>N>U ?L>l3o^ ^- A)^-^— ^-^ THE DEVELOPMENT OF THE CHICK AN INTRODUCTION TO EMBRYOLOGY THE DEVELOPMENT OF THE CHICK AN INTRODUCTION TO EMBRYOLOGY BY FRANK R. LILLIE PROFESSOR IN THE UNIVERSITY OF CHICAGO NEW YORK HENRY HOLT AND COMPANY 1908 Copyright, 1908, BY HENRY HOLT AND COMPANY PREFACE This book is a plain account of the development of the never- failing resource of the embryologist, the chick. It has been neces- sary to fill certain gaps in our knowledge of the development of the chick by descriptions of other birds. But the account does not go beyond the class Aves, and it applies exclusively to the chick except where there is specific statement to the contrary. Projected chapters on the integument, muscular sys- tem, physiology of development, teratology, and history of the subject have been omitted, as the book seemed to be already sufficiently long. The account has been written directly from the material in almost every part, and it has involved some special investigations, particularly on the early development undertaken by Doctor Mary Blount and Doctor J. T. Patterson, to whom acknowledgments are due for permission to incor- porate their results before full publication by the authors. As the book is meant for the use of beginners in embryology, refer- ences to authors are usually omitted except where the account is based directly on the description of a single investigator. A fairly full list of original sources is published as an appendix. Figures borrowed from other publications are credited in the legends to the figures. The majority of the illustrations are from original preparations of the author: Figures 46, 48, 50, 51, 52, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 71, 72, 73, 74, 75, 99, 105 and 106 were drawn by Mr. K. Hayashi; the remainder of the original drawings were executed by Mr. Kenji Toda. The photographs in Figures 118, 119, 120, 168, 181, 182, 189, 194, 197, and 231 are the work of Mr. Willard C. Green. Some of the figures may be studied with advantage for points not described in the text. Acknowledgments are also due my colleague, Professor W. L. Tower for much assistance, and to Doctor Roy L. Moodie for special work on the skeleton, and photographs of potash prep- arations reproduced in Figures 242, 246, 249 and 250. The best introduction to the problems opened up by the study iii iv PREFACE of embryology is a careful first-hand study of some one species. It is in this sense that the book may serve as an introduction to embryology, if its study is accompanied by careful laboratory work. In some respects it is fuller, and in others less complete, than other books with which it might be compared. On its comparative and experimental sides, embryology is the only key to the solution of some of the most fundamental problems of biology. The fact that comparative and experimental embry- ology receive bare mention is not due to any lack of appreciation of their interest and importance, but to the conviction that the beginner is not prepared to appreciate these problems at the start; to the belief that our teachers of embryology are com- petent to remedy omissions; and finally to the circumstance that no one book can, as a matter of fact, cover the entire field, except in the most superficial way. The development before laying and the first three days of incubation are treated by stages as far as possible, and this mat- ter constitutes Part I of the book. It involves the study of the origin of the primordia of most of the organs. The matter concerning the later development is classified by the organs concerned, which seems to be the only possible way, and this constitutes Part 11. The first part is complete in itself, so far as it goes, and no doubt it will be the only part consulted by some students. The attempt to present a consecutive account of the develop- ment of the form on which so many classics in the history of embryology have been based is no slight undertaking. The author can hardly hope that he has avoided omissions and errors, and he will be sincerely grateful to those who call such to his attention. CONTENTS INTRODUCTION PAGE I. The Cell Theory . 1 II. The Recapitulation Theory 3 III. The Physiology of Development 6 IV. Embryonic Primordia and the Law of Genetic Restric- tion 8 V. General Characters of Germ-cells 9 The Sperynatozoon 9 The Ovum 10 Comparison of the Germ-cells 12 YI. Polarity and Organization of the Ovum .... 14 PART I THE EARLY DEVELOPMENT TO THE END OF THE THIRD DAY CHAPTER I. THE EGG . 17 Chemical Composition of the Hen's Egg 20 Formation of the Egg 21 Abnormal Eggs 26 Ovogenesis . 26 CHAPTER II. THE DEVELOPMENT PRIOR TO LAYING 32 I. Maturation 32 II. Fertilization 35 HI. Cleavage of the Ovum 38 The Hen's Egg 39 The Pigeon's Egg 43 IV. Origin of the Periblastic Nuclei, Formation of the Germ-wall 47 V. Origin of the Ectoderm and Entoderm 52 CHAPTER III. OUTLINE OF DEVELOPMENT, ORIENTA- TION, CHRONOLOGY 61 Orientation 63 Chronology {Classification of Stages) 64 Tables of the Development of the Chick 68 V / 7 ^ ^f VI CONTEXTS PAGE CHAPTER IV. FROM LAYIXG TO THE FORMATION OF THE FIRST SOMITE 69 I. Structure of the Unincubated Blastoderm .... 69 II. The Primitive Streak 69 Total Views 69 Sections 74 The Head-process 80 Interpretation of the Primitive Streak 83 III. The Mesoderm of the Opaque Area 86 IV. The Germ-wall 90 CHAPTER V. HEAD-FOLD TO TWELVE SOMITES (From about the twenty-first to the thirty-third hour of incu- bation) • . . . 91 I. Origin of the Head-fold 91 II. Formation of the Fore-gut 93 III. Origin of the Neural Tube 95 The Medullary Plate 95 The Neural Groove arid Folds 97 Primary Divisions of the Neural Tube 10') Origin of the Primary Divisions of the Embryonic Brain 108 IV. The Mesoblast 109 Primary Structure of the Somites 114 The Nephrotome, or Intermediate Cell-mass (Middle Plate) 114 The Lateral Plate 115 Development of the Body-cavity or Coelome 115 Mesoblast of the Head 116 Vascular System 117 Origin of the Heart 119 The Embryonic Blood-vessels 121 V. Description of an Embryo with 10 Somites .... 122 The Nervous System 124 Alimentary Canal 126 Vascular System 126 General 127 Zones of the Blastoderm 127 CHAPTER VI. FROM TWELVE TO THIRTY-SIX SOMITES. THIRTY-FOUR TO SEVENTY-TWO HOURS . 130 I. Development of the External Form, and Turning of THE Embryo 1,30 Separation of the Embryo from the Blastoderm . . , 130 CONTEXTS vii PAGE The Turning of the Embryo and the Embryonic Flexures 133 II. Origin of the Embryonic Membranes 135 Origin of the Amnion and Chorion 135 The Yolk-sac 143 Origin of the Allantois 143 Summary of Later History of the Embryonic Membranes . 145 III. The Nervous System 147 The Brain 147 The Neural Crest and the Cranial and Spinal Ganglia 156 IV. The Organs of Special Sense (Eye, Ear, Nose) . 164 The Eye 164 The Auditory Sac 168 The Nose (Olfactory Pits) 169 V. The Alimentary Canal and its Appendages . . . 170 The Stomod(Pum 173 The Pharynx and Visceral Arches 173 CEsophagus and Stomach 179 The Liver 179 The Pancreas 181 The Mid-Gut 181 Anal Plate, Hind-gut, Post-anal gut and Allantois 182 VI. History of the Mesoderm 183 Somites 183 The Intermediate Cell-mass 190 The Vascular System 197 VII. The Body-cavity and Mesenteries . 205 PART II THE FOURTH DAY TO HATCHING, ORGANOGENY, DEVELOPMENT OF THE ORGANS CHAPTER VII. THE EXTERNAL FORM OF THE EM- BRYO AND THE EMBRYONIC MEMBRANES 211 I. The External Form 211 General. . 211 Head 213 11. Embryonic Membranes 216 General 216 The Allantois 220 , The Yolk-sac , 225 The Amnion 231 Hatching 232 viii CONTEXTS PAGE CHAPTER VIII. THE NERVOUS SYSTEM 233 I. The Neuroblasts 233 The Medullary Neuroblasts 233 The Ganglionic Neuroblasts 236 II. The Development of the Spinal Cord 239 Central Canal and Fissures of the Cord 242 Neuroblasts, Commissures, and Fiber Tracts of the Cord . 244 III. The Development of the Brain 244 The Telencephaloti 245 The Die?icephalo)i 249 The Mesencephalon 251 The Metencephalon 251 The Myelencephalon 252 Commissures of the Brain 252 IV. The Peripheral Nervous System 252 The Spinal Nerves 252 The Cranial Nerves 2G1 CHAPTER IX. ORGANS OF SPECIAL SENSE . . . .271 I. The Eye 271 The Optic Cup 271 The Vitreous Humor 275 The Lens 276 A?iterior Chamber and Cornea 278 The Choroid and Sclerotic Coats 279 The Eyelids and Conjunctival Sac 279 Choroid Fissure, Pecten and Optic Nerve 281 11. The Development of the Olfactory Organ . . . 285 III. The Development of the Ear 288 Development of the Otocyst and Associated Parts . . . 289 The Development of the Tubo-tympanic Cavity, External Auditory Meatus and Tympanum 297 CHAPTER X. THE ALIMENTARY TRACT AND ITS AP- PENDAGES 301 I. Mouth and Oral Cavity 301 Beak and Egg-tooth 302 The Tongue 305 Oral Glands 306 II. Derivatives of the Embryonic Pharynx 306 Fate of the Visceral Clefts 307 Thyroid 307 CONTEXTS ix PAGE Visceral Pouches . 307 The Thymus 308 Epithelial Vestiges 309 The Postbranchial Bodies 309 III. The (Esophagus, Stomach and Intestine .... 309 Oesophagus 312 Stomach 313 Large Intestine, Cloaca, and Anus 314 IV. The Development of the Liver and Pancreas . . . 319 The Liver 319 The Pancreas 323 V. The Respiratory Tract 325 Bronchi, Lungs and Air-sacs 325 The Laryngotracheal Groove 331 CHAPTER XI. THE BODY-CAVITIES, MESENTERIES AND SEPTUM TRAXSVERSUM 333 I. The Separation of the Pericardial and Pleuroperi- TONEAL Cavities 333 Septum Transversiim 334 Closure of the Dorsal Opening of the Pericardium . . .337 Establishment of Independent Pericardial Walls . . . 338 Derivatives of the Septum Transversum ..... 339 II. Separation of Pleural and Peritoneal Cavities; Or- igin OF the Septum Pleuro-peritoneale . . . 340 III. The Mesenteries 342 The Dorsal Mesentery 342 The Origin of the Omentum 343 Origin of the Spleen 345 CHAPTER XII. THE LATER DEVELOPMEXT OF THE VASCULAR SYSTEM 348 I. The Heart 348 The Development of the External Form of the Heart . . 348 Division of the Cavities of the Heart 350 Fate of the Bulbus 357 The Sinus Venosus 357 II. The Arterial System 358 The Aortic Arches 358 The Carotid Arch : 361 The Subclavian A rtery 362 The Aortic System 362 X CONTENTS PAGE III. The Venous System 363 The Anterior Vena Cavce 363 The Omphalomesenteric Veins 364 The Umbilical Veins 367 The System of the Inferior Vena Cava 368 IV. The ExMBRyonic Circulation 372 CHAPTER XIII. THE URINOGENITAL SYSTEM ... 378 I. The Later History of the Mesonephros 378 II. The Development of the Metanephros or Permanent Kidney 384 The Metanephric Diverticulum 384 The Nephrogenous Tissue of the Metanephros . . . 387 III. The Organs of Reproduction 390 Development of Ovary and Testis 391 Development of the Genital Ducts 401 IV. The Suprarenal Capsules 403 Origin of the Cortical Cords 405 Origin of the Medullary Cords 406 CHAPTER XIV. THE SKELETON 407 I. General 407 II. The Vertebral Column 411 The Sclerotomes and Vertebral Sgmentation .... 412 Membranous Stage of the Vertebrce 414 Choyidrification 418 Atlas and Axis {Epistropheus) 420 Formation of Vertebral Articulations 421 Ossification 4-1 III. Development of the Ribs and Sternal Apparatus. . 424 IV. Development of the Skull 427 Development of the Cartilaginous or Primordial Cranium . 428 Ossification of the Skull 43- V. Appendicular Skeleton 434 The Fore-limb 434 The Skeleton of the Hind-limb 438 APPENDIX General Literature • « .... 443 Literature — Chapter I 443 Literature — Chapter II 444 Literature — Chapter III 44o Literature — Chapters IV and V 44a CONTENTS xi PAGE Literature — Chapter VII 447 Literature — Chapter VIII 449 Literature — Chapter IX 450 Literature — Chapter X 453 Literature — Chapter XI 457 Literature — Chapter XII 458 Literature — Chapter XIII 459 Literature — Chapter XIV 461 Index 465 THE DEVELOPMENT OF THE CHICK INTRODUCTION I. The Cell Theory The fundamental basis of the general conceptions of embry- ology, as of other biological disciplines, is the cell theory. The organism is composed of innumerable vital units, the cells, each of which has its independent life. The hfe of the organism as a whole is a product of the combined activity of all the cells. New cells arise always by subdivision of pre-existing cells, and new generations of the organism from liberated cells of the parental body. The protozoa, however, have the grade of organization of single cells, and the daughter-cells arising by fission constitute at the same time new generations. In some metazoa new gen- erations may arise asexually by a process of budding, as in Hydra, or of fission, as in some Turbellaria; such cases constitute excep- tions to the rule that new generations arise from liberated cells of the parental body, but the rule holds without exception for all cases of sexual reproduction. The body consists of various functional parts or organs; each of these again consists of various tissues, and the tissues are com- posed of specific kinds of cells. The reproductive organs, or gonads, are characterized by the production of germ-cells, ova in the female gonad or ovary, and spermatozoa in the male gonad or testis. However large the ovum may be, and in the hen it is the part of the egg known as the yolk, it is, nevertheless, a single cell at the time that it leaves the ovary in all animals. Similarly the spermatozoon is a single cell. An ovum and sper- matozoon unite, in the manner to be described later, and con- stitute a single cell by fusion, the fertilized ovum or oosperm. This cell divides and forms two; each of the daughter-cells divides, making four, and the number of cells steadily increases by suc- cessive divisions of all daughter-cells, so that a large number of cells is rapidly produced. Organs are formed by successive 1 2 THE DEVELOPMENT OF THE CHICK and orderly differentiation among groups of these cells. Among these organs are the gonads, consisting of cells which trace a continuous lineage by cell-division back to the fertilized ovum, and which are capable of developing into ova or spermatozoa according to the sex of the individual. The lives of successive generations are thus continuous because the series of germ-cells from which they arise shows no break in continuity. All other kinds of cells composing the body finally die. In view of this contrast the non-germinal cells of the body are known collectively as somatic cells. In some way the germ- cells of a species maintain very constant properties from gen- eration to generation in spite of their enormous multiplication, and this furnishes the basis for hereditary resemblance. The establishment of the fact that in all animals the ovum is a single cell, and that the cells of all tissues of the body are derived from it by a continuous process of cell-division, completes the outline of the cycle of the generations, and furnishes the basis for a complete theory of development. The full significance of this principle can only be appreciated by learning the condition of embryology before the establishment of the cell-theory in the eighteenth century. The history of our knowledge of the devel- opment of mammals is particularly instructive in this respect: some knowledge had been gained of the anatomy of the embryos, mostly relatively advanced, of a few mammals; but the origin of the embryo was entirely unknown; the ovum itself had not been discovered; the process of fertilization was not understood. In the knowledge of the cycle of generations there was a great gap, and the embryo was as much a mystery as if it had arisen by a direct act of creation. To be sure Harvey in 1651 had propounded the theorem, omne vivum ex ovo, but no one had ever seen the egg of a mammal, and there was no clear idea in the case of other forms what the egg signified. In 1672, de Graaf (who died in 1673 at the age of 32) published a work, "de mulierum organis generationis inservientibus," in which he attempted to show that the vesicles seen on the surface of the ovaries were the female reproductive material. But he could not reconcile the view that the Graafian follicle is the mjim- malian egg with the fact that the earliest embryos discovered by him were smaller than the follicles. For this reason his views were opposed by Leeuwenhoek and \^alisnieri; and the later re- INTRODUCTION 3 searches of Haller and his pupil Kuhlemann seemed to estabHsh a view which banished all possibility of a rational explanation of development, viz., that, in the highest group of animals (the mammalia) the embryo arose after fertilization out of formless fluids. In 1827 V. Baer discovered the mammalian ovum within the Graafian follicle. But no correct interpretation of this discovery was possible until the establishment of the cell-theory by Theo- dore Schwann in 1839; Schwann concluded as the result of his investigations that there was one general principle for the forma- tion of all organisms, namely, the formation of cells; that "the cause of nutrition and growth resides not in the organism as a whole, but in the separate elementary parts, the cells.'' He recognized the ovum as a single cell and the germinal vesicle as its nucleus. But on account of his erroneous conception of the origin of cells as a kind of crystallization in a primordial sub- stance, the cytoblastema, he was unable to form the conception of continuity of generations which is an essential part of the modern cell-theory. Schwann's theory as regards the ovum was not at once ac- cepted. Indeed, for a period of about twenty years some of the best investigators, notably Bischoff, opposed the view that the ovum is a single cell, and the so-called germinal vesicle its nucleus. It was not, indeed, until 1861 that Gegenbaur deci- sively demonstrated that the bird's ovum is a single cell. Even after that it was maintained for a long time by His and his fol- lowers that all the cells were not derived from the ovum directly, but that certain tissues, notably the blood and connective tissues, w^ere to be traced to maternal leucocytes that had migrated into the ovum while it was yet in the follicle. This view was decisively disproved in the course of time. II. The Recapitulation Theory Haeckel's formula, that the development of the individual repeats briefly the evolution of the species, or that ontogeny is a brief recapitulation of phylogeny, has been widely accepted by embryologists. It is based on a comparison between the embry- onic development of the individual and the comparative anatomy of the phylum. The embryonic conditions of any set of organs of a higher species of a phylum resemble, in many essential par- 4 THE DEVELOPMENT OF THE CHICK ticulars, conditions that are adult in lower species of the same phylum; and, moreover, the order of embryonic development of organs corresponds in general to the taxonomic order of organ- ization of the same organs. As the taxonomic order is the order of evolution, Haeckel's generalization, which he called the funda- mental law of biogenesis, would appear to follow of necessity. But it never happens that the embryo of any definite species resembles in its entirety the adult of a lower species, nor even the embryo of a lower species; its organization is specific at all stages from the ovum on, so that it is possible without any diffi- culty to recognize the order of animals to which a given embryo belongs, and more careful examination will usually enable one to assign its zoological position very closely. If phylogeny be understood to be the succession of adult forms in the line of evolution, it cannot be said in any real sense that ontogeny is a brief recapitulation of phylogeny, for the embryo of a higher form is never like the adult of a lower form, though the anatomy of embryonic organs of higher species re- sembles in many particulars the anatomy of the homologous organs of the adult of the lower species. However, if we conceive that the whole life history is necessary for the definition of a species, we obtain a different basis for the recapitulation theory. The comparable units are then entire ontogenies, and these re- semble one another in proportion to the nearness of relationship, just as the definitive structures do. The ontogeny is inherited no less than the adult characteristics, and is subject to precisely the same laws of modification and variation. Thus in nearly related species the ontogenies are very similar; in more distantly related species there is less resemblance, and in species from different classes the ontogenies are widely divergent in many respects. In species of lower grades of organization the ontogenetic series is a shorter one than in species of higher grades, so that the final stages of the organs of a lower species become inter- mediate or embryonic stages in species of higher rank. But the stage of the lower species does not appear in all the organs of the higher species simultaneously. Thus the chick never exhibits the grade of organization of a fish throughout; while its pharynx, for instance, is in a fish-like condition with reference to arches and clefts, the nervous system is relatively undifferentiated, and IXTRODUCTIOX 5 it has no vertebrae; on the other hand, it has a heart of an am- phibian rather than of a fish type. Some of these considerations may be represented graphically as follows : let us take a species D that has an ontogeny A, B, C, D, and suppose that this species evolves successively into species E, F, G, H, etc. When evolution has progressed a step, to E, the characters of the species established develop directly from the ovum, and are therefore, in some way, involved in the com- position of the latter. All of the stages of the ontogeny leading up to E are modified, and we can indicate this in the ontogeny 1. A B C D of E as in line 2; similarly, when evolu- 2. A^ B^ C^ D^ E tion has progressed to species F, seeing 3. A^ B^ C^ D^ E^ F that the characters of F now develop 4. A^ B^ C^ D^ E^ F^ G directly from the ovum, all the onto- 5. A'' B'' C^ D" E^ F^ G^ H genetic stages leading up to F are modi- fied, line 3. And so on for each successive advance in evolution, lines 4 and 5. It will also be noticed that the terminal stage D of species 1, becomes a successively earlier ontogenetic stage of species 2, 3, 4, 5, etc., and moreover it does not recur in its pure form, but in the form D^ in species 2, D^ in species 3, etc. Now if the last five stages of the ontogeny of species 5 be examined, viz., D"*, E^, F^, G^, H, it will be seen that they repeat the ph3'logeny of the adult stages D, E, F, G, H, but in a modified form. This is in fact what the diagram shows; but it is an essential defect of the diagram that it is incapable of showing the character of the modifications of the ancestral conditions. Not only is each stage of the ancestral ontogenies modified with each phylogenetic advance, but the elements of organization of the ancestral stages are also dispersed so that no ancestral stage hangs together as a unit. The embryonic stages show as much proportional modi- fication in the course of evolution as the adult, but this is not so obvious owing to the simpler and more generalized character of the embryonic stages. The recapitulation theory as outlined above is obviously a corollary of the theory of organic descent; it was in fact developed in essentially its present form, soon after the publication of the "Origin of Species," by Fritz Miiller and Ernst Haeckel. But the data on which it was based were known to the earlier embry- ologists; and ^leckel, for instance, insisted very strongly on the resemblance between the ontogenetic and the taxonomic series 6 THE DEVELOPMENT OF THE CHICK (1821). V. Baer opposed Meckel's view that higher organisms pass through the definitive stages of the lower organisms, and formulated his conclusions on the subject in 1828 in the following theses : 1. ''The more general features of a large division of animals arise in the embryo earlier than the more special features." 2. " From the most general features of structure arise those that are less general, and so on until the most specific features arise." 3. "The embryo of any definite species tends away from the specific forms of other species instead of passing through them." 4. "Fundamentally, therefore, the embryo of any higher species is never like a lower species, but only like its embryo." Some embryologists profess to prefer the laws of v. Baer to the recapitulation theory as a formulation of the actual facts. But it is obvious that the only possible explanation of the facts is found in the theory of descent, and that therefore they must be formulated in terms of this theory. The method of formula- tion will depend on the conception of the nature of the factors of organic evolution. Haeckel stated his theory in Lamarckian terms, which renders it inacceptable in many places to those who cannot accept the Lamarckian point of view. But as the basis of any theory of descent is heredity, and it must be recog- nized that ontogenies are inherited, the resemblance between the individual history and the phylogenetic history necessarily fol- lows. If one holds, as does the present writer, that phylogenetic variations are germinal in their character, then one must admit that every phase of development of every part has two aspects, viz.: the modern, specific, or coenogenetic, and the ancestral or palingenetic aspect. The latter aspect may be more or less com- pletely obscured in the course of evolution, but it can never entirely vanish because it is the original germ of the specific form acquired. It is not correct from this point of view to classify some features of development as coenogenetic and others as palin- genetic, though it is obvious that some characters may exhibit the ancestral conditions in more apparent and others in less apparent form. III. The Physiology of Development To explain how a germ possessed the potency of forming an adult, the jyreformationists of the eighteenth century assumed INTRODUCTION 7 that it contained a miniature adult, and that the process of development consisted essentially in enlargement and completion in detail of that which was already preformed. They solved the problem of development, therefore, by denying its existence: In the begininng the Creator had not only made all species of animals and plants in essentially their present forms, but had at the same time created the germs of all the generations that were ever to come into existence. The ovum of any species, therefore, contained encapsuled the germ of the next generation; this, likewise encapsuled, the germ of the generation next follow- ing, and so on to the predetermined end of the species. This was know^n as the doctrine of evolution or preformation. In opposition to this conception, those of the same period who be- lieved in epigenesis maintained the apparent simplicity of the germ to be real, and development to be actual. But, as there was no conception of the continuity of generations, the adherents of this point of view had to assume the spontaneous generation of the embryo. A great advance over the preformation theory of develop- ment was made in the modern theory of determinants. This conception, which forms the basis of Darwin's theory of pan- genesis as well as of Weismann's germ-plasm theory of develop- ment, is, essentially, that all the diverse components of the organism are represented in the germ by distinct entities (pangens of Darwin, determinants of Weismann) which are germs of the parts that they represent, and which are so distributed in the pro- cess of development that they produce all the parts of the embryo in their proper sequence and relations. This is not the place to enter into the numerous and diverse variations of the deter- minant hypothesis. It was an advance over the preformation theory of development in so far as it was reconcilable with the cell and protoplasm theories of organization, but it has a real relationship to the preformation theory inasmuch as it denies the simplicity of the germ and avoids any real explanation of the modus operandi of development. Development is as truly a physiological process as secretion, and as such is to be studied by similar methods, mainly experi- mental. The limits of pure observation without experiment are soon reached in the analysis of such a complex subject as the physiology of development; experiment then becomes necessary 8 THE DEVELOPMENT OF THE CHICK to push the analysis of the subject farther, and to furnish the true interpretation of the observations. In some cases experi- ments have confirmed the physiological deductions of pure ob- servation, and in many cases have decided between conflicting views. Not all embryological experiments, however, are essays in the direction of a physiology of development; some are directed to the solution of morphological problems, as, for instance, the origin of the sheath cells of nerves, or the order of origin of so- mites, or the relation of the primitive streak to the embryo. Experimental embryology is, therefore, not synonymous with physiology of development. Physiology of development must proceed from an investiga- tion of the composition and properties of the germ-cells. It must investigate the role of cell-division in development, the factors that determine the location, origin, and properties of the primordia of organs, the laws that determine unequal growth, the conditions that determine the direction of differentiation, the influence of extraorganic conditions on the formation of the embryo, and the effects of the intraorganic environment, i.e., of component parts of the embryo on other parts (correlative differentiation). Each of these divisions of the subject includes numerous problems, which have attracted many investigators, so that the materials for a consistent exposition of the physiology of embryonic development are being rapidly accumulated. This direction of investigation is, however, one of the youngest of the biological disciplines. It will be seen how far it is removed from attempts to explain embryonic development by a single principle. IV. Embryonic Primordia and the Law of Genetic Re- striction In the course of development the most general features of organization arise first, and those that are successively less general in the order of their specialization. For every structure, there- fore, there is a period of emergence from something more general. The earliest discernible germ of any part or organ may be called its primordium. In this sense the ovum is the primordium of the individual, the ectoderm the primordium of all ectodermal structures, the medullary plate the primordium of the central and part of the peripheral nervous system, the first thickening IXTRODUCTION ^ of the ectoderm over the optic cup the primordium of the lens, etc. Primordia are, therefore, of all grades, and each arises from a primordium of a higher grade of generality. The emergence of a primordium involves a limitation in two directions: (1) it is itself limited in a positive fashion by being restricted to a definite line of differentiation more special than the primordium from which it sprang, and (2) the latter is limited in a negative way by losing the capacity for producing another primordium of exactly the same sort. The advance of differen- tiation sets a limit in all cases, in the manners indicated, to sub- sequent differentiation, a principle that has been designated by Minot the law of genetic restriction. This law has not been sufficiently investigated in an experi- mental fashion to demonstrate its universal valicUty, but enough is known to establish its general applicability. A very impor- tant property of primordia in many animals is their capacity for subdivision, each part retaining the potencies of the whole. Thus, for instance, in some animals two or several embryos may be produced from parts of one ovum. Similarly two or more limbs may be produced in some forms by subdividing a limb- bud, etc. V. General Character of Germ-cells As already remarked the ovum and spermatozoon have the character of single cells in all animals. They are, however, specialized for the performance of their respective functions. The ovum is relatively large, inert, and usually rounded in form. Its size is due to the presence of a sufficient quantity of proto- plasm to serve as the primordium of an embryo, and of a greater or less amount of yolk for its nutrition. The spermatozoon, on the other hand, is relatively minute and capable of locomotion. It contains no food substances, and only sufficient protoplasm to serve as transmitter of paternal qualities and for organs of locomotion. The Spermatozoon. The spermatozo5n (Fig. 1) is an elon- gated flagellated cell in which three main divisions are distin- guished, viz., head (caput), neck (coUum) and tail (cauda). The head contains the nucleus, and the neck the centrosomes of the sperm mother-cell or spermatid. The tip of the head is often transformed into a perforatorium. Three parts may be recog- 10 THE DEVELOPMENT OF THE CHICK nized in the tail, viz., the connecting piece (pars conjiinctionis) next to the neck, the main piece (pars principalis) and the end- piece or terminal filament (pars terminalis). The entire tail is traversed by an axial filament; in the region of the connecting and main pieces the axial filament is surrounded by a protoplasmic sheath (involucrum) which may be variously modified in different animals. The end-piece is made up of the axial filament alone. The Ovum. The ova of difTerent phyla and classes of animals vary greatly in size, in or- ganization, and in the nature of their enve- lopes. In considering these variations we shall limit ourselves to the vertebrates. Within the ovary the ovum receives two envelopes, viz., a primary envelope, the so-called vitelline mem- brane, which is supposed to be secreted by the ovum itself, and a secondary or follicular mem- brane, which is secreted by the follicular cells. (See Chap. I). Theoretically the distinction be- tween vitelline membrane and follicular mem- brane (primary and secondary egg-membranes) is perfectly clear; but practically it is impossi- ble in most cases to make such a distinction. Therefore the membrane that surrounds the ovarian ovum will be termed the vitelline mem- brane or zona radiata without reference to its YiQ_ 1. Sperma- theoretical mode of origin. tozoon of the pig- The ovum escapes from the ovary (ovula- eon from the vas tion) by rupture of the wall of the follicle, and, deferens. (After -^^ ^^g^ vertebrates, is taken up by the oviduct Ballowitz.) through which it passes on its way to the ex- terior. Within the oviduct it may become surrounded by tertiary membranes secreted by the wall of the oviduct itself. Tertiary membranes are lacking in some vertebrates, in others they are of great importance. Thus in birds the albumen, the shell- membrane and the shell itself are tertiary membranes. The principal differences to be emphasized in the ova of ver- tebrates are, however, in the amount and arrangement of the yolk contained within the ovum proper. All ova contain more INTRODUCTION 11 or less yolk. In the case of mammals (excepting the monotre- mata: Ornithorhynchus, Echidna, etc., which have large ova) the yolk is scanty in amount, and quite uniformly distributed in the form of fine granules; the ovum is, therefore, relatively very small (mouse, 0.059 mm.; man, 0.17 mm.). Such ova are often termed alecithal, which means literally without yolk. In the literal sense, however, no ova are entirely alecithal, so that it will be better to use the term of Waldeyer, isolecithal. In the amphibia the yolk is much greater in amount and it is centered towards one pole of the ovum; the germinal vesicle (nucleus of the egg-cell), which occupies the center of the protoplasm of the ovum, is therefore displaced towards the opposite pole of the ovum. Such ova are termed telolecithal. In the ova of Selachia, reptiles and birds, the yolk is very much greater in amount and in consequence the protoplasm containing the germinal vesicle appears as a small disc, the germinal disc, on the surface of the huge yolk-mass. But no matter how large the ovum may become by deposi- tion of yolk, its unicellular character is not altered. The deposi- tion of yolk is simply a provision for the nutrition of the embryo. In the mammals the nutrition of the embryo is provided for by the placenta; therefore yolk may be dispensed with. In the absence of such provision the amount of yolk is a measure of the length of the embryonic period of development. In the amphibia, for instance, this is relatively brief, for the yolk is soon used up, and the larva must then depend on its own activities for its nutri- tion. Therefore the development involves a metamorphosis: the embryo is born in a very unfinished condition, as a larva (the tadpole in the case of amphibia) , which must undergo an exten- sive metamorphosis to reach the adult condition. In the reptiles and birds, however, the amount of yolk is sufficient to carry the development through to a juvenile condition, before an extrane- ous food-supply is necessary. The metamorphosis, therefore, which takes place in free life in amphibia, goes on within the egg in reptiles and birds. The first form of development is known as larval, the second as foetal. The amount and arrangement of yolk also influences very profoundly the form of the early stages of development. Ova are classified in this respect as holoblastic and meroblastic. Holo- blastic ova are those in which the process of cell division (cleav- 12 THE DEVELOPMENT OF THE CHICK age or segmentation of the ovum), with which development begins, involves the entire ovum. This occurs where the amount of the yolk is relatively small and where it is completely inter- penetrated by sufficient protoplasm to carry the planes of divi- sion through the inert yolk. But where the amount of yolk becomes very large, or where it is not interpenetrated sufficiently by the protoplasm, the division planes are confined to the proto- plasmic portion of the ovum, and the yolk remains undivided. Such ova are known as meroblastic. In these ova the cellular part of the ovum forms a blastodisc (germinal disc) on the surface of the yolk. The ova of Amphioxus, Petromyzontidse, Ganoi- dea. Dipnoi, Amphibia, Marsupialia, and Placentalia are holo- blastic; those of Myxinoidea, Teleostei, Selachia, ReptiUa, Aves, and Monotremata are meroblastic. It is obvious that transitional conditions between holoblastic and meroblastic ova may occur; such are in fact found among the ganoids. In Lepidosteus, for instance, the quantity of proto- plasm in the lower hemisphere is so slight that the division planes form with extreme slowness. On the other hand, it should be emphasized that the distinction between holoblastic and mero- blastic ova is not so much due to amount of yolk as to the defi- niteness of its separation from the protoplasm. Thus the ova of some teleosts, particularly of the viviparous forms described by Eigenmann, are many times smaller than the ova of Necturus or Cryptobranchus among amphibia. Yet the teleost ovum is meroblastic, because the protoplasm does not penetrate suffi- ciently into the yolk, and the amphibian ovum is holoblastic. Comparison of the Germ-cells. Although it is not within the province of this book to enter fully into a discussion of this ques- tion, yet it should be pointed out that, in spite of the extreme differences in the structure of the germ-cells, they are exactly equivalent in hereditary potency, as is proved by the similar nature of reciprocal crosses. Their resemblances are in fact fundamental and their differences must be regarded as adapta- tions to secure their union. The comparative history of the germ-cells, that is a comparison of ovogenesis and spermato- genesis, brings out their fundamental similarity as germ-cells. In both the ovogenesis and spermatogenesis three periods are clearly distinguishable, viz. : a period of multiplication, a period of growth, and a period of maturation. In the period of multiplication INTRODUCTION 13 the primordial germ-cells, known as ovogonia and spermatogonia are very similar in their morphological characters; both kinds are small, yolkless cells containing the typical or somatic number of chromosomes; they multiply rapidly by karyokinetic division. At the end of this period multiplication ceases and the germ- cells increase in size (period of growth). They are now known as ovocytes and spermatocytes of the first generation. The growth of the ovocyte is much greater than that of the sperma- tocyte; deposition of yolk occurs in the ovocyte during this period; and in some animals a small quantity of yolk-granules may be found even in the spermatocytes. Another character- teristic feature of the period of grow^th is the reduction of the number of chromosomes to one half of the typical number, which takes place, according to the current conception, by union of the chromosomes in pairs (synapsis) forming one half of the somatic number of chromosomes, which are, however, bivalent and are know^n as tetrads. At the end of the period of grow^th the ovocyte of the first generation is usually many times larger than the spermatocyte, owing mainly to the amount of yolk formed. But the tw^o kinds of cells are precisely alike in nuclear constitution. Then comes the period of maturation, w^hich is the same in both kinds of cells with reference to the nuclear phenomena, but very different as regards the behavior of the cell-body. The maturation consists of two rapidly succeeding karyokinetic divisions: in the case of the spermatocyte the first division results in the formation of two similar cells, the spermatocytes of the second order, and the second maturation division divides each of these equally, forming two similar spermatids, so that four equal and similar spermatids arise from each spermatocyte of the first order. Each spermatid then differentiates into a single spermatozoon. In the case of the ovocyte of the first order, the first maturation division is exceedingly unequal; the smaller cell is known as the first polar body, but both cells are ovocytes of the second order. The second maturation division usually involves only the large secondary ovocyte; it is as unequal as the first division and results in the formation of a second polar body. The division of the first polar body, where it occurs, is equal. Thus the net result of the matu- ration division of the ovum is the production of three cells (four if the first polar body divides), viz., the two (or three) polar bodies 14 THE DEVELOPMENT OF THE CHICK and the ovum. The size of the polar globules is usually so small that their elimination makes no appreciable difference in the size of the ovum proper, but they have, nevertheless, the same nuclear constitution as the ovum. The mature ovum (ootid) and the polar bodies are the precise equivalent of the four spermatids, but whereas each of the latter becomes a functional spermatozoon, only the ovum on the female side is functional; the polar bodies lack the necessary protoplasm and yolk for development, and they therefore die. The polar bodies must be regarded as abortive ova; and a teleological ex- planation of the form of maturation of the ovum is afforded by the consideration that equal maturation divisions would reduce the amount of protoplasm and yolk in the products below the minimum desirable for perfect development. Although the maturation divisions of the ovum and sperma- tozoon are so dissimilar externally, yet the nuclear phenomena are exactly alike. The net result of the maturation divisions is to produce definitive germ-cells containing one half of the somatic number of chromosomes owing to the reduction by pairing (syn- apsis) that occurs in both at the beginning of the period of growth. The somatic number is again restored when the sperm-nucleus and the egg-nucleus unite in fertilization. Questions of funda- mental importance for the problems of heredity arise in connec- tion w4th the phenomena of maturation and fertilization, but their consideration lies without the scope of the present book. VI. Polarity and Organization of the Ovum Although the ovum is morphologically a single cell, yet, as the primordium of an individual, it has certain specific properties that predelineate or foreshadow the main structural features of the embryo. Polarity is the most general of these features: all the axes of the ovum are not similar, though they may be equal; there is one axis around which the development centers ; the ends of this axis are known as the animal and the vegetative poles of the ovum, and the hemispheres in which they lie are named correspondingly. In telolecithal ova the yolk is centered in the vegetative hemisphere, the protoplasm in the animal hemisphere; even in ova which are called isolecithal there is a tendency for the yolk to be more abundant in the vegetative hemisphere. The polar globules are formed at the animal pole; hence their IXTRODUCTIOX 15 name; they often furnish the only clear indication of polarity before cleavage begins. With reference to the heteropolar ovic axis a series of meridia may be defined, drawn from pole to pole over the surface; likewise an equator and a series of horizontal zones parallel to the equator. Thus directions on the surface of the ovum may be defined as meridional, equatorial, or oblique. Cleavage takes place with reference to the axis of the ovum. Thus in holoblastic vertebrate ova the first and second cleavage planes are meridional, and the third usually equatorial. The mammalian ovum may form an exception to this rule, though little is known, as a matter of fact, about the polarity of the mam- malian ovum. The cleavage of meroblastic ova takes place likewise with reference to the polarity (see Chap. II); and the location of the primary germ-layers is determined by the polarity. Not only is the ovum heteropolar, but in many bilateral animals, and perhaps in all, it is bilaterally symmetrical before cleavage begins; that is to say, one of the meridional planes defines the longitudinal axis of the future embryo, and the direc- tion of anterior and posterior ends is also predetermined in this meridian, so that halves of the egg corresponding to future right and left sides of the embryo may be distinguished. In the frog's egg the plane of symmetry is marked by a gray crescent that appears above the equator on the side of the egg that corresponds to the hinder end of the embrvo. This crescent is bisected bv the meridional plane of symmetry. In the hen's egg the plane of sym.metry of the em]:>ryo appears on the surface of the yolk in a line at right angles to the axis of the shell, and the left side of the embryo is turned towards the broad end, the right side towards the narrow end of the shell. The same plane of sym- metry must exist in the ovum prior to cleavage for reasons ex- plained beyond, although there is no morphological differentiation in the ovum proper, i.e., the germinal disc or yolk, that indicates it. This predelineation of embryonic areas within the unseg- mented ovum has led to the idea that the ovum contains various materials, so-called formative stuffs, in typical arrangement, that determine in some physiological way the formation of specific structures. PART I THE EARLY DEVELOPMENT TO THE END OF THE THIRD DAY CHAPTER I THE EGG The parts of a newly laid hen's egg are the shell, shell-mem- brane, albumen, and yolk. In an egg that has been undisturbed for a short time the yolk floats in the albumen with a whitish disc, the blastoderm about 4 mm. in diameter, on its upper sur- face. If the yolk be rotated, it will return to its former position in a few minutes, owing to the slightly lower specific gravity of the hemisphere containing the blastoderm. The blastoderm is the living part of the egg, from which the embryo and all its membranes are derived. It is already in a fairly advanced stage of development when the egg is laid. The yolk and blastoderm are enclosed within a delicate transparent membrane (vitelline membrane) which holds the fluid yolk-mass together. We may now consider some details of the structure and composition of the parts of the egg. The shell is composed of three layers: (1) the inner or mam- millary layer, (2) the intermediate spongy layer, and (3) the surface cuticle. The mammillary layer consists of minute cal- careous particles about 0.01-0.015 mm. in diameter welded to- gether, with conical faces impinging on the shell-membrane; the minute air-spaces between the conical inner ends of the mammillae communicate with the meshes of the spongy layer, which is sev- eral times as thick, and which is bounded externally by the ex- tremely delicate shell cuticle. The spongy layer consists of matted calcareous strands. The shell cuticle is porous, but apparently quite structureless otherwise. The cuticular pores communicate with the mesh-work of the spongy layer; thus the entire shell is permeable to gases, and permits of embryonic respiration, and evaporation of water. 17 18 THE DEVELOPMENT OF THE CHICK The shell-membrane consists of two layers, a thick outer layer next to the shell and a thinner one next the albumen. Both are composed of matted organic fibers (more delicate in the inner than in the outer layer), crossing one another in all directions. At the blunt end of the egg the two layers are separated and form a chamber containing air that enters after the egg is laid (Fig. 2). The physical characteristics of the albumen are too well known to require description. A dense layer immediately next BI ML. ■hi^. VA/. Fig. 2. — Diagram of the hen's e^g in section to show relations of the parts. A. C, Air chamber. Alb., Albumen. Bl., Blastoderm. Chal., Chalaza. I. S. M., Inner layer of the shell membrane. L., Latebra. N. L., Neck of Latebra. N. P., Nucleus of Pander. O. S. M., Outer shell membrane, p' v. s., Perivitelline space. S., Shell. V. M., Vitelline membrane. W. Y., White yolk. Y. Y., Yellow yolk. to the vitelline membrane is prolonged in the form of two spirally coiled opalescent cords towards the blunt and narrow ends of the egg respectively; these are the chalazae, so called from a fanciful resemblance to hail stones. The two chalazae are twisted in opposite directions. In a hard-boiled egg it is possible to strip off the albumen in concentric spiral layers from left to right from the broad to the small end of the egg. THE EGG 19 The 3'olk and blastoderm are enclosed within the delicate vitelline membrane; the yolk is a highly nutritious food destined to be gradually digested and absorbed by the living cells of the blastoderm and used for the growth of the embryo. It is not of uniform composition throughout, but consists of two main ingredients known as the yellow and the white yolk. The yellow yolk makes up the greater part of the yolk-sphere; the main part of the white yolk is a flask-shaped mass, the bulb of which, known as the latebra, is situated near the center of the whole volk, the neck rising- towards the surface and expanding in the form of a disc (nucleus of Pander) situated imme- diately beneath the blastoderm (Fig. 2) ; at its margin this disc is continuous with a thin peri- pheral layer of white yolk that surrounds the entire mass. In addition there are several thin concentric layers of white yolk concentric to the inner bulb-shaped mass.^ If an egg be opened, a delicate hair inserted in the blastoderm to mark its position, and then boiled hard, a sec- tion through the hair and center of the yolk will show the above relations quite clearly. The white yolk does not coagulate so readily as the yellow yolk, and it may be distinguished by this property as well as by its lighter color. Both kinds of yolk are made up of innumer- able spheres which are, however, quite different in each (Fig. 3). Those of the yellow yolk are on the whole larger than those of the white yolk (about 0.025-0.100 mm. in diameter) with extremely fine granular contents. There is no fluid between the spheres. Those of the white yolk are smaller and more variable in size, ranging from the finest granules up to B Fig. 3. — Y o 1 k - spheres of the hen's egg; highly magnified. (After Foster and Bal- four.) A. Varieties of white yolk-spheres. B. Yellow yolk- sphere. ^ The assertion that the thin layers that define the concentric stratifica- tion of the yellow yolk are of the nature of white yolk is traceable to Meckel V. Hemsbach, Leuckart, and Allen Thomson. His was not able to satisfy himself that the characteristic elements of the white yolk occur within these thin concentric lamellae (Untersuchungen ueber die erste Anlage des Wir- beltierleibes, p. 2). 20 THE DEVELOPMENT OF THE CHICK about 0.07 mm. The larger spheres of the white yolk contain several highly refractive granules of relatively considerable size as compared with those of the yellow spheres (Fig. 3), and such granules may have secondary inclusions. As we shall see later, the smaller granules of the white yolk extend into the germinal disc (forerunner of the blastoderm) and grade into minute yolk- granules contained within the living protoplasm. The earlier investigators from the time of Schwann regarded the white yolk-spheres as actual cells (Schwann, Reichert, Coste, His). His especially laid great stress on this interpretation; he believed that they were derived from the cells of the ovarian follicle which migrated into the ovum in the course of ovogenesis, that they multiplied like other cells, and took part in the formation of certain embryonic tissues. Sub- sequently he abandoned this position as untenable. The white yolk spheres are now universally regarded as food matters of a particular sort. The yolk and albumen are complex mixtures of many difTerent substances, organic and inorganic, containing all the elements necessary for the growth of the embryo. Very little is known concerning the series of chemical changes that go on in them during incubation. Chemical Composition of the Hen's Egg. — The following data on the chemical composition of the hen's egg are taken from Simon's Physiological Chemistry. For details and literature the student is referred to the standard text-books of jDhysiological chemistry. GENERAL COMPOSITION OF THE YOLK PER CENT. Water 47.19-51.49 Solids 48.51-42.81 Fats (olein, pahnitin, and stearin) 21.30-22.84 Vitelline and other albumens 15.63-15.76 Lecithin 8.43-10.72 Cholesterin 0.44- 1.75 Cerebrin 0.30 Mineral salts 3.33- 0.36 Colorine: matters 1 r> --o ^1 \ O.oo3 Glucose J Analysis of the Mineral Salts Sodium (NasO) 5.12- 6.57 Potassium (K/J) 8.05- 8.93 Calcium (CaO) 12.21-13.28 THE EGG 21 PER CENT. Magnesium (MgO) 2.07- 2.11 Iron (Fe.Oa) 1.19- 1.45 Phosphoric acid, free (P2O5) 5.72 Phosphoric acid, combined 63.81-66.70 Sihcic acid 0.55- 1.40 Chlorine Traces. GENERAL COMPOSITION OF THE ALBUMEN Water 80.00-86.68 SoHds 13.22-20.00 Albumens 11.50-12.27 Extractives 0.38- 0.77 Glucose 0.10- 0.50 Fats and Soaps Traces Mineral salts 0.30- 0.66 Lecithins and Cholesterin Traces. Analysis of the Mineral Ash Sodium (Na.O) 23.56-32.93 Potassium (K2O) 27.66-28.45 Calcium (CaO) 1.74- 2.90 Magnesium (MgO) 1.60- 3.17 Iron (FeoOa) 0.44- 0.55 Chlorine (CI) 23.84-28.56 Phosphoric acid (P2O5) 3.16- 4.83 Carbonic acid (CO2) 9.67-11.60 Sulphuric acid (SO3) 1.32- 2.63 Sihcic acid (SiO.) 0.28- 0.49 Fluorine (Fl) Traces. The shell consists of an organic matrix of the nature of keratin impregnated with lime salts: calcium and magnesium carbonates about 97 %, calcium and magnesium phosphates about 1 %, keratin and water about 2 %, trace of iron. The shell-membrane and the vitelline membrane are stated to consist of keratin or a closely allied substance. Formation of the Egg. The organs of reproduction of the hen are the ovary and oviduct of the left side of the body. Al- though the right ovary and oviduct are formed in the embryo at the same time as those of the left side, they degenerate more or less completely in the course of development (see Chap. XIII), so that only functionless rudiments remain. This would appear to be correlated with the large size of the egg and the delicate 22 THE DEVELOPMENT OF THE CHICK Fig. 4. — Reproductive organs of the hen. (After Duval, based on a figure by Coste.) The figure is diagrammatic in one respect, namely, that two THE EGG 23 nature of the shell, as there is not room for two eggs side by side in the lower part of the body-cavity. The ovary lies at the anterior end of the kidney attached by a fold of the peritoneum (mesovarium) to the dorsal wall of the body-cavity. In a laying hen ova of all sizes are found from microscopic up to the fully formed ovum ready to escape from the follicle. Such an ovary is shown in Figure 4; the gradation in size of the ova will be noticed up to the one fully formed and ready to burst from its capsule. At 5 in this figure is shown a ruptured follicle, and the ovum which has escaped from this follicle is shown in the oviduct at 8. It will be seen that the part of the definitive hen's egg produced in the ovary is the so-called yolk. The blood-supply of the very vascular ovary is derived from the dorsal aorta, and the veins open into the vena cava inferior. The oviduct is a large coiled tube (Fig. 4) which begins in a wide mouth with fringed borders, the ostium tubce abdominale (funnel or infundibulum) opening into the body-cavity near the ovary. It is attached by a special mesentery to the dorsal wall of the body-cavity, and opens into the cloaca. The following divisions are usually distinguished: (1) the oviduct s. s., (2) the uterus, (3) the vagina (Fig. 4). The oviduct includes the entire tube from the funnel to the dilated uterus. The vagina is the short terminal portion opening into the cloaca (Figs. 4 and 5). In the oviduct proper we distinguish the funnel, the main glandu- lar part, and the isthmus. The formation of an egg takes place as follows: the yolk, or ovum proper, escapes by rupture of the follicle along a preformed band, the stigma (Fig. 4-4), into the infundibulum which swallows it, so to speak, and it is passed down by peristaltic contractions ova are shown in the oviduct at different levels; normally but one ovum is found in the oviduct at a time. 1, Ovary; region of young follicles. 2 and 3, Successively larger follicles. 4, Stigmata, or non-vascular areas, along which the rupture of the follicle takes place. 5, Empty follicle. 6, Cephalic lip of ostium. 7, Funnel of oviduct (ostium tuba? abdominale). 8, Ovum in the upper part of the ovi- duct. 9, Region of the oviduct in which the albumen is secreted. 10, Albu- men surrounding an ovum. 11, Ovum. 12, Germinal disc. 13, Region of the oviduct in which the superficial layers of albumen and the shell-mem- brane are formed. 14, Lower part of the oviduct (" uterus," shell-gland). 15, Rectum. 16, Reflected wall of the abdomen. 17, Anus, or external opening of cloaca. 24 THE DEVELOPMENT OF THE CHICK of the oviduct. The escape of the ovum from the follicle is known as the process of ovulation. During its passage down the ovi- duct it becomes surrounded by layers of albumen secreted by the oviducal glands. The shell- membrane is secreted in the isthmus and the shell in the uterus (Fig. 5). The ovum is fertilized in the uppermost part of the oviduct and the cleavage and early stages of formation of the germ-layers take place be- fore the egg is laid. The time occupied by the ovum in tra- versing the various sections of the oviduct is estimated by Kolliker as follows: Upper two thirds of the oviduct about three hours (formation of al- bumen), isthmus about three hours (secretion of shell-mem- brane) , uterus twelve to twenty- four hours (formation of shell and laying). These figures are only approximate and it is obvious that they are likely to vary considerably in different breeds of hens. Some of the details of these Fig. 5. — Uterus (shell-gland) of the hen cut open to show the fully formed egg. (After Duval.) 1, Cut surface of oviduct, region of remarkable processes deserve isthmus. 2 Reflected flap of uterus, attention: the observations of 3, Egg ready to be laid. 4, Lower extremity, or vaginal portion, of the several naturalists demonstrate thf o;^ucti,ft:1rclorca''''7':o1,en! that the ripe follicle i. em- ing of the rectum into the cloaca. 8, braced by the funnel of the ovi- ^^^^^^- duct before its rupture so that the ovum does not escape into the body-cavity, but into the oviduct itself. Coste describes the process in the following way: "In hens killed seventeen to twenty hours after laying I have observed all the stages of this remarkable process. In some the follicle, still intact and enclosing its egg, had already been swallowed, and the mouth of the oviduct, contracted THE EGG 25 around the stalk of the capsule, seemed to exert some pressure on it, in other cases the ruptured capsule still partly enclosed the egg which projected from the opening; in others finally the empty capsule had just deposited the egg in the entrance of the oviduct/' The existence of double-yolked eggs renders it probable that the oviduct can pick up eggs that have escaped into the body- cavity. But in some cases ova that escape into the body-cavity undergo resorption there. Immediately after the o\aim is received by the oviduct it appears to become softer and more flexible (Coste). The upper- most portion of the oviduct then secretes a special layer of albu- men which adheres closely to the vitelline membrane and is prolonged in two strands, one extending up and the other down the oviduct; these strands become the chalazse; the layer to which they are attached may^ therefore, be called the chalaziferous layer (Coste) of the albumen. The ovum then passes down the oviduct, rotating on the chalazal axis, and thus describing a spiral path; the albumen which is secreted abundantly in advance of the ovum is therefore wrapped around the chalaziferous layer and chalazae in successive spiral layers and the chalazse are re- volved in spiral turns. The main factor in propulsion of the ovum along the oviduct appears to be the peristaltic movements of the latter; it is probable that the cilia which line the cavity have something to do with the rotation of the ovum on its chalazal axis. The line joining the attachments of the chalazse is at right angles to the main axis of the ovum (that passing through the germinal disc); it is obvious, therefore, that there must be some antecedent condition that determines the position of the ovum in the oviduct; probably the position of the ovum in the follicle, i.e., the relation of the germinal disc to the stigma, for the fol- licular orientation is apparently preserved in the oviduct. The question is of considerable importance because, as we shall see. the axis of the embryo is later bisected by a plane passing through the chalazae, and is therefore certainly determined at the time that the chalazse are formed. Is the embryonic axis determined before or after ovulation, and how is it determined in either event? This question, to which there is at present no answer, furnishes an interesting problem for investigation. 26 THE DEVELOPMENT OF THE CHICK Abnormal eggs are of two main kinds: those with more than one yolk, and enclosed eggs (ovum in ovo). Doiible-yolked eggs are obviously due to the simultaneous, or almost simultaneous, liberation of two yolks, and their incorporation in a single set of egg-membranes. The two yolks are usually separate in such cases and are derived, presumably, from separate follicles. But two yolks within a single vitelline membrane have been observed; such are in all probability products of a single follicle. Cases of three yolks within a single shell are extremely rare. The class of enclosed eggs includes those in which there are two shells, one within the other. There are different cases: (1) those in which the contents of the enclosed and the enclosing eggs are substantially normal, though of course the enclosing shell is abnormally large. (2) the enclosed egg may be abnormal as to size (small yolk), or contents (no yolk). In all cases described, the enclosing egg possesses a yolk (Parker). Abnormal eggs of these three classes are of either ovarian or oviducal origin; double- yolked eggs and eggs with abnormal yolks are due to abnormal ovarian conditions; enclosed eggs to abnormal oviducal condi- tions, or to both ovarian and oviducal abnormalities. Assuming the normal peristalsis of the oviduct to be reversed when a fully formed egg is present, the egg would be carried up the oviduct a greater or less distance and might there meet a second yolk. If the peristalsis became normal again, both would be carried to the uterus and enclosed in a common shell. (For a fuller discussion of double eggs see G. H. Parker.) Ovogenesis. The ovogenesis, or development of ova, may be divided into three very distinct stages. The first stage, or period of multiplication, is embryonic and ends about the time of hatching (in the chick) ; it is characterized by the small size of the ova and their rapid multiplication by division. The multi- plying primitive ova are known as ovogonia. At the end of this period multiplication ceases and the period of growth begins. The ova, known as ovocytes of the first order, become enclosed in follicles; the size of the ovum constantly increases and the yolk is formed. The third period, known as the period of matura- tion, is characterized by two successive exceedingly unequal divisions of the egg-cell, producing two minute cells, the polar globules, that take no part in the formation of the embryo, but die and degenerate. The process of maturation begins in the THE EGG 27 fully ripe follicle and is completed after ovulation in the oviduct, while the ovum is being fertilized. The origin of the primitive ova, their multiplication and the formation of the primordial follicles is described in Chapter XIIL In the young chick all the cell cords and cell nests (de- scribed in Chapter XIII) become converted into primordial follicles. During the egg-laying period there is a continuous process of growth and ripening of the primordial follicles, which takes place successively; the immense majority at any given period remain latent, so that at any time all stages of growth of egg follicles may be found in a laying hen. A primordial follicle consists of the ovum surrounded by a single layer of cubical epithelial cells (granulosa or follicle cells); the fibers of the adjacent stroma have a concentric arrangement around the follicle forming the theca folliculi (Fig. 6 8tr.). The ovum itself is a rounded cell with a large nucleus which may be central in position or slightly ex- centric. In the protoplasm on one side of the nucleus is a con- centrated mass of protoplasm from which rays extend out into the protoplasm. This is the so- called yolk-nucleus; it probably corresponds morphologically to the attraction sphere of other cells. Fig. 6. — Primordial follicle from the ovary of the hen. (After Holl.) Or., Granulosa. N., Nucleus. Str., Stroma. Y. N., Yolk nucleus. Holl derives the follicular cells in birds from the stroma, but on insufficient grounds. The most re- cent and, in many respects, the best account is that of D 'Hollander. According to this author they are derived, like the primitive ova, from the germinal epithelium, in which he agrees with the majority of his predecessors. He states that the period of multiplication of the ovo- gonia ends about the time of hatching ; that the period of growth of the ovocytes begins at about the fourteenth day of incubation (seven days before hatching), and before the formation of the primordial follicle, which begins on the fourth day after hatching. Thus the periods of multiplication and growth overlap. He gives a detailed and well-illus- trated account of the nuclear changes accompanying the first stages of growth (synapsis, etc.) 28 THE DEVELOPMENT OF THE CHICK Although the nuculeus (germinal vesicle of authors) may be excentric in position in the youngest ovocytes, it always occupies an approximately central position in those slightly older. The nucleus increases in size with the growth of the cell-body; in the youngest ovocytes its diameter is about 9 /x. and Fig. 7. — Section of an ovarian ovum of the pigeon; drawn from a prepara- tion of Mr. J. T. Patterson. The actual dimensions of the ovum are 1.44 X 1.25 mm. f. s., Stalk of follicle. G. V., Germinal vesicle. Gr., Granvilosa. L., Latebra. p. P., Peripheral protoplasm, pr. f., Primordial follicles. Th. ex., Theca externa. Th. int., Theca interna. Y. Y., Yellow yolk. Z. r., Zona radiata. in the ripe ovum it is flattened and measures about 117 x 315 fi. It retains its central position until the ovum is about 0.66 mm. in diameter, and then moves to the surface where it lies almost in contact with the vitelline membrane (Fig. 7). It becomes elliptical, and later the outer surface is fiat- THE EGG 29 tened against the vitelline membrane, the inner surface re- maining convex (Fig. 8). The point on the surface to which the germinal vesicle migrates is situated away from the surface of the ovary, and thus in the position of the pedicle of the follicle, when the latter projects from the surface of the ovary (Fig. 7). . The formation of the yolk has not received the attention that the subject deserves; and it is possible to give only a very general outline. While the nucleus is still in the center of the egg a very dense deposit of extremely fine granules is formed around it, and gradually extends out towards the periphery of the cell, but does not involve the peripheral layer of protoplasm, which is slightly thicker at the innermost side of the follicle correspond- ing to the stalk. When the ovum has reached a size of approx- imately 0.66 mm, the nucleus moves tow^ards the thickening of the peripheral layer and enters it, lying very close to the vitel- line membrane. The very finely granular central aggregation of yolk-granules represents the primordium of the latebra or central mass of the white yolk. After the nucleus has reached the periphery, or while it is still on its way, the yellow yolk begins to be formed by the peripheral layer of protoplasm. Small yolk-granules arise near the inner margin of the peripheral layer of protoplasm and increase in size; each becomes enclosed in a vacuole which grows to a considerable size, so that the accumulation of vacu- oles on the inner surface of the peripheral protoplasm soon produces a kind of emulsion; this appears in section like a reticulum, the spaces of which are the sectioned vacuoles, and the strands the remains of the protoplasm in w^hich the vacuoles are embedded (Fig. 7). This layer lies between the peripheral unmodified protoplasm and the white yolk. A sim- ilar process is going on at the same time in the central pri- mordium of the white yolk, but the vacuoles and the granules are smaller, and the contrast between the white and yellow yolk is obvious (Fig. 7). Successive layers of yellow yolk are deposited around the central mass of white yolk by the activity of the peripheral layer of protoplasm. Thus one can understand in general how the concentric arrangement of the yellow yolk is produced. But the alternation of layers as seen in a section of the hardened yolk 30 THE DEVELOPMENT OF THE CHICK implies a certain periodicity in the deposition of the yolk that is not understood. 1 The germinal vesicle lies in a thickening of the peripheral layer of protoplasm known as the germinal disc, which is con- tinuous, like the remainder of the peripheral protoplasm, with the protoplasmic reticulum that forms the walls of the yolk- vacuoles. The protoplasmic reticulum gradually disappears as the vacuoles are completely converted into yolk-granules. In the last stages of growth a peripheral layer of white yolk is formed around the entire yolk-mass, presumably by the peripheral proto- plasmic layer. The germinal disc increases in extent and thick- ness; but whether this takes place by inflowing of the peripheral protoplasm, or growth of the original disc, is not known. It is certain that the peripheral protoplasm disappears over most of the yolk when the external layer of white yolk is formed. An inflow of the peripheral protoplasm into the disc appears very probable by analogy with the bony fishes where this process can be studied with great ease. The alternative assumption would be the complete utilization of the peripheral protoplasm in form- ing the superficial layer of white yolk. The method of formation of the neck of the latebra and the so-called nucleus of Pander, or peripheral expansion of the neck, follows more or less directly from the preceding account: As the circumference of the o\ami enlarges, the germinal disc is carried out and leaves behind it a trail in which yellow yolk is not formed. When the ovum is fully grown, the exact boundaries between the protoplasmic germinal disc and the yolk are not determinable. The disc itself is charged with small yolk-granules which grade off very gradually into the white yolk lying around and beneath the disc. The mode of nutrition of the ovum and the formation of the vitelline membrane remain to be considered. The nutrition is conveyed from the highly vascular theca folliculi by way of the follicular cells, or membrana granulosa, to the ovum. The nutri- ment enters by diffusion; at no stage is there any evidence of 1 Since the above was written, observations and experiments of Dr. Oscar Riddle have demonstrated that the periodicity is daily, and is correlated with the daily physiological rhythm of vitality of the hen. Acknowledgment is due Dr. Riddle for permission to make this statement in advance of his own publication. THE EGG 31 immigration of solid food particles, let alone entire cells, into the growing ovum. At an early stage a definite membrane is formed between the ovum and the follicular cells, the zona radiata or primordium of the vitelline membrane (Fig. 7). This is pierced by innumerable extremely minute pores which become narrow canals as the zona radiata increases in thickness. The follicular cells and the peripheral layer of protoplasm of the ovum are connected by extremely delicate strands of protoplasm that pass through the pores (Holl). In some way the nutriment of the ovum is conveyed through these strands. The discussion as to whether the zona radiata is a product of the ovum itself or of the follicular cells seems to me to be largely academic and wdll not be summarized here. There seems to be sufficient evidence of a primary true vitelline membrane secreted by the o\auu itself, though this may not represent the entire zona radiata of older ova. The third phase of ovogenesis, maturation or formation of the polar globules, is transferred to the next chapter, because it is overlapped by the process of fertilization. It is not definitely known if maturation in birds may be completed without fertiliza- tion, but it seems probable that, as in many other animals, the completion of maturation is dependent on the stimulus of fertiliza- tion. It is, however, essentially a process absolutely distinct from fertilization, and in some animals {e.g., echinids) is com- pleted without fertilization. CHAPTER II THE DEVELOPMENT PRIOR TO LAYING I. Maturation The phenomena of the maturation and fertiUzation of the hen's egg are almost entirely unknown. The observations of Holl demonstrate only that the wall of the germinal vesicle tends to disappear when the follicle is nearly ripe. He also doubtfully identified six rod-shaped bodies at the margin of the germinal vesicle as the chromosomes of the maturation spindle (Fig. 8). Ch!-. W0!f^- '^,^^^^p°' '=^.♦5 ^° ^ J'oV^<5y?ad''^%° o o ^iS^o o- o^ "o-V" Fig. 8. — Section of the germinal vesicle and surrounding parts of an ovarian ovum of the hen measuring 40 x 35 mm. (after Holl). Chr., Chromosomes. Gr., Granulosa. G. V., Germinal vesi- cle. Z. r., Zona radiata. But we have fortunately a very good account of the maturation and fertilization of the pigeon's egg by E. H. Harper, which fur- nishes the basis of the following description: The wall of the germinal vesicle begins to break down in ovarian eggs of about 18.75 mm. diameter, the full size of the egg of the pigeon being about 25 mm. Part of the fluid contents of the germinal vesicle flows out and forms a layer outside the disinte- 32 DEVELOPMENT PRIOR TO LAYING 33 grating wail (Fig. 9). Tlie cliromosomes and nucleoli form a group near the center of the upper plane surface of the germinal Chr •>?^ f<^ -tf^ ^ Fig. 9. — Vertical section of the germinal vesicle and part of the germinal disc of an ovarian ovum | inch in diameter; pigeon, x 385. (After Harper.) Chr., Chromosomes. Gr., Granulosa. G. V., Wall of germinal vesicle. vesicle. The first maturation spindle is formed before ovulation, containing eight quadruple chromosomes (tetrads). The spindle is still in the equatorial plate stage when the ovum is grasped by the mouth of the oviduct (Fig. 10). The bulk of the substance of the ger- minal vesicle soon forms a yolk-free cone extending from the maturation spin- dle deep into the superficial ^ ^ yolk. The outer end of the Fig. 10. — Vertical section of the germinal spindle is in almost imme- fh^c of the pigeon's egg showing the diate contact with the sur- ^^^^ maturation spindle. The egg was . clasped by the funnel or the oviduct, face of the ovum. In the g ^^ ^ ^^ ^ 2OOO. (After Harper.) later stages of formation of m.Sp. 1, First maturation spindle. Tetr., the first polar body each Tetrad. tetrad, or quadruple chromosome, separates into two dyads or double chromosomes, and the members of each pair of dyads separate and approach opposite ends of the spindle (anaphase). 34 THE DEVELOPMENT OF THE CHICK Thus at each end of the spindle there are eight dyads. Those at the outer end then enter a Uttle bud of protoplasm projecting above the surface of the germinal disc, and this bud with the dyads is cut off as the first polar body, which lies in a depression of the germinal disc beneath the vitelline membrane (Fig. 11). Eight dyads, therefore, remain within the germinal disc. A second maturation spindle is then formed almost imme- diately, apparently without the intervention of a resting stage of the nucleus, and takes a radial position similar to that occupied by the first, with the dyads forming an equatorial plate (Fig. 11). ^■-v.M. t.^sm '71. 3 p. 2. Fig. 11. — Second maturation spindle and first polar hody of the pigeon's egg; a combination of two sections. 8.15 p.m. x 2000. (After Harper.) m. Sp. 2, Second maturation spindle, p. b. 1, First polar body. v. M., Vitelline membrane. Each dyad then divides along the preformed plane of division, and the daughter-chromosomes diverge towards opposite poles of .the spindle. The outer end of the second maturation spindle then enters a superficial bud of the protoplasm of the germinal disc similar to that of the first maturation spindle; and this bud together with the contained chromosomes becomes cut off as the second polar body. The result of these processes of maturation is the formation of three cells, viz., the two polar bodies and the mature egg. The polar bodies are relatively very minute and soon degenerate completely. After the formation of the second polar body there remain in the egg eight chromosomes, each of which represents one quarter of an original tetrad. These form a small resting nucleus known as the egg-nucleus or female pronucleus. It is many times smaller than the original germinal vesicle (Fig. 12), and DEVELOPMENT PRIOR TO LAYING 35 it rapidly withdraws from the surface of the egg to a deeper position near the center of the germinal disc. (Concerning the Fig. 12.- — Egg nucleus (female pronucleus) and polar bodies of the pigeon's egg. (After Harper.) 8.30 p.m. x 2000. E. N., Egg nucleus, p. b. 1, First polar body. p. b. 2, Second polar body. p'v. S., Perivitelline space, v. M., Vi- telline membrane. general theory of the maturation process see E. B. Wilson, "The Cell in Development and Inheritance/' the Macmillan Company, New York.) II. Fertilization The spermatozoa traverse the entire length of the oviduct and are found in the uppermost portion in a fertile hen. The period of life of the spermatozoa within the oviduct is considerable, -as proved by the fact that hens may continue to lay fertile eggs for a period of at least three weeks after isolation from the cock. After the end of the third week the vitality of the spermatozoa is apparently reduced, as eggs laid during the fourth and fifth weeks may exhibit, at the most, abnormal cleavage, which soon ceases. Eggs laid forty days after isolation are certainly unfer- tilized, and do not develop (Lau and Barfurth). The so-called parthenogenetic cleavage of such eggs is merely a phenomenon of fragmentation of the protoplasm; there is no true cell-division. The ovum is surrounded immediately after ovulation, that is in the infundibulum,by a fluid containing spermatozoa in suspen- sion. In the egg of the pigeon a certain number of spermatozoa 36 THE DEVELOPMENT OF THE CHICK immediately bore through the egg-membrane and enter the ger- minal disc, within which the heads, which represent the nuclei of the spermatozoa, enlarge and become transformed into sperm nuclei (Fig. 13). The fate of the middle piece and tail of the sperma- tozoa is not known in birds, but it is improbable that they furnish any definitive morphological element of the fertilized egg. At the time of entrance of the spermatozoa the first maturation spin- dle is in process of formation; it lies in the center of a group of granules at the sur- face of the egg, which is bounded by a non-granular zone of protoplasm, called by Harper the polar ring, in which the sperm- nuclei accumulate. External to the polar ring the protoplasm is granular again (Fig. 14). The sperm-nuclei remain quiescent while the polar bodies are being formed, and, when the egg nucleus is reconstituted, one x2000. (After Har- per.) The order of stages is indicated by the letters a — g. Fig. 13. — Stages in the transformation of sperm heads into the sperm nuclei from the of them, which may be called the male pro- ovum of the pigeon, nucleus or primary sperm nucleus, moves inwards and comes into contact with the egg nucleus (Fig. 15). The opposed faces of the conjugating nuclei become flattened together, until the contours form a single sphere, the first segmentation nucleus, in which a partition sep- arates the original components, viz., the sperm and egg nucleus. The partition apparently disappears. However, it is very un- likely that a complete intermingling of the contents of the two germ-nuclei takes place, because in other groups of animals where the processes have been more fully studied, it has been determined that each germ-nucleus forms an independent group of chromo- somes of the same number in each. Shortly after its formation, the first segmentation nucleus prepares for division in the usual karyokinetic way. The first segmentation (or cleavage) spindle thus formed lies near the center of the germinal disc a short distance beneath the surface and its axis is tangential to the surface, or, in other words, at right angles to the axis of the egg. The fertilization may be considered to be completed at this stage. DEVELOPMEXT PRIOR TO LAYING 37 The entrance of several spermatozoa appears to be character- istic of vertebrates with large ova; thus for instance, it has been described in selachii, some amphibia, reptiles, and birds. Such a condition is known as polyspermy; it is normal in the forms mentioned, but occurs only under abnormal conditions in the iev'.'.V: ■■■;•- ■; a^': mm: fi:/::v '-iii 'Sp.N: :■•.•:) ■^SA:-^;;; "^^•-^^^'-■y^'^^^'■t^-•-"^^' '■^•^•f^i •"•—'■"''' '^'''^^^ mm '.-'•'•J- Fig. 14. — Horizontal section of the germinal disc of a pig- eon's ovum immediately after ovulation, x 125. (After Harper.) N., Nucleus, probably first maturation spindle, p. r., Polar ring. 8p. N., Sperm nuclei. K" \^ f^ 1,5" '^.V^-^ -J Fig. 15. — Vertical section of the pigeon's egg showing germ nuclei (pronuclei) in the center of the disc, x 2000. 10.40 p.m. (After Harper.) 38 THE DEVELOPMENT OF THE CHICK great majority of animals. Harper observed that the number of sperm-nuclei formed in the pigeon varied from twelve to twenty- five in different cases. Only one of these serves as a functional sperm-nucleus; the remainder or supernumerary sperm-nuclei migrate, as though repelled, from the center towards the margins and deeper portions of the germinal disc, where they become temporarily active, dividing and furnishing a secondary area of small cells (accessory cleavage) surrounding the true cleavage- cells produced by division of the central portion of the disc around the descendants of the segmentation nucleus. It has been sup- posed by some authors who studied the selachii that the de- scendants of the supernumerary sperm-nuclei form functional nuclei of the so-called periblast, but this view has been disproved for the pigeon (Blount), in which it can be demonstrated that the supernumerary sperm-nuclei have but a brief period of activity, and then degenerate. III. Cleavage of the Ovum The fertilized ovum is morphologically a single cell, with a single nucleus, the first segmentation nucleus. The living proto- plasm is aggregated in the germinal disc, and the remainder of the ovum is an inert mass of food material destined to be assimi- lated by the embryo which arises from the germinal disc. The first step in the development is a series of cell-divisions of the usual karyokinetic type, restricted to the germinal disc, which rapidly becomes multicellular. As the early divisions take place nearly synchronously in all the cells, there is a tendenc}' for the number of the cells to increase in geometrical progression, fur- nishing 2-, 4-, 8-, and 16- etc., celled stages; but sooner or later the divisions cease to be synchronous. All of the cells of the body are derived from the germinal disc, and the nuclei of all cells trace their lineage back to the first segmentation nucleus. The supernumerary sperm-nuclei do not take part in the forma- tion of the embryo. Cell-division is the most conspicuous part of the early de- velopment; hence this period is known as the cleavage, or segmentation, period. But it should be remembered first, that cell-division is as constant a process in later embryonic stages as in the cleavage period, and second, that it is probable, though little is known yet about this subject in the bird's egg, that DEVELOPMENT PRIOR TO LAYING 39 other important phenomena are going on during the cleavage period. The type of cleavage exhibited by the bird's egg is known as meroblastic, for the reason that only a part of the ovum is concerned, viz., the germinal disc. This is obviously due to the great amount of yolk (see Introduction, pp. 11 and 12). To understand the form and significance of the cleavage of the bird's egg, it is necessary first of all to gain a clear idea of the structure of the germinal disc and its relations to the yolk. At the time of the first cleavage the germinal disc is round in surface view and about 3 mm. in diameter; the center is white and is surrounded by a darker margin about 0.5 mm. wide. These \ two zones have been compared to the pellucid and opaque areas of later stages, but it is certain that the correspondence is not exact. We shall call the outer zone the periblastic zone, or simply periblast. In section, the germinal disc is biconvex, but the outer surface which conforms to the contour of the entire egg is much less arched than the inner surface. The disc is every- where separated from the yellow yolk by a layer of white yolk (Fig. 2) ; on the other hand, there is no sharp separation between the disc and the white yolk. The granules of the latter are largest in the deeper layers and there is a gradual transition from them to the smaller yolk-granules with which the disc is thickly charged (Fig. 19). It is practically impossible in a section to say w^here the protoplasm of the disc ceases; it is indeed probable that it extends some distance into the white yolk both beneath and around the margins of the disc. Thus in Figure 21a cone, ap- parently of protoplasm, extends into the neck of the latebra a considerable distance. In other cases it does not extend so far. The Hen's Egg. The form of cleavage of the hen's egg is illustrated in Fig. 16, A-E. The first cleavage appears in surface view as a narrow furrow extending part way across the germinal disc (Fig. 16 A). According to Coste the furrow is central in po- sition, but Kolliker describes it as excentric. Probably both con- ditions may be found in different eggs. While the ends of the first cleavage furrow are still extending towards the periblast, the second division begins. It is a vertical division in each cell like the first and the two furrows meet the first cleavage furrow at right angles. They may meet the first furrow at approximately the same point, in which case they form an approximately straight 40 THE DEVELOPMENT OF THE CHICK Fig. 16. — Five stages of the cleavage of the hen's egg. (After KolHker.) A. First cleavage furrow (x 14). The egg came from the lower end of the oviduct. B. Four-celled stage (x 17); from the uterus. C. Ten central and eleven marginal cells (x about 16). D. Nine central and sixteen marginal cells (x about 16). E. Late cleavage stage (x about 22). DEVELOPMENT PRIOR TO LAYING 41 line (Fig. 16 B), or they may meet the first cleavage furrow at separate points, in which case the intervening part of the first furrow becomes bent at an angle, forming a cross furrow. The third cleavage of the hen's egg has not been figured or described by any author, so far as I know. But it is probable from analogy with other similar forms of cleavage that in each of the four cells a furrow arises approximately at right angles to the second furrow and parallel to the first, thus producing eight cells in two parallel rows of four each. But the variable forms of the succeeding cleavage stages indicate a probable considerable variation in the eight-celled stage. Before describing the later cleavage stages, we should note certain important relations of the first four or eight cells: First, these are not complete cells in the sense that they are separate from one another. They are, indeed, areas with separate nuclei marked out by cleavage furrows in a continuous mass of proto- plasm. The furrows do not cut through the entire depth of the germinal disc, and the cells are therefore connected below by the deeper layer of the protoplasm; nor do the furrows extend into the periblast, and all the cells are therefore united at their^ margins by the unsegmented ring of periblast. Second, accord- ing to several observers, the center of the cleavage, i.e., the place where the first two cleavage furrows cross, is excentric. It is believed by those who emphasize this point, that the displace- ment is towards the posterior end of the blastoderm; but Coste, for instance, failed to note any excentricity. The number of observations is still too few to admit of a safe conclusion on this point; in the pigeon, according to Miss Blount's observations recorded below, excentricity appears to be exceptional; more- over, the excentric area may bear any relation whatever to the future hind end of the embryo, so that in the pigeon it will not bear the interpretation that has been placed on it in the hen's egg. The following cleavages (after the eight-celled stage) in the hen's egg are very irregular, but two classes of furrows may be distinguished in surface view: (1) those that cut off the inner ends of the cells, and (2) those that run in a radial direction. The furrows of the first class produce a group of cells that are bounded on all sides in surface view, but these are, at first, still connected below by the deeper protoplasm. They may be called 42 THE DEVELOPMENT OF THE CHICK the central cells. These are bounded by cells that are united in the marginal periblast, and thus lack marginal boundaries as well as deep boundaries; these may be called the marginal cells (Fig. 16 0). The distinction between central and marginal cells is one of great importance which should be clearly grasped. '' In the surface views of later cleavages the following points should be noted: (1) the group of central cells increases by the addition of cells cut off from the inner ends of the marginal cells, and by the multiplication of the central cells themselves; (2) the marginal cells increase by the formation of new radial furrows. The increase of the central cells is much more rapid than that of the marginal cells, and the cells themselves are much smaller than the marginal cells, both because of their mode of origin and also because of their more rapid multiplication. The area of the central cells is also constantly increasing, with consequent re- duction of the marginal zone (Fig. 16 E). Emphasis has been laid by several authors on the excentric position of the smallest cells, and the inference has been drawn that these represent the hinder end of the blastodisc. Similar excentricity in the pigeon's egg is without reference to the future embryonic axis (see Fig. 18). But the surface views do not show w^hat is going on in the deeper parts of the germinal disc. Sections show that after about the 16- or 32-celled stage an entirely new class of cleav- age planes arises in the central cells. These planes are parallel to the surface, and the superficial cells arising from such a division are therefore completed below. Of the two daughter-nuclei produced by such a division, one remains in the superficial cell and the other in the unsegmented deep layer of the germinal <^ disc, w^hich thus becomes nucleated. After this the nuclei mul- ^ p . tiply in this deeper layer and cells are constantly being produced, » which bud off from it and become added to the segmented part . '/ I of the germinal disc above. \j^'^ In this way the entire thickness of the central part of the J^ y germinal disc becomes gradually converted into cells. A cavity V^ arises between the cellular disc and the white yolk below, the seg- ^ mentation cavity, often called the subgerminal cavity. It is first formed in the center of the central group of cells and extends out gradually towards the margin, Vmt it never cuts under the mar- ginal cells, which remain united below^ and at their margins by the periblast. DEVELOPMENT PRIOR TO LAYING 43 Duval interprets a narrow space observed by him between the single superficial layer of cells and the deeper cells of the germinal disc as the segmentation cavity; it is thus entirely distinct from the subgerminal cavity which arises much later, according to his conception. Apart from the fact that his figures appear to represent the merely virtual space between the superficial cells and the underlying cells in an exag- gerated form, the interpretation appears to me to be incorrect. It is based on the theory that the deeper cells represent the primary entoderm, a view which I cannot accept; the interpretation of this space as cleavage cavity fails if it be shown (see beyond) that the underlying cells are not entoderm. The account given above of the deeper cleavages, those seen in section, is the conventional one, based on the observations of Kolliker, Duval, and others. The account, that follows, of the corresponding cleavages in the pigeon's egg, is different in some important respects, that bring it into agreement with the best known meroblastic eggs, those of the bony fishes. I have, however, allowed the above account to stand, though I consider it probable that a careful re-examination would bring the cleavage of the hen's egg into line with that of the pigeon and the teleost. The Pigeon's Egg. The cleavage of the pigeon's egg has been worked out in more detail than that of the hen's egg (Blount) ; as it offers some interesting features that have never been de- scribed for the hen's egg, and must be made the basis of the description of the formation of the germinal wall and the germ- layers in the absence of any consistent account for the hen's egg, it will next be described. The fundamental features of the cleav- age are the same as in the hen's egg, so that the description need not be repeated. The feature to be particularly emphasized in the cleavage of the pigeon's egg is the occurrence of a secondary or accessory cleavage in the marginal zone or periblast (Figs. 17 and 18 A). When the origin of these cells is traced it is found that they arisft around the supernumerary sperm-nuclei, which accumulate and multiply in the periblast. The complete history of these nuclei has been worked out by Harper and Blount, so that there can be no doubt as to their derivation. Another interesting point illustrated by the figures is that the marginal cells have a peripheral wall wherever the accessory cleavage occurs, but between the groups of accessory cleavage cells the marginal cells are continuous with the periblast (Figs. 17 and 18 A), as they are 44 THE DEVELOPMENT OF THE CHICK everywhere in the hen's egg. In a section of a germinal disc, showing the accessory cleavage (Fig. 20), it is seen that the peripheral boundary of the marginal cells cuts under the margin for a considerable distance. The accessory cleavage becomes manifest at the time of appearance of the first cleavage plane, and increases in amount Fig. 17. — Photograph of an eight-celled pigeon ovum (after Mary Blount). 2.45 a.m. Accessory cleavage (ac. cl.) in the marginal zone bounding the segmented area. Vesicles, appearing black in the photograph, are seen on the surface of the yolk beyond the mar- ginal zone of the germinal disc. Orientation as in Fig. 18. up to about the 32-celled stage, and thereafter gradually decreases until it completely disappears (Figs. 18 B, C, and D). The peripheral boundaries of the marginal cells disappear pari passu, and, when the accessory cleavage is finally wdped out, the mar- ginal cells are everywhere continuous with the periblast, as in the hen's egg (Figs. 18 B and C). In some eggs the accessory cleavage is much more extensive than in others; indeed, in some it appears to be entirelv absent, but this is relativelv rare. In the stage shown in Fig. 18 B, for instance, there is usually con- siderable accessory cleavage; but in this egg there is none. The variation is obviously due to variations in the number of super- numerary spermatozoa, such as may readily occur. DEVELOPMENT PRIOR TO LAYIXG 45 The question arises whether the disappearance of the cell- walls around the sperm-nuclei is caused by degeneration of the latter, or is simply a later syncytial condition in the periblast in B C D Fig. 18. — Photographs of the cleavage of the pigeon's ovum (after Mary Blount). The figures are so arranged that the axis of the shell is across the page with the large end to the left. The future axis of the embryo is therefore inclined 45° to the margin of the page with the anterior end to the right above. A. A very regular sixteen -celled stage; accessory cleavage well shown; though not well focused on the lower margin. 3.45 a.m. B. Approximate thirtv-two celled stage. There is no accessory cleavage in this case. The formation of the central from the margmal cells may be readily observed in this figure. 5.15 a.m. C. Later stage of cleavage. 7.10 a.m. , D. Cleavaiie at 9.80 a.m. The marjrinal cells are now becommg separated peripherally from the periblast which has received its nuclei from them. which the sperm-nuclei are embedded. There can be little doubt that the former alternative is correct. While in the stages of the accessory cleavage, sperm-nuclei are readily found both m 46 THE DEVELOPMENT OF THE CHICK the accessory cleavage-cells and also in the unsegmented periblast (Figs. 19 and 20), they decrease in number as the accessory cleavage planes disappear, and when the latter are entirely lost 1 •> Fig. 19. — Transverse section of the blastoderm of a pigeon's egg about 8| hours after fertilization (4.45 a.m.). (After Blount.) 1, Accessory cleavage. 2, Migrating sperm-nuclei, a, b, c, d, Cells of primary cleavage. the periblast is absolutely devoid of nuclei. Fragmentation of the sperm-nuclei is a frequent accompaniment of their disappearance. Thus the accessory cleavage is a secondary and transient feature of the cleavage of the pigeon's egg due to polyspermy. After it has passed, the ovum is in precisely the same condition Fig. 20. — Transverse section of the blastoderm of a pigeon's eg^ at the end of the period of multiplication of sperm-nuclei, about 10 hours after fertil- ization (6.30 A.M.). (After Blount.) 1, Accessory cleavage around the sperm -nuclei. 2, Marginal cells; sharply separated from the sperm-nuclei. 3, Central cells. 4, Sperm-nuclei. as the hen's ovum of the same stage of development. It is doubt- ful whether the absence of accessory cleavage in the hen's egg should be taken as evidence that the fertilization is monospermic. It may well be that supernumerary sperm-nuclei are present without producing the appearance of accessory cleavage, owing, perhaps, to a deeper situation in the periblast. This point requires investigation. Another feature brought out by these photographs requires emphasis. The periblast ring shows no definite outer margin, DEVELOPMENT PRIOR TO LAYING 47 but beyond the zone of the accessory cleavage there may occur two or three concentric circles variously indicated (Fig, 17). Vacuoles, appearing black in the photographs, are very common in the outer zones. These appearances indicate that the peii- blastic protoplasm extends farther out in the superficial white yolk than is usually believed to be the case; and this suggests an interesting comparison with the teleost ovum, where the peri- blastic protoplasm surrounds the entire yolk as a very thin layer. Sections confirm the idea that the periblastic protoplasm has an extension beyond the so-called margin of the blastodisc. Some eggs show a more definite margin than others; it may be that there is a periodic heaping of the periblast at the margins, for which again an analogy may be found in teleosts. Although the smallest cells may be more or less excentric in the segmented germinal disc of the pigeon, their position bears no constant relation to the future embryonic axis. They may lie in this axis in front of or behind the middle, or to the right or left of it (cf. Fig. 18 A-D). At the eight-celled stage a horizontal cleavage plane begins to appear beneath the central cells (Fig. 19). This marks the full depth of the blastoderm at all stages, and the several-layered condition arises by horizontal cleavages between this and the surface. Comparison of Figs. 19, 20, and 22, drawn at the same magnification, will show that the depth does not increase by addi- tion of cells cut off from below, as is usually supposed to be the case in the bird's ovum. The first horizontal cleavage plane not "? only marks the full depth of the blastoderm, but it also indicates ) the site of the segmentation cavity which arises gradually by accu- mulation of fluid between the cells and the underlying unseg-' mented protoplasm and yolk. The segmentation cavity gradually extends towards the margin of the blastoderm, but it is bounded peripherally by the zone of junction between the marginal cells and the periblast. IV. Origin of the Periblastic Nuclei, Formation of the Germ-wall Our knowledge of this part of the subject in the hen's egg is very incomplete, and the various accounts are contradictory. The reason for this is the great difficulty of securing a complete series of stages, and of arranging them in proper sequence. There ( 48 THE DEVELOPMENT OF THE CHICK is no way of timing the development, so that one has to judge the sequence of the stages, all of which come from the uterus, by the degree of formation of the shell, by the size of the cells and b}" the appearance of the sections. This can be at best only approximate; and, as the securing of any given stage is largely a matter of chance, no one has, as a matter of fact, secured a complete series. In the pigeon, on the other hand, the time since laying the first egg is a fairly exact criterion of the stage of development of the second egg. It has, therefore, been pos- sible to secure a complete series, and the subject has been worked out by Miss Blount, whose preliminary communication in Vol. XIII of the Biological Bulletin furnishes the basis of the following account. The periblast ring is entirel}^ devoid of nuclei after the super- numerary sperm-nuclei have degenerated. The marginal cells become greatly reduced in size owing to multiplication and continuous production of central cells, and their nuclei thus approach more and more closely to the periblastic ring. The scene then changes; the marginal cells cease to produce central cells; when their nuclei divide the peripheral daughter-nuclei move out into the periblast, which is thus converted into a nu- cleated syncytium. The periblastic nuclei multiply rapidly and invade all portions of the periblastic ring, which maintains its original connection with the white yolk. Not only do the peri- blastic nuclei invade the periblastic ring, but some of them also migrate centrally into the protoplasm forming the floor of the segmentation cavity. They do not, however, reach the center, but leave a non-nucleated sub-germinal area, corresponding approximately to the nucleus of Pander, free from nuclei. The subgerminal syncytium may be known as the central periblast to distinguish it from the marginal periblast. They are, of course, continuous. In sections one has the appearance of nuclei in the yolk, for there is no sharp boundary between periblast and yolk (Fig. 22). The syncytium, which has received its nuclei from the marginal cells, is the primordium of the germ-wall (Figs. 21, 22, 23, 24). There is a snarp contrast between the segmented blastoderm and the syncytial periblast not only in structure Init also as regards fate. The marginal cells constitute a zone of junction be- tween blastoderm and periblast. Thus in Fig. 22 it will be ob- DEVELOPMENT PRIOR TO LAYING 49 served that the large marginal cells on each side are continuous with the periblast, and nuclei are found in the periblast both central and peripheral to the zone of junction. The latter forms J.' .• .*■*•. •-■.■."■■■•>.••:•;■;■•..■:'.•••:■•"••'.•■■..••'•■•■..■.■• -% u 4« • « * '-' ^^^^•-V*!' • •J ■' o s . ~ •a _ -, _ J ^ 4r • • • :,v •♦. •© •4- • • • • • • •: ■ ■ . • "? •"■*,• ■? ■•'■.• •^ ...* fe • .• Fig. 21. — Longitudinal section of the blastoderm of a pigeon's egg at the time of disappearance of the sperm-nuclei. On the left (anterior) margin, the marginal cells have become open, that is, continuous with the peri- blast, as contrasted with Fig. 20. About 11 hours after fertilization (7.00 A.M.). (After Blount.) 1, Marginal cells. 2, Cone of protoplasm. 3, Marginal periblast. 4, Neck of latebra. 5, Yellow yolk. Fig. 22. — Transverse section through the center of the blastoderm of a pigeon's egg, 14^ hours after fertilization (10.30 a.m.). (After Blount.) 1, Marginal cells. 2, Marginal periblast. 3, Nuclei of the subgerminal periblast. 50 THE DEVELOPMENT OF THE CHICK a ring around the blastoderm. It persists during the expansion of the blastoderm over the surface of the yolk. The blastoderm now begins to expand, owing largely, at first, to additions of cells to its margin cut off from the germ-wall. The central as well as the marginal periblast contributes to the blastoderm, but the former appears to be rapidly used up. The marginal periblast on the other hand grows at its periphery while it adds cells to the blastoderm centrally, and thus it moves out in the white 3'olk, building up the margin of the blastoderm at the same time. The original group of central cells appears to correspond approximately to the pellucid area; the additions from the germ-wall would thus constitute the opaque area. 4 Fig. 23. — Posterior end of a longitudinal section through the blastoderm of a pigeon's egg about 25 hours after fertilization (8.50 p.m.). (After Blount.) 1, Nests of periblast nuclei. 2, Periblast nucleus in marginal position. 3, Syncytial mass derived presumably from the periblast, in process of or- ganization into cells. 4, Vacuoles. Some phases of these processes are illustrated in Figs. 23 and 24. In the vertical section, Fig. 23, the surface of the germ- wall next the blastoderm is indented as though for the formation of superficial cells. Along the steep central margin of the germ- wall groups of cells are apparently being cut off and added to the cellular blastoderm. In the horizontal section, Fig. 24, the process of cellularization at the central margin of the germ-wall is apparently proceeding rapidly. The superficial cells thus added to the margin of the cellular blastoderm become continuous with the ectoderm, and the deeper layers later form the yolk-sac entoderm which becomes continuous with the embryonic entoderm secondarily. The term germ-wall is usually applied to the primordium of the yolk-sac DEVELOPMENT PRIOR TO LAYING 51 entoderm and the periblast proper as weU. We shall follow this usage and distinguish two parts of the germ-wall. Fig. 24. — Part of the mar2;in of a horizontal section through the blastoderm of a pigeon's egg about 25 hours after fertilization (8.50 p.m.). (After Blount.) 1, Periblast nuclei. 2, 3, Cells organized in the periblast. 4, A cell apparently added to the blastoderm from the periblast. 5, Vacuoles. In later stages the inner margin of the periblast becomes much less steep, owing apparently to active proliferation of cells. This is illustrated in the outline drawings of Fig. 25. Later yet the Fig. 25. — Outlines of the margins of transverse sections of the blastoderm of pigeon's eggs; 26 (A), 28 (B), and 32 (C) hours after fertilization. (After Blount.) 52 THE DEVELOPMENT OF THE CHICK marginal cells extend out peripherally and form a short project- ing shelf beyond the zone of junction, appearing wedge-shaped in section (Figs. 28 A, etc.). This we shall call the margin of overgrowth. Thus we may distinguish the following zones: (1) margin of overgrowth; (2) zone of junction; (3) the inner zone of the germ- w^all, and (4) the original cellular blastoderm (pellucid area) Fig. 29. V. Origin of the Ectoderm and Entoderm The ectoderm and entoderm are the primary germ-layers, out of which all organs of the embryo differentiate; hence great importance attaches to the mode of their origin. But up to the present it has not been possible to decide between three con- flicting views. These are: (1) The theory of delamination, viz., that the superficial cells of the segmented blastoderm form the ectoderm and the deeper cells the entoderm; in other words, that the blastoderm splits into the two primary germ-la3^ers. This is the oldest view, but it has not lacked support in recent times, e.g., by Duval. (2) The theory of invagination, viz., that the primary entoderm arises as an ingrowth from the margin of the blastoderm. This view, which was supported by Haeckel, Goette, Rauber, and some others, brings the mode of gastrulation in the bird into line with lower vertebrates. (3) A third and relatively recent point of view is that the primary entoderm arises as an ingrowth of cells from the germ-wall, more particularly from the posterior portion. This view, put forward by Nowack, has been adopted in substance by O. Hertwig (Handbuch der vergl. u. exp. Entwickelungslehre der Wirbeltiere). The reason for the conflict of opinion appears to lie mainly in the fact that the critical stages occur prior to laying, and no one has investigated a complete series of stages. For this reason the subject was reinvestigated in the Zoological Laboratory of the University of Chicago, by Mr. J. Thomas Patterson, at the suggestion of Prof. C. O. \^/ hitman. A very complete series of stages of the pigeon's ovum was studied, with results that are consistent in themselves and that agree with the principles of formation of the primary germ-layers in the lower vertebrates. The author has had the opportunity of following the work step by step, and is convinced of its accuracy. It is therefore made the basis of the following account: DEVELOPMENT PRIOR TO LAYIXG 53 The first step in the process of gastrulation, or formation of the primary entodeiTn, is a thinning of the bla stoderm, which begins sUghtly post erior to tj ie^^center and rapidh^ involves a sector of the poste rior thi rd of the blastoderm. This process occurs between the twenty-first and tenth hours prior to laying. It is due apparently to the gradual rearrangement of the cells in a single layer. A late stage of this process is shown in Figure 26, which represents a complete longitudinal section through the blastoderm ten hours before laying. It will be observed that the anterior portion of the blastoderm is many cells thick (26 A), ; but as one passes towards the posterior end the number of layers | becomes less, and is reduced to a single layer at the extreme pos- terior end. Here and there, e.g., at X, the arrangement of the cells indicates that cells of the lower layer are entering the upper layer. It is obvious that such a process must result in increase of the diameter of the blastoderm, and Patterson states that the average diameter twenty hours prior to laying is 1.915 mm. and ^ 2.573 mm. ten hours later. The thinning also involves enlarge- ment of the segmentation cavity, which may now be known as the subgerminal cavity. Hand in hand with the thinning out there takes place an interruption of the germ-wall at the posterior end, so that in this region the margin no longer enters a syncytium but rests directly on the yolk (cf. anterior and posterior ends of Fig. 26). Figure 27 is a reconstruction of the stage in question. The germ-wall, represented by the parallel lines, is absent at the posterior end. Here the cells of the blastoderm rest directly on the yolk. The sector bounded by this free margin and the broken line represents the area of the blastoderm that is approximately one cell thick. The figures 2 to 7 indicate regions approximately two to seven cells thick. Gastrulation begins by an involution or rolling under of the | free margin, as though the free edge were tucked in beneath the blastoderm. The involuted edge then begins to grow forward, towards the center of the blastoderm, and thus establishes a lower layer of cells, the primary entoderm. As soon as this process is started the margin of the blastoderm begins to thicken, and thus the inner layer of cells (entoderm) and the outer layer of cells (ectoderm) are continuous with one another in a marginal thickening (Fig. 28). '•• r^i> >• o =« c3 C r—t ^-^ o Ph 1-H o <^ •2 .2 1— 1 m CO ^ "5 !:- fc£ o t, c3 &£— ' JO (U ^ :^Tc o o s !>>-S • ^^ «4-l a, C OJ (-1 cj £^ ^M K o m ^ t^ C5 «^ o c3 '^ )I2\ © ® s;^ $) ',•% -^ 's ^ o c « o r^ "3 r o O " o CO 2<.i t^ DEVELOPMENT PRIOR TO LAYING 55 The margin of invasination is known as the hp of the blasto- ] pore or primitive mouth; the space between this margin and the volk is the blastopore, and the space between the entodem -^ and yolk, derived from part of the subgerminal cavity, is tlie^ archenteron or primitive intestine. Fig. 27. — Diagrammatic reconstruction of the blasto- derm of which a longitudinal section is shown in Fig. 26. C-D.. Plane of Fig. 26. ■ - ■ i- , G. W., Germ-wall 1, 2, 3, 4, 5, 6, and ' >™lioate regions of the blastoderm which are approximately from 1 to 7 cells deep respectively. The broken line around 1 indicates the region where the blastoderm is approxi- mLtely one cell d?ep. x 27.2. (After Patterson.) The first stage in the formation of the entoderm is interpreted as involution of the free margin, and this view is supported by the fact, determined by Patterson, that the antero-poster.or diameter of the blastoderm is shorter than the transverse diameter during this process, whereas previously the blastoderm was approximately circular. An even stronger support of this view is furnished bv experiments which demonstrate that injuries to the margin made just prior to gastrulation appear later in an "X ^ .■ * 1..' ••^» • * ----o ^. oi r o fl . S ■^ ^ ^ O ^ ^ ^ ?. (2 -t^ ^ 03T3 —- o a; K^_^ -4— > ro.S ^ ;i3 > , c ci ^H"^ C^ C3 CC t<-c fcJC s S c.§ id d ^C rt c b£ ..S r p ^^-^o"^ r^ C > ^ ■+J cS .S ^ 45 <^ bJCC CO +2 — ^ ^ tn S '^ ^ S 2i 'T3 ^ s- S Oj r^ _; i- — < TS ^£:S 3^ B^IS' '3 C.*- c-7 Li-i >> fc- f^ b£ -^ b 0) c ^i-i o5 >--3 •r; *-^ t« "^ ojf^.S bC -C bX) fl -*-^ ■^ fc^ .J-H o3 ■*-^ '-5 CO >0 ^ cu Ts £W A S ^< S c3 > 1> t- 0; S^ c c -■ _c ^-r- s- Ci «^ 0^.2 CQ oWS -t^ • p— t •" Ji c c3 ^ «-(-l ti-i 0) ^ '. ° ^-^o^ ■& u-l ^ £i^ . »— 1 ja >< h-1 '^^ ^ °i> 00 • 1— 1 ^.t^ . (N r^ r- ci-Q , ^PlhW o'p: DEVELOPMENT PRIOR TO LAYING 57 anterior position in the entoderm (Patterson). But after the margin has thickened the farther extension of the entoderm is due, largely at least, to ingrowth from the marginal thickening. Patterson also believes that the thickening of the margin is due not so much to multiplication of cells in situ as to immigration of cells from the sides. This view is also supported by experi- ments. ^ E G Fig. 29. — Diagrammatic reconstruction of the blastoderm of a pigeon's egg, 36 hours after fertilization; from the same series as Fig. 28. X 27.2. (After Patterson.) E., Invaginated or gut entoderm. O., Margin of overgrowth. PA., Outer margin of pellucid area. R., Margin of invagmation (dorsal lip of blastopore). S., Beginning of yolk-sac entoderm. Y., Yolk zone. Z., Zone of junction. , ,. . . The arrows at the posterior margin indicate the direction of movement of the halves of the margin. The circles in the pellucid area indicate yolk masses in the segmentation cavity. Figure 29 is a reconstruction of a blastoderm in the stage of Fig. 28, that is at the height of gastrulation. The margin of overgrowth (cf. Fig. 28 O) is represented by the area O; the zone of junction by the ruled area Z; the inner portion of the -o o ' c3 .2 o 'T3 O l-y-^ :9 ; ■k: t^i O ^■■^■M xn u o 00 CO bJD fcJD O a; O a, o m o _biD 'q o **-M '- fc^ 03 «4-l %-H t< O c3 o; s -1^ m CD -rs ci M ^ o o ^Ph M r-H o c3 13 >< *S xn -^ ID o3 O <4-l o 03 o o • 1— 1 -:3 o o (1) o OQ • r ^ "■+3 ^^ Zj > "r c3 CO ■^ "^^ C ^ -*-3 "^ 'tJC o 'so ^ o c X/1 -^'s .S c3 ~ fcl -5 Ch '^-4^ ^ o , " o s o ^-s d CO 73 C 03 PQ 73 d t^ >< ^ << ° TJ K S >m». .1 b«in (Dunl) ■-'-^ MtdulUrr lold> m«(iiv{ in rrHw o( mld-tninj ld dxpcT A and? ,„ .- ..,«=. ",63, ■i>;Dmi. McdulUnloUiMlni FoiF-cui about .1(1 mm |ll™j i.l«d> behind em » ... .s l^ion. ni..S" •<•■»■■ MedulluT feldt ckMl .boul 0., mm. "ass- SJ.T" Fm-pil about o^ ram Fore-iul sbtrul 0,4 mm Fon.-«ni do«d .bou |ll.»ll« o( brJn luplurini Detwan 17 and 18 j>ft i.S „,™. "(utS) FintUuJIaUoDcJcroBul 3ll(hi mnii) Bam os» mm. ThfMprimar): brain i«l da ctcuir IndlQiial brain tnick dull on [mini of dulDC Opi.ni.Bu UUuo SUfbl nm Op>. TO Fore-gut doKd abou Fore.«ul doKd abou Hart^ttnl dightly lo Heart beBI 10 fighl Bctna IS aadift a If s to Rim. t.;>iin.l»u«ninlUk(n itifhl cnnul OuuR lad bctiu ID lum OB licdlr tl-cdPNiuro- anmxly toukoj indiaud Coniulnion It LOKi'l frod. ■bun SI llihtd Slullon (il pit ''",s",{S3 Mrs Fore-cut dosed 14 mm. Heart beM lo right He^ tli|hily Sihiped "polbfST""^'™ 1, s;. 14 S Ukm ot HoiTO "(USJII) Uk. MS Ukc.jS Uke ijS Ukc>5S Fore-giii doKd I.J mm. Uke ■) S Kead-folfl cotfri enlire loce-brainrecion " a>« >JS — ■ nlcriu p>fl or hEul bAlI Ukc'tS. RwEilhiDd Prifflirv ™itfc •Ulkcd UU .5S "s^Md viKciiii' ddi imliciied cn.nuUj S-.h.ii« ol hart ilightly llad-fokl coven to ollc """"» .0 ■BS (,. mm. «raia l.Wh nlirr hod n«m IhM lull lumod on Ml adc bon reooD in Inillud -onilPicilnn ippnn be- l»cTn dl- uul ulin' apLiloa "ta UkosS Lcnmh of lofc-gut 1.1 ro«i^.l'lij'd!^ Sahapc ol heart recy ttijlht end Idl auifda lladlold oven anterior (un ol hind brain Mandibulai uches begin lo project " "- .(r.»S ,tmm.pc*lo>k0(1h (UurLl) ad lulk lureid lo lelt- Cranul sod Hniol llnuns rouDded ob- Diilloit tnluxemcnl ol "s S^U? o( dMp pit No Inditalioa ol hyp ophyii. Lrngth ol lore-pil I.I mm. Hm dcit oaily thiouch \uriiuU> nolricutar •iiul indicilHl bv mn Uke IDS " " «S end 't'"^ ir«in-hipc la bnila ConsUitlioD Hypoijy^ barely iodi- Length oll-(. t.< mm ta & Id delta naily IhrouEb 3d .IK. pouch «B VcniridilM loop noir isite Surfda^walraUj^tu™ 8ih somite (doe t rapid groinb of am Dion?) j6-j; oS ..s -" "■"■ £'SF""°" :st,f »I1K««S Oorinoiwol Ukc !■ S UkfijS ^.0thtn.i«KiiI'^ Like II S Had-lold CTDsa id vU- ccral pouch and >d « il. J64 .)S . J mm. imlrM iFniih .«Ti»l (lout. uiKte ol q uil bud .... ^.iKinolnl Moulh ol |>l SS'SS""'^" LcoEth of lore-gut 1,1; ?^rly"UI^I^''S upptt BIlKtC Uko 11 S ^^t^ """ '" "'"' TiU-lold barely indialed i&'iii l''i S^ '*■*' ^ ,) mill. «r«i«l koxlh -irvlcalllnurelike jiS. Cnnlal Ijccnma IcH Hun ■ H«bl anvil! nl^ bnd CIonHridlan bEi««n dl viiioni of bnln mim Mnuih ol iJl'uo^n'' Rdntiiely BalioaioNo,jSi(>5S) IIOD iO ItOBt Ol Hul pbH L«ng'h ol fore-pit oboul \sm 111 v.niridc vcniial r-o"'\'^mpi'ei"'" aurilc "-Jit""--"' \olflil-lold *' "* »o s .« mm. fmioi IcnVh "(dSS) ncTBK of J S Same «..-.,& aij/a-^/hild*""" Mead fold covm 10 i;lh Tail lold indialed "• 1MB mm. r«l**> Iniih H boun cphjiUc fluurc ilithll) ka ibia inncdinc Mvuih ol loulm.1 Vtr, aaaU "So"" ■;oepiphi^ iFon veif laD< uid d Htccrnl pouch weU dmebped. Thyroid and luni dinilicuta .climated atiho. Iliild ncaity Hud-folil conn lo loUi Hiad-gut !«mi a bo; jiS •KS .■mm.|ralatlcni lomu illlhl evxinaUon SuDtujyS d vimral pouch asumea elongiled lorm Three canplele aortic Tall-foU formed, con Forms small >rlde open pwicb " " >liS ^ mm. iraioi mou Ukc M S Like iR i .... Jkcjxs \-o tppbydi Simt u J7 Likens. OralmFmbnne unUokcn Thm complele aortic UthH Budiofblnd dialed by picuic Had-lold coven U> »lb umile. Small [aU'lokl UkeiSS joS .bmn.«reilnltcniih nCTiuc ol CRticol Qa urr mikn inilt nf falDd hniD >od trunk .Miuitb > rtihi SDflc r-HuWcilon ol inbffiu -t icUii Utei •! oiml(« SJ"'" Dlic midei UkciSS null .pipliyu. di.UUy I'liunli tuc. pouch indi- cltd. Clodni mem- brane of 3d very thin, Ibin Three compleie aortic archn. Fourlhlncom- plele Biid> of [pre limba'"i^ dieted by lUdteoing Head-fold coven lo iSIh somile. SduU lalMold Allanloii calendlim Inli veniral manBUrj jiS aram Mtkull "nitnl flaura ■ lull riibi >Dtle Ion conilhtlioo sC ulh Kcllulliyei ■hiikci. Jlk vedcic doKd Ux» |lll mall epiphy»U Hypnihyv-ilinnglaitlan Same as joS Oval openiog Inlo .mcd h^Iim ?aVb^o"l^''b? hind ]IU Mmlie Same as jo 46 ro5 j*S .fmiii.nMkUl] 4 mm. nxck iiU "CX""-"^ =^si'""*^" umt u JI "i£\i'"n"Sl Ihio ,■ S) tuUow jiii m.U (plpbyiU itIBt U ]l sanieaj js itmea.,. Umb budi «'3liii^" Amniotic umtlllcvu al most entirely dart Mesoblaslofollanloi.be ■iiu lo aland laio body wvily Curving of tall and In aojS ol cer.ial Da ure accouoi lor de- crease in lepglh *" dioM. Inhmuj rit ■iuplj mvked 5.™„„ Mine as zi Sameaajo Umb buds Aniiiiailc umbilicus equal to widlb ol ii "fSiB Meublasl ol allanlniB be- Vo curving of .all "' »s .1 mm. nrck.likll. ■.« mm. Immld bnin Iriln Hne niU [on- Dmbbudi cloulr indlnlKl iraldc it hiKlv India inl JkejiS DlUil CDluiemcnl of ulu drptcMloD daily Oral membrane lucihcr rupiuinl. Thyroid di- venJculum dwd' Fourth via. noucb iharpty bounded pot- Founb pcudl dearly kco Same u^ 11 S lo^^dbmetre of ,.s Uke 3> S Cucvlag ol Ull begun , " *tS S mm. DAk Uil > t daily lodiuUd OfCinnlwol cndolrm- 11 1> l»m I.Eil.^c.1 InncUu- UtkotRjl dUUll* (Oil >U|hili|«dur>l*idu ShoniDlnt ol [nva^ai lion ol hTP. •^"•^ "I.S^ff, Founh aoDc utii lonned Cldsed begun Tip of laU begjns to be fUR^ed lonrardl IM 41 S « mm. wk Ull •« mm. rn J. II noi. ri:».mld Into *'!S!n'* "^.-.siirru -mhral bflnliit«a unailr Indiutctl Endot. dud ipbrrinl lomidrb- nllc fttf ieclloci UkejT "Kt^Codbi-^i '■&),=' ""■' \Kl.^l«id. Oart ayilt ol allalilol. b(- DtsulrtUrirmnHouko ■bote altantois Oaik. shaped '!» ., CHAPTER IV FROM LAYING TO THE FORMATION OF THE FIRST SOMITE I. Structure of the Unincubated Blastoderm There is more or less variation in the stage of development of unincubated blastoderms; in exceptional cases these variations may be extreme. However, the usual condition may be described very briefly as follows (see Fig. 34): Beneath the pellucid area is the subgerminal cavity bounded marginally by the germ-wall. The posterior part only of the pellucid area is two-layered. The lower layer or gut-entoderm terminates posteriorly at the germ- wall, with which, however, it is not united. It is composed of spindle-shaped cells which form a coherent layer, perforated by numerous small openings that appear as breaks in the layer in section. In front of the gut-entoderm a few scattered cells appear in the subgerminal cavity. The gut entoderm does not reach the germ-wall either laterally or anteriorly, but in the course of a few hours' incubation it spreads so as to unite with the germ-wall around the entire margin of the pellucid area. The germ-wall is slightly thicker at the posterior than at the anterior end, that is to say, that the nuclei extend deeper into the yolk (Fig. 34). There is a broad zone of junction and beyond this the margin of the blastoderm overlaps the yolk a short dis- tance. The germ-wall has not yet become organized as a layer separate from the yolk. The ectoderm is thicker in the region of the area pellucida~\ than in the area opaca; and slightly thicker in the center than at the margin of the area pellucida. This thickening is in part ^ the forerunner of the medullary plate. II. The Primitive Streak Total Views. The primitive streak appears early on the first day of incubation as an elongated slightly opaque band occupying 69 70 THE DEVELOPMENT OF THE CHICK N t :.:'■- >^^«* iif •:;••• ••••5 r .»5 f-^ I'; s is; \:-;^- S .b£ o .2 0; r, C > O c -r o ^ .2 H 3 H Z o O _:2 C3 r- o o a; o o . ■gtsj W r oj >.> > P O -V, -p.-T^ o fl a; -i 2 ^ 'So cu C 'Ts C o "B .2 CO Q, 6 ^ r o >^ O S3 cj < o ^^ FROM LAYING TO FORMATION OF FIRST SOMITE 71 the posterior half or two fifths of the circular pellucid area (Fig. 35 B). It is relatively narrow in front and widens posteriorly, where it is at the same time less dense. Its anterior end usually does not quite reach the center of the pellucid area. It rapidly increases in length; the anterior end appears to be practically a^ fixed point, and growth takes place posteriorly probably not by I addition, but between the two ends. The posterior half of the pellucid area elongates simultaneously, keeping pace with the Fig. 35. — Surface views of two stages of the blastoderm of the egg of the sparrow. (After Schaiiinsland.) A. Before the appearance of the primitive streak. B. The first appearance of the primitive streak. a. o., Area opaca. a. p., Area pellucida. Ent. Th., Thickening of en- toderm, pr. str., Primitive streak. primitive streak which lies entirely wdthin it in the chick and most other birds. Thus the area pellucida becomes oval, then pear-shaped, and the primitive streak bisects the greater part of its length (Figs. 35, 36, 44, etc.). According to KoUer the primitive streak takes its origin from a crescentic area at the posterior margin of the pellucid area, which he terms the sickle. The primitive streak appears as a process extending forward from the center of the sickle, and, as it growls forward, the lateral horns of the sickle are gradually taken into its posterior end. Roller's observations and interpretations have not, however, been con- firmed by subsequent investigators and they would appear to rest on rather exceptional and inessential conditions. 72 THE DEVELOPMENT OF THE CHICK a.p. Jfes. Ent. TJ). ■prp Ent.Th. -pr.gr. -pr.f. s.^r Fig. 36. — A. Intermediate stage of the formation of the primitive streak of the sparrow. (After Schaninsland.) B. Fully formed primitive streak of the spar- row. (After Schaninsland.) a. o., Area opaca. a. p., Area pellucida. Ent. Th., Thickenino; of entoderm. Mes., Mesoderm, pr. f., Primitive fold. pr. gr., Primitive groove. pr. p., Primitive pit. pr. str., Primitive streak, s. gr., Sickle groove. At first the surface of the primitive streak is even, but, as it elongates, a groove appears down its center. This groove is I known as the primitive groove ; it is bounded by the primitive \folds and terminates abruptly in front in a pit, the primitiv^e (pit, which corresponds to the neurenteric canal of other verte- FROM LAYING TO FORMATION OF FIRST SOMITE 73 brates (Figs. 35, 36, 44, etc.). The primitive groove does not involve the extreme anterior end of the primitive streak, which forms a Uttle knot in front of it, the primitive knot (''Hen- sen's knot"). The posterior end of the primitive streak termi- nates in an expansion which is not very obvious in surface view, and hence is not usually described; it may be called the primitive plate (Figs. 36, 44 A, 44 B, etc). In some cases the primitive streak and groove are bifurcated at the posterior end (Fig. 44 B). The primitive streak is the first clear indication of the axis of the embryo. The neurenteric canal is a canal that connects the posterior end of the central canal of the neural tube with the intestine. It arises from the anterior end of the primitive mouth, and is typically developed in Selachia, Amphibia, reptiles, and some birds {e.g., duck, goose. Sterna, etc.). It begins in the primitive pit and extends forward into the head- process (p. 80). Subsequently the primitive pit becomes surrounded by the medullary folds, and thus opens into the neural canal. An opening is later formed through the entoderm so that the definitive canal connects neural tube and hind-gut. In the chick the neurenteric canal is never typically developed. Usually it is represented only by the primitive pit. In exceptional cases I have found traces of it in the head-process. The so-called head-process appears in front of the primitive knot (Figs. ,36 B and 44 B). In surface view it appears not unlike the primitive streak itself, but is fainter and less clearly defined. It is continuous with the primitive streak at the primitive knot, but its axis is usually a little out of line with the axis of the primi- tive streak. Figs. 35 and 36 exhibit four stages of the development of the primitive streak of the sparrow (after Schauinsland). The darker area in the anterior part of the area pellucida is caused by a thicker region of the entoderm which in the course of time becomes of uniform thickness with the remainder. It will be ob- served that the primitive streak arises entirely within the area pellucida (Fig. 35 B). In later stages its posterior end is bifurcated (Figs. 36 A and B), and we have the appearance of a sickle some- what similar to Koller's description for the chick. The primitive groove begins near the anterior end of the primitive streak in an especially deep pit just behind the primitive knot, and extends back the entire length of the primitive streak into the horns of the sickle. The head-process is barely indicated in Fig. 36 B. 74 THE DEVELOPMENT OF THE CHICK The later history of the primitive streak is illustrated. in Figs. 44, 51, 61, 65, etc.: the embryo arises in front of it around the head-process as a center; the anterior end of the primitive strealH marks the hind end of the differentiated portion of the embryo^ As the embryo grows in length the primitive streak decreases (cf. measurements in table), until finally, when the completion of the embryo is indicated by the formation of the tail-fold, the primi- tive streak disappears. The primitive knot and primitive pit occupy its anterior end at all stages, and, as the embryo differen- tiates from the anterior end of the primitive streak, the primitive pit must be regarded as moving back along the line of the primi- tive groove, always representing its anterior end. Sections. The preceding sketch of the superficial appearance of the primitive streak must now be followed by a careful exami- nation of its structure and role in the development. **»~"'r|@S»iiS. YiG. 37. — Three sections through the primitive streak of a sparrow at a stage intermediate between Figs. 35 and 36. x 230. (After Schauinsland.) A. In front of the primitive streak. B. Through the anterior end of the primitive streak (primitive knot). C. About through the center of the primitive streak. All recent authors are agreed that the primitive streak owes its origin to a linear thickening of the ectoderm, from which cells are proliferated between the ectoderm and the entoderm, forming a third layer, the mesoderm. Figs. 37 A, B, C show three trans- verse sections through a blastoderm of the sparrow slightly more advanced than the stage shown in Fig. 35 B. The first section is just in front of the primitive streak. The ectoderm is thick in the center and thins gradually toward the margin of the area pellucida, becoming decidedly thin in the region of the area opaca. The thin entoderm of the area pellucida unites peripherally with the thick yolk-sac entoderm of the area opaca. The second FROM LAYING TO FORMATION OF FIRST SOMITE 75 section -passes through the anterior end of the primitive streak; the ectoderm is greatly thickened (primitive knot); the base- ment membrane is interrupted below, and the lowermost cells are becoming loose. The third section is through a more pos- terior portion of the primitive streak. The proliferation from the ectoderm is more extensive, the cells are looser and are begin- prJ-^.^. t^' fct'^mw^ ^ ' ^^-€sf^. % Mes. >J!S^ Fig. 38. — Transverse sections through a very short primitive streak of the chick. Incubated 17^ hours; no head-process. A. Through the anterior end of the primitive streak (primitive knot). Mesodermal cells are being proliferated from the ectodermal thickening; some are scattered between the two primary germ layers. The entoderm shows no proliferation, though some mesoderm cells are adhering to it. B. Fourteen sections posterior to A. (Entire length of the primitive streak is 80 sections.) The mesoblast wings are forming; the primitive groove and primitive folds are indicated. The entoderm is free from the mesoderm. Ect. Ectoderm. Ent., Entoderm. Mes., Mesoderm, pr. f., Primitive fold. pr. gr. Primitive groove, pr. kn., Primitive knot. ning to spread out laterally. The entoderm is a continuous membrane without any connection with the primitive streak, and there are no cells between ectoderm and entoderm save those derived from the primitive streak. Figs. 38 A and B show the structure of the primitive streak [ 76 THE DEVELOPMENT OF THE CHICK of the chick at a more advanced stage, but before the formation of the head-process. Sections in front of the primitive streak show no cells between ectoderm and entoderm. In the region of the primitive knot (A) the ectoderm is greatly thickened, forming a projection above and below. Cells become detached from the lower surface of the ectoderm, and are converted into migratory cells between the two primary layers. Immediately behind the primitive knot the primitive groove begins abruptly; it is the seat of active proliferation from the lower layer of the ectoderm, and the cells migrate out laterally forming wings of cells, which do not, however, reach the area opaca (Fig. 38 B). Conditions are very similar along the entire length of the primitive streak at this time; but near the posterior end a few cells of the mesoderm reach the area opaca and begin to insinuate themselves between the ectoderm and the germ-wall. There is no evidence at any place that any of the mesoderm cells are derived from the entoderm. The axial thickening of the primitive groove comes in contact with the entoderm and appears in places fused to it. Figures 39 A-E represent five sections through the head-process and primitive streak of a chick embryo at a time when the head- process is still very short. The first section through the head- process is described beyond. B is through the primitive knot; the ingrowth of cells is more extensive than in the preceding stage and it will be observed that they are now fused with the entoderm, so that the latter no longer appears as a distinct layer. C is through the primitive groove near its anterior end. D is a little behind the center of the primitive groove, and E is through the primitive plate. Behind the center of the primitive streak the entoderm is again free (D). It will be observed that the area of proliferation in the primitive plate is very wide. Fig. 39. — Five sections through the head-process and primitive streak of a chick embryo. The head-process is very short. A. Through the head-process, now fused to the entoderm. B. Through the primitive knot. C. Through the anterior end of the primitive groove. D. A Uttle behind the center of the primitive streak. E. Through the primitive plate. The total number of sections through the head-process and primitive streak of this series is 102. B. is 4 sections behind A. C. is 12 sections behind A. D. is 59 sections behind A. E. is 87 sections behind A. Ect., Ectoderm. Ent., Entoderm. G. W., Germ-wall. H. Pr., Head- process, med. pi., Medullary plate. Mes. Mesoblast. pr. f. Primitive fold, pr. gr., Primitive groove, pr. kn., Primitive knot. pr. pi., Primitive plate. medp/. .'•;• .?.»■ A.«_<»-- «-•• B ^-Y^ "■«^ Z''^/'/ '^^«^'- • a«*t *»i.^_^' L 78 THE DEVELOPMENT OF THE CHICK The mode of origin of the mesoderm of birds has been a very puzzhng question as is proved by the numerous views that have been in vogue from time to time. One of the earhest views was that the mesoderm arose by sphtting of the primary entoderm (Remak). This view sur- vives in part even at the present time (mesoblast of the opaque area). Balfour beheved that the mesoblast in the region of the embryo "ori- ginates as two lateral plates split off from the primitive hypoblast," and that the primitive streak mesoblast is extra-embryonic, or at most enters into the formation of mesoblast of the extreme hind end of the embryo (allantois mesoblast in part). This view is found in the "Elements of Embryology" of Foster and Balfour. A third view, now of historical interest only, was that the mesoblast cells arose peripherally and mi- grated between the two primary germ-layers (Peremeschko, Goette). The latter author even attempted to derive the primitive streak from an aggregation of such inwanclering cells. The view that the primitive^ streak arises as a thickening of the ectoderm and that it is the source of all the mesoderm was first stated by Kolliker, and has been accepted by Hertwig, Rabl, and many others. It may, indeed, be regarded as definitely established for the embryonic mesoblast. Others, however, believe with His that the mesoblast of the opaque area arises by delam- ination from the germ-wall ; this question is discussed beyond. It should also be noted that it is probable that the primitive embryonic mesoblast is supplemented in certain regions at later stages by cells proliferated from both entoderm and ectoderm, particularly in the region of the head. (See pp. 116, 117.) In early stages of the primitive streak the mesoblast cells are relatively sparse and bear every appearance of migrating separately. But as the ingrowth progresses and the cells become more numerous, the mesoderm becomes converted into coherent plates. These are wedge-shaped, the central broad ends fused with the primitive streak and the narrow margins extending laterally (Figs. 40 A, B, C). They soon overlap the margin of the opaque area and thus is produced a three-layered portion of Fig. 40. — Three transverse sections of a late stage (corresponding to about Fig. 44 B), through the head-process and primitive streak of a chick embryo. A. Near the hind end of the head-process. B. Through the primitive pit. C. A short distance behind the center of the primitive streak. The region between the lines A-A and B-B is represented under a high magnification in Fig. 41. Bl. I., Blood island, coel. Mes., Coelomic mesoblast. Ect., Ectoderm. Ent., Entoderm. G. W., Germ-wall. med. pi., Medullary plate. Mes., Meso- derm. N'ch., Notochord. pr. f., Primitive fokl. pr. gr., Primitive groove, pr. p.. Primitive pit. FROM LAYING TO FORMATION OF FIRST SOMITE 79 >i A» a i«^ • f. ^ ^ f S.pTe.r.Jt/'wJj •• I' . '0 << ♦'•• ••♦. * o 5^" 80 THE DEVELOPMENT OF THE CHICK the latter which corresponds to the future vascular area. The mesoblast grows out, not only from the sides of the head-process and primitive streak, but also from the hind end of the latter, that is from the primitive plate. The mesoblast thus extends into the opaque area behind the embryo at a very early stage (Figs. 42 and 44). This part of the mesoblast is homologous with the mesoblast of the ventral lip of the blastopore of reptiles and amphibia, and, like it, is the first place of formation of blood. The primitive groove must be regarded as an expression of the forces of invagination of the mesoblast, and the primitive folds as the lips of this invagination. .'■- "^^•-■ff* « oC Q ©• Fig. 41. — The part of the section shown in Fig. 40 C, between A-A and B-B more highly magnified. Abbreviations same as Fig-. 40. ^fe' The Head-process. Two stages of the head-process are shown in tranverse section a short distance in front of the primitive knot in Figs. 39 A and 40 A. It consists of a thicker central mass of cells with lateral wings; the central part, or primordium of the notochord, is continuous posteriorly with the axis of the primitive streak (Fig. 42); the lateral wings are mesoblast and they are continuous posteriorly with the mesoblast wings of the primitive streak. The head-process becomes inseparably fused with the entoderm in the middle line immediate!}' after its forma- tion; and this fusion is continued back along the axis of the primitive streak (Figs. 39 and 40). The fusion is particularly intimate and persistent at the extreme anterior end of the head- process; behind this point the notochord and entoderm soon sepa- rate again in the course of development. But the anterior end FROM LAYING TO FORMATION OF FIRST SOMITE 81 >>.JL 82 THE DEVELOPMENT OF THE CHICK of the notochord remains attached to the entoderm for a consid- erable period after the formation of the head-fold. A longitudinal section shows the head-process as an appendage to the anterior end of the primitive streak, or the primitive knot (Fig. 42). m.n. Fig. 43. — Diagrams to illustrate the theory of concrescence as applied to the primitive streak of the bird. The central area bounded by the broken line represents the pellucid area; external to this is the area opaca, showing as concentric zones the germ-wall (G. W.), the zone of junction (Z. J.), and the margin of overgrowth (M. O.). m. n., Marginal notch. For de- scription see text. The most obvious interpretation of the head-process is as an outgrowth from the primitive knot. But another, and more proba])le interpretation in view of all the facts, is that the head- process is a later stage of the anterior end of the primitive streak; FROM LAYING TO FORMATION OF FIRST SOMITE 83 that a gradual separation of the ectoderm takes place in the axis of the primitive streak beginning at the anterior end, and progresses posteriorly. That part in which the ectoderm is separated represents the head-process; it has therefore the same composition as the primitive streak, except that the ectoderm has become independent. Interpretation of the Primitive Streak. The discussion of the significance of the primitive streak involves two parts: (1) its morphological significance, and (2) its role in the formation of the embryo. The first question involves knowledge of comparative embryology, which is not assumed for the purposes of this book, and it will therefore be considered very briefly. The fundamental relations of the primitive streak must define its morphological interpretation; the first thing to be noted is that the germ-layers, more especially the ectoderm and mesoderm, are fused in the primitive streak; second, the differentiated part of the embryo is formed in front of it; third, the neurenteric canal occupies the anterior end of the primitive streak; fourth, the anus forms at its posterior end. Now these characters are exactly those of the blastopore or primitive mouth of lower vertebrates, that is of the aperture of invagination of the archenteron. For these reasons, and because in all other essential respects the primitive streak corresponds to the blastopore, it must be interpreted as the homo- logue of the latter. It is to be regarded, therefore, as an elongated blastopore, and the primitive groove as a rudimentary archenteric invagination. This interpretation raises the question as to its relation to the original marginal area of invagination of the entoderm. Can these two things be really different stages of the same thing? The concrescence theory gives a theoretical basis for their iden- tification. It will be remembered that the margin of invagina- tion represents a small section of the margin of the primitive blastoderm in the pigeon, and, by inference, in the chick also. The remainder of the margin w^here the zone of junction persists is the margin of overgrowth. Now we assume that the closure of the original marginal area of invagination proceeds by con- crescence or coalescence of its lips, beginning in the middle line behind, thus producing a suture which is the beginning of the primitive streak. Let the above circles (Fig. 43) represent the blastoderm in four stages of closure of the original area of invag- 84 THE DEVELOPMENT OF THE CHICK ination. The shaded margin represents the zone of junction, the unshaded portion of the margin represents the area of invagina- tion of the entoderm. The dotted contour represents the margin of the pelkicid area. In A the middle of the area of invagination is marked 1, and corresponding points to the right and left 2, 3, and 4. In diagram B it is supposed that the margin of invagina- tion is turned forward at 1, and that the lateral portions are brought together as far as 2, thus producing a suture in the middle line 1-2 continuous with the margin 3-4. The zone of invagina- tion is correspondingly reduced in extent and the zone of junction increased. In diagram C the lateral lips of the zone of invagina- tion are represented as completely concresced, thus producing a median suture 1, 2, 3, 4, extending through the posterior half of the area ppllucida to the margin. The zone of junction is on the point of closing behind the line of concrescence which is the primordium of the primitive streak. In diagram D, finally, the opaque area has closed in behind the line of concrescence which occupies the hinder half of the pellucid area. To apply this theory to the actual data of the development, it is only necessary to assume that the entoderm separates from the ectoderm along the line of concrescence, and that the primi- tive streak arises subsequently along the same line. The actual demonstration of the truth of this conception cannot be furnished by observation alone, however detailed. It is, however, possible to test it by experiment, though difficult because the concrescence must take place, if at all, prior to laying. The strong support of the theory lies at present in the data of comparative embry- ology; in the lower vertebrates the mesoderm and entoderm are both formed from the margin of invagination. Summarizing the matter, we may say that in the chick gastru- lation is divided into two separate processes: the first is the in- vagination of the entoderm from the margin, and the second is the ingrowth (or invagination) of mesoblast and notochord from the primitive streak, which represents the coalesced lips of the margin of invagination; the primitive groove is therefore the expression of a second phase of invagination. The genetic relation of the primitive streak to the margin of the blastoderm is well illustrated by an abnormal blastoderm described by Whitman in which the primitive groove was con- tinued across the area opaca to a marginal notch at the posterior FROM LAYING TO FORMATION OF FIRST SOMITE 85 end. A similar marginal notch at the hinder end of the blasto- derm in the line of prolongation of the primitive streak has been described also by His and Raiiber, but in the cases observed by them there was no connection with the primitive groove. It suggested to them, however, the idea of genetic connection between the two, and was used as argument for the derivation of the primitive streak from the margin by concrescence. The second question concerning the primitive streak, its role in the formation of the embryo, may be answered very briefly by saying that it is itself the primordium of the greater portion of the axis of the embryo; some indeed maintain that it represents the entire embryonic axis excepting the short pre-chordal part (Kopsch). The view of Balfour and Dursy that it takes no essen- tial part in the formation of the embryo, but atrophies as the embryo forms, is now of historical interest only. The question is how much of the embryo is represented by the primitive streak. But this question is by no means easy to answer, and there is no complete agreement in regard to it. The one point that is definitelv settled is that the anus arises at the hinder end of the primitive streak; but what point in the embryo corresponds to the anterior end of the primitive streak, or, in other words, how much of the embryo is laid down in the blastoderm in front of the primitive streak, is a disputed question. The attempt has been made to solve the problem by destroying the anterior end of the primitive streak by a hot needle, or by electrolysis, then sealing up the egg and permitting it to develop farther and finally locating the resultant injury in the embryo. But, while one worker finds the injury at the anterior end of the notochord (Kopsch), that is in the region of the fore-brain, another finds it in the region of the heart, that is in the hind-brain (Peebles). The reasons for this discrepancy in results are two : (1) the methods employed are not sufficiently exact, and (2) it is difficult in the living egg to determine the exact location of the anterior end of the primitive streak, and sometimes even to distinguish it from the head-process. Owing to the extremely rapid growth of all parts of the embryonic axis, a minute division of the primitive streak becomes a relatively long part of the embryonic axis in a very short time. It is obvious, therefore, that the slightest deviation of the injury from the point aimed at may lead to considerable error in the results. Until embryologists operate 86 THE DEVELOPMENT OF THE CHICK with instruments of greater precision one cannot feel certain of the results of such experiments. III. The Mesoderm of the Opaque Area We have seen that the mesoderm arises from the sides of the head-process and the primitive streak, and grows out between the ectoderm and the entoderm to the margin of the pellucid area; it then begins to overlap the opaque area at first behind, later at the sides, appearing between the ectoderm and the germ- wall. Figs. 44 A, B, C, and 45 illustrate its peripheral extension; at first it spreads most rapidly behind the embryo, but soon ex- tends with equal speed opposite the primitive streak, and thus a considerable portion of the area opaca becomes three-layered, consisting of ectoderm, mesoderm, and germ-wall (Figs. 40 C and 41). The contour of the anterior margin of the mesoderm iSa'^ first rounded, convex anteriorly (Figs. 44 A and B). Then the antero-lateral angles of the mesoblast begin to extend forward so that the anterior boundary becomes concave (Fig. 44 C) ; the lateral horns thus established continue to grow forward and ultimately meet in front of the head (Fig. 45) ; they thus bound a mesoblast-free area in front of and beneath the head, known as the proamnion, into which the mesoderm does not penetrate until a relatively late stage of development. Blood-islands (Figs. 44 C and 45) develop early in the three- layered part of the opaque area; appearing first behind the em- bryo, they rapidly differentiate forward opposite the sides of the embryo and follow the expansion of the mesoblast. This three-layered portion of the opaque area is known as the vascular area (area vasculosa) after the appearance of the blood-islands. It soon acquires a very definite peripheral boundary by the forma- tion of the vena (sinus) terminalis at its margin (Fig. 45). The two-layered peripheral portion of the opaque area is known as the vitelline area (area vitellina), and here again we distinguish two zones, an outer including the zone of junction, and an inner one (Figs. 32, 33). The first blood-islands are masses of cells lying on the germ- wall behind the embryo; the first blood-cells (erythrocytes) and blood-vessels arise from them, hence their name. Soon after their origin the blood-islands appear red owing to the formation of haemoglobin. Between the blood-islands and the ectoderm FROM LAYING TO FORMATION OF FIRST SOMITE 87 .<5l i^ o CO o X O 03 >iCl,C c3 O ^^ 8 ^ ^^ £ > ci -:' . .s . ,~ •— i_-i — c_i — ci--' O c 5=; S o '— £ o o !/: :5 -^ t- gcC £ -> £-:S S -c p X— = b C Cm X -* t: ■< . t- o^ c3 r S CI, X -^ o o !- -e ♦- t^ «4- ,^ h-" ^ g 2 ^ t: ;- . , ;v -^ w C ? rr ~; ^ "^ bf, -■ "^ ^ ^ 5 it — •— ^ aj "-^ c^ J Q -t:; .:- 88 THE DEVELOPMENT OF THE CHICK is a layer of the mesoderm (Fig. 41). If the blood-islands be reckoned as mesoderm we must distinguish two layers of the latter, yiz., a deep or vascular layer (angioblast) lying next the germ-wall, and an upper layer next the ectoderm, which may be called the coelomic mesoderm, inasmuch as the body-cavity (coelome) develops within it later. pre prstr d vase. Fig. 45. — Blastoderm and embryo at the stage of four- teen somites. The horns of mesoblast are on the point of meeting in front of the head. a. p., Area pelliicida. a. vase, Area vasculosa. a. v. i.. Area vitellina interna. Ht., Heart, n. F., Neural folds. pr'a., Proamnion, pr. str., Primitive streak. S. t., Sinus terminalis. There are two sharply contrasted views concerning the origin of the mesoblast in the area opaca. According to the one point of view it is simply a peripheral extension of the primitive streak mesoblast with which as a matter of fact it is continuous (Hert- wig, Rabl, and others). AccorcUng to the other point of view FROM LAYING TO FORMATION OF FIRST SOMITE 89 it is split off from the germ-wall (His and others). One thing is perfectly clear, viz., that the mesoderm of the opaque area arises in continuity with the primitive streak mesoderm; the second view would therefore be better expressed, as Riickert states it, that the primitive streak mesoderm grows in the region of the area opaca at the expense of elements of the germinal wall. If the cells of the primitive streak mesoblast be compared | with the cells of the forming blood-islands a sharp contrast is \ observed; the mesoblast cells of the area pellucida are devoid of yolk-granules; young blood-islands on the other hand contain yolk-granules of precisely the same character as those of the germ-wall (Fig. 41), which must have been derived from the latter. If the origin of the blood-islands be carefully traced, they are \ found to be rooted in the protoplasm of the germ-w'all; and prior i to the appearance of the blood-islands proper, protoplasm and nuclei of the germ-wall aggregate superficially in a manner that appears to foreshadow the blood-islands. Therefore, either the blood-islands are derived from the cells of the germ-wall, or cells of the mesoderm growing over the germ-wall burrow into \ the latter, engulf yolk-spheres, and reappear in masses as blood- islands. We shall not attempt to decide between these pos- sibilities. Another question concerns the origin of the layer of ccelomic mesoblast that overlies the blood-islands: is it derived from the primitive streak mesoblast, or is it split off from the blood-islands? When the latter first appear, in the periphery of the vascular area at least, there is no ccelomic mesoblast above them. It appears later, at first not as a coherent layer, but as scattered cells that rapidly unite to form a layer. In many places the microscopical appearances indicate strongly that the cells are split off from the surface of the blood-islands; but, as they are usually not far from the edge of the advancing co'lomic meso- blast, it may be that they are derived from the latter. Riickert states, however, that, in the case of some isolated blood-islands behind the embryo, a layer of mesoblast is formed over them Avhile they are still isolated. This would render the derivation from the blood-islands probable in such cases. It is possible, therefore, that the ccelomic mesoblast grows partly, at least, at the expense of the superficial cells of blood-islands. As rapidly as they are formed the various blood-islands con- 90 THE DEVELOPMENT OF THE CHICK nect and anastomose with one another, forming a vascular net- work lying between the coelomic mesoblast and the remains of the germ-wall. This network spreads throughout the vascular area, and appears later in the pellucid area, and communicates with the blood-vessels of the embryo (Figs. 44 and 45). In the next chapter we shall consider the manner in which the extension takes place, and the origin of the blood-vessels and blood-cells. IV. The Germ-wall The germ-wall arises, as we have seen, through infiltration of the superficial white yelk by the periblast. These cells mul- tiply and anastomose and form a multinucleated S3mcytium with the yolk-granules in its meshes. By degrees the protoplasm itself takes up the yolk-granules, Avhich are gradually digested, and the germ-wall thus becomes organized as a coherent layer. It then separates from the underlying yolk. The next period in the history of the germ-wall is its differentiation, which takes place in the vascular area concomitantly with the formation of the blood- islancls: a considerable proportion of the protoplasm and nuclei of the germ-wall accumulates at the surface and forms the vascu- lar mesoderm in the manner already described. The part of the germ-wall that remains after the separation of the mesoderm then differentiates into the characteristic entodermal epithelium of the opaque area, which is known as the yolk-sac epithelium (ento- derm) because it is destined to form the lining of the yolk-sac. After the formation of the vascular area the term germ-wall must be restricted to the lower layer of the vitelline area, because within the vascular area it has already differentiated into the mesoderm and yolk-sac entoderm. The development of the germ-wall takes place in a centripetal direction; at any period during the overgrowth of the yolk the three stages of the germ- wall may be found in the concentric zones. The first stage,- that of periblast, is found in the zone of junction (area vitellina externa); the second stage, that of organization of the germ- wall, is found in the area vitellina interna; and the third stage, that of differentiation, is found at the margin of the area vascu- losa. Within the latter area the differentiation is completed. CHAPTER V HEAD-FOLD TO TWELVE SO:\IITES (From about the twenty-first to the thirty-third hour of incubation) I. Origin of the Head-fold At the end of the period described in Chapter IV, the embryo is represented by a central differentiated area of the blastoderm, lying within the area pellucida, distinguished anterioriy by the medullary plate and head-process, and posterioriy by the primitive streak. The layers of the embryonic area are everywhere continu- ous with the corresponding layers of the extra-embryonic blasto- derm, with no clear line of division between the two. In the course of the second and third days the embryo becomes cleariy defined by its own growth, and by the formation of bounding folds. The delimitation of the embryo from the blastoderm begins immediately after the formation of the head-process by the for- mation of a fold at the anterior end of medullary plate known as the head-fold (Fig. 42). Seen from the surface, this fold has a semicircular outline, the concavity of which is directed posterioriy (Fig. 44). It involves both the ectoderm and entoderm. A later stage is shown in sagittal section in Figs. 46 and 47: the ecto- derm and entoderm immediately in front of the medullary plate make a sharp bend downwards and backwards, and then turn forward again. The head-fold thus produces an internal bay in the entoderm, the beginning of the fore-gut. There is similariy an external bay, the posterior angle of which is the head-fold proper, lying beneath the projecting head. These bays are of course turned in opposite directions, the internal one opening into the subgerminal cavity posterioriy, and the external one opening anteriorly on the surface of the blastoderm. The transition from the ectoderm of the medullary plate into that of the under surface of the head and the proamnion is a grad- ual one. The difference is, however, very strongly marked (Fig. 47). The formation of the head-fold is due to the more rapid 91 92 THE DEVELOPMENT OF THE CHICK a m >>7^ O c-^ o ;: li o o ffi ^ ^ ^ o c c = ^ s o o +i o 1:: "^ M CQ ':^ '^ -^ -■■■ -^ (0 ms^ f=H growth of the medullary plate, which causes the latter to extend forward above the thinner and more pliable membrane in front. The entoderm is attached to the inner surface of the anterior end of the medullary plate (Fig. 47), and is apparently carried forw^ard with the latter to form the anterior portion of the fore-gut. The actual form of the fold depends upon the mechani- cal properties of the membranes concerned, especially the unequal thickness of their parts produced by unequal grow^th. Although the head-fold thus ap- pears to be a single fold involving the two primary la3'ers, it is con- venient, for purposes of description, to consider it as two separate folds, ectodermal and entodermal. The deepening of these folds takes place at the same rate up to the time when four somites are formed (Fig. 49). At about this time the paired primordia of the parietal cavity (amnio-carcUac vesicles), which ap- pear in the mesoblast in the lateral extensions of the head-fold (Fig. 50), push in towards the mid- dle line so as to separate the ecto- dermal and entodermal limbs (Figs. 52 and 58). When six somites are formed, these cavities fuse in the middle line, thus effecting a complete separation of the two limbs. The further progression of the head-fold, after this union, takes place separately in the two limbs. \ HEAD-FOLD TO TWELVE SOMITES 93 11. Formation of the Fore-gut The extension of the amnio-cardiac vesicles between the ectodermal and entodermal layers of the head-fold introduces a section of the body-cavity (pericardium) between these layers and at the same time converts the ectodermal limb into a portion of the somatopleure, and the entodermal limb into a portion of the splanchnopleure. (See p. Ho.) The splanchnopleuric head-fold extends posteriorly very rapidly after the invasion of the body-cavity, while the somatopleuric fold apparently remains fixed for some time, though the head-fold appears to Fig. 47. — Head-fold region of Fig. 46 highly magnified . For abbreviations see Fig. 46. become deeper, owing to the forward extension of the head above the blastoderm. The posterior extension of the splanch- . nopleuric head-fold lengthens the floor of the fore-gut; it is caused by the median growth and concrescence of folds of the splanchnopleure (Fig. 53). Along with this process is involved the development of the heart described farther on. The growth in length of the fore-gut may be realized by a comparison of Figs. 50, 52,62, etc. Thus by the 12 s stage a considerable section of the fore-gut is already established (Fig. 63); this is the pharyngeal division; from the first it is extremely broad, and lunate in cross-section (Fig. 54), the floor being composed of columnar cells, and the roof 94 THE DEVELOPMENT OF THE CHICK of very flat cells. The lateral extensions may be regarded as diverticula; subsequently these grow more rapidly at four places along their length, and come in contact with the ectoderm. Thus four pouches are established on each side as described in detail /i.gr &o. ■£t^t. -prjr. Fig. 48. — Stage of first intersomitic groove drawn from an entire mount in balsam by transmitted light. a. c. v., Amnio-cardiac vesicle, a. o., In- ner margin of Area opaca. Ect., Ectoderm. Ent., Entoderm H. F., Head-fold, i.s.f.l., First intersomitic furrow, med. pi., Medullary plate. Mes., Mesoderm, n. gr., Neural groove, pr. gr., Primitive groove. Pr'a, proamnion. in the next chapter. At the 12 s stage one such place of contact is alreadv formed, Iving a short distance in front of the thickened ectoderm destined to form the auditory pit. HEAD-FOLD TO TWELVE SOMITES 95 Another place of fusion between the fore-gut and the ecto- derm is tlie so-called oral plate (pharyngeal membrane), which occupies a mid-ventral position at the extreme anterior end. The parietal cavities meet posterior to the oral plate (Figs. 67 and 75). Transverse sections show the oral plate to be depressed beneath the level of the ventral surface of the head at the stage of 10 somites (Fig. 55), a condition that increases, as development J7 /=" /Tff. ^IjO. £/)t. £-c/. Fig. 49. — Median sagittal section of the head at the stage of 4 s. a. i. p., Anterior intestinal portal. F. G., Fore-gut. Ect., Ectoderm. Ent., Entoderm. H. F., head-fold. Mes., Mesoblast. n. F., Neural fold, or. pL, Oral plate. proceeds, by the formation of the cranial flexture, and by the up- growth of the tissues behind and at its sides; thus will be estab- lished a deep depression lined by ectoderm, the floor of which is formed by the oral plate, and which is destined to form a large part of the mouth. The depression is known as the stomodseum. III. Origin of the Neural Tube The Medullary Plate. The medullary plate is the primordium of the central nervous svstem. At the time of formation of the head-fold it is broad in front and narrower posteriorly, ending opposite the posterior end of the primitive streak. Its central portion is not a separate plate of cells in the region of the primi- 96 THE DEVELOPMENT OF THE CHICK tive streak, but this part becomes distinct as the primitive streak spUts into its derivatives. It is therefore only when the latter is entirely used up that the entire length of the medullary plate is established. However, long before this time the greater por- tion has become converted by folding into the neural tube, a process that proceeds in general from in front backwards. Thus a.o. re. - — pr.gr. Fig. 50. — Embryo of 3 s from above, drawn in bal- sam with transmitted light. a. c. v., Amnio-cardiac vesicle, a. o., inner margin of Area opaca. F. G., Fore-gut. N'ch., Notochord. n. F., Neural fold. pr. gr., Primitive groove, s. l,s. 2, s. 3, First, second and third somites. successive stages may be studied in serial sections of the same embryo; an anterior section, for instance, showing the completed tube, one farther back, the folded medullary plate, and yet more posteriorly the central part of the medullary plate disappears in. HEAD-FOLD TO TWELA E SOMITES 97 the undifferentiated mass of the primitive streak. These condi- tions must be born in mind in the following description. The Neural Groove and Folds. Shortly after the formation of the head-fold the center of the medullary plate becomes sunk in the form of a deep groove beginning a short distance behind the "■;*-■■■"'■ "-'.>; j/.-.-Tj. *;-"'.--■;?••■ ■.'•'.'.■.■■ -j P'/"-^'!-t.' '"■■'•■.;■ .'.T.-':'-' '-.'■• '.' ■'-. '■■■ '.v'V- ■':;'.'- '.•'•» ''-". '■'■''- ■■■ "■ ' .■ >'':'^- ::■':" '■ ; ■■ . .•■'-V.-.. ■:•.-•■ ■■'■'', :'^^:■'•.■-^''■^S•v'.■■ ..:-';'■ 7. ';■-•■■•. ■-:-»■• • v.\-*^v jfyi :<7l'J^-^i--^. '^>' ;.'i-- ;:-'i .■.-■'• ". •..-,; . ;■ -■ ',' ■■■•:■ ^ ■'■■- -■■ .. ■" .■■;--;;\v':r ^t^v-^-;-v.v. •'!•.■• -.•..'s-'^.^'.' jt> ~^t/ pA: r.z^y. xv ■ V.'-. ■•Viv^;,-..-'; .-,.':.■. -:.:v,v ..'..'■ v.-'::-,/; Fig. 51. — Embryo of 4 s from above, drawn in alcohol by reflected light, a. c. v., Amnio-cardiac vesicle, a. p., Area pellucida. a. v. i., Inter- nal vitelline area. med. pL, Medullary plate, n. F., Neural fold. Pr'a., Proamnion, pr. str., Primitive streak, s. 1, s. 3, First and third somites. 98 THE DEVELOPMENT OF THE CHICK anterior end of the plate (Fig. 48) (the neural groove) ; the mar- gins of the anterior portion of the medullary plate then become elevated somewhat above the surrounding blastoderm, forming MM mm mm ■J ■Si 'U 1 ( i Fig. 52. — The same embryo from beneath. a. c. v., Amnio-cardiac vesicle, a. i. p., Anterior intestinal portal. H. F., Head-fold. Pr'a., Proamnion. the neural folds (Figs. 51 and 56). The latter rise very rapidly, thus deepening the neural groove, and bend in towards the middle line (Figs. 53,54, etc.,) meeting, by the time four or five somites are HEAD-FOLD TO TWELVE SOMITES 99 formed, a short distance back of the anterior end of the medullary plate (Figs. 50 and 51). The posterior ends of the neural folds do not, at this time, reach the region of the first somite. The region where the neural folds first come in contact corresponds approximately with the region of the future mid-brain, or ante- rior part of the hind-brain. &i 00 0^0- Q^O®o ^^y»r,XOWf: %<^0 -s%pi%^;-:roo%o:-oj 5 0O® o°« OOOQVP®^ ^«.^-s^4§%%S^##£-1?gis.3§s^^o-^ 1PQ.P «©0o® ©Oqo i-/7/f 7,OPO, ?o"g^^'®p®o:©:oo Q^J'S^§o'i^Pj,O^J^^ a cs> <0^1- onpJ £nt. Fig. 52 A. — Median longitudinal section of the head, stage of 4 s. The sec- tion passes through the length of one of the neural folds just behind the anterior end. (Cf. Fig. 5L) a. i. p., Anterior intestinal portal. Ect., Ectoderm. Ent., Entoderm. F. G., Fore-gut. H. F., Head-fold. Mes., Mesoderm. Mes. H. C, Meso- blastic head cavity, n. F., Neural fold. or. pi., Oral plate. The process of closure itself is essentially the same in all regions of the neural tube. Each neural fold has two limbs: an inner thick limb, belonging to the medullary plate, and an outer, thin limb, continuous with the general ectoderm (cf. Fig. 68 B). When the folds of opposite sides come in contact, the inner limbs of the two sides become continuous with one another, and also the outer limbs, the ectoderm then passing continuously over a closed neural tube. Certain cells in the suture and in the walls of the tube next 100 THE DEVELOPMENT OF THE CHICK i !lfc o rz Si o H c r-> l-H 1 Ti t; O y o r o -= -^ CQ 31 1> c3 o 3 a o5 >5 G o 72 O -t-^ ^ • ^H ■+^ u* n o C3 «+- CI) C,J -*^ H t/2 — O 53 -O o is . o O "TS t" - .2 6 _ '/J -»J r', yj U. ;_; • ^^ -k-1 rf, r^ >— ' '•¥-' ^tJ o ^^ fl) o ^/i o ^M 'Jl 0* a* > a o H >. . o I— I a' xn % I S r "^ '^ ^ HEAD-FOLD TO TWELVE SOMITES 101 the ectoderm 'are destined to form the neural crest, a structure of great significance, inasmuch as the series of cranial and spinal ganglia is derived from it. (See following chapter.) So//}'//. ^'■•''l^/./>/. *ry>^'^ Fig. 54. — Transverse section through the same embryo a short distance in front of the anterior intestinal portal. For explanation of letters see preceding figure; in addition: Ph., Pharynx. Som'pl., Somatopleure. Spl'pl., Splanchnopleure. v. M., Ventral Mesentery. :)4?/>/ Fig. 54 A. — Transverse section through the head of a 10 s embryo. The region of the section is near the center of the hind brain. Ao., Aorta. End'c, Endocardium. End'c. S., Endocardial septum. H. B., Hind brain. My'c, Myocardium, p. C, Parietal cavity. Ph., pharynx. So'pl., Somatopleure. SpFpl., Splanchnopleure. v. M., Ventral mesentery. The Neuropore. From the place where the neural folds first meet, the elevation and fusion proceed both forwards and back- wards in a continuous fashion (cf. Figs. 59, 61, 65, etc.). Although the open anterior stretch of the neural tube is very short in com- parison to the posterior open part, it is not until about the 12 s 102 THE DEVELOPMENT OF THE CHICK stage that the former closes completely (cf. Fig. 64). The final point of closure at the anterior end, known as the neuropore, is supposed by some to be a point of great morphological signifi- cance, and to mark the extreme anterior end of the original neural ax. Mes. 'S^. pcf.C. /^ Mes. N ^ Fig. hb. — Transverse section through the head immediately behind the optic vesicles; stage, 10 s. Ao., Aorta, ax. Mes., Axial mesoblast. Ect., Ectoderm. Ent., Entoderm. F. B., Fore-brain. Mes., Mesoderm, or. pi., Oral plate, p'a. c, Periaxial cord. p. C. Parietal cavity. Pr'a., Proamnion. Ph., Pharynx, v. Ao., Ventral aorta. axis. It is identified by these writers with the permanent neuro- pore of Amphioxus. However, this is open to question. Poste- riorly the closure of the neural tube proceeds much more rapidly, though, of course, it is not fully completed until after the disap- pearance of the primitive streak. ^■F. med.p/. '^-^ -;§_?& mmiSif^^^^^s& ^ ^mmmmm^ ?^3£Ect. ^®l«5;i?'- A'U Fig. 56. — Early stage of the neural folds. Transverse section through a 4-5 s embryo between the last somite and the anterior end of the primitive streak. Ect., Ectoderm. Ent., Entoderm, n. F., Neural fold. N'ch., Noto- chbrd. med. pi., Medullary plate. Mes., Mesoderm. The question as to the position of the anterior end of the original neural axis is one of great morphological significance. Accompanying the closure of the neural tube in this region the HEAD-FOLD TO TWELVE SOMITES 103 /?/='. /?.0 r ■gc. med.pl r-l /? Cr. -. f" •** «^?s*J«> ^^^^^^^ ^cl ^o/r?J/es. O 6^ ^.?>;^: yT^/. 6"ce/ Fig. 57. — Later stage of the neural folds. Section through the head of an embryo of 2-3 s; corresponding to about the future mid-brain region. Coel., Coelome. g. C, Germinal cells, med. pi., Medullary plate. Mes., Mesoblast. n. F., Neural fold. n. Cr., Neural crest. N'ch., Notochord. som. Mes., Somatic layer of mesoblast. spl. Mes., Splanchnic layer of mesoblast. anterior end rapidly grows forward beyond the anterior end of the fore-gut. The floor of the neural tube does not, however, take part in this extension, the consequence being that the sum- mits of the neural folds form arching knees extending in front of the original anterior end of the medullary plate (Figs. 51 and 52). The extreme anterior end of the neural tube formed in this way has a ventral as well as a dorsal defect, and when it closes there is a ventral as well as a dorsal suture. The ventral end of this suture marks the original anterior end of the me- dullary plate, and this lies at the stage of 10 somites a short distance in front of the ante- rior end of the oral plate in the region of the future re- cessus opticus (Fig. 62). (Go- ronowitsch calls the anterior fissure, sutura cerchralis ante- rior; His divided it into two Fig. 58. — Ventral view of the head region of an embryo of 5 somites, drawn in balsam with transmitted light. X 30. a. c. v., Amnio-cardiac A-esicle. a. i. p.. Anterior intestinal jiortal. F. G., Fore-gut. My'c, Myocardium. N'ch., Notochord. n. F., Neural fold. s 2, s 4, Second and fourth somites. 104 THE DEVELOPMENT OF THE CHICK -ytv: ^/y. Fig. 59. — Embryo of 7 s from above drawn in balsam with transmitted light, x 30. a. c. s., Anterior cerebral suture, ceph. Mes., Cephalic Mesoblast. F. G., Fore-gut. N'ch., Notochord. n. T., Neural tube. op. Ves., Optic vesicle. Pr'a., Proamnion, pr. str., Primitive streak, s 2, s 7, Second and seventh somites. V. o. m., Omphalo-mes- enteric vein. HEAD-FOLD TO TWELVE SOMITES 105 parts, sutura neurochordalis sen ventralis and sutura terminalis anterior.) The neuropore question resolves itself into this: What part of the sutura cerebralis anterior is to be called neuropore? As the suture extends from near the infundibulum to the pineal region at least, there is a wide range of choice. However, there is a point in the suture near its dorsal end where the separation of the ectcderm from the neural tube takes place later than elsewhere. This may be regarded as the equivalent of the neuropore. The suture is the site of formation of the lamina terminalis (Chap. VIII). d.i.p. Fig. 60. — The head of the same embryo from below X 30. a. i. p., Anterior intestinal portal. End'c. s., Endocardial septum. F. G., Fore-g;ut. Ht., Heart. N'ch. T., Termination of Notochord. op. Ves., Optic vesicle, p. C, Parietal cavity. Pr'a., Pro- amnion. V. o. m., Omphalo-mesenteric vein. It will be seen that according to this account most of the primary fore-brain includes no part of the original floor of the neural tube. Primary Divisions of the Neural Tube. The neural tube is the primordium of the brain and spinal cord. Its cavity becomes the ventricles of the brain and the central canal of the cord. There 106 THE DEVELOPMENT OF THE CHICK pr.str. ■ ■■■-■■:■■:: -■■ft:- ' ■ ■■■^ Fig. 61. — Embryo of 9 s from above drawn as a transparent object with transmitted light. X 30. Abbreviations same as before; in addi- tion: H. B., Hind brain. M. B., Mid brain. n. S., Neural suture. HEAD-FOLD TO T\VEL^'E SOMITES 107 a.cs. ceph. Mes End'c.5. d.l.p. s.z. Pig 62. — The head of the same embryo from beneath more highly magnified. In this drawing an attempt is made to show different levels of the embryo superposed: thus the heart is uppermost in the figure, beneath this the fore-gut (F. G.), beneath this the notochord, and at the lowest level, the neural tube, a. c. s., Anterior cerebral suture. Inf., Infundibulum. p. C, represents the anterior boundary of the parietal cavity, or. pi., Oral plate, v. Ao., Ventral aorta. Other abbreviations as before. is no clear distinction between brain and cord at first, the one passing without any anatomical landmark into the other. Now the brain is the central nervous system of the head, so it is not until one can determine the posterior boundary of the embryonic head that it becomes possible to determine the hind end of the 108 THE DEVELOPMENT OF THE CHICK brain. The first clear landmark is given by the mesoblastic so- mites, because it is known that the four anterior somites are cephalic. All of the neural tube in front of the fifth somite is therefore cranial. What a large proportion of the neural tube this is in early stages may be seen by comparison of figures of embryos in the period covered by the chapter (cf. Fig. 61). Be- fore the appearance of the first somite the entire medullary plate in front of the primitive streak is in fact cranial. Origin of the Primary Divisions of the Embryonic Brain. The embryonic brain is divided into three divisions of unequal length, viz., the jore-hrain (prosencephalon), mid-brain {tnesencephalon) , and hind-hrain (rhombencephalon). The first division is character- ized in the period we are considering by its very considerable lateral expansions, the rudiments of the optic vesicles (Figs. 59, 61, 63, etc.), and also by the fact that there is a suture in the anterior portion of its floor owing to the mode of its origin (Fig. 62). A definite constriction between it and the following division first appears in embryos with six or seven somites (Fig. 59). At the stage of 9-10 somites the next division (mid-brain) becomes clearly marked off by a constriction from the hind-brain (Fig. 61). The latter is relatively very long, and its anterior half is characterized in the 12-somite stage by the existence of five divi- sions (neuro meres) separated by constrictions (Fig. 63). It will be noted that the first neuromere of the hind-brain appears about twice as large as the succeeding ones ; it really includes two neuro- meres according to some authors. Similarly, it is maintained that the mid-brain includes two neuromeres and the fore-brain three. According to HilFs account the entire brain of the embryo chick is composed of eleven neuromeres or neural segments, which are formed even in the 1 s stage. The first three enter into the composition of the fore-brain ; the next two, viz., 4 and 5, form the mid-brain, and the last six the hind-brain. The three that enter into the composition of the primary fore-brain have the following fate according to Hill: the first forms the telen- cephalon, the second the anterior division (parencephalon) and the third the posterior division (synencephalon) of the diencephalon. The cere- bellum arises from the first neuromere of the hind-brain, sixth of the series. This question is more fully discussed in Chapter ^ I. (See Fig. 83.) HEAD-FOLD TO TWELVE SOMITES 109 Ji.B. Vo.m. ^.2. 3.JZ. ■pr str. Fig. 63. — Embryo of 12 s, from above, drawn as a transparent object with transmitted light. X 30. Abbreviations as before. IV. The Mesoblast The changes in the mesoblast during this period are of great importance. At the time of appearance of the head-fold it con- sists of two great sheets of cells between ectoderm and entoderm no THE DEVELOPMENT OF THE CHICK beginning on each side of the head-process and primitive streak, and extending laterally and posteriorly to the margin of the vascular area. The lateral margins at this time extend anterior to the embryonic axis, so that the anterior margin of the mesoblast forms a curve with the concavity directed forward. c O P- Fig. 64. — Head of the same embryo from below. X 30. Abbreviations as before. The mesoblast in the region in front of the primitive streak is known as gastral mesoblast, and in the region of the primitive streak as prostomial mesoblast; the latter is fused with the primi- tive streak. However, the distinction between the gastral and prostomial mesoblast is not of permanent significance, because the latter is being continually converted into the former as the primitive streak vmdergoes separation into ectoderm, notochord^ and mesoderm. Confining our account now to the gastral mesoblast: a trans- verse section across an embryo in which the head-fold is forming shows a sheet of cells lying on each side of the notochord between the ectoderm and entoderm. It is several cells deep near the notochord, and thins gradually peripherally (cf. Fig. 56). The thicker portion next the notochord is distinguished as the paraxial mesoblast (vertebral plate) from the more peripheral portion or lateral plate. The mesoblast is sparser, the cells more scattered, HEAD-FOLD TO TWELVE SOMITES 111 and the whole tissue of much looser texture in the more anterior portions of the embryo. The paraxial mesoblast increases rapidly in thickness and thus becomes clearly distinguishable from the lateral plate. Shortly after the formation of the head-fold a transverse split appears in the paraxial mesoblast a short distance in front of the anterior end of the primitive streak (Fig. 48). This is soon fol- lowed by a second split, a very short distance behind the first, and thus a complete mesoblastic somite is established. The split- ting is accomplished rather by segregation of the cells than by an actual folding. The mesoblast cells immediately in front of the first split aggregate so as to form a somite continuous anteriorly with the mesoblast of the head and thus lacking an anterior boundary; this is the first somite, and the one formed between the first two splits in the mesoblast is the second. The first somite established is first, not only in point of time, but also in position, all the remainder forming in succession behind this (cf. Figs. 48, 50, 51, 59, 61, etc.). As this is a point of con- siderable importance for understanding the topography of the embryo, and as previous text-books have a different account of it, it is worth while to give the evidence for this position in some detail. It has been believed up to a very recent time that from two to four somites were formed in front of the first one. This belief was due very largely to a misconception of the nature of the primitive streak, which was believed by some to be extra- embryonic, that is to lie behind the embryo and not to be a part of the embryo itself. The first somite lies so near to the anterior end of the primitive streak that it was difficult to believe that room could be made by growth between it and the primitive streak with sufficient rapidity to accommodate the rapidly form- ing somites. In the entire absence of differentiated organs it was impossible to find landmarks by which to distinguish the first somite among the first five or six; hence it was natural to suppose that a certain number of somites arose in front of the first, espe- cially as it was not know^n how much of the anterior portion of the embryonic axis represented the head. However, in the absence of natural landmarks identifying the first somite formed, it is quite possible to create artificial ones, and in this way to identify it in later stages. This has been done by one of my students, Miss Marion Hubbard, in the following manner: In the 112 THE DEVELOPMENT OF THE CHICK first place the position of the first somite was marked with a deUcate electrolytic needle which left a permanent scar. The eggs thus operated on were closed up and permitted to develop to a stage of 10-12 somites or more; and then the mark was found F.B. jY.A^ ^u. ep. S.2 •S.J2. 2> G- pr.sir. ^t^■ Fig. 65. — Embryo of 12 s, from above, drawn in alcohol with re- flected light. au. ep., Auditory epithelium. Other abbreviations as before. to coincide with the first somite of the series. In the next place it was possible by similar means to mark out the topography of the embryonic head in the stage of one or two somites. Thus it was determined that a mark made immediately in front of the first somite formed appeared later in the region of the otocyst; HEAD-FOLD TO TWELVE SOMITES 113 but this arises normally at the stage of 12-14 somites, a very short distance in front of the first somite of the series, which is thus shown to have the same position as the first somite formed On the other hand, if one assumed that the first somite formed r? — ■y . g; .fs i'^ .' : ;> ' jib;4-vs Ht A.CV. A/'ch. Vo.m. d.i.p. Fig. 66. — The same embryo from beneath, drawn in alcohol with reflected light. Abbreviations as before. became the third or fourth of the series, it is clear that one would have to make a mark some distance in front of the first somite formed, to strike the place of origin of the otocyst. Marks made on this theory were always found a considerable distance m front of the otocyst. Altogether a large number of experiments 114 THE DEVELOPxMEXT OF THE CHICK was made, the concurrent testimony of which was perfectly conclusive.^ We shall then proceed on the assumption that the first somite formed is also the first of the series, and that the remainder arise in succession behind it as transverse sections of the paraxial mesoblast. There is always a stretch of unsegmented paraxial mesoblast between the last somite and the anterior end of the primitive streak. The first four somites belong to the head, and enter into the composition of the occipital region. The more anterior part of the mesoblast of the head never becomes segmented in the chick. In the anamniote vertebrates, segmentation of the mesoblast extends farther forward, and there is a greater number of cephalic somites. This may be taken as evidence that a large part, at least, of the head was primitively segmented like the trunk. As we shall see later, the primitive metamerism of the head is also expressed in other ways: neuromeres, branchiomeres, etc. The segmentation of the mesoblast finally extends to the hind end of the tail, new segments being continually cut off from the anterior end of the paraxial mesoblast until it is all used up. This is not complete until the fifth day. The number of somites thus formed is perfectly constant, as is also the fate of the individual somites. Primary Structure of the Somites. Each somite is primarily a block of cells arranged in the form of an epithelium around a small central lumen, towards which the inner ends of all the cells converge (Fig. 68 B). The central cavity (myocoele) is, however, filled with an irregularly arranged group of cells, and, though the cavity must be regarded as part of the primitive body-cavity, or coelome, it has no open communication with it. After the somites are formed they rapidly become thicker so that their lateral boundary becomes very sharply marked; this is not clue to a longitudinal constriction external to the paraxial mesoblast, as usually stated. Each somite has six sides, of which five are free, viz., dorsal, ventral, anterior, posterior, and median. The sixth or lateral side is continuous with the nephrotome. The Nephrotome. or Intermediate Cell-mass (Middle Plate). ^ Since the above was written, J. T. Patterson has obtained the same results (Biol. Bull. XIII, 1907). HEAD-FOLD TO TWELVE SOMITES 115 The somites and the lateral plate are not in immediate contact but are separated by a short stretch of cells continuous with both, known as the nephrotome or intermediate cell-mass or middle plate. The intersegmental furrows do not extend into the intermediate cell-mass, and the latter therefore remains unsegmented like the lateral plate. It consists fundamentally of two layers of cells, dorsal and ventral, of which the former is continuous with the dorsal wall of the somite and the somatic layer of the lateral plate, and the latter with the ventral wall of the somite and the splanchnic layer of the lateral plate (Fig. 68 B). Thus if the two layers of the intermediate cell-mass were separated the space between them would be continuous with the coelome that arises secondarily in the lateral plate. This condition actually exists in some of the Anamnia (Selachii, for instance) in which the intermediate cell-mass is also segmented. The Lateral Plate. This name is given to the lateral meso- blast within which the body-cavity arises. It is separated from the somite by the nephrotome and its lateral extension coincides with the margin of the vascular area. Development of the Body-cavity or Ccelome. The ccelome or body-cavity arises within the lateral plate as a series of sep- arated small cavities, distributed throughout its whole extent, which appear first in the anterior portion (1-3 s stage). By successive fusion of these cavities and their extension centrally and laterally, there arises a continuous cavity, the ccelome, w^hich extends from the nephrotome to the margin of the vascular area (Fig. 68), and w^hich becomes the pleuro peritoneal and per- icardial cavities in the embryo, and the extra-embryonic body- cavity beyond the boundaries of the embryo. Of the two layers of the lateral mesoblast thus established, the external is known as the somatic and the internal as the splanchnic layer. In the course of development the somatic layer becomes closely bound to the ectoderm, thus constituting the somatoplenre, and the splanchnic layer becomes similarly united to the entoderm, thus establishing the splanchnoplenre. The somatoplenre is destined to form the body-wall and the extra-embryonic membranes known as the amnion and chorion; from the splanchnoplenre is derived the alimentary canal with all its appendages, and the yolk-sac. As described in detail in the next chapter, this splitting of the mesoblast progresses with 116 THE DEVELOPMENT OF THE CHICK the overgrowth of the yolk until it extends completeh' around the latter Returning now to the first stages in the formation of the coe- lome. In the 3 s stage it undergoes a precocious expansion in the region lateral to the head of the embryo (Figs. 51, 52, etc.), forming a pair of large cavities known as the amnio-cardiac vesicles, because they participate in the formation of the amnion and pericardium. These cavities extend in rapidly towards the middle line, and enter the head-fold in the 4-5 's; stage (Figs. 52, 58). At the stage of 6-7 s they meet in the floor of the fore-gut immediately behind the oral plate and fuse together, thus divid- ing the head-fold into somatic and splanchnic limbs, as previously described. A median undivided portion of the body-cavity known as the parietal cavity (forerunner of the pericardium) is thus established beneath the fore-gut; and it extends back- ward wdth the elongation of the fore-gut in the manner already described. A pair of blind prolongations of this cavity extends a short distance forward at the sides of the oral plate at the 10-12 s stage (cf. Fig. 62), lying lateral and ventral to the ventral aortse. The median angle of the body-cavity, where the somatic and splanchnic layers meet, is a point of fundamental morpho- logical importance. In the region of the somites the nephrotome is attached here, and in the head the wdng of cells leading to the axial mesoblast (cf. Figs. 68 B, 53, and 54). In an embryo with ten somites this angle may be traced forward to near the hinder end of the oral plate, lying beneath the lateral angles of the pharynx. Mesoblast of the Head. Mesoblast exists in two forms in the embryo: (1) in the form of epithelial layers or membranes (mesothelium), and (2) in the form of migrating cells which usually unite secondarily to form a syncytium in the form of a network, the meshes of wdiich are filled with fluid; the nuclei lie in the thickened nodes. This form of the mesoblast is known as mesenchyme. It is always derived from a pre-existing epi- thelial layer, usuall}", but not necessarily, mesothelium, for, as we shall see, parts of it are derived from ectoderm and entoderm; on the other hand, mesenchyme may secondarily take on an epithelial arrangement (endothelium). The terms mesothelium and mesenchyme have therefore merely descriptive significance in the early embryonic stages. The mesenchyme has no single HEAD-FOLD TO TWELVE SOMITES 117 embryonic significance either as to origin or fate, but is to be regarded as a mixed tissue. The mesoblast of the head is derived from several sources: (1) from a continuation forward of the paraxial mesoblast; (2) by proliferation from the fore-gut; and (3) from proliferations of ectoderm. (1) The axial mesoblast of the head is an anterior continua- tion of that of the trunk ;^ terminates at the anterior end of the fore-gut with which it is continuous from the stage of the head- process up to about the 6 s stage>.(Figs. 43 and 49). In the anterior part of the head it is mesenchypial in its general struc- ture, grading posteriorly into the mesothelial paraxial mesoblast of the hinder part of the head and trunk. It* is continuous at first with the lateral mesoblast in whl<»h the ■ a-mnio-cardiac vesicles are forming; but this connection is los^ in iUe anterior part of the head that projects forward above the blastoderm; that is, in front of the head-fold. (2) The anterior end of the fore-gut proliferates mesenchyme from the time of its first formation to about the 6 s stage (Fig. 49). The proliferation is so rapid that it may give rise to the appearance of diverticula. The extreme anterior end of the floor forms a sac which lies just in front of the oral plate at the 4 s stage (Fig. 52 A), but soon after breaks up into mesenchyme. There is a considerable mass of mesenchyme formed from this source in the space bounded by the anterior end of the fore-gut, the neural tube and the ectoderm; at the 4 s stage this appears fused with the floor of the neural tube and the surface ectoderm, and probably receives cells from both; the anterior end of the notochord also disappears in this mass (cf. Fig. 67). (3) Ectodermal proliferations forming mesenchyme in the head. (This subject is discussed in the next chapter.) Vascular System. The origin of the blood-islands in the opaque area was described in the preceding chapter. They lie between the coelomic mesoblast and the yolk-sac entoderm de- rived from the germ-wall. When the somatopleure and splanch- nopleure are formed the blood-islands lie between the two layers of the latter, and the somatopleure is entirely bloodless. About the stage of 1 somite a vascular network continuous with the original network of the opaque area begins to appear in the pellucid area, at first at the margin of the opaque area, but by 118 THE DEVELOPMENT OF THE CHICK degrees nearer and nearer to the embryo, until, by the 7 or 8 s stage, blood-vessels begin to appear in the embryo itself. It is important to note that the order of appearance of the vascular primordia is first in the area opaca in the order previously de- scribed, then in the pellucid area and finally in the embryo itself. Moreover, the parts appearing later are, usually at least, in con- tinuitv with those first formed. Before discussing the way in which the blood-vessels arise in the pellucid area and in the embryo, we should consider the first differentiation within the original, or peripheral, blood- islands. Between the 3 and 5 s stage it may be noticed in sections that vacuoles are forming within the peripheral blood- islands near the entodermal surface. The expansion of these vacuoles carries the peripheral layer of cells away from the main mass of cells composing the blood-islands, and by degrees the process is carried completely around the blood-island, so that the peripheral layer becomes entirely separated from the central mass and encloses it (See Fig. 68 C). The enclosing cells become flattened during this process to form an endothelium; inasmuch as the blood-islands are not separate, but anastomose to form a network, the process results in the formation of a network of endothelial tubes enclosing cell-masses. Thus arise the first blood-vessels. The enclosed masses of cells rapidly acquire haemoglobin, become separated from one another, and form blood-cells. There is a great difference in the relative amounts of blood- cells formed in different regions. Thus in the anterior part of the opaque area and in the pellucid area the original blood- islands are relatively small (Figs. 44 and 45), and furnish material sufficient only for the formation of the blood-vessels. On the other hand, in the peripheral part of the vascular area, especially towards its posterior end, the largest masses of blood-cells are found; and these conditions grade into one another. In other words, the formation of blood-cells is restricted at this time to the opaque area, and is most abundant posteriorly. In the pellucid area only empty blood-vessels are formed. Similarly the blood-vessels of the embryo itself are at first empty; they become filled secondarily from the opaque area when circulation begins. The appearance of blood-vessels within the pellucid area HEAD-FOLD TO TWELVE SOMITES 119 and the embryo has been interpreted in two principal ways: (1) that they are an ingrowth from the original vascular primor- dium of the opaque area; and (2) that they arise by differentia- tion in situ. The first view was originally stated by His, and has been supported by Kolliker and others. The second is sup- ported by Riickert, P. Mayer and others. The observations, on which the ingrowth theory of His w^ere based, were made originally on whole blastoderms of the chick, and concerned primarily the order of origin of the blood-vessels, which is cen- tripetal. But it is obvious that such observations do not in themselves demonstrate the existence of an independent ingrow- ing primordium; they are not in the least inconsistent with the view that the blood-vessels cUfferentiate from the cells in situ. Within the embryo itself parts of certain vessels appear to arise separately, and form secondary connections with the vessels formed at an earlier time; this is the case for instance with the dorsal aorta in the region of the head. The histological appear- ances accompanying the first origin of blood-vessels in the pel- lucid area appear to favor the view that they arise from detached cell-groups of the splanchnic mesoderm. It would, however, be going too far to assert that the embryonic blood-vessels have no independent power of growth; certain appearances cannot be satisfactorily explained in any other way. Origin of the Heart. The embryonic heart possesses two layers: an internal delicate endothelium, the endocardium, and an external strong muscular layer, the myocardium. The endo- cardium arises in continuity with the blood-vessels of the pellucid area, and is in no Avise different from them; the myocardium, on the other hand, arises from the splanchnic mesoblast. The heart is thus to be regarded as a portion of the embryonic vascular system, specially provided with a muscular wall for the propul- sion of the blood. The first indication of the heart is a thicken- ing of the splanchnopleure of the amniocardiac vesicles, which forms the primordium of the m\l^ardium. This is situated a short distance lateral to the hind-brain region of the embryo, and makes its appearance between the stage of 3 and 5 somites. The endocardium soon appears between the thickened ento- derm and the myocardium, in the form of a delicate endothelial vessel on each side, continuous with the extra-embryonic blood- vessels. This is, indeed, the place where the blood-vessels first 120 THE DEVELOPMENT OF THE CHICK reach the embryo. The myocardium then becomes arched towards the body-cavity and includes the endocardium in its concavity (Fig. 53). The heart thus comes to consist of two parts on each side: a myocarcUal gutter semicircular in cross section, open towards the entoderm, and an endothelial tube lying in the gutter, and in contact with the entoderm. At this time the lateral limiting sulci appear in the splanchnopleure just central to the endocardium on each side, and, as the fore- gut closes from in front backwards, the following changes take place (Figs. 54 and 54 A): (1) the entoderm withdraws completely from the fused apices of the lateral folds in the splanchnopleure, and thus a wide separation is made between the floor of the pharynx and the splanchnopleure below; (2) the right and left endocardial tubes come into immediate contact in the floor of the pharynx; (3) the two myocardial gutters coming together form a single tube around the endocardium, suspended by a double mesoder- mal membrane (mesocardium or dorsal mesentery of the heart) to the floor of the pharynx, and attached by a similar mesentery (ventral mesentery of the heart) to the splanchnopleure beneath (Fig. 54). The latter connection is ruptured almost as soon as formed, so that the floor of the myocardium becomes complete (Fig. 54 A). Soon after the completion of the floor of the phar- ynx the two endocardial tubes press together until the common wall becomes reduced to a vertical partition, which then ruptures; and finally (10-12 s) all traces of the original duplicity of the heart disappear (Figs. 60, 62, 64). The heart thus arises from two lateral halves which fuse sec- ondarily to form a single tube. This fusion takes place from in front backwards, hence the anterior end of the heart is formed first. Indeed, the full length of the cardiac tube is not formed in the period covered by this chapter; the definitive hindermost division is established by concrescence after the 12 s stage. But the actual hind end is always continuous with the extra-embryonic network of blood-vessels and this connection develops into the main splanchnic veins. As a rare abnormality the lateral primordia of the heart may meet and fuse dorsal to the embryo, instead of in the floor of the pharynx. This condition is known as omphalocephaly; in other rare cases the lateral halves may fail to unite, and two hearts may be formed. There are three views concerning the origin of the endocardium: HEAD-FOLD TO TWELVE SOMITES 121 (1) that it is an ingrowth of the extra-embryonic vessels, (2) that it arises from the mesoblast in situ, (3) that it arises from the entoderm in situ. Appearances such as that shown in Fig. 53 favor the last view. The heart is then a double-walled tube attached to the floor of the pharynx. The posterior end rests squarely against the an- terior intestinal portal and is continuous with the rudiments of the splanchnic veins running in the diverging folds of the portal; / the anterior end of the heart is continued as a simple endothelial tube (ventral aorta) as far forward as the oral plate, where it is divided in two (Figs. 62, 64, etc.). This primitive simplicity of the cardiac tube continues through- out the period considered in this chapter without substantial alteration. The heart increases in length with considerable rapidity, but being attached at its anterior and posterior ends by the aortic and venous roots respectively, it is forced to bend, nearly always to the right, so that a convexity of the heart appears to the right of the embryonic head, at about the 11-12 s stage (Figs. 63, 64). About this time the mesocardium (dorsal mesentery of the heart) disappears except at the posterior end, and the cardiac tube thus becomes free except at its two ends. The Embryonic Blood-vessels. The dorsal aorta arises from the median edge of the vascular network, which extends across the pellucid area in the splanchnopleure. At the stage of 7-9 somites, it has reached the nephrotomic level. The marginal meshes gradually straighten themselves out into a longitudinal vessel, continuous with the net- work at the sides and behind. Only the trunk part arises in this manner. The cephalic part arises by forward growth of the trunk part or from mesenchyme in situ. In some embryos the most anterior portions of the cephalic aortse are much better developed than the posterior por- tion; indeed, in places there appears to be a complete hiatus between cephalic and trunk aortse. The impression gained is that a large part of the cephalic aortse arises in situ. Some series of sections are practically conclusive in this respect (8-9 somites). A connection is then formed around the anterior end of the fore- gut with the ventral aortse (Fig. c>b), and an arterial pathway is thus established from the heart by way of the ventral and dorsal aortae to the vascular network of the splanchnopleure. The arterial system consists at thi rty-three hours (12 s stage) of the following parts: (1) ventral aorta; (2) first visceral or 122 THE DEVELOPMENT OF THE CHICK mandibular arteries connecting 1 and 3; (3) dorsal aortse; (4) seg- mental branches of the dorsal aortse. The ventral aorta is, as we have seen, the anterior prolongation of the endocardium extending between the extreme anterior end of the heart proper and the oral plate. At the oral plate it divides into two branches, right and left mandibular arteries or arches, that surround the anterior end of the fore-gut, and arch over to be continued into the two dorsal aortae. The tissue in which these arches run is destined to form the mandibular arch or lower jaw. The two dorsal aortse are very large vessels running above the roof of the pharynx near its lateral angles. They give off no branches in the head. In the trunk they pass backwards in the splanchno- pleure beneath the somites (Fig. 68 B), and are connected at intervals with the extra-embryonic blood-vessels. These con- nections are more important in the region of the primitive streak (Fig. 63) where the dorsal aortae disappear in the general extra- embryonic network. Slight diverticula of the dorsal aortae ascend in the interspaces between successive somites (segmental arteries). Concerning the veins in the period under consideration there is nothing additional to be said. V. Description of an Embryo w^ith 10 Somites It will now be in place to describe rather fully the anatomy of the stage at which we have arrived; this will serve as a point of departure for the next chapter. The blastoderm is a circular membrane covering a consider- able portion of the yolk (cf. Fig. 32 A). The embryo appears to the naked eye as a whitish streak in the central pear-shaped pellucid area. The surface views and the two views of the em- bryo viewed as a transparent object show the topography of the various parts of the embr3^o (Figs. 63-66). A section across the entire blastoderm at the stage of 10 s, through the sixth somite (Fig. 68), shows the following parts: The ectoderm bounds the section above; it is thickened in the angle between the neural tube and the somites, and becomes thinner as it is traced peripherally; at the extreme periphery of the blastoderm it merges into a mass of cells that interpenetrate the yolk. Ventrally the boundary of the section is formed by ike entoderm which is slightly arched upwards in the middle line. HEAD-FOLD TO TWELVE SOMITES 123 In the region of the area pellucida the entoderm is very thin; at its boundary it passes rather abruptly into the large rounded vesi- cular cells of the yolk-sac entoderm, which becomes continuous at the margin of the vascular area with the germ-wall; the latter continues to the periphery where it merges in the undifferen- tiated cell-mass (zone of junction) (Figs. 68 A-68 E). The large neural tube is not yet closed. Beneath the neural tube is a sec- tion of the solid rod-like notochord. Fig. 67. — Median longitudinal section of the head of an embryo of 13 s. Ectam., Ectamnion. F. B., Fore-brain. H. B., Hind-brain. Inf., In- fundibulum. M. B., Mid-brain, pr'c. pi., Precardial plate. T. p., Tuber- culum posterius. Other abbreviations as before. The mesoderm (Fig. 68 A, B, C) lies between the parts already named; it consists on each side of the middle line of the following parts: (1) the mesoblastic somite, a block of cells that radiate from a central cavity filled with irregularly disposed cells; (2) the intermediate cell-mass or nephrotome, forming a narrow connect- ing bridge between the somite and the lateral plate; (3) the lateral plate, split into two layers, external, known as the somatic layer, and internal or splanchnic layer. The cavity between the two layers is the ccelome or body-cavity; it is very narrow next the nephrotome, but widens as it extends laterally to the margin of the vascular area, and is divided by various strands of cells extending from somatic to splanchnic layers, thus indicating its origin by fusion of coelomic vesicles. The ectoderm plus the somatic layer constitute the somato- pleure, from which the body-wall, amnion, and chorion are derived, and the entoderm plus the splanchnic layer form the splanchno- 124 THE DEVELOPMENT OF THE CHICK pleure, from which arises the intestine and all its appendages, including the allantois and the yolk-sac. Blood-vessels lie be- tween the splanchnic mesoblast and the entoderm. The large vessels beneath the somite and nephrotome are the dorsal aortce; small vessels are present in the area pellucida, and there are many large ones in the area vasculosa. The walls of the vessels are constituted of a single layer of flat endothelial cells bulging in the region of the nuclei; in the vascular area are true blood- islands with embryonic blood-cells more or less fully filling the cavity. In a median sagittal section (Fig. 67) the following points should be noticed: (1) the neural tube is enlarged in the region of the head to form the brain, more fully described below; (2) the entoderm forms a tube in the head known as the pharynx or cephalic enteron (cephalic part of the fore-gut), opening behind the heart into the space between the entoderm and yolk. The floor of the anterior end of the fore-gut is fused to the ectoderm in the middle line forming the oral plate. The entoderm forming the floor of the fore-gut turns forward around the hind end of the heart, and beneath the anterior part of the head forms part of the proamnion or mesoderm-free region of the pellucid area; (3) the large pericardial (parietal) cavity lies beneath the floor of the fore-gut. Attached to the posterior wall of the pericar- dium one sees the hind end of the heart with its two walls, the endocardium and the myocardium a fold of the mesoblastic lin- ing of the pericardium. Between the anterior end of the pericar- dium and the oral plate is seen the endothelial ventral aorta; (4) the notochord lies between the fore-gut and neural tube and ends anteriorly in a mass of mesenchyme lying between the infundib- ulum and fore-gut. The Nervous System. The neural tube is closed at the 12 s Fig. 68. — A. Transverse section across the axis of the embryo and the en- tire blastoderm of one side. The section passes through the sixth somite of a 10 s embryo, and is intended to show the topography of the blastoderm. The regions B, C, D, E are represented under higher magnification in the Figs. B, C, D, E. a. V. e., Area vitellina externa, a. v. i.,Area vitellina interna. Bl. i., Blood island. Bl. v., Blood vessel. Coel., Coelome. G. W., Germ-wall. M. O., Margin of overgrowth. N'ph., Nephrotome. S., Somite. Som'pl. Soma- topleure. SpFpl., Splanchnopleure. Som. Mes., Somatic layer of mesoblast. spl. Mes., splanchnic layer of the mesoblast. S. T., Sinus terminalis. Y. S. Ent., Yolk-sac entoderm. Z. J., Zone of junction. Somp, V B/v BJ/ B/.!. -M,-*' *i J,'-2.' ■*^-- 3 D ^•'^iV^r:^^ d.i/:s. 126 THE DEVELOPMENT OF THE CHICK stage (Figs. 63 and 65) to a point a little behind the last meso- blastic somite; beyond this the medullary folds diverge and are lost to view towards the hind end of the primitive streak. We may distinguish a cephalic portion (brain or encephalon) and a trunk portion (spinal cord or myelon) of the neural tube; the boundary lies between the fourth and fifth somites, for the first four somites enter into the composition of the head. The brain is thus at this time about as long as the portion of the cord formed or indicated by the medullary folds. For description, see p. 108. Alimentary Canal. The alimentary canal and its appendages exist only potentially in this embryo in the form of the splanchno- pleure, except in the head. The cephalic enteron of this stage corresponds to a large part of the pharynx. The oral plate has already been described in connection with the sagittal section (Fig. 67). In transverse section the extreme anterior end of the fore-gut is quite narrow, elsewhere it is very wide laterally, and in one place its lateral expansions come in contact with the ectoderm on each side and fuse to it, thus indicating the hyoman- dihular cleft. The floor and lateral walls of the pharynx are com- posed of columnar cells, the roof of flattened squamous cells (Fig. 54). Vascular System. The heart lies in the parietal cavity be- neath the pharynx; it is bent near its middle to the right. It is an undivided double-walled tube, the internal wall or endocardium being a continuation of the blood-vessels, and the external wall, myocardium or muscular heart, being a duplication of the wall of the pericardium. It has not yet reached the stage of regular contraction, though it may be observed to twitch from time to time. The chambers of the heart are indicated, but not clearly defined at this time; one can only say that the posterior end is the venous end from which the sinus and auricles are to form, and the anterior two thirds, the arterial end, destined to form the ventricles and bulbus. The endocardium is continued anteriorly into the ventral aorta, which divides on each side of the oral plate (Fig. 64), to form the mandibular arches that describe a loop around the anterior end of the fore-gut and are continued posteriorly as the dorsal aortw, which run above the roof of the pharynx, lateral to the notochord, into the trunk, where they lie ventral to the nephrotome, and send off short blind branches (segmental arteries) HEAD-FOLD TO TWELVE SOMITES 127 between the somites. Near the primitive streak they disappear by merging in the vascular network of the blastoderm. The posterior end of the endocardium divides in two branches that pass out along the postero-lateral margins of the fore-gut into the general vascular network of the blastoderm (Fig. 64). This connection constitutes the beginning of the vitelline veins, through which the blood from the yolk-sac enters the posteriori end of the heart. General The elongated form of the entire embryo and the preponderance of the head are marked features of this stage. The latter condition is largely due to the order of origin of parts: the anterior parts preceding the more posterior in their appear- ance. The head is really, therefore, in a more advanced stage of development than the trunk, hence larger. The elongated condition of the head and the arrangement of all its organs in longitudinal sequence, however, are probably conditions of phylogenetic significance, and point towards an ancestral con- dition. The topographical values of the cUvisions of the em- bryonic head are very different from those of the adult, to attain which certain regions develop to a relatively enormous extent, and others comparatively little. A number of features in the anatomy of the 12 s stage are purposely omitted from this description, as they represent the primordia of structures described more fully beyond; such, for instance, are the neural crest, the pronephros, etc. Zones of the Blastoderm. The following zones may be recog- nized in the blastoderm : (1) the pellucid area surrounding the embryo; (2) the vascular zone of the opaque area; (3) area vitel- lina interna; (4) area vitellina externa. The pellucid area is readily defined by its transparency and by the existence of the sub- germinal cavity beneath it. The vascular zone is most readily defined by the extension of the blood tissue which has a very definite margin, coincident with the extension of the mesoblast. The area vitellina includes all of the blastoderm peripheral to the vascular area, and it is characterized by the presence of two layers only, ectoderm and entoderm (germ-wall). It is again divided into two concentric zones, internal and external. The internal is much the wider (Fig. 32 A), and is characterized by the existence of a perilecithal space, i.e., a slight fluid-filled cavity between the entoderm and yolk continuing the subgerminal 128 THE DEVELOPMENT OF THE CHICK cavity peripherally. The external vitelline area is relatively narrow, and consists (1) of the zone of junction adjoining the internal vitelline area, and (2) a free margin separate from the yolk (margin of overgrowth). The zone of junction is the per- sistent embryonic or formative part of the blastoderm from which the extra-embryonic ectoderm and entoderm arises. Thus as it spreads peripherally over the surface of the yolk, it leaves on its central margin the differentiated extra-embryonic ecto- derm and entoderm; in other words, the zone of junction is the youngest part of the blastoderm, and the concentric zones that may be drawn within it represent successively older stages in a centripetal direction. Therefore in a transverse section through the entire blastoderm successive stages of differentiation of the ectoderm and particularly of the entoderm are met as one passes from the zone of junction towards the center. The free margin arises from the zone of junction in the manner already described in Chapter II. It may be considered as a part of the ectoderm and it terminates in a row of enlarged cells that often exhibit amoeboid prominences on their margins. It would appear that these cells have the function of a marginal wedge that separates the vitelline membrane and yolk. The germ-wall has been the subject of many extended re- searches, but a definitive solution of its origin and function has not hitherto been obtained, mainly on account of the incomplete knowledge of its early history. The ground here taken is in some respects different from that of the various authors, but it is based on a study of its early history given in Chapter II. There is no deviation from the mode of formation of the zone of junction in the stage under consideration from what was found in earlier stages, and there is no reason to believe that its subsequent history varies in any important respect. It appears to be produced by continuous proliferation of the cells in the yolk as in earlier stages (see Fig. 68 E). These cells actively engulf the large yolk gran- ules, and the histological structure becomes in consequence diffi- cult of analysis. The cells lose their individuality and constitute an extended syncytium, the protoplasm of which is packed with yolk-granules. In removing the blastoderm from the egg in salt- solution one finds always, after removing the yolk that may be washed off, a narrow submarginal zone of adherent yolk that is not readily removed, and this is the site of the zone of junction. HEAD-FOLD TO TWELVE SOMITES 129 Centrally to the zone of junction we have the differentiated ectoderm and germ-wall sharply separated from the yolk by the perilecithal space. The ectoderm of the inner zone of the vitelline area requires no extended notice; it consists at this time of a sin- gle layer of flattened cells. The germ-wall next to the zone of junction consists of two or three layers of large, more or less rounded, cells with definite boundaries, each of which contains one or more yolk-spheres and smaller yolk-granules (Fig. 68 E). We may say roughly that whereas in the zone of junction we have cells in the yolk, in the vitelline area we have yolk in the cells. This may indicate sufficiently the way in which a several layered epithelium becomes differentiated from the zone of junc- tion. As this epithelium is traced centrally we find usually a short distance from the zone of junction a thinner area (Fig. 68 D), and beyond this again the several layers of cells even more laden with yolk-spheres and granules than previously; so that it would appear that these cells may actively engulf yolk- granules. At the margin of the vascular area the entoderm be- comes one-layered, and is composed of columnar cells with swollen free margins turned towards the yolk and still containing some yolk-granules and spheres (Fig. 68 C). At the margin of the pellucid area there is a rather sudden transition to the flat ento- dermal epithelium characteristic of this area. CHAPTER VI FROM TWELVE TO THIRTY-SIX SOMITES. THIRTY- FOUR TO SEVENTY-TWO HOURS I. Development of the External Form, and Turning of THE Embryo In the embryo of twelve somites only the head is distinctly separated from the blastoderm; and there is no sharp boundary between the embryonic and extra-embryonic portions of the blastoderm in the region of the trunk; but this changes very rapidly. The progress of the developmental processes, that have marked out an embryonic axis in the blastoderm, produces in the course of about eighteen hours a sharp distinction everywhere between embryo and extra-embryonic blastoderm. The latter, together with an outgrowth of the embryonic hind-gut (allantois), then constitute the so-called embryonic membranes, which become very complicated, and which provide for the protection, respira- tion, and nutrition of the embryo. We shall consider the forma- tion of the embryonic membranes separately in order not to confuse the account of the development of the external form of the embryo. In considering the development of the external form of the embryo, we must distinguish between those processes that sepa- rate it from the extra-embryonic blastoderm, and those that occur within its own substance leading to various characteristic bend- ings and flexures; we may consider them separately, although they are going on at the same time. Separation of the Embryo from the Blastoderm. The separa- tion of the embryo from the blastoderm takes place by the formation of certain folds or sulci that may be named: (1) the head-fold or anterior limiting sulcus; (2) the lateral limiting sulci, appearing as prolongations of the head-fold along the sides of the embryonic axis; and (3) the tail-fold or posterior limiting sulcus. The head-fold has been described in detail in the preceding 130 FRO:\r TWELVE TO THIRTY-SIX SOMITES 131 chapter. The lateral limiting sulci are a continuation of the lateral limbs of the head-fold; they owe their origin to the folding of the splanchnopleure and somatopleure adjacent to the embryo towards the yolk, at the line of junction of embryonic and extra- embryonic parts. The tail-fold arises about the stage of 26 to 27 somites (Fig. 93), and is similar to the head-fold, except that it is turned in the opposite direction. The sulci combine to form a continuous ring around the embryo and gradually pinch it off, so to speak, from the extra-embryonic blastoderm. In the splanchnopleure the limiting sulci (Fig. 69) come to- j.i.p. Mesacsrd 'o/- Fig. 69. — Transverse section through the fifth somite of the 23 s stage. Amn., Amnion. Ao., Aorta, a. i. p., Anterior intestinal portal. Coel., Coelome. Chor., Chorion. Ectam., Ectamnion. E. E. B. C, Extra-embry- onic body-cavity. Int., Intestine. 1. 1. s., Lateral limiting sulcus. My., Myotome, s. a., Segmental artery. So 'pi., Somatopleure. Spl'pl., Splanch- nopleure. s.. Somite, s. 5, Fifth somite. V. O. M. R. and L., Right and left omphalo-mesenteric veins. V. V., Vitelline vein. gether and fuse both in a caudal direction from the fore-gut, and subsequently in a cephalic direction from the hind-gut (see below), so as to convert the splanchnic gutter into a tube (the alimentary canal). There is thus a ventral suture along the alimentary canal in which the entoderm of the alimentary canal becomes separated from the extra-embryonic entoderm, thus leaving a double layer of the splanchnic mesoblast (ventral mesentery) connecting the alimentary canal with the extra-embryonic splanch- nopleure; but this disappears everywhere as soon as formed, except in the region of the posterior part of the heart and the liver where it forms the doreal mesocardium and gastro-hepatic ligament (Fig. 118), and in the region of the neck of the allantois. 132 THE DEVELOPMENT OF THE CHICK The fore-gut is thus being continually lengthened backwards by fusion of the lateral limbs of the splanchnopleure. At the 31s stage this has proceeded about to the fourteenth somite. At about the 26 s stage the tail-fold appears in the splanchno- pleure, thus establishing the hind-gut (Fig. 70) which gradually f.f.So'pl. So'pl. Sp'pl. Fig. 70. — Median longitudinal section through the hind end of an embryo of about 21 s. an. pi., Anal plate, an. t., Anal tube. p. i. p., Posterior intestinal portal. T. B., Tail-bud. t. f. So'pl., Tail fold in the Somatopleure. t. f. Sp'pL, Tail fold in the splanchnopleure. Other abbreviations as before. elongates forwards. There remains then an open portion of the alimentary tract, where its walls are continuous with the extra- embryonic splanchnopleure or 3'olk-sac. This is known as the yolk-stalk. The entrance from the yolk-sac into the fore-gut is known as the anterior intestinal portal, and that from the yolk-sac into the hind-gut as the posterior intestinal portal (Fig. 70). At the 27 s stage the yolk-stalk is long and narrow (Fig. 106) ; the stems of the splanchnic (omphalo-mesenteric) veins run to the heart in its anterior portion, and the omphalo-mesenteric arteries pass out about its center. As it gradually closes, the stems of the omphalo-mesenteric arteries and veins are brought closer together. At about five days it becomes a tubular, thick- walled stalk, connecting intestine and yolk-sac, and so remains throughout embryonic life. The limiting sulci in the somatopleure lead to the formation of the body- wall. In the trunk the somatopleure is separated from the splanchnopleure by the coelome (Fig. 69), and the folds in the somatopleure take the same general direction as those in the splanchnopleure; they thus lead to the formation of a tube (body-wall) outside of a tube (alimentary canal), the intervening FROM TWELVE TO THIRTY-SIX SOMITES 133 cavity being the body-cavity. The unclosed part of the bod}- wall is continuous with the extra-embryonic somatopleure, more specifically the amnion (see below), and this connection is known as the somatic stalk or umbilicus. The yolk-stalk and neck of the allantois pass out of the body-cavity through the somatic stalk, which therefore remains open until near the end of incubation. The Turning of the Embryo and the Embryonic Flexures. We have described the separation of the embryo from the extra- embrvonic blastoderm without reference to its turnino; from a prone to a lateral position or to the formation of the flexures of the entire head and body that are so characteristic of amnio te embryos generally. These changes begin about the 14 s stage and are first indicated by a slight transverse bending of the origi- nally straight axis of the head in the region of the mid-brain (Fig. 67). By means of this bending, known as the Qranial flex- ure, the fore-brain is directed toward the yolk; but almost simul- taneously another tendency manifests itself, viz., rotation of the head on its side, at first affecting only the extreme end. (See Figs. 71, 73, 99, etc.) By the 27 s stage these two processes have resulted in the conditions shown in Fig. 105: by the cranial flexure the fore-brain is bent at right angles to the axis of the embryo, and owing to the rotation the head of the embryo lies on its left side. But inasmuch as the trunk is still prone on the surface of the volk the axis of the embrvo is twisted in the inter- mediate region. This twist is transferred farther and farther backwards as the turning of the head gradually involves the trunk, until finally, at about ninety-six hours, the embryo lies entirely on its left side. Exceptionally the rotation may be in the inverse direction (heterotaxia) ; in such cases it is often associated with situs in- versus viscerum. Heterotaxia has been produced experimentally (Fol and Warynsky). After the appearance of the cranial flexure a second trans- verse flexure appears in the embryo, this time at about the junction of head and trunk, hence known as the cervical flexure (Figs. 73, 99, etc.). This flexure gradually increases in extent until the head forms a right, or even smaller, angle with the trunk; thus the fore-brain is turned to such an extent that its anterior end points backwards and its ventral surface is opposed to the ventral surface of the throat (Fig. 117). 134 THE DEVELOPMENT OF THE CHICK KFAm, S.16. Pr.atr: Fig. 71. — Entire embryo of 16 s, drawn from above as a transparent object. Note the cranial flexure, also the rotation of the head on its left side, au. P., Auditory pit. F. B., Fore-brain. H. B. 1, First division of the hind brain. H. F. Am., Head-fold of the amnion. Hm. F., Hyomandibular furrow. Pr'am., Proam- nion. M. B., Mid-brain, op. Ves., Optic vesicle, pr. str., Primitive streak, s 2, s 4, s 16, Second, fourth, and sixteenth somites. V. o. m., omphalo-mesenteric vein. VII-VIII, The acustico-facialis primordium. IX-X, Primordium of the glossopharyngeus and vagus. The entire trunk tends also to bend ventrally, i.e., to develop a dorsal convexity, and this approximates its posterior end to the tip of the head. These flexures are characteristic of amniote FROM TWELVE TO THIRTY-SIX SOMITES 135 vertebrate embryos; the cause appears to lie in the precocious development of the central nervous system, of which more here- after. Only the cranial flexure remains as a permanent con- dition. II. Origix of the Embryonic Membranes The period from about 12 to 36 somites also includes the early history of the embryonic membranes, amnion, chorion, yolk-sac and allantois. The first three arise from the extra-embrvonic blastoderm, and the allantois arises as an outgrowth of the ven- tral wall of the hind-gut. -Pr'am. V.Ao. Fig. 72. — The head of the same embryo from below. a. i. p., Anterior intestinal portal. B. a., Bulbus arteriosus. Inf., Infimdibulum. or. pi., Oral plate. Tr. a., Trunciis arteriosus. Ven., Ventricle, v. Ao., Ventral aorta. Origin of the Amnion and Chorion. The amnion is a thin membranous sac, forming a complete investment for the embryo and continuous with the body-wall at the umbilicus; it lies beneath the chorion to which it remains attached throughout incubation by a plate of tissue (sero-amniotic connection), and it arises in common with the chorion from the extra-embryonic somatopleure. The entire somatopleure external to the embryo is used up in the formation of these two membranes. The amnion arises from a portion immediately adjoining the embryo itself; the remainder of the somatopleure peripheral to the amniogenous part forms the chorion. Thus the extra-embryonic somatopleure may be divided into two zones; an amniogenous zone immediately adja- 136 THE DEVELOPMENT OF THE CHICK lenc. Vel.tr. Cr.Fl. Me ten c. My e lenc./. au.P. Myefenc.2. 3.Z. ^^f^—A.o.m. Fig. 73. — Entire embryo of 20 s, viewed as a transparent object from above. The cranial flexure and the rotation of the head of the embryo have made considerable progress. A. o. m., Omphalo-mesenteric artery. Cr. Fl., Cranial flexure. D. C, Duct of Cuvier. Dienc, Diencephalon. Mesenc, Mesencephalon. Metenc, Metencepha- lon. Myelenc. 1, and 2, Anterior and posterior divisions of the myelencepha- lon. Telenc, Telencephalon. Vel. tr., Velum transversum. Other abbrevia- tions as before, x 30. cent to and surrounding the embryo, and a choriogenous zone, comprising the remainder. The method of formation of amnion and chorion is as follows: FROM TWELVE TO THIRTY-SIX SOMITES 137 (a diagrammatic outline is first given and a detailed description follows). The somatopleure becomes elevated m the form ot a told surrounding the embryo; this fold begins fii-st m front ot the head of the embryo as the head-foU of the amnion, which lenc: — ojo.l^es. Mesenc op.Yes Pjq 74, _ Head of the same embryo from the ventral side. Abbreviations as before. V /sfh. ///"Am. Rec.opf Fig 75. - Median sagittal section of the head of an embryo of 18 s^ H. F. Am., Head-fold of the amnion. P^. Pharynx^ Isth Region o the isthmus pr'o. g., Preoral gut. or. pi., Oral plate. Rec. opt., Kecessus opticus! S 'v., Sinus venosus. Other abbreviations as before. immediately turns backwards over the head, forming a complete cap (Fio-s 67, 71, 75, etc.); the side Umbs of the head-fold are then elongated backwards, and are here known as the lateral fold, of the amnion; these rise up and arch over the embryo 138 THE DEVELOPMENT OF THE CHICK (Figs. 109 and 110). In each fold one can distinguish an amniotic or internal limb, and a chorionic or external limb meeting at or near the angle of the folds, the line of junction being marked by an ectodermal thickening, the ectamnion. Fusion of the right and left lateral folds begins at the head-cap, and progresses backwards in such a way that the right and left amniotic limbs become continuous with one another, similarly the right and left chorionic limbs; and, when fusion is complete, the amnion and chorion become separate continuous membranes. In this way the amnion extends, by the 27 s stage, back to the seventeenth somite (Fig. 105). At this time a new^ fold arises behind the rudimentary tail-bud and covers the tail precisely as the head- fold covers the head (Fig. 105); the tail-fold of the amnion then apparently is prolonged forward a short distance and soon meets the anterior lateral folds, forming a continuous lateral fold. Fu- sion continues until, about the 31 s stage, the opening into the am- niotic cavity is reduced to a small elliptical aperture lying above the buds of the hind-limbs (Fig. 99). This then rapidly closes, but a connection, sero-amniotic connection, remains at the place of final closure. Elsewhere the separation of chorion and amnion is complete. The formation of the amnion is an extremely interesting process from the standpoint of developmental mechanics, and involves a number of details that are best understood after such a general review of the process as has been given in the preceding paragraphs. Returning then to the 12 s stage for consideration of these details, we must first note that the extension of the meso- blast prior to this period has left an area situated in front of the head free from mesoblast (Figs. 65, 67, 71, 75, etc.). This area, in which the ectoderm and entoderm are in contact, is known as tlie proamnion. The formation of the amnion begins within this area by a thickening in the ectoderm (ectamnion) near the anterior boundary of the proamnion at a stage with about eight or nine somites. The thickening, which is very narrow, extends right and left, and turns backwards along the sides of the head to about the region of the middle of the heart, gradually becoming more peripheral in position and fading out (Fig. 76). It represents the junction of the amniogenous and choriogenous somatopleure and thus corresponds to the angle of the future amniotic folds. The head of the embryo lies in a FROM TWELVE TO THIRTY-SIX SOMITES 139 e.a. depression bounded in front by the ectamnion, and on the sides by the amnio-cardiac vesicles of the body-cavity (Fig. 65). The floor of the depression is the proamnion. Just before the for- mation of the head-fold proper, the ectamnion in front of the head becomes irregularly thickened to such an extent as sometimes to present an actually villous surface (Fig. 77; cf. Fig. 67). The head-fold of the amnion begins to form at about the same time as the cephalic flexure. The great expansion of the body-cavity on each side of the head (amnio-cardiac vesicles) causes an elevation of the anterior angle of the ectamnion, and a pocket is formed by fusion of its lateral limbs. This slips over the head of the embryo with aid of the ventral flexure of the head just developing. Inasmuch as the anterior angle of the ectamnion is in the pro- amnion, where there is no mesoderm, and where the ectoderm is in immediate contact with the entoderm, the ento- derm as well as the ectoderm of the pro- amnion is drawn into the head-fold, so that the latter is not at first a fold of Fig. 76. — Entire embryo of 13 s, to show the rela- tions of the ectamnion. a. c. Inner margin of the somatopleure. But in the chick the amnio-cardiac vesicles. proamniotic part of the head-fold is e. a Ectamnion i^ ^ A. Region of the soma- never very extensive and does not at any topleure destined to form time extend back of the beginning of the body-wall. . B. Amniogenous soma- the mid-brain. Moreover, it is soon m- topleure. vaded (Fig. 75) by the body-cavity, and ^^ ^^^^^''^'^''^^^'''^^ ^°"'^" then the entoderm is withdrawn and becomes part of the general splanchnopleure. The proamnion ventral to the head is not invaded by mesoderm until a much later period. The ectodermal thickening marking the junction of amniotic and chorionic somatopleure extends backwards very rapidly and always precedes the origin of folds in any region. The lateral folds themselves appear to owe their origin to the progressive 140 THE DEVELOPMENT OF THE CHICK fusion of the ectodermal thickenings of the opposite sides, beginning at the posterior angle of the head-fold and proceeding backwards. The energy of fusion is sufficient in itself to lift the somatopleure up in the form of a fold around the body of the embryo. Thus new parts of the ectodermal thickening are con- stantly being brought together and the fusion progresses steadily, and this in its turn prolongs the lateral amniotic folds. These possess no independent power of elevation of any considerable amount, for, when the initial fold of one side is destroyed by cauterization, the fold of the opposite side remains as an insig- nificant elevation in the somatopleure a long distance lateral to the embryo. ? Fig. 77. — Transverse section through the anterior angle of the ectamnion a few sections in front of the tip of the head. Stage of 14-15 s. b. c, Extra-embryonic body-cavity, c, Cavity in the entoderm, e. a., Ectamnion. The tail-fold arises in an analogous manner to the head-fold, except that there is no proamnion here. The progress of the various folds and their final fusion follows from what has already been said. Practically all of the somatopleure of the pellucid area is amniogenous with the exception, naturally, of that part internal to the limiting sulci that forms the body-wall. What effect has the turning of the embryo on its left side on the amniogenous somatopleure? We will suppose that the latter is primitively of equal width on both sides and that the notochord represents approximately the axis of rotation. During the process of rota- tion, the embryo sinks and the lateral limiting sulci become deeper. A direct consequence of the rotation must be, therefore, a strong tension on the somatopleure belonging to the under (left) side, a-h, and practically none on the upper (right) side, c~d. (See Fig. 78 A, B). / FROM TWELVE TO THIRTY-SIX SOMITES 141 Even though the difference may be partly compensated by drawing of the embryo to tlie left, the tendency ^YOuld be to stretch a-h. If there were no such compensation and a and h were practically fixed points, the length of a-h at the conclusion of the rotation would much exceed that of c-d (Fig. 78 B), and J^ a C Fig 78. A, B, and C. Diagrams to represent the effect of rotation of the embryo on the amniogenous somatopleure. a represents in all figures the position of the ectamnion on the left (lower) side; d represents in all figures the position of the ectamnion on the right (upper) side, b and c repre- sent the junction of amnion and body-wall on left and right sides respectively. In Fig. A, a-h and c-d are equal. In Fig. B, rotation of the embryo is assumed to have taken place without formation of the amnion; the distance a-b has become greater than c-d. In Fig. C is represented rotation of the embryo with synchronous formation of the amniotic folds, as is actually the case; c-d is inevitably thrown into secondary folds. The vertical lines at the extreme right and left represent the margins of the pellucid area. if, during this process, there were actual independent growth of a-h and c-d, the latter would of necessity be thrown into folds, but not the former. Finally, if the amniotic folds were forming at the same time (as is actually the case), the right one would 142 THE DEVELOPMENT OF THE CHICK inevitably be thrown into secondary folds by the approximation of points c and d (Fig. 78 C). Study of the fusion of the amniotic folds in actual sections shows, that the line of fusion of the opposite amniotic limbs is over the dorsal surface of the embryo only so long as the latter lies flat on the yolk; it does not follow the turning of the embryo on to its left side, and the consequence is that, after rotation of the embryo, the line of fusion lies over the upper (right) side of the embryo, often opposite the horizontal level of the intestine (Fig. 79). Thus one fold of the amnion passes all the way from the under side over the back of the embryo and around on the other side to the line of fusion, and thus is several times as long as the opposite limb. Moreover, the amniotic fold of the right Fig. 79. — Section of an embryo of about 60 hours to show the sec- ondary fold (s. f .) of the amnion on the right side, e. a., Ectamnion. s. f., Secondary fold. 1., Left, r., Right. side is invariably thicker than that of the left side, and is always thrown into secondary folds at the place of turning (Fig. 79). These conditions are satisfactorily explained, as noted above, by the mere turning of the embryo on its side. One must therefore distinguish in the upper limb of the am- nion two kinds of folds: (1) The ordinary amniotic fold induced by the fusion of the right and left folds, and (2) secondary folds formed simply by the process of twisting of the embryo. These secondary folds of the amnion are very transitory, except in two regions: (1) Above the hind end of the heart (apex of ventricle), and continuing a short distance behind it; (2) in the region immediately in front of the allantois, at sixty to seventy hours, thus in the neighborhood of the final closure of the amniotic FROM TWELVE TO THIRTY-SIX S0:MITES 143 folds. The former are of very constant occurrence and persist a long time (Fig. 93). Elsewhere the effect of the twisting of the embryo is rapidly compensated so that the secondary folds of the right half of the amnion do not persist long. The subsequent history of the amnion and chorion is given in another place. It should be noted here that the chorion, at the stage of seventy-two hours, is continuous peripherally with the splanchnopleure at the margin of the vascular area, and that it becomes separate from it only as the body-cavity extends more and more peripherally. The sero-amniotic connection remains throughout the entire embryonic period and modifies in an important fashion the subsequent history of the membranes. The yolk-sac is the name given to the extra-embryonic splanchnopleure, because in the course of expansion of the blasto- derm and extension of the extra-embryonic body-cavity over the surface of the yolk, it finally becomes a separate sac enclosing the yolk. It remains connected by the yolk-stalk with the intes- tine until finally, some time after hatching, it is absorbed com- pletely. The yolk is absorbed by the entodermal lining and is carried to the embryo in solution by means of the vitelline veins. Origin of the Allantois. The allantois arises as a diverticulum of the hind-gut soon after the formation of the latter by the tail- fold. It is not indicated before the formation of the tail-fold as stated by some authors, but the tube identified by them as the primordium of the allantois at this early stage is really the in- testinal diverticulum leading to the anal plate (Fig. 70). At the stage of twenty-eight somites the allantois is indicated by the depth of the hind-gut, the ventral portion of which in front of the anal plate soon becomes constricted from the upper portion, and forms the primordium of the allantois. In longitudinal sec- tions of an embryo of about thirty-five somites it can be seen to include nearly the entire floor of the hind-gut between the anal plate and the posterior intestinal portal (Fig. 80). It is lined with entoderm and has a thick mesodermal floor in which numer- ous small blood-vessels are already present. A transverse section (Fig. 81) shows that the thick mesodermal wall is broadly fused with the somatopleure in the region of the neck. In other words, the allantois is developed within the ventral mesentery. It will also be seen by comparing these figures that the amnion 144 THE DEVELOPMENT OF THE CHICK arises from the neck of the allantois both behind and also at the sides, (cf. Fig. 82.) During the fourth day the distal portion of the allantois pushes out into the portion of the extra-embryonic body-cavity beneath the hind end of the embryo and rapidly expands to form a relatively large sac. But the neck of the allantois remains embedded in the ventral mesentery and does not expand; the terminal portion of the intestine has in the meantime formed Ect -3pl'pJ. Mesam- Fig. 80. — Sagittal section through the tail of an embryo of about 35 s. All., Allantois. An. pi., Anal plate, c. C, Central canal of the neural tube. CI., Cloaca. Ectam., Ectoderm of the amnion. Mesam., Mesoderm of the amnion, p'a. G., Post-anal gut. p. i. p., Posterior intestinal portal, s. A., Segmental arteries. Other abbreviations as before. the primordium of the cloaca, from which, therefore, the neck of the allantois appears to arise (Fig. 183); at all stages of incuba- tion the neck of the allantois forms an open connection between the cloaca and the allantoic sac. The Umbilicus. The closure of the body-wall progressively reduces the communication between the embryonic and extra- embryonic body-cavity to a narrow chink between the yolk-stalk FROM TWELVE TO THIRTY-SIX SOMITES 145 and allantoic stalk on the one hand and the attachment of the amnion on the other. The umbilical cord thus consists of an outer tube continuous with the body-wall, enclosing the yolk- stalk and the stalk of the allantois, together with the arteries and veins of yolk-sac and allantois. It is important to bear in mind that in the region of the neck of the allantois the amnion is attached to the latter at the sides and behind; only the anterior wall of the allantoic stalk is free (Fig. 82). In other words, the somatic umbilical stalk is fused with the lateral and caudal wall of the neck of the allantois, a relation that is common to all amniota. Am. Fig. 81. —Transverse section through the hind-gut and allantois of an em- bryo of 35 s; the section passes through the thirtieth somite. Details diagrammatic. All., Allantois. H. G., Hind-gut.^ L. B.,^ Leg bud. v.^ M., Ventral mesentery. W. D., Wolffian duct. Other abbreviations as before. Summary of Later History of the Embryonic Membranes. The full history of the embryonic membranes will be given later (Chap. VII), but it seems desirable to give an outline here in order to avoid repeated recurrence to this subject. The extension of the body-cavity in the blastoderm is at first very rapid, but about the fifth day it becomes slow, and the yolk-sac is never com- pletely separated from the chorion. The allantois extends out into the extra-embryonic body-cavity as a small pear-shaped vesicle by the end of the fourth day. It then enlarges very rapidly and extends in the form of a flattened sac over and around the embryo immediately beneath the chorion with which it forms 146 THE DEVELOPMENT OF THE CHICK an inseparable union. As the extra-embryonic body-cavity extends, the allantois continues its expansion between the chorion and the yolk-sac, and finally wraps itself together with a duplica- tion of the chorion, completely around the albumen of the egg, which has become very viscid, and aggregated in a lump opposite to the embryo. The allantois is very vascular from the start, and serves as an embryonic organ of respiration. It also receives the excretion of the embryonic kidneys and absorbs the albumen. Fig. 82. — Model of the caudal end of a four-day chick to show the relations of the amnion to the allantois and umbilicus. (After Ravn.) All., Neck of the Allantois. Am., cut surface of the amnion. A. o. m., Omphalo-mesenteric artery, an. pL, Anal plate. L. B., cut surface of leg bud. T., Tail. The yolk-sac becomes much shriveled during incubation owing to absorption of its contents, and on the last day of incubation is withdrawn into the boch-cavity through the umbilicus, which finally closes. The chorion, amnion, and allantois shrivel up when the chick begins to breathe air, and are cast off with the shell at hatching. FROM TWEL^'E TO THIRTY-SIX SOMITES 147 III. The Nervous System The Brain. The description of the nervous system in the pre- ceding chapter forms our starting-point. During the period now under consideration the foundation of the main parts of the adult brain are laid down, and its five chief divisions become sharply characterized. It is important to correlate these with the earliest morphological characters (original anterior end of medullary plate, neuromeres, etc.) in order to trace these fundamental landmarks through to definitive structures. As we have already seen, the primary fore-brain includes the first three neuromeres, the mid-brain the fourth and fifth, and the hind-brain the sixth to the eleventh, as well as the region opposite to the first four mesoblastic somites. It is clear that a second point of fundamental morphological significance is the original anterior end of the medullary plate which would naturally form the center for a description of the anterior part of the neural axis, if recognizable throughout the development. This point may be recognized for a considerable period after the closure of the anterior part of the neural tube, as the ventral end of the anterior cerebral fissure (Fig. 62), opposite the center of the primary optic vesicles, thus in the region of the recessus opticus (Figs. 87 and 88), which is to be regarded as marking the original anterior end of the neural axis. Even after closure of the anterior cerebral fissure a connection remains at its dorsal end between the ectoderm and the neural tube. To this we may apply the name neuropore, though no actual opening is found here at this time. The median stretch of tissue between the recessus opticus and the neuropore constitutes the lamina terminalis which remains as the permanent anterior wall of the neural tube. It must not be forgotten that the original anterior end of the medullary plate lies at the ventral end of the lamina terminalis, i.e., in the re- cessus opticus. A third landmark of fundamental morphogenic significance is the infundibulum, which coincides in position, as we have seen, with the anterior end of the notochord. Thus we may distinguish prechordal and suprachordal portions of the neu- ral axis (cf. Fig. 67). Dorsal and Ventral Zones in the Wall of the Brain. The con- ception of His, that the walls of the neural tube may be consid- ered as formed of four longitudinal strips, viz., floor, roof, and 148 THE DEVELOPMENT OF THE CHICK Fig. 83. — Five stages in the history of the neuromeres of the brain of the chick. (After Hill.) All figures drawn from preparations of the embryonic brain dissected out of the embryo. A. Neural groove in an embryo with 4 somites. Right profile view, x 44. B. Brain of a 7 s embryo, 26 hours old. Dorsal view; the three anterior neuromeres are practically obliterated, x 44. C. Brain of 14 s embryo. Dorsal view, x 44. The neuromeres have now disappeared in the mid-brain region. D. Right side of the brain of a chick embryo, 47 hours old. x 44. E. Right side of the brain of an embryo, 80 hours old. x 17. 1-11, Neuromeres 1 to 11. Ill, V, VH, interneuromeric grooves. A'f., Root of acustico-facialis (seventh and eighth cranial nerves), au. vs., Audi- tory pit. ep., Epiphysis, r., Groove between the tel- and diencephalon. s., Groove between the par- and synencephalon. Tr., Root of trigeminus. FROM TWELVE TO THIRTY-SIX SOMITES 149 two lateral walls, is a useful one. Each lateral wall may also be divided into a dorsal and ventral zone, the former of which is related to the sensory nerve roots and the latter to the motor. Cerebral Flexures. The cerebral flexures correspond to the cranial and cervical flexures of the entire head already described. Their form and rate of progress may be more readily learned from the figures (Figs. 67, 73, 83, etc.) than from any verbal description. Only the cranial flexure is permanent, and the angle thus formed ventrally in the floor of the mid-brain is known as the plica encephali ventralis. A third flexure is formed later in the anterior portion of the hind-brain, hy a ventral bending of the floor which is barely indicated in the period now^ under de- scription, but becomes much more pronounced later; this is known as the pontine flexure. We may now take up separately the changes in each of the primary cerebral vesicles. The Prosencephalon. The principal events in the early de- velopment of the prosencephalon are: (a) the separation of the optic vesicles; (6) the delimitation of the tel- and diencephalon; (c) special differentiation of the walls. (a) A section across the optic vesicles of the 12 s chick shows the prosencephalon as a central division with its cavity widely confluent with the cavities of the optic vesicles. This wide com- munication is rapidly narrowed by a ventrally directed fold of the roof at the line of junction of the optic vesicles and prosencephalon proper (Fig. 84) ; the fold also involves to a certain extent the anterior and posterior line of junction. In the 20 s embryo the connection of the optic vesicles and prosencephalon has been re- duced in this way to about one third of its original diameter (from actual measurements), forming a narrow tubular stalk, the optic stalk, attached to the ventral portion of the fore-brain (Figs. 73 and 74); the cavities of the optic vesicles are still con- tinuous through the stalk with the cavity of the prosencephalon, dipping into the recessus opticus; the ventral wall of the optic stalk thus becomes continuous with the floor, and the dorsal wall with the lateral wall of the prosencephalon (Fig. 84). Growth of the mesenchyme situated above the original optic stalk appears to be an active factor in the separation; at least it grows at a rate sufficient to fill in the space produced by the constriction. At the same time there is a slight increase in the dorso-ventral 150 THE DEVELOPMENT OF THE CHICK diameter of the fore-brain itself, though this is relatively slight up to twenty somites, but it enhances the general effect of the change in position of the optic stalk. The subsequent history of the optic vesicles is given beyond. (h) The deUmitation of the tel- and diencephalon is initiated by a forward expansion of the anterior end of the primary fore- brain, which becomes the telencephalon or secondary fore-brain, the remainder being then known as the diencephalon or 'tween brain. The expansion proceeds very rapidly from the 14 s stage, and it is probable that it involves only the dorsal zones. It is, Ectdw Am.F. £5. B.C. ^^^^ :-:M yfm§ ^O:... .. ^>^ ■ -^ ■" '^\kj^&^i-^ '4, \>, Fig. 84. — Transverse section through the fore-brain and optic vesicles of a 16-s embryo. Am. F., Amniotic fold. Ectam., Ectamnion. L., Left side, op.st., Optic stalk. R., Right side. Other abbreviations as before. however, difficult to establish an exact line of demarcation be- tween the two subdivisions of the primary fore-brain, until about the 18 to 20 s stage, when a slight transverse fold or indentation in the roof (velum transversum) gives a dorsal landmark (Figs. 73, 85); the recessus opticus forms the ventral boundary between the two. The velum transversum lies a considerable distance above the dorsal end of the lamina terminalis, but it is difficult to say just how far, owing to the inclefiniteness of this point for some time after the disappearance of the neuropore. A line drawn between the velum transversum and the recessus opticus mav be taken as the boundary between the two divisions of the FROM TWELVE TO THIRTY-SIX SOMITES 151 primary fore-brain; but, owing to the simultaneous lateral expan- sion of the telencephalon, the line of separation in the lateral walls forms a curve with the convexity directed posteriorly (Figs. 83 E and 86). (c) The next stage in the differentiation of the telencephalon (20 s to 36 s) is characterized by a rapid expansion and evagina- tion of its lateral w^alls, while the entire median strip extending from the velum transversum to the recessus opticus remains prac- FiG. 85. — Optical sagittal section of the head of an embryo of 22-23 s. The heart is represented entire. Atr., Atrium. Hyp., Hypophysis. Inf., Infundibiilum. Md., Man- dibular arch. or. pl.^ Oral plate. Pr'o. G., Preoral gut. Th., First in- dication of thyroid. T. p., Tuberculum posterius. V. tr., Velum transversum. Other abbreviations as before tically unaltered; and thus acts like a rigid band stretched over the surface between these tw^o points. The effect of this is to form a pair of outgrowths that soon begin to project dorsally, anteriorly, and posteriorly (Fig. 83 E); these are the primordia of the cerebral hemispheres, the cavities of which thus appear as lateral diverticula of the median cavity of the telencephalon (Fig. 86). The central part of the telencephalon may be called the telencephalon medium, and the lateral outgrow^ths the hemi- spheres. The walls of the hemispheres become considerably thicker in this period, but quite uniformly at first, so that the distinction between mantle and basal ganglia is indicated only by position. (See Chap. VIII.) 152 THE DEVELOPMENT OF THE CHICK The median strip includes the tela choroidea, beginning at the diencephalon, and the lamina terminalis, which ends at the recessus opticus. These divisions are of great prospective signifi- cance, though at the stage of 36 s they are but sHghtly differen- tiated, save by their position. A shght thickening of the lamina terminalis just in front of the recessus opticus marks the site of the future anterior commissure (Figs. 87 and 88). Metenc. Mesenc \r C/) opt. /?PC Opt D/enc Vtr\~, Te/enc Fig. 86. — Inner view of the brain of a chick of about 82 hours, drawn from a dissection. Ch. opt., Chiasma opticus. Ep., Epiphysis (pineal gland). Isth., Isth- mus. PI. enc. v., Plica encephali ventralis. Rec. opt., Recessus opticus. V. tr., Velum Transversum. Other abbreviations as before. The Diencejphalon. The portion of the primary fore-brain pos- terior to the telencephalon is known as the diencephalon. It in- cludes the second and third neuromeres and probably also the ventral zones and floor of the first (Fig. 83). A slight constriction in the roof that appears about the 18 to 20 s stage near the junc- tion of the middle and last third may represent the boundary be- tween the second and third neuromeres; this persists for a long time and may be traced in the lateral walls to the region of the FROM TWELVE TO THIRTY-SIX SOMITES 153 infimdibulum (Fig. 83 E) ; thus the diencephalon may be divided into an anterior and posterior division, parencephalon and synen- cephalon (Kupffer) (Fig. 87). The optic stalks are attached to the floor and ventral zones at the extreme anterior end. The diencephalon includes part of the roof, floor, and dorsal and ven- tral lateral zones of the original neural tube. These may be de- scribed as follows (Figs. 87 and 88): 5tOJ77. Oes. - Jyy?e/)c Am. Fig. 87. — Optical longitudinal section of the head of an embryo of 30 s. The heart is represented entire. Atr., Atrium (auricles). B. a., Bulbus arteriosus. D. v., Ductus venosus. Lg., Laryngo-tracheal groove. Oes., Oesophagus, or. pi., Oral plate, which has begun to rupture. Parenc, Parencephalon. Ph., Pharynx. Stom., Stomach. Synenc, Synencephalon. Th., Thyroid. S. v., Sinus venosus. Ven. R., Right ventricle. Other abbreviations as before. The roof rises quite sharply from the velum transversum, and is indented between the parencephalic and synencephalic divi- sions as already noticed. It is relatively thin. About the 30- 35 s stage the epiphysis (pineal body) begins to form as an evagination from about the middle, and by the 36 s stage is a small hemispherical protuberance (Figs. 86 and 88). The floor becomes extremely thin in the center of the recessus opticus, which marks its anterior end; immediately behind this is a sudden and 154 THE DEVELOPMENT OF THE CHICK conspicuous thickening, the optic chiasma, which is continued as a ridge in the lateral ventral zones on each side (Fig. 86). The infundibulum follows just behind this, and constitutes a considerable pouch-shaped depression from which the saccus infundibuli grows out later. The posterior wall of this depression rises sharply and joins the thickened tuberculum posterius which is the end of the floor of the diencephalon. The diencephalon is compressed laterally (Fig. 97); the dorsal zones are slightly thickened, indicating the future thalami optici. Fig. 88. — Optical longitudinal section of the head of an embryo of 39 s. Abbreviations as before. The hypophysis should be mentioned here, although it is not embryologically a part of the brain. It arises as a median tubu- lar invagination of the ectoderm of the ventral surface of the head immediately in front of the oral plate at about the 20 s stage (Fig. 85), and grows rapidly inward in contact with the floor of the diencephalon. At about the 30 s stage its end reaches nearly to the infundibulum (Fig. 87). At first part of its wall is formed by the oral plate, and when this ruptures the effect is to shorten the apparent length of the hypophysis (Fig. 88). At about the 36 s stage its distal portion flattens laterally FROM TWELVE TO THIRTY-SIX SOMITES 155 and shows indication of branching. Subsequently it becomes much branched and quite massive and unites with the infun- dibulum to form the pituitary body. (See Chap. VIII.) The Mesencephalon. This portion of the brain comes to occupy the summit of the cranial flexure, which indeed owes its origin largely to the rapid growth in extent of the roof of the mesencephalon. In longitudinal section it thus appears wedge- shaped, with short floor and long arched roof (Figs. 87 and 88). Its walls remain of practically uniform thickness up to the seventy-second hour. The lateral walls expand more rapidly than the roof and thus form the optic lobes. But these are barely indicated at the 36 s stage. Isthmus. The great expansion of the mesencephalon does not involve the portion immediately adjacent to the hind-brain^ which is henceforth known as the isthmus (Figs. 87, 88). The Rhombencephalon {Primary Hind-brain) . Two divisions of the embryonic brain arise from the rhombencephalon, viz., the metencephalon and the myelencephalon; the former becomes the region of the cerebellum and pons of the adult brain, and the latter the medulla oblongata. The metencephalon is a relatively short section of the original rhombencephalon, and includes only the most anterior neuromere of the rhomben- cephalon or the sixth of the series (Fig. 83 D, E). It may be distinguished at the beginning of the period under consideration by the fact that its roof remains as thick as that of the mesen- cephalon. At the end of this time, i.e., seventy-two hours, the roof in sagittal sections appears to rise sharply from the isthmus and thins towards the summit, where it passes into the thin epi- thelial roof of the myelencephalon (Figs. 87 and 88). The rudi- ment of the cerebellum is slightly thicker on each side of the middle line at seventv-two hours. The myelencephalon becomes sharply characterized by the thinness of its roof and thickening of ventral lateral zones and floor. The epithelial roof has a triangular form, the base resting against the metencephalon. The neuromeres remain very distinct (Figs. 83, 89), but change their form. Up to about twenty-three somites they still form external expansions, but as the wall thickens the external surface becomes smooth, and the neuro- meres may now be recognized as a series of concavities in the lateral wall, with intervening projections (Fig. 89). The arrange- 156 THE DEVELOPMENT OF THE CHICK ment of the nuclei leaves thin non-nucleated strips (septa) be- tween adjacent neuromeres. The interneuromeric projections are most pronounced laterally and fade out dorsally and ventrally. Behind the neuromeric portion of the hind-brain is a portion extending to the posterior end of the fourth mesoblastic somite from which the twelfth cranial nerve arises. The Neural Crest and the Cranial and Spinal Ganglia. The cranial and spinal ganglia owe their origin to a structure known as the neural crest, which is a practically continuous cord of cells, lying on each side in the angle between the neural tube and the ectoderm, extending from the extreme anterior to the pos- terior end. Like other meristic structures the anterior portion y//-y/// ^// ^/^^/:s^r^ YS Fig. 89. — Frontal section of the hind-brain region of an embryo of about 36 s. Ot., Otocyst. N. 6, N. 7, N. 8, N. 9, N. 10, N. 11, Neuromeres, 6 to 11, according to Hill's enumeration, s. 1, s. 2, s. 3, First, Second, and third somites. V, Primordium of the trigeminus. VII-VIII, Primordmm of the acustico-facialis. of the neural crest is the first to arise (at about 6-7 s stage), and the remainder appears in successive order during or shortly after the closure of the neural tube in each region; thus it is not until after the completion of the neural tube that the last portion of the neural crest is established. But before this time successive enlargements of the cranial part of the crest have formed the primordia of the cerebral gan- glia, and similar successively arising enlargements of the parts of the crest opposite the mesoblastic somites form the rudiments of the spinal ganglia. The intervening portions of the crest form the so-called interganglionic commissures, w^hich subsequently FROM TWELVE TO THIRTY-SIX SOMITES 157 appear to form mesenchyme. The formation of mesenchyme from certain parts of the neural crest is most marked in the region of the brain. The primordia of the gangha contain the cells (neuroblasts) which form the dorsal root fibers of the spinal nerves and parts of certain cranial nerves. They also appear to contain the cells from which the sheaths of the nerve fibers are formed; thus three kinds of cells at least are found in the neural crest, viz., mesenchyme forming cells, neuroblasts, and sheath cells. The Cranial Neural Crest and its Derivatives. The neural crest in the head may be divided into pre- and post-otic divisions, and these arise at different times. n.Cr. Si.'/- cer. op.Ves. -Mhh. Fig. 90. — Transverse section of the fore-brain, and optic vesicles at the stage of 7 s. M'ch., Mesenchyme, n. Cr., Neural crest. Ph., Phar- ynx. Sut. cer., Anterior cerebral suture. X., Mass of cells in which the anterior end of the intestine, the neural tube and the notochord fuse. (1) The pre-otic division, which extends from the extreme anterior end of the neural tube to about the center of the audi- tory pit, is well developed at a stage of 7-8 somites, but it is not found at the 5 s stage. The origin is everywhere the same, viz., from the dorsalmost cells of the medullary plate and the ecto- derm immediately adjacent; it arises at the time of contact of the medullary folds and is thus thickest in the region of the suture. Fig. 90 is a section through the developing optic vesicles, and shows the neural crest continuous with the tube and ectoderm 158 THE DEVELOPMENT OF THE CHICK in the neural suture; it is separated from the mesenchyme in the region of the fore-gut by a considerable space. (We shall call the latter portion of mesenchyme the axial mesenchyme of the head, to distinguish it from the mesenchyme derived from the neural crest, which later lies lat- eral to it, and which may thus be known as the periaxial layer.) The crest may be followed ante- riorly to the extreme tip of the neural tube, and posteriorly to the region of the anterior intesti- nal portal, which lies at about the transverse level of the future au- ditory pit (cf. Fig. 91). In the region of the mid-brain it spreads out laterally until its peripheral cells reach the axial mesenchyme. Goronowitsch divides the pre-otic portion of the neural crest into pri- mary and secondary ganglionic crests, the post-otic portion being the terti- ary crest. According to his account there is a decided difference in time of origin of the primary and second- FiG. 91. - Diagram of the cephalic ary crests; the primary, involving the neural crest of a chick of about region of fore- and mid-bram, aris- 12 s. (After Wilhelm His.) ing before the secondary which in- cludes the region of the trigeminus and acustico-facialis. I have not, however, found such a difference in my preparations. At the stage of 10 somites the cells of the pre-otic neural crest have lost their connection with the neural tube. Behind the optic vesicles they have spread out laterally between the axial mesenchyme and the ectoderm, where they form a prac- tically continuous periaxial layer, distinguishable from the axial mesenchvme bv its greater densitv, and hence deeper stain; but apparently mingling with it at the surface of contact. In the stages immediately following (10-20 s), the portions of the periaxial layer lying above the mandil^ular and the hyoid arches condense and thicken, and form strong cords extending FROM TWELVE TO THIRTY-SIX SOMITES 159 from the superior angles of the neural tube into the arches in question; here they form connections with the ectoderm of the arches, which proliferates so as to contribute to their substance (Fig. 92). Elsewhere the periaxial layer gradually merges with the axial mesenchyme. The periaxial cords are the primordia of the trigeminus and acustico-facialis ganglia, and mark the paths of the trigeminal and facial nerves. Their connection with #5 i^i^'t.f/*"*'- Fig. 92. — Transverse section immediately be- hind the first visceral pouch of a chick embryo of thirteen somites. (After Gorono- witsch.) Note connection of the periaxial cord with the ectoderm of the visceral arch. Ad., Aorta descendens. c. Rounded me- senchyme cells, g. Place where cells derived from neural crest unite with the mesenchyme cells of the periaxial cord. f. Fusion, p. Spin- dle-shaped peripheral mesenchyme cells. the ectoderm in the neighborhood of the first visceral pouch must not be confused with the so-called branchial sense-organs, for the primary connection is soon lost, and secondary connec- tions arise at about the 27 s stage, and constitute the true branchial sense-organs of these arches. 160 THE DEVELOPMENT OF THE CHICK The acustico-facial periaxial cord attains definiteness some time before the trigeminal (cf. Fig. 71), and indeed appears almost from the first as a specially strong part of the periaxial layer: whereas in the region of the trigeminus the cells of this layer are first widely dispersed and secondarily aggregate, between the stages of 14 and 18 somites. Both cords are attached to the brain, the trigeminus to the first neuromere of the myelencepha- lon, and the acustico-facialis to the third (Fig. 83 E). The trigeminal and facial periaxial cords are supplemented^ as we have seen, by proliferations of the ectoderm on each side of the first visceral pouch; the trigeminal cord then enters the mandibular arch, and the facial the hyoid arch, and in the stages between 20 and 27 somites form at least part of the mesenchyme of these arches. The axial mesoblast likewise contributes to the mesenchyme of these arches, and it becomes impossible in later stages to separate these two mesenchymal components. The ganglia proper differentiate from the upper portions of the cords. The trigeminal periaxial cord divides over the angle of the mouth and sends out a process into the rudimentary maxillary process. A third projection of the same cord towards the e3^e forms the path of the ophthalmic division of the trigeminus (Fig. 117). At the stage of about 27 s the trigeminus forms a connection with a thickening of the ectoderm (placode of the trigeminus) situated in front of and above the first visceral cleft; and the facial connects similarly w^ith a larger ectodermal thickening (placode of the facialis) situated on the posterior margin of the uppermost part of the first visceral furrow. These ectodermal thickenings are rudimentarv structures of verv brief duration, representing parts of the sensory canal system of the head of aquatic vertebrates. Their occurrence in the chick is an interest- ing example of phjdogenetic recurrence. A third and fourth like organ arises in connection with the post-otic ganglia. At the stage of 72 hours there are two ectodermal thicken- ings (placodes) in connection with the trigeminus, one in front of the other, derived probably by division of the original first. The facialis placode is more fully developed. (2) The post-otic ganglionic crest is a direct continuation of the pre-otic behind the ear, and it is at first difficult to make an exact boundary between them. At the stage of 13 s the pre-otic crest extends beneath the auditory epithelium nearly to its middle FROM TWELVE TO THIRTY-SIX SOMITES 161 in the form of a thick mass of cells in the roof of the neural tube. Towards the posterior end of the auditory epithelium the crest becomes smaller, and this is the beginning of the post-otic crest. Behind the ear the crest becomes larger again and extends later- ally so as to form a periaxial layer between the ectoderm and the axial mesoblast which extends back, above the first, second, and third somites to the middle of the fourth. The part between the ear and the first somite is, however, by far the best developed, the continuation behind being a relatively slight cord of cells. At about the stage of 17 somites the anterior part of the crest condenses to form a well-defined periaxial cord, which arises from the neural tube above the middle of the auditory pit, arches back behind its posterior margin and extends down into the third visceral arch, w^here it enlarges. This is the glossopharyn- geal periaxial cord. There is an enlarged portion of the crest just behind this overlying the site of the future fourth and fifth arches, but its substance is not yet condensed to form a distinct periaxial cord. At the stage of 20 somites the anterior cardinal vein and the duct of Cuvier form the posterior boundary of the enlarged por- tion of the post-otic crest (Fig. 73). The part of the periaxial layer immediately in front of this is somewhat condensed to form the periaxial cord of the vagus, and this is only indistinctly separated from that of the glossopharyngeus. The formation of the third visceral cleft definitely splits the periaxial layer into the periaxial cords of the glossopharyngeus and vagus (25 s). This division is carried up indistinctly, at first, into the roots which occupy the space between the auditory sac and the first somite. The formation of the fourth visceral pouch similarly divides the distal portion of the vagus cord, so that part of it lies in front of the pouch and part behind. At the stage of seventy-two hours the ganglion petrosum (glossopharyngeus) is definitely formed by an enlargement of the cord just above the third visceral arch, and the ganglion nodosum (vagus), similarly formed from the vagus cord, lies above the fourth visceral pouch, thus extending over the fourth and fifth arches. Branchial sense organs are formed at the dorsal angles of the second and third visceral furrows in connection with the IX and X nerves respectiveh^ It would appear that the neural crest in the head is the 162 THE DE\ ELOPMENT OF THE CHICK source of much of the mesenchyme, and it is an interesting ques- tion whether or not such mesenchyme has a different fate from that of different origin. Nothing definite, however, is known in regard to this, owing to the impossibiUty of separating the various kinds after they have once merged. The Neural Crest in the Region of the Somites. The neural crest is very sUghtly developed in the region of the first five so- mites, which is correlated with the fact that these somites are devoid of ganglia. But the mode of origin is the same through- out the somitic region. Shortly after the closure of the neural tube in any region the neural crest forms an aggregation of cells in the roof, more or less sharply separated from the remainder of the tube both by the arrangement of the cells and also by their lighter stain (Figs. 107, 109, 112, 113). The early history may be followed in a single embryo, by comparing the conditions opposite the last somite with that of more anterior somites where develop- ment is more advanced. Figs. 107, 108, 109, 110 represent transverse sections through the twenty-ninth, twenty-sixth, twentieth, and seventeenth somites of a 29 s embryo. In Fig. 107 the cells of the crest are extending towards the upper angle of the somite, with which they are connected by protoplasmic strands. The aggregation in the roof of the neural tube is thus decidedly diminished; the lateral wings of the crest lie in the angle between the neural tube and the ectoderm. In the twenty-sixth somite (Fig. 108) the lateral wings extend farther from their point of origin, and appear to have a more intimate connection with the myotome. In the more anterior and older somites, twenty and seventeen (Figs. 109 and 110), the process has progressed much farther and the neural crest cells are completely expelled from the neural tube, which closes after them (Fig. 110). A yet later stage is shown in Fig. Ill, through the twenty-third somite of a 35 s embryo. The dorsal commissure uniting the right and left sides of the crest ruptures, and the cells of the crest aggregate so as to form a pair of ganglia in each somite. Thus, although the neural crest is primarily a median structure, it becomes divided into two lateral halves, and although it is primarily a continuous structure it becomes divided into a series of pairs of metameric ganglia. The fate of the interganglionic commissures is conjectural. The ganglia are ill-defined from the mesenchyme when they are first FROM TWELVE TO THIRTY-SIX SOMITES 163 Metenc /3t/). cr.rj. •vr .Ns s .■.: :!c"-r'- ■:.^';^■^w':-,^^^ ■ '■- • ^-7r. ,-. ;^<. \>" v-v .; />- :|? '2;?- i '^:;&$A.o.^f. C'*^ . ■-,- ,W;J1 ■ '-^.'■'' <7 •*?3i ^ i;^- " Wi ..,- .;§ :^ i'%., 1^ iS; 1 Fig. 93. — Entire embryo of 27 s viewed as a transparent object from above. a. a. 1, a. a. 2, a. a. 3, First, second, and third aortic arches. Car., Carotid loop. Ret., Retina. V. C. 1, V. C. 2, First and second visceral clefts. Other abbreviations as before, x 20. 164 THE DEVELOPMENT OF THE CHICK formed, but they rapidly become well differentiated. (See Chap. VIII.) IV. The Organs of Special Sense (Eye, Ear, Nose) Embryologically a sharp distinction must be drawn betw^een the essential percipient part of the organs of sense (retina -of the eye, olfactory epithelium, and epithelium of the membranous laby- rinth) and the parts formed for protection and for the elaboration of function. The sensory part proper is the first to arise in the embryo, and is protected later by modifications of surrounding tissues or parts. We may thus distinguish between primary and secondary parts in the case of all organs of sense. Only the early history of the primary parts falls within the period covered by this chapter, except the formation of the lens in the case of the eye. The Eye. The primary optic vesicles arise, as we have seen, as lateral expansions of the anterior end of the neural tube; their position is indicated by an enlargement of the neural tube even before the meeting of the medullary folds in this region. The shape and relations of the early optic vesicles have already been described and figured. The origin of the optic stalk by constriction of the base of the vesicle was described in a preced- ing section of this chapter (p. 149). The stalks remain attached to the ventral end of the lateral walls of the diencephalon in the region of the recessus opticus, and constitute tubular con- nections between the vesicles and the brain, in the walls of which the optic nerve develops later (Fig. 84). Locy found six pairs of " accessory optic vesicles " occurring in series immediately behind the true optic vesicles; they form low rounded swellings of the side-walls of the neural folds before the true brain vesicles are indicated, and last only about three hours in the chick (twenty-fourth to twenty-seventh hours of incubation). ''Their exist- ence supports the hypothesis that the vertebrate eyes are segmental, and that the ancestors of vertebrates were primitively multiple-eyed." (Locy.) The external surface of the optic vesicle early reaches the ectoderm, to which it appears to be cemented at the 10 s stage. In the 17-18 s stage, the optic vesicles project decidedly behind the attachment of the optic stalk, and the external wall is slightly thicker than that next the brain. The ectoderm then becomes thickened over a circular area in contact with the optic vesicle FROM TWELVE TO THIRTY-SIX SOMITES 165 and this constitutes the primordium of the lens (Fig 94). The thickening of the external wall of the optic vesicle and of the lens primordium now proceed rapidly, and soon an invagination is formed in each (Fig. 95). Fig. 94. — Section through the primordium of the eye of a chick embryo of 21 s. (After Froriep.) d., Distal wall of optic vesicle, p., Proximal wall of optic vesicle. Fig. 95. — Section through the primordium of the eye of a chick embryo at the end of the second day of incubation. (After Froriep.) It is probable that a stimulus is exerted by the optic vesicle on the ectoderm with which it is in contact, causing it to thicken and become the primordium of the lens. This has been demonstrated experimentally to be the case in the embryo of the frog, and the morphological rela- tions are the same in the chick. The invagination of the primary optic vesicle to form the secondary optic vesicle is not mechanically produced by the grow^th of the lens, as some have supposed, for it has been shown (see Fol and Warynsky) that the secondary optic vesicle is formed in the absence of the lens. We may now consider the formation of the optic cup and of the lens separately. The Optic Cup. The invagination of the outer wall of the primary optic vesicle gradually brings this wall into contact with the inner wall and obliterates the primary cavity. Thus 166 THE DEVELOPMENT OF THE CHICK is established the secondary optic vesicle or optic cup. Special attention must be given to the form of the invagination, for this determines relations of fundamental importance. The invagina- tion may be stated to consist of two parts. The first is directly internal to the lens primordium, and the second, which is continu- ous with the first, involves the ventral wall of the primary optic vesicle as far as the optic stalk. The secondary optic vesicle established by these invaginations thus has two openings into its cavity, (1) the external opening, which becomes the pupil of the eye, and (2) the ventral opening, continuous with the pupil, which is known as the choroid fissure. Figs. 96 A, B, and C exhibit these relations better than a detailed description. The choroid fissure is a transitory embryonic structure, sub- sequently closing by fusion of its lips. However, it establishes a relation of fundamental importance in that the ventral wall of the optic-stalk is kept continuous in this way with the inner or retinal layer of the secondary optic vesicle (Figs. 96 B, and 97), and thus a path is provided for the development of the optic nerve (see Chap. IX). It also provides an aperture in the wall of the optic cup for the entrance of the arteria centralis retinae. The optic primordium at the 36 s stage, with the omission of the lens, is composed as follows: (1) Optic-stalk attached to the floor of the brain; this is still tubular. (2) The optic cup or secondary optic vesicle consisting of two layers, viz., (a) a thick internal or retinal layer continuous at the pupil and choroid fissure with {h) the thin external layer. The cavity of the cup is the future posterior chamber of the eye; it has two openings, viz., the pupil filled by the primordium of the lens, and the slit-like choroid fissure extending from the pupil to the optic stalk along the ventral surface. The retinal layer is continuous with the floor of the optic-stalk, and thus with the diencephalon. The optic cup expands with extreme rapidity between the stages of 26 and 36 somites, as may be seen from the figures by comparing the relative size of the lens and optic cup at different stages. The Lens. The invagination of the thickened ectoderm external to the optic vesicle soon leads to the formation of a deep, thick-walled pit which rapidly closes (26-28 somites) and thus FROM TWELVE TO THIRTY-SIX SOMITES 167 forms an epithelial sac, which at first practically fills the cavity ot the optic cup. However, it very soon becomes detached from the posterior wall of the optic cup, which expands with o-reat rapidity, and the lens is left at the mouth of the cup The walls --^--^^5^=^ ^"^-^^^Xr^ — ■ • V N \' -.^ / / C O ^ O t^ 5 L cj o a> " fi C fc. .aj c o ^ • -^ 2 o tn ~ ^ ^ ^ ~ "tj A I I l—i 4J o— c ^ o 9^ .0 c — — +^ ^ -^ ^' "^ tj r O O I— I o a; of the lens sac are at first of practically even thickness (28 s) but by the 35 s stage a great difference has arisen by the elono-a- tion of the cells of the inner wall, which are destined to form lens fibers: the cells of the anterior (outer) wall elongate much 168 THE DEVELOPMENT OF THE CHICK less during this period, and are destined to form the epithelium of the lens (Fig. 97). Intermediate conditions are found around the equator of the lens. The subsequent history is given in chapter IX. The Auditory Sac. At about the 12 s stage the first evidence of the auditory sacs is found in the form of a pair of circular patches of thickened ectoderm situated on the dorsal surface of the head opposite to the ninth, tenth, and eleventh neuromeres, and thus a short distance in front of the first mesoblastic somite; it lies between the rudiments of the acustico-facialial and glosso- pharyngeal ganglia. In the 14 s stage the auditory epithelium Lens. p.C/i Lens . Fig. 97. — Transverse section through the eyes and heart of an embryo of about 35 s. The plane of the section will be readily understood by com- parison of Fig. 117. ch. Fis., Choroid fissure. D. C, Duct of Cuvier. Lg., Lung. pi. gr., Pleural groove. V. c, Posterior cardinal vein. Y. S., Yolk-sac. Other abbreviations as before. is slightly depressed, and in the 16 s stage it forms a wide-open pit. At about the 20 s stage the mouth of the pit narrow^s slightly, and gradually closes (28-30 s), thus forming the auditory sac or vesicle (otocyst) (cf. Figs. 71, 73, 89, and 93). The method of closure of the pit, which is of interest, may readily be observed in mounts of entire embryos; at first the lips fold over most rapidly from the anterior and posterior mar- gins; thus the mouth of the pit becomes elliptical with the long axis vertical (stage of 22 somites) and extending from the apex nearly to the base. The ventral lip then begins to ascend (stage of 24 somites) and the closure gradually proceeds tow^ards the fro:m twelve to thirty-six somites 169 D.cndl apex, so that by the stage of 29 somites the opening is reduced to a minute ellipse situated on the external side of the dorsalmost portion of the otocyst (see Fig. 93). This portion of the otocyst now begins to form a small conical elevation, and the final closure takes place on the external side of this elevation, which is destined to form the endolymphatic duct. The latter remains united to the epi- deTmis at this point for a consid- erable period of time by a strand of cells which may preserve a lumen up to 104 hours (Fig. 98). The final point of closure of the oto- cyst is thug very definitely placed, and it coincides with the middle of the endolymphatic duct, that is, with the junction of the later formed saccus and ductus endolymphaticus. In the Selachia this duct remains in open communication with the exterior throughout life; the rela- tively long persistence of its con- nection with the epidermis in the chick may thus be interpreted as a phylogenic reminiscence of the an- cestral condition. The Nose (Olfactory Pits). At about the 28 s stage, the ectoderm on the sides of the head a short dis- tance in front of the eyes appears thickened. Two circular patches of the surface to the otocyst. D. ectoderm are thus marked off, the end'l. Endolymphatic duct D, . . . Dorsal. Ect., Ectoderm of the begmnmg of the olfactory epithe- surface of the head. Gn., Audi- Hum; at first this grades almost im- tory ganglion. L., Lateral. M., ... . - .... Median, v., ventral, perceptibly nito the neighboring ectoderm. In the stages immediately following the olfactory plates appear to sink down towards the ventral surface of the head, due no doubt to more rapid growth of the dorsal portion of the head. Thus they appear at the ventro-lateral angles of V. iWin^P Fig. 98. — Section of the otocyst of an embryo of 104 hours. The original opening of the otocyst is drawn out into a narrow ca- nal which connects with the endolymphatic duct (recessus labyrinthi). a., Ball of cells in the otocyst (otolith?), b., Canal leading from 170 THE DEVELOPMENT OF THE CHICK the anterior part of the head at the stage of 36 somites. During the displacement a depression appears in the center of each olfac- tory plate, and as this becomes deeper, the olfactory pits are formed (Figs. 99 and 117). At the stage of 36 somites each is a deep pit situated at the junction of the sides and ventral sur- face of the anterior portion of the head, with the wide mouth opening outwards and ventrally. The olfactory epithelium now becomes sharply differentiated from the ectoderm of the head, owing to the formation of a super- ficial layer of cells (teloderm, see p. 285) above the columnar cells in the ectoderm, but not in the region of the sensory epithelium, where the cells still form a single layer. In the center of the olfactory pit the epithelium is very much thickened owing to elongation of the cells, and the nuclei lie in five or six layers; there is a gradual thinning of the epithelium to the lips of the pit and then a sudden, but graduated, decrease to the general ectoderm. The line of junction of olfactory epithelium and indifferent ectoderm of the head is a little distance beyond the margin of the pit, as may be determined by the edge of the^telo- clermic layer; in other words, all of the olfactory epithelium is not yet invaginated. It is probable that the invagination of the olfactory plates is due mostly, up to this time, to the processes of growth within the plates themselves, although there has been considerable accumulation of mesenchyme in this region. But the subsequent deepening of the pits appears to be due largely to the formation of processes around the mouths of the primary pits. (See Chap. IX.) V. The Alimentary Tract and Its Appendages We have already learned that the main portion of the alimen- tary tract arises from the splanchnopleure; a portion of the mouth cavity is, however, lined with ectoderm and arises from an inde- pendent ectodermal pit, the stomodceum, which communicates only secondarily with the entodermal portion; similarly the last portion, external to the cloaca, arises from an ectodermal pit, the proctodceum, which communicates only secondarily with the entodermal part. We shall thus have to consider the origin of the stomodaium and the proctoda^um in connection with the alimentarv tract. FROM TWELVE TO THIRTY-SIX SOMITES 171 r%^ cery.fJ. .. 5MW7 Of 'J'f-VH/ -V'm V. MeZ-erfC. f^^^- u;£i;i^c^ ,2. -Mesenc. ,. ..- ex.o.c. -.DC.V/. f-;r 320^^^^ 3J0^y 7.B- vr.. €^^ % < V -y|.K ^1 ^- A .# '' ■ -■ %i-v^'--^--K dm. Umh. ■f ^V<^ \W#'/ J Fig. 99. — Entire embryo of 31 somites viewed as a transparent object, am. Umb., Amniotic umbilicus. B. a., Bulbus arteriosus, cerv. Fl., Cervical flexure, ch. Fis., Cho- roid fissure, cr. Fl., Cranial flexure. D. C, Duct of Cuvier. ex. o. c, External layer of the optic cup. int. o. c, Internal layer of the optic cup (retina.) N'm., Neuromere of myelencephalon. olf., Olfactory pit pc. W., Line of attachment of amnion to peri- cardial wall. V. C. 1, 2, 3, First, second, and third visceral clefts. Other abbreviations as before. 172 THE DEVELOPMENT OF THE CHICK From the embryological point of view the alimentary tract may be divided into fore-, mid-, and hind-gut. The fore-gut includes the anterior portion as far back as the liver diverticulum, the mid-gut extends from here to the coecal appendages, and the hind-gut includes the remainder. From each division there arise certain outgrowths which may be termed collectively appendages of the alimentary tract, and these will also be considered here, so far as they arise within the period covered by this chapter. Thus from the fore-gut there arise the visceral pouches, the thyroid and thymus glands, the postbranchial bodies, the respiratory tract, and the liver and pancreas; from the mid-gut the yolk-sac, and from the hind-gut the ccecal appendages and allantois. The enlargement of the body-cavity towards the middle line gradually reduces the broad mesodermal septum situated between its inner angles to a relatively narrow plate, which forms the dor- sal mesentery of the intestine (Figs. 107, 109, 110, and 111). This elongates in the course of development and forms a sheet of tissue suspending the intestinal tube to the mid-dorsal line of the body- cavity. It is composed of two layers of mesothelium (peritoneum) continuous with the lining of the body-cavity and enclosing a certain amount of mesenchyme; the dorsal mesentery extends along the entire length of the intestinal canal. A ventral mesentery uniting gut and yolk-sac is also estab- lished by the meeting of the limiting sulci in the splanchnopleure. When the body-wall closes, the ventral mesentery consists of two layers of mesothelium attaching the intestinal canal to the mid-ventral line of the body-wall. The dorsal and ventral mesen- teries, together with the alimentary canal, thus constitute a complete partition between the right and left halves of the body- cavity. However, the ventral mesentery is a very transient structure except in the region of the fore-gut and liver, and in the extreme end of the hind-gut. In these places it is persistent and is the seat of formation of important organs. The wall of the intestine contains three embryonic layers: viz., entoderm, mesenchyme, and mesothelium. The first forms the lining epithelium of the intestine, and of all glandular attach- ments, as well as of the respiratory tract and allantois; the last forms the serosa; and the mesenchyme the intermediate layers. We shall now consider the development of each region of the FROM TWELVE TO THIRTY-SIX SOMITES 173 alimentary tract and the appendages proper to each in the follow- ing order: (1) Stomodaeum, (2) Pharynx, (3) (Esophagus, (4) Stomach, (5) Hepato-pancreatic division of the fore-gut, (6) Mid- gut, (7) Hind-gut. The stomodaeum owes its origin to an expansion of the em- bryonic parts surrounding the oral plate, and it gives rise to a large part of the buccal cavity, which is therefore lined by ecto- derm. (See Chap. X.) It will be remembered that at the 12 s stage the oral plate lies between the pericardium and the head- fold (Fig. 67), and that it consists of a fusion between the ectoderm of the ventral surface of the head and the entoderm composing the floor of the anterior end of the fore-gut. It lies in a slight depression on the under surface of the head which is the beginning of the oral cavity. This small beginning owes its enlargement (1) to the cranial flexure, by which the ventral surface of the head becomes bent at right angles to the oral plate instead of forming a direct continuation of it, and (2) to the formation and protrusion of the mandibular arches and maxillary processes at the sides and behind. (See fuller account in Chap. VII.) In this way it becomes a deep cavity closed internally by the oral plate. The series of figures of sagittal sections through the head illustrates very well the gradual deep- ening of the stomodaeum by these processes (Figs. 75, 85, 87, 88). The oral plate thins gradually from the 12 to the 30 s stage when it breaks through (Figs 87 and 88), thus establishing an opening into the alimentary tract. The remnants of the oral membrane then gradually disappear and leave no trace. The subsequent extension of the maxillary region to form the upper jaw greatly enlarges the extent of the ectodermal portion of the buccal cavity. It will have been noted (Figs. 85 and 87) that the hypophysis opens in front of the oral plate on the ectodermal side, and this constitutes a most important landmark for deter- mining the limit of the ectodermal portion of the buccal cavity in later stages. The Pharynx and Visceral Arches. The pharynx may be briefly defined as the alimentary canal of the head. It is the most variable part of the alimentary canal in the series of vertebrates. Modified, as it is in all vertebrates, for purposes of respiration, the transition from the aquatic to the terrestrial mode of respira- tion brought about great changes in it. It is thus marked em- 174 THE DEVELOPMENT OF THE CHICK bryologically first by the development of structures, the viscera! arches and clefts, whose primary function was aquatic respira- tion, and second by the development of the air-breathing lungs. Such fundamental changes in function have left a deep impression, not only on the embryonic history of the pharynx itself, but also on the development of the nervous and vascular systems. The extreme anterior end of the pharynx extends at first some distance in front of the oral plate, and may hence be called the pre-oral gut (Figs. 75, 85, etc.). After the rupture of the oral plate, the pre-oral gut appears like an evagination of the pharynx immediately behind the hypophysis and is noAv known as Seessel's pocket (Fig. 87), but it gradually flattens out and disappears (Fig. 88). The form of the pharynx at thirty-three hours has been already described; briefly, it is much expanded laterally, exhibiting a crescentic form in cross-section (Fig. 54 A). The horns of the crescent are in contact with the ectoderm in front of the auditory pit, marking the site of the future hyoynandihular cleft, which arises by perforation in the fused area at about forty-six hours. A second pair of lateral expansions brings about a second fusion of the lateral wings of the pharynx just behind the auditory pit at about the stage of 19-20 somites. This is followed by the formation of a third and a fourth pair of lateral evaginations of the pharynx which reach the ectoderm at about 23 s and 35 s respectively. The walls of the pharynx appear considerably constricted between the evaginations which are known as vis- ceral 'pouches (Figs. 100 and 101). Corresponding to each visceral pouch there is formed an ectodermal invagination of much lesser extent, which may be known as the visceral furrow. The furrows do not form directly opposite the pouches, but slightly behind them so as to overlap the margins of the latter (Fig. 101). The ectoderm of the visceral furrows forms a close union with the entoderm of the pouches, and openings arise within these areas, excepting the fourth, forming transitory visceral clefts. There are thus four pairs of visceral pouches and furrows, known as the first, second, third, and fourth; the first is some- times called the hyomandibular. According to Kastschenko, there are evidences of three pairs of FROM TWELVE TO THIRTY-SIX SOMITES 175 visceral furrows in front of the first at the 14-16 s stage. These he in- terprets as phyletic rudiments. It is certain that the lower vertebrates had pouches posterior to the fourth. The post-branchial bodies (see p. 309) are probably rudiments of a fifth pair of pouches. The tissue between the visceral pouches thickens, by accumu- lation of mesenchyme, to form the visceral arches, of which there are five, viz.: (1) the mandibular in front of the first pouch, form- ing also the posterior boundary of the oral cavity, (2) the hyoid between the first and second pouches, (3) the third visceral arch between the second and third pouches, (4) the fourth visceral arch between the third and fourth pouches, and (5) the fifth vPl yC.(]J vPPvC.a^ ^, Fig. 100. — Reconstruction of the fore-gut of a chick of 72 hours. (After Kastschenko.) Hyp., Hypophysis, lar-tr. Gr., Laryngotracheal groove. Lg., Lung. Md. a., Mandibular arch. Oes., Oesophagus, pr'o. G., Preoral gut. Stom., Stomach. Th., Thyroid, v. C. d, 1, 2, Dorsal division of the first and second visceral clefts, v. C. v. 2, Ventral division of the second visceral cleft, v. P. 1, 2, 3, 4, First, second, third, and fourth visceral pouches. visceral arch behind the fourth pouch. Each arch is bounded internally by entoderm, externally by ectoderm. The main portion of its substance is formed of mesenchyme; each contains also a branch of the ventral aorta (aortic arch) and a branch of a cranial nerve. Understanding of their relations is therefore essential to knowledge of the develoj^ment of the nervous system, vascular system, and skull. We shall now^ consider the history of each visceral pouch and arch separately: The first visceral pouch becomes adherent to the ectoderm of the first visceral furrow at its dorsal and ventral ends, leaving 176 THE DEVELOPMENT OF THE CHICK an intermediate free portion. At about the 26 s stage an opening (cleft) forms at the dorsal adhesion, but none at the ventral; thus the first visceral cleft is confined to the dorsalmost portion of the pouch (Fig. 100). This opening closes about the end of the fourth day; the ventral portion of the pouch then flattens out, and the dorsal portion expands upwards towards the otocyst (Fig. 102). The first visceral (mandibular) arch thickens greatly between the 14 and 35 s stages, the ventral ends project a little behind the oral invagination, and subsequently meet to form the primor- dium of the lower jaw (Figs. 125 and 126, Chap. VII). A pro- ■ iv" -p I . Cj^r j4/r, Jff J yc/ vfi2. ^pj /isr-f/:^/: ^z VS. da.2. i/jr£ u/:3. Fig. 101. — Frontal section through the pharynx of a 35 s embryo. a. a. 1, 2, 3, 4, First, second, third, and fourth aortic arches. Hyp., Hypo- physis. J., Jugular vein, lar-tr. Gr., Laryngotracheal groove, or., Oral cavity. Ph., Pharynx. Th., Thyroid, v. A. 1, 2, 3, First, second, and third visceral arches, v. C. 1, First visceral cleft, v. F. 2, 3, Second and third visceral furrows, v. P. 2, 3, 4, Second, third, and fourth visceral pouches. Ill, Third cranial nerve. jection of the upper anterior border just behind the eye is the beginning of the maxillary process, or primordium of the maxillary portion of the upper jaw. The second visceral pouch likewise becomes adherent to the ectoderm of the second visceral furrow at its dorsal and ventral ends, and openings are formed in each adhesion b}^ the 35 s stage (Fig. 100); the dorsal opening is small and ow^l (later becoming more elongated) while the ventral one is a long, narrow fissure; they are separated only by a narrow bridge of tissue, and close during the fourth day. The third visceral pouch behaves like the second, forming a FROM TWELVE TO THIRTY-SIX SOMITES 177 small round dorsal, and long fissure-like ventral cleft at about the 40 s stage (Fig. 102). These close during the fifth day. The significance of the separate dorsal and ventral divisions of the visceral clefts is an interesting question. It is probable that the dorsal division had a special function, as they have a special connection with the branchial sense organs. Fig. 102. — Reconstruction of the pharyngeal organs of the chick at the end of the fourth day of incubation. (After Kastsch- enko.) a. a. 3, a. a. 4, a. a. 6, Third, fourth, and sixth aortic arches. Car. e.. External carotid. Car. i., Internal carotid. G. Gn., (ge- niculate ganglion. G. n. X., Ganglion nodosum. G. pr., Gan- gHon petrosum. ot., Otocyst. p. A., pulmonary artery. Ih. Thyroid, v. P. 1, 2, 3, 4, First, second, third, and fourth visceral pouches. _ . , , 1- V, VH, VIII, IX, X, XII, Cranial nerves and ganglia. The fourth visceral pouch connects with the ectoderm at its dorsal end, about the 35 s stage, but no cleft develops. Its pos- terior wall develops an evagination (postbranchial body) which by some is considered to be a rudimentary fifth pouch, and w^hich contributes to the formation of the thymus. (See Chap. X.) 178 THE DEVELOPMENT OF THE CHICK The second visceral arch is the largest of the arches and over- laps both the first and third. See Figs. 117 and 125 in place of description. All of the arches are wedge-shaped, corresponding to the wedge-like form of the hind-brain region. The fourth arch is small and incomplete ventrally; the fifth a mere transitory rudiment. The greatest development of the arches is at about the end of the fourth day. According to Kastschenko the closure of the visceral clefts takes place external to the meeting-place of the visceral furrows and clefts, and in this way some of the ectoderm of the furrows remains attached to the visceral pouches. The thyroid arises as a small, spherical evagination of the epithelium of the floor of the pharynx situated between, and a little in front of, the ventral ends of the second pair of visceral pouches (Figs. 85, 87, 88, 101). In the 18-20 s stage, it is repre- sented by a sharply defined plate of high, columnar cells in the same situation, which may be recognized even at the stage of 12 s. At the stage of 26 s this plate forms a deep, saucer-shaped depression, and at the 30 s stage it is a well-developed sac with wide-open mouth which gradually closes, thus transforming the sac into a small spherical vesicle lying beneath the floor of the pharynx (Fig. 102). The Pulmonary Tract. The portion of the pharynx that includes the visceral pouches may be called the branchial portion, because it is homologous to the gill-bearing portion in fishes and amphibia, and because the visceral pouches are phylogenetic rudiments of branchial clefts. The larnyx, trachea, and lungs develop from the ventral division of the postbranchial portion of the pharynx. At about the 23 s stage a reconstruction shows this respiratory division of the pharynx slightly constricted from the broader branchial portion, enlarged on each side at its pos- terior end and w^ith a ventral depression; the latter rapidly deepens to form a narrow groove, the primordium of the larynx and trachea, while the posterior lateral expansion begins to form outgrowths, the primordia of the lungs and air-sacs. By the stage of 35 s (Fig. 100) the postbranchial portion of the pharynx has become narrow transversely and its ventral half is a deep groove (laryngotracheal groove) leading back to the lung pri- mordia. A true median sagittal section at this time shows the FROM TWELVE TO THIRTY-SIX SOMITES 179 floor of the laiyngotracheal groove directl}^ continuous with the floor of the branchial portion of the pharynx at its hind end; the former bends up at about right angles to enter the narrow oesophagus (Figs. 87 and 88). Thus the whole pulmonary tract communicates widely with the pharynx at the 35 s stage. Its complete delimination falls within the period covered by Chapter X. The continuity of the expansions that form the lung primordia, w^ith the series of visceral pouches as shown in Fig. 100, is especially noteworthy as suggesting a theory of the phylogenetic derivation of the lungs. Fig. 103. — Reconstructions of the liver diverticula of the chick. (After Hammar.) A. On the third day of incubation; from the left side; the divertic- ula arise from the anterior intestinal portal. B. Beginning of the fourth day; from the left side. a. i. p., Anterior intestinal portal. D. V., Indicates position of ductus venosus. g. b., Gall bladder. 1. d. d. (cr.)., Dorsal or cra- nial liver diverticulum. 1. d. v. (caud.), Ventral or caudal liver diverticulum, pc. d., Dorsal pancreas. X., Marks the depression in the floor of the duodenum from which the common bile duct is formed. CEsophagus and Stomach. Immediately behind the pharynx, at the stage of 36 s, the intestine narrows suddenly (primordium of oesophagus) and enters a small, spindle-shaped enlargement, the primordium of the stomach (Figs. 87, 88, 100). The liver arises in the chick as two diverticula of the entoderm of the anterior intestinal portal, one situated immediately above and the other below the posterior end of the ductus venosus, or fork of the omphalomesenteric veins (Fig. 103 A). This portion 180 THE DEVELOPMENT OF THE CHICK of the anterior intestinal portal becomes incorporated in the floor of the intestine as the anterior intestinal portal retreats backwards, and the original dorsal liver diverticulum therefore becomes anterior or cephalic and the ventral becomes posterior or caudal (Fig. 103 B). Before this transposition occurs, how- ever, the diverticula have grown forward towards the sinus venosus in the ventral mesentery of the stomach, the anterior diverticulum above and the posterior diverticulum below the ductus venosus. The stretch of entoderm between the two liver diverticula thus lies in the angle made by the union of the two omphalomesenteric veins. At the stage of 26 somites, the anterior diverticulum has grown forward above the ductus venosus to the level of the Cuvierian veins and is large and flattened laterally. The posterior diverticulum is barely indicated at this time. The anterior diverticulum was originally described as left and the posterior as right (Goette, 1867), and this description was taken up by Foster and Balfour. This was corrected by Felix (1892). Subse- quent writers do not agree exactly as to the time or precise relations of the diverticula; however, it is generally agreed that the two diver- ticula are subdivisions of a common hepatic furrow, inasmuch as the entoderm between them lies below the level of the entoderm in front and behind (Fig. 103 B). Brouha maintains that at first the hepatic furrow lies in front of the anterior intestinal portal, and that the latter secondarily moves forward so as to include the hepatic furrow, which later again comes into the floor of the intestine with the definitive retreat of the anterior intestinal portal. This view does not rest on very secure evidence, and is probably based on interpretation of slight individual variations as successive stages of development. Choronschitzky places the time of appearance of the hepatic diverticula at about the thirty-sixth hour. It is probable, however, that this is too early. I have found the first unmistakable diverticulum at a stage of 22 somites, a slight rudi- ment of the anterior diverticulum in the anterior intestinal portal. At the 30 s stage the anterior or dorsal diverticulum has ex- panded much more, mainly to the left of the middle line, as though to embrace the ductus venosus, and the posterior or ventral diverticulum has an even greater development and embraces the right side of the ductus venosus, but it does not extend as far forward as the anterior diverticulum. Both diverticula now branch rapidly and profusely, forming secondary anasto- FROM TWELVE TO THIRTY-SIX SOMITES 181 moses where branches meet, so that a complete ring of anas- tomosing columns of hepatic cylinders is rapidly formed around the center of the ductus venosus (Figs. 103 B and 104, cf. also Figs. 119 and 120). But the anterior and posterior ends of the ductus venosus are not yet completely surrounded by the basket-work of liver substance, owing to the ab- sence of any part of the posterior diverticulum in its anterior por- tion, and of the anterior divertic- ulum in its posterior portion. The floor of the intestine be- tween the anterior and posterior liver diverticula is depressed; later it becomes separated from the intestinal cavity to form a tem- porary common bile-duct; which then receives the two primary di- verticula (Figs. 103 B, 104 and 187). The pancreas arises from a dor- sal and a pair of ventral primorclia. The former is an outgrowth of the dorsal wall of the intestine immediately above the posterior liver diverticulum (Figs. 103 B and 104). At the 35 s stage it is a solid thickening of the dorsal wall of the intestine of consider- able extent; a little later the base of the thickening is hollowed out, and the free margin sends ofT solid buds into the dorsal mesentery just behind the stomach. The ventral primordia arise from the posterior liver diverticulum in a manner to be described later (Chap. X). Mid-gut. At the 35 s stage the mid-gut is still open to the yolk-sac. Its subsequent history is given in Chapter X. Fig. 104. — Reconstruction of the liver of the chick at the end of the fourth day of incubation. (After Hammar.) du., Duodenum. L., Substance of liver. Other ahbreviations as before. 182 THE DEVELOPMENT OF THE CHICK Anal Plate, Hind-gut, Post-anal Gut. and Allantois. At about the 14 s stage a thickening of the ectoderm in the middle line just behind the primitive streak extends towards the entoderm which is folded up so as to neariy meet it, thus cutting off the extra-embryonic mesoblast from the primitive streak. The ecto- derm and the entoderm then come into contact here, and form a firm union, the anal plate (Fig. 70), which is subsequently perforated to form the anus. At first, however, the anal plate lies entirely behind the embryo, and the post-anal portion of the embryo arises from the thickened remnant of the primitive streak (tail-bud) which grows backwards over the blastoderm beyond the anal plate. Even before this, however, the hind-gut begins to be formed by a fold of the splanchnopleure directed forwards beneath the tail-bud, and the hind end of the tube thus formed ends at the anal plate (Fig. 70). The entoderm in front of the anal tube is fused with the substance of the tail-bud, and as the latter grows backwards beyond the anal plate it carries with it a pocket of the hind-gut, and this forms the post-anal gut (Fig. 80). The formation of the tail brings the anal plate on to the ven- tral surface of the embryo at the junction of tail and trunk, and the post-anal gut then appears as a broad continuation of the hind-gut extending behind the anal plate, and ending in the tail at the hind end of the notochord (Fig. 80). The further elonga- tion of the tail draws out the post-anal gut into a narrow tube lying beneath the notochord in the substance of the tail; it then gradually disappears and leaves no trace. The formation of the hind-gut takes place prior to the for- mation of the embryonic body-cavity at this place. It thus happens that the splanchnic mesoderm, forming the floor of the hind-gut, is directly continuous with the somatic mesoderm. When the body-cavity does penetrate this region it is without direct lateral connections with the extra-embryonic body-cavity, so that the connection of the splanchnic and somatic mesoderm persists, forming the ventral mesentery of the hind-gut (Fig. 81). This is a thick mass of mesoblast binding the hind-gut to the somatopleure. The hind-gut is deep from the first, and its ven- tral division soon begins to extend into the ventral mesentery as a broad evagination, the allantois (see p. 143). FROM TWELVE TO THIRTY-SIX SOMITES 183 VI. History of the Mesoderm The history of the extra-embryonic mesoderm is considered sufficiently in the first part of this chapter. The history of the embryonic mesoderm will be considered under the following heads: (1) Somites, (2) Intermediate Cell-mass, (3) Vascular System, (4) Lateral Plate and Body-Cavity, (5) Mesoblast of the Head. 0tMj^jff§^3^: Fig. iUo. — Embryo of aboul 27 somites drawn in alcohol by re- flected light; upper side, x 10. Am., Amnion, ot., Otocyst. t. F. Am., Tail fold of amnion. (1) Somites. The rate of formation of the somites from the segmental plate and their number at different times is given in the normal table of embryos (p. 68), and may be seen in various 184 THE DEVELOPMENT OF THE CHICK figures of entire embryos. The formation of new somites con- tinues after the end of the period discussed in this chapter, up to about the sixth day. Each somite has a definite value in the developmental history. Fig. 106. — The same embryo from beneath, x 10. a. i. p., Anterior intestinal portal. A. V., Vitelline artery. Int., Intestinal groove. In an embryo of 42 somites (about ninety-six hours), the value of the somites as determined by their relations and subsequent history is as follows: 1 to 4. Cephalic ; entering into the composition of the occipital region of the skull. Prebrachial; i.e., entering into the region between the wing and the skull. Brachial. Between wing and leg. 5 to 16. 17 to 19. 20 to 25. FROM TWELVE TO THIRTY-SIX SOMITES 185 26 to 32. Leg somites. 33 to 35. Region of cloaca. 36 to 42. Caudal. More somites are formed later, the maximum number recorded being 52 (see Keibel and Abraham, Xormaltafeln). In an eight-day chick the number of somites is again about 42, including the four fused with the skull. Thus the ten somites formed last are again lost. This points towards a long-tailed ancestry for birds. Each somite is composed of an epithelial wall of high, columnar cells, enclosing a core of cells that nearly fills the cavity (Figs. 112, 113, etc.). From each somite there arise three parts of fundamental significance, viz., the sclerotome, the muscle plate, and the cutis plate (dermatome), the primordium of the axial Fig 107. — Transverse section through the twenty-ninth somite of a 29 s embryo, n. Cr., Neural crest. Neph., Nephrotome. W. D., Wolffian duct. Other abbreviations as before. skeleton, the voluntary muscles (excepting those of the head), and derma respectively. The manner of origin of these parts may be studied fully in an embryo of 25 to 30 somites, by com- paring the most posterior somites, in which the process is begin- ning, with somites of intermediate and anterior positions in the series, which show successively later stages. Figs. 107, 108, 109, and 110 represent transverse sections through the twenty-ninth, twenty-sixth, tw^entieth, and seven- teenth somites of a 29 s embryo. In the twenty-ninth somite 186 THE DEVELOPMENT OF THE CHICK (Fig. 107) the primitive relations of the parts are still preserved. In the twenty-sixth somite (Fig. 108) it will be seen that the cells of the core and of the ventral and median wall of the somite extending from the nephrotome to about the center of the neural tube are becoming mesenchymal; they spread out towards the notochord and the space between the latter and the dorsal aorta. These cells constitute the sclerotome. The muscle plate extends from the dorsal edge of the sclerotome to the dorso-median angle of the wall of the somite, and the dermatome from this point to the nephrotome. nSr Mj/. ..ft 'j^- ^^jst'-i=^-!^ l^cp ^^^.: ~\ ^A. S26'. Sc/er -,£i-iv5'^ ^■j3;^v Coe/. 'A/ep/i. /;■.//. ;«g9^_, Fig. 108. — Transverse section through the twenty-sixth somite of a 29 s embryo. (Same embryo as Fig. 107.) Derm., Dermatome. My., Myotome. Scler., Sclerotome. V. c. p., Pos- terior cardinal vein. Other abbreviations as before. Fig. 109 is a section through the twentieth somite of the same embryo. The sclerotome is entirely mesenchymal, and its cells are extending between the notochord and aorta, and along the sides of the neural tube. The muscle-plate has now bent over so that its inner surface is being applied against the dermatome, but there is still a considerable cavity (myocoele) between the two, at the lateral angle of the dermo-myotomic plate. The lateral edge of the dermatome is freed from the nephrotome, and turns in to a slight extent. Other details are readily understood from the figure. The growth of the free edge of the muscle-plate towards the free lateral edge of the dermatome continues as illustrated in FROM TWELVE TO THIRTY-SIX SOMITES 187 t ,^«>-9^ -1^ "Si a .^s'sd^!^^ n y : V 7 h5* • T-H ;-< C4-1 tt a ^' ' -O ^-H r/7 cS >i o > __ Lh -.^ ^ 1^ ^ -* rt v. Sh .-^ , -^^ ^ r^ 3 !/) »5 02 Oi C^ (/J c; (-1 2 -t-> (-H X ^ H -^ rH r' a. H* -tj 1^ Oi -f^ 8 ^ '^ if >> £; i«^ o ^ s 7J • TJ t:; cc i-4 Vm > ^ ^H H <^ -. PSH C_) 6 < 188 THE DEVELOPMENT OF THE CHICK Figs. 109 and 110, until complete union of the two takes place (Fig. Ill) and there is established a complete dermo-myotomic plate in each somite. This is usually known as the myotome, which therefore includes two layers; the external cutis-plate or dermatome, and the internal muscle-plate. With the eleva- tion of the axis of the body, the myotome gradually assumes a nearly vertical position. C/zor Jy^. J Fig. 110. — Transverse section through the seventeenth somite of a 29 s embryo. (Same embryo as Fig. 107.) am. Cav., Amniotic cavity. E. E. B. C, Extra-embryonic body-cavity. Gn., Ganghon. mes'n. V., Mesonephric vesicle. S.-Am., Sero-amniotic con- nection. Other abbreviations as before. Other details concerning the early history of the sclerotome are given in Chapter XIII, and it remains to add here only a short description of certain changes in the cells of the myotome (myo- blasts). In longitudinal sections the cells of the myotome are seen to become spindle-shaped soon after the folding towards the dermatome begins. The nuclei of the myoblasts are large and stain less deeply than those of adjoining tissues. They become elliptical in correspondence with the form of the cell- bodies. Each myoblast soon stretches from anterior to pos- terior faces of the somite, and this represents the first stage in the differentiation of the voluntary muscles. In later stages the myotomes send outgrowths into the limb- buds and ventral body-wall for the formation of the voluntary FRO:\r TWELVE TO THIRTY-SIX SOMITES 189 o f) Zj 1 K Si SI ;-i 0; " > « m xn *^ c3 , > 190 THE DEVELOPMENT OF THE CHICK muscles of these parts. The vohmtary muscles of the head, on the other hand (excepting the hypoglossus musculature), arise in front of the somites; the mesoblast from which they arise is, however, part of the original paraxial meroblast, in large part at least. It is important to note that the voluntary muscles are epithelial in origin. The involuntary, or smooth, muscle fibers, on the other hand, are mesenchymal in origin. The dermatome remains epithelial in all the somites well into the third day; the cells then begin to separate and form mesenchyme; this process begins at the anterior somites and proceeds backwards. The mesenchyme thus formed is the foundation of the derma. The Intermediate Cell-mass. This is the cord of cells uniting somite and lateral plate; it reaches its typical development only from the fifth to the thirty-third somites, in which it contributes to the development of the excretory system. Behind the cloaca, that is in the region of the tail, there is no lateral plate and no nephrotome. Origin of the Excretory System. The history of the excretory system in Amniota is of particular interest, because it shows a succession of three separate organs of excretion or kidneys, the first of which is a mere functionless rudiment, the second is the principal organ of excretion during embryonic life (at least in reptiles and birds), and the third finally becomes substituted for the second, which degenerates and is mostly absorbed; however, parts of the second remain and contribute to the formation of the organs of reproduction. The first, known as the head kidney or pronephros, is probably homologous to the permanent kidney of Amphioxus; the second or mesonephros, is the homologue, in part, of the permanent kidney of Anamnia, and the third or metanephros is the permanent kidney. The secreting parts of all arise from the intermediate cell-mass, though not in the same manner. The development of the metanephros does not begin until the fourth day; it is therefore not considered in this chapter. Pronephros and Wolffian Duct. The pronephros extends over only eleven or twelve somites, viz., from the fifth to the fifteenth or sixteenth inclusive; it consists originally of as many parts or tubules as the somites concerned. Each tubule arises as a thickening of the somatic layer of the intermediate cell- FROM TWELVE TO THIRTY-SIX SOMITES 191 mass, which grows out tow^ards the ectoderm in the form of a blind, soUd sprout. The distal end of each turns backwards and unites with the one behind so as to form a continuous cord of cells, which is thus united with the intermediate cell-mass in successive somites by the original outgrowths. This cord of cells is the beginning of the Wolffian duct. Behind the sixteenth somite, the latter grows freely backwards just above the inter- mediate cell-mass until it reaches the cloaca with which it unites about the 31 s stage. Fig. 112. — A. Transverse section through the twelfth somite of a 16 s em- bryo. B. Three sections behind A to show the nephrostome of the same pro- nephric tubule. V. c. p., Posterior cardinal vein. c. C, Central canal. Ms'ch., mesen- chyme, n. Cr., Neural crest. N'st. Nephrostome. n. T., Neural tube, pr'n. 1, 2, Distal and proximal divisions of the pronephric tubule. The primary pronephric tubules are originally attached to the nephrotome opposite the posterior portion of the somite, about half-way between the somite and the lateral plate (Figs. 112 and 113). The part of the nephrotome between the attach- ment of the primary tubule and the lateral plate is continuous with the primary tubule and forms a supplementary part of the complete pronephric tubule; the remainder of the nephrotome then becomes converted into mesenchyme and the connection with the somites is lost (Figs. 112 and 113). Thus each pro- nephric tubule forms a connection between the Wolffian duct 192 THE DEVELOPMENT OF THE CHICK and the angle of the body-cavity; it consists of two parts, viz., the primary tubule and the supplementary part. It never pos- sesses a continuous lumen, though there is often a cavity in the supplementary part, which opens into the body-cavity through the nephrostome (Fig. 112 B). The pronephros of the chick is a purely vestigial organ, of no apparent functional significance. Its development is accord- ingly highly variable, and it often happens that the right and left sides of the same embryo do not correspond. It is also of very short duration and is usually completely lost on the fourth day. The tubules in the fifth to the tenth somites, moreover. /Ms'cA fir'/?. /'/SJ 'A So'pJ. S/?lpI. Fig. 113. — Transverse section tlu-ough the fifteenth somite of the same embryo. pr'n. (14), (15), Pronephric tubules of the fourteenth and fifteenth somites, respectively. hardly pass the first stage when they appear as thickenings of the somatic layer of the somitic stalk; thus the Wolffian duct does not extend into this region, and the best developed pronephric tubules are confined to the tenth to the fifteenth somites. The pronephric tubules do not form Malpighian corpuscles; but glomeruli develop as cellular buds at the peritoneal orifices of the posterior tubules, projecting into the coelome near the mesentery. Curiously enough these do not form at the time of greatest development of the tubules, but subsequently to this when the tubules themselves are in process of degeneration. Moreover, they are extremely variable as to number, and degree of development. They appear to be best developed on the third and fourth days. They agree in many respects with the so-called external glomeruli of the pronephros of Anamnia, and should be FROM TWELVE TO THIRTY-SIX SOMITES 193 homologized with these. On the other hand, they appear at the same time as the first glomeruU of the mesonephros (q. v.) and possess, by way of the intermediate tubules, undeniable resem- blance to the latter. At the stage of 10 somites the pronephros is represented by a series of thickenings of the somatic layer of the intermediate eell-mass extend- ing from the fifth somite backward to the segmental plate. In an embryo of 13 somites the connection between the somite and nephrotome is lost, and the pronephric tubules from the ninth to the thirteenth somites have united to form the beginning of the Wolffian duct. In an embryo of 16 somites a single pronephric tubule was found at the level of the hind end of the fifth somite, and was very distinct on one side but hardly discernible on the other. Its posterior continua- tion was soon lost, and the next distinct tubules were between the ninth and tenth somites ; from here back there was a tubule opposite the hind end of each somite to the fifteenth, which was the last, and the duct was continuous. In an embryo of 21 somites, one finds only isolated remnants of the pronephros in front of the eleventh somite; from here to the fifteenth the tubules are well developed and retain their connection both with the Wolffian duct and the lateral plate. The Wolffian duct extends back of this place to the region of the posterior half of the segmental plate. At the 35 s stage the pronephric tubules are much degenerated, but the nephrostomes usually remain. In one embryo there was found a well-developed pronephric tubule on each side in the thirteenth somite. That of the left side had a wide nephrostome, the lumen of which stopped short of the tubul ; the nephrostome of the right side was rudimentary. On the right side the Wolffian duct extended no farther forward, but on the left side it was continued to the eleventh somite, and rudimentary pronephric strands uniting it to the coelomic epithelium existed in both eleventh and twelfth somites. Here the W'olflaan duct stopped. But isolated pronephric rudiments and minute nephrostomes were found on both sides as far forward as the tenth somite. The Wolffian Duct. The Wolffian duct consists according to the foregoing account of two parts, (1) an anterior division formed by the union of the pronephric tubules, and (2) a posterior divi- sion that arises as an outgrowth of the anterior part. The latter grows backward above the intermediate cell-mass as a solid cord (Fig. 107), apparently by active multiplication of its own cells, without participation of the neighboring mesoderm or 194 THE DEVELOPMENT OF THE CHICK ectoderm, until it reaches the level of the cloaca at about the sixtieth hour (30-31 s). It acquires a narrow lumen anteriorly at about the 25 s stage; but the remainder is solid. At about the sixtieth hour the ends of the ducts fuse with broad lateral diverticula of the cloaca, and the lumen extends backwards until the duct becomes viable all the way into the cloaca (at about seventy-two hours, 35 s stage). The Mesonephros or Wolffian Body. The mesonephros de- velops from the substance of the intermediate cell-mass between the thirteenth or fourteenth somites and the thirtieth somite. There are slight local differences in the relations of the tubules in front and those behind the nineteenth and twentieth somites, but in general the tubules may be stated to arise as epithelial vesicles derived from the intermediate cell-mass, which become transformed into tubules, one end of which unites with the Wolffian duct and the other forms a Malpighian corpuscle in the manner described below. It will be seen that the anterior mesonephric tubules which are relatively rudimentary and of brief duration overlap the posterior pronephric tubules; they may possess neph- rostomes, whereas the typical mesonephric tubules formed behind them, which constitute the main bulk of the mesonephros, never possess peritoneal connections. An embryo with 29-30 somites is in a good stage for consid- ering the early development of the mesonephric tubules. If one examines a section a short distance behind the last somite, one finds that the intermediate cell-mass is a narrow neck of cells uniting the segmental plate and the lateral plate, and that the cells composing it are arranged more or less definitely in a dorsal and ventral layer, though some occur between. The primordium of the Wolffian duct occurs in the angle between the somatic mesoblast and the intermediate cell-mass, and the aorta lies in the corresponding angle of the splanchnic mesoblast. In the last somite (Fig. 107) one finds two important changes: (1) the intermediate cell-mass is much broader owing to multi- plication of its cells, and as a consequence the two-layered arrange- ment is lost; (2) whereas the cells of the intermediate cell-mass in the region of the segmental plate could not be delimited accu- rately from either the segmental or lateral plate, it is now easy in most sections to mark its boundary on both sides. It now constitutes, therefore^ a rather well-defimed but unorganized mass FROM TWELVE TO THIRTY-SIX SOMITES 195 of cells between the somite and lateral plate, aorta and Wolffian duct; the posterior cardinal vein appears above the Wolffian duct. The next change, found to begin in about the twenty-sixth somite, is a condensation of a portion of the cell-mass lying median to and below the Wolffian duct (Fig. 108), rendered evi- dent by the deeper stain in this region; the condensed portion of the original intermediate cell-mass is not, however, sharply separated from the remainder, but shades gradually into it both dorsally and ventrally, so that it can be seen to represent approximately the central part of the original middle plate. In view of its prospective function it may be called the nephrogenous tissue. Following it yet farther forward one finds that it is a continuous cord of cells with alternating denser and less dense portions, until in the twentieth somite (Fig. 109), the denser portions become discrete balls of radially arranged cells. In the eighteenth and seventeenth somites (Fig. 110) these become small thick-walled vesicles, which are situated median and ventral to the duct. Each vesicle is the primordium of a complete mesonephric tubule. Farther developed tubules are found in the fifteenth and sixteenth somites, and it is probable that the nephrogenous tissue forms mesonephric tubules in the four- teenth, thirteenth, and perhaps yet more anterior segments. The formation of the tubules proper from the vesicles may be studied satisfactorily in a 35 s embryo (seventy-two hours). In the twenty-third somite of such an embryo the nephrogenous tissue and the nascent tubules lie lateral to the Wolffian duct and below the median margin of the cardinal vein (Fig. 111). The Wolffian duct is triangular in cross-section with its longest and thinnest side next the coelome. The most advanced vesicle in this region possesses a hollow sprout extending laterally to the Wolffian duct with which it is in close contact; this is the pri- mordium of the tubular part of the mesonephric tubule (cf. Fig. 114 A and B). In more anterior somites it is found that such sprouts have fused with the wall of the duct in such a manner that the lumen of the tubule now communicates with that of the duct. Simultaneously the median portion of the original vesicle has been transformed into a small Malpighian corpuscle in the following manner: it has first become flattened so that the lumen is reduced to a narrow slit; then this double-layered disc becomes concave with the shallow cavity directed posteriorly and dorsally; 196 THE DEVELOPMENT OF THE CHICK at the same time the convex wall becomes thin, and the concave thick. The entire tubule thus becomes S-shaped. Figs. 114 A, B, C, D illustrate the corresponding processes in the duck, which are similar in all essential respects to the chick. B C D Fig. 114. — From a transverse series through a duck embryo of 45 s, to show the formation of the mesonephric tubules. (After Schreiner.) Fig. 218 shows the position of the sections A, B, and C. V. c. p., Posterior cardinal vein. W. D., Wolffian duct. A. and B. represent tubules of the twenty-ninth segment. C. of the twenty-seventh segment. D. of the twenty-fourth segment. In the chick embryo of 35 somites the only differentiated tubules are in front of the twentieth somite, a region of the mesonephros that never develops far, and such tubules do not appear ever to become functional. In the region of the subse- quent functional mesonephros (twentieth to thirtieth somites) the development has not progressed beyond the stage of the vesicles showing the first indications of budding. FROM TWELVE TO THIRTY-SIX SOMITES 197 The main part of the mesonephros is thus between the twen- tieth and thirtieth somites. In the anterior half of this region three or four rudiments of tubules are formed in each somite by the seventy-second hour. Subsequently five or six tubules are formed in each segment between the twentieth and thirtieth. Tubules are formed first from the ventral portions of the neph- rogenous tissue (see Fig. Ill); those formed later arise from the unused portions. There is no evidence that they ever arise in any other way. The tulniles may thus be divided according to the time of origin into primary, secondary, tertiary, etc., sets, but there is no morphological or functional distinction between the successive sets. (See Chap. XII.) The collection of tubules causes a projection or fold on each side of the mesentery into the body-cavity, known as the Wolffian body, the detailed history of which is given in Chapter XII. In conclusion it should be noted that the most anterior tubules of the Wolffian body possess peritoneal funnels like the pronephric tubules. Thus in an embryo of 30 somites I have noticed open perito- neal funnels in the eighth, ninth, twelfth, thirteenth, fourteenth, fifteenth, sixteenth, and seventeenth somites. It seems quite certain that the last of these belong to the mesonephros, though the most anterior are undoubtedly pronephric rudiments. In the eighteenth, nineteenth, twentieth, and twenty-first somites, small depressions of the peritoneum were noticed opposite tubules, but not communicating with them. The Vascular System. Soon after the thirty-third hour the heart begins to twitch at irregular intervals, and by the fort}^- fourth hour its beatings have become regular and continue unin- terruptedly. The contraction proceeds in the form of a rapid peristaltic wave from the posterior to the anterior end of the cardiac tube, and the blood, already present, is forced out in front. Through the aortic arches it reaches the dorsal aorta which distributes part to the body of the embryo, but most of the blood enters the vascular network of the yolk-sac. It is returned to the heart by various veins in the yolk-sac and em- bryo, and recommences the circuit. The development of the vascular system will be more readily understood if we preface the account with a brief description of the anatomy of the system early in the fourth day (Fig. 115, cf. also Figs. 135 and 136). The heart consists of four chambers, viz., the sinus venosus, 198 THE DEVELOPMENT OF THE CHICK the atrium, the ventricular loop, and the bulbus arteriosus (Fig. ^^^The truncus arteriosus lies in the floor of the pharynx and gives off the following vessels: (1) a short branch, the external carotid, extending into the mandibular arch; (2) complete arches in the second, third, and fourth visceral arches which jom the Fig 115 - The circulation in the embryo and yolk-sac between the eightieth and ninetieth hours of incubation, drawn from a photograph by A. H^ ^oie. The arteries are represented in solid black; the veins in neutral tint, fold of the yolk-sac covers the fore part of the head, a. a. 2, 3, 4, Second, third, and fourth aortic arches ^o forta^ Ar , Atrium B a Bulbus arteriosus. Car. ext., External carotid. ^V,^;- Internal carotk . D. C, Duct of Cuvier. D. V., D^^^^us ^^^^^'^l./-' ^'^^y W vein (anterior cardinal). 1. a. V., Left anterior vitelline ^^m Posterior vitelline vein. S. V., Sinus venosus. V. c. p., Posterior caruii veTn VenrVentricle. V. O. M. L., Left omphalomesenteric vein. FROM TWELVE TO THIRTY-SIX SOMITES 199 dorsal aorta; these are known as the second, third, and fourth aortic arches; the third arch is the largest. The original mandibular aortic arches unite with the anterior ends of the dorsal aortse, forming a loop on each side at the base of the fore- brain (Fig. 93), and they have, therefore, a different relation from the other aortic arches; it seems probable also that they have a different morphological value. The ventral limb of this loop disappears in its pre-oral part after this stage and a new vessel is formed entirely within the mandibular arch, bearing the same relation to the visceral arch as the other aortic arches. At the stage of 35 somites it is a complete arch, in some embryos at least (Fig. 117), though of very small caliber and very transitory, possibly sporadic, in its occurrence. It is possible that this is the true mandibular arch, and the pre-oral portion of the original mandibular arch should have another interpretation. Kastschenko suggests that it may have been related to lost pre-mandibular gill- clefts. The roots of the dorsal aorta above the pharynx receive the aortic arches and are continued forward as the internal carotid arteries, branching in the fore part of the head. Posteriorly the right and left aortic roots unite just behind the fourth visceral pouch to form the dorsal aorta, and this continues as an undi- vided vessel to about the level of the twenty-second somite, w^here it divides into right and left dorsal aortse, and at the same time sends out a large omphalomesenteric artery into the yolk-sac on each side, and these branch as shown in Figure 115 into the capillary network of the yolk-sac. The dorsal aortae, now much diminished in size, continue back into the tail where they are known as the caudal arteries. The dorsal aorta also sends off a pair of segmental arteries into each intersomitic septum, and a pair of small allantoic arteries into the primordium of the allantois. The veins enter the heart through three main trunks: (1) the ductus venosus, (2 and 3) the paired ducts of Cuvier. These are made up as follows: (1) the ductus venosus is formed at the level of the posterior liver diverticulum by the right and left omphalomesenteric veins, which arise in the yolk-sac by union of the capillaries of the vascular area; the right vitelline vein also receives two veins coming directly from the anterior and posterior ends respectively of the sinus terminalis, the anterior of these is frequently partly double owing to its mode of origin. (See beyond, Chap. VII.) The vascular area in the yolk-sac thus 200 THE DEVELOPMENT OF THE CHICK appears strikingly bilateral at this time. (2 and 3) The ducts of Cuvier are made up by the union of all the somatic veins. Each is formed primarily by the union of the anterior and posterior cardinal veins. The anterior cardinal vein receives all the blood of the head, and thus includes the first three segmental veins. It also receives at its point of junction with the posterior cardinal vein a branch from the floor of the pharynx, the external jugular vein. The posterior cardinal vein receives (1) all the segmental veins of the trunk, of which there are twenty-nine pairs, running in the intersomitic septa between the fourth and thirty-third somites, and the veins of the Wolffian body of which there are several to each somite concerned, as described in the account of that organ. The development of the vascular system up to the stage just described will now be taken up. Development of the Heart, (a) Changes in the External Form. In the last chapter we traced the origin of the heart up to the time when it is a practically straight, undivided, somewhat spindle-shaped tube lying below the floor of the pharynx, to which it is attached by its dorsal mesentery (mesocardium). Posteriorly its cavity divides into the omphalomesenteric veins which run in the side-walls of the anterior intestinal portal. The heart is lengthened backwards by the concrescence of the omphalo- mesenteric veins and the most posterior division of the heart (the sinus venosus) is established in this way between the stages of 12 and 18 somites; it is marked by a broad fusion with the somatopleure (mesocardia lateralia) through which the ducts of Cuvier enter the heart. At the stage of sixteen somites the duct of Cuvier lies opposite the hind end of the second somite on the right side, and a little farther back on the left side; and the somato-cardiac fusion (mesocardium laterale) in which it lies is of the width of about one and a half somites. On the right side the duct of Cuvier lies a little in front of, and on the left side a little behind, the point of union of the omphalomesenteric veins; thus the posterior end of the heart is not fully formed at the stage of 16 s, but is at the stage of 18 s. The subsequent fusion of the omphalomesenteric veins produces the so-called ductus venosus, or main splanchnic vein, which is therefore a posterior continuation of the sinus venosus. The cardiac tube proper lies between the origin of the aortic FROM TWELVE TO THIRTY-SIX SOMITES 201 arches at the anterior end and a point Uttle behind the entrance of the ducts of Cuvier into the heart at the posterior end. Two main changes characterize the development of the heart in the period under consideration: (1) folding of the cardiac tube and (2) differentiation of its walls in successive regions to form the four primary chambers of the heart, viz. (from behind for- wards), the sinus venosus, the auricular division (atrium), the ventricular division and the hulhus arteriosus. The folding of the heart is caused by the rapid growth between its anterior and posterior fixed ends, and the places of folding are determined largely by differences in the structure of the walls at various places. The folding begins by a curvature to the right, and this proceeds until the tube has an approximately semicircular curvature (Fig. 72). At a certain place in the curved tube a very pronounced posterior projection takes place (Figs. 73 and 74), and at the same time this bent portion turns ventrally; the apex of the bend represents the future apex of the ventricles. The continuation of these two directions of folding then brings the ventricular division of the heart immediately beneath the sinu-auricular division which is attached dorsally by the somato-cardiac connections; further continuation brings the apex of the heart a little behind the auricular portion (Figs. 85, 87, 88, 93, 99). During all this period the distance between the two fixed ends has remained practically constant. During the process of folding, constrictions have arisen between successive portions of the cardiac tube, owing to expan- sion of intervening portions, and thus at the stage of seventy-two hours the heart shows the following divisions and form. From the dorsal surface (in a dissection. Fig. 116) one sees (1) the sinus venosus, broad behind and narrow in front where it joins the auricular division; it receives three veins: (a) the large ductus venosus, appearing as a direct posterior continuation of the sinus, and separated from it by only a slight constriction; and (b and c) the right and left ducts of Cuvier entering the sinus laterally and dorsally near its enlarged posterior end; (2) the sinus enters the atrium through the dorsal wall; the atrium shows two lateral expansions, the future auricles, of which the left is much the more expanded at this time; the sinus appears partly sunk in the right auricle. (3) Only the right limb of the ventricular loop is visible from the dorsal surface at this time, and is separated 202 THE DEVELOPMENT OF THE CHICK from (4) the bulbus arteriosus by a slight constriction. The bulbus thus Ues on the right side; it sweeps around the atrium anteriorly to the middle line and then bends up to enter the floor of the pharynx. From the ventral side one sees the looped ventricular division behind, in which we distinguish right and left limbs, the former of which enters the bulbus in front, and the latter the auricles. These two limbs represent ap- proximately the future right and left ventricles (Fig. 198, Chap. XII). In an ordinar}^ entire mount of this stage the heart is seen from the right side, and the dis- position of the parts may be readily understood by reference to Fig. 117, and the preceding description. Another change that should be noted here is the disappear- ance of the mesocardium during the folding of the cardiac tube, except in the region of the sinus venosus w^here it remains permanently and becomes much broadened (seventy-two hours). (6) Changes in the Internal Structure o] the Heart. We have already seen that the heart consists of two primary layers, viz., the endocardium, which is endothelial in nature, and the myo- cardium, which is derived from the splanchnic mesoblast. The distinction between the sinu-auricular and the bulbo-ventricular divisions of the heart is indicated internally at about the time the first external evidence is seen, by the fact that the endocar- dium is more closely applied to the myocardium in the former than in the latter division. In the sinus and atrium but little change takes place in the period under consideration. In the ventricle, on the other hand, and especially in the right Umb, the wide space originally existing between endocardium and Fig. 116 . — Heart of a chick embryo of 72 hours, dissected out and drawn from the dorsal surface. Aur. 1., Left auricle. Aur. r., Right auricle. B. a., Bulbus arteriosus. D. C. r. 1., Right and left ducts of Cuvier. D. V., Ductus venosus. S. V., Sinus venosus. Tr. a., Truncus arte- riosus. V. r., Right limb of ventricle. FRO:^r TWELVE TO THIRTY-SIX SOMITES 203 myocardium becomes more or less filled by multiplication of the endocardial cells. On the side of the myocardium there is first a thickening, and then anastomosing processes are sent out towards the endocardium. Cavities also arise within the thick- ened myocardium and all communicate. The endocardial cells then form a covering to all myocardial processes and cavities, and the cavities thus lined communicate with the main endo- cardial cavity. Thus the wall of the ventricles becomes spongy and all the cavities in it are lined by a layer of endocardium and communicate with the endocardial cavity. In the bulbus finally there is a great thickening of the endocardium produced by multiplication of its cells, but no corresponding change in the myocardium; thus the bulbus at seventy-two hours shows a thin myocardial and a thick endocarcUal wall. The later development is described in Chapter XII. The Arterial System. The description of the development of the arterial system proceeds from the stage of 12 somites described in the last chapter. The following should be added to the account there given. At this stage Kastschenko finds three pairs of small arterial vessels in front of the first visceral pouch running from the dorsal towards the ventral aorta, which, however, they do not meet. At about forty-six hours the first two of these have disappeared. The third, however, has become almost as large as the hyoid aortic arch. Kastschenko thinks it prob- able that this is the true mandibular arch. Though he did not find it in connection with the ventral aorta, he thinks it may form such a union of short duration. I have actually found such a vessel joining the man- dibular arch to the dorsal aorta in an embryo of 35 somites. On the other hand, what we have previously called the mandibular arch may be the true one displaced in the course of phylogeny. The Aortic Arches. An arch of the aorta is formed in each vis- ceral arch; they arise successively as buds from the roots of the dor- sal aorta in the order and time of formation of the visceral arches. Thus the first or mandibular aortic arch is formed at the stage of 9-10 somites; the second or hyoid aortic arch arises from the dor- sal aorta at about the stage of 19 s and joins the ventral aorta at about the 24 s stage. The third is completely formed at the stage of 26 somites. The fourth is completely formed at the stage of 36 somites; and the fifth and sixth arise during the fourth and fifth days. (See Chap. XII for account of the fifth and sixth arches.) 204 THE DEVELOPMENT OF THE CHICK The first aortic arch loses its connection with the dorsal aorta at about the stage of 30 somites, and the second arch similarly during the fourth day; the ventral ends of these arches retain their connection with the ventral aorta and constitute the begin- ning of the external carotid. Thus the third, fourth, fifth and sixth aortic arches remain. Their transformation belongs to the subject-matter of Chapter XII. The pulmonary artery appears as a posterior prolongation of the ventral aorta on each side at about the 35 s stage. It thus appears successively in later stages as a branch from the base of the fourth and sixth aortic arches. The Internal Carotids. The loop where the mandibular arch joins the dorsal aorta may be called the carotid loop; it is situated in front of the oral plate at the base of the fore-brain on each side (Fig. 93). It enlarges to form a sac, and when the connec- tion with the mandibular arch is lost, sends out branches into the tissue surrounding the brain. These are of course a direct continuation of the dorsal aorta on each side. The segmental arteries are paired branches of the dorsal aorta in each intersomitic septum. They pass dorsally to about the center of the neural tube and arch over laterally to enter the segmental veins, and thus unite with the cardinal veins. The Development of the Venous System. The main outlines of the development of the venous system have been already considered. The somatic veins, i.e., the anterior and posterior cardinal veins and their branches, enter the sinus venosus through the ducts of Cuvier. The original position of this duct as we have seen is about the level of the second somite. The formation of the cervical flexure, however, carries a number of somites forward above the heart, so that at about the stage of 32 s it comes to lie in the region of the eighth and ninth somites. The relation between the somatopleure and the heart in this region has been already described. The anterior cardinal veins are the great blood-vessels of the head, and become the internal jugulars in the course of develop- ment. Owing to the order of development of the body, the anterior cardinals are formed before the posterior cardinals. At the 15-16 s stage: they lie at the base of the brain, dorsal and lateral to the dorsal aortse, and extend forward to the region of FROM TWELVE TO THIRTY-SIX SOMITES 205 the diencephalon. They he internal to the cranial nerves and pass just beneath the auditory pits. As the brain develoi:>s many branches of the anterior cardinal veins arise, the most conspicuous of which at seventy-two hours are a large branch just behind the auditory sac, one between the auditory sac and the trigeminal ganglion, an ophthalmic branch extending along the base of the brain to the region of the optic stalks and a network of vessels on the lateral surfaces of the fore-brain. The other branches of the anterior cardinal vein are the three anterior intersomitic veins (Fig. 115); the external jugular from the floor of the pharynx enters the duct of Cuvier just beyond the union of the anterior and posterior cardinal veins. Up to about forty-eight hours the anterior cardinal veins lie median to the cranial nerves, but between this time and seventy- two hours the facial and glossopharyngeal nerves cut completely through the vessel and thus come to lie median to it; the trigem- inus and vagus continue to lie lateral to it. The posterior cardinal arises as a posterior prolongation from the duct of Cuvier and grows backward above the Wolffian duct, keeping pace with the differentiation of the intermediate cell- mass, as far as the thirtv-third somite. It does not enter the caudal region of the body. As already described it receives twenty-nine intersomitic veins and the veins of the Wolffian body. At first its connection with the duct of Cuvier is by means of a network of vessels, which gradually gives place to a single trunk (cf. Fig. 117). The Splanchnic Veins. The ductus venosus is the unpaired vein immediately behind the sinus venosus, formed by fusion of the two omphalomesenteric veins. It is fully formed at the stage of 27 somites. Its relations to the liver have already been de- scribed in connection with that organ. Its subsequent changes are described in Chapter XII. The vitelline veins are united at about the stage of seventy- two hours by a loop passing over the intestine immediately behind the pancreas. (See Chap. XII.) YII. The Body-cavity and Mesenteries The origin of the dorsal and ventral mesenteries was con- sidered in the section of this chapter dealing with the ali- mentary canal. As noted there, the dorsal mesentery extends 206 THE develop:\iext of the chick ceri^/7. MM Oj?MJ//^ /s/A Fig. 117. — Entire embryo of 35 s, drawn as a transparent object. a. a. 1, 2, 3, 4, First, second, third, and fourth aortic arches. Ar., Artery. A. V., Vitelhne artery, cerv. Fl., Cervical flexure, cr. Fl., Cranial flexure. D. C, Duct of Cuvier. D. V., Ductus venosus. Ep., Epiphysis. Gn. V., Ganfjlion of trigeminus. Isth., Isthmus. Jug. ex., External jugular vein. Md., Mandibular arch. M. M., Maxillo-mandibular branch of the trigeminus, olf. P., Olfactory pit. Ophth., Ophthalmic branch of the trigeminus. Ot., otocyst. V., vein. W. B., Wing bud. V. c. p., Posterior cardinal vein. V. umb.. Umbilical vein. V. V., Vitelline vein. V. V. p.. Posterior vitel- line vein. FROM TWELVE TO THIRTY-SIX SOMITES 207 the entire length of the ahmentary canal, while the ventral mesentery persists only in the region of the fore-gut and the cloaca. The embryonic body-cavity shows two divisions from a yery early stage, viz., (1) the large cephalic or parietal cavity situated in the pharyngeal region of the head and containing the heart, and (2) the general coelomic cavity of the trunk. After the heart is established in the middle line the parietal cavity is bounded posteriorly by the ^vall of the anterior intestinal portal (Figs. 75, 85, etc.), but it communicates with the pleuro peri- toneal cavity around the sides of the portal, in which the vitelline veins run. Laterally the parietal cavity communicates with the extra-embryonic body-cavity. ' The mesocardia lateralia are also an important landmark in the embryonic body-cavity because from them proceed the par- titions that subsequently separate the pericardial and pleural cavities on the one hand, and the pleural and peritoneal body- cavities on the other. (See Chap. XI.) The primordium of the lateral mesocardia may be recognized in the 10 s stage: just behind the heart the median portion of the body-cavity is thick-walled, the peritoneal cells being actually columnar. At this place, a short distance lateral to the median angle of the body-cavity, and at the junction of the cylindrical and flat mesothelium, a fusion of considerable longitudinal extent is formed between the somatopleure and the proximal portion of the vitelline veins, projecting up from the splanchnopleure; this fusion is the begin- ning of the lateral mesocardiam. It separates a more median portion of the body-cavity from a more lateral, and in it the duct of Cuvier soon develops. When this portion of the body of the embryo becomes ele- vated (forty to fifty hours) the portion of the body-cavity lateral to the mesocardia lateralia comes to lie ventrallv to the median portion (cf. Fig. 69), and at the same time the lateral mesocardia rotate around a longitudinal axis through an angle of about 90°, so that the original median border becomes dorsal, and the original lateral border becomes ventral. The dorsal divisions, right and left, of the pleuroperitoneal cavity may now be called the pleural grooves. Inasmuch as the parietal cavity has receded considerably at the same time into the trunk with the elongation of the fore-gut, it comes to lie beneath the pleural grooves 208 THE DEVELOPMENT OF THE CHICK instead of in front of them as before. Therefore in cross-sections^ in front of the lateral mesocardia, the pleural grooves appear as dorsal projections of the parietal cavity, separated from one another in the middle line by the oesophagus (Fig. 118). The relations of the three divisions of the embryonic Ijody- cavity thus established may be described as follows: the parietal cavity contains the heart, and is therefore the prospective peri- I \ m ■<:-f:^£ ^.■^■t.cfors- juteni^ D-C-r Lg. Ajn. ?.■).:.■; //fi '■^- 5 9 ■■«.-; >--.^V; i-jifj p.c^ ^.j^ .Atr / / L^wS Fig. 118. — Transverse section of an embryo of 35 s, imme- diately in front of the lateral mesocardia. Ao., Aorta. Atr., Atrium. B. a., Bulbus arteriosus. D.C. r., and ]., Right and left ducts of Cuvier. Lg., Lung, m's'c. dors., Dorsal mesocardium. m's't. dors., Dorsal mesentery. P. C, Pericardial cavity, pi. gr., Pleural groove. Rec. pul. ent., Recessus pulmo-entericus. S. V., Sinus venosus. cardial cavity. It is not, however, a closed cavity, but communi- cates in front of the lateral mesocardia with the pleural grooves (Fig. 118), and by way of the latter above the lateral mesocardia with the peritoneal cavity (Figs. 119 and 120); a second communi- cation of the parietal cavity with the peritoneal cavity is beneath the lateral mesocardia around the sides of the anterior intestinal portal, now being converted into the septum transversum (cf. FROM TWELVE TO THIRTY-SIX SOMITES 209 Fig. 120). A more complete description of the cavities is given in Chapter XI. The median wall of the pleural grooves forms much mesobla.st during the formation of the lung diverticula, and thus initiates the formation of lobes enclosing the lungs (Figs. 118 and 119). These lobes descend ventrally and unite with the septum trans- versum (see below), thus producing blind bays of the coelome nX,-. ^<5Y. /?ec/' jris-tdor^- stom. '/v'a't.irQTl. / Fig. 119. — Transverse section of the same embryo through the lateral mesocardia. Liv., Liver, m's'c. lat., Lateral mesocardiiim. m's't. access., Accessory mesentery, m's't. ven., Ventral mesentery. Other abbreviations as before. at the sides of the oesophagus, known as the superior recesses of the peritoneal cavity or pulmo-enteric recesses. The ventral mesentery extends from the anterior end of the sinus venosus to the hind end of tlie fore-gut, where it unites with the ventral body-wall. It includes the sinus venosus and the ductus venosus, together with the hepatic diverticula. The median and lateral mesocardia, together with the ventral mesen- tery of the fore-gut, form a mass known as the septum transversum. 210 THE DEVELOPMENT OF THE CHICK At the stage of seventy-two hours, then, the pleural, pericar- dial and peritoneal divisions of the body-cavity are indicated, but all are in communication. The pleural cavities connect with the peritoneal cavity posteriorly, and with the pericardial cavity anteriorly in front of the lateral mesocardia (Figs. 118, 119, 120); and the pericardial cavity communicates with the Fig. 120. — Transverse section of the same embryo immediately behind the lateral mesocardia. ant. hep. Div., Anterior hepatic diverticulum. Duod., Duo- denum. End 'c, Endocardium. D. V., Ductus venosus. My'c, Myocardium. PI. m's'g., Plica mesogastrica. S-am., Sero-am- niotic connection, ven. r., 1., Right and left limbs of the ven- tricle. V. umb., Umbilical vein. peritoneal cavity beneath the lateral mesocardia around the roots of the vitelline veins (sides of the anterior intestinal portal). Thus the ducts of Cuvier and the vitelline veins are the agencies that introduce the separation of the body-cavities. The tail-fold forms blind ccelomic pockets in the region of the hind-gut, which end in the region of the thirty-third somite. (Cf. Fig. 81.) PART II THE FOURTH DAY TO HATCHING ORGANOGENY, DEVELOPMENT OF THE ORGANS CHAPTER Vn THE EXTERNAL FORM OF THE EMBRYO AND THE EMBRYONIC ME:\IBRANES I. The External Form General. The development of the external form of the em- bryo is conditioned by the order of development of the organs. The early form is thus given by the nervous system, somites A B Fig. 121. — A. Embryo of 3 days' and 16 hours' incubation, x 5. B. Embryo of 5 days' incubation, x 5. (After Keibel and Abra- ham.) and viscera. The development of muscles, bones, limbs, etc., that define the form of the fowl, begins relatively late, and only gradu- ally conceals the outlines of the internal parts. Figs. 121 to 124 illustrate the development of the external 211 212 THE DEVELOPMENT OF THE CHICK form from three days sixteen hours to ten days. In Fig. 121 A (three days sixteen hours) the form of the head is defined by the brain, eyes, and visceral arches. The cervical flexure is strongly marked. There is no neck. The heart makes a large protuberance immediately behind the head. The limb-buds are rounded swellings. In Fig. 121 B (five days one hour) the cer- vical flexure is less marked; the enlargement of the mid-brain Fig. 122. — Embryo of 7 days' and 7 hours' incubation x 5. (After Keibel and Abra- ham.) makes a more pronounced protuberance of the head in this region; the heart has retreated farther back into the thorax, and the neck is. thus indicated. The main divisions of the limbs are beginning to appear. In Fig. 122 (seven days seven hours) there are marked changes: The cervical flexure is practically lost. The elevation of the head and retreat of the heart into the thorax have produced a well-marked neck. The upper EMBRYO AND EMBRYONIC MEMBRANES 213 portion of the first visceral cleft alone is visible as the external auditory meatus; the other visceral arches and clefts have prac- tically disappeared, excepting the mandibular arch, forming the lower jaw. The abdominal viscera begin to protrude. In the next stage, Fig. 123 (eight days), the contours of the body are decidedly bird-like; feather germs have appeared in definite Fig. 123. — Embryo of 8 days x 5. (After Keibel and Abraham.) tracts; the fore-limbs are w^ing-like. The contours of the head are much smoother, and determined more by the development of the facial region and skull than by the brain. The protuber- ance of the ventral surface caused by the viscera is strongly marked. Fig. 124 finally shows a ten-day embryo. Head. The embryonic development of the head depends on the changes in three important classes of organs, together with 214 THE DEVELOPMENT OF THE CHICK their supporting and skeletal structures and accessory parts: (a) the central nervous system, (5) the organs of special sense, and (c) the visceral organs, mouth and pharynx. The origin of all these parts has been considered, and it is proposed to take X' . / '&yy \- •^. -^ li Fig. 124. — Embryo of 10 days and 2 hours x 5. (After Keibel and Abraham.) up here only the development of the external form of the head. The preceding section gives an account sufficient for our present purposes, exc'ept in the case of the facial region. At four days this region appears as follows (Fig. 125): the mouth is a large, ill-defined opening, bounded behind by the mandibular arches, EMBRYO AND EMBRYOXIC MEMBRANES 215 at the side by the maxillary processes, and in front by the naso- frontal process, which is a broad projection below the cerebral hemispheres overhanging the mouth. On each side of the naso- frontal process are the olfactory pits, the cavities of which are continuous with the oral cavity. Lateral to the olfactory pits are the external nasal processes, abutting against the eye and separated from the maxillary process b}^ the lachrymal groove. The portion of the naso-frontal process bounding the olfactory pits on the median sides may be called the internal nasal process. 5. p. nss. 'Hem. /' ~ / - 0//. ... ^ye. Mx.' Or. - pi-. __:^_^>^^— Md. ^ . Hy. '^€ vA.3 - ^ Fig. 125. — Head of an embryo of 4 flays' in- cubation, from the oral surface (N. L. 6 mm.) Ep., Epiphysis. Hem., Cerebral hemi- sphere. Hy., Hyoidarch. 1. nas. pr.. Lateral nasal process. Md., Mandibular arch. Mx., Maxillary process, nas.fr., Naso-frontal pro- cess. Olf., Olfactory pit. Or., Oral cavity. Ph., Pharynx, v. A. 3, Third visceral arch. During the fourth and fifth days a fusion is gradualh' formed between the internal nasal process on the one hand, and the external nasal and maxillary processes on the other (Fig. 126), thus forming a bridge across the open mouth of the olfactory pits and dividing the openings in two parts, one within the oral cavity, which becomes the internal nares or choanae, and one without, which becomes the external nares or nostrils. During the same time the whole naso-frontal process begins to project 216 THE DEVELOPMENT OF THE CHICK forward to form the tip of the upper jaw. The two mandibular arches have also fused in the middle line and begin to project forward to form the lower jaw\ This projection of upper and lower jaw causes a great increase in the depth of the oral cavity (Fig. 147). The upper jaw is thus composed of three independent parts: viz., the median part formed from the naso-frontal process and ex-72ar. l.ms.pn Fig. 126. — Head of an embryo of about 5 days from the oral surface. (N. L. 8 mm.) ch. F., Choroid fissure. E. L., Eye-lid (nic- titating membrane), ex. nar., External nares. 1. Gr., Lachrymal groove. Other abbreviations as before. the two lateral parts formed from the maxillary processes. The former becomes the intermaxillary and the latter the maxillary region. II. Embryonic Membranes General. The extension of the blastoderm over the surface of the yolk goes on very rapidly up to the end of the fourth day of incubation (Fig. 33), at which time there is left a small cir- cumscribed area of uncovered yolk, that may be called the umbilicus of the yolk-sac, which remains uncovered for a long time. Its final closure is associated with the formation of the albumen-sac. EMBRYO AND EMBRYONIC MEMBRANES 217 The splitting of the mesoblast is never complete; but on the contrary the undivided margin begins to thicken after the fourth day, and gradually forms a ring of connective tissue that surrounds the umbilicus of the yolk-sac (Figs. 128 and 129). When this ring closes, about the seventeenth day, it forms a mass of con- nective tissue uniting the yolk-sac and albumen-sac. (See below). During the first few days of incubation the albumen loses water rapidly, and becomes more viscid, settling, as this takes place, towards the yolk-sac umbilicus. Thus the amniotic sac containing the embryo lies above; beneath the amniotic sac comes the yolk, and the main mass of the albumen lies towards the caudal end of the embryo (Figs. 128 and 129). The allantois expands very rapidly in the extra-embryonic body-cavity, and the latter extends by splitting of the mesoblast into the neighborhood of the yolk-sac umbilicus. When the allantois in its expansion approaches the lower pole of the egg, it begins to wrap itself around the viscid mass of the albumen accumulated there. In so doing, it carries with it a fold of the chorion, as it must do in the nature of the case, and thus the albumen mass begins to be surrounded by folds of the allantois with an intervening layer of the duplicated chorion. These relations will be readily understood by an examination of the accompanying diagrams (Figs. 128 and 129). In this way an albumen-sac, which rapidly becomes closed, is established out- side of the yolk-sac, and the two are united by the undivided portion of the mesoblast around the yolk-sac umbilicus. This connection is never severed, and in consequence the remains of the albumen-sac is drawn with the yolk-sac into the body-cavity towards the end of incubation. The sero-amniotic connection, which persists throughout incu- bation, has an important effect on the general disposition of the embryonic membranes. It is formed, as we have seen, in the closure of the amnion, by the thickened ectoderm of the suture; this ectodermal connection is, however, absorbed and replaced on the fifth to the seventh days by a broad mesodermal fu- sion, which maintains a permanent connection between amnion and chorion. One important result of this relation is that the albumen-sac, which is formed by the duplication of the chorion, is prolonged by a tubular diverticulum to the sero-amniotic 218 THE DEVELOPMENT OF THE CHICK plate (see Figs. 128 and 129). The latter becomes perforated after the eleventh day, and there is thus direct communication between the albumen-sac and the amniotic cavity. Hirota Figs. 127, 128, and 129. — Diagrams of the relations of the embryonic mem- branes of the chick, constructed from preparations, and from figures and descriptions of Duval, Hans Virchow, Hirota and Fiilleborn. In these figures the ectoderm and entoderm are represented by plain Hnes: The mesoderm by a cross-hatched line or band. The yolk-sac is represented by broken parallel lines. In Fig. 127 the allantois is represented as a sac. In Figs. 128 and 129, where it is supposed to be seen in section, its cavity is represented by unbroken parallel lines. The stalk of the allantois is exaggerated in all the diagrams to bring out its connection with the em- bryo. The actual relations of the stalk is shown in Figures 33 and 82. Alb., Albumen. Alb. S., Albumen-sac. AIL, Allantois. AIL L, Inner wall of the allantois. All. C, Cavity of allantois. All. S., Stalk of allantois. AIL + Am., Fusion of allantois and amnion. Am., Amnion. Am. C, Amniotic cavity. Chor., Chorion. C. T. R., Connective tissue ring. Ect., Ectoderm. E.' E. B. C, Extra-embryonic body-cavity. Ent., Entoderm. Mes., Mesoderm. S.-Am., Sero-amniotic connection. S. Y. S. U., Sac of the yolk-sac umbilicus. Umb., Umbilicus. V. M., VitelHne membrane. Y. S. S., Septa of the yolk-sac. Fig. 127. — Fourth day of incubation. The embryo is surrounded by the amnion which arises from the somatic umbilicus in front and behind; the sero-amniotic connection is represented above the tail of the embryo; it consists at this time of a fusion of the ectoderm of the amnion and chorion. The allantois is represented as a sac, the stalk of which enters the umbihcus behind the yolk-stalk ; the allantois lies in the extra-embryonic body-cavity, and its mesoblastic layer is fused with the corresponding layer of the chorion above the embryo. The septa of the yolk-sac are represented at an early stage. The splitting of the mesoderm has progressed beyond the equator of the yolk-sac, and the undivided portion is slightly thickened to form the beginning of the connective tissue ring that surrounds the yolk-sac umbilicus. The ectoderm and entoderm meet in the zone of junction, beyond which the ectoderm is continued a short distance. The vitelline membrane is ruptured, but still covers the yolk in the neighborhood of the yolk-sac umbilicus. The albumen is not represented in this figure. Fig. 128. — Ninth day of incubation. The yolk-sac umbilicus has become much narrowed; it is surrounded by the mesodermal connective tissue ring, and by the free edges of the ectoderm and entoderm. The vitelline membrane still covers the yolk-sac umbilicus and is folded into the albumen. The allantois has expanded around the amnion and yolk-sac and its outer wall is fused with the chorion. It has pushed a fold of the chorion over the sero-amniotic connection, into which the mesoderm has penetrated, and thus forms the upper fold of the albumen-sac. The lower fold of the albumen-sac is likewise formed by a duplication of the chorion and allan- tois; it must be understood that lateral folds are forming also, so that the albumen is being surrounded from all sides. The stalk of the allantois is exaggerated so as to show the connection of the allantois with the embryo; it is supposed to pass over the amnion^ and not through the cavity of the latter, of course. EMBRYO AND EMBRYONIC ME^IBRANES 219 All: Am.' Chor. Umb. Fig. 127 Al/C Fig. 128 220 THE DEVELOPMENT OF THE CHICK states that, after this connection is estabhshed, the amniotic fluid coagulates in alcohol, "just like the fluid in the albumen- sac; owing, presumably, to the presence of albumen which has found its way through the perforations into the amniotic fluid." This observation is confirmed by Fiilleborn. The AUantois. The part of the w^all of the allantois that fuses with the chorion may be called the outer wall; the remainder of the sac of the allantois constitutes the inner w^all. The distal intermediate part of the allantois is specialized as the wall of the albumen-sac. AlhAm. AlIC. jSliS. Fig. 129. — Twelfth day of incubation. The conditions represented in Fig. 128 are more advanced. The albu- men-sac is closing; its connection with the cavity of the amnion by way of the sero-amniotic connection will be obvious. The inner wall of the allantois has fused extensively with the anmion. The umbilicus of the yolk-sac is much reduced, and some yolk protrudes into the albumen (sac of the yolk-sac umbilicus). In the outer wall there are three layers, viz., an internal epi- thelial laver, formed bv the entoderm of the allantois; a thick very vascular middle or mesodermal layer, formed by fusion of the mesoblast of allantois and chorion; and a thin, outer, ecto- dermal laver derived from the chorion. EMBRYO AXD EMBRYONIC MEMBRANES 221 Rate of Growth of the Allantois. As the embryo lies on its left side, the allantois grows out on the right side of the embryo (Figs. 127 and 130 A) and unites with the chorion about the one hundredth hour. It then spreads rapidly as a flattened sac over the embryo, increasing the extent of the fusion with the Fig. 130. — Diagrams showing the relations of the allantois, represented by the tinted area, at different ages. (After Hirota.) Alb., Albumen. Alb. S., Edge of albumen-sac. All. V., Allantoic vein. am. C, Amniotic cavity. S.-Am., Sero-amni- otic connection. Y. S., Yolk-sac. A. At 120 hours showing only the amniotic cavity and al- lantois X 2. • 1 1 B. At 144 hours, showing only the amniotic cavity and al- lantois X 1.2. C. At 192 hours; the entire yolk x .66. The dotted out- line represents the amniotic cavity. D. At 214 hours. The entire egg after removal of the shell, X .66. The allnimen mass is at the left ; the albumen-sac is be- ginning to form. chorion, hence of its outer wall pari passu. At the end of the fifth day it covers more than half of the embryo (Fig. 130 A); at the end of the sixth day the embryo is entirely covered by 222 THE DEVELOPMENT OF THE CHICK the allantois (Fig. 130 B) ; at the end of the eighth day the allan- tois has covered half of the yolk-sac (Fig. 130 C). At the end of the ninth day, the formation of the albumen-sac is begun (Fig. 130 D). At the end of the eleventh day, the albumen-sac is practically closed at the lower pole. On the twelfth day, the albumen-sac is closed, and on the sixteenth day the contents are practically entirely absorbed. Blood-supply of the Allantois. There are two allantoic arteries and one allantoic vein. (See Chap. XII). Both arteries persist throughout the period of incubation, but the left is much the best developed. It passes out along the stalk of the allantois to the inner wall of the allantoic sac, where it divides in two strong branches, one running cephalad and the other caudad to the margins of the sac where they pass over to the outer wall; The allantoic vein runs in the inner wall and passes over to the outer wall near the sero-amniotic connection. Both arteries and veins inhibit the expansion of the allantoic sac w^here they sur- round the margin; but the vein has by far the greatest effect, as its action is supplemented by the sero-amniotic connection. Thus indentations, gradually growing deeper, are established along the margins of the allantoic sac, and the outgrowth of the latter on each side of the indentations produce overlapping lobes (Figs. 130 C and D). The arrangement of the smaller vessels and capillaries is very different in the outer and inner walls. In the outer wall the arteries and veins branch and interdigitate in the deeper portions of the mesoblast, and end in an extraordinarily fine- meshed capillary netw^ork situated immediately beneath the thin ectoderm. "The capillaries form such narrow meshes, and have relatively so wide a lumen, that they can be compared only with those of the lungs of higher animals, and of the choroidea of the eye; indeed, instead of describing it as a vascular network embedded in tissue, one could as well describe it as a great blood-sinus interrupted by strands of tissue" (Fiilleborn.) This capillary network of the outer wall constitutes the respiratory area of the allantois. At the margins it passes gradually into the incomparably wider meshed capillary network of the inner wall. An extensive system of lymphatics is developed, both in the outer and inner walls of the allantois, accompanying all the blood-vessels, even to their ultimate terminations. EMBRYO AND EMBRYONIC :\IEMBRAXES 223 Structure of the Allantois. (1) Inner wall. The inner wall of the allantois consists primarily of two layers, an inner ento- dermal and outer mesodermal layer. The latter soon becomes differentiated into two layers, an external, delicate, limiting layer of flat polygonal cells, with interlocking margins, and an inter- mediate layer of star-shaped cells embedded in a homogenous mucous ground substance. Parts of the inner wall become extremely thin, and in these regions the intermediate layer may become entirely absent. Elsewhere, particularly around the larger arteries and veins, the intermediate layer may attain considerable thickness. The entoderm becomes reduced to a layer of flat, interlocking cells. On the eighth day, spindle- shaped muscle cells begin to appear in the mesoderm of the inner wall, and undergo rapid increase in numbers. Their dis- tribution is somewhat irregular; in certain places they may even form several layers, and in others are practically wanting. On the seventh dav the inner wall of the allantois begins to fuse with the amnion in the neighborhood of the sero-amniotic connection, and this fusion rapidly extends over the area of contact between the two membranes. Within the area of fusion the muscle lavers of the allantois and amnion mutuallv reinforce each other, and in places no boundary can be found between them (Fiilleborn). But during the latter half of incubation the musculature of the fused area of allantois and amnion degener- ates almost completely. Towards the end of incubation, part of the inner wall of the allantois fuses also with the yolk-sac, and is therefore carried with the latter into the body-cavity of the chick. (2) The Outer Wall of the Allantois. As already noted, the outer wall of the allantois fuses with the chorion. The compound membrane, which is respiratory in function, must be considered, therefore, as one. Over the entire respiratory area the ectoderm, belonging primarily to the chorion, which is elsewhere two layers of cells in thickness, becomes reduced to an exceedingly thin layer in direct contact with the walls of the capillaries internally and the shell membrane externally. According to Fiilleborn, the ectoderm cannot be distinguished as a separate layer in the latter half of incubation, and the capillaries appear to be in immediate contact with the shell-membrane. No muscular tissue appears to develop in the outer wall of the allantois. 224 THE DEVELOPMENT OF THE CHICK (3) The Albumen-sac. The allantois in the course of its expansion over the embryo, between amnion and chorion, reaches the sero-amniotic connection; it must then either divide and grow round on each side of the connection, or evaginate the chorion above the connection and carry it as an overlapping fold on beyond. The latter is what actually happens, and there is established as a consequence an overlapping fold of the chorion containing an extension of the allantois (Fig. 128); the space beneath this fold terminates, naturally, at the sero-amniotic connection. In the meantime the cleavage of the mesoblast has separated chorion and yolk-sac as far as the neighborhood of the yolk-sac umbilicus, where the viscid albumen has accumu- lated. The latter is situated not opposite to the yolk-stalky but near the posterior pole of the yolk-sac, with reference to the embryo, i.e., usually towards the narrow end of the shell. Now the allantois growing around the yolk-sac from all sides reaches the neighborhood of the albumen and enters an evagina- tion of the chorion that wraps itself around the albumen, thus initiating the formation of a double sac of the chorion enfolding the albumen and containing between its two layers an extension of the allantois. The latter is therefore separated everywhere from the albumen by the thickness of the chorion. The superior fold of the albumen-sac is the same fold that overgrows the sero-amniotic connection, and the albumen-sac is therefore pro- longed beneath this fold to the sero-amniotic connection itself, which, as we have seen, becomes perforated, thus admitting albumen into the amniotic cavity. The ectoderm lining the albumen-sac is two-layered, and the cells next the albumen tend to be cubical or swollen, and fre- quently vesicular, owing apparently to absorption of albumen. In the neighborhood of the yolk-sac umbilicus, papilla-like pro- jections of the ectoderm into the albumen are common (Fig. 129). But these do not occur over the remainder of the albumen-sac of the chick, as described by Duval for the linnet; nor do they possess a mesodermal core. Prior to the union of the mesoderm over the yolk-sac umbili- cus, the yolk forms a hernia-like protrusion into the albumen- sac (sac of the yolk-sac umbilicus, see Fig. 129), which is, however, retracted as the mesoderm ring closes over the yolk-sac umbilicus. The vitelline membrane ruptures at an early period of the incu- EMBRYO AND EMBRYONIC MEMBRANES 225 bation over the embryonic pole and gradually slips down over the yolk, and is finally gathered together in the albumen-sac. (4) The allantois also serves as a reservoir for the secretions of the mesonephros, and subsequently the permanent kidney, which reach it by way of the cloaca and neck of the allantois. The fluid part of the embryonic urine is absorbed, but the con- tained salts are deposited in the walls and cavity of the allantois. If the connection between the Wolffian ducts and cloaca be inter- rupted, the former become enormously extended by the secre- tions of the mesonephros. The Yolk-sac. The j'olk-sac is established in the manner already described; it is constituted by the extra-embryonic splanchnopleure, and is permanently united to the intestine by the yolk-stalk. A narrow lumen remains in the stalk of the yolk-sac throughout and even after incubation, but the yolk does not seem to pass through it into the intestinal cavity. The walls of the yolk-sac, excepting the part derived from the pellucid area, are lined with a special glandular and absorbing epithelium, which digests and absorbs the yolk and passes it into the vitel- Une circulation, through which it enters the hepatic portal circu- lation and comes under the influence of the hepatic cells. The yolk-sac is thus the primary organ of nutrition of the embryo, and it becomes highly elaborated for the performance of this function. Contrary to the statements found in many text-books, it does not reach its maximum development until the end of incubation. Throughout incubation it steadily increases in complexity and efficiency so as to provide for the extremely rapid growth of the embryo. The functions of the yolk-sac manifestly require a large sur- face area, which is provided for by foldings of the wall projecting into the yolk. At the height of its development the inner surface of the yolk-sac is covered with numerous folds or septa projecting into the yolk, w'hich are highest at the equator and decrease in both directions away from the equator. In general, these folds follow the direction of the main arteries, i.e., they run in a meridional direction, repeatedly bifurcating distally (Fig. 132). Moreover, each one is perforated by numerous stomata, and the yolk-sac epithelium covers all free surfaces, and a capillary net- work is found in every part. So far do they project into the interior towards the close of incubation, that those of opposite 226 THE DEVELOPMENT OF THE CHICK sides may be approximate!}^ in contact, and the cavity of the volk-sac is thus broken up into numerous connecting compart- ments filled with yolk. The outer wall of the yolk-sac is smooth and not involved in the folds. The beginning of the folds of the volk-sac mav be found at the time of appearance of the vascular area of the blastoderm, and they develop pari passu, with the vessels of the yolk-sac (Fig. 131). Fig. 131 shows the appearance of the folds at the stage of twelve somites. It is a view of the blastoderm from below. '^ Fig. 131. — Septa of the yolk-sac as seen on the lower surface of the blastoderm at the stage of 12 s. (After Hans Virchow.) m. R., Marginal ridge of entoderm overly- ing the sinus terminalis. drawn as an opaque object, and it shows the incipient folds of the yolk-sac in an arrangement that corresponds roughly, but not accuratelv, with that of the blood-islands, which lie in large part in the bases of the folds. The site of the vena terminalis is marked bv a circular fold of the entoderm. The folds of the yolk-sac thus coincide in their distribution Avith the vascular area and are so limited at all times, being absent in the vitelline area. There is thus a close connection between the vitelline blood- EMBRYO AND EMBRYONIC MEMBRANES 227 vessels and the folds of the yolk-sac, which will be considered more fully bevond. The interior of the yolk-sac is lined with entoderm which differs in its structure in different regions: In the area pellucida the cells are flattened; in the vascular zone of the area opaca are found the columnar cells with swollen ends described pre- viously. After the third or fourth day these are found filled with yellow fatty droplets, which give a yellow tone to the interior of the living yolk-sac, and which are so abundant in later stages as to render the layer perfectly opaque. These cells do not con- FiG. 132. — Part of the interior of the yolk-sac of a duck at the time of hatchng. In the upper part of the figure the septa are seen from the side showing the stomata. In the lower part they are seen on edge. Note the sinuous course of the arteries along the free edges of some of the septa. (After H. Virchow.) tain entire yolk-granules; apparently, then, the yolk-granules are digested l^efore absorption in this region. In the region of the inner zone of the vitelline area, the entoderm is composed of several layers of large cells containing yolk-granules, and in the outer vitelline zone we come to the germinal wall. The germinal wall and inner zone of the vitelline area represent the formative region of the yolk-sac epithelium in the manner already described (Chap. V). Blood-vessels of the Yolk-sac. The development of the circu- 228 THE DEVELOPMENT OF THE CHICK lation in the yolk-sac may be divided into the following stages (following Popoff): 1. Indifferent network bounded peripherally by the vena terminalis, connected by two anterior vitelline veins with the heart; no arterial trunks. 2. Origin of an arterial path in the network; the right anterior vitelline vein begins to degenerate. 3. Origin of intermediate veins; the (left) posterior vein begins to develop. 4. Development of collateral veins; further degeneration of the right anterior vein; complete formation of the posterior vein. 5. Further branching; development of a rich venous network; the vena terminalis begins to degenerate. 6. Definitive concUtion; development of a rich venous net- work in the folds or septa of the yolk-sac; anastomosis of vessels of the yolk-sac and allantois. The changes can be followed only in outline. The earliest condition has been described in Chapters IV and V. Fig. 133 shows a condition intermediate between stages 1 and 2 above. The network is entirely arterial, except towards the anterior end, i.e., the blood flows outwards away from the heart. It enters the vena terminalis and is returned by right and left an- terior vitelline veins to the heart. The beginning of arterial trunks in the network is indicated particularly on the left side (right side of the figure). The connection of the arterial network with the dorsal aorta is still net-like. Fig. 134 shows an advance of the same processes. The trunks of the vitelline arteries are better differentiated from the network, and the blood is still returned to the heart entirely by way of the vena terminalis and the right and left anterior vitelline veins, which have come in contact distally, circumscribing in their proximal parts the mesoderm-free area of the blastoderm. The beginning of the lateral vitelline veins is indicated, particularly on the right side (left of the figure). Fig. 135 represents a great advance. The vitelline arteries arise from the dorsal aortse as single trunks, and branch in the vascular network, some of them reaching as far as the vena terminalis. The two anterior vitelline veins have fused in front, and the right anterior vein is reduced in size so that most of the blood reaches the heart through the left anterior vein. But the Fig. 133. — Circulation in the embryo and the yolk-sac. Stage of about 16 s; from below. The vitelline arteries are beginning to differentiate out of the vascular network particularly on the left side. (Observer's right.) Injected. (After Popoff.) 1, Marginal vein. 2, Region of venous network. 3, First and second aortic arches. 4 r, 41, Right and left anterior vitelline veins. 5, Heart 6, Anterior intestinal portal. 7, Aortse. differentiation. 9, Blood islands. 8, Vitelline arteries in process of W. Fig. 134. — Circulation in the embryo and the yolk-sac at the stage of about 22 s, drawn from below. Note differentiation of branches of the vitelline arteries. Injected. (After Popoff.) 1, Marginal vein. 2, Region of venous network. 3, Carotid loop. 4 r, 4 1, Right and left anterior vitelline veins. 5, Heart. 6, Anterior intes- tinal portal. 7, Dorsal aorta. 8, Branches of vitelline arteries. EMBRYO AND EMBRYONIC MEMBRANES 229 most striking change is the transformation of part of the vascular network into channels in which the blood flows towards the heart. Of these there may be recognized the following: 1. Intermediate veins arising from the vena terminalis at various places and gradually losing themselves centrally in the vascular network. 2. The vascular network immediately behind the embryo has assumed a venous character and likewise a large part of the network immediately surrounding the embryo. 3. Lateral vitel- line veins are beginning to develoj) from the anterior intestinal portal backwards. Fig. 136, representing the circulation at a stage of about 40 somites, shows the completion of the primary circulation in the yolk-sac. The vitelline arteries branch richly, and end in a capillary network; very few arterial branches reach the vena terminalis as such, and then only very fine ones. The vena terminalis itself is relatively reduced; the lateral vitelline veins have absorbed the network between themselves and the inter- mediate veins, which now appear as prolongations of the lateral veins. The right anterior vitelline vein has disappeared almost entirely and the posterior vitelline vein is well developed, empty- ing into the left lateral vein. The lateral vitelline arteries and veins are superposed as far peripherally as the original intermediate veins, which lie between the arterial trunks. Wherever there is superposition of arteries and veins, the latter are superficial and the former deep in position as seen from above. The figure also shows the vascular network in the budding allantois, and some of the em- bryonic blood-vessels. In the later stages of development the arteries are carried in by the septa of the yolk-sac and lie near their free edges; the veins, on the other hand, remain superficial in position. The terminal vein becomes progressively reduced in importance up to about the tenth day, and then gradually disappears as such, being taken into the terminal capillaries. After the tenth day the anterior and posterior vitelline veins decrease in importance and finally become almost unrecognizable. The lateral veins, on the other hand, increase in importance and return all of the blood to the embrvo. The rich network of venous capillaries in the septa of the yolk-sac is shown in Fig. 137. It lies immediately beneath the 230 THE DEVELOPMENT OF THE CHICK epithelium over the entire extent of the septa and forms loops along the free border. The arteries do not communicate directly with this network according to Popoff, and the course of the circulation from arteries to veins is not clearly described by this author. The allantois fuses with the yolk-sac in the region of the yolk-sac umbilicus, and anastomoses arise between the veins of the allantois and those of the yolk-sac. Ultimate Fate of the Yolk-sac. On the nineteenth day of incubation, the yolk-sac slips into the body-cavity through the umbilicus; which thereupon closes. The mechanism of this process is of considerable interest. The yolk-sac is still a volu- minous organ, and equal to about one sixth the weight of the embryo. It is therefore inconceivable that it could be "drawn into" the body-cavity by means of its stalk, which has only the intestine for attachment. The process is much more complex and may be briefly described as follows: We have already seen that the inner wall of the allantois fuses with the amnion on the one hand; distally it is connected with the yolk-sac. Now this wall of the allantois is muscular, and it is probable that its con- traction is the first act in the inclusion of the yolk-sac within the body-wall. It is aided in this, however, by the inner wall of the amnion, i.e., that part of the amnion arising from the umbili- cus and not fused with the allantois. This part of the amnion surrounds the yolk-stalk, and is itself richly provided with muscle cells, forming a crossing and interlacing system. It is carried down and over the yolk-sac to about its equator by the allantois, and when the yolk-sac is half taken into the body-cavity, it reaches its distal pole and fuses there. Now if the egg be opened at this stage in the process and this wall of the amnion cut through, it contracts rapidly to a fraction of its former area (Virchow). It is apparent, then, that the tension of this membrane on the yolk-sac must exert a continuous pressure that tends to force it into the body-cavity. It is in this way, then, by contraction of the inner walls of the allantois and of the amnion, that the yolk-sac is pressed into the body-cavity. The umbilicus is therefore closed by the mere act of inclusion of the yolk-sac, for the inner amniotic wall is attached on the one hand to the body-wall, and on the other to the distal pole of the yolk-sac. A minute opening is left in the center of the Fig. 135. — Circulation in the embryo and yolk-sac after 74 hours' incuba- tion. Stage of about 27 s from below. Injected. (After Popoff.) 1 Marginal vein. 2 r, 21, Right and left anterior vitelline veins sur- rounding the mesoderm-free area. 3, Anterior intestinal Portal. 4 In- termediate veins connecting with the venous network centrally. 5, Right dorsal aorta. 6, Posterior vitelline vein in process of formation. /, Vitel- line arteries. Note that the right anterior vitelline vein (2 r) is much atrophied. Fig. 136. — Circulation in the embryo and yolk-sac of an embryo of about 40 s, showing the later development of the lateral and intermediate vitel- line veins. Reduction of vena terminalis (marginal vein). Almost com- plete atrophy of the right anterior vein. Injected. (After Popoff.) 1, Marginal vein. 2 r, 21, Right and left anterior vitelline veins. 3, Arch of aorta. 4, Left posterior cardinal vein. 5 r, 51, Right and left omphalomesenteric veins. 6, Aorta. 6 a. Left dorsal aorta. 7, Vitelline artery. 8, Posterior vitelline vein. 9, Vascular network in the allantois. EMBRYO AXD EMBRYONIC MEMBRANES 231 umbilical field, through which dried remnants of the inner wall of the allantois, which is likewise attached to the distal pole of the yolk-sac, protrude for a short time. On the inner side the yolk-sac is attached to the umbilicus by its distal pole, and by its stalk to the intestine. The absorption of the yolk-sac then goes on with great rapidity, being reduced from a w^eight of 5.34 gr. twelve hours after hatching to 0.05 gr. on the sixth day after hatching, according to a series of observations of Virchow. The Amnion. The amnion invests the embryo closely at the time of its formation, but soon after, fluid begins to accumulate within the amniotic cavity, which gradually enlarges so that the embryo lies within a considerable fluid-filled space, which in- creases gradually up to the latter part of the incubation, and then diminishes again, so that the embryo finally occupies most of the cavity. The connections of the amnion with the chorion, and later with the allantois, albumen-sac, and yolk-sac, have been already described. Muscle fibers appear in the walls of the amnion on the fifth or sixth day and gradually increase in number; though they subsequently degenerate over the area of fusion with the allan- tois. They persist elsewhere, however, and are active in the inclusion of the yolk-sac in the manner already described. Shortly after the appearance of the muscle fibers slow vermicular or peristaltic contractions of the amnion begin, and the embryo is rocked within the amniotic cavity. Apparently, adhesions are thus prevented, but they are sometimes formed and lead to various malformations of the embryo. In some cases the amnion fails to develop; in such cases, the embryo usually dies at a relatively early stage, though Dareste records an anamniotic embryo of thirteen days, apparently full of life and vigor. The amnion apparently acts first as a protection against all mechanical shocks and jars which are taken up by the fluid; second, by protecting the embryo against the danger of desicca- tion; third, by protecting it against adhesions with the shell- membrane and embryonic membranes, and lastly by providing space for the expansion of the allantois and consequent increase of the respiratory surface. It also has secondary functions in the chick in connection with the absorption of the albumen and the inclusion of the yolk-sac. It will be readily understood, then, why anamniotic embryos usually do not develop far. 232 THE DEVELOPMENT OF THE CHICK Hatching (after von Baer). About the fourteenth day the growing embryo accommodates itself to the form of the egg so as to he parallel to the long axis with its head usually towards the broad end near to the air-chamber. Sometimes, however, the embryo is turned in the reverse position (von Baer). The head is bent towards the breast, and is usually tucked under the right wing. Important changes preparatory to hatching take place on the seventeenth to the nineteenth days. The fluid decreases in the amnion. The neck acquires a double bend so that the head is turned forward, and, in consequence, the beak is towards that part of the membranes next to the air-chamber. The intestine is retracted completely into the body-cavity, and on the nineteenth day the yolk-sac begins to enter the body- cavity. On the twentieth day the yolk-sac is completely included, and practically all the amniotic fluid has disappeared. The chick now occupies practically all the space within the egg, outside of the air-chamber. The umbilicus is closing over. The ductus arteriosi begin to contract, so that more blood flows through the lungs. The external wall of the allantois fused with the chorion still remains very vascular. Now, if the chick raises its head, the beak readily pierces the membranes and enters the air-chamber. It then begins to breath slowly the contained air; the chick may be heard, in some cases, to peep within the shell two days before hatching, a sure sign that breathing has begun. But the circulation in the allan- tois is still maintained and it still preserves its respiratory func- tion. When the chick makes the first small opening in the shell, which usually takes place on the twentieth day, it begins to breathe normally, and then the allantois begins to dry up and the circulation in it rapidly ceases. It then becomes separated from the umbilicus, and the remainder of the act of hatching is completed, usually on the twenty-first day. Fig. 137. — Part of a septum of the yolk-sac. Injected. 20 days' incuba- tion. The free edge is above. (After Popoff.) Ar., Artery. St., Stomata. V. an., Longitudinal anastomoses of venous network. V., vein. CHAPTER VIII THE NERVOUS SYSTEIVI I. The Neuroblasts The account given in Chapters V and VI outlines the origin of the larger divisions of the central nervous system and ganglia. The subsequent growth and differentiation is due to multiplica- tion of cells, aggregation of embryonic nerve-cells, or neuro- blasts, in particular regions or centers, the formation and growth of nerve-fibers which combine to form nerves and tracts, and the origin and differentiation of nerve-sheaths, and the support- ing cells, neuroglia, of the central system. The most important factors are the origin of the neuroblasts and of nerve-fibers in connection with them; these fibers form the various nerve-tracts and commissures within the central nervous system and the system of peripheral nerves. The origin of neuroblasts and the development of fibers is the clue to differentiation in all parts of the nervous system. Neuroblasts are found in two primary locations in the embryo; (1) in the neural tube, and (2) in the series of ganglia derived from the neural crest; these are known as medullary and gang- lionic neuroblasts respectively.^ The Medullary Neuroblasts. In the neural tube of the chick, up to about the third day, there are present only two kinds of cells, the epithelial cells and the germinal cells (Fig. 138). The epithelial cells constitute the main bulk of the walls, and extend from the central canal to the exterior; their inner ends unite to form an internal limiting membrane lining the central canal, and their outer ends to form an external limiting membrane. Each cell in the lateral walls of the tube is much elongated and usually shows three enlargements, viz., at each end and in the region of the nucleus, the cell being somewhat constricted between the nucleus and each end. In different 1 Neuroblasts arise also in the olfactory epithelium. (See Chap. IX.) 233 234 THE DEVELOPMENT OF THE CHICK cells the nuclei are at different levels; thus in a section several layers of nuclei appear. These cells are not closely packed together, except at their outer ends, but are more or less separated by intercellular spaces that form a communicating system of narrow channels. ep.Crr, Fig. 138. — Structure of the wall of the neural tube. Trans- verse section through the region of the twenty-first somite of a 29 s embryo. Drawn with Zeiss 2 mm. oil-immersion. c. C, Central canal, ep. C, Epithelial cells, g. C, Ger- minal cells. Ms'ch., Mesenchyme. The germinal cells are rounded cells situated next the central canal between the inner ends of the epithelial cells; karyokinetic figures are very common in them. According to His the germinal cells are the parent cells of the neuroblasts alone; it is probable, however, that they are not so limited in function, and that they represent primitive cells from which proceed other epithelial cells and embryonic neuroglia cells as well as neuroblasts. THE NERVOUS SYSTEM 235 A narrow non-nucleated margin, known as the marginal velum, appears in the lateral Avails of the neural tube external to the nuclei (Fig. 138). This is occupied by the outer ends of the epithelial cells. At this time, therefore, three zones may be distinctly recognized in the walls of the neural tube, viz., (1) the zone of the germinal cells, including also the inner ends of the epithelial cells, (2) the zone of the nuclei of the epithelial cells, (3) the marginal velum. No distinctly nervous elements are yet differentiated. Such elements, however, soon begin to appear: Fig. 139 repre- sents a section through the cord of a chick embryo of about the end of the third day; it is from a Golgi preparation in which the distinctly nervous elements are stained black, and the epithelial and germinal cells are seen only very indis- tinctly. The stained elements are the neuroblasts, and it will be observed that they form a layer roughly intermediate in position between the marginal velum and the nuclei of the epithelial cells. They are usually regarded as derived from germinal cells that have migrated from their central position outwards; but it is mi3.^ M/4f MI .4. Fig. 139. — Transverse section through the spinal cord and ganglion of a chick about the end of the third day; prepared by the method of Golgi. (After Ramon y Cajal.) C, Cones of growth. Nbl. 1, 2, 3, 4, Neuroblasts of the lateral wall (1 and 2); of the spinal ganglion (3); of the ventral horn (motor neuroblasts) (4). possible that some of them may have been derived from epithelial cells. However this may be in such an early stage, it is certain that the neuroblasts formed later are derived from germinal cells. It will be observed that each neuroblast consists of a cell- body and a process ending in an enlargement. The process arises as an outgrowth of the cell-body, and forms the axis cylin- der or axone of a nerve-fiber; the terminal enlargement is known as the cone of growth, because the growth processes by which the axone increases in length are presumably located here. It may be stated as an invariable rule that each axone process of a medullary neuroblast arises as an outgrowth, and grows to its 236 THE DEVELOPMENT OF THE CHICK final termination without addition on the part of other cells. The body of the neuroblast forms the nerve-cell, from which, later on, secondary processes arise constituting the dendrites. The view that each nerve-cell with its axone process and dendrites is an original cellular individual, is known as the neurone theory. For the central nervous system this view is generally held, but its extension to the peripheral system is opposed by some on the ground that the axone in peripheral nerves is formed within chains of cells, and is thus strictly speaking not an original product of the neuroblast, though it may be continuous with the axis cylinder process of a neuroblast. This view is discussed under the peripheral nervous system. Each medullary neuroblast is primarily unipolar and the axone is the original outgrowth. Soon, however, secondary proto- plasmic processes arise from the body of the nerve-cell and form the dendrites. These appear first in certain neuroblasts of the ventro- lateral portion of the embryonic cord, whose processes enter into the ventral roots of spinal nerves (Fig. 140). The extent and kind of de- velopment of these dendritic pro- FiG. 140. — Transverse section cesses of the nerve-cells varies through the spinal cord of a extraordinarily in different regions; chick on the fourth day of jrigg. 1.39, 140, and 141 give an idea of their rapid development in the motor neuroblasts up to the eighth day. The Ganglionic Neuroblasts. The ganglionic neuroblasts are located, as the name implies, in the series of ganglia derived from the neural crest. It must not be supposed, however, that all of the cells of the ganglia are neuroblasts, for the ganglia contain, in all probability, large numbers of cells of entirely different function. (Sheath-cells, see peripheral nervous system.) It is probable also that the neuroblasts of the spinal ganglia and some cranial ganglia, at least, are of two original kinds, viz., the neuroblasts of incubation; prepared by the method of Golgi. (After Ra- mon y Cajal.) C. a., Anterior commissure. D., Dendrite, d. R., Dorsal root. Ep. Z., Ependymal zone. W., White matter (marginal velum). Nbl. 4, Neuroblast of the ventral horn (motor). THE NERVOUS SYSTEM 237 the dorsal root and of the sympathetic system. The first kind only is considered here, and they are usually called the gan- glionic neuroblasts s.s., because they alone remain in the spinal ganglia. Like the medullary neuroblasts these neuroblasts form outgrowths that become axis cylinder processes; but they differ from the latter in that each ganglionic neuroblast forms two axones, one from each end of the spindle-shaped cells, which are arranged with their long axis parallel to the long axis of the ganglion (Fig. 139). Thus we may distinguish a central process and a peripheral process from each neuroblast, the former grow- ing towards and the latter away from the neural tube (Fig. 139). In other words each ganglionic neuroblast is bipolar, as contrasted w^ith the unipolar medullary neuroblasts. The central axone enters the dorsal zone of the neural tube, and the peripheral one grows out into the surrounding mesenchyme. Fig. 141. — Transverse section through the spinal cord of a 9-day chick, prepared by the method of Golgi. (After Ramon y Cajal.) Col., Collaterals, d. R., Dorsal root. G., Gray matter. Gn., Ganglion. Nbl. 4, Neuroblast of the ventral horn (motor), v. R., Ventral root. W., White matter. In the course of the later development the cell-body moves to one side so that the central and peripheral branches appear nearly continuous (Fig. 141). Farther shifting of the cell-body produces the characteristic form of the ganglionic nerve-cell with rounded body provided with stem from which the central and peripheral branches pass off in opposite directions. The central process enters the marginal velum near its dorsal boundary and 238 THE DEVELOPMENT OF THE CHICK there bifurcates, producing two branches, one of which grows towards the head and the other towards the tail in the dorsal OoJ- FiG. 142. — Six centripetal axones of the dorsal root, rigorously copied from a good preparation prepared according to the method of Golgi. From a longitudinal and tangential section of the dorsal column of the spinal cord of an 8- day chick. (After Ramon y Cajal.) Col., Collaterals. 1, 2, 3, 4, 5, 6, the axones entering the cord. column of the white matter. The ascending and descencUng branches send off lateral branches, collaterals, which pass deeper into the cord, and ramify in the gray matter of the dorsal horn. THE NERVOUS SYSTEM 239 Fig. 142 represents six central processes of ganglionic neuroblasts entering the cord and branching as described. After this preliminary account of the neuroblasts we may take up the development of the spinal cord, brain, and peripheral nervous system. II. The Development of the Spinal Cord We have seen that the epithelial cells of the neural tube stretch from the lumen of the central canal to the exterior, and that the nuclei are arranged so as to leave the outer ends free, thus forming the marginal velum. In the roof and floor the epithelial cells are relatively low, and in the lateral zones much elongated. The epithelial cells are added to at first by transformation of some of the germinal cells; but they do not appear to multiply by division, and as development proceeds they become more and more widely sep- arated, the interstices being filled up by neuroblasts, embryonic glia cells, and fiber tracts. As the wall of the neural tube grows in thickness, the epithelial cells become more and more elongated, seeing that both external and internal connections are retained; and, as the growth takes place mainly external to their nuclear layer, the latter becomes reduced, relative to the entire thickness of the neural tube, to a comparatively narrow zone surrounding the central canal, and is now known as the ependyma (Fig. 143). Cilia develop on the central ends of the ependymal cells in the central canal, and from the outer end of each a branching process extends to the periphery anastomosing with neighboring epen- dymal processes so as to form a skeleton or framework enclosing the other cellular elements and fibers of the central system. Beginning with the third day a new layer appears between the nuclei of the epithelial cells and the marginal velum. This layer, known as the mantle layer, is composed of neuroblasts and embryonic glia cells, and represents the gray matter (Figs. 139 and 140). The white matter of the cord is laid down in the marginal velum. The sources of the cells composing the mantle layer may be twofold, viz., from the young epithelial cells or from the germinal cells. According to some authors young epithelial cells may be transformed into either neuroblasts or neuroglia cells. Thus the form of the youngest neuroblasts in Fig. 139 indicates derivation from epithelial cells, but this 240 THE DEVELOPMENT OF THE CHICK cannot be regarded as proved. Similarly intermediate stages between epithelial and true glia cells are apparently shown in Fig. 143. However, there can be but little doubt that the prin- cipal source of the neuroblasts of the mantle layer is the germinal cells, that migrate outwards between the bodies of the epithelial cells. The germinal cells continue in active division up to at least the eleventh day, and their activity seems sufficient to provide for all the cellular elements of the mantle layer, whereas the epithelial cells apparently do not divide at all. Moreover, mitoses are not infrequent in some cells of the mantle layer itself. Fig. 143. — - Transverse section of the cord of a nine-day chick to show neuroglia and ependymal cells; prepared by the method of Golgi. (After Ramon y Cajal.) D., Dorsal. Ep., Ependymal cells. N'gl., Neu- roglia cells, v., Ventral. The form of the gray matter in the cord in successive stages is shown in Figs. 144, 145, and 146, representing sections of the cord at five, eight, and twelve days. It will be seen that the gray matter gains very rapidh^ in importance between the fifth and the eighth days. Attention should be directed to a group of neuroblasts situated at the external margin of the white matter just above the ventral roots. This is known as Hoffmann's nucleus; it extends the entire length of the cord (Fig. 146, twelve days). The white matter of the cord gains in importance at an equal rate (Figs. 144, 145, 146). Its production is due to ascending THE XERVOUS SYSTEM 241 and descending tracts of fibers derived from medullarv and ganglionic neuroblasts. The dorsal and ventral roots of the spinal nerves divide it on each side into three main columns. viz., dorsal situated above the dorsal root, lateral situated be- tween dorsal and ventral roots, and ventral situated below the <%i^ 7 / . i ^P w. ^u \s M'lr Fig. 144. — Transverse section through the cervical swelhng of the spinal cord of a chick, middle of the fifth day. (After V. Kiipffer.) bl. v., Blood vessel. C. a., Anterior commissure. C, Cen- tral canal, d., Group of axones at the level of the dorsal root. Ep., Ependyma. N'bl., Neuroblasts. V. Ventral column of white matter. ventral roots. The dorsal column begins first as a bundle of fibers at the entrance of the fibers of the dorsal root (Fig. 144). Subsequently, other fibers come in this region and gradually extend towards the dorsal middle line, displacing the ependyma 242 THE DEVELOPMENT OF THE CHICK and gray matter (Fig. 145, eight days), but the dorsal columns of the two sides are still separated in the median line by a broad septum of epend3^mal cells. Later (Fig. 146, twelve daj-s) this septum becomes very narrow, and the accumulation of fibers in the dorsal columns causes the latter to project on each side of the middle line, thus forming an actual fissure between them. •3/}. OrJf. m.M I- ^\'- -1^ -■;-> V < • • ' -'e* ^.'/V J?, com ''y{Gn.^i//np. Fig. 145. — Transverse section through the spinal cord, and the eighteenth spinal ganghon of an eight-day chick. Centr., Centrum of vertebra, d. R., Dorsal root. Ep., Ependyma. Gn., Spinal Ganghon. Gn. symp., Sympathetic ganghon. Gr. M., Gray matter, m. N., Motor nucleus. R. com., Ramus communicans. R. d., Ramus dor- sahs. R. v., Ramus ventrahs. Sp., Spinous process of vertebra, v. R., Ventral root. Wh. M., White matter. Central Canal and Fissures of the Cord. The central canal passes through a series of changes of form in becoming the prac- tically circular central canal of the fully formed cord. Up to the sixth day it is elongated dorso-ventrally, usually narrowest in the middle with both dorsal and A^entral enlargements. About THE NERVOUS SYSTEM 243 the seventh day the dorsal portion begins to be obhterated by fusion of the ependymal cells, and is thus reduced to an epen- dymal septum. On the eighth day this process has involved the upper third of the canal; the form of the canal is roughly wedge- shaped, pointed dorsally and broad ventrally (Fig. 145). The continuation of this process leaves only the ventral division as the permanent canal. At the extreme hind end of the cord the central canal becomes dilated to form a relatively large pear-shaped chamber with thin undifferentiated walls (Fig. 148) ; the terminal wall is still fused with the ectoderm at eight days, and the chamber appears to have a maximum size at this time. At eleven days the fusion with the ectoderm still exists, and the cavity is smaller. r / >v;:-.'-i. iW'NJi. Fig. 146. — Transverse section through the cervical swelling of the spinal cord of a 12-day chick. (After v. Kupffer.) C, Central canal, d. H., Dorsal horn of the gray matter. Ep., Ependyma. N. H., Nucleus of Hoffmann, s. d., Dorsal fissure, s. v., Ventral fissure, v. H., Ventral horn of the gray matter. The development of the so-called dorsal and ventral fissures is essentially different. The entire ventral longitudinal fissure of the cord owes its origin to growth of the ventral columns of gray and white matter which protrude below the level of the original floor (Figs. 145 and 146), and the latter is thus left be- tween the inner end of the fissure and the central canal. The dorsal longitudinal fissure on the other hand is for the most part 244 THE DEVELOPMENT OF THE CHICK a septum produced by fusion of the walls of the intermediate and dorsal portions of the central canal; there is, however, a true fissure produced by protrusion of the dorsal columns of white matter (Fig. 146). This is, however, of relatively slight extent. The original roof of the canal is therefore found between the dorsal septum and the fissure. Neuroblasts, Commissures, and Fiber Tracts of the Cord. The medullary neuroblasts may be divided into four groups: (1) The first group, or motor neuroblasts, form the fibers of the ventral roots of the spinal nerves. These are situated originally in the ventrolateral zone of the gray matter (Figs. 144, 145, 146): they are relatively large and form a profusion of dendrites (Figs. 140, 141). As they increase in number and size they come to form a very important component of the ventral horn of the gray matter and contribute to its protrusion. (2) The second group may be called the commissural neuroblasts. These are situated originally mainly in the lateral and dorsal portions of the mantle layer, but are scattered throughout the gray matter, and their axis cylinders grow ventrally and cross over to the opposite side of the cord through the floor (Figs. 139 and 140), and thus form the anterior or white commissure of the cord. (3) The cells of the fiber tracts are scattered throughout the gray matter, and are characterized by the fact that their axis cylinders enter the white matter of the same side; here they may bifurcate, furnishing both an ascending and a descending branch, or may simply turn in a longitudinal direction. (4) Finally there are found certain neuroblasts with a short axis cylinder, ramifying in the gray matter on the same side of the cord. These are found in the dorsal horn of the gray matter and develop relatively late (about sixteen days, Ramon y Cajal). III. The Development of the Brain Unfortunately the later development of the brain of birds has not been fully studied. The following account is therefore fragmentary. It is based mainly on a dissection and sections of the brain of chicks of eight days' incubation. Fig. 147 is a drawing of a dissection of the brain of an eight- day embryo. The left half of the brain has been removed, and the median wall of the right cerebral hemisphere also. The details of the cut surfaces are drawn in from sections. Figs. 148 THE NERVOUS SYSTEM 245 and 150 show median and lateral sagittal sections of the same stage. The flexures of the brain at this stage are: (1) the cranial flexure marked by the plica encephali ventralis on the ventral surface, (2) the cervical flexure at the junction of myelencephalon and cord, somewhat reduced in this stage, and (3) the pontine flexure, a ventral projection of the floor of the myelencephalon. Cpzn.post. 0/A Fig. 147. — Dissection of the brain of an 8-day chick. For description see text. The arrows shown in the figure lie near the dorsal and ventral boun- daries of the foramen of Monro. ch. PI., Choroid plexus. Com. ant., Anterior commissure. Com. post., Posterior commissure. C. str., Corpus striatum. Ep., Epiphysis. H., Hemisphere. Hyp., Hypophysis. L. t.. Lamina terminalis. Myel., Myel- encephalon. olf., Olfactory nerve, op. N., Optic chiasma. op. L., Optic lobe. Par., Paraphysis. Paren., Parencephalon. pi. enc. v.. Plica en- cephali ventralis. pont. Fl., Pontine flexure. Rec. op., Recessus opticus. S. Inf., Saccus infundibuli. Tel. med.. Telencephalon medium. Th., Tha- lamus. T. tr.. Torus transversus. Tr., Commissura trochlearis. X. J^?.^^^^^ ^~^' ^~^' ^~^' *^"^' ^"«' ^-f' represent the planes of section A, B, C, D, E, and F of Fig. 151. Telencephalon. The telencephalon is bounded posteriorly, as noted in the last chapter, by the line drawn from the velum transversum to the recessus opticus. The telencephalon medium has grown but little since the fourth day, but the hemispheres 246 THE DEVELOPMENT OF THE CHICK Cp. ..^ V.I- Te/./ned CA Nem. l.t. Jfec. op. CAop. ' SJnf. JYyp. I J?.//(/p JVas. Aom Fig. 148. — Median sagittal section of an embryo of eight days. a. A., Aortic arch. All., AUantois. An., Anus. A. o. m., Om- phalomesenteric artery. B. F., Bursa Fabricii. b. P., Basilar plate. C. A., Anterior commissure, c. C, Central canal. Ch. op., Optic chiasma. C. p., posterior Commissure. CI., Cloaca. Cr., Crop, d. Ao., Dorsal aorta. D. Hyp., Duct of the hypophysis. Ep., Epi- physis. Fis. Eus., Fissura Eustachii. Hem., Surface of hemisphere barely touched by section. Hyp., Hypophysis. L. t., Lamina ter- minalis. n. A. 8, neural arch of the eighth vertebra. Nas., Nasal THE NERVOUS SYSTEM 247 have expanded enormously, particularly anteriorly and dorsally, and their median surfaces are flattened against one another in front of the lamina terminalis, which forms the anterior boundary of the telencephalon medium (Figs. 148, 149). Posteriorly the cerebral hemispheres extend to about the middle of the dien- cephalon and their lateral faces are rounded. The lateral walls of the hemispheres have become enormously thickened to form the corpora striata (Figs. 147 and 151 A), and the superior and lateral walls have remained relatively thin, forming the mantle of the cerebral hemispheres (pallium). Thus the cavity of the lateral ventricle is greatly narrowed. The dissection (Fig. 147) shows the corpus striatum of the right side forming the lateral wall of the hemisphere, and extend- ing past the aperture (foramen of Monro) between the lateral and third ventricles towards the recessus opticus, where it comes to an end. The olfactory part of the hemispheres is not well differen- tiated from the remainder in the chick embryo of eight days. There is, however, a slight constriction on the median and ventral face (Fig. 147) which may be interpreted as the boundary of the olfactory lobe. The telencephalon medium is crowded in between the hemi- spheres and the diencephalon; its cavity forms the anterior end of the third ventricle, and communicates anteriorly through two slits, the foramina of Monro, with the lateral ventricles in the hemisphere. In Fig. 147, the upper and lower boundaries of the foramen of Monro, are indicated by the grooves on either side of the posterior end of the corpus striatum. A hair intro- duced from the third ventricle into the lateral ventricle through the foramen of Monro in the position of the arrow in Fig. 147, can be moved up and down over the whole width of the striatum. The lateral walls of the telencephalon medium are formed by the posterior ends of the corpora striata and are therefore very thick. The lamina terminalis passes obliquely upwards and forwards cavity. Oes., Oesophagus, p. A., Pulmonary arch, par., Paraphysis. P. C, Pericardial cavity. Rec. op., Recessus opticus. R., Rectum. S. Inf., Saccus infundibuli. T., Tongue. Tel., Med. Telencephalon medium. Tr., Trachea. V. 1, 10, 20, 30, First, tenth, twentieth and thirtieth vertebral centra, r. A., right auricle. Vel. tr., Vehun transversum. V. o. m., Omphalomesenteric vein. V. umb., Umbilical vein. 248 THE DEVELOPMENT OF THE CHICK from the recessus opticus to the region between the foramina of Monro. It is very thin, excepting near its center, where it is thickened to form the torus transversus, containing the anterior commissure. At its dorsal summit it is continuous with the roof of the telencephalon medium, which has formed a pouch- like evagination, the paraphysis. Just behind the paraphysis Fig. 149. — Median sagittal section of the brain of a chick embryo of 7 days. (After v. Kupffer.) c, Cerebellum, ca., Anterior commissure, cd., Notochord. ch., Pro- jection of the optic chiasma. cp., Posterior commissure, e., Epiphysis, e'., Paraphysis. hy., Hypophysis. I., Infundibulum. It., Lamina termi- nahs. Lop., Optic lobe. M., Mesencephalon. Mt., Metencephalon. opt., Chiasma of the optic nerves, p., Parencephalon. ro., Recessus opticus, s., Saccus infundibuH. se., Synencephalon. tp., Mammillary tubercle, tp., Tuberculum posterius. tr., Torus transversus. Tr., De- cussation of the trochlear nerves. Va., Velum medullare anterius. Vi., Ventriculus impar telencephali. vp., Velum medullare posterius. is the velum transversum, where the roof bends upwards sharply into the roof of the diencephalon. The epithelial wall around the bend is folded to form the choroid plexus of the third ven- tricle, which is continued forward into the lateral ventricle along THE NERVOUS SYSTEM 249 the median wall of the hemisphere, ending anteriorly in a free branched tip (Fig. 147, ch. PL) The principal changes in the telencephalon since the third day comprise: (1) great expansion of the hemispheres and thickening of the ventro-lateral wall to form the corpora striata; (2) origin of the paraphysis which arises as an evagination of the roof just in front of the velum transversum about the middle of the fifth (lay; (3) formation of the choroid plexus; (4) origin of the anterior commissure within the lamina terminalis; (5) develop- ment of the olfactory region. The general morphology of the adult telencephalon is thus well expressed at this time. The Diencephalon has undergone marked changes since the third day. The roof of the parencephalic division has remained very thin, and has expanded into a large irregular sac (Figs. 147 and 148), situated between the hinder ends of the hemispheres. The attachment of the epiphysis has shifted back to the indenta- tion between parencephalic and synencephalic divisions, and the epiphysis itself has grown out into a long, narrow tube, dilated distally, and provided with numerous hollow buds. In the roof of the synencephalic division the posterior commissure has de- veloped (Fig. 147). In the floor the chiasma has become a thick bundle of fibers, and the infundibulum a deep pocket, from the bottom of which a secondary pocket (saccus infundibuli) is grow- ing out in contact with the posterior face of the hypophysis. Following the posterior wall of the infundibulum in its rise, we come to a slight elevation, the rudiment of the mammillary tubercles; just beyond this is a transverse commissure (the in- ferior commissure) ; and the diencephalon ends at the tuberculum posterius. The hypophysis has become metamorphosed into a mass of tulniles enclosed within a mesenchymatous sheath; the stalk is continuous with a central tubule representing the original cavity from which the other tubules have branched out (Fig. 148), and it may be followed to the oral epithelium from which the whole structure originally arose. The lateral walls of the diencephalon have become immensely thickened, both dorsally and ventrally, and a deep fissure (Fig. 147) is found on the inner face at the anterior end, between the dorsal and ventral thickenings. The deepest part of the fissure is a short distance behind the velum transversum; from this a 250 THE DEVELOPMENT OF THE CHICK -~M. OJf.l. Olf.N. / /yy# ) // X ^'30 Fig. 150. — Lateral sagittal section of an embryo of 8 days. Right side of the body. All. N., Neck of the allantois. Cbl., cerebelkim. Cr., Crop. E. T., Egg THE NERVOUS SYSTEM 251 short spur runs forward, a still shorter one ventrally, and the longest arm extends backwards, gradually fading out beyond the middle of the diencephalon. This fissure is not a continuation of the sulcus Monroi, or backward prolongation of the foramen of Monro, but is, on the contrary, entirely independent. The lateral thickenings of the diencephalon constitute the thalami optici, each of which may be divided into epithalamic, mesothalamic, and hypothalamic subdivisions. In the chick at eight days there is a deep fissure between the epi- and meso- thalamic divisions (the thalamic fissure, Fig. 147). The substance of the epithalamus forms the ganglion habenulae. The meso- thalamic and hypothalamic divisions are not clearly separated. The transition zone between the diencephalon and mesencephalon is sometimes called the metathalamus. The mesencephalon has undergone considerable changes since the third day. The dorso-lateral zones have grown greatly in extent, at the same time becoming thicker, and have evaginated in the form of the two large optic lobes. Hence the median portion of the roof is sunk in between the lobes (Fig. 147), and is much thinner than the walls of the lobes. The dorso-lateral zones and roof thus form a very distinct division of the mesen- cephalon, known as the tectum lohi optici. The ventro-lateral zones and floor have thickened greatly and form the basal divi- sion of the mesencephalon. The ventricle of the mesencephalon thus becomes converted into a canal (aqueduct of Sylvius), from which the cavities of the optic lobes open off. In the metencephalon likewise there is a sharp distinction between the development of the dorso-lateral zones and roof, on the one hand, and the ventro-lateral zones and floor on the other. From the former the cerebellum develops in the form of a thickening overhanging the foiu'th ventricle. This thick- ening is relatively inconsiderable in the middle line (cf. Figs. 148 and 150). Thus the future hemispheres of the cerebellum are tooth. Eiist., Eustachian tube. Gn. 1, 13, First and thirteenth spinal ^angha. Gon., Gonad. Hem., Hemisphere. Lag., Lagena. Lg., Lung. M., Mantle of Hemisphere. Msn., Mesonephros. Olf. L., Olfactory lobe. Olf. N., Olfactory nerve. P. C, Pericardial cavity. Pz. 5, The fifth post-zyga- pophysis. R. C. 1, 2, Last two cervical ribs. R. th. 1, 5, First and fifth tho- racic ribs. S. pc-per., Septum pericardiaco-peritoneale. S'r., Suprarenal. Symp., Main trunk of the sympathetic. Str., Corpus striatum. V. 1, 10, 20, 30, First, tenth, twentieth and thirtieth vertebral arches. V. C. I., Vena cava inferior. V. L. L., Ventral ligament of the liver. 252 THE DEVELOPMENT OF THE CHICK indicated. The surface is still smooth at the eighth day, but on the tenth and eleventh days folds of the external surface begin to extend into its substance, without, however, invaginat- ing its entire thickness. These are the beginnings of the cere- bellar fissures. The floor and ventro-lateral zones of the metencephalon enter into the formation of the pons. In the roof of the isthmus, or constricted region between cerebellum and mesencephalon, is found a small commissure produced by decussation of the fibers of the trochlearis (Fig. 147). In the wall of the myelencephalon the neuromeres have dis- appeared. The thin epithelial roof has become more expanded in the anterior part (Figs. 147 and 148). Floor and sides have become greatly thickened. Commissures. The brain commissures existing at eight days are the anterior, posterior, inferior, and trochlearis (Fig. 149). In the next four or five days two more appear, viz., the com- missura pallii anterior (KupfTer), corresponding to the corpus callosum of mammalia and the commissura habenularis. The development of the various nuclei and fiber tracts of the bird's brain is entirely unknown and affords an interesting topic for research. IV. The Peripheral Nervous System The peripheral nervous system comprises the nerves which span between peripheral organs and the central nervous system. There are fifty pairs in a chick embryo of eight days, of which twelve innervate the head, and thirty-eight the trunk, distin- guished respectively as cranial and spinal nerves. It is con- venient for purposes of description to consider cranial and spinal nerves separately, and to take up the spinal nerves first because they are much more uniform in their mode of development than the cranial nerves, and also exhibit a more primitive or typical condition, on the basis of which the development of the cranial nerves must be, in part, at least, explained. The Spinal Nerves. Each spinal nerve may be divided into a somatic portion related primarily to the somatopleure and axis of the embryo, and a splanchnic portion related primarily to the splanchnopleure and its derivatives. In each of these again a motor and sensory component may be distinguished. Thus each THE NERVOUS SYSTEM 253 hLK / C.jfr. A ^:>::^ ^^^/r z./. H;^ :< :U iIl(Telme(f.) D CN. G/i.V' Fig. 151. — Six transverse sections through the brain of an 8-day chick in the planes represented in Fig. 147. Cbl., Cerebelkim. F. M., Foramen of Monro. Gn. V., GangHon of the trigeminus. Isth., Isthmus. It. d., Diverticuknn of the iter. lat. V., Lateral ventricle. Other abbreviations as before (Fig. 147). 254 THE DEVELOPMENT OF THE CHICK spinal nerve has four components, viz., somatic motor, somatic sensory, splanchnic motor, and splanchnic sensory, the two latter constituting the so-called sympathetic nervous system. It is obvious, of course, that the splanchnic components must be missing in the caudal nerves. The somatic and splanchnic com- ponents will be considered separately. Somatic Components. Each spinal nerve arises from two roots, dorsal and ventral (Fig. 145). The fibers of the former arise from the bipolar neuroblasts of the spinal ganglia; the fibers of the ven- tral root, on the other hand, arise from a group of neuroblasts in the ventral portion of the cord. The roots unite in the interver- tebral foramen to form the spinal nerve. Typically, each spinal nerve divides almost immediately into three branches, viz., a dor- sal branch, a ventral branch, and a splanchnic branch to the sym- pathetic cord; the last is known as the ramus communicans. Fig. 145 represents a section passing through the twentieth spinal nerve of an eight-day chick. The dorsal and ventral roots unite just beneath the spinal ganglion; fibers are seen entering the sympathetic ganglion {ramus communicans); the ventral branch passes laterally a short distance where it is cut off; beyond this point it can be traced in other sections in the next posterior intercostal space more than half-way round the body-wall; that is, as far as the myotome has extended in its ventral growth. The dorsal branch arises at the root of the ventral and passes dorsally in contact with the ganglion to branch in the dorsal musculature. This nerve may be regarded as typical of the spinal nerves generally. There are thirty-eight spinal nerves in an embryo of eight days. The first two are represented only by small ventral roots. The first two spinal ganglia are rudimentary in the embryo and absent in the adult, hence the ganglion illustrated in Fig. 145 is the eighteenth of the functional series (see Fig. 149); it lies between the nineteenth and twentieth vertebrae. The fourteenth, fifteenth, and sixteenth are the principal nerves of the brachial plexus, and have unusually large ganglia. The twenty-third to the twenty-ninth are the nerves of the leg plexus, the thirtieth to the thirty-second innervate the region of the cloaca and the remainder are caudal. The special mor- phology of the spinal nerves does not belong in this description. THE NERVOUS SYSTEM 255 There are one or two vestigial ganglia behind the thirty-eighth nerve, evidently in process of disappearance at eight days. The early history of the spinal nerves is as follows: The axis cylinder processes of the fibers begin to grow out from the neuro- blasts about the third day (cf. p. 235). At this time the myo- tomes are in almost immediate contact with the ganglia; thus the fibers have to cross only a very short space before they enter the myotome. The further growth is associated with the growth and differentiation of the myotome between which and the embryonic nerve there is a very intimate relation of such a sort that the nerve follows the myotome and its derivatives in all changes of position. Thus nerves do not need to grow long distances to establish their connections, but these are formed at a very early period. This accounts for the motor fibers; the way in which the sensory fibers, that arise from the spinal ganglia, reach their termination is not known. Sheath-cells and Cell-chain Hypothesis. No embryonic nerve consists entirely of axones, but, from the start, each nerve trunk contains numerous nuclei. The latter belong to cells which have been given two radically different interpretations, corresponding to two distinct theories concerning the neuraxone. (1) The first theory, known as the neurone theory, is the one tacitly followed in the preceding description and may be stated as follows: the nerve-cell, dendrites and axone, including the terminal arborization, constitute a single cellular individual or unit, differentiated from the neuroblast alone. The nuclei in the embryonic nerves therefore belong to cells that are foreign to the primary nerve. Their function is to form the various sheaths of the nerves, viz., the sheaths of the individual axones and the endo-, peri-, and epineurium. The sheath of Schwann arises from such cells that envelop the individual fibers at suitable distances and spread longitudinally until neighboring sheath cells meet; each such place of meeting constitutes a node of Ranvier. Until recently it has been universally believed that the sheath cells arose from the mesenchyme; but recent observations on Am- phibia and Selachia have shown that they arise from the gangfia in these forms ; their original source is therefore the ectoderm. It is probable that they have the same origin in the chick, though this has not been demonstrated by direct observation or experiment. (2) The second theory is known as the cell-chain hypothesis. 256 THE DEVELOPMENT OF THE CHICK According to this the axones of peripheral nerves arise as differ- entiations of the sheath-cells in situ; continuity of the axone is established by arrangement of these cells in rows, and union with the neuroblast is essentially secondary. The entire axone is thus by no means an outgrowth of the neuroblast; at most its proximal portion is. Bethe (1903) expresses the idea thus: "Between the cord of the embryo and the part to be innervated there is formed primarily a chain of nuclei around w^hich the protoplasm is condensed. This is fundamentally an extended syncytium in which the nuclei of the neuroblasts and of the nerve-primordium lie. Within the denser protoplasm which appears as the body of the nerve- cells, axones differentiate by condensation, and these extend from one cell to the next, and so on to the condensations which are called neuroblasts. The differentiated axones tend more and more to occupy the center of the embryonic nerve, where they appear to lie free, though as a matter of fact they are still embedded in the general plasma which is no longer distinctly visible on account of its lesser density. Since the axones remain in firm connection with the neuroblasts, it appears in later stages as if they were processes of these and had nothing to do wdth their original formative cells." This view is essentially that of Balfour, Beard, and Dohrn; the neurone hypothesis was first clearly formulated in embryo- logical terms by His, and has been supported by the investiga- tions of a considerable number of observers, notably Ramon y Cajal, Lenhossek and Harrison, The neurone hypothesis has far stronger embryological sup- port than the cell-chain hypothesis so far as peripheral nerves are concerned; moreover, it is the only possible hypothesis of the development of nerve tracts in the central system, because cell-chains are entirely lacking here during the formation of these tracts, in which axones may have as long a course as in most peripheral nerves. It still remains to be seen w^hether the neurone hypothesis will be modified in any important w^ay by observa- tions on the development of peripheral nerves. Splanchnic Components {Sympathetic Nervous System). Two views have been held concerning the origin of the sympathetic nervous system: (a) that it is of mesenchymal origin, its elements arising in situ; (h) that it is of ectodermal origin, its elements THE NERVOUS SYSTEM 257 migrating from the cerebro-spinal ganglia to their definitive positions. The first view was held b}- the earlier investigators and was originally associated with the extinct idea that the spinal ganglia were mesenchymal in origin; the view has been largely, but not entirely, abandoned. The second view was partly established with the discovery that the spinal ganglia are of ectodermal origin, and that the ganglia of the main sym- pathetic trunk arise from the spinal ganglia; but there is some difference of opinion yet in regard to the peripheral ganglia of the sympathetic system, and especially the plexuses of Meissner and Auerbach in the walls of the intestine. However, the preponderance of evidence and logic favors the view of the ectodermal origin of the entire sympathetic nervous system. The first clear evidences of the sympathetic nervous system of the chick are found at about the end of the third or the begin- ning of the fourth day; at each side of the dorsal surface of the aorta there is found in cross-section a small group of cells massed more densely than the mesenchyme and staining more deeply. Study of a series of sections shows these to be a pair of longi- tudinal cords of cells beginning in the region of the vagus, where they lie above the carotids, and extending back to the beginning of the tail; the cords are strongest in the region of the thorax, and slightly larger opposite each spinal ganglion. Cells similar to those composing the cords are found along the course of the nerves up to the spinal ganglia, and careful study of earlier stages indicates that the cells composing the cords have migrated from the spinal ganglia. The two cords constitute the primary sym- pathetic trunks. Fig. 152 is a reconstruction of the anterior spinal and sym- pathetic ganglia of a chick embryo of four days. The primary sympathetic trunk is represented by a cord of cells enlarged opposite each ganglion and united to the spinal nerve by a cellu- lar process, the primordium of the ramus communicans. In the region of the head the segmental enlargements are lacking. No other part of the sympathetic nervous system is formed at this time with the exception of a group of cells situated in the dorsal mesentery above the yolk-stalk; these are destined to form the ganglion and intestinal nerves of Remak. They have not been traced back to the spinal ganglia, but it is probable that such is their orisiin. 258 THE DEVELOPMENT OF THE CHICK In the course of the fourth and fifth days aggregations of sympathetic gangUon cells begin to appear ventral to the aorta, and in the mesentery near the intestine. The connection of these with the primary cord is usually rendered evident by agreement in structure, and by the presence of intervening strands of cells; moreover, in point of time they always succeed the primary cord, so that their origin from it can hardly be doubted. About the sixth day the secondary or permanent sympathetic trunk begins to appear as a series of groups of neuroblasts situ- ated just median to the ventral roots of the spinal nerves. They Fig. 152. — Reconstruction in the sagittal plane of the anterior spinal and sympathetic gan- glia of a chick embryo of 4 days. (After Neumayer.) Ceph. 8., Cephalic continuation of the sym- pathetic trunk. S. C, Sympathetic cord. Sg., Sympathetic ganghon. sp., Spinal nerve, spg., Spinal ganglion. R. C, Ramus communicans. are thus separated from the spinal ganglia only by the fibers of the ventral roots between which neuroblasts may be found, caught apparently in migration from the spinal to the sympa- thetic ganglion. The number of these secondary sympathetic ganglia is originally 30, one opposite the main vagus ganglion, and each spinal ganglion to the twenty-ninth (Fig. 150). Soon after their origin they acquire three connections by means of axones, viz., (a) central, with the corresponding spinal nerve- THE NERVOUS SYSTEM 259 root by means of strong bundles of fibers; (6) peripheral, with certain parts of the original primary sympathetic cord; (c) longi- tudinal, the entire series being joined together by two longitudinal bundles of fibers uniting them in a chain. The central connec- tions constitute the rami communicantes , and are as numerous as the sympathetic ganglia themselves; but so close is the approxi- mation of the sympathetic ganglion to the roots of the spinal nerves that they are not visible externally, the ganglion appear- ing to be sessile on the root (Fig. 145); sections, however, show the fibers. The peripheral connections constitute the various nerves of the abdominal viscera; these are not metameric; their number and arrangement is shown in Figure 153. In the period between the fourth and the eighth day the pri- mary sympathetic cord becomes resolved into the various ganglia and nerves constituting the aortic plexus, the splanchnic plexus, and the various ganglia and nerves of the wall of the intestine. Remak's ganglion has grown and formed connections with the splanchnic plexus, and other parts of the primary sympathetic cord. The details of these various processes are too complex for full description; they are included in part in Figs. 153 and 154. Ganglia and Nerves of the Heart. The development of the cardiac nerves is of special interest on account of its bearing on the physiological problem of the origin of the heart-beat. The heart of the chick begins to beat long before any nervous connections with the central system can have been established; indeed, the rhythmical pulsation begins at about the stage of 10 somites when the neural crest is yet imdifferentiated, and no neuroblasts are to be found anywhere. Either, then, the heart-beat is of mus- cular origin (myogenic), or, if of nervous origin, the nerve-cells concerned must exist in the wall of the cardiac tube ah initio. The first trace of nerve-cells is found in the heart of the chick about the sixth day. These cells are at the distal ends of branches of the vagus, with which they have grown into the heart. Pre- vious to this time these neuroblasts are found nearer to the vagus along the course of the arteries. There can be but little doubt that they have arisen from the vagus ganglion and that they reach the heart by migration. Such an origin has been demon- strated with great pro])ability for all the known nervous elements of the heart of the chick. (See Wilhelm His, Jr., Die Entwickelung des Herznervensystems bei Wirbelthieren.) 260 THE DEVELOPMENT OF THE CHICK cQ I o .2 -2 • =^ S Ok, - • O - ^ ti • . ■ ■ '-^ ^ S:^ S C3 OJ "^ r^ £ "^ -^ -tJ ^ • CO Oh a 03 s o o CO b£} c3 03 o o3< =» . > ^l3 o 0) "^—1 '~; r" C >5 O ^ in -^ fe U O CO QJ > £ fi >>o3 ^^ 05 >.•-;: ^ b£ _^ c o3 o • ^■^ - .. P • O O CO *" "-dec O o3 ^ a THE NERVOUS SYSTEM 261 If any cardiac nervous elements arise in situ, they certainly remain undifferentiated until those that have a ganglionic origin have already entered the heart. The Cranial Nerves. The nerves of the head exhibit a much greater degree of heteronomy than the spinal nerves, and, in spite of much study, knowledge of their embryonic development is still in a very unsatisfactory condition. The same principles, however, apply to the development of both cranial and spinal nerves; the axones of the former like those of the latter arise either from medullary or ganglionic neuroblasts which are re- spectively unipolar and bipolar; but the cranial ganglionic and Fig. 1.54. — Diagram of the relations of the parts of the sympathetic nervous system as seen in the cross-section. (After His, Jr.) M., mesentery. Msn., Mesonephros. Other abbreviations same as Fig. 153. medullary nerve-nuclei are not similarly segmented, as in the case of the spinal nerves, and hence the axones are not related as dorsal and ventral roots of single nerve trunks; nor has the attempt to interpret the cranial nerves as homologues of dorsal and ventral roots respectively been successful in the case of the most important nerves. Moreover, the olfactory and optic nerves differ from the spinal type even more fundamentally. The olfac- tory is a sensory nerve that arises apparently from the olfactory 262 THE DEVELOPMENT OF THE CHICK epithelium, and the optic is really comparable to an intramedul- lary nerve tract, seeing that its termination lies in a part of the original wall of the neural tube, viz., the retina. Groups of medullary neuroblasts giving rise to axones of motor cranial nerves are located in the brain as follows, according to His: Oculo-motor nucleus in the mid-brain. Trochlearis nucleus in the isthmus. Motor trigeminus nucleus in the zone of the cerebellum, including the descending root. Abducens and facialis nuclei, beyond zone of greatest width of the fourth ventricle (auditory sac zone). Glossopharyngeus, vagus, in the region of the calamus scrip- torius. Accessorius and hypoglossus, in the region extending to the cervical flexure. These constitute the cranial motor nerve nuclei, and are more or less discontinuous. The ganglionic nerves or nerve-components of the head arise from the following primitive embryonic ganglion-complexes: 1. Complex of the trigeminus ganglia. 2. Complex of the acustico-facialis ganglia. 3. Complex of the glossopharyngeus ganglia. 4. Complex of the vagus ganglia. The early history of these ganglion-complexes has already been considered; the}^ are called complexes because each forms more than one definitive ganglion. It is probable also that each con- tains sympathetic neuroblasts, which may separate out later as dis- tinct ganglia, thus resembling the spinal sympathetic neuroblasts. There is no close agreement in the segmentation of the motor neuroblasts within the brain and that of the ganglion complexes. For instance, in the region of the trigeminal ganglionic complex, the motor nuclei of the oculo-motor, trochlearis, and trigeminus are found, and in the region of the vagus ganglionic complex, the motor nuclei of vagus, accessorius, and hypoglossus. Thus the medullary and ganglionic nerves of the head are primitively separate by virtue of their separate origins. They may remain entirely so, as in the case of the olfactory, trochlearis, and abdu- cens, or they may unite in the most varied manners to form mixed nerves. THE NERVOUS SYSTEM 263 The motor nuclei of the ociilo-motor, trochlearis, abdiicens, and hypoglossus nerves lie in the same plane as the motor nuclei of the spinal nerves, i.e., in the line of prolongation of the ventral horns of the gray matter. The motor nuclei of the trigeminus, facialis, glossopharyngeus, vagus, and spinal accessory on the other hand lie at a more dorsal level, and the roots emerge there- fore above the level of origin of the others. It will be noted that these are the nerves of the visceral arches, whereas those cranial nerves that continue the series of spinal ventral roots innervate myotomic muscles, like the latter. Similarly the ganglia of the pharyngeal nerves (V, VII, IX, and X) differ from spinal ganglia in certain important respects: the latter are derived entirely from the neural crest, whereas a certain portion of each of the primary cranial ganglia is derived from the lateral ectoderm of the head, as noted in the preceding chapter. Thus the pharyn- geal nerves form embryologically a class by themselves, both as regards the medullary and also the ganglionic components. 1. The Olfactory Nerve. The embryonic origin of the olfactory nerve has been a subject of much difference of opinion: thus it has been maintained by a considerable number of workers that it arises from a group of cells on each side situated between the fore-brain and olfactory pits; some of these maintained that these cells arose as an outgrowth from the fore-brain, others that they came from the epithelium of the olfactory pit, and yet others that this group of cells, or olfactory ganglion, was derived from both sources. This group of cells was supposed by some to include a large number of bipolar neuroblasts, one process of which grew towards the olfactory epithelium and the other tow^ards the fore-brain, entering the olfactory lobe and ending there in terminal arborization. This view^ is, however, in conflict with the ascertained fact that the fibers of the fully formed olfactory nerve are centripetal processes of olfactory sensory cells situated in the olfactory epithelium. The most satisfactory account of the origin of the olfactory nerve in the chick is that of Disse. This author finds two kinds of cells in the olfactory epithelium of a three-day chick, viz., epithelial cells, and germinal cells which become embryonic nerve-cells or neuroblasts. At this time the olfactory epithelium is separated from the wall of the fore-brain by only a very thin layer of mesenchyme. Early on the fourth day axones arise 264 THE DEVELOPMENT OF THE CHICK from the central ends of the neuroblasts and grow into the mesenchyme towards the fore-brain. At the same time groups of epithelial cells free themselves from the inner face of the olfactory epithelium, and come to lie between this and the fore- brain. The axones of the neuroblasts grow between these cells until they reach the base of the fore-brain over which they spread out, entering the olfactory lobe about the sixth day (Figs. 155 and 156). In the meantime the peripheral ends of the olfactory neuroblasts have extended out as broad protoplasmic processes to the surface of the olfactory epithelium, and thus form the per- cipient part of the olfactory sense-cells. Fig. 155. — Olfactory epithelium of a chick embryo of 5 days, prepared by the method of Golgi. (After Disse.) a, b, and c indicate different forms of neuroblasts in the olfactory epithelium. The epithelial cells between fore-brain and olfactory pit, through which the axones of the olfactory neuroblasts grow, are for the most part supporting and sheath-cells of the nerve, but they in- clude a few bipolar neuroblasts (Fig. 156). The latter are to be considered as olfactory neuroblasts with elongated protoplas- mic processes. Rubaschkin finds a ganglion, which he calls ganglion olfactorium nervi trigemini, situated beneath the olfactory epithelium in a nine- day chick. The bipolar cells send out processes peripherally which end in fine branches between the cells of the olfactory mucous membrane, and centrally, which go by way of the ramus olfactorius nervi trigemini towards the Gasserian ganglion. 2. The Second Cranial or Optic Nerve. The course of this THE NERVOUS SYSTEM 265 nerve is entirely intramedullary, the retina being part of the wall of the embryonic brain; its development will therefore be considered in connection with the development of the eye. NM. o/A/i Fig. 156. — Sagittal section through the head of a chick embryo of 5 days, showing the floor of fore-brain, olfactory pit, and developing olfactory nerve between. (After Disse.) a.. Unipolar neuroblasts near the olfactory epithelium, b., Bipolar cell in the olfactory nerve, c, Unipolar cell near the brain. F. B., Floor of fore-brain. N'bl., Neuroblast in the olfactory epithelium, olf. Ep., Olfac- tory epithelium, olf. N., Olfactory nerve, olf. P., Cavity of olfactory pit. 3. The third cranial or oculo-motor nerve arises from a group of neuroblasts in the ventral zone of the mid-brain near the median line, and appears external to the wall of the brain at about sixty hours (about 28-30 somites). At this time it appears as a small group of axones emerging from the region of the plica encephali 266 THE DEVELOPMENT OF THE CHICK ventralis, and ending in the mesenchyme a short distance from its point of origin. At seventy-two hours the root is much stronger, interpenetrated with mesenchyme and ends between the optic cup and floor of the brain behind the optic stalk (cf. Fig. 101). At ninety-six hours the root is broad and fan-shaped, the nerve itself is comparatively slender, and passes downwards and backwards behind the optic-stalk where it enters a well- defined ganglion situated just median to the ophthalmic branch of the trigeminus; this is the ciliary ganglion; beyond it the fibers of the oculo-motor turn forward again to enter the region of the future orbit. According to Carpenter (1906) the ciliary ganglion arises from two sources: (a) migrant medullary neuroblasts that pass out into the root of the oculo-motor, and follow its course to the definitive situation of the ciliary ganglion, and (b) a much smaller group of neuroblasts that migrate from the gangUon of the trigeminus along the ophthalmic branch, and by way of a ramus communicans to the ciliary ganglion. The adult ciliary ganglion shows correspondingly two component parts: (a) a larger ventral region composed of large bipolar ganglion cells, and (6) a smaller dorsal region containing small ganglion cells with many sympathetic characters. It is probable that the medullary fibers of the oculo-motor nerve are distributed entirely to the muscles innervated by it, viz., the superior, inferior, and internal rectus and inferior oblique muscles of the eye. The fibers arising from the neuroblasts of the ciUary ganglion ter- minate peripherally in the intrinsic muscles of the eye-ball, and centrally (in the case of the bipolar cells) in the brain, which they reach by way of the medullary nerve. The motor branches leave the trunk of the nerve a short distance centrally to the ciliary ganglion. 4. The trochlearis or fourth cranial nerve is peculiar inas- much as it arises from the dorsal surface of the brain in the region of the isthmus. It arises entirely from medullary neuro- blasts and innervates the superior oblique muscle of the eye. Marshall states that it may be readily seen in a five-day embryo; in an embryo of eight days it is a slender nerve arising from the dorsal surface of the isthmus immediately in front of the cere- bellum; the fibers of the two sides form a commissure in the roof of the isthmus (Fig. 148). THE NERVOUS SYSTEM 267 5. The trigeminus or fifth cranial nerve consists of motor and sensory portions. The latter arises from the trigeminal ganglion, the origin of which has already been described. The ganglionic rudiment appears roughly Y-shaped even at an early stage (cf. Figs. 105 and 117), the short stem lying against the wall of the brain and the two branches diverging one in the direc- tion of the upper surface of the optic cup (ophthalmic branch) and the other towards the mandibular arch. The original con- nection of the ganglion with the roof of the neural tube is lost during the second day and permanent connection is established during the third day, presumably by growth of axones into the wall of the brain. The new connection or sensory root of the trigeminus is attached to the myelencephalon in the region of greatest width of the fourth ventricle near the ventral portion of the lateral zone. During the fourth day the peripheral axones follow the direc- tion of the ophthalmic and mandibular branches of the ganglion and grow out farther as the ophthalmic and mandibular nerves; the former passes forward between the optic vesicle and the wall of the brain; the latter runs ventrally towards the angle of the mouth, over which it divides, a smaller maxillar}^ branch entering the maxillary process of the mandibular arch, and a larger one, the mandibular nerve, runs into the mandibular arch. (For an account of the branchial sense organ of the trigeminus, see Chap. VI.) A medullary component of the trigeminal nerve arises from the wall of the brain just median to the ganglionic root during the fourth day; it runs forward parallel to the ganglionic ophthal- mic branch, and sends a twig to the ciliary ganglion. Beyond this point it unites with the ganglionic branch. A connection of the trigeminus with the olfactory sensory epithelium is described under the olfactory nerve. 6. The sixth cranial or abducens nerve is stated to arise about the end of the fourth day. It is a purely motor nerve, and has no ganglion connected with it; it innervates the external rectus muscle of the eye. At 122 hours it arises by a number of slender roots attached to the myelencephalon near the mid- ventral line, beneath the seventh nerve. Its roots unite into a slender trunk that runs directly forward beneath the base of the brain to the region of the orbit. The sixth nerve thus corresponds more 268 THE DEVELOPMENT OF THE CHICK nearly than any other cranial nerve to a ventral spinal nerve- root. 7 and 8. The Facial and Auditory Nerves. The ganglia of these nerves at first form a common mass, the acustico-facialis. But during the course of the fourth day the anterior and ventral portion becomes distinctly separated from the remainder and forms the geniculate ganglion; the remainder then forming the auditory ganglia (cf. Fig. 102). The acustico-facialis ganglion complex moves from its original attachment to the dorsal surface of the brain and acquires a permanent root during the third day, attached ventrally just in front of the auditory sac. (a) The .seventh cranial or facialis nerve arises during the fourth day from the geniculate ganglion which is situated just above the second or hyomandibular branchial cleft. It grows first into the hyoid arch (posttrematic branch), but towards the end of the fourth day a small branch arises just above the cleft and arches over in front of it and runs down the posterior face of the mandibular arch (pretrematic branch). The origin of the motor components is not known. (6) The further history of the auditory nerve is considered with the development of the ear. 9. The ganglion of the ninth cranial or glossopharyngeal nerve (ganglion petrosum cf. Fig. 102) arises from the anterior part of the postotic cranial neural crest as already described. Early on the fourth day the ganglionic axones enter the base of the brain just behind the auditory sac and establish the root, which con- sists of four or five parts on each side. From the ganglion which is situated at the summit of the third visceral arch a strong peripheral branch develops on the fourth day, and extends into the same arch; a smaller anterior branch develops a little later which passes over the second visceral pouch and enters the second visceral arch. About the same time an anastomosis is formed with the ganglion of the vagus. 10. The tenth cranial or vagus (pneiunogastric) nerve is very large and complex. Its ganglion very early shows two divisions, one near the roots (ganglion jugulare) and the other above the fourth and fifth visceral arches (ganglion nodosum cf. Fig. 102). It arises by a large number of fine rootlets on each side of the hind-brain behind the glossopharyngeus, and the roots converge in a fan-like manner into the proximal ganglion; from here a stout THE NERVOUS SYSTEM 269 nerve passes ventrally and enters the ganglion nodosum situated above the fourth and fifth visceral arches. Branches pass from here into the fourth and fifth arches, and the main stem is con- tinued backward as the pneumogastric nerve s.s. From the hinder portion of the spreading roots a strong commissure is continued backward parallel to and near the base of the neural tube as far as the fifth somite; this is provided with three small ganglion-like swellings. This condition is found about the end of the fourth day. Later this commissure unites with the main sympathetic trunk, and part of the vagus ganglion separates from the remain- der as the ganglion cervicale primum of the sympathetic trunk. During the fifth and sixth days the main stem of the vagus grows farther back and innervates the heart, lungs, and stomach. Neuroblasts of the sympathetic system accompany the vagus in its growth, and form the various ganglion cells of the heart, and other organs innervated by the vagus. During the fifth and sixth days the ganglion nodosum, which originally lay at the hind end of the pharynx, is carried down with the retreat of the heart into the thorax, and on the eighth day it is situated at the base of the neck in close contact with the thymus gland. 11. The Eleventh Cranial or Spinal Accessory Nerve. No ob- servations on the development of this nerve in the chick are known to me. 12. The twelfth cranial or hypoglossus nerve appears on the fourth day as two pairs of ventral roots opposite the third and fourth mesoblastic somites; each root is formed, like the ventral roots of the spinal nerves, of several bundles that unite in a com- mon slender trunk; ganglia are lacking, as in the first and second cervical nerves. The roots of the hypoglossus are a direct con- tinuation of the series of ventral spinal roots, and as they are related to somitic muscle plates in the same way as the latter, there can be no doubt of their serial homology with ventral roots of spinal nerves. The first four mesoblastic somites are subse- quently incorporated in the occipital region of the skull, and thus the hypoglossus nerve becomes a cranial nerve. No nerves are formed in connection with the first and second mesoblastic somites. As the occipital region of the skull forms in the region of the occipital somites, two foramina are left on each side for exit of the roots of the hypoglossus (Figs. 150 and 244). 270 THE DEVELOPMENT OF THE CHICK During the fourth and fifth days the nerve grows back above the roof of the pharynx, then turns ventrally behind the last visceral pouch and forward in the floor of the pharynx. According to Chiarugi minute ganglia are formed in the second, third, and fourth somites: but they soon degenerate (fourth day) without forming nerves. CHAPTER IX ORGANS OF SPECIAL SENSE I. The Eye The development of the eye up to the stage of 36 somites has been ah-eady described. We shall now consider the subsequent changes in the following order: (1) optic cup, (2) vitreous body, (3) lens, (4) anterior chamber, cornea, iris, etc., (5) choroid and sclerotic, (6) the conjunctival sac and eyelids, (7) the choroid fis- sure and the optic nerve. 1. The optic cup at the stage of 36 somites is composed of two layers, an inner, thicker layer, known as the retinal layer, and an outer, thinner layer, known as the pigment layer; these are continuous with one another at the pupil and choroid fissure. The inner and outer layers come into contact first in the region of the fundus, and the cavity of the original optic vesicle is gradu- ally obliterated. The choroid fissure is in the ventral face of the optic cup; it is very narrow at this time, and opens distally into the pupil; centrally it ends at the junction of optic stalk and cup, not being continued on the stalk as it is in mammals (Fig. 157). The walls of the optic cup may be divided into a lenticular zone {pars lenticularis or pars caeca) and a retinal zone; the former includes the zone adjacent to the pupil, not sharply demarcated at first from the remainder or retinal zone, but later bounded dis- tinctly by the ora serrata. The retinal zone alone becomes the sensitive portion of the eye; the lenticular zone develops into the epithelium of the iris and ciliary processes. In the lenticular zone the inner and outer layers become actu- ally fused, but in the retinal zone they may always be separated; indeed, in most preparations they are separated by an actual space produced by unequal shrinkage. The differentiation of the lenticular from the retinal zone begins about the seventh day, when a marked difference in thick- 271 272 THE DEVELOPMENT OF THE CHICK ness appears. The transition from the thinner lenticular to the thicker retinal zone soon becomes rather sudden in the region of the future ora serrata. About the eighth or ninth day a further differentiation arises within the lenticular zone, marking off the regions of the iris and ciliary processes (Fig. 159). The region /^v-;.^ if; 157 158 Fig. 157. — Section through the eye of a chick embryo at the beginning of the fourth day of incubation. (After Froriep.) ch. Fis. 1., Lip of the choroid fissure. Di., Lateral wall of the diencephalon. Y, I", Distal and proximal walls of the lens, st., Optic stalk. Fig. 158. — Section of the distal portion of the eye of a chick, second half of the fifth day of incubation. (After Froriep.) 0. ep. int., Internal epithelium of the cornea. Corn, pr.. Cornea propria. Ect., Ectoderm, ep., Epidermis, ir., Iris, mes., Meso- derm, p.. Pigment layer of the optic cup. r., Retinal layer of the optic cup. of the iris is a narrow zone bounding the pupil in which the two la3'ers of the optic cup become blended so that pigment from the outer layer invades the inner layer; the epithelia are decidedly ORGANS OF SPECIAL SENSE 273 Fig. 159. — Frontal section of the eye of an eight day chick. ant. ch., Anterior chamber of the eye. ch., Choroid coat, cil., CiHary processes. Corn., Cornea. 1. e. 1., Lower eyehd. n. m., Nictitating mem- brane, olf.. Olfactory sac. op. n., Optic nerve, o. s., Ora serrata. p., Pigment layer of the optic cup. post, ch., Posterior chamber of the eye. ret., Retina, scl.. Sclerotic coat. scl. C, Sclerotic cartilage, u. e. 1., Upper eyelid. thinner than in the ciliary region. The mesenchyme overlying the iris early becomes condensed to form the stroma of the iris; the epithelia form the uvea of the developed iris (Fig. 159). The muscles of the iris (sphincter pupillse) are stated by 274 THE DEVELOPMENT OF THE CHICK Nussbaum, Szily, and Lewis to arise from epithelial buds of the pupillary margin and the adjacent portion of the pigment layer of the iris. The marginal buds (Fig. 160) begin to form during the seventh day, the more peripheral ones somewhat later; the former are less numerous and larger than the latter. The observations are well supported, and appear to leave no doubt that the specificity of the ectoderm cells of the iris are not fixed. According to Lewis the wandering pigmented cells of the ante- rior portion, at least, of the choroid also arise from the pigment layer of the optic cup. The ciliary processes begin to form from the ciliary region of the lenticular zone on the eighth day (Fig. 159) ; the epithelium ^Pft" Sph. B. Fig. 160. — Two sections of the pupillary margin of the eye of a chick of 13 days' incubation. A., X 260. B., 130. (After Lewis.) c. P., CiHary process. E. B., Epithelial bud. P., Margin of pupil, p. 1., Pigment layer of Iris. r. 1., Retinal layer of iris. 8ph., Bud for the forma- tion of the sphincter muscle of the iris, derived from the margin. Sph.', Sph.", Submarginal buds of the sphincter. becomes thrown into folds projecting towards the posterior cham- ber, the cavity of the folds being filled by the mesenchyme of the developing choroid coat. The muscles of the ciliary body de- velop from the mesenchyme of the processes, which acquire a connection with the lens through a special differentiation of the vitreous body, the zonula ciliaris (zonula Zinnii). In the retinal portion of the optic cup the inner layer forms the entire retina proper from the internal limiting membrane to the rods and cones inclusive. The outer layer forms the pig- ORGANS OF SPECIAL SENSE 275 ment layer of the retina. About the middle of the fourth day pigment begins to develop in the outer layer and extends through- out it, even to the distal portion of the optic-stalk at first (Ucke, '91). The histogenesis of the retina of the chick has been de- scribed by Weysse (1906). 2. The Vitreous Humor (Corpus Vitreum). Until compara- tively recently embryologists have adhered to the view stated by Schoeler (1848) and Kolliker (1861) that the vitreous body arises from mesenchymal cells that enter the eyeball through the choroid fissure. The fact that the embryonic vitreous humor of birds is almost entirely devoid of cells was a serious difficulty. The cells are in fact so scanty as to be absent in many entire sections. Moreover, in character thev resemble embrvonic blood-cells and not mesenchyme, and disappear entirely by the eighth day. It seems impossible that they should play any important part in the origin of the massive vitreous body. Re- searches of the last few years have demonstrated that the vitreous body is primarily of ectodermal origin, its fibers arising as processes of cells of the inner layer of the optic cup and the matrix as secretion. According to some the cells of the lens are responsible wholly (Lenhossek) or in part (S/ili) f