Paper - The anterior end of the neural tube and the anterior end of the body

From Embryology
Embryology - 20 Aug 2019    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)

Dart RA. The anterior end of the neural tube and the anterior end of the body. (1924) J Anat. 58(3): 181-205. PMID 17104010

Online Editor  
Mark Hill.jpg
This historic 1924 paper by Dart describes the anterior end of the neural tube and the anterior end of the body

Dart RA. (1924). The Anterior End of the Neural Tube and the Anterior End of the Body. J. Anat. , 58, 181-205. PMID: 17104010

Modern Notes: neural |

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 | Postnatal | Postnatal - Neural Examination | Histology | Historic Neural | Category:Neural
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" 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 and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

The Anterior End of the Neural Tube and the Anterior End of the Body

By Raymond A. Dart, M.Sc., M.B., Cu.M.

Professor of Anatomy, University of the Witwatersrand, Johannesburg, Late Senior Demonstrator of Anatomy, University College, University of London.

Remove not the ancient landmark which thy fathers have set.” Proverbs xxii. 28.

In July 1922 I advanced certain criteria which I considered to be of the utmost importance in the classification of musculature. It was necessary, in order to bring the facts of the present contribution into their proper perspective, to show in that article that the use of the term “ visceral,” as applied to any portion of the segmented mesodermal musculature, was entirely misleading. By an historical examination of the introduction of the term “visceral” into osteological, myological and finally into neurological literature, I showed the violence done to the original meaning of the term in each new acceptance. I examined the musculature de novo from phylogenetic, embryological, histological and physiological points of view and suggested that all musculature be grouped under the two headings “‘ dermal” and “mesodermal”? as shown in the following table.

(table to be formatted)


DERMAL — unsegmented L (unstriped) \ ENDODERMAL

MUSCLE \ / DORSAL (extensor) \ usopERMaL—segmented (striped) \ VENTRAL (flexor)

It was shown further that the so-called.‘ visceral” musculature of the head region (eye musculature, jaw musculature, etc.) was comparable in all points with trunk musculature of the somatic (mesodermal) type. The present communication is designed to examine the embryonic history of the segmented mesoderm and then to show what light this embryonic history throws upon the two questions raised in the title of this paper.


In 1894 Lwoff put forward two conceptions concerning the origin of the segmented mesoderm at that time regarded as extremely unorthodox, which are given in his own words in the following statement:

Das Hauptergebniss dieser Untersuchung ist, dass die Einstiilpung bei Amphioxus keineswegs als einfache Gastrulation zu betrachten ist, als es bisher angenommen. Es sind vielmehr hier zwei verschiedene Processe zu unterscheiden: erstens die Einstiilpung der Entodermzellen, aus denen der Darm entsteht; zweitens die Einstiilpung der Ektodermzellen vom dorsalen Umschlagsrande aus, welche die ektoblastogene Anlage der Chorda und des Mesoderms bildet. Die Einstiilpung der Entodermzellen ist als Gastrulation zu betrachten. Es ist ein palingenetischer Process, den die Chordaten von ihren Vorfahren ererbt zu haben scheinen, wo dieser Process gleichmassig und radial symmetrisch vor sich ging, wie es sich bei einigen wirbellosen Thieren beobachten lasst. Die Einstiilpung der Ektodermzellen ist dagegen als ein caenogenetischer Process zu betrachten, der mit der Bildung des Darmes nicht zu thun hat und durch den die Bildung der ektoblastogene Anlage der Chorda und des Mesoderms eingeleitet wird.

Google Translate  
The main finding of this study is that the invasion in Amphioxus is by no means considered to be a simple gastrulation than previously thought. Rather, two different processes are to be distinguished here: first, the insertion of the endodermal cells from which the intestine arises; second, the insertion of the ectoderm cells from the dorsal border, which forms the ectoblastogenic system of the chorda and the mesoderm. The Einiilpung the endoderm cells is to be regarded as gastrulation. It is a palingenetic process which the chordates seem to have inherited from their ancestors, where this process proceeded uniformly and radially symmetrically, as is observed in some invertebrate animals. The ectodermal cells, on the other hand, are to be regarded as a caenogenetic process, which has nothing to do with the formation of the intestine, and which causes the formation of the ectoblastogenic system of the chorda and the mesoderm.

It will be readily appreciated that the two significant conceptions introduced here by Lwoff are (a) the postulate of a caenogenetic second phase in chordate development; the tissue of which gives rise to the chorda dorsalis and the segmented mesoderm and (b) the postulate that this second phase which provides the tissue giving rise to the chorda dorsalis and the segmented mesoderm is an invagination of ectoderm cells.

Nobody need be reminded of the fact that the first phase (the palingenetic phase or gastrulation proper), to quote Brachet (1921, p. 117), is “‘le processus grace auquel, aux dépens de la masse des cellules issues de la segmentation de l’ceuf, se constitue une larve 4 deux feuillets, l’un externe, l’autre interne, qui restent en continuité entre eux en un point déterminé de leur étendue. La gastrula est donc, dans tous les cas, un embryon didermique.”” This phase has for its objective the laying-down of the gut and its derivatives and the formation of that ancestral gastrula type postulated by all embryologists since the appearance of the classical research of Huxley (1859) when he established the homologies of the ectoderm and. endoderm of Coelenterata with the same layers in Vertebrata. It is the same ancestral type whether we regard its endoderm as being produced by invagination (as in Haeckel’s gastraea, 1872), by delamination (as in Lankester’s planula, 1877), by transverse fission (as in Biitschli’s plakula, 1884), or by immigration (as in Metschnikoff’s parenchymella, 1886)—vide McMurrich (1890). The term didermique used by Brachet is therefore singularly apt in summarising the end result of this first phase.

We may therefore assume that all investigators are unanimous in recognising a didermic or first phase of “‘gastrulation,”’ whatever the differences of opinion may be concerning the actual method of achievement of the phase.

As regards the second phase, it is perfectly true that even before the time of Lwoff various observers had recognised that the process of invagination in Chordata was not a single but a double one. However, Lwoff appears to have been the first to realise that the second phase had one objective only and that this objective was the formation of the tissues of directive movement, i.e. the segmented mesoderm and the notochord. He therefore was the first to really define it and certainly did a great service by conceiving it as a unified and separate process superimposed upon the first phase. Since the time of Lwoff, this second phase has been repeatedly recognised—by Brauer (1897), Brachet (1902), O. Hertwig (1903), Hubrecht (1905), Keibel (1905), and champs (1910)— and has become practically part and parcel of present day embryological orthodoxy.

The actual nature of the events which lead to the production of the second phase of gastrulation will be apparent immediately if fig. 1 (representing diagrammatically the invagination process in an Amphibian form) be examined. This diagram (after Brauer) shows the blastoporal opening posteriorly. The ventral (or posterior) boundary of the blastopore is formed by large rounded and richly yolk-laden cells which present the strongest possible morphological contrast to these endodermal cells as regards their size, shape and cellular content. They are deficient in yolk and are richly supplied with pigment so approximate much more closely in character to the ectoderm. These cells may be traced forward as a continuous plate lying under the ectoderm. This plate terminates suddenly and is supplanted anteriorly by two endodermal cells (i.e. at the point of transition). This ‘‘dorsal plate” (or deutenteric arch of Brachet) gives rise to the notochord and the segmented mesoderm. We see in it therefore the accomplishment of the second phase in gastrulation.

Fig. 1 (after Brauer) to show the two phases of vertebrate gastrulation.

So obtrusive are these two phases in Petromyzon that Selys Longchamps (1910) ultimately named the cavity lying under the primordium of the chorda and the segmented mesoderm the deutenteron, as opposed to the archenteron which is present more anteriorly in the embryo and arises before the deutenteron.

The “enteric” cavity covered over by this dorsal plate of mesodermal tissue in the amphibian (vide fig. 1) is the deutenteron (of Selys Longchamps); anterior to it we recognise the archenteron which becomes formed gradually and lined by true endoderm cells.

This conception of the second phase has found admirable treatment in Brachet’s (1921) work, where it has been demonstrated for all the vertebrate classes. Consequently, there is little need here for a revision of the detailed evidence.

On en a déduit (says Brachet, p. 181) cette conclusion, d’une importance capitale pour la morphologie des Chordes, que l’écusson médullaire. et les organes qui lui sont sous-jacents, par conséquent le dos entier de l’embryon, au cerveau antérieur 4 l’anus, se forment dans la domaine exclusif du deutenteron, par différenciation sur place de l’ectoblaste et de l’entoblaste qui en constituent la paroi superficielle. Nous donnerons 4 cette paroi le nom de votite deutenterique.

Scarcely has the first phase been completed in lower Chordata (Amphioxus and Petromyzon) when the second phase begins. Indeed, the higher the animal, the greater is the tendency for the second phase to be initiated early and to be expedited so that in the mammalia the second phase is ushered in so rapidly that it may be described as anteceding the first phase. In this precocity in appearance of the second phase we recognise another example of the tendency frequently evidenced in embryonic forms to reproduce, as early as possible in foetal life, the tissues which are more early called into activity irrespective of the order of their appearance in phylogeny. This is of interest because certain authors (vide Lwoff’s paper) have been misled by this precocity into regarding the second phase as palingenetic instead of caenogenetic, a deduction which other considerations will not allow us to admit.

It is evident, therefore, that the first postulate of Lwoff—the unified and separate process superimposed upon the first phase—is well grounded in embryological fact. Further, since this process gives rise to the segmented tissues it is evident that the second phase and the deutenteron are peculiar to the possessors of a segmented mesoderm.

The second postulate of Lwoff—the ectodermal origin of the notochord and segmented mesoderm—has not been generally admitted by embryologists. Selys Longchamps and Brachet, amongst the recent Continental workers, still regard these tissues which arise from their ‘“‘deutenteric arch” as endodermal in origin—a view dating from Kowalevsky’s and Hatschek’s classical researches upon Amphioxus and maintained by Conklin (1905) for the Ascidian.

An insistence upon the endodermal origin of these tissues has been rendered very easy by its general adoption in most embryological text-books since Hatschek’s time. Nor is the reason for this insistence far to seek. The cogent impulse behind the attempt to derive the segmented mesoderm from the endoderm is the establishment of an homology between the “‘outpouchings of the archenteron” in Echinodermata and the so-called ‘“outpouchings of the archenteron” in Chordata; i.e. the demonstration of an homologous process of mesoderm-formation in both phyla.

There are two reasons why such a doctrine is untenable. In the first place, the two tissues called ‘“‘mesoderm” produced by these supposedly homologous processes are in no way comparable with one another. Thus the “ outpouchings of the archenteron” in Echinodermata give rise to a water-vascular system and other structures peculiar to Echinodermata; in brief, the resultant “mesoderm”’ is unsegmented and is innervated by a characteristic nerve-net system. The embryological investigation of Echinoderms has tended to emphasise this endodermal origin of the unsegmented mesoderm in these creatures; there is present in them, however, an equally important, even if not so obvious, ectodermal constituent of this unsegmented mesoderm also innervated by the nerve-net system. The whole “mesoderm” so constituted of endodermal and ectodermal elements and innervated by the nerve-net system may be broadly compared only with the peripheral mesoderm of Rickert (the peristomial mesoderm of Rabl) in Chordata. In Chordata, this "mesoderm" is equally unsegmented and innervated by the sympathetic system—the nerve-net of these forms.

But the unsegmented mesoderm, common to all Metazoa, is a very different type of tissue from that mesoderm which results from the so-called “outpouchings of the archenteron” in Chordata. This mesoderm is the true or segmented mesoderm—the azial mesoderm of Riickert (the gastral mesoderm of Rabl). There is no tissue of Echinodermata comparable with this segmented mesoblast which is not only innervated by the sympathetic nerve-net system (Boeke, 1918; J. T. Wilson, 1921) but has an additional and characteristic innervation presenting the synaptic neurones of Waldeyer with their specific reflexes and invariability of response—phenomena entirely without counterpart in Echinodermata (Dart, loc. cit.).

The second reason why the echinoderm homology is untenable has been adequately dealt with by Lwoff whose statement will be used here. Lwoff has shown that, even though “outpouchings” occur in Chordata, the cavities of the outpouchings do not form the coelom as has been assumed in the past; but these cavities always disappear and the coelom appears separately and elsewhere. This repudiation is so important and has been so generally neglected that it should be quoted in full. He says:

Daraus ist klar, dass die Mesodermfalten mit ihren Héhlen bei Amphioxus nur ein dussere zufallige Erscheinung darstellen, der man keine besondere phylogenetische Bedeutung zumuthen kann. Die Leibeshéhle hat hier mit den vermeintlichen Urdarmdivertikeln nichts zu thun. Es ist also nur eine scheinbare Enterocoelie, die in Wirklichkeit nicht existiert, da die Leibeshéhle, wie bei allen Wirbelthieren, durch Auseinanderweichen der Zellen gebildet wird. Es darf darum keine Rede davon sein, das Amphioxus streng genommen ein Enterocoelier ist, geschweige denn davon, dass alle Wirbelthiere, was die Mesoderm—und Leibeshéhlenbildung betrifft, von einem Enterocoelier abzuleiten sind. Denn alle Versuche, bei niederen oder héheren Wirbelthieren die Urdarmdivertikel resp. Mesodermfalten und einige Spiiren von Enterocoelie zu finden, sind so gezwungen, dass man sie im Ernst nicht nehmen kann.

Dasselbe muss auch iiber die Coelomdivertikel gesagt werden, welche Van Beneden und Julin bei Clavellina beschreiben, weil die Héhlen dieser Divertikel bald verschwinden und das Mesoderm jederseits solide Zellenhaufen darstellt. Uebrigens konnte der spitere Untersucher des Gegenstandes—Davidoff eben so wenig wie der friihere—Seeliger die Bilder auffinden wie sie van Beneden und Julin zur Annahme der Divertikel fiihrten. Davidoff kommt in seiner Zeit zum Schlusse dass die Entstehung des Mesoderms bei Distalpia und Clavellina in keiner Weise an Ausstiilpungen der Gastrulahéhle (Coelomdivertikel) gebunden ist, und dass die Mesodermentwickelung bei diesen Chordaten keineswegs auf eine den Hertwigschen Enterocoeliern entsprechende Form zurickgefiihrt werden kann (loc. cit. pag. 600 und 607). Da bei anderen Tunicaten bis jetzt keine Spur von Coelomdivertikeln gesehen wurden, so beruht die ganze Coelomlehre, was die Chordaten betrifft, nur auf der Entwickelung des Amphioxus. Da aber meine Untersuchungen gezeigt haben, dass auch beim Amphioxus diese Divertikel zur Leibeshéhle (Coelom) nicht werden, sondern verschwinden, so kann angenommen werden, dass ein wahrer Enterocoelier unter allen Chordaten nicht existiert.

It is very evident, then, that the coelom in Chordata is not archenteric in origin at all, and certainly cannot be homologised with the “archenteric outpouchings” in Echinodermata. This lacuna in the evidence for homologies has never been filled in; and even if it were, a still greater one remains. The “‘outpouchings” themselves in Chordata are not outpouchings of the archenteron but of the deutenteron; and the deutenteron (with its derivatives) is something characteristic of Chordata which has no counterpart whatever in Echinodermata.

The assumption which has been used to justify “‘the hypothesis that the mesoblastic somites of segmented animals are derived from a diploblastic coelenterate-like ancestor with folded gut walls,” in the past, has been the belief that “the folding has arisen as a result of the necessity for an increase in the extent of the vegetative surfaces in a rapidly enlarging animal” (ef. Sedgwick, 1884).

Such a conception is entirely inadequate to explain either the embryological or the physiological facts. The second phase in development, which gives rise to the segmented mechanisms, teaches us that the distinctive factor in the differentiation of the higher segmented forms is the segmented nervous system and dependent upon it but coincidently with it a segmented mesoderm (vide Kleinenberg).

The fundamental physiological distinction between the movements ofanimals without this segmental apparatus and those of creatures which possess it in even the crudest form was detected by Agassiz and Gould (1848) when they said:

The jelly fishes (Medusae) swim by contracting their umbrella-shaped bodies upon the water contained within and its resistance urges them forwards, ...others contract small portions of the body in succession which, being thereby rendered firmer, serve as points of resistance, against which the animal may strive, in urging the body forwards, The earth-worm whose body is composed of a series of rings united by muscles, and shutting more or less into each other, has only to close up the rings at one or more points to form a sort of fulerum against which the rest of the body exerts itself in extending forwards.

It was this serial arrangement of successive rings (segmentation) which provided a point of departure from the peristaltic response of the nerve net system of lower forms. It depended upon a segmental autonomy in reflex response — the reflex are and the neurone of Waldeyer. In short, the achievement of a neuromuscular mechanism of segmental type has been the starting point in the production of forms invertebrate and vertebrate, endlessly diversified in type, all of which have solved in the characteristic segmental fashion the problem of orientating the body adequately with reference to its environment. Such forms, as I have shown elsewhere (loc. cit. 1922), proved especially capable in dealing with the problem of land progression. A segmented nervous system was therefore the initial event which led on the one hand to that pre-onychophoran form which was ancestor to Annelida, and on the other hand to the ancestor of Chordata.

It might be anticipated from these neurological and physiological considerations that an ectodermal origin of the segmented mesoderm is not only possible but highly probable. Huxley pointed out (1877) that the fundamental reason for the differentiation of ectoderm and endoderm was the physiological division of labour, providing in the first a tissue for protection and locomotion and in the second a tissue for nutrition. The segmental apparatus is first and foremost a locomotor tissue—for directive movement of the body as a whole. By second intention this mechanism of body movement is protective—i.e. for directive movement away from danger. Last of all, various portions of this system, so arising, become adapted in response to the demands of nutrition— i.e. the mastication and digestion of food, etc.; but such adaptations are rather final and not initial factors in the appearance and transformation of segmented mesoderm. It is for this reason that such segmented musculature as becomes secondarily adapted for mastication, deglutition, respiration, micturition, defaecation and parturition is amongst the most altered muscular tissue of the animal body and has preserved, with difficulty as it were, the traces of its primitive segmental simplicity and autonomy.

In consequence of these realities, histological criteria are available which support the ectodermal conception of the origin of the “deutenteric arch” (or “head process” or ‘“‘dorsal plate”). The first criterion is that the cells arise at a definite site (the dorsal or anterior lip of the blastopore) and are progressively invaginated along the roof of the enteric canal—hence the “invagination”? is real (cf. fig. 1). Secondly, this region of “‘ecto-mesoderm” is in a state of extraordinarily active mitosis while the definitive endoderm is relatively passive and amitotic. In consequence, there is a marked contrast between the anterior and posterior lips of the blastopore. Finally there is an intimate histological similarity (e.g. yolk content—vide Brauer) between the cells of the ectoderm and of the segmented mesoderm, while the cells of the ectoderm and the endoderm are in marked histological contrast.

This invagination, from the ectoderm, of cells of different character from true endoderm cells was certainly recognised by the three British investigators, Balfour (1880-81), Scott (1882) and Shipley (1887). Indeed Shipley was revolutionary enough to state of Petromyzon that “‘its dorsal wall is composed of columnar cells resembling those of the general epiblast; the cells forming the floor have the same characters as the yolk cells.”

But Lwoff showed that identical conditions obtained through all Chordata and in this he has been confirmed, more especially by the admirable embryological researches of Brauer (1897). The revision of discordant views by Brauer is of the first importance. Moreover he has given support to Lwoff’s view, by a careful study of the pigment content of the mesoderm (p. 457). The work of Lwoff and Brauer offers a better interpretation of the facts concerning the lower Chordata and has allowed of a direct homologisation of the process of mesoderm formation there with the process as followed by Amniota. Their points of disagreement from other workers are not points of fact but of interpretation. In this interpretation they have received noteworthy support from Carl Huber (1918), who has summarised the literature upon the subject, carried out a meticulous research upon the development of the mammalian chorda dorsalis and expressed himself finally as follows:

Since the endoderm takes no active part in the histogenesis of the head process, chordal canal and chordal plate and since the chordal plate becomes only partially and temporarily incorporated in the endoderm, there seems no justification for classing the chorda dorsalis as an endodermal derivative. And since the head process, the anlage of the chordal canal and derived structures, has its anlage in the cranial portion of the primitive node, a region of active ectodermal cell proliferation; and since the chordal canal and plate retain their continuity with the primitive node, which serves as a growth zone; there seems justification in regarding the head process—chordal canal and derived structures, chordal plate and chorda dorsalis—as a derivative of the ectoderm of the primitive streak region of the embryonic shield.

It is futile to embarrass chordate embryology with analogies which the science has long since outgrown. Not only the first but also the second postulate of Lwoff is justified—the justification not merely resting on a histogenetic basis but being confirmed by neurology, physiology and phylogeny. For just as Brachet (1921, p. 179) has insisted that the chorda dorsalis is.a neoformation so, too, are the segmented mesoderm and neural tube that arise coincidentally with it. These neoformations are, as we have seen, the basis of the physiologist’s voluntary apparatus; and it is rather to be expected that since this apparatus was designed to answer the call for locomotion of the body as a whole (Dart, loc. cit.) the stimulus behind segmentation was a more precise appreciation of environment by the ectoderm (Huxley) rather than some fortuitous internal expansion of the endoderm (Sedgwick). This conception is in entire agreement with the heterodoxy of Lwoff’s derivation of the segmented mesoderm from the ectoderm in Chordata and with Kleinenberg’s derivation of the segmented musculature and nervous system from a “gemeinsame Neuromuskelanlage” in the ectoderm of segmented Invertebrata. It is expressed in the following scheme of classification:

Coelenterata, Platyhelminthes, Nemathelminthes, Echinodermata, Mollusca.


uxszommnren MESODERM. ENDODERM’ Anterior End of Neural Tube 189

Annelida, Arthropoda, Chordata.




The traditional text-book pictures founded upon Hatschek’s figures of Amphioxus, and an imperfect conception of the nature of the segmented mesoderm, are misleading. The fact of bilateral symmetry in this mesoderm so impressively represented in Hatschek’s figures is undoubtedly valuable; but the equally fundamental truth revealed by this series of investigators is thereby lost; namely, that the whole segmented mesodermal mass is an “invagination”’ of the ectodermal anterior lip of the blastopore.

Meantime a wholly different series of investigators including Kastschenko (1888), Goronowitsch (1892), Miss Platt (1898), ete.—as Landacre (1921) has shown in his admirable summary—have declared that much of the “head mesoderm” is proliferated from the ectoderm directly—a fact which has ultimately received important experimental confirmation in the work of Stone (1921).

This “head mesoderm,” so arising, is part of the segmental mesoderm— it gives rise to the “‘trabeculae, Meckel’s cartilage, the palatoquadrate and all of the branchial cartilages except the second basi-branchial or urohyal” (according to Landacre). It would, indeed, have been an extraordinary fact if this mesoderm had been ectodermal and the mesoderm giving rise to the segmental musculature had not been ectodermal (as so many have maintained).

The cumulative evidence of the two series of observers mentioned, has shown that not only the so-called “‘head mesoderm” but also the segmented mesoderm of the trunk is primarily ectodermal; so the two together form an indivisible harmonic structural unity.

This unity in character is illustrated not only by the embryological evidence adduced. We have already recognised that the whole of the striated (mesodermal) musculature (derived from the deutenteric arch formation) is to be contrasted with all other unstriated (or dermal) musculature, because the dermal (unstriated) musculature is innervated by the sympathetic (nerve-net) system only, whereas the true mesodermal (striated) musculature is innervated by the sympathetic system and by the central nervous system as well. In this fundamental fact of double innervation we recognise the unity of the adult tissues derived from the deutenteric arch formation and appreciate in addition that we have, in this tissue, an evolutionary advance on the dermal musculature which was there before it—that the ‘mesodermal musculature, as a whole, is a new formation.

It is equally significant that from the striated musculature there flow continuously into the medullary tube those impulses of “‘muscle sense” and “tendon sense” which provide the organism with information concerning its location in space. This property of the mesodermal (striated) musculature, which is unknown in the dermal (unstriated) musculature, is at one and the same time a token of its separation from the (endoderm) tissues which possess only an “enteroceptive” quality and its affinity with the tissues (ectodermal only) which possess the “‘exteroceptive” quality so preponderantly. Unfortunately these terms “‘extero-, proprio-, and enteroceptive” have meaning only in neurology—as denoting the different kinds of afferent impulses recognised in the animal body. The terms have had no morphological meaning hitherto, nor is it suggested here that they should be given a morphological meaning. Attention is merely drawn to the fact that the distinctions in the “afferent”? system, inculeated by Sherrington, have as their morphological foundation one fundamental fact and one only: that the enteroceptive impulses are mediated primarily by a nerve-net (vegetative) nervous system, the exteroceptive and proprioceptive primarily by a segmented (central) nervous system; the former is initially the chaos of a network, the latter the order of serial arrangement.

This serial segmental arrangement is found in the ectoderm and it is found in a certain restricted portion of the mesoderm. It has never been found in the endoderm of any organism. Are we then to attribute to this endodermal parent, fertile as she has been, such an unnatural offspring?

But the unity in character, or better the maternal-filial relationship between the ectoderm and the segmented mesoderm is equally patent if we consider the other derivative of the deutenteric arch (or dorsal plate). I refer to the skeletal parts, the bones, the fulera and points of attachment upon which the segmented musculature acts. The most exacting histological observation by observers of all nationalities over a long period of time has only served to reveal the intimacy of the homology between the various processes of bone formation in the ectoderm and in the segmented mesoderm, whether the end results be scales, the so-called “‘dermal” bone, “cartilage” bone, or “membrane” bone. What evidence have we that the endoderm preserved the faculty to produce such structures itself, much less to produce a whole sheet of body tissue which should give rise to such structures at a considerably later period of evolutionary history? In other words the tissues to which this sheet of invaginated ectoderm gives rise vindicate their ancestry by their close histological resemblance to similar tissues whose origin from the ectoderm is indisputable; and so the whole segmented mechanism of musculature and skeletal parts manifests in its lineaments (embryonic or adult) that it sprang from the same womb—twin births as it were—which gave origin to the tissue controlling its activities, namely, the segmented medullary tube. It is by these criteria of histology and of neuro-physiology as well as, or rather, in preference to those other hypothetical and very erroneous criteria of echinoderm comparison, that these questions should be examined and an answer returned.

Superficially, the recognition of this fact of ectodermal invagination might seem to be merely a verbal quibble; but actually it is a generalisation of the greatest significance not only for physiology, neurology and embryology, but for all orientation and consequently for all descriptive morphology, whether of the skeletal, osseous or nervous systems—indeed of the whole body. Because orientation is of such paramount importance in descriptive morphology I wish to examine, in the light of the information gathered concerning the segmented mesoderm, two ancient questions of orientation.


In this fact of the second phase of invagination we have the ontogenetic repetition of a fundamental incident in phylogenetic history, namely, that the segmental tissues were introduced into the framework of a creature which previously had no segmentation at all but already possessed a mouth and an anus. Since the segmental tissues are the tissues of directive movement, this is only another way of saying that the creature understood how to deal with food before it appreciated how to seek for it. ,

Fig. 2 (after Goette) for comparison of “‘head”’ and “tail” growth.

Whether we agree with those who (like Sedgwick and O. Hertwig) derive the mouth from the division of the blastoporal slit, or, with those (like Huxley) who postulate for its production a new formation, we must recognise the general admission that the laying down of these segmental tissues has taken place along a central linear axis extending between these two sites. The proof of this statement lies in the fact that, when the ectodermal invagination has given rise later to the segmented tissues, the segmental tissues (“le dos entier de l’embryon’’) are limited in front and behind respectively by these two sites (oral and anal membranes). This fact is picturesquely emphasised by a comparison of the behaviour of the segmental tissues posteriorly and anteriorly during development.

Fig. 2 is a schematic sagittal section of Bombinator igneus (after Goette) to illustrate the changes which take place after the invagination (pictured for another amphibian in fig. 1). The original blastoporal site is now represented, as is well known, by the anus. The zone of growth still lies (morphologically 192 R. A. Dari

speaking) anterior to the blastopore (anus), and gives rise to the segmented tissues of the tail. Under it lies the neurenteric canal; dorsally, the canal opens into the neural tube and ventrally into the post-anal gut. At this stage the ‘“deutenteric arch” is no longer exposed in the gut cavity but has become separated from it by the undergrowth of true endoderm and has itself become differentiated into notochordal and somitic tissues. Concerning the relation of the tail to the anus Brachet (p. 255) has suitably summarised our present information:

Quant a la queue, elle procéde de l’allongement et de la différenciation du bourgeon caudal. Elle surplombe d’abord l’anus, puis le dépasse d’avant en arriére. Par définition, elle est, non seulement chez tous les Amphibiens, mais chez tous les Vertébrés, la portion postanale de la larve. Or l’anus étant, par son origine méme, l’extrémité terminale de la partie ventrale du corps, il s’ensuit que le bourgeon caudal n’est qu’un prolongement de sa partie dorsale et ne contiendra dans sa substance que le systéme nerveux central, la chorde, les parties juxtachordales du mésoblaste et la votite du tube digestif.


Fig. 3 (after Bonnet) to show anterior end of the dorsal plate (deutenteric arch).

Now, no embryologist would seriously put forward the notion that the tip of the tail or the posterior termination of the central nervous system in the filium terminale represents the morphologically terminal end of the chordate body. It is equally impossible to regard the projecting tip of the snout, much less any supposed termination of the neural tube, as the true anterior end of the body as has been done by Hatschek (1909) and others.

Fig. 8 (a sagittal section of a dog embryo after Bonnet) presents a certain stage in the development of the anterior region of the body. The oral membrane O, hypophysis H, and Seessel’s pocket S, present a fixed and striking relationship with one another. The ‘“‘voite deutenterique” of Brachet is -represented anterior to the notochord N by the so-called “‘ prae-chordal Platte ”’ of Oppel (“protochordal plate” of other authors). The continuity of. this structure with the notochord and the direct part it plays (cf. K. M. Parker, 1917) in the production of the anterior head cavities and in consequence (cf. E. A. Fraser, 1915) of the eye musculature show its continuity with and genetic relationship to the “‘dorsale Einstiilpung”’ of Goette (i.e. the “‘dorsale Platte’? of Lwoff), which gives rise throughout Chordata to the axial skeleton and the segmented mesodermal tissues generally.

Fig. 4 (after Brauer) to emphasise the linear morphological orientation of the segmented tissues between the oral membrane and anus.

Reference to fig. 4, taken from the work of Brauer, will make these facts obvious. We must recognise that, as far as the head is concerned, it developes by a lengthening out of the neural tube and the segmented mesoderm headwards over the oral plate which is directly comparable with the prolongation of these structures tailwards beyond the anus. Just as the prolongation of these structures posteriorly causes a dorsal diverticular extension of the gut to form the post-anal gut, so their anterior prolongation is the causative factor in the production anteriorly of the dorsal diverticular extension of the gut known as Seessel’s pouch or pre-oral gut. Subsequently, just as the post-anal gut disappears in ontogeny so the pre-oral endoderm “does not play any part in later development” (K. M. Parker, 1917, p. 195).

It follows from these facts that the caenogenetic tissue of the protochordal plate and its derivatives ends in the vicinity of Seessel’s pocket and that the protochordal plate is morphologically entirely posterior to the oral membrane just as the same tissue is morphologically entirely anterior to the anal membrane. Consequently the so-called prae-oral gut is morphologically post-oral and the post-anal gut is prae-anal. Finally, the most valuable morphological anteroposterior orientation which we possess for all Chordata is the oral plate-anal membrane orientation; and the oral membrane is the true anterior end of the chordate body.

The criterion of antero-posterior orientation so established needs a word of comment because the vertebrate mouth has been provocative of such extraordinary speculations. Even at the present time we find some investigators (e.g. Neal, 1921), who hold that ‘“‘the present vertebrate mouth may be considered as the third—or fourth, if the neuropore be added—mouth in chordate cephalogenesis.”’

’The contribution of Beard (1888) has done much to eliminate the crudities of the earlier speculations of Dohrn, Owen, J. T. Cunningham, and others concerning the fourth ventricle, infundibulum and other routes which the “‘palaeostoma”’ was supposed to have taken. At the same time Beard’s own speculation concerning the hypophyseal “palaeostoma”’ is searcely happier. Ontogeny provides no evidence whatever to support a ‘‘ palaeostoma” other than the oral membrane itself.

From such bizarre interpretations it is refreshing to turn to the single postulate which Huxley (1877) regarded as sufficient to account for the appearance of a stoma in animals above Coelenterata, namely “the development of a secondary aperture near the anterior end of the body, which becomes the permanent mouth.” The only modification of this conception suggested here is that this secondary aperture was not merely near the anterior end of the body, but was itself the actual anterior end.


The recognition of the “second phase”’ in chordate development has further a most important bearing upon the problem of the anterior end of the neural tube. We have recognised as one of the constituent portions of “le dos entier de l’embryon” the neural tube itself. In direct consequence of this fact, the one and only point where the neural plate and axial mesoderm (the coincident phylogenetic new formations) come into most intimate relationship with one another, and at the same time with the ectoderm (hypophyseal inpocketing) and the endoderm (Seessel’s pocket) which they are separating—this point is at the anterior end of both the neural plate and of the axial mesoderm. This site, common to all these structures, may be termed the infundibular point, and the infundibular recess, which is in most intimate contact with this site, is to be regarded as the anterior extremity of the neural tube.

It is generally laid to the credit of His that he was the first to differentiate clearly between an alar and a basal lamina of the neural tube and to give a clear and precise statement of the development of that structure, placing the anterior end of the neural tube in the recessus infundtbuli.

As a matter of actual fact, von Baer (1828) had pointed out many years before, against the teaching then current, that ‘der hohle Cylinder (i.e. the neural canal) bestiinde aus zwei urspriinglich vereinigten Halften.” He showed (S. 64) that “‘Jede Seitenhalfte des Riickenmarkes ist durch eine mittlere helle Furche in einen obern und einen untern Strang getheilt” and from considerations of primary position and subsequent growth had already termed the infundibulum “das wahre urspriingliche Ende vom Centraltheile des Nervensystems.”

Various authors since the time of von Baer, however varied their line of approach, have. been led to identical or approximately identical views. The outstanding features in regard to the infundibular location is its fixity from the embryological and osteological point of view. Thus J. E. Frazer (1921) states:

Some years ago (Lancet, 1916) I described the pituitary region of the brain as the most fized place within the skull, and referred the formation of the midbrain flexure to the forward growth of the hind-brain acting against this fized point....The definite adhesion of stomodaeal and neural ectoderm at this place, and their association with the upper end of the bucco-pharyngeal membrane, Seessel’s pocket (when present), and the adherent notochord, seem to me to give a guarantee of fixation, supported by every sagittal section, which must influence the shape of a brain growing up against its fixation here. The brain growing more rapidly than the skull-base, piles itself up into curves, so to speak, and these then owe their existence mechanically, and at least in part, to this fixation of the forebrain.

Equally definite is the dictum of Sir Arthur Keith (Keith and Campion, 1921) in the following excerpt:

The superimposition has been made so that pituitary fossa falls on pituitary fossa, and cribriform plate on cribriform plate, because a prolonged experience

has shown one of us (A. K.) that the pituitary or sphenoid region serves best as a fixed point in comparing the development of one skull with that of another.

Similarly Brachet (1921) in discussing the development of the brain expresses himself (p. 333):

Comme celle-ci est fivée 4 ses deux extrémités, en avant par l’hypophyse et la bouche, en arriére par sa continuité avec le trone, elle est obligée de se soulever en une voussure proéminente, qui surplombe et dépasse de plus en plus la région de la membrane pharyngienne.

That the cranial flexures are a measurable sign of the degree of twisting that has taken place about the end of the notochord is well recognised. Hertwig (p. 425, Mark’s translation) states:

The extent of these curvatures is very different in the various classes of vertebrates. Thus the cephalic flexure is only slightly emphasised in the lower vertebrates (Cyclostomes, Fishes, Amphibia); it is, on the contrary, much greater in reptiles, birds and mammals; but in Man especially, whose brain is most voluminous, all of the flexures are developed to a very high degree.

This fixity is the conditio sine qua non of any landmark and these weighty statements, representing the results of experienced observation, are corroborative of von Baer’s view. They might be expected sufficiently to solve the problem which we have here engaged, but it is necessary to examine closely other views which have been advanced at different times over an extensive period and which lay claim to being more adequate interpretations of the evidence.

This evidence, gleaned from the study of the ontogeny of the chordate neural tube, has given the most diverse results in the hands of various workers.

His (1893) regarded the anterior end of the neural tube as closing during development by a process of concrescence, which extends from the infundibulum forwards to the upper edge of the lamina terminalis. Consequently, although His did place the extreme anterior end in the recessus infundibuli, he gave no morphological reason for so doing.

Von Kupffer (1898) chose the recessus neuroporicus at the upper edge of the lamina terminalis, but Keibel (1889), J. B. Johnston (1909) and others have selected the optic chiasma as the point in question. It is evident that by means of the “raphe” conception His was able to bring his views ‘more or less closely in line with von Kupffer’s notion of the neuropore and yet to state that he differed from Keibel’s optic chiasma “‘nur um weniges.”’

But this rhetorical treatment of the matter is unsatisfactory, unscientific and inadmissible. We will discuss the two points in turn considering the claims, first, of the neuropore; and, second, of the optic chiasma.

Neuroporic recess.

Von Kupffer put forward in various papers a series of postulates which have not received corroboration: (1) that the last point of closure of the neural tube and its separation from the ectoderm is constant for Vertebrata and is marked in the adult by the recessus neuroporicus; (2) that the point in question is homologous with the “neuropore” of Amphioxus; and (3) it is homologous also with the “‘lobus olfactorius impar” in Cyclostomi.

In the first place, as regards the lamprey homologisation, Kappers (1921) has said quite recently that von Kupffer’s point (neuropore) loses much of its importance because of Woerdeman’s (1914) demonstration that the olfactory placode is in contact with the “‘ Vorraum” of the hypophysis. But quite apart from any embryological demonstrations at all, it is clear that the homology of the unpaired olfactory lobe of Cyclostomes with the embryonic neuropore of vertebrates by von Kupffer is pure speculation; for it is inconceivable how any homology of a cavity with a definite cellular mass can be contemplated. The only homology the lobus olfactorius impar can have is with the paired lobes of higher forms. As Karl Peter (1906) pointed out (Hertwig’s Entwickelungslehre) nobody has ever confirmed von Kupffer’s “triple” origin of the olfactory organ in Cyclostomes nor his unpaired olfactory nerve passing to the unpaired groove.

In the second place, Hatschek (1909) by a comparative study of Petromyzon and Amphioxus has dealt with von Kupffer’s second contention. He has refuted his homology of the neuropore of Amphioxus with the structure, so-named, in other Chordata and has insisted upon the “roof” position of the lamina terminalis and associated structures.

Finally, other investigators have rendered yeoman service to embryology and neurology by a critical examination of the first and really basal postulate in von Kupffer’s thesis. Koltzoff (1902), by a study of the development of the ammocoete larva of Petromyzon, showed that the “‘neuropore” of von Kupffer is not the terminal point of the neural tube at all but can be regarded at best as merely that place where the neural tube last loses contact with the ectoderm. He further insisted that the “‘recessus opticus”’ belongs to the dorsal or roof structures of the neural tube and that in later stages the apex of the infundibulum serves as the boundary mark between the floor and roof of the neural tube. In this way, he urged, the floor of the tube lies in contact with the notochord, the roof with the ectoderm.

Equally vigorous is the attack of Fanny Fuchs (1907-8) upon the speculations of von Kupffer. The latter had assumed, as stated previously, that the last point of separation of the neural tube in ontogeny is represented in the adult by the recessus neuroporicus and that it constitutes the most anterior point of the brain in all Chordata. Of this Fanny Fuchs (1907-8) has said:

Dieser Zusammenhang (of brain and ectoderm)...kann bleiben erhalten langer an irgend einer Stelle. Der Prozess der Ablésung des Hirns vom Ektoderm geht wohl etwa von der Mitte des Gehirns aus und schreitet von da nach vorn und hinten fort, so dass schliesslich irgendwo in der Nahe des vorderen Hirnendes dieser letzte Zusammenhang bestehen bleibt. Ebensowenig aber, wie die Stelle genau zu fixieren ist, wo die Ablésung beginnt, ist auch die, wo sie aufh6rt, als fester Punkt zu betrachten, der im allen Hirnen an der gleichen Stelle liegt. Offenbar wird diese Stelle wesentlich durch die Formbildung des Hirns bestimmt; sie wird immer da liegen, wo die geringsten Veranderungen durch Wachstum vor sich gegangen sind.

Daraus geht schon hervor, dass der Neuroporus nie an der stark wachsenden Vorderseite des Hirns (lamina terminalis), sondern héchstens an ihrem obern Rande liegen kann. Dort wird er auch sehr haufig angetroffen, wie ein Blick auf v. Kupffer’s Medianschnitte zeigt (Selachier, Ganoiden). Bei den Anuren aber liegt er viel weiter hinten, etwas vor der vorderen Grenze des Mittelhirns.

Aus dem Gesagten geht wohl zur Geniige hervor, dass die Stelle, wo der Neuroporus liegt, wechselt und dass sie niemals an der Vorderseite des Gehirns, sondern immer an seiner Oberseite liegt.

That this important statement of Fuchs is not captious but well-grounded has been amply shown by the still more recent embryological researches of Schulte and Tilney (1915). In a careful study of successive stages of development in the domestic cat these authors found that:

the neural folds first meet in the region of the future mesencephalon but the closure is not simply progressive from this point in both directions. On the contrary it is incident simultaneously at several points which may be rather widely separated. In the eight-somite embryo, in addition to the closure of the mid-brain, which extends from the optic anlage to the quintal ganglion, there is a second closure between the quintal and acoustico-facial anlages; and again, after an interval at a third point the folds seem on the verge of meeting. There is also some fusion cephalad at the ventral margin of the neuropore. This is of some theoretical importance and diminishes the significance of the neuropore as a morphologic landmark.

The discussion of this point by these authors is worth quoting in eatenso.

To accept the last point of attachment of the ectoderm marked by the recessus neuroporicus as the extremity of the axis, implies that the raphe below this point is a suture between the basal plates;...further it would seem the necessary consequence of the acceptance of this landmark (recessus neuroporicus) that the mammillary and infundibular regions and the ventral half of the optic vesicles themselves were derived from the basal laminae. To accept the recessus neuroporicus as the ontogenetic pole of the brain seems, therefore, to disregard the ventral deflection of the neuraxis and the composition of the wall of basal and alar plates.

It is clear, then, that every single one of the postulates of von Kupffer has proved decidedly unsatisfactory and that the neuropore, which changes and has no homologies of the type suggested cannot be regarded seriously as a morphological landmark. The vast amount of detailed comparative embryological research which they have elicited has failed entirely to ratify his original conceptions,

Optic chiasma.

It should be called to mind, in the first place, that the traditional arrangement of the olfactory nerve, as the first, and the optic nerve, as the second cranial nerve respectively has not always remained uncontested. Thus, Mrs Susanna Phelps Gage (1905), assuming the infundibulum to be the anterior end of the neural tube, insisted that the position of the chiasma denoted that the eyes were the first pair of the segmented organs unless the infundibular organs of Boeke (1901-2) precede them in the series. In this respect Mrs Gage has not been alone, for van Wijhe (1882) suggested ‘‘Da der Olfactorius vor der Entstehungsstelle des Opticus auftritt, ist er zwar scheinbar der vorderste, in Wirklichkeit aber der zweite (nicht segmentale) Kopfnerv”’; and again he states, ‘“‘Der Opticus ist morphogenetisch der vorderste Hirnnerv, der Olfactorius der zweite.”” With this conception Hatschek (1909) concurs and so, logically, should all who, like Keibel (1889), Johnston (1909) and Kingsbury (1920-21) believe that the optic chiasma is the true anterior end of the neural tube. Sedgwick certainly upheld this view (Encyclopaedia Britannica) as recently as 1911.

The complete discussion of the evidence against such a reversal of the agehonoured sequence of cranial nerves would lead us far from the central issues of this thesis, but several comparative anatomical facts make its acceptance impossible. In the first place, the optic chiasma cannot be used as a fixed morphological landmark since, like the neuropore, it is in ‘“‘no way constant” and is sometimes situated entirely anterior to the corpora striata. As Victor Franz (1912) has stated, ‘‘Mit Kappers und friihern Autoren mache ich noch darauf Aufmerksam, dass das Chiasma nervorum opticum keineswegs konstant an einer und derselben Stelle liegt, sondern bei manchen Arten (Gadus) sogar bis oral von den Lobi anteriores verschoben ist.”

It is impossible to regard any structure which has this degree of variability as a point of morphological value. In the second place attention must be drawn to the discovery which Nils Holmgren (1918) has made recently in certain bony fishes and which he calls “ein Kuriosum.’’ That it is no mere curiosity of spasmodic occurrence is shown by its repetition in Osmerus, Cottus, Perca, ete. (cf. Nils Holmgren, 1902). This “Kuriosum” is the fact that the nerous terminalis and the nervous olfactorius utilise the optic tract as portion of their sensorial distribution through the medium of the so-called terminalis-opticus and olfactoopticus tracts. These tracts, according to Holmgren, form an appreciable part of the so-called “efferent” fibres of the optic tract.

Many years ago Edinger pointed out that the so-called “olfactory field” and the thalamencephalon is “‘completely covered laterally through the great fibre system of the optic tract” and it has been generally recognised (e.g. Judson Herrick, 1908) that “functionally and genetically, the retinae, optic nerves, chiasma, and tracts and the optic thalamus (sensu stricto) should be associated with the optic tectum of the mid-brain to form an ophthalmencephalon whose boundaries cross freely those of the classic encephalic regions.” Indeed, so clearly is this embracing relationship of the optic tract to the prosencephalon recognised nowadays that Kappers has insisted on terming the habenular commissure—which is the most posteriorly situated structure of this “embraced” area—‘“‘ telencephalic.” .

This placing of a large portion of the telencephalon posterior to the chiasm and the embracing relationship of the ophthalmencephalon to the telencephalon, form one of the most intricate puzzles in neurology. But the facts, viz. the embracing relationship of the ophthalmencephalon (Edinger, Herrick) and the variable position of the chiasma in certain fishes (Franz, Kappers and others), the absence of any decussation whatever in lowly forms like Myxine (Harris and Parsons) are only to be explained by the assumption that the optic tract, in the higher Vertebrata, has actually undergone much shifting from the primitive situation of its constituent elements; and further, that the extent of this shifting and the degree of chiasmatic formation itself has been very variable in different Vertebrata, particularly amongst Fishes.

The external situation of the eyes themselves is capable of extensive migration, in ontogeny, in fish of certain families (e.g. sole and plaice). Such migrations of the peripheral optic apparatus, in ontogeny, prepare us to expect that extensive and very varied modifications of the central or receptive apparatus were attempted before the successful chiasmatic solution was achieved. The optic chiasma was not “created.”

The researches of Holmgren, which have revealed hitherto unrecognised components of the optic tract, indicate the probable lines along which the modifications in structure proceeded to give rise finally to the optic chiasma. Presumably the original sensorium for the sense of sight was simply restricted to a segment of the neural tube lying posterior to the olfactory (and terminal) segment, Its original site may be most reasonably regarded as that now occupied by the tectum opticum. Forward growth of a portion of the optic sensorium (i.e. of the optic tract itself) in phylogeny would entail to external appearance the embracing of the olfactory field by the ophthalmencephalon, and in actual fact, the involvement of the optic sensorium (optic tract) with the sensorium of the olfactory (and terminal) nerve (tractus terminalis-opticus and tractus olfacto-opticus). The involvement of the optic tract with this primitive sensorium, now so intimate as to be apparently inextricable, can be rationally explained by this conception of a forward migration in phylogeny not only of the eyes but also of their sensoria.

The forward migration of the optic sensoria to form the optic chiasma did not occur in such a way that the ophthalmencephalon embraced the whole vertical height of the neural tube, but in such a way that the ophthalmencephalon became incorporated with the olfactory (and terminal) sensorial tissue (alar lamina of His) which primitively lay anterior to (not vertical to) it. Thus, whatever be our criterion of the anterior end of the neural tube, the caenogenetic and migrating optic chiasma is a will-o’-the-wisp the pursuit of which will merely serve to entangle neurologists in a mental morass just as long as embryology remains unchecked by the findings of comparative anatomy.

Considerations of the distribution of the histological architecture of their respective sensoria teach us therefore that the optic nerve is morphologically posterior to the olfactory nerve, and that the olfactory nerves precede the optic nerves in the segmental series suggested by Mrs Gage. It should not be surprising either, that the involved series of changes which has obviously taken place in this difficult region to produce the optic chiasma should have been productive of a developmental picture (repli cerebral transverse of Brachet) which has too frequently defied the interpretations of embryologists who fail to take account of the same facts of comparative anatomy and neurology.

The chiasma is caenogenetic and the olfactory (and terminal) apparatus lies more anteriorly. So the olfactory apparatus must be the most anterior of all chordate sense organs unless Boeke (1918) is correct in believing that the ancestral vertebrate had paired infundibular organs such as certain larval fishes ordinarily possess. Relying upon Boeke’s evidence, Mrs Gage (1905) has already suggested that the infundibular organs are the first in the segmental series of sensory mechanisms in Chordata. This line of evidence, therefore, if valid, corroborates the ancient conclusion of von Baer (and the identical one of Keith, Frazer, and many others) that the infundibular region is the most anterior and most fixed portion of the neural tube.

Certain other investigators have attempted to provide evidence towards a solution of this problem of the anterior end of the neural tube by a histological testing of the hypothetical brain morphology advanced by His. His interpreted the facts elucidated by von Baer as demonstrating an early subdivision of the neural tube into motor (basal) and sensory (alar) laminae. Consequently these observers have examined the anterior parts of the neural tube histologically to find the anterior termination of this ventral lamina (or column).

Tretjakoff (1909), by examination of the brain of Ammocoetes, showed that in front of the infundibulum there does not exist a single primary “effector” element in the neural tube; in short, that the whole of this region, topographically anterior to the infundibulum in the adult, is entirely set apart for “‘receptor” and “assoeiation” functions. In his later work, Tretjakoff (1913b) pointed out further that the researches of Boeke, Kappers, Dammerman and others bore out his own contentions that the infundibulum itself was a receptor mechanism and so reinforced his central argument that the whole of the region anterior to it in the adult was “alar” in type, and therefore was morphologically posterior to the infundibulum (on His’s own hypothesis).

But the subsequent investigation of mammalian forms by Malone (1910 and 1914) has clinched this type of evidence concerning this region for all vertebrates by showing that even in such advanced forms “the basal optic ganglion, the nuclei tuberis lateralis, nucleus paraventricularis and the nucleus tubero-mammillaris are composed of cells whose histological character indicates that they are not efferent, but are concerned in receiving and correlating incoming impulses; these cells do not possess the relatively large, diverse Nissl bodies characteristic of efferent cells.”

It needs no detailed logic to demonstrate that this “laminar” criterion of His, and those who follow him, is entirely contradictory of his previous criterion of the “raphe” as the anterior limit. If this sensory-motor subdivision of the neural tube, adopted by His, were regarded as the criterion for determining the point in question not only the neuropore but the optic chiasma, the whole ‘‘raphe,” and even the infundibulum itself are worthless as anterior landmarks. To accept the His doctrine in its entirety is impossible—either his “raphe” or his “laminae” must succumb.

These facts have been brought out to some extent in the recent contribution to the question by Kingsbury (loc. cit.), who has given us an original interpretation of the His morphology. Still, there is no reason why our morphological views should be handicapped in any way by the “‘laminar”’ conception attributed to His. It has value as a generalisation, through the pioneering work of Bell and Waller, but should not be conceived any longer in the sense originally intended by His.

Moreover, in collaboration with Dr Shellshear, I have recently put forward the conception (1920-21) that the motor neuroblasts are developed extraneurally, i.e. outside the neural tube, and are subsequently incorporated within it during ontogeny. If this conception is correct—and a very large body of facts supports it—the so-called motor or basal lamina of the neural tube only achieves that significance secondarily. The neural tube is primarily merely a coordinating mechanism composed of intercalated elements. To find the infundibulum composed of these elements does not destroy its title to being the anterior end upon our hypothesis; it would destroy that title upon the hypothesis of His. But quite apart from this discussion the motor “column” of nuclei has long been recognised as a segmentally repeated succession of cell-groups—groups which are widely separated particularly in the intracranial part of the neural tube. They form no true colwmn at all and do not end at any well-defined morphological point. This quasi-“‘column,” since it lacks continuity, can scarcely be used as defining the true anterior end of the neural tube.

Although the motor “column” has no importance in deciding the question positively, it does have great significance in supplying certain negative information, namely, that the anterior end of the neural tube, wherever it lies, is certainly not behind any motor cell group in the tube. Kingsbury (loc. cit.) has recently put forward the suggestion that the anterior end of the floor-plate of His lies a great distance posteriorly to the infundibulum and posteriorly to the oculo-motor nucleus. Creditable as is any attempt to introduce accord where so much discord has been apparent the suggestion advanced by Kingsbury is entirely retrograde. It is impossible to regard a point located at the junction of the mesencephalon with the metencephalon as the anterior end of the neural tube.

If, however, we retain the division of the neural tube which von Baer first suggested (and His later adopted) and appreciate that the “‘sensorymotor” lamination has no morphological significance (save the indirect one above referred to) it is possible, for argument’s sake, to let the sulcus terminalis end in the infundibular recess (where von Baer regarded it as ending) and not in the optic recess (where Johnston and others would have it end), nor in any other later out-pouching of the neural tube, such as the cerebral vesicle itself (where one might just as reasonably regard it as ending).

If we retain this subdivision of the tube and this termination of the sulcus terminalis, then the “raphe” of His would meet the ventral laminae and the floor-plate of His in the infundibular recess; the infundibular recess would be recognised for the true anterior end of the neural tube and the whole district of the “‘raphe”’ from infundibulum to neuropore would be recognised for the roof which von Baer, Hatschek, Tretjakoff, Malone, and so many others have recognised it to be.


Most of the discord revealed by a study of the literature of the last two decades is due to the fact that the question of the anterior end of the neural tube has been approached without regard to that other most significant issue— the anterior end of the body. Consequently the selection of the point in question has been arbitrary in practically every case, while collectively the results obtained have been highly conflicting.

The internal evidence reveals the fallacy of accepting any of the criteria hitherto suggested. The neuropore has been found to be incapable of true demonstration or homologisation. The optic chiasma, as a landmark, rests on equally unsatisfactory foundations. The “laminar” conception has not ratified any point hitherto taken and, in the nature of things, is incapable of doing so.

If we go back to first principles we will recognise that the segmented neural tube is only as old as the segmented skeletal and muscular systems. These phylogenetically coincident mechanisms are distinctive of the Chordata—their most obtrusive characteristic. There is but one single point—the only point not in conflict with the evidence of comparative neurology and ontogeny, and corroborated by teratology and osteology—which lies morphologically at the anterior end of all of these segmented systems. Lying morphologically posterior to the oral membrane (the anterior end of the body), this site is simultaneously the anterior end of the axial mesoderm and of the neural tube. It is the infundibular point.

It is from this fived point that the brain, when it became expanded, was prolonged forwards and so caused its already fused roof to face ventrally. Thus the district lying between the infundibulum and the lamina terminalis (including the chiasma) faces ventrally in most vertebrata. But this ventral “facing” of structures originally dorsal is no novelty in cerebral morphology. We do not hesitate to regard the cerebrum or cerebellum as expansions of the alar (sensory) folds, when considerable areas thereof are exposed ventrally.

It is from behind this same fixed point that the mesodermal rudiments of striated muscles have slipped laterally and have been thrust forwards to give rise to the eye musculature, jaw musculature, etc., and it is from ectoderm lying morphologically posterior to the oral membrane (and because of their. juxtaposition posterior to the infundibular point) that “mesoderm” has been derived, which sank under the ectoderm, and provided on the one hand encapsulations for various specified areas of the expanding neural tube and the sense organs connected with it, and on the other hand, sites of attachment for the already mentioned musculature. Hence the trabeculae, the cartilages of the sense organs of the skull and the branchial arches (however far forwards or ventrally they may become displaced) have arisen from tissue entirely posterior to the oral membrane and posterior to the infundibular point.

Many arguments might be urged in demonstration of this statement but, after all, they are all implicit, in some form or other, in the embryological and anatomical data which we have already considered, and further discussion would merely serve to swell unnecessarily the size of the present article. I shall merely state, in conclusion, that I do not know of a single new fact or observation amongst those which I have here brought forward, nor have I attempted to bolster up my argument with a single new picture or diagram. On the contrary, it has been my object to utilise as widely as was convenient, within the limits of a brief article, the most diverse discoveries of a long series of distinguished observers, whose statements are the more compelling because they have no prejudice; reckoning it a happy and worthy service if it should fall to my lot merely to show that these diverse discoveries, in so many spheres of investigation, were pertinent to the difficult problem in hand and afforded simultaneously the same answer.


Agassiz, L. and Gout, A. R. Principles of Zoology, 1848.

Ayers, Howarp. “Some brain landmarks, with a consideration of the neuropore, nose, hypophysis and mouth.” The Lancet Clinic, December, 1907. .

von Bakr, C. E. Ueber Entwickelungsgeschichte der Thiere, Konigsberg, 1828.

Batrour, F. M. Comparative Embryology, London, 1880-1.

Bearp, J. “The old mouth and the new.” Anat. Anzeig. Jahrb. 3, No. 1, 1888. 204. R. A. Dart

Bozxg, J. ‘Die Bedeutung des Infundibulums in der Entwickelung der Knochenfische.” Anat. Anz. Bd. xx, 1901-2.

Bracuet, A. Traité d Embryologie des Vertébrés, Paris, 1921.

Braver, A. Zool. Jahrb. Anat. u. Ontog. Bd. x, 1897.

Butscuur, O. ‘Bemerkungen zur Gastraea Theorie.” Morph. Jahrb. Bd. 1x, 1884.

ConKLIN, Epwarp G. “The organisation and cell-lineage of the Ascidian egg.” Journ. Acad. Nat. Sci. Philadelphia, second series, vol. xm, Part 1, 1905.

Dart, Raymonp A. “The misuse of the term ‘visceral.’” Journ. Anat. July, 1922.

Dart, Raymonp A. and SHELLSHEAR, J. L. Anat. Record, vol. xx1, No. 1, 1920-1.

Dourn, A. “Der Ursprung der Wirbelthiere und das Princep des Functionswechsels.” Genealogische Skizzen, 1875.

—— “Studien zur Urgeschichte des Wirbelthierkérpers. III. Die Entstehung der Hypophysis bei Petromyzon planeri.” Mitteilungen aus der zoologischen Station zu Neapel, Leipzig, Bd. tv, 1883.

Epincer, L. Bau der Nervésen-Zentralorgane, Leipzig, 1904.

Franz, Victor. Folia Neurobiologica, 1912.

Fraser, E. A. “The head-cavities and development of the eye-muscles in Trichosurus vulpecula, ete.” Proc. Zool. Soc. Lond., 1915.

Frazer, J. E. Journ. Anat. vol. tv1, Part 1, 1921.

Fucus, Fanny. Zool. Jahrb. Bd. xxv, 1907-8.

Gace, Mrs S. P. “A three weeks’ human embryo with especial reference to the brain and nephric system.” Journ. Anat. vol. Iv, No. 4, 1905.

Gorttze, A. Die Entwickelungsgeschichte der Unke (Bombinatorigneus) als Grundlage einer vergleichenden Morphologie der Wirbelthiere, 1875.

—— Abhandlungen zur Entwicklungsgeschichte der Thiere, Heft 5, 1890.

— Entwicklungsgeschichte des Flussneunauges (Petromyzon fluviatilis). Hamburg u. Leipzig, 1890.

Goronowrtsca, N. “Die axiale und die laterale Kopfmetamere der Végelembryonen. Die Rolle der sog. ‘Ganglionleisten’ im Aufbaue der Nervenstamme.” Anat. Anz. Bd. 7, 1892.

HaerckEL, E. “Die Gastraea-theorie, die phylogenetische Classification des Thier-reichs und die Homologie der Keimblatter.” Jenaische Zeitsch. Bd. van, 1874.

—— “Die Gastrula und die Eifiirchung der Thiere.” Jenaische Zeitsch. Bd. 1x, 1875.

—— ‘“Nachtraige zur Gastraea-Theorie.” Jenaische Zeitsch. Bd. x1, 1877.

Harris, WILFRED. “Binocular and stereoscopic vision, etc.” Brain, vol. xxvu, 1904.

HartscueEk, B. “Studien tiber Entwick. des Amphi. lanceol.” Arb. Zool. Inst. Wien, Bd. Iv, 1882.

—— “Mittheil. iber Amphioxus.” Zool. Anz. Bd. vu, 1884.

—— “Studien zur Segmenttheorie des Wirbelthierkopfes. 2. Mitteilung: Das primitive Vorderende des Wirbelthierembryos.” Morph. Jahrb. Bd. xxxrx, 1909.

Herrick, C. Jupson. Journ. Comp. Neur. and Psych. vol. xvut, No. 4, 1908.

Hertwic, 0. “Die Lehre von den Keimblatten.”” Handb. Entw. Wirbelt, Jena, Bd. 1, 1905.

His, W. “Zur allgemeinen Morphologie des Gehirns.” Archiv f. Anat. u. Phys. Anat. Abt. S. 346-383, 1892.

—— “Ueber das frontale Ende des Gehirnrohres.” Archiv f. Anat. u. Phys. Anat. Abt. S. 157-171, 1893

Hoimaren, Nits. Folia Neurobiologica, Bd. x1, 1918.

—— Acta Zoologica, vol. 1, 1920.

Huser, G. Cart. ‘‘On the anlage and morphogenesis of the chorda dorsalis in Mammalia, in particular the guinea-pig (Cavia cobaya).” Anat. Rec. vol. x1v, No. 4, 1918.

Husrecat, A. A. W. “ Die Gastrulation der Wirbelthiere.” Anat. Anz. Bd. xxv1, 1905.

Huxtey, T. H. Anatomy of invertebrated animals, 1877.

Jounston, J. B. “The morphology of the forebrain vesicle in vertebrates.” Journ. Comp. Neur. vol. xx1x, 1909. Kappsrs, C. U. Arrens. Die vergleichende Anatomie des Nervensystems. Haarlem, 1921.

KastscHEenko, N. “Zur Entwickelungsgeschichte des Selachier-Embryos.” Anat. Anz. Bd. mm, No. 16, 1888. Anterior End of Neural Tube 205

Keser, F. “Zur Entwickelungsgeschichte der Chorda bei Séugern.” Archiv f. Anat. u. Phys. S. 329, 1889.

——- “Zur Gastrulationsfrage.” Anat. Anz. Bd. xxvt, 1905.

Kerrn, A. and Campton, G. G. A contribution to the mechanism of growth of the human face, London, 1921. ,

Krnaspury, B. F. “The extent of the floor-plate of His and its significance.” Journ. Comp. Neur. vol. xxxu, No. 1, 1920-1.

’ Kinastey, J. S. Comparative anatomy of vertebrates, 1912.

Kuaatscu, H. “Zur Frage nach der morphologischen Bedeutung der Hypochorda.” Morph. Jahrb. Bd. xxv, 1898.

Kourzorr, N. K. Entwickelungsgeschichte des Kopfes von Petromyzon Planeri, Moskau, 1902.

von Kurrrer, C. Studien zur vergleichenden Entwickelungsgeschichte des Kopfes der Cranioten. Minchen, Heft 1, p. 84, 1893.

—— “Entwick. des Kopfes.” Ergeb. der Anat. und Entwick. Merkel and Bonnet, Bd. 1, 1893.

LanpacreE, F, L. “The fate of the neural crest in the head of urodeles.” Journ. Comp. Neur. vol. xxxm, No. 1, 1921.

LanKEsTER, E. Ray. “Notes on embryology and classification of the animal kingdom, etc.’ Quart. Journ. Micr. Sct. vol. xvm, 1877.

Lonecuames, Sgtys. Archives de Biologie, T. xxv, 1910.

Lworr, Bastivus. Zeitsch. f. wiss. Zool. Bd. Lv1, 1893.

—— “Die Bildung der primaren Keimenblatter.” Moskau Universitdtsdriickerei, 1894.

MatonE, E. F. “Ueber die Kerne des menschlichen Diencephalon.” Verlag. der konigl. Akad. der Wissensch. Berlin, 1910.

— “Nuclei tuberis laterales and the so-called ganglion opticum basale.” Johns Hopkins Hospital Bulletin, 1914.

McMorricu, J. PLayratr. “The Gastraea theory and its successors.” Marine Biological Laboratory, Lecture fourth, 1890.

Merscunikorr, E. Embryologische Studien an Medusen, Wien, 1886.

Moraan, T. H. “On the amphibian blastopore.” Stud. Biol. Labor. Johns Hopkins Univ. Baltimore, vol. Iv, 1890.

Neat, H. V. Anat. Record, vol. xx, No. 2, 1921.

Owen, R. “On the homology of the Conario-hypophyseal tract, etc.” Proc. of Linn. Soc. vol. xv1, No. 90.

Parker, K. M. “The development of the hypophysis cerebri.” Journ. Anat. vol. Lt (third series, vol. xm), April, 1917.

Parsons, Sir Joun H. “The evolution of visual perceptions.” Trans. Ophthal. Soc. vol. x11, 1921.

Pratt, Jura B. ‘Ectodermic origin of the cartilages of the head.” Anat. Anz. Bd. vit, 1898.

—— “The development of the cartilaginous skull and of the branchial and hypoglossal musculature in Necturus.” Morph. Jahrb. Bd. xxv, 1898.

Scuu.te, H. von W. and Trnery, F. “Development of the neuraxis in the domestic cat to the stage of twenty-one somites.” Annals of the New York Academy of Sciences, vol. XxIv, 1915.

Scort, W. B. Morph. Jahrb. vol. vu, 1882.

Surpiey, A. E. Quart. Journ. Mier. Sci., vol. xxvm, 1887.

Stong, L. 8. “Experiments on the development of the cranial ganglia and the lateral line sense organs in Amblystoma.” Anat. Rec. vol. xxi, No. 1, 1921.

Trersakor¥, D. Archiv f. mikr. Anat. Bd. txx1v, 1909.

—— “Die zentralen Sinnesorgane bei Petromyzon.” Archiv. f. mikr. Anat. Bd. Lxxxm, 1913.

Wenoxesacn, K. F. “Der Gastrulationsprocess bei Lacerta agilis.” Anat. Anz. Jahrb. v1, No. 2 u. 3, 1891.

van WuuE, J. W. Uber die Mesodermsegmente und die Entwicklung der Nerven des Selachierkopfes. Amsterdam, 1882.

—— Anat. Anz. Bd. vim, No. 5, 1893.

Wu, L. Zoologischer Jahrb. Anat. u. Ontog. Bd. v1, 1893.

WorErDEMAN. ‘‘Vergleichende Anatomie der Hypophysis.” Archiv f. mikr. Anat. Bd. Lxxxv1, 1914. ,

Cite this page: Hill, M.A. (2019, August 20) Embryology Paper - The anterior end of the neural tube and the anterior end of the body. Retrieved from

What Links Here?
© Dr Mark Hill 2019, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G