Studies on neuroglia tissue. Contributions to Medical Research, Univ.

of Michigan, Dedicated to Victor Clarence Vaughan, p. 578. George Wahr, Ann Arbor.

. 1913. The morphology of the sympathetic nervous system. XVIP’’ International Congre.'JS of Medicine, London.

. 1928. New method of fixation and staining of the central nervous system for

purpo.se of stud}' of cytoarchiteeture. Contributions to Medical Research, Univ. of ^Michigan, Dedicated to Aldrcd Scott Warthin, p. 1, George Wahr, Ann Arbor.

Hucer, G. Carl, and Crosbv, E. C. 1929. Somatic and Visceral Connections of the Dicnccphalon. Arch. Neur. and Psych. Vol. 22, p. 187.

. 1930. See Picnsol, G. A., Human Anatomy.

Huber, G. Carl, and Guild, S. R. 1913. Observations on the histogenesis of protoplasmic processes and of collaterals, terminating in end bulbs, of the neurones of peripheral scnsor 3 ' ganglia. Anal. Rec., vol. 7, no. 10, p. 253.

Hunter, J. I. 1924. The influence of the sympathetic nervous system in the genesis of the rigidity of striated muscle in spastic paralysis. Surg., Gynec. and Obst., vol. 39, p. 721.

. 1925. The sympathetic innervation of striated muscle. I. The dual innervation of striated muscle. Brit. Med. J., vol. 1, p. 192. II. The functions of the two groups of muscle fibers. Ibid., pp. 251 and 298. III. The practical applications. Ibid., pp. 350 and 398.

Hu'nter, j. L, and Latham, 0. 1925. A contribution to the discussion of the histological conception of a somatic and sjunpathetic innervation of voluntary muscle. Australian Med. J. (Jan. 10), p. 27.

Ide, K. 1929. On the postnatal growth in area in the median and sciatic nerves of the albino rat — according to sex. Fixation in Bouin’s solution. J. Comp. Neurol., vol. 48, p. 355.

Ingbert, C. E. 1903. An cnvimeration of the medullated nerve fibers in the dorsal roots of the spinal nerves of man. J. Comp. Neurol., vol. 13, p. 53.

. 1903. On the density of the cutaneous innen'ation in man. J. Comp. Neurol.,

vol. 13, p. 209.

. 1904. An enumeration of the medullated nerve fibers in the ventral roots of the

spinal nerves of man. J. Comp. Neurol., vol. 13, p. 209.

Ingv.ar, S. 1919. Zur Phylo- und Ontogenese des Kleinhims. Folia neuro-biol., Bd. 11, S. 205.

Jacobsohn, L. 1908, Ueber die Kerne des menschlichen Riickenmarks. Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin, p. 389.

Jaquet, M. 1901. Anatoraie comparde du systeme nerveux sympathique cervical dans la sdrie des vertdbrds. Bull. Soc. d. Sc. de Bucarest, vol. 10, p. 240.

Johnson, S. E. 1921. An experimental study of the sacral sympathetic trunk of the cat with special reference to the occurrence of intrinsic commissural neurons. J. Comp. Neurol., vol. 33, p. 85.

. 1925. Experimental degeneration of the extrinsic nerves of the small intestine in

relation to the structure of the myenteric plexus. J. Comp. Neurol., vol. 38, p. 299.

Johnson, S. E., and Mason, M. L. 1821. The first thoracic white ramus eommunicans in man. J. Comp. Neurol., vol. 33, p. 77.

Johnson, S. E., and Palmer, M. 1931. Further observations on the components of the myenteric plexus. J. Comp. Neurol., vol. 53, p. 169.

JuTCHENKo, A. J. 1898. Sur les rapports du ganglion sympathique mesenterique inferieur avec Pinnervation de la vessie. Arch. Soc. biol. de St. Petersbourg , vol. 6, p. 53G.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 323

Kohnstamn, 0. 1900. Ueber die Koordinationskerne des Hirnstammes und die absteigende Spinalbahn. Nach den Ergebnissen der kombinierten Degenerationsmethode. Monatschr. f. Psychiat. u. Neurol., Bd. 8, S. 261.

. 1900a. Ueber die gekreuzt aufsteigcnde Spinalbahn und ihre Beziehung zum

Gowersschcn Strang. Neurol. Centralbl., Bd. 19, S. 242.

. 1902. Zur anatomischen Grundlegung der Kleinhirn-Physiologie. Arch. f. d.

ges. Ph 3 ’-siol., Bd. 89, S. 243.

VON KoLLiKEn, A. 1889-02. HandbuchdcrGewebelehredesMenschen., Bd. 1. Allgemeine Gcwebelehre und die Systeme der Haut, Kjiochen und Muskeln. W. Engelmann, Leipzig.

. 1890. Ueber den feineren Bau des Riickenmarks. Sitzungsb., d. phys.-med.

Gesellsch. zu Wurzburg, S. 44.

Kopsch, F. 1897. Das Riickenmark von Elephas indicus. Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin.

Kosaka, K., and Yagita, K. 1905. Experimentelle Untersuchungen iiber den Ursprung des N. vagus und die centralen Endigung der dem Plexus nodosus entstammenden sensibilen Vagusfasern, sowie iiber den Verlauf ihrer secundaren Bahn. Shinkcigaku Zasshi, Tokyo, Bd. 4, S. 29. (Quoted from Allen, ’27.)

Kotzenberg, W. 1899. Untersuchungen iiber das Riickenmark des Igels. J. F. Bergmann, Wiesbaden.

Kraxieu, F. 1910. Periphere Sensibilitiitsstorungen. Handbuch der Neurologic, M. Lewandow’sky, Bd. 1, part 2, S. 559. J. Springer, Berlin.

Krediet. 1910. Ueber die sympathischen Nerven in der Bauch- und Beckenhohle des Pferdcs, der Wiederkauer (insbesondcre der Ziege) und des Hundes. Inaugural Dissertation, Bern.

KijHNE, W. 1865. Ueber den Farbstoff der Muskeln. Virchow’s Arch. f. path. Anat., Bd. 33, S. 79.

Kuiper. 1890. Over het gevoel van koude en warmte. Dissertatie. Amsterdam.

Kulchitsky, N. 1924. Nerve endings in muscles. J. Anat., vol. 58, p. 152.

. 1924-25. Nerve endings in the muscles of the frog. J. Anat., vol. 59, p. 1.

Kunitomo, K. 1918. The development and reduction of the tail and of the caudal end of the spinal cord. Contrib. Embryol., Publ. Carnegie Inst., Washington, vol. 8, no. 26, p. 163. (Publ. no. 271, Carnegie Inst.)

Kuntz, a. 1910. The development of the sympathetic nervous system in mammals. J. Comp. Neurol., vol. 20, p. 211.

. 1913. The development of the cranial sympathetic ganglia in the pig. Brain,

vol. 23, p. 71.

. 1920. The development of the sympathetic nervous system in man. J. Comp.

Neurol., vol. 32, p. 173.

. 1927. On the occurrence of sympathetic nerve fibers in muscles of the extremities

following experimental degeneration of the spinal nerves. J. Comp. Neurol., vol. 43, p. 357.

. 1929. The autonomic nervous system. Lea and Febiger, Philadelphia.

Kuntz, A., and Farnsworth, D. I. 1928. Peripheral distribution of myelinated nerve fibers through communicating rami in dog. Proc. Soc. Exp. Biol, and Med., vol. 25,

p. 808.

Kuntz, A., and Kerper, A. H. 1924. Experimental observations on the functional significance of the sympathetic innervation of voluntary muscles. Proc. Soc. Exper. Biol, and Med., vol. 22, p. 25.

. 1924a. The sympathetic innervation of voluntary muscle. Proc. Soc. Exper.

Biol, and Med., vol. 22, p. 23.

. 1925. The sympathetic innervation of voluntary muscles and its effect on their

contractile power and resistance to fatigue. Anat. Rec., vol. 29, p. 366.

324 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Kuntz, a., and Keeper, A. H. 1925a. Posture of the extremities as related to the sympathetic nervous system. Anat. Rec., vol. 29, p. 366.

. 1926. Qualitative experimental data on the sympathetic innervation and tonus

of the triceps brachii muscles. Proc. Soc. Exper. Biol, and Med., vol. 23, p. 367.

. 1926a. An experimental study of tonus in skeletal muscles as related to the

sympathetic nervous system. Am. J. Physiol., vol. 76, p. 121.

Kur 6, K. (with collaborators). 1927-28. See also 1930, 1931. On the spinal parasympathetic system. Quart. J. Exp. Physiol., vol. 18, p. 333, 1928 ; vol. 20, p. 51, 1930, vol. 21, p. 1, and p. 103, 1931 ; Z. Zellforsch. u. Mikr. Anat., vol. 22, 1935.

Kue6, K., Shinosaki, T., and Shinagawa, F. 1925. Die morphologische Grundlage fiir die doppelte (cerebrospinale and autonome) Innervation des quergestreiften Muskels. Zeitschr. f. d. ges. exper. Med., Bd. 44, S. 144.

Laignel-Lavastine, M. 1903. Le plexus solaire. Steinheil, Paris.

DE Lange, S. J. 1907. Opstijgende degeneratie na gedieltetijke doorsnijding van het ruggemerg. Verslag. kon. Akad. v. Wetensch., te Amsterdam, vol. 161, p. 350.

Langelaan, j. W. 1915. On muscle tonus. Brain, vol. 38, p. 233.

. 1922. On muscle tonus, de Boer’s experiment on the frog. Brain, vol. 45, p. 434.

. 1925. Le tonus musculaire et le reflexe tendineux. L’enc5phale, vol. 20, p. 629.

Langley, J. N. 1890. On the physiology of the salivary secretion. J. Physiol., vol. 11, p. 123.

. 1891. On the course and connections of the secretory fibres supplying the sweat

glands of the feet of the cat. J. Physiol., vol. 12, p. 347.

. 1891a. Note on the connection with nerve-cells of the vasomotor nerves for the

feet. J. Physiol., vol. 12, p. 375.

. 1892. The origin from the spinal cord of the cervical and upper thoracic sympathetic fibres, with some observations on white and gray rami communicantes. Phil. Tr. Roy. Soc., London, Ser. B., vol. 183, p. 85. '

. 1892a. On the larger medullated fibres of the sympathetic system. J. Physiol.,

vol. 13, p. 786.

. 1894. The arrangement of the sympathetic nervous system, based chiefly on

observations upon pilo-motor nerves. J. Physiol., vol. 15, p. 176.

. 1895. A short account of the sympathetic system. Physiological Congress,

Berne.

. 1896. Observations on the medullated fibres of the sympathetic system and

chiefly on those of the gray rami communicantes. J. Physiol., vol. 20, p. 55.

. 1898. The S 3 rmpathetic and other related system of nerves. Textbook of

Physiology edited by E. A. Schafer. Vol. 2. Y. J. Pentland, London.

. 1903. Die sympathischen und verwandten nervosen Systeme der Wirbeltiere

(anatomes nervoses System). Ergeb. d. Physiol. (Asher u. Spiro), Bd. 2, Abt. 2, S. 818.

. 1904. On the question of commissural fibres between nerve-cells having the

same function and situated in the same sympathetic ganglion, and on the function of postganglionic nerve plexuses. J. Physiol., vol. 31, p. 244.

Langley, J. N., and Anderson, H. K. 1892. The action of nicotin on the ciliary ganglion and on the endings of the third cranial nerve. J. Physiol., vol. 13, p. 460.

. 1895. The innervation of the pelvic and adjoining viscera. Part II. The

bladder. Part III. The external generative organs. Part IV. The internal generative organs. Part V. Position of the nerve cells on the course of the efferent nerve fibers. J. Physiol., vol. 19, p. 71.

. 1904. The union of different kinds of nerve fibres. J. Physiol., vol. 30, p. 365.

Langley, J. N., and Magnus, R. 1905. Some observations of the movements of the intestine before and after degenerative section of the mesenteric nerves. J. Physiol., vol. 33, p. 34.

THE COMPARATHTL ANATOMY OP THE SPINAL CORD 325

Langley, J. N., and SnEnniNGTON, C. S. 1884. Secondary degeneration of nerve tracts following removal of the cortex of the cerebrum in the dog. J. Phvsiol vol. 5, p. 49. ’

Lankesteh, R. 1871. Ueber das Vorkommen von Haemoglobin in den Muskeln der ^lollusken und die Vcrbreitung desselben in den lebendigen Organismen. Arch. f. d. ges. Plo’siol. (P/luger’s), Ed. 4, S. 315.

. 1872. A contribution to the knowledge of haemoglobin. Proc. Roy. Soc.,

London, Scr. B., vol. 21, p. 70.

Lapinsky, M. 1903. Zur Frage der fur die einzelnen Segmenfe der Extremitiiten und der Muskelgruppen bestimmtcn Riickenmarkscentren beim Hunde. Arch. f. Anat. u. Phj-.siol., Physiol. Abt., Suppl.-Bd., S. 427.

Lapinsky, M., and Cassirer, R. 1901. Ueber den Ursprung des Halssympathicus im Riickcnmark. Deutsche Zeitschr. f. Nen'cnh., Bd. 19, S. 139.

L.\slett, E. E., .and Warrington, W. B. 1899. Observations on the ascending tracts of the spinal cord of the human subject. Brain, vol. 22, p. 586.

Lassek, a., Dowd, L., and Weil, A. 1930. The quantitative distribution of the pjTamidal tract in the dog. J. Comp. Neurol., vol. 51, p. 153.

Lawrentjew, B. j. 1929. Experimentcll-morphologische Studien uber den feineren Bau des autonomen Nen'ensystems. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 233.

. 1931. Zur Lchrc von dcr Qdoarchitectonik des peripherischen autonomen

Nervensj'stems. I. Die Cytoarchitektonik der Ganglien des Verdauungskanals beim Hunde. Jahrb. Morph, u. Mikrosk, Anat., Abt. 2, Zeitschr. Mikrosk.-Anat. Forschung, vol. 23, p. 527.

VON LENHOssfiK, M. 1889. Ueber die Pyramidenbahnen im Riickenmarke einiges Saugetierc. Anat. Anz., Bd. 4, S. 208.

• . 1889. Untersuchungen tiber die Entwicklung der Markscheiden und den

Fasereerlauf im Riickenmark der Maus. Arch. f. mikr. Anat., Bd. 33, S. 7l.

. 1895. Der feinere Bau des Nerv'ensystems im Lichte neuester Forschungen.

2** Aufl., Fischer, Berlin.

. 1898. Bemerkungen tiber den Bau der Spinalganglienzellen. Neurol. Centralbl.,

Bd. 17, S. 577.

Leriche, R. 1921. Spinal anaesthesia. Ann. Surg., vol. 81, p. 38.

LESZL&m, O. 1912. Studien tiber die Lissauersche Randzone des Hinterhoms. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 19, S. 252.

Levi, G. 1908. I gangll cerebrospinali. Studi di Istologia comparata e di Istogenesi. Arch. ital. di anat. e. di cmbriol., vol. 7 (Suppl.), p. 1. Referred to in Folia neurobiol., Bd. 6, S. 188, 1912.

Lewandowsky, M. 1904. Untersuchungen fiber die Leitungsbahnen des Truncus cerebri und ihren Zusammenhang mit denen der Medulla spinalis und des Cortex cerebri. G. Fischer, Jena.

Lhermitte, j., and Khaes, W. M. 1925. On the form of the anterior horn cells. Anat. Rec., vol. 31, p. 123.

Linowiecki, a. j. 1914. The comparative anatomy of the pyramidal tract. J. Comp. Neurol., vol. 24, p. 509.

Lissauer, H. 1886. Beitrag zum Faserverlauf im Hinterhom des menschlichen Rtickenmarks und zum Verhalten derselben bei Tabes dorsalis. Arch. f. Psychiat., Bd. 17, S. 377.

Ljubuschin, a. 1902. Zur Lehre von den endogenen Fasem in dem Vorderseitstrang des Rtickenmarks. Neurol. Centralbl., Bd. 21, S. 184.

VAN London, D. M. 1910. The extent of the sensory disturbances in tabes dorsalis, and its significance in the study of segmentation. Folia neuro-biol., Bd.

4, S. 6.

'326 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Lo^venthal, N. 1883. Ueber den Unterschied zwnschen der secundaren Degeneration des Seitenstrangs nach Him- und Ruckenmarksverletzungen. Arch. f. d. gcs. Physiol. (Pfluger’s), Bd. 31, S. 350.

. 1885. Degenerations secondaires de la moelle epinifire. Dissertation, Geneve.

. 1893. Neuer e.xperimentell-anatomischer Beitrag zur Kenntnis einiger Bahncn

im Gehim und Riickenmark. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 10, S. 168.

. 1911. Etude historique et critique sur quelques nouvelles systematisations dans

le cordon antero-lateral de la moelle epinidre. Rev. med. de la Suisse rom., vol. 31, p. 217.

. 1913. Les voies ascendantes du cordon antero-lateral de la moelle. Compt.

rend. A. de anat., 1'‘ reunion, Lausanne, p. 78.

Lugaro, E. 1906. Fibre aberranti, fibre centrifuge e fibre ricorrenti nolle radici posteriori. Monitore zool. ital., vol. 17, p. 217.

jMacNalty, a. S., ANn Horsley, V. 1909. On the cervical spino-bulbar and spinocerebellar tracts and on the question of topographical representation in the cerebellum. Brain, vol. 32, p. 237.

McCotter, R. E. 1916. Regarding the length and extent of the human medulla spinahs. Anat. Rec., vol. 10, p. 559.

Magnus, R. 1924. Korperstcllung. J. Springer, Berlin.

Magnus, R., ano de ICleijn, A. 1912. Die Abhangigkeit des Tonus der Extremitatenmuskeln von der Kopfstcllung. Arch. f. d. gcs. Physiol. (Pflilgcr’s), Bd. 145, S. 455.

Marburg, 0. 1902. Die absteigenden Hinterstrangsbahnen. Jahrb. f. Psychiat. u. Neurol., Bd. 22, S. 243.

. 1903. Zur Frage des “Anterolateral Tractes von Gowers.” Monatschr. f.

Psychiat. u. Neurol., Bd. 13, S. 486.

Marchi, V., AND Algeri, G. 1887. Sulle degenerazioni discendenti consecutive a lesioni della cortcccia cercbrale. Riv. sper. di freniat., vol. 12, p. 208.

Margulies, a. 1897. Experimen telle Untcrsuchungen iiber den Aufbau der Ilinterstrange beim Affcn. Monatschr. f. Psychiat. u. Neurol., Bd. 1, S. 277.

Marie, P., and Guillain, G. 1904. Les degenerations du cordon antcricur de la moelle. Pev. neurol., vol. 12, p. 776.

Marinesco, M. G. 1906. Quelques recherches sur la morphologie normale ct palhologique des cellules des ganglions spinaux et sympathiques de I’homme. Ia) Nevra.xe, vol. 8, fuse. 1, p. 7.

May, W. P. 1906. The afferent path. Brain, vol. 29, p. 742.

Mayer, C. 1895. Zur pathologischcn Anatomic der Rtickcnmarkshintcrstrangc. Jahrb. f. Psychiat. u. Neurol., Bd. 13, S. 57.

Meissner, G. 1853. Beitnige zur Anatomic und Physiologic der Haul. L. Voss, lAiipzig.

. 1853. Zur lAihrc vom Tast.sinn. lAiipzig. See nkso Zcitschr. f. rat. Med.

HeidcUicrg, N. F., Bd. 4, S. 260.

Meyer, E. 1875. Uclx;r rothc und blas-cc qucrgcstrciftc Mu.“kcln. Arch. f. Anal., Physiol, u. wisscnsch. Med., S. 217.

MERznACHEU, L., AND SpiELMEYER, W. 1903. Bcitriigc zur Kenntnis.s de.s fliederinnu^gehini.s, l)c.«ondcrs der corticomotori.schcn Bahnen. Neurol. Ccnfrnlhl., Bd. 22, S. 10,50.

Miller, M. M. 1913. Prenatal growth of the human spinal cord. J. Cornp. Neimil., vol. 23, p. 39.

Mingazzini, G. 1928. Medulla oblongata und Brticke. G. von Mfillendorf’s Handbuch der mikro'kopischc Anatomic des Men.'-chen. Nen'ensj-stem 4/1, S. 579.

J. Springer, Berlin.

THE COIMPAHATIVE ANATOMY OF THE SPINAL CORD 327’

VON Monakow, C. 1883. Experimenteller Beitrag zur Kenntnis des Corpus restiforme, des “ausseren Acusticuskemes” und deren Beziehungen zum Riickeiunark. Arch. f. Psychiat., Bd. 1, S. 104.

Morat, J. P. 1897. Troubles trophiques cons4cutifs a la section des racines post6rieures mddullaires. Gaz. d. h6p., vol. 70, p. 647.

MossOj A. 1904. Th(5orie de la tonicitd musculaire has6e sur Ja double miiBTvntioD des muscles stries. Arch. ital. de biol., vol. 41, p. 183.

Mott, F. W. 1888. Alicroscopical examination of Clarke’s column in man, the monkey, and the dog. J. Anat., vol. 22, p. 479.

. 1890. The bipolar cells of the spinal cord and their connections. Brain vol

13, p. 433. ’

. 1892. Ascending degenerations resulting from lesions of the spinal cord in

monkeys. Brain, vol. 15, p. 215. See also Phil. Tr. Roy. Soc., London, 1892.

Muller, L. A. 1924. Die Lebensnerven. J. Springer, Berlin.

Murray, W. S. 1929. Studies on developmental anomalies in the descendants of X-rayed mice. Papers Michigan Acad. Sc., Arts and Lett., vol. 10 (1928), p. 509. Published in 1929.

Nageotte, j. 1904. Contribution h I’dtude anatomique des cordons postdrieurs. Un cas de Idsion de la queue de cheval et un cas de tabes incipiens. Nouvelle iconographie de la salpetriere, vol. 17, p. 17.

NEEnHAM, D. M. 1926. Red and white muscle. Physiol. Rev., vol. 6, p. 1.

DE Neef, C. 1901. Rccherches experimentales sur les localisations motrices m^dullaires chez le chien et le lapin. Le Ndvraxe, vol. 2, p. 71.

Negrin t: Lopez, J., and von Brucke, E. T. 1916. Zur Frage nach der Bedeutung des Sympathicus ftir den Tonus der Skelettmuskulatur. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 166, S. 55.

Neumayer, L. 1906. Histo-und Morphogenese des peripheren Nervensystems der Spinalganglien und des Nenuis sympathicus. Handbuch der vergleichenden und experimentellen Entwicklungslehre, 0. Hertwig, Bd. 2, Teil 3, S. 513. G. Fischer, Jena.

Newton, F. C. 1924. Researches on the alleged influence of sympathetic innervation on warmth production in skeletal muscles. Am. J. Physiol., vol. 71, p. 1.

Ngowyang, G. 1930. On the growth of the motor cells, from birth to maturity, at four levels in the spinal cord of the albino mouse. J. Comp. Neurol., vol. 50, p. 231.

Nichols, H. N. T. 19ll. The occurrence in man of double ganglia upon the dorsal roots of the spinal nerves. Anat. Rec., vol. 5, p. 229.

Nieuwenhuijse, P. Die pathologisch-anatomische Untersuchung eines Falles von Herpes zoster. Zeitschr. f. d. ges. Neurol, u. Psychiat., Bd. 22, S. 45.

Obersteiner, H. 1900. Bemerkungen zur Helweg’schen Dreikantenbahn. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 7, S. 286.

. 1901. Ueber das Helweg’sche Biindel. Neurol. Centralbl., Bd. 20, S. 546.

. 1912. Anleitung beim Studium des Baues der nervosen Zentralorgane im

gesunden und kranken Zustande. 5" Aufl., F. Deuticke, Leipzig, u. Wien.

O’Donnell, J., and Windle, W. 1933. An experimental study of the sensory neurons in the ventral spinal roots of the cat. Anat. Rec., vol. 55, p. 117.

Onuf (Onufrowicz), B. 1900. On the arrangement and function of the sacral region of the spinal cord in man. Arch. Neurol, and Psychopath., vol. 3, no. 3, p. 387.

Onuf, B., and Collins, J. 1898. Experimental researches on the localisation of the sympathetic nerve in the spinal cord and brain, and contributions to its physiology.

J. Nerv. and Ment, Dis., vol. 25, p. 661.

. 1900. Experimental researches on the central localization of the sympathetic

with a critical review of its anatomy and physiology. Arch. Neurol, and Psychopath., vol. 3, p. 1.

328 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Oppenheim, H. 1913. Lehrbuch der Nervenkrankheiten. S. Karger, Berlin.

Papez, J. W. 1926. Reticulo-spinal tracts in the cat. Marchi method. J. Comp. Neurol., vol. 41, p. 365.

. 1929. Comparative neurology. Thomas Y. Crowell Co., New York.

Petr^;n, K. 1902. Ein Beitrag zur Frage vom Verlaufe der Bahnen der Hautsinne im Riickenmarke. Skandin. Arch. f. Physiol., Bd. 13, S. 9.

. 1910. Ueber die Bahnen der Sensibilitat im Riickenmarke, besonders nach

den Fallen von Stichverletzung studiert. Arch. f. Psychiat., Bd. 47, S. 495.

. 1911. Sur les voies de conduction de la sensibility dans la moelle ^piniere. Rev.

neurol., vol. 21, p. 548.

Piersol, G. A. 1930. Human anatomy. 9th ed., revised under supervision of G. Carl Huber. J. B. Lippincott Co., Philadelphia and London.

Pines, L. 1931. Allgemeine Ergebnisse unserer Untersuchungen iiber die Innervation der innersekretorischen Organe. Arch. f. Ges. Physiol. (Pfliiger’s), Bd. 228, S. 373.

PiRiE, J. H. H. 1908. The middle cells of the gray matter of the spinal cord. Proc. Roy. Soc., Edinburgh, vol. 28, p. 595.

PoLjAK, S. 1924. Die Struktureigentiimlichkeiten des Riickenmarkes bei den Chiropteren. Zugleich ein Beitrag zu den Frage iiber die spinalen Zentren des Sympathicus. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 74, S. 509.

. 1924a. Ueber die Intermediarzone im Riickenmark der Siiuger und ihr Ver haltniss zu dem Vegetativ-Nervensystem. Lijecnic Vjesnik.

Pollock, L. J. 1920. Nerve overlap as related to the relatively early return of pain sense following injury to the peripheral nerves. J. Comp. Neurol., vol. 32, p. 357.

. 1929. The pattern of sensory recovery in peripheral nerve lesions. Surg.,

Gynec. and Obst., vol. 49, p. 160.

Popper, E. 1904. Ein Marsupialier Riickenmark. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 11, S. 94.

Porter, W. T. 1895. The path of the respiratory impulse from the bulb to the phrenic nucleus. J. Physiol. (English), vol. 17, p. 455.

Pressley, H., and Cobb, S. 1929. Observations on the spinal cord of Phocaena. J. Comp. Neurol., vol. 47, p. 75.

Probst, M. 1902. Zur Kenntnis der Schleifenschicht und iiber centripetale Riickcnmarksfasem zum Deiters’schen Kem, zum Sehiigel und zur Substantia reticularis. Monatschr. f. Psychiat. u. Neurol., Bd. 11, S. 3.

Quensel, and Kohnstamm, 0. 1907. Neurol. Centralbl., p. 1138.

Ram6n y Cajal, S. 1890. Ryponse 5. Mr. Golgi 5 propos des fibrilles collatdrales de la moelle ypiniere, et de la structure g4nyrale de la substance grise. Anat. Anz., Bd. 5, S. 579.

. 1890. Sur I’origine et les ramifications des fibres nerveuses de la moelle embryon naire. Anat. Anz., Bd. 5, S. 85 u. 111.

. 1899-1904. Textura del sistema nervioso del hombre y de los vertcbrados.

Vol. 1. N. Moya, Madrid.

. 1907. Die Structur der sensiblen Ganglien des Menschen und der Tiere. Anat.

Hefte, Bd. 16, S. 177.

. 1909-11. Histologie du syst^me nerveux de I’homme et des vertybrAs. A.

Maloine, Paris.

Ram6n y Cajal, S., and Oloriz, F. 1898. Los ganglios sensitives craneales de los mamiferos. Rev. trim, microgr., vol. 2, p. 129.

Ranson, S. W. 1908. The architectural relations of the afferent elements entering into the formation of the spinal nerves. J. Comp. Neurol., vol. 18, p. 101.

. 1911. Non-medullated nerve fibers in the spinal nerves. Am. J. Anat., vol. 12,

p. 67.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 329

Ranson, S. W. 1912. The structure of the spinal ganglia and of the spinal nerves. J. Comp. Neurol., vol. 22, p. 159.

. 1913. The course within the spinal cord of the non-medullated fibers of the

dorsal roots: A study of Lissauer’s tract in the cat. J. Comp. Neurol., vol. 23, p. 259.

. 1913a. The fasciculus cerebro-spinalis in the albino rat. Am. J. Anat., vol. 14,

p. 411.

. 1914. The tract of Lissauer and the substantia gelatinosa Rolandi. Am. J.

Anat., vol. 16, p. 97.

. 1914a. An experimental study of Liss.auer’s tract and the dorsal roots. J. Comp.

Neurol., vol. 24, p. 531.

. 1914b. A note on the degeneration of the fasciculus cerebro-spinalis in the albino

rat. J. Comp. Neurol., vol. 24, p. 503.

. 1915. Unmyelinated nerve-fibres as conductors of protopathic sensation. Brain,

vol. 38, p. 381.

. 1918. An introduction to a series of studies on the sympathetic nervous system.

J. Comp. Neurol., vol. 29, p. 305.

. 1926. Studies on muscle tonus. I. Contractile and plastic factors in decerebrate rigiditj’. II. A comparison of the synapse-blocking action of nicotine and chloral hydrate. III. Sublaminal injection of chloral hydrate in decerebrated cats. J. Comp. Neurol., vol. 40, p. 1.

. 1928. The role of the dorsal roots in muscle tonus. Arch. Neurol, and Psychiat.,

vol. 19, p. 201.

. 1930. A discussion of the uses of surgery in conditions involving the sympathetic

nervous system. Graham’s Surgical Diagnosis, W. B. Saunders Co., Philadelphia.

. 1931. The anatomy of the nervous system. 4* ed., W. B. Saunders Co.

Ranson, S. W., and Billingsley, P. R. 1916. The conduction of painful afferent impulses in the spinal nerves. Studies in vasomotor reflex arcs. II. Am. J. Physiol., vol. 40, p. 571.

. 1916a. Afferent spinal paths and vasomotor reflexes. Am. J. Physiol., vol. 42,

p. 16.

. 1918. The superior cendcal ganglion and the cervical portion of the sympathetic

trunk. J. Comp. Neurol., vol. 29, p. 313.

. 1918a. The thoracic truncus sympathicus, rami communicantes and splanchnic

neiwes in the cat. J. Comp. Neurol., vol. 29, p. 405.

. 1918b. An experimental analysis of the sympathetic trunk and greater splanchnic

nerve in the cat. J. Comp. Neurol., vol. 29, p. 441.

Ranson, S. W., and von Hess, C. L. 1915. The conduction within the spinal cord of the afferent impulses producing pain and the vasomotor reflexes. Am. J. Physiol., vol. 38, p. 128.

Ranson, S. W., and Hinset, J. C. 1926. Studies on muscle tonus. IV. The role of the sympathetic nervous system in muscle tonus. J. Comp. Neurol., vol. 42, p. 69.

Ranson, S. W., and Morris, A. W. 1926. Studies on muscle tonus. V. Tetanus contracture. J. Comp. Neurol., vol. 42, p. 99.

Ranvier, M. L. 1874 (also 1889). Trait6 technique d’histologie. F. Savy, Paris.

. 1880. Legons d’anatomie gdngrale sur le systdme musculaire. J. B. Bailli^re

et fils, Paris.

Rasmussen, A. 1932. The principal nervous pathways. The Macmillan Co., New York.

Raitione, G. 1884. Sur I’existence de cellules ganglionaires dans les racines post4rieures des nerfs rachidiens de I’homme. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 1, S. 53.

330 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Reich, Z. 1909. Vom Aufbau der Mittelzone des Riickenmarks. Arb. a. d. neurol.

Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 17, S. 314.

Reveley, I. L. 1915. The pyramidal tract in the guinea-pig (Cavia aperea). Anat. Rec., vol. 9, p. 297.

Reveley, I. L., and Simpson, S. 1909. The cortico-spinal tract in the guinea-pig.

Rep. Brit. A. Adv. Sc., Winnipeg meeting, p. 645.

Riesser, 0. 1925. Ueber den Tonus der Muskeln. Klin. Woch., Bd. 4, no. 1, S. 1 u. 52.

VAN Rijnberk, G. 1907. Sulla metameria nel sistema nervoso simpatico. II. I’lnneia'azione pilomotrice. Arch, di fisiol., vol. 4, p. 349.

. 1908. Versuch einer Segmentalanatomie. Ergeb. d. Anat. u. Entwicklungs gesch. (Merkel u. Bonnet), Bd. 18, S. 353.

. 1918. Recherches sur le tonus musculaire et son innervation. I. Actions ner veuses pseudomotrices et innervation du tonus. Arch, nderl. de physiol., vol. 1, p. 257.

. 1918a. Idem. II. Tonus musculaire et rigiditd de d6c6rd‘bration. Arch.

nderl. de physiol., vol. 1, p. 726.

Rivers, W. H. R., and Head, H. 1908. A human experiment in ner\'e division. Brain, vol. 31, p. 323.

Roberts, F. 1916. Degeneration of muscle following nerve injury. Brain, vol. 39, p. 297.

Rosenzweig. 1905. Beitriige zur Kenntnis des feinen Baues der Substantia gclat^ inosa Rolandi des Riickenmarks. Inaugural Dissertation, Berlin.

Rossoli’mo, G. 1886. Zur Frage iiber den weiteren Verlauf der Hinterwurzelfasern im Riickenmarke. Neurol. Centralbl., Bd. 5, S. 391.

Rothmann, M. 1902. Ueber die spinalen Athmungsbahnen. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 11.

. 1906. Ueber die Leitung der Sensibilitat im Riickenmark. Berlin Med.

Gesellsch., S. 562.

Rdffini, a. 1905. Sur les expansions nen^euses dc la peau chez I’hommc ct quclqucs autres mammiferes. Rev. gen. d’histoL, vol. 1, p. 421.

Saegusa, G. 1931. Ueber die "spinalparasympathctischen” F.oscrn. Mitteil. Med. Ges., Tokio, vol. 45, p. 1864.

Sakai, S. 1913. Vergleichende Untcrsuchungen des Conus terminalis bei Mann and Frau. Arb. a. d. neurol. Inst. a. d. Wein. Univ. (Obersteiner’.s), Bd. 20, S. 47. Sandmever, W. 1892. Sekundare Degeneration nach Extirpation motorischer Centra.

Zcitschr. f. Biol., Bd. 28 (N. F. Bd. 10), S. 177.

Sano, T. 1904. Ia:s localisations des fonctions motoriccs dc la moelie (5piniero. Rapport du Congres des medicins .alicncstcs ct neurologistcs de France et des p.a 3 ’s de languc francjaisc. Pau.

. 1909. Vcrgicichend-anatomische und physiologische Untcrsuchungen fiber die

Substantia gelatinosa des Hinterhorns. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Olierslciner’s), Bd. 17, S. 1.

Schaffer, J. 1891. Vcrgicichend-anatomische Untcrsuchungen hber Riickenmarksfa.'-erung. Arch. f. mikr. Anat., Bd. 38, S. 157.

. 1891. Beitriige zur Kenntniss des Sttitzgeriistes im menschlichcn llfickcnmarkc.

Arch. f. mikr. Anat., Bd. 44, S. 26.

Schaffer, K. 1898, Beitrag zum Fn.«cr\'er]nuf der Ilintenvurzcln irn Cervicalmarkc des ^ipnschen. Neurol. Centndbl., Bd. 17, S. 434. va.n der Scheer, W. M. Beitrag zur Frage nncli der Bedcutung der Herpes zoster und der Ileadsclion h>-pcralgcti.«chen Zonen. Zcitschr. f. d. ges. Neurol, u. P.sychiat., Bd. 10, S. 313.

van der Scheer, W. M., and Stuur.man, F. .1. 1910. Fin Fall von Herpes zoster nut

anatomischem Befund. Zcitschr. f. d. ges. Neurol, u. Psychiat., Bd. 34, S, 119.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 331

\

Schmidt, C. A. 1930. Distribution of vagus and sacral nerves to large intestine.

Proc. Soc. exp. Biol, and Med., vol. 30, p. 739.

Scholl, K. 1910. Sammelreferat uber Head’sche Sensibklitatslehre und ihre Kritiker.

Zeitschr. f. d. ges. Neurol, u. Psychiat., Referate.

ScHOONDEMARK, A. 1918. Over gevoelsstoornissen bij een aantal klinische gevallen van periphere zenuwlaesies. Psychiat. en neurol. Bl., Amsterdam, Bd. 22, S. 1. ScHULTZE, F. 1883. Beitriige zur Lehre von der secundiiren Degeneration im Riickenmark des Menschen nebst Bemerkung iiber die Anatomie der Tabes. Arch. f. Psychiat., Bd. 14, S. 359.

Selle, H. 1931. Die Nervenbahnen in der hinteren Wurzeln und im Spinalganglion mit besonderer Beriicksichtigung ihrer Beziehungen zum vegetativen Nervensystem. Sitzungsber. Phys. med. Soc. Erlangen, vol. 62, S. 303.

Sheehan, D. 1933. On unmyelinated fibers in the spinal nerves. Anat. Rec. vol. 55,

p. 111.

Sherrington, C. S. 1885. On secondary and tertiary degenerations in the spinal cord of the dog. J. Physiol., vol. 6, p. 177.

. 1890. On out-lying nerve-cells in the mammalian spinal cord. Phil. Trans.

Roy. Soc., London, Ser. B, vol. 181, p. 33.

. 1893. Experiments in examination of the peripheral distribution of the fibers

of the posterior roots of some spinal nerves. Phil. Trans. Roy. Soc., London, Ser. B, vol. 184, p. 641.

. 1894. The nen'es of skeletal muscle. J. Physiol., vol. 17, p. 211.

. 1897. On the question whether any fibers of the mammalian dorsal (afferent)

spinal root are of intraspinal origin. J. Physiol., vol. 21, p. 209.

. 1897a. Further note on the sensory nerves of muscle. Proc. Roy. Soc., London,

vol. 61, p. 247.

. 1897-98. Decerebrate rigidity and reflex coordination of movements.

J. Physiol., vol. 22, p. 319.

. 1898. Further note on the sensory nerves of eye muscles. Proc. Roy. Soc.,

London, Ser. B, vol. 64, p. 120.

. 1901. The spinal roots and dissociative anaesthesia in the monkey. J. Physiol.,

vol. 27, p. 360.

. 1906. The integrative action of the nervous system. C. Scribner’s Sons, New

York.

. 1915. Postural activity of muscle and nen-e. Brain, vol. 38, p. 191.

. 1917. jNIammalian physiology ; a course of practical exercises. The Clarendon

Press, Oxford and New York.

Sherrington, C. S., and Laslett, E. E. 1903. Remarks on the dorsal spino-cerebellar tract. J. Physiol., vol. 29, p. 188.

Shimazono j. 1914. Beitrag zur Lokalization der Muskeln der Hand, des Beins und des Zwerchfels im Riickenmarke der Menschen. Mitt. med. Fakult. k. Univ. Tokyo, Bd. 11, H. 2. Reviewed in Folia neuro-biol., Bd. 9, S. 147, 1915.

Simpson, S. 1912. The pyramid tract in the Canadian porcupine (Erethizon dorsatus, Linn.). Proc. Soc. Exper. Biol, and Med., vol. 10, p. 4.

1914 , The pyramid tract in the red squirrel (Sciurus hudsonius loquax) and

chipmunk (Tamias striatus lysteri). J. Comp. Neurol., vol. 24, p. 137.

. 1914a. The motor areas and pyramid tract in the Canadian porcupine (Erethizon dorsatus, Linn.). Quart. J. Exper. Physiol., vol. 8, p. 79.

. 1915. The pyramid tract in the striped gopher (Spermophilus tndecemlmeatus).

Quart. J. Exper. Physiol., vol. 8, p. 383. , , tt j

Singer J 1881 Ueber sekundare Degeneration im Ruckenmarke des Hundes. It’zungsb dK. Akad. d. Wissensch. zu Wien, Math.-nat. Cl., Bd. 84, Abt. 3, S. 390.

332 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Slinger, R. T., and Horsley, V. 1906. Upon the orientation of points in space by the muscular, arthrodial, and tactile senses of the upper limbs in normal individuals and in blind persons. Brain, vol. 29, p. 1.

SoKOLOWA, M. L. 1931. Zum Lehre von der Cytoarchitektonik des peripherischen autonomen Nervensystems. II. Die Architektur der intra-muralen Ganglien des Verdauengstrakts des Rindes. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr. -anat. Forschung, Bd. 23, S. 552.

Spiegel, E. A. 1901. Die zentrale Lokalisation autonomer Funktionen. Zeitschr. f. d. ges. Neurol, u. Psychiat., Referate, Bd. 22. (For 1932, see Chapter VIII.)

Spiegel, E. A., and Bernis, W. J. 1925. Studies of sympathetic nervous system; paths of visceral sensibility in spinal cord. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 210, S. 309.

Spiller, W. G. 1899. A contribution to the study of the pyramidal tract in the central nervous system of man. Brain, vol. 22, p. 563.

. 1902. Ueber den directen ventro-lateralen Pyramidenstrang. Neurol. Cen tralbl., Bd. 21, S. 534.

Spitzka, E. C. 1886. The comparative anatomy of the pyramid tract. J. Comp. Med. and Surg., vol. 7, p. 1.

Starling, E. H. 1920. Principles of human physiology. 3'‘’ edition. J. and A. Churchill, London. Later edition in 1930.

Starr, M. A. 1894. Local anaesthesia as a guide in the diagnosis of lesions of the upper portion of the spinal cord. Brain, vol. 17, p. 481.

Stern, R. 1908. Beitrag zur Kenntnis der Form und Grosse der Rtickenmarksquerschnittes. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 14, S. 329.

Stieda, L. 1869. Studien fiber das centrale Nervensystem der Vogel und Siiugethiere. Zeitschr. f. wissensch. Zook, Bd. 19, S. 1.

Stilling, B. 1859. Neue Untersuchungen ueber den feineren Bau des Riickenmarks. H. Hotop, Cassel.

Stohr, P., Jr. 1928. Das peripherischen Nervensystem. Das Handbuch der mikroscopischen Anatomic des Menschen, edited by von Mollendorf. G. Fischer, Jena.

. 1930. Mikroskopische Studien zur Innervation des Magen-Darmkanales.

Zeitschr. Wiss. Biol., Abt. B, Zeitschr. Zellforsch. u. mikr. Anat., Bd. 12, S. 66.

Stotsenburg, j. M. 1909. On the growth of the albino rat (Mus norvegicus var. albus) after castration. Proc. A. Am. Anatomists, Anat. Rec., vol. 3, p. 233.

Streeter, G. L. 1909, 1913. Laboratory guide in anatomy. G. Wahr, Ann Arbor.

. 1919. Factors involved in the formation of the 61um terminale. Am. J. Anat.,

vol. 25, p. 1.

Strumpell, a. 1904. Ueber die Bedeutung der SeusibilitatsprufUngen mit besondercr Beriicksichtigung des Drucksinnes. Deutsche med. Wochenschr., Bd. 30, S. 1411 u. 1460.

Taft, A. E. 1920. A comparison of the anterior horn cells in the normal spinal cord and after amputation. Arch. Neurol, and Psychiat., vol. 3, p. 41.

Takahasiii, D. 1913. Zur vergleichcnden Anatomic des Seitenhorns im Riickcnmark der Verlcbratcn. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Oberstciner’s), Bd. 20, S. 62.

Teding van Berkiiout. 1893. Experimentell-anatomischer Beitrag zur Kenntnis dor .secundiiren Degeneration im Riickenmarke. Inaugural Dissertation, Bern.

Testdt, L. 1921-23. Trait<5 d’anatomie humainc. 0. Doin, Paris.

Tiio.mas, M. a. 1897. Sur les fibres d’union do la mocllc avec Ics autres centres nerveux ct principalcment sur les faisceaux cdr^belleux ascendants. Compt. rend. Soc. de biol., Ser. 10, vol. 4, p. 88.

Thompson, T. 1912. A case of .subacute combined degeneration of the spinal cord demonstrating the nature of the afferent impulses in the posterior columns. Brain, vol. 34, p. 510.

THE COMPARATIVE ANATOMY OF THE SPINAL CORD 333

Thorburn, W. 1887. Cases of injury to the cervical region of the spinal cord. Brain, vol. 9, p. 510.

. 1888. On injuries of the cauda equina. Brain, vol. 10, p. 381.

. 1889. Spinal localisations as indicated by spinal injuries. Brain, vol. 11, p. 289.

. 1893. The sensory distribution of spinal nerves. Brain, vol. 16, p. 355.

Tigerstadt, II. 1909-10. Lehrbuch der Physiologie des Menschen. S. Hirzel, Leipzig.

Tilney, F., and Elsberg, C. A. 1926. Sensory disturbances in tumors of the cervical spinal cord : arrangement of the fibers in the sensory pathways. Arch. Neurol, and Psychiat., vol. 15, p. 444.

Tinel, J. 1916. Les blessures des nerfs. Paris.

Tower, S. S. 1926. A study of the sympathetic innervation to skeletal muscle. Am. J. Physiol., vol. 78, p. 462.

Trostanetzky, M. M. 1928. Zur Frage des Baues der sympathischen Ganglien des Darmgeflechtes. ' Zeitschr. f . wissensch. Biol. ; Abt. B, ^itschr. f . Zellforsch. u. mikr. Anat., Bd. 8.

Trotter, W., and Davies, H. M. 1909. Experimental studies in the innervation of the skin. J. Physiol., vol. 38, p. 134.

VAN Valkenburg, C. T. 1909. Zur Anatomie der Hinterstrange (kreuzende Fasem). Neurol. Centralbl., Bd. 28, S. 2.

. 1912. Algemcine klinische lokalisatic in het zenuwstelsen. Wolters, Groningen.

. 1914. De organisatie der sensibiliteit in het zenuwstelsel. Psychiat. en neurol.

Bl., Amsterdam, Bd. 18, S. 387.

. 1917. Le paraRllisme entre la sensibility cutanye et la sensibility profonde.

Arch, nyerl. de physiol., Ser. 3C, vol. 1, p. 653.

Vermeulen, H. a. 1916. On the conus medullaris of the domestic animals. Kon. Akad. V. Wetensch. te Amsterdam, Proc. sect, sc., vol. 18’, p. 780.

. 1916. Ueber den Conus medullaris der Haustiere, sein besonderes Verhalten

beim Pferd und dessen Bedeutung. Berliner tierarztl. Wochenschr., Bd. 32, S. 13.

. 1916a. Over den nervus sympathicus van huisdieren. Verslag. kon. Akad. v.

Wetensch. te Amsterdam, vol. 24, p. 1222.

. 1916b. Ueber den Nervus sympathicus der Haustiere. Anat. Anz., Bd. 49, S. 301.

VAN DER Vloet. 1906. Ueber den Verlauf der Pyramidenbahn bei niedern Siiugetieren. Anat. Anz., Bd. 29, S. 113.

VoRis, H. C. 1928. The morphology of the spinal cord of the Virginian opossum (Didelphis virginiana). J. Comp. Neurol., vol. 46, p. 407.

. 1928a. The arterial supply of the brain and spinal cord of the Virginian opossum

(Didelphis virginiana). J. Comp. Neurol., vol. 44, p. 403.

VuLPiAN, A. 1876. Destruction de la substance grise du gyrus sigmoide du cote droit sur un chien. Arch, de Physiol., p. 814.

Wagner, R., and Meissner, G. 1852. Ueber das Vorhandensein bisher unbekannter eigenthiimlicher Tastkorperchen (Corpuscula tactus) in den Gefuhls warzchen der menschlichen Haut und neben die Endausbreitung sensitiver Nerv'en. Nachr. V. d. k. Gesellsch. d. Wissensch. u. d. Georg.-Aug. Univ., Gottingen, S. 17. Waldeyer, H. 1889. Das Gorilla -Riickenmark. Abhandl. d. kon. preuss. Akad. d.

Wissensch., Berlin, Phys.-math. Classe. 1888, Abt. III. Berlin, 1889. Wallenberg, A. 1898. Beitrage zur Topographie der Hinterstrange des Menschen.

Deutsche Zeitschr. f. Nervenh., Bd. 13, S. 441.

Wang, C. C. 1927. On the postnatal growth in the area of the optic nerve in the albino and in gray Norway rats. J. Comp. Neurol., vol. 43, p. 201.

Warrington, W. B., and Griffith, F. 1904. On the cells of the spinal ganglia and on the relationship of their histological structure to the axonal distribution. Brain, vol. 27, p. 297.

334 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Watson, J. B. 1905. The effect of the bearing of young upon the body weight and the weight of the central nervous system of the female rat. J. Comp. Neurol., vol. 15, p. 514.

Weed, L. H. 1914. Observations upon decerebrate rigidity. J. Physiol., vol. 48, p. 205.

Weil, A. 1927. A comparative quantitative study in proprioceptive tracts of mammals. J. Comp. Neurol., vol. 44, p. 61.

Wertheim-Salomonson. 1888. Stereognosis. Dissertation, Leiden.

Wilkinson, H. J. 1929. The innervation of striated muscle. Australian Med. J., vol. 2, p. 768.

. 1930. Experimental studies on the innervation of striated muscle. J. Comp.

Neurol., vol. 31, p. 129.

. 1930a. Observations on the sole plates of motor end organs. J. Comp. Neurol.,

vol. 50, p. 133.

Wilson, J. T. 1921. The double innervation of striated muscle. Brain, vol. 44, p. 234.

WiNDLE, W. F. 1923. Unmyelinated nerve fibers of the dorsal roots. J. Anat., vol. 57, p. 360.

. 1926. Non-bifurcating nerve fibers of the trigeminal nerve. J. Comp. Neurol.,

vol, 40, p. 229.

. 1926a. The distribution and probable significance of unmyelinated nerve fibers

in the trigeminal nerve of the cat. J. Comp. Neurol., vol. 41, p. 453.

. 1929. The relation of the level of transection of the brain stem to the occurrence

of decerebrate rigidity in young animals. J. Comp. Neurol., vol. 48, p. 227.

. 1931. Neurons of the sensory type in the ventral roots of man and of other

mammals. Arch. Neurol, and Psychiat., vol. 26, p. 791.

Winkler, C. 1903. Ueber die Rumpfdermatome. Ein experimenteller Beitrag zur Lehre der Segmental-Innervation der Haut. Monatschr. f. Psychiat. u. Neurol., Bd. 13, S. 161.

. 1918. Le faisceau sacral post4ro-m4dial dans les cas de tabes, et la demonstration

que ce faisceau nel renferme acune fibre spinale endogSne et est formd de fibres radiculaires descendantes. Opera omnia, Bd. 5, S. 461. E. F. Bohn, Haarlem.

. 1919. Anatomie du systeme nerveux. Vol. 1. E. F. Bohn, Haarlem.

Winkler, C., and van Rijnberk, G. 1901. On function and structure of the trunkdermatoma. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., .vol. 4, p. 266. Idem., Part II, vol. 4, p. 308.

. 1902. Structure and function of the trunk-dermatoma. III. Kon. Akad. v.

Wetensch. te Amsterdam, Proc. sect, sc., vol. 4, p. 509.

. 1903. Over het uitgroeien van de laterale rel den der rompdermatomen op de

caudale afdeeling der bovenste extremiteit. Verslag. kon. Akad. v. Wetensch. te Amsterdam, vol. 12, p. 570.

WooLLARD, H. H. 1926. The innervation of the heart. J. Anat., vol. 55, p. 345.

Yoshida, I. 1924. Ueber die Nissl-degeneration nach Kleinhimlasionen. I Teil. Ueber die Nissl-degeneration in den Vestibularisendkemen und anderswo nach Kleinhimlasionen. Okayama-Igakkwai-Zasshi, Nrs. 415-16.

Zeehandelaar, I. 1921. Ontogenese und phylogenese der Hinterstrangkeme in Verband mit der Sensibilitat. Folia neuro-biol., Bd. 12, S. 1. Also as Dissertatie, Amsterdam, 1919.

Ziehen, T. 1887. Secundare Degeneration nach Exstirpation motorischer Rindenregionen. Arch. f. Psychiat., Bd. 18, S. 300.

. 1899. . Zur vergleichenden Anatomie der Pyramidenbahn. Anat. Anz., Bd. 16,

S. 446.

. 1900. Ueber die Pyramidenkreuzung des Schafes. Anat. Anz., Bd. 17, S. 237.

. 1899a, 1903, 1913. Makroskopische und mikroskopische Anatomie des Zentral nervensystems. J. Fischer, Jena.

Chapter III The Medulla Oblongata

Taste and the General Afferent Systems

The medulla oblongata, as this term is applied in vertebrates below mammals, is that segment of the central nervous system situated between the midbrain orally and the spinal cord caudally. This places its cephalic boundary in a plane which passes through the anterior medullary velum and usually through the region of the decussation of the trochlear nerve. The caudal level is less readily determinable ; ob^’iously the region of passage from the cranial cavity into the spinal canal cannot serve as a r^iable guide in the face of the increase in the size and caudal extent of the skull from low.er to higher vertebrates. In most vertebrates the region just oral to the region of the emerging root fibers of the first cervical nerve is regarded as medulla oblongata. In mammals the medulla oblongata, as defined in the Basle anatomical nomenclature, extends from a plane in front of the level of emergence of the most cephalic root fibers of the first spinal nerve (or through the lower border of the motor decussation or the foramen magnum) to a plane through the stria medullaris dorsally and the caudal border of the pontine fibers ventrally. The segment of nervous tissue bordered ventrally by the pontine fibers, extending dorsally between the stria medullaris acustici and the trochlear decussation, arbitrarily separated at each side by a section through the trigeminal roots, and forming the cephalic portion of the submammalian medulla oblongata, is known in mammalian anatomy as the pons. In the present account it will be considered with the medulla oblongata of these forms.

A study of the finer structure of the medulla oblongata indicates that three general types of constituents form the major portion of its walls. These may be grouped as : (1) the nuclei of origin and termination of the cranial nerves, together with their secondary connections; (2) the nuclear groups of the region other than those of the cranial nerves, and their secondary connections ; (3) the fibers of passage.

The central canal of the spinal cord, which in the highest mammals is only partially patent as a rule, continues forward into the medulla oblongata where it widens out in the region of the calamus scriptorius to form the fourth ventricle. This ventricle has a roof formed partly by the overlying cerebellum and partly by the choroid plexus of the ventricle, which passes over into the cerebellum in the region of the posterior medullary velum and into the thicker side walls along the so-called taenia choroidea ventriculi quarti. The shape of the ventricle is that of a rhomboid, the caudal angle lying at the calamus scriptorius and the cephalic angle where the fourth ventricle narrows down into the aqueduct of

335

336 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Sylvius, while the two lateral angles of the rhomboid, as represented in the widest portion of the ventricle, lie at about the level of entrance of the acoustic nerve or in the region of the so-called lateral recess of higher forms. From its general shape the fourth ventricle is termed at times the ventriculus rhomboidalis, and the brain segment from the aqueduct region (or the decussation of the trochlear) to the beginning of the cord, that is, the medulla oblongata of vertebrates

Fig. 135. Dorsal view of medulla oblongata of Petromyzon marinus. The tela choroidea has been removed. The IXth and Xth roots are not shown. Note that the greatest width of the medulla oblongata is found at the entrance of the largest dorsal roots.

below mammals, is called the rhombencephalon. The choroid plexuses have been discussed in the first chapter (p. 47) and do not require further consideration here. As was implied above, the sides and floor are massive, consisting of fiber bundles and nuclear centers (fig. 135).

Even on gross inspection the floor and walls of the medulla oblongata show some differentiation, which is a reflection of its functional pattern. In its simpler and more instructive form this differentiation is seen in the medulla oblongata of adult selachians, but it is evident in the embryonic stages of the higher verte

THE MEDULLA OBLONGATA

337

brates. In the shark (fig. 136), in each half of the floor^of the fourth ventricle, there is to be noted a deep sulcus which runs longitudinally through the medulla oblongata. This is the sulcus limitans, which separates the sensory areas dorsolateral to it from the motor and effectory areas ventromedial to it. The

Fig. 136. Dorsal view of the medulla oblongata of a shark (Carcharias glaucus). The tela choroidea has been removed.

motor areas are again separable grossly into a more medial somatic motor or efferent area, appearing as a longitudinal column projecting into the ventricle on either side of the median sulcus, and a lateral, more depressed longitudinal column representing the position of the visceral efferent areas. The somatic efferent area is to be regarded as the direct forward continuation of the ventral

338 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

horn region of the spinal cord ; the visceral efferent region is in line with, and in part comparable to, the intermediolateral column. The afferent regions of the cord, represented by the dorsal horn gray, continue laterally in the medulla oblongata with the widening out of the central canal into the fourth ventricle. They lie in the region lateral to the sulcus limitans and are represented by two longitudinal columns which project out into the ventricle. The more ventromedial of these columns is beaded in appearance in the selachians. Within it and below it lie the visceral afferent centers. The more dorsolateral of the columns marks the position of somatic afferent centers and varies in size somewhat at different levels, depending upon the regions of entrance of somatic afferent nerves. In the medulla oblongata of adult higher forms this fundamental pattern is somewhat modified by the increase and fusion of certain columns and the migration of certain nuclear masses. Such modifications lead, among other changes, to a partial obliteration of the sulcus limitans ; thus, for example, in the human medulla oblongata a single region, the ala cinerea, on the floor of the medulla oblongata overlies both visceral efferent and visceral afferent centers, and the sulcus limitans which intervenes between the two regions in the embryo is obliterated in the adult.

The columns or centers — somatic efferent, visceral efferent, visceral afferent, and somatic afferent — thus indicated on the floor and walls of the medulla oblongata are characterized by the presence of, and receive their name from, appropriate nuclei of origin or nuclei of termination of the cranial nerves (sec fig. 137).

The nuclei of origin and termination, their arrangement, and the pattern of distribution of their a.ssociated nerves will be considered in detail in this and the succeeding chapters. It will be seen that individual nerves may carry several types of components but that the central distribution of each fiber group will depend more particularly upon the kind of impulses which it carries than upon the peripheral nerve over which they have passed. This rearrangement within the brain has its ex-planation in the so-called doctrine of nerve components which had its inception with the work of Gaskell (’86, ’89, etc.), was extendwl and amplified by the work of Oliver Strong (’90, ’92, ’95), of Herrick (’99), and of Johnston (’02), and has received support and corroboration through the work of numerous obser\'ers.

In the medulla oblongata of craniotes, as in their spinal cord, both ventral and dorsal roots are present, but whereas the ventral roots in this region are

Fig. 137. A schematic cross section through the posterior portion of the medulla oblongata of .Scyllium canicula. The somatic sensory portion of the alar plate is indicated by perpendicular lines; the visceral sensory portion by horizontal lines. The floor plate is unlined.

THE MEDULLA OBLONGATA

339

considerably reduced in size, the dorsal roots of the medulla oblongata are extraordinarily well developed. This relative increase in the size of the dorsal roots, which apparently contributes to the lateral exversion of the area and thus contributes to the formation of the fourth ventricle, is of two kinds. On the one hand, there is a special differentiation of the somatic sensory components in the trigeminal and acoustic nerves, in consequence of the development of the mouth and face region and of the acoustico-lateral system. On the other hand, the visceral components of the dorsal roots are much greater than those of the corresponding roots of the cord, due to the presence of gill arches or their derivatives and the appearance of the gustatory terminations with their associated fibers.

The branchial nerves proper also show a considerable increase in the visceral efferent as well as the visceral afferent components. At their point of entrance to the brain, such visceral efferent fibers are associated in position with the cutaneous and visceral afferent fibers, a relationship which exists wholly or in part in the dorsal roots of the spinal nerves of many vertebrate types. The evidence indicates that the branchial nerves must be considered merely as special differentiations of such primitive dorsal roots containing these components. The muscles supplied by the somatic efferent fibers of the spinal nerves and by the hypoglossal and abducens nerves of the rhombencephalon (and the trochlear and oculomotor nerves of the mesencephalon) are derived from the myotomes of the somites. Typically, their centers are situated close to the floor of the ventricle, usually near the medial longitudinal fasciculus. Thps they fall within the somatic efferent column. The visceral efferent nerves fall into two groups : (1) a general visceral efferent or preganghonic group, the cells of origin of which lie within the visceral efferent column situated typically just lateral to the somatic efferent centers and consisting in mammals of the Edinger-Westphal nucleus of the oculomotor nerve (a midbrain center), the superior salivatory nucleus of the facial, the inferior salivatory nucleus of the glossopharyngeal, and the dorsal efferent nucleus of the vagus and accessory nerves, and (2) a special visceral efferent group of nerves which supplies the striated gill musculature of lower forms or their derivatives in higher forms. These muscles are derived from lateral plate mesoderm {van Wijhe, ’82, and others). Thus, from their more primitive relations to the viscera and their embryonic origin, these branchial muscles are to be regarded as visceral muscles and their nerves as visceral efferent nerves. Such branchial muscles or their derivatives, such as the muscles of facial expression or those for the jaws, are striated and, in the adult, resemble in appearance the ordinary segmental muscle derived from somites. Hence the nerves are termed special visceral efferent, to differentiate them from those supplying cardiac and smooth muscle. The nuclei of origin for the special visceral efferent fibers are the motor nuclei of the trigeminal, the facial, the glossopharyngeal-vagus-accessory group (nucleus ambiguus of man), and possibly the accessory nucleus of the cervical cord (see page 588). With the exception of the motor nucleus of the trigeminal nerve, which often is situated relatively far dorsally, in many mammals these special visceral efferent centers lie deep within the substance of the medulla oblongata.

340 NERVOUS SYSTEMS OF VERTEBRATES AND OF AL4N

The afferent nerv’-es of the brain stem fall also into visceral and somatic groups, associated respectively with the visceral afferent and the somatic afferent columns.

The \asceral afferent fibers carried by the branchial ners’^es are found in facial, glossophaiyngeal, and vagus ner\^es. Their sensory fibers have origin from extramedullary cranial ganglia (fig. 138), which are relatively large and which, in most vertebrates, consist of unipolar neurons. The number of visceral

Fig. 1.38. Reconstruction of the cranial ganglia and head sympathetic of a trout 2.5 cm. long. L. Sdtwarlz.

A7.//. (Klcinhirn), cerebellum ; jlf.//., midbrain ; M.nhl., medulla oblongata ; //, optic nerve; III, oculomotor nerve; IV, trochlear nerve; V, trigeminus nerve; IV, abducen.s nerve; IV/, facial ner\-c; VIII, vestibular ncta'e; IX, glo.s.sopharjTigcus nerve; A', vagus nera’c; S‘~ , spinal nerves; L.V., L.VII, L.IX, L.X, N.N. lateralcs.

The visceral scn.sory fibers and their ganglia arc colored black; all other rool.s are black lined. G.C., ganglion ciliare (of oculomotor); K^~ (Kopfsympathicu.“gnnglicn) cranial sympathetic ganglia; K connecting branches to the branchial nerves; <7.5. (Grentastrang), sympathetic chain.

afferent fibers in any of these nerves is large compared with Ihc number in a dorsal root of the cord, since the inner surface of the brancliial region, due to the enlargement of mucous surfaces, particularly in its more cephalic portions, provides the greater part of the sensory fibers found in the.se three branchial nerves. Such visceral afferent elements fall into a special visceral sensory or afferent type, carrying gustatory impulses from taste buds, and a general visccr.al afferent type concerned with general .sensations from visceral surfaces. The cutaneous elements arc relatively reduced in the three branchial nerves under consideration, although in lower forms demonstrable in each. In higher fonns the facial carries no cutaneous sensory, and its presence in the glo.s,‘=opharjmgc.'ii

THE MEDULLA OBLONGATA

341

is questioned by many, while the somatic sensory fibers in the vagus, although demonstrable in the auricular ramus, are so few as to be practically negligible. The progressive loss of cutaneous sensibility in these nerves is compensated for by a gradual increase in this type of fibers in the trigeminus, for the major number of the cutaneous fibers for the head, in so far as it is supphed by cranial nerves, run in the trigeminal nerve. This nerve, although one of the branchial nerves, has no visceral afferent components, but has a large number of somatic afferent fibers of both exteroceptive and proprioceptive type (as well as the special visceral efferent fibers mentioned earher).

To the nuclear groups of the medulla oblongata, other than those of the cranial nerves, belong the reticular nuclei of the region and the inferior olive. These are discussed in Chapter VI. The centers such as the superior olive, trapezoid nuclei, and pontine gray are treated with the fiber systems with which they are in relation. Nucleus gracilis and nucleus cuneatus are important nuclear constituents of the medulla oblongata of mammals and are represented in certain lower forms such as birds and reptiles.

The fasciculus gracilis and fasciculus cuneatus become infiltrated with the gray of their respective nuclei at about the level of transition from the cord to the medulla oblongata, nucleus gracilis appearing slightly caudal to nucleus cuneatus. Followed forw'ard, the nuclei increase in size as the fascicular fibers decrease, but occupy the same relative positions as did their related fiber paths within the cord. With the opening of the fourth ventricle, the nuclei swing slightly lateralward. In forms in which these nuclei are well developed they form eminences on the outer wall of the medulla oblongata, the clava and the cuneate tubercle of human anatomy. The details of their development and of the structure and the arrangement of their gray matter in various higher vertebrates have been noted in Chapter II (pp. 262-265), and reference is made here to the description there given. There also their fiber connections have been described. However, certain of the fiber connections form important fiber components of the medulla oblongata and thus deserve consideration here. Arising from nucleus gracilis and nucleus cuneatus are a series of fiber fascicles which swing down toward the midline and constitute the internal arcuate system of the medulla oblongata. After crossing the midline many of these fibers turn directly forward as the medial lemniscus. This medial lemniscus, as it develops, occupies a vertical position parallel to the midsagittal plane to about the level of entrance of the cochlear nerve (the stria medullaris acustici, or the beginning of the pons in mammals), at which level it gradually shifts to a horizontal plane as it passes forward. At first it lies near the midline, but gradually moves lateralward and then swings obliquely dorsalward and so proceeds to the ventral nucleus of the thalamus. The medial lemniscus thus is composed of the second neuron (bulbo-thalamic neuron) in the chain for proprioceptive impulses from the extremities and trunk. It carries also impulses for two-point discrimination, at least for the trunk and upper extremities (p. 521). Other internal or deep arcuate fibers arising from the nucleus gracilis and nucleus cuneatus, after decussating in the midline, swing downward, usuallj" internal but now and again

342 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

external to the pyramids, and, after a variable amount of synapse in the arcuate nucleus, pass dorsalward over the periphery of the medulla oblongata as the ventral superficial arcuate system to enter the cerebellum by way of the restifonn body, carrying proprioceptive impulses to the cerebellum. The fibers of passage found in the medulla oblongata are discussed with the respective nuclei of origin or termination. They require no reference here except to indicate their shift of relative position as they pass from the cord to the medulla oblongata or from the medulla oblongata to the cord. The most marked changes in position are seen in the cortico-spinal paths, which in mammals interdigitate with the fibers of the pons and below the level of this brain segment form the pyramids, until the caudal portion of the medulla oblongata is reached, where 75 per cent to 90 per cent of them decussate to form the motor decussation and then continue into the cord as the dorsal (or dorsolateral) cortico-spinal fasciculus, those which remain uncrossed passing directly caudalward as the ventral (or medial) cortico-spinal fasciculus. Lateral to the fibers of this motor decussation lie the fasciculi which formed the important elements of the ventral funiculus of the cord. With the increase of the fibers in the pyramids they are forced dorsolateralward and then to a dorsal position on either side of the midline. At first they overlie the pyramids, but almost at once are separated from them by the gradually increasing internal arcuate fibers which cross in the sensory decussation to form the medial lemniscus. In a section of the medulla oblongata taken at the level of the vagus, the fiber paths or fasciculi present the following order, beginning with the floor of the fourth ventricle : the medial longitudinal fasciculus, the medial tecto-spinal fasciculus, the ventral spinothalamic fasciculus, the medial lemniscus, and the pyramids.

The fasciculi of the lateral funiculus of the spinal cord are continued directly into the lateral field of the medulla oblongata, in which they occupy essentially similar positions. Within the medulla oblongata the dorsal spino-cerebellar fasciculus overrides the nucleus of the descending root of the trigeminal to enter the corpus restiforme and thus proceeds to the cerebellum ; the olivo-spinal and spino-olivary fasciculi terminate at the level of the inferior olivary nucleus, the reticulo-spinal in the reticular nuclei and the lateral (or ventrolateral) vestibulospinal and spino-vestibular fasciculi extend dorsalward to the region of the spinal vestibular and lateral vestibular nuclei. The remaining fasciculi of the lateral funiculus of the cord (ventral spino-cerebellar, rubro-spinal, lateral tectospinal and spino-tectal, and lateral spino-thalamic tracts) are found in the lateral field of the medulla oblongata, dorsolateral to the inferior olivary nucleus and in substantially the same relative positions as in the spinal cord. They retain these relations until the upper border of the rhombencephalon is reached, in which region the ventral spino-cerebellar swings dorsalward and then caudalward to enter the cerebellum along with the superior cerebellar peduncle. In the mammalian pons the ponto-cerebellar tracts overlie these fasciculi and separate them from the periphery.

One of the conspicuous elements of the rhombencephalon is the pontocerebellar system of fibers. These fibers are discussed in detail in the chapter on

TASTE AND THE GENERAL AFFERENT SYSTEMS

343

the cerebellum and only their general morphologic relations will receive attention here. As a clearly defined system they are present only in mammals, and they increase progressively from lower to higher mammals mth the corresponding increase in the cerebral cortex. They delimit ventrally and give name to the pons of the B.N.A. nomenclature. Passing between the ponto-cerebellar fibers and at right angles to them are to be found cortico-pontine, cortico-bulbar, and cortico-spinal fibers. The last two of these fiber paths send collaterals to the scattered gray of the ponto-cerebellar fibers and the cortico-pontine fibers synapse with the cells of this gray, which is known as pontine gray or pontine nucleus, its cells serving chiefly as neurons of the second order in the two-neuron path (corticopontine and ponto-cerebellar path) between the cerebral cortex and the cerebellar cortex.

The Structure, Distribution,

AND Number of Taste Buds IN Vertebrates

The presence of gustatory fibers gives special significance to the sensory roots of the vagus, glossopharyngeal, and facial nerves. Such being the case, it appears advisable to begin the discussion of this section with a brief account of the structure of typical taste buds and of their distribution, in general, in various vertebrates (for details see Kolmer, ’27, and for early accounts. Sappy, ’47, Compt. Rendus Acad., Paris; Schwalbe, ’68, Arch. f. mikr. Anat., Bd. 4 ; Loven, ’68, Arch. f. mikr. Anat., Bd.4 ; Krause, ’70 ; Merkel, ’92; and Graberg, ’99).

In general plan the taste buds are much ahke in form and structure in all vertebrates, but there are certain minor variations. There is a considerable difference in their shape in different forms. Avian taste buds are narrow ; in amphibians and mammals (newborn, 42^1 by 60m, Kolmer, ’27), they are much broader. Many taste buds may be compared to closed rosebuds, which are wide at the base and in the middle portions and which narrow down toward the distal end. They extend entirely through the epithelium. A small gustatory canal connecting the bud with the surface was described long ago by von Ebner (’97) for cats, monkeys, and man. Its inner and outer openings are termed respectively the inner and outer pores. The size of the canal varies in different forms. Three types of cells are concerned in the formation of the taste buds : basal cells, sustentacular cells, and neuroepithelial or sense cells. The basal cells, which are relatively few in number (not present, von Ebner, ’02), may vary in size (Kallius, ’05), may show mitosis {Hermann), and are broad and connected

Perigemmal

fibers

Taste pore

Taste rods

Fig. 139. Taste bud on the tongue of a hedgehog. Boeke.

344 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

by protoplasmic strands with the sustentacular cells. As their name imphes, they lie at the base of the taste buds, but the other cells extend through the bud with spaces between them. The sustentacular cells form layers, several cells thick, about them and between them. The sustentacular or supporting cells are of two types, an outer or tegmental cell, broader, with a centrally placed nucleus and cuticular process, and an inner or rod-shaped cell, narrower, with a basally situated nucleus and no cuticular process {Hermann, ’88). The neuroepithelial or gustatory cells are narrow, spindle-shaped cells, with the nucleus in the thickest portion, and with a smaller, sometimes branched inner part, and an outer part having the diplosome and ending in a stiff, dehcate process which extends into the gustatory canal but not to the outer pore. Terminal fibers of the gustatory nerves end in the taste buds in varicose, richly branching fibers. They terminate around the gustatory cells. Some observers, such as Boeke (’26), beheved this termination to be partly within the cells, but others regarded the relation as one of contact. Some fibers terminate on sustentacular cells. Certain fibers within the bud are termed intragemmal fibers ; other fibers of the same nerve, terminating between the buds, are called intergemmal (perigemmal) fibers {Retzius, ’92; fig. 139). It is to be emphasized that the gustatory cells are in no sense to be regarded as nerve cells, for they do not give rise to nerve fibers. They are supporting elements for nerve terminations, and by certain observers {Hermann, ’85 ; Kolmer, ’10) are believed to be concerned to some extent in the reception of gustatory stimuli. Herrick (’25) described a third system of fibers in relation to the taste buds in Amblystoma. These fibers, which he termed circumgemmal fibers, are arranged in dense coils around the palatine taste buds, such a coil encircling a bud at the region of its greatest diameter. According to this observer, the fibers described by him are similar to those seen by Strong (’95, fig. 16) in the frog and by von Lenhossek (’93, fig. 3 ; ’94, fig. 12) in the conger eel. Herrick was unable to trace the fibers forming this plexus to their source. He discussed various possibilities : (1) that they might be collaterals of gustatory fibers, the main stems of which terminate as intragemmal fibers ; (2) that they might be another type of sensory fibers (presumably tactile over the trigeminal through its anastomosis mth the palatine plexus) ; (3) that they might be postganghonic neurons forming a sympathetic plexus around the taste buds. The possible reinforcing action of these circumgemmal fibers was discussed.

Variations in distribution of the taste buds in different vertebrates are of great importance. Although originally of entodermal origin {Johnston, ’10, for amphibians), they become widespread and in some forms occupy large ectodermal areas. Taste buds are lacking in invertebrates and in Amphioxus. In the larvae of cyclostomes {Retzius, ’93 ; Johnston, ’02, '05 ; Wiedersheim, ’09) they are found only in the pharynx and the gill cavities, but in the fully grown animal they occur on the outer skin also. So far as is known, they are confined to the mouth and pharynx in selachians {Op-pel, ’00; Sheldon, ’09), but in ganoids and teleosts they are found on the external surface of the head and in many teleosts over the body as well {Herrick, ’01, ’02, ’03a, ’08, etc. ; Landacre, ’07). This latter

TASTE AND THE GENERAL AFFERENT SYSTEMS

345

condition leads to an enormous increase in the number of taste buds {Johnston, ’01 ; Herrick, ’00, ’01, 07), so that in certain fishes, such as the siluroids, the buds may number approximately a hundred thousand. Taste buds on the outer surface of the head and body are always innervated by a branch of the facial nerve (ner\Tis recurrens facialis). In animals above fishes taste buds are not found on the surface of the body. In amphibians, particularly the tailed variety, the number of such buds in the mouth and pharynx is very large. A tongue makes its first appearance in these forms and taste buds are found on it {Hartmann, ’QZ] Relzius, ‘92, ’05 ; 0ppel,’00; Johnston, ’10). The relative number of those situated on the tongue increases in general in higher forms, until the greater number of the taste buds is found on this organ. However, taste is reduced in the terrestrial animal, in reptiles and land amphibians as well as in higher forms. Thus in the snake the tongue is a tactile organ, although numerous taste buds are found along its posterior margin as well as on the palate. The tongue of the alligator is said to be provided with but few taste buds {Bath, ’06). The taste buds are found principally in the pharynx and on the choana of these animals. According to Tuckerman (’92), a fairly large number of taste buds is present on the anterior as well as the posterior half of the turtle tongue. A similar condition has been described for saurians by Leydig (’72) and by Merkel (’92). The greatest atrophy of gustatory sensibility is to be found in birds. The horny tongues of these animals are provided with only a few taste buds ; such as are present occur at the root of the tongue, on the palate near the choana, along the pharynx, and on the posterior side of the. epiglottis. In certain birds a few buds are found also along the margin of the lower jaw. On the whole, the number of taste buds is very small in birds, usually from forty to sixty; an exception to this is found in parrots, which, according to Bath (’06), may have as many as 400.

The real development of taste buds as the special sense organs of the tongue occurs in mammals, due probably to the fact that they are the only animals which chew their food. Although in most mammals there are buds on the palate, on the pharynx, on the posterior side of the epiglottis, and even occasionally on the larynx, the number of these, in comparison with those on the tongue, is exceedingly small. They are present in the human infant on the fungiform papillae, but disappear very early from these papillae (except for an occasional bud) and from the inner side of the cheeks {Stahr, ’02), and .remain only on the phylogenetically younger papillae, the circumvallates, and upon the epiglottis.

The number of taste buds (the numbers here listed are not to be regarded as exact numbers) varies considerably in different forms. According to Boulton (’83a ; see also ’83 and ’83b) they number about 10,000 in the larger marsupials, while Tuckerman (’89a) estimated that the small bat has about 800; the squirrel, 4000 to 6000 ; the hare, 900 ; the rabbit, 17,000 ; the pig and goat, 15,000; the sheep, 10,000; and the bullock, up to 35,000. An adult man is said to have about 9000 taste buds. It is to be remembered that the tongue is an exploratory organ in mammals and that taste serves it in carrying out this

346 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

function. The part of taste in such activities finds expression in the arrangement of the bulbar taste centers (see p. 355).

In view of the fact that fishes in general have such enormous numbers of taste buds, it is strange that the cetaceans — which are also aquatic animals — should possess so very few. Rawitz (’03) found that taste buds are lacking in the cetacean, Delphinus delphus. These animals have no olfactory nerves either. This probably is in accordance with the law of Dollo, which states that if the function of an organ has once changed from one t3q)e (in this case perception in a fluid medium) to another (terrestrial life adaptations) it can never reassmne the first function (the aquatic adaptation).

The question as to whether or not taste buds can regenerate in the adult animal has been a matter of considerable interest to anatomists and physiologists. Griffini (’87), after removal of circumvallate and foliate papillae in dogs and rabbits, obtained in the region huge taste-bud bearing fungiform papillae, except in a case where the cut was made very close to the surface, in which case normal foliate papillae appeared. Drasch (’87), Hermann (’88), and Stahr (’02) foimd mitotic figures in taste buds, but the last mentioned observer did not believe that the whole bud regenerates. On the contrary, Patzelt (’23) favored the idea that such regeneration is occurring constantly and Whiteside (’26) appears to have shown that in rabbits fully formed circumvallate and foliate papillae reappear after they have been entirely destroyed, that in this redevelopment they follow, as far as circumstances permit, the normal course of development, and that in their epithelium normally developed taste buds are foimd. To what extent the presence of the proper innervation affects the ability of the epithelium to regenerate taste buds has been (and possibly by some observers is still regarded as) a somewhat disputed question. The work of Olmsted (’20 and ’20a) on Ameiurus appears to indicate that, in that fish at least, the innervation is necessary for the regeneration of the taste buds, for when the end of the barbel is cut the barbel and taste buds regenerate if the facial nerve is intact but fail to regenerate if it is severed. Harrison (’04) has drawn attention to the fact that degeneration of taste buds in amphibians after severance of the nerve supplying them cannot be regarded necessarily as evidence that regeneration will be brought about if the nerve is intact. He has emphasized the fact that this proves no more than that the nerve exerts a trophic influence over the cells about which it terminates. Patzelt (’23) believed that taste buds in mammals develop in relation with stratified pavement epithelium, and occasionally in relation with stratified ciliated columnar epithelium, and that they may develop in regions which gustatory fibers have not supplied and into which they grow secondarily in order to supply the newly-formed taste buds. In a paper reported at the 1928 meeting of the American Association of Anatomists, Whiteside stated that in the rat bilateral section of the glossopharyngeal nerve caused the degeneration of taste buds on the circumvallate papillae, that no buds were present four v’eeks after the operation, and that none had reappeared four months after the operation, although many buds regenerated if either glossopharyngeal nerve was present. The foliate papillae, which are innervmted in the rat by both facial and glosso

TASTE AND THE GENERAL AFFERENT SYSTEMS

347

pliarjTigeal nerves, showed generally a complete degeneration of taste buds on the posterior fold and a variable amount of degeneration of taste buds on the anterior fold after four weeks. Four months after the operation the number had increased greatly over that at four weeks. It is to be inferred that this regeneration was due to the presence of an intact facial nerve.

Chemical Sense and Taste

Certain points with regard to the physiology of taste and the differences between gustatory and chemical sensibility must be reviewed briefly. It has been known for some time that cutaneous sensory nerves carry chemical sensibility from the skin regions and that this sensibility is lost as the skin becomes cornifled. In an animal such as a young cyclostome, a shark, or a frog, the skin is sensitive to salts, alkalies, acids, and bitter substances, but not to sugar. When the recurrens branch of the facial is cut, which supplies the taste buds on the surface of the body and head of such fishes as siluroids, it is possible to demonstrate clearly that chemical sensibility, mediated by spinal nerves, is still present in the skin of these animals. Animals in general, and particularly fishes, tend to react most strongly to the above-mentioned chemical substances when they are applied to the head ; the tail and extremities are less sensitive, and the trunk the least so. This chemical sensibility is believed to be due to the stimulation of free endings of nerves, rather than to the excitation of some specialized nerve termination. Certain facts support this view. In the first place, chemical sensibility is present in animals in which there are no specialized endings. Furthermore, chemical sensibility is best developed in those regions where there is the greatest richness of so-called free sensory endings. The question might be raised as to the possibility that the nervous impulse produced by a drop of acid or other chemical might be due to tactile stimulation. Van Wayenburg (’97) has demonstrated rather clearly that such is probably not the case. This observer was able to point out that fluctuations in the reflexes produced by chemical stimuli were proportional to the concentration of the chemical substance, when the tactile stimulation was unchanged. In his work on frogs he was the first to show that the sensitivity of the skin to chemical stimuh follows the laws of Weber and Fechner, and in this respect is analogous to other sense qualities. "When the tactile or temperature stimuli remain comparatively constant, a geometrical gradation of the chemical stimuli is accompanied by an arithmetical gradation of the reflexes.

The work of von Anrep (’80), Sheldon (’09), and Parker (’12) indicates very strongly that the free sensory endings capable of stimulation by proper chemicals are not the same as those which carry general tactile sensibility, since application of cocaine causes first the loss of tactile sensibility and later that of chemical sensibility. A similar test of gustatory and tactile sensibility indicates that the former is the first to disappear. This affords evidence, then, that different types of sensation are under consideration when chemical and gustatory sensibilities are tested, even though the substances ^ used as stimuli and their con1 This does not apply to sugar, which does not produce chemical sensibility.

348 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

centrations may be identical and although it is the negative ion (not the metallic one) which in each case determines the stimulation.

A brief review of certain of their physiologic characteristics emphasizes even more clearly the differences between gustatory and chemical sensibility. Thus van Wayenburg (’97) and Herrick (’02 and ’03) have both shown that reflexes brought out by the chemical sense are always negative ; i.e. are defense reactions. This is in harmony with the teachings of Sherrington (’00), who regarded the free endings in the skin as having a nocireceptive function. The reflexes brought out by stimulation of the taste buds are entirely different. C. J. Herrick (’03) performed a series of very careful experiments upon various teleosts. His results showed that those animals in which the skin of the body is supplied with taste buds (as in the siluroids, cyprinoids, and Gadida) react positively toward proper food substances brought into contact with the skin, while fishes (such as Prionotus and Opsanus) which have no taste buds on the surface of the body either do not react to the stimulus or swim away from the tactile or chemical stimulus thus received, giving in this case the avoiding reaction. Parker (’08) confirmed these observations. He found that although after cutting the nerve supply to the taste buds of the body the fishes still give the characteristic avoiding reaction to various salty, bitter, and acid substances applied to the body surface, they do not give a positive reaction to food substances. Yet feeding experiments have brought out a fact of fundamental significance, that the undisturbed existence of taste buds alone is frequently not suflRcient to produce the seeking reaction. The positive reaction toward food suffers a pronounced interruption when those nerves are resected which provide a particular area of taste buds with tactile fibers, so that the correlation of these two types of sensation clearly plays an important part in determining the animal’s ability to recognize and to locate the proffered nourishment (see p. 361). This observation is an exceedingly important one. This correlation between gustatory and tactile sensibility at the periphery finds expression in the anatomy of the central nervous system through the relative location of the taste and tactile centers of the tongue and their secondary connections, particularly where the two types of sensation pass in from a given area over different nerves.

Now the reader is invited to turn to a consideration of the sensory components of the branchial nerves, to introduce which a brief review of the nerves in Amphioxus is presented.

The Peripheral and Central Relations of the Afferent Components OF the Vagus, Glossopharyngeal, and Facial Nerves

AMPHIOXUS ^

Vestibular and lateral line organs are not present in Amphioxus. Consequently a motive for specialization and hypertrophy of exteroceptive branches is lacking, as it is also for visceral sensory branches in the absence of true gill arches and taste buds. As a result, the dorsal roots of the region do not attain

^ For the bibliography for this section on Amphioxus, reference is made to Chapter II, pp. 292 to 294.

TASTE AND THE GENERAL AFFERENT SYSTEMS 349

a size comparable to that of similar roots in craniotes. The lateral exversion of the sensory medulla oblongata, which occurs in craniotes under the influence of these hj^pertrophied dorsal roots and which causes the formation of the fourth ventricle, is not e\ddent in these forms. The presence of welldeveloped cutaneous components and of corresponding but unconnected ventral occipital roots makes so marked the resemblance between the peribranchial nerves and the dorsal roots of the spinal nerves in Amphioxus, that the term branchio-spinal seems particularlj' applicable to these cephalic, mixed dorsal roots. Since up to the present time it has not been possible to differentiate either structurally or functionally between the nerves of the medulla oblongata and those of the spinal cord, the nerves in this animal are numbered in a single series from the head to the tail. In this enumeration, the nerve entering the ventral side of the brain, in front of the infundibulum (fig. 53) is termed the first nerve (Nervnis I or nervus terminalis). This is the nervus apicis of van Wijhe (’14). According to Kutchin (’13), this nerve is purely somatic sensory; according to Langerhans (’76) and van Wijhe (’93), peripheral sympathetic ganglia lie in association with it. Possibly it is the nervus terminalis of higher forms. It is analogous to the somatic sensorj’’ fibers which enter the brain dorsally, behind the infundibulum. Its entrance on the ventral side is due to the fact that the dorsal sensory portion of the nervous system, the alar plate of the medullary tube, extends cephalad beyond the ventral motor part and consequently all of the most cephalic portions of the brain, dorsal and ventral alike, are derivatives of the dorsal sensorj’- or alar plate of the medullary tube.

The next ner\'e, Nervus II of various authors, enters the brain on the dorsal side, caudal to the infundibulum. It is purely general somatic sensory. It supplies, as does the pre\’ious nei^-^e, the skin of the rostrum and a part of the so-called dorsal fin. As a rule it is unequally developed on the two sides due to the asymmetri-^ of Amphioxus. Most authors consider that it has no somatic efferent fibers, but Ayers (’21a) described somatic efferent fibers in it to the first myotome. It carries neither visceral afferent nor visceral efferent fibers. Its point of entrance to the brain is in front of the second myotome. For this reason it is to be considered as the homologue of the most cephalic dorsal nen^e of craniotes ; that is, the nervus ophthalmicus profundus trigemini, which emerges in a similar position with reference to the second myotome {van Wijhe)}

Nervus III of Amphioxus emerges beh nd the second myotome. It is the representative of the radix maxillo-mandibularis trigemini of higher forms. It contains, in this animal, somatic or cutaneous sensory and visceral sensory fibers, but no special visceral motor as in craniotes, because jaw musculature is lacking in Amphioxus.

The roots entering dorsally, caudal to Nervus HI, are about thirty-six in number. Tliey supply the caudal and peribranchial regions and are to be

® With the exception of nervus terminalis, nemis ophthalmicus profundus is the only interseptal nerx'e entering in front of the second myotome. The nemis mesencephalicus and thalaraicus, described by Tretjakoff (’09) for Ammocoetes, and by Mesdag (’09) for birds, is not constant. It represents, in all probability (Ariens Rappers), simply a remnant of the ophthalmicus profundus which has not followed the caudal migration of the main nen'e mass.

350 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

regarded as forerunners of the branchial nerves of craniotes lying caudal to the trigeminal. They resemble the preceding nerve, but carry, even in Amphioxus, many visceral efferent fibers to those striated peribranchial muscles concerned in breathing (musculi transversi). In craniotes a reduction of the caudal part of the branchial region occurs ; consequently the more posterior branchio-spinal nerves either disappear or become greatly reduced. Such fibers as remain probably become concerned in the innervation of the gut, serving as sensory and preganglionic components. In the most cephalic portion of the branchial region, with the appearance of active gill breathing, there is an increase in the visceral components of those branchial nerves situated directly behind the pseudotrigeminus of Amphioxus. These components become progressively larger with the greater development of the mucous membrane and musculature of the gill region and form the facial, glossopharyngeal, and vagus nerves of true vertebrates.

How many dorsal roots of Amphioxus are used in the formation of each of these nerves in the different craniotes is at present uncertain. There is a constantly progressive atrophy of branchial arches. Thus there are 13 in myxinoids, 8 in Petromyzon, 7 in Heptanchus, 6 in Hexanchus, and 5 in other selachians. In Amphioxus the branchial region may be very large and perhaps covers the whole preatrial repon, i.e. about half of the so-called cord. All the dorsal roots of this region would then correspond to the branchial nerves of fishes, which is quite possible since the line of emergence of the vagus is in direct continuation with that of the dorsal roots of the cord.

There is a pronounced tendency for the pseudo-trigeminus of Amphioxus and the trigeminus of craniotes to play a major role in the somatic sensory innervation of the head. All visceral sensibility, as far as it is represented in cranial nerves, is carried by the homologues of the facial, glossopharyngeal, and vagus nerves, while somatic sensory components in higher forms are either very small or entirely absent (see p. 340). Thus there is a distinct functional demarcation on the sensory side between the first branchial nerve and the succeeding ones. Consequently, for convenience in description, they are subdivided in the following account into two groups ; an anterior group represented by the trigeminal, and a posterior group represented by the facial, glossopharyngeal, and vagus nerves.

CYCLOSTOMES

In cyclostomes the somatic sensory fibers of the vagus and glossopharyngeus have retained their dorsal as well as their ventral branches, and consequently distribute to the dorsal and ventral parts of lateral areas in the repon of the gill arches (Johnston, ’05, ’06). Sensory branches of the facial are found on the surface of the head caudal to the orbit and beneath it. The dorsal branch is lacking (Johnston, ’05, ’06). The trigeminus sends sensory branches to the region previously supplied by this dorsal part of the facial.^ Visceral sensory fibers

The visceral sensory branches of the facia] innervate the anterior half of the first gill pouch,

those of the glossopharyngeal, the inner part of the posterior half of the first gilt pouch and the inner part of the anterior half of the second, while the vagus supplies fibers to the remaining gill pouches.

352 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

PLAGIOSTOMES

True branchiomeric nerves have increased considerably in plagiostomes as compared with cyclostomes. They still carry the sensory components which were found in cyclostomes, but the reduction of somatic sensory fibers has been carried further and the vagus and glossopharyngeus have only dorsal skin branches {Ewart, ’89; Cole, ’96; Merrill Hawkes, ’06). Likewise the somatic sensory branch of the facial is still further reduced and may even be lacking in some plagiostomes. The somatic sensory branch is still present in Heptanchus and Hexanchus, although its peripheral relations have not been studied thoroughly. Visceral and somatic sensory components, however, have been recognized because of their central relations (Ariens Kappers, ’14). Merrill Hawkes

found small branches of the facial passing to the skin in the ventrolateral region of the head iw CWamydosftlachws. Iw figure 141A the relations of the vagus are shown. A figure taken through the level of the glossopharyngeal will present a very similar picture. It is possible in these forms to demonstrate more clearly than in cyclostomes that the root is divided into three portions in the medulla oblongata; the more lateral portion passes "^0 the descending (or spinal) root of the trigeminal and the intermediate portion 0 the visceral sensory area, while the medial portion arises from the visceral fferent nucleus. For a short distance the somatic sensory root of the vagus is [istinct from (although parallel to) the descending root of the trigeminal. This ondition is indicated in the figure (fig. 141 A; somatic sens, root X). The [escending fibers of the visceral afferent division of the vagus form a small but distinct postvagal fasciculus solitarius. This primary fasciculus solitarius increases slightly in size farther caudalward, but is never very large in plagiostomes. There are gustatory fibers as well as general visceral afferent fibers in the vagus of plagiostomes, but it is believed that this postvagal portion of the fasciculus solitarius is concerned chiefly with fibers of general sensibility {Ariens Kappers, ’14; see also ’06).

Sensory fibers entering by the facial nerve distribute according to the components which they carry, either to the nucleus of the descending root of the trigeminal with which they are in close relation caudally (fig. 141B), or to the visceral sensory area dorsomedial to this descending root. This visceral sensory area forms an eminence on the ventricular floor (fig. 136). The visceral sensory fibers are of finer caliber but they outnumber the somatic sensory and terminate not only at the level of entrance, but also in the gray of the visceral sensory

Fig. 141. A. The various components of the vagus nerve in Heptanchus.

B. The various components of the facial nerve in Heptanohus. Note the position of the cutaneous branch (Hautaat des VII) of the facial in the vicinity of the descending root of the trigeminus.

TASTE AND THE GENERAL AFFERENT SYSTEMS

353

column as far back as the point of entrance of the glossopharyngeal. Some of them teiminate directly in a bundle associated with the column — the prevagal part of the fasciculus solitarius — which extends caudalward to the level of the vagus, where it becomes continuous with the postvagal portion of the fasciculus. This prevagal portion is sufficiently well developed to be visible macroscopically in properly prepared specimens, and the associated visceral gray is sufficiently great in amount to form eminences on the floor of the ventricle as far back as the calamus scriptorius (see fig. 136).

Behind this level, the colmnn is visible microscopically as a mass of gray with associated fibers. This gray fuses over the top of the ventricle to form the unpaired nucleus of the commissura infima. In plagiostomes it is possible to differentiate to some extent between general and special visceral afferent fibers once they have entered the medulla oblongata. Most of the special visceral afferent or gustatory fibers terminate in the visceral sensory area indicated by the pearl-like eminences on the ventricular floor ; the general visceral afferents appear to be represented chiefly in the short, longitudinally running tract, the forerunner of the fasciculus solitarius of higher vertebrates {Ariens Kappers ; see ’06 and ’14).

From the visceral afferent column of the medulla oblongata and from the nucleus of the commissura infima, fibers are sent to the neighboring efferent nuclei of the facial, glossopharyngeal, and vagus nerves, and apparently, after crossing, to more caudal centers, although the destination of these latter is not known with certainty. Provision is thus made for local and for aboral reflexes. The efferent centers of these nerves in plagiostomes are to be considered later, but there are several points concerning them which need consideration in relation to the sensory branches. In these animals the efferent fibers emerge medioventralward from, but in close relation with, the afferent roots. The nuclei of origin of the efferent fibers of the facial, glossopharyngeal, and vagus nerves form a practically continuous coliunn of cells. This relationship differs from that in cyclostomes, in which the motor nuclei of the glossopharyngeal and vagus are situated considerably caudal to the motor nucleus of the facial, which latter nucleus lies at the point of entrance of its fibers. Its migration caudally in plagiostomes is a consequence of the greater development of the more caudal portions of the visceral sensory area and is considered to be due to neurobiotaxis.

GANOIDS AND TELEOSTS

The reduction of the somatic sensory components of the facial, glossopharyngeal, and vagal nerves, which is more pronounced in plagiostomes than in cyclostomes, has gone still farther in ganoids and teleosts, in consequence of the development of an operculum in these latter animals. IVith certain exceptions, the cutaneous sensory branches of both facial and glossopharyngeal nerves are lacking in teleosts. Thus, in Amia and in Lepidosteus (Norris and Eupkes, ’20) and in Albula (van der Horst, ’27), the facial root has a cutaneous branch which may be large, as in Albula, and which, curiously enough, does not run with the descending root of the trigeminal but more laterally, ending in the richly

354 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

developed gelatinous substance of the spinal cord. Only the dorsal cutaneous branch of the vagus, which innervates the occipital region of the head and the upper part of the operculum, is pre.sent in all ganoids and teleosts. The trigeminal nerve supplies the rest of the head with cutaneous branches.

The special visceral afferent fibers or gustatory fibers reach a very high degree of development, particularly in some of the teleosts, and in many of these latter animals extend not only over the head, but even over the trunk. Taste buds occur on the surface of the head even in cyclostomes, but they are relatively few in number and it is not until ganoids are reached in the phylogenetic scale that there is any widespread distribution of taste buds over the body. Such a distribution reaches its maximum in the teleosts and particularly in the cyprinoids and siluroids, where the head and body have many taste buds. This increase in

the number and the distribution of the taste buds is associated with a hypertrophy of the visceral sensory roots of the facial nerve, since all such peripheral buds transmit their impulses over the recurrent branch of the facial, for the facial nerve carries fibers from the anterior part of the mouth (ectodermal portion) and from the skin, while the vagus and glossophar 3 mgeal nerves supply taste buds to the mucous membrane of the back part of the mouth and associated areas. The former supplies external surfaces ; the latter supply visceral surfaces. An increase in taste buds at the sirnface of the body will not produce hypertrophy of the vagal and glossopharyngeal nerves in teleosts. Occasionally, however, there is an increase (in cyprinoids, particularly carp-like forms) in the taste buds on the back part of the mouth and on the palatal organ. Such an increase leads to an enlargement of the visceral afferent roots of the glossopharyngeal and vagus nerves. Ganoids and teleosts show no marked increase of general visceral sensibility over that found in plagiostomes. Consequently the hypertrophy of the three nerves just discussed must be due almost entirely to the increase in taste buds.

The nucleus of termination of visceral sensory fibers of the facial, glossopharyngeal, and vagus nerves in ganoids such as Acipenser and Polyodon is a mass of gray extending from the level of entrance of the sensory facial root caudalward to the calamus scriptorius region. To this area Johnston (’01) applied the name of lobus vagi, for the' nucleus is largest at the level of entrance of this nerve. There are relatively few cells in this nucleus, and the number of secondary fibers is small. A secondary gustatory tract (secondary vagus tract of Johnston), poorly medullated in Polyodon and unmedullated in Acipenser, has been followed forward to a secondary gustatory nucleus situated in the isthmus region caudal to the nucleus isthmii and ventral and lateral to the nucleus lateralis cerebelli {Johnston, ’01 ; Hocke Hoogenboom, ’29). Of course root fibers of the facial, glossopharyngeal, and vagus end not only in relation to the vagal

Kio. 142. The distribution of the gustatory fibers of the facial nerve in the skin of Ameiurus melas. C. J. Herrick.

TASTE AND THE GENERAL AFFERENT SYSTEMS 355

lobe or visceral sensory area, but take part in the formation of a fasciculus solitarius, which is well developed in these forms.

Essentially similar relations are foimd in the dipnoan, Ceratodus {Holmgren and van der Horst, ’25), where the roots of the nerves in question enter the visceral sensory area of the medulla oblongata. A fasciculus solitarius is present, having a well-developed prevagal position, which reaches its greatest size at about the level of entrance of the cephalic sensory roots of the vagus and then narrows down as it is followed caudalward. It comes into relation with the nucleus of the commissura infima, where the fibers end. Either root or secondary visceral fibers cross in the visceral portion of the commissure. Holmgren and van der Horst (’25) carried fibers forward, in Ceratodus, in what they believed to be an ascending secondary gustatory tract, although they were unable to trace it to its termination.

The wide distribution of taste buds in many teleosts leads, of necessity, to a wide distribution of the recurrent or communis branch of the facial nerve, and carries this nerve into territory which is ordinarily supplied by spinal nerves and by the trigeminal nerve for general cutaneous sensibility. Centrally, the increase in the taste fibers is reflected in the enormous enlargement of the visceral sensory areas of the teleostean medulla oblongata (fig. 143). This enlargement indicates that the local centers of the medulla oblongata are more particularly concerned with gustatory impulses. The following brief account of the relations of gustatory centers in teleosts is based particularly on the fundamental work of Herrick (’99, ’02, ’05, ’06, ’07, ’08, and elsewhere) and on the observations of Berkelbach van der Sprenkel (’15) and of Aliens Kappers (’09, ’14). Purposely, emphasis has been placed on the account of those teleosts in which the gustatory centers are hypertrophied, as in siluroid^ for example. Special visceral afferent fibers and general visceral afferent fibers enter the medulla oblongata from the cranial ganglia situated on the roots of the facial, glossopharyngeal, and vagus. The neuraxes enter the enlarged visceral afferent column, which is called, at the level of the facial nerve, the facial lobe and, at the level of the glossopharyngeal and vagus nerves, the vagal lobe. Many of the taste fibers are believed to terminate directly in this central gray.

The vagal lobe is very highly differentiated in certain fishes and detailed accounts of it, in examples of this type, have been given by Mayser (’81) and by Herrick (’05). In the carp, where the vagal lobes are greatly enlarged, Herrick described the following layers : (1) a peripheral sheet of incoming sensory root fibers, which break up into small fascicles and enter (2) the layer of secondary neurons (2nd layer of Mayser) ; (3) the secondary gustatory fibers (3rd layer of Mayser) which are followed by (4) the so-called motor layer (4th layer of Mayser), consisting really of cells of origin for motor root fibers of IX and X. The layer of secondary neurons can be subdi\'ided further into seven bands. Of these, the more peripherally placed cells, large, pale-staining ones, give rise, at least mainly, to the long secondary gustatory tracts. The other neurons probably diffuse incoming stimuli, serving thus as intrinsic neurons, or connect wifh near-lying motor centers of the medulla oblongata. The motor layer of certain cyprinoids is replaced in siluroids by the nucleus intermedius vagi.

35G NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

In cyprinoids the facial lobe is relatively simpler than the vagal, and may fuse with its fellow across the midline, forming the so-called tuberculum impar (fig. 145). In .siluroids a fusion across the midline does not occur (fig. 1433, 146). The central part of the lobe has a great number of small neurons in peculiar “rosette” arrangement. The neurons giving rise to the secondary gustatory tracts are similar in appearance to the homologous neurons of the vagal lobe. Incoming facial fibers, after entrance to the medulla oblongata, swing caudalward as the prevagal fasciculus solitarius and enter the lobe on its lateral side. There is no well-differentiated motor layer such as is found in the vagal lobe, but at the level where the lobe goes over into the remainder of the medulla oblongata, there is a differentiated nuclear region, the nucleus intermedius facialis of Herrick. The small lobe between the facial and vagal lobes of cyprinoids {Haller, ’96) is

Fig. 143. A. Dorsal and lateral views of the brain of Carpiodc.svelifer. Herrick. B. Dorsal view of a brain of a siluroid, Malapterus electricus.

supplied by the glossopharyngeal nerve and, therefore, may be called the lobus glossopharyngei {Herrick, ’05). However, it receives root fibers of the facial and vagus as well as those of the glossopharyngeal nerve.

In addition to visceral sensory fibers, direct root fibers from the descending root of the trigeminal pass into the deeper portion of the facial lobe. Thus are correlated tactile and gustatory impulses, probably from homologous body regions. From this area, discharge is made to the nucleus intermedius facialis, which in turn sends the impulses to the motor centers of the medulla oblongata. Thus from both vagal and facial lobes, impulses are sent by secondary neurons to motor nuclei of the same side (through the neuraxes of smaller neurons of the lobes) and of the opposite side (through neuraxes of secondary gustatory neurons). These connections are for reflex movements of barbels, mouth parts, gills, and pharynx. From the facial lobe (in siluroids and cyprinoids) and from the vagal lobe, a path passes fonvard, made up of the neuraxes of the secondary gustatory cells. This is the secondary ascending gustatory tract of Herrick. Its course is peculiar (figs. 144 to 148). It runs ventralward from its origin and ascends ventral and medial to the descending root of the trigeminal, sometimes almost completely inclosing the root, to the entrance of the sensory trigeminal root. There it turns dorsomedialward and thus reaches the superior secondary gusta

TASTE AND THE GENERAL AFFERENT SYSTEMS

357

torj* niiclcu‘5, where it terminates largely on the same side but partly in the nucleus of the other side after eiossing in the intrinsic commissure of these nuclei. This superior sccondaiy gustatory nucleus (fig. 147) is situated in the dorsal isthmus region. It is in front of the motor nucleus of the trigeminal and is in synaptic relation with this nucleus through the dendrites of motor neurons of the trigeminal nerve. There is a suggestion that there may be a similar relation with eye muscle nuclei, or at least with the trochlear nucleus, and there are known to be connections with the vahmla cerebelli and cerebellum, and with the reticular

Capj*ular fibrrs

Somatic root X

Tig 14 1. The vnnoiis components of the vnciis nerve in Tmc.o, a cj-prinoid.

formation. The secondary gustatory nucleus consists of small intrinsic neurons and larger ncr\'c cells. The neuraxes of the latter form, in part, the chief tertiary gustatory tract to the inferior lobes. This is heavily medullated. Burr (’28) was able to identify the secondary ascending gustatory tract in Orthagoriscus mola, but was unable to identify definitely its nucleus of termination, which quite presumablj' may be small in this fish, where taste is not so highly developed. In addition to its termination in the superior secondary gustatory nucleus, Brickner (’30) stated that certain bundles swing forward, ventral to the nucleus. He believed that such rostral bundles reach the tectum. Evidence to be published later (Pearson and Barnard) ^ indicates that such rostral fibers in large part pass to the hypothalamus.

» Dissertations, Ijiboratoo’ of Comparative Neurology, The University of Michigan.

358 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The facial gustatory centers are related not only to the tactile centers of the head, but also to centers receiving tactile sensations from the body. This correlation finds expression in the so-called descending secondary gustatory tract. This tract has a distinct pars facialis only in cyprinoids and siluroids {Herrick, ’05). It arises from both the facial and vagal lobes ; but the tract, in the carp for example, forms a large medullated bundle from the facial lobe, while the descending fibers from the vagal lobe are unmedullated and mostly scattered. It has been described {Herrick) as arising from the more cephalic part of the facial lobe, turning ventrolateralward with the fibers of the ascending system, and running caudalward, dorsal and medial to the descending root of the trigeminal nerve. For the most part, the tract terminates in the so-called inferior secondary gustatory nucleus. Some of its fibers continue caudalward in the ventrolateral tracts of the cord. From the vagal lobe there are faintly medullated scattered fiber bundles passing caudalward to the nucleus commissurae infimae and the inferior gustatory nucleus.

The inferior secondary gustatory nucleus, Herrick (’05 ; or lateral funicular nucleus, ’06) is reticular gray which is continuous frontally with the vagal lobe but which shows a particular enlargement at the level of the funicular nuclei. The secondary inferior gustatory nucleus is a center of correlation for tactile stimuli (from the funicular nuclei and descending root of the trigeminal) and gustatory and possibly general visceral sensations (from the facial and vagal lobes by way of the secondary descending gustatory path). It is large in fishes with highly developed gustatory sensibility ; very small in those in which taste is poorly developed. In his 1908 paper Herrick mentioned terminations of the secondary descending visceral tract within both lateral and median funicular nuclei in the carp and in Ameiurus, and spoke of these nuclei as gustatorytactile correlation centers. In the carp he was able to demonstrate, from the median funicular nucleus and the reticular gray ventral to it, a path (tractus funiculo-ambiguus) which decussates in the somatic commissura infima and then runs forward to the motor centers of the glossopharyngeal and vagal nerves (the nucleus ambiguus of Herrick), which supply the gill muscles. It is of interest that this descending secondary gustatory path arises largely from the facial lobe, that is, from that nucleus which receives gustatory fibers from the cervical regions of the body. It makes possible vuthin a single nucleus, a central correlation of gustatory and tactile impulses from a given peripheral area.

The glossopharyngeal and vagal lobe is connected with another nucleus which comes into relation with the more caudal centers. However, in all probability, this is concerned with general visceral sensation, and the nucleus with which it connects is the nucleus of the commissura infima (fig. 147, n. comm.). This nucleus receives not only the fibers from the more posterior roots of the vagus, which have practically no gustatory fibers, but also fibers from the glossopharyngeal and vagus nucleus, which is analogous vfith the nucleus intermedius of the facial. From the nucleus commissurae infimae fibers are sent to nearIjnng motor centers. Herrick (’08) described a somatic commissura infima

TASTE AND THE GENERAL AFFERENT SYSTEMS

359

(ventral to the visceral commissura infima in fishes) which is chiefly, although not exclusively, a fiber band related to the funicular nuclei.

The relations just described do not exist in all fishes. Herrick (’07) made clear that while the peripheral distribution of taste buds and their facial branches

nucleus of the glossopharyngeus; racdialward, the tuberculum impar of the sensory root of the Vllth.

in Gadus (codfish) is essentially the same as in Ameiurus, the former show different central relations. In the codfish there is no facial lobe which is capable of clear definition. The incoming visceral sensory fibers of the facial end largely in the lateral part of the vagal lobe. The median portion of the vagal lobe (the median lobule), separated from the lateral lobule (Sens, lobe of VII, fig. 148) by a longitudinal septum, receives primarily the gustatory fibers entering over the

360 NERVOUS SYSTEMS OF ^TERTEBRATES AND OF ^LA.N

glossopharyngeal and vagus nerv^es. Herrick suggested that the fibers of the facial which reach the median lobule may be those distributing to the inside of the mouth, and that the median lobule functions as a visceral, the lateral as a somatic center. The median lobule of the codfish, which is continuous caudalward with the visceral nucleus commissurae infimae, has most of the connections tj^jical of the vagal lobe of fishes. The ascending secondary gustatory tract comes almost entirely from this lobule, little if any from the lateral. However, no part of the vagal lobe has such an evident descending secondary gustatory tract as that found in many other fishes. The secondarj’^ connections of the lateral lobule are

Fia. 1-10. The Riistntory nucleus (A'urZ. VII sens.) and the pustntory path (tr. gust, ant.) of a siluroid, Ariiis. ran dcr Horst.

not those tj^tical of tlie facial lobe of other fishes. This lobule extends directly into the funicular region and is connected with that region by' a thinly mcdullatcd tract which, collecting on the ventrolateral region of the lobule, extends directly back to the funicular nucleus. This is probably' the homologue of the Ixittcr developed .secondary descending gustatory' tract described for certain other fishes. From the lateral lobule throughout its extent lic.'ivily' mcdullatcd fibers run to the region of the ventral commissure and distribute to the ventral funicular arca-s of tiic same and oppo.'iitc sides. This is a direct connection to motor centers. It i.s suege-^ted that this short circuiting to motor centers is n.s.socintcd with the more active life and with the nipid and precise movements of Gadus ns compared with those of .\meiurus. In general it may' be said that in Ameiurus there i.s a central correlation, through secondary .ascending and dc.«ccnding gu-statory' path.*?, with

361

TASTE AND THE GENERAL AFFERENT SYSTEMS

tactile centers, and then a final common path to motor centers. In the cod the major connection is directly to the motor centers, for responses to tactile and gustatory impulses, the final common path being represented in the efferent tract.

The above discussion leads to two important conclusions :

(1) Those systems which are related from the standpoint of peripheral stimulation will show the results of such a relationship in their central connections.

This will hold true whether they supply areas which adjoin those supplied by analogous nerves, as is the case of the gustatory branches to the mucous membrane supplied by facial, glossopharyngeal, and vagus nerves, or whether they cooperate peripherally with nerves bearing different stimuli in the innervation of an area.

(2) It is in accordance with physiologic facts that the cooperation of the perceptions of taste and touch, although not necessary to, is very favorable for the accomplishment of the feeding reflexes. Thus the efferent tracts to motor

Fig. 147. Gustatory nuclei and paths in a cyprinoid. Schematic drawing. Herrick (’05). (A^F., nuclei funiculi.)

Sens root of VII

Sens. lobe of VII ( Sens lobe of IX aud X

Fig. 148. Contralateral relations of the sensory nuclei of the Vllth, IXth, and Xth nerves in the cod. Secondary gustatory tract {Second gust, tr.) from the sensorj' nuclei of the IXth and Xth, and a descending path {Crossed desc. fibers) from the facial lobe.

regions take their origins from the gustatory and tactile correlation centers and represent final common paths in the sense of Sherrington (’06 ; see p. 128).

In most teleosts, the interrelations between gustatory sensations and tactile sensations brought in over the trigeminal nerve provide the most important

362 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

central correlations. In Gadida the anterior fins serve as organs of exploration for the animal. They are innervated by the facial nerve for taste and by the cutaneous branches of the cervical nerves for tactile sensibility. The importance of the correlation of these two types of sensation is seen in some slight modification of the central pattern. Thus in siluroids and c3T3rinoids, in which the trigeminal-facial correlations are of great importance, the facial gustatory center lies chiefly cephalad to the lobe of the glossopharyngeal and the vagus and receives mainly trigeminal collaterals, while in Gadida the sensory facial nucleus extends backward along the other side of the glossopharyngeal and vagal lobe, the entering facial fibers passing in part to the more cephalic portions of the facial lobe and to the glossopharyngeal and vagal lobe for correlations of gustatory sensibility with the sensations from the mucous membrane of the gill arch region. The lateral facial nucleus represents mainly a nucleus of termination for gustatory fibers from the fins.

AMPHIBIANS

There is a marked decrease in the number of taste buds in amphibians as compared with the number in teleosts, and this is reflected in the more simple arrangement of the central visceral centers and their connections in the former animals. The conditions are even more simple than in the shark, but in general — and particularly so in the tailed amphibians, such as the salamander — the central pattern of the visceral afferent components of the vagus, glossopharyngeal, and facial nerves of amphibians bears considerable resemblance to that of the homologous components in plagiostomes. An even greater resemblance is seen between the distribution of the general somatic sensory components in these two forms.

The relations of the visceral centers of the brain for various amphibians have been the subject of much study. The present knowledge of them is based on the investigations of Osborn (’88), Strong (’95), Kingsbury (’95), Norris (’09 and ’13), Herrick (’14, ’25, and ’30), Rothig (’13 and ’27), and others. The following account treats of the relations in Necturus, Amblystoma, and the frog. The visceral afferent column is medial to the sulcus acusticus (fig. -149).

Several of the more cephalic roots of the vagus carry visceral afferent fibers in Necturus, but the visceral sensory components of the glossopharyngeal enter by a single root in this animal {Herrick, ’30). In the frog the vagus has only two main roots {Ariens Kappers and Hammer, ’18), differing in this respect from certain other forms such as urodeles and particularly most sharks, where this nerve enters by many roots. After entrance to the medulla oblongata, the cutaneous branches separate off from the rest of the fibers in the usual manner and pass to the nucleus of the descending root of the trigeminal (fig. 150). The remainder of the nerve runs dorsalward and the visceral afferent components thus reach the visceral region on the floor of the medulla oblongata. In part, the visceral afferent or sensory fibers terminate in relationship w'ith the gray of this area at the level of entrance of the nerve. The major portion of the fibers, however, bends caudalward and forms a conspicuous fasciculus solitarius which

TASTE AND THE GENERAL AFFERENT SYSTEMS

363

runs caudalward, medial and dorsal to the descending root of the trigeminal. This postvagal fasciculus solitarius contains also descending root fibers of the glossopharyngeal nerve, which, like the vagus, sends fibers in part to the visceral gray at its level of entrance and in part into the fasciculus solitarius. The gray associated with the fasciculus solitarius is usually termed the nucleus of that

Fig. 149. Dorsal view of the medulla oblongata of Eana mugiens (catesbiana). Wax reconstruction by Hammer. The dotted line gives the line of attachment of the choroid plexus.

fasciculus. The postvagal fasciculus solitarius is highly developed in amphibians and is comparable in its relations to the homologous tract of higher vertebrates (fig. 150). Caudalward it becomes remarkably deficient in medullated fibers (fig. 150A) and then shifts farther and farther dorsalward. It decussates in part in the commissura infima, fibers reaching the homolateral and the contralateral nuclei of the commissure (Wallenberg, ’07). The postvagal fasciculus solitarius in the frog extends to the level of the second or third spinal segment (see Wallen

364 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

herg). In its course through the medulla oblongata the bundle is accompanied on its medial side by a column of gray, the visceral afferent column or the nucleus of the fasciculus solitarius, which is particularly well developed at the level of entrance of the glossopharyngeal nerve, and which receives collaterals and medul

Fig. 150. A and B. Two cross sections through the level of the vagus in Rana mugiens; above (A) the more caudal, below (B) the more frontal.

lated stem fibers throughout its extent. Due to the termination of the medullated fibers of the fasciculus solitarius in this associated gray, the bundle becomes progressively smaller in its passage caudalward. The visceral sensory components of the glossopharyngeal and vagus nerves present essentially the same pattern in the tailed amphibians as has been described for the frog. The visceral

TASTE AND THE GENERAL AFFERENT SYSTEMS 365

component of the commissura infima, formed by crossing fibers of the fasciculi solitarii, is composed of very small, unmedullated fibers apparently few in number in Necturus {Herrick, ’30).

The visceral sensory root of the facial nerve enters the brain immediately dorsal to the acoustic root in Necturus {Herrick, ’30). A similar relation exists in the larval Amblystoma {Herrick, ’14), but in the adult Amblystoma {Coghill, ’02) the visceral sensory fibers of the facial enter between the two roots of the acoustic nerve, while in the frog they are chiefly ventral to the acoustic nerve in both the tadpole {Strong, ’95) and the adult {Black, ’17).

The sensory portion of the facial is much larger in tailed amphibians than in the frog. In urodele amphibians {Norris, ’09 ; Norris and Buckley, ’ll) at least, it carries cutaneous branches which, after entrance to the brain, join the descending fibers of the trigeminal. In this the amphibians resemble the plagiostomes.® The visceral afferent fibers turn caudalward in both tailed and tailless forms, and in part reach the levels of entrance of visceral sensory fibers of the glossopharyngeal and the vagus nerves ; there they terminate in relation with the associated gray which may show an enlargement in the vagal region {Herrick, larval Amblystoma, ’14). In part they synapse in the gray along the fasciculus. In larval Amblystoma He^r^c^ described a bifurcation of the incoming visceral sensory fibers. After such bifurcation, one branch ascended to form a prefacial fasciculus solitarius. In Necturus {Herrick, ’30), after entrance to the medulla oblongata, the heavily medullated visceral afferent fibers swing medialward and caudalward into the fasciculus solitarius. On their entrance to this fasciculus, certain (but not all) of the fibers bifurcate and the larger branch continues caudalward in the postfacial part of the fasciculus, while the smaller, unmedullated branches form the prefacial fasciculus solitarius, which can be traced forward to a plane cephalad to the trigeminal roots. Earlier, Herrick (’14) had described a similar prefacial fasciculus solitarius in Amblystoma. In these animals it extends forward dorsal to the trigeminal root fibers and reaches almost to the cephalic end of the auricular lobe, where it terminates, in the larva, in an ill-defined area internal to the nuclei of the trigeminal and acoustic nerves, and, in the adult, in a region of neuropil rostral to the motor nucleus of the trigeminal. Slightly in front of and lateral to this area of termination of the prefacial fasciculus solitarius in the isthmus region is the superior secondary gustatory nucleus {Herrick, ’14). The descending branches of the visceral afferent fibers of the facial run spinalward in both Amblystoma and Necturus, uniting with similar components of the glossopharyngeal and the vagus nerves to form the main fasciculus solitarius. In adult Amblystoma and adult Cryptobranchus alleghaniensis {Herrick, ’14), a few fascicles of the sensory root of the trigeminal either join the fasciculus solitarius or run at the side of it. A similar connection of the lingual

® Cutaneous sensory fibers are found in teleosts in the trigeminus and vagus. In both the plagiostomes and amphibians, cutaneous branches occur not only in trigeminal and vagal roots, but also in those of the glossopharyngeal and facial nerves {Norris, ’09). Neither of these latter groups has an operculum such as is characteristic of teleosts, and it appears reasonable to suppose that the disappearance of cutaneous fibers in the teleostean facial and glossopharyngeal is associated with the appearance of an opercular cover for the jaws.

366 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

branch of the trigeminal nerve has been described by Wallenberg (’97, see mammals) for man. As yet it has not been possible to distinguish in amphibians between those central fibers which carry taste and those which carry general visceral sensibility. It is probable, however, that the portions of the fasciculus solitarius which extend down to the lower end of the medulla oblongata and into the cord are concerned mainly with general visceral sensations.

The secondary connections of the visceral sensory column are not as completely known for amphibians as for teleosts. In both Necturus and Amblystoma Herrick was able to trace uncrossed neuraxes from gray associated with the fasciculus solitarius (post-facial part) into the ascending secondary nucleus {Herrick, ’14, ’30). Other bundles run forward in the cerebral peduncle. A tertiary gustatory tract from the superior secondary visceral nucleus swings ventralward, lateralward, and forward to join this secondary gustatory tract. Herrick followed the bundle thus formed to levels through the superficial origin of the oculomotor, where it was lost among fiber bundles of the region. This secondary gustatory nucleus has not been identified in all amphibians and it seems possible that it may not be as prominent in the tailless as in the tailed forms, but Larsell (’23) found such a center in the frog and identified its typical connections. From the visceral sensory column, in tailed amphibians and also in the frog, there are fibers which decussate in the region below the fasciculus longitudinalis medialis and then pass caudalward; these suggest the descending secondary gustatory tract of bony fishes. Other secondary connections in the region of the nucleus of the fasciculus soUtarius are established by means of crossed and uncrossed fibers.

On the whole, the ascending and descending gustatory tracts appear to be less well developed in amphibians than in many bony fishes, while the postvagal portion of the fasciculus solitarius is much better developed in the former animals, or at least in the tailless amphibians. The reasons imderlying the greater development of the postvagal portions are not entirely understood. Amphibians do not appear to exhibit any special differentiation of taste which might call for such an increase in the visceral tract. It seems more probable that this postvagal portion increases with the greater development of general visceral sensibility. The facts which justify, but do not entirely confirm, such a point of view are as follows : (1) In the first place, the postvagal portion of the fasciculus solitarius is associated with the commissura infima and its nucleus. This nucleus, even in fishes, is a significant correlation center for visceral afferent and somatic afferent impulses. In tailless amphibians its visceral afferent components (presumably general visceral afferent) have increased, and this increase is associated with an increase in the cutaneous fibers of the trigeminal, facial, and vagus nerves, which run caudalward in the descending root of the trigeminal nerve, terminate in its nucleus, and then are connected by short fibers with the nucleus of the commissura infima. (2) In the second place, the terrestrial life of these amphibians introduces respiration by means of lungs. These organs send afferent impulses to the central nervous system over both cranial and spinal nerves. Impulses passing in over the vagus nerve will enter the medulla oblongata and

TASTE AND THE GENERAL AFFERENT SYSTEMS 367

become correlated there with visceral afferent impulses from mouth, throat, and tongue. They will also pass by way of the postvagal fasciculus solitarius to the upper levels of the cord. In its course caudalward, this postvagal portion of the fasciculus solitarius lies at the base of the dorsal median septum {Wallenberg, 07), near the base of the dorsal horn of the spinal cord, and not far from the preganglionic centers of that region. Either directly or by way of secondary neurons from the nucleus of the commissura infuna, impulses are carried over to such preganglionic centers and pass out by way of preganglionic fibers to synapse on sympathetic neurons, which in turn innervate smooth muscle associated with the bronchi of the lung, and so make possible respiratory reflexes.^

Other factors may also play a role in increasing the caudal extension of the visceral afferent fibers. One such probable factor is the increase in importance of the correlations between cutaneous fibers of the trigeminal and the visceral afferent fibers of the facial and glossopharyngeal in amphibians, for in these animals a muscular tongue with sensory innervation from the trigeminal, facial, and glossopharyngeal nerves first appears. Such correlation might well take place at the caudal end of the descending root of the trigeminal nerve, and if it did so, would inevitably lead to an increase in the number of descending fibers of the postvagal part of the fasciculus solitarius. In birds, where the number of gustatory fibers is extremely small, the postvagal portion of the fasciculus solitarius is very large, suggesting again that it is concerned in general rather than in special visceral sensibility (see section on birds, p. 371).

Localization of centers and consequent marked specificity of secondary tracts are lacking in amphibians, and perhaps particularly in the tailed forms. Consequently it is not surprising that cells located in the acoustico-lateral area of the medulla oblongata and giving rise to the general bulbar lemniscus system of Necturus {Herrick, ’30 ; the bulbo-tectal tract of Amblystoma, Herrick, ’14) should receive impulses by way of their dendrites, not only from terminals of the sensory trigeminal root, but also from the fasciculus solitarius (see page 387 for a further discussion of this point), thus carrying visceral sensibility along with other types of impulses to the optic tectum.

REPTILES

Considerable differences in detail with respect to the arrangement and development of the visceral afferent system in reptiles depend upon whether the animal considered is a turtle, a crocodile, a snake, or a lizard. The following description is based primarily on crocodiles and lizards because the relations of the visceral afferent systems are somewhat better understood there.

Consideration of figure 151 shows clearly that in the crocodile, as compared with the teleost, there is relatively a greater development of the somatic afferent and a less development of the visceral afferent systems. Scarcely any eminence is formed on the floor of the fourth ventricle by the gray of the visceral afferent or visceral sensory area. The visceral afferent fibers of the glossopharyngeal

’’ A true diaphragm is not present, but the pericardium in amphibians receives fibers from the fourth nen'e (Furbringer).

368 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and vagus nerves terminate in part in the visceral afferent area at their level of entrance (see fig. 152). In part, the visceral components of these nerves turn directly caudalward in the fasciculus solitarius. Secondary fibers arise from

the visceral sensory area and course medialward into the region of the medial longitudinal fasciculus, but it has not been possible as yet to trace them beyond that level. The postvagal part of the fasciculus solitarius decreases considerably as it runs caudalward. The visceral sensory areas of the two sides unite over the ventricle to form a nucleus commissurae infimae, and the fasciculus, after a partial decussation in the commissura infima, terminates in this nucleus. Certain of the fibers may continue into the spinal cord but their final destination is not known.

The visceral afferent or visceral sensory fibers of the facial root, which are fewer in number in the alligator than in a lizard such as Varanus, enter at the level of the vestibular nerve (fig. 154) . Below the floor of the fourth ventricle the fibers turn caudalward to become continuous with the remainder of the fasciculus solitarius. These roots of the facial form the so-called prevagal portion of the fasciculus. In Varanus the more posterior roots of the vagus have been forced dorsalward by the fairly large descending root of the trigeminal nerve, so that the most caudal fibers run across from the dorsal to the ventral side (X. S., fig. 153). Such roots give off cutaneous fibers in the usual manner to the descending root of the trigeminal. While the greater portion of the visceral root fibers pass directly into the postvagal portion of the fasciculus solitarius, the remainder synapse in the visceral sensory area at the level of entrance of the nerve. This visceral sensory area in Varanus and in Alligator is divided into a dorsal nucleus and a medial nucleus, the latter lying beside the raph6 {Ariens Kappers). This medial nucleus separates the efferent nucleus of the vagus from the medial longitudinal fasciculus (fig. 153). The medial nucleus is present in both Caiman and Varanus, but is relatively smaller in the former animal. The sensory or afferent facial root in Varanus is separated from the efferent root by vestibular root fibers, in consequence of which it enters farther dorsally than

Fra. 152. The components of the vagus in Caiman sklerope.

—Habenula

—IV nerve — V nerve

-—Poet, cercb. fieaure — Deac. root of VIII V>Deac. root of V

Fig. 151. Dorsal view of the brain stem of Crocodilus porosus. Notice the compression of the ventricle due to the strong development of the somatic sensory region (and especially the descending root of the trigeminal).

Pars impar of mesenceph "*j Optic tectum — Cerebellum-*.

VIII nerve ,,

TASTE AND THE GENERAL AFFERENT SYSTEMS

369

■VDore. Duel. X f-'Sol fasc., med. part

—Secondary Bbers of X — Med. nucl. X

is usual (fig. 154). On reaching the ventricle it turns caudalward and joins the fasciculus solitarius at the level of the glossopharyngeal nerve, but soon exhausts ® itself in the gray that accompanies the bundle. In both

types of reptiles the descending \

fasciculus solitarius is larger than \

in amphibians. It is particularly / \ d

well developed in Caiman, although / -Sol fasc., med. part

the taste buds are relatively much hu

less in number in this reptile than in ( Jjp Secondary fibera of X

the frog. This, too, offers confirma- ^

tion of the notion that the postvagal \ portion of the fasciculus solitarius I is concerned with visceral sensibility V

rather than ivith taste. It is prob- \

able that the centers for taste are \

located in reptiles, as in lower ver- \

tebrates, vdthin the medulla oblon- \.

gata itself. Presumably both the j,

dorsal and medial visceral sensory

nuclei are concerned in this func- The compojrns^of the vagus in Varanus

tion. This medial sensory nucleus,

the more cephalic end of which is between the medial longitudinal fasciculus and the efferent nucleus of the vagus, increases in size as it extends caudalward and is still fairly well developed at the level of the conunissura infima. At the

Fio. 153. The components of the vagus in Varanus Salvator.

Fig. 154. The entrance of the sensory and motor roots of the facial in Varanus Salvator.

level at which the hypoglossal nucleus makes its appearance, this medial visceral sensory nucleus lies between the nucleus of the hypoglossal and the efferent nucleus of the vagus, and is often termed the nucleus intercalatus (Aliens Kappers, ’14; for Allen’s and Du Bois’ interpretations of the mammahan nucleus intercalatus reference is made to p. 377). It may be stated in passing that the efferent nucleus of the vagus lies farther lateral and dorsal in reptiles than in sharks, and consequently in closer

‘ Beccari (’14) believed that the sensory VII fibers dichotomize, one branch passing to the cerebellum. This, however, needs further confirmation.

370 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

relation with the dorsal nucleus of the visceral sensory column and also with the nucleus of the descending root of the trigeminal nerve. The glossopharyngeal nucleus is united with the nucleus of the facial nerve in both Caiman and Varanus.

In the former, the nuclear mass is situated relatively far dorsally and frontally ; in the latter, it is farther ventral and caudal. The cells of origin for efferent fibers to the tongue are more prominent in reptiles, but are still continuous with the ventromedial column of the ventral horn of the spinal cord and lie below rather than above the dorsal arcuate fibers.

A

BIRDS

Dora. lat. nucl. X

B

Birds furnish very instructive material for an estimate of the value of sensory impulses in determining the pattern of the medulla oblongata (Bo/c, ’15). In these animals taste is very atrophic. This is particularly true of regions supplied by the facial nerve. As a consequence, the afferent or sensory root of the facial is very poorly developed. The glossopharyngeus and (particularly)' the vagus

Fig. 155. A. The vagus root, t.ic fasciculus soliiarius, the Carry larger afferent roots dorsomedial gray, and its a.s.sociated net in Ca.suarius. The section t.h.nn Hops the facial. Their is rather far caudal. Besides the medial longitudinal fa.sciculus , . i

three cells of the Xllth nucleus are shown. ’ roots enter the Clorsai

B. The glossopharyngeal root and fasciculus solitarius, with fibers to the dorsomedial fiber net in Casuarius. This section is frontal to that of fig. 155 A.

part of the medulla oblongata and, after entrance, run directly medialward

through the upper third of the descending root of the trigeminal nerve. During

their course they contribute cutaneous branches to this root and its nucleus.

The visceral fibers from the mucous membrane, on approaching the floor of the medulla oblongata, end to some extent in scattered gray of the region, but the

TASTE AND THE GENERAL AFFERENT SYSTEMS

371

majority of the fibers enter the fasciculus solitarius and pass caudalward in that bundle. The postvagal portion of the fasciculus solitarius is generally very well developed in birds. Its high degree of development in the cassowary is to be seen in figmes 155 A and B. In the sparrow Ramon y Cajal (’09) distinguished two fasciculi solitarii on each side, of which the dorsolateral one received crossed fibers, particularly from the glossopharyngeal nerve. However, at more caudal levels the double bundles disappear. The poor development of taste in these animals makes it impossible to consider that these large fasciculi solitarii are associated here with this special sense. Rather, it would seem that such development is associated with an increased number of general visceral sensory fibers in the bird. A re\dew of mammalian conditions confirms this point of view. Thus a comparison of two sections, one through the medulla oblongata of the cassowary (fig. 155) and the other through a corresponding region from a rabbit brain, indicates that, although the two sections are approximately of tlje same size, the fasciculus solitarius in the bird is very much larger than in the rabbit (fig. 158 A). Since the number of taste buds in the bird as compared with the rabbit is about 100 to 17,000, it follows that a direct relationship between the size of the fasciculus solitarius and the degree of development of gustatory sensibility is not probable.

About three-fourths of the fasciculus solitarius of birds decussates in the commissura infima and terminates in its associated gray. The remaining fourth passes directly caudalward and is lost to view in the cervical cord {Brandis, ’94). Although the great majority of the afferent root fibers of the vagus and glossopharyngeus can be found in the fasciculus solitarius, a much smaller portion of the visceral sensory fibers of these nerves follows a different course. Frequently, after having accompanied the fasciculus for a short distance, they turn medialward and end in the dorsolateral visceral sensory area of the medulla oblongata (fig. 155 ; the homologue of the dorsal visceral nucleus of the crocodile), which is much smaller in birds than in mammals. Others enter nucleus dorsomedialis, where they form a network which is particularly characteristic in Casuarius, but much less so and sometimes scarcely to be found in certain other birds (Spheniscus, Colymbus, Ciconia). This nucleus, in Casuarius, as in Varanus, extends to the level of the commissura infima. At those levels where the hypoglossal nucleus is distinguishable, the dorsomedial nucleus extends over the hjqjoglossal cells and lies between them and the efferent nucleus of the vagus and glossopharyngeal (fig. 155). It forms, then, a primitive nucleus intercalatus, which is regarded here as gustatory {Aliens Kappers, ’14 ; for other interpretations of nucleus intercalatus see the discussion of Allen’s work on p. 377).

Since only from 40 to 60 taste buds are present in most birds, and since the majority of these lie in the region innervated by the glossopharyngeal nerve, it is evident that the sensory facial component will be smaller and that the most of the fibers which it carries will of necessity be concerned with general visceral sensory rather than gustatory sensibility. It follows, then, that these animals show a remarkably strong development of the postvagal portion of the fasciculus solitarius, that this is associated with an increase in general visceral sensibility.

372 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and that it is not associated with taste, which in these forms is considerably atrophied.

There is no need at the present juncture to consider the motor nuclei in birds, since a discussion of them is given in Chapter IV. Mention, however, may be made of the fact that, with the poor development of taste and consequently of the more caudal portions of the visceral sensory area of the facial, the efferent nuclei of the facial retain their primitive position at the level of entrance of the root or even migrate frontalward. They do not pass caudalward to join the motor (or efferent) nucleus of the glossopharyngeal as is the case in reptiles. The tendency toward frontal migration is due to the increased importance of the interrelation of the motor nuclei of the facial and trigeminal nerves, this being due to the peripheral relations of these two nerves, the facial, for example, supplying the posterior belly of the digastric while the trigeminal supplies the anterior belly of this muscle. This is another example of the importance of neurobiotaxis in determining the position of the nuclei of origin and of termination of the cranial nerves.

MAMMALS

Visceral sensory fibers enter the medulla oblongata of mammals by way of the vagal, glossopharyngeal, and facial nerves. The fibers terminate in part at the level of entrance of the respective nerves in the gray of the visceral afferent column associated with the fasciculus solitarius. There are said to be no ascending fibers. The majority of the fibers turn caudalward. The portion of the fasciculus solitarius in front of the level of entrance of the glossopharyngeal nerve is small in mammals, particularly in the highest manunals (man), but the postglossopharyngeal and postvagal portion of the fasciculus is larger. This postglossopharyngeal and postvagal portion consists chiefly of glossopharyngeal fibers {Bechterew, ’99, and Brun, ’12), but also carries many fibers of the vagus (von Kolliker, ’89 to ’02 ; Bruce, ’98 ; Kosaka and Yagita, ’05 ; Stuurman, ’13). In mammals about three-fourths of the fasciculus crosses in the commissura infima and ends on the contralateral side in the nucleus associated with this commissure, the so-called nucleus commissurae infimae, or the nucleus of Cajal. The remaining one-fourth descends into the cervical cord. The final termination of this fasciculus is not known with certainty. In the mouse and the cat, fibers are present caudal to the nuclei of the posterior funiculi (Ramdn y Cajal, ’09). It is possible that they reach as far as the second cervical segment.

A column of gray accompanies fasciculus solitarius throughout its course in the medulla oblongata. The larger portion of this lies on the medial side and is known as the nucleus of the fasciculus solitarius. The gray matter on its ventrolateral side is sometimes termed the nucleus parasolitarius or the nucleus of Kohnstamm and Wolfstein (’07). The work of Tumbelaka (’16) suggests that this latter nucleus, in man at least, may have connections with the thalamus, for this observer found that an injury to the middle third of the ventrolateral nucleus of the thalamus produced degeneration of the large cells of the nucleus parasolitarius in man. A somewhat similar result was obtained by von

TASTE AND THE GENERAL AFFERENT SYSTEMS 373

Monakow (’85). Nucleus parasolitarius is regarded by some observers as concerned in the regulation of the movements of respiration, since it has connections with the lateral reticular formation, which in turn gives descending fibers, mostly crossed, to the cord for innervation of motor centers supplying the diaphragm and the intercostal muscles. This is the so-called fasciculus Thomasi {Thomas, ’99 ; Lewandowsky , ’04 ; Kohnslanun and Wolf stein, ’07). Rothmann’s experiments

(’02) indicated that this tract is found in the ventrolateral funiculus of the cord, in so far as it is concerned in the innervation of the diaphragm. The fibers concerned with the respiratory centers of the thorax are said to reach the lateral part of the ventral funiculus. In the dog Kosaka and Yagita (’05) demonstrated the presence of a fasciculus solitario-spinalis from the gray surrounding the fasciculus solitarius to the centers of respiration of the spinal cord. The fibers run through the fasciculus longitudinalis medialis and the piedorsal or medial tecto-spinal bundles, decussate in part, and become continuous with the longitudinal bundle of the ventral funiculus of the cord and hence distribute to the phrenic nucleus

374 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and to the ventromedial column of the ventral horns (fig, 157). This bundle extends at least as far caudalward as the fourth cervical segment and is undoubtedly concerned with carrying impulses to the nucleus of the phrenic nerve, from which, in turn, they are sent to the diaphragm. It is probable that it also reaches the motor centers for muscles to the ribs. This connection was confirmed for the rabbit by Hirose (’07, fig. 157). A diagram involving the use of a solitario-spinal tract for man is to be found in such standard texts of neurology as those of Ramdn y Cajal (’09), Herrick (in various editions), and Ranson (’31). Allen (’27 ; see also p. 377) believed that impulses from the gray associated with fasciculus solitarius relay in reticular gray before being discharged to the efferent neurons (see also Rasmussen, ’32, p. 41).

Since the development of the postvagal portion of the fasciculus solitarius is associated with respiration (along with other fundamental reflexes), this fasciculus becomes particularly important as respiration by means of lungs appears. The visceral afferent fibers concerned in respiration include not only the sensory branches from bronchi and lungs but also fibers from associated organs such as the mouth, larynx, and trachea. Thus, not only is the vagus nerve involved, but the lingual branch of the trigeminal and the branches of the facial and the glossopharyngeal, carrying impulses which have their effect on respiration, as the experiments of van Melle (’00), using the method of Lahorde for artificial respiration by tongue traction, indicate. This association with respiration and other equally important relations of the fasciculus solitarius suggest that its postvagal and postglossopharyngeal portion, at least, is concerned, as in lower vertebrates, in carrying general sensations from visceral surfaces. Such general visceral sensibility is very large in manamals. Since the posterior part of the tongue and the pharynx region are supplied for general visceral sensibility by the glossopharyngeal nerve, and sensations from the remaining viscera from the pharynx to about the region of the descending colon travel over the vagus, it is evident that the number of fibers so concerned must be very large and their representation within the brain, as embodied in the fasciculus solitarius, relatively very well developed.

Gustatory fibers in mammals pass to the medulla oblongata from cells in the geniculate ganglion of the facial, the petrosal gangUon of the glossopharyngeal, and the nodosal ganglion of the vagus nerve. In some higher mammals — and certainly in man — the greater number of these fibers enter over the glossopharyngeal nerve, which supplies the taste buds located on the circumvallate papillae and the foliate folds of the tongue. The gustatory fibers in the vagus are relatively few in number, being confined to the epiglottis region. In many mammals a considerable number of gustatory fibers enter through the facial nerve from taste buds in the fungiform papillae of the tongue, but in man taste buds which are present during development later almost or entirely disappear from the fungiform papillae. Consequently in man the number of gustatory fibers in the facial nerve is very small.

Not all of the fibers of the vagus nerve and the glossopharyngeal nerve enter the fasciculus solitarius. Certain fibers appear to run on the medial side of

TASTE AND THE GENERAL AFFERENT SYSTEMS

375

the fasciculus to end in the dorsolateral gray of the visceral sensory column (fig. loSA). This nucleus is considered as a caudal continuation of the nucleus triangularis of the vestibular, but Sala (’93) has pointed out that on its medial aspect it receives also certain root fibers of the glossopharyngeal and anterior vagal roots. A few sensory fibers {Brun, ’12; Fuse, ’13a) run in a medial di

Fig. 157. The ascending (above) and the descending (below) degenerations following a lesion of the fasciculus solitarius and the regions immediately adjoining it, according to Hirose. Operated side to the right.

Marchi preparation.

rection and appear to enter the nucleus intercalatus of Staderini (’94 to ’07 ; Luna ’10). In their course to this nucleus the visceral sensory fibers in mammals pass ventral to the dorsal efferent nucleus of the vagus. This is in contrast to their course in reptiles and birds, where they are found dorsal to the nucleus, between it and the floor of the ventricle. This shift in position is due not only to a phylogenetic change in these fibers, but to a shift lateralward of the efferent nucleus in question (consult fig. 155 and fig. 158). Up to the present, the distribution of the visceral sensory impulses brought in by

376 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the vagus to the nucleus intercalatus has been observed most readily in the edentates, in Tamandua, and especially in the ant-eater, Myrmecophaga jubata, but it is also very distinct in the rabbit {Ariens Kappers). The fact that the nuclear mass is largest in mammals is perhaps an argument in favor of its importance as a gustatory center in these forms, particularly as regards taste from the tongue. There is certainly evidence that it is larger in such animals as rodents and edentates, in which the tongue is well provided with taste buds, than in the dolphin, where the gustatory sensibility is almost entirely lacking. The

absence of a nucleus intercalatus in fishes might be explained on the ground that these animals have no real tongue. The position of the nucleus in mammals between the dorsal efferent nucleus of the vagus and the motor nucleus of the hypoglossal is favorable for reflexes involving gustatory impulses. Nucleus intercalatus, and in general the gray of the medial visceral sensory column, extends very far caudalward in these animals and in mammals in general. The nucleus intercalatus is not only intercalated between the dorsal efferent nucleus of the vagus and the efferent nucleus of the hypoglossal, but also extends medialward into the region of the so-called nucleus funiculi teretis (see Chapter IV), and covers the nucleus hypoglossus like a cap. This medial extension permits the nuclei intercalati of the two sides to come into relationship with each other in the posterior regions of the medulla oblongata and to exchange fibers. The frontalward extension of this medial area unites with the dorsal sensory nucleus of the glossopharyngeal and vagus and thus emphasizes their functional relationship. Internuclear fibers from this dorsal nucleus to the nucleus intercalatus, which were present in the alligator, have also been demonstrated in mammals {Ariem Kappers, ’14 and ’20, and Fuse, ’13a). DuBois (’29), however, was unable to find them in the opossum. According to Ariens Kappers, both the dorsal visceral sensorj' nucleus and the nucleus intercalatus are gustatory, the latter serving as a secondary rather than a primary gustatory center, a point of view with which the results of Berkelbach van der Sprenkel (’24) are in accord. Fuse (’14) described gustatory fibers to the nucleus intercalatus in man under tlie name of the fascicu

Fio. 158. A. Vagus root, fasciculus solitarius, the dorsolateral gray, the dorsal efferent nucleus of the Xth, the nucleus of Staderini, and the nucleus of the Xllth. Rabbit.

B. The relations of fibers of the third trigeminal root, especially the ramus lingualis V, to the fasciculus solitarius in man. Wallenberg.

TASTE AND THE GENERAL AFFERENT SYSTEMS 377

lus triangularis intercalatus. Allen (’23, ’23a), in his experimental work on the visceral sensory centers in the medulla oblongata of the guinea pig, was unable to trace gustatory fibers to the nucleus intercalatus, and regarded it as unrelated to the visceral afferent system. He considered the possibility of its being a vestibular center. DuBois thought it was not gustatory because of its position caudal to the entrance of glossopharyngeal and vagus roots (as Allen had emphasized) and because of its intimate relation with the medial vestibular nucleus. He made two suggestions regarding the function of this center : (1) that it may be considered a proprioceptive center for the tongue, and (2) that it may also receive proprioceptive fibers from the pharynx and larynx. General somatic afferent fibers carrjdng exteroceptive impulses are present in the vagus and possibly in the glossopharyngeal nerve. Such cutaneous fibers of the vagus constitute its ramus auricularis, which distributes to the external ear and which has its cells of origin in the jugular ganglion. Neuraxes for the cell bodies in this ganghon enter the medulla oblongata with other sensory fibers of the vagus but distribute to the nucleus of the descending root of the trigeminal, which is the nucleus of termination for other exteroceptive fibers of cranial nerves. A few cutaneous fibers distributing to the external ear are believed to be components of the glossopharyngeal nerve (Herrick, ’31), with cells of origin in the ganghon superius (or superior part of the petrosal ganghon) and a nucleus of termination in the nucleus of the descending root of the trigeminal nerve. A satisfactory demonstration of proprioceptive fibers in the glossopharyngeal and vagus nerves has not been given.

The facial nerve carries both special and general \dsceral sensory fibers in mammals, although, as was stated previously (p. 374), in adult man there are practicahy no taste buds on the anterior two-thirds of the tongue, though such buds appear during development. Consequently the number of gustatory fibers in the hmnan facial nerve is smah. Fibers carrying sensory impulses in this nerve arise, according to Ramsay Hunt (’09), from portions of the external, middle, and internal ear, from the Eustachian tube, and from the mastoid cells. Some observers are of the opinion that sensory fibers are also present in the facial from the anterior two-thirds of the tongue (Cushing, ’03 ; Oppenheim, ’13), and that such enable this organ to retain in part its sensibihty to tactile stimuh even after extirpation of the trigeminal.® From their peripheral distribution, these latter fibers, as well as those from the external ear, must be classed as somatic afferent, while those from the entodermal surfaces of the middle and internal ear and the Eustachian tube are general visceral afferents. Certainly visceral sensibihty from the submaxillary and sublingual salivary glands (Huber, ’96), and possibly from the parotid (see Bremer, ’27), passes centraiward over the chorda tympani. The visceral afferent fibers entering by way of the facial nerve ter ® According to Wallenberg (’97), the lingual branch of the trigeminal enters fasciculus solitarius. This observer regarded this as proof that taste fibers run in the trigeminal nerve, a point of view which later experimental and particularly clinical work appears to have disproved. If such distribution of the lingual branch does occur, it would appear that it was to interrelate centrally the somatic with the visceral sensibility of the region.

378 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

minate in the gray associated with the fasciculus sohtarius and in part extend caudalward as far as the visceral sensory center of the glossopharyngeal nerve. Probably the gustatory fibers terminate near their level of entrance.

The results of comparative neurological studies indicate that the fasciculus sohtarius, and particularly its postvagal portion, probably develops primarily with the development of general visceral sensibility, while the relatively prominent development of taste, as in fishes, is associated with definite development of the bulbar nuclei and not with an increase in descending fibers, as represented in the fasciculus sohtarius. In fact this fasciculus, particularly its postvagal portion, is small in those animals where the taste is best developed. Consequently in mammals it is to be expected that the visceral sensory centers at the level of entrance of the facial, glossopharyngeal, and vagus nerves are to be regarded as mainly associated with taste, while the fasciculus sohtarius, and particularly its more posterior portion, is associated with visceral sensibihty.

General somatic afferent fibers of proprioceptive t3q)e have been described in the facial nerve (Davis, ’23) . Such fibers, after passing along the motor nerves, are said to enter the medulla oblongata with the sensory roots. Wakeley and Edgeworth, ’33, J. Anat. vol. 67, located their cells of origin in the geniculate ganglion of the facial nerve.

Discussion of the Peripheral Taste Paths in Relation to the Trigeminus

The question as to which nerves carried gustatory fibers has been given extensive consideration and there is as yet no unanimity of opinion regarding this matter. All observers are agreed that the posterior part of the mouth and the pharynx region receive their gustatory innervation from the glossopharyngeal and vagus nerves. In man taste buds on the epiglottis send their impulses over fibers of the vagus, and those on the circumvallate papillae over the glossopharyngeal nerve. The anterior two-thirds of the tongue has practically no taste buds in adult man, although many are present in that region in other mammals. There is a gradually increasing number of observers who regard the gustatory fibers from the anterior two-thirds of the tongue as terminal branches of the facial nerve. Some still attribute such innervation to the trigeminal nerve. The following arguments appear to favor in all animals the innervation of the anterior two-thirds of the tongue for gustatory impulses through the facial nerve :

(1) In fishes the increase in taste buds in the mouth and the under surface of the body is associated with a marked increase in size of the afferent root of the facial and of its visceral centers within the medulla oblongata.

(2) The increase in size of the proximal gustatory area, which occurs in frogs, is always coupled with the increase in size of the sensory facial root, and never with the enlargement of the sensorj" trigeminal root.

(3) The atrophy of the proximal gustatory area is always associated with an atrophy of the sensory facial root. In birds, where taste buds are almost entirely lacking on the anterior two-thirds of the tongue, the chorda tjmipani and the intramedullarj’- sensory root of the facial are extremely small. In most birds the trigeminal nen.’e is very well developed.

TASTE AND THE GENERAL AFFERENT SYSTEMS

379

(4) Those authors who consider the trigeminal the gustatory nerve for the anterior part of the tongue describe its fibers as passing in with the chorda tympani. However, this chorda tjnnpani, in so far as it carries afferent fibers, receives its fibers from neurons, the cell bodies of which are in the geniculate ganglion of the facial nerve, since four-fifths of the cells of the ganglion degenerate after section of the chorda tympani (Amabilino, '98 ; de Gaefani, ’06) . There does not appear to be much proof for the statement that the neuraxes of cells of this ganglion become part of the trigeminal nerve by way of the nervus petrosus superficialis major.

(5) The chorda tympani is derived phylogenetically from a branch of the facial nerve, according to many authors. Dixon (’96, ’99), Cole (’96, bibliography, p. 420), Herrick (’99, bibliography, p. 420), Weigner (’05, bibliography, p. 425), and Streeter (’08) considered the ramus petrosus superficialis major a branch of the facial nerve. Consequently lesions of this nerve, even if they did produce disturbances of taste, could not be regarded as proof that the gustatory fibers are components of the trigeminal nerve. Ewart (’89, bibhography, p. 420), Cole (’96, bibliography, p. 420), Green (’00), Strong (’03, bibliography, p. 420), and Bender (’06) regarded it as a derivative of the ramus mandibularis internus of plagiostomes. Herrick (’99, ’01, bibliography, pp. 420, 421) traced it to the same nerve in teleosts, and Coghill (’02, bibliography, p. 421) and Bender (’06) demonstrated that it has a similar origin in amphibians. In Bender’s (’06) experimental work on reptiles and birds he arrived at similar conclusions.

The arguments of those who ascribe gustatory function to the trigeminus depend particularly upon the statements of Krause (’96) that extirpation of the Gasserian ganglion leads to disturbances of taste in the anterior part of the tongue. Sherrington (’98) reached a similar conclusion in his experiments on apes. TT^aflenberg (’97, bibliography, p. 425) found that tactile sensibihty and taste on the left side of the tongue were interrupted in a case following partial degeneration of the semilimar ganglion of the corresponding side and a degeneration of the lingual branch of the trigeminal nerve. Strange to say, taste was not interrupted on the tip of the tongue. Roster (’00), in his work on the facial nerve, supported the conclusion that this nerv^e does not carry gustatory fibers. An argument against this clinical and degeneration work is to be found in a statement by Bruns (’89) that, after extirpation of the Gasserian ganglion, the amount of disturbance of taste varies considerably, and the results of Dana (’06), who, in a case of paralysis of the trigeminus, found taste imdisturbed. The former observer presented a clinical case of great significance for the consideration of this question. He studied a patient in which there was total extirpation of the Gasserian ganglion on the left side, while taste on the left was undisturbed. This same patient had a paralysis of the facial on the right side with a lack of gustatory sensations on the frontal right half of the tongue. Particular credit belongs to Cushing (’03 and ’04) for his thorough examination of this problem. He stated decisively that in 17 out of 18 cases of extirpation of the trigeminal nerve he was unable to establish any disturbances of taste in examinations conducted a month after the operation. Davies (’07) found no disturbance of gustatory sensibility in 15 out

380 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

of 17 cases of extirpation of the trigeminal. Lussana favored the idea that taste is carried by the facial nerve, and Delprat ('90) found that taste was greatly disturbed on the tongue in a case of central facial paralysis on both sides. However, there are cases reported in which a paralysis of the trigeminus appears to have been associated with a disturbance of taste on the anterior portion of the tongue. In many of these cases at least, -as has been noted by Krause (’96) and Wallenberg (’97), smell was likewise disturbed. It seems possible, then, that the disturbance of taste and smell as well was not due to an injury to the nerves carrying these impulses, but to changes in the mucous membrane of the mouth and nose due to the operation. Undoubtedly there is a correlation between tactile and gustatory sensations from the tongue, and a loss of tactile sensibility, as represented in lesions of the trigeminal nerve, probably would interfere with the perceptions of finer gustatory stimuli, particularly since in cases of trigeminal lesion there is often a more or less atrophic condition of the mucous membrane covering the tongue. In conclusion it may be said that the evidence indicates most strongly that gustatory fibers are carried by the facial, glossopharyngeal, and vagus nerves.

The Sensoey Components op the Trigeminal Nerve

The trigeminal nerve is a branchial nerve, but during its phylogenetic development differs in certain fundamental ways from the other nerves of the group. Thus while the vagus, the glossopharyngeus, and the facial, in their development from the branchio-spinal dorsal roots of Amphioxus, exhibit a gradual loss of their cutaneous components and an hypertrophy of their visceral afferent fibers, the sensory portion of the trigeminus shows a development which is the reverse of this. Even in Amphioxus the first dorsal nerve (Nerve II), which corresponds to the ophthalmic portion of the trigeminal nerve, has lost its visceral sensory fibers, but such fibers are present in the second dorsal nerve of this animal (Nerve III), which probably corresponds to the maxillo-mandibular ramus of the trigeminal. In craniotes the trigeminal no longer supplies mucous membranes nor contains visceral sensory fibers. A part of its fibers are cutaneous to the anterior part of the mouth and tongue, but only to that portion which is ectodermal in origin. They do not reach the region behind the embryologic buccopharyngeal membrane. The sensory fibers of the trigeminal nerve, then, are purely somatic afferent, and many of them serve as a cutaneous supply to the outer surface of the head. These skin branches increase in number during phylogeny, with the decrease of cutaneous branches in the other branchial nerves. In addition to these cutaneous sensory fibers and the fibers to the mouth region, the trigeminal has sensory fibers with cells of origin within the central nervous system and dendrites which arise from neuromuscular and neurotendinous terminations in the jaw muscles. These fibers carry proprioceptive impulses from these muscles to the central nervous system. They come from the muscles supplied by the motor branches of the trigeminal, but inside of the medulla oblongata they separate to form the so-called mesencephalic root of the trigeminal nerve, and run dorsalward and forward to their cells of origin, which are situated along the course of the root and as far

TASTE AND THE GENERAL AFFERENT SYSTEMS 381

forward as the posterior commissure. These intramedullary ganglion cells are suggestive of dorsal root cells found in the spinal cord of lower animals, either in adult stages, as in cyclostomes, or during development, as in amphibians. Their extension forward to the region of the midbrain indicates their intimate relation with incoming optic and static stimuli and with primitive general sensibility carried forward from the spinal cord to the tectal centers. The great development of the somatic sensory exteroceptive and proprioceptive components of the trigeminal nerve and its lack of general visceral afferent and gustatory fibers differentiate it clearly from the facial, glossopharyngeal, and vagal branchial nerves.

As the foregoing account indicates, the trigeminal really represents the fusion of two nerves (the second and third dorsal nerves of Amphioxus). The first of these nerves, the ophthalmic, emerges in front of the second myotome, while the second ner\'’e, the ramus maxillo-mandibularis, emerges behind this myotome. The present knowledge of these relations has been contributed to very considerably by the researches of van Wijhe (’82 ; see plagiostomes), whose work has been confirmed frequently. Results of work of this type indicate that the sensory ganglion of the ophthalmic ramus in the earliest stages of development is situated in craniotes at the level of the midbrain, and that it shifts caudalward secondarily and so unites with the ganglion of the nervus maxillo-mandibularis. This occurs in all classes of craniotes, including man. It is of interest to note that in Amphioxus, where the ophthalmic and maxillo-mandibular nerves have their primitive position, and where the branchio-spinal nerves that occur behind the trigeminus carry many cutaneous branches, the central descending trigeminal root does not extend nearlj" so far caudalward as does the descending root of the trigeminal in craniotes. This is a condition which is to be expected. The course caudalward of the descending root of the trigeminal nerve in craniotes is undoubtedly due to the great reduction of the somatic afferent or somatic sensory fibers in the other branchial nerves and the consequent extension of the cutaneous components of the trigeminal farther back in order that they may come into relation with the cutaneous sensory centers and fibers of the cervical cord.

Evidence is lacking to prove that any of the intracerebral spinal ganglion neurons, which are the cells of origin for sensory fibers of both trigeminal roots, as well as for those of other dorsal roots in Amphioxus, are the forerunners of the mesencephalic intracerebral cells of craniotes. Certainly such might offer a plausible explanation for the mesencephalic position of the cells of origin of the proprioceptive fibers of the trigeminal nerve.

CYCLOSTOMES

In cyclostomes the trigeminal nerve has acquired those peculiarities which distinguish it among vertebrates. In both petromyzons and myxinoids the nerve is very large. It appears particularly large in the latter forms because other nerves of the medulla oblongata are so greatly atrophied there {Worthington, ’06 ; Rothig and Ariens Kappers, ’14 ; Jansen ’30). In the larval form of Petromyzon, Ammocoetes {Tretjakoff, ’09), the trigeminal nerve enters as two separate roots : a ramus ophthalmicus and a ramus maxillo-mandibularis, both of which dichoto

382 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

mize in the medulla oblongata. The ascending branches, however, are very short and act apparently as collaterals of the stem fibers. The descending branches of each root can be traced in more or less separate bundles well back into the medulla. Those of the ophthalmic ramus, which are joined centrally by cutaneous fibers from the facial, glossopharyngeal, and vagus, are particularly easily followed. They originate from a ganglion on the root, enter cephalad to the maxillo-mandibular root, and occupy a somewhat ventral position to it. They descend as far as the cervical cord, where they enter into relation with cutaneous fibers of the upper cervical nerves and where stimuli carried by them from the head region are transferred, after a .synapse, to the motor centers of the trunk. Sensory fibers of the maxillo-mandibular branches, which arise from the ganglion on the roots, lie dorsal to the fibers of the ophthalmic division. Their descending branches are short, terminate in the gray of the medulla oblongata, and have nearly all ended before the region of the calamus scriptorius is reached. Van Valkenburg (’11a) has emphasized the point (and there is frequent confirmation of it) that the sensory fibers of the ophthalmic branches descend farther and are thus able to bring about reflexes involving trunk musculature, while the cutaneous fibers of the maxillo-mandibular branch are shorter, terminate in medullar gray, and are concerned particularly in sending impulses by way of secondary fibers to the motor centers of the medulla oblongata.

The descending course within the brain of the root fibers of the trigeminal nerve, which carry cutaneous sensibility from the head, and the ascending course within the spinal cord of the dorsal roots of the spinal nerves, carrj'ing cutaneous sensibility from the body, illustrate well the theory of neurobiotaxis, for these courses indicate the tendency of centers which frequently receive simultaneous or successive stimuli to approach each other. A true sensorj" nucleus of the trigeminal is not present in Petromyzon, but scattered larger cells, similar to the gray of the dorsal horn, are found along the course of the descending roots and are to be regarded, then, as a primitive nucleus of the descending root of the trigeminal. From these cells secondary fibers, largely crossed, pass to efferent centers of the medulla oblongata, particularly to the motor nuclei of the trigeminal and facial nerves and to the motor cells of the cord ; some pass forward (bulbar lemniscus, p. 872).

A nucleus of the mesencephalic root of the trigeminal ner\'e has not been demonstrated with certainty in the cyclostomes. Due to the cooperation of the motor trigeminal and the motor facial nerves in the establishment of movements of the sucking apparatus in response to sensory stimuli over the trigeminal, the motor nucleus of the trigeminal lies near the ependyma of the ventricle and is separated slightly from the motor nucleus of the facial in Petromyzon. In myxinoids both nuclei are ventral in position, in close relation to the sensory root of the trigeminal {Rothig, ’13 ; Rothig and Ariens Kappers, ’14 ; Black, ’17 ; see fig. 215, Chapter IV).

PLAGIOSTOMES

In plagiostomes the more caudally entering fibers of the trigeminal nerve, the maxillo-mandibular branch, occupy the more dorsal portion of the descending

TASTE AND THE GENERAL AFFERENT SYSTEMS

383

root of the trigeminal, while the fibers from the ophthalmic branch which enter farther fonvard he in the ventral part of the root {van Valkenburg, ’ll). Of the two, the more ventral branches are the more compact. In sharks cutaneous fibers, entering by way of the facial nerve (fig. 141B), Join this ophthalmic branch.

and in all plagiostomes cutaneous fibers from the glossopharyngeal and the vagus

nerves (fig. 141A) are associated with it. In part the ^

fibers of the descending root — more particularly those ^

of the maxillo-mandibular branch — terminate in relation

with associated gray (the nucleus of the descending root of Mcsractpk red V ■ ' f

the trigeminal) in its course through the medulla oblon- I

gata. The ophthalmic portion carries many long fibers

which extend for some distance into the cervical cord,

and so come into relation with cutaneous impulses brought

in by cervical spinal nerves. From the gray of the

nucleus of the descending root, secondary fibers pass to ' :

the ventral horns of the spinal cord and to the efferent oi&ib!.'

nuclei of the medulla oblongata. Some fibers originating Fig. iso. The position of here run forward through the medulla oblongata lateral nucleus

to the fasciculus longitudinalis medialis and end in the course of its root in Scyltectum of the midbrain ventrolateral to the octavo- or Hum, according to ./oSnsfon. acoustico-lateral path. These secondary ascending tri- ’ > >

geminal fibers represent, as in cyclostomes, an early projection of impulses from the nucleus of the descending root of the trigeminal upon the midbrain. They carry forward general tactile sensibility and probably pain (vital or protopathic sensibility) from the head, the impulses being similar to those carried from the body by the spino-mesencephalic tracts of these animals.

The mesencephalic nucleus of the trigeminal nerve is very evident in plagiostomes (fig. 159, Scyllium). It consists of large, round or pear-shaped cells / / (fig. 159), accumulated in the roof of the midbrain near the

/ — ^ ( intertectal commissure and extending from the posterior

f ) commissure to the anterior medullary velum. In certain

j cases the cells are found scattered among the ependymal cells of the ventricle. Short dendrites can be demonstrated. It is probable that the fibers constituting the mesencephalic eiv j-oot of the trigeminal enter with the maxillo-mandibular

t)t{su.IY '

MmicfyL re«< 0p6ettd.'

Fig. 159. The position of

9i. J MeKBttpk

reoiV

Fig. 160. Entrance but this statement requires demon root of “rscyuL^m' stration. Fibers of the mesencephahc root (Johnston, ’05, van Valkenburg. iqq . Valkenburg, ’11a) can be identified at the level of

the motor nucleus of the trigeminal, lying dorsal and lateral to that nucleus and lateral to the motor root (fig. 160). The root (fig. 159) can be traced fonvard to the dorsal side of the aqueduct, where it reaches the cells from which it takes origin. Collaterals are given off from it at the level of the motor nucleus of the trigeminal to the cells of that nucleus, and such collaterals are to be regarded as the neuraxes of the cells of the mesencephalic root of the tngemmal. Such a connection makes possible motor reflexes in response to incoming proprio

384 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ceptive impulses. The sensory roots of the trigeminal are termed portio major because of their greater size ; the motor roots constitute the portio minor.

TELEOSTS

In teleosts both the ophthalmic and the maxillo-mandibular divisions of the descending root of the trigeminal are usually very distinct. These divisions are well indicated in the figures of Lophius piscatorius (see figs. 161-162) and of Gadus (fig. 148). Here again the ophthalmic ramus is more compact, is

ventral to the looser fascicles of the more caudally entering ramus maxillo-mandibularis, and is distinguishable in the Weigert preparations through its darker color (see figs. 148 and 161). These dorsal and ventral divisions of the descending root of the trigeminal are entirely separated in their caudal course in Lophius {van Valk(}nburg,’l\o). Between the two divisions passes the vagus nerve (fig. 162). The maxillomandibular portion swings farther and farther dorsalward and ends medially in the associated gray, terminating entirely in front of the entrance of the spinal nerves. However, according to Herrick (’06), the amount of gray is relatively small except at the posterior end of the medulla oblongata. The ventral portion extends farther caudalward but soon loses a considerable nmnber of its fibers. This caudal bundle is surrounded by masses of fibers — ascending fibers from the first and second cervical nerves (see the account of the spinal cord), which send large bundles of root fibers in a frontal direction (fig. 81). The ophthalmic branch of the trigeminal terminates in the spinal cord, both in the funicular nuclei and caudal to their level. In some teleosts there is a clearly defined nucleus for the fibers of the descending root of the trigeminal. In others, these cells are intermingled with the funicular nuclei. In all, the number of fibers which pass behind these latter nuclei into the dorsolateral funiculus of the spinal cord {Herrick) is considerable. The above account indicates that, as in selachians, the maxillo-mandibular fibers terminate in medulla oblongata regions, and are concerned largely in carrying impulses which are passed by secondary fibers to motor centers of the medulla oblongata, while the ophthalmic branch is concerned chiefly in establishing interrelations with sensory impulses brought in by spinal cord nerves and gustatory nerves from the skin and in making connections with motor neurons of the spinal cord. By means of this descending ophthalmic branch it is possible for sensory impulses from the skin of

Fig. 161. Differentiation in staining and position of the pars maxillo-mandibularis {r.d.lrig.p.m.m.) and the pars ophthalmicus (r.d.lrig.p.o.) of the trigeminus in Lophius piscatorius. (Dramng by von Drooglccver Fortuyn.)

TASTE AND THE GENERAL AFFERENT SYSTEMS 385

the head to set up, relatively directly, aboral reflexes through the breast flns. Just before the descending root of the trigeminal reaches the fimicular region it is joined by the cutaneous sensory fibers of the vagus nerve. This latter

Sens. lobe of X

Fig. 162. The separation of the ramus maxillo-mandibularis from the ramus ophthalmicus of the Vth by the entering Xth root. The ramus maxillomandibularis terminates. The ramus ophthalmicus runs caudalward.

branch is small in some teleosts. It appears to be very large in Prionotus (Herrick, ’06).

The cephalic portion of the nucleus of the descending root of the trigeminal is somewhat larger than the remainder of that nuclear mass in certain teleosts

Mesenceph. Bucl< of V

Fig. 163. The position of the mesencephalic nucleus of the Vth under the anterior part of the tectum in Monopterus albus. van der Horst.

(Ariens Kappers, ’06). Herrick, however, could find no appreciable chief sensory nucleus in the fonns studied by him. Very few, if any, secondary fibers can be traced from this cephalic region forward to the midbrain. However, from the

386 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

more caudal portions of this gray nuclear mass there are numerous fibers which decussate to the opposite side and pass forward to the midbrain roof. They are similar in character to those described for selachians and carry general tactile and pain sensibihty from the head and possibly from the cervical region of the body to the midbrain.

A nucleus of the mesencephalic root of the trigeminal nerve was identified in teleosts by Mayser (’81) . Goronowitsch (’88) recognized such a nuclear mass in the brain of Acipenser. Van Gehuchten (’95) described it and its root in Trutta fario and Tello (’09) gave an account of it in the young trout. Detailed studies of the root and its nucleus as these occur in various fishes have been made by J ohnston (’05, ’09), van Valkenburg (’11a), Weinberg (’28), and Woodbume (’36, seep. 1239).

AAIPHIBIANS

The ophthalmic and maxillo-mandibular branches are less easily differentiable from each other in amphibians than in teleosts, since the differences in stainiog reaction and in compactness of arrangement, which made such a differentiation possible in the latter animals, do not occur in these forms. Moreover, the relations of the peripheral branches to the dorsal and ventral parts of the descending root of the trigeminal have not been established with certainty. There is indication that the more dorsal portion terminates to a considerable degree in the medulla oblongata, in the gray that fies close to the ventricular floor {Wallenberg, ’07). Using degeneration methods, this observer showed that a portion of the descending root of the trigeminal enters the frog cord and extends in a caudal direction as far as the lumbar region (the eighth spinal segment), in which region the fibers occupy the lateral part of the dorsal or posterior funiculus and so contribute to the enlargement of this area. By such a distribution the cutaneous sensations from the head, in amphibians as in teleosts, enter into correlation with the cutaneous sensibihty brought in by cervical nerves. Moreover, by means of secondary neurons, impulses reaching the gray of the cord through the descending root of the trigeminal are distributed to the motor neurons, which in turn supply the muscles of the posterior extremities {van Valkenburg, ’11a). Thus motor responses of the body to stimulations of the head are made possible.

In larval Amblystoma, Herrick (’14) found that the sensory trigeminal system arises from various sized, unipolar neurons of the Gasserian ganglion and that the larger root fibers are connected with the larger cells. Such larger fibers, after entrance to the medulla oblongata, occupy a relatively deeper position. They divide into smaller (and even in the adult, unmedullated) ascending and larger descending fibers. The smaller branches of the incoming trigeminal nerve bifurcate also. They are more superficial in position. Throughout the extent of the descending root of the trigeminal Herrick described collaterals from these descending fibers to the ascending secondary visceral (or gustatory) tract. He believed that this may indicate a common nucleus of synapse for both types of fibers, thus establishing tactual-gustatory correlations. In these larval Amblystoma, scaU tered, eoarse fibers pass from the region of the entering root through the ventral commissure to the other side of the medulla oblongata, possibly to the motor

TASTE AND THE GENERAL AFFERENT SYSTEMS

387

nucleus of the trigeminal. The general cutaneous fibers of the facial (described by Norns, ’13, for Siren) and the cutaneous fibers coming in by way of the vagus enter the descending root of the trigeminal and bifurcate into ascending and descending branches. Direct root fibers of the trigeminal pass to the cerebellar auricle {Herrick, ’30).

Secondary fibers from the gray associated with the descending root of the trigeminal have been described in Amblystoma and Necturus {Herrick, ’14, ’30) as decussating and then passing forward to the higher centers. In Necturus, Herrick (’30) found fibers from the more caudal portion of the nucleus joining the spinal lemniscus (of his terminology) and those from the more rostral part joining the trigeminal lemniscus. Both lenmisci terminate partly in the tectum and partly in the dorsal thalamus. It is interesting that these cells of origin for the secondary trigeminal fibers have dendrites which lie in relation to the gray of the fasciculus solitarius. Such a bulbo-tectal or secondary trigeminal connection carries exteroceptive (or somatic vital or protopathic) sensibility as well as visceral sensibility. At the place of entrance of the trigeminal root, cells scattered in neuropil are the forerunner of the chief or superior sensory nucleus of the trigeminal nerve {Herrick, ’30). This nucleus gives rise to fibers of the intertrigeminal commissure of Bindcwald (’ll) and Rolhig (’27) or the commissura cerebclli lateralis of Herrick (’14, ’14a, and ’30), through which the nucleus of one side appears to be related to its fellow of the opposite side and probably to the corpus cerebelli. Other connections of the superior trigeminal nucleus appear to be provided for through a broad mass of decussating fibers, intermingled in part with crossing secondary acoustico-lateral fibers. Certain of such decussating fibers lie close to the surface (as external arcuates) and, after crossing the midline, turn caudalward. They are coarse and heavily myelinated. Herrick has described certain fine myelinated fibers arising from this nucleus which, after decussation, run forward to higher centers, constituting his trigeminal lemniscus. A further connection of the superior trigeminal nucleus with hypothalamic areas was suggested but not regarded as certainly proved by Herrick (’30).

A mesencephalic root of the trigeminal is present in amphibians {Ramdn y Cajal, ’09 ; Johnston, ’05 ; Herrick, ’14, ’14a, and ’30). The cells of origin for this root vary in position somewhat, depending upon the amphibian under consideration. Thus in Necturus {Johnston, ’09; Herrick, ’17) and in Cryptobranchus {Johnston, ’09) the cell bodies of these neurons extend throughout the length of the tectum, although there is only a small number of them in the most cephalic part of the area. In the frog the main cell masses of the mesencephalic nucleus of the trigeminal nerve are found in the cephalic third of the tectum, and the cells of the nucleus, as a rule, are not found in the caudal half of the tectal region {Weinberg, ’28) . A few aberrant cells were found by Larsell (’23) in the region of the velum. A detailed account of this nucleus in the frog (consult fig. 164) is to be found in Weinberg’s paper (’28), where measurements of the nucleus itself and of its constituent cells are given. According to this observer, the 434 cells of the nucleus are arranged in medial, intermediate, and lateral groups, and, as in fishes, the ratio of the lateral to the medial group (with which the intermediate

388 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

was counted) is as 1 : 2. The cells of the lateral group are situated usually in the tectal layer immediately bordering on the ventricular ependyma or even in the ependymal layer itself, while the more medial cells are situated at a greater distance from the ventricular surface. In the toad somewhat similar relations of the cells of the mesencephalic root of the trigeminal have been described, with the addition of a large number of cells extending across the midline and interconnecting the two lateral groups {Johnston, ’09). These cells of the mesencephalic root are in all probability sensory neurons, proprioceptive in character and comparable in a general way to the intramedullary ganglion cells described for fishes. Weinberg could trace a long, straight process from many of these cells which ran toward

the surface of the tectum until it reached the deep medullary layer, where in some preparations it turned and assumed a course parallel to the ventricular surface, in common with other fibers of the layer. In certain cells the process directed toward the surface bifurcated, and in others several short processes arose from the cell and distributed to certain deeper tectal layers. Various other modifications were recognized, all of which indicate that many of the neurons of the mesencephalic nucleus of the trigeminal are not unipolar in amphibians and that, in these animals at least, there is an intimate relation between such neurons and those constituting the collicular region proper. The processes entering the deep medullary layer appear to constitute the mesencephalic root. These root fibers swing downward in front of the nucleus isthmi and then ventral to that nuclear mass. In their course caudalward they reach a position dorsomedial to the sensory root of the trigeminal and there become intermingled with the fibers of this latter root {van Valkenburg, ’11a). Certain fiber bundles passing with the trochlear nerve were traced into the mesencephalic root by McKibben (’13) in Necturus, although Herrick (’14) could not identify such fibers in his material {Herrick found both crossed and uncrossed fibers in the mesencephalic root in Necturus). Black (’17) and Larsell (’23) offered some confirmatory evidence for similar relations in frogs, but their results are not conclusive. Collaterals, probably to be regarded as neuraxes, were traced by Herrick to the motor nuclei of the facial and trigeminal nerves.

Fia. 164. Sagittal section through the midbrain of a frog slightly lateral to the trochlear root, showing the position of the nucleus of the mesencephalic root of V. Chrom-silver preparation. cer., cerebellum; mes.V, cells of the mesencephalic root of V ; n.isth., nucleus isthmi; N.IV, trochlear nerve; opt.v., optic ventricle; T.mes.V, the mesencephalic root of V; sir. alb. pr., stratum album profundum; tor.semicir., torus semicircularis ; ir.sp.cer., tractus spino-cerebellaris. Weinberg (’28).

REPTILES

The sensory trigeminal root varies greatly in different species of reptiles. It is largest in crocodiles due to the great extension of the snout in these animals. This increase in size is strikingly evident macroscopically, since the great development of the descending root of the trigeminal compresses the fourth ventricle

S g ? 2 S

^ ^ tm *3 O O w, Cj Ci 5 d

fc- - = o 2 c « ^ *5 o ^ ^ = a ®.

« c M w ^ i:

'§ g > S* C C .. «*S OS'"

.2

t3

&.ll

to

_s **^ •* eJ "S O o

3>ie ^ c t- S f3 O g "

o

C O

O c g

c “ O

'U « *■

se fc-’

S’.! g

S.-5 -=.:.- B--- --i; — ’o'? -o' 2 _5 ^ 'S § “ o 5

g § o _

§ fe S3 b O o s: ^ X5

g’^’

" "a ^ - w

> >,*o 5 ^ S*

CO o

c

S E

o --.

3 c 3

O X

O

rs

n ^

3^ fe.O

^ £ s: s 2 5 £ 5:5 S'

l^^-l

ft >. c o

^-■3 -ip

s e

£ ^ _g Ph _

C t> . ‘

t5«= S-S

o A J

i S43;§ § I

t;’S “ £

S •- X o _ JL

S g .-s -n ^ ^ 2

V •- M C 5 > o.s

is g| §• ^ g S J " ” ®

““ - “ E 5

...oi .2'^'® -“to'smJS .^ac -=g'=oo_ ® ^ c "“ ^ c t; -S

5 i" 0 '**Sc 2 ®&ix *c> if — _ c S s“r;-^_£»— e

E. S

gjI'SS^Su

5 ^ — 45 •S 5 '"«i £*'

— S 3 t- r.srr^^ 03

^ g - a

- O K.' ®

c; c;

c "

E g = s£ gfc

g'=-'’S o g, V «■ "■ 5 >,

o t- 5^

‘x C

“-Si •cS'S'S^ '*^^**‘

r J i S^iT'o J‘S

55 .^rcS!!^ 5 cs,^ec^x

rfjD C 3 .« 2 — r' — o X

£- _•§ |c JS> s g-f

rrowft'^ ro

^ ---o .2 o 5 « s -<* V

C — S 2 S >i-~ is *S O

,x 5 oP'^' 5 o>£.S 8 ’ 5 .

2 •£ S - s.” «r.£ X ” ^ E ft- o ®

o ^ tJ

C3 >

Ci ta_i

C o P *

TO .5 c

g ?f feci ^ 5 i si '=i>!!: g

1^ X ^mm

o « .5

-^ o c.

.£■ c c « X *5 .S ® x'^^’rt «'£5’g — «o-£*5^o§fefe^

-5 =r • 4»

= § 5- «

o ^ ®

0X2? ' =

P .E C c.“

|3

f 2 '■§ I > j §

5

u

w t)

V "

o P

c rs

S ® O wi to O

M’S i g S

S s l^-l

o o ft^ .p •— «

< c .2

, X

5 5.0 .

^ ^ P e ‘p ““ . • J 5 c*

^ Oj O

i=“2

° -i g s s s --i g g I

5 E 2 •-■= S-S g-o § >.S g

.£* 5 x 5 := -'^ST.SxP = E ft b TJ •= ft.

— ft ts.WS § S C'ftot^ftOSo - fto 3 X_ijV O — t- ej S P _X "S _ *"

^b 2 ^J 3 oPC.ft 2 >-p{ 'cn ft .t-. ft t- ‘

£-.b "o

e ■

o ft

, X — 1:^

■

>

• •. O

O -o

ft **• t

-3 J C I

£o' 2 ' 5 ^x

C .a O c »>• O

- " ' 5 ..S

s3 5 -o ®

p - S

S ^

X

lT 03 i

O -X

J = =

ECh

-ft P ...o _

. p

I p

■i '-^3 i

> °

< Ci

>

-H --T b ^ £

i-s.:rftla

a.

I ft= ft p r o

! _ S-s-3 h-a

“ E-H = x-='o >. E 2 g's g '

S I 8 ;r

P -tf H-» gc ,

X .ft l-H .= ^ P

o.“>’e S ”

^ ^ p c

g:-= S3 ■

> S H .2 .’2 M

■ S ( 5 . t, 03 c

P P X

.£f “3 > t:^

t. .j t 2 tc X *- S'

.ft

... p

^ P

2.2 c

c- X g.

■ft “ ..^ p tc S 03 e

Sft P .ft ^ u ^ CU cl .

o1^' v5

5 2 S 2 g

5 g i’i .5^^.'=^ ■

H M O 03 £

= 03 O .5P o Si ..;

■’.“fe: >.

p •

ft

J jft ^

“ c ^ •

I s>

a .to

i: P o* o

S > ft

=. t3 g

ft ft p

i P CQ .ft P ^ ^ E >.£=>-§

389

TASTE AND THE GENERAL AFFERENT SYSTEMS

391

fonvard to the region of the optic tectum. These are comparable to the secondary trigeminal fibers of fishes and amphibians. Short fibers, crossed and unci ossed, to the motor nucleus of the trigeminal and to the other motor centers of the medulla oblongata aiise from the chief sensory nucleus and the nucleus of the descending root of the trigeminal. This latter nucleus is believed to send short fibers to the motor centers of the upper cervical cord. The increase in terminal nuclei and in secondary trigeminal fibers in reptiles is associated with the appearance in these foi-rns of tactile and Pacinian corpuscles in the region supplied peripherally by the trigeminal nerve. (For the various tracts see figure 165.)

A mesencephalic root of the trigeminal has been described by Johnston (’09) for turtles and by van Valkenburg (’11a) and Weinberg (’28) for various reptiles. In Alligator and Chelone, the cells of origin are found throughout the length of the tectum. Thej^ form into lateral and medial nuclei, the I

cereb

Nucl mc5cnceph root of V in Opt t€ct

- -

Fig 107 The mesencephalic nucleus of the Vth in a snake, Eunectes Sagittal section. I an Valkenburg.

168 The me.sencephahc nucleus m a young boa constnctor Huber and Crosby

medial nuclei of the two sides fusing to form an unpaired median group. In addition to these two groups, an indistinctly separated intermediate group {van Valkenburg, ’11a; TFei7i6er<7, ’28) has been recognized in the alligator but is not identified easily in the turtle. It was recognized by van Valkenburg in Chelone, but was not differentiable from the cells of the lateral group in the series of Chrysemys studied by Weinberg. The difference in distribution of the cells of the mesencephalic root of the trigeminal nerve in the alligator and the turtle (fig. 166) is indicated in the counts made by Weinberg (’28), who found in the former animal 1295 cells in the lateral gioup and 1398 in the combined intermediate and medial groups, making a total of 2693 cells, ^\dth an approximate ratio of 1 ; 1 for the two groups. In the turtle Weinberg counted 192 cells in the lateral group and 345 cells in the medial group (no intermediate cells were present). This made a total of 537 cells, the ratio of the lateral to the medial group being as 1 : 1.8. The medial cells in reptiles are predominantly multipolar, but the great majority of the cells in the lateral group are unipolar as in mammals. In lizards and snakes there are very few cells in the midline, the cells being mostly arranged in the lateral nucleus which extends throughout the length of the tectiun. In snakes (Eunectes murinus, van Valkenburg, ’11a) they have accumulated par

392 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ticularly in the more posterior portion of this lateral group (fig. 167 ; fig. 168, boa constrictor), where they form a very large nucleus. This increase in size in snakes is due to the development of the musculature of the lower jaw, to which they supply proprioceptive fibers. The root runs caudalward in the usual manner, lying lateral to the trochlear root. In snakes it forms a very thick bundle lateral to the medial longitudinal fasciculus. On emerging it runs parallel to the fibers of the motor root, which it joins. Functionally the mesencephalic root is believed to consist largely of proprioceptive fibers from sensory endings in the muscles supplied by the motor trigeminal nerve. There is evidence that it is joined centrally by proprioceptive fibers entering with the eye muscle nerves and having their cells of origin in the mesencephalic nucleus of the trigeminal.

BIRDS

Highly differentiated sensory terminations occur on the peripheral branches of the trigeminal nerve in birds. These peripheral terminations, corpuscles of Grandry, are found in the beak and the tongue; those of Herbst in the beak. A description of these corpuscles and figures illustrating them are found on pages 31 and 32. The majority of the trigeminal fibers end as free sensory endings or as tactile corpuscles in the skin, and as neuromuscular terminations in the muscles. The size of the sensory branch of the trigeminal in birds varies with the size and formation of the head, and particularly the size and form of the beak.

Within the central nervous system, the nuclei of termination for the cutaneous sensory fibers are a chief sensory nucleus (see also fig. 169) and a nucleus of the descending root of the trigeminal nerve. Unlike reptiles, these two nuclei are discontinuous in birds. The chief sensory nucleus lies at the level of entrance of the root. It contains many large cells and is distinctly separated from the surrounding gray by a cap of myelinated fibers. It receives a very considerable number of the incoming sensory fibers of the trigeminal. The nucleus of the descending root consists of scattered gray along the course of the path. It is relatively small in birds. It receives not only root fibers from the descending root of the trigeminal, but also cutaneous sensory fibers entering by way of the vagus nerve. At its caudal end it lies somewhat farther dorsally than throughout the rest of its course

Fig 169 Chief sensory and the motor nuclei of the trigeminus in a bird (Catharistes urubu).

TASTE AND THE GENERAL AFFERENT SYSTEMS 393

and increases in size. Tliis latter part alone is sometimes termed the spinal sensory nucleus of the trigeminal. As in reptiles, it is not easy to distinguish between the ramus maxillo-mandibularis and the ophthalmic ramus of the trigeminal nerve once thej"^ have entered the central neiwous system. There is reason to think that the chief sensory nucleus of the trigeminal receives largely maxillo-mandibular fibers, while the ramus descendens is concerned particularly with ophthalmic branches. Since the maxillo-mandibular branches carry the fibers from beak and tongue regions, where the highly specialized corpuscles are situated, the chief sensory nucleus might be regarded in birds, as in mammals, as concerned primarily with the more discriminative tjTDes of cutaneous sensibility, ha\dng differentiated as such in correspondence with the appearance of specialized nerve endings. This would agi'ee very well with the results obtained in mammals.

It is questionable whether or not in birds the direct trigeminal root fibers pass to the cerebellum, although certain preparations certainly suggest such a connection. From the chief sensory nucleus of the trigeminal there are connections, crossed and uncrossed, ^\dth the motor centers of the medulla oblongata and the cerebellum {Biondi, T3 ;

Craigie, ’28; Sanders, ’29). Furthermore, from this nuclear mass arises the quinto-frontal tract of Wallenberg (’03,

’04; see also Schroeder, ’ll, and Huber and Crosby, ’29). After a partial decussation in the medulla oblongata, this tract runs forward toward the basal portions of the forebrain and thus brings cutaneous sensations from the head — “oral sense’’ (Ariens Kappers; Edinger, ’08) — into correlation with olfactory sensations. Secondary fibers from the chief sensory nucleus, the so-called quinto-mesencephahe tract, decussate and run forward to the nucleus lateralis mesencephali and to the optic tectum. From the lower part of the nucleus of the descending root of the trigeminal or the so-called spinal trigeminal nucleus, at least in the cassowary, there are crossed fibers which end largely in the reticular gray, but which, in part at least, run forward as a trigeminal mesencephalic tract to the midbrain. YTiether or not any fibers from the terminal nuclei of the trigeminal nerve reach the thalamus directly is uncertain.

The mesencepahlic root in birds has certain distinguishing peculiarities. Scattered cells belonging to the root are found throughout the whole tectum. In birds both medial and lateral groups of cells associated with the mesencephalic root of the trigeminal nerve have been described (van Valkenburg, ’11a ; Weinberg,

Cells of meaencepb root V

trigeminus in Cathanstes urubu.

394 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

’28), but the relative number of cells in the two groups varies in different birds. Thus the ratio of lateral to medial cells in the chicken is 1 ; 1.25, in the duck 1 ; 2.5, m the dove, 1:3, in the sparrow 1:1. In some birds the cells are located largely in or near the midline (Catharistes, fig. 170), although lateral groups also occur. In other birds (stork) the cells are situated laterally. In the sparrow Weinberg counted 676 cells in the lateral group and 757 cells in the medial group. The cells vary in size in different birds. Usually the cells of the lateral group are the larger, particularly in such birds as the duck, where they approach the size of the homologous cells in man. The largest number of these cells is found at the level of the nucleus of the oculomotor. Such cells accompany the mesencephalic

Cells of the mesenceph. root of V

Fig. 171. The mesencephalic cells of the trigeminus in a turkey. Kosaka.

root as far caudalward as the level of the trochlear. Although critical evidence is lacking, there is every probability that the mesencephalic root fibers enter with the maxillo-mandibular root iJVeinherg, ’28). Collaterals from the root have been traced to the cerebellum, to the motor trigeminal nucleus, and to the reticular formation as far as the level of entrance of the glossopharyngeus (IF alienberg, ’04). There is some evidence that there is a relation between the trochlear root and the mesencephalic root of the trigeminal, and it is probable that a similar relation exists between this nerve and the other eye muscle nerves, particularly the oculomotor. Presumably this mesencephalic root and its nucleus are concerned with proprioceptive impulses brought in from the muscles supplied by the trigeminal and eye muscle nerves. The nucleus itself, in such cases, is comparable to the mesencephalic nucleus of the trigeminal found in other forms.

MAMMALS

The impulses of pain, temperature, and tactile sensibility carried from the head region are often particularly well developed and important in mammals.

TASTE AND THE GENERAL AFFERENT SYSTEMS

395

where the snout region (together with the sense of smell) frequently serves as an organ of exploration. This is particularly true of animals which root in the ground, such as the hedgehog, the mole, and the pig. The organ of Eimer in the mole and the modified snout in the pig represent organs of special sensation. In some animals, such as the mouse and the cat, the whiskers, which are innervated at their base b}' trigeminal fibers, act as organs of localization as do the antennae of insects. In certain animals, as the bullock, the innervation of the so-called sense hairs is important (fig. 174). The size of the sensory root of the trigeminal

Mu ettpYEdesc Hutrian^

Fig 172 The descending root of the trigeminus in Echidna Schepman

nerve is not always correlated directly unth the specificity of the sensation carried by it. Such specificity is determined by the specialization of the peripheral area. Thus, in the monotremes the trigeminal nerve is very large because its area of peripheral distribution is very great, but in the field mouse the large size of the root is not associated with distribution of its fibers to a large peripheral area, but with the richness of their distribution within restricted parts of that area. The distribution of the sensory fibers of the trigeminal nerve in man is shown in figure 173. The right side of the figure shows the distribution after a preparation of BoJk (’98 to ’00) ; the left side shows the loss of sensibility after extirpation of the semilunar ganglion in a case of trifacial neuralgia, and follows the work of Cushing (’04, fig. 173).

Centrally ophthalmic and maxillo-mandibular rami are differentiable. The

396 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ophthalmic portion enters in front of the maxillo-mandibular portion and runs ventral to it within the central nervous system. This latter arrangement within the central nervous system was pointed out by Bregmann (’92), Bochenek (’01), and TV allenberg (’05) for the rabbit. Wallenberg was able to show that in man the lingual branch of the trigeminal is the most dorsal constituent of the descending root and lies in close relation to the fasciculus solitarius (see fig. 158 B). Van Valkenburg (’ll) confirmed the ventral position of the ophthalmic ramus in man and found that it constitutes the most caudal portion of the descending root, as has been shown to be the case in lower vertebrates.

Fig. 173. Area of peripheral distribution of the trigeminal nerve in man. On the left, the peripheral distribution of the trigeminal nerve after extirpation of the Gasserian ganglion, as determined by Cushing, and on the right, as determined by Bolk by means of anatomic preparations. /, 11 , and 111 denote the distribution areas of the respective major branches.

The position of the descending root of the trigeminal is not the same in all animals (van Valkenburg, T1 a). In certain animals this half-moon-shaped bundle has a peripheral position and even forms a swelling on the lateral wall of the medulla oblongata (fig. 175, hedgehog). In others, the upper point of the half moon, the maxillo-mandibular portion, is turned inward and approaches more nearly the floor of the ventricle (see fig. 175, anl^eater). This difference in position is correlated with a difference in function and hence a difference in the secondary connections of the trigeminal. Thus in the ant-eater the development of the extraordinarily long tongue, by means of which the animal obtains its food, leads to a particularly great development of the lingual fibers and a consequent increase of this portion of the root and a close functional relationship of it with the visceral sensory centers of the tongue and the pharynx. As van Valkenburg (’ll) pointed out, the variations in the position and shape of the descending root of the trigeminal arc not to be regarded as chance conditions. They are due undoubtedly to the operation of neurobiotactic influences which tend to bring these fil)crs

Fia. 174. The nerve termination on a tactile hair, after Treljakoff, redrawn in single color. Bullock.

A, outer lamella of sheath; Ar, artery; B, sinus trabecula; G, glassy layer; F, connective tissue fibers; H, hair; J, inner lamella of sheath; K, conical body; Ki, sinus cushion; P, papilla; S, sinus space; Sch, umbrella-like swelling of outer root sheath; T, sebaceous gland; V, vein; Z, swelling of outer root sheath.

Nervous elements: 1, entering nerve funiculi; 2 (27), lower annular ple.\us ; 3, simple side branches beneath the annular plexus on inner surface of outer root sheath; 4y more complex endings in relatively the same position ; 5, side endings on bundles which ascend along the inner surface of the outer follicle sheath ; <?, side endings in the outer follicle sheath; 7, side endings in connection with the two formed endings in the outer follicle sheath ; 8, side ending in the middle region of the hair follicle ; 9, upper annular plexus or the upper nerve ring; 10, nerve fibers into the subpapiUary connective tissue; II, nerve arborizations in the outer follicular sheath, below the lower annular nerve ring; 12, similar endings in the outer portion of the outer follicular sheath ; IS, similar ending in the middle region of the follicle; bulb w'ith an axial end fiber; J5, end bulb with a branched central fiber (Golgi Mazzoni corpuscle); 16, encapsuled ending with leaf-formed expansions; 17, tree-formed endings; 18, similar ending with spindle-formed expansion; 19, similar ending with skein-formed terminations; 20, endings on sinus trabeculae; 21, sensory end disks below the swelling of the root sheath; 22, sensory end disks on the swelling of the root sheath; 23, sensory end disks on the conical body; 24, branched end disks beneath the sebaceous gland; 25, tactile disks in the upper half of the root sheath swelling ; 26, tactile disks in the lower half of the root sheath swelling ; 27, lower annular plexus or lower nerve ring.

397

398 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

into closer relationships with the centers with which they are most intimately concerned, considered functionally. Illustration of this is found on the one hand in the more intimate relation of visceral sensory centers of the mouth and the pharynx and cutaneous sensory centers represented by the trigeminal. On the other hand, there is a tendency for the cutaneous fibers from the face, particularly those originating around the eyes, and from the eyes themselves, to take a position close to the longer ascending and descending tracts of the ventral tegmental area, such as the tecto-bulbar and tecto-spinal paths and the reflex tracts of the acoustic system.

The terminal nuclei (fig. 178), within the brain stem, for the exteroceptive fibers of the trigeminal nerve, have been recognized by many observers as represented in the chief sensory nucleus and in the nucleus of the descending root. The

former nucleus lies at the level of entrance of the trigeminal roots, in position lateral to the motor nucleus of this nerve, and separated from it by trigeminal root fibers. In nearly all mammals the chief sensory nucleus is very well developed. It is larger in certain mammals than in birds, but usually does not occupy so dorsal a position, and contains dominantly small nerve cells intermingled with larger ones. The nucleus of the descending root of the trigeminal is continuous with the caudal pole of the chief sensory nucleus, from which position it extends backward toward the spinal cord, where it becomes continuous with the dorsal horn gray, vath the level of the termination varying slightly in different mammals. It is thought to extend, as a rule, as far caudal as the second cervical segment and thus to provide for sensory correlation between the head and trunk regions, since usually in mammals the first cervical nerve does not present a sensory root. The longer ophthalmic fibers reach this caudal part of the nucleus of the descending root.

There has not been unanimity of opinion with regard to the distribution of the root fibers of the trigeminal and consequently TOth regard to the functional significance of these two nuclear groups. In the work of Held ('92) and von Kolliker ('89 to ’02), reference is made to bifurcating fibers of the trigeminal root, but the first clearly presented and conclusive evidence of the bifurcation of sensory fibers of the trigeminal appears to have been offered by Ram6n y Cajal (’9G). This observer traced sensory trigeminal fibers which bifurcated, one

Impulses from the top of the head, carried by ophthalmic mrvcH, are very frequently responded to by motor impulses from the cervical cord, ^liilc Fensoiy impulfcs over the maxillomaudihular roots tiave their motor respon'-cs, such ns chewing and other jaw movements, largely by way of the medulla oblongata reflexes.

Fia 175. The differences in position of the descending root of the Vth (black) in the hedgehog (at the left) and in the ant-eater (at the right), tan Valkenhurg.

TASTE AND THE GENERAL AFFERENT SYSTEMS

399

branch passing forward and dorsalward into the chief sensory nucleus, the other passing caudalward to form the descending root and to terminate in its associated gray, the nucleus of the descending root. Obersteiner (’12) found that the great majority of the fibers bifurcated and Gerard (’23) held a similar opinion, although she was unable to demonstrate a bifurcation of the nonmedullated or the smaller medullated fibers, which, according to her statement, constitute approximately one-fourth of the total number of fibers. Windle (’26) found three types of exteroceptive trigeminal fibers : (1) a medullated fiber passing directly to the chief sensory nucleus ; (2) a medullated fiber which bifurcates and sends one branch to the chief sensory nucleus and the other branch to the nucleus of the descending root ; (3) a smaller medullated or nonmedullated fiber distributing to the descending root. Fibers of types 1 and 2 are regarded as carrying tactile impulses, fibers of type 3, as cariying pain impulses (see also Windle, ’27). According to this interpretation {Windle), the chief sensory nucleus is a tactile brain center, the nucleus of the descending root a center for general tactile sensibility and pain. These conclusions, as reached by Windle, agree with the clinical observations made by Winkler (’14), Spiller (’15), Gerard (’23), and Stopford (’25). Brouioer (’15) has also pointed out that lesions of the caudal portion of the trigeminal nucleus interrupt the passage of sensations of pain and temperature, while tactile sensibility shows very slight abatement, owing to the fact that fibers which carry tactile impulses terminate to a great extent in the chief sensory nucleus.

In many ways there is a marked resemblance between the chief sensory nucleus of the trigeminal and the nuclei of fasciculus gracilis and fasciculus cuneatus, for just as the chief sensory nucleus receives tactile impulses (and more particularly, the more specialized type of tactile sensibility) from the head, so the nucleus gracihs and especially the nucleus cuneatus receive the impulses for twopoint discrimination from the trunk. A similar comparison may be established between the nucleus of the descending root of the trigeminal nerve and the dorsal horn gray of the spinal cord, since both receive impulses of general tactile sensibihty, pain, and temperature. In so far, then, as the cutaneous sensory elements are concerned, the trigeminal nerve is comparable with the sensory nerves of the spinal cord.

Short fibers interrelate the various portions of the nucleus of the descending root of the trigeminal, which do not degenerate with experimental degeneration of the root {Breuer and Marburg, ’02), and are spoken of as fibrae concomitantes trigemini. Secondary tracts from both the nucleus of the descending root and the chief sensory nucleus of the trigeminal nerve pass to the motor nuclei of the medulla oblongata and possibly to the upper part of the spinal cord. These are largely, but not entirely, crossed. Many of such fibers arise from the dorsomedial portion of the chief sensory nucleus and pass to the motor trigeminal nucleus of the other side, and, as is postulated, also to the oculomotor nucleus. The secondary ascending fibers from the spinal trigeminal nucleus appear to be wholly crossed (Wallenberg, ’05). From the chief sensory nucleus of the trigeminal nerve, the secondary fibers arise and pass forward as the homolateral and contra

400 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

lateral dorsal secondary ascending tract of the trigeminal. At first they are separated from the other lemnisci systems lying farther dorsalward, but they gradually swing somewhat lateral ward and join the other great sensory ascending paths, passing with them to the thalamus, where they terminate in the ventral nucleus. They are supposed to carry mainly tactile discrimination. A somewhat similar path, carrying pain and general tactile sensibility from the head, arises from the nucleus of the descending root of the trigeminal, and crosses and ascends as a ventral secondary tract of the trigeminal to the ventral nucleus of the thalamus (for details see fig. 521 and Chapter VIII, p. 1150). This ventral tract is joined in the pons region by fibers from the chief sensory nucleus. At

Leeion

that level, it lies in relation with the other lemnisci systems. These secondary ascending bundles from the trigeminal sensory nuclei are often spoken of as the trigeminal lemniscus. In amphibians Herrick (’30 ; see bibliography for amphibians) spoke of an ascending tract from the chief sensory nucleus and the upper part of the nucleus of the descending root as a trigeminal lemniscus, and stated that secondary fibers from the lower part of the nucleus of the descending root ascended in the spinal lemniscus. Sometimes the name “ trigeminal lemniscus ” is applied only to the dorsal tract. Thus von Monakow (’10), Lewandowsky (’04), and Wallenberg (’05) regarded the trigeminal lemniscus as arising only from the chief sensory trigeminal nucleus. In their course, the bundles, particularly the ventral one, give off collaterals, and perhaps stem fibers, to the midbrain. This part of the bundle is comparable to the trigemino-mesencephalic or -tectal path of lower forms. Reference is made to figure 176.

The nucleus of the mesencephalic root of the trigeminal extends in many mammals throughout the entire length of the optic tectum as far as the plane of

TASTE AND THE GENERAL AFFERENT SYSTEMS 401

the posterior commissure. Examples of such an extent are to be found in Echidna, Onychogale (fig. 177), Vesperugo, Tamandua, Phoca, and other forms. Castaldi (’26), who made a careful study of the nucleus of the mesencephalic root of the trigeminal in the guinea pig, found the main mass of the nucleus appearing

NiicI mescnceph root of V

Fig. 177. The nucleus of the mcsenceplmlic root of the trigeminal nerve of Onychogale frenata (sagittal section), van Valkenburg.

in sections through the rostral end of the oculomotor, and the caudal pole of the nucleus gradually diminishing and terminating at the caudal end of the chief sensory nucleus of the trigeminal. In man the cells extend relatively far frontalward {van Valkenburg, ’11a).

In the series of animals studied by Weinberg (’2S) , the greatest numbers of cells of the mesencephalic root of the trigeminal were found in monkej’s and man, where

Fig. 178. The mesencephalic, the chief sensory, and the motor nuclei of the Vth nerve in the rabbit.

totals of 4869 and 5735 cells were obtained, respectively. An increase in size of the caudal part of the nucleus in mammals (rat, cat, mouse, rabbit, monkey, and man), as compared with the nucleus in forms below mammals, was recognized,

402 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

such an augmentation occurring in the cat and in rodents at about the level, or behind that, of the decussation of the trochlear nerve, in monkeys slightly in front, and in man behind that level {Weinberg, ’28). In the guinea pig, Castaldi (’26) found an increase in the number of cells at the level of the inferior colliculus, a decrease where the root passes downward and through the anterior medullary velrnn, and then an increase in planes through the motor nucleus of the trigeminal nerve. Hulles (’07) found the trochlear root in guinea pigs and pigs surrounded by cells of the mesencephalic root and identified relatively large numbers of these cells at the level of the motor nucleus of the trigeminal in rats, moles, and rabbits. In the rodents studied by Weinberg, the nucleus is not so well developed in the oculomotor region as in the cat, the monkey, and man. The division into lateral, medial, and intermediate groups, such as is evident in many lower forms, is lacking in most mammals. Cells of the mesencephalic root of the trigeminal are present in a mediodorsal position in monotremes and marsupials, according to van Valkenburg (’11a). This same observer found some cells of the root in the anterior medullary velmn in rabbits, horses, and man, but Weinberg’s work indicates that the number of these cells in man, as well as in the cat and the monkey, are few in the aqueduct region, and that cells of this type are lacking in this region in the mouse and the rat. On the whole, cells of the mesencephalic root of the trigeminal lie farther ventralward in most mammals than in lower vertebrates, and in man extend to the motor nuclei of the oculomotor and trochlear nerves, and are even intermingled with cells of these nuclei. Various measurements of the cells of the mesencephalic root of mammals have been made. The following are several measurements given for man :

45-50^1 Meynert

60-70/i (largest measurement) Schwalbe ('81)

45-60/1 Obersleiner (’01)

18-20/1 (small dorsal cells) Weinberg (’28)

28-35/1 Weinberg (’28)

In general the cells of the mesencephalic root in adult mammals are unipolar, but occasional bipolar or multipolar cells are found and they are of frequent occurrence in embryos {Ramon y Cajal, ’96 ; Johnston, ’09 ; Castaldi, ’26 ; Weinberg, ’28, and others). In the dog, Schneider (’28) described large cells in the ventral part of the mesencephalic nucleus of the trigeminal and small and medium-sized cells in the dorsal portion, with unipolar, bipolar, and multipolar cells present, but with the unipolar the predominant type with a considerable number of the small neurons being bipolar and multipolar in type. He considered that the cells have the appearance of sensory neurons. The tendency toward imipolarity and the general appearance of the neurons of this root have led numerous observers to compare them with the spinal ganglion cells and to regard them as intramedullary ganglion cells {Deiters, ’65 ; Meynert, ’72 ; Kure, ’99 ; Ram6n y Cajal, ’09 ; Johnston, ) Willems, ’ll] Weinberg, ’28; and others). Their gradual shifting caudalward in mammals in such a case would appear to indicate neurobiotactic influences which lead them to migrate along the root (toward the source of

TASTE AND THE GENERAL AFFERENT SYSTEMS

403

stimulation). Johnston (’09) considered these neurons as neural crest elements which have not been cut off from the central nervous system, and such an origin has been suggested for them by other workers. If such be the case, they are comparable to the intramedullary ganglion cells of fishes. The thick fibers of the mesencephalic root are considered dendrites, the thin collaterals are regarded as neuraxes. Further evidence of the sensory character of the mesencephalic nucleus of the trigeminal is offered in some recent work by Sheinin (’30). This observer, in his study of the cells of the mesencephalic nucleus in dogs, found, in addition to the large, pear-shaped type (a), three other types of cells : (6) a medium-sized neuron with spheroidal cell body and coarse, somewhat elongated Nissl granules ; (c) a small or medium-sized cell, with spheroidal or in some cases oval cell body, and either irregular or coarse, or at times fine Nissl granules; and (d) very small, very deeply stained cells of varying shapes with few and fine Nissl granules. For the details of their cellular structure the original article should be consulted. If these cells are all neuronic elements of the nucleus of the mesencephalic root of the trigeminal — and in this connection it should be remembered that van Valkenburg (’11a), Willems (’ll), and Schneider (’28) have all recognized small cells within this nucleus, and the second observer has noted the chromatolysis of their Nissl substance following lesions of the mesencephalic root — then the fact that their cell types correspond to those described for the dorsal root and semilunar ganglia and that they occur in the same ratio in the mesencephalic nucleus and the semilunar ganglion is a matter of considerable interest. This offers a further argument in favor of the sensory character of the mesencephalic root of the trigeminal nerve.

The mesencephalic root itself has been traced in many animals {von London, ’07; van Valkenburg, ’11a; May and Horsley, ’10; Kosaka, ’12; Castaldi, ’26; Weinberg, ’28). Many of these observers have used degeneration preparations. Kosaka found that in the rabbit, at least, these fibers are almost exclusively in the mandibular branch of the trigeminal. Willems (’ll) found that the mesencephalic root fibers enter with the mandibular branch in the rat. This observer pointed out that the number of fibers (4800) in the portio minor of the trigeminal of the rat amounts to about the same as the number of mesencephalic (1600) and motor cells (2900) together (4500). Having entered with this branch, the fibers run dorsolateralward, passing the motor trigeminal nucleus to which their neuraxes are distributed {Ramon y Cajal, ’09 ; and Willems, ’ll), and then extending forward to their cells of origin. Probst traced a portion farther caudad to the level of entrance of the facial and glossopharyngeal nerves. No fibers of the mesencephalic trigeminal root have been observed in the ophthalmic branch of the trigeminal, although phylogenetically this is the branch which is the most intimately related with the midbrain. Even in the dog, where a portion of the mesencephalic root fibers is believed to be found in the maxillary ramus of the trigeminal, none are present in the ophthalmic branch. The consideration as to which branches carry mesencephalic root fibers has been discussed at some length by Kosaka (’12), but the explanation as to the distribution appears to lie in the fact that in peripheral distribution and functional significance the mesencephalic

404 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

fibers are most intimately related with the mandibular branch, for these fibers carry proprioceptive impulses from the muscles of mastication.”

Possibly the forward extension of the nucleus of the mesencephalic root of the trigeminal nerve may be accounted for through its relation to proprioceptive

Fig. 179. A, cross section through the region of the oculomotor nucleus of man. B, cross section through- the frontal portion of the pons in man, showing the relation of the mesencephalic root of V to the nucleus loci coerulei. Chrora-silver preparation. Weinberg (’28). a, cell of the mesencephalic root of V lying very close to the oculomotor nucleus. The process of this cell is directed toward the root fibers of the third nerve; ag., aqueduct; br.conj., brachium conjunctivum; c.guad.anL, corpus quadrigeminum anterius; f.l.m., fasciculus longitudinalis medialis; mes.V, cells of the mesencephalic root of V; n.Lcoer., nucleus loci coerulei; n.mot.V, motor nucleus of V; n.III, oculomotor nucleus; r. mes.V, mesencephalic root of V; r.mot.V, motor root of V ; r.sens.V, sensory root of V ; slr.alh.pr., stratum album profundum.

fibers of the trochlear and, more particularly, the oculomotor nerve, for in addition to the proprioceptive impulses from the muscles of mastication, there is an in “ Herrick (T4) thinks it possible that the mesencephalic root fibers carry only one type of muscular sensitivity, that represented in the neuromuscular terminations, while other sensory trigeminal fibers may be concerned in the appreciation of pressure within the muscle or recognition of pain.

TASTE AND THE GENERAL AFFERENT SYSTEMS

405

creasing body of evidence favoring the presence of proprioceptive fibers from eye muscle nerves within the mesencephalic root of the trigeminal nerve. The presence of proprioceptive nerve endings within the eye muscles has been known for many years {Huber, ’99, ’00 ; Tozer and Sherrington, ’10, and many others), and their course peripherally in the eye muscle nerves has been established. Ganglion cells have been described along the roots of the oculomotor and abducens nerves of man {Nicholson, ’24), but Clark (’26) has pointed out that in the arrangement of their Nissl substance and in other general characteristics, they do not resemble known proprioceptive neurons. Neither resection of the eye muscles, as was carried out by Kohnstamm and Quensel (’08), nor section of trochlear and oculomotor nerves, as was done by Tozer (’12), led to changes in the cells of the mesencephalic root of the trigeminal, but a degeneration of the neurons, together with their root fibers, occurred when all structures were removed from the orbit in the cat {Freeman, ’27). Anatomical evidence presented by Weinberg (’28 ; fig. 129) favors his conclusion that “the muscle sense from the eye muscles is, at least partly, transmitted by the neurons of the mesencephalic root of V.’’ In the preparations of hmnan material, Weinberg identified cells of the type found in the nucleus of the mesencephalic root of the trigeminal not only near the nuclei of the oculomotor or trochlear, but intermingled with their cells. Davis (’23) appears to have shown that proprioceptive impulses from the face musculature pass centralward over the facial nerve. Possibly their nuclei of origin lie in the geniculate ganglion {Wakeley and Edgeworth, 'ZZ ; see p. 378).

But while an increasing munber of students of the central nervous system appear to accept the idea of a proprioceptive function of the mesencephalic root of the trigeminal, certain other interpretations have been held with regard to the fimction of this root, some of which are still current. Space does not permit a full discussion of these divergent views, though a few are here listed : (1) A group of observers, of whom we may fist van Gehuchten (’95), Bruce (’98), Ramdn y Cajal (’09), and Jacobsohn (’08), regarded the nucleus and the root as efferent in function, suppljdng fibers to the motor root of the trigeminal or to the tensor veil palatini {von Kolliker, ’96 ; Terterjanz, ’99) and the tensor tympani muscles {von Kolliker, ’96) . A recent exponent of the view that the mesencephalic nucleus of the trigeminal nerve is motor and not sensory is Castaldi (’26), who, among other arguments, advanced the objection (raised from his study of the embryologic development of the nucleus) that the nucleus of the mesencephalic root develops precociously as does the motor nucleus of the trigeminal and is larger in the embryo, at a stage where the muscles are relatively retarded in development, than it is in the adult. This observer also stated that the long processes of its neurons show the characteristics of neuraxes of neurons of the central nervous system, particularly since they do not have Nissl substance and are myelinated, and that to regard these processes as dendrites would imply that the dendrites developed here before neuraxes, which is contrary to the usual course of embryologic development. The physical appearance of the processes can scarcely be regarded as affording unquestionable criteria in this case, since in the peripheral nerves the dendrites of sensory neurons are indistinguishable anatomically.

406 NERVOUS SYSTEMS OF VERTEBRATES AND OF ]\IAN

Nevertheless, the judgment of so experienced and careful an observer merits serious attention. (2) Others have regarded this nucleus as having a form of visceral function, serving either as a trophic {Mendel, ’88), a secretory {Merkel, ’74), or a vasomotor center {Huguenin, ’79). There appears to be little e\ddence in support of such a parasympathetic (or craniosacral sympathetic) functioning of these cells {Oppenheim, ’13 ; clinical material of Darus, ’07).

Collaterals given off from the fibers constituting the mesencephalic root of the trigeminal establish connections in mammals with the deeper tectal graj' {Held, ’93 ; Weinberg, ’28) and with the reticular gray of the mesencephalon {Willems, ’ll). As the root fibers pass the upper border of the cerebellum, collaterals enter that brain region {Wallenberg, ’04 ; Weinberg, ’28, and others), and small fibers have been seen to enter the locus coeruleus {Held, ’93 ; Willems, ’ll ; and others). In addition to the motor nucleus of the trigeminal {Ramon y Cajal, ’96, ’09; Weinberg, ’28, and many others), other cranial nerve nuclei, such as Deiters’ nucleus, the nucleus ambiguus {Probst, ’99 ; Weinberg, ’28), and the motor facial nucleus {Weinberg, ’28) are believed to receive collaterals from the fibers of the mesencephalic root.

Before closing the account of the nucleus of the mesencephalic root of the trigeminal, some mention must be made of its relation to the nucleus of the locus coeruleus. Meynert (’72), Obersteiner (’92), Cramer (’94), von Kollikcr (’9G), Terlerjam (’99), and Allen (’19) have regarded this latter cell mass as a nucleus of origin for mesencephalic root fibers. Kohnstamm (’10), May and Horsley (’10), and Kosaka (’12), testing such relations by cutting the trigeminal root and studying the cells of the nucleus loci coerulei, found no degenerations. Some cells of the nucleus of the mesencephalic root appear to be intermingled with those of the locus coeruleus, and such cells, and those alone, give rise to mesencephalic root fibers, according to Weinberg (’28 ; see also Schwalbe, '81, and Ram6n y Cajal, ’09). Possibly differences in interpretation as to what constitutes the nucleus of the locus coeruleus may offer an explanation of these differences of opinion, as Sheinin (’30) suggested. The locus coeruleus with the pertinent literature is discussed in Chapter VI, p. 662.

Resume of the Developjient of the Bhanchiae Nerves and of Their Afferent Connections

Tlie medulla oblongata differs from the spinal cord in the greater development of its dorsal roots as compared with its ventral roots. The underlying cause for the hypertrophy of the dorsal roots is to be found in the development of the gill apparatus and the sense organs of the region. These latter arc considered in the next chapter. The nerves of the branchial arch apparatus must be regarded as .specialized components of the dorsal roots which originally (see Amphioxus) contained, in addition to somatic afferent and visceral afferent fibers, visceral efferent fibers as well. Such roots, even in primitive vertebrates, do not join the ventral efferent roots of tlic region.

In .•Vmphioxus the most cephalic of the dorsal roots show differentiation. Thus the most cephalic dorsal root fNer\’e II) of Amphioxus, in con.sequence

TASTE AND THE GENERAL AFFERENT SYSTEMS

407

of an atrophy of the proximal portion of the body, contains only somatic afferent fibers. This nei’ve probably corresponds to the ramus ophthalmicus trigemini of craniotes, since it emerges in front of the second head myotome. The secondary dorsal nerve, emerging behind the second myotome, corresponds with the ramus maxillo-mandibularis of higher forms. In Amphioxus this nerve carries both somatic afferent and visceral afferent fibers, but lacks the visceral efferent fibers characteristic of the more caudally emerging dorsal roots of the buccal and peribranchial region. These roots of the buccal and peribranchial region are the forerunners of the facial, glossophar 5 mgeal, and vagus nerves, since their visceral efferent fibers are concerned in innervating the muscles of respiration (in Amphioxus the transverse muscles of the peribranchial ca%dty).

In Amphioxus the branchial region is relatively very large and it has not been possible to determine the line of separation between the brain and the cord. Hence nerves suppljdng this indeterminate region are called branchio-spinal nerves. Their number is much greater than that of the true branchial nerves in craniotes. This is not surprising, because even among craniotes there is a progressive reduction of the caudal branchial nerves, or at least of the branchial arches with which they are associated. Thus, while in Amphioxus there are 36, there are 13 in Myxine, 8 in Petromyzon, 7 in Heptanchus, 6 in Hexanchus, and in all other sharks only 5 branchial arches. YTiile the caudal branchial nerves are undergoing reduction, the anterior branchial nerves, that is, the branchial nerves as the term is applied to craniotes, are becoming increasingly important. This is due to the increase in visceral afferent and visceral efferent fibers resulting from the enlargement of the mucous membrane of the gills and the development of true branchial musculature.

Other secondarj changes of the original dorsal roots occurring in the vagus, glossopharyngeal, and facial nerves are :

(1) The addition of elements from the branchial cleft organs of Froriep (epibranchial placodes of von Kupffer) to their ganglia.

(2) The development of taste buds in the regions supplied by these nerves (these are still lacking in Amphioxus).

(3) The progressive reduction of fibers carrying ordinary tactile sensibility from the skin. Tliis reduction runs parallel with the specialization of the somatic afferent elements of the labyrinth and the lateral fine system, and the increase in the cutaneous fibers within the trigeminal nerve.

The enormous increase in the somatic sensory systems, particularly at the level of the facial, acoustic, and lateral line roots, pulls lateralward the dorsal parts of the alar plates of the medulla oblongata so that the central canal is greatly widened out to form a fourth ventricle {Ingvar; Ariens Kappers) and the calamus scriptorius, which in Amphioxus hes behind the third dorsal nerve, is shifted caudalward.

Cutaneous sensory fibers are found in the trigeminal, the facial, the vagus, and possibly the glossopharyngeal nerves in cyclostomes and amphibians. In some selachians and teleosts the facial root still has cutaneous fibers, and certain

408 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

observers consider that the nerve has a few cutaneous fibers even in mammals. However, this is still to be confirmed. In other animals cutaneous branches to the head, in so far as they are supplied by cranial nerves, run in the trigeminal and the vagus, with possibly a few in the glossopharyngeus. The trigeminal carries the major portion of such fibers. In man the only cutaneous branch of the vagus is the ramus auricularis to the external ear, and cutaneous fibers of the glossopharyngeal, if present, are fimited to a few fibers to the same region. After entrance to the medulla oblongata, the somatic afferent fibers of the facial glossopharyngeal, and vagus, carrying pain and general tactile sensibility, separate from the visceral afferent fibers of these nerves, become associated wholly or in part with the descending root of the trigeminal nerve, and end in its associated gray. This is an excellent example of the reorganization within the medulla oblongata of fibers into patterns governed by functional relations of the components rather than by their peripheral courses.

The general visceral afferent fibers and the taste fibers enter by way of the facial, glossopharyngeal, and vagus nerves. The visceral afferent fibers of the glossopharyngeal and vagus originally terminate close to their planes of entrance in the visceral sensory area near the floor of the ventricle. Incoming sensory root fibers of the facial turn caudalward, forming the prevagal (and preglossopharyngeal) portion of the fasciculus solitarius, and likewise come into synaptic relations with this visceral sensory area. Such a condition is found in many lower forms. In those fishes in which taste buds are present over the surface of the body as well as on the inside of the mouth, the number of visceral sensory fibers in the facial is very large and the visceral sensory areas reached by its incoming fibers are increased greatly in size. Thus a conspicuous enlargement of the visceral sensory area of the medulla, the gustatory lobe of the facial, is formed. A similar lobe may appear in these animals at the level of the glossopharyngeal and vagus, usually termed the vagal lobe. A postvagal fasciculus solitarius is present in these forms but is small.

Above fishes the primary gustatory centers decrease in size mth the decrease in the relative number of taste buds at the periphery. The amount of visceral sensory or visceral afferent gray in the nuclei of the fasciculus solitarius and the size of the descending facial root in this fasciculus in front of the level of entrance of the glossopharyngeal nerve vary with the degree of development of taste buds in the region supplied by the nerve. In certain mammals this number is relatively large on the anterior two-thirds of the tongue, but in man taste buds disappear from this area and the amount of taste fibers in the facial is almost negligible. Since the number of visceral afferent fibers carried by the facial nerve is relatively very small, being mainly from the submaxillary and sublingual glands, the visceral sensory root of the nerve also is small and the fasciculus solitarius back to the level of entrance of the glossopharyngeal is hard to differentiate in ordinary preparations.

Even in fishes there are certain descending fibers of the glossophaiyngeal and the vagus which pass caudalward and form the post-glossopharyngeal and postvagal portion of the fasciculus solitarius. They come into synaptic relation

TASTE AND THE GENERAL AFFERENT SYSTEMS

409

with the nucleus of the commissura infima, after a partial decussation in that commissure. From fishes on, this postvagal portion becomes of greater importance. It is better developed in amphibians and reptiles and forms a conspicuous bundle in mammals and particularly evident fiber tracts in birds. This increase in the postvagal portion of the fasciculus solitarius and the further differentiation of the commissura infima and its associated gray are associated with the increase in the general visceral sensory components of the glossopharyngeal and vagus nerves. The relative sizes of the visceral sensory centers of the medulla oblongata are probably associated with the degree of development of gustatory sensibility. Evidence for this interpretation is found particularly in the small size of the visceral sensory centers in the medulla oblongata and the great size of the postvagal fasciculus solitarius in animals such as birds, where taste is greatly atrophied. It is of interest that a development of the postvagal fasciculus solitarius is first clearly evident in those animals in which respiration through the lungs substitutes for that by means of gills. This is perhaps due to the development within the cord of specialized motor centers for the innervation of the diaphragm and the consequent tendency of the visceral sensory impulses to reach such centers by the shortest possible path. In some animals the fasciculus soUtarius has been seen to descend into the cervical cord, and secondary descending neurons arising from the gray associated with this fasciculus have been traced at least as far caudalward as the fourth cervical segment. The taste fibers probably remain localized in the medulla oblongata because the most important reflexes produced by the stimulation of taste buds involve muscles innervated by motor nuclei of this region. Nucleus intercalatus appears to be best developed in those animals, such as mammals, in which the tongue has become an organ of exploration. It is lacking in animals which do not have a muscular tongue and is larger in mgTnTnals with many taste buds on the anterior part of the tongue (such as rodents and edentates) than in those in which the sense of taste on the tongue is atrophic (Cetacea). It occupies a position between the hypoglossal nucleus which supplies motor fibers to the tongue and the dorsal efferent nucleus which carries preganglionic fibers for innervation of the viscera, including the stomach. It is possible, then, that this nucleus is to be regarded as a gustatory center {Ariens Kappers), and that its position is such as is most favorable for the furthering of gustatory reflexes. The various views regarding the functional relations of nucleus intercalatus have been referred to earlier and need not be reviewed here (see pages 376 and 377).

The pathways for gustatory impulses to higher centers are best known in certain bony fishes. In general, from the bulbar gustatory centers fibers pass to a secondary gustatory nucleus in the isthmus region ; after a sjmapse there, the impulses are carried to the hypothalamus. The details of these connections are illustrated in figure 147. The secondary gustatory nucleus in the frog has also been identified, as well as in the chameleon, but little is known of the details of ascending gustatory connections in any forms above fishes. In mammals, projection fibers from the gustatory centers of the medulla oblongata to the ventral nucleus of the thalamus have been described and other observers have

410 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

assumed a hypothalamic connection, but the evidence for neither of these connections is conclusive.

The sensory trigeminal root probably originates from a union of the two frontal dorsal nerves of Amphioxus (second and third septal nerves), one of which emerges in front of the second myotome and the other behind it. The ophthalmic and maxillo-mandibular rami of the trigeminal keep these same relations during development in higher forms. Two such well-separated roots are still present in mammals during embryonic development. These nerves, in contrast with the other branchial nerves of craniotes, show an atrophy of the visceral sensory components and an increase of the somatic sensory component. There is neither sufficient phylogenetic nor pathologic evidence to prove that the trigeminal nerve innervates taste buds in any form. Taste buds situated in the region of its peripheral distribution receive their innervation through the sensory branches of the facial. Those fibers of the trigeminal nerve innervating the oral cavity supply that portion of it which is embryologically ectodermal in origin, and carry pain, temperature, and tactile impulses.

Incoming cutaneous sensory fibers of the trigeminal nerve, after entrance to the medulla oblongata, turn caudalward to form the descending root of the trigeminal. Such fibers are present from cyclostomes to man and carry pain and general tactile sensibility. In many higher animals those carrying pain are smaller and have either a thinner medullary sheath or are nonmedullated. They terminate in large part in the nucleus of the descending root. In many forms the ophthalmic branches of this root are the more ventral and extend farthest caudally, even reaching the cervical cord, while the more dorsally placed maxillo-mandibular fibers terminate largely in the medulla oblongata in relation to the gray associated with the root. Thus stimulations on the top of the head lead to motor responses from trunk and limbs, while those coming in from the mouth region find expression on the motor side in movements of the face and the jaw through innervation of cranial nerves. In higher animals, beginning with frogs and evident in reptiles, there appears a distinct chief sensory nucleus at the level of entrance of incoming fibers. This nucleus increases in size in birds and mammals. In the birds it is distinct from the gray accompanying the descending root, but in mammals the two are continuous. There is evidence that this chief sensory nucleus is concerned particularly with the terminations of the larger medullated fibers. It appears phylogenetically with the appearance of peripheral tactile corpuscles and has to do, in all probability, particularly with tactile discrimination. In the cat, the larger, more heavily medullated sensory trigeminal fibers, on entrance to the medulla oblongata, dichotomize and terminate partly in the chief sensory nucleus and partly in the nucleus of the descending root. Other medullated, non-bifurcating fibers end only in the chief sensory nucleus. Both types probably carry tactile sensibility {Windle). The smaller medullated fibers and the unmedullated fibers swing caudalward to terminate in relation with the nucleus of the descending root. Such fibers probably carry pain, temperature, and more indefinitely localized tactile sensibility. Short fibers connect the sensory nuclei of the trigeminal nerve with near-lying motor centers.

TASTE AND THE GENERAL AFFERENT SYSTEMS

411

Secondary ascending pathways from both the chief sensory nucleus and the nucleus of the descending root pass to the cerebellum in reptiles, birds, and mammals. In mammals two major ascending pathways connect these trigeminal sensory nuclei with the thalamus. The first of these paths, the ventral ascending secondary tract of the trigeminal, arises from the nucleus of the descending root and probably from the chief sensory nucleus as well. Its fibers decussate and then turn forward with other of the great ascending lemnisci systems to pass (at least in man) to the ventral nucleus (and perhaps other centers) of the thalamus (see Chapter VIII). In its course this tract gives off fibers to the midbrain. In forms below mammals this tract has been demonstrated with certainty only as far as the midbrain. The dorsal secondary ascending tract of the trigeminal nerve, or trigeminal lemniscus proper, arises from the chief sensory nucleus, decussates in part at its level of origin, and passes forward to the ventral nucleus of the thalamus, also giving off in its course connections to the tectum. The trigemino-tectal portion of this path is present in reptiles and birds. Its thalamic component has not been shown with certainty below mammals. Although such direct trigeminal connections to the thalamus are yet to be demonstrated, the influence of the tectal component upon thalamic development in reptiles and birds is made clearly possible through' the tremendous development of their tecto-thalamic system.

A quinto-f rental tract, directly from the chief sensory nucleus to the hemisphere, has been demonstrated in birds. A homologous tract has been described for rabbits, although the connection in this latter case has only been carried as far as the entopeduncular nucleus. The projection of tactile, temperature, and pain impulses directly upon the hemisphere, presumably for correlation with olfactory impulses, has led to the theory of hemisphere development through the influence of the oral sense. An analogous tract is the dorsal secondary ascending tract of mammals {Woodbume).

The cells of the mesencephalic root of the trigeminal are comparable in many respects to intra- and suprameduUary ganglion cells of lower forms. Their usual course peripheralward in association with the mandibular root is due probably to the similar peripheral distribution in jaw muscles. The mesencephalic root is concerned in carrying proprioceptive impulses from the muscles of mastication to the central nervous system and there setting up reflexes through their connection with the motor nucleus of the Vth which makes possible coordinated jaw movements. It is probable that the so-called mesencephalic nucleus of the trigeminal receives proprioceptive impulses from eye-muscle terminations running peripherally in the eye-muscle nerves. In selachians the cells of the mesencephalic nucleus of the trigeminal occur throughout the whole tectum, close to the midline. In teleosts they lie near the anterior border of the tectum. In amphibians they are in the posterior part of the tectmn near the midline. In reptiles, with the exception of the hydrosaurians, they extend throughout the length of the tectum, occurring in greatest numbers in the posterior part. In hydrosaurians they are found throughout the tectum but are most numerous in the frontal portions. In birds the cells of the mesencephalic

412 KERVOUS SYSTEMS OF ^^RTEBRATES AND OF :MAN

root lie in or near the anterior medullarj'- velum and lateral to it. In manj* mammals they have shifted still farther caudalward, and in rodents they show a particular accumulation at the level of the motor trigeminal nucleus. It seems relatively evident from the above statement that the primitive position of these cells is in the midbrain region, owing to the fact that this portion of the brain is primitively associated wdth the ophthalmic ramus. Their caudal migration must be considered due to neurobiotaxis, that is, their tendenc)’’ to migrate along their dendrites toward the source of their stimulation and also, in this case, toward the major nucleus with which they lie in sjmaptic relation, that is, the motor nucleus of the trigeminal. This is most striking in mammals which chew their food. The retention of a midbrain position of certain other of the cells may be due to their intimate functional relations with the oculomotor and trochlear centers. The secondary central connections of the mesencephalic trigeminal nucleus, so far as knowm, have been listed on page 406. There is considerable e\ddence that the cells give off collaterals to the tectum, the lateral reticular nucleus of the midbrain, the cerebellum, and the locus coeruleus, and most observers are agreed that their neuraxes terminate in the motor nucleus of the trigeminal and possibly in those of facial and glossophar^mgeal nerves.

BIBLIOGRAPHY

CEnTAiN General References on Medulla Oblongata and the Theory of the

Nerve Components

{The lUcralitre on choroid plexus is found in Chapter 1.)

AriKns Kappers, C. U. 1920-21. Vcrpleichcndc Anatomic dcs Nervens^’stems.

(German edition of present text.) E. F. Bohn, Haarlem.

Blake, J. A. 1900. The roof and lateral recesses of the fourth ventricle considered morphologically and embiymlogicallj'. J. Comp. Neurol., vol. 10, p. 79.

Cannieu, a. 1897. Contribution it IT-tude dc la voGve du quntriome ventriculc chez Ics mammifdres ; le trou dc Magendie. J. dc m6d. dc Bordeaux, vol. 27, p. .'147.

Edinger, L. 1908. ^'orlcsungen iiber den Ban der nciwoscn Ccntralorgane dcs Mcnschen und der Thicrc. 7‘' .'Viifl., F. C. W. Vogel, lA?ipzig.

Engel, E. A. 1908. Ueber die Sekretionscrschcinungcn in den Zcllcn der Plexus chorioidci de.-= Mcnschcn. Arch. f. Zcllforsch., Bd. 2, .S. 191.

Gabkell, W. II. 18SG. On the structure, distribution, and function of the nerves which innen'ate the visceral and vascular .systems. ,1. Pliysiol., vol. 7.

. 1889. On the relation between the stnicture, function, distribution, and origin

of the cranial nerves, together with a theory of the origin of the nervous .system of vertebrates. .1. Physiol., vol. 10.

IlEunicK, C. .1. 1899. The cranial and first spinal nerx'cs of Mcnidia. A contribution

upon the nerve components of the bony fishes. ,1. Comp. Neurol., vol. 9, P- 21. See aho Arch. Neurol, and P.sychopafh., vol. 2, p. 1.

. 191-4. Cnini.al nerves. Wood’s Reference Handbook of Medical Science, vol. .9,

p. 921.

. 1924. Neurological foundation"^ of liehavior. H. Holt and Co., New York.

. 192.') The innervation of palatal taste bud.“ and teeth of Arnbly.^tornn. J. Comp.

Neurol., vol. ',‘S, p. 989.

TASTE AND THE GENERAL AFFERENT SYSTEMS

413

Herrick, C. J. 1931. An introduction to neurology. S**" edition. W. B. Saunders Co., Philadelphia and London.

Hess, C. 1885. Das Foramen Magendii und die Offnungen an den Recessus laterales des IV Vcntrikels. Morphol. Jahrb., Bd. 10, S. 578.

Heuser, C. H. 1913. The development of the cerebral ventricles in the pig. Am. J. Anat., vol. 15, p. 215.

His, AV. 1890. Die Entwickelung des menschlichen Rautenhirns, vom Ende des ersten bis zum Beginn des dritten Monats. S. Hirzel, Leipzig.

. 1903. Die Haute und Hohlen des Korpers. Arch. f. Anat. u. Physiol., Anat.

Abt., S. 368.

. 1904. Die Entwickelung des menschlichen Gehirns wahrend der ersten Monate.

S. Hirzel, Leipzig.

Hworostuchin, H. 1911. Zur Frage fiber den Bau des Plexus chorioideus. Arch, f. mikr. Anat., Bd. 77, S. 232.

Jelgersma, G. 1932. Atlas anatomicum cerebri humani. Amsterdam.

Johnston, J. B. 1902. An attempt to define the primitive functional divisions of the central nervous system. J. Comp. Neurol., vol. 12, p. 87.

. 1902. Das Gehim und die Cranialnerven der Anamnier. Ergeb. d. Anat. u.

Entwicklungsgesch. (Merkel u. Bonnet), Bd. 11, S. 973.

. 1905. The morphology of the vertebrate head from the viewpoint of the functional divisions of the nenmus system. J. Comp. Neurol., vol. 15, p. 175.

. 1906. The nervous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

Key, E. a. H., and Retzius, G. 1875-76. Studien in der Anatomie des Nervensystems und des Bindegewebes. Teil 1. Samson and Wallin, Stockholm.

Kuntz, a. 1931. A textbook of neuro-anatomy. Lea and Febiger, Philadelphia.

VON Kupffer, K. 1906. Die Morphogenie des Centralnervensystems. Handbuch der vergleichenden und experimentellen Entwickelungslehre der Wirbeltiere, 0. Hertwig, Bd. 2, Teil 3, S. 1. J. Fischer, Jena.

VON Luschka, H. 1855. Die Adergeflechte des menschlichen Gehimes. G. Reimer, Berlin.

Marburg, 0. 1910. Mikroskopisch-topographischer Atlas des menschlichen Zentral nervensystems. 2" Aufl., F. Deuticke, Leipzig und Wien.

Mingazzini, G. 1928. Medulla oblongata und Brficke. G. von Mollendorf’s Handbuch der mikroskopische Anatomie des Menschen. Nervensystem, Bd. 4, S. 579. J. Springer, Berlin.

VON Monakow, C. 1905. Grosshimpathologie. 2** Aufl., A. Holder, Wien. S. 128.

Obersteiner, H. 1901. Anleitung beim Studium des Baues der nervosen Zentralorgane. 4‘” Aufl., F. Deuticke, Leipzig und Wien. S. 490 u. 330.

Osborn, H. F. 1888-. A contribution to the internal structure of the amphibian brain. J. Morphol., vol. 2, p. 151.

Papez, j. W. 1929. Comparative neurology, Thomas Y. Crowell Co., New York.

Ramon y Cajal, S. 1909-11. Histologic du systfime nerveux de I’homme et des vert^bres. A. Maloine, Paris.

Ranson, S. W. 1931. The anatomy of the nervous system. 4‘ edition. W. B. Saunders Co., Philadelphia and London.

ScHROEDER VAN DER Kolk, J. L. C. Bau und Functionen der Medulla spinalis und oblongata. Aus dem Hollandischen fibertragen von Dr. F. W. Thiele. F. Vieweg und Sohn, Braunschweig.

Sherrington, C. S. 1906. The integrative action of the nervous system. C. Scribner’s Sons, New York.

Sterzi, G. 1915. Anatomia del sistema nervoso centrale dell’uomo. Vol. 1, p. 356. A. Draghi, Padova.

414 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Stheetee, G. L. 1912. The development of the nervous system. Manual of human embryology, vol. II, Keibel, F., and Mall, F. P., J. B. Lippincott Co., Philadelphia and London.

Strong, 0. S. 1890. The structure and homologies of the cranial nerves of the Amphibia as determined by their peripheral distribution and internal origin. Zool. Anz., Bd. 23, S. 598.

. 1892. Idem. Anat. Anz., Bd. 7, S. 467.

. 1895. The cranial nerves of the Amphibia. A contribution to the morphology

of the vertebrate nervous system. J. Morphol., vol. 10, p. 101.

Tilney, F. 1926. The brains of ape and man. P. Hoeber, New York.

Tilney, F., and Riley, H. 1921. The form and functions of the central nervous system. P. Hoeber, New York.

ViLLiGEE, E. 1925. Brain and spinal cord. Edited by Addison. J. B. Lippincott Co., Philadelphia and London.

Weed, L. H. 1917. The development of the cerebro-spinal spaces in pig and in man. Contrib. Embryol., Publ. Carnegie Inst., Washington, vol. 5, no. 14, p. 1. Publ. no. 225 of the Carnegie Institution.

VAN WijHE, J. W. 1882. Ueber die Mesodermsegmente und die Entwickelung der Nerven des Selachierkopfes, J. Muller, Amsterdam. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, vol. 22, 1883.

Wilder, B. G. 1886. Notes on the foramina of Magendie in man and the cat. J. Nerv. and Ment. Dis., vol. 13, p. 206.

. 1893. The metapore (foramen of Magendie) in man and Orang. Med. News,

Philadelphia, vol. 63, p. 439.

Winkler, C., and Potter, A. 1911. An anatomical guide to experimental researches on the rabbit’s brain. W. Versluys, Amsterdam.

. 1914. An anatomical guide to experimental researches on the cat's brain.

W. Versluys, Amsterdam.

Ziehen, T. 1899. Anatomic des centrales Nervensystems. Bd. 4, S. 1. G. Fischer, Jena.

. 1905. Die Morphogenie des Centralnervensystems der Saugetiere. Handbuch

der vergleichenden und experimentellen Entwickelungslehre der Wirbeltiere, O. Hertwig. Bd. 2, Teil 3, S. 273. G. Fischer, Jena.

Literature on the Sensory Root of the Vagus, Glossopharyngeal, and Facial

Nerves

the distribution op the taste buds and papillae Cyclostomes

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

. 1905. The cranial nerve components of Petromyzon. Morphol. Jahrb., Bd. 34,

S. 149.

Retzius, G. 1893. Ueber Geschmacksknospen bei Petromyzon. Biol. Untersuch., N. F., Bd. 5, S. 69.

Wiedersheim, R. E. E. 1909. Grundriss der vergleichenden Anatomic der Wirbelthiere. 7‘' Aufl., G. Fischer, Jena.

Plagiostomes

Oppel, a. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomic der Wirbeltiere. Mundhohle, Bauchspeicbeldriise und Leber, Bd. 3. G. Fischer, Jena, 1896-1914.

TASTE AND THE GENERAL AFFERENT SYSTEMS

415

Sheldon, R. E. 1909. The reactions of the dogfish to chemical stimuli. J. Comp. Neurol., vol. 19, p. 273.

Ganoids and teleosts

Allis, E. P. 1897. The cranial muscles and cranial and first spinal nerves in Amia calva. J. MorphoL, vol. 12, p. 487.

Herrick, C. J. 1900. A contribution upon the cranial nerves of the codfish. J. Comp. Neurol., vol. 10, p. 265.

. 1901. The cranial nerves and cutaneous sense organs of the North American

siluroid fishes. J. Comp. Neurol., vol. 11, p. 177.

. 1902. The organ and sense of taste in fishes. Bull. U. S. Fish. Commission,

1902; also. Bull. Sc. Lab. Denison Univ., vol. 12, 1903, art. 6, p. 39.

. 1903. On the morphological and physiological classification of the cutaneous

sense organs of fishes. Am. Naturalist, vol. 37, p. 313.

. 1903a. On the phylogeny and morphological position of the terminal buds of

fishes. J. Comp. Neurol., vol. 13, p. 121.

. 1907. A study of the vagal lobes and funicular nuclei of the brain of the codfish.

J. Comp. Neurol., vol. 17, p. 67.

. 1908. On the phylogenetic differentiation of the organs of smell and taste.

J. Comp. Neurol., vol. 18, p. 157.

Horschelmann. 1866. Ein Beitrag zur Anatomie der Zunge der Fische. Dissertation, Dorpat.

Johnston, J. B. 1901. The brain of Acipenser; a contribution to the morphology of the vertebrate brain. G. Fischer, Jena.

Kamon, K. 1904. Ueber die “Geruchsknospen.” Arch. f. mikr. Anat., Bd. 64, S. 653.

Landacre, F. L. 1907. On the place of origin and method of distribution of taste buds in Ameiurus melas. J. Comp. Neurol., vol. 17, p. 1.

VON Lenhoss^k, M. 1893. Der feinere Bau imd die Nervenendigungen der Gesmachsknospen. Anat. Anz., Bd. 8, S. 121.

. 1894. Die Endknospen der Barbe und des Aales. Beitrage zur Histologie des

Nervensystems und der Sinnesorgane. J. F. Bergmann, Wiesbaden.

VON Lbtdig, F. 1853. Anatomisch-histologische Untersuchungen fiber Fische und Reptilien. G. Reimer, Berlin.

. 1857. Lehrbuch der Histologie des Menschen und der Thiere. Meidinger,

Sohn u. Comp., Frankfurt a Main.

May, R. 1925. The relation of nerves to degenerating and regenerating taste buds. J. Exper. Zool., vol. 42, p. 371.

Olmsted, J. M. D. 1920. The nerve as a formative influence in the development of taste buds. J. Comp. Neurol., vol. 31, p. 465.

. 1920a. The results of cutting the seventh cranial nerve in Ameiurus nebulosus

(Lesueur). J. Exper. Zool., vol. 31, p. 369.

Oppel, a. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomie der Wirbeltiere. Mundhohle, Bauchspeicheldrfise und Leber, Bd. 3. G. Fischer, Jena, 1896-1914.

Parker, G. H. 1908. The sense of taste in fishes. Science, N. S., vol. 27, p. 453.

Amphibians

Beigel-Klaften, C. 1917. Ueber Plasmastrukturen in Sinnesorganen und Drfisenzellen des Axolotls. Arch. f. mikr. Anat., Bd. 90, S. 39.

Fixsen, C. 1857. De linguae raninae textura disquisitiones microscopicae. H. Laakmanni, Dorpat.'

416 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Harrison, R. 1904. Experimentelle Untersuchungen iiber die Entwicklung der Seitenlinie bci den Amphibien. Arch. f. mikr. Anat., Bd. 63, S. 35.

Hartmann, R. 1863. Ueber die Endigungsweise der Nerven in den Papillae fungifonne.s der Froschzunge. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 634 u. 710.

Herrick, C. J. 1925. The innervation of palatal taste buds and teeth in Amblystoma. J. Comp. Neurol., vol. 38, p. 389.

Johnston, J. B. 1910. The limit between ectoderm and entoderm in the mouth, and the origin of taste buds. I. Amphibians. Am. J. Anat., vol. 10, p. 41.

Kolmer, W. 1910. Ueber Strukturcn im Epithel der Sinnesorgane. Anat. Anz., Bd. 36, S. 281.

VON Leydig, F. 1853. Anatomisch-histologische Untersuchungen iiber Fische und Reptilien. G. Reimer, Berlin. (Also contains contributions on Amphibia.)

Oppel, a. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomic der Wirbeltiere. Mundhohlc, Bauchspeicheldriise und Leber, Bd. 3. G. Fischer, Jena, 1896-1914.

Retzius, G. 1892. Die Nervenendigungcn in dem Gcschmacksorgan der Saugetieren und Amphibien. Biol. Untersuch., N. F., Bd. 4, S. 19 ; also Bd. 13, 1906.

. 1892a. Ueber die sensiblen Nervcnendigung in den Epithelien bei den Wir beltieren. Biol. Untersuch., N. F., Bd. 4, S. 33.

. 1905. Zur Kenntnis dor Nervenendigungcn in den Papillen der Zunge der Amphibien. Biol. Untersuch., N. F., Bd. 12, S. 61.

Strong, 0. S. 1890. The structure and homologies of the cranial nerves of the Amphibia as determined by their peripheral distribution and internal origin. Zool. Anz., Bd. 23, S. 598. See also Anat. Anz., Bd. 7, S. 467.

. 1895. The cranial nerves of the Amphibia. A contribution to the morphology

of the vertebrate nervous system. J. Morph., vol. 10, p. 101.

Reptiles

Bath, W. 1906. Die Geschmacksorgane der Vogel und Krokodilc. Arch. f. Biontol., Bd. 1, S, 1.

VON Leydig, F. 1853. Anatomisch-histologische Untersuchungen uber Fische und Reptilien. G. Reimer, Berlin.

. 1872. Die in Deutschland lebenden Arten der Saurier. H. Laupp'sche Buch handlung, Tubingen.

. 1872a. Zur Kenntniss der Sinnesorgane der Schlangen. Arch. f. mikr. Anat.,

Bd. 8, S. 317.

Merkel, F. 1880. Ueber die Endigungen der sensiblen Nerven in der Haut der Wirbelthiere. H. Schmidt, Rostock.

. 1892. Sinnesorgane und Geschmacksorgane. Zeitschr. f. d. Anat.; Abt. 3,

Ergebn. d. Anat. u. Entwicklungsgesch., Bd. 2-3 (1892-’93).

Oppel, A. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomic der Wirbeltiere. Mundhohlc, Bauchspeicheldriise und Leber, Bd. 3. G. Fischer, Jena, 1896-1914.

Tuckerman, F. 1892. On the terminations of the nerves in the lingual papillae of the Chelonia. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 9, S. 1.

Birds

Bath, W. 1906. Die Geschmacksorgane der Vogel und Krokodile. Arch. f. Biontol., Bd. 1, S. 1.

Oppel, A. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomic der Wirbeltiere. Mundhohle, Bauchspeicheldriise und I^eber, Bd. 3. G. Fischer, Jena, 1896-1914.

TASTE AND THE GENERAL AFFERENT SYSTEMS

417

Mammals

Boeke, J. 1925a. Die intracellulare Lage der Nervenendigungen im Epithelgewebe und ihre Beziehungen zum Zellkern. Jahrb. f. Morphol. u. mikr. Anat., Abt. 2, Zeitschr. f. mikr. Anat. Forschung., Bd. 2, S. 341.

DE Castro, F. 1916. Nota sobre la disposicion de aparato reticular de Golgi en los botones gustativos. Trab. d. lab. de invest, biol. Univ. de Madrid, vol 14 p. 107.

Davis, C. 1877. Die becherformigen Organe des Kehlkopfes. Arch. f. mikr. Anat., Bd. 14, S. 158.

Disse, J. 1895. Ueber Epithelknospen in des Regio olfactoria der Sauger. Anat. Hefte, Bd. 6, S. 21.

Dbasch, 0. 1887. Untersuchungen tiber die Papillae foUiatae et circumvallatae des Kaninchen und Feldhasen. Abhandl. d. math.-phys. IGasse d. kon. siichs. Gesellsch. d. Wissensch., Bd. 14, S. 229.

VON Ebner, V. 1897. Ueber die Spitzen der Geschmacksknospen. Sitzungsb. d. K. Akad. d. wissensch. zu Wien, Math.-nat. Cl., Bd. 106, Abt. 3, S. 73.

Grabehg, j. 1899. Zur Kenntnis des cellularen Baues des Gesmacksknospen beim Menschen. Anat. Hefte, Bd. 12, S. 337.

Griffini, L. 1887. Sulla reproduzione degli organi gustatorii. Rendic. d. R. 1st. Lombardo di sc. e lett., Ser. 2, vol. 20, p. 667.

Grossman, B. 1921. Ueber das Vorkommen von Gesmacksknospen an der Vorde^ wand der Pars laryngea pharyngis beim Menschen. Monatschr. f. Ohrenh., Bd. 5^ S. 1174.

Hermann, F. 1885. Beitrag zur Entwicklungsgeschichte des Geschmacksorgans beim Kaninchen. Arch. f. mikr. Anat., Bd. 24, S. 216.

. 1888. Studien iiber den feineren Bau des Gesmacksorgane. Sitzungsb. d.

math.-phys. Cl. d. k.-bayer. Akad. d. Wissensch. zu Munchen, Bd. 18, S. 277.

Heymann, R. 1889. Beitrag zur Kenntnis des Epithels und der Driisen des menschlichen Kehlkopfes im gesunden und im kranken Zustande. Virchow’s Arch. f. path. Anat., Bd. 118, S. 320.

Kalltus, E. 1905. Geschmacksorgan. K. von Bardeleben’s Handbuch der Anatomic des Menschen. G. Fischer, Jena, 1896-1909.

Kolmer, W. 1910. Ueber Strukturen im Epithel der Sinnesorgane. Anat. Anz., Bd. 36, S. 281.

. 1927. Geschmacksorgan. G. von Mollendorf’s Handbuch der mikroscopischen

Anatomie des Menschen, Bd. 1, S. 154. J. Springer, Berlin.

Krause, G. 1908. Ueber die Papillae filiformes des Menschen. Dissertation, Konigsberg.

Krause, W. 1870. Die Nervenendigungen in der Zunge des Menschen. Nachr.

V. d. k. Gesellsch. d. Wissensch., Gottigen, Math.-phys. Klasse, no. 18, S. 331.

von Lenhoss^k, M. 1893. Der feinere Bau und die Nervenendigungen der Gesmachsknospen. Anat. Anz., Bd. 8, S. 121.

Merkel, F. 1880. Ueber die Endigungen der sensiblen Nerven in der Haut der Wirbelthiere. H. Schmidt, Rostock.

Olmsted, J. M. D. 1921. Effect of cutting the lingual nerve of the dog. J. Comp. Neurol., vol. 33, p. 149.

Oppel, A. 1900. Lehrbuch der vergleichenden mikroskopischen Anatomie der Wirbeltiere. Mundhohle, Bauchspeicheldriise und Leber, Bd. 3. G. Fischer, Jena, 1896-1914.

Patzelt, V. 1921. Die Ergebnisse eines Untersuchung fiber die Histologic und Histogenese der menschlichen Epiglottis unter besonderer Berficksichtigung der Metaplasiefrage. Anat. Anz., Bd. 5, S. 161.

418 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Patzelt, V. 1923. Ueber die menschliche Epiglottis und die Entwicklung des Epithels in den Nachbargebieten. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 70, S. 1.

Plate. 1925. Allgemeine Zoologie. Vol. 2.

PouLTON, E. B. 1883. The tongue of Perameles nasuta, with some suggestions as to the origin of taste bulbs. Quart. J. Micr. Sc., vol. 23, p. 69.

. 1883a. On the tongues of the Marsupialia. Proc. Zool. Soc., London, p. 599.

. 1883b. The tongue of Omithorrhynchus paradoxus ; the origin of taste bulbs

and the parts upon which they occur. Quart. J. Micr. Sc., vol. 23, p. 453.

Rawitz, B. 1903. Beitrage zur mikroskopischen Anatomic der Cetaceen. II. Ueber die Zunge von Delphinus delphis. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 20, S. 429.

Retzius, G. 1892. Die Nervenendigungen in dem Geschmacksorgan der Saugetieren und Amphibien. Biol. Untersuch., N. F., Bd. 4, S. 19; see also Bd. 13, p. 117, 1906.

Schaffer, J. 1914. Kleinere histologische Mitteilungen. II. Ueber ein paariges Geschmacksorgan am Eingang der Speiserohre bei Spitzmausen. Verhandl. d. anat. Gesellsch. (Erganzungsheft zum Bd. 46, Anat. Anz.), S. 95.

Schofield, R. H. A. 1876. Observation on taste goblets in the epiglottis of the dog and cat. J. Anat., vol. 10, p. 475.

Stahr, H. 1902. Ueber die Papillae fungiformes der Kinderzunge und ihre Bedeutung als Geschmacksorgan. Zeitschr. f. Morphol. u. Anthropol., Bd. 4, S. 199.

Tuckerman, F. 1888. The tongue and gustatory organs of Fiber zibethicus. J. Anat., vol. 22, p. 135.

. 1889. On the gustatory organs of the American hare, Lepus Americanus.

Am. J. Sc., vol. 38, p. 277.

. 1889a. The gustatory organs of Mammalia. Proc. Boston Soc. Nat. Hist.,

vol. 24.

. 1889b. On the development of the taste-organs of man. J. Anat., vol. 23,

p. 559.

. 1889c. The gustatory organs of Belideus ariel. J. Anat., vol. 24, p. 85.

. 1890. On the gustatory organs of some of the Mammalia. J. Morphol., vol. 4,

p. 151.

. 1890a. The gustatory organs of Procyon lotor. J. Anat., vol. 24, p. 156.

. 1891. On the gustatory organs of Sciurus hudsonius. Intemat. Monatschr. f.

Anat. u. Physiol., Bd. 8, S. 137.

. 1892. The gustatory organs of Ateles ater. J. Anat., vol. 26, p. 391.

Verson, E. 1868. Beitrage zur Kenntniss des Kehlkopfes und der Trachea. Sitzungsb. d. K. Akad. d. Wissensch. zu Wien, Math.-nat. CL, Bd. 57, Abt. 1, S. 1093.

Whiteside, B. 1926. The regeneration of the gustatory apparatus in the rat. J. Comp. Neurol., vol. 40, p. 33,

. 1928. Nerve overlap in the foliate papilla of the rat. (Abstract.) Anat. Rec.,

vol. 38, p. 35.

CHEMICAL sense AND TASTE

VON Anrep, B. 1880. Ueber die physiologische Wirkung des Cocain. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 21, S. 38.

Braeuning, H. 1904. Zur Kenntnis der Wirkung chemischer Reize. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 102, S. 163.

Cole, L. W. 1910. Reactions of frogs to chlorides of ammonium, potassium, sodium, and lithium. J. Comp. Neurol., vol. 20, p. 601.

Herrick, C. J. 1902. Organ and sense of taste in fishes. Bull. U. S. Fish Commission. Bull. Sc. Lab. Denison University. Vol. 12, 1903, art. 6, p. 39.

TASTE AND THE GENERAL AFFERENT SYSTEMS

419

Herhick, C. J. 1906. On the centers for taste and touch in the medulla oblongata of fishes. J. Comp. Neurol., vol. 16, p. 403.

Nagle, W. A. 1894. Ergebnisse vergleichend-physiologischer und anatomischer Untersuchungen fiber den Geruchs- und Geschmackssinn und ihre Organe; mit einleitenden Betrachtungen aus der allgemeinen vergleichenden Sinnesphysiologie. Biol. Centralbl., Bd. 14, S. 543.

Parker, F. H. 1908. The sense of taste in fishes. Science, N. S., vol. 27, p. 453.

. 1912. The relation of smell, taste, and the common chemical sense in vertebrates.

J. Acad. Nat. Sc., Philadelphia, Ser. 2, vol. 15, p. 219.

Sheldon, R. E. 1909. The reactions of the dogfish to chemical stimuli. J. Comp. Neurol., vol. 19, p. 273.

Sherrington, C. S. 1900. Cutaneous sensations. Textbook of physiology, edited by E. A. Schafer, vol. 2, p. 965. Y. J. Pentland, London.

VAN Watenbtjrg, G. A. M. 1897. De beteekenis van reflectorische bewegingen voor de zintuigelijke waameming in verband met de wetten van Weber en Fechner. Scheltema en Holkema, Amsterdam.

ZwAARDEMAKER, H. 1903. Geschmak. Ergeb. d. Physiol. (Asher u. Spiro), Bd. 2^, S. 699.

The Peripheral and Central Relations of the Vagus, Glossopharyngeal,

AND Facial Nerves

CYCLOSTOMES

Ahlborn, F. 1883. Untersuchungen fiber das Gehim der Petromyzonten. Zeitschr. f. wissensch. Zool., Bd. 39, S. 191.

Edinger, L. 1906. Ueber das Gehim von Myxine glutinosa. Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin.

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

. 1905. The cranial nerve components of Petromyzon. Morphol. Jahrb., Bd. 34,

S. 149.

. 1906. The nervous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

. 1908. Additional notes on the cranial nerves of petromyzonts. J. Comp.

Neurol., vol. 18, p. 569.

. 1908a. A note on the presence or absence of the glossopharyngeal nerve in

myxinoids. Anat. Rec., vol. 2, p. 233.

Rothig, P., and Ariens Kappers, C. U. 1914. Further contributions to our knowledge of the brain of Myxine glutinosa. Kon. Akad. v. Wetensoh. te Amsterdam, Proc. sect, sc., vol. 17S p. 2.

Schilling. 1907. Untersuchungen fiber das Gehirn von Petromyzon fluviatilis.

Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main, S. 423.

Tretjakofp, D. 1909. Das Nennusystem von Ammocoetes. II. Gehim. Arch, f. mikr. Anat., Bd. 74, S. 626.

Worthington, J. 1905. The descriptive anatomy of the brain and cranial nenns of Bdellostoma Dombeyi. Quart. J. Micr. Sc., vol. 49, p. 137.

PLAGIOSTOMES

Ariens ICappers, C. U. 1906. The structure of the teleostian and selachian brain.

J. Comp. Neurol., vol. 16, p. 1. . . tt t-.- i

. 1908. Weitere IMitteilungen fiber Neurobiotaxis. II. Die phylogenetische

Entwickelung des horizontalen Schenkels des Facialiswurzelknies. Folia neurobiol., Bd. 2, S. 255.

420 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ariens Kappehs, C. U. 1914. Der Geschmack, perifer und central. Zugleich eine Skizze der phylogenetischen Veranderungen in den sensibelen VII, IX und X Wurzeln. Psychiat. en neurol. Bl., Amsterdam, Bd. 18, S. 82.

Bender, 0. 1906. Ueber die Schleimhautnerven der VII, IX und X in Semens

Forschungsreisen. G. Fischer, Jena.

Cole, F. J. 1896. On the cranial nerves of Chimaera monstrosa ; with a discussion of the lateral line system, and the morphology of the chorda tympani. Tr. Roy. Soc., Edinburgh, vol. 38, p. 631.

Ewart, J. C. 1889. On the cranial nerves of Elasmobranch fishes. Proc. Roy. Soc., London, vol. 45, p. 524.

Ewart, J. C., and Cole, F. J. 1894-95. On the dorsal branches of the cranial and spinal nerves of Elasmobranchs. Proc. Roy. Soc., Edinburgh, vol. 20, p. 475.

Gegenbatjr, C. 1871. Die Kopfnerven von Hexanchus und ihr Verhaltniss zur “Wirbeltheorie” des Schadels. Jenaische Zeitschr. f. Naturw., Bd. 6, S. 497.

Houser, G. L. 1901. The neurones and supporting elements of the brain of a selachian. J. Comp. Neurol., vol. 11, p. 65.

Merritt Hawkes, Mrs. 0. A. 1906. The cranial and spinal nerves of Chlamydoselachus anguineus. Proc. Zool. Soc., London, p. 959.

Strong, 0. S. 1903. The cranial nerves of Squalus acanthias. Science, N. S., vol. 17,

p. 254.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehims der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

GANOIDS AND TELEOSTS

Allis, E. P. 1897. The cranial muscles and cranial and first spinal nerves in Amia calva. J. Morphol, vol. 12, p. 487.

Ariens Kappers, C. U. 1907. Untersuchungen fiber das Gehirn der Knochenganoiden Amia calva und Lepidosteus osseus. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main.

. 1909. Weitere Mitteilungen fiber Neurobiotaxis. III. Ueber den Einfluss

der Neurone der Geschmackskeme auf den motorischen Facialis- und Glossopharyngeuskern und ihr Verhalten zur Radix descendens nervi quinti. Folia neuro-biol., Bd. 3, S. 259.

. 1914. Der Geschmack, perifer und central. Zugleich eine Skizze der phylogenetischen Veranderungen in den sensibelen VII, IX und X Wurzeln. Psychiat. en neurol. Bl., Amsterdam, Bd. 18, S. 82.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

Berkelbach van der Sprenkel, H. 1915. The central relations of the cranial nerves in Silurus glanis and Mormyrus caschive. J. Comp. Neurol., vol. 25, p. 5.

Brickner, R. M. 1930. A new tract in Herrick’s gustatory system in certain teleosts. J. Comp. Neurol., vol. 50, p. 153.

Burr, H. S. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol., vol. 45, p. 33.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbeltniere mit Berficksichtigung der Wirbellosen. Teil 2, W. Engelmann, Leipzig.

Haller, B. 1896. Der Ursprung der Vagusgruppe bei den Teleostiem. Festschr. f. Carl Gegenbaur, Bd. 3. W. Engelmann, Leipzig. (Quoted from Herrick ’05.)

Herrick, C. J. 1899. The cranial and first spinal nerves of Menidia; a contribution upon the nerve components of the bony fishes. J. Comp. Neurol., vol. 9, p. 153.

. 1900. A contribution upon the cranial nerves of the codfish. J. Comp. Neurol.,

vol. 10, p. 265.

TASTE AND THE GENERAL AFFERENT SYSTEMS

421

Hereick, C. J. 1901. The cranial nerves and cutaneous sense organs of the North American siluroid fishes. J. Comp. Neurol., vol. 11, p. 177.

. 1902. Organ and sense of taste in fishes. Bull. U. S. Fish. Commission.

. 1905. The central gustatory paths in the brains of bony fishes. J. Comp.

Neurol., vol. 15, p. 375.

. 1906. On the centers of taste and touch in the medulla oblongata of fishes.

J. Comp. Neurol., vol. 16, p. 403.

. 1907. A study of the vagal lobes and funicular nuclei of the brain of the codfish.

J. Comp. Neurol., vol. 17, p. 67.

. 1908. On the commissura infima and its nuclei in the brains of fishes. J. Comp.

Neurol., vol. 18, p. 409.

Hocke Hoogenboom, K. J. 1929. Das Gehim von Polyodon folium Lacep. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 311.

Holmgren, N., anb van der Horst, C. J. 1925. Contribution to the morphology of the brain of Ceratodus. Acta Zool., Bd. 6, S. 59.

VAN DER Horst, C. J. 1918. See bibliography on p. 299.

. 1927. A cutaneous branch of the facial nerve in a teleost. Kon. Akad. v.

Wetensch. te Amsterdam, Proc. sect, sc., vol. 31, p. 24.

Johnston, J. B. 1901. The brain of Acipenser; a contribution to the morphology of the vertebrate brain. G. Fischer, Jena.

Landacre, F. L. 1910. The origin of the cranial ganglia in Ameiurus. J. Comp. Neurol., vol. 20, p. 309.

. 1914. Embryonic cerebral ganglia and the doctrine of nerve components.

Folia neuro-biol., Bd. 8, S. 601.

Mayser, P. 1881. Vergleichend-anatomische Studien fiber das Gehirn der Knochenfische mit besonderer Berficksichtigung der Cyprinoiden. Zeitschr. f. wissensch. Zool., Bd. 36, S. 259.

Norris, H. W., and Hughes, S. 1920. The cranial, occipital, and anterior spinal nen^es of the dogfish, Squalus acanthias. J. Comp. Neurol., vol. 31, p. 293. Pearson, Jr., Anthony A. 1933. The Acoustico-lateral Centers and the Cerebellum, with Fiber Connections, of Fishes. Dissertation.

Schwartz, L. 1918. Ueber die Entwicklung der Kopfganglien und des Kopfsympathicus bei der Forelle. Folia neuro-biol., Bd. 11, S. 37.

VAN DER Sprenkel. (See Berkelbach van der Sprenkel.)

Stannius, H. 1849. Das peripherische Nervensystem der Fische, anatomisch und physiologisch untersucht. Stiller, Rostock.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehims der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

van Wijhe, j. W. 1879-82. Das Viszeralskelett und die Nen-en des Kopfes der Ganoiden und von Ceratodus. Nederl. Arch. f. Zool., Bd. 5.

AMPHIBIANS

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomic des Nen'ensystems.

(German edition of present text.) E. F. Bohn, Haarlem.

Ariens ICappers, C. U., and Hammer, E. 1918. Das Zentralnervensystem des Ochsenfrosches (Rana catesbyana). Psychiat. en neurol. BL, Amsterdam, Bd. 22, S. 368.

Black, D. 1917. The motor nuclei in phytogeny. II. Amphibia. J. Comp. Neurol., vol. 28, p. 379.

CoGHiLL, G. E. 1902. The cranial nerves of Amblystoma tigrinum. J. Comp. Neurol., vol. 12, p. 203.

422 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Gegenbaxjb, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

Herrick, C. J. 1914. The medulla oblongata of larval Amblystoma. J. Comp. Neurol., vol. 24, p. 1.

. 1925. The innervation of palatal taste buds and teeth of Amblystoma. J. Comp.

Neurol., vol. 38, p. 389.

. 1930. The medulla oblongata of Necturus. J. Comp. Neurol., vol. 50, p. 1.

Kingsbury, B. F. 1895. On the brain of Necturus maculatus. J. Comp. Neurol., vol. 5, p. 139.

Larsell, 0. 1923. The cerebellum of the frog. J. Comp. Neurol., vol. 36, p. 89.

Norris, H. W. 1909. The fifth and seventh cranial nerves in Plethodon glutinosus. Proc. Iowa Acad. Sc., vol. 16, p. 189.

. 1913. The cranial nerves of Siren lacertina. J. Morphol., vol. 24, p. 245.

Norris, H. W., and Buckley, M. 1911. The peripheral distribution of the cranial nerves of Necturus maculatus. Proc. Iowa Acad. Sc., vol. 18, p. 131.

Osborn, H. F. 1888. A contribution to the internal structure of the amphibian brain. J. Morph., vol. 2, p. 51.

von Plessen and Rabinowitz. 1891. Die Kopfnerven von Salamandra maculata. Miinchen.

Rothig, P. 1913. Contributions upon Neurobiotaxis. (The arrangement of the motor nuclei in Myxine glutinosa, in Ciyptobranchus, Necturus, Rana fusca and Bufo.) Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 16‘, p. 296.

. 1927. Beitrage zur Studium des Zentralnervensystems der Wirbeltiere. Jahrb.

f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 10, S. 381,

Strong, 0. S. 1895. The cranial nerves of Amphibia. J. Morphol., vol. 10, p. 101.

Wallenberg, A. 1907. Die kaudale Endigung der bulbo-spinalen Wurzeln des Trigeminus, Vestibularis, und Vagus beim Frosche. Anat. Anz., Bd. 30, S. 564.

reptiles

Ariens ILappers, C. U. 1914. Der Geschmack, perifer und central. Zugleich einc Skizze der phylogenetischen Veranderungen in den sensibelen VII, IX und X Wurzeln. Psychiat. en neurol. Bl., Amsterdam. Bd. 18, S. 82.

Beccari, N. 1914. II IX, X, XI e XII pajo di nervi cranici e i nervi cervical! negli embrioni di Lacerta muralis. (Contribuzioni alio studio del significato morfologico dei nervi della testa.) Arch. ital. di anat. e di embriol., vol. 13, p, 1.

Bender, O. 1906. Ueber die Schleimhautnerven der VII, IX und X in Semons Forschungsreisen. G. Fischer, Jena.

Edinger, L. 1908. Vorlesungen fiber den Bau der nenmsen Centralorgane des Menschcn und der Thiere. 7*” Aufl., F. C. W. Vogel, Leipzig.

Gegenbaur, C. Vergleichende Anatomie der Wirbelthiere mit Berficksichtigung der Wirbellosen. W. Engelmann, Leipzig.

Brans

Ariens Kappers, C. U. 1914. Der Geschmack, perifer und central. Zugleich cine Skizze der phylogenetischen Veranderungen in den sensibelen VII, IX und X Wurzeln. Psychiat. en neurol. Bl., Amsterdam, Bd. 18, S. 82.

Bok, S. T. 1915. Die Entwicklung der Himner\'en und ihrer zentralen Bahnen. Die Stimulogene Fibrillation. Folia neuro-biol., Bd. 9, S. 475.

Brandis, F. 1894. Untersuchungen fiber das Gehirn der Vogel. II. Theil : Ursprung der Nerven der Medulla oblongata. Arch. f. mikr. Anat., Bd. 43, S. 96.

Mesdag, T. M. 1909. Bijdrage tot de ontwikkclingsgeschiedenis van de structuur der herscnen bij het kip-embryo. Dissertatie, Groningen.

TASTE AND THE GENERAL AFFERENT SYSTEMS 423

Ram6n y Cajal, S. 1909. Contribuci6n al estudio de los ganglios de la substancia reticular del bulbo con algunos detalles conciemientes i los focos motores y rias reflejas bulbares y mesocefalicas. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 7, p. 259.

MAMMALS

Allen, W. F. 1923. Origin and distribution of the tractus solitarius in the guinea pig. J. Comp. Neurol., vol. 35, p. 171.

. 1923a. Origin and destination of the secondary visceral fibers in the guinea pig.

J. Comp. Neurol., vol. 35, p. 275.

. 1927. Experimental anatomical studies on the visceral bulbospinal pathway

in the cat and guinea pig. J. Comp. Neurol., vol. 42, p. 393.

Ariens ICappers, C. U. 1914. Der Geschmack, perifer und central. Zugleich eine Skizze der phylogenetischen Veranderungen in den sensibelen VII, IX und X Wurzeln. Psychiat. en neurol. BL, Amsterdam, Bd. 18, S. 82.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

VON Bechterew, W. 1899. Die Leitungsbahnen im Gehim und Riickenmark. 2** Aufl., Leipzig.

Berkelbach van der Sprenkel, H. 1924. The hypoglossal nerve in an embryo of Erinaceus Europaeus. J. Comp. Neurol., vol. 36, p. 219.

Biondi, G. 1910. Sul decorso e sulle connessioni della porzione distale del fasciculus solitarius nell’uomo. Ricerche n. lab. di anat. norm. d. r. Univ. di Roma, vol. 15, p. 253.

Bremer, J. L. 1927. A textbook of histology arranged on an embryological basis. P. Blakiston’s Son and Co., Philadelphia.

Bruce, A. 1898. On the dorsal or so-called sensory nucleus of the glossopharyngeal nerve, and on the nuclei of origin of the trigeminal nerve. Brain, vol. 21, p. 383.

Brun, R. 1912. Ein Fall von doppelseitigen, symmetrischen Erweichungscysten im verlangerten Mark. Arb. a. d. himanat. Inst, in Zurich, Bd. 6, S. 273.

Cushing, H. 1903. The taste fibers and their independence of the nervus trigeminus. Bull. Johns Hopkins Hospital, vol. 14, p. 71.

Davis, L. E. 1923. The deep sensibility of the face. Arch. Neurol, and Psychiat., vol. 9, p. 283.

DuBois, F. S. 1929. The tractus solitarius and attendant nuclei in the Virginian opossum (Didelphis virginiana). J. Comp. Neurol., vol. 47, p. 189.

Fuse, G. 1913. Die Randgebiete des Pons und des Mittelhims. Arb. a. d. himanat Inst, in Zurich, Bd. 7, S. 211.

. 1913a. Beitrage zur Anatomie des Bodens des IV Ventrikels. Arb. a. d.

himanat. Inst, in Zurich, Bd. 8, S. 213.

VAN Gehuchten, a. 1900. Recherches sur la terminaison centrale des nerfs sensibles pdriph6riques. I. Le nerf intermSdiaire de Wrisberg. Le N^vraxe, vol. 1, p. 3.

. 1900. Idem. II. Le faisceau solitaire. Le N6vraxe, vol. 1, p. 171.

Herrick, C. J. 1931. An introduction to neurology. 5““ edition. W. B. Saunders Co., Philadelphia and London.

Hirose, K. 1907. Ueber eine bulbospinale Bahn. Folia neuro-biol., Bd. 10, S. 371.

Huber, G. Carl. 1896. Observations on the innervation of the sublingual and submaxillary glands. J. Exper. Med., vol. 1, p. 281.

. 1919. Textbook of histology. J. B. Saunders and Co., New York. (Printed in

1904, reprinted several times including 1919.)

Hunt, J. R. 1909. The sensory system of the facial nerve, and its symptomatology.

J. Nerv. and Ment. Dis., vol. 36, p. 321.

424 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Kohnstamm, 0., AND WoLFSTEiN, J. 1907. Versuch einer physiologischen Anatomie der Vagusurspriinge und des Kopfsympathicus. J. f. Psychol, u. Neurol., Bd. 8, S. 177.

VON Kolliker, a. 1889-02. Handbuch der Gewebelehre des Menschen. Bd. 2. W. Engelmann, Leipzig.

Kosaka, K., and Yagita, K. 1905. Experimentelle Untersuchungen iiber den Ursprung des N. vagus und die centralen Endigung der dem Plexus nodosus entstammenden sensiblen Vagusfasem, sowie iiber den Verlauf ihrer secondaren Bahn. Shinkeigaku Zasshi, Tokyo, Bd. 4, S. 29.

Lewandowsky, M. 1904. Untersuchungen iiber die Leitungsbahnen des Truncus cerebri und ihren Zusammenhang mit denen der Medulla spinalis und des Cortex cerebri. G. Fischer, Jena.

Luna, E. 1910. Zur Morphologie und zur Entwickelung des Nucleus intercalatus (Staderini). Folia neuro-biol., Bd. 4, S. 242.

VAN Melle, M. a. 1900. Some reflexes on the respiration in connection with Laborde’s method to restore, by rhythmical traction of the tongue, the respiration suppressed in narcosis. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 2, p. 640.

VON Monakow, C. 1885. Neue experimentelle Beitrage zur Anatomie der Schleife. Neurol. Centralbl., Bd. 4, S. 265.

. • 1913. Zur Kenntnis der Grosshimteile (Vago-glossopharyngeus-schleife. Fasc.

bulbo-thal.). Neurol. Centralbl., Bd. 32, S. 331.

Nageotte, j. 1906. Le Nerf interm^diare de Wrisberg et le Noyau de la Gustation chez I’Homme. Rev. Neurol, et Psychiat., T. 14, p. 1152.

Oppenheim, H. 1913. Lehrbuch der Nervenkrankheiten. S. Karger, Berlin.

Ram6n y Cajal, S. 1909-11. Histologic du systeme nerveux de I’homme et des vert6br4s. A. Maloine, Paris.

Ranson, S. W. 1931. The anatomy of the nervous system. 4“" edition. W. B. Saunders Co., Philadelphia and London.

Rasmussen, A. 1932. The principal nervous pathways. The Macmillan Co., New York.

Rothmann, M. 1902. Ueber die spinalen Atmungsbahnen. Arch. f. Anat. u. Physiol., Physiol. Abt., S. 11.

Sala, L. 1893. Ueber den Ursprung des Nervus acusticus. Arch. f. mikr. Anat., Bd. 42, S. 18.

VAN DER Sprenkel. (See Berkelbach van der Sprenkel.)

Staderini, R. 1894. Sopra un nucleo di cellule nervose intercalate fra i nuclei di origine del vago e dell’ipoglosso. Monitore zool. ital., vol. 5, p. 178.

. 1896. Ubicazione e rapporti di alcuni Nuclei di sostanza grigia della midolla

allungata. (Nucleo triangolare dell’Acustico, Nucleo terminale del Vago, Nucleo dellTpoglosso e Nucleus funiculi teretis.) Intemat. Monatschr. f. Anat. u. Physiol., Bd. 13, S. 326.

. 1898. Per una questione di priorita sul “Nucleo intercalate.” Anat. Anz.,

Bd. 14, S. 317.

. 1906. “Nucleo intercalate” e “Pars inferior fossae rhomboideae.” Anat. Anz.,

Bd. 29, S. 329.

. 1907. Sopra alcune particolarit^ anatomiche della midolla allungata. Anat.

Anz., Bd. 30, S. 316.

Streeter, G. L. 1908. The peripheral nervous system in the human embryo at the end of the first month. Am. J. Anat., vol. 8, p. 285.

Stuurman, F. j. 1913. Over den oorsprong van den Nervus Vagus bij het Konijn. Dissertatie, Amsterdam.

Thomas, A. 1899. Etude sur quelques faisceaux descendants de la moelle. Jphysiol, et de path, gen., vol. 1, p. 47.

taste and the general AFP^ERENT SYSTIiaiS 425=

TumbeIiAka., R. 1916. Konsekutive Veranderungen eines kleinen Herdes in dem ventrolatei'alen Thalamuskern und das deraselben zugrunde liegende klinische Bild. Psychiat. en neural. BL, Amsterdam, Bd. 20, S. 377.

Wallenbeeg, a. 1897. Das dorsale Gebiet der spioalen Trigeminuswurzel und seine Beziehungen zum solitaren Biindel beim Menschen- Deutsche Zeitschr. f. Nervenh., Bd. 11, S. 391.

VAN WayenbuPG, G. a. M. 1914. Discussie : Notulen der Wetenschappelijke Vergadering der Nederlandsche Vereeniging voor Psyckiatrie en Neurologie te Utrecht. Psychiat. en neurol. Bl., Amsterdam, Bd. 18, S. l91 Weigner, K. 1905. Ueber den Verlauf des Nervus iptermedius. Anat. Hefte, Bd. 29, S. 97.

Winkler, C. 1917. Anatomie du systeme nerveux. E. F. Bohn, Haarlem.

Certain Literature Concerned with the Origin of the Chorda Titupani and THE Discussion of the Supposed Relation of the Trigeminal to Taste

Amabilino, R. 1898. Sui rapporti del ganglio geniculato con la corda del timpano e col faciale. Pisani, Palermo, vol. 19, p. 123.

B'Ender, O. 'Vh'it'b. Yeber die ‘ScY\eim'na\ii,neT\e'n Ykk, I'X ■CiTid "X in ‘SeTntms Forschungsreisen. G. Fischer, Jena.

Bruns, L. 1889. Multiple Hirnnervenlasion nach Basisfractur. Ein Beitrag zur Frage des Verlaufs des Geschmacksnerven. Arch- 1- Psychiat., Bd. 20, S. 495.

Cushing, H. 1903. The taste fibres and their independence of the N. trigeminus. Deductions from thirteen cases of Gasserian ganglion extirpation. Bull. Johns Hopkins Hospital, vol. 14, p. 71.

. 1904. The sensory distribution of the fifth cranial nen^e. Bull. Johns Hopkins

Hospital, vol. 15, p. 213.

Dana, C. L. 1886. A case of paralysis of the trigeminus followed by alternate hemiplegia — its relations to the nen'C of taste. •!- Nerv. and Ment. Dis., vol. 13,

P-65. . . . . j . . .

. 1906. The question of protopathic and epicritic sensibility and the distribution

of the trigeminus nerve (third branch). J. Nerv. and Ment. Dis., vol. 33, p. 577.

Davies, H. M, 1907. The functions of the trigeminal nerve. Brain, vol. 30, p. 219.

Delprat, C. C. 1890. Centrale, dubbelzijdige faciahs parese. Nederl. tijdschr. v. geneesk., vol. 26, part 2, p. 697.

Dixon, A. F. 1896. On the development of the branches of the fifth cranial nerve in man. Sc. Tr. Roy. Dublin Soc., vol. 6, Ser. 2.

. 1899. The sensory distribution of the facial jierv'e in man. J. Anat., vol. 33,

P- 471.

DE Gaetani, L- 1906. Del nervo intermediario di Wnsberg e della corda del timpano. Le Ndvraxe, vol. 8, p. 67. ..... ^

Green, H. A. 1900. On the homologies of the chorda t junpani in selachians. J . Comp. Neurol., vol. 10, p. 411.

Hunt, J. R. 1915. The sensory field of the facial nerve : A further contribution to the sjTnptomatology of the geniculate ganglion. Brain, vol. 38, p. 418.

Kehrer, F. A. 1892. Ein Versuch bei einem neugcbornen Kinde ubcr den Sitz der Atmungscentrcn. Zeitschr. f. Biol., Bd. 28 (N. F._Bd. 10), S. 450.

Koster, G. 1900. Klinischer und e.xpcrimentcllcr Bcitmg zur Lchre von der Liihmung des NervUS facialis, zuglcich ein Beitnag zur Physiologic des Geschmackcs, der Schweiss-, Spcichel- und Thriincnabsonderung. Deutsches Arch. f. klin. Med., Bd. 68, S. 3«. .

ICrause, F. 1890. Die Neuralgic des Trigeminus, nebst der Anatomic und Phj'siologie dieses Nerven. F. C. W. Vogel, Leipzig.

426 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ltjssana. Sui nervi del gusto. Gaz. med. ital., Prov. Veneto Fisc. 42-46. (Quoted from Ariens Kappers ’20-’21.)

Ram6n y Cajal, S. 1899-1904. Textura del sistema nervioso del hombre y de los vertebrados. Vol. 2. N. Moya, Madrid.

Shekrington, C. S. 1898. Experiments in examination of the peripheral distribution of the fibres of the posterior roots of some spinal nerves. Part 11. Phil. Tr. Roy. Soc., London, Ser. B., vol. 190, p. 45.

Streeter, G. L. 1908. In the Manual of human embryology by Keibel, F., and Mall, F. P. J. B. Lippincott and Co., Philadelphia and London, 1912.

ZwAARDBMAKER, H. 1903. Geschmalc. Ergeb. d. Physiol. (Asher u. Spiro), Bd. 2^, S. 699.

Wallenberg, A. 1897. Das dorsale Gebeit der spinalen Trigeminuswurzel und seine Beziehungen zum solitaren Bundel beim Menschen. Deutsche Zeitschr. f. Nervenheilkunde, vol. 11, S. 391.

The Peripheral and Central Relations of the Trigeminal Nerve

CYCLOSTOMES

Ayers, H., and Worthington, J. 1907. The skin endorgans of the trigeminus and the lateralis nerves of Bdellostoma dombeyii. Am. J. Anat., vol. 7, p. 327.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny : a study of the phenomena of neurobiotaxis. Part I. Cyclostomi and Pisces. J. Comp. Neurol., vol. 27, p. 467.

Jansen, J. 1930. The brain of Myxine glutinosa. J. Comp. Neurol., vol. 49, p. 359.

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

Rothig, P. 1913. Beitrage zur Lehre der Neurobiotaxis. Die Verscheibung des motorischen Kerne in der Oblongata bei Myxine glutinosa und bei einigen Amphibien. Verslag. kon. Akad. v. Wetensch. te Amsterdam, Bd. 22, S. 315.

Rothig, P. , and Ariens Kappers, C. U. 1914. Further contributions to our knowledge of the brain of Myxine glutinosa. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 17*, p. 2.

Tretjakofp, D. 1909. Das Nervensystem von Ammocoetes. II. Gehirn. Arch, f. mikr. Anat., Bd. 74, S. 636.

VAN Valkenbuhg, C. T. 1911. Zur vergleichenden Anatomie des raesencephalen Trigeminusanteils. Folia neuro-biol., Bd. 5, S. 360.

Worthington, J. 1905. Contribution to our knowledge of the myxinoids. Am. Naturalist, vol. 39, no. 465, p. 607.

. 1906. The descriptive anatomy of the brain and cranial nerves of Bdellostoma

dombeyi. Quart. J. Micr. Sc., vol. 49, pt. 1, p. 137.

plagiostomes

Johnston, J. B. 1905. The radix mesencephalica trigemini. The ganglion isthmi. Anat. Anz., Bd. 27, S. 364.

. 1909. The radix mesencephalica trigemini. J. Comp. Neurol., vol. 19, p. 593.

VAN Valkenburg, C. T. 1911. Zur Kenntnis der Radix spinalis Nervi trigemini. Monatschr. f. Psychiat. u. Neurol., Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalon Trigeminusanteils.

Folia neuro-biol., Bd. 5, S. 360.

VAN Wijhe, J. W. 1882. Ueber die Mesodermsegmente und die Entwickelung der Nerven des Selachierkopfes. J. Miiller, Amsterdam. Verhandl. d. kon. Akad.

V. Wetensch. te Amsterdam, vol. 22, 1883.

TASTE AND THE GENERAL AFFERENT SYSTEMS

427

TEI/EOSTS

Ariens EAppers, C. U. 1906. The structure of the teleostean and selachian brain. J. Comp. Neurol., vol. 16, p. 1.

VAN Gehuchten, a. 1895. De I’origine du path4tique et de la racine sup4rieure du trijumeau. Bull, de I’Acad. roy. d. sc., d. let. et d. beaux-arts de Belgique, Ann4e 65, S4r. Ill, vol. 29, no. 3, p. 417.

Goronowitsch, N. 1888. Das Gehirn und die Cranialnerven von Acipenser ruthenus. Morphol. Jahrb., Bd. 13, S. 515.

Herrick, C. J. 1906. On the centers of taste and touch in the medulla oblongata of fishes. J. Comp. Neurol., vol. 16, p. 403.

Johnston, J. B. 1905. The radix mesencephalica trigemini. The ganglion isthmi. Anat. Anz., Bd. 27, S. 364.

. 1909. The radix mesencephalica trigemini. J. Comp. Neurol., vol. 19, p. 593.

Matser, P. 1881. Vergleichend-anatomische Studien iiber das Gehirn der Knochenfische mit besonderer Berucksichtigung der Cyprinoiden. Zeitschr. f. wissensch. Zool., Bd. 36, S. 259.

Tello, F. 1909. Contribucidn al conocimiento del encdfalo de los Teledsteos. Los niicleos bulbares. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 7, p. 1.

VAN Valkenburg, C. T. 1911. Zur Kenntnis der Radix spinalis nervi trigemini. Monatschr. f. Psychiat. u. Neurol., Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalen Trigeminusanteils.

Folia neuro-biol., Bd. 5, S. 360.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehirns der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

Weinberg, E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol., vol. 46, p. 249.

AMPHIBIANS

Bindewald, C. 1911. Fine Commissure intertrigemina in Amphibiengehirn. Anat. Adz., Bd. 40, S. 243.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny. A study of the phenomena of neurobiotaxis. J. Comp. Neurol., vol. 28, p. 379.

Gaupp, E. 1904. Anatomie des Frosches by Ecker, A. Wiedersheim, R. and Gaupp, E. Teil 2. 2“® Aufi., F. Vieweg and Sohn, Braunschweig.

Herrick, C. J. 1914. The medulla oblongata of larval Amblystoma. J. Comp. Neurol., vol. 24, p. 343.

. 1914a. The cerebellum of Necturus and other urodele Amphibia. J. Comp.

Neurol., vol. 24, p. 1.

. 1917. The midbrain and thalamus of Necturus. J. Comp. Neurol., vol. 28,

p. 251.

. 1930. The medulla oblongata of Necturus. J. Comp. Neurol., vol. 50, p. 1.

Johnston, J. B. 1905. The radix mesencephalica trigemini. The ganglion isthmi. Anat. Anz., Bd. 27, S. 364.

. 1909. The radix mesencephalica trigemini. J. Comp. Neurol., vol. 19, p. 593.

Larsell, 0. 1923. The cerebellum of the frog. J. Comp. Neurol., vol. 36, p. 104. McKibben, P. S. 1913. The eye muscle nerves in Necturus. J. Comp. Neurol., vol. 23, p. 153.

Norris, H. W. 1913. The cranial nerves of Siren lacertina. J. Morphol., vol. 24, p. 245.

Ram6n, P. 1904. Origen del nervio masticador en las aves, reptiles y batracios.

Trab. d. lab. de invest, biol. Univ. de Madrid., vol. 3, p. 153.

Ram6n y Cajal, S. 1909. See bibliography for mammals.

428 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Rothig, P. 1927. Beitrage zum Studium des Zentralnervensystems der Wirbeltiere, no. 11. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 10, S. 381.

Strong, 0. S. 1913. The cranial nerves of Amphibia. J. Morphol., vol. 10,

p. 101.

VAN Valkenburg, C. T. 1911. Zur Kenntnis der Radix spinalis Nervi trigemini. Monatschr. f. Psychiat. u. Neurol., Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalen Trigeminusanteils.

Folia neuro-biol., Bd. 5, S. 360.

Wallenberg, A. 1907. Die kaudale Endigung der bulbo-spinalen Wurzeln des Trigeminus, Vestibularis und Vagus beim Frosche. Anat. Anz., Bd. 30, S. 564. Weinberg, E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol., vol. 46, p. 249.

REPTILES

Beccari, N. 1912. La costituzione, i nuclei terminali, e le vie di connessione del nervo acustico nella Lacerta muralis. Men. Arch. ital. di anat. e di embriol., vol. 10, p. 646.

Huber, G. Carl, and Crosby, E. C. 1926. On thalamic and tectal nuclei and fiber paths in the brain of the American alligator. J. Comp. Neurol., vol. 40, p. 97. Ingvar, S. 1918. Zur Phylo- und Ontogenese des Kleinhirns. Folia neuro-biol., Bd. 11, S. 171.

Johnston, J. B. 1909. The radix mesencephalica trigemini. J. Comp. Neurol, vol. 19, p. 593.

de Lange, S. J. 1913. Das Zwischenhim und das Mittelhim der Reptilien. Folia neuro-biol., Bd. 7, S. 67.

. 1917. Das Hinterhim, das Nachhim und das Ruckenmark der Reptilien. Folia

neuro-biol, Bd. 10, S. 385.

Larsell, 0. 1933. The cerebellum of reptiles : chelonians and alligator. J. Comp.

Neurol, vol. 56, p. 299.

Ramon, P. 1904. Origen del nervio masticador en las aves, reptiles y batracios.

Trab. d. lab. de invest, biol. Univ. de Madrid, vol 3, p. 153.

VAN Valkenburg, C. T. 1911. Zur Kenntnis der Radix spinalis Ner\d Trigemini. Monatschr. f. Psychiat. u. Neurol, Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalen Trigeminusanteils.

Folia neuro-biol, Bd. 5, S. 360.

Weinberg, E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol, vol. 46, p. 249.

Weston, J. K. 1933. The Reptilian Vestibular and Cerebellar Gray with Fiber Connections. Dissertation.

Bmns

Biondi, G. 1913. I nuclei d’origine e terminali del nervo trigemino nel polio. Rev. ital. di neuropat., psich. ed elettroterapia, vol. 6.

Bok, S. T. 1915. Die Entwicklung der Himnerven und ihrer zentralen Bahnen. Die Stimulogene Fibrillation. Folia neuro-biol, Bd. 9 , S. 475.

Brandis, F. 1895. Untersuchungen fiber das Gehini der Vogel. III. Der Ursprung des N. Trigeminus und der Augenmuskelnerven. Arch. f. mikr. Anat., Bd. 44, S. 534.

Craigie, E. H. 1928. Observations on the brain of the hummingbird (Chrysolampis mosquitus Linn, and Chlorostilbon caribaeus Lawr.). J. Comp. Neurol, vol 4o, p. 377.

TASTE AND THE GENERAL AFFERENT SYSTEMS

429

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7*® Aufl., F. C. W. Vogel, Leipzig.

Huber, G., and Crosby, E. C. 1929. The nuclei and fiber paths of the avian diencephalon, with consideration of telencephalic and certain mesencephalic centers and connections. J. Comp. Neurol., vol. 48, p. 1.

Kosaka, K. 1912. Zur Frage der physiologischen Natur der zerebralen Trigeminuswurzel. Folia neuro-biol., Bd. 6, S. 1.

Ram6n, P. 1904. Origen del nervio masticador en las aves, reptiles y batracios. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 3, p. 153.

Sanders, E. B. 1929. A consideration of certain bulbar, midbrain, and cerebellar centers and fiber tracts in birds. J. Comp. Neurol., vol. 49, p. 155.

ScHROEDER, K. 1911. Dcr Faserverlauf im Vorderhirn des Huhnes. J. f. Psychol, u. Neurol., Bd. 18, S. 115.

VAN Valkenbdrg, C. T. 1911. Zur Kenntnis der Radix spinalis Nervi trigemini. Monatschr. f. Psychiat. u. Neurol., Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalen Trigeminusanteils.

Folia neuro-biol., Bd. 5, S. 360.

Wallenberg, A. 1898. Die secundare Acusticusbahn der Taube. Anat. Anz., Bd. 14, S. 353.

. 1903. Der Ursprung des Tractus isthmo-striatus (oder bulbo-striatus) der Taube.

Neurol. Centralbl., Bd. 22, S. 98.

. 1904. Neue Untersuchungen fiber den Himstamm der Taube. II. Die cere brale Trigeminuswurzel. Anat. Anz., Bd. 25, S. 526.

. 1904a. Nachtrag zu meinen Artikel fiber die cerebrale Trigeminuswurzel der

Vogel. Anat. Anz., Bd. 25, S. 621.

Weinberg, E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol., vol. 46, p. 249.

MAMMALS

Allen, W. F. 1919. The application of the Marchi method to the study of the radix mesencephalica trigemini in the guinea pig. J. Comp. Neurol., vol. 30, p. 169.

. 1925. Identification of the cells and fibers concerned in the innervation of the

teeth, J. Comp, Neurol., vol. 39, p. 325.

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Bochenek, V. A. 1901. La racine bulbo-spinale du trijumeau et ses connexions avec les trois branches p6riph6riques. Le N6vraxe, vol. 3, p. 107.

Bolk, L. 1898-1900. Die segmental differenzirung des menschlichen Rumpfes und seiner Extremitaten. Morphol. Jahrb., Bd. 25,. S. 465; Bd. 26, S. 91; Bd. 27, S. 630 and Bd. 28, S. 105.

Bregmann, E. 1892. Ueber experimentelle aufsteigende Degeneration motorischer und sensibler Himnerven. Jahrb. f. Psychiat. u. Neurol., Bd. 11, S. 73.

Breuer, R., AND Marburg, 0. 1902. Zur Klinik und Pathologie der apoplectiformen

Bulbarparalyse. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 9, S. 181.

Brouwer, B. 1915. Die biologische Bedeutung der Dermatomerie. Beitrag zur Kenntnis der Segmentalanatomie und der Sensibilitatsleitung im Rfickenmark und in der Medulla oblongata. Folia neuro-biol., Bd. 9, S. 225.

Bruce, A. 1898. On the dorsal or so-called sensory nucleus of the glossopharyngeal nerve, and on the nuclei of origin of the trigeminal nerve. Brain, vol. 21, p. 383.

Castaldi, L. 1926. Studi sulla struttura e sullo sviluppo del mesencefalo. Ricerche in Cavia cobaya. Arch. ital. di anat. e di embriol., vol. 23, p. 481.

430 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Clark, S. 1926. Nissl granules of primaiy afferent neurons. J. Comp. Neurol., vol. 41, p. 423.

Cramer, A. 1894. Beitrag zur feineren Anatomie der Medulla oblongata und der Briicke mit besonderer Beriichsichtigung des dritten bis zwolften. Renewed in Ergeb. d. Anat. u. Entwicklungsgesch. (Merkel u. Bonnet), Bd. 4, S. 254.

Cushing, H. 1903. The taste fibres and their independence of the N. trigeminus. Bull. Johns Hopkins Hospital, vol. 14, p. 71.

. 1904. The sensory distribution of the fifth cranial nerve. Bull. Johns Hopkins

Hospital, vol. 15, p. 213.

Davis, L. E. 1923. The deep sensibility of the face. Arch. Neurol, and Psychiat., vol. 9, p. 283.

Deiters, 0. 1865. Untersuchungen iiber das Gehim und Riickenmark der Saugetiere.

Braunschweig. S. 91.

Donaldson, H. H. 1912. Review of Willems; “Localisation motrice et kineslhesique.” J. Nerv. and Ment. Dis., vol. 39, p. 67.

Freeman, W. 1927. The columnar arrangement of the primary centers in the brainstem of man. J. Ne^. and Ment. Dis., vol. 65, p. 379.

VAN Gehuchten, A. 1895. De I’origine du pathdtique et de la racine supdrieure du trijumeau. Bull, de I’Acad. roy. d. sc., d. let. et d. beaux-arts de Belgique, Anndc 65, Ser. Ill, vol. 29, no. 3, p. 417.

Gerard, M. W. 1923. Afferent impulses of the trigeminal nerve. Arch. Neurol, and Psychiat., vol. 9, p. 306.

Giglio-Tos, E. 1902. Sull’origine embrionale del nervo trigemino nell’uomo. Anat. Anz., Bd. 21, S. 85.

Held, H. 1892. Die Endigungweise der sensiblen Nerven im Gehirn. Arch. f. Anat. u. Physiol., Anat. Abt., Heft I u. II, S. 33.

. 1893. Beitrag zur feineren Anatomie des Kleinhims und des Himstammes.

Arch. f. Anat. u. Phy.siol. Anat. Abt., Heft 4, S. 435.

Herrick, C. J. 1931. An introduction to neurology. 5‘’* edition. W. B. Saunders Co., Philadelphia and London.

Hosel, 0. 1893. Ein weiterer Beitrag zur Lehre vom Vcrlauf der Rindenschlcife

und centraler Trigeminusfasem beim Menschen. Arch. f. Psychiat., Bd. 25, S. 1.

Huber, G. Carl. 1898. The innervation of the tooth pulp. Dental Co.smos, vol. 40, p. 797.

. 1899. A note on sensory nen^e endings in the extrinsic eye muscles of the rabbit.

Atypical motor endings of Retzius. Anat. Anz., Bd. 15, S. 335.

. 1900. Sensory nerve terminations in the tendons of the extrinsic eye-muscles

of the cat. J. Comp. Neurol., vol. 10, p. 152.

Huguenin, G. 1879. Anatomic des centres nerveux. Paris. (Quoted from Tertcrjanz ’99.)

Hulles, E. 1907. Zur vergleichenden Anatomic der ccrebralcn Trigcminu.swurzcl. Arb. a. d. neurol. Inst. a. d. Wien. Univ*. (Obcrstciner’s), Bd. 16, S. 469.

Jacorsohn, L. 1908. Ueber die Kerne des menschlichen Riickenmarks. Abhandl. d. kon. Preuss. Akad. d. Wis.?cnsch., Berlin, S. 389.

Johnston, J. B. 1909. The radix me.sencephalica trigemini. J. Comp. Neurol., vol. 19, p. 593.

Koh.n'.stam.m, 0. 1910. Studien zur phy.siologischcn Anatomie des Himstammes. III.

J. f. P.sychol. u. Neurol., Bd. 17, S. 33.

Kohn.stamm, O., and Quensel, F. 1908. Studien zur phy.siologischcn Anatomie des Himstammes. I. J. f. P.sychol. u. Neurol., Bd. 13, S. 89.

VON Kulliker, a. 1889-02. Handbuch dor Gcwebclehre des Menschen. 6“ Aufl., Bd. 2. W. Engelmann, Leipzig.

TASTE AND THE GENERAL AFFERENT SYSTEMS 431

VON KQllikee, a. 1901. Die Medulla oblongata und die Vierhiigelgegend von Omithorhynchus und Echidna. W. Engelmann, Leipzig.

Kosaka, K. 1912. Zur Frage der physiologischen Natur der zerebralen Trigeminuswurzel. Folia neuro-biol., Bd. 6, S. 1.

Kue^, S. 1899. Die normale und pathologische Struktur der Zellen an der cerebralen Wurzel des Nervus trigeminus. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 6, S. 159.

Lewandowsky, M. 1904. Untersuchungen fiber die Leitungsbahnen des Truncus cerebri und ihren Zusammenhang mit denen der Medulla spinalis und des Cortex cerebri. G. Fischer, Jena.

VAN Londen, D. M. 1907. Untersuchungen bei einigen Saugetieren und beim Menschen. Den zentralen Verlauf des “Nervus trigeminus" nach intracranialer Durchneidung seines Stammes. Petrus Camper. Nederl. bijdr. t. de anat., vol. 4, p. 285.

Mabbueg, O. 1911. Ueber die neueren Fortschritte in der topischen Diagnostik der Pons und der Oblongata. Deutsche Zeitschr. f. Nen'enh., Bd. 41, S. 41.

May, 0., AND Hoesley, V. 1910. The mesencephalic root of the fifth nerve. Brain, vol. 33, p. 175.

Mendel, E. 1888. Zur Lehre von der Hemiatrophia facialis. Neurol. Centralbl., Bd. 14, S. 401.

Meekel, F. 1874. Die trophische Wurzel des Trigeminus. Enters, a. d. anat. Inst, zu Rostock. Reviewed in Jahresb. fi. d. Fortschr. d. ges. Med., Bd. 1, S. 68.

Meyeesohn, F. 1913. Die Dicke der spinalen Trigeminuswurzel bei verschiedenen Siiugem. Folia neuro-biol., Bd. 7, S. 217.

Meyneet, T. 1872. The brain of mammals. In Strieker’s manual of histology. American translation, pp. 704 and 708. W. Wood and Co., New York.

VON Monakow, C. 1895. Experimentelle und pathologisch-anatomische Untersuchungen fiber die Haubenregion, den SehhUgels und des Regio Subthalamica, nebst Beitragen zur Kenntniss fruh erworbener Gross- xmd Kleinhirndefecte. Arch. f. Psychiat., Bd. 27, S. 1.

. 1910. Der rote Kem, die Haube und die Regio hypothalamica betreffend.

Teil I. Arb. a. d. himanat. Inst, in Zfirich, Bd. 5, S. 103.

Nicholson, H. 1924. On the presence of ganglion cells in the third and sixth nerves in man. J. Comp. Neurol., vol. 37, p. 31.

Obeesteinee, H. 1912. Anleitung beim Studium des Baues der nervosen Zentralorgane im gesunden und kranken Zustande. 5*® Aufl., F. Deuticke, Leipzig u. Wien. See also editions of 1888 and 1901.

Oppenheim, H. 1913. Lehrbuch der Nervenkrankheiten. S. Karger, Berlin.

Peobst, M. 1899. Ueber vom Vierhfigel von der Brficke und von Kleinhirn absteigende Bahnen. Deutsche Zeitschr. f. Nervenh., Bd. 15, S. 192.

Ram6n y Cajal, S. 1896. Beitrag zum Studien der Medulla oblongata, des Kleinhims und des Ursprungs des Gehimnerven. J. A. Barth, Leipzig.

. 1909-11. Histologie du systime nerveux de I’homme et des vert4br4s.

A. Maloine, Paris.

ScHNEiDEE, A. J. 1928. The histology of the radix mesencephalica n. trigemini in the dog. Anat. Rec., vol. 38, p. 321.

Schwalbe, G. 1881. Lehrbuch der Neurologie, S. 678. E. Besold, Erlangen.

Sheinin, j. j. 1930. Typing of the cells of the mesencephalic nucleus of the trigeminal nerve in the dog, based on Nissl granule arrangement. J. Comp. Neurol., vol. 51, p. 109.

Spillee, W. G. 1915. Remarks on the central representation of sensation. J. Nerv. and Ment. Dis., vol. 42, p. 399.

Stopfoed, j. S. B. 1925. The function of the spinal nucleus of the trigeminal nerve.

J. Anat., vol. 59, p. 120.

432 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Terterjanz, M. 1899. Die obere Trigeminuswurzel. Arch. f. mikr. Anat., Bd. 53, S. 632.

Tozer, F. M. 1912. On the presence of ganglion cells in the roots of III, IV, and VI cranial nerves. J. Physiol., vol. 45, p. 15.

Tozer, F. M., and Sherrington, C. S. 1910. Receptors and afferents of the third, fourth, and sixth cranial nerves. Proc. Roy. Soc., London, Ser. B., vol. 82, p. 450. VAN Valkenburg, C. T. 1911. Zur Kenntnis der Radix spinalis Nervi trigemini. Monatschr. f. Psychiat. u. Neurol., Bd. 29, S. 407.

. 1911a. Zur vergleichenden Anatomie des mesencephalon Trigeminusanteils.

Folia neuro-biol., Bd. 5, S. 360.

Wallenberg, A. 1896. Zur Physiologie der spinalen Trigeminuswurzel. Neurol. Centralbl., Bd. 15, S. 873.

. 1904. Neue Untersuchungen iiber den Hirnstamm der Taube. III. Die cere brale Trigeminuswurzel. Anat. Anz., Bd. 25, S. 526.

. 1904a. Nachtrag zu meinen Artikel iiber die cerebrale Trigeminuswurzel der

Vogel. Anat. Anz., Bd. 25, S. 621.

. 1905. Sekundare Bahnen aus dem frontalen sensibeln Trigeminuskerne des

Kaninchens. Anat. Anz., Bd. 26, S. 145.

Weinberg, E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol., vol. 46, p. 249.

Willems, E. 1911. Localisation motrice et kinesth4sique. Les noyaux masticateur et m4senc6phalique du trijumeau chez le lapin. Le Ndvraxe, vol. 12, p. 1.

WiNDLE, W. F. 1926. Non-bifurcating nerve fibers of the trigeminal nerve. J. Comp. Neurol., vol. 40, p. 229.

. 1927. Experimental proof of the types of neurons that innervate the tooth

pulp. J. Comp. Neurol., vol. 43, p. 347.

Winkler, C. 1914. A case of occlusion of the arteria cerebella posterior inferior.

Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 17, pt. 2, p. 914.

. 1918. Anatomie du syst^me nerveux. Vol. 1. E. F. Bohn, Haarlem.

CHAPTER rv

THE LATERAI^LINE AND ACOUSTIC SYSTEMS

As compared with those in Amphioxus, the branchial nerves in craniotes, with the exception of the trigeminal nerve, show an hypertrophy of the visceral afferent and visceral efferent components and an atrophy of the somatic afferent components supplying general sensibility to ectodermal surfaces. However, new systems of afferent fibers appear in craniotes in close association with these branchial nerves. These are the lateral-line components associated with the facial and the

Fia. 180. The ectodermal areas of contact in a torpedo embryo of twelve millimeters length. The lateral-line placodes are cross lined.

The epibranchial placodes are black. The nerves are lined and the branchial clefts are numbered (after Froriep from Edinger’s Vorlesungen).

glossopharyngeal or vagus nerves in lower vertebrates through water amphibians, the vestibular, present with few exceptions in all craniotes, and the cochlear components of higher forms.

American neurologists, in particular, have pointed out that fibers innervating the static organs belong to the category of somatic afferent nerves, since this type of sense organ is derived from the ectoderm and not, as was originally supposed, from the entoderm. In view of their relations, it is not surprising that such nerves should arise in that region of the medulla oblongata in which ordinary somatic afferent fibers are greatly reduced, that is, in the region of emergence of the facial and glossopharyngeal nerves. The acoustic nerve emerges immediately behind the facial nerve and practically at the same level (figs. 184 and 188). The lateral-line nerves have their entrance at the level of emergence of the facial and of the glossopharyngeal or the vagus nerves. They are termed the nervus lateralis anterior (nervus lateralis facialis) and the nervus

433

434 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

lateralis posterior (nervus lateralis glossopharyngei or vagi). These static nerves differ from the ordinary cutaneous nerves in their embryonic origin.

The latter nerves arise from the neural crests. The former are only in part derived from that region ; in part they arise from thickenings of the lateral ectoderm termed dorsolateral placodes (fig. 180). In 1891, von Kupffer, working on Petromyzon, distinguished between the dorsolateral placodes concerned with the lateral line, and the epibranchial placodes concerned with the visceral sensory components of the branchial nerves. Miss Platt (’96) first made clear the distinction between the primordia of the lateral-line organs and the ganglia of the lateral-line nerves, and demonstrated clearly in Necturus the origin of the latter from the dorsolateral placodes. In Ameiurus Landacre (’10) found a marked thickening of the ectoderm on either side of the neural plate, to which he gave the name of the lateral mass. From this lateral mass he derived the dorsolateral placodes represented in the preauditory placode, the auditory vesicle, and the postauditory placode, as well as in the primordia of other ganglia, in the cells homologous to the neural crest of most other animals, and in the mesectoderm. From the lateral mass Landacre derived the lateral-line ganglia of the facial nerve, from the auditory vesicle (and possibly other cells of the lateral mass) the ganglion of the acoustic nerve and the lateral-line component of the glossopharyngeal nerve, and from the postauditory placode the lateral-line ganglion cells of the vagus nerve. Later Landacre and Conger (’13) studied the development of these nerves in Lepidosteus. Coghill (’16) and Stone (’22) studied the development of the cranial ganglia in Amblystoma, including the ganglia of the lateral-line nerves. Slone laid particular emphasis upon the study of the lateral-line sense organs, also. He found that the lateral-line ganglia are derived entirely from placodes and that the preauditory placode, which gives rise to the major part of the lateral-line ganglion of the facial nerve, also gives rise, in Amblystoma, to the anlagen for the sense organs of the supraorbital canal. The removal, at the proper stages, of the ectoderm in front of the auditory vesicle is followed by the disappearance of the greater part of the lateral-line ganglion of the facial and of the supraorbital sense organs. Stone’s results indicated that the supraorbital, infraorbital, and hyomandibular sense organs, as well as the ventral hyomandibular and mandibular sense organs, arise from separate anlagen. In 1926, Landacre considered the presence in Amblystoma jeffersonianum of certain primitive ectodermal thickenings or primitive lines which had been described in Necturus by Platt (’94). From the acoustico-lateral thickening he derived the preauditory and postauditory placodes and the auditory vesicle “which are intimately concerned with the origin of the lateralis ganglia of the VII, IX, and X nerves and also with the origin of the migratory lateral-line placodes of the head and body.’’ Landacre pointed out that even those lines which disappear in development mark the position of later developing lateral-line sense organs. Stone (’28) found both the permanent and transient ectodermal thickenings in Amblystoma punctatum to agree in position with those described in the form studied by Landacre. He carried out a series of transplantation experiments, and his results indicated that the transient thickenings are unessential to the

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

435

development of the organs, since the presence of transplanted, stained ectoderm in the region is followed by a normal migration of the lateral-line primordia and the development of a normal colored organ, and that the thickenings appear in indifferent ectoderm transplanted over axial mesoderm before the differentiation of the somites. Stone found that if, in an embryo of the proper age, a transplantation of the postauditory placode is made to a position on the body ventral to that in which the midbody lateral-line will develop, the transplant wiU develop caudally, forming a lateral-hne anlage parallel to the development of the midbody lateral hne and in a position not normally containing such an anlage. If a placode of this type is rotated, it still lays down a lateral-hne anlage caudally without reference to transient thickenings of the ectoderm. Stone suggested “that these transitory thickenings are expressions of the temporary contour of underlying structures” (p. 189), and that the ectodermal thickenings in the head region are due to its function as a source of migrating ectodermal cells to various regions. Rotation of 180° and then transplantation of the postauditory placode above the eye in Amblystoma (Stone, ’29) are followed by the development of a ganglion of the lateral-hne system which supphes fibers to lateral-hne sense organs developing from the transplant. These fibers join those of the lateral-hne nerve to the supraorbital canal. The lateral-line primordium for the supraorbital canal may develop side by side with the primordium from the transplant. However, the primordium which normaUy produces the supraorbital canal, under these operative conditions usuaUy is prevented from extending farther, once it has reached the transplanted tissue, because of the development of the ganghon. However, when it has reached the transplanted lateral-hne primordiiun, this latter grows forward to produce the cephahc end of a normally appearing supraorbital canal.

Although the functions of the lateral-hne organs and the labyrinth are not identical, nevertheless their central connections show intimate interrelation. Physiologically they have important characteristics in common. The lateralhne nerves, as they occiu' in fishes and water amphibians, are concerned in maintaining the position of the body with reference to its surroundings. They are important in the ordinary reflexes concerned with the protection and welfare of the animal. They represent a degree of differentiation comparable to protopathic or \utal sensibility in cutaneous nerves. The vestibular and, much more particularly, the associated cochlear are concerned in discriminative types of response, which consequently are comparable to the epicritic sensibihty of cutaneous nerves. The physiological and psychological differences come to expression in the central connections. Thus only the cochlear centers — those concerned with discriminatory reactions — have highly developed thalamic, and hence cortical, connections. Before dealing with the central connections of these nerves, a brief description of certain general characteristics of the peripheral end-organs seems advisable. The special differentiations of these end-organs, and particularly those of the acoustic nerve, will be considered under the different classes, since they show such differences in differentiation in the various forms.

436 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

In all three of these types of end-organs, the specialized epithelium shows certain similarities in structure. It consists of cone-shaped cells, shorter than the supporting cells, and bearing on their free surfaces long, fine hairs, giving them the name of hair cells. Such neuroepithelial or sense cells differ from the neuroepithelial cells of the gustatory organs in being relatively much .shorter and in having long, thin hairs, easily put in motion by the fluid or covering membrane, rather than the shorter, stiffer hairs characteristic of gustatory organs. The fluid which sets in motion the neuroepithelial hairs of the static organs may be either the water which surrounds the body and enters through appropriate openings, or may be a fluid inclosed within the cavity of the sense organ itself. The

supporting or sustentacular cells belong to several types, these types varying with the particular organ under consideration (figs. 181, 182, 183).

Both the lateral-line organs and the labyrinth, at some time during their life history, communicate directly with the surface. The lateral-line canals usually retain this open communication, but it is absent in the vesicles of Savi (’44) and in the canals of some of the suBtentacuiar ceUa higher fishes. The lab 5 Tinth of the adult

Fig. 181. Lateral-line organ in a Catostomus animals maintains a direct connection larva. Johnston. ,, ^ i • i

With the surrounding water only in selachians. This connection is the ductus endolymphaticus. In higher vertebrates the labyrinth is cut off from the surface during development and the hairs of the neuroepithelial cells are set in vibration by means of an inclosed fluid. All of these organs are concerned with the appreciation of vibrations. In Amphioxus, special peripheral sense organs for perception of movement of the surrounding medium do not exist. The slight sensitivity to vibration which is shown by this animal is probably due to the appreciation of rhythmic alterations in pressure on the part of the ordinary tactile endings, in the tentacles of the velum (Dogiel, ’03; Parker, '08; Kutchin, ’13). It is only within craniotes that special endorgans for the perception of vibrations occur. The least complicated apparatus for the appreciation of this type of stimuli is to be found in the organs of the lateral-line system. These vibrations are of less frequency than those appreciable to the human cochlea, being as low as 6 vibrations per second. According to Parker (’04, ’10), Hofer (’07), and certain other observers, these organs are for the perception of periodic changes in pressure caused by the movements of the animal and by the reflection of the water from firm objects. Thus a fish can find its way in the dark without running against solid obstacles.

The lateral-line canals of the head and body are conspicuous structures in certain fishes. Usually four such canals are present on a side. The lateral canal, which gives to the entire apparatus its name, extends in a straight line along the side of the body. The other three canals (fig. 186) are found on the head, and

Hair celle

THE LATERALr-LINE AND ACOUSTIC SYSTEMS

437

are so arranged that the supraorbital canal (C. S. 0.) runs over or around the eye, the infraorbital (C. I. O.) below it, while the hyomandibular canal (C. H. M.) extends over the lower jaw.

That these canals contain sense organs was discovered by Leydig (’50 and ’68) in fishes, was confirmed by Schultze (’61, ’70) for fishes and amphibians, and has been reconfirmed by many observers. The simplest form of the lateralline organ is foimd in cyclostomes, as, for instance, in Petromyzon, and in the

Fig 182 Neuroepithelial and sustentacular cells in crista acustica of Proteus anguineus Retzms

amphibians. In these forms the lateral-line sense organs lie in grooves rather than canals, with the free end of the organ usually about level with the surface epidermis or slightly below it. In plagiostomes (as Mustelus canis and Squalus acanthias, Sydney Johnson, ’17) the lateral-line sense oigans are either in canals or present as pit organs. The development of the canals from open grooves during the embryonic development is discussed briefly on page 436. In certain fishes the canals are closed almost completely, so that the direct perception of vibrations of the water is not possible. However, the covering is so thin that a transmission of vibrations of the water to the fluid inside of the canals occurs easily. Two specialized t 3 npes of terminations, the ampullae of Lorenzini (’78) and the vesicles of Savi (’44), lie considerably deeper than the lateral-line canals. The first of these specialized organs is found in plagiostomes (fig. 186). On the heads of these animals there are openings, from each of which an epithelial tube extends far below the epidermis and terminates in an ampullar enlargement

438 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

having small lateral indentations. These organs are supplied by branches of the anterior lateral-line nerves. Within each ampulla is a structure similar to the organs of the lateral-line canal. This is covered by a gelatinous substance that extends as far as the opening (hence the German name Gallertrohre or “jellytubes”)- The vesicles of Savi (’44) are entirely separate from the surface. They are found only in the torpedo. In this animal they lie near the electric organs. They have large cells provided with sensory hairs. The vesicles are filled with a fluid, the oscillations of which stimulate the sense cells. Thus in the various animals there is a transition between an open and a closed system of organs for perceiving vibrations. This is of particular interest since another organ specialized for the perception of vibrations, the vestibular apparatus.

Tectorial membrane

at first communicates with the surrounding medium by means of the endolymphatic duct. Its separation from the surface occurs only above elasmobranchs.

In ganoids, the water reaches the lateral-line organs through openings between or in the scales. The largest lateral-line organs are found in the teleosts, where the skeleton takes part in their formation (Gegenbauer, ’98). A considerable development of this apparatus is found on the heads of some of the deep-sea fishes (Macruridae) and of Mormyrus. Lateral-line sense organs are still present in aquatic amphibians, particularly in Perennibranchiata and Caducibranchiata. As was stated earlier, they are no longer inclosed in canals, but show a primitive form comparable to that in cyclostomes. In land-living urodeles they disappear from the surface during dry weather, but reappear on the surface the following spring. The organs are found in the larvae of Anura, but disappear during metamorphosis. In these forms, their former position is marked by a deficiency of pigmentation. Such areas are known as “touch-spots, ” but their significance in tactile innervation is yet to be proved {Gegenbauer).

The nerves innervating the lateral-line organs, the anterior and posterior lateral-line nerves, usually emerge from the medulla oblongata in a dorsal and a ventral group. The former nerve, generally described under the name of the nervus lateralis facialis, or rarely as the nervus lateralis trigemini, enters the medulla oblongata at the level of entrance of the facial. Peripherally it is dis

THE LATERAL-LINE AND ACOUSTIC SYSTEMS 439

tributed with the sensory branches of the trigeminal. It reaches the lateralhne organs of the head. The nervus lateralis posterior, called also the nervus lateralis vagi or glossopharyngei, enters the medulla oblongata at the level of entrance of the vagus or glossopharyngeus. Its peripheral distribution is to the lateral line of the body.

The Acoustico-lateral Systems of Cyclostomes The lateral-line sense organs of Petromyzon are less well developed than those of other fishes, but resemble those of certain amphibians. They occur as pit organs and not as canal organs in closed lateral-line canals. The myxinoids have very simple lateral-fine organs also. Ayers and Worthington (’07) were the first observers to describe these organs in Bdellostoma. They found that in this

Fig 184 The arrangement of the dorsal, medial, and ventral nuclei in the static or acoustico-lateral area of Petromyzon marinus Huber and Crosby

cyclostome there are canals which are exceedingly small and are only faintly indicated on the surface. Such canals are arranged in an anterior and a posterior group, corresponding to the anterior and posterior lateral-line nerves, through which they are innervated. The anterior group consists of about four canals situated on the side of the head in front of the eye ; the posterior group, of an inner series of canals running mesolaterally and an outer series running at a slight angle to the midaxis of the body. The canals remain throughout life in the epidermis of Bdellostoma and the sense organs are poorly developed.

In Petromyzon, both the lateral-line nerves and the acoustic (essentially vestibular) nerves are well developed. Both end in the area statica (tuberculum acusticum or area acustico-lateralis). This area in Petromyzon is characterized by the possession of cells similar to those found in the dorsal horn of the cord and in the nucleus of the descending root of the trigeminal (Johnston, ’02). These cells are large. Their dendrites are richly branched and are directed toward the periphery. Their neuraxes give rise to arcuate fibers. Among these larger cells are small cells of the granular type, the exact connections of which are not known as yet. The distribution of the gray and the white of the area is seen most distinctly at the level of entrance of the vestibular nerve (fig. 184). In this

440 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

region, three areas may be distinguished : a dorsal area or nucleus, a medial nucleus, and a ventral nucleus. The dorsal nucleus (Dors, nucl., fig. 184) receives chiefly the larger fibers of the nervus lateralis anterior. The area is often called the anterior lateral-line lobe. The medial nucleus (dorsomedial nucleus of Johnston; Medial nucl., fig. 184) receives the finer fibers of the anterior lateral-line nerve, and more especially the fibers of the posterior lateral-line nerve. The ventral nucleus (the ventrolateral nucleus of Johnston; Vent, nucl., fig. 184) is primarily the region of termination of the vestibular nerve. The two kinds of cells referred to above occur in each of these nuclei. Large, scattered, spindle

Fig. 185. The nucleus octavo-motorius anterior in Petromyzon. Drawing by van Hoevell.

shaped cells are present in the dorsal nucleus. An accumulation of such cells forward, near the cerebellum, constitutes the nueleus octavo-motorius anterior {Schilling, ’07). Others of these cells lie in the caudal portion of the lobe, where they constitute a nucleus octavo-motorius posterior.

The entire area statica is covered by a continuation of the molecular layer of the cerebellum — the crista cerebellaris. In petromyzonts this is a relatively thin layer of neuropil which receives dendrites of the larger cells which lie beneath it and the terminal branchings of many root fibers of the lateral line and vestibular nerves. These larger cells also send their dendrites into the central gray of the lobe. Frontally, these cells gradually pass over into the primordial Purkinje cells of the cerebellum. In so doing, the dendrites running medialward are lost, while the lateral dendrites extend into the molecular layer. The crista cerebellaris also contains a great many neuraxes similar to the parallel fibers of the cerebellum. These fibers run through the entire length of the crest and in part

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

441

enter the molecular layer of the cerebellum. Of the three nuclei mentioned above, it is the medial nucleus which passes over into the cerebellum (see Pearson, ’33, Dissertation). Furthermore, it is the large cells of this nucleus which, for the most part, become continuous with the Purkinj e-like elements of the cerebellum (Johnston, ’02).

The relation of the two lateral-line nerves to their central gray is as follows ; The posterior lateral-line nerve in Petromyzon enters the medulla oblongata at the level of the nervus glossopharyngeus, but somewhat cephalad and dorsal (Tretjakoff, ’09 ; Ariens Kappers, ’20) to its sensory root. It enters usually as a single bundle, conspicuous for the thickness of its fibers which run frontalward along the dorsolateral periphery of the medulla oblongata and end chiefly in the medial nucleus of the static area ; a large number of its fibers (Ariens Kappers, ’20) enter the crista cerebellaris and there terminate in synaptic relations with the dendrites of the Purkinje-like cells. Part of the fibers of the posterior lateral-line nerve enter the cerebellimi itself ( Tretjakoff, ’09 ; Ariens Kappers, ’20), where they are partly crossed and partly uncrossed. The decussating fibers form the coarse fibers of the commissura cerebelh. Root fibers of the posterior lateral-line nerve are not beheved to dichotomize, but part of them are supposed to descend (Tretjakoff, m).

The anterior lateral-line nerve (ner\ms lateralis facialis ; fig. 184) enters the medulla oblongata at the entrance of the facial and vestibular nerves, slightly dorsal and cephalad to the latter. It consists of two distinct root bundles, distinguishable through the caliber of their fibers (Johnston, ’02; Tretjakoff, ’09). The fibers of the ventral bundle are finer than are those of the dorsal. They end chiefly in the medial nucleus, into which the nervus lateralis posterior also enters. Others pass to the ventral nucleus; fibers ascend to the cerebellum (Johnston, ’02 ; Tretjakoff, '09 ; Ariens Kappers, ’20) and to the dorsal nucleus, (Johnston, ’02). The dorsal bundle consists of coarser fibers. It comes into relationship with the dorsal nucleus. Some fibers reach the medial nucleus while others dichotomize and extend forward and backward. The ascending fibers of t.his root come into synaptic relation with the cells of the anterior octavo-motorius nucleus (fig. 185), and with the cells of the cerebellum. They exhibit a peculiar cap-shaped type of synapse in relation with the cells which form the nucleus. The descending fibers of the dorsal root branch around the smaller and larger cells of the anterior lateral-line lobe. These larger cells (Schilling, ’07) lie principally in the most caudal part of the lobe and there form a more or less pronounced nucleus which might be termed “nucleus octavo-motorius posterior” (Ariens Kappers, ’20), in contrast to the anterior nucleus mentioned above. Since, as has been stated, anterior lateral-line fibers end in the medial nucleus (the principal nucleus of the posterior lateral-line nerve), the two lateral line systems are not strictly separated from each other and a correlation of their stimuh may take place.

The vestibular organ of the cyclostomes has a much more complicated structure than do those of the lateral-line systems. In the myxinoids it consists of a saccus communis, from the lateral ends of which a single arch arises. The

442 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

vestibular nerve ends in two branches along the saccus communis. These branches are termed the ramus posterior and the ramus anterior. In Petromyzon the saccus communis consists of two divisions and the arch has also differentiated into two canals, a canalis anterior and a canalis posterior. Each of these shows an ampullar enlargement. The ramus anterior of the vestibular nerve terminates in relation to the ampulla of the anterior canal. The ramus posterior has a similar relation to the posterior canal. Small branches of the nerve innervate the saccus.

There is an intimate relation between the central terminations of the vestibular and of the lateral-line nerves, particularly those of the anterior lateral-line nerve. The vestibular nerve enters by two roots, a posterior or dorsal root and a ventral root. After a partial decussation the fibers of the dorsal root ascend and descend. Among them is a group of less coarse fibers which extends toward the area statica and there divides into ascending and descending bundles {Tretjakoff, ’09). One ascending bundle enters the dorsal part of the area statica and thus provides for a correlation of the lateral line and vestibular systems. Another ascending bundle runs to the cerebellum. The descending bundle terminates in the caudal part of the dorsal nucleus. Non-dichotomizing fibers of this root run forward to the ventral (ventrolateral) nucleus (Tretjakoff, ’09).

The fibers of the ventral root vary greatly in size. Part of them are said not to divide but to pass directly into a descending root which terminates chiefly in the caudal part of the ventral nucleus. Other thick fibers of the ventral root dichotomize after their entrance to the medulla oblongata and then ascend and descend (Tretjakoff, ’09). The ascending fiber may terminate in the ventral nucleus or may extend farther forward. The finer descending branch terminates in the posterior part of the ventral nucleus, which is reached also by a considerable number of non-bifurcating fibers. It appears probable that both roots of the vestibular nerve have analogous functions, since both have ascending and descending fibers, but the ventral root is concerned particularly with aboral reflexes while the dorsal root provides a means of correlation with the more frontal centers (cerebellum?).

A number of larger cells ‘ are present in the ventral (or ventrolateral) nucleus. The neuraxes of these cells decussate and then, for the most part, descend with bundles of the fasciculus longitudinalis medialis. They are concerned in vestibulo-motor reflexes, since they transmit stimuli by way of the medial longitudinal fasciculus to the motor centers of the medulla oblongata and the cord (Tretjakoff, ’08). From the acoustico-lateral area as well as from other sensory centers of the medulla oblongata ascending fibers pass to the midbrain, eonstituting a primordial bulbar lemniscus (see Pearson,, ’33; Dissertation).

It is to be noted that the cephalic end of the area statica is continuous with the cerebellum. Caudally this lobe extends into the region of the dorsal root

It is possible that among the cells of this nucleus may lie the forerunners of the Mauthner

cells, specialized neurons occurring in the higher fishes, which transmit stimuli from the octavolateral (acoustico-lateral) region to the tail region of the spinal cord. Other more oval and slightly more ventrally placed cells are more closely related to the tangential nucleus described for fishes by Ram6n y Cajal (’08 ; Ariens Kappers ’20).

THE LATERAL-LINE AND ACOUSTIC SYSTEMS 443

fibers of the spinal cord and the area supplied by the descending root of the trigeminal nerve. Such relations demonstrate its intimate interrelation with other somatic sensory stimuli. Furthermore, the cerebellum, in its most primitive form, is to be regarded as a continuation and further differentiation of the static area. This latter point will be considered again in the discussion of the cerebellum (see Chapter VII).

There is apparently less differentiation in the lateral-hne area of myxinoids than of petromyzonts. Ayers and Worthington (’08) were unable to distinguish as separate nuclei the nucleus ventrolateralis and the nucleus dorsomedialis, and found the lobus lineae lateralis only partly separable. They termed the whole area the nucleus acusticus. Jansen (’30) found the system small and the central area of reception represented by a nucleus of the acoustico-lateral system, consisting of cells similar in type to those of the nucleus of the descending root of the trigeminal. Scattered along the acoustico-lateral fibers he identified certain larger elements which form a group of large cells in the caudal part of the area, the nucleus acustico-lateralis magnocellularis. This latter nucleus is said to receive both cutaneous and lateral-line fibers.

Lateral-line fibers enter the acoustico-lateral area (or nucleus acusticus or acustico-lateralis) in myxinoids, as in other cyclostomes, over the anterior and posterior lateral-line nerves and over the two roots of the vestibular nerve — the ramus acusticus utricularis and ramus acusticus saccularis (Ayers and Worthington ’08; Jansen, ’30). In Bdellostoma, the posterior lateral-line nerve, on entrance to the medulla oblongata, passes directly to the acoustic nucleus, ultimately reaching a position on its medial border about midway between its dorsal and ventral poles. From this position it rims forward along the medial border of the nucleus until near its frontal termination, and there distributes through dorsally and ventrally directed branches. Some fibers turn backward to reach more caudal portions of the nucleus, and occasionally a bifurcating branch is seen, but for the most part, Ayers and Worthington found the posterior lateral-line nerve in Bdellostoma unbranched. The anterior lateral-line nerve, as described for Bdellostoma, consists of a series of bundles accompanying other nerves of the region. Two of such bundles are said to accompany the posterior and anterior trunks of the trigeminal nerve, forming the posterior and anterior parts of the anterior lateral-line nerves (Ayers and Worthington, ’08). Such fibers often have been considered as trigeminal nerves (Sanders, ’94; Holm, ’01). The posterior part of the anterior lateral-line nerve, with the accompanying trigeminal fibers, becomes entangled on its entrance to the brain with the fascicles of the ramus acusticus utricularis, and Ayers and Worthington were unable to separate it centrally from that ramus. The anterior part, in company with the trigeminal fibers enters the lower border of the nucleus acusticus, turns caudalward, and then distributes to the nucleus, the trigeminal fibers continuing into the general cutaneous area of the medulla oblongata. Ayers and W orthington expressed the opinion that still another branch of the lateral-line nerve, a bundle accompanying the facial nerve, is present in Bdellostoma, although they were unable to demonstrate it to their complete satisfaction.

444 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ramus utricularis and ramus saccularis of Bdellostoma send fibers forward and medialward to reach the acoustic nucleus, which they enter. They divide dichotomously and one branch ascends and the other branch descends. The ascending branch may redivide a number of times to distribute to various parts of the acoustico-lateral area.

The secondary connections of the acoustico-lateral system of Bdellostoma, according to Ayers and Worlhinglon, are as follows : (1) with the motor centers of the opposite side through bundles which decussate in the ventral raph4 ; (2) with the nucleus funiculi through an acoustico-funicular path ; and (3) with the cerebellum by means of a small acoustico-cerebellar path. Those interested in the details of these connections should consult the original paper. The distribution of the acoustico-lateral fibers to the acoustico-lateral area of the medulla oblongata (with special fascicles to the nucleus acustico-lateralis magnocellularis) has been given with less detail for Myxine glutinosa by Jansen (’30).

The Acoustico-lateral Systems of Plagiostomes

Both the lateral line and vestibular systems are developed much better in the plagiostomes than in the cyclostomes. The increased development is associated with changes in the relations at the periphery of the body, which include an increase in the number of end-organs of the lateral-line system. Thus more than one hundred lateral-line organs are found in the canals on the body and head. The ampullae of Lorenzini (78, p. 437) are found also on the head in the torpedo ; the vesicles of Savi (’44) are present. Various accounts of the plagiostome lateral-line canals and sense organs are to be found in the literature {Ewart, ’92; Ewart and Mitchell, ’92; Klinkhardt, ’05; Johnson, ’17, and others).

In Mustelus canis and Squalus acanthias {Sydney Johnson, ’17), lateral-line sense organs occur either as surface or pit organs, or as canal organs. The former are comparable to those found in Petromyzon and tailed amphibians, but the canal organs are inclosed within the epithelial lateral-line canals, appearing as differentiated epithelial structures (usually on the superior wall). The method of formation of these canals is interesting. In the thickened ectodermal line which indicates the position of the future canals, a groove appears and then spreads along the thickened ectodermal line, usually in the direction of growth of this line. Thus in the formation of the lateral canal, the groove appears at the cephalic end first and gradually extends caudalward. In this groove are an almost continuous row of enlargements which represent differentiating sense organs. With the deepening of the walls of the groove, short portions of it close over and fuse, leaving, between the fused portions, openings which are the anlagen of the canal tubules. The openings or pores gradually become smaller. In the lateral-line sense organs of these plagiostomes Johnson (’17) found four cell types : sense cells, columnar cells, spindle-shaped cells, and basal cells, suggesting those found later in tailed amphibians by Charipper (’28) and Chezar (’30). However, Johnson’s description of the nerve terminations shows important differences. This latter observer carried the nerve fibers to the region

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

445

of the hair or sense cells and found that there the fibers divide at various levels, many of them branching between the basement membrane and the hair or sense cells. Such branches of a single nerve may supply a munber of organs. From these primary branches, after repeated division, fine fibrillae are given off and these pass between the hair cells and may extend from a fourth to three-fourths of the length of such a cell. Often they terminate on end knobs. According to Johnson, they do not form “cups” around the base of the cell, and their ultimate termination is on and not within the cell.

N. lat anterior N.vest

« » (

CHnx A A

Fig. 186. The system of lateral-line organs and ampullae of Lorenzini in Laemargus boreaUs. Ewart.

On the head ; C.S.O., canalis supraorbitalis (runs over the orbit) ; C.I.O., canalis infraorbitalis; C.Hm., canalis hyomandibularis; A., ampulla of Lorenzini.

On the body : C.L., canalis lateralis.

The nerves supplying the lateral-line apparatus and their central terminations are the same in general in plagiostomes as in Petromyzon, but they are greatly increased in size in the former animals. The posterior lateral-line nerve in plagiostomes (as in Petromyzon) enters at the level of entrance of the glossopharyngeal nerve, and dorsal to its sensory component. Peripherally the nerve is distributed for some distance with the vagus. The anterior lateral-line nerve enters the medulla oblongata dorsal to the facial and vestibular (figs. 186-188).

The static area is much more highly developed in plagiostomes than in cyclostomes, and particularly so in those forms such as Hexanchus which have an everted cerebellum. The dorsal nucleus forms a distinct mass connected to the rest of the medulla oblongata by means of a narrow zone (fig. 188). The medial and ventral nuclei occupy the dorsal part of the medulla oblongata and the area immediately ventrolateral to it. The dorsal nucleus is called the lobe of the anterior lateral-line nerve, while the medial nucleus is frequently termed the lobe of the posterior lateral-line nerve. This latter nucleus, however, receives fibers from both the anterior and posterior nerves of the lateral-line system, although the component from the posterior nerve is the larger. Both lobes are

446 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

covered by the crista cerebellaris, which is also much more clearly developed and larger in the plagiostomes than in the cyclostomes. The crista extends caudalward along the medulla oblongata almost to the level of the calamus scriptorius. As in the cyclostomes, larger cells and smaller granule-hke cells are found in the area statica. The larger cells are of two types. The one type lies embedded in the gray substance of the lobe itself. Its dendrites pass in all directions from the cell body, although they may be somewhat more numerous laterally. The neuraxes of these neurons enter the arcuate fibers. The second type of larger

fibr desc.let ant

Fig 187 The entrance of nervus lateralis posterior in Scyllium canicula Schcpman

cells lies near the crista and its neurons send the majority of their dendrites into this area. These dendrites are thorny and resemble, in this particular, the dendrites of Purkinje cells. However, they are oriented differently, for they extend medialward into the substance of the lobe.

The ventral nucleus of the area statica has less of the granule-like cells and a greater number of the larger neurons (especially in its lowest portions) than do the dorsal and medial nuclei. The largest cells of all are obviously homologous with certain differentiated elements (see p. 456) found in the teleosts.

The connections of the lateral line and vestibular nerves with their nuclei are analogous in certain respects to those in cyclostomes. The nervus lateralis posterior (fig. 187) for the most part extends forward along the dorsolateral periphery of the medulla oblongata as far as the level of entrance of the antenor

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

447

lateral-line nerve and the vestibular nerve into the crista cerebellaris. This point marks the cephalic end of the granular layer of the posterior lateral-line lobe, or the medial nucleus of the area statica. A few fibers of this root run farther cephalad and, lateral to the ascending fibers of the vestibular nerve, enter the auricle of the cerebellum. Part of the fibers terminate in the auricle and part of them pass to the eminentia lateralis cerebelli. This eminentia is formed

Fig. 188. The entrance of nervus lateralis anterior and nervus vestibularis in Scyllium canicula.

Schepman (’18).

by the thickening of the lateral wall of the cerebellum at the level where it passes over into the medulla oblongata. A small number of root fibers of the posterior lateral-line nerve turn caudalward at their place of entrance and descend lateral to the descending root of the trigeminal nerve, probably distributing to acousticolateral gray.

The anterior lateral-line nerve (fig. 188) in plagiostomes, as in petromyzonts, consists of two branches. These are not differentiable in the plagiostomes through differences in the caliber of the fibers, but the dorsal branch is more deeply stained in the Weigert material. The ventral branch enters the medial

448 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

nucleus (posterior lateral-line lobe) as did the fine fibers of the anterior lateralline nerve in petromyzonts. Such fibers are easily distinguishable from those of the vestibular nerve in the more primitive forms, but are intermingled with them in such plagiostomes as Acanthias {Schepman, T8). A small part of this ventral branch reaches the medial nucleus by way of the narrow connecting bridge of gray between the anterior and posterior lateral-line lobes.

The dorsal root of the anterior lateral-line nerve ends in a separate enlargement, the anterior lateral-line lobe, or the dorsal nucleus. In most selachians this lies medial to the posterior lateral-line lobe, but in Hexanchus it covers the lobe on the outer side. Its fibers, which apparently dichotomize, ascend and

descend. From this dorsal branch a very few fibers pass to the cerebellum. The branch breaks up within the substance of the anterior lobe. The fibers appear to distribute also to the molecular layer and to the crista cerebelli, ending around the dendrites of the Purkinje-hke cells, after the fashion of climbing fibers.

The labyrinth of plagiostomes is much enlarged compared with that of cycloFig. 189. The labyrinth of Acanthias vulgaris. Retzivs. Stomes. ThuS while the petro myzonts have only two canals, an anterior and a posterior, the plagiostomes have a third one, the external canal. These three canals are present throughout all higher vertebrates. The saccus communis of cyclostomes has differentiated into a sacculus and utriculus in plagiostomes. Likewise there is a beginning, in these latter forms, of a lagena, which, however, is poorly developed. The macula neglecta has also appeared. Thus the acoustic nerve of the plagiostomes receives its fibers from a greater number of end-organs than does the corresponding nerve in the cyclostomes.

Centralward, the fibers run in the two roots characteristic of vertebrates. The more cephalic branch, ramus anterior, has its fibers from the ampullae anterior and externa, and from the utriculus. The more caudal branch, ramus posterior (which in higher vertebrates is joined by the cochlear branch), is composed of fibers from the ampulla posterior, the sacculus, the macula neglecta, and the papilla lagenae. Centrally, the ramus posterior becomes the dorsal root and the ramus anterior the ventral root.

Some of the fibers of the dorsal root terminate centrally, near their level of entrance, around large cells homologous with the cells of origin of the arcuate fibers in petromyzonts and with the cells of Deiters in higher vertebrates. In plagiostomes, such cells lie lateral and dorsolateral to the descending root of the trigeminal, and for the most part send their neuraxes in a caudal direction. Fibers

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

449

of the dorsal root divide, then ascend and descend ; the branches thus formed are analogous to each other in size and shape. The descending bundle lies very near to the ventricle of the medulla oblongata. It forms here an oval, compact mass (fig. 190), which can be traced caudalward to the transition from the medulla oblongata to the spinal cord. The ascending fibers likewise lie near the corner of the ventricle medial to the ascending lateral-line fibers. They can be traced as far forward as the auricle and the eminentia lateralis cerebelli, but they do not enter the body of the cerebellum. The ventral, or anterior, root of the eighth nerve lies directly adjacent to the root of the facial nerve, then swings medial to it and runs caudalward along the base of the medulla oblongata.

Fia. 190. The entrance and course of vestibular fibers in Acanthias vulgaris.

Schepman.

Gradually it acquires a somewhat more dorsal position. About midway between the level of entrance of the facial and glossopharyngeal nerves a part of its fibers cross and become lost in the ventral tegmental gray of the medulla oblongata. Some of them can be traced as far caudalward as the spinal cord, where they break up in the ventro-lateral part of the ventral horn, in relation to the motor nuclei. They may be accompanied by secondary fibers of the same type. The tract thus formed is called the tractus octavo-spinalis cruciatus. The nuclei of termination of the lateral line and vestibular nerves are intercoimected by means of fibers decussating in the ventral tegmental region. From the whole area statica, secondary fibers arise for reflex and coordinating functions. These fibers ascend and descend in the medial longitudinal fasciculus and form the tractus octavo-motorius. In their course to the midline they run with the cerebellomotorius system close to the ventricular gray, and enter the bundle in the midline as crossed and uncrossed fibers. The large cells, from which many of these fibers

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

451

the S3^stem is greatly hypertrophied. Moreover, due consideration is given to the conditions in ganoids as described by Johnston (’01), Ariens Kappers (’07, ’20), and Hocke Hoogenboom (’29) and in teleosts as studied by Haller (’98), Wallenberg (’07), Ramon y Cajal (’08), Tello (’09), Franz (’ll), Bartlemez (’15), Burr (’28), and others. The region has been studied in the Dipnoi, Ceratodus, by Holmgren and van der Horst (’25).

Because of morphologic changes in the region it is often more difficult to delimit sharplj’- the lateral-line nuclei of the acoustico-lateral system in certain of the more highly specialized fishes than in plagiostomes. Nevertheless, in manj”- cases typical nuclear masses are demonstrable. The dorsal nucleus of the acoustico-lateral area (the anterior lateral-line lobe) is well developed in certain ganoids (Acipenser, Johnston, ’01 ; Polyodon, Hocke Hoogenboom, ’29) and teleosts (such as the siluroids, Berkelbach van der Sprenkel, ’15 ; possibly present in Orthagoriscus, Burr ’28) but is believed to be absent in certain ganoids (as Amia) and certain teleosts. Apparently invariably present is the medial nucleus of the area (the posterior lateral-line lobe), which receives fibers of the anterior lateralline nerve {Hocke Hoogenboom, ’29, and others) but which is particularly the nucleus of termination of the posterior lateral-line nerve (see below). The dorsal nucleus (where present) and the medial nucleus are covered by a crista cerebellaris which is particularl}"- large in siluroids and consists there {Berkelbach van der Sprenkel, ’15), as in selachians, of two divisions each of which corresponds with one of the above mentioned dorsal and medial nuclei. In certain siluroids the crista of the anterior lateral-line lobe extends farther caudalward than that of the posterior lateral-line lobe ; in others, as Arius, this relationship is reversed. Laterally the two cristae are independent of each other. Frontally, the crista of the posterior lateral-line lobe is much larger than that of the anterior lobe and unites with the molecular layer of the cerebellum early. The half moon shaped crista of the anterior lobe extends forward for a short distance beneath the cerebellum and then, in Siluris glanis at least {Berkelbach van der Sprenkel, ’15), fuses with that of the opposite lobe. This fused portion becomes continuous with the cerebellum in the same way as does the crest of the posterior lateral-line lobe. Frontalward the granular layer of the posterior lobe (lob. lin. lat., fig. 191) increases greatly and is said to form the massa granularis of certain teleosts {Franz, ’ll).

The fibers of the posterior lateral-line nerve enter at the level of the glossopharyngeal nerve. They run frontalward along the dorsolateral periphery of the brain (fig. 191, n. lat. post.). A small portion of them are thought to turn caudally, coursing beside the descending fibeis of the vestibular, to near the level of the spinal cord {Ariens Kappers, ’06; see also ’07). The ascending and descending fibers of this nerve are said to be produced by bifurcation of the root fibers {Tello, ’09) . The frontal fibers of the posterior lateral-line nerve end chiefly in the enlarged portion of the posterior lobe (the medial nucleus) and appear to distribute also to the overlying crista and to the massa granularis of the cerebellum. This is particularly true in Mormyrus, where this nerve and its end station attain an enormous development (fig. 192), but is also true in other tel

452 NERVOUS SYSTEMS OF "^TERTEBRATES AND OF MAN

eostomi, including the ganoid Polyodon (Hocke Hoogenboom, ’29). In siluroids these branches of the posterior lateral-line nerve very evidently are accompanied by fibers from the anterior lateral-hne nerve and from the acoustic-vestibular nerve. Thus the medial nucleus in both teleosts and plagiostomes does not belong exclusively to the posterior lateral-line system. Certain fibers of tliis posterior lateral-line system reach the cerebellum or at least that portion of it which conesponds to the auricle.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

453

entrance to the lobe, the fibers ascended and descended, apparently following bifurcations. This observer followed the ventral root of the anterior lateral-line nerve in Polyodon to the medial nucleus (or posterior lateral-line lobe), to a less extent to the dorsal lobe (or anterior lateral-line lobe), and to the cerebellum and the crista cerebellaris. Dorsal and ventral roots of the anterior lateral-line nerve, corresponding to those described above, were identified and described in Ceratodus by Holmgren and van der Horst (’25). In Orthagoriscus, Burr (’28) traced fibers of the ventral root of the anterior lateral-line nerve into a small celled continuation of his dorsal vestibular nucleus.

The ventral root of the anterior lateral-line nerve contains remarkably thick fibers in siluroids and these run in a caudal direction and end in or near the vicinity of the nucleus tangentialis and Mauthner’s cells. Although a part of these thicker fibers also reach the posterior lateral-line lobe, the majority of them accompany the vestibular system. These connections demonstrate the interrelations existing between the lateral-line system and the vestibular apparatus. The peripheral connections of these coarse fibers of the anterior lateral-line nerves do not appear to have been demonstrated. Their thickness and their direct connection with the motor tegmentum of the medulla oblongata suggest that centrally they are concerned with the establishment of reflex pathways to motor centers.

In Mormyrus the area statica of the medulla oblongata reaches a relatively large size and covers the other medullar centers like a cap (fig. 192).

The labyrinth in teleosts — unlike that of plagiostomes — does not open at the surface, for the ductus endolymphaticus in the former animals ends blindly and is covered by the meninx. The lagena shows considerably more differentiation in certain teleosts," such as Cyprinoidea and Cryptinoidea, than in plagiostomes. Retzius (’81), Ramdn y Cajal (’08), and Tello (’09) have studied

® An unusual relationship is seen in the labyrinth of Cynoscion regales, where the sacculus and utriculus are not connected with each other (Parker),

Cr.l. L.orc. M. otL.

Tr.Bp.c.

Fig. 192. Excessive development and fusion of the lobi laterales posteriores of the two sides (these are the nuclei of termination of N.lai.post.) in Mormyrus caschive. Berkelbach van der Sprenkel.

Cr.l., Crista lobi lateralis posterioris; L. of C., cell layer of lobus lateralis posterioris ; M. of L., medullated fiber layer of lobus lateralis posterioris; Tr.sp.c., Tractus spino^eerebellaris.

454 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the vestibular nerve and its connections most carefully in teleosts. Retzim (’81) and de Burlet (’29) identified the seven areas concerned with the peripheral distribution of the vestibular nerve. These are the cristae of the ampullae of the three semicircular canals, the macula utricuh and macula sacculi, the macula, crista, or papilla neglecta ® (which is lacking in the flat fishes and in certain others, as for example Reniceps, Gobius, Lophius, Zeus, Callionymus, Gadus — Reizim ’81), and the papilla lagenae (macula lagenae of de Burlet, ’29, which is absent in certain fishes). The fibers distributing to the three ampullae are partly

coarse and partly fine ; those to the macula or crista neglecta are very coarse, those to the utriculus and sacculus are of medium size, and those to the papilla lagenae are relatively few in number and of very fine caliber. The ventral vestibular root, which corresponds to the ramus anterior vestibularis, contains coarse and fine fibers of the ampullae externa and anterior, and fibers of medium size from the utriculus. The dorsal vestibular root, which corresponds to the ramus posterior, carries coarse and fine fibers from the ampulla posterior and the macula or crista neglecta, medium sized fibers from the macula sacculi,'* and fine fibers from the macula lagenae. Each root thus carries fibers of different calibers.

The two major vestibular nuclei which are known to be the recipients of the vestibular nerves of teleosts are the tangential nucleus (of Ram6n y Cajal, ’08,

’ This is generally called a macula neglecta, but Benjamins (’13) showed it to be a crista. De Burlet (’29) called it the papilla neglecta.

‘ In certain teleo.sts such ns Salmo salar, Clupca harengus, and Anguilla, the ramus anterior gives oil a branch to the sacculus.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

455

and Tello, ’09) and Deiters’ nucleus. The former nucleus is divisible in certain fishes into two portions, a more cephalic part lying ventral to the Deiters’ nucleus and a caudal part situated farther dorsally and continuous caudally with the Deiters’ nucleus. A descending vestibular nucleus has been identified in teleosts (see Pearson, ’33 ; Dissertation). This nucleus is formed by the fusion caudally of the ventral and tangential nuclei.

The manner of termination of the vestibular fibers about the cells of the tangential nucleus is very characteristic. The endings are pecuhar, pericellular baskets or caps, formed by the terminal fibers around the cell body {Beccari, ’31), and are similar to those found about the spindle shaped cells of the nucleus octavo-motorius of cyclostomes (Ariiins Kappers, ’20) and the neurons of the tangential nuclei of reptiles and birds (see pages 468 and 476). After entrance to the medulla oblongata, the coarse fibers of the vestibular root separate from the fine and medium-sized fibers. The very coarse fibers, coming from the three ampullae and from the macula or crista neglecta, terminate in the tangential nucleus soon after their entrance to the medulla oblongata. The coarse fibers of the ramus lateralis of siluroids, and the homologous fibers of certain other fishes, also reach this nucleus. From this nucleus there is made immediate provision for reflex connections with motor centers through secondary and largely through contralateral paths. Such secondary fibers are said to be accompanied by certain primary or root fibers {Wallenberg, ’07).

The medium sized fibers from the utriculus end around the somewhat more dorsally situated nucleus of Deiters. Those from the sacculus, the crista neglecta, and the lagenaend partly in reticular cells (see below), together with finer fibers from the macula, and partly form ascending and descending root bundles in the dorsal part of the medulla. The ascending fibers end in the medial nucleus and in the crista cerebellaris. They are accompanied by fibers of the nervi laterales anterior and posterior. They give off crossed and uncrossed fibers to the cerebellum. The descending fibers run at first between the descending root of the trigeminal and the root of the facial, and then join the descending fibers of the lateral-line nerves. From Deiters’ nucleus descending bundles, largely homolateral, pass to the motor centers.

The finer fibers — part of those from the ampullae and sacculus, and all of those from the macula lagenae — form ascending and descending fibers in the dorsal part of the static lobe (fig. 193, Fibr. dors. Vestib.). They end in its deeper gray and in the crista cerebellaris that covers it. From there, stimuli are transferred to the roof of the midbrain by means of the fasciculus longitudinalis lateralis or the lemniscus acustico-lateralis. Thus the functional localization which exists in the central nervous system, as evinced by the presence of distinct nuclear groups, is indicated at the periphery by the different caliber of fibers. Those bundles which are carrying impulses that are immediately relayed to motor centers are composed peripherally of coarser fibers. The lagena fibers belong to the group which terminates in the more dorsal part of the medulla. From that region the impulses are relayed to the midbrain.

456 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Among the cells of the nucleus motorius tegmenti (which nucleus receives fibers from Deiters’ and the tangential vestibular nuclei) are neurons similar to the cells of Muller found in cyclostomes. These are, for the most part, large cells which primitively lie in the region of the motor nuclei of the medulla oblongata, but which secondarily, through the action of neurobiotactic factors, have come into more intimate relation with the primary sensory nuclei (Bartelmez, ’15) and have formed a distinct nuclear mass. This migration toward the primary sensory centers provides for more rapid reflexes involving these centers and the motor centers for supply of the somatic musculature. This tendency toward migration has led to the accumulation of a large number of these cells near the region of entrance of the vestibular, since such connections facilitate the responses involving the maintenance of equilibrium. Some of these cells are found in the region of the raphd, where they receive collaterals of the crossed vestibular and lateral-line fibers. It is these latter cells that are to be homologized particularly with the cells of Muller or wdth the large neurons of the static area of cyclostomes.

In this differentiation of the reticular elements there is one element which must be emphasized particularly. This is the cell of Mauthner (fig. 194) which, because of its size, has interested many authors. Beccari (’07), Bartelmez (’15 ; with Hoerr, ’33), Marui (’18 and ’19), and Tiegs (’31, see p. 131) have studied this neuron. The cell body and the dendrites of the cell of Mauthner are enormous. The cell body is situated in the midline near the floor of the ventricle, but its dendrites extend lateralward and ventralward to very near the periphery of the medulla oblongata. Its lateral dendrite is in relation with the incoming vestibular fibers, particularly those {Beccari, ’07) from the sacculus. Homolateral and contralateral vestibular fibers send collaterals to the axon cap and to the network of telodendria surrounding the cell body. In addition to these direct vestibular connections, the cell has a number of secondary octavo-lateral or acoustico-lateral connections, since both crossed and imcrossed collaterals of decussating fibers and some accompanying stem fibers are from the dorsal and ventral nuclei of the static area and terminate in the pericellular network or on the axon cap. Also, fibers of the cerebello-motorius system, fibers from the sensory trigeminal as well as gray bundles from the mesencephalon, and from the optic tectum synapse with Mauthner’s cell. Those fibers from the cerebellum and from the chief sensory nucleus of the trigeminal terminate chiefly in relation to the ventral dendrites. In some teleosts such as Ameiurus an upper and a lower ventral dendrite are present, the latter having the position of the single ventral dendrite found in certain bony fishes. It is the upper dendrite, where two are present, which synapses with the cerebello-motorius fibers and those from the trigeminal nucleus. The lower ventral dendrite is in synaptic relation with the ventral nucleus of the static area, and with the crossed and uncrossed tecto-bulbar paths. The neuraxis of Mauthner’s cell is traceable to the caudal end of the spinal cord. It provides a “final common path” for impulses to the tail. These connections and relations show the great importance of this cell in the preservation of equilibrium and in such acts, necessary to maintain equilibrium, as involve movements of the tail

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

457

This latter fact explains the great development of the cell in fishes. In some fishes such as the Dipnoi, Ceratodus {Holmgren and van der Horst ’25), the Mauthner fiber arises from four or five axons which, coalescing, lose their individual myelin sheaths, becoming surrounded by a single myelin sheath in which, however, each axon can be recognized. A similar relation is found in the giant

Fig. 194. Mauthner’a cell in a young salmon. Bartelmcz. F.L.M., fasciculus longitudinalis medialis; Ax., axon; Ax.Cp., axon cap; Nud.D; nucleus of Deiters.

nerve fibers of the Annelids (Ashworth). The reticular cells of the nucleus motorius tegmenti pars postmauthnerea of Bartelmez (’15) have synaptic relations with the vestibular centers similar to those of the lower ventral dendrites of Mauthner’s cell.

The other secondary connections of the static area may be summarized briefly as follows. The areas of the two sides are interconnected, as in plagiostomes, by commissural fibers, but while the octavo-lateral or acoustico-lateral commissure in plagiostomes is represented as a ventral system of decussating fibers, in teleosts such fibers eross dorsally. Usually such dorsally crossing fibers lie near the floor of the ventricle, ventral to the medial longitudinal fasciculus, but in Gadus they decussate in a connnissural plate which lies above the fourth ventricle and in this position connects the lateral-line lobes of the two sides. Sometimes two such commissural bundles may be discerned. Two kinds of fibers are likewise distinguishable in the secondary systems ; those concerned directly with motor centers, and those which carry impulses to the midbrain.

458 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Of the first group, those arising from the tangential nucleus and the nucleus of Deiters have been mentioned. In addition to these, there are bundles from the dorsal part of the medulla oblongata comparable to the octavo-motorius tract

of plagiostomes. These, increased by cerebello-motorius fibers, accompany the medial longitudinal fasciculus. They consist of crossed and uncrossed fibers which both ascend and descend. The ascending fibers terminate in the motor nuclei of the oculomotor and trochlear nerves. The descending ones pass to the nucleus of the abducens and to the motor neurons of the spinal cord.

The ascending tract to the tectum, the fasciculus longitudinalis lateralis, or the acoustico-lateral lemniscus {Herrick, ’14 ; see bibliography for amphibians) originates in teleosts from the dorsal and medial nuclei of the area statica. It crosses the midline directly below the medial longitudinal fasciculus and runs forward to the torus semicircularis of the midbrain. Direct terminations of this tract within the

Root of IV Nucl. IV

Fig. 195. A cross section at the transition of the medulla oblongata into the midbrain in Monn 3 TH 8 caschive. Van Gieson preparation.

Note the great development of the acoustico-lateral lemniscus {Lai. Urn.).

Inf. lobe, lobus inferior hypothalami; Vah. eerebel., Carp, eerebel., valvula and corpus cerebelli. Berkelbach van der Sprenkel.

Optic tectum have not been demonstrated. Wallenberg traced processes of secondary neurons of the octavo-lateral or acoustico-lateral region to the thalamus and hypothalamus. Since the acousticolateral lemniscus arises principally

from the dorsal and medial nuclei,'^ and these nuclei receive, in addition to fibers from the vestibular root, many fibers of the lateral-line system from

both anterior and posterior lateral-line nerves, it is not surprising that an

^ These nuclei grow to a relatively large size in Mormyrus. The medial nucleus, or lobus lateralis posterior, attains such a size that a bilateral union in the form of a large cap occurs. This cap incloses the smaller anterior lateral-line lobe, which likewise fuses with its homologous lobe of the other side (fig. 192). The fusion of the anterior lateral-line lobes occurs doreal and caudal to that of the posterior lobes. Although the vestibular apparatus in these animals is relatively small, the acoustico-lateral system and its nucleus of termination is considerably enlarpd, as was stated above. Moreover, the valvula cerebelli, which is connected with the torus, exhibits a very distinct enlargement. As a matter of fact, the great size of the valvula in these animals is due to the stimuli which reach it directly, over the roots of the enlarged lateral-line nerves, and indirectly, by way of the lemniscus acustico-lateralis and the torus semicircularis.

459

THE LATERAI^LINE AND ACOUSTIC SYSTEMS

hj-pertrophy of these nerves, such as occurs in Mormyrus {StcjidcU, T4, 'Ha • licrlcclbach van dcr Sprcnlccl, ’15), is attended by a marked increase in size of the acoustico-lateral lemniscus.

The question — as to whether or not fishes are able to hear — has received much attention. Piper (’06) was of the opinion that a negative vibration of the nerve is produced in an eel or a pike when the water in its vicinity is .set in vibration by means of violin strings. Parker (’08) pointed out that certain fishes react reflexly to vibrations produced by a tuning fork of a frequency of 128 vibration.s per second. This is a frequency appreciable by the human ear. The response to this type of stimulation is a sudden little jump produced by the action of the tail and pectoral fins. It wa.s not interfered with by sectioning the lateral-line nerves. Hence, it wa.s concluded that the appreciation of the vibration is due to the stimulation of the acoustic and not the lateral-line nerves. For a full discussion of this matter consult Parker and can Heusen (’17) and Parker (’18) -As yet, these observers have not ascertained bj' which fibers of the acoustic nerve these stimuli were carried. However, it is known that the tail musculature is p.articularly under the influence of bulbar centers through Mauthner’s cell the lateral dendrite of which is in intimate relation with fibers from the sacculus {Bcccari, ’07 : sec also page 456). It is not impossible, therefore, that the primitive stimuli producing this acoustic reflex arise in the sacculus {Barlehnez ’15). This is the more probable since many physiologists (and particularly Henson, ’07) regard the sacculus even in mammals as concerned in the perception of noise. Moreover, dc Bnrlel (’24) thought that the sacculus, like the cochlea is surrounded by a free perilymphatic space. The papilla lagenae dilTerentiatos from the macula sacculi, and the papilla ba.silaris cochleae of higher vertebrates (amphibians), as will be seen presently, is developed from, or in connection with the papilla lagenae. It is probable, then, that the lagena fibers of fishes are concerned also in the appreciation of these vibrations. The experiments of Parker (’05, and elsewhere) and Parker and van Hcitsen (’17) indicate that the end-organs of the lateral-line system are capable of appreciating vibrations of a slow rate — about six per second. Such vibrations may be produced by the reflection of the water from stable objects and by movements of the fish in swimming. In connection with this discussion of hearing, it is to be noted that only in higher amphibians, where the lateral-line organs have disappeared and the papilla basilaris cochleae has developed, is there clear evidence of an appre, ciation of sounds. (For earlier work on hearing in fishes see Kreidl, ’95, ’96)_

The Lateral-line and Acoustic Systems of Amphibians

The degree of development of both the lateral-line and acoustic systems varies in amphibians depending upon whether tailed or tailless forms are under consideration. The present knowledge of the peripheral and central relations of these systems in the former animals is based upon the researches of Osborn ('§g. Strong (’95), Kingsbury (’95), Coghill (’02), Beccari (’08), Norris (’ll and ’I 3 j’ Herrick (’14, ’30), Rothig (’27), and others, while the tailless forms have studied particularly by Holmes (’03), Retzius (’81), Deganello (’06), WalkiH,:^^

460 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

(’07), Arims Kappers and Hammer (’18), Rothig (’27), and many others. Both types are provided with lateral-line nerves during the larval stages. The tailless forms, such as Anura, lose them during metamorphosis. Urodeles retain the lateral-line nerves much longer, but use them only as long as the animals live in the water. The system becomes functional periodically in these forms. After spawning, the animals crawl out on land and the organs sink below the surface of the skin, but return to the surface the following year. This occurs periodically. In animals of this type (see fig. 196 of Molge) a crista cerebellaris occurs, which is lost TOth the disappearance of the lateral-line nerves.

Differences in the peripheral nerves of the lateral line and vestibular systems are present, depending upon the habits of the animal. In general, animals possess lateral-line nerves as long as they are aquatic and have gills. Thus the

Fig. 196. The course of the root fibers of the lateral-line nerves under the crista cerebellaris of the medulla oblongata of Molge cristata, at the level of entrance of the fibers from the lagena and papilla basilaris.

Tracts a and b (Kingsbury) appear to consist of association fibers. (Compare fig. 199 A). Schepman.

tailed amphibians, the perennibranchiates (such as Axolotl) which breathe through gills during their whole life, also have lateral-line canals throughout their lifetime. The caducibranchiates (such as the salamander and the triton) have lateral-line organs only during the larval stage. Tailless amphibians (such as the frog) have gills for some time ; in harmony with this, they have lateralline organs during their aquatic life but lose them later. Thus in Necturus maculosus, Charipper (’28) found the lateral-line sense organs situated in a groove within the epidermis, sufficiently deep so that the free ends of the organ did not reach quite to the level of the surface of the skin. Similar relations were described in Triturus viridescens by Chezar (’.30), but this latter observer found the organs slightly deeper in Cryptobranchus, while in Siren the groove is hard to see, but lack of pigment in the underlying basement membrane makes the organs easily discernible. Both of the above mentioned observers described four types of cells in these sense organs : (1) sense cells ; (2) mangle (protective) cells ; (3) sustentacular (supporting) cells ; and (4) basal cells. For the c 3 ixDlogic characteristics of these cells, reference is made to the original papers. Both

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

461

Charipper and Chezar described the ultimate terminations of lateral-line fibers as intracellular with reference to the sense cells. They traced fine branches of the lateral-line nerves into the sense organs, which branches ran as delicate unmyelinated fibers between the sustentacular and basal cells, gi\dng off short and slender branches (fibrillae) which broke up in somewhat elaborate baskets or cups within the basal ends of the sense cells and formed a cover over the proximal end of the nucleus. They advocated, then, the intraprotoplasmic terminations of the ultimate nerve fibrillae.

The interest in the amphibian brain centers particularly about the development of the acoustic nerve in these forms. Here differences occur depending upon differences in the peripheral relations. In both types of amphibians the acoustic is divided into two branches, a ramus anterior and a ramus posterior. In both, the ramus anterior provides the ampullae externa and anterior and the utriculus with fibers, while the ramus posterior comes from the ampulla posterior, the papilla neglecta — absent, according to de Burlel (’29) in all the amphibians except the Gymnopheona — thepapilla lagenae, the papilla basilaris cochleae, and the papilla amphibiorum. In the tailed amphibians (that in other particulars approach most nearly the plagiostomes) the branch to the sacculus runs, as in selachians, in the ramus posterior {Retzius, ’81).

In tailless amphibians, such as the frog, this branch is present in the ramus anterior. In Ichthyophis glutinosus, the bundles to the macula sacculi are given off as a series of little nerve branches from the ramus posterior (de Burlel, ’29). In Amphiuma an intermediate condition is found, the sacculus fibers running in both ramus anterior and ramus posterior {Retzius). The beginning of a cochlea, the pars basilaris cochleae, develops first in amphibians. It appears in relation to the lagena and receives its innervation from the ramulus lagenae. The ramulus lagenae, in all amphibians in which it has been seen, runs in the posterior branch of the acoustic nerve. In those tailed amphibians in which the papilla basilaris cochleae is absent, probably the ramulus lagenae (nervus cochleae of Schepman) is lacking also ; at all events Herrick (’30) was unable to find it in his preparations of Necturus, although Schepman (’18) found it as the most dorsocaudal component of the posterior acoustic root in Molge cristata. The primordial cochlea is very small, and even absent in some tailed amphibians such as Proteus, Menobranchus (Necturus), and Amphiuma. In general, it is larger in the tailless (fig. 197) than in the tailed forms. One might predict, then, that in the former the auditory

d.endoJ.

sacc.

Fig. 197. Labyrinth of the frog. Retzius. Note in contrast to the condition in Acanthias (fig. 189) the presence of a pars basilaris cochleae.

462 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

function is better developed. It must be remembered, however, that the number of fibers which come from the papilla basilaris cochleae even in frog-like forms represents only a very small part of the posterior ramus. The major number of the fibers supplies the ampulla posterior, the macula neglecta, and papilla lagenae ; and this ramus is considered to be mainly vestibular in function.

There has been no intention of giving an account of the structure of the ear beyond mention of the general relation of its parts and their relations to its nerve supply. However, there is one part of its structure, whicli owing to its size and close relation to brain structures in the amphibians, deserves brief mention. This structure is the saccus endolymphaticus. For especially interested readers, reference is here made to the account of Dem'pster (’30), which includes a full review of the pertinent literature, and the brief r6sum6 here given is based on this account.

In Necturus and Cryptobranchus this observer found that the endolymphatic sac is small, is flattened somewhat in a vertical direction, and is outside of the dura above the choroid plexus of the fourth ventricle. In Amblystoma and Salamandridae, he described the sac as greatly dilated and as overlying the dorsolateral brain surface in the region. The sacs of the two sides were found to meet but not to fuse in the midline in the latter amphibians. In Salamandridae the sac is somewhat irregular, although not as lobulated as in Amblystoma. According to Dempster, the sac reaches its fullest development in Amblystoma at the time of metamorphosis, but in Salamandra it continues to expand after the adult stage is reached. Among plethodonts (Typhlotriton, Spelerpes, and Desmognathus) are found relatively small vesicular types of endolymphatic sacs. In others, such as Plethodon cinereus, he found a large sac, resembling those in the Salamandridae. It had been known for a long time that the sacs are very large in Rana {Blasius, 1681), although their connection with the ear was not well understood. In Rana {Gaupp, ’94 ; Dempster, ’30) the duct expands in passing through the endolymphatic foramen, into an enormous sac which lies within the extradural space and extends both rostrally and caudally. For details of relations, the original paper should be consulted. In general, one portion passes forward over the midbrain and fuses witli its fellow of the opposite side ; another portion extends around the brain as far forward as the h 3 T)ophysis, and extends at the sides but does not reach the dorsal midline. The caudally extending portion has one expansion over the upper part of the choroid plexus of the fourth ventricle and another which reaches along the sides of the plexus. This latter, the processus spinalis, meets the corresponding portion of the other side and the two unite to form the pars spinalis. According to Gaupp, beyond the seventh vertebra the process may become paired. From the pars spinalis, small sacs pass out through the vertical foramina and overlie the spinal ganglia. Descriptions of the conditions in other Ranidae and a review of the literature are to be found in the contribution of Dempster (’30). Fineman (’15), Whiteside (’22), and Michl (’25) have studied the development of the endolymphatic duct and sac.

The following account of the relations of the lateral-line and acoustic nerves in tailed amphibians is based particularly on work of Herrick (’30 ; sec fig. 198).

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

463

The anterior lateral-line nerve is composed of fibers from the supraorbital, infraorbital, and mandibular trunk, which enter the brain as dorsal (bi of Kingsbury, ’95), ventral, and medial (b 2 of Kingsbury) facial roots. The posterior lateral-fine nerves enter over the roots of the vagus (or, according to certain observers, the glossopharyngeal) nerve. The relative positions of the rootlets of these anterior and posterior lateral-fine nerves vary in different urodeles, but these variations are of no particular functional significance. In Necturus the acoustico-lateral area is a ridge projecting into the ventricle and forming its lateral wall for the greater part of its extent. The gray of the area is situated immediately under the ventricular floor, with the white of the area lateral and dorso Dors. ifiland of KioEsbury Area acoustica

Fio. 198. Transverse section through the medulla oblongata of Necturus at the level of entrance of the Vllth and Vlllth nerves. Huber and Crosby.

lateral to it. There does not appear to be any marked differentiation within this gray. In Amblystoma, Herrick found that cell bodies of neurons situated in the periventricular gray send out dendrites into practically all parts of the overlying stratum album. The majority of such dendrites were confined to the acoustico-lateral field, but some of them could be traced into adjoining areas such as the region of the descending root of the trigeminal and fasciculus sofitarius. Similar conditions were found in Necturus, though the spread of the dendrites is somewhat less than in larval Amblystoma. However, Herrick found that most of the neurons send dendritic branches to the region of more than one of the incoming lateral-line roots, and since such roots themselves are made up peripherally of fibers from several lateral-line rami and interchange fibers among themselves centrally, this observer reached the very evident conclusion regarding the forms studied that ; “Any physiological specificity that may be present must rest in large measure on some other feature than localization in space of the central receptive apparatus.”

Neuraxes of cells in the area acustico-lateralis enter the general bulbar lemniscus {Herrick, ’30), which has been termed the medial lemniscus {Papez, ’29), the lateral lemniscus {Ariens Kappers. ’20, and Rbthig, ’27), and is com

464 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

parable, probably, to the lemniscus bulbaris of myxinoids {Jansen, ’30) and to the fasciculus longitudinalis lateralis described for various fishes. This bulbar lemniscus, by whatever term it may be designated, carries not only impulses from the acoustico-lateral area, but also from other sensory areas of the medulla oblongata. Herrick did not consider it the anlage of the lateral lemniscus, but believed that this latter system is represented in another tract, the tractus bulbotectalis lateralis, or primordial lemniscus, which through its connection with the inferior colliculus, is to be regarded as the true lateral lemniscus, and that its presence in Necturus, where papilla basilaris, cochlear root, and cochlear centers are lacking, is explainable either by regarding Necturus, as Norris (’ll) has done, as an arrested larval stage of some more highly developed but extinct ancestral form, or less probably, by conceding that the secondary connections have preceded the primary connections in their development. As yet, the cells of origin for this tractus bulbo-tectalis lateralis do not appear to have been identified, and until their position and relations have been established, in the opinion of the present writers {Huber and Crosby), the interpretation of this tract as a lateral lemniscus system can only be accepted provisionally. In weighing the evidence for the acceptance of this system of fibers rather than the bulbar lemniscus (or lateral lemniscus of many obseiw’ers) as the primordium of the lateral lemniscus of higher forms, it must be remembered that this last mentioned tract connects wth the optic tectum as well as with the inferior colliculus in reptiles and probably mammals.

The central relations in amphibians, where the primordial cochlear apparatus is relatively well developed, will be considered now. The account is based primarily on the frog, and particularly the bullfrog {Ariens Kappers and Hammer, ’18 ; Roihig, ’27a), as well as preparations from the Amsterdam collection studied by Ariens Kappers. In ^this amphibian there are two acoustic roots and two separate ganglia, the ganglion acusticum anterior and the ganglion acusticum posterior. The fibers of the two roots are fairly distinctly separated at their entrance to the medulla oblongata, although they enter at about the same level. The fibers entering more dorsally correspond to the posterior ramus and hence carry impulses from the ampulla posterior, the macula neglecta, the lagena, and the papilla basilaris cochleae (fig. 199A). The ventral root, which corresponds to the anterior ramus, is purely vestibular in character. As in fishes, it enters the medulla oblongata somewhat farther ventrally than does the posterior ramus. The central terminations of the two roots are different. The fibers of the ventral root, which end ventral to the nucleus dorso-magnocellularis, terminate in a variety of ways. On entering, many of them divide dichotomously and some of their branches terminate in the ventral nucleus of the Vlllth. This nucleus is, in a general way, homologous to the Deiters’ nucleus of higher animals. In some of the lower tailed amphibians it is more primitive in character {Beccari, ’07, and Herrick, ’14), for there a Mauthner’s cell is present, the neuraxis of which runs caudalward in the usual manner beside the fasciculus longitudinalis medialis. As in fishes, this neuraxis provides a final common path for impulses giving tail reflexes. The cell of Mauthner is not found in tailless amphibians.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

465

Other dividing fibers of the ventral root form longer ascending and descending branches. The former run to the vicinity of the cerebellum, the latter have been traced by Wallenberg (’07) to the sixth spinal segment. Deganello (’06) has

Fig. 199. A. The entrance of the dorsal and ventral roots of the acoustic nerve in Rana mugiens. The dorsal and ventral nuclei of the acoustic are seen. The crista cerebellaris has disappeared (compare with fig. 196).

B. Rana catesbiana. Cross section through nuclei of VIII and through caudal end of VI and accessory VI nuclei. Weigert-Pal-paracarmine. X40. Addcns, ’33.

observed degenerations in the ventral funiculi of the spinal cord after injuries of the labyrinth. Other secondary tracts have been described on the basis of degenerations, but the results are not always conclusive. There seems to be a

466 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

general agreement among workers that crossed fibers as well as homolateral ones are to be found among the ascending and descending tracts and in the branches to the ventral nucleus of the Vlllth. These fibers decussate as part of the fibrae arcuatae dorsales.

The dorsal root ends mainly around a large-celled nucleus in the dorsolateral part of the area statica. This nucleus magnocellularis dorsalis (fig. 199A, B) lies in a region corresponding to the position of the medial nucleus of fishes, which latter nucleus, it will be remembered, received some vestibular bundles but was particularly an end station for lateral-line fibers. In the frog, the nucleus occupies the most dorsal part of the medulla oblongata, the nucleus dorsalis of fishes disappearing in amphibians with the disappearance of the anterior lateral-line nerve. The disappearance of posterior lateral-line fibers as well may be regarded as leaving the cells of the medial nucleus under the influence of the acoustic nerve, in which case this nuclear mass is believed by certain observers to develop as a cochlear center {Ariens Kappers, ’20). In fishes this repon (that is, the medial nucleus) receives also vestibular fibers. Whether or not such fibers are present in the nucleus magnocellularis of the frog is not known with certainty. However, it is rather doubtful, particularly since certain of the dorsal fibers appear to end in the region below the magnocellular nucleus in the nucleus of the ve.stibular VIII. There is not unanimity of opinion as to the existence of a direct genetic relation between the lateral-line centers of lower forms and the cochlear centers as found in higher amphibians. Thus Herrick (’30) questioned such a genetic relation and apparently regarded the cochlear centers as new structures in amphibians.

The secondary connections of the acoustic centers arc not thoroughly known for the frog. From the ventral nucleus and from the gray associated with the ascending and descending roots, fibers ari.se which decussate to the opposite side of the medulla oblongata and then accompany the ascending and descending roots. Of these, the coarse, somewhat more isolated fibers found with ascending, and particularly with descending bundles, from the ventral nucleus, deserve special mention. They pass to the motor centers of midbrain, medulla oblongata, and cord, and are comparable to the coarse reflex fibers previously described for fishes. They are the tractus octavo-motorius cruciatus, shown in Marchi preparations of Rana by Rolhig (’27 ; see also fig. 199).

From the dorsal part of the medulla oblongata, in all probability from the nucleus dorsalis magnocellularis itself, there arises in the frog'a lateral lemniscus system. This arises from two bundles, a dorsal one which passes directly medialward from its cells of origin near the ventricular floor in order to cross the midline and enter the ascending tract, and a ventral bundle, described by Holmes (’03) as the trapezoid body, which runs near the .surface of the medulla oblongata and contributes fibers to the superior olivary nucleus during its course. After decu-ssation it joins the crossed dorsal bundle (tr. bulbo-mes. (lat. lem.) in fig. 199A and fig. 199B), and together they pass to the torus semicircularis or nucleas posterior tecti and to the optic tectum. It also contributes fibers to the nueleas isthmi {Larsell, ’23). This tract, together with the tractus isthmo

THE LATERAIr-LINE AND ACOUSTIC SYSTEMS

467

bulbaris and tractus bulbo-isthmicus, has been shown in Marchi preparations of Rana by Rothig (’27), who termed it the lateral lemniscus or the tractus bulbotectalis or the bulbar lemniscus. With the presence of a pars basilaris cochleae in the frog, supplied by fibers of the dorsal root of the acoustic nerve, and the association of such fibers with the cells of origin (within the nucleus magnocellularis dorsalis) of the ascending tract just described, the recognition of this tract as carrying true lateral lemniscus fibers in certain amphibians is justified. A new nucleus, associated with the development of cochlear fibers, likewise appears in amphibians. This is the superior olive. It appears in the course of the lateral lemniscus of the frog at the cross sectional level of the nucleus magnocellularis, midway between the points of emergence of the facial and glossopharyngeal nerves. It consists of small cells, which correspond in type and position to those of the superior olive of higher animals. It has not been clearly demonstrated as yet that collaterals from the lateral lemniscus reach this nuclear mass. Schepman (’18) has indicated that the nucleus has a dorsal extension similar to that described for various reptiles {Ariens Kappers, ’20 ; Huber and Crosby, ’26 ; see bibliography for reptiles).

The nuclei posteriores tecti of the two sides are fused, in certain amphibians, and a part of the ventricle is shut off looking toward the formation of a true aqueduct of Sylvius. These nuclei are separated at their cephalic ends, a reminder of the more primitive condition. They are covered entirely in amphibians by the superior collicular portions of the tectum so that from the surface only two swellings are seen and these are spoken of as the corpora bigemina. It is only in certain reptiles and more clearly in mammals that corpora quadrigemina are seen in surface view. Their appearance, it will be evident, is associated with a relative increase in size of the inferior colliculi, with the elaboration of the cochlea and the lateral lemniscus and with a relative diminution of the superior colliculus as the optic thalamic center (the lateral geniculate of higher forms) becomes more important. The relation between the tori semicirculares of fishes, the nuclei posteriores tecti of amphibians, and the inferior colliculi of mammals is discussed on page 944.

The Vestibular and Cochlear Systems of Reptiles

The lab 3 Tinth of reptiles (fig. 200), with its innervation, has been studied most carefully by Relzius (’84). Two rami are present in the acoustic nerve of reptiles, a ramus anterior and a ramus posterior. In the ophidians and the saurians the ramus anterior innervates the utriculus and the ampullae anterior and externa, while the ramus posterior receives its fibers from the macula sacculi, the ampulla posterior, the papilla or macula neglecta, the macula lagenae, and the papilla basilaris cochleae. In the hydrosaurians, at least in the crocodiles, the macula sacculi is innervated in part by the ramus posterior and in part by a branch running independently through a foramen acusticum medium {de Burlet, ’29), and in Cheloniae or turtles, the macula sacculi has a similar innervation. In Iguana the macula sacculi is supplied through the ramus posterior ; in the chameleon it is supplied by an independent branch (de Burlet, ’29). The greater

468 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ean.anl. —

....can.po5{.

.can. ext.

development of the cochlea and its associated fibers permit the connections of the posterior ramus to be regarded to a considerable extent as cochlear in nature. Moreover, such vestibular fibers as are present in this ramus tend to separate from it and to distribute ventral to the distribution of the cochlear fibers. The ramus posterior becomes the dorsal root, the ramus anterior the ventral root. Particularly in crocodiles the two acoustic branches enter the brain at some distance from each other. The dorsal root is somewhat farther caudal than the ventral root. Two types of fibers, coarse and fine, are distinguishable in the

ventral root in many reptiles {Beccari, ’12).

The reptilian vestibular nuclei have not been thoroughly analyzed as yet.® The following nuclei receive vestibular root fibers : (1) a tangential nucleus, similar to that described for teleosts, which lies at the entrance of the vestibular root fibers, a little ventral and slightly caudal to the ventrolateral vestibular nucleus ; (2) the ventrolateral vestibular nucleus, often called Deiters’ nucleus, rather generally recognized by observers and consisting of large, multipolar elements with certain intermingled smaller neurons which may be subdivided into a central (Deiters’ nucleus of mammals), a dorsal, and a ventrocaudal group according to Beccari (’12 ; see also Aricns

p.bas.cochl.

Fig. 200. Labyrinth of the alligator. Rclzius. (Note the greater size of pars basilaris cochlcaris as compared with that of the frog, fig. 197).

Kappers, ’20); (3) a descending or inferior vestibular nucleus {Beccari, ’12; Weston, ’33), a caudal continuation of the smaller neurons of the last described vestibular nucleus ; (4) a nucleus vestibularis superior {Beccari, ’12, Ariens Kappers, ’21, Larsell, ’26, Hindenach, ’31) or anterior {van Hocvcll, ’16) which extends forward and dorsalward between the ventrolateral vestibular nucleus and the cerebellar nuclei ; (5 and 6) two nuclear groups described by Weston (’33), a chiefly small-celled ventromedial vestibular nucleus, ventral and medial to the ventrolateral vestibular nucleus, and a specialized superior group in turtles.

Certain fibers (finer in Lacerta muralLs, Beccari, ’12) from the ampullae, on entrance to the medulla oblongata, terminate in part in typical, pericellular,

' A compreliensivc ficrics of investigation.s on the vestibular centers and the cercliclluin in progress at the Tjiboratory of Comparative Neurology of the University of Michigan will present a relatively complete analysis of this region in reptiles (IFcsfon, ’33).

THE LATERAIr-LINE AND ACOUSTIC SYSTEMS

469

spoon-shaped endings around the cell bodies of the neurons of the nucleus tangentialis (fig. 479). Occasionally a coarse fiber forks, and one branch enters the ascending vestibular root while the other synapses with the spindle and stellate neurons of this nucleus in the typical manner (see Beccari, ’12). Occasionally both branches terminate in the nucleus. This tangential nucleus receives fibers from the ventral vestibular root as well as from the vestibular portion of the dorsal root. Beccan (’12) has subdivided the tangential nucleus into foim secondary groups, best seen in horizontal planes. He found the anterior and medial groups related particularly to the ventral vestibular root; the posterior and lateral groups, to the dorsal vestibular root fibers. The neuraxes of these cells, as in teleosts and cyclostomes, decussate and ascend and descend in the medial longitudinal fasciculus. From the lateral and posterior groups, fibers appear to join the tractus vestibulo-spinalis, although as yet not traced for their entire course {Beccari, ’12).

The finer vestibular fibers originate from the sacculus, the utriculus, and the ampullae. On entrance to the medulla oblongata they divide dichotomously into ascending and descending bundles. The ascend itio- THITI fnTOrarrl fn 201. The entrance of the ventral acoustic root (vestibular)

mg noers run lorwaru w m alligator de Lange

various parts of the cerebellum. During this course, the fibers lie in relation with the scattered cells in front of the level of entrance of the nerve, and in this relation extend forward and dorsalward to the anterior vestibular nucleus of various observers or the superior vestibular nucleus of Beccari (’12), which is probably the homologue, in part at least, of the mammahan nucleus of that name, the so-called nucleus of Bechterew (which lies incorporated in the ascending vestibular root near the margin of the ventricle and medial to the corpus restiforme) but more particularly the forerunner of the avian dorsolateral nuclear group. The descending vestibular fibers extend to the beginning of the spinal cord and possibly even traverse it for a shorter or longer distance. Certain of these fiber bxmdles, soon after entrance, terminate in the ventrolateral nucleus

470 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

(fig. 479 ; n. vest. VIII, fig. 202), which, in saurians and hydrosaurians, lies in front of, and at the level of, the facial root. This nucleus is well developed but does not extend so far caudally as in mammals. In general, it consists of large, polygonal cells. According to Beccari (’12), these are arranged in three groups, a central, a dorsal, and a posterior group. However, in part at least, this grouping appears to be due to a purely secondary arrangement occasioned by the passage of root fibers. The central group is connected chiefly with the ventral vestibular root, tbe dorsal group with the dorsal vestibular root. The latter group lies near the dorsal cochlear nucleus and is said to receive collaterals

Fio. 202. Transverse section Itirough the cnlrnnco of llic Vllltli nerve sliowing two of the main cochlear nuclei in Alligator mississippicnsis. Chrom-silver preparation, //iitcrand Crosby (’2G). The crossed and uncrossed dorsal secondary cochlear fibers; (st.vicd.ac.) arc shown in the drawing and something of their relation with the triangular-shaped portion of the sviperior olive. Secondary cochlear fibers running lateralward along the surface of the medulla, particularly to the upper portion of superior olive arc to be seen on the right of the figure. Cajal preparation. XI5. dcc.scc.VIIIf., decussating secondary Vlllth fibers; f.l.m., fa.sciculus longitudinalis mcdialis; Icm.lnt., lemniscus lateralis; N.cnrii. VIII, cochlear component of the Vlllth nerve; n.dors.mngnoccll., nucleus dorsalis magnoccllularis ; n.lam., nucleus laminnris; n.sp.V.fr. fr., spinal Vih nucleus and tract; n.vcsl.VIII, nucleus vc.stibulnris of Vlfith nerve; N.irst. VIII, vestibular Vlllth nerve; N.VI, Vlth nerve; src.rorh.f., secondary cochlear fibers; sl.mcd.ar., dorsal secondary cochlear fibers; aup.ol., suix’rior olive; tr.siic<T., tractus spino-cercbcllaris; trjip.mcs., tractus spino-me.scnccphalicus; lr.thid.b>dh,, tractus thalamo-bulbaris.

from cochlear fibers {Arims Kappers, ’20). From both central and dorsal groups, partly uncrossed but mainly crossed fibers enter the medial longitudinal fasciculus. Tlie homolatcral tractus vcstibulo-spinalis arises from the ventrolateral vestibular nucleus (the posterior group in Lacorta according to Bcccari, ’12) and lies in close relation with the descending vestibular nucleus in many reptiles, in relation with which it probably terminates. This latter nucleus (fig. 479) receives spino-vestibular fibers (IVc-sfon, ’33). Connections of the ventrolateral nucleus with the cerebellum (largely cerebello-vcstibular) have Ix'cn described by several observers, including A nens A'appers (’21), Iluhcr and Croshij (’20), and Shartidin (’30).

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

471

The nucleus vestibularis superior (see fig. 479) lies in the course of ascending vestibular fibers, certain of which pass directly to the cerebellum (Beccari, ’12, Huber and Crosby, ’26, Larsell, ’33, Wesion, Dissertation, ’33). It has connections with such fibers and other ascending tracts to the cerebellum, and is connected to the cerebellum itself, probably by vestibulo-cerebellar and cerebellovestibular fibers, and sends bundles (probably ascending, Weston, ’33) to the medial longitudinal fasciculus.

The dorsal branch of the acoustic nerve in reptiles carries the cochlear fibers as well as those from the ampulla posterior and bundles from the macula

Fia. 203. Nucleus angularis and nucleus lammans in the crocodile. Schepman.

sacculi. From these latter areas, the vestibular fibers also arise, but such fibers have the same central relations as do those represented in the ventral root. The cochlear fibers show different central relations. These relations have been studied in a number of reptilian forms by different observers {Holmes, ’03 ; Beccari, ’12; de Lange, ’17; Schepman, ’18; Ariens Kappers, ’20; Huber and Crosby, ’26). Two nuclei — the nucleus dorsalis magnocellularis {Schepman, ’18; Ariens Kappers, ’20; Huber and Crosby, ’26) or the nucleus cochlearis anterior, Beccari (’12), and the nucleus angularis or the nucleus cochlearis dorsalis or superior {Holmes, ’03 ; Beccari, ’12) — are clearly associated with cochlear bundles (figs. 202, 203). Nucleus magnocellularis lies entirely dorsal. It consists of fairly large, goblet-shaped cells and is comparable to the similarly named nucleus of amphibians. It is found at a level midway between the entrance of the facial and glossopharyngeal roots. Nucleus angularis lies somewhat farther forward. It is situated in the dorsal part of the medulla oblongata along its periphery (fig. 203). This nucleus, which has been described by many observers {Holmes, ’03; Beccari, ’12; Schepman, ’18; Ariens Kappers, ’20; Huber and Crosby, ’26; Shanklin, ’30, and others), is the homologue of the nucleus angularis of birds, to which it corresponds in relative position and in fiber relations. In close relation with the nucleus magnocellularis is a peculiarly shaped row of cells termed nucleus laminaris (see figs. 202 and 203) which, from

472 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

its topographic relations and general appearance, in all probability is the homologue of the avian nucleus of that name. Nucleus laminaris {Holmes, ’03; de Lange, ’17; Ariens Kappers, ’20; Huber and Crosby, ’26; Shanklin, ’30, and others) is present in many reptiles. Holmes (’03) believed that he could trace acoustic fibers into the nucleus in many reptiles, while de Lange (’17) denied their presence in Varanus. Ariens Kappers (’20) stated that it does not receive fibers of the dorsal root of the cochlear nerve, but that it is in intimate relation with nucleus magnocellularis, and that the basis of its homology with the avian nucleus of the same name is that it is an end station for crossed and uncrossed fibers of the nucleus dorsalis magnocellularis. The nucleus appears to contribute ventral arcuate fibers to the lateral lemniscus. A different interpretation of the nucleus laminaris is found in Shanklin’ s (’30) account of the chameleon, for this observer stated that he carried to the nucleus fibers of the vestibular or ventral root of the acoustic nerve, and gave as its secondary connections bundles to the medial longitudinal fasciculus and cerebellum, to the region of the superior olivary nucleus, and to the contralateral nucleus laminaris. The connection with the cerebellum had been commented on earlier by Holmes (’03). It is probably the tractus cochleo-vestibulo-cerebellaris of Ariens Kappers (’20) and Huber and Crosby (’26) from the nucleus ventrolateralis and nucleus laminaris (fig. 203).

A superior olivary nucleus is present in reptiles {Schepman, ’18; Ariens Kappers, ’20 ; Huber and Crosby, ’26 ; Shanklin, ’30, and others). In mammals the superior olive lies in the ventrolateral part of the tegmental region of the pons, ventromedial to the motor nucleus of the facial nerve. In Anura it has a distinct ventral extension. In reptiles, and particularly in the saurians and hydrosaurians, the superior olive has shifted caudalward, and a caudal and more dorsal portion and a ventral and more cephalic portion may be recognized. Accompanying illustrations (fig. 201 ; see also figs. 260 and 262) indicate that the small celled mass of the superior olive is found, in part, in a very dorsal position, being actually embedded m the course of the arcuate fibers arising from the nucleus laminaris. Another part of the olive (fig. 203) appears farther ventral, in the course of the lateral lemniscus. The parts just mentioned are continuous with each other. This condition is found in the bird as well as in the alligator and points to the double origin of the superior olive {Ariens Kappers, ’20). It also suggests that the superior olive, like the inferior olive and the pontine gray, arises from the more dorsal part of the medulla oblongata, that is, from the somatic sensory area, and that its more ventral position later is due to the neurobiotactic influence of secondary acoustic fibers such as are represented in the bundles of the lateral lemniscus. Scattered along the course of the lateral lemniscus are detached masses of gray of a type which warrants regarding them as homologous with the mammalian ventral nucleus of the lateral lemniscus (p. 498).

We may now consider the secondary ascending paths, which have received the attention of numerous observers {Holmes, ’03 ; Beccari, ’12; de Lange, ’17 ; Joustra, ’18; Schepman, ’18; Ariens Kappers, ’20; Huber and Crosby, ’26; Shanklin, ’30, and others), and thus need only brief review. From the nucleus magnocellularis and the nucleus laminaris (figs. 201 and 203) arise secondary

THE LATERAL-LINE AND ACOUSTIC SYSTEMS 473

arcuate fibers which, partly crossed and partlj"^ uncrossed, swing into relation with the lateral lemniscus. Those from the nucleus magnocellularis pass ventralward, after a partial decussation under the medial longitudinal fascicuU, and reach a position dorsal to the ventralmost portion of the superior olive. They may send collaterals to this nucleus, but the majority of the fibers swing lateralward to enter the lateral lemniscus. The fibers from nucleus laminaris turn ventralward on the same side (at least for the most part) and come into relation with the more dorsal portion of the superior olive. Here some of them synapse ; others pass through and join the more lateral fibers which run at the side of the nucleus. The bundle then enters the lateral lemniscus. Whether or not fibers arising from the superior olive join this lemniscus is uncertain. From the nucleus angularis, fibers pass ventralward along the periphery of the medulla oblongata and come into relation with the more ventral portion of the olive of the same side, and, after decussation, with that of the other side. Here some of the fibers synapse while others pass directly into the bundles of the lateral lemniscus. This lemniscus runs forv'ard, accompanied by other ascending systems, and after some synapse in the nucleus of the lateral lemniscus, terminates in the nucleus isthmi and the torus semicircularis. From the nucleus isthmi impulses are passed to the optic tectum, the torus semicircularis, and the contralateral nucleus isthmi. Both the nucleus isthmi and the torus semicircularis are highly developed in reptiles. An account of these nuclei, of their connections and possible homologies, is to be found in Chapter VIII, page 982.

The above account indicates that the cochlear system is well developed in reptiles and shows great resemblance to the mammahan system both in its nuclear masses and its fiber connections. The nucleus magnocellularis dorsalis is to be regarded as the homologue of the ventral cochlear nucleus, according to the studies of Brandis (’94) and Ramon y Cajal (’08a, for birds ; see bibliography for birds). The nucleus angularis is probably the primitive homologue of the mammalian dorsal cochlear nucleus, or tuberculum acusticum. The nucleus laminaris has an avian, but no known mammalian equivalent. It will be remembered that the dorsal root contains not only cochlear and lagena fibers, but a variable number of vestibular fibers. The careful work of Beccari (’12), on fizards, indicates that such vestibular fibers of the dorsal root have central connections similar to those of the ventral root and are not related to the cochlear centers, 'svith the possible exception of a few vestibular fibers to nucleus laminaris.

The Vestibular and Cochlear Systems of Birds

The peripheral relations of the acoustic system in birds have been studied particularly by Retzius. Schepman (’18) traced the connections of the peripheral organs with the central nuclei, and the central connections have been very exhaustively studied by Ramon y Cajal (’08a) and by Bok (’15). The work of Retzius (’84) indicates that there is a strong resemblance between the peripheral organs in birds and those in saurians and ophidians. In birds the posterior branch of the nerve innervates the ampulla posterior, the sacculus, the macula

474 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

neglecta, including the papilla lagenae, and the papilla basilaris cochleae. This membrana basilaris cochleae has increased considerably in most birds, but the cochlea does not have the snail-shell shape characteristic of most mammals. The lagena, which lies at the tip of the cochlea, is always well developed. On rare occasions Retzius found that the sacculus (as in hydrosaurians) is innervated by the ramus anterior. This occurs in Columba and Turdus. It is evident then, that a certain amount of variability in the peripheral innervation is present in birds as well as in amphibians and reptiles (see fig. 204).

can. ant.

The avian vestibular nuclei and fiber connections have been studied by a number of observers. Due partly, perhaps, to differences in the birds studied (which, according to Bartels (’25), and others, appears to be considerable), and partly to differences in interpretation, great confusion exists as to the subdivisions of the vestibular areas in these forms. The terminology here employed is that found in the Sanders paper (’29), the work for which was carried out on the collection of the Laboratory of Comparative Neurology, Department of Anatomy, at the University of Michigan. The following account deals primarily with conditions in sparrows. Six vestibular nuclear groups were identified : (1) nucleus vestibularis tangentialis, homologous with the nucleus of the same name described by Ramon y Cajal (’08a) and by Ariens Kappers (’20), as given in the earlier edition of this text, by Bartels (’25), Craigie (’28), and others; (2) nucleus vestibularis ventrolateralis, the Deiters’ nucleus of Wallenberg (’98, ’00), Ramdn y Cajal (’08a), and Craigie (’28) and the ventral nucleus of Bartels (’25) ; probably the mammalian lateral vestibular nucleus as a whole or in part ; (3) nucleus vestibularis descendens (inferioris), rather generally recognized by students of avian brains; (4) nucleus vestibularis dorsomedialis or nucleus triangularis, identified by Wallenberg (’00), Holmes (’03), Ramdn y Cajal (’08a), Bartels (’25),

THE LATERALr-LINE AND ACOUSTIC SYSTEMS

475

Craigie (’28), and others; (5) nucleus vestibularis dorsolateralis, Bartel’s dorsal nucleus of Deiters, which is redivided into a medial division (medial portion of “noyau jumeaux” of Ramon y Cajal, the nucleus gemelli of Craigie and part of this latter nucleus, according to Bartels), an intermediate division (homologous with the lateral part of the “noyau jumeaux” or Ramdn y Cajal’ s nucleus gemelli of Bartels), and a lateral division, nucleus quadrangularis (nucleus quadrangularis and nucleus vestibulo-cerebellaris of Ramdn y Cajal and Craigie) ; (6) nucleus vestibularis superior (nucleus oralis of Bartels and Craigie).

Fio 20S. The vestibular nuclei of a bird, Passer domesticus Ramdn y Cajal

Nucleus vestibularis tangentialis, homologous with the nucleus of the same name described by Ramdn y Cajal (’08a), Anens Kappers (’20-’21, in the earlier edition of the present text), Bartels (’25), Craigie (’28), and others, consists of small cells lying in the course of the incoming root fibers. Nucleus vestibularis ventrolateralis is composed of large, multipolar neurons. It is medial to the nucleus vestibularis tangentialis, ventral to the direct or crossed vestibular root fibers, and in the course of the ventral vestibular fibers. Secondary subdivisions of this nuclear group have been made by Bartels (’25). The nucleus vestibularis descendens or inferior is a small celled nuclear group caudal to the other vestibular nuclei. Dorsal to the dorsal arcuate system and medial to the cephalic portion of the nucleus magnocellularis is the nucleus vestibularis dorsomedialis or triangularis, the claims of which, as a vestibular center, have been disputed by Ramdn y Cajal (’08a), although other observers (Wallenberg, Bartels, Craigie, and Sanders) believed that vestibular fibers can be traced to it (figs. 205 to 207).

The three divisions of the nucleus vestibularis dorsolateralis are based on a study of the cytoarchitecture of the region, together with its fiber connections. Much of the previous work has been based on fiber pattern alone, with the result that a single nucleus has been designated by different names and two or more nuclear masses by the same name. The nucleus (fig. 206) lies dorsal to the posi

476 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

tion of the nucleus vestibularis ventrolateralis, occupying a position in the dorsolateral part of the medulla oblongata in the region of attachment of the cerebellum. Its medial, intermediate, and lateral divisions are dorsoventrally extending portions of the main nuclear mass. The three parts — ventral, central, and dorsal — of the medial division occupy the relative positions indicated by their names. The ventral is distinctly small celled and is quite obviously the part represented by the nucleus pyriformis of Ramdn y Cajal (fig. 205) ; the scattered cells extending ventrolaterally from it, with their accompanying fiber bundle, constitute the pedicle of nucleus pyriformis described by that observer. The central part consists of medium sized cells which approach, near the caudal

Fig. 206. Vestibular nuclear group in the sparrow. For the terminology, see the text. Toluidinblue preparation.

nucleus and extends forward from this cephalic terminations of the medial ant

end, the large size and multipolar character of the dorsal part. The central and dorsal parts, together, form a concavity in which the intermediate division of the dorsolateral vestibular nucleus rests. This intermediate division consists of a rather definitely delimited group of relatively large, somewhat angular cells, which extends caudalward to a plane through nucleus angularis. The lateral division of the dorsolateral vestibular nucleus makes its appearance in transverse planes through the ventrolateral vestibular level to a region well in front of the 1 intermediate divisions of the nucleus.

On the whole, the most conspicuous portion of this division is the nucleus quadrangularis, a compact cell mass which, in spite of its name, varies in outUne from section to section, sometimes being more or less quadrangular, sometimes crescent-shaped, and frequently irregular in outline. Its cells are less deeply stained than the cells in the two divisions of the dorsolateral vestibular nuclei previously described. Included with nucleus quadrangularis in the lateral division is a small group of cells triangular in outline, probably the “noyau vestibulo-cer6b611eux ” (fig. 205) of Ramon y Cajal, ’08a. This nucleus lies ventral to the intermediate division of the nucleus dorsolatcralis

vestibularis, in planes through the frontal end of the nucleus magnocellularis.

A superior vestibular nucleus, described as nucleus oralis by Bartels (’25) and Craigic (’28), is seen best at its greatest extent, where it has a more or le.‘=s H-outline and is situated so far dorsally that a part of the lateral cerebellar nucleus lies in the upper fork of the H. The nucleus disappears frontally at a plane through the chief sensory nucleus of the trigeminal nerve and caudally at the

level of the inferior vestibular nucleus.

Certain of the coarse fibers terminate in part, soon after their entrance into the medulla oblongata, in the nucleus vestibularis tangentialis, around the cell bodies of which they form curious spoon-shaped pericellular synapsc.s, and in the

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

477

nucleus vestibularis ventrolateralis. Direct vestibular root fibers, many of them of smaller caliber, accompanied by secondary fibers from nucleus vestibularis tangentialis, cross the midline to the vestibular region of the other side {Winkler,

area in a chicken embryo of 16 days The right side is cut farther dorsahvard than the left. Schepman.

’07, and others). Other incoming vestibular fibers swing dorsalward. One bundle passes to the lateral division of the dorsolateral vestibular nucleus, running in its course along the ventral border of the “noyau vestibulo-cer6b611eux” (or nucleus vestibulo-cerebellaris) and then dorsalward on the medial side of the nucleus quadrangularis. It contributes fibers to both of these nuclei. Direct vestibular fibers can be followed to both the medial and lateral sides of the

478 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

medial division of the nucleus vestibularis dorsolateralis. The bundles on both sides supply the medial division and send direct vestibular root fibers to the cerebellar nuclei. Direct vestibular fibers have also been traced to the intermediate division of the nucleus vestibularis dorsolateralis, to nucleus vestibularis dorsomedialis, and also to the nucleus vestibularis superior. The above account follows the distribution of the root fibers as given by Sanders (’29), but essentially similar distributions of these fibers have been found by other authors with the exception that Ramdn y Cajal could find no vestibular fibers to the dorsomedial vestibular nucleus or nucleus triangularis, although Wallenberg (’00), Bartels (’25), and Craigie (’28) believed that such fibers are present.

The connections of the superior vestibular nucleus are with the cerebellum. Secondary fibers reach the medullar centers of the opposite side by way of neuraxes of the nucleus vestibularis tangentialis, which accompany the decussating vestibular root fibers. The ventrolateral vestibular nucleus sends fibers to the spinal cord and across the midline to the contralateral fasciculus longitudinalis medialis. The neurons comprising the intermediate and medial divisions of the dorsolateral vestibular nucleus show a marked orientation, for their dendrites come into synaptic relation with the cerebello-motorius system at the one side, and the incoming vestibular fibers and the cerebello-petal path at the other side. Their neuraxes appear to enter the cerebello-motorius system. The dendrites of the nucleus quadrangularis are in synaptic relation with the spino-cerebellar tracts. Bartels (’25) placed the dorsolateral vestibular nucleus within the Deiters’ complex and the fiber connections of the group indicate certainly that it belongs to the efferent nuclear groups rather than to the afferent.

It is quite impossible to attempt to review the great accumulation of experimental studies which have resulted from operations on the semicircular canals, the vestibular nerves, or the vestibular centers of birds. The following observations are quoted because they emphasize particular anatomic relations or connections.

According to Winkler (’07), the loss of one labyrinth has no immediate effect on pigeons. He found that atony was present on the operated side but that the forced attitudes did not begin until the third day (due presumably to the effect of operative shock on the uninjured side). He emphasized that the presence of crossed fibers as well as homolateral fibers in these animals permitted the uninjured nerve to affect the efferent centers sufficiently so that automatic retention of the usual position (with lack of forced attitudes) characterizes the resting animal. In this retention of the normal position, Winkler believed the unchanged optic, motor, and kinesthetic impulses play a part. Any change of these stimuli or of central stimuli (such as desire for movement on the part of the animal) upsets the condition of equilibrium, and the destruction of the vestibular nerve (which means lesser innervation on the contralateral side) is evidenced by the appearance of forced movements, i.e. forced positions of the head. Winkler believed that such differences as appeared in birds as compared with mammals, after a onesided destruction of the vestibular nerve, are due to the greater number of crossed octavus root fibers in the former as compared with the latter animal. Muskens

THE LATERAL-LINE AND ACOUSTIC SYSTEMS 479

(’30), in his study of forced movements in birds, has laid particular emphasis upon the importance of the vestibular connections through the medial longitudinal fasciculus, with the nuclei of this fasciculus and of the posterior commissure. Forced movements (and particularly Muskens’ contributions to this important matter) are discussed at some length under the discussion of this fasciculus in mammals (Chapter VIII, p. 1081) and need not be considered further here.

Groebbels (’27, ’27a, ’28, ’28a) carried on a considerable number of experiments on birds in order to determine the functional relations of the vestibular apparatus in these forms. The following account is a very brief summary of certain of his results. For the details the original papers should be consulted. This observer believed that the crista macula sacculi plays a role, together with ampullae of the semicircular canals, in supplying impulses for certain compensatory neck reflexes. He came to these conclusions from a series of experiments in which the semicircular canals and ampullae were destroyed, first on one side and later on the other side, while the maculae remained functional. Such cases were contrasted with those obtained by complete destruction of the labyrinth portion on one side and later on the other, the animals being studied at various stages of the experiments.

Groebbels (’27) found also that the caudal root of the vestibular (said by him to degenerate after destruction of the ampullae and otoliths) terminates centrally in relation to the ventral nucleus of Deiters (usually termed the lateral or ventrolateral vestibular nucleus) and the nucleus vestibularis descendens or inferior. The oral root or anterior root terminates in the dorsal nucleus of Deiters (included at least in the nucleus vestibularis dorsolateralis) and the processus cerebelli. The most important secondary connections are with the medial longitudinal fasciculus and the cerebellum, by means of homolateral and contralateral fibers.

Groebbels (’28 and ’28a) studied also, in a nmnber of experimental contributions, the fimctions of the cerebellum and the vestibular apparatus and their interrelations. He pointed out that the two centers are intimately interrelated in function. “Vestibularis-zentren und Paleocerebellum bilden viel mehr eine imtrennbare anatomische-physiologische Einheit,” and further : “Bei der Taube gehen dies Beziehungen so weit, dasz man uberhaupt keine Labyrintheffekte erhalten kann, bei denen nicht Kleinhirneffekte beteiligt waren, und umgekehrt, man bekommt keine ICleinhimeffekte, die nicht eine Veranderung der auf die Vestibularis-zentren einwirkenden Erregung hervorrufen.” Groebbels showed that impulses from vestibular centers for turning and lifting the neck were affected reflexly by the cerebellum, which has a “checking” or inhibitory, or perhaps regulatory influence on them. Such impulses appeared to be carried over the cerebello-vestibular tract, a crossed path. Destruction of the cerebellar nucleus or restiform body on one side led to a bending of the neck to the opposite side. Destruction of the vestibular center of the same side produced the same effect. Destruction of both the lateral cerebellar nucleus and the vestibular area on the same side, after a few days, led to the turning of the head to the same side, and then to the other side. Destruction of both lateral cerebellar nuclei and both vestibular areas naturally were not followed by any twisting of the neck.

480 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Groebbels described another pathway, regulatory or “checking” in function, from the cerebellum. This runs from the lateral lobes to nuclei tecti (nuclei fastigii) and is relayed from there through the uncinate fasciculus to the vestibular centers. An opisthotonus resulted if both nuclei or the commissure connecting them were cut, but the neck was not turned backward if only one nucleus was destroyed. An opisthotonus was obtained also by a lesion of the lateral lobes. Thus these centers appear to be concerned with the reflex turning of the head and tail and with the regulation of the labyrinthine righting reflexes of the head. The observer stated that the results obtained do not appear to differ from those following a destruction of the frontal ampullae. He also described a new labyrinthine reflex, terming it the labyrinthine reflex of the trunk.

The ramus posterior, commonly called the cochlear ramus, carries, in addition to cochlear fibers, fibers from the ampulla posterior, sacculus, and macula neglecta. The vestibular components in this ramus separate from the cochlear components and terminate in the most dorsal part of the nucleus tangentialis. Winkler (’07) and Sanders ('29) believed such to be the case, the latter observer tracing certain of the fibers to pericellular terminations arotmd the cell bodies of the neurons, although she regarded the majority of them as fibers of passage. Wallenberg ('00) and Ramdn y Cajal (’08a) believed that the fibers merely run through the nucleus without synaptic relation to it.

There are at least two nuclei of termination for cochlear fibers in birds, the nucleus (dorsalis) magnocellularis and nucleus angularis (figs. 207 to 210). There is question as to whether incoming cochlear fibers pass directly to nucleus laminaris. These centers are homologous to the similarly named centers in reptiles. Bok noted that even in birds some cells of the nucleus magnocellularis occupied a more ventral position. The nucleus (dorsalis) magnocellularis is probably the forenmner of the mammalian ventral cochlear nucleus, as has been made clear by Ramon y Cajal ('08a) and Brandis ('94). The nucleus angularis is regarded as homologous to the tuberculum acusticum or dorsal cochlear nucleus of higher animals. This nucleus angularis is phylogenetically younger than is nucleus magnocellularis, thus it is present in reptiles but not in amphibians, and it appears (Bok, ’15) a day later in embryonic development than does the latter nucleus. In addition to these nuclei of termination, Bok traced fibers directly to the superior olive.

Nucleus angularis lies along the dorsolateral siudace of the medulla oblongata. The cochlear fibers which enter the brain at the level of this nucleus swing dorsalward in order to enter the nucleus from its ventromedial side. They separate its caudal portion from the nucleus magnocellularis. The nucleus is continued

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

481

forward to a plane through the caudal end of the abducens nucleus (sparrow, Sanders, ’29). The cells of this nucleus are medium sized, multipolar, and tend to be somewhat spindle shaped in outline.

Sanders described three portions in the nucleus magnocellularis of the sparrow, a mediomedial, a mediolateral, and a ventrolateral part, but pointed out that

these subdivisions do not appear to indicate any fundamental cytological difference. Of these, the first two are the more cephalic and both have, through a part of their course at least, a lamellar arrangement of cells. Ultimately the two portions become fused, and immediately behind the level of this fusion, the smallercelled ventrolateral nucleus makes its appearance. Somewhat different subdi\asions of the nucleus magnocellularis have been described by Ramon y Cajal (’08a) and still others by Holmes (’03).

Nucleus laminaris forms a crescent-shaped mass around nucleus magnocellularis. In the sparrow (Sanders, ’29), the crescent consists for the most part of a smaller-celled dorsal portion and a larger-celled ventral part, although the amount of differentiation between the two parts varies at different levels, being less clear at the cephalic end of the nucleus.

482 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Nucleus laminaris consists of small cells, the dendrites of which radiate out in both dorsal and ventral directions. The cells are arranged in a definite row with processes on either side, hence the name of the nuclear mass. In an effort to homologize this nucleus with the cell masses in mammalian brains, Ramon y Cajal (’08a) suggested that this nucleus might be regarded as the homologue of a mammalian accessory olive. This homology is based on a certain degree of similarity in shape and on the presence in each of a somewhat common type of cell. It is uncertain at most, since in lower mammals the inferior olive is not lamellar in shape.

Nucl. aagi]laris

Coch. nerve

Vent. coch. cell

Vent. coch. fib. in'

Vent, arcuate eyetem

Sup, olive and lat. lem.

Ventral extension ol

Nucl. magnocellularia

Nucl. laminaria

Cocbleo-cerebellar tr.

Dorsal cochlear, com.

Vent, cochlear, com.

Fig. 210, The course of the primary and secondary cochlear fibers in a hen. The diagram is according to Bok,

The superior olive consists of two portions, a dorsal portion and a ventral one (Schepman, ’18, Ariens Kappers, '20). The dorsal part, which is also the more caudal part of this nuclear mass, is usually the larger of the two. It corresponds to the dorsal cephalic portion of the nucleus in reptiles, but is relatively farther ventral in birds, with the exception of runners, but still within the course of the arcuate fibers from the nucleus laminaris. The thinner ventral portion of the nucleus may either be connected with the dorsal part by a narrow band of gray or may be distinct from it. This ventral portion in many birds extends far forward along the course of the longitudinally running lateral lemniscus. This dorsal portion is presumably that portion of the olivary nucleus described by Sanders as forming, in the sparrow, a more or less round nuclear mass with cells arranged in somewhat concentric rings around a central portion. Probably the thinned down anterior portion, which was traced to the caudal end of the ventral trigeminal nucleus, corresponds to the ventral part of the nucleus described by Ariens Kappers and others. A nucleus of the trapezoid body has not been demonstrated as yet in avian forms. Since it is already present in the lowest mammals, it may be represented in birds by scattered cells of the reticular gray or of the raph4. A considerable number of larger scattered cells, near the origin of the lateral lemnis

THE LATERAL-LINE AND ACOUSTIC SYSTEMS 483

cus, is sometimes regarded as belonging to the superior olive {Ariens Kappers). Similar groups are seen farther forward where the superior reticular nucleus shows a tendency to relate itself to the lateral lemniscus.

Having now described the main cochlear centers of the medulla oblongata, the fiber relations may receive consideration (fig. 210). The more lateral incoming fibers of the cochlear root run lateralward and dorsalward to end in the nucleus angularis. The greater part of the medial group of cochlear fibers, including all of its most caudal fibers (Sanders ’29, for the sparrow), enters the lateral side of the dorsal and caudal areas of nucleus magnocellularis, where it distributes. Farther forward, cochlear fibers enter the ventrolateral border of the nucleus magnocellularis. In order to reach the dorsal and cephalic parts of the nucleus, the cochlear fibers swung through the nucleus laminaris, and, according to Winkler (’07) and Sanders (’29), distribute to this nucleus to some extent during their passage through it. Sanders described such terminal branches as unmeduUated collaterals of the medullated stem fibers. Such collaterals terminate in pericellular sjmapses around the cell bodies of nucleus laminaris. Brandis (’94), Wallenberg (’98 and ’00), and Ramdn y Cajal (’08a) regarded the cochlear fibers as merely passing through the nucleus without entering into synaptic relation with its cells. Bok (’15) believed that he was able to trace fibers of the cochlear root directly into the superior olivarj' nucleus, and Sanders (’29) w^as of the opinion that the material studied by her showed a few cochlear fibers joining the ventral arcuate system directly.

From nucleus magnocellularis ' fibers run ventralward close to the periventricular gray and decussate to the opposite side under the median longitudinal fasciculus. Joined by uncrossed fibers, they run lateralward and slightly ventrolateralward and become a part of the lateral lemniscus. During their course, they give some fibers to the superior olive. This nucleus is in synaptic relations w’ith nucleus laminaris, even if there is no clear evidence at present that this latter nucleus receives direct cochlear fibers, since fibers previously regarded as such are now thought to be concerned with other cell masses. Wallenberg (’98, ’00) proved that section of the cochlear root produces no degeneration within the nucleus. Ramdn y Cajal (’ll) and Bok (’15) showed that fibers passing to the nucleus, both internally and at its periphery, were neuraxes of neurons of the nucleus magnocellularis dorsalis. Such fibers were both crossed and uncrossed, the former distributing to the ventral side and the latter to the dorsal side of the nucleus. Nucleus laminaris and nucleus magnocellularis as w-ell are said to have connections with the cerebellum. Nucleus laminaris gives rise to crossed fibers to the lateral lemniscus. The course of these fibers is at first ventralw'ard. They send collaterals to the homolateral superior olive. These fibers have been demonstrated in degeneration material by Wallenberg (’98, ’00). Cochlear fibers like ’’ The connection of nucleus (dorsalis) magnocellularis with the cerebellum has been described by Bok (T5), as tractus cochleo-cerobellaris. Tlie termination of these fibers within the cerebellum is not’knowTi. Schepman (’18) believed that vestibular fibers of the same side join this tract. Mesdag (’09) described a similar tract but regarded it as originating from nucleus laminaris. Should such connections receive more definite confirmation, they would open interesting possibilities with regard to the function of the cochlear nerv'e.

484 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

wise terminate in the nucleus angularis. This nucleus sends crossed and possibly uncrossed ncuraxes to the nucleus laminaris. From it arises a tract which runs along the ventrolateral surface of the medulla oblongata and becomes connected with the lateral lemniscus. It appears to give collaterals to the more ventral part of the superior olive.

The lateral lemniscus, then, arises from the cells in nucleus angularis, nucleus magnocellularis dorsalis, and nucleus laminaris. It is not believed to receive fibers from the superior olive. From the .superior olivary nucleus, according to Yoshida (’25), neuraxes pass to the median longitudinal fasciculus and the predorsal fasciculus and descend to the cervical cord. This is considered of first importance for head movements in the direction of sound. The superior olivary nucleus also distributes largely to the motor nuclei and reticular gray of the medulla oblongata. The lateral lemniscus runs forward in company with other ascending systems and terminates in the nucleus semilunaris, nucleus isthmi, and nucleus mesencephalicus lateralis pars dorsalis. The details of its relations in the midbrain will be discussed in Chapter VIII (see also fig. 210).

The Vestibular and Cochlear Systems of Mammals

Before entering upon a description of the central relations of the acoustic nerve in mammals, consideration will be given to the peripheral branches of this nerve. As in other vertebrates, in mammals the acoustic nerve gives off in the meatus auditorius internus, a ramus anterior and a ramus posterior. The former arises from the pars superior of the ganglion vestibulare and distributes to the ampullae anterior and externa and to the utriculus. The ramus posterior consists of two portions, one arising from the pars inferior of the ganglion vestibulare, with peripheral distribution to the ampulla posterior and the sacculus, and the other, the cochlear branch of the nerve, from neurons of the spiral ganglion, with dendritic branches from the membrana basilaris cochleae. This has been verified recently by de Burlet (’29). Sometimes the branches from the ampulla posterior and the sacculus are united into a common branch called the ramus medius. The branch supplying the cochlea is often called the cochlear ramus or cochlear nerve, or the ramus posterior sensu strictoris. In the rabbit and the cat Oort (’18) described an anastomosis between the cochlear ramus and the saccular branch. This is of considerable phylogenetic interest, since it is a reminder of the condition in fishes, where the ramulus lagenae, from which during later development the cochlear ramus arises, is a branch of the ramus saccularis. However, a lagena, and thus a ramulus lagenae, is not present in higher mammals, although both are present in monotremes {Alexander, ’04). Likewise the papilla neglecta has not been demonstrated with certainty in higher mammals (see fig. 211).

The ganglion of the vestibular nerve has been studied by many observers {His, ’88 ; Retzius, '92 ; von Lenhossek, ’93 ; Alexander, ’99 ; Ramon y Cajal, ’09 ; Streeter, ’07, and others). The evidence agrees that the neurons of the adult ganglion are bipolar and not imipolar. The dendrites of such neurons havesensory endings on the hair cells of the peripheral organ. The neuraxes enter the brain stem to end in the vestibular nuclei. Ramdn y Cajal (’09) described certain

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

485

fine, afferent fibers ramifying among the cells of the vestibular ganglion in the guinea pig. The origin of such fibers appears to be unknown although Ramon y Cajal believed that the neuraxes from which such fibers arose ran in the nerve of Wrisberg.

The central vestibular apparatus is less complicated in mammals than in birds and the number of nuclei is less. This is probably due to the fact that animals

Fig. 211. .Schematic diagram of the acoustic branches in the rabbit. Oort.

The ramus anterior, which supplies the macula utriculi, gives off side branches to both vestibular ampullae; from it a small twig goes to the macula sacculi iVoil). The ramus posterior divides into two branches.

The branch which supplies the macula sacculi innervates also the posterior ampulla. This same branch also gives rise to a small bundle which runs with the nerviis cochlearis to the cochlea. This bundle provides a connection between this portion of the cochlear nerve and the vestibular nerve.

which fly make special demands on this apparatus. The vestibular root enters in front of the cochlear root. The tangential nucleus, which in birds, reptiles, and teleosts lies at the entrance of the vestibular root, is represented in mammals by clusters of cells among the intramedullary vestibular fibers {Ramon y Cajal, ’09, Klosscnvsky, ’32). The other vestibular nuclei generally recognized in mammals are the medial or Schwalbe’s nucleus, the superior or Bechterew’s nucleus, the lateral or Deiters’ nucleus, and its caudal continuation, the descending or the spinal vestibular nucleus. An addition to these nuclei, according to many observers, is the accessory or external cuneate nucleus (nucleus of Burdach-von Monakow or nucleus of von Monakow) which is lateral to, or in close association with, the main cuneate nucleus. The medial vestibular nucleus (fig. 212) lies near the floor of the ventricle at the level of entrance of the vestibular root, in a position just medial to the lateral or Deiters’ nucleus, the whole vestibular area at this level being situated internal to the inferior cerebellar peduncle. From this position, in many mammals, the nucleus extends forward to a plane through the abducens

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

487

the upper part of the body with proprioceptive impulses entering by way of the vestibular. Its neuraxes pass directly to motor centers of the cord and thus mediate immediate reflex responses.

Turning now to a consideration of the distribution of the vestibular fibers to the several vestibular nuclear masses and to the secondary connections there formed, it becomes evident that this has been considered by many observers, based on observations obtained by the use of various methods, with results that are far from a general agreement. It is beyond the scope of this account to enter into a discussion of the questions here involved. The interested reader is referred to the original sources. Among such contributors may be listed Baginsky (’86, ’90), von Bechterew (’87), Flechsig (’90), Held (’91 and ’93), Obersteiner (’93), von Kolliker (’93), Cramer (’94), Winkler (’07, ’08, ’10, ’ll), Ramon y Cajal (’09), Lewy (’10), Brouwer (’12), Fuse (’13), Brouwer and van Walree (’14), Muskens (’14, ’22) Schepman (’18), Ariens Kappers (’20), Yoshida (’24 and ’25), Gray (’26), and Rasmussen (’32) .

The great majority of vestibular root fibers, soon after their entrance into the medulla oblongata, bifurcate into two branches, an ascending and a descending one. The latter carries, as in birds, the thicker fibers. From the ascending branch, fibers pass directly to the cerebellmn {van Gehuchten, ’07 ; Ramon y Cajal, ’09, and others). On their way to the cerebellum they cross the superior vestibular nucleus or nucleus of Bechterew (which is connected by commissural fibers with its contralateral homologue, Yoshida, ’24), to which they give collaterals, and then pass medial to the corpus restiforme and reach the centrum medullare of the cerebellum. Ingvar (’19, fig. 213) was able to show that vestibular fibers terminate not only in cerebellar nuclei but also in the nodule, uvula, lingula, and flocculus of the cerebellar cortex. Thus in the mammalian cerebellum there is a basal portion which receives direct vestibular impulses and consequently forms the vestibular foundation upon which the rest of the cerebellum is built. Figure 506 illustrates the following discussion.

The vestibular root sends fibers into all four vestibular nuclei. The ascending fibers terminate particularly in relation to the superior vestibular nucleus. Other fibers pass to Deiters’ nucleus (lateral vestibular nucleus) and to the medial vestibular nucleus. Descending fibers distribute to the lateral vestibular nucleus and its caudal continuation, the inferior, descending, or spinal vestibular nucleus and, according to many observers, to the external cuneate nucleus or the nucleus of von Monakow. However, there is no general agreement as to this last mentioned distribution. There is not accord at present as to whether or not the vestibular nerve sends direct root fibers to the vestibular centers of the other side. Ramon y Cajal (’97 and ’09) described a bundle of fibers crossing directly to the opposite side near the level of the nucleus of the abducens, constituting a transverse or crossed vestibular bundle. Recent work suggests that this may be a decussating facial bundle rather than a vestibular fasciculus. Leidler (’13 and ’14) believed this bundle to be constituted of efferent fibers of the nucleus superior salivatorius, and Schepman (’18) was able to demonstrate the presence of crossing fibers in a cat in which de Kleijn had cut the vestibular root. Gray (’26) believed

meddia^obionga'tf^^rrcat. ° 5. Preparation of cerebellum and

cau“ad‘’Ld ccDhahd’'ioTr.'f;n!, 9 ‘T^ aections 1 and 3 are re.spectivcly

ifere numelus fil^m are (o h^nc^ >" section 2 the fiber path which pas-scs to the flocculus,

each other *" cortex of the nodulus and the lingula, which lobas lie near

488

THE LATERAI^LINE AND ACOUSTIC SYSTEMS

489

he could trace crossed vestibular fibers in his experimental work on the vestibular centei-a in the cat. Rasmussen (’32), working with experimental material, thought that he was able to demonstrate crossed root fibers.

There is no gener.al agreement as to the details of the secondary connections of the vestibular centers in mammals. Whether the disagreement is due to actual dilTerences in the forms studied cannot be determined at present. The superior and medial vestibular nuclei, and probably also the lateral vestibular nucleus, are interconnected with the cerebellum by way of cerebcllo-vestibular and vestibulocerebellar fibers. According to Yoshida (’24), the vestibulo-cerebellar connections are formed by dichotomous division of the fibers, one branch of which passes to the cerebellum and the other into the vestibulo-spinal tract, so that impulses from the part of the lateral nucleus from which these fibers arise are related simultaneously to both centers. According to the same author, the ventral part of the lateral vestibular nucleus and the spinal vestibular nucleus send their fibers into the spinal cord only, distributing both contralatcrally and homolaterally (see also Winldcr, ’08). Both the lateral and the spinal vestibular nuclei receive fibers from the uncinate f.asciculus which pro\ide a cerebello-vestibular connection {Lewandowshy, ’04, and others). The superior and medial vestibular nuclei are connected also with the cerebellum. Connections of the medial vestibular nucleus with uvula .and nodule have been described. The medial vestibular nucleus and the inferior or spin.al vestibular nucleus are believed to send fibers to the contralateral medial longitudinal fasciculus. Such fibers from the medial vestibular nucleus bifurcate, then ascend and descend. This has been substantiated recently by experimental work ’of Gray (’26) and of Rasmussen (’32) on the cat. The fibers of the inferior or spinal vestibular nucleus m.ay descend {Gray) or may bifurcate, then ascend and descend {Rasmussen). Opinions vary as to whether the lateral vestibular nucleus contributes fibers to the medial longitudinal f.asciculus. The various studies (as those on birds, Wallenberg, ’98, ’00), on forms below mammals and lower mammals, would lead to such conclusions. Rasmussen (’32) described homohateral fibers to the meditil longitudinal fasciculus in the cat ; Gray (’26) did not find them. The superior vestibular nucleus is believed to contain homolateral fibers to the eye muscles through the medial longitudinal fasciculus {Gray, '26; Rasmussen, ’32; and others). Leidler (’13 and ’14) was of the opinion thiit the contribution from the superior vestibular nucleus is very small and that an injury to this nucleus alone does not produce any pronounced effect upon eye movement. It is to be understood that these connections through the medial longitudinal fasciculus produce eye and head movements in response to changes of position of the body. The lateral vestibular nucleus {Gray ’26), and, according to certain investigators, the spinal vestibular nucleus as well {Yoshida, ’27), gives rise to a descending vestibulo-spinal system which runs ventralward from its origin in the floor of the medulla oblongata. In the medulla oblongata it lies ventral to the lateral spino-thalamic system and is found in the spinal cord in the ventrolateral part of the lateral funiculus. In man it extends throughout all or nearly all of the spinal cord. It constitutes an important path conducting impulses for the changes in position of the body,

490 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

conducive to the maintenance of equilibrium, (Certain connections of the vestibular nuclei are shown in fig. 506.)

With the results of Gray (’26) and Rasmussen (’32), as above quoted, it is desirable to contrast the results of Muskens (’14), obtained in his study of the relations of the medial longitudinal fasciculus (posterior longitudinal fasciculus) to forced movements. Muskens traced contralateral vestibulo-mesencephalic fibers from the vestibular area,* chiefly the nucleus triangularis (probably the medial vestibular nucleus of the foregoing terminology). Homolateral vestibulomesencephalic fibers, from the superior vestibular nucleus, were traced to the medial longitudinal fasciculus, where they occupied a position just e.xternal to the eontralateral vestibulo-spinal path. Muskens found degeneration of this path associated with conjugate deviation toward the normal side. Mtiskciis found no definite descending or ascending connections in association with the inferior, descending, or spinal vestibular nucleus, although animals which exhibited a lesion of this nuclear gray revealed rolling movements toward the uninjured side. These movements persisted for several days after the operation. Injury to tlie magnocellular part of Deiters’ nucleus (the lateral vestibular nucleus) were associated with rolling movements toward the normal side, and demonstrated the presence of an ascending tract to the tegmentum, the tractus vestibulo-tcgmcntalis dorsalis of Muskens. This observer traced homolatcral and contralateral descending fibers from Deiters’ complex of his terminology, and a homolatcral descending group from the superior {Bechterew} vestibular nucleus. Such descending fibers were believed to be concerned particularly with the rolling movements which this experimenter obtained. Injury to the entering vestibular root fibers produced such a rolling movement toward the injured side, and usually “a .skew deviation and conjugate rotation of the eyeballs on their antero-postcrior axis.” The vestibular connections through the medial longitudinal fasciculus as described by Muskens are discussed on pages 1076 to 1083 (see also fig. 50G).

No discussion of the vestibular centers and their connections could be considered as closed without some reference to the experimental work on the functional relations of the vestibular apparatus as this has been evidenced fhrough the studies of de Klcijn and Magnus and their associates. These observers havc been working for a number of years on the various reflexes of the normal labj'rinth and on determining means of varying them under experimental conditions. They distinguished three groups of tonic labyrinthine reflexes: (1) tonic labyrinthine reflexes on the body muscle of the extremities and the neck ; (2) laltyrinthine righting reflc.xes ; and (3) a tonic labyrinthine reflex on eye. muscles. Wicn both labyrinths arc destroyed (dc Klcijn and Magnus, ’21), certain reflexes concerned in progressive movements in various directions in space are destroyed, but

• Accordina to Afux/.ciix, there nre present in tlie ent tlircc vcHtibulnr tiuclcar groups : (1)» de-ceiidiiig liniiicli of tlie vcstibtiliir nucleus (inferior, descending, or spinnl vestibular nucleus) ; (2) II Hcrhlcrctr nucleuK (suiierior vestibular nucleus) ; (a) a Deiters' nucleus, eomiw-ed of a ventral c-audal division tenned the nucleus triangularis, and a dorsal inagnoeellular division or Iteiters nucleus pro|)er. Although the relative position differs slightly from those usually deserilK'd, il is pnibable that the nucleus triangularis is the media! vestibular nucleus and the niagnocelluhir nucleus is the lateral vestibular nucleus of the preceding account.

THE LATERAH-LINE AND ACOUSTIC SYSTEMS 491

positional reflexes are unchanged. The inhibition or augmentation of such reflexes for movements of progression is due probably to impulses from the cerebellum, but the reflexes persist in the absence of this structure. The reflexes are not changed by the removal of the cerebral cortex. De Kleijn and Magnus found that “ Drehreaktionen und -nachreaktionen auf Kopf und Augen sind Bogengangsreflexe und konnen beim Fehlen der Otolithenmembran normal Zustande kommen.” De Kleijn showed the effects of changes of position of the head on the tonic state of the eye muscles. He found that this tonic state is regulated by two groups of reflexes ; these are a tonic labyrinth reflex and a tonic neck reflex. This observer believed that every new position of the head has a corresponding position of the eyes in the orbit.

Ivy (’ 19) studied nystagmus in various animals. This observer discussed the quick component of nystagmus and the slow component. Tilney and Pike (’25) had thought that the cerebral cortex plays a part in restoring the eyes to a normal position. Ivy (’29) believed that the quick component is not dependent upon the cortex, since in the rabbit complete removal of the hemispheres, together with destruction of the thalamus, does not destroy this component. Ivy stated, therefore, that the quick component depends upon segmental proprioceptive impulses while the slow component is related to the vestibular apparatus. This observer suggested that the cerebral cortex may exert an inhibitory influence.

The finer structure of the organ of hearing is outside of the limits set for the present account. Those interested are referred to the excellent discussion and bibliography of Kolmer (’28) in Mollendorf’s “Handbuch der Anatomic des Menchen.” Special mention is made here of the work of Held (’97a, ’04, ’09, etc.), Shambaugh (’07, see p. 504, and elswhere), Hardesty (’08, ’15, etc.), Guild (’21, ’24, ’25) and the very recent paper by Crowe, Guild and Polvogt (Bui. Johns Hopkins Hosp., vol. 54, p. 315).

The first question which arises in connection with the cochlear division of the acoustic nerve is concerned ^vith the central relations of the fibers from the ampulla posterior and the sacculus. These fibers, in the rabbit and the cat, run in the cochlear root peripherally and enter centrally with its dorsal part, but on entering the brain pass directly around the corpus restiforme to a lateral portion of the nucleus triangularis, thus ultimately distributing to a vestibular center. Ramon y Cajal (’09) mentioned the presence of fibers of the cochlear root entering the region of the lateral vestibular nucleus and the central reticular substance directly along the corpus restiforme. That fibers from the sacculus and posterior ampulla are found in the cochlear root is evidenced by the fact that, after destruction of the cochlea, certain fibers of the cochlear ramus remain uninjured and can be traced into the vestibular nucleus {Winkler, ’07).* The vestibular root fibers are myelinated in the 23 cm. human embryo, but m 3 'elin sheaths do not appear in the cochlear root fibers until the 28 cm. stage is reached. Embryological proof of the vestibular character of certain fibers accompanying the cochlear root is offered by the presence of myelin sheaths on the fibers of the vestibular root and on the

s It should be added that Winkler regarded certain fiber bundles entering the dorsomedial part of Deiters’ nucleus as true cochlear fibers.

492 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Can. ext.

Can. post. .

Foramen rotundum \ \\

Gan. ant.

Utriculus

‘Foramen ovale

•• CochUa

Fig. 214. The labyrinth in Echidna. Note the relatively limited development of the cochlea in Echidna as compared with the cochlea in higher mammals. Gray.

fibers of the deeper portion of the cochlear root in the 23 cm. human embryo, at which stage the remaining cochlear fibers are unmyehnated. In a congenitally deaf cat, after labyrinth extirpation, Schepman (’18) found that fibers caudal to

the radix anterior, on entering the brain passed around the inferior cerebellar peduncle, and, after apparently undergoing dichotomous division, distributed to the nucleus triangularis and the cerebellum. Such evidence suggests that in mammals, as in birds and reptiles, the cochlear root carries not only cochlear fibers but also vestibular fibers, which in manunals, as shown by Beccari (’12 ; see bibliography for reptiles) for reptiles, do not pass to cochlear centers but to vestibular centers.

The position of the cochlear nuclei, together with certain central connections, have received quite extensive discussion. Their relations, as compared with submammalian types, are perhaps best understood by presenting a description of the conditions as found in certain of the monotremes and marsupials, since these forms offer transition stages between conditions found in reptiles and birds and those in liigher mammals. In monotremes and marsupials the peripheral apparatus associated with the cochlear branches of the acoustic is relatively simpler than in higher mammals. The cochlea itself is cnn.post.

much more simple, exlnbits little or no coiling, and is similar to that in birds (fig. 214).

The central relations in Echidna have been studied particularly by von KoUiker (’01) and Schepman (’18), and in marsupials have engaged the attention of Holmes (’03) and Stokes (’12). Schepman found that in Echidna the incoming roots bifurcate and then terminate within two nuclear masses.

The ascending branch ends in synaptic relations with a nucleus containing spindleshaped colls, similar to tho.se of the nucleus angularis of reptiles (nu. dors, coch., fig. 21GB). This nucleus is relatively far mcdialward and lies close to the nucleus incdialis or triangularis of the vestibular nuclear group. The descending branch terminates in a large-celled nuclcu-s similar to the nucleus magnocellularis of reptiles and birds. The nucleus magnoccllularis extends fonvard in two prolongations, one of which lies lateral, the other

Cbd

Ett.

8iu«ulu«

aDwil.r m.nib. «*■>'•

Fig. 21.0.

Tlio liibyrintli in a ralibit.

llelriu’

494 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

medial, to the inferior cerebellar peduncle or the restiform body (fig. 216A). The spindle-celled nucleus (the dorsal cochlear nucleus) lies entirely medial to this peduncle (fig. 216B). These relations are repeated in the marsupial brain. Gordon Holmes (’03) pointed out that, as in birds and reptiles, the cochlear nuclei of Macropus lie medial to the corpus restiforme rather than lateral or ventral to it, as is the case in higher mammals. In the opossum, Stokes (’12) described and sketched two nuclei ; a dorsal nucleus, the tuberculum acusticum, and a largecelled nucleus, nucleus ventralis, which lies farther medialward than does the homologous nucleus in higher mammals (fig. 217). The dorsal nucleus in the opossum is medial to the restiform body. The nucleus dorsalis cochlearis receives ascending cochlear fibers. It is probably the homologue of the nucleus angularis of birds and reptiles. Its position in opossum is surprisingly like its position in birds and offers further justification for the statement that the nucleus angularis of birds is the homologue of the dorsal cochlear nucleus of higher mammals. It is less well developed in man than in certain lower mammals, such as the guinea pig, cat, and rabbit {Ram6n y Cajal, ’09). This last mentioned observer has divided the nucleus into four zones ; an epithelial, a plexiform, a zone of granular cells, and a zone of large nerve cells. The cell types are illustrated and described in detail in his text, and reference is made here to that account.

The ventral cochlear nucleus (fig. 219), in many higher mammals, receives its name from its distinct ventral position, since it lies far ventrolateralward along the periphery of the medulla oblongata and ventral to the corpus restiforme. In some mammals it even lies in a projection on the wall of the medulla oblongata. The nucleus dorsalis magnocellularis of birds and reptiles is distinctly dorsal in position and at first glance it may seem surprising that it should be regarded as the homologue of the ventral cochlear nucleus. However, a study of the conditions in lower mammals indicates that the ventral, large-celled nucleus in these forms is intermediate in position between that occupied by the nucleus in birds and that held by the nucleus in higher mammals. Thus, in Didelphis the anterior part of the large-celled ventral nucleus is entirely dorsal and is medial to the corpus restiforme. Only the posterior part of the nucleus has shifted somewhat farther ventralward. A comparison of the conditions in the opossum and bat (figs. 217 and 409) shows the importance of this intermediate stage in the former animal. Stokes (’12) was inclined to regard this migration of the nucleus ventralward and lateralward as due to an increase in bulbar structures and a consequent forcing of the nucleus outward toward the plane of entrance of its incoming fibers. This purely mechanical explanation offers no particular justification for its passing ventralward rather than medialward. More probably the shifting is due to the tendency of the cells to move toward the source of their stimulation, in other words due to neurobiotactic influence {Ariens Kappers, ’20). This has gone so far that there is a tendency for the nucleus actually to pass into the incoming root fibers. A further argument against Stokes’ theory is the fact that the shifting ventrolateralward is not of necessity greatest in those animals with a medulla oblongata complicated by the appearance of many new structures, but most marked in those forms in which the cochlea is particularly well developed

THE LATERAIr-LINE AND ACOUSTIC SYSTEMS

495

fiu.ventr.~-Cjf R cochl. -At

Rvest.-'

corp.trap.

and where there are the richest provisions for the carrying of cochlear impulses, as

in Chiroptera, rodents, and carnivores, and in the whale. The ventral migration

of the nucleus magnocellularis ^ a

, , , Ru dors R. cochl.

or nucleus ventralis cochleae i

is most marked in the bat and

the whale, although the me- dulla oblongata in these forms

has a relatively simple devel opment (fig. 409). The mi- iVj

gration of the magnocellular

or ventral cochlear nucleus nuyg^tr

toward the entering root fibers

is regarded {Ariens Kappers,

’20) as ha\’ing exerted an in fluence on the nomenclature p ^,gjj

of the acoustic roots in mam maE. Thus the term dorsal

root is applied to the cochlear

root in reptiles and birds, /

since it enters dorsal to the /

vestibular root, which is con- corp.trap, i

sequently called the ventral ol.sup.

root. In most mammals this

relation has been somewhat nu. dom. r. cocu.

altered by the ventrolateral /

shift of the nucleus magno- .. oV ' ^ /' ^ ^T'

cellularis, so that the ramus ( y\-:A C i T'x^ “'■■“ed.vest

cochlearis enters the ventral

nucleus ventral to the point i

of entrance of the vestibular \

root. Consequently the cochlear root is termed in mammahan anatomy the ventral root and the vestibular the dorsal root. The details of

the cytologic structure of this j

ventral cochlear nucleus are available in the text of Ramon y Cajal (’09), to which refer ence is made. ^

The changes in the loca- 217. A and B. The relations of the acoustic nuclei in , . , . , , . , r opossum, after Slakes.

tion of the nuclei have of

necessity influenced the course of the secondary paths. Moreover, the enlargement of the nuclei has led to differences in their forms and relations. Both the tuberculum acusticum, or the dorsal cochlear nucleus, and the ventral cochlear

su. dors. R. cochl.

I DU ventr. K. cochl.

$

f / tr desc. N. vest.

~3

.;•■■■* !,,• V.'M'Jil’

496 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Nu. lai cochl. (tub. acuse.)

nucleus have become extraordinarily enlarged. This enlargement has occurred largely in a caudal direction in the dorsal nucleus and principally, though not entirely, in a frontal direction in the ventral nucleus. With these changes in position, the dorsal nucleus receives not only ascending but descending root fibers as well, and the ventral cochlear nucleus receives likewise ascending fibers. Thus there is not a complete homology in the distribution of ascending and descending root fibers in mammals as compared with the distribution in birds or reptiles, but this difference is due to the enlargement of the nuclei rather

than to any fundamental difference in their character.

Having discussed briefly the evolution of the cochlear nuclear groups and their peculiarities in various mammals, there now remains a consideration of the various fiber connections and certain secondary and tertiary nuclear masses associated with the system. Close to their place of entrance to the medulla oblongata, from their cells of origin in the spiral ganglion, the fascicles of the cochlear root come into relation with the nucleus magnocellularis or nucleus ventralis cochlearis. Here the fibers dichotomize; the ascending branch terminates in the so-called baskets or calyces of Held (’93), around the cells of the ventral nucleus {Ram6n y Cajal, ’09), while the descending fibers rim mainly to the dorsal cochlear nucleus or the tuberculum acusticum. According to Held (’93) and Winkler (’07), some of the root fibers pass directly to centers which are usually regarded as secondary in character ; thus the fibers are believed by them to pass directly through the trapezoid body to the superior olive. Figure 529 illustrates the following account.

From the dorsal cochlear nucleus, secondary fibers (not striae medullares acustici, see p. 804; Fuse, ’ll, ’12) pass over the corpus restiforme and then near the floor of the ventricle to the midline. Here most of them decussate in the region immediately ventral to the medial longitudinal fasciculus. Crossed and uncrossed fibers then swing ventrolateralward to the region of the superior olive. According to certain observers (von Bechierew, and others), they give collaterals to this nucleus. The fibers from the dorsal cochlear nucleus unite with bundles from the ventral cochlear nucleus to form the lateral lemniscus. Fascicles from the ventral cochlear nucleus run toward the midline in the more ventral portions of the medulla oblongata (or pons), dorsal to the pontine fibers and interweaving in their course with the fibers of the medial lemniscus. These fibers are termed the corpus trapezoideus. This bundle is larger in birds and mammals than are the secondary fibers from the dorsal cochlear nucleus. According to Held (’93), Winkler (’07), and Schepman (’18), vestibular root fibers are

Ku. magn. VIII.

Fia. 218. The tuberculum acusticum and the nucleus magnocellularis VIII or nucleus ventralis in the cat. Fxise.

THE LATERAI^LINE AND ACOUSTIC SYSTEMS

497

present in the trapezoid body, but most observers do not agree with this statement. The relatively more ventral position of the trapezoid body in mammals than in birds is undoubtedly due to the more ventral position of the ventral cochlear nucleus, which certainly is its chief center of origin in the lower animals (Anens Kappcrs). In Echidna,, where the ventral nuclei of the trapezoid body are small and the corpus trapezoideum is only a thin band close to the outer surface of the medulla oblongata, the secondary cochlear connections are smaller from the ventral than from the dorsal cochlear nucleus. As far down in the mam

Fio 219 Entrance of the Vlllth nerve in man (right). Left • tlie ventral cochlear nucleus and the tuberculum acusticum (The leader for nucleus VII should be slightly extended )

malian scale as Didelphis, however, there is a strong ventral crossing, while the dorsal crossing is poorly developed. In many mammals, at least, there are welldeveloped nuclei of the trapezoid body with which the fibers have special synaptic relations. Collaterals and perhaps stem fibers are also given off from the trapezoid body to the superior olive of the same and of the opposite side. Lateral to the olive the trapezoid body and the fibers from the dorsal cochlear nucleus unite to form the lateral lemniscus. Another component to the lateral lemniscus has been described under the name of the intermediary bundle or the decussation of Held (’93). The fibers constituting this bundle are said to arise from the dorsal part of the ventral cochlear nucleus {Ramon y Cajal, ’09 ; Yoshida, ’24), run over the corpus restiforme, and then swing somewhat ventralward so that they cross the raph4 in a position intermediate between the decussation of the corpus trapezoideum and the secondary dorsal cochlear fibers and then swing ventralward to join the former and enter with it the lateral lemniscus. (For a diagram of these paths see figure 529, to which reference was made on the preceding page.)

As the lateral lemniscus turns forward in the pons region toward mesencephalic and diencephalic centers, it is found to consist of secondary neurons of the

498 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ascending cochlear paths situated in the contralateral and homolateral nucleus dorsalis cochlearis, in the contralateral nucleus ventralis cochlearis, and in the trapezoid nuclei. In addition to these generally recognized connections, certain observers (Yoshida, ’25) believe that the lateral lemniscus receives contributions from the superior olivary nucleus and possibly from accessory superior olivary nucleus, described by Ramdn y Cajal (’09) and regarded by him as the possible homologue of the nucleus laminaris of birds. A connection of this accessory superior olivary nucleus with the lateral lemniscus has not been demonstrated as yet. As far as concerns the contribution of the superior olive, there is still some doubt. It is of interest that Malone (’23) found the superior olive of efferent cell type, a fact which would suggest that at least its main connection was not with higher centers.

In its course frontalward the lateral lemniscus lies at first lateral to the superior olive. In front of that nucleus it comes into close relation with the medial lemniscus, the two forming — together with the ventral and lateral spino-thalamic paths and the secondary trigeminal fibers — a great ascending sensory system. On entering the midbrain, the lateral lemniscus swings dorsalward to enter the inferior colliculus and the medial geniculate nucleus. Along the course of the lateral lemniscus are two nuclear masses, the upper and lower nuclei lemnisci laterales, lying medial and against the lateral lemniscus. The dendrites of these neurons receive stimuli from the neighboring tegmental regions. Their neuraxes enter the lateral lemniscus of the same side and of the opposite side (after crossing in the so-called commissure of Probst). Moreover, the large reticular cells of the nucleus reticularis superior lie very close to the lateral lemniscus and in synaptic relation with it in the isthmus region (see Chapter VI).

The inferior colliculus, which is homologous with the corpus posticum of lower forms, is the midbrain auditory center. From this center impulses are relayed in many mammals directly into the tecto-bulbar and tecto-spinal paths. In other mammals, including man, the major number of fibers of the tecto-bulbar and tecto-spinal system arise at least largely, if not exclusively, in the superior colliculus. To associate such efferent pathways with these secondary ascending eochlear pathways, impulses are sent directly to the superior colliculus through certain small fascicles of the lateral lemniscus, or indirectly to this center after a synapse in the inferior colliculus, by way of the acoustico-optic tract. The inferior colliculus may relay auditory impulses to the medial geniculate by way of the peduncle of the inferior colliculus (pedundle or brachium of the medial geniculate nucleus). It does not contribute fibers directly to the cortex (many observers, including Yoshida, ’24). The medial geniculate nucleus is a metathalamic auditory center, a necessary relay station in the pathway to the auditory cortex (superior transverse temporal gyrus, or Heschl’s convolution). The inferior colliculus and medial geniculate nucleus are considered in detail in Chapter VIII ; the auditory cortex in Chapter X.

In lower forms a part of the so-called nucleus isthmi stands intercalated in the course of the lateral lemniscus fibers. Considerable evidence has been produced to indicate that in type of connections it resembles the medial geniculate of

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

499

mammals. However, its position makes it relatively certain that this homology is not correct. A suggestion has been offered recently by Ariims Kappers as to the presence of a possible homologous nuclear mass in man. According to this observer, the nucleus dorsalis lemnisci lateralis located in front of the level of entrance of the trigeminal nerve in the lateral part of the field close to the lateral lemniscus in many mammals, from its position as well as from the general character of its cells suggests itself as the homologue of a part at least of the reptilian nucleus. Traces of this nucleus isthmi may be discerned in the human nucleus lemnisci dorsalis of Marburg (’24). The matter receives further consideration in Chapter VIII (p. 1170).

VTiether or not the superior olivary nucleus contributes fibers to the lateral lemniscus (advocated bj' Yoshida, ’25, and questioned by certain other observers), its major efferent connections are quite different in character. From the superior olivarj' nucleus neuraxes are sent to various efferent centers. Of such connections, one is represented bj' a fasciculus of fibers to the motor nucleus of the abducens nerve. This fasciculus is designated the peduncle of the superior olive. It is concerned in providing a pathway from auditory centers to the motor nuclei of the eye and neck muscles by way of the nucleus of the abducens and the medial longitudinal fasciculus, into which Yoshida (’25) carried fibers from the medial part of the superior olivary nucleus. An illustration of the functional significance of this path is to be found in the ordinary reflex of turning the eyes and head toward the sound (see fig. 529). The superior olive also is believed to have connections vith the facial nucleus, producing reflex movements of the ear muscles, although this connection as yet has not been demonstrated with certainty.

It may be recorded that the superior olivary nucleus in certain mammals is corrugated or folded so as to suggest a field of the inferior olivary nucleus and has Ij'ing in fairly close proximity the accessory superior olivary nucleus. Two olivary nuclei were found in certain mammals by Hoffmann (’08) . IVhether these are comparable to the two divisions described for birds and reptiles is uncertain. The larger, more dorsallj’^ situated part is best developed in those animals in which auditory reflexes are of greatest importance. In general, the superior olivary nucleus is ventral in mammals, a condition associated with the ventral position of the ventral cochlear nucleus in these animals. There is no direct evidence that stimuli of the vestibular system are carried forw'ard to the inferior collicular region. No such connections have been demonstrated in mammals. The phylogenetic history of the lateral lemniscus suggests that such connections might possibly be present. Winkler (’07) particularly has emphasized the interrelation of the two systems.

In this connection, an interesting experiment of Hornbostel might be mentioned. This observer found that there was a tendency for body movements to conform to the pitch of the music in dancing. Thus when the pitch gradually became higher there was a tendency to lift the head, limbs, and thorax, while when the pitch was gradually lowered, there was a drooping of the limbs and expiration. Thus the body movements strove to reproduce the pitch of the soimds

500 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

heard. A study of the body responses to dance music and to marches suggests a response in terms of body orientation to musical sounds. The theory of Rutz, later elaborated by Sieve that attitudes of the body are often called forth by ideas, expressed through music, poetry, or appropriate prose selections, has its basis also in the relation between the mechanisms for sound and for orientation in space (Ariens Kappers, ’20).

Resume or the Laterae-Line and Acoustic Systems

Both anatomically and physiologically, the lateral-line nerves, the vestibular nerve, and the cochlear nerve are to be regarded as related groups of fibers. Physiologically all three of them are specialized for the appreciation of vibrations, either directly or by synchronously vibrating fluids and membranes. Anatomically these receptive organs show certain similarities of structure. Each of them has neuroepithelial cells differentiable from the surrounding epithelium by the presence of hairs on the free surface and by the greater shortness of the cell. Primitively these organs are in communication with the surrounding medium ; that is, the water. Secondarily the vestibular and cochlear apparatus, and in some cases the lateral-line canals, lose this open communication. Centrally the relationships of the terminal fibers of these three types of nerves are distinctly pronounced. All three have their centers in the dorsolateral somatic sensory area of the medulla oblongata. Lateral-line organs and their accompanying nerves and centers are present in aquatic vertebrates from cyclostomes through aquatic amphibians. A distinct cochlear system is present from land amphibians through higher mammals. The vestibular apparatus is found, with few exceptions, in all vertebrates. In gill-breathing vertebrates (fishes, perennibranchiate amphibians, and caducibranchiate amphibians during the time they possess gills) the lateral-line organs are found on the side of the body and on the head. They are stimulated by longitudinal sinusoidal vibrations of the surrounding fluid of such a low rate of frequency (6 per second) that they fall below vibrations of a rate (10-20,000 per second) that can be perceived by the cochlea. Nerve fibers passing centralward from these lateral-line organs reach the most dorsal portion of the medulla oblongata, a portion of which is covered by a continuation of the molecular layer of the cerebellum. This suggests the intimate relation in function between the lateral-line organs and the cerebellum. The central endings of the lateral-line nerves are largely to homolateral gray. Incoming fibers of the nervus lateralis anterior divide, then ascend and descend. Those of the nervus lateralis posterior ascend chiefly, extending as far forward as the entrance of the anterior lateral-line nerve. Terminations of the lateral-line system in aquatic animals are closely associated with those of the vestibular nerve.

The acoustic nerve in lower vertebrates consists of an anterior and a posterior ramus. The anterior ramus, and originally the most ventral one as well, in most cases carries into the medulla oblongata fibers from the anterior and external ampullae and from the utriculus. The ramus posterior, and primitively the one which enters more dorsally, contains, in fishes, fibers from the posterior ampulla, from the macula neglecta, the sacculus, and the papilla lagenae. In amphibians

THE LATERAL^LINE AND ACOUSTIC SYSTEMS

501

it also carries fibers from the region of the lagena, the papilla basilaris, which is the beginning of the cochlea. This latter portion ultimately becomes so large that by far the majority of fibers of the ramus posterior are cochlear branches. In mammals the entire ramus is called radix cochlearis. It is frequently forgotten that even there it carries some fibers from the sacculus and the ampulla posterior, and consequently retains at least a small degree of vestibular function. The presence of these vestibular fibers in the cochlear root has been demonstrated in birds.

Centrally the connections of the ramus anterior are found to be fairly constant during phylogeny, while those of the ramus posterior varj’^ in harmony with the degree of development of the cochlea. The true vestibular connections are in principle fairly analogous throughout all classes of vertebrates. On entering, the roots divide and give off ascending and descending fibers. Some of the ascending branches extend as far as the cerebellum. The descending fibers enter the spinal cord in amphibians, and in many vertebrates reach the place of transition from the medulla oblongata to the spinal cord. Thus thej' come into relationship vuth neurons which are concerned in the control of the equilibrium of the body. The mammalian nucleus of von Monakow, with its intimate relation to the nucleus of Burdach, suggests the correlation between vestibular and proprioceptive functions in these forms. In the region of the medulla oblongata the nucleus of Deiters, or its forerunner, is present in all vertebrates. Phylogenetically it is derived from the large reticular elements which progressively become differentiated into a distinct nuclear mass. As far up in the scale as reptiles this nucleus is confined to the medulla oblongata at about the level of entrance of the vestibular root, but in manunals it has a considerable caudal extension. In fishes, the cell of Mauthner represents a particular reflex mechanism associated with vestibular impulses concerned in tail movements in response to vestibular and other sensory stimuli. This cell is also present in aquatic amphibians. The tangential nucleus of Ram6n y Cajal occurs in fishes, reptiles, birds, and mammals. The medial (principal or triangular) vestibular nucleus has been demonstrated wnth certainty only in mammals although evidence is accumulating that it has representation in reptiles and birds. It has been suggested that it may be concerned in the fine elaboration of eye movements. All of the above nuclei, together with the specialized representatives found in birds — that is, nucleus vestibulo-cerebellaris, nucleus pyriformis, and nucleus quadrangularis — receive incoming vestibular impulses. The details of the connections are to be found under the discussion of the several forms.

The posterior ramus alters considerably in its central relations with the development of the cochlea. That portion of it which carries proprioceptive impulses from the semicircular canals terminates in the vestibular nuclei. In fishes the fibers from the lagenae terminate in relationship wth lateral-line fibers in the posterior lateral-line lobe. In land amphibians, where a true papilla basilaris cochleae is present, fibers from this papilla pass to a differentiated nucleus near the dorsal surface of the medulla oblongata, termed the nucleus (dorsalis) magnocellularis. In reptiles two nuclei of termination for cochlear

502 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

fibers are to be differentiated; the one, nucleus (dorsalis) magnocellularis, the homologue of the amphibian nucleus of that name, the other, nucleus angularis. Intimately associated with nucleus magnocellularis in this form and in birds is a nucleus laminaris which does not, however, receive directly incoming cochlear fibers. In birds more highly differentiated nuclei, but nuclei comparable to those of reptiles and bearing similar names, are to be found. In mammals two nuclei of termination, the dorsal and ventral cochlear nuclei, are present. The nucleus (dorsalis) magnocellularis of lower forms is the homologue of the mammalian ventral nucleus, and nucleus angularis is the homologue of the mammalian dorsal nucleus. Nucleus laminaris has not been demonstrated with certainty in mammals. The secondary acoustic center, the superior olive, has its beginning in amphibians and reaches a considerable development in reptiles and birds, in which latter form it is differentiable into a more caudal and a more cephalic portion. Incoming cochlear fibers distribute to the nucleus magnocellularis in amphibians, to the nucleus of this name and the nucleus angularis in birds, and to the dorsal and ventral cochlear nuclei in mammals. From these nuclei secondary fibers, crossed and uncrossed, ascend in the lateral lemniscus ; secondary fibers pass to the superior olive. In forms below mammals, the lateral lemniscus terminates chiefly in nucleus isthmi and the corpus posticum (the mammalian inferior colliculus). In its course, the tract gives off fibers to associated gray nuclei which are all of the character of a nucleus of the lateral lemniscus, although they bear different names in different forms. In mammals the lateral lemniscus terminates in part in the inferior colliculus, but in part in the medial geniculate nucleus of the thalamus. From the latter arise the auditory projection fibers to the cortex. On the basis of its fiber connections and relations with the inferior colliculus, the frontal part of the nucleus isthmi of lower forms (sometimes all of the nucleus) has been regarded by certain observers as the possible homologue of a differentiated portion of the nuclei of the lateral lemniscus of higher forms. These homologies are discussed on page 1170 to which reference is made here. The caudal magnocellular part of the nucleus isthmi (perhaps all of the nucleus) of the Sauropsida may be represented by a nuclear group lying lateral to the lateral lemniscus in the pre-trigeminal region of the mammalian medulla oblongata.

It will be seen, then, that the differences in central connections and nuclear masses within the acoustic system are to be explained in the various forms on the basis of the relative number of cochlear fibers and their relative complexity. The pattern is similar throughout and the steps in the development are unusually clear. One further point should receive consideration, and this point relates to the more ventral entrance of the cochlear fibers in mammals. This is probably due to the ventral shift of the nucleus (dorsalis) magnocellularis under neurobiotactic influences so that the termination of the root becomes considerably more ventral.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

503

BIBLIOGRAPHY

GENERAL INTRODUCTION

Benjamins, C. E. 1913. Bcitrag zur Kpnntnis dcs hautigen Labyrinthes. Ueber cine vierle Crista acustica. Zcitschr. f. Ohrenh. Bd. 68, S. 101.

CoGHiLL, G. E. 1916. Correlated anatomical and physiological studies of the growth of the nervous system of Amphibia. II. The afferent system of the head of Amblj-stoma. J. Comp. Neurol., %'ol. 26, p. 247.

Gegenbaur, C. 1898. Vergleichcnde Anatomic der Wirbelthiere mit Beriicksichtigung der Wirbellosen. Bd. 1. W. Engclmann, Leipzig.

Gray, A. A. 1907. The labyrinth of animals. J. and A. Churchill, London.

. 1908. An investigation on the anatomical structure and relationships of the

labyrinth in the reptile, the bird, and the mammal. Proc. Roy. Soc., London, Ser. B, vol. 80, p. 507.

VON Helmholtz, H. L. F. 1865. Die Lehre von den Tonempfindungen als physiologische Grundlage fur die Theorie der Musik. F. Vieweg u. Sohn, Braunschweig.

HciFER. 1907. Ueber die Hautsinnesorgane der Fische. Berichte d. Bayer, biol. Versuchs-station, Munchen, Bd. 1.

Johnson, S. 1917. Structure and development of the lateral canal system of selachians (Mustclus canis and Squalus acanthias). J. Comp. Neurol., vol. 28, p. 1.

Johnston, J. B. 1906. The ner\'ous system of vertebrates. P. Blakiston’s Son and Co., Philadelphia.

VON Kupffer, C. 1891. The development of the cranial nerves of vertebrates. J. Comp. Neurol., vol. 1, p. 246.

Landacre, F. 1910. The origin of the cranial ganglia in Ameiurus. J. Comp. Neurol., vol. 20, p. 309.

. 1926. The primitive lines of Amblystoma Jeffersonianum. J. Comp. Neurol.,

vol. 40, p. 471.

Landacre, F., and Conger, A. L. 1913. The origin of the lateral-line primordia in Lepidosteus osseus. J. Comp. Neurol., vol. 23, p. 607.

Leydig, F. 1850. Ueber die Schleimkanale der Knochenfische. Arch. f. Anat., Physiol, u. wissensch. Med. (IMuller’s), S. 170.

. 1868. Ueber Organe eines sechsten Sinnes. Nova Acta Acad. C. L. Nat. Curio,

vol. 34 (Quoted from Johnson T7).

Lorenzini, S. 1678. Osser\"azloni intomo alle Torpedini. Firenze. (Quoted from Johnson ’17.)

Mangold, E. 1913. Gehorsinn und statischer Sinn. B. Gehorssinn und statischer Sinn bei Wirbeltieren. H. Winterstcin’s Handbuch der vergleichenden Physiologie, Bd. 4, S. 907. G. Fischer, Jena.

Mullenix. 1909. The peripheral terminations of the eighth cranial nerve in vertebrates, especially in fishes. Bull. Mus. Comp. Zool., Harvard, vol. 53.

Parker, G. H. 1902. Hearing and allied senses in fishes. Bull. U. S. Bureau Fisheries, vol. 22, p. 45. Washington, 1903.

. 1904. The function of the lateral-line organs in fishes. Bull. U. S. Bureau

Fisheries, vol. 24, p. 183. Washington, 1905.

. 1908. Structure and functions of the ear of the Squeteague. Bull. U. S. Bureau

Fisheries, vol. 28, pt. 2, p. 1211. Washington, 1910.

. 1908. The sensory reactions of Amphioxus. Proc. Am. Acad. Arts and Sc.,

vol. 43, p. 415.

. 1910. Sound as a directing influence in the movements of fishes. Bull.

U. S. Bureau Fisheries, vol. 30, p. 97. Washington, 1912.

504 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Parker, G. H. 1910. Influence of the eyes, ears, and other allied sense organs on the movements of the dog fish, Mustelus cams. Bull. U. S. Bureau Fisheries, vol. 29, p. 43.

Piper, H. 1906. Aktionsstrome vom Gehororgan der Fische bei Schallreizung. Zentralbl. f. Physiol., Bd. 20, S. 293.

Platt, J. B. 1894. Ontogenetische Differenzierung des Ektodermis in Necturus. I. Studie. Arch. f. mikr. Anat., Bd. 43, S. 911.

. 1896. Ontogenetic differentiation of the ectoderm in Necturus. Study II.

On the development of the peripheral nervous system. Quart. J. Micr. Sc., N. S., vol. 38, no. 152, p. 485.

Retzius, G. 1881. Das Gehororgan der Wirbelthiere. I. Das Gehororgan der Fische und Amphibien. Kongel. Boktr., Stockholm.

. 1884. Das Gehororgan der Reptilien, Vogel und Saugetiere. Stockholm.

. 1892. Die Endigungsweise des Gehornerven. Biol. Untersuch., N. F., Bd. 3, S. 29.

. 1905. Ueber die Endigungsweise des Gehornerven in den Maculae und Cristae

acusticae im Gehorlabyrinth der Wirbeltiere. Eine historisch-kritische Uebersicht. Biol. Untersuch., N. F., Bd. 12, S. 21.

ScHEPMAN, A. M. H. 1918. De Octavo-laterale zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

Schulze, F. E. 1861. Ueber die Nervenendigung in den sogenannten Schleimkanalen der Fische und fiber entsprechende Organe der durch Kiemenathmenden Amphibien. Arch. f. Anat., Physiol, u. wissensch. Med., S. 759.

. 1870. Ueber die Sinnesorgane der Seitenlinie bei Fischen und Amphibien.

Arch. f. mikr. Anat., Bd. 6, S. 62.

Shambaugh, G. E. 1907. A restudy of the minute anatomy of structures in the cochlea with conclusions bearing on the solution of the problem of tone perception. Am. J. Anat., vol. 7, p. 245.

Stone, L. S. 1922. Experiments on the development of the cranial ganglia and the lateral-line sense organs in Amblystoma punctatum. J. Exper. Zook, vol. 35, no. 4, p. 421.

. 1928. Primitive lines in Amblystoma and their relations to the migratory

lateral-line primordia. J. Comp. Neurol., vol. 45, p. 169.

. 1928a. Experiments on the transplantation of placodes of the cranial ganglia

in the emphibian embryo. III. Preauditory and postauditory placodal material interchanged. J. Comp. Neurol., vol. 47, p. 117.

. 1928. Experiments testing the phenomenon of growth direction of migratory

lateral line primordia in embryos of Amblystoma. Abstract in Proc. A. Am. Anatomists, Anat. Rec., vol. 38, p. 30.

. 1929. Experiments on the transplantation of placodes of the cranial ganglia

in the amphibian embryo. IV. Heterotopic transplantations of the postauditory placodal material upon the head and body of Amblystoma punctatum. J. Comp. Neurol., vol. 48, p. 311.

Wittmaack, K. 1909. Zur Frage der Schiidigung des Gchororgans durch Schallcinwirkung. Zcitschr. f. Ohrenh., Bd. 59, S. 211.

Yoshi, U. 1909. Expcrimentelle Untersuchungen fiber die Schiidigung des Gehororgnns durch Schalleinwirkung. Zcitschr. f. Ohrenh., Bd. 58, S. 201.

cvclostomes

Ahlborn, F. 1883. Untcr.suchungen fiber das Gchirn der Petromyzonton. Zcitschr. f. wissensch. Zook, Bd. 39, S. 191.

AriKns KAPPims, C. U. 1920-21. Vergicichende Anatomic des Ncrv'cnsystcms. (German edition of present text.) E. F. Bohn, Haarlem.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

505

A-i-EKS, H., AND WonTHiNGTON, J. 1907. The skin end-organs of the trigeminus and lateralis nerves of Bdellostoma Dombeyi. Am. J. Anat., vol. 7, p. 327.

. 190S. The finer anatomy of the brain of Bdellostoma Dombeyi. I. The

acouslico-lateral system. Am. J. Anat., vol. 8, p. 1.

Holm, J. 1901. The finer anatomy of the nervous system of M 3 odne glutinosa. Morphol. Jahrb., Bd. 29, S. 364.

Jansen, J. 1930. The brain of My'xine glutinosa. J. Comp. Neurol., vol. 49, p. 359.

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

Pe-ahson, a. a. 1933. The acoustico-lateral centers and the cerebellum, Mth connections, of fishes. Dissertation.

Sanders, A. 1894. Researches in the nenmus system of Myxine glutinosa. (Quoted from Jansen ’30.)

Schilling. 1907. Ueber das Gehim von Petromy'zon fluviatilis. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main, Bd. 30, S. 425.

Tretjakoff, D. 1908. Die peripherische und zentrale Endigung des Gehomerven bei Ammocoetes und Petromyzon fluviatilis. Folia neuro-biol., Bd. 1, S. 14.

. 1909. Das Nen-ensystem von Ammocoetes. II. Gehim. Arch. f. mikr.

Anat., Bd. 74, S. 636.

PLAGIOSTOMES

Allis, E. P. 1901. The lateral sensory canals, the eye-muscles, and the peripheral distribution of certain of the cranial nen-es of Mustelus laevis. Quart. J. Micr. Sc., vol. 45, p. 87.

AriEns ICappers, C. U. 1906. The structure of the teleostean and selachian brain. J. Comp. Neurol., vol. 16, p. 1.

Benjamins, C. E. 1913. Beitrag zur Kenntnis des hautigen Labyrinthes. Ueber eine vierte Crista acustica. iJeitschr. f. Ohrenh., Bd. 68, S. 101.

Ewart, J. C. 1892. The lateral sense organs of Elasmobranchs. I. The sensory canals of Laemargus. Tr. Roy. Soc., Edinburgh, vol. 37, p. 59.

Ewart, J. C., and Mitchell, J. C. 1892. On the lateral sense organs of Elasmobranchs. II. The sensoiy canals of the common skate (Raia batis). Tr. Roy. Soc., Edinburgh, vol. 37, p. 87.

Houser, G. L. 1901. The neurones and supporting elements of the brain of a selachian. J. Comp. Neurol., vol. 11, p. 65.

Johnson, S. 1917. Structure and development of the sense organs of the lateral canal system of selachians (Mustelus canis and Squalus acanthias). J. Comp. Neurol,, vol. 28, p. 1.

Klinkhardt, K. 1905. Beitrage zur Entwickelungsgeschichte der Kopfganglien und Sinneslinien der Selachier. Jenaische Zeitschr. f. Naturw., Bd. 40, S. 423.

Schepman, a. M. H. 1918. De octavo-laterale zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

Sterzi, G. 1909-12. II sistema nervoso centrale dei vertebrati ; ricerche anatomiche ed embriologiche. Vol. 2, Pesci ; pt. 1, Selaci. A. Draghi, Padova.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehims der Teleostier und Selachier, Anat. Anz., Bd. 31, S. 369.

GANOIDS AND TELEOSTS

Allis, E. P. 1889. The anatomy and development of the lateral line system in Amia calva. J. Morphol., vol. 2, p. 463.

Ariens Kappers, C. U, 1906. The structure of the teleostean and selachian brain. J. Comp. Neurol., vol. 16, p. 1.

506 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ariens Kappers, C. U. 1907. Das Gehim des Knochenganoiden Amia calva und Lepidosteus osseus. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

. 1929. The evolution of the nervous system. E. F. Bohn, Haarlem.

Bartelmez, G. W. 1915. Mauthner’s cell and the nucleus motorius tegmenti. J. Comp. Neurol., vol. 25, p. 87.

Bartelmez, G. W., and Hoerr, N. L. 1933. The vestibular club endings in Ameiurus. J. Comp. Neurol., vol. 57, p. 401.

Beccari, N. 1907. Ricerche sulle cellule e fibre del Mauthner e sulle loro connessioni in pesci ed anfibii (Salmo fario, S. irideus e Salamandrina perspicillata). Arch. ital. di anat. e di embriol., vol. 6, p. 660.

. 1931. Studi comparativi sopra i nuclei terminali del nervo acustico nei pesci.

Arch. zool. Ital., vol. 16, p. 732.

Benjamins, C. E. 1913. Beitrag zur Kenntnis des hautigen Labyrinthes. Ueber eine vierte Crista acustica. Zeitschr. f. Ohrenh., Bd. 68, S. 101.

Berkelbach van der Sprenkel, H. 1915. The central relations of the cranial nerves in Silurus glanis and Mormyrus caschive. J, Comp. Neurol., vol. 25, p. 5. Bonnier, P. 1896. Sur le sous lateral. Compt. rend. Soc. de biol., S6r. 10, vol. 3, p. 917.

Bugnion, E. 1873. Bull. Soc. Vaudoise Sci. Nat., s^r. 2, vol. 22, p. 259.

DE Burlet, M. 1924. Einiges fiber den perilymphatischen Raun und das Labyrinth. Nederlandsch. tijdschr. f. geneesk., Bd. 68, p. 1228.

. 1929. Anatomisches zur Horfahigkeit der Siluroiden. Zeitschr. f, Anat. u.

Entwicklungsch., Bd. 89, p. 11.

. 1929a. Zur vergleichenden Anatomie der Labyrinthinnervation. J. Comp.

Neurol., vol. 47, p. 155.

Burr, H. S. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol., vol. 45, p. 33.

Edinger, L. 1908. Vorlesungen fiber den Bau der nervosen Centralorgane des Menschen und der Thiere. Teil 2. 7" Aufl., F. C. W. Vogel, Leipzig.

Franz, V. 1911. Das Mormyridenhim. Zool. Jahrb., Abt. f. Anat., Bd. 32, S. 465. Fuchs, S. 1894. Ueber der unter der Haut liegenden Canalsysteme bei den Selachiem.

Arch. f. d. ges. Physiol. (Pflfiger’s), Bd. 59, S. 454.

Herrick, C. J. 1899. The cranial and first spinal nerves of Menidia: a contribution upon the nerve components of the bony fishes, J. Comp. Neurol., vol. 9, p. 153.

Hocke Hoogenboom, K. J. 1929. Das Gehim von Polyodon folium Lacep. Jahrb. f. Morphol. u. mikr. Anat, ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 311.

Hofer. 1907. Ueber die Hautsinnesorgane der Fische. Berichte d. Bayer, biol. Versuch-s-station, Mfinchen, Bd. 1.

Holmgren, N., and van der Horst, C. J. 1925. Contribution to the morphology of the brain of Ceratodus. Acta Zool., Bd. 6, S. 59.

VAN DER Horst, C. J. 1918. Die motorischen Kerne und Bahnen in dem Gehim der Fische, ihr taxonomischer Wert und ihre neurobiotaktische Bedeutung. Tijdschr. d. Ned. Dierk. Vereen., vol. 16, p. 168.

Johnston, J. B. 1901. The brain of Acipenser; a contribution to the morphology of the vertebrate brain. G. Fischer, Jena.

Kreidl, a. 1895. Ueber die Schallperception der Fische. Arch. f. d. ges. Physiol. (Pfluger’.s), Bd. 59, S. 450.

. 1896. Ein weiterer Versuch fiber das angebliche Horen eines Glockenzeichens

durch die Fische. Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 58, S. 581.

THE LATERAL-LINE AND ACOUSTIC SYSTEMS

507

Landacre, F. L., and Conger, A. C. 1913. The origin of the lateral line primordia in Lepidosteus osseus. J. Comp. Neurol., vol. 23, p. 575.

Lee, F. S. 1898. The functions of the ear and the lateral line in fishes. Am. J. Physiol., vol. 1, p. 128.

Leydig, F. 1894. Integument and Hautsinnesorgane der Knochenfische. Zool. Jahrb., Abt. f. Anat., Bd. 8, S. 1.

Marui, K. 1918. On the finer structure of the synapse of the Mauthner cell with especial consideration of the " Golgi net ” of Bethe, nervous terminal feet and the nenmus pericellular terminal net of Held. J. Comp. Neurol., vol. 30, p. 127.

. 1919. The effect of over-activity on the morphological structure of the synapse.

J. Comp. Neurol., vol. 30, p. 253.

Mayser, P. 1881. Vergleichende anatomische Studien fiber das Gehirn der Knochenfische mit besonderer Berficksichtigung der Cyprinoiden. Zeitschr. f. wissensch. Zool, Bd. 36, S. 259.

Parker, G. H. 1905. The function of the lateral line organ in fishes. Bull. U. S. Fisheries, vol. 24, p. 183.

. 1908. Structure and functions of the ear of the Squeteague. Bull. U. S.

Fisheries, vol. 28, p. 1213.

. 1918. A critical survey of the sense of hearing in fishes. Proc. Am. Philos. Soc.,

vol. 57, p. 69.

Parker, G. H., and van Heusen, A. P. 1917. The responses of the catfish, Amiurus nebulosus, to metallic and non-metallic rods. Am. J. Physiol, vol. 44, p. 405.

. 1917a. Reception of mechanical stimuli by the skin, lateral-line organs and ears

of fishes, especially in Amiurus. Am. J. Physiol, vol. 44, p. 463.

Pearson, A. A. 1933. The acoustico-lateral centers and the cerebellum, with fiber connections, of fishes. Dissertation.

PfOller, a. 1914. Beitrage zur Kenntnis der Seitensinnesorgane und Kopfanatomie der Macruriden. Jenaische Zeitschr. f. Naturw., Bd. 52, S. 259.

Piper, H. 1906. Aktionsstrome von Gehororgan der Fische bei Schallreizung. Zentralbl. f. Physiol, Bd. 20, S. 293.

Ram6n y Cajal, S. 1908. Sur un noyau special du nerf vestibulaire des poissons et des oiseaux. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 6, p. 1,

Retzius, G. 1881. Das Gehororgan der Wirbelthiere. I. Das Gehororgan der Fische und Amphibien. Kongel Boktr., Stockholm.

ScHEPMAN, A. M. H. 1918. De octavo-laterale zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

VAN DER Sprenkel. See Berkelbach van der Sprenkel.

Stendell, W. 1914. Die Faseranatomie des Mormyridengehirns. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main.

. 1914a. Morphologische Studien an Mormyriden. Verhandl. d. deutsch. zool.

Gesellsch., Bd. 24, S. 254.

Tbllo, F. 1909. Contribucifin al conocimiento del enc4falo de los Teleoste6s. Los nucleos bulbares. Trab. d. lab. de invest, biol. univ. de Madrid, vol. 7, p. 1.

Wallenberg, A. 1907. Beitrage zur Kenntnis des Gehirns der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

AMPHIBIANS

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Ariens Kappers, .C. U., and Hammer, E. 1918. Das Zentralnervensystem des Ochsenfrosches (Rana catesbyana). Psychiat. en neurol. Bl, Amsterdam, Bd. 22,

S. 368.

508 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Beccari, N. 1908. Ricerche sulle cellule e fibre del Maulhner c sullc loro conne.s.sioni in pesci ed anfibii (Salmo fario, S. irideu.s e Salamandrina penspicillata). Arch. ital. di anat. e di embriol., vol. 6, p. 660.

Benjamins, C. E. 1913. Beitrag zur Kenntnis dc.s hiiutigcn Labyrinthe.s. Ueber eine vierte Crista acustica. Zeitschr. f. Ohrcnh., Bd. 68, S. 101.

DE Burlet, M. 1929. Zur vergleichcnde Anatomic der Labyrinthinnervation. J. Comp. Neurol., vol. 47, p. 155.

Charipper, H. a. 1928. Studies on the lateral-line system of Amphibia. I. Cytology and innervation of the lateral-line organs of Necturus maculosus. J. Comp. Neurol., vol. 44, p. 425.

Chezar, H. H. 1930. Studies on the lateral-line system of Amphibia. II. Comparative cytology and innervation of the lateral-line organs in Urodela. J. Comp. Neurol., vol. 50, p. 159.

CoGHiLE, G. 1902. The cranial nerves of Amblystoma tigrinum, J. Comp. Neurol., vol. 12, p. 205.

Deganello, U. 1906. Degenerazioni nel nevrasse della Rana, consecutive all’asportazione del labirinto dell’orecchio. Atti d. r. 1st. Veneto, vol. 65, pt. 2, p. 829. Venezia.

Dempster, W. 1930. The morphology of the amphibian endolymphatic organ. J. Morphol., vol. 50, p. 71.

Ecker, a. WiEDERSHEiM, R., AND Gaupp, E. 1894. Anatomic des Frosches. Teil 2. F. Vieweg u. Sohn, Braunschweig.

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7'® Aufl., Teil 2. F. C. W. Vogel, Leipzig.

Fineman, G. 1915. Beitrage zur Kenntnis der Entwicklung des Ductus endolymphaticus bei dem Menschen und einigen Wirbeltieren. Anat. Hefte, Bd. 53, S. 1.

Gatipp, E. See Ecker, Wiedersheim, and Gaupp.

Herrick, C. J. 1914. The medulla oblongata of larval Amblystoma. J. Comp. Neurol., vol. 24, p. 343.

. 1930. The medulla oblongata of Necturus. J. Comp. Neurol., vol. 50, p. 1 Holmes, G. 1903. On the comparative anatomy of the nerv'us acusticus. Tr. Roy. Irish Acad., vol. 32, sect. B, p. 101.

Kingsbury, B. F. 1895. On the brain of Necturus maculatus. J. Comp. Neurol., vol. 5, p. 139.

Larsell, 0. 1923. The nucleus isthmi of the frog. J. Comp. Neurol., vol. 36, p. 309.

Lewis, W. H. 1907. On the origin and differentiation of the otic vesicle in amphibian embryos. Anat. Rec., vol. 1, p. 141.

Michl, R. 1925. Ueber die Entwicklung des Gehorlabyrinthes bei Rana temporaria. Spisy Rkarske fakulty Mosorykovy univ. v. Brue, vol. 3, no. 7. (Quoted from Dempster, 1930.)

Moodie, R. L. 1915. A further contribution to a knowledge of the lateral line system in extinct Amphibia. J. Comp. Neurol., vol. 25, p. 317.

Norris, H. W. 1911. The rank of Necturus among the tailed amphibians as indicated by the distribution of its cranial nerves. Proc. Iowa Acad. Sc., vol. 18, p. 137.

. 1913. The cranial nerves of Siren lacertina. J. Morphol., vol. 24, p. 245.

Osborn, H. 1888. A contribution to the internal structure of the amphibian brain. J. Morphol., vol. 2, p. 1.

Papez, j. W. 1929. Comparative neurology. Thomas Crowell Co., New York. Retzius, G. 1881. See bibliography for ganoids and teleosts.

Rothig, P. 1927. Beitrage zum Studium des Zentralnervensystpms der Wirbeltiere. No. 11. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 10, S. 381.

THE LATERAE-LINE AND ACOUSTIC SYSTEMS 509

ScHEPiiAN, A. M. H. 1918. De octavo-laterale zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

Strong, 0. S. 1895. The cranial nerves of Amphibia. J. MorphoL, vol. 10, p. 101. Wallenberg, A. 1907. Die kaudale Endigung der bulbo-spinalen Wurzeln des Trigeminus, Vestibularis und Vagus beim Frosche. Anat. Anz., Bd. 30, S. 564. Whiteside, B. 1922. Development of the endoljTnphatic sac in the frog. Am. J. Anat., vol. 30, p. 231.

REPTILES

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Beccari, N. 1912. La costituzione, i nuclei terminali, e le vie di connessione del nervo acustico nella Lacerta muralis, Merr. Arch. ital. di anat. e di embriol., vol. 10 p. 646.

Benjamins, C. E. 1913. Beitrag zur Kenntnis des hautigen Labyrinthes. Ueber eine vierte Crista acustica. Zeitschr. f. Ohrenh., Bd. 68, S. 101.

DE Bdrlet, H. M. 1929. Zur vergleichenden Anatomie der Labyrinthinnervation. J. Comp. Neurol., vol. 47, p. 155.

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7 ^ Aufl., Bd. 2, F. C. W. Vogel, Leipzig.

Gray, A. A. 1907. The labyrinth of animals. J. and A. Churchill, London.

Hindenach, j. C. R. See bibliography for reptiles. Chapter VII.

Holmes, G. 1903. On the comparative anatomy of the nervus acusticus. Tr. Roy. Irish Acad., vol. 32, sect. B, p. 101.

Huber, G. Carl, and Crosby, E. C. 1926. On thalamic and tectal nuclei and fiber paths in the brain of the American alligator. J. Comp. Neurol., vol. 40, p. 97.

JousTRA, N. 1918. Over de Homologie van het Ganglion Isthmi. Psychiat. en neurol. Bl., Amsterdam, vol. 22, p. 361.

DE Lange, S. J. 1917. Das Hinterhim, das Nachhim und das Riickenmark der Reptilien. Folia neuro-biol., Bd. 10, S. 385.

Larsell, O. 1932. The cerebellum of reptiles: chelonians and alligator. J. Comp. Neurol., vol. 56, p. 299.

Retzius, G. 1884. Das Gehororgan der Reptilien, Vogel und Saugetiere. Stockholm.

ScHEPMAN, A. M. H. 1918. De octavo-laterale zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

Shanklin, W. M. 1930. The central nervous system of Chameleon vulgaris. Acta ZooL, Bd. 11, S. 425.

Weston, J. K. 1933. The reptilian vestibular and cerebellar gray with fiber connections. Dissertation.

BIRDS

Ariens ICappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems.

(German edition of present text.) E. F. Bohn, Haarlem.

Bartels, M. 1925. Ueber die Gegend des Deiters und Bechterewskernes der Vogeln. Zeitschr. f. d. ges. Anat. ; Abt. 1, Zeitschr. f. Anat. u. Entwicklungsgesch., Bd. 77, S. 726.

Benjamins, C. E. 1913. Beitrag zur Kenntnis des hautigen Labyrinthes. Ueber eine vierte Crista acustica. Zeitschr. f. Ohrenh., Bd. 68, S. 101.

Bok, S. T. 1915. Die EntMcklung der Himnen-en und ihrer zentralen Bahnen. ‘ Die stimulogene Fibrillation. Folia neuro-biol., Bd. 9, S. 475.

Brandis, F. 1894. Untersuchungen fiber das Gehim der Vogel. II. Theil : Ursprung der Nen^en der Medulla oblongata. Arch. f. mikr. Anat., Bd. 43, S. 96.

510 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Craigie, E. H. 1928. Observ'ations on the brain of the hummingbird (Chrysolampis mosquitus Linn, and Chlorostilbon caribaeus Lawr.). J. Comp. Neurol., vol. 48, p. 377.

Edinger, L. 1908. Vorlesungen tiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7** Aufl., Bd. 2, F. C. W. Vogel, Leipzig.

Gray, A. A. 1907. The labyrinth of animals. J. and A. Churchill, London.

Groebbels, F. 1927. Die Lage- und Bewegungsreflexe der Vogel. VI. Mitteilung. Degenerationsbefunde im Zentralnervensystem der Taube nach Entfernung des Labyrinths und seiner Teile. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 218, S. 89.

. 1927. Idem. VII. Mitteilung. Die Wirkung zweizeitiger Labyrinthopera tionen auf die Lage- und Bewegungsreflexe der Haustaube. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 218, S. 408.

. 1928. Idem. XI. Mitteilung. Die Analyse der Stiitzreaktion. (Mit Unter stiitzung der Notgemeinschaft der Deutschen Wissenschaft.) Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 221, S. 50.

Holmes, G. 1903. On the comparative anatomy of the nervus acusticus. Tr. Roy. Iri.sh Acad., vol. 32, sect. B, p. 101.

Mesdag, T. M. 1909. Bijdrage tot de ontwikkelingsgeschiedenis van de structuur der hersenen bij het kip-embryo. Inaugural Dissertation, Groningen.

Muskens, L. j. j. 1930. On tracts and centers involved in the upward and downward associated movements of the eyes after experiments in birds. J. Comp. Neurol.,, vol. 50, p. 289.

Ram6n y Cajal, S. 1908. Los ganglios centrales del cerebelo de las aves. Trab. d. lab. de invest, biol. Univ. de Madrid, vol. 6, p. 177.

. 1908a. Les ganglions terminaux du nerf acoustique des oiseaux. Trab. d. lab.

de invest, biol. Univ. de Madrid, vol. 6, p. 195.

. 1909-11. Histologic du syst4me nerveux de I’homme et des vertdbrds. Vol. I,

1909 ; vol. II, 1911. A. Maloine, Paris.

Retzius, G. 1884. Das Gehororgan der Reptilien Vogel und Saugetierc. Stockholm.

Sanders, E. B. 1929. A consideration of certain bulbar, midbrain, and cerebellar centers and fiber tracts in birds. J. Comp. Neurol., vol. 49, p. 155.

ScHEPMAN, A. M. H. 1918. De octavo-lateralc zintuigen en hun verbindingen in de hersenen der vertebraten. Dissertatie, Amsterdam.

Sinn, R. 1913. Beitrag zur Kenntnis der Medulla oblongata der Vogel. Monatschr. f. Psychiat. u. Neurol., Bd. 33, S. 1.

Wallenberg, A. 1898. Die sccundare Acusticusbahn der Taube. Anat. Anz., Bd. 14, S. 353.

. 1900. Ueber centrale Endstiltten des Nervus octavus der Taube. Anat. Anz.,

Bd. 17, S. 102.

Winkler, C. 1907. The central course of the nervus octavus and its influence on motility. Verhandl. d. kon. Akad. v. AVetensch. te Amsterdam, Tweedc Scctic, Decl 14, no. 1.

Yosiiida, I. 1924. Ein Beitrag zur Kenntnis der zentralen Aestibularisbahn. Folia Anat. Japon., Bd. 2, S. 283.

. 1925. Ueber die funktionelle Bedcutung der oberen Olive nebst ihrer Fascr bahnen. Folia Anat. Japon., vol. 3, S. 111.

MAMMALS

Alexander, G. 1899. Zur Anatomic des Ganglion vestibularc der Saugethicre. Sitzungsb. d. K. Akad. d. wissensch. zu Wien, Math.-nat. Cl., Bd. 108, Abt. 3, S. 449. •

. 1904. Entwicklung und Bau des inneren Gehororgans van Echidna. Semons

Forschung-sreisen., Bd. 3, Tcil 2.

THE LATERAI^LINE AND ACOUSTIC SYSTEMS

511

Alexander, G. 1905. Zur Frage der phylogenetischen vikariierenden Ausbilcung der Sinnesorgane. Zeitschr. f. Psychol, u. Physiol, d. Sinnesorg., Bd. 38, S. 24.

Ariens ICappers, C. U. 1920-21. Vergleichende Anatomic des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Baginskt, B. 1886. Ueber den TJrsprung und den centralen Verlauf des Nervus acusticus des Kaninchens. Virchow’s Arch. f. path. Anat., Bd. 105, S. 28.

. 1890. Ueber den Ursprung und den centralen Verlauf des Nervus acusticus des

Kaninchens und der Katze. Virchow’s Arch. f. path. Anat., Bd. 119, S. 81.

BXrAny , R., AND WiTTMAACK, K. 1911. Funktionellc Priifung des Vestibularapparates. Verhandl. d. deutsch. otolog. Gesellsch., Bd. 20, S. 37.

VON Bechterew, W. 1887. Zur Frage iiber den Ursprung des Gehornervens und Tiber die physiologische Bedeutung des N. vestibularis. Neurol. Centralbl., Bd. 6, S. 193.

Benjamins, C. E. 1913. Beitrag zur Kenntnis des hautigen Labyrinthes. Ueber eine vierte Crista acustica. Zeitschr. f. Ohrenh., Bd. 68, S. 101.

Brouwer, B. 1910. In Quix and Brouwer’s IBeitrag zur Anatomic der kongenitalen Taubstummheit. J. F. Bergmann, Wiesbaden.

. 1912. Das Gehirn einer congenital tauben Katze. Folia neuro-biol., Bd. 6,

S. 197.

Brouwer, B., and van Walree, D. 1914. Ueber den Hirnstamm eines Taubstummen. Folia neuro-biol., Bd. 8, S. 589.

DE Buhlet, H. M. 1929. Zur vergleichenden Anatomic der Labyrinthinnervation. J. Comp. Neurol., vol. 47, p. 155.

Camis, M. 1930. The physiology of the vestibular apparatus. Oxford Univ. Press.

Cramer, J. B. 1894. Beitrage zur feineren Anatomie der Medulla oblongata und der Briicke mit besonderer Berilcksichtigung des 3-12 Hirnnerven. G. Fischer, Jena. (Quoted from Ram6n y Cajal ’09, p. 760.)

Deiters, 0. 1865. Untersuchungen iiber Gehirn und Riickenmark des Menschen und der Saugetiere. Braunschweig.

Edinger, L. 1908. Vorlesungen iiber den Bau der nervosen Centralorgane des Menschen und der Thiere. 7*® Aufl., F. C. W. Vogel, Leipzig.

Ewald, j. R. 1887. Zur Physiologic der Bogengiinge. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 41, S. 463.

. 1889. Zur Physiologic der Bogengange. Fortsetzung: Ueber Bewegungen

der Perilymphe. Arch. f. d. ges. Physiol. (Pfliiger’s), Bd. 44, S. 319.

. 1892. Physiologische Untersuchungen uber das Endorgan des Nervus octavus.

J. F. Bergmann, Wiesbaden.

Fischer, M. H. 1928. Die Regulationsfunktionen des menschlichen Labyrinthes und die Zusammenhange mit verwandten Funktionen. Ergeb. d. Physiol., Bd. 27, S. 209.

Flechsig, P. E. 1890. Weitere Mittheilungen iiber die Beziehungen des unteren Vierhiigels zum Homerven. Neurol. Centralbl., Bd. 9, S. 98.

Fraser, E. H. 1902. An experimental research into the relations of the posterior longitudinal bundle and Deiters’ nucleus. J. Physiol., vol. 27, p. 372.

Fuse, G. 1911. Striae acusticae v. Monakowi beim Menschen. Rev. di patol. nerv. e ment., vol. 30, p. 912.

. 1912. Striae medullares acusticae (Piccolohomini). Rev. di patol. nerv. e

ment., vol. 31, p. 463.

. 1913. Das Ganglion ventrale und das Tuberculum acusticum bei eimgen Sauge tieren und beim Menschen. Arb. a. d. himanat. Inst, in Zurich, Bd. 7, S. 1.

VAN Gehuchten, a. 1902. Recherches sur la voie acoustique centrale. Le Ndvraxe,

vol. 4, p. 253.

■ . 1903. La deg^ndrescence dite retrograde.

Le Ndvraxe, vol. 5, p. 1.

512 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

VAN Gehuchten, a. 1904. Connexions centrales du noyau de Deiters et des masses grises voisines. Le N4vraxe, vol. 6, p. 21.

. 1907. Anatomie du systSme nerveux de I’homme. edition. A. Uys pruyst-Dieudonn6, Louvain.

Gray, A. A. 1907. The labyrinth of animals. J. and A. Churchill, London.

. 1908. An investigation on the anatomical structure and relationships of the

labyrinth in the reptile, the bird, and the mammal. Proc. Roy. Soc., London, Ser. B., vol. 80, p. 507.

Gray, L. P. 1926. Some experimental evidence on the connections of the vestibular mechanism in the cat. J. Comp. Neurol., vol. 41, p. 319.

Griffith, C. R. 1922. An historical survey of vestibular equilibration. Bull. TJniv. Illinois, vol. 20, no. 5, p. 1.

. 1929. Vestibular sensations and the mechanisms of balance. Psychol. Bull.,

vol. 26, p. 549.

Guild, S. R. 1921. A graphic reconstruction method for the study of the organ of Corti. Anat. Rec., vol. 22, p. 141.

, 1924. Experimental evidence on the circulation of the endolymph in mammals.

Anat. Rec., vol. 27, p. 205.

. 1925. Structure and normal contents of the ductus and saccus endolymphaticus

in the guinea pig. Anat. Rec., vol. 29, p. 358.

Hardesty, I. 1908. On the nature of the tectorial membrane and its probable rfile in the anatomy of hearing. Am. J. Anat., vol. 8, p. 109.

. 1915. On the proportions, development, and attachment of the tectorial membrane. Am. J. Anat., vol. 18, p. 1.

Harrison, H. S. 1903. The homology of the lagena throughout vertebrates. Anat. Anz., Bd. 23, S. 627.

Hashimoto, T. 1928. Experimentelle Untersuchungen iiber die spinale Bahn des N. vestibularis, mit besonderer Beriicksichtigung auf dem Gcbietc des Deiters dcscendens et ascendens Winklers. Folia anat. Japon., Bd. 6, S. 537.

Held, H. 1891. Die zentralen Bahnen des Nervus acusticus bei der Katze. Arch, f. Anat. u. Physiol., Anat. Abt., S. 271.

. 1893. Die centrale Gehorleitung. Arch. f. Anat. u. Physiol., Anat. Abt.,

S. 201.

. 1897. Beitragc zum Studium der Nervenzellen und ihre Fortsatze. Arch. f.

Anat. u. Physiol., Anat. Abt., S. 204.

. 1897a. Zur Kenntnis der peripheren Gehorleitung. Arch. f. Anat. u. Physiol.,

Anat. Abt., S. 350.

. 1904. Untersuchungen iiber den feincren Bnu des Gehororgans der Wirbeltiere.

I. Zur Kenntnis des Cortischen Organs und der tibrigen Sinnesapparate des Labyrinthes bei Saugeticren. Abhandl. d. math.-phys. Klasse d. kon. siichs. Gcsellsch. d. Wissensch., Bd. 28, S, 1.

. 1909. Untersuchungen iiber den feineren Bau des Ohrlabyrinthes der Wirbcl thiere. II. Zur Entwicklungsgeschichte des Cortischen Organs und der Macula acustica bei Saugeticren und Vogeln. Abhandl. d. math.-phys. Klasse d. kon. siichs. Gcsellsch. d. Wissensch., Bd. 31, S. 193.

Henson, V. 1907. Die Empfindungsarten des Schalls. Arch. f. d. gcs. Physiol., Bd. 119, S. 249.

His, W. 1888. Zur Geschichtc des Gchirns sowie der centralcn und peripherischen Nervenbahn bcim menschlichcn Embryo. Abhandl. d. math.-phys, Klasse d. kon. .sichs, Ge.«ellsch. d. Wi.«sen.sch. Bd. 14, S. 339.

Hoffman, F. 1908. Die obere Olive der Saugcticrc nebst Bemerkungen iiber die Lagc der Cochlcariscndkeme. Arb. a. d. ncurol. Inst. a. d. Wien. Univ. (Oberstciner's), Bd. 14, S. 76.

THE LATER Aly-LINE AND ACOUSTIC SYSTEMS

513

Hoi.Mr,s, G. 1903. On the comparative anatomy of the nerviis acusticus, Tr. Roy. Irish Acjid., vol. 32, sect. B, p. 101.

Hunr.n, G. Caui,, and Cnosnv, E. C. 1929. Somatic and visceral connections of the dicncephalon. Arch. Neurol, and Rsychiat., vol. 22, p. 187.

. 1930. See Piersol, G. A., Human Anatomy.

Ingvah, S. 1919. Zur Phylo- und Ontogencse des Kleinhirns. Folia neuro-biol., Bd. 11, S. 205.

Ivy, a. C. 1919. E.\pcrimenl.al studic.s on the brain stem. II. Comparative study of the relation of the cerebral corte.v to vestibular nystagmus. J. Comp. Neurol., vol. 31, p. I.

. 1929. The physiology of vestibular nystagmus. Arch. Otolarjmgol.,vol.9,p. 123.

Ka!Da, Y. 1929. Ucl>cr den Ursprung und den peripheren Verlauf dcr sogenannten rentrifug.alcn Vestibularisncrvcn nach Lcidlcr (Fasciculus vestibularis medialis nnch Kaplan). Arch. f. Oh^cn-, Nasen- u. Kehlkopfh., Bd. 123, S. 62.

Kai’d.an, M. 1913. Die spinale Acusiicuswurzel und in ihr eingelagerten Zcllsysteme. Nucleus Deiters — Nucleus Bcchtcrcw. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Oberstcincr’s), Bd. 20, S. 375.

DE Ku'.tj.v, A., and Magnus, R. 1921. Ueber die Funktion dcr Otolithen. I. Otolithenstand bci den tonischen Lab}’rinthrofle.\e. Arch. f. d. ges. Physiol. (Pfliigcr’s), Bd. 186, S. 6.

. 1921a. Idem. II. Isolierte Otolithcnausschaltung bei Mecrschweinchen.

Arch. f. d. gc.s. Physiol. (Pfluger’s), Bd. 126, S. 61.

Klossowsky, B. 1932. Ueber cine bishcr noch nicht bcschricbcnc Zellgruppe im intramcdullaren Toil dcr Wurzcl dcs Nervus vestibularis beim Menschen und bei einigen Saugcticrcn. (Nucleus intraradicularis nervi vestibularis.) Arch. Psychiatr. u. Ncn'cnkrankh., vol. 98, p. 255.

VON KOllikeii, A. 1893. Handbuch dcr Gcwcbclchrc dcs Menschen. 6" Aufi., Bd. 2. W. Engclmann, Ivcipzig, various editions from 1889-’02.

. 1901. Die Medulla oblongata und die Vicrhugelgegend von Omithorhynchus

und Echidna. W. Engclmann, Leipzig.

Kreidl, a., AND Kato, T. 1913. Zur Frage der sekunduren Horbahnen. Folia neurobiol., Bd. 6, S. -195.

Leideer, R. 1913. Expcrimentcllc Untersuchungen iiber das Endigungsgebeit des Nen-us vestibularis. I. Mitteilung. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Oberstcincr’s), Bd. 20, S. 256.

. 1914. Idem. II. Mitteilung. Arb. a. d. neurol. Inst. a. d. Wien. Univ.

(Obcrstcincr’.s), Bd. 21, S. 151.

VON LENiiossfiK, M. 1893. Die Ncia'cnendigungen in den Maculae und Cristae acusticae. Versamml. d. anat. Gesellsch. zu Gottigen (Mai 23). Also in Anat. Hoftc, Bd. 3, S. 229.

Lewandowsky, M. 1904. Untersuchungen iiber die Leitungsbahnen des Truncus cerebri und ihren Zu.sammenhang mil denen der Medulla spinalis und des Cortex cerebri. G. Fischer, Jena.

Lewy, F. H. 1909. Dcgencrationsversuchc am akustischen System des Kaninchens und der Katze. Zugleich ein Beitrag zur Anwendung der Marchischen Methode. Folia neuro-biol., Bd. 2, S. 471.

. 1910. Der Deitersche Kern und das Deitero-spinale Biindel. Arb. a. d. hirna nat. Inst, in Zurich, Bd. 4, S. 227.

Lloto, R. E. 1900. On chromatolysis in Deiters’ nucleus after hemisection of the cord.

J. Physiol., vol. 25, p. 191.

Magnus, R. 1914. Welche Teile des Zcntralnervensystems miissen fiir das Zilstandekommen dcr tonischen Hals- und Labyrinlhreflexe auf die Korpermuskulatur vorhanden sein? Arch. f. d. ges. Physiol. (Pfluger’s), Bd. 159, S. 224.

THE LATERAIv-LINE AND ACOUSTIC SYSTEMS 515

SniEETEU, G. L. 1907. On tlic development of the membranous labyrinth and the 'acoustic and facial ncn’c.s in the human embryo. Am. J. Anat., vol. 6, p. 139.

Tienev, F., ANn Pike, F. H. 1925. Muscular coordination experimentally studied in its relation to the cerelx;llum. Arch. Neurol, and P-sychiat., vol. 13, p. 289.

Veiiatti, E. 1900. Su alcune particolaritil di struttura dei centri acustici nei mammiferi. Pavia.

ViNCENzi, L. 1905. Del nucleo del corpo trapezoide studiato coi metodi di Cajal per Ic neurofibrille. Anat. Anz., Bd. 27, S. 20.

VoiT, M. 1907. Zur Frape der Venlstelung des Nervus acusticus bei den Saugetieren. Anat. Anz.., Bd. 31, S. 035.

Wilson, .1. G., and Pike, F. H. 1911. The effect of stimulation and extirpation of the ear and their relation to the motor system. Phil. Tr. Roy. Soc., London, vol. 203, p. 158.

. 1913. The effects of .stimulation of the car in the living animal. Proc. Soc.

Exper. Biol, and Med., vol. 10, p. 81.

. 1915. The mechanism of labj’rinthine nystagmus. Arch. Int. Med., vol. 15,

p. 31.

WiNKLEn, C. 1907. The central course of the ncr\-us octavus and its influence on motilit}’. Vcrhandel. d. kon. Akad. v. Wetcnsch. te Amsterdam, Tiveede Sectie, Dcel 1*1, no. 1.

. 1908. Hot ccntralc zenuwstelscl cener wittc doofgcborcn kat. Vcrslag. kon.

Akad. v. Wetcnsch. to Amsterdam, Decl 16.

. 1910. Die Folgcn der Abtragung des Tuberculiim acusticum bei junggeborenen

Kaninchen. Folia ncuro-biol., Bd. 3, S. 275.

. 1911. Ex-perimcntcllcr Beitrag zur Kenntniss der .sccundiiren Horbahnen der

Katze. Folia neuro-biol., Bd. 5, S. SG9.

. 1917. Anatomic du systeme nerveux. Vol. 1, E. F. Bohn, Haarlem.

Yosiiida, I. 1924. Ein Beitrag zur Kenntnis der zentralcn Vestibularisbahn. Folia anat. Japon., Bd. 2, S. 283.

. 1925. Ueber die funktioncllc Bedeutung dcr oberen Olive nebst ihrer Faser bahnen. Folia anat. Japon., Bd. 3, S. 111.

Zwaaudemaker, H. 1915. Lccrbock der physiologic. E. F. Bohn, Haarlem. Bd. 2.

CHAPTER V

THE EFFECTORY SYSTEM OF THE MIDBRAIN AND THE MEDULLA OBLONGATA

Before beginning a consideration of the midbrain and medullar motor roots and their nuclei in craniotes, a brief review will be given of the relations of these elements and of their peripheral end-organs, the muscles, in acraniotes and in

Amphioxus, and the embryologic development of the muscles will be discussed.

Striated voluntary muscle is derived embryologically from the mesoderm of the somite, arising from its dorsal and sometimes dorsolateral portion, the myotome. The somites lie on either side of the developing nervous system. Through proliferation, the myoblasts of the myotome furnish the anlagen of muscle ; these migrate from this primitive position to that of the muscle which they will form. During the migration of the muscle anlagen and the cytomorphosis of the myoblasts, elongated, multinuclear cells are differentiated, in the cytoplasm of which the striated myofibrils develop which characterize striated voluntary muscle. While the myoblasts still form a part of the somite, neuraxes of ventral horn neurons reach them ; such processes elongate as the myoblasts migrate away from their primitive positions. The innervation of adult body muscle is determined, therefore, by its somitic origin, that is, by the myotome or myotomes from which it is derived.

Visceral, nonstriated, involuntary muscles, and also cardiac, which is a syncytial muscle with striated myofibrils, are, with a few exceptions, of mesoder

Fig 220. A and B. Two cross sections of a triton embryo. Herlmq

A Through the body region in which the medullary plate (top) is not as yet closed and the primary body segments (ush) are beginning to separate from the celomic cavity (Ih).

B. Through the body region in which the medullary tube (n) IS closed and the primary body segments have been formed. ch, chorda dorsalis; dh, enteric cavity; ep, epidermis; mk', parietal layer of celom ; mk'‘, visceral layer of celom.

THE EFFECTORY SYSTEM

517

mnl origin, but they are derived from mesenchyme scattered through many body legions rather than from the myotomes. They are innervated by the sympathetic system through a two neuron chain, with the cell bodies of the preganglionic neurons in the central nervous system and those of the postganglionic neurons situated in sympathetic ganglia (p. 237). Both the preganglionic and postganglionic fibers aic classified as general visceral efferents.

Besides the t 3 ^pes of muscle above discussed, there is found a vaiiety of striated muscle which is known as branchiomeric or branchial muscle. In higher forms this tj'pe of muscle is indistinguishable from ordinary voluntary striated muscle by its general structure. An example of it is to be found in the highly specialized muscles of mastication. However, such branchiomeric muscles are derived from lateral plate rather than myotomic mesoderm, and, during the phjdogenctic development, have developed in relation with gill arches, and thus aie visceral; the differences in developmental history arc indicated by the innervation, which, for branchiomeric muscles, is from the lateral motor roots (special ^’isceral efferents) and not from the ventral motor roots (somatic efferents). Such branchiomeric musculature is confined to the head, the neck region (including the larjmx), and a part of the oesophagus, and special \dsceral efferent components arc found only in cranial nerves and po‘-sibly the spinal part of the accessory nerve.

Phyloguny or Cr.\nial Efferents

Amphioxus possesses no eyes and conse- musofPctromyzonraannus.Bhowingaspmo, IT- T occipilnl nucleus or nucleus of the first

quently the eye muscles are lacking. In spmnl nerve, depending on terminology used this form, a body space occurs in the region B Rostral motor vagus nucleus of Petro. , , 1 , \ mjzon marinus IJubcr and Crosby

occupied by the first (or oculomotor) myotome of craniotes. Neither visceral nor somatic musculature is to be found in the region. This space is between the point of emergence of the so-called nervus terminalis (see p. 136), and a sensory nerve emerging dorsally, the first dorsal nerve or nervus ophthalmicus profundus. The second myotome is present but lies in relation to the visceral plate which has a modified character Its changed branchiomeric character is indicated by its innervation — which is from doisolateral motor fibers of the second dorsal nerve — for this part of the visceral plate is the forerunner of the musculature innervated by the trigeminus. The perichordal part of the myotome, however, receives ventral root fibers. According to van Wijhc (’82, repeated later), the superior oblique muscle arises from the myotome in this region. The ventral root fibers innervating it in Amphioxus become the trochlear nerve of craniotes which emerges dorsally.

618 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Other observers (Plall, ’91, and Hoffmann, ’94; see p. 630, who found ganglion cells along the course of the trochlear) are of the opinion that the trochlear nerve is a visceral nerve related to musculature of branchiomeric origin. This opinion finds support in the interesting discovery of Bok (’15, .see p. 636), who found that the trochlear nerve is activated with the series of visceral motor rather than with the somatic motor nerves. Bok found that the visceral efferent roots develop from 20 to 24 hours before the somatic efferent roots in the chick, and that the trochlear nerve develops with the visceral efferent system and considerably earlier (rather than later) than the oculomotor. Its dorsal emergence also suggests such a visceral origin, since visceral efferent fibers leave the brain dorsal to the place of emergence of somatic efferent fibers. When to these facts is added the doubtful embryologic history of the superior oblique muscle, there appears to be a considerable body of evidence favoring the classification of the trochlear nerve as a visceral efferent nerve. As regards the possibility of its having associated ganglion cells along its course there is still some doubt. Hoffmann and Plali were of the opinion that such exist ; others regarded the cells as sheath cells rather than neurons. Whatever may be said with regard to the character of the trochlear in craniotes, however, its position is occupied in Amphioxus by a ventral or somatic motor nerve supplying muscles of myotomic ori^n. Such muscle may disappear in higher forms.

From the third myotome of Amphioxus arises the external rectus muscle of craniotes, which is supplied by the abducens nerve. From the corresponding visceral plate, contributions to the musculature supplied by the craniote trigeminal nerve arise. Only in cyclostomes, where the abducens arises ventromedial to the trigeminal nerve, do these two nerves show a topographic relationship to each other (fig. 224). In higher vertebrates, as a result of the shifting of the nucleus, the abducens root leaves the brain stem farther caudalward and becomes separated by a greater space from the region of emergence of the trigeminal nerve. The post-trigeminal branchial musculature (that supplied by facial, glossopharyngeal, vagus, and accessory nerves) arises from the visceral plate behind the third myotome, which corresponds to a large number of myotomes, most of which have disappeared on the ventral side. However, the sixth, seventh, eighth, and ninth myotomes of the region give rise to musculature innervated in lower craniotes by the spino-occipital nerves. In higher craniotes the muscles for the tongue, innervated by the hypoglossal, arise in this region. The tenth myotome develops into the more cephalic portion of the trunk musculature, innervated by motor neurons from the spinal cord, while the corresponding visceral musculature receives its innervation eventually through the sympathetic system.

In fishes, with the exception of many teleosts and ganoids, in the region of transition between the medulla oblongata and the spinal cord, are certain ventral roots which lie entirely outside of the skull (the paleocranium), which in these animals terminates at the labyrinth. In selachians (page 531) the skull extends farther caudalward and a number of the more frontal spinal nerves become included within the skull (protomeric assimilation). These nerves, which have no remaining sensory roots, often are called occipital nerves. By later auximeric

THE EFFECTORY SYSTEM

519

assimilation, a second group of nerves, the occipito-spinal nerves, is included. The two groups are termed the spino-occipital nerves by Furbringer (’97), and they are so designated in the following account. From such nerves, which were primaril}’- spinal cord nerves and later cranial nerves, is the hypoglossal nerve differentiated. Not only do changes in the relations of the skull occur, but

nuclear changes also appear, so that the spino-occipital nerves become a true head nerve, not only because the skull extends farther caudalward, but also because the nuclei extend farther frontalward. The spino-occipital and the hypoglossal are somatic efferent nerves.

The central cells of origin of the somatic and visceral musculature form, in their early development, a common column of neuroblasts (Soi, ’15), which occupies a positiou near the midline. This position is brought about through the influence of nervous impulses passing over the primary central path, the medial longitudinal fasciculus. Such impulses activate the neuroblasts in certain regions (which are determined by the correlated contraction of the myotomes with each passage of an impulse) leading thus to the formation of nerve roots (law of stimulogenous fibrillation. Boh, ’15 ; see p. 636). The secondarily acquired positions of the efferent neurons are due to migration from their primitive position under the influence of sensory centers correlated in function with

the muscles they supply (neurobiotaxis, Ariens Kappers).

Not only Boh but Mesdag (’09) and Tello (’22, Trav. Lab. Rech. Biol., Madrid, T. 21, p. 1) for birds and Windle (’33; see also

B

Fig. 222. A. Cervical motor column and caudal motor vagus nucleus of Petromyzon Iluviatilis. X40. Addens.

B. Rostral motor vagus nucleus of Petromyzon Iluviatilis. X40. Addens.

Shaner, J. Anat., vol. 68, p. 314) for mammals have found a common coliunn

of origin for the motor neurons ; moreover, the visceral efferents are said to lie

medial to the somatic efferents in this column- Coghill (’26; see p. 302) re

garded the special visceral efferents as arising from a “primarily integrated total motor matrix.” To those interested in the development of the behavior

pattern and in the migration of motor neurons from their embryonic to their adult positions, the above references are suggested. Special reference is made to the work of Coghill (’29, and elsewhere, see bibliographies of Chapters II and IX) on behavior in amphibian embryos, the studies of Windle and Griffin (’31, J. Comp. Neurol., vol. 58, p. 149), Windle, O’Donnell, and Glashagle (’33,

520 NERVOUS SYSTEMS OF ^VERTEBRATES AND OF FIAN

Physiol. Zool., vol. 6, p. 521) and Windle ('34, J. Comp. Neurol., vol. 59, p. 487) on the development of the beha^^o^ pattern in cat embryos, and on the series of studies of Windle (’32, ’32a, ’33) on the development and migrations of the nuclei in cat embr}^os. Of interest in this connection is the work of Hogg (’28, J. Comp. Neurol., vol. 44, p. 449) on the differentiation of the motor

nuclei in the rat from birth until adult life, and the studies of Craigie (’24) on changes in the vascularity of the region in the rat between these ages and his interpretation that functional activity needs greater vascular richness than growth processes.

Before turning to a consideration of the relations and connections of the motor nerves in the various vertebrate classes, a few w’ords of explanation arc necessary with regard to the charts and certain of the figures in this chapter. The charts are intended to show the relative positions and extents of the motor nuclei of the nerves as these nuclei arc projected on a sagittal plane. The upper base line in each diagram is the floor of the ventricle and the under base line is the ventral border of the medulla oblongata and the midbrain. Of cour.se the vi.sccral efferent columns lie lateral to the somatic efferent columns. In a figure with nuclei projected on a single sagittal plane, it has not been possible to illustrate this relationship, but this is shown adequately in the figures of the cross sections. The charts in the original German edition were furni.shcd by various observers who had studied the motor nuclei under the direction of, and in collaboration with, AricTis Kappers, In the present edition, these charts are supplemented and in part replaced by a scries of charts most kindly prepared and loaned by Dr. J. L. Addens, also a member of the Neurological Institute of .•\mslcrdnm, who has worked for many years on the positions and relations

Fir.. 223. A. Sagittnl spclion of the medulla oblonpala of Petromyzon marinu.s near the midline, showinR the continuation of the ccr\-ic.al motor cells into the calamus region (Motor cells), Jn front of the motor cells arc larger reticular cells (Reticulnr eelh).

li. .Sagittal section of the medulla oblongata of Petromyzon marinu-s, lateral to figure A, showing the caudal and the frontal visccrri-molor columns in part, lluhcr and Crorhy.

THE EFFECTORY SYSTEM

521

of the motor nuclei in vertebrates, mile the charts illustrate the points emphasized in the following account, they contain certain details to which reference cannot be made, since these details, and the evidence supporting them, are the outcome of work as yet unpublished. Under each chart the name of the observer responsible for it is found. The conventions used to indicate the various nuclei are shown with each figure.

The Effectory System of Cyclostomes

Y+rai IX+X 12 3

Fig. 224. A projection on the sagittal plane of the motor nuclei and motor nerve roots m two cyclostomes.

A. Petromyzon fiuviatilis Addens. IIlc + F/, caudal III root plus VI root (abducens of authors, oculomotor-abducens, Addens),

B. Bdellostoma stouti. Addens. In this 6gure Addens has interpreted a part of the facial root as joined vith the trigeminal nerve. The part of the facial thus joined has been regarded hitherto as a second trigeminal root.

There is a great difference in the relations of the motor nuclei in the two orders of cyclostomes, the petromyzonts and the myxinoids (see fig. 224). The more primitive and more complete arrangement in the petromyzonts will be described first, and then this ivill be compared with the relatively reduced condition in the myxinoids.

Two groups or columns of cells can be distinguished in the caudal part of the medulla oblongata of Petromyzon flu\datilis ; a dorsomedial (or ventral) column and a dorsolateral (or lateral) column {Johnston, ’02; figs. 221 to 223). Both columns consist of relatively large, multipolar neurons. The cells of the dorso

522 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

medial column differ from those of the dorsolateral column in being somewhat larger. The dorsomedial neurons send their fibers ventrally into what are termed the spino-occipital nerves; the neurons of the dorsolateral column are cells of origin for laterally emerging vagus fibers. Caudally the dorsomedial column becomes directly continuous with the mediocentral cell column of the spinal cord.

The cephalic limit of the column cannot be determined very easily, for large reticular cells, which do not send their neuraxes out through spino-occipital roots but to medullary centers, are intermingled with the neurons giving rise to spino-occipital roots, and the limit of the dorsomedial column can be determined only by tracing the nerve roots. The large reticular elements continue forward beyond the cephalic boundary of the dorsomedial colimm. They form the primitive homo

Fio. 225. A. Motor facial nucleus of Petromyzon fluviatilia. X40. Addens.

B. Motor trigeminus nucleus of Petromyzon fluviatilia. X40. Addens.

C. Caudal oculomotor nucleus of Petromyzon fluviatilia. X40. Addens.

logue of the nucleus magnocellularis inferior, which in these animals occupies a periependymal position in correspondence with the primitive character of the medulla oblongata. The spino-occipital roots pass somewhat caudalward and ventralward from their nucleus of origin near the floor of the medulla oblongata to their places of emergence, the cell bodies of the neurons lying somewhat farther forward than the emergence of the corresponding fibers, a condition which becomes more marked in plagiostomes and teleosts. The extent of the dorsomedial column and its relation to the dorsolateral or visceral column are clearly visible in the accompanying diagram (fig. 224). Reference to this diagram indicates that both the cells and the roots of the spino-occipital system have a distinctly caudal position as compared with their position in certain higher animals

THE EFFECTORY SYSTEM

523

(figs. 294 and 315). Addem termed the spino-occipital the first spinal nerve (fig. 222 A).

In petromyzonts the visceral efferent column is divided into a series of nuclear groups corresponding to the visceral efferent nerves. These nuclear masses lie close together with the exception of the neurons associated with the vagus, which are arranged in two groups forming a frontal and a caudal vagus nucleus, between which there is a space {Addens, ’28, ’33; figs. 221, 222, 224). Root fibers have been followed with certainty from the cephalic vagus nucleus but less certainly from the caudal nucleus (see, however, Johnston, ’02). Addens regarded the cephalic nucleus as a center for branchiomeric muscle, the caudal as a preganglionic center. The caudal and cephalic boundaries of the caudal vagus nucleus are difficult to determine (figure 224 from Addens). This cell group was seen by Treljakoff (’09) but not interpreted by him. It was recognized first as a vagus nucleus by Addens (’33 ; see also fig. 222). There is no distinct separation (fig. 224) between the frontal end of the vagus nucleus and the glossopharjmgeal portion of the visceral efferent group. There is no indication of a ventral shifting of the group, such as occurs in higher vertebrates. Dendrites of these neurons extend out in all directions; certain of the larger dendritic processes can be traced lateralward and ventrolateralward to come into relation with the impulses from the periphery, probably -with somatic sensory impulses carried by the nucleus of the descending root of the trigeminal and visceral sensory fibers brought in over sensory fibers from the gill arch regions. Often, but not always, the motor glossopharjmgeal root emerges with the sensory root.

The cephalic end of the visceral efferent or dorsolateral cell column contains cells of origin for efferent fibers of the facial and for similar fibers of the trigeminal nerve. In Ammocoetes, Treljakoff (’09) found that the trigeminal nerve arises from the brain bj*^ four roots, of which the two dorsal are sensory, the intermediate one motor, and the ventral one composed of mixed bundles. In the larval form the motor nucleus of the trigeminal is separated from the motor nucleus of the facial by a Muller’s cell. The cell bodies of the neurons composing the trigeminal nucleus, according to Treljakoff, are somewhat smaller, and the cross sectional area of the trigeminal column is smaller than the corresponding neurons and column of the facial. The fiber bundles arising from the nucleus swing forward, in frontal and caudal bundles, to the place of emergence of the roots. The caudal bundle, which is three or four times thicker than the frontal, unites ivith a sensory bundle forming the mixed or ventral root ; the frontal leaves as a pure motor root, which Addens (’33) regarded as oculomotor (p. 524). In the adult the relations are different, for here the trigeminal nucleus is differentiable from the facial not only through its greater breadth but also through the greater size of its cellular constituents; again they are separated by a Muller cell (see figs. 225, A and B, fig. 226).

The whole dorsolateral \dsceral efferent column lies near the flooi of the ventricle. None of its component nuclear masses shoves the tendency to migrate ventralward seen in the visceral centers of mammals. Thus the efferent nucleus of the facial occupies its primitive position near the floor of the ventricle, appear

624 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ing cephalically slightly behind the level of emergence of the root fibers and extending caudahvard for some distance. The nucleus of the trigeminal, likewise dorsal in position, has in general the relations as illustrated in figures 224 and 225B.

Paul Furbringer ('75) and his brother, Max Fiirbringer (’97), found that the abducens bundle of petromyzonts becomes separated from the trigeminal nerve in the orbit and then gives off two branches, one to the external rectus muscle (the muscle usually supplied by the abducens nerve), the other to the inferior rectus muscle, which is suppUed in other forms by the oculomotor nerve. Their results, which were carried out on Petromyzon marinus, were substantiated for

this animal by Coming (’02), Addens (’28) and Cords (’29). Similar conditions were found by Nishi (’22) in Entosphenus japonicus. Slightly different results were obtained by Tretjaltoff (’26, ’27, and ’29), who found that in Petromyzon fluviatilis not only fibers from the abducens complex but also a small twig from the oculomotor nerve reaches the rectus inferior.

The innervation of the rectus inferior of Petromyzon through fascicles associated with the usual abducens bundles, together with certain central relations, led Addens (’28) to the conclusion that the so-called abducens nerv’^e of these forms really carried both oculomotor and abducens fibers. The cells of origin of the abducens nerv’^e are not known vuth certainty for petromyzonts. J ohnston (’02) suggested an origin which needs further confirmation and the accotmt of this nucleus as part of an oculomotor-abducens nuclear group given by Addens (’28) has recently been retracted (’33). Addens (’33) believed that the neurons usually identified as the frontal end of the motor nucleus of the trigeminal nerve really give rise to oculomotor fibers. Two rootlets arise from this caudal oculomotor nucleus and join the abducens fibers outside of the brain, giving rise to his oculomotor-abducens complex. Tretjakoff (’29) had regarded this caudal oculomotor nucleus of Addens as a trigeminal nucleus and the root as the cephalic trigeminal root. It should be mentioned in this connection that Corning (’02) suggested that the oculomotor rootlets supplying the inferior rectus in Petromyzon join the abducens rootlets intracranially (see fig. 224 and fig, 230 A).

The nuclei of trochlear and oculomotor nerves (figs. 227, 229), which supply the eye muscles, show certain relations in cyclostomes which are of interest phylogenetically . Thus the trochlear nucleus evidently lies dorsal to the ventricle (Ahlborn, ’83; Tretjakoff, ’09; Huet, ’ll; Ariens Kappers, ’12; Addens, ’28)

Fig. 226. Motor trigeminal nucleus from Petromyzon marinus, showing its most caudal portion. Lateral-line fibers enter above the ganglion. Huber and Crosby.

nucl.isthm.' cell. Miill.isthm. A B

Fig. 227. A. Emergence of trochlear root of Petromyzon fluviatilis. X40. Addens. B. Trochlear nucleus of Petromyzon fluviatilis. X40. Addens.

tecl.opi

.^plex. chon tect.opt.v

mesenc, "r'. .O

p/€X. chon mesen

{Sulc Itm.

^Ic lim.

‘S-l -'.f ^ ^ aq.clud^P

]|r!|fr)esencl

I*.

4^ V i> , \\ y -v.*

nucl.n rostr. \ venlrolal. —

r.in rostn reefc.r M /, cruc.'

^ — nucl. interpedN.,^ / ■•■;•■

1'/'^“'^- ^ ^ tiucl.!! r. dorsom/^'^';^

1 / parvocelK 7 ^

Pnucl. in rostr. dorsomed. det tr.octTmetanl/ 1 magnocell. ^^^1 i^terped./

A B

'dec.r.Iirrt

'com.ans

Fig. 228. A and B. Rostral and caudal portions, respectively, of the rostral oculomotor nucleus of Petromyzon fluviatilis. X40. Addens.

525

526 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

in the anterior medullary velum, although with this interpretation Kuhlenbeck (’27) and Saiio (’31) disagreed. Moreover, it lies very far caudal to the plane of emergence of its root, at about the level of emergence of the trigeminal nerve (figs. 227B, 229A), and much nearer the nucleus of this latter nerve than that of the oculomotor. This relation may be in harmony with the discovery of Hoffmaiin (’94) that in Acanthias the superior oblique muscle breaks off from the musculature innervated by the trigeminus and that the trochlear root itself separates from the trigeminus secundar (which appear in amphibians to be a single root), and is in complete agreement with the interpretation of Bok (’15 ; see p. 518 and bibliography for birds), who described the trochlear as a visceral efferent nerve {Ariens Kappers, ’20).

Tretjakoff (’09) and Ariens Kappers (’12 and ’20) regarded the caudal position of the nucleus of the trochlear as primitive. Recently Addens f’28) described this eye muscle nucleus in Petromyzon. He also regarded the caudal location of this nucleus as primitive, calling attention to the fact that this caudal position, while most marked in Petromyzon, is found also in other animals, since a distinctly caudal location of the trochlear nucleus has been described in urodeles by Rolhig (’13), and part of the nucleus in man frequently is separated from the main trochlear cell mass to form a nucleus trochlearis posterior (Tsvchida, ’06 ; Ariens Kappers, ’12 ; van Valkenburg, ’12). Addens (’28) was of the opinion that neurobiotactic factors — the influence of vestibular and lateral-line impulses — operate in producing the dorsal position of the trochlear nucleus in petromyzonts.

The oculomotor nuclear gray may be divided into three nuclear groups (Tretjakoff, ’09) in Ammocoetes — a ventral nucleus giving rise to crossed fibers, and a dorsal and a lateral nucleus sending out homolateral fibers. The dorsal nucleus is separated caudally from the trigeminal nucleus by a cell of Muller. The nucleus lies in part at the plane of the motor cells of the trochlear and its cephalic tip represents the frontal end of the somatic efferent column. Above the nucleus is a coordinating zone, to which dendritic processes of oculomotor neurons contribute. The ventral nucleus lies below and in part among the fiber bundles of the tractus octavo-motorius in Ammocoetes. Sometimes this nuclear group is not reckoned with the oculomotor group, but its fusion in certain cases with the dorsal nucleus (Ariens Kappers, ’20) and the fact that its extent corresponds reasonably well with that of the emergence of the oculomotor nerve favor its interpretation as an oculomotor center. Moreover, Tretjakoff was able to trace certain crossed (and a few uncrossed) fibers into the oculomotor root from the gray. The lateral oculomotor nucleus, described by AJdborn (’83), Johnston (’02), Tretjakoff (’09, ’29), and others, lies at the level of emergence of the oculomotor and slightly in front of it. Both Johnston and Tretjakoff have traced fibers from this nucleus into the homolateral oculomotor root. The latter observer carried to this nucleus bulbo-oculomotor fibers, possibly of vestibular origin, running directly or, for part of their course, with the medial longitudinal fasciculus and ha\dng specialized endings around the cells.

Certain differences of opinion from those cited above, as well as certain points of agreement, are to be found in the recent careful accounts of the oculomotor

THE EFFECTORY SYSTEM

527

nuclei of Petromyzon by Addcj}s (’28, ’33). He divided the oculomotor nuclei into two parts, a rostral and a caudal portion. The caudal portion of his oculomotor complex, a part of the oculomotor-abducens complex, has received consideration (p. 524) and need not be discussed further here. The rostral oculomotor cell group was rediv’ided by Addens into two nuclei, a dorsomedial nucleus and a ventrolateral nucleus. The dorsomedial nucleus is smaller celled and corresponds to the ventral nucleus described by Treljakoff. This dorsomedial

Fia. 220. A. Tlic position of the trochlear nucleus in the velum cerebelli at the frontal level of the trigeminal root in Petromyzon marinus.

B. Dorsal oculomotor nucleus and root in Petromyzon marinus.

C. Ventral oculomotor nucleus in Petromyzon marinus. Huber and Crosby.

From the larger celled ventrolateral nucleus (see figs. 228 and 229), which corresponds to the lateral nucleus of Johmton (’02), Treljakoff (’09), and Ariens Kappers (’20), only homolateral fibers arise. Addene (’28, ’33) reemphasized the extreme ventral position of this nuclear mass. He even found certain of the cells extending outside of the brain among the root fibers of the oculomotor nerve, as had Johnston (’02) for Larapetra wilder! and Treljakoff (’29) for adult Petromyzon fluviatilis. Addens ('28, ’33) discussed the possibility that this ventral position is primitive, but reached the conclusion that in all probability it is secondarily acquired as a result of the decrease of optic impulses in these forms. None of the authors dealing with the oculomotor nucleus in petromyzonts has found evidence of the existence of an Edinger-Westphal nucleus in these animals, but since intrinsic eye musculature and a ciliary ganglion are lacking here, the presence of such a nucleus is hardly to be expected.

The brain in Myxine (and in Bdellostoma, also, according to Black, ’17) is greatly compressed in the frontocaudal direction (Furbringer, ’75). The massive character of the brain is associated with a marked reduction in its ventricular system ; thus the fourth ventricle has the appearance of being little more than a

THE EFFECTORY SYSTEM

529

medulla oblongata near its caudal end, Jansen (’30) applied the name of the glossopharyngeal-vagus complex. Furbringer (’75), Holm (’01), and Worthington (’05a) have described a glossopharyngeal nerve. The motor fibers of this complex arise from a cell column which in Myxine glutinosa {Jansen, ’30) lies close to the lateral surface in the caudal part of the medulla oblongata, with its cephalic end overlapping dorsolaterally the caudal end of the trigeminal-facial nuclear column and swinging somewhat farther medially at more caudal levels where the medulla oblongata has decreased markedly in size. Addens (’33) divided the nucleus into a larger rostral and a smaller caudal part. In general the cells are typical motor neurons, but Jansen found a few aberrant cells (considered sensory by Adde?is, ’33), unipolar in type, the single process proceeding medialward and there bifurcating, after which one branch enters the peripheral nerve and the other rebranches near the nucleus. A few paler staining cells at the cephalic end of the column, slightly ventral to its main nuclear mass and lateral to the caudal end of the efferent facial cell column, were regarded by Jansen as giving rise to efferent fibers for the two most cephalic rootlets of the glossopharyngeal-vagus complex, which small rootlets he regarded as homologous with the glossopharyngeal nerve of other forms.

The trigemino-facial nuclear group in Myxine (fig. 231) is an elongated cell mass extending from behind the plane of entrance of the trigeminal roots caudalward to a point slightly behind the cephalic end of the glossopharyngeal-vagus complex, so that the cephalic portion of the visceral efferent colmnn overlaps the caudal part {Jansen, ’30). The facial nerve as usually recognized is small. According to Jansen (’30), its coarse motor fibers arise from the whole extent of the trigemino-facial nuclear complex but chiefly from the caudal end of the nucleus. However, Addens (in his recent publication, ’33 ; see fig. 224) believed that in addition to this usually recognized facial root there is a second bundle of facial root fibers (usually regarded as the second root of the trigeminal) which has joined the trigeminal root fibers, producing again the condition defined by this observer as central anastomosis.

The trigeminal motor neurons fall into two cell groups {Ariiins Kappers, '10, ’12; Black, ’17 ; Jansen, ’30), a cephalic or anterior division, pars magnocellularis, consisting of large neurons having short, thick dendrites, and a caudal or posterior portion, pars parvocellularis, situated immediately behind the magnocellular portion, both portions providing fibers to the motor trigeminal root. According to Addens (fig. 224B), however, the small-celled portion is a motor facial nucleus. He divided the large-celled portion into a medial, small-celled

r'Oufa

Iranjvept

630

Fia. 232. Acanthias acanthias (L.).

THE EFFECTORY SYSTEM

531

part and a lateral, large-celled part. The visceral efferent column is much farther ventrolateral in the myxinoids than in the cyclostomes. Jansen’s Golgi preparations indicated that the neuraxes of many of the motor neurons form a long dorsolateral loop before entering the nerve roots. It is possible that this change in position is brought about by the great relative significance of the sensory trigeminal system in these forms. The descending root of the trigeminal is hypertrophied here. Its course parallels that of the visceral efferent column in the medulla oblongata, and neurobiotactic influences may bring this latter column into close positional relation with the sensory center from which it receives its most important stimulations. The importance of the cutaneous sensory impulses in these animals is exemplified by their mode of life, for they live mostly in the dark at fairly great depths (Ayers and Worthington, ’07 ; see also ’08 and ’ll). The great reduction of the eyes in myxinoids is associated with a lack of eye muscle nuclei (Sanders, ’94; Holm, ’01, and others).

The Effectory System of Plagiostomes

To illustrate the arrangement of the motor nuclei in plagiostomes, there are reconstructions given (figs. 234, 239) of these nuclei in rays and sharks. These show that the somatic efferent colunm, which has a ventral position in the cervical cord, be^ns to swing into a more ventromedial position in the region immediately behind the calamus scriptorius, although still ventral to the ventricle. The occipital roots (with the other spinal gray) are indicated in black in the figures, with the exception of figure 232, where they are shown in the diagram by a black field with white circles, with roots indicated by the letters w, x, y, according to the scheme of Furhringer (’97) . Such occipital roots leave the greatly enlarged cranium of the plagiostomes. Their nmnber varies in sharks. They are lacking in the rays, where all caudal ventral roots emerge behind the cranium. To what extent these topographic differences indicate intrinsic differences, and to what extent the absence of occipital nerves (as in rays and, as will be seen later, in teleosts) is in relation -with a corresponding lack of peripheral muscles, are as yet uncertain. However, it is known that the shortening of the cervical region in rays, for example, is associated with a reduction of skeleton and muscles. For a further discussion of this matter, the reader is referred to accounts of Furhringer (’97), Neal (’97), Edgeworth (’ll), and van der Horst (’18).

The visceral efferent column (Ariens Kappers, ’08, ’ll, ’12, ’12a, ’14; Black, ’17) shows certain differences in plagiostomes as compared with that in cyclostomes. The most marked difference is the more caudal position of the facial nucleus in the former animals. In plagiostomes the efferent nucleus of the facial nerve has become a part of a facial-glossopharyngeal-vagus efferent column (figs. 232, 234). The nucleus of origin lies considerably caudal to the place of emergence of the facial root, and the root fibers pass forward along the floor of the ventricle and then turn ventrolateralward to their point of emergence from the medulla oblongata. No genu is present in the course of the facial root where it turns ventralward, caudal to the motor trigeminal nucleus. Its nucleus occupies a relatively dorsal position just under the ventricular floor. The tendency toward

532 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

a future more ventrolateral position of the nucleus is indicated by the direction in which the principal dendrites develop. The gustatory component of the sensory root of the facial nerve is markedly large in plagiostomes and it turns caudalward to terminate in the common facial-glossopharyngeal gustatory center at the level of entrance of the glossopharyngeal root (page 253). This center is the chief center in establishing reflex circuits, through the efferent nucleus of the facial nerve. Thus the caudal shifting of the efferent nucleus of the facial nerve

in plagiostomes corresponds with the increased development and importance of gustatory impulses in these animals, the nucleus moving caudalward through

the neurobiotactic action of the gustatory centers so that it may lie nearer the centers from which it receives its major stimulations. This addition of the efferent facial nucleus to that of the glossopharyngeal necessarily lengthens the caudal nucl. visceral efferent column.^ The nuclei of the glossopharyngeal olivary nuci. vagus nervBS resemble

Fio. 233. A cross section of the meduUa oblongata of Galeua the nucleus of the facial. Howcanis, at tne calamus region.

ever, there are certain differences in the course of the root fibers. A fascicle of the glossopharyngeal may pass through the motor facial nucleus (fig. 235) on its way to the periphery, and, in addition, make a very short loop through the lateral part of the fasciculus

longitudinalis medialis. Possibly this course indicates that the primitive posi

tion of the visceral efferent column is farther medial than its position indicates

in the full-grown animal.^ The caudal border of the visceral efferent column is

approximately the same in all sharks. Addens (’28, ’33 ; see p. 628) divided the vagal part of the column into rostral and caudal parts in Selache maxima and Holocephali (figs. 234:B, and 239B).

The abducens nucleus in sharks (Ariens Kappers, ’10, ’12 ; Black, ’17) consists of large neurons not compactly arranged but scattered along the medial longitudinal fasciculus. Only by tracing the root fibers is it possible to delimit this nucleus from the surrounding gray. The abducens cells are far dorsal in all sharks and their dendritic processes spread out over the middle third of the distance between the entrance of facial and of glossopharyngeal roots, while the abducens root fibers lie between the levels of emergence of facial and of glossopharyngeal fibers, but nearer the nudline, as is also the case with the occipital nerves. Within the medulla oblongata; the neuraxes of the abducens

, accessory as distinct from the glossopharsmgeal-vagus complex has

been distinguishable m these forms up to the present time, although sharks are said to possess an accessory nerve (fig. 233).

= The lateral position is even more striking in reptiles and birds (see figs. 152 and 155A). e course the glossopharyngeal root fibers through the facial nucleus produces noticeable consequences jn teleosts.

THE EFFECTORY SYSTEM

533

neurons pass through the lateral part of the medial longitudinal fasciculus and swing ventralward and slightly ventrolateralward through the medulla oblongata, in their course describing a curve with the convexity toward the midline (fig. 42). The dorsal position of the abducens nucleus in the shark undoubtedly is associated with the enormous development of the dorsal vestibular and optic

ttj

_ imL 111! not ID

_ todiainollV

cod ud not VD _ nod VI

md lod tool X mil XI; ctSs rlaBcsscKcl)

md aid not? DodadnolIX

L oodDoiltij

cud ami tool spill, occ. am! XD

]- nod Ediiijer-Wealplial

inferior oEntj nod

tYn,IX= {eiioVD,IX J calans stnplorins

H

1^^^1111111

HU

VI XI ott.l(») cct2(t) occJ(j)

Fig. 234. Charts showing the topographic relations of the motor roots and nuclei of certain sharks.

A. Hexanchus griseus Ariens Kappcrs.

B. Selache maxima. Black.

C. Acanthias vulgaris, van der Horst.

systems, from the brain centers of which paths pass to the highly developed medial longitudinal fasciculus (from the vestibular nuclei at the level of the abducens nucleus and probably determining that level). In a certain sense, the abducens nerve may be regarded as the functional ventral root of the acoustic nerve, although it does not belong primitively to the same neuromere {Ariens Kappers; see also p. 518).

534 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The trigeminal nucleus has about the same relations in all plagiostomes. As contrasted with the position of the corresponding nucleus in petromyzonts, it shows a ventrolateral shifting which is evident throughout nearly the whole

length of the nucleus. The position of the nucleus of the descending root of the trigeminal very probably has an influence on that of the motor trigeminal nucleus, for the mandibular part of the trigeminal nerve constitutes its more dorsal portion on entrance to the brain (see figures 234, 236, and 239).

The nuclei of the oculomotor and trochlear nerves show certain significant relations when compared with the corresponding centers in cyclostomes. The trochlear nucleus has migrated forward in plagiostomes and occupies a. position on the floor of the aqueduct (fig. 234C). Frequently the trochlear root does not take part in this frontal migration, in which case the nucleus lies in front of the plane of emergence of the root (figs. 234, 237). In Spinax, among the plagiostomes studied, the plane of entrance of the trochlear root is shifted forward to approximately the level of the trochlear nucleus. It is probable that differences in position of the velum anticum cerebelli, which in turn depend upon mutual relations between the cerebellum and the tectum opticum, and the size and the shape of the caudal end of this latter area may be determinative of the region of emergence of the trochlear, but whether other factors play a part in such a determination is uncertain at present. The position of the trochlear nucleus with reference to the medial longitudinal fasciculus and the ventricle is the same for all sharks. The nucleus, which is crescent shaped, lies close under the ependyma of the ventricle, dorsolateral to, and partly embedded within, the medial longitudinal fasciculus. Its length, compared with that of the oculomotor nucleus, is indicated in the diagram.

The oculomotor nucleus in plagiostomes has retained the position at the level of entrance of its root fibers characteristic of the rostral part of the oculomotor nucleus of cyclostomes. However, it shows a marked difference in position in cross sections, since it lies in plagiostomes far medially and dorsally (in part between the left and right medial longitudinal fasciculi, fig. 238, Nu. III). In Hexanchus the cephalic part of the nucleus lies somewhat farther ventrally. In

Fia. 235. Course of the motor root fibers of the glossopharyngeal nerve through the motor facial nucleus of Scyllium.

Mot. Duel. V

Fig. 236.

The position of the motor trigeminal nucleus of Raja clavata.

THE EFFECTORY SYSTEM

535

these forms there is a very deep indentation of the midline portion of the aqueduct which extends ventralward almost to the commissura ansulata. In other

selachians the depression on the floor of the aqueduct is not so deep, and the nuclei of the oculomotor, lying between the two medial longitudinal fasciculi and, caudally, extending over their lateral borders, do not reach so far ventralward as in Hexanchus.

It is not possible to make any clear division of the oculomotor neurons into distinct nuclear groups. At most, a dorsal and medial group, lying between the medial longitudinal fasciculi, and a lateral group, might be distinguished which may be termed respectively, the dorsal nucleus and the lateral nucleus (fig. 238). This latter' nucleus corresponds in position in transverse series with the

p')!'*; '-a,

trochlear nucleus, with Fio. 237. The position of the trochlear nucleus in relation to 1 . , . j. ,, . . the medial longitudinal fasciculus and the course of the trochlear

which it IS directly continu- „ 0 t in AcanthL. van der Horst.

ous caudally. The two nu- ^ v.

clear groups, the dorsal and

the lateral, are separated •v. _ \

by no clear boundary and \

are distinguishable from 5 ^ ' ' mVi /

each other only through dif- 1' v-V " ' ‘

ferences in position and not ' •/

11 •* ’ * • ' «l

through differences in cell - 7

type. The oculomotor and ' ' 'f&f-y

trochlear nuclei are not sur- '•

rounded by the small-celled

reticular groups character- ^

istic of the region in higher „ ^ ^ , , o , . . ,

forms ; there is to be found

only a mass of large cells with no clear arrangement into nuclear groups. Whether certain parts of the oculomotor nucleus are more concerned with crossed than

Fia. 238. The oculomotor nucleus of Selache maxima.

536 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

with uncrossed fibers it is impossible to state at present, but the number of crossed fibers appears to be less than that of the uncrossed.

In conclusion, it may be stated that the relations of the efferent cranial nuclei and their roots in the rays (fig. 239A) are very similar to those in the sharks, with such exceptions as have been noted.

Kej

rad mils

~ rad aad ltd ly

rad rai root Vn _ rad VI

md and root X uj XI, cdb rlaiioudlci.)

i

■dulttdV

■odudnetK

radlMen

_ md ud root xpm, occ. ud XD _ dferwr o5tji 7 md

|_ mdEiEijtrWejliW

tYII.IX=.it«VIl,IX 1_ talaBontriptorim

Fia 239 Charts of the topographic relations of the motor roots and nuclei m a ray and a holocephahan

A Raja clavata Arxens Kappers B Chunaera monstrosa Anens Kapperg.

The Effectory System of the Ganoids and Tbleosts

The arrangement of the motor nuclei as seen in plagiostomes continues, on the one hand, to the arrangement found in ganoids and through ganoids to that found in teleosts (fig. 240), and on thr* other hand, through the arrangement found in Dipnoi to that found in amphibians. The great differences found in teleosts, as van der Horst (’18) has pointed out so clearly, have a taxonomic significance and are best understood by approaching them through a study of the relations in ganoids (Crossopterygii). The present knowledge of these latter fishes is based, to a very con.siderable degree, on the researches of Droogleever Fortuyn (’12), Theunissen (’14), van der Horst (’18), and Hocke Hoogenhoom (’29).

The position of the cells of origin for the accessory nerve in the ganoids, as in certain other vertebrates, is not agreed upon by all observers. By most workers the accessory nerve is regarded as a visceral efferent nerve, and its cells of origin

Ktj

_ cududnelin

_ cad in] ml YD

_ cad ml tod lY cad ind red Y

_ cad an] tool X lad XI, ctSj rXa!»stck(tl)

_ tadin!reoli5si,oec.mlXn I

_ cfoior cSni7 cad

cad ml tool K I— and Daltn

j I _ cad E&ger Weslpis!

tYU, Ko laa YH, K 1 calamiB scnploriaj

YI X

Fig 240 Charts showing the topographic relations of the motor roots and nuclei of three ganoids and a teleost.

A Calamoichthys calabaricus van der Horst B. Acipenser ruthenus Theumssen (7. Amia calva Droogleever Forluyn D, Megalops cyprmoides. van der Horst.

637

Fio. 241. Megnlops cyprinoides (Brousa). The nrrangemont of tho motor roots and nuclei in a relatively primitive teleost. van der Horst.

638

THE EFFECTORY SYSTEM

539

are believed fo lie in the caudal part of the vagal column. Certainly this would bo true of any representation of the bulbar accessory roots, whether the special \nsceral elTerent or the general visceral efferent (preganglionic) components were under consideration. With regard to the spinal accessory portion, there is still not general accord. Those who regard this component also as special visceral efferent in character consider that its cells are primitively in line mth the remainder of this visceral column. Another point of view advocated for the spinal accessory nerve is that it is somatic efferent in origin. Evidence confirmatory of this latter point of view, based on a careful study of a mde range of material, including fushcs, has been published hy Addons (’33). According to this observer, the evidence indicates that fibers of the foremost ventral root or roots of the somatic efferent column of the cord, which primitively emerge in the typical ventral position, have acquired secondarily a dorsal course. Any final decision as to the character of this .spinal portion of the accessory ner\-e must await further studies on its embr 3 mlogical development and peripheral distribution (see fig. 240).

According to various observers (see fig. 240) the facial, glossopharjmgcal, and vagus efferent fibers have their cells of origin in a common caudal vi.«ccral cffcrcnl column in such ganoids as Amia and Acipenser.

In the more frontal regions of the medulla Fio- 242 . Vagus and spinal nuclei oblongata of ganoids lies the abducens nucleus, P’seatonus.

which is very small and verj' diffuse in Calamoichthys (fig. 240A) and not well delimited in Lcpidostcus. It has a somewhat less dorsal position than does the homologous nucleus in plagiostomes, perhaps because the medial longitudinal fasciculas is less well developed in these ganoids. On the contrary, the trigeminal nucleus has a less ventral position and a greater length than in plagiostomes. The most outstanding difference between plagiostomes and ganoids lies in the relations of the trochlc.ar and oculomotor nuclei. These two nuclei are separated entirely from each other in the ganoids and the trochlear root always emerges far caudally (for Polyodon, Black, ’17). (Note fig. 240.)

The motor system in teleosts varies greatly, depending upon the animal studied. In the most primitive teleosts the relations resemble those described for ganoids. Such a transitional form is Amia calva, which although a ganoid differs from tj^pical ganoids (Crossopter 3 ^gii) in the somewhat more ventral position of the nucleus of the abducens {van dor Horst, ’18). Somewhat analogous conditions were found by van dor Horst (’18) in Megalops cyprinoides (figs. 240D and 241) and in Salmo. In these latter animals both the more ventral position of the abducens nucleus and the ventral shifting of the facial nucleus suggest the conditions common to many teleosts.

The occipital roots and the first spinal root are absent in many teleosts. The somatic efferent cell column differentiates from the occipital column of lower fishes and extends somewhat farther ventralward. As its dorsal border remains

540 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

practically the same, the dorso ventral diameter of this nucleus is quite large in teleosts and the cephahc end of the nucleus begins to take on the characteristics of a horn of the spinal cord. Occasionally these horns are divided into two parts — a central or dorsal portion near the central canal, and a ventral or ventrolateral

cod lod root m

_ inid u<) root IV .•tv-vl; _ oodandrMlY

mxl. ind root VO

rod VI

9_ rodudtwdX

rodiiidrootXuilXI;cdli rXedw»d(cJ.)

rodDalcn

rod inl not spin, occ. jad XD I I - rod Efinjer-VltslpU

olerior oEnir rod {.VlllXa pesVltlX 1= ciluns in^ttiis

Fig. 243. Charts showing the topographic relations of the motor roots and nuclei in a tench, a siluroid, and a flying fish.

A. Tinea vulgaris. Ariens Kappers.

B. Siluris glanis. Berkelbach van der Sprenkel,

C. Exocoetes evolans. van der Horst.

part by internal arcuate fibers, which take origin from the dorsal horns of the spinal cord and are to be regarded in part as processes of secondary neurons. In the teleostean embryos the central (or dorsal) cells are slightly larger and have a distinct, quite deeply staining fibrillar structure. Their branches extend out into different parts of the substantia reticularis. In part they reach the fasciculus longitudinalis medialis, where they come into relation with the giant fiber of

THE EFFECTORY SYSTEM

541

Mauthncr’s cell. The ventrolateral (or ventral) cells are larger than those of the central (or dorsal) group and have finer and still more deeply stained neurofibrils. Their dendrites can be traced in all directions {TcUo, ’09). The absence in telcosts of those cell groups of plagiostomcs which give rise to the occipital nerves (compare figures 239 and 243) is possibly in agreement with the reduction of the ventromedial (hypobranchial) musculature in telcosts, where the neck region has been greatly reduced. Usually the central and ventrolateral cell groups both extend fonvard to the level of the calamus scriptorius {van der Horst, ’IS). In ccrt.ain telcosts, however, van der Horst (’18) found the ventrolateral (or ventral) cell column extending much farther forward than the central portion (see fig. 241), an example in such cases of the importance of certain specific reflex paths. TcUo (’09) and Ariais Kappers (’20) were of the opinion that the position of the central or dorsal cells (which, unlike their homologucs in plagiostomcs, arc found to extend somewhat caudally in telcosts) is influenced by Alauthner’s fiber and the medial longitudinal fasciculus, while the position of the ventrolateral cells is determined to 2-14. The efferent nuelens and root of the a considerable extent by the influence p>scatonus.

exerted upon them by the descending tractus tecto-bulbaris ventralis and the ventrolatcrally situated vestibular tracts, particularly the tractus octavo-motorius cruciatus vcntrolatcralis, first described by Wallenberg (’07). This latter tract, after its decussation at the level of entrance of the acoustic nerve, passes caudalward along the ventrolateral border of the medulla oblongata, then courses along the caudal abducens nucleus and along the ventrolateral side of the brain st^m directly beneath the peripheral ventral horn cells (fig. 242). This path breaks up among the peripheral cell groups.

YTiether the ventrolateral and central groups serve separate muscular systems or are concerned in particular movements has not as yet been decided. One point of difference between teleosts and plagiostomcs, with regard to this nuclear column, must be emphasized. In plagiostomcs the frontal portion of the cell column only innervates frontal median muscles, whereas the cephalic end of the spinal cell column of teleosts supplies neuraxes to the muscles of the shoulders and fins (as well as to certain more caudal ventral trunk muscles), particularly where the first spinal nerve is lacking (fig. 243 A, B, and C). By analogy with the localization within the spinal cords of higher animals, including mammals, as described in Chapter II, it seems altogether probable that the more ventrolateral group is concerned with extremity musculature, since in general the ventrolateral cell group of the spinal cord is concerned in innervating such muscles. Of interest is the discovery of van der Horst (’18) that in several teleosts the ventrolateral cell group extends even farther forrvard than the central or dorsal group.

542 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The %nsceral cell column is better differentiated in most teleosts than in plagiostomes and ganoids. Although certain teleosts and ganoids have no true trapezius muscles, and consequently no nucleus of the spinal accessory nerve, on the whole the visceral column does not extend much less far caudally in them than

j— Bad wJ root in

_ ■d.tfdrtidT *X**'t*M - md.$dmiy

■ _ rad laJ rtd \

adurintta

rad lad ml X rd XI, cdb tXmbwiACeJ.) tsMlliml — *4 Dtim

rad id root cec iflJ Xn I | _ ■dEfccsWtafU

donor tSnn od. tYn,IX= {esBVI!,IX I olan u i^ori *

I in 2Ui CIi irts illiistratinp the lopof^npliic rclntions of the motor roots and nuclei in Gndus, Lopliiiii, and Ortli iRonocii’

d C! iilii'i ardflinur tnn drr Han't

II lyrpliliis pi“( norms Draaqlrnrr Forlwjn.

C OrtliaRori'Ciis rnnla Kinder Uor^t

THE EFFECTORY SYSTEM

543

explanation in his interpretation of the spinal accessory nucleus as a constituent of the somatic efferent column.

In the more caudal part of the visceral efferent column of teleosts lie closely packed cells of origin of efferent fibers (figs. 242 to 249) of the vagus, continuous forward in primitive teleosts with the cells of origin of the glossopharyngeal and facial nerves. In less primitive teleosts only a part of the fibers of the facial can be traced to this cell column. The others arise in part in a separate nucleus. In certain species (siluroids, Siluris glanis, fig. 243B) all of the facial fibers arise from this separate nucleus (van der Horsl, ’18). Also, the glossopharyngeal neurons may separate from the caudal visceral column (fig. 245). In that case they lie in intimate relation with those of the facial nucleus. In animals in which the glossopharyngeal cells of origin remain united with the nucleus of the vagus, while the cephalic part of the facial nucleus has separated, the root fibers of the glossopharyngeal have a very peculiar course. After their entrance to the medulla oblongata, the glossopharyngeal fibers run far frontalward and cross the frontal facial nucleus (see fig. 246) in order to take a position on the medial side of this cell mass. Then they pass caudalward along the side of the medial longitudinal fasciculus, ventral to, and in company with, the root of the facial, which passes (fig. 246) to the caudal facial nucleus, and together with that root enter the cephalic end of the facialglossopharyngeal-vagus visceral column. This frontal loop described by the root of the glossopharyngeal is due to the forward migration of the cephalic end of the facial nucleus through which the glossopharyngeal fibers (as far down as sharks) pass in order to reach their cells of origin. If separated from the nucleus of the vagus (fig. 245), the glossopharyngeal nucleus, together with the facial nucleus, may occupy a quite ventral position, presumably due to the influence of impulses passing over the ventrolateral spino-bulbar tract (compare figs. 243 and 245) . This position is particularly evident in those species in which the cutaneous sensory system is hypertrophied.

The most striking characteristic of the facial efferent center in teleosts is its division into two nuclei, of which the more cephalic is a distinct nuclear entity while the more caudal is joined to the vagus-glossopharyngeal nucleus in most teleosts. The causes underlying this difference are not as yet understood. It appears possible that the more cephalic nucleus, which corresponds topographically to the trigeminal nucleus, may be concerned in the innervation of gullet musculature, while the more caudal nucleus innervates muscles associated with the gill apparatus, such as the levator and adductor operculi, which in teleosts receive root fibers of the facial nerve. The positions of both cephalic and caudal

Fig. 246. A drawing to illustrate the course of the efferent root of the glossopharyngeal nerve through the facial nucleus of Gadus.

544 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

portions vary in different teleosts (figs. 243, 245, and 249). Thus in figure 247 (Gadus), the cells of the cephalic facial nucleus are found scattered along the

Sens, root VII Mot. root VII

Fio. 247. The anterior ctTcrcnt nucleus of the fncinl nerve and tlio posterior abducens nucleus in Gadus morrhua.

gray of the anterior secondary gustatory tract. In a form such as Tinea, where the taste fibers are much greater in number than in Gadus, this relation is still more noticeable, while it is most evident in the siluroids {Berkelbach van dcr

Sprenkel, T5 ; fig. 243 A and B) . Further evidence in favor of the statement that the position of this nucleus is dependent upon the influence of this gustatory tract is to be found in the fact that the amount of ventrolateral shifting of the nucleus is proportional to the degree of development of the tract. In Gadus (fig. 245A) is found an example in which the caudal facial nucleus is slightly separated from the main caudal or posterior column. No distinct caudal facial nucleus exists either in Lophius or Orthagoriscus (fig. 245B and C). The cells of origin for the facial root of Lophius (fig. 248)

iM'

/'V’Vy

SrtJ

0<v rw« •( V

. . Iff! WV ffirt

Fi'i. 2),S. Tlic dor.'.-.'il (Dors. Mnt. mid. VII) nnd veiilnil 0 ml M„t. mid. 17/) fnciiil nuclei of Lopliiiis Drnoghrirr I'tirtui/n.

pi.ic.'iloriii.'!.

n ‘'I " ° n caudal) facial

he tvo cell masses being interconnected by a narrow bridge of cells

°Pco

Fig. 249. The motor roots and nuclei of Tetrodon. van der Horst,

546 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

In Orthagoriscus {Burr, ’28) the conditions are the same except that the band is wide and the nucleus has about the same width throughout. At the caudal end of

the facial nucleus lie the cells of origin for the glossopharyngeal nerve. Possibly in these animals, as in Tetrodon, the position of the caudal facial nucleus is influenced by impulses passing over the descending root of the trigeminal nerve.

The abducens nuclei in teleosts (fig. 250) show certain marked differences from the corresponding nuclear masses in plagiostomes. Thus, in almost all teleosts, the cells associated with this nerve have a ventral rather than a dorsal position (figs. 249 and 250), and are divisible into two distinct nuclear groups (figs. 243, 245, 249). Since the ventral reflex tracts, particularly the tectobulbar system, are of major importance in these animals, while the fasciculus longitudinalis medialis is poorly developed in / contrast to the corresponding fasciculus in plagiostomes, the ventral shifting of the abducens nuclear complex is not surprising, but is to be regarded as a very evident example of the action of

Tta 250 Tetrodon speciosus The posterior root and the posterior nucleus of the abducens nerve, and the anterior efferent nucleus of the facial nerve The fiber bundle medial to the abducens nucleus is the tractus tectobulbaris ventrahs van der Horst /

\ nurl V lormation

I III 251 The trigeminal ntirlc I of L<iphuis piscatorius

THE EFFECTORY SYSTEM

547

neurobiotaxjs. Tlie presence of two distinct abducens nuclei appears to be characteristic of many teleosts. As a rule, each of the two abducens nuclei gives rise to a nerve root. One of these roots emerges in the same transverse plane as does the facial root, but in a more medial position. The other leaves the brain between the levels of emergence of the facial and the glossopharyngeal nerves. The roots often leave the brain at points somewhat cephalic to the planes of their respective nuclei. i i

Fig. 2.‘;2. Arius. At the left, the nuclcu.s molorius trigemini posterior; at the right, the nucleus motorius trigemini anterior; in the ventral part of the Geld, the nucleus reticularis superior, van dcr Hard.

In Tinea vulgaris the fiber bundles constituting the cephalic root of the abducens have been traced to three nuclei of origin (Anens Kappers, ’12, ’20). The largest part of the fibers arises from an anterior ventral nucleus ; part of the fibers come from a ventrolateral nucleus,^ embedded in the substantia reticularis ventrolateralis, and turn medialward and tlien ventralward to their point of emergence, while the third and smallest portion arises from a small cell group

Telia (’09) found more than two abducens nuclear groups in embryological material. Since

this nucleus is formed during development from the periventricular gray and only gradually migrates to a ventral position, it is not surprising that it should be represented early by several coll groups. In primitive teleosts, van der Horst (’18) found more than two cell groups associated with abducens root Bbers (compare Hgs. 240D and 241, Megalops).

” In the diagrams of Tinea it was not possible to depict this group since it lies at the same sagittal and dorsoventral levels as does the main ventral nucleus.

548 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

situated in the vicinity of the medial longitudinal fasciculus. Thus in this animal, the neurobiotactic differentiation has gone extraordinarily far.

It must be emphasized that in the flat fishes the chief nuclei of the abducens and their corresponding roots lie somewhat farther forward than in many other

Fig 253 The trochlear nucleus of the eel, Ancuilla mg represents a composite figure made by projectmg sections on a single plane van der Horsl

The drawa number of

teleosts. Thus the anterior abducens nucleus falls in the plane of emergence of the facial root, and the nerve fibers from this nucleus in Pleuronectes and Rhombus emerge in front of the facial root fibers. This forward shifting of the nucleus in flat fishes is quite in harmony with the greater development of the frontal reflex paths in these animals. Among such paths are the ventral tecto-bulbar tracts, which are developed to such an enormous extent in flat fishes, probably in conjunction with the peculiarities of their eye

„ • , , , , movements, and which

lx considerable degree nothin the nuclei of the abducens nerve

mimtlln succounding them. Van derHarsl found this frontal

esDMiallv devolo™? conspicuous m Gasterosteus and in several other fishes with on?v (or It, least “y® (Tetrodon, fig. 249, and Orthagoriscns, fig. 245C). Whv the whllfnncl™ ^ *'■= *=iucens nucleus, instead of

Inl strse of th ’ 1“ fvv “ ““ '““"g « -ot PC« wlbablTat thll, "P ‘ccto-bulbar tract, it is

with the terminatin ^ *^f fin ^ Pecleus comes into direct synaptic relations Sfctld neXXtl 11 K X” ““ P”"™ of ‘I-e nucleus

iXn tXtX h i; ^ T. over the tract. This does not

of the abducens niiel ^ ^ eomewhat similar position to that

«vl poStn Ta “ P'-C'P-Pably tte more primi t ml Ttte nrimb- 5' X* “ factor in the reten nudeus W IZ k " “ verted upon this portion of the

trolaterahl (Zm.‘^ Petavo-motorius cruciatus ven it dels not ZeZoT “ *” P“®‘PP°' nMucens nucleus while

appear to have any connection with the anterior abducens nucleus.

THE EFFECTORY SYSTEM

549

The efferent trigeminal nucleus shows distinct differences in different representatives of the group, but has certain constant features. In harmony with the conditions in ganoids, it extends far caudalward in teleosts to the level of emergence of the facial root and even behind it (Siluris, fig. 243B ; Lophius and Orthagoriscus, fig. 245B, C).

However, in almost all bony fishes, caudal to a group of more dorsally situated cells, is a second group which takes a more ventral position (fig.

251). This latter group is large in proportion to the development of the secondary gustatory tract of Herrick (’06, bibliography, p. 421). Thus the ventral shifting involves many efferent trigeminal neurons in Tinea and Arius, where this gustatory tract is large (compare figs. 145 and 252), and this enlargement of the nucleus is most conspicuous at its caudal end.

As was stated previously (Chapter III), this gustatory path takes a more dorsal course at about the place of emergence of the trigeminal root. Thus it is the more cephalic part of the trigeminal nucleus which extends farthest dorsalward, and the caudal part which reaches farthest ventralward. In teleosts, then, the trigeminal nuclear gray is separated completely into cephalic and caudal parts (figs. 243 and 245). In embryos Telia ^09) divided the anterior or cephalic trigeminal nucleus into two parts, a ventral and a dorsal portion. According to this observer, the dendrites of neurons in the ventral portion are in synaptic relation with the tractus tecto-bulbaris ventralis. According to Berkelbach van der Sprenkel (’15), a marked ventral shifting of trigeminal cells occurs in siluroids, in correlation with tli^ presence of a very large gustatory tract.

The trochlear and oculomotor nuclei show great peculiarities in teleosts. The trochlear nucleus lies lateral and slightly dorsolateral to the fasciculus longitudinalis medialis (figs. 253, 254), from which it receives many collateral fibers, although, according to Beccari (’12), no terminal fibers. In reality it lies at about the same level as in sharks, although apparently it is farther frontal because of the relatively caudal plane of emergence of the trochlear root in teleosts, which makes the distance between the place of emergence of the root

Fig. 254. The course of the trochlear root m the fresh-water perch. van der Horst.

550 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and the plane of the trochlear nucleus greater in Ihcsc latter animals than in the sharks. A marked peculiarity of the trochlear root in certain tcicosts is its subdivision by the tractus me.scnccphnIo-cerebellaris posterior (tr. mes. ccr,, fig. 254). For the details of this peculiarity in the course of the root fibers, reference is made to the publications of find (’ll), of Vidor Franz (’ll), and particularly of van dcr Horsl (’18). Only a very brief statement follows. After its origin from the nucleus, in many teleosts one p.art of the trochlear root pas.'ies directly around the lateral angle of the aqueduct and enters the velum meduliare antcrius, in which it cro.sses and then turns latcralward before emerging; another part, after leaving the nucleus, turns more peripheralward, bre.aks through the tractas mesencephalo-cerebellaris posterior, pa.ss&s around its lateral side, and again runs fonvard within the anterior medullary velum. In animals which have a large valvula cerebelli which extends far under the tectum, still other variations

nooi of iiirr

Vcntro-Iet. DU**!. II!

Annulate cotn

Fig. 255. The oculomotor nuclei of Gadui morrhun

|Dnr,o-i.t.nuoi III troclilcar root are found, certain of which have been described by van dcr Hard. Frequently these variations are due to the influence exerted by the transfonnation of the velum into a vahaila and the resulting changes in the surface of the velum. In most teleosts the trochlear nucleus passes over directly _ . intothenucleusoftheoculo ^ exists between the two nuclei through which fibers pass. IS latter condition is found in many primitive teleosts (but also in Perea, modytes, Synbranchidae, Anguilla, Hippoglo.ssus, and Rhombus). VTiat factors determine the continuity of these nuclei in teleosts or the lack of it (as the fS.urtn " yet unknown. Such a difference in relations is the more

flat fishpq occurs wdthin a related group, as for example in

not in the nuclei are continuous with each other, but

not in certain other flat fishes.

cells^thir,l°r'”^”^r ^ ventral shifting of one part of its

of the brain onW^h *'^^t they are separated from the base

o ten Ltemis fpS ' -n^ulata (fig. 255). This ventral portion

uporthembvte" neurobiotactic influence exerted

rtecto busbar t^ (’10) «h°"'n that

bu!bar diu ^ l<^vels. The dorsal tecto portion of the opiilnm^ ^”i ^ ^ ^ the cephalic end of the dorsal system appears onlv ^ ^ creasing of the ventral tecto-bulbar

nucleus reachpp ift: a- f f where the ventral portion of the oculomotor

grea s ventral extent. Impulses reach the dorsal nuclear

THE EFFECTORY SYSTEM

551

mass by way of crossed and uncrossed fibers of the dorsal system and the ventral nuclear portion by similar fascicles of the ventral system. Since obviously crossed fibers could distribute to the nucleus only after decussation, it is evident that each portion of the nucleus has migrated to the position most favorable for its reception of stimulations. The oculomotor nucleus receives collaterals and terminal fibers from the fasciculus longitudinalis medialis, apparently arising from nuclei of the acoustic (vestibular) nerve (Beccari, ’12, ’31 ; see p. 506).

It is not possible to state with certainty from which part of the oculomotor nucleus arise crossed and from which part uncrossed fibers. In general, the ventral part of the nucleus in teleosts — as in birds, but not in mammals — appears to give origin to crossed root fibers. Such fibers are much less numerous than the uncrossed fibers, and less in number than those of birds and mammals.

In Ceratodus {Holmgren and van der Horst, ’25 ; fig. 256A) a relatively simple arrangement of the motor nuclei is found, resembling that in plagiostomes and in certain ganoids. The caudal visceral efferent colunrn is formed by motor neurons of the facial, glossopharyngeal, and vagus nerves, but with another small motor facial nucleus in front of the nucleus of the abducens in Ceratodus (fig. 256 A). The trigeminal and abducens nuclei lie approximately at the level of entrance of their respective roots. The trochlear nucleus is farther caudal than in plagiostomes, approaching the level of its root. The oculomotor nucleus appears to be independent of the trochlear and to lie at the level where its fibers leave the brain.

The Effectory System of Amphibians

The arrangement of the motor roots and of their nuclei in the medulla oblongata and midbrain of amphibians is more primitive than the arrangement found in teleosts. In certain respects these relations resemble those in Dipnoi, as brought out particularly by the excellent work of Holmgren and van der Horst (’25, see p. 632), and in plagiostomes; in other respects they form the basis for the arrangement found in higher vertebrates, including mammals. The researches of Aricns Rappers (’12), Rothig (’13), and Herrick (’14 and ’30) indicate that the urodele amphibians show a particularly close resemblance to the above mentioned fishes. Not only the perennibranchiates but the caducibranchiates as well, which breathe by means of gills for only a short time, show this resemblance to the Dipnoi and to plagiostomes. This is indicated by the positions of efferent roots supplying the gills, particularly the roots of the facial nerve.

The ventral horn gray extends about as far forward in the amphibians as in the lung fishes (fig. 256). According to FUrbringer (’97), urodele amphibians usually have no occipital nerves. However, exceptions to this are found. Thus in Triton, Driiner (’01) described such a nerve, which, after its exit from the skull by a separate foramen, fused with the ventral root of the first spinal nerve. Spino-occipital nerves have been described for Cryptobranchus alleghaniensis {Driiner, ’01) and japonicus {Fiirbringer, ’97), and for Salamandra {Driiner, ’01).

Kingsbury (’95), Norris and Buckley (’ll), and Herrick (’30) found that dorsal roots and dorsal root ganglia are lacking in the first two spinal nerves of

552 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Necturus. In the specimen described by Herrick, three rootlets made up the first spinal nerve. Of these, the two more cephalic roots emerged in front of the calamus scriptorius, and the third either at its level or slightly caudal to it. The first two rootlets arise from large bipolar neurons, the cell bodies of which occupy

K'T I ' L mclaalrootin |||||||||||||| _ md ud rool IT adailrMlV

micLuidnjolTII

Bid. VI

bkLuIimIIX

_ docL and ml X lad XI; ctSt rlaiIiosid(cl)

■idDdIcn

imclai]dmts;iii.oa.iiidXII I | _ ad L&ftr.Wedpy

inltrioroErarjiiod tYII,IX= ftraVUIX 1 olamciIjitNiB

lung fish, CerSodus^orsterh of Triton “

A. Ceratodus forsteri. van der Horst

B. Triton cristatus. van der Horst.

principal abducens^nudeus, has'been added^by^ddens abducens nucleus, lateral to the

LTwargf lon^tudinal fasciculus. Each of such cells

the raph6 and the cells of tL motorT^^^^^ laterally, reaching respectively first spinal root The oplk f • • while the neuraxis enters the

sp nZerlroLrlL'i the first

roT' SZirL“*h cells „I origin the hypoglossal nucleos by

THE EFFECTORY SYSTEM

553

they represent a stage toward the development of the hypoglossal nucleus of tailless amphibians, as described by Black (T7) in the frog. Norris (T3) found that the hypoglossal nerve is formed in Amphiuma from the ventral ramus of the first spinal nerve. Usually, however, it is formed by a fusion of the ventral rami of the first two spinal nerves, as in Amblystoma and presumably in Triton {Coghill, ’02, ’06) and in Siren lacertina (Norris, ’13).

In all the adult opisthoglossal anurans studied by Fiirbringer (’97), he found that the first spinal nerve, or occipito-spinal nerve, was absent ; a similar condition was found by Gaupp (’99) in the Ranidae which he examined. Consequently Black (’17) considered the first spinal nerve in the adult Rana catesbyana as really the second spinal nerve. The more rostral rootlets of this nerve arise chiefly, if not entirely, according to Black, from a dorsomedial nucleus, which lies in the reticular gray forming the somatic efferent column. There is sufficient cytological differentiation of the cells of the dorsomedial nucleus so that the limits of this cell group can be predicted reasonably well. This nucleus and certain of the rootlets arising from it are considered a hypoglossal nucleus and nerve, respectively. Other motor rootlets, termed merely motor roots of the second spinal nerve by Black, arise to some extent from this dorsomedial nucleus, but largely from a ventrolateral nucleus which does not extend so far forward, and in reality is ventral horn gray of the cord.

From the above accounts, it is evident that in tailless amphibians it is possible to separate the gray of the somatic efferent column into a distinct cell group for the medulla oblongata and a group for the ventral horn region of the cord. It might be added further, that whether the hypoglossal is formed from the ventral ramus of the first or the second spinal nerve or by a fusion of both rami, its peripheral distribution is essentially the same.

The visceral efferent column presents certain points of interest in the historj’- of the phylogenetic development of the motor centers in vertebrates. In the vagal complex of Necturus, Herrick (’30) counted five clearly defined motor roots. One or more of the three more spinal of these roots probably carries accessory fibers, according to this observer. The first two of such roots arise from neurons which are situated well behind the plane of emergence of the root fibers. These are large motor neurons which lie near the ventricular floor, lateral to the medial longitudinal fasciculus, and send their neuraxes lateralward to the emergent vagus roots. Efferent fibers of the glossopharyngeal nerve follow a course similar to those of the vagus. Its fibers, arising from cells similarly situated near the ventricular floor, swing forward and lateralward to their superficial origin. In this same efferent nucleus for the vagus and glossopharyngeal nerves are neurons which give rise to neuraxes entering a facial root. The limits of this common motor nucleus for the facial, glossopharyngeal, vagus (and perhaps accessory) nerves are not defined sharply; its cells intermingled medially and laterally with cells of the motor tegmentum. The caudal and cephalic limits of the nucleus are not clearly determined (Herrick, ’30) ; it is known to extend both cephalic and caudal to the planes determined by emergent roots of the vagus and the glossopharyngeus. It is of interest that the neurons supplying

554 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

efferent fibers to the facial nerve do not occupy the most cephalic part of the nucleus. Root fibers of the facial nerve arise well caudal to a plane through the first motor roots of the vagus and extend forward through the nueleus to a point behind the plane of emergence of the glossopharyngeal root, at which level the facial fascicles, according to Herrick's account, take a position dorsal to the medial longitudinal fasciculus, where they extend forward as a medial bundle (constituting root fibers of the first motor root of the facial) and a lateral bundle (composed of root fibers of the second motor facial root). Near their plane of emergence the root fibers run lateralward and slightly forward, and pass through the descending root of the trigeminal to their superficial origin. Cells of origin of the facial nerve lie also in the nucleus motorius tegmenti, in a plane immediately caudal to the emergence of the roots. The neuraxes of these cells join the other motor fibers of the facial. Similar relations of the facial occur in larval Amblystoma {Herrick, ’14, ’30).

Judging from various accounts in the literature, the relations in the visceral efferent column vary among different tailless amphibians. Thus in Rana esculenta (Ariens Kappers, ’12) and in Rana fusca and Bufo (Rolhig, ’13), the motor nuclei of the glossopharyngeus and vagus form a continuous gray column very similar in general position and relations to that described for iirodeles (Herrick, >’14:). In Rana catesbyana, however, both Black (’17) and van der Horst (’18) found a hiatus in this caudal visceral efferent column between the nuclei of origin of the glossopharyngeal and vagal motor roots. The motor root of the facial nerve arises from a nucleus situated at the edge of the ventricular gray of the medulla oblongata. In the greater number of frogs studied (Ariens Kappers, ’12, ’20 ; Rolhig, ’13 ; Black, ’17), it lies mainly at the level of superficial origin of the root fibers, but extends also slightly caudal to that level. In Bufo, Rothig (’13) found this motor nucleus of the facial nerve in a plane entirely behind the place of emergence of its root fibers. It is to be noted that this nucleus is independent of the glossopharyngeal-vagal efferent column and presumably is represented in urodeles by the motor neurons of the facial, found in the nucleus motorius tegmenti of Necturus and even of larval Amblystoma by Herrick (’30). In these latter animals, as was stated previously, the larger part of the nucleus is related to the cephalic end of the glossopharyngeal-vagal column, a relation as yet not described for anural amphibians, but which resembles the conditions found in Dipnoi and plagiostomes (see fig. 256, A, B, C).

What causes underlie these differences between tailed and tailless amphibians as yet is not well imderstood. It is true there is a great resemblance between Dipnoi and plagiostomes, and those urodeles which possess gills either without an operculum or with only a very small one. This has resulted in the massing of the neurons supplying efferent fibers to the gill musculature into a single column under the influence of the caudally placed visceral sensory centers which receive impulses from the gill region and which show much the same relations as in plagiostomes. However, this explanation does not make clear the persistence of this arrangement in caducibranchiates, which are not gill breathing throughout their lives. It may be suggested that

THE EFFECTORY SYSTEM

555

the sensory facial nucleus, which in the first place is concerned with reflexes through the motor facial center and consequently determines the position of this latter center, has about the same position in caducibranchiates as in the gill breathing amphibians discussed above, and is better developed in perennibranchiates. This nucleus holds the efferent facial center to its primitive position in the glossopharyngeal-vagus column. Thus in Triton the position of this facial nucleus is comparable to that of its homologue in Necturus (fig. 25GB). In tailless ampliibians the gustatory impulses are no longer of such paramount importance (Chapter III).

The nucleus of the abducens nerve is a derivative of the floor plate of the embryonic neural tube and is situated in adult urodeles (Necturus) near the midline, among the cells of the primary motor column {Herrick, ’30). On the whole, its cells appear rather small, but its dendrites have a wide spread. The neuraxes leave the brain as two small but well medullated motor roots.

In transverse planes the main abducens nucleus occupies about the same position in urodele and anural amphibians. Adderrs (fig. 25GC ; also 1933 account, p. G28) has identified in Rana a cell group situated lateral to the main nucleus of the abducens, which he interpreted ns the homologue of the accessory abducens nucleus describetl by Tcrni in amniotes (see p. 5G3). Gage (’93) indicated that considerable variability in the superficial origin of the abducens nerve is to be met with among amphibians. In Rana fusca and Bufo, Holing (’13) identified two rootlets, while three were described in Rana catesbyana by Black (’17) and in Rana esculenta by Ariens Kappers (’10), and two in Rana catesbyana by van der Horst (’18) and Addons (see charts).

The trigeminal nuclei in Necturus and in larval Amblystoma {Herrick, ’14, ’30) produce a slight eminence on the floor of the fourth ventricle which has been termed the eminentia trigemini. The medial part of this elevation is occupied by the motor nucleus of the trigeminal, the lateral part by the sensorj’^ nucleus of this nerve. The limits of the motor nucleus, according to Herrick, are not easily definable, since it fades out into the nucleus motorius tegmenti. Rostral and caudal parts of the trigeminal nucleus, associated respectively with the first and second roots of the nerve, have been described in Necturus. Herrick suggested that this division may be due to neurobiotactic influences, although for this there is “no evidence except the topographic relations of parts.’’ He thought it probable that the lower part of the nucleus is affected principally by gustatory and other sensory impulses, while the upper part is primarily under the influence of such efferent systems as the tecto-bulbar and dorsal tegmental tracts.

Little need be added regarding the trigeminal efferent nucleus in tailed amphibians. It appears to be a single nuclear mass which lies partly at the lev'el of emergence of its root fibers, but with about half of the nucleus caudal to that level. It occupies a distinctly dorsal position. In the frog it is separated from the facial nucleus by a small but absolutely distinct gap. The cells of the facial and trigeminal nuclei are very similar, but in Rana {Black, ’h) the trigeminal nucleus is more sharply delimited than the facial and approximately

556 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

twice its size. After their origin from the nucleus, the more caudal trigeminal root fibers run somewhat frontalward, join those arising at that level, and emerge in a plane through the cephalic end of the nucleus (see fig. 256).

In the material of Siren lacertina available for study, Ariens Kappers (T2) was not able to identify a trochlear nucleus or trace a trochlear root, but Norris (T3) identified this greatly reduced nerve in his series of Siren. V on Plessen and Rabinomcz (’91) were unable to find this root in the salamander, but it has been seen in this form by other observers {Armns Kappers, ’12; probably others also). In general, Triton and Rana present different relations with regard to the trochlear and oculomotor nuclei, for the trochlear nucleus, which in urodeles lies at a very considerable distance caudal to the oculomotor nucleus, in Rana catesbyana {Black, ’17) approaches close to the oculomotor nucleus but does not join it, and in Rana esculenta {Ariens Kappers, ’12) comes into direct continuation with that nucleus (compare fig. 256 ; for Necturus and Cryptobran chus, see Rothig, ’13). Sometimes the trochlear nucleus shows a frontal extension which may be regarded as a beginning of an extension in the direction of the oculomotor nucleus {Rothig, ’13) and so as a transition stage. The root fibers of the trochlear nerve pass lateralward and then dorsal- and caudalward to their decussation in the anterior medullary velum. Uncrossed fibers entering with these root fibers at the superficial origin of the nerve may be related to cells of the mesencephalic root of the trigeminal, as McKibben (’13) suggested for Necturus.

In several amphibians the oculomotor nucleus may be divided into a medial and a dorsolateral portion, the forerunners of the distinct divisions observable in reptiles {Ariens Kappers, ’12; Black, ’17). No ventral oculomotor nucleus such as is present in teleosts is observable in these animals. No satisfactory analysis of the relations and localization of crossed and uncrossed oculomotor fibers has as yet been made for amphibians.

The Effectory System of Reptiles

The motor system of reptiles presents a number of stages of development which are of importance in the consideration of the origin of the mammalian and avian types. This is particularly evident in regard to the relations of the hypoglossal and facial nuclei and the nuclei of the eye muscle nerves. The cytological differentiation of the motor centers of the vagus and glossopharyngeal nerves is not great in reptiles. Such differentiation could not be expected in animals in which the higher organization and specialization of the muscles of the pharynx and larynx have not appeared as yet.

There is a difference of opinion with regard to the differentiation of the hypoglossal nuclei of reptiles. At least a forward shifting of this nuclear group is seen in these animals, in correspondence with the development of gustatory endings in the tongue. In Caiman sclerops (which is illustrated in figure 257 and which has been studied carefully by Addens), a distinct dorsal hypoglossal nucleus has been identified. Addens found a ventral hypoglossal nucleus, which he regarded as directly continuous with the accessory nucleus of the spinal cord

THE EFFECTORY SYSTEiH

557

Fio. 257.

Hypoglos-snl nucleus in Cnimnn. lan IIocicll

(for the evidence see the 1933 paper of Addciis). This relation resembles very much the conditions described in the Italian sparrow by Bcccari (’22) and is also confirmed by the avian material studied by Addons. The dorsal hypoglossal nucleus extends farther forward than the ventral. The reconstructions of the hypoglossalaccessory complex as illustrated for reptiles other than Caiman (fig. 259) are incomplete.

They are receiving further invc.stigation in line with the work on Caiman. Tmjc (’32) found a dorsal and a ventral hypoglossal nucleus in Clirj'scmys elegan.s. The course of the hypoglossal root fibers within the medulla oblongata is longer in reptiles than in fishes, and is comparable to that in mammals.

A study of the visceral cfTcrcnt sj'stem in reptiles presents certain points of especial interest.

The whole column lies farther laterallj' (fig.

2.'58) than in the fishes, and as a result its root fibers form a sm.all loop (fig. 152), since these roots are held to their primitive position in relation to the medial longitudinal fasciculus. The cephalic border of the column falls in approximately the same plane in all reptiles ; the caudal border varies in different reptiles, its position in hj'drosaurians, turtles, and lizards differing from its position in snakes (fig. 259.‘\.-D). The visceral efferent column extends farthest eaudalward in the turtle and in Caiman, in fact, it extends farther than

.luitmK of indicated in the diagrams (fig. odiiofnucLX 259A-B), being still clearly evident in the last slide of the series from which these diagrams were made. Certain observers who regard the spinal accessory nerve as a special visceral efferent nerve have endeavored to find its nucleus in the caudal part of the vagal column in many reptiles. The absence of this nerve, together with the shortening of this column in reptiles which lack extremi

Fia 258.

Figure illustrating the lateral shifting of the neurons of the vagus nerve in Caiman.

ties, has been regarded as evidence for such a localization of its cells of origin. The greater extension of the column in aquatic animals is explained then, as due to the presence of a fairly well developed trapezius musculature, as the researches of Furbringer (’00) have made clear. However, according to the work of Beccari (’13, ’14), only the cranial and not the spinal part of the accessory is to be found

Kej

_ Duel an] root m _ nod anJ rool VII

_ nod aid rool lY

end VI

rad ami root X and XI; ccOa rXcoIiaaad[(dL)

_ mdaodnolV _ md and root K _ nidDtdcn

- nod and root ipk, oo. and ffl | | — md Edbfrr-WejtpluJ

_ inferior oBrarj rad snpenor oBnij dqcL •I— calamos scr^nni

Fig. 259. Charts illustrating the relations of the motor nerve roots and their nuclei in various reptiles.

A, Chelone mydas. Ariens Kappers. C. Varanus salvalor. Anens Kappers,

a. Caiman sclerops. Addens ('33). /). Boa constrictor. Ariens Kappers.

558

THE EFFECTORY SYSTEM

559

in the caudal part of the column. The spinal part arises from the cervical cord gray which is continued caudally as the nucleus of origin of the fibers of von Lenhossck, which leave the cord through the dorsal roots. Addens, who regarded the spinal accessory nerve as a somatic efferent nerve, has advanced the theory that the spinal accessory is formed merely of parts of the first ventral spinal roots which have shifted from their primitive ventral emergence to the dorsal places of emergence characteristic of the spinal accessory of higher forms. In crocodiles the accessory root can be traced as very thin fascicles to about the second cervical segment. It emerges dorsally. It arises from a distinct cell column continuous with the hypoglossal nucleus, according to Addens (’33).

A second point of importance in the development of the posterior visceral column is the appearance of a ventrolateral portion of the nucleus in reptiles. The simplest type of shifting is found in Chelone, where cells in between the middle and posterior thirds of the column acquire a distinctly ventrolateral position although retaining a connection with the dorsal part of the column throughout. A similar ventrolateral shifting of a part of the posterior visceral efferent column is present in Caiman, but the nucleus thus formed is separated for a greater distance from the dorsal column (figs. 258 and 259). In Varanus, also, in the same topographic position, a shifting of cells is observable. It shows an independence comparable to that found in Caiman. In more caudal regions of the brain the visceral efferent column occupies a more ventrolateral position than is indicated in the diagram of Varanus (fig. 259C) and shows relations strikingly similar to the conditions found in embryonic sheep. In Boa this ventrolateral nucleus is either very small or lacking. Originally this was regarded {Ariens Kappers, ’ll, ’12) as the beginning of an accessory nucleus only. Careful comparisons with the condition in birds and lower mammals show that it corresponds to the posterior third of the nucleus ambiguus, which in this portion in mammals gives rise to both vagus and bulbar accessory fibers (see various texts on Nervous Anatomy, such as those of Ranson, '31, and Herrick, '31). In addition to the nucleus of the vagus just described, Addens (’33) found a second ventral nucleus in Caiman, situated rostral and ventral to it.

The nucleus of the glossopharyngeal nerve has not been identified with certainty in all reptiles. In most reptiles the root fibers of the glossopharyngeal nerve appear to take origin from the same column of gray from which the facial nerve arises. However, Addens has been able to distinguish a discrete glossopharyngeal nucleus in Caiman (see fig. 259B).

In Caiman (fig. 260) three facial nuclei are present — a dorsal, an intermediate, and a ventral nucleus. Of these, the two more dorsally situated (the dorsal and intermediate of the above terminology) were recognized by earlier observers (Ariens Kappers, ’12, and Black, ’20). The most ventral nucleus has been identified recently by Addens (’33 ; also fig. 259B). In Varanus the entire complex lies farther ventrally (fig. 261), and a facial genu, as in Caiman and similar to that in mammals, is seen clearly in figmre 315E. The migration of this complex occurs not only in a ventrolateral but to some slight extent in a caudal direction. The caudal migration is least marked in Caiman (compare

Nucl.VI

Nucl. VII I

Sup. oliva:

QUcl

Fig. 2G0. The facial and abducens nuclei in Caiman.

5G0

THE EFFECTORY SYSTEM

561

fip. 259R). Finally, it is to ho noted that, the facial nucleus lies medial to the superior olive in those reptiles in which the nucleus has been identified. This is particularly evident in Caiman, since there the facial nucleus and the superior olive lie in the same transverse plane and their relative j)ositions arc easily recoRnizahle. It is to be noted that in many mammals the facial nucleus is dorsolateral rather than medial to the superior olive.

The study of the eye muscle nuclei of reptiles is of particuhir interest, since within these animals apijcar those changes by means of which the slightly difTcrentiated and somewhat diffuselj' arranged nuclei of lower vertebrates go over into the highly difTercntiated, compact nuclear mas'^es of higher forms.

In Caiman, as well as in Chelone, the abducens nucleus shows many chanicterislics reminiscent of this cell group in amphibians. In the aquatic reptiles this nucleus still consists of scattered cells and the greater portion of it lies far caxidal to the plane of emergence of the facial root, extending even to that of the glossopharyngeal root (fig. 2.5t)B of Caiman). The position of the abducens cor rc.sponds with that of its nucleus (fig. 262), so that fascicles may leave the brain throughout all levels from the plane of emergence of the facial nerve to that of the glossopharyngeal nerve. In no other animals are they known to be scattered over so great a distance (fig. 259A and B), a condition brought about by a frontal migration of certain of the cells which persists from reptiles through higher forms in association with a caudal position of the main nuclear mass corresponding to the position of this cell group in certain subreptilian forms. This frontal migration is most marked in higher reptiles. Thus Boa, as compared with Chelone, has a distinctly less well developed caudal pole to the nucleus and the nucleus no longer extends over the whole distance between the facial and glossopharyngeal roots.

Fig. 202. TIic ntidiircns niirlcu.s of Cnimnn. lOTi lloei'clt.

562 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

At the same time, all of the root fibers have acquired a frontal position and emerge from the medulla oblongata in front of the level of the facial root (comyjjj pare figs. 259A-D). This process shows

its greatest development in Varanus, where conditions approaching those found in birds and mammals may be seen (figs. 259C and 304) . In this reptile the whole nucleus has migrated so far forward that it extends frontally beyond the level of entrance of the facial root fibers and, as in mammals, lies in part in front of the facial roots. At the same time, the nucleus has become very compact. This frontal migration oj the dorsal nucleus of the abducens in the reptilian group is evident not only by comparing it with the facial and glossopharyngeal roots, but by noting its position with respect to other structures of the

Fig. 264. The caudal portion of the trigeminal nucleus in Caiman aclerops. 6, reticular cells, van Hocvell.

medulla oblongata, as is indicated in the reconstructions given. The nucleus appears to be influenced by the reflex paths asso

ciated with the lateral vestibular or Deiters’ nucleus, since it evidently approaches this nuclear mass. In the diagrams for Caiman, Boa, and Varanus, the position of

the Deiters’ nucleus® is indicated

® The nucleus of Deiters is not so clearly developed caudally in reptiles (as is apparent from the diagrams) as in mammals. In these latter animals it extends to about the level of entrance of the glossopharyngeal nerve roots (see page 593). Leioy (TO ; see p. 513) thought that the caudal portion of the mammalian nucleus only atrophied when the cervical cord was cut, and that the anterior part, which remained unchanged, served the bulbar vestibular paths as a nucleus of origin. Rasmussen (’32; see p. 489) located the cells of origin for the reflex paths to the eye muscle nuclei in both the medial vestibular nucleus and in the nucleus of Deiters ; Gray (’26 ; see p. 489) placed them in the medial vestibular nucleus. These differences in interpretation are discussed in the preceding chapter. It seems evident that the vestibular cen

ters at this level (either the lateral Fig. 265. The trochlear nucleus in a turtle. S.J.de Lange. vestibular or the immediately adjoin ing medial vestibular area) exert a marked influence on the position of the abducens nucleus. This matter will be discussed again in the account of the abducens nucleus of mammals.

TIIE EFFECTORY SYSTEM

563

by the field D-D (fig. 262, Caiman), and it is apparent that the abducens nucleus gradually approaches this cell mass and tends to retain a position near its level.

An accessory abducens nucleus, situated ventromedial to the descending root of the trigeminal nerve and in close proximity to it, has been found in reptiles {Tcrni, '22, Gongylus occllatus). Adde7is (’33) >

also saw this nucleus in reptiles (see the chart of Caiman, fig. 259B). In this accessory abducens nu- / cleus lie the cells of origin for the fibers to the retractor > / bulbi and to muscles of the nictitating membrane. 1[

The impulses which set off this reflex enter through the sensory trigeminal root and terminate in the descending 'j •. 4’* nucleus, which accounts, presumably, for the close in- '/'.'.v terrelation in position between this latter nucleus and ' the accessory abducens nucleus (Tcrni, ’22a). This relation is tested easily, for when the snout of the reptile is touched the e^’e is retracted and the nictitating mem brane comes over it (Tcnii, ’22a, Arch. Fisiol., vol. /

20, p. 305). ‘ . ; ■ ' ^

In the aquatic reptiles the greater part of the tri gerainal nucleus lies dorsally (sec fig. 259). A small Fio 2C0 Tho oculomotor part of the nucleus, and alwap its caudal part, has nucleus of Caiman, moved into a ventrolateral position (Ariens Kappers, ’20). In Chelone (fig. 263), and particularly in Caiman (fig. 264), this ventral part of the trigeminal nucleus is ^^siblc ; in the latter animal it is separated completely from the dorsal portion, a condition occurring also in teleosts. In many teleosts this separation is due to the great development of the secondary gustatory tract. It is probable that

in reptiles its position is due impulses reaching it from

Dons, nua III- — ^ ‘'Vi'. ■ I* V.' the nucleus of the descending

pvji , V" trigeminal. In the

boa constrictor the whole tri Vcntnucim » ^ gcminal nucleus has moved

laterally, away from the ventricle toward the sensory tri't' f ' ' ' geminal root and its associated

r'K gray. There is still some ques .Iti ft! , I ‘ { I II I - ^ \ .

' l' whether or not this

Fia. 2G7. Tho oculomotor nucleus of a c^mclcon. shifting in position is related

directly to the marked development of the branchiomeric musculature in these animals and the intimate rela

Fia 2CG Tho oculomotor nucleus of Caiman.

Dors. cud. Ill

Vent, cud in

-TV

(- i •' .J . ’ ■

Fia. 2G7. Tho oculomotor nucleus of a chameleon.

tion existing between this musculature and the cutaneous sensory impulses from the mouth region. A third, still more ventrally situated trigeminal nucleus has been identified by Addens (’33) in Caiman (see fig. 259B). Part of the motor trigeminal fibers appear to have a contralateral origin in the alligator and possibly in other reptiles (Huber and Crosby, ’26).

564 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The trochlear and oculomotor nuclei of reptiles exhibit all the stages of differentiation from that represented by the simple, slightly differentiated nuclei of Chelone to that of the highly differentiated nuclei — particularly of the oculomotor — found in certain higher reptiles and suggestive of the conditions in birds and mammals. The mutual relations of the nuclei of the trochlear and oculomotor vary in different reptiles. In Caiman, Chelone, and Boa, the.se nuclei are not joined together — a distinct gap separates the two (fig. 259 — whereas in Varanus the nuclei are in close contact with each other. In one species of Varanus (the name of which is unknown, Ariilns Kappers, ’20), there is even a pronounced overlapping. The trochlear nucleus and its roots show the usual relations. The nucleus (fig. 265) lies dorsally, slightlj’’ laterally

mcdlal longitudinal fas ■d lT I 1 1 1 M 1 1 1 1 1 MTh^ J ,l 1 1 1 1 1 1 1 1 1 iRtH+ll I t ciculus. It consists of the usual

Ifirgc, polygonal colls and shows

rm 7 no division into smaller nuclear

^11 "i" Effl E ..iaUJ

oro A 1 - . ■„ groups. In reptiles the troch I'lo. 268. A diagram to illustrate the rclntion.s of the i • i a

trochlear and oculomotor nuclei in C; u-man scieropa. ^s yet sur rounded by the small and large reticular cells as is the case in mammals. Attention is called to the trochlear nucleus as this is seen in a specimen of Varanus, whore the nucleus is in the shape of an arch which does not arise, however, from the floor of the aqueduct,

^ ut which arises in the dorsolateral wall of the ventricle, a position of interest in view of the primitive relations of the trochlear nucleus in cyclostomes [Ariens Kappers, ’12, ’20).

The oculomotor nucleus varies in different reptiles. The simplest condition, accor ing to ricns Kappers, '20, is found in Chelone, where the midline 1^0 ongation of the ventricle penetrates deep into the midbrain to ivithin a ^ ventral surface. Lateral to this ventricular prolongation,

and between it and the medial longitudinal fasciculus, lies the oculomotor nucleus. A dorsolateral wing-like extension is barely discernible ; consequently the distinct dorsolateral and medial groups seen in certain reptiles cannot be recopized here. However, in Chrysemys Tugc (’32) described dorsal, intermediate, and ventral portions of the oculomotor nucleus, all portions lieing in contact with each other. The ventrolateral portion forms a more compact cell mass. In Caiman (fig. 266), a nuclear arrangement is easily made out. The w ole nucleus is confined to the more dorsal regions of the tegmentum, lying etween the raphe and the medial longitudinal fasciculus. The dorsal cells e ocu omotor extend laterally over the medial longitudinal fasciculus and again accumulate on its lateral side so that a ventromedial nucleus and a dorsorecognized. Caudally both groups begin at about the e level (fig 268) and here the dorsolateral group approaches the trochlear nucleus Cephahcally there is a great difference between the two nuclei, the dorsolaterd nucleus extending much farther fonvard than the ventromedial nuc eus. The separation of the oculomotor cells of origin into two principal uclei, a dorsolateral and a ventromedial, is more evident in the chameleon (fig.

THE EFFECTORY SYSTEM

565

267) and the lizard than in Caiman. In the two former animals, just as in the latter, the dorsolateral nuclear group extends farther frontalward, while caudaUy both nuclei terminate at about the same level, with the dorsolateral group going over gradually into the trochlear nucleus. In Varanus another cell group, present as a usual feature in birds and mammals, has made its appearance. In planes through about the middle third of the main dorsolateral nucleus and dorsal to that nucleus lies a definite cell group which, from its separation from the other groups and from the smaller size and somewhat different arrangement of its cells, is recognizable as a distinct nuclear entity.

From its relations and general appearance, it seems altogether probable that this cell mass just described is the homologue of the EdingerWestphal nucleus of higher forms. This nucleus was identified for the first time in reptiles (Varanus) by Ariens Kappers (see fig. 269).

It has been seen since in Caiman (sclerops) and in other reptiles by Addens (’33; see charts). As to whether or not this nucleus F'G’ 269. Cross section through the

actually contributes root fibers to the oculo- preparation colored in elderberry juice, motor nerve, even in mammals, has been a

matter of considerable dispute (see p. 616). Certainly it appears later, both ontogenetically and phylogenetically, than do other nuclear groups of the oculomotor. Experimental work on certain forms (such as that of Brouwer on the sparrow, p. 578) indicates that this nucleus is a preganglionic center, the fibers of which accompany other oculomotor root fibers from the brain, and that these preganglionic fibers synapse with postganglionic neurons of the ciliary ganglion for the supply of intrinsic eye muscles.

As yet it has not been possible to determine accurately the number of crossed and uncrossed oculomotor root fibers. It would appear that, on the whole, the number of crossed fibers is relatively less in reptiles than in mammals, and that the greater number of fibers arise from the ventromedial nucleus, as is the case in teleosts and birds.

Access. Duel.

Dorso-Iat nucl.

Ventro-med. nucl.

/

The,^ffectory System in Birds

The motor centers in birds present many interesting variations and transitional forms which throw light on certain factors involved in the differentiation of these centers in higher forms. The hypoglossal complex of birds presents a marked peculiarity as compared with that of reptiles. Ariens Kappers (’12) identified, as the cephalic representatives of the somatic efferent column of the spinal cord, a ventral and a dorsal column. Of these, the former is a direct continuation of the ventral horn of the cord. Further work by Beccari (’22) and by Addens (figs. 271A, B, C ; see also his 1933 paper) indicates that this

566 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

ventral column may be divided further into a smaller, more dorsal hypoglossal nucleus, and a larger, more ventral, cephalic continuation of the ventral horn gray. The ventral column has been regarded by Aricns Kappers (’ 12) , Black (’22), and Sanders (’29) as hypoglossal in character, although Kosaka and Yagila (’03) thought that it contained no true hypoglossal neurons. The more dorsal group, usually termed the nucleus intermedius, lies in relatively close relation to the cells of origin of the vagus. The intimacy of this relationship varies in different birds ; in many birds contact occurs at frontal levels of the vagus (Aritins Kappers, ’20) ; in the duck (Kosaka and Yagiia, ’03) and in Colymbus (Aricns Kappers, ’ll) the two nuclei are in contact near the caudal end of the vagus, while in Gallus there is no real contact (Kosaka and Yagila, ’03). The presence of hypoglossal neurons in this nucleus is indicated by tracing root fibers arising there into the emergent hypoglossal roots ; moreover, degeneration preparations indicate the presence of such cells of origin (compare this with Brandis,

’93a, and Kosaka and Yagiia, ’03). Such hypoglossal cells are found for only a short distance. They begin well behind the cephalic tip of the glossopharyngeal-vagus column and they never extend so far caudalward as does the dorsal vagus nucleus. The question as to whether or not vagus cells are found in this hjqjoglossal nucleus has been answered in the negative by Kosaka and Yagila (’03), Beccari (’22), and Groebhcls (’22). Brandis (’93a), Bok (’15), Ariens Kappers (’ll), and Sanders (’29) were of the opinion that such vagus cells are present, and that their presence points to a functional relationship between the centers, which will be discussed more fully later. Addens believed that they may be present (as in Sphenicus) or absent (as in Cacatua and Gallus) . The interpretation of Ariens Kappers will be considered first. It is sufficient to state here that the hypoglossal nucleus not only supplies the poorly developed tongue musculature in birds, but also — and chiefly — innervates those muscles which belong to the peculiar extension of the trachea just above the bronchi, known as the syrinx, an organ concerned in the production of sound. The musculature of the syrinx arises from the sternohyoideus, which is innervated in mammals by the ramus descendens hypoglossi. The so-called ramus laryngeus of the hypoglossal, which innervates the syrinx in birds, is homologous to this mammalian ramus (Ariens Kappers, ’20). According to Kosaka and Yagila^ ( 03) , the mammalian ramus descendens hypoglossi arises from the cephalic extension of the ventral gray of the cervical cord, while the nerve to the avian syrinx arises from the dorsal portion of the hypoglossal nucleus, suggesting in t is latter case a more intimate relation with the gray associated with the vagusgbssopharyngeal sensory area. Addens (’33) offered another explanation — that the laryngeal muscles are supplied from the ventral trigeminal nucleus.

Dora. nucl. X Intcrmcd. nucl.

Cervical nucl.

Fio. 270. A cross section through the caudal end of the medulla oblongata of the stork (schematized).

THE EFFECTORY SYSTEM

567

and hypoglossal neurons are added to this ventral nucleus providing for a central interrelation of impulses from the larynx and the syrinx (see figs. 270, 271 and 273 A and B).

_ BdaidmtT _ mdiailmllX

_ cod. DtSoj

lej

codisircdin calls] rod fH

_ esdudctdlY odVl

calaBlrcolXi3dXI,ct!b Tle!>93»l(cl)

cod aoj rod tpk, ecc. cd Xn

] _ end E5ajcr*WesIpla!

icfnior oSntT cod f.Vn,IXc: (aVIlIX 1 obrt idrglorin

Flo, 271. Diagrammatic representation of the motor roots and nuclei in certain birds. In Sphenicus it has not been possible to determine the exact limits of the spinal accessory nucleus (Addens). The nuclear mass usually recognized as the dorsal nucleus of the glossopharjmgeal nerve has been diagrammed as a part of the vagus nucleus (in keeping -nith Addens’ interpretation), and there is shown only a single glossopharyngeal nucleus, the ventral nucleus of many observers. In keeping with Addens’ interpretation of the spinal accessory nerve as a somatic efferent nucleus and nerve, it is shown in solid black, the convention used for other somatic efferent centers.

A. Sphenicus demersus. Accessory abducens nucleus is shown lateral to the principal nucleus of the abducens; added in accordance to Addens (’33).

B. Gallus domestious. Addens.

C. Cacatus roseicapiUa. Addens.

Many observers regard the spinal accessory nerve here and elsewhere as composed of special visceral efferent components. Beccari (’13, ’22) identified a distinct spinal accessory nucleus in the sparrow, in line with the somatic efferent column of the spinal cord as represented in the dorsal hypoglossal

568 NERVOUS SYSTEMS OF ^VERTEBRATES AND OF MAN

nucleus. This has been confirmed by Addens (details as published, ’33) in birds generally (see figures of Sphenicus, Gallus, and Cacatua). The bulbar accessory centers are in relation to the visceral columns of the vagus (see fig. 272). This has been seen by BeccaH (’13) and others. In general this caudal visceral efferent column of birds resembles that of reptiles in that a slight lateral shifting of the column (in comparison with the homologous column in plagiostomes) occurs, and is evidenced by the peculiar, caudally directed loop formed by the efferent vagus fibers, which, on leaving their nucleus, first penetrate the lateral part of the medial longitudinal fasciculus and then run in a lateral direction. In its frontocaudal relations, the column in birds varies from that in reptiles.

Dora. nucl. X - *

Intcrmed. nucl.

Cervical nucl. .

Ventro-Iat. nucl. X

Fio. 2 ( 2 . Tlie vagus nucleus and the upper cervical gray in the penguin.

frontanimu^ boundary of the visceral column in birds varies in e.xtent, the

If en atn " ; n beyond the level

-ome dis ann f ^“Pbaryngeal root, whereas in reptiles it terminates at

Te to tho ‘I n 7 ""T condition in the latter animals is

uLus of ° f.°f°Pbaryngeal nucleus wholly or in part with the

pemlent of 1^ '‘"l ' "I b^'^ever, the glossopharyngeal nucleus is indeE St bv in / e^°«^«Pbaryngeal nucleus was identified

1 CW -S; 1 ’28) bs presence confirmed

contiiot . i '^bo dorsal efferent column is said to

’63-1 • AriV-nv Vn Tharj ngeal and vagus components in most birds {Brandis,

'28- and others). AddJ, vagus imcS roL rT'" Rlossopharyngeal nucleus as actually a

vtt g oi^ t'T' ventralward to unite centrally

r ri’ " bis phenomenon of

nmttcr the reader is

THE EFFECTORY SYSTEM

569

In contrast to the relatively primitive relations of the glossopharyngeal complex, the more caudal portions of the posterior visceral column have more complicated relations. Three distinct nuclear groups within the dorsal efferent nucleus of the vagus were described by Sanders (’29) for the sparrow, the duck, the dove, and the chicken, the subdivision being based on neuronic characteristics.

Os

> t>

' ■'J C5

<>.<=& t!>

Fio. 273. A. Cross section through the dorsal motor or cfTcrent nucleus of the vagus and the nucleus intermedins in parrakect. Sanders (’29).

B. Cros.s section through a level similar to that of /I in sparrow. Sanders {'29). Attention is called to the relative difTcrcnccs in size of the motor nuclei in the two forms and the greater development of the inferior olivary nucleus in the parrakect. dors.m.n.X, dorsal motor (or efferent) nucleus of vagus nerve; dors.lal.v.s.n.X, dorsolateral visceral sensory nucleus of vagus nerve; n.ain., nucleus cuneatus; n.dcsc.r.V, nucleus of descending root of trigeminal nerve; n.grac., nucleus gracilis; n.inlcrmed.X-XII, nucleus intermedius of vagus and hypoglossal nerves; n.ol.inf., nucleus olivarius inferior; N.X, vagus nerve; vcnl.m.n.X, ventral motor nucleus of vagus nerve.

C, D, E, F. High-power camera-lucida drawing of the cell groups in the dorsal motor or efferent nucleus of the vagal and glossopharyngeal nerves. Sparrow. Toluidin-blue preparation.

C, dorsolateral nuclear group of dorsal efferent nucleus of the vagus; D, ventromedial nuclear group of the dorsal efferent nucleus of the vagus; E, neurons of ventrolateral nuclear group of the dorsal efferent nucleus of the vagus; F, neurons of nuclear group forming the dorsal efferent or motor nucleus of the glossopharyngeal nerve. Sanders (’29).

The most cephalic, and at the same time the most dorsal, of thesb groups has large, multipolar cells. Smaller cells, comparable in size, type, and general character to those of the dorsal efferent nucleus of the glossopharyngeal nerve form a ventromedial cell group, while a third portion, giving rise to the greatest number of root fibers, the ventrolateral part, has even smaller cells. Only one cell type was found by Sanders in the dorsal efferent nucleus of the vagus in the parrakeet, and in this latter animal the nuclear group itself was smaller than the nucleus intermedius. (Consult fig. 273.)

570 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The visceral efferent column as represented by neurons of the vagus contains cells of origin for preganglionic fibers supplying viscei-a such as the stomach and the heart, and also cells of origin for fibers to branchiomeric muscles, as represented in higher forms in striated muscles of the phaiynx, larynx, and oesophagus. In mammals the preganglionic neurons (as the dorsal efferent nucleus) retam their primitive position near the floor of the ventilcle, while those cells giving rise to fibers to branchiomeric muscle migrate ventrolaterahvard to form the nucleus ambiguus. Two .such ventral migrations appear to have occurred m birds, one forming all or part of the nucleus intermedius (p. 566), he other forming a more ventral vagus nucleus. By analogy, the first group of ventrally migrating cells in birds might be regarded as containing the cell bodies o bers suppling the larynx. Such an inteipretation is at least in accord with the small development and diffuse structure of the nuclear mass ; for it is gener wit6 in birds the laryngeal musculature is only poorly developed.

With this lack of development of the muscles of the larynx in birds is associated syrinx an organ particularly concerned with the production thp vpL although functioning conjointly ivith the larynx. Thus

effect unon ^r^ ' innervation of the larynx, and the hypoglossal, through its

nroduction nre both concerned in birds in sound

certain cp 11<5 nf i® appears o afford an explanation for the close relationship of

fmther wo!k J intermedius. However,

T^pll 7 . ? relationship beyond doubt,

e second and larger ventral migration of vagus cells is found behind the calamus scriptorius (figs. 271A B I louna oenina cne

Caial (’C]Qa\ j sroup was observed by Eowdn 7/

eSup Hid,?„r ‘,p emerging root fibers of (he vagus

SSorv ™ f ‘ f ™ ““"bed its fibers to the

rfiafrrt^vr

guish with (ertaintv H. t ftetehed a dorsal branch with it. To distin it would be necessaLfn from those of the vagus,

their termination in the mradK** a dTtalt’'*'* ’’'"'’=“1 distribution to

knowledge) as yet not taS tit for bWs “n

which althoiifrli itipvr nvv „ + oiras. ihere are, however, certain facts

that the cells here considered strongly support the statement

accessory fiberiThS al^^^^^ ‘’’“.“Ses rather than to the bulbar

ventral oeUs rftws occZVS "’e nucleus. Thus these

biguus of marim^VSer'^nTuo ^

animals gives rise to the accessory fibere but ‘be latter

ongin for the vagus. In mammals the ^ ^ ^ cells of

caudal limit of the inferior olivarv i bes entirely behind the

does not reach CsteSrulanrh^ W'-d®

Kappers, ’20), nor does it extend sTf^r inferior olive (Ariens

of bulbar accessory nerves Thp p 'ri ‘’^’^dalward as the levels of emergence y nerves. The evidence would appear to indicate strongly.

THE EFFECTORY SYSTEM

571

although it does not piove conclusively, that in this ventral nucleus is represented a portion of the nucleus ambiguus which gives rise to efferent vagus fibers {Ariens Kappers, ’20; Sanders, ’29, and others), presumably those to the larynx

Fio 27-1 The facia! and abdueons nuclei in (he stork

Tig 27.3 The motor and 'cn'or> nuclei of tlir trigeminal none in tlic stork (compare 111111 fipiro IR'i)

(and perhaps to other striated branchiomeric muscles ns veil). The limited frontal extent of the nucleus has been associated 331 th the fact that thehomologue^ of those dos'sophnrjTigenl neurons, 3\hich in mammals lie33ithin nucleus ambiguus, are regarded by certain observers (AriCns Kappers) as still occupying in some birds a dorsal position near the ventricular floor. That in certain birds.

nucl.a,nf

r cochl.asc., nuci, vest.v^ ggl Scarp

^nucl. magnocell. ant. cochleo-cereb.

r.^r/z:

r. Y des nucl.d &ca

ol.sup' j nucf ® mot.

S com. cochl. dors.

nucl-THprlnc. fp^fajc long. med.

Wil

fasc.long. verrtr.

u

7 days. X70. accessory abducens nuclei in the chick embryo, age

method^ nucleus. Cajal

572

THE EFFECTORY SYSTEM

573

at least, these neurons may have migrated from their primitive dorsal position to a more ventral situation, suggesting an anterior portion of the nucleus ambiguus, is indicated by the results discussed on page 570. Crossed vagus motor root fibers have been described by Ramon y Cajal (’09) and Beccari (’22), but questioned by Addens (’34).

The avian cephalic visceral column (figs. 271, 274 to 277) exhibits certain marked peculiarities which offer excellent illustrations of the part played by neurobiotaxis in the shifting of the nuclear groups. Thus in running birds, such as the cassowary (one of the lowestbirds), conditions corresponding to those described for Caiman (p. 559) are found.

However, in general, the avian facial nucleus 3ZL differs from that of all other animals (compare m figs. 259 and 271) in lying in front of the plane of emergence of its root fibers instead of behind that plane.

Frequently the nucleus consists of two portions, a dorsal and a ventral group (figs. 271A, 274).

In the chick and the cockatoo {Addens, ’33) it falls into three nuclear groups — a dorsal, an intermediate, and a ventral group — as in Caiman. The dorsal nuclear group fies in close relation to the motor nucleus of the trigeminal nerve. This intimate relation, as seen in Cacatua by Black (’22), and more recently studied hy Addens (see 1933 paper), has led to various interpretations of these neurons by different observers. Thus the frontal pole of the dorsal facial nucleus, as shown and interpreted in the figure by Addens, was regarded by Black as a trigeminal nucleus. The fibers traced from this cephalic portion by both Black and Addens join the trigeminal root, but such fibers, while regarded as trigeminal by the former observer, were regarded as facial by Addens, who considered this another example of central anastomosis. Addens regarded this condition as an illustration of fasciculation, a matter discussed in detail in his recent (’33) contribution. The cephalic position of the facial nucleus in these forms, in which gustatory functions are so poorly developed, is to be expected, if the caudal shifting of the homologous nucleus in other forms (particularly sharks and teleosts) is to be interpreted as due to the neurobiotactic influence of the gustatory centers {Ariens Kappers, ’20). By careful observations Bath (’06, see bibliography, p. 416) has been able to demonstrate the presence of a few taste buds in birds. In birds, with the exception of parrots, they compare with

THE

T dtic.

Oliva ;

Fia 277. Abduccns and facial nuclei in the cassowary.

574 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the number in mammals (rabbits) as 1 : 100 (see chapter on branchial nerves). With this great decrease there is a corresponding decrease in the development of the central gustatory centers and consequently a much less marked neurobiotactic influence is exerted by these centers on efferent nuclei of the brain stem. In birds, then, the influence of the sensory trigeminal nucleus upon the facial nucleus is to draw this latter nuclear group forward to near the former nuclear group. The more dorsal part of the facial nucleus — the part which is closely related to the motor trigeminal nucleus — innervates the depressor mandibuli, the intermediate group innervates the mylohyoideus posterior, and the most ventral group innervates the sphincter colli (Kosaka and Hiraiwa, ’08).

The motor trigeminal nucleus (figs. 275, 276B, charts in 271) falls also into several portions. A smaller medial trigeminal nucleus- {Arims Kappers) in Chrysometrius is of flattened form, with its root fibers arching first dorsally and then laterally. The lateral nuclear group is more nearly round in outline and extends farther cephalad than does the more medial trigeminal nucleus. In the cassowary {Ariens Kappers, ’12; not figured) quite a different

f .-.“I

Fig 278. The trochlear nucleus of the stork (Ciconia alba), van Giesen preparation.

Acc^a nuci III (Edingc r-W ea t phal)

Dor" lat nucI III —

Dors mcd nuci III —

Vent mcd. nuci III

i,

i

Fig. 279. The oculomotor nucleus of the chicken. Nissl preparation. Vermeulen.

condition is found, for in this bird the relations are more primitive, suggestive of^ t le^ relations in turtles and in Caiman. The cassowary has two efferent trigeminal nuclei, of which the smaller shows a migration to a ventrolateral position. The larger nuclear mass in this bird remains in a more dorsal posi

THE EFFECTORY SYSTEM

575

tion, close to the floor of the ventricle. Consequently, in the arrangement of the cells of origin of motor fibers, a more primitive relation of fibers persists in the cassowary than in the other birds studied. The facial and the trigeminal efferent nuclei have been considered together because of their close topographic and functional relationships. Consideration will now be given to the abducens nucleus and root.

The abducens root in most birds, as in higher reptiles and most mammals, emerges partly in front of the level of the facial roots ; consequently the term “ sixth ” nerve is applicable here. In the cassowary (Ariens Kappers, ’ 12) , which in other particulars has more primitive relations, the majority of the root fibers of the abducens leave the brain behind the emergence of the facial root.

This is in accord with the

Fjg. 280.

Fffl InMmdB

I wAarin

xtJocUD

The topography, in the sagittal plane, of the trochlear and oculomotor nuclei of the penguin.

uccount given by Jifesdoff of the presence of n part of the abducens root in a quite caudal position in the embr^’onic chick. There is reason to believe that the transition stages from a caudal to a more cephalic position are repeated during the embryological development of the nerve in the chick {Biondi, 13 , Bok, ’15). Suggestions of such primitive relations occur in other birds. Thus Sanders (’29) found that in the sparrow, where the number of abducens rootlets is large, some of them emerge caudal to the most caudal rootlets of the facial

nerve.

The cells of origin of the avian abducens nerve (fig. 277) form a nuclear mass which lies directly lateral to the medial longitudinal fasciculus, partly at the level of entrance of the facial root and partly cephalic to it. This abducens nucleus is quite large and is somewhat flattened on the lateral side. Sagittal planes show no outstanding peculiarities in the topography of the abducens nucleus in birds, since in its relations this nucleus duplicates essentially the relations of the homologous nucleus in Varanus, \vith reference to the facial nerve. The nucleus in birds is particularly large. A comparison with conditions later to be described for mammals shows that the abducens nucleus differs in avian and in mammalian forms in failing to show in the former any tendency toward the ventrolateral shifting so characteristic of it in at least certain of the mammalian forms (as in the rabbit and man). This difference is associated probably with the fact that the avian nucleus receives about as many contralateral as homolateral fibers from the vestibular nuclei and the cerebellum, while the mammalian nucleus receives more homolateral than contralateral tracts and tends to shift dorsolateralward in the direction from which it obtains its

major stimuU (see figs. 271, 276, 277). /-» i.

An accessory abducens nucleus has been described for birds by van Gehuchien (’93), and later by Terni (’22, ’22a), Preziuso (’24), Addens (cited by Craigie, ’28, ’33), and Sanders (’29). As in reptiles, the cells of this accessory nucleus lie far lateralward, in the neighborhood of the superior olivary nucleus and close

576 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

to the nucleus of the descending root of the trigeminal (fig. 276A) . In the sparrow, according to Sa7iders (’29), they lie lateral to the superior olive in planes through the latter nucleus, but caudal to such planes the accessory nucleus occupies the same general position, until its tip becomes continuous with the ventral part of the nucleus of the descending root of the trigeminal. It is certain that neuraxes of these cells extend in a dorsal direction toward the dorsally situated abducens nucleus and they join root fibers arising from this nucleus. According to Terni ( 22, 22a) and Preziuso (’24), the accessory nucleus of the abducens innervates the muscles of the nictitating membrane and receives its dominant impulses from the nucleus of the descending root of the trigeminal nerve. The nucleus described by Nakamura (’30) as a facial nucleus probably is to bo regarded as such, but it is extremely doubtful that it is homologous with the accessory VI nucleus of van Gehuchten and others (see also Addens).

The anterior eye muscle nuclei in birds show highly differentiated relationships. Thus the oculomotor cell mass can be subdivided into a number of mm ear groups (fig. 279). The trochlear and oculomotor nuclei overlap each other, so that the medial parts of the oculomotor extend between the trochlear nuclei of the two sides. The direct transition from the oculomotor to • ^11 nuclei, which is a secondarily acquired characteristic phylogenet ically, only appears in later stages ontogenetically, as Mesdag’s (’09) work indi ^ embryo a distance comparable in length to

that of the oculomotor nucleus intervenes between this nucleus and that of the roc ear nerve. n a chick embryo of nine days’ incubation the two nuclei almost touch each other, although by the thirteenth day of incubation they have qui e uni e . ater the continuity of one nucleus with the other occurs, and in such a manner that the trochlear becomes a caudal continuation of the dorsolateral oculomotor nucleus (fig. 280).

decussation, the trochlear nerve emerges near the caudal

InTifninrli nucleus itscIf lies dorsolateral to the medial

intn ascicu us, as in most higher vertebrates. It shows no subdivision

n are only slightly

w relntionl'rfi many interest obsertfrc r r recognized by many

’S ^ Kapvers,A2- Brornvcr,

s tuated bttw ventromedial group, which is

medid t^n, n A' T longitudinal fasciculus and the raphii, (2) a dorsomS oMoTh 1 prolongation of the former group but is

dorsolateral to dorsolateral group, which lies group reeularlv A ^“S'^^dinal fasciculus, and finally (4) a fourth nuclear Westphal bv 1?^® lioen termed the nucleus of Edinger (’18) and othpr <5 +v, ’ ^ Cajal (’09), Ariens Kappers (’12), Brouiocr

mentioned "«^“us III by Meskig (’09). This last

and is differentiahl 'T ® 'Slil'ly dorsolateral to the nucleus dorsolateralis

smaliL A oculomotor nerve through the

eep s ainmg of its cells. This nucleus was identified in

THE EFFECTORY SYSTEM

577

Varanus {AriC-ns Kappcrs, ’12), where it is directly comparable to that of birds in position and cellular characteristics.

The frontocaudal extent of the different nuclei of the oculomotor corresponds in general to the extent of homologous nuclei in Caiman. Thus the dorsolateral

- *’>

' -i

c

Fio. 2S1. The development of the oculomotor nuclei in the chick. Biondi.

A. Embrj-o, 4 days old. The migration of the cells has not begun. All the root fibers pa-ss on the side of their origin.

B. Embryo, 0 days old. The migration has begun.

C. Embryo, 8 days old. The migration is completed. A portion of the root fibers emerge on the side opposite their origin.

nucleus reaches farther cephalad than do the dorsomedial and ventromedial nuclei (compare figs. 268 and 280). The accessory nucleus of the oculomotor or the nucleus of Edinger-Westphal does not reach the caudal end of the oculomotor nuclear complex in birds (or in Varanus). It usually is found in planes from the frontal end of the dorsolateral nucleus (and so of the nuclear mass) to the caudal third of the oculomotor nucleus. In certain cases it gradually unites

578 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

with the dorsolateral nucleus. Evidence at present available confirms the probable homology of this nucleus with the Edinger-Westphal nucleus of mammals. As confirmatory, are to be regarded its general relationships ; thus both Ramon y Cajal (’09) and Mesdag (’09) traced fibers from it into the oculomotor roots. Moreover, complete atrophy of an eye in a sparrow was found by Brouwer (’18) to be associated with a degeneration of the nucleus, indicating that, as is the case with the Edinger-Westphal nucleus in mammals, this nucleus supplies preganglionic fibers for innervation of the intrinsic eye muscles through the cihary ganglion. In the oculomotor root of the hen Uarpenier (’ll) identified coarse and fine fibers. The fine fibers are situated largely at the periphery along the lateral side of the coarse fibers, and were followed by Carpenter (’ll) into the ciliary ganglion, indicating that they constitute the preganglionic fibers to which reference has just been made.^

Of the coarse root fibers of the oculomotor, a not inconsiderable number cross, particularly those from the ventromedial nucleus. The manner in which this crossing occurs is very peculiar. It was described originally by Biondi (’10), whose statements have been confirmed by Boh (’15). It appears that in the four-day chick embryo, all the root fibers are as yet uncrossed. On the sixth day, however, a migration of cells (particularly those of the ventromedial nucleus) occurs through the raphd, after the greater number of the neuraxes are formed. As a consequence, fibers arising from such migrating cells have a crossed origin. This migration (fig. 281) is completed during the eighth day of incubation. Probably it is caused by the nevuobiotactic influence of impulses acting in a contralateral way upon the oculomotor nuclei.

The Effectort System of Mammals

The arrangements of the motor centers in the medulla oblongata and midbrain of various mammals show certain marked similarities which may be considered as common to the class as a whole. On the contrary, certain differences are sufficiently pronounced to serve as characteristics of the various orders of this class. That the likenesses and differences may receive due emphasis, the conditions as found in various lower forms will be discussed first, and then, with this as a basis for comparison, variations from this pattern as found in higher mammals will be considered.

In birds the hypoglossal nerve had begun to show evident marks of differentiation, since its cells were separated from the gray of the cervical cord. In most mammals an independent hypoglossal nucleus is found, and the nucleus lies entirely within the brain stem and is independent of the cervical gray. The differences between the hjqjoglossal nucleus of mammals and that of birds are largely that in the former animals the nucleus has a slightly more dorsal position,

■ The nucleus of Darkschewitsch is considered under the description of the posterior commissure and the medial longitudinal fasciculus, since it is regarded here, as previously described by Ramon y Cajal (’09a), as consisting of cells giving origin to fascicles of these fiber complexes. Topographically, in the cassowary and in Sphenicus (Ariens Kapyers, ’12) at least, it lies in close relation to the nucleus accessorius III (but it extends farther frontalward and it is broader; see Chapter Vlll).

THE EFFECTORY SYSTEM

579

that it extends somewhat farther cephalad, and that it is considerably larger. The migration forward of the nucleus of the hypoglossal is associated with a shifting frontalward of its roots so that they emerge in planes medial to the

racLirfroeini ||||[[|(||| md aid r»l lY fe.'-A-l nudajiJroolV

nacL lid root Vn nod ¥1

cod. znd root X ud XI, ceSIs rlediostt&Cel) cud asd root spin., occ aid XO ofonor o5yai7 cud

cod aod root IX nod Defters

1 cod Edinger*Weslplul

fifportoro&nrxEod 1 ca’inas scr^lonns

Fig. 282. A diagramniatic representation of the topographic relations of the motor roots and nuclei of eertain lower mammals In the mole the eye muscle nerves (with the exception of the oculomotor) are lacking. The two vertical lines (II) mdicate the position of the genu of the facial nerve. (The ventral nucleus of the vagus — nucleus ambiguus — extends somewhat farther caudalward than is indicated in the diagram; compare fig 291).

A. Echidna. Anens Rappers. B. Didelphys. Ariens Rappers. C. Talpa. Aneiis Rappers.

inferior olivary nucleus, beginning usually immediately behind the plane of emergence of the glossopharyngeal nerve.® This is an excellent illustration of the way in which roots may shift with changes in position of their cells of origin. While such changes of the roots may occur, it does not follow, by any means,

In a specimen of Macropus, one fascicle of the hypoglossal root has been found even in front

of the level of emergence of the glossopharyngeal nerve (Anens Kappers, ’20).

5S0 NER^'OUS SYSTEMS OF VERTEBRATES AND OF MAN

that they always do occur. Thus the trochlear and facial roots afford clear examples of the fact that changes in root fibers need not of necessity keep pace with changes in nuclear position. Space must be available for the shifting of the roots. Thus the presence of the quadrigeminal bodies prevents the migration fon^’ard of the trochlear roots, and the relation of the facial with the vestibular prevents a change in position of the facial roots.

In order to understand the factors M’hich produce the changes in position of the hypoglossal nucleus in man, it is necessary to consider the reflex paths which most closelj' involve this nuclear mass. In lower vertebrates the muscles innervated by the spino-occipital nerves do not take part in the formation of a tongue, but from amphibians on this organ develops gradually. The sensory impulses from this organ are of two major tj^^es, those of general sensation (carried by trigeminal and glossopharyngeal fibers) and those of gustatory sensibility (mediated by the facial nerve, the glossopharyngeal nerve, and, for the epiglottis region, a few fascicles of the vagus nerve). Thus; in a functional sense, the.se nerves have become the sensorj’^ roots

I in. 2S.’!. ,y//, (lie hypoKlo'wal nucleus; r, the for ward corituiuiitiou of tlic ventral horn pray; a, the nucleus ainliipuiis. The .Portion i.s from the medulla oblongata of PlKtraciia. \'rnnrulcn.

of the hypoglossal, since they innervate the epithelial and connective tissue areas ovcrlj'

ing the muscles innervated by the liypoglos.'ial nerve, and since from their nuclei of termination within the medulla oblongata (and pons) the major impulses affecting the motor nucleus of the hypoglo.'i.sal emanate. According to Wallcnhcrg (’04), the trigeminal nerve .1 .'-o send.s a part of its fibers, those from the inside of the mouth, into the nuclear gra> of the region of the fasciculus soHtarius. From this gray the "via central ^ ^ y ^ Ramon y Cajal (’09), con.si.sting of short, cro.sscd and uncros.scd fibers runs medialward, directly above the hypoglossal nucleus, to which it rontn niles fillers. 1 his fiber .sj'stcm tends to draw the hypoglo.ssal nucleus orwari and dorsahyard. So, undoubtedlj', does the nucleus intcrcalatus or the •iiK ( u- of Staderini, which lies between the motor nucleus of the hypoglo.ssal am 1 1 ( dor.'-al efferent nucleus of the vagus, if the interpretation of it as a gustatory renter be accepted fsce the chapter on branchial nerves). Tliat gustatory impu '■f*' pl.aj an important role in determining the position of the hypoglo.ssal nucleus Imcii .yhown by the work of Vrrmaden (’Ifi; fig. 2a'l), which shows ttiat tht« nucleus m Cetacea, where taste is atrophied (Rawilz, ’03, Delphinus

THE EFFECTORY SYSTEM

581

delphis), has shifted less far frontalward and at least, as far as its spinal half is concerned, less far dorsalward, and has retained its connection with the gray of the cervical cord (compare fig. 65A). In the camel and in the giraffe, relations suggestive of those of lower forms are to be found, but in no mammal are they so indicative of primitive conditions as in Phocaena (Vermeulen, T5, T5a, ’16). A brief survey of the detailed structure of the hypoglossal nucleus in various mammals follows.

In Echidna, the hypoglossal nucleus is small and occupies a lateral position extending outward in the direction of the dorsal efferent nucleus of the vagus (see fig. 282A). No division into distinct groups was possible in the material available for study. In the opossum, Didelphis virginiana, Eon's and Hoerr

Fro. 2S4. The hypoglossal nucleus of the opossum (Didelphys marsupialis).

(’32) found “large, pyramidal-multipolar’’ neurons constituting the hypoglossal nucleus. Such neurons showed the chromaphilic characters of somatic efferent neurons. The nucleus was situated close to the midline and at all levels separated from the dorsal efferent nucleus of the vagus by the nucleus intercalatus. The measurements of the nucleus in the opossum by Voris and Hoerr showed it to have a total length of about 6 mm. According to these observers, a not sharply definable Roller’s nucleus is present in the opossum. (For opossum see figs. 282B, 284).

In carnivores, such as Canis familiaris, at least three cell groups can be distinguished for the hypoglossal nerve. Of these, the ventromedial group is the longest, being found throughout the length of the nucleus. A ventrolateral group, present in the cephalic half of the nucleus, follows somewhat the direction of the root fibers lying laterally, and consists of a fairly large number of cells. This group increases in size from the middle of the nucleus forward. The third cell group, appearing about a fifth of the way from the caudal end of the nucleus, at first is dorsolateral in position, but farther forward occupies a position lateral to the ventromedial nucleus. It is in direct contact, particularly in its caudal part, with the dorsal efferent nucleus of the vagus. It appears to be that portion of the hypoglossal nucleus from which Kosaka and Yagita (’05) traced the ramus descendens hypoglossi in the rabbit and the dog, a localization in harmony with the earlier experiments of Parhon and Goldstein (’99). Somewhat above the cells of origin of the ramus descendens hypoglossi (which lie caudomedialward,

582 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

■

but in their neighborhood), are the neurons the neuraxes of which supply the hyoglossal and styloglossal muscles (Sluurman, ’16a). Fibers supplying the geniohyoid and genioglossal muscles have their origin from the cephalic portion

of the ventrolateral nucleus. The vertical and transverse muscles of the tongue are innervated by fibers from the ventromedial cell group {Sluurman, in the mouse).

The nuclei in Myrmecophaga jubata (figs.

285 and 286) and Tamandua tetradactyla

(analogous with Myrmecophaga in general)

' 2 - f . . have ventromedial, ventrolateral, and dorso lateral portions, comparable to those described above for Canis. As before, the ventromedial part extends through most of the length of the nucleus, while the dorsolateral portion is not found in the caudal part of the region (in Myrmecophaga not in the caudal third, see fig. 285). The dorsolateral nucleus lies very close to the dorsal efferent nucleus of the vagus. The ventrolateral nucleus does not differ in any significant way from the homologous nucleus in the dog.

The hypoglossal nuclear group of the ant-eater is remarkable for its very great size and for its differentiation ; a condition associated with the great length of the tongue in this animal. Of particular interest is the fact that here the ventromedial nuclei of the two sides lie very close together and form, in front

Fig. 285. The caudal pole of the hypoglossal nucleus of an ant-eater (Myrmecophaga).

Fig. 286. The middle portion of the hypoglossal nucleus of an ant-eater (Myrmecophaga).

of the caudal third, a raph6 group which becomes very large. Frontally this raph6 group terminates before the caudal pole of the remainder of the ventromedial nucleus is reached. This correlates with the high degree of development of the vertical and circular muscles of the tongue in these animals. Whether the absence of a median septum here and the strong integration in action of the

THE EFFECTORY SYSTEM

583

muscles of the two sides has led to the formation of a raph6 nucleus is not clear, although probable. The relations in the hedgehog have been studied very carefully by Berkelbach van der Sprenkel (’24). The following localization ■ndthin the h3T5oglossal nucleus {Slilling's chief nucleus, ’43) is indicated by his studies. A center for the geniohyoid muscle lies in the cephalic fifth, distributing through the first hj’^poglossal rootlet, and centers for the hyoglossal muscle in the dorsal and for the styloglossal and genioglossal muscles (the latter the more caudal) in the ventral part of the succeeding fifth. These last muscles are supphed through the medial and lateral rootlets, respectively, of the second hjTDoglossal root. In the middle fifth, the dorsal part has the centers for the longitudinal and the ventral part for the transverse muscles of the tongue. The motor fibers for these muscles run in the lateral and medial rootlets respectively of the third hypoglossal root.

In the caudal two-fifths are centers (extending across nearly the whole cross section of the nucleus, Ijdng one behind the other) for the ansa muscles, the thyrohyoid, the sternoh 3 '’oid, the sternothyroid, and the omohyoid.

The fibers supplying this last muscle group pass out either with Cl (those for the thyrohyoid muscle) or with C2 (those for the remaining muscles). Their cells of origin are much farther forward than the plane of emergence of their roots, a clear indication of the functioning of neurobiotaxis. This caudal two-fifths of the nucleus is in reality a center supplying spinal components (the ansa hj^joglossi), and the true hypoglossal nuclei, according to Berkelbach van der Sprenkel (’24), are represented in the cephalic three-fifths of the principal hypoglossal nucleus of Stilling. Attention should be called here to the suggestion of Berkelbach van der Sprenkel (’24) that not all of the neurons of the hypoglossal nucleus are to be regarded as somatic efferent, since the geniohyoid may be considered as a branchiomeric derivative, and, in that case, its roots as special visceral in tj^De.

The hypoglossal nucleus of primates and of man (fig. 287) shows clearly a subdivision into smaller groups {Parhon and Papinian, ’04 ; Mingazzini, ’09 ; Hudovernig, ’08 ; Goldstein and Minea, ’09). An essential difference depends on the fact that the ramus descendens hypoglossi no longer arises fiom the hypoglossal nucleus but from the cervical cord. A group of smaller cells in the region of the dorsal accessory olivary nucleus gives rise to hypoglossal root fibers, according to Mingazzini. This is a true hypoglossal nucleus (“Der wahre Duvalsche Kem ”) according to this observer, who believed that the cells scattered through the reticular gray constituting the nuclei anterolaterales or accessorii of

I. 287. The human hypoglossal nucleus The middle portion of the nucleus Huber and Crosby

584 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Duval are not contributors to the hypoglossal root fibers. Mingazzini studied carnivore and primate material. A nucleus of Duval, contributing to hypoglossal root fibers, has been described in lower and higher mammals, including man. In addition to efferent neurons supplying muscle of the voluntary striated type, according to R. B. Wilson ® (J. Comp. Neurol., vol. 58, p. 419), in the whale the hypoglossal nucleus contains general visceral efferent neurons which are

situated on its lateral side in the cephalic half of the nucleus. In Erinaceus, Berkelbach van der SprenJcel f’24) described a small-celled neuron group on the medial side of the hypoglossal, in a position corresponding to Roller’s nucleus (Roller, ’81) and interpreted these cells as constituting a general visceral efferent center, serving as a vasomotor, more particularly as a vasodilator center. A preganglionic group, other than Roller’s nucleus, was described in the hypoglossal nucleus of man by Schwentker (’27, Anat. Rec., vol. 35, p. 345).

There still exists reasonable doubt as to the possible presence of proprioceptive fibers in this nerve. By experimental procedure Langworthy (’24) appears to have demonstrated that an ataxia of the tongue muscle and a degeneration of the neuromuscular spindles within the tongue result from a section of the hypoglossal nerve. Neither severing the lingual branches of the trigeminal nor severing the glossopharyngeal nerve produces this result. The experiments of Langworthy were planned to simulate those carried out by Tozer and Sherrington (’10) to determine the sensory innervation of the eye muscles. In Erinaceus,

Fig. 288. Relative position and according to Berkelbach van der Sprenkel (’24), the extent of the accessory roots in : A, a. • i* , t.

shark; B, Caiman; C, a mammal, propnoceptive sensory fibers Supplying the hyoglossal, styloglossal, and thyroglossal muscles (and probably also those muscles innervated by the ansa hypoglossi) have their cells of origin in the second cervical dorsal root ganglion, their dendrites passing through the corresponding cervical nerve to the ramus descendens hypoglossi. Hinsey and Corbin (’34, J. Comp. Neurol., vol. 60, p. 37) found no proprioceptive fibers from the first four cervical ganglia in the cat.

There have been considerable differences of opinion as to whether or not hypoglossal fibers may arise from the contralateral as well as from the homolateral hypoglossal nucleus, but the evidence at present indicates the homolateral origin of these fibers. Mingazzini (’28) laid stress, however, on the close prox ® That the nucleus referred to in the whale is autonomic in character was concluded by R. B. Wilson by reason of its distinct chromaffin character ; also, its cells were smaller than ordinary root cells of the hypoglossal.

C

THE EFFECTORY SYSTEM

587

canal separately {Lesbre, ’03, and Vermeulen, ’15a), and finally runs to join ventral root fibers to corresponding segments at the peripherj'. The bulbar portion of the accessory nerve takes origin from two centers, the dorsal efferent nucleus and the nucleus ambiguus. The dorsal efferent nucleus gives origin to preganglionic fibers which distribute with the vagus. This nucleus, which occupies a position near the central canal, dorsolateral to the midline, is continuous forward with the dorsal efferent nucleus of the vagus. It has been recognized in various mammals, the account of Precechtel (’25) for elephant being one of the more recent. The caudal end of the nucleus ambiguus gives rise to fibers of the bulbar accessory nerve which, through the internal ramus, supply branchiomeric muscle. A fusion of the .spinal acce.'sorj'^ nucleus with the nucleus ambiguus, which has been found in a few cases (fig. 290 ; Vermeulen, ’18), is regarded as a secondary condition, due to a lengthening of the latter nucleus.

Consideration must now be given to the classification of the functional components carried by the accessory nerve. There is fairly general agreement that the bulbar portion consists of general ^dsceral efferent fibers from the dorsal efferent nucleus and special visceral efferent fibers from the nucleus ambiguus. With regard to the components represented in the spinal accessory nerve there is not general agreement, as yet. The interpretation of Addens, according to which the spinal accessory ner\'^e of mammals is to be regarded as a somatic efferent nerve, is in conformity with his views mth regard to this nerve in submammalian vertebrate forms, to which views reference has been made in preceding pages. The evidence for such an allocation of the nerve has been considered recently by Addens (’33) and cannot be discussed further at this time. In the majority of the modem textbooks of anatomy and of anatomy of the nervous system, as also in the German edition of this text (Herrick, ’31 ; PiersoVs Anatomy, Hnfeer edition, ’30; and others), the spinal accessory nerve is classified as a special visceral efferent ner\'^e. Justification for this classification depends ultimately on the recognition of branchiomeric elements within the developing trapezius and sternocleidomastoid muscles, which requires further embrjmlogic investigation. Certain morphologic features of the developing visceral efferent column of vertebrates, whether considered phylogenetically or ontogcnetically, offer support for such an interpretation of the ner\'e.

The morphologic details and the e\'idence for the development of the accessorj' nerv’e in vertebrates have been considered by Ariens Kappers (’12), with a brief abstract in the German edition of this text. The communication of AriC-ns Kappers (’12) should be consulted for particulars. The views advocated there are given here in brief revdew : The caudal visceral efferent column in plagiostomes occupies a relatively dorsal position and by Ariens Kappers is believed to contain cells of origin of the accessory nerve. In lizards, a ventral shifting of t!ic \dsceral efferent column occurs at about the level of the calamus scriptorius and in the region immediately caudal to it. On the basis of a comparison with conditions in birds and mammals, it appears safe to conclude that this ventrally placed part is comparable with the caudal part of the nucleus ambiguus, containing bulbar accessory and vagus fibers in mammals. The most caudal portion of

588 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the visceral efferent column has not been affected by the ventral migration just described, but shows a special caudal migration of the dorsal visceral column as compared with this same column in selachians and amphibians. In general, the most caudal cells of this spinal extension of the column gradually lose their position near the central canal and take up a somewhat more lateroventral position between the bases of the dorsal and ventral horns. This extension is only limited

c

Fig. 289. Cross sections through the upper cervical cord and the medulla oblongata of a sheep embryo of 33 mm crown-rump length (preparations from Dr. Rolhig).

A. 128 sections in front of the caudal pole of the inferior olive.

B. 2 sections caudal to the inferior olive.

C. 62 sections caudal to the inferior olive.

in amount in these animals ; nevertheless, it is interpreted as indicating that the accessory nucleus is a spinal prolongation of the dorsal visceral efferent column.

Since the differences between the nuclei of the accessory nerve in mammals and in lower forms are too great to give an adequate notion of the ontogenetic development of these centers, a study (Ariens Kavpers) of the development of the region was made on sheep embryos of selected ages, placed at the observer’s disposal for this purpose by Dr. Rolhig of Berlin. This material, serially cut and well stained, consisted of embryos of 12, 17, 23, and 33 mm. crown-rump length (see figs. 289A, B, C).

In an embryo of 12 mm. and, better developed, in an embryo of 17 mm. is an accessory nucleus, differentiable on the basis of the size of its cells from the

THE EFFECTORY SYSTEM

589

surrounding regions. It is even more distinct in 23 and 33 mm. stages, and its relations are established better, since by then the inferior olive is well developed and gives the general topographic characteristics to the region. Following the spinal accessory nucleus forward, it is found to be continuous with the dorsal efferent group containing cells of origin for the preganglionic components of the accessory nerve rather than with the nucleus ambiguus which contains the neurons supplying branchiomeric muscle.

The fusion of the spinal accessory nucleus with the nucleus ambiguus, which sometimes occurs, is to be regarded as a secondary condition in mammals (figs. 290, 291 ; Vermculcn, ’18) due to a caudal lengthening of the nucleus ambiguus. The spinal accessory nucleus may be increased by appositional growth within the cord due to the appearance of certain neuron groups and their fusion with the original

A.-

Pio. 290. Two cross sections tlirough the upper cervical cord of a llama. Vermeitten. In A, which is somewhat the more frontal of the two levels, the accessory and vagus nuclei are connected.

center. This growth appears to be at the expense of more dorsally situated cells rather than of those of the ventral horn and finds confirmation in the fact that the most caudal accessorj' roots are the most dorsal at their level of emergence, while if this caudal differentiation were occurring in the region of the ventral horn, it appears probable that they would emerge farthest ventral. YTiile the derivation here given is opposed to the assumption that the accessory nucleus is derived from ventral root elements, it is not necessarily out of harmony with the suggestion of Beccari (’13) that the spinal portion of the accessory is represented in the coarse von Lenhossek fibers of birds and reptiles, which emerge with the dorsal roots although their cells of origin lie in the lateral part of the ventral horn. A condition found in certain mammals by Lesbre (’03) and Vermeulen (’18) seems to favor the homology suggested by Beccari (’13 ; see p. 559), for they found that in certain ungulates the accessory, in its extramedullary course, does not unite into a common stem but remains as separate rootlets which run close to the dorsal roots. The fact that the accessory nucleus at the appropriate level lies close to the spinal centers for the ventral root neurons supplying the trapezius muscle

590 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

{Vermeulen, T8) does not imply necessarily that the accessory nucleus is somatic efferent nor does the fact that accessory fibers may join ventral root fibers (Bolk, ’98) outside of the cord carry such an implication.

Although both the spinal and the bulbar portions of the accessory nerve are usually recognized as predominantly efferent in character, it has been known forsome time that ganghon cells in scattered groups, both in man and mammals, may be found along its course. These may be found along the intracranial rootlets or may lie extracranially (Fahmy, ’27). Streeter (’08) regarded the accessory as well as the vagal portion of the general vago-accessory complex as possessing sensory as well as motor roots in the hmnan embryo, and believed that certain ganglion cells of sensory type found along the rootlets of the accessory nerve of man by Weigner (’01) were derived from this embryonic root ganglion of the accessory nerve. He also was able to show a very close relation in position between the spinal accessory nerve and the dorsal root ganglion of the first cervical nerve. Weigner (’01) made further interesting observations with regard to the spinal accessory, for he had noted many anastomoses between this nerve and the dorsal roots of the upper spinal nerves and called attention particularly to those cases in which the dorsal root ganglion of the first cervical nerve is incorporated within the accessory root. In his study of the nerves of a three-months-old child, Fahmy (’27) showed that clusters of ganglion cells may be found on the accessory, both intracranially and extracranially. Recently Windle (’31a), using cat material, has re\'iewed the question of sensory components of the accessory nerve. He found that the cell bodies of the sensory neurons are situated, as had been indicated before, partly within the first spinal ganglion and partly along the course of the root fibers. This observer followed the central course of these sensory fibers and believed he could trace them into the caudal end of the tractus Bolitarius and that these fibers are possibly largely proprioceptive {Windle and DeLozier, J. Comp. Neurol., v. 54, p. 97). (One of the most interesting of the observations made by Windle (’31, ’31a) relates to his statement that the entrance of collaterals from the mantle layer of the developing nervous system into this bundle of longitudinally running fibers (which occurs before the appearance of movement in the embryo) completes the ontogenetically oldest spinal reflex path.) Hinsey and Corbin (’34, J. Comp. Neurol., vol. 60, p. 37) found no proprioceptive fibers in the cat arising from cervical ganglia (C 1 to C 4).

Attention has been directed to the fact that the spinal portion of the spinal accessory nerve, after its emergence from the cord, forms a bundle of fibers which passes through the vertebral canal and foramen magnum to join the cranial portion of the nerve, constituting the external ramus. The roots constituting the preganglionic portion of the bulbar accessory pass directly ventrolateralward from their cells of origin in the dorsal efferent nucleus, to emerge along the lateral wall of the medulla oblongata in line with the fascicles of the vagus nerve. Such general visceral efferent fibers are accompanied by fibers which pass dorsomedialward from the nucleus ambiguus to join them near the floor of the ventricle, forming the loop characteristic of the course of many special visceral efferent fibers. These bulbar components form the internal ramus of the accessory nerve.

THE EFFECTORY SYSTEM

593

neck region to about the level of the descending colon (human), the synapse occurring in peripheral sympathetic ganglia. According to Malone (’13), within this nucleus in man lie the cells of origin for preganglionic fibers carrying inhibitory impulses to the heart by way of synapse in the cardiac gangha. This observer identified a group of larger cells, intermediate in type and size between the remaining cells of the dorsal efferent nucleus and the neurons constituting the hypoglossal nucleus, which he regarded as forming a cardiac center. According to the observations of Kosaka (’09), the heart center in the dog is localized within the distal third of the nucleus ambiguus and within its most ventral part. It would be surprising if there should be confirmation of this localization of the heart center in the nucleus ambiguus, since the other pregangUonic

centers for the vagus lie in the dorsal efferent nucleus. Molhani (’10), Wainstein (’21), and von Husien (’24) located the cardiac center in the dorsal efferent nucleus.

The inferior salivatory nucleus (fig. 293),“ detached from the dorsal efferent nucleus but in the same general

Fig. 292. The nucleus motorius commissuralis Fig. 293. The salivatory center for the glossoof the vagus in the llama. Vcrvieulcn. pharyngeal nerve in the dog. Yagila.

column, provides preganglionic fibers to the glossopharyngeal nerve. These synapse in the otic ganglion with postganglionic neurons supplying the parotid gland.

Aside from internuclear connections with the adjoining visceral afferent column, the pathways of incoming impulses to the dorsal efferent nucleus and the nucleus salivatorius inferior do not appear to be known with much certainty. The results obtained from clinical cases and physiological experimentation suggest direct or indirect connections with the hypothalamus (see Chapter VIII), and one possible pathway for such impulses, by way of the diencephalic periventricular system to the dorsal tegmental nucleus, and through the dorsal longitudinal fasciculus of Schiitz, has been suggested {Huber and Crosby, ’29, ’30 ; see also figs. 512 and 530).

” Voris and Hoerr (’32) stated that the nucleus dorsalis motorius n. vagi (the dorsal efferent nucleus of the above account) “sends only a few visceral efferent fibers to the glossopharyngeal nerve” in the opossum. Later they speak of the possible interpretation of a small group of neurons ventrolateral to the fasciculus solitarius as a nucleus salivatorius inferior.

Rtl

1 — md. ud r*o( IV

md ud root I rod ud root Vn

mdVI

rod ud root X and XI; CCDs r.LaiI)orKk<t.l)

- nod ud root Y _ rod ud root DC

- BodDtiters

rod ud root tpio., occ. ud XII

e.Vn,IX= |en!iYn,IX

_ inlerior oSnrj rod

i =

_ alaniB scriplorias

Fia. 294. A,B,C,D. A comparison of a series of projections on the sagittal plane of the motor nuclei and motor roots in the respective forms considered.

A, Petromyzon fiuviatihs {lllc + P/, equals III root plus VI root = abducens of authors, oculomotor-abducens of Addons ; the position of the VI nucleus could not be ascertained by Addons) Addons; B, Scyllium canicula, Ariens Kappcrs; C, Leuciscus rutilus, van dor Horsl; D, Triton cristatus, Rdthig.

594

lod usd net m

isdtsiInellT

isseL usd not V

Icj

_ tad usd not Vn tad TI

cadu:dnolXtssdXI,ctib rXes!so»d(cl)

tadusdn^K

I— eadDeilen

|_ bxL usd not spai, ecc. usd Xn

gVn.K=ieBaYII.IX

tsfersor oEnij tad

scT^oms

596 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

While during phylogenetic development the preganglionic neurons of the vagus and the glossopharyngeal nerves retain their position near the floor of the ventricle, presumably under the neurobiotactic influence of the visceral afferent centers associated with the fasciculus solitarius, from which the preganglionic neurons receive their most important stimulation, the neurons supplying branchiomeric muscle and constituting the mammalian nucleus ambiguus gradually shift ventrolaterally from their primitive position near the floor of the ventricle. This change in position of these efferent neurons is said to be due to neurobiotactic influences of cutaneous impulses distributed to the nucleus of the descending root of the trigeminal, near to which nucleus the nucleus ambiguus takes its position. This shifting affects the caudal part of the vagal complex first, and is evident in reptiles. A shifting of a part of the vagal complex occurs in birds, but a comparison of the portion of the vagus nucleus which has a ventral extension in birds with the mammalian nucleus ambiguus shows that the nucleus in the latter animals is larger. This increase in size is due to several causes. In the first place, the single mammalian nucleus includes the neurons of the glossopharyngeal nerve which supply the stylopharyngeus muscle and which form a separate nucleus in birds (Addens). In the second place, the vagal portion of the nucleus is enlarged with the enlargement and more precise differentiation of the muscles, and particularly the laryngeal muscles, innervated by these neurons. With the exception of birds, no animals other than mammals produce highly complex sounds. Birds have a special apparatus, the syrinx, concerned with the production of sound, and their lar 3 mgeal musculature is much less well differentiated than that of mammals. For the progressive evolution of the caudal visceral efferent column see the several parts of figure 294.

In lower mammals the nucleus ambiguus (fig. 295) frequently extends in front of the level of the cephaUc pole of the dorsal efferent nucleus, the cephahc part being clearly thicker than the caudal portion. Caudally, particularly in the last third of the nucleus ambiguus, the cell mass becomes so scattered that in many sections either no cell bodies at all, or possibly only one or two neurons, can be identified. According to Kosaka’s (’09) work on the dog, this last third can be subdivided further into a dorsal portion supplying the laryngeal musculature (except the cricothyroid) and a ventral portion supplying the heart. As has been previously stated, Malone’s work on man appears to indicate the presence of the heart center in the dorsal efferent nucleus (see page 594). The middle third of the nucleus ambiguus, according to Kosaka, is concerned chiefly with the innervation of the veh palatini. From the anterior third of the nucleus arise fibers for the cricothyroid muscle, for the muscles of the pharynx (including the stylopharyngeus muscle) , and for the striated muscle of the oesophagus. There is some question among certain observers as to the exact place of separation of the glossopharyngeal and vagus centers. (For the ambiguus nucleus in the pig consult Fazzari, ’29).

In addition to short intemuclear connections from near-lying medullar centers, afferent paths to the nucleus ambiguus are provided from several higher centers of the nervous system, chief among which are cortico-bulbar fibers from the contra

THE EFFECTORY SYSTEM

697

lateral and homolateral {Favill, ’33) precentral gyri of the cerebral cortex, corticostriate (Muskens, ’22) fibers, lateral tecto-bulbar fibers, and cerebello-motorius fibers. The lateral tecto-bulbar fibers are partly crossed (in the tectum), the number of crossed fibers depending on the mammal under consideration. The cerebello-motorius fibers are both homolateral and contralateral.

Special visceral afferent or gustatory impulses from the epiglottic region and general \dsceral afferent impulses from the neck to the descending colon are carried by neurons of the vagus, vdth cells of origin in the ganglion nodosum and neuraxes in the visceral area at the level of entrance of the nerve and in the gray associated with the postvagal fasciculus solitarius.

Special visceral affeient or gustatory impulses from the posterior third of the tongue and general ^^sceral sensibility ^

from the same region of the tongue, as well as from the pharynx region, pass over neurons of the glossopharjmgeal nerve, with cells of origin in the ganglion petrosum and neuraxes which terminate in the visceral sensory area at the “

level of entrance of the glossopharyngeal nerve and in the gray of the descending fasciculus solitarius. The details of this distribution have been discussed in Chapter III (pp. 374 to 377) and need not receive further consideration here. Cutaneous sensory fibers (general somatic afferent) are supplied to the external ear over the ramus auricularis of the vagus, and possibly over the glossopharyngeal {Herrick, ’31). Such fibers arise from cells in the ganglion jugulare and ganglion superius (or superior part of ganglion petrosum). The neuraxes of such cells distribute to the nucleus of the descending root of the trigeminal, along with other cutaneous impulses from the face (see Chapter III, p. 377).

Before closing the account of the vagus and glossopharyngeal nerves, attention must be directed to certain peculiarities of their root fibers. In the first place, these nerves belong to the so-called lateral motor nerves, and their fibers emerge (and enter) in a lateral position immediately above the inferior oHvary nucleus. The preganghonic fibers of the vagus pass directly ventrolateralward from their cells of origin in the dorsal efferent nucleus to their place of emergence. They are accompanied on their lateral side by incoming vagus fibers which pass directly to the descending root of the trigeminal if somatic afferent, and to the fasciculus solitarius and associated gray if visceral afferent. The special visceral

598 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

efferent components have a longer course, which is a reminder of their position near the floor of the ventricle during ontogenetic and phylogenetic development. Such special visceral efferent fibers rvm dorsomedially from their cells of origin in the ambiguus nucleus until a position in relation to the dorsal efferent nucleus is reached, and then accompany the general visceral efferent fibers peripheralward. The courses just described for the vagus fibers are repeated in all essential respects by the root fibers of the glossopharyngeal; for this nerve also, the preganglionic fibers and afferent fibers follow relatively direct courses within

the brain, while the special visceral efferent fibers form a conspicuous dorsal loop to come into association with the other components of the nerve. The root fibers of the glossopharyngeal nerve, in their course toward the periphery of the brain in many mammals, for the most part pass through the most dorsal portion of the descending root of the trigeminal nerve, while the vagus fibers pass through a more ventral portion of the trigeminal descending root fibers. These differences in relation are due to changes in the position of the descending root of the trigeminal.

In mammals, the facial nerve is known to carry general and special visceral efferent and general and special visceral afferent and is thought to carry general somatic afferent (proprioceptive) fibers. The general visceral efferent components, for the most part, have origin in the nucleus salivatorius superior. This nucleus has been identified in many mammals, including man. In general, it occupies a dorsal position in line with the dorsal efferent nucleus of the vagus and the inferior salivatory nucleus. It may lie in close relation with this latter nucleus, as was found to be the case in the dog by Yagita and Hayama (’09), and Yagiia (’09). In man the two nuclei are in the same general line and at no great distance from each other. In the dog {Yagita and Hayama, Yagita; fig. 296) the superior salivatory nucleus in part occupies a dorsal position, but certain cells forming a portion of it are scattered along the nucleus of the descending root of the trigeminal. However, it is to be noted that in many cases the nucleus maintauis a position relatively near to the visceral sensory centers of the medulla oblongata, another example of the neurobiotactic influence of these visceral centers. Fibers (partly crossed, Kohnstamm, ’02, Lorente de N6, ’22, Kaida, ’29, etc.) arising in the superior salivatory nucleus distribute through the chorda tympani to the submaxillary ganglia in the chorda lingual triangle. From this group of ganglia, postganglionic fibers distribute to the sublingual glands. Certain of the preganglionic fibers, without synapse, pass through the submaxillary ganglia in order to reach the submaxillary gland to terminate in subcapsular, pericellular synapses in small, scattered ganglia at the hilus of the submaxillary gland and along its major duct, collectively known as the ganglion of Langley.

Flo. 296. The superior salivatory nucleus of the facial nerve in the dog. Yagiia and Hayama.

\

THE EFFECTORY SYSTEM

599

Another general visceral efferent or preganglionic center of the facial nerve has been demonstrated experimentally by Yagita (’14). This observer regarded the center as concerned with supplying preganglionic fibers through the greater superficial petrosal nerve to the sphenopalatine ganglion, the impulses passing from there by way of postganglionic fibers to the lacrimal glands.

This latter nuclear group may constitute the dorsal facial nucleus of van Valkcnburg (’10), identified and studied by him in various mammals and man. This cell mass previously had been identified by Pacetti (’98), who regarded it as an abducens nucleus. Van Valkcnburg believed this inteipretation to be incorrect, since after destruction of the abducens nerve, this nucleus presented no evidence of injury. However, it should be noted that the nucleus identified by van Valkcnburg (’10) as the dorsal facial nucleus, appears to have been regarded in the more recent work of Prcziuso (’24) as an accessory abducens nucleus ; this is in agreement with the results of Addens (see fig. 294H of the rabbit) . Incoming fibers to the general \’isceral efferent centers of the facial, other than those from the nearlying afferent centers of the medulla oblongata, including the gray associated with the fasciculus solitarius and possibly the short internuclear connections with the nucleus of the descending root of the trigeminal nerve, are not well known as yet. Presumably they receive, directl}’’ or indirectly, impulses from the hypothalamic and probably epithalamic regions, and it is believed that certain of such impulses may course along the dorsal tegmental nucleus and the dorsal longitudinal fasciculus of Schiitz {Huber and Crosby, ’29, ’30; see also figs. 512 and 530).

Special visceral efferent fibers of the facial take origin from cells of the socalled facial (or ventral facial) nucleus, which may present secondary divisions. The position of the facial nucleus, owing in part to neurobiotactic influences and in part to mechanical factors, varies to a considerable extent in mammals. A brief account of its relations in the several orders in this class follows. The facial nucleus of monotremes shows entirely different relations than does the corresponding nuclear group in other mammals. In Echidna (fig. 297), as well as in Ornithorhjmcus {von Kolliker, ’01), observers have described two facial nuclei. The larger, more ventral group extends from a short distance in front of the glossopharyngeal root to far beyond the frontal extent of the facial root. This nucleus is situated about midway between the midline and the periphery of the medulla oblongata. The smaller group of cells lies dorsal to the cephalic tip of the main facial nucleus and extends somewhat farther frontalward than does this latter nucleus {Aliens Kappers, ’20) P In the ventrolateral part of the medulla oblongata these nuclei have a less ventral and consequently more primitive position than that of the nucleus in higher mammals. The facial efferent nucleus in monotremes extends farther forward than in higher mammals, and the e\ddence indicates that this is due to an increase in the size of the nuclear mass rather than to a frontal shifting of its constituent cells. In order to understand these differences, the development of the facial muscles and their relative position in Echidna

“In position, this nucleus suggests the dorsal facial nucleus described by van Valhenburg (’10) and regarded by certain observers as an accessory abducens nucleus. The significance of this dorsal group is not known ; it awaits further study (see p. 610).

600 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and in higher mammals (figs. 298A, B, C) must be understood. The musculature supphed by the facial nerve, which as yet is very simple in reptiles and birds, forms an elaborate system in mammals. Particularly great changes are found in the circular muscular system which has arisen in reptiles as the sphincter colli. The superficial layer of this sphincter forms the platysma in mammals, which is found as subcutaneous muscle partly in the neck region and partly in the head region. From the platysma there differentiates the orbicularis oculi, the quadratus labii inferioris, and the mentalis {Ernst Huber). Other changes associated with the

Fig. 297. Dorsal facial nucleus (at the right) and ventral facial nucleus (at the right and left) in Echidna. {Ariens Kappers, ’20, see discussion.)

developing of the facial musculature in mammals are to be found in the ear muscles and those of the nose. The deeper layer of sphincter musculature goes even further in differentiation and development, particularly in connection with the mouth, where the orbicularis oris, the triangularis, and the buccinator are developed. It is apparent that the term “facial” nerve is particularly appropriate, then, in mammals where the muscles of the face are highly developed. These muscles are of great significance in eating and in connection with the greater efficiency of special senses, such as the muscles associated with the eye, the ear, and the nose. In monotremes there are certain differences in the positions of the muscles supplied by the facial nerve, for while portions of the musculature receiving fibers from this nerve are in the mouth, ear, and eye regions in Echidna, other portions of the same muscle system cover the muscles of the pectoral region and intermingle with those to the anterior extremities. This distribution of muscles is supposed to be associated with the presence of spines on the surface, to

THE EFFECTORY SYSTEM

601

which fibers of this system can be traced. The fact that Echidna has a larger facial center than any other mammal is undoubtedly due to the wide distribution of its fibers in association with the enormous development of the musculature which it supplies. It will be seen that this great development of the facial is in direct contrast to the limited development of the trigeminal motor centers in these forms, in which there is only slight development of the masticatory musculature.

Fio 298 The phylogenetic relationship of the facial nucleus and the facial musculature A, facial nucleus, Scylhum canicula, and facial musculature of a shark (Rttge); B, facial nucleus of Varanus salvator and facial musculature of Varanus {Ruge ) ; C, facial nucleus of Mus musculus and facial musculature of Mus musculus {Parsons)

The sensory components of the trigeminal are large in Echidna, and it is through them that the impulses are brought to the central nervous system, where they are distributed by short internuclear fibers to the facial nuclei. The importance of these reflex connections undoubtedly has had some neurobiotactic influence in shifting forward the neurons of the facial nucleus. This influence is increased somewhat by the harmonious action of the trigeminal and facial motor centers in these animals, for the poorly developed masticatory action is supplemented by a sucking action on the part of the animal, which is made possible through the

602 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

contraction of muscles supplied by the facial nerve. The factors producing the two facial nuclear groups in Echidna are not understood as yet.

In other mammals the facial nuclear mass occupies a ventral position. Usually

it lies caudal to the point of entrance of the facial root, but in higher apes and man, with the caudal shifting of the root fibers in consequence of the development of the pons, the fibers and their cells of origin are found at about the same level. The caudal position of the facial nucleus in most mammals is in harmony with

THE EFFECTORY SYSTEM

603

the relations described previously for reptiles, urodele amphibians, and gnathostome fishes. Its position, then, is to be regarded as an inherited one, which is retained in mammals because of the development of gustatory centers at a level far caudal to the entrance of the facial root. Embryologically the facial nucleus occupies a dorsal position. Later this primitive position is abandoned by all but possibly a very small group of cells, and the nucleus moves in a ventral direction. In lower mammals it is caudal to and somewhat medial to the superior olive (figs. 299, 300). With the increase in size of this olive, frequently the nucleus is forced even farther caudalward, as is to be seen in the dog (fig. 302A), the cat, and in Phoca. In other higher mammals it may take a position lateral to the superior olive and even somewhat frontal to it (fig. 301). Since this facial nucleus contains centers for the musculature associated with the face and with the sense organs (eyes, nose, and ears), it is not surprising that it should occupy a ventral position, since the reflex paths which bring the stimuli to these facial motor cells likewise are situated in the ventral part of the medulla oblongata.

Of all of these sensory systems, the most important for the facial are the chief sensory nucleus and the nucleus of the descending root of the trigeminal. The trigeminal branches which innervate the

eye, the forehead, and the skin of the Fla. 301. Dorsal and ventral facial nuclei in nose (fig. 173) run in the most ventral man, following i4n^AappCTs(’20). The Dors.

,. j j j- T 4 - nucl. of VII probably is the accessory abducens

portion of the descending root. It is nucleus ofilt/cfens (seep. 610). Huher and Crosby.

evident that this branch carries the most

important impulses to be transferred to the facial musculature. Ram6n y Cajal (’09) found also that the cells of the superior olive give rise to a series of short neurons which terminate in the motor nucleus of the facial nerve. The influence of soimd on the stapedius muscle and the muscles of the ear (in the horse, for example) is another evidence of the significance of the close positional relation of the facial nucleus to the superior olive (figs. 296 and 301). It is not unlikely that the position of the cortico-bulbar fibers may have some influence on the position of the facial, although researches of Toyojiiku (’10) indicated that in certain cases in which there was a lack of development of this tract in man, the ventral shifting of the facial nucleus occurred uninfluenced. However, H. Vogi (’05) showed that in two cases in which not only the cortico-bulbar S 3 "stem was lacking, but also the reflex paths of the medulla oblongata, and particularlj'^ the superior olive, were atrophied, only a partial shifting ventralward of the facial nucleus had occurred. The fact that in the highest mammals and man the facial nucleus shifts away from the cortico-bulbar system toward the descending trigeminal root (as evdneed by its change of position from the medial to the lateral

604 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

side of the superior olive as well as its tendency to shift somewhat farther frontalward) indicates that of the two factors — a cortical or a medullar reflex influence — the latter is more important in determining the position of the nucleus.

Various observers have attempted to localize within the facial nucleus centers for the specific muscle groups supplied by this nerve. Among such observers may be mentioned van Gehuchlen (’99 and ’03), Marinesco (’98 and ’99), Yagila (’10), Pa-pez (’27), and others. Yagila found that the chief or ventral facial nucleus, with the exception of its extreme distal and proximal ends, shows a distinct division into a dorsal and a ventral part. In man the dorsal division is the larger, but in the lower mammals studied (such ds-the rabbit) the converse is true. In its central portion the nucleus has, in addition to the dorsal and ventral groups, an intermediary group that is well developed in the dog, poorly developed in man, and in some animals, such as the rabbit, is intermingled with the ventral part. The dorsal group in the dog, for example, can be redivided into medial and lateral portions, and the ventral nucleus into middle, medial, and lateral parts. These subdivisions, of course, do not occur at the proximal and distal ends of the nucleus. In a general way, the dorsal part gives rise to the upper facial branch of mammals, the intermediary portion to the branches to the ear, and the ventral part to the lower facial branch.

Y agita (’ 10) — in agreement with most other authors — has shown that the upper facial branch arises from the dorsolateral part in the dog and from the entire dorsal part in the rabbit. The branches which innervate the external ear musculature arise in the dog from the lateral part of the intermediary division ; in the rabbit, apparently from the medial ventral group. He was not able to localize the center for the stapedius, but he found that the removal of the facial root in the foramen stylomastoideum caused no degeneration of the anterior part of the facial nucleus. This would suggest that the branch to the stapedius muscle was intact. In dogs the ventral nuclear division forms the central portion of the under facial branch in that the medial part gives rise to the nervus subcutaneous colli, while the middle and lateral ventral groups send fibers to muscles around the mouth supplied by the facial. Marinesco and van Gehuchlen recognized a dorsal, a lateral (or external), a middle, and a medial group of cells within the facial nucleus ; the three last groups probably corresponding to the subdivisions of the ventral nucleus as given by Yagila. Papez (’27) studied the localization within the facial nucleus inrats, guinea pigs, dogs, and cats by extirpating the various branches singly and studying the resulting chromatolysis. He obtained the following results :

Branch destroyed

zygomatico-orbital superior labial inferior labial

anterior auricular

posterior auricular posterior belly of digastric

Changes in

large dorsal (dorsolateral) group lateral group ventrolateral group

(connected with lateral) intermediate group, lateral to medial group

dorsal part of medial group small ventromedian cell group.

THE EFFECTORY SYSTEM

605

Pa-pez was of the opinion that the platysma, and perhaps the stapedius, are innervated from neurons with cell bodies in the ventral part of the medial group, and that cell bodies in the intermediate group send neuraxes to the sphincter colli profundus.

Reference has been made in the preceding paragraphs to many of the paths which terminate in the special visceral efferent nuclear gray of the facial nerve. Such connections are here briefly summarized : In addition to short internuclear fibers connecting the facial center with the surrounding gray (for example, the nucleus of the descending trigeminal, and probably its chief sensory nucleus as well, and the superior olivary nucleus), there are fibers from the other motor nuclei, particularly the oculomotor and the contralateral hypoglossal nuclei, through the medial longitudinal fasciculus (see page 1078, also fig. 506), connections with the tectal region through the lateral tecto-bulbar tract, with the red nucleus through the rubro-bulbar tract, and with the precentral gyrus of the cerebral cortex through the cortico-bulbar tract. The latter tract carries fibers of contralateral origin to all parts of the special visceral efferent nucleus and also homolateral fibers to the dorsal part. Clinical and pathological observations document this double innervation of the more dorsal neurons of the facial nerve, for destruction of the cortico-bulbar connections on one side does not produce paralysis of the upper part of the face and regions around the eye, although the lower part of the face shows an upper motor neuron paralysis.

The course of the facial root in mammals is well kno^vn. There has been considerable question as to whether or not contralateral as well as homolateral special visceral efferent fibers are present, but the consensus of opinion at present seems to favor the imilateral origin, although Windle (’32, ’33) found crossed fibers in cat embryos. The literature on the subject is discussed in Yagita’s 1909 paper and also in Addens’ 1934 report. Arising from cell bodies in the facial nucleus, the fibers swing forward, dorsalward, and medialward, producing a so-called knee around the nucleus of the abducens, and then s\ving ventrolateralward in order to emerge. In the majority of mammals, this knee of the facial root is fixed by the abducens nucleus, but that this nucleus does not serve necessarily as a fixed point is evident from the fact that the knee of the facial is present in forms such as teleosts, where the abducens nucleus lies ventrally, or in others such as Talpa, in which an abducens nucleus appears to be lacking. In such instances the dorsal arcuate fibers may keep the knee in this position. That part of the genu of the facial which extends around the abducens nucleus is sometimes termed the anterior genu and the fibers which are found caudal to this nucleus are then called the posterior genu. In addition to these loops, in certain animals there is a third looping of the fibers, which may be termed the genu inferius or the ventral genu. This last is formd only in those animals that have a very large trigeminal nucleus and in which a marked development of the trapezoid bod}’- has pushed the facial root fonvard. In such forms the trigeminal nucleus, or more probably the reticular cells that immediately surround it, exercises a caudal pressure upon the facial root at about the middle of its course through the medulla oblongata. This produces a knee which is very marked in such animals as the

606 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

cat, the dog, and the horse. The posterior border of this caudal convexity of the root is at the caudal pole of the trigeminal nucleus behind the entrance of the facial root. It is evident that this hne in the.jdog lies far caudal to the level of entrance of the facial root. In the anthropoid apes and man, the increase in

V

Itj

Bdulmlin

axludmlYII

udarimlT ■dial ml DC..

_ ■cli«lmlXuJXI;e*rJM»«rf(£l)

acLDeilm

■cluJrM(i)ik,tcc.aiJ]CO

■cLEJ^tr-VtiifU

■fcriorttnijad.

Irs

tliniTMl J.- calm lOflinil

In^the and nuclei of dog ape, and man.

somewhat more caudalward than is sho^-n in th^c We ^ amb.guus extends

^idog; -5, Oedipomidae, one of the lower apes; C, man.

ITA!? caudalward shifts the facial rootlets in a caudal direction (comdfsappeam'^^"'”"^

abducens nerve presents points of interest for discussion.

SL abducens nucleus occur in submam mahan forms. Such vanations have been recounted in the preceding pages and are summanzed on page 623. The phylogenetic history of the abducens is re

THE EFFECTORY SYSTEM

607

peated to a considerable extent in its ontogenetic history. This is seen in the frontal shifting of its nucleus from a more caudal embryonic (and phylogenetically more primitive) position {Streeter, ’08; fig. 303).

In human and other mammahan embryos at an early stage (10 mm. in man) small grooves, which have a branchiomeric or neuromeric significance, are apparent on the floor of the fourth ventricle. If, following Streeter (’08), these small grooves are termed A, B, C, D, E, and F, then the grooves A and B correspond

V J vn/y vi"MVk VVk

Fig. 303. Topographic relations of V, VI, VII, IX, in certain vertebrates. A, shark; B, teleost;

C, frog; £), turtle; B, desert lizard; F, man. Ariens Kappers.

to the trigeminal, C to the facial, D to the abducens, E to the glossopharyngeal, and F to the vagus. In young embryos the nucleus of the abducens and the beginning of its root fibers are discernible below the groove D and occasionally partly below that of E. This corresponds to a position behind the facial (C) and (at E) approaching the level of the glossopharyngeal, an ontogenetic repetition of the phylogenetic condition pointed out in Chelone and the alligator. During the progress of development, however, the nucleus of the abducens shifts forward and attains the level of the facial, which, in its turn, during ontogenetic as well as phylogenetic development, is shifted both caudalward and lateralward. Bremer (’08) and others have found that in young embryos the abducens not infrequently shows a number of small rootlets and a caudal continuance of its nucleus as far

608 NEEVOUS SYSTEMS OF VERTEBRATES AND OF MAN

back as the fifth neuromere of the rhombencephalon. Such rootlets are usually aberrant and do not reach a normal termination in muscle but run into the loose mesenchyme. Where sufficiently developed, according to Bremer ('21), they pass to the anterior spinal musculature, the branchial musculature, or the dorsal muscles of the head. Bremer (’21) found that among the mammals studied by him, only in man did the abducens rootlets appear in order from behind forward, with the oculomotor making its appearance later.

For adult mammals the abducens nucleus has a relatively constant position and sends its root fibers out at the level, and slightly in front of the level, of the

emergent facial roots, so that it has become in truth a sixth nerve. In a few mammals, as carnivores (fig. 304) and Phocaena, the abducens nucleus has a medial position, beside the medial longitudinal fasciculus. Usually the abducens nucleus of mammals differs from that of birds in that its cells do not lie quite so close to the medial longitudinal fasciculus, but lie somewhat farther dorsolaterally. As a result the neurons do not occupy a position under the horizontal arm of the facial root, but, to a very considerable extent, lie lateral or even dorsolateral to it, much nearer to the ventricular floor, as is evident particularly in rabbits, horses, and man (figs. 305 and 330) . This shifting finds expression in a laterodorsal turn made by the abducens root. The laterodorsal shifting of the abducens nucleus in certain mammals is an indication of the importance of the connections between the nucleus in question and the nuclear centers of the vestibular nerve, particularly the nucleus of Deitersand thenucleus triangularis or principalis. F use (’12) has pointed out that the abducens nucleus may be divided into a portion situated nearer the ventricular floor and a reticular portion. The first-mentioned portion, which is the more dorsolateral and lies along the genu of the facial, consists chiefly of cells of varying sizes, but the second part, the principal nucleus, contains large cells. Modifications of this are found in various mammals ; for example, in the cat and the dog large cells are found below the genu of the facial in the reticular portion, but in the rabbit, the goat, and the macaque they lie in the ventricular portion lateral to the genu of the facial nerve. The small and middlesized cells which were included in the abducens nucleus by Fuse (’12), according to his own experiments, do not give rise to abducens root fibers. Indeed, the ventricular portion is present in the mole, in which, as Fuse (’12) himself stated, an abducens nucleus is completely lacking. He came to the conclusion, therefore, that the true nucleus of origin for abducens fibers (der Hauptkern) in cats and dogs lies farther under the facial in the substantia reticularis dorsalis beside the medial longitudinal fasciculus, but that in rabbits and in goats (and also in rodents, ungulates, and primates) it lies to a considerable extent lateral to the

Fig 304 The position of the abducens nucleus in the cat, by the side of the media! longitudinal fasciculus below the horizontal part of the facial root Huber and Crosby

THE EFFECTORY SYSTEM

609

genu of the facial, as had been stated earlier {Ariens Kappers, ’10). That, as a matter of fact, the small and medium-sized cells of the nucleus of the abducens are not cells of origin for root fibers is supported by experiments based on the cutting of the root. It hardly seems justifiable to describe elements as nuclei of an efferent nerve which do not give rise to root fibers. That many nuclei of the medulla oblongata during their phylogeny are surrounded by smaller elements, as well as by large-celled leticular nuclei, has been generally recognized {van Hoevell, see Chapter VI ; Ariens Kappers in various papers on motor nuclei , and others).

Horizontal part Pnncipal of root of VII nucl VIII

Lat vest nucl. (Dei tens)

Nucl VI

^ Hoot of VI

An accessory abducens nucleus has been recognized in mammals by Terni (’22), Preziuso (’24), and Addens (’33, shown in figure 294H of the rabbit).^® Adders (’33) described the nucleus in the human embryo. In figure 301, a dorsal facial nucleus is shown following the terminology of Ariens Kappers (’20, fig. 280) and also the interpretation of van Valkenburg (’10). This nuclear group, which consists of only a relatively few neurons, may correspond to the accessory abducens nucleus described in the human embryo by Addens (’33). In the material available at the University of Michigan it has not been possible to trace neuraxes of cells of this group into either the abducens or facial roots although the cell bodies he close to facial root fibers. If the neurons in question be regarded as representative of the accessory abducens nucleus, they probably must be regarded either as a vestigial nucleus or as serving some other function than that of supplying the retractor bulbi — which is ascribed to them in various other forms — since this is said to be lacking in primates and man. In view of

“ Addens is inclined to believe that the dorsal facial nucleus descnbed by van Valkenburg (’10) is the accessory abducens nucleus of Terni, although van Valkenburg appears to have been quite convinced that it did not give nse to fibers of the abducens nerve.

610 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

the inadequacy of their present knowledge of this nuclear group, Huber and Crosby must join those observers who are unwilling to express an opinion as to whether or not the cell group labeled “ dors. nucl. of VII ” in figure 301 supplies fibers to the facial or abducens root bundles.

Entering the abducens nucleus of mammals are collaterals and stem fibers from the medial longitudinal fasciculus, which interrelate this nucleus with the other eye muscle nuclei and the vestibular nuclei. The abducens nucleus also receives connections from the superior olivary nucleus by way of the peduncle of the superior olivary nucleus. The foregoing connections are illustrated in figure 506, Chapter VIII. Cortico-bulbar fibers from the motor cortex also

terminate in the abducens nucleus, crossing at the upper border of the pons.

The trigeminal nucleus of mammals (fig. 306) differs in no marked degree in the various representatives in this class. In nearly all of the mammals, but especially in the higher mammals, the trigeminal nucleus can be differentiated from that of birds and reptiles through the presence of a relatively larger number of small and middle-sized cells, which surround the nucleus and are strewn throughout it. This is particularly the case in the caudal end of this

Fia. 306 . Motor trigeminal nucleus in cell mass. In the monotremes the trigeminal the cat. Huber &na Crosby. i , i -i i , it

nucleus holds a somewhat more dorsal position,

and, as is to be expected, is relatively small, since there is noted a marked atrophy of musculature {Gegenbaur, ’98), especially of the masseter and temporal muscles, innervated by the trigeminal nerve. Also in Echidna, which has no teeth and does not chew its food, the motor trigeminal nucleus is small. The trigeminal nucleus attains its greatest size in carnivores, particularly in Phoca and Canis, and it is also very well developed in the horse (fig. 330). There is a quite perceptible shifting of the nucleus frontalward, or again caudalward, depending on the form studied. In Echidna it occupies a more frontal position, no doubt as a result of the marked frontal extension of the chief sensory nucleus in these forms. The upper and middle portions of the trigeminal root, carrying sensory fibers of the maxillary and mandibular branches of the nerve, which innervate the jaw region and the mouth, influence particularly the topographic position of the motor nucleus of the trigeminal nerve, not only in Echidna but also in many other mammals, including man. In many mammals the efferent trigeminal nucleus extends beyond the frontal limit of the sensory trigeminal nucleus, due perhaps to its reception of many collaterals of the mesencephalic root of the trigeminal (see the chapter on branchiomeric nerves). Van Valkenhurg’s study (’10) of the development of the trigeminal nucleus in man shows an interesting correspondence between its ontogenetic and phylogenetic history. Thus, both ontogenetically and phylogenetically considered, this nucleus occupies a dorsal position and only gradually moves into a somewhat more ventrolateral position. In the mam

THE EFFECTORY SYSTEM

611

malian embryo the more caudal cells are the first to wander ventralward, while for a long time the more cephalic cells occupy a dorsal position. In his study of the development of the motor trigeminal nucleus, van Valkenburg (TO) found a small nuclear group (the nucleus trigeminus posterior) caudal to the main nuclear mass. This small nucleus unites later with the chief nucleus. T his phase of the ontogenetic development shows resemblance to a phylogenetic stage found in fishes and reptiles, where a small part of the under portion of the motor trigeminal nucleus migrates to a position ventral to the main nucleus. Whether the nucleus trigeminus posterior of van Valkenburg represents such a ventral migration in mammals is uncertain as yet. Windle (’33) described a medial and a lateral trigeminal nucleus in the younger (8 and 10 mm.) cat embryos — as had Beccari (’23) in the 11 mm. guinea pig embryo — but most of the medial cells had migrated laterally by the 11.5 mm. stage of the cat embryo.

The main incoming fibers to the motor nucleus of the trigeminal, aside from short internuclear fibers from the sensory nuclei associated vdth the nerve and perhaps with the visceral efferent centers of the pons and the medulla oblongata, are : cortico-bulbar fibers from the contralateral and from the homolateral {Favill, ’33) motor cortex ; collaterals and perhaps stem fibers of the descending tracts from the tectum and the red nucleus ; homolateral and contralateral cerebello-motorius fibers ; and probably fascicles from the medial longitudinal fasciculus, which connect the trigeminal motor centers with other motor centers of the brain stem. Neuraxes of the mesencephalic root of the trigeminal {Weinberg, ’28, and others ; see Chapter IV, p. 406) also end in relation with the motor neurons of the trigeminal.

The motor root fibers of the trigeminal pass directly from their cells of origin toward their place of emergence at the lateral border of the pons. They do not form the loop characteristic of the course of other special visceral efferent fibers in mammals. In accordance with the extreme frontal position of the cells of origin of the motor root, the fibers of the trigeminal root lie well forward in monotremes. The gradual increase in the ponto-cerebellar fibers forces the trigeminal root fibers farther and farther back in higher mammals, so that the place of emergence of the efferent trigeminal root approaches that of the efferent facial root. However, during phylogeny, gradually the trigeminal fibers begin to emerge through the fibers of the pons, while the facial root continues to be forced back by the pons formation, increasing again the distance between the planes of emergence of the two nerves.

The cephalic eye muscle nuclei have been identified and studied by observers of the mammalian midbrain to an extent precluding the necessity of presenting here anything approaching a complete review of the mass of pertinent literature that has accumulated in this field. An attempt is made to describe and locate these nuclear masses and discuss the conceptions of structural and functional relations which have been codified out of these various contributions. The interested reader will find further references in the texts of Obersteiner (’01), Ramdn y Cajal (’ll), and Edinger (’08), in the papers of Gastaldi (’23), in Mollendorfs "Handbuch der Mikroskopischen Anatomie” {Mingazzini, ’28), and in many other of the larger texts treating of this region.

Raph6 cella

Nucl IV

Fia. 307 . The trochlear nucleus of the rabbit.

612 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The cephalic eye muscle nuclei of mammals, as represented in the mesencephalic centers for the trochlear and oculomotor nerves, have certain features in common. Three different types of topographic relations exist between them : (1) The two nuclei may be separated completely, as in Echidna, a relation reminiscent of the conditions found in certain reptiles. A gap between these

nuclei has been described also for the cat by van Valkenhurg (’12). (2) The two

nuclei may be separated merely by a more or less definite layer of myelinated fibers, analogous to the fiber bands separating the various parts of the oculomotor nuclei, particularly in birds. (3) Trochlear and oculomotor nuclei may form a continuous nuclear column, as has been described for the opossum by Voris and Hoerr (’32). In the majority of mammals it is with the dorsolateral nucleus of the oculomotor that the trochlear nucleus is continuous. Even in the horse, where the trochlear nucleus is situated The oc illnf" T lon^tudinal fasciculus while the dorsolateral nucleus of is continiim ’tb ^ above this bundle, the trochlear nucleus

lateraf dorsolateral nucleus rather than with the ventro imoniZZlf Z.b TZ- The cells connecting the two nuclei lie

ventrofloriinl °"Situdinal fasciculus, occupying an oblique,

LlmotoZ T u the trochlear nucleus is distinct from the

oculomotor nucleus. It shifts forward gradually

dunng pyiogenetic development, its close approximation to the oculomotor nucleus being a secondarily acquired relation.

Streeter’s study (’08) of the ontogenetic development of the trochlear nucleus in man indicates that the phylogenetic history is repeated during the ontogenetic development of this nu- ii

clear mass. In a human embryo of 10 mm 3

crown breech length, a considerable distance in eijenes between the nucleus of the oculomotor Fig. 308. The trochlear nucleus of the and that of the trochlear. Such separation per cTJcmIv ‘I'O "“'lei n>ay b" ”

connected by scattered cells (Tsuchidu, ’06 ; Obenteiner, ’12). .

the medialTon^Ur’U f "“'1'“® 1'" below the aqueduct in close relation to rrs^a^d "■“y OlEe- 30^. 308) it lies

the fasciculus or ev ’h ” tf b“l it may extend well over the side of

V L -iST '“'bedded within it. In the horse (Mida, ’06;

crmerfen, 14) the nucleus is said to lie ventral to the medi^ longitudinal

Hapb^

Nucl. IV.

xijf'i

4

THE EFFECTORY SYSTEM

613

Root of IV

Post trochlear nuci

fasciculus. The neurons constituting the trochlear nucleus are large and multipolar ; they are typical somatic efferent neurons. Now and again the nucleus may be subdi\dded into a larger cephalic and a smaller caudal portion, as is indicated in figure 310. Tsuclnda f’06) noted such a differentiation in 20 per cent to 30 per cent of the human material studied, but this differentiation and separation, in most cases, were present only on one side. Van Valkcnbitrg (T2) confirmed the observations of Tsiichida in this respect, for he found that either the nucleus was subdmded only on one side or was more completely separated on one side than on the other. In the three cases described by va?i Valkcnhurg, and in the case figured here by Ariens Kappers (fig. 309), the separation occurs on the left side. Small reticular elements, similar to those found in relation with the hypoglossal, abducens, and trigeminal nuclei, are present in considerable number in the region immediately sui rounding the trochlear nucleus and may e.xtend into the nuclear mass. In those cases in which the nucleus occupies its usual position dorsal to the fasciculus, small periventricular gray elements are clustered above the nucleus They do not intermingle with the trochlear nucleus but lie near the floor of the aqueduct at some distance from the nuclear group. They suggest similarly arranged cells at the level of the hypoglossal nucleus. Among the cells of the trochlear nucleus of the rabbit Weinberg (’28) found certain neurons which were identified as of the type of the

309 Nucleus trochleans posterior of a left-sided posterior trochlear nucleus

xiHIICb

H 1 1

Fig 310 Topographic relation of the several groups forming the human oculomotor nuclei and the relative position of the trochlear nucleus in a case where a nucleus trochleans posterior was present

mesencephalic nucleus of the trigeminal. Such neurons, together with others found along the course of the mesencephalic root of the trigeminal, provide a general somatic afferent component of the proprioceptive tjqie. Their neuraxes enter into synaptic relations with special somatic efferent components of that nerve, which constitute the main mass of the trochlear nucleus. Other connections of the trochlear nucleus are with the other eye muscle nuclei through the medial longitudinal fasciculus (fig. 506, p. lOrO) and with the centers for eye muscle movement in the cortex by bilateral cortico-bulbar paths.

614 KER\'OUS SYSTEMS OF VERTEBRATES AND OF MAN

The course of the trochlear root fibers from their cells of origin in the midbrain, dorsalward and caudahvard to tbeir decussation above the ventricle in the anterior medullar}’- velum, is so generally known as to require no further description. This caudal course is due in part, perhaps, to the shiRing forward of the cells of origin, in part to the development of the inferior colliculus and other structures in the region which prevent its forward migration.

The oculomotor nerve carries special somatic efferent, general visceral efferent, and general somatic afferent (proprioceptive) components. As may be anticipated, several nuclear groups are concerned in supplying the nerve fibers of the oculomotor. Variations in size or position of the respective component cell groups of the oculomotor nuclei of mammals are largely in terms of differences in

Medial larce*" nucl. Ill I Lateral / (EdinRcr-Weatphal)

Fifi. 311. A pcriion llirougli the middle third of the human oculomotor nucleus.

Medial \ acccps. nucl. Ill

Lateral / (Edmgcr-Weatphal)

Fig. 312. Section through the anterior third of the human oculomotor nucleus.

.«izc or differentiation of these gray masses. The following nuclear groups are recognized most generally as giving rise to oculomotor root fibers: (1) a chief nuclear group consisting of a dorsal or dorsolateral nucleus and a ventral or ventrolateral nucleus; (2) a central nucleus, the so-called central nucleus of Perlia ; (3) an Edingcr-Westphal nucleus ; (4) a central caudal nucleus. Following a discusMon of these .several nuclei, certain nuclear groups a.ssociated in position and, in .‘■ome instances, perhaps accessory in function, will be considered (see figurc-s 310 to 312).

1 he dorsal and ventral oculomotor nuclei are present in nearly all mammals, and in the consideration of the various forms but slight differences are seen in their .arrangement. 1 he more dorsal cells may be displaced more laterally and may lie partly in or partly above the medial longitudinal fasciculus, while the \entral nucleus is usually .situated between the medial longitudinal fasciculus and the raplif-. .\eeording to Pacctli (’98), the dorsal nucleus has a more or lc.‘’S circular outline, while the ventral nucleus is longer. Further subdivisions of both don-.d and ventral oculomotor nuclei of man into anterior dorsal and posterior dor-al ami anterior ventral and posterior ventral groups have been made (Ming'l.-int, 2S}. In carnivores the ventral nuclei approach each other at the mid

THE EFFECTORY SYSTEM

615

line, and a median cell group, the forerunner of the central nucleus of Perlia, is formed {Brouwer, T8). In higher apes and man this nucleus increases in size at its frontal pole and extends forward between the Edinger-Westphal nucleus of the two sides as the central nucleus {Bach, ’99 ; Brouwer). This latter nucleus

lies in close relation with {Bernheimer, ’96 and ’97 ; Edinger, ’08) , or it contains {Brouwer, '18), the cells of origin for the fibers to the internal rectus muscle, and is regarded by certain observers as the mesencephalic center for convergence {Edinger, ’08; Brouwer, ’18). Le Gros Clark (’26), however, found no relation between the degree of binocular vision in a species and the amount of development of the central or paramedian group.

The exact locahzation of neurons supplying the various striated voluntary eye muscles through the oculomotor nerve has interested many observers, and various schemes of such a localization in man have been plotted {Starr, ’88 ; Bernheimer, ’96 and ’97 ; Edinger, ’08 ; Brouwer, ’18; Mingazzini, ’28, and others). Of such plans, several are shown in figure 313, and of these, that of Brouwer appears to be most nearly in accord with the present knowledge of this area. All of these four observers found the center for the levator palpebrae occupying a position at the frontal end of the somatic efferent column (see Frank, p. 618), close to the EdingerWestphal nucleus. This is of interest in connection with the researches of Economo on the localization of the sleep center in the wall of the midbrain near the tuberculum

the oculomotor nuclear group, according to Starr, Bernheimer, Edinger, and Brouwer. The Edinger-Westphal nuclei are colored black, the chief nuclei are designated by checks, but the nuclei recti interni with the central nucleus of Perlia are indicated by stripes. The trochlear nucleus is left clear.

posterius thalami, since one of the first

phenomena of approaching sleep is the drooping of the eyelids and changes in

the pupils.

The major incoming fibers to the somatic efferent centers of the oculomotor are provided from the motor area of the cortex (cortico-bulbar tracts), from the overlying superior colliculus (tecto-oculomotor fibers), from nuclei of the cranial nerves, particularly the vestibular, trochlear, and abducens nerves, and from the nucleus of the medial longitudinal fasciculus, through this fasciculus. Certain less understood connections of the midbrain and diencephalon will be discussed with their centers of origin. Outgoing fibers connect the oculomotor nuclei

616 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

with the other eye muscle nuclei and the facial nucleus, through the medial longitudinal fasciculus.

The Edinger-Westphal nucleus of man, identified in fetal material by Edmger (’85) and in adult human material by WcsLphal (’87), is a smaller celled nucleus situated on either side of the midline and in close apposition to the central nucleus. With it we have grouped here the nucleus medianus anterior. It is not so well developed in lower mammals as in man. In Echidna it is represented by a few neurons scattered in and above the raphd, which have wandered into this position from the periventricular gray during development. A similar condition is found in marsupials. Such cells form the beginning of an unpaired nucleus, and although this nucleus becomes somewhat better organized in ungulates and progressively better in rodents, edentates, and carnivores, it still remains unpaired. A paired accessory nucleus is found first in monkeys and is present in higher apes and man. The Edinger-Westphal nucleus in man is divisible into two portions : a medial portion, oriented in a more or less vertical direction, and a more horizontally placed lateral portion, although there may be found some fusion of the two groups of cells. According to Panegrossi (’98, ’03), both cell groups may have a horizontal extent, one nucleus overlying the other.

A great number of observers (Bernheimer, ’96, see also ’97, ’97a, ’99 ; Siemerling and Boedeker, ’97; Brouwer, ’18; compare Zweig, ’21) think that the nucleus of Edinger-Westphal is to be regarded as containing cells of origin for the general visceral components of the oculomotor nerves. Such preganglionic fibers pass to the ciliary ganglion (a peripheral craniosacral sjunpathetic ganglion), where they form pericellular, subcapsular synapses {Huher ’96) with postganglionic sympathetic neurons, the neuraxesof which supply the ciliary muscle and the circular fibers of the iris (the sphincter iridis).’ That other points of view are still current is evident from the statement by *Mingazzini (’28, p. 561), for he wrote; “Das Pupillarreflexzentrum muss hochstwahrscheinlich ins zentrale Hohlengrau verlegt werden, auf welches der Edinger-Westphalsche Kern und der Nucleus raphes post, einwirkt (Redlich),” and in the interpretation of Frank (’21) that the Edinger-Westphal nucleus constituted a midbrain center for convergence. The central nucleus {Tsuchida, ’06), is related in position and development with the Edinger-Westphal nucleus (Le Gros Clark, ’26).

The major incoming path to the Edinger-Westphal nucleus is the tecto(or pretecto-) oculomotor tract from the pretectal nucleus (and superior colliculus of man), a tract by which light impulses reach the cells of the nucleus.

The ontogenetic history repeats the phylogenetic development (fig. 314) of the oculomotor nuclei, for the nuclei which appear latest in development are those which were added secondarily during phylogeny. Thus in a fetus of the fourth and fifth month no nucleus of Edinger-Westphal is present, nor can a typical unpaired central nucleus or a paired medial central nucleus be pointed out. In the fetus of seven months the nucleus of Edinger-Westphal makes its first appear nThe Edinger-Westphal nucleus of mammals appears to be homologous with a similarly situated nucleus m birds. The avian nucleus is known to contribute root fibers to the oculomotor nerve. Lalmuten {Le Gros Clark, ’26), however, found no root fibers in mammals.

TIIE EFFECTORY SYSTEM

617

ance. This develops rapidly and later differentiates into two nuclear groups. The unpaired cent ral nucleus of Perlia is present also, but not as yet the smallcelled nucleus medianus anterior. The work of various observers and a consideration of conditions in submammalian forms suggest that both the nucleus medianus anterior and the Edinger-Westphal nucleus develop ontogenetically, as phylogenetically, as accessory structures to the main oculomotor nuclei.

In addition to the three major groups of oculomotor nuclei thus far discussed, there are certain nuclei which lie in the region of the oculomotor nuclei and which

Fig. 3M. The (icvciopraent of tlie Edingcr-Weslphal nucleus for accommodation (Gray) and the fii-sing of the nuclei of convergence (nuclei recti interni) in thcraph6in Carnivora and Arctopithcci to form the central nucleus of Perlia, which, in anthropoid apes and man, is pushed fora-ard (o the level of the nuclei of accommodation. Brouwer.

have been regarded as related to them after one fashion or another. The evidence that any of these nuclei actually contribute root fibers to the oculomotor nerve appears to be unsatisfactory to the present -writers. Several of these nuclei must be considered again in the account of the various tegmental centers of the midbrain. To this group belong the nucleus of Darkschewitsch, the nucleus medianus anterior, the nucleus dorsocentralis posterior, and the nucleus subfascicularis. The nucleus of Darkschewitsch has been regarded by Perlia (’89), Tsuchida (’06), and Mingazzini (’28, p. 649, fig. 56) as giving rise to oculomotor fibers. Such a distribution of its neuraxes has been denied by various observers {von Kolliker, ’96; Obersteiner, ’01; Ramon y Cajal, ’ll; Ariens Kappers, ’20, and many others). By the majority of observers, the nucleus of Darksche-witsch is regarded as giving rise to fibers of the posterior commissure (see p. 1077). In

THE EFFECTORY SYSTEM

619

R^sum^: of the Effectory System of the Medulla Oblongata

AND Midbrain

A study of the comparative anatomy of the effectory systems of the medulla oblongata and midbrain indicates considerable variations in position of the efferent nuclei of the cranial nerves of the region in the various classes of vertebrates. Moreover, it teaches that these changes in position are associated with the functional development of other centers within the nervous system.

One of the most interesting migrations observed is the moving forward of the anterior spino-occipital nerves during their transformation into the hypoglossal. In animals which have no muscular tongue the corresponding center is only a continuation of the ventral horns of the cervical cord, and its cephalic boundary is far caudal to that of the vagus column. Indeed, in Petromyzon (fig. 294A) it does not reach the caudal end of the frontal vagus nucleus. In animals which possess a tongue, the center gradually loses its connection with the cervical cord and undergoes a dorsofrontal shifting to such an extent that in mammniR its frontal end reaches to about the same plane as does the cephahc tip of the dorsal vagus column. With this extension forward, the cells likewise move dorsalward so that to a great extent they he dorsal to the medial longitudinal fasciculus.

The spinal portion of the accessory nerve arises in mammals from the accessory nucleus, which is situated in a lateral position within the gray matter of the cervical cord (to the sixth or seventh cervical segment in man). In the preceding pages, the two current interpretations of its cells of origin, either as constituents, phylogenetically and embryologically, of the somatic efferent column or as a derivative of the special visceral efferent column, have been discussed and need not receive reconsideration here. The manunalian spinal accessory nerve probably is represented in avian and reptilian forms by the fibers of von Lenhossek. The cranial or bulbar portion of the accessory is a part of the primitive caudal visceral efferent column, which, on separation into a dorsal and ventral portion, leaves the general visceral efferent neurons of the accessory (with comparable preganglionic neurons of the vagus) in the dorsal efferent nucleus, while the special visceral efferent neurons accompany those of the vagus and glossopharyngeus to form the nucleus ambiguus.

The entire nuclear mass of the vagus has a dorsal position except in certain reptiles, birds, and mammals (figs. 294, 315). In birds one part of the nucleus has shifted in the direction of the hypoglossal nucleus and formed, in conjunction with the cells of origin of this last nerve, a special center, the nucleus intermedius. This latter nucleus appears to give origin to fibers which are distributed to the syrinx. Another more caudal portion of the avian nucleus lies relatively far ventralward and is representative of the mammaUan nucleus ambiguus. In mammals those cells of the vagus column which give rise to preganglionic fibers retain their position near the floor of the ventricle. According to Malone, these include preganglionic neurons for the innervation of the heart. According to Kosaka, the cell bodies of the neurons supplying the cardiac musculature have migrated ventralward with the neurons of the vagus which supply

G20 NERVOUS SYSTEMS OF '^TSRTEBRATES AND OF MAN

striated muscle of brancliiomeric origin. The cell bodies of such neurons of the vagus supplying branchiomeric muscle, together ^vith the cell bodies of the special visceral efferent neurons of the cranial accessory and the glossopharyngeal nerves.

D

I'Ki. .Tl.'j. Comparative lopoKrapliie dinplay of the motor nuclei and motor nerve roots, represent inj; grapliic rceonatructioiiB from sagittal M-etioiiK of the l)rnin stem. A, Pctromyzon; U, Scyllium; C, Tinea;

1), It-ana.

constitute tlic mammtilian nucleus ambiguus, which lies dorsal to the inferior olive and ventromcditil to the nucleus of the descending root of the trigeminal nerve.

THE EFFECTORY SYSTEM

621

The efferent center for the glossopharyngeal nerve has a dorsal position in cyclostomes and selachians and in many teleosts. In Petromyzon (fig. 294A) it lies in line with the vagus nucleus caudally and the facial nucleus frontally, but is

Fig. 315 {Continued). Comparative display of the motor nuclei and nerve roots, representing graphic reconstructions from sagittal sections of the brain stem. F, Varanus; F, Ciconia; G.Erinaceus; Homo.

separated from each by a short interval. In Bdellostoma (fig. 224B) it is continuous with the vagus column. In plagiostomes and ganoids and in many of the teleosts studied, the glossopharyngeal nucleus is in the caudal visceral

622 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

efferent column, between the facial and vagus centers (fig. 294B, C, fig. 315B, C, figs. 234, 239, 243). In other teleosts the glossopharyngeal (fig. 24.5) is separated from the vagus nucleus. In such cases it occupies a more or less ventral position in association with the efferent nucleus of the facial. In the tailed amphibians (fig. 294D) the glossopharyngeal nucleus has much the same relations as in selachians, but in Rana (fig. 294E) a glossopharyngeal nucleus distinct from both facial and vagus centers occurs, which lies practically in line with the latter center. In reptiles (fig. 294F, also fig. 259) two glossopharyngeal nuclei are present, a more dorsal nucleus, forming the cephalic end of the caudal visceral efferent column, and a more ventral nucleus, w'hich is independent in Caiman (fig. 259B) but fused with one of the facial nuclei in Crocodilus porosus (fig. 294F). In the birds studied, conditions comparable to those in Caiman are found. The major portion of the glossopharyngeal nucleus occupies a distinctly ventral position in mammals, forming the cephalic portion of the nucleus ambiguus (see page 595). A fewer number of somewhat smaller neurons of the glossopharyngeal nuclear complex retains a dorsal position as the nucleus salivatorius inferior, which supplies preganglionic fibers by way of the otic ganglion to the parotid gland. This nucleus usually lies in line with the nucleus salivatorius superior of the facial nerve.

The efferent nucleus of the facial has a dorsal position in cyclostomes. In Petromyzon (fig. 294A) it is an independent nucleus, slightly caudal to the level of emergence of its roots ; in Bdellostoma (fig. 224B) it lies at the level of the main root, fused with the trigeminal nucleus. A secondary facial root, associated ;vith the trigeminal, has been identified by Addcns (shown in fig. 224B), and represents in this latter form an excellent example of the condition which he has termed central anastomosis. In selachians, ganoids, certain teleosts, and urodele amphibians (fig 294A to D, also figs. 234, 239, 240, 256B) the facial nucleus forms the cephalic end of the caudal visceral efferent column. In certain fishes, such as Orthagoriscus (fig. 245C), the facial nucleus has a relatively large extent, reaching from a dorsal position into a ventral and caudal position, its ventral, caudal tip being fused with the glossopharyngeal nucleus. A somewhat similar condition is seen in Lophius (fig. 245B) except that a constriction of the gray mass is to be noted so that a separation into a ventral and a dorsal portion is foreshadowed here. In certain fishes two efferent facial nuclei have been described. Such are noted in Ceratodus (fig. 256A), where one group of facial neurons forms the cephalic tip of the visceral efferent column while the other occupies an independent position at the level of emergence of the facial root. In Gadus (fig. 245A) both such nuclei lie caudal to the plane of emergence of the facial fibers, and both are slightly ventral with reference to the vagus column. The more cephalic of the two nuclei is independent ; the more caudal is fused with the glossopharyngeal efferent nucleus. In Rana (fig. 294E) the facial nucleus is an independent nuclear group situated slightly caudal to the plane of emergence of the fibers. Several small nuclear groups associated with the facial root fibers have been seen in reptiles, particularly by Addens (see 1933 paper ; also the three facial nuclei in Caiman and the four in Crocodilus illustrated in Addens' figures, 259B

THE EFFECTORY SYSTEM

623

and 294F). These consist of a dorsal and of several ventral groups. The reptilian findings are repeated in a general way in birds, except that in the latter forms the nuclei lie at the level of entrance and more particularly in front of the level of emergence of the root fibers, while in the reptiles studied, the facial root leaves the brain stem in front of the plane of the nuclei. In crocodiles the most ventrally situated facial nucleus is fused with the ventral efferent nucleus of the glossopharyngeal ; in Sphenicus the most ventrally lying facial group is fused w’ith the ventral part of the trigeminal complex. The general visceral efferent preganglionic centers of the mammalian facial nerve retain their dorsal position near the ventricle, the respective neurons falling into one or more secondary groups (page 600). Of these the more generally recognized is the nucleus salivatorius superior, which supplies preganglionic fibers to the submaxillary ganglion, postganglionic fibers passing from there to submaxillary and sublingual glands.

Most of the variations in position of the abducens nucleus occur in forms below mammals (see fig. 315) , while its position in mammals is relatively constant. In cyclostomes (fig. 294A) the nucleus is forward, medial to the frontal end of the trigeminal nucleus (Addens). In selachians it is dorsal and more caudal, between the facial and glossopharjmgeal roots, and the long abducens nucleus of Ceratodus (fig. 256A) lies also between these roots but in a slightly more ventral position. In teleosts such as Gadus, Orthagoriscus, and Leuciscus (fig. 294, also fig. 245A, C) there are two ventrally situated abducens nuclei, one of which is in front of the other. Of these two gray masses, the more frontal nucleus lies near the level of emergence of the facial root, but in front of the facial nucleus. The more caudal abducens nucleus may lie in the plane of the facial nucleus or in front of it. The number of abducens nuclei may be increased as in Megalops (fig. 240D). In tailed amphibians such as Triton (fig. 256B) the abducens nucleus is dorsal, between the planes of emergence of the facial and glossopharyngeal nerves and in front of the facial nucleus, but a part of its rootlets emerge caudal to the glossopharyngeal nerve. In Rana (fig. 256C) there are two nuclei of the abducens nerve, which lie caudal to the facial nucleus, but in this form the root fibers emerge in front of the glossopharyngeal nerve. In reptiles the main abducens nucleus approaches the position characteristic for it in higher forms, lying at the level of the facial nuclei and slightly in front of them in the crocodile (fig. 294F), while it overlaps these nuclear masses caudofrontally in Caiman (fig. 259B). The accessory abducens nucleus lies ventrally in association with the facial nucleus. The rootlets, which are relatively numerous, emerge for the most part, in Caiman and Crocodilus, at a level between the planes of emergence of the facial and glossopharyngeal roots. In birds the chief abducens nucleus lies immediately caudal to the most dorsal trigeminal nucleus, which is relatively far forward in these forms. The accessory abducens nucleus occupies a ventral and, in many birds (as Callus and Cacatua, fig. 294G, fig. 271B and C), a slightly more caudal position, its caudal pole extending behind the plane of emergence of the facial root. The abducens rootlets, which may be numerous, leave the brain stem in an area which may extend forward to a plane passing through the caudal third of the trigeminal root and as far caudal (at least in Callus and Sphenicus) as a plane behind the facial

624 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

root. In mammals such as the rabbit (fig. 294H), the rootlets of the abducens have practically the same plane of emergence as has just been mentioned for birds. The abducens nucleus has an essentially similar position, lying caudal to the plane of the trigeminal nucleus. It overlaps the cephalic pole of the facial nucleus, which, however, is relatively farther forward in birds than in mammals. From the foregoing account it is evident that only in certain forms, among which are birds and mammals, does the abducens nerve form a true sixth nerve.

The efferent nucleus of the trigeminal (fig. 294), in nearly all the various classes of vertebrates, lies at the plane of entrance of the root fibers. The nuclear complex extends sometimes in front of (as in selachians) and sometimes behind such planes (as in Petromyzon). Usually a single nuclear complex has been recognized, although this may show some slight tendency toward secondary division ; but in certain teleosts two nuclei, and in certain reptiles (Caiman and Crocodilus, see fig. 2.59B and fig. 294F) and birds (such as Gallus, fig. 294G), three nuclei have been recognized, and even more groups are present in certain avian forms. The primitive position of the efferent trigeminal nucleus is very near the floor of the ventricle (Petromyzon and amphibians) ; in teleosts and reptiles it lies somewhat deeper and occupies a still more ventral position in many mammals (carnivores), although its dorsoventral position varies quite markedly in mammals with the form under consideration. It lies relatively near the ventricular floor in man.

The trochlear nucleus (fig. 294A) occupies primitively a dorsal position far behind the level of entrance of the oculomotor nerve and in the velum cerebelli in Petromyzon. In other vertebrates it lies below the floor of the fourth ventricle and in a more frontal position. Especially in amphibians, but also in some fishes and reptiles, there is still a considerable distance between it and the oculomotor nucleus, but in most vertebrates it lies either in the vicinity of the nucleus of the oculomotor or in direct continuation with it. The nucleus of the oculomotor (fig. 294) is situated usually in a dorsal position immediately below the ventricular space in the midbrain. In Petromyzon, in addition to the dorsal nucleus, there is found a group of cells situated far ventrally, in close relation to the root at its emergence, and, according to Addens, a caudal group in relation to the trigeminal nucleus.

A careful study of the migration of the efferent nuclei and a comparison with such other phylogenetic alterations in the medulla oblongata and the midbrain as have been described in the preceding pages indicate that alterations in the position of the efferent nuclei have as their cause the differentiation of the stimuli which effect them. Such alterations have their counterpart in the differences of development of corresponding dorsal roots and of reflex paths and centers connected with the efferent nuclei. These various factors are often supplemented by changes in the muscular system innervated. The nuclei tend to shift in the direction from which they receive the greatest number of their stimuli or their most important stimuli. One of the most striking examples of the migration of an efferent center, as this is brought about by changes in the effective stimuli and associated reflex patterns, is found in the phylogenetic hi.story of the hypoglossal complex. As long as no muscular tongue is present, corresponding musculature is found in the

626 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

which provides for common reflexes. The localization of the center for the innervation of heart musculature is still in dispute. The more definite localization of its position must be left for further verification. The ventral shifting of the glossopharyngeal cells supplying branchiomeric muscle is probably due to their tendency to migrate in the direction of the nucleus of the descending root of the trigeminal. The retention of a dorsal position of the nucleus salivatorius inferior of the glossopharyngeal is undoubtedly due to the importance of its connections with the gustatory centers of the medulla oblongata.

The caudal migration of the neurons constituting the main efferent nuclei of the facial undoubtedly is an expression of the tendency of this center to approach the gustatory centers. This migration is least marked in those animals in which gustatory sensation is poorly' developed, as in cyclostomes and especially in birds ; in fact, in the latter forms the facial may be in part fused with the trigeminal nucleus. The nucleus of the descending root of the trigeminal and its chief sensory nucleus exert a most important influence on the dorsoventral position of the facial nucleus. The skin area innervated by the trigeminal for its greater extent overlies the musculature innervated by the facial nerve, and with the great development of this musculature in mammals, there is every incentive for a close interrelation centrally between the nucleus of termination of incoming cutaneous sensory fibers and the cells of origin for outgoing motor fibers. Consequently in these forms it is not surprising that the majority of the neurons giving rise to facial root fibers ndgrate ventralward and, in the highest manunals, lateralward, between the superior olive and the nucleus of the descending root of the trigeminal, with which centers they are in synaptic relation. Only the preganglionic centers of the facial retain a relatively dorsal position. To such preganglionic centers belongs the nucleus salivatorius superior, which supplies preganglionic fibers to the submaxillary ganglion and the ganglion of Langley, and then impulses through postganglionic fibers to the sublingual and submaxillary saUvary glands, the functional activity of which is in intimate relation with the gustatory centers.

The efferent trigeminal nucleus has primitively a position in the dorsal part of the brain stem, close to the ventricular floor. In certain higher forms, including many manunals, it migrates away from the ventricle and takes up a position approximately midway between the dorsal and ventral brain surfaces, or even ventral to this position. This migration is believed to be due to the influence of impulses from the sensory trigeminal nuclei, especially from the upper two-thirds of the sensory gray, which receives sensory fibers from the maxillary and mandibular roots of the trigeminal nerve, peripherally supplying the jaw and mouth regions. The ventral shifting of the efferent trigeminal nucleus appears to be most marked in those forms in which the jaw muscles are developed especially well. The efferent trigeminal nucleus of man has a relatively dorsal position, not far from the floor of the ventricle. The frontal extension of the efferent nucleus of the trigeminal nerve is influenced, probably, by the importance of impulses reaching it over the collaterals (or stem fiber neuraxes) of mesencephalic root fibers.

THE EFFECTORY SYSTEM

G27

The positions of the nuclei of origin of the eye muscle nerves are determined by the interconnections of these nuclei through the medial longitudinal fasciculus and by their relations to impulses from the vestibular centers. Frequently these nuclei either lie very close to each other or are actuall 3 ' continuous, the trochlear nucleus being fused with the dorsal (or dorsolateral) nucleus of the oculomotor. Both the trochlear and the oculomotor nuclei lie in very close relation to the medial longitudinal fasciculus, through which they are interrelated with other efferent nuclei of the brain stem (particularly \rith the nucleus of the abducens nerve) and wath the vestibular centers. Through the medial longitudinal fasciculus the trochlear nucleus receives, directly or indirectly, optic impulses, under the influence of which the nucleus probablj' moves frontalward. Wiile (with a few exceptions) the oculomotor nuclei do not show much shifting in position frontocaudally, there are to be noted variations in a dorsoventral direction. Thus in cyclostomes part of the somatic efferent neurons have a dorsal position, while others he ventrolly, near the region of emergence of the rool.s. In selachians, where the medial lonptudinal fasciculus is very well developed, the whole oculomotor nuclear mass occupies a dorsal position. The ventral position of the ventral tecto-bulbar tracts affects the relativelj' ventral position of the oculomotor, but the degree of development of the.se tracts, and hence the influence exerted by them, varies in different forms. Probably the medial longitudinal fasciculus is the most determinative factor in the establi.'jhmcnt of the position of the oculomotor nuclear complex. According to Ariens Kappers, the various positions of the smaller nuclei making up the oculomotor nuclear group can be explained on the basis of their functional connections.

It is evident, then, that the arrangement of the principal effectorj’ centers is dependent on the influence exerted by stimulations of different origin, the ma.vimal stimuli serving as tropisms leading to the shifting of the various groups. It is evident that an increase in the development of a certain afferent tract docs not lead to the migration of all efferent nuclei into the region of that tract. The selectivity shown is determined by the corrclativ'e relation between certain sensory patlis and certain effcctory centers. This correlation is an evidence of the working of the principal law of ncurobiotaxis, the action of simultaneously correlated stimuli. Thus an increase in the medial longitudinal fasciculus lends to a migration of the neurons of the abducens nucleus into its vicinity, since the impulses which it carries are of major importance in the functioning of the abducens nerv'c. That impulses over the medial longitudinal f.a=ciculus do not affect the nucleus of the facial nerve in a similar m.'inner is due to the fact that the impulses carried bj' this Fasciculus are not related primarily to the functional nctiv'itj' of the facial nerve. A conv'crse relation is found in the visceral tract.=, which influence markcdlv'’ the position of the facial nuclei but do not affect the position of the cj-e muscle nuclei. Thas the primary law of neurobiot.axis (.as evidenced by the descriptions in this chapter) is cxprcs.=cd best by saying that the position of a nucleus is determined by those fiber systems which arc correlntetl with it and over which the major number of stimuli arc conducted to it. This matter is discussed further in Chapter 1, pp. 73 to 92.

628 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

BIBLIOGRAPHY

CYCLOSTOMES

Addens, J. L. 1928. The eye muscle nerves of petromyzonts, especially in their general morphological Significance. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 31, p. 1.

. 1933 . The motor nuclei and roots of the cranial and first spinal nerves of vertebrates. Zeitschr. f. Anat. u. Entwicklungsgeschichte, I. Abteil. d. Zeitschr. f. ges. Anat., Bd. 101, S. 307.

Ahlborn, F. 1883. XJntersuchungen fiber das Gehim der Petromyzonten. Zeitschr, f. wissensch. Zool., Bd. 39, S. 191.

. 1884. Ueber den Ursprung und Austritt der Hirnnerven von Petromyzon.

Zeitschr. f. wissensch. Zool., Bd. 40, S. 286.

Ahiens Kappers, C. U. 1908. Weitere Mitteilungen fiber Neurobiotaxis. II. Die phylogenetische Entwickelung des horizontalen Schenkels des Facialiswurzelknies. Folia neuro-biol., Bd. 2, S. 255.

. 1910. Idem. V. The migrations of the motor cells of the bulbar trigeminus,

abducens and facialis in the series of vertebrates and the differences in the course of their root-fibers. Verhandel. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, Deel 16, no. 4.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wur zelkerne in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-Erganzungs-Heft, S. 1.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

Ayers, H., and Worthington, J. 1907. The skin end-organs of the trigeminus and lateralis nerves of Bdellostoma dombeyi. Am. J. Anat., vol. 7, p. 327.

. 1908. The finer anatomy of the brain of Bdellostoma dombeyi. I. The acus tico-lateral system. Am. J. Anat., vol. 8, p. 1.

. 1911. Idem. II. The fasciculus communis system. J. Comp. Neurol.,

vol. 21, p. 593.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny : A study of the phenomena of neurobiotaxis. Part I. Cyclostomi and Pisces. J. Comp. Neurol., vol. 27, p. 467.

Butschli, 0. 1912. Vorlesungen fiber vergleichende Anatomie. W, Engelmann, Leipzig.

Conel, j. LeR. 1929. The development of the brain of Bdellostoma stouti. I. External growth changes. J. Comp. Neurol., vol. 47, p. 343.

. 1931. The development of the brain of Bdellostoma stouti. II. Internal

growth changes. J. Comp. Neurol., vol. 52, p. 365.

Cords, E. 1924. Zur Frage des M. retractor bulbi der Sauger. Zeitsch. f. Anat. u. Entwicklungsgesch., Bd. 71, S. 240.

Corning, H. K. 1902. Ueber die vergleichende Anatomie der Augenmuskulatur. Morphol. Jahrb., Bd. 29, S. 94.

Edinger, L. 1906. Ueber das Gehim von Myxine glutinosa. Abhandl. d. kon. preuss. Akad. d. Wissensch., Berlin.

FiiUBRiNGER, P. 1875. Untersuchungen zur vergleichenden Anatomie der Muskulatur des Kopfskelets der Cyclostomen. Jenaische Zeitschr. f. Naturw., Bd. 9, S. 1.

Furbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologic. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

THE EPFECTORY SYSTEM

G29

Holm, J. F. 1901. The finer anatomy of the nervous system of Myxine glutinosa. Morphol. Jahrb., Bd. 29, S. 364.

Huet, W. G. 1911. Notes on the trochlear and oculomotor nuclei and the trochlear root in the lower vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pt. 2, p. 897.

Jansen, J. 1930. The brain of Myxine glutinosa. J. Comp. Neurol., vol. 49, p. 359.

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Neurol., vol. 12, p. 1.

. 1908. Note on the presence or absence of the glossopharyngeal nerve in myxi noids. Anat. Rec., vol. 2, p. 233.

VON Kupffeu, C. 1906. Die Morphogcnic des Centralnervensystems. 0. Hertw'ig’s Handbuch dcr vcrgleichenden und experimentellen Entwickelungslehre der Wirbeltiere. G. Fischer, Jena.

Neal, H. V. 1897. The development of the hypoglossus musculature in Petromyzon and Squalus. Anat. Anz., Bd. 13, S. 441.

Nisni, S. 1922. Beitragc zur vcrgleichenden Anatomic der Augenmuskulatur. Arb. a. d. anat. Inst. d. K.-japon. Univ. Sendai, Bd. 7, S. 65.

Retzius, G. 1893. Das Gehirn und das Auge von Myxine. Biol. Untersuch., N. F., Bd. 5, S. 55.

Rothig, P. 1913. Contributions upon neurobiotaxis. (The arrangement of the motor nuclei in Myxine glutinosa, in Cryptobranchus, Necturus, Rana fusca and Bufo.) Kon. Akad. v. Wetensch. tc Amsterdam, Proc. sect, sc., vol. 16, p. 296.

Rothig, P., and Ahiens Kappehs, C. U. 1914. Further contributions to our knowledge of the brain of Myxine glutinosa. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 17, pt. 1, p. 2.

Sandehs, a. 1894. Researches in the ner\’ous system of Myxine glutinosa. London.

Schilling, K. 1907. Ueber das Gehirn von Petromyzon fluviatilis. Abhandl. d. Senckenb. nat. Gesellsch., Frankfurt am Main, Bd. 30, p. 425.

Steuzi, G. 1909. II sistema nervoso centrale dei vertebrati. Vol. I, Ciclostomi. A. Draghi, Padova.

Thetjakoff, D. 1909. Das Nervensystem von Ammocoetes. II. Gehirn. Arch, f. mikr. Anat., Bd. 74, S. 636.

. 1926. Das Skelett und die Muskulatur im Kopfe des Fluszneunauges. Zeitschr.

f. wisscnsch. Zool., Bd. 128, S. 267.

. 1927. Das periphere Nervensystem des Fluszneunauges. Zeitschr. f. wissensch.

Zool., Bd. 129, S. 358.

. 1929. Der Oculomotorius vom Neunauge. Anat. Anz., Bd. 68, S. 225.

Tsuchida, U. 1906. Ueber die Ursprungskeme der Augenbewegungsnerven und liber die mit diesen in Beziehung stehenden Bahnen im Mittel- und Zwischenhirn. Arb. a. d. hirnanat. Inst, in Zurich, Bd. 2, S. 1.

VAN Valkenbuhg, C. T. 1912. On the splitting of the nucleus trochlearis. Kon. Akad. V. Wetensch. te Amsterdam, Proc. sect, sc., vol. 14, p. 1023.

Worthington, J. 1905. Contribution to our knowledge of the myxinoids. Am. Naturalist, vol. 39, p. 625.

. 1905a. The descriptive anatomy of the brain and cranial nerves of Bdellostoma

dombeyi. Quart. J. Micr. Sc., vol. 49, p. 137.

plagiostomes

AniiiNs ILappers, C. U. 1906. The structure of the teleostean and selachian brain.

J. Comp. Neurol., vol. 16, p. 1.

. 1908. Weitere Mitteilungen iiber Neurobiotaxis. II. Die phylogenetische

Entwickelung des horizontalen Schenkels des Facialiswurzelknies. Folia neurobiol., Bd. 2, S. 255.

630 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Aeiens ICappers, C. U. 1910. Idem. V. The migrations of the motor cells of the bulbar trigeminus, abducens and facialis in the series of vertebrates, and the differences in the course of their root-fibers. Verhandl. d. kon. Akad. v. Wetcnsch. to Amsterdam, Tweede Sectie, Deel 16, no. 4, S. 1.

. 1911. Idem. VI. The migrations of the motor root-cells of the vagus group,

and the phylogenetic differentiation of the hypoglossus nucleus from the spinooccipital-system. Psychiat. en neurol. Bl., Amsterdam, vol. 15, p. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wurzcl kerne in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-ErgiinzungsHeft, S. 1.

. 1912a. The arrangement of the motor nuclei in Chimaera monstrosa compared

with other fishes. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 14, pt. 2, p. 1176.

. 1914. Weitere Mitteilungen fiber Neurobiotaxis. VIII. Ueber den motorischen Glossopharyngeus und Facialis bei niedcren Vertebraten. Folia neurobiol., Bd. 8, S. 383.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny. A study of the phenomena of neurobiotaxis. Part I. Cyclostomi and Pisces. J. Comp. Neurol., vol. 27, p. 467.

Butschli, 0. 1912. Vorlesungen fiber vergleichende Anatomic. W. Engelmann, Leipzig.

Edgeworth, F. H. 1911. On the morphology of the cranial muscles in some vertebrates. Quart. J. Micr. Sc., vol. 56, p. 167.

Furbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologic. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

Gegenbaur, C. 1898. Vergleichende Anatomic der Wirbelthiere mit Berucksichtigung der Wirbellosen. W. Engelmann, Leipzig. Bd. 1.

Hoffmann, C. K. 1894. Zur Entwicklungsgcschichtc des Selachier Kopfes. Anat. Anz. Bd. 9, p. 638.

VAN DER Horst, C. J. 1918. Die motorischen Kerne und Bahnen in dem Gehirn der Fische, ihr taxonomischer Wert und ihre neurobiotactische Bedcutung. Tijdschr. d. Ned. Dierk. Vereen., Deel 16, no. 2, p. 168.

Neal, H. V. 1897. The development of the hypoglossus musculature in Petromyzon and Squalus. Anat. Anz., Bd. 13, S. 441.

. 1914. The morphology of the eye-muscle nerves. Tufts College Studies, vol. 3,

no. 4. J. Morphol., vol. 25, p. 1.

Platt, J. B. 1891. Further contributions to the morphology of the vertebrate head. Anat. Anz. Bd. 6, p. 251.

Sterzi, G. 1909-12. II sistema nervoso centrale dei vertebrati. Vol. 2, Pcsci, Libro I, Sclaci. A. Draghi, Padova.

Tiestng, B. 1896. Ein Beitrag zur Kenntnis der Augen-, Kiefer-, und Kiemeninuskulatur der Haie und Rochen. Jenaischc Zcitschr. f. Naturw., Bd. 30, S. 75.

Vetter, B. 1874. Untersuchungen zur vcrgleichenden Anatomic der Kicmen- und Kicfermusculatur dor Fische. I. Thcil. Jenaischc Zcitschr. f. Naturw., Bd. 8, S. 405.

. 1878. Idem., II. Thcil. Jenaischc Zcitschr. f. Naturw., Bd. 12, S. 431.

VAN WijHE, J. W. 1882. Ueber die Mesodermsegmente und die Entwickelung der Nerven des Selachicrkopfcs. J. Muller, Amsterdam. Verhandl. d. kon. Akad. v. Wetcnsch. tc Amsterdam, vol. 22, 1883.

GANOIDS AND TELEOSTS

Addens, j. L. Sec bibliography for cyclostomes.

Agassiz, L. J. R., AND Vogt, C. 1845. Anatomic des Salmones. H. Wolfrath, Neuchdtel.

THE EFFECTORY SYSTEM

631

Allis, E. P. 1897. The cranial muscles and cranial and first spinal nerves in Amia calva. J. Alorphol., vol. 12, p. 487.

. 1903. The skull, and the cranial and first spinal muscles and nerves in Scomber

scomber. J. Morphol., vol. 18, p. 45.

AniENs ICappers, C. U. 1906. The structure of the teleostean and selachian brain. J. Comp. Neurol., vol. 16, p. 1.

. 1909. Weitere Mitteilungen iiber Neurobiotaxis. III. Ueber den Einfluss

der Neurone der Geschmackskeme auf den motorischen Facialis- und Glossopharyngeuskem und ihr Verhalten zur Radix descendens nervi quinti. Folia neurobiol., Bd. 3, S. 259.

. 1910. Idem. IV. The migrations of the abducens-nucleus and the con comitating changes of its root-fibers. Psychiat. en neurol. Bl., Amsterdam, vol. 14, p. 271.

. 1910a. Idem. V. The migrations of the motor cells of the bulbar trigeminus,

abducens and facialis in the series of vertebrates, and the differences in the course of their root-fibers. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Seclie, Deel 16, no. 4, S. 1.

. 1911. Idem. VI. The migrations of the motor root-cells of the vagus group,

and the phylogenetic differentiation of the hypoglossus nucleus from the spinooccipital system. Psychiat. on neurol. Bl., Amsterdam, vol. 15, p. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wur zelkeme in Oblongata und Mittclhirn. Folia neuro-biol., Bd. 6, Sommer-Erganzungs-Hcft, S. 1.

. 1914. Idem. tTII. Ueber die motorischen Glossopharyugeus und Facialis

bci niedern Vertebraten. Folia neuro-biol., Bd. 9, S. 383.

. 1920-21. See bibliography for cyclostomes.

Beccari, N. 1912. Sopra alcuni rapporti del fascicule longitudinale posteriore con i nuclei di origine del nervo III e IV nei Teleostei. Monitore zool. ital., 1912.

. 1923. Intomo al primo differenziamento dei nuclei motori dei nervi cranici.

Monitore zool. ital., vol. 34, p. 161.

Berkelbach van ber Sprenkel, H. 1915. The central relations of the cranial nerves in Silurus glanis and Mormyrus caschive. J. Comp. Neurol., vol. 25, p. 5.

Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny : A study of the phenomena of neurobiotaxis. Part I. Cyclostomi and Pisces. J. Comp. Neurol., vol. 27, p. 467.

Burr, H. S. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol., vol. 45, p. 33.

Cunningham, J. T. 1890. A treatise on the common sole considered both as an organism and as a commodity. Plymouth (England).

Dietz, P. A. 1912. Vergelijkende Anatomic van de Kaak- en Kieuwboogspieren der Teleostei. Inaugural Dissertation, Leiden.

Droogleever, FoRTunr, A. B. 1912. Notiz fiber den Eintritt der motorischen Nervenwurzeln in die Medulla oblongata und fiber die Lage der motorischen Kerne bei Amia calva L. Folia neuro-biol., Bd. 6, S. 27.

Edgeworth, F. H. 1911. On the morphology of the cranial muscles in some vertebrates. Quart. J. Micr. Sc., vol. 56, p. 167.

Franz, V. 1911. Das intracraniale und intracerebrale Verhalten des Nervus trochlearis bei den Knochenfischen. Anat. Anz., Bd. 38, S. 592.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Berficksichtigung der Wirbellosen. W. Engelmann, Leipzig.

Herrick, C. J. 1899. The cranial and first spinal nerves of Menidia: A contribution upon the nerve components of the bony fishes. J. Comp. Neurol., vol. 9, p. 153.

632 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Hocke Hoogenboom, K. J. 1929. Das Gehim von Polydon folium Lacep. Jahrb. f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 311.

Holmgren, N., and van der Horst, C. J. 1925. Contribution to the morphology of the brain of Ceratodus. Acta Zook, Bd. 6, S. 59.

VAN DER Horst, C. J. 1918. Die motorischen Kerne und Bahnen in dem Gehirn der Fische, ihr taxonomischer Wert und ihre neurobiotaktische Bedeutung. Tijdschr. d. Ned. Dierk. Vereen., vol. 16, no. 2, p. 168.

Huet, W. G. 1911. Notes on the trochlear and oculomotor nuclei and the trochlear root in the lower vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pt. 2, p. 897.

JuGE, M. 1899. Recherches sur les nerfs c6r4braux de la musculature c6phalique de Silurus glanis. Rev. suisse de zook, vol. 6, fasc. 1, p. 1.

DE Lange, S. J. 1910. The descending tracts of the corpora quadrigemina. Folia neuro-biok, Bd. 3, S. 633.

McMurrich, j. P. 1884. The myology of Ameiurus catus (L.) with a consideration of the relations of the postoccipital and hypoglossal nerves in various vertebrate groups. Proc. Canadian Inst., Toronto, N. S., vol. 2, no. 3, p. 311.

. 1885. The cranial muscles of Amia calva. Johns Hopkins Univ., Studies Biol.

Lab., vol. 3, no. 3, p. 121.

Mayser, P. 1881. Vergleichend anatomische Studien tiber das Gehim der Knochenfische mit besonderer Beriicksichtigung der Cyprinoiden. Zeitschr. f. wissensch. Zook, Bd. 36, S. 259.

Schreiner, K. E. 1902. Einige Ergebnisse iiber den Bau und die Entwicklung der Occipitalregion von Amia und Lepidosteus. Zeitschr. f. wissensch. Zook, Bd. 72, S. 467.

VAN DER Sprenkel. See Berkelbach van der Sprenkek

Stannius, H. 1849. Das peripherische Nervensystem der Fische, anatomisch und physiologisch untersucht. Stiller, Rostock.

. 1854. Zootomie der Fische. Handbuch der Anatomic der Wirbelthiere. 2** Aufl., Heft 1, Veit u. Co., Berlin.

Tello, F. 1909. ContribuciOn al conocimiento der encfifalo de los Teleosteos. Trab. d. lab. de invest, biok Univ. de Madrid, vol. 7, p. 1.

Theunissen, F. 1914. The arrangement of the motor roots and nuclei in the brain of Acipenser ruthenus and Lepidosteus osseus. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 16, pt. 2, p. 1032.

Thilo, 0. 1901. Die Vorfahren der Schollen. Bulk Imptir. Acad. Sc. St. Petersbourg, 5 S(5r., vol. 14, p. 315.

. 1902. Die Umbildungen am Knochengeruste der Schollen. Zook Anz., Bd. 25,

S. 305.

\ alenciennes, a., AND CuviER, G. 1828-1849. Histoire naturelle des poissons, Paris.

\ ETTER, B. 1874. Untersuchungen zur vergleichenden Anatomie der Kiemen- und Kiefermusculatur der Fische. I. Theik Jenaische Zeitschr. f. Naturw., Bd. 8, S. 405.

. 1878. Idem. II. Theik Jenaische Zeitschr. f. Naturw., Bd. 12, S. 431. allenberg, a. 1907. Beitrage zur Kenntnis des Gehims der Teleostier und Selachier. Anat. Anz., Bd. 31, S. 369.

AMPHIBIANS

Addens, j. L. 1933. See bibliography for cyclostomes

AriKns Kappers C. U. 1910. Weitere Mittcilungen iiber Neurobiotaxis. IV. The migrations of the abducens nucleus and the concomitating changes of its root fibers.

1 .sychiat. cn neurok Bk, Amsterdam, vol. 14, p. 271.

THE EFFECTORY SYSTEM

633

ARiiiNS ICappers, C. U. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wurzelkerne in Oblongata und Mittelhirn. Folia neuro-biol., Bd. 6, Sommer-Erganzungs-Heft, S. 1.

. 1920-21. Vergleichende Anatomie des Nen'ensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

Black, D. 1917, The motor nuclei of the cerebral nerves in phylogeny ; A study of the phenomena of neurobiota.xis. II. Amphibia. J. Comp. Neurol., vol. 28, p. 379.

Bruner, H. L. 1899. Description of new facial muscles in Anura, with new observations on the nasal muscles of Salamandridae. Anat. Anz,, Bd. 15, S. 411.

Butschli, 0. 1912. Vorlesungen fiber vergleichende Anatomie. W. Engelmann, Leipzig.

CoGHiLL, G. 1902. The cranial ner\'es of Amblystoma tigrinum. J. Comp. Neurol., vol. 12, p. 205.

. 1906. The cranial nen'cs of Triton taeniatus. J. Comp. Neurol., vol. 16, p. 247.

Druner, L. 1901. Studien zur Anatomie der Zungenbein-, Kiemenbogen- und Kehlkopfmuskeln der Urodelen. I. Theil. Zool. Jahrb., Abt. f. Anat., Bd. 15, S., 435.

. 1904. Idem. II. Theil. Zool. Jahrb., Abt. f. Anat., Bd. 19, S. 361.

Edgeworth, F. H. 1911. On the morphology of the cranial muscles in some vertebrates. Quart. J. Micr. Sc., vol. 56, p. 167.

Furbringer, M. 1897. Ueber die spino-occipitalen Nerven der Selachier und Holocephalen und ihre vergleichende Morphologic. Festschr. f. Carl Gegenbaur, Bd. 3, S. 349. W. Engelmann, Leipzig.

Gage, S. P. 1893. The brain of Diemyctylus viridescens from larval to adult life, and comparisons with the brain of Amia and Petromyzon. Wilder Quarter Century Book, p. 259.

Gaupp, E. 1899. Anatomie des Frosches, by Ecker, A., Wiedersheim, R., and Gaupp, E. 2 Aufl., F. Vieweg u. Sohn, Braunschweig.

Gegenbaur, C. 1898. Vergleichende Anatomie der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

Herrick, C. J. 1914. The medulla oblongata of larval Amblystoma. J. Comp. Neurol., vol. 24, p. 343.

. 1930. The medulla oblongata of Necturus. J. Comp. Neurol., vol. 50, p. 1.

VAN DER Horst, C. J. 1918. Die motorischen Kerne und Bahnen in dem Gehim der Fische, ihr taxonomischer Wert und ihre neurobiotactische Bedeutung. Tijdschr. d. Ned. Dierk. Vereen., Deel 16, no. 2, p. 168.

Kingsbury, B. F. 1895. On the brain of Necturus maculata. J. Comp. Neurol., vol. 5, p. 139.

McIGbben, P. 1913. The eye muscle nerves in Necturus. J. Comp. Neurol., vol. 23, p. 153.

UE Man. 1873. Vergelijkende myologische en neurologische Studien over Amphibien en Vogels. Leiden.

Norris, H. W. 1913. The cranial nerves of Siren lacertina. J. MorphoL, vol. 24, p. 245.

Norris, H. W., and Buckley, M. 1911. The peripheral distribution of the cranial nerves of Necturus maculata. Proc. Iowa Acad. Sc., vol. 18, p. 131.

VON Plessen, and Rabinovicz. 1891. Die Kopfnerven von Salamandra maculata im vorgeriickten Embryonalstadium. Miinchen.

Rothig, P. 1913. Contributions upon neurobiotaxis. (The arrangement of the motor nuclei in Myxine glutinosa, in Cryptobranchus, Necturus, Rana fusca, and Bufo.) Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., yol. 16, p. 296.

Strong, O. S. 1895. The cranial nerves of Amphibia. J. MorphoL, vol. 10,

P- 101.

Wiedersheim, R. 1875. Salamandrina perspicillata und Geotnton fuscus. Genua.

634 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

REPTILES

Addens, J. L. 1933. See bibliography for cyclostomes.

Ariens Kappers, C. U. 1910. Weitere Mitteilungen uber Neurobiotaxis. IV. The migrations of the abducens nucleus and the concomitating changes of its root-fibers. Psychiat. en neurol. Bl., Amsterdam, vol. 14, p. 271.

. 1910. Idem. V. The migrations of the motor cells of the bulbar trigeminus, abducens and facialis in the series of vertebrates, and the differences in the course of their root-fibers. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, Deel 16, no. 4, S. 1.

. 1911. Idem. VI. The migrations of the motor root-cells of the vagus group, and the phylogenetic differentiation of the hypoglossus nucleus from the spinooccipital system. Psychiat. en neurol. Bl., Amsterdam, vol. 15, p. 408.

1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wurzelkerne in Oblongata und Mittelhim. Folia neuro-biol., Bd. 6, Sommer-ErganzungsHeft, S. 1.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of present text.) E. F. Bohn, Haarlem.

Beccari, N. 1913. Sulla spettanza delle fibre del Lenhoss^k al sistema del nervo accessono e contribute alia morfologia di questo nervo (Osservazioni in Lacerta muralis). Arch. ital. di anat. e di embriol., vol. 11, p. 299.

j- T ® nervi cranici e i nervi cervicali negli embrioni

di Lacerta muralis. Arch. ital. di anat. e di embriol., vol 13 p 1

• comparativi sulla struttura del Rhombencefalo ’ I Nervi spino eTritvdTp/S^"' Arch. ital. di anat. "edi

°e^es in phylogeny ; A study of Bo^s J-Comp.Neurol.:vol.32,p. 61.

mentatus est. J Zawidzkt”vihSe° Europeae, indagavit, depivxit, com iTcertnia^' ZoT JahTb! ASr^AnaT.? movements of the skull in

ATa/d°W?sTpn;.y,^^^^ Chamaeleonen. Sitzungsb. d. K.

Akad. d Wissensch. zu Wien, Math.-nat. CL, Bd. 8 S 65

^^^l^ipzig.*^' liber vergleichende Anatomie. W. Engelmann,

^“"befRe^LTen.^^LrSrjtLb^S^I.'^S.^”'’^

^'"'x'de l^naf e?phJjoL,^vor3r Camdldon.

Quart. L Micr. Sc.,

^™rrd:?schuiteZskefm^;s^^^

^^hHaddpWaS'Londo^^^ neurology. 5 ‘h edition, W. B. Saunders Co.,

differenrUrtebrate?’ Korn Ak^rv’^'CtcMch'^e^A" ?”"] *P

vol. 13, pt. 2, p 1047 "cienscn. te Amsterdam. Proc, sect, sc.,

TTiier-Reicl^^ C^F^ WintP^'^T^" S’ ^“"n’s Klassen und Ordnungen dcs

nier neicns. c. 1- . Winter, Leipzig and Heidelberg, 1889.

THE EFFECTORY SYSTEM

635

Hobeh, G. Cabb, and CnosDY, E. C. 1926. Sec bibliography for reptiles, Chapter IV.

Kiluan, G. 1890. Die Ohrmuskcln dor Krokodilcs, ncbst vorlaufigen Bemerkungen liber die Homologie des Musculus stapedius und dcs Stapes. Jenaische Zeitschr f. Natun\-., Bd. 24, S. 632.

DE Lange, S. J. 1917. D.as Hinterhim, das Nachhim und das Ruckenmark der Repti• lien. Folia ncuro-biol., Bd. 10, S. 385.

Mn-AUT, St. G. 1867. Notes on the mj’ology of Iguana tuberculata. Proc. Zool. Soc., London, p. 766.

. 1869. Notes on the myologj’ of Menopoma alleghaniense. Proc. Zool. Soc.,

London, p. 254.

. 1869. Notes on the myology of Mcnobranchus lateralis. Proc. Zool. Soc.,

London, p. 450.

. 1870. Notes on the myology of Chamaclcon parsonii. Proc. Zool. Soc., London,

p. 850.

OsAWA, G. 1898. Bcitn'igc zur dcr Hatteria punctata. Arch. f. mikr. Anat., Bd. 51, S. 481.

Ranson, S. W. 1931. The anatomy of the ncr\-oussj'stem. 4"’ edition, W. B. Saunders Co., Philadelphia and London.

Ruge, G. 1897. Ueber das pcriphcrische Gebiet des Neiams facialis bei Wirbelthieren. Fest.schr. f. Gegenbaur, Bd. 3, S. 193. W. Engclmann, Leipzig.

Terni, T. 1922, 1922a. Riccrchc sul nervo abducente c in special modo intomo al significato del suo nuclco acccssorio d’origine. Folia neuro-biol., vol. 12, p. 277. 1922a, Arch. Fisiol., vol. 20, p. 305.

Tege, H. 1932. Somatic motor mechanisms in the midbrain and medulla oblongata of Qii^’semj’s elegans (Weed). J. Comp. Neurol., vol. 55, p. 185.

Verslevs, j. 1904. Ueber ILaumuskcln bei Lacertilia. Anat. Anz., Bd. 24, S. 641.

Waltoer, F. 1887. Das Visccralskelctt und seine Muskulatur bei den einheimischen Amphibicn und Rcptilicn. Jenaische Zeitschr. f. Naturw., Bd. 21, S. 1.

nmos

Addens, j. L. 1933, 1934. See bibliography for mammals.

AriKns Kappers, C. U. 1908. Phylogenetische Verlagerung der motorischen Oblongata-Kemc, ihre Ursache und Bedcutung. Folia neuro-biol., Bd. 1, S. 301.

-. 1910. ‘Weitere Mitteilungen uber Neurobiotaxis. V. The migrations of the

niotor cells of the bulbar trigeminus, abducens and facialis in the series of vertebrates, and the differences in the course of their root-6bers. Verhandl. d. kon. Akad. v. Wetcnsch. tc Amsterdam, Tweede Scctie, Dcel 16, no 4, S. 1.

. 1911. Idem. VI. The migrations of the motor root-cells of the vagus group,

and the phylogenetic differentiation of the hjTJOglossus nucleus from the spinooccipital system. Psychiat. en neurol. Bl., Amsterdam, vol. 15, p. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wur zelkeme in Oblongata and Mittelhim. Folia neuro-biol., Bd. 6, Sommer-Erganzungs-Heft, S. 1.

• 1913. Phenomena of neurobiotaxis in the central nerv'ous system. Transactions of the XVTI International Congress of Medicine, London.

. 1920-21. Vergleichende Anatomic des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

Beccari, N. 1913. Sulla spettanza delle fibre del Lenhoss6k al sistema del nervo acccssorio e contributo alia morfologia di questo nenm (Osservazioni in Lacerta muralis). Arch. ital. di anat. e di embriol., vol. 11, p. 299.

■ . 1922. Studi comparativi suUa struttura del rhombencefalo. Arch. ital. di anat.

e di embriol., vol. 19, p. 122.

636 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Biondi, G. 1913. Osservazioni suUo Bvilupo e sulla struttura dei nuclei d’origine du nervi oculomotore e trocleare nel polio. Riv. ital. di neuropat., psichiat. ed el ettrot., vol. 3, p. 302.

Black, D. 1922. The motor nuclei of the cranial nerves in phylogeny. A study of the phenomena of neurobiotaxis. IV. Aves. J. Comp. Neurol., vol. 34, p. 233. Bok, S. T. 1915. Die Entwicklung der Himnerven und ihrer zentralen Bahnen. Die stimulogene Fibrillation. FoHa neuro-bioL, Bd. 9, S. 475.

Brandis, F. 1893. Untersuchungen fiber das Gehim der Vogel. I. Theil : Uebergangsgebiet vom Ruckenmark zur Medulla oblongata. Arch. f. mikr. Anat., Bd. 41, S. 168.

. 1893a. Idem. II. Theil : Ursprung der Nerven der Medulla oblongata. Arch, f. mikr. Anat., Bd. 41, S. 623.

. 1895. Idem. III. Der Ursprung des N. trigeminus und der Augenmuskelnerven. Arch. f. mikr. Anat., Bd. 44, S. 534.

Brouwer, B. 1918. Klinisch-anatomische Untersuchungen fiber den Oculomotoriuskem. Zeitschr. f. d. ges. Neurol, u. Psychiat., Bd. 40, S. 152.

Carpenter, F. W. 1911. The cihary ganghon of birds. Folia neuro-bioL, Bd. 5, S. 738.

Craigie, E. H. 1928. Observations on the brain of the humming bird (Chiysolampis mosquitus Linn, and Chlorostilbon caribaeus Lawr.). J. Comp. Neurol., vol. 45, p. 377.

Gad^, H 1889. H. G. Bronn’s Klassen und Ordnungen des Thier-Reichs.

C. F. Winter, Leipzig u. Heidelberg.

VAN Gehuchtcn, a. 1893. Le syst^me nerveux de I’homme.

Groebbem, F. 1922. Der Hypoglossal der Vogel. Zool. Jahrb., Abt. f. Anat., i5a, 43, D. 465.

. 1924 Untersuchungen fiber den Thalamus und das Mittelhim der Vogel. Anat. Anz., Bd. 57, S. 385.

Jelgersma, G. 1896. I^ verbindingen van de groote hersenen bij der vogels met de oculomotonuskem. Feestbundel uitgegeven doore Nederlandsche Vereeniging von Psychiatne, vol. 49, p. 241.

Faciahskeme des Huhnes. Jahrb. f.

Psychiat. u. Neurol., Bd. 25, S. 57.

Kosa^, K., jwd Yagita, K. 1903. Experimentelle Untersuchungen fiber die

Pqvphipt^ ^ ^^°8lossus und seines absteigenden Astes. Jahrb. f.

Psychiat. u. Neurol., Bd. 24, S. 150.

Lubmch, 1899 Vergl^chend-anatomische Untersuchungen fiber den Ursprung Mes^g T M tenq Willisii. Arch. f. mikr. Anat., Bd. 54, S. 514.

der hpiwnpn hit h f L- ontwikkelingsgeschiedenis van de structuur

der hersenen bij het kip. Dissertation, Groningen.

"^^^z^Mrisphen Untersuchungen fiber den sogennantcn

PREzrosT? N'^'^'-ol. Instit. Wien, Bd. 32, S 262.

feri. ^ Nuovo Ercolanr^^ acessono del nervo abducente in alcuni uccelli e mammi Trab rte in! p^f terminaux du nerf acustique des oiseaux.

A. Maloine Paris ° ^ systSme nerveux de I’homme ct des vcrtebrds.

bulbo^on al™nn!^^pf5u los ganglios de la substancia reticular del

y mesocefdlicas Tmh ^°“cemientes d Jos focos motores y vias reflejas bulbarcs

SandeS^ r B 1090 de Madrid, vol. 7, p. 259.

centers and fiber tntpf certain bulbar, midbrain, and cerebellar

centers and fiber tracts in birds. J. Comp. Neurol., vol. 49, p. Ififi!

THE EFFECTORY SYSTEM 637

Tello, J. F. 1923. Les differenciations neuronales dans I’embryon du poulet, pendant les premiers jours de I’incubation. Travaux Cajal, T. 21, p. 1.

Terni, T. 1922, 1922a. Ricerche sul nervo abducenti ed in special modo intomo al significato del suo nucleo accessorio d’origine. Folia neurobiol., vol. 12. For 1922a, Arch. Fisiol., vol. 20, p. 305.

Una, F. M. 1922. Uber die embryonare Entwicklung des Oculomotoriuskeme Arch. Ophthamol. (von Graefc’s), Bd. 107, p. 123.

ILAMMALS

Addens, j. L. 1933. See bibliography for cyclostomes.

. 1934. A critical review of the occurrence of crossing root fibers in facialis,

vestibular, glossopharyngeal and vagus nerves. Psychiat. en neurol. Bladen, Jaarg. ’34, p. 274.

Allen, W. E. 1923. Origin and distribution of the tractus solitarius in the guinea pig. J. Comp. Neurol., vol. 35, p. 171.

. 1923a. Origin and distribution of the secondary visceral fibers in the guinea pig.

J. Comp. Neurol., vol. 35, p. 275.

Ariens ICappers, C. U. 1909. Weitere Mitteilungen fiber Neurobiotaxis. III. Ueber den Einfluss der Neurone der Geschmackskerne auf den motorischen Facialisund Glossopharyngeuskem und ihr Verhalten zur Radix descendens nervi quinti. Folia neuro-biol., Bd. 3, S. 259.

. 1910. Idem. V. The migrations of the motor cells of the bulbar trigeminus,

abducens and facialis in the series of vertebrates, and the differences in the course of their root-fibers. Verhandl. d. kon. Akad. v. Wetensch. te Amsterdam, Tweede Sectie, Deel 16, no. 4, S. 1.

. 1911. Idem. VI. The migrations of the motor root-cells of the vagus

group, and the phylogenetic differentiation of the hypoglossus nucleus from the spino-occipital system. Psychiat. en neurol. BL, Amsterdam, vol. 15, p. 408.

. 1912. Idem. VII. Die phylogenetische Entwicklung der motorischen Wur zelkeme in Oblongata und Mittelhim. Foha neuro-biol., Bd. 6, Sommer-Erganzungs-Heft, S. 1.

. 1913. Phenomena of neurobiotaxis in the central nervous system. Transactions

of the XtTI International Congress of Medicine, London.

. 1920-21. Vergleichende Anatomie des Nervensystems. (German edition of

present text.) E. F. Bohn, Haarlem.

Bach, L. 1896. Experimentelle Untersuchungen fiber die Localisation im Oculcmotoriuskemgebiet, sowie fiber den Trochleariskem. Zentralbl. f. Nervenh. u. Psychiat., Bd. 19, S. 611. See also Arch. f. Ophth. (von Graefe’s), 1899.

1903. Pupillenstudien. Arch. f. Ophth. (von Graefe’s), Bd. 57, S. 219.

Beccari, N. 1923. In tomo al primo differenziamento dei nuclei motori dei nervi cranici. Monitore Zool. ital., vol. 34, p. 161.

VON Bechterew, W. 1897. Ueber die Kerne der mit den Augenbewegungen in Beziehung stehenden Nerven (des Oculomotorius, Abducens und Trochlearis) und fiber die Verbindung derselben imter einander. Arch. f. Anat. u. Physiol., Anat. Abt., S. 307.

. 1899. Die Leitungsbahnen im Gehim und Rfickenmark. 2*® Aufl., Leipzig.

Beck, W. 1895. Ueber den Austritt des N. hypoglossus und N. cervicalis primus aus dem Centralorgan beim Menschen und in der Reihe der Saugetiere unter besonderer Berficksichtigung der dorsalen Wurzeln. Anat. Hefte, Bd. 6, S. 249.

Berkelbach VAN DER Sprenkel, H. 1924. The h3TDOglossal nerve in an embryo of Erinaceus europaeus. J. Comp. Neurol., vol. 36, p. 219.

638 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Bernheimer, S. 1896, 1897. Zur Kenntnis der Lokalisation im Kemgebiete des Oculomotorius. Wien. klin. Wochenschr., Bd. 9, S. 73. Arch. f. Ophth. (von Graefe’s), 1897.

. 1897, 1898. Die Ergebnisse experimentellen Untersuchungen iiber die Augen muskelkerngebiete. Die Heilkunde, 1897. Arch. f. Ophth. (von Graefe’s), 1899. DE Beule, F. 1901. Contribution I’^tude des lesions des cellules de I’hypoglosse apres I’arrachement du nerf. Le N4vraxe, vol. 3, p. 143.

VAN Biervliet, J. 1899. Noyaux d’origine du Nerf oculo-moteur commun du Lapin. La Cellule, vol. 16, p. 5.

Bischoff, E. 1899. Ueber den intramedullaren Verlauf des Facialis. Neurol. Centralbl., Bd. 18, S. 1014.

Black, D. 1917 . The motor nuclei of the cerebral nerves in phylogeny : A study of the phenomena of neurobiotaxis. J. Comp. Neurol., vol. 27, p. 467.

Bolk, L. 1898. Die Segmentaldifferenzierung des menschlichen Rumpfes und seiner Extremitaten. Morphol. Jahrb., Bd. 25, S. 520.

Bremer, J. L. 1908. Aberrant roots and branches of the abducens and hypoglossal nerves. J. Comp. Neurol., yol. 18, p. 619.

. 1921. Recurrent branches of the abducens nerve in human embryos. Am. J. Anat., vol. 28, p. 371.

Brouwer, B. 1918. Klinisch-anatomische Untersuchungen tiber den Oculomotoriuskem. Zeitschr. f. d. ges. Neurol, u. Psychiat., Bd. 40, S. 152.

RUCE, A, ANn PmiE, J. H. H. 1908. On the origin of the facial nerve. Rev. Neurol, and Psychiat., vol. 6, p. 685.

Bun^Fbdern, E. 1897. Ueber den Kem des Nervus accessorius. Monatschr. f. Psychiat. u. Neurol., Bd. 2, S. 427.

^ J o centrale Ursprung des N. vagus. Monatschr. f . Psychiat. u. Neurol,

IdcI, 5, S. 1.

Castaldi, L. 1923. Studi sulla struttura e sullo sviluppo del mesencefalo. Ricerche in Ca^a cobaya Parte prima. Arch. ital. di anat. e di embrioL, vol. 20, p. 23. y^RAMER, J. B 1894. Beitrage zur feineren Anatomic der Medulla oblongata und der Brucke. G. Fischer, Jena.

Darkschewitsot, L. 1889. Ueber den oberen Kem des N. oculomotorius. Arch, f. Anat. u. Physiol, Anat. Abt., S. 107.

Ursprung und den centralen Verlauf des Nervus accessorius Wilhsii. Allg. Zeitschr. f. Psychiat., Bd. 43, S. 453.

. 1888. Uslsr die Beziehung des Nervus accessorius zu den Nn. vagus und hj^o glossus. Allg. Zeitschr. f. Psychiat., Bd. 44, S. 655.

EXMR, H. 1899. Die Nervenkrankheiten des Pferdes. F. Deuticke, Leipzig u.

mo?' “°<'ralen Bahnen des Kau- und Schluckactes. Arch. f. d. ges. Physiol (Pfluger’s), Bd. 91, S. 629.

fiber den Bau der nervosen Centralorgane des^Menschen und der Thiere. 7^ Aufl., F. C. W. Vogel, Leipzig. 8‘« Aufl. in

^ intracranial and extracranial parts of the ninth, tenth,

and eleventh nerves. J. Anat., vol. 61, p. 298

FtS/i cranial nerves. Univ. of Chicago Press, Chicago.

Ann Plin iV/f ^ conoscenza del nucleo ambiguo in “Sus scropa.”

Ann Chn. Med. e Med. Sperim., vol. 19, p. 785.

^'ichiat ^ vo^ 7 ^p^l6 nuclei nel mesencefalo dell’uomo. Quademi di

ven ^np.I? mesencefalo dell’uomo. Arch,

gen. d. neurol. e psichiat., vol. 2, p. 120. (Quoted from Castaldi, ’23.)

THE EFFECTORY SYSTEM 639

Froriep, a. 1882. Ueber ein Ganglion des Hypoglossus und Wirbelanlagen in der Occipitalregion. Arch. f. Anat. u. Physiol., Anat. Abt., S. 279.

. 1885. Ueber Anlagen von Sinnesorganen am Facialis, Glossopharyngeus und

Vagus, liber die genetische Stellung des Vagus zum Hypoglossus, und liber die Herkunft der Zungenmusculatur. Arch. f. Anat. u. Physiol., Anat. Abt., S. 1.

Fuse, G. 1912. Ueber den Abducenskem der Sauger. Arb. a. d. himanat. Inst, in Zurich, Bd. 6, S. 401.

Gegenbaur, C. 1898. Vergleichende Anatomic der Wirbelthiere mit Beriicksichtigung der Wirbellosen. W. Engelmann, Leipzig.

VAN Gehuchten, a. 1899. Sur I’existence on la non-existence de fibres crois4es rlnns le tronc des nerfs moteurs craniens. J. de neurol., vol. 4, p. 7.

. 1903. Recherches sur I’origine rdelle et le trajet intracir^bral des nerfs moteurs

par la mdthode de la ddg^n4rescence wall^rienne indirecte. Le N6vraxe, vol. 5, p. 263.

van Gehuchten, A., and van Biervuet, J. 1901. Le noyau de I’oculo-moteur commun 16, 19, et 21 mois aprfes la resection du nerf. Le N6vraxe, vol. 2, p. 205.

Goldstein, M., and Minea, I. 1909. Quelques localisations dans le noyau de I'hypoglosse et du trijumeau chez I’homme. Folia neuro-biol., Bd. 3, S. 135.

Golgi, C. 1893. Sur I’originc du quatridme nerf c6r6bral (path6tique) et sur un point d’histo-physiologie g6n6rale qui se rattache k cette question. Arch. ital. de biol., vol. 19, p. 454.

Graboiver, Doctor. 1889. Das Wurzelgebiet des motorischen Kehlkopfnerven. Zentralbl. f. Physiol., Bd. 3, S. 505.

. 1895. Ueber die Kerne und Wurzeln des N. accessorius und N. vagus und deren

gegenseitige Beziehungen; ein Beitrag zum Studium der Innervation des Kehlkopfs. Arch. f. Laryngol. u. Rhinol., Bd. 2, S. 143.

Gray, L. P. 1926. Some experimental evidence on the connections of the vestibular mechanism in the cat. J. Corap. Neurol., vol. 41, p. 319.

Grosssun, M. 1916. Zur Anatomie und Physiologie des N. vagus-accessorius. Wien. med. Wochenschr., Bd. 66, No. 26, S. 983.

Gudden, H. 1891. Beitrag zur Kenntnis der Wurzeln des Trigeminus-nerven. Allg. Zeitschr. f. Psychiat., Bd. 48, S. 16.

Harman, N. B. 1907. The origin of the facial nerve. Brit. Med. J., pt. 2, p. 1296.

Hepburn, D., and Waterston, D. 1904. A comparative study of the gray and white matter, of the motor-cell groups, and the spinal accessory nerve, in the spinal cord of the porpoise (Phocaena communis). J. Anat., vol. 38, p. 105.

Herrick, C. J. 1931. An introduction to neurology. 5‘'* edition, W. B. Saunders Co., Philadelphia and London.

Huber, E. See account of Dr. E. Huber’s work in The Evolution of the Nenmus System by Ariens Kappers, ’29, p. 304. For further references see Anat. Anz., Bd. 51, 58, 59 u. 61.

Huber, E., and Hughson, W. 1926. Experimental studies on the voluntary motor innervation of the facial musculature. J. Comp. Neurol., vol. 42, p. 113.

Huber, G. Carl. 1896. Observations on the innervation of the sublingual and submaxillary glands. J. Exper. Med., vol. 1, p. 281.

Huber, G. Carl, and Crosby, E. C. 1929. The somatic and visceral connections of the diencephalon. Arch, neurol. and Psychiat., vol. 22, p. 187.

. 1930. (See Piersol, G. A. ; Human Anatomy.)

Hudovernig, C. 1907. Beitrage zur mikroskopischen Anatomie und zur Lokalisationslehre einiger Gehimnervenkeme (Nervus Hypoglossus, Vagus und Facialis).

J. f. Psychol, u. Neurol., Bd. 9, S. 137.

. 1908. Idem. J. f. Psychol, u. Neurol., Bd. 10, S. 247 ; Bd. 11, S. 26.

640 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

VON Husten, K. 1924. Experimentellc Untcrsuchungen ubcr die Beziehungen clcr Vaguskeme zu den Brust- und Bauchorganen. ^itschr. f. d. ge.s. Neurol, u. Psychiat., Bd. 93, S. 763.

Ikegami and Yagita, K. 1907. Ueber den Ursprung des Herzvagu.s. Mitt. d. Med. Gesellsch. in Okayama.

Kaida, Y. 1929. t)ber den Ursprung und den peripheren Verlauf der sog. zentrifugalen Vestibularisnerven nach Leidler (Fasciculus vestibularis mcdialis nach Kaplan) Arch. Ohren-, Nasen-, Kehlkopfheilkund., Bd. 123, S. 62.

Kaplan, L., and Finkelnburg, R. 1900. Beitriige zur Kcnntnis des sogenannten ventralen Abducenskems (van Gehuchten’scher Kern). Arch. f. Psychiat., Bd. 33, S. 965.

Kohnstamm, 0. 1902. Vom Centrum der Speichelsecretion, dem Ncrvus intermedins und der gekreuzten Facialiswurzel. Verb, Congr. Tunerc Mcdizin. Congr. 20, p. 361.

Kohnstamm, 0., and Wolfstein, J. 1907. Versuch einer physiologischen Anatomic der Vagusurspninge und des Kopfsympathicus. J. f. Psychol, u. Neurol., Bd. 8, S. 177.

VON Kolliker, a. 1896. Handbuch der Gewebelehrc des Menschen, Teil 2. W. Engelmann, Leipzig, 1889-02.

■ Medulla oblongata und die Vierhiigelgegend von Omithorh^chus

und Echidna. W. Engelmann, Leipzig.

Kosaka, K. 1909. Ueber die Vaguskeme des Hundes. Neurol. Centralbl, Bd. 28, b. 406.

Kosa^, K., Aim Yagita, K. 1903. Experimentelle Untcrsuchungen uber die Urspmnge des Nervus hypoglossus und seines absteigenden Astes. Jahrb. f. Psychiat. u. Neurol., Bd. 24, S. 150.

«Jes R. descendens N. hypoglossi beim Kaninchen.

Mitt. d. Med. Gesellsch. in Okayama.

Kdr 6 S. 1899 Die nomale und pathologische Stmctur der Zcllen an der cerebralen urze ^ es ervus trig^inus, die Kreuzungsfrage der Ictzteren und der motorischen ^ n^mmuswurze . Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 6,

Langwortoy, 0. 1924. A study of the innervation of the tongue musculature with particular reference to the proprioceptive mechanism. J. Comp. Neurol., vol. 54,

^ 1926. The mammalian oculomotor nucleus. J. Anat., vol. 60,

‘■“ThIsI'

Lubokh, JY 1899. Vergl«chend-anatomische Untersuchungen iiber den Urspmng LunrFtnmrln pT N. accessorius Willisii. Arch. f. mikr. Anat., Bd. 54, S. 514. Malone E F Bayern. 1884. Zur Anatomie der Zunge. Miinchen.

of nerve cells whinh ' ^ cardiacus nervi vagi and the three distinct types

?ol l^P i Jl. tbree different types of muscle. Am. J. Anat.,

^^S^m^d Bup^rieur. Rev. neurol., vol. 6, p. 30. La

presse med., vol. 9, no. 65, p. 85, 1899.

- mirnten bis zwolften Kopfnerven bei der

™l, moidologische sul Meleo del ipogloeso. Arch.

S. 273^’ zentrale Hypoglossusbahn. J. f. Psychol, u. Neurol., Bd. 29,

THE EFFECTORY SYSTEM

641

Mingazzini, G. 1928. Mcdulln oblongata unci Briicke. G. von Mollendorf’s Handbuch dcr mikroskopischo Anatomic dcs Mcnschcn. Ncrvensystcm, Bd. 4, S. 579. J. Springer, Berlin.

Mingazzini, G., and Poumanti, 0. 1910. Ucber die korticalen und bulbaren Verbindungen dc.** lIypoglos.sus. Monatschr. f. Psychiat. u. Neurol., Bd. 27, S. 187.

Molha.nt, M. 1910. lx* nerf vague. Elude anatomique et experimental. Partid I. Lc noyau dorsal du vague, sc.s connexions analomiques et sa valeur fonctionelle, Ix! Ndvmxe, vol. 11, p. 137.

Musskn, a. T. 1909. Note on the movement.s of the tongue from stimulation of the twelfth nucleus, root, and nerve. Brain, vol, 32, p. 205.

N<j, R. Ixmr.NTO DF,. 1922. Conlribricion al conocimiento del nervio trigemino. Libro on honor. Cajnl, T. 2, p. 13.

OiiKii.STEi.NEn, H. 1912. Anleitung beim Studium dcs Baues der nervosen Zentralorgane. F. Dcuticke, Leipzig u. Wien.

Oi'PEL, A, 1900. Lohrbuch dcr vcrgleichcndcn mikroskopischen Anatomic der Wirbeltierc. Bd. 3. G. Fischer, Jena, 1896-1914.

Pacetti, G. 1895. Sopra il nuclco di originc del Nervus abducens. Ricerche n. Lab. di anat. norm. d. r. Univ. di Roma, vol. 5, H. 2, p. 121.

Paneghossi, G. 1898. Contributo alio studio anatomo-Ssiologico dci centri oculomotori dcH’uomo. Ricerche n. lab. di anat. norm. d. r. Univ. di Roma, vol. 6, fa.'c. 2 and 3, p. 103.

. 1903. Wcitcror Beit rag zum Studium dcr Augcnmuskclnervcnkeme. Monat ,schr. f. P.sychiat. u. Neurol., Bd. 16. S. 258, u. S. 344.

Papez, j. W. 1927. Subdivisions of the facial nucleus. J. Comp. Neurol., vol. 43, p. 159.

Paiuion, C., and Goldstein, M. 1899. Sur Poriginc do la branche descendante de l’h>’poglossc. Roumanic mt'dicalc.

PAnuoN, C., AND Nadejde, G. 1905. Nouvelle contribution 5. 1’dtudc des localisations dans les noyaux dcs nerfs craniens et rachidiens chez I’homme et chez le chien. Arch, dc neurol., Scr. 2, vol. 20, p. 241 ; also J. de neurol., vol. 11, p. 129, 1906. Congres fran^ais dcs mcdccins alicnistcs et ncurologistcs, Rennes, 1905.

Paiuion, C., and Papinian, J. 1904. Contribution & I’etudo des localisations dans les noyaux bulbo-protubGrantiels (hypoglosse et faeial) chez I’homme. Semaine mcd., vol. 24, p. 401.

Pahiion, C., and ]?ARnoN, Mme C. 1903. Contribution i IMtude des localisations dans !e noyau de I’hypoglospc. Rev. neurol., vol. 11, p. 461.

Parsons, F. G. 1895. Myology of Rodents. Part II. An account of the Myomorpha, together with a comparison of the muscles of the various suborders of rodents. Proc. Zool. Soc., London, p. 159.

Perlia Doctor. 1889. Die Anatomic des Oculomotoriuscentrums beim Menschen. Arch. f. Ophth. (von Graefe's), Bd. 35, Abt. 4, S. 287.

PiERsoL, G. A. 1930. Human anatomy. 9"* .edition revised under the supervision of G. Carl Huber. J. B. Lippincott Co., Philadelphia and London.

pRkcECUTfiL, A. 1925. Some notes upon the finer anatomy of the brain stem and basal ganglia of Elephas indicus. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 28, p. 1.

pREZinso, L. 1924. Sul nuclco acessorio del nervo abducente in alcuni uccelli e mammiferi. Nuovo Ercolani.

Ram6n y Cajal, S. 1895. Apuntes para el estudio del bulbo raquideo, cerebello y origen de los nervios encefalicos. XVI. Nucleo del facial. Madrid. Anales de Historia Natural, Sesi6n del 6 de Febrero de 1895.

. 1897. Nueva contribucion al estudio del bulbo raqiiideo. Los micleos de origen

del nervio espinal. Rev. trimest. micrograf., vol. 2, p. 67.

642 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ram6n y Cajal, S. 1909-11. Histologic du systSme nerveux de I’homme et des vert6br4s. A. Maloine, Paris.

Rawitz, B. 1903. Beitriige zur mikroskopischen Anatomic dcr Cctacecn. II. Ueber die Zungc von Delphinus delphis. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 20, S. 429.

Rhinehart, D. 1918. Nervus facialis of the albino mouse. J. Comp. Neurol., vol. 30.

p. 81.

Roller, F. C. W. 1881. Ein kleinzelliger Hypoglossuskem. Arch. f. mikr. Anat., Bd. 19, S. 383.

Romano, W. 1898. Zur Frage von den centralen Verbindungen der motorischen Himnerven. Neurol. Centralbl, Bd. 17, S. 593.

Ruge, G. 1887. Untersuchungen iiber die Gesichtsmuskulatur der Primaten. W. Engelmann, Leipzig.

. 1893. Die Hautmuskulatur der Monotremen. Semon’s Forschungsreisen, Jena.

. 1897. Ueber das peripherische Gebiet des Nervus facialis bei Wirbelthieren. Festschr. f. Carl Gegenbaur, Bd. 3, S. 193. W. Engelmann, Leipzig.

Sabin, F. R. 1901. An atlas of the medulla and midbrain. Friedenwald, Baltimore. Schwab^ H. 1896. Ueber Gliederung des Oculomotoriushauptkems und die Lage der den einzelnen Muskeln entsprechenden Gebiete in demselben. Neurol. Centralbl., Bd. 15, S. 792.

SiEMERLiNG, E 1891. Zur Syphilis des Centralnervensystems. Arch. f. Psychiat., Joa. 22, S. 257.

SiEiiERLiNG, E., AND BoEDEKER, J. 1897. Chronische fortschreitende Augenmuskellahmung und progressive Paralyse. Arc. f. Psychiat., Bd. 29, S. 716.

Staderini, R. 1895. Nucleus intercalatus. Arch. ital. de’biol., vol. 23, p. 41.

. 1895a. Ricerche sperimentali sopra I’origine reale dell’ipoglosso. Intemat. Monatschr. f. Anat. u. Physiol., Bd. 12, S. 27.

Starr, M A. 1888. Ophthalmoplegia externa partialis. J. Nerv. and Ment. Dis., vol. lo, p, oOl,

S™ER G^L

. development of the cranial and spinal nerves in the

i one human embryo. Am. J. Anat., vol. 4, p. 83.

Pr °ngin of the cranial nerves in the 10 mm. human embryo.

Anatomists, 1908. Abstract in Anat. Rec., vol. 2, p. 111.

iiv Tfoii, I nervous system. Manual of human embryology

L^Sn ’ ’ "" ’ ■ Lippincott Co., Philadelphia and

cIm Opiilnm^rft Lehre von der Lage und der Function der einzelnen Zellgruppen Stutoman F T ^raefe’s), Abt. 2, Bd. 41, S 1.

___Inaug;ral‘ Dissertation linst^rdl'^^^^

■ Hypoglossuskem der Ameisenfresser. Anat. Anz., Bd. 49, S. 342.

Anz., M. 4? S 593 ^ Zungenmuskeln im Nucleus hypoglossi. Anat.

del'suo nnoW abc^centi ed in special modo intomo al significato

1920 11 ° Tt ° Holia Neurobiol., vol. 12, p. 277.

vol. 40 p 349 ^ umano nello stadio presomitico. Monitore zool. ital.,

Arb™' d newnT Laprung der motorischen Kerne im Hirnstamme.

fourth a;rsk?r3‘'T°’ ^nd aflerents of the third,

lourth, and sixth cranial nerves. Proc. Roy. Soc., London, Ser. B, vol. 82,

THE EFFECTORY SYSTEM

643

Tsuchida, U. 1906. Ueber die Ursprungskerne der Augenbewegungsnerven und iiber die mit diesen in Beziehung stehenden Bahnen im Mittel- und Zwischenhirn. Arb. a. d. himanat. Inst, in Zurich, Bd. 2, S. 1.

VAN Valkenburg, C. T. 1910. Nucleus facialis dorsalis, nucleus trigemini posterior, and nucleus trochlearis posterior. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pt. 1, p. 143.

. 1912. On the splitting of the nucleus trochlearis. Kon. Akad. v. Wetensch. te

Amsterdam, Proc. sect, sc., vol. 14, pt. 2, p. 1023.

Vermeulen, H. a. 1913. Note on the size of the dorsal motor nucleus of the X‘ nerve in regard to the development of the stomach. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 16, pt. 1, p. 305.

. 1914. Das Kehlkopfpjeifen beim Pferde. van Dishoek, Utrecht.

. 1915. The vagus-area in Camelopardalus giraffe. Kon. Akad. v. Wetensch. te

Amsterdam, Proc. sect, sc., vol. 18, pt. 1, p. 647.

. 1915a. The vagus area in Camelidae. Kon. Akad. v. Wetensch. te Amsterdam,

Proc. sect, sc., vol. 17, pt. 2, p. 1119.

. 1916. On the vagus and hypoglossus area of Phocaena communis. Kon. Akad.

V. Wetensch. te Amsterdam, Proc. sect, sc., vol. 18, pt. 2, p. 965.

. 1918. Die Accessoriusfrage. Psychiat. en neurol. BL, Amsterdam, vol. 22,

p. 729.

Vincenzi, L. 1884. Note histologique sur Torigine de rdelle quelques nerfs c6r4braux. Arch. ital. de biol., vol. 5, p. 109.

. 1903. Sulla presenza di fibre incrociate nel nervo ipoglosso. Anat. Anz., Bd. 22,

S. 567.

Vogt, H. 1905. Ueber die Anatomie, das Wesen und die Entstehung mikrocephaler Missbildungen. Wiesbaden. See also 1905, Monatschr. f. Psychol, u. Neurol., vol. 17, p. 337.

VoRis H. AND Hoerr, N. 1932. The hindbrain of the opossum, Didelphis virginiana. J. Comp. Neurol., vol. 54, p. 277.

Wainstein Z. 1921. Ueber die motorischen Vaguskeme. Dissertation, Petersburg. Referred to in Zentralbl. f. d. ges. Neurol, u. Psychiat., Bd. 28, S. 185, 1922.

Weed L. H. 1914. A reconstruction of the nuclear masses in the lower portion of the human brainstem. Publ. Carnegie Inst., Washington, No. 191.

Weigner K. 1901. Beziehungen des Nervus accessorius zu den proximalen Spinalnerven. Arb. a. anat. Inst., Abt. I, Bd. 17, S. 551.

Weinberg E. 1928. The mesencephalic root of the fifth nerve. A comparative anatomical study. J. Comp. Neurol., vol. 46, p. 249.

Westling. 1889. Anatomische Untersuchungen fiber Echidna. Svenska Vet. Akad. Handb., Bd. 15, Afd. 4.

Westphal, C. 1887. Ueber einen Fall von chronischer progressiver Lahmung der Augenmuskeln (Ophthalmoplegia externa) nebst Beschreibung von Ganglienzellengruppen im Bereiche des Oculomotoriuskems. Arch. f. Psychiat., Bd. 18, S. 846.

Willems E. 1911. Localisation motorice et kinesthdsique. Les noyaux masticateur et mesencephalique du trijumeau chez le lapin. Le N^vraxe, vol. 12, p. 1.

WiNDLE, B. C. A., AND Parsons, F. G. 1897. On the myology of the terrestrial Carnivora. Part I. Muscles of the head, neck, and fore-limbs. Proc. Zool. Soc.,

London, p. 370. t. t

. 1899. On the myology of the Edentata. Proc. Zool. Soc., London. Part I.

Muscles of the head, neck, and fore limb, p. 314. Part II. Muscles of the hind limb j and summary of conclusions respecting the musculature of the order, p. 990.

. 1901. On the muscles of the Ungulata. Part I. Muscles of the head, neck,

and fore-limb. Proc. Zool. Soc., London, p. 656.

644 N^WUS SYSTEMS OF VERTEBRATES AND OF MAN

jy^ti^SLE, W. F. 1931. The neurofibrillar structure of the spinal cord of cat embryos "■ correlated with the appearance of early somatic movements. J. Comp. Neurol., vol. 53, p. 71.

. 19^ia. . The sensory components of the spinal accessory nerve. J. Comp.

Neurol., vol. 53, p. 115.

. 1933. (See also 1932, 1932a.) Neurofibrillar development in the central

nervous system of cat embryos between 8 and 12 mm. long. J. Comp. Neurol., vol. 55, p. 643. See also J. Comp. Neurol., vol. 55, p. 99 and p. 315.

Yagita, K. 1909. Weitere Untersuchungen fiber das Speichelzentrum. Anat. Anz., Bd. 35, S. 70.

. 1910. Experimentelle Untersuchungen fiber den Ursprung des Nervus facialis. Anat. Anz., Bd. 37, S. 195.

1914. Einige Experimente an dem Nervus petrosus superficialis major zur Bestimmung des Ursprungsgebietes des Nerven. Folia neuro-biol., Bd. 8, S. 361. Yagita, K., and Hayama, S. 1909. Ueber das Speichelsekretionscentrum. Neurol. Centralbl., Bd. 28, S. 738.

Ziehen, T. 1897. Der Aufbau des Cervicalmarks und der Oblongata bei Marsupialiem und Monotremen. Anat. Anz., Bd. 13, S. 171.

CHAPTER VI

THE COORDINATING APPARATUS

The Reticular Centers of the Medulla Oblongata and the

Midbrain

Any discussion of the motor centers of the brain stem would be incomplete without some account of the effector apparatus, such as the reticular elements and their associated fiber tracts, which are concerned particularly in coordinating motor reflexes. Primitively the reticular elements appear, for the most part, in the ventral region of the medulla oblongata below the sulcus limitans and near the primary motor cells. During phylogenetic and ontogenetic development they migrate frequently to other portions of the medulla oblongata under the influence of various stimuli (neurobio taxis). They often differ, then, from the primary motor neurons in their positions in the adult and in their size. However, certain reticular elements arise in the dorsal sensory area of the medulla oblongata and retain this position. Sometimes such neurons are regarded as sensory nuclei, since they are connected in particular with the collaterals or stem fibers of a definite sensory root. However, as a rule they are connected in such a way as to receive more than a single type of stimulus and their cells give rise to a final common path (Sherrington, '06), which passes either to motor centers in the vicinity or to such centers situated at other and particularly more caudal levels.

It is self-evident that sensory impulses are frequently responded to by more than a single group of muscles. A bright object appearing within the range of vision of an animal leads not only to a movement of the eyes toward it, but frequently to a turning of the head and — particularly in animals with an immovable head, as fishes — of the whole body toward the source of the stimulation. In feeding reactions, especially in lower animals, groups of muscles innervated by distinct and widely separated nervous centers are brought into play. Thus in seizing prey, not only the jaw musculature, but frequently that of the whole body and tail is involved. The above are merely illustrations of the amount of coordination necessary on the motor side of the arc for the performance of acts of everyday occurrence. Such coordinations are accomplished to a very considerable extent by means of reticular cells. These cells are either arranged in groups or scattered throughout the gray. Gradually certain of these reticular groups become arranged to form recognized nuclear masses in higher forms ; others remain as small, scattered groups. Their ultimate arrangement is dependent, to a considerable degree, upon their relations with the sensory centers. Thus, when the stimulations from a given sensory center greatly predominate, the reticular

645

646 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

.cells tend to concentrate near it and form a definitely circumscribed nuclear mass, "sensitivo-reflectory in type for the sensory nerve in question. The nucleus of Deiters and the tangential nucleus of higher animals are examples of nuclear masses formed by the organization of the reticular cells of the medulla oblongata under the influence of vestibular fibers near to their point of entrance to the brain stem. The majority of reticular cells are not dominated by a particular system, but are the cells of origin for the final common path of several systems, and consequently do not accumulate into a common nuclear mass but remain scattered or in small groups. Such scattered cells almost invariably have several large dendrites which branch out in all directions. Many of the scattered cells have no special designation. The larger groups have been named by various investigators. The present knowledge of these groups is based on the researches of Beccari (’08, ’21), Tretjakoff (’09), Ramon y Cajal (’09), van Hocvell (’ll), Barlelmez (’15), Castaldi (’23, ’26), Payez (’26), Allen (’27, ’27a, see p. 310), and others. The following account is an attempt to summarize some of the more important observations of these workers.

Cyclostomes

The reticular elements of cyclostomes have been described by Johnston (’02) and Tretjakoff (’09), who differentiated certain groups which are situated more or less in contact with appropriate motor centers. Herrick and Obenchain (’13) plotted the position of several large cells of this type in their description of Ichthyomyzon concolor, however, without describing their connections, and Jansen ( 30) described a nucleus reticularis mesencephali and a nucleus motorius tegmenti in Myxine glutinosa. According to Tretjakoff, these reticular cells in cyclostomes are scarcely ever under the direct influence of sensory roots. The most conspicuous representatives of these cells are known as the cells of Muller.

The nucleus reticularis inferior (fig. 316; nucleus motorius tegmenti of Jansen, 30) lies at the level of the efferent nucleus of the vagus, being practically coextensive with that nuclear mass, and falls into a dorsal and a ventral portion The dorsal portion lies in or near the ventral part of the central ^ay of the region. In Myxine {Jansen) this consists of relatively large cells, while the ventral portion consists of relatively small cells, some near the ventral surface and others along the caudal part of the lemniscus bulbans. The functions of these two parts of the nucleus reticularis inferior are no un erstoo very well. Some of the dendrites of the larger cells have a re a ive y ^ e spread and consequently must receive impulses from various cen ers, possi y rom visceral areas (for visceral reflexes), and from trigeminal

L'Cll 1 /v3tS •

'The ventral part of the nucleus reticularis medius (fig. 316), lying at the level of entrance of the vestibular nerve and, in Myxine {Jansen, ’30) at least, practically coextensive with the facial and trigeminal motor nuclei, has large neurons, arranged with more widely branching and larger dendrites than the inenor reticular nucleus. One or two very large cells occur. Of the two giant

647

648 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

cells forming the dorsal part of the nucleus one lies directly behind and the other directly in front of the facial nucleus. Their dendrites are widespread while the neuraxes reach the caudal part of the cord by way of the medial longitudinal fasciculus.

A third group, consisting of much smaller cells (fig. 316), lies directly in front of the motor nucleus of the trigeminal nerve, at the level of entrance of the sensory root of that nerve. This is designated as the trigeminal group (Ammocoetes, Tretjakoff, ’09 ; also fig. 316). Its dendrites have a greater spread in the lateral than in the ventral regions of the medulla oblongata. Its neuraxes run caudalward in the medial longitudinal fasciculus of the same side. Occasionally cells of this type are found farther ventralward in Petromyzon. This trigeminal group was described by J ansen (’30) in Myxine glutinosa under the name of the nucleus ventralis tegmenti and was regarded by him as extending forward in this animal into relation with the interpeduncular nucleus.

The so-called isthmus group (fig. 316) consists of very large reticular elements, the largest in the whole brain. It lies at the level of the trochlear nucleus but somewhat ventral to it. The trigeminal and isthmus groups together may be termed the nucleus reticularis superior. The dendrites of the large cells extend in part to the nuclei of the trochlear and oculomotor, and consequently should receive the same stimuli as do the motor cells of these cranial nerv'es. A mechanism is thus provided for coordinate movements of eye and body muscles.

Nucleus reticularis mesencephali has a function somewhat similar to that of the last reticular group described. It forms the midbrain group of Muller’s cells. It begins in the region of the oculomotor nucleus and extends to the region of the posterior commissure.’ In certain cyclostomes (see fig. 316) it consists of both small and large cells, the dendrites of which are in relation with the nucleus of the oculomotor. The neuraxes of many of these cells run in the medial longitudinal fasciculus of the same side to the caudal end of the spinal cord. During their comse they give collaterals to the nuclei of the eye muscles and so provide for coordination of eye and tail musculature. The neuraxes of neurons of the nucleus reticularis mesencephali, the nucleus reticularis superior, and the nucleus reticulans medius form the Muller fibers described in the chapter on the spina cor (see page 156). The region of termination of the neuraxes of the smaller cells of the more frontal reticular groups is not definitely known. It has been said that the nucleus reticularis mesencephali contributes fibers to the postenor commissure. Jansen (’30) found that in Myxine the anterodorsal neurons of his nucleus reticularis mesencephali send large dendrites in a dorsal irection to intermingle with posterior commissure fibers. The nucleus mesencephah of myxinoids consists of four to six large, multipolar giant cells with large, radiating dendrites and is situated at the plane of the posterior commissure.

s neuraxes form Muller’s fibers, which have the usual relations with the medial longitudinal fasciculus.

’ It may be subdivided dorsal produces an elevation

into smaller groups (Herrick and Obenchain, ’ 13 ) of which the most Of the sulcus limitans.

THE COORDINATING APPARATUS

649

Selachians

The reticular cells in selachians show somewhat different topographic relations. Large, multipolar cells are found scattered along nearly the whole extent of the fasciculus longitudinalis medialis, lateral and ventrolateral to it. Such cells are not primary motor neurons but reticular cells.

Fia. 317. The reticular elements in the lower part of the medulla oblongata (nucleus reticularis inferior) in a ray. van Hoevell,

Fig. 318. The reticular elements in the acoustic or octavus region (nucleus reticularis medius) in a ray. van Hoevell.

Fig. 319. The motor nucleus of the trigeminal nerve and the nucleus reticularis superior of Raja clavata, van Hoevell.

The reticular elements in the region of the vagus are relatively very numerous. They are found ventral to the nucleus in the region of the raph4. They are often so numerous that the term nucleus reticularis inferior {van Hoevell, ’ll) is justified. In such cases they extend not only into the raph6, but lateral to it (as in Acanthias, Saito, ’30, Folia Anat. Jap., vol. 8, p. 323), and form garland-like masses about the medial longitudinal fasciculus. This cell group is relatively clearly

G50 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

delimited both caudalward and frontahvard (fig. 317). It is the myclenccphalic group of Beccari (’30).

Further cephalad there arc also special accumulations of reticular cells. These are found at the same cross-sectional level as were the similar cells in Petromyzon ; that is, at the entrance of the vestibular nerve (as the nucleus reticularis medius), at the level of the trigeminal nuclei, and in the tegmental portion of the midbrain.

At the level of entrance of the vestibular nerve only a verj’’ few cells are found in the region of the raph6. The major number of the reticular elements arc bilaterally arranged, lying lateral to the raph6 and near the vmntrolateral margin of the medial longitudinal fasciculus, where from G to 8 such cells ma}’’ be seen on each side of a given section. However, certain cells form a group situated more ventrolaterally and clearly in relation with the incoming fibers of the vestibular. From these cells arises a special bundle of fibers, the fasciculus medianus of Stieda, which takes a position lateral to the medial longitudinal fasciculus and extends far down into the cord (sec fig. 318).

The number of reticular elements decreases immediately in front of the level of the vestibular nerve, but increases again in the region of the trigeminal nerve to form the nucleus reticularis superior (fig. 319). The cells of this reticular nucleus (which is the nucleus pretrigeminus of Bcccari, ’30) are arranged simiarly to those at the level of entrance of the vestibular nerve; that is, lateral to the raph4 and in garlands around the medial longitudinal fasciculus. They lie chiefly in the upper third of the cross-section.

In front of the nucleus reticularis superior the number of reticular cells is ^ea y re uce , a though the cells do not entirely disappear. Another accumulation of them is to be found at the base of the midbrain, lateral and in front of the roots of the oculomotor. This metencephalic principal group {Bcccari, ’30), or the nucleus reticu ans mesencephali, is the most frontal and dorsal of these elemen s an s in front of the nucleus of the oculomotor nerv'e into the

caudal part of the posterior commissure, with fibers of which the cells are intermg e . IS possible that these reticular elements in selachians are the orerunners of interstitial cells of higher animals, which send their neuraxes caudalward in the medial longitudinal fasciculus.

UANOIDS AND TeLEOSTS

nuclear pattern of the nei^mus system is more definite arounc? i elements are arranged in somewhat more distinct

nouncpd Cfitr of the sensory reflex centers are more profound /n ft ^ ^ ‘^"ossopterygian. Polyodon, Hocke Hoogenboom (’29)

fascifidur No " ventrolateral to the medial longitudinal

area whptp . r f ^ the more rostral part of the

situat Jd ^th the somewhat more medioventrally

level of entrnn ^ "ooleus reticularis medius is represented best at the

irt ventro 1 ' its cells occupy an

tirely ventrolateral position, due no doubt to the entrance of the vestibular

THE COORDINATING APPARATUS

651

fibers and those of the lateral-fine system. In Polyodon the Mauthner cell lies at this level, as also in Acipenser (J ohnston, ’01). The neuraxis of the Mauthner cell has the usual relation to the medial longitudinal fasciculus. Between the nucleus reticularis inferior and the nucleus reticularis medius are to be found scattered reticular cells, but Hocke Hoogenboom found the majority of the neurons constituting the nucleus situated in front of the Mauthner cell. This observer recognized also a nucleus

reticularis superior, at the ' PSupa ^

level of the trigeminal \ ^>PSup

rnnf. nnH n r»nn1<inc ro- aW PPretl^'. 'MulC^

. Lob'CH

VO..

U--Nm

N ac vtnV

N Mot Teg fnfc J

>Mlj; C

"'N /I mb

N.ac-dprs.

Mot ]?

'Mot m

reticularissuperior, atthe ' PSupa. /'w

level of the trigeminal \ /PSap

root, and a nucleus re ticularis mesencephali ff-F-'

which contributes fibers 'W Npr.Y — ^ • \

to the medial longitudi- / , ' \

nal fasciculus and the / '^^'^^N.ac-dars. 1

posterior commissure. V '. -i M 'f'\ /

At the level of en- / \ Lob 'CUM .' / 'v y

trance of vagus root fibers ,, >/ t V:*^ '}\

inthetrout(Reccan,’ 22 ) \

is found a group of teg- ^ , ' Xf / X jC

mental cells, situated \ \yMijl C yjl

close to the vagus center. I Mot Te^ infc'|K. f^bXk I 11 Mot CH

This dorsal tegmental 1 )|-llx

column or nucleus pre- 'd ‘I

sumably is under the 5 g V'v'v ■ ".j S

influence of secondary

neurons from the gusta- ” /

tory and general visceral N.Ccrv.I w- • y-*N. R

sensory area, dominated 'A i j

by the fibers of the vagus, I j

and is concerned in the , . , , . ,

. i r ' 1 Fig* 320, Projection of sensory, motor, and reticular nuclei of

carrying out of visceral Ameiurus in a horizontal plane. Bartelmez.

reflexes. The neuraxes of The reticular elements are cross hatched; the motor nuclei are

Ur, «r,iir, 4 - 1 ^ r^/-I T o 1 mdicatcd by obliquc lincs ; thesensory areas are dotted. Mauthner’s

Its cells enter the medial and Muller’s cells (iim.C.) are indicated by black dots,

longitudinal fasciculus.

Beccari recognized two further reticular groups in the trout at more frontal levels of the vagus, a nucleus of the raph5 and a ventral tegmental column or nucleus. The nucleus of the raph 6 consists of small cells situated in the median fine near the ventral border of the medulla oblongata. The ventral reticular column or nucleus lies in the lower half of the medulla oblongata on either side

SgR.

N.Cen.I"'

of the raph 6 , although the cells are well away from the ventral surface. The dendrites extend among the fibers (see Beccari, ’22, figure 5 IE, p. 227) of the acoustico-lateral lemniscus or fasciculus longitudinalis lateralis, the fasciculus octavo-motorius, and the tractus tecto-bulbo-spinalis cruciatus. The three groups mentioned constitute the nucleus reticularis inferior. The tegmental cells corresponding to the nucleus reticularis medius, in the trout, according

652 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

to Beccari, consist of a lateral group of cells, between the raph6 and the acoustico-lateral lemniscus and relatively close to the medial longitudinal fasciculus, and a ventral group at the level of entrance of the vagus, with dendrites extending among the bundles of the crossed and imcrossed tecto-bulbar system. The ventrolateral vestibular nucleus or Deiters’ nucleus, which is recognized as a specialized portion of the tegmental gray (Ariens Kappers, ’20, and elsewhere; Beccari, ’22, and others), lies at this level. It has been described in Chapter IV, under the account of the acoustico-lateral system. There, also, was described Mauthner’s cell (page 456), which is a constituent of the lateral tegmental coliunn (of Beccari) at this level. Here attention is called to the important distribution of its processes. Our present knowledge of this distribution is based on the work of various observers — Beccari (’07), Bartelmez (’15), Tiegs (’31, see p. 131), and Bartelmez and Hoerr (’34) , and others. The chief, lateral dendrite and the cell body of Mauthner’s cell are in direct synaptic relation with vestibular fibers ; its upper ventral dendrite extends into the region of the chief sensory nucleus of the trigeminus, while its lower ventral dendrite is in synaptic relation with the ventral nucleus of the vestibular. The reticular cells lying behind the cell of Mauthner (pars postmauthneriana, Bartelmez, ’15) have the same connections as the lower ventral dendrite of the Mauthner cell. Together with the cells in their immediate vicinity, they probably constitute the homologue of the nucleus reticularis medius of sharks. The reticular elements lying in front of the cell of Mauthner have the same connections as the upper ventral dendrite of that cell. Their principal synaptic connections, then, are with the chief sensory nucleus of the trigeminal and with the cerebello-motorius system {Bartelmez). This premauthnerian group probably corresponds, at least in part, to the trigeminal group found in selachians. Some of the cells of this anterior group extend as far forward as the level of the trigeminal root. In the trout {Beccari, 21) there are two groups or columns of tegmental cells in this region, a lateral and a ventral, which have positions corresponding to the similarly named groups of the nucleus reticularis medius. These groups at and near the^ level of entrance of the trigeminal are the homologue of the nucleus reticularis superior of lower forms. Their neurons are in relation with collaterals of the acoustico-lateral lemniscus (fasciculus longitudinalis lateralis), a fact of importance in the consideration of the further development of this fiber system.

At the base of the imdbrain in teleosts, large reticular elements are found ; these are particularly evident in the region immediately in front of the nucleus of t e oculomotor. They are homologous to the nucleus reticularis mesencephali of selachians. In teleosts, as in selachians, they are undoubtedly concerned in coor mating movements of eye and body. Apparently they are in synaptic relation with the cerebellum through collaterals or direct cerebello-tegmental ers, other tegmental nuclei. Their neuraxes run caudalward in or beside e me la ongitudinal fasciculus. Several of the nuclear groups from which constituents of this bundle arise, in the midbrain as well as in the vestibular region, may be considered as belonging to the tegmental cell groups.

653

THE COORDINATING APPARATUS

Amphibians

Our knowledge of the reticular elements in amphibians is based very largely on the observations of Beccari (’07) and of Herrick (’14, ’30) ; less on the work of other observers. However, these workers studied particularly larval types. The reticular elements of adult amphibians are not so completely known as might be desired. Special nuclear groupings of the reticular elements — such as have been described for cyclostomes, selachians, and teleosts — have not been observed in amphibians, and probably are not present.

The nucleus motorius tegmenti (Edinger, ’08), as defined by Herrick (’30) for Necturus, “comprises the gray substance (except the neurons of the peripheral motor nuclei) beginning a short distance from the median sulcus and extending laterally to merge by insensible gradation into the gray of the reticular formation.” He included thus not only the larger elements to which the name nucleus motorius tegmenti or nucleus reticularis as a rule is limited, but also smaller cells of the area; these smaller cells probably serve local reflexes. Herrick traced neuraxes of certain of the larger cells into the contralateral medial longitudinal fasciculus, while certain others were observed turning spinalward on the same side, and other decussating components could not be traced. In tailed amphibians a Mauthner’s cell is present on each side of the medulla oblongata at the level of the entering vestibular root. As in teleosts, its neuraxis crosses and courses caudally in the medial longitudinal fasciculus. In Necturus {Herrick, ’30) two main dendrites pass out from this cell, a lateral and a medial dendrite. The lateral dendrite extends into the region of entering vestibular fibers {Herrick). The medial dendrite forms synaptic relations with stem fibers and collaterals of the medial longitudinal fasciculus. Ventrally directed dendrites {Ariens Kappers) come into relation TOth other fiber systems of the medulla oblongata, particularly the tecto-bulbar and spino-bulbar paths. In adult tailless amphibians there is no cell of Mauthner. The e^ddence is not at hand to enable determining whether or not it is present in the larval stage of the frog. At the base of the midbrain in amphibians are large, polygonal nerve cells which lie near the nucleus of the oculomotor. Their connections are not known vdth certainty. They are probably related to the posterior commissure and the medial longitudinal fasciculus, but their neuraxes are said to be directed caudalward. Herrick (’30) differentiated, from the nucleus motorius tegmenti, reticular gray which he regarded as “ a synaptic field interposed between sensory and motor fields in the interest of reflex adjustments.”

Reptiles

The arrangement of the reticular cells in reptiles is much more distinct than in amphibians. Van Hoevell (’ll) has studied these cells quite thoroughly, and the following account is based on his work, that of Beccari (’22, ’22a) and of Tuge (’32, bibliography, p. 1216).

The posterior group, nucleus reticularis inferior, is represented by a cluster of cells (fig. 321, a) in the raphe, as in selachians. Lateral to the raph4 (fig. 321, h) is a garland-shaped mass throughout the entire region of the vagus. Whether or

K«L«(Yin

SttJedXI

XdWuc

Uf rtbcacKb)

W nbc. Md (i) U rtfic. bcL (0

Fig. 321. The nucleus reticularis inferior of Caiman, van Hoevell.

654 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

not the most laterally situated cells of the region are to be regarded as a part of this nucleus is questionable. The neuraxes of the neurons constituting the clearly recognizable nucleus reticularis inferior enter the medial longitudinal fasciculus

and descend to the spinal cord. The course of the neuraxes of the lateral group is not known. It is possible that these cells are the forerunners of the lateral reticular nucleus of mammals. It is thought that they are associated with the ventrolateral areas of the medulla oblongata and that their neuraxes ascend along the ventrolateral margin of the region.

The lower, large-celled reticular elements in Caiman are intermingled with, or better, surrounded by, small cells (van Hoevell, ’ll). This becomes more and more evident from reptiles on. In mammals the ring of smaller cells around the larger cells is a conspicuous characteristic, not only of the reticular nuclei, but of the motor nuclei as well. About the level of the glossopharyngeal nerve there is a break in the reticular nuclei, only a few scattered reticular cells being present. Farther frontalward the large polygonal elements make their reappearance. However, here they are no longer found in the raph4, but along the side of it, beside the emerging roots of the abducens (fig. 322) and at the level of entrance of the vestibular root. They constitute the nucleus reticularis medius.

This arrangement persists to the level of the nucleus of the trigeminus. In the pretrigeminal region the cells show a distinct ventrolateral shifting (fig. 323) in the direction of the anterior portion of the superior oUve and of the lateral lemniscus. It was noted in teleosts that the more anterior reticular cell group received numerous collaterals from the lateral lemniscus. Therefore, this ventrolateral shifting toward the lemniscus in reptiles is not surprising, since the cells in lower forms had migrated there under the influence of the lateral lemniscus. There are small cells (see fig. 323) associated with the raph4 at this level and the reticular cells at the angle of the lateral ventricle, although they contain no pigment, may be the forerunners of loci coerulei (Ariens Kappers, ’20).^ The above account is based primarily on the studies of van Hoevell and Ariens Kappers. Tuge (’32, bibliography, p. 1216) described in Chrysemys : (1) a pars

1, * Certainly the presence of pigment is no conditio sine qua non for this homology. Frequently

the substantia nigra is pigment free.

hdilMkn-'

IUJhi/mc]

IU.WT1

- f

IU.rtfic.Md

... Ax .

0)

ldrt&.Md

(0

>■•'¥ *•

. lUtfdViA

titnnmi

Fig. 322. The nucleus reticularis medius (6 and c) and other reticular elements in the medulla oblongata of Caiman, van Hoevell.

THE COORDINATING APPARATUS

655

superior of Caiman. vanHoevell. Compare this figure with figure 322, which is from a more caudal level.

inferior nuclei motorii tegmenti which, with cells of the nucleus raph6 pars inferior, represents the nucleus reticularis inferior ; (2) a pars lateralis nuclei motorii tegmenti ; (3) a pars superior nuclei motorii tegmenti, which falls within the nucleus reticularis medius ; (4) a nucleus reticularis diffusus, wluch includes the more scattered tegmental cells of the midbrain region. “Centro tegmentale metencefalico ” of Beccari (’22) corresponds to (3) of the above list and to the nucleus raphd pars superior; the “centre tegmentale mielencefalico ’’ to (1) and (2) and the nucleus raph4 pars inferior. An uncrossed path, the paramedian longitudinal fasciculus of Beccari (’22a, and possibly the laterodorsal bundle of Tuge), descends from metencephalic regions (see account for birds). The myelencephalic region {Beccari) has descending paths through the medial longitudinal fasciculus. The internal uncrossed and the contralateral paths described from the tegmentum in the chick {Ramdn y Cajal) are said to be probable homologues of the lateral and medial bundles of Tuge (’32, p. 264). Cerebello-tegmental paths have been seen by various observers (Edinger, ’08, and others). For further details the above-quoted papers should be consulted.

The number of large reticular cells has increased at the base of the midbrain in reptiles, as compared with amphibians and lower forms. They form a large celled nucleus which is situated lateral to the roots of the oculomotor (fig. 324). These cells mark the beginning of the large-celled part of the nucleus ruber (de Lange, ’12 and ’13 ; Beccari, ’23, and others). They are coordinating in function and give rise to a crossed, descending path. In front of the oculomotor nucleus a similar group constitutes the nucleus interstitialis of Cajal. A nucleus interstitialis of Cajal and two nuclei of the medial longitudinal fasciculus have been described for Varanus by Beccari (’23) and for Alligator mississippiensis by Huher and Crosby (’26). These will be considered further in the chapter on midbrain and diencephalon.

Fig. 324. The nucleus ruber in Anolis.

Birds

The reticular cells of birds are somewhat differently arranged from those of reptiles. The nucleus reticularis inferior is not so well developed within the raph6 as at the side of it. The neuraxes enter the medial longitudinal fasciculus after a partial decussation, the greater number of them being crossed {Ramon y Cajal,

656 NERVOUS SYSTEMS OP VERTEBRATES AND OF MAN

’09) . The cells within the raph6 are not only few in number but are much smaller than those at the side of it (van Hoevell, ’ll ; fig. 325).

Many large cells are found in the raph6 as constituents of the nucleus reticularis medius. These are the more conspicuous since other large cells are not

found in the raph6 in this region. The number of reticular elements is very large, and these increase rostrally at the level of the facial nerve (fig. 326). There the lateral cells assume a position close to the ventrolateral surface of the medulla oblongata, near the superior olive. Certain neuraxes of these cells enter the medial longitudinal fasciculus ; others are lateral to it. These latter constitute the larger bundle, which courses caudalward, uncrossed, at the side of the fasciculus and is usually designated the tractus homolateralis internus (Ramdn y Cajal, ’09). ... The nucleus reticularis superior is highly de^eriofo^f cSniraE" mn Swit' ^elpped in birds. It shows marked resemblance

to its homologous nucleus in Caiman. The neuraxes of its cells are almost entirely uncrossed, and in their course caudalward do not form a part of the medial longitudinal fasciculus but run at the side of this bundle, forming the so-called tractus homolateralis externus of Ramdn y Cajal (’09) or the fascicule paramediano longitudinals of Beccari (’22).

Flo. 32G. The nucleus reticularis medius of Passer domcsticus. RamOn y Cajal.

'The large-celled elements of the reticular substance in the midbrain show considerable nuclear differentiation as compared with those in lower form.-?, atera to the oculomotor root is a well-circumscribed nucleus ruber (fig. 327),

THE COORDINATING APPARATUS

657

forerunner of the large-celled portion of that nucleus in manunals. Neuraxes arising from it run caudalward but are not associated in their course with the medial longitudinal fasciculus, but run ventral to it and form a true tractus rubro-bulbo-spinalis. This nucleus receives cerebello-tegmental fibers and is a coordinating center for somatic responses. In every way, then, it is analogous to the midbrain reticular centers of lower forms. It is not usually counted among the reticular elements, however, because of the special course of its efferent tract.

The large cells dorsal, lateral, and frontal to this level, which constitute the nucleus interstitialis of Cajal, resemble much more typical reticular elements. Their neuraxes probably enter the medial longitudinal fasciculus. In addition to this nucleus, two nuclei of the medial longitudinal fasciculus, presumably homologous to the dorsal and ventral nuclei of this fasciculus in reptiles, have been seen in birds {Rendahl, ’24 ; Huber and Crosby, ’29). These various midbrain centers are discussed more fully in Chapter VIII.

Fig. 327. The nucleus ruber in the stork (Ciconia alba). de Lange.

Mammals

The arrangement of the large reticular cells in mammals is analogous in principle to that in lower forms, but there are certain differences, particularly in the nucleus reticularis superior, which must be emphasized. The smaller reticular elements, which are more numerous here than in lower forms, are arranged in various groups which frequently are delimited very poorly.

The followng account is based particularly on the studies of von Bechterew (’94), Kohnstamm (’99), Probst (’02-’02a), van Gehuchten (’04), Lewandowsky (’04), Jacobsohn (’09), Ramon y Cajal (’09-’ll), van Hoevell (’ll), and Papez (’26). Voris and Hoerr (’32), in their figures, have shown scattered cells of the nucleus motorius tegmenti through practically the whole extent of the midbrain of Didelphis virginiana. The nucleus motorius tegmenti centralis consists of cells scattered among the smaller cells of the reticular gray, sometimes lying near or even in the raph6 and sometimes in a more lateral position. The nucleus motorius tegmenti ventralis consists of a group of cells, presumably of less frontocaudal extent, situated dorsal to the inferior olivary nucleus. Their figures show also a well-defined nucleus funiculi lateralis, consisting of several clusters of cells dorsolateral to the inferior olivary nucleus. Apparently it is present throughout mammals, although it is usually designated the nucleus reticularis lateralis’ or merely the nucleus lateralis in higher forms and in man. In those forms, including man, in which its connections have been studied, it has been found to receive

659

THE COORDINATING APPAR/VTUS

From the nucleus pigantocellularis reticularis Papez (’26) traced fibers to the medial longitudinal fasciculus of the same side and, to some extent, of the opposite side. Such fibers constitute bundles of his medial reticulo-spinal tract, which arise in part from the nucleus just mentioned and the nucleus reticularis lateralis, and in part from the larger tegmental cells of the upper pons region and the isthmus region. The medial reticulo-spinal tract runs caudalward on the lateral side of the medial longitudinal fasciculus, with the bundles from the higher centers overlying t hose from more caudal centers. It enters the ventral funiculus of the spinal cord. It may be repeated that Allen (’27, ’27a; .‘^cc p. 374) regarded the solitario-spinal reflex path as relayed on the larger cells of the medullar reticular formation in its course to the cord ; thus the rcticulo-.spinal tract .scrvc.s important functions in respiratory and other visceral reflex paths.

Crossed and uncrossed ccrcbellorcticular fibers reach the nucleus reticularis inferior.

In front of the level of entrance 329- The nucleus reticularis medius of Macropus.

, ran Hocvcll.

of the acoustic or octavus roots the

number of reticular cells decreases, but increases again at the level of the motor nucleus of the trigeminal nerve, and in general in the more frontal region of the pons. According to wn Hocvcll ('ll), at these levels giant cells are no longer present in the raph6 of the kangaroo and the horse. However, other reticular cells are present in the raph6 associated with the nuclei pontis and farther frontahvard in tiic region just behind the corpora quadrigemina. The larger reticular cells in the trigemin.al and pretrigeminal regions, constituting the nucleus reticularis superior (see von Bcchlcrcio, ’94), show a marked peculiarity in many mammals in their arrangement. However, in Phocaena relatively simple conditions e.xist. The nucleus reticularis superior is not divisible into subgroups. Tlie cells are fairly evenly scattered throughout the reticular formation but are somewhat more plentiful in the dorsal part of the brain stem, probably because of the more dorsal position of the lateral lemniscus in these forms. In a great many mammals, as in reptiles, the cells are arranged in two more or less distinct groups (van Hocvcll, ’ll). One of these groups, the nucleus superior dorsalis, has a dorsal position nothin the brain stem. The nucleus superior ventrolateralis extends in a ventrolateral direction, taking up its position close to the lateral lemniscus. This lateral nucleus is particularly distinct in the horse, where it lies in part among the fibers of the lateral lemniscus (fig. 330, h' group). In the kangaroo, as in reptiles, the two groups are clearly continuous with each other. In certain other mammals, among which is the rabbit, van

660 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Hoevell (Tl) found a condition intermediate to that in the horse and in the kangaroo. The ventrolateral group in the horse (fig. 330) accompanies the lateral lemniscus and so lies at first ventrolateral and then farther dorsal (compare A and B of fig. 331)^ The ventrolateral shifting of the reticular cells of the nucleus reticularis ^perior and the consequent formation of a specialized ventrolateral nucleus may be due to the neurobiotactic influence of the lateral lemniscus fibers {Ariens Kappers, ’20). Variation in the position of this fiber tract determines the differentiation and position of the reticular cells in the trigeminal and pretrigeminal regions. Papez (’26) found that a lesion of the large cells of the cat nucleus reticularis superior, which is situated medial to the motor trigeminal nucleus, produces a contralateral degeneration of the lateral reticulo-spinal tract. This tract passes caudalward, medial, and largely ventromedial, to the descending root of the trigeminal nerve. At the proper level it is in close relation to the nucleus ambiguus, and enters the cord between the

lateral cortico-spinal and the rubrospinal tracts (Papez, ’26), close to the spinal gray. From the cells of the nucleus reticularis superior fibers are also contributed to the medial reticulo-spinal tract of Papez (see page 1078; also fig. 506).

The nucleus superior dorsalis (h) of van Hoevell is the homologue of the nucleus centralis superior or lateralis of von Bechterew (’94). In the cat and in the rabbit Ram^n y Cajal (’09- 11) found in this nucleus cells of medium size with many greatly branched and somewhat varicosed dendrites. This observer found the nucleus traversed by numerous fibers and giving origin to bundles the destinations of which were unknown, but which obviously decussated to the other side. The nucleus superior ventrolateralis is the homologue of the nucleus tegmenti lateralis described by von Kolliker (’91) as present between the lateral lemniscus and the superior cerebellar peduncle of man. It is probable that the nucleus paralemniscalis inferior, described by Kohnstamm (’10) for rabbits, is analogous to the b' position of the nucleus superior ventrolateralis of the above terminology. The nucleus paralenmiscalis of Kohnstamm (’10) is fittingly named, since it occupies the position which the name suggests (see figs. 330 and 331, A, B — h, b').

In figure 331^ the dorsal extent of the nuclei pontis is indicated by a dotted me. In the neighborhood of the raph6 the dorsal extension of these pontine nuclei is particularly emphasized. The cells in this region are not sharply separable from the remainder of the nucleus pontis in the horse. However, in several mammals, as, for example, in the rabbit, and particularly in Phocaena, cells of this domal group are clearly distinct from the rest of the nuclei pontis. Peculiar radiations of cells in the lateral and dorsal directions suggest that this mass has some special significance ; it constitutes the nucleus reticularis tegmenti

Fig. 330. The nucleus reticularis superior of the horse, van Hoevell.

THE COORDINATING APPARATUS

661

pontis of von Bechlerevu (’94), which was termed nucleus reticularis of von Bechterew by van Hoevell (’ll). This reticular nucleus (figs. 331B, 332) is connected with the cerebellum through the contralateral superior cerebellar peduncle {Lewandowsky, ’04; Papez, ’26). Lewandowsky (’04) also described fibers passing into it through the brachium pontis. Papez (’26) was unable to obtain degeneration of such fibers after lesions involving the brachium pontis. He suggested that these fibers might be reticulo-cerebellar fibers, but this point requires further investigation. Lesions of the nucleus reticularis produce homolateral and contralateral degenerations of the ventral reticulo-spinal tract, as described by Papez. However, such degenerations are due to the decussation of this tract in the region of the nuclei reticulares of the two sides. The observ^er suggested that the tract takes origin from the nucleus ventrolateralis of the reticular formation at levels just lateral to the nucleus reticularis. The ventral reticulo-spinal tract is lateral to the medial longitudinal fasciculus and dorsal to the medial lemniscus. In its course to the ventral funiculus of the cord it lies relatively close to the dorsal accessory olive.

Certain tegmental nuclei of the pretrigeminal region extend well forward through the isthmus region to the caudal end of the mesencephalon. Conspicuous among these are the dorsal and ventral tegmental nuclei. The dorsal tegmental nucleus (nucleus tegmenti dorsalis of von Gudden) is situated caudal to the nucleus of the trochlear nerve, in the isthmus region. Cells in this nucleus are relatively small compared with other reticular cells. They are either oval or triangular in outline and multipolar in character, giving off three relatively long and somewhat branched dendrites which spread out among the cell bodies constituting the nucleus, and also among the fascicles of the medial longitudinal fasciculus. The neuraxes of these cells are believed to enter the dorsal longitudinal fasciculus of Schiitz,' an observation first made by von Kdlliker (’96) . The dorsal tegmental nucleus receives habenulo-tegmental, mammillo-tegmental, and pedunculo-tegmental fibers. It is, then, a rela.y station for impulses passing between the diencephalic olfactory correlation centers and efferent centers of the

Sip.n&,Bd

<W)

Fig. 331. A. Section through the nucleus reticularis superior of the horse at the level of the trigeminal nerve.

B. Section in front of A, showing the nucleus reticularis superior of the horse, tan Hoevell.

6(52 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

brain stem. The ventral tegmental nucleus lies ventral to the dorsal tegmental nucleus, separated from it by the bundles of the medial longitudinal fasciculus, and dorsal to the nucleus centralis superior. Its cells consist partly of medium sized and partly of larger neurons with very long, little branched dendrites. Its connections are not very clearly understood. According to Papcz (’26), the medial reticulo-spinal tract arises in part from large reticular cells of the upper pons and isthmus regions, and certainly a portion of the degenerated fibers shown in his figures appear to involve the region of this ventral tegmental nucleus, but it cannot be stated with certainty that its neurons are contributors to that tract.

Casialdi (’26) termed a mass of gray dorsolateral to the nucleus dorsalis of the tegmentum the nucleus latero-dorsalis tegmenti. This latter nucleus, found in the aqueduct region in close relation with the mesencephalic root of the trigeminal nerve, extends forward into the mesencephalon and backward along the mesencephalic root behind the aqueduct. As seen in the guinea pig, this cell mass consists of cells of several sizes, and while it occupies in part the position of the nucleus of the locus coeruleus of man, it includes other gray (extending beyond the limits of the human locus coeruleus) and is unpigmented. Casialdi termed it the nucleus latero-dorsalis tegmenti (" nuclei latero-dorsale del t^mento ”) and considered it to be associative in function. He stated “Dalla mia descrizione resulta che il n. latero-dorsale fa parte della categoria dei nuclei associativi secondari ; come tutti gli altri che abbiamo visto format! de cellule non impregnabili col methode Cajal, che non hanno una connessione bene spiccata e definite, ma molteplici connessioni colie parti vicine.” He discussed its possible relations to the trigeminal nerve and concluded that, while it might receive some fibers or possibly collaterals of the mesencephahc root of the trigeminal, it could not be considered a true and proper (“vero e proprio”) sensory nucleus of the trigeminal nerve. (For Casialdi’s views with regard to the functions of the mesencephalic root and nucleus of the trigeminal, see Chapter TV, p. 405).

There have been various opinions with regard to the nucleus of the locus coeruleus {Held, ’93 ; F orel, ’77 ; many observers since) or, as it is sometimes called, the substantia ferruginea {Meynerl, ’72 ; Ford, ’77, and others).’ The human locus coeruleus itself is a “faintly bluish,’’ shallow groove, which extends to the cerebral aqueduct from the superior fovea of the ventricle (Ranson, ’31). By whatever name the cell mass is designated, there appears to be difference of opinion in regard to both its cell types and its relations to the trigeminal nerve. Thus Meynert (’72), Obersteiner (’93), Cramer (’94), von Kolliker (’96), Terterjanz ( 99), Johnston (’09), and Allen (’19) believed that it supplies fibers to the mesencephalic root of the trigeminal nerve, and the two latter observers could find no differences between the cells of the mesencephalic root of the trigeminal and those of the nucleus of the locus coeruleus. Meynert described the cells of the locus

(’ll) used both terms, calling the cells internal to the mesencephalic root of the

not external to the root the substantia ferruginea. This is

not a customary usage of the terms.

THE COORDINATING APPARATUS

663

coeruleus as fusiform in outline, Kure (’99) found large and small cells, of which the former were directly comparable to those of the mesencephalic root, while Weinberg found cells which were histologically different from those of the nucleus of the mesencephalic root, which he regarded as constituting the nucleus of the locus coeruleus, but intermingled with these at certain levels, he found true cells of the nucleus of the mesencephalic root of the trigeminal nerve. The statements of Kure and Weinberg are essentially in agreement. Schwalbe (’81), Ramon y Cajal (’09), and Weinberg (’28), then, believed that no contribution to the mesencephalic root of the trigeminal nerve comes from the nucleus of the locus coeruleus. Kure had obtained chromatolysis of the large cells of his nucleus on cutting the trigeminal nerve, but Kohnstamm (’10), May and Horsley (’10), and Kosaka (’12) failed to obtain alteration of the cells of the nucleus of the locus coeruleus after a comparable operation. Obviously the differences in opinion with regard to the functional relations of these neurons is dependent largely upon differences in the establishment of the limits of the nucleus of the locus coeruleus {Weinberg, ’28; Sheinin, ’30). Sheinin found pigment in the cells of this nucleus in preparations of the dog brain studied by him, but this pigment was not confined to the region of the aqueduct, nor, indeed, to the nucleus in question, but occurred also in the nuclei of the trochlear, oculomotor, and trigeminal nerves.

The dorsal nucleus of the lateral lemniscus, which, with the more ventrally situated cells scattered along this ascending system, is probably a derivative of the reticular gray, is considered to better advantage in other connections (see

p. 11).

Among the various groups of nuclei in the midbrain, which either belong to the reticular nuclei or have been derived from them m phylogeny, may be mentioned the red nucleus, the nucleus interstitiahs of Ramon y Cajal (’ll), the nucleus of Darkschewitsch, the nucleus lateralis profundus, and possibly cells of the annulus fasciculi longitudinahs medialis {Castaldi, ’23). All of these nuclei must be considered again in Chapter VIII, in order to be able to present a clear picture of the relations and functional activities of the tegmental portion of the midbrain. The pertinent literature will receive full consideration in Chapter VIII. Consequently, in this relation the nuclei above listed are described very briefly, without attempting to cover the literature at this time, beyond making references to respective pages in Chapter VIII, where a more complete account is presented.

Fia 332 The reticular nuclei at a frontal level of the medulla oblongata Cat Huher and Crosby

664 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The studies of Hatschek (’07) on the phylogenetic development of the red nucleus in manunals, the results of which are in accord with those obtained by de Lange (’12) for reptiles, indicate that in marsupials (Didelphis and Macropus) this nucleus consists almost exclusively of relatively large neurons, with large, widespread dendrites, having the characteristics of large reticular cells. In the majority of mammals the red nucleus is surrounded by a definite capsule of fibers. In the rabbit, Winkler and Potter (’ll) described a magnocellular portion which is surrounded in front and then laterally (dorsolaterally and to some extent ventrolaterally) by a parvocellular portion. At the outer side of the red nucleus lie small reticular elements. The reticular character of the red nucleus is indicated in the figures of Castaldi (’23). According to Ariens Kappers (’20), the large celled portion is very distinct in Edentates. The red nucleus of carnivores has been studied by a number of observers (von Monakow, ’95 ; Winkler and Potter, ’14; Rioch, ’29; Davenport and Ranson, ’30, and others). In both the dog and the cat (Rioch; Davenport and Ranson) it consists of a compact caudal portion, composed chiefly of relatively large cells, and a less clearly defined rostral portion, presenting an intermingling of large and small cells. Rioch found that in the dog the cephalic pole of the nucleus is slightly in front of the level of emergence of the oculomotor root, and Davenport and Ranson foimd that in the rabbit and in the cat it extends forward to approximately the same level. In man the nucleus consists of the usual mesencephalic magnocellular portion and a more rostral parvocellular part, which extends forward into the subthalamic region. Thus the parvocellular portion has become a conspicuous portion in the higher mammals. Its development is associated with a gradual increase in the range of functional activity of the red nucleus, associated with the progressively greater development of forebrain regions in higher forms. The small-celled portion, a way station between the cerebellum and higher centers, increases with the increase of diencephalic and telencephalic centers. The large-celled portion in man, as in lower forms, is a center for motor coordination and a nucleus of origin, at least in many mammals, for the rubro-spinal tract which decussates, immediately after arising from the red nucleus, in the ventral tegmental decussation or the fountain decussation of Forel, and then courses caudalward along the lateral border of the medulla oblongata in company with, and dorsal to, the lateral tecto-spinal tract. It contributes collaterals and stem fibers to the motor centers of the medulla oblongata and perhaps to the inferior olive, and then enters the spinal cord, within which it occupies a position internal to the ventral spinocerebellar system, Ijdng ventral to the lateral cortico-spinal tract and dorsal to the lateral tecto-spinal fasciculus. Other important connections of the red nucleus are pro\'ided by contralateral dento-rubral and homolateral corticorubral and strio-rubral paths. It is connected with the dorsal thalamus through the rubro-thalamic tract and receives crossed and uncrossed fibers from the optic tectum. Short internuclear fibers connect it vuth the substantia nigra, the zona incerta, the subthalamic nucleus, and the tegmental region of the midbrain, including the nucleus profundus lateralis mesencephali. The details of the structure, the connections, and the probable functions of the red nucleus are

THE COORDINATING APPARATUS

665

further discussed on pages 1085 to 1091, together with a review of certain experimental contributions.

The large cells which lie in the dorsal part of the tegmental region of the midbrain in front of the oculomotor nucleus have retained their reticular character more clearly than have the cells of the red nucleus. Here is found the nucleus interstitialis of Ramon y Cajal (’ll). This is the nucleus of origin for the most frontal fibers of the medial longitudinal fasciculus. The cells are lateral to the midline; they are not very large, are polygonal in shape, and have dendrites which branch among the descending tectal tract and the fibers of the posterior commissure. This nucleus is the chief representative of the nucleus reticularis mesencephali of lower forms. The greater development of other nuclei in the base of the midbrain makes it relatively less clear in mammals.

The nucleus of Darkschewitsch is placed with the tegmental cells of the midbrain, owing to the general character of its cells and because of its relation with the posterior commissure. Ram6n y Cajal (’ll) traced numerous collaterals from fibers of the medial longitudinal fasciculus, possibly fibers ascending from vestibular centers, to this nucleus. He also found neuraxes of the neurons of the nucleus entering the medial longitudinal fasciculus. One of the major fiber contributions of the nucleus of Darkschewitsch is to the posterior commissure. Possibly the nucleus plays a r61e in the bilateral coordination of eye-muscle movements {Aricns Kappers, ’20). Muskens’ views are discussed on pages 1078 to 1082.

The nucleus lateralis profundus mesencephali {Ziehen, ’20; Castaldi, ’23; and others) has received various designations, such as the following : nucleus lateralis mesencephali of Marburg (’03), nucleus lateralis der Haube s. pararubralis of von Monakow (’09-’10), nucleus motorius dissipatus formationis reticularis medullae pedunculi of Jacohsohn (’09), and ganglion profundum laterals mesencephali of Edinger (’ll). A long list of synonyms is to be found on page 144 of the 1923 paper of Castaldi. The nucleus lateralis profundus, which consists chiefly of scattered multipolar cells of medium size and with a general reticular character {Sterzi), Castaldi (’23) regarded as the direct continuation forward of the nucleus reticularis of the pons. At the level of the red nucleus it lies lateral and dorsolateral to that nuclear mass. The nucleus lateralis profundus is a region of passage for various fiber bundles, including the descending tectal and tegmental systems, to the latter of which it probably contributes. Neuraxes of its cells have been followed medially, probably to the medial longitudinal fasciculus ■ (see Castaldi, ’23). Nucleus lateralis profundus is connected with other mesencephalic centers, such as the red nucleus and the substantia nigra. It probably receives fibers from the cerebellum.

Various other nuclear groups to be found in the midbrain region may be wholly or in part derivatives of primitive reticular cells. Certain of these are discussed in Chapter VIII, in which the midbrain as a whole is considered. To such centers belong the cell clusters which constitute the annular nuclei of the medial longitudinal fasciculus (“Ringkem des hinteren Langsbundel ” ; see Ziehen, ’20, and Castaldi, ’23). The cell groups encircle the medial longitudinal

666 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

fasciculus, being found in a region limited by a plane passing through the frontal part of the dorsal tegmental nucleus to a plane passing through the oculomotor nucleus. The relative position and relations of this nucleus are well shown in figure 16, page 66, of the paper of Castaldi (’23) . This nucleus may be subdivided into several nuclear groups. In addition to these larger cells of the midbrain region, smaller cells of the reticular formation are found scattered among the more conspicuous nuclear masses, and a well-organized group of raph6 cells of the so-called linear nuclei (see Castaldi, ’23) and the cells intercalated in the midtegmental decussation, including the nucleus of the commissure of Wernicke, are present and serve as reminders of the raph6 nuclei of the pons and medulla oblongata.

R6sum6 of the Relations of the Reticular Cells of the Medulla

Oblongata and Midbrain

In all vertebrates a number of large, polygonal cells, distinctly coordinating in character, are found in the medulla oblongata and the midbrain. For the most part, their neuraxes course caudalward, usually in or near the medial longitudinal fasciculus. At first these cells show intimate topographic relations to the motor roots of the cranial nerves. Later they shift somewhat in position under the influence of stimuli reaching them. Since they represent the cells of origin for a final common path for stimuli from various sources, they seldom accumulate in large, clearly defined nuclei but remain in smaller, more scattered groups. The cells of such groups send dendrites out in various directions. In higher animals these nuclei take on a more complex appearance, due to the presence of numerous smaller cells here as in the motor nuclei.

A posterior group, nucleus reticularis inferior, is found at the level of the vagus. In the lowest vertebrates, such as cyclostomes, it is merely the dorsal continuation of the spino-occipital column of the spinal cord. A second group appears at the level of the octavus and facial nerves. It also is primarily dorsal in position and constitutes the nucleus reticularis medius. Nucleus reticularis superior is present at the trigeminal and pretrigeminal levels, while nucleus reticularis mesencephali occurs near the nucleus of the oculomotor. Many of the neuraxes arising from these nuclei are uncrossed.

The most important changes in positions and relations which these nuclei undergo are as follows : The nucleus reticularis inferior, in nearly all animals, assumes a ventromedial position ; that is, it is situated partly in the raph6 and partly beside it. It is believed to receive impulses from the visceral sensory and trigeminal nuclei (possibly also from the inferior olive) and to transmit them to more caudal motor centers. The position of these cells in the raph6 is most evident in selachians, reptiles, and mammals. It is less evident in teleosts and birds.

The nucleus reticularis medius in fishes is chiefly under the influence of root fibers of the vestibular and lateral-line nerves and fibers from their nuclei, and of the tecto-bulbar and cerebello-motorius systems. The importance of these connections is reflected in an increase in size of the dendrites of the reticular cells

THE COORDINATING APPARATUS

667

and a sliifting caudahvard of flic cell bodies. A cell on either side differenfiates in a very special manner. This is the cell of Mauthner. It is concerned in swimming rofle.\cs, since it transmits impuls&s brought in by the vestibular and lateral-line nerves directly to the motor centers for tail musculature. Its dendrites spread laterahvard and ventrahvard under the ncurobiotactic influence of certain fiber systems, notably the incoming vestibular roots.

The nucleus reticularis superior shows marked dilTerentiation in higher forms. Certain of the cells retain a position medial to the motor nucleus of the trigeminal ; others move farther ventrahvard and some cells, presumably under the neurobiotactic influence of the lateral lemniscus, form the nucleus paralemniscalis of Kohnstamm. At this level, also, diflerentiales the nucleus reticularis of von Bechtcrew, and from neurons of the nucleus reticularis superior, as well as from those of the nucleus reticularis inferior, arise important homolateral and contralateral rcticulo-spinal tracts. Particular!}' marked in the isthmus region are the dorsal, the ventral, and the superior central tegmental nuclei. Of these, the dorsal, and possibly the ventral, tegmental nuclei are way stations in descending paths from diencephalic olfactorj' correlation centers.

In the midbrain of lower vertebrates the cells constituting the nucleus reticularis mcsencephali usually fall into two groups. One group lies lateral to the root fibers of the oculomotor nerve, the other, in front of its nucleus. The first group in fishes consists of scattered cells. In reptiles the cells are organized into a wcll-circumscribcd nucleus, the forerunner of the nucleus ruber. In reptiles, birds, and lower mammals, the nucleus ruber, which receives dento-rubral and strio-rubral connections, has chiefly large, reticular cells, but in higher mammals both largo and small cells are found and the small cells predominate in man. In man this important gray mass remains the nucleus of origin of the rubrospinal tract, with its accompanying rubro-bulbar and probably rubro-olivary components. It is a motor coordinating center and a relay station for impulses from the cerebellum and from the striatum to the spinal cord. There is an important path concerned with the rubro-thalamic connections. By way of it, impulses reach the thalamic centers and, after sjmapse, the higher centers. It develops with the development of thalamic and forebrain centers.

The nucleus intcrstitialis of Ram6n y Cajal is a homologue, in part at least, of the nucleus reticularis mcsencephali of lower forms. It is probably concerned in the transmission of optic reflexes over the medial longitudinal fasciculus, in which its neuraxes run. The nucleus of Darkschewitsch, which also lies in this region, is a nucleus of termination and probablj also of origin for the medial longitudinal fasciculus, and contributes fibers to the posterior commissure. The nucleus lateralis profundus is a reticular center lateral and dorsolateral to the red nucleus and in the course of fiber systems which have passage through the midbrain. Probably it contributes to the descending tegmental paths and is believed to be connected with the red nucleus and the cerebellum. Certain other reticular centers which have been mentioned do not require rediscussion here.

From the foregoing account it is evident that the reticular elements of the medulla oblongata and midbrain constitute an integral part of the effector

668 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

apparatus of the brain stem. They differ from the primary motor neurons in being intrinsic to the nervous system. Their neuraxes do not terminate in peripheral end-organs but end in synaptic relation with motor neurons of the brain and spinal cord. But while the intrinsic character of the reticular elements may serve to differentiate them from peripheral motor neurons, it cannot be regarded as a criterion which may be used in discriminating between them and other secondary centers of the central nervous system. Thus a great number of secondary centers send fibers to motor neurons as well as to the reticular cells. Examples of such centers are the lateral vestibular nucleus (nucleus of Deiters) and the tangential nucleus. These latter even contribute fibers to the medial longitudinal fasciculus, yet once they are so fully developed in phylogeny they are no longer regarded as true reticular nuclei. The differentiating characteristic of the reticular nuclei is that they are not under the dominance of any one system but receive stimuli from various systems. Such stimuli may be interoceptive, exteroceptive, or proprioceptive in character. An illustration of this is to be found in the connections of the nucleus reticularis medius. The cells of this nucleus are in synaptic relation not only with the root fibers of the acoustic and secondary vestibular centers, but also with collaterals from the cerebellotegmental system and fibers from the chief sensory nucleus of the trigeminal nerve and from the tectobulbar tracts. Its neuraxes form the final common path for the discharge of stimuU reaching this nuclear mass. Its dendrites branch in all directions. This nucleus is usually scattered. The predominance of a single system, as indicated by the shifting of the whole or a part of the cell mass, is evident only in special cases.

As soon as a reticular group becomes dominated by a single system it assumes a more definite nuclear organization and usually migrates toward the source of its main stimulation. It now becomes a sensory-motor coordinating center for that system. The term of sensory reflectory nuclei has been applied to such nuclei. Examples of them are the lateral vestibular nucleus (of Deiters) and the tangential nucleus. Such differentiation into sensory reflectory centers usually occurs in reticular centers during phylogenetic development. Such differentiation has been emphasized in the accounts of selachians and cyclostomes. The variability in direction and source of the stimuli entering a given reticular center prevents its assuming the compact arrangement characteristic of the nuclei of motor roots. It is to be emphasized that the reticular centers are primarily coordinating centers which provide a “final common path ” to the motor centers.

Certain Other Coordinating Systems of the Medulla Oblongata

THE INFERIOR OLIVARY NUCLEUS

^ The reticular nuclei of the brain stem represent a system in which various mpulses, frequently of different types, are passed over to effectory centers. However, centers of coordination of another type are present in the brain stem; centers having a coordinating function, discharging the impulses which they have brought together, not into effectory centers, but into higher centers.

THE COORDINATING APPARATUS Various representatives of coordinating nuclei of o

with in the consideration of the brain stem as in “et

nucleus cuneatus, which receive impulses of two-noinrHi"^

proprioceptive types from the body over the fasci.dn«

cuneatus. Such impdses. after s^iapse in thnucJ J

cerebellum by way of arcuate fibers or forward by wav^Mi.

the sensory decussatim and the medial lemniscus tn cZ arcuates,

after synapse, they projected on the cortex Tb^ ‘Calamus, from whence,

no way show a segmntal character, Zeehandelaar >97^ “

relation and distrib’tion of impulses to higher centers the cor been described in erlier pages and need not receive fnrVb

(see pages 262 to 26). further consideration here

One very imporint correlation center

center is represent^ by a gray mass found in the lowe^ This

oblongata, which, ecause of similar topographic^ relation the cerebellum, hregarded as homologLs in all ver Jif "f connections with homology cannofoe established until its fiber connerH. ^^^phte

-- orgray r^hTinlt'

level of entanwf the .nterior epino^occipital reot. As yet™e7

of the nucleus f unknown, although tbo coni^tions

eellB appear tcurse ip a dorSal^d”

this observatioii desirable. ^ ^'^rther coloration of

Among thelagiostomes, the inferior oUvro,.,v t . A , , ,

{Kooy, ’17) in 1 sharks. In the lower nnrt f tif is At developed

forms, at the el of entrap of T the medulla o/ngata in these

accumulation small pear-shaped cells ! root, /ere is found an

olivary Uelt SeSriftSp^aJ

the lateral sid‘cre is a concavity (fig. 333C) which iAled by a fiber system.

rior Cfitr ^33‘tr'pp raph6, lies the nucleus reticularis infe nor fig. 333C»e a^so pap 649). This pucleus rSe, not form any part of

The connens cl the inferior olive with tb/ rpinal cord have not been defimlely estaied m plagiostomes. It is possible that the olive receives arcuate fib»s froae dorsa portion of the cord..' The cerebellar eonnSns “ p giostomes feiy evident. They consist of senate fibers which run dorsal

cJltben “'l™. decussate in the »“dline, and then ascend to the

cpeMum alehe upper border of the mednll, oblongata. They terminate in the body but f, the auncle of the cerebellurtfroorW, W ; see also p. 722)

670 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

The inferior olivary nucleus in the skate is similar to that in the shark, but the dorsally and ventrally coursing fiber bundles are somewhat less clear {Kooy, 17).

A slightly different type of development of this nucleus is evident in the Holocephali. In Chimaera monstrosa (Kooy, T7)_the inferior olivary nucleus is transitional in type between the nucleus found in selachians and that found in teleosts. In this fish the greater portion of the inferior olivary nucleus lies close to the raph6. At more frontal levels a second nuclear mass of a similar

RapbC

Inf olive —

Lower border — med obi

B

Fia. 333. A, a sagittal section through the medulla oblongata of ScylLum; B, a wax reconstruction ; and C, a cross section of the inferior olive in a selachian. Kooy. (The part indicated by crosses is the nucleus reticularis inferior.)

cell type is found, which has a more lateral position and only an indistinct connection with the more caudal nucleus. Close to the more frontal nucleus is a third nucleus, consisting of larger cells than those of the other two cell masses. Its relation to the inferior olivary complex is uncertain.

Mayser (’81) identified the inferior olivary nucleus in teleosts. According toKooy (’17), it is represented by a mass of small spindle-shaped cells present m all groups of teleosts, but present as a more or less compact nuclear mass in only a few forms (Anguilla and Clupea). It is situated in a position analogous to that of the lateral part of the inferior olive in Chimaera. Medial from this mass and near to the raph6, in a position analogous to that occupied by the inferior olive in selachians, there is a small area consisting of small cells and finely

THE COORDINATING APPARATUS

671

branching fibers, similar in appearance to the so-called gelatinous substance n Orthagonscus Burr (’28) was unable to recognize any nuclear mass just lateral corresponding to the inferior olivary nucleus as described by Koov ( 17). However, he did find a midline cell group extending from the medial lonptudmal fasciculus to the ventral surface of the brain, between planes through the first motor roots o' the vagus and the caudal nucleus of the abducens nerve Burr (and Anc/is Kapoors; see Burr, '28, p. 75) suggested this as a "generalized

c

Fio. 333. — {Conlinved).

ancestor of the live,” and its presence was confirned in other teleosts. This matter requiresirther investigation.

The inferioPhvary nucleus of teleosts rece'ves external arcuate fibers, especially evide io Lophius, from the dorsal h(rn of the spinal cord {Aliens Kappas), Als' certain fibers of the tecto-bdbar tract teimnate in this nucleus. RMn-.>ggcsted a connection of the iiencephalon with the inferior olivary nucleus ™ngh medial longitudinal asciculus : a fact of considerable interest if confi'cd by further observation. .

In amnhibis a nucleus homologous tothe mfenor ohve is not clearly differentiated i ordinary cell or fiber development

here, as in cy(Stomes, is probably associ?^d with the small size of the cerebellum.

672 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

In many reptiles no definite accumulation of cells homologous to an inferior olive has been demonstrated. However, along the ventral margin of the medulla oblongata, close to the midline, there are scattered cells which suggest the beginning of an inferior olivary nucleus. They are evident in hydrosaurians, Shanklin (’30, bibliography, p. 1215) showed an olive in Chameleon.

In birds the inferior olive (fig. 334) is sharply differentiated. It consists of two lamellae parallel to each other and to the surface of the medulla, an arrangement which was first described by Williams (’09) and Yoshimura (’10). An exception to this arrangement is found in runners such as the ostrich, where {Yoshimura, ’10, Groebbels, ’23) only a dorsomedial lamella is present, the place of the ventrolateral layer being occupied by a small group of laterally placed cells. Kooy (’17) regarded this exception as only apparent. He pointed out that the inner part of the ventrolateral layer, that is, the part turned toward the raphd, in all birds has fewer cells than the external portion. He regarded the ostrich as representing merely the extreme of the usual condition. Very careful examination showed that typical olivary cells, which characterize the inner part of the ventrolateral layer, are present, but that they are so few in number as not to be noticed readily. At frontal levels, this inner portion of the ventrolateral lamina constitutes the cormection between the dorsal and ventrolateral laminae of the inferior olive.

In all birds, the connection between the inner and outer portions of the dorsal lamina is not clear in more frontal sections, and the inferior olive appears to consist of three portions, a medial, a dorsal, and a ventrolateral part (fig. 334 ; see both parts). The outer portion of the dorsal lamina extends farthest caudalward (black, fig. 334, 1). Frontalward this layer increases in size on its medial side ; it is crossed by the emerging roots of the hypoglossal and is thus divided into an outer and dorsal portion and an inner and medial portion. The latter part becomes greatly thickened (fig. 334). At this same level the ventrolateral lamella, which at first was distinctly lateralward, becomes ventral to the dorsal portion of the medial lamina and unites with the ventral portion of it. In a general way the cell types of these three parts of the olive are the same. In parrots, at the periphery of the highly developed medial portion, there is a layer of large, flat cells, similar to those described in the nucleus cuneatus, but smaller. These cells will be found again in mammals. It is of interest that large birds in general have larger inferior olivary cells than do small birds. It has not been possible as yet to point out definite morphologic types of cells for the different orders of birds, nor to associate different types with different habits of life. In the runners, which are the oldest order, it is apparent that the ventrolateral lamina of the olive is least developed. This probably represents the most primitive condition. Such an interpretation is in accordance with our present knowledge of the inferior olive in mammals. The main inferior olivary nucleus of these latter animals has representation in the ventrolateral lamina in birds. Yoshimura (’10) believed that the development of the lower or ventrolateral lamella is concerned with flight in birds. Edinger (’08, bibliography for amphibians), Williams (’09), and Groebbels (’23) considered that the accessory

Fio. 334 pve, a wax reconstruction of thcinferior olivary nucleus of Lophortyx califomiciLS. The'oss sections are taken at differen' levels of the inferior olivary nucleus. (The raph6 portion ithe nucleus reticularis inferior i-designated by coarse dots.) Kooy.

,7S

674 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

olive is concerned with static and locomotive functions in fluid media, the chief nucleus with these functions under more stable conditions. Groebbels related the inferior olive functionally with the motor tegmentum, regarding it as serving static and locomotive functions in association with the cerebellum, a view supported by the presence of crossed and uncrossed paths between these centers {Yoshimura, TO; Shimazono, T2). Sinn (T3, Monatsschr. f. Psychiat. u. Neurol., Bd. 33, S. 1 ; see also Groebbels, ’23) described interolivary fibers.

The conditions in Echidna show a very distinct relation to those in birds. Von Kdlliker (’01) and Ziehen (’08) demonstrated that the inferior olivary com

Fig. 335. Above, a wax reconstruction of the inferior olivary nucleus of Eclidna, seen from the lateral side. ,

Below, a caudal view of the frontal part of this reconstruction. Kooy. 1

plex in Echidna consists of a dorsolateral or lateral arched portion and a cylindrical middle (ventromedial) portion (fig. 335), with the parts connoted by a portion which projects farther dorsalward than the other parts of the nucleus. This recalls the condition in birds, where the dorsolateral portion of the inferior olive is connected with the medial part by a lamella which lies dorsal to the rest of the olive. Fimthermore, in both avian forms and in Echidna this connecting portion does not extend so far caudalward as do the dorsal and medial parts of the arch (a condition which is true for the ventrolateral or chief inferior olivary nucleus of other mammals). The parts of the inferior olive in Echidna are analogous to those of birds. The avian dorsomedial lamellae and the dorsal curve and lateral plate (fig. 335) of Echidna are probably homologous to the accessory olives of other mammals. This is the more interesting because of the fact that in some other mammals the medial and dorsal accessory olives are con

THE COORDINATING APPARATUS 675

nected by a band of cells. The ventrolateral lamella in birds and the middle piece in Echidna correspond to the ventrolateral or principal nucleus of higher mammals. For a discussion of special points of correspondence, the work of Kooy (’17) should be consulted.

Three distinct portions can be demonstrated in the inferior olivary complex of higher mammals : a medial accessory, a dorsal accessory, and a chief inferior olivary nucleus. During the course of phylogenetic development the last of the

Med. arc, Dora. aec. Med. aec. ol. nucl. oL nucl. ol. nucl.

, j ^ — — • — -p j

Med. aec. nud* Med. acc. ol. cud. Vent. Ut. or chief ol. nucl. Med. acc. ol. nucL

Fia 338 Abovftateral view and below, a medial view of the inferior olivary nucleus of Phocaena

commi- Kooy. The great size of the medial accessory olive should be noted.

three division's gradually increased relatively as well as actually until in man it quite overslows the accessory olives. In the highest mammals these parts are distinct D each other. Nevertheless this increase in the chief nucleus is gradual and itions remain sufficiently similar, in passing from a lower to a higher form fbat it is possible to formulate certain statements concerning the shape and poon of the three parts of the inferior olivary complex which will

apply to all lomals. , . , -.t, .i, ,

The medi^cessory olivary nucleus, which is homologous with the medial part of the d(Daedial arch of birds, forms in many mammals, alone or with the dorsal accessioHvary nucleus, the caudal pole of the inferior olivary complex.

676 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and two crura, a ventral and a medial crus, are distinguishable in it at these more caudal levels. Over the medial crus, at the plane of the caudal end of the chief olivary nucleus, there is a cap of cells. In lower mammals the dorsal accessory olive and the chief inferior olive are often connected with this cap of cells, and in this way show primitive relations. In Phocaena communis (fig. 336, both parts) the medial accessory olive is remarkably well developed and its two divisions are very distinct.

Brunner (T8) has interpreted the nucleus just described as a medial accessory olive, as a new nucleus distinct from the inferior olivary nucleus, which according to the results of Kooy (’17) and Ariens Kappers (’20) does not appear to be correct. The cell mass identified by Brunner as the homologue of the medial accessory olive of other mammals is only the caudal part of that nucleus (see fig. 338). It corresponds in all respects to the caudo-ventral component of the medial accessory olive of other mammals. The nucleus situated more orally is triangular in cross section and is very typical in Cetacea. However, it probably represents a strongly hypertrophied medial component ofja ventromedial accessory olive, which in these forms has become entirely separated from the main nuclear mass. It bears the same relation then to the i^edial accessory olive as that latter nuclear mass bears to the inferior olive proper.) Histologically this nucleus of Brunner belongs to the olivary complex. Suci differences as can be cited — such as the vesicular shape of the cells and their mirginal arrangement, associated with the central core of fibers — are insignificant, for they are repeated in enlarged portions of the olive of other animals where the mass is relatively compact. Examples of this are found in the dorsal jlamella' of the inferior olive proper of elephants.

The dorsal accessory olive, the dorsolateral part of the arch in lower animals, extends nearly if not quite as far caudalward as does the medial accessory olive. On its first appearance at caudal levels it lies dorsal to the medial accessory olive. In more rostral sections it shifts farther lateralward, but returns to a more medial position at the lower level of the inferior olive (compare figs. 337 and 338). The wax model, figure 337, indicates that it has a U-shaped outline, the apex of the U coinciding with the caudal pole of the inferior olive. The descending limb is oval and the ascending limb is plate shaped.

The ventrolateral olivary nucleus, the chief inferior olivary nucleus of mammals, has much the same shape in all mammalian forms. It is situated at the frontal end of the inferior olivary complex and never extends as /ar caudalward as do the accessory olives. This is clearly shown in figure 337, where the caudal pole of the main inferior olivary nucleus falls at about the middle of the dorsal accessory olive. In cross section it is U shaped in outline, with the open part of the U directed dorsomedially in many mammals and medially in man, thus forming a hilus. This characteristic outline is visible in lower animals only in certain sections. In higher mammals it can be demonstrated thloughout much of the extent of the nuclear mass, being absent only at the closed caudal and oral ends of the nucleus. Occasionally the dorsal limb of the U is connected for a few sections with the dorsal accessory olive, while the ventral limb is con

Caudal

Frontal

r*vid r^v!#* arc ol

fhjfJnl nurl ntirl

Fio. 337_ Above, a lateral view of a wax reconstruction of the inferior olivary nucleus of the dog. Below, cross sections taken at different levels of the inferior olivary nucleus of this animal. Kooy.

677

678 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

tinuous with the medial accessory olive. Kankeleit (’13) found this “Vierblattertypus” (four-leaved type) in all mammals, although demonstrable only at certain levels.

After having demonstrated the homologies of the different parts of the inferior olivary complex in various mammals on the basis of certain evident characteristics, it is easy to trace the relative development of the various parts, and particularly the development of the main inferior olivary nucleus, from marsupials to the highest primates. During this development the accessory olives show little change in the principles of structure from the lowest to the

I

n

m

IV

Hypothetic Stage

Lower mammals

Higher mammals

Highest mammals

Medial complex only

Fig. 338. A scheme of the course of development of the inferior olivary nucleus (white) in contrast to that of the accessory olives, which are stippled. Kooy.

highest mammals. Such differences as do occur have to do principally with the development of the major inferior olivary nucleus. The development of this nuclear mass apparently occurs in a caudal direction. This appears to indicate that this portion receives a larger number of caudally running fibers, which enter it at the ventrolateral margin of the medulla oblongata {Ariens Kappers, ’20) . Examples of variation in extent are to be found throughout the mammalian phylum. Thus in the marsupials, as in Echidna, the anterior part of the olivary complex has a size such that the frontal half of the whole olivary complex is represented only by a ventromedially directed outgrowth of the medial accessory olive. In higher animals this nuclear mass occupies not only the entire frontal half but the greater part of the caudal half of the whole complex. At

679

Fio. 339. A cross section of the olivary complex in various mammals.

631

THE COORDINATING APPARATUS 683

The lateral part of the inferior olive, as has been stated before, forms the larger portion of the surface of that nucleus and has the most and the deepest fissures. Moreover, in all mammals this lateral part extends farther caudalward than does the medial portion. The medial layer is well developed only in the more frontal sections. These relations are true of man as well as of lower mammals, with the exception that the portion of the medial layer which extends farthest dorsalward is narrower in man and is separated from the rest of the

caudal and dorsal

nuclear mass. This thinning out of the layer is probably due to the great development of the fiber systems in the region of the hilus. A sun'C}' of the phylogenetic relations and the ontogenetic development of the olivarj" complex indicates that the inferior olive of mammals, as compared with that of man, is similar in its development and varies in the degree rather than in the kind of differentiation.

The development of the inferior olive is associated with the evolution of the cerebellum. This conception is confirmed by the pathologic ^ an anatonuc researches of Holmes and Stewart (’08), already referred to in this chapter, and by those of Bromeer (’18 ; also fig. 346). The observations of the firsUmentioned

684 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

workers indicate that when only the hemispheres of the cerebellum are injured, the accessory olives remain intact, but that these nuclei degenerate when the vermis is affected. These observers found, also, that the medial portion of the main inferior ohvary mass is connected with the vermis. Brouwer (T8) confirmed these results with the reservation that only the medial part of the rostral sections of the inferior olive and the whole rostral pole are connected with the vermis. The accessory olives and the rostromedial portions of the main inferior olive are the first to differentiate and are to be regarded as phylogenetically the oldest elements of the olivary complex of mammals. In the same way the vermis is the oldest portion of the corpus cerebelli. The remainder of the main inferior olivary nucleus is phylogenetically younger and is connected with the cerebellar henuspheres which are also relatively new. The fiber relations of the olivary complex to the cerebellum, as described by Holmes and Stewart (’08), are best understood by reference to figure 347. In this figure it will be seen that the dorsomedial portions of the inferior olive are connected with the superior

Fia 343 A wax reconstruction of the inferior olivary nucleus of a human fetus (14 cm. length) seen from the lateral side Kooy

medial portions of the cerebellum of the opposite side, the ventromedial with the inferior medial portions of the opposite cerebellum, while the dorsolateral and ventrolateral areas are related to the superior lateral and inferior lateral regions of the contralateral cerebellum. In other words, a given region of the inferior olivary nucleus is related to a locaUzed region of the contralateral cerebellum. It is noteworthy that thus far no connection of the inferior olive with the flocculus has been described, although this portion of the cerebellum is one of its oldest parts phylogenetically. Perhaps this lack of connection is due to the rather limited functional scope of the floccular portion, since it is dominated to a large extent by the vestibular (and lateral-line) areas of the medulla oblongata and receives no fibers of the spinocerebellar system. Kooy (’17) believed that the anterior portion of the medial accessory olive may be related to the parafiocculus since both are so large in Cetacea (see Chapter VII) . This latter point needs further research. Langworthy’s (’33, bibliography for Chapter X) studies on the infant brain showed that neuraxes arising from neurons of the accessory ohves and the medial part of the inferior olivary nucleus myeUnate before fibers from the remainder of the inferior olivary complex and pass to the vermis.

In addition to spino-olivary, olivo-cerebellar, and cerebello-olivary fibers, there are descending olivo-spinal fibers from the inferior olive to the motor nuclei of the cervical cord, including the accessory nucleus supplying the trapezius

TV

,-w v/ •:-m<-^im

Fiq. 345. A lateral view of a reconstruction of the inferior olivary nucleus of an adult man. X 9. Kooy.

Fig. 346. Atrophy of the human inferior olivary nucleus in two cases of neocerebellar atrophy. Brouwer. The atrophy of the cerebellar hemispheres is associated with an atrophy of the portion of the inferior olivary nucleus colored black in the above diagrams. The accessory olives and the frontal part of the inferior olivary nucleus, indicated by lined portions in the figures, remain normal, sinoe the vermis cerebelli was unaffected.

687

688 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

and sternocleidomastoid muscles. Also, there is a relatively large path, entering the nucleus on its lateral side, termed the thalamo-olivary fasciculus, but its origin from the thalamus is uncertain. WinkUr (’33 ; see p. 1239) described an olivo-striatal tract and favored a pallido-rubro-olivary tract in the central tegmental tract. Interolivary fibers connect the nuclei of the two sides. Tilney (’27) believed that the inferior olivary nucleus receives secondary trigeminal

fibers, at least in certain mammals such as the elephant. In addition to the typical spino-olivary tract from the spinal cord, decussating fibers from the upper part of the cervical cord in fishes and from the nuclei of the posterior funiculi in mammals, are said to reach the inferior olive. However, these latter connections are by way of collaterals rather than through terminal fibers. It is probable that in fishes the inferior olive likewise receives fibers from the optic tectum. Homologous fibers may run

for mammals in the rubro- or tecto-spinal paths. It has many short connections to the surrounding gray.

Fio. .’il7. The (opopraphic relations exist itiR between tlic inferior olivary nucleus and the cerebellum. Gordon Holmes and Groinger Sicwarl.

R^isuMfi OF THE Phylogenetic Development OF THE Inferior Olive

On comparing the inferior olivary nucleus in those classes of vertebrates in which it is well developed (fishes — particularly plagiostomes — birds, and mammals), it appears that in lower vertebrates only the medial portion of the complex is found, while in higher vertebrates (birds and mammals) the nuclear mass grows out in dorsolateral and ventrolateral directions. The dorsolateral portion forms the dorsal accessory olive. The ventrolateral portion develops into the main mass of the mammalian in

ferior olivary nucleus. The progressive development of the ventrolateral or main inferior olivary mass in mammals is accompanied by the development of the hemispheres of the cerebellum. It develops in the mammalian phylum in a frontocaiidal direction, and in anthropoids and in man extends throughout nearly the vhole length of the olivary complex. Its increase in size corresponds with an inereiU'C in its connections with the cerebellum. Consequently it must be regarded as a dependency of that center, and as concerned in carrying to the cerebellum proprioeeptixe impulses from at least the upper regions of the body by way of spinoolivary .and olivo-cercbellar paths. It is also an area of synapse for ccrebcllo

THE COORDINATING APPARATUS

689

olivary and olivo-spinal fibers. Over this latter system of fiber paths the cerebellum discharges to the motor centers of the cervical cord, to those supplying neck muscles. It is placed under the control of higher centers by a large, socalled thalamo-olivary path. It is connected with its fellow of the opposite side by interolivary fibers, in fishes (and probably in mammals), with the tectum by tecto-olivary bundles, and, in mammals at least, is believed to receive fibers from the nucleus gracilis and the nucleus cuneatus {Mussen, 27, bibliography, p. 1232), and from the red nucleus. Tilney (’27) believed it to be connected by the central tegmental bundle with the nucleus of the mesencephalic root of the

trigeminal nerve. . .

The presence of a medial accessory olivary nucleus in such lower animals as are good swimmers is probably due to the very great development of the trunk and tail musculature in these forms and the accompanying development of a simple reflex mechanism for the body. In Cetaceans the enormous development of the medial accessory olive may be regarded as relate to t eir swimming habits. The greater development of the chief inferior olivary nucleus m primates is a consequence of the finer regulation of the various movenMnts o e ex rem ities. Having carried on a series of experiments on dogs, von ec i erew cone u e that the inferior olivary nucleus plays a r61e m static conduction. ^

in discussing the probable functions of the inferior o move came to the conclusion that it may be concerned in ocu o-cep ^ ments,” where movements of the hands, and ^ °

associated mth movements of the head and eyes. j pnrtpx Insofar as

connections, this nucleus is connected vdth the cerebellar u SaUt is a

its connection is with the cortex of the hemispheres it seen^ probab et^^^^^^^^ relay station, as suggested for afferent m^^^ muscular

spinal cord centers to the neocerebellum, ess ^

coordinations involved in some phase of neokinesis. jq seen to be

The phylogenetic development of ^^y ^'j^^^older vermis and newer related with the differentiation of the cerebel flnnpiiliis has not yet

hemispheres. A connection of the inferior olive thTthl

been pointed out. Tliis lack is in harmony onrl that it is not in con concerned n-ith vestibular and lateral-line tunctrons end that it

nection with the spino-cerebellar systems.

bibliography

■Rttt'R and the Midbkain The Reticular Elements of the Bulb and t

CYCLOSTOMES

TT T 1011 Notes on the anatomy of a cyclostome

Herrick, C. J., and Obenchain, J. . ^ol. 23, p. 635. _

Johnston^ J.‘b7' 1902. The brain of Petromyzom tesf* IL Das Gehim.

Tretjakoff, D. 1909. Das Nen^ensystem von Ammocoe

Arch. f. mikr. Anat., Bd. 74, S. 636.

690 NERVOUS SYSTEMS OF '^'ERTEBRATES AND OF MAN

SELACHIANS

Beccari, N. 1930. I centri tegmentale nell 'asse cerebrale dei Sclacei. ' Arch. Zool. ital. vol. 14, p. 411.

van Hoevell, J. J. L. D. 1911. Remarks on the reticular cells of the oblongata in different vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pt. 2, p. 1047.

TELEOSTS

AnikNs Kappers, C. U. 1920-21. Verglcichende Anatomic des Nerv’ensystcms.

(German edition of present text.) E. F. Bohn, Haarlem.

B.artelmez, G. W. 1915. Alauthner’s cell and the nucleus motorius tegmenti. J. Comp. Neurol., vol. 25, p. 87.

Beccari, N. 1907. Ricerche sullc cellule e fibre del Mauthner e sulle loro connessioni in pesci cd anfibii. (Salmo fario, S. irideus e Salamandrina perspicillata.) Arch, ital. di anat. e di embriol., vol. 6, p. 660.

. 1922. Studi comparati\n sulla struttura del rhombencefalo. Arch. Ital. di

anat. c embriol., vol. 19, p. 122.

Hocke Hoogenboom, K. J. 1929. Das Gchim von Polydon folium Lacep. Jahrb.

f. Morphol. u. mikr. Anat. ; Abt. 2, Zeitschr. f. mikr.-anat. Forschung, Bd. 18, S. 311. Johnston, . I. B. 1901. The brain of Acipenser ; a contribution to the morphology of the vertebrate brain. G. Fischer, Jena.

Tello, F. 1909. Contribucidn al conocimiento del cncdfalo de los Telefistcos. Los miclcos bulbarcs. Trab. d. lab. de invcist. biol. Univ. de Madrid, vol. 7, p. 1.

AMPHIBIANS

AriKns Kappers, C. U. 1920-21. Verglcichende Anatomic des Nen'cnsystcms.

(German edition of pre.'sent tc.vt.) E. F. Bohn, Haarlem.

Beccari, N. 1907. Ricerche sulle cellule e fibre del Mauthner e sullc loro connc.ssioni in pe.sci cd in anfibii. (S.almo fario, S. irideus c Salamandrina perspicillata.) Arch, ital. di anat. c di embriol., vol. 6, p. 600.

Edinoer, L. 1903. Vorlc.sungcn (iber den Ban dor nervosen Ccntralorganc des Mrn=chcn und dcr Thicrc. 7" Aufl., F. C. W. Vogel, Ijcipzig.

Hebrick, C. j. 1914. The medulla oblongata of larv'al Amblystoma. J. Comp. Neurol., vol. 24, p. 343.

. 1930. The medulla oblongata of Ncctunis. J. Comp. Neurol., vol. 50, p. 1.

REPTILES

AriI'.ns Kappers, C. U. 1920-21. Verglcichende Anatomic des Nervensystems.

(Gomian edition of pre.^ent te.xt.) E. F. Bohn, Haarlem.

Biccari, N. 1922. See bibliography for ganoid.s and tcleosts.

. 1922a. I centri tegmentcdel rombenccfalo. Rev. di pat. nen'., vol. 27, p. 1.

. 1922b La natura del ganglio ottico ba.'^.alc o ictomamillarc deH’Iixlinger

i fa^ci di--cendenti genicolare c prctcttnlc cd i loro nuclei intcrstiziali nci rettile. Monitore zool. ital., vol. 33, p. 72.

. 1923 II centro tegnicntale c interstizialc ed altrc formazioni poco note nel

nietencefalo c nel diencefnlo di iin rettile. Arch. ital. di anat, c di embriol., vol. 20,

p .'00

I-.fiiNGi R, L 1903. See bibliogmphy for amphibians.

VAN liojvn.i., J J. L D. 1911. Remarks on the reticular colls of the oblongata in diflfren! vertebnite« Kon. Akad. v. Wetensch. to Amsterdam, Proc. sect, sc., Vol. 13, pt. 2, p. 1057.

THE COORDINATING APPARATUS

691

Hobeh, G. Cahl, and CnosBY, E. C. 1926. On thalamic and tectal nuclei and fiber paths in the brain of the American alligator. ' J. Comp. Neurol., vol. 40, p. 47.

DE Lange, S. J. 1912. The red nucleus in reptiles. Kon. Akad. v. Wetensch. te Amsterdam, Proc. .■^ect. sc., vol. 14, pt. 2, p. 1082.

. 1913. Das Z\vi.schcnhim und das Mittclhim der Reptilien. Folia neuro-biol.,

Bd. 7, S. 67.

Rjim6n y Cajal, S. 1909-11. Histologic du systdme nerveux de I’homme et des vert4brfe. A. Maloine, Paris.

nmns

Bok, S. T. 1915. Die Entwicklung der Hirnnerven und ihrer zentralen Bahnen. Die stimulogene Fibrillation. Folia neuro-biol., Bd. 9, S. 475.

VAN Hoevell, j. j. L. D. 1911. Remarks on the reticular cells of the oblongata in different vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pt. 2, p. 1047.

Hubeu, G. Cahl, and Crosby, E. C. 1929. The nuclei and fiber paths of the avian dienccphalon, with consideration of telencephalic and certain mesencephalic centers

and connections. J. Comp. Neurol., vol. 48, p. 1. , i u ^ •

IL\1!6n y Cajal, S. 1909. Contribucidn al estudio de los ganglios de la substancia reticular del bulbo con algunos detallcs concemientes d los focos motores y vim refiejas bulbarcs y mcseccfdlicas. Trab. d. lab. de invest, biol. Umv. de Madrid,

vol. 7, p. 259. J , ,, ,

. 1909-11. Histologic du systfime nerveux de I’homme et des vertebras.

RENDAii^^IL^'^im'^ Embrj'ologische und morphologische Studien iiber dasZwischenhim beim Huhn. Acta Zool., Bd. 5, S. 241.

MAMMALS

Allen, W. F. 1919. Application of the March! method to the study of the radix

mescncephalica trigemini in the guinea pig. J. Comp, euro • / A cninal

. 1927. Ex-perimcntal anatomical studies on the onpn of the v.sceral bulbo-spinal

pathway in the cat and guinea pig. J. Comp. Neurol., vo . , p- , g

AriKns Iyappers, C. U. 1920-21. Verglcichende Anatomic des Nerv y

(German edition of present text.) E. F. Bohn, aar em. Riickenmark

VON Bechterew. W. 1894. Die Leitungsbahnen im Gehim und Ruckenmark.

CAST^Lm '!:. 1922. Contribute alio studio dei nuclei degli ®

— 1923. Studi sulla struttura c sullo sviluppo del

cobaya. Parte prima. ^/jgj'Jliesenc’efalo Ricerche in Cavia

. 1926. Studi sulla struttura e sullo sviluppo del mesenceta

cobaya. Arch. ital. di anat. e di embrioh, yol. , P- ’ oblongata und der Cramer, A. 1894. Beitrag zur feineren ^Ys Jena.

Briicke mit besondercr Beruchsichtigung des CMerkel u Bonnet), Bd. 4.

Reviewed in Ergeb. d. Anat. u. Entwicklungsgesch. (Merkel

w’ .m The ™cle«. ..d cell

EmSI’L ne„5se. Ce..r..org..e de. Me. schen und der Thiere. Vogel, Leipzig. ..nd ihre oberen Verkniip J^OREL, A. 1877. Untersuchungen fiber die Ha'ibenreg ^ Psychiat.,

fungen im Gehime des Menschen und einiger g Bd. 7, S. 393.

692 NERVOUS SYSTEMS OF VTSRTEBRATES AND OF IVIAN

VAX Gehuchten, a. 1904. Conne.xions centrales du noyau de Deiters et des masses prises voisinc.'!. Le Nevraxe, vol. 6, p. 19.

. 1904. Lc corps rcstiforme. Le Nevraxe, vol. 6, p. 135.

Hatschek, R. 1907. Zur vergleichenden Anatomie des Nucleus ruber Tegmenti. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Oberstciner’s), Bd. 15, S. 89.

Held, H. 1893. Beitriige zur feineren Anatomie des Kleinhirns und des Hirnstammes. Arch. f. Anat. u. Physiol., Anat. Abt., S. 435.

VAN' Hoevell, J. J. L. D. 1911. Remarks on the reticular cells of the oblongata in different vertebrates. Kon. Akad. v. Wetensch. te Amsterdam, Proc. sect, sc., vol. 13, pf. 2, p. 1047.

Huber, G. Carl, and Crosby, E. C. 1929. The somatic and visceral connections of the dicncephalon. Arch. Neurol, and Psychiat., vol. 22, p. 187.

. 1930. (Sec Piersol, G. A. ; Human Anatomy.)

Jacobsohn, L. 1909. Ueber die Kcmc des menschlichen Hirnstammes (der Medulla oblongata, des Pons und des Pedunculus). Neurol. Ccntralbl., Bd. 28, S. 674.

Johnston, J. B. 1909. The radix mescncephalica trigemini. J. Comp. Neurol., vol. 19, p. 593.

K 01 J.N.STA.M.M, 0. 1899. Ueber Ursprungskemc spinnlcr Bnhnen im Himstamm, speciell

ul>cr das Atcmccntrum. Referred to in Neurol. Centralbl., Bd. 18, S. 668.

. 1912. Der nucleus paralcmniscalis inferior als akustischer Reflexkern und als

Glicd dcr zcntralcn Horlcitung (nebst einer Bemerkung fiber den Bechtcreu'sehen Kern und den nucl. lateralis pontis). Arch. f. Ohrenh., Bd. 89, H. 1, S. 59.

voN KOlliker, a. 1891. Dcr feinere Bau dcs verlangcrtcn Marks. Anat, Anz., Bd. 6, S. 427.

. 1896. Handbuch dcr Gewebclchrc dcs Mcnschcn. W. Engelmann, Leipzig,

1889-02.

Kosaka, K. 1912. Zur Fragc dcr physiologischcn Natur der zcrebralen Trigeminuswairzcl. Folia ncuro-biol., Bd. 6, S. 1.

Kunf;, S. 1899. Die normalc und pathologische Stnictur der Zellen an dcr ccrcbralen Wurzel dp.« Nervus trigeminus, die Kreuzungsfrape dcr letztem und der motorischen Trigominuswurzel. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obcrsteincr’s), Bd. 6, R. 1.59.

Lewandow.skv, M. 190^1. Unfersuchungen iiber die Leitunp-sbahnen des Truncus cerebri und ihren Zusammcnhanp mit denen dcr Medulla spinalis und dcs Cortex eerebri. G. Fischer, Jena.

MAitnriio, O. 1903. Mikro.skopi.sch-topopraphischcr Atln.s de.s menschlichen Zcnlralnen.Tn.sy.«tems. F. Dcutickc, Ixiipzip u. Wien. 2'" Aufl., 1910.

May, 1\ . P., AND IloiusLEY, V. 1910. The mesencephalic root of the V"" ncrv'c. Brain, vol. 33, p. 175.

Mei'Nert, T. 1872. Vom Gchimc der Saupetierc. (S. Strieker's Handbuch dcr Ix'hrc und der Gewelien), Bd. 2, S. 694. W. Engelmann, Ixsipzip.

vo.N Mo.vakoiv, C. 1895. Experimentcllc und pafholopisch-anafomischc Untcrsurhuncen iilyr die Haulynrepion, den Schhupel und die Repio subthalamica. Arrh f. IVychiat,, Bd. 27, S. 1, u. S. 386.

. I9W. Der rote Kern, die Hauly und die Repio hypothalamica l>ci cinipen Safipctieren und lyim Men-rhen. Arb. a. d. himanat. In.st. in Ztirich, Bd. 3, S. 49. A!w) Bd 4, S 103, 1910.

Mo-ev. AT. 1927. Exfyrimental invc.^fipation.s on the cerebellum. Brain, vol. 50, p 313

till) i'.>,Ti.iN'j'T!, II. 1893. .•\nleitunp lyim Studiurn des. Baue.s der nervosen Ccnlntlorv'tnf. Fn’neh tnin^lnlinn. Dr. Corofnne,

1912. Anl'itune lyirn Studiurn de« Baue.s der neno'yen Zenf ralorpane im pf «nndn und knmken Zu'-taride. 5*' Aufl., F. Deuticke, lyipzip u. Wien.

THE COORDINATING APPARATUS

693

Papez, J. W. 1926. Reticulospinal tracts in the cat. Marchi method. J. Comp. Neurol., vol. 41, p. 365.

PiERSOL, G. A. 1930. Human anatomy. 9* edition revised under the supervision of G. Carl Huber. J. B. Lippincott Co., Philadelphia and London.

Probst, M. 1902. Zur Anatomie und Physiologic des Kleinhims. Arch. f. Psychiat., Bd. 35, S. 692.

. 1902a. Ueber centripetale Ruckenmarksfasem zur Substantia reticulans.

Monatschr. f. Psychiat. u. Neurol., Bd. 11, S. 3.

. 1902b. Experimentelle Untersuchungen iiber die Anatomie und Physiologie der

Leitungsbahnen des Gehirnstammes. Arch. f. Anat. u. Physiol., Anat. Abt., Supplement-Band, S. 147.

Ram6n y Cajal, S. 1909-11. Histologic du systeme nerveux de 1 homme et des

vert^br4s. A. Maloine, Paris. ^ d a j

Ranson, S. W. 1931. The anatomy of the nervous system. 4*'- edition, W. B. Saunders

Co., Philadelphia and London. „ . t i

Rioch, D. M. 1929. Studies on the diencephalon of Carnivora. I. Ihe nuclear

configuration of the thalamus, epithalamus, and hypothalamus of the dog and cat.

J. Comp. Neurol., vol. 49, p. 1. .r, u -n i

Schwalbe, G. 1881. Lehrbuch der Neurologie. E. Besold, Erlangen. . • Sheinin, j. j. 1930. Typing of the cells of the mesencephalic nucleus of the tngemmal nerve in the dog, based on Nissl granule arrangement. J. omp. euro • »

Sterzi,^g! 1915. Anatomia del sistema nervoso centrale dell’uomo. A. Draghi,

Padova. Bd. 1, 1914, Bd. 2, 1915. a t, r Annt Bd 53

Terterjanz, M. 1899. Die obere Trigeminuswurzel. Arch. f. mikr. Anat., B . ,

VoRis, H., AND Hoerr, N. 1932. The hindbrain of the opossum, Didelphis virgmiana.

-»t of the fifth .en., A .o„pnr.,ive

anatomical study. J. Comp. Neurol., P; . ' y novniix masticateur

Willems, E. 1911. Localisation motrice et kinesth4si^e. 12 n. 1.

et m4sencephalique du trijumeau chez le lapin. pxDerimental researches

Winkler, C., a^d Potter, A. 1911. An anatomical guide to expenmental

on the rabbit’s brain. W. Versluys, Amsterdam brain.

. 1914. An anatomical guide to experiment

Yos"oJrX“BeTwg » Konolols dor zenWoo YosllhoI.risb.hn. Folio

und

Bardeleben’s Handbuch der Anatomie des Menschen), G. k .seller,

1903 ; II Abt. 1 T., 1913 ; II Abt. 2 T. 1920.

Literature Concerned with the Phi'logenetic Parts

Olivary Nucleus and the Significance of Its Several

CYCLOSTOMES

Johnston, J. B. 1902. The brain of Petromyzon. J. Comp. Eeurol.,

plagiostomes and holocephali

ARikNs Kappers, C. U. 1906. The structure of the teleostean an J. Comp. Neurol., vol. 16, p. 1

694 NERVOUS SYSTEMS OF VERTEBRATES AND OF MAN

Ahiens Kappers, C. U., and Carpenter, F. W. 1911. Das Gehim von Chimaera monstrosa. Folia neuro-bioL, Bd. 5, S. 127.

Edinger, L. 1908. Vorlesungen iiber den Ban der nervosen Centralorgane des Menschen und der Thiere. 7*® Aufl., F. C. W. Vogel, Leipzig.

Kooy, F. H. 1917. The inferior olive in vertebrates. Folia neuro-biol., Bd. 10, S. 205. Sterzi, G. 1909. II sistema nervoso centrale dei vertebrati. Selaci. A. Draghi, Padova.

ganoids and teleosts

-Burr, H. S. 1928. The central nervous system of Orthagoriscus mola. J. Comp. Neurol., vol. 45, p. 33.

Johnston, J. B. 1901. The brain of Acipenser ; a contribution to the morphology of the vertebrate brain. G. Fischer, Jena.

Kooy, F. H. 1917. The inferior olive in vertebrates. Folia neuro-biol., Bd. 10, S. 205. Mayser, P. 1881. Vergleichend anatomische Studien iiber das Gehim der Knochenfische mit besonderer Beriicksichtigung der Cyprinoiden. Zeitschr. f. wissensch. Zool., Bd. 36, S. 259.

reptiles

Kooy, F. H. 1917. The inferior olive in vertebrates. Folia neuro-biol., Bd. 10, S. 205. DE Lange, S. J. 1917. Das Hinterhim, das Nachhim und das Ruckenmark der Reptilien. Folia neuro-biol., Bd. 10, S. 385.

birds

Kooy, F. H. 1917, The inferior olive in vertebrates. Folia neuro-biol., Bd. 10, S. 205. Groebbels, F. 1923. Die untere Olive der Vogel. Anat. Anz., Bd. 56, S. 296. Shimazono, j. 1912. Das Kleinhim der Vogel. Arch. f. mikr. Anat., Bd. 80, S. 397. Williams, E. W. 1909. Vergleichend anatomische Studien iiber den Bau und die Bedeuting der Oliva inferior der Saugetiere und Vogel. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s), Bd. 17, S. 118.

Yoshimura, K. 1910. Experimentelle und vergleichend anatomische Untersuchungen iiber die untere Olive der Vogel. Arb. a. d. neurol. Inst. a. d. Wien. Univ., Bd. 18, S. 46.

MAMMALS

Ariens Kappers, C. U. 1920-21. Vergleichende Anatomie des Nervensystems.

(German edition of present text.) E. F. Bohn, Haarlem.

Brouwer, B. 1913. Ueber Hemiatrophia neocerebellaris. Arch. f. Psychiat., Bd. 51, S. 539.

. 1915. Anatomische Untersuchung iiber das Kleinhim des Menschen. Psychiat.

en neurol. Bl., Amsterdam, vol. 19, p. 104.

. 1918. Klinisch-anatomische Untersuchungen iiber den Oculomotoriuskem.

Zeitschr. f. d. ges. Neurol, u. Psychiat., Bd. 40, S. 152.

Brouwer, B., and Coenen, L. 1919. Ueber die Oliva inferior, J. f, Psychol, u. Neurol., Bd. 25, Heft 2, S. 52.

Brunner, H. 1917. Beitrage zur Kenntnis der unteren Olive bei den Saugetieren. Arb. a. d. neurol. Inst. a. d. Wien. Univ. (Obersteiner’s) Bd. 22, S. 113.

. 1918. Bemerkungen zum Aufbau des Himstammes der Cetacean mit besonderer

Beriicksichtigung der unteren Olive. J. f. Psychol, u. Neurol., Bd. 24, S. 138. • Essick, C. R. 1907. The corpus ponto-bulbare — a hitherto undescribed nuclear mass in the human hindbrain. Aon. J. Anat., vol. 7, p. 119.

. 1912. The development of the nuclei pontis and the nucleus arcuatus in man.

Am. J. Anat., vol. 13, p. 25.

THE COORDINATING APPARATUS

695

Goldstein, K. 1910. Ueber die aufsteigende Degeneration nach Querschnittsunterbrechung des Ruckenmarks (Tractus spino-cerebellaris posterior, Tractus spinoolivaris, Tractus spino-thalamicus). Neurol. Centrabl., Bd. 29, S. 898.

His, W. 1889. Ueber die Entwickelung des Riechlappens und des Riechganglions und liber diejenige des verlangerten Markes. Verhandl. d. anat. Gesellsch., Bd. 3, S. 63.

. 1891. Die Entwicklung des menschlichen Rautenhorns vom Ende des 1. bis

Beginn des 3. Monats. I Verlangertes Mark. Abb. d. math.-phys. Kl. d. Konig. Sachs. Gesellsch. d. Wiss. Bd. 17, S. 1.

Holmes, G., and Stewart, G. T. 1908. On the connection of the inferior olives with the cerebellum in man. Brain, vol. 31, p. 125.

Jelgersma, G. 1899. Ueber den Bau des Saiigethiergehirns. Morphol. Jahrb., Bd. 15, S. 61.

Jenkins, G. B. 1916. A study of the morphology of the inferior olive. Anat. Rec., vol. 10, p. 317.

Kankeleit, 0. 1913. Zur vergleichenden Morphologic der unteren Saugetierolive

(mit Bemerkungen iiber Kerne in der Olivenperipherie). Inaugural Dissertation. Berlin. Arch. f. Anat. u. Physiol., Anat. Abt., 1913, S. 1. . ■

VON Kolliker, a. 1901. Die Medulla oblongata und die Vierhiigelgegend von Ornithorhynchus und Echidna. W. Engelmann, Leipzig.

Koot, F. H. 1917. The inferior olive in vertebrates. Folia neuro-biol., Bd. 10,

S 205

. 1920. The inferior olive in Cetacea. Folia neuro-biol., Bd. 11, S. 647.

Ram6n t Cajal, S. 1904. Textura del sistema nervioso del hombre y de los vertebrados. Vol. 2. N. Moya, Madrid.

. 1909-11. Histologic du syst^me nerveux de 1 homme et des vert^br^s.

SABm' R R°'° I’gS"^ An atlas of the medulla and midbrain.

Streeter, G. L. 1912. The development of the nervous sptem.

embryology, vol. II, Keibel, F., and Mall, F. P., J. B. Lippmco .,

TiLN^i^ R "mV. The brain stem of Tarsius. A critical comparison with other pn mates. J. Comp. Neurol., vol. 43, p. 371. . nnrtion of

Weed, L. H. 1914. A reconstruction of the nuclear masses the lower portion

the human brain stem. Publ. Carnegie Inst., Was mg on, ' Marsupialier

Ziehen, T. 1908. Das Zentralnervensystem der Monotremen u

(Zweiter Teil, Microscopische Anatomic. Semon’s Forschungsreisen,

im Gehim von Echidna und Omithorhynchus). Semons worsen g

— ' Makroskopische and mlkro.kopi^ha An.lome des ZenW nervensystems. G. Fischer, Jena. -o ,ip 5 flehims. (In Barde — . 1920. Makroscopische und mikroscopische Anatom edes^G^^^^

leben’s Handbuch der Anatomie des Menschen),

Talk:Book - The comparative anatomy of the nervous system of vertebrates including man 1

Contents

The comparative anatomy of the nervous system of vertebrates including man

Introduction

Contents

CHAPTER I THE EVOLUTION AND MORPHOLOGY OF NERVOUS ELEMENTS

Chapter II The Comparative Anatomy Of The Spinal Cord

Chapter III The Medulla Oblongata