Talk:Paper - Contribution to the structure and development of the vertebrate head

From Embryology


Decke-»Er, JFJF XVI-»F« F.







c 0NTENTS. PAG GENEIUL IHTRODUCTION ........................................................................................ .. 498



Neecl of closer series of specimensssp Scope of the work.-Eat1iest phase of the conception that the head exhibits metamerism.Goethe, Olceth Huxlezy Gegenbautz Balkoutn

PART I.—— Mximumxetsu oF THE Hmu

Basxs For: THE Dxscussron .......................................................................... .. so; As) cranial nett-es. G) Mesoblastic head cavities. (-.·) segments of. the neural ruhe.

HIsToRIcAx REVIEW OF« THE Wen( ON NEUROMERES ........................ .. so; Von Baer ’28, Bist-hoff so, Remalc ’5o, Dursy ’69, Beraneclc ’84, Kupffer ’8 5 and ’93, Rabl ’8 s, Gegend-sur, Ort ’87, Oscak Hertwig ’88Hoffmann ’88, Mcclure ’89, Platt ’89, Zimmerman ’91, Waters ’92, Froriep ’92, Herriclc ’92, Locy ’94. DESCIUPTIONS m·- STAGES org« Acxunsgxks .............................................. 513 stages previous to the formation and closure of the neural gtoove.-stages after the closure of the groove.

SIIPPLEHENTARY OBSERVATIoNS on OTHER Adam-us .......................... .. 528 Amblystoma, Rana Palasts-is, Diemycty1us, Torpedo ocellata, chiclh 498 Lock. [v01.. x1.

PAGE If. GENERAL coNsIDnRaTIoNs .......................................................................... .. 532 I. Wes» Kapers« Cz« Bis-e Eis-HEXE»- czf ÆeZa-«-«Zk»x. Eos-ce- zo später» MJFFMEXZI i» Eos-rh- rrcxgas .................. ·; .................................................... » 332 e. Hi« rlzeyxlsixjocxask ............................................................................... » 533

(a) Different reagents used. G) Coherent history in Acanthias (x) Presence in various kinds of geruht-Yes. (o«) 0bservations on fresh materiaL

Z. Do May oft-Xa« c: CZEW sc) »Es.- Meäawewisw ej Bär: Byczfx ? .............. ». 535 4. The-J- a» joymesck Jsecfezåeiecseæzsh of Merockeryyzic Jajfzxewce ................ » 536 F. Presse-««- ezf Hex« Segment-«- ia Emåyyoøeic Ei» wen? P7«-"»«·z«i7-e Genosse« 539 H. Mem-o«- qj Fegwcazk Jkcpycxeisexsck i» Läg Brei« .................................. .. · s 39 J. Eselswies- izf «« Mezxroipzcysr tu) .S"-«rg-O-;.;»a-e.s am? Tag-»Ja! New-s .... « 541 8. Eeeck irr-ed III-Mk .................................................................................... » 543 9. Fmyxmory .................................................................................................. .. 544

Pan-r II. —- THE SUCH-Cassius.

I. THE LATERAL Evns ........................................................................................ .. 349 II Accnssoav OPTIC VESICLES ......................................... ............................ .. 556 III. THE: PINEAL SENSDORGANS ...................................................................... .. 561

I. Fragen-l- sgf Xxowzekigc reger-ji«;- LÆJ Pia-o! Fgiszskcygoar ................ .- 561 a. Lassen« ANY-i» A· IV» Einen! OMFWMZÆ .............................................. » 565 Z. cowpoyiwyx Hex-zog»- Epzpfzysiaf Ozxxgyowtfir i« Peiyowyzom Island-Er, ais-J Lesers-Zic- ........................................ ............................................ .. 571 4. Log-XI·- Moswye oft-T«- Ezozpåjmjk ............................................................ .. 576 s. Parojdäyxiæ .............................................................................................. « 576 IV. THE: Bnetnntne ori- TIIE Aunrronv Onoan .......................................... .. 577


A GnowING interest has been manifested in the prob1ems of craniaI morphology as their importance in comparative anatomy has been more fully rea1ized, and a number of streng investigators have taken up this particular Held of study. Through their researches much has been accomplished ; the views on crania1 anatomy have gradually changed, as advance after advance has been made, unti1 morphologists have come to Iook upon that wonderful comp1ex——-the head-not as a structure «« Fasse-als, but as the most extensively modified part of the anima1, formed by differentiation and specialization

. from parts that are structurally homo1ogous with those that

follow in the trunk. According to this conception, the distinction between head region and trunk region is one of degree of differentiation and not one of kind. No. 3.] THE NEXTEEXH TE Eis-in. 499

The modiiications which the head has undergone have been brought about gradually, and are so comprehensive in their range that if we could know their complete history, even in one animal, we should have a key to the leading questions of vertebrate phylogeny. But there are so many causes tending to modify the course of deve1opment, that we cannot depend on the steps of phy1ogenetic history being repeated in a complete and order1y way in any one anima1 form. 0ften a balance of probabi1ities must be struck to determine what is ancestral and whatis secondarily acquired. The chain of evidence is indeed very incomplete, and must always be supplemented by a certain amount of inferencez but what has not been fully enough recognized in the practical study of embryo1ogical development, is that the shortest interva1s of time may be very important in keeping the connection. Coherency of the history must be preserved ; and the difficulty of doing so is great1y increased by the fact that the new is made to proceed out of the old, and, frequent1y, one organ insidiously takes the place of an earlier formed one. Too great stress cannot be laid on the desirabi1ity of having a more complete series for study, and this is especially important in cranial anatomy. The traditional method has been to study one stage and then another «a little older,« and to fill in the intervening gap with inferences. This has proved to be inadequate and mis1eading. It is now required that we shall have stages close enough together to trace the history of the transitory as well as the permanent Organs. The practical diiiiculties in obtaining a sufficiently complete series are very great, and, in many cases, well-nigh impossible. Great effort has been expended in getting the materia1 for the present research; it is a kind of material in which the stages cannot be contro11ed, and my series cannot be regarded in any sense

pas a complete one. Nevertheless there is represented in it

several distinct stages that have not heretofore been described by students of e1asmobranch embryology. Beard, in his study

on the Transient Ganglion Cells and their nerves in Raja batis.

says : « My series of the embryos of this form is what many might judge to be complete, numbering as it does Some zoo 500 Lock? [VoI« XI.

specimens of all ages and sizes. Nevertheless there are gaps in the collection and these are often of a kind that it may not be easy to ftll in. No two embryos are exactly alike in all the pictures which they yield of this apparatus, and in half a dozen specimens that would be taken to be of the same age from their sizes, from a comparison with Balfour’s stages, or from the more certain criteria of number of gi1l—clefts, protovertebræ, etc» etc., it is quite common to lind this transient nervous System, like other Organs, in widely different stages of deve1opment and presenting great variations in detailed characters. In my researches- on Raja I have been compelled to give up completely any attempt to make use of Balfour’s nomenclature. With a limited number of embryos at one’s disposal, it is easy to tit them into one or the other of the well-known stages ; with an increased number this becomes more difiicult or even impossible, so great are the variations met with« I have had a sirnilar experience with my embryos of Squalus acanthias (-4c-mtJ-z2r.r -z-2«Iga»«.s). I am conscious there are gaps in the Material, and, after making the best use of it that I can, I have, no doubt, missed many things not represented in my collection.

The studies recorded in this paper deal almost exclusively with the early history of the brain and sense—organs. Some of the questions upon which direct evidence is brought are: What was the primitive condition of the nervous system of the Vertebratesk What was the number and nature of the primitive neural segments entering into the braink What has been, in generaL the line of modification along which they have been converted into the braini And what were the early steps in the differentiation of Sense-Organs?

I have been greatly indebted to Professor c. 0. Whitman for courtesies both at Woods Holl Biological Laboratory and at the University of chicago, and I have also to thanlc him for advice and suggestions on the subjecvmatter of this paper!

I soon after the appearance of the Zieglers’ paper (Beiträge zur Entwicklungsgeschichte von Torpedo : Art-Eis- jiir Ali-E. ais-at» Bd. 39, Akt. I, Jan. I892) Dr. Whitman suggested to me that I should work over the same gkouncl—— the gastrw lation and formation of the germinal layers—in one of our North American No. 3.] THE« IZEJEIEBJPÄTE HEXE. 501

In glancing over the views regarding cranial anatomy that have been he1d since the beginning of this century, we find a suggestive hint in the shift of opinion, as to the decided1y relative nature of all our views on morphological Problems. There has been a distinct rise in the point of vier-v, but scarcely any of the earlier Problems are yet solved ; new ones have been added, but those that have been handed down have grown broader and increased in proportions as a mountain on nearer approach. There has been a gradual change in front, keeping pace with the advance of our knowledge and with the changes in our methods of interpretation. During that period the very sou1 has been breathed into the body of comparative morphology, and as a resu1t, our interpretations are directed towards

explaining anatomica1 structures in connection with their past deve1opment. At the beginning of this century, on account of superiicial

»appearances, the conception took rise that the head is divided

into definite segments. The obvious division of the slcull by sutures was gradually worked into the theory that the cranium represents a number of modified vertebræ; and then began on the part of anatomists, the efforts to determine the number of

these segrnents or vertebræ in the sku1l. Olcen and Goethe

formu1ated the theory which was taken up by that eminent anatomjst Richard 0wen, who lavished upon it an amount of attention and labor that was worthy of a more substantial theory, but the mistalces of that great man have proved of benefit to other rnorpho1ogists.

This was the beginning of the idea that the head region represents a deiinite number of segments. Originally founded on

external features of no segmental importance whatsoever, and

not essential to the question which they served to introduce, the problem gradually widened and deepened and reached the essential parts connected with this segmental condition. The Hrst conspicuous change of front came with the delivering of

species of Elasmobranchii I collected material (Galeus canis and Squalus

acanthias) for that purpose, and. began studies along that 1ine, but, tinding new .

and undescribed conditions in the head region, my attention was gradually drawn off from the original purpose and directed towards cranial anatomy. 502 Lock: kvizkn XI.

the croonian Lectures, in 1869, by Huxley, in which he completely overthrew the vertebral theory of the sku1l, withdrawing attention from the superHcial sutures in the cranium and directing it to the cranial nerves and branchiæ as bearing evidence to the segmentation of the head. Gegenbaun in I872, studied the cranial nerves especially in relation to the branchial clefts and reached the con(:1usion that there are nine segments represented in the head. Another distinct advance was made by Balfour, who first studied the segmental divisions of the mesoblast in the head of the Elasmobranchs, and identified by this means eight head-somites clearly represented. He also expressed the conviction that there were primitively a larger number of segments but, owing to extreme modifications of the head region, they are no longer clearly represented. This was at bottom the same problem, but it was now shifted upon Organs that are truly segmental.

From the time of its discovery, this segmental division of the mesoblast in the head became a great favorite with morphologists in elucidating the problem of head segmentation. The mesoblastic divisions seemed, so far as the evidence went, to embody the most direct survivals of the original segmentation and, therefore, to be the most promising line along which to work out the problem. Valuable contributions have been made along this line since Balfouns time, by Marshall, Van Wijhe, D0hrn, Killian, 0ppe1, and others, and the myotomes of the head have continued to hold their high Position in the minds of morphologists as the most signiiicant remnants of the original segmentation.

Notwithstanding all these researches the original problem is still unsolved. There is no agreement as to the number or the nature of the primitive segments, and about the only point that may be regarded as settled beyond controversy is that the head and brain were primitively divided into segments.

Reeentlzg there has been added as a factor in the diseussiom observations on segmental divisions of the neura1 tube. Although such divisions have attracted the attention of several observers, their importance in the problem of cranial segmentation has not been appreciated. They have been regarded as of secondary origin, depending on the segmentation of the mesoblast. Nevertheless, I hope to show that these segments are the first to appear in the head region, and that they are entitled to more serious attention on the part of anatomists, as representing the most primitive segmentation of the head of which any traces are preserved.


I. Basis Fon THE DIscussxoim

The more recent discussions on the metamerism of the head are based upon segmental divisions as shown in (I) cranial nerves and branchial clefts, (2) mesoblastic head cavities and (3) segments of the neural tube.

The firsdmentioned basis may now be set aside as involving too much conjecture. Mcclure has stated the objections to it as follows : «We have positive proof that the degeneration of certain branches has taken placeY This being the case, we have every reason to assume that who1e segmental nerves may have once existed, which have completely degenerated, leaving no trace whatever of their previous existence If such be the case, the segments origina1ly connected with these degenerated nerves must necessarily be overlooked, if the existing nerves are made use of as a means of determining the original number of segments.

« Furtherrnore, the vagrant changes in the position of some of the cranial nerves must necessarily cause confusion. For examplcz take the sixth nerve, which in the frog and tadpo1e xstages is situated between the first and second roots of the ninth nerve, a position somewhat posterior to its place of origin. This remarkable shifting clearly shows not only what great

schanges in position the cranial nerves are capable of under—

going, but it also goes to prove that we can find no reliable

means of determining the primitive segments by means of their sconnection with the exit of the existing cranial nerves. Beard

in taking up this problem made use of san important series of

Sense-Organs for which he has proposed the name of sBrans

chial Sense 0rgans,’ from their development from thickenings 504 Lock. kvon XI.

of the epiblast over each branchial c1eft. The dorsal branchesv of certain crania1 nerves fuse with these epiblastic thickenings ; the superficial part of the thiclcening giving rise to a branchial sense-organ, while the deeper portion becomes the gang1ion of the dorsal root of the crania1 nerve. This close relation which exists between the dorsal branches of the crania1 nerves and their corresponding sense—organs is undoubtedly of segmental Character. But this line of research is beset by a great dikticu1ty, namely, that the degeneration of certain branchial sense-organs would, in time, involve the degeneration of their corresponding crania1 nerves, and such degeneration has certain1y taken place, in part. or in whole, leaving in doubt the primitive segments with which they were connected."

The second and third p0ints mentioned are more important c1ews to the metamerism of the head. Muscle and nerve are, physiologically, so fundamentally related that we should natural1y expect Some close correspondence between muscle segments and neural segments, and metamerism of the head region should be studied in light of the work done on both sets of structures.

The myotornes (or muscle segments) have received by far the most attention as they are the more conspicuous, but it is timely to ask whether they afford the most re1iable evidence as. to the primitive number of brain segmenta Comparative study shows that the neural segments are the Hrst to appear and are less subject to rnodifications than the muscle segments of the head. The large number of myotomes described in the head of se1achian embryos by Dohrn and Killian are more transitory than the neural segments. The period in which they are exhibited is a short one, and soon the seventeen or eighteen Segment-s of Killiartz and the eighteen or nineteen of Dohrn,. become reduced, by fusion, or absorption, or both, to the nine head segments of Van Wijhe

The neural segments, on the other band, begin very early, as shown in this paper, and preserve their original number and characteristics through several embryonic periods. It will bei seen as we proceed in the account of these. segments, that the assumption cannot be sustained, that the segmental divisions of the middle germ—layer (protovertebræ) are primitive. NO. 3.] THE IXEKTEBÆÄTE HEXE. 505


The question of Metamerism of the Head as based upon myotomes has been comp1etely reviewed by Dohrn, Ki11ian, and others; Isha11say nothing on that side of the problem, but shall limit the historical review, and confme the discussion to the side of the question that has been less cultivated

It is a fact of comparatively recent discovery that the whole neura1 tube of vertebrates is divided by constrictions into Simi1ar segments. Each Segment is bounded, anteri0rly and posterior1y, by transverse folds; and the elevated area between them constitutes the Segment to which the name metamere is given. These segments may be picturecl to the mind as a series of transverse ridges and furrows occupying each side of the neural tube and not extending across the median plane. They are exhibited in very young embryos of Vertebrates and disap— pear before what may be called the middle embryonic period. The existence of such folds in the wa1ls of the hind-brain has been known since the time of Von Baer, who in I828, first observed them in the embryonic tchiclc of the third day of development ; but it was not unti1 1889 that they were known to extend throughout the length of the neura1 tube.

since Von Baer’s time they have been observed and commented upon by various anatomists. Bischoff1 Hgures the neural segments, but does not mention them either in the text or in the descriptions of the iigurea His figures show seven folds in the region of the fourth ventricle of a dog embryo of the twentyssiifth day of development. There are also shown three additional folds in the region of the mid-brain.t

Remalg in I850, made important observations, and suggested that the segments in the hind-brain are connected with the origin of the nerves in that region. He noted five or— six quadrilateral Helds on each side of the hind—brain wa11s, calling attention to the fact that they correspond (:1osely in Position

I I am greatly indebted to Hoffmanns historical review of the literature in Bronrks Klassen und Ordnungen des Thierreichs I have consulted nearly all the literature referred to there, but, in some few cases, where the original papers have been irr-accessible, I have depended wholly upon bis review of it. 506 Lock: kvon XI.

with the viscera1 arches, and with the cranial nerves «which grow with them« According to his observations they fade away after the iifteenth day.

Dursy observed them in I869, in the embryonic cow, of 6.5 mm. in 1ength. He recorded the occurrence of six folds in the regioni of the fourth ventricle Foster and Ba1four, in I874, noted the same structures in the chick, and suggested that they were of segmental importance. Dohrn, in I8y5, called attention to the occurrence of eight or nine neural segments in the fourth ventricle of bony Hshes He contrasted this early segmentation with segmental divisions in insects. Gotte figures such segments in the hind-brain of welldeveloped ernbryos of Bominator. In I877, Mihalkovics was inclined to interpret these segments as due to mechanical pressure of the mesob1ast, and, therefore, not a fundamental feature of the medullary tube.

s Beraneck shovved, in I884, that there is a definite connection between certain of these segments and crania1 nerves, thus giving the first real foundation for establishing their segmental re1ations. In his earlier paper, he describes five pairs of transverse folds in the hind-brain of embryos of Lacerta agilis from 3 to 4 mm. tin length He noticed that they rapidly fade away and disappear in embryos 5 or 6 mm. in length. In I887, he studied the relations of these « replis medullaires « in the chick, and, as regards their connection vsksith cranial new-es, reached similar conc1usions. I »

Kupffer maintained in 1885 that these segments indicate a primary metamerism of the medullary tube. He has published several brief notices on these structures In I884, he gave a record of studies on» the brain of the trout, in which he found five pairs of neural segments in the Und-brain. In sagittal Sections he noted, in addition to these, three pairs in the mid— brain. , He found no segments further forwards, and concluded that the foresbrain is not to be included in the segmented region.

In the following year («85) he gave the results of his studies on embryos of Salamandra atra. In embryos of that form, showing as yet no traces of protovertebræ, he found No. 3.] THE« yEJe III-sen DE« HEXE. 507

eight pairs of neural segments in the median part of the hindbrain. It is to be carefu1ly noted that these segments observed by Kupffer were in embryos with a wide open neural groove, and occupied the median part of the cephalic p1ate, thus giving a ssmediane Gliederung des Hirnes.« In I893, in his « Vergleichende Entwicklungsgeschichte des Kopfes der Kranioten,« he gives figures (20xx and 20 F) of the forms described in I885. These iigures of Salamandra atra show segmental folds only in the median part of the neura1 plate, and none in the neural ridges. This is interesting when compared with my observations on amphibian eggs (see p. 529).

In addition to the eight segments in the brain region he (:ounted thirteen or fourteen in the cord, extending baclcwards to a point a litt1e in front of the blastopore.

Rabl («85) spealcs of unmistakable segmentation in the hindbrain of chick embryos of from fifty to ninety hours« incubation. He found seven or eight segments in the region of the fourth ventricle-——-not being able to determine deHnitely whether there were seven or eight. Again, in 1892, Rabl has most ably discussed the question of the metamerism of the head, but as his paper deals almost exclusively with segmentation in the mesoblast, it does not come in for attention in the present connection.

Oscar Hertwig gives the matter passing attention in the third («88) edition of « Lehrbuch der Entwicklungsgeschichte.« He is not inc1ined to attach much importance to the neura1 segments.

Gegenbaun also, does not loolc upon these particular segments as important factors in the metamerism of the head. His position on the question is shown by the following quotation so frequently met with: «So interessant und so vielversprechend diese Thatsachen sind, so wenig scheinen sie mir gegenwärtig geeignet, zur Beurtheilung der Metamerie des Kopfes selbst als Factoren in Geltung gebracht zu werden»

0rr, in I88y, traced very definitely the connection between these segments in the hind-brain and cranial nerves. In describing the segments he made use of the term « neuromeres," which has been generally adopted on this side of the Atlantio 508 Lock. kvon XI.

He describes six in the hind—brain of the lizard (Anolis), giving their anatomical characteristics with great clearness He observed no neuromeres behind the point of origin of the tenth nerve, nor did he find them in the fore— and mid—brain, but he conc1uded, hypothetica11y, that they were present in the anterior brain regions. Orr found the Hfth, seventh and eighth, ninth, and tenth nerves, respectively, connected with the first, third, fifth, and sixth neuromeres of the hind-brain.

Hoffmann, in Bronn’s « Klassen und Ordnungen des Thierreichs « (1888), records his observations on these segments in Lacerta and Tropidonotus He found seven in the hindbrain of these forms. In the following year he added further details in the Zoolog-STIMM- Änaezszszm He differs somewhat from 0rr as regards the relationships of the crania1 neu-es, assigning the Hfth nerve to the second neuromere of the hindbrain, the seventh and eighth to the fourth, the ninth nerve to the sixth, and the tenth nerve to the seventh neuromere. From the first Segment the fourth nerve arises, and subsequent1y shifts its Position forwards.

Mcc1ure, following 0rr’s work, demonstrated the segmentation of the neural tube throughout its who1e extent, and published a preliminary announcement of the same in 1889. He

showed the presence in the spina1 cord of segments continuous

with those in the brain, and histologically simi1ar to them. He

examined these structures in the chicken, Amb1ystoma, and the

lizard (Anolis). He fixed upon six in the chicken and lizard, and five in Amblystoma, as the number in the Und-brain of each respectively. He found two in the fore-brain, but left the number in the mid—brain undetermined, expressing the view, however, that there are two neuromeres in that brain region. Thus he identifies ten neuromeres in the entire brain region, and agrees with 0rr in the assignrnent of nerves to the neuromeres of the hind-brain.

Miss Platt’s work (1889), on «Axia1 Segmentation of the,

chickenZH agrees, in so kar as neurorneric segmentation is con-cerned, with that of her predecessors, except as regards the relation of the nervedibres to their corresp0nding neuromeres. According to her observations they spring, primarily, from the NO. 3.] THE« IJEJETEBÆÄTE HEXE. 509

concavity between two segments, and not from the crest of a neuromere. My observations on the motor roots agree with those of Miss Platt in that particular.

Zimmerman («9:I.) states that he Hnds in embryos of the rabbit and chick, shortly before the c1osure of the neural groove, the segments observed by Kupffer in Salamandra atra. He noticed at first eight of these segrnents (encephalomeres) in the brain region. The three anterior ones were much larger than the five lying behind them in the medu1la. The three front ones are the vesicles of the fore-, mid-, and hind—brains, and they straightway undergo secondary division as follows: The first divides into two, the second into three, and the third into three, making a total of eight secondary divisions arising from three primary ones. These added to the five of the medulla give a total of thirteen segments in the brain region. He also observed these structures in Acanthias and Muste1us, and found them very clearly defined In mamma1s the metameres of the nnd-brain are not so distinct. Zimmerman goes on to say that these folds cannot be accidenta1 appearancea since in all classes of vertebrates corresponding nerves arise from corresponding segments He gives a kahle, showing nerve re1ations, with too

i much detail to reproduce here.

Waters, whose complete paper appeared in June, I892, studied especially the mid-brain of Teleosts He confirmed and extended the observations of Orr and McC1ure. He counted eleven neuromeres in the entire brain regiom six in the hindbrain, two in the mid-brain, and three in the fore—brain. He did not lind neuromeres in the brain of the Cod earlier than the ninth day of development He assigns the olfactory and optic nerves to the anterior two neuromeres The two neuromeres of the midsbrain give origin to the third and fourth nerves, and from the six segments in the hind—brain the näh, seventh, eighth, ninth, and tenth nerves arise as designated by Meclure The sixth nerve he found to occupy its theoretical position when the neuromere exists; when fusion has taken place between the trigeminus and abducens neuromeres the sixth nerve has been shifted backwards between the seventh and eighth nerves. 5 ro Lock. Not« XI.

Froriep gave a noteworthy contribution to the subject of neuromeric segmentation in very early stages, before the Anatomische Gesellschaft of Germany, at the June meeting in I892. He described anew the so—ca1led neuromeres that he had previously observed in mole embryos, but concluded that they are not of true morphogenetic signiiicance He further described the conditions in Triton embryos, and concluded that the so-called primary neuromeres detected by Kupffer in those animals are simply the result of underlying mesoblastic somites.

Froriep agrees with Kupffer in tinding segmental folds while the neural groove is widely open, and in locating them in the median part of the cephalic plate. But, whereas Kupffer linds eight in the brain region of Salamandra atra, he tinds only four in the corresponding region of Salamandra maculosa, and five in Triton cristatus (Compare with my observations, p. 529.) His general conclusion is that «the jointing of the vertebrate body is originally determined by the middle germlayer; when ectodermal structures exhibit segmental arrangement, it is the result of secondary adaptation.« «

Herriclc («92), in a preliminary paper, gives an account of neuromeres in the Ophidian embryo, in stages after the com— plete closure of the neural groove, and after the formation of the ear vesicle. His figures show six neuromeres in the medulla. He states a proposition that will be of use later, in helping to distinguish between primary metamerism, and metameres of secondary origin which show after the closure of the neural tube: « If neuromeres once existed in the fore-brain they would be visible only at an early stage . . . The so—called fore-brain neuromeres differ from those of the medulla and cord in involving only dorsal structures."

The present writer, in I894, gave the first account connecting the earliest formed neuromeres with those of later stages He showed that in sharlcs they arise very early, and may be traced without a break through all the stages of the open neural groove into the structures that have, in later periods, been designated neuromeres. The segmental divisions extend to the anterior tip of the fore-brain, and are distinguishable in that region for a brief time after the closure of the neural groove NO. 3.] THE FEÆIEBJBÄ TE III-IF. 51 I

He also recorded the presence of neuromeres in Amblystoma ernbryos with wide open neural groove, but differs from Kupffer in locating them in the neural ridges instead of the median neural p1ate. (See further on this point, p. 530.)

From the foregoing historical survey it appears that our knowledge regarding metamerism in the neural tube has passed through the phase of simple observation of its occurrence (Von Baer, «28, Remalg Ho, Dursy, So, and others), and has grown by successive additions to the recent conception of its segmental importance. In reaching this point, Hrst came the work of Beraneck («84) and Orr (-87), showing that the neuromeres of the hindsbrain are definitely connected with nerves; following this it was demonstrated by Kupffen partly, in I886, and by Mcclure, definitely, in 1889, that the neuromeres extend throughout the neural tube, and that those of the trunk region merge gradually into those of the head region. Lastly, the neuromeres of the mid-brain have been especially studied by Waters («92), the condition of that brain region having been left undetermined by previous observers. Remak was the Erst (I85o) to suggest the segmental relations of nerves and neural segments; Beraneck the first (I884) to demonstrate it. From this time onward the definite relation of nerves and neuromeres began to be studied, and both Orr (-87) and Hoffmann («88) are pioneers in this line of study.

It will be observed that previous to the appearance of my paper just referred to, no one but Kupffer and Froriep had claimed a very early appearance for neural segments, and these two authors had recorded their appearance only in the median part of the neural p1ate, and not in the neural ridges. They had not shown the structures of the open neural groove stage to be in any way connected with the neuromeres of later periods, which are present in the lateral walls of the neural tube. g

In the case of Froriep’s observations I thinlc there is reason to doubt whether the segments observed are really the «neural segments « of other writers. The small number (4 or s) which he observed in the head region does not correspond with the number of neuromeres observed by any other author in the 5 1 2 Lock. [Vo1-. XI.

same region. I thinl(, also, there is a way to bring Froriep’s observations into reconciliation with my own (see p. 529).

At all events, it has been understood from the work of previous observers that the neural segments arise after the neural groove is closed, or while it is in process of closing. Waters («92) carries the idea throughout his paper that the metameric segmentation arises relatively late, especially in the Teleosts, where he was unab1e to lind any traces of this segmentation earlier than the ninth day of deve1opment, after the auditory pit is formed. 0rr and Mcclure do not in every case state ages, but from their flgures and the text I understand that they have not detected this segmentation in very young stages. Miss Platt mentions the fact that these segments sometimes occur in, the chick while the groove is open. Rabl mentions them as being especially clear from the iiftieth to the ninetieth hour of incubation in the chicken

I have been fortunate enough to lind these neural segments in a number of animal forms in extremely young stages, and in Squalus acanthias, to trace them coherently onward into the latet stages. In this form, the division of the embryo into segments takes place before the neural groove is formed, and, before any protovertebræ have made their appearance, the metameres not only extend the whole length of the embryo, but they are continued for Some distance into the embryonic rim. They occur under such conditions in this animal that they cannot be interpreted as depending on mesoblastic segmentation. In Amblystoma and the newt (Diemycty1us) the metameric segmentation is present in the rudiments of the neural fo1ds, just after their Erst appearance and during their period of broadest expansion. In living chick embryos of about the twentieth hour of incubation they can be made out with clearness along the walls of the beginning neural folds. It is only in Squalus acanthias, however, that I have traced the complete history of these neuromeric segments. NO. 3.] THE PEJETEBÆÄTE HEFT-D. 513


The ear1iest stage in which I have detected this rnetarneric segmentation is represented in Pl. XXVII, Fig. 25. This is an age somewhere between Balfour’s stages B and C. It is, in reality, the youngest embryo of Squa1us to which I have had access since I began to observe especially the metameric segments in that animal. Whether or not they occur in still younger embryos I do not know, but they are already clearly defmed in the stage referred to, and it is reasonable to suppose that they may be seen in still earlier stages.

The axial embryo (Fig. 25) is just fair1y established, and has reached a length of 1kIzmm. The headsend is already wider than the rest of the embryo. It has begun to show that tendency to broaden that is characteristic of the head-end of the embryo. The gastrular cavity is broad, and extends to the extreme anterior end of the embryo. In the iigure it is seen even protruding beyond the head-p1ate. The primitive furrowz that has often been confused with the neural groove in these Elasmobranchs, is broadened at its anterior end. Fig. 63, Pl. XXIX, is a sketch of a horizontal Section of this embryo to show the general appearance of the metameres in Section. i

The segmental divisions in this embryo extend from the anterior end backwards along the margins of the axial part of the embryo, and out into the norkaxial part or embryonic rim. There are seven or eight pairs of these segments in the embryo, and as many more, directly continuous with them, in the embryonic rim. The latter is segmented to the points where it is broken from the rest of the blastodernr Whether

or not these segments extend further into the blastodermic

rim, I am unable to say. The segments are most clearly defmed a1ong the inner margin of the embryonic rim, and extend more faintly across it.

In the axial part of the embryo they are not in such a favor ab1e position to be observed from above—-—-they are on the rounded mai-ging; but if the embryo be rolled into such a position that the margin is brought into view, its division into 514 LOCK [Vo1.. XI.

segments is more plain1y seen. Near the middle part of the embryo the 1ines of segmentation are faintly traceable from

the margins towards the median furrow. The two lines of

segments are joined in front by a single median piece or segment. This unsegmented anterior tip becomes more prorni— nent in the irnmediately following stages There is no evidence to show whether this represents the primitive anterior Segment or several aggregated anterior segmenta These segments, once estab1ished in this very early stage, may be traced onward in an unbrolcen continuity until they become the neuromeres of other observers, and sustain deiinite relations to the spinal

and crania1nerves. Ryder, in I88I, observed segmental divi sions extending into the embryonic rim of Elacate, one of the Teleosts. In I885, he Hgures such structures in a stage in which the neural groove is c1osed and the eye vesicles are well established. Although the iigure shows a considerab1y later stage than we are now dealing with, and he does not speak of their earliest origin, nevertheless, the feature of their extending beyond the embryonic axis into the blastoi derrnic rim agrees with my observations on Acanthias and I thinlc it not improbable, that Rydefs segments correspond with those I have described. These segrnents, observed by Ryder under such unusual conditions, have generally been interpreted by morphologists as due to precocious segmentation in the non——axial mesoderm. The segmentation I have just described is not capable of such interpretation, for sections show that the mesoderm is not yet divided into protovertebræ at this stage, and that the epiblast is the seat of the tsegmental divisions. The mesodermic somites of Squa1us are formed later in the usual way, and the first ones appear in the trunk or neck region at a later period.

In Fig. 26 the embryo is relative1y more slender in the trunk region, and there is coming to be an observable distinction between the broadly expanded cephalic plate and the narrower bodzn Upon the anterior end there is being forrned a pro— truding unsegmented median tip, which is much better seen in Figs. 27 and s. The median furrow ends in front in a broadly expanded depression. The gastrular cavity has become narNo. 3.] IYJE FXEJETEBJBÄ IIE HEXE-D.

rowed in fr0nt by folding in of the sides, so that the embryo, when viewed directly from in freut, seems mounted on a kee1; the kee1, however, is not so1id, but contains the anterior part of the gastrular cavity, which still reaches to the anteri0r 1imit of the head. For the purpose of fixing the stages as defmitely as possible, I give anatomical characteristics not immediately connected with the metameric segmentatioIL The embryo from which the ftgure was Inade measured ITBZ mal« CUT I.—-— Embryo and blastoderm oi Acanthias

in Iengkjh but· there is so just before the kormation ok the neun! Leids. d( about 8 diameters

much individual variati0n in the size of embryos of apparently the same age, that the length is not very signiöcant There are, in that part of the embryo

» «» sc.

« »«- . » As« - H. X r XX? , «, — -« . »; —


Cin- 2. —The same ernbryo x about 40 diameters Meeatneres no: ehe-sen. The travsverse lines and number-s indicate the Plane of the Sections shown in the succeeding out.

behind the cephalic region, three or four mesodermic somites rather imperfectly differentiated The metameric segmentation »5 I 6 LOCJT [vor.. XI.

is very clearly exhibited al0ng the lateral margins of the neural plate, extending from the unsegmented tip backwards, and, as in Fig. 25, is continued in the embryonic rim to the points on either side of the latter, where it is broken from the rest of the blastoderm While it is the lateral margins that are most clearly divided into segments, in the trank region the lines of division may be traced inwards towards the median furrow. This is probably due to the appearance of the mesodermic

ctrr z. ——Twelve traust-erst: Sections ok the embryo Figur-ed. in the pteeeciing sent. d( about Zo die-merkte. The number-s reier to the positions ok the. Sections in the set-ice. ·

sornites in that region. Fig. 64 represents a horizontal Section of this embryo showing metameres in the ectob1ast.

Fig. 27 is in many respects similar to Fig. 26 ; it is slightly older and has reached a length of 2 mm. and shows about five mesodermic somites. Diverging furrows have appeared upon the cephalic plate that include between them a wedge—shaped Central piece which terminates in the anterior unsegmented tip before mentioned. The cephalic plate is thus separated into a median and two lateral parts. lt will also be noted in this iigure that the lateral margins are marked off from the rest of the medullary plate by two furrows running lengthwise of the ern— bryo, so that the plate is bordered, as it were, by marginal bands. The furrows are most disti11ct in the head region, but No. 3.] THE: NEXTEBJM TE Eis-im. 517

they extend also, with 1ess distinctness, into the trunjc and fade away without reaching the hinder extrernity The furrows do not show so distinct1y in cross—section as one would at first suppose, and they are in part an optical effect, arising from the way in which the neural folds are formed This will be understood on reference to cut s. There is, neverthe1ess, a distinct notch to be seen in the crosssectioris of many specimens, while in others it is laclcing. I am inc1ined, with my present Iight,

CUT 4. — Embryo of Äcanthias just after the formation of the neural fo1ds. d( about 40 Manier-ers. Metameres not shown. The traust-esse lines indicate the plane of the Sections in cut s.

to consider these furrows as pure1y mechanical effects Sections show (cut z) that the marginal hands, at a stage just younger than this one, are composed of an accurnulation of cells forming thickened cords running along the margins of the embryo. These bundles of cells and their immediate derivatives are the rnaterial out of which the neuraI ridges and a large part of the medullary fo1ds are straightway produced.

As noted in the preceding embryo, the tnetarneres are most isclearly seen from below, but the reason for this is not far to 518 Lock. kvon XI.

see1(. The condition of the neura1 folds in this animal is very unusual: when Hrst formed they are 1ateral, wing-1ike expansions, extending along each side of the embryo, overhanging the yolk (cut 5). No sooner are they formed than they bec0me ventrally cum-ed, and, in this way the most clearly segmented

Co? s. -—«I’welve ttansverse Sections of the emhryo of the pkecckling out. z( about zo diameteka The neural folds are expandecl laterally beyonkl the body and veuttally curved.

parts of the embryo are brought ventra1wards, and this accounts for the metameres being most distinct when viewed from the ventral surface

Figs. 28 and 29 represent two views of the same embryoz it is an older stage than that represented in Fig. 27. The neural fo1ds are now fully formed and their ventral curvature is very marked. In Fig. 28, the optic vesicles OF) are seen on each side of the Central tongueslike Process to which attention has already been ca11ed. No. 3.] THE PEJBTEBJEÄTE HEXE. 519

In Fig. 29,1 the view is taken from below ; the embryo has been removed from the b1astoderm and placed upon its dorsal surface and the recurved edges of the neural folds are thus brought prominently into view. This is of course the most favorable position for making observations The dish contain— ing the embryo should be placed over a black, non—reflecting background, and the embryo rotated into the most favorable Position with a, fme artist’s brush.

In the actual specimen from which the figures were made the segments showed most beautifully. They appear like a row of beads running along the ventrally recurved margin, and extend with great distinctness the entire 1ength of the embryo. Those in the trunk region are continuous with those in the head and pass into the 1atter without any transition forms. There is, however, some individual variation in size of the neuromeres, and they are not abso1utely symmetrical on the right and left sides, but the signif1cant thing is, there is uniform1y the same number on each side in a given region, such as the hind—brain, or the brain region as a who1e. Fig. 29 shows the Central unsegmented piece from below with three segments on either side of it, occupying a part of the headfolds that is directed forwards Following the beaded edge, from the head into the trunk region we find it disappeap ing from view beneath the expanded walls of the gastrular cavity. viewing the same embryo from above (Fig. 28), the metameric segmentation is seen to extend the entire length of the embryo and, as in the earlier stages, lateral1y into its expanded parts. The segments are so plain that they may be easily counted There seems now to be a natural landmark sseparating the cephalic plate from the rest of the embryoz this is an abrupt downward bending (j) in the medullary folds which, as I have determined, lies just in front of the future origin of the vagus nerve. There are eleven metameres in the lateral margins of the cephalic plate, including the ones embraced inthis fold.

Fig. 3o represents an older stage, in which the medullary

I ln the process of transferring and shading, the outlines in this ögure have been rendered too symmetricah giving to it a semidiagrammatic characters. 520 Lock: kvon In.

folds have unrolled from their ventrally curved Position and are in the process of growing upwards Those of the head are at this stage nearly in the horizontal plane (cut 6). The outer margins of the folds are p1ain1y divided into segments, and the segrnentation extends backwards, also, into the trunk region, but not so clearly deftned. The optic vesicles are clearly seen on the head-plate, but the accexsory optic vesicles (see p. 57) have not yet made their appearance Beginning at the front end and counting baclcwards eleven segments on either side, we come to the point where the broadly expanded cephalic plate

CUT H. — six transverse Sections ok an embrzso older than the preceding one. X about so diameters. The numbers below the Sections reker to their Position in the series. The neural folds are growing upwards ancl in Section 24 liave reachecl the horizontal Plane. The depressions in Sections 3 and 12 are the optic vesicles The embryo from which the Sections are made is shown on Pl. XXVL Fig. J.

passes into the narrower neclc and trunk. This, as before indicated, is the point of future origin of the vagus nerve. It seems to me to be a natural line of division which may be of Service in determining the limits of the embryonic head. The question will be returned to on p. 543.

It should be borne in mind that all the stages so far described are very young ; the ear1iest ones are before the formation of the embryonic medullary folds, and the oldest one is just when the Inedullary fo1ds are arching tipwarcls to form, for the first time, a medu11ary groove The mesenchymic somites have, in the interim, appeared in the trunk region, and have NO. 3.] THE« IXEJBTEBJLEL TE III-CI. 5 21

produced faint surface indications in the median parts of the medullary plate.

Taking a considerable step forwards in the history of these segments, we come to the condition represented in Fig. II. This iigure showS a stage in which the medullary folds have attained a nearly vertical Position; they are about to bend towards each other and meet in the median plane, but, as yet, they have not become approximated in any part of their course, and, therefore, we have an open neura1 groove (see cut 9, p. 553) extending the who1e length of the embryo. In this Hgure the embryo is viewed obliquely from the right side. The rudiments of several organs have now appeared upon-the head ; the most anterior of these organs is the primary optic vesiclez just back of this, near the margin of the head—folds, is seen a similar e1evation that represents the combined vesic1e of the mid—brain and the first act-array- optic vesicle (see p. 556). Still further back in the same 1ine, is another simi1ar but smaller elevation, which, I think, represents the second accessory optic vesicle. Behind the latter structure the niargin of the medullary fold is bent abruptly downwards; this is a normal condition in this stage, and is of course found in the earlier stages. Baclc of the primary optic vesic1e and somewhat between it and the midbrain vesic1e, is a rounded eminence which is the external indicati0n of the mandibular cavity, and behind this is the branchial pouch from which the branchiae are subsequently formed. The front end of the gastrular cavity is being cut oft· a1ong with the increase in the head flexure. i

Directing our attention to the margin of the right medullary fold, we note that it is clearly segmented through the head region, and backwards into the trunk region, where, in the Hgure, it disappears behind the yo1l(. The metameres extend in reality to the posterior part of the embryo. There has been a slight change in position of the foremost segments with reference to the rest of the head-p1ate. The three anterior ones are no 1onger, as in Fig. 29, on a part of the margin that looks forwards, but they have been shifted backwards, and that part of the margin that was anterior, now constitutes a part of the lateral border. Of course this shifting of position is brought 522 Lock. kvon x1.

about by changes in the medullary fo1ds. The first three metameres are, at this stage, in front of the eye, the fourth, f1fth, and part of the sixth, in front of the accessory optic and midssbrain vesic1e. The following iive segments (seven to e1even) occupy the refiected part of the neural fo1d. The eleventh, as has been indicated, lies in front of the vagus nerve.

The next stage to be considered (Pl. XXVII, Fig. 32) is just after the c1osure of the neural groove, which is, however, still open in front by a small neuropore. The original segments are still visible throughout the head region; those in the fore and mid-brains still show from the outside. The surfacemarkings on the head are similar to those in Fig. II. In front is the conspicuous optic vesic1e, and behind it is another similar rounded eminence, the mid-brain vesic1e. Further back there is an area, circular in out1ine, resembling the midbrain, except it is not so prominent. It is, however, merely a pad of mesoderm applied to the walls of the cerebellum There is another surface indication of considerable interest, via, a line of mesob1ast running over the branchial region; it connects in front with the mandibular cavity and behind with the body protovertebraen

Fig. 33 shows a slightly older stage than Fig. 32. The neura1 groove is complete1y c1osed. The extreme anterior end of the gastrular cavity has been obliterated The cranial flexure is quite marked. This is the last stage in which the metameres are visible throughout the length of the embryo; those in the fore- and mid-brain have become indistinguishab1e in surface views; they are, however, still to be detected in longitudinal Sections. We possess now a particular advantage in dealing with these segments, because anatomical landmarks of the head regions have become established, and these enable us to say with defmiteness what are the relations of the segments to the rest of the headJ This is just prior to the appearance of the auditory vesicle; when first established its center occupies the space of the Segment marked Io. Some— times, in its earliest stages, the circu1ar area spreads over the space of the three segments marked 9, to, and II, but I No. 3.] THE PEÆTEBJEÄTE JJEXJD. 523

should say from my observations that, more frequent1y, it is not so widely expanded It always settles down in squalus acanthias, to occupy the position first indicated, and, subscquent1y, it is shifted backwards The topography of the head region is similar to what it was in Fig. 32. The Chief differences to note are the further differentiation of the branchial arches and clefts; we may now distinguish the position of the future first visceral cleft, and, faintly outlined, the boundaries of branchial arches.

The Segment marked 8 serves as an important landmark in all subsequent changes that affect the segments It is seated above a depressed region in which the first visceral cleft subsequently appears, and, during all the time the segments are distinguishable from the outside, it has no nerve root.

Only a few words of description will be needed to enable us to follow the history of the metameres through the later

stages. In Fig. 34, the auditory vesicle (x:2-) has been formedp

and the first visceral cleft has broken through. The anterior metameres 1ying in front of the one marked 6 are no longer distinguishable from surface observation. The lines of neuromeres have been brought into contact in the median plane by the closing of the neural groove, but they are soon forced apart by the growth of the dorsal wall of the neural tube. We have now reached the stages, approximate1y, in which these neuromeres have been described by previous writers, —-—that is, the stage just after the appearance of the auditory vesicle, but it is to be remembered that Kupffer and Froriep have noted a form of segmentation in very early stages of Amphibia affecting the neural plate, but not the neural ridges (see p. 529).

In Fig. 35 Some characteristic changes are to be noted ; the auditory vesicle has shifted backwards till it occupies-a position opposite the eleventh metamere ; the metameres are being forced apart laterally by the growth of the dorsal wall of the hind-brain. The eighth metamere still serves as a landmark ; there are now two clearly marked metameres in front of it and three behind it. The only metameres discernible from surface view are those belonging to the hind-brain. r The mid-brain is considerably increased in expanse, and the first accessory 524 Lock: kvon XI.

vesicle has been crowded forwards into the region of the thalamencephalon This may be seen after the removal of the overlying layers of mesoderm, etc. (Pl. XXVIIL Fig. 44). The kifth nerve is already well begun, and nerve iibres are also given off from the segments numbered 9 and I0.

Fig. 36 shows an embryo slightly older than that in Fig. 3 s. The line of neuromeres have been forced further apart by the 1ateral growth of the dorsal wa1l of the hind-brain. The ear— vesicle is no longer circular in outline, but is fast becoming a closed pouch. The eye shows the beginning of the lens and the choroid f1ssure. The Anlagen of the Iifth, seventh, eighth, ninth, and tenth nerves are distinctly visible from surface observation. The branchial arches are all clearly outlined and the first two gill—clefts have broken through. The specimen shows about forty-f1ve mesodermic somites. A line of surface elevations over the hinder branchiae mark the beginning of the lateral line.

In Fig. 37 the neural segments are undergoing sorne changes in outline that are lilcely to lead to confusion in identifying them in later stages. If, for example, we loolc along the lower margin of the segmented border, we shall see that the elevations and constrictions are substantially as they have been in all the previous stages, but those along the upper margin no longer correspond with them. In all the preceding Hgures the boundaries of the segments correspond on both upper and lower margins. In Fig. 37, however, the upper margin shows e1evations just above the constrictions on the lower margin. These newckormed elevations become very quickly prominenh while the segments along the lower margin lose their individuality, and the segmented area becomes more and more an irregular sinuous band with crests upon its upper margin. The entire line of segments finally becomes indistinguishab1e, but if they be studied in stages irnmediately following that represented in Fig. 37 it will be the crests along the upper margin that first catch the eye. If the observations are made from above, these crests are seen to be transverse folds on each side of the medulla, and when counted will, of course, be one less than the original segments. It is only by viewing NO. 3.] THE« IXEÆTEFÆÄTE HEXE. 525

them from the side, and comparing them with ear1ier stages, that we shall be able to identify the boundaries of the original segments.

Just what is taking place during the appearance of the crests is not now sclear to me. I have heretofore assumed that it signified a union of the original sog-meinte, the anterior half of one with the posterior half of the Segment lying just in front of it, but at present I am inclinecl to question that interpretation The crests on the upper margin are between two neuromeres as designated by 0rr, and they correspond in Position to his inner ridge. The point of origin (motor Hbres) of the Hfth, seventh, and eighth nerves (so far as it may be determined by surface view) is now c1ear; they arise, as Miss Platt has described them in the chicken, from the concavity (on the lower margin) between two neuromeres. This will receive fuller consideration under the headingz The Nerves. lt will be interesting to note incidenta1ly, in this Hgure, the very large development of mid-brain over that in Figs. 33 and 34, and the consequent crowding forwards of the first accessory optic vesicle. The latter structure is also much reduced in size, and with its fellow is in the region of the thalamencephalon

None of these Hgures have shown the condition of the entire neural tube. The segmentation so clearly seen in Figs. 34, 35, and 36, represents only a part of the actual segmenta-· tion, sei-Ja, that in the uppermost part of the neural tube. The rest of the tube is too much covered by mesoblast to be seen without dissection The upper part of the tube has——-—in the region of the hind-brain-—-two thiclcened lateral bands of cells, which form a border on either side of the neural tube; these are the segmented parts that are visib1e from surface observation. I have found it necessary to remove the overlying layer of mesoblast and the outer epidermic stratum, and completely expose the walls of the brain. When thus laid bare, the walls of the neural tube show in a most satisfactory manner. The ten or twelve ögures following those just described show dissections of this kind. i

Pl. XXVIIL Fig. 4I, shows the surface view of an embryo slightly older than that represented in Fig. 33, and Fig. 42 526 LOCJÄ [VoI.. XI.

shows the same embryo with the overlying tissues removed from the brain wal1s. It is clear from a comparison of these two figures that the line of metameres seen from the surface view are those occupying the upper part of the neural tube, and that below them the entire neural tube is divided into corresponcling segments. The segments (lo not reach acrosS the median plane, but they are alike in number and position on both sides of the tube.

Fig. 43 represents an embryo in which the auditory vesicle is just forming. The embryo was intermediate in age between those shown in Figs. 41 and 34.

Fig. 44 shows a dissection of an embryo of the same age as the one represented in surface view in Fig. 34.

Fig. 45 is talcen from an embryo just younger than that represented in Fig. 35. By the growth of the dorsal wall of the hind—brain the line of metameres have been forced apart. This expanded dorsal wall, being a new growth, is not divided into segments. i

Fig. 46 represents a slightly older embryo with the optic vesicle and also the auditory vesicle removed. There are in this figure eight segments in the hind-brain that show plain1y, and faint indications of a ninth.

Fig. 47 represents a dissection of the embryo from which Fig. 35 was made, and therefore a direct comparison can be made between the two figures

Fig. 48 shows the condition of the brain walls in an enibryo just older than that represented in Fig. 37. From the continued growth of the dorsal wall themetameres on each side have become widely separated. The eapcapsule has not been removed. i Fig. 49 represents a slightly older embryo than the for-egoing one. The auditory and the optic vesicle have been removed. There are now distinctly nine segments in the hindbrain region. The U-shaped Segment, No. I2, in the hind brain lies opposite the ear-capsule. This Hgure shows well the

condition of the neuromeres described in Fig. 37, in which there is no longer (as in earlier stages) a correspondence between the ridges on the upper margin and those on thelower margin of thesegmented lateral bands of the neural tube. No. 3.] TJJE IJEJE TEFJM TE HEÄ D. 5 2 7

Up to this point the figures described have all been magni— Hed uniformly 45 diameters; but on account of the increase in size of the embryos it will be better to carry forward the history of these segments with ftgures drawn on a smal1er scale. Accordingly Figs 50-60, inclusive, are magnified only ten diameters.

Fig. 50 represents an embryo of the same age as that shown in Fig. 32.

Fig. 51 shows the entire embryo, partly dissectecL of which Fig. 43 is a portion more highly magniiied Behind the figure is seen the 1ine of fusion of the 1ips of the blastopore

Fig. 52 is a sketch of the embryo of which Fig. 47 is the enlarged view of a partia1 dissection. They all show well the segmented condition of the walls of the hind—brain.

soon after the age represented in Fig. 55 is reached, the neural segrnents fade away. Figs. 57 and 60 (PI. XXIX), represent the head region of older embryos in whichthe segments are no longer visible

Taken togethekz the iigures now described give a comparatively full view of the neura1 segments in different ages. They show them from their first appearance to the time they fade away. We learn from this examination that the neural segments are established before any embryonic organs appear, and that in the early stages they extend not only throughout the length of the embryo, but into the embryonic rim. In the earliest stages the segments are a1ike, and there is no Structura1 distinction to be made between those in the head and those in the trunk, or even those in the embryonic rim.

In sagittal Sections the neuromeres are well exhibited Fig. 72 shows a Section of a specimen just after the closure of the neura1 groove in which the five neuromeres belonging to the fore- and mid—brain are exhibited The second neuromere coincides in the median plane with the neuropore. This is also to be seen in stirface view in Fig. 32. Very soon the anterior brain regions become so much modified and expanded that the original segmentaI divisions are no longer visible

»Figs. 66—7I (Pl. XXIX), show six successive Sections of a specimen slightly younger than the one just described. In 528 Lock [vox.. x1.

front are seen three of the brain vesicles, —-—those of the fore—» brain, the mid-brain, and the cerebellum. The thalamenc:epha1on and the prosencephalon do not show as separate parte. In the region of the hind-brain are seen three neuromeres espe— cial1y well developed They present the appearance of three bars ; they are the seventh, eighth, and ninth neuromeres respectivelzn The other neuromeres are present, but they do not stand out with such distinctness as the three mentioned. When the ear vesicle first arises it makes its appearance opposite the ninth neuromere.

Figs. 73, 74, and 75 are Sections of a somewhat older embryo after the ear vesicle is estab1ished. In these figures eight neuromeres of the hind—brain are visib1e. The ear vesicle is opposite the tenth neuromere. Just in front of it, in Fig-s. 73 and 74, are the roots of the eighth and seventh nett-es, respec— tively, those from the former nerve being connec:ted with the tenth neuromere, and those of the latter with the ninth neurowere.

In Fig. H, the iifth nerve is seen to have connections with the first and second neuromeres of the hind—brain, ins» the sixth and seventh neuromeres respectivelzn

As already indicated, the eighth neuromere bears no nerve,

and Hoffmann remarks, « This seems to be the case in all Vertebrates."


I have also made some supplementary observations on these neural segments in Amblystoma, Diemyctylus, Rana palustris, Torpedo ocellata, and the chiclc In all of these forms the metameric divisions are to be found in very early stages before any of the embryonic organs have been formedz and they are in all essential features like those I have described for squalus acanthias.

Figs. 113 and 114 (Pl. XXX) show camera sketches of an Amblystoma egg, with broadly expanded neural plate and widely open neura1 groove. "The neural folds or ridges are divided throughout their length into a series of segments with no espe— No. 3.] THE IJEJZTEEKH H: Eis-im. 5 29

i cial distinguishing features between those of the head and those

of the body region. The median plate included between the neural ridges is smooth at this stage; at a slightly later peri0d, however, while the groove is still widely open, the median plate exhibits very faint transverse markings

The contrast between this condition in Amblystoma and that in a closely related egg (Rana palustris) is szvery instrucs tive. In the 1atter(Fig. III) the cephalic plate is very obviously segrnented, while the folds in the neural ridges are extremely difftcult to see. We thus have in these two closely related eggs strikingly different conditions In the one it isthe median plate material that is thrown into obvious folds, while those in the neural ridges are well—nigh indiscerniblez and in the other the conditions are reversed. This is not, however, to be taken as indicating profound differencesz for a little (:arefu1 observation shows that the median divisions do not correspond to those in the neural ridges, and therefore the median fo1ds in Rana palustris are not to be cornpared to the primitive segmental fo1ds in the neural ridges of Amblystoma. careful observation shows, also, that there are seg— mental divisions in the neural ridges of Rana palustris that do correspond, in number and general charaoteristics, with those in the neural ridges of Amblystoma. These latter segs mental divisions are extremely diflicult to see in Rana palustris. There are four or Ave median transverse divisions in the cephalic plate of that form, while there are ten or eleven segments in the neural ridges of the same region.

These facts throw light upon the apparent discrepancy between the observations of Kupffer and Froriep and those I published in a preliminary artic1e. Both the former authors observed segrnents in the median plate of amphibian embryos,

s and none in the neural ridges, while I have ftgured segments

in the neural ridges of Amb1ystoma, and none in the median plate. i But my later observations show that the appearances

even in close1y related eggs of the same age are not necessarily identical. »

Froriep agrees with Kupffer as to the position of the seg ments -in the early stages, but not as to their number. In H 30 « LOCK [Vo1«. XI.

Salamandra maculosa he found only four segments in the head region, while Kupffer found eight in the same region of Salamandra atra (Fig. 1 Iy). All other observers have found eight or more in the head region, and Froriep stands alone in identify— ing so small a number of segments in the head. In Rana palustris there are about four obvious divisions in the median plate of the head (but, in addition, there are other segments in the neural ridges to the number of ten in the same region). This affords a suggestion that the segments observed by Froriep (see Fig. II6) correspond to those I have observed in the median plate of Rana palustris, and not to the segments in the neural ridges, which are evidently homologous with those in the same Position in Amblystoma, Acanthias, and other forms.

Whether we Hnd the median plate smooth in Amblystoma or faintly segmented depends upon the stage at which the ex— amination is made, and we recognize that the appearances in any one egg are not constant throughout the open groove stagez and, further, that eggs of closely related animals are by no means necessarily similar at corresponding stages. The failure to take things of this nature (which are really common enough) into consideration has, I thinlc, given origin to many differences of opinion and formed the basis of many a contro— versy. This should make us careful in comparing results to have the same material in precisely the same stages.

In Diemyctylus the conditions are very similar to those in Amblystoma. The neural folds show metameric divisions in the stages with wide-open neural groove. In the three am— phibian forms I have examined there are about ten pairs of

segments in the broadly expanded neural folds of the head.

In Torpedo ocellata I have also noted the occurrence of this metameric segmentation in several stages. The youngest embryos of that animal I have had corresponds to the stage designated « c « by the Zieglers, and beginning at that point I have traced it along through several stages with a widely open neural groove. This form is so closely related to the one I have especially studied that we would naturally expect—as is the case-—marlced similarity between the neural segments. NO. 3.] IJJE FEJZTEFJBÄ DE? READ. 531

Torpedo is not so favorable for the study of the segments as» Acanthias The folds are much fainter in the former than in the 1atter, but the significant thing f is, that the number in a given region in Torpedo corresponds to that in Acanthias.

In the chick these segments may be detected as soon as the neural folds are establishechbefore there is any trace of protovertebrae A small number is visible in th—e b1astoderm of the twe1fth hour of incubation just as the head-fo1ds are first out1ined. Their history has been carefully traced by one of my students, Mr. F. A. Hayner, and it agrees in all essentiai features with the history of the same segments in Acanthias Cut 7 shows the appearance of these segments from surface view in a chick embryo of about twentyckour hours incubation There are eleven segments in front of the firsdformed protovertebrae The cell-arrange— ment in the metameres, as shown in Sections, is the same as that described by 0rr in neurorneres of older stages of the lizard (Ano1is).

From these supp1ementary observations it is c1ear that the occurrence of primitive neural segments, as the Erst dehnite expression of rnetamep «« ism, is not an isolated case in C» ,.,,E,,,bsp» «« »die« »Hm Ho» Squa1us. They occur in the same MOSOITOTIIIIO stsmitss »« Ihm« 45 very early period in al1 the other L: forms examinech and in all of them vided throughoutthsircntitslcngtb precede any adivision of the meso— Tat« primitive metamericsegments’

» » while there are but kour mesodekmicz derm into protovertebraa spmzkcs kanns, LOCR [VOI«. XI.


I. Wege) Erz-Erz« oj »Es» Proåzwkz oj JVaZcZMer2·F-ø.

The neural segments have now been shown to occur in extremely young stages of a number of animala The fact of their presence in these early stages once established they assume new importance They have too deflnite a history to admit of being set aside as mere beadings or undulations of no metameric signilicance When I Erst hegen, Some two years ago, to notice these segments in extremely young embryos, I attached no particular significance to them ; but a comprehensive study has convinced me of their segmental importance and everything taken into consideration they furnish, I thinlg a more satisfactory basis for interpretation of metamerism of the head than we have had before. I have pointed them out to several observers, in Amblystoma and the chick, and have found no one who had ever noticed them before ; but through the kindness of others I have had my observations as to the number in the head region confirmed. Since these Structur-es have been overlooked in the earliest stages, it will not be out of place to say a word about their general appearance, and how to see them.

It is extremely difiicult to represent them on paper just as they look. All my drawings are camera tracings, and so far as number and arrangement of the segrnents are coneerned, have been verilied and reverijied over and over again. They are a little too distinct in Some of my drawings ; in making the figures clear enough so there shall be no doubt as to what I mean, their distinctness has been somewhat exaggeratekl On Pl. XXVI I have given a few photographs from untouched negatives that show these segments more as they appear in

the actual specimens. To observe them successfully is largely

a question of getting shadowa 0ne can loolc directly at the crenated margins without seeing the segments at all, but by tilting and rotating the speeimens until the proper angle of illutnination is found, they may be seen with more or less distinctness, according to the general state of preparation of the No. 3.] THE IXEJE TEEXH TE fix-m. 5 3 3

Material. I have never seen a shark embryo, of the proper age, of any method of preparation, in which I could not detect them. A dead—black background is, of course, the best surface for observing anything of this kind by reflected light A white surface is occasionally recommended, but it is desirab1e to cut off all reflected rays except those coming from the specimen, and one may see delicate structures on a black surface that cannot be seen at all on a white background

The care with which the specimens are prepared makes a great difference in the c1earness with which these structures may be seen. The conditions under which most material is hardened are unfavorable There is usually an albuminous fluid in contact with the embryo, and this, together with minute fragments of yo1k, coagulates when the Hxing reagent is used, and forms a coating over the embryo. The embryos should be washed by a very gentle jet of the reagent immediate1y after their immersion, and the clouded reagent should be removed and rep1aced by clear fluid. This makes the preparations beautifully clear.

2. Ä» »Es-J- Jlyxsfcxcxs ?

There is one question that must be answered for all new structures, wie» are they artifacts —produced by the reagents used? Too great precaution cannot be taken in considering this question.

I have used the following reagents in preparing my material: picro-sulphuric acid, picro-nitric acid, Flemming’s stronger solution, DavidofPs c0rrosive-acetic, chromic acid with a trace of osmic, corrosive sublimate removed with iodine. I have observed Acanthias material prepared by all these various methods, and in every case have found the segments as I have described them. I have always taken the precaution to count the number in the head region, and uniformly have found the same number of segments there, regardless of the hardening reagent used. I have had a complete parallel series of specimens hardened with DavidofFs corrosivesacetic and Kleinen— berg’s picro-sulphuric, and have compared the two, step by step, 5 34 Lock: kvon x1.

one set serving to fu11y corroborate the observations made on the other.

In Acanthias I have fol1owed the history of these segments very carefu11y, and have traced the earliest-formed ones wie-Izozzx ·:- Zsracxå into the later stages, and identiiied them with the neuromeres. If, therefore, the segrnental f0lds of the open neural groove stage are artifacts, it may with equal force be claimed that the so-ca1led neuromeres, which are later stages of the same segments, are also artifacts.

It should also be borne in mind that simi1ar segments exist in correspondingly early stages in Arnblystoma, Rana palustris, Diemycty1us, and the chic1(. It is tru1y interesting to observe the way in which these different kinds of material bear out the interpretatiom that we are dealing with veritable anatomical structures

The most satisfactory reply to be given to the question is based on study of fresh Material. Fortunately, the chick offers at all times a source where we can get living embryonic material at any desired age. I have repeatedly observed these segments in living chick embryos I in the eighteen to twenty-twohour stages, and have treated them with reagents while they were actually under observation. The effect of the addition of picrcksnlphuric aeid is to render immediately the walls of the neural groove opaque and more clearly dejined, but not to affect the number or arrangement of the segments camera sketches have been made before the addition of the reagent and compared with those made after the specimens were hardeneii The two correspond, as regards number and arrangement of the segments I have also noted these structures in living embryos between the twelfth and fifteenth hours of incubation Their consecutive history has been traced in my laboratory by

one of my students, and the results may be expected to appear later in this Jovis-rat» o

I have likewise observed them in living embryos of Amblyss

I The embryo was removed with a part of the blastodermto normal salt solation, which was lcept at the proper temperature, and the specimens tilted and rotated with a very fme sable brush over a black background The structures are, of course, faint, and de1icately, although detinitelztz outlinecL No. 3.] THE »Er-e THE« H« III-i D. 5 3 5

toma in the open neural groove stage and have compared them with hardened specimens of the same age.

Z. Dr) III-ej- czjmii a: THE-Ze- w M« Jlfexæmerzsrw czf Mk« Brei« ?

If we grant that these are tru1y segmenta1 structures, the question still remains : Do they afford the best or even a good clew to the number of segments in the primitive brain ? If so, they must be shown to be equal1y important in this direction with myotomes, branchiæ and cranial nerves.

In estimating the claims of these various forms of segmental divisions, to rank as primitive, the question of the time of their appearance in developmental history will be signiiicant On this point I wish to observe, that in all forms I have studied—-— embracing representatives of Birds, Amphibietz and Selachians —the neura1 segments are among the first anatomica1 Structures to be establishedz before the vestiges of any organs have appeared, the embryo is divided throughout its length into similar segments These metameric divisions, therefore, antedate myotomes, branchiæ, crania1 nerves, or any other Structures that exhibit metamerism. They persist through the early stages of developmenh and become deiinitely related to segrnental nerves and segmental Sense-organs. 1n the light of their very early appearance and their history, I think we are justiiied in saying that they are the most satisfactory traces of primitive metamerism that is presented in the group of Verte— brates. They may be loolced upon as a survival of that genera1 segmentatiom which all agree in assuming for the (not too remote) ancestra1 form.

1t should also be observed that the entire embryo is segmented, and the term «Metamerism of the Head « should be understood to signify merely regional metamerism and not a different kind of segmental division from that occurring in the rest of the embryo.

The study of Sections shows that the neural segments are distinctly differentiated groups of cel1s, and not merely a series of undulations. As has been already pointed out the arrange ment of ce1ls is like that described by Ort« for the neuromeres in Anolis 5 36 Lock. [v0x«. XI.

4. Tfzey o« joymezs Jssøciepesedewxzsx H« Maroziaymic ZEIT-Maske.

The next point I wish to maintain with regard to these segments is that they are formed independently of mesodermic influence

In a preliminary account, I went so far as to state that the segmentation is epiblastic, but that statement should be qualiHed. It is certain1y most clearly expressed in the epiblast, but the other germdayers apparently feel the same iniluence, and partake of this segmenta1 division in a modified degree. It is least c1ear in the rnesoderm — in Sections it may be made out in this layer, and also in the endoderm. The so—called neural segments are developed throughout the entire 1ength of the embryo before any protovertebræ are formed and, therefore, they must be independent of any formative influence of the latter.

There is also a notable difference in the two forms of metameric division ; the formation of protovertebræ, as is well known, proceeds from a certain point forwards and baclcwards, and if the segmentation I have described were rnoulded over the former, we should expect the neural segments to appear gradually, keeping pace with the formation of protovertebræ

Neverthe1ess, the majority of authors who have touched upon the question have talcen the position that segmental division ok the mesoderm is primary and that the neural segments are moulded over it, but all this time the early history of the neural segments has not been known. Kupffer announced, in I885, that segments appear upon the median neural plate of Salamandra atra while the groove is widely open and before the appearance of any protovertebrae He designated this primary metamerism Seven years later, Froriep («92) studied the question of metameric segmentation upon closely related forms (Salamandra maculosa and Triton cristatus), but reached different conclusions as regards the nature of the segments.

Froriep’s conclusions are based on both surface study and Sections. He shows in very young Triton embryos a seg mented condition in the neural plate included between the No. 3.] THE PEKTEBJEH In« JJEHZZ 537

incipient neural fo1ds. He expressly states that the segmental divisions do not extend beyond the plate into the neural ridges. The latter, he says, become segmented laterz only, when in the process of c1osing of the neural groove, they are brought directly over the originally segmented plate, and are made to feel the effect of the under1ying segmented mesoblast. The number of segments which he detects in the neural plate at this period is very small. He fmds only four in the greatly expanded anterior part.

My observations (already given on p. 53o), differfrom his in

a fundamental way: showing, in Amblystoma, Rana palustris and the newt, the presence of unmistakable segments, in the ear1iest formed neural ridges, and not less than ten pairs in the broadly expanded head—p1ate. · In Squalus acanthias, also, it is the incipient neural folds that are the most plainly segmented, and in this animal they are so situated that they cannot possibly come under the influence of the mesoblast in any mechanica1 way. It will be remembered from descriptions of Figs. 27 to 3o and from Sections of the same (cuts 3, s, 6) that the neural folds are formed as wing—like expansions, extending laterally beyond the body. The mesoblast does not accompany these outfo1dings of the epiblast, and when they best exhibit the segmenta1 divisions no mesoblast enters into them.

As for the rest, Froriep gives figures of Sections to show that the mesoblast is at this very early period actually divided into somites that correspond to the external segments His figures certainly show segmental divisions, but being wood—cuts they are not entirely satisfactory when the question arises whether

they are real1y protovertebrae Froriep interprets them as

such and closes his article as follows: « Die Gliederung des Wirbeltierkörpers ist ursprünglich an das mittlere Keimblatt gebunden; wo sich als Produkte des äusseren Keimblattes segmentale Anordnungen finden, sind dieselben durch Anpassung an die Metamerie des Mesoblastes sekundär entstanden«

The internal condition Hgured by Froriep is interesting when compared with that in Squalus acanthias at corresponding stages. 1 have found in that animal an undulated condition of 538 Lock: Not. XI.

the mesoblast, but it is not clear in my mind that we are justified in looking upon the divisions as primitive mesoblastic somites ; before seeing Froriep’s paper I had considered them as undulations, probably due to the same influence that has thrown the epiblast so clearly into segments. In many embryos of Squalus, showing epiblastic segmentations, they are, so far as I can determine from Sections, entirely lacking It seems to me that they are rather to be looked upon as a consequence of primitive segmentation and not as protoverte— brave. This view is substantiated by the fact that the formation of the latter takes place in the usual manner at a later period. Considering the late stages in which the neuromeres have genera1ly been described, it was a natura1 inference that they arise after the neural tube is established Minot, in expressing his conception of the formation of the neural segments, based on the descriptions of 0rr and McClure, says : « Their appearance seems to depend on the development of the primitive segments of the mesothelium. When the segments are fully formed, and before their inner wall has changed into mesenchymal tissue, they presst against the medullary tube and oppose its enlargement; at 1east one Sees that the tube becomes s1ight1y constricted between each pair of segments and slightly enlarged opposite each intepsegmental space.« But the neural segments appear so much earlier than the primitive segments of the mesothelium, that this interpretation is no 1onger tenab1e. i i · The facts made known will tend to materially modify the current theory of metamerism, which assumes as fundamental that metameric divisions begin in and depend on the mesoblast. As Adam sedgwick says, in a recent paper: « The Segmentation which obviously persists in the trunk region, and which begins with segmentation of the mesoderm, and is moulded upon it in the manner characteristic of all metamerically seg-— mented animals."« It seems to me a strained inference, that the middle 1ayer—the last germxlayer to be formed—should be the bearer of this primitive metameric division, but, of course, the final appeal must be made to Sections and my sectioned material has given evidence entirely corroborative of No. 3.J THE IJEXTEBJJH DE: Erz-ro. 5 39

the surface observationsz Sections of the embryo i11ustrated in Fig. 25 show that the mesob1ast is nowhere divided into segments, while the epiblastic somites extend throughout its entire length When the stage represented in Fig. 26 is reached, the first mesoblastic somites (three in number) are to be detected, in the trunlc regionz the rest of the mesob1ast remains undivided, but the epib1astic segmentation is as before. I have also made a carefu1 study of the Sections of the chicken embryo illustrated in cut J, and can Hnd no segmenta1 divisions of the mesoblast outside the protovertebræ (iive in number), but the metameric divisions of the epib1ast extend from the extreme anterior tip back to the point where the neura1 fo1ds fade out.

s. Presse-e« rj ist«-Fa Fegmwzrs i» EMÆJJOJZFC Eis» am? Prämisse-sFrone-sc.

Another point to be noted regarding these primitive segments is their range. They do not end at the posterior 1imit of the axial embryo, but extend on either side into the embryonic rim. As the embryo increases in 1ength, those in the embryonic rim are apparently drawn into the axial embryoz at any rate, I do not find evidence of extension by budding in the axial embryo, and, in the early stages, the number of neural segments therein increases inidirect ratio to the lengthening of the embrya This would indicate that at least some of the materia1 of the embryonic rim contributes to the e1ongation of the embryo.

In the chick the walls of the primitive groove are also divided into segments that are similar to those that appear in the neural fo1ds.

6. JVIJJEZZY oj Fegkøzaipzxs ygpkaxejøxmk F» IX« Brei-z.

There is considerable variance in the conclusion of different observers as to the number of neura1 segments represented in the brain. 0rr, McC1ure, Waters, agree in identifying six in kheihindbkain ok the— chick and ward, and Ave in the hindbrain of Amb1ystoma. In estimating the entire number« in 540 Lock. [Vo1«. XI.

the brain, there is to be added, according to Mcclures count, four for the c0mbined fore— and mid-brain, and, according to Waterss five for these regions. Thus, Mcclure considers the brain to represent a total of ten neuromeric segments, and Waters makes a total of eleven segments.

The European investigators, in general, have found a larger number of these segments in the hind-brain, and usually have not observed them in the fore—brain. Dohrn gives eight in the bony ftshes, Rabl seven or eight in the chiclg Hoffmann seven in Reptilia ; Froriep counted twelve in the entire brain of mole embry0s, and Zimmerman gives thirteen in front of the vagus The fore-part of the brain has been regarded as non—metameric, and the question of the anterior extension of segmentation is an important one. In regard to that point, the evidence furnished by the neural segments, shows that the segmentation extends throughout the fore—brain. This corresponds with the investigations of Whitman on the nervous System of C1epsine, in which he shows conclusively that the cerebra1 ganglia of that anima1 belong to the metameric series.

Recent observations on a different set of segments, sei-Jan, segmental divisions in the mesoderm of the head, have brought to light a re1atively large number of cephalic segments Dohrn, in I890, observed eighteen or nineteen mesodermic segments in the head of Torpedo marmorata of 3 mm. in length. Killian («91) in the following year describes seventeen or eighteen in Balfour stages F« and J of Torpedo ocel1ata. He also makes a correction to Dohrn’s enumeration, making it correspond with his own. ln later stages of Selachians, as Van Wijhe has shown, they are reduced to nine, or to ten according to Miss P1att.

There is such a fundamental relation between muscle and

nerve that we might expect myotomes and neural segments to

bear a direct numerical relation. My observations show a

smaller number of actual segments in the brain-walls than that tarrived at by Dohrn and Killian through studying the myotomes of the head region. In the hind-brain I have found» eight

segments represented in the earlier stages, and a ninth segment which, to all appearances, belongs to those of the spinal cord, but later this ninth Segment becomes clearly associated N0.3.] THE· FEÆTEBÆÄTE HEXE. 541

with those of the hind-brain. I have taken this to signify that the hind-brain has encroached on the territory of the .spinal cord, and has embraced at least one Segment origina11y be1onging to the cord. It is reasonable, on theoretical grounds, to suppose that such a process has taken place and has been several times repeated

In order to bring the segments of the brain into satisfactory evidence I have completely laid bare the brain-walls by removing the overlying layers of mesoderm and epidermis The segments after being exposed in this way stand out so clearly that I feel considerable coniidence in my count of them in Squalus acanthias. In addition to the nine segments in the hind-brain, I have found five in the combined fore- and mid-brain. In the latter particular I agree with Waters, but, of course, differ as regards the total number in the brain. Mcclure says (p. 39): «I will show that Dr. Hoffmann is probab1y wrong in considering the hind—brain as consisting of seven segments, and that the Segment considered by him as the iirst Segment of the hind-brain is rather the posterior Segment of the mid-brain; in other words, it is the second neuromere of the mid-brain." The recently adduced evidence is all in favor of Hoffmanns observation. According to my observations, there are still represented in the otogeny of squalus at least fourteen paired neural segments belonging to the brain region. There should be added to this enumeration the median unsegmented tip which terminates the line of segments in front, and, as stated above, there are reasons to suspect (a1though no direct evidence in that structure) that there may be more than one segment included in that terminal piece. If the terminal piece represents a single pair consolidated, then there are ftfteen neural segments in the earliest condition of the brain of Acanthias

J. Æekxxioøs qf M« Magre-May« sc) Fææsskcygcms gis-ei Erst-ein! New-s.

The Sense-Organs and cranial nerves, undoubtedly, at Hrst sustained definite segmental relations to the neural segments. In the spinal cord there is still a pair of nerves for each neuro542 Lock: Not. XI.

mere, but in the brain the primitive re1ations have been greatly modified or ob1iterated. There is not sufåcient evidence to show what might have been the primitive arrangement in that region. Of course, we bear in mind that the sensory fibres grow centripeta11y, but if they appertained to a particular segment, we might make a tentative estimate of numerica1 relations as follows : 1. First N euromere of F ore-brain, Olfactory II. second N euromere of Fore-brain, 0ptic. III. Third N euromere of F ore—brain, possibly nerve to Pineal Senseorgan ? IV. F irst N euromere of Mid-brain, 0culo-motor. V. second N euromere of Mid-brain, Trochlearis

It is easier to assign relations for the segments of the hind-brain. In my assignment I agree substantia1ly with Hoffmann.

VI. First Neuromere of Hindibraikn anterior root of Trigeminua VII. Main root of Trigeminus VIII. Third Neuromere of blind-main; no nerve root, at least in

ear1y stages.

IX. Facialis. X. Auditorzn The roots of the F acialis and the Auditory arise Separately in squalus acanthias. XI. G1osso—pharyngea.1. x11,i xIII, xIv. Roots of the vagus.

At the time of its first appearance the auditory saucer 1ies in front of the tenth neuromere, but is soon shifted backwards opposite the eleventh.

Minot has directed attention to the fact that Miss PIatt and McClure ignored the difference between ganglionic and medu1lary nerve-f1bres, and this is important ; but the ganglion ridges are divided in the same way as the neura1 ruhe, and we may spealc of the neuromeres as including the segments of the ganglion ridge.

The re1ation of the nerveckibres to the neuromeres will be considered more in detail under the heading of The

Nerves. No. 3.] THE IJEJETEEEH E« Haar-o. 543

8. Etwa? can! Frass-e.

It would be a great convenience to anatomists to have Some means of distinguishing between the head and trunk of very young embryos. It is genera11y regarded as impossib1e, on account of the laek of deHnite landmarks, to assign such a line of division, in early stages, before the origin of the auditory vesic1e. As Sedgwick says, in the artic1e referred to above, « The term head here must be regarded as meaning the anterior end of the body, for it is not possible in these young embryos t·o distinguish the head from the trunk." Nevertheless, in the young embryos of Squalus acanthias, there seems to be a natural line of division. The neural folds in this anima1 run forwards with the margins nearly parallel to one another, and then expand in front into the broad cephalic p1ate. The expansion takes place rather abrupt1y, and it is possib1e, in very young stages, to draw a line indicating where the expanded part of the cepha1ic pIate joins the nonexpanded part of the embryo. This line may be drawn without any reference to the number of primitive segments that it will cut off. The position of such a line is indicated by fix-i in Fig. 26. This is, in Squalus acanthias, just in front of the point where, subsequently, the vagus nerve begins. As before indicated, there are uniform1y eleven neural segments in front of this line. Their number remains the same from the ear1iest stages until the anatomical landmarks (auditory vesicle and nerves) appear that enable us to determine the limits of the head -region. In this animal, we may identify that part of the head that 1ies in front of the vagus nerve by counting the first e1even neural segments. It will be merely a question of agreeing upon the number of primitive segments be1onging to the vagus, to enable us to locate with defmiteness the hindermost limit of the head. Besides being of use in other ways, this would enable us to say, even in the earliest stages, what is headmesoblast and what is trunk-mesob1ast.

It is interesting, in this connection, to notice that there is

in Amblystoma a similar greatly expanded cephalic part with an assignable line of its union with the norkexpanded trunk 544 Lock: kvon x1.

region, and there are also in this form ten or eleven segments

included in the expanded head part. In the chick there are

eleven segments in front of the first formed protovertebrasx

9. F-«-2»e·-2»J-.

Some of the more important facts and conclusions regarding the neural Segment may be briefly stated:

I. They are natural structures, not artifacts produced by

· the reagents This has been shown by their constancy in ap pearance and general characteristics when different reagents are used; their consecutive history; their similarity in different kinds of embryos; and lastly, by their presence in fresh rnaterial before any reagents have been used.

2. Eos-Z)- xZPpea-2mce. —The so-called neuromeric segmentation is the first to appear. It arises long before there are any segmental divisions of the mesoderm, and therefore cannot depend upon segmental divisions of the middle germ-layer. Neuromeric segmentation is more primitive than mesodermic segmentation. i

Z. Fxmcz««se. —The cells in these segments are characteristica1ly arranged even in the earliest stages, and their arrangement and their structure would indicate that they are definite differentiations of cell areas, not merely mechanical undulations.

4. Exzesse-Z. —- The entire embryo is divided into similar segments, and passes through an arthromeric condition simi1ar to that of arthropods and worms. i ln Squalus the neural segments extend also into the germ-ring, and in the chiclc at times into the primitive streak.

s. JV2-»zåe-.·—-There are in Squa1us acanthias eleven segments in the brain region in front of the vagus nerve, and fourteen paired segments in the entire brain region, as follows: nine in the hind-brain, two in the mid-brain, and three in the fore-brain; this does not include the anterior tip, which may represent several consolidated segments.

6. Bcrokeeicxyzigyoeoxfk — There is Some evidence to show that

the spinal cord region is being encroached upon by baclcward No. 3.] TJJE Kaiser-Laien TE fix-ro. 545

differentiation Early in the history of these segments there are seven that clearly belong to the hind-brain, and later there are successively added two more.

J. Nase-», pas-Pfad« -——-These segrnents are clearest in the epiblastz the other layers are slightly afkected by the segmental influencez the mesob1ast least of all. The segments are serially homologous

8. XXVII-Eises. —They are directly (if not genetically) related to the cranial and spinal nerves. The neural ridges are divided in the same manner as the neural tube. The segments are also directly related to the Sense-Organs through non-es.

9. Twcxæsosweczxiows w- MociMc-2sz«2·o-2F. —The modifications are most extreme in the anterior part with the early obliterasp tion of those belonging to the foresi and mid—brains. Those in the hind—brain region are clearly deiined for some time after« the establishment of the cranial nerves, and then they fade away. i

This is about all that can be said about their transforma— tions, for the modiftcations have not yet been worked out in detaiL


The Sense-Organs of Vertebrates embrace all those structures that are endowed with Special sensibility They differ from one another mainly in degree of differentiation, and form a series, at the lower end of which are simple sensory papillæ, and at the upper end are complex organs like the eye and ear. We recognize two orders of Sense-Organs — simple generalized ones, the ganglionic Sense-Organs, and more specialized ones» belonging to the so—ca1led five senses.

The question of the origin and relationship of these sense-— Organs is full of interest. From the combined results of investigations on both Invertebrates and Vertebrates it seemsaltogether probable that the higher· Sense-Organs have been derived from those of a lower order, and, indeed, that they have all been differentiated from a common sensory basis, and are therefore related in a direct way. i 546 Lock Ijvon XI.

. This view has been entertained by morpho1ogists for upwards of ten years, and has a firm foothold in philosophical discussions. The evidence favoring such an interpretation has been accumulating, and although still incomplete, the hypothesis was never so well supported as at present.

Professor Whitmarks researches were among the first to bring out this conception. Ten years ago he demonstrated in 1eeches the genetic relation between eyes and tactile papil1ae, and since that time it has no longer been a matter of pure assumption to say that certain end-organs are modifications of a common sensory basis. He shows that the sensory papillæ located on each bodycking in the leeches have a very interesting relation. In the hinder part of the body the papillæ are purely tactile, but passing headward they become gradually modified, and are at first mixed visual and tactile, and tinally the anterior ten pairs are purely visual. We have thus had for upwards of ten years a welbauthenticated case of the derivation of Sense-Organs of a higher grade from those of a lower order.

In a more recent publication, «The Metamerism of Clep— sine,« 1892, Professor Whitman has added to the def1niteness of his earlier discoveries by observations on a new species, ask-r» Clepsine hol1ensis. He shows by a comprehensive study of the elements of the nervous System, the nerve distribution, and the sensillæ the complete homodynamy of all the segments.

Regarding the relationship of the eyes to other metameric

Sense-Organs, he says: «0ne more peculiarity of this species

may be noticed The median rows of sensillæ are quite well developed as eyes, at least on segments IV, V, and VI. The visual elements in Segment IV are quite numerous, and they are placed in a pigment cup quite like that of the principal eyes, only thinner and sma1ler, and directed backward instead

of forward. The sensi1la of Segment V is a little sma1ler, but

still presents the same general features. In Segment VI the organ is still eye-lilce, butiits visual elements are fewer andthe pigment cup imperfectly represented by loose pigment. In Segment VII we find one or two visual eells and a little scat . teredpigment In the following segments we usually End the No. 3.] THE FEETEZJEH TE HEXE. 547

sensillæ in about the. same condition. In no other species hitherto described do we find the sensillæ passing by such gradations into the principal eyes. IX«- Fæiriaz Æomozogy cj Zlzcxe wjgsmscx ZEJZZJZ Mk:- FJEF IF, HEXE, z: ja« okjrøoyzszkezxgci »« nie-s;åy ff« wyøåkjxologitczl xicexæ!oziwekx-skzz«, He« als» Zy «« Fxrzxcxzmczf gracixzziom egxlziåizezi i« Zlza rufe-Z! zwei-setz! « (pp. 39I, 392).

The existence of a series of rudimentary sense—organs in Vertebratos was brought to 1ight in 1885 by Froriep and Beard. While it is by no means clear that they are the homo1ogues of the sensory papillæ of anne1ids, it is nevertheless probab1e. It is now generally admitted that they constitute the basis from which the higher Sense-Organs of Vertebrates are derived. These organs are arranged in 1ongitudinal series. The series has become rudimentary or lost in the adult forms of higher existing Vertebrates, but they are present in the embryos Kupffer and others have shown that there are at least two series of these Organs, the upper one corresponding to the lateral line of comparative anatomy, and the 1ower one embrac— ing the so-ca11ed branchia1 Sense-Organs of Beard. From the upper series the ears and nose are probab1y derived, and possibly the eyes.

Allis (’88), in his well-known memoir on the lateral line of Amia, gives Hgures that show a connection between the epithelium of the nasal pit and that of the1ateral1ine. His Fig. I, P1. XXX, representing a larval form one day old, shows the nasal epithelium forming part of the lateral line. In subsequent growth it follows the same course that the surface organs of the lateral System do when they become canal Organs. Allis says, p. 537: «The nasal pits are inclosed in the same way

that the lateral cana1s are, and the short canal so formed is at

first continuous with the canal inclosing organ 5 infra-orbita1.«

There is now genera1 agreement that the ears be1ong to the lateral line series, but there has been much dissent on the part of Some morphologists to admitting the eyes to the same category; but the grounds for the Opposition seem to me to be largely removed. It must be said, moreover, that the Organs of taste and touch have not as yet been shown to have any genetic relation to the ganglionic Sense-Organs. 548 Lock: kvon XI.

I am g1ad to acknowledge my indebtedness to Minot’s intro— ductory remarks to his chapter on the Sense-Organs. He says: «Very suggestive in this connection are the observations of H. V. Wilson («91), of a thickening of the nervous layer of the epiderrnis on either side of the head in the bass embryo (Serranus atrarius). This thickening forms a long, shallow furrow,

which subsequently divides into three parts, of which the first

becomes a Sense-Organ over the gil1-cleft, the second, the auditory invagination, and the third, the Anlage of the sense-organs of the lateral line. This peculiar development confirms the notion that all these organs belong in one series, but the appearance of a continuous thickening as the Anlage of them all has, as yet, been observed only in this fish, and may not indicate a corresponding ancestral condition Unfortunately, Wilson was unable to make out anything as to the connection of the sensory plate with the ganglia. The sense-organ above the gi1lcleft, although differentiated, is a larval structure only, and disappears in the adult.«

A quite similar condition is now known to obtain in some elasmobranch forms. Mitrophanom in 1 89o, published a preliminary report of his observations on the lateral line of Acanthias and other Elasmobranchs In 1893 he published a full report of the same, illustrated by many iigures He describes a continw ous thickening of the epidermis along the sides of the head embracing the territory of the roots of the seventh to tenth nerves. From this thiclcening there is separated the material for the auditory saucer, the branchia1 sense-organs, and the beginning of the 1ateral line. My observations on this region in Squalus agree with those of Mitrophanow

In looking over the literature on the sense-organs one cannot fai1 to be struck by the remarkable uniformity of the sensescells in the different kinds of sense-organs. , They are easily reducible to one type, and this, of course, favors the view that they

have been derived from a common form. NO. 3.] THE FEÆTEBÆÄ TE III-TO. 549


Squalus acanthias is an especially favorab1e form for observing the beginnings of the optic vesic1es. They make their appearance in this animal at a very early stage, while the neura1 p1ate is broadly expanded, and before the neural folds have arched upwards in any part of the embryo. I should estimate their appearance in this animal to be as early as in any other anima1 in which they have been described. Previous to I889 it had been current opinion that the optic vesic1es appeared very early only in the class of Mammala Bischoff, Kö11iker, His, Van Beneden, Heape, and others had recorded their early appearance in that group ; but it was regarded as a precocious deve1opment for some reason confined to mamma1ian forms.

More careful observations on the earlier stages have shown that the optic vesic1es customarily arise in different anima1s much earlier than was supposed. In I889 Whitman called attention to the very early appearance of the optic vesicles in Necturus: « Its basis being discernib1e as a circular area—-— after treatment with osmic acid, followed by MerkePs fluid —- long before the closure of the neura1 folds of the brain." Dura1noted the early appearance in birds. In 1893 Eyc1eshymer gave notes upon their appearance and structure in Necturus and other Amphibia His most noteworthy observations are upon Rana pa1ustris, a form hitherto unstudied, in which he ftnds a remarkably early deve1opment of the primary optic vesicIes. They are very evident from surface study in this form, on account of the distribution of pigment granules in them, just after the neural ridges have begun to form. cross-sections- in these early stages show a considerable amount of histological differentiation from the surrounding ce11s.

The early differentiation of the optic vesicles had not been recorded in any e1asmobranch form unti1 the publication of my preliminary papers in 1893 and I894, where the optic vesic1es are described and their serial relation to other structures on the cepha1ic plate is show-vix. Since then I have had opportunity to coniirm my observations on Squa1us, and to extend them to some other forms ; for instance, in Diemycty1us and 550 Lock kvon x1.

Amblystoma the optic vesicles are well deve1oped in very early stages In Torpedo oce1lata they show fairly well, and in the chick they may be readily made out while the groove is wide1y open.

In Squalus acanthias the optic vesicles are the first ruchments of Sense-Organs to appear. counting out the gastricular cavity (and the neural segments) they are the Hrst organs of any kind to be established. Strangely enough, their early history in these animals has been entirely overlooked. Balfour’s statement that « The eye does not present in its early development any very especial features of interest," seems to have withdrawn attention from that organ in the elasmobranch Hshes.

In order to iix clearly the time when the optic vesicles first appear in Squa1us, we must glance over the early steps in the formation of the head-plate. Pl. XXVL Figs. I, 2, Z, represent three stages that immediately precede the formation of the eyes. In Fig. I the embryo is well established; it is a stage slightly older than Balfoufs stage B. The front part of the embryo is not broader than the rest, and the curve of the anterior margin is uniform and unbrokerr In Fig. 2 we note two changes, vie» that the front end is broader than the part just behind it, which is apparently constricted, and the anterior margin is drawn out into a rounded median cusp. The headend continues to expand laterally (Figs. 3 andr4) until we may (with some appropriateness) speak of it distinctively as a headplate. In the meantime the trunk region has grown longer, and is relatively narrower, so that the entire axial embryo has a Characteristik: appearance of a narrow body terminated by a rounded plate-like head. When this stage has been reached the first satisfactory traces of the eyes become visible ; when first formed they present the appearance that might be produced by pressing down lightly upon a plastic surface with two rounded dies. They are circular areas slightly concave upwards, and occupy a position far forwards on the cephalic p1ate. The early stage of these vesicles may easily escape observation, as they become evident only when the light strikes them properly, and when they are in a favorable position for the observer. No. 3.] III-I: PEJZTEFJEH TE Eis-ro. 551

Fig. 5 is engraved from a photograph in swhich the optic vesicles show very well. The specimen from which the photograph was made showed about six mesoblastic somites, and was ZHF mm. in length The circu1ar areas may be made out satisfactorily in slightly earlier stages, in which only three mesoblastic somites are differentiated, and in which the neural fo1ds of both head and trunk are not only broadly expanded, but are even ventrally —curved. The circular depressions are at first separated from one another by a raised welt. This has already been spoken of in Part I as atonguedike process extending from the mediananterior tip bacliwards to two—thirds the length of the cephalic plate. It continues to be a prominent feature of the cephalic plate for some time. I can offer no suggestion as to its signifis cance, outside of the obvious suspicion that it may represent a proboscis of Some lcind, or that it may be related to the 1arge notochord of this region. On this point 1 have no conjectures to make. The depression to form the infundibulum starts a little after the first appearance of the optic vesicles, and cuts the median process in two, so that the anterior tip is separated from the rest, and the depression for optic vesicles and infundibulum combined reaches across the median plane of the cephalic plate (Figs. 6, J, 8, 9, ers-J. The optic vesicles, however, do not reach to the 1ateral margins of the neural folds ; the latter are much expanded beyond them.

It is evident that we cannot, in a strict Sense, speak of such vesicles as «diverticula of the fore-brain.« The «diverticula" are found before the fore-brain. The depressions once started grow deeper and press outward laterally ; when fully formed they are cupped, almost rounded in out1ine, concave from within, and they form rounded elevations where they come in contact with the outer layer of epiblast. Figs. H, J, 8, 9, Io, show very well the appearance presented by these primary optic vesicles when viewed from above. Fig. 7 is magnified higher than the others, and is, therefore, larger, but the others are all uniformly en1arged. These figures all show the interior of the optic pits. They represent a range of stages only slightly older than that·shown in Fig. s. In all of them the depression in which the optic vesicle lies extended across the median 552 Lock Ijvon x1.

p1ane, and in its median portion, the depression develops into the infundibu1um.

The1atera1depressions, which form the optic vesicle, sink

p downwards more rapidly than the median part of the depression,

and as a resu1t there is left a ridge or keel separating the optic vesicle and connecting the anteriopmedian tip of the headp1ate with the Central tonguelike process. Miss P1att, in her studies on the head of this animaL says: « It may be here

cis-r 8. —-Eight transverse Sections of the embryo shown in Fig. s, Pl. XXVL d( about zo diameters The nnmbers below the Sections refer to their Position in the set-les.

noticed that the downward growth in the Hoor of the brain,y which gives rise to the infundibulum, is prirnitively bilateraL and not median," and I think, therefore, that she saw these circular optic pits in an early stage, but interpreted them as 1atera1 pockets of the infundibu1um. There might be reason to doubt whether they are the optic vesicles, if they could not be traced with perfect continuity into the eyes, and if there were not evidence of histological differentiation in their areas.

By the time the neural folds of the head begin to rise the

optic vesicles are very prominent, and may be observed from the outside and from within. Fig. I 3 shows an embryo viewed No. 3.] THE »Er-E TEBJM TE JJEHJZ 5 5 3

obliquely from the 1eft side, and the optic vesicle of that side is seen from the external surface, where it presents the appearance of a rounded eminence 0n the opposite side the optic vesicle is seen from within. cut 9 shows transverse Sections of the head of the same embryo. Figs. 16 and 17 show embryos in which the optic vesicle Ho. A) shows from without, and also from within («:)P.i-). 1n Figs. I9 and 22 the optic vesicle shows clearly from the outside. Transverse Sections of the

CUT 9.-— Eight transverse Sections of the embryo shown in Pl. XXVL Fig. is. I( about so characters. The accessory optic vesicles H. Ha) have been formed.

embryo photographed in Fig. 23, after partial closure of the neural groove, are illustrated in cut I0.

These optic vesicles are so well developed at an early period in Squa1us, that it seemed to me probab1e that similar early formed vesicles might exist in Torpedo oce11ata, and have been overlooked by the Zieglers, who have studied the early stages of that form in 1892; accordingly I procured embryos of Torpedo oce1lata from the Zoölogical Station at Naples, and studied the head region with some care. The first thing noted was that the embryos of Torpedo ocellata aire not nearly as favorable objects for observations as those of SquaIus. They are smaller and the beginnings of Sense-Organs, branchiæ, and other structures about the head are by no means so clear as 5 5 4 LOC P. [Vo1«. XI.

they are in Squa1us. As far as I am ab1e to discern, there is no very distinct differentiation of the optic vesicles, in the nearly stages of this form; nevertheless, they are present at the stage designated C, by the Zieglers, and perhaps a little earlien In studying the actual specimens, I had occasion to note that

cui« 1o.—'I’we1ve transverse Sections of the embryo shown in Pl. XXVL Fig. 23. x about zo diameters The number-s below the Sections reker to their positions in the series. In Section 14 the optic vesicles show below and the depressions for the mid-brain vesicles above.

the Zieglers’ models are admirable representations of the embryos of the form they are intended to represent The optic vesicles are much fainter than they are in Squa1us, but net-ertheless are simi1ar to them. I have studied them in Torpedo both in surface views and in Sections. One noteworthy fact is that they have been sectioned and actua11y Hgured by Ziegler in his Pl. IV, Fig. 19I. It will be noted in this iigure, that the brain-wa11s bulge outwards on either side and these depressions No. 3.] » THE IJEEIEEJZH IE Iris-IV. 555

mark the interior of the optic vesicles P1. XXIX, Fig. 65, is a drawing of one of my Sections of Torpedo s1ight1y younger than the one of Zieg1er’s just referred to. The optic vesicles are indicated at pp.

The external appearance of the optic vesicles in later stages is shown in Pl. XXVII, Figs 34, 35, 36, 37. When the stage representecl in Fig. 35 has been reached the lens shows externally ; it is forrned in the usual vertebrate way. The choroid fissure shows in Figs. 36 and 37. My studies have not been made to include the later stages of development of the optic vesic1e, but are coniined to its earliest differentiation

Sections of the earliestckormed circular areas, show something in the direction of histo1ogical clifferentiation Mitoses are more frequent and more of the cells are e1ongated and pear-shaped than in the other parts of the cepha1ic plate. The differentiation is by no means as marked as in the Rana pa1ustris Hgured by Eycleshymer —- nevertheless indications of change are not wholly lacking My Sections are too thick (I0«- to I In) for satisfactory histological study, but one can see in them that the middle of the wa11s of the optic cups are areas of differentiation The differentiation does not progress far until later, but the frequency of dividing cells, and their change of form continue, in these early stages, to be marks for distinguishing visual epithelium from that of the surrounding brain—wal1s. The dividing cells are neurob1asts, and these are known to be points from which thenervedibres spring. Their presence in any considerable number would, therefore, indicate a differentiation in the direction of increase of sensibi1ity.

The eyes are the highest developed of the Sense-Organs, and in that particular stand at the head of the series. But, do the eyes belong to the same series with the other Sense-Organs, or do they occupy a position by themse1ves, can they be homo1ogized with the restk This is a puzzling question, over which there has been much controversy, and upon which there is still much difference of opinion among anatomists The stock objection to their being classed with the other Sense-Organs, is this:—— they have apparently been derived from a different basis, and their nerves are developed in a different way. The 556 Lock. Not. x1.

eyes are formed out of neural-plate material as diverticula of the fore-brain, while the other Sense-Organs arise outside the neural plate ; they have an independent epiblast origin.

It is questionable, whether any particular stress should be put upon that argument, as the neural plate is at best only a part of the modified epiblast ; moreover, the suggestion that the neural plate is, undoubtedly, very much widened and expanded from its ancestral condition, and that certain sensory area, originally lying outside, may have been included in it, does much to 0ffset that objection. The neural plate has been a prodigiously long time in forming, and the eyes have been brought into closer relationship with it. The plate is broadest in front, and the eyes are the most anterior sense-organs, and have become included in this expansion While there is not sufficient data to give a wholly satisfactory answer to the question propoundech the assumption that the eyes are closely related to all the others is not without foundation, and we are now in the attitude of awaiting further facts.

II. AccEssoRv Oprxc VEs1cLEs.

The neura.1 plate, while still in very young stages, and while the neural grooves are widely open, becomes the seat of some accessory differentiations that resemble the optic vesicles. So far as I know no observations upon these structures have been recorded except those in my preliminary account in the JOURNAL OF« MORPHOLOGL I quote from that paper: But, more interesting than the fact of their (the optic vesicles’) very early appearance in Elasmobranchs, is their apparent relation— ship to other depressions that are formed upon the cephalic plate, behind the already established optic vesicles. The new involutions referred to, make their appearance upon the cephalic plate just baclc of the optic vesic1e. Two of them (Fig. 13, as. v) and Fig. U, as. Eil, ask. IN) take precedence of all others in deve1opment, and are so distinctly formed as to afford a good basis for cornparison with the optic vesicles. They are circular depressions formed in front of the latter, and they produce upon the exterior corresponding rounded e1evations. No. 3.J THE »Es« Paar« JIE HEXE. 5 5 7

The optic vesicles are formed first, and when, at a very litt1e later stage, the others arise behind them, it appears as if the process of eye-formation were repeating itself serial1y.

These circular pockets not only arise in a simi1ar way, but structura1ly resemb1e the optic vesic1es. In cross and 1ongitudina1 Sections the ce11s of the sunken patches are similar to those in the eye pockets. They may be designated access-»Jozjszz«c TIZFZ«CZFF.

The assumption that they are primitively visual in character may be going too far, but the basis upon which it rests will be shown directly. The 1east that can be said of them (from their structure, the place and manner of their appearance), I think, is that they are segmental sensory patches.

These structures begin to appear soon after the eye vesic1es. There are several pairs of them stretching in rows baclc of the eyes, as stated above, the two anterior pair are the most prom— inent. They gradually grow fainterz the anterior one is the best differentiated; the second one is not so clear, and the succeeding ones grow fainter as we pass backwards I have counted as many as four pairs in surface studyz but Sections show that the series is more extensive and contains not less than eight pairs of these circular pits. The two anterior ones are well shown in Figs. 9 and ro. These are the earliest surface indications. In Fig. V, which is slightly more advanced, the accessory structures are clearly developedz the ftgure shows four pairs behind the eyes. In longitudinal Sections of the same specimen, Pl. XXIX, Figs. y6——87, eight pairs may be counted. The form of the depression, as seen at first from the surface view, is ovate; they soon become circular patches, and form themselves into concave, shallow cups. I have seen them in many specimens.

It will be noted that they make their Erst appearance while the neural plate is broadly expanded, and spring into prominence while the neural folds are growing upwards It is during the rise of the neural folds that similar, but fainteky vesicles can be made out, in surface study, behind the two anterior pairs.

Pl. XXX, Figs. 88 to 112 show twentyckive transverse sections of an embryo very slightly older than that represented in 5 58 LOCPI IJVo1.. XI,

Pl. XXVL Fig. 16. The Sections 88 to 100 lie in the region of the cepha1ic plate. The following Sections, 105 to 1I2, lie in the neck and trunk region. These Sections are remarkable in bringing to light serial depressions a1ong the walls of the neural folds. They show that the serial cup-like differentia— tions extend back of the cephalic plate. I have not been able to determine the number of serial differentiations of this character in the embryo, but it is clear that there are several pairs behind the cephalic plate upon which I have noted, in surface study, four pairs in addition to the primary optic vesicles.

My Sections are not favorable for a critical study of the histological conditions, but it is clear, from them, that« a differentiation starts in the anterior patches simi1ar to that mentioned above, for the true eye vesicles. There is a greater frequency of dividing cells and many of the cells become elongated and pear—shaped.

comparisons of the two anterior pairs with the eye vesicles give the following points of resemblance They are formed in precisely the same way, and present the same general appearancez viewed from within they are all similar concave cups; they produce corresponding elevations on the outer surface, and, if observed from the« outside, they form a series of rounded knobs serially arranged, the eye being the anterior terminations of the row. The marked differentiation in the eye takes place in later stages. There are, however, the histological features spoken of above that serve to distinguish it in early stages, and the accessory vesicle shows the same characteristics

These resemblances, coupled with the existence of serial eyes in Some of the Invertebrates, has led to the assumption that in Vertebrates these rudirnents represent accessory optic vesicles. The fate of the anterior pair would also favor this

view ; they pass into the pineal sense—organ, a structure whose visual character is acknowledged

If the view expressed above is true, we have a multiple-eyed condition in the embryos of these animals. This is common

enough in Invertebrates, but has not been previously noticed in Vertebrates. No. 3.] THE« IJEXTEEKH TE III-Hm. 5 59

These accessory structures are present for a brief time only. That region of the head, behind the optic vesicles and in front of the ears, becomes the seat of great modifications, and, as the neural groove closes, the structures described give way to later formed ones. They are, therefore, not only embryonic structures, but are also transitory, It is a truism in develop— mental history that the more primitive characteristics appear first, and the secondary modiiications come in later. The structures described are among the most primitive that have been preserved in this very ancient group of Vertebrates, and should be of signiiicance in indicating the ancestral relations of the eye. I have spoken of the disappearance of the Organs, but I have been able to trace the anterior pair further along, an account of which will be given in the next Section on The Pineal Sense-Organs.

I have also traced out simi1ar rudimentary organs in the embryo chick, at about the conventional 24-hour stage ; there is a series of seven vesicles behind the optic vesicles These structures are rudimentary, and very quickly fade away.

All this acquires new interest when taken in connectiion with the researches of Whitman on the eyes of leeches. As indicated above, he showed that the eyes of the leeches are derived from segmental sensory papillæ. It now appears that we have traces of a somewhat simi1ar line of segmental sensory organs in Vertebrates It is essential to say Drecke-F, for these structures are transitory, and do, not rise to a high grade of differentiation, except in the case of the pineal Sense-Organs. They may never, in the vertebrate group, have been raised to the rank of eyes, but it is certain that in Annelids simi1ar segmental sensory patches have been so raised. It is, however, reasonable to suppose that members of this series have been functionaL even in Vertebrates, when we recollect the highly deve1oped condition of the pineal eye in Some forms.

There is really very little direct evidence to support the proposition that the Vertebrates are derived from multiple—eyed ancestors, although it is aconception that has been in the minds of morphologists for Some time. Besides the indications of such a condition afforded by the facts of this paper there 560 LOCPT [Vo1.. XI.

has recently come more evidence bearing in the same direction. This consists in the discovery of multip1e pineal eyes, in several distinct forms, and, in one case, the existence of three distinct nerves to the pineal organs (see Section on The Pineal Sense-Organs)

Whitman, in his paper just cited, says: ssAlthough the evidence appears to me conclusive that the eyes and the segmenta1 papillæ were, originally, morpho1ogical as well as physiological equivalents, it does not, of course, follow necessarily that both Organs now have the same fnnctional signifk cance. The original papillæ may have represented Sense-Organs of a more or less indifferent order, among which, inthe course of the historical development of the leech, a division of Iabor was introduced, a few at the anterior end becoming specialized as lighbperceiving Organs, the rest either remaining in their early indifferent condition or becoming specialized in Some other direction.

«The discovery that these papillæ are Sense-Organs might lead us to speculate on aHinities of a distant and somewhat uncertain nature, such as are supposed by the writerz in common with many others, to exist between annelid worms and Vertebrates At all events, the existence of such organs in the leech furnishes a broader basis for the discussion of the question whether the Vertebrates and Annelids have been derived from a common form possessing metameric Senseorgans, as was first argued by Dr. Eisig of the Naples station. Assuming that the Sense-Organs of the Vertebrate and the segmental papillæ of the leech may be traced to a common origin in some remote ancestral form, it does not follow that they should now present close structural resernblances It is far more important to show that they possess certain general features in common. The most important of their common features is undoubtedly their metameric origin. The nervesupply forms another feature of fundamental importance, in which, according to the interesting observations of Mr. Beard, on the segmental Sense-Organs of the lateral line (Z·:)x)Z. des» VIL Nos. 161 and I62) of the Vertebrate there is essential agreement. The developmenta1 history of these No. 3.] THE »Der-Tagen TE Erst-ro. 561

latera1 organs in the Hsh, where they make their first appearance as FFFJJHFJZZZZZPZZPJJZZE in the strictest sense of these words, cannot be explained on a more satisfactory hypothesis."


I. Gram-J- oj lfisssozeyZecige ragmwssbøg IX«- Piøeaaz Same-Organs.

The pineal outgrowth has attracted much attention since the discovery, in I886, of its eye-like structure. Since then it has been accepted by morphologists as a rudimentary Sense-organ. The earlier 1iterature bearing on the subject has been theroughly reported on by Spencer («86) and Francotte («89). several recent1y published papers taken together put the subject in a new light, and, as Ritter («94) has said, «At no time since the epiphysis and pineal eye have been the topic of

investigation has it been a more interesting or a more inviting

one than it is just now." The growth of our knowledge regarding this remarkable

Sense-organ, scattered through the various publications of

Spencer, Beraneclg Francotte, Strahl and Martin, Leydig, and others, may be reduced to the following summary : i First came the demonstration of its eye-lil(e structure in Amphibia, by De Graaf, and in Lacertilia by Spencer. De Graaf’s publication is much briefer, and was published only a few months before Spencer’s extensive study. Both authors investigated the structure in adult forms. Spencer studied its condition in twenty-eight different species of Lacertilia, and demonstrated very clearly that even in the

adult forms of these animals, this organ has an eye-like Struc ture, with a 1ens, a pigmented retinal layer, and a nerve. It seems doubtful, in the light of recent work, whether the kibrous Strand connecting the eye-capsu1e with the epiphysial Stall-i, and designated the nerve by Spencer, is really nerve or not.

Following the study of the adult structure came investigations of the embryonic conditions The first studies dealingv with the embryonic development of this organ, and with its. 5 62 LOCK [Voi:.. XI.

minute structure in the early stages, were those of Strahl and Martin. Along this same line were the studies of Båraneclg Francotte, Leydig, and others. By these investigators the organ was traced backwards to a certain point in its history, that is, to the time when it springs from the roof of. the thalamencephalon as a tubular outgrowth, like the tip of a finger of a glove. To this period also belongs the discovery of a nerve of transitory existence The investigations were coniined almost entire1y to two groups of animals, Amphibia and Reptilia. In the embryos of these animals many points of minute structure were worked out, such as the distribution of the nerve within the eye—like capsule, the division of the retinal part into distinct layers, the fact that the pineal eye is higher deve1oped in the embryonic periods than later, etc.

The next step was the discovery that there is more than one epiphysial outgrowth. Se1enka, Leydig, Eycleshymen Hill, successive1y ca11ed attention to the existence of two distinct outgrowths, or vesicles, from the roof of the thalamencephalon in Anguis, Amblystoma, and Teleosts.

Hi11’s studies on these two vesicles are the most complete He has shown the existence in Teleosts and Amia of two vesicles, anterior and posterior, and has traced them through the stages of deve1opment. The posterior develops into the epiphysis and the anterior degenerates. The dista1 portion of the former is histologically very complex, and receives a nerve from the posterior commissure. Hill made out the distribution of nervedibres in this part of the epiphysis. He traced quite clearly the history of the vesic1es.

Two vesic1es, an upper and a lower, have long been known to exist in Petromyzon, in both embryos and adu1ts. It is much to be regretted that we do not know their embryonic history, for much depends on this. The most recent paper on the epiphysial Organs in Petromyzon—that of studniöka—— does not clear up the question of the origin of the two vesic1es. Histological material was 1acking in just the stage required. The two outgrowths in embryos outside the Cyclostomes have been designated epiphysis, for the hinder, and pineal eye for the anterior. Studniöka designates them pineal and parapineaL No. 3.] THE NEJZTEEJEA If: fix-w. 363

respectively, in the Cyclostomes As I shall show latet, the so-ca11ed paraphysis is a different Structur-e.

Beraneck showed that the pineal eye and epiphysis have been confused, and, on account of the existence of the nerve from outside sources, he argued for the complete independence of the pineal eye.

Klinckowstrom has p1aced himse1f in Opposition to this opinion, showing the eye vesicle is formed as an outgrowth from the epiphysis.

Next to the discovery of two or more epiphysial outgrowths may be mentioned the existence of accessory pineal eyes The ear1iest publication containing an account of such Structur-es is that of Duva1 and Kalt, in I889. In a brief note (Des Yeux Pinåaux Multiples chez 1’Orvet) these authors record the fjnding of two or three accessory pineal vesic1es in Anguis

They do not say whether adu1t or embryonic forms were


carriere («89), in the same year, noted the occurrence in embryos of Anguis of a very rudimentary accessory pineal vesic1e.

Leydig («90) observed these structures independently in embryos of the same anirnals He showed that there is much variabi1ity, both as to the presence and the histological Structure of these accessory capsu1es. In Some individuals they are lacking, and in others of the same age present and well deve1oped. He records the occurrence, in Some cases, of two accessory Organs, a. larger and a smaller one. The larger one resembles the pineal eye in structure and arrangement of pigment. The smaller one is very rudimentary Leydig com— pared them to the ocelli of insects. It is c1ear from his account that variabi1ity and inconstancy are characteristics of these accessory Organs.

Prenanh in the latter part of I893, again records the occurrence of one accessory pineal eye in the embry0s of Anguis He directs attention to the fact that up to that time they had been discovered only in a single species, VIII» Anguis fragilis, and further, that they were all in ernbryos, and not in adults. Ritter (-94) has recently recorded the, occurrence of such an» accessory org-an in the aclult Phrynosoma 564 Lock: kvon XI.

The above observations, taken together, give suflicient positive data to establish the fact that there appears, from time to time, in some animals the vestiges of accessory pineal Organs, and that they are variab1e and inconstant in their occurrence.

The most recent advances consist in carrying the history of these organs baclcwards, and showing them to be connected with patches of sensory epithelium that arise on the cephalic plate in very young stages, and in the discovery, in Iguana, of three distinct nerves connected with the epiphysial outgrowths. The former work was done by the present writer, and the latter by Klinckowstrom.

As indicated in the preceding Section, there are several pairs of cups on the cephalic plate back of the optic vesicle, formed while the neural groove is widely open. I have designated them accessory optic vesicles In the process of closure of the neural groove the anterior pair are brought together, and form part of the thalamencephalon, from the roof of which the pinea1 outgrowth is derived. This process will be described more in detai1.

Klinckowström (-98) has recent1y shown in Iguana the presence of two nerves connected with the anterior outgrowth or pineal eye, and sometimes a third connected with the posterior soutgrowth or epiphysis.

These observations suggest new interpretationsz still it is not an auspicious moment to indulge in speculation. It is clearly evident that we need more data regarding these organs and their relationships But the trend of these discoveries is to strengthen the suggestion already made in this paper, that, primitively, the Vertebrates were multiple-eyed. The presence of accessory pineal eyes, the discovery of serial sensory areas on the cephalic plate from which epiphysial outgrowths arise, and the presence of two or three pinea1 nerves, are all consistent with this interpretation so far as the evidence goes, there is more than one epiphysial outgrowth, and therefore I have headed this Section, The Pineal Sense-Organs. In what

way it will be necessary to qualify this proposition will depend on subsequent discoveries. NO. 3.] THE« IJEJBTEBJEÄTE HEXE. 565

The results of all the embryo1ogica1 studies on the epiphysia1 outgrowth between 1887 and 1893 were to settle on one point of common agreement regarding its origin. Each investigator successiveiy found it to arise as a tubular outgrowth from the roof of the tha1amencepha1on comparative1y late in ontogeny, after the parts of the brain are estab1ished. No earlier trace of it had been found; but there was a previous undiscovered history, and it is now time1y to make the inquiry, What is the very beginning of the pinea1 Sense-Organs?

Z. Las-Hofe Fragt-«» czf Mk· Pisa-se! O«z«g-o2e-x-Tz.

As already indicated, there are upon the cephalic plate accessory eye vesicles which have made their appearance while the neural groove is widely open; and in tracing their fate I shall attempt to f111 up that gap in the history of the pineal outgrowth from the time these cup—1ike structures appear upon the cepha1ic p1ate to the time the tubu1ar outgrowth begins from the tha1amencephalon.

As the wa11s of the neural groove grow upwards these Cuplike structures are carried up vvith them; and there comes a time (Figs. I3, 3I) when they may be seen from the outside

as rounded eminences, and from the inside as concave cups.

When the edges of the neural groove meet in the middle line the cups are approximatech and come to form part of the tha1amencepha1on.

While these changes have been going on, further differentis

.szations have been taking place in the wa11s of the brain. The

bulging of the wa11s to form the mid-brain vesicle has come on insidious1y, and has taken a position behind the vesicles of the paired eyes, in apparent1y the same position previously

ssoccupied by the accessory vesicles. These transformations are

confusing, as the wa11s of the mid-brain resemble the accessory optic vesicles grown largerx 1n an ear1ier published paper I made the mistake of identifying the mid-brain with the accessory optic vesicles; but it was merely an error of identification, and did not affect the main contentionof that artie1e.

In working out the details of the formation of pineal outgrowth in Squa1us, it soon becomes apparent that the surface 566 Lock. kvon XI.

contours of the head are considerably altered by the distribution of mesoblastic Hcells lying between the external layer of epiblast and the brain-waIIs. The mesoblast forms a pad of varying thickness in close contact with the brain—wa1ls; the depressions are tilled, and in some places thick patches of mesoblast give rise to external rnarkings that render the surface appearances untrustworthy. There is, for instance, a deep pad of mesoblast tilling up a depression at the sides of the cerebellum, and also extending forwards on to the midbrain; at the external surface the pad assumes a circular form, and it is that pad of mesoblast which is seen in Fig. 32. The true mid-brain vesicle in that Hgure is the eminence marked wo. The accessory optic vesicles are present, but are not well marked externally. They are in front of the protuberance marked »O.

In order to get a view of the brain-walls I have found it necessary to remove the mesoblast and its coverings of epib1ast, and thus to complete1y expose the brain-wal1s. The brain thus laid bare, enables one to see with complete satisfac— tion its different parts and their relation to one another. A very interesting relation comes to light through the dissections. Somewhere between the stage with an open neural groove (Fig. 3I) and the completely closed groove (Fig. 33) the midbrain vesicle insidious1y takes the former position of the accessory optic vesicle while the latter is carried forward by the pr0cess of cranial flexure During the formation of the midbrain vesicle the two structures become incorporated into one faintly bi1obed protuberance (PI. XXVIIL Figs. 38 and 39). The anterior part is the accessory optic vesicle, and the posterior one the mid—brain. The separation soon becomes com— p1ete, and by the stage represented in Fig. 40 the two are completely separatedz but owing to the arrangernent of the mesoderm the accessory optic vesicle is rendered indistinguishable from the outside. When the mesoderm is entirely removed it may be seen. All this takes place in very brief intervals of time, and a complete series of embryos representing the different phases of the closure of the neural groove is required to see it all. I must say also that the changes are so perplexing No. z] THE YEETEEJZH THE: III-m. 567

that there exists in my mind Some doubt as to the adequacy of the above account. 1t may have to be modified, but I have given the facts as they now appear to me.

F igs. 38-54 show a series of embryos that have been partly dissected in the way just indicated They show characteristic changes in the brain-wa1ls as seen from the exteriotr

In Fig. 38 the vvalls of the neural groove have not yet met in any part of their course, and the thalamencephalon cannot be distinguished from the rest of the brain. The first pair of accessory vesic1es are visible; they soon become inc0rporated in the walls of the tha1amencephalon. I have not been able to determine whether it is the epithe1ium of the first pair of accessory vesicles only, or whether epithe1ium from the second pair is also included. Either the epithe1ium of the two pairs is incorporated into the walls of the tha1amencepha1on by being carried together, or the epithe1ium of the second pair fades into the surrounding substance of the brain-wa11, which almost immediately grows into the vesicle of the mid-brain. Although I am doubtful as to whether the second pair of accessory vesicles pass into the inter——brain, I am sure that the epithe1ium of the anterior pair is so included.

Fig. 39 shows the first accessory optic vesicle and the midbrain vesicle forming a common protuberance, but they are very quiclcly separated, and the anterior vesicle goes into the tha1amencepha1on. e

In Fig. 4o, which is somewhat o1der, this has occurred The lateral wa11s of the thalamencephalon now consist of two shallow cups, approximated so that the structure looks lenticular when viewed from above. As a usual thing, the thalamencephalon is not evident at this stage before the removal of the

overlying tissues. Fig. 4I shows, however, a specimen in

which it could be seen before any dissection. Compare this with Fig. 32, which is the more usual appearance of an embryo

of this age.

Fig. 42 is a dissection of the embryo shown in Fig. 41. The mid-brain, which, from external view, appears like a single rounded eminence, is shown after exposure to be

"bilobed. 568 Lock. Not. x1.

Figs. 43, 44, and 45, which are successively o1der, show an

increase in the size of the thalamencephalon, and marked

changes in the mid-brain.

In Figs. 46 and 47 a faint furrow has appeared in front that serves to mark the boundary between the thalamencephalon proper and the cerebral lobes. This furrow is clearly deiined in Figs. 48 and 49, so that we have now the thalamencepha1on distinctly marked off from the rest of the fore-brain.

In Fig. 48 the optic vesicle and all the mesob1ast have been removed, and the view is talcen from the right side. The brain— walls show very clearly. The tha1amencephalon is now bounded anterior1y and posterior1y by furrows. The two furrows run nearly parallel to one another, and they serve to mark oft« the inter-brain very distinct1y.

The roof of the thalamencephalon is at this stage raised into two rounded eminences of nearly equal dimensions, an anterior and a posterior one. The posterior is, from the beginning, somewhat in advance of the other. It becomes the pineal out— growth, while the posterior eminence goes on deve1oping. The anterior one becomes greatly reduced by compression from the rapidly growing adjacent brain-walls. It very soon loses its rounded Character, and becomes pressed into a semicircular fo1d lying in front of the epiphysis. It is, I thinl(, equivalent to the Zirbelpolster of German author-s. If longitudinal sections be made at the stage represented in Fig. 48, the two e1evations show as in Minot’s Fig. 3I9, p. 5y2, of his Hasen« EmZ--;1-·c)!og·j-. In that Hgure they look like two equal vesicles springing from a single elevation of the brain roof, and«opening by a common wide passage into the brain vesicle No especial signif1cance, I think, is to be attached to the fact that

there are at first two equal embossrnents from the roof of the

tha1amencephalon. The anterior one is converted into a fold of the roof of the ’tween—brain that has long been recognized in different anima1s. It is designated Zirbelpolster by Burckhardt It does not in any sense correspond to the anterior epiphysial vesicle discovered by Hill in Teleosts, but rather to the fold of ’tween-brain that lies in front of both epiphyses. Studniölca

has, however, marked the fold in diagramrnatic figures of Tele—No. 3.] THE PEETEBJZA TE XII-IV. 569

osts, the anterior vesicle of Hi1l, but this identification is not right.

We have in this specimen the first appearance of the tubular outgrowth, which, accorciing to previous authors, marks the beginning of the epiphysia It is somewhat important to fix the stage at which this outgrowth begins. Its very first appearance as an elevation of the roof of the thalamencephalon is in the stage represented in Fig. 36 after the saucersstage of the ear is past.

In Fig. 49 the posterior protuberance of the roof of the thalamencephalon is assuming a tubular form. It becomes the epiphysis The surface of the mid-b1«ain is now considerab1y changed. Two furrows have made their appearance, which divide the 1ateralwalls into three nearly equal 1obes. These do not show at all from the exterior before the removal of the epidermis and the mesoblast.

The gradual reduction of the thalamencephalon by compression between the Central hemispheres and the mid-brain is shown consecutively in Figs. 53, 54, II, If, and 6o. In Fig. 54, and 1ater, the epiphysis appears pear—shaped from the side, but from in front the enlargement on the distal end is seen to be broader than thick. It is borne upon a hol1ow sta11(.

Fig. 57 represents the brain of an embryo considerably older than in Fig. 56. The thalamencephalon is now very much compressed. The anterior half of it, which originally had a rounded protuberance growing from its roof, is now reduced to a small semicircular fold in front of the epiphysis.

Fig. 59 is a view upon the epiphysis of the same embryo from directly in front. It was obtained by removing the cere— bral 1obes, which lie in front of, and partly hide the epiphysia The epiphysis is seen to be composed of a Stall( and an en1arged dista1 extremity ; leading from the Stall( on either side are the epiphysialpedunc1es. In this hgure we are looking through the ventricle of the thalamencephalon into that of the midbrain. On the inside of the lateral walls of the opening are two enlargements, the beginnings of the thala1ni optici. These contain the ganglia hebenulæ 570 Lock. s kvon XI.

In this specigmen the paraphysis had made its appearance as an outgrowth from the cerebrum. It is relatively late in arising, and is entirely distinct from any outgrowth from the thalamencephalon The lateral wal1s of the mid-brain are no 1onger10bed as in earlier stages. Fig. 58 is the same brain viewed from above ; we are looking directly into the cavity of the fourth ventricle.

Fig. 6o shows the brain of a still older embryo. It is the largest embryo of squalus in my collection, measuring 35 mm. in length. The cerebral lobes are divided into right and left halves by a sha11ow furrow. From the back of the cerebrum rises the paraphysis (P); just behind this is seen the semicircu1ar fo1d, belonging to the ’tween-brain, and behind that is the epiphysis, rising above and resting upon the anterior wal1 of the mid-brain. Fig. 6I shows a front view of the cerebral lobes after being removed from the rest of the brain ; the paraphysis is also brought into view. Fig. 62 is the same

I brain with the cerebral lobes removed, and p1aced so as to give

a ful1-face view on the epiphysisz the stalk is considerably longer than in Fig. so.

I have not had materia1 to work out the subsequent history of the epiphysis In specimens about six inches long it is situated in the wall of the cranium, directly over the cerebral lobes, and is connected to the roof of the ’tween-brain by a long threadJilce stallc The enlarged distal end is about the same size as in Fig. 62. cattie shows it in the adult with several capsules on the end of the stalk

Sections of the stages just studied serve to coniirm the observations made from the outside. They do« not add very much to its features, speaking strictly morphologically The surface views have been prepared in such a way that they may stand for reconstruction, but they are more accurate than any actual reconstruction can be, as they show the relations entirely undisturbed.

Figs. I I9—I24 show a series of sagittal Sections covering the i

period of the first appearance of the epiphysis, the reduction of the anterior part of the thalamencepha1on, and the beginning of the paraphysis and the choroid plexus N0.3.] THE FEÆIIEBÆÄTE HEXE-D. 571

Fig. I19 is a specimen of the same age as that shown in Fig. 48. It shows that the roof of the thalarnencephalon is raised into two nearly equa1 protuberances.

Fig. I20 shows the beginning of the reduction of the anterior of these elevations, and the increased growth of the posterior one. In Fig. 121 the posterior e1evation, or epiphysis, has become a tubular outgrowth, while the anterior one is reduced and depressed.

In Fig. 122 the tubular Stall( of the epiphysis is considerably elongatech while that part of the roof of the thalamencephalon in front of it has become reduced to a narrowfo1d that has been carried ventracL The paraphysis has now arisen from the roof of the prosencepha1on. The beginning of the choroid plexus extends from the structure down into the ventric1e of the fort-brain, and its posterior part is connected with the fold in front of the epiphysis The anterior and posterior Commissures are now distinguishab1e.

Fig-s. 123 and 124 exhibit snbstantially the same relations ; the choroid plexus has considerably increased in extent, and is

rapidly becoming much convolutedz the brain vesicles have become reduced. In Fig. 124 the dista1 end of the epiphysis is inserted into a cavity in the roof of the cranial Wall, that is,

the mesenchyme forms a cup over its rounded end.

Z. Cw-e,7)-z«·.ck)-F Hexen-we EPHOÆIFZZIZ O2««g-c)2e-z«JzF.

The recent studies have done much towards establishing a basis for comparison between the epiphysial outgrowths in different groups of anima1s. Beraneclg Francotte, and others claim that the anterior epiphysial vesic1e—common1y designated the pineal eye — is formed independently, having no generic connection with the posterior one, or epiphysis Hill’s observations on the Teleosts coincide with this view. Klinclcowström, on the other hand, claims that the anterior one is formed at the expense of the posterior one, and Substantiates his conclusions with figures of Sections of 1guana, showing the two vesiclesin union with one anothetc It is not slifficult to conceive how the condition represented by Klinc572 Lock: [vo1..xI.

kowström might be brought about, and still not be inharmonious with the other observations. The two vesicles might, in ear1ier stages than he represents, have been formed indepencL ently, side by side, and thrown into communication by an upward growth of the brain roof involving them both. Or, indeed, there may be variations in the details of the formas tions of these two vesic1es.

However this may be, the resu1ts of the different observers all go to show conclusively that there are two (not counting the paraphysis) distinct outgrossvths from the roof of the thalas mencephalon of Petromyzon, Teleosts, and Lacerti1ia. The fate of the anterior vesicle varies in the different groups, while the posterior one persists in all as the epiphysis The anterior one may be formed as a rudimentary organ of transitory existence, as in Teleostsz or it may be developed into the pineal eye, in front of the epiphysis, as in Lacertilia

The sources of nervesupply to the two vesic1es have been made known in Cyc1ostomes, Teleosts, and Lacerti1ia, and these facts are important in establishing homologies. It has been shown (studniåka, Gaskelh KlinckowströrrO that the

upper capsule of Petromyzon receives its nerves from the

posterior commissure, and that the 1ower vesicle is innervated from the left ganglion hebenu1a. Hi1l, in embryos of Te1eosts, has traced a nerve from the posterior commissure into the epipipzysis and has studied its distribution in detail in that structure The anterior vesicle of Te1eosts disappears before it has formed any nerve connection with the brain. Klinckowström («9s) has observed very interesting conditions in embryos of Iguana. In this anima1 he found in one case three nerves distributed to the parietwpineal Organs: one coming from behind and entering the epiphysis, and two coming from right and left ganglion hebenula, respective1y, and entering the pinea1 eye. He found the right and left nerves in three different embryos of Iguana eighteen days old. In one case the nerve from the left ganglion hebenula was nearly the same size as that from the right, but in the other

two individuals the left nerves were much reduced l It should

be added that Klinckowstrom has found the nerve to the pineal N0.3.] THE TXEÆTEBJPÄTE ABBE.

eye in this animal to come normally from the right ganglion hebenu1a, while in Petromyzon it comes from the ganglion of the left Ziele. But, his discovery of the occasiona1 presence of two nerves in Iguana, the left one being smallen affords an interesting transition between the two. It also enables us to account for the marked asymmetry in the ganglia hebenulæ in Petromyzon, on the hypothesis that the right ganglion, in Cyclostomes has, for some reason, ceased to have any nerve connection with the pineal Organ, while the left ganglion has retained its connection.

As Beraneck has shown, a nerve leading to the pineal eye arises in front of the epiphysis, and has but a transitory existence

Putting the facts together and basing relationships on thenervesupply, and comparative study of the vesic1es, we may express the probable homologies in the following diagrams and tab1e :


I. Uppser vcsicle in Petkomyzon, epiphysis in Teleost and Lacerti1ia. II. Lower vesicle in Petrotny2on, HiIPs anterior vesicle in Teleosts, pineal eye in laccrtilia IV. Nerve to posted-im« seminis-sure. XI. Nerve from lower vesicle in Petromyzon to Gan— glion hebenuliy embryonic nerve of transitory existence in Lacertiliix

Z.P. «Zirbelpolster."

s I PETRoMYzoN. THE-USE. . LACERTILIA. I H . . . « Posterior epiphysial Is the superior Is the epiphysis Is the epiphysis vesicle of I-Ii1l. vesicle. Nerve- with nerve con- net-ve- s u p pl y

niislca. « posterior com— posterior com- commissure. Epiphysis of others. [ missure. missure.

Pinealorgan of stucl- I. 1 supply from necting it to from posterior

Anterior epiphysial f "Is the inferior Is formed as in Becomes pineal

vesicle of Hill. ] l l vesicle. the two other eye supplied

Parapineal Organ of Ell. groups, but with a nerve

studniislca. j l aborts. which early cle[ generates 5 74 LOCR [VOL. XI.

If these comparisons are well taken, we have in the Fette— myzon, the epiphysis in its highest state of development Its distal end is enlarged into an eye—like capsu1e that is more differentiated than their inferior vesicle which represents the pinea1 eye. This is the reverse of what is true in most Lacertilia, and, on this account the upper capsule of Petrornyzon has usually been eompared with the pineal eye of Reptilia and Amphibia But the fact that the superior capsule in Petrornyzon receives its nervesupply from the posterior commissure, corresponding in this particular with the epiphysis of other animals, has much more weight in determining its homo1ogies than any purely structural resemblances We are on a better foundation, therefore, in comparing the upper capsule of Cyclostomes to the epiphysis of other animals, than in taking the other Position. It may be noted, in passing, that the superior capsu1e is in the same relative position as the epiphysis, and the inferior one corresponds in position with the pineal eye.

This view would make the epiphysis eye—like in Character, and there are many facts that weigh in that direction. Hil1’s studies speak strongly for the visual nature of the epiphysis He shows a high degree of differentiation in the nerve (:ells to which the nerve from the posterior commissure is distributed. There is also formed in Iguana, Plica, and Phrynosorna an eyelike enlargement at the distal extremity of the epiphysis.

Between the condition in Petrornyzon and that in the Lacertilia many gradations have already been brought to light

In Teleosts, the anterior vesicle arises, as in the other groups, but it is transitory, ancl soon degenerates It would appear from a recent paper by Ritter, that the anterior vesicle in Phrynosoma is more persistent than in Teleosts, but does not reach the grade of differentiation exhibited in the Lacertilia. He gives a Hgure showing the rare occurrence of the anterior vesicle in an adult Phrynosoma, and although large this capsule is not so highly differentiated as the capsule behind it, but to compensate for this the distal end of the epiphysis is highly developed.t· ln other korms of Lacertilia, No. 3.] THE PEXTEBJZH H; Erst-IF. 5 7 5

the anterior vesicle reaches a high grade of perfection, while the epiphysis is not well developed. In most cases, however, there is a deposit of pigment in the distal Position of the epiphysis. The nerve distribution in that organ is known only in the Teleosts.

We might carry our comparisons further and inquire whether the structure present in selachians is to be homo1ogized with the epiphysis or the pineal eye of other forms. The nerve relations are not known in the Se1achians and, theref0re, we have not, as yet, a satisfactory basis for comparison. In its structure and persistent attachment to the brain roof by a sta1k, the outgrowth in Se1achians resembles the epiphysis and one would be inc1ined to say that only the epiphysis is represented in the Elasmobranchs, and that the pineal eye is lacking. Klinckowström has offered the ingenious suggestion that in Se1achians and birds the second or anterior epiphysis is never ful1y separated from the posterior, and this is worth thinking about. There is no positive evidence to support it.

However the above question may be settled, it seems to me that the fundamental features of the morphology of the epiphysis are tolerably clear, and that future work on this subject will be main1y in the direction of ampliiication and discovery of details.

The invertebrate homologies of the pineal eye are not known. Spencer supposed it to be homologous with the median eye of Tunicates, but that relation is a strained one.

Various authors have compared the pineal outgrowths to the ocelli of Arthropods Leydig compares them to the stemmata of Insects, and shows that there are many resemblances to support the comparison

Patten has recent1y studied the deve1opment of the eyes of Limulus, and argues for a homo1ogy between the median eyes of these Arthropods and the pineal eye of Vertebrates He shows the median eye of Limulus to. arise by the fusion of paired eye-stallcs, giving us a case of undoubted union of originally paired Sense-Organs. I have c1aimed a similar occurrence in the epiphysis of Squa1us. But it seems to me-that the 576 Lock: kvon XI.

evidence is too circumstantial at present to bear the interpretation that we already know the invertebrate homologue of the vertebrate pinea1 Organ.

4. Tätig» Don-He Max-»r- oj II«- EPZPÆJFZF

receives support from two sources. It is diflicult to interpret Klinckowströmk diScovery of a double nerve in Iguana on any other hypothesis That interpretation is also strengthened by the claim I have made of tracing the epiphysis in Se1achians back to a pair of epithelial cups on the cepha1ic plate.

Hill (-94) regards the two vesicles he has discovered in Teleosts as having been primitively side by side; and Ritter («94) has recently suggested that the epiphysis and the parapineal organ of Studniälca are right and left mates rather than independent eyes of double origin. It will be remembered, in this connectiorn that Klinckowström found three nerves in one individual of Iguana, two from the ganglia hebenulae, and one from the posterior commissure; that Leydig discovered two accessory pineal organs in Anguis ; Duval and Kalt found as many as three of these Organs, and I have found several pairs of epithelial patches back of the eyes on the cephalic plate, and have traced one pair into the epiphysis It seems to me that all this evidence is more favorable to the idea that the pineal Sense-Organs are multip1e, and individually of paired origin, than to the idea expressed by Ritter.

s. Posapkysisx

considerab1e confusion has arisen in identifying the paraphysis in different forrns. As soon as it was made known that there are two outgrowths from the roof of the fore-brain, it was at once assumed that the most anterior one is the paraphysis, and the posterior one the epiphysis But there is now evidence that there are, at times, more than two outgrowths Hill has shown in Amia the presence at one and the same time of three tubular outgrowths from the roof of the fore-brain. Two of these come from the thalamencephalom andthe anterior one arises from the prosencephalon He points out that N0.3.] THE« IZEÆTEBJEÄTE JJEATZZ

the 1atter is the paraphysis proper, and that in Amia we have to deal with two other outgrowths from the roof of the thalamencepha1on. These he makes homo1ogous with the two vesicles he has described in Teleosts. Hill concludes that in the 1atter group the paraphysis is lacking on account of the nomdevelopment of the choroid plexus This identification is in harmony with the original description of the paraphysis Francotte, who described it in 188 J, and Selenka (-90), to whom the credit of its discovery is usually accorded, both state that it arises from the prosencephal0n. Selenka says: « Wie das Zwischenhirn seine Epiphyse, so hat das Vorderhirn seine Paraphyse.« Francotte suggested that it represents the beginning of the choroid plexus p

Minot (-92), p. 69o, designates the anterior vesicle discovered by Hill (-91) in Coregonus the «paraphysis," but the Structure in question arises from the roof of the thalamencepha1on, and between it and the prosencephalon is a marked downward fold in the brain-wal1, which in many other forms is present behind the paraphysis. It is more probable, as Hill suggests in his1ater paper («94), that the paraphysis is lacking in the Te1eosts.

It is to be understood that in some forms there are two epiphysia1 outgrowths from the thalamencepha1on that are entirely independent of the paraphysis When the latter Structure is present in conjunction with the former two, as in Amia, there are then three separate outgrowths from the roof of the fore-brain.

As already indicated in this paper, the paraphysis is present in Squalus as an outgrowth from the prosencephalon, and is connected with the choroid plexus


The formation of the auditory saucer is preceded by a thickening of the epidermis along the sides of the head in the

auditory region. The thickening portion occupies a larger area »

than that used by the ear vesicle when it is Erst formed, and it is coniiuent with the epidermal thickening just above the 578 Lock. » [Vor«. XL

branchiae As Mitrophanow has pointed out, this thickenecl patch of epithelium is composite in character; it embraces not only the epithelium for the formation of the auditory plate, but that to form the so—called branchial Sense-organs, and also the beginning of the lateral line. I had noted this relationship in Squa1us before seeing Mitrophanoufs paper, and 1 lind no com— parison that my resu1ts correspond —— so far as the earlier Stages are concerned—very closely with his. He has, however, carried his studies considerably farther, and has shown how the different branches of the lateral-line System arise.

The general relationship noted between the auditory plate, the branchia1 Sense-organs, and the lateral line is very similar to what Wilson found in sea bass. In that form the Connection between ear, branchial Organs, and lateral line is evidently more c1ear on account of the presence of a distinct sensory furrow and the way in which the separation into three parts takes place.

The general thickening in the sharks is not so well circumscribed. The fact that these different organs proceed from a common epithelial thickening indicates relationship of a fundamental kind. After separation the ear gives further evidence of its relationship with the lateral organs by preserving its canal (endolyrnphatic duct) connection with the exterior and developing in a manner that is characteristic of the canal organs of the lateral line.

Mitraphanow departs from the usual point of view that the organs of the lateral line are metarnericz and in that particular, I think, I should be inclined to follow him.

The auditory plate is at first differentiated from the general thickening, and its epithelium then becomes gradually rounded up into a circu1ar area, that is depressed in the centre—the auditory saucer. This structure sometimes covers the space of three neuromeres, and I have one surface preparation in which the outer epithelium shows three bars running vertically across it; these bars correspond with the underlying neuromeres.

We may interpret these superficial bars either as being moulded upon the underlying neural segments, or as being No. 3.] THE IXEJBIEBJZAIIE READ. 579

due to the same general cause as the neura1 segments I am of the opinion that the latter alternative is the correct one.

When iirst formed, the auditory saucer is opposite the ninth neuromerez but subsequent1y, while it is still in the saucer condition, it shifts backwards, and cornes to lie at the side of the tenth neural segrnenh and later still, as a capsule, it lies opposite the eleventh neuromere

The history of the auditory vesicle in sharks has been worked out beyond this period by Ayers (-92), and it is very clear from its mode of growth that it is directly related to the canal organs of the lateral 1ine.

A consideration of the so—ca11ed branchia1 Sense-Organs and their ganglia is reserved to be published later with the part on the Nerves

NOT-E. ——I have been indebted to the persons mentioned below for material that has helped jill up the gaps in rny collection of embryos, and I desire here to express Iny appreciation of their lcindness

To Miss Julia B. platt, for the loan of Sections of Acanthias ; to Dr. E. L. Mark, for young stages of Squalus ; to Mr. F rank smith, for a considerable number of embryos ; to Professor J. E. Reighard for the loan and free use of his entire collection of elasmobranch embryos; to Professor A. D. Morrill, for early stages of squalus

LAKE FoREsT, ILL1No1s, January, I895. 580


«84 ’88


’28 ’85


’92 ’84

«87a ’87b








Lock: kvon x1.


AHLBORN, F. Untersuchungen iiber das Gehirn der Petromyzontem Zeixrclznjc gis-irr. Zooä Bd. xxxix, pp. 19I—294, Taf. I3—I7.

Eint-Bonn, F . Ueber die Segmentation des Wirbelthier—I(örpe1-s.

ALLE, E. P., Jr. The Anatomy and Development of the Lateral Line System in Ewig: its-wo. »Ja-tm. oj Mark-Cz. ji, p. 463.

AYERS Hei-Man. Vertebrate Cephalogenesis Il. A contribution to the Morphology of the Vertebrate Ear, with a Reconsideration of its Functions ji«-m. ej Max-pl« VI, Nos. I and 2, pp. I—32o, p1s. i—xii.

BAEIH KARL E. voN. Th. i, S. 64.

BEARIV JOHN. The System of Branchia1 Sense-Organs and their Associated Ganglia in JcÆZJzyoYJ-2Z-z. gutes. ji«-»O. Max-«. sc.

BEARD, JOHN. Morphological Studies. I. The Parietal Eye of the Cyclostome F ishes. gis-of. Ins-XX. XII-By. sc. xxix, pp. 53——73, pls. 6 and 7.

Bis-um, JOHN. The transient Ganglion cells and their Nerves in Æcxjcx Hex-Er. Alter«. Ämse-Yes. vii, S. I91——2o6, Fig. I—8.

BIZRANECIS E. Recherches sur le developpement des nerks cräniens ehe: les Lezards Bezirks? Zwä Fzxäixra T. i, pp. 5I9-—6o2, 4 pls.

BERANECIS E. Replis medullaires du Poulet. Besitzes? Zur-Z. sei-Irre.

BERANECIS E. Ueber das Parietalauge der Reptilien. »Ja-ers. Zeärrcfm Bd. xxi, pp. 374-4ko.

BFJRANECK, E. sur le nerf parietal et la morph0logie du troisieme oeil des Vertöbres Arg-«. Klasse-ges. Bd. VII, pp. 678—689.

Brig-kurzen, E. contribution å Pembryogenie de la glande pineale des Arnphibiens Basis« Zwist-« ci- Zoo!czgsie. T. i, pp. 2s5—288, 3 pls.

B1scIIoFF. Entwiclcelungsgeschichte des Hundeeis.

BURCKHARDT, R. Die Zirbel von Jcfxxlryxxpäzlr gzzxrimprxxr und ProHopierixr Ascesi-»F. Als-Hex. Eises-Yes. vi, pp. 348—349.

Blmctcnnnnæz R. Die Homologieen des Zwischenhirndaches und ihre Bedeutung fiir die Morphologie des Hirns bei niederen Vertebraten. Eis-or. -4»-ge-;g-y. IX, pp. t52—I IF, I tig.

BURcIcHnRD-r, R. Die Homologieen des Zwischenhirndaches bei Reptilien und Vögeln. Ase-«. Akt-argen IX, pp. 32o—324, 3 ligs

CARR1TRE, J. N euere Untersuchung über das Parieta1organ. Beding. Ckyziyafåä fix, pp. I 36—I49.

CATTIEH J. Recherches sur la glande pineale (Ezåx«pfz·j-.rz·r ers-esse) des Plagiostomes, des Ganoides et des Teleosteens. xlnrlk sie Bis-Z. iii,

pp. 1oI—I94, p1s. 4—6.

Ueber Entwickelungsgeschichte der Thiere. NO.

«89 ’75 ’90

’69 78






se· ’92










THE »Es TEBJEH TE fix« D. 58 1


DUvAL et Kam: Des yeux pineaux multiples chez1’0rvet. Eos-»F. Kaiser! d, Zcz Fort. sc? Bis-Z. No. 6, pp. 85-86.

DOHRN, ANTON. Der Ursprung der Wirbelthiere und das Prineip des Functionswechsels Leipzig.

DOHRN, ANTON. Neue Grundlagen zur Beurteilung der Metamerie des Kopfes. HEXE-XI. a. d! Zool. Fzcxzc Z« Magst-Z. Bd. ist, p. 334.

DURSY. Entwickelungsgeschichte des Kopfes. Tiibingerx Atlas Taf. iii.

EHLERS, E. Die Epjphysis am Gehirn der Plagiostomeix Zeiss-»Es. J. wiss. Inn-Z. Bd. xxx, S. 607—-634.

EYCLESHYMER, A. C. Paraphysis and Epiphysis in Amblystorneu Herze: Ase-Zacken vii, S. 21 F—-2I J.

F RANCOTTE, P. Recherches sur le developpement de l’epiphyse. Erde. sie Bioä T. viii, pp. 757—-82I.

FRANCOTTZT P. Note sur Poeil parietah 1’epip11yse, la paraphyse et les Pia-«» cleosofsfer du troisieme ventricule FULL ei« Bären! Ray. IS'-Figura. T. xxvii, pp. 84—112, 2 pls.

FRORIEP, A. Zur Frage der sogenannten N euromerie. Äms-«. Schaf-Trick. o. VersammL Wien. S. I62—I67.

GASRELI., W. H. 0n the Origin of Vertebrates from crustaceandike Ancestors. gar-r. Jus-m. Mir-«. Fa, pp. 379—444.

GEGENBAUIV CARL. Die Metamerie des Kopfes unddie Wirbel— theorie des Kopfslcelettes Mai-pls. Joleyzu I 3. Bd., p. 37.

HERRIcIg c. L. Embryological Not-es on the Brain of the Snalce. Jomua ej· co--«-O. Mem-ei. pp. I6o—-1 76, F pls.

HERR-Ists, 0sCAR. Lehrbuch der Entwiclcelungsgeschichte des Menschen und der Wirbelthiere

Ell-L, cui-is. Developrnent of the Epiphysis in Cur-Fuss« wieder. Jota-». oj May-Mr. Vol. v, pp. Eos-sie.

HIL1«, cHAs. The Epiphysis of Teleosts and Amia. Jus-»Ja. ej« Worzu-X. Vol. irr, pp. 237—266, 2 pls.

HOFFMANLH C. K. Reptilien in Bronn’s Klassen und Ordnungen des Thier-Reichs. S. I966—I 971, 1989.

HOFFMANW c. K. Ueber die Metarnerie des N achhirns und Hinter— hirns und ihre Beziehung zu den segmentalen Kopfnerven bei Reptilien-Embryonen. Zuoz -4«z-x;gey. xii, S. 337—339.

KILLIALL Zur Metamerie des selachierkopfes Verleg-IN. Ärzte-f. Garczärxfa F. VersammL München. S. 85—Io7.

K1.1NcI(owsTRöM, A. DE. Le premier developpement de Poeil pineah Pöpiphyse et le nerf parietal chez Jgxxxzszzz Kasse-»Mein. · Hans. Äzxaezsgem viii, pp. 289-299.

KLINcKOwsTRöM, A. DE. Beiträge zur Kenntniss des Parietalauges Zwä »Ja-Ists. VI, s. 249—-28o, 2 Taf.

KUPFFER, C. vors. Primäre Metamerie des N euralohrs der Vertebraten. Eis-Z. ei. Aktien! a» XII-wägen, PJx·)-.r.-Mk2z·-E. Cz, vom F. December.

Passe-»F. 5 8 2 «91 see -94 seo 298


’94b ’94o ’90 78 ’81 ’82 «« "92 PS


Ja? see



Lock( krank. XI.

Kur-Eisen, C. vors. Die Entwickelung des Kopfnerven der Vertebraten. Kerkers-XX. X. Heer. Geer-MAX. München.

KUPFFEIIH c. vors. Studien der vergleichenden Entwickelungsgeschichte des Kopfes der Kranioten Heft t.

Kopf-THE, c. von. Ibid. Heft 2. München und Leipzig.

Las-»Die, F Im. D·as Parietalorgan der Amphibien und Reptilien Anatomisclkhistologische Untersuchung. Erkennt-II. as. Fee-take«åeyg. Mai-»F( GFXFZZFCL XVI, Z. S. 44I-55I.

L0cY, WM. A. The Derivation of the Pineal Eye. -4-.c2r.«4«ae-Jgey. ix, S. I69—-I8o, F iigs. Nachtrag zu diesem Aufsatze Anat Ase-Zeiger. ist, S. 23I—232.

L0cY, Wirt. A. The Optik: Vesieles of Elasmobranchs and their Serial Relation to 0ther Structur-es on the cephalic Plate. · Ins-m. ej May-»Ja. Vol. ix, pp. II5—-I22, 6 tigs.

L0cY, WM. A. Metameric Segmentation in the Medullary Folds and Embryonic Rim. Erz-«. Lege-Yes. IX, S. 393—-41F, II läge. LocY, Wm. A. The Mid-Brain and the Accessory Optic Vesicles.

Hans-«. Exzesse-«. ix, S. 486—-488. MCCLURFY CHARLES The Segmentation of the primitive Vertebrate Brain. Jene-». ej« Alex-Hefe. its, pp. 35—56, I pl.

Maximal-L, A. M. Development of the cranial N erves in the chielin Leser. Jena«-». Wiss. sc. xviii.

Magst-ihn, A. M. Observations (with W. B. Spencer) on the cranial Nerves of sag-Eines. »Es-as. »Ja-»He. ej· Wiss. sc. xxin

Marien-ihr» A. M. The Segmental Value of the cranial Nerven Jene-». DIE-mit. ais-d EVEN. Vol. XVI.

MIHALxov1cs, Entwielcelungsgeschiehte des Gehirns. Nach Unter— suchungen an höheren Wirbelthieren und den Menschen. Leipzig.

Mucor, CHARLES S. Human Ernbryo1ogy. New York.

MITnoPIIAN0w, Paul» Etude embryogenique sur les Selaciena Zins-E. «? Zeig! Enge. T. i, pp, 16t—22I. ·

PATTEJHH WILLIAIW On the Morphology and Physiology of the Brain and Sense-Organs of Zier-Max. gern-«. Jan-». Mit-«. sc. muss, pp. I—96.

0RR, H. A contribution to the Embryology of the Li2ard. Jene-». cjMeyjblk Vol. I.

Ost-man, HENRY F. A contribution to the Internal Structure of the Arnphibian Brain. jin-www. ej« May-pl« ii, pp. 5I—96.

0WsJANNIKOw, P. H. Ueber das dritte Auge bei Paxsnwjawzjkixøsiizs Ziizsssx nebst einigen Bemerkungen über dasselbe Organ bei anderen Thieren. Mär-z. cis.- Zhslcad list-Käf. n?- Fz«.-Pe’xey.5e3o»yg. xxxvh p. at, I Taf. ·

Faktor-kann, S. A. La glande pineale et le troisidme oeil des Ver— tebres These. Paris, I887. No. 3.] THE FEÆTEBÆÄ IJE III-TO. 583

’9s EIN-ZEIT, A. sur Pæil parietal access-one. XIV-er. Away-Zeiss. ix. S. los-I I2.

«94 PRIZNANT, A. Les yeux parietaux accessoires dDslwgzxix jsczgz·!2«.s. BZZZZQJFXEPJL XII-Cz. No. H, N ov.-Dec., pp. I—7.

’89 PLATL JULIA B. studies on the Primitive Axial segmentation of the Chick. FULL. XII-«. case-P. Zoäz Vol. xvii, pp. I7I—I9o, 2 p1s.

’8s RAE-L, C. Bemerkungen iiber die segmentierung des Hirns. Zcswk -4-.-«.28-Jg.e-«.

’50 REMMZ Untersuchungen iiber die Entwickelung der Wirbelthiere. Berlin, 18Fo—-5F.

VI. RITTER, W. E. The Parietal Eye in Some Lizards from the Western United States. BUT-IX. Markt. Ost-»F. XVI. Vol. XX, No. Z, 4 pIs.

«94- RITTER, W. E. On the Presence of a Parapineal Organ in Pfzyymck Eos-m. XIV-XX. Eises-Xexes. irr, S. 766—772, I Hg.

’85 RYDER, JOHN. On the F ormation of the Embryonic Axis of the Teleostean Embryo by the concrescence of the Rim of the Blastos derm. ANY. Max. sei-c, pp. 614—6I s, I jig.

«87 Sc0TT, W. B. The Embryology of Pciskiwkygow Jomw ojxlsosplk Vol. i, No. 2.

’90 SELENKA, E. Das Stirnorgan der Wirbelthieren Bis-Z. Texts-EIN. Bd. X.

’86 SHIPLEIL 0n Some Points in the Development of Pexyomjzosx JE"««z-z·czxz«ZeTr. gis-m: »Am-». Mehr. sc. xxv.

"’94 SORENSEIV A. D. comparative Study of the Epiphysis and Roof of the Diencephaloth Jan-»O Cis-»F. New-CI. Vol. its, pp. I2-72; I53—I7o.

’86 SPENCER, W. B. On the Presence and structure of the Pineal Eye in Lcxc--z«-·Zx"cx. Lichts. Joixrw Miit-c sc. Vol. Jst-ev, p. 239.

’88 STRAHL und MARTIKL Die Entwickelung des Parietalauges bei ÄAFYFJJTAFZFF und Lerci-»Sk- TJFQØEJW Ast-E. J. xljmt «. Ewkza S. I46—I6I, I Taf.

’82 VAN WIJHE. Ueber die Mesodermsegmente und die Entwickelung der· Nerven des Selachier-Kopfes. Mai. Feffzcxxrizl X. Eis-Es. Läg-Z. -Hi--zsz«a«icwx. xxii.

’92 WATERS B. H. Primitive segmentation of the Vertebrate Brain.

gzzxxjc Jus-IX. Miit-c sc. xxxiiih pp. 457—475, I pl.

’84 Wann-ais, c. O. The External Morphology of the Leech. Preise. Hin-g. Ärmel. Er« am! sc. Vol. XX, No. F, pp. 76—87, I pl.

’92 WHITMADH c. O. The Metamerism of cis-Preises. Fuss-HEXEN gar« sz·cöc»zzsgsrc» Geåicyxxzwge Ækcciozj lL«--»c-kæ-«!.s. S. 385-39F, 2 Taf.

VI. ZIMMERMADL Ueber die Metarnerie des Wirbe1thierlcopfes. Prslxxrmiä

rä XIV-ex. Gereizt-Ä. F. VersamtnL München. S. 1o7-—It3. 584 » Lock: kvop XI.


END-»m- May-Es.

Ue, F, H, etc. Metameric Segment. »Ja-r. wisse-blast.

XII« line drawn in front of the ante- Max. metameric Segment. rior neuromere of the vagus d km. nasal epitheliutth net-vie. OF. ». Optik: net-se.

tm. auditory vesicla ex. .k. optic stalk acn ». accessory Optik: Ists-Ziele. pp. z« « optic vesicle.

«. s. anterior portion of alimentary F. paraphysis can-Eil. 4 w. serial Sense-Organ.

cåä cerebe11um. III. tha1amencepha1on. Mk. prosencephaloxx « Mel. ex. thalamus opticus. Hei. epiphysis · New. 4 fourth ventric1e. II. c. manclibulax head cavitzk Zp. Zirbe1po1ster. ynä mid-brain. N().3.] THE ZEÆTEBÆÄTE HEXE.


The ligures are from untouched negatives and are noteworthy in showing the very early condition of the optic vesicle, the accessory vesicles and in some cases the primitive rnetameric segrnenta They are all Photographs of spart-«« irre-träfe.and, with the exception of Fig. J, are X about 2o diameters

Fig. 1. PhotograPh of embryo between Balfour’s stages i? and c.

FIg. 2. somewhat older embryo. The embryonic rim on the left side shows faint1y Some of the metameric segments (not reproduced by the artist).

F1g. Z. Slightly older embryo to show the forrnation of the heachplate and the Central wedgeshaped Process thereon. This is the stage in which the neural folds are started along the margins of the body. They are ventrally curved.

Fig. 4. stage with a rounded head-Plate when the optic vesicles first become evident.

Fig. s. Another embryo, about the same age as the Preceding, in which the optic vesic1es and one of the accessory vesicles are shown.

F1g. H. Somewhat older embryo showing the infolding for the optic vesicles extending across the median Plane.

Fig. 7. Older embryo somewhat higher magnifred The headsplate is very broad, the trunlc narrow. The neural folds of the head lie nearly in the horizontal Plane. The rnetarneric segments show well on the left margin of the head-P1ate. They exist in the earlier stages, but are very diiiicult to catch with the camera.

FIg. 8. speeirnen in which the dePressions for the optic vesicles show very distinctly

FIgs. 9 and ro. Two ernbryos older (although smaller) than the preceding. In both two Pairs of accessory optic vesicles are to be seen on the cephalic Plate baclc of the Primary optic vesicles.

Fuss. II and Ia. Two specimens slightly older than the preceding two, seen from different Point of observatiow

FIg. r Z. Embryo after the neural folds have begun to grow uPwards, seen obliquely from the left sicle. Gives an external view of— the vesicles on one side and an internal view of them on the oPPosite side of the neural folda

FIg. I4. Embryo from which Fig. 29, Plate XXV1I, is drawjx Shows metaIneric segments on the exPosed ventral surface of the neural folds.

FIg. I s. Embryo of same age as Fig. 13 and Fig. Hi, Plate XXVII. shows metarneric segments on the neural folds in front of the eye vesiclea

FIg. ro. Embryo of the same age as Fig. 9 just above it. Seen in a Position more favorable to bring out the accessory vesicles on the cephalic plate. The optic vesicle on the right side shows as an external Protuberance.

Fig. r7. somewhat older embryo viewed obliquely from above. shows the optic vesicle of the right side as an external rounded Protuberance, and that of the left side from within as a cup. Behind the optic vesiele on the left side of the cephalic place are four accessory vesiclea They show their serial reiation with the Primary optic vesicle.

FIg. 18. specimen showing a large development of the central tongueislilte Process. Embryo slightly older than that in Fig. 7. 586 Lock.

Fuss. 19 and 2o. Side view of two heads of embryos vvith open neural groove to show the external appearance of the optic vesicle

FIG. 21. Embryo of same age as Fig. 13 viewed obliquely from above.

F1G. ge. Side view of embryo of the same age (with broadly open neural gross-re) to show the external appearance of the optic vesicie and the other vesicles behind it.

Mo. 23. Embryo after partie-l closure of the neural groove Showing two openings, an anterior and a posterior one, into the neural canaL

FIG. 24. Some-what older specimen with head broken off. showing three openings (sma1ler than those in the preceding embryo) into the neural canaL .Jc«i».««Z M· xwxyfzofozgzy !«F2!..-l7.

THE-Fast. s« Dir» e.- s SICH-rege: Fmxcskfnnfål THE-is. Lock. know. DIE IJEJE III-Fa re« III-Hi D. 587


All the figures are of Fee-Z« cxcxxmbäiax ,- they are drawn with the aid of thecamera, and are all d( 45 diameters.

FIG. 2 s. Young embryo intermediate between Balfoufs stages B and C. The embryo so far as formed is divided into eight pairs of metameres and these arecontinued without brealc, or any change in Character, into the halves of the embryonic rim. The eleventh Inetamere which, in later stages, lies in kront of the vagus nerve is now on either side the third one from the axial embryo.

Fig. 26. Somewhat older embryo, showing the change in form of the head region. The axial embryo now includes about fifteen pairs of rnetamerea

Fig. 27. Slightly older than the precedingz there are about eighteen pairs of rnetameres in the axial part of the embryo and,·as in the forrner instances, are continued into the embryonic rim. Two longitudinal marginal furrows have appeared, separating two marginal bands from the rest of the embryo. Along the line of these furrows are seen four depressions that mark the very beginning of segmental Sense-Organs. ·

FIG. as. View from the upper side of the same embryo illustrated in Fig. 29. The cephalic plate is now clearly marked off from the more slender trunk region. The depressions for the optic vesicles (c-P.) have made their appearancqz

F1G. 29. View of the same embryo from the ventral aspect. The yolk has been cornpletely removed, and we get a view directly into the gastrular cavity. There are eleven pairs of metameres in the broad part of the cephalic plate. The neural folds are ventrally curved The outlines of the figure and the neural segments are too syrnrnetricaL

FIG. Ho. Older embryo with neural folds lying in the horizontal plane. The broad cephalic plate is in marked contrast with the slender trank.

Fig. Hi. Embryo in which the neural folds have nearly attained the vertical Plane. The neural groove is still open. The optic vesicle (qk).) and the combined vesicle of mid—brain and accessory optic Eis-z. -l- A. tax-L) vesicle show on the sides of the head. There is also the beginning of rnandibnlar head cavity El. c) and the branchia1 pouch. The original Inetameric divisions are still very plain.

FIG. Ja. Embryo just after the closure of the neural groove in the anterior end. The posterior part of the neural canal is not cornpletely closed. Note the metarneric divisions indicated by numbers J, e, F, etc. ·

Fritz. 33. Embryo after complete closure of the neural groove and before the appearance of the ear vesicle.

FI(;. 34. Embryo after the differentiation of the ear saucen The five anterior metarneric divisions are no longer distinguishable, those of the hind-brain are prominent and are approximatecl in the middle plane. 0ne gilbcleft has broken through. The nasal pit has started. ·

Flci 35. slightly older embryo showing several characteristic changea The— line of neural segments are being forced apart by lateralgrowth of the roof of the hind-brain. The fifth nerve is plainly visib1e. Over the gillsclefts runs a continuous nodulated thiclcening containing the branchial Sense-Organs and the radirnent of the lateral 1ine. 588 Lock:

Flcsp 36. Some-what older embryo differing from the preceding mainly in show ing the rudiments of the seventh, eighth ninth, and tenth next-es. Note the lens and choroid Hssure in the eye vesic1e.

Eis. 37. Slightly older than the precedjng The line of neural segments are undergoing some ehanges whereby the concavity on the lower margin is made to correspond vvith a crest on the upper margitx In embryos of about the age represented in Fig. 36 or a very little older the epiphysial outgrowth arises from the roof of the tha1amencepha1on. lot-Aal of· Jløæzøleologg Ist-I. 590 LOCJT


A series of partial dissections of embryos of syst-III« in such a way that the brainswalls have been laid bare. .Figs. 46—49 X 45 diameters Figs. »so-so X io diameters.

Fig. 38. Embryo with open neural groove. The epidermal layer and the mesoderm have been removed from the sides of the brain-Wall; behind the optic vesicle is seen a bilobed protuberance—t·he combined mid-brain vesicle and the anterior accessory optic vesicle. ·

Fig. 39. 0lder embryo with neural canal partly formed. The specimen shovvs the same condition of mid-brain vesicle and that of the anterior accessory optic vesicle »

Fig. 4o. The braiwwalls of an embryo of the same age as that shown in Fig. Ja.

Fig. 4I. Embryo before dissectiom showing especially well the contours of the brain-wal1s.

Fig. 42. The same embryo after exposure of the braiwwalls by dissection The thalamencephalon is well exhibited · The mid-brain is bilobed.

Fig. 43. slcetch of partially dissected embryo just after the appearance of the auditory vesicle.

Fig. 44. The exposed brainwvalls of an embryo slightly older than that represented in Fig. 34. ..

Fig. 45. Brain of embryo about same age as that in Fig. 35. The auditory vesicle has been Ieft in Position. The midbrain is now indistinctly trilobed.

Fig. 46. Brainwvalls of embryo with the optic vesicle removed. About the same age as the preceding

Fig. 47. Dissection of the brain of the embryo of which Fig. 36 is an external view. The mid—brain is distinctly tri1obed. There are eight clearly marked segments in the hind-brain.

Fig. 48. Dissection of embryo slightly older than the one represented in Fig. 37. The thalamencephalon is now dekinitely marked out by furrows ; it bears upon its summit two rounded confluent pr0tuberances.

Fig. 49. Dissection of brain of embryo somewhat older than the preceding The thalamencephalon is clearly defined. The posterior protuberance from its roof has grown much faster than the anterior one. The former is the beginning of epiphysis. There are nine neural segments in the hind-brain.

The embryos are now too large to represent advantageously on the same scale, and in the following figures the scale of magniiication is reduced from 45 to to— diameters

Fig. so. Shows embryo of same age as that represented in Fig. Ja.

Fig. Hi. Embryo of nearly the same age as that represented in Fig. 34.

Fig. se. Partly dissected embryo of about the same age as that represented in Fig. 36 and again in Fig. 47.

Fig. II. Braiiikwalls of an embryo just older than that shown in Fig. 49.

Flgs. 54 and IF. Successively older embryos to show especially the changes in the tha1amencephalon and the outgrowth from its roof of the epiphysis.

Fig. so. The same brain shown in Fig. IF, with the cerebral lobesreinoved

and turned so as to vievv directly against the epiphysia »Jetzt-seit? of JXGXZJJZAZCJYV list-IT J THE-I»

si !

« «« -..« 592 zociz


Figs 57—62 a continuation of the series of dissections shown on the two previous plates. All X io diametera Figs. 65—87 X 45 characters.

Fig. 57. Side view of brain of an older embryo than that slcetched in the fore— going ügure The neural segments have now become ob1iterated, and the three lobes of the mid—brain (secondary«divisions) are no Ionger distinguishable The epiphysis is well developed ; in front of it is a semicircular fold of the braiwwallz this structure is the remnant of the elevation which started in front of the epiphs ysis on the roof of the thalamencephalon (see Figs. 48 and 49). It has become rednced by compression between the rapidly growing adjacent brain regions. The paraphysis is also visible in this drawing

Fig. 58. View of the same brain from above looking into the cavity of the fonrth ventriclia

Fig. so. The same brain after removal of the cerebra1 lobes and arranged so as to show the epiphysis in front view.

Fig. Ho. Brain of older embryo showing substantially the same features as the preceding

Fig. 6I. The cerebral lobes of the same brain after removal. Note the paraphysis arising from the posterior part of the roof of the prosencephaloiic

Fig. Ha. The same brain after removal of prosencephalon and arranged so as to show to best advantage the epiphysis with its Stall-r.

Fig. 63. Horizontal Section of ernbryo shown in Fig. 25 to show the general appearance of the metameres in Section. X so.

Fig. 64. Horizontal Section of embryo sketched in Fig. 26 showing metameres in Section, and adjacent layer of mesodernn X Ho.

Fig. 6I. Section of heacl of Tasse-ob ocelzaro through the optic vesicles at the time of their first appearance compare this with Zieglers’ Pl. IV, Fig. i93.

Fig-s. 66, 67, 68, 69, 7o. successive sagittal Sections of embryo just after closure of the neural groove and prior to the appearance of the ear vesicle show the primary for-obtain, the inid— and hind—brains, and three very prominent neural segments of the hindsbrain The segments are the seventh, eighth, and ninth respectivelyn When the ear is first differentiated it arises opposite the ninth Segment.

Fig. 71. Deeper section of the same embryo showing a curved line of mesoss derni over the mandibular cavity.

Fig. Je. Sagittal Section of ernbryo near the median plane after the formation of the auditory saucer and the complete closure of the neural groove. The five neural segments of the fore- and mickbrain are exhibited The second neural Segment nearly coincides in Position with the neuropore

Fuss. 73, 74, 7 H. Three successive sagittal Sections of an embryo of about the age of that drawn in Fig. 34. It is slightly older, shows well the neuromeres of

· the hind-brain. The ear capsule is in the space of the tenth neuromere. In front

of it in Figs. 73 and· 74 are seen the roots of the seventh and. eighth nett-es.

Fig-s. 76-87. Sagittal Sections of the specimen photographed as Fig. 17, Pl. XXVL Giving evidence of a series of cupälike depressions on the neural plate of headand trank. The series of these cupped areas is terrninated in front by the optic vesicles 0n the cephalic plate they are relativelzss large and resemble the primary optic vesicles in mode of origin and in structure How far the series extends into the trank I have not been able to determine. ji«-IS« Z »M«k2.syzsgckssfx-jejy , «;

«. .-.«—-,x7s.«k.MPOMPO4

se«  ·- -.-. -. H 94 LOC P.


Flog. 88—1 la. Twentystive transverse Sections of an embryo very slightly older than that represented in Fig. I6. The Sections 88-1oo 1ie in the region of the cephalic p1ate. The following Sections los-I 12 lie in the neck and trunlc regions

These Sections are remarlcable in bringing to Iight serial depressions along the walls of the neural folds. They show that the serial cup—like differentiations extend back of the cephalic plate. I have not been able to determine the number of serial differentiations of this Character in the embryo, but it is clear there are several pairs behind the cephalic plate upon which I have noted in surface study four pairs in addition to the eyes. X 45 diameters.

[Figs. II 3——1 I s were drawn from nature by Miss Tanetta Gilleland Figs. 117 and 118 are talcen from Kupffer’s «Studien zur vergleichenden Entwickelungsgeschichte des Kopfes der KraniotenX and Fig. I 16 is after FroriepJ

FIG. Ir3. View of the head-end of embryo of xlmzhxxoma in the open neural groove stage, X about Io diameters. shows segmental folds in the neural folds

and a srnooth neural plate.

Eis. rr4. View on the caudal extremity of the sarne embryo.

Fig. II s. Embryo of Kam: Pelz-sinks- showing large obvious segmental folds in· the median plate and also srnaller fainter folds in the neural ridges. The latter correspond to the segments in the neural ridges in -4-zäJ·J-.kzo»-a.

FIG. 116. Embryo of THE» cyisiazms after Froriep. showing large obvious folds in the median plate with unsegmented neural ridges These median folds

probably correspond to those in Reiz« Feier-Iris and nof to the Inetameric divi—

isions in the neural folds.

Eis. I17. View on the head-end of Falæwamlsa »Im, according to Kupffen showing segmental folds in the median plate but none in the neural ridges com pare vvith Fig. II s.

Ho. Its. caudal end of same embryo.

Figs I 19-124 are sagittal Sections of syst-il« arg-Esset. X about 45 diameters.

Eis. I19. Section of head of embryo about same age as those represented in Figs 37 and 48, showing the roof of the thalamencephalon raised into two eleva-tions. · ·

Fig. I2o. somewhat older ernbryo showing the tubulaklilre grovvth of the .epiphysis. ·

pro. I21. still older stage showing reduction of the anterior part of roof of thalamencephalon and great increase in depth of the furrow in front of the thalamencephalotn

Fu; r22. 0lder stage in which the paraphysis has made its appearance from sthe roof of the prosencephalotn The choroid plexus has also started.

Flcs 123 and 124. Two older stages showing the increase in length of the epiphysis ; its distal end is somewhat enlarged and is inserted into a concavity in the cranial roof. The paraphysis is indicated at pp. — The choroid plexus is considerably increased in extent and hangs into the cavity of the fore-brain. III-«. .I7. Pf, . Akt.