Talk:Book - Manual of Human Embryology 14

mT

mm*

MANUAL

OF

Human Embryology

WRITTEN BY Charles R. Bardeen, Madison, Wis.; Herbert M. Evans, Baltimi Md. ; Walter Felix, Zurich; Otto Grosser, Prague; Franz Keibei Freiburg i.Br. ; Frederic T. Lewis, Boston, Mass. ; Warren H. Lewis, Baltimore, Md. ; J. Plavfair McMurrich, Toronto; Franklin P. Mall, Baltimore, Md. ; Charles S. Minot, Boston, Mass.; Felix Pinkus, Berlin; Florence R. Sabin, Baltimore, Md. ; George L. Streeter, Ann Arbor Mich. ; Julius Tandler, Vienna ; Emil Zuckerkandl, Vienna.

EDITED BY

FRANZ KEIBEL Professor in the University of" Freiburg i.Br.

and FRANKLIN P. MALL Professor of Anatomy in the Johns Hopkins University. Baltimore, U.S.A.

IN TWO VOLUMES

VOLUME II.

With 658 Illustrations

PHILADELPHIA Of LONDON J. B. LIPP1NCOTT COMPANY

1012

Copyright, 1912 By J. B. Lippincott Company

\

Printed by J. B. Lippincott Company The Washington Square Press, Philadelphia, U. S. A.

PREFACE

The second volume of this Manual makes its appearance almost half a year later than the editors had hoped. Explanations of the delay are hardly necessary; they are self-evident in the case of a book in which a large number of authors participate. The delay would, however, have been greater had nol Professor Tandler and Dr. Evans undertaken the sections on the development of the heart and of the blood-vessels, which had originally been placed in other hands. To these two collaborators the editors are under special obligations. The account of the development of the sense-organs, for which also other plans were made, had to be undertaken by one of the editors, Keibel, an arrangement which precluded the treatment of the subject entirely on the basis of personal observation. A number of the contributions to the volume have been completed for a considerable time (some for more than a year), and consequently it has been impossible to include in them all the references to the most recent literature (since the beginning of 1910).

The editors regret that this second volume has considerably exceeded the limits originally set for it. If it had been written by a single hand, a greater condensation would certainly have resulted in many chapters, but under the circumstances this has not been possible, notwithstanding the earnest endeavors of the editors.

On the whole, however, the editors feel that they have reason both to congratulate themselves on the completion of the work, which, in spite of many minor defects, is undoubtedly an important one, and to hope that it will give further impetus to the study of human embryology.

In conclusion the editors wish to express their heartiest thanks to all those who have assisted in the completion of the work, to the collaborators, to Professor J. P. McMurrich for the excellent translation of the chapters by Zuckerkandl, Keibel, Tandler, and Felix, and, above all, to the publishers S. Hirzel and the J. B. Lippincott Company. They have done everything in their power to make the work a success, and, especially, have made possible the illustration of the text by so many and such excellent figures.

Franz Keibel. Franklin P. Mall.

CONTENTS

CHAPTER XIV. By G. L. Streeter.

PAGE The Development of the Nervous System 1 1 . The Histogenesis of Nervous Tissue ] 2. The Development of the Central Nervous Sj'stem 29 (a) The Development of Central Nervous System during the First Month. . 31 (b) The Central Nervous System at the End of the First Month 40 (c) The Spinal Cord from the End of the First Month to the Completion of its Development 47 (d) The Development of the Hindbrain from the End of the First Month-. . 59 (e) The Development of the Midbrain 74 (f) The Development of the Diencephalon 77 (g) The Development of the Telencephalon 82 3. The Peripheral Nervous System 117 (a) The Spinal Nerves 117 (b) The Cervical Plexus 120 (c) The Nerves of the Arm and Leg 122 (d) The Cerebral Nerves 128 (e) The Nerves of the Special Sense Organs 129 (f) The Somatic Motor Group 132 (g) The Visceral Group 136 4. The Sympathetic Nervous System 144 CHAPTER XV.

By E. Zuckerkandl.

The Development of the Chromaffin Organs and of the Suprarenal Bodies 157 The Suprarenal Glands 170 The Development of the Accessory Suprarenal Glands ' 17" CHAPTER XVI. By F. Keibel.

The Development of the Sense-Organs 180 General Considerations 180 The Touch-cells 180 The Epibranchial Sense-organs 182 The Gustatory Organ L82 The Olfactory Organ 188 The Development of the Eye 218 The Development of the Ear CHAPTER XVII. By O. Grosser, F. T. Lewis, and J. P. McMfrrich.

The Development of the Digestive Tract and of the Organs of Respiration 291 Introduction 291 The Early Development of the Entodermal Tract and the Formation of its Subdivisions v 295

vi CONTENTS.

The Mouth and its Organs 335 The Development of the (Esophagus 355' The Development of the Stomach 368 The Development of the Small Intestine 381 The Development of the Large Intestine 393 The Development of the Liver 403 The Development of the Pancreas 429 The Development of the Pharynx and of the Respiratory Apparatus 446 A. The Pharynx and its Derivatives 446 I. General Morphology of the Pharyngeal Pouches 447 II. Differentiation of the Pharyngeal Pouches; the Second Pharyngeal Pouch and the Tonsils 457 III. The Third, Fourth, and Fifth Pharyngeal Pouches; the Branchio genic Organs 460 B. The Development of the Respiratory Apparatus 473 I. The Earliest Anlage 473 II. The Trachea 475 III. The Larynx 476 IV. The Lungs 482 CHAPTER XVIII.

By C. S. Minot, H. M. Evans, J. Tandler, and F. R. Sabin.

The Development of the Blood, the Vascular System, and the Spleen 498 I. The Formation of the Angioblast and the Development of the Blood .... 498 II. The Development of the Heart 534 III. The Development of the Blood-vascular System 570 1. General 570 2. The Special Development of the Blood-vessels. 587 A. Origin of the Blood-vascular System 587 B. Description of the Blood-vascular System in a Series of Human Embryos from the Youngest Stages up to a Length of 7 mm. 588 C. The Arteries 613 D. The Development of the Veins 669 IV. The Development of the Lymphatic System 709 V. The Development of the Spleen 745

CHAPTER XIX.

By W Felix.

Tbe Development of the Urinogenital Organs 752 I. The Development of the Excretory Glands and their Ducts 753 Introduction 753 Differentiation of the Mesoderm 754 The Development of the Pronephros 759 The Urogenital Fold 783 The Mesonephros 796 The Development of the Metanephros 831 The Development of the Ureter, the Primitive Renal Pelvis, and the System of Collecting Tubules 833 The MetanephrogenieTiesue during the Development of the Collecting Tubules 842 The Development of the Uriniferous Tubules from the Metanephrogenic ' Tissue 846

CONTENTS. vii The Definitive Renal Pelvis and the Reduction of the Collecting Tubules 859 The Formation of Cortex and Medulla — Medullar}- Rays — Papilla*. . . . 861 Relation between the Right and Left Kidney 863 Change of Position of the Kidney 863 Vessels of the Kidney 864 Capsule of the Kidney 86f) Later Development of the Ureter 8f>5 Malformations of the Kidney 867 The Phylogenetic Development of the Metanephros 867 The Function of the Mesonephros 868 Cloaca, Bladder, Urethra, and Urogenital Sinus 869 The Separation of the Bladder and the Urethra 878 Later Development of the Bladder 879 II. The Development of the Reproductive Glands and their Duct.- 881 Introduction 881 The Genital Cells 882 Development of the Indifferent Reproductive Glands 885 The Differentiation of the Reproductive Glands 890 Transformation of the Indifferent Reproductive Gland into the Testis. 891 Malformations of the Testis 897 "^^Transformation of the Reproductive Gland into the Ovary 897 Malformations of the Ovary 909 Development of the Blood-vessels of Both Reproductive Glands 909 Comparison of Testis and Ovary 910 '^-Development of the Female Ducts 911 ^Formation of the Utero- vaginal Canal 915 •^Formation of the Wall of the Utero-vaginal Canal 916 -^Development of the External Form of the Female Uterus 918 Development of the External Form of the Male Uterus 918 Formation of the Definite Wall of the Tubes 919 Transformation of the Tubar Portion of the Uterus and of the Utero-vaginal Canal into the Uterus and Vagina 919 Development of the Vagina 92 1 Development of the Uterine Wall 024 Development of the Musculature of the Vagina and Uterus 926 Growth of the Uterus in the Postfetal Period 928 Degeneration of the Tube and Utero-va^inal Canal in the Male Embryo 929 Inhibitions of the Development of the Uterus and Vagina 930 III. The Urogenital Union 937 The Further Differentiation in the Male Sex 937 Degeneration of the Urogenital Union and of the Primary Excretory Duct in the Female 939 The Ligaments of the Reproductive Glands 940 Development of the Ligamentum Ovarii Proprium, the Ligamentum Uteri Rotundum, and the Chorda Gubernaculi 942 IV. Development of the External Genitalia 947 The Indifferent Phallus 949 Sexual Differentiation Development of the Penis and Scrotum 951 Development of the Clitoris and of the Labia Majora and Minora .... 954 Homologies of the Male and Female External Genitalia 958 Further Development of the Urogenital Sinus 960 Malformations of the Urogenital Sinus 964 Glands of the Urogenital Sinus 965 Development of the Bulbo-Urethral (Cowper's) and the Vestibular (Bartholin's) Glands 966 Development of the Small Sinus Glands 967 Descensus Testiculorum , 967 Formation of the Saecus Vaginalis 969 Diagrammatic Representation of the Fate of the Mesonephros, Primary Kxcretory Duct, and Miillerian Duct in both Sexes 973 CHAPTER XX.

By F. Keibel.

The Interdependence of the Various Developmental Processes 980 Subject Index 1003 Index of Authors 1023

HANDBOOK of HUMAN EMBRYOLOGY

XIV. THE DEVELOPMENT OF THE NERVOUS SYSTEM.

By GEORGE L. STREETER, Ann Arbor, Mich.

The entire nervous system, as will presently be described, has a common origin. It is derived from an anlage that is very early differentiated from the ectoderm as the neural or medullary plate. The description of its development will be taken up under the following headings: 1, the histogenesis of the nervous tissues; 2, the development of the central nervous system; 3, the development of the peripheral nervous system; and, 4, the development of the sympathetic nervous system.

I. HISTOGENESIS OF NERVOUS TISSUE.

In Chapters III, IV, and V the segmentation of the ovum and the grouping of its cells in the form of three germ layers have already been described. It has also been described how a portion of the ectoderm becomes thickened and differentiated into the elongated axial neural plate and neural groove, and how eventually the elevated edges of this groove come together and fuse, thereby converting it into the neural tube. It is the anterior portion of this neural tube that is enlarged and converted by constrictions into the three primary cerebral vesicles from the thickened walls of which the brain is derived. The caudal portion of the neural tube remains nearly uniform in diameter and forms the spinal cord. The lumen of the caudal portion becomes converted into the central canal of the cord, and the enlarged cavities of the vesi cles of the anterior part of the tube form the several ventricles of the brain. The nerves which serve to connect the central nervous system with the periphery either sprout out from the wall of the neural tube or from the ectodermal cellular crest that is attached along its dorsal seam. Thus, if we except the mesoderms Vol. II.— 1 1

2 HUMAN EMBRYOLOGY.

elements that subsequently penetrate the neural tube, accompanying the invasion of blood-vessels into its substance, we may say that all the essential nervous tissues of the body are derived from the thickening and differentiation of the walls of the neural tube and its ganglionic crest, and hence are ectodermal.

Development of the Walls of the Neural Tube.

In the absence of suitable human material our information regarding the changes that occur early in the differentiation of

mli

B

g mli

1 P

Fig. 1. — Three early stages in the development of the wall of the neural tube, showing the conversion of the single layer of discrete cells into a richly nucleated syncytial framework. A, medullary plate of embryo rabbit just before closure of neural tube; B, similar section of 5 mm. pig embryo just after closure of neural tube; C, wall of neural tube of 10 mm. pig embryo; a, ependymal layer; g, germinal cell; m, marginal layer; mle and mli, external and internal limiting membranes; n, mantle or nuclear layer; p, mesoderm. (After Hardesty.) the wall of the neural tube is based on the chick, rabbit, and pig, notably through the investigations of His, Cajal, Schaper, and more recently Hardesty. The essential feature in this development, as can be seen in Fig. 1, consists in the conversion of the original single layer of cuboidal ectodermal cells into a thick wall whose elements are arranged in the form of three definite layers or zones. These three zones constitute respectively the anlages of the ependyma, the gray substance, and the white substance, the individual cells being accordingly differentiated into supporting cells and nerve-cells proper.

DEVELOPMENT OF THE NERVOUS SYSTEM. 3 In Fig. 1, A, is shown the single layer of ectodermal cells which constitutes the original nenral plate. It is to be noted that they consist of individual cells, distinctly outlined and definitely arranged. During the closure of the neural tube they proliferate and the cell boundaries partially disappear, and there thus results a thick wall of fused cells, a compact nucleated protoplasmic syncytium as seen in Fig. 1, B. As the growth continues the syncytium expands into a looser framework or myelospongium and its outer and inner margins are condensed in a continuous sheet to form the external and internal limiting membranes, as seen in Fig. 2, which is taken from a human embryo. In Fig. 1, C, which shows the appearance at the end of the first month, the myelospongium consists of radially arranged cellular strands still united in a

M. Iimi

tans «tJS mk

&

--M. limitans int.

Fio. 2. — Wall of the neurfl tube in a human embryo about two weeks old, showing its syncytial character. This stage is between that of B and C, Fig. 1. (After His.) syncytium by numerous branching processes. Further examination shows that, owing to the grouping of the nuclei, the wall may be subdivided into three primary zones or layers: (1) the ventricular or ependymal layer; (2) the intermediate or nuclear or mantle layer; and (3) the outer or non-nuclear or marginal layer (Randschleier). It may be added that the term nuclear layer is applied in the early stages, while the boundary between them is indistinct, both to the ependymal and mantle layers, thus contrast ing them to the marginal or non-nuclear layer. In this sense one can say that the wall at first consists of two layers, nuclear layer and non-nuclear layer, and that later the former or nuclear layer becomes differentiated into the ependymal and mantle layers. Here the term nuclear layer is used as synonymous with mantle layer, which is its chief derivative. The ependymal layer consists of a single layer of elongated nuelei connected with the internal limiting

4 HUMAN EMBRYOLOGY.

membrane by strong protoplasmic processes. Among these are seen prominent mitotic nuclei, the so-called germinal cells shown in Fig. 1, g, and Fig. 3. According to Hardesty, the germinal cells are found in greatest number in embryos (pig) between 10 and 20 mm. long, and gradually disappear, ceasing altogether at 40 mm. The nuclear or mantle layer consists of nucleated branching strands of the myelospongium forming a radially arranged protoplasmic framework. Rapidly proliferating nuclei are embedded in the strands and about them the protoplasm is becoming more condensed and granular. Mitotic nuclei are frequently found in this layer after the disappearance of the germinal cells of the ependymal layer, and it is suggested by Schaper

Ependymal Cells

Q .© Germinal Cells

6 Indifferent Cells

Mitotic Indifferent Cells

^, ^S CGB G „°c£<^ oOo p Q « 

Neuroglia Cells Neuroblasts Fig. 3. — Diagram showing the differentiation of the cells of the wall of the neural tube and the theoretica derivation of the ependymal cells, neuroglia cells, and neuroblasts. (After Schaper.) (1897) that they are a proliferation phenomenon: Those that persist until late stages are regarded by him as material for regeneration processes. Whether the proliferation of the nuclei of the mantle layer is always mitotic or whether it is to some extent amitotic remains to be determined. The mantle layer may be considered as the anlage of all gray substance. The marginal layer forms the anlage of the white substance, and through its meshes pass all the principal longitudinal fibre tracts of the cord and brain. For a considerable period it remains devoid of nuclei and consists simply of the network of the myelospongium.

The further steps by which this comparatively simple wall of three primary layers becomes subsequently converted into the complicated spinal cord and brain will be considered later. We will only consider at this time the histogenesis of its elements.

The cells, which in Fig. 1, C, are still fused in a common framework, gradually in their further differentiation undergo extensive change in form and position, and soon it is possible to class them

DEVELOPMENT OF THE NERVOUS SYSTEM. 5 into two groups: (1) the spongioblasts which form the supporting framework or the neuroglia tissues; and (2) the neuroblasts or true ganglion-cells which come to lie within the meshes of the former. Up to the time the two groups can be distinguished from each other they are spoken of as indifferent cells (Schaper).

The differentiation into neuroblasts and spongioblasts does not occur in all cells simultaneously, some cells being more precocious than others; so that in a given section one may recognize well-defined neuroblasts, or spongioblasts, among other cells that are still in the indifferent stage, as is schematically shown in Fig. 3. In general, the spongioblasts develop somewhat in advance of the neuroblasts, and we find in some sections a fairly complete spongioblastic framework having as yet no differentiated nerve cells or fibres. In their further development the spongioblasts and neuroblasts will be treated separately.

Development of the Neuroglia Framework.

Embraced under the term neuroglia we must distinguish (a) ependymal cells, (b) neuroglia cells, and (c) neuroglia fibres, all of which elements serve as a supporting framework for the central nervous system, and all of which are differentiated products of the spongioblastic syncytium already referred to.

It was seen in Fig. 1, C, how the spongioblasts formed a radial framework traversing the entire wall of the neural tube. As the wall thickens the radial arrangement becomes still more marked and the spongioblastic strands become drawn out. These strands take the precipitate intensely in silver methods and form the characteristic G-olgi picture seen in Fig. 4. From the radial strands numerous processes extend laterally and unite the whole in a complete framework. At the junction of the nuclear and marginal layers there exists a close felt-work of these lateral processes which is supposed to prevent temporarily the invasion of nuclei into the latter layer (Hardesty, 1904). A distinct group of spongioblasts maintain their radial arrangement near the ventricular border and eventually constitute the ependymal cells of the adult. In the remainder of the wall the radial arrangement is finally obliterated, which is due to the proliferation and growth of the neuroblasts, the spaces in the framework being determined by the shape of the elements that it supports. It is this portion of the framework that furnishes the neuroglia cell- proper.

The ependymal cells represent the most primitive form of neuroglia, such as is found prevailing in certain lower forms. They are characterized by their radial arrangement and proximity to the lumen of the neural tube. Through their continuity witli the general spongioblastic framework they extend through the whole thickness of the wall, from the internal to the external limiting membrane, as shown in Fig. 1, C. As the ependymal cells proliferate and accumulate more body-protoplasm they come to form a compact mass of deeply staining columnar cells, constituting the lining membrane of the central canal of the cord and the ventricles of the brain. Their connection with the remainder of the neuroglia framework is maintained by processes which extend from the basal portion of the cells a variable distance into the underlying parts.

In the region of the anterior median fissure of the cord and the median raphe of the hind-brain, corresponding to the Bodenplatte of His, the neuroglia maintains its primitive ependymal type of simple radial fibres extending from the lumen to the sur

Fig. 4. — Development of neuroglia framework. Combined drawing, after Golgi and Benda methods, showing section of spinal cord of embryo pig 30 mm. long. It can be seen that the wealth of communicating processes of the neuroglia cells is not shown in the Golgi portion, ep, ependymal layer; m, marginal layer; n mantle or nuclear layer; p pia mater. (After Hardesty.) face of the tube. It is this region that is traversed by the fibres of the anterior white commissure of the cord and the transverse arcuate fibres of the hind-brain. The persistence of this simple type of neuroglia may be explained by the absence of any mantle or nuclear layer with its consequent complications at this place. In a similar way in the posterior median region of the cord there is formed the posterior median septum, consisting of persistent radial neuroglia fibres extending from the lumen to the surface of the cord. Here, however, the conditions are modified by the partial fusion of the walls of the central canal and the fusion of the posterior funiculi, so that instead of a deep fissure we have in the end only a shallow groove. The development of the posterior median septum will be described in detail in the section on the development of the spinal cord.

DEVELOPMENT OF THE NERVOUS SYSTEM. 7 Thus in the adult we find that the ependymal neuroglia is persistent only as septa in the anterior and posterior median planes of the nervous system and as a lining membrane for its central canal and ventricles. Otherwise the whole supporting framework of the nervous system consists of neuroglia cells proper and the neuroglia fibres developed in connection with them, that is if we do not regard the blood-vessels and the accompanying mesodermal tissue, which subsequently enter the neural tube, as a true supporting framework.

The neuroglia cells are differentiated somewhat later than the ependymal cells. They can be traced back to the time when they constituted units in the primary spongioblastic syncytium. As the extension and branching of the latter continues, its component cells can be seen to proliferate and develop additional protoplasm. They at the same time become moulded into shape by the growing nerve cells and fibres which they enmesh. Due to this moulding process there result the characteristic much-branched cells that we know as spider-cells, glia-cells, astrocytes, or Deiters's cells, as shown in Fig. 5. Instead of the simple spongioblastic framework seen in the earlier stages, we now have a dense felt-work produced by the anastomosing branches of these cells, intimately intertwining between the developing nerve-cells and their processes.

Fio. 5. — Combined drawings, after Golgi and Benda methods, of the spinal cord of fetal pig, 20 cm. long, showing syncytial character of neuroglia framework and the first appearance of neuroglia fibres. o, neuroglia cells after the Benda method; a', similar cell after the Golgi method; e and /, neuroglia fibres beginning to take the neuroglia stain; b, pseudocell due to staining of a portion of the syncytium such as seen at b; «, seal-ring cells. (After Hardesty.)

The variation in the shape of the individual neuroglia cells is determined by the nature of the nerve tissue supported, and consequently there are different types described for the cerebral cortex, the cerebellum, and the white and gray substance of the cord. Such variations involve the shape and size of the cell body as well as the character of their cell processes. The earlier conceptions of neuroglia cells were based on silver precipitation methods (Golgi) which failed to reveal the true wealth of their anastomosing branches, and there thus existed a false impression of neuroglia as consisting of scattered and independent cells.

Fig. 6. — Section of spinal cord of suckling pig of two weeks, showing fully developed neuroglia fibre and fibres in the process of transformation, a and c, early stages of neuroglia cells, multinucleated protoplasmic masses; d, g, h, and k, stages of karyolysis which many of the free neuroglia cells undergo; s, sealring cell. (After Hardesty.)

The neuroglia fibres constitute the third element of the supporting tissue, and they resemble in their process of development the fibres in white fibrous connective tissue. They do not appear until late in uterine life. According to Hardesty (1904), they can first be recognized in pig embryos between 16 and 20 cm. long, and from then on gradually increase in number until after birth. There is some evidence that the process of differentiation continues even into adult life.

It was formerly supposed (G-olgi, v. Lenhossek, Erik Miiller, and others) that neuroglia cells existed as separate units and that neuroglia fibres were simply the modified processes of these cells. With the development of the differential neuroglia stains it soon became apparent that this was not the case. From Weigert 's studies (1895) on human tissues, followed by Huber's (1903) studies on several vertebrate classes, it was shown that neuroglia fibres are not to be regarded as processes of the neuroglia cells, but rather as fibres having specific physico-chemical properties which lie in close contact with or actually embedded in the peripheral layer of the protoplasm of the neuroglia cells from which they have become differentiated, as is shown in Fig. 7. The fibres are not

Fig. 7. — Neuroglia fibres in adult human spinal cord, showing their relation to 'the protoplasm of the neuroglia cell and its processes. (After Huber.) interrupted by the protoplasm of the cells, and a single fibre may even be continuous from one cell to another.

To understand properly the manner in which these fibres form, one should keep in mind the syncytial character of the neuroglia framework. This has been especially emphasized by Hardesty (1904), on whose studies the description of this development is based. It has already been seen how the protoplasm of the syncytium in its development tends to accumulate about the nuclei and thus forms masses which we have described above as neuroglia cells. The separate nucleated masses or cells remain more or less continuous with each other by means of attenuated branching portions of the syncytium. In these attenuated portions at a certain period (pigs 16 to 20 cm. long) it can be seen that the protoplasm is becoming more condensed, stains less deeply, and is beginning to show fibrillation. The same process can be recognized in the peripheral portions of the nucleated masses or cells. The fibrillation continues to become more marked, and eventually distinctly outlined individual fibres of varying length can be recognized as lying in the attenuated portions of the syncytial protoplasm and passing through the "domain" of one or more nuclei, as shown in Fig. 5. Whether all the protoplasm in the attenuated portions is consumed in the process of fibrillation, or whether some remains as a partial coating of the fibre is not yet clearly determined.

The final step in their development is the chemical transformation that the fibres undergo, owing to which they exhibit a blue reaction when stained with special neuroglia stains. When that is attained they present the characteristic picture seen in Figs. 6 and 7. A portion of a fibre may give this reaction before the differentiation of the remainder of the fibre is complete. Sometimes a fibre ripens in interrupted areas along its course and hence temporarily appears as a row of fine dots. By the time the chemical transformation is completed the fibres have attained a size about the same as found in the adult. There is some variation in the distribution and form of the neuroglia fibres, depending on whether they are located in the region of gray matter or white matter or in the region of ependymal cells; the essential points in their development, however, are the same in all regions.

Those writers {e.g., Rubaschkin, 1904) who consider neuroglia cells as separate and independent structures describe the neuroglia fibres as modified cell processes. A special cell is also distinguished from which they are derived (gliogenetic cells), which forms an intermediate stage between the spongioblast and neuroglia cell proper. Several fibres may be derived from one process, and eventually they detach themselves from the cell and lie free in the tissue.

Development of Neuroblasts and Motor Nerves.

We have seen above how the cells forming the wall of the neural tube in the early stages are fused in a common syncytial framework, and how the constituent cells of this framework gradually differentiate themselves into spongioblasts and neuroblasts. The former maintain their syncytial arrangement relatively late and continue even in the adult to show permanent traces of that condition. The neuroblasts, however, toward the end of the first month apparently detach themselves from the general framework and form separate clusters within its meshes. These clusters of proliferating neuroblasts form a prominent part of the nuclear or mantle layer, and it is from these that all the true nerve-cells of the brain and cord are derived.

The neuroblasts can be recognized by their characteristic shape (see Fig. 9). They possess a prominent nucleus and a tapering protoplasmic body which is continued into a slender primary process which is to become the future axis-cylinder of

DEVELOPMENT OF THE NERVOUS SYSTEM.

the cell. There is always the tendency for the primary processes of adjacent neuroblasts to come together and form a common strand, in this way giving rise to the arrangement in clusters. It has not yet been shown whether the processes of these cells come together secondarily, or whether the arrangement is the result of the manner of cell cleavage, the processes being the last portion of the proliferating cells to split apart. The primary process may either extend to some other part of the neural tube or may leave the neural tube and extend through the mesoderm to some peripheral structure, as is shown in Fig. 8. The motor nerve roots are an example of the latter, and in Fig. 10 there is shown how clusters

Fig. 8. — Diagram showing distribution of neuroblasts in human embryo of four weeks. On right side both neuroblasts and spongioblasts are shown; on left side neuroblasts only. The converging processes in the ventral part of cord unite to form the anterior nerve-roots. The processes forming the posterior roots enter the marginal layer and extend upward and downward as a longitudinal bundle, thus constituting the anlage of the posterior white column. (After His.) of such motor neuroblasts form rootlets which pass through the marginal layer of the neural tube and enter the mesoderm and unite in a common nerve-trunk. A group of neuroblasts from the same section is shown under higher power in Fig. 9.

Later the original pear-shape of the neuroblast is altered by the increase in body protoplasm and the development of secondary processes which branch out into the neighboring spaces. "With the increase in size the cell becomes moulded by the supporting framework and adjacent cells into its eventual shape, upon which the nucleus retires to the centre of the cell and enters upon the resting stage. The development of neurofibrils within the substance of the cell body and its processes occurs early; with special staining methods they can be demonstrated in embryo pigs 10 mm. long.

12 It was formerly thought (His) that the neuroblasts were derived directly and exclusively from the previously mentioned mitotic cells that are found during a brief period along the ventricular border of the wall of the neural tube, the so-called germinal cells (see Fig. 1, g). His pictures these germinal cells and shows them dividing and after assuming a pear-shape migrating out into the mantle or nuclear layer, being all the time sharply sepa

Fig. 9. — Cluster of neuroblasts from nucleus of origin of n. oculomotorius, showing characteristic shape and grouping of cells. Taken from same specimen shown in Fig. 10.

, mill ; . -o-v/

Fig. 10. — Section through floor of mid-brain of human embryo one month old (length 10 mm., Huber collection No. 3), showing the origin and formation of the trunk of a typical motor nerve (c. oculomotorius).

rated from the spongioblasts. This process has not been confirmed by the more recent investigators, and it is now generally thought that the germinal cells are only a sign of cell proliferation, and belong in common to the whole syncytial framework or myelospongium and thus are ancestors of the neuroblasts and spongioblasts alike.

The development of the neuroblast forms a problem that has been the subject of much investigation and discussion. Particular interest has been manifested in the growth of its attached nervefibre, both on account of the great length of the latter as compared with the size of the cell and on account of the intricate maze of

DEVELOPMENT OF THE NERVOUS SYSTEM. 13 fibres through which it traces its way to reach the organ for which it is intended. Of the many interpretations that have been offered in explanation of this problem the one that seems now to be most generally accepted is the so-called outgrowth theory of His, according to which each nerve-trunk is the process of a single nerve-cell. The fibre is a simple elongation of the primary process produced by an actual outflow of substance from the ganglion-cell toward the periphery, and it thus makes its way through various tissues to reach its proper end organ with which it comes into relation secondarily. It is this theory that has been ably substantiated by the histological studies of Eamon y Cajal and conclusively established by the recent experimental studies of Harrison.

Among the other hypotheses that have been advocated regarding the development of the nerve-fibre the most prominent is the cell-chain theory, according to which each nerve-fibre is the product of a chain or pathway of ectodermal (according to some, mesodermal) cells extending from the neural tube to the periphery. From the protoplasm of these cells fibrillar are differentiated which join to form a continuous fibre connecting secondarily the ganglion-cell with its end organ (Schwann, Balfour, Dohrn, Bethe). This theory was later modified by Apathy and 0. Schultze, who conceived of a syncytium of anastomosing cells which from the beginning connects the neural tube with the peripheral end organs, somewhat like the older description of Hensen, according to which protoplasmic bridges persist between dividing cells everywhere in the embryo, and by fibrillar differentiation some of these protoplasmic connections become eventually converted into nerve-fibres and thus connect the ganglion-cells with, the organs they supply. A further modification of this interpretation is that of Held (1907). Like Hensen, he conceives of a general syncytium connected everywhere by protoplasmic bridges, and later the nerve paths are formed by the transformation of these bridges ; but, according to Held, the growth of these fibrillse is from the nerve-cell outward, instead of simultaneously along the whole path, as maintained by his predecessors. In the protoplasm surrounding the nucleus of each neuroblast a plexus of neurofibrils forms, thus marking off neuroblasts from spongioblasts. From each of these a bundle of the fibrils extends out through the substance of the syncytial framework along a definite path and constitutes its axis-cylinder process. Subsequently, in addition to the primary bundle of fibrillae, other fibrillae extend out from the perinuclear network along other protoplasmic bridges and thus form the dendrites. In all these views the nerve-fibre is considered as a product of a group of cells instead of as an appendage of a single cell, as is postulated in the outgrowth theory.

14

Fig. 11. — Isolated ganglion-cells, from embryonic spinal cord of frog, and growing in clotted lymph. B is an isolated cell from tissue taken from branchial sense organ. The drawings are made from live specimens; two views of C are shown, taken four and three-quarters hours apart, showing rate of outgrowth of the nerve-fibre and the manner of its branching. (After Harrison.)

Fig. 12.— r-Three views, taken at intervals of Ik and 8i hours, of the same living nerve-fibres growing from a mass of spinal-cord tissue (frog embryo) out into clotted lymph. They show the rate of growth and the longitudinal splitting of fibres, and the characteristic growing ends, a larger example of which is represented by n.f. (After Harrison.)

DEVELOPMENT OF THE NERVOUS SYSTEM. 15 In his experiments carried out on amphibian larvas Harrison demonstrated four essential facts: In the first place he showed that no peripheral nerves would develop in an embryo from which the nerve-centres had been removed, thus establishing the fact that the ganglion-cells are an essential element in the development of the nerve-fibre. He next showed that the sheath cells of Schwann, upon the influence of which in the formation of the fibre many of the histologists had placed much emphasis, were not essential to the growth of the nerve-fibre, and that the axis-cylinders will develop and extend out in the surrounding tissues in the normal way and reach their normal length in specimens where the sheath cells have been eliminated. Thirdly, he showed, by modifying the environment of the developing nerve, that fibres will form in surroundings entirely different from their natural path and establish completely foreign connections. Finally, he succeeded in growing ganglion-cells outside of the body in an unorganized medium (clotted lymph), where all possibility of contributions on the part of other living tissues was eliminated, and he was able with this method to demonstrate directly under the eye the outgrowth of the nerve-fibre from the ganglion-cell as it developed hour by hour, proving conclusively the unity and continuity of the two.

Isolated ganglion-cells developing in an artificial medium are shown in Fig. 11, where A, B, and C represent three cells in different stages of growth. Two views of C are shown, one being taken four and three-quarters hours later than the other. The different cells shown in Fig. 11 are all represented under the same magnification, and on comparing them it is evident that the growth of the fibre is the result of a flowing out of the protoplasm of the cell into a thick primary process which as it elongates soon attains a uniformly slender width. The growing end of the process is characterized by an enlarged branched tip which continually undergoes amoeboid changes in form. One of these characteristic muchbranched growing ends is shown in Fig. 12 (»./.) under higher magnification. Occasionally the growing tip completely bifurcates, which results in permanent bifurcation or branching of the nerve-fibre at intervals along its course.

In Fig. 12 there can be seen a further phenomenon in the growth of the nerve-cell consisting in the apparent longitudinal splitting of the cell and its main process, a form of retarded cell division that occurs subsequently to the formation of the primary process. In the figure a, Id, and c represent the same nerve-cell mass as it appears at intervals of one and one-half and eight and one-half hours. It shows how that which is apparently a single cell and process becomes converted by this longitudinal cleavage into four individual cells with processes.

16 It is to be remembered that the forms shown in Figs. 11 and 12 represent the growth of ganglion-cells as it occurs in clotted lymph, and it is to be presumed that they depart in some details from that which occurs in living tissues ; however, the structure of the fibres and the characteristic enlarged branching ends conform exactly to the appearances seen in sections of specimens where the fibres have developed under normal conditions, and they no longer leave room for doubt as to the essential validity of the His outgrowth theory.

Development of Spinal Ganglia and Sensory Nerve=Roots.

As the elevated borders of the neural plate come together and fuse to form the closed neural tube, it detaches itself from and becomes completely covered in by the general ectoderm. During Ganglion crest

Ectoderm

Neural tube

Ectoderm jg^f^^^^^A'^t^i, ^>as^— Sensory roots Fia. 13. — Transverse .sections through dorsal region of human embryos showing three stages in the development ]ofJth3 ganglion crest and the anlage of the spinal ganglia. (After von Lenhoss^k and Kollmann.) this process a row of ectodermal cells situated along the borders of the neural plate undergo special changes ; they neither become incorporated in the neural tube nor do they become skin. It is these cells that form the so-called ganglion crest from which the cerebrospinal ganglia and sensory nerve roots are derived, in addition to the chromaffin cells and the sheath cells of Schwann.

The cells of the ganglion crest differentiate themselves from the cells of the neural tube during the closure of the latter. As shown in Fig. 13, they form a cellular ridge or crest along the line of closure. As they continue to develop they become apparently completely detached from the tube. Separating bilaterally into right and left crests, they migrate lateralward and ventralward in between the myotomes and neural tube. Here they continue to proliferate, and, as the constituent cells become differentiated into

DEVELOPMENT OF THE NERVOUS SYSTEM.

17

ganglion-cells, rootlets appear along the dorsal border of the crest and attach it secondarily to the neural tube. The ganglion crest thus extends as a flattened cellular band on each side uninterruptedly from the caudal tip of the neural tube forward to the region of the ear vesicle (compare Fig. 83). In the spinal region it becomes notched with segmental incisures along its ventral border, producing a series of ganglionic masses which for a time remain connected with each other by a dorsal cellular bridge. Eventually the latter disappears and there results a complete segmentation of the crest into separate spinal ganglia.

In the head region the structures to be supplied lose their simple segmental character, and there are introduced the complications of the gill arches and the lateral line system. These factors

iL

• Vi\. ?:**

-:.j>i

^ : » : " :i if .

i ¥* 

K$r'-,. MOP '/^^m«Jw

\ - ^' \

m -'•

Fig. 14. — Section through spinal ganglion of human embryo 18 mm. long, about 6 weeks old (Huber collection No. 5). On the left the cells are mostly in the indifferent stage, and ganglion-cells and supporting cells look much alike; on the right the ganglion-cells stand out sharply, due to the accumulation of granular protoplasm around the nucleus.

modify the character of the ganglion crest, though in its derivation from the border of the neural tube it apparently follows the same process as described for the spinal region. In the region of the midbrain it is believed by some writers (Meyer, Johnston) that a portion of the ganglion crest becomes incorporated in the neural tube, and the cells derived from this portion never obtain a peripheral position. The gross features in the development of the ganglia of the head and spinal regions will be referred to again more in detail under the peripheral nervous system.

At the time the ganglion crest detaches itself from the neural tube and begins to spread ventralward (embryos 2-3 mm. long), examination of it reveals a moderately compact mass lying in the space between the neural tube and the myotomes. The constituent Vol. II.— 2

18 cells possess oval or rounded nuclei with multiple small chromatin bodies such as are found in cells during active proliferation. The cell bodies possess ill-defined outlines and fuse with each other in a syncytial mass.

As the cells of the ganglion crest proliferate and become further differentiated it is possible to divide them into two different groups, — i.e., ganglion-cells and supporting cells, — in the same way that neuroblasts and spongioblasts are developed in the neural tube. (Compare Figs. 3, 4, and 5.) The steps in this procedure are shown in Figs. 14, 15, and 16. On comparing these the most apparent difference between the ganglion-cell (ganglioblast) and the supporting cell (capsule and sheath cells) is that the ganglion

tv 9

(K '

< «o

©. §^'

Fig. 15. — Section through cervical spinal ganglion of human fetus 8.5 cm. long, about 3 months old, showing large ganglion-cells with eccentric nuclei. Surrounding them are the much-branched supporting cells, some of which are to become capsule cells and the others sheath cells.

cells detach themselves more or less completely from each other and develop isolated large compact protoplasmic bodies, while the supporting cells consist almost entirely of branching processes which remain united in an extensive network or syncytium in the meshes of which lie the other cells.

By the end of the fourth month (Fig. 16) this differentiation is far advanced, and one can plainly recognize the supporting cells arranged in the form of capsules around the ganglion-cells proper. The next younger stage, at the end of the third month (Fig. 15), the supporting cells can be distinguished, but the capsule arrangement at that time is very incomplete. Going still further back, to the six-weeks embryo (Fig. 14), we find a portion of the ganglion mass consisting of undifferentiated neural crest cells, the ganglioncells and supporting cells both having the same appearance. How

DEVELOPMENT OP THE NERVOUS SYSTEM.

19

ever, scattered in among these "indifferent cells" are some more advanced cells that can be recognized as ganglion-cells by their increased body protoplasm. As a rule, the ganglion-cells attain a characteristic appearance earlier than the supporting cells.

The differentiation of the ganglion-cell consists in the condensation of the body protoplasm and the development of fibrillar processes. As an indifferent cell of the ganglion crest syncytium it has no definite outline, and in a teased preparation it appears as a ragged granular mass around a nucleus. Later, as the protoplasm of the cell body accumulates, it detaches itself more or less

Ly „ A 2.

<3 <o> ° &

Fig. 16. — Section through sixth cervical ganglion of human fetus 10.5 cm. long, about 4 months old. Here the branched supporting cells are arranged in distinct capsules around the ganglion-cells, and in the lower part of the section they form a characteristic framework for the nerve-fibres. The ganglion-cells are nearing the completion of their development; the nuclei are retiring to the centre of the cell, and the basic stainable substance is appearing at the periphery of the cell bodies which later becomes clumped to form the Nissl bodies.

completely from the neighboring cells and attains discrete outlines. It is possible then to make teased preparations such as are shown in Fig. 17.

The earlier ganglion-cells usually have a simple fusiform shape with a process extending out at each pole, as is shown in Fig. 17, group A. These lie in clusters, like the neuroblasts in the spinal cord, and their processes are still fused so that in teasing them apart we get the appearances seen in group B. By Golgi methods it can be demonstrated that the central processes of such cell groups form rootlets which enter the dorsal part of the wall of the rrural tube, where for the most part the fibres bifurcate and form a longitudinal bundle within the marginal layer, as seen

20 in Fig. 8, and after a longer or shorter course upward or downward they end in an arborization among the cells of the mantle layer. In the spinal region the sensory fibres thus entering form one continuous path, which eventually constitutes the dorsal funiculi of the cord. In the cranial region the entering fibres of the sensory nerves form separate paths, the vestibular tract, the trigemi

c

D

E

F

Fig. 17. — Isolated cells teased from spinal ganglia of embryo pigs 20-40 mm. long, showing the variation in the form of the early ganglion-cells. B shows the grouping of the cells; A, C, D, E, and F represent selected types arranged according to the form of the cell body and the character of the processes.

nal tract, and the tractus solitarius, to which are contributed fibres from the seventh, ninth, and tenth nerves. The peripheral processes unite into bundles, which join with the fibres of the anterior roots and with them form the main nerve-trunk, the further development of which will be referred to later in the section on the peripheral nervous system.

In examining a given ganglion in embryos about two months old it will be found that the cells are present in many different stages of growth, and that in addition to the simple bipolar earlier

DEVELOPMENT OF THE NERVOUS SYSTEM. 21 cells there are younger cells that must make room for themselves in the intervening spaces. Such cells possess great irregularity in their shape. Their form and the manner in which the processes leave the cell body is apparently determined by the pressure that is exerted upon them by the adjoining cells. Some of the forms that are frequently seen in teased preparations are shown in groups C, D, E, and F, Fig. 17. The only feature common to all of them is the eccentric position of the nucleus, which is retained until the cell nearly reaches its full size, upon which the nucleus retires to the centre of the cell and assumes a resting appearance. (Compare Figs. 15 and 16.) It will be seen that there is a great variety in the number, size, and situation of the cell processes. In group F there are multipolar forms, some of which persist as such in the adult forms. A very common form is shown in group E, which are technically bipolar cells, but in each case one of the processes is much better developed than the other, the small one having the appearance of undergoing retrogression. It suggests the possibility that these represent bipolar cells that are being converted into unipolar cells by absorption of the smaller process. All through the nervous system there is evidence of cell processes and nerve-fibres that grow out for a certain distance and then for some reason cease to develop any further and finally disappear, as in the case of the hypoglossal nerve. This phenomenon is not limited to the temporary existence of vestigial structures, but also represents what might be called a ' ' tentative growth, ' ' under which we may understand that of the great number of processes and fibres which start out a certain number fail to establish adequate connections and subsequently disappear. On the other hand, in the present case of the ganglion-cell the smaller process may represent the slender axon that enters the cord, while the heavier process corresponds to the peripheral dendrite.

The conversion of bipolar cells into unipolar T-shaped cells of Eanvier is commonly supposed to be due to the unilateral growth of the cell body towards one side, which brings about an approximation of its two processes so that they fuse in a common extension from the cell. The cells shown in Fig. 17, C, are regarded as transitional cells undergoing this change, and cells 8, 9, and 10 would represent the successive steps in the process. But if this is the explanation, it should be expected that no cells would show the T division until they reach the stage of growth represented by cell 10. This, however, is not the case, for in group D we have cells in which the T division is already completed before that time, and moreover cell 1 in group D possesses the T division and the opposite process is still intact. Furthermore, as will be presently seen, the growth of the enmeshing capsule cells is already

22 well advanced while many of the cells are still of the simple bipolar type {e.g., Fig. 18, a and b), and this would tend to prevent the approximation of opposite processes. There is therefore much reason to believe that the T division is simply the result of the bifurcation of the growing end of the main process of the cell in the manner so clearly demonstrated by Harrison's growing nervecells (Fig. 11), and that the so-called transitional cells are merely accidental forms due to the moulding influence of adjacent cells. Certainly a great many of the cells can never undergo the supposed transformation.

Concerning the migration of spinal ganglion cells and their relation to the formation of the sympathetic ganglia the reader

Fig. 18.— Teased preparations from spinal ganglia of pig, showing development of sheath and capsule cells. In a and b (pigs 3-4 cm. long) the supporting tissue consists of a loose syncytium enmeshing the ganglion-cells; in c and d (pigs 20 cm. long) the protoplasm is condensed and arranged in the form of distinct capsules. In d the capsule cells are directly continuous with the sheath coating the main process of the cell. Compare with Figs. 15 and 16.

is referred to the section on the development of the sympathetic system, and for the chromaffin or phssochrome cells and organs to Chapter XV which describes those structures.

The development of the supporting cells can be followed by comparing Figs. 14, 15, and 16, which show the appearance as seen in cross sections, and Fig. 18, which represents teased preparations. Like the ganglion-cell the supporting cell is derived from the indifferent cells of the ganglion crest, and also like the ganglion-cell the chief phenomenon of its differentiation consists in the condensation of the body protoplasm.

When first recognized the supporting cells form a loose poorly defined granular syncytium from which the ganglion-cells have become almost completely detached. The subsequent condensation takes the form of branching processes instead of full rounded bodies like the ganglion-cells, and there results a sharply outlined nucleated framework in the meshes of which lie the nerve elements proper (Fig. 18, c). In their derivation and behavior it will be

DEVELOPMENT OF THE NERVOUS SYSTEM. 23 noticed that the supporting cells are directly analogous to neuroglia cells. Eventually they become subdivided into capsule cells and sheath cells, according to whether they form an envelope around some ganglion-cell or are situated along the course of a cell process. The two kinds are identical in character and are directly continuous with each other, as can be seen in Fig. 18, d. It is now believed, as will be referred to again under myelinization, that it is some of the supporting cells of the spinal ganglia that migrate outward along the course of the nerves and form all the sheath cells of the peripheral nerves.

Myelinization of Nerve=fibres.

The final development of the nerve-fibre is characterized by the formation of a fatty enveloping sheath. This process does not become apparent until about the fourth month, and it is not completed until after birth. Many details concerning the acquisition and structure of the nerve-sheath are still involved in dispute, and especially in human material there is no adequate description yet available. Our knowledge is mostly based on studies of the pig and sheep (Vignal 1883, Gurwitsch 1900, Bardeen 1903, and Hardesty 1904).

In the development of the nerve-sheath we have to distinguish between the formation of the non-nucleated myelin sheath and the nucleated membranous sheath which are its two subdivisions, the latter being synonymous with the sheath of Schwann or neurolemma. Either of these may make their appearance before the other. In the central nervous system the myelin sheath appears first, and until recently was supposed to be the only covering possessed by these fibres. In the case of the peripheral nerves, on the other hand, the axon is usually completely inclosed by the membranous sheath before the myelin appears, and in the case of some fibres, particularly in the sympathetic system, this remains the permanent condition. Such fibres are known as non-medullated fibres.

The cells that give rise to the membranous sheath of the peripheral fibres appear relatively early. They are easily distinguished toward the end of the third week as spindle-shaped cells (cells of Schwann or sheath cells) closely attached to the developing nerve-fibres. That these cells are ectodermal and are derived from the ganglion crest has been shown experimentally in amphibian larvae by Harrison (1906). The experiments of this investigator show that the source of the sheath cells can be removed by early extirpation of the ganglion crest, and that in such cases there develop naked fibres which normally are found covered with sheath cells. It should, however, be remembered that this may not be the only source in all forms ; in fact in elasmobranchii it is said that there are in addition a large number of sheath cells that wan

24 der out somewhat later from the ventral part of the spinal cord along the motor roots (Neal, 1903). Up to the time of Harrison's experiments it was supposed that the sheath cells were derived from the mesenchyme and were foreign cells that invaded the nerve-trunks and ganglia very early (Vignal, Gurwitsch, and Bardeen).

In the human embryo during the transformation of the ganglion crest into spinal ganglia it is possible to trace step by step the differentiation of its constituent cells into ganglion-cells proper and supporting cells, as shown in Figs. 14, 15, and 16. It is these supporting cells that form the capsule cells and sheath cells. Of the latter some are for the spinal ganglia themselves while others

^a

Fig. 19. — Isolated fibres showing development of medullary sheath. The specimens are taken from the intercostal nerve of pig fetuses about 15 cm. long and prepared by formalin and osmic acid. (After Bardeen.) wander out along both the sensory and motor nerve-trunks, accompanying them in their growth forward, and eventually forming their membranous sheath.

The sheath cells are at first arranged so as to inclose a group of axons in a common trunk. By subsequent proliferation they invade the trunk and divide it into smaller bundles, until finally they close in around each axon and form individual sheaths. Each sheath cell corresponds to a segment of the sheath, and the interval between two adjacent cells corresponds to a node of Ranvier. The differentiation of the sheath cells and their conversion into the nerve-sheath is shown in Figs. 19 and 20. In the earlier stages they unite in the formation of a segmented thin-walled sheath, which in formalin specimens can be seen loosely enveloping the axis-cylinder, and apparently containing fluid (Bardeen, 1903) ; in osmic acid specimens it is found shrunken closely against the axis-cylinder (Fig. 20, A). Each nucleus maintains its position near the centre of a segment and continues to maintain a small quantity of granular protoplasm around itself. There then follows a deposit of myelin within the membranous sheath around the axis-cylinder. In the case of most peripheral fibres the axis

DEVELOPMENT OF THE NERVOUS SYSTEM.

25

cylinder becomes completely inclosed by the sheath of Schwann before the formation of the myelin sheath begins. As regards the origin of the myelin there have been three theories presented: first, that it is formed by the sheath of Schwann ; second, that it is a differentiated portion of the outer part of the axis-cylinder (Kolliker, 1904) ; and according to the third it is precipitated from the fluid immediately surrounding the axon by some reaction between the two (Bardeen, 1903). At first the myelin is evenly distributed along the axis-cylinder, but later it accumulates more rapidly in some areas than in others, which produces the beaded appearance shown in Fig. 19, c. Eventually the sheath is completely filled out (Fig. 19, d). At the nodes of Ranvier the formation of myelin continues for some time after the completion of the myelin deposit in the remainder of the sheath (Fig. 19, e), A

s:

Fig. 20. — Isolated fibres of the sciatic nerve of sheep fetus 15 cm. long, treated with osmic acid and showing development of the nerve-sheath. A shows axon inclosed in membranous sheath with proliferating nuclei. In B the development is more advanced, and a deposit of myelin shown in dark stipple has formed within the membranous sheath: n.R., node of Ranvier; S.z., nucleus of membranous sheath. (After Vignai.) apparently thus allowing for the lengthening of the fibre. During the later stages of development further provision is made for the lengthening of the fibre by the intercalation of new segments of the sheath of Schwann at the nodes of Ranvier.

In the central nervous system there seems to be no distinct separate membrane investing the fibres that exactly corresponds to the sheath of Schwann of the peripheral medullated fibres. There are, however, sheath cells present. They are apparently most numerous and possess most protoplasm during the active period of myelin development. On adult fibres they are relatively fewer in number and are less conspicuous since they then possess less body protoplasm. According to Hardesty (1904), these sheath cells of the central nervous system in the pig are derived from the indifferent supporting cells which we have already seen for the most part are converted into neuroglia tissue. Shortly after the beginning of myelin formation they become differentiated into typical sheath cells, and form around each nervefibre an extended lamellated reticulum, in the meshes of which the small globules of myelin are supported.

26 Myelin formation commences about the fourth month, but it does not make its appearance in all parts of the nervous system at the same time, nor simultaneously in all parts of an individual fibre. As a rule, it appears first in fibres that become functional first and in fibres that are phylogenetically the oldest. In an individual nerve the myelin is first seen at the central end and spreads from there toward the periphery. According to Westphal (1897), the medullation of the spinal nerves is completed between the second and third years, while the cranial nerves are completely medullated by the ninth or tenth post-embryonal week. Of the

vermis sup.

^Sx

JLf \ i# imt.

fvrn

a

J^y. '

a Jkns ;

Fio. 21. — Section through hind-brain of new-born child, showing myelinization of fifth, sixth, seventh, and eighth cranial nerves and associated fibre tracts (fasciculus longitudinalis mediahs and corpus trapezoides). The lemniscus medialis is also myelinated. These stand out in sharp contrast to the other fibres of the pontine region, including the pyramidal tract, which are still devoid^of myelin. (After Edinger.) cranial nerves the motor fibres are medullated first, and the sensory fibres a little later, excepting the acusticus, the vestibular division of which is medullated about as soon as the motor fibres. In premature births the medullation of the cranial nerves seems to be accelerated. Westphal states that there is a gradual increase in the size of the sheath with increasing age. He finds that on comparing the adult with the child the largest sheath in a given motor nerve in the adult is twice as large as the largest sheath in the same nerve in the child. He finds also that the sheaths are more uniformly large in the adult, while in the child there is a large proportion of small sheaths, so that the average of large sheaths in an adult nerve is four or five times greater than in the same nerve in a child. Regarding the time of myelinization of the different areas and fibre paths of the central nervous system

DEVELOPMENT OF THE NERVOUS SYSTEM. 27 the reader is referred to Westphal's paper and the work of Flech sig (1896, 1904). As can be seen in Fig. 21, the relative time of appearance of the myelin is an important factor in the identification of the different fibre paths.

LITERATURE.

Bardeen, C. R. : The Growth and Histogenesis of the Cerebrospinal Nerves in Mammals. Arner. Jonm. Anat. Vol. 2, No. 2. 1903. Biervliet, vax : La substance chromophile pendant le cours du developpement de la cellule nerveuse (chromolyse physiologique et chromolyse experi mentale). Journ. de Neurol. — Ref. im Centralbl. f. Nervenheilk. u Psych.

Jahrg. 24. 1901. Brodhann, K. : Bemerkungen iiber die Fibrillogenie und ihre Beziehungen zur Myelogenic mit besonderer Beriicksichtigung der cortex cerebri. Neurol.

Centralbl. Jahrg. 26, S. 338-349. 1907. Cajal, S. Ramon: Nouvelles observations sur devolution des neuroblastes avec quelques reinarques sur Thypothese neurogenetique de Hensen-Held. Anat.

Anz. Bd. 32. 1908. Cameron, J.: The Development of the Vertebrate Nerve-cell: A cytological study of the neuroblast-nucleus. Brain : Part 115. 1906. Capobianco, T. : Recherches ulterieures sur la genese des cellules nerveuses. Ann.

Neurol. Anno 23.— Ref. Arch. ital. Biol. T. 44. 1905. Collin, R. : Recherches cytologiques sur le developpement de la cellule nerveuse.

Le Nevraxe. T. 8 1906 et These en med. Nancy 1907. Colucci, C, et Piccinino, F. : Su alcuni studii di sviluppo delle cellule del midollo spinale umano. 12 Fig. Ann. Neurol. Anno 18. p. 81-110.

1900. Ref. in Centralbl. f. Nervenheilk. u. Psych. Jahrg. 24, S. 521. 1901. Flechsig, P. : Einige Bemerkungen iiber die Untersuchungsmethoden der Gross hirnrinde insbesondere des Menschen. Bericht d. math.-phys. Kl. d. Kgl.

Sachs. Gesell. d. Wissensch. z. Leipzig. 11. Jan. 1904. Gehirn und Seele. 2. Aufl., Leipzig. 1896. Gierlich : Uber die Entwicklung der Neurofibrillen in der Pyramidenbahn des Menschen. Vortrag a. d. Vers, siidwestdeutscher Neurologen, Mai 1906.

Deutsche Zeitschr. f. Nervenheilk. Bd. 32, S. 97. 1906. Gurwitsch, A. : Die Histogenese der Schwann'schen Scheide. Arch, f . Anat. u.

Physiol. Anat. Abt. 1900. Hardesty, I. : On the Occurrence of Sheath Cells and the Nature of the Axone Sheaths in the Central Nervous System. Amer. Jour. Anat. Vol. 4. 1904. On the Development and Nature of the Neuroglia. Amer. Jour, of Anat.

Vol. 3. 1904. Harrison, R. G. : Experiments in Transplanting Limbs and Their Bearing Upon the Problems of the Development of Nerves. Jour, of Exper. Zool.

Vol. 4, p. 239 ff. 1907. Further Experiments on the Development of Peripheral Nerves. Amer.

Jour. Anat. Vol. 5. 1906. Hatai, S. : On the Increase in the Number of Medullated Nerve-fibres in the Ventral Roots of the Spinal Nerves of the Growing White Rat. Jour.

Comp. Neurol. Vol. 13, p. 177-183. 1903. Hensen, V. : Die Entwicklungsmechanik der Nervenbahnen im Embryo der Sauge tiere. Ein Probeversuch. 4°. 51 S. Lipsius & Tischer, Kiel u. Leipzig.

1903. His, W. : Histogenese und Zusammenhang der Nervenelemente. Arch. f. Anat. u.

Physiol. Anat. Abt. Suppl. Bd. 1890. Histogenese und Zusammenhang der Nervenelemente. Arch. f. Anat. u.

Physiol. Anat. Abt. Suppl.-Bd., S. 95-117. 1890.

28 HUMAN EMBRYOLOGY.

Huber, G. C. : Studies on Neuroglia Tissue. Contributions to Medical Research dedicated to V. C. Vaughan. University of Michigan, Ann Arbor. 1903. Kappers, C. U. A. : Recherches sur le developpement des gaines dans le tube nerveux. 2. Teil. 2. Aufl. Petrus Camper. 1903. Kolliker, A. von : Uber die Entwicklung der Elemente des Nervensystems, contra Beard und Dohrn. Verb. Anat. Gesellsch. Wien. S. 76-78. 1892. Die Entwicklung der Elemente des Nervensystems. Zeitschr. f. wissenseh.

Zool. 4 Taf. u. 12 Textabb. Bd. 82. 1905. Uber Entwicklung der Nervenfasern. Anat. Anz. Bd. 25. 1901. Kronthal, P. : Zum Kapitel Leukocyt und Nervenzelle. Anat. Anz. Bd. 22, Nr. 21/22. 1903. Lawdowski, M. D. : Uber anastomotische Verbindungen zwischen Nervenzellen.

Russki Wratsch. Bd. 1, Nr. 12. 1902. (Russisch.) Lenhossek, M. v. : Nervensystem. Ergebn. Anat. u. Entwicklungsgesch. Bd. 7.

S. 110-178. 1898. Minot, C. S. : Die fruben Stadien und die Histogenese des Nervensystems.

Ergebn. Anat. u. Entwicklungsgesch. Bd. 6, S. 687-739. 1897. Muhlmann, M. : Uber die Veranderungen der Nervenzellen in verschiedenem Alter. Verb. d. Gesell. Deut. Naturf. u. Arzte. 1900. Bd. 2, S. 20-21. 1901. Weitere Untersuchungen uber die Veranderung der Nervenzellen in verschiedenem Alter. Arch. f. mikrosk. Anat. Bd. 58, S. 231-246. 1901. Nussbaum, M. : Nerv und Muskel : Abhangigkeit des Muskelwachstums vom Nervenverlauf. Yerhandl. d. Anat. Gesellsch. 8. Vers, in Strassburg. 1894. Retzius, G. : Biol. Untersuch. N. F. IV, p. 59, Kenntniss der Ganglienzellen der Spinalganglien ; V, p. 24, Ependym und Neuroglia; VI, p. 1, Neuroglia; VIII, p. 102, Entwicklung der Riickenmarkselemente. Stockholm, 1892, 1893, 1894, 1898. Rubaschkin, W. : Studien uber Neuroglia. Arch, f . Mikr. Anat. Bd. 64. 1904. Schaper, A. : Die friihesten Differenzierungsvorgange im Centralnervensystem.

Arch. f. Entwicklungsmechanik. Bd. 5. 1897. Schultze, O. : Uber den friihesten Nachweis der Markscheidenbildung im Nervensystem. Sitz.-Ber. d. Phys.-med. Gesellsch. zu Wiirzburg. (Friihe Entste hung auch beim Menschen.) 1906. Seggel: Uber das Verhaltnis von Schadel- und Gehirnentwicklung zum Langen wachstum des Korpers. Arch. f. Anthropol. N. F. Bd. 1, Heft 1. 1904. Strasser, H. : Alte und neue Probleme der entwicklungsgeschichtlichen Forschung auf dem Gebiete des Nervensystems. Anat. Hefte, Abt. 2, Ergebnisse.

Bd. 2, S. 565-603. 1893. Takahashi, K. : Some Conditions Which Determine the Length of the Internodes Found on the Nerve-fibres of the Leopard Frog, Rana Pipiens. Jour.

Compar. Neurol, and Psychol. Vol. 18. 1908. Vignal, W. : Memoire sur le developpement des tubes nerveux chez les embryons de mammif eres. Arch, de Physiol. 3me ser. T. 1. 1883. Vogt, O.: Zur Kritik der sogenannten entwicklungsgeschichtlichen anatomischen Methode. Ber. lib. die Jahresvers. d. Ver. deutsch. Irrenarzte, 1900. Ref.

Centralbl. f. Nervenheilk. u. Psych. Jahrg. 23, S. 228-289. Valeur de l'etude de la myelisation pour ranatomie et la physiologie du cerveau.

Journ. de la Physiol, et de la Pathol, generate. T. 2, p. 525-538. 1900. Waldeyer, W. : Uber den neuesten Stand der Forschungen im Gebiet des Nervensystems. Arcb. f. Anat. u. Physiol., Physiol. Abt., Jahrg. 1895. Weigert, C. : Beitrage zur Kenntnis der normalen menschlichen Neuroglia. Frankfurt a. M. 1895. Westphal, A.: Uber die Markscheidenbildung der Gehirnnerven des Menschen.

Arch. f. Psychiatric Bd. 29. 1897. Wlassak : Die Herkunft des Myelins. Ein Beitrag zur Physiologie des Nervenstiitzgewebes. 4 Taf. Arch. f. Entwicklungsmech. Bd. 6, S. 453-i93. 1898.

DEVELOPMENT OF THE NERVOUS SYSTEM. 29

II. DEVELOPMENT OF THE CENTRAL NERVOUS SYSTEM The nervous systems of all vertebrates develop from the same ectodermal germ layer and within certain limits they always pass through the same series of developmental stages. The difference between the higher and lower forms consists only of a variation in size and degree of development of certain of the component structures, owing to the variation in the demand made upon those particular structures by the special life habits, size, and requirements of the respective animals. The nervous system is delicately responsive to the requirements of each individual animal and all of its parts are subject to a greater or lesser development in accordance with their functional necessity.

In this variability in the development of its different parts the nervous system does not adhere to a phylogenetic order. Hence it is impossible to definitely determine the relative position of two animals by the character of their nervous systems. With the exception of one portion the nervous system of the lower vertebrates possesses all the essential features and the same general grouping found in the highest vertebrates. In fact in certain instances they possess structures that are more complicated and presumably more efficient than the analogous structures in higher forms. Even the same structure may exhibit in closely related families a marked difference in the degree of its development, particularly in case of the higher co-ordinating apparatuses such as is represented by the cerebellum. One portion (pallium), however, contrary to the rest of the nervous system, varies in its development directly according to each animal's phylogenetic position. It follows a definite phylogenetic curve. It is present as a small rudiment in fish and amphibians and gradually increases in relative size and complexity as we ascend the vertebrate scale until it finally forms the prominent cerebral hemisphere of man.

In all vertebrates the central nervous system consists of a dorsally placed hollow tube, the walls of which are connected by the peripheral nerves with the different parts of the body. "Where many nerves enter or arise the central apparatus is larger, as in the enlargements of the cord, the medulla oblongata, midbrain roof and thalamus. Further enlargement is produced by the establishment of fibres connecting different levels. A third group of enlargements is produced by the establishment of nuclear masses which serve as higher receptive and co-ordinating centres. In general, it may be said that the nervous system consists of afferent and efferent tracts, or peripheral system, and a central mass that serves to connect and co-ordinate them, or central system. Of the former, the afferent tracts are made up of receptor neurones and the efferent tracts of effector neurones. The latter, or central

30 HUMAN EMBRYOLOGY.

nervous system, in addition to the central extension of peripheral neurones, consists of intersegmental neurones, uniting different levels, and suprasegmental neurones, that form the receptive and co-ordinating centres for control over the lower neurones. The location of the elaborated sense organs at the forward end of all vertebrates, to facilitate the gathering of food and the detection of enemies and mates, results in the highest development of the nervous system in the head region. We thus speak of cephalization of the central nervous system and distinguish the enlarged anterior end, with which the special sense organs are connected, as brain, and the remainder as spinal cord.

At the time of the closure of the neural tube that portion corresponding to the brain undergoes a threefold constriction, forming three primary brain vesicles which are constant in all vertebrates and form a definite morphological basis for the subdivision of the brain into three primary portions. These are known as the prosencephalon (the most oral one), the mesencephalon, and the rhombencephalon (the most caudal one). Subsequently the first of these (prosencephalon) becomes further subdivided into an end portion or telencephalon and an intermediate portion or diencephalon. The last, or rhombencephalon, is also often subdivided for descriptive purposes into the metencephalon, from which the pons and cerebellum are developed, and the myelencephalon, from which the medulla oblongata is developed. Owing to the tendency to metamerism of the mesencephalon and the rhombencephalon and their conformity with the type seen in the spinal cord, it has been suggested (Strong) that we recognize an epichordal system including all that portion of the central nervous system from which the true cerebrospinal nerves arise, and which lies dorsal to the chorda. The remainder is designated as prechordal, the boundary between the two being the primary ventral infolding of the brain wall (plica encephali ventralis). These terms of subdivision will be constantly made use of in the following pages.

We will first consider the anlage of the nervous system as it is found in the earliest stages, and will then follow its conversion into the medullary tube, and trace the development of the latter during the first four weeks. This will be followed by a description of the central nervous system as it is found at the end of the first month. Up to that time it is necessary to consider the different portions more or less in conjunction with each other. After the first month, however, the further development of the individual subdivisions will be described separately.

DEVELOPMENT OF THE NERVOUS SYSTEM.

31

(a) Development of Central Nervous System during First Month.

The first evidence of the nervous system is the neural groove that forms as an axial furrow in the median line of the germin.-i! plate, as can be seen in Fig. 22. The thickened ectoderm on each side of this groove forms the neural plate. In the Von Spee embryo Grle the borders of the neural plate are beginning to be indicated, and just a little later, in embryos about two weeks old, the elevated edges of the neural plate are clearly marked off from the rest of the ectoderm, as shown in Fig. 23, A.

Sacculus vitellinus •

Canalis neurentericu* Linea primitiva Pedunculus abdominal is

Villi _ choriales "^g^.

Fig. 22. — Dorsal view of human embryo in the neural-plate stage. The amnion is opened, exposing the germinal plate lying directly on the yolk-sac. (Graf Spee, from Kollmann.) The further elevation of the edges and their approximation and fusion across the median line to form the neural tube is shown in four stages in Fig. 23. The comparison of these figures shows that the formation of the neural tube is most advanced in the middle of the germ plate corresponding to the junction of brain and spinal cord. From this region the differentiation and closure of the tube extends caudally and orally, the last portions to close being called the anterior and posterior neuropores (see Figs. 24 and 26). The process of closure, though it always begins in about the same region, shows some variation in the time of its occurrence. In Fig. 23, C, it is further advanced than in Fig. 23, D, which, judging from the number of somites, is the older embryo of the two.

The anterior neuropore is found closed in embryos of about 23 somites and the posterior neuropore a little later in embryos

32 of about 30 somites, the end of the third week. The neuropores do not exactly represent the anterior and posterior ends of the neural tube. 1 The anterior and posterior margins of the neural plate are rounded off laterally so that the extreme anterior and

Amnion

Canalis neurentericus

Encephalon

— Pedunculus abdominalis

Neural tube not yet closed

~p*p— Hindbrain

Midbrain

Provertebra VII

Amnion removed

Protovertebra VII

Canal neurentericus

L-inea primitiva

Midbrain

Amnion (cut)

Protovertebra XI\

Neural tube not yet close

Forebrain

Mouth-fissure

Amnion (cut)

Sacctilus vitellinus

Fig. 23. — Four stages in the closure of the neural tube in human einbryos possessing from 5 to 14 somites. A, embryo Klb, 5-6 somites, from Keibel u. Elze (Normentafeln); B, embryo No. 391, 7 somites, Mall collection, after Dandy; C, Eternod's embryo, 2.11 mm. long, 8 somites, after Kollmann; D, embryo 2.4 mm. long, 14 somites- estimated age 14 to 16 days. (C and D after Kollmann.) posterior ends of* the neural plate are found in the median line. In the closure of the tube it is the lateral portions of the anterior and posterior margins that unite last, and so the neuropores are found dorsal to what was originally the extreme ends of the neural

1 Compare also the chapter describing the development of the eye.

DEVELOPMENT OF THE NERVOUS SYSTEM.

33

tube. This is particularly marked in case of the posterior neuropore (Fig. 24). The region in the adult brain corresponding to the anterior end of the neural plate was placed by His at the infundibulum. Johnston (1909) from comparative embryological studies places it at the optic chiasma. The anterior neuropore forms dorsal to this, and by its closure completes the lamina terminalis.

FiQ. 24. — Profile reconstruction of human embryo 3.2 mm. long, showing neural tube closed except for anterior and posterior neuropores. The chorda lies closely against its ventral surface. The optic evagination is shown by'a dotted line and is connected with the neural tube by a wide slit. The ear vesicle is indicated lateral to the hindbrain. In the area reuniens the neural tube, chorda, and foregut are still undifferentiated from one another. (After His, 1904.) Ab., truncus arteriosus; All., allantoic duct; Bs., abdominal stalk; CL, cloake; Lb., liver anlage; Lg., lung anlage; Nh., stalk of the umbilical vesicle; Vh., atrium; Vt., ventricle. ..^ While still in the neural plate stage before its closure the central nervous system is differentiated into an anterior portion that is to form the brain and a posterior portion that is to form the spinal cord, as shown in Fig. 23, A. The brain portion is wider, more irregular, and projects beyond the yolk sac, bending forward nearly at a right angle. It forms about one-half of the neural plate. We cannot yet with any clearness recognize the subdivisions of the brain. The spinal cord portion is narrower and uniform in width. It extends caudally to the neurenteric canal, extending on each side a little beyond it. Most of the spinal por Vol. II.— 3

34 tion of the neural tube that is differentiated at this time is that which is to form the cervical cord. The lower cord is formed by the caudalward expansion of the neural plate. In embryos of seven somites, Fig. 23, B, where the neural tube is just beginning its closure, there appear two distinct constrictions in the brain

p.-c.

Coe.

Fig. 25. — Dorsal view of model of human embryo possessing 7 somites, being the same embryo shown in Fig. 23, B. Portion of ectoderm of right neural plate is removed, showing thickness of wall and its relation to deeper structures. The three primary cerebral vesicles are indicated. (After Dandy.) All., allantois; Ch., chorda; Coe., ccelom; Fg., foregut; Hg., hindgut; Ht., heart; Mes., mesoderm; P. c, pericardial ccelom; U., umbilical arterial sinus; V., umbilical vein.

wall, subdividing it into a prosencephalon, mesencephalon, and rhombencephalon. These are best seen in Fig. 25, which is from the same embryo, but the ectoderm is partially removed, showing the cut edge of the right half of the neural plate. With the closure of the tube they become still more marked, as shown in Fig. 23, D. In these young stages the neural tube follows the flattened curve of the yolk sac, the only bending in the axial line being anteriorly

DEVELOPMENT OF THE NERVOUS SYSTEM.

35

and posteriorly where it projects free beyond the surface. It is now supposed that the sharp ventral convexity sometimes found in the spinal region in these young specimens is artificially produced by the preserving medium.

The character of the neural tube at the beginning of the third week is shown in Figs. 24, 26, and 27. The age of embryo from which these reconstructions were made was estimated by His at two weeks, but on comparing it with other embryos we are probably safer in considering it as being in the third week. The

Pallium

Diencephalon

Neuroporus ant Corpus striatum

y Vesicula optica

• 5 Mesencephalon

Site of future pontine flexure

y Rhombencephalon

Fig. 26. — Median view of a model of the brain of a human embryo 3.2 mm. long, being the same specimen shown in Fig. 24. (After His.) closure of the tube is now complete except for the small anterior neuropore and a small portion at the caudal end. The cranial portion of the tube is distinctly subdivided into the three primary vesicles. The most caudal one, or rhombencephalon, is much the largest. Its roof is not yet thinned out, the wall of the neural tube being everywhere of about the same thickness.

The first flexure to form in the neural tube is the cephalic flexure. It is well marked at this time, being a sharp bend in the neural tube in the region of the midbrain so that the axis of the forebrain forms approximately a right angle with that of the

36 hindbrain, the notch formed on the ventral surface being known as the ventral cephalic fold. The location of the future pontine flexure is marked by a ventral bulging of the floor of the tube. The anterior primary vesicle or forebrain at the completion of its closure is already marked off into its main subdivisions. The optic evagination exists as a depression in the lateral wall before the closure is complete, but at the time under consideration it forms a distinct pocket projecting later alward and caudalward. Extending from its dorsal border is a fold (margo thalamicus) which marks the bound

Diencephalon

Pallium

Mesencephalon

\

Isthmus

Neuroporus anterior

Vesicula optica

Rhombencephalon *'

Fig. 27. — Lateral view of same model shown in Fig. 26.

ary between the telencephalon and diencephalon and separates the anlage of the thalamus from that of the pallium. The wall adjacent to and forming the anterior border of the evagination forms the anlage of the corpus striatum. The wall posterior to the evagination forms the ventral half of the diencephalon or hypothalamus.

The changes occurring in the central nervous system between the third and fourth weeks may be seen by comparing Figs. 28 and 29 with Figs. 26 and 27. The tube is now completely closed and the neuropores have disappeared. The walls due to their differentiation begin to show a variation in thickness. This is particularly marked in the thinning of the roof of the rhomben

DEVELOPMENT OF THE NERVOUS SYSTEM.

37

cephalon. The flexures of the tube are more marked. The cephalic flexure has increased from a right angle to an acute one,

Mesencephalon

Isthmus

' Cerebellum

Corpus striatum Evaginatio

im / / io oculi Hypothalamus

Medulla oblongata

Fig. 28. — Median view of a model of the brain of a human embryo 6.9 mm. long. This figure. should be compared with Fig. 26. (After His.)

Diencephalon (Thalamus)

Mesencephalon

Isthmus t

Metencephalon '— - v

Telencephalon , (Pallium)

Myelen cephalon c

Fig. 29. — Lateral view of same model shown in Fig. 28. Compare with Fig. 27.

so that the axis of the forebrain is about parallel with that of the hindbrain. A distinct pontine flexure is now present in the rhombencephalon.

38 The formation of the flexures of the neural tube is shown in Fig. 30. It will be seen that there are three distinct flexures, cephalic, pontine and cervical. Two of them have already been

Fig. 30. — Profile views of the brains of human embryos as seen during the third (A), fourth (B), and eighth (C) weeks, showing the conversion of the three primary cerebral vesicles into their chief subdivisions and the formation of the flexures of the neural tube. A, optic vesicle; Br, pontine region; Gb, auditory vesicle; H, telencephalon; Hb, metencephalon; J, isthmus; M, mesencephalon; N, myelencephalon; NK, neck bend; Pm, mammillary recess; Rf, posterior medullary velum; Tr, infundibular recess; Z, diencephalon. (After His.) mentioned. The third, or cervical flexure, marks the junction of brain and spinal cord and is formed about the same time as the pontine flexure. They are formed, in part at least, in consequence of unequal growth of different parts of the neural tube. They

DEVELOPMENT OF THE NERVOUS SYSTEM.

39

probably influence and also are influenced by the growth of the surrounding structures. The cephalic and cervical flexures involve the surrounding structures to a considerable extent so that there is a corresponding bend of the axis of the whole head, and thus the presence of them can be recognized on the exterior of the embryo. The pontine flexure, however, is limited to the nervous system. The cephalic flexure persists into adult life. The pontine flexure finally disappears and the cervical flexure nearly does.

The* formation of the pontine flexure marks a line dividing the rhombencephalon into an oral portion (metencephalon), from which the cerebellum and pons are developed, and a caudal portion (myelencephalon) which forms the medulla oblongata. The constricted portion separating the rhombencephalon from the mesencephalon (Fig. 28) is known as the isthmus. It differs from the adjoining portions of the neural tube in never undergoing any special development. It eventually forms the velum medullare anterius and through its lateral and ventral walls pass fibre tracts connecting other, parts of the neural tube.

The mesencephalon and diencephalon do not differ materially from the preceding (three weeks) stage, though in the hypothalamic region one can recognize two shallow pockets in the midline in the floor, the more oral one being the anlage of the infundibulum and the more caudal one being the mammillary recess. The optic evagination has undergone considerable modification, the inversion Table showing Subdivisions of Neural Tube and their Derivatives.

Brain

Main divisions.

Subdivisions.

Derivatives

Lumen.

Prosencephalon (Anterior vesicle)

Telencephalon Rhinencephalon Corpora striata

Cerebral hemisphere Oral end of third

Diencephalon

Optic thalamus Optic tract Hypothalamus

Colliculi Mesencephalon Mesencephalon Tegmentum (Middle vesicle) Crura cerebri

ventricle Lateral ventricle

Third ventricle

Aqueduct of Sylvius

Rhombenceph- Metencephalon alon (Poste

Cerebellum Pons

rior vesicle) Myelencephalon Medulla oblongata

Fourth ventricle

Spinal cord

Spinal cord

Canalis centralis

of its lateral wall changing it from a simple vesicle to an eye cup which remains connected with the brain wall by a narrow hollow stalk. The former constitutes the eyeball and the latter the optic

40 HUMAN EMBRYOLOGY.

nerve. The subdivisions of the telencephalon are more clearly marked in Fig. 28 than in Fig. 26. It is easy to recognize the protruding pallium separated from the optic evagination by the corpus striatum and from the thalamus by the margo thalamicus. The wall in front of the corpus striatum and adjacent to the median line (lamina terminalis) constitutes the anlage of the rhinencephalon. The subdivisions of the neural tube that can be recognized toward the end of the fourth week, in embryos about 7 mm. long, may be summarized as in the accompanying table (p. 39).

(b) The Central Nervous System at the End of the First Month.

By the end of the first month there is completed what we may call the primary stage in the growth of the nervous system. The primary neurons forming the peripheral nerves are by that time well laid down; all their chief peripheral branches and plexuses are indicated, and centrally the nerve roots can be traced into the substance of the walls of the neural tube, where the nuclei from which the motor roots arise can be definitely outlined, and the sensory roots can be recognized as forming definite longitudinal fibre bundles, extending upward and downward in the outermost layer. The higher neuron systems, however, are still in a rudimentary state, and in sections through the brain and cord at this time we see only the primary apparatus differentiated. Such co-ordinating centres as the pons, olive and cerebellum are still undeveloped, and the forebrain, further than presenting the beginning threefold division into ependymal, mantle, and marginal zones, shows little evidence of differentiation and still remains a relatively simple thin-walled tube. This period in development thus corresponds to a rudimentary nervous system in which there is found only the apparatus necessary for the simple cerebrospinal reflexes, the system of primary neurons.

The outer form of the brain and spinal cord and their relation to the body outline are shown in Fig. 86. A series of transverse sections through the same specimen is shown in Fig. 31. It is at once seen that the greatest bulk of the central nervous system is formed by the rhombencephalon and the spinal cord. Of these two the rhombencephalon is relatively the larger; it is approximately two-thirds as large as the whole spinal cord and is as large as the midbrain and forebrain taken together.

The spinal cord is largest in the cervical region and from there gradually tapers down to the coccyx, except in the lumbosacral region, where it is somewhat larger again. The tendency toward cervical and lumbar enlargements is plainly indicated. The cord in cross section (Fig. 31, A) presents a rounded quadrilateral outline. It consists principally of two thick lateral walls. These

DEVELOPMENT OF THE NERVOUS SYSTEM.

41

Ependymal zone Mantle zone

Marginal zone

Cross-section through the spinal cord at the height of the fourth cervical segment.

Lamina alaris 7 Tractus | solitarius

Tr. spinalis n. V.

Lamina, Abasalis \ N.x\

Caudal portion of the rhombencephalon with the vagus and hypoglossal nerves.

N. V. N. V. sens. motor Cranial portion of the rhombencephalon with the sensory and motor roots of the trigeminal nerve.

N. Ill

Mesencephalon with the oculomotor nerve.

Prosencephalon showing plan of the hemispheres, the optic vesicles, the third ventricle, and the infundibulum.

Fig. 31. — Series of transverse sections through the central nervous system of human embryo one month old, made from tracings taken from the model shown in Fig. 86. These are all on the same scale of enlargement and thus graphically represent the relative size and thickness of the neural tube in the different regions. S.I., sulcus limitans. Enlarged 25 : 1.

42 HUMAN EMBRYOLOGY.

are united ventrally and dorsally in the median line by what we have previously recognized as the floor plate and roof plate, which are here reduced to narrow seams. The thick lateral walls are subdivided by a longitudinal furrow, the sulcus limitans, into ventral and dorsal portions, known respectively as the basal and alar plates. The ventral or basal plate is the thicker of the two. Its thickness is largely due to the proliferating cells of the mantle layer, the anlage of the future anterior horn. Among these are grouped clusters of neuroblasts whose growing processes extend out through the marginal zone and are assembled to form the anterior nerve roots. The dorsal nerve roots enter the cord opposite the sulcus limitans and form in the marginal zone a longitudinal strand which is the forerunner of the posterior funiculi of the cord. In the marginal zone, in addition to the prominent bundle of axons derived from the posterior roots, there can be seen scattered axons running both longitudinally and transversely which belong to the co-ordinating neurons of the mantle zone. It is these that later form in large part the anterolateral tracts of the cord.

In the transition from the spinal cord to the rhombencephalon the most striking difference consists in the widening out of the thin roof plate and the accompanying flaring apart of the alar and basal plates. As a result of this the lumen of the neural tube widens out from the narrow lanceolate cleft characteristic of the spinal region to the capacious fourth ventricle. Opposite the entrance of the trigeminal nerve the lumen is larger at this time than in any other part of the neural tube.

The lateral walls of the rhombencephalon have a larger area in cross section than the lateral walls in the spinal region (Fig. 31, B, C). Their form and degree of differentiation, however, are essentially the same. There is the same sulcus limitans separating them longitudinally into basal and alar plates ; and, as in the spinal cord, there is the marginal zone containing the longitudinal fibre tracts, and the mantle zone consisting of clusters of proliferating neuroblasts which in part form the nuclei of the motor nerve roots. The ependymal zone is made up of closely packed and deeply staining primitive cells. The layer is several cells thick and the uneven line separating it from the mantle zone indicates that it is still active and giving off cells to the latter.

The arrangement of the nuclei of origin of the nerves connected with the rhombencephalon and the entering fibres of their sensory rootlets is on the same general plan as in the cord. The motor nuclei are grouped in a longitudinal column in the mantle zone of the basal plate. The fibres supplying somatic muscles pass directly ventralward through the marginal zone and emerge as the rootlets of the hypoglossal and abducens nerves, as shown in

DEVELOPMENT OF THE NERVOUS SYSTEM.

43

44 HUMAN EMBRYOLOGY.

Figs. 31, B, and 32. In the lateral part of the basal plate is the nuclear series supplying motor fibres to visceral musculature. These fibres pass lateralward and unite with the entering sensory fibres of the corresponding nerves. It is these fibres that constitute the motor elements in the trigeminal, facial, glossopharyngeal, vagus and spinal accessory nerves. This series of lateral motor nuclei may be subdivided on the one hand into a part that lies directly lateral to the somatic motor group, constituting eventually the nucleus ambiguus and the nucleus of the facial nerve, and on the other hand into nuclei massed nearer the entrance of the sensory fibres. The latter is well represented in case of the trigeminal nerve and in lesser degree in the dorsal motor nuclei of the ninth and tenth nerves. (Compare Figs. 31, B and C, and 32.) The sensory fibres enter the marginal zone near the junction of the basal and alar plates, and immediately form longitudinal tracts analogous to the posterior funiculi of the spinal cord. The entering fibres of the seventh, ninth and tenth nerves in this manner unite to form the tr actus solitarius, as shown in Figs. 32 and 96. The entering fibres of the trigeminal nerve form a similar but separate bundle. In the latter case we can recognize a cephalic limb extending to the anlage of the cerebellum and midbrain and a caudal limb extending toward the spinal region. A similar but smaller tract is formed by the entering fibres of the acoustic nerve, which at this time consists mostly of vestibular fibres. In addition to the tracts mentioned there are present in the marginal zone a few early representatives of the correlating fibres which later form the formatio reticularis and system of arcuate fibres and their longitudinal extensions. Near the median line the marginal zone is somewhat thicker from the presence of such fibres.

That portion of the alar plate in front of the trigeminal nerve constitutes the anlage of the cerebellum, but as yet it shows no apparent difference from the alar plate of the caudal half of the rhombencephalon.

The so-called rhombic grooves or transverse furrows, shown in Figs. 33, 34, and 95, are sharply marked at this time in the floor of the fourth ventricle. These grooves evidently form an important feature in the early growth of the rhombencephalon. They have been reported in a variety of different mammals (pig, sheep, dog, cat, rabbit, and rat) beside man and seem to be fairly constant in form and number. In man they are best seen during the third and fourth weeks. At first they are described as involving the whole thickness of the wall so that they can be seen both on the inner and outer surfaces of the brain wall (Gage, 1905). At the fourth week the outer surface of the wall is smooth and the grooves involve only the ependymal layer. After the fourth week the grooves rapidly disappear, leaving no marking that can be seen in the adult.

DEVELOPMENT OF THE NERVOUS SYSTEM.

45

There are six rhombic grooves. The most cephalic one is in the region of the pontine bend, and they extend from there caudally as shown in Fig. 34. They bear a constant relation to the cranial nerves, which is indicated in Figs. 33 and 95. If the grooves are labelled a,b, c, d, e, and /, then it can be seen that we have the following relations : the trigeminal nerve arises conjointly from a and b; the facial nerve (motor root) runs transversely beneath the floor of groove c, which usually is the deepest and most

N triqem (motor) N trigem (5ensJ ...-N. facialis '-'.--•- N.acusticus N.abducens N qlo55opharyrvq N vagus

hypoqlossus

Fig. 33

Fig. 34

Fig. 33. — Diagram showing relation of the cranial nerves to the floor of the fourth ventricle and the rhombic grooves, being the same specimen as shown in Fig. 34.

Fig. 34. — Dorsal view of model showing rhombencephalon of human embryo one month old, being the same model shown in Fig. 86. The thin tela chorioidea is removed, exposing the floor of the fourth ventricle and rhombic grooves. The anlage of the cerebellum is formed by the alar plates cephalad to the pontine flexure. Compare with Fig. 33.

sharply cut of all six grooves; the acoustic nerve has its attachment to the alar plate adjoining grooves c and d; the abducens nerve arises from d, a shallow and somewhat quadrilateral groove ; the glossopharyngeal nerve (motor portion) runs under the floor of the narrow groove e, and the motor roots of the vagus arise from /, which groove merges caudally into the general floor of the ventricle.

This nerve distribution is constant in the different mammals, and it is very likely that in this we have an explanation of the

46 HUMAN EMBRYOLOGY.

significance of these grooves. The predominant view regarding them heretofore has been that they are neuromeric and in a series with the spinal segments and the coarser transverse divisions of the mid- and forebrain. Instead of this, if emphasis is laid on the fact that they stand in constant relation to the lateral group of cranial nerves (fifth, seventh, ninth and tenth), then they may be fitted in with and form part of the branchiomeric system. This view has in its favor the fact that they are not only united by nerve trunks, but also numerically correspond to and are embryologically contemporary with the branchial and facial arches in the manner shown in the following table : Maxillary process \ N tri eminus f Groove a Mandibular arch J ° \ Groove b Hyoid arch N. facialis Groove c — N. abducens Groove d Third branchial arch N. glossopharyngeus Groove e Fourth branchial arch N. vagus Groove / The one discordant feature is groove d, which has no corresponding branchial arch. As yet we have no satisfactory explanation for either the aberrant course of the abducens nerve or its connection with this particular groove.

In the region of the midbrain the basal plate is much like that of the rhombencephalon and is in about the same stage of differentiation. In its mantle zone are the clusters of neuroblasts constituting the nuclei of the third and fourth cranial nerves. The fibres from the former pass through the marginal zone directly ventralward, as shown in Figs. 31, D, 32, and 86, and emerge from the hollow of the mesencephalic bend. The fibres of the fourth nerve on the other hand pass dorsalward just beneath the ependymal layer and decussate in the roof at the junction of the mid- and hindbrains and emerge directly after their decussation. No satisfactory explanation has as yet been given for this dorsal decussation of the fourth nerve.

The alar plates of the midbrain are thinned out and extend around dorsally to meet in the median line, the roof plate thus being reduced to a narrow seam. The only evidence suggesting the later development is found in the ependymal layer which forms an extensive germinal bed from whose cells are to be derived the neuroblasts composing the future corpora quadrigemina. The ependymal and mantle zones still exist as one layer. It is the outer portion that gradually becomes differentiated as the mantle zone and that gives origin to the quadrigeminal neuroblasts.

The outlines between the mesencephalon, thalamencephalon, and prosencephalon can be distinctly made out both externally and internally. The prosencephalon is characterized by a prominent lateral evagination whose lumen is to form the lateral ventricle

DEVELOPMENT OF THE NERVOUS SYSTEM. 47 and is connected with the main lumen of the neural tube by the large foramen of Monro. It is the wall of this pouch that is to form the future cerebral hemisphere, as will be described later. Two other evaginations are developed from the floor of the prosencephalon to form the special sense organs of smell and sight. The former at the end of the first month is just making its appearance in the form of a slight depression in the lumen of the anterior brain wall just lateral to the lamina terminalis and does not yet form a distinct pouch. The visual apparatus is much further advanced. As seen in Fig. 31, E, we have a well-formed optic cup connected by a hollow optic stalk with the floor of the prosencephalon at its junction with the thalamencephalon.

The thalamencephalon as yet shows little sign of differentiation, though it is possible to divide it into a ventral portion that is to form the hypothalamus and infundibulum, and a dorsal portion that is to form the thalamus. The latter is continuous with and resembles the alar plate of the midbrain.

(c) The Spinal Cord from the End of the First Month to Maturity.

A general sketch of the formation of the neural tube and the differentiation of its walls and its change in form up to the completion of the first month has been given. The further changes by which it becomes converted into the adult spinal cord now remain to be considered.

In considering the different elements taking part in its further development it is important to follow the subdivision of the wall of the cord into its three constituent layers or zones (ependymal, mantle, and marginal), tracing the fate of each up to the adult condition. At the same time one should keep in mind the foursided form of the cord, consisting of two lateral walls united ventrally by a floor plate and dorsally by a roof plate. The roof and floor plates retain their primitive characteristics throughout and are only modified secondarily, due to the changes in the adjoining portions of the lateral walls. The lateral walls on the other hand undergo enormous growth, and it is the character of their thickening that determines the shape of the cord.

The lateral walls are subdivided into a ventral or basal plate and a dorsal or alar plate, the junction between the two corresponding to the sulcus limitans. These are also known as the anterior "Markcylinder" and posterior "Markprisma" of His. From the first of these are developed the anterior gray columns or horns, the motor nerve roots, and the surrounding funiculi of longitudinal fibres (anterolateral ground bundles). From the dorsal or alar plates are developed the posterior gray columns, the substantia gelatinosa and the dorsal funiculi into which the posterior nerve

48 roots enter. The basal plate is primarily a motor apparatus and the alar plate is primarily a sensory one.

The basal and alar plates are united by an isthmus-like intermediate portion known as the "Schaltstiick" of His. This area, owing to the shrinking of the tissues, particularly the radial framework fibres, in prepared specimens is usually sharply demarcated on the surface of the cord by two longitudinal furrows, the marginal groove directly in front of the entering posterior roots and the cylinder groove at a point about midway between there and the emerging anterior roots. From the gray substance of this area are developed many of the internal arcuate fibres, and the adjoining marginal zone becomes converted into the formatio reticularis and supplies a pathway for the spinocerebellar and cerebrospinal tracts.

In Figs. 36, 37, 38, and 39 are shown the typical stages in the conversion of the neural tube with its three primitive zones into

ex..

mm.

30 mm.

45 mm.

80 mm.

Fig. 35. — Diagram showing the fate of the ependymal layer and the formation of the posterior median septum and the central canal of the spinal cord. The approximate length (crown-rump) of the embryo is indicated in mm. below each stage, s. g., dorsal gray column and substantia gelatinosa.

the more complicated solid cord as found in the adult. Starting with the ependymal zone and tracing it through these successive stages we meet with the appearances that are schematically shown in Fig. 35. To facilitate comparison these are all drawn on the same scale of enlargement.

It can be seen that at the beginning of the second month (15 mm.) the lumen of the cord is still relatively large, and that from that time on up to embryos of 80 mm. it decreases in actual size and still more so in size relative to the size of the cord. The shape of the lumen passes through characteristic changes. As seen in transverse section it is at first (15 mm.) an elongated oval slit, and is wider in the dorsal half. At 30 mm. the condition is reversed and the lumen is" wider in the ventral portion, while the dorsal portion is reduced to a narrow slit. Eventually the dorsal portion becomes obliterated and there onlv remains the rounded ventral

DEVELOPMENT OF THE NERVOUS SYSTEM.

49

portion, and it is this that forms the permanent central canal. The longitudinal furrows that have been described as indenting the sides of the lumen are probably to be classed as shrinkage phenomena, since they are not present in razor sections of unembedded pig embryos. The most constant of these furrows is the sulcus limitans at the junction of the basal and alar plates.

The ependymal cells that form the ventral portion of the lumen differ already in the 15 mm. stage in character and arrangement from those forming the dorsal portion. They are more compactly arranged and form a narrow band that is more sharply separated from the adjacent mantle zone than is the case in the dorsal

Funic, posterior

Zona epend- h '

Radix post.

Zona mant.

Zona marg.

Radix ant.

Col. ant.

Funic, anterolat.

Fiq. 36. — Cervical spinal cord of a human embryo 15.5 mm. long. Enlarged 60 : 1. Dp., roof plate; Bp., floor plate, in which the anterior commissure is developing. (After Bryce.) portion of the cord. If we may regard the ependymal zone as a germinal bed delivering proliferating cells to the mantle zone then it is apparent that this phenomenon is practically completed in the ventral portion and the remaining cells are entering the resting stage, while the process is still in active operation in the dorsal portion. The ventral portion of the ependymal zone, like the ventral portion of the cord in general, may be regarded as further advanced in its development than the corresponding dorsal portions.

In embryos from 15 mm. on, coincident with the formation of the dorsal columns of gray matter, there is a gradual subsidence of the proliferation of ependymal cells around the dorsal portion, and the whole ependymal border comes to a resting stage Vol. II.— 4

50 and forms a narrow, sharply demarcated border for the central canal. In doing this the size of the canal is decreased through the approximation and fusion of the walls of its dorsal portion. In Fig. 37 this process of fusion is in active operation. As the walls come together the cells lose their radial direction and form a unilateral compact strand which is soon replaced by a sparsely nucleated seam of supporting tissue like that forming the framework of the mantle zone. It is the extension of this seam of closure that eventually forms the posterior median septum. The ventral end of the septum is a fixed point and its further growth

Septum medianum posterius Funic, gracilis Funic, cuneatus * i * \ / Substantia gelatinosa »

Radix post

h S.m.

Col. ant. -4 -- —

Radix ant.,-.

— -"Funic, lat.

•I. » ' >

Fiss. med. ant.

Funic, ventr.

Fig. 37. — Cervical spinal cord of a human embryo 30 mm. long (Huber collection, No. 15). Enlarged 60 : 1. S.m., junction of lateral and posterior funiculi, which point in shrunken specimens is marked by a deep groove.

and elongation, coincident with the development of the dorsal funiculi, must be considered as taking place principally at the dorsal end.

The development of the mantle zone is closely associated in the earlier stages with that of the ependymal zone and the line of demarcation between them is ill-defined. As has been described in the section on histogenesis of nervous tissues the mantle and ependymal zones were originally one common layer, and the mantle zone may be regarded as a proliferation and differentiation of the outer ependymal cells. Later, as the anterior and posterior columns of gray substance begin to take form (15 mm.), the ependyma

gradually enters upon its resting stage and from then on becomes sharply marked off from and takes no further part in the development of the mantle zone.

On comparing Fig. 36 to 39 it is seen that in 15 mm. embryos we can already speak of an anterior gray column (horn) which is composed of a supporting framework and clusters of developing neuroblasts. The processes from the neuroblasts are assembled into rootlets which emerge on the ventrolateral border of the cord as the anterior nerve roots. The anterior columns in the later stages enlarge and their contour becomes irregular and eventually

Septum medianum posterius

Funic, gracilis

Funic, cuneatus

Radix, posterior Substantia, gelatinosa

Canal is centralis

Columna "~S~~ post.

•Funic, lat.

.Columna lateralis

N Columna anterior

Commissura anterior

Fissura med. ant.

Radix ant.

Fig. 38. — Cervical spinal cord of a human fetus 45 mm. long (Huber collection, No. 18). Enlarged 45 : 1. This figure should be compared with Figs. 36, 37, and 39, all of which represent the same region in its different stages of development.

we can recognize a lateral division, the so-called lateral horn. In the 80 mm. embryo (Fig. 39) there is presented practically the adult form. The enlargement from 30 mm. on consists partly in the elaboration of the supporting framework and partly in the increase in size of the contained neuroblasts. The growth of the latter involves also their processes, so each ventral column would become larger through the growth of its own processes as well as the invading processes from other portions of the mantle zone. The elaboration of the supporting framework and development of the processes of the neuroblasts results in a greater separation of the neuroblasts from each other and gives the mantle zone the appearance of being more sparsely nucleated. There arc also cer

tain supplementary factors in the growth of this tissue due to the process of vascularization and later due to the acquisition of myelin sheaths.

The formation of the dorsal gray columns (posterior horns) occurs somewhat later than the ventral ones. In the 15 mm. embryo (Fig. 36) it is possible to outline that portion of the mantle zone that is to form them. They possess, however, very little at that time either in structure or form that is characteristic; the cells of the ependymal zone of that region are still actively crowding outward to become incorporated in the mantle zone. In em

Septum medianum posterius

Radix post

Funic, gracilis Funic, cuneatus

,V",m-'

Substantia gelatinosa

Funic. "- ~ ~ \ - -lateralis

Radix ant — Canalis centralis Fissura med. ant. Columna anterior Fig. 39. — Cervical spinal cord of a human fetus 80 mm. long. Enlarged 30 : 1.

bryos of 30 mm. the outline of the dorsal columns commences to take form and masses of neuroblasts group themselves so as to form a cap at the dorsal border, which eventually becomes the ganglionic mass known as the substantia gelatinosa. The further elongation of the mantle zone into typical posterior columns can be seen by comparing Figs. 37, 38, and 39.

As has already been shown the roof plate as such disappears with the formation of the posterior median septum. The floor plate, however, persists and is formed in part by the ependymal cells which extend from the lumen radially outward to the surface of the cord, forming the bottom of the anterior longitudinal fissure. In addition to the ependymal cells the floor plate is made i up of the processes from the heteromeric neuroblasts of the mantle

DEVELOPMENT OF THE NERVOUS SYSTEM. 53 zone, whose decussation eventually constitutes the anterior commissure. Neuroblasts of this character are found in considerable numbers in the inner portion of the mantle zone and are analogous to the internal arcuate fibres found in the hindbrain. The processes forming the anterior commissure cross the median line in the mantle zone. A marginal layer can scarcely be said to exist in the region of the floor plate.

The development of the marginal layer (Eandschleier of His) and its conversion into the white substance of the cord are dependent upon the foreign cells whose axons penetrate this zone and thread their way through its meshes forming the longitudinal fibre tracts of the cord. These axons may be classed into five main groups: (a) dorsal root fibres arising from the spinal ganglia; (b) short intersegmental fibres (ground bundles) arising from neuroblasts of the mantle zone and serving to connect adjacent levels of the cord; (c) long suprasegmental fibres connecting the cord nuclei with higher centres; (d) long descending fibres connecting nuclei of the hindbrain with lower levels in the cord; and (e) long fibres forming the descending palliospinal tracts. Of these fibre groups (a) and (b) make their appearance first (end of first month). Very soon afterward groups (c) and (d) appear (beginning of third month). The last fibres to appear are those belonging to group (e) (end of fifth month). The axons belonging to group (a) form a tract of fibres (funiculi posteriores) that always remains separated from the remaining fibres. Groups (b-e) partly merge into one another forming the anterolateral funiculi and thus cannot be so sharply outlined. In shrunken specimens the radial fibres of the marginal zone contract and cause the longitudinal fibres to present the appearance of being arranged in distinct bundles. The posterior funiculi in such specimens stand out with exaggerated distinctness.

The anterolateral funiculi do not meet in the middle line owing to the interposition of the floor plate. The latter plays only a passive part in the further growth and thus subsequent to the growth of the adjacent mantle and marginal zones there is formed the deep anterior median fissure, in which are found the nutrient blood-vessels of the cord. The formation of this fissure is readily seen by comparing Figs. 36 to 39.

The growth of the posterior funiculi is dependent on the addition of new fibres derived from the spinal ganglia and also on the extension upward of overlapping fibres from lower segments of the cord. At the end of the first month they form an oval bundle in the marginal zone, and when a piece of cord is examined with a low-power lens these fibres can be seen forming a white longitudinal band on the dorsolateral surface of the cord. In the median line the marginal zone is very thin or absent, so that the

54 HUMAN EMBRYOLOGY.

nuclear substance of the ependymal and mantle layers can be seen projecting between the posterior funiculi of the two sides. As these bundles enlarge they become thicker and spread toward the median line, where they eventually meet. As they meet in the median line they bend forward, as is shown in Fig. 35, and fill in the space left by the receding ependyma. It may be supposed that their presence plays a part in the stimulation of the production of the posterior median septum, which forms between the right and left halves. As these fibres crowd ventralward the posterior gray columns extend dorsolateralward and the resulting form of the combined posterior funiculi is wedge-shaped as found in the adult. In the cervical region in embryos between 20 and 60 mm. (Figs. 37 and 38), there is a V-shaped portion of these funiculi in the dorsal part at the median line that differs in appearance from the remainder. This is regarded (His) as the primitive column of Goll. The difference in appearance is doubtless due to the fact that it consists entirely of longitudinal fibres destined for the gracile nucleus at the cephalic end of the cord, and is not constantly giving off collaterals to the gray substance as the other fibres of the funiculus seem to do. In the process of shrinkage such collaterals draw in or flatten the remaining funiculus, while it leaves the dorsomedian wedge (column of Goll) unaffected. The latter consequently stands out prominently in all shrunken specimens. In older cords, as the supporting framework becomes more complete, the contrast between these two portions becomes less noticeable.

The myelinization of the fibres of the cord does not begin until about the fifth month of fetal life and is not completed until between the 15th and 20th year (Flechsig, '90, Popoff, '88, Bechterew, '87, and Trepinski, '98). It becomes first apparent in the anterior and posterior roots and the ventral commissure. Very soon afterward the ventrolateral ground bundles begin to show scattered myelinated fibres and likewise a portion of the posterior funiculi.

Three typical stages in the process of myelinization of the cord are shown in Fig. 40. On comparing these it will be seen that the formation of myelin occurs along certain tracts or systems of fibres, and owing to this difference in the time at which they acquire their myelin we are able to map out the different levels of the cord into definite functional areas. One of the latest systems to become myelinated is the pyramidal tract, which can still be seen at the seventh month (Fig. 40) as an open area. It becomes myelinated between the ninth month and second year.

Though the anterior and posterior roots show the beginning of myelinization about the same time, the process is simpler in the anterior roots, and their myelinization is uniformly completed at a

DEVELOPMENT OF THE NERVOUS SYSTEM.

55

time when not more than one-half of the fibres of the posterior roots show any myelin. The fibres of the posterior roots become myelinated in a series of rather definite stages. Each root can thus

Rz.

ant.

...

-j.

\ Tȣ

% m» itfr

Z. med.

Lumbar cord, fifth month

Cervical cord, fifth month Fasc. gracilis

-/*

Rz.

med.

Rz. ant.

_ a

US

k' . -M 1

c

Lumbar cord, at five and a half months . Rz. pott. med.

• *

E

Lumbar cord, seventh month

~ Pyr. lot.

". ^v

Fasc. lot. proprim

Pyr. vent, D Cervical cord, at five and a half months

Z. med.

Rz.'post. lot.

Rz.

post.

med Fasc. cerebello- \ spinalis \ \ a •

$»

91

— ».'

Pyr. lot.

)

'Cervical cord, seventh month

Fig 40. — Diagram showing the myelinization of the spinal cord at different periods in fetal life. Rz. ant., anterior root zone; Rz.med. median lu "t zone; Rt post med., median posterior root zone; Rz. post tat., lateral posterior root zone (Lissauer); Z. med., median zone; Pyr. tat., fasciculus cerebroepinalis lateralis; Pyr. ventr., fasciculus cerebrospinalis v entrails. (After Trepinski.) be subdivided into a number of different fibre groups. Likewise on the basis of this difference in their myelinization the fibres of the posterior funiculi have been subdivided into five different em

56 HUMAN EMBRYOLOGY.

bryonic fibre systems, which present a characteristic grouping in the different levels of the cord, as shown in Fig. 40. These systems are as follows : 1. Anterior root-zone, consisting of fibres from posterior roots which after a longer or shorter course in the posterior funiculi disappear in the anterior portion of the posterior horns (Rz. ant.. Fig. 40). It extends throughout the whole length of the cord and is the first system in the posterior funiculi to begin myelinization.

2. Middle root-zone consists of a group of fibres derived from the posterior roots, which lie between the anterior and posterior root-zones. This zone may be divided into two divisions, first and second, the former becoming myelinated in embryos 19 to 20 cm. long and the other somewhat later. The fibres of the first system after a short course enter the column of Clark, and in regions where this is absent they are lost in the gray substance connecting the anterior and posterior horns.

3. Posterior root-zone is divided into a median portion and a lateral portion. The fibres of the median portion pass mostly forward and after entering the posterior horn they extend to the region of large ganglion cells in the anterior horn. The fibres of the lateral portion are apparently derived from the posterior roots and end in the substantia gelatinosa, constituting Lissauer's column.

4. Median zone lies against median septum and is most distinct in the cervical and upper two-thirds of the thoracic region. The course of its fibres is not known, but it is apparently distinct from the fasciculus gracilis.

5. Fasciculus gracilis (column of Goll), commencing at the tenth thoracic segment, extends upward in a compact bundle to reach the nucleus gracilis. The source of its fibres is not known.

It will be seen that three of these systems are derived from fibres in the posterior roots, and in two of them the source of the fibres is not definitely known. In the order of their myelinization the different systems may be grouped as follows : First stage Anterior root-zone Second stage Middle root-zone (first division) Median zone Third stage Fasciculus gracilis, Middle root-zone (second division) Posterior root-zone (median division) Fourth stage Posterior root-zone (lateral division) The caudal end of the spinal cord exhibits certain departures from the uniform development characterizing the rest of it, to which special attention may be directed. If one examines a sagittal series through an embryo 11 cm. long, as shown in Fig. 41, it can be seen that the extreme tip of the cord lying in the tail anlage has been closed off to form a simple epithelial sac. The lumen of

DEVELOPMENT OF THE NERVOUS SYSTEM.

57

the cord above this point becomes obliterated and there results a slender solid strand of nervous tissue which we know as the filum terminale. The epithelial sac becomes the vestiges medullaires coccygiens of Tourneux and Herrmann, whose development is described by Tourneux (Precis d'embryologie humain, second edition, 1909, pp. 348, 349) as follows: At the beginning of the third month the neural tube still extends to the extreme end of the vertebral column into the tail bud, and its slightly enlarged tip is closely united to the deep layers of the skin. Toward the end of the third month the spinal column, developing faster than the soft parts, draws along the part of the neural tube that is adherent to it and whose extreme tip remains attached to the skin. As a

^ ,'Conus medullaris] ..Vertebra sacralis

Ventri cuius erminalis

Arteria sacralis media

Periosteum

Vestiges coccygiens

Ligament urn caudal"

Filum terminale

Fig. 41. — Schematic median section through the caudal end of a human fetus 11 cm. long (crownrump), showing the formation of the vestiges meclullaires coccygiens and its relation to the filum terminale(After Unger and Brugsch.) result of this unequal growth the terminal or coccygeal portion of the neural tube becomes bent in the form of a loop, the more deeply situated limb of which is attached to the posterior surface of the coccyx (segment coccygien direct), and the other more superficial limb extends obliquely from a caudal and ventral position to one more dorsal and cranial (segment coccygien reflechi). During the course of the fourth month the more deeply situated limb, the segment coccygien direct, atrophies and disappears, while the more superficial one, the segment coccygien reflechi, continues to develop into the fifth month and gives origin to cell cords or cell masses which contain cavities lined either with prismatic or pavement epithelium; these are the vestiges medullaires coccygiens or paracoccygiens. These structures from the sixth month on suffer a progressive atrophy, but it is possible to recognize traces of them up to the time of birth.

58 The caudal end of the central canal extends through the conus medullaris to the beginning of the filum terminale. At its lower end it undergoes a conical expansion out of which open irregular side pouches and occasionally an elongated blind sac giving the canal the appearance as though the lower end were bent on itself (see Fig. 42). This caudal enlargement of the canal is known as the ventriculus terminalis.

During the earliest stages, up to the time of the highest development of the anlage of the tail, the spinal cord, chorda and mesoderm develop at the same rate. With the reduction of the tail bud the reduction first occurs in the mesoderm, and thus the spinal

Ventriculus terminalis

Conus medullaris

Filum terminale

Fig 42 Fig. 42. — Caudal end of the central canal of the spinal cord in a human fetus 9 cm. long (crown-rump), showing the formation of the terminal ventricle. (After Unger and Brugsch.) Fia. 43. — Spinal cord exposed from behind in a three months old embryo, at which time the cord still extends to the caudal tip of the vertebral canal. (After Kolliker.) cord temporarily becomes longer than the vertebral column. With the formation of the coccygeal process there begins a relative increase in the rate of growth of the vertebral column. At the third month, Fig. 43, the cord is about the same length as the vertebral canal, but from the third month on into adult life the vertebral column becomes increasingly longer than the cord. The cord is more precocious than the skeletal system and reaches its full size before the latter has finished its growth.

Owing to this unequal growth during the latter part of fetal life there is a gradual change in the position of the cord in the vertebral canal. Since the cephalic end of the cord is fixed this unequal growth results in its caudal end being drawn upward away from the lower end of the canal. At the time of birth that part of the cord from which the coccygeal nerves arise is found

DEVELOPMENT OF THE NERVOUS SYSTEM. 59 opposite the third lumbar vertebra; in the adult it is opposite the first lumbar. In this process of shifting, the caudal tip of the cord remains attached to the coccyx, and becomes stretched out into the slender filum terminale. Likewise the nerve roots, their ganglia, with the exception of the ganglion coccygeale, having already become attached in the intervertebral foramina, become stretched out and come to lie in an oblique direction, the most caudal root being longest and most oblique. There is thus formed the cauda equina.

Concerning the development of the blood-vessels of the spinal cord the reader is referred to the corresponding section in the chapter on the development of the blood and vascular system.

The development of the membranes of the cord has been worked up in greater detail in other mammals than in man. According to Sterzi (1900), in the sheep embryo 15 mm. long there is a meningeal mesenchyme, which in embryos 20 mm. long forms a definite membrane. In the 80 mm. embryo this membrane becomes differentiated into an outer layer or dura mater and an inner layer or meninx secundaria, the two being separated by an intradural space. The dural layer is separated externally by the epidural space from an endorhachide, which resembles the dura but is always distinct from it. Finally in the 157 mm. embryo the meninx secundaria is further differentiated into an outer layer or arachnoideal coat and an inner layer or pia mater as seen in the adult.

(d) Development of the Hindbrain from the End of the First Month on.

Up to the end of the first month the rhombencephalon passes through the same general process of development that has been described for the spinal cord. It undergoes the same differentiation into ependymal, mantle and marginal zones. Its walls also are divided into two lateral plates, united ventrally by a floor plate and dorsally by a roof plate. The latter is very broad and is thinned out so as to form an extensive membrane covering in the lumen of the tube. The lateral walls, as in the case of the spinal cord, are subdivided longitudinally by the sulcus limitans into a median basal plate which is chiefly motor, and a dorsolateral alar plate which is chiefly receptive. From the first month on, however, the exuberant growth of intersegmental neurons composing the reticular formation and the development of the suprasegmental ganglion masses and their respective tracts rapidly diminish the resemblance between hindbrain and spinal cord.

At about the fifth week certain alterations occur in the outward appearance of the rhombencephalon, indicating the changes going on within its walls. The most conspicuous is the bending of the axis of the tube resulting in the pontine flexure, as shown in Fi<r.

60 30. Whether this is simply due to an overgrowth of the tube in its long axis or whether there is the additional factor of unequal growth of different portions of the wall has not been clearly shown.

Along with the bending there occur noticeable changes in the broad ependymal roof of the fourth ventricle. As shown in Fig.

31, B, and C, the fourth ventricle is completely covered in by the expanded roof plate consisting of a thinned-out layer of ependymal cells, which is attached laterally to the border of the alar plate,

Cerebellum

ii. Fig. 44. — Reconstruction showing cephalic portion of the rhombencephalon and adjoining midbrain at end of second month (human embryo 30 mm. long, Mall collection. No. 86). The thickened alar plates form the anlage of the cerebellum and the two halves are still separate. Caudally they are continuous with the nucleus of the acoustic nerve.

the transitional line forming the rhombic lip. Owing to the changes occurring in the alar plate that, particularly in the cephalic half of the rhombencephalon, cause it to become everted and folded back on itself, the rhombic lip becomes partly fused to what was originally ventricular surface of the alar plate. The changes occurring in the rhombic lip may be seen by comparing Figs. 34, 44, and 45, where it is represented as a cut edge, the whole ependymal roof being removed. It is this rhombic lip that forms the taenia of the fourth ventricle and the obex at its caudal apex.

DEVELOPMENT OF THE NERVOUS SYSTEM.

61

At the same time with the formation of the pontine flexure there is produced a transverse fold (plica chorioidea) in the ependymal roof. This extends outward on each side into the lateral recess which is formed by the overgrowth of that part of the alar plate that is to form the cerebellum and tuberculum acusticum (see Fig. 45). This marks the beginning of the chorioid plexus of the fourth ventricle. The chorioid plexus is formed by ependymal epithelium and its covering of vascular mesoderm. They unite in forming minute villous-like folds which project within the lumen of the ventricle. These folds are first found along the line of the

Brachium quadrigeminum inferius I

Vermis

Lobus laterahVcerebelli / /

Lemniscus , y lateralis

N. V ' v Rhombic lip Corpus pontobulbare N. VII N. VIII Fiq. 45. — Reconstruction showing the outer form of the hindbrain at the end of the third month (human fetus 9.5 cm. long, Mall collection, No. 146. 1

transverse chorioid fold, and from there the formation spreads caudally until nearly the whole posterior medullary velum is converted into a chorioid mass, the tela chorioidea inferior. Originally the fourth ventricle is completely roofed in as has been seen, but later there are found apertures which are formed secondarily, one in the caudal portion, the foramen of Magendie, and one in each lateral recess, the foramina of Luschka. The presence of these foramina has been denied by some investigators.

Before referring to further changes in the outward form of the rhombencephalon we will consider the histological changes that produce them.

As has already been seen, by the end of the first month the primary neurons belonging to the cranial nerves are clearly differentiated. In Figs. 86, 32, and 96 their relation to each other and

62 HUMAN EMBRYOLOGY.

to the walls of the neural tube are shown. Their relation to the rhombic grooves is shown in Fig. 33. The further growth of the rhombencephalon results from the elaboration of neuroblasts from the ependymal and mantle zones which form the following structures: (a) receptive nuclei for the cranial nerves, comparable to the dorsal horns of the spinal cord, the axons of which form the median and lateral lemniscuses; (b) intersegmental neurons, constituting the reticular formation for co-ordination of different groups of primary neurons; (c) suprasegmental nuclei with afferent tracts which are laid down subsequent to and hence are external to the reticular formation, of which the olive is a conspicuous example; (d) efferent tracts from cerebellum and midbrain; and (e) the descending pyramidal tract from the cerebral cortex.

The neuroblasts derived from the basal plates are chiefly those that form effector fibres for the cranial nerves. As they will be described at length with the peripheral nervous system, their description here will be omitted. It should be pointed out, however, that their growth is relatively precocious, and that the differentiation of the basal plate begins first and is finished before that of the alar plate. The ependymal layer comes to the resting stage early, about the end of the second month. The subsequent development of the basal plate, aside from the part it takes in the differentiation of the formatio reticularis, is passive, and is dependent on the invasion of neuroblasts from the alar plate and the ingrowth of foreign fibres in its marginal zone.

The differentiation of the formatio reticularis is not confined to the basal plate, though it is first apparent there. As in the spinal cord, it consists of intersegmental neurons derived from the mantle zone, the processes of which to a large extent cross the median line as internal arcuate fibres. The crossing of these fibres marks the beginning of the raphe. After crossing they form a. longitudinal bundle in the marginal zone analogous to the ventral ground bundle in the spinal cord, and corresponding in position to the median longitudinal fasciculus of the adult oblongata, though, as will be pointed out, the latter contains other fibres in addition to these. The processes of some of the more lateral neuroblasts, instead of crossing within the mantle zone, penetrate the marginal zone and make their way along its surface, thus forming external arcuate fibres. In addition to the internal and external arcuate fibres (heteromeric), the formatio reticularis is early characterized by radially directed neuroblasts whose processes extend toward the marginal zone to form tautomeric intersegmental fibre tracts. The reticular formation is eventually subdivided into a gray portion (formatio reticularis grisea) containing cell bodies and shorter tracts, located in mantle zone, and a white portion (formatio reticularis alba) consisting of long tracts

DEVELOPMENT OF THE NERVOUS SYSTEM.

63

and located in the marginal zone. The great development of the reticular formation is an important determinative factor in the morphology of the adult oblongata. Three stages in its development are shown in Figs. 46 and 47. In the same figures is shown the conversion of the original floor plate into the median septum and raphe. It will be noticed that, as in the case of the spinal cord, the marginal zones of the two sides do not fuse, being always separated by the prominent radial processes of the ependymal cells, which extend from the lumen to the surface of the Drain.

Wandering cells from the alar plate

Wandering cells from the rhombic lip

Tractus solitarius

Marginal zone Mantle zone

N.XII

Septum medulla

Fig. 46. — Transverse section of medulla oblongata of a human embryo at end of fifth week (10.5 mm. long), showing ventromedial migration of neuroblasts from the alar lamina and rhombic lip. (After His, 1891.) It is in the development of the alar rather than the basal plates that the rhombencephalon departs so widely from the type found in the spinal cord. While the initial changes have been taking place in the basal plate the cells of the ependymal zone of the alar plate have been actively separating off to join the mantle zone, preparatory to forming receptive nuclei for the peripheral nerves, as well as other nuclear masses, making up intersegmental and suprasegmental tracts and centres. Originally, as has been seen, the afferent peripheral fibres on entering the wall of the neural tube unite to form longitudinal tracts which extend upward or downward in the marginal zone over a varying number of s< ments. In the rhombencephalon such tracts are represented by

64 the tractus solitarius, spinal limb of the n. trigeminus, and fibres coming up from the posterior funiculi of the spinal cord. Later there are added the fibres of the restiform body and fibres from the lateral funiculi of the cord. Thus originally we have here as in the spinal cord a central gray portion sharply marked off from a peripheral white portion, the latter consisting of distinct funiculi. This resemblance is, however, very soon diminished

Tractus solitarius

M !'( i - i .

ife

N.XII

Tela chorioidea

N.XII Tractus solitarius

Formatio reticularis grisea

Rhombic lip

Corpus resti forme Tractus spinalis trigemini

Formatio reticularis alba Anlage'of the nucleus olivaris access, med.

Fig. 47. — Transverse sections of the medulla oblongata, showing the development of the formatio reticularis and olivary nucleus. A, embryo 13.6 mm. long (5 weeks). Enlarged 40 : 1. B, embryo 22 mm. long (8 weeks). Enlarged 10 : 1. (After His, 1891.) by the profuse growth of the alar mantle zone which invades the marginal zone partially enveloping the tracts and spreading them apart.

The profuse proliferation of the neuroblasts in the alar mantle zone is indicated in Fig. 46, and by comparing this with Fig. 47, A and B, the eventual fate of these cells can be seen. They form clusters along the tractus solitarius, and the descending fibres of the n. vestibularis and n. trigeminus which become the receptive nuclei for these particular fibres (nucleus tracti solitarii, nucleus vestibularis spinalis and substantia gelatinosa). At the caudal

DEVELOPMENT OF THE NERVOUS SYSTEM. 65 end in a similar way the gracile and cuneate nuclei are formed in which the fibres from the posterior funiculi of the cord terminate. From these receptive nuclei axons are developed which make their way largely as internal arcuate fibres through the formatio reticularis, decussating to the opposite side to form a longitudinal tract near the median line (lemniscus medialis) which forms an afferent path to the midbrain and thalamus.

In addition to these receptive nuclei there are other nuclei formed from the alar neuroblasts that serve as connecting paths to the suprasegmental centres, cerebellum and forebrain. The most conspicuous of these are the pontine nuclear mass and the nucleus of the olive. These are formed by virtue of the extensive power of migration possessed by the cells of the alar plate. The formation of the olivary nucleus is shown in Figs. 46 and 47. The migratory process begins at the beginning of the second month. At this time massed cells can be seen making their way through the mantle zone toward the median line, the cells from the median portion of the alar plate passing median to the tractus solitarius and the cells from the lateral border passing lateral to it and separating it from the marginal zone to which it originally belonged. Toward the end of the second month a distinct group of these migratory neuroblasts have assembled near the median line, constituting the anlage of the median accessory olive (Fig. 47, B). By the third month subsequent groups are added laterally to form the convoluted inferior olivary nucleus. The majority of the olivary neuroblasts are probably the products of proliferation of migratory cells after the completion of their migration. Their axons decussate and join with spinal cord fibres to form the restiform body. The restiform body can be recognized by the eighth week.

At the extreme lateral border of the alar plate at the rhombic lip are found cells which retain their primitive embryonic appearance into adult life. Others invade the marginal zone and emerge on the surface of the wall. Some of them then migrate toward the median line by a superficial path peripheral to the marginal zone. The exact fate of these cells remains to be studied. It is possible that the more caudal ones take part in the formation of the arcuate nuclei, and possibly also the olivary nuclei. The more cephalic ones just back of the acoustic region form a narrow migratory path from the rhombic lip around to the ventral surface of the pontine flexure. This path persists in the adult as a fibroganglionic band known as the corpus pontobulbare, described by Essick (compare Fig. 49). It is possible that it is the proliferation of these cells that produces the nuclei of the pons. It is also possible that the pontine neuroblasts come from the mantle zone of that region and reach the surface by emerging through the mar Vol. II.— 5

66 ginal zone, as apparently happens with the cortical cells of the cerebellum. The pons makes its appearance between the second and third month. The axones from its proliferating neuroblasts decussate across the median line in front of the formatio reticularis and the lemniscus medialis, and pass to the cerebellum on the opposite side forming the brachium pontis. As the pons is developing the corticospinal fibres (pyramidal tract) make their way along its ventral surface and become enveloped among its proliferating cells and fibres. Accompanying the corticospinal tract are other fibres from higher centres which terminate among the cells of the pons and thus become connected with the cerebellum.

The alar plates of the cephalic end of the rhombencephalon undergo extensive and specialized development. They form that which eventually becomes the largest part of the hindbrain, i.e.,

Tractus solitarius

Fig. 48. — Transverse section of medulla oblongata of a human embryo 9.1 m m . long, showing folding of rhombic Up (Rl.). This feature is most marked in the region of the pontine flexure.

cerebellum, and in addition the acoustic nucleus. As can be seen by comparing Figs. 34, 44, and 45, the acoustic nucleus is formed from the thickened rhombic lip at the pontine flexure, at the point where the lateral recess develops. The rhombic lip becomes everted and is folded against and is partially fused with the lateral surface of the remaining alar plate, as is schematically shown in Fig. 48. The fibres of the acoustic nerve enter at the lower edge of this mass. As the trunk of the nerve differentiates itself into cochlear and vestibular portions (embryos 20-30 mm. long), the terminal nucleus also becomes differentiated into a median portion (vestibular) and a lateral portion (cochlear). These become more sharply separated from each other by the development of the restiform body whose fibres pass between them. The vestibular nucleus remains closely connected with the cerebellum and most of its axons terminate there. The cochlear nucleus remains quite independent of the cerebellum. Its axons pass across the ventral

DEVELOPMENT OF THE NERVOUS SYSTEM.

border of the reticular formation and decussate to form the trapezium. The superior olive in which many of these fibres terminate can be seen by the eighth week, apparently developing from migrant alar neuroblasts in a similar way to the inferior olive. The axones from the superior olive and trapezium extend forward along the ventrolateral border of the formatio reticularis to the inferior colliculus constituting the lateral lemniscus. This path is partly nuclear and partly fibrous, and is virtually a forward extension of the superior olive. Later the fibre element predominates through the increase in number and length of fibres, resulting in the separation of the superior olive from a smaller nucleus, the nucleus of the lateral lemniscus, which were originally one continuous structure. From its position it can be seen that the trapezium is laid down subsequent to the formatio reticularis. Still later occurs the descent of the corticospinal tract which with the growth of the pons nearly entirely covers in the trapezium on its ventral surface.

Before considering the development of the cerebellum it may be pointed out that the characteristic features of the adult rhombencephalon are only the result of the further growth of the structures that have been mentioned. In the floor of the ventricle we meet with swellings produced by the nuclei of the hypoglossal and abducens nerves, lateral to which is a longitudinal furrow representing the sulcus limitans. Lateral to this furrow are the structures derived from the alar plate, including the vestibular field and the terminal nuclei of the trigeminal and vagoglossopharyngeal nerves (ala cinera). Secondary tracts and nuclei invade the floor of the ventricle producing the characteristic striae acusticae, nucleus intercalatus and funiculus teres. The ventricle is closed in caudally by the rounded elevations (clava and cuneus) caused by the large gracile and cuneate nuclei. The olivary nuclei produce lateral swellings on each side (olives), and ventrally emerging through the pontine nuclear mass are the prominent corticospinal tracts (pyramids).

THE CEKEBELLTJM.

The character of the cerebellum at the end of the first, second and third months is shown in Figs. 34, 44 and 45 respectively. This covers the period from the time when it exists as simple bilateral alar plates to the time when it fuses across the median line as a transverse mass consisting of a median vermis and two lateral lobes. Its later enlargement and the formation of its characteristic lobes and fissures are shown in Fig. 49.

At the end of the first month the alar plates of the rhombencephalon cephalad to the pontine flexure differ very little from the rest. They are a little thicker and present a moderately convex

68 Pyramis

Lobus lateralis,

I Uvula ! i Nodulus

"Flocculus Corp. pontobulbare

A. 9.5 cm. (3 Mo.)

Uvula i

Fissura prima \ v S.postlunaris

Lobus lateralis

P V raml5 \ ! Nodulus Tonsilla

'Flocculus

Corp. pontobulbare

Paraflocculus-Fiocculus

B. 4 Mo.

S.floccularis ef postnodularis

S posttonsillar ! Pyramis Tonsilla (Fiss. secunda) | r ,, , ! . Uvula i u e • I. I Lob. postero-mfenor '. ! ! | Tuber ! I <- , • S. parapy rarrwdalis ; ! j i ! ; S.posrlunans r r y , , , ' ; Nodulus

Fissura prima / / 5. horiz. maq.

S.horizontalis magnus ',

Lob. biventer

Paraflocculus,

flocculus \ 'Corp. pontobulba / Flocculus F'araf locculus Fig. 49. — Three stages in the development of the fissures and convolutions of the cerebellum, as seen from behind. At the left are shown lateral views of the same specimens on a smaller scale of enlargement Ajs from the same model shown in Fig. 45; B and C are drawings made from dissected specimens.

DEVELOPMENT OF THE NERVOUS SYSTEM. 69 surface toward the ventricle. Anteriorly they converge toward the median line, being united by a narrow seam just behind the exit of the trochlear nerve. Attached along their free edge is the tela chorioidea forming the roof of the fourth ventricle.

During the second month the cerebellar plates, owing to the active proliferation of cells in their mantle zone, rapidly thicken and bulge inward toward the ventricle. They also come to lie transversely so that what was originally a longitudinal dimension becomes a transverse one. This change in position is apparently due to the marked increase in the pontine flexure that occurs at this time. The cerebellar plates not only increase in thickness but also in length (i.e., transverse dimension), so that they become cramped in position and show a tendency to be thrown in folds. Further irregularity may be due to unequal growth in different portions. The growth is more marked in the cephalic half than in the caudal or lateral half. Near the median line on each side can be seen a swelling that corresponds to the vermis, which like the cerebellum itself originally consists of bilateral halves separated from each other by the roof plate.

During the third month (Fig. 50) the cerebellar mass comes to bulge outward, instead of inward toward the ventricle as before, which is evidently due to the fact that the proliferating mantle zone cells find less resistance in that direction, the marginal surface being more yielding than the ependymal surface. The cerebellum now consists of two convex masses (lateral lobes) connected laterally by a ridge (brachium pontis) with the developing pons. At the same time the fusion across the dorsal median line has commenced. The fusion begins on the dorsal surface and gradually involves the whole thickness of the wall, the last portion involved being the ependymal membrane and rhombic lip. Before the fusion is completed the outer surface of this region has commenced to show transverse fissures marking off the primitive lobes of the vermis.

We have already seen how the rhombic lip takes part in the development of the acoustic nucleus ; likewise in the cerebellum it plays an important part, giving origin to the nodulus and flocculus. At the third month it forms a distinct ledge, still notched in the median line, and along its free edge is attached the tela chorioidea. In conjunction with the acoustic lip it forms the lateral recess.

Between the third and fifth months the outer form of the cerebellum is completed by the formation of its principal lobes and fissures, the steps of which process are shown in Fig. 49. This lamellation is evidently due to the fact that the cortical region undergoes greater cell proliferation than the deeper portion, and since this growth is chiefly in the longitudinal axis it results in

70 fissures that run transversely to that axis. As the cortical differentiation and development of fissures occur together, we have in the latter an index of the former. It will thus be noticed that the vermis, and the anterior portion adjoining it are the first to show signs of this process. The floccular region begins to show fissures at about the same time. These may be regarded as the more primitive parts of the cerebellum, and are found to be the most constant in different vertebrates. The lateral lobes which form the so

Fiss prima

Mantle zone ^Stratum marqinale embryonale Tela chonoidea " v E.pendyma Ventr.quartus

Velum medullare ant

Stratum marginale embryonale / /Stratum molecular? ,5fratum qranulare ..-Fiss postnodularij Plx. chonoideus -Ventriculus quartu:

A 30 mm. Crl.

B 70 mm. Crl.

Fissura prima

Fissura prima

Stratum qranulare ^Stratum moleculare Pyramis Fiss. secunda '—-Uvula Fiss. postnodularis

Monticulus

L. centralis /

, Tuber ,.Pyramis Fiss.secunda -----Uvula "— --Nodulus Ventr. quartus

C 95 mm. Crl.

D 12 cm. Crl.

Fig. 50. — Sagittal sections through the cerebellum at or near the median line, showing the development of the fissures of the vermis and the formation of the cerebellar cortex.

called hemispheres of the adult organ are still smooth at the fourth month and it is not until the fifth month that they receive their fissures, which are partly intrinsic and partly extensions of fissures from the vermis. It is evident that phylogenetically the lateral lobes are recent structures. Later they increase enormously in size and eventually cover in the entire posterior portion of the vermis. It is supposed that their extensive growth in man is correlated with the large pons, and that in turn with the large pallium. Bolk, 1906, suggests that if we are to consider the cerebellum as the co-ordinator of muscle contraction then we may assign the muscles of the median line, which are common to all

DEVELOPMENT OF THE NERVOUS SYSTEM. 71 vertebrates, to the vermis, and the muscles of the extremities to the lateral lobes, the high development of the upper extremities in man explaining the marked development of his lateral lobes.

The lateral lobes are the lateral extensions of the median lobe. The median lobe is bounded anteriorly by the primary fissure and posteriorly by the secondary fissure. Everything in front of the primary fissure comprises the anterior lobe, and everything posterior to the secondary fissure comprises the posterior lobe (tonsil, uvula, nodule, and flocculus). The detailed development of the surface markings of the cerebellum can be seen by comparing Figs. 49 and 50. It should be noted that the great horizontal sulcus that is so prominent in the adult is relatively late in appearing; at the fifth month it only exists as a shallow furrow. The median longitudinal fissure, which is found on the posterior and inferior surface in the adult, is produced by the excessive growth of the lateral lobes whereby they close in over the vermis posteriorly. This rolling in of the lateral lobes toward the median line has already commenced at the fifth month (Fig. 49, C).

On tracing the fate of the rhombic lip in Fig. 49, where it is stippled darker than the rest of the cerebellum, it is seen that from the median portion there is developed the nodulus, which is the last portion of the vermis to show its bilateral character. From the lateral portion is developed the flocculus. The paraflocculus is derived from the lateral portion of the cerebellar plate immediately adjoining the rhombic lip.

The alar plate from which the cerebellum is formed, like the rest of the neural tube, at the outset has the typical ependymal, mantle and marginal zones, the ependymal zone toward the ventricle and the marginal zone toward the outer surface. It is through secondary development and migration that the cerebellar plate becomes covered with the layer of cells which eventually form its cortex, as will presently be seen.

At the end of the second month (30 mm.) the demarcation between ependymal and mantle zones is still poorly defined, and it is apparent that cells from the ependyma are still being contributed to the mantle zone. Later (70 mm.) the ependyma, with the exception of the portion situated at the rhombic lip, gradually enters upon its resting stage and assumes the form seen in adult specimens. The ependymal cells at the rhombic lip differ from the rest of the ependyma in that they continue to show active proliferation late in embryonic life. The same feature is shown throughout the whole rhombic lip but is more marked in the portion belonging to the cerebellum.

The most characteristic feature of the cerebellum is its cortex. We have already seen that originally the outer surface of the cerebellar plate is formed by the marginal zone and is devoid

72 HUMAN EMBRYOLOGY.

of nuclei. The neuroblasts which are found there later, constituting the cerebellar cortex, reach the surface by a process of migration. The steps by which the non-nuclear marginal zone becomes converted into a ganglionic layer by the invasion of these neuroblasts are shown in Fig. 50. This figure represents sagittal sections at or near the median line in four successive stages. At 30 mm. it can be seen that a layer of closely packed cells (stratum marginale embryonale) from the rhombic lip is spreading over the surface of the cerebellar plate, covering half of it in, the remaining portion being still non-nuclear. At 70 mm. this invasion of surface cells has extended so as to cover in the whole cerebellar plate, excepting the thinned-out portion that is to become the anterior medullary velum. At the same time a second layer of cells (stratum granulare) may be seen spreading from the rhombic lip forward and lateralward in the same direction as the marginal layer but beneath the surface in the outer part of the mantle zone. The space between it and the marginal layer corresponds to the stratum moleculare. In fetuses 95 mm. long these layers form well-marked strata running parallel with the surface of the cerebellum, dipping down where there are fissures. In fetuses 12 cm. long the surface of the cerebellum is greatly increased in extent, in the first place by actual growth of the whole cerebellum and in the second place by the infolding of the surface due to the rapidly increasing number of fissures. The cortex through further proliferation presents thicker and more sharply defined strata. We now (12 cm.) have an arrangement possessing a close similarity to the adult, viz., a central portion, fibrous and sparsely nucleated, covered in by a convoluted cortex consisting of three distinct strata. Concerning the source of the cells forming the cortex there remains some doubt. It is apparent that the cortex formation begins at the rhombic lip and spreads from there forward and lateralward; that is, it spreads from the free edge of the original alar plate toward the junction of the latter with the basal plate. This applies to all three of the primitive cortical layers. It is also evident that the outermost layer (stratum marginale embryonale) is directly continuous with the ependymal zone of the rhombic lip. On this account it has been suggested by Schaper, 1894, that the rhombic lip constitutes a germinal bed from which cells are given off and that these cells migrate along the surface and so form the outer layer as indicated in Fig. 51. It is conceivable that the deepest layer (stratum granulare) is also derived in the same way, its cells migrating out from the region of the rhombic lip along the outer border of the mantle zone. There is, however, no proof that the outer cells of the mantle zone do not take part in the formation of any one or all three of the cortical layers.

DEVELOPMENT OF THE NERVOUS SYSTEM.

73

The later histogenesis, as seen in G-olgi specimens in other mammals, is shown in Fig. 52. The development is not completed

£-•

• 9* f

®%

•"

— . ^ LX A. ii. wn a

© v

Fia. f 51. — Schematic drawing showing the differentiation and migration of cerebellar neuroblasts in the teleost. Arrows indicate the migration of cells from the rhombic lip over the surface of the cerebellar plate. The different cells are indicated in the same way as shown in Fig. 3. (After Schaper.) until very late, sometime after birth. The development of the granule cells seems to occur through a process of unipolarization, like the T-formation described in the spinal ganglion cells, and

S.marqinale embryonale

S.rnoleculare

" S.qranulare

A

B

Fio. 52. — Sections showing the histogenesis' of the cerebellar cortex. (After Cajal.) A, schematic section showing the migration of granule cells from the surface to the stratum granulare, and different stages in their differentiation, o being the youngest and k the fully formed cell. B, section through the cerebellar cortex of new-born dog, showing two Purkinje cells with partially formed dendrites, axones, and terminal aborizations. , subsequent migration inward to the granular layer. The succes • sive steps in this process are shown in Fig. 52, A, of which a is the earliest undifferentiated stage and "h" the fully formed

74 HUMAN EMBRYOLOGY.

granule cell. The dendrites of the Purkinje cells do not form until the migration of granule cells is completed. The outer layer (stratum marginale embryonale) disappears in man several years after birth.

Many of the mantle zone cells take no part in the formation of the cortex. Some of them form the neuroglia framework through which the fibres of the central white substance pass. Others become neuroblasts which are grouped to form the internal nuclei of the cerebellum. In embryos toward the end of the third month (50-95 mm.) the nucleus dentatus can be outlined in the interior of the lateral lobe, conforming to the outer form of the lobe. From its anterior border strands of axones emerge passing forward to be assembled in the two brachia conjunctiva, which can be traced forward to their decussation and connection with the red nuclei of the midbrain (see Fig. 53).

Later, as the nucleus dentatus becomes more sharply outlined, it assumes the convoluted form seen in the adult. Median to the dentate nucleus, in the vermis, the mantle zone neuroblasts are grouped to form the paired tegmental nuclei to which acoustic fibres can be traced by the end of the third month. Other nuclei formed supplementary to the dentate nucleus are the emboliform and globular nuclei. Of the centripetal tracts to the cerebellum the restiform body and the centripetal fibres from the acoustic and trigeminal nerves are the first to become well marked (second month). By the end of the third month the middle cerebellar peduncles containing fibres from the pontine nuclei may be distinctly traced to the lateral lobes of the cerebellum. The final development and completion of the arborizations of these fibres is not finished until sometime after birth.

(e) Development of the Midbrain.

As has been previously mentioned the midbrain is a portion of the epichordal brain and is closely affiliated in its manner of development and general form with the rhombencephalon and spinal cord. As in the latter two we can recognize on each side a basal plate and alar plate, the alar plates being large and united above by a narrow seam (roof plate). The wall forming them at first consists of a combined ependymal and mantle layer covered in by a non-nucleated marginal layer. Later the mantle layer becomes clearly differentiated from the ependymal layer, forming thereby three distinct strata (ependymal, mantle, and marginal). The differentiation between ependymal and mantle layers is completed in the basal plates about the end of the first month (Fig. 31, D). In the alar plates it is considerably later, about the third month. In their later development the alar plates form

DEVELOPMENT OF THE NERVOUS SYSTEM.

75

suprasegrnental ganglion masses (corpora quadrigemina) in a manner analogous to the development of the alar plates of the hindbrain that form the cerebellum.

The basal plates conform even more than the alar plates to the form seen in the rest of the epichordal system. The points of difference are mainly dependent on extrinsic factors. As in the hindbrain they give origin to motor nerves which become covered in by a modified formatio reticularis and the marginal zone is traversed by long suprasegrnental fibre tracts (peduncles of cerebrum). In Fig. 53 is shown a reconstruction of the midbrain of an embryo about three months old, in which the more prominent structures of the basal plate are indicated. The nuclei of the third

N. trochlearis

N. oculomotorius

Nucl. habenul le

Cerebellum

Nucl. dentatus

Fasciculus retroflexus

Fasc. long. med.

-I Nucl. ruber

Nucl. interpeduncularis Brachium conjunctivum (decussatio)

Fiq. 53. — Reconstruction of the midbrain of a human embryo 80 mm. long (Mall collection, No. 172) showing the relations of the red nucleus and the decussation of the brachium conjunctivum connecting it with the dentate nucleus of the cerebellum.

and fourth cranial nerves maintain their position near the floor of the lumen like the hypoglossal and abducens nerves in the hindbrain. The trunk of the fourth nerve, however, bends around dorsally to decussate in the roof as we have already seen. It runs directly dorsalward at first, but later owing to the growth of the inferior colliculus it is crowded backward and assumes a distinctly caudalward course before its decussation.

Associated with the third and fourth nerves is the median longitudinal fasciculus, which forms an intersegmental bundle belonging to the formatio reticularis, extending throughout the whole length of the mid- and hindbrains and into the anterolateral fasciculus of the spinal cord. It is apparently for the most part made up of axons which run a short course in the bundle and serve to connect the motor nuclei of the eye muscles (Nn. oculomotorius trochlearis and abducens^. It probably contains axons from

76 HUMAN EMBRYOLOGY.

other cranial nerve nuclei (Nn. facialis, acusticus, and hypoglossns). It also contains the fibres of the rubrospinal tract and fibres from the superior colliculus. Next to the entering fibres of the sensory nerves this is one of the earliest tracts in the epichordal system to become well outlined.

The origin of the nucleus ruber is not definitely known. It forms in the mantle layer as a portion of the formatio reticularis and by the end of the third month it is sharply outlined, and the decussating brachia conjunctiva connecting it with the dentate nucleus can be clearly recognized, as seen in Fig. 53. From its resemblance to the inferior olive we may assume that it develops in a similar manner. The fibres of the fasciculus retroflexus (Meynerti) traverse it on their way to the interpeduncular nucleus.

The mantle zone structures of the midbrain region become closed in ventrally and laterally by the tracts of the marginal zone. The first of these are the median lemniscus and the lateral lemniscus, which are usually included with the formatio reticularis as comprising the tegmentum. Later there are added ventrally the fibres connecting the cerebral cortex with the pons, medulla oblongata and spinal cord, which can be recognized toward the end of the third month. The subsequent increase in size of these ventral tracts produces the projecting masses known as the peduncles of the cerebrum.

The alar plates by their large size indicate that they are to form a large organ, though in man owing to the recession of the optic lobes the superior colliculus never attains the size found in lower vertebrates. The alar plates are at first separated by a narrow seam or furrow, and in young specimens, if there is any maceration, this seam (roof plate) is easily stretched and the alar plates may then overlap one another. With the subsequent thickening of the alar plates this median furrow disappears. This thickening also causes the lumen to decrease in proportionate size. Instead of a considerable cavity or midbrain ventricle we eventually have the narrow aqueduct of Sylvius connecting the third and fourth ventricles.

The details in the differentiation of the alar plates have not been fully studied in man, but we know that like the cerebellar plates they are characterized by the migration of neuroblasts to their outer surface. These neuroblasts proliferate and develop into more or less stratified ganglionic masses which, together with the deeper lying cells, form the superior and inferior colliculi. The more superficial layers correspond to the cortex of the cerebellum, and the deeper cell masses correspond to the dentate and tegmental nuclei. The fibres from the optic tract and lateral lemniscus can be plainly traced to the colliculi by the end of the third month.

DEVELOPMENT OF THE NERVOUS SYSTEM. 77 The optic tract fibres disappear beneath the superficial ganglion layer of the superior colliculus, while the lateral lemniscus spreads over the surface of the inferior colliculus. At about the same time the inferior brachium connecting the inferior colliculus with the median geniculate body can be recognized as shown in Fig. 45.

(f) Development of the Diencephalon.

The division of the prosencephalon into the telencephalon and diencephalon has already been referred to. The telencephalon from the outset differs widely from the type seen in the epichordal portion of the nervous system, and at a casual glance would seem to have nothing in common with it. The diencephalon, however, forms an intermediate link, and though it merges directly into the telencephalon, yet it resembles the epichordal system in many ways, particularly in the early stages.

As seen in Fig. 54, we can speak of an alar plate and a floor plate, united dorsally by a roof plate and ventrally by a floor

Ma.

Fig. 54. — Section through the diencephalon of a five weeks human embryo. Dp., roof plate (chorioid plexus); Th., alar plate (thalamus); S.M., sulcus limitans (sulcus hypothalamicus): P. s., basal plate (hypothalamus); Ma., mammillary recess. (After His.) plate, differing from the spinal cord up to the fifth week only in the absence of a ganglion crest and motor nerves. The alar plate like the alar plate of the epichordal system is predominantly sensory. It rapidly thickens, due to the proliferation of the neuroblasts which are to form the receptive nuclei for the optic and cochlear tracts and for the fibres of the medial lemniscus. These nuclei are massed together to form the thalamus which constitutes the largest part of the diencephalon. The lateral nuclei (geniculate bodies) are spoken of as the metathalamus.

The thalamus is separated from the hypothalamus by the sulcus hypothalamicus which extends forward to the optic recess. This sulcus apparently is analogous to the sulcus limitans. It persists into adult life. The hypothalamus or basal plate portion

78 lacks the motor elements which form so large a part of the basal plate in the epichordal system. Its principal function seems to be in connection with the special structures which are developed in the floor plate, the hypophysis and mammillary bodies (see Figs. 55 and 56). It is also from this portion that the corpus Luysi is developed, and through the caudal part of its marginal zone the large tracts pass from the internaUcapsule constituting the pedunculi cerebri.

The anterior boundary of the floor of the diencephalon may be regarded as constituted by the transverse ridge formed in the floor by the optic chiasm (Johnston). This ridge extends later

Sulcus hypothalamics [Monroi] Epithalamus (Corpus pineale) { { Metathalamus (Corpora geniculata)

Thalamus

Fissura chorioidea

Pallium

ora quadrigemina

Pedunculus cerebri

Rhinencephalon / Corpus striatum Hypothalamus Chiasma opticum' Hypophysis'

Pons [Varolii]

Cerebellum

MyeleflC^pHja^"^ Fossa rhomboidea

Fig. 55. — Brain of a human embryo 5 weeks old (13.6 mm.), median view[of the right half.

model, from Spalteholz.)

(After a His

ally into the basal plate as the pars optica hypothalamica. Caudal to it (Fig. 56) is a pouch representing the beginning of the infundibulum. During the fourth week an extension of this pouch (infundibular process) comes into contact with a similar pouch formed from the stomodseal epithelium (Rathke's hypophyseal pouch). The latter finally becomes detached from the oral epithelium and becomes incorporated with the infundibular process to form the hypophysis. The nervous and epithelial elements remain distinct throughout and constitute its two lobes. The epithelial pouch is at first flat and lies in front. It later develops two horns which envelop the infundibular pouch laterally. During the latter half of the second month vascular epithelial sprouts are developed from the pouch forming a mass of tortuous tubules, and finally

DEVELOPMENT OF THE NERVOUS SYSTEM.

79

(third month) obliterate the original cavity, converting it into a solid glandular organ. In the meantime the lumen of the nervous infundibular process has become shut off from the rest of the infundibular cavity, though the process always remains attached to the infundibulum. It becomes converted into a solid mass of tissue resembling neuroglia, and is closely united with the epithelial portion by a connective-tissue capsule and forms the posterior lobe of the organ.

The diencephalic floor caudal to the infundibulum forms the tuber cinereum, and still further caudal the mammillary recess from the walls of which the mammillary bodies are formed.

Pallium

Corpus striatum

EpithaJamus . 'Corpus pineale) '.'Metathalamus k Corpora geniculata

/Corpora quadrigemina

— Pe dunculus cerebri

Rhinencephalon ' Pars optica hypothalami Chiasma opticum Hypophysis Pars mamillaris hypothalami Pons [Varoli

Cerebellum — Fossa rhomboidea — Medulla oblongata

Fiq. 56. — Brain of a human'fetus in the third month, Jmedian view of right half. (After a His model, from Spalteholz.)

The roof plate of the diencephalon is bounded anteriorly and posteriorly by two transverse grooves. The posterior one appears toward the end of the first month and always remains a wellmarked groove. The posterior commissure crosses through its substance. The anterior boundary is formed by the velum transversum in a line with the foramen of Monro. The identity of this groove is best recognized in the early stages, 5-10 mm., after which the complications of the forebrain development obscure it. It marks the boundary between a part of the third ventricle that belongs to the telencephalon and a part that belongs to the diencephalon (see Fig. 26).

80 A dorsal view of the roof plate is shown in Fig. 57 and a median view in Fig. 56. It consists of a thin ependymal plate uniting the two thalamic plates. At the fourth week it is smooth. Proliferation of its cells causes it to expand and form an outward ridge which is soon thrown into longitudinal folds, as shown in Fig. 54. These folds project into the ventricle as the ectodermal lining of the tela chorioidea of the third ventricle. The increase of these folds and the development of a vascular mesodermal coat

Lamina terminals'

Fissura prima

Plexus chorioideus

Hippocampus

Thalamus

Epiphysis

Corpus striatum

Roof plate

Colliculus superior

Fiq. 57. — Dorsal view of a model of the prosencephalon of a human embryo at the beginning of the sixth week (13.6 mm.). The pallium is partly removed, exposing the interior of L the lateral ventricles. This is from the same specimen shown in Fig. 55. (After His.) complete the formation of a typical chorioid plexus. Orally this chorioid roof is continued into the telencephalon where it forms a pointed pouch overlapping the lamina terminalis and the contained commissures. At the foramen of Monro it is continuous with the similarly formed chorioid body of the lateral ventricle. At the posterior, or epiphyseal end of the roof, there is another small chorioidal pouch formed which overlaps the epiphysis. The anterior chorioidal pouch apparently corresponds to the paraphysis of the lower vertebrates.

Laterally the roof plate is attached along the borders of the thalamic plates. At the line of junction there are formed the

DEVELOPMENT OF THE NERVOUS SYSTEM.

81

epithalamic structures known as the ganglia habenulse and the epiphysis. The habenular apparatus bears a relation to the thalamic plate similar to that of the rhombic lip to the alar plate in the hindbrain. The habenular nuclear mass can be recognized by the fifth or sixth week as a longitudinal ridge on the dorsal surface of the diencephalon along the edge of the thalamic plate. In fetuses 80 mm. long, as seen in Fig. 53, the nucleus can be outlined and the stria medullaris and fasciculus retroflexus traced to their terminal connections. The former extends forward along the edge of the thalamic plate and spreads out over the surface of the future anterior nucleus of the thalamus.

Nucleus caudatus

Fissura chorioidea

Hippocampus

Mesencephalon

Cerebellum

Capsula interna

Insula Bulbus olfactorius

Fia. 58. — Lateral view of brain of a human embryo about three months old (crown-rump length 53 mm.). Part of the hemisphere wall is removed, showing its thickness and exposing the [interior of the lateral ventricle. (After His.) The epiphysis is formed at the caudal end of the diencephalic roof. At the fifth week, Fig. 57, it appears as a rounded elevation of the roof. In the groove behind it the posterior commissure crosses and in the groove in front of it the dorsal or habenular commissure. It thus originally consists of a thin ependymal diverticulum between these two commissures. Subsequently its* walls are thickened and incorporate some of the adjacent vascular mesoderm to form the adult organ. In the human embryo the epiphysis never reaches the advanced stage of development seen in reptiles (pineal eye).

The ventricle of the diencephalon, at first a relatively broad space, becomes thinned down to a narrow cleft owing to the thiek Vol. II.— 6

82 HUMAN EMBRYOLOGY.

ening and crowding in of the lateral walls. The space is still further reduced by an actual approximation and fusion of a portion of the thalamic plates. In this manner there is produced the commissura mollis, the extent varying in different subjects. It is very large in lower mammals.

We thus see that the diencephalon consists of three main regions, the hypothalamus, the epithalamus, and, largest of all, the thalamus proper (including the metathalamus or geniculate bodies). The hypothalamus and epithalamus are the most primitive in character and their fibres are the first to develop. During the second month the following tracts become established in the hypothalamus: (a) fasciculus mammillotegmentalis ; (b) fasciculus thalamomammillaris ; and (c) columna fornicis. At the same time in connection with the epithalamus there are developed: (a) stria medullaris; (b) commissura habenularis; (c) fasciculus retroflexus; and (d) commissura posterior. Advanced development of the thalamus is characteristic of the higher vertebrates. In the human embryo, though it develops somewhat slower, yet it eventually predominates over all the rest of the diencephalon. By the end of the second month the acoustic fibres have reached the median geniculate bodies, the optic tract fibres the lateral geniculate bodies, and the fibres from the median lemniscus the ventrolateral thalamic nucleus. At the same time these nuclei give off fibres that extend into the corpus striatum forming the thalamic radiation. Some of them can be seen passing through to reach the developing neopallium (compare Figs. 63 and 73).

(g) Development of the Telencephalon.

When we come to the extreme oral end of the neural tube it is no longer possible to clearly recognize an alar plate or basal plate as seen everywhere else in the tube. The sulcus limitans, however, is usually considered as curving downward along the posterior border of the optic evagination to the median line just in front of or along the transverse ridge caused by the optic chiasm. (Compare Figs. 26 and 28.) In this sense practically the whole telencephalon may be regarded as corresponding to an elaboration of the alar plate.

The broad expanse of this portion of the neural plate in the earliest stages (Fig. 23) indicates its importance. As we have already seen, before the closure of the tube is completed (Figs. 26 and 28), the telencephalon has become differentiated on each side into a bulging portion marking the future pallium or hemisphere and a basal portion which is to form laterally the corpus striatum and medially the rhinencephalon. The optic evagination is on the boundary line between the latter and the diencephalon.

DEVELOPMENT OF THE NERVOUS SYSTEM. 83 The further steps in the differentiation are shown in Figs. 55, 56, 59, and 60. Comparison of these figures shows that though these primary regions of the telencephalon undergo great change in size, form and position, yet they maintain their identity throughout. It is the extensive development of the pallium that is the most striking feature ; whereas the rhinencephalon, which is so massive in lower vertebrates, at its most favorable embryonic stage in the human embryo composes not more than one-twentieth part of the bulk of the telencephalon, and in the adult a far smaller proportion. The corpus striatum is closely united with the thalamus, and it is through this that the connection between telencephalon and diencephalon is principally maintained. It shares with the

Mesencephalon Diencephalon \

Pallium

/ Recessus mammillaris

\ Lobus olfactorius

Saccus hypophyseos Pedunculus opticus

Fig. 59. — Lateral view of the forebrain of a human embryo 10.2 mm. long. Taken from a model. It shows the beginning of the overlapping of the diencephalon by the pallium. Compare with Figs. 29 and 86. After His.) thalamus in the development of the pallium. Through these two centres pass all the paths of communication to and from the cortex, excepting the insignificant portion belonging to the olfactory system.

Corpus Striatum and Pallium. — The change occurring in the telencephalon toward the end of the first month is a very important one and should be carefully noted in order to understand the development of this region of the brain. As can be seen by comparing Figs. 28 and 55, it represents the transition from the neural tube type to the typical paired hemispheres opening out laterally through the foramen of Monro. The latter is produced not as an actual constriction but secondarily through the fact that its boundaries remain nearly stationary while the pallial walls undergo enormous expansion. The expansion of the pallial walls is shown in Figs. 59 and 60. From around the borders of the corpus striatum they expand orally, dorsally and caudally, gradually covering in the whole diencephalon and more caudal parts of the

84 HUMAN EMBRYOLOGY.

brain. In Fig. 60 we can recognize the oral end as the frontal lobe, its inclosed cavity being the anterior horn of the lateral ventricle. The caudal end curves downward to form the temporal lobe, its inclosed cavity being the descending horn of the lateral ventricle. Later, fetuses 10 cm. long, the caudal portion presents two lobes or poles, the temporal lobe having become more ventral and the new occipital lobe forming the extreme caudal end, its cavity corresponding to the posterior horn of the lateral ventricle (see Figs. 76 and 77).

During the expansion of the pallial walls the median lamina uniting them does not share in the growth, and there is thus formed Pallium

Lobus temporalis Lobus parietalis Lobus frontalis

Corpus mammillare Tuber cinereum X ' ^"^ Pars ant - Uobus Pars post. jolfactonus / i \ Infundibulum Saccus Pedunculus opticus hypophyseos Fig. 60. — Lateral]view of the forebrain of a human embryo 13.6 mm. long. Taken from the same specimen shown in Fig. 57. Comparison with Fig. 59 shows the growth of the pallium and the manner in which it_overlaps the diencephalon, (After Hi's.) the great longitudinal fissure between the two hemispheres, which eventually becomes occupied by a mesodermal septum, the falx cerebri. The lamina uniting the two hemispheres is continuous anteriorly with the lamina terminalis. and might properly be regarded as belonging to it. Posteriorly it is continuous with the diencephalic roof plate. We have already seen how the latter becomes folded and vascularized to form a chorioidal roof for the third ventricle. A similar change occurs in the pallial wall near its junction with the diencephalon. The wall becomes very thin and folds into the lumen of the lateral ventricle, carrying with it vascular mesoderm (Fig. 62) and thus finally forms a chorioidal body within the ventricle. If the chorioidal body is removed there is left a cleft in the wall, the approximate position of which is shown in Fig. 64, and which corresponds to the fissura chorioidea. That portion of the hemisphere wall ventral to the fissure apparently never undergoes active development. In the portion

DEVELOPMENT OF THE NERVOUS SYSTEM. 85 dorsal to it is developed the hippocampal system. Anteriorly the chorioidal formation is continuous with the chorioidal roof of the third ventricle. The whole chorioidal mass forms an irregular Y, the stem being the roof of the third ventricle and the two arms being the chorioidal bodies of the lateral ventricles. The arms begin at the foramen of Monro and differ from the stem in being better developed and projecting into the ventricle. The relations of these structures are considerably modified later by the changes occurring in the lamina terminalis due to the formation of the interforebrain commissures which will be spoken of again in connection with the hippocampus. In all other regions of the neural tube we find chorioidal formation limited to the roof plate and this gives a ground for considering the roof plate of the telencephalon as bifurcated and represented by the two chorioidal fissures. There is, however, no other evidence of such a bifurcation of the oral end of the tube.

Attention has already been called to the ridge formed by the corpus striatum in the floor of the telencephalon (Figs. 28, 57, and 55). It can be seen at the outset that the corpus striatum is directly continuous with the thalamic plate of the diencephalon. In its development it resembles the thalamus and becomes closely co-ordinated with it, but the two always remain distinctly separated from each other, at first by a deep groove and later by the taenia semicircularis. It consists at first of a ridge which spreads out anteriorly in three limbs (Fig. 28), marking off the two divisions of the rhinencephalon. Later, with the expansion of the pallium, the ridge becomes more prominent. It is elongated caudally and curves around the developing stalk of the hemisphere to the tip of the inferior horn, forming the tail of the caudate nucleus. As the wall thickens it projects into the ventricle, and the lateral surface of the same portion of the brain wall, Fig. 60, presents a shallow fossa which continues to become deeper as the surrounding pallium develops. The thickening of the wall at first involves chiefly the ependymal zone, which undergoes an exuberant growth, and exactly in the area corresponding to the future caudate nucleus. Gradually from the ependymal zone a mantle zone is elaborated and furnishes the neuroblasts which become assembled into a typical corpus striatum. The fibre strands from and to the thalamus become arranged in a sharply marked lamina which subdivides the corpus striatum into the caudate and lenticular portions, thus forming the limbs of the internal capsule. It should be noted that this subdivision of the corpus striatum and the formation of an internal capsule are due to the manner in which the fibres traverse it. In some mammals the fibres pass diffusely through the striatum and then the capsule-like arrangement of the fibres is absent.

86 The division between thalamus and corpus striatum is most evident in midembryonic life. At the end of the third month a deep groove separates them. Subsequently as they become larger and as the nerve-fibres connecting them increase, this groove be

Fiq. 61. — Schematic horizontal sections through the forebrain of human embryos, showing three stages in the fusion of the thalamus and corpus striatum. A, embryo of about 6 weeks (15 mm.); B, fetus during the fourth month; C, fetus during fifth month (crown-rump length 150 mm.). F. M., foramen of Monro. (After Goldstein.) comes flattened out, and they come to form one solid ganglionic mass separated from each other only by the taenia semicircularis. It is thought by some that in this process an apposition and fusion occurs between the anterior end of the thalamus, the medial pallia!

Fissura chorioidea

Corpus striatum

Ventric. Ill

Ventric. lateralis

Foramen Monroi

Fusion of the thalamus and corpus striatum

Infundibulum

Fiq. 62. — Transverse section through forebrain of human embryo of about six weeks (16 mm. long), showing on one side the fusion of thalamus and corpus striatum. Compare with Fig. 61, A. (After His.) wall and the corpus striatum, as shown in Fig. 61. The same result, however, would occur if it were a simple thickening of the wall produced by the massive connections developed between thalamus and corpus striatum. In Fig. 62, on comparing the two sides it would seem as though such a fusion had occurred; but it

DEVELOPMENT OF THE NERVOUS SYSTEM.

87

should be remembered that the two sides are not cut at the same level, that one is through the foramen of Monro and the other just below or caudal to it.

At the end of the fifth month, as seen in Fig. 63, the relation and form of the corpus striatum are practically those of the adult. The form of the internal capsule can be clearly made out. It contains: (a) fibres connecting the thalamus with the corpus striatum and pallium; (b) optic and acoustic fibres from the metathalamus to the pallium; (c) projection fibres from the pallium (pyramidal

Cornu anterius —

Corpus pratermi nale et columna fornicis Foramen of Monro

Ventric. Ill

Glomus chorioideum

Cornu posterius

Nucleus caudatus Claustrum Nucleus lentiformis Capsula interna

— Falx cerebri

FlQ. 63. — Horizontal section through the forebrain of a fetus about five months old (crown-rump length 160 mm.). This represents a stage intermediate between B and C in Fig. 61. (After His.) tract). The stria semicircularis, though it is on the border line between diencephalon and telencephalon, probably belongs to the latter. It can be recognized early (80 mm.), as it curves around the thalamic border. It is shown in Figs. 66, 67, and 68.

Rhinencephalon. — The olfactory apparatus consists of a basal portion and a cortical or pallial portion. The basal portion includes the olfactory bulb, olfactory stalk, the median and lateral olfactory tracts and the region of the anterior perforated space which merges on the one hand with the tip of the temporal lobe and on the other with the preterminal body which partly forms the septum pellucidum.

88 HUMAN EMBRYOLOGY.

All of these parts are derived from the basal portion of the telencephalon median to the corpus striatum (see Fig. 55). The cortical portion of the rhinencephalon belongs to the pallium and is designated as archipallium in contradistinction to the remainder or neopallium. It consists of an extension of the basal preterminal area dorsalward and forms the median margin of the pallium along the dorsal border of the chorioidal fissure. In the adult we know it as the hippocampus, the dentate fascia, and it probably includes a strip of cortex bordering along the hippocampus. It differs from the basal portion in that the cells composing it are arranged in the form of a cortex with characteristic strata.

Fig. 64. — Median wall of the telencephalon of same specimen shown in Fig. 58, showing the fissura chorioidea and under it a narrow strip of the lamina infrachorioidea. The fissura prima extends upward from the olfactory lobe marking off the anterior border of the prseterminal body. In front of it is the accessory' arcuate fissure. The calcarine fissure is sharply marked. The hippocampal fissure curves around parallel with the inner margin of the specimen. (After His.) In Figs. 66, 67, and 68 the median view of the brain is shown in its later stages of development exposing this general region. That portion of the pallium possessing a uniform cortical layer is shown in lighter color than the remainder and corresponds in general to the neopallium. The darker portions form the archipallium and thus represent the olfactory apparatus. If we are to consider that portion of the cortex adjoining the hippocampus as olfactory and belonging to the archipallium (G. Elliot Smith), then to complete the pallial portion of the olfactory system the dark shading should be spread wider to include this. The basal portions of the olfactory apparatus are shown in the same figures. Lateral views of the same stages are shown in Figs. 76 and 77. The relations of the basal portions can be best seen, however, in a ventral view, as given in Fig. 65, where its different parts are indicated. The boundaries of the olfactory apparatus in this figure are marked by the dorsal border of the lateral olfactory gyrus and the gyrus ambiens.

DEVELOPMENT OF THE NERVOUS SYSTEM.

89

The olfactory apparatus can be traced back to a still youn._ stage in Figs. 55 and 60, which present median and lateral views of the brain at the end of the fifth week. At this period, as seen in a lateral view, there is a distinct field marked off ventral to the Sylvian depression which represents the basal portion. It consists of two elevations. The anterior one is formed by a shallow pocket opening out of the ventricular cavity. The further evagination of this results in the formation of the hollow olfactory bulb whose form in later stages we have already seen. It can be readily understood how this extending tubular process becomes narrowed down in its proximal portion to form the olfactory stalk, commonly spoken of as the first pair of cranial nerves. In man the lumen of this process is eventually obliterated.

Gyrus olfactor. medialis Gyrus olfactor. medius Gyrus diagonalia

Cerebellum

Insula Gyrus olfactor. lat.

Gyrus ambiens Gyrus semilu:.

Oliva

Fig. 65. — Ventral view of brain of a four months old human fetus, showing the olfactory apparatus. v After Kollmann.) The posterior olfactory elevation becomes a thickened portion of the wall instead of an evagination and represents the anterior perforated space. Mesially (Fig. 55) it extends upward, bordering along the lamina terminalis, and forms the preterminal body of Elliot Smith; and is directly continuous with the olfactory pallium along the fissura chorioidea. In the older stages the olfactory tracts were already laid down. At this stage they are just beginning to appear. According to His. a little before this time, in embryos Nl. 10.9 mm., filaments can be recognized connecting the nasal epithelium with the olfactory pocket i anterior olfactory elevation) of the brain wall, i.e., before we can yet speak of an olfactory bulb or stalk. About the same time the fibres of the median olfactory tract begin to extend dorsalward from the olfactory pocket to the preterminal body. By the time the olfactory bulb has become partially constricted from the general brain wall

90 HUMAN EMBRYOLOGY.

fibres can be seen extending- backward from it along the dorsolateral border of the posterior olfactory elevation (anterior perforated space) and constituting the lateral olfactory tract, whose fibres are distributed to the archipallium covering the apex of the temporal lobe.

In embryos under four weeks old the rhinencephalon can only be recognized as the space between the lamina terminalis and the internal ridge formed by the corpus striatum (Fig. 28). According to His the three limbs of the striate ridge mark out the two fossae which we have seen as prominences on the lateral surface corresponding to the olfactory bulb and the anterior perforated space.

Inteefobebeain Commissuees.- — In connection with the olfactory apparatus there are the tracts connecting it with the hypothalamus (fornix) and the commissural tracts uniting the opposite sides of the telencephalon (anterior commissure, commissure of fornix and corpus callosum). The development of these structures can best be understood by comparison of Figs. 66, 61, and 68, which represent three stages in their development.

In these figures the lamina terminalis is shown as though cut in the median sagittal plane, while the commissures are left longer so that their cut ends project from the surface. It will thus be seen that they all cross through the substance of the thickened lamina terminalis and are thus confined to the original walls connecting the two hemispheres. The use of the term " lamina terminalis " is made in a broad sense. It is not restricted to the ependymal seam that originally closes off the anterior end of the tube, or its immediate derivatives ; but includes also a certain amount of neuroglial tissue from the adjacent wall which becomes incorporated with it. In this sense we may speak of a fusion of the median walls of the precommissural bodies, but the process only occurs in a narrow line immediately in front of the lamina terminalis and the derived tissue becomes a definite part of the latter. Its further enlargement is produced by the stretching of its boundaries by the entering commissural fibres. A narrow lamina terminalis suffices originally, as only a few slender bundles cross at first. As further fibres are added the fibre mass spreads open a space for itself, in which process a portion of the precommissural body is appropriated, and the eventual lamina terminalis presents a large surface in the cut median section, including the whole corpus callosum and the septum pellucidum. The widening of the boundaries of the lamina terminalis occurs rapidly in fetuses between 80 and 150 mm. It is distended dorsalward and anterolateralward through the growth of the corpus callosum, the shape of which in turn is determined by the expanding pallium. As a result of this tension there is a new arrangement of its tissue, and

DEVELOPMENT OF THE NERVOUS SYSTEM. 91 in the readjustment a ventricle is formed, the so-called fifth ventricle or cavum septi pellucidi. This becomes lined with a smooth neuroglial membrane. The ventricle is only present where there is a large corpus callosum.

It should be mentioned that, according to some authors (Goldstein, 1903), the commissures are developed entirely within the lamina terminalis, in the narrower sense, the adjacent walls not contributing anything to it. On the other hand, according to Zuckerkandl, 1901, there is an approximation and fusion of a considerable area of the median walls with resorption of the previously interposed mesodermal falx. This forms a "massa commissuralis" through which the fibres subsequently cross. In the process of fusion he describes an active proliferation of the cells of the wall, forming "wulstartige Vorspriinge, ' ' which meet and fuse in the median line. The description given above corresponds essentially with that given by G. Elliot Smith (1895) and Marchand (1909).

Of the three commissures the anterior commissure and the commissure of the fornix are the more primitive, and they both serve as commissures for the archipallium. The fibres of the fornix make their appearance early along the chorioidal margin of the hippocampus and form a bundle increasing in size as it extends forward. It passes over the foramen of Monro to reach the precommissural body where it gives off and receives fibres. It then extends ventralward to the hypothalamus in the region of the mammillary body (pillar of the fornix). In the region of the precommissural body the two fornix systems exchange fibres forming a commissure between the two hippocampal gyri. Originally this commissure lies directly dorsal to the anterior commissure, as seen in Fig. 66. Subsequently it is drawn backward' owing to the change in the position of the hippocampi, which in turn are carried backward by the ventral extension of the temporal lobe.

The development of the corpus callosum is closely connected with that of the fornix commissure, being practically a derivative of it. The latter is a commissure of the archipallium and the former of the neopallium, the one starting where the other stops. Owing to the great development of the neopallium in man the corpus callosum soon predominates.

In fetuses 80 mm. long (Fig. 66) its fibres can be recognized in the medial wall of the hemisphere streaming toward the upper part of the lamina terminalis, where it crosses together with the commissure of the fornix, forming a rounded bundle on the dorsal surface of the latter. The first discoverable fibres contributed to it are found in the median wall in the vicinity of the point of crossing. As further fibres are added they form a layer that can be gradually traced spreading backward and lateral ward through

92 Taenia semicircularis

Corpus callosum

~"- Fissura chorioidea

Commi8sura hippocampi Commissura anterior Fia. 66. — Median view of a model of the telencephalon of a human fetus three months old (80 mm. long, Mall collection, No. 234a). It shows the cut ends of the commissural bundles crossing in the lamina terminalis. The thalamus is removed, exposing through the fissura chorioidea, the lateral ventricle and nucleus caudatus. Comparison of this figure with Figs. 67 and 68 shows the development of the corpus callosum and its relation to the commissure of the hippocampus.

Taenia semicircularis

Corpus callosum \

\

Cavum septi pellucidi

Commissura anterior

Fissura hippocampi

Columna fornicis

Fascia dentata \ Fissura chorioidea

Fig. 67. — Median view of a model of the telencephalon of a human fetus about 4 months old (95 mm. long, Mall collection, No. 146). It shows the same structures seen in Figs. 66 and 68. The lamina terminalis has become thickened at the expense of the corpus praeterminale, and a cavity has developed in it forming the cavum septi pellucidi. The fissura hippocampi has deepened, it being the first step in the covering in of the fascia dentata.

DEVELOPMENT OF THE NERVOUS SYSTEM.

93

the pallium to the regions more distant from the lamina terminalis. It is quite probable that the growth of these fibres starts simultaneously in all parts of the pallium, and it is natural that it is in the region of the lamina terminalis that the accumulation of them is first sufficient to be recognized as a definite layer. The form of the corpus callosum in fetuses 95 mm. long is shown in Fig. 70, where its relation to the brain wall can be seen. It lies nearer the ependyma than the outer surface. The model represents only

Ca^nim septi pellucidi Corpus callosum

Taenia semicircularis Commissura hippocampi

Commissura anterior

\ Fornix

r\ Fascia dentata \ i Radiatio thalami

Columna fornicis

Fig. 68. — Median view of a model of the telencephalon of a human fetus of the fifth month C150 mm. long, Mall collection). Here the commissural systems are practically those of the adult. The fascia dentata extends up over the splenium of the corpus callosum and reappears at the front end (genu) and is continuous with the preterminal body and enlarged lamina terminalis. Compare with Figs. 66 and 67

that part that could be distinguished as a stratum. It is probable that callosal fibres reach much further than this.

The addition of fibres occurs interstitially, the new fibres growing in everywhere between the old ones. As seen in cross section, it very early takes on a typical form. In fetuses 95 mm. long, Fig. 67, we can recognize the anterior end as the genu and the posterior end as the splenium, the latter always remaining in contact with the fornix commissure. With the further growth of the pallium and the addition of new fibres to the corpus callosum we have in 150 mm. fetuses, Fig. 68, relations which are practically those of the adult. It will be seen that like the pallium it has

94 HUMAN EMBRYOLOGY.

grown both posteriorly and anteriorly. As the splenial end spreads caudalward it covers in the diencephalon, carrying the commissure of the fornix with it. The latter gradually becomes flattened out on the under surface of the corpus callosum to form the psalterium.

The striae of Lancisi represent tissue connecting the dentate fascia with the precommissural body. By comparison of Figs. 66, 67, and 68 it can be seen how the two are closely connected at first (practically continuous). The tissue connecting them may be regarded as regressive fascia dentata. It soon is stretched out in a narrow strand by the enlarging corpus callosum. Some of the fibres in this substance are frequently laid down in advance of the corpus callosum, so that the fibres of the latter pass between them and incorporate them. Thus the anterior end of the adult corpus

Taenia semicircularis Commissura anterior W * Fig. 69. — Right half of the anterior commissure dissected out from the brain of anl80 mm. pig fetus, viewed from above and from the median side. Its constituent elements are shown andjits connection with the taenia semicircularis callosum is found to be traversed by some of these fibres, on their way through the wall of the septum pellucidum to the precommissural body.

The anterior commissure, as it appears when dissected out in the pig embryo, is shown in Fig. 69. It consists of two divisions, an anterior or olfactory division and a posterior division. The olfactory division arises principally (in pig embryos) in the brain wall in the neighborhood of the olfactory evagination. The posterior division arises between the corpus striatum and the overlying cortex and thus corresponds in position to the fossa of Sylvius. It forms a concave lamina in which the corpus striatum rests. Owing to the fibres streaming from the corpus striatum to the pallium it is difficult to determine whether its fibres are derived mesially from the corpus striatum or laterally from the superimposed cortex. In Fig. 69 it is shown how the fibres composing it form confluent fan-like bundles which point ventralward and become incorporated into two main bundles, anterior and

DEVELOPMENT OF THE NERVOUS SYSTEM.

95

posterior, which in turn unite and point toward the lamina terminals to meet the similar formation from the opposite side. In the earlier stages it is possible to recognize the anterior commissure fibres before they have reached the median line. As it approaches the median line it receives a communication from the taenia semicircularis. In Fig. 70 is shown the relation of the anterior commissure in a 95 mm. human fetus, being essentially like

Ventriculus lateralis

Capsula interna'

Corpus callosum

Commissura hippocampi

Commissural anterior

| Pars | post.

Pars ant.

Bulbus olfactorius Chiasma

Fig. 70. — Anteromedian view of a model of the fibre tracts of the same brain shown in Fig. 67. One half of the brain stem is preserved intact. In the front part of the telencephalon everything is removed excepting the fibre tracts, exposing the corpus callosum, pillar and commissure of the fornix, two divisions of the anterior commissure and the internal capsule. The last subdivides the corpus striatum into the caudate and lenticular nuclei. On the left side the connection from the taenia semicircularis to the anterior commissure is shown.

that seen in dissections of pig fetuses. Of the olfactory division some strands apparently come from the olfactory stalk, and others from the hemisphere wall in the immediate neighborhood.

DEVELOPMENT OF THE WALL OF THE HEMISPHERE.

Up to the end of the second month the wall of the hemisphere remains thin and relatively undifferentiated. The increase in thickness and development of the wall that begins to be noticeable at that time does not occur uniformly in all regions at the same time, but is always more advanced in the basal portions adjoining the corpus striatum, and from there it gradually extends toward the median line over the whole pallium. In the accompanying table is given the thickness of the wall and its constituent zones in

- X IX

' - - ~ ~ "' : i. -.:.-:. 111 taken at appro-ximateiv the same region.

Apprcaiinaw *js !-:• aai

y 1 :' r^z; 1

rttooD

' .6

l.i

.16 15

-" AY 1: r,^

I :

-.

.* r.C— Z. . .'.. f:.l.~ ' ; "' i ; . . \ " f *

Mi.t.'j: :;:: rim::

- - -•-.

T- -.

si - 1 tube,, of an outer irr t.:'z1~ iiifi:^

—

A :••:•: r ii: ::n: i

ll - f7

a ike the wall of the original cleated or marginal zone and .ml :'-r 7ii .ills ::rziiiz rftion of the latter are farther differentiated and are irr.izr-ri. "1 i :«:-:-;:; - :iiri :r 7iii:'.e :::e. trueture the wall eons: ~ : s : i i " i — j i - -:n~ : ::i_ .:: r-l.ii ;i :ii: us i described imder the h _ 77tSt7.:5 It :1y7.:it:ii: :ii: is i:: itmn cribrosam. This is situated in : f lateral processes from the so arranged as to form a thin nucleated and outer part of this zone.

ff£ 17. V v 111-: 111 --' ---i 21 ^ 7 rill 2 ^.

i:i:i l.iii :. iiriiiiiil .ilis ~li:i

Ill "lr -" : s :' : 77i r :

£

the t ?. 717 f '. Z "If 7 .: . .171 .

lie region of the eorpns

.1 : i f : ii ~~' * r f i : r : f :i f sarface as far as the olfactory margin, where it form the olfactory cortex. It is formed by the mi if~f - iiim if ii: zli.sii :': ri 71 f fif 177111*. 11 i 711.1:". -_ in mil cording to Tfis (1904),

rion of this : hi. iii„j " :i- 71 -11" modified to

7i:i:i Hi" fntnTr 1 - 7-amidaI cells advance toward the onter surface as far as :i nil ::::: : mi. ""ii - 11 pact layer. This migration is most actinre daring the third month '-'. 111 :-= — f_ iii: -..- 1:1 71 7 1 "_ 1 - — 1 - 7 ::i:i

DEVELOPMENT OF THE NERVOUS SYSTEM.

97

Stratum cribrosum

Lumen"

Zona .'ependymalis ^matrix)

Pars nuelearh

Mantle zone

Zona marginalis

Limitans meningea

Fig. 71. — Section through hemisphere wall of a human embryo at about the sixth week (16 mm. long) before the development of the cortical layer. (After His.)

Lumen

Matrix (ependy ma)

Intermediate zone (Zwischenschicht)

Wandering neuroblasts

\ Cortical layer

Marginal i-eil

Fig. 72. — Hemisphere wall of a human fetus at about the eleventh week (crown-rump length 46 mm.), showing the wandering of the neuroblasts from the interior of the wall toward the outer surface where they form the pyramidal or cortical layer. (After His.)

Vol. IT.

98 of the brain wall of an embryo about 3 months old. On the inner surface of the marginal zone is a well-developed cortical layer, toward which numerous other neuroblasts can be seen wandering from the mantle zone. The mantle zone has become much broader, and can be subdivided into an inner nuclear portion, consisting of proliferating neuroblasts and spongioblasts, and an outer wider portion, or intermediate layer (Zwischenschicht), which seems to

Matrix (Z. ependymalis)

Intermediate zone (mantle layer)

Cortical layer Stratum cribrosum Marginal veil

f.<

Fig. 73

Fig. 74

Fig. 73. — Schematic drawing showing structure of hemisphere wall at the end of the third month. On the left side is shown only the spongioblasts framework; on the right side are shown the wandering neuroblasts and their accumulation to form the cortical layer. The arrows indicate ectogenous fibres from the internal capsule. Compare with Fig. 72. (After His.) Fig. 74. — Pyramidal neuroblasts during the period of migration. The lower process is broader and more irregular. It becomes the apical process. The other is more slender; it is pointed toward the ependyma and becomes the axone. (After His.) be sparsely nucleated, owing to the extensive invasion of fibres from the internal capsule. A schematic section of the wall at this time is shown in Fig. 73. On the left is shown the spongioblastic framework and its formation into different layers, the nuclei representing future neuroglial cells. On the right are indicated the wandering neuroblasts moving radially outward to form the cortical layer. In their journey they have to make their way through the meshes of the framework and through the entering fibres from the thalamus and corpus striatum. During this period of migration the pyramidal cells usually possess a bipolar character, as shown in Fig. 74. The advancing end is broader and more irregu

DEVELOPMENT OF THE NERVOUS SYSTEM.

99

lar and becomes the apical process. The slender central end (i.e., toward the lumen) becomes the axone. The development of lateral dendrites and attainment of the characteristic adult form does not occur until about the time of birth.

Up to the fourth month the wall remains relatively thin and the ventricle large. From then on the wall rapidly thickens, owing mainly to the great increase in fibres in the intermediate layer. These fibres at first are all ectogenous fibres, from the thalamus and corpus striatum. Subsequently there are added the axones from

Cortical layer

Intermediate zone (mantle zone)

Matrix (ependyma)

Fia. 75. — Section through hemisphere wall at the end of the fourth month (crown-rump length 120 mm.). The intermediate (Zwischenschicht) zone is rapidly becoming thicker, owing to the increase of incoming and outgoing fibres. It is this layer that eventually forms the central white substance of the brain. Its inner portion is subdivided into the following layers: a, outer transitional layer; b, outer striated layer; c, inner transitional layer; d, inner striated layer. At this period the wandering of the cortical neuroblasts is completed. (After His.) the developing neuroblasts of the cortical layer (autochthonous). It is this fibrous layer that eventually forms the massive white substance of the hemisphere. With the increase in the thickness of the wall the ventricle apparently becomes smaller. The appearance is due in part to the difference in relative growth of the two, and in part to the change in shape of the ventricle, from a wide vesicle to a narrow slit.

As shown in Fig. 75, the inner portion of the mantle zone still possesses many nuclei, belonging to the supporting framework,

100 HUMAN EMBRYOLOGY.

which are now arranged in layers. The ependyma does not appear as active as heretofore, although it apparently is still giving off spongioblasts that are to form the neuroglial elements of the white substance. The cortical or pyramidal layer has taken up all its wandering neuroblasts from the deeper layers and is sharply marked off from the subjacent intermediate layer. It is already beginning to subdivide itself into two separate layers, the outer portion being somewhat denser than the inner portion. During the sixth and seventh months, with the further differentiation of the cortical cells, they become grouped into six distinct layers, corresponding to the stratification of the adult cortex. Certain portions of the cortex exhibit modifications of this six-layered arrangement, the strata being increased or decreased in number or varying in thickness. Thus the adult cortex presents various histological areas, each possessing its own characteristic stratification. The visual cortex is a particularly marked example. Its characteristic consists in the subdivision of the internal granular layer into two layers, between which is formed a conspicuous white line, the so-called line of Gennari. Another departure from the general type is found in the hippocampus and fascia dentata, which differ in the well-known way from the cortex seen in other regions. On the outer surface of the cortical layer there is frequently seen during the fourth month, as shown in Fig. 75, an irregular fungiform clumping of cells, the so-called Eetzius papillae. These, however, are an artefact, being a result of partial maceration of the tissues.

FORMATION OF SULCI AND CONVOLUTIONS.

In the further growth of the brain wall the white matter continues to increase in thickness, while the cortical zone (gray matter) remains spread out in a relatively thin layer, its expansion taking place in a plane parallel with the surface of the brain wall. To accommodate this increase in the extent of its surface the outer surface of the brain wall is thrown into folds. The lines along which the principal folds form correspond (usually) to the boundaries of different histological areas, due perhaps to the difference in their time of development or possibly in consequence of their different reaction to the tension existing between the gray and white substance. Since the different histological areas represent constant functional areas, the fissures are therefore more or less constant.

The first fissures to appear (about the third month) are those associated with the primitive olfactory system, the hippocampal and rhinal. The hippocampal or arcuate fissure forms along the border of the dentate fascia, as shown in Fig. 67. An accessory arcuate fissure on the median wall dorsal to the hippocampal fissure

DEVELOPMENT OF THE NERVOUS SYSTEM is described by His as occurring about the same time. By other writers (Hochstetter, Goldstein) this and the so-called fissura prima are regarded as post-mortem phenomena. The fissura prima, or anterior arcuate fissure, is described by His as being the first fissure to appear (the second month). It is found on the median wall near the olfactory bulb, and extends upward in front of the preterminal body of G. Elliot Smith (that is, the trapezoid body of His). Part of it is thought to persist as the fissura parolfactoria posterior. According to His, an extension of the pia mater extends into and corresponds in form to the fissure, and he argues that therefore it must be a real fissure. His opponents deny its presence in well-preserved brains. The rhinal fissure Lobus parietal is

Lobus occipitalis .

Insula

-- Lobus frontalis

Lobus temporalis Bulb us olfactorius Traetus olfactorius lateralis Fig. 76. — Lateral view of reconstruction of cerebral hemisphere of a human fetus about three months old (crown-rump length 80 mm., Mall collection, No. 234a). The rhinal fissure is situated along the upper border of the traetus olfactorius lateralis. The dark-shaded portion at the base represents that portion that is not covered in by a typical cortical layer.

(Fig. 76) separates the lobus piriformis from the neopallium, but in man, owing to the suppression of the olfactory apparatus, it always remains insignificant.

The development of the Sylvian fissure is not completed until after birth, but the first stages in its formation can be seen at the third month. On comparing Figs. 76, 77, and 79 it will be seen that its formation is dependent on the fact that the brain wall in the region of the corpus striatum does not enlarge as rapidly as the parts adjacent to it. This at first expresses itself in the formation of a shallow depression, the fossa Sylvii (see Fig. 76). As the neighboring temporal, frontoparietal, and orbital portions become thicker they form in-rolling walls or lips, the so-called opercula, which finally cover in the retarded portion or insula. The lines along which the lips meet constitute the fissure of Sylvius. The temporal and frontoparietal opercula are formed first. The

102 frontal and orbital opercula are very late in development. They do not begin to form until the insula is already partly covered in by the temporal and frontoparietal opercula, and they do not

Lobus parietalis

/

Lobus occipitalis -9

Lobus frontalis ',

Insula 

Bulbus olfactorius

Lobus temporalis

Fia. 77. — Lateral view of reconstruction of cerebral hemisphere of human fetus at the beginning of the fifth month (crown-rump length 150 mm., Mall collection). A portion of the^tractus olfactorius lateralis can. be .seeni'at the lower border of the insula.

come into apposition with each other and the other two opercula, so as to close in the anterior part of the insula, until after birth. In Fig. 77 the advanced growth of the temporal, occipital, and parietal lobes, corresponding to their primitive functions, is very evident. The frontal lobe whose functional activity is the last to be required is correspondingly backward. In consequence the brain in fetuses between 50 and 150 mm. long, when seen from above or in front, as in Fig. 78, resembles in its outward appearance the smooth forward tapering brain seen in some of the lower mammals, — e.g., rabbit. Variations in the degree of development of the frontal operculum determine the shape of the two anterior limbs of the Sylvian fissure between which it lies. When well developed, it separates the two Sylvian limbs from each other so that they assume a U shape; when less developed, it forms a V shape; or, if so poorly developed that the orbital and frontoparietal opercula meet so as to occlude the frontal operculum from the main limb of the fissure, then we have a Y shape. In some forms of arrested development the anterior portion of the fissure of Sylvius is defective and the insula remains partly exposed.

Fig. 78. — Anterior view of brain of a human fetus of about the same age as shown in Fig. 76. (After His.)

DEVELOPMENT OF THE NERVOUS SYSTEM.

108

The hippocampal, rkinal, and Sylvian fissures develop along boundary lines of brain areas that differ markedly both in structure and in rate of development. The calcarine, parieto-occipital, and central fissures also represent boundary lines of areas that differ histologically, but the differences are not so marked as in the former cases, and the fissures appear somewhat later, during the fifth month. They are soon followed by the collateral, inferior and superior precentral, postcentral, superior temporal, superior and inferior frontal, parolfactory, interparietal, callosomarginal, and orbital fissures, all of which appear during the sixth or seventh month. 2 The character of these fissures at the end of the seventh month is shown in Figs. 79, 80, and 81. These fissures first appear as shallow furrows, and an individual fissure may first appear as

Sulcus postcentralis

Sulcus centralis

Lobus parietalis superior Region of gyrus supramargin.

et angularis Ramus posterior

Sulcus temporalis medius

Polus posterior

Sulcus frontalis inferior Ramus anterior ascendens Fissura cerebri lateralis (Sylvii)

Lobus temporalis

Gyrus temporalis superior Gyrus temporalis medius Fig. 79. — Lateral view of the cerebral hemisphere of a human fetus at the end of the seventh month, showing the formation of the early fissures. Compare with Figs. 76 and 77. (After Kollmann.)|

several short furrows which as they deepen subsequently unite into a continuous furrow. The calcarine fissure, somewhat like the hippocampal fissure, involves the whole thickness of the brain wall and produces an elevation within the ventricle, the calcar avis. The parieto-occipital fissure apparently also causes at first

'A distinction is frequently made between sulci and fissures, based on the conditions found in the adult. The term sulcus is used for the more shallow grooves, and the term fissure for the deeper ones, those which in their development involve the whole thickness of the brain wall and influence the form of the ventricle. Thus we speak of the fissure of Sylvius, the longitudinal, the hippocampal, the collateral, the calcarine, and the parieto-occipital fissures; all other grooves are designated as sulci. From the embryological stand-point it might be better still further to restrict the term fissure and limit it to the fissure of Sylvius and the longitudinal fissure. We would then have the term sulcus as representing all those furrows which are formed as actual grooves in the brain wall, and the term fissure would be limited to the clefts that result secondarily from the unequal growth of major brain regions, and are not true indentations in the original surface of the brain wall.

104 an infolding of the brain wall. This, however, disappears with the thickening of the wall. It is possible that it is to be classed as an artefact together with the so-called transitory fissures. The transitory fissures consist of sharply marked furrows frequently found indenting all parts of the cortex during the third and fourth months. They are irregular in form and position and are only found at this time. It is to be remembered that at this period the human hemisphere consists of a large thin-walled vesicle whose walls have not yet developed a firm framework, and, even with good material and with great care in its treatment, it is difficult to prevent artificial foldings of the wall when the specimen is immersed in preserving fluids.

Corp. Sulcus corp. callosi Gyrus cinguli callosum | Splenium Fissura parieto- occipitalis

Cavum septi pellucidi Lamina rostralis Area parolfactoria

Cuneus

Fissura calcarina

Nervus olfactorius ' | Fissura rhinica Nervus opticus Lobus temporalis Fig. 80. — Median view of the cerebral hemisphere of a human fetus at the end of the seventh month, being the same specimen shown in Fig. 79. (After Kollmann.) These primary fissures constitute boundaries of primary functional areas of the pallium. Subsequently association areas are developed around them, forming new cortical territories, and as these expand new furrows develop to accommodate the growing cortex. The secondary gyri thus formed may crowd the older ones into new forms or even partly replace them by burying them under, as we have seen the insula buried by the adjoining opercula.

DEVELOPMENT OP THE MYELIN SHEATHS.

The final phase in the development of the hemisphere wall consists in the process of myelinization of its nerve-fibres. This does not begin until about the time of birth and it continues from then until the end of puberty. For the details of this process the reader is referred to the studies of Flechsig, to whom we are indebted for almost the whole of our present knowledge of this, subject. We will here only give a brief outline of the process.

DEVELOPMENT OF THE NERVOUS SYSTEM.

105

The process begins in the projection fibres of the four primn sensorimotor fields : 1, the olfactory ; 2, the visual ; 3, the acoustic ; and 4, the somatic. These possess both efferent and afferent elements. In the visual and acoustic mechanisms the efferent element is very small. In the somatic area both the efferent and afferent elements are largely represented, each occupying a definite portion of the total area, the afferent forming the postcentral area (somaesthetic) and the efferent the precentral area (motor). The afferent projection fibres are probably myelinated shortly before the efferent ones.

Gyrus frontalis medius Gyrus frontalis inferior Gyrus frontalis superior Gyrus prsecentralis Gyrus centralis posterior Lobulus parietalis superior Lobulus parietalis inferior Lobus occipitalis

Sulcuslfrontalis'superior Sulcus frontalis inferior

— Sulcus prsecentralis

Sulcus centralis Sulcus postcentrals

— • -4 — Sulcus interparietals

— Fissura parieto-occipitalis

Fig. 81. — Dorsal view of the cerebral hemisphere of a human fetus at the end of the seventh month, being the same specimen shown in Figs. 79 and 80. (After Kollmann.) The process next spreads to a series of intermediate areas whose projection fibres serve to connect the primary cortical areas with the thalamic and pontine nuclei. The terminal areas to become myelinated are those made up almost entirely of association neurones, whose axones cross in the corpus callosum to the opposite hemisphere or extend to distant or near parts of the same hemisphere. .

In concluding this chapter, there is added a table, taken from His (1904), showing the order of development of the different fibre tracts of the central nervous system. The size and approximate age of the embryo or fetus is indicated, and in the columns the first recognizable appearance of a given tract is indicated by a check.

106 •q^nora

+ + + + + + + + + + + + + + + + + + + +

§ a .a

§ + + + + + + + + + + + + + + + + + + + +

CO 00

++++++++++++++++++

o to

+++++++++++++++

&5

a o

CO

to

+++++++++++++++

s +++++++++++++++

to

++++++++++++++

8 ++++++++++++++

05 CI

++++++++++++++

3 § a 3

CM

+ + + + + + + + + + + + + +

+ + + + + + + + + + + +

U9 00

+ + + + + + + + + + + +

00

+ + + + + + + + + + +

+ + + + + + + + + + +

£ + + + + + + + + + +

+ + + + + + +

+ + + + + + +

00 CO

+ + + + + +

©

+ + + + + + :

©

+ + +

•q^notu

to

+ + +::::

a"J oJa w .a •OSS fl 1) t.

§ > a m K ^.2 2 <c » a ? a V c3 - 2 a N a o 5

.a a •c

00 .— 73 cO G "^ •3 a C O oo on u d GO CO cO cO

35

G G CD

a^-fi

8J a •c

60 Si

G cO g ,2 a * • • a o o o a §^ fe fe ^

O CD

ifi 6 3 a « £5 ° 3 00 +s GO CD

a oo O CO

• 00 • <D t • -Q ^ <D .2 .2 > -3

IB ^ a a M g 00 C .1 S'§ a 2« a §1 » « -a hJ Q H

-si G

-G +a

00 s 'E

2 oo a s ° s 1"? o a) "-3 «&• ,;

G

C

° .2 « G « G o "G Oj ~ < g g «*-> •-> Oh

G *^

7i w C G G cO g K — ~>

G oo O =3 CO u

7}

00 — 2

00 5-°

c 13 as

a

t_ ' ^N* w* u— , www* •*-« ^

DEVELOPMENT OF THE NERVOUS SYSTEM. 107 LITERATURE.

Adamkiewicz, A.: Die Blutyefiisse des menschlichen Riiekenniarkes. Sitz. d. Kgl.

Akad. d. Wiss. in Wien. Bd. 84. III. Abt. S. 469. 1881. Argutinsky, P. : Ueber eine regelmassige Gliederung in der grauen Snbstanz des Riickenmarks beim Nengeborenen und iiber die Mittelzellen. Arch. f.

mikrosk. Anat. Bd. 48. " 1897. Ueber die Gestalt und die Entstebungsweise des ventrieulus terminalis und iiber das filuni terminale des Riickenmarks bei Nengeborenen. I. Mitt. Arch. f.

mikrosk. Anat. Bd. 52, S. 501-534. 1898. Barker, L. F. : Specimens Illustrating the Medullating Cerebrum in Human Beings.

Journ. of Nerv. and Ment. Diseases, p. 343. 1898. On the Relation of the Third Fetal System of Trepinski to the Direct Cerebellar Tract. Amer. Journ. of Anat. Vol. 2. No. 2. Proc. Assoc.

Amer. Anat., p. 15. 1902-3. Bechterew : Ueber die Entwicklung der Zellelemente in der Grosshirnrinde des Menschen. Neurol. Centralbl. Bd. 18, S. 770-772. 1899. Compare also Bechterew's work reported by Ziehen in Hertwig's Handbuch, S. 504-505. Ueber die Darstellung der Riickenmarksysteme mit Hilfe der Entwicklungs methode. Arch. f. Anat. u. Physiol. Anat. Abt. S. 280-296. 1901. Berliner, K. : Beitrage zur Histologie und Entwicklungsgeschichte des Kleinhirns.

Inaug.-Diss. Breslau 1904. Beitrage zur Histologie und Entwicklungsgeschichte des Kleinhirns, nebst Be merkungen iiber die Entwicklung der Funktionstiichtigkeit desselben. 1 Taf.

u. 19 Fig. Arch. f. mikrosk. Anat. Bd. 66, S. 220-269. 1905. Bertachini, P. : Intorno alia struttura anatomica dei centri nervosi di un embrione umano lungo 4.5 mm. Intern. Monatsschr. f. Anat. u. Physiol. Bd. 14.

1897. Descrizione di un giovanissimo embrione umano con speziale riguardo alio sviluppo dei centri nervosi. Internat. Monatsschr. f. Anat. u. Physiol.

Bd. 15, p. 3-25. 1898. Bertelli, D. : Le sillon intermediaire anterieur de la moelle humaine dans la premiere annee de vie. Arch. ital. de Biol. T. 11, 1889, p. 420 and Atti di Soc. Tosc. d. Sc. nat. Vol. 10 u. 11. 1889. Bischoff, Th. L. W. : Die Grosshirnwindungen des Menschen mit Berucksichtigung ihrer Entwicklung bei dem Fetus und ihrer Anordnung bei den Affen. Abh.

d. 2 Kl. d. Kgl. bayr. Akad. d. Wiss. Bd. 10. II. Abt. S. 391^97. 1868. Blake, J. A. : The Roof and Lateral Recesses of the Fourth Ventricle, Considered Morphologically and Embryologically. Journ. Compar. Neurology. Vol. 10, p. 79 to 108. 1900. Bltjmenau, L. : Zur Entwicklungsgeschichte und feineren Anatomie des Hirn balkens. Arch. f. mikr. Anat. Bd. 37, S. 1-15. 1890. Bolk, L. : Das Cerebellum der Saugetiere. 3 Taf. u. 183 Fig. Fischer, Jena 1906. Over de ontwikkeling van het cerebellum bij den mensch. 1 Taf. Erste med.

Vers. wiss. nat. Afd. Acad. Wet. Amsterdam. D. 13, p. 635-641. 1905. Bradlet, O. C. : The Mammalian Cerebellum: Its Lobes and Fissures. Journ.

Anat. and Physiol. Vol. 38, p. 448-475; Vol. 39, p. 99-117. 1904. Broca, P.: Sur le cerveau a l'etat foetal. Bull, de la Soc. d'Anthrop. Ser. 2.

T. 12, p. 216-222. 1877. Brodmann, K. : Bemerkungen iiber die Fibrillogenie und ihre Beziehungen zur Myelogenie mit besonderer Berucksichtigung der cortex cerebri. Neurol.

Centralbl. S. 338-349. 1907. Brugsch, Theodor and Unger, E. : Die Entwicklung des ventrieulus terminalis beim Menschen. 8 Fig. Arch. f. mikrosk. Anat. Bd. 61, S. 220-232. 1902

108 HUMAN EMBRYOLOGY.

Cajal, S. R. : Estudios sobre la eortezza cerebral umana (cortezza visual, cortezza motriz). Rev. trim, niicrogr. Vol. 4, p. 1-63 and p. 117-200. 1899. Apuntes para el studio estructural de la cortezza visual del cerebro humano.

5 Fotogr. Rev. Ibero-Americana de Cienc. med. 1899. Die Struktur des chiasroa opticum nebst einer allgeineinen Theorie der Kreuz ung der Nervenbahnen. Aus dem Spaniscken von J. Bresler. Mit einera Vorwort von Flecksig. Leipzig 1899. Studien liber die Himrinde des Menseken. Deutsck von Bresler. Leipzig 1900,. 1901, 1902. Callendar, G-. W. : Lectures upon tke Formation and Early Growtk of tke Brain of Man. Brit. Med. Journ., p. 731. 6 June, 1874. Carus, C. G. : Versuck einer Darstellung des Nervensystems und insbesondere des Gekirns usw. S. 258-297. Leipzig 1814. Clarke, J. L. : Researckes on tke Development of tke Spinal Cord in Man, Mammalia, and Birds. Pkil. Transact. R. Soc, London. Part 2. Vol. 152, p. 911 to 938. 1863. Colucci E Piccinino : Su alcuni stadii di sviluppo delle cellule del midollo spinale umano. Ann. di Nevrol. Vol. 18. 1900. Cramer, A. : Das kintere Langsbiindel, fasciculus longitudinalis dorsalis, nack Untersuckungen am menscklicken Fetus, Neugeborenen und 1-3 Monate alten Kindern. 4 Taf . und 14 Fig. Anat. Hef te. 1 Abt. Bd. 13, S. 97-202. 1899. Cunningham, D. J. : Intraparietal Sulcus of tke Brain. Journ. of Anat. and Physiol., p. 135-155. 1890. The Complete Fissures of the Human Cerebrum, etc. Journ. of Anat. and Physiol. Vol. 24, p. 309-345. 1890. The Development of the Gyri and Sulci on the Surface of the Island of Reil of the Human Brain. Journ. of Anat. and Physiol. Vol. 25, p.

338-348. 1891. Contribution to the Surface Anatomy of the Cerebral Hemispheres. Mem.

Roy. Sc. Acad. Dublin Acad. House. 1892. A Demonstration Illustrating the Development of the Cerebral Sulci in the Human Brain. Journ. for Mental Science. Vol. 40. 1894. The Rolandic and Calcarine Fissures, a Study of the Growing Cortex of the Cerebrum. Journ. of Anat. and Physiol. Vol. 31, p. 586-598. 1897. The Insular District of the Cerebral Cortex in Man and in the Man-like Apes. Journ. of Anat. and Physiol. Vol. 32, p. 11-22. 1898. Cutore, G. : Ricerche anatomo-comparative sullo sviluppo, sull' istogenesi e sui carratteri definitivi dell' esterno caudale del midollo spinale. 7 Taf. Arch.

ital. Anat. e Embriol. Anno 4. 1905. Dandy, W. E. : A Human Embryo with Seven Pairs of Somites. Amer. Journ.

of Anat. Vol. 10. 1910. Danielbekof : Materialen zur Frage iiber das Gewicht des Gehirns und der med.

obi. bei Kindern beiderlei Geschlechts. Petersburg 1885. Deniker, J. : Reeherches anatomiques et embryologiques sur les singes anthro poi'des etc. Arch, de Zool. exper. et gen. Ser. 3. T. 3. 1886. Doellinger, I. : Beitrage zur Entwieklungsgeschichte des menschlichen Gehirns.

Frankfurt a. M. 1814. Dollken: Zur Entwicklung der Schleife und ihrer zentralen Verbindungen.

Neurol. Centralbl. S. 50-61. 1899. Donaldson, H. H. : The Growth of the Brain. 1895. Dursy, E. : Zur Entwieklungsgeschichte des Kopfes des Menschen und der hoheren Wirbeltiere. Tubingen 1869. Vgl. auch Centralbl. f. d. med. Wiss. 1868. Ecker, A. : . Icones physiologicae. Leipzig 1851/59.

Zur Entwieklungsgeschichte der Furchen und Windungen der Grosshirnhemi spharen im Fetus des Menschen. Arch. f. Anthrop. Bd. 3,S. 203-225. 1868.

DEVELOPMENT OF THE NERVOUS SYSTEM. 109 Die Hirnwindungen des Menschen Qaeh eigenen [Jntersuchungen, insbesondere iiber die Entwieklung derselben beim Fetus asw. '_' Aufl. Braunsehw( 18S3. Eichhorst, H. : Ueber die Entwieklung des menschliehen Riiekenmarks und seiner Formelemente. Vircbow's Arch. Bd. 64, S. 425 475. L875. Engel: Beitrage zur Entwieklung des Gehirns. Sitz.-Ber. d. ruath.-natui Kl. d. Kais. Akad. zu Wien. Bd. 12, S. 209. 1854. Erchia, F. d' : Di un piccolo corpo vesicolare posto sulla volta del cervello anteriore di un embrione umano, lunge 3.5 mm. in linea retta. Monit.

zool. ital. Vol. 12, p. 40-43. 1901. Falcone, C. : Sopra alcune particolarita di sviluppo del midollo spinale. Note di embriogenia eomparata. Mit 4 Taf. Arch. ital. Anat. Embriol. Vol. 1, p. 97-119. 1902. Fere, Ch. : Sur le developpement du cerveau considere dans ses rapports avec le crane. Rev. d'Anthropol, p. 661-674. 1879. Fischer, E. K. : Ueber die Furcben und Windungen des kindlichen Gehirns im ersten Halbjahr des Extrauterinlebens. Diss. St. Petersburg 1902. (Russian.) Flechsig, P.: 1st die tabes dorsalis eine " System-Erkrankung " ? Neurol. Centralbl.

Bd. 9. 1890. Neue Untersuebungen iiber die Markbildung in den menschliehen Grosshirn lappen. Neurol. Centralbl., Jahrg. 17. S. 977-996. 1898. Weitere Mitteilungen iiber die entwicklungsgeschiehtlichen (myelogenetischen) Felder in der menschliehen Grosshirnrinde. Neurol. Centralbl. Jahrg. 22.

Nr. 5, S. 202-206. 1903. Die innere Ausbildung des Gehirns der rechtzeitig geborenen menschliehen Frucht. Versammlung mitteldeutscher Psychiater und Neurologen 1903.

Neurolog. Centralbl. Bd. 22, S. 1079-1080. 1903. Forli, V.: Sulla mielinizzazione del lobo frontale. Ann. Instit. Psich. R. Univ.

Roma. Vol. 2, p. 152-216. 1903. Fragntto, O. : Alcune questioni su lo sviluppo della corteceia cerebrale nei mammiferi. Ricerche. 4 Fig. Ann. di Nevrol. Anno 24. Fasc. 5, 6. p.

369-387. 1907. Fuchs, S. : Zur Histogenese der menschliehen Grosshirnrinde. Sitz.-Ber. der Wiener Akad. T. 3. Bd. 88, S. 157-184. 1884. Gage, S. P.: The Total Folds of the Forebrain, their Origin and Development to the Third Week in the Human Embryo. Amer. Journ. Anat. Vol. 4.

Proc. Assoc. Amer. Anatomists. 1905. Gaupp, E. : Zirbel, Parietalorgan und Paraphysis. Ergebn. d. Anat. u. Entwiek lungsgesch. Literatur 1897. Bd. 7, S. 208-285. 1898. Gehuchten, A. van : La neurologie dans le cervelet de l'homme. Bibliogr. anat.

Annee 2. 1894. Gemelli, E. : Nuove ricerche sull' anatomia e sull' ernbriologia dell' ipofisi. Bull.

Soe. med.-ehir. Pavia. Nr. 3. 1903. Gierlich, N. : Kurze Bemerkungen iiber Fibrillogenie im Centralnervensj des Menschen zur Arbeit Brodmanns: Bemerkungen iiber die Fibrillogenie.

Neurolog. Centralbl. Jahrg. 26. 1907. Giese, E. : Ueber die Entwieklung der Neuroglia im Riickenmark des Menschen.

Wissensch. Vers. Arzte St. Petersburger Klin. f. Nerven- u. Geisteskranke.

Sitz. 27. Marz 1897. Ref. in: Neurolog. Centralbl. Jahrg. 17. S. 139.

1897. Ueber die Bestandteile der weissen Snbstanz des menschliehen Riickenmarkes nach der entwicklungsgeschiehtlichen Methode. Deutseher Auszug aua der russ. Inaug.-Diss. 1899. Giglio-Tos, E. : Sulle cellule germinative del tubo midoilare embrionale dell' umano. 6 Fig. Anat. Anz. Bd. 20, p. 472 480. 1901.

110 HUMAN EMBRYOLOGY.

Goldstein, K. : Beitrage zur Entwicklungsgeschichte des nienschlichen Gehirns.

I. Die erste Entwicklung der gTossen Hirnkommissuren und die Verwachsung von thalamus und striatum. Arch. f. Anat. u. Physiol. Anat, Abt. Heft 1, S. 29-60. 1903. Zur Frage der Existenzberechtigung der sogenannten Bogenfurchen usw. Anat.

Anz. Bd. 24, S. 579-595. 1904. Groschuff, K. : Ueber sinnesknospenahnliche Epithelbildungen im Centralkanal des embryonalen Riickenmarks. Sitz.-Ber. Ges. Morphol. Physiol. Munch. 1897. Gratiolet, P. : Anat. comparee du systeme nerveux von Leuret et Gratiolet. T. 2.

Kap. 7. p. 224-265. 1839-1857. Hamilton, D. J. : On the Corpus Callosum in the Embryo. Brain. Vol. 8, p. 145-163. 1886. Hamy, E. T. : Contribution a l'etude du developpement des lobes cerebraux des primates. Rev. d'Anthrop. T. 1, p. 424-431. 1872. Hensen, V. : Beitrag zur Morphologie der Korperform und des Gehirns des menschliehen Embryos. Arch, f . Anat. u. Physiol. Anat. Abt. S. 1-8. 1877. Herrick, C. L. : The Histogenesis of the Cerebellum. Journ. of Comp. Neurol.

Vol. 5. 1895. Herxheimer, G. : Zur Entwicklung der Nervenf asern im Gehirn und Riickenmark.

Verb. Deutsch. patholog. Gesellsch. 10. Tagung Stuttgart 1906, erschien Jena 1907. S. 139-143. 1906-07. Hess, C. : Das foramen Magendie und die Offnungen an den recessus laterales des 4. Ventrikels. Morphol. Jahrb. Bd. 10, S. 578-602. 1885.

His, W. : Ueber das Auftreten der weissen Substanz und der Wurzelf asern am Riickenmark menschlicher Embryonen. Arch. f. Anat. u. Physiol. Anat. Abt. S. 163-170. 1883. Ueber die Entstehung und Ausbreitung der Nervenfasern. Anat. Anz. Bd. 1, 5. 284 u. 285. 1886.

Zur Geschichte des menschliehen Riickenmarks und der Nervenwurzeln. Abh.

d. math.-phys. Kl. d. Kgl. Sachs. Gesellsch. d. Wiss. Nr. 6. Bd. 8. 1886. Zur Geschichte des Gehirns, sowie der centralen und peripherischen Nerven bahnen beim menschliehen Embryo. Abh. d. math.-phys. Kl. d. Kgl. Sachs.

Gesellsch. d. Wiss. Bd. 14, S. 339-392. 1888. Die Formentwicklung des menschliehen Vorderhoms vom Ende des 1. bis zum Beginn des 3. Monats. Abh. d. math.-phys. Kl. d. Kgl. Sachs. Gesellsch. d.

Wiss. Nr. 8. Bd. 15, S. 675-736. 1889. Ueber die Entwicklung des Riechlappens und des Rieehganglions und uber diejenige des verlangerten Markes. Verh. d. Anat. Gesellsch. Anat. Anz.

Erg.-Heft. Bd. 4. 1889. Die Neuroblasten und ihre Entstehung im embryonalen Mark. Abh. d. math.phys. Kl. d. Kgl. Sachs. Gesellsch. d. Wiss. IV, und Arch. f. Anat. und Physiol. Anat. Abt. S. 249-300. Jahrg. 1889. Die Entwicklung des menschliehen Rautenhirns vom Ende des 1. bis zum Beginn des 3. Monats. I. Verlangertes Mark. Abh. d. math.-phys. Kl. d.

Kgl. Sachs. Gesellsch. d. Wiss. Bd. 17, S. 1-74. 1891. Zur allgemeinen Morphologie des Gehirns. Arch. f. Anat. u. Physiol. Anat.

Abt. S. 346-383. Jahrg. 1892. Ueber das frontale Ende des Gehirnrohres. Arch. f. Anat. u. Physiol. Anat.

Abt. S. 157-171. 1893. Ueber das frontale Ende und iiber die naturliche Einteilung des Gehirnrohres.

Verh. d. Anat, Ges. 1893. Vorschlage zur Einteilung des Gehirns. Arch. f. Anat. u. Physiol. Anat.

Abt. ' S. 172-179. 1893. Address upon the Development of the Brain. Trans. R. Acad. Med. Ireland.

Vol. 15. 1898. Die Entwicklung des menschliehen Gehirns wahrend rler ersten Monate.

Leipzig 1°04.

DEVELOPMENT OF THE NERVOUS SYSTEM.

Hochstetter, F. : Ueber die Beziehung des thalamus opticus zum Seitenventrikel der Grosshirnheniispharen. Anat. Anz. Bd. 10, S. 295-302. 1895. Beitrage zur Entwieklungsgesebichte des Gebirns. 4 Taf. Bibliogr. med.

Abt. A. Heft 2. Stuttgart 1898. Ueber die sogenannten transitorisehen Furcben der Grosshirnheinispharen uiensehlieher Enibryonen, nebst Beruei'kungen iiber die Bogenfurche. Wien.

kiin. Woebensebr. Jabrg. 10. Nr. 17. S. 409. Ueber die Nicbtexistenz der sogenannten Bogenfureben an den Gehirn lebens frisch konservierter menscblieber Embryonen. Yerh. d. Anat. Gesellsch.

S. 27. Anat. Anz. Erg.-Heft. Bd. 25. 1904. Hoesel, O. : Beitrage zur Markscheidenentwieklung im Gebirn und in der medulla oblongata des Menseben. 11 Fig. Monatsschr. f. Psyeh. u. Neurol. Bd. 6, S. 161-192. Bd. 7, Heft 4, S. 265-282, 11 Fig. u. 1 Taf.; Heft 5, S. 345-365. Hole, M. : Ueber die Insel des Menseben- und Anthropoidengehirns. Arch, f . Anat.

u. Physiol. Anat. Abt. S. 1-44. 1902. Huschke, E.: SehadeL Hirn und Seele. Jena 1854. Johnston, J. B. : The Problem of the Correlation Mechanisms. Proceed. Amer.

Assoc, of Anat. Boston, Anat. Record, Vol. 4. 1910. Kaes, Theodor : Ueber Grosshirnrindenmasse und iiber Anordnung der Markf aser systeme in der Rinde des Menseben. Verb. d. Vers. Deutsch Naturf. u.

Artze. Bd. 2. Abt. 2, and Wien. med. Wochenschr. Jahrg. 45. Nr.

41 and 42. S. 1733-1739 and S. 1769-1775. 1895. Die Rindenbreite als wesentlicber Faktor zur Beurteilung der Entwicklung des Gehirns und namentlieh der Intelligenz. 6 Fig. Neurol. Centralbl.

Jahrg. 24. S. 1026-1047. 1905. Die Grosshirnrinde des Menseben in ihren Massen und ihr Fasergehalt. 79 Taf. Jena 1907. Karusin, P. : Die Faserziige des Riickenmarks, entwicklungsgeschichtlich unter sueht. Moskauer Doetor-Diss. Moskau 1894. Kastanajan, E. M. : Phylogenese und Ontogenese des Kommissurensystems des Vorderhirns. Obosr. Psich. Jahrg. 8. Nr. 4. St. Petersburg 1903.

(Russian.) Review of literature, no new investigations. Keibel, Fr. : Ueber den Schwanz des menschlichen Embryo. Anat. Anz. Bd. 6, S. 670-675. 1891. (Compare also Arch. f. Anat. und Physiol. Anat.

Abt. 1891 u. 1896.) Zur Entwieklungsgeschichte der Affen. Verh. d. Anat. Gesellsch. S. 156-163.

Jena 1904. Anat. Anz. Erg.-Heft. Bd. 25. Kolliker, A. v. : Entwieklungsgeschichte des Menseben. 2 Aufl. Engelmann, Leipzig 1879. Ueber den feineren Bau des Riickenmarks menscblieber Embryonen. Sitz. Ber. d. phys.-med. Gesellsch. zu Wurzburg. S. 126. 1890. Handbucb der Gewebelebre. Leipzig 1S93. KOLLMANN, J. : Handatlas der Entwieklungsgeschichte des Menseben. Jena 1907. Kuithan, "W. : Die Entwicklung des Kleinhirns bei Saugetieren. Diss. Miinchen.

Also : Miinchener med. Abb. Reihe 7. Arb. a. d. Anat. Institut. Heft 6.

1895. Kupffer, K. v. : Die Morpbogenie des Centralnervensystems. In : O. Hertwig'a Handbuch. 1906. Lachi, P. : La tela corioidea superiore e i ventrieoli cerebrali dell' uomo. Atti della Soc. Tosc. d. Sc. nat. Vol. 9, Fasc. 1. 1888. Sul rapporto del talamo ottico col ventricolo Iaterale dell' emisfero cerebrale.

Anat. Anz. Bd. 10. p. 537-538. 1895. Lahousse, E. : Recherehes sur l'ontogenese du cervelet. Arch, de Biol. T. S.

p. 43-110. 18S8.

112 HUMAN EMBRYOLOGY.

Laigxel-Lavastine : Note sur la presence des cellules pyramidales binucleees dans l'ecorce cerebrale d'un nouveau-ne. Bull. Mem. Soc. anat. Nr. 7. Paris 1903. Lexhossek, M. v. : Der feinere Bau des Nervensystenis im Lichte neuester Forsehungen. 2. Aufl. Berlin 189-3. Levi, G. : Ueber die Entwicklung und Histogenese der Amnionsbornf ormation.

1 Taf. Arch. f. mikrosk. Anat. Bd. 64, S. 389-104. 1904. Loewe, L. : Beitrage zur Anatoinie und zur Entwicklungsgescbichte des Nerven systerns der Saugetiere und des Menseben. Berlin 1880. Beitrage zur vergleickenden Morpbogenesis des zentralen Nervensystems der Wirbeltiere. Sehenk's Mitteil. a. d. embryol. Inst. Bd. 2. 1880. Long, M. : On the Development of tbe Nuclei Pontis During the Second and Third Months of Embryonic Life. 4 plates. Bull. Johns Hopkins Hosp. . Vol. 12, p. 123-126. 1901. Lubimopf, A. : Embryologiscbe und histogenetische Untersuchungen iiber das sympatbische und centrale Orebrospinalnervensystem. Virch. Arch. Bd.

40, S. 217-273. 1874. Lugaro, E. : Sulla genesi delle circonvoluzioni cerebrali e cerebellari. Riv. di Patol. nerv. e ment. Vol. 2. 1897. Mall, F. P. : On the Transitory or Artificial Fissures of the Human Cerebrum.

Amer. Journ. Anat. Nr. 3. Vol. 2, p. 333-340. 1903. A Human Embryo Twenty-six Days Old. Journ. of Morph. Vol. V. 1891. Manouelian, Y. : Sur le mode de develop pement des arborisations grimpantes du eervelet. C. R. Soc. Biol. p. 333-334. 1899. Marchand, F. : Ueber die Entwicklung des Balkens im menschlichen Gehirn. Arch.

f. mikrosk. Anat. Bd. 37, p. 333-340. 1891. Die Morpbologie des Stirnlappens und der Insel der Anthropomorphen. 1893. Ueber das Gehimgewicht des Menschen. Abh. d. math.-phys. Ell. d. Kgl. Sachs.

Gesellseh. f. Wiss. Bd. 27, S. 394-481. 1902. Ueber die normale Entwicklung und den Mangel des Balkens im menschlichen Gehirn. Abb. d. matb.-phvs. Kl. d. Kgl. Sachs. Gesellseh. d. Wiss.

Nr. 8. Bd. 31. 1909. Maschtakow, A. W. : Die Entwicklung der Rinde der Centralwindungen im Kindesalter. Diss. St. Petersburg 1902. (Russian.) Meckel, J. F. : Versucb einer Entwicklungsgeschicbte der Zentralteile des Nervensystems in den Saugetieren. Deutscb. Arch. f. Physiol. Bd. 1, p. 76 ff.

and 334 ff. (Compare also Beitr. z. vergl. Anat. Bd. 2, Heft 1.) 1815. Meyer, A. : Zur Homologie der Fornixkommissur und des septum lucidum bei den Reptilien und Saugern. Anat. Anz. Bd. 10. 1895. Mies, J. : Ueber das Gehimgewicht des heranwacbsenden Menschen. Mitteil. d.

Antbropol. Ges. in Wien. Bd. 24. (N. F. Bd. 14.) 1895. Mihalkovics, V. v. : Die Entwicklung des Gehirnbalkens und des Gewolbes.

Centralbl. f. d. med. Wiss. Nr. 19. 1876. Entwicklungsgeschiehte des Gehirns. Leipzig 1877. Mingazzini, G. : Intorno ai solchi e le circonvoluzioni cerebrali dei primati e del feto umano. Estratto dagli Atti della N. Accad. med. di Roma. Anno 15.

Vol. 4. Ser. 2. Roma. 1888.— Bull. d. R. Accad. med. di Roma. T. 9.

1887-1888. Ueber die Entwicklung der Furchen und Windungen des menschlichen Gehirns.

Moleschott's Untersuchungen zur Naturlehre. Bd. 13, S. 498-562. 1888. Ueber die Furchen und Windungen des Gebims der Primaten und der menschlichen Frucht. Moleschott's Untersuchungen zur Naturlebre des Menschen und der Tiere. Bd. 14. 1889. Mirto, D. : La mielinizzazione del nervo ottieo come segue di vita extrauterina produtta nei neonati prematuri ed a termine. 1 Taf. Pizani. Vol. 23, p. 5-31. 1902.

DEVELOPMENT OF THE NERVOUS SYSTEM. 113 Obersteiner, H. : Ueber das hellgelbe Pigment in den Nervenzellen und das Vorkommen weiterer fettahnlicher Korper im Zentralnervensystem. Arb.

Neurol. Instit. Wien. Univ. Bd. 10. S. 245. 1903.— Ref. in: Neurol.

Centralbl. Jahrg. 23. Nr. 6. S. 259. 1904. Osborn, H. S. : The Origin of the Corpus Callosum. a Contribution upon the Cerebral Commissures of the Vertebrates. Morphol. Jahrb. Bd. 12, p. 4" 529. 1887. Pansch, A. : Ueber die typisehe Anordnung der Furchen und Windungen auf den Grosshirnhemispharen des Menschen und der Affen. Arch. f. Anthropol.

Bd. 3, S. 227-257. 1868. Paton, S. : Die Histogenesis der Zellenelemente der Hirnrinde. Neurolog.

Centralbl. Bd. IS, S. 1086-10SS. 1899. The Histogenesis of the Cellular Elements of the Cerebral Cortex. Johns Hopkins Hospital Reports. Vol. 9. — Ref. in Neurolog. Centralbl. Jahrg.

20. 1900. A Study of the Neurofibrils in the Ganglion Cells of the Cortex. Journ.

Experim. Med. (N. Y.), Oct., 1900. Pfister, H. : Das Hirngewieht im Kindesalter. Arch. f. Kinderheilk. Bd. 23.

1897. Ueber die occipitale Region und das Studium der Grosshirnoberflache. Stuttgart 1S99. Zur Anthropologie des Riickenmarks. Neurol. Centralbl. Bd. 22, Nr. 16 and 17, p. 757-762 and 819-824. 1903. Teilwagungen kindlicher Gehirne. Arch. f. Kinderheilk. Bd. 37, S. 243-251.

1903. Ueber das Gewicht des Gehirns und einzelner Hirnteile beim Saugling und alteren Kinde. Neurol. Centralbl. Bd. 22, N. 12, S. 562-572. 1903. Neue Beitrage zur Kenntnis des kindlichen Hirngewichts. Arch. f. Kinderheilk. Bd. 37, S. 239-242. 1903. Popoff, N. : Recherches sur la structure des cordons posterieurs de la moelle epiniere de Fhomme. Arch, de Neurol. Nr. 50. Vol. 17, p. 177. 1889. Prekant, A.: Criteriums histologiques pour la determination de la partie per sistante du canal ependymaire primitif. Intern. Monatssehr. f. Anat. u.

Physiol. Bd. 11, p. 281-296. 1894. Reichert, C. B. : Der Bau des mensehliehen Gehirns. Bd. 2, S. 1-90. Leipzig 1861. Retzius, G. : En hittils obeaktad egendommeligned i menniskohjernans embryonela utveckling. (A hitherto unobserved peculiarity in the development of the human embryo.) Svenska Lakare-Sallskapets Fbrh. — Ref. in: Centralbl.

f. Nervenheilk. Bd. 14, S. 347. 1891. Das Menschenhirn. Studien in der makrosk. Morphol. 1896. 4. Ueber das Auftreten des sulcus centralis und der fiss. calcarina im Menschenhirn. Biol. Unters. N. F. Bd. 8, S. 59-64. 1898. Zur Frage von den sogenannten transitorischen Furchen des Menschenhirns.

Anat. Anz. Erganz. Heft. Bd. 19. 1901. Zur Frage der transitorischen Furchen des embryonalen Menschenhirns. 4 Fig. Biol. Unters. N. F. Bd. 10, S. 65-66. 1902. Zur Morphologie der insula Reilii. 3 Taf. Biol. Unters. N. F. Bd. 10.

S. 15-20. 1902. Renbold, W. : Zur Entwicklungsgeschichte des mensehliehen Gehirns. Festschr.

f. d. Univ. Wiirzburg v. d. med. Fakult. Wiirzburg. Bd. 1. S. 167 1882. Richter, A.: Ueber die Enstehung der Grosshirnwindungen. Virch. Arch. Bd.

108, S. 398-422. 1887.

Von. II.— 8

114 HUMAN EMBRYOLOGY.

Righetti, R. : Sulla mielinizzazione delle fibre della corteccia cerebrale umana nei primi mesi di vita. Riv. di Pathol, nerv. e merit. Vol. 2, p. 347-354.

1897. Robinson, A. : The Development of the Posterior Columns, of the Posterior Fissure, and of the Central Canal of the Spinal Cord. Studies in Anatomy from the Anatomical Department of Owens College. Vol. 1. Manchester 1892. Roemer, P. : Beitrag zur Auffassung des Faserverlaufs im Gehirn auf Grund des Studiunis von Kindergehirnen. 6 Fig. Diss. med. Marburg 1900. Romiti, G. : Sull' ordine di successione, con il quale appajano le scissure cerebrale.

Proc. verb. d. Soc. Tosc. d. Sc. nat. 1882. Rosenberg, E.: Ueber die Entwicklung der Wirbelsaule und das centrale carpi des Menschen. Morph. Jahrb. Bd. 1, S. 83-198. 1876. Rosenzweig, E. : Beitrage zur Kenntnis des feineren Baues der substantia Rolandi des Riiekenmarks. 2 Taf. u. 8 Fig. Journ. Psych, u. Neurol. Bd. 5, S. 49-72. Also: Berliner med. Diss. 1905. Rudinger, N. : Vorlaufige Mitteilungen iiber die Unterschiede der Gehirnwindungen nach dem Geschlecht beim Fetus und Neugeborenen. Mfinchen 1877. Ein Beitrag zur Anatomie des Sprachcentrums. Stuttgart 1882. Salvi, G. : L'istogenesi e la struttura delle meningi. Atti Soc. Tosc. di Sc. anat.

Mem. Vol. 16, p. 187-225. 1898. Sopra lo sviluppo delle meningi cerebrali. Mem. della Soc. Tosc. di Sc.

nat. Pisa 1897. Sanctis, S. de: Untersuchungen fiber den Bau und die Markscheidenbildungen des mensehlichen Kleinhirns. Monatsschr. f. Psych, u. Neurol. Bd. 4, S.

237-246 and 271-284. 1898. Ricerche intorno alia mielinizzazione del cervelletto umano. Ric. Labor.

Anat. norm. Univ. Roma. Vol. 9, Fasc. 4. 1903. Schaper, A. : Zur Morphologie des Kleinhirns. 10 Fig. Verh. d. Anat. Gesellsch.

S. 102-115. Anat. Anz. Erg.-Heft. Bd. 16. 1899. Zur Frage Existenzberechtigung der Bogenfurchen am Gehirne menschlicher Embryonen. Anat. Verhandl. S. 35-37. Anat. Anz. Erg.-Heft. Bd. 25.

Jena 1904. Schmidt, F. : Beitrage zur Entwicklungsgeschichte des Gehirns. Zeitschr. f .

wissensch. Zool. Bd. 11, S. 43-61. 1862. Schoenlein, J. L. : Von der Hirnmetamorphose. Diss. Wfirzburg 1846. Schwalbe, G. : Beitrag zur Entwicklungsgeschichte des Zwischenhirns. Jenaische Sitzungsber. S. 2-7. 1880. Sernow: Individuelle Typen der Hirnwindungen des Menschen. Moskau 1877. Siemerling : Ueber Markscheidenentwicklung des Gehirns und ihre Bedeutung f fir die Lokalisation. Jahrssitz. d. Ver. deutsch. Irrenarzte, Bonn. Ref. in : Neurol. Centralbl. Jahrg. 17. S. 961-963. 1898. Smith, G. E. : Morphology of the True Limbic Lobe, Corpus Callosum, Septum Pellucidum, and Fornix. Journ. of Anat. and Physiol. Vol. 30, p. 157-167 and 185-205. 1895. Note on the So-called " Transitory Fissures " of the Human Brain, with Special Reference to Bischoff's " Fissura Perpendicularis Externa." Anat.

Anz. Bd. 24, Nr. 8, p. 216-220. 1904. Studies in the Morphology of the Human Brain, with Special Reference to that of the Egyptians. No. 1. The Occipital Region. 2 plates. Rec.

Egypt. Gov. School Med. Vol. 2, p. 123-172. Cairo 1904. Morphology of the Occipital Region of the Cerebral Hemisphere in Man and the Apes. Anat. Anz. Bd. 24, S. 436-451. 1904. The Fossa Parieto-occipitalis. Journ. of Anat. and Physiol. Vol. 38, p. 164 to 169. 1904.

DEVELOPMENT OF THE NERVOUS SYSTEM. 115 Note on an Exceptional Human Brain Presenting a Pithecoid Abnormality of the Sylvian Region. Journ. of Anat. and Physiol. Vol.38, p. 158-161. 1904. New Studies on the Folding of the Visual Cortex and the Significance of the Occipital Sulci in the Human Brain. Journ. of Anat. and Physiol.

Vol. 41. 1907. Soemmering, S. Th.: Hirnlehre und Nervenlehre. Frankfurt a. M. 1791. Sperino, C. : Contributo alio studio dei rapporti fra lo sviluppo degli arti e quella dei centri nervosi. Giorn. R. Accad. medic. Torino. Anno 55, Nr. 2, p. 153-160. 1892. Sterzi, G. : Anatomia comparata ed all' ontogenesi delle Meningi midollari. Atti del Reale Inst. Venento di Scienze. Letters ed. Arti. Tomo 60. 1900. Gli spazii linfatizi delle meningi spinali ed il loro significato. Monitore zool. ital. Anno 12. 1901. Ricerche intorno alia anatomia comparata ed all' ontogenia delle meningi.

Considerazioni sulla filogenesi. Parte prima. Meningi midollari. 1 Taf.

Atti Instit. Veneto Sc. Lett, ed Arti. 1900-1901. Vol. 60. 1901. Strasser, H. : Ueber die Hiillen des Gehirns und des Riickenmarks, ihre Funktion und ihre Entwicklung. C. R. de 1' Assoc, des Anat. 3. Sess. p. 175-184.

Lyon 1901. Straussler, E. : Zur Morphologie des normalen und pathologischen Riickenmarks und der Pyramidenseitenstrangbahn. Jahrb. Psych, u. Neurol. Bd. 23, Heft 3, S. 260-298. 1903. Streeter, G. L. : The Cortex of the Brain in the Human Embryo During the Fourth Month, with Special Reference to the So-called " Papillae of Retzius." The Amer. Journ. of Anat. Vol. 7. 1907. Stroud, B. B. : The Mammalian Cerebellum. 1. The Development of the Cerebellum in Man and the Cat. Journ. of Compar. Neurology. Vol. 5, p. 71-118. 1895. Suchanow, S., et Czarniecki, F. : Sur l'aspect des prolongements protoplasmiques des cellules nerveuses des comes anterieures et posterieures de la moelle epiniere chez des enfants nouveau-nes. (Methode chromo-argentique.) Nouv. Iconographie de la Salpetriere. Annee 15, Nr. 6, p. 530-539. 1902. Symikgton, J. : On the Temporary Fissures of the Human Cerebral Hemispheres, with Observations on the Development of the Hippocampal Fissure and ' Hippocampal Formation. Rep. 71st Meet. Brit. Assoc. Advanc. Sc. Glasgow. 1901. Tiedemann, Fr. : Anatomie und Bildungsgeschichte des Gehirns im Fetus des Menschen nebst einer vergleichenden Darstellung des Hirnbaues in den Tieren. Nurnberg 1816. Tkatschenko, M. P. : Die Leitungsbahnen des menschlichen Kleinhirns auf Grund der Untersuchung ihrer Entwicklung. Diss. St. Petersburg 1903.

(Russian.) Tourneux, F., et Hermann, G. : Les vestiges du segment caudal de la moelle epiniere et leur role dans la formation de certaines tumeurs sacro coccygiennes. Compt. rend. T. 104. p. 1324-1326. 1SS7. Sur la persistance de vestiges medullaires coccygiens pendant toute la periode foetale chez l'homme et sur le role de ces vestiges dans la production tumeurs sacrococeygiennes congenitales. Journ. de l'Anat. et de la PhyeioL p. 498-529. 1887. Tourneux, F. : Sur la structure et sur le developpenient du fil terminal de la moelle chez l'homme. C. R. hebd. de la Soc. de Biol. Ser. 9. T. t, p. 340-343. 1892. Tourneux, F., et Soulie, A.: Sur les premiers developpements de la pituitaire chez l'homme. C. R. Soc. Biol. Paris. Nr. 29, p. '7. L898.

Trepinski: Die embryonalen Fasersysteme in den Hinterstriingen und ihre Degeneration bei der tabes dorsalis. Arch. f. Psych iai rie. Bd. 30, S. 54-S1. 1S97.

116 HUMAN EMBRYOLOGY.

Unger, E., und Brugsch, Th. : Zur Kenntnis der fovea und fistula sacrococcygea s. caudalis und der Entwicklung des lig. caudale beim Menschen. Arch. f.

mikr. Anat. Bd. 61, S. 151-219. 1903. Valenti, G. : Sur le developpement des prolongements de la pie-mere dans les scissures cerebrales. A. ital. de Biol. T. 20. 1894. Valenza, G. B. : Sur une disposition partieuliere en peleton des tubes nerveux dans la moelle de Pembryon bumain. Compt. rend. Soe. de Biol. T. 27.

III. Part. 1897. Vanhersecke, G. B. A. : La morphologie des eirconvolutions cerebrales. Origine, developpement, valeur morphologique et medicale des plis corticaux du cerveau. These de Lille. 1891. Vignal, W. : Sur le developpement des elements de la moelle des mammif eres.

Arch, de la Physiol. Nr. 7, p. 117-237.— C. R. T. 99. Nr. 9. 1884. Reeherches sur le developpement de la substance corticale du cerveau et du cervelet chez l'homme et les mammif eres. Arch, de la Physiol. Ser. 4.

T. 2, p. 228-254 and 311-335. 1888. Developpement des elements du systeme nerveux cerebrospinal etc. Paris 1889. Vogt, C. : Etude sur la myelinisation des hemispheres cerebraux. These de Doctorat en Med. Paris 1900. Vogt, C, und Vogt, 0.: Die Markreifung des Kindergehirns wahrend der ersten 4 Lebensmonate und ihre methodologische Bedeutung. (1. Mitteil.) Atlas.

T. 1. 124 Lichtdrucktafeln. — Denkschr. med. naturw. Ges. Jena. Bd. 9.

Lief. 2. Bd. 12. — Also: Neurobiologische Arbeiten. 1904. Vogt, H. : Ueber die Entwicklung des Cerebellums. Journ. Psych, u. Neurol.

Bd. 5. 1905. Volz, R. : Das foramen " interventriculare." Laupp, Tubingen 1907. Vulpius, O. : Ueber die Entwicklung und Ausbreitung der Tangentialf asern in der m;enschlichen Grosshirnrinde wahrend verschiedener Altersperioden.

Arch. f. Psych. Bd. 23, S. 775-798. 1892. Waldeyer, W. : Hirnf urchen und Hirnwindungen. Ergebnisse d. Anat. u. Ent wicklungsgeseh. Bd. 5 (lit. to 1896), S. 146-193. 1896. Hirnwindungen. Nachtrag zu dem Referat iiber Hirnwindungen in Bd. 5 der Ergebnisse. Ergebn. d. Anat. u. Entwieklungsgesch. Bd. 6, S. 171-183. 1897. Hirnfurchen und Hirnwindungen. Hirnkommissuren. JJirngewicht. Ergebn.

d. Anat. u. Entwieklungsgesch. Bd. 8, S. 362-401. 1899. Weber, L. W. : Der heutige Stand der Neurogliaf rage. Zusammenf assendes Referat. Centralbl. f. allgem. Path. u. path. Anat. Bd. 14, Nr. 1, S. 7-33.

(Literature.) 1903. Wenzel, J. et K. : Prodromus eines Werks iiber das Gehirn des Menschen und der Siiugetiere. Tiibingen 1809. De penitiori structura cerebri hominis et brutorum. Tubingae 1812. Wolpin, L. : Gewichtsbestimmungen iiber das Wachstum des Gehirns bei Kindern.

Diss. St. Petersburg 1903. (Russian.) Zander, R. : Beitrage zur Morphologie der dura mater und zur Knochenentwick lung. 2 Taf. Festschr. zum 70. Geburtstag von C. v. Kupffer. Jena 1899. Ziehen, Th. : Die Morphogenie des Centralnervensystems der Siiugetiere. In : O. Hertwig's Handbuch. 1906. Die Histogenese von Hirn und Riickenmark. Entwicklung der Leitungsbahnen und der Nervenkerne bei den Wirbeltieren. In: O. Hertwig's Handbuch.

1906. Zuckerkandl, E. : Zur Entwicklung des Balkens und des Gewolbes. Mit 3 Tafeln.

Sitz.-Ber. k. Akad. Wissensch. Bd. 110. Wien 1901. Ueber die Affenspalte und das operculum occipitale des menschlichen Gehirns.

14 Fig. Arb. a. d. neurol. Inst. d. Wien. Univ. Bd. 12, S. 207-242. 1905. Zur Entwicklung des Balkens. Arbeiten aus Neurolog. Inst, an der Wiener Univ. Bd. 17. 1909.

DEVELOPMENT OF THE NERVOUS SYSTEM.

117

III. PERIPHERAL NERVOUS SYSTEM.

SPINAL NERVES.

The ventral roots of the spinal nerves are derived from the mantle layer of the neural tube, as has been previously referred to. Processes from the neuroblasts situated in the mantle layer are assembled into rootlets which emerge in a continuous longitudinal series along the ventrolateral border of the tube. Outside of the tube these rootlets are grouped into segmental bundles, and after being joined by the fibres of the dorsal roots they constitute complete segmental nerves. The direction taken by the ventral root fibres on emerging from the tube varies according to the size and position of the ganglion crest. In the human embryo it is almost

X-X/ Ganglionic crest

XI Rootlets

Bridge

Fig. 82. — Reconstruction of a portion of the peripheral nerves of a human embryo 4 mm. long'. (Hertwig collection, No. 137). Enlarged 22 : 1. Ot. v., ear vesicle.

directly lateral. In some sections of the 5.5 mm. pig in the author's possession the ventral roots extend dorsalward through the mesenchyma at an angle of 30° to reach the ventral border of the ganglion mass.

The dorsal roots are derived entirely from the ganglion crest. This structure can be seen in the 4 mm. embryo, Fig. 82, as a flattened cellular band which extends caudalward from the auditory vesicle along the lateral border of the neural tube to its extreme tip. As will be referred to later, the ganglion crest of the hindbrain is similar to and by some described as continuous with that of the spinal cord (p. 139). That part of the crest which corresponds to the spinal cord is characterized at this time by segmental incisures along its ventral border. The dorsal border of the crest remains intact until the appearance of the dorsal rootlets, in the meantime constituting a cellular bridge connecting the more ventral ganglionic clumps. In embryos at the end of the fourth

US

HOIAN EMBRYOLOGY.

week _ B fibrous processes can be seen cropping out from the dorsal border of the crest and attaching themselves to the spinal cord. They appear first in the cervical region and somewhat later can be seen in the more caudal part of the crest. These are the primitive dorsal rootlets. They enter the marginal zone of the tnbe and eventually form a longitudinal bundle corresponding to the dorsal funiculus of the adult cord. Peripherally they

FK. 83.— a

of the peripheral nerves of a hrrman embryo 6.9 mm. long (His collection. Br. 3). rged 16.7 : 1.

can be traced back to cell clusters in the crest, the processes from ^ral cells uniting in a common fibrous strand. With the formation of these rootlets there is a gradual disappearance of the dorsal bridge, and there is thereby produced a complete segmentation of the ganglion crest. The ventral end of the ganglia extend forward and end diffusely among the fibres of the ventral roots. The cells are in a state of active differentiation, and the developing fibrous pro' can be seen joining the more precocious ventral roo

DEVELOPMENT OF THE NERVOUS SYSTEM.

119

In embryos 9 to 10 mm. long (Fig. 84) the differentiation of the ganglion-cells and their fibrous processes has advanced to a point where the chief parts of a typical spinal nerve may be recognized. There is the dorsal root and its sharply outlined ganglion and a well-defined ventral root joins it, the two together forming the nerve-trunk. At the same time that the dorsal and ventral roots unite to form the main trunk, they both give off lateral fibres which form the dorsal branch, the so-called posterior primary division, which breaks up among the cells that are to form the long muscles of the back, supplying these and extending through to reach the integument. The remainder of the nervetrunk is continued forward as the ventral branch, or anterior primary division. From its median side there is given off the ramus

Ganglion spinale

Dorsal muscles

Ventrolateral muscles

Arcus vertebralis \ Ramus posterior n. spinalis v^Y. Art. thorlcica Proc. costal is Ramus terminalis "\ lateralis Ramus terminalis \ anterior

Fio. 84. — Diagrammatic transverse section through a thoracic segment of a 9 mm. human embryo.

Enlarged 25 : 1. (After Bardeen and Lewis, 1901.)

communicans, which extends toward the aorta and ends in the sympathetic ganglion cord. The main trunk terminates in two branches, the anterior and lateral terminal branches, from which arise the anterior and lateral cutaneous branches of the adult, and which in the thoracic and abdominal regions give off branches to the musculature of the front and lateral body wall. The relation of these branches to the individual muscles is shown at a later stage in Fig. 85. In the same figure can be seen the loop-formation in the intercostal space that occurs before the bifurcation ot' the trunk into the lateral and anterior terminal branches is completed.

Throughout the spinal region there is a tendency for the ad cent nerve-trunks to unite at the place where the lateral term; branches arise, and there is formed thereby a ieri inters mental loops. This loop- or plexus-formation may involve either

120 the lateral or the anterior terminal branches, or both. Its degree of development depends upon the complexity of the parts supplied.

Arcus vertebralis Mm. dorsales

Radix posterior ; Ramus" posted or

medial is

^--R. lateralis

intercost. ext.

M. in tercost . v int.

Fig. 85. — Diagrammatic transverse section through a thoracic segment of a 17 mm. human embryo (Huber collection, No. 14), showing a typical thoracic nerve. Enlarged 15 : 1.

In three regions it is particularly marked, and there are thus produced the cervical, brachial, and lumbosacral plexuses.

Cervical Plexus.

In the cervical region the anterior and lateral terminal branches form two separate plexuses; the former produces the deep cervical plexus, and the latter the superficial cervical plexus. The superficial cervical plexus consists of the union of the lateral terminal branches into loops, from which are given off the cutaneous branches to the auricular, cervical, and occipital regions. The deep plexus results in the formation of the ansa hypoglossi and the phrenic nerve. The former is produced by the fusion of the second and third cervical nerves into the descendens cervicis, which unites in a loop with the hypoglossal, together with which the first cervical has been incorporated above. From this loop are given off the short branches which end among the cells that are to form the hyoid musculature.

The phrenic nerve is formed by anterior terminal branches principally from the fourth and fifth cervical nerves. A contribution on the part of^the sixth and third nerves may occur. This

\

DEVELOPMENT OF THE NERVOUS SYSTEM.

121

iien ~ can be seen through the transparent arm in Fig. 8G. Owin^ to the position of the diaphragm at this time the course of the nerve is almost directly ventral. Later, as pointed out by His and Mall (Mall, 1901), the points of origin and insertion of the nerve

Gang, acusticum N VI Gang, semilunare n. trigemini Cerebellum

Yesicula auditiva Gang, radicis n. IX ! Gang, petrosum

Gang, radicis n. vagi

Gang. Froriep

,-^N. hypoglossus

Gang. cerv. I

=-N. XI "Gang, nodosum N.desc. cerv. "(.ansa hypoglosei) Rami hyoid.

N. musculocutan. \' . axillaris *N. phrenicus N". medianus N. radialis -F. ulnaris

-Gang. thor. I

N. tibialis N. peronaus Tubus digest. ' Gang. sacr. I

R. posterior ! R. terminalis lateralis i i \ R. terminalis anterior N. femoral. ! | ! Gang. lumb. I N. obturator. Mesonephros Nn. ilio-ing. et hypogastr.

Fig. 86. — Reconstruction of the peripheral nervous system of a human embryo 10 mm. long (Huber collection, No. 3). The arm and leg are represented as transparent masses, into the substance of which the branches of the brachial and lumbosacral plexuses may be followed. Enlarged 12 : 1.

draw gradually apart, due on the one hand to the descent of the diaphragm and the lengthening of the thoracic cavity, and on the other hand to the subsequent elevation of the cervical nerves which accompanies the development of the structures of the neck. It is thus that there results the long caudal course of this nerve that is characteristic of the adult.

122 HUMAN EMBRYOLOGY.

Nerves of the Arm and Leg.

In the arm and leg we meet with special conditions which will be better understood if we first refer to two essential factors that were established by the recent experimental work on amphibian larvae by Harrison (1907). In the first place, he has shown that the nerves which take part in the innervation of a limb are determined by the position and width of the limb bud. A limb bud transplanted to some other part of the body acquires a complete system of nerves, supplied by the region in which the limb is implanted. In the second place, the distribution of the nerves within a limb is determined by its own component structures. The segregation of the developing structures within the limb has a directive action upon the growing nerve-fibres, and determines their grouping into definite characteristic bundles. Even foreign nerves entering a transplanted limb bud are likewise controlled so that they form intrinsic nerves to the limb and assume normal terminal ramifications. These two factors are to be kept in mind in interpreting the normal embryology of these nerves. Our knowledge concerning the details in the development of the nerves of the arm in the human embryo is based principally upon the work of Lewis (W. H., 1902) and of the leg on the work of Bardeen (1907). Large use has been made of their papers in the following description.

When the limb buds first form they consist, to all appearances, of homogeneous mesenchyma and contain no nerves. Very soon, however, opposite the base of each limb bud, presumably stimulated by its presence, the anterior primary divisions of the spinal nerves undergo an exuberant growth and form a solid sheet or plexus of fibres extending toward the base of the limb. In embryos a few days older, coincident with the condensation of the skeletal core, branches from this nerve-plexus can be seen advancing into the limb bud and entering the areas of premuscle tissue that in the mean time have formed a sheath around the skeletal core. The premuscle sheath is not evenly distributed, but from the very start is arranged in the form of muscle groups, and it is between these groups through the intermuscular spaces that the nerves make their way. As the differentiation of the limb continues the nerve trunks extend distalward in the limb and give off muscle branches which enter the muscle groups and supply the individual muscle anlages. The site of nerve entry into a muscle is constant. It is situated near the centre of the anlage on the side toward the main trunk. This point is the seat of the earliest differentiation of the muscle, and according to the Nussbaum law muscle growth extends from here in the direction of the intramuscular nerve branching. Though nerve-fibres and muscle groups seem to make their appear

DEVELOPMENT OF THE NERVOUS SYSTEM.

123

ance simultaneously, and even at times the nerves seem to precede the muscles, yet it must be remembered that the experimental evidence clearly shows that the situation and branching of the nerves are entirely dependent on the structural segregation of the developing muscles and skeleton, and this is why the main nerve-trunks are developed in paths situated in the intermuscular areas, and likewise why the larger nerve branches are in the intramuscular septa of individual muscles. As we do not have a metameric distribution of the muscles of a limb, we consequently do not have a true metameric distribution of the nerves.

,v

Fig. 87. — Diagram illustrating the influence of the direction of the fibre bundles in different muscle types upon the character of their nerve supply. (After Bardeen, 1907.) Within the muscle the course of the chief branches is determined by the direction of the fibre bundles, whether they run parallel or are transverse to the main trunk from which the nerve arises. When the fibre bundles of a muscle are transverse to the main nerve-trunk, its nerve and chief branches pass across the fibre bundles midway between the points of attachment, giving off branches on each side which go to form the intramuscular nerve-plexus. When the fibre bundles of a muscle run parallel with the main nerve-trunk, the branches to this muscle, as a rule, enter the proximal third of the muscle belly and extend distally parallel with the muscular fibres, giving off at the same time the branches to the intramuscular plexus. This relation between the

124 HUMAN EMBRYOLOGY.

direction of the muscle-fibres and the course of the supplying nerve is shown for different muscle types in Fig. 87. A further influence on the course of nerves is exerted by the migration of the muscles which they supply. The muscle masses, having received their nerves at an early stage, may by subsequent migration draw the nerves a long way out of their original course. This is illustrated by the latissimus dorsi, the trapezius, the diaphragm, and the muscles of the tongue.

The arm bud develops somewhat in advance of the leg bud. In the 4.5 mm. embryo there is a well-defined arm bud, consisting of an apparently homogeneous mesenchyma and having as yet no nerves. A leg bud in a similar stage of development is not met with until we come to embryos about 7 mm. long. The base of the arm bud is usually situated opposite the lower four cervical and first thoracic vertebrae, and the leg bud opposite the five lumbar and first sacral. There is some variation in the position of the limb buds relative to the spinal axis at the time of entrance of the spinal nerves. Consequently there may be a variation of as much as three segments in the origin of the spinal nerves which eventually enter a particular limb, the more cephalic nerves supplying the limb when the limb bud has a more cephalic position and vice versa.

In the 9 mm. embryo the central mesenchyma of the leg bud is condensed into sclerogenous tissue corresponding to the hip-bone and proximal part of the femur. This sclerogenous tissue divides the bundles of nerve-fibres streaming into the leg into the main nerve-trunks. The lumbosacral plexus in an embryo of about this age is shown in Fig. 86. The nerves forming it unite into a flattened mass or sheet of fibres which enters into the base of the leg bud, the division into anterior and lateral terminal branches being lost in the formation of the plexus. The further course of the fibres is determined by the framework of the leg. Owing to the cell masses of the bony pelvis and the femur the fibres become grouped into four bundles arranged in two pairs, each consisting of a median and lateral trunk. Of the upper pair the median trunk corresponds to the n. obturator, and the lateral to the n. femoralis. The lower pair represent the n. sciaticus, the median bundle constituting the future n. tibialis, and the lateral the n. peronseus communis. In the 11 mm. embryo, as shown in Fig. 88, the leg is differentiated externally into foot-plate, cms, and thigh, and internally, surrounding the skeletal core, a distinct myogenous zone can be recognized, consisting of muscle groups with intervening intermuscular spaces. The main nerve-trunks have grown well down into the limb between the muscle groups, and the chief muscular and cutaneous branches can be seen. In Fig. 89 is represented a median view of the leg and the adjoining part of the

DEVELOPMENT OF THE NERVOUS SYSTEM.

125

trunk C)f an embryo 20 mm. long, showing the relations of the thoracic, lumbar, and sacral nerves to the abdominal musculature and the skeleton of the leg. Both muscular and sensory branches are shown, and it will be seen that the nerve supply of the leg at this time must be considered as essentially complete.

The nerves of the arm in a 9 mm. embryo are shown in Fig. 90. The brachial plexus consists of a continuous sheet of fibres which on reaching the developing humerus is split into a dorsal N. sp. ext. N. il. ing. N. 1. ing- N. il hypog.

Mm. add Mm. fern. post. Urachus Mm. cr. post. sup. Mm. cr. post. prof. Foot plate

N. cut. lat. N. cut. ant N. sart. M.

N. thor. 11. M. rect. abd.

Cloaca

N. bi. prox. port.'et semit.

Isch.

N. cocc.

M. obt. int. N. pud. 1 N. cut. fern. post.

M. quadr. fern.

Fig. 88. — Nerves of the leg in an embryo 11 mm. long, age about five weeks. Enlarged 17 : 1. The principal muscle anlages are shown in lighter color. Isch., os ischii; Mm. add., musculi adductores; Mm. cr. post, sup., musculi cruris post, superficiales; Mm. cr. post, prof ., musculi cruris post, profundi; Mm. fern, post., musculi femorales posteriores; M. obt. int., muse, obturator int.; M. quadr. fern., muse, quadratus femoris; M . rect. abd., muse, rectus abdominis; N. bi. prox. port, et semit., n. proximalis portionis muse, bicipitis et muse, semitendinosi; N. cocc., n. coccygeus; N. cut. ant., n. cutaneus femoris anterior; N . cut. fern, post., n. cutaneus femoris posterior; N. cut. lot., n. cutaneus femoris lateralis; N.il.hyp., n. iliohypogastricus; N. il. ing., n. ilio-inguinalis; iV. I. ing., n. lumbo-inguinalis; N. pud., n. pudendus; N. saph., n. saphenus; N. sart., n. muse, sartorii; N. sp. ext., n. spermaticus ext.; N. thor. 11, n. thoracalis 11; Pu„ os pubis. (After Bardeen, 1907.) and ventral division, the former corresponding to the posterior cord and the latter to the outer and inner cords. These cords are immediately broken up into the large branches which pass down in the intermuscular spaces, where the fibres abruptly fray out to enter the premuscle masses. The formation and branches of the brachial plexus, as seen in the 10 mm. embryo, are shown in Fig. 86. The brachial plexus is split by the skeletal anlage into two laminae from which the various nerves arise. From the anterior or ventral lamina arise the n. musculocutaneus, n. medianus, and n. ulnaris, and from the posterior or dorsal lamina the n. axillaris

126 and n. radialis. As compared with the lumbosacral plexus in the same embryo it is considerably in advance. It is not until

N. thor. 6

N. dig. plant. II-I1I \

Nn. il. ing. et^sp. ext Nn. cut. ant, N. quadr. fern

N. popl. N. flex. dig. 1. 'N. flex. hal. I.,

N. plant. med N. plant. lat. % N.abd. hal. N. flex, hal.br. \

Nn. vise.

Nn. pud.

N.'cocc. Nn. semit. et bi, prox. port. Nn. add. m. semim., semit. et'.bi. cap. 1. N. bi. cap. br. N. gastroc.

R. prof.

N. sural. N. quadr. pi. Nn. sol.

Fig. 89. — Nerves of the leg and adjacent abdominal wall in an embryo 20 mm. long.'age about' seven weeks. Enlarged 10 : 1. (After Bardeen, 1907.) For the abbreviations see Fig. 88.

later (20 mm.), secondary to the caudal migration of the arm, that we meet with a decided posterior inclination of the brachial plexus ; but, aside from this and aside from the proportionate large size

DEVELOPMENT OF THE NERVOUS SYSTEM.

127

of the nerves as compared with other structures, there is little in their gross morphology to distinguish the nerves of the arm at the

Spinal gangl.

-Sympatliicus I . i A '" i

Disc, interv.

Fig. 90. — Reconstruction of the nerves and skeleton of the arm in a human embryo 9 mm. long. (After W. H. Lewis, 1902.) end of the first fetal month from those of the adult. In the following table is given the origin of the fibres of the larger nerves of the arm as traced by Lewis (1902) in an embryo of this age:

Cervical.

Thoracic.

N. suprascapularis IV, (VI) N. subscapularis V, VI, VII N. thoracalis longus VII, VLTl Nn. thoracales anteriores V, VI, VII, VIII N. musculocutaneus V, VI, (VII) N. medianus V, VI, VII, VIII N. axillaris V, VI, VI I N. radialis V, VI, VII, VIII N. ulnaris (VI), VII, VIII

According to Bardeen (1907) the cutaneous nerves first approach the superficial fascia along the line corresponding to the primary margin of the limb bud, and from these areas send branches of distribution over the dorsal and ventral surfaces of the developing limb, as shown in Fig. 91. There exi-N a considerable variation in the extent of distribution of these branches to the skin. The extensive development of one nerve tends to retard the growth of the neighboring nerves, and diminished development

128 stimulates them to more active growth. A further source of variation, which is equally true of motor nerves, is that any two nerves that arise in succession — such as the twelfth thoracic and hypogastric, the hypogastric and inguinal, or the lateral cutaneous

T12

iliohypogastric, ilio-inguinal. sperm, ext. lumbo-inguinal.

obturator

plantar med.

plantar lat.

1^ "" --^r\ sural j ^f ^prz^zrr.... cut. fern. post.

-- - rami perineal.

r n. dors. pen.

nn. perinei n. hsemorrh. inf.

A

iliohypogastric.

ilio-inguinal.

sperm, ext.

lumbo-inguinal.

cut. fem. lat.

cut. fem. ant.

ram. cut. med. (n. obturat.) saphenus peronaeus prof, peronseus superf .

cut. surse lat. cut. fem. post, rami perineal, (n. cut. fem. post.)

Fig. 91. — Diagram of the cutaneous nerves of the lumbosacral plexus, showing the superficial areas along the margin of the limb bud which are first reached by the tips of the growing nerves. Extending from these foci branches spread eventually over the dorsal and ventral surfaces of the limb. A, ventral group; B, dorsal group. (After Bardeen, 1907.) and femoral— may be combined into a single trunk for a greater or lesser part of their course. On the other hand, two or more nerve-trunks may carry fibres ordinarily assembled in a single trunk, such as extra iliac and genital branches or an accessory obturator.

CEREBRAL NERVES.

For purposes of description the nerves of the head will be grouped according to their function, following as far as possible the functional systems of Gaskell ; i.e., the activities of the organism are separated into somatic and visceral, in each of which there

DEVELOPMENT OF THE NERVOUS* SYSTEM. 129 is the double activity on the part of the nervous system, motor and sensory, making in all four primary functional divisions. Some of the cranial nerves consist of elements belonging exclusively to one functional division, — for example, the n. abducens, which consists entirely of somatic motor fibres, — while others are complex nerves containing elements of more than one system, such as the n. vagus, which contains elements from three functional divisions, somatic sensory, visceral sensory, and visceral motor. The nerves will therefore be grouped according to the predominance of their functional elements as follows: Somatic sensory. Somatic motor. Visceral (motor and sensory).

Olfactory Oculomotor Trigeminal Optic Trochlear Facial Acoustic Abducens Glossopharyngeal Hypoglossal Vagus and accessory In general the basal plate of the neural tube is motor and the lateral or alar plate is sensory; thus, the somatic motor group arises entirely from the basal plate, while the visceral group is connected in larger part (sensory) with the alar plate and in lesser part (motor) with the basal plate. The nerves included under the somatic sensory group are all specialized nerves and have individual processes or lobes of the nervous system devoted to them, — i.e., the olfactory bulb, the eye bulb and stalk, and the tuberculum acusticum.

Nerves of Special Sense Organs. (Special Somatic Sensory. ) These nerves belong to the group of afferent nerves which connect the integument with the central nervous system, and the union of nerve and integument has resulted in the formation of special sense organs, — that is, the olfactory organ, the eye. the ear, and the lateral line system, composed partly of nerve elements and partly of integument. This nerve group is shown in Fig. 92, which represents schematically a typical vertebrate &ead in which the mtegumental part of the special sense organs is shown in red. The nose, eye, and ear are the same as seen in man. The lateral line system, however, is absent in man except for a temporary trace which may be seen in 7 mm. embryos in the form of areas of thickened epidermis situated over the second, third, and fourth branchial arches, probably representing the lateral line ganglia incorporated with the seventh, ninth, and tenth cranial nerves of lower- vertebrates. By comparing the nerve portions of these organs in Fig. 92 it can be seen that the olfactory nerves, the retinal ganglion-cells, and the acoustic nerve, though differing so widely in their adult morphology, must all be considered as analogous structures.

The nn. olfactorii and the n. opticus will be described with the special sense organs to which they belong.

Vol. II.— 9

130 The n. acusticus develops from a small group of ganglion-cells which can be recognized by the end of the third week, lying closely against the cephalic border of the ear vesicle. Nerve-fibres arise from the proximal pole of the mass connecting it with that part of the neural tube which is to form the tuberculum acusticum, and from the distal pole connecting it with the auditory vesicle. Concerning the origin of these ganglion-cells there still remains some uncertainty. They are evidently not derived from the neural crest; but whether they migrate out from the brain wall or the walls of the developing ear vesicle, or are derived from the ectoderm immediately adjacent to the auditory pit, remains to be determined.

[in.systemat. li n. lateralis (VII) LobuS lin. lateralis Tube re. acusticum

Vesicula aud. et \i. acu st icu:

|M.f acial is

Fig. 92. — Diagrammatic vertebrate head showing the special sense organs that are formed by the union of nerve and integument, the in tegumental portion being shown in red.

The successive* stages in the development of the acoustic ganglion mass and its branches are represented in Fig. 93, which shows its appearance in embryos 4, 7, 9, 20, and 30 mm. long. This should be compared also with Fig. 86. The first step consists in the elongation of the ganglion and partial subdivision into a pars superior and pars inferior, each of which develops its own separate group of peripheral nerve branches. The pars superior forms the ganglion of the nerves to the utricle and the ampullar of the superior and lateral semicircular canals. The pars inferior forms the ganglion of the nerves to the saccule and the ampulla of the posterior canal. In addition there is derived from the pars inferior a cell mass that becomes differentiated into the ganglion spirale. This makes its appearance in embryos about 9 mm. long, where it can be seen that some of the ganglion-cells on the ventral border of the pars inferior have become massed together

DEVELOPMENT OF THE NERVOUS SYSTEM.

131

-n ve stib -•

'4mm 7mm

I pars sup

-pars inf

pars Sup pars inf., "

i

^sS-n vestib

-9mm

.-/-:^ji '~ri COch 'vji&£- - -qanc) spirale

pars sup -pars mf

,camp sup. ,-r amp. lat . r utn c r sacc

.y.- .••.•.•;.••:'•:••.'•.:•.•/ x M°"H 4&":. : '•':•': ??' _ -> spirale^-.

' ; *?£.;.V->ii - - - - "

qanq

r. amp sup.,

Fia. 93. — Development of the left ganglion acusticum and its nerve branches. The vestibular portion is shown by fine dots and the cochlear (ganglion spirale) by large dots. In the 9, 20 and 30 mm. stages a median view is shown on the left and a lateral view on the right.

132 HUMAN EMBRYOLOGY.

to form the anlage of this ganglion; in other words, the ganglion acusticum at this stage consists of an upper division entirely vestibular and a lower division partly vestibular and partly cochlear. The vestibular part of the pars inferior constitutes the so-called Zwi sell en ganglion of His, jun.

As the differentiation proceeds and the fibres elongate, the group of cells forming the ganglion spirale becomes separated from the parent ganglion mass, and eventually assumes the spiral form of the adult. The pars superior and pars inferior usually become completely separated, accompanying the subdivision of the vestibular nerve. The embryonic connection between the two may, however, persist. Likewise the path of separation between the pars inferior and the ganglion spirale may be more or less completely bridged over in the adult by a persistent connecting chain of ganglion-cells.

It will be seen that the n. cochlearis is made up entirely of fibres derived from the ganglion spirale, while the n. vestibularis consists of two portions derived respectively from the pars superior and pars inferior of the ganglion vestibulare. Owing to the contiguity of the pars inferior and the n. cochlearis, they become closely united by the developing mesenchymal elements, and this gives rise to the misleading appearance in the adult of the saccule and posterior ampulla being supplied by the cochlear nerve. The vestibular and cochlear divisions of the acoustic complex present the following contrasting characteristics : The ganglia belonging to the vestibular division develop midway along the trunk of the nerve; the ganglion of the cochlear division is situated at the extreme distal end of the nerve and lies closely against the cochlear duct, being later incorporated with it in the cartilaginous capsule, and hence the cochlear terminal branches are short and form a continuous fringe of fibres, while the vestibular terminal branches form discrete nerves of some length; the main trunk of the cochlear division is characterized by the compactness of its fibres and their spiral arrangement, which is already apparent in the 30 mm. embryo, while the fibres of the vestibular division are less compactly bundled and do not have a spiral character. In these particulars the vestibular nerve is the more primitive of the two, and the cochlear nerve must be considered as a portion of the former that has undergone special development.

Somatic Motor Group.

This group consists of the hypoglossal and the three nerves to the extrinsic eye muscles (nn. oculomotorius, trochlearis, and abducens), and in common they all arise from the basal plate and maintain their position near the median line directly beneath the

DEVELOPMENT OF THE NERVOUS SYSTEM. 133 floor of the ventricle. Their nuclei of origin are considered as a cephalic continuation of the ventral motor column of the spinal cord. The series is shown in red in Fig. 100.

The n. liypoglossus, as it appears in the 4 mm. embryo, is shown in Fig. 82. Its rootlets arise from the basal plate of the neural tube in three or four segmental groups in a longitudinal series directly continuous with the ventral roots of the cervical nerves. During the fourth week they grow forward and fuse in a common trunk, which by the end of the first month has made its way around the lateral border of the ganglion nodosum, and there breaks up in its terminal branches in the anlage of the tongue. It is now generally considered that the hypoglossal is a composite nerve made up of the ventral roots of three or four segmental spinal nerves which in the course of phylogenesis have become inclosed by the bony cranium (occipitospinal nerves). This view is supported by the identity in appearance that exists in the earliest stages (Fig. 82) between the hypoglossal rootlets and the ventral roots of the spinal nerves. Furthermore the nucleus of origin of the hypoglossal forms a continuous column with the ventral horn of the spinal cord, as seen in Fig. 100. The fact that there are no dorsal roots and ganglia as in the other spinal nerves is explained on the ground of retrogression of the sensory part of these nerves, involving especially their more cephalic rootlets. Occasionally in the embryo a ganglion, and at times also a dorsal root, is found in connection with the more caudal rootlets of the nerve (Froriep's ganglion), as in Fig. 94. In other cases, where the sensory retrogression is more extreme, there not only is no ganglion of the hypoglossal, but the ganglion of the first cervical is also partially or wholly absent.

In the course of its development the hypoglossal unites with the cervical nerves in the formation of the ansa hypoglossi. The steps in the formation of this plexus may be seen in Figs. 82, 83, and 94. The fibres of the hypoglossal and the upper three cervical nerves start out perpendicularly from the neural tube, and due to the curve of the latter they are brought together at a common focus, like spokes in a wheel, and enter together- the muscle mi Froriep's Schulterzungenstrang, that is to form the tongue and hyoid muscles. With the formation of nerve sheathe adjoining fibres become bound together, so that, as the individual muscles take form and draw apart, the nerves are separated out in the plexus that is characteristic of the nerves supplying these mus< in the adult. As seen in Fig. 86, the essential features of ansa hypoglossi are completed in the 10 mm. embryo. The f cervical unites with the trunk of the hypoglossal, and lower down leaves it as the descendens hypoglossi, carrying along a var] number of hypoglossal fibres. The descendens hypoglossi on

134 with the descendens cervicis, a fused branch from the second and third cervical nerves, and thereby forms a loop, the ansa hypoglossi, from which are .given off branches to the hyoid musculature. The n. oculomotorius arises from a group of neuroblasts situated in the ventral part of the mantle layer of the mesencephalon. These neuroblasts converge to form small rootlets, which emerge on the ventral surface of the neural tube in the

Vagus with gang, jugulare Root ganglia of the n. accessorius

Gang, jugul. of. n. IX'

' Froriep's ganglion

N. X

Gang, nodos N. laryng. sup

N. XII with r. descend.

N. phi

Sympathicua

N. vagus

Fig.' 94. — Reconstruction of the peripheral nerves on the left side of a human embryo 14 mm. long (Mall collection, No. 144). Enlarged 16.7 : 1.

concavity of the cephalic flexure. Here they unite, as shown in Figs. 10, 83, and 86, in a common trunk, which passes ventralward to the region median to the first and second divisions of the trigeminal nerve, where it breaks up in the cellular mass that is to form the eye muscles. It eventually supplies a root to the ciliary ganglion. There is no sensory ganglion in connection with the oculomotor in the human embryo. In the chick and torpedo, however, large bipolar cells have been described as migrating along

DEVELOPMENT OF THE NERVOUS SYSTEM.

135

its trunk to take part in the formation of the ciliary ganglion (Froriep, 1902, Carpenter, 1906).

The ii. trochlearis arises from a cluster of neuroblasts similar and lying just caudal to those of the oculomotor. The rootlets derived from them, instead of emerging directly ventralward, curve dorsalward over the roof of the aqueduct, where they decussate and emerge as a slender trunk which passes ventralward to reach the anlage of the superior oblique muscle. No satisfactory explanation has ever been given for the peculiar dorsal decussation of this nerve.

Fig. 95. — Composite sagittal section through a human embryo 10 mm. long, the same shown in Fig. 86, and representing the rhombic grooves and their nerve connection with the branchial arches. The nerve arising from groove d is the n. abducens, which passes median to the ganglion semilunare.

The n. abducens arises from a group of neuroblasts in the median part of the mantle zone directly beneath the fourth rhombic groove or neuromere (Figs. 95 and 100). These neuroblasts converge and form rootlets which pass through the marginal zone and emerge on the ventral surface of the neural tube. On emerging, as can be seen in the 10 mm. embryo, they are gathered together in a single trunk, which immediately bends forward at an angle of 90°, and passes forward mesial to the semilunar ganglion to reach the anlage of the external rectus muscle. It has been shown by Bremer (1908) and Elze (1907) that it is not uncommon for this

136 HUMAN EMBRYOLOGY.

Derve to have a series of multiple rootlets, with a corresponding caudal prolongation of its nucleus of origin backward into the region of the fifth rhombic groove. On the other hand, it has been further shown by Bremer (1908) that the nucleus of origin of the n. hypoglossus may extend forward and more or less completely bridge in the gap existing between it and the n. abducens. The root fibres arising in such cases from the extreme cephalic end of the hypoglossal nucleus and from the extreme caudal end of the abducens nucleus, instead of joining with their respective nervetrunks, show a tendency to form "aberrant roots," which pass out in various directions and are lost in the loose mesenchyme. These aberrant roots eventually (during the second month) entirely disappear.

Visceral Group.

The facial, glossopharyngeal, and vagus form a series of similar nerves which consist almost wholly of visceral fibres. Their motor visceral fibres arise from a column of neuroblasts (nucleus ambiguus and nucleus facialis) continuous with the lateral horn cells of the cord. Their sensory visceral fibres arise from the peripheral ganglia and enter the alar plate of the neural tube and form a longitudinal strand which in the adult we know as the tractus solitarius. In addition to these visceral fibres there are a few somatic sensory fibres for the supply of the integument of the adjoining region, which arise and have a course similar to the visceral sensory fibres. In aquatic vertebrates there are also the special somatic sensory fibres of the lateral line system, whose fibres join the roots of the facial, glossopharyngeal, and vagus to reach the brain, and the ganglia from which these fibres are derived become incorporated in the geniculate, petrosal, and nodosal ganglia. A trace of these organs is seen in the human embryo in the form of a temporary thickening of the ectoderm directly over the ganglia of these three nerves. The fourth member of this group, the trigeminal nerve, is distinguished by a larger admixture of somatic sensory fibres. The ganglia of all four nerves are derived from the neural crest.

The n. facialis is characterized by a large predominance of visceral motor fibres which make up the large motor root of the adult. These motor fibres can be seen in the 10 mm. embryo (Figs. 96 and 100) arising from a group of neuroblasts situated beneath the third rhombic groove or neuromere. The fibre bundles are assembled and pass directly lateral under the floor of this groove and gradually converge to form a solid trunk which emerges from the neural tube just median to the acoustic ganglion. On emerging the motor trunk curves caudalward (Fig. 100) and terminates among the cells of the hyoid arch which are destined to form the muscles of expression.

DEVELOPMENT OF THE NERVOUS SYSTEM.

137

The sensory fibres of this nerve spring from the geniculate ganglion, which is apparently a derivative of the neural crest. It can be clearly made out toward the end of the third week, lying in front of and separate from the ganglion acusticum. Shortly after a path of loosely grouped fibres can be seen extending from it to the neural tube constituting its proximal root, the n. intermedins. On entering the alar plate it forms, as can be seen in the 10 mm. embryo (Figs. 96 and 100), a fibre path which bends caudalward to join the tractus solitarius. From the peripheral end of the ganglion fibres pass down as the chorda tympani, which finally

Nucl. mot. n. IX Rr. sens. IX et X I

Tractus solitarius Nucl. mot. n.'X (ambiguus)

Nucl. mot n. VII

Pars intermedia n. VII

Gang, genie

Lamina alaris

Lamina basalis

Gang, radicis n. vagi

N. chorda tymp. I

N. facialis Gang, petros. n. IX

Gang, nod os. n. vagi

Fig. 96. — Reconstruction of the left facial, glossopharyngeal, and vagus nerves of the same embryo shown in Fig. 86. A transverse section of the neural tube is included in the reconstruction to show it* relation to the different nerve-roots. This should be compared with Fig. 100.

leaves the main trunk of the nerve to enter the mandibular arch, eventually joining the third division of the trigeminal nerve. The great superficial petrosal nerve is another peripheral derivative of this ganglion, which makes its appearance shortly after the chorda tympani and extends forward to reach the anlage of the sphenopalatine ganglion. Most authors in comparing the seventh nerve with a typical branchial nerve describe the great superficial petrosal as representing the prsetrematic branch and the chorda tympani as the posttrematic branch. In the early embryo the sensory root or pars intermedia, as compared with the motor root, is larger than in the adult. The subsequent marked growth of the

138 motor root and the relative, standstill of the sensory root result in the former becoming the main trunk of the nerve.

Owing to the close relation existing between the facial and acoustic nerves the two are frequently classed as the facial-acoustic complex. However, aside from the fact that the acoustic nerve, in its development in the higher vertebrates, crowds in against the facial nerve and becomes more or less fastened together with it by mesodermal elements, it has no other thing in common, the two being nerves of entirely different embryological and functional significance. The relation between the facial and abducens becomes reversed in the adult from that of the early embryo. The facial at first lies directly under the third rhombic groove, while the abducens is more caudal and is under the fourth rhombic groove. As shown in Fig. 97, the two nerves gradually shift their relative

Genu internum n. facialis

Sulcus

Sulcus

B

Sulcus

Fig. 97. — Diagram illustrating three stages in the development of the genu facialis, the youngest, A, corresponding to the 10 mm. embryo, and the oldest, C, the new-born child. The relative position of the nucleus of the n. abducens is represented in outline. Sulcus, sulcus medianus fossa? rhomboidese.

positions, the abducens moving forward. This migration results in the bending of the motor root of the facial out of its original course and produces the genu facialis.

The n. gloss opharyng eus possesses a ganglion of the root and ganglion of the trunk, the latter being temporarily connected with the placode over the third arch. As can be seen from the relative size of the ganglia in Fig. 86, the nerve consists almost wholly of sensory fibres, connected peripherally with the structures developing from the second (r. tympanicus) and third (r. lingualis) arches. The tympanic branch is not well defined until we come to embryos between 12 and 14 mm. long. Centrally the rootlets enter the brain wall and, joining with the fibres of the facial, extend caudally (Fig. 96) as the tractus solitarius. The motor rootlets of this nerve arise from a group of neuroblasts in the nucleus ambiguus series, situated beneath the floor of the fifth rhombic groove. The motor bundles extend directly lateral beneath this groove and pass under the spinal tract of the trigeminal and then emerge from the brain wall and join the main trunk of the nerve. This nerve forms a more typical branchial nerve than either the facial or vagus. Its tympanic branch is regarded as the prsetrematic branch and the lingual as the posttrematic branch.

DEVELOPMENT OF THE NERVOUS SYSTEM. 139 The vagus complex (n. vagus et n. accessorius) represents several branchial nerves, the motor fibres of which in man have undergone special development for the purpose of supplying the group of muscles derived from its branchial arches. The facial nerve is also a similar nerve; its large motor trunk, as we have seen, is distributed to the muscle-cells of the hyoid arch, and, as these cells group themselves into the muscles of expression and spread forward over the face, the facial branches are drawn along with them. In a similar way the more caudal rootlets of the vagus become predominantly motor, and form a distinct bundle which we know as the spinal accessory nerve, and this bundle is distributed to a group of muscle-cells originally belonging to the more caudal branchial arches, and in man is destined to form muscles for the arm girdle, the mm. sternocleidomastoideus and trapezius. As these muscles spread out into their eventual position the nerve is drawn down across the neck with them. Coincident with the increased importance of this musculature as we ascend the vertebrate scale we meet with increased development of the accessory nerve, and it obtains additional rootlets of origin by spreading down into the region of the spinal cord. As can be seen in Fig. 86, it may reach as far down as the fourth cervical segment. The nucleus of origin of the spinal accessory and other motor rootlets of the vagus constitutes the nucleus ambiguus of the medulla oblongata and a portion of the lateral horn of the spinal cord, the two being continuous (see Fig. 100).

The early stages in the growth of the glossopharyngeal and vagus nerves are shown in Figs. 82, 83, 86, and 94-. Their sensory elements are derived from the ganglion crest of the hind-brain in the same manner that spinal ganglia are derived from the ganglion crest of the cord. The crest of the hind-brain and that of the cord are described by some as continuous (Dohrn, 1901), but Froriep (1901) distinguishes between a ganglion crest of the head and one of the trunk, the two overlapping in the occipito spinal region. The cranial ganglion crest migrates ventrally down the side of the neural tube and is soon joined by visceral motor fibres that emerge from the lateral border of the neural tube. In the 4 mm. embryo, Fig. 82, a bundle of such fibres can be seen running along the ventral border of the crest and constituting the primitive n. accessorius. Soon after this the cells of the crest show signs of differentiation and are gradually converted into sensor} glion-cells with fibrous processes, which attach themselves to the neural tube on the one hand and extend peripherally on the other, — i.e., dorsal rootlets. This fibre development results in the break up of the ganglion crest into cell masses, which is not a metameric segmentation such as appears in the spinal region. These masses constitute the ganglia of the roots. The ninth nerve has one. the

140 ganglion of Ehrenritter ; while the tenth, being a composite nerve, has a series of them, the most oral one being the largest. The ganglion of the trunk (ganglion nodosum), as is also true of the ganglion petrosum of the glossopharyngeus, when first identified does not seem to be definitely connected with the root ganglia. Furthermore it differs from the root ganglia in being connected with a rudimentary sense organ (epibranchial placode), as shown in Fig. 98. It, however, is generally considered as a derivative of the ganglion crest.

Sang, petros et Organ sens.

Fig. 98. — Reconstruction of the nerves in the occipital region of a 7 mm. human embryo (Mall collection, No. 2), showing sensory placodes in connection with the petrosal and nodosal ganglia. Enlarged 16.7 : 1. Ot. v., ear vesicle; gang, crest., ganglionic crest.

The formation of the vagus rootlets at a somewhat later stage is shown in Figs. 96 and 100. The motor neuroblasts point dorsolateralward and assemble in a series of rootlets which emerge just ventral to the entrance of the sensory roots. After emergence they turn forward and form a common trunk, which in the spinal region lies between the dorsal roots and the side of the spinal cord. Connected with the motor roots are the sensory roots and ganglia. The development of the ganglia is more pronounced in the cephalic end of the nerve. The more caudal ganglia usually disappear in the adult, except for traces of scattered ganglion-cells found occasionally on the rootlets of the accessory division. Owing to the tendency to regression on the part of the more caudal of the vagus root ganglia, the vagus complex becomes differentiated into a fore part or vagus division which is predominantly sensory, and a hind part or accessory division which is almost wholly motor. In other words, in the course of phylogenesis as the vagus invades the spinal region it is the motor elements that play the prominent part. The more caudal vagus ganglia are not to be mistaken for the Froriep ganglion, which represents a persistent precervical

DEVELOPMENT OF THE NERVOUS SYSTEM. 141 ganglion. In the one case we have a series diminishing from I head toward the tail, and in the other it is in the opposite direction. The sensory fibres derived from these ganglia enter the wall of the neural tube and immediately nnite in a longitudinal tract continuous with similar fibres from the facial and glossopharyngeal, thus completing the formation of the tractus solitarius, whose form and relation to a section through the oblongata region are shown in Fig. 96.

The n. trigeminus possesses on its sensory root the largest ganglion of the whole embryo, the ganglion semilunare. Processes which grow out from its constituent cells connect it with the brain on the one hand, and extend as three large trunks on the other into the ophthalmic, maxillary, and mandibular region-. 'This ganglion has generally been considered as a single undivided mass of cells derived entirely from the ganglion crest (Dixon, 1896). A human embryo of 15 segments has, however, been described by Giglio-Tos (1902), in which the anlage of the semilunar ganglion consists of three separate proganglia, " proganglia neurales, ,; which are connected by a cellular lamina with three other distal or epibranchial proganglia, the whole group being eventually fused into a single ganglion mass. If this is the case, it is probable that the semilunar ganglion arises in part from the ganglion crest and in part from the epibranchial ectoderm. Such a composite character would correspond in some degree with the condition found in lower vertebrates. It has been suggested by Johnston (1908) and others that perhaps the sensory ganglion-cells belonging to the midbrain and the most oral part of the hind-brain become included in the wall of the neural tube, instead of becoming detached with the neural crest. Such cells then never become incorporated in the semilunar ganglion. It is these cells that are supposed to form the mesencephalic root of the trigeminal nerve, their processes extending caudalward beneath the central gray substance to join the main sensory trunk of the nerve at its point of entrance into the wall of the tube.

Processes grow out from the constituent cells of the ganglion and extend peripherally as three large trunks or divisions (see Figs. 83, 99, and 86). The ophthalmic division passes forward and soon subdivides into the frontal and nasociliary nerves ; the latter in the 10 mm. embryo is a well-defined branch just dorsal to the eye stalk. The maxillary and mandibular divisions extend downward and break up in their terminal branches among the cells of the maxillary process and mandibular arch respectively. By the beginning of the sixth week the chief branches of these divisions can be recognized. Centrally the ganglion becomes connected with the brain by a large single root, consisting of both somatic and visceral sensorv fibres. This enters the wall at the pontine flexure

142

HUMAN EMBRYOLOGY

opposite the first and second rhombic grooves. TYithin the wall in the marginal zone the fibres form a flattened longitudinal tract, part of which extends caudally as the spinal tract, and part extends forward and upward to enter the cerebellar rid^e. as shown in Fig. 100.

In its motor elements the trigeminal nerve departs somewhat from the type represented in the other three nerves of this visceral group. In the others the nucleus of origin is in the basal plate,

sac : - . ~oh.

-cochlea ~ -3t.aud.ext

motor YE.

Fig. 99. — Reconstruction of the facial and trigeminal nerves of a human embryo 14 mm. long (Mall collection. No. 144), showing motor and division of the facial sensory. Enlarged 8 : 1.

and the nerve rootlets exhibit a characteristic curved course to reach the point of emergence ; while in the trigeminal the nucleus is more lateral and lies directly against the entering sensory fibres, so that the fibres of the motor root pass directly ventralward to fuse with the mandibular division. Its nucleus corresponds to the dorsal motor nuclei found in the adult ninth and tenth nerves, though it is much larger. On analysis it is found that a typical visceral cranial nerve has three central terminations. — sensory root (tractus solitarius), ventral motor root (nucleus ambiguus),

DEVELOPMENT OF THE NERVOUS SYSTEM.

144 HUMAN EMBRYOLOGY.

dorsal motor root (nucleus nervi dorsalis). The ventral motor root always has a characteristic curved course between its nucleus and point of emergence, while the nucleus of the dorsal motor root is clustered near the sensory division of the nerve, and from it the root extends to the surface of the brain in a straight line. These three elements are represented in the different nerves in different proportions. The ninth nerve approaches the mean and all elements are fairly represented. In the vagus the curved ventral motor roots are increased in the caudal portions and form the spinal accessory. In the facial the sensory root (n. intermedius) is diminutive, while the curved ventral root becomes the main trunk of the nerve. In the trigeminal it is the straight dorsal motor root that forms the principal motor supply, while the curved ventral motor root must be considered as absent. It should be mentioned, that though the motor root eventually fuses with the mandibular division, in the embryo the motor fibres corresponding to the n. masticatorius have been observed passing on the median side of the ganglion and extending from its distal border, not with the mandibular division, but as a separate trunk (Streeter, 1904). The cranial sympathetic ganglia derived from the trigeminal nerve will be described under the sympathetic system.

IV. SYMPATHETIC NERVOUS SYSTEM.

The sympathetic system arises in common with the spinal ganglia from that portion of the ectoderm which forms the lateral border of the neural plate, and in common with the spinal ganglia it takes part in the formation of the neural crest. As the neural crest becomes detached and its segmenting parts invade the space between the myotomes and neural tube, certain ganglioncells separate themselves from its ventral border and independently migrate ventralward into the neighborhood of the aorta. It is these cells that form the connected chain of errant ganglia which we know as the sympathetic system. It is a true derivative of the rest of the nervous system. Originally the two were continuous ectoderm, and the establishment of the former as a distinct and separate system is solely due to its detachment and forward migration.

To witness the successive steps in the development of the sympathetic system it is necessary to commence with embryos that are still in the neural crest stage, 2-3 mm. long. At 7 mm. the cell migration is in active progress, and cellular rami communicantes are already present in some regions ; in the 9 mm. embryo the ganglionic cord and splanchnic nerve-plexus are definitely outlined; and finally, in the 16 mm. embryo the differentiation has advanced far enough so that it is possible to recognize the more outlying visceral ganglia and the ganglia of the head and their connecting branches. Thus, before the completion of the sixth week of embryonic life the essential features of the entire sympathetic apparatus can be clearly made out.

A representation of the derivation of the sympathetic cells is shown in Fig. 101, in which A, B, C, and D represent the successive stages and show schematically the increase in number and forward migration of cells and the subsequent formation of connecting fibre trunks. The sympathetic cells are shown in black in contrast to the other derivatives of the neural crest, — i.e., the sheath cells and the spinal ganglion-cells. The sheath cells are shown as plain white rings, while the spinal ganglion-cells are dotted.

Fig. 101. — Diagram representing the origin and migration of the sympathetic ganglion-cells and their relation to the other components of the neural crest. Sympathetic cells, black; spinal ganglion-cells, dotted circles; sheath cells, white rings. A and D are based on embryos Nos. 12 and 143 of the Mall collection, and B and C on the Buxton embryo, Gage collection.

The three varieties of cells mentioned can be traced back to the neural crest, as indicated in stage A. They are, however, not to be distinguished from each other histologically at this time; it is only for schematic purposes that they are represented in Vol. II.— 10 that way in the drawing. In the next drawing, B, can be seen the increase in size and ventral advance of the entire ganglion mass. The ventral advance is in part secondary to the increase in the number of cells and the consequent forward crowding in the direction of least resistance, and in part it is due to the intrinsic migratory energy of the individual cells. What appears as an ill-defined ragged ventral border, on closer examination can be seen to be made up of branching ganglion-cells reaching forward into the mesoderm in the form of a loose syncytium. In the drawing, for the sake of simplicity the processes are not shown. As seen in B, cells from the loose ventral border of the ganglion mass detach themselves and extend ventralward in advance of the ventral root fibres. The latter can be seen emerging from the neural tube and following along in the wake of the migrating ganglion-cells. Simultaneously with the forward growth of the ventral root fibres and the formation of a definite nerve-trunk, as shown in C, the sympathetic cells continue their migration medianward toward the aorta and thereby form a cellular connected strand which is the primitive ramus communicans. The difference between drawings B and C is that existing between the upper and lower parts of the spinal cord of the same embryo. B corresponds to the upper thoracic region in the 4.5 mm. embryo.

The completion of the segmental type is shown in D, where the sympathetic cells have completed their wandering, and by rapid increase in numbers form a compactly clumped ganglion mass which by fusing from segment to segment extends as a continuous longitudinal cord along the lateral border of the aorta. The cellular ramus communicans is in the meantime replaced by centrifugal fibres from the nerve-trunk which have followed along the migration path, the same fibres that form the white ramus in the adult. Somewhat later centripetal fibres sprout out from the sympathetic ganglia and either work their way backward along the path of the centrifugal fibres, or else form an independent bundle, the future gray ramus. The ramus communicans thus represents a portion of the path along which the sympathetic cells originally migrated. At first consisting of a chain of ganglioncells spun out by the wandering cells, it is later replaced by an ingrowth of fibres representing spinal axones on the one hand and sympathetic axones on the other ; the cells forming the temporary connecting bridge, having in the meantime completed their journey, form a compact ganglion mass near the aorta.

If we judge from the adult conditions we must conclude that there are some sympathetic cells which never wander out from the spinal ganglion mass. Some of these stationary cells are shown in the drawings C and D. Likewise some of the sheath cells do not leave, but remain in the ganglion mass, where they either

DEVELOPMENT OF THE NERVOUS SYSTEM. 147 form sheaths to the nerve-fibres or else form cell nests encapsulating the ganglion-cells proper. The wandering sheath cells, as seen in B and D, advance simultaneously with the sympathetic cells at the tip of and along with the nerve-fibres, and by the time a well-defined nerve-trunk makes its appearance sheath cells are found scattered along its whole course, as well as along the ramus communicans.

The development of the individual sympathetic ganglion-cell may be divided into three stages. The first, or indifferent stage, covers the period during which it is one of the cells forming the neural crest and spinal ganglion mass ; the second, or intermediate stage, corresponds to the period of migration; the third, or terminal stage, is from the time it reaches its permanent position to the attainment of its adult form.

During the first stage the three cell groups derived from the neural crest — i.e., the sympathetic ganglion-cells, the spinal ganglion-cells, and the sheath cells — cannot be distinguished from one another, and may be considered as indifferent ectoderm cells. Together they constitute a moderately compact mass lying between the neural tube and the dorsal border of the myotome. Their body-outlines are ill-defined and show more or less fusion. As compared with the adjacent mesodermal cells they possess more body protoplasm and are not branched. The nuclei are oval or rounded and are marked by deeply staining nucleoli.

During the second or wandering stage the sympathetic cells, as is common with all wandering cells, are characterized by the development of slender protoplasmic processes. The sheath cells also develop similar processes about the same time, and the two cannot be easily distinguished from each other. The nuclei of the latter, however, are somewhat more elongated and take the stain less intensely. The spinal ganglion-cells may be readily identified by the well-defined protoplasmic body which the more advanced ones have in the meantime developed. It is proposed by Kohn (1905) that the sheath cells of this period and the sympathetic cells, because of their similarity and their prevailing presence along the developing nerve, be grouped together under the term neurocytes. A syncytial strand of this type of sympathetic cells forms the initial ramus communicans. which can be seen making its way through the mesoderm toward the aorta, entirely devoid of fibres. See accompanying Fig. 102, which shows a similar picture in the rabbit.

With the beginning of the third stage the sympathetic cells undergo their terminal differentiation. They rapidly proliferate and clump themselves into ganglia; while the individual cells develop condensed and sharply outlined protoplasmic bodies and their processes become fibrillar and extend out to take part in the formation of the various communicating trunks. In passing through these three stages the sympathetic cells do not all develop with equal rapidity, and there is consequently an overlapping of the successive periods; for example, the cells belonging to the ganglionated cord are well along toward the completion of the third stage at a time when the visceral ganglia are still in the second stage, and likewise some parts of the ganglionated cord develop

•

• 2>k

e

•4 TC

B-j

Fig. 102. — From a cross section through a rabbit embryo, showing chain of sympathetic cells forming the primitive ramus communicans {re), which leads from the fibrous spinal nerve (n) medianward toward the aortic region. Enlarged 300 : 1. nz. neurocytes ; bk, blood-cells. (After Kohn, 1907.) more rapidly than others, the thoracic region is always in advance of the lower lumbar and sacral and the cranial ganglia.

The earlier writers (Remak, 1847) were of the opinion that the sympathetic system was of mesodermal origin, until Balfour (1877) showed that in selachians the sympathetic ganglia develop as buds or outgrowths from the trunks of the spinal nerves, and hence are ectodermal. Some observers (Pater son, 1890) still adhered long after to the mesodermal origin, yet Balfour's observations were thoroughly confirmed and the ectodermal origin of the sympathetic was generally regarded as thereby established. Later investigators (Schenk and Birdsall, 1878, Onodi, 1886), who worked

DEVELOPMENT OF THE NERVOUS SYSTEM. 149 on higher forms, including human, modified Balfour's view by describing the sympathetic ganglia as detached parts of the spinal ganglia which are separated off and linked together into a longitudinal chain. It was not until His, jun. (1891), introduced the principle of the wandering of individual sympathetic cells that we approached our present conception. It was surmised by His (1890) and immediately supported by His, jun. (1891), that the development of the sympathetic system is dependent on the active ventral migration of germinating sympathetic cells, which cells, however, do not migrate until a preliminary nerve-fibre framework is laid out in the form of rami communicantes and connecting longitudinal commissures. Along these fibre paths the sympathetic cells wander forward to form ganglia.

The only essential difference between the description of His, jun., and that as given above in the present article, consists in the time of cell migration. The writer is in accord with the recent work of Kohn (1907) in believing that the migration occurs earlier than described by His, and that the cells wander through mesoderm rather than along compact nerve-fibre paths. The same picture is presented in human material that Kohn describes in the rabbit : migrating cells can be recognized in advance of the loose strands of the tip of the growing nerve, and extending through the mesoderm as a bridge of cells toward the aorta; by the time a well-defined nerve-trunk is established the sympathetic cells have already completed that part of their migration, and the cells then found on the nerve-trunk are sheath cells only. A divergent view has been recently published by Cajal (1908), according to which the sympathetic cells in the chick are true motor cells and are derived from the spinal cord. During their germinative stage they migrate out from the cord at the same place at which later the ventral roots emerge.

The cranial sympathetic system departs from the uniform segmental type found in the trunk. It, however, adheres in three ways to the general type: first, the ganglion-cells are apparently derived from a cerebrospinal ganglion mass; secondly, they migrate ventralward and eventually assume a position outside of the bony canal, and thirdly, the ganglia give off fibres which form a communicating trunk along the internal carotid artery, this trunk serving to unite the cranial sympathetic ganglia with each other, analogous to the longitudinal system of communications of the ganglionated cord, with which it is continuous.

The early history of the cranial sympathetic ganglia is less definitely known than that of the trunk system, due to the relatively smaller size, lesser number, and lack of symmetry of the former ganglia, as well as the complexity of the surrounding struc tures. From the evidence at hand it seems probable that all the

150 HUMAN EMBRYOLOGY.

cranial sympathetic ganglia — gg. ciliare, sphenopalatinum, oticum, and submaxillare — are originally derived from the semilunar ganglion mass.

The ciliary ganglion presents certain modifications. In the chick it consists (Carpenter, 1906) of two portions, a smaller dorsal sympathetic portion derived from the semilunar ganglion, and a larger ventral portion containing large bipolar cells (supposedly sensory), derived from the neural tube by migration along the oculomotor nerve. In the same way in the torpedo it is known that cells wander out from the medullary tube and migrate ventralward in company with the growing oculomotor fibres, and eventually fuse with the cells derived from the trigeminal ganglion, together forming a composite ganglion (Froriep, 1902). If there are any oculomotor ganglion-cells in the human embryo, such as are found in the chick and torpedo, they do not apparently pass through a migrating stage, and never leave the neural tube. Consequently the ciliary ganglion here consists exclusively of migrant cells from the semilunar ganglion. We may assume that they behave like the sympathetic cells of the trunk and pass through their wandering stage very early. Their detachment from the parent ganglion mass and forward migration must occur just in advance of the developing nerve-trunks, and it is not until they reach their permanent position that they undergo active proliferation and form a compact cell group.

Both the sphenopalatine and submaxillary ganglia are probably derived entirely from the semilunar ganglion, but it must be borne in mind that they are connected with the geniculate ganglion of the facial, and there is the possibility that the former contains contributing cells which have migrated along the path of the great superficial petrosal nerve and the latter cells which have migrated along the path of the chorda tympani. In the same way the otic ganglion, though developed intimately with the semilunar ganglion, may in part consist of cells derived from the glossopharyngeal nerve through its tympanic branch. However, both their comparative and embryological histories indicate that the facial, glossopharyngeal, and vagus nerves constitute a definite group, one of the characteristics of which is that they retain within their root or trunk ganglia or within the brain tube itself whatever sympathetic ganglion-cells they possess, and consequently there are in connection with them no rami communicantes and no derivative ganglia.

The origin of the four cranial ganglia may be represented as in the adjoining scheme, Fig. 104. The arrows indicate the paths of primary migration, and the dotted lines paths of subsequent intercommunication, by which all the cranial ganglia establish connection with the carotid plexus and so become directly continuous

DEVELOPMENT OF THE NERVOUS SYSTEM.

151

with the gangliated cord of the trunk. This figure should be compared with Fig. 103, which represents a profile reconstruction of an embryo 16 mm. long. For the purpose of contrast the sympa

A.carot int. et

Q nodos.

Tr'uncus sympathicus Fio. 103. — Profile reconstruction of the sympathetic nervou9 system in a 16 mm., nearly six weeks, embryo (Huber collection, VI), enlarged 10 : 1. In order to expose the cceliac plexus and suprarenal gland the stomach is represented as raised forward and to the right.

thetic system is shown in black. The cranial sympathetic ganglia at this time are connected with the semilunar ganglion by longer or shorter branches which are analogous to rami communicantes.

152 HUMAN EMBRYOLOGY.

The ganglion ciliare lies closely against the oculomotor nerve, from which it receives some fibres. A true ramus communicans, identical with the radix longa of the adult, connects it with the ophthalmic division of the trigeminal nerve. The sphenopalatine ganglion is connected with the parent ganglion by two or more rami communicantes, the nn. sphenopalatine These are in part sensory fibres, which pass' directly through the ganglion without interruption, connecting the periphery with the semilunar ganglion. It is fibres % of this sort that form the major part of the peripheral branches of the ganglion. In case of the otic ganglion there is a less distinct ramus communicans, owing to the fact that the ganglion lies closely against the nerve-trunk and thus there results in the adult a short plexus uniting the two. The submaxillary ganglion presents an even more close union between the ganglion and the nerve-trunk, and the latter in 16 mm. embryos can be seen making its way directly through the substance of the ganglion mass. So here, as in the case of the otic ganglion, there results a plexus of communication between ganglion and nerve-trunk. In speaking of this as the ganglion submaxillare it is done in the sense of including both the submaxillary and sublingual ganglia. As has been shown by both Langley and Huber (Huber, 1896), that which is ordinarily referred to as the submaxillary ganglion is in reality sublingual, while the submaxillary ganglion proper consists of multiple ganglia situated in the substance of the gland along the course of its ducts.

It was shown in Figs. 101 and 104 how the sympathetic cells in the spinal region migrate forward and form ganglia, and it has been also mentioned that these ganglia fuse from segment to segment and thereby form a continuous longitudinal cord of ganglioncells, the so-called ganglionated cord. This structure is at first purely cellular. Later, fibres make their appearance among the cells; in the 16 mm. embryo, Fig. 103, they are already abundant, particularly in the cervical and thoracic regions. The fibre growth continues in such a manner as to break the continuity of the cellular cord, and produces a longitudinal series of ganglionic masses connected by intervening fibrous bridges, the ganglionic chain as seen in the adult. For the greater part these ganglia are segmental, but in the cervical and upper thoracic region the cells remain massed in larger clumps, and there result ganglia corresponding to from two to five segments.

The prevertebral and visceral sympathetic ganglia are considered by most writers as derivatives of the neural crest in common with the rest of the sympathetic system. They differ only in that their migration extends further than that of the latter: instead of stopping at the side of the aorta, they migrate ventralward through the loose mesoderm into the region of the preverte

DEVELOPMENT OF THE NERVOUS SYSTEM.

153

bral plexuses, and some of them still further forward to become incorporated in the walls of the viscera to form the submucous plexus. These paths of migration are diagrammatically shown in Fig. 104. These ganglia reach their eventual position relatively early, so the distance covered in their migration is therefore not so great. Cajal (1908) describes in the chick of 52 hours visceral

Q ciliare

G. sphenopal Q. submax

Pix coeiiacus Pix submucosus

Pix hypoqastr

Fig. 104. — Diagram showing the migration paths of the sympathetic cells. Dotted lines indicate secondary subsequent communications which link the ganglia together and form a longitudinal chain continuous throughout head and trunk. Secondary and tertiary migrations result in the formation of prevertebral and visceral plexuses.

sympathetic cells which have completed their migration. In human embryos 16 mm. long the cardiac ganglia are to be recognized. The main features of the cardiac plexus are completed by the time the embryo reaches 19 mm. neck-rump length, as shown in Fig. 105. The cceliac and hypogastric plexuses together with the splanchnic nerves present the picture seen in Fig. 103, and differ from the adult only in the incomplete differentiation of the cells. Continuous with the cceliac plexus is a group of sympathetic

154 cells which extend through its median surface directly into the substance of the suprarenal gland, and constitute its nerve supply. A portion of these cells, instead of becoming typical ganglioncells, undergo special development, the details of which are not yet satisfactorily understood. On account of the affinity of these

N. vagus

Aorta

Bulbar plexus

Auricle

Truncus sympathicus

Connecting plexus

Atrial plexus

Fig. 105. — Cardiac rlexua in human embryos between 10 and 19 mm. long. (After Kollmann, 1907, and iHis, jun.. 1891.) special cells for the chrome salts, they are designated as chromaffin cells. They are found also in other portions of the sympathetic system, such as in the ganglia of the ganglionated cord and of the abdominal plexuses; to some extent they also form independent bodies, the chromaffin bodies of Zuckerkandl, to which group the carotid glands belong. These structures will be described in detail in the following chapter.

LITERATURE.

Balfour, F. M. : The Development of Elasmobranch Fishes. Development of the Spinal Nerves and of the Sympathetic Nervous System. Journ. of Anat. and Physiol. Vol. 11. 1877. Bardeen, C. R. : Development and Variation of the Nerves and the Musculature of the Inferior Extremity and the Neighboring Regions of the Trunk in Man. Anier. Journ. of Anat. Vol. 6. 1906-07. Bardeen and Lewis: Development of the Limbs, Body-wall, and Back in Man.

Amer. Journ. of Anat. Vol. 1. 1901. Bechterew, W. : Ueber die binteren Nervenwurzeln usw. Archiv f . Anat. u.

Physiol. Anat. Abt. P. 126. 1887. Bremer, J. L. : Aberrant Roots and Branches of the Abducent and Hypoglossal Nerves. Journ. of Compar. Neurol, and Psychol. Vol. 18. 1908. Carpenter, F. W. : The Development of the Oculomotor Nerve, the Ciliary Ganglion, and the Abducent Nerve in the Chick. Bulletin Museum Compar.

Zool. Harvard CoUege. Vol. 48. 1906. Chiarugi, G. : Contribuzioni alio studio della sviluppo dei nervi encefalici nei mammiferi in confronto con altri vertebrati. IV. Sviluppo dei nervi oculomotore e trigemello. Publi. d. R. Instit. di Studi super. Firenze 1897.

DEVELOPMENT OF THE NERVOUS SYSTEM. 155 Dixon, A. F. : On the Development of the Branches of the Fifth Cranial Nerve in Man. The Scientific Transactions of the R. Dublin Soc. Vol. 6, Series 2. Dublin 1896. The Sensory Distribution of the Facial Nerve in Man. 4 Fig. Journ. of Anat. and Physiol. Vol. 33, N. S. Vol. 13, p. 471^92. London 1899. Eisler, P. : Ueber die Ursache der Geflechtbildung an den peripheren Nerven. Verh. Anat. Gesellsch. 16. Vers. P. 200-207. Halle a. S. 1902. Anat. Anz. Bd. 21. Erg.-Heft. Futamura, R. : Ueber die Entwicklung der Facialismuskulatur des Menschen.

Anat. Hefte. Arbeiten. Bd. 30. 1906. Giglio-Tos, Ermano: Sugli organi branchiali e laterali di senso nell' uomo nei primordi del suo sviluppo. Monit. zool. ital. Vol. 13, p. 105-119. 1902. Sull' origine embrionale del nervo trigernino nell' uomo. 4 Fig. Anat. Anz.

Vol. 21, p. 85-105. 1902. Sui primordi dello sviluppo del nervo acustico-faciale nell' uomo. Anat. Anz. Vol. 21. 1902. Grafenberg, Ernst: Die Entwicklung der Knochen, Muskeln und Nerven der Hand und der fur die Bewegung der Hand bestimmten Muskeln des Unterarms. 19 Fig. Anat. Hefte. Bd. 30, p. 1-54. 1906. Grosser und Frohlich : Beitrage zur Kenntnis der Dermatome der menschlichen Rumpfhaut. 14 Fig. Morphol. Jahrb. Bd. 30, p. 508-597. 1902. Hensen, Victor: Die Entwicklungsmechanik der Nervenbahnen im Embryo der Saugetiere. Ein Probeversuch. Kiel und Liepzig 1903. His, W. : Zur Geschichte des menschlichen Ruckeninarks und der Nervenwurzeln. Abh. math.-phys. Kgl. Sachs. Akad. d. Wiss. Bd. 13. 1886. Ueber die Entstehung und Ausbreitung der Nervenfasern. Anat. Anz. Bd. 1.

1886. Die Entwicklung der ersten Nervenbahnen beim menschlichen Embryo. Uebersichtliche Darstellung. Arch. f. Anat. u. Physiol. Anat. Abt. P. 368-378. 1887. Ueber die embryonale Entwicklung der Nervenbahnen. Anat. Anz. Bd. 3, p. 499-506. 1888. Ueber die embryonale Entwicklung der Nervenfasern. Verh. Anat. Gesellsch.

Wiirzburg 1888. Anat. Anz. Bd. 3. Histogenese und Zusammenhang der Nervenelemente. Arch. f. Anat. u.

Physiol. Anat. Abt. Suppl. 1890. Histogenese und Zusammenhang der Nervenelemente. Verh. 10. internat. Kongr. Bd. 2. Berlin 1891. — " Bemerkungen zu obigem Auf satz." Arch, f . Anat. u. Physiol. Anat. Abt. 1891. Ueber den Aufbau unseres Nervensystems. Verh. Gesellsch. deutscher Naturf. u. Arzte. 65. Vers. Teil 1. 1903. Also compare the papers of His mentioned in the Normentafel of Keibel and Elze. His, W., jun. : Zur Entwicklungsgeschichte des Acustico-Facialis-Gebiets beim Menschen. Arch. f. Anat. u. Physiol. Suppl. 1889. Die Entwicklung des Herznervensystems bei Wirbeltieren. Abh. d. math.-phys.

Kl. d. Kgl. Sachs. Gesellsch. d. Wiss. Bd. 18. 1891. Ueber die Entwicklung des Sympathicus bei Wirbeltieren mit besonderer Beriicksichtigung der Herzganglien. Verh. Anat. Gesellsch. Wien. P. 69~7~>. 1892. Anat. Anz. Bd. 7. Erg.-Heft. Ueber die Entwicklung des Bauchsympathieus beim Hiinchen und Menschen. Arch. f. Anat. u. Physiol. Suppl. 1897. His, W., jun., und Romberg, E.: Beitrage zur Herzinnervation. Fortsehr. d.

Med. Bd. 8. 1890. Huber, G. C. : Observations on the Innervation of the Sublingual and Submaxillary Glands. Journ. of Exper. Med. Vol. 1. 1896.

156 HUMAN EMBRYOLOGY.

Four Lectures on the Sympathetic System. Journ. Comp. Neurol. Vol. 7.

1897. Kohn, A.: Ueber die Entwicklung des peripheren Nerveusystems. Verh. Anat.

Gesellsch. 19. Vers. Anat. Anz. Erg.-Heft. Bd. 27. 1905. Ueber die Scheidenzellen (Randzellen) peripherer Ganglienzellen. Anat. Anz.

Bd. 30. 1907. Lenhossek, M. v. : Die Entwicklung der Ganglienanlagen bei dem menschlichen Embryo. Arch, f . Anat. u. Physiol. Anat. Abt. 1891. Lewis, W. H. : The Development of the Arm in Man. Amer. Journ. of Anat.

Vol. 1, p. 145-183. 1902. Lubimoff, Alexis: Embryologisehe und histogenetisehe Untersuchungen iiber das sympathische und centrale Cerebro-Spinal-Nervensystem. Virchow's Arch.

Bd. 40, p. 217-273. 1874. Muller, Erik : Beitrage zur Morphologie des Gefasssystems. I. Die Armarterien des Menschen. Anat. Hefte. Bd. 22, p. 377-385. Compare p. 531 ff. 1903. Neumayer, L. : Histo- und Morphogenese des peripheren Nervensystems, der Spinalganglien und des nervus sympathicus. Hertwig's Handb. 1906. Onodi, A. D. : Ueber die Entwicklung des sjnnpathisehen Nervensystems. Arch.

f. mikr. Anat. Bd. 26. 1886. Paterson, A. M. : Development of the Sympathetic Nervous System in Mammals.

Phil. Trans. Royal Soc. London. Vol. 181. 1890. Popowsky, L. : Zur Entwicklungsgeschichte des nervus facialis beim Menschen.

Morphol. Jahrb. Bd. 23, p. 329-374. 1895. Rabl, K. : Ueber das Gebiet des nervus facialis. Anat. Anz. Bd. 2. Nr. 8, p. 219-227. 1887. Ram strom, M. : Ueber die Innervation des Peritoneums der vorderen Bauchwand.

Verh. Anat. Gesells. Jena. Anat. Anz. Erg.-Heft, Bd. 25, p. 44-51. 1904. Remak, R. : Ueber die Entwicklung des Huhnchens im Ei. Arch, f . Anat., Physiol.

und wiss. Med. (Muller). 1843. Schenck und Birdsall: Uebor die Lehre von der Entwicklung der Ganglien des Sympathicus. Mitteil. aus d. embryol. Inst. z. Wien. 2. Heft. 1878. Schumacher, S. v.: Ueber die Nerven des Schwanzes der Saugetiere und des Menschen; mit besonderer Beriicksichtigung des sympathischen Grenz stranges. Anz. Kais. Akad. d. Wiss. Wien. Nr. 18. 1905. Smirnow, A. E. : Eihige Beobachtungen iiber den Ban der Spinalganglienzellen bei einem viermonatlichen menschlichen Embryo. Arch. f. mikr. Anat.

Bd. 59. 1902. Streeter, George L. : The Development of the Cranial and Spinal Nerves in the Occipital Region of the Human Embryo. 4 PI., 14 Fig. Amer. Journ.

of Anat. Vol. 4, p. 83-116. 1905. On the Histogenesis of the Spinal Ganglia in Mammals. Proc. Assoc. Amer.

Anat. 18th Sess. Philadelphia (Penn.) 1904. Ref. Amer. Journ. of Anat.

Vol. 4, p. 13. 1905. On the Development of the Membranous Labyrinth and the Acoustic and Facial Nerves in the Human Embryo. Amer. Journ. of Anat. Vol. 6. 1907. Tapt, M. B.: De l'histogenese des fibres du grand sympathique chez l'homme et les mammiferes. These. Paris 1902. Weigner, K. : Bemerkungen zur Entwicklung des ganglion acustico-faciale und des ganglion semilunare. 6 Fig. Anat. Anz. Bd. 19, p. 145-155. 1901. Westphal, A. : Ueber die Markseheidenbildung der Gehirnnerven des Menschen.

Arch. f. Psych u. Nervenkrankh. Bd. 29, p. 474-527. 1897. Zimmermann, W. : Demonstration einer Rekonstruktionszeichnung des Abducens, der ventralen Wurzeln des Glossopharyngeus, des Vagus und des Hypoglossus eines menschlichen Embryo anfangs des zweiten Monats. Verh. Anat.

Gesellsch. Gottingen. 1893. Anat. Anz. Bd. 8. Erg.-Heft.