i t n ^mw^^^'^^hw.^'^ ^^^^' ■.J I dforncU Imttersita ffitbrary Stliata, Sfeiu Uork ..a-R.i.€.Er.e.at.. QM 601.K°2™" """"''"' '•"'™^ V.1 'Manual of human embryolog 3 1924 024 561 817 The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://archive.org/details/cu31924024561817 MANUAL OF Human Embryology WRITTEN BY Charles R. Bardeen, Madison, Wis.; Herbert M. Evans, Baltimore, Md.; Walter Felix, Zurich; Otto Grosser, Prague; Franz Keibel, Freiburg i. Br.; Frederic T. Lewis, Boston, Mass.; Warren H. Lewis, Baltimore, Md.; J. Playfair McMurrich, Toronto; Franklin P. Mall, Baltimore, Md.; Charles S. Minot, Boston, Mass.; Felix Pinkus, Berlin; Florence R. Sabin, Baltimore, Md ; George L. Streeter, Ann Arbor, Mich.; Julius Tandler, Vienna; Emil Zuckerkandl, Vienna. EDITED BY FRANZ KEIBEL and FRANKLIN P. MALL Professor in the University at Freiburg i. Br. Professor of Anatomy in the Johns Hopkins University, Baltimore, U.S.A. IN TWO VOLUMES VOLUME I With 423 Illustrations PHILADELPHIA & LONDON J. B. LIPPINCOTT COMPANY 1910 V. I A-' Copyright, 1910 Bt J. B. LiPPINCOTT COMPAiSrT A. M^' I '^^" Printed by J. B. Lippincott Company The Washington Square Press, Philadelphia, V. S. A. CONTENTS PAGE Preface ix Introduction xi-xviii CHAPTER I. By F. Keibel. The Germ Cells 1-17 CHAPTER II. By F. Keibel. Fertilization 18 CHAPTER III. 'By F. Keibel. Segmentation 19-20 CHAPTER IV. By F. Keibel. Young Human Ova and Embryos up to the Formation of the First Primitive Segment 21-42 CHAPTER V By F. Keibel. The Formation of the Germ Layers and the Gastrulation Problem . . . 43-58 CHAPTER VI. By F. Keibel. Summary of the Development of the Human Embryo and the Differ- entiation OF ITS External Form 59-90 CHAPTER VII. By O. Grosser. The Development of the Egg Membranes and the Placenta ; Menstru- ation • .' 91-179 I. Introduction 91 II. Menstruation 97 III. Observations on Young Ova 103 A. Review of the^ Descriptions of Young Ova observed in situ . . . 105 B. R^sum^ of the First Processes of Development up to the For- mation of the Villi and the Appearance of the Intervillous Space 116 vi CONTENTS. C. The Stages from the Appearance of the Villi to their Complete Formation 127 IV. The Formation of the Placenta; Relations of the Embryonic Mem- branes up to their Maturity 137 (a) Differentiation of the Chorion ; Chorion laeve, Decidua parietalis and capsularis 137 (b) The Placenta 143 V. The Mature Afterbirth; the Amnion, AUantois, and Yolk Sack up to Maturity 16» VI. The Uterus post partum 174 CHAPTER VIII. By F. p. Mall. Determination^ of the Age of Human Embryos and Fetuses 180-201 CHAPTER IX. By F. p. Mall. The Pathology of the Human Ovum 202-242 CHAPTER X. By F. Pinkus. The Development of the Integument 243-291 A. The Epidermis 243 (a) Early Stages 243 (b) Further Development 246 (c) Formation of the Stratum Corneum 249 (d) Granule Inclusions of the Epidermal Cells 251 B. The Corium 254 C. The Connection between the Corium and Epidermis 256 D. Dermal Ridges and Folds 257 (a) Dermal Ridges produced by Surface Growth; Growth Folds 257 (b) Tension Folds 260 E. The Metamerism of the Skin 261 F. The Hair 262 G. The Sudoriparous Glands 276 H. The Nails 281 CHAPTER XI. By C. R. Bahdeen. The Development of the Skeleton and op the Connective Tissues 292-453 Part I. The Histogenesis of the Connective Tissue.s 293 (a) The Early Mesodermic Syncytium 293 (b) Formation of the Mesodermic Somites 294 (e) The Axial Mesenchyme 296 (d) The Parietal and Visceral Layers of the Mesoderm 296- (e) The Mesenchyme of the Head 297 (f) The Origin of the Connective Tissues 298 Part II. The Morphogenesis of the Skeletal System 316 A. General Features o^r B. Origin and Fate of the Chorda Dorsalis 327 CONTENTS. vii C. Vertebral Column and Thorax 331 Regional Differentiation 341 D. Skeleton of the Limbs 366 Inferior Extremity 367 Superior Extremity 379 E. The Development of the Skull, the Hyoid Bone, and the Larynx . . 398 CHAPTER XII. By W. H. Lewis. The Development op the Musculab System 454-S22 Histogenesis 458 The Segmentation of the Mesoderm 469 The Differentiation of the Primitive Segments 470 The Muscles of the Trunk 473 The Muscles of the Perineum 478 The Ventrolateral Muscles of the Neck 480 The Musculature of the Extremities 482 The Muscles of the Head 505 CHAPTER XIII. By F. p. Mall. The Development of the Coelom and Diaphragm 523-548 The Body Cavities 525 The Septum Transversum 530 The Separation of the Pericardial, Pleural, and Peritoneal Cavities 535 PREFACE The circumstances tliat led to the publication of tMs Manual of Human Embryology, of wliicli the present is the first volume, have been recorded in the Introductory Chapter, where the aims of the work are also set forth. A number of German and Amer- ican embryologists have collaborated in its production, and the work appears simultaneously in Grermany and America. The trans- lation of the chapters originally written in Grerman has been made by Professor McMurrich, and the publishers desire to express their thanks to him for his careful work. The English chapters have been translated for the German edition by one of the editors, Professor Keibel. Valuable assistance has been rendered in the correction of the proofs by Dr. C. Elze, and to him also the thanks of the publishers are due. The editors desire to express their in- debtedness to the publishers, Messrs. S. Hirzel, of Leipzig, and J. B. Lippincott Company, of Philadelphia, for their generosity in making it possible for the collaborators to enrich the text with numerous excellent illustrations and for the aid they have ren- dered in bringing the work to a successful completion. The second volume will appear at an early date. Fkanz Keibel, Fkanklin p. Mall. INTRODUCTION. Bt FEANZ KEIBEL, Feeibueg i. Br. Vesal has rightly been regarded as the founder of human anatomy. He emancipated anatomy from the dogmas of Galen- ism and showed that Galen's anatomical observations had really been made upon apes and that he had attributed to the human body the structure observed in these forms. Vesal studied the human body and based his immortal work upon that study. Embryology was, however, only incidentally considered by him, and in his stud- ies in this field he was false to his principles in that, like Galen in anatomy, he attributed to man what he observed in the embryos of animals.^ However, the path which Vesal opened in anatomy was also followed by his successors in embryology. In this connection Gabriele Fallopia (1523-1562) deserves mention as the first who gave a correct description of the placenta and of the chorion and its vessels, and also, on the basis of his own observations, denied the occurrence of an allantois in man. Bartholomeo Eustachi (d. 1574) studied the development of the teeth in human embryos, and Julio Caesare Aranzi (1530-1588) expressly noted that differences exist between the early stages of development of man and those of the lower animals. Vesal 's successor and pupil, Matteo Realdo Colombo (d. 1559), endeavored to do for human embryology what Vesal had done for human anatomy, promising a consideration of human embryology and not animal embryology, since Nature had formed man and animals upon different plans. His efforts, how- ever, to establish an embryology on the basis of observations on human material must, as Bloch has pointed out, be regarded as a failure. And the reasons why Colombo was able to record only exceedingly incomplete observations are not far to seek. The num- ber of human embryos accessible to a single investigator in his day must have been very small ; in addition they were, in part, exceed- ingly altered pathologically, and, what was still more important, the phenomena of greatest significance for embryology occur at such an early period of development that they could not possibly ' Those interested in the history of Embryology are referred to the paper of Bruno Bloch, Die gesehichtliehen Grundlagen der Embryologie bis auf Harvey, Nova Acta Leop.-Carol. Akademie, No. 3, 1904, p. 295 et seq., a work written under the stimulus and influence of Rudolf Burckhardt. sii INTKODUCTION. be observed with the methods of investigation available at that time. Even in the case of animals these phenomena were still alto- gether obscure, and almost three hundred years were to elapse before the necessary preliminary work in tliis field was accom- plished. Ulisse Aldrovandi (1526-1605) must first lay a founda- tion for modern embryology. He was the first to trace systemati- cally the development of the chick up to the time of hatching and to give a consecutive account of the development. After Aldrovandi followed his pupil Volcher Koyter (1534-1600), and then Hieronymus Fabricius ab Aquapendente (1537-1619). Har- vey and Malpighi need be only mentioned here, but a few words mav be devoted to the less known Adrian van den Spieghel (Spi- gelius) (1578-1625). Spieghel, as may be seen from the introduction to his work De formate fetu, published the year after his death (1626), had a clear appreciation of the value of embryology to the physician, -and while bis predecessors studied especially the fetal accessory organs, he turned his attention mainly to the development of the fetus itself and that of its organs. Since his object was a history of the development of the human embryo, he was met by the diffi- culties mentioned above, but where these difficulties were less pro- nounced, as in the history of the development of the bones, he made valuable contributions and materially advanced the investi- gations of Koyter and Fallopia in this field. In this connection he made the first attempt at a description based on observation of thei genesis of a tissue, distingtiishing those osseous elements which were formed in preformed cartilage from those which arose in membrane and descri])ing the growth of membrane bones by the apposition of osseous spicules at the margins. What Spieghel had to say concerning the development of the other organs of the body does not compare favorably with his description of the de- velopment of the bones, and he also devoted more than half the con- tents of his book to the consideration of the chorion, the placenta, and the umbilical vessels ; he recognized the occurrence of an allan- tois in man. And Spieghel also rendered important service by his observations on the fetal circulation. In his time the teaching of Galen that the direction of the blood-stream was the same in both the umbilical arteries and veins was generally accepted, and with tliis false belief it was naturally impossible to obtain a correct understanding of the fetal circulation. Spieghel, for a number of good reasons, concluded that the direction of the flow in the umbili- cal arteries was centrifugal; he believed, however, that the vessels carried not blood but Galen's spiritus vi talis, and this prevented him from obtaining a clear idea of the true relations and an accurate knowledge of the circulation of the blood. The difficulties which stood in the way of this discovery were exceedingly great. INTRODUCTION. xiii and the service rendered by Harvey, as Bloch points out {l. c, p. 333j note), can be rightly appreciated only when, from a study of the literature, an idea is obtained of how confused were the notions of even the most distinguislied physiologists of the time concern- ing the movements of tlie blood. Without entering into details the names may be mentioned here of (lualtherus Needham, Nicolaus Hoboken, Nicolaus Steno, and Thomas Wharton, who rendered services in connection with the anatomy and physiology of the egg-membranes. The discoveries of Eegner de Graaf, Swammerdam, Hamm, and Leeu- wejihoek are generally known and have many times been recounted ; and the much-admired ''Icones Ossium Foetus Humani" of Albinus gave a certain amount of completeness to the knowledge of the development of the human skeleton. Albrecht von Haller and Kaspar Friedrich Wolit are worthy of mention for more than the foimulation of the theories of evolution and epigenesis. Haller wrote important works on the development of the osseous system and of the heart, and Kaspar Friedrich Wolff was the founder of the theory of the germ layers, which was further developed by Dollinger's pupil, Christian Pander, and especially by Karl Ernst von Baer and Remak. Karl Ernst von Baer finally discovered the long-sought-for ovum of man and of the mammalia, and is the real founder of comparative embryology. His work, "Ueber Entwicklungsgeschichte der Tiere, Beobachtung und Eeflexion" (Konigsberg, 1828 and 1837), must be considered, according to Kol- liker ("Entwicklungsgeschichte," 1879, p. 14), "the best that em- bryological literature of all times and all peoples has to show." The origin of the germ layers and the origin of the organs from these were now subjected to most careful investigation on all sides and in all classes of animals. Here Eemak's work, which exercised a very deep and lasting influence, need only be mentioned. The progress in human embryology did not, however, at first keep pace with that in comparative embryology. William Hunter's "Anato- mia Uteri Gravidi" (Birmingham, 1774) gave splendid represen- tations of the egg-membranes and of the gravid uterus, but did not advance our actual knowledge of embryology, the condition of which at the close of the eighteenth century can be well under- stood from the book by D. Ferdinand Georg Danz ("Grundriss der Zergliederungskunde des ungebornen Kindes," vol. i, Frank- fiirt and Leipzig, 1792; vol. ii, Giessen, 1793), which was accom- panied by annotations by Sommerring, the first authority of that period. Sommerring himself published in 1799 his "Icones Em- bryonum Humanorum," which does not, however, contain satis- factory representations of even the later stages of human develop- ment, although it is interesting as citing all important earlier observations. How small was the amount of useful material of xiv INTEODUCTION. the early stages of human development accessible even to Karl E. von Baer can readily be appreciated if one examines the con- cluding part of his great work, edited by Stieda in 1888. In addi- tion the methods of satisfactory fixation, of sectioning, to say noth- ing of serial sections and of reconstruction, were still lacking. Consequently, the most important early stages in the development of man remained unknown, and progress even in the case of ani- mals, of which abundant material was available and could be stud- ied under the lens while still fresh and transparent, was slow and difficult, even when the method of fixing the embryos in alcohol and dissecting them under the lens with fine needles and knives was learned. Indeed, it is remarkable what was accomplished in spite of these arduous and uncertain methods, but results could be ob- tained only by oft-repeated observations, and, in the case of man, the material for these was wanting. Thus, we read in Friedrich Tiedemann's "Anatomic und Bildungsgeschichte des Gehirns im Fetus des Menschen" (Niirnberg, 1816) that in embryos of the first month the place of the spinal cord and brain is occupied by a clear fluid. His observations begin with an embryo which had been preserved for some time in alcohol and measured seven lines in length (from the figure it is to be concluded that Ehenish lines of 2.18 mm. are meant; the Parisian line corresponds to 2.25 mm.), that is to say, about 15 mm. Johannes Miiller in his "Bildungs- geschichte der Grenitalien" (Diisseldorf, 1830) begins his observa- tions on human embryos with an embryo of 8 lines ^ in length (measurement of tlie figure gives a length of 20 mm.). What the younger stages were like could only be conjectured from observa- tions on animals, and so matters practically remained until His published his "Anatomie menschlicher Embryonen." It is true that human embryos and the egg membranes were extensively investigated and to a certain extent very well figured at this time — as in Coste's " Embryogenie " (Paris, 1837), in the works of Pockel (Isis, 1825), Seller ("Die Gebarmutter und das Ei des Menschen," Dresden, 1831), Breschet ("Etudes anatomiques sur I'reuf humain," Paris, 1832), Th. L. W. Bischoff ("Beitrage zur Lehre von den Eiliiillen des menschlichen Fetus," Bonn, 1831), and in the briefer publications of E. H. Weber, Joh. Miiller, E. Wagner, Von Baer, Wharton Jones, Allen Thomson, and Esch- richt— but the observations remained isolated, they were not suffi- cient to round out the whole story, and frequently the results failed to attain completeness owing to a disinclination to destroy a valu- ""'A wonderful human embryo, 3i lines in len^h measured along its curva- ture, with a long-stalked umbilical vesicle and traces of branchial clefts," he could not sa^riflee for investigation ; he described it in Meckel's Archiv, 1830. He also refers to a similar embryo in Meckel's possession (Beitrage zur vergl Anat vol. i, pp. 71 and 72). '' INTRODUCTION. xv able specimen. The technic of investigation made little progress, and it is in this particular that His made a change. He had pre- pared himself by a study of the embryology of the chick and in connection with this study had worked out a successful technic; he fixed his embryos, he had constructed an apparatus for section cutting which, primitive as it now seems, led the way to our modem microtomes, and he had thought out a method of recon- struction which has been gradually improved, especially by Born, into a method that is now absolutely indispensable in embryological investigation. With this equipment His began the study of human embryology. Although he was fortunate in obtaining material, yet this was still scanty, and he was far from imagining that with it he could study the complete development of man ; but with the help of the method of reconstruction he first thoroughly worked out the anatomy of the individual embryos so far as his technic permitted, and the results so obtained formed a sure foundation for human embryology. His models, as well as those he had con- structed from the chick, were reproduced by Dr. Ziegler in Frei- burg, and, since these came to be regarded as indispensable by every anatomical institute, thej" served, more than anything else, to increase the knowledge and interest in human embryology, gradually the gaps which existed were filled in by His's own continued observations and by those of others who studied their material by similar methods, and in this connection there may be mentioned the work of Fol, of Phisalix, of Eternod, and, espe- cially, of Graf Spee. Monographs of individual embryos were contributed by Piper, Mall, Frederic T. Lewis, Susanna Phelps Oage, and Bremer, and in connection with the "Normentafel zur Entwicklungsgeschichte des Menschen" of Keibel and Elze, to be considered later, by Thompson, Ingalls, Elze, and Low. In the meantime the investigation of the individual organ-systems was not standing still. It was very natural that the foundation stones upon which certain systems of organs are built up should be brought together sooner than those of other systems, and, this happening. His directed his attention to these systems, although the development of the organism as a whole still remained the chief object of his thoughts. Thus were achieved the results re- corded in the third part of his "Anatomie menschlicher Embry- onen,'* of which those treating of the general differentiation of the digestive tract and those concerning the heart may be especially mentioned. He also published other larger works on the develop- ment of the nervous system, one of which appeared shortly before his death. There should also be mentioned in this connection the work of A. von Kolliker (eye and olfactory organ); of Hammar (the branchial arch region) ; of Broman, Mall, and Swaen (coelom and xvi INTRODUCTION. diaphragm) ; of Hochstetter, Tandler, Erik Miiller, and de Vriese (vascular system; also Elze) ; of Bardeen (skeleton); of W. H. Lewis (musculature) ; of Nagel, Keibel, and Bayer (the urogenital system) ; of the younger His, Hammar, and S'tr-eeter' (iear) ; and of the younger His and Romberg (the sympathetic system). The above enumeration makes no pretence of completeness, but is merely intended to show how human embryology was ap- proached from all sides and how the material for a full exposition of the subject was gradually accumulated. As has been stated, the idea of working out a complete account of the development of the human body was always before the mind of His, but as time went on the hope of accomplishing the task single-handed failed him; and so he suggested that I should col- laborate with him: in writing a text-book on human embryologj^ Unfortunately, the plan was never carried out ; I have, however, gone further with it and prepared the way for its fulfilment by another work. In the pages of my "Normentafel zur Entwick- lungsgeschichte der Wirbeltiere" I have published, in conjunction with Curt Elze, a Normentafel for the development of man. In this I have had the assistance of a large number of other investi- gators, some of whom, such as Hammar and Tandler, have made personal contributions, while others have contributed valuable material, frequently material obtained by operation and most admirably preserved. Thus I have been able to follow the develop- ment of the body in an almost perfect series of embiyos from about the twelfth day ^ up to the end of the second month and to obtain many data concerning the first appearance and degree of development of the organs. An explanation of the stages of de- velopment earlier than those that I examined had been furnished by the admirable investigations of Graf Spee and Hubert -Peters,* already mentioned, and the time seemed to me to be propitious for giving an account of the development of the human body, based throughout on human material. We have already, it is true, a whole series of text-books on human embryology, some of them ex- cellent; but they are based, for the most part, on other than human material. They have been written from the comparative embryo- ° According to the age estimate visually employed; if the estimates of Bryce and Teacher, to which I shall refer later, are correct, the above statement should be "from about the eighteenth day." ( Compare Chapter VIII, The Age of Human Embryos and Fetuses.) •Quite recently the observations of Spee and Peters have been confirmed and extended by Ph. Jung ("Beitrag zur friihesten Eieinbettung beim menseh- lichen Weibe," Berlin, 1908) ; Bi-yee. Teacher, and Kerr ("Contribution to the Study of the Early Development and Imbedding of the Human Ovum," Glasgow, 1908) ; and Erassi ("Ueber ein junges menschliehes Ei in Situ," Arch. f. mikr! Anat., vol. Ixx, 1907, and Weitere Ergebnisse des Studiums eines jmigen mensch- liehes Eies in Situ, ibid., vol. Ixsi, 1908). INTRODUCTION. xvii logical stand-point, and, in the interests of continuity, an endeavor to conceal the gaps in our knowledge is frequently evident. This -will not be the case in the present book. On the contrary, its endeavor will be to indicate clearly and precisely these gaps, for thus only can they be filled in; and frequently, as I know, the material for filling them in is already available. I do not propose, of course, to dispense with the assistance of comparative embry- ology and anatomy, but when a gap has been indicated the manner in which it is probably to be bridged will be pointed out, and, similarly, attention will be given to those facts of comparative embryology and anatomy which serve to render intelligible to us special processes of development in man. As a rule, however, shell considerations will be printed in smaller type, in ofder that there may be a clear distinction between facts and deductions. So far as I can see, all the material necessary for a human embryology, with the exception of the earlier stages which concern the develop- ment of the germ layers and the first stages of placentation, is already available or at least is comparatively readily obtainable. It would be possible, therefore, to describe the development of all the organs exclusively from human material if it were possible to bring together all the material which is now available. If my colleagues and myself have not succeeded in this, nevertheless we have been able to approach the goal, and we may hope that it will fare better with us than with Realdo Colombo, and even if the first attempt has not been perfectly successful, the second, whether made by us or by others, will come so much nearer the goal. The material is in hand and to be obtained ; it will be brought forward if we only point out clearly what is lacking. When working at my Normentafel I had very striking evidence of the community of purpose which to-day inspires our scientific world. A considerable number of investigators deprived them- selves for long periods of time of valuable material which they themselves had not yet had opportunity to study thoroughly, in order that a larger undertaking might be completed. In the time of Kealdo Colombo such scientific co-operation did not exist; we can with pride regard this as a great step in advance. Without it the completion of such a work as the Normentafel would have been unthinkable; to-day ^he post and the telegraph, railways and steamboats, are at our service and may be turned to the ser- vice of science. And so it has been possible to bring to completion such a "Handbook of Human Embryology" as His had planned with me in Germany with the assistance of one of his most loyal pupils on the far side of the ocean, Franklin P. Mall, of the Johns Hopkins Medical School in Baltimore. It has seemed proper to Mall and myself to enlist the services of a considerable number of collaborators, the necessary similarity of treatment being secured xviii INTRODUCTION. by the common purpose and by editorial supervision ; in this way the book would the sooner be brought to completion and, what is more important, the various chapters would be written with a complete mastery of the subject. Considered purely objectively and scientifically the embryology of man has no more interest than that of any other mammal or vertebrate, and from this stand-point special endeavors to work out human embryology by itself may seem to be misplaced. Our interest in the development of the human bodj^ is justified, however, first, by the fact that we are human beings, and, secondly, because, as Spieghel long ago pointed out, it is of importance to phj'sicians. That we do not undervalue the information to be obtained from comparative embryology and anatomy has already been stated. HANDBOOK of HUMAN EMBRYOLOGY I. THE GERM-CELLS. By FRANZ KEIBEL, Freibueg i. Be. The germ-cells of man are the ovum (oide, ovium, mature ovum), formed in the ovary, and the spermium, formed in the testes. The mature human ovum (the oide of Korschelt and Heider, the ovium of Waldeyer) is not yet known, nor have the processes which bring about its maturation yet been certainly observed. NageP (1888) in connection with his PI. XXI, Fig. 7, speaks of remains of polar bodies, but this interpretation, as Waldeyer (1902, in Hertwig's Handbuch, vol. i, part i, p. 333) has pointed out, cannot be accepted in view of the occurrence in the ovum figured of a perfectly unaltered nucleus. The human ovum which has reached its full size in the ovary is a true cell, with cytoplasm, in the ovum termed ooplasm (yolk), a nucleus, frequently spoken of as the germinal vesicle, and a nucleolus, also termed the germinal spot. In the nucleus there is, in addition to the nucleolus, a fine, somewhat scanty nuclear reticulum containing chromatin. We must assvmie that processes of maturation occur in such an ovum, similar to those which occur in the ova of other animals. Briefly stated the maturation consists in that from the ovum, by two rapidly succeeding divisions, four cells, one large and three small, are formed. The large oell is the mature ovum, the oide or ovium, the three other cells are the polar bodies (polocytes), formerly known as the directive corpuscles. At the first division the first polar body is separated, at the second division the second one, the first one at the same time also undergoing division. The divisions take place by mitosis and the final products possess only half the number of chromosomes characteristic of the ordinary cell-divisions of the species. A reduction of the number of chromosomes by one-half is thus brought about, but how this reduction is accomplished and what may be its significance is still a matter of discussion ; indeed, whether the reduction always actually takes place during the nuclear divisions which produce the polocytes — whether at the one or the other of these divisions there is really a reduction division — is yet uncertain. For further 'W. Nagel: Das mensehliche Ei, Arch. f. mikr. Anat., vol. xxxi, 1888. 1 2 HUMAN EMBRYOLOGY. consideration of this point reference may be had to Korsehelt and Heider/ Haecker,' Waldeyer/ and E. B. Wilson." Individually many variations occur m the process; the division of the first polocyte may not take place, the formation of one of the polar bodies may be suppressed, and, what is noteworthy, modifications occur m individuals of one and the same spegies. The centrosome of the egg-cell seems usually to degenerate duruig these processes. The time, with reference to the fertilization, at which the formation of the polar bodies takes place, varies; the first one is often formed while the ovum_ is still in the ovary. In mammals these phenomena have been especially well studied in the mouse (Sobotta," Leo Gerlach,' and others), and they have also been investi- gated in the guinea-pig (Rubaschkdn). In man, as has been stated, nothmg is known of these things. We may with perfect certainty assume that polocytes are formed and that there is a reduction of the chromosomes to one-half the typical number; but whether variations from the general plan occur, whether or not the formation of one of the polar bodies is suppressed, we do not know. Similarly we can say nothing as to the time at which the polocytes are formed. It may be that the first one is formed while the ovum is still in the ovary, and observations on this point are much to be desired. But even although, as is clear from what has just been said, the mature human ovum is unknown, nevertheless descriptions have frequently been given of a so-called mature human ovum, that is to say, of an ovum which was near maturity, and usually the figure and description given by Nagel has formed the basis of these accounts. 0. Hertwig follows Nagel's description and Waldeyer quotes it almost verbally; it may be given here as well as the figure. The human ovum retains its transparency in all stages of development, so that all its anatomical characteristics can be fully made out in the living object. The cell substance, which may be termed the ooplasma, and which is frequently spoken of as the yolk, is separated into two layers. In the inner (central) layer is found the greater part of the deutoplasm, that is to say, the inclusions of the ovum, which are usually regarded as nutritive or reserve substances; they produce merely a slight opacity in comparison with what is found in the ova of other mammals. The deutoplasm consists partly of feebly and partly of strongly refractive, finer and coarser granules; but a distinct delimitation of the deuto- plasmic elements, such as one finds not only in the ova of the lower vertebrates but also in those of many mammals, cannot be made out. The outer layer, the marginal zone of the ooplasma, is much ' Korsehelt and Heider : Lehrbuch der vergleichenden Entwieklungsgeschichte, 1902. " Haecker : Die Chromosomen als angenommene Vererbungstriiger, Ergeb- nisse und Fortschritte der Zoologie, herausgegeben von J. W. Sprengel, vol. i, 1907. * Waldeyer: Chapter I in 0. Hertwig's Handbueh der vergleichenden und experimentellen Entwieklungslehre, 1906. °E. B. Wilson: The Cell in Development and Inheritance, New York, 1906. °J. Sobotta: Die Bildung der Richtungskorper im Ei der Maus, Anatom. Hefte, evi, 1907. The remaining literature is cited here. ' L. Gerlach : Ueber die Bildung der Richtungskorper bei Mus musculus Wies- baden, 1906. THE GERM-CELLS. 3 more finely granular and transparent. It contains the germinal vesicle, that is to say, the nucleus of the cell, in which is to he seen a large germinal spot or nucleolus. In ova examined while fresh in the liquor folliculi Nagel observed amoeboid movements in the nucleolus, and such movements have also been described as occur- ring in the nucleoli of other ova, although Flemming ("Zellsub- stanz, Kern, und Zellteilung," Leipzig, 1882, p. 157), to whom Nage] ("Die weibliche Geschlechtsorgane," in K. von Bardeleben's Fig. 1, — A fresh ovum from an ovarian follicle of a woman of thirty years. The side of the yolk upon which is the nucleus is towards the observer, so that one looks directly down upon the nucleus and this rests upon the deutoplasm. (From W. Nagel: Arch. f. mikr. Anat., vol. xxxi, 1888, PI. XX, Fig. 5.) "Handbuch der Anatomic des Menschen," vol. vii, p. 59, 1896) refers in connection with the egg of Paludina, expressly states that in spite of many endeavors he had not succeeded in perceiving changes of form in living nucleoli. Further observations on this point are much needed. The nucleus appears to be homogeneous in the freshly stained ovum, but with proper staining a scanty chromatin network becomes visible. The membrane which encloses the ovum, the zona pellucida, is remarkably broad and is finely striated radially. It is separated from the ooplasma by a narrow perivitelline space and is surrounded by cells, derived from the cumulus ovigerus which surrounds the ovum in the ovarian follicle. 4 HUMAN EMBRYOLOGY. These cells form three or four layers, those of the layer nearest to the zona pellucida having a markedly radiating arrangement and forming in the fully formed ovum the corona radiata of Bischoff. This author regarded the corona as an indication of the maturity of the ovum, a view with which Waldeyer coincides to the extent that he regards a well-developed corona as a sign of approaching maturity, without, however, acknowledging it to be an indication of complete ripeness. In this he agrees with Van Beneden and Nagel. To this description of the approximately ripe ovum it must be added that the existence of a perivitelline space is in dispute. Nagel believes that the space is of importance in that it permits a rotation of the ovum so that the nucleus usually lies upon its upper surface, for this was the position in which he found it in all fresh, approximately ripe ova. Ebner (KoUiker's "Handbuch der Gewebelehre, " vol. iii, p. 517, 1902) disputes both the existence of a perivitelline space and the assumption that the ovum rotates within the zona pellucida, the nucleus thereby coming to lie upon the upper surface. He is of the opinion that the nucleus ascends through the almost fluid ooplasma on account of its lesser specific gravity, and that the ovum as a whole does not rotate; he bases his opinion on the fact that when a fresh ovum is burst the greater part of the ooplasma together with the nucleus is expelled, but the apparently denser marginal zone of the ooplasma always remains adherent to the zona pellucida. This could not occur if an interval filled with fluid was interposed between the marginal ooplasma and the zona. Furthermore it could be observed, by carefully focus- sing an equatorial optical section of an uninjured fresh ovum, that the marginal ooplasma was in intimate contact with the inner surface of the zona pellucida. "What Nagel figures as a cleft is a line of refraction, which, with deep focussing, appears to be in the deeper layer of the yolk (ooplasma), and apparently sepa- rates the yolk (ooplasma) and the zona, but is really a purely optical phenomenon which depends on the cur\'ature of the zona and resembles the line of refraction which one may observe under similar circumstances in cartilage cavities." Waldeyer (1. c, p. 332), on the other hand, mantains the cor- rectness of Nagel 's observation, since he has been able to convince himself of it from the ovum figured by Nagel. He is not, however, convinced as to the rotation of the ovum within the zona, but agrees with Ebner that the fact observed by Nagel may be explained by the ascent of the lighter nucleus through the almost fluid ooplasm. He himself shows in his Fig. 134 (l. c), which is reproduced here as Fig. 2, an almost ripe human oVum (a fully grown, oocyte), taken from an ovary while it was still warm, in which there was not the slightest indication of a perivitelline space. Waldeyer explains this discrepancy by assuming that the ovum described by Nagel THE GERM-CELLS. 5 was further advanced towards maturity. The ooplasm is also arranged somewhat differently in this ovnm from what it is ia that described by Nagel. Close under the zona there is a very narrow zone of a finely granular material, the ooplasm cortex, which was clearly seen by Waldeyer as well as by Ebner; beneath this is a broader, clearer zone of ooplasm in which the nucleus lies in almost ripe ova, and then comes the central darker mass of the ooplasm. Fig. 2. — AJmost mature human ovum (fully grown oocyte) taken from an ovary while still warm. On the outside appears the epithelium, with the clear zona pellucida ; within this, a broad stratum of ooplasm with little yolk; and in the centre the ooplasm rich in yolk. Above to the left, is the nucleus, with nucleolus: and a few subzonal nuclei are seen. X500. (After Waldeyer in Hertwig's Handbuch, vol. i, part i, p. 330, Fig. 134.) It will be seen that in this connection important questions still await decision, and if gyn?ecologists and embryologists work with a common purpose a solution of them may be expected. Another question I regard as already settled. The human ovum has no micropyle, that is to say, no preformed opening in the zona pel- lucida for the entrance of the spermium. The question as to the 6 HUMAN EilBETOLOGY. existence of a micropyle has been recently revived by HoU {Anat. Anz., 1891, p. 554; and Sh. K. Acad. Wien., vol. cii, 1893), after it had long been regarded as settled. It seems to me to be certain that Holl has been the victim of a mistake. Ebner (KoUiker's "Gewebelehre," vol. iii, p. 518) says "HoU's figure of a section of an apparently degenerated human ovum shows, traversing the greatly shrunken zona, a small oblique cleft, probably formed by accident, perhaps by a wandering cell. True micropyles, so far as known, traverse the zona radially, not obliquely." The development of the ovum will not be discussed here, but in the chapter on the urogenital apparatus; yet, since it has to do with the question of the relation of the follicle cells to the ooplasm, it must be noted here that the mode of formation of the zona pellu- cida is not yet certain. Waldeyer inclines to the opinion that it is a product of the ooplasm and is therefore a primary egg-m.em- brane. E. Hertwig (O. Hertwig's "Handbuch") and others re- gard it as a chorion, that is, as a secondary egg-membrane, a product of the follicle cells ; while for others it is a double product of the follicle cells and the ooplasm, at least so Waldeyer interprets the observations of Eetzius,* Flemming,^ and Ebner.^" According to these authors there is first formed, from processes extending from the follicle cells to the ooplasm, a delicate network, which rests closely upon the surface of the ovum. The network is the first indication of the zona, aud, later, a homogeneous substance appears between its fibres and forms with them the zona. Whence this homogeneous substance arises is uncertain; according to Waldeyer it may come from the ooplasm. A portion of the cell bridges that originally extend between the follicle cells and the oijplasm are retained in the forming zona substance in a proto- plasmic condition, but whether such unions persist in the approxi- ma^tely ripe ovum is uncertain ; certainly their existence can hardly be reconciled with the presence of a perivitelline space. Kolliker (according to Von Ebner in KoUiker's "Handbuch der G-ewebe- lehre," vol. iii, p. 511) gives the diameter of the approximately ripe ovum as 0.22-0.32 mm., but Waldeyer remarks {I. c, p. 352, note) that he has never seen a human ovum of over 0.25 mm. As regards further measurements Ebner gives the diameter of the nucleus (germinal vesicle) as 30-45 /x, that of the nucleolus (ger- minal spot) as 7-10 M, that of the zona pellucida as 10-11 /*, and that of the deutoplasm granules (yolk granules. Von Ebner) as 2-3 /i. ' Retzius : Zur Kenntnis vom Ban des Eierstockeies und des Graaf 'schen Follikels, Hygiea Festband, Stockholm, 1889. "Flemming: Zellsubstanz, Kern und Zellteihrag, Leipzig, 1882, p. 35. '°Von Ebner: Ueber das Verbalten der zona pellucida zum Ei, Anat Anz vol. xviii, 1900. ' '' THE aErai-CELLS. 7 The production of ova begins at a very early period of life in the human species. According to Waldeyer (1. c, p. 373; and "Eierstock und Ei," 1870) at birth or shortly thereafter all the oogonia have become oocytes of the first order, and so have before them only further growth and maturation. (The contrary opinion of Paladino " I do not consider well founded.) Already in the ovary of the child the ova may approach ripeness (see C. Hennig; Ueber friihreife Eibildung, Sh. d. Leipzig. Naturf. Ges., p. 5, 1878). Also Waldeyer says, "One finds in the ovaries of newly-born and young children follicles the size of a pea with normally devel- oped ova." On the other hand, those ova which ripen only after the cessation of ovulation require for their development about fifty years. Into the broad field of the pathology of the ovum I cannot enter here; never- theless, follicles with several ova, multinucleated ova, and the fragmentation of the ovum may be briefly mentioned. Follicles with several ova may be explained (Schottlander : Arch. f. mikr. Anat., vol. xli, 1893; M. and P. Bouin; C. R. Soo. Biol., vol. lii, p. 17 and 18, Paris, 1900 [dog] ; Ch. Honore : Areh. de Biol., vol. xvii, p. 489^97, 1900 [rabbit] ) by supposing that the different ova of an egg mass in the embryo and child have not completely separated, so that several ova have be- come enclosed within a common follicle wall. They may very naturally tend to the production of twin pregnancies. They may also be supposed to have arisen by the fusion of originally distinct follicles. Multinucleated ova have been accounted for in various ways and perhaps have various methods of origin. They may be formed by direct nuclear division ( Stockel : Arch. f. mikr. Anat., vol. liii, 1899 ; Falcone: Monitore zool. ital., Suppl., 1899) or one may suppose that originally distinct ova have subsequently fused (H. Rabl : Mehrkemige Eizellen und mehreiige Follikel, Arch. f. mikr. Anat., vol. liv, 1899 ; S. von Schumacher und C. Schwarz : Anat. Anz., vol. xviii, 1900). Finally, they may be produced by the division of the nucleus of an oogonium, without the corresponding division of the cytoplasm taking place, as sometimes occurs in spermatogonia. Cases in which several nuclei occur in an ovum as the result of an immigration of leucocytes need not be considered here. In mammals a division or fragmentation of ripe ova after the expulsion of polar bodies has been observed (Henneguy, Janosik, H. Rabl, Gurwitsch, Van der Stricht), and I have also seen such a condition in human ova. The phenomenon is one leading to the degeneration of the ovum; some authors have compared it with segmentation and have seen in it a parthenogenetic process, but Bonnet ^^ has disposed of such notions. We may now turn to a consideration of the male germ-cell, the spermium, which is formed in the testis. It is well known that for a considerable time it was uncertain whether the male cells were " Paladino : La renovazione del parenehima ovarico nella donna, Atti dell' XI Congr. internaz. med. di Roma, vol. ii, Anatomia, 1894, p. 19. Compare also Arch. ital. de Biologic, vol. xxi, p. 15, 1894; and Monitore zoolog. ital., Anno V, p. 140, 1894; also, Per il tipo di struttura dell' ovaja, Rendic. Acad. Sc. fis. math., Napoli, vol. iii, p. 232, 1897; also, Sur le type de structure de I'ovaire, Arch. ital. de Biol., vol. xxix, p. 139, 1898. ^^ Bonnet : Gibt es bei Wirbeltieren Parthenogenesis? Ergebnisse d. Anatomic und Entwicklungsgeschichte, vol. ix, 1900. 8 HUMAN EMBRYOLOGY. not parasites in the seminal fluid— the name spermatozoa is a re- minder of this idea. The development of the spermium first clearly showed that this structure is nothing else than a modified cell. Spermatogenesis has also been studied in man and some figures from Meves^s (^^f./,. f_ mikr. Anat, vol. liv, p. 378) may be reproduced here. Other figures have been given by Ebner (KoUiker's "Handbuch," vol. iii, p. 454). In the study of human spermatogenesis attempts have been frequently made to deter- mine the number of the chromosomes. Dues- berg," who also cites the literature bearing on 9 k Fig. 4. — A typical human spermium, straightened out. In a the head is seen from the surface and in b from the edge. In both figures there are shown the head cap, the neck piece with the centro- somes lying close to the head, the connecting piece, and the princi- pal and end pieces of the tail. In a is shown in the anterior part of the head a darlc dot. (After G.Retzius, Biol. Untersuchungen, neue Folge, vol. X, PI. 15, Figs. 1 and 2.) Fig. 3, — Four stages in the spermatogenesis of man. (After Meves, Arch. f. mikr. Anat., vol. liv, p. 378, 1899.) the question, finds that in the spermatocytes there are in all probability twelve. If this be correct in the spermatogonia and the soma cells there would be twenty-four, as Flemming had already (1898) supposed. Good figures of hu- man spermia have recently been given by Ret- zius {Biol. Unters., neue Folge x, 1902), and an excellent diagram by Meves is reproduced by Waldeyer in Hertwig's "Handbuch," vol. i, p. 146. In the human spermium, which is essentially similar to that of other mammals, there may be recognized a head and a tail ; a neck piece is not clearly distinguishable. In the tail, if the indis- tinct neck piece be disregarded, there are a con- necting piece, a principal piece, and an end piece. Seen from the surface the head is oval, and in side view it is elongated pear-shaped, the tail being attached to the broader end ; upon each sur- face of the head there is a slight depression. '^ See, also, Meves : Zur Entstehung der Achsenfaden mensehlieher Spermato- zoen, Anat. Anz., vol. xiv, 1897; and Ueber das Verhalten der Centralkorper bei der Histogenese der Samenfaden von Mensch und Ratte, Verb. Anat. Ges. (Kiel), 1898. " J. Duesberg : Sur le nombre des chromosomes chez I'homine, Anat. Anz., vol. xxviii, 1906. THE GERM-CELLS. 9 According to Waldeyer one sees with very strong magnification a constriction between the head and the connecting piece, and this is an indication of the neck; and in this situation Krause and Waldeyer describe a small depression in the head, which receives the neck together with the connecting piece. By staining the head cap can be brought into view, its posterior border marking off an anterior and a posterior portion of the head. The anterior sharp border of the cap represents the perforatorium; special perfora- te ria, such as Nelson 15 and Bardeleben ^^ have described, may be produced by special conditions and perhaps have been confused with attached bacteria. The neck has the form of a disk, which is formed by the anterior centrosome bodies, the noduli anteriores (Fig. 5, A and B, Nd.a, dark), and a homogeneous intermediate substance, the massa intermedia {Ms.int., clear). The succeeding connecting piece (pars conjunctionis) begins with the noduli pos- teriores (the posterior centrosome bodies), represented in the diagram as a black stripe, and ends with the annulus; it includes, therefore, the region of the posterior centrosome, which during spermatogenesis has divided into these two portions. The filum principale of the tail traverses the axis of the connecting piece, extending from the proximal portion of the posterior' centrosome. In this region the filum principale has a delicate investment which probably passes over posteriorly into the thicker sheath of the tail and finally ends at the beginning of the filum terminale. Around this delicate sheath is the spiral sheath, and external to this the mitochondria sheath. The spiral sheath consists of a spiral filament, not recognizable in the mature spermium, and an intermediate substance, the substantia intermedia, represented as clear in the diagram. The mitochondria sheath is the matrix of the spiral filament and is characterized by the presence of mito- chondria granules. At the beginning of the principal portion of the tail the spiral and mitochondria sheaths terminate, but the inner thin sheath is probably continued into the involucrum of the tail. A spiral membrane has been described for the human spermium by several authors, but does not really occur. The measurements of the human spermium are, according to W. Krause ("Handbuch der menschl. Anat.," vol. i, p. 559, 1876), as follows : Entire length 52-62 ^, of which the head measures 4.5 i^, the connecting piece 6 ^u, and the tail 41-52 ix.. The width of the ""E. M. Nelson: Some Observations on the Human Spermatozoon, Joum. Quekett Micr. Club, London, ser. 2, iii, pp. 310-314, 1889. " Von Bardeleben : Ueber die Entstehung der Aehsenfaden im menschliohen und Saugetierspermatozoon, Anat. Anz., vol. xiv, 1897; also, Beitrage zur Histologie des Hodens und zur Spermatogenese beim Menschen, Arch. f. Anat. und Entwick- lungsgeseh., Supplementband, 1897; also, Weitere Beitrage zur Spermatogenese beim Mensehen, Jenaische Zeitsehrift, vol. xxxi, 1898. 10 HUMAN EMBRYOLOGY. -Ann -Jnv. Cd, \ — P.pr Fig. 5, A. Fig. 5, B. Fig. 5, A and B. — Diagram of a human spermium ac- cording to Meves. Cp., caput (head); CI., coUum (neck): Cd., Cauda (tail); P.c, parsconjunctionis (connecting piece); P.pr., pars principahs (principal portion of the tail); P.t,, pars terminalis (end piece); A^'d.p., noduli posteriores (ante- rior boundary of the connecting piece); Ann., annulus (pos- terior boundary of the connecting piece); L.P.pr., limes partis principalis (posterior boundary of the principal por- tion); P. a,, pars anterior capitis (anterior portion of the head); L.GaL, limes galece (boimdary of the head cap); P.p., para posterior capitis (posterior portion of the head; Nd.a., noduli anteriores (anterior centrosome bodies); Ms.ini., massa intermedia (intermediate substance of the neck); Spir., spiral filament; Inv., involucrum (sheath of the axial fila- ment in the connecting piece and principal portion of the taiU; Mtch., mitochondria; Sb.ini., substantia intermedia (intermediate substance of the spiral sheath); F.pr., filum principalis (axial filament). (From Waldeyer : Die Ge- schlechtszellen, in Hertwig's Handbuch, vol. i, p. 14^ Figs 43, A and B.) - e. - THE GERM-CELLS. 11 head is 2-3 <«, its thickness 1-2 /*. Giant spermatozoa also occur. Such a structure was figured by Widersperg" in 1885. Since then abnormal forms of human spermatozoa, after G. Retzius i* had described double-tailed forms in 1881, have been recently specially studied by Ivar Broman.i^ Some of his figures are here reproduced. Fig. 6, a and h, shows a giant and a dwarf sperm; the tails have in both cases about the normal length. Fig. 6, c and Fig. 6. — a-/, abnormal human spermia; g, a normal sperm under the same enlargement, o, giant sperm ; 6, dwarf sperm ; c and d, double-tailed forms ; «, a four-tailed, and j, a double-headed form. (After Ivar Broman, Anat. Anz., vol. xxi, 1902.) d, shows double-tailed spermia ; e, one with four tails ; /, a double- headed one ; and g, a normal sperm under the same enlargement. Broman remarks that certain forms of abnormal spermia may be the result of a general weakening of the body by illness and by the influence of morphine, coffee, and alcohol. The idea of Bardeleben ^^ that two different forms of human spermia occur " Gustav von Widersperg : Beitrage zur Entwicklungsgescliichte der Samen- korper, Areh. f. mikr. Anat., vol. xxv, 1885 (PI. VI, Fig.'lS). ^ G. Retzius : Zur Kenntnis der Spermatozoon, Biol. Untersnch., 1881. ^Ivar Broman: Ueber atypische Spermien (speziell beim Menschen) und ihre mogliehe Bedeutnng, mit 107 Abbildungen, Anat. Anz., vol. xxi, p. 455^61, 1902. ^ Von Bardeleben : Dimorphismus der mannlichen Geschlechtszellen bei Saugetieren, Anat. Anz., vol. xiii, 1897. 12 HUMAN EMBRYOLOGY. has been disproved, although such a dunorphism occurs among the invertebrates. The spermia are motile, being propelled by movements of the tail. They swim against a current, as was determined indepen- dently by Both 21 and Adolphi," although this fact had previously been observed by Lott (1872) and Hensen (1876), whose results had been forgotten. The current exercises a directing force upon the spermia, but for this it must have a certain strength. The influence begins with currents flowing with a rate of 3^ /* per second ; in currents of 4—20 /^ the spermia move forward, the more slowly the more rapid the current ; in a flow of 25 /* they can jusj; hold their own ; and in more rapid currents they are carried back- ward. The absolute rapidity of the spermia is 23-26 /^ ; only at the beginning of its action does the current increase their activity. Since dead spermia also move with the head against the current, the explanation of the effect must be sought in their physical structure. This adaptation is of great importance for fertiliza- tion, since it determines that the spermia will direct their course against the outwardly moving current produced by the cilia of the uterus and tubes and pass in the most direct route toward the infundibulum. Waldeyer (1. c, p. 209), following Henle (" Allgem. Anat.," p. 954), makes the rapidity of movement of the spermia considerably higher than Adolphi; in seven and a half minutes they cover a distance of 27 mm., this being at a rate of 60 ;u. per second. How long the human spermia may remain motile and capable of producing fertilization is uncertain. I found still motile spermia on the third day in the testes of an executed criminal, the organs having been placed unopened in picro-sublimate ; that they may remain motile for just as long a period in the cadaver is known (F. Strassmann, "Lehrbuch der gerichtlichen Medizin," p. 61, 1895). Bossi^* found still living spermia in the vagina twelve to seventeen days and in the cervix five to seven and a half days after the last cohabitation; Diihrssen {Sh. Ges. Geburtshilfe vmd Gynaekol., Berlin, May 19, 1893; also Zweifel: "Lehrbuch der Greburtshilf e, " 3 Aufl., 1902) observed living spermia in a dis- eased tube nine days after the admission of the patient to the ^ A. Roth : Ueber das Verhalten beweglicher Mikroorganismen in stromenden Flussigkeiten, Deutsche med. Wochenschrift, Jg. 19, p. 351-352, 1893; also, Zur Kenntnis der Bewegung der Spermien, Arch. f. Anat. uiid Physiol., Physiol. Abt., 1904. '' H. Adolphi : Die Spermatozoen der Saugetiere schwimmen gegen den Strom, Anat. Auz., vol. xxvi, 1905. '^ Bossi : Etude clinique et experimentale de I'epoque la plus favorable a la fecondation et de la vitalite des spermatozoi'des sejoumant dans le nidus seminis Rivista di ostetrie. e ginecol., 1891, no. 10, also in Nouv. Arch, d'obstetr. et de gynecol., April, 1891. THE GERM-CELLS. 13 clinic; according to the statements of the patient the last co- habitation had occurred three and a half weeks previously. Alil- feld succeeded in keeping spermia alive for eight days at body temperature, and Wederhake ("Zur Technik der Spermaunter- suchungen," Monatsschrift Urol., vol. x, p. 520-525, 1905) found, in sperm that had been kept sterile, still living spermia on the eighth day. All these data indicate that human spermia in the female genitalia remain still capable of fertilization for a con- siderable period, certainly over a week. In animals spermia capable of producing fertilization may remain in the female genitalia for months, as in the bats, in which, as I have satisfied myself, copulation occurs in the autumn while the fertilization does not result until the following April or the beginning of May. COMPARISON OF THE OVIUM AND SPERMIUM. A comparison of the ovium and spermium is possible only when the development of both cells is considered. This will be done in the chapter on the urogenital apparatus and it will be only briefly treated here. In the development of the male and female germ-cells three periods may be distinguished. The first period is that of increase, in which the germ-cells — at this stage termed oogonia in the female and spermatogonia in the male — undergo a rapid increase by mitotic division. At a certain period of development the divisions cease and the cells produced by the last divisions — the oocytes of the first order in the female and the spermatocytes of the first order in the male — then enter upon a period of growth, which, especially in the female, leads to a great increase in size. At the close of this period that of maturation begins, during which the polocytes are formed in the female cell. A corresponding period occurs in spermatogenesis. Two divisions follow quickly on one another; each spermatocyte of the first order divides first into two spermatocytes of the second order and each of these again divides into two spermatids. Just as in the case of the matured ovum, the oide or ovium, and the three polocytes, which are to be regarded as rudimentary oides, the chromatin elements are reduced to half their original number, so too in each of the four spermatids that are formed by the division of a spermatocyte of the first order the chromosomes are reduced by one-half. A marked difference, however, occurs: whereas the four oides (the ovium and the three polocytes) are very unequal in size, the ovium being many times larger than each of the polocytes, the four spermatids from each spermatocyte of the first order are equal in size. And another difference lies in this, that while the ovium is capable of being fertilized immediately after or even during the second maturation division, the four spermatids must still undergo a complicated process of develop- 14 HUMAN EMBRYOLOGY. -E- M A l\ W W . A l\ l\ W J 2 5 e 7 s 1 - ei' I 3Jg^ M \ ^^pz,' ment, that has no equivalent in the ovium; they must be trans- formed into spermia before they are capable of fertilizing. A comparison of spermatogenesis and oogenesis is clearly shown by Boveri's diagram (Fig. 7), taken from Waldeyer (p. 225). The male and female germ-cells are shown by this comparison to be essentially equivalent, and yet the subordinate differences are of the greatest importance. Both are cells in which the num- ber of the chromosomes has been reduced to one-half by quite comparable pro- -f ♦ cesses; their differ- ences are associated with a division of labor. In the ovum nutriment is stored up for the new being that will be formed; it consequently grows to a considerable size. During the maturation divisions all the nutritive ma- terial is retained by one cell; the polo- cytes receive none of it, the ovium contains it all. On account of the mass of nutritive material the ovium becomes heavy and is not in a condition to move toward a union with the male germ- cell ; it must await it. In the spermatocytes of the first order there is, in the first place, much less nutritive material stored up, and, in the second place, it is equally distributed among all the four spermatids dur- ing the maturation divisions ; and, in order that they may be able to seek out the ovium, each of these four spermatids must be further modified into spermia. Wliile, therefore, from each oocyte of the first order but one ovium capable of being fertilized is produced, from each spermatocyte four spermia arise, and thus there are formed from an equal number of spermatocytes and oocytes of the first order four times as many fertilizing spermia as there are fertilizable oides. And yet this difference as we V -^v* •^ •* w pz^ Fig. 7. — Diagram of the development of the primitive germ- cells (I in the zone of increase) — the primitive sperm cells to the left, the primitive ova to the right — into spermatids and spermia or into ovia; II and III in the zone of increase represent the spermato- gonia and oogonia. The upper small cells in the zone of growth enlarge to form spermatocytes or oocytes (ei^) of the first order. By the division of these cells (I in the maturation zone) there are formed (to the left) two spermatocytes of the second order (pre- spermatids) or (to the right) an oocyte of the second order (ei^) and a first polocyte (pz^). The succeeding division, represented at II in the maturation zone, produces the spermatids (1, 2, 3, 4, to the left) or a mature ovium (ei^), together with the second polocyte (pz'^). A division of pz^ may also take place and then there will be formed, to the right as well as to the left, four descendants from the cells represented by I of the maturation zone. (After Boveri and O. Hertwig, from Waldeyer's " Geschlechtszellen " in Hertwig's Handb., vol. i, part i, p. 225, Fig. 56.) THE GERM-CELLS. 15 shall see later, does not represent, even approximately, the nu- merical relations of the mature ova and the spermia. It must be noted, however, that so far as man and the mam- mals are concerned the comparison of the oogenesis and sperma- togenesis shown in Boveri's diagram is not quite free from objec- tion, for it is doubtful if the oogonia and spermatogonia can be exactly homologized in these forms. For although both arise from the germinal epithelium, nevertheless they appear to belong to ditferent cell generations. In the male, according to recent observations, the germ-cells have their origin from cells which correspond to those of the medullary cords of the ovary; the ingrowths which give rise to the germinal cords, in which the ova develop, have no homologue in the testis. The annexed diagram will make this clear. Bonnet " imagines that exceptionally the polocytes may also be fertilized and give rise to embryomata. As regards the number of germ-cells which may normally be produced, Hensen estimates that a human female develops in each ovary about 200 ova ripe for fertilization. Lode ^^ found in 1 c.mm. of a human ejaculate 60,876 spermia, and from that calculates for the entire ejaculate, which averages 3373 c.mm., over 200 million spermia, so that during his life a man may produce about 340 billion spermia. Consequently for every mature ovium there are about 850 million spermia. Finally, the question is to be considered whether the sex of the future individual is in any way determined in the germ-cells, the ovium or spermium. In the chapter on the development of the urogenital organs it will be shown that with our present meth- ods of observation the sex of the human embryo cannot be de- termined before the fifth or sixth week of development; as is well known it has been supposed that during the earlier stages of pregnancy influences may be brought to bear which will de- termine the formation of the one or the other sex. This, however, seems to be impossible, since whether the developing organism shall be of the male or the female sex is, apparently, already determined at fertilization. This is indicated by the fact that twins and double monsters formed from a single ovum are always of the same sex. Most observers now incline to the belief that the sex is ahvays determined before fertilization in the ovium, that there are mature ova (ovia) from which only females and others "" Bonnet : Zur Aetiologie der Embryome, Greifswald. med. Verein, Bericht in Miinchener Med. Wochenschrift, 1901, p. 315. ^ A. Lode : Untersuchungen liber Zahlen und Regenerationsverhaltnisse der Spermatozoiden bei Hunde und Mensch, Arch. f. ges. Phys., 1891; also, E.xperi- raentelle Beitrage zur Physiologie der Samenblasen, Sb. K. Acad. Wiss. Wien., Abt. 3, vol. civ, p. 33, 1895. Hi HUMAN EMBRYOLOGY. THE GERM-CELLS. 17 from which only males can be formed. In my opinion the question is not yet settled so far as man and the vertebrates are concerned, but it is certain that in many invertebrates the sex is already de- termined in the egg. How the germ-cells of the human female or male may be influenced so that they will produce either the one or the other sex has been frequently discussed, but none of the con- clusions will stand serious criticism. Nussbamn ^o has shown that in the rotifer Hydatina senta the sex of the progeny is determined while they are still within the body of the mother, since all poorly nourished females deposit eggs from which males develop, while from the eggs deposited by well-nourished females only females were formed. All the eggs of any one female pro- duce the same sex and neither fertilization nor any treatment after fertilization has any effect. In an extensive series of observations on a mammal, the mouse, 0. Schultze^^ did not succeed in influ- encing the sex character ; no influence was exerted by the age or the difference of age of the parents, by the age of the sexual products, by in-breeding or by incest-breeding, by sexual appetency, etc. For further consideration of the question reference may be made to Henneberg ^* and Lenhossek.^^ "° M. Nussbanm : Die Entstehung des Geschleehts bei Hydatina senta, Arch, f. mikr. Anat., vol. xlix, p. 227-308, 1897. ^ 0. Schultze : Znr Frage von den geschleehtsbildenden Ursachen, Arch, f . mikr. Anat., vol. Ixiii, p. 197-257, 1903. ^° B. Henneberg : Wodnrch wird das Gesehleehtsverhaltnis beim Mensohen nnd bei den hoheren Tieren beeinflusst? Ergebnisse der Anat. nnd Eutwieklungs- gesch., vol. vii, p. 697-721 (Lit., 1897), 1898. ^ M. von Lenhossek : Des Problem der geschleehtsbestimmenden Ursaehen, Jena, 1903. Vol. I.— 2 II. FERTILIZATION. By FRANZ KEIBEL, Feeibxieg i. Br. Nothing is knoA^m concerning the fertilization of the human ovum, but it may be presumed that it takes place in essentially the same manner as in other mammals, and for an account of the process in these forms reference may be had to the works of Sobotta and Leo Gerlach, cited in the preceding chapter. The usual place for fertilization must be the first portion of the tube. That the spermia penetrate into the tube has been observed (Diihrssen: 8b. d. Ges. f. Gyn. und Geburtsh. in Berlin, 1893; and Zweifel: "Lehrbuch der Geburtshilf e, " 3 Aufl., 1902), and is also made certain by the occurrence of tubal pregnancies. Occa- sionally the union of the ovum and sperm-cell may take place upon the surface of the ovary or even in the interior of the Graafian follicle, as is definitely shown by rarely occurring cases of ovarian pregnancy (see Freund and Thome,^ and Bryce, Teacher, and Kerr^). Accurate estimates of the rapidity of penetration of the spermia into the uterus and tubes are not available, but it may be supposed that in one or two hours after coitus the spermia have penetrated far into the tubes. That fertilization of the ovum may take place after it has reached the uterus seems to me from observations on other mammals very improbable; at all events no good grounds for such a supposition seem to exist. Nevertheless Wyder ^ regards the uterus as the normal site of fertilization; and other gynaecologists — Hofmeier, for example — agree that fertilization may occasionally take place in the uterus. Waldeyer also comes to this conclusion and states (Hertwig's "Handbuch," vol. i, p. 370) that, "It cannot be denied, however, that under certain circumstances fertilization may first occur in the uterus"; he does not state, however, what these circumstances may be. 'PI. W. Freimd and R. Thome: Eierstoeksehwangersehaft, Virch. Arch., vol. clxxxviii, pp. 54-91. ^ Bryee, Teacher, and Kerr : Contributions to the Study of the Early Develop- ment and Imbedding of the Human Ovum, Glasgow, 1908. ° Th. Wyder: Beitrage zur Lehre von der Extrauterinschwangerschaft und dom Orte des Zusammentreffens von Ovulum und Spennatozoen, Arch. f. Gynakol., ■\ol. xxviii, 1886. 18 III. SEGMENTATION. By FRANZ KEIBEL, Treibueg i. Br. The segmentation stages of the human ovum have not yet been observed. We may with certainty assmne that the early stages of fe'rtihzation are passed through during the passage of the ovum through the tube, but whether the entire segmentation takes place during this passage, or in what stage of seg- mentation the ovum reaches the uterus, cannot even be con- jectured. In mammals there are apparently differences in this respect. The time, also, that the human ovum requires for the passage of the tube is very difficult to estimate; accord- ing to the data obtained from other mammals it cannot well be believed that the uterus is reached before the fifth day. Similarly, it is unknown whether the ovum becomes imbedded in the mucous membrane immediately after it has reached the uterus. That it is possible by good fortune and persistency yet to observe segmenting human ova in the tube is shown by the observations of Letheby ^ and Hyrtl (in a work by Bischoff ^ and in Froriep's " Neue Not.," 1852, No. 603), who discovered ova in the tube, although they were not able to make observations of the segmentation, partly on account of the imperfections of their technic and partly because the ova were unfertilized. The relative certainty with which experienced embryologists are able to-day to obtain the segmentation stages of even large mammals is an encouragement for further efforts in this direction. The force which propels the ovum through the tube into the uterus is the ciliary action of the tubal epithelium, and injury to this epithelium may be the cause of the retention of the ovum in the tube or in a diverticulum of it and so the cause of a tubal pregnancy. If the ciliary current is not impaired, the ova are readily driven over any diverticula that may exist (Kromer^). 'H. Letheby: An Aeconnt of Two Cases in which Ovules or Their Remains Were Discovered in the Fallopian Tubes of Unimpregnated Women who Had Died during the Period of Menstruation, Philos. Transact. Royal See. London, 1852 (altogether unsatisfactory). See also Froriep's Neue Notizen, 1852, No. 603. ' Th. L. W. Bischoff : Beitrage zur Lehre von der Menstruation und Bef ruch- tung, Zeitschrift fiir rationelle Medizin, neue Folge, vol. iv, 1854. P. Kromer : Untersuchungen iiber den Bau der menschlichen Tube, Leipzig, 1906. 19 20 HUMAN EMBRYOLOGY. It may be regarded as quite certain not only that the human ovum undergoes a seo-mentation quite similar to that of the other mammals, but also that it is a secondary total segmentation. The ancestors of the human species, like those of other mammals, must have once possessed yolk-laden meroblastic ova. A separation of the segmentation cells according to their developmental potencies has been variously postulated for the earlier stages of the mammalian segmentation, but conclusive evidence for this is lacking, the observations hitherto made not being capable of such an interpretation. This is true also of the observations which have been supposed to indicate the existence of a gastrulation process in the later stages, but this question will be considered in the chapter dealing with the forma- tion of the germ layers and the gastrulation prob- lem. Finally, it may be noted here that Hubrecht has recently succeeded in finding the ovum of a monkey in segmentation. It has been figured in the posthumous paper by Selenka which I have edited (" Menschenaffen," 5 Lief., Zur vergleich- enden Keimesgeschichte der Primaten, Wiesbaden, 1903) and is reproduced here (Fig. 9), since it is the only primate ovum in a segmentation stage at present known. Selenka states concerning this ovum, which was found in serial sections of a tube of a Macaeus nemestrinus from Java : " At about the middle of the oviduct was the ovum, having a diameter of 0.04 mm. and loosely attached to the somewhat frayed out ciliated cells. The largest of the approximately ripe ovarial ova were of about the same size. Four segmentation cells of about equal size are clearly to be distinguished; two of these (the central and left upper ones in the figure) are irregularly oval, the other two are almost spherical. The cells are naked; no trace of an enclosing membrane is to be observed. The shrinkage which the tissues of the oviduct show suggests the idea that the segmenting ovum no longer retains its natural condition. It is, however, of importance to note what the preparation reveals: The segmentation begins in a manner similar to that of other higher mammals, and it is probable that it is com- pleted as soon as the ovum has entered the uterine enlargement." This last conclusion I cannot accept. Fig. 9. — Ovum of a monkey in segmentation, from the tube of a Macaeus nemestrinus Demarest. X 400. (From Selenka: "Men- schenaffen," 5 Lief., Fig. 1, p. 331, 1903. ) IV. YOUNG HUMAN OVA AND EMBRYOS UP TO THE FORMATION OF THE FIRST PRIMITIVE SEGMENT. (a ceitical account) By FKANZ KEIBEL, Freibueg i. Be. By the term ovum is understood in human embryology not only the egg-cell but also later the entire structure developed from the egg-cell, the embryo or fetus surrounded by the amnion and chorion. In this sense the word is used here. I do not intend to enumerate and describe here all young and very young human ova that have been observed, but only those which may be regarded as normal or approximately so, and as such I can regard only those in which an embryo has been observed. The observations of Graf Spee and Peters on human ova, and of Selenka on those of monkeys, have shown that in man and the primates the chorion grows much more rapidly than the embryo, that, consequently, a relatively large ovum may contain a very small embryo, and that even in the youngest ova yet studied the amnion and the yolk sack are already formed. I shall show later that in all probability the embryo never lies free upon the surface of the ovum, as it does in the birds and in many mammals, but that from the beginning it is sunk in the interior of the ovum, and that the amniotic cavity arises as a cleft and not by the formation of folds, and is always closed. The extraordinary minuteness of the embryonic anlage is a sufficient explanation why in early times, when the methods of investigation were imperfect, it was overlooked or unrecog- nized ; many of these earlier described ova may have been normal or nearly so. But when no embryonic anlage is found in an ovum that has been investigated according to all the rules of modern technic, as is the ease with that which Gr. Leopold ^ has lately studied with so much care, that ovum is certainly to be regarded as pathological; and the occurrence of maternal blood in the interior of the ovum is further evidence in this direction, as Spee has pointed out in Schwalbe's JahresbericM. Leopold's ' G. Leopold: Ueber ein sehr junges mensehliches Ei, Arb. Kgl. Erauenklinik, Dresden, vol. iv, Leipzig, 1906. 21 22 HUMAN EIMBRYOLOGY. ovum does not, therefore, require consideration here ; on the other hand, some older observations may be noticed, such as those of Eeichert, Wharton Jones, and Breuss. The ovum of Eeichert especially has played and is still playing, though improperly, an important role in human embryology. Reichert" found the ovum in the uterus of a suicide and estimated its age at twelve to thirteen or thirteen to fourteen days. It was completely enclosed by the mucous membrane of the uterus. On the side of the capsule which was turned towards the uterus there was a transparent area measuring 3 mm., which Reichert termed the capsule scar, believing that at the sides of it the mucous membrane of the uterus had grown up to surround the ovum. The ovum itself was a lenticular vesicle, whose diameters were 5.5 and 3.3 mm. The surface of the vesicle which was turned towards the uterus, the basal surface, was almost flat, that turned toward the lumen of the uterus somewhat curved. The marginal zone was richly furnished with small villi, the largest of which were 0.2 mm. in length and were already partly provided with lateral branches. From the margin small villi, diminishing in size, extended for some distance upon the surface of the vesicle turned toward the uterus wall (the basal surface of Reichert) ; but at 'the centre of the surface an area of about 2.5 mm. diameter remained free from them. In the centre of this free area Reichert described a dull circular spot. The surface of the ovum turned toward the lumen of the uterus was free from villi. The statements that Reichert makes concerning the finer structure of the ovum are in part insufficient and in part quite erroneous. Thus the wall of the ovum could not have been, as he supposed, purely epithelial, nor the villi hollow epithelial structures, but the wall must have been formed of mesodermal tissue with an epithelial coveriag and the axes of the villi occupied by mesodermal tissue. Reichert, indeed, perceived this mesodermal tissue, but regarded it as coagulated material. Also what he says concerning the ingrowth of the villi into the uterine glands is undoubtedly incorrect. As regards the structure of the dull spot on the uterine surface of the ovum, he supposed that it was formed by a layer of small, finely granular, nucleated, polyhedral cells, situated within the epithelial wall. It was taken for the embryonic anlage, and His has estimated the diameter of this " embryonic spot " as 1.6 mm. That this spot really represents the un- injured embryonic anlage is improbable, and Kollmann's statement in his " Hand- atlas der Entwieklungsgeschichte des Menschen " (1907) — "From what we know from the mammals this spot would now be regarded as the embryonic shield " — is, as will be shown later, absolutely disproved. A definite opinion cannot be given on account of the insufficiency of Reichert's description; nevertheless, I regard as well founded the conclusion of Spee, that Reichert had destroyed the actual embryonic structure during his preparation of the ovum and that in its degree of development it would have occupied a place between the ovum of Peters, to be described in detail later, and the Van HerflE o^um of Spee. Probably Reichert had casually obser^'ed the embryo; it may have been the spherical body on the basal wall which he mentions on p. 26. Another ovum which deserves mention is that described by Wharton Jones ^ in 1837; it was of the size of a pea. The fis:ure drawn from the preparation in alcohol shows a diameter of 6.2 X 4.7 mm. The surface turned toward the lumen ° Reichert : Beschreibung einer f riihzeitigen mensehliehen Frucht im blaschen- formigen Bildungszustande, etc., Abh. Kgl. Akad. d. Wiss., Berlin, 1873. ' Thomas Wharton Jones : On the First Changes in the Ova of the Mammi- fera in Consequence of Impregnation and on the Mode of Origin of the Chorion, Philosoph. Transact. Royal Soe. London, 1837, p. 2. OVA AND VERY YOUNG EMBRYOS. 23 of the uterus was free from villi. Imbedded in the cavity of the ovum was a spherical body, with a diameter of 1.5 mm., which His, probably correctly, identified as_ the embryonic structure, that is to say, the actual embryonic anlage together with the amnion, yolk sack, and belly stalk. His assumes that this embryonic structure may have been artificially displaced. Next comes an ovum described by Breuss ' in 1877. It was expelled together with the entire lining of the uterus. The wall of the ovum, which had a diameter of 5 mm., consisted of two layers, the outer of which was epithelial and the inner formed of connective tissue. The villi were for the most part unbranched and were about 1 mm. in length and 0.07 mm. in diameter; they left free a roundish area 2 mm. in diameter. Vessels could not be distinguished in their interior. A projection which occurred in the interior of the ovum, consisting of nucleated cells and having a length of 1 mm. and a diameter of 0.5 mm., may have been the embiyonic structure. If it is assumed that the ovum was normal, we must suppose that Breuss overlooked the amniotic cavity and the cavity of the yolk sack, a supposition which I regard as possible. Mention may also be made of two other ova, described by Allen Thomson.' In one of these the embryonic structure must be regarded either as pathological or as not corresponding to the degree of development of the ovum. Thomson estimates the age of the smaller of the two ova at twelve to thirteen days. It had a diameter of 6.6 mm., was everywhere surrounded with villi, and was almost com- pletely filled by a vesicle which was apparently the yolk sack. Upon the yolk sack was an embryo 2.2 mm. long and with both its cranial and caudal ends separated from the sack. Thomson makes no mention of an amnion, but we may suppose that it was present and covered the embryo on the surface opposite the yolk sack; and a belly stalk must also have been present, since Thomson states that the embryo was attached by its dorsal surface to the external egg membrane, that is to say, to the chorion. Although the ovum is larger than that of His, to be described below, the second observation of Thomson may be recorded here; it concerns an ovum measuring 13.2 mm. in diameter, whose age was estimated at fifteen days. It was oval in shape and was evenly surrounded with villi. In the interior was a large cavity filled with fluid; and at one spot was the embryo, closely attached to the chorion and with a yolk sack and the remains of the amnion. The embryo had a length of 2.2 mm. and its cranial and caudal ends projected somewhat beyond the yolk sack. Viewed from the surface turned toward the chorion, it showed distinctly the medullary folds, which manifested a tendency to fuse at the middle of their length; ventrally was the heart. The diameter of the yolk sack was also 2.2 mm.; nothing is said concerning an amnion, but a lobe which is shown in Thomson's figure at the head end of the embryo is apparently the remains of an amnion that had been destroyed during the preparation of the embryo. It is interesting to note that Kolliker" in 1879 regarded this second ovum described by Thomson as not quite normal on account of the large space which separated the embryo and yolk sack on the one side from the inner surface of the chorion on the other. We now know that this is the normal condition in embryos of this stage; and it is rather the smaller of Thomson's ova, which Kolliker was inclined to regard as normal, that shows abnormal conditions, since the yolk sack never fills the chorion so completely either in human or mammalian ova of this stage. In all the ova so far mentioned a correct identification of the embryo, the ' K. Breuss : Ueber ein menschliches Ei aus der zweiten Woche der Graviditiit, Wiener med. Wochenschrift, 1877, pp. 502-504, 'Allen Thomson: Edinburgh Med. and Surg. Journal, vol. ii, 1839; and Froriep's Neue Notizen, vol. xiii, 1840. ■°A. Kolliker: Entwicklungsgesehiehte des Mensehen, 1879. 24 HUMAN EilBRTOLOGY. yolk sack, amnion, belly stalk, and chorion is possible only in the two described by Thomson, which contained embryos already rather well developed; and even in these the identifications were only general ones, as may be seen from Kolliker's comments upon the ova. He regarded, on the basis of the information available at that time, the normal ovum as abnormal and the abnormal one as normal, and the same conclusion was reached by Eeker, another distinguished embryologist of the time. A definite idea of the relations of the amnion and the belly stalk was also impossible for KoUiker. A correct interpretation of the discoveries mentioned could not be given at the time of their publication and, indeed, in part, not for some time after. The investigators who sought such interpretations were led from the right path, and Keichert's ovum, as I have stated, gave rise to many false ideas, even up to recent times. Consequently, as Elze and I ' have already pointed out in our "Normentafel zur Entwieklungsgeschichte des Menschen," an observation by His * mai-ks an important advance. The ovum in question (No. XLIV [Bff.] of His's collection) had a greater diameter of 8 mm. and, at right angles to this, a diameter of 7 mm.; it was somewhat flattened and at one point the villi were somewhat fewer than elsewhere. " On opening it there was found, on one wall, a small body measuring 1.4 mm. in its longest diameter and consisting of an ellipsoidal opaque body with a transparent vesicle attached to it. The opaque body, which seemed fi-om partial foldings of its surface to be hollow, had a greater diameter of 0.85 mm. and a diameter at right angles to this of 0.6 mm. The transparent vesicle surrounded by its border one end of the ellipsoid. The connection with the chorion was by means of a short stalk, which stood in relation to both the vesicle and the ellipsoid." " I regard," His says in another place, " the more solid body as the umbilical vesicle and the transparent part as the amnion, and conclude from this that the embryonic anlage, so far as it is present, lies at the boundary between the two. With this idea the manner in which the structure is attached to the chorion agrees. That is to say, the place of attachment lies on the boundary between the vesicle and the opaque body." His's interpretation, we can say to-day with all certainty, is in agreement with the actual facts, and His was the first to give a perfectly correct interpretation of a human ovum of this stage. He further reported con- cerning this ovium that to the lower pole of the yolk sack there were attached threads of that looser tissue "which traverses the cavity of the ovum, one of these threads being especially distinguished by its tougher consistency and its opacity." Later observations of young human embryos, carried out with the methods of modem technic, and especially with the aid of well stained and perfect series of sections, have, as has been already stated, confirmed His's views and have led to further, partly unexpected, results. It was the observations of Graf Spec, especially, that opened the way, and, later, H. Peters rendered great service; but for the sake of continuity the ova in question will not be described in the order in which they were discovered, but according to their degree of development. Consequently I shall begin with the ovum described by Bryee and Teacher." The ovum was obtained from an abortion, and although the preservation of the embryo proper is not perfect yet it is of the greatest importance ; Fig. 10 shows the ovum as it lay in the uterine mucous membrane, according to a diagram by Bryce. TVith the exception of a small area it is completely surrounded by ' Franz Keibel and Curt Elze : Normentafel zur Entwicklungsgesehichte des Menschen, Jena, 1908. ° W. His : Anatomie menschlieher Embryonen, Leipzig, 1882, part ii, pp. 32 and 87 et seq. See also Mall, Journ. of Morph., vol. six, p. 1.51. " Bryce,, Teacher, and Kerr : Contributions to the Study of the Early Develop- ment and Imbedding of the Hmnan Ovum, Glasgow, 1908. OVA AND VERY YOUNG EMBRYOS. 25 decidua, and the opening in the deeidua (capsularis) is closed by coagulated fibrin containing leucocytes. A large opening with fungoid tissue (the closing coagulum of Bonnet) is wanting. The ovum, surrounded with blood, lies in a relatively large chamber, with whose walls it is not united; the maternal and fetal tissues are quite separate. The innermost layer of the decidua, which forms the capsule of the ovum, is in an advanced stage of coagulation necrosis and, together with some deposits of fibrin, forms around the ovum a capsule of dead tissue, which is interrupted only at one or two places, where blood-vessels open into the egg chamber, and at one where a hemorrhage has broken through into it. pi'. cap. Fig. 10. — Diagram of the ovum of Teacher and Bryce, after Bryce. P.e., point of entrance; cyt., cytotropho blast; pi., plasmoditrophoblast; n.z., necrotic decidua zone; gl., glands; cap., capillaries; pi'., vacuolized Plasmodia which are penetrating capillaries. The cavity of the ovum is completely filled with mesoblast, and in this the meduUo-amniotic and entodermic (intestine-yolk sack) vesicles are imbedded. The natural proportions of the various parts have been accurately preserved. X 50. (From Contribu- tions, etc., 1908, p. 41, Fig. VII.) The wall of the ovum consists of an inner layer (the cytotrophoblast, the layer of Langhans's cells), whose cells are poorly separated from one another and externally pass over into a very irregularly arranged plasmoditrophoblast ; this has a distinctly plasmodial character and forms a loose network, whose spaces are filled with maternal blood. The cytotrophoblast is nowhere continued into the plasmoditrophoblastic trabeeulee. The cavity of the ovum is occupied by a delicate tissue which has the characters of mesenchyme. This primitive mesoblast shows no traces of a splitting into a parietal and a visceral layer; there is, accordingly, no coelom. Also, projections of the mesoblast toward the cytotrophoblast (mesodermic villi) are not yet present. The embryonic structures (the embryo together with the amnion and yolk sack) are represented by two vesicles which have an exeentric position, and are completely separated from th^ cytotrophoblast by the mesenchymatous tissue. 26 HUMAN EMBRYOLOGY. The cavity of the larger vesicle is supposed to be the amniotic cavity and that of the smaller the cavity of the yolk sack. The cells which enclose the amniotic cavity are cubical and those enclosing the yolk sack are flattened, but in neither vesicle do they show individual differentiation. The egg-chamber is oval in form, the longest axis, lying paraUel to the surface of °the uterus, measuring 1.95 mm.; perpendicular to this and also parallel to the uterine surface the lumen of the egg-capsule measures 1.1 mm. and its depth (perpendicular to the surface of the uterus) is 0.95 mm. Smee the wall of the ovum itself is folded the measurements of the cavity can be stated only approximately as 0.77 and 0.63 mm. The relative sizes of the ' amniotic cavity and of the yolk sack may be perceived from Fig. 10; unfortunately these structures were not intact. The estimate of the age of the o^^lm made by Teacher and Bryce is very interesting and important, since it is based on exact data concerning the men- struation and the cohabitations. They estimate the age at thirteen to fourteen days, the ovum having been expelled sixteen and one-half days after the fertilizmg coitus. The cause of the abortion is supposed to have been a later coitus. If this estimate be correct the other ova to be described later on are all older than has hitherto been supposed. The table given by Bryce and Teacher may be reproduced here. TABLE I. (From Bryce and Teacher.) Chronological table of twelve well-described early pregnancies. Fertilization is assumed to be effected about 24 hours after insemination, and 24 to 48 hours are allowed for the completion of abortion. The leading data are supplied by the histories of Nos. 1, 4, 5, 8, 9, 10; and the position of the remainder is adjusted according to their dimensions and state of development. The ages according to the conven- tion of His are shown in the column headed, " Days elapsed from omitted period. 1 Teachek-Bryce. 2 Peters . Jung 4 Merttens . 5' Beneke . .. Von Spee (Von Herff) Leopold 8 Reichert . . . 9 Rossi Doria. 10] Etebnod 11' Fbassi Dimensions in millimetres. Ovum. Embryo. External. Internal. 1.9.5x0.95x1.10 2.4x1.8 1.0x3.0 6.0x4.5 6.0x6.5 9.0x8.0 10x8.2x6.0 13x5.0 12 Von Spf.e (Glae- vecke) 0.77x0.63x0.52 1.6x0.8x0.9 2.5x2.2x1.0 3.0x2.0 4.2x2.2x1.2 4.0 4.0x3.7 5.5x3.3 6.0x5.0 6.0x4.8x3.6 9.4x3.2 10x8.5x5.5 fan" .= o So Eli about 0.15 0.16 like 6 but young' r 0.37 Days elap-sed §■3 1.3 1.17 1.54 34 10 12 Age in days. 13-14 14-15 UJ-loJ 14 5-15 J 16-17 17-18 17-18 17-18 18-19 18-19 18-19 19-20 Remarks. How obtained. Abortion 16J days after cohabitation. Suicide. Autopsy. Periods every 5-6 weeks. Curetting on account of leucor- rhoea. Curetting. Curetting. Abortion two days after beginning of in- fluenza. Hysterectomy: carci- noma of cervix. Men- struation duringpreg- nancy (?) . Sudden death. .Au- topsy. Abortion with .sudden beginning and three days' retention. A single cohabitation 21 days before abor- tion. Hysterectomy ; metri- tis chronica. Recent abortion. OVA AND VERY YOUNG EMBRYOS. 27 TABLE II. (From Bryee and Teacter.) „.l„. l.fiT'"^ the relation of the dates of fertilization and of imbedding to the menstrual cycle calculated from the data given m Table I. The higher figure in the age column is arbitrarily chosen in each mstance, and allowance is made for the special circumstances of each case. Fertilization. Days of menstrual period. Imbedding. Merttens . . . Eossi Doria. Beneke Eternod . Peters . Jung .. Von Spee (Glaeveoke) . Von Spee (Von Herff). Frassi Teacher-Bryce and Reichert . Leopold 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 . . . Merttens .Rossi Doria Beneke Eternod . Peters . . Jung Von Spee (Glaevecke) Von Spee (Von Herff) Days of succeeding menstrual period. 1 2 3 4 5 6 7 8 9 10 Frassi Teaclier-Bryce and Eeicliert Leopold We now come to the ovum described by Hubert Peters"; it was obtained at the autopsy of a woman who had poisoned herself with caustic potash. Death had occurred within three hours after the taking of the poison and the autopsy was performed on the same day, a few hours after death. The ovum, to- gether with the entire egg capsule, fixed by Prosector Kretz in Miiller's fluid and hardened in alcohol, was successfully stained and microtomized by Hochstetter; it found in Peters an exceedingly careful observer. The uterus, from which the ovum was taken, was the size of a goose's egg, and thick-walled, and felt some- what more doughy and softer than a normal uterus. " The decidua of the corpus uteri was traversed by numerous furrows which crossed one another at various angles and occasionally formed grooves, so that the mucous membrane between " Hubert Peters : Ueber die Einbettung des mensehlichen Eies und das friiheste bisher bekannte mensehlichen Placentationsstadium, Leipzig und Wien, 1899. 28 HIBIAN EMBRYOLOGY. these bounding furrows formed root-like or occasionally rounded projections toward the uterine lumen. In the middle of the posterior waU Prosector Ivretz noticed a small area which was of the size of a hemp seed, which was somewhat paler but not prominent"; it contained the ovum. This was ellipsoidal in form, its diameters being 1.6X0.8X0.9 mm., these measurements, however, being of the cavity of the egg capsule. The part of the decidua m which the ovum lay presented, as the study of the sections showed, a slight, rounded elevation toward the cavity of the uterus. " While the decidua stretched as a very thm sheet, m the form of a capsularis, over the lateral portions of the ovum, the summit of the ovum was quite free from maternal tissue and projected freely into the lumen of the uterus by means of a blood-granulation mass which rested upon it "; this mass Peters terms the fungoid tissue (Gewebspilz). If the interpretation which Peters gives of the fimgoid tissue is correct and it is not a post-mortem phenomenon or a result of the poisoning, we have for the first time a human ovum that is not yet quite covered by a capsularis. Later Spee {Verh. Anat. Ges., 1902 (discussion of Marchand's paper); compare Schwalbe's Jahresber., n. P., vol. viii, No. 2, p. 298) stated that occasionally young ova were to be found in which the egg capsule had an opening toward the uterus at the place where the scar tissue occurred, and at the correspondmg place the ovum of Bryee and Teacher showed, as has been mentioned, only a small opening and no well-marked fungoid tissue. Upon the mesoblast layer of the ovum, which shows indications of the first few and as yet but slightly developed villi, follows a layer of epithelial cells, which reaches in places a thickness of more than 0.5 mm. and is traversed in a honeycomb manner by smaller and larger blood lacunse, but still remaining continuous at the periphery. Peters interprets this rightly, as I believe, as an " eetoblast shell " and names it the trophoblast, adopting the term which Hubrecht had proposed as a result of his observations on the hedgehog. Its further significance will be fully considered in the chapter on placentation and it yn\l there also he compared with the corresponding structures of other mammals (cheiroptera) and with what is found in older human ova (Kastschenko, Merttens, ^'on HerfE). As in the ovum of Bryce and Teacher differentiation of the cells of the trophoblast investment was e^^dent, and the cytotrophoblast (Langhans's cells) and the spongiotrophoblast were distinguishable." The embryonic anlage in this remarkable ovum was extraordinarily small, the embryo measurmg, as estimated from the sections, 190 /i in length. I quote, as Peters has done, the description which Graf Spee has given of the cavity of the o\Tiin. " The entire cavity of the ovum enclosed by the chorionic eetoblast [trophoblast shell, cytotrophoblast, and spongiotrophoblast] is filled up to the cavities of the embryonic anlage with mesoblast. This latter is very irregular as regards its possession of mesoderm cells. These are more numerous in the mesoderm layer resting upon the chorion, this consisting of two or at the most four cell-layers, presenting a greater thickness only in the region where the embryonic anlage occurs." Exception must be made of those places where there occur the mesodermic rudiments of the villi, already mentioned. " The more central portions of the egg cavity are very poor in cellular elements. Only scattered tracts of spindle-shaped mesoderm cells traverse it. In the intervals there is a feebly staining fibrogranular mass which occupies most of the cavity. The cellular tracts frequently unite with the mesoderm enclosing the embryonic anlage and with that of the opposite wall of the ovum." Spee has found similar cords in all younger human ova. " Those portions of the trophoblast which come into relation with the tissues of the uterine wall and take an active part in the implantation of the ovum and in the excavation of the egg chamber have been termed trophoderm by Minot (Charles S. Minot: The Implantation of the Human Ovum in the Uterus, Trans. Americ. Gynaecol. Soc, 1904). OVA AND VERY YOUNG EMBRYOS. 29 "The embryonic anlage (cut obliquely in the preparations) shows two very small epithelial cavities ( amnion and yolk sack), completely surrounded by mesoderm and contained within a thickening- of the chorionic mesoblast. The amniotic cavity is completely closed. Its wall is differentiated into a very thin amniotic membrane, lying nearer the surface of the ovum, and a plate con- sisting of high cylindrical cells, the germinal disc (germinal shield, embryonic shield). Between these and the wall of the yolk sack, composed of entoderm cells occasionally difficult to recognize, a layer of mesoderm is interposed. At one (the cranial) end (section 49 [-±4, 43?]) the cellular portion of the mesoblast does not appear to reach the median line. It lies on the yolk sack and is separated from the ectodermal territoi-y by a ' membrana prima.' This mem- brana prima (Hensen) always develops as a fine contour at the boundary between the ectoblast and mesoblast. It extends across the middle line in the preparations. . . . The series of sections clearly reveals the relations of most portions of the embryonic anlag-e. One end of the series only presents difficulties in the way of observation, partly on account of the unfavorable plane of the sections and partly, perhaps, on account of some complications in this region; for in- stance, it is impossible to determine the continuity of the ectoblast and mesoblast in some sections, and the condition of the yolk sack cannot be made out in this region." Spee considers this to be the caudal end. "An isolated cord con- necting the embryonic structures with the chorion cannot be said to exist, since almost the entire embryonic anlage seems to be imbedded in a thickening of the chorionic mesoderm. Whether the first small rudiment of an entoblastic diverticulum (the allantoic duct) has begun to bud out and is represented by a ring of epithelium-like cells arranged around a lumen, is altogether uncertain." Grosser figures a section through the ovum in situ (see Fig. 96, from Peters's figure) in the chapter on the development of the egg membranes and the placenta. I have reconstructed the embryonic structures from plates left by Selenka and found neither an allantoic nor an amniotic duct. The surface of the yolk sack appeared warty, as if the blood and vessels were beginning to form upon it; naturally, the wax plates and the model give no definite informa- tion on this point. The ovum which Graf Spee has described, unfortunately only briefly (Verh. deutsch. Ges. Gyndk., 1905, pp. 421-423; compare Schwalbe's Jdhresh., n. F., vol. xi. No. 2, p. 241), comes nearest to that of Peters. In a woman poisoned by oxalic acid one of the swollen areas of the mucous membrane on the ventral wall of the uterus immediately in front of the opening of the right tube was markedly prominent and its depressed summit showed a distinctive coloration. In it was found, in an egg capsule of 1.5 X 2.5 mm. diameter, an ovum poorly provided with villi and with a very small embryonic anlage that was surrounded by a quantity of blood. The summit of the egg capsule showed an implantation open- ing, which was 0.8 mm. in diameter and covered by a very flat blood-clot; in its neighborhood the villi were more numerous than elsewhere. Very near is also an ovum studied quite recently by Ph. Jung"^; it was obtained from an abrasio mucosae. The preparation is splendidly preserved and apparently normal. Jung confirms in general the observations of Peters and Spee; comparatively little attention was devoted to the embryonic structures, and it is very desirable that these should be thoroughly studied. The cavity of the ovxun measured 2.5 X 2.2 mm. The Von Herff ovum described by Graf Spee " takes its place in the series here. It was expelled after a menopause of five weeks on the second " Ph. Jung : Beitrage zur f riihesten Eieinbettung beim menschlichen Weibe, with 20 figs, on 7 plates, Berlin, 1908. " Graf von Spee : Neue Beobachtungen liber sehr f rlihe Entwieklungsstuf en des menschlichen Eies, Arch. f. Anat. u. Physiol., Anat. Abt., 1896. 30 HUMAN EMBRYOLOGY. day after a severe attack of influenza, probably as a result of the illness, and was apparently normal; it was throughout richly supplied with villi. Spee's remark that the diameters of the egg capsule, which he believes must have been really about 7 and 5i mm., were actually greater than these, is based upon the fact that the capsule was strongly distended with blood. Since, however, the periphery of the ovum must have reached the maternal tissues such a condition cannot be regarded as normal, but must have arisen shortly before or during the abortion. Spee estimates the diameter of the space within the chorion at barely 4 nun. The thickness of the chorion was 0.9 nun. j the villi measured 0.16-0.18 mm. at the base and were separated from each other by intervals of 0.2-0.78 mm. where the distance could be determined. The villi were covered by a double layer of cells, the Langhans layer and the syncytium, the latter, although not so well developed as in later stages, being nevertheless quite distinct. Both layers are regarded as differentiations of the ectoblastic trophoblast shell, as cytotrophoblast (the Langhans layer) and spongioblast (the syncytium). The embryonic structure had the form of an elongated thick papilla, attached at only one of its extremities to the chorion and elsewhere projecting quite freely into the interior of the cavity of the ovum (that is to say, into the periembryonic mesoderm space, the exocoelom of Selenka). Its long axis cuts the chorion at a very acute angle. A superficial furrow marks off on the papilla two elliptical portions. The larger of these forms the free pole of the papilla and proved to be the relatively vei-y large j'olk sack; the smaller one contains, on the surface which lies close to the chorion, a completely closed cavity, which was the amniotic ca\-ity with its ectoblastic lining, but for the rest it is a compact cord composed of mesoderm, which extends from the mesoblastic cover- ing of the yolk sack to the chorion, surrounding almost three-fourths of the amnion, so that this structure seems to be sunken into it. This part is the actual belly stalk and the sole connection with the chorion. In it there was an allantoic duct extending from the yolk sack. The portion of the ectoblastic lining of the meduUo-amniotic cavity that rests on the yolk sack consisted of cylindrical cells and formed a thick plate, evidently the embryonic shield (the germinal disk). The plane of the embi-yonie shield is somewhat perpendicular, that is to say, radial, to the surface of the chorion, the head end being nearest it. In the model the embryonic shield presents an oval outline and a median furrow lying between lateral portions which are convex dorsally and are somewhat unequal in size in the transverse direction. At the same time the dorsal surface of the shield is adapted to the form of the amniotic cavity and is, on the whole, concave. Spee gives the following measurements : " Direct measurements of the embi-yonic papilla : Longest diameter, 1.84 mm. ; diameter through the constricted portion, 0.475 mm. Almost perpendicular to these the longest diameter of the yolk sack is 1.054 mm. The amnion together with the belly stalk measures 0.76 mm. ; the greatest leng-th of the two latter structures is 0.76 mm. ; the greatest breadth of the yolk sack, 1.083 mm. ; its thickness about the same. "Measurements taken on the model (divided by 100 and so reduced to the actual size) : Length of the gemiinal disk, 0.37 mm. ; its breadth, 0.23 mm. (this is the ectoblast plate of the germinal disk) ; height of the amniotic cavity, up to 0.34 mm.; thickness of the belly stalk together with the amnion, 0.62 mm.; length of the allantoic duct, 0.35 mm.'' An amniotic duct or cord was not present. The entire anlage of the germinal disk (embryonic shield) was apparently, according to Spee, only a portion of the primitive streak region, notwithstanding that the typical fusion of the ectoblast and mesoblast could not be recognized in the sections, probably as a result of the preparation. No trace could be found of a differentiation of the medullary plates or of the chorda. " The primitive streak region extends right up to the cranial end of the germinal disk" (the embryonic shield). OVA AND VERT YOUNG EMBRYOS. 31 The waJls of the yolk sack seem to have advanced further in development than any other part of the embryonic anlage. The lining of its cavity is through- out smgle-layered and formed by cubical cells. Its mesoblastic covering forms irregular elevations and knobs, which project like small papilte, especially over the pole that is turned away from the embryonic shield. In each papilla a blood island was interposed between the mesoblast and entoblast and produced a bulging of the mesoblast, but very little irregularity of the entoblast. The formation of blood islands ceased at a much less distance from the embryonic shield m this ovum than in Von Spee's Glaevecke embryo, to be described later. The youngest stages of the blood islands lay nearest the embryonic shield- the oldest, at the distal pole of the yolk sack. ' Near to this Von Herff ovum— indeed, according to the opinion of its finder somewhat younger— is that which Beneke " found in a curetting done for therapeutic reasons. The curetting was made March 30, 1903, the last menstruation having lasted from March 5 to 10. No cohabitation had occurred after March 22. The cavity of this ovum was 3.8 mm. long, 2.2 mm. broad, and 1.2 mm. high; and the embryo itself had a length of 1.74 mm.," its greatest thickness in the dorsoventral diameter being 0.6 mm. The medullo-amniotic cavity was elongated caudally in a fusiform manner and was connected with the chorionic ectoblast by a cord of epithelial cells. It is stated that a typical medullai-y epithelium was already present in the anterior part of the germinal disk, and mention is made of a head process and of a chorda-like mass of cells. A neurenteric canal was present, but an allantoic duct was " not distinct." Although' the statements m the brief description are not always as clear as could be desired, yet it seems to me, from the presence of a canalis neurentericus, of an amniotic duct or cord, and of an anlage of the meduUa at the anterior end, that the embryo is more developed than that of the Herff ovum described by Spee. ^ A thorough study of the ovum is expected and when its results appear more definite conclusions will be possible; a thorough description may also make clear the meaning of certain peculiar structures that have been taken for blood-vessels, but which I shall not discuss here. The embryonic anlage of an ovum described by Carlo Giacomini'° was probably in about the same stag-e of development as the Spee embryo Von Herff; but on account of its poor preservation it would not have deserved mention here were it not that Giacomini states that it was expelled eleven days after a single cohabitation, so that its age may be estimated at nine or ten days, an estimate that does not agree with the conclusions of Bryee and Teacher that have been thoroughly discussed and reproduced above. The occurrence may also be noted of a small duct that opened on the surface of the chorion near the point of fixation of the embryonic structures and has been identified by Marchand " as the remains of an amniotic duct. A young ovum described by Mall " may also be mentioned here, although since it possessed a well-developed "Beneke: Ein sehr junges menschliches Ei (Ost.-Westpreuss. Gesellsch. f. GynakoL), Deutsche med. Wochenschrift, Jahrg. xxx, 1904; and Mitteilungen und Demonstrationen mit deni Universalprojektionsapparat iiber ein sehr jimges menschliches Ei, Marburger Sb., 1908, pp. 29-38. "^ Surely a misprint. ^° Carlo Giacomini : Un uovo humano di 11 giorni, Giomale della Reals Academia di Medicina di Torino, vol. iii, anno 60, Fasc. 10-11, Torino, 1897. " r. Marchand : Beobachtungen an jungen menschlichen Eiem, Anat. Hefte, vol. xxi, 1903. "Franklin P. Mall: A Contribution to the Study of the Pathology of Early Human Embryos, Johns Hopkins Hospital Reports, Festschrift for Welch, vol. ix, 1900. Also Joum. of Morph., vol. xix, p. 144, 1908. 32 HUMAN EMBRYOLOGY. allantoic duct, it may have been somewhat older. Its diameter favors this view. Its long diameter was 10 mm. and its short one 7 mm., and, like the Reichert ovum, it had villi only axound its greatest circumference, two areas being thus bare. The vUh were 0.5-0.7 mm. long and were branched. Mall now regards the ovum, probably eoiTectly, as being pathological. A very interesting ovum, similar, but probably in a slightly older stage of development, is described by Siegenbeek van Ileukelom,'" who unfortunately considers the embi-yonie structures only casually. It was obtamed from a woman who received some burns during an epileptic attack and died six hours later. An autopsy was performed fourteen hours after death. With the exception of the bm-ns the body showed no noteworthy departures from the normal. The entire uterus was placed in 3 per cent, formalin for some days and was then washed and fixed in alcohol. Since the ovum was collapsed as the result of a slight tear, accurate measurements could not be made; but Van Heukelom estimates its meridian at about 16^ mm., which would give a diameter of 0.1 mm. The ovum was completely covered mth villi; the insertion of thirty- one of these could be counted in a meridional section, fifteen occurring on the basal portion of the section (the portion nearest the uterine wall), twelve at the opposite pole, and two on each side at the equator. The villi of the embryonic pole were better developed than those of the opposite pole, and those at the equator were " especially heavy and thick." They varied greatly in length, some being small and short, the majority 0.75 mm. long and those at the equator as much as 1 mm.' Some were little branched, others, especially the basal and equatorial ones, had many offsets, and some divided into numerous branches. The epithelial covering of the villi was composed of two layers; the outer showed no cell limits, so that we have to deal again with the Langhans layer and the syncytium, a cytotrophoblast and a spongiotrophoblast. All the large branches of the villi were connected peripherally by means of epithelial trabeeulse (ectoblast trabeculse, cell columns) to form an epithelial shell (ectoblast or trophoblast shell) provided with large and small spaces. This shell was often very thin, consisting of only a single cell layer; at other times it was very thick and in its thicker parts there were peculiar blood laeunse, resting directly upon the maternal com- pacta. The remaining conditions resembled in essence those of the Von Herff ovum of Spee. At the basal wall of the ovum was the embryonic papilla (the embryonic structures) united to the chorion at a very acute angle by a stalk. A distinct allantoic duct was present and the amniotic cavity was lower than in the Von Herfif ovum described by Spee. Only one of the sections through the embryonic shield is figured, and this is reproduced here as Fig. 11. In it, in the region of the primitive streak, a mass of cells is seen projecting beyond the surface of the shield; Van Heukelom suggested that this might be Hensen's node, a suggestion which I, as well as the author, would mark with a large note of interrogation. Nothing is said of an amniotic duct. Somewhat older again is the ovum that I''" described in 1890; it contained an embryonic shield with a well-developed primitive streak and was the first of this stage to be described. The ovum was expelled in an abortion with the egg capsule, which was lenticular in shape and measured 12 X 9i X 7 mm. ; it had a scar measuring 2i mm. Like all human ova of this period it was easily separated from the capsule, and measured 8i X 7f X 6 mm. It had two areas free from villi : a larger one, measuring 6 J X 5| mm., at the pole opposite the embryonic anlage, that is, the pole toward the lumen of the uterus; and a smaller one at " Siegenbeek van Heukelom : Ueber die menschliche Plazentation, Arch, f . Anat. u. Physiol., Anat. Abt., 1898. "" Franz Keibel : Ein sehr junges menschliches Ei, Arch, f . Anat. u. Physiol., Anat. Abt., 1890. OVA AND VERY YOUNG EMBRYOS. 33 the embryonic pole situated excentrically in front of the point of attachment of the belly stalk and measuring 2 mm. in diameter. The tissue of the Reichert scar showed neither glands nor blood-vessels, nor was it bounded by epithelium The structure of the chorion and villi was the same as in the Spee ovum Von Hertt, except that the syncytium was more strongly developed. The embryonic shield was about 1 mm. long and showed in the sections a well-developed primitive streak The yolk sack had a diameter of 1 mm. and showed numerous blood and blood- vessel anlagen. The endothelial walls of the blood-vessels were already clearly distmguishable from the blood-corpuscles. There was an allantoic duct. i!i^^>E=^^ Fig. 11. — Section through the basal portion of the ovum described by Siegenbeek van Heukelom, with the embryonic structures (embryonic papilla), a, intervillous space; 6, villus; c, ectoblast mass of a villus not included in this section; d, outer, epithelial layer of chorion; e, inner, mesoblastic layer of chorion (parietal mesoblast); /, clot and granular deposits; g, mesoblastic stalk; h, part of the stalk with many cells; i, amnion; k, embryonic shield in cross-section (ectoblast); I, mesoblast; m, hypoblast; n, hypoblast of the yolk sack; o, visceral layer of mesoblast; p, maternal blood. (After Siegenbeek van Heukelom: Arch. f. Anatomie u. Physiol., Anat. Abt., 1898.) The ova described by Merttens/' by Leopold " in his Atlas, by Marchand and by Rossi need be mentioned here only in so far as they present especially striking- peculiarities. Merttens had for study only four sections found accidentally in the investigation of a uterine curetting. Leopold in his ovum obtained by opera- tion and estimated by him, " with doubtful accuracy," according to Spee, to be seven to eight days old — it was undoubtedly older — found no distinct embryonic " J. Merttens : Beitrage zur normalen und patholog. Anat. der menschl. Placenta, Zeitschr. f. Geburtsh. u. Gynakol., vol. xxx, 1894, and vol. xxxi, 1895, '°' G. Leopold : Uterus und Kind, Leipzig, 1897. 3 34 HUMAN e:\ibryology. papilla, either on account of unsatisfactory preservation, as Van Heukelom sug- gests, or, as Spee believes, because the ovum was abnormal. At all events the ovum, whose diameter was 4X3.7 mm., need not be further considered here. Concerning the ova described by Marchand (1898) I may remark briefly that that author describes (Marburger Sb., 1898, pp. 150-153), in an imperfectly presen-ed ovum of the size of a pea, at that portion of the surface of the chorion where the remains of the embryonic anlage occurred, a funnel-like de- pression of the surface of the chorion which led into a canal filled with syncytium; this he interpreted as the remains of an amniotic duct.^ From two later publications by Marchand (" Beobachtungen an jungen menschlichen Eiern,"' Anat. Hefte, No. 67, vol. xxi, 1902, pp. 217-278 ; and " Einige Beobachtungen an jungen mensch- lichen Eiern," Verh. Anat. Ges., 1902) it may be noted that he found in one ovum ''* that the intervillous space was not filled with maternal blood, notwith- standing that the blood-vessels of the neighboring portions of the mucous membrane were ex- cessively engorged. The inter- villous space seemed to be sepa- rated from the cavity of the egg chamber peripherally by strong ectoblastic proliferations; blood- vessels opening into the cavity^ of the egg chamber could not be found. From this it would fol- low that the blood must normally enter the intervillous space only at a later period of development; a precocious entrance of the blood may, according to Marchand, lead to an abortion. Also, according to Rossi Doria,^ who described an ovum of less than 9X8 mm. contained in a completely closed egg capsule, an actual circulation of Fig. 12, A and B. — Figures of the embryonic shield of the Frassi ovum, from a model by Elze. The belly stalk, together with the portion of the chorion to which it is attached, is shown to the right; the yolk sack and amnion have been removed. In A one is looking down upon the shield; at about its middle is the dorsal opening of the canalis neurentericus, to the left of this is the shallow medullary groove flanked by elevations with indistinct boundaries, to the right are the primitive streak and primitive groove. The plane of the sec- tions is shown by an arrow. B shows the shield from the left side. In the belly stalk the allantoic duct has been exposed so that its origin from the yolk sack can be seen. X 25. (Fig. 12, B, from Frassi: Arch. t. mikr. Anat., 1908, vol. Ixxi, Fig. 9.) ""In a later publication (Anat. Hefte, 1903, p. 223) he says: "We have to do therefore with a narrow canal traversing the chorion, which from its opening at the surface is lined or, more properly, filled by a prolongation of the so-called surface epithelium, and ends blindly a short distance below the inner surface, just where the remains of the embryonic anlage occur." He ascribes the same significance to a depression on the surface of the chorion of an ovum of 14 X 3 mm. diameters, which was laterally compressed by a blood-clot. ^ The ovum was obtained from the body of a woman who had died as the result of a gunshot wound. It was studied in situ, but the ovum and egg capsule were folded; the latter, according to Marchand, had a length of about 1.5 cm. and a breadth and depth of about 5-6 mm. The entire ovum was covered with branched villi. The embryo was completely disintegrated. ^ Tullio Rossi Doria : Ueber die Einbettung des menschlichen Eies, studiert an einem kleinen Ei der zweiten Woche, Arch. f. Gynakol., 1905, vol. Ixxvi, pp. 433-505. OVA AND VERY YOUNG EMBRYOS. 35 the blood does not take place at the beginning of the second week of develop- ment, because at that time the maternal blood has not yet gained access to the intervillous space; he regards the so-called prickle processes on the surface of the syncytium as a deposit formed from degenerated blood-corpuscles and the "scar" of the egg capsule as formed by regressive changes of the summit of the reflexa. Nothing is stated concerning the embryonic anlage, and the entire ovum was apparently little favorable for the settlement of important questions. We may now consider an ovum obtained by operation and studied by Frassi '" under my direction. The entire uterus, removed per vaginam, was at once placed in a warm 5 per cent, solution of formalin; it remained there forty-eight hours and was then washed for twelve hours and finally passed through alcohols of gradually increasing strength. Only then was it opened, cut into portions, and these imbedded in celloidin, in which condition it came into the hands of Frassi. The ovum, together with the portion of the uterus that contained it, was cut into serial sections. The ovum and embryonic structures were undoubtedly normal. ^i?'^^"^^^^*^, Fio. 13. — Section of the embryonic anlage of the Frassi ovum, talcen 30 f cranial to the dorsal open- ing of the neureuteric canal; the head process is cut obliquely and the ventral opening of the neurenteric canal practically tangentially. X 50. (From Frassi: Arch. f. mikr. Anat., vol. Ixxi, 1908.) The diameter of the egg capsule parallel to the surface of the uterine lumen, in the plane of the sections, was 13 mm., perpendicular to this surface it was 5 mm. at the middle of the ovum; the corresponding diameters of the cavity of the ovum were 9.4 and 3.2 mm. A scar could not be detected in the decidua oapsularis. The ovum was completely covered with villi, which were especially developed in the equatorial zone; their length varied between 0.5 and 1.9 mm. Both blood-vessels and glands opened into the intervillous space, but with regard to the latter it could be perceived that they had been laterally eroded, so that their communication with the space was secondai-y. It is remarkable that practically no blood was contained in the intervillous space, notwithstanding that blood- vessels opened into it; it must be that the blood had completely escaped during the operation. The Langhans layer, syncytium, and cell columns were present; and of these the Langhans layer and the cell columns may be regarded as cyto- ^ L. Frassi : Ueber ein junges mensehliches Ei in situ, Arch, f . mikr. Anat., vol. Ixx, 1907; and Weitere Ergebnisse des Studiums eines jungen mensehliehen Eies in situ, ibid., vol. Ixxi, 1908. 36 HUIMAN EilBRYOLOGY. trophoblast and the syncytium as spongiotrophoblast. The embryonic shield was cut somewhat obliquely; it showed the anlage of a well-developed primitive streak, at the anterior end of which was a neurenteric canal and at the posterior end the cloaeal membrane. The section shown in Fig. 14 passed directly through the neurenteric canal. In front of the primitive streak is a flat medullary groove, bounded by still indistinct medullary folds. Anlagen of blood and blood-vessels occurred on the yolk sack. Anlagen of blood-vessels could be seen with certainly in the mesoderm of the chorion only in the neighborhood of the insertion of the Ch. Fio. 14. — Section of the embryonic anlage of the Frassi ovum through the dorsal opening of the neurenteric canal. At the opposite pole of the yoli sack are anlagen of blood and of blood-vessels, and in addition a small cyst lined with ccelomic epithelium; it is shown more highly enlarged in Fig. 14a. Over the amnion is a portion of the chorion (Ch.) with a cut origin of a villus. X 50. (From Frassi: Arch, f. mikr. Anat., vol. Ixxi, 1908.) belly stalk; none could be detected in the mesodermal axes of the villi. Models were made of the embryonic structures as well as of the embryonic shield, but only those of the shield need be figured here, together with some of the sections. The measurements, made on the model, were: 1. Length of the embryonic shield 1,17 mm. 2. Breadth of the embryonic shield 0.6 mm. 3. Length of the primitive streak 0.5 mm. 4. Diameter of the yolk sack, a. Greatest 1.9 mm. 6. Least 0.9 mm. The embryonic structures were attached to the inner surface of the chorion by a typical belly stalk, in which vessels could be made out. We come now to the ovum Gle (Glaeveeke), the careful study of which by Graf Spee has done so much to advance our knowledge of human embryology. It was an aborted ovum that was expelled, together with the entire uterine mucous OVA AND VERY YOUNG EMBRYOS. 37 membrane, five weeks after the cessation of the menses. The diameter of the egg capsule parallel to the surface of the decidua was 10 X U mm., and perpendicular to this, the thickness of the decidua basalis being included, 7.2 mm. The ovum was everywhere, but not very thickly, covered with villi. It was somewhat oval. '^ f« es ® (lb ■®-®" ®' — D.E. — C. E. ©S®.®^ ?.. !,». '©.:® •"■© ® iS«a>*. -E.d.C. -v^.- ^;j^ <_® ms -' an Fig. 31. — Young Trockophura of Polygordius in which the body is beginning to grow out. an, anus; m, mouth; ms, mesoderm; sp, apical plate. (Simplified from Hatschek, from Jablonowski: Anat. Anz., vol. xiv, 1898.) an Fig. 31a. — Older Trochophora of Polygordius. The body region has grown larger and a number of segments have formed in the mesoderm, an, anus; m, mouth; Tns, mesoderm; sp, apical plate. (From Jablonowski; Anat. Anz., vol. xiv, 1898.) but a budding zone is formed in the region of the blastopore from which the segments of the vertebrate body are budded off. At this stage a comparison with the Trochophora (Fig. 31), a widely distributed larva among the annelids and molluscs, is quite possible, as Kopsch '^ and J. Jablonowski " have shown. Just as we can distinguish in the body of the larva an anterior unsegmented portion " This has already been done quite logically by Lwoff (B. Lwoff : Ueber einige wichtige Punkte in der Entwicklung des Amphioxits, Biol. Zentralbl. vol. xii, 1892; and Die BUdung der primaren Keimblatter und die Entstehung der Chorda und des Mesoderm bei den Wirbeltieren, Bull. Soc. Imper. des Natural, de Moscou, 1894) in the case of Amphioxus. He regards as the principal result of bis in- vestigations the conclusion that " in Amphioxus the invagination is in no wise to be regarded as a simple gastrulation, as has hitherto been done. There are, rather two different phases to be distinguished in it: in the first place, the invagination of the entoderm cells, from which the intestine is formed; and, in the second place, the invagination of ectoderm cells from the dorsal transition border, which form the ectoblastogenic anlage of the chorda and mesoderm." '" Kopsch : Gemeinsame Entwicklungsformen bei Wirbeltieren und Wirbellosen, Verb. Anat. Ges., 1898. " J. Jablonowski : Ueber einige Vorgange in der Entwicklung des Salmoniden- embryos und ihre Bedeutung fiir die Beurteilung der Bildung des Wirbeltierkorpers, Anat. Anz., vol. xiv, 1898. GERM LAYERS AND GASTRULATION. 51 which has been formed by the gastrulation process, and a posterior segmented portion which owes its existence to a budding process that succeeds gastrulation, so too is it possible in the vertebrate embryo. Hubrecht consequently distinguishes between cephalogenesis and notogenesis ui the development of vertebrates. By eephalogenesis the anterior unsegmented portion of the vertebrate body is formed as a result of gastrulation; by notogenesis, a process of budding, the succeeding portion is formed. One must not, however, misunderstand these expressions. The limits of the two portions of the body must not be sought where the head now joins the trunk; trunk segments in unknown number have been taken up into the head and " the question concerns, on the one hand, only the most anterior region of the head, to which the olfactorius and opticus belong; and, on the other hand, the remaining portion of the brain, together with the base of the skull with the remains of the chorda and the visceral arches, and the entire trunk." Fig. 32. — a, gastrula with open blastopore; b, a radially symmetrical actinian-like ccelenterate dia- gram- c, a bilaterally symmetrical, elongate, worm-like, and actinian-like animal with stomodeeum and gut pouches; d, a worm-like protochordate with differentiation of a head, trunk, and chorda, and with a beginning metamerism. (From Hubrecht: Anat. Anz., vol. xxvi, 1905, Figs 3-6.) While it is far from my intention to attempt a derivation of the vertebrates from the actiniae at all directly, yet it may be well to consider the possibility of derivation from such simple forms. With this object I give here Hubrecht's speculations (Quart. Journ. Micr. Sei., vol. xlix, p. 410) (Fig. 32) : " Once the didermic gastrula-stage reached, a second phase of ontogenetic development is inaugurated which is also of high phylogenetie importance. In this phase the bilaterally symmetric metameric animal gradually appears which we have to com- pare with possible phylogenetie transition forms that have connected the Vertebrates with radially symmetrical ancestors. This attempt at a plausible and rational reconstruction of the Vertebrate ancestry is, of course, hampered by the circum- stance that no trace of those forms is any longer in existemee. Still, an actinia-like, vermiform being, elongated in the direction of the mouth slit, imposes itself upon our imagination, such as has served for the theoretical speculations of Sedgwick on this same subject, and has once been accepted by van Beneden for the precursors 52 HIBIAN EMBRYOLOGY. of the Chordata." Hubrecht has pointed out that " the processes of growth by which a Coelenterate gastrula becomes fixed aiid gradually changes into a sessile Actinian can hardly be looked upon as protracted phases of gastrulation. This will be more difficult yet when the animal has already acquired a higher degree of complication than that of the Coelenterates, and swims about in the shape of a worm-like, lower chordate animal. We know of Polygordius and of other primi- tive worm types that to the radial, didermic larval stage — the Trochophora — an- other developmental phase succeeds, during which we observe proliferation in the anal region, leading to an increase in the distance between the anus and the apex of the metamerical worm, the latter budding off, so to say, from the radial trochophora. " We find similar processes in the Vertebrates, but without a free trochophora larva, and to this latter radial and didermic primitive stage corresponds in the Craniata the rapidly passing earliest phase in which delamination calls forth two germinal layers. Both in Elasmobranchs and in mammals we notice that the cellular material which is present in those very earliest stages contributes especially — as it does in the trochophora — towards the formation of the anterior part, the head, and that, following upon this, a proliferation-process is inaugurated (com- parable to the origin of the metamerieal worm out of the trochophora larva) by which the notochord and the somites, i.e., the bilaterally symmetrical metameric animal, are called into existence." Hubrecht, going far back in the phylogeny, compares this proliferation with the growth of an elongated actinian. He imagines the coelom pouches of the actinian, still in communication with the intestine, to represent the somites; the nerve ring on the oral disk would represent the spinal cord; the stomodaeum, the chorda ; and the actinian mouth, which is not to be regarded as its primitive mouth or blastopore, but is a secondary formation, represents the primitive streak, that stands in such intimate relation to the chorda (that is, to the actinian stomodseum). The ingerent opening of the actinian gastrula elongates to form the actinian mouth that leads into the stomodeeum; the blastopore of the mammalian gastrula, which never opens, elongates posteriorly to form the primitive groove, whose floor, the primitive streak, furnishes the material for the chorda. " There is, then, during ontogeny an unbroken continuity between the blasto- pore of the Actinian and its oral slit, between the blastopore of the Vertebrate (often only potential in mammals and not identical with the opening that is called by that name in Sauropsids) and its primitive groove. A phylogenetic continuity has to be statuated between this oral slit of the Actinia and the peculiar spot (be- hind the so-called anterior lip of the blastopore) which on the Vertebrate embryonic shield gradually moves backwards and establishes in many cases an open communi- cation between a portion of the Vertebrate intestine and the exterior. The primitive streak, however, the solid material that proliferates downwards from the ectoderm, coalesces with the entoderm, and brings forth the notochord from its median (though really paired) portion and the somites from its lateral wings — this primitive streak can never be identified with a blastopore. For we have above attempted to demon- strate that in this primitive streak we encounter the material which, also in the Actinia, (1) proliferates downwards from the ectoderm and produces the stomo- dseum, (2) coalesces with the entoderm, (3) is in direct continuity with those parts which are preparing to give rise to coelomic pouches but are yet continuous with the primitive enteron." Hubrecht proposes a modification of the nomenclature in that for that portion of the vertebrate embryonic disk which he has compared with the actinian mouth and stomodaeum he suggests the use of the name " dorsal mouth " instead of primitive mouth, gastrula mouth, or blastopore. In this manner the contrast with the phyla of the annelids and molluscs will be more plainly brought out. GERM LAYERS AND GASTRULATION. 53 If now we summarize what has been stated above and apply it to the formation of the germ layers in the mammals, we must say: 1. In the mammals the entoderm (yolk and intestinal entoderm) is formed by delamination ; and it is only this delamination process that one can term gastrulation. The two cell-complexes which are formed as the result of gastru- lation are to be termed ectoblast (ectoderm) and entoblast (hypoblast, entoderm). That the superficial layer of the wall of the embryonic vesicle, for a time the only layer that is present, must also be regarded as a portion of the ectoblast seems self- evident; the so-called covering layer belongs to it also. In so far as it is con- cerned in the nourishment of the ovum it may be termed, following Hubrecht, the trophoblast. As trophoderm — in the sense in which that word is used by Minot — a portion of the trophoblast is again to be distinguished, which in certain eases produces the penetration of the ovum into the wall of the uterus. Morphologically the trophoblast may be divided into the cytotrophoblast, from which the Langhans layer of cells is formed in the human ovum, and the spongiotrophoblast (plasmodi- trophoblast), which gives rise to the syncytium. 2. A typical blastopore has not yet been certainly observed in the mammals. An invagination blastopore need not be expected to occur; some observations by which a eomiection of the ectoderm and entoderm was shown in early stages, before the formation of the mesoblast, have suggested the occurrence of a rudi- mentary blastopore. The primitive node and primitive streak — whether or not a primitive groove forms in the primitive streak is a matter of subordinate impor- tance — are certainly associated with the blastopore, but cannot be directly homol- ogized with it. 3. The anterior end of the primitive streak, at the time when the streak is at its greatest extension anteriorly, but not always its most anterior part, must be regarded as indicating the position of the blastopore. 4. The invagination processes by which m the mammals the formation of the chorda and mesoblast is initiated, are comparable to the corresponding processes in Amphioxus, but can be regarded as the gastrulation process neither in Amphioxus nor the vertebrates. We must accordingly say that the chorda and mesoblast arise from the ectoblast, the chorda partly directly and partly after it has be- longed to the mesoblast for some time. The enclosure of the chorda in the entoblast is an entirely secondary phenomenon. 5. Even although the region of the primitive streak, with or without a primitive groove, cannot be compared directly with the fused lips of the blastopore of a gastrula, nevertheless they must be regarded as modified structures that have arisen in association with a typical blastopore and must be compared with what it has been the custom to call the lips of the blastopore in the vertebrates, from Amphioxus upwards. The processes which are here concerned are comparable to the budding processes which in the Trochophora larva succeed the gastrulation processes. This becomes especially clear when one considers the later transforma- tion that the primitive streak undergoes and by which it becomes converted into the so-called " tail bud," a structure from which (as has already been pointed out) not only the tail arises, since the line between the tail and the trunk is merely conventional. The principal limit that concerns us here lies far forward in the head region, between those portions of the body which are formed by the gastrulation process and those that owe their existence to the succeeding proliferation process. The anterior end of the primitive streak, at the time of its greatest anterior extension, marks this limit. There result from these theoretical considerations certain conclusions that are of interest from the more practical side. A longitudinal splitting of the embryo, which may extend into the head region, may, under some circumstances, be regarded as produced by inhibition of growth, which can be referred to the primitive streak and the processes which take place in it. 5J: HmiAN EMBRYOLOGY. Roux's hemitheria anteriora such as the calf studied by Roux's pupil Eek- hardt," which Roux regarded as due to the early degeneration of the two seg- mentation cells which contained the anlagen of the caudal half of the body, are much more probably due to disturbances in the territory of the primitive streak. According as these occur early or late, a greater or smaller portion of the posterior extremity of the body will be wanting. Bob-tailed cats and dogs also belong to this class of developmental inhibitions; in these the inhibition first occurred after a portion of the tail had developed from the tail bud. Embryology throws light upon the occurrence of coccygeal tumors and their varied structure by the fact that in the " tail bud " an indifferent cell material is present from which all the germ layers may be produced. In conclusion, we may now return to man and endeavor, from what we at present know concerning his development and that of animals, to form a picture of the formation of his germ layers. We must assume that the human ovum, similarly to that of the guinea-pig, burrows into the mucous membrane of the uterus, destroys the maternal tissues, and so makes a cavity for itself. At the time of the penetration into the mucous membrane the diameter of the ovum can scarcely amount to 0.5 mm. Very early, but probably only after the ovum has burrowed into the mucous membrane, the formation of the coelom and of the germ layers begins. I assume that this formation can begin only at that time because the burrowing process would otherwise encounter diffi- culties from the enlargement of the ovum made necessary by the formation of the coelom and mesoblast; and it may also be supposed that those cells would first be formed which are intended for the destruction and absorption of the maternal tissue, the trophoblast cells, which belong to the later ectoblast complex. At all events, one must assume, as Spee pointed out as long ago as 1896, and as is now rendered almost certain by the Peters ovum, that at the time of the first formation of the mesoblast the diameter of the ovum does not exceed 0.5 mm. The formation of the mesoblast follows immediately upon that of the amniotic (medullo-amniotic) cavity and that of the cavity of the yolk sack and intestine. That the coelom in the human ovum is formed by a splitting process is indicated by the scattered mesoblastic strands which stretch between the yolk sack and the chorionic mesoblast, as well as by observations in many other mammals.' ° The human ovum is, consequently, to be assigned to the category of schizocoel ova. It is, however, not yet quite clear how the cavity traversed by scattered strands of mesoblast and lying between the yolk sack and the chorion in the Peters ovum is "Eekhardt: Ueber Hemitheria anteriora (Eoux), Dissert., Breslau, 1889. "^ This process may now be regarded as certain as a result of the observations of Biyce and Teacher (I.e.), which were published only after the first chapters of this " Handbook " had already been written and which must be noticed subsequently. GERM LAYERS AND GASTRULATION. 55 to be interpreted. It may be supposed to represent the extra- embryonic coelom ; but it may also be imagined that it bas arisen from an extensive loosening up of the tissue and not by a splitting of the mesoderm, and tbat the triangular space beside the caudal extremity of the embryo (see Fig. 96, p. 108, in the chapter on the development of the egg membranes and the placentation), wliich is lined with flat cells having an epithelial arrangement, is the first anlage of the ccBlom. As to the amniotic (meduUo-amniotic) cavity, it may be said with a probability bordering upon certainty that it arises by a splitting of a solid cell mass; consequently, amniotic folds never occur in man. The amniotic duct or cord — a connection between the epithelium of the amniotic cavity and the surface of the chorion, indications of which have been observed by Eternod and Marchand and more distinct evidence by Beneke in human ova and by Selenka in apes — appears to arise later, and, indeed, it is ques- tionable if it is of regular occurrence ; it is a phylogenetic memory from dim ancestral times. Selenka (1903) believes that this amniotic umbilicus, as he calls it, following Bonnet, does not reach complete development in apes ; that may also be the case in man; at all events it is wanting in the youngest stages yet observed, and, if it is of constant occurrence, its existence must be limited to a very short period of time. As the amniotic (meduUo-amniotic) cavity, so also the cavity of the yolk sack is formed by the splitting of an originally solid mass of cells ; in the Peters ovum it is still so small — smaller than the ripe ovarian ovum — that one can hardly imagine it to be formed by being enclosed by a surrounding epithelial lamella of entoblast. No important difficulties stand in the way of its origin by a splitting process. Graf Spee in his paper of 1889 makes use of ova which present an inversion of the germ layers for an explanation of the conditions which obtain in the human ovum. On the other hand, I maintained (1890) that the peculiarities of the human ovum were to be explained by the early formation of the extra-embryonic coelom and of the amnion, and that a sinking of the human embryonic anlage deep into the yolk sack, as in oVa with inversion of the germ layers, could not be imagined. In the endeavor to obtain a clear picture of the processes just discussed, the extraordinary minuteness of the embryonic structures must constantly be borne in mind. The diameter of the yolk sack in the Peters o\Tim is 0.19 mm., that is to say, it is not quite the size of a ripe human ovum.'" " According to Kolliker and Ebner (" Handbuch der Gewebelehre," 6 Aufl., vol. iii, 1902), the diameter of the ripe human ovum is 0.22-0.32 mm.; and Waldeyer remarks in Hertwig's "Handbuch" concerning this statement, that he has never seen human ova measuring more than 0.25 mm. 56 HU^ilAN EMBRYOLOGY. Elze and I have endeavored in our "Normentafel zur Ent- •wicklungsgescliiclite" to embody the results of the above con- siderations in a series of diagrams. Fig. 33, A, represents an ovum towards the end of segmentation as a whole object. Fig. 33, B-F, are sections, and it is assumed that all sections pass through the median sagittal plane of the future embryo. The figures are of such a size that they may be regarded as enlarged twenty-five times. The ovum shown in Fig. 33, A, is still sur- rounded by the zona pellucida; it may have just reached the uterus. B represents an ovum which has already eaten its way into the mucous membrane of the uterus. Four groups of cells are to be recognized in it. The periphery consists throughout of ectoblast cells — the trophoblast, represented in gray. Inside this trophoblast mantle are three cell-complexes. That which gives rise to the ectoderm of the embiyo and of the amnion is repre- sented in black; to the right, corresponding to the caudal end of the later embryo, we have left indistinct the boundary between it and the mesololast complex, represented in red, in order to indicate that in this region there is perhaps a transition between the two complexes. We have represented this connection in the diagram, because wherever the development of the mesoblast has been sufficiently studied, the greater portion of it, at least, is found to arise from the ectoblast complex, and we find later, even in human ova, a primitive streak. It is extraordinarily difficult to picture to oneself how the processes actually take place in man, owing to the minuteness of the human ovum at this stage. The meso- blast complex is everywhere forcing the entoblast complex, repre- sented in green, away from the trophoblast shell. Fig. 33, C, shows an older stage. By a process of splitting in the ectoblast and entoblast complexes, the amniotic (medullo- amniotic) cavity in the one case and the cavity of the intestine and yolk sack in the other have been formed, and in a similar manner the extra-embryonic coelom has formed in the mesoblast complex. We have already pointed out that another mode of formation of the ccelom is possible. In this case we must suppose that in this stage the extra-embryonic coelom is not yet formed and we must imagine the cavity inside the red, which is left white in C, to have a pale reddish tinge and to represent solid but very loosely compacted mesoblast. Around the amnion, the yolk sack, and at the periphery next the trophoblast, the mesoblast, which as a whole is still solid, is somewhat denser. The amnion at this stage may still be in contact with the trophoblast shell. Diagram D, Fig. 33, shows the conditions which obtain in the Peters ovum. The trophoblast mantle surrounding the ovum has developed lacunae which are filled with maternal blood, and the mesoblastic axes of the villi have begun to grow out into the GERM LAYERS AND GA STRULATION. 57 Fig. 33. — Stages in the development of germ layers. Gray, trophoblaat; black, embryonic and amniotic ectoblast; green, intestinal and yolk-sack ectoblast; red, mesoblast. a, cranial end; p, caudal end. X 25. (From Keibel and Elze: Normentafel, p. 13, Fig. 2a-f.) 58 HmiAN EMBRYOLOGY. trophoblast. At the caudal extremitj' the belly stalk has become distinct, but an allantoic duct is not yet formed. Whether or not a very small primitive streak was present in the Peters ovum must remain doubtful; we assume that the delimitation of the ectoblast from the mesoblast was not quite sharp at the caudal end, and this we take to be the anlage of a primitive streak. The embryonic coelom is represented as completely formed, although a somewhat different interpretation of the conditions in the Peters ovum has been mentioned. The ectoblastic covering of the embryo and the amnion are everywhere being forced away from the trophoblast by mesoblast cells. Diagram -E,- Fig. 33, represents a median sagittal section through an ovum of the stage seen in the Frassi ovum. The anterior half of the section, in the region of th'rembryonic shield, is occupied by the floor of the medullary groove; behind it is the neurenteric canal; and then, lying in the same plane as the anterior half of the embryonic shield, the region of the primitive streak, which occupies about half the shield. At the caudal end of the primitive streak the cloacal membrane is already recog- nizable. The chordTi is enclosed within the entoblast; anlagen of blood and blood-vessels occur in the yolk sack. An allantoic duct is present. Diagram F shows a median sagittal section through the stage seen in Spec's Glaevecke embryo. Especially to be noted is the recession of the primitive streak and the fact that the now quite short primitive streak is bent down at an angle to the plane of the cranial extremity of the embryonic disk. A cloacal membrane must have been present at the caudal end of the primitive streak, but it is not represented in the diagram because it was not observed in the Spee embryo, probably on account of the direction in which the sections were made. VI. SUMMARY OF THE DEVELOPMENT OF THE HUMAN EMBRYO AND THE DIFFERENTIATION OF ITS EXTERNAL FORM. By FRANZ KEIBEL, Freiburg i. Be. The first relatively satisfactory synopsis of the development of the external form of the human body is that given by His ^ in his "Anatomie menschlicher Embryonen" and in the Normentafel published with it. In the latter there is shown a series of human embryos dating from the end of the second week to the end of the second month. With this latter period the development of the embryo is so far advanced that the human in it is recognizable even to the laity; His' designates this as the embryonic period and that succeeding it up to birth he terms the fetal period. A com- prehensive account of the development of the body during the fetal period, with abundant illustrations, has been given by Gustav Retzius - in his memoir "Zur Kenntnis der Entwicklung der Korperform des Menschen wahrend der fetalen Lebensstuf en, " published in 1904. Disregarding studies of individual embryos there must also be mentioned here the ' ' Normentafel zur Entwicklungsgeschichte ' ' of Keibel and Elze,^ Carl Rabl's "Entwicklung des Gesichtes"* and the splendid heliogravures of human embryos that Hoch- stetter ^ has published. Those who desire a comparison of human development with that of animals I would refer to Hertwig's "Handbuch"" in which I have considered the development of the external form in vertebrate embryos. On account of the fundamental importance of His's "Ana- tomie menschlicher Embryonen," I here give a view of the ^W. His: Anatomie menschlicher Embryonen, Leipzig, 1880-1885. ' Gustav Retzius : Biologische Untersuehungen, neue Folge xi, 1904. ° Keibel and Elze: Normentafel zur Entwicklungsgeschichte des Menschen, Jena, 1908. * Carl Rabl : Die Entwicklung des Gesichtes, Leipzig, 1902. °r. Hochstetter: Bilder der ausseren Korperform einiger menschlicher Embryonen aus den beiden ersten Monaten der Entwicklung, Munich, 1907 (pub- lished by F. Bruckmann). °F. Keibel: Die Entwicklung der ausseren Korperformen der Wirbeltier- embryonen, etc., Hertwig's Handbuch, 1906, vol. i, chap. 6 (published 1902). 59 60 HU:\IAN E]\IBRYOLOGT. Fig. 34, o-p. — The embryos of His's Normentafel, from the Normentafel of Keibel and EIze (Fig. 1, p. 6). X 5. His's numbers are given in parentheses. DEVELOPMENT OF HUMAN EMBRYO. 61 development of human embryos as shown in His's Normentafel the first fifteen figures, as in the original, being enlarged five tmies, the remammg ones only two and a half times. Fig. 34, q~z. — The embryos of His's Normentafel, from the Normentafel of Keibel and Elze (Fig. 1, p. 6) X 2.5. His's numbers are given in parentheses. The individual embryos are lettered, His's numbers being given in parentheses. To this reproduction of His's Normentafel I append a synoptic table from which it may be seen how His designated each embryo, its size, and its age, as estimated by His. G-2 HUMAN EMBRYOLOGY. Fig. Xo. His'sdeejgnation. o (1) b (2) C (3) d (4) e (5) / (6) g (7) ft (8) i (9) tClO) ((11) m (12) n (13) 0(14) J (15) 9(16) r(17) 8(18) «(19) U(20) f (21) w(22) a; (23) y(2f) 2(25) Embryo E (VII) Embryo SE (VI) Embryo Lg (LXVIII) . Embryo Sch (LXVI)^.. Embryo M (IV) Embryo Lr (LXVII) . . Embryo a (III) Embryo R (LVII) . . . . Embryo A (II) Embryo Pr Berlin anat. collection. . Ruge's collection Embryo M (X) Embryo Br (XXIX).. Embryo Rg (LXXIV) . Embryo Sj (XXXV) . Embryo CH Embryo Schz (XLVI)... Ruge's collection Embryo Dr (XXXIV) Embryo Sj (XXXVI) Embryo XCI Embryo Ltz Embryo Zw Embryo Wt (LXXVII) Estimated age in days. L. L. L. L. L. L. Nl, Nl. Kl. Nl, Nl. Nl. Kl. Nl. Nl. Nl. Nl. L. L. L. L. L. L. L. L. 2.1 2.2 2.15 2.2 2.6 4.2 4.0 5.5 7.5 10.0 9.1 9.1 10.5 11.0 11.5 12.5 13.7 13.8 13.6 14.5 15.5 16.0 17.5 18.5 23.0 Uterus Uterus Uterus Uterus Extra-uterine 12-15 12-15 12-15 12-15 18-21 18-21 23 24-25 27-30 27-30 27-30 27-30 31-34 31-34 31-34 31-34 31-34 About 35 About 35 About 37-38 About 39-40 About 42-45 47-51 62-54 60 To avoid repetition I shall not proceed to describe His's Normentafel, but will consider a series of embryos which, in my opinion, present the best summary now available of the develop- ment of the external form of the human body; and in doing so I shall have occasion to consider the embryos of the Normentafel in their appropriate places. The so-called egg membranes will be described only in so far as they influence the form of the body. Nothing need be added here concerning the youngest human embryonic disks to what has already been said in the chapter on ' ' The Youngest Human Ova and Embryos. ' ' The embryonic disk of the Peters ovum had a length of 0.19 mm. ; whether it possessed a very small primitive streak must remain doubtful. In Spee's Von Herff ovum the disk had an oval outline and presented a median groove lying between the dorsally convex lateral portions, which were somewhat unequal in the transverse direction. This groove is the primitive groove, and it may be supposed that the primitive streak extends through the entire length of the embryonic disk, since Spee says: "The entire anlage of the embryonic disk is apparently only a portion of the actual primitive streak region." DEVELOPMENT OP HUMAN EMBETO. 63 Close upon the embryonic disk of Spee's Von Herff ovum fol- lows that of the Frassi ovum. In the former the primitive streak was at the height of its development, extending as it did throughout the entire germ; in the Frassi ovum it is about half the length of the embryonic disk, and at its cranial end it shows the anlage of a neurenteric canal, while at the caudal end, as has been noted (p. 36), the cinlage of a cloacal membrane has appeared. In front of the primi- tive streak there is a shallow medullary groove bounded by low medullary folds. The entire em- bryonic disk is slightly convex in the craniocaudal direction and from right to left, and it covers the yolk sack like a lid. ^'°- ^^■ Following this embryo that of Spee's Grlaevecke ovum (Fig. 37) may be mentioned. The actual embryo had a somewhat "constricted pear-shaped" out- line, and within this the outline of the biscuit-shaped medullary plate was distinctly marked. The caudal end of the embryo was bent sharply centrally, al- most at a right angle, and, consequently, when looked at from above, appears greatly foreshortened. Slightly cranial to this downwardly bent portion there is a somewhat circular swell- ing, that in position corresponds to Hen- sen's node and surrounds, like a low wall, a roundish triangular opening, the neu- renteric canal; behind this swelling is the primitive groove, resting upon the primitive streak. Anteriorly the primitive streak is embraced by the medullary folds. For the measurements of the embryo consult Chap. IV, p. 39. Next to this embryo comes embryo 1 of His's Normentafel. This embryo was obtained from an ovum that measured in the fresh condition 8J X 5i mm. and was completely surrounded by villi. The length of the embryo, including the belly stalk, was 2.6 mm., and without the belly stalk was 2.1 mm. The yolk sack was somewhat flattened and measured 2.3 X 1-6 mm. The embryo rested upon the yolk sack for about 2 mm. of its leno-th and was enclosed in an amnion which also surrounded the FlG3. 35 and 36. — The Frassi embryo (Nor- mentafel of Keibel and Elze, Fig. I, plate 1). Fig. 35 shows the embryo from above and Fig. 36 from the left side. Fig. 35 is from a model by Frassi, X 20; Fig. 36, from a model by Elze, X 25. Fig. 37. — Graf Spee's Glae- vecke ovum. X 20. (From the Normentafel of Keibel and Elze, Fig. II, plate 2. ) 64 HUMAN EMBRYOLOGY. \ upper surface of the belly stalk. In the embryo itself, the section- ing of which was unsuccessful, the medullary folds and medullary groove could be recognized, and laterally from the medullary folds at the anterior extremity the anlage of the heart was visible. I place next the embryo Klb. of the Normentafel of Keibel and Elze, Figs. Hid and Ilia. The ovum which contained the embryo was obtained by a laparotomy and both it and the embryo may be regarded as quite normal. The embryo had five to six pairs of primitive somites. The vertex bend had already begun to form, otherwise the embryo lies flat on the yolk sack; it has no dorsal bend. Fig. 38 shows it from the dorsal surface, and, the amnion being cut away, it is seen to be attached to the chorion, a small portion of which is represented, by a short belly stalk. To the right and left of the cut edges of the amnion the yolk sack projects beyond the embryonic struct- ures. Both the cra- nial and caudal ends of the embryo are separated from the yolk sack, but the middle portion is still spread out flat. The medullary groove is widely open, but rather deep. Caudally the medullary folds embrace the dorsal opening of the neurenteric canal and the cranial end of the primitive streak, which immediately bends downward so that it cannot be perceived in its entire extent; indeed, it has already undergone much retrogression. The well-marked medullary anlage shows the vertex bend, so that its cranial end cannot be seen in a dorsal view. The brain part of the medullary anlage shows a separation into three portions, the most caudal of which extends to about the fourth pair of primitive somites and then passes without any marked boundary into the spinal cord. I may here note that by far the greater part of the dorsal region of this embryo belongs to the later head region. The boundary between the head and trunk regions passes through the fourth primitive somite. The most anterior somite of the neck region is first differentiated. Fig. 39 shows the embryo from in front. The amnion has again been removed; ventrally the relatively large yolk sack is seen. Of the embryonic anlage only the ventrally bent portion, as far back as about the vertex bend, can be seen; but it may be observed that the most anterior portion of the brain anlage is Figs. 38 and 39 The embryo Klb. from the Normentafel of Keibel and Elze, Figs. Hid and Ilia. X 20. DEVELOPMENT OF HUMAN EMBRYO. 65 relatively greatly developed. The medullary groove terminates a little behind the cranial end of the medullary anlage, so that the latter shows anteriorly a transverse ridge. Nothing is yet to be seen of the optic pits, the forerunners of the optic vesicles. Regarding the measurements of this embryo Kromer makes the following statements : " The greatest length of the embryonic anlage from the anterior amniotic border of the head cap to the chorionic end of the belly stalk was 1.95 mm. ; the length of the embryo, without the belly stalk, from head to tail was 1.8 mm. ; the greatest breadth of the yolk sack was barely 1.2 mm. ; the breadth of the embryonic disk at the junction of the amnion and yolk sack (measured in an anterior view) was 0.9 mm." The size of the yolk sack was 1.1 mm. in height, 1.4 mm. in breadth, and 1.5 mm. in length. After this embryo, that represented in Fig. 2 of His's Nor- mentafel may follow, although on account of its possessing a dorsal flexure I am somewhat in doubt if it is normal in form. Its greatest length was 2.2 mm. The greatest di- ameter of the somewhat collapsed yolk sack was 1.9 mm.; and the embryo rests upon it in such a way that anteriorly the head end, for a distance of 0.4 mm., and posteriorly the caudal end, for a distance of 0.5 mm., project beyond the yolk sack navel, which is 1.3 mm. in length. The body of the embryo shows the middle of the dorsal surface somewhat depressed and the head and tail ends are somewhat bent downwards. The margins of the medullary plate are still wide apart, and a number of primitive somites, how many could not be exactly determined, were formed. The anlage of the heart was still paired, and upon the surface of the yolk sack there were numerous wart- like elevations (anlagen of blood-vessels). The BuUe embryo of Kollmann (Fig. 40) may well come in here or even after the embryo Klb. Both the head and tail ends project beyond the yolk sack, which communicates widely with the intestine, so that one cannot yet speak of a yolk stalk. The volk sack has been cut away, except for the part by which it is continuous with the embryo; and similarly the amnion has been cut away not far from its root ; under the caudal end of the embryo is to be seen the belly stalk. The figure shows the embryo from behind and from the right, so that the heart swelling is hidden. The brain portion of the medullary canal, in which the segments of the brain can be seen, is still open, and the caudal end of the canal is also open, although this cannot be perceived from the figure If we reckon three of the fourteen primitive somites as belonging to the head and eight to the neck region, there still 5 Fig. 40.— The BuUe embryo of Kollmann. X20. (From the Normentafel o f Keibel and Elze, Fig. IV, plate 5.) 66 HUMAN EMBRYOLOGY. remain three as representing the thorax; in the relatively small caudal end of the embryo almost all the segments of the trunk and tail must still be represented. In the region of the sixth to the fourteenth somites the dorsal surface of the embryo is slightly depressed. Kollmann says : "In the region of the sixth primitive somite the flexure of the dorsal region, which later becomes so striking, is noticeable." I cannot recognize in the figures of the embryo a flexure in the region of the sixth somite, and in any case the sixth segment belongs to the neck region. That a dorsal flexure occurs at this stage or later in normal human embryos I regard as disproved, and in this Spee^ is in agreement with me. In 1905 * I had come to the conclusion that even if a dorsal flexure normally occurs in' human embryos it can only be at a stage of development in which six or at the most twelve primitive somites are present. I do not, however, regard the evidence of its occurrence at this stage as sufficiently based upon the embryo that Eternod » has had modelled by Ziegler and upon an embryo with seven ])airs of primitive somites belonging to Graf Spee.i" It is a question whether in this and other cases one has not to deal with a de*- f ormity produced by swelling. (For further consideration of this matter see the Normen- tafel of Keibel and Elze, pp. 22-23.) The greatest length of the embryo, measured after it had been preserved in alco- hol, was 2.36 mm. The embryo Pfannenstiel III (Figs. 41 and 42), figured in Keibel and Elze's Normentafel, has the same number of primitive somites as the Bulle embryo of Kollmann, yet it is more de- veloped. It was obtained from a hysterectomy and was modelled by Elze. The medullary tube is open in the brain region as well as caudally ; but, in addition to the vertex bend, present in the embryo Klb., a nape bend has appeared ; there is no indication of the dorsal ' Schwalbe's Jahresbericht, Jena, 1906, neue Folge, vol. xi, 2 Abt., p. 225 (literature of 1905). " Keibel : Zur Embryologie des Menseben, der Affen und der Halbaffen, Verb. Anat. Ges. (Genf), 1905; also in C. R. Soe. des Anatomistes, 1905. " Compare A. C. F. Eternod : II y a un canal notocbordal dans I'embryon humain, Anat. Anz., vol. xvi, 1899. ^° Graf Spee : Mitteilungen iiber den Verein sebleswig-holsteiniseher Aerzte, vol. xi, article 8, 1887. Figs. 41 and 42. — The embryo Pfannenstiel III. X 20. (From the Normentafel of Keibel and Elze, Figs. Vr and Vv.) DEVELOPMENT OF HUMAN EMBRYO. 67 flexure. The optic pits were distinct and the auditory epithelium was slightly depressed. The greatest length of the embryo meas- ured 2.6 mm. The flexure which I have termed the nape bend was very striking in this embryo. If it be normal, it is very doubtful if the nape bend that is found in later stages is comparable with it, and if it does not later disappear before the well-known nape bend of later stages develops. Older embryos of mammals and also of man, such as embryo 3 of His's Normentafel, still lack a nape bend, as does also an orang-outang corresponding to this last embryo.^'- At this point a great gap unfortunately occurs in our series of human embryos, so that a settlement of the question is at present impossible. The embryo shown in Fig. VI of the Normen- tafel of Keibel and Elze had already twenty-three pairs of primitive somites; in addition, our ideas as to its external form are based upon a not very successful model, whose head region presents a very improbable form. I shall not, therefore, give a figure of the embryo here, but merely describe it briefly. It shows a well-marked vertex bend, but no distinct nape bend; and it is spirally coiled, so that its tail end comes to he to the right of the very large belly stalk. The medullary tube is still open at its caudal end. The heart swelling lies entirely in the amniotic cavity, and through its thin wall the heart must certainly have been shad- owed. Three branchial arches and as many branchial furrows were recognizable, and dorsal to the second furrow there was the already greatly narrowed opening of the ear vesicle. The extremities have not yet formed. Embryos 3, 4, and 6 of His's Normentafel, Fig. 34, c (3), d (4), and / (6) I regard as abnormal; c (3) and d (4) show the much' discussed dorsal flexure and/ (6) seems to me for other reasons to be excluded from the series. The embryo shown in Fig. 34, e (5), seems to me more worthy of confidence; and I repeat it here, more highly magnified, as Fig. 43. It came from an oviim which measured 7.5-8 mm. in diameter and was completely surrounded with villi. The amnion Fig. 43.- His's embryo M (His's Normentafel, Fig. 5 = Fig. 35e.) X20. " See r. Keibel : in Selenka's Menschenaffen, 9 Lieferung, 1906. 68 HIBIAN EMBRYOLOGY. lies close to the embryo and does not yet entirely enclose the heart. The body of the embryo is bent upon itself anteriorly and at the same time twisted about its axis so that the head end is turned toward the left and the pelvic end toward the right. The dorsal curvature is very regular, and the nape prominence is not yet pronounced. The anterior portion of the head is bent ven- trally to such an extent that the vertex is formed by the mid- brain. Behind the anterior part of the head there is a deep depression, which indicates the entrance of the oral sinus and is continued dorsally in the orbitonasal groove. Behind the mouth cleft is a broad mandibular process, separated by a groove from the second visceral arch; and the posterior boundary of this second arch can also be made out. The study of the entire embryo revealed no clear picture of the third and fourth arches, although their presence was quite evident in sections. The anlage. of the heart projected from the ventral surface of the body as a broad transverse swelling; its prolongation on the right side extends forward as the aortic bulb to the edge of the mandibular process. To the atrial portion of the heart belongs an outpouching which is seen ventral to the hinder portion of the head, on the lateral wall. Immediately behind the heart the umbilical vesicle (yolk sack) projects from the umbilicus, which has the form of a longi- tudinal cleft; the%olk sack was somewhat sunken in and was pyriform in shape. The pelvic end of the body is curved forward in a hook-like manner and on account of the twisting of the embyro cannot be seen from the left side. In the posterior half of the trunk one sees four parallel longitudinal ridges, two of which, the medullary and somitic ridges, belong to the axial zone, while the other two, the Wolffian and marginal ridges, belong to the parietal zone. No traces of the extremities are yet visible. According to His's estimate there were thirty-five pairs of primitive somites. His gives the following measurements: Greatest length in a straight line 2.6 mm. From vertex to behind the mandibular process 0.7 mm. From vertex to behind the heart 1.4 mm. Height of yolk sack where it projects from umbilicus. ... 0.6 mm. Maximal height of yolk sack 1.7 mm. Length of yolk sack 2.6 mm. Length of the posterior portion of the body, measured from the point of emergence of the yolk sack 0.6 mm. The first embryo of Hochstetter's series (Fig. 44), which belongs to Professor Fischel of Prague, presents a very great advance in development. As the figure shows, it is not only very greatly curved upon itself, but it is also strongly twisted spirally. In the head pne sees the primary optic vesicles and on the first DEVELOPMENT OP HUMAN EMBRYO. 69 branchial arch an early indication of the maxillary process. Both ■upper and lower extremities are recognizable, the former being plate-like. The definitive nape bend is strongly developed. The greatest length of the embryo was 4.02 mm. ; it was obtained as an abortion. The embryo which I would place next in the series is embryo G 31 of the Anatomical-biological Institute of Berlin (Fig. 45) (Fig. VIII of the Normentafel of Keibel and Elze) ; it stands close to embryo 7 (Fig. 34, g) of His's Normentafel. Its greatest length is the nape-breech length (nape line) and is 4.9 mm.; the vertex- breech length is 4.7 mm. It is rather strongly curved upon itself, but more regularly than the preceding embryo ; the nape bend is strongly marked, and the spiral twisting is still quite distinct. The tail lies to the right. The upper extremities are plate-like in form, the lower ones ridge-like. The anterior end of the head shows in the middle line a slight depression, and the anlagen of the eyes show through the integument. The region of the later sinus eervicalis is still but little depressed, so that the third and fourth branchial arches are quite easily seen. The atrial and ventricular portions of the heart can be distinguished through the thin wall of the pericardial cavity. A few words may be said concerning embryo 7 of His's Normentafel (Fig. 34,^). This embryo has reached the high- est degree of curvature that has yet been observed in the human embryo. The length from the forehead to the tip of the coccyx following the curvature was 13.7 mm.; the straight line from the nape prominence to the twelfth thoracic segment, the nape- line (NL), was 4 mm. in length. The tail end was curved forward to the level of the ventricle of the heart and lay upon the left side of this. In the curve which is formed by the embryo from the forehead to the tip of the coccyx there are four regions where the bend- ing is stronger than elsewhere: (1) the region of the mid-brain; (2) the nape prominence; (3) the boundary between the neck and thoracic regions; and (4) the boundary between the abdom- FlG. 44. — The Fischel embryo. X 10. (After Hochstetter: Bilder der ausseren Korperform men- Bchlicher Embryonen, Munich, F. Bruckmann.) 70 HUMAN EMBRYOLOGY. Fig. 45. — The embryo G 31 of the Anatomical-biological Institute in Berlin. '10. (From the Normentafel of Keibel and Elze, Fig. VIII, plate 14.) inal and pelvic regions. The plane of symmetry of the embryo is warped and is twisted in such a way that the head looks to the right and pelvic end to the left. The upper extremities are plate-like, the lower ridge-like. At the head end the shape of the cerebral hemispheres, the tween-brain, mid-brain, hind-brain, and after-brain, is plainly recognizable, and the bound- aries of the fourth ventricle are sharply defined. In a view from the dorsal surface the lateral walls of the fourth ventricle show a series of regu- lar and bilaterally symmetrical trans- verse folds (neuromeres). The optic vesicles form on each side a circular circumscribed projection measuring 0.35 mm. in diameter, and the audi- tory vesicles are ver^- evident as oval structures lying at the level of the sec- ond visceral groove. In addition to the moderately developed maxillary and mandibular processes the lateral wall of the head shows on each side three visceral arches, the fourth of which is still fully exposed. The distance from the anterior border of the maxillary process to the fourth branchial groove was 1.4 mm. ; a line drawTi through the ventral ends of all four arches was almost straight and cut the fore-brain some distance in front of the optic vesicles. The an- terior borders of these were about 0.55 mm. from the anterior pole of the fore-brain, so that the prominence of the fore-brain characteristic of human embryos in all later stages of development is already present. Im- mediately behind the ends of the vis- ceral arches is the heart, the three portions of which are plainly visible when the embryo is examined from both sides, with this difference, how- ever, that the atrial swelling is more distinct on the left side and the swell- ing of the aortic bulb on the right side. the swelling produced by the aortic bulb from the facial surface of the head. The visceral region of the body lying behind the heart shows on the right side a distinct "Wolffian ridge and the transition border of the amnion ; on the left side no characteristic elevations are visible. Fig. 46. — Embryo 112 of Keibel's collection. X 10. (From the Normentafel of Keibel and Elze, Fig. IX, 1, plate 6.) No deep cleft yet separates DEVELOPMENT OP HUMAN EMBRYO. 71 Neax Fig. 8 of His's Normantafel (Fig. 34, h) stands the embryo shown in Fig. 46. It was obtained as an artificial abor- tion. It shows the vertex and nape prominences and is not only strongly curved on itself but is also spirally twisted. The olfac- tory area is beginning to show a better delimitation dorsally and laterally, and the maxillary process can be seen in a profile view. In the region of the sinus cervicalis which is still widely open, the third and fourth branchial arches are plainly exposed; the heart swelling is very prominent and the atrial and ventricular portions of the heart are recognizable. The liver swelling is still but poorly developed, and the tail is curved in between the heart swelling and the belly stalk. Dorsal to the primitive somites one sees O. Schultze's ^^ division of the sclero- tomes. The embryo had thirty-six pairs of so- mites. Between embryos 8 and 9 of His's Nor- mantafel (Fig. 34, h and i) there is a rather wide interval, into which the embryos shown in Fig. X and Fig. XI of the Keibel-Elze Normen- tafel fit. The embryo shown in Figs. XII and XIr corresponds fairly well with the Braus em- bryo shown in Figs. 2 and 3 of Hochstetter's series. For explanation of the figure of this, shown in Fig. 47, a few words will suffice. The olfactory area is beginning to be delimited more sharply dorsally and caudally ; the first and second branchial arches have now become relatively large and have begun to l)e further modified ; the triangular area caudal to them, in which the third and fourth arches are situated, is slightly depressed, forming the sinus cervicalis ; the upper extremi- ties have passed from the plate-like into the stump-like stage. Following closely upon this embryo comes embryo 9 of His's Xormentafel (Fig. 34, i). In it the nape bend is more strongly marked, so that the head is much more bent down upon the heart ; "^ 0. Sehultze : Ueber ambryonale nnd bleibende Segmentierung, Yerli. Anat. Ges., 1896, pp. 87-93. Fig. 47. — The Braus embryo. X 10. (From Hochstetter: Bilder der iiusseren Korperform menschlicher Embryonen, Munich, F. Bruckmanu.) 72 HUMAN EMBRYOLOGY. the distinctly circumscribed olfactory pit rests upon the heart. The tips of the anterior extremities have become bent ventrally, and the roots of the extremities are correspondingly angled. In the trunk one sees three tumor-like projections, of which the two situated more cranially and ventrally are produced by the ven- tricular and atrial portions of the heart, while that lying more caudally and dorsally is formed by the liver. The embryo shows a very distinct external tail. The external configuration of the head is principally determined by the subdivisions of the brain, whose shapes are clearly recognizable through their thin covering. At the base of the fore-brain is the olfactory pit, and a slight dis- FiG. 48. — Embryo 304 of Robert Meyer, Berlin. X 10. (From the Normentafel of Keibel and Elze, Fig. XIV, plate 32.) tance in front of this is the eye with the lens groove. The small- ness of the eye compared with the great development of the fore- brain is characteristic. A prominence behind the eye marks the position of the ganglion of the trigeminus; it lies in the angle be- tween the mid-brain and hind-brain. At the level of the second ATsceral groove the auditory vesicle, together with the ganglion acusticum lying in front of it, forms a slight elevation. Elevations are beginning to appear on the mandibular process and hyoid arch; the third and fourth branchial grooves lie at the bottom of a triangular depression, which will become the sinus cervicalis. Embryo 10 of His's Normentafel is comparatively large for the degree of development it presents ; it resembles the preceding embryo, and was taken from a uterus. DEVELOPMENT OP HUMAN EMBRYO. 73 Figs 49-51 — The embryo Ma. 3 of Hochstetter. X 10. (From the Normentafel of Keibel and Elze, Figs, X VII 1 XVII d, and XVII v, plate 41.) 74 HmiAN EMBRYOLOGY. Embryo XIV of the Keibel-Elze Normentafel (Fig. 48) is not quite so large, but is more developed. This is shown especially by the condition of the extremities and of the sinus cervicalis. Behind the dorsal part of the hyoid arch one may perceive the entrance into the glossopharyngeal organ (the second branchial groove organ). In the upper extremities the hand plates are dis- tinct. Forty trunk somites were counted, and the last six of Fig. 52. — The embryo P. 1 of Hochstetter. X 10. (From the Normentafel of Keibel and Elze, Fig. XIX, plate 45.) these were beginning to be transformed into a tail filament. The embryo was obtained from an artificial abortion. The embryo to which we now come may be considered not only from the left side (Fig. 49) but also from the dorsal (Fig. 50) and ventral (Fig. 51) surfaces. It is the Hochstetter embryo Ma. 3, Fig. XVII of the Keibel-Elze Normentafel and Figs. 7, 8, and 9 of Hochstetter 's series. The trunk region has again begun to elongate. The spiral twisting is evident only to a slight extent in the tail region, the tail lying to the left of the belly stalk. The nape bend is almost a right angle; the cerebral hemispheres are DEVELOPMENT OP HUMAN EMBRYO. 75 recognizable externally; and the cerebellum shows out plainly, especially in the ventral view (Fig. 51). The axes of the upper limbs are almost parallel to the dorsal line ; the hand plates are almost circular; the elbows show especially distinctly in the dorsal view (Fig. 50) ; and in the same view one may perceive, on the surface of the upper limb which looks towards the trunk, a small tubercle. In the lower limb the foot plates are marked off. Dorsal to the row of primitive somites one can clearly distinguish Schultze's segmentation of the sclerotomes. The maxillary pro- cesses have come into relation with the median nasal process, the openings of the nasal pits look towards the wall of the pericardium and are no longer to be seen from the side, especially from the left. A distinct sculpturing is present on the mandibular process and especially on the hyoid arch. The entrance into the sinus cervicalis is still vis- ible as a small triangular hole. In part somewhat more and in part less developed than this are embryos 11, 12, and 13 of His's Normentafel (Fig. 34, I, in, and n) ; that shown in Fig. 11 (l) was taken directly from the uterus. In that shown in Fig. 12 (m) a small nape fur- row has formed beneath the strong nape prominence and the position of the upper extremi- ties does not seem normal. In that shown in Fig. 13 (n) the body is relatively slender and a heart swelling cannot be made out. The body of Hochstetter's embryo P. 1 (Fig. 52) (No. 10 of Hochstetter's series. Fig. XIX of the Normentafel of Keibel and Elze) is already rather elongated. The elbows have moved out from the body and the forearm makes an acute angle with the dorsal line. The fingers are beginning to become distinct on the hand plates, and the foot plates have become circular. Close to this embryo come those shown in Figs. 14, 15, 16, and 17 of His's Normentafel (Fig. 34, o, p, q, and r) ; in Figs. 15, 16, and 17 the thumbs are distinctly recognizable. In the lower limbs of Figs. 18, 19, and 20 of His's Normentafel (Fig. 34, s, t, n, and v) the anlagen of the toes can be seen. The head is gradually bending to the erect position, and in correspond- FlG. 63. — The embryo Chr. 2 of Hochstetter, X 5. (From the Normentafel of Keibel and Elze. Fig. XX, plate 61.) 76 HUMAN EMBRYOLOGY. DEVELOPMENT OF HUMAN EMBRYO. 77 ence with this the neck is forming ; the formation of the external ear is also making progress. These stages may be represented here by Hochstetter's embryo Chr. 2 (Fig. 53, No. 15 of Hochstet- ter's series, Fig. XX of the Keibel-Elze Normentafel), Rabl's em- bryo C (Figs. 54, 55, and 56, Nos. 16, 17, and 18 of Hochstet- ter's series) and the embryo No. 302 of Eobert Meyer's collection (Fig. 57, Fig. XXI of the Normentafel of Keibel and Elze). In the embryo Chr. 2 the nape bend is somewhat more than a right angle. The mouth lies in close contact with the peri- cardium, and the nose has become free from it. The folds of the pinna, the trag-us and the antitragus, are formed; the concha is widely open and fiat. Behind the nape prominence the contour of the back shows a distinct depression. The fingers have become quite pronounced and the toes are just indicated, but the position of the great toe is already to be recognized. The elbows and knees are evident, and the angle that the axis of the forearm makes with the dorsal line is almost a right angle. Slightly further developed than this is Rabl's embryo C (Figs. 51, 55, and 56). The nose is more marked off, the eyelids are further advanced, and so also the ear. The finger-tips are distinctly projecting beyond the border of the hand plate. In the ventral view one may note the position of the eyes and of the limbs. Between the foot plates may be seen the physiological umbilical hernia. The tail has become transformed into a coccy- geal tubercle. In the dorsal view the sharp slope of the caudal portion of the trunk is striking. The embryo 302 of Eobert Meyer's collection (Fig. 57) shows a distinct depression between the root of the nose and the fore- head. The pinna, tragus, and antitragus are evident; and while the concha is still widely open it shows a tendency to deepen. Early indications of the eyelids are to be seen. The nape flexure is still very evident, but it forms an obtuse angle, and the depres- sion of the dorsal line just below it is less marked. The nose and mouth have both separated from contact with the pericardium ; the mouth is slightly open and the mandible lies close upon the breast. The shoulder is distinctly marked off from the body, the upper arm makes an angle with the forearm, and the elbow projects strongly. The hand plate is beginning to turn ventrally, and the finger anlagen are very distinct ; their tips are becoming free and the thumb is strongly abducted. In the lower limb the knee is very distinct and anlagen of the toes are appearing on the foot plate. The dorsum of the foot is not yet marked off from the crus. When we pass from embryo 21 (Fig. 34, v) to embryo 22 (Fig. 34, w) of His's Normentafel, we are passing from the embryonic to the fetal stage. The profile of the latter embryo already shows clearly a nose, upper lip, lower lip, and chin ; and the neck is also HmiAN e:mbiiyology. present. The eyelids are formed, and above the eyes there is a distinct supra-orbital swelling. The upper limb, in which the fingers have become distinct, has increased in length considerably and shows a distinct division into upper arm, forearm, and hand. The shoulder has formed, and the characteristic position of the arms is worthy of note. In the lower limb the foot, crus, and thigh can be distin^Tiished. The anlagen of the toes are not yet sepa- rated from one another, but that of the great toe is especially well marked. The nape promi- nence and nape depression are much reduced, and the coccygeal tubercle still ap- pears as a short stumpy tail. With the three remain- ing stages of His's Nornien- tafel, Figs. 23, 24, and 25 (Fig. 34, X, y, and z) the transition from embryo to fetus has been accomplished. In Fig. 34, X, one sees the toes separated from each otlier, and the great toe has a characteristic abducted po- sition, recalling the position- of the thumb in the corre- sponding stage in the devel- opment of the hand. In Fig. 34, y, the foot is better formed ; the legs have under- gone a twisting so that the knee looks more upwards and the foot more down- wards. Fig. 34, z, still shows a slight nape prominence and a very shallow nape de- pression. The small crea- ture shoAvs a distinctly human character in its features. Some ad- ditional figures may complete the story. The embryo shown from in front in Fig. 58 corresponds some- what to Fig. 22 (Fig. 34, w) of His's Normentafel. The position of the eyes, the broad nose, and the position of the limbs may be noted. A' fetus measuring 25 mm. in its greatest length is shown from the left side and from the ventral aspect in Figs. 59 and 60, re- produced from the Normentafel of Keibel and Elze. It may be regarded as standing between Figs. 24 and 25 (Fig. 34, y and z) Fia. 57 — Embryo 302 of Robert Meyer, Berlin, X 5. (From the Normentafel of Keibel and Elze, XXI, plate 64.) Fig. DEVELOPMENT OF HUMAN EMBRYO. 79 of His's Normentafel. Again I would call attention to the position of the limbs. In Fig. 59 we see touch pads on the sole of the right foot. At the summit of tlie coccygeal tubercle there is a small knob, as in all well-preserved embryos of this stage, and it is also seen in tlie ventral view (Fig. 60), in which the physiological umbilical hernia is indicated by the coils of the intestine showing tlirough the wall of the cord. We may now again pass in review the stages which the human embryo must traverse in order to acquire its human form. The earliest known embryos are flat, shield-like plates which rest upon the yolk sack. At a certain stage a primitive streak extends through- out the entire length of the shield; in front of the primitive streak the embryo is formed, and it grows at the expense of the streak, which retrogresses in a cranio- caudal direction. The streak is thus con- verted into a growth zone, which may be termed at first the trunk bud and later the tail bud. To correctly imderstand these processes of development we must bear in mind the extent to which the head region predominates in young embryos. Beturning again to the primitive streak, during its modification its caudal end becomes bent ventrally; tlie cloacal membrane had previously formed in this caudal region, and it now comes to lie on the ventral surface. This bending process is associated with the constric- tion of the embryo from the yolk sack; cranially and caudally, from the right and from the left, the boundary grooves cut inwards and soon convert the yolk sack into a stalked vesicle; in the meantime the dorsal portion of the embryo grows more rapidly than the ventral, a condition caused by the accelerated growth of the meduUan'' anlage. Be- fore the "medullary plate is converted into a tube the brain, with its principal subdivisions, becomes distinct at the anterior end, and the optic pits are also formed. By the rapid growth of tlie brain portion, while the medullary tube is still open, there as produced a bending down of its cranial portion (the vertex bend) and, if the Pfannenstiel embryo III (Figs. 41 and 42) is to be regarded as normal, also the more caudal nape bend. It has been already pointed out, however, on pp. 66 and 67, that a diffi- Fig. 58. — Bulius embryo, 13/2. (From the Normen- tafel of Keibel and Elze, Fig. XXII, plate 70.) 1907. 80 HUMAN EMBRYOLOGY. culty exists in this particular. Embryos that are properly be- lieved to be older do not show the nape bend distinctly, and these agree with the embryos of other mammals, especially with that of the orang-outang which I described in 1906. There are two possibilities : either the nape bend of the embryo shown in Figs. 41 and 12 was an abnormality; or it is a primary nape bend that normally disappears, the long known nape bend later appearing independently of it. On account of the lack of the necessary stages this question cannot be decided at present; but in any event the entire embryo soon becomes curved on account of the much greater growth of the dorsal surface, and coils itself in a spiral. The spiral turns sometimes to the right and some- times to the left. While these developmental processes are taking Figs. 59 and 60. — The Marburg fetus No. 21. X 2.5. (From the Normentafel of Keibel and Elze, Fiea. XXV 1 and XXV v, plate 82.) place there has formed from the tail bud a small but unquestion- able external tail. In these stages also the head region still pre- dominates to an extraordinary degree, but the embryo no longer consists, as in the stages of the early primitive somites, only of the future head, so to speak. The extensive coiling toward the ventral surface is followed by a very gradual uncoiling; first the trunk straightens out and the nape bend slowly becomes obliterated. These processes are accompanied, perhaps determined, by a rapid growth of the ventrally situated organs, especially of the heart and liver, which produce on the surface of the body the heart and liver prominences. The Wolffian bodies do not play a very im- portant part in this respect in man. With the straightening out of the nape bend opportunity is afforded for the formation of the neck. Originally the face lies closely upon the heart promi- nence and the ventral portion of the neck is entirely wanting, while its lateral parts are occupied by the branchial or visceral DEVELOPMENT OF HUMAN EilBRYO. 81 arclies. The transformation of these interesting structures will be thoroughly described in other places ; here the following may suffice. The third and fourth arches— quite transitorily there is also recognizable the anlage of a fifth— remain small in comparison with the first and second, the mandibular and hyoid arches, and consequently the region in which they occur becomes depressed, forming the sinus cervicalis, which becomes covered in by the second or hyoid arch and its opercular process. In the region of the first branchial groove there are formed on the mandibular and hyoid arches a number of elevations and folds from which the ex- ternal ear is formed ; its development will be considered in detail with the sense organs. In determining the form of the head the brain is at first almost the controlling organ, and its various parts can be dis- tinguished until relatively late stages through the thin walls of the head. G-radually the mesenchyme increases in the walls, and, above all, the face begins to develop in the service of the principal sense-organs and of the respiratory and digestive tracts; its formation will be considered later on. The ventral wall of the trunk is at first very thin, and the heart with its various parts, the liver, and other viscera may be seen through it. From the right and left the anlagen of the skeleton and the musculature grow into the walls ; inliibitions of this growth may occur and produce ectopia cordis and fissura sterni, which depend upon disturbances of this process in the thoracic region. A portion of the intestine normally projects like a hernia into the umbilical cord, in association with the outward growth of the abdominal walls; thus the hernia funiculi umbili- calis physiologica is produced. Normally this hernia disappears with the further formation of the abdominal walls, but occasionally it may persist as an inhibition structure. On the ventral wall of the body the region from the umbilicus to the end of the trunk is the original primitive streak territor3^ That portion of the streak which was originally, before it became bent ventrally, the caudal portion, but which is now directed cranially, becomes encroached upon by the developing skeletal and muscle anlagen of the trunk. If inhibitions occur in this region ectopia vesicae and pelvic clefts may be produced, and epispadias also belongs in this category. The openings of the urogenital sinus and the anus and the intervening perineum arise in the territory of the portion of the cloacal membrane which per- sists after the formation of the abdominal wall below the umbilicus. For an account of these developmental processes reference may be had to the chapter on the urogenital apparatus. That a tail occurs in the human embryo has already been noted; embryos of 4-12 mm. NL. have a typical external tail, in Vol. I.— 6 82 HUMAN EMBRYOLOGY. -whieli a caudal gut occurs and wliicli possesses more segments and spinal ganglia than persist. This external tail (cauda aperta) becomes transformed in its distal portion into a tail filament, which is knob-like in the human embryo and is later cast off, while the remaining portions become overgrown by the neighbor- ing parts and so disappear beneath the surface ; in this situation remains of it may persist, forming what is termed the internal tail (Braun, 1882) or cauda occulta (Bodenacker, 1898).i* If one now glances at the body as a whole, one can hardly fail to be struck by the fact that the cranial portion is much more fully developed, up to relatively later stages, than the caudal portion ; this is shown very clearly by the dorsal views of embryos, and I would call attention, in this connection, to Figs. 51 and 55. The extremities first appear as ridges which later are con- verted into plates, becoming more sharply defined cranially and caudally. The portion which is first formed corresponds essen- tially to the anlage of the hand or foot ; gradually the anlagen of the forearm and crus, and then those of the upper arm and thigh, grow out from the body. The further differentiation of the ex- tremities, as well as that of the face and head, will now be con- sidered more thoroughly, and for this purpose good figures are much more important than extensive descriptions. The figures showing the development of the face in the first stages of its formation are rather few in number, since the pro- cesses cannot be well observed in the human embryo without dis- section, and dissection prevents the obtaining of a continuous series of sections. I shall start with a stage that Rabl estimates at nineteen to twenty days ; according to the estimate of Bryce and Teacher (see pp. 26 and 27) it would be considerably older. It corresponds essentially to Fig. VI of the Normentafel of Keibel and Elze and to embrj^o 5 (Fig. 84, e) of the His Normentafel. Fig. 61 shows a profile \aew of it. Three branchial arches are recognizable; on the first the maxillary process cannot be seen in this view, although it is present, as may be seen from the front view (Fig. 62). The optic vesicles, which are directed laterally, show through the wall of the head, and over the second arch are the auditory vesicles, not yet quite closed. In the front view (Fig. 62) one looks into the pentagonal oral sinus; it is bounded above by the frontal process, laterally and above by the maxillary processes, and below by the mandibular processes. ]\Iuch more developed is the face that I reproduce (also from Rabl) in Fig. 63. It is from an embryo which measured 8.3 mm. in the preserved condition and belonged, according to Eabl's estimate, to the fourth or perhaps the beginning of the fifth week. It corresponds fairly well with Fig. 11 "^ Compare Keibel in Hertwig's Handbnch, vol. i, part 2, p. 151. DEVELOPMENT OF HUMAN EMBRYO. 83 Figs. 61 and 62, — Head end of an embryo en face and from the left. The embryo corresponds essen- tially to embryo M of His (here Fig. 43). X20. CFrom C. Rabl: Die Entwicklungsgeschichte des Gesichtes, Leipzig, 1902.) (Fig. 34, I) of His's Normentafel and to Fig. XIV of those of Keibel and Elze. The pentagonal opening of the oral sinus has transformed into a broad mouth-cleft, whose lower border shows a notch between the two adjacent ends of the mandibular processes. The nasal pits are distinct; they are deeper dorsally and are flattened out ventrally. One may also speak of the two nasal processes, that is to say, the ridges which bound the nasal pits laterally and medi- ally. The portion of the frontal process between the medial nasal processes is still very broad. The advance that is to be seen in Fig. 64 (again from Eabl) is quite marked. The figure represents the face of an embryo that corresponds to Fig. 14 (Fig. 34, o) of His's Nor- mentafel and to Fig. XIX of those of Keibel and Elze; its nape- breech length was 11.3 mm. and its age was estimated by Eabl at ^ thirty to thirty-one days. In the head the H ^ --^ fore-brain region is markedly prominent; ■ (f \ below the forehead the area triangularis I /■ "^ (His) projects, and below the middle of this P^ % there runs what is only a moderately broad groove through the mouth opening into the palate, separating the medial nasal processes, which at their first appearance were so far apart. The two nasal openings are not only relatively but absolutely nearer each other than in the preceding stage, and look directly forward. The lower ends of the medial nasal processes, which are uniting with the maxil- lary processes, are separated by a slight de- pression from the lateral processes and are the processus globulares (His). The lateral nasal processes are consequently excluded from the formation of the upper boundary of the mouth, which becomes the upper lip. Fig. 65 shows the face of a 15 mm. embryo from in front, after Retzius. The upper lip is still divided, the nasal septum is still somewhat incomplete, and the processus globulares are still Fig. 63. — Head of an embryo of 8.3 mm. NL., seen en face. X 10. (From Rabl, I.e. It corresponds essentially to Fig. XIV of the Normen- tafel of Keibel and Elze, here Fig. 48.) 84 HmiAN EMBRYOLOGY. visible. From the upper angle of the nose two folds run from above and medially downward and laterally, the medial one passing to the angle of the mouth; these are the oblique maxillary folds, that is to say, the nasolabial and the sub- orbital folds. The eyes are directed distinctly laterally and are compar- atively wide apart; the mouth is very wide ; and at the middle of the lower jaw the anlage of the chin is visible, although the lower jaw and chin are as yet very poorly devel- oped. Figs. 66 and 67 show the head of an embryo of IS mm. en face and in profile (after Eetzius). The view en face appears rather remarkable; Retzius thinks that the embiyo may not have been quite normal, but gives no further reason for such a sup- position. The description given of the preceding embryo will answer for this one also, if it be added that the nasal septum is now completely formed. In the profile view the posi- tion of the ear should be noticed; if the mouth cleft be produced dorsally the ear would lie below it. Figs. 68 and 69 show profile and en face views of the head of a fetus of about eight weeks and 25 mm. in length (after Retzius). The eyelids are in process of formation and supra-orbital folds also occur, in addition to the nasolabial and suborbital. The supranasal groove is strongly marked. The distance be- tween the eyes is still considerable, and they are distinctly directed laterally. Figs. 70 and 71 represent the head of a fetus 42.5 mm. in its greatest length (ninth week), in profile and en face. In the profile view the great development of the forehead region is striking, and below this the root of the nose is deeply depressed. The nose is still low, but the lower jaw and chin are well marked. The eyes are completely closed by the lids, but the distance between the inner angles of the eyelids is very consider- able; from medially and above the interpalpebral fissures are directed laterally and downward. The upper eyelid is relatively Fig. 64. — Head of an embryo which corresponds essentially to Fig. XIX of the Normentafel of Keibel and Elze (here Fig. 62). X 10. (From C. Rabl, I. c). Fig. 65. — Head of an embryo of 15 mm., seen en jace. X 5. (After G. Retzius, I. c, Plate XVI, Fig. 1.) DEVELOPMENT OF HUMAN EMBRYO. 85 small and is bounded above by a sharply-marked arched groove and below by the interpalpebral fissure; from the inner angle of^ eye a supra-orbital groove extends obliquely laterally and Figs. 66 and 67. — Head of an embryo of 18 mm., seen en face and in profile. X 5. (After Betzius / c , Plate XVI, Figs. 3 and 4.) Figs. 68 and 69. — Head of a fetus of 25 mm., seen en face and in profile. X 2.5. (After Retzius, I. i Plate XVI, Figg 5 and 6.) Figs. 70 and 71. — Head of a fetus of 42.5 mm., seen en face and in profile. X 2.5. (After Retzius, I. t., Plate XVI. Figs. 8 and 9.) upward. The nose is very broad in proportion to its height (172: 100), and the external nares are closed by the epidermal plugs which are continuous with an epidermal thickening on the upper lip. 86 HUMAN EMBRYOLOGY. Finally, the profile view of the head of a fetus 117 mm. in leng-th may be shown (Fig. 72), and in it I would draw especial attention to the projecting upper lip and the receding chin, to the double lip, and to the shape of the nose. The pinna has almost the position it holds in the adult. With regard to the mouth it may be observed that the vertical median furrow which becomes the phil- trum appears in the fourtli month. The wall-like projecting margin of the upper lip is separated from the inner portion, which in the fourth and fifth months shows more or less projecting tubercle-like elevations. The middle portion of the margin early becomes the tuberculum labii superioris. In a similar manner the wall-like margin of the lower lip becomes delimited from the inner portion. In the first half of the third month the two lips project about equally, but later tlae border of the upper lip and the lip itself grow more rapidly, so that in the fourth and fifth months it projects markedly beyond tlie lower lip; by a stronger growth of the lower jaw and lip this difference is gradually over- come in the sixth to the ninth months, but by a kind of inhibition process the early fetal arrangement may be re- tained in the adult to a marked de- gree. Retzius has given especial at- tention to the time of occurrence of individuality, and comes to the con- clusion that in man it is recognizable even in the fourth month of intra- uterine life, and becomes more marked in the succeeding months. The first phases of the development of the extremities up to the formation of the fingers and toes have already been considered in connection with the development of the entire embrj^o; an account of the further changes in the hand and foot may now be presented, the descriptions given by Eetzius being followed closely. The hands, after the fingers are formed, earlv assume their human form and even in the third month have acquired their most important characters ; they are then comparatively broad and the fingers are flexed. Of the persistent palmar furrows only the largest, the lines of Venus and ]\Iars, are distinct in the "third month. Four distal metacarpal pads (touch pads) appear at the begmmng of the third month opposite the interdigital clefts, and in the course of this month they develop into strongly marked elevations. In addition, there is a distinct ulnar marginal pad on FfG. 72. — Head of a fetus of 117 mm., in profile. Natural size. (After Retzius, I. c, Plate XVII, Fig. 10.) DEVELOPMENT OF HUMAN EMBRYO. 87 the metacarpus and one or sometimes two carpal pads. On the terminal phalanges strongly prominent, hemispherical touch pads develop. These, as well as the metacarpal pads, become relatively lower in the fourth and fifth months, and their boundaries be- come gradually more indistinct; only in individual cases do they persist in a more evident form with the later half of the fetal period. The feet are always some- what behind the hands in their develop- ment. Even long before the separation of the individual toes by the interdigital clefts the abducted position of the great toe is striking; and verj^ early, even in the second month, the prominence of the heel is recognizable. The great toe is from the beginning somewhat thicker and the little toe somewhat smaller than the other three. The soles, like the palms, are at first directed medially, and consequent- ly are opposed to one another. The dor- sum of the foot is relatively very high in the third month, and also broad towards the roots of the toes. Compared with their position in the adult the feet as a whole have now an " oblique " position (the varo- equinus position) ; the arch of the sole is beginning to develop. Fig. 73. — The right hand of a fetus of 25 mm., seen fiom the volar surface. X 10. (After Ret- zius, I c. Plate XXIII, Fig 20 ) IP„ 7i The DOSterior end of a fetus of 25 mm., seen from the dorsal surface. X 10. (After Retziua, -''^ Z. c, Plate XXIII, Fig. 23.) 88 HmiAN EJIBRYOLOGY. During this time, at the beginning of the third month, a row of distal metatarsal pads appears as four or five roimdish or oval elevations, and one soon sees clearly that they lie opposite the in- terdigital clefts ; it seems as though a shifting toward the lateral (fibular) side occurred. In the next stage, that is to say, in the latter half of the third month, there are four metatarsal pads which cor- respond to tlie interdigital clefts, the fifth seems to have shifted proximally on the lateral border of the foot. At this time the four first-mentioned pads are relatively at their highest stage of devel- opment, and simultaneously the pads on the terminal phalanges have developed to hemispherical plantar elevations. In the fourth Figs. 75 and 76. — The right arai and foot of a fetus of 25 mm., seen from the dorsal and the plantar sur- faces. X 10. (After Retzius, I. c. Plate XXIII, Figs. 22 and 20.) and fifth months the distal metatarsal pads undergo a relative retrogression and their outlines become gradually indistinct; the phalangeal pads remain well marked, although their outlines be- come less pronounced. Some figures may make these points clear. Figs. 73-76 repre- sent the extremities of a fetus of 25 mm. Fig. 73 shows the right hand from the volar surface with the touch pads ; Fig. 74, the two lower extremities seen from behind and dorsally. The feet are seen from their fibular borders, and their plantar surfaces are turned toward one another. The high dorsum and the malleolar eminences should be noted. Figs. 75 and 76 show the right foot from the dorsal and plantar surfaces ; the abducted position of the great toe and the metatarsal touch pads cannot be overlooked. DEVELOPMENT OP HUMAN EMBRYO. 89 Fig. 77 shows the sole of the right foot of a fetus of 44 mm. with its touch pads ; Fig. 78 represents the middle finger of a fetus of 52 mm. from the side. In the descriptions of the development of the face, hands, and feet the conditions in the fetal period have already been con- sidered. For the development of the form of the rest of the body reference must be made to Gus- tav Retzius (I. c). Retzius has been the first to stucfy thoroughly the proportions of the human body during the fetal period, and a resume of his most important results may bring this chapter to a close." He finds as fol- lows: i. The entire body length, measured from vertex to heel, increases during the fetal period to a greater extent than the vertex-breech length, that is to say, the lower limbs continually increase in length. 2. The relation of the height of the head to the vertex-breech length, gradually diminishes. 3. A comparison of the length of the entire vertebral column with the height of the head and with the different regions of the column shows that : a. The height of the head diminishes relatively to the length of the vertebral column. b. In general, only a slight change can be obser\'ed in the ratio of the cervical vertebrae to the entire column, al- though there is a certain tendency towards a relative shortening of the cervical vertebrae in the earlier stages. c. Scarcely any change, apai't from individual variations, can be seen in the relation of the thoracic vertebras to the entire column. d. Tlie relation of the lumbar vertebrae to the en- tire column shows no appreciable change. e. The relation of the sacrococcygeal vertebrae to the entire column also shows no material change; indi- vidual variations are, however, especially great, and the difficulties in the way of making exact measurements are worthy of note. 4. The relation of the circumference of the head to the body length diminishes from the earlier stages. 5. As regards the relation of arm length to body length, it was found that during the second and third months the arm grows to such an extent that often even in the third month, and more certainly in the fourth and beginning of Fig. 77. — The right foot of a fetus of 44 mm., seen from the plantar surface. X 10. (After Retzius, I. c, Plate XXIV, Fig. 8.) Fig. 78. — The middle finger of the right hand of a 52 mm. fetus, seen from the right side. X 10. (After Retzius, I. c, Plate XXIV, Fig. 9a.) " Compare also Chapter VIII. 90 hu:man embryology. the fifth, it reaches its greatest relative leng-th for the fetal period, its first maximum (37-42 per cent, of the body length). 6. As regards the relation of the upper extremity to the lower and of the le<5- leng-th to the body leng-th, it may be said that the lower limb grows more slowly than the arm during the fetal period; at the end of that period it is scarcely as long as the arm, but after birth it soon surpasses it. The relative maximum of length for the fetal period (36-39 per cent, of the body lengih) is acquired by the leg at about the fifth month. 7. A comparison of the arm length with the lengths of the upper arm, f oreann, and hand shows that : a. During the period from the third to the tenth month the upper arm is about 39^2 per cent, of the entire arm leng-th. b. In the relation of the arm length and that of the entire distal part of the ai-m (forearm and hand) there are no perceptible changes either of progression ■or regression during the fetal period. c. Also the arm length compared with the forearm length (without the hand), and d. The arm length compared with the hand length show no noteworthy ■changes of proportion during the fetal period. 8. The relations of the leng-th of the lower limb to those of the three portions of which it is composed may be stated as follows: a. The relation of the thigh length to the leg length shows no noteworthy ■changes from the third to the tenth month. 6. In the relation between the leg length and that of the cms (omitting the height of the foot), a slight relative elongation of the crus is evident about or before the middle of the fetal period. c. In the relation of the length of the foot to the leg length there is a definite relative elongation of the foot, especially from the sixth to the eighth month. 9. In the relation of the breadth of the iliac crests to the body length no actual change occurs during the fetal period from the third to the tenth month. 10. As regards the proportions of the head and face during the fetal laeriod, Ketzius' results are as follows : a. Ratio of head leng-th to head width: In the first months, while the cerebral hemispheres are still developing posteriorly, no measurements can be obtained that allow satisfactory comparison. Notwithstanding- that Retzius worked -with Swedish embryos and that the Swedes are a typically dolichocephalic race, it seems that there is a strong tendency to brachycephalism and, indeed, to a quite high degree of it. 6. Ratio of head length to head height : This index is very high in the early months (112..5, 111.1, 100, etc.); it is still high m the third month (108.5, 104.2, etc.); but toward the end of this month it sinks (86.0, 81.6), and remains at about the same level from the fourth to the seventh months, with only a few individual variations upwards. If the figures given above are interpreted according to the standard employed for adult skulls they all denote hypsicephalism (75.1 and over). c. Ratio of head length to head circumference: In general, the circum- ference of the head is about or almost three times as great as the length. The index, which at first is smaller, increases during the third month to this value and remains about the same, with individual variations, to the seventh month. d. Ratio of head width to head height : This index shows a definite tendency to dimmish, apart from individual variations. e. Ratio of head circumference to face height : The figures show no regular change until the seventh month; it is remarkable that during these stages -almost the same values recur. VII. THE DEVELOPMENT OF THE EGG MEMBRANES AND THE PLACENTA; MENSTRUATION. By otto grosser, Peague. I. INTRODUCTION. The difficulties in the way of a comprehensive description of human placentation have been mentioned so often that a detailed re-enumeration of them is unnecessary here. The first stages, so necessary for the understanding of all the later ones, are lacking, just as they are in the case of the formation of the germinal layers, and, as in this case, must be conjectured by deduction and analogy. In the following description an endeavor will be made to state what has been determined with certainty, and, in connection with this, to call attention to disputed questions and to the probable significances of the phenomena described. A statement of our knowledge in the field of comparative placentation may also be dispensed with, since it has repeatedly been given in detail within recent years. ^ The position which man occupies among the Mammalia on the basis of the structure of the placenta may, however, be indicated; and, in connection with this, the nomenclature employed in placental classification and the general morphological and histological processes involved in the formation of the placenta may be described. Placentation is (in mammals) the intimate union (apposition or fusion) of the mucous membrane of the uterus with the outer layer of the ovum, the chorion, which becomes vascularized from the allantois ^ for the purpose of providing for the respiration and nutrition of the embrj^o and for carrj^ing away its waste products. ^ 0. Schultze : Grnndriss der Entwieklung-sgresehiehte, 1897 ; Strahl : Embry- onalhiillen der Sauger mid Placenta, in Ilertwig's Handbuch, 1902 ; Bonnet : Lehr- buch der Entwicklungsgesehiehte, 1907, and, most recently, 0. Grosser: Vergleieh- •ende Anatomie und Ent"wicklungsgeschiehte der Eihaute und der Placenta, Lehrbuch fiir Studierende und Aerzte, Wien, 1909. From this last work the majority of the illustrations of this chapter have been taken. A very complete list of the literature on human placentation is to be found in the work of P. Keibel and C. Elze: Normentafel des Menschen, Jena, 1908. ' In some mammals also from the yolk sack. According to the view of Hubrecht (see especially Resink, Tijdsehrift Ned. Dierk. Vereen, 1903, 190.5; Hubreeht, Quart. Joum. Micr. Sc, 1909), however, the chorion possesses from the beginning a vasifactive mesoderm. Compare Grosser's Lehrbuch. 91 92 HUMAN EMBEYOLOGY. Since the union of the chorion and the uterine mucous membrane is either an apposition or a fusion, the expulsion of the chorion sack after birth either may take place without injury to the uterine mucous membrane or a portion of the latter, the decidual mem- brane (membraiia decidual may be expelled with it; hence the old division of the Mammalia into the lower Adeciduata and the higher Deciduata. But tissue destruction frequently takes place during preg-nancy in the former, and, on the other hand, in many highly organized forms the placenta contains no considerable quantity of maternal tissue, if the maternal blood be disregarded, so that in these it is hardly proper to speak of decidua (Strahl). Accordingly, Strahl * has employed the relations of the maternal blood as a basis of classification and has designated these placenta" "in which post partum the spaces of the placenta which carry maternal blood are separated and fexpelled" complete placentcB or placentae verce; while those simpler placentae, in which during and after birth the maternal blood-spaces remain intact, are termed hcdf placentce or semiplacentce. The classification pro- posed by Eobinson (1904) is practically the same, since his "ap- posed placentce" include those in which there is merely an apposi- tion of the chorion to the uterine mucous membrane, while those in which there is fusion of the two he terms "conjoined placentae." In the same way the two groups proposed by Assheton (1906), that of the placentce plicatcs with simple, non-proliferating chorionic epithelium, and that of the placentce cumidatce with a greatly proliferated and thickened chorionic epithelium, traversed by lacunae for the maternal blood, agree essentially with the two divisions of Strahl. The idea of placental types which the author ^ has conceived takes its origin from another standpoint. The nutritive material which passes from the maternal blood into that of the fetus in the lowest types of placentae passes in succession through maternal endothelium, connective tissue, uterine epithelium, portions of the uterine cavity, the chorionic epithelium, chorionic connective tissue, and the endothelium of the chorionic vessels. At the commence- ment of development all the maternal walls are present in the highest types of placentae also. But while the fetal layers are always retained, or, in the highest types, are gradually formed. ' As a rale, however, the term decidua is not only applied to the superficial layer which is expelled, but also includes the entire thickness of the mucous membrane. This is the case with its application to the human placenta. 'In Hertwig's Handbuch, 1902, and recently in a fuller somewhat modified statement in Der "Uterus puerpuralis von Erinaceus europceus, Verhandl. K. Ak. Wetensch., Amst., 1907. °Verh. morph. Gesellsch., Wien, 1908; Zentralblatt fiir Physiologie, 1908; and Lehrbuch. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 93 during the early stages of development, the maternal partitions disappear one after the other in the course of the phylogenesis or ontogenesis, the chorionic epithelium penetrating farther and farther toward the source of its nutrition, the maternal blood. But the blood spaces themselves, those of the mother on the one hand and those of the fetus on the other, remain sharply separated under all circumstances. The penetration of the fetal tissue may halt at any stage and so determine the structure of the mature placenta and also the name which may be applied to it, this indi- cating the maternal tissue which is in immediate contact with the chorionic epithelium. At the beginning of the series stand placentae such as those of the pig, in which all the maternal parti- tions are retained; the uterine and chorionic epithelia are in con- tact; and the placenta is a placenta epitheliochorialis. If the maternal epithelium disappears, at least to a considerable extent, as in the ruminants, the chorionic epithelium comes into contact with the connective tissue and a placenta syndesmochorialis is formed. If the connective tissue also disappears, so that the chorionic epithelium is in contact with the endothelium of the maternal blood-vessels, as is the case in the Carnivora, according to Schoenfeld, then the placenta is to be termed a placenta en- dotheliochorialis. And, finally, if the endothelium disappears, so that all the maternal partitions have vanished and the maternal blood directly bathes the chorionic epithelium, then the highest possible stage of placentation has been reached and the placenta is a placenta hcemochorialis. Thus the most important morpho- logical character of a placenta is directly indicated by its name. The further subdivisions may, following Strahl, be based on the form of the placenta. Thus there may be recognized a placenta diffusa, in which the chorionic proliferations or villi are uniformly distributed ; a placenta midtiplex, with the villi arranged in groups; a placenta zonaria, in which they have a girdle-like arrangement ; and a placenta discoidalis, m which they are aggre- gated to form a disk-like structure. The last group, which in- cludes the highest types of placentae (the htemochorial of the classification given above), may be divided, again following Strahl (1905), into labyrinth placentce, with narrow capillary-like channels for the maternal blood, and boiul placentce [placentce olliformes), in which the maternal blood has the form of a large sinus, the floor of the space (bowl) being formed by decidua and the roof by the chorion, from which the villi project into the space. The human placenta is a placenta vera (conjugata, cumulata) discoidalis ollif ormis, or, according to my nomenclature, a placenta hcemochorialis discoidalis olliforntis. It represents the highest development of its type, a development which even the placentae of the anthropoid apes have not quite reached. 94 HUMAN EilBRTOLOGT. The nutrition of the embryo takes place, in general, in two ways : on the one hand, by the transference of nutritive material from the blood of the mother to that of the child ; and, on the other, by the direct absorption by the chorionic epithelium of products of the maternal mucous membrane, these products frequently being subjected to a kind of digestive process before they pass into the embryonic circulation. These maternal substances are partly products of secretion, partly waste products, together with extra- vasated maternal blood, and have been included by Bonnet under the term emhryotrophe and by English authors have been desig- nated pabulum. In the lower types of placentae the emhryotrophe plays an important role throughout the entire duration of preg- nancy; in hasmochorial placentae, and therefore in man, we find (as has been noted, for instance, by Pfannenstiel and Jung), at the beginning of development, up to the establishment of a definite circulation in both the maternal and fetal blood spaces, a very distinct absorption of emhryotrophe consisting of degenerated maternal tissues, while later, emhryotrophe is entirely wanting, at least in the region of the placenta." In hasmochorial placentae, therefore, two phases or stages may be distinguished: an embryo- trophic phase, at the commencement of development; and a later hcBmotropMc phase, not sharply distinguished from the former in time, but during which the nutritive material is received from the maternal blood exclusively. This absorption of material cannot, however, be regarded as a simple process of diffusion. This could be the case only with crystalloid substances at the most; colloids, on the other hand (such as the albumins, for instance), are taken from the maternal blood by a process of resorption, associated with a partly con- structive and partly destructive activity on the part of the epithelium of the villi; and certain highly complex substances, such as many immunity substances, cannot pass the placenta at all. The chorionic villi of the placenta have a certain similarity to the intestinal villi (Hofbauer), the maternal blood corresponding to the digested food material. Up to the present the wandering of fat, glycogen, and iron, the last as haemoglobin or its derivatives, has been followed histologically from the maternal blood through the chorionic epithelium into the fetal vascular system. The fat, which penetrates into the chorionic epithelium in a state of solu- tion (saponification), is reconverted into fat globules within the e]3ithelium at the bases of its cells.'^ T he hfemoglobin comes from °A modification of this statement is necessary in connection with the maternal blood. See below. 'Holsti (1908) lays special weight upon fatty degeneration of the decidna, upon fat formation m the glands, and upon the transportation of fat from other organs by leucocytes; this fat is directly absorbed by the chorionic epithelium up to the close of pregnancy. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 95 the maternal blood-corpuscles which, degenerate in the placenta itself, perhaps ia contact with the chorionic epithelium. Oxygen is set free from the oxyha^moglobin of the mother, probably by ferment action (for further consideration consult Hofbauer and Kehrer). The maternal blood, therefore, often assumes in later stages, even in placentse of the highest type, the role of the embryo- trophe, although not in a manner easily recognizable histologically ; and also for this reason the term haemotrophic phase is justifiable. In placentation many cytological phenomena occur that are not observable elsewhere. The most striking are those that lead to the formation of multinucleated masses of protoplasm. Bonnet (1903) has brought order into the exceedingly confused nomen- clature of these structures; he designates (Lehrbuch, 1907) as syncytia, "deeply staining nucleated masses of protoplasm formed by the fusion of originally separate cells ; plasmoclia, on the other hand, arise by repeated nuclear division imaccompanied by cor- responding cell division. . . . Syncj^tia and plasmodia are always living and active formations, endowed with especially energetic metabolism, together with histolytic or phagocytic properties, and also with the power of amoeboid movement. . . . They may subse- quently split again into separate cell territories. . . . Quite dif- ferent are the deeply staining nucleated masses produced by the confusion of originally distinct cell boundaries and by aggregation, hut ivhich show unmistakable signs of commencing degeneration." Such masses are termed symplasmata. Syncytia and plasmodia are chiefly formed by fetal tissues, namely, by the chorionic ectoderm ; symplasmata, on the contrary, arise from the maternal tissues. Yet, for a precise definition of the structure, mention should be made of its origin; so we speak of a syncytium fetale epitheliale, of a symplasma maternum con- junctivum, etc. If the masses in question remain relatively small they are known as multinuclear giant cells or simply as giant cells ; the above classification is applicable to these also. Mononuclear giant cells, which, however, never reach a special development in the human placenta, are merely greatly enlarged cells, usually derived from the fetal epithelium. While extensive histological modifications may affect almost all the constituents of the maternal mucous membrane and find ex- pression there in the formation of the decidua mentioned above, these modifications affect only the chorionic epithelium among the fetal tissues, the chorionic connective tissue and vessels showing great uniformity of condition. The chorionic epithelium has been termed the trophoblast by Hubrecht ; this distinguishes the ecto- derm of the chorion from that of the embryo and that of the amnion. In the region of all placenta; belonging to the higher types it shows, at least in parts, active proliferation phenomena. Wliere 96 IimiAX EMBRYOLOGY. ii comes into relation with the maternal tissues it usually becomes transformed at its surface into a syncytium (according to Hubrecht's terminology, a Plasmodium), and this portion has been termed the ■plasmoditrophoblast (Vernhout, the plusinodihlast of Van Beneden), now more properly the sync yi iotrophohlast ; while those portions in which the cell boundaries are still retained form the cytotrophoblast (the cytoblast of Van Beneden). The trophoblast in Hubrecht's sense is a morphological con- cept, based upon the views of that author as to the phylogenesis of the Mammalia; it occurs in all of this group and covers the entire ovum : it may also enter into entirely passive relations with the maternal mucous membrane. The term is not used in Hubrecht's sense when it is applied to the proliferating' tropho- blast, as frequently happens in the literature.** This proliferating portion of the trophoblast. which is provided with histolytic properties and especially makes possible the formation of placenta:^ of the higher types, is quite different from the portion known as the inactive trophoblast, and has been termed by Minot the trophoderm. Only in man (and the anthropoid apes) do the two ideas coincide, since in these eases the entire trophoblast un- dergoes lively proliferation and is, therefore, converted into trophoderm. Finally, as regards the position of the ovum in the uterus, different types are recognizable (Bonnet, 1903). The union of the ovum with the mucous membrane is known as the implantation or nidation. If the ovum remains in the main cavity of the uterus, the implantation is termed a central one. This is the most fre- quent U-j)e {Adeciduata, Carnirores, the rabbit, the lower apes, etc.). If, however, the ovum becomes implanted in a furrow or diverticulum of the uterus and subsequently is shut off from the uterine lumen by a fusion of the lips of the furrow or diverticulum, then the implantation is of the excentric type (hedgehog, mouse). Finally, if the ovum penetrates into the mucous membrane by producing a destruction of the uterine epithelium and develops in the mucous membrane after the closure of the point of entrance, that is to say, outside the cavity of the uterus, as in the guinea- pig and in the rodent Gpomys, then the implantation is of the interstitial type. In man the occurrence of this last type has now been almost certainly proved. In the last two types of implantation the ovum is separated from the uterine cavity by a layer of maternal tissue, the decidua capsularis. It arises in the first case by a fusion of the margins 'Hubrecht, however, in his earlier works (Placentation of the Hedgehog, 1S90) employed the expression to denote only the proliferating chorionic ectoderm. Compare Hubrecht, Science, 1904, and foot-note 2. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 97 of the walls of the furrow or of the lips of the diverticulum; in the second case, by the fusion of the lips of the implantation cavity ; and, in later stages, when the ovum bulges out toward the lumen of the uterus, it covers like a shell the part of the ovum turned away from the placenta. II. MENSTRUATION. Menstruation, which occurs at regular periodic intervals in man and the apes, is the expression of changes in the uterine mucous membrane which are associated with preparations for the reception of a fertilized ovum. A consideration of it is therefore necessary as an introduction to an account of placentation. The mucous membrane of the corpus uteri has, in general, a very simple structure. A single-layered, cylindrical or cubical surface-epithelium, with varying amounts of cilia tion (Mandl, 1908), simple or sparingly branched tubular glands varying con- siderably in number in different individuals (Hitschmann and Adler), a stroma with fine connective-tissue fibrils which are diffi- cult to demonstrate by ordinary histological methods (Bjorkenheim, Hitschmann, and Adler), and, finally, the entire absence of a sub- mucosa — these are its most important characteristics. The ends of the glands frequently penetrate to between the irregularly de- fined innermost layers of the muscularis and so obtain for the mucous membrane a firm adhesion to the muscularis and a well- protected position for their basal portions, a circumstance of considerable importance both intra and post partum. Hitschmann and Adler ^ have shown by their comprehensive, recently published observations, upon which are based the state- ments that follow, that this mucous membrane is never in a com- pletely resting condition in a fertile female. Growth and degener- ation alternate regularly and form together a menstrual cycle of normally twenty-eight days. The cycle may be divided into certain more or less clearly marked periods or phases. The longest of these is the interval (between two menstruations) or the inter- menstrual period, which lasts for about fourteen days, during which the mucous membrane is almost at rest and only undergoes a very gradual increase in thickness. Upon this follows, without a sharp limitation, the premenstrual period, which lasts about six or seven days and is characterized by intensive proliferation and swelling of the mucous membrane, finally leading to hemorrhages. These last for about three to five days, the period of menstruation, •during which the mucous membrane again decreases in thickness and undergoes extensive degeneration. In the remaining period •of the cycle, the postwi,enstrual period, of about four or six days duration, the mucous membrane is regenerated. " These authors cite the literature of the question. Vol. I.— 7 98 HTOIAX EMBRYOLOGY. The mucous membrane during the interval is in the condition usually described as normal (Fig. 80). The mucosa is, on the average, about 2 mm. thick; in the fresh condition it is grayish red 11 y A r,\i sv- %>^.. />5 r-, < f Qs o o 1% .0- Fig. 81. Fig. 82. Figs. 79-82. — Figures of the uterine mucous membrane in the various phases. Fig. 79. Post- menstrual mucous membrane, one day after menstruation. Fig. 80. The condition during the interval. Fig. 81. Premenstrual condition. Fig. 82. Condition on the third day of menstruation, showing separa- tion of the superficial layer. (After Hitschmann and Adler.) and rather smooth. The glands have a slightly spiral course and, for the most part, are directed obliquely to the surface, their lower ends being, as a rule, bent upon themselves. Their lumina are circular in transverse section and at first empty. The gland cells DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 99 are at first small with closely-set nuclei; but in tlie second half of the period they become enlarged, their plasma becoming homogeneous and acidophilous. The stroma cells (Fig. 89) are fusiform or stellate, with large nuclei, richly provided with chro- matin, and possess but little plasma, so that the tissue resembles adenoid or embryonic connective tissue. Lymphocytes and small lymph-nodes occur in it. Fig. 84. Fig. 85 Fig. 86. ,g^' ffk — '^- FiG. 87. Figs. 83-87. — The form of the glands in the different phases of menstruation under the same magnification. Fig. 83. A postmenstrual gland, small and elongated. Fig. 84. A gland from the inter- val period, spirally coiled and enlarged. Fig. 85. A premenstrual gland, wide, with secondary alveoli and in secretion. Fig. 86. Glands at the third day of the menstrual period, one still of the premenstrual type, the other contracted and degenerating. Fig. 87. Gland from a young decidua, wide, with secondary alveoli and in secretion. (After Hitschmann and Adler.) Already toward the close of the interval the gland cells begin to produce secretion granules (Fig. 84), which are also expelled into the lumina of the glands; and the stroma begins to show a diminution in compactness and some cedematous infiltration. In the premenstrual stage (Fig. 81) the mucous membrane rapidly thickens to two or three times its previous thickness. This depends partly upon an increase of the oedema, and partly upon 100 HUMAN EMBRYOLOGY. an enlargement of the individual elements. This shows itself in the gland cells (Fig. 85) by a swelling of the nuclei and of the plasma and by abundant secretion, which produces a frayed ap- pearance on the inner surfaces of the cells; the secretion, which is also to be found in the lumen of the uterus, is now clearly recog- nizable histologically as a mucous secretion and contains flakes of the older, still acidophilous secretion. The enlargement of the cells produces, on the one hand, a formation of folds and out- pouchings of the walls of the glands, the stroma projecting like Fig. 88 Fig. 90. Figs. 88-91.— The cyclic changes of the stroma cells of the uterine mucosa. Fig 88 Postmen- Btrual condition. Fig. 89 The condition occurring in the interval. Fig. 90. Premenstrual condition ^r to .ax days before the menstruation). Fig. 91. Condition immediately before menstruation. (Alter Hitschmann and Adler.) papilla into the folds of the mucous membrane; on the other hand, in conjunction with the secretion, a great enlargement of the lumina of the glands results. In consequence, the walls of neigh- boring glands are brought nearer together, the stroma being compressed between them ; and an appearance as if there was an increase in the number of the glands is produced. The glandular changes are most striking in the deepest layer of the mucous membrane; in the superficial layer the stroma cells enlarge (Fig. 90) and become roundish or polygonal with a clear, feebly staining plasma and large, also feebly staining nuclei; they represent a preliminary stage of decidual cells (Fig. 91). By the localization of the glandular changes, on the one hand, and those of the stroma DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 101 cells, on the other, the mucosa is differentiated into two layers, which, as in the case of the mucous membrane of the gravid uterus, are designated, the deep one as the spongy and the super- ficial one as the compact layer (compare also Fig. 92). The 4 Wff '? \, if Fig, 92. — The uterine mucous membrane in the first day of menstruation. BL, hemorrhage into tlie compacta; Co., compacta; H., hsematoma under the epithelium; M., muscularis; I'o., gland of the post- menstrual type; Pr., gland of the premenstrual type; Sp., spongiosa. (From Hitschmann and Adler.) former acquires a spongy consistency as the result of the great enlargement of the glands, and the latter is characterized by the closely packed, decidua-like stroma cells and the straighter course of the glands. Toward the end of the premenstrual phase there occurs an engorgement and dilatation of the blood-vessels; and the mucous membrane, which at first was pale, becomes bright red in color. 102 HU.AIAN EI\IBRTOLOGY. Small hemorrhages occur at the same time ; these become confluent and destroy the continuity of the tissue, subepithelial haematomata are formed, portions of the epithelium are torn away, blood makes its way to the surface of the mucous membrane, and the actual menstruation begins. With the onset of this there is an effusion of blood and of the cedema fluid, on the one hand, and an expulsion of the glandular secretion, on the other, whereby a rapid shrinkage of the mucous membrane occurs. Frequently, but not always, there is also a desquamation of the glandular epithelium. The outpouchings of the glands rapidly disappear and these assume an almost straight form with narrow or collapsed lumina (Figs. 82, 86, and 92) ; the emptied cells become low and small. Of the stroma cells those that have been most profoundly altered break down and are expelled or carried away by leucocytes; the rest again diminish in size. The surface epithelium may be for the most part retained, or, even in normal menstruation, may be expelled together with the greater part of the compact layer (Fig. 82), painful contractions of the musculature aiding in the separation of this rigid swollen layer; yet the epithelium is in all cases regenerated before the close of the menstruation. In the postmen strual stage the mucous membrane is thin, with almost straight glands (Fig. 83) and long, narrow, fusiform, closely packed stroma cells (Fig. 88) ; little remains of the hemor- rhages, and even these remnants quickly disappear. After a few days the glands again become larger and begin to assume a wavy outline, the stroma cells become more succulent, and the mucous membrane returns to the relatively quiescent stage of the interval ; nevertheless, in the first half of the interval numerous mitoses are to be found in the glandular epithelium. The similarity of the premenstrual mucous membrane to that of the decidua indicates that the premenstrual changes (the loosen- ing up of the tissue, enlargement, increased glandular activity, swelling of the stroma cells, formation of two layers in the mucosa) are a ripening process, a preparation for the reception of a fertilized ovum, and that thev are physiologically the most im- portant part of the entire cycle, while menstruation itself is only a secondary process, a degeneration of the mucous membrane, which from a failure of pregnancy has not been able to fulfil its purpose. As to the relation of menstruation to " beat " of animals, as -^vell as con- cerning the question of the occurrence of regnlar cyclical changes in the genital mucous membranes, see, for example, W. Heape : " The Sexual Season of Mammals and the Relation of the ' Prooestmm ' to Menstruation," Quart. Joum. Mier. Science, vol. xliv, 1906; M. Van Herwerden : " Bydrage tot de Kennis van den menstrueelen Cyclus, Tijdscbr. Nederlandsehe Dierkundige Yereeniging, Deel, x, 1906; and the auto-abstract of the author: "Beitrag zur Kenntnis des men- struellen Zyklus," Monatssehrift fiir Geburtshilfe und Gynakologie, vol. xxiv, 1906. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 103 Bryee and Teacher (1908) lay special weight upon the possibility of an implajitation m any portion of the intermenstrual cycle (compare Sect III), and so reach the conclusion that the " menstrual deeidua " is not a preparation for the reception of an ovum and that menstruation cannot be regarded as "the abortion of au unfertilized ovum." The object of menstruation is merely to maintain the endometrium at all times ready for the formation of a deeidua; the premenstrual tumidity and deeidua are by chance similar, but actually are merely degenerative phenomena of an over-ripe mucous membrane. Both periods, the menstrual and the premenstrual taken together, are compared by these authors, in agreement with Heape, to the phase of animal " heat " that the latter author has termed the " procestrum." During this the vulva is swollen and red, blood and mucus exude from the vagina, but the animal is not capable of conception. The time for conception, the "oestrus," corresponds to the postmenstrual period, with its somewhat increased libido, observable also in the human species; whUe the interval is equivalent to the resting stage in animals, the " metoestrum." This view of the matter is difficult to reconcile with the histological phenomena of menstruation; see also later. Section III. III. OBSERVATIONS ON YOUNG OVA. {Implantation, the Embryotrophic Phase of Placentation, and Transition Stages.) Ova which, indeed, do not directly reveal the processes of implantation, bit are young enough to permit definite conclusions concerning it, are that of Bryce and Teacher (1908) and then that of Peters (1899) and that of Leopold (1906). These preparations show, on the one hand, the so-called implantation opening," and, on the other, the extensive proliferation of the chorionic ectoderm or trophoblast which precedes the development of true chorionic villi containing mesoderm. They do not suffice, however, for a certain solution of all the questions which suggest themselves." Thus confirmation, based upon the study of older ova, is much needed of the views regarding the mode of development of the extensive intervillous space and of the formation of a double- layered epithelium on the villi. Of modern, well-described prepara- tions the thoroughly studied and very beautiful o^nim of Jung ^° The implantation opening is also evident in some older ova (Graf Spee, Beneke) ; in others, some of which are very young (such as that of Jung), it is no longer so. " The ovum of Leopold is undoubtedly extensively altered, so that while it is of value for the confirmation of ideas derived from other ova, it is in itself of little significance; the Peters ovum, whose discovery has effected a revolution in our ideas of placentation, and the more recent ovum of Bryce and Teacher are the most important sources of our information concerning the beginning of human development. That the ova described' by older writers (Breuss, Allen Thomson, and especially the celebrated ovum of Reichert) were younger, as Stratz, for example, supposes, is very improbable, since the measurements of their egg capsules were much greater. The methods by which these ova were studied were too imperfect to allow wide-reaching conclusions; and the ova themselves need not be further considered here. Furthermore, as regards the Reichert ovum, Kolliker and later Hofmeier (1896) have, on sufficient grounds, reached the con- elusion that it was not normal. (See also note, p. 104.) 104 HUMAN EMBRYOLOGY. (1908) is the most important; then the equally well-preserved ovum of Siegenbeek van Heukelom, the first that was described under the influence of the newer ideas concerning implantation and placentation; and the preparation of Frassi (1907 and 1908), the study of which has led to conclusive information concerning many of the processes succeeding implantation. Other important objects are the young ova which Graf Spee (1905) and Beneke (1902) exhibited at Congresses, but concerning which only quite brief notices exist ; and, further, the thoroughly described prepara- tions of Friolet (1904), Eossi Doria (1905), and Cova (1907), as well as those of Pfannenstiel (1903) and Marchand (1903). A very young, but unfortunately poorly preserved, ovum is that of Stolper (1906). For a number of questions the older preparations of Merttens (1894), Graf Spee (1896), and Leopold (1897) are of interest.^ ^ The preparations considered here are arranged according to their size in the appended table. Certain difficulties, however, become apparent in the arrangement, since the measurements employed by the authors are not identical.'^ Author. Bryce-Teacher . Peters Leopold . . . . Stolper CTraf Spee . . .Tnng Beneke . . . . Siegenbeek . Eossi Doria . Frassi Friolet Cova Dimensions ill mm. Remarks. 0.6.S (X0.77) 1.6X0.9X0.8 1.4X0.9X0.8 2..5X1-'.2X1.0 2.5X1.5 2.5X2.2 (X 1.1) 4.2X2.2X1.2 5.5X4.5 6 X5 9.4X3.2 11-12X9 Exclusive of epithelium (trophoblast) . Internal space of the egg capsule (exclusive of trophoblast) . Diameter of cavity. Diameter of egg capsule. Exclusive of epithelium (trophoblast). Measurement of cavity of ovum. Including epithelium between the bases of the villi, but excluding the villi themselves. Diameter of the cavity. Embryo with open auditory vesicles, anlagen of liver, hypophysis, etc. '"The number of young ova described in the literature is much greater. Of well-preserved ova there may be mentioned especially those of Etemod and Hitseh- mann and Lindenthal (figures of the latter in Schauta : Lehrbuch der gesamten Gynakologie; and in the Lehrbuch by the author, already mentioned); but of these thorough descriptions have not yet been published. The numerous ova de- scribed before the publication of the works of Siegenbeek and Peters, few of which were observed in situ, the majority being aborted or separated from the egg capsule, are for the most part of little interest in connection with the questions under discussion here, since up to that time the problems were imperfectly under- stood. A mention of these ova would occupy too much space; compare the comprehensive reviews of Peters, Pfannenstiel, and Frassi. " The determination of any measurements from the figures is hardly ever possible; the authors almost never state the magnification. A statement of the objective and ocular, which is generally preferred, is worthless, since for deter- mination of the enlargement the tube length, the height of the stage, and the kind and dimensions of the drawing apparatus are also necessary'. DEVELOPMENT OF EGG MEMBEANES AND PLACENTA. 105 Finally, some preparations of the attachment of the ovum in atypical situations, such as tubal and ovarian pregnancies, are of importance, since they offer opportunities for observing, as in an experiment, the development under modified external conditions and for separating fetal and maternal derivatives. To this group belong, for example, the tubal ova of Fiith (1898), Pfannenstiel (1903), etc., and the ovarian ova of Freund and Thome (1906), Busalla (1907), and Bryce, Kerr, and Teacher (1908; this also contains literature references). A. REVIEW OF THE DESCRIPTIONS OF YOUNG OVA OBSERVED IN SITU. The youngest known human ovum, that of Bryce and Teacher, was already imbedded in the mucous membrane. It consisted of a loose, almost spherical mass of mesoderm, averaging 0.63 mm. in diameter, with wide intercellular spaces, but no coelom : in this mass were two small epithelial cavities (probablv the medullo- r .•?^\ «. _ ^V^ .<:; • ■! - «• V- -M-' &,', Fig. 93. — Transverse section of the Bryce-Teacher ovum (Verh. Anat. Ges., 1908) magnified 60 diameters. The section shows "the point of entrance, with a conical mass of fibrinous material below it pointing to blastocyst ; the implantation cavity is bounded by a necrotic decidua layer and is filled with maternal blood, which bathes a very extensive and irregular plasmodial formation." amniotic and yolk-sack cavities) and it was enclosed by a thick investment of tissue, which is probably to be regarded as chorionic ectoderm only, the trophoblast shell (Fig. 93). This shell, the blastocyst wall, consists (see Verh. Anat. Gesellsch., 1908) (1) of an inner lamella in which the cell outlines are not sharply defined, the nuclei are very irregular in size, and many cells show double, treble or even multiple nuclei; (2) of an extremely irregular formation which has definitely plasmodial characters. These two layers differ very markedly in the characters of the nuclei and in the staining reactions of the protoplasm, but they clearly form parts of one formation. The cellular layer we name, after Hubrecht, the eytotrophoblast, and the plasmodial layer, the plasmoditrophoblast. The eytotropho- blast is confined to the immediate wall of the blastocyst, and there is no sign of 106 HUirAX EMBRYOLOGY. protrusions of tlie cellular layer into the strands of the Plasmodium, although at one or two points a minute bud of cytotrophoblast is seen extending outwards. "The Plasmodium" forms an extremely irregular network, the spaces of which are filled with maternal blood (Fig. 94). Isolated masses of the formation show all stages of vacuolation, from multiple small vacuoles to a spun-out reticular condition. °This vacuolation of the Plasmodium is probably produced by the -"wd^"^ ■?-- "- '^B * .-«■ ■ 'mfr . r-' '^^^^^■B^/ •"- ^^B g E *.!JF^ - '/jy^JBlJPBfc/ '"'= Sb m tfl^V '"*' L.^^Hr^_jBl>^fcZT ' ? e gtJK i'_^^ ^/''^^ ' - '^ j:^^r- '■■ ^'Ww* * •*/ / '^ M^% ' *9^'m^' ^* » /* * '* -»-^ /*- 0e/ 3 '•''^^^r ^ • ^ *' yJTpx-^iM'xr ' "''■ J '^'''■••^^i!l^^"'*'^''^\ ""^ ' £SjO^^^BJBvo irt,^j^ '' „ Bim^HRoi''-' A^k ' a m^^^^^t^^^K^^ ^'^^^^l ^ ^M^'^ '""^^ '•' W '' JUr ' '^^^^^^^jh-' , ':yi^*^-- '" "";"^'^^^sw% ^-- ^ *. '• 1 .^' ■ •••) dec. Fig. 94. — Bla-stocyst wall with cytotiophoblast and syncytium, decidua, and opening of a dilated sinus-like capillary in the implantation cavity, cyt., cytotrophoblast; dec, decidua; end,, endothelium of a maternal capillary; n. z., necrotic zone of the decidua; pi., Plasmodium (syncytium). X 250. (From Bryce-Teacher, Plate V.) secretion of a fluid containing digestive ferments, which cause coagulation necrosis followed by solution of the decidua, thus leading to enlargement of the im- plantation cavity. As the vacuoles enlarge the Plasmodium is reduced to fine strands, and when these break through, the maternal blood takes the place of the secretion in the spaces of the mesh-work." The oval cavity of the decidua, in which the ovum lay, had diameters of " Corresponds to the syncytium of Bonnet (p. 95), since it contains no nuclear divisions. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 107 1.9X0.95X1.1 mm. A small opening, closed by fibrin, and about 0.1 mm. in diameter, placed the cavity in communication with the lumen of the uterus- the opening was not covered by a blood-clot. The wall of the implantation cavity, ex- cept at points where maternal vessels opened into the cavity, was formed of necrotic decidua and fibrm deposits. Only at individual pomts did the Plasmodium quite reach the wall. The glands were enlarged and filled with blood, their epithelium having separated; and the greatly dilated vessels form, especially beneath the ovum, a regular cushion. The decidua is traversed by numerous leucocytes, and all the portions of the uterine mucous membrane that were examined showed Fig. 95. — A portion of the necrotic zone of tlie decidua and the large cells situated on its inner sur- face, n. z., necrotic zone of the decidua: m. c, large, probably maternal, degenerating cells; cav., im- plantation cavity filled with blood. X 350. (From Bryce-Teacher, Plate VI.) decidual changes. In the necrotic decidua zone around the ovum and also lying free in the blood-containing implantation cavity was an almost continuous peripheral layer of large, mostly mono-nucleated cells (Fig. 95), which are perhaps to be regarded as degenerating decidua cells set free by the breaking up of the necrotic zone.'" '^ The comparison of these cells with a layer of fetal cells which occurs upon the surface of the placental anlage in the guinea-pig, a comparison drawn by the authors on the basis of a demonstration by Graf Spee, does not seem to be justified, since the cell layer in question (Duval's ectoplacental entoderm; see also the author's Lehrbuch) owes its existence to the greatly modified inversion of the germinal layers in the giiinea-pig. 108 HUMAN EMBRYOLOGY. The preparation was obtained from an abortion which occuri'ed sixteen and a half days after the only cohabitation that needs consideration and ten days after the failure of the expected menstruation. The microscopic picture is very strange and striking and cannot be compared with any stages of placenta formation known in animals ; " it is however, as the authors state, quite reconcilable with the newer theoretical deductions concerning implantation and the commencement Tr. a. m. Chz. Fig. 96. — A section through the Peters ovum and the surrounding portion.^ of the uterine mucous membrane. BL, blood lacunge; Ca., capsularis; m. C/i2., mesodermal axis of the first chorionic villi; Co., decidua compacta; Dr., glands; .ff., embryo; G., maternal vessels; Sc, closing coagulum (Peters's fungoid tissue); 5y., syncytium; Tr., trophoblast; t/e., uterine epithelium; C/z,, zone of enclosure. The opening in the capsularis extends from a to 6. X 50.* (After Peters, 1899. Compare also Fig. 97.) of the placenta formation. Among the most striking peculiarities in comparison with what is found in older preparations are: (1) the structure of the trophoblast shell; (2) the smallness of the implantation opening; (3) the necrotic character of the wall of the egg chamber and the absence of a mutual penetration of the fetal and maternal elements. Whether the preparation can be regarded as "Whether the extensive syncytial formation described by Strahl (1906) in young stages of Mijrmecophaga, Dasi/pus, Dendrohi/rax, and Aluata, and by Duck- worth (1907) in 3Iacacus, is comparable with that in the human ovum cannot be determined, since Strahl gives no figures and those of Duckworth concern a somewhat later stage. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 109 absolutely normal must provisionally be left undecided; it comes, on the one hand, from an abortion, and, on the other, it was preserved only after having re- mamed for twenty hours in a mixture of urine and blood serum. Nevertheless, mitoses are still distinguisliable in the cytotrophoblast cells, and the general im- pression furnished by the preparation, which was demonstrated at the Congress of Anatomists at Berlin, 1908, is distinctly favorable. In the absence of other equally young ova our views concerning placentation must, for the time being, be brought into harmony with this preparation. The conditions in the Peters ovum are quite different. The cavity of the ovum contains the magma reticulare with the anlage of the embryo and the body cavities" (Figs. 96 and 97), and has diameters of 1.6X0.9X0.8 mm.; ex- ternal to the chorionic mesoderm is a layer of closely packed cells, which is Fig. 97. — Cytotrophoblast and syncytium of the Peters ovum. The embryonic structures are shown diagrammatically. Ah., amniotic cavity; Ds., yolk sack; Ic.~R., intercellular cavities of the mesoderm; Lh., body cavities (cf. Grosser; Lehrbuch); M. r., magma reticulare; Sy., syncytium; Tr., (cytojtropho- blast. (From Peters, Plate I, copied under control of the preparation itself.) traversed by wide blood spaces and surrounds the entire ovum like a shell or mantle having a thickness of 0.5 mm. or more. Into this cell mantle, which is thicker toward the muscularis than toward the surface of the mucous membrane, there project everywhere short, stout processes of the mesoderm, the anlagen of the mesodermal axes of the villi. The cell mantle, on account of its relation to the mesoderm of the ovum, can hardly be interpreted otherwise than as the chorionic ectoderm, trophoblast, and trophoderm." Peripheral to this trophoblast shell lies a layer of tissue which Peters terms the transition zone and which contains, imbedded in an oedematous stroma, a confused mass of maternal, and apparently also of fetal, cells, together with a large number of free blood-corpuscles. The entire ovum, vrithout projecting beyond the level of the mucous membrane, lies beside a fold of the membrane, " For details concerning the formation of the body cavities of the Peters embryo consult Grosser's Lehrbuch. " Compare, however, the ovum of Beneke described below, and Disse's in- terpretation of it. 110 HUMAN EMBRYOLOGY. imbedded in the decidua eompacta, which over the dorsal surface, the summit of the ovum, is defective over an area of about 1 mm.; throughout this region the uterine epithelium, elsewhere well preserved, is wanting. The egg does not project through this defective area freely into the uterine lumen, but is separated from it by a fibrin clot that closes the opening in the eompacta and spreads out laterally like a fungus growth (Figs. 96 and 98). This clot is termed by Peters the fungoid tissue or blood-fungus, and later by Bonnet the dosing coagulum. The decidua over the entire surface of the uterus is high and swollen, and is divided by furrows into distinct areas; its separation into compact and spongy layers is distinct only in the neighborhood of the ovum, for, although enlarged glands with epithelial papillse occur elsewhere, yet these occupy almost the entire thickness of the mucous membrane, so that a superficial compact layer is not distinct. Typical decidua cells cannot be found, although some large cells of Ue. Sc. i^^^jstatijwiu.-. ■O* - «'~»j5,-^'-^ \ ^V^l^^ /f•„'.=;:v,..„"-v':.®.•:c■.■• b. Tr. Fig. 98. — Summit of the Peters ovum. BL, blood lacunae; Ca., capsularis; 5c., closing coagulum; St,, its stalk; Sy., syncytium; Tr., trophoblast; Ue., uterine epithelium; Ue. R., the crumpled border of this; a., trophoblast nucleus in the syncytium; b. and c, preparatory stages of the syncytium (wreath-like deposit iu a blood lacuna.) (From Peters, 1899.) irregular shape and with large, deeply staining nuclei occur in the vicinity of the ovum; the significance of these is, however, obscure. The entire mucous membrane, in which very greatly enlarged blood-vessels occur, especially in the neighborhood of the ovum, shows signs of oedematous infiltration, which increases in distinctness nearer the ovum; in this region extravasated red and white blood- corpuscles also occur. A new formation of blood-vessels occurs especially in the zone of tissue which intervenes between the ovum and the uterine lumen. The glands in the neighborhood of the ovum curve around this and open near it upon the surface of the uterus; beneath the ovum are closed glandular spaces filled with blood, which show no connection with the egg capsule. A number of important points are still to be noticed concerning the tropho- blast layer. The principal part of the layer consists of completely separated cells with pale protoplasm and large, deeply staining, round or oval nuclei. On account of the size and staining properties of the nuclei the entire trophoblast shell appears dark even under weak magnification. Throughout its entire extent it is traversed by blood lacunae, some large and some small, which are continuous one with the DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. Ill other; these lacunae, some of which approach so closely to the chorionic mesoderm as to be separated from it only by one or two layers of cells, are everywhere com- pletely filled with well-preserved maternal blood. At various places they are in connection with venous vessels, which possess an endothelial wall only in the transition zone; the opening of arteries or capillaries into the lacunae cannot be made out. " The most peripheral lacuna are, for the most part, separated from the decidual tissue by a thin covering of ectoblast arranged in concentric layers; but in places diverging tracts of trophoblast stream out into the compacta, and the blood spaces lying between these lack the ectodermal covering on their periph- eral surfaces." In the most central portions of the trophoblast cells are to be found with feebly staining and distended nuclei, with vacuoles, nuclear fragments, Fig, 99. — A portion of the periphery of the trophoblast shell of the Peters ovum. Degenerating patches of syncytium with greatly enlarged nuclei ; the blood-corpuscles, for the most part, only resting upon the syncytium. Tr., cytotrophoblast; Uz., enclosiug zone. X 350. and flakes. More peripherally the distention and degeneration of the nuclei in- creases, the cell boundaries vanish, and there are formed very irregular, large, vacuolated masses of protoplasm, with numerous, irregularly contoured, and ex- ceedingly large nuclei (Figs. 99 and 102). These masses constitute the syncytium of the Peters ovum, which is, accordingly, united by all possible transitions with the cellular trophoblast; it is never separated from this by a limiting membrane. Prickle processes cannot be seen ; at most there is " a delicate and thin, strongly refractive deposit, slightly frayed at the edges, on the surface of the syncytium." The syncytium completely clothes, except at a few places, the blood lacunee with a thin layer; indeed, according to Peters, the formation of the syncytium seems to be produced by the contact of the trophoblast with the maternal blood. Further- more, it would seem, according to his ideas, that degenerating red and white blood- corpuscles may be " transformed " into a syncytium, which applies itself to that formed by the trophoblast, so that the blood with its own structural elements may be concerned in the formation of the syncytium (compare Fig. 98, "wreath-like 112 HU:\IAX EMBRYOLOGY. deposits in the lacunffi representing preliminary stages of the syncytium )• The syncytium occurs, as a rule, only at regions where there is contact ^^1th the maternal blood Trophoblast and synevtium are frequently mingled with elements of the maternal tissues in the transition zone, and, like these, undergo degeneration m that region, since free maternal and fetal nuclei can be found m it; the tropho- blast and syncytium also frequently replace the wall of a gland and project mto its lumen, and they may form the walls of the maternal blood-vessels in the peripheral portions of the trophoblast shell— sometimes by forming one wall of the vessel while the opposite one remains formed by normal epithelium, sometimes in that over the entire wall onlv the epithelium, either intact or in fragments, separates the syncytium from the cavity of the vessel (Fig. 100). A transition between the endothelium and the syncytium is never recog-nizable. Peters's preparation was obtained from the uterus of a suicide, poisoned by caustic potash on the third day after the omission of a menstrual period. The Bl.-L. p. En. Fig. 100. — Formstion of the intervillous spaces in the Peters ovum. " The capillary still possesses an epithehal wall on the side towards the trophoblast; the blood from the lacuna seems to have broken through at two places (a and 6). Tr., trophoblast; U.-Z., enclosing zone; Ca., capillary; p. En., peripheral endothelium; En., endothelium of trophoblastic side of capillary; Bl.-L., blood lacuna; Sy., syncyt- ium." CAfter Peters, 1899.) mode of death may not have been without influence on the blood engorgement of the uterine mucous membrane. Peters estimates the duration of the pregnancy at from three to four days; details concerning this are given elsewhere. The preservation of the ovum (the autopsy was performed a few hours after death) with the exception of the caudal end of the embryonic anlage, was very good^ even although mitoses are not recognizable in the trophoblast cells. The Leopold ovum, which was obtained from a case of phosphorus poisoning, concerning which further data are not available, differs in several points from that of Peters. Thus, the trophoblast growth was less extensive ; and the blood la- cunEe, which were in open connection with the neighboring capillaries, were unusually wide. The trophoblast cords were, for the most part, reduced to one or two layers of cells, and were to a large extent covered by syncytium. In addition to the ovum, separated portions of the syncytium were also to be found in the decidua, " Peters, in a private communication, now regards these and similar structures as rather the expression of embryotrophic processes. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 113 having evidently wandered into it; but, on the whole, there was less syncytium than m the Peters ovum, and its individual parts were smaller. It was every- where clearly distinguishable from the vascular endothelium. The contents of the trophoblast shell, the mesoderm of the ovum, were irregularly shrunken and con- tained maternal blood-corpuscles; an embryonic anlage could not be distinguished (it had probably been already destroyed). The internal diameters of the cavity of the ovum were 1.4 X 0.9 X 0.8 mm. This ovum was also imbedded, somewhat superficially, in the mucous membrane in the neighborhood of a furrow, but a differentiation of the mucous membrane into compacta and spongiosa was not distinct. The regions surrounding the ovum were very rich in glands, which curved around the ovum. Doreal to the ovum the deeidua, in contrast to that of the Peters ovum, is closed except for a small opening, so that a deeidua capsularis is present. Extending outward from the opening upon the surface of the mucous membrane is a clot, consistmg of blood and fibrin, which corresponds to the fungoid tissue or closing eoagulum and is termed by Leopold the fibrin cover. On account of the absence of the embiyonic anlage the stage of development cannot be accurately determined; the measurements are not sufficient for this purpose, since, as has been pointed out, the contents of the ovum were shrunken and the blood laeunse enormously distended. The relatively scanty growth of the trophoblast may indicate, as Bi-yce and Teacher remark, that the ovum was younger than that of Petere; the small size of the opening in the capsularis may also have the same significance (see below, p. 117). However, definite conclusions cannot be drawn from the jareparation. The ovum described by Stolper (1906) seems to represent a very young stage, but it had evidently died some time before its abortion. The diameters of the egg capsule (2.5 X 2.2 X 1-0 mm.) are, at all events, smaller than in the two ova to be described next, but for which the corresponding measurements are not given. The embryo was macerated. The ovum is characterized by a very extensive develop- ment of syncytium. Wide blood spaces, probably intended for a diminution of the blood pressure in the communicating vessels, are regularly arranged around the inteiTUlous space. The results obtained by Graf Spee (1905) from a young ovum studied by him have been stated only briefly, and until a thorough study of the ovum and figures are available, a comparison with the ova already mentioned cannot be made. The ovum was obtained from a case of oxalic acid poisoning; the mucous membrane of the uterus " showed the areas, divided by furrows, that are character- istic of pregnancy," and in one of these at a point marked by a slight depression of the surface was the ovum. " Beneath about two-thirds of the free surface of the prominent area of mucous membrane, imbedded in a cavity in the inter- glandular connective tissue of the uterine mucosa, was an ovum measuring 1.5 X 2.5 mm. in its greater diameters, poorly provided with villi, and with very small embryonic structures in the anterior. Between the surface of the chorion and the uterine tissue were here and there small quantities of blood from open blood- vessels. The walls of the egg chamber consist throughout of elements of the interglandular connective tissue of the uterus. The lumina of all the glands open into the uterine lumen ; none into the egg chamber. The portion of the mucous membrane (serotina) inter^'ening between the ovum and the muscularis holds a large mass of blood (just as in the ovum of Peters) contained in enormously enlarged endothelial canals and apparently stagnant even in life ; it may very well have furnished nutrition to the ovum and at the same time have served as a rampart protecting the parts of the mucous membrane near the muscularis from the destructive contact action of the ovum. The walls of the egg chamber, separating the ovum from the lumen of the uterus, consisted of a thicker or thinner layer of the interglandular connective tissue next the ovum and a single-layered epithelial covering next the uterine cavity. Only in the region of the surface Vol. I.— 8 114 HUMAN EMBRYOLOGY. depression is the uterine tissue interrupted by an opening, which may be regarded as the point of entrance of the ovum into the uterine mucous membrane, the implantation opening; it is closed only by a flat expanded blood-clot (fibrin with enclosed leucocytes and red blood-corpuscles). The conditions are, accordingly, very similar to those occurring in the human ovum described by Peters. " The implantation opening, at this stage 0.8 mm. in diameter at the most, has probably increased somewhat in size from what it was when first produced by the ovum by stretching and growth, and perhaps also by histolysis of the chamber wall, for I imagine that the ovum during the seven days which probably intervene between fertilization and implantation cannot have increased much in diameter and therefore cannot have measured much over 0.2 mm." Among these data the most striking are the small amount of blood in the iromediate neighborhood of the ovum and the small number of villi. Nothing is stated concerning the character of the chorionic epithelium, the syncytium, and the intervillous space. The ovum, nevertheless, does possess villi and in this respect is further developed than Peters's preparation. It is questionable, how- ever, if it is to be regarded as quite normal. The ovum described by Jung agrees excellently in its general character with that of Peters. It was obtained from a curetting, and, completely surrounded with mucous membrane, was preserved, while still fresh, in 80 per cent, alcohol. Its age was not determined. The egg capsule was completely separated from the lumen of the uterus, but was situated somewhat superficially. The somewhat compressed but uninjured cap of the ovum was composed of coagulated blood and tufts of fibrin, together with numerous leucocytes and degenerating decidua cells, and passed gradually over into the transition zone. The diameter of this necrotic cap was 1.7 mm.; it corresponds, according to the opinion of the author, to the closing coagulum of other ova and its extent indicates that of the distended implantation opening.™ The diameter of the ovum exclusive of the chorionic epithelium was 2.5 X 2.2 (Xl-l) mm. The mesoderm of the chorion had already sent processes, the mesodermal axes of the villi, into the extensively developed trophoblast (Jung avoids the use of the term trophoblast and speaks only of ectoblast). The ovum was completely surrounded by rudiments of villi, all of about the same length; and on the chorion membrane and at the roots of the villi was an epithelium, consisting of two layers, a basal and a covering layer. The basal layer, composed of distinctly defined cells, passed over into stout columns of cells, which frequently united and so formed the shell around the ovum. Only occasionally did free cell-columns occur, the representatives of free villi. In the peripheral portions the individual cells were somewhat larger and clearer, but everywhere abundant mitoses could be observed and there were no signs of degeneration in the peripheral elements. Individual mitoses were so placed that one of the daughter cells passed into the covering layer, this, the syncytium, showing no mitoses although the nuclei were, for the most part, well preserved. The protoplasm of the covering layer presented, in general, a foamy structure; prickle processes projected toward the intervillous space, at least in certain places. Only at certain regions of the periphery was the covering layer, which frequently streamed into the maternal tissues, in degeneration and forming a symplasma syncytiale in Bonnet's sense, perhaps as the result of the action of maternal leucocytes. The vacuoles of the covering layer at places contained what seemed to be altered maternal blood, but everywhere '° In this case the capsularis would not actually be closed. The gradual transition of the cap into the transition zone on the lateral portions of the ovum seems to be opposed to Jung's view. It is possible that the capsularis had at one time been complete, but was again undergoing degeneration. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 115 a continuity of the syncytium with the maternal tissues, that is to say, with endothe- lium, was lacking. The boundary between the maternal and fetal tissues was almost everywhere easily recognizable, with some difiaeulty only in the regions where symplasma structures occurred. Between the trophoblast columns, that is to say, the anlagen of the villi, there was a very irregular intervillous space; this was abundantly filled with blood, was in continuity with the maternal vascular system by means of gaps in the sieve-like trophoblast shell, and was lined by maternal tissue (endothelium) only in the neighborhood of these gaps. The communications with the maternal vessels were always narrow and the circulation must have been very slow. In the tran- sition zone, situated outside the trophoblast shell, degenerating maternal tissue occurred, partly associated with the formation of symplasmata, but always with- out signs of active proliferation; also no new formation of blood-vessels could be found. Around the ovum was a strip of fibrin of varying thickness, produced by degeneration of the maternal tissue (stroma, endothelium, gland epithelium). Leucocytes occurred abundantly in the transition zone, but not in the fetal tissues. The glands were frequently destroyed by the trophoblast, but appeared to with- stand its attacks for a longer time than the rest of the decidua; none of them opened into the intervillous space, but they took a curved course around the ovum. External to the transition zone a separation of the compacta and spongiosa had taken place. The former was oedematous, beset with numerous lymphocytes, and its gland ducts were contorted; typical decidua cells and hemorrhages were want- ing, although the glands frequently contained clotted blood. The representatives of the later decidua cells showed numerous mitoses. No oedema occurred in the spongiosa. The ovum of Beneke was obtained from a curetting twenty-five days after the omission of a menstrual period and was fixed in alcohol. It contained an embryo measuring 1.86 mm. in length, and had a cavity of 4.2 X 2.2 X 1-2 mm., surrounded by a trophoblast measuring 0.4-1.0 mm. in thickness. " As regards the structure of the trophoblast the author can only confirm in general the obser- vations of Siegenbeek, Peters, Marchand, and others." The syncytial giant cells are throughout of fetal origin and symplasma formation is not recognizable. The syncytia had encroached upon the endothelium of the decidual vessels and also upon the epithelium of the glands; by the development of extensive clefts in the interior of the giant cells the intervillous blood spaces are being formed. Scattered giant cells with prickle processes occur in the chorionic connective tissues of the investment of the ovum; they wander to a certain depth into the decidual tissue, where they may be recognized by their characteristic nuclei and by con- taining glycogen. The decidual cells, extensively swollen, take part in the formation of the so-called transition zone to a greater extent than has been supposed by Peters, for example. The closing ' tissue plug ' which fills the opening in the reflexa corresponds in general in its histological constituents, blood, fibrin, leucocytes, etc., with what Peters has described." The measurements of Siegenbeek's ovum (4.5 X 5.5 mm.) were not made directly, since the ovum had been opened by a tear at one spot and was collapsed, but were estimated from the perimeter. It was obtained from a woman who had met an accidental death from burning; the entire uterus was preserved in formalin fourteen hours after death. The ovum was completely covered with villi, those on the basal surface being stronger than the peripheral ones, and those about the equator of the ovum the strongest of all. Free villi occurred; the majority were continued into cell columns (ectoblastic trabeculae), which united ^ Disse, who has subsequently studied the specimen, regards the entire trophoblast of the ovum as maternal tissue and also transfers this same inter- pretation to the Peters ovnm. 116 HUMAN EMBRYOLOGY. together peripherally and formed an ectoblast shell traversed by large and small spaces, and varying in thickness in different regions. In general it was thicker on its peripheral than on its basal side. The ectoblast cells situated near the maternal tissue were larger than those having a more central position and frequently showed degenerating nuclei, but mitoses occurred in all portions of the ectoblast shell. The boundai-y between the fetal and maternal tissues was difficult to make out in certain regions. The intervillous space was formed by blood-lilled lacunas which were lined only by cellular ectoblast or by syncytium; the endothelium was also frequently wanting in the blood-vessels at their communication with the intervillous space. In the space were very many leucocytes and perhaps also sjjecial nucleated elements of the maternal blood. The syncytium was only to be found in the region of the blood paths; it showed no prickle processes and no cuticula on the side next the ectoblast. The derivation of the syncytium from maternal tissues (endothelium, epithelium, or connective tissue) was excluded, but its continuity with the cellular ectoblast could not be made out, so that the origin of the tissue could not be determined. The ovum lay in the compacta, whose basal portion had the same structure as the capsularis (reflexa), except as regards the occurrence of glands. The capsularis, for the most part, lacked an epithelium and contained in its interior filirin striae. Basally there were greatly distended glands filled with blood; around the periphery of the ovum the glands were arranged concentrically, and the glandular epithelium did not form syncytia. Characteristic decidual cells were nowhere present. A sharp separation of the compacta and spongiosa had not yet occurred in the decidua vera; the compacta was oedematous."" The ovum of Rossi Doria was obtained from an abortion. It was injured by a tear and does not seem to have contained an embi-yo. The egg membranes were rather well preserved. The theoretical considerations of the author will be discussed in note 23. Frassi's ovum was obtained from an operation fourteen days after the omission of a menstrual period ; the unopened uterus was preserved in formalm. The ovum of Friolet was also obtained from an operation and fixed in the unopened uterus. Both ova already showed, for the most part, a two- layered epithelium over the villi; a criticism of the obsen-ations made upon these ova will follow in the resume. The ova of Peters, Jung, Beneke, and Siegenbeek, with their extensive development of the trophoblast, form a single harmonious group, which may be derived from conditions such as Bryee and Teacher have described. To the older stages, on the other hand, a natural and easy transition is formed by the Siegenbeek ovum. B. RESUME OF THE FIRST PROCESSES OF DEVELOPMENT IT TO THE FORMATION OF THE VILLI AND THE APPEARANCE OF THE INTERVILLOUS SPACE. From the foregoing tlie course of the first stages of develop- ment may, with a good deal of certainty, be concluded. Especially is this so with regard to the implantation of the ovum. Of the types of implantation mentioned in the introduction the inter- stitial is the only one that concerns us here; as in the guinea-pig, so in the human species, the ovum penetrates like a parasite, through an opening that it forms for itself, into the mucosa and '^ The ovum has more recently been studied by Veit (1905) ; the author has not been able to accept the idea of an active penetration of the ectoderm into the maternal tissues. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 117 develops there (Berry Hart, Graf Spee, Von Herff, Peters ^s). It must at this time be very small, since otherwise such a penetration of the entire ovum could not be readily understood. The opening in the surface of the mucous membrane had a diameter of 0.1 mm. in the Bryce-Teacher ovum, in that of Leopold its margins were in contact, in that of Peters its diameter was 1 mm., and in that of Graf Spee 0.8 mm. It is probably enlarged very quickly by the growth of the ovum, and the diameter of the ovum at the time of implantation is probably about 0.2 mm. (Graf Spee). The forma- tion of the mesoderm cannot have begun, and it is questionable if at this time even the cavities of the ovum (the blastocoel, meduUo- amniotic cavity, cavity of the yolk sack) have appeared. The ovum of the guinea-pig forms at the moment of implantation a solid cell mass (Graf Spee). A marked growth of the ovum is dependent on favorable conditions of nutrition, and these are furnished only after implantation. The ovum does not undergo implantation in a furrow,^^ but at any portion of the smooth mucous membrane where perhaps a special thickening or an extravasation of blood, which Taaj serve as an embryotrophe, facilitates implantation. The spot is usually on either the anterior or posterior wall of the uterus, and determines the situation of the placenta. The implantation usually occurs between two glands, and the glands are later forced apart by the growth of the pene- trated ovum, so that they bend around it in curves. Implantation in a gland is not probable, since the diameter of the ovum is always greater than that of the lumen of a gland. During the implantation the superficial epithelium and connective tissue are dissolved and probably serve the ovum as embryotrophe ; the solu- tion of the maternal tissues may perhaps be produced by the action of ferments secreted by the ovum, and to this Bryce and Teacher refer the vacuolation of the syncytium seen in young stages. The penetration of the ovum is not determined by gravity, since the minuteness of the ovum places this out of the question and, further- more, the implantation takes place just as often contrary to the ^ Rossi Doria believes in a kind of combination of penetration and ciroum- vallation, since the ovum observed by him projected, for the most part, beyond the level of the mucous membrane. The ovum was, however, much too old to settle the question. He had to do, apparently, with a superficially implanted ovum. " The mucous membrane of the non-gravid uterus never shows, even at the greatest development of the premenstrual swelling, a formation of furrows and elevations (compare Hitschmann and Adler) ; consequently the idea, frequently expressed, that the implantation takes place in a furrow, as in the hedgehog, fails. The formation of furrows is actually a symptom of pregnancy (Graf Spee) and as such may direct the attention to the possibility of a young pregnancy in autopsies, at a time when the ovum itself can scarcely be recognized. The furrows are not preformed, but are produced as foldings of the continually thickening mucous membrane. 118 HUMAN EMBRYOLOGY. direction of gravity as in accordance with it on either of the opposite walls of the uterus. Nor can the action of an internal pressure by the uterus (Pfannenstiel) be assumed, since the ovum floats in a quantity of detritus which it produces and which cannot flow away on account of the swelling of the mucous membrane, but is rather increased in quantity by the flow of additional material from neighboring tissue spaces. There remains then only the sup- position of an active penetration on the part of the ovum which may be due to an amoeboid activity of the superficial cell layers of the trophoblast (Peters), in favor of which evidence has been obtained within recent times. Indications of an active penetration by the ovum have also been furnished by young tubal pregnancies (Filth, Aschoff, and others), in which the ovum has completely destroyed the thin mucous membrane and has penetrated into the muscularis. The fact that the tube is open at one extremity is ample evidence of the non-existence of an internal pressure which could force the ovum into the muscularis. The duration of the implantation process, which in the guinea-pig is about eight hours, may be estimated at about one day in man (G-raf Spee). ^Vs to the behavior of the ovum before implantation we rely solely on conjecture based on a comparison of what occurs in animals. The fertilization of the ovum set free from its follicle probably takes place, as a rule, in the pars ampullaris tubae, to which the spermatozoa penetrate and where they may remain capable of fertilization for days or perhaps for weeks (see, for example, His: "Anatomic menschlicher Embryonen," vol. ii). The fertilized ovum then wanders down the tube and through the uterus until it reaches the place of implantation ; this movement is a passive one on the part of the ovum, being caused by the action of the cilia of the surface of the tube and uterus. During this time the ovum loses its corona radiata ^^ and zona pellucida, and passes through the first stages of development, that is to say, the segmentation; it obtains the necessary oxygen from the serum which moistens the mucous membrane and perhaps employs the secretions of the mem.brane as embryotrophe (p. 1 19) . The passage through the tube to the implantation region is by no means rapid; even in the white mouse, where the distance to be traversed is very short, it occupies five or six days (Sobotta, Melissenos), in the guinea-pig seven days (Graf Spee), and in larger animals, such as the cat, dog, pig, and sheep, from eight to ten days (Bonnet). Taking into account the length of the human tube, the assumption that the wandering of the human ovum occupies eight or ten days is quite reasonable, notwithstanding that the human ovum is rela- tively small and the rapidity of the wandering increases, in general, " The theory of Hofmeier that the corona radiata is retained and becomes transformed into the syncytium is of only historical interest. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 119 with the smallness of the ovum (Minot places the wandering period at eight days, Graf Spee at seven days, Pfannenstiel at from five to seven days, and Bryce and Teacher at seven days). Of the various phases of the menstrual cycle, the premenstrual is the most important for implantation; at least so the study of the phenomena of menstruation seems to indicate. The pre- menstrual loosening of the tissues would favor the penetration of the ovum, the secretion of the glands would serve as embryotrophe until the completion of implantation, and the mucous membrane of the uterus in the cases of Peters, Leopold, Jung, and Siegen- beek resembles much more a premenstrual membrane than a decidua. The connective-tissue cells are, in Peters 's case, for ex- ample, less plainly altered toward the decidual condition than they are normally immediately before the appearance of the menses, and this even although the time of menstruation was several days overdue.^'' Indeed, even in older ova, such as that of Frassi, typical decidual cells occur only in the neighborhood of the ovum ; and among all the young ova a distinct decidual alteration is to be found only in that of Bryce and Teacher. We must assume that the implantation exercises an inhibiting effect on the premenstrual changes, for otherwise menstruation would not be omitted during pregnancy ; and the delaying of the decidual changes in the uterine connective tissue may be regarded as the visible expression of this inhibition. Yet a certain amount of time must be granted the ovum for the development of this inhibitory action; an ovum im- planted immediately before menstruation may well be sacrificed to this process; and such menstruations would then perhaps be abundant in quantity. Normally (typically) therefore the im- plantation must take place several days before the time for the appearance of the menses, but whether two or five days previously cannot at present be determined. Perhaps two days is too short an interval to allow the inhibitory action to become efficient. If the times required for the passage through the tube, the implantation, and the inhibition of menstruation be added together, it follows that the expulsion of the ovum from its follicle and its fertilization must normally occur at a minimum of about from eleven to fourteen days before the date of the expected menstru- ation. But this entire interval has been almost always neglected in gynaecological literature, in accordance with the tables established by His, and the age of the ovum has been determined from the '° The decidual cells are in any event to be derived from the stroma cells of the uterine mucous membrane, and the various older theories (derivation from perivascular cells, the now almost forgotten " perithelia," or leucocytes, etc.) are negligible. Concerning the mitoses observed by Jung in the preparatory stages of the decidual cells it is to be remarked that they furnish an explanation of the at first rapid increase of the decidua. 1^0 HUMAN EMBRYOLOGY. estimated time of appearance of the omitted menstruation. Conse- quently nearly always the age estimates have been too low by the amoujit given above. The interval between implantation and the beginning of the expected menstruation has been considered by Peters and Leopold, for instance, but they neglected the time required for the passage through the tube. If one reckons from the moment of fertilization, the Peters ovum must have been at least fourteen days old (and implanted for about five days). Implantation may, however, be possible in other phases of the menstrual cycle than the premenstrual, and it may be that the stimulus arising from the ovum may also have the property of accelerating the occurrence of the premenstrual changes. Per- haps certain pathological phenomena may be associated with pre- cocious implantation (see Grosser "Lehrbuch"). The view stated here is, however, scarcely in agreement with the age estimates that have so far been published of various young human ova. Bryce and Teacher, on the basis of an analysis of twelve cases, reach conclusions quite at variance with that given above, — namely, that menstruation is actually without influence on conception and implantation; that, indeed, the latter may take place on the day immediately before or after the calculated date for the first omitted menstruation; and that, accordingly, it is not the implantation that is responsible for the inhibition of the ap- proaching menstruation, but the fertilization which has already taken place in the ampulla of the tube. These authors, however,^ start with the assumptions that fertilization occurs, on the average, twenty-four hours after coition, and, secondly, they base their calculations on a series of aborted ova as well as upon some others which were obtained by operative interference necessitated by pathological conditions of the uterine mucous membrane. If one considers, on the one hand, how much uncertainty exists regarding the time relations of the processes of fertilization and, on the other hand, the fact that only two cases of normal pregnancies termi- nated by extrinsic causes (Peters, Eeichert) occur in their tables, it may seem venturesome to set aside as without significance the relationship of the premenstrual mucous membrane to the decidua, which is capable of being directly observed. Cases in which a spontaneous abortion occurred or in which there was a catarrh of the mucosa which called for curetting and which, if longer continued, would have produced a spontaneous abortion, may, in- deed, be associated with an implantation in an improperly prepared mucous membrane. The occurrence of typical decidua in the Bryce-Teacher ovum is strange when compared with other results (see p. 119). But at all events these authors have rendered the service of having thrown full light upon the obscurity which pre- vails concerning the course of the phenomena under discussion. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 121 The normal period of ovulation is also still quite uncertain. Ovulation may take place at any time ; the prevailing view is that it coincides with menstruation ^^ (see Nagel: "Handbuch der Physiologie"), while Ancel and Villemin (1907), on the ground of their observations of freshly ruptured follicles, suppose that it occurs, on the average, twelve days before the begianing of men- struation. The latter period is in excellent agreement with the view that the premenstrual phenomena are preparations for preg- nancy ; this can hardly be said of the former one. The ovum ceases its penetration in the decidua compacta; the implantation opening is closed by the coagulation of the tissue lluids exuding from the mucous membrane, and the product of this coagulation is the closing coagulum (fungoid tissue, tibrin cover) which occurs in a whole series of young ova (Peters, Leo- pold, Beneke, Graf Spee) and, consequently, can hardly be re- garded, as Pfannenstiel would wish, as an abnormal occurrence. In the Bryce-Teaeher ovum the coagulum is wanting and the authors suppose that it is first formed after the ovum has increased in size and the implantation opening enlarged. The implanted ovum begins to grow rapidly and presses further into the mucous membrane, so that it divides this into a superficial and a deep layer (Fig. 101). The superficial layer becomes the covering of the ovum on the side toward the cavity of the uterus, it becomes the decidua capsularis ,^^ which at first bears the implantation opening, ^'According to Leopold and Ravano (Arcliiv f. Gyn., vol. Ixxxiii, 1907) ovulation coincides with menstruation in about two-thirds of the observed cases, but in one-third of them it occurred quite independently of it ; conception is possible at any time. The authors estimate the period of ovulation from the condition of the corpus luteum; but this estimate must necessarily be uncertain, since, in view of the uncertainty of the time of ovulation, a basis for a thorough knowledge of the time required for the development of the corpus luteum is lacking. Also the observations of H. Bab (Deutsch. med. Wochenschr., 1908) indicate that impregnation, and consequently also ovulation, takes place some days before menstruation; nevertheless, it is as yet hardly possible to draw conclusions as to the time of impregnation from the size of the embryo, as this author does. Com- pare, for instance, the data furnished by Bab concerning his first two cases with those given by Tandler (Anat. Anz., vol. xxxi, 1907) concerning an almost equally developed embryo. The discussion whether the ovum belongs to the first omitted (Lowenhardt-Sigismund) or the last completed menstruation, a discussion in which Bab declares himself in favor of the former idea, arises from the old notion that ovulation and menstruation, on the one hand, and fertilization and implanta- tion, on the other, coincide. The latter coincidence has been disproved; the former is improbable, or at least requires demonstration. "" The decidua capsularis is the decidua reflexa of the older terminology. The latter name is an expression of the older theories (W. Hunter, Reichert) of its origin, to the effect that the mucous membrane was reflected or curved over the ovum and fused over it. Since, however, young stages are opposed to this view and older ones show no conditions that cannot be explained as well or even better as the results of interstitial implantation, this theory, which up to ten years ago was the only prevailing one, is now regarded as disposed of. 122 HUMAN e:mbryology. but later completely closes (see below). The deep layer, or what remains of it, forms the basis of the later placenta and is the decidua basalis (the decidua serotina of the older nomenclature) ; lateral to the ovum is the decidua marginalis, whose fate is of great importance for the later stages of development. The re- maining mucous membrane forms the decidua vera, recently very appropriately termed the decidua parietalis by Bonnet. Fig. 101. — Pregnancy of the first month. The ovum expelled with the entire decidua; the decidua capsularis and chorion have been cut through and the intervillous space and extra-embryonic body-cavity ■opened. Ch., chorion; Dz,, decidua capsularis; Dp,, decidua parietalis; E,, embryo in amnion. X IJ. There are four structures that still require thorough discus- sion: the trophoblast shell, the syncytium, the blood lacunae, and the transition zone. The trophoblast shell is to-day regarded unanimously, if we neglect Disse's view, as embryonic ectodermal tissue, as tropho- blast (cytotrophoblast, trophoderm).^^ To it is also generally ascribed the power of dissolving and absorbing the maternal tis- " The view advanced at one time by Langhans, but since relinquished, that the layer of separate cells upon the surfaces of the villi, arising from the tropho- blast, was derived from the fetal mesoderm and that only the syncytium corre- sponded to the chorionic ectoderm has recently (on the last occasion in 1904) been revived by Van der Hoeven, but without sufficient evidence. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 123 sues. Also analogies for the fact that it surrounds the ovum as an extensive growth are to be found among animals, namely, in the hedgehog.3« That during its growth toward the maternal tissues portions of the trophoblast also are destroyed,^^ and that a zone of mutual penetration by the tissues, a transition zone, occurs, are also phenomena frequently to be observed in animals. The trophoblast shell usually develops more extensively on the basal side of the ovum, where the nutrition is best (Peters), and there, for the same reason, ai'e formed the embryonic anlage and, later, the placenta ^^ (Von Franque, Peters). The syncytium and the blood lacunce. are associated topograph- ically and perhaps genetically also. The former has been the most disputed tissue in the whole field of histology, and even to-day it is not yet thoroughly understood. Of the different opinions as to its origin that have been advanced from time to time only two need further consideration ; ^^ the one derives the syncyt- ^ The investigation of the hedgehog we owe to Hnbrecht and his school. In the literature only the first work on this animal, that by Hubrecht himself (1890), is generally known. According to this certain important differences exist between the hedgehog and man, but more recent observations made by Resink under Hubrecht's direction (1903) have corrected a number of inaccuracies and thereby revealed a greater resemblance to the human conditions. For instance, the tissue formerly termed the trophospongia and derived from the decidua is now assigned to the trophoblast. (See also Grosser: Lehrbuch.) ^' Jung, in agreement with Langhans, will not admit, at least in young stages, the occurrence of a destruction of the peripheral portions of the trophoblast shell, described especially by Peters. This author's results are regarded as post- mortem phenomena. "^ This superiority of the basal growth is not always pronounced; apart from the Bryce-Teacher ovum, which showed an especially strong equatorial development of the syncytium, there was in the Jung ovum an almost equal development of the trophoblast shell, in the Spee ovum villi occurred on the peripheral surface, and in the Siegenbeek ovum there was again a superiority in the equatorial villi. A purely equatorial villous girdle, such as the frequently figured Reichert ovum (1873) showed, cannot be regarded as normal, since it can hardly be reconciled with the idea of interstitial implantation. The occurrence of variations within certain limits is, however, not unthinkable, since they may be produced by factors extrinsic to the ovum, such as the distribution of the embiyotrophe, local pathologi- cal changes in the mucous membrane, etc. ^ So long as the mechanism of implantation was unexplained, speculation concerning the origin of the syncytium had free rein. The early view, supported by the most prominent investigators (Langhans and his school, Strahl) and which assigned its origin to the uterine epithelium, is irreconcilable with interstitial implantation. Also the glandular epithelium need hardly now be considered as a possible source. For a consideration of the early views consult, for example, the well-known account of Waldeyer, also Peters and Strahl. Directly opposed to the idea of its origin from the uterine epithelium are cases of pathological im- plantation, such as are seen in ovarian pregnancies, for in these the villi have a typical syncytial covering. 124 Hr:\IAN EMBRYOLOGY. ium from the troplioblast, the other from the endothelium of the maternal vessels.^^ The trophoblastic origin of the syncytium is upheld by all supporters of Hubrecht's views and especially by all recent stu- dents of the problem. The Bryce-Teacher ovum is especially illuminating in this connection: in it a connection of the syncyt- ium with the cytotrophoblast is, on the one hand, clear ; and, on the ctlier hand, an anchoring of the ovum to the maternal tissues, that is to say, a direct contact of syncytium and decidua, is wanting. Peters, Leopold, and Jung expressly mention the occurrence in their preparations of gradual transitions between the cytotropho- blast and the syncytium (for example, the passage of nuclei from the former into the latter, Jung) and the absence of similar tran- sitions between the syncytium and the endothelium. These facts overthrow the opposed view of Pfannenstiel, based upon older preparations, which view brings him into accord with a number of older authors and for support of which he relies upon one uterine ovnm which he himself investigated and one tubal ovum; at the same time other authors, such as Frassi and Bonnet, find no sup- port from older ova for an origin of the syncytium from the endothelium, but declare themselves in favor of its fetal origin. The figures given by Pfannenstiel, which seem to speak for a derivation of the syncytium from the endothelium, are, apparently, capable of another interpretation (Frassi). But although the fetal origin of the syncytium is no longer doubtful, the beginning of its formation has not yet been sufficiently studied. Hubrecht, Marchand, Bonnet, and others suppose that the syncytium is the expression of a special vital energy and is produced by the penetration of the troplioblast into the maternal tissues. Peters, however, is of the opinion that the syncytium is formed from the cytotrophoblast by a kind of degeneration process influenced by the maternal blood. The syncytium of both the Biyce-Teacher and the Peters ovum ^^^ is undoubtedly materially different from that of later stages, which forms a layer of almost even thickness over the chorionic villi. In the Bryce-Teacher ovum there is a thick spongy syncytium shell resting upon a thin layer of cytotrophoblast; in the Peters preparation there is a great quantity of cytotrophoblast and a very irregular distribution of the sATicytium. This forms often large masses, which frequently ^ Graf Spee does not express himself definitely on the question, but from the remarkable occurrence in one instance of a cuticle between the syncytium and the cell layer he is rather inclined to accept a maternal origin for the syncytium, deriving it eventually from the giant marrov? cells of the mother. This idea is no longer tenable. ™ The avithor is greatly indebted to Professor Peters for permission to study and make use of this valuable preparation. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 125 project freely into the blood, and at many places are provided with relatively few, but greatly enlarged, nuclei, so that, as Peters points out, they give the idea not of a progressive but of a regressive form of tissue and according to Bonnet 'vS terminology deserve to be termed symplasmata (Figs. 99 and 102). Only at certain places does the syncytium form a lining for the blood lacunse, so as to recall well-known figures. In the Leopold ovum, on the other hand, it has many more of the usual characters; still more pro- nounced, perhaps, is this condition in the Jung ovum, in which degenerating syncytium, termed symplasma syncytiale by the author, can be observed only locally. Following a view similar to Fig. 102. — A portion ot the trophoblast shell of the Peters ovum. The syncytium has a mesh-like ar- rangement, blood occupying the spaces. Mes., chorionic mesoderm; Tr., cytotrophoblast. X 350. that of Bryce and Teacher, and assuming that their ovum was v;vH:^::. ® l'::^:^:&-A'-:;v'--'^<'^l:- ■■A M'^'" ■■■ * Dr. ■ ^~ ^■i\\:'\\}'-l,':''.X.-'h\ jj'-'-' '-.V;*-"^? '^^'J' '''::h^\-'^i::'-:iiyy^'-'''' ,■'1 ■;.'*'■'.'.;■ ',v.v ' --.■'j A ''•'• ■'"., decidua; Dz,, decidual cell sunounded by tro- phoblast; Sy., syncytium; Z., villus. X 200, been described by Franque and Vassmer, but denied by Giese (1905), their supposed presence being based on an error of ob- servation. The derivatives of the trophoblast, syncytium and "fibrin" (see pp. 151 et seq.), are of constant occurrence in the "The so-called decidual columns (Deciduabalken, Leopold) will be discussed later with the decidual pillars. Happe (1907), like Pfannenstiel (1903) and Webster (1906), regards the islands as formed principally of trophoblast, but also maintains that they contain decidual tissue (more spindle-shaped cells, loosely con- nected together and partly with a finely granular intercellular substance). DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 135 islands, which disappear in the course of the first months, being for the most part converted into "fibrin." The "basal ectoderm" is a term applied by Langhans and his school to that portion of the trophoblast which occurs on the outer wall of the intervillous space; in somewhat older ova it there forms a stratified layer (Fig. 109) and is frequently retained until the end of pregnane}', if not as a continuous stratum, at least in masses of cells arranged in groups (Figs. 129-132). Its rela- tions in younger stages, at the commencement of the formation of a continuous intervillous space, have been studied by Frassi. In such cases there is found upon the outer surface of the intervillous space the "covering layer," a simple layer of cells, resting as an almost continuous sheet upon the decidua. "The nuclei of these I Fig. 110. — Covering layer of the ovum capsule (Al.) at its transition into a cell column (at Al.*); R.Z., giant cell; D., decidua; L., leucocytes X 300. (From Frassi, 1908.) elements are larger and take the stain more deeply and regularly (than the nuclei of the decidual cells). With strong magnification a distinct difference can be perceived between the endonuclear sub- stance of such cells and that of the decidual cells." The covering layer is everywhere one-layered in Frassi 's preparations; it is in places separated from the decidua by fibrin and is lacking only in a few places. It occurs here and there between the cell columns or forms an external covering for these (Fig. 110). Transitions into decidua are absolutely wanting, but, on the other hand, they occur into the syncytium, so that the covering layer is of fetal origin. The boundary between the fetal and the maternal elements is not always easy of determination ; assistance is rendered in this connection, according to Frassi and Jung, by the leucocytes which occur abundantly and are always to be found in the neighborhood of the ovum. It would appear that they cannot pass beyond the 136 HmiAX EMBRYOLOGY. boundary of the ovum, which penetrates like a parasite into the mucous membrane, and consequently they make possible the de- termination of that boundary. The term transition zone indicates that the boundary is not a sharp one. Bonnet (1904), in the cases of older ova (those containing an embryo 3 mm. in length), preferred to speak of a detritus zone between the chorionic villi and the decidua. In association with it are symplasma formations of the decidual cells and enlarged glands and in the lumen exudations of secretion, blood, and leucocytes. The intervillous space has also received different interpreta- tions from different investigators. According to the older im- plantation theory, which held that the ovum became attached to the mucous membrane only superficially and that the uterine epithelium was retained, transformed into syncytium, the space was necessarily regarded as a portion of the cavity of the uterus enclosed between the ovum and the surface of the uterus; the occurrence of blood within it was only accidental, or, at most, a regular phenomenon only in later stages of development, its place being taken in young stages by a secretion of the mucous mem- brane, a kind of uterine milk. As a matter of fact the space was usually found to be empty in aborted ova (Fig. 107) and even in those obtained by operation and observed in situ (Fig. 118). These observations were taken as evidence opposed to a regulated cir- culation in the intervillous space; and the condition occurring in the Peters ovum, for instance, in which the lacunte were engorged with blood, was explained as the result of the action of the poison taken by the mother. Frassi, who also foimd the space empty in his ovum, although open communications with maternal blood- vessels could be determined at various places, rightly maintained, on the contrary, the existence of a regulated circulation, and pointed out that, after the inflow of blood had ceased as a result of the cessation of the heart-beats of the mother or of the ligation of the arteries during operation, an outflow of blood through the veins was still quite possible and, furthermore, would be aided by the final contractions of the uterine musculature. Such con- tractions, indeed, occurring as they do, though to a lesser degree, throughout the whole period of pregnancy, may form an important accessory factor in promoting a circulation, which at the best must be difficult and slow, through the very irregular space (Von Herff). The views of Pfannenstiel regarding the formation of the inter^'illous space will be considered later (p. 167). With the formation of the intervillous space and the gradual disappearance of the trophoblast shell the ovum passes from the embryotrophic into the hEemotrophic phase of placentation. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 137 IV. THE FORMATION OF THE PLACENTA; RELATIONS OF THE EMBRYONIC MEMBRANES UP TO THEIR MATURITY. {Hcsmotrophic Phase of Placentation.) a. DIFFERENTIATION OF THE CHORION; CHORION L^VE, DECIDUA PARIETALIS, AND CAPSULARIS. At first the trophoblast shell completely surrounds the ovum and villi are formed over the entire surface of the chorion; the entire chorion is at first a chorion frondosum. As the ovum in- creases in size and projects more and more beyond the general level of the mucous membrane, the decidua capsularis, which covers it and is only poorly supplied with nourishment, is gradu- ally distended more and more, the circulation in the intervillous space over the convexity of the ovum becomes more and more difficult, and the villi on the surface directed toward the capsularis finally atrophy, so that the convexity of the chorion becomes smooth, becomes a chorion Iceve, while the basal portion of the chorion frondosum becomes the placenta fetalis. According to the observations of Pfannenstiel (1903) ova of the fourth week (from the cessation of menstruation) already distinctly show a bare spot at the capsularis pole, and even at the end of the second week this pole may be almost destitute of villi. Ova of the second to the fourth week project beyond the general level of the mucous membrane to very varying extents, either as far as the equator or even further. This condition is referred by Pfannenstiel to varying depths of implantation ; the shallower the implantation the more the ovum later projects beyond the level of the mucosa. The depth of the implantation, on its part, depends upon the intensity of the original growth of the trophoblast and its action on the maternal tissues. Concerning the relation which probably obtains between the depth of the implantation and certain abnormal forms of jDlacenta (placenta marginata, reflexa, accreta) see Grosser 's "Lehrbuch." The further fate of the chorion lajve will be considered in connection with that of the decidua capsularis. The decidua parietalis (vera), in accordance with its pre- menstrual relations, is already more or less distinctly differenti- ated into a pars compacta and a pars spongiosa at the time of implantation.*^ The former is essentially the region of stroma changes while the latter shows characteristic gland forms. But both layers during the first weeks of pregnancy still present the " Peters did not observe the compact layer and believed that it develops later, and Siegenbeek notes the lack of a distinct boundary between the two ; never- theless it must be remembered that in this respect variations occur also in the premenstrual mucous membrane. In the Jung ovum the layers are separated. 138 HUMAN e:mbryologt. premenstrual type, iB accordance with the inhibitory effect exer- cised by the implanted ovum upon the changes of the mucous membrane (p. 119) ; in the stages now under consideration (Fig. Ill) they are differentiated. The decidua compacta (Fig. 112), in addition to the straighter tenninal portions of the glands and greatly enlarged blood-vessels, also contains decidual cells, which are formed from stroma cells -•■J0 m^MIMma Comp.<{ Spong. Dr.l. AIusc, Dr. 2. Fig. 111. — Decidua vera (parietalis) of the second month. Comp., decidua compacta; Spong,, decidua spongiosa; Muac, mu9cularis uteri. For Dr. 1. and Dr. 2. see the detail, Figs. 112 and 113. X 12. by the continuation of the changes that are characteristic of the end of the premenstrual stage. They are large, clear, vesicular cells, as much as 50 /j. in diameter, and are round or, from mutual pressure, polygonal, resembling epithelial or epithelioid cells. The changes by which they are produced do not occur simultaneously in all the stroma cells ; and even at the height of the formation of the decidua one may find here and there stroma cells but slightly altered and showing division and growth phenomena, so that Marchand (1904) recognizes two types of decidual cells, large and small. The mature (large) decidual cells show, at the most, only DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 139 direct nuclear division (they contain frequently two nuclei and indications of a cell boundary between the nuclei) and, as fully differentiated cells, are capable of no further progressive or regressive development. Many of them degenerate during the second half of pregnancy and are disposed of by leucocytes ; but the majority are thrown off either during or after birth. After the fourth month they all become smaller again and more spindle- Fig. 112. — Detail of Fig. Ill; the gland duct there indicated by Dr. 1. in the decidua parietalis compacta of the second month. The gland contains secretion and around it are typical decidual cells and a few leucocytes. X 350. shaped and are arranged parallel to the surface (Pfannenstiel). According to Wederhake (1906), Unna plasma-cells also occur in the decidua, and transitions between these and typical decidual cells. The significance of the formation of the decidual cells lies, according to Marchand, in the storing up of glycogen;*'' the ma- *^ Driessen (1907), who recently has again taken up the older observations of Langhans on the oeeiirrence of glycogen in the decidua, finds that substance chiefly in the glandular epithelium of the spongiosa; it is not always recognizable in the decidual cells. In the second half of pregnancy it gradually disappears. 140 nU.MAX EitBRYOLOGT. jority of other authors see in their formation a provision against the too intensive jjenetration of the ovum into the mucous mem- brane, without furnishing sufficient evidence for such a view. Ac- cording to Marchand, spindle-shaped epithelial cells grow out as wandering epithelial cells from the degenerating glands of the compacta into the stroma and may there fuse to form multi- nucleated masses. Tn the decidua spongiosa are to be found at first the glands of pipgnancy, also characterized by the further development of the premenstrual changes (Figs. 87 and 113). They are greatly Fig, 113.— Detail of Fig. Ill; the gland of the decidua parietalis spongiosa there indicated by Br 2 Papilla in the wall of the gland-bearing epithelium, secretion in the lumen. X 150. enlarged and tortuous, irregular in section, and filled with secre- tion. The enlarged epithelium projects into the lumen in the form of papilte borne upon small elevations of the stroma; it is composed of high cylindrical cells, with clear marginal zones tilled with secretion. Between the glands are very small connective- tissue septa with scattered decidual cells; only near the larger vessels are the septa broader. After the second month the epithelial papillte disappear and the cavities of the glands become low and broad as a result of the stretching of the entire decidua, due to the increase in size of the uterus. The epithelial cells con- tinue to grow broader and lower (Fig. 114) until, finally, they resemble an endothelium and are lacking in places. The cavities of the glands then appear as small, elongated clefts, with thin intervening walls, resembling in mass an empty venous plexus DEVELOPMENT OP EGG filEMBRANES AND PLACENTA. 141 ■ ^■f»^~:^%-!^:i k-''''^y'^^^^V'^'%^^- ~V^^P''^«^V -A. -Ch.-B. -Ch.-Z. r B'lG. U4.^The egg membranes and uterine wall opposite the placenta in the fourth month; from the same case as Fig. 130. (Embryo 13^ cm. in vertex-breech measurement.) 4.. amnion; CA,-B., chorionic connective tissue; Ch.-Z., degenerated chorionic villi; Comp., decidua parietalis compacta and capsularis; Spong., decidua parietalis spongiosa; Muse, musoularis uteri. For detail see Fig. 115. X 80. Ch.-B. ^^ Dec. Fig. 115. — Detail of Fig. 114. The degenerated villi of the choiion leeve in the fourth month (the villi have shrunken somewhat during the preparation of the object). Chz., chorionic villi; Ch.-Ep., cho- rionic epithelium; Ch.-B., chorionic connective tissue; Dec, decidua capsularis and parietalis compacta, with leucocytes. X 300. (Fig. 117). The separation of the decidua in an abortion or at birth can therefore take place easily in the spongiosa. Only the deepest portions of the glands (the boundary layer of His), which lie between the irregularities of the surface of the musoularis, 142 nmiAN EilBRYOLOGY. retain tlieir cubical epithelium and form the starting point for the post-partum regeneration of the mucous membrane. The surface epithelium becomes flattened and loses its cilia (according to Marchand) ; furthermore, fat globules are formed B.C. Ch.-Z. Sp. Fig. 116. — Egg membranes and decidua capsularis at the fourth month of pregnancy, from the region over the internal OS uteri (from the same case as Fig. 118). A., amnion; C/i., chorion Iseve; Ch.-Z., degenerated chorionic villus; Sp., cleft (remains of the intervillous space?). X 70. in the cells and symplasmic formations occur, and toward the end of the third month the epithelium has practically disappeared. At the same time the cavity of the uterus, the perional space (the space surrounding the ovum (mov) : Webster), which has at this time only a potential existence, disappears as the capsularis comes r — ' Amn. I \Zw. ;^ ^>ii;K\ i'-^A,- Fig. 117. — Section through the mature egg membranes with the adhering decidua, expelled spon- taneously. Amn., amnion; Ch.-B., chorionic connective tissue; Zw., intermediate zone (chorionic epithe- lium, remains of the villi, decidua capsularis, and decidua parietalis compacta); D. sp., decidua parietalis spongiosa; Dr., glandular remains. X 90- into contact with the parietalis. The capsularis (Figs. 114 to 117), by stretching and by degeneration as well, has become greatly reduced and its remains now fuse with the decidua parietalis. The view that it remains recognizable as a streak of cells up to the close of pregnancy is probably based on an error, the chorionic DEVELOPMENT OF EGG MEIMBRANES AND PLACENTA. 143 epithelium being mistaken for it. The degeneration of the cap- sularis can be demonstrated beyond question in tlie region of the internal os uteri, where its fusion with the decidua parietahs is impossible. Even at the fourth month the capsularis (Fig. 116) consists in that region of only a very thin layer of flattened ele- ments with some elongated clefts, probably remnants of the inter- villous space. The chorion IcBve also shows extensive degenerative changes. The epithelium of the villi disappears, their stroma undergoes hyaline degeneration (Figs. Ill to 116), and between the hyaline masses so formed one finds the detritus of cells and leucocytes. At the summit of the ovum even these hyaline re- mains of the villi vanish (Fig. 117), but they persist in the neigh- borhood of the placenta. The epithelium of the chorionic mem- brane itself is, however, usually recognizable in the mature egg membranes ; external to it is a zone of detritus with the remains of the villi, the capsularis, and the decidua parietalis compaeta, in which also hyaline degeneration, as well as fusion and destruction of the cells, has occurred (Fig. 117). Still more externally are the remains of the spongiosa, which at the close of pregnancy is re- duced to a thickness of 1-2 mm., but which still contains remains of the gland cavities. The fatty degeneration of the decidua parietalis, which was formerly regarded as the rule, occurs at most only in exceptional instances. b. THE PLACENTA. In the formation of the placenta the chorion frondosum and the decidua basalis participate, the former constituting the placenta fetalis and the latter the placenta materna.** The placenta fetalis consists of the chorion plate and the chorionic villi; both contain a mesodermal stroma and an ecto- dermal (trophoblastic) epithehum. The stroma of the villi is at an early period distinctly fibrillar and provided with fusiform cells in the principal stems and in the chorion plate ; in the lateral branches it is at first formed of stellate cells with wide intercellular spaces, but even in these portions it soon assumes a fibrillar character. In the meshwork of the con- nective tissue there frequently occur in young ova lymphocyte-like structures and some especially large cells, with highly vacuolated plasma and large nuclei (Fig. 119), to which Hofbauer has called attention and which he brings into relation with the plasma cells. Their significance is, however, still uncertain. The capillaries of ■" This latter term has varied somewhat in its significance. Kolliker terms the entire basalis the placenta materna and divides it into a pars non caduca sen -jixa, which corresponds to the spongy portion, and a pars caduca, which is expelled at birth and is usually known as the basal plate. However, the latter alone is frequently termed the placenta materna. 144 HmrAN E]\IBRYOLOGY. Ch.-P- R.F mm- ^f "\ x^ ^k ^^ ^^ Dr. M~ 6'>iV-'^.Sr"*W^^*»»'" Fig. 118. — Anlageof the placenta from the second month. From a uterus obtained per operalionem. The embryo had a vertex-breech length of 28 mm. The same case as is shown in Figs, 111, 119, and 136, C/i.-P., chorion plate; Dr., glands; 6. jE., basal ectoderm.; Hz., anchoring villi; Af., rauscularis uteri; m.A., maternal artery in a placental septum (decidual pillar); N. F., Nitabuch's fibrin stria; R. F., E,ohr's fibrin stria; Z.-7., cell island. X 15. the fetal vascular system lie, for tlie most part, near the surface of the villi. According to Bonnet (1903) lymph-vessels also occur in the stroma of the villi and can be followed to larger vessels in the chorionic membrane. Nerves are not recognizable in the DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 145 H.-Z. Fig. 119 — Detail from Fig. 118. Chorionic villus from the second month; the syncytium provided with prickle processes, f . Cap., fetal capillary; H,-Z., Hofbauer cell. X 400. Fig. 120. Villi from a mature human placenta, injected with carmine and teased out in the fresh condition. X 30. placenta (Bucura). Fossati has described a network of tibres, characterized by special histological peculiarities, as occurring around the chorionic vessels. In the stroma of the chorion plate Vol. I.— 10 146 HUMAN EMBRYOLOGY. Langhans has described a more superficial subchorial vascular layer and a deeper fibrillar one, which shows no sharply defined boundary from the coelom. These layers become distinguishable only at about the third month. Glycogen is found in young ova chiefly in the connective tissue of the chorion plate and of the larger villi (Happe, Driessen). (For further particulars concern- ing the stroma of the villi see Happe, 1907 ; and regarding elastic fibres consult Fuss, 1906.) Fig. 121. — Injected villi from the mature human placenta; arteries dark, veins light. The apparently free endings of the vessels due to incomplete injection. Fresh preparation. X 350. The form of the villi, which is determined largely by the stroma, changes during pregnancy in that, on the one hand, the branchings of the villi become continually more numerous and the villous trees larger, and, on the other hand, the branches them- selves become more slender and longer (Fig. 120) ; yet even in the mature placenta variations in this respect occur. In each villus one or two arteries occur and one or two somewhat stronger veins, the two sets of vessels being connected by a capillary net- DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 147 work lying immediately beneath the epithelium (Fig. 121). (Con- cerning the form of the villi see Minot, 1889, and Happe, 1907.) The Choi-ionic epithelium, as has already been stated, is two- layered after the formation of the villi (Fig. 129). The deeper layer, which is composed of distinctly separated cells, is usually named from its discoverer the Langhans layer, but is also termed the cell layer. The superficial layer is termed simply syncytium or also syncytial layer or covenng layery^ The two layers together form the diplotrophoblast of Hubrecht. As a rule, the cells of the Langhans layer are arranged in contact with each other in an epithelial manner, but frequently the syncytium extends between the cells (Fig. 119) to the basement membrane of the epithelium (Bonnet, 1903). As a result of this it appears in places as if the cells were arranged in separate cell territories enclosed in a ground substance and with a kind of capsule or bounding layer. This condition is regarded as the rule by Happe (1907) among recent authors. The syncytium generally forms a layer of almost the same thickness as the cell layer and has but a single layel- of nuclei. Vacuoles, that are so striking in the syncytium in early stages, are also to be seen at later periods and may be the expression of degeneration or of the absorption of material. On its outer surface it is provided with a delicate membrane, which proves to be composed of prickle processes, stiff hairs or rodlets, stereocilia (G-raf Spee, Von Lenhossek, Bonnet). This membrane is perhaps existent only under certain functional conditions and cannot always be perceived ; the rudimentary basal bodies described by Lenhossek as occurring in the cilia have not been found again by Bonnet. Indications of absorption in the form of fat globules, basophile granules, and mitochondria occur in the syncytium, and it takes up haemoglobin in a soluble form. On its outer surface it frequently bears irregular, multinuclear elevations or buds (proliferation nodes; Fig. 122), which occa- sionally become separated and may be carried in the circulation far from the intervillous space (the deportation of syncytial ele- ments of Veit). They are probably indications of amoeboid activit}^ which, in all probability, occurs in the syncytium.*" De- generation (the formation of symplasma syncytiale with spiny nuclei and the dissolving of the plasma into clouds or drops) may be observed, according to Bonnet, in the syncytium in younger stages; in older stages it takes on other forms (see p. 151). " To the two ■ layers of chorionic epithelium have been ascribed by different investio-ators very various and somewhat remarkable significances. For a review of the different origins suggested for the epithelium, which have been copied in a number of papers, see Waldeyer, 1890. "Also entire villi may be torn away by the blood stream and enter the maternal vascular system. 148 HUMAN EMBRYOLOCY. Glycogen occurs in the cytotrophoblast (in the cell columns and cell islands, less regularly in the Langhans cells) ; it is lacking in the syncytium. It disappears completely with increasing maturity of the placenta (Driessen, Happe, 1907). Mitoses occur only in the Langhans layer; in the syncytium only direct division occurs, and it is rare (Van Cauwenberghe). The direct passage of a cell from one layer into the other has not yet been observed in older stages and occurs only occasionally in younger ones (see p. lU). Nevertheless, the distribution of the nuclear divisions must be taken as evidence that even in later stages the cell layer is the source of the syncytium and adds to Fig. 122. — From a mature placenta (after birth). Formalin. /. G., fetal chorionic vessel; m. B., maternal blood-corpuscles in the intervillous space; Sy., syncytium; Sy.-Sp., syncytial process (prolif- eration node). X 250. it. The older opinion of Kastsehenko, which has recently been revived by Happe (1907), to the effect that the cell layer arises from the sjneytium, seems to be overthrown by this. Between the syncj'tium and the cell layer there is, according to Graf Spee and Van Cauwenberghe, frequently but not regularly a cuticula or deep syncytial membrane, which, however, is believed by most other authors to be an artefact; beneath the cell layer is a basement or hyaline membrane. (For details concerning the e])ithelium of the villi see INIarchand, Friolet, Van Cauwenberghe, Happe.) In addition to the occurrence of proliferation nodes there is also another phenomenon that speaks in favor of amoeboid activi- ties in the syncytium; this is the relation of the basal (serotinal) or syncytial giant cells (Figs. 12.3 and 124). In the decidua basalis DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 149 one finds even in young stages multinucleated masses of proto- plasm wlaich cannot be distinguished liistologicallv from syncytium and are of great importance in connection with the significance of the syncytium. In the Frassi ovum they are throughout (except at one doubtful spot) in connection with the syncytium of the villi but m the preparations of Beneke, Pfannenstiel," and Friolet for example, free masses of syncytium occur in the decidua basalis- mdeed, even m the superficial layers of the muscularis and fre- quently m the neighborhood of vascular endothelium. They are most frequent at about the middle of pregnancy, when they may reach a very considerable size (Fig. 123) and may penetrate far into the muscularis (Fig. 124) ; nevertheless their abundance is Muse. ua.*.> ^'"•*- v^^-t y. 's.^ * ■aa ..ass Fig. 123. — Giant cells beneath the placenta in the muscularis uteri in the fourth month. En., endotheUum of a vein of the muscularis; Muse, smooth muscle bundle. X 300. From the same case as figs. 130 and 131. subject to rather great individual variations. Toward the end of pregnancy they diminish in number. Pfannenstiel finds support, in their occasional topographic relations to the vascular endothelium, for the derivation of the entire syncytium from the endothelium; but syncytial growths arising from endothelia are denied by Friolet and Frassi. Friolet leaves the possibility of their origin from the connective tissue an open question; Pels Leusden, Web- ster, Frassi, and others regard them as derivatives of the (fetal) syncytium, that penetrate individually into the maternal tissues,*'' " That they may also penetrate toward tlie centre of the ovum is indicated at present only by the very definite statement of Beneke (p. 115). Sie^-enbeek also records the remarkable occurrence of a syncytium mass between the chorionic epithelium and connective tissue, but considers it to have occurred by active immigration into the ovum through a tear in its wall, which he assumes to have occurred during life. 150 HUMAN EilBKYOLOGY. and this view is certainly the most probable. The fact, also, that they degenerate post jjartum, without taking any part in the regeneration of the mucous membrane, is in favor of this view (Wormser). r-' mm N. Fs. ■■4 i^'^ Muse, ( ; ■; ^.^:■"?^ ■;;..- : K J ^ ,.-■'■ ■ K"- '," . ,>•» ■.V^ :r^''Z 'V»iV ^^ii^' ?,- ^■-"^^: '•- m}'-- j^ - ". ■ / ,:>&- --^ . ( '. ' .. '■. ■ »■- ,".■■■' ^^ . 1 ■■^' ■'■'• ^^; Ge/. s. ffz. '^.m;'m& Fig, 124. — Placenta in situ from the second half of pregnancy, with numerous subplacental syncyt- ial giant cells, D.b., decidua basalis; Dr., glands; Gef., maternal vessels; Hz., anchoring villi; Muse, mus- cularis uteri; A^, Fs., Nitabuch's fibrin stria; 8. Rz., syncytial giant cells; V., larger uterine vein, X 27. Even in the fourth month the cell layer is present only in patches, the syncytium resting, for the most part, directly on the stroma of the villi. Toward the end of pregnancy individual Langhans cells are still to be found beneath the syncytium, ac- cording to Van Cauweuberghe, yet this is certainly a by no means frequent occurrence. The layer is partly spread out and stretched over a constantly increasing area by the growth of the villi and is partly used up in the foi-mation of syncytium (Fig. 122). The DEVELOPMENT OF EGG AlEMBRANES AND PLACENTA. 151 syncytium also frequently shows signs of degeneration ; it becomes greatly attenuated and may even be wanting at many places on In such cases so-called placental fibrin the surface of the villi, occurs on the villi. This "fibrin" has also frequently been the object of investi- gation and of controversy. In quite young ova (Peters, Leopold) it does not occur; in older ones, in which the intervillous space has formed, it first appears usually as a stria situated some dis- tance from the space in the basalis or even in the capsularis. Fig. 125. — Chorion plate with subchorial closing ring and "canalized fibrin," from a mature placenta. Chp., chorion plate; k. F., canalized fibrin; s. Sr., subchorial closing ring; Z., villus; Zs., con- nective-tissue stroma of degenerating villi. X 80. The time of its appearance does not seem to be constant (see p. 127) ; but having once appeared it persists until the close of preg- nancy. It was first described in a dissertation written by one of Langhans' pupils and was named, from the authoress of the paper, Nitahuch's fibrin stria. A second stria also occurs, though not constantly, immediately in the wall of the intervillous space and has been termed Rohr's stria*^ (Fig. 118). In addition, there is a third stria which is constant in its occurrence close beneath the chorion plate; it appears, however, later, only in the second half of pregnancy, and is known as Langhans' stria. In the same region is to be found especially the "canalized fibrin" of Lang- hans (Fig. 125), and, finally, quantities of fibrin occur everywhere " Rohr himself names this the upper stria, terming Nitahuch's the lower stria. 1.52 HUaiAN EMBRYOLOGY. in the mature placenta and between the villi; these fibrin masses are, for the most part, microscopic in size, but frequently increase to extensive structures. The small ones are termed fibrin nodes, and the larger are known as white infarcts. The latter may occa- sionally form almost half the mass of the placenta. Langhans and his pupils regard the Nitabuch stria as marking the boundary between the maternal and fetal tissues. It is certain ■■^^'\- ,--.■•—#■' . ^- J Fig. 126— a small fibrin node between degenerated villi from a mature placenta, with degenerat- vmu?" X^go" "P'"''''^^*- "'• ■^^•' proliferating epithelium of villi; Zs., connective-tissue stroma of that it occurs in the transition zone and that the maternal tissues that may be between it and the ovum quickly degenerate. Jung derives the stria from the boundary zone of the maternal tissue (see p. 115). It is at all events basal to typical decidual tissues, and is between the ovum and the basal ectoderm and degenerating tissue, whose origin cannot always be certainly ascertained. It IS traversed by maternal (uteroplacental) vessels. That it is practically the boundary between the mother and the ovum is also DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 153 shown by its relations to the leucocytes ; these, as a rule, do not pass beyond the stria. The fibrin of the Nitabuch stria is ap- parently the first to appear in the placenta, yet the methods for detecting the material in other portions of the placenta are insuf- ficient. In general this so-called fibrin is by no means typical blood-fibrin, it shows only occasionally the histological reactions of fibrin. Hitschmann and Lindenthal believe that the fibrin reaction is shown only at the commencement of its formation, and that it later alters with age. It would be better, therefore, to designate the substance by some indifferent term, such as fibrinoid or fibrinoid substance. At all events, the stimulus for the forma- tion of fibrinoid is afforded by degenerative changes, those occur- ring in the decidua and chiefly in the region of the basalis being responsible for the formation of the Nitabuch stria and the smaller basal fibrin masses ; yet the amount of participation of the indi- vidual tissues of that region cannot as yet be strictly defined. At other places the trophoblast must be regarded as the seat of the fibrin formation; for very large quantities of fibrin are found at places where the decidua is wanting, as, for instance, beneath the chorion and between the villi. The maternal blood may also participate, forming true fibrin; examples of this are shown in places where the fibrin contains red and white blood-corpuscles. It is also possible that the blood fibrin may become so altered in course of time that it can no longer be detected by the usual histo- chemical methods. It may be such fibrin that occurs in the decidua basalis and in the white infarcts, deposited as the result of the disturbances in the circulation produced by the degeneration of the epithelium of the villi and the fusion of these structures, or as a result of the imperfect outflow of blood from the intervillous space produced by villi being carried into the maternal veins, as Giese suggests. The subchorial "canalized fibrin" also presents a peculiar layered structure (Fig. 125), which may be explained as the result of the deposition of successive layers of blood fibrin, especially since the blood stream is undoubtedly greatly retarded in the roof of the intervillous space.*^ The derivation of fibrinoid from the trophoblast is based upon the study of the formation of fibrin nodes on the villi; and it is also supported by what can be seen in the formation of the Lang- hans stria. In the villi one finds the first stages of fibrinoid forma- tion partly between the syncytium and the connective-tissue stroma, partly where the epithelium of the villi has almost dis- appeared, as it does in every mature placenta. The occurrence of the fibrinoid between the syncytium and the stroma points to the cell layer as the seat of its formation, and this indication becomes ^ Langhans spoke of the canalized fibrin as a tissue ; but this conception of it is incorrect, since it cannot be considered as living material. 154 HUMAN EMBRYOLOGY. stronger in regions where the villi are closely packed together at the time when the fibrin formation begins. One can then observe how the formed masses of fibrin produce a cohesion of the villi and how the Langhans cells occur between the fibrinoid and the stroma of the villi ; also in the mature placenta the Langhans cells, prac- r^;^v. /*'■ w.Ze. Fig. 127. — The formation of fibrin between fused villi of the mature placenta, with proliferation and degeneration of the epithelium of the villi (trophoblast). K., calcareous deposit; w. Ze., prolifer- ating epithelium of villi; Zs., connective-tissue stroma of villi. tically wanting elsewhere, may be seen at the surface of the villi (Fig. 127), usually in a continuous and sometimes in a double row; and, furthermore, these cells occur free in the formed fibrinoid, where they become vesicular in appearance and gradually lose their staining properties, persisting for some time as "ghosts" of their former selves and eventually becoming completely converted into fibrinoid. The process of conversion cannot be termed a direct necrosis of the cells ; it has great similarity to what is seen in the DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 155 formation of the matrix of many kinds of connective tissue, as, for instance, when cells and even whole cell territories become transformed into matrix in some kinds of cartilage. It is a process apparently intermediate between secretion and the direct transformation of the peripheral portions of the cells, and by the continuation of it the entire cell becomes transformed. The fibrinoid so formed lacks the property that inhibits coagulation and that is possessed by the living epithelium of the villi, and so the mass of the fibrin node is increased by the formation of fibrin from the blood. Whether the syncytium also takes a direct part in the formation of fibrinoid, or degenerates, or, perhaps, first divides into separate cells which are then transformed into fibrin- oid, has not yet been determined. The degenerating stroma of the villi later undergoes a very similar (hyaline) transformation and disappears in fibrinoid. The chorion plate is the seat of formation of the Langhans fibrin stria. In it even in the fourth month a partial transforma- tion of the epithelium into fibrinoid can be detected, and in the sixth month the epithelium in the middle of the placenta is replaced by the fibrin stria. The lack of epithelium, on the one hand, and the retarded blood flow, on the other, then produce blood coagula- tion in layers, the result being canalized fibrin. An important source of fibrinoid is to be found in the cell islands, that are so frequent in the first month and later disappear. In these also the beginning transformation into fibrinoid is evi- dent very early (Fig. 108). Possibly these trophoblast masses give a stimulus for the formation of the larger fibrin masses. The description of fibrin formation given above resembles closely that given by Schickele. The derivation of the fibrin from the trophoblast, in part at least, has also been maintained, among others, by Kermauner, Hitschmann and Lindenthal, and Giese. These authors advocate even more strongly than has been done above the occurrence of disturbances in the circulation and the formation of fibrin from the blood. Steffeck (1890) has derived all the fibrin from proliferated decidua, a view that is to-day untenable; he has apparently throughout mistaken the swollen degenerating- trophoblast cells for decidua cells. The transformation of the trophoblast may, however, be carried even further and produce a hquefaction of the formed fibrin; thus arise the placental cysts which are of very frequent occurrence in mature placentae. Superficially situated cysts, oc- curring in the Langhans stria, may reach the size of a hazel-nut or even larger; more frequently are small microscopic cysts in the middle of the tissues (Fig. 128). They are always enclosed within a mantle of fibrinoid containing degenerating trophoblast cells and are frequently lined by a flattened, endothelium-like layer of these cells (Giese). The white infarcts form, as the result of a combination of fibrinoid and blood-fibrin formations, larger solid masses, which 156 HUMAN EMBRYOLOGY. may enclose large villi, causing the degeneration and death of their connective tissue, which eventually becomes unrecognizable as such. The destruction of the connective tissue and chorionic vessels m the infarct is, however, a secondary process, and not primary, as Ackermann supposed, since the villi are nourished from the inter- villous space and not by the chorionic vessels (see p. 164). Red infarcts, which are of much rarer occurrence and owe their name to their color, are due to mass coagulation of the blood in the intervillous space. d.Z. ~ — #'/^'K' •■/■■■".-•-' ^>w ■>w/..'.:'.-jw.~^ ; - '■■ ■''"- ■-ii:.--^ Zs. Fig. 12S. — Small cyst in a fibrin node, from a mature placenta. Around the cyst are degenerating masses of troplioblast. d. Z., degenerating villi; Zs., connective-tissue stroma of villi. X 90. It is generally supposed that the cytotrophoblast, with the exception of some scattered Langhans cells lying beneath the syncytium of the villi, has entirely disappeared in the mature placenta; in speaking of the formation of fibrinoid, however, at- tention has been called to the occurrence of isolated trophoblast cells, and they are to be found rather constantly in two other situations : in the floor of the intervillous space, their occurrence in this region will be discussed in connection with the description of the decidua basalis ; and in the region of the subchorial closing ring (Waldeyer). In this region they form a cell plate of varying breadth, circular, in correspondence to the margin of the placenta, DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 157 and often wanting; when present, however, it lies between the chorionic connective tissue and the Langlians fibrin stria, extends inwards from the margin for about 1-2 cm. and laterally passes over insensibly into the epithelial layer (p. 143) which persists on the outer surface of the chorion lasve (Fig. 125). The closing ring is the remains of the epithelium of the chorion plate, which further e.Sr. f.G. I '•'"Tk"-^, Zs. ^h.^ N 'Z 1 •A *'i^«» rib. D p m.O. b.Tr Fig 129 —Margin of a mature expelled placenta. A marginal sinus lias not formed at this place. Bv basarplate- Chh Che. chorionic connective tissue and epithelium belongmg to the chorion laeve; ?L' chorion p^te;D p., decidua parictalis (and capsularis); Fib., fibrin; /. G.mG., etal and maternal vessels 7 infarct K calcareous deposit; iv. fi., intervillous space; s. Sr subchorial closing ring fre- ^rently interrupted; 6. Tr., remains of basal trophoblast; Z. 1 ., villus which traverses the subchorial closing ring; Za., connective-tissue stroma of villi. X 17. in has completely disappeared, that is to say, has become com- pletely transformed into fibrinoid, and at the margin of the placenta has, it is true, lost its syncytium and has formed fibrmoid, but yet has largely persisted or has even become many-layered as the result of proliferation. In these cells, even in the mature placenta, the transformation into fibrinoid can be observed, and the cell plate is usually not continuous, but shows local defects {Fig. 129). 358 HUMAN EilBRTOLOGY. Winkler originally described a continuous plate of cells as existing beneath the connective tissue of the chorion plate, naming it the dosing plate and deriving it from the decidua. Kolliker pointed out that a complete plate does not exist and speaks of a decidua siibchorialis occurring at the margin of the placenta, and by this name the tissue is now generally known. Pfannenstiel endeavors to explain it as produced by an undermining of the marginal decidua by the marginal villi. The idea of a decidual origin for the layer is based upon the pale and swollen appearance of the cells, but this, as has already been several times noted, is 6. E. (F.) Mu&c. Fig. 130. — Margin of a placenta of the fourth month (vertex-breech length of the embryo 133^ cm.). A decidua .subchoriaiis is wanting. The gland cavities of the spongiosa (Dr.) have been separated in the preparation. Am., amnion; Ch., chorion; b. E., basal ectoderm; b. E. {F .), basal ectoderm in process of transformation into fibrinoid; Fib., fibrinoid; Muse, muscularis; iv. R., intervillous space; Zk., cell nodes. X 18. not a proof of its origin. Langhans in 1877 expressed doubts as to the decidual nature of the layer and later (1891) decided in favor of an ectodermal origin; Hitschmann and Lindentlial agree with him in this. In the mature placenta the closing ring is occa- sionally broken through by villi (Fig. 129), which are usually atrophic and belong to the portion of the chorion which is transitional between the chorion f rondo sum of the placenta and the chorion Iseve, and this fact is sufficient in itself to overthrow the idea that one has to deal with an undermined marginal decidua. There remains, therefore, for consideration only the assumption of an active growth on the part of the decidua cells of the placental margin towards the middle of the placenta, a supposition which has little probability on account of what has already been said as DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 159 to the degeneration of the decidua in the later stages of pregnancy. But in addition this possibility is also excluded by the study of the margin of the placenta in different months of pregnancy. This shows (c/. Fig. 130) that the transition of the placenta into the chorion Iseve during the progress of pregnancy is a gradual one and the two structures are first sharply marked out towards the end of it. Normally a proliferation of the marginal decidua or an undermining of it is never to be seen ; but, on the other hand, such preparations directly show the gradual degeneration of the epithelium of the chorion plate in the middle of the placenta and its transformation into the subchorial closing ring at its margin. Perhaps a portion of this epithelium may persist throughout the entire extent of the placenta, just as it does at its margin; at least this would explain cases such as that described by Steffeck, in which an actually complete subchorial epithelial plate is said to have been present, and which was identified by that author as decidua. The conditions in a placenta marginata are quite dif- ferent, for this may well owe its origin to an actual undermining of the marginal decidua (c/. Pfannenstiel, 1903, and Grosser, "Lehrbuch").^'^ The decidua hasalis {placenta materna) is divisible in later stages of pregnancy into a compact layer, the basal plate of the placenta, and a spongy layer. The former, which closes the inter- villous space on the basal side, is formed from the portion of the premenstrual compacta that is not destroyed by the ovum during implantation; the latter is formed from the premenstrual spongiosa. This shows (Figs. 118 and 124), in general, the same changes as the decidua parietalis spongiosa; only the layer is thinner, and the gland spaces appear to be less frequent, perhaps because some of the glands, which in the early stages of the de- velopment were filled with blood, have degenerated; between the glands are the syncytial giant cells already mentioned. The basal plate (Figs. 131 and 132) consists of decidua, which toward the end of pregnancy again contains more fusiform cells, and also of the fibrin stria; already described and of the remains of trophoblast derived from the basal ectoderm and the cell columns of the anchoring villi. These trophoblast remains, again, are partly syncytial in character and partly appear as large, swollen, mono- nuclear cells (mononuclear giant cells) similar to those found in the early stages of fibrinoid formation in other places ; these cells ^ Sfameni takes a different view of the origin of the placenta marginata. He refers it to a lack or insufficiency of pressure within the ovum and to insufficient stretching of the wall of the uterus. Compare also Kromer (1907) and Liepmann (1906), as well as the polemic of these authors in Centralbl. f. Gynak., vol. xxxii, 1908. 160 HraiAN EMBRYOLOGY. lie, sometimes singly and sometimes in groups, in degenerating tissue and in later stages may extend peripherally beyond the Nitabuch stria. The trophoblast masses are no longer, however, in direct connection with the anchoring villi ; the tips of the latter p'°' ^^n~^°°'i'°"K*''"!! ?•"'' ''^^^' P'''*^ °f "^^ placenta in the fourth month. From the same tt^. u^' ^ «C • (Vf I^^'^-breech length of embryo, 13^ ^m.) 6. E.. basal ectoderm ; Dr.. glands ; NF the Nitabuch fibrm stna ; Sy., syncytium ^80. . . s , . - ., usually dip into masses of fibrinoid, which extend to the stroma of the villi. Frequently this is more or less degenerated, as it is in the infarcts. In the basal plate there also occur the so-called choriodecidunl vessels (Euge). They are relatively large stems, visible to the naked eye in the expelled placenta, provided that they are well DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 161 filled with blood, or, better, injected from the vessels of the villi ; they pass out from the villi, and either terminate in the basal plate or enter the stem of a villus ascending from the basal plate. This fact indicates that the villi concerned have come into contact with the basal plate and have lost their epithelium and stroma by de- generation, while their vessels have persisted and have become enlarged by the blood-pressure, partly on account of degeneration in their walls. Usually remains of trophoblast and of the stroma N.I- Fig. 132. — Basal plate of mature expelled placenta. A.u.-p., arteria utcroplacentalis ; Db., decidua baaalis ; Ep., epithelial remains in the floor of the intervillous space; Fib., fibrin; Hz., anchoring villus; N.Fa., the Nitabuch fibrin stria; Rz., mononuclear giant cells (remains of trophoblast); Zs., connective-tissue stroma of villi enclosed by fibrin. X 70. of villi (Fig. 133) are to be found in the neighborhood of the ves- sels, provided their entire surroundings have not been transformed into a fibrinoid mass. Ruge believed that his discovery of these vessels indicated a vascularization of the decidua, that is, of maternal tissues, by fetal vessels ; the explanation given above has already been advo- cated by W. Wolska, working under Langhans' directions, and has been confirmed by other authors. The statement, which has ap- peared in some articles, to the effect that Ruge has described an anastomosis of fetal and maternal vessels, is erroneous. From the basal plate there extend towards the intervillous space the decidual pillars, which represent the septa placentce of later stages. The decidual pillars (Fig. 118) are to be regarded Vol. I.— 11 162 HTJ]\L4N EMBRYOLOGY. as portions of the decidua basalis compacta which have been spared during the penetration of the trophoblast. In structure tliey re- semble the basal plate. As regards their number and arrange- ment in young stages adequate investigations are as yet lacking. Occasionally, but certainly not frequently, the pillars seem to project for a considerable distance into the intervillous space, even to near the chorion plate, and, in sections, appear to have no connection with the basal plate, so that they seem to be "decidua islands" or decidual trabeciilcB (Leopold). (That free decidua islands do not, in all probability, occur has already been noted.) The septa placenta in later stages and at the close of pregnancy divide the placenta into separate lobes or cotyledons, which do not, Fig. 133. — Choriodecidual vessels injected from an umbilical artery, in a mature expelled placenta. Bp., basal plate; Fib., fibrin stria; G., cfioriodecidual vessels; iv, R., intervillous space; Tr., basal degen- erating trophoblast; Za., degenerating connective-tissue stroma of villus. X 40. however, represent closed areas, since the septa in the middle of the placenta are ver}- low and even at the margin reach the chorion plate only in small part. The original formation of these septa from decidua cells is almost always very difficult of determination in the mature placenta (Fig. 134). The decidua cells have, for the most part, degenerated and disappeared, and in their place there remains only an em])ty mesh-work, into which the trophoblastic anchoring villi penetrate. These frequently grow completely through the septa, so that again fibrinoid formation and also the inclusion of neighboring villi, together with their stroma, may be found in the septa. ]\Iany septa at the close of pregnancy seem to be composed entirely of fetal elements. The basal plate is traversed by maternal or uteroplacental vessels, destined for the intervillous space. Arteries and veins DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 163 Tr.( ^.■;;' Fig. 134. — Septum of a mature expelled placenta. Tr., trophoblast; Z., stroma of villus that has fused with the septum. X 50, pass through the plate in sinuous courses, but at their entrance into it they lose their muscularis and are represented by channels lined only by endothelium. And even this is usually lost^^ when °' The opinion of Waldeyer that the endothelium is continued for some distance upon the wall of the intervillous space, though in accordance with the older views, has not been confirmed by later investigators. 164 HmiAN EMBRYOLOGY. the vessels pass tlirough the fibrin layers of the plate. The arteries generally open in the region of the septa or close beside them, while the veins arise rather towards the middle of the cotyledons. This arrangement favors, to a certain extent, the intervillous circulation. Both arteries and veins, as a rule, traverse the wall of the intervillous space obliquely, yet villi are frequently sucked into the veins or they may, as has already been mentioned, be torn away and, passing into the veins, close them, thus producing disturbances in the circulation. Fig. 135. — Marginal portion of a mature expelled placenta with well-developed marginal .sinus distended with air. Ch.l., chorion Iseve ; D., fragments of decidua ; wp. G., uteroplacental vessels; K., superficial calcareous deposit; 0, opening into the marginal sinus (a tear, produced at birth by the division of a uterine vein), villi, which project into the marginal sinus, being seen through the opening; Ra., marginal sinus. X 1^. TEe conditions of the circulation in the placenta are quite peculiar and are found nowhere else in the body. The arteries open into a wide, very irregular space, extending throughout the entire placenta and limited only by fetal elements (and by fibrinoid). The space is formed from trophoblast lacunse and is filled with blood by the erosion of maternal vessels; these at first are of merely capillary size or but little greater. With the increasing importance of these afferent and efferent vessels they become gradually larger. Originally the opening of the vessels occurs in the region of the transition zone, and even later the union of the vessels with the intervillous space is still characterized by the fact that their endothelium has no continuity with any of the cellular elements that line the space (syncytium, basal ectoderm). The inter^dllous space is both the functional and the nutritive vascular space of the placenta; from it the villi derive their nourishment, the fetal circulation playing merely a subordinate role in this respect. Even after the death of the embryo the placenta may persist for a considerable time and may continue to grow, although only in an atypical manner. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 165 One of the efferent channels of the placenta is the marginal smus, Meckel's blood-channel, the sinus circularis; it is situated in the angle between the margin of the placenta and the chorion la!ve n> X. Fig. 136. — Margin of the placenta in the second month. From the same case as Fig. 118. Bos., decidua basalis ; Caps., decidua capsularis; Chp., chorion plate; Par. comp. and Par. sp., decidua parietalis compacta and spongiosa; Rs., marginal sinus; U.-L., lumen of the uterus. X 15. (Fig. 135), and does not encircle the entire placenta (it is wanting at the region shown in Fig. 129), but only about one-quarter, or at most seven-eighths, of the circumference (Budde) ; occasionally it can scarcely be recognized. It is not to be regarded as a con- 166 HUMAN EMBRYOLOGY. tinuous, regular vessel, but as a rather irregular space of varying diameter, which presents peripherally gaps and openings, cor- responding to uterine veins that have been torn away; and gaps of varying size also occur on its inner wall through which bunches of villi project and by which it is in communication with the intervillous space (Figs. 135 and 137). Budde agrees with earlier iv. R.r, Ch.l. Fig. 137. — Marginal sinus of a mature expelled placenta with an endothelium-like lining. Ch, Z., Chorion IsEve; /)., decidua parietalis; iu.i?., intervillous space; Z., remains of villi. X 18. authors in regarding the entire sinus as the marginal portion of the intervillous space, its inner wall, so far as it is formed, being produced by cohesion of the marginal villi and by the formation of fibrin. This explanation does not seem to suffice for all cases ; in younger ova greatly enlarged veins occur, which have a circular course and have been regarded as the marginal sinus {e.g., by Friolet and in the case shown in Fig. 136), and in the mature DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 167 placenta a complete endothelium-like lining is occasionally found in the sinus (Fig. 137). Nevertheless, this endothelium rests upon a necrotic substratum resembling an infarct and containing de- generating villi, and it is not impossible that the flat cells are derived from the epithelium of villi ; the question, however, needs further investigation. The relations of the circulation in later stages are not essen- tially different from those that obtain in younger ova; in the ma- ture expelled placenta also the intervillous space is, as a rule, but incompletely filled (Fig. 122) or almost empty (Fig. 129). Pla- centae fixed in situ are usually gorged with blood (Bloch, 1889). The circulation is determined, on the one hand, by the arrangement of the uteroplacental vessels, already described, and on the other, by the pregnancy pains already referred to (p. 136). The poverty in blood of the expelled placenta is further evidence of a compara- tively well-determined circulation. In considering the growth of the placenta a distinction must be made between growth in thickness and growth in area. The growth in thickness is chiefly due to an elongation of the villi and the associated separation of the chorion plate from the basal plate. Somewhat different from this is the view of Pfannenstiel (1903). As has already been pointed out, this author derives the syncytium from the endothelium of the maternal blood-vessels and regards the first-formed blood lacunse as greatly enlarged capillaries; he terms them the primary intervillous space. This space then enlarges by the veins which open on its floor and into which bunches of villi project, enlarging as the result of a degeneration of their walls, this degen- eration extending as far as the arteries which open between the veins; the tissue surrounding the arteries, however, persists to form the decidual pillars, that is to say, the septa placentae. The new spaee so formed is termed the secondary intervillous space. The projection of villi into the veins is incontestable (see p. 164) ; the process assumed to occur by Pfannenstiel cannot, however, have anj' important significance, since, firstly, anchoring villi are at that time attached to the basal plate in the neighborhood of the openings of the veins and their attach- ment would be broken by such a degeneration, without the possibility of forming a new attachment after the destruction of the cell columns. Secondly, the basal plate is not thicker in the early stages of development than it is later, and such a melting away of the portion of it that is in relation to the intervillous space would require a very intensive proliferation of the deeidua basalis for the replace- ment of the lost layers. But there is no histological evidence of the occurrence of such a proliferation. The question of the groirth in area of the placenta presents great difficulties, apart from that which occurs during the embryo- trophic stage of placentation. The most important factor in this latter growth is the splitting of the marginal deeidua, and the occurrence of this process makes it intelligible how the ovum, at first a mere dot but later increasing enormously in size, forms for itself a capsule, which projects beyond the level of the mucous membrane as far as the equator of the ovum or even beyond it 168 HU.AIAN EMBRYOLOGY. and fills a considerable portion of the uterus. But the growth in area of the placenta is not confined to the embryotrophic period; the placenta reaches its greatest relative extent at about the fourth month (Von Herff), at which time it occupies almost the half of the inner surface of the uterus. At this time it is depressed in a cup-hke manner, at the centre, in correspondence with the almost even curvature of the uterus. Later its growth is relatively less rapid than that of the uterus and it becomes flatter in correspond- ence with the stretching of the pregnant uterus and the slight flattening of its anterior and posterior walls. This, at first, rapid enlargement of the placenta has been re- garded by Hitschmann and Lindenthal as the result of a gradual inclusion in the placenta of the decidua capsularis and the portions of the chorion frondosum opposite to it, there being at the same time a stretching of the summit of the capsularis and of the chorion Iri've — in other words, it is due to a kind of unrolling and stretching of the chorion frondosum; but no evidence that such a process occurs is forthcoming. The possibility of a shifting of the margin of the placenta, as a result of unequal growth, is also worthy of consideration. Similarly, the later relative diminution of the placenta may be due to two different causes; either there is a degeneration of its marginal portions or there is again a shifting of the margin as the result of unequal growth. Pfannenstiel chiefly inclines toward a shifting of the margin, which has, indeed, the greater probability ; j'et the process cannot be a simp!e one on account of the manifold connections between the placenta and the subjacent tissues by means of the blood-ves- sels. The decidua basalis spongiosa with its large gland cavities forms, it is true, a very adaptable substratum. But an obliteration of the marginal portions in later stages is also very probable, since younger placentae (Fig. 130) show a relatively gradual transition into the chorion lasve and only later does a distinct placental margin appear. A solution of the problem may pei^raps be obtained in tlie following manner: Villi are probably formed only in early stages by the ingrowth of mesoderm into the trophojjlast shell (Hitschmann and Lindenthal) ; after the formation of the chorion plate and its two-layered epithelium the new-formation of villi must cease. The destruction of smaller villi in the subchorial fibrin stria is probable, but the larger ones must persist until the mature placenta, in which, as a rule, a villus corresponds to each cotyledon. An enumeration of the basally directed villi at dif- ferent stages would show whether the placenta was enlarged by taking into its territory new portions of the chorion and later diminished by excluding them again, or whether a certain con- stancy in the numbers occurred. Such enumerations have not, however, as yet been attempted. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 169 The situation of the placenta, which is determined at implanta- tion, IS just as frequently upon the anterior as upon posterior wall of the uterus ; more rarely it is lateral, in which case it may cover the opening of a tube.^^ According to Holzapfel (1898), in 107 cases the placenta was situated on the anterior wall in forty-two, on the posterior wall in forty-five (both inclusive of cases of extra- median positions and extension upon the fundus), in a tubal angle in fourteen, laterally below the opening of a tube in five. One case was a placenta praevia (covering the os internum uteri), but this condition is of rare occurrence, since according to Schauta but one case occurs in 1500-1600 pregnancies. It may result from a per- sistence of a portion of the capsularis with its intervillous space and supply of villi (placenta reflexa), or from an abnormally low implantation of the ovum, near the ostium internum, or perhaps from a double implantation on both walls of the uterus at the same time, as has occasionally been observed (Graf Spee) in guinea-pigs (Hitschmann and Lindenthal). V. THE MATURE AFTERBIRTH; THE AMNION, ALLANTOIS AND YOLK SACK UP TO MATURITY. The immediate causes of birth are still unknown, yet it may be said that in general the placenta is so altered at the close of preg- nancy by the continued modifications of the epithelium of the villi, the disappearance of the Langhans layer and syncytium, the forma- tion of fibrinoid and infarcts, that it can no longer perform its function of affording nourishment to the child, or can do so at best only insufficiently. The trophoblast and syncytium have a limited span of life and its close is reached with the close of pregnancy. The plane of separation of the afterbirth lies in the region of the decidua spongiosa, in which preparations for it have been made by anatomical conditions (the thin gland partitions) (Lang- hans) ; nevertheless, local separations occur also in the compacta, and, indeed, are regarded as the rule by Webster. The placenta, chorion laeve, and decidua compacta possess a certain amount of firmness and are actually separated from the spongiosa by any diminution of the inner surface of the uterus ; the result of this is the formation of a retro placental hcematoma, which begins to be formed at the first rupture of the vessels. The expelled placenta is, as a rule, disk-shaped and has a diameter of 16-20 cm., a thickness of 2^2-3 cm., and an average weight of something over 500 Gm. Variations are not rare in all dimensions and do not stand in any accurate relation to the "' The covering of an opening of a tube is evidence in favor of the splitting of the decidua marginalis during the growth of the ovum. 170 HUMAN EMBRYOLOGY. development of the child. The maternal or outer surface (Fig. 135) after the removal of the adhering blood-clots appears dark reddish gray, frequently with small, somewhat pale spots; is divided by furrows into 15-20 irregular lobes, (cotyledons) ; and is, in general, rather smooth, except for occasional adhering shreds of tissue (portions of the spongiosa). The paler spots correspond to anchoring villi and the bunches of villi which surround them, and the darker areas between them are caused by the blood of the intervillous space; yet these differences of color are evident only when the basal plate is relatively thin and the maternal blood has been retained in considerable quantity. Furthermore, small white scales (calcareous deposits), very variable in number, are usually to be seen; they usually occur in fibrin nodes (Figs. 127 and 129) and are usually near the basal plate. Larger white or yellowish masses projecting above the general surface are caused by the white infarcts. The uteroplacental vessels traverse the basal plate usually as greatly contorted canals (Bumm, Klein) ; the veins have a greater calibre and more delicate walls than the arteries. In each cotyledon there are about two or three veins situated centrally and from three to five peripheral arteries (see also p. 162). (Con- cerning the visibility of the choriodecidual vessels see p. 160.) The cotyledons are incompletely separated by the septa placentae, which extend from the furrows of the outer surface and by manipulation of the placenta, or even by birth-trauma, may be readily separated into two layers, so that the furrows appear markedly deepened. The fetal inner surface is whitish, and at first is covered by the amnion; after this is removed it is rather smooth, with the fetal vessels projecting from the surface. A marking of the sur- face, usually visible, consisting of paler spots on a darker back- ground, is due to the same causes as the corresponding marking of the outer surface, but is dependent, as regards its visibility, on the thickness of the subchorial layer of fibrin. From the sur- face project occasionally the placental cysts (p. 155) filled with a clear fluid. On section the fresh placenta is dark red and shows a spongy structure; the septa as well as the basal plate are more grayish-red. The arrangement and size of the cotyledons are determined chiefly by the chorionic villi. In general', each cotyledon corre- sponds to a villus, which, with its branches, fills the cavity of the cotyledon and is attached by numerous anchoring villi to the basal plate and the septa. Close to the origin of tlie villus from the chorion plate branches are given off, which divide immediately beneath the plate and are frequently included in the Langhans stria and the canalized fibrin. Small or rudimentary villi occur on the chorion plate only in the region of the placental margin. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 171 The chorion (chorion lagve) forms with the placenta a sacJi, which is opened in the afterbirth by a sht, often irregular or triradiate, and corresponding to the lower pole of the ovum. It is a grayish- or yellowish-red, easily torn membrane, whose outer surface is rough and has attached to it shreds of decidua and blood-clots. Its microscopic structure, and also that of the placenta at birth, has been described in Section IV. Hyrtl describes in the chorion la^ve, close to the margin of the placenta, occasionally occurring arteriovenous anastomoses be- tween fetal placental vessels of the size of a needle or more. The vessels which occurred in the chorion Iseve when it still possessed villi have completely disappeared at birth, so that the membrane is non- vascular ; yet occasionally some weak stems occur in the marginal parts of the mature membrane. The amnion is fused with the placenta and chorion, but may be separated from these as an independent, transparent, glistening Fig. 138. — Amniotic villi from a mature human afterbirth. X 100. membrane. After its separation its outer surface, and also the inner surface of the chorion, has a fibrillar appearance, as delicate connective-tissue strands which connect the two membranes have been torn by the separation. The amnion consists of a connective- tissue stroma and a cubical or cylindrical epithelium, in which Lonnberg finds fat granules of frequent occurrence. Earely (and principally in the placental region) the epithelium presents irregu- lar whitish growths, the amniotic villi (Fig. 138) (Ahlfeld), struc- tures which are of normal occurrence in ungulates. The epithelium over them is many-layered and the cells undergo comification and desquamation. Blood-vessels are wanting in the amnion. The fusion of the amnion and chorion is secondarily pro- duced by the disappearance of the extra-embryonic coelom towards the end of the second month (Strahl), as the result of the rapid growth of the amniotic sack. The amniotic epithelium is at first quite low and endothelium-like, and becomes cubical only in the second half of pregnancy (Bondi). (Compare also Figs. 96, 114, and 117.) G-ranules appear in the cells after the third month and Bondi also describes granules in the mature amnion that stain with neutral red. Stomata he could not find. 172 HUMAN EMBRYOLOGY. The liquor amnii, whose quantity amounts to about a litre at the end of pregnancy, is a secretion of the amniotic epithelium (Mandl, Bondi, Kreidl and Mandl). The evacuations of the fetal urinary bladder have no noteworthy significance in its production (compare also Wargaftig, 1907). The umbilical cord (funiculus umbilicalis) is a cord which is usually twisted anti-clockwise (to the left) ; it is normally about the same length as the child, but may be reduced almost to nothing or may reach three times the normal length. The twisting depends upon the unequal growth of the two umbilical arteries, and this again is associated with the sHght difference which exists in the diameters of the two arteries from the beginning and with the difference of blood-pressure in the arteries as a result of the dif- ference in frictional resistance (Neugebauer). The embryo, which floats freely in the amniotic fluid and is almost sustained by it, is passively rotated as the result of the gradual twisting of the cord, and local growths of the arteries produce the false nodes which occur in this. The angle at which the cord is inserted into the placenta varies between 0° and 90° and is greatest in cases where the insertion is central; when the attachment is at an acute angle there occurs between the cord and the placenta what is known as SchuUze's amniotic fold, produced by the incomplete apposition of the amnion to the cord and placenta, owing to the persistence of the yolk sack and its vessels. Microscopically there may be distinguished upon the surface of the cord the single-layered cubical or flattened amniotic epithe- lium, in which, according to Koster, stomata occur ; the connective- tissue layer of the amnion is not distinguishable as a separate sheet. The stroma of the cord consists of a gelatinous tissue, Wharton's jelly, which is characterized by possessing stellate cells, resembling embryonic cells, a scanty development of fibrillse, and wide intercellular spaces. At the periphery and in the neighbor- hood of the vessels the tissue is arranged in concentric layers (Fig. 139). The vessels, especially the arteries, show stout longi- tudinal muscle-bundles beneath the circular musculature, and these, when contracted, form strong projections which facilitate the com- plete closure of the vessels when the cord is severed (Henneberg, Bucura). Vasa vasorum can be distinctly injected in the veins, according to Gonner, and while they cannot be injected in the arteries yet their origins can be distinguished when the arteries are laid open. Nerves can be followed from the abdominal cavity of the child only to the umbilicus or a short distance beyond it; they do not occur in the free portion of the cord (Bucura). Finally, remains of the allantoic duct occur in the mature cord. The allantoic duct is, according to Lowy, still hollow through- out its entire length in embryos with a greatest length of 8 mm., DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 173 but has a very variable diameter, and in embryos of 9 mm. its lumen is partly obliterated toward the peripheral end, while in those of 14 mm. it is open only in a small portion of its extent and that portion is irregularly expanded ; even in the fourth month remains of the duct are still to be foimd in the neighborhood of the embryo with a lumen and cubical epithelium (Fig. 139) ; and in the mature cord there are to be found occasional epithelial pearls and occa- sionally coils and lateral outpouchings as remnants of the duct. The yolh stalk becomes divided shortly after the closure of the umbilicus. In the umbilical cord remains of it mav be found r --^Sk ."X^^^; Allg A. u. Fig 139 —Central portion of a transverse section of an umbilical cord of an embryo of the fourth month (vertex-breech length 1314 cm.). The Wbartonian jelly is arranged in concentric layers. A. u. and F.M.,iimbihcal vessels; Aiiff., allantoic duct. The yolk stalk has disappeared. X30. up to the third month, but at maturity these remnants have prob- ably completely vanished. Portions of the omphalomesenteric vessels, occasionally filled with blood, are also to be found in the third month and, rarely, they persist until the close of pregnancy (Lonnberg). . i --j. j. The yolli sack (vesicula umbilicalis) is a normal constituent of the mature afterbirth (B. S. Schultze), but on account of its minuteness and the irregularity of its situation it is readily over- looked It occurs between the chorion and amnion, on the placenta or the chorion Iseve, or even at the opposite pole of the ovum ; very rarely it even appears to lie in the umbilical cord itself (Lonnberg) Its variability in position is due to the length of the yolk stalk 174 HUMAN EMBRYOLOGY. and the width of the exocoelom. The Schultze amniotic fold may serve as a guide to it, but between the direction of the fold and the connecting line between the sack and the umbilical cord there may be a divergence of even 90° (Lonnberg) . Macro scopically the mature yolk sack is usually a round or oval, flattened, white or yellow body with a diameter of 1-5 mm. ; microscopically it presents a mesodermal investment and its contents are flake-like and partly calcified, but no epithelium can be detected. At the commencement of its development the yolk sack shows a certain amount of differentiation. The blood and the vessels of the ovum first appear in its wall and, later, for a considerable time it is a region of blood formation and consequently richly vascular (Fig. 140). Its epithelium forms gland-like invagina- FiG, 140.- -Wall of the yolk sack from a human embryo of a greatest length of 9 mm. cysts; C, blood-vessel; Spl., splanchnopleure. X 300. Cy., intra-epithelial tions (Graf Spee) or intra-epithelial cysts, produced by cell de- generation (Fig. 140). (Compare also Branca.) Graf Spee also describes the occurrence of giant cells in the epithelium of younger stages and regards them as associated with the blood formation. According to Meyer and Jordan there occur at the end of the first month epithelial buds, solid or hollow outgrowths, which project into the mesoderm, yet these structures are rather variable in their occurrence. Later the epithelium becomes flat and, finally, degenerates with the thickening up of the contents of the vesicle.' VI. THE UTERUS POST PARTUM Since the plane of separation of the afterbirth passes through the decidua spongiosa, the numerous flattened gland cavities of the latter are opened at birth. In the region of the placenta the spongiosa is somewhat thinner than elsewhere (see p. 159). Hem- orrhage from the opened vessels is prevented by the compression DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 175 of the vessels within the contracted muscularis and the compressed mucosa, but blood-serum exudes for a considerable time from the opened tissue spaces. The epithelium of the superficial gland cavities, which Langhans believes is responsible for the renewal of the superficial epithelium of the mucous membrane, has httle sig- nificance in this respeet,"*^ having undergone extensive modifica- tions during pregnancy (p. 140). The regeneration is rather from the deepest gland zone, the limiting layer of His ; but since this is exposed only in spots, the more superficial layers of the spongiosa and the persisting remains of the compacta must be destroyed by coagulation necrosis and be expelled. Previously to their ex- pulsion they form a whitish-yellow layer resting upon the mucous membrane and have been usually regarded as necrosed compacta. The line of demarcation between the necrotic and persisting layers "is formed by the basal surfaces of all the gland cavities occurring in its neighborhood, which thus later become the sur- face" (of the mucosa) (Wormser). This line becomes distinct on the second day after birth ; the expulsion of necrotic tissue begins on the fifth day and is completed everywhere on the tenth to the twelfth day after birth. The gaps between the gland cavities are covered over after the expulsion "by lateral shifting, flattening, and amitotic increase" of the epithelium; mitoses seem to be at first entirely wanting. As the epithelium grows out from the deeper portions of the glands many-layered zones of epithelium and multi- nuclear masses of protoplasm are formed, and, at the same time, degenerations occur everywhere in the epithelium; the formation of vacuoles, shrinkage of the nuclei, and degeneration of cells and nuclei are frequently to be found at the surface. Mitoses first ap- pear about two weeks after delivery, and, finally, probably only those epithelial cells persist which have been newly formed by the mitotic process. Leucocytes wander in rather large numbers through the mucous membrane. The decidua cells degenerate in the vicinity of the line of demarcation, probably by a fatty de- generation; and the connective tissue scaffolding thus persists as an empty mesh-work. This process Wormser terms areolar degeneration, and he imagines the meshes to be eventually re- occupied by inwandering connective-tissue cells ; this idea, how- ever, is rather improbable. The decidual modifications which have occurred in the deeper layers of the mucous membrane disap- pear, the syncytial giant cells degenerate and vanish, and in two or three weeks after birth the regeneration of the mucous mem- brane, accompanied by an increase in its thickness, is so far com- pleted that stroma, tubular glands, and a surface epithelium are "^ The account given here is principally based on the observations of Wormser (1906). 176 HUMAN EMBRYOLOGY. already present; nevertheless, the epithelium continues to show degenerations and regenerations for some time. The mucous membrane of the cervix uteri becomes looser dur- ing pregnancy and shows serous infiltration and an increase m the glands, while after birth it presents zones of traumatic hemor- rhages, the epithelium, however, being retained. Leucocytes also wander out through this mucous membrane in considerable numbers. The muscularis uteri increases during pregnancy to about twentv-four times its original size, partly by hypertrophy (the formation of new fibres by the division of those already present) and partly by hyperplasia of the individual fibres (KoUiker) . The reduction is produced l)y a diminution of the size of the fibres and perhaps also by the complete degeneration of some of them (Von Ebner). The peritoneal covering of the uterus and of the parts in its neighborhood shows here and there during pregnancy decidua-like growths (Schmorl), and similar growths occur in the tunica albuginea of the ovaries (Lindenthal). The tubes, with the excep- tion of increased blood-supply and some serous infiltration, are but little altered. BIBLIOGRAPHY. Ahlfeld, r. : Ueber die Zotten des Amnion, Archiv f . Gynak., vol. vi, 1874. Zur Genese der Amnionzotten, ibid., vol. vii, 1875. Bab, H. : Konzeption, Menstruation u. Schwangersehaft, Deutsche med. Woeh., 1908. Beneke : Ein sehr junges menschliches Ei. Deutsche med. Wochenschr., Jahrg. 30, 1904; also Monatssehr. f. Geb. und Gyn., vol. xix, 1904. Bloch: Ueber den Bau der menschlichen Placenta, Beitr. z. path. Anat., vol. iv, 1889. BoNDi, J. : Zur Histogenese des Amnionepithels, Zentralbl. f . Gynak., Jahrg. 29, 1905. BOKNET, R. : Ueber Syncytien, Plasmodien und Symplasma in der Placenta der Siiugetiere und des Menschen, Monatssehr. f. Geb. und Gyn., vol. xviii, 1903. Lehrbuch der Entwicklungsgeschichte, Berlin, 1907. Beanca, a. : Sur I'endoderme ombilical de I'embryon humain, Bibl. Anat. Supple- ment, 1908 (C. R. Ass. Aiiat.). Bryce, T. H., and Teacher, J. H. : An Early Ovum Imbedded in the Decidua; also conjointly with J. M. Munro Kerr, An Early Ovarian Pregnancy, under the title : Contributions to the Study of the Early Development and Imbedding of the Human Ovum, Glasgow, 1908. Bucura, C. : Ueber den physiologischen Verschluss der Nabelarterien und iiber das Vorkommen von Langsmuskulatur in den Arterien des weibliehen Genitales, Zentralbl. f. Gyniik., Jahrg. 27, 1903. Ueber Nerven in der Nabelschnur und in der Placenta, Zeitsehr. f . Heilkunde, vol. xxviii, 1907. (See also Zentralbl. f. Gynak., vol. xxxii, 1908.) BuMM, E. : Ueber Uteroplacentargefasse, Archiv f . Gynak., vol. xxxv, 1889. Ueber die Entwicklung des miitterlichen Blutkreislaufes in der menschlichen Placenta, Archiv f. Gynak., vol. xliii, 1893. BuSALLA : Beschreibung und histologisches Untersuchungsergebnis eines neuen Falles von Eierstocksschwangerschaft, Archiv f. Gynak., vol. Ixxxiii, 1907. DEVELOPMENT OF EGG MEMBRANES AND PLACENTA. 1 1 1 CovA, E. : Ueber ein menschliches Ei der zweiten Woche, Arehiv f. Gynak., vol. Ixxxiii, 1907. DissB, J. : Die Eikammer bei Nagern, Insektivoren und Primaten, Ergebnisse d. Anat. und Entw., vol. xv for 1905, Wiesbaden, 1906. FossATi, G. : Ueber Nen'en in der Nabelschnur und in der Placenta, Zentralbl. f . Gynak., vol. xxxi, 1907. Franque, 0. V. : Die Entstehung der velamentosen Insertion der Nabelschnur, Zentralbl. f. Gynak., 1901. Frassi, L. : Ueber ein junges menschliches Ei in situ, Arehiv f . mikrosk. Anatomie, vol. Ixx, 1907. Weitere Ergebnisse des Stadiums eines jungen menschliehen Eies in situ, ibid., vol. Ixxi, 1908. Friolet, H. : Beitrag zum Studium der menschliehen Placentation, Beitrage z. Geburtsh. u. Gynak., vol. ix, 1904; also Dissert., Basel. Fuss, S. : Die Bildung der elatischen Faser, Arch. f. pathol. Anat., vol. clxxxv, 1906. GiESB, H. : Histologische Untersuchungen liber den weissen Inf arkt der Placenta, Dissert., Halle, 1905. GOENNER, A. : Ueber Nerven und emahrende Gef asse im Nabelstrang, Monatsschr. f. Geburtsh. u. Gynak., vol. xxiv, 1906. Grosser, 0. : Vergleichende Anatomie und Entwicklungsgeschichte der Eihaute und der Placenta mit besonderer Beriicksiehtigung des Menschen, Lehrbuch fiir Studierende und Aerzte, Wien, 1909. Happe, H. : Beobachtungen an Eihauten junger menschlicher Eier, Anat. Hef te, vol. xxxii, 1906. Henneberg, B. : Beitrage zur feineren Struktur, Entwicklungsgeschichte und Physiologie der Umbilicalgefasse des Menschen, Anat. Hefte, vol. xix, 1902. Herpp, 0. V. : Beitrage zur Lehre von der Placenta und von den miitterlichen Eihiillen, Zeitschr. f. Geb. u. Gyn., vol. xxxv, 1896, and vol. xxxvi, 1897. HlTSCHJlANN, F. . Die Deportation der Zotten und ihre Bedeutung, Zeitschr. f . Geb. u. Gyn., vol. liii, 1904. HiTSCHMANN, F.j and Lindenthal, 0. : Ueber das Wachstum der Placenta, Zentralbl. f. Gyniik., 1902. Ueber die Haftung des Eies an atypischem Orte, ibid., 1903. Der weisse Infarkt der Placenta, Arehiv f. Gyniik., vol. Ixix, 1903. Hofbauer T. : Grundziige einer Biologie der menschliehen Placenta mit besonderer Beriicksiehtigung der Fragen der fetalen Emiihrmig, Wien u. Leipzig, 1905 (literature). Die menschliche Placenta als Assimilationsorgan, 1907. HOPJIEIBR, M. : Die menschliche Placenta, Wiesbaden, 1890. HoLL, M. : Ueber die Blutgefiisse der menschliehen Nachgeburt, Sitz-Ber. k. Akad. Wiss., Wien, vol. Ixxxiii, 1881. HoLSTi, 0. N. : Weitere Beitrage zur Kenntnis der Embryotrophe, II. Ueber die Fettzufuhr zum menschliehen Ei, Anat. Hefte, vol. xxxiii, 1908. HoLZAPPEL, K.: Ueber den Placentarsitz, Beitrage z. Geburtsh. u. Gyniik., vol. i, 1898. Zur Pathologie der Eihiiute, Beitr. zur Geburtsh. und Gynijk., vol. vm, 1903. (The author regards the amniotic villi as transplanted embryonic epidemiis.) Hyrtl, J. : Die Blutgef asse der menschliehen Nachgeburt in normalen und abnormen Verhaltnissen, Wien, 1870. IwASE, Y. : Ueber die zyklisehe Umwandlung der Utemssehleimhaut, Zeitschr. f. Geburtsh. und Gyniik., vol. Ixiii, 1908. Jordan, H. E. : The Histology of the Yolk-sac of a 92-mm. Human Embryo, Anat. Anzeiger, vol. xxxi, 1907. Jung, Ph.: Beitrage zur friihesten Ei-Einbettung bemi menschliehen \\ eilie, ' Berlin, 1908. Vol. I.— 12 178 HUMAN EJIBRYOLOGY. Kastschenko, N. : Das mensehliche Chorionepithel und dessen Rolle bei der Histogenese der Placenta, Archiv f. Anat. u. Phys., An at. Abt., 1885. Keheer, E. : Der placentare Stoifaustausch in seiner physiologischen und patho- logischen Bedeutung, Wurzburger Abhandl., vol. vii, parts 2 and 3, 1907 (literature). Klein, G. : Mikroskopisches Verbalten der Uteroplaeentargefasse, in Hofmeier: Die mensehliche Placenta, Wiesbaden, 1890. Zur Entstehung der Placenta marginata und succenturiata, ibid. KoLLiiANN, J.: Kreislauf der Placenta, Chorionzotten und Telegonie, Zeitsehr. f. Biologie, vol. xlii, 1902. Langhans, Th. : Zur Kenntnis der menschlichen Placenta, Archiv £. Gynak., vol. i, 1870. Untersuchungen iiber die mensehliche Placenta, Archiv f. Anat. u. Phys., Anat. Abt., 1877. Ueber die Zellsehicht des menschlichen Chorion, Festsehr. f. Henle, Beitrage z. Anat. u. Embryol., Bonn, 1882. Syncytium und Zellsehicht, Beitrage z. Geburtsh. u. Gynak., vol. v, 1901. Leopold, G. : Uterus und Kind, mit Atlas, Leipzig, 1897. Ueber ein sehr junges menschliches Ei in situ, Leipzig, 1906. Leopold and Ravano : Neuer Beitrag zur Lehre von der Menstruation und Ovula- tion, Archiv f. Gynak., vol. Ixxxiii, 1907. LoNNBERG, J. : Studien iiber das Nabelblaschen an der Naehgeburt des ausgetragenen Kindes, Stockholm, 1901 (literature). Mandl, L. : Histologische Untersuchungen iiber die sekretorische Tatigkeit des Amnionepithels, Zeitsehr. f. Geb. u. Gjti., vol. liv, 1905. Weitere Beitrage zur Kenntnis der sekretorischen Tatigkeit des Amnionepithels, ibid., vol. Iviii, 1906. Maechand, r. : Beobachtungen an jungen menschlichen Eiern, Anat. Hefte, vol. xxi, 1903. Beitrag zur Kenntnis der normalen und pathologischen Histologie der Decidua, Archiv f. Gynak., vol. Ixxii, 1904. Mebttens, J. : Beitrage zur normalen and pathologischen Anatomie der mensch- lichen Placenta, Zeitsehr. f. Geb. u. Gyn., vol. xxx, 1894, and vol. xxxi, 1895. Meyer, A. W. : On the Structure of the Human Umbilical Vesicle, Amer. Journal of Anatomy, vol. iii, 1904. MiNOT, Ch. S.: Uteras and Embryo, Journal of Morphology, vol. ii, 1889 (literature). The Implantation of the Human Ovum in the Uterus, New York Med. Journ., vol. Ixxx, 1904. Neugebauee, L. a. : Morphologie der menschlichen Nabelschnur, Breslau, 1858. NiTABUCH, R. : Beitrage zur Kenntnis der menschlichen Placenta, Dissert., Bern, 1887. Peters, H. : Ueber die Einbettung des menschlichen Eies und das friiheste bisher bekannte mensehliche Placentationsstadium, Leipzig and Wien, 1899 (literature). Zum Kapitel: Langhans'sche Zellsehicht, Zentralbl. f. Gyniik., 1900. Beitrag zur Kasuistik der Vasa praavia und Gedanken zur Theorie der Insertio velamentosa, Monatsschr. f. Geb. u. Gyniik., vol. xiii, 1901. Pfannenstiel, J. : Die ersten Veranderungen der Gebiirmutter infolge der Schwangerschaft, — Die Einbettung des Eies, — Die Bildung der Placenta, der Eihiiute und der Nabelschnur, — Die weiteren Veranderungen der genannten Gebilde wahrend der Schwangerschaft, in Winckel: Handbueh der Geburtshilfe, vol. i, Wiesbaden, 1903 (literature). Reinstein-Mogilowa, A.: Ueber die Beteiligung der Zellsehicht des Chorions an der Bildung der Serotina und Reflexa, Virchow's Archiv, vol. exxiv, 1891. DEVELOPMENT OP EGG MEMBRANES AND PLACENTA. 179 ROHR, K. : Die Beziehungen der mutterlichen Gef iisse zu den intervillosen Raume.i der reifen Placenta, speziell zur Thrombose derselben (weisser Infarkt) Virchow's Archiv, vol. cxv, 1889. ROSSI-DOBIA, J.: Ueber die Einbettung des menschliclien Eies, studiert an einem klemen Ei der zweiten Woehe, Archiv f. Gyuiik., vol. Ixxvi, 1905. Huge, C. : Die Eihiillen des in der Geburt beflndlichen Uterus, Bemerkungeii iiber den Ort und die Art der Ernahrung des Kindes in demselben, in : SCHROBDER, K. : Der schwangere und kreissende Uterus, Beitrage zur Anatomie und Physiologie der Geburtskunde, mit Atlas, Bonn, 1886. SCHICKELE, G. : Die Chorionektodermwueherungen der menschliehen Placenta, ihre Beziehungen zur Entstehung der Cysten und Pibrinknoten der Placenta, Beitr. z. Geb. u. Gyn., vol. x, 1905. Sellheim, H. : Physiologie der weiblichen Geschlechtsorgane, in Nagel, Handbuch der Physiologie, vol. ii, 1907. SiEGENBEEK VAN Heukblom : Ueber die mensehliche Placentation, Arch. f. Anat. u. Phys., Anat. Abt., 1898. Spee, p., Graf: Ueber die mensehliche Eikammer und Decidua reflexa, Anat. Anzeiger, vol. xiv, Suppl., 1898. Neue Beobachtungen iiber sehr friihe Entwieklungsstufen des menschliehen Eies, Arch. f. Anat. u. Phys., Anat. Abt., 1896. Epidiaskopische Demonstration eines jungen Stadiums der menschliehen Eiein- bettung, Verhandl. Deutsch. Ges. f. Gynak., xi Meeting in Kiel, 1905, Leipzig, 1906. Steffeck, p.: Der weisse Infarkt der Placenta, in Hofmeier: Die mensehliche Placenta, Wiesbaden, 1890. Stolper, L. : Zur Physiologie und Pathologie der Placentation, Monatsschr. f. Geb. u. Gyn., vol. xxiv, 1906. Steahl, H. : Die mensehliche Placenta, Ergebnisse d. Anat. u. Entwickl., vol. ii, 1893 (literature). Neues iiber den Bau der Placenta, ibid., vol. vi, 1897 (literature). Placentaranatomie, ibid., vol. viii, 1899 (literature). Die EmbryonaJhiillen der Sanger und die Placenta, Hertwig's Handbuch der vergl. u. experiment. Entwicklungslehre, vol. i, part ii, Jena, 1906 (pub- lished 1902) (literature). Der Uterus post partum, Ergebnisse d. Anat. u. Entw., vol. xv, 1906 (literature). Van" Cauwenbebghe, Recherches sur le role du syncytium dans la nutrition embryonnaire de la femme. Archives de Biol., vol. xxiii, 1907 (literature). VoiGT, J. : Ueber das Verhalten von miitterliehen und kindlichen Elementen an der Einnistungsstelle jiingerer menschlicher Eier, Zeitschr. f. Geburtsh. u. Gyniik., vol. liv, 1905. Waldbtbb, W. : Ueber den Placentarkreislauf des Mensehen, Sitz.-Ber. k. preuss. Akad. Wiss., 1887. Bemerkungen iiber den Bau der Mensehen- und AfEenplacenta, Archiv f . mikr. Anat., vol. xxxv, 1890 (literature). Wargaftig, Gr. : Der augenblickliche Stand der Lehre von der Herkunft, der Physiologie und Pathologie des Fruchtwassers, Diss., Freiburg, i. Br., 1907. Webster, J. C. : Human Placentation, Chicago, 1901. Die Placentation beim Mensehen, Transl. by Kolischer, Berlin, 1906. Wederhake, J.: Ueber Plasma- und Deciduazellen, Monatsschr. f. Geburtsh. u. Gynak., vol. xxiv, 1906. WoLSKA, W. : Ueber die von Ruge beschriebene f etale Vasknlarisation der Serotina, Dissert., Bern, 1888. WORMSBR, E. : Die Regeneration der Uterussehleimhaut nach der Geburt, Archiv f . Gynak., vol. Ixix, 1903 (literature). VIII. DETERMINATION OF THE AGE OF HUMAN EMBRYOS AND FETUSES. Br FRANKLIN P. MALL or Baltimoee. It would be relatively easy to determine the age of human embryos were it possible to fix with certainty the time of concep- tion, that is, the time at which the spermatozoon enters_ the ovum. However, this question, which is directly associated with that of the duration of pregnancy, and must be discussed with it, has been a most important one in anatomy for ages, and it appears to be gradually approaching a satisfactory solution. In ancient times it was generally believed that the duration of pregnancy in man, unlike that in lower animals, was of very uncertain length ; and it was not until the seventeenth century that it was more accurately fixed, by Fidele of Palermo, at forty weeks, counting from the last menstrual period. In the next century Haller found that if pregnancy is reckoned from the time of a fruitful copulation it is usually thirty-nine weeks, and rarely forty weeks in duration. In general these results are fully confirmed by the thousands of careful data collected during the nineteenth century. The difficulties encountered in determining the age of an embryo are due to the impossibility of determining the exact time of fertilization, for this does not necessarily follow immediately after copulation, and it is related only in a loose way with men- struation, the error in calculation in the second case often being a full month; but to the present time it has been most convenient, and probably most nearly correct, to rate the age of an embryo and the duration of pregnancy from the last menstrual period. However, from thousands of records it is found that the mean duration of pre,gnancy varies in first and second pregnancies, is more protracted in healthy women, in married women, in winter, and in the upper classes. As in lower animals, it varies very much in individual cases independently of any assignable cause. In general, it is longer when the new-bom infants are over 50 cm. long, the mean difference being, according to Issmer, fifteen days between those that are 48 cm. and those that are 53 cm. long. Furthermore, it is well known that other mammalian embryos of the same age vary much in size, and, although we have a variety of data which bear upon the time of conception and 180 AGE OF EMBRYOS AND FETUSES. 181 the age of young human embryos, none are of more value, as Von Baer has pointed out, than those obtained from comparative embryology. The first step toward the solution of the problem was made by Von Baer when he discovered the human ovum. Next it was proved by Bisehoff that ova are usually liberated periodically during the menstrual period, independently of copula- tion, and that the Graafian follicles of the ovary which contained these are converted into corpora lutea. Thus the first step was taken, for it had been shown that a recent ovulation is marked by a fresh corpus luteum and that in turn this usually takes place during the menstrual period. The excellent work of Bischoff on fertilization in the rabbit and dog demonstrated that soon after copulation the spermatozoa pass through the uterus into the tube, where the ovum is usually met. When copulation takes place during the period of rut the ovum is usually fertilized within twenty-four hours, and in case the ovum is not fertilized upon the surface of the ovary or in the upper end of the tube it soon degenerates. However, this second point is not well established for mammals, but it is known that unfertilized hen's egg's can not be fertilized in the lower part of the oviduct. Since segmentation takes place in the mammalian ovum while it is in the uterine tube it is highly improbable that a human ovum could be fertilized after it has reached the uterus, but instead is probably always fertilized upon the surface of the ovary or in the upper part of the tube, as the frequency of tubal pregnancies indicates. In 1868 Eeichert obtained a very small human ovum measur- ing 5.5 by 3.3 mm. from a woman who commited suicide exactly two weeks after her menstrual period failed to appear. This ovum was studied with great care and described at great length by Reichert, and has become the classic specimen upon which to reckon the age of young human ova as well as to fix the time of fertilization. In one ovary -there was found a well-developed corpus luteum with but very little blood in its centre. Eeichert then studied the condition of the ovaries during menstruation and found that in nineteen specimens out of twenty-three the Graafian folHcles had ruptured in the beginning of the period, while in four they were still unruptured. From these observations he concluded that, as a rule, ovulation takes place just before menstruation and that in case the ovum is fertilized menstruation is missed. This view changed the entire aspect of the question at once. Formerly it was believed that the ovum came down the tube slowly and was fertilized some days or weeks later, and now Eeichert was led to believe that ovulation and fertilization took place but a few days before menstruation and that the presence of a microscopic fer- tilized ovum in the upper end of the tube arrested entirely the menstrual hemorrhage which was about to appear. With great 182 HU.MAN EMBRYOLOGY. force he discussed the whole question and decided that his specimen must be two weeks and not six weeks old. Already Von Baer had noted that the human ovum was precocious in its early development, but Eeichert's conclusion made it much more so. However, it was easier to believe that Eeichert's ovum was two weeks than six weeks old and there seemed to be no other possibility. The theory of Reichert regarding the time of conception was accepted by His as the most probable, and accordingly he gave the probable age of embryos up to 35 mm. long. Due to his great influence the Reichert theory has been generally accepted and the most remarkable distortions have been made to fit individual specimens into it. The best known case is the Peters ovum, a specimen 1.6 X 0.9 mm. which was obtained thirty days after the last period, and Peters estimates its age to be three or four days. According to Weysse a pig's ovum of the same size is from nine to eleven days old, and according to Bischoff and Bonnet a dog's ovum 2 mm. in diameter is from twelve to fourteen days old. Unfortunately Peters does not describe the condition of the corpus luteum in this specimen, at present the only reliable index by which we can hope to determine the age of young ova. It is imperative that we standardize the corpus luteum of the first weeks of pregnancy and that in all cases when embryos are obtained at autopsy the ovaries should be carefully described and sections of the corpus luteum should be made and pictured. The ovum of Merttens, 3x2 mm., was obtained from a uterine scraping twenty-one days after the last period. Here there was no lapsed period, and twent^^-one or fewer days does not seem to me to be an unreasonable age for a human ovum 3 mm. in diameter. That the Peters ovum was older than four days is indicated by the morning sickness that preceded the lapsed period, and, if I am not mistaken, morning sickness is sometimes the first sign of pregnancy. Among the records of embryos under 5 mm. long I have been able to find thirty-six with menstrual history given (Fig. 147). According to the Reichert-His theory about twenty-five days would have to be subtracted from the ages of twenty-seven of them to make them correspond with the remaining nine. In other words. His would rate nine of them from the last period and twenty- seven from the first lapsed period. We are evidently dealing with two groups of young embryos which correspond with ovulations of two menstrual periods, but which two is still uncertain. Must we add twenty-eight days to each of the group of nine or subtract twenty-eight days from the group of twenty-seven? The investigations of Bischoff, Dalton, Williams, Reichert, Arnold, Leopold, Leopold and Mironoff, and Leopold and Ravano have shown conclusively that ovulation and menstruation are AGE OF EMBRYOS AND FETUSES. 183 usually synchronous, but menstruation often occurs without ovula^ tion and sometimes ovulation takes place in the intermenstrual period. In Leopold and Mironoff's forty-two cases ovulation occurs thirty times with menstruation, once without, and ten times there is menstruation without ovulation. The ninety-five selected cases of Leopold and Ravano show that ovulation and menstru- ation coincide in fifty-nine and are independent of each other in the remaining thirty-six cases. In other words, the connection between ovulation and menstruation is very loose and the two coincide in only two-thirds of the cases. Furthermore, ovulation occurs frequently during pregnancy. [Eavano.] The data of the other investigators give a similar distribu- tion of ovulations, and there is no marked evidence that ovulation precedes menstruation, as is required if Reichert's theory is true. It is to be hoped that this subject will be carefully studied in some large clinic where many normal ovaries are examined in abdominal operations. Then the age of corpora lutea could be standardized and subsequently applied to autopsy and other cases in which young ova are found in the uterus. In the recent work of Leopold and Eavano only the estimated age of the corpus luteum is used to determine the time of ovulation. in relation to menstruation. At the present time their determination of the age of the corpus luteum is the best which we possess, but I believe that it is pos- sible to standardize better the corpus luteum of the first week, that is, those which are formed during menstruation. This must be studied first, then that of the second week, and so on. Through this method we can determine with much greater precision the probable age of a corpus luteum. At any rate, the separation of young human embryos of the same size into two groups to correspond with two previous men- strual periods indicates that pregnancy usually takes place in the neighborhood of menstrual periods, and facts regarding the dura- tion of pregnancy bear this out. Leuckart tabulated 110 cases of births during the first ten months of married life and found that the maximum number were on the two hundred and seventy-fifth day, after which they fell off and increased again to a second maxi- mum on the two hundred and ninety-third day. He l)e]ieves that in those cases which came in the first maximum the ovum was obtained from an ovulation which preceded marriage and those that fell in the second maximum belonged to the first menstruation after marriage. He was able to collect eight cases in which the menstrual history was given. In four of them, in which marriage occurred during the third and fourth week after the beginning of menstruation, the women menstruated once after marriage. In the remaining four, in which the marriage followed immediately upon the cessation of menstruation, three did not menstruate again and one menstruated but once before pregnancy. In the second 184 HUMAN EMBRYOLOGY. case where newly-married women do not menstruate at all, we must assume that the ovulation of the last period gave rise to the pregnancy; that ovulation delivered the ovum into the upper end of the tube and soon became fertilized. In the first case the spermatozoa reach the ovary and there await the ovum from the ovulation which takes place with the first menstruation after mar- riage. It follows from the above that a fertilization immediately before menstruation does not cause the period to lapse, which contradicts the main proposition in the Eeichert-His theory. In fact women quite frequently menstruate more than once after the beginning of pregnancy, and at present there are no data to show that a woman who has not copulated since the last menstruation cannot be pregnant. Some additional light is thrown upon the question of the time of conception by a study of the duration of pregnancy as estimated from the last menstrual period as well as from the time of copulation. According to the more recent statistics of Issmer the average duration of pregnancy, in 1220 cases, is 280 days when estimated from the first day of the last menstrual period, and, in 628 cases, 269 days when estimated from the fruitful copulation. In general these two figures cor- respond with those of Ahlfeld, Hecker, and Hasler, who also collected about 500 cases in which the time of fruitful copulation was given. So in a group of about 1200 cases the duration of pregnancy is fully ten days longer when reckoned from the last period than when from the time of copulation. It may be noted that the data regarding fruitful copulation must be taken with the greatest reserve, for many of them are. from unmarried women and in but few of them does the fruitful copulation precede the menstrual period. However, it is remarkable that the results of the different observers are practically the same, each time giving a difference of a week or ten days. Alilfeld further classified the cases, giving the time of copulation in relation to menstruation. On last day of menstruation. First twelve days after beginning of men.struation. First seven days after end of menstruation. Married women . Per cent. 35.55 ■25.49 Per cent. 88.44 70.98 Per cent. 88 88 70.58 Similar figures are given by Issmer. Time of copulation. No. of cases. Average duration of pregnancy. First week of menstrual period 172 164 72 45 277 days 279 days 287 days 285 adys Second week of menstrual period Third week of menstrual period Fourtli week of menstrual period AGE OF EMBRYOS AND FETUSES. 185 From these figures it is seen that most pregnancies take place during the first week after menstruation and that the duration of pregnancy is longer if copulation takes place towards the end of the intermenstrual period. And this is explained if we assume that in the first week, especially the first few daj^s after the cessation of menstruation, the ovum is in the upper end of the tube awaiting the sperm and that conception immediately follows copulation. When the fruitful copulation takes place in the latter two weeks of the month the opposite is usually the case; the sperm wanders to the ovary and there awaits the ovum; and, therefore, on an average, pregnancy is prolonged in this group of cases, when de- termined from the time of copulation. This explanation fits all the facts but opposes the Reichert-His theory. We have finally the argument given by comparative embryol- ogy. Why should the human ovum be precocious in its early growth? We have good data upon the rabbit, dog, pig, and sheep ; and, in general, if we apply the Eeichert-His theory, the growth of the human ovum is at first far more rapid than any of them, and is then overtaken by the rabbit when the embryo is 5 mm. long, by the pig at 15 mm., and by the dog at 20 mm. The duration of pregnancy in the rabbit is 30 days ; in the dog, 63 ; in the pig, 120 ; in the sheep, 154; — why should these animals grow slower at first than man! If the age of human embryos is estimated from the last menstrual period or a few days later, a curve of growth is obtained which corresponds fairly well with that of lower animals. Possibly it may be allowable to compare early human develop- ment with that of the dog. According to Marshall ovulation in the bitch occurs after bleeding from the external opening has been going on for some days, or when it is almost or quite over. It takes place quite independently of coition. Up to this time the bitch will not copulate and unless the act is repeated fertilization does not always take place. Usually the ova are fertilized in the upper end of the tube, and segmentation is practically completed before they enter the uterus. According to Bischoff the growth of the dog's ovum is as follows : Diameter of ovum in mm. 0.15 0.14 0.14 0.16 0.16 0.18 0.20 0.21 Age in days 1 2 3 4 5 6 7 8 Diameter of ovum in mm. 0.28 0.30 1.0 2.0 3.0 4.0 5.0 5.0 6.0 Age in days 9 10 11 12 13 14 15 16 16i Twelve days after copulation are required before the ovum is as large as Peters 's and thirteen days before it is as large as Mert- tens's. When it is recalled that Merttens's ovum was scraped out of the uterus twenty-one days after the beginning of the last period we are inclined to believe that sixteen to twenty-one days 186 HUMAN EMBRYOLOGY. represents its true age. Peters 's ovum, on the other hand, must be over "three or four days old"; fifteen is much more nearly its correct age. In determining the age of human embryos it is probably more nearly correct to count from the end of the last period, for all evi- dence points to that time as the most probable at which pregnancy takes place. The group of cases from which His did not subtract twenty-eight days in forming his curve (for instance, Hensen's embryo, which is 4.5 mm. long and was aborted on the twenty-first day) are probably much older than His thought. They belong to those cases in which women menstruated once after becoming pregnant. Having determined the time at which pregnancy probably occurs, it is necessary to fix that at wliich it ends. Not only is it necessary to determine the day but also the probable size of the child, for there is as much variation in the size as in its age. Issmer gives the following figures : Size of child Ko. of cases. Age in days, from the beginning of the last in cm. menstrual period. 48 203 271 49 272 278 50 252 277 51 211 282 52 123 283 53 34 286 54 18 290 The mean length of the child at birth is 49.5 cm. Hecker found it to be 51.2 cm. in 985 cases, and Ahlfeld a little over 50 cm. If a week is allowed to elapse between the beginning of menstruation and conception then the mean new-born child is 271 days old and is 50 cm. long. Having fixed the probable relation between ovulation and men- struation it is next necessary to relate each embryo and fetus first to the first day of the last menstrual period and then correct the same to correspond with the probable time of conception. In order to do this it is necessary to establish some standard measurements of the embryo and, if possible, to determine their deviations when expressed in time. It is known that embryos of other mammals of the same age vary considerably in size unless they are from the same litter, when they are usually very much alike. Undoubtedly there are variations in different animals, and this must be taken into account in comparing human embryos with one another. Also we must not forget that in early abortions there are many pathological specimens, and even if the embryo is normal in appearance patho- AGE OF EMBRYOS AND FETUSES. 187 logical conditions are usually the cause of the abortion. This being true the menstrual periods are also far from normal, and it is not unlikely that ovulation is more irregular than normal in these cases. Thus it is often difficult to determine accurately the last period, for it may be complicated by more or less continuous hemor- rhage. With all these uncertain factors before us it is certainly remarkable that the specimens can be arranged as well as they are, especially their falling into two sharply defined groups during the first two months of pregnancy. Until quite recently no serious attempt has been made to de- termine the age of embryos; it was usually estimated. In order to do this with some precision Arnold measured embryos from head to breech and Toldt from the crown to the soles of the feet (His, 1904). Although this second measurement is called an un- certain one I think that my measurements show that it varies no more than Arnold's (Figs. 145 and 146). These two measure- ments I consider the best that have been proposed. The first — the crown-rump, vertex-breech, or sitting height — and the second — the vertex-heel, crown-heel, or standing height — are standard ones used by anthropologists in measuring the body after birth. In addition to these. His has introduced a measurement for very young embryos from the elevation on top of the back of the head to the breech, the Nackenlinie; but this is of little value in measur- ing older embryos, and easily leads to confusion. In measuring my own specimens, as well as all those I have found suitably pictured in the literature, my attention was called to the neck- breech measurement and its meaning. As it is usually taken it is of value from the time the embryo is well curled upon itself until the neck is fairly well developed, that is, from the fourth to the seventh week. During this period this measurement is the longest, or is as long as any other, that can be made upon the embryo with- out stretching the legs. In later stages it equals practically the length of the vertebral column. In order to make satisfactory measurements upon the bodies of young embryos it is necessary to measure them from more fixed points than is usually done. According to the position of the head the upper end of the longest measurement of an embryo may fall over any portion of the brain, and from a study of numerous speci- mens I find that the middle of the mid-brain is usually just below the highest point of the head; but whenever this is not the case, as it is found to be in young embryos, I think the measurement should still be taken from a point immediately over the mid-brain, as is shown in Fig. 141, C. The other point which I suggest as a desir- able one to measure from lies in the mid-dorsal region just above the first cervical nerve, as shown in Figs. 141 and 142, which have the outlines of this nerve drawn in. In Figs. 143 and 144 this point 188 HUMAN EMBRYOLOGY. is marked by passing a straight line from tlie middle of the lens through the external auditory meatus to the back of the head. In both of these specimens this line passes between the atlas and the occipital bone. This gives an upper point, between the skull and the vertebral column, which is below the one from which His drew his Nackenlinie and above the depression in the neck from which a number of embr^i'ologists make their neck-breech measurements. I have found from numerous measurements of embryos, fetuses, infants, and adults that a line drawn from the middle of Fig. 141. -Embryo No. 163, X 10 diameters. C, crown immediately over the mid-brain; R, rump; A^ point between the occipital bone and the first vertebrae; ec, eye-ear line. the eye through the middle of the ear and extended to the back of the neck always passes just below the foramen magnum, or slightly higher. For practical purposes it cuts the skull from the body, and according to our knowledge of the position of the eye and ear this should be the case. This line, which I have termed the oculo- auricular, or eye-ear line, is of fundamental importance in measur- ing the length of the spinal column in embryos. Anthropologists obtain the same point between the skull and vertebral column by extending the plane between the two rows of teeth to the back of the head; while art anatomists determine it by projecting a hori- zontal line through the nasal spine, just below the nares, to the AGE OF EMBRYOS AND FETUSES. 189 back of the head; in both cases the skull is cut off. All three of the lines meet in the adult at the foramen magnum; but in the embryo only the eye-ear line is of practical use, for it can be determined early and with certainty. The height of the skull, which forms the submodulus in the Fritsch-Schmidt canon, can be obtained in any embryo by measuring the distance at right angles from the above-mentioned horizontal line, through the nasal spine, to the crown (Figs. 141-144, C), that is, the point immediately over the mid-brain. Fig. 142. — Embryo No. 144, X 7 diameters. Letters as in Fig. 141. H, heel; h, hip-joint; K, knee- joint; X, point in leg which equals the distance from n to R. By adding xH to CR the standing height of the embryo is obtained. The two upper points from which to measure being fixed just above the atlas and just over the mid-brain, it is necessary to have a lower point in order to measure the length of the head and trunk. All embryologists agree that it be placed at the lowest point of the breech. The line AR approximates the length of the spinal column and the line CR equals the sitting height of the embryo. These two lines mark respectively the atlantosacral and the mesencephalosacral measurements. In Figs. 141 and 142 the point R is exactly below the sacrum, but as the embryo grows longer (Figs. 143 and 144) the ischium gradually recedes; at birth 190 HUMAN EMBRYOLOGY. it is considerably below the level of the sacrum. For practical purposes, therefore, the line from the foramen magnum to the rump, AR, equals the length of the spinal column. In the adult the tip of the sacrum is at the level of the middle of the acetabulum, and this latter point is naturally chosen by Fritsch in the construc- tion of his canon. On account of the high position of the ilium in both the embryo and the fetus, and on account of the close relation between the lower end of the sacrum and the rump in them, I believe it most desirable to measure to the rump and not to the Fig. 143. — Embryo No. 22, X 5 diameters. acetabulum. Furthermore, it makes one of the measurements, AR, equal the length of the spinal column, and the other, CR, the sitting height of the embryo. A comparison of these two lines upon all four figures shows that in all cases they are the longest lines that can be drawn from the mid-brain and atlas to the rump in each case. Furthermore, as the embryos increase in size the angles these form at the rump become more and more acute. In Fig. 141 the crown-rump line falls far in front of the eye ; in Fig. 142 it is just in front, and in Fig. 143 just behind the eye; in Fig. 144 it nearly strikes the ear. AGE OP EMBRYOS AND FETUSES. 191 The entire length of the body, the mesencephalocalcanean line, or the standing height of the embryo, is really the best single measurement of the embryo, for it is the one usually made by obstetricians as well as by anthropologists. It has been said that the standing height of embryos and fetuses is an unsatisfac- tory measurement on account of its uncertainty, but my experience obtained from the measurement of many embryos, Fig. 146, shows that it is no more variable, probably less so, than either the sitting height or that of the spinal column. In Fig. 141 the sitting and Fio 144 Embryo No. 131, natural size. Length of " vertebral column," 68 mm., sitting lieight (crown- rump or vertex-breech length), 90 mm.; standing height (90 + 21 + 23), 134 mm. the standing heights still equal each other, for, as is easily seen, the leg bud cannot be stretched beyond the rump. The other figures show that by extending the legs the standing height becomes greater than the sitting. In each of the figures the hip- and knee- joints and heel are indicated by letters. If a circle is described around the head of the femur, as has been done in the figures, the portion of the length of the leg to be added to' the sitting height in order to obtain the standing height is easily ascertained. _ In Figs. 142 and 143 this amount is only a portion of the leg, while in Fig. 144 it includes most of the thigh and all of the leg. A number of fresh embryos were measured in this way, the legs were then 192 HUMAN EJIBRYOLOGT. straightened and specimens were again measured from crown to heel, and it was found that the two measurements agreed exactly. By this method, then, the standing height of an embryo can be eoMM To 20 30 40 50 60MM. Fig. 145.— Chart giving the standing height (CH), sitting height (C7J), and vertebral column (AR) measuiements of embryos less than 90 mm. long. The abscissas are CR, and the two series of ordinates are C H and AR measurements. Each dot represents two measurements of an embryo. determined without stretching a fresh specimen or injuring a valuable one after it has been hardened. By making a large number of measurements of the human body, Pfitzner has demonstrated that the most constant ratio of AGE OF EMBRYOS AND FETUSES. 100 200 300MM. Fig. 146. — Chart shown in Fig. 145 extended to include all fetuses. The lower row of specimens marked X are all from Burtscher's measurements. any is obtained by dividing the breadth of the head by its length. The mean index for individuals of every year, from birth to old age, is 83 in males and 82 in females. I gather from the figures of embryos and fetuses published by Eetzius that in all months of Vol. I.— 13 194 HU.MAX EMBRYOLOGY. uterine life the index is the same as after birth, for in the indi- vidual cases given it ranges between 80 and 85. Were it possible to apply these measurements to all fetuses, I think either the length or the breadth of the head would prove the best standard, and all other measurements could be adjusted to it as the art anatomist has adjusted all proportions to the submodulus. That other measurements are required of the body of the embryo than those that are usually made, including that of the entire length of the body, is indicated by various writers, including His, who was the first to use the Nackenlinie. More recently he employed a new measurement which he calls Kopftiefe, and which he says corre- sponds about to the height of the head, measured from the chin to the crown. The Kopfldnge is the length of the head, a measurement which can easily be obtained if this part of the embryo is not dis- torted. The point between the occipital bone and atlas having been determined, as is done by the eye-ear line, a second line may be introduced connecting the spine of the nose with it. The longest line within the head of the embryo parallel with this measures the length of the head, and a line at right angles to it extending to the crown measures the height of the head. Thus it is seen that it is possible to make some of the ordinary head measurements of the adult upon the head of the embryo. It may be that the submodulus of Fritsch- Schmidt may yet prove to be the standard measurement in human embryology, comparing all of the other measurements of the body with it, as is the case in the Fritsch-Schmidt canon of the adult. However, this possibility appears to be remote. It seems to me that for the present we must continue to employ the sitting height or the crown-rump measurement as the standard. Next in importance is the standing height, and, judging by the form of a curve made by abscissas and ordinates to determine the means by the graphic method, I do not find that one is more vari- able than the other (Figs. 145 and 146). The sitting height is the measurement most easily, and, therefore, the one usually made upon young specimens, and the standing height upon older ones. These two measurements can be compared directly with the two standard measurements made after birth. By means of the eye- ear line the point between the head and neck can be marked and from it the length of the head, and of the skull, may be obtained. That the standing height is just as good a measurement as the sit- ting height is further found by the experience of Pfitzner, who was at first opposed to it, but after having made many more measure- ments he selected it as the best standard measurement with which to compare all others. This last statement is based upon the careful measurements of 5000 cadavers. All the measurements that I have been able to collect from the literature, by correspondence, and from my own specimens, are AGE OP EMBRYOS AND FETUSES. 195 given in the two curves (Figs. 145 and 146)/ These were tabulated with the crown-rump measurements as abscissas and the standing heights as ordinates. In the embryos and smaller fetuses a second set of ordinates gives the length of the vertebral column, and it is seen that the deviations here are quite marked. The rows of dots were then divided by curves which included half of the cases, leaving one-quarter on one side and the other quarter on the other. The dots which fell between the two lines mark probable deviations and a line drawn midway between them gives the probable mean. By this method a probable mean is determined in a graphic way from a relatively small number of cases. From the two curves the means of all the measurements in any specimen may be obtained at a glance. This is necessary, for the age of embryos with the standing height given had to be compared with those in which the sitting height is given. In the embryos with a CR measurement less than 13 mm. long there is considerable deviation on account of the irregularity of these young specimens, their smallness, and the great probable error in making the CR and AR measurements. In embryos 13 mm. long the legs begin to grow and the CR and CH measurements form a very even curve, but the deviation of the AR measurement is very marked, showing that it is not altogether satisfactory. Having remeasured in three directions after a uniform plan all the embryos I could collect it is now possible to tabulate them in relation to the menstrual history, and the curve is by no means as satisfactory as I had hoped it to be. However, it must stand for the present, and new and much better material is needed before it can be revised. Even if we should limit ourselves to specimens got from mechanical abortions, operations, and autopsies we must still reckon on 6 per cent, of abnormalities, which are present in all pregnancies. The cases given in the two curves have been shifted and tested, and again and again controlled by the curves of Hecker, Ahlfeld, Toldt, His, Issmer, and Michaelis, and it seems to me that they are the best that can be done with the data at hand. Towards the end of pregnancy I have allowed Ahlfeld 's and Issmer 's data to influence my curve a little ; for a number of the measurements of my older fetuses came from negroes, and their statements are 'From the literature: Retzius, Merkel, Burtscher, Sommerring, Ecker, Kolliker, O. Schultze, Kollmann, Heisler, Minot, His, Frederic, Fraser, Keibel, Bade, Bonnet, Fiersol, Rabl, Tandler, Merttens, Reichert, Peters, Graf Spee, Frassi, fitemod, Thomson, Hensen, Janosik, Meyer, Stubenrauch, and Wagner. Through eorrespondence : From Professors Graf Spee, Laguesse, Hasse, Robin- son, Edwards, Hrdlieka, Streeter, Jackson, Bruner, Lee, Meyer, Waldeyer, Braehet, Keibel, Gage, Thomson, Austrian and Mandelbaum. Together there are over 1000 measurements, of which over 500 have data relating to the age. Fully half of both these are from my own collection. 196 HUMAN EMBRYOLOGY. not as reliable as they might be. At the beginning of the curve I have deviated considerably from His for reasons given above. Toldt's curve is largely an opinion, as he states in his article. The other specimens from the latter months of pregnancy (Issmer, Ahlfeld, and Michaelis) are from exact data and are very rehable. In transferring Michaelis 's means I placed it in the middle of the month and not at the end. The His curve is constructed from measurements taken from his " Normentaf el, " and the higher line gives his Nackenlinie. In all other cases the CH measurement is given as soon as the legs begin to develop. The great amount of scattering of early specimens, as shown in Fig 147, is due no doubt in part to an arrest of development, on the one hand, and continued menstruation after pregnancy, on the other. In order to get any kind of agreement His, in the con- struction of his curve of growth, deducted about twenty-eight days from the age of many specimens in order to make them agree with the rest. However, his curve which I have introduced is an irregu- lar one, unlike the probable curve obtained by tabulating any growing organic body. Days between the beginning Only possible time of Length of embryo of the last copulation between the last in mm menstruation menstruation and the and the abortion. abortion. Embryo anlage, 0.15 38 Exactly 16 days Bryce-Teacher's monograph, 1908. Eeichert. Ovum, 5.5X3.3 42 20 days before and earlier Embryo anlage, 1.3. 34 Exactly 21 days Eternod : Anat. Anz., vol. XV, 1899. Embryo, 3.2 48 40 days before and later His : A. M. E., II. 50 49 38 days before and later 39 and 41 days before 45 days and later KoUmann's Atlas. No. 208. Embryo, 7J 57 His. Embryo, 8.8 42 Exactly 38 days Tandler : Anat. Anz., vol. xxxi, 1907. Embryo, 10 60 49 days and earlier His. Embryo, 11 55 31 days (?) Eabl : Entwickl. d. Gesch. Embryo, 13.6 63 53 days and later His. Embryo, 14 65 Exactly 47 days Rabl. Embryo, 30 75 Exactly 56 days No. 26. There are a few cases in which the time of copulation as well as that of menstruation is given in the history of young embryos. These I have brought together in the above table, and I have also entered them with a * in Fig. 147. That is, the age of the embryo, as rated by the only copulation between the menstruation and the abortion, is also given. From this it is evident that the most probable time of conception is during the first week after the menstrual period, as advocated by Hensen and most obstetricians. B H! AGE OF EMBRYOS AND FETUSES. 197 3 '^r^ S 4 ig the specimens wit *ments. The embryt (CH) was aborted fif \ V\3= is-B 2 \ r \ . .^o >■ ... . \ \ n \.\ • o '-^ 9 HI V. <\ o X > 6 9 1 3 less than 80 mm. long (CH). be traced back to a single cop pulation and seventy-five days •• • V ^v\ • \ \ \» • • • • e • \ N^ ^ N^ ,\ \ •• <. > "^^ 1 The curve Illation arc after the • • • X >- < •\^ ^ ^ 1 1 s of His, T inserted beginning \ • "n^. ^ 2 1 oldt, and : a second t of the las \ \ ^^ X. B^g- S =! S T- • • \ a CL £— CD 5:7 n> W CO ft ^33 ^9 O as o o e •a o 198 HlfMAN EMBRYOLOGY. This view is supported by all the evidence, including that obtained by the study of early human embryos. Among civilized races copulation does not take place during the menstrual period, and it is believed that it is most likely to be followed by pregnancy if it occurs immediately after the period. Furthermore, competent authorities recommend that women who are anxious to become pregnant should copulate before the period is fully over. If it is true that ovulation takes place towards the end of the period the ovum is most likely to become fertilized if it is met by a host of spermatozoa in the tube. After the sper- matozoa have passed through the tube the probability of fertiliza- tion is reduced. In such cases pregnancy should occur within twenty- four hours after copulation. If the sperm awaits the ovum, as is the case when the copulation takes place some time before the period, the probability of fertilization is greatly reduced, and if it occurs it is only after the lapse of a number of days. This accounts for the discrepancies given by Issmer. That conditions regarding the relation of fertilization to menstruation and to the cestrous cycle are identical is further proved by the habit of American negroes, who, I am informed by Professor Williams, prefer to copulate dur- ing the menstrual period. At this time the odors of the negress are said to excite the passion of the negro and are very attractive to him. The following table gives the probable deviation of the length of embryos and of the menstrual age, that is, the age as computed from the first day of menstruation. I have also included in it the figures given by Michaelis. Since he gives the mean for each month I have given his data for the last day of the second week, for, as I take it, his mean measurements apply to the middle of the month. The probable age which is the basis for this table is given in the form of a curve in Fig. 148. It may be noted that its form during the first month (Fig. 147) is somewhat drawn out and does not correspond any too well with the curve during the remaining nine months of pregnancy. However, in embryos up to 10 mm. long the CR measurement is less than that of the spinal column, while later on it exceeds it. In fact it is the diameter of a circle in very yoimg embryos, while later on it is half of the circumference. This Toldt attempted to correct. The His curve is very irregular in form and for this reason, if for no other, cannot be correct. It falls between Toldt's and mine. AGE OF EMBRYOS AND FETUSES. 199 k Probable Measurement Measurements 1 g a deviation S~ Probable of embryo of embryo > s > of G deviation iCH). iCB). Si menstrual ^^ (CH).* According to ^ According to "0 O) age.* 1 Michaelis.t 3 Michaelis.t u a 1 1 ., decidua; 5., syncytium; y., villus; Ch. chorion; .1/., mucoid substance between the villi rich in leucocytes. that the percentage is nearly the same in each collection when the figures are arranged in parallel columns. I always considered it remarkable that His never observed an ovum without an embryo, Fig. 172. — Photograph of an embryo 11 mm. long, attached to the chorion. for 28 per cent, of my specimens are of that kind. However, many of them are bloody or fleshy moles, while others are ova which appeared to be perfectly normal until they had been cut into serial sections. I have also compared my vesicular forms with His's nodular forms, for no doubt they are the same in most cases. PATHOLOGY OF THE OVUM. 221 Classification. Ova without amnion or embryo Ova with amnion but without embryo Vesicular (nodular) forms Embryos 2-2.5 mm. long Embryos 2.5-4.5 mm. long Embryos 5-8 mm. long Embryos 9-14 mm. long *This number includes some of my specimens of the seventh week. His. Number. 5 9 10 11 12 Per cent. 53.4 24.3 22.3 Mall. Number. 29 15 18 J 21 26* Per cent. 46.6 23.9 29.5 i.e., all embryos less than 15 mm. long. His did not cut sections of the nodular forms, and had he done so he would probably have found them, as I did, often composed of a single umbilical vesicle without an embryo, which was frequently not attached to the chorion. The large percentage (12) of vesicular forms in my collection is probably due to the Pig. 173.— Section of the embryo shown in Fig. 172. Most of the head is destroyed, but the face is adherent to the chest below. refined method I have employed in examining these specimens, most of them having been cut into serial sections. Giacomini notes especially that it is unnecessary to make a group to include the vesicular or cystic forms of pathological ova, for they may be scattered under the various headings of his classifi- cation. Under this Group I include those in which the entire 222 HUMAN EMBRYOLOGY. embryo has been destroyed, leaving only the umbilical vesicle and sometimes a portion of the amnion. The remnants of the embryos of this group correspond well with Panum's II, 2, in which the embryo is destroyed but the area vasculosa remains and gives rise to blood, just as the vesicle in this form of human monster is composed mainly of an umbilical vesicle with its primary blood- vessels. It may be noted that some of these umbilical vesicles have been confused with the amnion in Giacomini's fantastic group, in which the embryo has wandered out. It is further stated by Giacomini that the openings through which the embryo wan- dered often healed up, for they could not be found. Fig. 174. — Section of the deoidua, villi, and chorion of the specimen shown in Fig. 172. There is a mass of mucus between the villi which contains many leucocytes. In the nest table I have arranged nearly all of my pathological ova under various headings, omitting only a few embryos over nine weeks old. Geoxjp I. — In the first group are the vesicular forms in which the main remnant of the embryonic mass is composed of the umbilical vesicle (Figs. 176a-178). In some of them the amnion is formed and in others it is destroyed entirely. Group II. — In the second group there is neither amnion, em- bryo, nor umbilical vesicle ; only the chorion remains. This group must have formed from that variety of Group I in which there is no amnion present. Vesicular and solid moles may arise from this group. PATHOLOGY OF THE OVUM. 223 Group III. — In this group the embryo was destroyed after the amnion had been formed; usually it lines the chorion. All Fig. 175. — Section through the tips of the villi of an ovum 24 mm. in diameter. The embryo within, 2 mm. long, is deformed but nearly normal. S., syncytium; V., villus; F., fragmented nuclei; 71. S., necrotic syncytium. _,37--Af. Ch. Fig. 176. Mucoid mass between the villi and the chorion from the same specimen pictured in Fig. 175. Ch., chorion; V., villus; M., mucoid mass rich in leucocytes and containing nests of syncytium. stages of the complete destruction of the embryo are found in this group, from a necrotic, granular mass to a vesicular ovum lined by the amnion with but a very short stump of the umbilical cord left (Fig. 179). 224 HUMAN EMBRYOLOGY. IHstribution of 159 Pathological Ova in a Collection of 434 Specimens, giving also the Condition of the Chorion. CE length in mm. Number. Per cent. Condition of the chorion. Normal. Path- ological. No record. 2J 5i 8 11 17 25 32 43 19 29 15 4 18 21 13 27 10 2 1 12 18 10 3 11 13 8 17 6 1.4 .6 6 6 3 1 1 2 11 22 11 3 11 15 10 23 4 2 1 2 II. Ova with neither amnion nor 1 III. Ova with amnion but without embrvo 1 IV. Embryos of the 4th week Embryos of the 5th week Embryos of the SJth week Embryos of the 6th week Embryos of the 7th week Embryos of the 8th week Embryos of the 9th week Embryos of the 10th week 1 6 6 3 3 4 Total 159 100.0 19 113 27 GrROUP IV. — The embryo is present in this group and is more or less degenerated. In case it is much degenerated it may pro- duce a nodular embryo of His or an amorphous embryo of Panum. Usually after the fifth week it is quite easy to recognize the stage in which the embryo became pathological. The younger ones correspond with His's abortive, atrophic, or degenerated forms, Fig. 176a.— 0\rum 36 X 33 X 13 mm. without villi. A piece of the chorion is turned over to show a small nodule, the remnant of the embryo. Fig. 177. — Ovum measuring 12 X 9 X 5 mm. with ragged villi and a large vesicle within. the older ones often with his cylindrical forms (Figs. 180-183). I have found it more convenient to arrange them in weeks accord- ing to the age of the embryo at the time the pathological process began. The embryos of any given week may contain any of His's atrophic forms according to the extent, degree, and duration of the pathological process. It is noteworthy that there are so few path- PATHOLOGY OF THE OVUM. 225 ological embryos of the fourth week in my collection, while relatively there are four times as many in His's collection. Just the opposite is the case with the vesicular or nodular forms. It Fig. 178.— Section through the vesicle within the ovum shown in Fig. 177. The double vesicle measures 5.5 X 3.3 mm The smaUer vesicle between the larger one and the chorion may be the amni- otic cavity or possibly the dilated allantois, although it is not attached to the chorion. may be that I have had a tendency to class with these embryos those that he classes with the nodular form. The vesicular forms are intermediate between ova without embryos and ova with path- ological embryos of the fourth and fifth weeks. Fig. 179. — Umbilical cord without an erabi-yo in an ovum of 55 X 40 X 40 mm. The largest number of pathological embryos are formed dur- ing the first seven weeks of pregnancy; their number falls off markedly in the eighth and ninth weeks ; and but very few occur Vol. I.— 15 22G HUMAN EMBRYOLOGY. Fig. 180. — Photograph of a pathological embryo 6 mm. long. Fig. 182. — Sagittal section of a much deformed embryo 6 mm. long. The dissociation of the tissues is nearly complete. Fig. 181. — Section through the embryo shown in Fig. 180. Fig. 183. — Sagittal section of a dissociated embryo 9 mm. long. after the tenth week. Pathological embryos that survive the second month will probably continue through the normal period of pregnancy and give birth to monsters. From statistics given above, this should be the case in every twelfth pathological ovum. PATHOLOGY OF THE OVUM. 227 It may also be suggested, with Giacomini, that threatened abortion in early pregnancy should be encouraged, for the cause of it is probably a pathological ovum and the uterus should be relieved of it. Careful investigations should also be made in these cases regarding the cause of the primary trouble. Sterility, or tendency towards sterility of women, especially if it is acquired, should be studied much more carefully than it has been for the sake of scientific teratology and the scientific treatment of abortion. The study of pathological ova has shown that the embryos within are deformed and that there are structural changes in the chorion which appear to be associated with inflammatory processes in the uterus. The villi are usually fibrous or are otherwise de- generated, the syncytium is atrophic or necrotic, and there is an Fig. 184. — Embryo 3% mm. long from a tubal pregnancy. excess of blood and mucus rich in leucocytes between the villi. These are also often invaded by syncytial cells and leucocytes. The picture indicates that the chorion is affected by an inflamed uterus, which naturally interferes with its nutrition. It is prob- able, however, that the process is somewhat more complicated, for the trouble often seems to lie within the decidua, especially in tubal pregnancies, which nearly always contain pathologcal ova (Fig. 184). In such cases the inflammatory process around ttie chorion is not so marked, but the decidua is deficient and there is an excessive amount of blood between the villi (Fig. 170). In both cases the nutrition of the ovum is affected, in the uterus by inflammatory and in the tube by hemorrhagic processes, which interfere with its implantation. As a result of faulty implanta- tion the chorion degenerates or its further growth is retarded and the embryo suffers and becomes atrophic. On the other hand, it is also possible to view the change in the chorion as secondary, as a result of primary changes in the embryo which are germinal in origin. In fact this view of them is entertained by many pathologists, who would consider the ovum 228 HUMAN E.MBRYOLOGT. as a foreign body after the death of the embryo, and all of the inflammatory changes found within it as of a secondary origin, as would be produced by a sponge if it were put in its place. This second attitude, which considers the changes within the embryo and in the chorion as a coincidence, is, I believe, incorrect. To be sure, it is well known that a woman who aborts a pathological ovum or gives birth to a monster is more likely to do so again, and this is the great argument in favor of the theory that the primary trouble lies in the germ and not in its environment. How- ever, if pathological ova and monsters are due to a diseased condi- tion in the uterus which interferes with the implantation of the ovum, this fact speaks equally as well for the environmental as it does for the germinal theory. The facts bearing upon these two theories I shall give briefly and in the order of their value. 1. It is shown in the table given in the beginning of this chap- ter that of all pregnancies 7 per cent, end in pathological ova. In case the pathological condition is present in either the germ or sperm that same percentage of pathological ova should be found in ectopic pregnancies. I have taken considerable trouble to investi- gate the evidence obtained from tubal pregnancies, and in general find stated in the literature that more deformed embryos are found in them than should be; biit this statement is rather an opinion than a demonstration based upon actual records. The answer to the question is complicated and rendered difficult by the rupture of the tube, which is of frequent occurrence, through which the embryo is easily lost in case it was present. I have col- lected all the cases of unruptured tubal pregnancies from Dr. Kelly's gynaecological laboratory and find that there are forty-six in which the tube has been examined by refined modern methods. The enlargement in the tube containing the ovum measured in these specimens from 1 to 6 cm. in diameter, and in thirty-nine of them remnants of the chorion were found with villi ramifying through the blood-clot. Five contained pathologica,l embryos, and but two contained normal embryos which were of the second month. In these two the chorion was well implanted, having a well-formed decidua, as is usually the case in the uterus. In all the rest the villi were of irregular shape, usually atrophic and degenerate, sometimes very long and thin with much blood between them, and the decidua was irregular and scanty. In these the implantation was faulty, and as a result 96 per cent, instead of 7 per cent, became pathological or produced monsters. This is the strongest argument against the germinal theory. 2. Von Winckel has done us a great service in collecting the data regarding the condition of live fetuses which had been re- moved from ruptured ectopic pregnancies by surgical operations. PATHOLOGY OF THE OVUM. 229 They, of course, were derived from the 4 per cent, of normal embryos mentioned above, for the pathological changes in the 96 per cent, were so radical that they could not develop into fetuses of any kind. Von Winckel's specimens are especially valuable, inasmuch as they show the possible fate of the normal embryos I found in tubal pregnancies obtained from Dr. Kelly's clinic. Forty-seven out of eighty-seven fetuses were much deformed and twelve were markedly monstrous; but eight were really normal. Among the monsters there were six specimens of hydrocephalus, one each of hydromeningocele, spina bifida, encephalocele, anen- cephalus, omphalocele, and hypospadia. In addition, the head was deformed fifty-seven times, legs forty-four, and arms thirty-five, with club-feet in twelve and amniotic bands in four cases. The placenta was usually deformed, sometimes multiple and sometimes broad, thin, or short, and often very hemorrhagic. In general, the poles of the body suffer most, the head being deformed in 75 per cent., legs in 50 per cent., arms in 40 per cent., and the trunk in 4 per cent, of the cases. It is clear that the difficulty is due largely to ordinary mechanical causes which interfere with the growth of the placenta and the poles of the embryo and frequently produce typical monsters. From the data given, it is seen that but very few of the embryos in tubal pregnancies produce normal individuals. 3. Comparative experimental teratology has shown us that all varieties of monsters found in man can be produced in large numbers from normal ova after fertilization as well as from normal embryos. I can only enumerate the results, the literature and general discussion having been given in my larger mono- graph, as well as in Hertwig's "Handbuch." a. Polysomatous monsters can be produced by a variety of mechanical methods from normal eggs. Thus Vejdowsky showed in 1892 that the number of monsters produced from Lumhricus eggs was greater in the summer months than in cool weather ; and somewhat later Driesch succeeded in producing monsters from sea-urchin eggs by separating the cells in the two-cell stage by mechanical means, or by increasing their temperature, which acted upon them in a similar way. Somewhat later Loeb pro- duced double monsters from sea-urchin eggs by changing the chemical composition of their surrounding sea water. In case it is diluted its rapid absorption by the egg causes the cell mem- brane to rupture, through which some of the protoplasm often escapes; upon returning the egg to normal sea water segmenta- tion begins, nuclei wander out into the extruded protoplasm, and a double monster develops. These important discoveries were next extended to verte- brates by Wilson, who experimented upon Amphioxus; and then 230 HUMAN EMBRYOLOGY. by 0. Sehultze, who experimented upon frog eggs. Botli terato- logists used mechanical means to produce double monsters. Wil- son shook his eggs after segmentation to form hour-glass shaped eggs, and Sehultze fixed the eggs between two glass slides and inverted them after segmentation had begun. The partly sepa- rated blastomeres gave the anlagen for the bodies of the two embryos, and recently Spemann has produced double monsters in the frog by tying its eggs with a fine thread at the right time. Furthermore, Tornier has produced double legs or even clusters of legs from the single anlage of the leg. These experiments all show that polysomatous monsters are produced from normal eggs. b. Another variety of monster, not well developed and poly- somatous, but atrophic and merosomatous, is found in lithium larvae. In 1893 Herbst found that there was often an inversion of the blastodermic membranes in case developing sea-urchin eggs were subjected to the action of lithium salts. Morgan ex- tended these experiments to frog eggs and found that the inversion of the layers was due to a failure of the upper protoplasmic contents of the egg to move downward, and he concludes that this arrest is due to the physical and chemical action of the lithium. These monsters are similar in appearance to the irregular nodular forms produced by Panum, Dareste, and Fere, and are interesting inasmuch as they show that the action of lithium upon a normal egg is specific; the lithium produced a definite action upon the egg, interfering with its internal growth and also with its nutrition. c. It has also been shown by Loeb that the action of calcium salts upon eggs has a specific action upon the growth of the heart and blood-vessels, by preventing the heart beat and retarding the growth of the blood-vessels, as well as of the embryo in general. Although Loeb states expressly that the action of the calcium is specific, as the rest of the embryo remains normal, I am inclined to believe him in error regarding this point, because he did not examine his specimens microscopically and because Knower has recently shown that mechanical enucleation of the heart in young embryos is followed by the gravest consequences. In such em- bryos the pronephros becomes cedematous and the lymph- and blood-vessels and body cavities become distended. There is a general arrest of development of the embryo; the coils of the intestine are atrophic and there is histolysis of the mesentery and vacuolation of the muscle cells. Teratologists recognized long ago that the heart must be affected more or less in monsters, on account of the frequent occurrence of an oedematous condition of the tis- sues, as well as of accumulation of fluid in the serous cavities and of the hydramnios and hydrocephalus. Such conditions are often seen in pathological embryos, as well as in the monstrous chicks which were produced experimentally by Panum and Dareste. PATHOLOGY OF THE OVUM. 231 d. In 1892 Ilertwig published his remarkable essay on spina bifida, which is of far-reaching importance. However, it was Morgan who discovered that spina bifida may be produced experi- mentally by subjecting frog eggs to the action of common salt. It was found that a 0.6 per cent, solution delays the development of the egg (the chorda, intestine, myotomes, and nervous system developing normally), but gastrulation is postponed for from twelve to twenty-four hours. Posterior spina bifida naturally results. Later in development the exposed cord undergoes cytolysis and histolysis. Subsequently Hertwig extended Mor- gan's sodium experiment to axolotl. Here the reaction is sharper than in the frog and there is also often anencephaly. It was found that a 0.5 per cent, solution had no effect upon the embryo at all, a 0.6 per cent, solution made half of them monsters, and in a 0.7 per cent, solution all of them developed spina bifida. Schaper has removed the brains of tadpoles mechanically and Harrison has done the same with the spinal cord. In these experiments the embryo grows normally without spinal nerves or cord unless the operation destroys the lymph hearts also ; then dropsy follows. In fact this seems to be always the case when the heart is involved by either mechanical or chemical means. e. The great precision by which spina bifida is produced by the action of sodium salts is equalled in a more striking manner by Stockard's magnesium experiments, in which typical cyclopia is produced in 50 per cent, of the fishes (Fundvlus) experimented upon. Teratologists have speculated upon the cause and the de- velopment of cyclopia for centuries, and now with one stroke all is clear. Ten years ago Born occasionally produced cyclopia by splitting the head of the embryo through its sagittal plane. Later Spemann produced the same by ligature of the head, and Leyj" by cutting off the front tip of the head. Harrison also often pro- duced a new variety of cyclopia by removing the brain of the embryo; the eyes then wandered to the back of the head in the region of the pineal eye and appeared to unite. By a very dif- ferent method Lewis succeeded in producing cyclopia in a large percentage of the specimens experimented upon. He pricked the extreme end of the embryonic shield of Fundulus and from such eggs embryos with all degrees of typical cyclopia developed. Of course, the striking experiment is Stockard's, and recently he has given an account of the anatomy of his embryos. At any rate, all these experiments show that all kinds of monsters, including spina bifida and cyclopia, are produced from normal embryos due to external influences. 4. The consensus of opinion of gyntecologists is that patho- logical ova are due to a diseased uterus, but they are not inclined to associate pathological embryos with monsters. Neither do they 232 HUMAN EMBETOLOGY. speak of curing the uterus of women who have given birth to monsters in order to prevent them from doing so again. How- ever, the evidence I have given above proves that monsters are produced from normal eggs by conditions which either interfere with their nutrition or poison them, and that in tubal pregnancies there is a great excess of pathological ova and monsters. How is it with pathological ova which come from the uterus f Is the uterus usually normal, or pathological! The chorion in nearly all pathological ova examined shows signs of inflammation, often severe, which is, of course, uterine in origin. Taking all of the pathological ova in my collection, thirty-three altogether, in which any data regarding the women from whom they were obtained are given, it is found that they are easily arranged in three groups. (1) In the first group of eleven cases the main trouble preceding the abortion was a severe hemorrhage extending over a number of days. (2) The second group of twelve specimens were abor- tions from first pregnancies in newly married women or relatively sterile women who had been married for some time and were anxious to have children. (3) The third group of ten specimens were from women who had given birth to a number of healthy chil- dren and then began to abort, often a second or a third time. The first group throws no light upon the question we are discussing, but the second is of value because it comes from sterile women. The third group is more easily explained. The women, perfectly healthy at first, gave birth to one or more children and then con- ceived and aborted quite regularly. In these cases the uterus was normal at first, but later, due to a variety of infections, became inflamed, and thereafter the fertilized ovum could not implant itself, became pathological, and was aborted. My records also state that seven of the women are healthy and twelve have uterine disease. In general, those with uterine disease belong to the second and third classes mentioned above. It may be noted that all the pathological conditions of the ova of the third group could not be due to germinal causes, for all these women had given birth to healthy children and the probabilities for any class are but 1 to 14. The data only confirm those obtained from tubal pregnancy, as well as those from experimental teratology, that is, the primary cause is in the environment. 5. Especially interesting are those cases in which two path- ological embryos are obtained from the same woman. Five such sets are in my collection, and in four of them the changes in those of a set are alike. Two sets are duplicate twins and one is com- posed of two twin ova from a woman who had aborted before. The fourth set are about a year apart from a woman who had had nine children, after which her health failed (ten years ago) ; since then she has conceived regularly and aborted every time. The PATHOLOGY OP THE OVUM. 233 chorions of these two specimens are well infiltrated with leuco- cytes, the vilU are largely destroyed, and the changes in the two embryos are severe and much alike. The fifth specimens are from a young woman, mother of two children, and the first of these appeared normal with the exception of an excessive amount of granular magma in the amnion, with leucocytic infiltration of the placenta. Nine months later a second typical pathological embiyo was obtained from the same woman. Disease of the uterus began with the birth of the second child. Later she aborted again. Although these cases do not prove the point made they at least indicate that the same environment affected the ova of either successive or twin pregnancies in the same way. The theory that merosomatous monsters are produced by mechanical influences was established by Lemery, defended by the Saint-Hilaires, but was antagonized to the utmost by Meckel, Bischoff, and others. In case they are produced from normal embryos by means of external influences it follows that the embryo must become wholly or in part diseased or pathological, a view entertained by Morgagni. Frequently the pathological changes in the fetus were compared with those in the adult and it was be- lieved that they were also due to a variety of diseases, such as syphilis, tuberculosis, rickets, or to inflammation. However, it was impossible to show that the destruction of tissue necessary to produce a monster was associated with pathological changes peculiar to these diseases, but instead they nearly always appeared to be normal in character. Panum defended the nosological theory and asserted, with good reason, that only the fundamental characters of the changes within the embryo in a given disease should be like those in the adult. In fact he asserted that the nomenclature for the pathological changes in the embryo cannot be the same as that for the adult, and this opinion is borne out by the numerous investigations during the last fifty years. At best, Panum states, the etiological factors are the same for diseases of the embryo and of the adult. He had found in experimental chick, monsters due to malnutrition that there are constant tissue changes, such as will produce exudates, bring about adhesions, and cause atrophy with scar formation. And since these changes are found constantly it indicates that they are due to a common path- ological cause. Local softening and necrosis, which often accom- pany the above-mentioned processes, are of sufficient importance to account for the changes in development, which is otherwise normal, to produce spina bifida and the like. These changes, which often take place in the embryo before the blood-vessels are formed, may be likened to those accompanying inflammation in the adult. However, there is a multiplication of cells as well as a cell necrosis, and Panum thinks himself justified in calling the process parenchy- matous inflammation of the embryo. 234 HUMAN EMBRYOLOGY. The tissue changes found by Panum in experimental chick monsters were subsequently seen and recognized by Giacomini, His, and myself in the human embryo, and are well described by Giacomini in his general article. By means of serial sections of pathological embryos it is easily seen that the sharp normal lines of demarcation of the structures are largely lost in young embryos ; and in older embryos the elements of difPerent organs become more or less mixed, which often gives them the appearance of lymph-glands. However, certain tissues like the ectodermal are more resistant than others. The changes in older embryos were described with great accuracy by His. He stated that the blood- vessels of pathological embryos enlarge and become gorged with Fig. 185. — Section through the head, marked in the adjoining diagram, showing cytolysia and dissociation of the tissues in an embryo 2 mm. long. M., hind-brain; P,, epithelial hning of the pharynx. blood, and that many of the cells wander into the surrounding tissues, thus converting the whole embryo into an even structure, as described by Giacomini. It seems to me, in view of what has been said above, as well as by the results of Born upon grafted embryos, and of Hertwig, Morgan, and others upon numerous experimental monsters, that we are dealing with a condition in which there is more or less correlation of growth, which may represent a fundamental type of inflammation. When, however, the embryonic tissues become mixed, which is generally due to malnutrition followed by some cytolysis, we have a new condition quite unlike any pathological change found in the adult. The repair of a simple wound in the embryo is always associated with further development of the sur- rounding parts, and in case the process ends in a perfect result normal development still remains, with or without regeneration; PATHOLOGY OF THE OVUM. 235 Fig. 186. — Embryo 2 mm. long, with an atrophic head, large neu- ropore, and spina bifida. but if tliere is a lack of correlation, pathological conditions arise, already recognized by Morgagni, and well described by Panum, Giacomini, and His. This pathological condition I shall term dissociation. The growth of dissociated tissues may be checked by excesssive cytolysis or they may be de- stroyed entirely by histolysis. There are several young embryos in my collection in which dissociation is just beginning. One of them, 2 mm. long, which is practically normal in form, is from an ovum which was curetted from the uterus and included part of the de- cidua (Fig. 185). The decidua is infil- trated with leucocytes and the coelom has in it an excess of magma reticule; other- wise the chorion appears normal. The front end of the amnion is torn and well packed in with magma, showing that it too is not of mechanical origin. In general, this specimen shows that in young path- ological ova the embryo is extremely susceptible and about the first to suffer. In this specimen the mesodermal tissue of the em1)ryo and the ventricles of the fore-brain are filled with round cells containing fragmented nu- clei. Most of the blood-corpuscles are still within the blood-vessels, and those that are free in the tissues are well defined and per- fectly normal in appearance. However, it may be noted that the. cells of the mesenchyme di- minish as the free round cells in- crease, showing that they are dis- sociating, as are also those of the brain tube. Another specimen, somewhat more advanced in de- velopment, has an atrophic head and a wide open spinal cord be- low, anencephaly and spina bifida (Fig. 186). In this there are slight signs of its pathological nature in the chorion, and there is an excessive amount of magma in the coelom. So far there is no dissociation of the tissues of the embryo; there is only an arrest of development of the central nervous system. As the pathological process continues in later Lx ,,- ^ \, \ i ^^ li^ r yF Fig. 187. — Embryo 35 mm. long, with de- formed hands and feet. The largest diameter o£ the ovum la 90 mm. The abortion took place 18 weeks after the last menstrual period. 236 HU.MAN EMBRYOLOGY. embryos of the fourth week, the amnion is often destroyed and the embryo rapidly degenerates, usually leaving the umbiHcal vesicle, which is quite resistant. Subsequently this is also de- stroyed, leaving only the chorion, without amnion or embryo, which may continue to grow into an irregular mole. The embryo gradually becomes more and more resistant during the fifth week, the brain and heart showing somewhat greater resistance than the other organs. Between the fifth and sixth weeks, when the peripheral nervous system appears, the Fig. 188. — Embryo 19 mm. long, with dropsical back. delineation of the organs becomes sharper. Here we also often find dissociation of one or more of the tissues or organs, a gorged vascular system, and frequent hydrocephalus, all probably due to an arrest of the heart action. Such embryos rapidly undergo secondary changes and within a month most of them abort, or if the chorion remains it may form the nucleus of a mole. During the sixth week, owing, no doubt, to the unequal dif- ferentiation of the tissues, some of them become more resistant than others. The more central tissues stand the action better and the more peripheral tissues are more susceptible. Thus the spinal cord and medulla do not dissociate and atrophy so easily PATHOLOGY OF THE OVUM. 237 as do the head, face, brain, and ex- tremities. The vascular system also suffers very much, probably on account of the effect of the impaired nutrition of the chorion upon the heart. As a result of the weakened heart the cavi- ties of the brain and body become dropsical, and the tissues of the ex- treme ends of the embryo dissociate and develop poorly. The precartilages and cartilages suffer least of all. In the beginning of the seventh week the cartilages of the extremities are outlined, and at the end of this week ossification centres make their ap- pearance. Coincidently the peripheral nerves ramify through the body and the muscle anlagen appear. On account of the high degree of differentiation of the structures of the embryo, impair- ment of its nutrition produces very unequal effects upon its organs and tissues. In the earliest stages the umbilical vesicle is the most resist- ant, then the nervous system, and now it is the skeletal tissues. Before the development of the heart, the blood- vessels were very resistant; now that they are dependent upon the heart they are least resistant, and structures which are dependent upon the circulation for their nutrition suffer in a secondary way. The changes in embryos of the seventh week can be followed easier than those in earlier embryos, for they are less rapid and the differentiation of the structures aids the observer very much. Now the extreme ends of the embryo are pro- foundly changed (Fig. 187), due probably to affections of the heart of the embryo ; there is much cytolysis and dissociation of the nervous system ; and the face, head, and extremities are often atrophic. To- wards the eighth week there is a great Fig. 190.— Reconstruction of the diminution of the uumbcr of pathological central nervous system in an embryo . n j_- i j.i 24 mm. long, showing spina bifida, ova lu my collectiou, as was also tlie case destructionofthemedulla.andatro- Jj^ ^^^^ ^f JJig_ J^ foUoWS that mOSt pny oi the brain. Fig. 189. — Sagittal section through the swelling in the back of the embryo pictured in Fig. 188, showing the blister of the epidermis. 238 HUMAN EMBRYOLOGY. monsters are formed before tlie eighth week; those with radical changes in them are aborted, while those that are slightly affected continue to develop until the end of pregnancy. However, many of those that are aborted show considerable growth of a variety of structures, such as the epidermis, which proves conclusively that we are not dealing with post-mortem changes in the embryo. The longer such a specimen remains in the uterus the more radical are the secondary changes in the embryo and the more pronounced are the primary changes in the chorion. In order that a monster shall continue throughout pregnancy the changes in the embryo must not be extreme enough to eliminate the heart, and the chorion must be normal enough to permit the formation of a healthy placenta, which begins to differentiate at the time (end of the second month) monsters cease to form. In the following table I have arranged the data regarding the percentage of the varieties of monsters found in pathological ova, and at birth. In general they agree very well. However, the percentage of spina bitida is greater in the embryo than at birth, indicating that the mortality is greatest in this variety of monster. If the larger number of cases of dropsy of the head were reduced or omitted, the proportion of monsters in pathological ova and those at birth would agree very well. No doubt water on the brain is an affection primarily of later fetal life, but this question remains to be investigated. Percentage of Monsters in Pathological Ova and at Birth. (Pantm). (Von Wikckel). (Mall). Varieties of monsters. 1 S 15 1 75 monsters in 12,378 birtlis. C S S 79 monsters in 163 pathological ova from 434 specimens collected. 1 g B 1 Anencephalus .... Hydrocephalus . . . Hydrocephalocele . 119 26 93 77 16 ■48 Head 23 12 4 9 17 3 7 31 [21 [35 4 9 Atrophic head Malf'd face & neck Displaced eyes Def'd extremities. Spina bifida Exomphalos 24 17 3 18 12 5 31 /Face Cvclonia 1 Eyes missing 9 I'' \Neck ^25 23 15 Def'd upper jaw. . . Def'd extremities. . Spina bifida 3 115 38 J 23 8 f Upper extremity ( Lower extremity Back Abdomen 6 Total 496 100 75 100 79 100 The specific action of salt solution upon amphibian eggs, pro- ducing a large percentage of spina bifida monsters, has been mentioned above. The work of Torneau and Martin, and more recently that of Fischel, has shown that spina bifida in man is not only to be viewed as an arrest of development of the medullary PATHOLOGY OF THE OVUM. 239 Fig. 191. — Embryo 16 mm. long, showing harelip, displaced ears, exom- phaly, and spina bifida. Fig. 192. — Sagittal section of the em- bryo shown in Fig. 191. The central ner- vous system consists largely of a mass of vascular tissue. Fig. 193. Section of the chorionic villi of the specimen shown in Fig. 191. There is an extensive mucoid reticulum between the villi, which contains many leucocytes and syncytial cells. 240 HUMAN EMBRYOLOGY. plate, leaving tke neural tube open, but that there is also a sec- ondary destruction of the niembrana reunions behind, at least in all cases of spina bifida occulta (Figs. 188-190). Deformities of the head, such as anencephaly (Figs. 191-193) and, what may often follow it, cyclopia (Figs. 194 and 195), are now easily understood, since we have the splendid experiments of Stockard and of Lewis upon this question in Fundulus. That the great varieties of dropsy, as pictured by Kollmann and as are frequently seen in embryos and fetuses, are due to an impairment of the action of the heart is now definitely proved by the enucleation experiments of Knower. It is no longer necessary for us to seek for mechanical obstructions which may compress the umbilical cord, such as amniotic bands, for it is now clear that the impairment of nutri- FlG. 194. — Cyclopian embryo, 20 mm. long, with anencephaly. The ovum measm-ed 80 X 60 < SO mm. Fig, 195. — Diagrammatic reconstruction of the Cyclops of Fig. 194, showing the extent of the central nervous system. tion which naturally follows faulty implantation, or the various poisons which may be in a diseased uterus, can do the whole mis- chief. That monsters group themselves, both in nature and when made experimentally, rather shows that certain tissues are in- fluenced at crucial periods in their development, and not that given substances have specific influences upon the embryo as a whole. PATHOLOGY OF THE OVUM. 241 BIBLIOGRAPHY. AiiLFELD: Geburtshilfe, 1903. Die Missbildungen des Mensehen, Leipzig, 1880-1882. Baedebn : Jour, of Experimental Zool., 1907. Ballantyne: Antenatal Pathology, 2 volumes, Edinburgh, 1904. Bischofp: Wagner's Handworterbuch, 1842. Bokn: Roux's Archiv, vol. iv, 1897. Dareste: Recherches sur la production de monstruosites, Paris, 1891. Driesch : Zeit. f . wiss. Zool., vol. Iv, 1892. Eycleshymer: Amer. Jour. Anat., vol. vii, 1907. Eischel: Ziegler's Beitrage, vol. xli, 1907. Eorster: Die Missbildungen des Mensehen, Jena, 1865. GiACOMiNi, Merkel, and Bonnet, Ergebnisse, vol. iv, 1894. Gerlacii: Doppelmissbildungen, 1882. Granville: Graphic Illustrations of Abortion, London, 1834. Harrison: Amer. Jour. Anat., vol. iii, 1904. Hertwig, 0. : Arch, f . mik. Anat., vol. xxxix, 1892 ; vol. xliv, 1895. Gegenbaur Festschrift, vol. ii, 1896. Handbuch d. Vergl. u. Experiment. Entwieklg. d. Wirbeltiere, vol. i. Hegar : Monatsch. f iir Geburtskunde, vol. Ixi, 1863. Hirst and Pieesol: Human Monsters, Philadelphia. His : Anatomie menschl. Embryonen, vol. ii, 1882. Virchow Festschrift, vol. i, 1899. Kelly: Operative Gynaecology, vol. ii, Philadelphia. Knower: Anatomical Record, vol. i, 1907. Koch : Beitrage zur Lehre von spina bifida, Kassel, 1881. Kollmann: Archiv fiir Anat. u. Entwicklgesch., Supplement-Band, 1899. Leopold : Arbeiten aus der dresdener Frauenklinik, vol. iv. Leuckart: De Monstris, Gottingem, 1845. Levy: Roux's Archiv, vol. xx, 1906. Lewis : The Experimental Production of Cyclopia in the Fish Embryo, Anatom. Record, vol. iii, 1908. Loeb : Biological Lectures at Woods Holl, 1893 ; Pfliiger's Archiv, vol. liv, 1893 ; ibid., vol. Iv, 1894; Roux's Archiv, vol. i, 1895; and Studies in General Physi- ology, Chapter X, Chicago, 1905. Mall : Welch Festschrift, Johns Hopkins Hospital Reports, vol. ix, 1900. Vaughan Festschrift, Contributions to Medical Research, Ann Arbor, 1903. Johns Hopkins Hospital Bulletin, 1903. Anatomical Record, January 1, 1907. A Study of the Causes Underlying the Origin of Human Monsters, Jour. Morph., vol. xix, 1908 (contains a full description of all the specimens used in this chapter). Marchand : Missbildungen, Eulenburg's Real-Eneyclopffidia, third edition, 1897, vol. XV. Meckel, J. F. : Handbuch d. pathol. Anatomie, Leipzig, 1812. Morgan : Ten Studies in Roux's Archiv, vols, xv-xix, 1902-1905. Morgan and Tsuda: Quart. Journ. Micr. Sci., N. S., vol. xxxv, 1894. MiJLLER : Meckel's Archiv, 1828. Panum : Entstehung der Missbildungen, 1860. Piersol: Teratology, Ref. Hndbk. Med. Sci., new edition, vol. vii. Von Recklinghausen: Virch. Arch., vol. cv, 1886. Richter: Anat. Anz., vol. iii, 1888. Schaper: Jour. Bost. Soe. Med. Sci., 1898; and Roux's Archiv, vol. vi. ScHULTZB, 0.: Verhandl. d. anat. Gesellsch., 1894; and Roux's Archiv, vol. i, 1895. Vol. I.— 16 242 HU3IAN EMBRYOLOGY. ScHWALBE, E. : Die Morphologie d. Missbildungen des Menselien und der Thiere, Jena, part i, 1906, part ii, 1907. Speiiann : Stizungsber. d. phys.-med. Gesellsch., Wurzburg, 1900. Roux's Archiv, vol. xv, 1903; and Zool. Jabrbiicber, vol. vii, Supplement, 1904. Stockaed : Roux's Arcbiv, vol. xxiii, 1907 ; also Jour. Ex. Zool., vol. vi, 1909. Taedffi: Storia delli Teratologia, 8 volumes, Bologna, 1881-1895. Toeneau and Martin: Journal d'Anat. et Pbysiol., vol. xvii, 1881. Tornier: Roux's Arehiv, vol. xx, 1905. Valentin: Handworterbuch d. Physiologie, vol. i, 1842. Vejdovsky : Entwieklsg. Untersuchungen, Prag-, 1890. Voigt: Anatom. Hefte, vol. xxx, 1906. Wetzel: Arch. f. mik. Anat., vol. xlvi, 1895. Williams: Obstetrics, New York, 1903. Wilson : Jour, of Morpb., 1893. Von Winckel: Ueber die Missbildung von ektopiscb. entwichten Eruchten, Wiesbaden, 1902. Ueber die menschl. Missbildungen, Samml. klin. Vortriige, Leipzig, 1904. X. THE DEVELOPMENT OF THE INTEGUMENT. By FELIX PINKUS of Berlin. A. THE EPIDERMIS. From the beginning of development the epidermis forms the outermost investment of the body. It consists of a uniform two- layered sheet, the upper layer forming a sort of hard covering- layer while the lower one remains soft and gives rise to new cells and to all the epidermal appendages of the integument. This two- layered stage persists over most portions of the body until into the fourth month, but even at the end of the second month it is not altogether unmodified. The regions which show the first signs of further development are all upon the ventral surface of the body, the skull and back remaining covered by an unaltered, two-layered, indifferent epi- dermis. In an embryo of 15 mm. Kallius found the first indica- tions of the milk ridge, and Tandler observed it later in one of the 9.75 mm. But the modifications of the epidermis are not confined to the regions of the milk ridges at the sides of the body ; also on the ventral surface, anteriorly over the branchial arches and posteriorly as far as the tail, changes occur which indicate a strong formative tendency. In somewhat older stages (32 mm., 40 mm.) an increased tendency towards development shows itself, especially over the facial region, on the anterior surface of the face and neck by the height and regularity of the basal columnar cells, and in the region of the eyebrows, the upper lip, and chin by the distinct commencement of hair formation. a. EARLY STAGES. Where its formation is most simple the epidermis consists of : 1. A superficial layer of flat cells, the epitricMum, or, better, the periderm (W. Krause, 1902). 2. A layer of cells greater both in height and breadth, the stratum germinativum (see Fig. 196). Beneath the latter and sharply marked off from it is the fibrous and very cellular connective tissue. 1. The periderm is the outermost layer of the epidermis. It consists, for the most part, of flat cells, which in transverse sec- 243 24i IiroiAX EilBRYOLOGY. tions of the integnament appear to be spindle-shaped with deeply staining, thin nuclei, while from above they appear as a layer o± large polygonal cells with large roundish nuclei, liven m very eadv stages the peripheral portions of the cells flatten out so ttaat onlv the central portions containing the nuclei remain thick Fig. 19M) Gradually they become quite flat and unusually large (Minot 1894) Frequently one finds some of these cells separated from the rest, so that they are seen from the surface m transverse sections, in which cases they appear as slightly irregular roundish disks with centrally placed nuclei, which are either still round or have become irregular. Around the nuclei there are frequently a larc^e number of roundish cavities, which give to the central por- tions of the cell the appearance of a coarse network. These are the cells which Eosenstadt (1897) found in the beak of an embryo chick where thev were full of large keratohyalm granules by whose solution the cavities are formed. Zander (1886) described them in the skin of all fingers and toes, where they were also observed by Kolliker (vesicular cells) and by Okamura (1900). As the outermost layer of the epidermis the periderm cells have the func- tion of the later-formed corneous Z ^ -^ — -c. layer, and they actually form an investment of a horny character e> (as shown by their reactions: indigestibility, Unna, 1889; yel- Fio. 196. — Hnman fetus, 5 cm. in greatest Imir cfainino- -nri f ll niprif apid length, female. (Collection of Prof. Robert lOW Staining WllU piCllC dClU, Meyer, No. 249.) Integument of abdomen, right Oedercreutz, 1907 ) . IntllCmOre side; '■pi'Iermistwo-layered. P., periderm;. .B., la>- ., . ^ -. . -. erofba.saicells;C.,corium^^ithfe^vcells. X200.1 deVelOpCd pOrtlOUS Ol tllC epider- mis (as on the forehead) these cells become heaped up in two or several layers, and may even form distinct elevations, as at the nostril and mouth openings, in which cases the cells are especiallj' large (Fig. 198). In man the periderm is not a layer which requires to be espe- cially distinguished as the oldest or specifically embryonic invest- ing layer. It is only the outer layer of epidermis, whose cells are no longer turgid and have become firmer and incapable of repro- duction. It merely occupies the place of the later homy layer and receives additions from the subjacent germinative layer, just as throughout life all the more superficial layers are recruited from the deepest layer, the stratum cylindricum. That the periderm is added to is shown 1. By the desquamation of its cells. ' Fig-s. 196-198 were drawn without nse of a camera and consequently the enlargements cannot be given with certainty. The remaining figures were drawn with a Zeiss- Abbe camera. DEVELOPMENT OF TPIE INTEGUMENT. 245 2. By the arrangement of its cells in a regular layer, not- withstanding the increased growth of the skin-surface. 6. Bj the ocal heaping up of layers of completely and simi arly formed periderm cells in the course of development. Each of these three phenomena indicates an increase in the number of periderm cells. 2. The deeper layer, thestratumgerminativuni, is the reproducing layer of the epidermis. Its cells are at first low, the breadth being equal to or even greater than the height; their nuclei are round or slightly oval, stain beauti- fully with a distinct chro- matin network, and are very large in proportion to the entire volume of the cells. The basal surfaces of the cells, turned towards the connective tissue, are flat or slightly concave, and at first are but sligiit- ly connected with the co- rium, so that they readily separate in spots after the death of the fetus (mac eration) or as the result of preparation. The lat- eral walls are variously curved, but in general but slightly, in correspond- rn II ?• ;~; '^"■S", '^*"^' ^2 mm. in greatest length. (Collection of Prof. Robert Meyer, No. 307.) Integument from the right side of the body; the epidermis is teginning to become three-layered. P., periderm; /., stratum inter- medium; B., stratum germinativum; C, corium, rich in cells; G., vessel. X 400. Fig. 198. — The same fetus as Tig. 197 (32 mm.); peri- derm elevation of the right upper lip. P., periderm, consist- ing of large vesicular cells, probably cut somewhat obliquely; /., stratum intermedium, with cells somewhat smaller, flat- tened; B., stratum germinativum, of high columnar cells with high, smaller, dark nuclei; C, corium separated from the epithelium, very cellular; Ci, the uppermost layer of the corium, especially rich in nuclei. X 100. ence with the pavement- like apposition of the essentially cubical cells. The outer surfaces are for the most part more or less convex. No special contents can be distinguished in their protoplasm by ordinary methods of preparation. In those regions which already, in these early stages, show an advance in development, the cells of the deep layer become higher, and finally columnar; the nuclei are closer together and form a quite regular layer, parallel with the lower surfaces of the cells, as may be recognized by weak magnification of not too 246 HUMAN EMBRYOLOGY. thin (15 /^) sections. They stain distinctly darker and are round or oval, the long axis being perpendicular to the surface. The cells are arranged palisade-like, close together, with perpendicu- lar side walls ; and their upper surfaces are rather straight, form- ing a slightly wavy line beneath the stratum intermedium. Their lower surfaces are no longer smooth as in the first stage, but are drawn out into small projecting feet. The cell bodies are much clearer than those of the superposed layers ; they are homogeneous, without any granular contents. Since the nuclei all lie in tlie outer portions of the cells, the lower portions appear as a clear- band between the row of dark nuclei and the dense mass of nuclei which occupies the most superficial portions of the corium. b. FURTHER DEVELOPMENT. Very early there appears between the periderm and the stratum germinativum a middle layer of cells, the stratum inter- medium. Previous to its appearance the cells of the stratum germinativum become higher and more closely approximated, and Fig. 199. — Human fetus, 85 mm. vertex-breech length, male. Epidermis distinctly three-layered. P., periderm with partly separated cells; /., stratum intermedium; B., basal layer of cells with mitosis; C, corium, rich in cells (mostly spindle-shaped). Mammary region. X 430. their nuclei become round and large. First individual cells appear between the two primary layers (Fig. 197), and then a complete row of them (Fig. 199), their nuclei being small and transversely oval and the cell bodies smaller than those of the basal cells, and they take the nuclear stain (carmine) somewhat. These simple conditions occur from the youngest up to rather advanced stages of development (end of the fourth month), where the integument has not yet formed any special organs. In those places where a modification occurs, as, for example, in the DEVELOPMENT OF THE INTEGUMENT. 247 region of the moutli and nose, the epithelium assumes quite early a very considerable thickness. Toward the end of fetal life the layer (layer of prickle cells) situated between the stratum ger- minativum and the corneous layer becomes the principal constit- uent of the epidermis. It is a solid layer, varying in thickness in different regions of the body, and its under surface forms an irreg- ular network of ridges and convexities, which increase its surface of contact with the corium from which it is nourished (rete Malpighi). In vertical sections of the skin these ridges appear as Fig. 200. — Human fetus, eighth month, male. D., stratum disjunctum of the corneous layer ; H., deeper portion of the corneous layer (the keratohyalin and eleidin layers are not recognizable) ; St., layer of prickle cells with epithelial bridges; .6., stratum germinativum, partly separated from the corium and "with the processes of the basal portions of the cells; C, corium. From the right mammary region. X 430. the so-called rete papillae. The layer of prickle cells is composed of a mosaic of closely apposed and regularly spaced large cells; their nuclei are large, they have a polygonal outline in section, and are variable in form and size within narrow limits (Fig. 200). Between the layers of the two- or three-layered epidermis epithelial bridges cannot yet be made out with certainty; but as the epidermis increases in thickness, or in early stages where it has already thickened, they become distinct. With ordinary stains or when unstained they appear as prickles (Riff el, Max Schultze; filaments d'union, Eanvier), but with specific stains (Kromayer, IJnna) they appear as epithelial fibres, which extend throughout a whole series of cells. 248 HmiAX EilBRYOLOGT. These epithelial fibres form only in the peripheral portions of the cells; these become denser and are distinguished as exoplasm from the endoplasm which contains the nucleus (Studnicka, 1903). The epithelial bridges arise by the forma- tion of vacuoles at the boundaries between cells; the fibres differentiate from the exoplasm. In cell division the entire cell divides and both daughter cells again form on their contact surfaces a new exoplasm layer containing vacuoles. In a similar manner Ide (1889) regards the outer layer of > the epidermis cells as a membrane, the prickles being foimed by a process of drawing out, as is especially evident after division when an intei-vening wall is formed between the two young cells. Almost the same idea, that the outer parts of the epithelial cells are a membrane, is expressed by Unna (1903). According to his view the epithelial cells are in close contact, the apparent clear intervals between them (readily visible in ^ ° the case of cells rich in protoplasm) not being intercellular spaces, but the outer layer of the cells, which stains with diffi- culty and is practically a membrane. The epithelial fibres are not empty spaces or spaces merely filled with intercellular fluid ; empty spaces have a vei-y different appear- ance, as may be seen where the protoplasm has retracted from around material (leu- cocytes) which has penetrated it. The limits between the cells are at the so-called nodes of Bizzozero, situated approximately at the middle of the epithelial bridges. These nodes lie in the very narrow clefts between the cells and appear as nodes on account of differences in refraction on staining. In comification it is only this membrane-like exoplasm layer that be- comes cornified, and the remains of the nodes are retained on its surface. The question whether the nodes are actually foi-m elements or merely the result of light interference by superposed networks, is not yet definitely settled; it would appear that the cell walls traversed by fibres may be confused with nodes in the thin (unstained) section, for nodes are frequently seen to be united by a narrow streak parallel to the cell wall (see Fig. 201, a, ch e). Fig. 201. — Epithelial fibres and cell bridges from the layer of prickle cells. From a pointed condyloma, adult, a, Unna's epithelial fibre stain; long fibres traverse the protoplasm of the epithelial cells, nodes on the intercellular bridges, (d is from the same preparation, more highly magnified.) ^C 430. 6, unstained; nodes partly double. > 860. c and e, stained with iron-hfiemato-xyhn; nodes visible in addi- tion to the fibres, >, 860. That the epithelial fibres arise from the exoplasm is generally admitted. They do not merely unite neighboring cells, but may extend through a whole series of cells. According to Schridde (1906) certain regular fibre systems may be recognized: in the deepest layers of the epidermis they form perpendicularly placed ovals, which are found also in higher layers; nearer the surface they form circles ; and at the surface horizontally placed ellipses. The form of the fibre arrangement consequently follows that of the cells, which nearer the corium are columnar, while those higher up are equal in all their diameters, and, finally, those at the surface are flattened. DEVELOPMENT OF THE INTEGUMENT. 249 c. FORMATION OF THE STRATUM CORNEUM. Those regions in which the epidermis consists of many layers show a cornification, but also in other regions of the body there early appear indications of it. These are distinctly visible in the second month, and in the third month the entire skin is undergoing cornification. Cedererentz (1907), using the method of Zilliacus, obtained the following colorations in a fetus 3.5 cm. in length : Yellow : the face, especially in the region around the mouth and nose (most marked in the epithelial plugs of the nostrils) and in front of the pinna. Yellowish: the lateral portions of the back, especially in the lower part, and also the lateral portions of the abdomen. The arrangement of the yellow spots on the body was rather distinctly symmetrical. Bluish-violet : the umbilical cord, the pinna, and the tingers and toes. In all other regions the skin assumed a dirty bluish-brownish green color. In a fetus 5 cm. in length the entire body was distinctly colored yeUow. The pavement epithelium, which becomes yellow with this stain, picric sub- limate, and hsemalum, must be regarded as eomified. Bjorkenheim (1906) has shown that the same regions that stain yellow also resist pepsin and trypsin digestion — a peculiarity which in the skin and mucous membranes is associated only with eomified epithelium. The cornification of the periderm, however, does not pass through the same stages as are to be seen in the formation of the definitive corneous layer. This is shown by the observation of Ernst (1896), who found that in the fourth month cornification could nowhere be observed in the hand or foot, except in the nails, by the Gram method of staining, and that a uniform corneous layer first appeared on the toes in the sixth month. In young embryos whose cells of the corneous layer still retain their nuclei, keratohyalin and eleidin, which are later the constant by-products of cornification, are comj)letely wanting. Keratohyalin and eleidin (parakeratose) are also lacking in the adult skin in places which eomify (pathologically) in such a manner that the nuclei of the corneous cells remain colorable with nuclear stains (hematoxylin, methylene blue). These substances first appear at the end of the third month in places where the epithelial cells are arranged in many layers. In fetuses 10 cm. in length keratohyalin granules are rather abundant in the face, but at this stage they are to be found else- where on the body only in places where longer outgrowths of the epidermis occur, as, for example, at the mouths of the long epithe- lial appendages around the nipples, in the anlage of the mammary gland, and, especially, in the epidermis of the nail bed. At the beginning of the fifth month keratohyalin is still lacking in the skin in general (Stohr, 1903) ; but trichohyalin has formed in the hairs and keratohyalin in the epidermis of the hair follicles. With the continued development of the skin the process of cornification 250 HUMAN EilBRYOLOGT. takes an entirely different course. The same substance that had already appeared in the early fetal stages, although in much smaller quantities, keratin, seems to result from the process ; but at the surface of the skin in later stages the cornification is usually accompanied by the formation of the by-products already men- tioned. As a consequence the corneous layer, on account of char- acteristic refractive properties and staining peculiarities, may be divided from below upwards into certain readily distinguishable layers. These are: 1. The stratum granulosum, with keratohyalin granules. The keratohyalin (Waldej^er), in the form of round or irregularly shaped granules (clumps, threads, occasionally bent at an angle or branched), is situated between the nuclei and the fibrillar layer of the epithelial cells, the exoplasma always forming an external investment of the cells free from granules. From it and its fibrillse the keratin is formed (Unna) ; the keratohyalin forms in tlie rest of the protoplasm. That the nucleus is concerned in its formation does not seem to be definitely shown by the similar staining prop- erties of the keratohyalin and the nuclear chromatin, by the similar non-polarizing refraction of the nucleus and stratum granulosum (the layer of prickle cells, on the other hand, being doubly refrac- tive, as well as the superficial corneous layers), and by the diminu- tion of the nucleus as the keratohyalin increases. Arcangeli (1908), it is true, claims to have directly observed (in the oesopha- gus of the guinea-pig) the extrusion of keratohyalin granules from the nucleus; and Cone (1907) has seen the same outpouchings and expulsions of chromatin from the nucleus in human skin which had been kept for eighteen to twenty-four hours in a thermostat. The keratohyalin is often situated at first at the periphery of the protoplasm (Weidenreich, 1901; Apolant, 1901), but collects later and preferably in the neighborhood of the nucleus. It is insoluble in our hardening fluids, and especially in water, alcohol, and ether ; but is soluble in alkalies and acids. In contrast to keratin it is soluble in hydrochloric-pepsin. It stains deeply with nuclear and acid stains (methyleosin. Zander, 1886; Rosenstadt, 1897; acid fuchsin), but not with osmic acid or Sudan (Rabl, 1902). Unna believes that with its appearance the glassy transparency of the skin of the young fetus disappears, since the keratohyalin (and eleidin), on account of its high refractive index, would render the skin opaque white. 2. The stratum lucidum (Oehl's layer) with eleidin drops. Immediately over the keratohyalin stratum is a thin layer which also remains unstained when treated with osmic acid, but after such treatment stains red with picrocarmine (Unna's basal cor- neous layer, Eanvier's stratum intermedium). Upon this there follows a thicker layer, that blackens with osmic acid (Unna's DEVELOPMENT OP THE INTEGUMENT. 251 superbasal corneous layer). This layer contains the eleidin, a name proposed by Ranvier for both keratohyalin and eleidin, the latter having been distinguished from keratohyalin by Unna and Buzzi in 1889. Eleidin is soluble in water and, in the fresh con- dition or after hardening, in strong alcohol; it stains with picro- caxmine and nigrosin (Buzzi). Rabl (1902) regards it as softened keratohyalin, the softening giving it the consistency of a fatty oil (keratoeleidin). According to the microchemical investigations of Ciliano (1908) it is an albumin. It does not appear in the form of granules, but either as a thickish fluid extending through- out the entire layer, or else in large drops or pools. Dreysel and Oppler (1895) could not detect it in a five-months fetus; it was abundant in the skin of one of eight months. With ordinary stains or when examined unstained in glycerin the stratum lucidum appears as a clear band. In this layer the cell membranes are already composed of keratin. 3. The stratum corneiim forms the outermost portion of the skin. In its cells only the exoplasmic portion and the thickened fibrillar layer are cornified (Unna; Ranvier; Weidenreich, 1901), and on their surfaces remains of the prickles can still be recog- nized. In the cells themselves a fat which reduces osmic acid collects and becomes very distinct in sections on treatment with osmic acid after fixation in Miiller's fluid (Rabl, 1902) ; according to Ranvier it has some resemblance to beeswax. The cells which contain this substance are flat and thin, but possess the property of swelling after an infiltrating injection of fluid into the skin. Probably the fat (pareleidin, Weidenreich) is formed in some way from eleidin (Rabl). Apolant (1901) assumes that the eleidin must be expelled from the flat corneous scales which represent completely cornified cells. The most superficial portion of the corneous layer stains diffusely with osmic acid. It has been termed by Ranvier the stratum disjunctum and by Unna (1883) the super- ficial corneous layer. In the homy substance there are two kinds of substances which react differently to chemical reagents (Unna and Golodetz, 1908) : (1) those which are not digested by hydrochloric-pepsin and which color red with Millon's reagent (presence of tyrosin), but are insoluble in fuming nitric acid or in sulphuric acid + hydrogen peroxide (keratin A) ; and (2) those which also are undigested by hydrochloric-pepsin and stain red with Millon's reagent, but are soluble in the strong acids mentioned (keratin B). d. GRANULE INCLUSIONS OF THE EPIDERMAL CELLS. While the cells, as they pass toward the surface, become cornified and in so doing form keratohyalin, eleidin, a wax-like substance, and tyrosin, the deeper layers contain other substances which are not found nearer the surface. 252 I-TOIAX EMBRYOLOGY. 1. Pigment (melanin) occurs only in the deepest layers of the epidermis and develops chiefly only after birth. In a six months fetus Dreysel (1S95) found no pigment; the granules which blackened with osmic acid did so also after previous treat- ment with chromic acid, a reaction which indicates fat but always destroys melanin. The extra-uterine development of pigment is much more distinct in the dark races than in Europeans. Negro children are quite light-colored at birth; they become brownish yellow on the second day and thereafter darken rapidly (Wieting and Hamdy, 1907), so that in six weeks they have reached the normal degree of darkness (Falkenstein; at four months accord- ing to Frederic, 1905). The children of the Australian aborigines are pale yellow with the exception of some black lines around the mouth, eyes, and nails— but by the broadening of these lines they become black in a few days (G-unn, according to Merkel). The pigment of the epidermis is much more abundant than that which occurs in the corium ; indeed, the latter may be entirely lacking. The epidermal pigment seems to be formed in situ. That it may be formed there is shown by experiments on the adult (exposure to Finsen light rays for two hours, accompanied by cooling and deprivation of blood, after which treatment excised portions of skin, examined immediately, show an increase in the amount of pigment), by the pig-mentation of skin from a cadaver (in incuba- tor, Meirowsky, 1906, 1908), and by the pigmentation of vitiligo spots in the absence of pigment in the corium (Buschke, 1907). r'urthermore, the iireferential presence of pigment in the skin, chorioid, and epend^-ma is an indication of a special pigment-form- ing faculty of the ectoderm (chromatophoroma of the central nervous system). In the epidermis the pigment granules occur: a. In the ordinaiy epithelial cells, at first surrounding the nucleus (Grund, 1905; Meirowsky, 1906, who regards the pigment as a transformation of the nucleolar substance), but later mostly in a cap-like mass on its outer surface. b. In stellate cells (chromatophores), which resemble the Langerhans cells. Whether these apparently stellate cells are really branched or whether the processes extending out from them are not streams of pigment granules passing out into the spaces between the small contact surfaces of adjacent ordinary rete cells (Eabl), is not as yet determined. According to Meirowsln^'s (1908) obsem^ations these cells develop from ordinary epidermis cells. The similarity of the chromatophores to the greatly branched Langerhans cells is striking, these latter cells, as well as the former, being demonstrable with especial distinctness and in large numbers by the gold and silver methods (Ramon y C'ajal's and Levaditi's silver method), so much so that on the basis of this reaction the Langerhans cells have already and again recently been actually termed colorless pigment cells (Schreiber and Schneider, 1908; Bizzozero, 1908). The DEVELOPMENT OF THE INTEGUMENT. 253 questions in discussion turn upon wliether tlie epithelium always forms pigment in situ (Post, 1893, the development of feathers; Jariseh, 1892, no pigment in the liair papillae ; Sehwalbe, 1893, the change of hair coat in the ei-mine, in which the white winter coat contains no pigment, while the summer coat makes its appearance pigTQented; Rosentadt, 1897), whether the epithelium also forms the spider-like cells (melanoblasts) (Grand, 1905; L. Loeb, 1898; Wieting and Hamdy, 1904, the gradual pigmentation of the nose in new-born dog-s), which then may wander from the epidermis into the cutis (pigment stored up in the lymph-nodes, shown by Jadassohn, 1892, in pityriasis rabra and by Schmorl, 1893, in the negi'o), or whether the chromatophores are originally pigmented connective-tissue cells which wander into the epidermis and supply its cells with pigment (Ehrmann, 1896; in the unpigmented egg of the triton there is formed in the mesenchyme, when the development of the blood begins, a series of dark cells, melanoblasts, which later become rich in pigment granules and give rise to all the pigment in the body. No pigment is formed in the epithelium, it is carried there by the melanoblasts; they appear about the hair anlagen at an early period, even before the fonmation of the papillas). According to all recent works the idea that pigment is formed in the epidermis itself seems to be well founded. 2. Glycogen is abundantly present in the embryonic epidermis (S. H. Gage, 1906). After the sixth month it diminishes in quan- tity and is finally to be found only in the cells in which it occurs in the adult (Lombardo, 1907). The epithelium of the sudoripa- rous glands, especially, regularly contains glycogen, the more the more actively they are secreting; furthermore, the outer root sheath of the growing hair, from the bulb to the insertion .of the muscle, contains it, while in that of the bulb hairs it is wanting (Brunner, 1906; Lombardo). 3. Fat is especially evident in the basal columnar cells and in the stratum granulosum even from the fifth month. The layer of prickle cells which intervenes between these is practically fat-free, its fat having presumably been used up during the divisions of the cells. As evidence for the absence of fat in the prickle cells it may be stated that post mortem no fat appears in them when they are placed in a thermostat, while in twenty-four to seventy-two hours the quantity of it in the stratum cylindricum and stratum granulosum is greatly increased (Cone, 1907). In the layer of columnar cells the fat, made evident by fettponceau, surrounds the nucleus in the form of granules of varying size (up to one-quarter the size of the nucleus) and streams out toward the periphery of *the cells. In the stratum granulosum it is equally distributed from the cell periphery to the edge of the nuclear cavity, some loops lying even within this (according to Unna it is completely filled by a fat drop). Much fat is also present in the stratum lucidum, where it is more irregular both in form and distribution than in the stratum granulosum. In the actual stratum comeum it lies especially between the cells. In between the epithelial cells processes of brnnched con- nective-tissue cells filled with fat granules (lipophores, Albrecht) extend from below. 25i HraiAN EMBRYOLOGY. B. CORIUM. The connective-tissue portion of the skin, the corium, is de- veloped from the most superficial portions of the somites. Ven- trally it arises from the outer part of the dermomuscular plate and is applied to the mj'^otomes so that, together with the epider- mis, it sinks down into the furrows between successive myotomes (sub-epithelial segments, Osc. Schultze, 1897). The case is similar dorsally, where the segmental furrows are continued between the sclerotomes which surround the spinal cord. The segmental fur- rows, however, soon disappear. The corium or dermis is at first very cellular, and the two portions into which it later divides, the corium proper and the subcutaneous fascia (tela subcutanea), are not distinguishable in the second month. Its oval cells elongate and begin to form fibrils (observed by Spalteholz, 1906, in the fifth week) in their own protoplasm, and these soon anastomose and form a delicate, somewhat irregular network, which, according to Spalteholz, remains throughout life unsheathed by the protoplasm mass and in connection with its original cells. A regular arrange- ment of the bundles of fibrils makes its appearance at about the end of the third month. In embryos 7-8 cm. in length the bundles arrange themselves in the lower part of the body in parallel bands running around the body (a result of the stretching of the skin by the growing liver). A little later the parallel arrangement of the bundles appears in the remaining portions of the skin as a result of the tension produced in it by growth (Otto Burkard, 1903). The regularity of the fibrils is interfered with by the development of the hairs (from the fourth month onwards), since the bundles separate around the down-growing hairs and form meshes, which after certain modifications are transformed into Langer's (1861) rhomboidal meshes. In the later embryonic period the superficial layer of the corium separates into the papillary bodies and the stratum retic- idare. The papillary bodies form a superficial layer whose fibres —collagenous connective tissue— stain reddish with Van Gieson's stain (picric acid and acid fuchsin) and are arranged horizontally, but with many vertical fibres that extend to the boundary of the epidermis. The stratum reticulare consists of coarse and fine bundles of connective-tissue fibres, which, interwoven, run more or less parallel to the surface of the skin and take the picric acid of Van Grieson's stain somewhat more strongly. The varieties of fibres which can be clearly distinguished microchemically differentiate rather early in certain regions of the body. According to earlier accounts the elastic fibres of the corium first appear in the seventh or eighth month, but according to Spalteholz (1906) they are pres- ent in the truncus arteriosus of the chick even in the third day, in DEVELOPMENT OF THE INTEGUMENT. 255 pig embryos of 9.2 mm., and in calf embryos (in the ligamentum nucbae) of 35 mm. Tbey arise intracellularly, either directly as fibres without any granular prophases in the protoplasm of the cells (Spalteholz; Gemmill, 1906, in tendons; Schiffmann, 1903), or as rows of granules (Jones, 1907, iu the epicardium of chick embryos; Teuffel, in the fetal lung; Nakai, 1905, in the vessels of chick embryos). The elastic substance can, apparently, form in all fibroblasts; special elastoblasts (Passarge, 1894; Loisel, 1897) are not recognizable. Pigment cells occur in the corium in varying numbers; they are partly small, like the epithelial chromatophores, partly espe- cially large and deeply seated. These latter appear during the fourth fetal month. According to Grimm (1895) and Adachi (1902) they are the cells which produce the blue gluteal spots (Mongolian spots). The corium pigment appears to be formed in the connective-tissue cells themselves, as the result of activities by which pigment is produced from uncolored constituents (accord- ing to Meirowsky, 1908, the cell nuclei are concerned in the process ; extruded nucleoli are transformed into pigment). The young, richly cellular corium contains wide blood-vessels with a distinct endothelium, but without any of the other con- stituents of the wall. Gradually the abundant and specialized vas- cular supply of the skin of the child develops, with its superficial vascular network and a deeper one parallel with the surface of the skin and connected with the superficial network by vertical branches, and with the vascular supply to the glands and hairs and some superficial fat islands. Beneath the corium is a looser tissue characterized by the formation of fat islands. According to Toldt these begin to form at definite places, so that the fat lobe is a special organ, with a peculiar blood-supply — a view with which, according to Eabl (1902), the sharp delimitation of the fat lobes is in agreement. More generally accepted is the idea, first suggested by Czajewicz (1866) and later more definitely by Hemming (1876), that in the subcutaneous tissue every cell may become a fat cell, even although there are constant areas in which fat develops by preference (Unna, 1881). The cells are at first branched and contain the fat in the form of small droplets, from which larger drops are formed, while the processes of the cells become less distinct. Gradually a large fat drop is formed, surrounded by a thin layer of protoplasm which contains the nucleus. The fat cells lie in groups, surrounded by ordinary connective tissue and richly supplied with blood- vessels. The masses of fat assume various forms, such as lobes into which the principal blood-vessel enters from below, cords extending along the blood-vessels, or islands standing isolated on the blood-vessels of the hair follicles and lacking a special vessel. 256 HUMAN EMBRYOLOGY. With the modern fat stains (fettponceau) it is possible to find even in the adult skin branched fat cells packed with small and large drops, whose processes extend into the epidermis (near to which they usually lie). These are Albrecht's lipophores_ (Cone, 1907). Fat granules may also be detected by staining in cells with pigment granules and in those with enzyme granules. C. THE CONNECTION BETWEEN THE CORIUM AND EPIDERMIS. The two-layered epidermis lies flat upon the corium through- out the greater portion of the body. In those places where the basal layer consists of high columnar cells, even in early stages (30 mm.) a closer connection occurs, the bases of the epidermal cells being divided into fine processes which fit into corresponding depressions of the corium. With increasing growth this connection becomes continually more intimate, and in fetuses of 10 cm. the basal portions of the columnar cells consist of a finely fibred protoplasm, which stains deeply and represents the rooting feet of the epidermis cells, firmly connected with the especially cellular surface of the dermis. This condition becomes more marked with increasing growth. In the adult skin the central parts especially of the cell bases seem to have formed fibrous processes. By maceration in 10 per cent, salt solution (Merk, 1904), in pyro- ligneous acid (Loewy), or in weak acetic acid (Blaschko, 1888) the epidermis separates from the corium without losing its own continuity, so that it seems that a change in the arrangement of the protoplasm of the foot portions of the epidermal cells has occurred (Merk), a change which is also indicated by their dif- ferent staining properties (bright red with eosin; reddish yellow Avith picric-acid-fuchsin, in contrast with the yellow color of the rest of the cell, a darker brown than the epithelial cells with saffranin after fixation in Flemming's fluid; a brown or grey color, the epithelium remaining uncolored, with the elastic fibril stain of Unna and Weigert, Rabl, 1902). Quite as delicate as the fibre structure of the epithelial root feet is that of the subjacent super- ficial lasers of the corium. But while the foi'mer are arranged perpendicular to the surface, in correspondence with the arrange- ment of their fibrils (epithelial fibres), the delicate fibrillation of the corium is, in general, parallel with the surface, only the finest of connective-tissue and elastic fibres ascending vertically towards the epithelium boundary. A penetration of corium fibres into the epidermis cannot be made out, but whether basal epithelial cells separate from tbe epidermis and wander into the corium or not is less certain. According to Kromayer (1905, dermoplasia), and especially according to Retterer (1904), the superficial portion of DEVELOPMENT OF THE INTEGUMENT. 257 the corinm is of epithelial origin, being formed by cells separated from the epidermis; and recently Krauss (1906), as the result of the employment of an especially distinctive staining method, has regarded a portion of the corinm in the Eeptilia as derived from the epidermis. On account of the irregTdarity of the contact sur- face between the epidermis and the corium, even the thinnest sec- tions are more or less oblique and give the appearance of an inter- stratification of the deepest portions of the epithelium and the connective-tissue fibres (the elastic fibres, namely), and so of the penetration of the fibres into the epithelium. So extensive a migration of epithelial cells and their conversion into connective- tissue cells has certainly not been demonstrated. A certain mingling in the course of development of living epithelial elements, such as touch corpuscles, pigment cells, and perhaps also un- pigmented nsevus cells, with corium constituents has been observed by many authors. In pathological processes separated epithelial cells (as in lichen ruber and vesicular eruptions) or epithelial islands or appendages (as in tuberculosis and trauma) usually gradually degenerate and only rarely find opportunities for further growth ; when they do so the growth is always of an epithelial form (traumatic epithelial cysts, milia). The connection of the epidermis and eorium, notwitlistanding the ease with -which they may be separated by maceration, is exceptionally intimate and is not broken by the displacements of the integ-ument which occur in the course of normal development. The epidermis follows every outgTowth of the corium and the latter yields to every epithelial projection, closely surrounding it on all sides. Since the •epidermis and superficial corium (in later stages separable into the papillai-y bodies and the subpapillaiy layer) constitute anatomically a single tissue-mass, and also are exposed in common to all changes, physiological and pathological, Kromayer (1899) has united them under the term parenchyma skin. D. DERMAL RIDGES AND FOLDS. DERMAL RIDGES PRODUCED BY SURFACE GROWTH, GROWTH FOLDS. For a long time the under surface of the epidermis remains smooth or slightly and irregularly wavy. With the development of hairs on the eyebrows and lips in the second month its first deep ingrowths into the corium occur. Much later, when the rest of the hair has begun to form everywhere, the lower surface of the epidermis begins to increase in certain regions, the rete ridges begin to form on the palms of the hands and the soles of the feet, and some time after these the more delicate outgrowths which form the papillary bodies and the rete Malpighi appear. The rete ridges make their appearance on the previously formed touch balls (see Figs. 73, 76, 77, and 78, p. 87-89). Each terminal phalanx of the fingers and toes bears one of these, four occupy the inter- digital spaces between the heads of the metacarpal and metatarsal Vol. I.— 17 258 nu:\rAX ejibryology. bones, and one corresponds to each of the lateral swellings of the hand and one to the side of the little toe. These are recognizable after the sixth week and reach their greatest relative development in the fifteenth week. After that they begin to disappear and their place is taken by corresponding systems of papillary ridges (papil- laiy ridge patterns, which have a triradiate form, consisting of three lines arranged in the form of a triangle, with diverging lines from each angle, Schlaginhaufen, 1905; Whipple, 1904). In the eighth week (Evatt, 1907), in fetuses of 3-4 cm. (Wilson, 1880), the epidermis lies flat on the corium and is separable by slight maceration; no trace of ridges is visible. In the eleventh week (Evatt), in fetuses of 9 cm. (Wilson), the skin appears streaked when seen from the surface, showing alternating light and dark lines, the beginnings of the formation of the rete ridges. While the outer surface of the epidermis is still smooth, with- out any pattern, there arise on its lower surface simple ridges, that are triangular in section with the apex directed downwards. They produce the striated appearance of the skin and are Blaschko's gland ridges, the sudoriparous glands forming later on their lower angles. The outer surface is not raised into corresponding ridges ; rather it may show grooves (W. Krause, 1902). Only in the eighteenth week do ridge-like elevations of the outer surface ap- pear, one corresponding to each of those on the lower surface. The development of the ridges begins at the tips of the fingers and toes and proceeds proximally over the entire surface (not radially from a centre), at the same time producing the ridge patterns with their whorls and spirals. These patterns even in this early stage show the same individual differences as may be observed in adults, and this is sO' even before the ridges of the outer surface are formed. First the gland ridges are formed and then the sudori- parous glands begin to form from them. At about the same time tlie same process takes place in the palm of the hand. In the cases of the gland ridges (papillary ridges, Hepburn; epidermic ridges, Whipple, 1904) an epithelial elevation of the outer surface (crista epidermidis superficialis, Heidenhain, 1906) corresponds to an epidermal ridge of the under surface (crista epidermidis pro- funda intermedia, Heidenhain; subdermal ridge, Evatt, 1907). At regular intervals sudoriparous glands arise from them, their ducts later traversing them in a cork-screw fashion to reach the exterior. Then a second series of lower ridges, destitute of glands, is inter- posed between the gland ridges, forming Blaschko's folds (cristse epidermidis profundse limitantes, Heidenhain) ; and finally delicate low transverse bridges are formed between the two series. Cor- responding to the folds on the outer surface, and therefore cor- responding to the grooves between the ridges, are depressions of the epidermis (sulci superficiales, Heidenhain). DEVELOPMENT OP THE INTEGUMENT. 259 According to Whipple (1904) the gland ridges are formed by the union of isolated epithelial papillae, each of which is traversed by a sudoriparous gland; and the transition of a ridge into a series of such papillae with few or but one gland duct is also to be seen in certain regions of the adult skin, constantly on the radial sides of the fingers and especially of the index finger. The development of the ridges on the palms and soles is completed at the end of the fifth month. From them there arise as secondary outgrowths the rete papillae, between which the true papillary bodies occur. In extra-uterine life the ridges lose much of their regularity, since new irregularly arranged outgrowths make their appearance. Simultaneously with the formation of the rete ridges the outer surface of the corium develops elevations and papillte, as a nega- tive, as it were, of the epidermis; for, on account of the continuity of the entire parenchyma, skin, an elevation of the epidermis must produce a practically corresponding depression of the corium. The dermal ridges of the hands and feet, so far as these are not movement folds but are the result of the enlargement of the epi- dermis surface (Lewinski, 1883), have their arrangement deter- mined by a series of mechanical conditions. In general, they are arranged transversely to the direction of the limb in grasping or progressing (friction skin, Whipple) and, consequently, at right angles to the directions of the most delicate sensations for the substratum ( Schlaginhauf en ; their distally overlapping, imbri- cated arrangement at the tips of the fingers, pointed out by Kidd, 19(J5, also serves to increase sensation) ; they constitute neutral curves, so that they are neither stretched nor compressed by ten- sions of the surface, and, consequently, the touch sensation is not disturbed bv the sensation of a surface tension (Kolosoff and Pankul, 1906). In the regions of the body where hair occurs, longer or shorter meshed networks form between the hairs ; the rete papillae become much lower in the vicinity of the hairs, so that these, when fully developed, occupy the centre of a star of low rete papilte, which extend more or less distinctly upon the hair follicle (Philippson, 1906). As a result of the formation of the papillce of the corium and of the ridges and papillae of the epidermis, the under surface of the latter becomes greatly increased in comparison with its earlier smooth condition, and a very much greater nutritive and functional surface is thus secured for the epidermis. The rete ridges follow in general the tension lines of the skin, so that a correspondence exists between the tension lines and the development of the ridges. 260 HUMAN EMBRYOLOGY. TENSION FOLDS. In addition to the folds produced by the development of the skin— the growth folds— a second variety occurs, produced by the innumerable repeated bendings and foldings of the skin during movements of the parts (Lewinsky, 1883 ; Blaschko, 1888; Loewy), or by the strains resulting from these movements (Philippson, 1889; distention folds, Charpy, 1905). In the palm of the hand, where the systeni of growth folds is most marked, some of the strongest flexion folds have the same direction; but elsewhere a contrast between the two systems seems to exist in tliat, in parts with strong flexion folds (joints, hands, nape), these cross the lines of tension of the connective-tissue bundles at right angles. The tension of the tissue is the cause of Langer's lines, the connec- tions between the slit-like clefts formed when the skin of the cadaver is pierced by a round instrument. With these also cor- respond, in addition to the rete ridges, the arrangement of the hairs. The investigation of the tension lines shows that in the fetus great displacements of the skin occur during the course of development. At first no directions of cleavage can be distinguished in the skin, the tensions to which it is subjected being equal in all direc- tions ; but with the commencement of the parallel arrangement of the connective-tissue bundles there arises a definite cleavage (Otto Burkard, 1903). The cleavage lines pass transversely from the dorsum ventrally, diverging, according!)', ventrally; in the ex- tremities they run longitudinally. This cleavage arises gradually, those portions of the skin that are more strongly stretched by the growth of the deeper parts showing it at an earlier period than those ]iarts in whicli the conditions of tension are indifferent. It begins in the third month and lasts until the end of the fourth or the beginning of the fifth. By the formation of the hair in the fourth month, the originally parallel bundles of connective tissue become arranged so as to form rhombic meshes, and then suddenly rearrange themselves as a result of the hair development in the fifth month, so that their new arrangement is at right angles to the original one and the cleavage lines become longitudinal in the trunk and circular on the arms though less so on the legs. The gluteal region now shows cleavage for the first time, the lines run- ning circularly and diverging from the gluteal cleft. In the neck they run horizontally around and remain thus until the adult con- dition, in which they run horizontally from the occipital region to the thorax. Transitions between the courses of the first cleavage lines and the second do not occur. It is not a question of a gradual development as the result of growth processes, but of a very rapid rearrangement of the originally transversely directed rhombic DEVELOPMENT OF THE INTEGUMENT. 261 meshes into longitudinally directed ones, with a brief intervening stage in which the meshes are quadrate and in which no definite lines of cleavage occur. These second cleavage lines vanish in the fifth and the begin- ning of the sixth month and gradually become horizontal, return- ing to the primary direction and being at right angles to the secondary one. This change is the result of gradual growth dis- placements (growth in length) and from it there gradually de- velop the oblique cleavage lines of the adult, directed from above downwards. The cleavage lines of the head and face change but little during development. (See representation by Otto Burkard, 1903.) E. THE METAMERISM OF THE SKIN. The segmental plan upon which the human body is constructed suggests that a metameric arrangement exists also in the skin (Blaschko, 1888). In some mammals (mouse, pig, rabbit), and also in man, the skin in early stages of development follows the metameric arrangement of the underlying structures with wliich it is connected (the myotomes ventrally and the sclerotomes dor- " sally, 0. Schultze, 1897) ; this is the subepithelial segmentation. In fishes the segmental arrangement of pigment cells (Bolk, 1906) indicates such a metamerism of the skin. In mammals none of the well-known metameric marking-s (siich as those of young animals, the zebra aiicl tiger), nor yet the metameric arrangement of the hair (trichomerism, Ilaacke), nor the circular arrangement of scales on the trunk and tail, can with certainty be referred to a primary metamerism of the skin. That this may exist, however, although invisible to our eyes, is indicated by pathological conditions, which are only to be explained as due to a certain predisposition to disease of growth zones or their boundaries (uEBvi, linear inflammations). The method of investigation that represents to us the development of skin segments is the comparison of the course of development of the peripheral nerves with that of the skin areas which they supply. No other tissue, neither the skeleton nor the musculature, corresponds with the segmental structure of the skin. Even the blood-vessels are less satisfactory in this respect; for the blood follows the most convenient paths it can find. The blood-vessels accompany the nerves and frequently confuse, by their anastomoses, the distmct metameric course maintained by the nerves. The nerve connection between the skin and the spinal cord is established in early stages, and it may be supposed that it remains unchanged until the com- pletion of development in spite of all displacements and modifications dependent on grovrth. Those portions of the skin which are supplied, for instance, by the sixth and seventh thoracic nerves, and which Grosser and Frohlich (1902) have followed throughout the development from an embryo of 14J mm. onwards, are to be regarded as identical in both the fetus and the adult. The territory which is supplied by a definite segmental spinal nerve remains the same from the beginning to the end and is known as a dermatome. At first the dermatomes form rings around the body, narrower ventrally and broader dorsally, in correspondence with the curvature of the embryo, in which the thorax and abdomen 262 HUJIAN EMBRYOLOGY. are much shorter than the back. The spinal cord at first grows more rapidly than the skin, and the lagging behind of the latter is shown by the fact that the cutaneous nerve branches pass proximally from, for example, an intercostal trunk, in order to reach their cutaneous areas. They are held back because their areas of distribution are of slower growth than the spinal cord. Later on the spinal cord lags behind, and the skin, with the out- growths of the arms and legs, grows more quickly, drawing the nerves peripherally with it. The growth of the extremities draws the skin out, and the more distant cutaneous areas of the trunk must follow towards the region from which the limbs arise. The dissection of the nerves (Voigt, 1857) shows the situation of the corresponding skin areas, wliich are to be regarded as skin meta- meres. In spite of all displacements, such as occur, for example, in the extremities, the common supply of an area of skin by the branches of the metameric nerve indicates its individuality (Head, 1898; Sherrington, 1893; Seiffer, 1898); and comparative studies (Grosser and Frohlich, 1902, 1904) show that it is to be regarded as a genetic anlage (dermatome). The conditions are much the clearest in the trunk, although even in this region great displacements occur. To the skin pass: ' 1. Twigs of the ramus posterior of each spinal nein^e, the medial tAvigs in the upper regions and the lateral in the lower. Both pass downwards from the intravertebral foramen on their way to the sldn, the medial (upper) ones to a lesser extent than the lower (lateral), which occasionally descend rapidly over three to four rib regions, since their cutaneous areas have been drawn downwards to this extent by the development of the lower limbs. 2. The rami laterales and the rami mediales are carried out of their original course, parallel to the ribs, to a much smaller extent. On the ventral surface of the body the skin follows more regularly the growth of the deeper parts ; it is drawn downwards to a certain extent, but at the same time the ribs and the greatly enlarged abdominal wall are also displaced caudally. The skin and the deeper parts have a common growth and retain their rela- tive positions. F. THE HAIR. The development of the hair begins on the eyebrows, the upper lip, and chin at the end of the second month. On the evebrows it has been seen in fetuses of 27 mm. (Keibel and Elze, "Normen- tafel," Xo. 80. 1908), but the number of anlagen was still verv small. In a fetus 32 mm. in its greatest length (No. 307 of the collection of Professor Robert IMeyer) I found on the eyebrows 89 anlagen on the left and 81 on the right; on the upper lip 73 ankgen on the left and 57 on the right; on the chin only one of which I could be certain and several uncertain ones. DEVELOPMENT OP THE INTEGUMENT. 263 In another fetus 30 mm. in length (No. 310 of the same collec- tion) the anlagen were somewhat more numerous and to a certain extent further advanced in developuient. At this time hair anlagen are present in no other regions. The general hair coat begins to form at the beginning of the fourth month, and at this time only the earliest anlagen occur, except in the face region, where they already show a more ad- vanced development. On the body the anlagen are quite closely placed, on the head they are somwhat further apart, and on the face they are especially close, being arranged laterally on the upper lip in a row which at places is unbroken. The hairs arise singly, but in many regions another appears early on each side of the first- formed hair, so that groups of three are formed. Stohr (1907) saw on the back of the neck two additional hair follicles sprouting out beside a regularly arranged hair group, but the nature of these could not be determined on account of their slight development. The phylogenetic origin of the hairs has not yet been definitely ascertained. Certain facts in connection witli tlieir structure and aiTang'ement have, liowever, been taken as the basis for theories whereby the mode of formation of the hair might be explained. Of these theories the two most important are the following: 1. Maiirer's theory of the origin of hairs from the epithelial sense organs (lateral-line organs) of the Amphibia (Stegocephali) is based upon the similarity of these organs to the first epithelial hair anlagen (hair germs) and upon the comparable arrangement of the sensoi-y hairs (sinus hairs) of the mammals and the lateral-line organs of the head in fish and Amphibia. It explains, according to Maurer, all the conditions of the evolution of the hairs. The transverse rows of hair groups, which are of the gxeatest importance for the following theory, constitute merely a topographic arrangement, and are independent of the arrange- ment of the scutes of the Stegocephali. 2. Weber's theory of the identity of the scutes of mammals with those of their reptilian ancestor's (1893), a theoi-y which has been further worked out by Reh (1896) and De Meijere (1891). As the type of the hair arrangement rows of hairs standing in groups of three have been taken, the groups in successive rows frequently alternating regularly (quincunxial aiTangement). This plan is shown in the development of the hairs at the posterior margin of regularly arranged scutes (t^upposed to represent the scutes of the Promammalia ) . The coat of the most thickly covered mammals is derived from the groups-of -three arrange- ment by secondary hair foimations, which are usually recognizable in the embryo, or by the arrangement of the hair muscles and the sudoriparous glands. The hair groups belong genetically to the scutes, behmd (De Meijere), or, better, upon (Reh), which they arise. This theory receives support from my discovery of the hair disks, in case these may be identified with the touch spots of the reptilian scutes. Just as the reptilian and stegocephalan scutes are bilaterally symmetrical structures, whose planes of symmetry eon-espond with the longitudinal axes of the trunk or the extremities, so, too, each skin area which is regarded as belonging to a group of three hairs is to he regarded as a bilaterally symmetrical structure from'its first beginning, the middle liair of each group of three being situated in its longitudinal axis and being flanked on either side by the two additional hairs, or by all "the additional hairs when the group consists of more than three; on the posterior surface of the hairs it has associated with it the sebaceous gland, the sudoriparous gland, the muscle, and the hair plate, the whole fomiing a hair ten-i- 264 HUMAN EMBRYOLOGY. toi-y which must be regarded as corresponding- to a scute of the promammalian integument. According to this theoi-y the haii-s ai'e to be regarded as new acqui- sitions by the Mammalia, if the entire scute did not in much earlier periods suiTOund a lateral-lme organ. Such a condition would be analogous with what occurs in the fishes, in which the scales of the lateral line also bear the lateral-line organs, and it would render possible a union of the .present theoi-y with that of Maurer. As the first anlage of a hair, in small areas of the three-layered epidermis the nuclei of the columnar cells become higher and more closely packed, so that more cells rest upon a given area of the corium surface than is the case elsewhere (Fig. 202, primary hair germ). In addition, there may perhaps be a very slight downward convexity of the germ, but the nuclei of the corium usually do not show any increase in number. These structures are rather ^ .»«r*."^'S^'*^*^^--**'*.jr;* '^ uncommon; thev seem to represent a very transitory stage, and only when they occur among more developed hair anlagen can they be re- cognized as the first stage of these. A similar appearance is presented by marginal sec- tions through structures ^ ,. ' * ~* '^ which are distinctly hair Fig. 202— Primary hair germ of human fetus, 8.5 germS ( Stohr, 1903) aud they cm. in length, male, from the mammary region. Two vvTiicf r»f 1 £» f\ -Pr\ r\ A '-fin successive sections in which the circumscribed approxi- UlUSL UOt Oe COnlO'UnQeCl WltU mation of the nuclei is visible. The cells of the stratum flnoco Tlicnr pnTT-ci,aT>r>Tirl in oirjo intermedium have not yet increased in number and UieSe. iUey COrreSpOUQ m SlZe there is no increase of the connective-tissue cells of the witll tlie hair germS measur- ing 45-60 \i in diameter. The hair germ (Fig. 203) differs from them in the distinct bulging out of the layer of cohnxmar cells towards the corium. The nuclei of the columnar cells are arranged radially to a somewhat distant centre, and are frequently slightly curved so as to be concave toward the centre of the structure (^laurer, 1895, the kiln-like arrangement of the hair cells). At the point toward which the cells converge there is occasionally a roundish opening. Even in this early sta.ge the germs do not possess a radial structure, but are bilaterally symmetrical (Figs. 204 and 205). On one side (the side of the later acute angle between the hair follicle and the under surface of the epidermis, the anterior surface of the hair) the cohmmar cells come quite up to the hair germ; on the other side (the posterior surface of the hair) lower cells occur in its neighborhood ( Stohr 's anlage of the hair canal). The kiln- shaped anlage is covered by some cells placed horizontally. The corium beneath the colunmar cells, which are increasing in height, usually contains more nuclei than are present in yet undiffer- DEVELOPMENT OF THE INTEGUMENT. 265 entiated regions, and beneath many hair anlagen the increase of the nuclei is especially pronounced (anlage of the papilla). No differences seem to exist between the early formed hair anlagen of the face and head and those of other portions of the body. H.-K. Fig. 203.— Hair germ from the same region as Fig. 202. H.-K., hair germ; I., increase of the cells of the stratum intermedium, and C, of the connective-tissue cells of the corium; Bl., periderm cell. X 430. }I.-Ki -*-<# Fig. 204. — Longitudinalsectionof a hair germ of another 8.5 cm. human fetus, more advanced stage. H.-Kan.. hair canal cells (Stohr). Spaces exist among the cells of the germ. V., anterior surface; H., posterior surface of the hair germ. X 430. Fig 205 Transverse section of a hair germ in the same stage as Fig. 204, from pectoral region of the fetus of Figs. 202 and 203. Shows the distinctly symmetrical structure and the kiln-like arrangement of the cells. X 430. With the further development of the hair germ Stohr 's stage of the hair papilla is reached, in which the hair anlage, while retaining approximately its original diameter, gradually grows downward into the corium (Fig. 206). The hair papilla projects downward from the stratum cylindricum; it consists of an outer layer of 266 HU:MAX EilBRYOLOGY. cylinder cells, wliicli are already beginning to differentiate in the deeper parts of the foUicle, and of polygonal cells, arranged more irregularly, which fill the space bounded by the columnar layer. The latter, when the length of the hair papilla is about three times its breadth, shows on the posterior surface two low outgrowths with outwardly diverging, regularly arranged nuclei. The upper of these is the aniage of the sebaceous gland, which in rare instances may be double (one above the other. Diem, 1907) ; the lower one is that of the sireUing (hair bed, Unna, 1883), which, usually, be- coming lower, extends around to the anterior surface of the follicle. The lowest portion of the hair pa- pilla consists of high columnar cells arranged in a kiln-like man- ner and forming a convexity which is higher than that of the simi- larly arranged cells of the hair germ (matrix plate, Garcia, 1891). There is also occasionally to be found in this region, at the point toward which the cells of the Fig. 206. — Hair papilla from the upper lip of the fetus of Figs. 202, 203, and 206. M., hair matri.K with spaces at the converging point of the matrix cells; Pa., aniage of papilla; Hba.. con- nective-tissue hair sheath; Bl., vesicular cells of the periderm; V., anterior surface of the fol- licle; //., its posterior surface with the (scarcely recognizable) differentiated arrangement of the nuclei (sebaceous gland and region of the hair swelling). X 430. II. -K. H.-Kan. Fig. 207. — Hair papilla in a more advanced stage, transition into the bulb papilla. Pa., aniage of papilla; 1!'., swelling; T.-Dr., aniage of sebaceous gland; /f.-itan., hair canal cells; //.-i^., tangential sec- tion of a neighboring hair (rerm. X 230. (After Stohr: Lehrbuch der Histologic, Fig. 300; Entwicklung des menschlichen Wollhaares, Fig. 9.) DEVELOPMENT OF TPIE INTEGUMENT, 267 matrix converge, a roundish cavity (Fig. 206). The under surface of the hair papilla is flattened or even somewhat concave. The epidermis in front of the hair is unmodified and three-layered, consisting of periderm, stratum intermedium, and stratum cylin- dricum; the posterior surface consists of proliferated, flatter, and elongated cells (the hair-canal anlage), which in part already show a beginning cornification (Stohr). The hair papilla is enclosed laterally within a layer of connective tissue with numerous cells. On the under surface this is continued into a mass of cells with perpendicularly arranged, concave nuclei, the concavity looking upwards, which are immediately adjacent to the matrix (connec- tive-tissue papiUa). All these peculiarities of the hair papilla may be more or less pronounced. Thus a very distinct swelling may occur on one which does not show any marked apical flattening; or on a very short epidermal papilla there may be a distinct connective-tissue one, or vice versa. Further, in these early stages there may be opposite the swelling- a marked aggregation of nuclei, which is to be regarded as the anlage of the muscidus arrector pili (Stohr). The hair anlagen, almost two months old, on the brows and lips have not advanced beyond this stage when the formation of the anlagen of the general hair coat begins. Some similar structures (gland anlagen) are alone larger; these will be con- sidered in connection with the perimammillary epithelial appen- dages. In the further course of development the layers of the hair and its sheath begin to form (Fig. 208). The epithelial papilla grows longer and thicker below (the bulb-papilla stage, Stohr). Its anterior surface abuts upon thin unaltered epidermis, while the posterior is continuous with the anlage of the hair canal, already recognizable in the two younger stages. This assumes the form of an elongated, partly cornified mass of flat cells, projecting from the under surface of the epidermis behind the hair. The epithelial papilla is still surrounded by a high columnar epithelium, which at two regions on the posterior surface, that of the sebaceous giand above and that of the hair swelling below, projects more markedly than formerly. The anlage of the sebaceous gland does not always lie exactly in the posterior surface of the hair; occasionally it lies more laterally and later may surround the entire periphery of the hair or come to be principally lateral or anterior to it. The specific fat formation begins at an early period in the central cells. The under surface of the epidermal papilla or bulb is at first only slightly concave for the reception of the connective-tissue papilla, but later (apparently very quickly, Stohr) it becomes deeply concave. The high columnar cells which line the concavity, the matrix plate, become the source of an upwardly projecting, conical, pointed mass of cells (the hair cone), which extends up- 268 nmiAx e:mbkyology. wards into the still rather irreg'ularly arranged cell material in the interior of the follicle. The outer boundary of this cone is formed by a layer of cells (the anlage of Henle's sheatli), which extends from the point where the outer surface of the follicle bends into the under surface to the smnmit of the cone. It is the first formed and outermost layer of the cone and arises from the most peripheral cells of the matrix plate ; it also cornifies sooner than any of the other layers of the hair. The inner sheaths and the hair are formed later from the middle cells of the matrix plate. Above the apex of the hair cone tlie cells arrange themselves to form an axial column, which eventually comes into relation with -- GI.-H. Fig 208.— Sheath hair still near the bulb-papilla stage. Pa., papilla; Gl.-H., vitreous layer; W swelling; u ,trr musculus arrector pili; H.-Sch., inner root sheath; T.-Dr., sebaceous gland; W -Sch H7.tnln^° p-^'"qAo -^"f ■ f, *''^ *'"''■ <=™^' °°* y«* differentiated. •. 230. (After Stohr; Lehrbuch der ilistologie, iig. i02; Entwicklung des mensehlichen Wollhaares, Fig. 14.J hair canal cells abutting upon the supei-fieial epidennis and indi- cates the path along which the hair sheaths, and with them the hair, must grow. The connective-tissue portions of the hair become more distmct. The concavity on the under surface of the bulb becomes filled by the large connective-tissue papilla, which consists of abundant, transversely arranged cells and which as vet shows no neck-hke constriction. It seems to be covered by a very thin con- tmuation of the vitreous layer and to be separated by this from the epithelium. From the richly cellular connective tissue of the hair follicle a denser homogeneous layer separates, especially in the region of the swelling; this is the vitreous layer. The mus- DEVELOPMENT OF THE INTEGUMENT. 269 cuhis arrector has at first the form of elongated cells, which show the oblique course from the bulb to the ef»idermis that is char- acteristic for the muscle; it becomes more distinct in this stage. The follicle, gradually enlarging, grows obliquely downwards, and all its constituent parts undergo further development until it becomes the anlage of the actual hair. The sebaceous gland and the swelling assume noticeable dimensions and the connective- tissue papilla increases in height in correspondence with a deepen- ing of the concavity of the matrix plate. While the bulb forces its way downwards the cornified hair sheaths which arise from it are pushed towards the surface. As the last stage, that may be taken in which all constituents of the hair are laid down (the sheath hair, Stohr). The hair elong'ates, especially in that part which lies below the hair bed. The upper part, with the swelling, sebaceous gland, and orifice, grows somewhat with the further development, but in general retains the same proportions. And, furthermore, it is the portion of the hair follicle intervening between the surface of the skin and the hair bed which remains unchanged throughout life, while the processes connected with hair change and the subsequent death of the hair take place only in the portions of the follicle below the hair bed. The outer root sheath, except in its lowest portion, from which the hair and its sheaths are developed, consists of two or three laj-ers of cells, the innermost of which is flattened against the inner root sheath and later is united with it (Stohr). The outer layer consists of more or less high columnar cells, which, in younger follicles, are all directed outwards and downwards (Fig. 209), probably in correspondence with the downwardly directed growth pressure of the follicle; later they become arranged per- IDendicularly to the axis of the hair. Towards the completion of development, when the hair change begins, the columnar cells be- come very high and their nuclei round or hemispherical with the flat surface directed outwards ; they lie in the inner portion of the cells, while the outer portions, which rest upon the Adtreous layer, are clear and unstainable. On the outer surface of these high columnar cells a homogeneous layer is secreted. The portions of this layer, at first separated, fuse together to form the inner vitreous lamella and unite intimately with the outer vitreous layer, which is formed by the innermost layer of the connective-tissue portion of the follicle. Later the two vitreous lamelte become closely connected together. The columnar cells stand in small transverse grooves of the vitreous layer, these grooves corresponding in width with the cell bases and being readily recognizable in microscopic sections, in which the vitreous layer is easily separated from the follicle epithelium, as slight elevations between the rows of cylinder cells. The vitreous layer terminates at the swelling, and above its 170 HUMAN E.T^IBRYOLOGY. H -Ka. T -Dr. upper edge an especially strong layer of elastic fibres lies close against the epithelium and ex- tends upward as far as the sebaceous gland. In the swelling the epithe- lium becomes many-lay- ered, more so on the anterior surface of the follicle than on the pos- terior, which, from the beginning, shows the strongest development of the swelling. At this point the arrector mus- cle, surrounded from below upwards with elas- tic fibres, is usually in- serted. Above the swell- ing the epithelium again becomes thinner and its cells lower, and it is es- p e c i a 1 1 y thin in the region of the sebaceous gland, at w h o s e orifice the hair follicle has its smallest diameter (the isthmus). At tliis point the funnel of the follicle begins, not being formed by a depression from above, but by the cornifi- cation, accompanied by the formation of kerato- hyalin, of the central cells of the follicle and of the h a i r - c a u a 1 cells, whereby a long streak of cornified cells, which ex- tends far into the epi- dennis, is formed. This indicates the path which the hair will shortly take, and, after it has broken through, the funnel becomes usually much widened and its walls strongly cornified. The lowermost part of the outer root sheath encloses the ele- ments of growth for the hair. It has already been seen that in P.h. ' -jS^!^''_ ' Fig. 209. — A completely formed lanugo hair, from the mam- mary region of an eight months male fetus. Partly schsmatic, reconstructed with the camera lucida from an oblique series of sections. P.p., cushion of connective-tissue papilla: P.h., neck of papilla; P.-S., tip of papilla; M ., hair matrix; Gl.-H., ^■itre- ous layer, for the most part separated from tlie outer root sheath; H ., shaft of hair; Tr.-.Sc/^., outer root slieath; Hbg., connective-tissue portion of follicle; S.-D,-., sudoriparous gland; W., swelling; T.-Dr., sebaceous gland; H.~Ka., hair canal. Black, inner root sheath + cuticuloc. X 120. DEVELOPMENT OF THE INTEGUMENT. 271 yoiinger stages Henle's sheath could be followed down to the matrix and was the most external and at the same time the oldest of all the structures which arise from the matrix. It also comities the earliest and the most strongly of all the hair sheaths. Some- where about the level of the connective-tissue papilla (according to Gravazzeni, 1908, even in the matrix cells) there is formed in it a ring of cells containing granules ; the nuclei of these cells do not become smaller, as is nearly always the case in the stratum granulosum, and their granules differ from keratohyahn in both their chemical and staining reactions (staining with eosin and fuchsin). It would seem that these granules, as also those of Huxley's layer and of the hair itself, are not keratohyalin, but a different chemical substance (trichohyalin, Vomer, 1903). The granules of Henle's layer became converted into elongated rods and soon disappear, the layer itself staining diffusely with eosin and becoming completely cornified. Henle's layer covers all the central structures and extends to the region of the hair canal, where it breaks up and is perforated by the hair in its upward growth (Fig. 209). Huxley's layer, the inner lamella of the inner root sheath, cannot be followed quite down to the matrix plate, although it is certain that it has its origin from a definite ring of matrix cells. In the lanugo hair it extends at first far beyond the tip of the connective-tissue papilla into the apparently undifferentiated mass of cells at the base of the follicle, and in later stages of develop- ment its trichohyalin-containing cells can be distinguished further down toward the matrix. Its granule cells extend far upward upon the shaft of the hair, which has formed in the meantime, and they then become cornified. It stands in intimate connection with the Henle layer, which has already become completely cornified, and between the cells of this layer those of Huxley's layer send processes containing granules. According to Garcia (1891) the sheaths can be followed quite to the matrix plate in the head hairs of fetuses of eight to nine months, when the hairs have just attained their complete develop- ment. Of the forty to fifty cells which occur in a longitudinal section through the summit of the matrix, Henle's and Huxley's layers correspond to four to six cells on either side. "Wlien the hair, in its full strength, has broken through the inner root sheath, this terminates in a sharp edge surrounding the hair in a circular manner, at the level of the isthmus below the orifice of the seba- ceous gland ; at first it extends somewhat higher than this, as far up as the hair canal. Beginning somewhat higher than Huxley's layer, there are formed from the more internal regions of the matrix the cuticle of the inner root sheath (the sheath cuticle) externally and the hair 272 HUMAN EMBRYOLOGY. cuticle internally. These two cuticles arise from the four to six cell-rings of the matrix internal to those which form the inner root sheath. Their cells contain no granules (whetlier they are also destitute of granules in the adult condition is still uncertain) , and they cornify probably before Huxley's layer, forming scales which in the sheath cuticle are small and directed obliquely inwards and downwards, while in the hair cuticle they are large and directed obliquely outwards and upwards; in the fully developed follicle the two sets of scales (imbrications) fit into one another. The sheath cuticle is almost inseparable from Huxley's sheath, and the hair cuticle, similarly, from the hair; and their imbrications ap- parently determine the equal ascent of the hair and the hair fol- licle, which their disappearance promptly disturbs (Von Ebner, 1876). The hair itself arises from the large central portion of the matrix plate. It forms at first an acute cone-shaped structure, just as the sheaths do, and is covered by the cornified sheaths as with a cornucopia until it breaks through the torn sheaths in the vicinity of the hair canal. Grradually it becomes broader, and arises from the greater portion of the matrix. On the head (in the eighth to ninth fetal month) each hair arises from twenty- four to thirty of the cells seen in a greatest longitudinal section of the matrix plate (Grarcia, 1891), and Stohr's figures of the lanugo hairs show about the same nmnber. The large roundish nuclei above the matrix gradually become elongated and the cells cornify without forming trichohyalin. The comification begins deep down, the nuclei vanishing at the level of the junction of the lower and middle thirds of the follicle (head hairs, Garcia). Between the matrix cells of the hair branched pigment cells appear, which ascend into the hair and contribute pigment granules to its other cells. According to the most recent views they take tlieir origin from the epithelial rather than from the connective-tissue cells (see p. 253). The diameter of the hair gradually diminishes as it grows away from the papilla, and its smallest diameter is reached with the completion of its comification. But before it becomes hardened, the succulent hair mass is pressed, as in a mould, by the elastic compression of the sheaths. Henle's sheath is a stiff tube, which on account of its net-like structure (flat cells separated by meshes, very well shown by Giinther, 1895, in his Fig. 202), may exercise an elastic pressure. In it the soft hair with its soft sheaths is compressed and moulded. The remaining sheaths form a softer cushion for the forming tube. After the comification of Huxley's layer the hair with the cuticles and the inner root sheath forms a compact cylinder, which is pushed upwards by new forma- tion at the base of the follicle. The inner part of the outer root sheath follows in the upward movement — the imbrications of the DEVELOPMENT OF THE INTEGUMENT. 273 hair cuticle, firmly united to the hair, and those of the sheath cuticle, firmly adherent to the hair sheath, seeming by their inter- lockmg to determine the regularity of this upward growth. The epithelial hair follicle is enclosed in a layer of connective tissue, sharply marked ofP from the surrounding eorium connec- tive tissue, which is, in general, arranged horizontally. This con- nective-tissue sheath consists of an outer layer of longitudinal fibres and an inner transverse or circular layer. The inner layer secretes the outer vitreous layer, whose connection with the colum- nar cells of the follicle by means of the inner vitreous layer formed by their bases has already been described. Beneath the hair bulb, the papilla (papilla pili), which fills the spacious cavity of the hollow hair bulb, projects from the connective tissue. It projects from the mass of transversely arranged cells (the papilla cushion), which lie below the entrance into the hair bulb, and extends some distance upwards to form a papilla terminating above in a point (the tipof the papilla). At the lower border of the hair bulb, the papilla is constricted in a neck-like manner (papilla neck, Garcia). The cells of the papilla are partly directed obliquely upward and partly are arranged transversely, as they are at their first appearance. They are distinguished from all other connective- tissue cells by their epithelial-like structure, being closely set cells with large, roundish or elongated, darkly staining nuclei. The first-formed hairs have only a short life. Even before birth the first hair change begins in the human species. This is total, compressed into a brief space of time, and associated with a change in the quality of the hairs. Some hairs cease to grow even before they have broken through the surface of the skin, a condition which is often shown in later life by lanugo hairs (such as hairs of the face). In hair change the cessation of growth of the hair seems to begin externally and to jaroceed internally. The cells of Henle's layer and then those of Huxley's layer cease to proliferate and are earned upwards by the still gTOwing hair by means of the interlocking of the imbrications. Then the matrix, the cuticles, and the hair itself cease their gTOwth. The hair becomes cornifled right to its tip and, jDrobably because it is no longer compressed by Henle's layer at its lower end, this enlarges to form a brush-like structure, the hair being then known as a hulb hair. A cell mass, the bulb cushion (Garcia), is formed by the matrix and occupies the space left vacant by the hair. This is carried out- wards rapidl3', as if squeezed out from the tube formed by the outer root-sheath. The matrix and connective-tissue papilla pass outwards more slowly, and the outer cells of the outer root sheath lose their columnar foiTn (Aubertin, 1896, in adult head hairs). A diminution seems to occur in the pressure which the growing hair exerts downwardly and which is at first greater than that exercised by the surrounding tissues; the pressure of the tissues is now alone active and the space left vacant is filled by their being forced inwards. During the ascent of the hair and its follicle the connective-tissue invest- ments of the follicle thicken, especially the circular fibrous and the outer vitreous layers. Perhaps these thickened layers exert a compression on the thinning lower Vol. I.— 18 274 IimiAN EMBEYOLOGY. part of the follicle, whereby the hair is forced outwards. But the thickening may also be regarded as a protection against the pressure of the surrounding tissue, or as the simple contraction of aii overstretched membrane. Indeed, all mechanical theories are to be advanced with great caution (Stcihr), since in every step the growth of the hairs apparently follows old inherited paths which, like the phylogeny of the hair, are unfortunately poorly understood. All arrangements are naturally intelligible mechanically and explicable as strain and pressure conditions; but whether these mechanical explanations are correct is a question. While the separated hair and the papilla ai-e ascending the epithelial root cylinder between the two becomes thinner; it becomes composed of cubical in- different cells and the connective-tissue papilla becomes smaller (diminishing at the most to about half its original size in section). When the papilla has reached its highest position, a new life begins in the root cylinder. It covers itself anew from above with new columnar cells, becomes thicker, and develops a new swelling- like outgTOwth, which later applies itself to the old swelling (Garcia). Gradually a new hair papilla forms in this, the hair matrix producing first a new inner root sheath and then a new hair, just as on the first formation of the hair. As the new hair grows out, its matiix and papilla are forced downwards by the renewed growth, and the inner root sheath is broken through just below the orifice of the sebaceous gland. The tip of the hair pushes its way, frequently in a tortuous course, through the old follicle canal, and after a considerable enlarge- ment of this the old hair, which projects considerably, eventually falls out, as the result of some mechanical cause. The new hair is no simple replacement of the old, but has a quite different character. While the hairs of the first generation are practically all alike, there begins to appear in the second generation the great difference between the head and body hairs, and this increases in generation after generation, until, at the beginning of puberty, the genital and axillary hairs begin to differ from the lanugo of the remaining portions of the body; the lanugo of tlie face gives place to the hairs of the beard, etc., in the male; and the apparently unaltered lanugo, as well as the head hairs, eyelashes, and eyebrows, assume a dif- ferent type. In later years still more of the lanugo becomes transformed into strong body haire (terminal hairs, Friedenthal, 1898). Each change of hair is at the same time a change in the character of the hair (type change, Unna, 1893). An accurate enumeration of the lanugo hairs on the human ear (Oshima, 1907) seems to show that the number of the fetal lanugo hairs is in places much greater than that of the lanugo of the adult. Tlie direction of the hairs is determined from the beginning. In the individual hairs it is recog-nizable in the hair-germ stage on account of the bilateral form of the germ, and in later stages it reveals itself by the hairs spreading out over the surface in the manner permanently determined for them (Blaschko) and not radially from some centres that arise. The same conditions have already been described as occurring in the development of the dermal ridges, which, in the same way, extend out over the surface of the skin from the regions of their first formation (the finger tips). AVith the completion of the hair formation the skin shows an arrangement of the hairs which is definite, unchangeable in any individual, and varying but slightly in different individuals, for the knowledge of which we are indebted to Eschricht (1837) and especially to Voigt (1857). DEVELOPMENT OP THE INTEGUJMENT. 275 The latter regarded the direction of the hairs as the result and an in- dication of the mode of growth of the skin especially, but also, to a certain extent, of the underlying portions of the body. The general spiral arrangement of the lines of the larger hair streams indicated a spiral growth of the enlarging skin, such as is the rule in plants, a mode of growth which was thoroughly studied by Ohlert (1854-55) and later by Schwendener (1909), and was regarded as a peculiar law of growth for the animal body, the law of torsion, by Fischer (188G). The centres of the spirals are the hair whorls, around which the hairs arise at intervals and in curves which are regular even although they have not hitherto been expressed mathematically. Voigt pointed out that the hairs at the ends of a stream are further apart than they are near the whorl. A considerable number of constantly occurring hair centres have the form of whorls, from which the hairs diverge in spirals (the direction of the free hairs being towards the periphery, diverging whorls). The more important of these are: 1. The crown or vertex whorl, curving towards the right in more than half the cases, towards the left in about a third, and doubled in the remaining eases (curving to the right on the left side and towards the left on the right side), or, in rare eases, trebled. 2. A right and a left brow whorl. 3. A right and a left ear whorl. 4. A right and a left axillary whorl. 0. A right and a left lumbar whorl. 6. Occasionally one or frequently two whorls, right and left at the side of the body (often only on one side). 7. Hand and foot whorls. In other regions the hairs converge from all directions to form converging whorlS; the most constant of which are : 1. The frontal whorl, at the root of the nose or at the edge of the scalp, or in both places. 2. Lateral cei-vical whorls. 3. Elbow whorls. 4. Umbilical whorl. 5. Penial whorl. 6. Coccygeal whorl. The hair lines meet one another at acute angles in streams; when they meet at right angles crosses are formed, such as the nasal cross, the hyoid cross, the pectoral cross, the abdominal cross, the penis cross, and the coccygeal cross in the middle line of the body; the brow crosses, the nape crosses, the supra-auricular crosses, and the lateral crosses, one on each side; and the shoulder crosses, the ulnar crosses, the carpal crosses, and the crural crosses on the extremities. Many are doubled, and very frequently one is absent on one side of the body. The diverging whorls, according to Voigt, are regions of least growth, of comparative rest; the converging whorls correspond to especially gi-eat stretching of the skin, in regions where (either in the ontogeny or phylogeny) some organ projected from the body (Wiedersheim: penis, umbilicus, branchial clefts, and, in animals, horns), or where especially strong growth resulted from the pressure of adjacent parts (coccyx, elbow). The crosses are regions of relative rest, lymg between forces acting from either side; the converging hair streams are regions which became stretched during growth. Of the an-angement of the hairs in transverse rows, correspondmg to the arrangement of the scutes in transverse girdles around the body and limbs of reptiles, and of the development of such rows, we have as yet no comprehensive investigation. 276 HUilAX EilBRYOLOGY. G. THE SUDORIPAROUS GLANDS. The development of the sudoriparous glands begins on the finger tips (Blaschko, 1888), the palms of the hands, and soles of the feet (Grefberg, 1883) in the fourth month; according to Kolliker (1889) in the fifth month. It follows immediately upon the formation of the dermal ridges. Their anlagen resemble closely those of the hairs, except that they lack the close aggregations of cells in the corium, from which, in the case of the hairs, the papillaj are formed. The anlagen project downwards as solid flask-shaped rete papilla?, which, becoming long and slender, begin to become tortuous in the sixth month. In the seventh month a lumen begins to form in each, l^he beginning secretion producing intercellular clefts which later unite to form a continuous cavity. In the mean- time the lower end, bending upon itself, forms the anlage of the coiled portion of the gland. The outer terminal portion of each gland forms its own lumen, which later Unites with that of the glandular portion. The two-layered epithelium of the duct portion becomes transformed at its passage into the glandular portion into two distinct layers; the inner of these remains the large-celled secretory layer, while the outer becomes flattened and forms the epithelial muscular layer (Kolliker, 18S9). The gland canal is then composed of an inner layer of cubical, or even higher, glandular cells, with large, round nuclei, and of an outer layer of flat muscle cells, whose nuclei are flattened and whose angles pro- ject between the cells of the inner layer; these cells form an outer investment of closely set parallel striae around the gland. At the time of birth the sudori])arous glands, like the hairs, seem to be completely laid down, so far as their number is concerned. In the hairless palms of the hands and soles of the feet it is certain that the sudoriparous glands arise from the surface epi- dermis, and in most of the other portions of the skin they seem to liave a similar origin in man. In many places, however, a rela- tion of the sudoriparous glands to the li^ir follicles, of general occurrence in animals, persists, the sudoriparous glands partly developing directly from the uppermost jwrtions of the hair fol- licles ("Wimpfheimer, 1907; Diem, 1907), or partly having at least a connection with its follicles at their mouths (Fig. 209). The idea that the follicles and the sudoriparous glands belong to genetically- single areas has been somewhat generally accepted; according to my observations the hair disks (not constant in their occurrence) also belong to these areas, each of which, with all its appendages (vessels, ner\'es, muscles) corresponds in its original fomi to a promammalian scute (scute area or hair area). According to this view the completely isolated sudoriparous glands of the palms and soles must each represent the remains of a hair area (Whipple). That in spite of an imperfect development of the hairs the sudoriparous glands may actually retain their proper places in the hair areas, I have been able to show in the sole of the foot of DEVELOPMENT OF THE INTEGUMENT. 277 Ornithorhynchus (Pinkus, 1905) where the paradoxical arrangement of a complex consisting of a sudoriparous gland behind, a simple hair follicle in the middle, and a hair disk anteriorly, can only be explained as a semicircular arrangement of the elements of the hair area, peculiar to Ornithorhynchus, and a disappearance of all the hairs with the exception of the middle one. A further small step in the reduction would leave nothing remaining but the sudoriparous gland. The con- nection of the sudoriparous gland with the hair follicle, regarded as merely a topographical relation by Maurer (1895), but by most authors {De Meijere, 1894; Eggeling, 1904) as genetic, has been demonstrated to be of the latter nature by the embryological observations of Stohr's pupils, Wimpfheimer and Diem, who found that in the majority of the mammals examined the sudoriparous glands arise from the follicle epithelium and their orifices only later migrate to the surface epidermis. Usually the sudoriparous gland is formed, like the other appendages of the hair (sebaceous gland, swelling, muscle, hair disk), on its posterior surface. It begins to form even in the hair germ stage and becomes distinct in the papilla stage (although not visible in man). Its cell nuclei, in contrast to those of the col- umnar layer of the hair, are small (Eggeling) ; in contrast to the regular an'angement in the anlagen of the seba- ceous glands they are irregular (in the mole) ; and the increase in the number of the connective-tissue cells, which usually begins early in the case of the hair germ, is wanting beneath them. Unfortunately, in man, as well as in many other mammals, the original mode of development of the sudoriparous glands cannot be obser^'ed. In these forms they appear to ai-ise, for the most part, not from the follicle epithelium, but from the surface epidermis. H.-Ka. Fig. 210. — Human sudoriparous gland open- ing into a hair follicle, from the axillary region. T.-Dr., sebaceous gland {the lower leader does not extend quite clearly enough to the gland); S.-Dr., sudoriparous gland; H.-Ka., hair canal. X 165.' After Stohr, from Diem: Entwicklung der Schweiss-drusen an der behaarten Haut der Saugetiere, Fig. 7. Departures from the usual structure of the sudoriparous glands occur in certain regions of the body. If we make exception of the eyehds with their speciahzed glands, these regions are the mammary and axillary regions, the inguinal folds with the scrotum, and the anus. In some cases the characteristic peculiarities are recognizable in the anlagen (region of the mammary gland), in others they first appear at puberty (the axillary glands). . ,. i A The Glands of the Mammary Region.— The ventrolateral surfaces of the embryo at an early stage (6.25-6.75 mm., Keibel and Elze, 1908, " Normentaf el, " Nos. 21, 24, and 25, Figs. 11 and 12) are occupied by a broad diffuse area of high epithelium (Schivalbe's milk streak), that has a variable development, ex- tending in some cases forward over the branchial arches and back- wards, over the limb buds, until it reaches the tail ; but m otlier cases it is of less extent and may be completely wanting. This 278 HU.MAN EMBRYOLOGY. epithelial thickening represents a formative region, which occurs also in the lower mammals and in birds (Heinrich Schmitt, 1898). In it, especially in its anterior portion, there develops in embryos 9 mm. in their greatest lengtli (" Normentaf el, " Nos. 37-39) the milk Hue or milk ridge (0. Schultze, 1892 ) , a band of thickened epithelimn which is ap- proximately lenticular in transverse section. The milk ridge is a structure comparable to the gan- glionic or dental ridges, glandular anlagen appear- ing in it from place to place (0. Schultze, 1897; Brouha, 1905). In it the anlage of the mammary gland develops in man. Its primary epithelial an- lage is, according to Rein (1882), (a) mound-shaped, consisting of an aggregation of epithelial cells projecting beyond the surface of the epidermis. Even in embryos with a greatest length of 9.5 mm. it .1.- ^-"v ,^^'^'\'if '••'.•.•..f'lr-f'" .<■■.■''.• ^' Tig. 211. — Section through the middle of the mammary an- lage in a human fetus 8.5 cm. in length, male. X 120. N.-MDr. N.-Dr ''■ ?^>:i .. u . '-*i3a2" '■'^Sf^rt-' Pig. 212. — Section through the mammary gland of an eight months male fetus. O., opening of the nipple; Mg., milk duct; N.-MDr., accessory milk glands (modified sudoriparous glands); F., fat lobes. X 44. has become (6) flatly lenticular, projecting somewhat beyond the epithelium both above and below, and being surrounded by an aggregation of corium cells (nipple soiie). As the result of strong downward growth there is formed (c) the papilla-shaped anlage (Fig. 211), from which (d) the bulb-shaped anlage is formed as DEVELOPMENT OF THE INTEGUMENT. 279 the lower parts increase in breadth while the more superficial por- tion becomes constricted in a neck-like manner. Later the milk ducts begin to sprout out, the anlage becoming polygonal and beset with short buds, (e) the period of bud formation. The develop- ment progresses slowlj^ Several simple papillas project down- wards into the connective tissue from the epithelial mass, and in the eighth month these become hollow and somewhat branched, their terminal portions being large (Fig. 212). The primary epi- thelial bulb from which these milk ducts arise becomes cornified in its central portion and a cavity forms in it, with which the lumma of the gland ducts become continuous (Fig. 212). When the hair development begins the region around the mammilla becomes conspicuous, on account of the absence of hairs on its surface, and forms the nipple area (0. Schultze, 1897). In addition to the paired mammte, supernumerary mammary glands are frequently developed. As a rule, they occur along a line from the anterior axillary fold to the ingniinal region and make it seem as if several mammary glands had formed from the milk ridge ; in other cases the hyperthelial structures are arranged irregularly in the mammary region. The anlage of the mammary gland is usually compared with those of the sudoriparous glands. In the monotremes the glands of the mammary pouches in their functional condition differ from the ordinary sudoriparous glands only in their exceptional size (Gegenbaur, 1886, who, however, originally derived the mammary glands from sebaceous glands). Their genetic origin from sudoriparous glands is shown both by their development and by their comparative anatomy. Futhermore, in the adult condition their hidradenoid structure, represented by the two-layered epithelium of the ducts and the simple epithelium in the glandular alveoli, is an indication of their sudoriparous character. Their glandular alveoli, like those of the sudoriparous glands, produce their secretion without destruction of the cells (Bertkau, 1907) and are enclosed within a muscular network (Benda, 1893), which resembles the basket-like muscular net- work of the sudoriparous glands in the snout of many animals (Kormann, 1906). Their similarity to sudoriparous glands is rendered still greater Ijy the fact that such glands, modified along the lines of the mammary glands, are formed in the neighborhood of these structures. Rein (1882) has already identified these as the anlage of the Montgomeiy's glands. Close around the nipple a number of sudoriparous glands is fomied from a primary epithelial anlage similar to that of the mammary gland itself, but smaller; they resemble milk ducts by possessing peculiar outpouchings and wide lumina. Occasionally sebaceous glands and small hair anlagen are to be seen arising from the same eisithelial papillae, so that the hair areas, whose remains are represented by the accessoi-y mammary glands (Eggeling, 1904), are in these cases not absolutely rudimentary (Brouha, 1905). In rather early stages (embryos of 15.5 mm., Walter, 1902) some other •epithelial appendages begin to develop around the mammaiy glands and were at first regarded as hyperthelial stmctures (accessory mammai-y glands, Hugo Schmidt, 1896). From another standpoint they were interpreted as marsupial anlagen (Walter), similar to the marsupial pouches which form around each mammary orifice in the opossum and fuse by their outer ends on the appearance of the marsupium (Bresslau, 1902). The number of these anlagen is occasionally very 280 HUilAX EMBRYOLOGY. large in human embryos, reaching in some eases forty; they are scattered around the nipple and as far up as the axilla, and are usually smaller than the milk o-lands proper. Up to the present they have not been found in fetuses longer than 60 mm. To the same eategoiy perhaps belongs also the mgumal epidermal anlage, which has been described by Brugsch and Tnger (1903). In the same situation as these epithelial thickenings I have found around the mammary gland during my preparation for the present work, in a fetus of 85 mm., five epithelial structures, which must be regarded as something quite distinct. They are long tubules, surrounded by a thin longitudinal layer of con- nective tissue and formed of a two- to four-layered epithelium, which bounds a small lumen, not opening to the surface. Near the epidermis the most superficial Fig. 213. -Epithelial appendage occurring near the mamma of the fetus from which Fig. 211 was taken. //"., cornified plug with keratohyalin; Z/., lumen; Af., muscle (?J. X 180. portion of the tubule contains a small cavity, filled with comifled cells and sur- rounded by keratohyalin cells; it resembles Stiihr's hair canal, and neither opens to the exterior nor communicates with the lumen of the tubule. Below, the tubule, beyond the termination of its lumen, ends in an epithelial thickening, which sometimes forms a very regular, almost conical structure, covered externally by columnar cells (Fig. 213). These epithelial tubules are, with the exception of the mamma, the lai-gest epithelial appendages of the fetus. At this stage the hair anlagen are still in the stage of the much shorter bulb papilla and the sudoriparous glands are not yet formed. I would assign the tubules without question to the mammary apparatus, were it not that similar, longer structures occurred in the nasal mucous membrane (Fig. 214). Possibly those in the pectoral region are a form of the further development of the epithelial structures of Schmidt (1896), mentioned above; probably they are forerunnei-s or early stages of the Montgomeiy glands of the nipple area. That they are glands of the sudoriparous system is shown by their complete similarity to the ciliary sudoriparous glands, IMciU's DEVELOPMENT OF THE INTEGUMENT. 281 glands, whose development from the epithelium of the hair follicles and occasional occurrence quite separated from these has been described by Ask (1908). The ciliary glands are, at this stage, however, younger, and only in fetuses of 170-250 mm. show the grade of development in these structures seen in 85 mm. fetuses. On the same side of the ciliary follicle as the sebaceous gland and swelling (the outer surface of the eyelid, but the posterior surface of the cilia) there extends above the level of the sebaceous gland a thin elongated epithelial cord, which, after passing over the sebaceous gland, bends downwards and is continued down- wards for some distance parallel to the anlage of the cilium, to terminate in a pear-shaped enlargement composed of small rounded epithelial cells. Later a lumen appears at the orifice and another in the coiled portion of the gland. The connective-tissue cells around the anlagen of Moll's glands are not, however, in- creased in number. Fig. 214. — A similar epithelial structure from the nasal cavity of the same fetus. X 180. B. The further development of the axillmy glands begins in the ninth year in the female, but not until the time of puberty in the male (Liineberg, 1902). They are formed from the ordinary sudoriparous glands of this region, a large number of which per- sist as small glands. The large axillary glands form a continuous layer of large, partly branched tubules, with an inner layer of high secreting cells and an outer single-layered muscle coat. Nothing special is known concerning the development of the inguinal and scrotal glands. H. THE NAILS. At a very early period the place where the nail will be formed is recog-nizable on the dorsum of a terminal phalanx. The primnry nail field (KoUiker, 1888), primary nail base (Zander, 2S2 HUirAN EMBEYOLOGY. 18S4), is recognizable in microscopic sections in fetuses of -4.5 cm., wMle'in tliose of 2.75 cm. the region in which it will develop is quite undifferentiated (Okamura, 1900). Externally, the primary nail field somewhat later becomes evident on account of its smooth appearance and its firmer adherence to the subjacent tissue, as well as by its sharp anterior, posterior, and lateral boundaries. It extends from the finger tip almost to the articular process of the terminal phalanx — an extent which the fully formed nail also pos- sesses and which is only relatively diminished transitorily during development by the increased growth of the ball of the finger. At its proximal (posterior) edge the epidermis invaginates to form a transverse, posteriorly convex pouch (the root lamella of Kolliker, the posterior nail fold of the adult), in which later the nail matrix is formed and whose roof is termed the nail wall. The epithelial invagination is the posterior liiiiiting furrow (Kol- liker), and laterally, on each side, it is continued into a lateral groove (the lateral nail fold, bounded externally by the lateral nail wall). Anteriorly the primary nail field abuts upon a shallower depression, which bounds it anteriorly in a transversely arched manner (the anterior groove, Unna ; nail fringe, Kolliker). The primary nail field is flatly convex and is at first distinguished from the two- to three-layered skin of the fingers by its three- to four- layered epithelium with a germinal layer of cubical cells. Later on, also, the cells of the lowest layer of the nail field (the later nail bed) remain cubical, in contrast to the columnar cells of the nail fringe, the nail fold, and the germinal layer of the rest of the epidermis of the finger. In front of the nail fringe a transverse ridge appears, which is of great importance from the comparative standpoint. It corresponds to the region of the skin in which later the looser cornification, seen under the free edge of the nail and equivalent to the sole plate of the other mammals, is formed (Boas, 1894). In front of this ridge the epidermis is depressed to form a groove (the distal limitiug furroiv, Kolliker). The germinal layer in this region is formed of high columnar cells and the epithelium is five- to six-layered. In front of the sole plate the germinal layer becomes still more columnar. The periderm and the stratum intermedium extend over the entire anlage, levelling its surface. The most superficial layer of veiy flat cells desquamates to a remarkable degree ; and one finds, still adhering to the surface, cells swollen to a vesicular form and with small pyknotic, darkly staining nuclei (for a description of these cells see especially Zander, 1884; Kolliker, vesicular cells; Okamura; for those of the chick embryo, Eosenstadt, 1897). In the course of further development the region of the sole plate becomes higher and eight- to ten-layered, and the cells of the uail field become flat, so that bv its cell form alone the dorsal sur- DEVELOPMENT OP THE INTEGUMENT. 283 face of the terminal phalanx can be readily distinguished from the volar surface, the germinal layer of which remains columnar. Only in the neighborhood of the nail wall, and, later, still further proximally also large vesicular, round cells persist at the surface, to a greater extent on the toes than on the fingers. The flat peri- W. N.-F. di. Gr.-F. Fig. 215. — Median longitudinal section through the index finger of a human fetus of 8.5 cm. N.-F., nail fold; N .-W ., nail wall; pr. Gr.-F., proximal limiting furrow; Ke., Iceratohyalin layer; N.-S., nail fringe; di. Gr.-F.. distal limiting furrow; L., commencing dermal-ridge formation. X 44. derm cells, beginning at the nail fringe, fuse to form a tough super- ficial layer, and between them and the cubical germinal layer there lie distally several layers of prickle cells and proximally, at the nail fold, an area of pale vesicular cells (Zander's limit- ing layer). These vesicular cells remain large imtil much later on the toes, but on the fingers they become flatter, probably in cor- respondence with the greater general flattening of the anlagen of the fing'er nails. Below Zander's limiting layer there is formed from the nppermost portions of the stra- tum intermedium, in somewhat larger embryos (10 cm.. Zander; 9.3" cm., Kolliker; 8.5 cm., ac- cording to my own observa- tions), a stout layer with abun- dant keratohvalin granules, which are partly large and roundish and partly quite fine and dust-like. Beneath this granule layer, whose keratohyalin nature was first determined by PoUitzer (1889), there lies a layer of pale vesicular cells, which also, espec- ially at the cell peripheries, contain granules similar to those of the superjacent laver. These three layers form Unna's eponychium (1883), and the nail itself is formed from them according to X.-F. T.-B. Fig. 216. — Transverse section through the ter- minal phalan.'c of another finger of the same fetus from which Fig. 215 is taken. The basal cell layer is cubical dorsally, columnar volarly. N . F., lateral nail fold; Ke., keratohyalin layer; T.-B., touch ball. X44. 284 Hu:\iAX e:mbryology. KoUiker, while according to Zander it is formed from tlie limiting layer, wMcli is contimially reinforced from the subjacent kerato- hyalin layers. The eponychium with its granule layer represents, according to Zander, only a precocious cornification, of the same type as that which occurs in the epidermis over the whole body at a later stage, when the granule layer, at the first sign of cornifi- cation, forms from the layer of prickle cells. This kind of cornifi- cation is found in such early fetal stages not only in the nail anla- gen, but in all places where large appendage structures of the skin are being formed. It occurs also at the openings of the peculiar long epithelial appendages to be found in the nasal mucous mem- brane and around the mammary glands, and it occurs in slightly later stages in the development of the hairs, in the anlagen of the hair canals. It appears to be merely an indication of increased fonnative force in the epidermis. This layer of actual cornifica- tion, already similar to the later cornification, cannot, however, be directly identified with the periderm (eponychium). This is an actual horn layer, which remains partly retained throughout the rest of life (as the fringe at the proximal edge of the nail; see below). Okamura (1900) is correct when he regards as the peri- derm merely the outermost layer of desquamating cells, which lies over the keratohyalin layer, for we have already seen, in the consideration of the general skin, that the periderm in man is not a special layer, but merely the outermost layer of the quite young epidermis, which in the later development of the skin is perma- nently cast off. The nail itself is formed from the region of the nail fold, quite without keratohyalin formation, in the manner described by Unna (1883) and confirmed by Okamura (1900) and Apolant (1901). The actual nail formation begins at the posterior part of the nail field, under the eponychium, as a layer of cornified cells. It begins in the fifth month (Unna), in fetuses 17 cm. in length, the first indications of its cells being visible in those of 16 cm. (Okamura, 1900). These cells lie at the entrance of the nail fold, at the level of the distal end of the nail wall. They present fine granulations, which do not take the stains which color keratohyalin, but stain with picric acid, swell in strong alkali, and are insoluble in all ordinary reagents such as water, alcohol, xylol, chloroform, ammonia, acetic, hydrochloric, and nitric acid, alkalies, and digestive ferments ; therefore they cannot be keratohyahn. They have been termed by Eanvier onychogenous substance; but they neither represent a special substance (onychin), nor are they really granules, but probably nothing but the cross sections of fibrillag. They consist of keratin and are the cross sections of Jceratin fibrils which occur in the nail cells (Unna, Apolant). The cornification takes place, without being accompanied with DEVELOPMENT OF THE INTEGUMENT. 285 a foi-mation of keratohyalin, in the cells of the nail matrix, which extends from the posterior (proximal) margin of the fold to the anterior (distal) border of the lunula. The cells flatten and be- come converted into platelets, which compose the solid substance of the nail. The formative region for this reaches, with advancing development, from the original point to the deepest part of the posterior nail fold, and anteriorly to the anterior border of the lunula. Over this area the nail is formed and pushed forward Fig. 217. — Longitudinal section through the keratohyalin layer of the fetal nail bed, anterior part. From a human fetus of 8.6 cm. (see Fig. 215). X 380. from the beginning, just as it is throughout the whole of later life. The newly formed nails are for some time completely covered by eponychium, but later this is thrown off and the surface of the nail is exposed. Nevertheless a portion of the eponychium per- sists throughout life as a small membrane resting upon the proxi- mal end of the nail ; this grows forward with the nail and sooner or later desquamates at a distance of from 1-3 mm. from the pos- terior nail wall. The anterior edge of the nail is at first very thin -^^i-^ Fig 218 — Transverse section through the keratohyalin layer of the nail bed, posterior part. From a human fetus of 8.8 cm. (see Fig. 216). X 380. and projects in older fetuses and after birth at the anterior end of the phalanx as a delicate membrane, but it is very quickly re- moved by the accidents of extra-uterine life (washing, movements) (Unna). The nail then grows, becoming gradually stronger, throughout the rest of life continuously, unless interrupted by some pathlogical condition. The portion of the skin on which the nail rests in front of the lunula takes no part whatever in the forma- tion. Its rete papillee, which form at the same time as the rete papilte and ridges of other regions, produce the longitudinal ridges of the nail. ^ 2&6 IimiAX EMBRYOLOGY. BIBLIOGRAPHY. Adachi, B.: Hautpigment beim Menschen und beim Affen, Anatom. Anzeiger, vol. xxi, 1902, p. 16. Apolant: Ueber den Verhomungsprozess, Arch. f. mikr. Anat., vol. Ivu, 1901, p. 766. Aecangeli, a.: Einige histologische Beobachtungen iiber das Deckepithel des Oesophagus beim Meersehweineheii (mit besonderer Beriieksiehtigung des Keratinohyalins), Monatsh. f. prakt. Dermatol., vol. xlvii, 1908, p. 297. Ask, F. : Ueber die Entwieklung der Lidrander, der Tranenkarunkel und der Nickhaut beim Menschen, nebst Bemerkungen zur Entwieklung der Triinen- ableitungswege, Anat. Hefte, No. 109, vol. xxxvi, part ii, 1908, p. 189. Aubeetin, G. : Das Vorkommen von Kolbenhaaren und die Verdnderungen derselben beim Haarwiederersatz, Arch. f. mikr. Anat., vol. xlvii, 1896, p. 472. Beard, J. : The Birth-Period of Trichosurus vulpecula, Zool. Jahrb., vol. xi, 1908. Benda, C. : Das Verhaltnis der Milchdriise zu den Hautdriisen, Dermat. Zeitsehr., vol. i, 1893. Bertkau, F. : Ein Beitrag zur Anatomie und Physiologie der Milchdriise, Anat. Anz., vol. XXX, 1907, p. 161. Bjorkenheim, E. A. : Zur Kenntnis des Epithels im Uterovaginalkanal des Weibes, Anat. Anzeiger, vol. xxviii, 1906, p. 447. Bizzozero : Miinchener med. Wochenschr., 1908. Blaschko, A. : Beitrage zur Anatomie der Oberhaut, Arch, f . mikr. Anat., vol. XXX, 1888, p. 495. Boas, J. E. V.: Zur Morphologie der Wirbeltierkralle, Morph. Jahrb., vol. xxi, 1894, p. 281. Bolk, L. : Ueber die segmentale Anordnung der Melanoblasten bei jungen Teleos- tieren. Verb. d. Anat. Gesellsch., Berlin, 1908, p. 135. Bonnet, R. : Haarspiralen und Haarspindeln, Morph. Jahrb., vol. xi, 1886, p. 220. Bresslau, E. : Beitrage zur Entwicklungsgeschiehte der Mammarorgane bei den Beuteltieren, Zeitsehr. f. Morphol. u. Anthropologic, vol. iv. No. 2, 1902. Brouha : Sur la bande et la erete mammaires et sur les pretendues ebauches hyper- theliales chez I'homme et le murin, Anat. Anzeiger, vol. xxvii, 1905, p. 462. Sur la signification morphologique de la mamelle, Anat. Anzeiger, vol. xxvii, 1905. p. 311. Brugsch, Th., and Unger, E. : Ein warzenformiges Gebilde der vorderen Bauch- wand bei einen menschlichen Embryo von 44 cm. Scheitelsteisslange, Anat. Anzeiger, vol. xxiii, 1903, p. 216. Brunner : Ueber Glycogen in der gesunden und kranken Haut, Deutsche Dermatol. Gesellsch., Bern, 1906. BuBKAED, 0. : Ueber die Hautspaltbarkeit menschlicher Embryonen, Arch, f . Anat. u. Physiol., Anat. Abth., 1903, p. 13. Buschke : Notiz zur Behandlung des Vitiligo mit Licht, Mediz. Klin., No. 33, 1907. Buschke and Mulzer: Weitere Beobachtungen iiber Liehtpigment, Berlin klin. Wochenschr., No. 49, 1907. Brzzi : Keratohyalin und Eleidin, Monatsh. f . prakt. Dermatol., vol. viii, 1889. Ueber Eleidin, Monatsh. f. pl-akt. Dennatol., vol. xxiii, 1896. Cedeecreutz : Ueber die Yerhornung der Epidermis beim menschlichen Embryo, Arch. f. Dermatol, u. Syphil., vol. Ixxxiv, 1907, p. 172. Charpt, a. : Les plis de la peau, Arch. med. de Toulouse, Ref . Zentralbl. f . norm. Anat., 1905, p. 335. CiLiANO, p.: Eleidin, Monatsh. prakt. Dermatol, vol. xlvi, 1908, p. 435. Cone, Claribel: Zur Kenntnis der Zellveranderungen in der normalen und pathologischen Epidermis des Menschen, Frankfurter Zeitsehr. f. Pathol., 1907, p. 37. DEVELOPMENT OF THE INTEGUMEKT. 287 CzAjEwicz: Mikroskopische Untersuchungen iiber die Textur, Entwicklung, Eiiekbildung und Lebensfiihigkeit des Fettgewebes, Arch. f. Anat u Physiol, 1866, p. 289. Diem, F. : Beitrage zur Entwicklung der Schweissdriisen an der behaarten Haut der Saugetiere, Inaug. Dissert., Wiirzburg, 1907. Dreysel, M. : Pig-ment und osmierbare Substanzen in der menschlichen Haut, Verb. d. V. Deutschen Dermatologen-Kongresses, 1895. Dreysel, M., and Oppler, P.: Beitrage zur Kenntnis des Eleidins in normaler und pathologiseh veranderter Haut, Arch. f. Dermatol, u. Syphil., vol. xxx, 1895. Von Ebner, V. : Mikroskopische Studien iiber Waehstum und Wechsel der Haare, Sitz. Ber. d. K. Akad. d. Wiss. Wien., vol. Ixxiv, 1876. Eggeling, H. : Ueber die Stellung der j\Iilchdriise zu den iibrigen Hautdriisen, Semon's Forschungsreise, Jena, 1895. Ueber ein wiehtiges Stadium in der Entwicklungsgeschichte der menschlichen Brustdriise, Anat. Anzeiger, vol. xxiv, 1904, p. 595. Ehrmann, S. : Das melanotische Pigment und die pigmentbUdenden Zellen des Mensehen und der Wirbeltiere in ihrer Entwicklung, nebst Bemerkungen iiber Blutbildung und Haarwechsel, Bibliotheca medica, vol. ii, part vi, 1896. Ernst, P. : Studien iiber normale "\'erhornung mit Hilf e der Gram'schen Methode, Arch. f. mikr. Anat., vol. xlvii, 1896, p. 683. Eschhicht: Miiller's Arch. f. Anat. u. Physiol., 1837 (cited by Voigt). EvATT, J. E. : The Development and Evolution of the " Papillary " Ridges and Patterns on the Volar Surface of the Hand, Journ. Anat. and Physiol., vol. xli, 1907, p. 66. FiCK, J. : Ueber metachromatische Farbung des Keratohyalins durch Cresylecht- violett, Centralbl. f. allgemein. Pathol., 1902, p. 987. Fischer, E. : Das Drehungsgesetz bei dem Waehstum der Organismen, Strassburg, published by the author, 1886. Flemming: Centralbl. f. d. med. Wissensch., July, 1870. Beobaehtungen liber Fettgewebe, Arch. f. mikrosk. Anat., vol. xii, 1876, p. 434. Ueber die Entwicklung der Fettzellen und des Fettgewebes, Arch. f. Anat. u. Physiol., 1879. Frederic : Zur Kenntnis der Hautfarbe der Neger, Zeitsehr. f . Morphol. u. AnthropoL, vol. ix, 1905, p. 41. Friedenthal, H. : Beitrage zur Naturgeschichte des Mensehen (Lief. I, Das Wollhaarkleid des Mensehen; Lief. II, Das Dauerhaarkleid des Mensehen; Lief. IV, Entwicklung, Ban und Entstehung der Haare), 1908 (with a most complete bibliography). Gage, S. H. : Glycogen in a 56 Days Human Embryo and in Pig Embi-yos of 7-10 mm., Amer. Journ. Anat., vol. v; Proe. Assoc. Amer. Anat., 1906, p. xiii-xv. Garcia, S. A.: Beitrage zur Kenntnis des Haarwechsels bei menschlichen Embryonen und Neugebomen, Morph. Arbeiten, vol. i, part ii, 1891, p. 136. Gavazzeni, G. a.: Trichohyalin, Monatsh. f. prakt. Dermatol., vol. xlvii, 1908, p. 229. Gegenbaur: Zur Morphologie des Nagels, Morph. Jahrb., vol. x, 1885, p. 465 (cited by Unna, 1889). Gemmil: Notes on the Origin of Elastic Fibre in Tendon, Journ. Anat. and Physiol., vol. xl, 1906, p. 397. Grepberg," W. : Die Haut und deren Driisen in ihrer Entwicklung, Mitth. a. d. embryol. Institut zu Wien, vol. ii, part iii, 1883, p. 126 (cited by Blaschko). Grimm : Pigment in der Haut, Dermatol. Zeitsehr., vol. ii, part iv, 1895. Grosser, 0. : Die Metamerie der Haut, Sammelreferat, Centralbl. f. die Grenzgebiete der Med. u. Chir., vol. vii, Nos. 1, et seq., 1904. 288 HU-^IAX EMBRYOLOGY. Geossee, 0., and Feolich, A. : Beitriige zur Keiintuis der Dermatome der mensch- lichen Rumpfhaut, Morph. Jalirb., vol. xxx, 1902. Geund: Experimentelle Beitrage zur Geiiese des Epidermispigments, Ziegler's Beitrage, Suppi. 7, 1905. GuNTHEE, M. : Haarkiiopf und iimere Wurzgelsclieide des Saiigetierhaares, Inaug. Dissert., Berlin, 1895. Head, H.: Die Sensibilitatsstorungen der Haut bei Yiseeralkrankungen, edited by Seiffer, Berlin, 1898. Heidenhain, M. : Ueber die gegenseitige Entsprechung von Epidermis- und Coriimileisten an der Beugeflacbe von Hand und Fuss beim Affen und Menschen, Anat. Hefte, No. 91, 1906, p. 419. HiESCHLAND : Beitrage zur ersten Entwicklung der Mammarorgane beim Menschen (Embryo von 4 mm.), Anat. Hefte, vol. xi, 1898. Ide, M. : Kouvelles observations sur les cellules epitheliales, La Cellule, vol. v, 1889, p. 319 (abstr. in Monatsb. f. prakt. Dermatol., vol. ix, 1890, p. 550). Jadassohn : Ueber Pityriasis rubra und ihre Beziebungen zur Tuberculose, nebst Bemerkungen iiber Pigment vei'scbleppung aus der Haut, Arch. f. Dermatol. u. SyphiL, 1892. Jaeisch : Ueber die Bildung des Pig-ments in den Oberhautzellen, Arch, f . Dermatol. u. SyphiL, vol. xxiv, 1892. Joees, L. . Ueber die feineren \'organge bei. der Bildung und "Wiederbildung des elastischen Bindegewebes, Ziegler's Beitrage, vol. xli, part i, 1907. Kallius, E. : Ein Fall von Milchleiste bei einem menschlichen Embryo, Anat. Hefte, vol. viii, 1897. Keibel, F., and Elze, C. : Normentafel zur Entwieklungsgesehichte des Menschen, Jena, 1908. IviDD, W. : On the Imbrication of the Papillary Ridges in Man, Journ. of Anat. and Physiol., vol. xxxix, 1905, p. 413. The Papillai-j' Ridges and the Papillary Layer of the Corium in the Mammalian Hand and Foot, Journ. of Anat. and Physiol., vol. xli, 1907, p. 35. IvoLLiKER, A. voK : Die Entwicklung des menschlichen Nagels, Zeitsehr. f . wiss. ZooL, vol. xlvii, 1888 (cited by Unna). Handbuch der Gewebelehre des Menschen, vol. i, 1889. KoLOSOFP, G., and Pankul, E.. Yersuch einer mathematischen Theorie der Haut- leistenflguren der Primaten-Palma und -Planta, Morphol. Jahrb., vol. xxxv; also Centralbl. f. nonn. Anat., 1906, p. 319. KOESIANK, B. : Ueber die Modiflkation der Haut und die subkutanen Driisen in der Umgebung der jMund- und Nasendftnungen, die Formationes parorales und paranaricffi der Haussiiugetiere, Anat. Anzeiger, vol. xxix, 1906, p. 113. Ivrause, W. : Die Entwicklung der Haut und ihrer Nebenorgane, Handb. d. vergleich. u. experiment. Entwicklungsgesch. der WirlDeltiere, herausg. von 0. Hertwig, vol. ii, 1902. Keauss, F. : Der Zusammenhang zwisehen Epidermis und Cutis bei Sauriern und Krokodilen, Arch. f. mikr. xVnat., vol. Ixvii, 1906, p. 313. Krojiatee : Die Parenchymhaut und ihre Erkrankungen, Arch, f . Entwicklungs- mech. del- Organe, 1899, p. 253. Die Desmoplasie der Epithelzellen in der menschlichen Haut, Monatsh. f. prakt. Dermatol., vol. xli, 1905, p. 477. Laxgek, K. : Zur Anatomie und Physiologie der Haut, I. Ueber die Spaltbarkeit der Cutis, Sitzber. d. k. Akad. "\Vis.s. Wien, math.-naturwiss. Kl., vol. xliii, 1861, p. 19. Lewinski : Ueber die Furchen und Fallen der Haut, Yirchow's Arehiv, vol. xcii, 1883. LoEB, L. : Ueber Transplantation von weisser Haut auf einer Defekt in schwarzer Haut und umgekehrt am Obr des Meerschweinchens, Arch. f. Entwicklungs- mech. der Organ., vol. \'i, 1808. DEVELOPMENT OF THE INTEGUMENT. 289 LoEWT, J.: Beitrage zur Anatomic und Physiologie der Oberhaut, Arch. f. mikr. Anat., vol. xxxvii, 1907, p. 159. LoiSEL, G. : Formation et evolution des elements du tissu elastique, Joum. de I'Anat. et de la Physiol., vol. xxxiii, 1897, p. 129. LoMBAEDO, C. : II glicogeno della cute, Giorn. ital. dalle mall, ven e della pelle, vol. xliii, 1907, p. 448. LuNEBURG, E. : Beitrage zur Entwicklung und Histologie der Knaueldriisen in der Achselhohle des Mensohen, Inaug. Diss., Rostock (Abstr. in Hofmann- Schwalbe, Jahresb. f. Anat. u. Entwieklungsgeseh., neue Folge, vol. viii, part iii, 1902, pp. 659, 670. Matjeeb, F. : Die Epidermis und ihre Abkommlinge, Leipzig, 1895. Das Integument eines Embryo von Ursus arctos, Festschr. f. Ernest Haeckel, Jena, 1904. De Meijere: Ueber die Haare der Saugetiere, besonders iiber ihre Anordnung, Morph. Jahrb., vol. xxi, 1894. Meirowsky, E. : Die Entstehung des Oberhautpigments in der Oberhaut selbst, Monatsh. f. prakt. Dermatol., vol. xlii, 1906. Die Entstehung des Oberhautpigments aus der Substanz der Kemkorperchen, Monatsh. f. prakt. Dermatol., vol. xliii, 1906. Ueber den Ursprung des melanotisehen Pigments der Haut und des Auges, Leipzig, 1908 (contains a full bibliography). Merk, L. : Die Verbindung mensehlicher Epidermiszellen unter sich und mit dem Corium, Monatsh. f. prakt. Dermatol., vol. xxxviii, 1904, p. 361. Meekel: Tastzellen und Tastkorperchen bei den Haustieren und beim Menschen, Arch, f . mikr. Anat., vol. xi, 1875. Mebkel, Fr. : Die Hautf arbung neugeborener Farbiger, Korrespondenzbl. d. Deutschen Gesellsch. £. Anthropol., Ethnol., und Urgesch., 1907, p. 15. Minot, C. S. : Lehrbuch der Entwicklungsgeschichte, 1894. Nakai, M. : Ueber die Entwicklung der elastische Fasem im Organismus und ihre Beziehungen zu der Gewebsfunktion, Virchow's Archiv, vol. clxxxii, 1905, p. 153. Ohlert, B. : Ueber die Gesetze der Blattstellung, Poggendorff's Annalen der Physik., vols, xciii and xcv, 1854-1855 (cited by Voigt). Oshima, T. : Die Beziehungen des WoUhaares der Neugeborenen zu den Haaren des Erwachsenen, Pfliiger's Archiv, vol. cxvii, 1907, p. 341. Okamura, T. : Ueber die Entwicklung des Nagels beim Menschen, Arch. f. Dermatol, u. Syphil., vol. Iii, 1900, p. 223. Passaege: Monatsh. f. prakt. Dermatol., Erganz-Heft, 1894. Philippson, L. : Ueber die Herstellung von Flachenbildern der Oberhaut und der Lederhaut, Monatsh. f. prakt. Dermatol., vol. viii, 1889, p. 389. PiNKUS, F. : Ueber Hautsinnesorgane neben dem menschliehen Haar (Haar- scheiben) und ihre vergleichend anatomisehe Bedeutung, Arch. f. mikr. Anat., vol. Ixviii, 1905, p. 121. POHL, J. (PiNCUS) : Die Querschnittsform des Kopfhaares der Kaukasier, Zeitschr. f. Ethnol. Verb., 1897, p. 483. Pollitzer, S. : Ueber die Natur der von Zander im embryonalen Nagel gefundenen Kornerzellen, Monatsh. f. prakt. Dermatol., vol. ix, 1889, p. 346. Post: Ueber normale und pathologisehe Pigmentienmg der Oberhautgebilde, Virchow's Archiv, vol. cxxxv, 1893. Prope, 0.: Beitrage zur Ontogenie und Phylogenie der Mammarorgane, Anat. Hefte, vol. xi, 1899. Eabl, H.: Histologie der normalen Haut des Menschen, Mraeek's Handbl. der Hautkrankheiten, vol. i, part i, 1902, p. I. Reh: Die Sehuppen der Saugetiere, Jen. Zeitschr. f. Naturwiss., vol. xxix, 1896. Rein, G.: Untersuchungen iiber die embryonale Entwicklungsgeschichte der Milchdriise, Arch. f. mikr. Anat., vols, xx and xxi, 1882. Vol. L— 19 290 HUMAN EMBRYOLOGY. Retterer, E.: Structure et evolution du tegument externa, Joum. de I'Anat. et de la physiol., 1904. RoSENSTADT, B. : Ueber das Eponychium des Hiihnehens, Arch. f. mikr. Anat., vol. xlix, 1897, p. 561. Studien iiber die Abstammung und Bildung des Haarpigments, Arch. f. mikr. Anat., vol. 1, 1897, p. 350. ScHiFPMANN, J. : Die Histogenese der elastischen Easem bei der Organisation des Aleuronatexsudats, t'entralbl. f. allg. Pathol, u. pathol. Anat., vol. xiv, 1903, p. 833 (bibliography). ScHLAGiNHAUFEN, O. : Ueber die Hautleisten der Planta der Primaten, Korrespon- denzbl. d. Deutsch. Gesellseh. f. Anthrop. Ethnol. u. Urgeseh., 1905, p. 336. Ueber die Leistenrelief der Hohlhand und Fusssohlenflaehe der Halbaffen, Ergebn. d. Anat. u. Entwicklungsgesch., vol. xv, 1905, p. 628-662. Schmidt, H. : Ueber nonuale Hyperthelie mensehlicher Embryonen, Anat. Anzeiger, vol. xi, 1896, p. 702. ScHiliTT, H. : Ueber die Entwicklung der Milchdriise und die Hyperthelie mensehlicher Embryonen, Inaug. Dissert., Freiburg (also in Morph. Arbeiten, vol. viii, 1898, p. 236). SCHMORL, G. : Ueber Pigmentversehleppung aus der Haut, Centralbl. f . allg. Pathol. u. pathol. Anat., vol. iv, 1893. ScHREiBER, L., and Schneider, P. : Eine Methode zur Darstellung von Pigmenten und ihren farblosen Vorstufen, mit besonderer Beriicksichtigung des Augen, und Hautpigments, Mlinchen. med. Wochenschr., No. 37, 1908, p. 1918. ScHRiDDE : Die Protoplasmaf asern der mensehliehen Epidermiszellen, Arch. f. mikr. Anat., vol. Ixvii, 1906, p. 291. ScHULTZE, 0. : Ueber die erste Analge der Milchdriisenapparate, Anat. Anzeiger, vol. viii, 1892. Ueber embryonale und bleibende Segmentierung. Verhandl. Anat. Gesellseh. Berlin, 1896, p. 87. Grundriss der Entwicklungsgesehichte des Menschen und der Saugetiere, Leipzig, 1897. ScHWALBE : Ueber den Farbenwechsel winterweisser Tiere, Morph. Arbeiten, vol. ii, 1893. ScHWElsTDENpR : Vorlesungeu iiber mechanische Probleme der Botanik, bearbeitete von C. Holtermann, Leipzig, 1909 (contains a bibliography of Schwendener's original papers on p. 61). Seipper, see Head. Sherrington, C. S. : Experiments in Examination of the Peripheral Distribution of the Fibres of the Posterior Roots of Some Spinal Nerves, Phil. Trans. Roy. Soc, vol. clxxxiv, 1893. Spaltbholz, W. : Ueber die Beziehungen zwischen Bindegewebsfasern und -zellen, Verhandl. Anat. Gesellseh., Rostock, 1906, p. 209. Stohr, p.: Ueber Intereellularbriicken zwischen ausserer und innerer Wurzel- scheide, Verhandl. Anat. Gesellseh., Heidelberg, 1901, p. 24. Ueber die Entwickelung der Glaschaut des mensehliehen Haarbalges, Verhandl. Anat. Gesellseh., Heidelberg, 1901, p. 26. Entwicklungsgesehichte des mensehliehen Wollhaars, Anat. Hefte, vol. xxiii, 1903. Lehrbuch der Histologie und der mikroskopischen Anatomie des Menschen, 12 Aufl., 1906. Ueber die Sehuppenstellung der mienschlichen Haare, Verhandl. Anat. •Gesellseh., Wiirzburg, 1907, p. 153. Strahl, H. : Die erste Entwicklung der Mammarorgane beim Menschen, Verhandl. Anat. Gesellseh., Kiel, 1898. Studnicka, F. K. : Schematische Darstellungen zur Entwicklungsgesehichte einiger Gewebe, Anat. Anzeiger, vol. xxii, 1903, p. 537. DEVELOPMENT OF THE INTEGUiMBNT. 291 Tandler, J. : Ueber einen menschlichen Embryo von 38 Tagen, Anat. Anzeiger, vol. xxxi, 1907, p. 49. Teuffel, see Schifpmann. Unna, p. G. : Das subkutane Fettgewebe, Monatsh. f . prakt. Dermatol., vol. i, 1881, p. 81. Entwicklungsgeschichte und Anatomie, Ziemssen's Handb. d. spez. Path. u. Ther., Leipzig, 1883, vol. xiv, part i, p. 3. Die Fortsehritte der Hautanatomie in den letzten ftinf Jaliren, lY. Der Nagel, Monatsh. f. prakt. Dermatol., vol. viii, 1889, p. 79. Der Nachweis des Fettes in der Haut durch seoundare Osmierung, Monatsh. f. prakt. Dermatol., vol. xxvi, 1898, p. 601. Eine neue Darstellung der Epithelfasern und die Membran der Stachelzellen, Monatsh. f. prakt. Dermatol., vol. xxxvii, 1903, pp. 289, 337. Unna and GoLODETZ : Neue Studien iiber die Homsubstanz, Monatsh. f . prakt. Dermatol., vol. xlvii, 1908, p. 62. Zur Chemie der Haut, II. Der mikroehemische Nachweis der Keratine durch Millon's Reagens, Monatsh. f. prakt. Dermatol., vol. xlvii, 1908, p. 595. VoiGT, C. A. : Abhandlung iiber die Richtung der Haare am menSehlichen Korper, Denkschr. k. Akad. Wissensoh. Wien. math.-naturwiss. Kl., vol. xiii, 1857. VoRNER, H. : Ueber Trichohyalin, Dermatol. Zeitsehr., vol. ix, 1903, p. 357. Walter, H. E. : On Transitory Epithelial Structures Associated with the Mammary Apparatus in Man, Anat. Anzeiger, vol. xxii, 1902, p. 97. Weber, M. : Bemerkungen iiber den Ursprung der Haare und iiber Schuppen bei Saugetieren, Anat. Anzeiger, vol. viii, 1893, p. 413. Weidbnreich, F. : Ueber Bau und Verhornung des menschlichen Oberhaut, Arch. f. mikr. Anat., vol. Ivi, 1900, p. 169. Weitere Mittheilungen iiber den Bau der Homschieht der menschlichen epidermis und ihren sog. Fettgehalte, Arch. f. mikr. Anat., vol. Ivii, 1901, p. 583. Whipple, Inez L. : The Ventral Surface of the Mammalian Chiridium with Special Reference to the Conditions Found in Man, Zeitsehr. f. Morphol. u. Anthrop., vol. viii, 1904, p. 261. Wiedeesheim, R. : Der Bau des Menschen als Zeugnis fiir seine Vergangenheit, 4th Aufl., 1908. WiETiNG and Hamdy: Ueber die physiologisehen und pathologischen Melaninpig- mentierung und den epithelialen Ursprung der Melanoblastome, Ziegler's Beitrage, vol. xlii, 1907. Wilson, H. C. : Beitrag zur Entwicklungsgeschichte der Haut des Menschen, Mitth. a. d. embryol. Inst, in Wien, part iv (abstr. in Monatsh. f. prakt. Dermatol., vol. i, 1880, p. 46. WiMPPHElMER, C. : Zur Entwicklung der Schweissdriisen der behaarten Haut, Anat. Hefte, No. 104, 1907. Zander: Die friihesten Stadien der Nagelentwicklung und ihre Beziehung zu den Digitalnerven, Arch. f. Anat. u. Phys., Anat. Abth., 1884, p. 103. Untersuchungen iiber den Verhomungsprozess, Arch. f. Anat. u. Phys., Anat. Abth., 1886, p. 373. Untersuchungen iiber den Verhomungsprozess, II. Mitteilung, Der Bau der menschlichen Epidermis, Arch. f. Anat. u. Phys., Anat. Abth., 1888, p. 51. XI. DEVELOPMENT OF THE SKELETON AND OF THE CONNECTIVE TISSUES. By CHARLES R. BARDEEN, Madison, Wis. GENERAL FEATURES. In the bodies of most living tilings certain tissues are differ- entiated for the purpose of passively supporting or protecting the physiologically more active structures. These tissues are charac- terized in the higher vertebrates by the predominant amount of extracellular substance, usually fibrous in nature, which, in large part at least, is differentiated during embryonic development from the peripheral portions of branched anastomosing cells. According to the nature of the intercellular substance the supports ing tissues are subdivided into white fibrous and yellow elastic tissues, reticulum, cartilage, and bone. In early embryonic stages the branched anastomosing cells which compose the supporting tissue or mesenchyme, form an extensive continuous framework. Certain parts of this framework are differentiated into the definitive skeleton and other parts into connective-tissue structures which protect and support the paren- ■chyme of the various organs of the body and attach these organs to the skeleton. The development of the various connective-tissue structures may be considered from two aspects, that of histogenesis and that of organogenesis. The histogenesis of the connective tissues has been most carefully studied in the lower vertebrates, in which the cells are large and the conditions are relatively simple. Organo- genesis has been more carefully studied in man than in any of the lower forms. The histogenesis of the connective tissues is appar- ently similar in the different vertebrates. Study of the histogene- sis of these tissues in man and the higher mammals in general ser-^'es to confirm the results found in the lower vertebrates-. Or- ganogenesis is peculiar for each species, although there are funda- mental similarities to be observed in related forms. We shall first give a brief account of the histogenesis of the connective tissues with especial reference to man and then treat with more detail the morphogenesis of the human skeleton. The specific development of the intrinsic supporting connective-tissue 292 HISTOGENESIS OF THE CONNECTIVE TISSUES. 293 framework of the various organs is most conveniently taken up m connection with each of these organs, and will therefore not be attempted here. PART I. Histogenesis of the Connective Tissues. (a) Early Mesodermic Syncytium. In the youngest human embryos which have been described there is present a well-developed layer of tissue composed of branched anastomosing cells. This tissue layer surrounds the amniotic and yolk-sacs and lines the chorionic vesicle (Fig. 219, A). It forms a continuous sheet between the epithelium lining the Fig. 219. — Diagrammatic sections through two young human embryos described by Graf Spee. A. (V. Spee, Arch. f. Anat. u. Physiol., Anat. Abt., 1896, Taf. I, Fig. 3.) Half-schematic sagittal section through Graf Spec's embryo v. H. B. (Ebenda, Taf. I, Fig. 1.) Half-schematic sagittal section through Graf Spee's embryo Gle. C. (Ebenda, 1889, Taf. XI, Fig. 14.) Transverse section through the embryonic anlage near the region opposite M. in Fig. B. al., allantoic canal; am., amniotic cavity; b., belly stalk; c, chorda anlage; cA., chorion; en., caaalis neurentericus; d., yolk-sac; en., entoblast; f., medullary groove; (?., blood islands; h., heart; kp., region of primitive streak; M., medullary plate; me., mesoblast. amniotic cavity and that lining the yolk sac. The origin of the tissue is nncertain. It evidently is homologous with the mesoderm which in many of the mammals is known to arise from the primitive streak and head process. In man the primitive mesoderm is apparently formed before the appearance of the medullary plate, the neurenteric canal, and 294 HUMAN EMBRYOLOGY. the primitive streak. After these structures appear the mesoderm disappears in the mid-axial line anterior to the primitive streak, and a chordal plate is differentiated in the entoderm. (See Fig. 219.) (b) Formation of the Mesodermic Somites. In Graf Spee's embryo Gle (Fig. 219, B and C) the mesoderm extends on each side of the neurenteric canal and of the medullary groove (Fig. 219, C) to the anterior extremity of the embryonic anlage where the mesodermic sheets of both sides become united (Fig. 219, B, h). Posterior to the neurenteric canal the mesoderm is intimately united to the tissue of the primitive streak, a region of active production of mesenchymal tissue. At the outer margin of the embryonic anlage the mesoderm is continuous posteriorly with the mesenchymal lining of the chorion, laterally and ante- riorly with the mesodermic covering of the yolk sac and amnion (Fig. 219, B and C). At a slightly later period the sheet of mesoderm on each side of the neural tube becomes longitudinally separated from the more laterally situated mesoderm (Fig. 220, A and C) and at the same time divided into a series of segments (mesoblastic somites). In the chick the first somites formed are the occipital somites (J. T. Patterson, Biological Bulletin, 1907, vol. xiii, p. 121), then follow in turn the cervical, thoracic, Imnbar, sacral, and coccygeal. It is probable that the first somites formed in the human embryo belong to the occipital region. In the latter half of the first month of development in the human embryo there are found anterior to the cer^dcal myotomes three incomplete occipital myotomes. The rela- tions of these myotomes to the first somites differentiated have not yet been definitely determined. Eternod (Anat. Anz., 1899, p. 131) has described an embiyo with eight somites, KoUmann (Anat. Anz., 1890, Arch. f. Anat. u. Physiol., Anat. Abt., 1891) one with fourteen (Fig. 220, A), and Mall ( Journ. Morphol., 1897) likewise one with fourteen. Mall considers the first three somites in his embryo to be occipital somites. They probably correspond with the first three somites in the embryo described by Kollmann and possibly with the first three in the embryo described by Eternod. The occipital somites are probably not completely divided off either from one another or from the lateral mesoderm (Figs. 229, 230, A).i The cervical somites, at least the more distal ones, on the other hand, become completely separated, and the tissue in ' Kollman states that in his embryo Bulle, with fourteen somites, the seonnen- tation externally appears well mai'ked in the post-otic region, but internally is apparently incomplete. (Personal communication to the author.) HISTOGENESIS OF THE CONNECTIVE TISSUES. 295 Somites Foregut Medullary tube Midgut ;^ 1 i Aorta _ Medullary tube Chorda anlage - -Entoderm .-• Intermediate cell mass ' ^1^ Parietal layer of the Y moonblast Coeloni Visceral layer of the mesoblast ~" Fig, 220. — A. (After Kollmann, Lehrbuch der Entwicklungsgeschichte des Menschen, Fig. 119.) Himian embryo with fourteen somites, 2.5 mm. long. Magn. 30 : 1. B. (Ebenda, Fig. 56.) Transverse section through the region behind the heart of the embryo shown in Fig. 220, A. C. (After Kollmann, Arch. f. Anat. u. Physiol., Anat. Abt., 1891, Plate III, Fig. 3.) Transverse section through the tenth pair of somites of the embryo shown in Fig. 220, A. 296 HUMAN EilBRYOLOGT. each assumes an epithelial character and becomes arranged about a central cavity or myocoel (Fig. 230, d). At the posterior end of the cervical region a solid column of cells marks for a short period the remains of the neurenteric canal. Beyond this in the axial region lies the tissue of the primitive streak which is continued into the mesenchyme of the allantoic stalk. In subsequent develop- ment mesoderm is differentiated from the anterior end of the primitive streak on each side of the posterior end of the neural groove. In this mesoderm successive somites are formed. Finally, as the differentiation of the body extends posteriorly, a definite primitive streak gradually gives way to a mass of mesenchymal cells situated between the ectoderm and entoderm, and then, in the caudal process, to a mass of cells entirely surrounded by ecto- derm. From this mass of cells are successively differentiated the more caudal mesodermic somites. (c) Axial Mesenchyme. As the chorda dorsalis becomes differentiated (see below) marked changes take place in the somites. For a time these consist of epithelial tissue which surrounds a central cavity or myocoel (Fig. 220, C). Toward the end of the third week the cervical and thoracic myoeoeles become gradually filled with branched spindle- shaped mesenchyme cells wliich come from the surrounding epithe- lium. The medial wall of the somite opens, and the mesenchyme cells wander out toward the neural tube and the chorda, and give rise to a tissue which ensheathes these organs (Fig. 221). The mass of mesenchyme derived from each somite represents a sclero- tome. _ The successive sclerotomes soon fuse so as to give rise to a continuous mass of mesenchyme. The mesenchyme of the two sides becomes fused. After giving rise to the sclerotomes the somites become converted into myotomes, the further fate of which is described in the section on the development of the muscular system. In many of the lower vertebrates the lateral layer of the myotomes gives rise to dermis, but in mammals the dermis comes chiefly, if not wholly, from axial mesenchyme. (Bardeen Johns Hopkins Hospital Reports, vol. ix, 1900.) ' (d) Parietal and Visceral Layers of the Mesoderm. During the formation of the embryonic ccelom, the lateral unsegmented mesoderm plates become divided into two layers, a parietal layer and a visceral layer (Figs. 220, C\ and 221). The cells facing the coelom assume an epithelial character. The deep strata of the parietal layer give rise to scleroblastema, from which some of the skeletal apparatus and connective tissues of the trunk HISTOGENESIS OF THE CONNECTIVE TISSUES. 297 and limbs are derived. The deeper strata of the visceral layer give rise to the connective tissues as well as to the musculature of the thoracic and abdominal viscera. (e) Mesenchyme of the Head. The axial mesoderm of the trunk is continued forward on each side of the chorda dorsalis to the region of the base of the midbrain. From it arises a large part of the mesenchyme of the head, including most of that which gives rise to the skeletal struc- tures of the cranium and the upper part of the face. The trans- formation of mesoderm into definitive skeletal structures is more direct in the cranial than in the spinal region. The formation of somites for the axial region of the head is restricted to the postotic region, and even here it is, as mentioned above, less complete than Myotome Mesenchyme from somile" Intermediate _ cell-nia?s Parietal layer of the meso ■ blast Visceral layer of the misoblast Fig. 221. — (After Kollmann, Arch. f. Anat. u. Pnysiol., Anat. Abt., 1891, Plate III, Fig. 8.) Transverse section through the posterior part of the trunk of a human embryo of the third week. in the trunk. The cranial mesoderm apparently is largely con- verted into mesenchyme without going through that process of division into somites characteristic of the spinal mesoderm. The mesenchyme near the chorda in the occipital region shows no segmentation in the latter half of the first month. More laterally segmentation is indicated by the formation of myotomes from the dorso-lateral portion of the mesoderm. Near the myosepta the mesoderm may show a slight condensation. In the prechordal mesenchyme of the head there are differ- entiated in many vertebrates vesicular cavities, "head cavities," 298 HUMAN EMBRYOLOGY. lined by epithelium, from wliicli musculature and mesenchyme arise. There are four such cavities in selachians and in reptiles. Their relation to the somites is undetermined. In man very transitory structures of this nature have been reported (Zimmer- mann, Ueber Kopfhohlenrudimente beim Menschen, Arch. f. mikr. Anat.', 1898, vol. liii), but they are rare and play no essential part in development. The dorsal portion of the lateral mesoderm plate of the trunk is continued anteriorly into the branchial region, where it gives rise to the mesenchyme of the branchial arches and partly also to that of the head. Ventral to the branchial arches the lateral mesoderm of the trunk is continued into the pericardial mesoderm. The coelom does not extend into the branchial region of the lateral mesoderm of the head. (f) Origin of the Connective Tissues. From the mesenchyme, derived in part directly from the primitive embryonic mesodermic tissue, in part from somites ditferentiated from this primitive tissue, and in part from the primitve streak, there arses a syncytial tissue which in turn gives origin to the various connective tissues and skeletal structures of the body as well as to some other structures, for instance, muscles and blood-vessels. In the adult connective tissues the bulk of the tissue substance is usually described as extracellular.^ The chief problem for those who have studied the histogenesis of the connective tissues has been to determine whether the substances which are intercellular in the differentiated tissues have an intracellular or an inter- cellular origin. The weight of evidence seems at present to be decidedlv in favor of the intracellular origin (Fleming, 1891, 1897, 1902, Retierer, 1892-1906, Spuler, 1897, Mall, 1902, and Spalteholz, 1906 ) . Among recent investigators who believe that the connective- tissue fibres have an intercellular origin may be mentioned E. Laguesse (1903) and Fr. Merkel (1895, 1909). Golowinski, while contending that the fibres appear between the cells, admits that they rise close to the cell body. According to him, most investiga- tors have described essentially the same phenomena, but some con- sider the mother substance in which the fibres arise as ectoplasm, while others consider it an intercellular substance. The majority of those who adopt the view that the "intercellular" portions of the adult connective tissues are intracellular in origin describe the primitive mesenchymal cells as becoming differentiated into endoplasmic and ectoplasmic portions. In the ectoplasm the " Spalteholz (Anat. Anz., 1906) lias, however, shown that even in the adult many, if not all, of the fibrils have an intracellular position. HISTOGENESIS OF THE CONNECTIVE TISSUES. 299 intercellular elements characteristic of each, of the various kinds of connective tissue are differentiated while the endoplasm becomes converted into the cells of the adult tissue. Eetterer (1892-1906) gives a different description of the process. According to him the primitive tissue from which the various kinds of connective tissue are differentiated consists of a homogeneous syncytium in which nuclei are scattered about. This homogeneous syncytium becomes differentiated into two parts, a hyaloplasm and a granular chromophilic portion. The granular chromophilic portion sur- rounds the nuclei and gives rise to branching processes which anastomose so as ultimately to form an extensive network. The hyaloplasm lies in the meshes of this network. The fibres of recticulum, elastic fibres, and the branched anastomosing processes which fill the canaliculi of bone arise from the chromophilic net- work, while white fibrous tissue and the chief part of the ground substance of cartilage and of bone are differentiated from the hyaloplasm. Eecently still another view of the origin of the fibrils of the connective tissues has been advanced. It has been known for some time that in the vitreous humor before the entrance of blood-vessels and mesenchyme cells there exists a fibrillar structure the compo- nents of which may be looked upon as branched anastomosing processes of cells of the retina and lens. From this fibrillar net- work the fibrils of the adult vitreous humor are probably derived. Aurel V. Szily (1908) has described a fibrous network filling in spaces throughout the embryonic body before the origin of the mesenchyme. The fibrils of the network are branched anasto- mosing processes of the epithelial layers bounding tiie various cavities. Szily tliinks that when the mesenchyme cells arise they wander into meshes of this fibrillar network and enter into intimate relations with the component fibrils. The fibrils subsequently lose connections with the epithelial cells from which they arise. According to Szily the fibrils of the early embryonic syncytium are thus of epithelial origin, while the cell protoplasm is of the mesen- chymal origin. Although the early connective-tissue fibrils are thus according to this view of epithelial origin, at a later stage connective-tissue fibrils are also differentiated in the ectoplasm of cells derived from the mesenchyme. According to Retterer (1904 and 1906) the syncytium of the cutis arises partly from the epidermis. The following account of the origin of the connective tissues is based chiefly on the paper of Mall, who has taken up the problem in connection with the pig and man. At an early stage there appear to be many individual cells in the mesenchyme "which multiply rapidly, so that in certain regions the nuclei are closely packed together. Then the cells 300 Hr:MAN EMBRYOLOGY. unite to form a syncytium and the protoplasm of the syncytium increases more rapidly in amount than the nuclei, so that the latter appear more widelv separated from one another than at first. The nuclei at an earlv stage lie within the protoplasm of the syn- cytium, but gradually differentiation takes place. Immediately about the nuclei the protoplasm becomes granular and forms an endoplasm which is distinct from the rest of the syncytium or ectoplasm. From the granular endoplasm about the nuclei processes may extend into the surrounding ectoplasm. In the ectoplasm fibrillation becomes more and more distinct. The nuclei surrounded by the endoplasm come to lie in certain of the meshes of the network formed by the ectoplasm. In other of the meshes merely a fluid substance "is seen. From this embryonic syncytium the various types of connective tissue are differentiated. Eetictjltjm.— Reticulum seems to be the least highly differ- entiated form of tissue which arises from the embryonic connec- tive-tissue syncytium. The reticulum develops directly in the syncytial ectoplasm, while the nuclei and endoplasm are converted into cells which lie upon the reticulum fibres. In the liver the origin of the reticulum differs from that in other parts of the body in that it arises from Kupffer's endothelial cells instead of from mesenchyme. The endothelial cells form a syncytium in which the reticulum fibres are differentiated. According to Bet- terer the reticulum fibres arise from chromophilic processes of the perinuclear protoplasm. White Fibrous Tissue. — In the development of white fibrous tissue from the embryonic syncytium Mall distinguishes two stages. In the first or prefibrous stage a tissue much resembling reticulum is differentiated, in the second or fibrous stage true white fibrous tissue appears. (Fig. 222, A and B.) In the first stage the syncytium grows very rapidly. The ectoplasm increases in amount much more rapidly than the endoplasm. The nuclei, however, multiply, and the endoplasm about each nucleus becomes drawn out spindle-like, giving rise to the well-known embryonic bipolar cells. The tips of these cells are extended into the ecto- plasm, and here the endoplasm appears constantly to contribute to the ectoplasm. The ectoplasm becomes steadily more fibrillated. The strands of ectoplasm become more and more drawn out, in tendons and fascia into parallel, in areolar tissue into interweaving bundles of fibres. In flie fibrous stage the embryonic fibres are converted into true white fibrous tissue, their chemical nature meanwhile changing. The fibres at first occasionally anastomose, but during further development the anastomosing bridges begin to break down. According to Mall the larger fibres become split into the individual fibrils of white fibrous tissue. The embryonic spindle-shaped cells become converted into the adult connective- HISTOGENESIS OF THE CONNECTIVE TISSUES. 301 tissue corpuscles. According to Retterer the fibres in the pre- fibrous stage belong to the chromophilic processes of the peri- nuclear protoplasm. On the other hand, the collagenous fibres arise from the hyaloplasm (ectoplasm). _ The body of the cornea is composed of a tissue the origin of which is similar to that of white fibrous tissue. It retains more features characteristic of the embryonic connective tissue than does the ordinary white fibrous tissue. It contains no elastic fibres. 0'-- ^.r-Z^fV ■^ E Fig. 222. — (After Mall, American Journal of Anatomy, 1902.) To illustrate the development of the connective tissues. A. (Fig. 12, Mall.) Section through the skin of a pig 5 cm. long. White fibres are forming in the ectoplasm. Magn. 250 : 1. B. (Fig. 13, Mall.) Section through the skin of a pig 16 cm. long. The nuclei and endoplasm on the left are immediately below the root of a hair. Magn. 250 ; 1. C. (Fig. 14, Mall.) Elastic tissue just beginning to appear in the syncytium of the umbilical vein of a pig 7 cm. long. Magn. 250 : 1. D. (Fig. 10, Mall.) Section through the occipital cartilage of an embryo pig 20 mm. long. The ground substance is deposited in the ectoplasm of the syncytium. Magn. 250 : 1. E. (Fig. 11, Mall.) Section through the frontal bone of a pig 20 mm. long. Magn. 250 : 1. Elastic Tissue. — With the exception of the tissue of the cornea probably all white fibrous tissue contains a greater or less number of elastic fibres intermingled with the bundles of white fibrils. The elastic-tissue fibres apparently are differentiated directly in the same syncytial ectoplasm in which the bundles of white fibrils develop (Fig. 222, C). The youngest pigs in which Mall found elastic fibres were four centimetres long. These fibres were found in the aorta and neighboring arteries. Fenestrated membranes are formed by the coalescence of neighboring fibres. Spalteholz (1906) has found elastic fibres in the truncus arteriosus of pig embryos 9.2 mm. long. Ranvier held that elastic fibres arise from the fusion of rows of elastic granules. According to Mall, elastic fibres are never formed by the fusion of rows of such granules. Spalteholz has likewise found that the elastic 302 HUMAN EMBRYOLOGY. fibres are directly differentiated. According to Retterer, the elastic fibres arise in the perinuclear chromophilic protoplasm and from the chromophilic processes which spring from it. Adipose Tissue. — Adipose tissue appears in the fourth month in the human embryo. In the regions where the adipose tissue is formed the embryonic mesenchymal tissue becomes differentiated on the one hand into blood-vessels and a supporting fibrous-tissue framework, on the other into cells in the protoplasm of which granules of fat appear. The granules of fat in each cell gradually become consolidated, so that finally there arises a single large globule of fat which greatly distends the cell. The protoplasm of the cell now forms a thin covering for the globule of fat. The nucleus surrounded by a small amount of granular protoplasm lies at one side. The fat cells are arranged more or less definitely with relation to the blood-vessels and frequently form well-marked clusters. (See Bell, 1909.) Caetilage. — In the formation of cartilage the ectoplasm of the syncytium becomes more and more dense. The nuclei sur- rounded by endoplasm come to lie in spaces in the ectoplasm, thus forming precartilage cells which in turn become converted into cartilage cells (Fig. 222, D). The syncytial ectoplasm undergoes chemical changes which make it exhibit the reactions characteristic of hyaline ground substance. Not infrequently the ectoplasm before becoming converted into hyaline ground substance becomes marked out into cell territories by the appearance of membranes between the cell units. These membranes appear as fine lines in cross section and have staining reactions similar to hyaline car- tilage. "When this condition is found, the cartilage has an epithe- lioid appearance (cellular cartilage). The endoplasmic units or cartilage cells exliibit a differentia- tion into perinuclear and peripheral portions. From the peri- pheral portion hyaline substance is differentiated so as to form a capsule (Max Schultze). The capsule appears lighter than the surrounding tissue and has slightly different staining reactions. Meanwhile the endoplasm increases in amount, the nuclei multiply, and from time to time cell division takes place in the endoplasmic units, but this division does not extend into the surrounding ecto- plasm. When cell division takes place, the line of separation between the two daughter cells usually becomes marked by a fine septal membrane composed of a substance that has some of the staining qualities of the cell capsules. This septum then becomes divided into two lamellae, each of which together with half of the old capsule surrounds a daughter cell. Sometimes the capsules of several successive generations of cells remain distinct for a con- siderable period, so that a capsule which first surrounded a single cell comes to surround several groups of daughter cells, each group HISTOGENESIS OF THE CONNECTIVE TISSUES. 303 and each daughter cell having in turn a capsule of its own. Usually, however, the primitive capsules become indistinguishably fused with the surrounding matrix, so that capsules about single cells or pairs of cells alone remain distinct. G-rowth of cartilage is in part interstitial, in part perichondral. The interstitial growth is due (1) to the direct increase in amount of the ectoplasm or ground substance, (2) to the formation of cell capsules at the periphery of the cells and the fusion of these capsules with the matrix, and (3) to cell multiplication. Peri- chondral chondrification is due to the formation of new cartilage beneath the perichondrium. The ground substance increases in amount faster than the cells multiply.^ In white-fibrous cartilage bundles of fibrils develop in the syncytium while the hyaline substance is being deposited. In elastic cartilage, according to Mall, elastic fibres are formed after the hyaline substance has been differentiated. According to Spalteholz (1906), however, elastic fibres appear before the hyaline ground substance in the ear cartilage of the pig. In the arytenoid cartilage clumps of elastic granules are deposited. While Eanvier held that elastic fibres arise from the fusion of rows of these granules. Mall, as mentioned above, believes that neither here nor elsewhere are the elastic granules fused to form elastic fibres. Bone. — The histological structure of bone is still a matter of dispute. Most investigators seem to consider the ground sub- stance to be composed of bundles of fibrils resembling those of white fibrous connective tissue embedded in a homogeneous "cement" substance. V. Kolliker, who considered the cement sub- stance to be slight in amount, believed the calcium salts to be embedded both in this and in the fibrils. V. Ebner, 1875, believed the calcium salts to be embedded chiefly in the cement substance. Retterer, 1905 and 1906, believes the ground substance of bone to be composed of a chromophilic reticulum embedded in a hyalo- plasm impregnated with calcium salts. It is well known that the ground substance of bone contains a collagenous substance similar to that of white fibrous tissue. Bone, like other connective tissues, is formed from a blastemal syncytium. Ectoplasm becomes distinct from nucleated endo- plasmic cell units. In the ectoplasm calcium salts are deposited. Two stages may thus be distinguished, — a pre-osseous, previous to the deposition of calcium salts, and an osseous, after these salts have been deposited. During ossification about two parts of inor- ganic salts combine with one part of organic matter. The cells which give rise to bone may appear similar to ordinary immature connective tissue cells or they may pass Tor details concerning the development of cartilage see Retterer (1900). 30i HUMAN EMBRYOLOGY. through a stage in which they appear epithelioid in character. Cells of the latter type are frequently found in regions where layers of bone are being applied to pre-existing bone or to calcified cartilage. The epithelioid cells, which Gegenbaur called osteo- blasts, form a layer from the deep surface of which certain cells branch, anastomose, and give rise to an osteogenetic syncytium which becomes converted into bone (Fig. 223, A). According to v. Kolliker (Grewebelehre) and to many other investigators, the osteoblasts secrete the ground substance, which, therefore, is to be looked upon rather as intercellular than as intracellular. To ^Valdeyer (1865) we are indebted for the first clear description of the differentiation of the ground substance of bone in the peripheral protoplasm of the osteoblasts. The endoplasmic units, or bone corpuscles, have branched processes which anastomose, freely through the canaliculi with those of neighboring cells. Before birth (Neumann) the periphery of the bone corpuscles becomes differentiated into a resistant cuticle which has staining reactions similar to elastic tissue (Ret- terer) and which is resistant to strong acids and alkalies. Brosike (1885) considered this cuticle (bone-cell capsule) to be composed of keratin, but Kolliker has shown it to be soluble in boiling water. According to Retterer the protoplasm of the branching processes which lie in the canaliculi is converted into a similar substance. In the human embryo bone arises chiefly in connection with a transitory cartilaginous skeleton which it gradually in large part replaces. Thus the vertebrae, ribs, sternum, the skeleton of the extremities, and most of the base of the skull are first formed of cartilage, and the cartilage is later replaced by bone (substitution bone). Centres of ossification may appear within the cartilage (endochondral ossification) or beneath the perichondrium (sub- periosteal ossification). On the other hand, most of the bones of the face and the flat bones of the skull are formed directly in membranous tissue (intramembranous bone). When bone is first formed in the embryo, it consists of a coarse plexiform or spongy framework, in the meshes of which lies a vascular embryonic marrow. To the walls of the spaces in this primitive spongy bone successive layers of bone are added by osteoblasts, so that the spaces come to have lamellated walls. Similarly beneath the periosteum lamellae of bone are laid down, so that the surface of the bone comes to consist of a series of suc- cessive lamellae. The formation of definite lamellae of compact bone is not, however, well marked until after birth. Previous to this period the vascular spaces in the bone are relatively large, so that the coarse spongy structure mentioned above is long retained. In long bones Schwalbe found compact lamellar bone formed about the marrow cavity and in the Haversian canals in HISTOGENESIS OP THE CONNECTIVE TISSUES. V I I / 305 Giant cell Lacuna (Osteoclast) ^•« / i:- <. 'i ;• ■-3 y^' 3one Marrow cells .\' Fig. 223. — A. (After Retterer, 1906.) Section through the jaw of a dog at birth. 1, layer of osteoblasts beneath the periosteum; 2, osteoblasts with reticular connective tissue (6) in a medullary cavity in the bone; 3, layer of preosseous tissue; 4, nucleated, granular, cytoplasmic masses of the preos- eeous layer; 5, osseous lamina; v, blood-vessel. B. (After Szymonowicz.) From a longitudinal section of the femur of a rabbit embryo. Vol. I.— 20 306 HmiAX EMBRYOLOGY. the sixth month after birth, but beneath the periosteum not until the fourth year. Kolliker (Gewebelehre) found lamellar subperi- osteal bone as early as in the first year after birth. During the period of the growth of bone new bony tissue is being constantly added in some regions, while in other regions the bone already formed is absorbed to make way for new vascular marrow cavities. In this process of bone absorption large cells, osteoclasts, containing, according to v. Kolliker who first described them, from one to sixty nuclei, play a chief part (Fig. 223, B). These osteoclasts vaiy in size, being from 43 to 91 ^ long, 30 to 40 1^ wide, and 16 to 17 /x thick. They apparently have the power of dissolving bone or calcified cartilage. The depressions which they cause in bone are called Howship's lacunae. According to Kolliker, they arise from osteoblasts, and may again divide up into osteoblasts or after remaining for a greater or less length of time in the bone marrow they may disappear. The nuclei within the cell multiply by direct division. The changes of form which bones undergo through the process of growth by apposition of new layers of bone to pre-existing layers and the absorption of bone previously laid down are well illustrated by comparing the jaw of the infant with that of the adult (Fig. 224, C). Under the term Sharpey's fibres, according to Eetterer (1906), several distinct structures have been described: (a) prolonga- tion of the periosteum into the bone; (b) granular elastic proto- plasmic processes of the lamellar system; (c) portions of the bone in which calcium salts have disappeared from the hyaloplasm and fibrous tissue has been differentiated. The true Sharpey's fibres are probably prolongations of the periosteum left behind as suc- cessive layers of bone are differentiated beneath the periosteum. To this brief description of the general nature of the process of ossification we may add a short account of the special features which characterize intramembranous, subperiosteal, and endo- chondral types of ossification. Intramembranous Ossification (Fig. 222, E, Fig. 224). — In this type of ossification bone first appears in the form of a network of spicules interwoven with a network of blood-vessels. Ossifica- tion begins at a centre from which it radiates peripherally. As one passes from the centre towards the periphery in the early period of ossification, one finds all stages from fully formed bone to an undifferentiated embryonic connective-tissue syncytium. In ossification in very young embryos the connective-tissue syncytium appears to be directly transformed into bone. The transforma- tion is marked first by the fibrils of the ectoplasm becoming more clearly marked, and then by the appearance of a basophilic sub- stance in the ectoplasm. In older embryos the ectoplasm is, accord- ing to Mall (1906), transformed into prefibrous tissue and the HISTOGENESIS OF THE CONNECTIVE TISSUES. 307 Bone corpuscles Osteoblasts Blood-vessel Connective tissue Fig. 224. — A. (After Szymonowicz, Text-book of Histology, trans, by MacCallum, Fig. 199.) From a transverse section of the parietal bone of a human fetus. B. (Quaiu, after Sharpey, Quain's Anatomy, 10th ed., vol. 1, Pt. 2, Fig. 308.) Parietal bone of a fetal sheep. Size of fetus 2^ in. Magn. about 12 : 1. a-b, height of subjacent cartilaginous lamina. C. (After Kcilliker, Gewebelehre, Fig. 271.) Mandible of a new-born infant, contrasted with that of an adult. 308 HUMAN EJIBRYOLOGY. latter is transformed into bone. The diameter of tlie embryonic bone corpuscles, according to v. Kolliker, varies from 13 to 22 ^. The primitive plexiform bone is thickened by deposit of osseous substance beneath the periosteum. The latter appears soon after bone-formation has commenced. The spaces in the plexiform network of bone at an early stage become converted into canals containing blood-vessels and primitive marrow. In bone of membranous origin cartilage may subsequently be developed beneath the periosteum. Examples of this are to be found in the temporomandibular joint. Hyaline cartilage Zone of calcification Capsules containing many cartilage cells Fig, 225, A. — (After Szymonowicz, Text-book of Histology, translated by MacCallum, Fig. 193.) From a longitudinal section of a finger of a three-and-a-half-months human fetus. Two-thirds of the second phalanx are represented. At X a periosteal bud is to be seen. Magn. about 85 : 1. Subperiosteal Ossification (Fig. 225, A and B). — Bone is formed in the deep layer of the periosteum (perichondrium) essen- tially as bone is formed in membrane which is not closely applied to cartilage. The bone formed beneath the periosteum has at first a coarse plexiform structure. The meshes of the osseous frame- work enclose vascular embryonic marrow. As mentioned above, dense subperiosteal lamelltB are formed in human long bones in the first year after birth, according to Kolliker (Gewebelehre), while according to Schwalbe they are not formed until the fourth HISTOGENESIS OF THE CONNECTIVE TISSUES. 309 year. Subperiosteal ossification is the sole method of substitution of osseous for cartilaginous tissue i'n some of the bones (in the ribs, for example), while in others it is closely associated with endochondral ossification (diaphyses of the long bones). Wlien it is the sole method of ossification, the underlying cartilage fre- quently undergoes changes similar to those preceding endochon- dral ossification (see below). ^^^^^ieof '^Ceogenetic j tissuo "^— ..-' Periosteal bud Blood-vesseli filled with blood I Calcified cartilag( ^^^^ i Enlarged cartilage - cell cavities li^ndochondral bone ^Primary marrow space Perichondral bone Fig. 225, B. — (Afer Szymonowicz, Text-book of Histology, translated by MacCallum, Fig. 195.) From a longitudinal section of a finger of a four-months human fetus. Only the diaphysis of the second phalanx is represented. Magn. about 85 : 1. Endochondral Ossification, — In endochondral ossification processes from the osteogenic layer of the perichondrium, or periosteum, extend into the substance of the cartilage, and these give rise on the one hand to destructive activities which break down the cartilage and on the other to constructive activities which result in the formation of bone. Endochondral ossification is pre- ceded by well-marked changes in the cartilage (Pig. 225). The cartilage cells first multiply rapidly in number and then enlarge so that the matrix becomes relatively reduced in amount. Neigh- 310 HUMAN EMBKYOLOGY. boring cartilage cells may so expand that the matrix between them disappears. Meanwhile calcium salts are deposited in the matrix in the form of granules which may become confluent. The proc- esses from the periosteum break into the cavities occupied by the cartilage cells, enlarge them, and thus give rise to primary marrow cavities. In the phalanges and other long bones of limited size, the cartilage at the centre of the shaft may be completely absorbed before the endochondral ossification begins. The primitive marrow is vascular and contains an embrj^onic syncytium not highly ditferentiated. Osteoblasts and osteoclasts and embryonic connective tissue, however, appear in it at an early stage, fat and marrow cells at a later period. About the primitive marrow cavi- ties bone is quickly laid down and there thus arises spongy endo- chondral bone. The bone first laid down is later again absorbed during development of the larger central marrow cavities. From the cavities into which the marrow first penetrates it gradually extends into neighboring cartilage-cell spaces. As the osteogenic tissue spreads, the surrounding cartilage undergoes changes simi- lar to those which took place at the primary centre of ossification. Thus, so long as the process of ossification continues, the cartilage farthest removed from the centre of ossification shows the least modification from the type of primitive hyaline cartilage, while as one passes toward the centre of ossification one finds the successive changes of cell multiplication, cell expansion, and calcification of the matrix. In long bones these successive stages are especially well marked. The multiplication of cartilage cells gives rise to groups which liecome arranged in long columns which are parallel to the long axis of the bone. The boundary between the zone of ossification and that of the highly modified cartilage is usually fairly sharp (Fig. 225, B). Capillary loops extend close to the limit of the advancing ossification. The extremities of these loops are often dilated. The fate of the cartilage cells in the calcified matrix is still in dispute. Most modern investigators, including Kolliker (Gewebelehre, 1889), seem to follow Sharpey and Loven in con- cluding that the cartilage cells are destroyed as osteogenic tissue derived indirectly from the periosteum enters the cell spaces. Numerous accurate observers, however, among who may be men- tioned H. Muller (1859), Eanvier (1865), and Retterer (1900), believe that the cartilage cells become converted into osteogenic tissue, each cartilage cell giving rise to several smaller cells and to reticular tissue. The old view, that cartilage may become directly converted into bone, seems to have few modern adherents.'' 'According to Strelzoff (1873), this metaplasia is constant in some regions, for example, in the lower jaw of human embryos. HISTOGENESIS OF THE CONNECTIVE TISSUES. 311 In epiphyses centres of ossification arise at a comparatively late period. Blood-vessels, wliicli spring from the periosteum and from the bone marrow, penetrate into the epiphyseal cartilage long before ossification begins. Friedlander (1904) gives good pictures of the blood-vessels in the epiphyseal cartilages of' the long bones. In some cartilages the blood-vessels appear in the third fetal month. In the seventh all the larger cartilaginous areas show rich vascular plexuses. Grontli of Bone. — The question as to whether or not there is an interstitial growth of bone has given rise to extensive investiga- tions. The evidence is fairly conclusive that there is no well- marked interstitial growth in bone. Hales, Duhamel, John Hunter, and others showed, during the eighteenth century, that two pegs driven into a bone do not move apart during development unless there is a non-ossified region between the two pegs. Z. Q. Strelzoff (1873), however, brought forward a certain amount of evidence to show that under some circumstances there may be a slight inter- stitial growth of bone.s Experiments made with madder go to show that growth of bone takes place entirely by apposition. Madder stains newly forming bone, and by feeding it to young animals the successive applications of layers of bone may be followed. Experiments along this line were first performed by Duhamel and J. Hunter. Duhamel also showed that a ring placed on the outside of a long bone of a j'oung animal may eventually be found in the marrow cavity. Regeneration. — In case of fractures union is effected by osteo- blasts which give rise to new bone which unites the broken ends. These osteoblasts in young animals may apparently be derived either from the marrow or from the periosteum, but in the adult chiefly, if not wholly, from the periosteum. Bonome (1885) has, however, brought forward evidence to show that the bone corpuscles in certain conditions where they are supplied with abundant nutrient blood may give rise to osteoblasts. Not infre- quently temporary cartilage is produced in places at the site of the fracture. In man fibrous tissue is often produced if the broken ends of the fractured bone are not closely approximated. The experiments of Oilier and others have shown that the bone-forming power of the periosteum may be exercised even when this is trans- planted into the tissues at some distance removed from any bone. If the periosteum is preserved it has the power of restoring in nearly normal form large parts of bone. = See also Eg-ger (1885) and J. Wolff (1885). 312 HUMAN EMBRYOLOGY. ADDENDUM. Since the preceding section on the development of the connective tissues was written, there have appeared several important articles on the development of the connective tissues in mammals. Fr. Merkel (1909) brings forth new evidence in favor of the intercellular origin of the connective-tissue fibrils. He pays particular attention to the development of limiting membranes which in places sharply mark off epithelium from the underlying connective tissue. These mem- branes, according to Merkel, arise from the connective-tissue matrix independently of the connective-tissue cells. They may become fibrillated. Similar non-cellular connective-tissue substances are formed at an early stage in the septa between myotomes, and later between muscle cells of various types and in lamellated connective tissues. The sarcolemma of striated muscle cells has a similar origin, according to Merkel. Disse (1909), on the other hand, describes the osteogenetio tissue as arising from the cell protoplasm. Each osteoblast becomes divided into two parts, a perinuclear granular portion and a peripheral, usually basilar, hyaline portion. The hyaline substance derived from osteoblasts fuses to form a mass in which tibrUs differentiate after the hyaline substance is separated from the peri- nuclear protoplasm. Bell (1909) gives a clear description of the development of adipose tissue. He supports the view of the histogenesis of the connective tissues adopted by Mall. BIBLIOGRAPHY. (On the development of the connective tissues, with especial reference to man.) Aeby, C. L. : Uber die Symphysis ossium des Menschen nebst Beitragen zur Lehr& vom hyalinen Knorpel und seiner Verknocherung. Zeitschr. f. rat. Med. Bd. 4. 1858. AsKANAZY, M. : Uber das basophile Protoplasma der Osteoblasten, Osteoklasten und anderer Gewebszellen. Centralbl. f. allg. Pathol, usw. Bd. 13, S. 369- 378. 1902. Baedeen, C. R., and Lewis, W. H. : Development of the Limbs, Bodywall, and Back in Man. Amer. Joum. of Anat. Vol. 1. 1901. Beclabd : Uber die Osteose, oder die Bildung, das Wachstum und die Altersab- nahme der Knochen des Menschen. Meckel's Arch. Bd. 6, S. 405^46. 1820. Bell, E. T. : I. On the Occurrence of Fat in the Epithelium, Cartilage, and Muscle Fibres of the Ox. II. On the Histogenesis of the Adipose Tissue of the Ox. Amer. Joum. of Anat. Vol. 9, p. 401. 1909. Bidder, Alp. : Osteobiologie. Arch, f . mikr. Anat. Bd. 68, S. 137-213. 1906. BoLK : Die Seg-mentaldifferenzierung des menschlichen Rumpfes und seiner Ex- tremitaten. III. Beitrage zur Anatomie und Morphogenese des menschlichem Korpers. Morphol. Jahrb. Bd. 27. 1899. IV. Ebenda, Bd. 28. 1899. Boll: Untersuchungen iiber den Ban und die Entwicklung der Gewebe. Zweite Abteilung: Die Entwicklung des fibrillaren Bindegewebes. Arch. f. mikr. Anat. Bd. 8, S 28. 1872. BoNOME, A. : Zur Histogenese der Knoehenregeneration. Virchow's Arch. Bd. 100. 1885. Brosike, G. : Uber die sogenannten Grenzscheiden des Knochenkanalsystems, nebst Bemerkungen iiber die Keratinsubstanzen. Arch. f. mikr. Anat. Bd. 26. 1885. BRiJNiNG, Fr. : Uber das Auf treten des Fettes im Knochenmark in den ersten Lebensjahren. Diss. Freiburg i. Br. 1903. Coca, Arth. F. : Die Bedeutung der Fibroglia-Fibrillen. Eine embryologische Studie. Virchow's Arch. Bd. 186. 1906. Disse, J.: Die Entstehung dds Knochengewebes und des Zahnbeines. Arch, fv mikr. Anat. Bd. 73, S. 563-606. 1909. HISTOGENESIS OP THE CONNECTIVE TISSUES. 313 Ebnee, v.: Uber den feineren Bau der Knochensubstanz. Sitzungsberichte der Kais. Akademie der Wissenschaften. Wien. Bd. 72, S. 49. 1875. Sind die Fibrillen des Knocliengewebes verkalkt oder nicht? Arch f mikr Anat. Bd. 29. 1887. Egger, G. : Experimentelle Beitriige zur Lebre vom interstitiellen Knoehen- wachstum. Virchow's Arch. Bd. 99. 1885. FiBlCH, K. : Beitrag zur Kenntnis der Histologie des hyalinen Knorpels Anat Anz. Bd. 24, S. 209-214. 1904. Flemming, W. : Zur Entwicklungsgeschichte der Bindegewebesflbrillen. Intemat. Beitrage zur ■wissensch. Medizin. Virchow's Festschrift, Berlin, 1891. Uber die Entwicklung der eollagenen Bindegewebesflbrillen bei Amphibien und Saugetieren. Archiv. f. Anat. u. Physiol. S. 171. 1897. Die Histogenese der Stiitzsubstanzen der Bindegewebsgruppe. Hertwio^'s Handb. d. vgl. Entwicklungsgesch. 1906, ersehienen 1902. Feiedlander, Friedr. v.: Beitrag zur Kenntnis der Architektur spongioser Knochen. Anat. Hefte. Bd. 23, Heft. 72. 1904. .FrOmm, Wald. : Beitrag zur Kasuistik der kongenitalen Knorpelreste am Halse. Diss. Miinchen 1904. (Literatur. 63 Falle.) Feoriep, a. : Kopf teil der Chorda dorsalis bei menschlichen Embryonen. Festschr. f. Henle. 1882. Gaemaschew, W. p. : Veranderungen des Knochenmarks wahrend des WachstumS. Diss. St. Petersburg, 1902. (Russisch.) Gegenbaue, C. : Uber die Bildung des Knochengewelies. Jen. Zeitschr, f . Med. u. Naturw. Bd. 1, S. 343-369. 1864; Bd. 3, S. 206-246. 1867. Geipel, p.: Uber elastisches Gewebe beim Embryo und in Geschwiilsten. Cen- tralbl. f. allg. Pathol, usw. Bd. 17, Nr. 14. 31. Juli 1906. Of. Nakai, Virchow's Arch. Bd. 182. 1906. (Man and fowl.) GoLOWlNSKi, J. : Zur Kenntnis der Histogenese der Bindegewebsfibrillen. Anat. Hefte. 1. Abt. Bd. 33. .1907. Grandis e Copello : Studi sulla composizione chimica delle ceneri delle cartilagini in relatione col proeesso di ossiflcazione. Arch. Sc. Med., p. 175-183. Vol. 26. 1902. Geekow, a. W. : Uber die morphologischen Veranderungen bei Rohrenknochen im Verlaufe des Kindesalters. Diss. St. Petersburg. 1903. (Russian.) Hammae, J. A. : Zur Kenntniss des Fettgewebes. Arch, f . mikr. Anat. Vol. 45. 1895. Hansen, C. C. : Uber die Genese einiger Bindegewebsgrundsubstanzen. Anat. Anz. Bd. 16, S. 147. 1899. Hellbe, G. : Beitrage zur Histogenese der elastisehen Fasem im Netzknorpel und Ligamentum nuchae. Monatsschr. f. prakt. Dermat. Bd. 14, S. 217-237. 1892. Hbbtwig, 0. : Uber die Entwicklung und den Bau des elastisehen Gewebes im Netzknorpel. Arch. f. mikr. Anat. Bd. 9, S. 80. 1873. HoEWiTZ, K. : Uber die Histologie des embryonalen KnochenmarkeB. Wien. med. Wochenschr. Jg. 54, S. 1449-1453, 1499-1503, 1544-1547, 1582-1584, 1631 bis 1634. 1904. HuGONNENQ : Sur la fixation des bases alcalines dans le squelette mineral du foetus pendant les cinq demiers mois de la grossesse. C. R. Acad. Sc. Paris. T. 130, p. 941. 1900. Kassowitz, M. : Die normale Ossifikation usw. Strieker's med. Jahrb. 1879-1880. KoLLiKEE, A. : Die normale Resorption des Knochen^ewebes und ihre Bedeutung fiir die Entstehung der typischen Knochenformen. Leipzig 1873. Zur Entwicklung des Fettgewebes. Anat. Anz. Bd. 1, S. 206-208. 1886. KoLLER, H. : 1st der bindegewebig vorgebildete Knochen imstande, Knorpel zu bilden? Arch. f. Eutwicklungsmechanik. Bd. 3. 1892. 314 HU:\IAN EiIBRY0L0C4T. KoLLjiANN, J. : Die Entwicklung der Chorda dorsalis beim Menscben. Anat. Anz. Bd. 5. 1890. Die Rumpfseg-mente mensehlieher Embryonen von 13-35 Invirbeln. Arch. ±. Anat. u. Physiol., Anat. Abt. 1891. Korff: rber die Entwicklung der Zahnbein- und KnoeheugTundsubstanz der Siiugetiere. Anat. Anz. Ergh. S. 132. 1906. KuSKOw, N.: Beitriig-e zur Kenntnis der Entwicklung des elastischen Gewebes. Arch. f. mikr. Anat. Bd. 30, S. 32-38. 1887. Lagues.se. E :. Sur I'histogenese de la fibre collagene et de la substance fondamen- tale dans la capsule de la rate chez les selaciens. Arch. d'Anatomie micro- scop. Vol. 6, p. 99. 1903. Landois, L.: Untersuchungen iiber die Bindesubstanz und den Yerknocherung- sprozess derselben. Zeitschrift f. Zool. Bd. 16, S. 1-26. 1866. Leboucq, H.: Etudes sur I'ossification. Evolution du Cartilage embryonnaire chez les mammiferes. Bull. Acad. Roy. Belg. Vol. 14, p. 457-459. 1877. Leser, E.: Uber histologische Yorgiinge an der Ossiflcationsgrenze, mit besonderer Beriicksichtigung des Verhaltens der Knorpelzellen. Arch. f. mikr. Anat. Bd. 32, S. 214-222. 1888. LoiSEL, G. : Formation et evolution des elements du tissu elastique. These de Paris 1896. Lowe, L. . Uber die Umwandlung der Osteoklasten im Knoehenmark, nebst Bemer- kungen fiber Knochenwachstum. Arch. f. mikr. Anat. Bd. ]6, S. 618-626. 1879. Lwoff: Uber die Entwicklung der Fibrillen des Bindegewebes. Sitz.-Ber. d. Wiener Akad. d. Wiss. 1\I.-N. Kl. Bd. 98, Abt. 3, S. 184. 1889. Mall, F. : Das retikulierte Ciewebe und seine Beziehungen zu den Bindegewebes- fibrillen. Abhandl. d. K. Siiehs. Gesellsch. d. Wiss. Bd. 17. 1891. The Development of the Connective Tissues from the Connective Tissue Syncy- tium. Amer. Journ. Anat. 1902. Masuoka : Die Regeneration des Knorpelgewebes. Virehow's Arch. Bd. 175. 1904. Merkel: Zur Histogenese des Bindegewebes. Verhandl. d. anatom. Gesellschaft, Basel. S. 45. 1895. Betrachtungen fiber die Entwicklung des Bindegewebes. Anat. Hefte, Bd. 38, S. 321. 1909. MiJLLER, Heinr. : Uber die Entwicklung der Knochensubstanz nebst Bemerkungen fiber den Bau rachitischer Knochen. Zeitsehr. f. wiss. Zool. Bd. 9, S. 147- 233. 1859. Uber Verknocheinmg. Eine Erwiderung an N. Lieberkuhn. Wurzburger naturw. Zeitsehr. Bd. 4. 1863. Nakai, Motokichi : Uber die Entwicklung der elastischen Fasern im Organismus und ihre Beziehungen zu der Gewebsfunktion. 1 Taf. Virehow's Arch. Bd. 1S2, S. 153. 1905. Neumaxx : Beitriige zur Kenntnis des normalen Zahnbein- und Knochengewebes. Leipzig 1863. Nicolas, A. : Developpement et structure des os. Traite d'Anatomie humaine par Poirier et Charpy. I. Paris 1899. Ollier: Du Perioste, 1863. Traite de la regeneration des os. 2 volumes. Paris 1867. Pacchioni, D. . Untersuchungen fiber die normale Ossifikation des Knorpels. Jahrb. f. Kinderheilk. Bd. 26, S. 327-340. 1902. POMMER, G. : Uber die Osteoklastentheorie. Virehow's Arch. Bd. 92. 1883. Rabl, C. : Theorie des Mesoderms. Morph. Jahrb. Bd. 15. 1889. Kasibaud et Renault: Origine et developpement des os. Paris 1864. HISTOGENESIS OF THE COXNECTIVE TISSUES. 315 Ranvier, L.: De quelques points relatifs a preparation et aux proprietes des cellules du cartilage. Joumal de la physiol. T. 6, p. 577. 1865. Quelques faits relatifs au developpement du tissu osseux. C. R Acad Sc Paris. Vol. 77, p. 1105-1109. 1873. Nouvelles recherelies sur la structure et le developpement des tendons Arch de Physiol. Vol. 6, p. 181-202. 1874. Developpement du tissu conjunctif. Traite technique d'Histoloe-ie, u 405- 409. 1S7.5. » ' 1- Renaut, J. : Sur la formation cloisonnante du cartilage hyalin foetal C R Acad Se. Paris. T. 104, p. 1452-1455. 1887. Sur la tramule du tissue conjunctive. Arch. Anat. Microsc. T. 6, p. 1. 1903. Retterer, Ed.: Les decouvertes reeentes relatives au developpement du tissu conjunctif. Journ. de I'Anat. et de la Physiol. Annee 28, 1892, p. 211. Sur le developpement des cavites closes tendineuses et des bourses muqueuses. C. R. de la Soc. de Biol., Ser. 10. T. 2. 1895. Sur le developpement morphologique et histologique des bourses muqueuses et des cavites peritendineuses. Journ. de I'Anat. et de la Physiol Annee 2'^ p. 256. 1896. ' Similitude des processus histogenetiques chez I'embrv'on et I'adulte. Journ. de I'Anat. et de la Physiol. Annee 36, p. 358. 190o! Evolution du' cartilage transitoire. Jouni. de I'Anat, et de la Physiol. Annee 36, p. 467-565. 1900. Structure et evolution du tegument externe. Journ. de I'Anat. et de la Physiol Annee 30, p. 337. 1904. Structure et histogenese de I'os. Journ. de I'Anat. et de la Phvsiol. Annee 41 1905. Histogenese des tissus flbreux et flbrocartilagineux. C. R. Soc. Biol. Pari'! T. 58, p. 240-243. 1905. Evolution du tissu osseux. Journ. de I'Anat. et de la Physiol. Annee 4'' p 193 bis 238. 1906. Des elements qui servent a la croissanee et a la renovation du derme, etc. Journ. de I'Anat. et de la Physiol. Annee 42, 1906. De I'osteogenese et du developpement variable des elements de la substance osseuse. Compt. rend. Soc. Biol. Vol. 64. 1008. De I'ossification infracartilagineuse ou enchondrale. f'ompt. rend. Soc. Biol. Vol. 64. 1908. Rollett, a.: Entwicklung des Bindegewebes. Strieker's Handbuch. Bd. 1, S. 61, 1871. Uber die Entwicklung des fibrillaren Bindegewebes. Unters. Inst. f. Phvs. u. Histolog. Graz. S. 257-265. 1873. Salvioli, I. : Contributo alio studio dell' accresciniento del tessuto connettivo. Arch, per le Sc. Med. Torino. Vol. 13, p. 281-290. 1889. ScHULTZE, 0. : rber embryonale und bleibende Segmentierung. Verb. d. Anat. Ges. 1896. SCHWALBE, G. : Uber das postembryonale Knochenwachstum. Sitzungsberichte der medizin.-naturwiss. Gesellschaft zu Jena 1877. Schwalbe's Jahresbericht. Bd. 6, S. 73. 1877. Schwegel: Die Entwicklungsgeschiehte der Knochen des Stammes und der Ex- tremitaten. Sitz.-Ber. d. Akad. Wien. math.-pliys. Kl. Bd. 30. 1858. Sharpey: Quain's Anatomy, 5th Ed. 1846. SOLGER, B. : Uber perieellulare und intercellulare Ablagerungen im Hyaliuknorpel. Arch. f. mikr. Anat. Bd. 34, S. 408^28. 1889. Ueber Knorpelwachstum. Verb. anat. Ges. Berlin, III. S. 67-71. 1889. Spalteholz, W. : Uber die Beziehungen zwisehen Bindegewebesfasern und Zellen. Anat. Anz. Ergli. Bd. 29. 1906. 316 HmiAN EMBRYOLOGY. Spina : Untersuohungen fiber die Bildung der Knorpelgrundsubstanz. Sitzb. Wieiu Akad. Bd. 81, S. 28-39. 1880. Spuler, a.: Beitrage zur Histologie uiid Histogenese der Binde- und Stiitzsub- stanz. Anat. Hefte. Bd. 7. 1897. Also Eilanger Habilit.-Schrift. Stieda, Ludw. : Die Bildung des Knorpelgewebes. Leipzig 1872. Studien fiber die Entwicklung der Knochen und des Knochengewebes. Arch. f. mikr. Anat. Bd. 11. 1875. Stbelzoff, Z. G. . Uber die Histogenese der Knoclien. Unters. a. d. path. Inst, zu Ziirich. Herausgegebeu von C. J. Eberth, Leipzig. 1873. Stricker : Uber die Bindesubstanzen im Allgemeinen und iiber die Gewebsentwick- lung im Centralnervensystem. Anz. Ges. d. Arzte Wien. S. 41^6, 48-52. 1879. Sbdinko, 0. V. : Beitrag zur Histologie des Knorpels. Anat. Anz. Bd. 22, S. 437 bis 446. 1903. Studnicka, F. K. : Uber kollagene Bindegewebsflbrillen in der Grundsubstanz des Hyalinknorpels, im Dentin und im Knochengewebe. Anat. Anz. Bd. 29, S. 334. 1906. SziLY, AuEBL V. : Uber das Entstehen eines flbrillaren Stfitzgewebes im Embryo und dessen Verhaltnis zur Glaskoperfrage. Anat. Hefte. Bd. 35, S. 649- 758 (literature). 1908. Tbiepel, H. : Die Anordnung der Knochenflbrillen in transf ormierter Spongiosa. Anat. Hefte. Bd. 33, S. 49. 1907. ViRCHOW : Uber die Identitiit von Knochen-, Knorpel- und Bindegewebskorperchen, sowie fiber Sehleimgewebe. Verhandl. der Wfirzburger phys. med. Gesell- schaft. Bd. 2, S. 150 und S. 314. 1852. Waldbyer, W. : Uber den Ossifikationsprozess. Arch, f . mikr. Anat. Bd. 1, S. 354-374. 1865. Uber Bindegewebszellen. Arch. f. mikr. Anat. Bd. 11, S. 176-194. 1874. Kittsubstanz, Grundsubstanz, Epithet und Endothel. Arch, f . mikr. Anat. Bd. 57, 1901. Welker, H. : Uber Wachstum und Bau des menschlichen Schadels. Leipzig 1862. WiNHOLD, H. : Uber das Vorkommen von Megaloblasten im Knoehenmark. Diss. Leipzig 1901. Wolff, J. : Das Gesetz der Transformation der Knochen. Berlin 1885. Uber die Theorie des Knochenschwundes durch vermehrten Druek und der Knochenanbildung durch Druckentlastung. Arch. f. klin. Chir. Bd. 13, S. 302 bis 324. 1891. PAKT II. Morphogenesis of the Skeletal System. A. GENERAL FEATURES. The definitive skeletal system is composed of bones and car- tilages united to one another at joints by means of ligaments. In the lowest vertebrates a cellular rod, the chorda dorsalis or noto- chord, situated in the mid-axial line ventral to the central nervous system, constitutes the chief part of the axial skeleton. In the higher vertebrates a chorda dorsalis is also formed during early embryonic development, though in mammals and man it lends little or no skeletal support to the embryo and mere derivatives of it are to be found in the adult. The definitive skeleton of the MORPHOGENESIS OP THE SKELETAL SYSTEM. 317 higher vertebrates, including man, is differentiated from the mesenchyme of the head, trunk, and limbs. The process of differ- entiation is somewhat complex. As a rule, the first visible step in the process is marked by condensation in the sclerogenous mesen- chyme or scleroblastema. Thus, in the development of the skele- ton of the inferior extremity, condensation begins in the vicinity of the future hip-joint and from here extends distally and proximally, so that there is produced a continuous mass of condensed tissue in which pelvic, femoral, tibio-fibular, and tarsal regions and five metatarso-phalangeal rays may be distinguished. The hard parts of the skeleton are developed from centres which appear in the scleroblastema. The joints are developed in the scleroblastema which intervenes between the hard parts. THE BONES. It has been mentioned in the section on histogenesis that most of the bones of the body are first formed of cartilage and then, during subsequent development, bone is gradually substituted for cartilage, substitution or cartilaginous bones (Figs. 277 and 278). Other bones are formed directly in the membranous sclero- blastema, membrane or investment bones. The bones of the extremities, with the partial exception of the clavicle, the bones of the spinal column and thorax, and the greater part of those of the base of the cranium, have a chondrogenous origin. The greater part of the bones of the cranial vault and of the face arise directly in the scleroblastema. It is to be noted, however, that during the formation of many of the typical substitution bones ossification may extend into membranes attached to the cartilage, so that certain processes on these bones are membranous in origin, and that, on the other hand, certain parts of bones of membranous origin may second- arily give rise to cartilage (temporomandibular joint). Several of the definitive bones of the skull have an origin partly cartilaginous, partly membranous. SUBSTITUTION BONES. As a rule, a centre of chondrifieation appears in the midst of condensed scleroblastema. (See femur, tibia, and fibula, Fig. 275.) It may, however, appear in tissue but slightly condensed, as in case of the vertebral bodies. Fig. 273. The cartilaginous centres expand rapidly, both by apposition from the surrounding blastema and by interstitial growth. Neigh- boring centres are thus soon brought into close approximation. Some of the centres fuse with one another in the region of approxi- mation. Between other centres joints are developed. The fate 318 HUilAX E.MBRYOLOGY. of the cartilaginous centres, therefore, differs considerably in different regions. The conditions in the skeleton of the limbs are the simplest. Here for each of the bones, including the pubis, ischium, and ilium, there is a single centre of chondrification (see Fig. 226 and Figs. 275 and 276). The clavicle forms an exception to the other bones in that the tissue at the centre of chondrification is not converted into typical embryonic hyaline cartilage (see pp. 380 and 388). The centres of chondrification for the pubis, ischium, and ilium soon fuse with one another so as to produce a continuous carti- laginous hip-bone, which gradually assumes definitive form (Figs. 276, 277, and 278). With the exception of a few cartilages in the wrist, the fate of which is treated elsewhere (p. 383), each of the other embryonic limb cartilages undergoes an independent develop- ment. In the region of the knee-joint, however, and possibly in some other articular regions of the limbs, independent skeletal elements become at an early period temporarily fused together by a kind of precartilage (Fig. 283). Temporary joints of this kind resemble the permanent joints of the shark's fin. A centre of ossification appears in the main body of each of the cartilages of the skeleton of the limbs ; in most of them early in fetal development, but not until after birth in those of the ankle and wrist, with the exception of the calcaneus, talus and cuboid and in the patella and other sessamoid bones. These chief centres of ossification establish bone in place of cartilage as growth pro- ceeds. In case of all the limb bones except those of the ankle and wiist secondary' epiphyseal centres of ossification appear early in childhood in those portions of the bone still cartilaginous, and as maturity is approached become fused with the main part of the bone. Growth in length of bone, as stated in the section on his- togenesis (p. 311), is dependent upon the growth of the cartila- ginous matrix and ceases when the epiphyses become fused with the main body of the bone. In the adult limb skeleton the only cartilage remaining is that upon the joint surfaces of the bones. In the vertebral column there are two bilaterally placed centres of chondrification for the body of each vertebra and one for each half arch (Figs. 239 and 249). The arch cartilages join the body considerably before they unite dor sally so as to complete the arch about the spinal cord. The ribs develop from separate centres of chondrification and do not fuse with the bodies. In the cervical, lumbar, and sacral regions there are more or less distinct centres of chondrification of costal elements which quickly fuse with the cartilage of the body. In the sacral region the various cartilages fuse to form a cartilaginous sacrum. The cartilaginous vertebral bodies are at first separated by thick blastemal discs, but as development proceeds the discs near the centre become thin MORPHOGENESIS OP THE SKELETAL SYSTEM. 319 and partially converted into a precartilaginous tissue, so that for a brief period there is a continuous vertebral axis composed of tissue of a cartilaginous nature but in which segmentation is clearly marked. The cartilage of the sternum arises mainly from the cartilage of the ribs, from which it is secondarily separated by the formation of costosternal joints. There are primary centres of ossification for the bodies of the vertebrae, each half arch, the ribs, and some of the costal elements of the sacrum. In addition, there are many epiphyseal centres. In the cranial blastema numerous centres of chondritication appear (Figs. 310 and 311). These, however, fuse to form a con- tinuous chondrocranium, in which no blastemal sutures remain to separate one cartilaginous element from another (Figs. 312 and 313). The incus and stapes remain distinct cartilages. The malleus is long continuous with Meckel's cartilage, the cartila- ginous skeleton of the mandibular arch. The cartilage of the hyoid arch becomes attached to the chondrocranium. While the chondrocranium is being formed, centres of ossifi- cation begin to appear in various parts of the cranial scleroblas- tema. From these centres of ossification, partly by expansion and partly by fusion of neighboring centres, there are produced the membranous bones of the skull (Fig. 321). Meanwhile, centres of ossification appear in the chondrocranium and by expansion and fusion give rise to the substitution bones of the skull. In the definitive skull some bones, like the parietal, frontal, and maxil- larv, are purely membranous in origin. Some, like the ethmoid, hyoid, incus, and stapes, are fairly typical substitution bones, while many of the bones, like the occipital, sphenoid, and temporal bones, arise partly from centres which appear in membranous tissue, partly from centres which appear in the chondrocranium. In the membra-nons tissue in which the centres for the investment bones appear the definitive form of the skeletal part is much less clearly marked than in the chondrocranium (compare Figs. 310, 311, 312, 313, 321). CAETILAGES. Not all the cartilage of the embryonic skeleton becomes re- placed by bones. Some of the embryonic cartilages become reduced to fibrous tissue, as in the case of the stylohyoid ligament; some give origin to the cartilages of the definitive skeleton, such as the costal cartilages and parts of the nasal capsule; some merely disappear. 320 HUMAN EilBRYOLOGY. JOINTS. When first differentiated the fixed parts of the skeleton are nnited to one another by dense blastemal tissue m which httle definite form is to be observed. In case of synarthroses this inter- vening blastemal tissue becomes directly or indirectly transformed into fibrous tissue (svndesmosis), into cartilage (synchondrosis), or into bone (svnostosis). While, as a rule, the fibrous tissue of a s}Tidesmosis comes fairly directly from the primitive blastema of the embryonic joint, it may arise as the result of retrograde metamorphosis of cartilage (lig. stylohyoideum). A synchon- drosis is usually preceded by an embryonic blastemal syndesmosis. A synostosis is usually preceded by a syndesmosis or a synchon- drosis.'' In a diarthrosis the joint cavity, synovial membrane, and the various ligaments characteristic of the joint are differentiated from the dense blastemal tissue which unites at first the two embryonic cartilages entering into the joint. Disci articulares and menisci articulares are also differentiated from this blastema. In case of the few diarthroses formed between membrane bones, as for instance between the mandible and the temporal bone, the blastemal tissue has the power of giving rise to cartilage which covers the joint surfaces of the bones. The various steps in the differentiation of a simple diarthrosis are well illustrated in the digital articulations (Figs. 226-228). In Fig. 226 are shown the cartilaginous anlages (a) of the three phalanges and the distal part of the metacarpal of a finger of an embryo 2.7 cm. long. These cartilaginous anlages are embedded in a dense blastema which shows lighter areas in the vicinity of the future joints (c). The term intermediate zone has been applied to the dense tissue lying between the two cartilages entering into a joint (b). As the cartilages expand they come into close approx- imation, as shown in the finger of a fetus 7 cm. long (Fig. 227). At this stage the cartilage is undergoing changes preliminary to ossification. The perichondrium about the joint surfaces of the cartilage entering into the joint is very dense. The joint cavity first appears at the periphery of the joint (Fig. 227). Gradually it extends in between the two cartilages entering into the joint and a variable distance over the head toward the shaft (Fig. 228, A, B, C). The form of the joint surfaces of the bones entering into the joint is highly differentiated before the joint cavity appears (Fig. 227). In the more complex joints in which menisci or intra-articular ligaments are differentiated, as in the knee-joint and hip-joint 'The nucleus pulposus of the intervertebral fibrocartilage (disc) arises from the tissue of the chorda dorsalis (see p. 341). MORPHOGENESIS OP THE SKELETAL SYSTEM. 321 a- m Fig, 226 Fig. 227. A. B. Fig. 228. c. FiGa. 226-228. — (After Schulin, Archiv f. Anat. u. Physiol., Anat. Abt., 1879.) Fig. 226. (Schulin, Fig. 1.) Finger of an embryo 2.7 cm. long, a, primordial cartilage; b, intermediate zone; c, region of loose tissue in preparation for the formation of a joint cavity. Fig. 227. (Schulin, Fig. 2.) Articulation of the second phalanx of the middle finger of a fetus 7 cm. long, with the basal and terminal phalanges, a, formation of joint cavity on the dorsal side. Fig. 228. (Schulin, Figs, 4, 5, and 8.) The first interphalangeal joint of the middle finger of a fetus 13 cm. long (A), of a fetua 20 cm. long (B), and of an adult (C). a, joint cavity; b, epiphyseal cartilage; c, bone. Vol. I.— 21 322 HUMAN EMBEYOLOGT. (Figs. 281, 282, 285), the cartilages of the bones entering into the joinl are less closely approximated at the time of the formation of the joint cavity than in simple joints, like those of the fingers. The external ligaments and the various intra-articular structures are differentiated directly from the intermediate zone of blastema, while the blastemal tissue next the joint surfaces of the cartilages entering into the joint becomes condensed into a dense perichon- drium. The rest of the tissue becomes less dense in character and is converted into mucoid tissue with a few cells scattered through the matrix (Fig. 282). As in all diarthroses the formation of the joint cavity begins at the side and extends toward the centre of the joint. The definitive cavity may be formed by the fusion of several cavities which appear at various places in the periphery of the joint (knee-joint, p. 372). The mucoid tissue disappears as the joint cavity enlarges. The capsular ligament which is formed from the periphery of the intermediate blastemal zone is continuous on each side of the joint at first with the perichondrium and later with the periosteum. The synovial membrane is formed on the inner surface of the capsular ligament. Synovial villi arise in the latter part of fetal life. At the time of the appearance of the joint cavity the bones entering into the joint are composed of cartilage in the region of the articulation, although ossification may be well under way at some distance from the articulation (Figs. 227 and 228). After the appearance of the joint cavity the articulating parts undergo an elaboration in form (Fig. 228), which may be quite extensive (Figs. 286, 288). This elaboration of form is due not only to interstitial growth of cartilage, but also to the appositional growth of bone. As the result of the ossification, all the cartilage near the joint becomes entirely replaced by bone except on the joint surface, where, as a rule, a layer of hyaline cartilage remains throughout life. The thin, dense layer of blastemal perichondrium which for a short time covers the joint cartilage, as a rule dis- appears early, although it may give rise to a permanent film of tissue or the joint cartilage may become in part composed of fibrocartilage (sternoclavicular, temperomandibular, costoverte- bral, sternocostal articulations). The relative positions of the articulating bones vary greatly in different regions at the time of the formation of the joints. The knee- and elbow- joints, for instance, are flexed at an angle of about 90° while the wrist-joint is nearly straight. SESAMOID BONES. Tendons are closely fused to the joint capsule in many articu- lations of the extremities. In certain regions where this occurs sesamoid bones are developed. The largest of the sesamoid bones MORPHOGENESIS OF THE SKELETAL SYSTEM. 323 is the patella. Well-marked sesamoid bones are fomid regularly on the flexor side of the metacarpo- and metatarsophalangeal joints, usually of the tirst and frequently of the other digits of the hand and foot. Dorsally placed sesamoid bones have also been seen in connection with the thumb. On the flexor surface of the thumb a sesamoid bone is frequently found at the interphalangeal joint. Fibrous interphalangeal sesamoids have been found in connection with the fingers. The sesamoid bones are better devel- oped in some of the lower mammals than in man, and, according to Pfitzner, are more frequent in the human embryo than in the adult. They are developed at the periphery of the intermediate blastemal zone. The blastema becomes condensed, and then in the better marked sesamoid bones becomes gradually transformed into cartilage. Ossification takes place relatively late in childhood. On the intracapsular origin of the sesamoid bones see Bradley (1906). In some tendons not intimately connected with a joint capsule a sesamoid bone may be developed in a region where the tendon is subjected to stress against a bone about which it turns. An example is the sesamoid bone often found in the tendon of the peroneus longus where this plays over the tuberosity of the cuboid. According to Lunghetti (1906), the sesamoid bone in the tendon of the M. peroneus longus develops in fibrous connective tissue, not in cartilage. It is commonly stated that it passes through a fibro- cartilaginous stage before becoming ossified. VARIATION IN THE DEVELOPMENT OF THE SKELETON. Variations in the bones of the adult human skeleton are frequent. Thus, for instance, skeletons with only eleven or with thirteen free ribs are not uncommon. Rosenberg, Pfltzner, Thilenius, and others would ascribe some of the variations found in the adult skeleton to the chance persistence of transitory conditions normally present in the embryonic or fetal skeleton and supposedly of phylogenetie importance. The studies of Thilenius, Bardeen, Mall, and others have shown, however, that the skeleton of the embryo is subject to fluctuating variations like those found in the adult. At present there are not sufficient data to determine definitely the relative frequency of skeletal variations in the adult compared with those in the embryo or fetus. ABNORMALITIES IN THE DEVELOPMENT OF THE SKELETON. The form of the skeleton as a whole and of the individual bones which compose it depends partly upon heredity, partly upon the mechanical and chemical influences to which it is subject during growth. The variations which are a normal inheritance of the race, including such extreme forms as individuals with six toes or six fingers, are to be distinguished from the abnormalities of structure due to unfavorable environment either within or without the body. In the main the shapes of the bones and joints are inherited, but to some slight extent both bones and joints are moulded by the experience of the individual. Abnormal stress of muscular or other origin, and abnormal lack of stress, as in cases of muscle paralysis, both give rise to bones and joints abnormal in form. 324 HU.MAX EAIBRYOLOGY. During development the skeleton is markedly influenced by internal chemical conditions affecting the growth or general nutrition of the body. The skeleton in some cases seems to be the part primarily affected. The skeletal lesions vary all the way from a retardation in the time of appearance of centres of ossification to the failure of a part of the skeleton to develop or to hyperplasia and abnormal form-differentiation. Agenesis, or failure of skeletal development, may be due either to primary lack of origin of a part or to an affection which destroys the skeletal anlage after it has begun to differentiate. It is most frequently found in the cranial vault and in the vertebral arches, less frequently in the vertebral bodies and the bones of the extremities. The osseous defect is usually, but not always, associated with other marked physical deformities. Hypoplasia, underdevelopment, of the skeleton, whether generalized or con- fined to a part, may be due either to prenatal or to postnatal conditions. The failure of the bones to develop normally may be due (1) to lack of active proliferation of cartUage (characteristic of cretins), (2) to inactivity in the process of ossification, membranous, subperiosteal or endochondral (see Michel, 1903, Lindemann, 1903), (3) to a premature union of epiphyses with the main part of a bone, (4) to growth of connective tissue between the growing cartilage of a bone and the region where ossification usually extends into the cartilage (micromelia chondromalacia, fetal rickets), and (5) to inflammation and other abnormal conditions affecting the growing parts of the bone. ^'arious congenital forms of hypoplasia are recognized, — microsomia, micromelia, micromelia chondromalacia (fetal rickets), cretinism, etc. In most instances while there is a general underdevelopment of the skeleton the long bones are especially affected and appear short and relatively thick; the pelvis and thorax are also usually abnormally small, and the root of the nose is broad and not infrequently sunken in. The causative factors of these conditions are obscure. In cretinism growth of cartilage is retarded and there is a delay in the appearance of centres of ossification and also in the fusion of epiphyses with the main parts of the bones (Wyss, 1900). In this disease there is good evidence that the failure of development of the body, including the skeleton, is due to lack of normal secretion by the thyroid gland. It is not improbable that the secretions of other glands of similar type may affect the development of the skeleton. Some diseases involving both the skeleton and the hypophysis have led to the belief that there is a relation between this gland and skeleton development. This relation has, however, been disputed (Arnold, 1894). K. Bach (1906) has recently discussed the apparent influence of the thymus on the growth of bones. Hyperplasia, overgrowth of the bones, is due (1) to an excessive activity of membranous or subperiosteal ossification or (2) to a prolonged persistence of actively growing epiphyseal cartilages, union of epiphysis with the main part of the bone being delayed, while endochondral ossification continues beyond the usual time. Hyperplasia may be local or general and may give rise to a well-proportioned or to disproportionate enlargement of the skeleton. It is stated that removal of the testicles early in infancy or congenital absence of the testicles may lead to an excessive prolongation of the activity of the epiphyseal cartilages and hence to gigantism (P. Launois and P. Roy, 1903, Poncet, 1903). Phosphorus and arsenic in small doses are said to promote bone growth. Partial hyperplasia is found most frequently in the skull and in the bones of the hands and feet. An irritative stimulus, such as a blow, may excite excessive growth of bone. In young people a small centre of inflammation (tuberculosis, osteomyelitis) in the diaphysis may excite activity in the processes concerned in ossification and induce abnoTOial growth in size of bone. If the centre of inflammation is near the epiphy- seal cartilage, ossification is apt to be very irregular. MORPHOGENESIS OF THE SKELETAL SYSTEM. 325 In congenital syphilis there are frequently, although not always, present characteristic irregularities in the deposition of calcium salts and in the formation of narrow cavities in the ossifying cartilage. This sometimes gives rise to marked abnormality of form. In rickets the process of bone absorption is abnormally active, while the formation of new bone is characterized by lack of deposit of the nonnal amount of calcium salts. In endochondral ossification there is no well-marked zone of calcification. The bones are abnormally thick, clumsy, and heavy and may be much distorted. In teratomata of various forms the skeletal abnormalities correspond with those of the rest of the body. BIBLIOGRAPHY. (On general features of the morphogenesis of the human skeleton.) Aeby, Chr. : Der Bau des menschliehen Korpers. Leipzig 1871. Abnold, J. : Weitere Beitrage zur Akromegaliefrage. Virchow's Arch. Bd. 135, S. L 1894. Bach, K. : Zur Physiologie und Pathologie des Thymus. Jahrb. f. Kinderheil- kunde. Bd. 64. 1906. Bade, P. : Die Entwicklung des menschliehen Skeletts bis zur Geburt. Arch. f. mikr. Anat. Bd. 55, S. 245-290. 1900. Baedeen and Lewis: Development of the Limbs, Bodywall, and Back in Man. Amer. Journ. of Anat. Vol. 1, p. 1. 1901. Bardeen, C. R. : Vertebral Variation in the Human Adult and Embryo. Anat. Anz. Bd. 25, S. 497. 1904. Beclaed : Uber die Osteose oder die Bildung, das Wachstum und die Altersabnahme der Knochen des Menschen. Meckel's Arch. Bd. 6, S. 405-446. 1820. Bernays, a.: Die Entwicklungsg-eschichte des Kniegelenkes des Menschen mit Bemerkungen liber die Gelenke im allegemeinen. Morphol. Jahrb. Bd. 4, S. 403. 1878. BOLK : Die Segmentaldifferenzierung des menschliehen Rumpfes und seiner Ex- tremitaten. Beitrage zur Anatomie und Morphologie des menschliehen Korpers. Morphol. Jahrb. Bd. 27. 1899. Bd. 28. 1899. Sur la signifieation de la sympodie au point de vue de I'anatomie segmentale. Petrus Camper Deel 1. 1901. Bradley, 0. C. : A Contribution to the Development of the Interphalangeal Sesamoid Bone. Anat. Anz. Bd. 28, S. 528-536. 1906. Bruch, C. : Beitrage zur Entwicklungsg-eschichte des Knochensystems. Neue Denk- schriften der allgemeinen Schweizerischen Gesellschaft fiir die gesamten Naturwissenschaften. Bd. 12, S. 16. Ziirich 1852. COREIDI, G. : Dei principale nuclei di ossificazione che possono invenirsi all' epoca della nascita. L'anomalo Napoli. Vol. 3, p. 143, 179, 231. 1891. DE COULON, W. : Uber Thyreoidea und Hypophysis bei Cretinen, sowie iiber Thyroichalreste bei struma nodosa. Virchow's Archiv. Bd. 147. 1897. Damany, p. le: L'adoption de I'homme a la station debout. Journ. de I'Anat. et' de la Physiol, p. 135-170. 1905. DoREiEN, Eenst : Uber Riesenwuchs und Elephantiasis congenita. Diss. Leipzig 1905. Eeissonius : Traite des os des enf ants. Cited by Le Double, 1906. Grawitz- Eetus mit kretinistischer Wachstumsstornng des Schadels und der Skelettknochen. Virchow's Arch. Bd. 100, S. 256-262. 1885. Hagen: Die Bildung des Knorpelskeletts beim menschliehen Embryo. Archiv f. Anat. und Physiol. Anat. Abt. S. 1-40. 1900. 326 HUMAX EMBRYOLOGY. Henke u. Rbther: Studien iiber die Entwicklung der Extremitaten des Menschen, insbes. der Gelenkflachen. Sitzungsb. der K. Akad. der Wiss. Math.-Naturw. Klasse. Wien. Bd. 80, S. 217. 1874. Hepburn, D. : The Development of Diathrodal Joints in Birds and Mammals. Journ. of Anat. and Physiol. Vol. 23, p. 507. 1889. Hdter, Carl : Anatomische Studien an den Extremitatengelenken Neugeborener und Erwaehsener. Virchow's Archiv. Bd. 25, 26, 28. 1862-1863. Kaufhann, E.: Untersuehungen iiber die sogen. fetale Rachitis. (Chondrodystro- phia fetalis.) Berlin 1892. Kerckring, Theod. : Spieilegium anatomiciun continens osteogeniam f oetuum. Amstelodami. 1670. Lambertz : Die Entwicklung- des mensehlichen Knochengeriistes wahrend des f etalen Lebens dargestellt an Rontgenbildern. Fortschritte auf dem Gebiete der Rontgenstrahlen. 1. u. 2. Erganzungsheft. 1900. Launois, p. E., et Roy, Pierre: Des relations qui existent entre I'etat des glandes genitales males et le developpement du squelette. C. R. Soc. Biol. T. 55. 1903. Lindemann, p.: Uber Osteogenesis imperfecta. Diss. Berlin 1903. LuBOSCH, \V. : Die Stammesgesehiehtlicbe Entwicklung der Synovialhaut und der Sehnen mit Hinweisen auf die Entwicklung der Kiefergelenke der Siiugetiere. Biol. Zentralbl. Bd. 28, S. 678, 697. 1908. Lundvall, Havar: Weiteres iiber Demonstration embryonaler Skelette. Anat. Anz. Bd. 27. 1905. LuscHKA : Die Halbgelenke des mensehlichen Korpers. Berlin 1858. LuNGHETTi, B. : Sopra I'ossification dei sesamoidi intratendinei. Monit. Zool. Ital. Anno 17, p. 321-322, 1906. Mall, F. P. : On Ossification Centers in Human Embryos. The American Journ of Anat. Vol. 5. 1906. Michel, P.: Osteogenesis imperfecta. Virchow's Arch. Bd. 174. 1903. Morin: Radiographies relatives a la foi-mation et a l^ccroissement du systeme osseux. Assoc. Fran?, pour I'Avancem. des Sc. Sess. 27, p. 678-683 Nantes 1899. Nobiling: Uber die Entwickelung einzelner Yerknocherungskerne in unreifen und reifen Friiehten. Miinchen 1899. PriTZNER, W. : Morph. Arbeiten. 1891-1898. PONCET, Antonin: De I'influence de la castration sur le developpement du squelette. Reeherches experimentales et cliniques. C R. Soe Biol T 55 1903. Rambaud et Renault: Origine et developpement des Os. Paris 1864. Regnault, F. : A propos de la morphogenie osseuse. Bull, et Mem. de la Soc dAnthrop, 1907. Retterer : Ebauehe squelettogene des membres et developpement des articulations. Journ. de lAnat. et de la Physiol. Annee 38, No. 5, p. 473, 580. 1902. Rosenberg, E. : Uber die Entwicklung der Wirbelsaule und des eentrale carpi des Menschen. Morphol. Jahrbuch. Bd. 1, S. 83. 1876. Bemerkungen iiber den Modus des Zustandekoinmens der Regionen an der Wirbelsaule des Menschen. Morphol. Jahrbuch. Bd. 36. 1907. Schubert, C. : Riesenwuehs beim Neugeborenen. Monatsschr. f. Geburtsh. u. Gynakol. Bd. 23, S. 453-456. 1906. ScHULiN, K. : rber die Entwicklung und weitere Ausbildung der Gelenke des mensehlichen Korpers. Arch. f. Anat. u. Physiol. Anat. Abth. S 240-274 1879. Schwegel: Die Entwicklungsgesehichte der Knochen des Stammes und der Extremitaten. Sitzb. Akad. Wien. Math. phys. Kl. Bd. 30. 1858. (Literature.) MORPHOGENESIS OP THE SKELETAL SYSTEM. 327 SiLBERSTEiN, A.: Ein Beitrag zur Lehre von den fetalen Knochenerkrankungen. Arch. klin. Chir. Bd. 70. 1903. Thilbnius, G. : Untersuchungen iiber die morphologische Bedeutung aecessoriselier Elemente am mensehliehen Cai-pus (und Tarsus). Morph. Arbeiten. Bd. 5. 1895. Accessorisehe und echte Skelettstiicke. Anat. Anz. Bd. 13. 1897. Wolff, J. : Uber die Theorie des Knochenschwundes durch vermehrten Druck und der Knoehenanbildung durch Druckentlastung. Arch, f . klin. Chir. Bd. 13, S. 302 bis 324. 1891. "Wyss, von: Beitrag zur Kenntnis der Entwicklung des Skelettes von Kretinen und Kretinoiden. Fortschr. auf d. Gebiete der Rontgenstrahlen. Bd. 3. 1900. B. ORIGIN AND FATE OF THE CHORDA DORSALIS. The cervical region of the chorda dorsalis appears to arise from the dorsal wall of the enterovitelline sac beneath the medul- lary groove, although it is not probable that these cells belong primitively to the entoderm. In many mammals the tissue from which it is derived comes primarily from the mesodermal head or chordal process (see p. 47). Bonnet,'^ however, has ascribed to the yolk entoderm the origin of the tissue for the anterior tip of the chorda in the dog and sheep, although the main part of the anterior portion of the chorda in these animals is derived from tissue which has become incorporated in the entoderm through fusion of the head process with the dorsal wall of the entero- vitelline sac. In the human embryo, as mentioned above (p. 293), there is at a very early period a layer of mesoderm formed between the ectoderm and entoderm in the mid-sagittal plane (Fig. 219, A). At a slightly later period (Fig. 219, B and C) the mesoderm has disappeared in this region. Possibly it is incor- porated with the entoderm which now beneath the neural groove presents a plate of tissue slightly thicker than the entoderm on each side. This is the chordal plate, the anlage of the anterior end of the chorda dorsalis. Kollmann (1890) has given an account of the origin of the chorda in a human embryo 2.11 mm. long with fourteen mesodermic somites. In this embryo (Fig. 220) the neurenteric canal has disappeared. Anterior to the primitive streak in the mid-sagittal plane a longitudinal ridge of cells pro- jects dorsally from the entoderm (Fig. 220, B). This ridge of cells gradually becomes pinched off from the entoderm (Fig. 220, C). In a slightly older embryo described hy Mall (1897) (Fig. 229) the neurenteric canal is represented by a solid column of cells. The chorda extends forward from this column of cells as far as the buccopharyngeal membrane. As Seessel's pocket develops, the chorda remains for a time attached to its posterior wall. This connection is lost^at about the period when the bucco- pharyngeal membrane is ruptured. 'Anat. Hefte, 1901. / 328 HUMAN EMBRYOLOGY. In the embryo described by Mall the posterior end of the chorda lies opposite what is probably the eighth cervical somite. During the development of the thoracic, lumbar, sacral, and coccy- geal regions of the embryo the chorda is gradually developed caudalwards. In this portion of its development the chorda is not first embedded in the entoderm and then again differentiated out, but is derived directly from the primitive streak and from the tissue which replaces the primitive streak caudalwards. In an embryo described by His (L, 2.4 mm. long) the noto- chord has a distinct lumen. This is not present in older embryos. At first there is no distinct membrane about the chorda (see Figs. 220, B, and 220, C). The cells are large, with clear proto- plasm. By the end of the fourth week of development a thin structureless membrane encircles the chorda, which is now about at the height of its development. The chorda is cylindrical. The cells are polygonal and are filled with a fine granular protoplasm. During this period differentiation of the base of the skull and of the spinal column is marked by condensations in the axial mesen- chyme (Fig. 231). Subsequenth" in the spinal region the inter- vertebral discs and the bodies of the vertebrae form about the chorda dorsalis. Between the chorda cells and the outer sheath of the chorda, there appears an inner membrane, apparently mucoid in nature (Williams). According to Minot (1907), at the period when the axial mesenchyme begins to be differentiated into vertebrae the notochord shows slight transient dorso-ventral segmental flexures. Just before ossification begins the chorda dis- appears in the vertebral bodies. In the intervertebral discs it becomes transformed into the tissue of the nucleus pulposus. In the cranial region the posterior and the anterior portions of the chorda dorsalis become embedded in the skeletal tissue of the base of the skull while the intermediate portion lies between the base of the skull and the dorsal wall of the pharynx (Fig. 266). Ultimately the cranial part of the chorda completely disappears. That part of the chorda which lies in the retropharyngeal tissue gives rise to numerous projections and swellings and is the first portion of the chorda to disappear. The posterior portion of the cranial part of the chorda comes to lie on the dorsal side of the basal occipital plate and disappears at the time of ossification of the basi-occipital. The anterior extremity of the chorda persists longer than the retropharyngeal portion, but usually disappears during the ossification of the base of the skull. A description of the changes undergone by the chorda is given in connection with the development of the vertebral column and sloill. Traces of the cranial part of the chorda dorsalis may persist in the adult and give rise to tumors. MORPHOGENESIS OF THE SKELETAL SYSTEM. Fig. 230b^ Fig. 230c.^ ^ ^-^^ 230d. A 329 Fto. 230ci. Fig. 229. — (After Mall, Journ. of Morphology, vol. 12, 1897, Fig. 16.) Outline drawing of a medial sagittal section of the model of Mall's embryo XII. Magn, 50: 1. The heavy line is the aorta. The muscle plates are numbered for occipital, cervical, and thoracic regions respectively. The mesoderm is striated. Am., amnion; a., border between fore-brain and mid-brain; X, X', extent of closure of spinal canal; S, Seessel's pocket; Ch., chorda; b', b", first and second branchial pockets; ov., otic vesicle; M., mouth; T., thyroid; H., pericardial space; Ph., pharynx; Ent., entoderm; St., septum transversum; L., liver; nc, neur^nteric canal; A^Z., allantois; M.r., membranareuniens; CZ., coelom. Fig. 230, a, b, c, d. — Transverse sections through the regions indicated in Fig. 229. Magn. 50 ; 1. The ccelom within the body is represented black. 0^ and O^, first and third occipital myotomes; C^ and C^, first and eighth ceir^ical myotomes; T^, first thoracic myotome; a., aorta; v., omphalomesenteric vein; ^, thyroid; Z., liver; PA., pharynx; i., intestine; tic, neurenteric canal: JIf .r., membrana reuniens. 330 HUMAN EMBEYOLOGY. ENTOCHORDA. A hypochorda or entochorda arising from the entoderm beneath the chorda dorsalis has been found in fishes, amphibia, birds, and reptiles, but apparently has not yet been described for the human embryo. In part the tissue of the hypochorda joins that of the chorda dorsalis. (See Ad. Reinhardt, Morphol. Jahrb., Bd. 32, 1904; Ph. Stohr, Morphol. Jahrb., Bd. 23, 1895; and S. A. Ussofi, Anat. Anz., Bd. 29, 1906.) BIBLIOGRAPHY. The chief paper on the early development of the chorda dorsalis in man is that of Kolhnami (1890). Important data concerning the chorda dorsalis are to be found in the various papers describing human embryos with fourteen somites or less. Bonnet : Beitrage zur Embryologie des Hundes. Anat. Hef te. 1897 und 1901. Eteenod : Communication sur un oeuf humain avec embryon exeessivement jeune. Arch. Ital. de Biologies Vol. 22. 1895. See also Monitore Zool. Ital. Vol. 5, p. 70-72. 1894. Sur un OBuf humain de 16.3 mm. avec embiyon de 2.1 mm. Arch, des Sciences Phys. et Nat. Annee 101. 4 Periode. T. 2, p. 609-624. 1896. II y a un canal notochordal dans I'embryon humain? Anat. Anz. Bd. 16, p. 131-143. 1899. Frassi, L. : Uber ein junges menschliehes Ei in situ. Arch, f . mikr. Anat. Bd. 71, S. 667. 1908. Feoriep, a. : Kopfteil der chorda dorsalis bei menschlichen Embryonen. Beitrage z. Anat. und Embryol. Als Festgabe fiir Jacob tienle. 1882. GiACOMiNi, C. : Un oeuf humain de 11 jours. Arch. Ital. de Biologie. Vol. 29. 1898. Heiberg, J. : Uber die Zwischenwirbelgelenke und Knochenkeme der Wirbelsaule. Mitt. a. d. Embryol. Inst, der K. K. Univ. Wien. I, S. 119-129. 1880. His : Anatomie menschlicher Embryonen. 1880. Janosik, J. : Zwei junge menschliche Embryonen. Arch. f. mikr. Anat. Bd. 30, S. 559. 1887. IvEiBEL : Zur Entvcicklungsgesehichte der chorda bei Saugem. Arch, f . Anat. und Physiol. Anat. Abt. 1889. KOLLsiANN : Die Entwicklung der chorda dorsalis beim Mensehen. Anat. Anz. Bd. 5, S. 308-321. 1890. Die Rumpfsegmente menschlicher Embryonen von 13-35 Urwirbeln. Arch. f. Anat. u. Physiol. Anat. Abt. 1891. Leboucq, H. : Recherches sur le mode de disparition de la chorde dorsale chez les vertebres superieurs. Arch, de Biol., p. 718-736. 1880. Luschka: Die Halbgelenke, 1852; 2. Aufl. 1858. Die Altersveranderung der Zwdschenmrbelknorpel. Virchow's Archiv. Bd. 9, S. 309-327. 1856. Uber gallertartige Auswiichse am Clivus Blumenbachii. Virchow's Archiv. Bd. 11, S. 8-12. 1857. Mall : A Human Embryo 26 Days old. Joum. of Morph. Vol. 5, p. 459-480. 1891. Human Coelom. Journ. of Morph. 1897. MiNOT, C. S. : The Segroental Flexures of the Notochord. Anatomical Record, Amer. Joum. of Anat. Vol. 6. 1907. MuLLER, H. : Uber das Vorkomnien von Resten der chorda dorsalis bei Mensehen nach der Geburt. Zeitschrift f. rationelle Med. Bd. 2. 1858. Musgrave, James : Persistence of the Notochord in the Human Subject. Joum. of Anat. and Physiol. Vol. 25. 1891. MORPHOGENESIS OF THE SKELETAL SYSTEM. 331 EoBiN, C. : Memoire sur revolution de la notocorde. Paris, 212 pp., 1868. RoMiTi, G. : Rigonflamenti della corda dorsale nella porzione cervieale nell'embrione umano. Notizie Anat. Siena. 1886. Spee, Graf v.: Beobachtungen an einer mensehliehen Keimscheibe mit offener Medullarrinne. Arch. f. Anat. nnd Phys. Anat. Abt. S. 159-176. 1889. Neue Beobaclitungen iiber sehr friihe Entwicklungsstufen des mensehliehen Eies. Arch. f. Anat. und Phys. Anat. Abt. S. 1-30. 1896. On the development of the notochord in the higher mammals see especially: Williams, L. W. : The Later Development of the Notochord in Mammals. Amer. Jour, of Anat. Vol. 8, p. 251. 1908. C. VERTEBRAL COLUMN AND THORAX. The development of the vertebral column and thorax may be divided into three overlapping periods: a membranous or blaste- mal, a chondrogenous, and an osteogenous. The Blastemal Period. The division of the axial mesenchyme into segments, sclero- tomes, which correspond to the myotomes and spinal ganglia, is marked at an early stage by intersegmental arteries (Fig. 233, A. is.). The segmental differentiation extends into the region dorsal to the spinal cord, but ventrally it does not reach the chorda dorsalis. Each sclerotome becomes divided into two portions, a' caudal half in which the tissue is condensed, and a cranial half in which the tissue is less dense (Fig. 234). In sections through hardened tissue a slight fissure, the intersegmental fissure (v. Ebner, 1888), may partially separate the two halves.^ From the condensed tissue of the caudal half there arises a primitive vertebra of Remak, or scleromere, with dorsal (neural) and ventral (costal) processes and chordal processes which unite these to the perichordal sheath, a dense layer of tissue forming a continuous sheath about the chorda dorsalis (Figs. 234, 237, 238, 240, 241, 242). From the tissue of the anterior halves of the sclerotomes arise "interdorsal membranes" which unite the dorsal processes of the scleromeres {M. id., Figs. 236, 244, 245, 247), and " interventral membranes" which unite the bases of the ventral processes (M. iv., Figs. 235, 243, 244, 245). The chordal processes become hollowed out caudalwards by a loosening up of their tissue and strengthened cranialwards by a condensation of tissue imme- diately bounding the intervertebral fissure (Figs. 234, 235, 243, " Schultze (1896) has described in a corresponding position in selachians and reptiles a diverticulum which communicates with the myoeoel. The fissure is aijparenl'y to be looked upon as an offshoot of the myoeoel. In birds the fissure is said to arise independently of and to fuse secondarily with the myoeoel. In mammals it appears after the myotome has become independent and the myoeoel has disappeared. 332 HUMAN EMBRYOLOGY. Fig. 231. Fig. 232. Fias. 231 and 232. — Uiagrammatic outlines to represent the development of the skeleton during the blastemal period. Fig. 231. Embryo II, length 7.5 mm. Fig. 232. Embryo CIX, length H mm. o, occipital; c\ first cervical: (i, first thoracic; (^ first lumbar; s^ first sacral; co^, first coccygeal vertebra. MORPHOGENESIS OF THE SKELETAL SYSTEM. 333 Fig. 233. Fig. 235. Fig. 234. Fig. 236. Figs. 233-236. — (After Bardeen, Amer. Joum. of Anat., vol. W, 1905.) Frontal sections through the thoracic region of several embryos during the blastemal period of vertebral development. Magn. 47.5 : 1. Fig. 233. Embryo CLXXXVI, length 3.5 mm. Fig. 234. Embryo LXXX, length 5 mm. Figs. 235 and 236. Embryo CCXTJ, length 6 mm. Fig. 235 through the region of the chorda dorsalis. Fig. 236 through a more dorsal plane. Figs. 233, 235, 236 represent sections cut somewhat obliquely so that the right side of the sections is ventral to the left. In Figs. 234 and 236 on the right side the bodies of several embryonic vertebrse are represented in outline. In Figs. 234 and 235, owing to artefacts, the myotomes are pulled away from the sclerotomes. A.is., arteria intersegmental] s; Coel., ccelom; Chd., chorda dorsalis; Der., dermis; F.v.E., fissure of v. Ebner (intervertebral fissure); M.id., membrana interdorsalis; Af.ip., membrana interventralis; ilf.sp., spinal cord; ilfyo., myotome; A''. sp. , nervus spina- lis; Pch.S., perichordal sheath; Pt.c, processus costalis; Pr.ch., processus chordalis; Pr.n., processus neurahs; ScL, sclerotome; V.c, vena cardinalls. 334 HUMAN EilBRYOLOGY. 244, 245). There is tlius formed about the intervertebral fissure a primitive intervertebral disc.'' The tissue lying between each two discs now becomes completely surrounded by a membrane of condensed tissue, which may be termed an viferdiscal membrane (Fig. 246, .1/. ir.). Meanwhile the perichordal sheath between each two discs becomes extended ventrodorsally, so that it gives rise to a "perichoi-dal" septum which divides into two parts the space surrounded by the interdiscal membrane (Figs. 239, 246, 247, Pch.s.). During the earlier stages of the blastemal period the sclero- meres are essentially similar throughout the length of the ver- tebral column. The differentiation of the scleromeres begins in the cervical region and extends caudalwards. At the end of the first month of development the scleromeres present the appearance shown in Figs. 231, 240, 241, and 242, although their margins are Jess sharply marked than it is necessary to represent them in the model. At this y)eriod of development the interdorsal and inter- ventral membranes have begun to appear in the cervical region, but are not represented in Fig. 231. Soon after this period the thoracic region of the spinal column becomes distinguishable from the neighboring regions through the great development of the costal process of the thoracic scleromeres (Figs. 232 and 239). Meanwhile centres of chondrification arise. These are described below.^" Tlie Occipital Region. — In man, as pointed out above, the primitive axial mesench3'me in the head posterior to the otic region undergoes a partial segmentation. At the end of the first month of development there are three fairly well-marked occipital myotomes which afterwards disappear. The axial mesenchyme corresponding to these mj^otomes is not definitely divided into sclerotomes, although that opposite the last occipital myotome becomes divided like each of the spinal sclerotomes into a light °I have elsewhere (1905) c-alled the united ehordal processes of the seleromere a primitive inter\-ertebral disc, but it seems better to restrict this term to the structure here described. According to Williams the primitive intei-vertebral discs are to be regarded as places in which the tissue remains dense while between them the differentiation of the bodies of the vertebrae is marked by a loosening up of the tissue. According to Williams the scleromeres are not true morphological units. " Charlotte Miiller (1906) has described a transitory, longitudinal ridge of cells which extends between the mid-ventral surface of the spinal column and the aorta. Opposite the primitive discs this ridge is connected to the anlages of the corresponding ribs by bands of tissue (hypochordal Spangen) which are not fused to the discs. Opposite the vertebral bodies the lighter tissue of the bodies is continued into the lighter tissue of the centre of the ridge of cells. The ridge extended from the second to the ninth thoracic vertebra in a 13 mm, embrj'o. There is no segmentation visible in the tissue of the ridge. MORPHOGENESIS OP THE SKELETAL SYSTEM. 335 Figs. 237-239. — (After Bardeen, Amer. Journ. of Anat., vol. iv, 1905.) Cross-sections through, midthoracic segments during the blastemal period of vertebral development. Magn. 55 : 1. Fig, 237. Embryo LXXVI, length 4.5 mm. The right side of the section passes through the middle, the left side through the posterior third of the fifth segment. Fig. 238. Embryo II, length 7 mm. Fifth thoracic segment. The right aide of the drawing represents a section anterior to that shown at the left. Fig. 239. Embryo CLXXV, length 13 mm. The left half of the sixth vertebral body, neural process, and rib are drawn in detail; the body-wall, spinal cord, and spinal ganglion are shown in outline. A.a., region of anastomosis of two successive intersegmental arteries; C.v., corpus vertebrse; Costa, rib; Ch.d., chorda dorsalis; Disc, intervertebral disc; G.sp., gangl. spinale; M.d., dorsal musculature; M.sp., medulla spinalis; Myo., myotome; N.sp., nervus spinalis; Pr.c, processus costalis; Pr.ch., processus chordalis; Pr.n., processus neuralis; Splm. perichordal septum. anterior and a condensed posterior half (scleromere). The lighter half is continuous apicalwards with the slightly condensed, unseg- mented mesenchyme which lies in the region of the more anterior occipital myotomes. This in turn is continued into a thin layer of 336 HUMAN EMBKYOLOGT. Ch.d.- 'Pch.s. A.is Fig. 242. Fig. 245. Ch.d. .M.Sl Ch.d. Pch.S. M.LV. Fig. 246. Fig. 247. Figs. 240-247. — (After Bardeen, Amer. Journ. of Anat., vol. iv, 1905.) Views of models repre- senting the blastemal stage of vertebral development. Figs. 240-242. Embryo II, length 7 mm. Magn. 33i : 1. Figs. 243-245. Embryo CLXIII, length 9 mm. Magn. 23 : 1. Figs. 246, 247. Embryo CIX, length 11 mm. Magn. 25 : 1. Figs. 240, 243, 246, views from in front; Figs. 241, 244, 247, views from the side; Figs. 242, 245, views from behind. A.is., arteria intersegmentalis; Ch.d., chorda dorsaiis; Disc, intervertebral disc; M.id., membrana interdorsalis; M.iv., membrana interventralis; Pr.n., processus neuralis; Pr.ch., processus chordali-s; Pr.c, processus ^^,. Sciuamosum' Squamosum iiF;.^ftyiS^^ Petrosum A^ .^BiUU^^ Tympanicum Tympanicum Petrosum A B Fig. 320. — (After Sappey, Trait<5 d'Anatomie, Figs. 32 and 33.) A, Temporal bone of a new-born Infant. B. Temporal bone of an infant of two years. The tympameum (Figs. 320, 321, 324) is a membrane bone. Its centre of ossification appears toward the end of the third month in the anterolateral part of the external membranous wall of the cavum tympani, near the angle between the capitulum of the mal- leus and Meckel's cartilage. It has a concave surface turned toward the latter. From this centre a band of bone grows first downwards, medialwards, and backwards and then upwards and lateralwards so as to form a semicircular bone surrounding the tympanic membrane. In the tenth fetal month first the free ends of the bone fuse with the squamosum and then the under part fuses with the petrosum. By addition of osseus tissue to the lateral '■' According to Eambaud and Renault, Toldt, Poirier, and others, the squa- mosum arises from three centres of ossification, one at the base of the zygomatic process, one in the squamosa, above this, and one behind the tympanicum. Among the variations found in the squamosal portion of the temporal bone are a division into a superior and an inferior part or into an anterior and a. posterior part. ""Fuchs {1905-1907), chiefly from the study of rabbit embryos, has come to the conclusion that the squamosum of mammals is composed of three parts, a squamous and a zygomatic (quadrato-jugale) part, each ossified in membrane, and an articular part preformed in cartilage (pars articularis quadrati). MORPHOGENESIS OF THE SKELETAL SYSTEM. 433 and medial margins of the bone, the primary narrow band becomes converted by the third year into a broad rolled-up plate of bone, which medially forms the ventral wall of the cavum tympani and laterally the ventral wall of the meatus acusticus externus. At this period the inferior surface of the tympanicum presents an aperture of some size which usually but not always becomes closed. According to Eambaud and Renault, Hammar, and others, the bone arises from several centres of ossification. Occipi- talesu- perius N. facialis Mandi- bula Squama temporalis Processus atyloideus Os tympanicum Cartilago Meckeli Fig 321 —{After Hertwig's model, from Kollmann's Handatlas, 1907, Fig. 263.) Lateral view of the cranium of a human fetus 80 mm. long. The chondrocranium and the overlymg membrane bones are shown. The parietal shows two centres of ossification. Tympanohyale (laterohyale) and stylohyale. See p. 439. The periotic portion, os petrosum (Fig. 322), arises from the ossification of the cartilaginous otic capsule. There are several centres of ossification. These centres arise during the fifth month and become fused with one another in the sixth. The descriptions of the centres of ossification in the labyrinth given by various authors differ considerably. That of Vrolik, as adopted by Gaupp (1906), is here chiefly followed. The first centre to appear is one in the region of the promontory between the fenestra vestibuli and the fenestra cochlear in a fetus 17 cm. long. Ossification extends Vol. I.— 28 434 HUMAN E.MBKTOLOGT. from here around the fenestra vestibuli and forms that part of the petrous bone which lies below the porus acusticus internus and the fenestra cochlearis. A second centre appears on the dorso- lateral surface of the central part of the capsule over the superior semicircular canal. Ossification extends from this into the region of the processus perioticus superior. It forms most of the cranial surface of the petrous bone, gives rise to the superior Vomer Occipitalei laterale Septum cartilagineum Capsula na- salis pars lateralis Alisphenoid Punctum ossificationis ^mediale of the basisphe- noidale Punctum .ossificationis of the petrosum Partes condyloideae Canalis condyloideus Occipitale superius Fig. 322.— CAtter Kollmann, Handatlas, 1907, Fig. 271.) Visceral surface of the base of the skull of a human fetus 18 cm. long. The mandible and maxilla have been removed. boundaries of the internal auditory meatus and the fenestra cochlearis, and also to a portion of the superior medial part of the mastoid. A third centre arises at the anterior proximal part of tlie cochlea near the incisura pro-otiea. ]\Iore caudally, medial to the fossa subarcuata, there arises a fourth centre of ossification. The fifth centre appears on the outer surface of the posterior part of the capsule in the region of the posterior semicircular canal ; the sixth centre arises slightly in front of the fifth. From the last MORPHOGENESIS OF THE SKELETAL SYSTEM. 435 two centres ossification extends into the parietal plate and the pars mastoidea of the tectum posterius. At the end of the sixth month the labyrinth is completely enclosed by bone. The tegmen tympani is ossified partly in membrane, partly in the cartilage of the processus perioticus superior by extension from the periotic capsule. There is occasionally found in man a separate bone in the anterior part of the tegmen tympani. This perhaps represents the ossiculum accessorium malleoli, which in some mammals has an independent centre of origin on the upper side of the proximal end of Meckel's cartilage (van Kampen, 1905). The embryonic epithelial labyrinth at first lies in a cavity surrounded by cartilage. The inner surface of this cartilage becomes transformed into membranous tissue, and this in turn in part furnishes a membranous covering for the labyrinth, and in part becomes ossified, forming the inner lining of the bony labyrinth, including the modiolus, the lamina modioli, and the lamina spiralis ossea. (See Kolliker, 1879; Bottcher, 1869.) The canalis caroticus is represented by a slight groove in the cartilaginous skull. It becomes converted into a canal during the period of ossification. At birth the central part only is roofed over. Between the apex pyramidis and the sphenoid a part of the chondrocranium persists as the fibrocartilago basalis, which lies in the foramen lacerum. Formation of mastoid cells does not begin until after birth, but in the second year they extend from the antrum into the mastoid process. The Canalis Facialis. — The short facial canal in the chondro- cranium is equivalent merely to the first part of the facial canal in the adult (part from the porus acusticus int. to the region of the geniculate ganglion). Before the chondrocranium is replaced by bone the nerve passes out from the canal above mentioned, then beneath the crista parotica and over the stapes, and thence back- wards and downwards beneath the tympanohyale, and then out- wards and ventralwards toward the surface of the body (Fig. 311). The nervus petrosus sup. major leaves the main trunk near the lateral orifice of the cartilaginous canal ^^ and runs forward on the lateral wall of the capsule. The chorda tympani separates from the main trunk behind the stylohyale, runs forwards lateral to this cartilage and then between the malleus and incus. It is not enclosed in a canal in the chondrocranium. When the lateral wall of the auditory capsule becomes ossified, the facial nerve is en- '"In the human fetus and in Talpa through a special opening in the external orifice of the facial canal (E. Fischer). 436 HUMAN EilBRYOLOGY. closed at first in a groove and later in a canal. While the facial nerve is being enclosed bony lamellae likewise enclose the stape- dius muscle and the chorda tjnnpani. The facial nerve and Eeichert's cartilage may be looked upon as caught between the tjTnpanicum and the mastoid part of the petrosum, the chorda tympani and Meckel's cartilage as caught between the tympanicum and the squamosum, the tuba auditiva (Eustachii) as enclosed between the tympanicum and the pars cochlearis of the petrosum (van Kampen). OS PARIETALE. The parietal arises as a membrane bone from two centres (Toldt), a superior and an inferior, which soon fuse into a single centre which lies in the region of the tuber parietale. The centres appear toward the end of the second month and apparently some- times arise as a single centre. The ossification radiates outwards from the combined centre of ossification toward the sphenoidal and occipital angles, so as to give rise for a time to an hourglass- shaped plate of bone (Mall). At a later period a notch is left in the sagittal margin of the bone in front of the occipital angle. The notches of the bones of the opposite sides together form the sagittal fontanelle, which toward the end of fetal life usually becomes closed, but may sometimes be recognized after birth. OS FEONTALE. This has two centres of ossification, one on each side of the body in the region of the tuber frontale (Fig. 321). These centres arise toward the end of the second month (56th day. Mall). From each a lateral half of the bone is formed. The orbital part of the bone appears in the ninth week. Toldt found no secondary centres, but these have been described by Eambaud and Renault, Serres, Jhering, v. Spee, and others. These accessory centres include, on each side of the bone, one for the spina trochlearis, one for the processus zygomaticus, one for the posterior part of the orbital region, and one which arises late in the spina frontalis lateral to the foramen CEecum. The two lateral halves of the bone are sepa- rated at birth, but during the first year become approximated in the midsagittal line. The middle part of the frontal suture becomes ossified in the second year. By the eighth year the suture is usually obliterated except inferiorly. Near the root of the nose the frontal suture is sometimes widened to form a fontanella metopica, in which an os metopicum may be formed. Ossicles may appear also in other parts of the frontal suture (Schwalbe, 1901; Fischer, 1901). The frontal sinuses begin to be developed early (first year, Toldt), but develop very slowly until toward puberty. They increase in size until late in life. MORPHOGENESIS OP THE SKELETAL SYSTEM. 437 FONTANELLES. The chief fontanelles are spaces covered by membranes which lie between the incomplete angles of the parietal and the neigh- boring bones." The anterior fontanelle (fonticulus frontalis) is situated between the frontal angles of the parietal and the postero- superior angles of the two parts of the frontal bone. It remains open until the third year. The posterior fontanelle (fonticulus occipitalis) is situated between the occipital angles of the parietal bones and the superior angle of the occipital. This fontanelle at birth is nearly obliterated, but the bones which bound it are still separated by membrane and are movable. It becomes closed between the third and sixth month. The lateral fontanelles (fon- ticulus mastoideus, fonticulus sphenoidalis) are situated between the sphenoidal and mastoidal angles of the parietal and the neigh- boring bones. The fonticulus mastoideus closes during the first half of the second year, the fonticulus sphenoidalis in the third year. The flat bones of the skull are united by definitive sutures by the end of the fourth year.^® Special small bones (Wormian bones) may develop in the fontanelles during the process of ossification. MAXILLA. The human maxilla consists of two distinct parts, a medial, the incisivum (premaxilla) of lower forms, and a lateral, or maxilla proper. According to Mall (1906), each of these parts of the bone ossifies from a single centre. The two centres appear at the end of the sixth week and become united at the end of the Sulcus lacrimalis maxillaris jr M Sutura incisiva <^^^^^^lm Sut.inc. Os incisivum ^^ Processus k alveolarig Processus Os Processus /ftli fe palatinus incisivum palatinus \r Fig. 323.— (After Toldt, Anat. Atlas, 1900, Heft 1, Figs. 176, 177.) The left maxilla of a fetus 30 cm. long. A. Medial surface, B. Inferior surface. second or early in the third month. Each of the centres gives rise to a part of the frontal process. The alveolar borders of the two bones unite before the frontal processes do. Authors differ greatly in the number of centres which they ascribe to the bone. The number given varies from two to six. Hertwig's model and " For an account of the transitory sagittal fontanelle between the parietal bones see above under Os parietale. Tor the metopic fontanelle see above under Os frontale. "' For the time and order of closure of the chief fontanelles see Adachi (1900). 438 HUMAN EMBRYOLOGY. Schultze's illustration of the bones of the skull of an embryo of the third month are cited by Mall in support of his o^n observations.^® At first the maxilla lies lateral to the cartilaginous nasal cap- sule. After this cartilaginous capsule is in large part absorbed the maxilla helps to bound the nasal cavity, the maxilloturbinale becomes joined to it, some of the ethmoid cells are closed off by it, and the sinus maxillaris is formed. The formation of the alveolar process begins in the fourth fetal month and is completed after the twentieth year. According to Milialkovics (1899), the proc. paranasalis of the nasal capsule is caught up in the ossifying maxilla. This is said to account for the islands of cartilage some- times described in the ossifying maxilla.*" The infraorbital nerve and vessels lie at first in a groove on the orbital surface of the maxilla, but later become enclosed by a lamina of bone which extends upwards on the lateral side, and then bends medialwards (see Fig. 321). The lateral part of the floor of the orbit and the infraorbital nerve lie at first very near the alveolar process of the maxilla. The development of the sinus maxillaris gradually serves to separate them. Compare the maxilla shown in Fig. 321 with that of an adult skull. The sinus maxillaris at birth is reiiresented by a slight depression on the medial surface of the maxilla opposite the second molar. After birth this depression extends laterally into the maxilla beneath the groove of the infraorbital nerve and blood- vessels. After the second dentition the sinus becomes greatly enlarged. OS ZYGOMATICUM. This appears on the 56th day as a small, three-cornered centre in the membranous tissue beneath and lateral to the eye (Fig. 321). On the 58th day it is four-cornered. Two of the comers give rise to processes partially encircling the orbit, one of the others extends to the maxilla, and one toward the temporal bone (Mall, 1906). From this period on, the adult form of the bone is steadily approached.*^ "" According to Le Double (1906), the number of centres described by various authors for the maxilla, including the incisivum, is as follows: One, Camper, Rous- seau, Cleland; two, Jamain; three, Serres, Meckel, Cruveilhier; five, Beclard, Sappey, Leidy, Poirier, W. Krause; five or six. Portal; six, Rambaud and Renault; seven, "\\'eber. '" In fetuses of the fourth to fifth month there may arise in the alveolar part small cartilaginous islands which have no connection with the nasal capsule and which disappear during the ossification of the upper jaw (Gaupp, 1906). " In the adult partially or completely bipartite and tripartite malar bones are not very infrequent. In the Japanese and Ainos there is found a considerable percentage of skulls in which the malar bone is partially or completely divided by a horizontal suture (os Japanieum, os Ainoicum). Various writers differ greatly MORPHOGENESIS OF THE SKELETAL SYSTEM. 439 AUDITORY OSSICLES. These begin to ossify during the last half of the fifth month. The malleus has a centre of ossification from which all parts of the bone arise except the processus anterior. This centre arises in the upper part of the coUum and from here ossification spreads to the other parts. The manubrium, the last part of the bone to become ossified, reaches its definitive form before birth. The processus anterior (Folii) arises at the end of the second month as a slender membrane bone on the medial side of Meckel's car- tilage. It reaches its definite length in the middle of the sixth month. The proximal end fuses with the collum mallei toward the end of the fifth month (Broman), at the time of ossification of the latter. When the malleus is ossified it becomes clearly demar- cated from the rest of Meckel's cartilage. The latter slowly atrophies, and is replaced by connective tissue, which first appears at its periphery. The incus is ossified from a single centre which appears in the upper part of the crus longum. Ossification extends from here into the other parts of the bone, including the processus len- ticularis. The ossification begins in fetuses 19-20 cm. long and by the time of birth has reached its definitive extension (Broman). In the stapes, a centre of ossification usually appears in the basal portion in fetuses 21 cm. long (Broman). From this centre the bone is ossified. The capitulum is usually ossified by the end of the sixth month. TYMPANO HYALE . The tympanohyale is probably derived from the cartilaginous tympanohyale described in connection with the chondrocranium, although tympanohyale and stylohyale are fused before ossifica- tion appears. Late in fetal life it is ossified from a special centre and becomes included between and fused to the petrosum and the tympanicum. It helps to bound the tympanic cavity medial to the OS tympanicum. in the number of centres of ossification which they ascribe to the bone. Le Double (1906) classified these writers as follows: Those describing one centre of ossification : Meckel, Beelard, Hyrtl, Sappey, Cruveilhier, Jamain, Leidy, Baraldi, Lachi, Romiti, Langer-Toldt, Stieda, Merkel, Graf V. Spee, Hartmann. Those describing one or two centres: Parker and Bettany. Those describing two centres : Kerckring, Lieutaud, Garbiglietti, Macalister. Those describing two or three centres : Morris and Rauber. Those describing one, occasionally two, rarely three : Breschet, Gruber. Those describing one, occasionally three: Virchow, Albrecht, Testut, Thane. Those describing three centres : 0. Sehultze, Kollmann, Frassetto, Minot, Spix, Calori, Quain, Rambaud and Renault, Schrenck, Poirier, Kolliker, C. Toldt, Le Double. 440 HUMAN EMBRYOLOGY. STYLO HYALE. The cartilaginous stylohyale gives rise to the styloid process. This ossifies after birth and is usually united to the tympanohyale by cartilage until middle age when it may become united to the latter by bone (Flower, 1870). KERATOHYALE. This is derived from the proximal part of Reichert's cartilage. It gives rise to the stylohyoid ligament. Occasionally it may become ossified and fused to the stylohyale and the lesser cornu of the hyoid. OS HYOIDEUM. This has five primary centres of ossification, — one for the body and one for each of the greater and each of the lesser cornua. Ossification begins in the body and greater cornua late in fetal life ; in the lesser cornua some time after birth. The greater cornua and body unite in- middle life. The lesser cornua are united by bone to the body of the hyoid only rarely. Usually a cartilaginous union remains throughout life. According to Eambaud and Renault, the centre in the body arises by fusion of two bilaterally placed centres. After puberty secondary centres for the tips of the greater cornua are described (Poirier, Traite d'Anatomie). MANDIBLE. This is ossified in the membranous tissue lateral to Meckel's cartilage. The centre of ossification may appear as early as the 39th day. By the 42d day the ramus and alveolar process may be distinguished. In a fetus 55 days old the beginnings of the coronoid process and condyle are visible. Before the end of the second month sockets for the teeth may be distinguished. By the middle of the third month the mandible has reached its char- acteristic shape (]\lall, 1906 ).*2 During the development of the mandible cartilage is produced in the membranous tissue of the tip of the condyle, in the angulus, the proc. coronoideus (see Fawcett, 1904; Low, 1905), and, according to Henneberg, in traces also on the superior lateral and the medial alveolar margins and the inferior lateral margin of the jaw. This cartilage^ has nothing to do with Meckel's cartilage. From Meckel's cartilage, however, there may be derived the cartilage in the symphysis of the jaw in which small bones, ossicula mentalia, develop in the eighth month. There are usually two pairs, or one unpaired bone and one pair of bones. They lie in the lower portion of the symphysis " Several investigators, among whom may be mentioned Eambaud and Renault and Wolff (1888), describe several centres of ossification for the mandible. MORPHOGENESIS OP THE SKELETAL SYSTEM. 441 and after birth become fused to the mandible and help form the protuberance of the chin; see v. Mies (1893), Toldt (1905-6), V. Bardeleben (1905). According to Low (1905), Meckel's car- tilage near the incisors becomes ossified and fused to the mandible. According to Fawcett (1904), Meckel's cartilage becomes ossified from the foramen mentale to the median line. The ossified carti- lage becomes enclosed in membrane bone. At birth the lower jaw usually consists of lateral parts united at the symphysis by fibrous tissue. Osseous union takes place in the first or second year after birth. Malleus Manubrium mallei Fig. 324. — (After Kollmann, Handatlaa, 1907, Fig. 273.) Mandible, Meckel's cartilage, malleus, and incus of a human fetus 8 cm. long. TEMPOROMANDIBULAR JOINT. This joint is developed between the membrane which covers the condyle of the mandible and the periosteum of the squamosum. In the loose tissue between the two a condensation marks the beginning of the differentiation of the discus articularis. On each side of this discus a joint cavity develops. Each joint cavity is throughout life lined by fibrous tissue. Beneath the joint periosteum of the mandible and of the temporal bone a thin layer of cartilage is produced (see Kjellberg, 1904). According to Wallisch (1906), in the new-born the tuberculum articulare is still undeveloped and the condyle is flatter than in the adult. The condyle reaches its definitive form and the tuberculum is devel- oped after the teeth appear.*^ "According to Fuchs (1905), the temporomandibular joint in rabbits, and hence by inference in other mammals, is homologous with the quadrato-articular joint of reptiles. As mentioned above, following Reichert, most investigators have come to the conclusion that the reptilian quadrato-articular joint is represented in mammals by the joint between the malleus and incus, wliile the temporomandibular joint of mammals is phylogenetically a new structure, a squamosodental joint. (See Gaupp, 1906.) 442 HUMAN EMBRYOLOGY. BIBLIOGRAPHY. (On the development of the human skull. Some references are also given to the literature on the comparative embryology of the skull and on variation in cranial structure.) Adachi, Buntaro: Uber die Seitenfontanellen. Zeitschrift fiir Morphol. und Anthropologie. Bd. 2. 1900. Uber die Knochelchen in der Symphyse des Unterkiefers. Zeitschrift fiir Morphol. und Anthropologie. Bd. 7, H. 2. 1904. Appel, Kxjet: Zur Lehre vom anatomischen Sitz der Hasensehartenkieferspalte. Miinchner med. Wochenschidft. Bd. 9, S. 41. 1894. Albeecht, p.: Sur les quatre os intermaxillaires, le bec-de-lievre et la valeur morphologique des dents incisives superieures de I'homme. Comm. Soc. d'Anthrop. de Bruxelles 1883. Uber den morphologischen Sitz der Hasensehartenkieferspalte. Biol. Central- blatt. Bd. 6. 1886. Anthony, R. : Contribution a I'etude de la regeneration osseuse du crane. Bull. Mem. Soc. I'Anthrop. Ser. 5, T. 7, p. 197-201. Paris 1906. AcQUA, DALL, U. : Sopra lo sviluppo delle suture. Monit. Zool. Ital. Vol. 9, p. 150. 1898. Aeai, Haeujiro : Der Inhalt des Canalis cranio-pharyngeus. 14 Fig. Anat. Hef te. Bd. 33, p. 411-151. 1907. Bardeleben, Kael v. : Uber das praefrontale und postfrontale des Mensehen. Verhd. der Anat. Gesellschaft auf der 10. Versammlung in Berlin. 1896. Anat. Anz. Erg. H. Bd. 12. Der Unterkiefer der Siiugetiere, besonders des Mensehen. Anat. Anz. Bd. 26, S. 104-111. 1905. Uber den Unterkiefer der Saugetiere. Sitzungsberieht der Gesellschaft naturf. Freunde Berlin. Jahrg. 1905. Baumuller, B. : Uber die letzten Veranderungen des Meckel'schen Knorpels. Zeitschrift fiir wissenseh. Zoologie. Bd. 32, H. 3. 1879. Baumgarten : Beitrage zur Entwicklungsgeschichte der Gehorknochelchen. Archiv fiir mikrosk. Anat. Bd. 40, S. 512. 1890. Berkhahn, 0. : Zur Entwieklung und Deutung der sogen. Azteken. — Mikro- cephalen. Globus. Bd. 73. 1898. Bessei.-Hagen, F. : Vorlauflge Mitteilung iiber die Entwieklungsgeschiehte des menschlichen Occiput und die abnormale Bildung des Os occipitis. Monats- bericht der Konigl. Preuss. Akademie der Wissenschaften zu Berlin aus dem Jahre 1879. S. 264. Berlin 1880. BiANCHi, S. : Sullo svilluppo delta squama oeeipitale, sul modo di originarsi e sul modo di originarsi delle varii forme delle ossa interparietali e preinterparie- tali nel cranio, umano. Monitore Zool. Ital. A''ol. 2, p. 69-91. 1891. Sopra alcune varieta del cranio osservate in feti umani ed in altri mammiferi. Monitore Zool. Ital. Vol. 4, p. 11-17. 1893. Sulla divisione dell' osso parietale e sul suo sviluppo. Atti. Accad. Fisiocritici Siena. Ser. 4. Vol. 13. Anno Acad. 210. 1901. Sullo svilluppo dell 'osso parietale umano. Arch. Ital. Anat. e Embriol. Vol. 2, p. 94-96. 1903. BiONDi, D. : Uber Zwischenkiefer- und Lippenkiefergaumenspalte. Arch, f . Anat. u. Physiol. Physiol. Abt. 1886. Zur Hassenschartenfrage. Wiener med. Blatter. S. 7-20. 1886. Lippenspalte und deren Komplikationen. Arch, fiir patholog. Anat. u. Physiol. und fiir klin. Medizin. Bd. 111. 1888. MORPHOGENESIS OF THE SKELETAL SYSTEM. 443 BiEKNEE, Feed. : Das Schadelwaehstum der beiden ainerikan. Mikrocephalen (sog. Azteken) Maximo und Bartola. Korresp.-Bl. Deutsch. Ges. f. Anthrop! Ethnolog-. Urgesehiehte. Jg. 28, N. 11 u. 12. 1898. Uber die sogen. Azteken. Arch. Anthropologie. Bd. 25. 1898. Uber eine sehr seltene Verknocherungsanomalie des Hirnschadels Petrus Camper. 2. Teil, 2. Afl. 1903. BoLK, L. : Uber eine sehr seltene Verknocherungsanomalie des Hirnsehiidels. Petrus Camper. Bd. 2, Afl. 3. 1903. EntwicklungsvorgiJnge in der occipitalen Region des Primordialcraniums beim Menschen. Petrus Camper. Bd. 2, Afl. 3, S. 325-327. 1903. Zur Frage der Assimilation des Atlas am Schiidel beim Menschen. Anat Anz Bd. 28, S. 497-506. 1906. BovERO, Alfonso: Ossieina medio-frontali nei crani di neonati. Giorn. Acad. Med. Torino. Anno 67, Nr. 5/6. 1904. In riposto ad una lettera aperta del dott. T Delia Vedova proposito di studi sullo sviluppo delle cavita nasali. Arch. Ital. Otol. Vol. 15. 1904. Beeschet, G. : Recherehes sur differentes pieces osseuses du squelette de I'homme et des animaux vertebres. Deuxieme memoire. De I'os malaire ou jugal. Annales des Sciences Naturelles. Ser. 3, Zoologie. Vol. 1. 1844. BnocA ET Lenoir: Note sur un cas de persistance du cartilage de Meckel ave<* absence de I'oreille externe du meme cote. Considerations sur le developpe- ment du maxillaire inferieur et des osselets de I'ouie. Journ. de I'Anat. et de la Physiol. Annee 32, p. 559. Paris 1896. Brock, J. : Uber die Entwicklung des Unterkiefers der Saugetiere. Zeitsehrift f iir wissenschaftliehe Zoologie. Bd. 27, S. 287. 1876. Broman, Ivar : Uber die Entwicklung der Gehorkuochelchen beim Menschen. Ver- handl. der Anat. Gesellschaft auf der 12. Versammlung in Kiel, 1898. Dis- sert. Lund, 1899. Anat. Hefte. Bd. 11, H. 4. Bithe: (1) Zwei seltene anatomische Befunde am Sehlafenbein. (2) Defekt des Bulbus vense jugularis und des sinus sigmoideus. Arch, fiir Ohrenheilk. Bd. 57. 1903. Callender, G. W. : The Formation and Early Growth of the Bones of the Human Face. PhUosoph. Trans, of the Royal Society of London. Vol. 159. 1870. On the Formation of some of the Subaxial Arches in Man. Philosoph. Trans, of the Royal Soc. of London. Vol. 161. 1872. Caselli, Aenoldo : Sulla permanenza del canale cranio-faringeo nell uomo. Gazz. di Ospedali. Anno 21; Arch. Ital. p. le Malatti Nerv. e Ment. Anno 37. Riv. sper. Freniatria. Vol. 27. 1900. Chaine, J. : Contribution a I'etude du cartilage de Meckel. C. R. Soc. Biol. T. 55, No. 5. Paris 1903. Chiarugi: II terzo condilo e i processi basilar! del cranio umano. (Rudknenti di un arco ipoeordale oecipitale). Monitore Zool. Ital. Anno 6. 1895. Cleland, J. : On the Relations of the Vomer, Ethmoid, and Intermaxillary Bones. Phil. Trans, of the Royal Society of London. Vol. 152. 1863. Debieere, C. : Developpement du segment occipital du crane. Journal de I'Anat. et de la Physiol. Annee 31. 1895. DiEULAFE, Leon: Les fosses nasales des vertebres (morphologie et embryologie). Journal de I'Anat. et de la Physiol. Annee 40, Nr. 3, p. 268-298; Nr. 4, p. 414-^44. 1904. DiEULAFE ET Herpin : Developpemeut de I'os maxillaire inferieur. Journ. de I'Anat. et de la Physiol. Annee 42, p. 239-242. 1906. Ii'apophyse angulaire du maxillaire inferieur (processus Sandifortii). Journ. de I'Anat. et de la Physiol. Annee 43, p. 332-340. 1907. Dixon, A. Francis: Some Specimens showing Indications of the Presence of an Occipital Vertebra. Trans. R. Acad, of Med. in Ireland. Vol. 24, p. 465. 1906. 444 HUMAN EilBRTOLOGT. Double, A. F. lb: Le canal cranio-pharyngien, hypophysaire ou pituitaire de rhomme. Bull, et Mem. Soe. dAnthrop. Ser. 5, T. 4. Paris 1903. Traite des variations des os du crane de I'homme et de leur signification an point de vue de I'anthropologie zoologique. Paris 1903. Traite des os de la face de rhomme et de leur signification au point de vue de I'anthropologie zoologique. Paris 1906. Dreytuss, R. : Beitrage zur Entwicklungsgeschichte des Mittelohres und des Trom- melfells des Menschen und der Saugetiere. Morphol. Arb., herausgegeben von G. Schwalbe. Bd. 2, S. 607. 1893. Druner, L. : Uber die Anatomie und Entwicklungsgeschichte des Mittelohres beim Menschen und bei der Maus. Anat. Anz. Bd. 24, S. 257-289. 1904. Uber die Wirbeltheorie des Schadels. Ber. Senckenberg naturf. Gesellschaft, S. 152-156. Frankfurt 1905. Durst, E. : Zur Entwicklungsgeschichte des Kopfes des Menschen und der hoheren Wirbeltiere. Tiibingen 1869. Engstler, G. : Uber den Liickenschadel Neugeborener und seine Beziehung zur Spina bifida. Arch, fiir Kinderh. Bd. 40. 1905. Fawcett, Edward : The Ossification of the Lower Jaw of Man. Journ. of Anat. and Physiol. Vol. 39, p. xlvii-xlviii. 1905. The Lancet. Vol. 167, p. 541. The Early Stages of the Ossification of the Pterygoid Plates of the Human Sphenoid. Proc. of the Anat. Soe. of Great Britain and Ireland, Feb., 1905. Journ. Anat. and Physiol. Vol. 39, N. S. Vol. 19, p. 6 and 7. 1905. On the Early Stages in the Ossification of the Pterygoid Plates of the Sphenoid Bone of Man. Anat. Anz. Bd. 26, S. 280-286. 1905. On the Development, Ossification, and Growth of the Palate Bone of Man. Journ. of Anat. and Physiol. Vol. 40, p. 400-466. 1906. FiCALBl, EuGENio: Sulla ossificazione delle capsule periotiche nell' uomo e negli altri mammiferi. Estratti degli Atti. della Reale Academia medica di Roma. Anno 13, Ser. 2. Vol. 3, 1886-1887. Considerazioni riassuntive sulle ossa accessorie del cranio dei mammiferi e dell' uomo. Monitore Zool. Ital. Anno 1, Nr. 7/8. 1890. FiLATOPF, D. : Zur Frage iiber die Anlage des Knorpelsehiidels bei einigen AVirbel- tieren. Anat. Anz. Vol. 29, S. 623. 1906. Fischer, Eugbn : Zur Kenntnis der Fontanella metopica und ihrer Bildungen. 2 Tafeln. Zeitschrift f. Morphol. und Anthropologie. Bd. 4. 1901. Zur Kenntnis' des Primordialcraniums der Affen. Anat. Anzeiger. Bd. 20, S. 410 bis 417. 1902. Zur Vergleichung des Menschen- und Affenschadels in friihen Entwicklungs- stadien. Korrespondenzblatt der Deutschen anthropol. Gesellschaft. 1902. Zur Entwicklungsgeschichte des Affenschadels. Zeitschrift fiir Morphol. und Anthropologie. Bd. 5. 1903. Demonstration von Modellen zur Vergleichung der Schiidelentwicklung von Mensch und Affe mit besonderer Berlicksichtigung der Nase. Verhandl. des Vereins siiddeutscher Laryngologen in der Dekade 1894-1903. Wiirzbur;;' 1904. Zur Frage der Kinnbildung und AValhoff's " Theorie." Deutsche ]\Ionatssehrif t fiir Zahnheilkunde. Jg. 23. 1905. Fleisohmann, a. : Das Kopfskelett der Amnioten. Morphogenet. Studien. Mor- pholog. Jahrbuch. Bd. 31. 1903. FoRSTER, Andreas : Beitrage zur Kenntnis der Entwicklungsgeschichte des Inter- parietale. 1 Tafel u. 45 Fig. Zeitschrift fiir Morphol. und Anthropologie. Bd. 4. 1902. Fraser : On the Development of the Ossicula Auditus in the Higher Mammalia. Phil. Trans. Vol. 173, p. 901. 1883. MORPHOGENESIS OF THE SKELETAL SYSTEM. 445 Frasetto, Fabio: Sulla probabile presenza di quatro nuclei di ossifieazione nel parietale dell' uomo e delle scimmie. 4 Fig. Verb, der Anat. Gesellsehaft, 14. Versammlung. Pavia 1900. Sulla legge che governa la genesi delle suture nel crania. Verb, der anat. Gesell- sehaft. 1900. Sur les fontanelles du crane chez rhomme, les primates et les mammiferes en general. 3 Fig. C. R. du CongTes Intemat. dAnthrop. et d'Archeol. Pre- bist. Ser. 12. Paris 1900. Contribute alia teoria dei quatro centri di ossificatione nell' osso parietale dell' uomo e dei primati. BoUetino dei Musei di Zoologia et Anatomia comparata delle R. TJniversita di Torino. Vol. 17. 1902. Le forme cranicbe degli Antropoidi (Simidse) in rapporto alle umane. Atti Soc. Roman, di Antbrop. Vol. 10, Fasc. 1-3. 1904. Parietali tripartiti in crani umani e di seunmie. Monit. Zool. Ital., Anno 15, N. 12. 1904. Suture, fontanelle ed ossicini fontanellari del palato duro. Anat. Anz. Bd. 27. Erganz.-Heft. Verb. d. Anat. Gesellsehaft. 1905. Frey, H. : Fin Beitrag zur Anatomie des Scbliifenbeins. Arch, fiir Ohrenbeil- kunde. Bd. 48, Hft. 1 u. 2, S. 44-62. 1905. Friedenthal, Adolph : Beitrag zur Kenntnis der embryonalen Scbadelentwicklung. Diss. Konigsberg 1900. Feoriep, a. : Kopf teil der Chorda dorsalis bei menscblieben Embryonen. Beitrage zur Anat. und Embryol. als Festgabe f. Jacob Henle, dargebracbt von seinen Scbulern. 1882. Entwicklungsgescbichte des Kopfes. Ergebnisse der Anat. u. Entwieklungsg. Bd. 3. 1894. Einige Bemerkungen zur Kopffrage. Ant. Anz. Erganz.-Heft zu Bd. 21. 1902. Verhandlg. der Anat. Gesellsehaft. 1902. FucHS, H. : Bemerkungen iiber die Herkunft und Entwicklung der Geborknochel- cben bei Kaninchenembryonen. Arch, fiir Anat. und Physiol. Anat. Abt. Suppl.-Band, S. 1-178. 1905. Morpbologische Bedeutung des squamosum am Saugetierschadel. Zeitschrift fiir Morphol. und Anthropologie. Bd. 10. 1907. FuSAEi, R. : A proposito di un cranio presentate I'osso parietale tripartito. Arch. Ital. Anat. e. Embriol. Vol. 2, Fasc. 3. 1903. Sulla divisione e suUa fessura marginali dell' osso parietale nella specie umana. Arch. Sc. Med. Torino. Vol. 28, Fasc. 1. 1904^ Ganfini^ C. ; Sopra alcune f aecette articolari del basioccipitale in rapporto ai proeessi basilari. Monit. Zool. Ital. Anno 17, p. 60-68. 1906. Gaupp, E. : Ontogenie vrnd Phylogenie der schalleitenden Apparate bei den Wirbel- tieren. Ergebnisse der Anat. u. Entwieklungsgeseh. Bd. 8. 1899. i Alte Probleme und neuere Arbeiten iiber den Wirbeltierschadel. 5 Fig. Er- gebnisse der Anat. u. Entwieklungsgeseh. Bd. 10. 1901. Uber die Ala temporalis des Saugeschadels und die Regio orbitalis einiger anderer "Wirbeltierschadel. Anat. Hefte. Bd. 19. 1901. Uber die Ala temporalis des Saugetierschadels. Verhandlung der Anat. Gesell- sehaft. 1901. Zum Verstandnis des Sanger- und Menschenschadels. Korresp.-Bl. d. deutsch. Anthropol. Gesellsehaft. Nr. 12. 1903. Die Nichthomologie des Unterkiefers in der WirbeltieiTeihe. Verhandlg. d. Anat. Gesellsehaft. 1905. Anat. Anz., Erganzungsbeft. Bd. 27. 1905. Neue Deutungen auf dem Gebiete der Lehre vom Saugetierschadel. Anat. Anz. Bd. 27. 1905. Das Hyobranehialskelett der WLrbeltiere. 46 Fig. Ergebnisse der Anat. und Entwieklungsg. Bd. 14, S. 808-1048. 1905. 446 HUMAN EMBRYOLOGY. Die Entwieklungsgeschiohte des Kopfskelettes. Hertwig-'s Handbuch. 1906. Uber allg. und spez. Fragen aus der Lehre vom Kopfskelett der Wirbeltiere. Anat. Anz., Erganzung-sheft. S. 21. 1906. Gegenbaur, C. : Uber die Pars facialis des Laerymale des Menseheu. Morphol. Jabrb. Bd. 7. 1SS2. Geiiiiell, J. F. : On tbe Origin of the Stapes and on its Continuity with the Hyoid Arch. Eep. 71st meeting Brit. Ass. Advanc. Sc. Glasgow. 1901. GiUFFEiDA-RuGGEEi : Sul significato delle ossa fontanellari e dei foramini parietal! e sulla pretesa penuria ossea del cranio umano. Atti Soc. Rom. di Antrop. Vol. 7, p. 81-92. 1901. Gkadenigo, G. : Die embryonale Anlage des Mittelolires. Die morphologische Be- deutung der Gehorknochelchen. Mitteilungen aus dem embryol. Institut der k. u. k. Universitat T\"ien. 1887. Geuber, J. : Beitrag zur Entwicklungsgeschichte des Steigbiigels und ovalen Fen- sters. Mitteilung aus dem embryol. Institut der k. u. k. Universitat in Wien. 1877. Monatsschrift fiir Ohrenheilkunde. 1878. GiJLKE, Karl: Verlauf und Verknocberung der Stimnaht. Diss. med. Gdttingen 1904. GuNTHEE, A. Fr. : Beobachtung iiber die Entwicklung des Gehororgangs bei Menschen und boheren Saugetieren. Leipzig 1842. Hammae, J. Aug. : Studien iiber die Entwicklung des Vorderdarms und einiger angrenzenden Organe. Arch, fiir mikr. Anat. Bd. 59. 1902. Hamy, E. T. : De I'osteogenie du frontal ehez I'homme, a propos d'une double anomalie d'ossification de cet os observe cbez. un monstre notencepbale. Bull, du Mus. d'Histor. Nat. 1900. Hannover, Adolph: Primordialbrusken og dens Forbening i det meneskelige Kranium for Fodselen. Danske Videnskabemas Selskabs Skrifter. Vol. 11, p. 349-528. 1880. Le cartilage primordial et son ossification ehez le crane humain avant nais- sance. Copenhag-ue 1881. Hasse, C. : Anatomische Studien. Entwicklung des Atlas. Leipzig 1872. Hegetschweiler, J. : Die embryologische Entwicklung des Steigbiigels. 2 Tafeln. Arch, fiir Anat. u. Physiol., Anat. Abt., Jahrg. 1898, S. 37. Heitzmann, Gael, und Boedeckek, C. F. W. : The Earliest Development of the Lower Jaw-bone. Dental Cosmos. Vol. 37. 1895. Hennebeeg, B. : Beitrage zur Entwicklungsgeschichte des Unterkiefers beim Men- schen. Med. Inaug. Diss. Berlm 1894. HoLL, M. : Uber die Bildung des Gesichtssehadels. Verhandlg. der Gesellschaft deutscher Naturforscher u. Arzte. 1894. Hrdlicke, Ales. : Division of the Parietal Bone in Man and Mammals. Bull. Americ. Mus. Nat. Hist. Vol. 19, p. 231. 1903. Hyrtl: Uber wahre und falsche Schaltknocben in der Pars orbitaria des Stirn- beines. Sitzungsbericht der k. Akademie d. AYissensch. zu AVien. Math. Natunvissensch. Kl. Bd. 42, Jhrg. 1860. Ihering, H. v. : Die Entwicklungsgeschichte des menschliehen Stimbeines. Arch. fiir Anat. u. Physiol, u. wiss. Medizin. S. 649-658. Jahrg. 1872. Jacobson : On Primordial-Craniet. Verhandlg. der Skandinavisehen Naturforsch.. Vers. Stockholm 1842. Jacoby, Martin : Ein Beitrag zur Kenntnis des menschliehen Primordialkraniums. Arch, fiir mikr. Anat. Bd. 44, S. 61-86. 1895. Joseph, L. : Osteologischer Beitrag iiber das Schlafenbein und den in ihm enthal- tenen Gehorapparat. Zeitschrift fiir rationelle Medizin. Bd. 28. 1866. Jurasz, a. : Eine seltene Anomalie des Schildknorpels und dessen Verbindung mit dem Zungenbein. Arch, fiir Anat. u. Entwicklungsg. S. 399^00. 1877. MORPHOGENESIS OP THE SKELETAL SYSTEM. 447 Kallius, E. : Beitrag-e zur Entwicklungsgeschiehte des Kehlkopfes. Anat Hefte. Bd. 9, S. 301. 1897. Kampen, van: Die Tympanalgegend des Siiugetiersehadels. Morjjh. Jahrb. Bd. 34, S. 321-422. 1905. Kazzandee, J.: Notiz iiber die Pneumatisation des Schlafenbeins beim Menschen. Anat. Anz. Bd. 26. 1905. Keibel, Fkanz: Zur Entwicklungsgeschiehte der Chorda bei Saugern. Arch, fiir Anat. u. Physiol. Anat. Abt. Jahrg. 1889. Keeckring, Theoi). : Spicilegium anatomieum continens osteogeniam f oetuum. Amstelodami 1670. KiKUTHi, J. : Der histologische Bau der Knochenblasen in der Nase nebst Bemer- kimgen iiber Wachstum und Entstehung derselben. Arch, fiir Laryngol. u. Rhinologie. Bd. 14, Heft 6. 1903. KiLLiAN, G. : Zur Anatomie der Nase mensehlieher Embryonen. Arch fiir Laryn- gologie. Bd. 2, Heft 2. 1895; Bd. 3, Heft 1 und 2. 1895; Bd. 4, Heft 1. 1896. Kjellberg, Knut : Beitrage zur Entwicklungsgeschiehte des Kief ergelenks. Mor- pholog. Jahrbuch. Bd. 32. 1904. KoLLiKEE, Th. : Zur Anatomie der Kieferspalte. Arch, fiir klin. Chirurgie. Bd. 26. 1881. Zur Zwischenkieferfrage. Centralblatt f. Chir. Jahrg. 11. 1884. Zur Odontologie der Kieferspalte bei der Hasencharte. Biolog. Centralblatt. Nr. 12. 1885. KoLLMANN, J. : Deux especes des variations correlatives de la crane faciale de I'homme. Assoc. Franc. Avanc. Se. Rouen. 1883. Varianten am os oceipitale besonders in der Umgebung des Foramen occipitale magnum. Verhandl. d. Anat. Gesellschaft. 1905. Anat. Anz. Erg. H. Bd. 30, S. 545-563. 1907. Der Canalis cranio-pharyngeus. Verb. d. Anat. Ges. Jena. 1904. Anat. Anz. Erff. H. Bd. 25, 1904. KtiSS, G., ET PiSSOT, L. : D'un prolongement constant observe sur les cartilages lateraux du nez de I'embryon humain. Bibliogr. Anat. Vol. 7. 1899. KtJSTNEE, Otto : Uber Trigonoeephalie. Ein Beitrag' zur Lehre von den fetalen Schadelsynostosen. Virchow's Arch. Bd. 83. 1881. Landzeet, Th. : Uber den Canalis cranio-pharyngeus am Schadel des Neuge- borenen. St. Petersburger med. Zeitschrift. Bd. 14. 1868. Centralblatt f. d. med. Wissensehaft. 1869. Lbche: Das Vorkommen und die morphologische Bedeutung des Pfannenknochens (os acetabuli). Internat. Monatsehr. f. Anatomie u. Histologie. S. 363, 1884. Langnick : Untersuchungen iiber das os Kerckringi. Inaug. Diss. Konigsberg i. Pr. 1898. Leuckaet, Friedr. S. : Untersuchungen iiber den Zwischenkiefer beim Menschen in seiner normalen und abnormalen Metamorphose. Stuttgart 1840. Levi, Guiseppe : Beitrag zum Studium der Entwicklung des knorpeligen Primordial- craniums des Menschen. Arch, fiir mikr. Anat. und Entwicklungsg. Bd. 55. 1900. Low, Alex.: On the Development of the Lower Jaw in Man. Proc. Anat. Soc. Great Britain and Ireland. May 1905. Journ. Anat. and Physiol. Vol. 39. 1905. LuBOSCH, W. : Die Stammesgeschiehtliche Entwicklung der Synovialhaut und der Sehnen mit Hinweisen auf die Entwicklung der Kiefergelenhe der Saugetiere Biol. Zentralbl. Bd. 28, S. 678-697. 1908. LuscHKA, H. : Das Nebentranenbein des Menschen. Arch, fiir Anat., Physiol, und wissensch. Med. Jahrg. 1858. 448 HmiAN EMBRYOLOGY. Macalister, a. : Notes on the Varieties and Morphology of the Human Lachrymal Bone and its Accessory Ossicles. Proc. Royal Soc. of London. Vol. 36. 1884 and Vol. 37. 1884. IIaggi, Leopoldo: Centri di ossifieazione e principal! varieta morphologische degli interparietali nell' uomo. Rendiconti del R. Inst. Lomb. di Scienze e Lettere. Ser. 2. Vol. 29. 1896. Centres d'ossifications et prineipales varietes morphologiques des interparietaux chez rhomme. Arch. Ital. de Biol. Vol. 26. 1896. Resultati di ricerche morfologische intorno ad ossa e fontanelle del cranio umano. Comm. preventura. Rend. Instit. Lomb. Scienze e Letteri. Vol. 29. 1896. Resultats de recherches morphologiques sur des os et des fontanelles du crane humain. Comm. preventure. Arch. Ital. de Biol. Vol. 27, Ease. 2. 1897. Altri resultati di ricerche morphologiche intorno ossa craniali, craniofaciali e fontanelle dell' uomo ad altri mammiferi. Rend, dell' Instit. Lomb. Ser. 2. Vol. 30. Milano 1897. Autres resultats de recherches morphologiques sur des os craniens et eranio- faciaux et sur des fontanelles de I'homme et d'autres mammiferes. Arch. Ital. de Biol. Vol. 30, Fase. 2. 1898. Suture ed ossa interparietali nell' cranio umano di bambino e di adulto. Rendiconti R. Istit. Lomb. di Scienze e Lettere. Ser. 2. Vol. 37, Fasc. 9. 1904. jMagitot, E., et Robik, Chr. : Memoire sur un organe transitoire de la vie foetale designe sous le nom de cartilage di Meckel. Annales des Sciences Naturelles, 4 Serie, Zoologie. T. 18. 1862. Martini, Vit. : Sullo sviluppo delle ossa piatte seeondarie del cranio de connettivo e suUa loro rigenerazione dopo la trapanazione. R. Univ. Siena, 36 pp. 1886. Masquelin, H. : Recherches sur le developpement du maxillaire inferieur de I'homme. Bull. d. I'Acad. Royale des Sciences, des Lettres et des Beaux- Arts de Belgique, p. 371-376. 1878. Matiegka : Tiber das " Os malare bipartitum." Anat. Anz. Bd. 16. 1899. Mayet: Quelques considerations sur le developpement du conduit auditif externa. Bull, de la Soc. Anat. de Paris. Annee 69. 1904. Mead, C. S. : Chondroeranium of an Embryo Pig. Amer. Jour, of Anat. Vol. 9, pp. 167-210. 1909. Meckel, .T. F. : Uber die Zwickelbeine am menschlichen Schadel. Beitrag z. vergl. Anat. Bd. 1, H. 2. 1809. Handbuch der menschlichen Anatomic. Bd. 2. Besondere Anatomic, Knochen- lehre, Banderlehre, Muskellehre. 1816. Bd. 4. Eingeweidelehre und Ge- schichte des Fetus. 1820. Menden", J. : Beitrag zur Casuistik der Ossificationsdefecte am Schadel des Neuge- borenen. Marburg 1S75. Meter, H. v. : Der Zwischenkieferknochen und seine Beziehungen zur Hasencharte und zur schragen Gesiehtspalte. Deutsche Zeitschrift f. Chirurgie. Bd. 26. 1884. ^IiES, Joseph : Uber die Knochelchen in der Symphyse des Unterkiefers von neu- geborenen Menschen (Ossicula mentalia). Anat. Anz. Bd. 8. 1893. MiHALKOVics, Victor v.: Wirbelsaite und Hirnanhang. Arch. f. mikr. Anat. Bd. 11. 1875. Bau und Entwicklung der pneumatischen Gesichtshohlen. Verh. d. Anat. Ges. 10. Vers. Berlin 1896. Anatomie und Entwicklungsgeschiehte der Nase und ihrer Nebenhohlen. Hey- mann's Handb. der Laryngol. u. Rhinologie. Bd. 3. 1896. Nasenhohle und Jacobson'sches Organ. Anat. Hefte. Bd. 11, H. 1/2. 1899. MORPHOGENESIS OF THE SKELETAL SYSTEM. 449 MiLOSLAwSKi, Fe. W. : Die Stimhohlen. Eine anatomisch-topographische und craniologische Untersuchung. Diss. med. Moscow. Ref. in Stieda's 6. Be- richt in: Ergebn. Anat. und Entwicklungsg. Bd. 13. 1903. Moldenhauer: Beitrage zur Anatomie und Entwicklungsgeschichte des mensch- liehen Gehororgans. Arch. f. Obrenheilk. Bd. 11. 1876. MouEET, Jules : Rapports du sinus frontal avec les cellules ethmoidales. 22 Fig. Bull, et Mem. de la See. Franc. d'Otol., de Laryngol. et de Rhinol. Con- gres 1901. MiJLLEE, Heineich: Uber das Vorkommen von Resten der Chorda dorsalis bei Menschen nach der Geburt und iiber ilir Verhaltnis zu den Gallertgeseh- wiilsten am Clivus. Zeitschr. f. rationelle Medizin. 3. Reihe. Bd. 2. 1858. Munch, F. E.: Uber die Entwicklung des Knorpels des ausseren Ohres. Morph. Arb. Bd. 7, S. 583. 1897. Musumeci: Sopra un caso singulare di terzo condilo. Monitore Zool. Ital. 1900. Nicola, Beniamino : Sullo sviluppo, sui eanali perf oranti e sulle f essure della por- zione laterale dell' ala magna dell' os sfenoidale nella specie umana. Mem. Accad. Sc. Torino. Ser. 2. V. 56, pp. 117-146. Torino 1906. Nicolas, A. : Recherches sur le developpement de quelques elements du larynx humain. Bibliographie anatomique. No. 5, p. 176-190. 1894. NooRDEKj W. van : Beitrag zur Anatomie der knorpeligen Schadelbasis bei mensch- lichen Embryonen. Arch. f. Anat. u. Physiol. Anat. Abt. S. 241-257. 1887. Olii : How many Months does the Larger Fontanelle of the Japanese Infant Re- quire to Close? The Tokio Jji-Shinoki. Nr. 1175. 1900. Papillault, G. : Ontogenese et phylogenese du crane humain. 8 Fig. Rev. ecole d'anthrop. Nr. 4, p. 105-153. Paris 1899. Paravicini, G. : Di alcune nuove ossicina suturofontanellari del cranio umano giovano ed adulti, appartenente ad alienati ed a normali. R. 1st. Lomb. di Se. e Lett., Ser. 2, Vol. 35. 1902. Paeiselle, H. : Des f ontanelles : anatomie et pathologic. These de doctorat en med. Paris 1900. Patel: Sinus frontaux et cellules ethmoidales anormalement developpes. Lyon med., p. 319-320. 1902. Paulisch, Otto: Das vordere Ende der Chorda dorsalis und der Frank'sche Nasenkamm. Arch. f. Anat. u. Physiol. Anat. Abt. 1887. Pauli, Simon : Uber Pneumaticitat des Schadels bei den Saugetieren. Morphol. Jahrb. Bd. 28, H. 4. 1900. Perna, Giovanni: Die Nasenbeine. Eine embryologische und vergleichend-ana^ tomische Untersuchung. Arch. f. Anat. u. Physiol. Anat. Abt. S. 119-154. 1906. Petee, Kael : Anlage und Homologie der Museheln des Menschen und der Sauge- tiere. Arch. f. mikr. Anat. u. Entwicklungsg. Bd. 60. 1902. PoLZL, Anna: Zur Entwicklungsgeschichte des menschl. Gaumens. Anat. Hefte. Bd. 27, S. 243-283. 2 Taf., 13 Fig. 1904. POLITZER, A. : Zur Anatomie des Gehijrorgans. 2. Uber den Processus styloideus. Arch. f. Obrenheilk. Bd. 4. 1875. Beitr. z. Anat. u. Physiol, als Fest- gabe f. Carl Ludwig. Leipzig 1875. Ranke, Johannes: Der Stirnfortsatz der Schlafenschuppe bei den Primaten. Sitzungsberichte der math.-phys. Klasse der k. Bayr. Akad. der Wissen- schaften zu Miinchen. Bd. 28. 1899. Die iiberzahligen Hautknochen des menschlichen Schadeldaches. Abhand- lungen der k. Bayr. Akademie der Wiss. 2. Klasse. Bd. 20. Miinchen 1900. Vol. I.— 29 450 HUMAN EilBEYOLOaY. Raubee, a. : Sehlaf enbein des Menschen, ohne pars tympaniea, mit Hammerrudi- ment. Morpholog. Jahrbuch. Bd. 38, S. 1. 1907. Regnault, Felix : Obliteration prematuree des sutures craniennes, meeanisme des deformations. 2 Fig. Bull, et Mem. de la Soc. d'Anthrop. de Paris. Ser. 5, Vol. 1. 1900. Reicheet, C. : Uber die Yiseeralbogen der Wirbeltiere im allgemeinen und deren Metamorphose bei den Vogeln und Saugetieren. Miiller's Archiv f. Anat., Physiol, und wissensch. Mediz. 1837. Rizzo, Agostino : Canale cranio-f aringeo, f ossetta f aringea, interparietali e preinter- parietali nel cranio umano. Monit. Zool. Ital., Anno 12, p. 211-252. 1901. Rousseau, Emjianuel: Description d'un nouvel os de la face chez I'homme ("lachrymal externe ou petit unguis"). Annales des Sciences Naturelles. Vol. 17. 1829. Rudinger : Uber die BUdung der Kanale und Hohlraume im menschlichen Schlafen- bein. Beitrage zur Anatomie des Gehbrorgans. Mlinchen 1876. Ruge, G. : Beitrage zum Wachstum des menschl. Unterkiefers. Inaug. Diss. Ber- lin 1875. Saint-Rejiy, G. : Reeherehes sur I'extremite anterieure de la corde dorsale chez les amniotes. Archives de Biologie. T. 14. 1896. SCHAEPrER, OsKAR : Untersuchungen iiber die normale Entwicklung der Diniensions- verhaltnisse des fetalen Menschenschadels. 51 pp. Miinehen und Leipzig 1892. ScHAPFER, J. : Die Verknocherung des Unterkiefers und die ]\Ietaplasiefrage. Arch. f. mikr. Anat. Bd. 32, S. 266-387. 1888. Zur Histologie, Histogenese und phylogenetischen Bedeutung der Epiglottis. Anat. Hefte. Bd. 38, S. 455-490. 1907. SCHATTAUEE, F. : Beitrag zur Kenntnis der Mikrognathie. Diss. Konigsberg 1903. ScHENK, Fr. : Die erste Anlage des Unterkiefers und der Zahnalveolen. S.-B. K. Akad. der Wiss. Wien, Math.-Naturw. Kl. Bd. 105. 1896. Viertel.i'ahrsschrift f. Zahnheilk. Jahrg. 12. 1896. ScHLAGiNHAuPEN, Otto : Ein Canalis craniopharyngeus persistens an einem Men- sehenschadel und sein Vorkommen bei den Anthropoiden. Anat. Anz. Bd. 30, S. 1. 1907. Schmidt, E. : Uber das postembryonale Weiterbestehen des Jacobson'schen Organs und Knorpels beim Menschen und die Beziehung derselbeu zueiauder. Diss. d. Med. Fakultat. Berlin 1896. SCHONEMANN, A. : Beitrag zur Kenntnis der Muschelbildung und des Muschel- wachstums. Anat. Hefte. Bd. 18, H. 1. 1901. Schumacher, Siegm. v. : Hamalbogen bei menschl. Embryonen. Anat. Anz. Bd. 29. Ergh. S. 173-178. 1906. Ein Beitrag zur Frage der Manifestation des Oceipitalwirbels. Anat. Anz. Bd. 31, S. 145. 1907. Schwalbe, E. : Entstehung der Hasenscharte naeh der Anschauung des vorigen Jahrhunderts. Miinch. med. Wochenschrift. 1899. Schwalbe, G. : Uber die Nasenmuscheln der Saugetiere und des Menschen. Sitz- ungsber. d. Physik.-Okon. Ges. zu Konigsberg. 1882. Uber accessorische Schiidelknoehen des Menschen und accessorische Schiidel- niihte. Wiener klin. Woehenschrift. 1899. Uber die Fontanella metopica und ihre Bildung. Zeitsehr. f. Morphol. u. Anthropologic. Bd. 3, H. 1. 1901. Uber den supranasalen Teil der Stimnaht. Zeitsehr. f. Morphol. und An- thropologic. 1901. Uber geteilte Scheitelbeine. Zeitsehr. f. MorpLol. und Anthropologie. Bd. 6, S. 361-434. 1903. Uber die Stirnnaht bei den Primaten. Zeitsehr. f. Morphol. u. Anthropologie. Bd. 7. 1904. MORPHOGENESIS OP THE SKELETAL SYSTEM. 451 ScHWEGEL, Andr. : Knochenvarietiiten. Zeitschr. f . ration. Med. Bd. 5. 1859. Sergi, G. : Le forme del cranio umano nello sviluppo f etale in relazione alle forme adulte. Rev. de Sc. Biol., Anno 2, p. 401-413, p. 831-847. 1900. Seydel, 0. : Uber die Nasenhohle der hoheren Saugetiere mid des Mensehen. Mor- phol. Jahrb. Bd. 17. 1891. SiEBENMANN : Die ersten Anlagen vom Mittelohrraum und Gehorknoehelchen des menschlielien Embryo in der vierten bis sechsten Woche. Arch. f. Anat. u. Physiol. Anat. Abt. S. 355. Jahrg-. 1894. Mittelohr und Labyrinth. Bardeleben's Handb. der Anat. des Mensehen. Bd. 5, 2. Abt. Jena 1897. Simon, Max : Uber angeborene knoeherne Choanalatresie im Anschluss an zwei neue Palle. Inaug.-Diss. Leipzig 1897. Smith, G. Elliott: On a Case of Fusion of the Atlas and Axis. Anat. Anz. Bd. 31, S. 166-168. 1907. SoKOLOw, Paul : Der eanalis craniopharyngeus. Arch, f . Anat. u. Physiol. Anat. Abt. 1904. SouLiB, A., ET Bardier, E. : Recherches sur le developpement du larynx. Joum. de I'Anat. et de la Physiol. Vol. 43, p. 219. 1907. SouQUES ET Heller: Paralysie faciale congenitale par agenesie du rocher. Arch. Intemat. d'Otol. 1903. Sparbel, H., and George, E. : The Articulation of the Vertebrate Jaw. Proc. Zool. Soc. London. Vol. 1, p. 114-123. 1906. Spee, F. Graf v. : Kopf . Bardeleben's Handbuch der Anatomie des Mensehen. Bd. 1, Abt. 2. Jena 1896. Spebino, G. : La ossificazione e la posizione della trochlea del musculus obliquus sup. oculi, la spina e la fovea trochlearis. Mem. R. Accad. di Seienze Lettere ed Arti Modena. Ser. 3. Vol. 6. Modena 1905. Spondli, Heinrich : Uber den Primordialsehadel der Saugetiere und des Meaischen. Inaug.-Diss. Zurich 1846. Spurgat, F. : Beitrage zur vergl. Anat. der Nasen- und Schnauzenknorpel des Mensehen und der Tiere. Morphol. Arb. Bd. 5, H. 3. 1896. Stadebini, R. : Intomo alle prime fasi di sviluppo dell Anulus stapedialis. Monit. Zool. Ital. Vol. 2, Anno 2. 1891. Staubenghi, C. : Ricerche intorno all' ossificazione della squama dell' oceipitale umano, a proposito di un esemplare di os incae tripartitum. Gazz. Med. Lomb., Anno 56. 1897. 2 Titel. Communieazioni preventive di cranio- logia. 4. Dell' ossificazione del frontale umano. Gazz. Med. Lomb., Anno 57. 1898. Dell' esistenza di ossa pre- e postfrontali nel cranio umano e dei mammiferi, eon un' appendice sulla questione dell' osso sfenotico dei mammiferi. Ref. in Monit. Zool. Ital., Anno 2. Milano 1891. Contribuzione alia osteogenesi dell' oecipitale umano a dei mammiferi con una nota suUo sviluppo del frontale e del parietale dell' uomo. 5 Taf. Pavia 1899. Steiner, F. : Uber die Entwicklung der Stirnhohlen und deren krankhafte Erwei- terung durch Ansammlung von Eliissigkeiten. Arch. f. klin. Chir. Bd. 13. 1872. Sternberg, M. : Ein bisher nieht besehriebener Kanal im Keilbein des Mensehen. Anat. Anz. Jahrg. 3. 1888. Ein bisher nieht besehriebener Kanal im Keilbein des Mensehen und mancher Saugetiere. Arch. f. Anat. u. Physiol. Anat. Abt. 1890. Stieda, Hermann: Die Anomalien der menschl. Hinterhauptsschuppe. Anat. Hefte. Bd. 2. 1892. Stieda, L. : Uber die Stirnnaht und den Stirnfontanellenknochen. C. R. 12. Congr. Intern. Med., Moscou 1897. T. 2. 1899. 452 HUMAN EilBRYOLOGY. Strasser, H. : Sur le developpement des cavites nasales et du squelette du nez. Arch, des Sc. Phys. et Nat. Genf. 1901. Sutton, J. Bland: Observations on the Parasphenoid, the Vomer and the Palato- pterygoid Arcade. Proceedings Zool. Soc. London 1884. On the Relations of the Orbito-Sphenoid to the Region Pterion in the Side Wall of the Skull. Journ. of Anat. and Physiol. Vol. 18. 1884. On the Development and Morphology of the Human Sphenoid Bone. Proceed- ings Zool. Soc. London 1885. SwjETSCHNiKOw : Uber die Assimilation des Atlas und die Manifestation des Occi- pitalwirbels beim Menschen. Arch. f. Anat. u. Physiol. S. 155-194. 1906. Symington : On the Relations of the Larynx and Trachea to the Vertebral Column in the Foetus and Child. Journ. of Anat. and Physiol. Vol. 9, p. 3. 1875. Tenchini, Lorenzo: Sopra il canale infra-squamoso di Gruber nell' uomo. Arch. Ital. Anat. e Embriol. Vol. 3. 1904. Tenchini, L., e. Zimmerl, U. : Sopra i ossa detti " Wormiani " della fossa cranii anteriori nell' uomo. Parma 1902. Thoma, R. : Synostosis suturae sagittalis cranii. Ein Beitrag zur Histomechanik des Skeletts und zur Lehre von dem interstitiellen Knochenwachstum. Vir- chow's Arch. Bd. 188. 1907. Thomson, Arthur: The Orhito-maxillary Frontal Suture in Man and the Apes, with Notes on the Varieties of the Human Lacrymal Bone. Journ. of Anat. and Physiol. Vol. 24. 1890. TOLDT, C. : Osteologische Mitteilungen. 1. Die Entstehung und Aushildung der conchae und der sinus sphenoidales Ijeim Menschen. 2. Uber die Entwick- lung des Scheitelbeines beim Menschen. Lotos. Jahrb. f. Naturwiss. N. F. Bd. 3 und 4. Prag 1882. Uber die Entwicklung des Scheitelbeines des Menschen. Prager Zeitschr. f. Heilk. Bd. 4. 1883. Uber die Kinnknochelchen und ihre Bedeutung f. die Kinnbildung beim Men- schen. Korresp.-Bl. d. Deutsch. Ges. f. Anthrop. u. Ethnol. Urgeschichte. Jahrg. 36, S. 115-118. 1905. Die ossicula mentalia und ihre Bedeutung f. die Bildung des menschlichen Kinnes. S. B. K. Accad. Wiss. in Wien. Bd. 114, Abt. 3. Wien 1905. Zur Frage der Kinnbildung. Korresp.-Bl. d. deutsch. Ges. f. Anthropol. u. Ethnol. Urgeschichte. Jahrg. 37, S. 9-17. 1906. ToLDT, Karl, Jr. : Entwicklung und Struktur des menschlichen Jochbeines. Sitz- ungsber. d. kais. Akad. d. Wiss. in Wien, Mathem.-naturwiss. Klasse. Bd. 109, Abt. 3. S. 562-569. 1902. Varietaten des menschlichen Jochbeines im Lichte der Entwicklungsgeschichte. Verb. d. Ges. deutsch. Naturf . u. Aerzte. Vers. 74. Bd. 2, S. 182-] 85. 1903. Die Querteilung des Jochbeines und andere Varietaten derselben. Sitzungsb. d. kais. Akad. d. Wiss. in Wien. Mathem. naturwiss. Klasse. Bd. 112, Abt. 3. 1903. Tomes, C. S. : Studien liber das Wachstum der Kieferbeine. Jahrbuch f . Zahn- heilk. Jahrg. 7, Nr. 9 und 10. 1892. ToYO, C. : Le forme dell cranio nello sviluppo f etale. Atti. Soc. Roman. Antropol. Vol. 11. 1905. ViRCHOW, Rudolph : Untersuehungen iiber die Entwicklung des Schadelgrundes im gesunden und krankhaften Zustande und iiber den Einfluss deselben auf Schadelform. Gesichtsbildung and Gehimbau. Berlin 1857. VoiT, Max : Das ehondroeranium und die Deckknoehen von lepus cunieulus. Anat. Hefte, Bd. 38 (116 Heft) 192 Stn. mit 16 Tafeln, 1909; also Freiburger Habilitationssehrift. Vrolik, a. J.: Studien liber die Verknocherung und die Knoehen des Sehadels der Teleostier. Niederland. Arch. f. Zool. Bd. 1. 1873. MORPHOGENESIS OF THE SKELETAL SYSTEM. 453 Walkhopp: Der mensohliehe Unterkiefer im Lichte der Entwicklungsmeehanik. Deutsch. Monatssehr. f. Zahnheilk. Jahrg. 18, S. 529-538. 1900; Jahrg. 19, S. 111-121. 1901. Die heutigen Theorien der Kinnbildung. Deutsch. Monatssehr. f . Zahnheilk. Jahrg. 23, S. 580-592. 1905. Wallisch : Das Kiefergelenk. Arch, f . Anat. u. Physiol. Anat. Abt. S. 303-310. 1906. Wartnski: Contribution a I'etude du bee de lievre simple et complexe. Vir- ehow's Arch. Bd. 112, p. 507-536. 1888. Weber, A. : Evolution de la region pterygoide, chez I'homme. C. R. Soc. Biol. Paris. T. 58, p. 1083-1084. 1905." L'orientation des ailes des apophyses pterygoides chez les primates. C. R. Soc. Biol. Paris. T. 59, p. 225-227. 1905. Les apophyses pterygoides du crane de I'homme. Developpement, orientation, variation. Bibliog. Anat. T. 15, p. 57-84. 1906. Les variations ethniques du trou ovale du sphenoide humain. Bibliogr. Anat. Paris 1906. Weidenreich, Franz : Die Bildung des Kinnes und seine angebliche Beziehung zur Sprache. Anat. Anz. Bd. 24. 1904. Die Kinnbildung beim Mensehen. Anat. Anz. Bd. 25. 1904. Walkhoif' s Theorie der Kinnbildung. Deutsche Monatsschrift f. Zahnheilk. Jahrg. 23, S. 747-752. 1905. Welker, H. : Uber Wachstum und Bau des menschl. Schiidels. Leipzig 1862. WiCHERT, Paul: Uber den Canalis ethmoidalis. Med. inaug. Diss. Konigsberg 1891. Williams, L. W. : The Later Development of the Notochord in Mammals. Amer. Jour, of Anat. Vol. 8, p. 251-284. 1908. Wolff, J. : Uber das Wachstum des Unterkiefers. Virchow's Arch. Bd. 114, S. 493 bis 547. 1888. Zabel, E. : Varietaten und voUstandiges Pehlen des Tranenbeins beim Mensehen. Anat. Hefte. Bd. 15, S. 153-201. 1900. Zander, Richard : Beitrage zur Morphologic der Dura mater und zur Knochen- entwicklung. Festschr. z. 70. Geburtstag von Carl v. Kupffer. 1899. Zanotti: La fontanella metopica ed il suo signiticato. Bolletino Science Mediche Bologna. Ser. 8, Vol. 2. 1902. Zondek, M. : Beitrage zur Entwicklungsgeschichte der Gehbrknochelchen. Arch. f. mikr. Anat. u. Entwicklungsg. Bd. 44, S. 499. 1895. Ztjckerkandl, C. : Zur Anatomic und Entwicklungsgeschichte der Nasoethmoidal- region. Medic. Jahrb. Herausg. v. d. k. u. k. Gesellsch. d. Aerzte. Wien 1878. Normale und pathologische Anatomie der Nasenhohle und ihrer pneumatischen Anhange. Bd. 1. Wien 1882. Uber die morphologische Bedeutung des Siebbeinlabyrinthes. Wien. med. Wochenschrift. Bd. 37. Wien 1887. Die Siebbeinmuscheln des Mensehen. Anat. Anz. Jahrg. 7. 1892. Die Entwicklung des Siebbeins. 6. Verhandlg. d. Anat. Gesellsehaft. (Wien) Anat. Anz. Bd. 7. Erg.-H. S. 261-264. 1892. Anat. norm, et path, des fosses nasales. Trad, frang. Paris 1895. Geruehsorgan. Ergebn. d. Anat. u. Entwicklungsgesch. Bd. 2. 1893. Anatomie und Entwicklungsgeschichte des Kehlkopfes und der Luftrohre. Handbuch der Laryngologie und Rhinologie. Bd. 1. S. 55-133. Wein 1896. Uber die Entwicklung der concha bullosa. Monatschr. f. Ohrenheilk. sowie Kehlkopf-, Nasen-, u. Rachenkrankheiten. N. F. Jahrg. 33. 1899. Uber den Jaeobsonsehen Knorpel und die Ossification des Pflugscharbeines. S. B. d. Wiener Akad. Wiss. math.-naturw. Kl. Bd. 117, Abt. 111. S. 1-55. 2 Tafeln und 26 Textfig. Wien 1909. XII. THE DEVELOPMENT OF THE MUSCULAR SYSTEM. By warren H. LEWIS, Johns Hopkins University, Baltimore, Md., U. S. A. The entire musculature, both cross-striated and smooth, with a few exceptions that are of ectodermal origin, arises from the mesoderm. Since contractility is a fundamental property of all cells it is not surprising that the ectoderm as well as the mesoderm should give rise to cells in which this function is highly developed. The ectoderm of the optic cup undoubtedly gives rise in most vertebrates, probably in all, to the musculus sphincter and the musculus dilatator pupillae: Nussbaum (1900, 1902), Herzog (1902), Heerfordt (1900), Szili (1901), Lewis (1903), Collin (1903). In mammals these muscles are of smooth fibres, but in birds and reptiles of cross-striated fibres. The muscles of the glandulas sudoriferae are also of smooth muscle derived from the ectoderm: Eanvier (1889), Koelliker (1889), Stohr (1902), Heidenhain (1893). The mesoderm, however, gives rise to the great bulk of the musculature, both smooth and cross-striated. The smooth and cross-striated muscles are not to be considered as fundamentally different ; they represent different grades of development of con- tractile tissue or different paths of differentiation from a common fundamental form. The smooth muscle shows a lower grade of differentiation. In insects and birds, for example, portions of the intestinal tract are supplied with cross-striated muscle, while in mammals the corresponding regions are supplied by smooth muscle. Marchesini and Ferrari (1895) found that in early de- velopment smooth and striped muscles show exactly the same structure. The fact that the myotomes give origin to many of the cross-striated muscles does not distinguish this variety from the smooth muscle, inasmuch as many of the cross-striated muscles — in the head, for example — arise directly from the mesoderm quite independently of the myotomes and in a manner similar to the origin of the smooth muscle. It has been customary to consider this voluntary musculature as being derived almost entirely from the primitive segments; yet in mammals the attempts to homologize the head muscles 454 DEVELOPMENT OF THE MUSCULAR SYSTEM. 455 with those of the trunk derived from the myotomes have failed, as there are no indications of preotic segments in the head region; the head muscles develop directly from the mesoderm of the branchial arches and the dorsal eye region. The muscles of the limbs likewise arise directly from the mesoderm of the limb buds ; but here uncertainty still exists as to what role migrating cells from the myotomes may play in their development. The muscles derived directly from the myotomes— namely, the deep muscles of the back and the intrinsic thoraco-abdominal mus- culature — are to be considered as both phylogenetically and on- togenetically the oldest of the skeletal muscles. The skeletal muscles of the head and limbs are, on the other hand, of later origin, and probably not derived from the more primitive seg- mental musculature, but directly from the mesoderm. Until, how- ever, we have a more complete picture of the developmental history, not only in mammals but also in the lower vertebrates, the relationships must remain obscure. The development of many of the muscles in man and mammals has never been traced, and of the remainder our knowledge is fragmentary and incomplete. It has commonly been supposed that the first differentiation of the muscles from the mesoderm takes place under the influence of the nervous system through the agency of the motor nerves, and that self-differentiation of muscles does not occur. Such a belief arose from our knowledge of the very early union of the motor nerves with the developing myotomes and muscle masses. Teratological evidence is at present conflicting. Neumann (1891), from his studies of acephalic and amyelic monsters, concludes that the influence of the motor nerves is necessary for the dif- ferentiation of the muscular system. LeonoAva (1893) and Fraser (1895) have described human monsters without brain and spinal cord in which the peripheral sensory nerves and musculature were normally developed. On the other hand, E. H. Weber (1851), Neumann (1901), and others have described cases in which ab- sence of certain portions of the central nervous system has been accompanied by total absence of musculature which is normally supplied by the lacking nerves, although skeleton, blood-vessels, and even tendons were normally developed. Neumann would rec- oncile these apparent differences by assuming that muscles first arise under the influence of the nervous system, but that their nourishment and further growth during the embryonic period take place independently of the central nervous system, and not until the post-embryonic life is reached is the dependence again established. Thus, the nervous system must have developed in the early stages of embryonic life up to a certain point and under- gone degeneration after differentiation of the muscular system had taken place. Herbst (1901) concludes, from the same data, 456 HUMAN EMBRYOLOGY. that the sensory nerves, including the cells of the spinal ganglia, and not the motor nerves, are necessary to stimulate the dif- ferentiation of the muscular substance in the embryo. The well- known fact that a muscle undergoes atrophic changes after its nerve supply has been cut off would at first sight uphold the view that the influence of the nervous system is necessary for the differentiation of contractile tissue. The study of the normal development likewise affords some evidence which might be in- terpreted as tending to support it, though it does not necessarily do so. In the embryos of lower vertebrates, for instance, the connection of the motor spinal nerves with the muscle plates is established just at the time when the contractile substance begins to be laid down, but in the pig embryo, according to Bardeen (1900), the musculature is differentiated to a considerable extent before the nerves establish a connection with it: Harrison (1904). It was only by the experimental method on the lower verte- brates that this question seems to have been finally settled, especially by the brilliant work of Harrison (1904). Harrison removed the spinal cord in a series of frog embryos before the histological differentiation in either the muscular or nervous system had begun, so that from the very beginning isolation of the musculature from the nervous system was complete, and all chance for the exertion of any peculiar formation stimulus emanat- ing from the nervous system as such was eliminated; and like- wise, owing to the consequent paralysis of the muscles in question, any possible stimulus resulting from the functional activity of the muscle itself was excluded. Still the differentiation of the contractile substance took place in the normal manner, as did the grouping of fibres into individual muscles. It is not likely that the conditions in mammals and man differ from those of the frog. Thus it is seen that all the constructive processes involved in the production of the specific structure and arrangement of the muscle-fibres take place independently of stimuli from the nervous system and of the functional activity of the muscles themselves. Cross-striated muscle tissue and the individual muscles are thus self -differentiating. At a later period during functional activity, as the experiments of severing the nerve to a muscle show, the muscle becomes dependent on the influence of the nervous system for its continued normal existence, either through a trophic in- fluence or functional actiAnty. At how early a period in the de- velopment of the ovum this power of self -differentiation of mus- cular tissue begins can be but problematical. Lewis (1907) found that in the early gastrula stage of the frog, tissue in the lips of the blastopore possesses the power of independent differentiation into muscular substance. Thus it appears that muscle tissue is already predetermined in the early gastrula. It is probable that DEVELOPMENT OF THE MUSCULAR SYSTEM. 457 this predetermination exists much earlier, even in the egg itself. Conklin (1895) has been able to determine in ascidian eggs, even before cleavage begins, the existence of organ-forming substances, one of which, the myoplasm, that has to do with the formation of muscle tissue, is clearly recognizable and can be followed through successive stages of development into formed muscle. Harrison's experiments likewise show that the formation of the individual muscles from the myotomes and muscle complexes takes place independently of the nervous system. It has often been assumed that this splitting up of the complex muscle masses into the individual muscles has come to pass through the active ingrowth of connective tissue, blood-vessels, and nerves. There is, however, no experimental evidence indicating that either the connective tissue, the blood-vessels, or nerves take an active part in this process, although descriptive observation might readily lead one to such a conclusion. It is more probable that the ex- planation lies primarily within the muscle mass itself and secon- darily to the relations which the muscle masses may have to shifting skeletal elements. Harrison's experiments would also eliminate functional activity as a necessary factor in this forma- tion of individual muscles. Although the nervous system does not influence muscle dif- ferentiation, the nerves, owing to their early attachment to the muscle rudiments, are in a general way indicators of the position of origin of many of the muscles, and likewise in many instances the nerves indicate tlie paths along which the developing muscles have migrated during development: Mall (1898), Nussbaum, Bar- deen and Lewis (1900), Lewis (1902), Bardeen (1907), Futamura (1906), G-rafenberg (1905). The muscle of the diaphragm, for example, has its origin in the region of the fourth and fifth cervical segments. The nervus phrenicus early enters the muscle mass and is carried with the muscle in its migration through the thorax. The Mm. trapezius and sternocleidomastoideus arise in the lateral occipital region as a common muscle mass, into which at a very early period the nervus accessorius extends, and as the muscle mass migrates and extends caudally the nerve is carried with it. The Mm. pectoralis major and minor arise in the cervical region and receive their nerves while in this position; then the mass migrates caudally and ventrally over the thorax. The Mm. latissimus dorsi and the serratus anterior are excellent examples of migrating muscles whose nerve supply indicates their origin in the cervical region. The M. rectus abdominis and the other muscles of the abdominal wall migrate or shift from a lateral to a ventrolateral or ventral position, carrying with them the nerves. The nervus facialis, which early enters the common facial muscle mass of the second branchial arch, is dragged about with 458 HUMAN EMBRYOLOGY. the muscle as it spreads over the head and face and neck, and the nerve splits into its divisions parallel with the splitting of the muscle mass into its various muscles. The nerve supply serves in part as a key to the common origin of certain groups of muscles. The nervus oculomotorius enters in the early embryo a common muscle mass which later splits into various eye muscles supplied by it. The nervus trigeminus first enters the common muscle mass in the mandibular arch which later splits into the various muscles supplied by this nerve and its branches. The lingual muscles arise from two muscle masses which are supplied by the two hj'poglossal nerves. The infra- hyoid muscles arise from a common mass supplied by the ramus descendens nervi hypoglossi. The M. trapezius and the M. sterno- cleidomastoideus arise from a single mass. Wliere a muscle is supplied by nerves from two or more segments, the indication is that such muscle has had a complex origin, as the Mm. rectus abdominis, obliquus externus and in- ternus ; but this is not always the case, for a muscle may receive secondarily new nerves, retaining at the same time its original nerve, as the M. trapezius, which was originally supplied only by the nervus accessorius and later receives branches from the cervical plexus. The M. digastricus also is, according to Futamura (1906), at first entirely supplied by the nervus facialis, later, as the anterior belly becomes constricted off from the posterior, the former obtains its motor nerve secondarily from the nervus mylo- hyoideus. The site of entry of a nerve into the muscle, as a rule, marks the region of earliest differentiation (Bardeen, 1907), and in many instances at least the distribution of the nerve within the adult muscle indicates the course of development or growth of that muscle (Nussbaum, 1894). HISTOGENESIS. Smooth Muscle. — We have already noted that smooth muscle may arise either from the ectoderm or the mesoderm; the great bulk of smooth muscle, however, arises in situ from the mesoderm, either directly from the mesenchymal derivatives of the meso- derm or from embryonal connective tissue. The various stages in its histogenesis from the mesoderm have never been carefully traced in man, and the following account is based on the excellent work of McGill (1907) on the histogenesis of the smooth muscle in the alimentary canal of the pig. The mesench;^Tiie which arises from the mesodermal germ layer is in the form of a syncytium, with protoplasmic continuity throughout the entire syncytial mass. " The nuclei of the DEVELOPMENT OP THE MUSCULAR SYSTEM. 459 syncytium are round or oval, with distinct nuclear walls and heavy chromatin reticulum." " The protoplasm shows a fine reticular structure, the strands of the reticulum being made up of rows of fine granules " (Fig. 325). This syncytium persists throughout intra-uterine life and even in the adult. "Separate and distinct smooth muscle-cells or fibres do not exist at any stage of development. ' ' Thus the term ' ' cell ' ' as here used refers to the enlarged, thickened por- tion of the syncytium surround- ing the nucleus. "Before muscle development begins, there is a general condensation of the mes- enchyme," with multiplication of the cells, followed by marked elongation of the nuclei in the region in which muscle is to differentiate. Not all smooth muscle develops from this prim- itive mesenchyme, for in later stages the muscle comes from the more developed mesenchyme or embryonal connective tissue. In the alimentary tract, for example, the longitudinal muscle appears much later than the circular and arises from the embryonal con- nective - tissue syncytium into which the primitive mesenchyme has been transformed. "In the areas of muscle formation not all of the mesenchymal cells (or, in later development, of the em- --iiiji^ -— ^g^ bryonal connective-tissue cells) Fig. 325— (After McGill: Intemat. Monats- 1 , + ■ +I1 ■ + 1 ^''^^ ^^ ^^^*- "■ Phys-. Taf. vii, Fig. 1.) From elongate 5 some retain tneir Stel- thecesophagusof a T-mm. pig, showing condensed la to aVifiTKi wifh mrni nnr'lpi nnrl mesenchymal syncytium (m) with the reticular late Snape, Wltn OVai nUCiei, ana structure of the granular protoplasm. from these, in later development, muscle-cells may arise; but they form in the main the anlage of the interstitial connective tissue." "As the elongation of mesenchymal and connective-tissue nuclei continues, in the formation of muscle tissue, there is an increase in the amount of protoplasm surrounding each nucleus. The perinuclear protoplasmic masses also elongate, corresponding to the nuclei, so that the cells change from stellate to spindle- shaped," without, however, losing the protoplasmic bridges which unite the entire mass into a syncytium; in fact, the bridges be- come larger in places. During the earliest stages the muscles 460 HUMAN EMBRYOLOGY. increase in size by additions from and transformation of the mesenchymal cells. Mitosis is abundant in the mesenchjnue and rare in the developing muscle. During the second period, in which the muscles are differentiating by the rapid formation of myo- fibrillfe and increase in the size of the elongated nuclei, there is little formation of new muscle tissue. Still later there is a second period of muscle growth in the circular layer of the oesophagus, which continues until the adult form is reached. "This increase is apparently due to two factors, — first, differentiation of the em- bryonal connective tissue both at the margins of the already 9P- , - Fig. 326. — (After MeGil!: Internat. Montassch. f. Anat. u. Phys., 24, Tat. vii, Fig. S.) From the cesophagus of a 13-mm. pig, showing marlced elongation of mesenchymal cells and nuclei, with the format tion of coarse myoJSbrillse (fc), granular spindle (fls), granules fused into a homogeneous spindle (ha), and their end-to-end union (hs), mesenchymal cells (m), intimately connected with muscle-cells. formed muscle layer and also apparently of that lying between the muscle elements, and second, by the mitotic division of the already formed nuiscle nuclei." "Immediately following the process of elongation of the mesenchyme, or in later stages of the embryonal connective tissue, the myofibrillse are formed in the protoplasm of the elongating cells or nuclear masses" (Fig. 326). "There are two kinds of myofibrillag, coarse and fine." "The coarse myofibrillse are the first to develop." "The protoplasm of the stellate mesenchyme appears to contain a granular reticulum," and, as the cells elongate to form muscle, the granules increase in number and take a more intense stain than the ordinary protoplasm. "As DEVELOPMENT OF THE MUSCULAR SYSTEM. 461 tlie elongation continues, the granular fibrils of the protoplasmic reticulum are stretched out more and more, and finally appear as more or less distinct longitudinal striations" (Fig. 327). "The protoplasm of the cell body appears to be made up largely of Mm. flex. pro. -Mm. fle.x. c. ul. " ^^ Mm. fem. post. ^ M add M. supraspin , N. C. V. ..- Plexus brachialis Mm. cr. post, pro Mm. cr. post. sup. M. quadra, fem. M. obi. int. Fig. 347. — (After Bardeen and Lewis.) Median view of body wall and limbs of an ll-mm. embryo. third rib, but the two muscles still form a single columnar mass attached to the humerus, to the coracoid process, and to the cla\'icular rudiment (Fig. 347). As the mass differentiates it flat- tens out and extends caudoventrally to the region of the distal ■ends of the upper ribs. In a 14-mm. embryo the caudal end of the muscle has extended near to the tip of the fifth rib and the "muscle has begun to assume more the adult form, with fibres arising from the upper five ribs and sternal anlage as well as from the clavicle. At this stage the proximal portion of the muscle 488 HUMAN EMBEYOLOGY. has split into the major and minor portions, the one attached by tendon to the humerus and the other to the coracoid process. Both muscles fuse together near the costal attachments. In a 16-mm. embryo the two muscles are quite distinct, the pectoralis major now extending to the sixth rib and showing a distinct cleavage between the costal and clavicular portions (Figs. 339 and 349). The pectoralis minor has now its distinct attachment to the second, third, and fourth ribs. The pectoralis major early splits into a series of overlapping bundles, and during the migration of the muscle the superficial fibres of each bundle move farther caudally than the deeper ones, giving the overlapping condition found in the adult. The tendon of insertion at first consists of a single sheet, but later from its distal end the second deeper sheet gradually spreads proximally and in an embryo of 40 mm. exceeds the superficial or ventral one in breadth. The M. pectoralis major is carried towards the midventral line with the growth of the ribs and sternal rudiments. The Muscles of the Arm. — The remaining or intrinsic muscles of the arm develop in situ and, as they differentiate from the arm blastema, have approximately the same position that they later occupy in the adult. In an embryo of 9 mm. in length the skeletal core has already begun to differentiate as a thick rod in the middle of the arm bud ending distally in the hand plate (Fig. 336). On all sides, however, and at the distal end this skeletal core gradually merges into the surrounding blastemal sheath in which the muscles later appear, although the positions of only the pectoral and latissimus premuscle masses are recog- nizable at this stage. As the miiscles differentiate their tendons likewise form in situ, and the muscles are thus from the first in connection with the skeletal structures by a condensed mesenchy- mal blastema out of which the tendons later develop. The Muscles of the Shoulder and Arm. — The Mm. deltoideus, teres minor, supra- and infraspinatus arise from a common pre- muscle mass continuous with the pectoral mass and the common arm sheath (Fig. 336). In an 11-mm. embryo the M. deltoideus has partially split-off from the mass towards its origin from the acromion and clavicula (Fig. 337). In embryos of 14 to 16 mm. in length it has much the adult form (Fig. 349), with usually a distinct slip arising from the fascia over the M. infraspinatus. In a 20-mm. embryo it has practically the adult form and attach- ments (Figs. 338, 339, 346, 348). The development of the acromion from the cephalic border of the scapula separates in part in an 11-mm. embryo the M. supraspinatus from the M. infraspinatus. The M. supraspinatus lies at first on the medial surface of the scapula (Fig. 347). Later it comes to lie along the cephalic border, as in 16-mm. and 20-mm. embryos (Fig. 350, 351, 346), and only DEVELOPMENT OF THE MUSCULAR SYSTEM. 489 in later stages, with the growth of the cephalic border, does the muscle acquire its position on the lateral surface of the' scapula. The Mm. infraspinatus and teres minor are from the first very closely associated and cover in an 11-mm. embryo only a portion of the lateral surface of the scapula (Fig. 337). In a 14-mm. embryo the muscle is quite distinct from the M. deltoideus, but does not cover the whole of the fossa infraspinata even in a 16-mm. or 20-mm. embryo (Figs. 349, 338, 346). M. deltoideus M. subscap^^^^ M. biceps ^ M. pect. maj. . M. brachialis Mm. fle-K. carp. rad. and pronator. M. interos. - M. pol.l.- M. fle-^. dig. s.-f; M. supraspin. M. trapezius M. quadri. fern. * Mm. adduct. , \ M. pop. , , I 1 r M. fle.x. hal. 1. \ M. flex. dig. 1. uad. lumb. Fig. 348. — (After Bardeen and Lewis.) Median view of body wall and limbs of a 20-mm. embryo. The M. subscapularis is more or less isolated from the other muscles from its first appearance, and occupies in an 11-mm. embryo only a small portion of the median surface of the scapula, and not until the embryo is more than 20 mm. in length does it occupy the entire medial surface of the scapula (Figs. 350, 351). The M. triceps brachii arises along the posterior and lateral surfaces of the humerus extending from the scapula to the ulna, and even in an 11-mm. embryo indications of the three heads are present, while in a 16-mm. embryo thev are very distinct (Figs. 337, 338, 346, 349, 350, 351). 490 HUMAN EMBRYOLOGY. The Mm. biceps bracbii, coracobracbialis, and bracbialis are intimately fused together in very early stages and probably arise from a common premuscle mass. The places of origin of the two heads of the biceps at this early stage are close together, and only by the later growth of the scapula do they become separated. The three muscles are to be recognized in embryos of from 1-1 to M. trap, M. infraspin. M. delt. ^ i\I. pect. maj. -v M bracliialis "^. ^ \ M. triceps \ M. brachiorad. n M. ext. carp. rad. , ^ ^ \ ; M. ext. dig. com., \ '^ ,^ M.abd.pol.l.*.text.pol.l^ \^ - ^a '^ * \ ^ ^ M. ext. car. ul. M. flex. carp, ul il. lect.abd. II M. obi. e.tt. CoBta IX M. lat. dorai Fig. 349. — (After Lewis.) Lateral view of arm of a IG-mm. embryo. 16 mm. in length, and the long tendon of the caput longum is to be recognized in an embryo of 14 mm. The distal end of the common muscle mass differentiates later than the proximal (Figs. 338, 339, 346, 347, 348, 349, 350, 351). The Extensor Muscles of the Forearm. — The extensors differ- entiate somewhat earlier than the flexors (Figs. 337, 338, 339, 346, 349). The common extensor premuscle mass on the laterocephalic side of the forearm first splits, in an embryo of about 11 mm. in length, into three groups of muscles, of which the largest and DEVELOPMENT OF THE MUSCULAR SYSTEM. 491 most superficial extends from the lateral condyle to the four ulnar digits. It is a thin layer spreading over the ulnar two-thirds of the forearm and distally joining the undifferentiated blastema of the digits. On its radial side proximally it is intimately fused M. supraspin. / M.subscap. J M. biceps & coracobrach. / M. brachialis M. brachiorad. \ ^ M. \ triceps ' JVM. M.pro. teres teres V^ maj. M.flex! j \ M.fle... ^ M.lat. cor. rad. \ \ dig. sub. / dorsi 1 M. pol.long. / M. flex. carp. ul. M. lurab. FiQ. 350. — (After Lewis.) Median view of arm of a IG-inin. embryo. M. supraspin. M. subyi M. flex. dig. sub. M. flex. carp. ul. M. triceps Epicond.med. ' M. teres maj. M. lat. dorsi Fig. 351. — (After Lewis.) Median view of arm of a 20-min. embryo deeply dissected. with the second or radial group. From the superficial extensor mass later differentiate the Mm. extensor digitorum communis, extensor carpi ulnaris, and the extensor digiti quinti proprius. (According to Grafenberg (1905), the ext. dig. V. prop, arises in common with the deep extensor mass.) The separation of the muscles of the superficial extensor mass begins in the carpal region and extends proximally (Grafenberg) in the later stages 492 HUMAN EMBRYOLOGY. as the tendons differentiate from the blastema of the digits. The radial group extends from the epicondylus lateralis and adjoining portion of the humerus distallj^ along the radial and adjoining extensor or dorsal surface of the forearm. It appears to arise in situ along the radial surface, and not to have wandered there from the extensor surface as claimed by Grafenberg (1905). The mass early divides into two parts at the distal end, one, the brachioradialis, fusing with the distal end of the radius, and the other, the extensor carpi radialis longus and brevis, passing be- neath the deep extensor mass to fuse with the blastema at the proximal ends of the second and third digits. The deep extensor mass lies beneath the radial portion of the superficial, becoming itself superficial over the distal radial surface of the radius and carpus and fusing with the blastema of the first and second digits. At a later stage this mass divides into two groups, a radial one for the Mm. abductor poUicis longus and extensor poUicis brevis, and probably the supinator, and an ulnar division for the extensor poUicis longus and extensor indicis proprius. Not imtil the embryo is about 20 mm. in length does the com- plete isolation process of the various extensor muscles reach an end. The extensor digiti Y. prop, is not split off until later. The Flexor Muscles of the Forearm. — The development of the flexors is more difficult to follow than the extensors, and, ow- ing to the concave volar surface of the forearm and carpus, the flexor muscle masses extend much farther distally than do the extensors (Figs. 347, 348, 350, 351). In an 11-mm. embryo, how- ever, one can distinguish a small superficial layer and a voluminous deep layer. The superficial layer lies more on the radial side of the volar surface and already shows indications of a radial mass extending from the epicondylus internus to the blastema at the distal end of the radius; later this mass splits into the Mm. flexor carpi radialis and the pronator teres. The latter extends farther distally on the radius in early stages, but, as the distal part of the radius grows faster than the proximal, the distal attachment of the pronator comes to lie farther and farther from the distal end of the radius. The remaining portion of the superficial layer develops into the M. palmaris longus. It is in- timately fused with the proximal part of the radial mass, but distally it extends on to the volar surface of the carpus, but with the elongation of the skeleton of the forearm the muscular belly comes to lie more and more over the proximal portion of the forearm. The deep flexor muscle mass is much more extensive and thicker than the superficial, extending from the epicondylus in- ternus over the entire volar surface of the forearm and carpus into the blastema of the digits. Even in an 11-mm. embryo the DEVELOPMENT OF THE MUSCULAR SYSTEM. 493 ulnar side already shows the beginning of the splitting off of the M. flexor carpi ulnaris which reaches to the blastema of the os pisiforme (Fig. 347). As the M. flexor carpi ulnaris increases in size it spreads over the ulnar side of the deep flexor mass and in an embryo 16 mm. in length is quite a distinct muscle. The remainder of the deep flexor mass, even in an 11-mm. embryo, shows indication of cleavage into a superficial flexor digitorum sublimis and a deeper flexor digitorum profundus. The mass extends from the epicondylus internus and volar surface of the forearm distally over the carpus without indications at this stage of a longitudinal splitting over the carpal region. The thick portion of the muscle in the carpal region has been designated as a separate M. flexor digitorum brevis by G-rafenberg, but it appears to represent merely the distal part of the mass which later recedes to the volar surface of the forearm as the latter increases in length. The tendons of the two layers of the deep flexor mass gradually differentiate from the blastema of the digits and develop in situ as the digits increase in length. In an embryo 20 mm. in length the characteristic adult arrangement of the tendons is attained, although the recession of the muscle bellies on to the forearm is not complete nor the longitudinal splitting which progresses in a distal proximal direction. At this stage the M. flexor poUicis longus is easily recognized, although it is not completely split off from the M. flexor digitorum profundus. The Mm. lumbricales appear to differentiate in situ from the distal portion of the deep flexor mass, and I have first been able to recognize them in an embryo 16 mm. in length, but they probably appear slightly before this stage. The origin of the M. pronator quadratus is uncertain, and I am not able to determine whether it splits otf from the deep flexor mass or arises independently. It is easily recognized in an embryo of 16 mm. According to Grafenberg, it has a greater proximal extent in early stages than in the adult. The Intrinsic Muscles of the Hand. — The Mm. interossei ap- pear to arise from a common volar muscle mass. In a 16-mm. embryo the first signs of differentiation appear ; later the common muscle mass splits into the various muscles which migrate from the volar surface towards the dorsal surface of the hand between the metacarpals. The Mm. abductor dig. V. and flexor digiti V. arise from a common muscle blastema, lying more on the ulnar side, and later wander volarwards (Grafenberg). The M. opponens dig. V. dif- ferentiates from the common interossei muscle mass (Grafenberg). The thumb muscles develop from a common muscle mass, ex- cept the M. adductor pollicis, which is always separated from the others by the tendon of the M. flexor pollicis longus. 494 HUMAN EMBRYOLOGY. THE MUSCLES OF THE LEG. The muscles of the leg, like those of the arm, develop from the premiiscle sheath of condensed mesenchyme which surrounds the axial skeletal rudiment. In embryos of about 11 mm. in length this common muscle blastema begins in the proximal part of the leg to differentiate into the muscles. According to Grafenberg (1904), the nerves enter a common muscle mass for each group and later this mass splits into the various muscles of the group. Grriif enberg 's description thus corresponds with that given for the arm. Bardeen (1907), however, finds that the individual muscles differentiate separately from the general muscle blastema, first as small centres about theii- nerves, and from these centres the muscle gradually extends towards origin and insertion. Whether this extension is brought about by differentiation of the mesenchyme into muscle or by active proliferation of cells of the original muscle nucleus is not clear from Bardeen 's account. If we assume, how- ever, that in these early stages the mesenchyme continues to be transformed into muscle about the original centre, Bardfeen's ob- servations come more into accord with the obsei*vations of Grrafen- berg and Lewis (1902). Bardeen 's (1907) account is here adopted as a basis for the following section on the development of the muscles of the leg. The Femoral Group. — The femoral group, comprising the Mm. iliopsoas, iliacus, pectineus, quadriceps femoris, and sar- torius, arises from the muscle blastema on the ventral side of the thigh and into it pass the N. femoralis and its branches. The M. iliopsoas arises from that portion of the femoral blastema which embraces the N. femoralis as it passes into the limb bud. In subsequent development the M. iliacus spreads out over the ilium and the M. ])soas major extends upwards along the roots of the femoral nerve to form its attachment to the vertebral column (Fig. 355), and in close union the two muscles extend distally to be attached to the trochanter minor. The M. psoas minor seems to split off from the M. psoas, but this is uncertain. The origin of the rudiment of the M. pectineus is uncertain, but it probably arises either from the iliopsoas mass or the M. adductor longus. In a 14-mm. embryo it is quite distinct, extend- ing from the pubis to the femur (Fig. 354). A portion of the muscle probably arises in connection with the M. obturator externus. The laidiment of the M. quadriceps femoris first appears in an embryo of about 11 mm. in length as a single mass lying over the anterolateral aspect of the middle of the shaft of the femur (Figs. 337, 347, 352, 353). In a slightly older embryo the mass DEVELOPMENT OF THE MUSCULAR SYSTEM. 495 N.t. U. N. hypog N.ing. N.g. N. ling.l N. cut. lat. N. cut. ant N.sart. \\\\ M. quadri. fem N. saph Mm. add. \ \ , , Mm. fem. post Urachus Mm. cr. post, sup Mm. cr. post. prof. Ft. pi. Cloaca N. bi. and semi. t. prox. port. Isch, N. caud. ^M. obt. int. N. pud.' N. cut. post. M. quadr. fem. Fig. 352. — (After Bardeen.) Median view of premutcle masses of leg of an H-mm. embryo. N. t. 12. /N. hypog. M. quadri. fem. Knee Fpl. N. cut. post, Mm. or. ant. (1. peron. . cap. br. N. caud. . hsemorrh. N. pud. Fig. 353. — (After Bardeen.) Lateral view of the premuscle masses of the leg of an 11-mm. embryo. 496 HUMAN EMBEYOLOGY. begins to show definite differentiation into the four portions, the Mm. rectus femoris, vastus lateralis, vastus intermedins, and vas- tus medialis (Fig. 354). In a 20-mm. embryo the various muscles of this group are all clearlj'^ demarcated and attached to the skeletal apparatus by distinct tendons (Figs. 338, 346, 355). The M. sartorius appears to arise from a distinct rudiment proximal to that of the quadriceps mass (Fig. 353), and differen- tiation gradually extends towards the ilium and tibia. The M. sartorius is at first relatively larger than the quadriceps group, but by the fourth month the quadriceps have overtaken it and at birth it is only a little larger relatively than in the adult (Figs. 354, 355, 358, 357, 339, 348). M. trans, abd. M. obi. abd. int. 1 M. obi. abd. ext. .hypog. N. ing. N.g. N. 1. ing. ^ -Sto-^ Proc. vag, N.il M. il. cost. Lig. ing. M. quadr. lumb . M.il. PS. N. cut, lat. N. cut. ant. M. pect. N. cut. med. N. tem Mjar. M. vast. med. m. tens. fasc. lat M.rect. femf M. vast. lat. Fig. 354. — (After Bardeen.) Lateral view of the muselea of the thigh of a 14-min. embryo. The Obturator Group. — The obturator group probably arises from a common muscle mass hing along the anteromedial portion of the femur (Figs. 347, 352). Cleavage is first apparent, in an 11-mm. embrj'o, in the proximal section of the region, into masses, one for the obturator portion of the M. adductor magnus and possibly also the M. obturator externus, the other for the Mm. adductor longus and brevis and the gracilis. In an embryo of 14 mm. the individual muscles may be clearly recognized, although the tendons are not as yet well differentiated (Figs. 356, 359). But by the time an embryo is 20 mm. in length the tendons of origin and insertion are well differentiated to the skeletal attach- ments (Figs. 348, 357, 358, 359, 360) and the obturator and sciatic portions of the M. adductor magnus have fused. The Gluteal Muscles. — According to Grafenberg, all the hip musculature, including the Mm. glutseus maximus, medius, and minimus, tensor fasciae latse, piriformis, obturator internus, gemellus superior and inferior, and quadratus femoris, arise from a cone- shaped mass found at the distal end of the pelvis during the fifth DEVELOPMEiNT OP THE MUSCULAR SYSTEM. 497 week. Bardeen derives these muscles from four rudiments which first appear about the ends of their nerves. The superior gluteal group consists of the Mm. glutseus medius and minimus, piriformis, and tensor fascite latse, which are intimately united in embryos 11 mm. in length (Figs. 337, 353). In a 14-mm. embryo the M. tensor fasciae lata has split off from the lateral edge of the two gluteals. According to Grafenberg, the M. tensor fascia latse is at first inserted into M. obi. abd. ext •> Costa 6 M. obi. abd. int. M. trans, abd.' N.il. N. g. and ing. N. 1. ing. M. tens. fasc. la- M. gl. max. N. sart. N. cut. lat. N. cut. ant. M. rect. fem. M. pect. proc. vag. M. add. 1. . cut. med. M.gt. M. add. magn. M. semim. M. semit. Fig. 355. — (After Bardeen.) Lateral view of the muscles of the thigh of a 20-mm. embryo. the rudiment of the trochanter major, but after splitting off from the gluteal it migrates laterally and loses its attach- ments to the trochanter (Figs. 338, 346). At this stage the M. piriformis is still closely fused with the gluteals, which lie over the acetabulum, extending from the femoral margin of the ilium to the rudiment of the trochanter major. The Mm. glutseus medius and minimus gradually extend over the surface of the ala oss. ilium. Grafenberg finds that the M. piriformis is from the first attached to the sacrum, but, according to Bardeen, it is at first separate and only later extends to its sacral attachment. Vol. I.— 32 498 HUMAN EMBRYOLOGY. The rudiment of the M. glutseus maximus is separate from that of the other gluteal muscles (Bardeen) and slightly fused with the rudiment of the short head of the biceps. At first it onljr slightly overlaps the M. glutseus medius in the trochanteric region, but it gradually extends over this muscle to the ilium, sacrum, and coccyx, and becomes separated into two portions, one being inserted into the femur and the other into the fascisB latte. In the embryo each portion has a separate nerve. The M. quadratus femoris (Figs. 347, 352, 359, 360) seems to arise early from a distinct rudiment lying between the anlage of the tuber ischiadieum and the trochanter major, but close to the rudiment for the Mm. obtiarator internus and gemelli on the ischial side of the hip-joint (Bardeen). The Mm. obturator internus N. 1. 1 Iscli. Fill. ./J>^iJ^J I N cut. post. N. TQe about the eyeball, and medially with the precartilag- DEVELOPMENT OP THE MUSCULAR SYSTEM. 507 N. VI \. IV Os. temp Os. occ. My. Ill •N.III Mm. oculi M. ser. ant. M. trap. / ,' \ Mm. mauilibularis Clavicula' / Mm. facialis M. stemomastoid Fin. 308. — Embryo 9 mm. in length. From a model of the premu.scle masses of the head and neck and antei'ior myotomes. N. VI s..;.n—-ff--^. N. phrenicus ; M. diaphra^ma \ \ Mm. linguie. \ N. XII Mm. infrahyoideus Fig. 369. — Embryo 9 mm. in length. Premuscle masses of the eye, tongue, infrahyoid, and diaphragm regions, from model. inoTis tissue about the optic nerve. The M. rectus lateralis has extended farther out along the path of the N. abducens and has as yet no skeletal attachment. The N. ophthalmicus passes above these muscle masses and nearly at right angles to them. 508 HOrAX EMBRYOLOGY. As differentiation progresses tlie muscle mass of the N. ocu- lomotorius gradually extends around the optic nerve and splits into the various muscles supplied by this nerve. In a 14-mm. embryo all the orbital muscles are to be distinguished and have nearly the adult relations to the bulbus oculi. Tlie M. obliquus inferior, however, does not completely separate off from the M. rectus inferior imtil a later stage. The Muscles of the Mandibular Arch. — The mesoderm of the mandibular arch gives rise to the muscles of mastication, includ- ing the Mm. temporalis, masseter, pterj-goideus externus and in- FiG. 370.-Embryo 11 mm. in length. Diagram of muscles of the head and neck, from graphic reconstruc- tion and model. tennis, and probably to tlie M. mylohyoideus. The Mm. tensor tympani and tensor veli palatini are also derived from this arch. In a 7-mm. embryo the mandibular arch is filled with a uniformly closely packed mesenchyme, with only the sUghtest trace of con- densation about the peripheral end of the N. mandibularis (Fig. .j(J7). In a 9-mm. embryo, however, this condensation is clearly to be recognized (Fig. 368). This premuscle mass, in which the N. mandibularis ends, lies at about the middle of the arch. It is a siinple mass, without indications of splitting and not sharply outlined from the mesenchyme which fills the arch. In an 11-mm. embryo this egg-shaped premuscle mass has increased in size, but DEVELOPMENT OF THE MUSCULAR SYSTEM. 609 still shows no indications of splitting into tlie various muscles (Fig. 370). The differentiation probably takes place in a manner very similar to that described by Eeuter in the pig. The pre- muscle mass is from the very beginning closely associated with the condensed mesenchyme for the mandible, and with the dif- ferentiation of the proximal end of the mandible the premuscle mass is partially split into a Y-shaped mass, the handle repre- senting the M. temporalis, the outer limb corresponding to the M. masseter, and the inner deeper limb, separated from the outer by the proximal end of the mandible, the mass for the Mm. pterygoideus externus and internus. In a 14-mm. embryo this process of differentiation has progressed still farther, and the Mm. pterj'goideus externus and internus are partially separated by Meckel's cartilage. The processus coronoideus only partially separates the masseter from the pterygoid mass at this stage. With the continued differentiation of the membranous mandible and adjoining portions of the skull to which the muscles of mas- tication are attached comes the gradual differentiation of these muscles. The muscle:; are at no time attached to Meckel's carti- lage, but are always in relation with the membranous mandible. In a 20-mm. embryo the various muscles of mastication are easily to be recognized, but have only a slight resemblance to the adult form. The M. temporalis is very small in proportion to the size of the head and gradually extends over a much greater proportional area during later stages. The M. masseter is at first attached only to the medial and lower surfaces of the arcus zygomaticus. The M. mylohyoideus apparently differentiates more rapidly than the other muscles of mastication, and is to be recognized by its nerve supply in an 11-mm. embryo, and in a 14-mm. embryo it has much the adult relations. I have been unable to determine its origin from the common muscle mass. In a 14-mm. embryo the Mm. tensor tympani and tensor veli palatini are to be recognized and are connected with the pterygoid mass from which they probably arise. The M. tensor tympani has already gained its insertion to the malleus. The later de- velopment of these muscles is bound up with the development of the base of the skull, the tuba auditiva, and the soft palate. The Facial Muscles {The Muscles of the Hyoid Arch.)— The facial muscle group includes all the muscles supplied by the N. facialis. The subcutaneous muscles of expression and certain muscles about the facial openings concerned with the vegetative functions belong to this group. Futamura (1906) has given the most elaborate recent account of their origin and development, and the following is based upon his work. 510 HUMAN E.MBRYOLOGY. This entire facial musculature arises from the closely packed mesenchyme which fills the second branchial or hyoid arch. In a 9-mm. embryo the main stem of the jST. facialis, which is very simple at this stage, passes into the hyoid arch to end brush-like in a single mass of premuscle blastema that occupies the ventro- lateral portion of the arch (Figs. 367, 368, 371). The condition is very similar to that described by Futamura as existing in a 10-mm. embryo. It is from this premuscle mass that the entire musculature supplied by the N. facialis arises (Futamura). This ^ ro J"' 371— (After Futamura, Anat. Hefte, Bd. 30, Fig. 27, on p. 440.) Human embryo, 27-30 days XSO dm. bagittal projection from frontal sections, a, eye; G.v, ganglion trigemini; ti ttl (in ramus primus secundus, and tertius N. trigemini; b. junction of the r. ii N. trigemini with N. petr'osus' superf maj.; (, N hngualis; C.t., chorda tympani; N.p.s. ma?., N. petrosus superficialis major; O vii gangHon gemcuh; G.viii, ganglion acusticum; P, the rudiment of platysma; G.ix, ganglion glossopharyngei ; lab., labyrmth; JV. (., N. tympanicus; e., epibranchialis; h., hyomandibular cleft. premuscle mass increases in size (Fig. 370), and in an embryo 13.7 mm. in length has begun to spread out not only ventralwards but also dorsalwards and aborally (Fig. 372). It also extends towards the shoulder to form the platysma. The medial side of the mass becomes thickened to form the rudiment of the Mm. stylohyoideus, digastricus, and stapedius. It extends from the root_ of the tongue along the posterolateral side of Eeiehert's cartilage towards the ear capsule. _ As the muscle rudiment spreads out the N. facialis becomes divided into several branches which follow the wanderings of the muscle tissue. Thus, in a 13.8-mm. embryo the nerve stem distal DEVELOPMENT OF THE MUSCULAR SYSTEiL 511 to the chorda tympani divides into three branches, a thin one directed medially into the rudiment of the Mm. digastricus and stylohyoideus, a second large branch, the N. auricularis posterior, and a third large branch which soon divides into the E. temporalis, maxillaris, and cervicofacialis. In a 15.5-mm. embryo the platysma rudiment has extended orally over the hyoid arch and caudally to the region of the sternum and shoulder-girdle. It has also pushed medially and begins to unite with its fellow of the opposite side (Fig. 373, Fio 372 —(After Futamura, Anat. Hefte, Bd. 30, Fig. 2, on p. 441.) Human embryo of 31-34 days. Sagittal projection from sagittal series. X38 dia. Explanation as in figure 371. N.p.s. mw.. N. petrosus superficialia minor. 374). The spreading of the muscle headwards takes place along two paths, separated by the anlage of the outer ear. The occipital portion gives rise to the Mm. auricularis posterior, transversus nuchse, transversus and obliquus auricula?, but at this stage still forming a continuous membrane. The facial portion of the platysma splits at the upper part of the neck into two layers, a superficial, lightly staining one and a deeper, more intensely staining one. A capillary network often separates the two layers. Futamura designates the super- ficial layer as the platysma faciei and the deep layer as the sphincter colli. . , The platysma faciei gradually extends over the lower ,iaw and )12 HUMAN EMBEYOLOGY. cheek to the forehead, eye, and temporal region, while the most anterior part goes to the lower lip. At the angle of the mouth the two layers are very difficult to separate from one another. In a 7-wk. embryo (Figs. 375, 376) the platysma faciei over the side of the head and above the ear unites with the platysma occipitale, which has also in the meantime extended cranialwards and broadened out to touch its fellow on the opposite side. The platysma occipitale has, however, now lost its connection with the platj^sma colli. o.OC. Fig. 373. — (After Futamura, Anat. Hefte, Bd. 30, Fig. 4, A, on p. 444.) Human embryo of six weeks. Sagittal projection from frontal series. X20 dia. Deep layer. o.oc., orbicularis oculi; o.n., orbicularis nasi; q.Ls.p., the anlage of the M. quadratus labii superioris proprius; a, the parts of the sphincter colli going to the lower lateral side of the eye; o.or., orbicularis oris; o.a., orbicularis auriculae; N.a.v., nervus auricularis posterior; b' and h", branches of the facial going to the posterior and anterior surfaces of the ear; R.i.i,, ramus temporofacialis; R.c.f., ramus cervicofacialis; s.c, sphincter colli. The sphincter colli, the deep layer, which underlies the platysma faciei, forms at first by the sixth week a sheet over the face, with sphincter-like differentiations about the mouth, nose, eye, and ear, to form the primitive Mm. orbicularis oris, orbicularis narium, orbicularis oculi, and orbicularis auriculae (Fig. 373). The differentiation of the sphincter colli begins earlier than the super- ficial layer and is far advanced in embryos 8-9 weeks old(Fig. 378). The lateral portion of the sphincter colli over the cheek degener- DEVELOPMENT OF THE MUSCULAR SYSTEM. 513 ates, as do also the primitive Mm. orbicularis oculi and narium. That part of the sphincter colli between the M. orbicularis oculi and angle of the mouth becomes attached to the maxilla to form the rudiment of the M. quadratus labii superioris. Lateral to this the M. zygomaticus appears to develop from the sphincter coUi. The M. orbicularis oris, which comes from this layer, continues to increase in size and, connected with the mouth but having attach- ment to the skeleton, are differentiated from the sphincter colli Fig. 374, — (After Futamura, Anat. Hefte, Bd. 30, Fig. 4, B, on p. 445.) Superficial layer of the platysma musculature of the same embryo as in Fig. 373. p.o., platysma occipitale; g, oral fossa; -p.f., platysma faciei; p.c. platysma colli. the Mm. caninus, incisivus labii superioris and inferioris, and nasalis. The latter replaces partly the M. sphincter narium. The M. triangularis is associated with the M. caninus. The M. risorius is derived later (thirteenth to seventeenth week) from the M. triangularis. The M. buccinator is derived from that portion of the sphincter colli lying between the upper and lower lips, and as the mouth decreases in size the muscle comes to lie deeper. The superficial layer, the platysma faciei, spreads out over the face and head in a continuous sheet, and in a 7-weeks embryo has united over the ear with the platysma occipitale to form the M. fronto-auriculo-occipitalis (Fig. 375). As the sphincters of Vol. I.— 33 514 HUMAN EMBRYOLOGY. the deep layer degenerate, there are gradually f ormed_ out of the superficial laver a new orbicularis oculi and orbicularis auriculae (Figs. 375, 377). The medial and lateral part of the M. quadratus labii superioris is derived from this superficial layer and also the Mm. mentalis and quadratus labii inferioris (Fig. 377). From the M. fronto-auriculo-occipitalis are derived the Mm. auricularis su- perior and anterior, and, through degeneration of the medial por- Fig. 375. — (After Futamura, Anat. Hefte, Bd. 30, Fig. 5, A, on p. 449.) Superficial layer of the same embryo as in Fig 376. F.a.o., M. fronto-auriculo-occipitalis; o.oc.o., M. orbicularis oculi (from the pla- tysma colli); K.;., R. temporalis; LLs., M. levator labii superioris proprius (alseque nasi); 0.or., orbicularis oris; R. mux., R. maxillaris; R. mar., R. marginalis; R. col., R. colli; R.c.f., R. cervicofaciahs; b" and b', the branches of the facial nerve going to the anterior and posterior surfaces of the ear; o.a., orbicularis auricula;; R.a.p., R. auricularis posterior. tion of the M. fronto-auriculo-occipitalis, the Mm. frontalis and occipitalis become widely separated but joined by the galea aponeurotica which probably represents the degenerated portion of the muscle. From the platysma occipitalis not only arises the M. occipitalis but also the Mm. auricularis posterior and transversus nuchse through degeneration of the intermediate parts. It has already been noted, that on the medial side of the platysma arise the rudiments of the Mm. digastricus and stylo- DEVELOPMENT OF THE MUSCULAR SYSTEM. 515 hyoideus. In the sixtli week this mass is already well developed, extending from the posterior side of the otic capsule in a concave bow towards the angle of the lower jaw. At this stage it is in intimate relation with Eeichert's cartilage. With the growth and lengthening of Eeichert's cartilage comes the complete separation of the digastric rudiment from the platysma. In the early stages this anlage is supplied entirely by the N. facialis, the N. mylo- can. Fig 376 —(After Futamura, Anat. Hefte, Bd. 30, Fig. 5, B, on p. 450.) Human embryo of 7 weeks. Sagittal projection from frontal sections. X 12 dia. Deep layer, o.c, M. orbicularis oculi (from sphmcter colli)- ml M maxillolabiales; o.or., M. orbicularis oris; 6i«;., M. buccmator; con., M. camnus; zi/., M. zygomaticiis; s.c, sphincter colli; R.zv., R. zygomaticus; B. 6uc., R. buccinatorius; N .a.-p., N. auriculans posterior; R.t., R. temporalis; R.c.j., R. oervioofacialis; R. max., R. maxiUaris. hyoideus entering only the M. mylohyoideus. At first the M. digastricus consists of only a single belly. From this rudiment splits off gradually the M. stylohyoideus. Parallel with this process occurs a constriction and degeneration in the middle of the digastricus, dividing it into two bellies. During the eighth to ninth week this process is quite complete. The innervation of the anterior belly of the digastric by the N. mylohyoideus is first to be recognized in a 7-W(5eks embryo when the tendon begins to form. 516 HUMAN EMBRYOLOGY. The M. stapedius appears to come from the proximal end of the rudiment of the M. digastricus. It gradually becomes separated and enclosed within the otic cartilages. In a 14-mm. embryo it is quite separate from the digastric and lies in close relation to the stem of the N. facialis. The Mm. levator veli palatini and uvuLt arise from the facial mass. Their development and complete separation from the rest of the facial muscles are bound up with the development of the Mill'!:!'; iiii'i|' " 'II 'l " 1 'i' ;:;,,! ill iJiiihS,!': Mil. R.man. R.max. Fig. 377. — (After Futamura, Anat. Hefte, Bd. 30, Fig. 8, A, on p. 456.) Human fetus of 8-9 weeks. Sagittal projection from sagittal series. X 14 dia. Superficial layer of the facial musculature, f.a.u., M. fronto-auriculo-occipitalis; o.c, M. orbicularis oculi; l.l.s., M. levator labii superioris alseque nasi; q.Ls.p., M. quadratus labii superioris proprius; n., M. nasalis; i.1.8., M. incisivus labii superioris; o.ot., M. orbicularis oris; c.z.q., caput zygomaticum M. quadrati labii superioris: zy., M. zygomaticus; i.l.i., M. incisivus labii inferioris; t., M. triangulari.^i; P.c, platysma colli; a.p., M. auricularis posterior; o.a., orbicularis auriculsp; N.a.p., auricularis posterior; R.t., ramus temporalis; R.c, ramus colli; R.mar., ramus marginalis mandibulie; R. max., ramus maxillaris. palate and the tuba auditiva. The M. uvulse is at first a paired muscle, but with the formation of the soft palate the two muscles unite in the midline. The Pharyngeal 21uscles. — A''ery little is known concerning the development of the pharyngeal muscles, but the constrictors of the pharynx, as well as the Mm. stylopharyngeus and palato- glossus, probably arise from the third gill arch. The premuscle tissue of this arch is already recognizable in a 9-mm. embryo and into it goes the ninth nerve. The third and fourth gill arches at this stage have already sunken some distance below the sur- DEVELOPMENT OP THE MUSCULAR SYSTEM. 517 face, and thus the premuscle mass comes to lie in the depths in this region. In an 11-mm. embryo the Mm. stylopharyngeus and con- strictors are closely miited together at the side of the pharynx, the M. stylopharyngeus having its attachment to the precartilag- inous horizontal styloid process near its medial end. With the growth and shifting of the various structures in this region this portion of the styloid gradually assumes a more lateral position, Fig. 378. — (After Futamura, Anat. Hefte, Bd. 30, Fig. 8, B, on p. 457.) Deep layer of the facial musculature of the same embryo as in Fig. 377, with higher magnification, can., M. caninus; n., M. nasalis; i.l.s. M. incisivus labii superioris; buc., M. buccinator; t., triangularis (abgeschnitten); i.l.i., inoisivu$ labii' inferioris; g.i.i., M. quadratus labii inferioris; m., mentalis. and in a 14-mni. embryo the muscle takes a medial direction toward the thyroid precartilage and the constrictors. One can also dis- tinguish at this stage the attachment of the M. constrictor pharyngis medius to the hyoid and the M. constrictor pharyngis inferior to the thyroid cartilages. These two constrictors are still united into a common mass which is rapidly extending over the dorsum of the pharjTix. Concerning the M. constrictor pharyngis superior nothing is known and I am uncertain as to 518 HmiAN EMBRYOLOGY. whether it arises in common with the others. Its relation with the buccinator suggests that it may arise from the facial mass. The M. constrictor pharj^ngis inferior is at first continuous with the M. cricothjTeoideus. In a 20-mm. embryo the pharyngeal muscles are quite distinct and the constrictors have grown round to the mid-dorsal line of the pharjTix. The Intrinsic Muscles of the Larynx. — The muscles of the larynx probably arise from the ventral ends of the third and fourth gill arches, which early fuse to form a mass of closely packed mesenchyme out of which later differentiate the cartilages and muscles. Soulie and Bardier (1907) found it difficult to recognize laryngeal muscles in a 14-mm. embryo. But in a 19-mm. embryo they were able to recognize clearly the Mm. interarytsenoideus, crico-arytaenoideus posterior, cricothyreoideus, and thyreo-crico- arytsenoideus. They were unable to discover any traces of a sphincter laryngis as described by Strazza (1889), but found that in 32- to 40-mm. embryos the M. thyreo-crico-arytjBnoideus is clearly distinguishable from the M. thyreo-arytsenoideus and towards the middle of the fifth month all the muscles are recogniz- able. I have noticed that in a 14-mm. embryo one can distinguish the various muscles, the Mm. arytsenoideus, crico-arytsenoideus posterior and lateralis, thyreo-arytsnoideus, and the crico- thyreoideus. I do not find, as Strazza, the muscles forming a common constrictor of the larynx. I find, with Strazza, that the constrictor of the pharynx is continuous with the M. crico- thjrreoideus. "With the farther differentiation of the cartilages the muscles become more distinct and in a 20-mm. embryo they have much the adult form. The Tongue Musculature. — It is usually assumed that the tongue musculature is derived from the occipital myotomes which appear to be serially related to the N. hypoglossus. There is, however, no direct evidence whatever for this statement, and we are inclined to believe from our studies that the tongue muscula- ture is derived from the mesoderm of the floor of the mouth. In a 7-mm. embryo the mesenchyme in the floor of the mouth is similar to that in the mandibular and hyoid arches from which later the musculature of these two arches develops. In a 9-mm. embryo the floor of the mouth has increased in thickness and the groove between the mandibular and hyoid arches has disap- peared. In the mesenchyme of this thickened floor are two bilateral masses similar in appearance to the jaw and facial masses found in the mandibular and hyoid arches at this same stage. These bilateral toneue premuscle masses extend from the region in which the mandible later develops to the hyoid region, and are here continuous with a medial mass of condensed mesenchyme which extends into the larynx region and also with the infrahyoid DEVELOPMENT OF THE MUSCULAR SYSTEM. 519 premuscle bands (Fig. 369). The caudal end of this latter mass is in turn continuous with the diaphragm premuscle mass. This lingual-infrahyoid-diaphragmatic band is probably a primitive ventral visceral muscle complex and in no way concerned with the myotomic system. The N. hypoglossus enters the caudal end of the tongue premuscle mass. One is unable at this stage to dis- tinguish any of the individual muscles. In an 11-mm. embryo (Figs. 370, 379) each homogeneous premuscle mass has split into two masses, a medial ventral mass for the Mm. geniohyoideus and genioglossus and a dorsolateral M.obl&up Fig. 379. — Embryo 11 mm. in length. Diagram of muscle masses of eye, tongue, and infrahyoid region, from graphic reconstruction and model. mass for the Mm. hyoglossus, styloglossus, and chondroglossus. The medial ventral mass extends from the region of the future symphysis menti to the prehyoid mass and dorsally it expands into the tongue region. The main stem of the N. hypoglossus enters its caudal end and passes longitudinally nearly to the an- terior end. The dorsolateral mass extends from the prehyoid and medial portion of the styloid process into the dorsolateral region of the tongue for a short distance. A branch of the N. hypoglossus enters its ventral surface. The styloid process at this stage has very nearly a horizontal position and extends nearly to the midline. 520 HUMAN EMBRYOLOGY. With the differentiation and development of these muscle masses the tongue gradually becomes raised more and more above the mandibular arch. In a 14-mm. embryo this process is ad- vanced, keeping pace with the differentiation of the mandibular arch in which are now plainly to be recognized Meckel 's cartilage and the partially enclosing membranous mandible. From the membranous mandible arise the radiating M. genioglossuS; extend- ing a considerable distance fan-like into the tongue, and the M. geniohyoideus, extending to the hyoid precartilage. Over the dorsum and dorsolateral region of the tongue extend the M. hyo- glossus and M. styloglossus, from the hyoid precartilage and styloid process respectively, to the tip of the tongue. They lie dorsal and lateral to the radiating genioglossus. Their origins, though distinct, are still close together and parallel. The N. hypo- glossus gives off branches to the M. geniohyoideus, then to the Mm. hyoglossus and styloglossus, and passes through the M. genioglossus to its tip, giving off numerous lateral branches into this muscle. There is at this stage apparently very little inter- lacing of the tongue muscle, nor are we able to recognize either the intrinsic muscles — the Mm. longitudinalis superior and in- ferior, transversus linguffi, and verticalis linguae — or the M. glosso- palatinus. lu a 20-mm. embryo, however, all the muscles are clearly dif- ferentiated and increased in size. The origin of the M. styloglossus has been carried to a more lateral position and enters the tongue at an angle to the M. hyoglossus. The latter has spread out more in its attachment to the hyoid cartilage. The M. glossopalatinus is now recognizable, extending from the laterally placed soft palate to the lateral surface of the tongue. The great development of the intrinsic muscles is most noticeable at this stage, but concern- ing their origin there are no observations. It is uncertain whether they are derived from the hyoid and mandibular tongue muscles or independently from the mesenchymal matrix. At this stage the interlacing of the tongue musculature is quite advanced. LITERATURE. Bardeen : The Development of the Jlnsciilatnre of the Body Wall in the Pig, Including- its Histogenesis and its Relation to the Myotomes and to the Skeletal and Nervous Apparatus, Johns Hopkins Hosp. Rep., vol. ix, 1900. Development and Variation of the Ner^-es and the Musculature of the Inferior Extremity and the Neighboring Regions of the Trunk in Man, Am. Journ. of Anat., vol. vi, 1907. Bardeen and Lewis: Development of the Back, Body Wall, and Limbs in Man, Am. Journ. of Anat., vol. i, 1901. Collin : Becherehes sur le developpement du muscle sphincter de I'iris ehez les oiseaux, Bibliographie Anatomique, t. 12, p. 18.3, 1903. DEVELOPMENT OP THE MUSCULAR SYSTEM. 521 Felix: Teilungserseheinuiig-en an quergestreiften Muskelfasern menschlicher Em- bryonen, Anat. Anz., Bd. 2, 1888. Ueber Waclisthum der quergestreiften Muskulatur naeh Beobachtungen am Mensehen, Zeitschrift f. wiss. Zool., Bd. 48, p. 224, 1888. FiscHEL, A.: Zur Entwicklung der ventralen Rumpf u. Extremitatenmuskulatur bei Vogeln und Saugetieren, Morph. Jahrbuch, Bd. 23, 1895. Futamura: Ueber die Entwicklung der Faeialismuskulatur des Mensehen, mit 27 Textabbildungen, Anat. Hefte, Bd. 30, 1906. GoDLEWSKi : Die Entwickelung des Skelet- und Herzmuskelgewebes der Saugetiere, Arehiv f. mikr. Anat., Bd. 60, 1902. Grapenberg, E. : Die Entwicklung der menschl. Beekenmuskulatur, Anat. Hefte, Bd. 23, 1904. Die Entwicklung der Knochen, Muskeln und Nerven der Hand und der fiir die Bewegungen der Hand bestimmten Muskeln ded Unterarms, mit 19 Abbildungen im Text, Anat. Hefte, Bd. 30, 1905. Harrison, R. G. : An Experimental Study of the Relation of the Nervous System to the Developing Musculature in the Embryo of the Frog, Am. Journ. of Anat., vol. iii, 1904. Heerfordt, C. F. : Studien ueber den Muse. Dilatator Pupillae, etc., Anat. Hefte, Bd. 14, p. 487 to 558, 1900. Heidenhain, M. : Ueber das Vorkommen von Intereellularbriicken zwischen glatten Muskelzellen und Epithelzellen, etc., Anat. Anz., Bd. 8, 1893. Herzog, H. : Ueber die Entwicklung der Binnenmuskulatur des Auges, Arehiv f. mikr. Anat., Bd. 60, p. 517-586, 1902. Ingalls, N. W. : Beschreibung eines menschlichen Embryos von 4.9 mm., Arehiv f. mikr. Anat., Bd. 70, 1907. Kotzenberg, W. : Zur Entwicklung der Ringmuskelschicht an den Bronchien der Saugetiere, Arehiv f. mikr. Anat., Bd. 60, p. 460, 1902. Kazzandek : Beitrag zur Lehre ueber die Entwicklung der Kaurauskeln, Anat. Anz., Bd. 6, p. 224, 1891. Keibel, F. : Studien zur Entwieklungsgeschiehte des Schweines II. Morphologisehe Arbeiten, Bd. 5, 1895. KoELLiKER : Handbueh der Gewebelehre des Mensehen, S. 253, 1889. KOLLMANN : Die Rumpfsegmente menschlicher Embryonen von 13-35 Urwirbeln, Arehiv f. Anat. u. Phys., Anat. Abt., 1891. Kroesing : Ueber die Riickbildung und Entwicklung der quergestreiften Muskel- fasern, Virch. Arch., Bd. 128, 1892. Leonowa, 0. VON : Zur pathologischen Entwicklung des Centralnerven^ystems (ein Fall von Anencephalie kombiniert mit totaler Amyelie), Neurol. Central- blatt, Jahrg. 12, 1893. Lewis, W. H. : The Development of the Ann in Man, Am. Jour, of Anat., vol. i, 1901. Observations on the Pectoralis Major Muscle in Man, Johns Hopkins Hosp. Bull., vol. xii, 1901. Wandering Pigmented Cells Arising from the Epithelium of the Optic Cup, with Observations on the Origin of the M. Sphincter Pupillse in the Chick, Am. Journ. of Anat., vol. ii, 1903. MacCallum, J. B. : On the Histogenesis of the Striated Muscle-flbre and the Growth of the Human Sartorius Muscle, Johns Hopkins Hosp. Bull., 1898. Mall : Development of the Ventral Abdominal Walls in Man, Jour, of Morph., vol. xiv, 1898. McGiLL, Caroline: The Histogenesis of Smooth Muscle in the Alimentary Canal and Respiratory Tract of the Pig, Internat. Monatschrift. Anat. u. Phvs., Bd. 24, 1907. JTarchesini and Ferrari : Untersuchungen ueber die glatten und die gestreiften Muskelfasern, Anat. Anz., Bd. 11, 1895. Meek, A. : Preliminarj- Note on the Post-embrj'onal Historj- of Striped Muscle- 522 HUMAN EMBRYOLOGY. fibres in Mammalia, Anat. Anz. 14, 1898, and 15, 1899. On the Post- embryonal History of Voluntary Muscles in Mammals, Journ. of Anat. and Physiol., London, vol. xxxiii, p. 546-608, 1899. MORPURGO, B. : Leber die postembryonale Entwieklung der quergestreiften Muskeln von weissen Ratten, Anat. Anz., 15, p. 200-206, 1898. Leber die Ver- haltnisse der Kernwucherung zum Langenwachstum an den quergestreiften Muskelfasem der weissen Ratten, Anat. Anz., Bd. 16, p. 88-91, 1899. Neumann, E. : Leber die vermeintliehe Abhangigkeit der Entstehung der Muskeln von den sensiblen Nerven, Arch. f. Entw.-Mech., Bd. 16, p. 642-650, 1903. NussBAUii, M. . Entwieklungsgeschichte des menschlichen Auges, Grafe-Samiseh Archiv, Bd. 2, 1900. Die Entwickelung der Binnenmuskeln des Auges der Wirbeltiere, Arch. f. mikr. Anat., Bd. 58, 1902. Paneth : Die Entwickelung von quergestreiften Muskelfasern aus Sarkoplasten, Sitzungsberichte d. Kais. Akad. d. Wissenschaft, III Abt., Bd. 92, 1885. POPOWSKT, J. : Zur Entwickelungsgeschichte der Dammuskulatur beim Menschen, 2 Taf., Anat. Hefte, Bd. 12, 1899. Reuter : Leber die Entwickelung der Kaumuskulatur beim Schwein, Anat. Hefte, Bd. 7, 1896. Leber die Entwickelung der Augenmuskeln beim Schwein, Anat. Hefte, Bd. 9, 1897. Ruge : Entwicklungsvorgange an der muskulatur d. mensehl. Fusses, Morph. Jahrb., Bd. 4, Suppl., 1878. SciiAFPER : Beitriige zur Histologie und Histogenesis der quergestreiften Mus- kelfasern des Menschen und einiger Wirbeltiere, S. Ber. d. Acad. d. Wiss. Wien, 3 Abt., B. 102, p. 7-148, 1893. SCHOMBURG, H. : Lntersuchungen der Entwickelung der IMuskeln u. Knoehen des mensehl. Pusses, Von der med. Facultat d. Lniversitjit Gottingen gekronte Preisschrift, 1900. Strazza, G. : Zur Lehre ueber die Entwieklung der Kehlkopf muskeln. Mittheil. aus dem Embryologischen Institute der K. K. Lniv. in Wien, Der Zweiten Polge, 3 Heft, 1899. Stohb, Ph. : Entwickelungsgeschichte der menschlichen WoUhaares, Anat. Hefte, Bd. 23, 1902. Soulie, a., et Bardier, E. : Recherches sur le developpement du larynx chez I'homme, Jour. dAnatomie et de la Phys., t. 43, p. 137-240, 1907. SziLi, A. : Zur Anatomie und Entwickelungsgeschichte der hinteren Irisschichten mit besonderer Beriieksichtigung des Musculus sphincter iridis des Menschen, Anat. Anz., Bd. 20, p. 161, 1901. ZiMMERMANN : Leber Kopfhohlenrudimente beim Menschen, Arch, f . mikr. Anat., Bd. 53, 1899. XIII. CCELOM AND DIAPHRAGM. By franklin P. MALL. The small ovum described recently by Bryce and Teacher is hardly two millimetres in diameter, and is covered with a reticular mass of syncytium, true villi with mesodermal cores not being present. Within there is an extremely small embryo anlage embedded in a delicate cellular reticulum. The mesenchymatous tissue shows no signs of cleavage into a parietal and a visceral layer nor has it arranged itself into a denser layer around the wall of the vesicle. The general appearance of the mesoderm is shown in Fig. 10, Chapter IV, and in Fig. 98, Chapter VII. It is there seen that no exoccelom is present. In comparing this speci- men with Peters' ovmn, which is somewhat larger, we must imagine a destruction of some of the mesodermal tissue, either in the centre of the ovum or near the embryo, in order to form the primitive exoccelom. In fact this interpretation can be given to the form of the exoccelom in Peters' ovum, as is shown in Figs. 96 and 97, Chapter VII. The first figure is taken from Peters' plate, and the second is an interpretation of this figure by Professor Grosser, who has compared the section from which the Peters drawiug was made with the drawing itself. It is seen, therefore, that the exoccelom is not present in an ovum two millimetres in diameter, and that it is well formed in an ovum somewhat larger, that is, at the beginning of the third week of pregnancy. All the other young human ova which have been studied by embryologists, possibly iucluding that by Peters and those by Graf Spee, have within them a large cavity lined entirely with a layer of mesoderm. The precocious development of this space, the exoccelom, appears to be peculiar to man and monkeys, for it has not been observed in other mammalian ova which have been studied with great care in a large number of species. Hanging freely within this cavity, but attached to one side of the chorion, there is always found in normal ova a relatively small embryonic mass composed of a closed amnion and an umbilical vesicle joined together by the anlage of the embryo, which in the youngest speci- mens contains only the three primitive blastodermic membranes. In general the diameter of the embryonic mass is but one-fifth of that of the exoccelom, which indicates that the latter began to form 623 524 HULIAN EMBRYOLOGY. by a splitting of the mesoderm when the ovum was very small, unless we assume that the embryonic mass becomes absolutely smaller while the exocoelom is forming. In the Peters ovum the embryonic mass measures about 0.2 mm. in diameter, and that of the exocoelom, which is egg-shaped, averages about one mm. Until the embryo is 2 mm. long the amnion hugs the embryo closely and does not encroach markedly upon the exocoelom. As the embryo grows a little larger some amniotic fluid accumulates around it, which naturally causes the embryonic mass to grow much more rapidly than the embryo. During this time the embryo also grows relatively faster than the exocoelom and therefore the embryo and amnion soon begin to obliterate the exocoelom. At the time the embryo is 4 mm. long the ratio in size between the embryonic mass and the exocoelom is about 5, and somewhat later it is but 3. When the embryo measures 7 mm. and the ovum is 18 mm. in diameter, the ratio is 2, and by the beginning of the eighth week the exocoelom is obliterated entirely. Now the amnion lines the entire chorion with the exception of a small region around the umbilical vesicle which lies immediately below the chorion and is surrounded by a small space, the last remnant of the exocoelom. The exocoelom is filled with an albuminous fluid which is held together by a delicate network of fibrils and gives it a jelly- like consistency. This mass, the magma reticule of the older authors (Velpeau), is also well marked in the exocoelom of monkeys' ova, and is probably more marked in normal ova than is generally believed (Keibel). In case the fibrils of the magma are scanty or beginning to disintegrate, or in case they are greatly increased in number, dense enough partly to obliterate the embryo, the specimen is certainly pathological. Magma which is trans- parent and contains just enough reticulum to hold it together is to be viewed as the normal constituent of the exocoelom. The nature of magma fibrils has not been definitely settled, but they appear to be connected with cells in some way as indicated in the ova described by Bryce and Teacher and Peters. Some embryologists are inclined to consider them the product of coagu- lation. However, they are present in the freshest specimens, and they cannot be stained by Weigert's fibrin stain. I have found reticular magma present in normal ova which were preserved in strong formalin immediately after their abortion. According to Eetzius, the magma reticule contains fibrils of a muco-fibrillar nature which become transparent in dilute acetic acid. Before the circulation of blood is established between the embryo and the chorion, — that is, before the embryo is 2 mm. long,— the nutrition of the embryo and its umbilical vesicle must pass from the chorion through the magma. This may be the CCELOM AND DIAPHRAGM. 525 reason wliy any alteration in the nutrition of the embryo, which affects its normal development, first manifests itself in a change in structure of the magma reticule. As the exocoelom is gradually obliterated the magma is pushed ahead of the advancing amnion and finally forms a delicate membrane of fibrils between the amnion and chorion. BODY CAVITIES. The ccBlom of the embryo arises independently of the exo- ccelom in embryos between 1.5 and 2 mm. long. In specimens less than 1.5 mm. long studied by Peters, Graf 8pee, and Minot, there is no trace of either blood-vessels or body cavity within the embryo, but in Graf Spec's well-known embryo Gle. (1.54 mm. long) traces of the beginning of the heart and its coelom are present. Fig. am. md. 380, which is through the head of this embryo, contains on either side of the ^^ body a few scattered cells between the mesoderm and entoderm. These if^c^^ll -^M^JSliJ— xd. Graf Spee believes to be the endothe- lial anlage of the heart. In this same section there is a split in the meso- derm of the embryo, which continues through other sections and marks the ™' hpo'iTlTllTIO- of thp nprir'arrlifll frplnm Fig. 380.— Last section (No. 24) through Ueginning Ol ine peilCaXUiai CtBiUm. the foregut of Graf Spee-s embryo GZe. It maV be noted that the endothelial (After Graf Spee.) am., amnion; md " . medullary plate ; me., mesoderm ; Kd., anlage Ot the heart and its CCKlom foregut ; a., chorda ; en., entoderm. The ni IT n t • n ^ n space in the mesoderm is the beginning of arise alter the loregUt is well formed, the co^lom, and between it and the ento- — that is, after the head of the embryo not"ho™L\hedrawing^ ''""*' '''''* '' is separated from the yolk-sac. The different sections show that the coelom is of irregular form and size and it is surrounded with numerous dividing cells (Fig. 381). Since no other human embryo has been studied which follows immediately upon Graf Spec's Gle., we must fill out the gap by observations upon other mammalian embryos; these are very complete and have been made with great care. They do not con- tradict one another nor any of the observations made upon the development of the human coelom. For these reasons much con- fidence can be placed upon facts derived from comparative embry- ology upon the early formation of the pericardial coelom and their bearing upon the same in human embryos. Bonnet's careful study of the development of the coelom in the embryo sheep fills out perfectly the gap between Spee's Gle. and older human embryos. In the sheep the pericardial coelom appears as irregular spaces on either side of the head much as in Graf Spee's embryo Gle. The spaces arise on both sides of the 526 HUMAN EMBETOLOGY. embryo, but are so irregular in position and form that their arrangement can not possibly be considered metameric. Next the spaces unite, thus forming in a relatively short time large spaces on either side of the body, which soon unite with each other at the extreme anterior end of the embryo, forming a horseshoe- shaped canal in the mesoderm on the ventral side of the head. Throughout its development the pericardial coelom is closed on its lateral sides and does not communicate with the exocoelom, with the exception of its later and indirect communication through the peritoneal coelom. It may also be noted that in the sheep the whole pericardial coelom is formed hand in hand with the invagina- FiG. 381. — Section No. 81 through Graf Spec's embryo Gle. (After Graf Spee.) p., pericardial coelom, near which on the entodermal side, a few loose cells may be seen which may represent the anlage of the heart. tion of the foregut and before the endothelial anlage of the heart has arranged itself into vascular tubes, much as is the condition found in Graf Spee's embryo Gle. That the pericardial coelom arises very early and independ- ently of the exocoelom is farther proved by the work of Strahl and Carius, who studied its development in the guinea-pig and the rahbit and confirmed fully the earlier observations of His upon the first formation of the pericardial coelom (Parietalhohle). In the rabbit the pericardial coelom ends in two dorsal and two ventral recesses, all four of which connect subsequently with the peritoneal coelom. However, only the dorsal recesses break into the peritoneal coelom in the human embryo, and it is this recess or canal wliich later on encircles the lung and probably forms the main anlage of the pleural coelom. We can now return to the human embryo, and the next stage after Graf Spee's Gle. which bears upon this point is found in a well-preserved normal human embryo 2 mm. long. The embryo CCBLOM AND DIAPHRAGM. 527 came from a self-inflicted mechanical abortion and was soon pre- served in formalin. Before transferring it to. alcohol it measured about 2 mm. in length, but due to its shrinkage while being Fig. 383. Fig. 386. Pigs. 382-386. — Sections through a human embryo 2 mm. long (No. 391). Enlarged 100 diameters. Fig. 382, section 27; Fig. 383, 49; Fig. 384, 58; Fig. 385, 76; and Fig. 386, 95. Ph, pharynx; H, heart; UVf umbilical vesicle; Ao, aorta; Ch, chorda; PC, pericardial coelom; P., peritoneal ccelom; Exoc, exoccelom; R., recess from pericardial ccelom. embedded it produced but 160 transverse sections each being 10 /* thick. The embryo has 7 myotomes. The figures (382-386) show that the pericardial coelom encircles the heart on its ventral side and reaches down into a recess which communicates with the 528 HUMAN EilBRTOLOGY. peritoneal spaces and through these with the exocoelom. The pericardial coelom is entirely closed on all sides. The peritoneal coelom is forming at numerous points in the mesoderm of the trunk of the emhryo, not in any regular fashion ; some of the spaces connect irregularly with one another, others connect with the exocoelom (Fig. 386), as is found to be the case in the sheep. Dandy has studied this embryo with great care, and a more detailed account of it, containing also a reconstruction of the coelom, may be found in his publication. Quite recently Keibel and Elze described an embryo nearly 2 mm. long in which the pericardial coelom is separated from the Fig. 387. — Profile reconstruction of an embryo 2.1 mm. long (No. 12). X 25 times. Am, amnion; OV, optic vesicle; AV., auditory vesicle; UV, umbilical vesicle; iJ,- heart; VOM., omphalomesenteric vein; mr., septum transversum; 0^, third occipital myotome; C^, eighth cervical myotome. exocoelom by a pronounced bridge of tissue. This specimen is of prime importance in the study of the coelom in the human embryo, for through it we can connect the development of the coelom in older and younger stages with one another, as well as with the condition found in the rabbit and in the sheep. Another embryo slightly older shows that the pericardial coelom communicates freely with the peritoneal coelom and this in turn with the exocoelom (Fig. 387). Probably this specimen is not altogether normal, because the brain is not well developed and the spinal cord behind appears to be too wide open. The liver and thyroid are beginning to form and two branchial pockets are well developed from the pharynx. In its caudal shifting the amnion has passed the heart, leaving the ventral body wall cov- ered with ectoderm. The attachment of the umbilical vesicle CCELOM AND DIAPHRAGM. 529 to the body is constricting from all sides, receding before the amnion in front, at 0, and from the allantois behind, 0\ Some- what later, when the constriction is more marked and the amnion V u. \ O.M, Fig. 388. — Section through the head of tiie embryo 2.1 mm. long. X 50 times. Coe, coelom ; Ph, pharynx; L, liver; ST, septum transversum; UV, umbilical vesicle. Fig. 389." — Section through the third occipital myotome of the embryo 2.1 mm. long, .04 mm. nearer the tail than Fig. 388. X 50 times. O3, third occip- ital myotome ; Coe, coelom ; V, vein ; ST, septum transversum ; L., liver; Ph, pharynx; UV, umbilical vesicle. Fig. 390.— Section through the first cervical myotome of the embryo 2.1 mm, long, .23 mm. nearer the tail than Fig. 389. X 50 times. C. Hrst cervical myotome ; Coe, cce- lom ; V. U., umbilical vein ; V. 0. M., omphalomesenteric vein; Umb., umbilical vesicle. Fig. 391. — Section through an embryo 3.5 mm. long (No. 164), X SO times. L, liver; Vent, -ventricle ; SR, sinus reuniens ; Coe, coelom. Fig. 392. — Section" through the embryo 3.5 mm. long, .18 mm. nearer the tail than Fig. 391, X 50 times. Coe, coelom; Int, intestine; VOM, omphalomesenteric vein; VU., umbilical vein; DC ductus Cuvieri. comes to lie upon the umbilical stalk, UV, the points and 0^ T\^iU be the points of communication between the peritoneal coelom of the two sides of the body. Other sections of this embryo are shown in Fig. 229, Chapter XI. Vol. I.— 34 530 HUMAN EMBRYOLOGY. In front the yolk-vein, VOM, connects freely with the capil- laries of the yolk-vesicle, and as it enters the heart it is bounded on its ventral side by a recess, Fig. 388, Coe, which, however, does not communicate directly with the peritoneal coelom or with the exoccelom. Capillary veins also arise in the septum trans- versum, around the liver bud, Fig. 389. More caudal. Fig. 390, VU, a branch of the vein extends to the lateral side of the body, which is no doubt the anlage of the umbilical vein. These veins are more pronounced in an embryo slightly older. Figs. 391 and 392, which shows the ductus Cuvieri, the umbilical vein, the omphalomesenteric vein, and a vein running from the head — • the jugular vein. All these veins unite in the sinus reuniens, which in turn enters the heart. SEPTUM TRANSVERSUM. It is seen from the description just given that the body cavities arise from two primary spaces, as was first well shown by His, and that each of these in turn is bilateral in origin. In addition to these there is the exoccelom as well as the cavities of the myo- tomes which in the human embryo soon disappear ; they communi- cate in part with the rest of the bodj^ cavity (Keibel and Elze). The two pericardial coelom cavities soon unite at the front end of the head, and this space later on encircles the whole ventral sur- face of the heart. Next, two pockets arise from it, extend dor- sally, and unite with the two peritoneal cavities. These diver- ticula barely exist as separate cavities, but blend immediately, as soon as they begin to form, through the peritoneal cavity with the exoccelom. By the end of the fourth week, as represented in Fig. 387, the five primitive cavities form a single coelom, from which at last seven body cavities arise. The primitive and most fun- damental septum by which the common body cavity is broken into compartments was discovered by His, which he described as the septum transversum. Figs. 383 and 384 show that the umbilical vesicle is probably tied to the heart by means of two blood-vessels, which are well embedded in a mass of tissue of the embryo as shown in Fig. 387. Within this mass the blood-vessels go to the heart and into it the liver grows. To the extent in which the pericardial coelom (Parietalhohle) does not communicate with the peritoneal coelom, its floor is formed by a mass of tissue which unites the two sides of the trunk of the body with each other, and also binds the sinus venosus to the foregut. It is this mass of tissue which His terms the primitive diaphragm or the septum transversum. From it the principal part of the permanent diaphragm is formed, and by its extension the primary pericardium and the primary diaphragm are completed. CCELOM AND DIAPHRAGM. 531 In an embryo slightly larger than the one just described, the septum transversum is well defined, as may be seen in Fig. 394. The lungs have just begnin to sprout (Fig. 393), and on the right FW. 1 ^, ^, Av%. _f. Fig. 393. — Section through an embryo 4.3 mm. long (No. 148) . X 25 times. T\, first thoracic myo- tome; C'e, Ci. and Cs, cervical myotomes; *9., stom- ach; B. bronchus; H, heart; T., thyroid; PC, peri- cardial cavity; L, liver; FW ... foranaen of Winslow. Fig. 394. — Section through the embryo 4.3 mm. long, .4 mm. deeper than Fig. 393. X 25 times. T, thoracic myotomes ; /. intestine ; L, liver ; V, ventricle ; BA, bulb of tlie aorta ; Am., amnion; UV, umbilical vein. /. \ ,'vc C^ ^ If vu side there is a diverticulum from the peritoneal coelom, which marks the beginning of the lesser peritoneal cavity. The liver is beginning to ramify within the septum and protrudes dorsally into the embryonic per- / ^^"-~^x itoneal cavity. In another embryo of about the same age the septum transversum is practically complete and in its form and position corresponds exactly with the des- cription of it given by His (Fig. 395). The pericardial, pleural, and peritoneal coe- ioms communicate as freely as they ever will, for there is as yet practically no sign of the formation of secondary septa. A reconstruction of the region of the septum of this embryo, showing the coelom, is given in Figs. 396 and 397. The pericardial cav- ity lies completely on the ventral side of the septum, from which it extends in the form of a TJ-shaped canal around the sep- tum on either side of the lung, stomach, and intestine. The pleural ccslom — that is, the ccelom which surrounds the lung buds ^is much the same in form on both sides of the body, but the peritoneal ccelom shows marked bilateral inequalities due to the changes which have taken place with the shifting of the stomach towards the left side of the body. r -^ Fig. 395. — Section through an embryo 4.5 mm. long (No. 7G). X 25 times. VC, cardinal vein ; A, aorta- ; VOM, omphalomesen terj c vein; FC/, umbilical vein; //.heart. 632 HUMAN EMBRYOLOGY. In Fig. 396, below the letters FW, there is a marked depres- sion, the recessus mesenterico-entericus, which corresponds with the shifting of the stomach to the left. (This is well shown in Fig. 405.) Above FW a sac or invagination passes on the dorsal median side of the lung to form the recessus pneumato-entericus dexter. This space also borders upon the liver and forms a third recess, the recessus hepato-entericus, over the region of the lobus Spighelii. These spaces together form the lesser peritoneal A. Ph. BA. FiQ. 397. Figs. 396 and 397. — Right and left views of a reconstruction of the embryo 4.5 mm. long. X 25 times. A., aorta; i»A., pharynx; 5^., bulbus aortae; Coe, coelom; P, pericardial ccelom; I/, lung; £j, liver; TTB, Wolffian body; S, stomach; fPT, foramen of Winslow; 57, sinus venosus; ST, septum transversum. cavity and were recognized as such by His in 1880. From this time onward it has been described by numerous embryologists, and the most satisfactory study of it in the human embryo is by Broman, whose terminology I have adopted. There is but little to be said about the left pleural recess except that its configuration is that of the lung over which it passes. However, between the lung and the stomach (Fig. 397, L and S) there is a slight depression. There is present at this point in reptiles and birds a marked pocket, or "left lesser peri- toneal cavity," which has also been described in mammalian embryos by Ravn. Eavn's discovery was doubted by a number CCELOM AND DIAPHRAGM. 533 ft " Oe 1 I 1 I ^ R.P.E.D. B.P.E.S. Fig. 398. — Section through the right and the left lesser peritoneal cavities in a human embryo 3 mm. long (No. 239). X 60 times. OE, oesophagus; L, liver; R.P.E.D., recessus pneumato-entericus dexter; R.P.E.S., recessus pneumato-entericus sinister. of embryologists, but recently Broman has found it in all mam- malian embryos examined, including the human. According- to Broman the left bursa appears in human embryos about 3 mm. long, and vanishes immediately, that is, before they are 4 mm. long. The left recessus pneumato-entericus, as Broman calls it, is decidedly smaller than the right, does not encroach upon the left lung as much as the right does upon the right lung; in fortunate trans- verse sections they appear to be symmetrical and equal (Fig. 398). Up to the present stage the septum transversum, which arose in the region of the head, has gradually receded until its dorsal end has fallen to the region of the fifth cervical nerve. This stage is of importance, for it shows how the phrenic nerve enters the septum transversum (Fig. 399, Ph). This section shows that the nerve grows at first between the subclavian and cardinal veins along the lateral wall of the pleu- ral coelom towards the ductus Cu- vieri. Subsequently, when the pericardial coelom is separated from the pleural by the growth of the pleuropericardial membrane from the wall of the ductus Cu- vieri to the root of the lung, it leiaves the nerve on the lateral edge of this separating membrane. Later, when the lung burrows to extend around to the side and the front of the heart (Fig. 414), the nerve is pushed into the pleuro- pericardial membrane between the heart and the lung. This addi- tional rotation in the development of the viscera makes this nerve the most important landmark by which we retain our conception of the relation of these organs. In addition to the rotation of the organs around the phrenic nerve in the transverse plane of the body, there is another and /*' iB^^\ .4 1 * 1 — Cs L —DC 1 i c.v. Ph. Fig. 399. — Section through an embryo 5 mm. long (No. 80). X 25 times. Ct, fifth cervical nerve; C.V.. cardinal vein; S., subclavian vein;. DC, ductus Cuvieri; L, lung; Ph, phrenic nerve. 534 HUMAN EMBRYOLOGY. greater shifting of the organs in their descent from the region of the head down into the thorax. This well-known descent of the diaphragm during development was first observed by von Baer, and later was more completely studied by His, Uskow, and myself. The accompanying scheme, Fig. 400, illustrates this process. On the right side of the schema, which is in the form of an embryo six weeks old, the numbers of the segments of the body are given. The space marked Co represents the communication Fig. 400. — Scheme showing the position of the septum transversum within the body in human em- bryos from 2 to 24 mm. long. For description see text. 0., occipital region; C, cervical; T., thoracic; L.f lumbar; Co^ coelom. between the pleural and the peritoneal coeloms. The rectangular blocks in front represent the position of the septum transversum in embryos of various sizes which are indicated by the accompany- ing numbers. Thus, the block in the head region represents the position of the septum in an embryo 2 mm. long and that in the lower thoracic region the position of the finished diaphragm in an embryo 24 mm. long. The arrows indicate that the septum is mov- ing more rapidly where they are located than elsewhere. In gen- eral there are two m.ain foci around which the diaphragm rotates: PR. CCELOM AND DIAPHRAGM. 535 (1) in the upper cervical region at the dorsal end of the septum, which is its position in embryos 4 to 7 mm. long, and (2) in the thoracic region at the ventral end of the septum, which is brought about by the growth of the heart and lungs within the thorax. Up to the end of the fifth week there is no sign whatever of a separation between the various portions of the C(Blom, but now a line of separation makes its appearance between the i)ericardial and pleural coeloms, due to a constriction of its walls along the course of the ductus Cu- vieri, or rather the ductus lies in a ridge of tissue encircling the canal of communication between the pleural and the pericardial coe- loms. This ridge is not present in one em- bryo 5 mm. long, but in another of this size there is an elevation on the dorsal wall of the pleural ccelom, Fig. 401, which encircles the lung and joins the dorsal end of the septum transversum. This ridge is one of the pillars of Uskow, or the beginning of the fm. 4oi.-sagittai section pulmonary ridge, as I have termed it. It lonrcNo.Tse)™ x^°25^ttoS; gradually widens to cover the whole lung ; on ff.'^^^^tum ^t;a™'mT'l: its cephalic end it gives rise to the pleuro- i™e; /S, stomach; a, arm; . J- 1 1 J! TT 1 n •, Pfl., pulmonary ridge. pericardial membrane ot Uskow, and on its caudal end to the pleuroperitoneal membrane of Brachet. For the present, while it still represents two structures, it should carry a single name, and I believe pulmonary ridge to be a suitable one. In later stages, when the pulmonary ridge gives rise to the pleuro- perieardial and the pleuroperitoneal membranes, the name should be dropped. SEPARATION OF THE PERICARDIAL, PLEURAL, AND PERITONEAL CAVITIES. The first steps required to bring about a separation of these cavities have been taken and are well under way in an embryo 7 mm. long (Figs. 402-405). The embryo is well kinked upon itself and the septum transversum in its wandering is entering the thorax, as Figs. 400 and 402 show. The communications between the pericardial and pleural cceloms have been reduced to narrow slits, as indicated by the arrow in Fig. 402 and in the cast of the ccelom as shown in Fig. 403. A section through the lung region. Fig. 404, shows the two ridges cut across four times. The recon- struction. Fig. 402, gives the relation of the pulmonary ridge to the surrounding organs and to the structures of the body in its imme- diate neighborhood. This ridge is welh shown in the His Atlas, embryos A and B, as well as in Piper's reconstruction, and it may 536 HUMAN EMBRYOLOGY. be seen in any liuman embryo of this stage. The relation of the ridge to the phrenic nerve, as well as its form in older embryos, makes of it the anlage of both pleuropericardial and pleuroperi- toneal membranes. It lies in the sagittal plane of the body in this embryo in the region of the fourth cervical nerve, immediately over the lung bud, and connects the dorsal end of the septum trans- versum with the Wolffian body. In sagittal sections of embryos of this stage the ridge may also be cut twice, as Fig. 406 shows. Fig. 402.— Embryo 7 mm. long (No. 2). Enlarged 14 times. B.C., ductus Cuvieri; Ph., phrenic nerve; I, lung ; Li, liver; S, stomach; W, Wolffian body. The embryos just described are kinked to their maximum, and in the next stage, with the disappearance of the sinus praecervicalis and the beginning of the development of the neck, the head begins to erect itself and the pulmonary ridge widens and spreads both towards the heart and the stomach, as Pig. 407 shows. The con- nection between the pleural and pericardial cavities is reduced to a small narrow slit, which is guarded on the pericardial side by a valve-like membrane, the pleuropericardial membrane. As the pulmonary ridge widens to encircle most of the lung, the dorsal end of the septum transversum sinks into the thorax CCELOM AND DIAPHRAGM. 537 W.B. Fig. 403. — Cast of coeloiu of the embryo 7 mm. long. X 14 times. P., pericardial coelom; L., ccelom encircling liver; W.B., position of Wolffian body; M., position of mesentery. Fig. 404. — Section through the seventh cervical segment of the embryo 7 mm. long. (No. 2). X 25 times. Cj, seventh cervical myotome; C. V., car- dinal vein; D. C, ductus Cuvieri; Br., brachial plexus; PR., pulmonary ridge; Ph., phrenic nerve; B., bronchus; i7, heart; B. A., bulbus aortffi. Fig. 405. — Section through the embryo 7 mm, long, .6 mm. deeper than 404. X 25 times. Ti, first thoracic myotome; C.V., cardinal vein; W.B., Wolffian body; S., stomach; LPC, lesser peritoneal cavity; L, liver; H, heart; S.T. septum transversum. ='>r^ ^^-^,, „ - Ph X' ) / 1 , J ie i' ' if - , \ t' L t .. L 1 \ „ V- J \ - , i Ph ■a 'j Fig. 418. — Sagittal section tlirough an embryo 14 mm. long (No. 144). X 10 times. PA., phrenic nerve; 10. tenth rib; .S, stomach; K, kidney; W, Wolffian bodj'. '\f Fig. 419. — Sagittal section through an embryo 16 mm. long (No. 43). X 8 times. 9, ninth rib. of the diaphragm. The true nature of these pillars and their re- lation to the permanent diaphragm was finally determined by Eavn and Brachet, and by Swaen for the human embryo. In a human embryo 11 mm. long, immediately after the com- pletion of the pleuropericardial membrane, it is seen that the rota- tion of the liver and septum transversum is accelerated, and that Fig. 420. — Transverse section through an embryo 19 mm. long (No. 74). X 10 times. 7, seventh rib. The pleuroperitoneal membrane, PP., is incomplete on one side. the pleuroperitoneal membrane extends rapidly down into the thorax as the Wolffian body recedes. (Compare Figs. 407 and 414.) The -<-shaped section of the septum transversum and pleuropericardial and peritoneal membranes (Fig. 414) is soon changed by the growth of the lung and the shifting of the dia- phragm (Figs. 417 and 418), which gradually places the pleuro- pericardial membrane at right angles to the diaphragm. The opening behind the pleural and peritoneal cavities gradually CGELOM AND DIAPHRAGM. 543 becomes smaller and smaller (Fig. 419), closes first on the right and then on the left side (Fig. 420) ; usually both membranes are complete in embryos 19 mm. long. As the lungs invade the thorax in the wandering of the dia- phragm they must carry their surrounding pleural space with them, which calls for a radical shifting of the mesenchyme between the parietal pleura and the ribs. In fact there is a great mass of mesenchyme just at this point (Figs. 417 to 420), which in embryos at this stage has many unusually large spaces in it indicating that they are normal tears. After the diaphragm has reached its permanent position and the lungs begin to grow relatively larger, they encroach upon this tissue and it is reduced in quantity to make room for them. The lung also burrows between the pleuro- pericardial membrane and the main body wall, increases the extent of the membrane, and pushes it with its inclosed phrenic nerve to the medial side of the lung, between it and the heart. To what extent the permanent diaphragm is formed from the pleuroperitoneal membrane is difficult to determine. It is probable that the portion dorsal to the attachment of the pleuro- pericardial membrane of the septum transversum is formed by the pleuroperitoneal membrane. At any rate, the point of entrance of the phrenic nerve may be viewed as the most fixed point, one difficult to shift, for it is closly associated with the muscle of the diaphragm which invades the septum transversum when it is still high in the neck of the embryo. However, this kind of reasoning is not altogether sound and should be taken with some reserve. The shifting of large masses of organs, their power to burrow and extend as the lung does around the heart, and the fact that the liver grows into the pleuroperitoneal membrane (Fig. 418) while it is being separated from the septum transversum on its ventral side (Fig. 417), should make us somewhat reserved in our statements regarding the origin of the dorsal and ventral dia- phragms. In reality we have gotten but a little further than to confirm His, who stated that the septum transversum is extended dorsally and separates the pleural from the pericardial and peri- toneal cavities. Defining the septum transversum as he did. proved to be the foundation-stone of all subsequent study regard- ing the separation of the body cavities. The ccelom cavities of the myotomes are in general inde- pendent of the peritoneal cavity in man, with the exception of the second, which, according to Keibel and Elze, communicate with this cavity in an embryo 1.38 mm. long. These cavities of the myotomes are well developed in embryos 2 mm. long, but after this stage they soon disappear. The diverticulum of the coelom into which the testis wanders begins to form as the Wolffian body atrophies during the third 544 HUMAN EMBRYOLOGY. month of pregnancy. At first there is an evagination of the abdominal wall in the inguinal region, forming the inguinal bursa, which is lined by a sac of peritoneum, the vaginal process. This in turn sinks into the embryonic scrotum, which has formed inde- pendently. Soon the bursa is partly filled again by a marked thickening of its apex to form the conus inguinalis or gubernacu- lum, which continues rato the genito-inguinal ligament to the testis. During the seventh month the final descent of the testis takes place. At this time the bursa becomes markedly enlarged, the conus retracts, and the testis moves into the embryonic tunica vaginalis, which becomes completely separated from the peritoneal cavity at about the time of birth. A similar but much less marked process takes place in the female. In the case of the ovary the migration is slight, although provision for the descent has been made in the formation of both inguinal bursa and ligament. The lumen of the vaginal process usually disappears, but may remain open to form what is known as the diverticulum of Nuck. In not all cases does the testis enter the inguinal bursa, but instead remains in the abdominal cavity or within the inguinal canal. In other cases the processus vaginalis does not close after the descent of the testis and some of the abdominal viscera may enter the canal, forming congenital inguinal hernia. A similar hernia of the diaphragm may occur when the communication between the pleural and peritoneal coelom is not completely cut off. This kind of anomaly is much more common in the left side of the body than on the right, probably on account of the corresponding unequal growth of the liver on the two sides of the body. Con- genital hernia may also occur in the umbilical cord when the coelom of the cord is not obliterated after the intestine returns from it into the abdominal cavity. In case the opening with the exoccelom of the cord is very large, most of the abdominal viscera — liver, spleen, large and small intestines — may extend into it. In such cases the extruded viscera are covered only by a thin membrane composed of peritoneum and amnion. There remains still one very fundamental pocket of the peri- toneal coelom; it is, the pocket which forms the lesser peritoneal cavity. This pocket, the bursa omentalis, was first recognized in the embryo by His, and later Eavn discovered that it developed not only on the right side of the body but on the left side also, and that the cephalic tip of the right cavity separated and formed one of the serous spaces which comes to lie in the region of the right lung. More recently the development of the lesser peritoneal cavity in man has been studied by Swaen and by Broman. Broman's admirable study is comparative and includes numerous CCELOM AND DIAPHRAG]\I. 545 liuman embryos; he confirms for the human embryos Eavn's dis- coveries in the rabbit, mentioned above. In an embryo 3 mm. long there are two peritoneal pockets on either side of the oesophagus, the right being somewhat larger than the left. The left is of but short duration in the human embrj^o ; it evaginates, Broman believes, but the right, the recessus pneumato-entericus dexter, is closely associated with the less- er peritoneal cavity. Just be- low this recess there is another depression (Fig. 396, below FW), the recessus mesenterico- entericus of Swaen, which soon extends between the stomach and liver and later gives rise to the omental bursa. A reconstruction of the right recess, in an embryo 5 mm. long, is shown in Fig. 421. The upper recess (Rpedx) reaches to the lung bud of the right side, the dorsal recess ex- tends into the mesentery of the stomach, and the ventral recess, recessus hepato-entericus, bor- ders on the liver and later en- circles the Spighelian lobe. Be- low they communicate with the peritoneal cavity through the hiatus communis recessuum (Her) or the primitive foramen of Winslow. Sections through the hiatus may be seen in three different stages in Figs. 393, 405, and 411. As the stomach grows and bends to the left, it gradually gives form to the lesser peritoneal cavity, as is shown by a cast of it from an embryo 11.7 mm. long (Fig. 422). It is easily seen that the hiatus of Fig. 421 has become the true foramen of Winslow, the recessus hepato-entericus has become the cavity of the lesser omentum (Bomin), and the recessus mesenterico-enteri- cus has become the cavity of the greater omentum (Bomaj). The cavity of the greater omentum is formed, according to Swaen, by a burrowing of the cavity into the wall of the stomach, and not by a simple bend of its mesentery, as is indicated, for instance, in Fig. 411. According to the B.N.A., the bursa omentalis is divided into Vol. I.— .35 Fig. 421. — Cast of the lesser peritoneal cavity encircling the intestine, from a human embryo 5 mm. long. (After Broman.) /2perf.r., recessus pneumato- entericus dexter; Rmse., recessus mesenterico-enteri- cus; Rhe., recessus hepato-entericus; Her., hiatus communis recessuum. 546 HUMAN EMBRYOLOGY. a vestibulum and a recessus superior, whicli form together Bro- man's bursa omenti minoris, and a recessus inferior, which is the same as the bursa omenti majoris. In my opinion, it would be better to apply the terms of the B.N. A. to these spaces in the embryo as soon as their fate is clear. For the same reason I have used the term pericardial cavity in describing the smallest embryos, for the relation of this space to the heart enables us to identify it. Additional unnecessary names only complicate the subject. Bomai Fig. 422. — Cast of the lesser peritoneal cavity from an embryo 11.5 mm. long. (After Broman.) Bomin,, bursa omenti minoris; Bomaf., bursa omenti majoris; Rcacoe.^ atrium bur.'fa omentalis; FW., foramen of Winslow; Bic, bursa infracardiaca. According to Broman, the recessus pneumato-entericus of Kavn extends upward in the human embryo to the bifurcation of the lungs. By the time the embryo is 11 mm. long, the recessus pneumato-entericus begins to separate from the lesser peritoneal cavity and is soon pinched off to form the bursa infracardiaca, as shown in Fig. 422. From now on, it can be found in all embryos as a closed sac lying between the right side of the oesophagus and the diaphragm. It gradually grows in size and is about one centi- metre in diameter at birth; in a specimen from an adult man (Fig. 423, Bic) it measures 42 x 20 mm. This third pleural cavity, very well developed in all animals which have an infracardial lobe of the lung, is therefore found in all human embryos and probably also in most adults. Its frequency will have to be established by statistics. 548 HUMAN EMBRYOLOGY, BeomAjST : Beschreibung eiiies meusehliehen Embrj'O von beinahe 3 mm. Liinge, Morphol. Arbeiten, Bd. 5, 1894. Entwicklung des Zwerchfelles beim Meuschen, "\'erliandl. d. anat. Gesellsch., 1902. Die Entwicklungsgeschichte der bursa omeiitalis, Wiesbaden, 1904. Bryce and Teacher : Contribution to the Study of the Early Development and Imbedding of the Human Ovum, Glasgow, 1908. DA>fDY : A Human Embryo with Seven Pairs of Somites, Measuring About 2 mm. in Length, Amer. Journ. of Anatomy, vol. 10, 1910. Herzog : Contribution to Our Knowledge of the Earliest Known Stages of Placeu- tation and Embryonic Development m Man, Amer. Jonrn. of jViintomy, 1909. His: Anatomie der menschl. Embi'yonen, Heft 1, 1880. Jlitteilung zur Embryolog'ie der Siiugetiere und des Menschen, Arch. f. Anat. u. Physiol., Anat. Abt., 1881. Hochstetter: Die Entwicklung des Blutgefiiss- Systems, Hertwig's Handb. d. vei-gleich. und exper. Entwickl. d. AVirbeltiere, Bd. 3, 1906. Keibel: Mensehenafi'en, IX. Liefenuig von Selenka's Werk, Wiesbaden, 1906. Keibel und Elze: Nnrmentafehi zur Entwickl. d. Wirbeltiere (Mensch), Heft 8, 190S. Klaatsch: rber den descensus testiculorum, Morph. Jahrbuch, Bd. 16, 1S90. Kollikek: Entwicklungsgeschichte des jMeuschen, Leipzig, 1S70. Mall: Development of the Lesser Peritoneal Cavity in Birds and Mammals, Journ. of Moi-phol., vol. 11, 1891. Devehipment of the Human Co?loni, Journ. of J[(U-phol., vol. 12, 1897. Supplementai-y Note on the Development of the Human Intestine, Anat Anz., Bd. 16, 1899. Contribution to the Study of the Pathology of Human Embryo, Johns Hopkins Hospital Rop., vol. 9, 1900, and Journ. of Moiph., vol. 19, 1908. On the Development of the Human Diaphragm, Johns Hopkins Hospital Bulletin, vol. 12, 1901; and Proceedings of Amer. Assoc. Anatom., vol 5, Washington, 1901. Mazilier: Contribution de Fembnclogie du diaphragma, Lille, 1907. iNICrLLER, E.: Beitrage der Anatomie des menschlichen Eetus, Kgl. Schwedische Akademie, Bd. 29, 1897. Muller, I. : Uber den Ursprung der Netze, Meckel's Arch., 1830. Piersol: Pluman Anatomy, Philadelphia, 1907. Piper: Tber ein ini Ziegler'schen Atelier hergestelltes Modell eines menschlichen Embryos von O.S mm. Nackenlinie, Anat. Anz., Bd. 21, 1902. Ravx: Uber die Bildung der Scheidewand zwischen Brust- und Bauchhohle in SiiugetierembiTonen, Arch. f. Anat. u. Physiol., Anat. Abt., 1889. Retzius : Biolog. Untersuch., Bd. 1, 1890. Spee, Graf v. : Beobachtungen an einer menschlichen Keimscheibe mit offener MeduUan-inne und canalis neurenterieus, Arch. f. Anat. u Physiol Anat Abt, 1SS<). ■ J ; Strahl : Beitrage zur Entwicklungsgeschichte des Herzens und der Korperhohlen Arch. f. Anat., 1889. SwAEx: Recherches sur le developpement du foi, Jouni. d'Anat et de Physiol vol. 33, 1897. ' Note sur la Topographie des Organs abdominaux, Bibliog. Anat., 1899. Toldt: Uber die Geschiehte der Mesenterien, Yerhandl. d. Anat. Gesellsch., 1893. UsKOw: Uber die Entwicklung des Zwerchfells, des Perieardiums und des Golems Arch. f. mikr. Anat., 1883. Yelpeau : Embryologie ou Ovologie humaine, etc., Pai-is, 1833. Waldeter : Lehrbuch der topographisch-ehirurgischen Anatomie, Bonn, 1899.