Paper - Further contributions to the study of the evolution of the forebrain 5

From Embryology
Embryology - 18 May 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Online Editor  
Mark Hill.jpg
This historic 1923 paper by Johnston describes development of the forebrain in many different species, including human.

See also his earlier paper: Johnston JB. The morphology of the forebrain vesicle in vertebrates. (1909) J. Comp. Neurol. and Psychol. 19: 457-539.
Our understanding of forebrain development has improved significantly over the last 100 years, some current research looks at early molecular patterning mechanisms as well as the development of connections between different regions.

See the links below for the current notes pages.

Neural Links: ectoderm | neural | neural crest | ventricular | sensory | Stage 22 | gliogenesis | neural fetal | Medicine Lecture - Neural | Lecture - Ectoderm | Lecture - Neural Crest | Lab - Early Neural | neural abnormalities | folic acid | iodine deficiency | Fetal Alcohol Syndrome | neural postnatal | neural examination | Histology | Historic Neural | Category:Neural
Historic Neural Embryology  
1883 Nervous System | 1893 Brain Structure | 1892 Nervous System Development | 1900 fourth ventricle | 1905 Brain Blood-Vessels | 1909 corpus ponto-bulbare | 1912 nuclei pontis - nucleus arcuatus | 1912 Diencephalon | 1921 Neural Development | 1921 Anencephaly | 1921 Brain Weight | 1921 Brain Vascular System | 1921 Cerebellum | 1922 Brain Plan | 1923 Neural Folds | 1904 Brain and Mind | 1904 Brain Structure | 1909 Forebrain Vesicle | 1922 Hippocampal Fissure | 1923 Forebrain | 1927 Anencephaly | 1934 Anencephaly | 1937 Anencephaly | 1945 Spinal Cord | 1945 cerebral cortex | Santiago Ramón y Cajal | Ziegler Neural Models | Historic Embryology Papers | Historic Disclaimer
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Further Contributions To The Study Of The Evolution Of The Forebrain V. Survey Of Forebrain Morphology

J. B. Johnston

Anatomical Laboratory, University of Minnesota

Thirty-Two Figures

In this section references are made to the description and figures contained in sections I to IV pub‘.ished in this Journal for August, 1923.

A. General Considerations

In preparation for the discussion of the evolution of the forebrain some reference must be made to the results of studies by many workers during fifty years upon the evolution and morphology of the vertebrate head. The process of cephalization has consisted largely of modifications in the anterior segments due to the locaton of mouth and special sense organs in that region. There have been modifications in the trunk as well as in the head. No segment in existing vertebrates presents unmodified the structures of a typical or primitive segment. Any discussion of the problem of cephalization must take into account all the elements which enter into a complete segment and in particular must give consideration to all four functional divisions of the nervous system. Notable modifications of the segments in the occipito—spinal and branchial regions need not be considered here since they do not directly affect the evolution of the forebrain.

1. In the ontogeny of the nervous system two broad stages are recognized: that of the neural plate and that of the neural tube. Between these occurs a transitional period during which the neural tube presents an open anterior neuropore. It is clear that the neural plate represents a ring of nervous ectoderm which surrounds the primitive blastopore and it is probable that the period of the ‘open neuropore represents the last part of the gastrula stage of evolution during which the neurenteric canal served for the ingestion of food. If so, the closing of the neuropore must have taken place after the formation of a direct opening to the enteric cavity (palaeostoma). On this hypothesis it would be assumed that the closed neural tube has existed during all the time necessary for the evolution of the enteric cavity, the development of the branchial apparatus, the formation of the mouth with jaws, etc.

2. During the neural plate and open neuropore stages there was established a front end or foremost border of the central nervous system. It has been shown that this anterior end is marked by the optic chiasma of the adult brain of all vertebrates (terminal ridge in the embryonic brain; Johnston, ’09 b). This has been confirmed by Kingsbury (’20, ’22).

All parts of the diencephalon and telencephalon including the cerebral hemispheres must be regarded as derivatives of the walls of the neural tube morphologically caudal to the chiasma ridge.

3. The whole diencephalic and telencephalic region is lacking in effective nerves or centers and this fact has offered serious difficulty to all attempts to compare the divisions of this region to the segments of the head and trunk caudal to the level of the oculomotor nerve. There has been great difficulty also in all attempts to trace forward the equivalents of the two columns of His (sensory alar plates and motor basal plates) into the diencephalon and telencephalon. The floor of these brain segments instead of resembling the motor basal plates is made up chiefly of centers related to the olfactory and gustatory systems. If the statement of His that’ the sulcus limitans, marking the boundary between sensory alar and motor basal plates, ends in the preoptic recess be accepted, the conclusion must follow that the floor of the diencephalon belongs to the motor plates. In many studies between 1901 and 1913 the writer struggled to harmonize this assumption with the obvious relationship of the whole hypothalamic region with sensory systems (olfactory and gustatory). The only conclusion arrived at was that motor elements have evidently disappeared and that the centers of this region are to be compared with tract and commissural cells having associational functions at all levels of the neural axis. This conflict is now resolved by the introduction of a new point of view by Kingsbury (see below).

Anterior to the myotome supplied by the oculomotor nerve (premandibular somite) there exist in embryos transient structures supposed to represent muscles once existing in this region of which we have no knowledge. These structures are the anterior head cavities and the median mesoderm connecting them in selachian embryos, and the corresponding median mesoderm plate in vertebrate embryos generally, known as the prechordal plate.

The lack of muscle segments and of motor nerves and centers places this anterior region in sharp contrast to the clearly recognized head and trunk segments. We have here only a portion of‘ the brain together with special sense organs, skeletal, vascular, and other subsidiary tissues. This region Neal includes in the first head segment.

4. A very important factor in the interpretation of the anterior part of the neural tube is found in the extent of the floor plate of His. This is brought forward by Kingsbury in two very illuminating discussions (’20, ’22). The floor plate of His is the median seam connecting the two basal plates. This seam is marked in neural plate stages of embryos by the primitive streak which owes its origin to the closing of the blastopore. Everyone who has studied these stages has noticed that the primitive streak does not extend throughout the whole length of the neural plate but stops some distance behind the anterior end. In later embryos and adults there is a median seam in the floor of spinal cord and hind-brain in which there is a strongly developed system of fibers derived from the ependyma cells along the median plane. This peculiar structure is known to neurologists as the raphé. This raphé does not extend through the whole neural tube but is absent at least in the telencephalon and diencephalon.

Professor Kingsbury, treating these facts from the standpoint of the blastopore theory, shows that the raphé is coextensive with the closed—up blastopore or neurochordal seam and plate, and that it is important to distinguish that part of the brain which does not possess a true raphé (prechordal brain) from that which does (epichordal brain). The point of View presented by Kingsbury leads to the solution of many perplexing problems connected with the fundamental morphology of the diencephalon and telen— cephalon.

First of all it is very clear that the region of the prechordal brain cannot be divided into segments which, taken with sense organs and transient or persistent nerve — and muscle — elements, may be described as head segments. The writer abandons his earlier recognition of two or more head segments in this region and accepts, in general terms, the views of Neal (’98, ’14, ’18, ’19) as to brain segmentation.

The myotomes are derived from the primitive streak (lips of the blastopore) and are approximately coextensive with the primitive streak. The anterior and premandibular head cavities are joined with their fellows across the median plane by the prechordal plate. This median plate may be regarded as a remnant of the mesoderm primitively confluent in front of the blastopore. We do not know whether the anterior head cavities represent vestiges of once existing muscle masses. What is clear is that the region anterior to the myotome suppliedby the oculomotor nerve has not been shown to possess these essential elements of a true head segment: a) myotomes which develop muscles in at least some existing vertebrates, b) somatic effective nerve components and nuclei of origin in the brain, c) Visceral effective nerve components arising from this part of the brain.

There is no definite evidence that these structures ever existed and have aborted. In the absence of these structures it is impossible to speak of head segments. The suggestion is offered that the region represented by the first brain segment of Neal is in reality a pre-metameric region. This pre-metameric region contains the forebrain, paired eyes and olfactory organs, the pineal eye, palaeostoma, hypophysis, infundibulum, and pre-chordal mesoderm. In a}! the rest of the head and body the metamerism is complete; that is, each metamere includes somatic and visceral effective nervous mechanisms and muscles as well as sensory systems and correlating nerve centers.

Adelmann (’22) from a study of the prechordal plate in selachians and the chick comes to the conclusion, “If, then, as I have shown, the mesoderm is continuous around the anterior end of the notochord and the premandibular somites are formed by the lateral growth of a single, medial pre-axial mesodermal mass, it follows that we have to do here with a line of somites which is continuous around the anterior end of the notochord.”

This contribution of Kingsbury’s has introduced a new and quite fundamental factor into our conception of the nature of the forebrain which must serve as the starting point in a study of its evolution in the vertebrate series. The earlier view, that the forebrain represents one or more typical metameres which once existed at least in vertebrate ancestors, and that these metameres have lost their effective systems (muscles, nerves, motor centers), implied that the forebrain is a continuation of the neural tube in which the effective centers have atrophied because the peripheral organs have been lost. On the other hand, special sense organs and correlating centers have had a special development, which accounts for great peculiarities of form and such neomorphs as the cerebral hemispheres.

Kingsbury brings forward evidence that what is here called the premetameric neural tube is fundamentally devoid of effective mechanisms, that the basal plates of His do not continue forward into this region, that the sulcus limitans does not extend into it, and that the floor of the neural tube in this region is formed by the alar plates of the two sides which are continuous across the median plane. This may be expressed by saying that the nervous ectoderm, like the mesoderm, surrounded the blastopore, and the pre-metameric brain represents that part of the neural plate which encircled the blastopore in front.

The results of Kingsbury’s observations do not mean that the anterior end of the nervous system was at the anterior end of the primitive streak, but that the streak (or neurochordal seam) marks the extent of the typical metameric portion of the nervous system. In front of this is a central nervous region without a true raphé and devoid of effective mechanisms, but consisting of certain special sensory mechanisms and of correlating centers to which impulses come from the metameres and from which directive impulses are sent to the metameres.

The neural plate in the ontogeny represents an area of ectoderm of high sensitivity which primitively surrounded the gastrula mouth or blastopore. That part which lay at the sides of the blastopore after its elongation came to be divided into segments corresponding to the myomeres, the anterior part remained undivided as a pre-metameric region or head segment. In the metameric part the nervous ectoderm came to be differentiated into longitudinal columns each of which contained the centers for one of the functional divisions of the nervous system: somatic receptive, visceral receptive, visceral" effective, somatic effective (Johnston, ’02 c). The arrangement of these in the neural plate is illustrated in figure 1 and in ’02 c, ’O6, ’09 a. In each half of the neural plate the four columns lay in the order above named, beginning with the lateral border. No superficial indication of the dividing lines between these columns is seen on the open neural plate but when the plate rolls up into a tube there


a. b., area basalis n. l., nucleus lentiformis a. o. l., area olfactoria latcralis op. 12., optic vesicle

b. 0., bulbus olfactorius r. z'., recessus infundibuli

cbl., cerebellum 'r. m., recessus mamillaris

ch., optic chiasma 1'. n., recessus neuroporicus

c. 12., commissura posterior r. p., recessus preopticus

c. p. p., commissura pallii posterior r. 120., recessus postopticus

fov., fovea isthmi s. l. H ., sulcus limitans of His

g. 12., general pallium s. m., somatic motor or effective column s

h., hippocampal formation . s. , somatic sensory or receptive column

hab., nucleus habenulae ten, terminal ridge

hem, cerebral hemisphere t. 12., tuberculum posterius

hy., hypophysis v. m., visceral motor or effective column hyth., hypothalamus v. s., visceral sensory or receptive column l. 3., lamina supra.neuroporica v. t. 0., ventricle of the tuberculum 01l. t., lamina terminalis factorium

m., mouth 2. l. m., zona limitans medialis

man., mandibular arch ' III , nervus oculomotorius

nch., notochord I V,- nervus trochlearis

Fig. 1 Two diagrams of the arrangements of the functional columns in the anterior portion of the neural plate and tube. Based on the work of Kingsbury and the discussion in the present paper.

In A, showing the open neural plate, the floor plate is represented by a heavy line ending forward in something like a spearhead indicating the fovea isthmi and the transverse grooves running from it in the lateral walls of the midbrain. About the fovea are grouped the nuclei of the III and IV nerves in agreement with Kingsbury’s figure 6 (’20). The functional columns are outlined by broken lines: s.s., somatic sensory; v.s., visceral sensory; v.m., visceral motor; s.m., somatic motor. The sulcus limitans of His, lateral to v.m. does not extend forward beyond the mammillary recess. The anterior end of the motor basal laminae is marked by the tuberculum posterius. The mammillary recess and infundibular recess lie in the region where the visceral sensory laminae are confluent across the middle line. The pits which evaginate to form the optic vesicles are located within the somatic sensory laminae and are connected with one another across the midline by the primitive optic groove which later becomes the postoptic reces-s (r.po.). The primitive optic groove is bounded in front by the terminal ridge (ten) which is the neural ridge bounding the neural plate and forms the anterior border of the plate in the midline.

In B the neural plate is supposed to have rolled up into a tube which is represented as laid open along the dorsal seam. The relationships are the same as in A except that the region of the future hemispheres (hem) has expanded forward and the terminal ridge has been converted into the optic chiasma, indicated by the crossing fibers of the optic tracts. The lamina terminalis is the anterior end of the dorsal seam which meets the anterior end of the brain floor (not floor plate) at the terminal ridge. fibers from the retinas entering the brain along the optic stalks cross in the terminal ridge forming the optic chiasma. The optic tracts occupy an oblique thickening in the brain wall leading from the terminal ridge toward the thalamus. This optic ridge cuts off the optic stalk from the primitive optic groove. The optic stalks then appear to be connected with one another at the lower border of the lamina terminalis. The pit or recess formed here is known as the recessus preopticus (r.p.), while the primitive optic groove behind the terminal ridge remains as the recessus postopticus (r.po.)appears on the inner wall of the tube on each side the sulcus limitans of His, who recognized that this groove marked the line of division between the effective and receptive columns.

His regarded the neural tube as a tube whose roof plate and floor plate were of equal length and whose anterior end was closed by a terminal plate (lamina terminalis). The writer has shown (’09 b) that the lamina terminalis must be regarded as the anterior portion of the roof plate, while Kingsbury (’20, ’22) has shown that the anterior part of the brain floor does not possess a floor plate in the sense of a seam connecting the two basal plates of His. Since all these facts go to show that the pre-metameric brain is a highly special region, differing in very important respects from the metameric brain, it becomes of the greatest interest to determine the line of demarcation between the two. Evidences upon this point are found in, a) the extent of the raphé in the brain, 1)) the relations of the notochord and brain in embryos, c) the structure of the brain walls, and d) the extent of the sulcus limitans of His in the embryonic brain.

Kingsbury states that the raphé stops anteriorly at the fovea isthmi and that the anterior end of the notochord in the embryo seems to him to lie at about the same level. He therefore regards the fovea isthmi as the mark of the anterior end of the epichordal brain.

I would point out that the characteristics of the floor plate seem to extend forward somewhat beyond the fovea and that the region between the fovea and the tuberculum posterius possesses peculiarities of structure which distinguish it from the mammillary, infundibular, and chiasmatic regions. In embryos of Acanthias, Amia, pigeon, pig, sheep and man there is found not a sudden stopping of raphé fibers at the fovea but a rapid change to a condition in which the ependyma fibers are more sparse and are separated by large numbers of infiltrated cells (figs. 22 and 24).

In Amblystoma embryos at the stage when the mammillary recess becomes recognizable as a ventro-caudal pouch with a thin wall of flattened cells there is a remarkably clear differentiation of the brain floor in the median plane caudal to this recess. Beginning behind the tuberculum posterius the floor in the median plane consists of a continuous layer of high columnar cells which are heavily pigmented on their internal ends (fig. 30), while no comparable pigmentation is found elsewhere in the neural tube at this stage. This condition continues ‘almost without interruption right back through the length of the spinal cord. These columnar pigmented cells form only a narrow streak at the median plane. These cells undoubtedly constitutethe floor plate in the sense of His and Kingsbury in the spinal cord and hind—brain. I find it difficult to believe that those in front of the fovea do not belong to the floor plate also. These pigmented columnar cells are probably the cells which laterugive rise to the ependyma cells and raphé fibers.

In Acanthias of 14 mm. the brain floor in the median plane consists of one layer of columnar cells up to a point near the mammillary recess, while the mammillary, infundibular, and optic chiasma regions are filled with neuroblasts many of which are pear-shaped and show the axone process. The brain floor in front of the tuberculum posterius seems to be constituted chiefly of neuroblasts in all the embryos studied.

5. The question of the relation of the anterior end of the notochord to the brain is very difficult, as Dr. Kingsbury recognizes. The extent of the notochord itself is uncertain, since it frequently blends with the prechordal mesoderm. The unequal growth of the brain producing sharp flexures moves the parts with reference to one another in successive stages of development.

Embryos of' Amblystoma illustrate in a striking way the shifting of position of the anterior end of the notochord with reference to the parts of the brain. In embryos of about the stage shown in figure 29 the notochord ends beneath the medulla oblongata far behind the cerebellum. Numerous older embryos, one of which is shown in figure 32, have the notochord in contact with the hypophysis and floor of the diencephalon, as a result of the brain flexure, although the end of the notochord still lies beneath the medulla oblongata. But now, early embryos in which the brain flexure merely follows the curvature of the egg (fig. 28) show the notochord following the same curvature and ending relatively far forward in contact with a thickening of the brain floor. This thickening contains a median sagittal cleft which is a ventro-caudal keel—like prolongation of the forebrain ventricle. This thickening with its ventricle represents the early downgrowth of the infundibular and mammillary regions. The thick walls contain many more mitotic figures than are found elsewhere in the brain walls at this stage. By the time those structures have clearl‘y differentiated, the growth of brain, myotomes and enteron have pushed the forebrain vesicle forward far from the end of the notochord, as shown in figure 29. In these early stages the end of the notochord lies close behind the mammillary recess.

6. The structure of the brain wall interests us here because of the presence of the nucleus and root of the oculomotor nerve rostral to the fovea isthmi. In Acanthias, Amia, Acipenser, Polyodon, Amblystoma, Necturus, Chelydra, and in human embryos the fovea is found just caudal to the third nerve. Neal (’19, figs. 7, 8) shows the fovea separating the nuclei of the third and fourth nerves. He says, “The lateral recess in the floor of the brain lies within the limits of the midbrain vesicle and does not indicate a boundary between the primary brain divisions.” Herrick (’21, fig. 65) has indicated that it lies between the third and fourth nerves in Necturus. In Acanthias, Amia, Amblystoma (fig. 32) and Chelydra the fovea is continued by a sulcus in the lateral wall which leads up into the caudal part of the tectum mesencephali. Whether the fovea and sulcus isthmi are considered as internal marks of the isthmus (Herrick) or are assigned to the mesencephalon (Neal) as their position seems to indicate, need not be discussed here. At least the greater part of the base of the midbrain including the nuclei of the third nerve lies in front of the fovea. Kingsbury clearly regards this as a region in which the basal plates of His are confluent in front of the blastepore, and notices the location here of a nucleus centralis nervi oculomotorii and other median nuclei and the existence of crossed origins of oculomotor root fibers as evidence of continuity between the two lateral halves. In accordance with this the midbrain must be regarded as part of the prechordal brain and the oculomotor nerves with the premandibular myotomes must be assigned to that preaxial common mass in front of the concrescence area which Kingsbury clearly recognized, but which His, Hertwig and Minot only vaguely appreciated. This means that the first segment of Neal contains the forebrain vesicle, eyes, olfactory organs, palaeostoma, infundibulum—hypophysis, and prechordal mesoderm but no muscles; the second segment contains midbrain, premandibular myotomes and their motor nerves (III), the ophthalmicus profundus nerve but no branchial element; the third and following segments present the essential elements of complete axial metameres. The.axis of these metameres is marked by the notochord and floor plate of the neural tube, or by the neurochordal suture.

7. The forward extent of the sulcus limitans of His which separates the sensory and motor columns in the lateral walls of the brain tube is of importance in this connection. His traced this forward to the preoptic recess. Kingsbury states with apparent approval that Schulte and Tilney (’15) have shown that the sulcus limitans in cat embryos cannot be traced farther forward than the midbrain. In my earlier work I adhered to the View of His that this sulcus ended in the preoptic recess. In 1912, however, in connection with the study of ventricular sulci in the diencephalon and telencephalon of cyclostomes it became clear that the sulcus limitans does not continue forward beyond its junction with the hypothalamic sulcus (’12 b, p. 347, and figs. 5 and 6). This led me to a study of the sulcus limitans of His in a wide range of comparative material. From the study of a large number of forms I came to the conclusion that the recognition of a sulcus limitans ending in the preoptic recess is based on the assumption that the groove which is observed ending in the preoptic recess is traceable through the midbrain and cerebellum to the undoubted sulcus limitans in the medulla oblongata. I was unable to trace the continuity of these sulci. I have recently reviewed the same material and studied other materials and am quite unable to find any clear evidence that the sulcus in the hindbrain and cord regions which is undoubtedly the sulcus limitans of His continues farther forward than the mammillary recess in any form studied.

In the brains of adult fishes the sulcus limitans of His is very clearly identified in the medulla oblongata. It forms the immediate ventral boundary of the vagal and facial lobes and is perfectly evident as far forward as the sensory root of the VII nerve. In Squalus acanthias and Scyllium the sulcus limitans is lost behind the cerebellum. A little lateral to it appears a deeper groove which marks the angle between the cerebellum and lateral wall. If this be considered the continuation of the sulcus limitans, it becomes broad and shallow and disappears at the level of the III nerve. A little farther forward the groove seems to reappear and goes up above the ridge containing the fibers of the posterior commissure to end in the tectum mesencephali.

In young embryos of Squalus and Scyllium (from 2.5 to 8.4 mm.) there is no sulcus limitans recognizable. In Squalus of 20.5 mm. a prominent groove in the medulla oblongata which one might suppose was the sulcus limitans runs forward to end in the dorsal wall of the cerebellum, behind the isthmus. In Squalus of 33.1 mm. (fig. 23) a small groove is clearly seen at the level of the VII and V nerves separating the sensory and motor nuclei of those nerves. This answers to the definition of the sulcus limitans. It is indistinct between the VII and V nerves and disappears a short distance forward from the V nerve. Two much more prominent grooves are seen in this region, one dorsal to the sensory nuclei of VII and V and one ventral to the motor nuclei of these nerves. Obviously neither of these is the sulcus limitans, but it would be easy for one who did not give careful consideration to their relations to the nerve roots and nuclei, to mistake one of those grooves for the sulcus limitans.

In adult Acipenser the sulcus limitans is lost at the root of VII. At the V nerve a new groove appears which goes up behind the posterior commissure as in selachians. In adult Amia the conditions are about the same. In Amia of 11 mm. and 25 mm. the sulcus limitans is lost in the auricular lobes of the cerebellum. The groove which may be considered to represent it farther forward stops behind the fovea isthmi. In Lepidosteus of 25 mm. likewise the sulcus can be followed forward past the V nerve but ends in the auricular lobe of the cerebellum. Although in these young ganoids a small sharp ependymal groove runs back from the preoptic recess, which can be followed either as a groove or thickening of ependyma to the isthmus (Johnston, ’11 b, p. 522), behind the isthmus this groove enters the dorsal part of the cerebellar ventricle and does not meet the sulcus limitans. Moreover, after seeing the complexity of ventricular grooves present in the diencephalon and mesencephalon, I would not be willing to say that this slender ependymal groove throughout this whole distance should be regarded as an entity distinct from the other grooves which it crosses between the preoptic recess and the isthmus. In Ameiurus of 25 mm. the sulcus limitans merges with the groove between cerebellum and lateral wall, rises to the dorsal part of the ventricle and disappears beneath the cerebellum.

In older stages of Amblystoma larvae what appears to be the sulcus limitans of the medulla oblongata enters the auricular lobe of the cerebellum. In earlier stages it is clear that the sulcus is continued through the cerebellum. This continuation groove in all stages descends in the lateral wall and ends at the side of the fovea isthmi. The two grooves give to the section of the ventricle the form of an inverted Y, between the arms of which lies the fovea, or the three are merged into a broad rounded fovea.

In embryos of Chelydra serpentina of 81/2 mm. carapace the sulcus which seems to represent the sulcus limitans in the cerebellar segment runs forward into the mammillary recess.

Human embryos of the following stages have been studied: C. R. length 9, 11, 13, 18, 19, 20, 21, 22, 24, 25 mm. The embryos of 9, 11, and 13 mm. are cut in such a plane that the course of the sulcus limitans can be studied only in models. The 9 mm. embryo was received in a perfectly fresh condit on and was admirably fixed, cut in a faultless series and well stained. Of this brain I possess a model made by an assistant years ago for an entirely different purpose and before this question came up for consideration. It is an excellent piece of workmanship and shows the grooves and ridges on the internal surface of the brain very clearly. The sulcus limitans has a somewhat zig-zag course across the transverse ‘neuromeric’ grooves in the wall of the medulla oblongata. I should be somewhat doubtful whether the groove which passes forward through the cerebellum and midbrain is really a continuation of the sulcus limitans. Granted that it is, this groove comes into close relation with the mammillary recess. From the mammillary recess a groove extends forward to meet the groove which rises from the preoptic recess to the interventricular forearm. While this model might bear the usual interpretation, I do not think that it would lead anyone to conclude that the sulcus limitans extends to the preoptic recess, unless the observer were influenced by a preconception based on the description of His. It is clear, however, that the longitudinal sulcus found in the cerebellar segment extends through the midbrain as far as the mammillary recess.

In the older embryos mentioned the sulcus limitans crosses the deepest of the neuromeric grooves at the level of the pontile flexure and then bends dorsad to end in the roof of the cerebellum behind the decussation of the IV nerve. In other words, the sulcus limitans merges with the transverse groove of the cerebellar neuromere and is lost.

So far as independent evidence from the sulcus limitans is concerned, then, the materials at my disposal do not make clear that it can be traced forward with certainty beyond the level of the trigeminus. There is a serious question whether the sulcus limitans has been properly identified in all studies of this subject. A sharply marked sulcus in the cerebellum and forward is not often present in young embryos. There is a tendency for the sulcus limitans to merge with one of the transverse grooves and then it has the appearance of ending in the roof (Squalus, Amei— urus, man) or the floor (Amblystoma). The utmost forward extent which the writer could admit on the basis of his material would be the mammillary recess (Chelydra and 9 mm. human embryos). The existence of sensory and motor centers in the midbrain comparable with those of the medulla oblongata and spinal cord would lead us to expect that a sulcus limitans would be found in the midbrain.

8. The writer is inclined to believe that all thefacts at present known are best explained by regarding the region of the midbrain, oculomotor nerves, and premandibular myotomes as a transitional region between the pre—metameric region (Neal’s segment I) and the true metameric region beginning with the segment of the cerebellum and mandibular myotome and arch (Neal’s segment III). In this second, transitional segment some of the elements of a complete metamere are lacking. The midbrain floor includes the nuclei of the most anterior pair of motor nerves and is traversed for a part of its length by the neurochordal suture. This is represented by the fovea isthmi and by columnar floor plate cells in the young embryos and raphé fibers in older stages continuing forward in the rostral wall of the fovea. Thus the midbrain segment includes the extreme anterior angle of the fused blastopore lips and also the confluent basal plates of the neural tube and confluent prechordal plate of the mesoderm, these being formed from the neural ectoderm and the entoderm which encircled the blastopore in front. At or near the tuberculum posterius occurs a sharp transition in the structure of the brain floor which marks the boundary between basal plates and alar plates of His. This is further evidenced by the convergence of the sulcus limitans of His, in the forms in which it is most fully preserved, at the mammillary region. The alar plates of His therefore give rise to the larger dorsal portion of the midbrain and the entire forebrain (diencephalon and telencephalon). Along With this exclusively sensory forebrain there entered into the premetameric region a non-muscle-forming mass of mesoderm including the prechordal plate and the anterior head cavities of selachians and the narrow preoral entoderm which served for the primitive connection of the enteron with the ectoderm (palaeostoma).

These conclusions are in essential agreement with the results of Neal on head segmentation which he has concisely summarized in his 1914 paper. In a later paper (’18) Neal has brought together reasons for considering the eye—muscle segments as members of a continuous series of metameres in the vertebrate body. My conclusions are also in agreement with Kingsbury in all important respects.

For the purpose of studying the evolution of the cerebral hemispheres, the general result of this survey of head segmentation of importance to us is that the embryonic forebrain Vesicle from which the telencephalon‘ and dieneephalon are derived is itself formed from a part of the neural plate which originally lay anterior to the cephalic end of the neurochordal seam (primitive blastopore) and consisted of nervous ectodermal tissue comparable to the sensory alar plates of His. Whether the supposition that the midbrain with the premandibular Inyotomes and the III nerves is penetrated by the cephalic end of the neurochordal seam is correct or not is immaterial to the further consideration of the forebrain

  • 1 In a recent study of Arnblystoma embryos Burr (’22) has come to the conclusion that: “Between the fovea isthrni and the preoptic recess the midventral portion of the neural plate is occupied by the continuity of the lateral basal laminae. Between the preoptic recess and the lamina terminalis the mid—ventral portion of the neural plate is occupied by the terminal ridge, the continuity of the lateral alar plates.”

It is necessary to correct the following statement of results attributed to the present writer: “The work of Johnston (’09), based on a comparative study of the forebrain vesicle in verteb1'ates, she“ ed clearly that the ventral lip of the neuropore was incorporaterl into the brain as the terminal ridge lying between the lamina terminalis anterior] y and the chiasrratie ridge posteriorly through the fusion of the lateral lip of the blastopore (sie., neuropore int ended 27). Ie concluded that the preoptic recess, utarking the terrnination of th« sulcus limitans and separating the terminal ridge from the chiasmatic ridge, reprerented the 1reeting»point of the roof and floor plate and hence the anterior end of the neural tube” (Burr, p. 278). In the paper referred to the writer stated: “After the brain is separated from the ectoderm the terminal ridge forms a distinct fold convex toward the ventricle, which in later stages is occupied by the fibers of the optic tracts in the chiasma” (p. 491; also p. 481). “The anterior boundary of the neural plate is formed by a transverse ridge, the terminal ridge which is continuous with the neural folds bounding the neural plate laterally. This terminal ridge is clearly seen in successive stages and is readily followed up to the time when the optic ehiasma is formed in it” (p. 504). “lrlarly stages show clearly that the basal axis of the brain ends not in the Basilarleiste but in the terminal ridge in which later the optic chiasn‘a appears” (p. 508). “The optic chiasma is formed in the terminal ridge and therefore occupies the anterior border of the brain floor” (p. 532).

The terminal ridge does not lie between the lamina terminalis and the ehiasmatie ridge, but is the ehiasmatie ridge.‘ The terminal ridge is not formed through the fusion of the lips of the neuropore but is the persistent neural ridge which bounds the neural plate anteriorly. The preoptic recess cannot separate the terminal ridge and ehiasmatie ridge which are one and the same thing. The body \\hich Burr figures as the terminal ridge is only a transitory mass of neural ridge cells caused by the coming together of the lips of the neuropore. It later thins out and forirs part of the lamina terminalis. The preoptie recess is form ed by thinning of the lo‘v\ er part of this adjacent to the true terminal or ehiasmatie ridge. My statements on these points were quite clear and explicit and they have been correctly understood and

9. The neural plate anterior to the primitive streak consisted of unsegmented ectoderm which primitively surrounded the anterior border of the blastopore. Throughout the head region the boundary of the neural plate was at first probably indistinct or irregular. Not only are nerve ganglia separated off from its border during the ontogeny, but such structures as the acusticolateral placodes, the visceral ganglion placodes, and the olfactory placode were formed from ectoderm more or less closely associated with the border of the neural plate. The recent description by Stone (’22) of the general cutaneous components of the cranial nerves in Amblystoma arising from ectodermal placodes further emphasizes the conception of the neural ectoderm as a broad irregular and undefined area only a part, and probably a variable part, of which enters into the neural plate and rolls up into a neural tube. The anterior boundary of this neural plate in the ontogeny is formed by the confluence of the neural ridge in the terminal ridge and this terminal ridge comes to be occupied by

quoted by Kingsbury ('20, ’22). Burr's further statement that, “In any event, it seems evident that this terminal ridge is svbsequently occupied by decussating fibers of the anterior commissurc” is erroneous. gllis so—called terminal ridge disappears and the bed of the anterior commissure is formed by a new thickening of the lamina terminalis formed by invasion from the striatal ridges.

Burr tried by experiment to determine the fate of the lower border of the neuropore. His l\Iile—blue sulphate experiment, however, was begun much too late for this purpose. The neuropore was already drawingtogether and the transitory plate which he calls terminal riige had already been formed before the experiment was begun. The Nile-hlue sulphate would need to be applied to the terminal ridge in the open neural plate stages. Burr states (p. 283) that, “The fundamental arrangements of the longitudinal columns of Iiis in the rostral end of the neural tube were as indicated by Kingsbury (’20) in his figure 6, ” but differs from hingsbury in assigning the optic ehiasma and hypothalamus to the basal plates. Lingsbury (’20, pp. 128-129) says, “ It is in my opinion a valid objection to the generally accepted interpretation that it includes in the basal, primary motor lamina optic ehiasma and hypothalamus —regions which possess no such significance. The motor zone ceases with the floor of the midbrain. . . . The telencephalon and diencephalon would then be entirely developed out of alar-plate material.” In his second paper Kingsbury (’22) states that the boundary between basilar and alar plates clearly lies in the region of the mammillary recess.

With regard to the first of the two conclusions above quoted, the w hole preceding discussion is opposed to the view that the hypothalamus and optic chiasma are derived frorn the basal laminae of I-l is. With regard to the second conclusion, the space between the preoptic recess and thelamina terminalis is nil. The terminal ridge is behind the preoptic recess.

the optic chiasma, as before stated. The considerations regarding the relation of the neural plate to the blastopore brought forward by Kingsbury do not in anyway affect this statement regarding the anterior end of the neural plate and tube.

In two previous papers (’09 a, ’09 b) the writer has discussed at length the origin of the lateral eyes in the ontogeny of several classes of vertebrates showing: a) that they arise as pits near the lateral borders of the neural plate; b) that these pits are connected with one another by a groove running across the median plane just behind the terminal ridge, forming there a primitive optic recess, which persists as the recessus postopticus; c) that by the development of an obliquely placed optic ridge running upward and backward from the terminal ridge in the wall of the neural tube the optic cups come to be separated from the primitive optic groove and connected by the optic stalks with a recess in front of the optic chiasma in the sphere of the lamina terminalis, the preoptic recess. The oblique optic ridge comes to serve as the bed of the optic tracts in the wall of the thalamus. The derivation of the -optic anlagen from lateral areas in the neural plate is supported by Bartelmez (’22) from the study of human embryos. The pineal eye or epiphysis seems also to be formed from the lateral border of the pre—metameric neural plate.

Kingsbury (’20, ’22) expresses some doubt whether the postoptic recess is independent from the infundibular recess. That these two recesses exist side by side both before and after the separation of the optic stalk from the primitive optic groove is clear from the accompanying drawings and photographs (figs. 7, 25, 26, 27, 31).

A part of the primitive nervous ectoderm which did not enter into the neural plate but lay close to its anterior border gave rise to the olfactory organ. The numerous cells collected in the olfactory placode perhaps bear some relation to the general sensory nerves of the skin which in lower vertebrates are sensitive to salt and sour. It may be supposed that here adjacent to the anterior end of the neural plate receptive cells of a general chemical sense developed a sensitivity to exceedingly minute quantities of the stimulant and hence became fit to act as distance receptors in the detection of food. If, and so long as the neurenteric canal served as a channel for the ingestion of food the olfactory organ in its position close to the lips of the anterior neuropore may have functioned to test food about to be ingested. After the formation of the palaeostoma and the closing of the neuropore, the functions of a distance receptor became characteristic of the olfactory organs.

In the most primitive living fishes the olfactory centers are of great size and have pouched out laterally to form the beginning of the cerebral hemispheres.

In the pre-metameric neural plate there came about a localization of function of a different pattern from that which produced the four functional columns in the metameric neural tube. In this region the peripheral effective mechanisms were absent and there was no demand for effective centers. What centers or functional areas should be formed and how they should be arranged in the pre-metameric nervous system were determined by three factors: the visual organ, the olfactory organ, and the afferent and efferent tracts connecting the pre-metameric with the metameric neural tube.

The first indications as to the probable primitive arrangement in the neural tube of the forebrain segment are found in the disposition of afferent tracts entering from behind. This point has been discussed in a previous paper (’10 b). In the lowest vertebrates correlating tracts arising from the tactual and proprioceptive centers in the cord and medulla oblongata enter the dorsal part of the diencephalon. fibers from the gustatory (and general visceral sensory) centers enter the hypothalamus.‘ These facts seem to show that at least in the diencephalon there was a retention of the relative position of the somatic and visceral receptive columns, but that the somatic column was much the larger of the two.

10. This view is supported by the relations of the eyes and the optic tracts. The retinas are derived from areas near the border -of the neural plate which clearly fall within the somatic sensory columns as outlined in figure 1. This is true whether the view of the writer (’06, ’09 a) or that of Stockard (’13) regarding the locus of the eye—anlage be accepted. They appear to owe their origin to the multiplication in these areas of the ectodermal cells which are sensitive to light and the heightening of their sensitivity. The fibers which go out from these retinal areas (optic tracts) enter the dorsal parts of ‘the diencephalon and mesencephalon. A large part of these fibers run along the anterior border of the neural plate and cross to the opposite side through the terminal ridge (optic chiasma). If the considerations regarding the extent of the floor plate and epichordal brain be accepted, it is clear that the retinas and the entire course of their central tracts lie wholly within the somatic sensory columns and the area of their confluence across the median plane. Moreover, the centers for the optic tracts in the diencephalon and especially the mesencephalon are in very close relation to the centers for the central tracts of the proprioceptive, tactual, and temperature senses. The grounds for considering the retina and its centers as derivatives of the somatic receptive column in the neural tube are clearer than when they were discussed by the writer in 1905, 1906, and 1909. It should be said, however, that the comparison of the optic tracts with the lemniscus system, because of their crossing in the brain floor, falls to the ground.

We may go one step further and say that in all fishes except cyclostomes, which have not yet been adequately studied, there exists a continuation of the somatic sensory column of the thalamus into the lateral wall of the telencephalon and that fibers arising in the tactual and visual centers of the thalamus enter this region of the telencephalon (Johnston, ’11 a, b). This fact offers strong presumptive evidence that the area in question is a primary part of the somatic receptive column. It has recently been shown by Holmgrcn (’22) that there exists in selachians a well—defined area in the roof of each hemisphere which corresponds to the dorsal or general pallium of reptiles and mammals. My own observations seem to show that this area is continuous with the somatic area in the lateral wall of the telencephalon medium from which the lentiform nucleus develops and that both are related to the corpus callosum in selachians. Thus it appears that the forebrain segment possesses in the fishes generally a clearly recognizable continuation of the somatic receptive column and that in selachians this shows the essential characters of a true general palliumz 1) reception of correlating tracts from tactual, visual, and proprioceptive centers in the metameric brain, 2) morphological unity and clear demarcation from other telencefhalic centers (Holmgrcn), and 3) the possession of a dorsal or pallial commissure independent of the hippocampal commissure and comparable with the corpus callosum of mammals. The presence of this somatic correlating organ in the telen— cephalon of selachians, the most primitive of the true fishes, is strong evidence that the anlage cf the somatic pallium is present from the beginning of vertebrate evolution. The somatic pallium is as old as the olfactory pallium.

11. The Visceral receptive column in the typical metameres is the center for receiving general impressions (tactual, thermal, et al.) from the visceral mucosa and muscle—sense impressions from the visceral musculature. To this column also were carried impressions from taste buds when they developed in the entoder— mal lining of branchial and mouth cavities. When taste buds appeared in the skin they also were supplied by fibers of the visceral receptive nerve—components and all gustatory impressions entered the visceral receptive column.

The taste buds are collections of epithelial cells which are sensitive to chemical stimuli. They differ from the olfactory sense cells in that they require stronger stimulation (i. e., more concentrated solutions) and that they do not send axones into the brain. In the fishes both gustatory and olfactory organs serve important functions in relation to feeding, the olfactory organ in detecting slight or distant indications of food, the gustatory organ in testing the food-object after it has been seized or taken into the mouth. The use of the gustatory" organs on the outside of the head and body in the search for food as in many bony fishes is to be regarded as a secondary development. The entrance of the gustatory fibers led to the great enlargement of the visceral receptive column in the region of the medulla oblongata (Johnston, ’02 c). From the hypertrophied visceral lobe in the oblongata correlating tracts ran forward to a complex of nuclei at about the level of the isthmus and from these to the hypothalamus (Herrick, ’05; Johnston, ’11 b). The area in which these tracts end (hypothalamus) appears to be in the territory of the visceral receptive column.

Now it is important to notice that the Whole gustatory system is a later addition to the general visceral receptive system. In other Words there existed a receptive center for general visceral impressions before the taste buds were developed. It is not surprising that when the taste buds appeared in the entodermal lining of the alimentary canal the fibers which supplied them should find their central endings in the visceral receptive column.

Indeed, it would have been very surprising if a separate central column had been developed for the reception of impulses from specialized sense cells in the general visceral surfaces.

Older than the gustatory system, the olfactory organ was the first special sense organ serving for the finding of food and for testing the quality of the water in which the animal lived. It may be supposed that substances in the Water (e.g., dissolved minerals or gases) unfavorable to respiration or otherwise deleterious to the life of the animal, would first be detected by the olfactory organ. This is probably the oldest of the special sense organs. The axones of the receptive cells entered the border of the neural plate. When the neuropore closed up the walls of the neural tube which received these fibers hypertrophied and pouched out in response to the incoming stimuli. We cannot say with certainty either that the somatic and visceral receptive columns remained distinct in this anterior end of the neural plate or that the olfactory centers were derived exclusively from either one. The nervus olfactorius and nervus terminalis were the only primary receptive nerves related to the Whole forebrain. The one has its receptive cells located in the ectoderm, the other, if sensory at all,-is probably general cutaneous. So far as the paradigm of functional columns in the neural tube goes, one would expect both these nerves to be related to the somatic receptive column. What is much more to the point is that the olfactory organ has stimulated the development in the forebrain of large centers peculiar to themselves and that the classification of these with somatic or visceral systems is a matter of convenience.

This olfactory organ plays a remarkable role. Arising and remaining in the ectoderm, it primarily serves, not to direct the movements of the animal or to protect it from external objects or enemies as the tactual and visual organs do, but to give information with regard to the chemical or metabolic relations of the animal. The special Olfactory centers eventually gain relations to both visceral and somatic receptive columns and the correlating centers derived from them. In the lowest existing vertebrates the olfactory centers send the overwhelming majority of their fibers to the visceral and gustatory centers in the hypothalamus. The hypothalamus becomes in consequence the chief correlating center for olfactory and visceral impressions and the olfactory centers gain their chief directive pathways through the tracts from the hypothalamus to the medulla oblongata. Secondary in importance to these are the tracts, through the habenula, which the Writer is inclined to believe is to be regarded as a derivative of the visceral receptive column. Into the habenula, however, come tracts from various sources and here olfactory impulses enter into relation with others belonging to somatic categories. In its earlier evolution, therefore, the olfactory system establishes relations chiefly with the visceral receptive systems and gains its connections with the centers for musclecontrol through these visceral correlating centers. Because of these facts and of the apparent primary importance of the olfactory system to Visceral activities the writer was led to classify the olfactory organ and centers with the visceral receptive system (’02 c). This View has been accepted by Herrick (’22) and others. The argument about the definition -of the term visceral (Dart, ’22) I leave to others. The functional and morphological relationships are clear. A rose by any other name would smell as sweet.

A further indication of the intimate association of the olfactory with the visceral receptive column is found in the relations of the tractus pallii in fishes. In addition to the basal forebrain bundle which connects the olfactory centers with the hypothalamus, there is found in selachians, ganoids, and teleosts a sharply defined tract connecting the hypothalamus with the most highly specialized part of the olfactory centers, the portion from which the hippocampus is formed in higher vertebrates. It appears that in the basal bundle descending fibers predominate, While in the tractus pallii fibers arising in the hypothalamus and ascending to the olfactory centers predominate. In whatever tracts they run the existence of ascending fibers from the hypothalamus to the olfactory centers is of great significance. It means that the final correlating centers to which visceral and gustatory impressions are carried in the cerebral hemispheres are at the same time part of the olfactory centers. However much stress we may lay upon the peculiar and individual character of the olfactory apparatusfland the writer is more inclined than heretofore to emphasize this—~these olfactory centers serve as the end station for the highest order of correlating fibers carrying visceral and gustatory impressions. The olfactory organ is highly special and independent but its centers and central tracts are‘ most intimately associated with those of the visceral and gustatory organs.

In higher vertebrates important relations are set up between the olfactory centers and the highly specialized correlating centers of the somatic receptive systems, situated in the telencephalon. These relations are already foreshadowed in some of the most active fishes (selachians) by the interchange of fibers between the lateral olfactory area and the anlagen of the lentiform nucleus and pallium respectively. As a result of the development of these olfacto-somatic correlations the olfactory organ changes its mode of functioning. Primitively it gave rise to reflex responses having to do directly with feeding or respiration. In higher forms the olfactory impressions may give rise to such reflex actions, but to an increasing degree such immediate responses are inhibited and the impressions are correlated with those received from the organs of touch, temperature, vision, and hearing to form a whole which we call a percept of the situation or environment at the moment. In this new integration of receptive systems in higher vertebrates brought about chiefly through the evolution of the general pallium, the relationship of the olfactory organ with the somatic receptive and correlating systems comes to be more extensive, if not more intimate, than its relation with the visceral systems. The differentiation of the pyriform lobe and of the newer parts of the amygdaloid complex as a result of this olfacto—somatic relationship has been discussed in the previous section (p. 464 ff.).

B. Form Changes in the Evolution of the Hemispheres

Effect of hypertrophy of the visceral receptive column

In the medulla oblongata of fishes the visceral receptive column is an elongated ridge projecting from the lateral Wall into the ventricle. In cyclostomes and selachians this is a rather slender ridge without great enlargement in any part. It is usually largest at about the level of the first roots of the vagus nerve. In some ganoids and teleosts this ridge is greatly enlarged, bulging out intogthe ventricle and pushing up dorsally until in many cases it thrusts the somatic receptive column to the side. Those changes are especially noticeable in the anterior part of the lobe related to the facial nerve. Those fishes in which this lobe serves chiefly general visceral functions have the vagal portion of the lobe largest, while those fishes in which the taste buds are very numerous have the facial lobe larger than the vagal. In extreme cases the facial lobes not only push dorsally as mentioned, but meet and fuse across the ventricle in the median plane. Some of these conditions are illustrated in previous papers (’06, ’11 b).

A similar condition is found in the telencephalon, which helps to explain the form of the hemisphere and the relative position of the functional centers in the telencephalon in early stages of evolution. In Petromyzon a part of the olfactory centers are evaginated into a hollow hemisphere, but that portion into which the tractus pallii comes up from the hypothalamus remains standing vertically in the lateral wall of the telencephalon medium. This structure-has been described as the primordium hippocampi (’12 b). Some ground has been given for regarding the ganglion habenulae as an hypertrophied part of the visceral receptive column which has risen dorsally in a similar manner (’11 a).

The tendency for the visceral receptive tissues to rise dorsally in the ventricular wall of the telencephalon accounts for the striking eversion in ganoids and teleosts and for certain relationships in other vertebrates as well. The somatic receptive column has not been recognized in the telencephalon of cyclostomes but in other fishes a condition similar to that described for Petromyzon is found in the telencephalon medium. In selachians along the dorsal border of this region there is a slender continuation of the primordium hippocampi imbedding the posterior pallial commissure and olfacto—habenular tracts, while the ventricular surface down to the recessus preopticus is an olfactory center. Clearly the olfactory centers in the telencephalon medium of sela— chains present the same relations as in cyclostomes. Just the same situation is found in ganoids and teleosts in the corresponding region (’11 b). ~ In selachians, ganoids, and teleosts the somatic receptive column in this same region is pushed outward and lies on the lateral surface. In the rising dorsally of the olfactory centers and tendency toward eversion the telencephalon medium closely resembles the medulla oblongata in forms with hypertrophied facial or Vagal lobe.

In Chimaera (Johnston, ’10d) the telencephalon medium is very greatly elongated and consists almost wholly of fiber tracts. In the massive portion of the forebrain the posterior part is slightly everted like that of teleosts, the anterior part is evaginated and possesses a massive roof like that of selachians. In ganoids and teleosts the tendency to eversion goes farther than in Chimaera so that the whole telencephalon up to the olfactory bulbs is everted like the telencephalon medium of cyclostomes. This is not strictly true, however, in all these fishes, since in Amia (Johnston, ’11 b) there is found a true dorsal commissure comparable to the hippocampal commissure of selachians. Practically, however, in ganoids and teleosts the olfactory bulb is the only part which is fully evaginated as hemisphere, the remainder being essentially everted telencephalon medium.

The selachian telencephalon presents a relatively enormous development of the medial olfactory nucleus, which‘ causes the olfactory peduncles to be carried far laterally. Closely related to this is the large size of the tuberculum olfactorium and the pallium. The everted condition of the ganoid and teleost telencephalon is due to the retention and exaggeration of a primitive eversion which belonged to the telencephalon medium. This condition was probably brought about under the influence of the gustatory organs and the tractus pallii which reach their greatest development in these fishes. The continued evagination and the formation of a massive roof in selachians was apparently due to the predominance of the olfactory and somatic systems. The medial nucleus, the tuberculum, and the hippocampal formation in the roof are primarily olfactory. The tractus pallii (gustatory) which comes up to the roof is relatively much smaller in selachians than in ganoids and teleosts. Greater volume for the olfactory centers in selachians was secured by evagination and as the gustatory correlation center was not hypertrophied it exercised only a subordinate influence in determining the form of the brain and was carried into a position in the roof. Since this center had already gained in cyclostomes a position adjacent to the choroid plexus, when it was carried out by evagination in selachians into the hollow hemisphere it retained its morphologicallypdorsal position and is located in the roof. Likewise the somatic receptive column retains its primitive lateral position and appears in the lateral surface of the telencephalon medium and in a dorso—lateral position in the roof of the hemisphere (Holmgren, ’22). The greater importance of the somatic centers in selachians is probably a strong factor in producing an evaginated hemisphere in the place of an everted telencephalon medium. The expansion of the somatic area in the lateral wall favored a continuation of the evagination process by which the olfactory bulb was formed. It would be of the greatest interest to know the morphology and history of the centers in ganoids and teleosts corresponding to the general pallium described by Holmgren.

Professor Holmgren’s very valuable paper (’22) reveals an organization in the selachian forebrain which heretofore has been regarded as characteristic of that of higher vertebrates. He recognizes well-defined areas in the developing forebrain of Acanthias which are clearly comparable morphologically with well-known areas in the reptilian and mammalian brain. The identification of an hippocampal formation and a general cortex and the demonstration that each of these is continuous with its fellow of the opposite side through a bridge of tissue across the median plane, are invaluable additions to our knowledge of forebrain morphology.

It seems to the writer that Professor Holmgren. has been unfortunate in his identification of the pyriform cortex. This cortex as described lies in the brain wall in front of the olfactory peduncle extending from the peduncle along the lateral and rostral wall of the frontal evagination or pouch. This topographically rostral region clearly belongs to the morphological medial wall of the brain, and its constituent parts are to be compared with structures lying in the medial or septal wall in the brain of a chelonian, marsupial, or mammal.

This view is clearly stated by Professor Holmgren in the second paragraph of page 399. In response to a letter from the present writer pointing out the inconsistency between that paragraph and the interpretation of a certain area as pyriform cortex, Professor Holmgren inserted in the manuscript the third paragraph on page 399, setting forth the hypothesis that the primary situation of the bulbus is a lateral one and that except in selachians the bulbus is dislocated during ontogeny to a terminal position. This hypothesis I cannot accept. It is undoubtedly true that the bulbus pediment moves from a more lateral to a more rostral position during ontogeny in various forms including some selachians, but I know no reason for considering this to be a dislocation or relative shifting of parts. The different (topographical) position of the bulbus in different forms is due to the differences in relative size of development of the centers or areas surrounding it. The bulbus is placed far laterally in selachians because of the relatively enormous sizeof the medial olfactory area, the tuberculum olfactorium, and the pallium. The apparent wandering of the bulbus in other forms toward the median plane which brings it into a (topographical) terminal position is due to the reduction of the medial olfactory center and tuberculum olfactorium and the increase of the lateral olfactory area and spreading of the general cortex laterally. No wandering of the bulbus takes place, but only a relative reduction of centers medial to it and enlargement of centers lateral to it. By common usage the bulbus olfactorius has been accepted as the mark of the morphological frontal pole, and it marks this pole just as truly in selachians where it is lateral in position as it does in teleosts and reptiles where it is rostral in position (I would not apply the word ‘terminal’ to the olfactory bulb because of confusion with the true anterior end of the brain at the terminal ridge or optic chiasma.)

The author describes a zona limitans medialis bounding the hippocampal formation medially, and a zona limitans lateralis bounding the pyriform cortex laterally. The relations of these two zonae limitantes is thus clearly described:

“Below the ventricle is a clear space crossing over the ventral part of the brain and delimiting the nucleus olfactorius lateralis and the tuberculum olfactorium cortex which lie below it. This space, poor in cells, medially bends up dorsad to the medial border of the hippocampal rudiment. The dorsal part of this clear space in the future development is destined to become the zona limitans medialis (2. l. m.): The lateral part of the space in question also bends dorsad to the lateral border of the pyriform lobe and forms the zona limitans lateralis (2. l. l.). Thus at the front pole of the forebrain Vesicle the zona limitans medialis and lateralis unite” (p. 405). On page 406 it is shown that what is called the zona limitans lateralis bends upward to pass over the olfactory crus. The reader of Holmgren’s paper will do well to remember that figures 5 to 15 and 17 to 20 from transverse sections of the forebrain of Acanthias embryos are taken in front of the bulbus olfactorius and therefore represent only the rostral pouch of the medial brain wall.

Thus the pyriform cortex of Holmgren lies in the topographical rostral and lateral wall of the rostral evagination of the hemisphere in front of the olfactory bulb. It ends at the level of the olfactory bulb in contact with the rostral end of the lateral olfactory nucleus. Now in reptiles, marsupials, and mammals the pyriform lobe lies in the lateral wall of the brain caudal to the olfactory bulb and represents a portion of the lateral olfactory nucleus lying along the lateral border of the general pallium and modified in function and structure because of that relationship. The ‘pyriform lobe’ here described lies in the topographically rostral wall in front of the bulbus olfactorius and therefore in the morphologically medial wall. Moreover, this ‘pyriform cortex ’ is continuous around the rostral surface with the hippocampal formation. Thus in the 4.5 cm. stage, “in the foremost part of the forebrain the pyriform cortex and the hippocampal cortex rudiment are constantly continuous with each other” (Holmgren, p. 403). This foremost part or pole of the brain where also “the zona limitans medialis and lateralis unite” is, however, the topographical and not the morphological frontal pole.

All these facts show clearly that what is found in Professor Holmgren’s sections is a continuous mass of cells bordering the general pallium medially and rostrally as far as the bulbus olfactorius. To the medial portion of this he has rightly given the name hippocampal formation. The lateral part also belongs to the hippocampal formation and throughout its extent it is bounded ventrally by a cell-free space which lies wholly in the morphological medial wall and is the zona limitans medialis. A zona limitans lateralis, if present, is represented by the doubtful or indistinct cell-free space in the lateral wall caudal to the bulbus olfactorius.

A very important relation of the hippocampal formation which has been clearly recognized since Elliot Smith ’s work on the marsupials is obscured by Professor Holmgren’s description for the selachians. I refer to the fact that the hippocampal formation extends right forward to the olfactory crus. This is clearly true also in Acanthias if the region described as pyriform cortex be interpreted as part of the hippocampal formation. Holmgren (p. 399) regards this extension of the hippocampal formation directly to the olfactory crus as a secondary condition, but this is only because he names this rostral part of the hippocampal formation in selachians the pyriform cortex. Of the true pyriform lobe developed from the lateral olfactory area Holmgren takes no cognizance.

The accompanying drawings (figs. 2 to 18) from a series of transverse sections and from a series of sagittal sections of the forebrain of Squalus acanthias of about 48 mm. serve to confirm Holmgren ’s description of pallial areas. The sagittal sections show especially well the continuity of the hippocampal formation from the median commissure bridge right round the rostral pouch to the base of the olfactory bulb. Further comments on the pallial areas Which appear in these figures are made on a later page.

The interpretation here suggested makes possible a direct and simple comparison of the general pallium and hippocampal formation of Acanthias with those of a primitive reptile such as the turtle.

The amphibian telencephalon is characterized by an evagination nearly as complete as that of reptiles together with an arrangement of commissures closely comparable with that of ganoids and teleosts. The fact that the hippocampal commissure crosses beneath the ventricle shows the affiliation of the amphibians with primitive ganoid-like ancestors. That the amphibians did not follow their ancestors in having the forebrain everted is accounted for by the sharp and probably sudden reduction in the gustatory system when the amphibians came to live on land. The primitive eversion has been retained in the telencephalon medium and in the arrangement of the commissures (Johnston, ’06, ’10 c, ’11 b), but the olfactory organ, being well developed and the new mode of life demanding a large development of the somatic centers, the anterior part of the forebra n cont'nued to evaginate very much as it does in selachians and reptiles. The posterior part of the telencephalon showing ganoid type of structure remains short while the evaginated part becomes greatly elongated. The amphibian telencephalon has been determined by a peculiar set of circumstances or a peculiar series of stages and it throws little light on the evolution of the hemispheres in higher vertebrates.

The selachians on the other hand, or selachian-like fishes, are undoubtedly the ancestors of reptiles, birds, and mammals. It is unfortunate that we have not a more complete account of the telencephalon of dipnoans, but enough is known (Kerr, Smith, Holmgren) to show that they possess a true dorsal pallium and commissure comparable with the hippocampal formation of reptiles. A somatic pallium seems also to be present but less developed than that of reptiles. The dipnoan forebrain belongs clearly to the selachian-reptilian-mammalian type and not to the ganoid-teleost-amphibian side branch. (figure 90 of ’11 b requires revision on this point.) To understand. the evolution of the mammalian forebrain we must follow the changes of form and structure in selachians, dipnoans, reptiles, and marsupials and try to see the influence of changes in habits and environment in bringing about those changes in structure.

Figs. 2 to 6 Transverse sections of the telencephalon of a Squalus acanthias embryo of 47.3 mm. Scammon no. 11. X 15 diam.

Fig. 2 Section taken at the junction of telencephalon and diencephalon. The upper part of the figure shows part of the thalamus and midbrain. Note that the dense cell lamina of the basal area extends back to this level. The cell-free area at the dorsal border of the telencephalon imbeds both the posterior pallial commissure (c.p.p.) and the stria medullaris. '

Fig. 3 On the left side the section touches the caudal surface of the olfactory bulb. The thick brain wall to which the bulb is attached probably includes parts of the lateral olfactory area and general pallium.

Extent or Degree of Evagination

As already noticed, in cyclostomes the bulb and part of the olfactory centers are evaginated, while the primordium hippocampi remains in the telencephalon medium. In Chimaera the evagination is somewhat greater. In ganoids and teleosts eversion becomes the dominant feature, while in amphibia, with the reduction of the gustatory apparatus, the eversion is subordinated to an evagination parallel with that in dipnoans and reptiles. In selachians most of the olfactory central mechanism is evaginated, but the telencephalon medium retains a slender continuation of the primordium hippocampi. (figs. 2, 3) and has a large part of the somatic column. The latter is destined to form the lentiform nucleus. Holmgren has not recognized this ‘somatic area’ in selachians, mistaking it for “part of an unusually large nucleus preopticus. ” Its extent is clearly defined on page 12 of ’11 a. Through this lentiform nucleus runs the primitive crus cerebri consisting of thalamic projection fibers (tactual, visual, etc.) to the somatic pallium and the descending fibers from the pallium. While the large fundamental divisions of the hemisphere can be recognized here, these parts have very different place relations in selachians and mammals. In selachians the palliumwincluding somatic and visceral parts—lies forward from the lentiform nucleus, the lateral olfactory area extends from the olfactory peduncle caudad lateral to the pallium to meet the lentiform nucleus. In the base, the anterior commissure crosses at about the caudal border of the lateral olfactory area and the tuberculum olfactorium lies forward from this commissure and connects the lateral and medial olfactory areas. In mammals the great occipital and temporal poles lie behind the level of the anterior cornmissure and lentiform nucleus and are occupied by somatic pallium and hippocampal- formation and by pyriform lobe and amygdaloid complex which have been derived from the lateral olfactory area and tuberculum olfactorium.

Fig. 4 On the left side the section passes through the olfactory bulb. The right side is still behind the bulb. The letter h marks the extreme lateral part of the hippocampal formation which meets the rostral side of the bulb. It is this which Holmgren calls the pyriform cortex.

Fig. 5 A section much farther forward, the right half of which touches the rostral surface of the bulb. The lower part of the ventricle corresponds to the ventricle of the tuberculum olfactorium in mammalian embryos. In the upper wall the hippocampal formations of the two sides meet in a median bridge as shown by Holmgren in his figure 15. The artist has not exaggerated the distinctness of the pallial areas in this specimen.

Fig. 6 A section a little farther forward than the last which shows the laminae of the general pallium meeting in the median plane. Rostral to this the general pallium disappears and the hippocampal formation extends across the upper partof the sections.

Figs. 7 to 18 Drawings from a series of sagittal sections of an embryo of Squalus acanthias of 48 mm. Scammon. X 16. figure 7 is in the median plane in the hypothalamic region; figure 8 is median at the neuroporic recess. The following sections extend to the olfactory bulb.

Figure 7 shows very clearly the distinctness of the postoptic recess and the infundibular recess. figures 8 to 18 show the pallium extending from the bridge in the median plane to the olfactory peduncle. The pallium is bounded ventrally by a continuous zona limitans medialis (Johnston, ’11 a) from the median plane to a point near the peduncle. Corresponding to this cell-free zona is a well-marked ventricular sulcus (s.l.). The bridge of pallial tissue which crosses the median plane in the lamina supraneuroporica (fig. 8, Ls.) contains both hippocampal and general pallial elements. This lamina supraneuroporica is to be compared with that Of cyclostomes (Johnston, ’12 b, fig. 23), and with that of the turtle, Opossum, bat, and man (’13 b, figs. 5, 22, 25, 8).

A short distance from the median plane (fig. 9) the general pallial and hippocampal masses become distinct and are clearly distinguished through all the following figures to the base of the Olfactory bulb. The hippocampal formation is a continuous mass from the median plane to the bulb. It is the lateral part of this which Holmgren calls the pyriform lobe. It is instead to be compared with the rostral extension‘ Of the hippocampal formation in reptiles, marsupials, and mammals which reaches to the Olfactory bulb.

Not only is the great size of the somatic pallium the chief characteristic of the mammalian forebrain, but the growth of this somatic pallium has been the dominating influence in the rearrangement of parts in the telencephalon from the selachians onward. The increasing importance of the thalamic radiations and the somatic pallium led to the enlargement of the lateral olfactory area and the evolution of the pyriform lobe and amygdaloid complex as described in section IV. At the same time the simple function of the olfactory apparatus as a distance receptor in finding food diminished in importance and the large medial olfactory nucleus was relatively reduced. The exransion of the general pallium in its lateral position caudal to the olfactory peduncleand the reduction of the medial olfactory nucleus in front of the peduncle brought the olfactory bulb into its characteristic rostral position near the median plane.

The further expansion of the hemisphere involved a bending laterad of the crus cerebri and of the lentiform nucleus through which it passes, and a narrowing of the interventricular foramen in such a way that the medial walls of the hemispheres approach one another closely and the corpus striatum becomes the floor of the lateral ventricle. There is nothing complex or intricate about the behavior of the lentiform nucleus or the crus in the evolution of the hemisphere. The crus bends laterad because the pallium expands backward. The lentiform nucleus increases in volume because it, too, is concerned in the increased functions of correlation of which the pallium is the chief mechanism. The volume of the corpus striatum is further increased by the increased number of fibers in the crus. finally from reptiles onward the nucleus caudatus is added to the corpus striatum by immigration from the border of the pallium. The thickening of the crus led to the attenuation of the nucleus caudatus and of the bed of the stria terminalis as explained in section I. The crus and lentiform nucleus form a relatively rigid and unchanging stem attaching the hemisphere to the thalamus as long ago explained by His. What remains to be explained is the movement cf other parts due to the expansion of the somatic pallium.

As the pallium expanded backward the lateral olfactory area elongated with it and came to cover the lateral surface of the lentif( rm nucleus and crus. This condition is seen during the ontogeny of reptiles (Johnston, ’16 b). Later in the ontogeny the cells of the lateral olfactory area are pushed aside by the bulging out cf the crus and recede to the borders of the striatal area leaving the bundles of the crus visible from the lateral surface. In marsupials the lateral olfactory area (lobus pyriformis) maintains its position covering the striatum laterally in the adult. In this condition is.gradual1y changed by the growth of the 1 allium so that the lobus pyriformis is pushed down on to the basal surface and the striatum is covered by pallium, thus producing the true insula. finally, the insula is covered in by the oyercula which bound the sylvian fissure.

The expansion of the sr-rr..atic pallium carried the hippocampal fornatir n with it _more or less mechanically. The hippocampus in re} tiles forms a broad curved band along the medial border of the <’orsal pallium and bending down into the temporal pole. Later. '~hrough the pressure from the general pallium and the thickening corpus callosum the dorsal hippocampus was compressed and attenuatedas shown by Elliot Smith in his early work. The hippocampus generally in mammals maintains its continuation forward beyond the corpus callosum to the base of the olfactory peduncle.

Caudally when the hemisphere extends beyond the level of the telencephalon medium and bends down to form the temporal pole, the structures which cover the ventricular surface of the lentiform nucleus and crus become involved in this pole. For the explanation of this the fact that the nucleus caudatus bends over the crus into the temporallobe is of little importance, since this nucleus is derived from the lateral border of the pallium and merely follows that border. The structures found in selachians on the internal surface of the telencephalon medium are: the continuation of the hippocampal formation imbedding the posterior pallial commissure, the ventricular gray corresponding to the lateral olfactory area Which becomes the bed of the stria terminalis, and other olfactory gray related to the preoptic region and imbedding part of the stria medullaris (tractus olfactohabenularis). The disposition of the stria medullaris in marsupials and mammals shows that this latter tissue is not carried out into the hemisphere. Great bundles of the stria medullaris arise from the medial forebrain bundle, the preoptic region and the adjacent parts of the thalamus (see sections I to IV, figs. 23, 24). The bed of the posterior pallial commissure in reptiles is carried down over the latero-caudal surface of the internal capsule, and the posterior pallial commissure runs up from the hippocampal formation with the temporal pole components of the stria medullaris. The bundles of the stria medullaris arising from the nucleus of the lateral olfactory tract, from the lateral border of the tuberculum olfactorium, from the lateral olfactory area (pyriform cortex), and the hippocampus are still present in the temporal pole at least in marsupials (Didelphys). The bundle from the hippocampus is homologous with the posterior pallial commissure of selachians.

The olfactory tissue which becomes the bed of the stria terminalis is elongated with‘the pallium and the lateral olfactory area and bends down into the temporal pole as described in earlier sections. This tissue remains massive behind the crus and gives rise to the central and medial amygdaloid nuclei, while it becomes attenuated over the crus where the stria terminalis fibers are accompanied by only a little gray matter.

The continuity of lateral and medial olfactory areas is maintained throughout the evolution of the hemispheres through the derivatives of the basal area of the selachian, namely, the tuberculum, anterior perforate space and diagonal band. As the hemisphere expands these structures are elongated and undergo some differentiation. The tuberculum and anterior perforate space owe their importance to the fact that they enter into the olfacto—somatic correlation system. The diagonal band becomes the chief avenue of connection between the more strictly olfactory portions of the amygdaloid complex and the hippocampus and medial olfactory areas. The relations of the anterior commissure to these structures are not materially changed in the vertebrate series. Its pallial elements run beneath the internal capsule and bend back in the external capsule to the temporal pole. These are especially strongly developed in marsupials and are important in mammals. (See ’13 b, p. 402, for a discussion of the anterior commissure in relation to the corpus callosum.) The lateral olfactory component of the anterior commissure is represented by the commissure of the nucleus of the lateral olfactory tract which forms a bundle of the stria terminalis. To what extent the temporal limb of the anterior commissure carries fibers arising in amygdaloid nuclei has not been determined.

It is not within the scope of this paper to discuss the evolution of the structure of the general pallium. That the essential homologue of the general pallium was present in selachians seems established by the work of Holmgren (’22) which the writer can confirm.

Figures 2 to 6 show two points of interest: the prolongation of the basal area far back into the telencephalon medium, and the differentation of the pallial areas, each of which is confluent with its fellow across the median plane at the level of the neuroporic recess and forward. The disposition of the basal area should be compared with that of the cell—proliferation in the human embryos of 21 mm. which gives rise to the tuberculum olfactorium, diagonal band and anterior perforate space (see sections I to IV, figs. 1 to 16). figures 7 to 18 show the disposition of the pallial areas in sagittal sections of a 48 mm. Squalus. It is noteworthy that both of the pallial areas cross the median plane and that the deeper one continues around the rostral wall without break to enter the olfactory peduncle just as the hippocampal formation does in mammals. figures 19, 20, 21 show the same two pallial areas in a young pup of Scymnus. In their general arrangement the pallial areas of both Squalis and Scymnus show a fundamental likeness to the hippocampus and general pallium of reptiles, especially the more primitive turtle. The hippocampus lies just above the level of the neuroporic recess and is separated from the medial olfactory (parolfactory) area by the medial sulcus limitans hippocampi and extends to the olfactory peduncle. It is the lateral part of this structure adjacent to the olfactory peduncle with Holmgren has identified with the pyriform cortex (p. 170).

The continuity of the hippocampal and general pallial cortical areas across the median plane is very interesting in relation to the problem of cortical commissures, the hippocampal commissure and the corpus callosum. We shall await with interest Holmgren’s report on the fiber relations in embryonic selachians to determine whether the bridges of cellular tissue described are the beds respectively for the two cortical commissures.

It is important to recognize that this general pallium was established and clearly distinct from the olfactory centers in selach— ians. Elliot Smith (’10, ’19) and Herrick (’10, ’21) have 1nain— tained the View that the general pallium (neopallium) arose from olfactory area because of the penetration of thalamic radiations into regions which were primarily olfactory. Since the work of Holmgren it is no longer necessary to discuss whether the pallium has arisen in amphibians or even in dipnoans. Already in selachians the general pallium is clearly present in simple form and independent of the olfactory centers. This seems to leave no ground for the theory that the neopallium has been differentiated

Figs. 19, 20, 21 Three transverse sections of the telencephalon of Scymmus lichia of about 150 mm. length. figure 19 falls immediately in front of the interventricular foramen and behind the olfactory bulb. figures 20 and 21 fall successively farther forward, passing through the olfactory bulb. In figure 19 both hippocampal and general pallial layers enter into a thick median (commissural) bridge. The lateral olfactory area which appears lateral to the basal area (tuberculum) is properly to be compared with the pyriform lobe of higher forms. figure 20 shows the neuroporic recess in the center, at either side of which the zona limitans medialis is clearly marked. The hippocampus in figures 20 and 21 occupies the same position as it does in reptiles. All three sections show the basal area of wide extent very much as in adult selachians. The ventricular pouch (v.t.o.) descending into this area should be compared with the ventricle of the tuberculum olfactorium in mammalian and human embryos (figs. 13 and 14 in section I). out of olfactory centers under the influence of thalamic radiations penetrating those olfactory centers. The thalamic radiations from somatic centers are already present in selach ans, ganoids, and teleosts (Johnston, ’11 a, b) and they penetrate centers which are not olfactory. These centers are to be regarded as primitive somatic correlating centers which give rise to the lentiform nucleus and the general pallium in higher forms. The special commissure of the centers in selachians I have compared with the corpus callosum (’11 a).

The evolution of the general pallium proceeds more rapidly than has generally been supposed. Whereas there has been serious discussion (P. Ramon, Edinger, Meyer, Unger, Cajal) of the question whether a true representative of the general pallium is to be found in reptiles, the writer has shown (’15 b, ’16 b) not only that a histologically differentiated pallium is to be recognized, but that along the anterior and lateral border of this is found an area which is structurally characterized by thickening and by larger size of its cells, an area which upon electrical stimulation proves to be the equivalent of the excitable (‘motor’) cortex of the mammalian hemisphere.

For this area Miss Crosby (’17) proposed the term primordium neopallii and Elliot Smith (’19) gives his hearty approval to this new term. But the application of this term to the pallial thickening in reptiles is too conservative by at least a few million years. We cannot properly speak of a neopallial primordium in any animals higher in the vertebrate scale than the selachians. The pallial thickening of reptiles is not only not a primordium from which the general pallium of mammals is evolved, but is a specialized part of the general pallium, namely, the so—called motor area probably homologous with the precentral area in man. The neopallium in reptiles is by no means in a primordial stage but has already a broad extent with localization of functions.

This position of the excitable pallium in the turtle agrees well with the condition in lower mammals where, as is well known, the excitable cortex is located far forward. The larger posterior part of the cortex in reptiles or mammals develops into the general sensory, proprioceptive, visual, auditory, and associational areas.

The frontal lobe which appears to have aesthetic and emotional functions in man develops later from tissue lying between the excitable area and the olfactory bulb. The relatively enormous development of this in man brings the olfactory bulb upon the basal surface of the frontal lobe.

Fig. 22 Sagittal section of the brain floor of a sheep embryo of 22 C?) mm. Minn. Anat. 533. X 24. The photograph shows the structure of the brain floor forward from the fovea isthmi. Compare with the upper portion of Kingsbury’s figure 7. The vertical fibers are comparable to the raphé fibers in and behind the fovea but are less numerous and are infiltrated by a larger number of cells.

Fig. 23 Transverse section of the brain of Squalus acanthias embryo of 33.1 mm. at the level of the seventh nerve roots, to show the relations of the sulcus limitans of His. The section also shows a typical raphé. Scammon no. V. X 46.

While for some years the attempt to make direct comparison between the human forebrain (diencephalon and telencephalon) and that of lower vertebrates fell into disrepute, this was because of the limited knowledge (partly due to the limited technique) upon which the comparisons by early authors (Rabl-Rijckhard, Edinger, Studnicka, Osborn, and others) were based. We have now a fairly complete solution of the head problem a11d of the relation of the neural plate to the blastopore, as well as a large collection of detailed and reliable data upon the structure, neurone connections, and embryology of the anterior regions of the brain. The farther this latter work has proceeded the more clear has become the continuity of phylogenetic development of brain structure. New structures have not appeared, and although old structures have been modified, the process of evolution leading to the human brain has been one of gradual development of the elements inherent in the structures or areas laid down in very early vertebrate ancestors with gradual shifting of emphasis, growth, or decline of this or that element, all under the influence of environment and changing habits. This is true of the following structures, of each of which we now have a fairly complete history; medial and lateral olfactory centers and hippocampal formation; amygdaloid complex; corpus striatum including putamen, globus pallidus, nucleus caudatus, bed of stria terminalis; general cortex or neopallium ; and the fiber systems connecting the telencephalon with lower centers. Future studies must be directed to the details of the differentiation of these structures and their modification in response to changes in environment, habits, and peripheral sense organs.

Fig. 24 Sagittal section of the brain floor of a human embryo of 15 mm. C. R. H. 35 Minn. Anat. X 110. The photograph shows that the raphé fibers extend forward beyond the fovea. The anterior end is to the right.

Fig. 25 Sagittal section of a pig; embryo of 24 mm. Anat. 508. X 24. The section shows the raphé at the level of the pens. In the floor of the diencephalon the recessus postopticus and recessus infundibuli are c-learly distinct.

Fig. 26 Sagittal section of a pig embryo of 9 mm. in the region of the lamina terminalis and floor of the diencephalon. Anat. 464 X 60.. To show the pre- and postoptic recesses in relation to the terminal ridge which becomes the chiasmatic ridge. Note that the recessus infundibuli is quite distinct from the postoptic recess.

Fig. 27 Sagittal section of an embryo of Chelydra serpentina of 8% mm. carapace. Johnson. X 20. The section shows the independence of the recessus postopticus and the recessus infundibuli. The section is not quite median in the region of the fovea isthmi and tuberculum posterius.

Fig. 28 Sagittal section of an embryo of Amblystoma tigrinum of about ten somites to show the relation of the notochord to the brain floor. X 46. The primitive hypothalamic downgrowth (hyih) is indicated by a thickening of the brain floor in which are seen two small cavities or clefts. These clefts communicate vsith the ventricle, but are irregular so that their connection does not show in a single section. The anterior end of the notochord is fairly sharply marked at the letters nch. The dense collection of small cells immediately above the letters nch. among which are several mitotic figures probably form the caudal wall of the hypothalamus.

Fig. 29 Sagittal section of an embryo of Amblystoma tigrinum of about seventeen postotic somites to show how the anterior part of the brain has been carried forward, leaving the anterior end of the notochord beneath the medulla oblongata. X 46.

The attempt to draw a diagram of the relations of somatic and visceral receptive columns in the forebrain region (fig. 1) may be premature. It may not be possible to determine the primitive relations of these columns in the forebrain. It is the writer’s present opinion, however, that a differentiation of these functional areas had taken place in the ancestors of vertebrates at a stage before the rolling up of the neural plate into a tube; that the axones of olfactory cells scattered in the skin found connections with neurones in the anterior end of the visceral receptive column; that general cutaneous ganglion cells established connections with the anterior end of the somatic receptive column; and that these two columns were more or less crowded and crumpled as the neural plate was rolled up into a tube. In the neural tube stage and in connection with all the changes attendant upon the development of the branchial apparatus and the abandonment of the palaeostoma in favor of the gnathostoma the rostral cutaneous area was greatly reduced and the innervation of most of it was taken over by the trigeminus. (Or, perhaps it is sufficient to say that the area innervated by the nervus terminalis was reduced to an insignificant vestige.) This situation left the anterior end of the somatic receptive area without dominating peripheral sensory fibers, such as enter all other segments of this column. The want of incoming sensory impulses of the first order, however, did not result in atrophy of this central area because it was already penetrated by secondary fibers coming up from lower segments of the same column, namely, visual, cutaneous, and proprioceptive centers. These secondary connections reaching the most anterior portion of the somatic receptive column maintained the functional activity of that region and pari passu with the reduction of its cutaneous area transformed it from a cutaneous center into a correlating center for all forms of receptive impulses coming to the somatic column from the skin and sense organs derived from the skin in all the more caudal segments. This process by which the anterior segment of the somatic receptive column was transformed into a somatic pallium or general cortex has been discussed at greater length in an earlier paper (’10 b). Although on pages 163 to 167 above I have refrained from speculation in treating of the olfactory centers, it has seemed worth while to state here at the end What I think is the most probable explanation of the origin of the olfactory and general (somatic) pallia.

November 20, 1922.

Literature Cited

BARTELMEZ, G. W. 1922 The origin of the otic and optic primordia in man. Jour. Comp. Neur., vol. 34.

BITRR, H. SAx'roN 1922 The early development of the cerebral hemispheres in Amblystoma. Jour. Comp. Neur., Vol. 34.

Cnosm‘, ELIZABETH C. 1917 The forebrain of Alligator mississippiensis. Jour. Comp. Neur., vol. 27.

HERRICK, C. .lI:DsoN 1905 The central gustatory paths in the brains of bony fishes. Jour. Comp. Neur., vol. 15. 1910 The morphology of the forebrain in amphibia and reptilia. Jour. Comp. Neur., vol. 20. 1922 What are viscera? Jour. of Anatomy, vol. 61.

HOLMGREN, NILS 1922 Points of View concerning forebrain morphology in lower vertebrates. Jour. Comp. Neur., vol. 34.

JOHNSTON, J. B. 1902 c An attempt to define the prirritive functional divisions of the central nervous system. Jour. Comp. Neur., vol. 12. 1906 Nervous system of vertebrates. Philadelphia: Blakiston. 1909 b The morphology of the forebrain vesicle in vertebrates. Jour. Comp. Neur., vol. 19. 1910 b The problem of the correlation mechanisms. Anat. Rec., vol. 4. 1910 c The evolution of the cerebral cortex. Anat. Rec., vol. 4. 1910d A note on the forebrain of Chimera. Anat. Anz., Bd. 36. 19119. The telencephalon of selachians. Jour. Comp. Neur., vol. 21. 1911 b Telencephalon of ganoids and teleosts. Jour. Comp. Neur., vol. 21. 1912 b The tclencephalon in Cyclostomes. Jour. Comp. Neur., vol. 22. 1913b The morphology of the septum, hippocampus, and pallial commissures in reptiles and mammals. Jour. Comp. Neur., vol. 23. 1915 b The cell masses in the forebrain of the turtle, Cistudo carolina. Jour. Comp. Neur., vol. 25. 1916 a Evidence of a motor pallium in the forebrain of reptiles. Jour. Comp. Neur., vol. 26. 1916b The development of the dorsal ventricular ridge in turtles. Jour. Comp. Neur., vol. 26.

KINGSBURY, B. F. 1920 The extent of the floor-plate of His and its significance. Jour. Comp. Neur., vol. 32. 1922 The fundamental plan of the vertebrate brain. Jour. Comp. Neur., vol. 34. fig. 30 Sagittal section of an older embryo of Amblystoma tigrinurn to show the columnar pigmented cells of the anterior end of the floor plate. X 110.

fig. 31 Sagittal section of a still older embryo of Amblystoma. tigrinum when the fovea isthmi is well established. The recessus postopticus and recessus infundibuli are quite distinct. X 110. L,-pr‘-I

fig. 32 Sagittal section of an advanced embryo of Amblystoma tigrinum to show the position of the anterior end of the notochord and the lateral sulcus in the mesencephalon connected with the fovea isthmi. X 46.

NEAL, H. V. 1898 The segmentation of the nervous system in Squalus acanthi-as. Bulletin of the Museum of Comparative Zoology, Vol. 31, no. 7. 1914 The morphology of the eye-muscle nerves. Jour. Morph., vol. 25. 1918 The history of the eye muscles. Jour. M0rph., vol. 30. 1919 Neuromeres and metameres. Jour. Morph., vol. 31.

SMXTH, G. ELLIOT 1908 The cerebral cortex in Lepidosiren. Anat. Anz., Bd. 33. 1910 The Arris and Gale lectures. The Lancet, Jan. 1, 15, and 22. 1919 A preliminary note on the morphology of the corpus striatum and the origin of the neopallium. Jour. of Anat., vol. 53.

SCHULTE AND TILNEY 1915 Development of the neuraxis in the domestic cat to the stage of twenty-one somites. Annals of the New York Academy of Sciences, vol. 24, pp. 319—346.

STOCKARD, CHARLES R. 1913 An experimental study of the position of the optic anlage in Amblystoma punctatum, with a discussion of certain eye defects. Am. Jour. Anat., vol. 15.

STONE, L. S. 1922 Experiments on the development of the cranial ganglia and the lateral line sense organs in Amblystoma punctatum. Jour. Exp. Zool., vol. 35.

Cite this page: Hill, M.A. (2024, May 18) Embryology Paper - Further contributions to the study of the evolution of the forebrain 5. Retrieved from

What Links Here?
© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G