_ ee oes ed : ‘ Lm Ae a A er We BeBe mw Spee ae . it. hh ee Se eo — : . Mee pole make Oy beeen be Ae Py ee : » Phe SS BReaee ae Ba * Aart a neiee bray Seah oo ae . . . ~%. pen ae ert ee rs ae are 7. a te oF ed Raw se BuOebew gre ae, “t Be Mirh le! sell des oan ee hd Saye 't, Nee LR, Yager: oe Meee ey Pe ae Ave Bare a reney ost . . an we tng. i ar eB Ae ver ‘ : career Pek, . ‘a oes . 5 oat va Vas . . < . on 7 . . ae , " - ‘ . ‘ Ne thet ot ee : . ; ; takes . .. ede nas . Gre ‘ ¢ ‘ Ze ‘ id ry - we 4H « pe . wre? ‘ ‘ ‘ . 33 . , ‘ ‘ oe ‘ . ‘ ‘ te , bis ¢ U id ’ . "ss ‘ ate Pi ee oe ‘ ’ ’ - -. - , . . 4 4 ‘ ‘ Pacer , . a) ‘ arr tat ear | rs g ’ ‘es , e Sh ee ree Xf, osteer rar) i pers ’ tom ot 9 Ser & R Fi _" A : : 7 Pep ete aloe Fs 6g LS EOE EOE Sage at se fie + see Te BRD OR PEE Are aie oe et # Fe SOW RS OE Oe Oe Gro ee oF A i ree ner oe cay eee ee P ; POR PEO FP aT ;. : LE tadtentededh dak . Sire yryteaeiswi eta? ee ee wa P row asa vider ye hen Br i, hs a “hb Morr ey Gaats OO FM Pek et? : ' ' . ’ . ’ J de yh he eracge sts eel * nese eer : : 7 CP Oar eed Tee ne F be ets gs et ee dads ¢ ‘ oe $4 Om ale Ft ee ome, . : i POPE PAP OTH BI Od ! ps0 Prelate alr ede. . 1 @ sa : ca 24 pM DES ET OF peering . <3 “% aA 4 Bd cook Luloal a om ; 3 Sp , : ? ‘ . = yf se . at oy Abie Vee galt V9 8 avo eee bea ‘ rm) ios wees ease aee © Loewe de ere : ; a Cee pbbot Tee he agee ee Oe ol asg eee ke nce pee ME peeret erat ed fades arate ome Ss 7 8A a8 eh ava is Harta Pee i oe re ce ae 4 Crs ea aN etre cere LOM OE UY 2 7 Pig": 210 9, a3 207s Beek? Be ‘ pate Ren Bs tee pe hE eT eee i et we ee be did eee AD Fate aehas gh acd te phe 4/4 pe tle ty, peg ater sage op Pte es BAP DEER hoe b Bek Cpe & pr ethees pees BAe O bork AR Ae ee? 7 6S 4, Ord PRS “fp o4 rte BY . ae Ree PAF Par wed creed PAF Sire hat woos sot ld LT ee wise a beeer $s tr bee x oo) both’ Pee oe OS€ rit bai ho ri bale > bes at oh ded ate, ps re eee ih oe ; ORBEA As ary an) BAe ey Cilazore! * ls sie b.? og A Ce Side Ge etal? Pi ae oe oe Oe ADE sep te de ee oe ee . tp warts Lt sue ep sot vied er ee bat : ee yh FOR. THE PEOPLE FOR EDVCATION FOR SCIENCE LIBRARY OF THE AMERICAN MUSEUM OF NATURAL HISTORY a rob F : a have Brien 4 Ts _ Dat et) Hie Ae i Le ‘ ; “ sf on by ihthe ne ae a PA | U ea | “a” ar. aad aA 7 a, 8S vO ae A > 7.) Lt : 5 7 ; me al i, 4 . | bay ah 4 A THE JOURNAL | Ca oae CZ s y ORF as | OF EXPERIMENTAL ZOOLOGY EDITED BY WILLIAM E. CastTLe * Franx R. LIne Harvard University University of Chicago EpwIn G. CoNKLIN JACQUES LOEB Princeton University Rockefeller Institute CHARLES B. DAVENPORT Tuomas H. MorGan Carnegie Institution Columbia University HORACE JAYNE GrorGE H. PARKER The Wistar Institute Harvard University HERBERT §. JENNINGS CHARLES O. WHITMAN Johns Hopkins University University of Chicago EpmunpD B. WILSON, Columbia University and Ross G. HARRISON, Yale University Managing Editor VOLUME 9 1910 THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY PHILADELPHIA, PA. y Ae Aad | : a} = | My iH : Gg a af ; a ‘i A ~~ x if | f | THE WILLIAM KEITH BROOKS MEMORIAL VOLUME EDITORIAL COMMITTEE ErxHan A. ANDREWS HERBERT §S. JENNINGS Johns Hopkins University Johns Hopkins University EpwIin G. CoNKLIN GEORGE LEFEVRE Princeton University University of Missouri Ross G. Harrison Tuomas H. MorGan Yale University Columbia University Wiiir1am H. HowE.u EpMuND B. WILSON Johns Hopkins University Columbia University HORACE JAYNE Henry H. WILson The Wistar Institute University of North Carolina THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY PHILADELPHIA, PA. 1910 YRARBII HTAD MUSEUM MAGILEMA VAOTEIN SARUIAY Go TO THE MEMORY OF WILLIAM KEITH BROOKS 1848-1908 THIS VOLUME IS DEDICATED AS A TOKEN OF AFFECTION AND RESPECT BY HIS PUPILS AND ASSOCIATES iPAG MUSE UM LAU tiEA YROLEU TARUTAN 40 aes es SOBS* os ae ee tha 4. hi he a 8 PREFACE The influence of Professor Brooks upon the present generation of American zodélogists was of such notable character that the presentation of a ‘‘Festschrift’’ by his pupils would have been a most natural and fitting act, and one that would undoubtedly have been consummated within the next few years, had not his untimely death prevented. Professor Brooks took so warm an interest in the activities of his former pupils, that it is certain he would have been deeply gratified to receive a tribute of this kind, and it is a source of much regret on the part of those who have helped to bring forth the present volume, that it was not issued during his lifetime. It seems none the less appropriate, however, to publish the volume as a memorial, and hence what otherwise might have | een done amidst gladness and felicitation is now. done with a sense of affectionate remembrance and regard. Plans to this end were proposed shortly after Professor Brooks’ death, at a meeting of his pupils and associates held in Baltimore on December 31, 1908. It was at that time suggested to issue a special volume of the Journal of Experimental Zodlogy in commemoration of his life and work. Brooks’ direct activities lay mainly in the field of comparative and philosophical mor- phology, and hardly extended into that of experimental zodlogy as now understood. But a man’s work may bear as abundant fruit in that of his pupils and followers as in his own immediate achievement. Some of those who have in later years devoted themselves to experimental studies, have borne witness to his awakening of their interest in these problems, and to his clear recognition of experiment as a first necessity of biological prog- ress. The Editorial Board of the Journal of Experimental Zodlogy, of which he was a member from its organization, there- fore gave cordial assent to the use of its pages for the purpose that has been indicated and a committee to carry out the pro- PREFACE ject was appointed by Prof. Samuel F. Clarke, who presided at the memorial meeting. Those who had been advanced students under Professor Brooks and a few others who had been intimately associated with him, were invited to contribute to the volume. The response to this invitation has been most cordial, and the Committee takes pleasure in expressing its thanks to the many who have codperated for the success of the undertaking. The account of the life and work of Professor Brooks which is here printed is of somewhat unusual character in that it has been contributed by a number of different persons, who have known him, at various periods of his life, or who are especially qualified to estimate the significance of his work. The result is a composite sketch, which, while unavoidably lacking in con- tinuity, nevertheless gives, we believe, a vivid impression of the man, the exact nature of whose charm and influence was such as would have been difficult for a single person adequately to por- tray. For contributions to this sketch the Committee is indebted to Professors Ethan A. Andrews, William Bateson, Samuel F. Clarke, Edwin G. Conklin, Henry H. Donaldson, Gilman A. Drew, Otto C. Glaser, Caswell Grave, Francis H. Herrick, Henry McE. Knower, George Lefevre, Alfred G. Mayer, Maynard M. Metcalf, and Henry V. Wilson. It has been a task of no small difficulty to fit together so many short sketches into an harmoni- ous whole, and it has, of course, been impossible to reproduce any of them verbatim. The Committee must therefore beg the indulgence of those who have so kindly contributed, hoping that no one will feel that his work has received inconsiderate treatment. CONTENTS 1910 No. 1. SEPTEMBER BrocrapHy. William Keith Brooks: A sketch of his life by some of his former Sire See MBCA A TCS POTLTAIIS....... 6... cen cew epee 1 EpMunD B. Witson. Studieson chromosomes. Five figures.................. 53 GEORGE LEFEVRE AND W. C. Curtis. Reproduction and parasitism in the Unionidae. Thirty-nine figures, five plates............................ 79 Otto C. GuasER. The nematocysts of Eolids. Twelve figures............... 117 ALBERT H. Turrie. Mitosisin Oedogonium. Eighteen figures .............. 143 J. Puayrarr McMurricu. The genus Arachnactis. Five figures............. 159 Francis H. Herrick. Life and behavior of the cuckoo. Twenty-three I ee aici vow H thee Sted pak etire ahs mam ee 169 No. 2. OCTOBER E. A. ANDREWS. Conjugation in the crayfish, Cambarus affinis. Eight Reed Mag aeeak, TERR eat OR eh ae ee 0 eee as ee 235 8.0. Mast. Reactionsin Amoebatolight. Two figures..................... 265 H.S8. Jennines. What conditions induce conjugation in Paramecium? Four ee es ee ee eet ee se ee Dole LP Pah tak ele ea 279 M.M.Mercatr. Studiesupon Amoeba. Forty-five figures.................. 301 SAMUEL RitTeNHOUSE. The embryology of Stomotoca apicata. Thirty-two re a Se TE ES a aS ee ee See 333 A.M. Reese. Thelateral line system of Chimaera colliei. Eighteen figures... 349 Epwin Linton. On anew Rhabdocoele commensal with Modiolus plicatulus. Ni ce wai awed wes sv oA dee wes 371 No. 2—ContTINUED te R. P. Cowes. Stimuli produced by light and by contact with solid walls as factors in the behavior of Ophiuroids. Thirteen figures................. 387 Epwin G. Conxkuin. The effects of centrifugal force upon the organization and development of the eggs of fresh water pulmonates. Forty-seven FP OD 5s i2es ws ie eS ea CTO Ee Oe eee 417 Henry F. Nacutrizes. The primitive pores of Polyodon spathula (Walbaum). Twelve Reuresy. 2. 026. h-c.2m ook veh 6 oe se eg ee ie 455 No. 3. NOVEMBER Serraro Goto. Ontwo species of Hydractinia living in symbiosis with a her- mit crab: Twenty-three figures... .. 0.8. 2... s.coeee se Oe ee 469 HvuBErtT LyMANn Ciarxk. The development of an apodous holothurian (Chiri- dota rotifera). Six figures. .... 2... 25.0000. 0.0.0 4s oe Oe 497 Henry L. Ossporn. On the structure of Cryptogonimus (nov. gen.) Chyli (n. sp.), an aberrant distome, from fishes of Michigan and New York. Seven FiPUTES: occ eee ee eek Sa wld wed eee wwWa oa «ue 517 H. V. Witson. A study of some epithelioid membranes in monaxonid sponges. Dwenty-one figures... ...0.62 0... ee ss eee ww 0 0s Se 537 CHARLES WILSON GREENE. An experimental determination of the speed of migration of salmon in the Columbia River. Two figures................. 579 T. H. Morean. Cytological studies of centrifuged eggs. One hundred and eightedn figures .>...... 2.0.0. 0.62.0 0028: 146 2. ee 593 No. 4. DECEMBER Davin H. TENNENT. Variation in Echinoid plutei. Twenty-four figures..... 657 Duncan 8. Jonnson. Studiesin the development of the Piperaceae. Seventy- GHOBPUTES . b.90 S25 6 ees os 5 Leip enn soe owe bb om es 715 Rosert P. Bic—ELow. A comparison of the sense organs in Medusae of the family Pelagidae. Thirty-eight figures.......:..........i.2) 751 Ross GRANVILLE Harrison. The outgrowth of the nerve fiber as a mode of protoplasmic movement. Three plates and three figures................ 787 GrorGE C. Price. The structure and function of the adult head kidney of Bdellostoma stouti. Four figures........ AW thes, 849 J. FRANK DANIEL. Observations on the period of gestation in white mice..... 865 WILLIAM KEITH BROOKS A SKETCH OF HIS LIFE BY SOME OF HIS FORMER PUPILS AND ASSOCIATES! William Keith Brooks, second of the four sons of Oliver Allen Brooks and Ellenora Bradbury Kingsley, was born at Cleveland, Ohio, March 25, 1848. His parents were both descended from the early settlers of Massachusetts, the first of the name having come to America from England, before the year 1634. His father, who was born in Middlebury, Vermont, had removed to Cleveland in 1835, where he was engaged in business. Asa boy, Brooks was studious and thoughtful. He obtained his early education in the public schools of Cleveland, and entered Hobart College at the age of eighteen. Two years later he entered the junior class at Williams College, from which he received the degree of Bachelor of Arts in 1870. Although in college he had 1 Biographical sketches of Professor W. K. Brooks have been published as fol- lows: 1. E. A. Andrews; Sketch of William Keith Brooks. Pop. Sci. Monthly, vol. 55 no. 3, July 1899, pp. 400-409, with portrait. 2. E. A. Andrews; William Keith Brooks. Science, N.S. vol. 28 Dec. 4, 1908, pp. 777-786, and Jan. 1, 1909, p. 31. : 3. Edwin G. Conklin; The Life and Work of Professor Brooks. Anatomical Record, vol. 3, no. 1, January 1909, pp. 1-13, with portrait. 4. Edwin G. Conklin; William Keith Brooks. Proc. Am. Phil. Soc., no. 190, 1909, pp. 3-10. 5. Edwin G. Conklin; Biographical Memoir of William Keith Brooks, 1848- 1908. National Academy of Sciences, Biographical Memoirs, vol. 7, February 1910. 6. E. A. Andrews; Biography of William Keith Brooks, in ‘‘ Leading Americans,”’ Holt & Co., New York. 1910. A meeting commemorative of Prof. Brooks was held in McCoy Hall, Johns Hopkins University, Nov. 12, 1908. Addresses were made by Professor B. L. Gildersleeve, Dr. H. M. Hurd, Professor W. H. Howell, Professor E. A. Andrews, Dr. Caswell Grave and Professor W. H. Browne. See Johns Hopkins Univer- sity Circular, January 1909. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 9, NO. 1. 2 WILLIAM KEITH BROOKS shown special interest in philosophy and in studies with the micro- scope, he was uncertain on graduation whether to devote himself to natural history, to mathematics or to Greek, in all of which subjects he excelled. After leaving Williams College he spent a short time with his father in business, but this occupation was not to his liking and he gave it up to become a teacher in a school for boys at Niagara Falls. After holding that position for two years he became a graduate student at Harvard College under Louis Agassiz, who was then at the zenith of his career, and at the seaside laboratory established by this great master in 1873 on the Island of Penikese, Brooks began a life-long devotion to the study of marine zodlogy. In 1875 he was appointed assistant in the museum of the Boston Society of Natural History and in the same year received the degree of Doctor of Philosophy from Harvard. It was during the summer of this year, while at home on his vacation, that he organized, together with Theodore B. Comstock and Albert H. Tuttle, a class for laboratory instruction in zoology and botany for teachers. With the opening of the Johns Hopkins University in 1876, one of the twenty fellowships was awarded Brooks, who thus at its very foundation entered the service of the institution with which he was to remain connected until his death. He was immediately advanced to the position of Associate and later was successively appointed Associate Professor of Comparative Anatomy, Asso- ciate Professor of Morphology, Professor of Animal Morphology, Professor of Zodlogy and Head of the Biological Department. In 1878 he was made Director of the Chesapeake Zodlogical Laboratory of the University, an institution which he organized and which became a potent adjunct to the Baltimore laboratory in the training of biologists. Professor Brooks was the recipient of numerous public honors. When but thirty-six years of age he was elected a member of the National Academy. Hewas chosen a member of the Ameri- can Philosophical Society in 1886, and of the Academy of Natural Sciences in 1887. He was Lowell lecturer in 1901 and gave one of the three general addresses before the International Zodélog- ical Congress at Boston, in 1907. He received the honorary de- A SKETCH OF HIS LIFE 3 gree of LL.D. from Williams College in 1893, from Hobart College in 1899, and from the University of Pennsylvania in 1906. For his discoveries on the life history of the American oyster he was awarded the medal of the Société d’Acclimatation of Paris, and for his work on the Stomatopoda, a Challenger medal. He was editor of the ‘‘Memoirs from the Biological Laboratory” of the Johns Hopkins University, joint editor of the ‘“‘Studies from the Biological Laboratory”? of the Johns Hopkins University, and one of the editors of the Journal of Experimental Zoédlogy. He was a member of the Boston Society of Natural History, the American Academy of Arts and Sciences, the American Society of Zodlogists, and of the Maryland Academy of Sciences, and was a Fellow of the American Association for the Advancement of Science and of the Royal Microscopical Society. On June 13, 1878, Professor Brooks married Amelia Katherine Schultz (deceased 1901), daughter of Edward Thomas Schultz, and Susan Rebecca (Martin) Schultz of Baltimore. Two chil- dren were born, Charles Edward Brooks and Mrs. Menetta White (Brooks) Daniel, both of whom survive him. A congenital defect of the heart had always caused Professor Brooks to lead a less active life physically than do most men, and to this trouble other bodily ills were added as life advanced. After a continuous prostrating illness of nine months he died at his home “Brightside,” near Baltimore, November 12, 1908. As a stimulating teacher, an ardent and successful investigator, and a philosophic naturalist, the influence of Brooks on the de- velopment of zodlogy in this country has been very great. His students are scattered widely in college and university, in muse- ums, and scientific stations in this country and abroad, and many have become eminent in their own fields of work. His discover- ies, numerous and important, have enriched zodlogy and have been incorporated into the permanent literature of that science. Certain of his memoirs are models of completeness and beauty. His brilliant and thorough work on the oyster fisheries of Mary- land has made his name familiar to economists and to intelligent legislators. In an age perhaps over-eager in the pursuit of new knowledge Brooks has called attention back to the fundamental 4 WILLIAM KEITH BROOKS nature of knowledge itself in such a way that his helpfulness has been gladly and gratefully recognized in many circles of science. In his personal character Brooks combined gentleness and strength and arare wisdom. In university matters and in all the affairs of life he was a lover of freedom and of justice tempered with kindliness. Although looked upon from the beginning as a master mind, he was totally free from formality and never ass 1med the authoritative air of the traditional professor, but met his students and associates in all simplicity and frankness as fellow student and inquirer. What he wasasaman and a student was fully revealed, and the singularly deep influence which he exerted upon those who worked with him constitutes a remarkable tribute to his genuine ability and worth. The close friend- ship between him and his students was evidenced in many ways in the daily life of the laboratory, and at the evening gather- ings at his home. It was given more definite expression on the occasion of his promotion to a full professorship, and again on his fiftieth birthday, when his pupils came together at Bright- side to present to him formally the portrait for which he had sat at their request. The appreciations, reviews and recollections embodied in the following pages and coming from former students and associates record some of the labors and some of the traits, human and pro- fessional, of a profound thinker and tireless worker. SOME RECOLLECTIONS OF PROFESSOR BROOKS CHRONOLOG- ICALLY ARRANGED 1876-79.2. Among the company of twenty young men who came together in Baltimore in 1876 as the first group of ‘‘Fellows”’ of Johns Hopkins University, were three biologists. One of these was of less than average stature, wearing a serious face, with close-set eyes, quiet and unhurried in his movements, speaking not frequently, and never with haste. This was W. K. Brooks of Willams and Harvard. The biological department was at once organized by Professor Martin with Dr. Brooks as an Asso- ? Professor S. F. Clarke, Williams College. A SKETCH OF HIS LIFE a ciate, and it is an illustration of that quiet impressiveness of Brooks’ simple manner that his appointment was immediately recognized by every one to be most eminently fitting. He quickly gained our respect and admiration by the con- stant seriousness of his thoughts, and the simplicity and gen- uineness of his statements: simple in expression but showing care- ful and deep reflection. Our affection was won and held by his genuine, never-failing interest in, and friendship for us. As L recall his reading to me of the then unfinished manuscript of his book on Heredity in my room on Centre street in 1876, of the many long talks on biological subjects, in either his roomor mine at the University, at Brightside, or at Crisfield, Fort Wool or Beaufort, I become aware again of the constant seriousness, and power of his thought, which awoke and continually increased an admiration for his intellectual ability. Brooks’ friendship was even and steadfast. It never found great expression in words, but it never wavered. [I felt this dur- ing my early years of association with him, and the conviction was but strengthened with the growing years. This steadfast- ness of affection and confidence in his friends, his perfect simpli- city and genuineness, and his serious and profound mind are to me the sources of Brooks’ great and lasting influence on men. 1883-84.? The first time I saw Professor Brooks was in 1883. The year before, while I was endeavoring to make out some of the points in the structure of Balanoglossus, then imperfectly known, it was announced in the Johns Hopkins Circular that a littoral species of that animal had been found at Hampton, Va. At Mr. Adam Sedgwick’s suggestion I wrote to Brooks asking if I might come over to investigate it. Brooks, as his friends will remember, did not habitually answer letters, but as it happened he did answer that one and sent me a cordial invitation to come and try. Such leave was no little thing to give, for Balanoglos- sus must have been known to be one of the prizes of the station, but in professional generosity Brooks was royal and lavish. * Professor William Bateson, Cambridge, England. 6 WILLIAM KEITH BROOKS From the first moment of meeting in the empty warehouse of the Normal College, which then served for a laboratory, we be- came friends. He was of course much my senior, but there is no other word which so well expresses the happy unconstrained feel- ing that I felt towards him and that he showed towards me. It had been settled that I was to live at Mr. Cock’s boarding house, across the creek, where the Brooks family had their quarters, and we thus spent several weeks in constant intimacy. He was not the least like any one else I had ever known; and I find it difficult to express the charm which his personality had for me then, and has had increasingly since. He was, as I soon found, on account of superficial eccentricities reputed a reserved and rather inaccessible man. In general company he would indeed often remain silent and I think he had moods in which a morbid shyness would take complete possession of him, but once at his ease he was another man. At such times he would talk abundantly, but his speech was always that of thetaciturnobserver, with the special, holding quality that the speech of such men has. He spoke in short incisive phrases, full of novelty, suggestion, and humorously inventive thought, sometimes, but not often, rising to enthusiasm. I see him now, with his short, round fig- ure, sitting on the piazza at Mr. Cock’s or lying flat on his bed— a posture he often took when in a talking mood—ruminating his thoughts, which, if the truth must be told, were periodically in- terrupted by his devotion to tobacco. What a strange combi- nation it was! The grave, kindly face, the earnest solemnity of philosophical speculation and the homely quid. Now, I suppose, no university professor, however contemplative, dare use tobacco in this particular way; but I wonder if any university professor ruminates spacious ideas as Brooks used to do, daily through long vacant hours of leisure, to the delight and elevation of a youthful listener. Those are the times of true education ‘when lofty thought Lifts the young heart above its mortal lair.’’ Many of Brooks’ pupils must look back on similar pleasant hours of intimate, informal summer laboratory life as critical moments in their development. For myself I know that it was A SKETCH OF HIS LIFE ( through Brooks that I first came to realize the problem which for years has been my chief interest and concern. At Cambridge in the eighties morphology held us like a spell. That part of biology was concrete. The discovery of definite, incontrovert- ible fact is the best kind of scientific work, and morphological research was still bringing up new facts in quantity. It scarcely occurred to us that the supply of that particular class of fact was exhaustible, still less that facts of other classes might have a wider significance. In 1883 Brooks was just finishing his book “‘ Her- edity’’, and naturally his talk used to turn largely on this subject. He used especially to recur to his ideas on the nature and causes of variation, and to the conception which he developed in “ Her- edity,” that the functions of the male and female germ cellsare distinct. The leading thought was that which he expresses in his book (p. 312) that “‘the obscurity and complexity of the phen- omena of heredity afford no ground for the belief that the subject is outside the legitimate province of scientific enquiry.’ He deplored the fact that he had no opportunity for the requisite experiments in breeding, but he saw plainly that such experiments were the first necessity for progress in biology. To me the whole province was new. Variation and heredity with us had stood as axioms. For Brooks they were problems. As he talked of them the insistence of these problems became imminent and oppressive. It all sounded rather inchoate and vaporous at first, intangible as compared with the facts of develop- ment which we knew well how to pursue, but with the lapse of time the impression became strong that Brooks was on the right line. ‘That autumn I went home feeling that though in technique we were a long way ahead of Johns Hopkins—I had the pleasure of showing off the Jung microtome, then the latest thing in pro- gress, to the admiring Baltimore men—yet somehow Brooks had access to novelties of a more serious description. In the following summer I was again with Brooks at Beaufort, N. C., but in that year I soon fell ill and was for a long time too weak for much talk of any kind. Indeed, but for the devoted ministrations of Brooks and his students, who for weeks performed for me the offices of the trained nurse, I might never have left 8 WILLIAM KEITH BROOKS Beaufort alive. The ‘Heredity’? had meanwhile appeared and I am afraid Brooks was disappointed with the reception it met, for it was noticed with little more than formal sympathy. Looked at in the light of subsequent knowledge its purpose was indeed rather, as he says, ‘‘to turn the attention of others into this channel” than to make an independent advance. In the preface he wrote: ‘‘I have little hope that my views will be accepted in the form in which they are here presented, but I do hope that they may serve to bind together and to vitalize the mass of facts which we already possess and that they may thus incite and direct new experiments.” That function he and his book did at length admirably perform for many, both in England and in America. 1885-89.4. In going over my memories of Dr. Brooks I find that my mind does not separate him from his environment. I con- tinually see him in the semicommunal life of the laboratory, whether in Baltimore or Beaufort, Woods Hole or the islands of the West Indian sea, which so stirred and charmed him. Even his home life with its restful, satisfying beauty was but a detached fragment of the other larger existence. I think of him as the cen- tral figure, wise and kind, of a circle of young men coming from many quarters, from New England, the Middle States, the West, and the South, from Canada, England and Japan, a society from which older members were always going out to honorable careers and into which new were coming to learn the ways and traditions of the school. Very different were we, but knit together from the start by the strong bond of a common interest, and presently by growing appreciation of him who made the school. It took us but a short time to learn that here was no mere work-shop, well organized and in which we might acquire the requisite degree of skill in a profession, but that we were in the company of a master mind, wide ranging in the fields of knowledge and inquiry, pro- found in contemplative thought, and with the acuteness of the observer who discovers what has been hidden. As I dwell on the man and try to single out mental habits and * Professor H. V. Wilson, University of North Carolina. A SKETCH OF HIS LIFE 9 attributes from the whole of his personality, I come to many that arrest and enchain my attention. It is interesting to consider his practice and advice to beginners in the study of Nature. It was to start out, not from a general principle, but from some phenomenon that had caught the eye and become a nucleus for thought. Continued, persistent obser- vation and reflection circling round such a.center would yield, he held, solid results in the shape of new facts and would sooner or later lead one into living contact with great questions. This method of work was eminently characteristic of his independent, individualistic temperament. The serenity of Dr. Brooks impressed every one. In a mind so strong, active, and keen, calm temperateness was doubly notice- able. This peace of mind must have been due in part to the fact that his critical insight was unobscured by self-seeking. v OOD O_ rrr Right Oral 11.52 +.12 1.89 =.09 16.48 +.80 100 2 Days Left Oral 11.37 +.12 1.92 +.09 16.89 +.82 100 Right Anal 14.86 +.18 2.68 +.12 18.05 =.88 100 Left Anal (14.26 +.19 2.92 +.13 20.52 +1.01 100 Right Oral 15.07 +.21 3.16 +.15 21.02 +1.04 100 3 Days Left Oral 14.79 =.20 3.08 +.14 20.86 +1.03 100 | Right Anal | 17.79 +.26 3.95 +.18 | 22.21 +1.11 100 /Left Anal 17.62 +.27 4.08 +.19 23.17 +1.16 100 { Right Oral 16.49 =.19 2.86 +.13 17.39 +.85 100 ‘tiave )\ Left Oral = 16.34 +.19 | 2.86 +.13 17.52 + .86 100 | Right Anal 20.16 +.21 3.13 +.14 15.57 +.75 100 || Left Anal 19.74 =.22 3.37 +.16 | 17.08 +.83 100 {, Right Oral 18.27 +.22 3.37 +.16 18.45 +.90. 100 5Days | LeftOral 18.05 +.23 3.43 +.16 19.00 +.93 100 )}| Right Anal | 22.54 +.22 | 3.34 +.15 14.82 +.72 100 {| + 2%) 3.61 =. + 79 | 100 ( j Left Anal (| 22.35 669 TABLE 4 Constants, Series 1II. Temperature at fertilization 27.6° C. Temperature during development 27.5°-29° AGE STRUCZURE re MEAN | SEoLurioe rome eee aN NUMBER | Right Oral | 11.46 +.10| 1.59 =.07 | 13.87 = .67/ 100 epee | Left Oral | 11.48 =.10| 1.59 +.07 | 13.90 = .67| 100 | Right Anal | 16.76 +.14| 2.16 +.10 | 12.93 = .62 100 Left Anal 16.54 + 15 2.30 +.10 | 13.92 + .67| 100 Right Oral | 11.82 +.19| 2.94 =.14 | 24.93 =1.26 100 oDan Left Oral | 11.79 +.20| 3.00 +.14 | 25.47 +1.29 100 Right Anal 16.55 =.21. 3.15 =.15 | 19.07 = .94 100 Left Anal | 16.41 +.21| 3.19 +.15 | 19.48 = .96 100 | Right Oral | 12.76 +.22| 3.37 =.16 |-26.45 +1.34 100 rane Left Oral | 12.73 +:22|° 3.37 =.16 | 26.53 =1.35) 100 Right Anal | 16.96 +.18| 2.77 +.13 | 16.37 = .80 100 Left Anal 16.89. +.18 2.77 +.13 16.40 = .80 100 Right Oral 15.79 +.29 4.33 +.20 | 27.47 +1.40| 100 Bina Left Oral | 15.81 +.29| 4.36 +.20 27.57 +1.41 100 Right Anal 20.16 +.28 4.18 +.19 20.74 +1.03 100 Left Anal | 20.03 +.28| 4.19 +.19 | 20.92 +1.04 100 TaBLE 5 Constants, Series IV. Temperature at fertilization 29° C. Temperature during development 28°-29° | STANDARD COEFFICIENT OF | J AGE STRUCTURE MEAN aE ana ae eee oA NUMBER es ae “Right Oral | 9.47 =.09| 1.47 +.07 | 15.54 +.75 100 ae | Left Oral | 9.49 +.09| 1.47 +.07 | 15.58 + 76 100 | Right Anal | 15.86 +.11| 1.65 +.07 | 10.48 +.50 100 Left Anal | 15.79 =.10| 1.56 =.07 | 9.89° = 47 100 | Right Oral | 9.92 =.08| 1.32 +.06'| 13.34 +.64 100 3am | Left Oral | 9.86 +.08| 1.27 +.06 | 12.90 +.62 | 100 Right Anal| 15.62 +.14| 2.08 +.09 | 13.32 +.64 100 || Left Anal | 15.20 =.13/ 2.03 =.09 | 13.38 +64 | 100 | Right Oral | 9.17 2.07) (£205 | 12093 aa 100 ity Left Oral | 9.14 ©.07| 1.06 +.05 | 11.67 = .56 100 \| Right Anal | 13.65 =.12| 1.79 =.08 | 13.11 = 763 | 100 | Left Anal | 13.41 +.12 1.86 +.08 13.87 +.67. 100 | Right Oral | 8.56 ©.07 | 1.08 £705 | 12.72 = 6t 100 oe Left Oral | 8.54 +.07} 1.12 +.05 | 13.18 +.64 100 Right Anal | 12.38 +.12| 1.80 =.08 | 14.59 =.71 100 Left Anal | 12.25 2:12) 4-85 2:.08)) 45.0199 100 TABLE 6 Constants, Series V. Temperature at fertilization 28.5° C. Temperature during development 27.5°-28.5° STANDARD COEFFICIENT OF agers sap tir ae a a DEVIATION VARIATION ——— NUMBER Right Oral 11.56 =.10 1.49 +.07 12.90 +.62 100 Us, Left Oral | 11.81 +.09 1.46 +.06 | 12.36 +.59 100 ays —}| Right Anal 17.26 =.19 2.95 =.14 | 17.10 +.83 100 || Left Anal | 17.50 +.19| 2.83 +.13 | 16.18 +.79 100 Right Oral 12.86 =.13 2.03 =.09 15.81 +.77 100 a | Left Oral | 12.81 +.13|) 2.07 +.09 16.17 +.79 100 he Right Anal | 17.57 +.13 2.05 +.09 | 11.66 +.56 100 (Left Anal 17.27 +.15 2.34 +.11 13.57 = .65 100 Right Oral 12.70 +.09 1.47 +.07 11.59 +.56 100 ree Left Oral (12.49 +.09 1.40 +.06 11.23 +.54 100 y Right Anal | 16.50 +.11 1.64 +.07 9.97 +.48 100 | Left Anal 16.40 +.12| 1.82 +.08 | 11.10 +.53 100 Right Oral 13.31 +.09 1.46 +.06 10.96 +.52 100 5 Dake Left Oral (13.41 +.09| 1.37 +.06 | 10.28 +.49 100 y Right Anal 18.07 +.13 1.98 +.09 10.99 +.53 100 | Left Anal 18.14 +.13 2.00 +.09 11.05 +.53 100 TaBLE 7 Constants, mean of the means, Series 1-V AGE STRUCTURE *MEAN STANDARD COEFFICIENT OF NUMBER DEVIATION VARIATION Right Oral 11.42 +.05 1.95 +.04 17.08 +.37 500 2 Days Left Oral | 11:45 =.05 1.95 =.04 | 17.04 +.37 500 | Right Anal | 15.77 +.07 2.55 +.05 16.18 +.35 590 Left Anal | 15.64 +.08 2.65 +.05 17.00 +.37 500 Right Oral 12.81 =.08 2.93 =.06 22.93 +.51 500 —— Left Oral 12.72 +.08 2.90 +.06 22.86 +.51 500 Right Anal 16.07 =.10 3.37 +.07 20.97 +.46 500 Left Anal 15.84 +.10| 3.42 +.07 21.59 +.48 500 Right Oral | 12.81 +.09 3.21 +.06 25.12 +.56 500 4Days /| Left Oral | 12.72 +.09 3.18 +.06 24.21 +.54 500 Right Anal | 15.80 +.11 3.83 +.08 24.28 +.54 500 Left Anal | 15.58 +.11 3.87 +.08 | 24.85 +.56 500 Right Oral 13.92 +.12 4.16 +.08 29.91 +.69 500 Days Left Oral (13.91 +.12 4.13 +.08 29.75 +.68 500 Right Anal 17.51 =.13 4.61 +.09 26.36 + .60 500 | Left Anal | 17.40 +.14 4.67 +.09 | 26.89 +.61 500 671 672 DAVID H. TENNENT TaBLE 8 Constants, Series VI-IX AGE STRUCTURE STANDARD DEVIATION | COEFFICIENT OF | VARIATION NUMBER —— | | | VI. Temperature at Fertilization 27.5°. During development 27°-27.5° (| Right Oral | 18.68 +.08 | 1.33 =.06| 7.13 +.34 | 3Days ) Left Oral | 18.62 +08 1.24 + .05| 6.69 +.32 | || Right Anal | 23.18 +.14/| 2.08 =.09 | 8.99 =.43 || Left Anal | 23.09 +.13| 2.06 =.09| 8.96 +.43 Right Oral | 19.26 =.09| 1.40 +.06| 7.28 =.34 | 4 Days Left Oral | 19.31 +.11 | 1.65 =.07/| 8.56 +.41 Right Anal | 23.48 +.14| 2.16 +.10| 9.19 +.44 | | Left Anal | 23.57 =.17| 2.54 +.12 | 10.80 +.52 Right Oral 18.62 +.10, 1.48 +.07| 7.99 +.38 SRage | Left Oral 18°65, =°09 | 1.45 © 06 | 7.80 =. Right Anal | 23.00 +.15 | 2.23 +.10| 9.70 +.46 | | Left Anal 23.03 +.14 | 2.17 +.10) 9.46 =.45 VII. Temperature at fertilization 24°. During development 24°-25° Right Oral 21.03 +.30 | 3.48 +.21 16.57 +1.0 3 4Days /| Left Oral | 20.80 +.30| 3.54 =.21 | 17.05 +1.0 7 ; }| Right Anal | 26.88 +.38| 4.40 +.26 | 16.40 +1.0 2 || Left Anal | 27.06 +.38| 4.42 +27 | 16.36 +1.0 2 VIII. Temperature at fertilization 26°. During development 26°-28° (| Right Oral | 18.35 + .23| 3.47 +.16 | 18.95 +.93 éDays | Left Oral | 18.52 +.24| 3.58 +17 | 19.33 +.95 ; || Right Anal | 23.12 + :26| 3.93 +.18 | 17.01 +.83 | || Left Anal | 22.93 +.27| 4.12 +.19| 17.99 +.88 IX. Temperature at fertilization 27°. During development 27°-28° Right Oral | 23.76 +.24/| 3.64 +.17 | 15.38 +.74 6Days {| Left Oral | 23.69 +.24) 2.59 =.17 | 15.15 +.73 | | Right Anal | 28.81 +.31 | 4.60 +.21 | 15.97 +.78 | 28.60 +.31| 4.69 + + 80 | | Left Anal | 22! 16.41 100 100 100 100 100 100 100 100 100 100 100 100 61 61 61 61 100 100 100 100 100 100 100 100 VARIATION IN ECHINOID PLUTEI 673 Since neither the defects nor the multiplicities appear as such in the tables of constants, although the defects have influenced the size of the mean in every case, aseparate treatment of these varia- tions is necessary. I mention these as percentages in the various series studied. Series I. 2 days. 3 days. 4 days. 5 days. Series IT. 2 days. 3 days. 4 days. 5 days. No anal arms (type, fig. 9), 1 per cent. No left oral arm, 1 per cent. No anal arms, 8 per cent. No oral arms, 7 per cent. No anal arms, 16 per cent. No oral arms, 28 per cent. No defects. No anal arms, | per cent. No oral arms, 6 per cent. No left anal arm (type, fig. 10) 2 per cent. With both multiple right and left anal arm (type, fig. 12), 3 per cent. With multiple right anal arm (type, fig. 11), 4 per cent. With multiple left anal arm, 5 per cent (one with three rods). No anal arms, 3 per cent. No oral arms, 4 per cent. Multiple right anal arm, 2 per cent. Multiple left anal arm, 2 per cent. Left anal arm cleft (type, fig. 17), 1 per cent. Slightly cleft pre-oral region, 5 per cent. No oral arms, 4 per cent. Multiple left anal arm, 2 per cent. Shghtly cleft pre-oral region, 4 per cent. No left anal arm, 1 per cent. No left oral arm, 1 per cent. Multiple right anal arm, | per cent. Multiple left anal arm, 2 per cent. Slightly cleft pre-oral region, 2 per cent. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOuw. 9, NO. 4. eed 674 DAVID H. TENNENT Serves ITI. 2days. Multiple left anal rods. (type, fig. 14) 2 per cent. 3 days. No anal arms, 1 per cent. No oral arms, 1 per cent. Multiple right anal rods, 1 per cent. Unceleft pre-oral region, 1 per cent. 4 days. Slightly cleft pre-oral region, 5 per cent. 5 days. Slightly cleft pre-oral region, 7 per cent. Series IV. 2days. Split left anal arm, 1 per cent. 5 days. No oral arms, 5 per cent. Series V. 2days. Noright anal arms, 4 per cent. No left anal arms, 1 per cent. No oral arms, 3 per cent. Cleft left anal arm, 1 per cent. 3 days. No anal arms, 1 per cent. No left anal arm, 1 per cent. 4days. No anal arms, 2 per cent. No oral arms, 3 per cent. Series VI. 3 days. No anal arms, | per cent. 4 days. One oral arm, 1 per cent. Multiple anal rods, 2 per cent. Multiple oral rods, 1 per cent. 5 days. Multiple anal rods, 1 per cent. Series VII (61 plutei). 4 days. No anal arm, | pluteus. One anal arm, 1 pluteus. No oral arms, 4 plutei. VARIATION IN ECHINOID PLUTEI 675 Series VITTI. No arms. Rods passing around the body (fig. 20), 1 per cent. One right anal arm, 4 per cent. One left anal arm, 3 per cent. Multiple skeletal rods right anal (type, fig. 14), 2 per cent. Multiple skeletal rods left anal, 3 per cent. Multiple right anal arm, 2 per cent. Multiple left anal arm, 2 per cent. No oral arms, 6 per cent. One right oral arm, 2 per cent. One left oral arm, 1 per cent. Series IX. Median analarm with multiple rods (type, fig. 6), 2 rods, 4 per cent, 3 rods 3 per cent. , No right anal arm, 5 per cent. No left anal arm, 4 per cent. Multiple right anal rods (type, fig. 14), 3 per cent. Multiple left anal rods (type, fig. 14), 2 per cent. Slightly cleft anal ridge, 1 per cent. No right oral arm, corresponding left with 2 bars, 2 per cent. No left oral arm, corresponding right with 2 bars, 2 per cent. Pre-oral region slightly cleft, 1 per cent. Pre-oral region uncleft, 1 per cent. 2. POTENTIAL EGG VARIATIONS The results noted in the preceding section indicated the pos- sible existence of definite type or line variations. Further in- formation on this point seemed desirable and I, therefore, under- took a very simple set of experiments. Experiment 1. Five females A, B, C, D, and E were selected and the eggs of each placed inaseparate aquarium. All of these were fertilized with sperm from a single male, F. The resulting embryos then were AF, BF, CF, DF and EF. Temperature at fertilization 27° C. During development 27°—28°. 676 DAVID H. TENNENT Two things were noted as a result of this experiment. First, the time of beginning of cleavage and the rate of cleavage are nearly constant in a given lot of eggs. Lot CF began its separa- tion into two cells in 39 minutes. Lot DF showed no constric- tion until 43 minutes after impregnation. Second, a variation; larvae of an auricularian type in lot AF. On the third day when a count of one hundred embryos was made 13 per cent were found to be of this type, the body elongated and with neither oral nor anal arms. A skeleton was present. Experiment 2. Five females A, B, C, D, and E were selected and the eggs of each divided into two portions, A;-E; and A,—Ep, each portion being placed in a separate aquarium. Lots A,-K, were fertilized with sperm from male F. Lots A,—E» were ferti- lized with sperm from male G. Temperature at fertilization 27°C. During development 27°-28°. Lots AiF and A.G showed the same variation, a divided condition of either the right anal or of the left anal arm. (Type, fig. 17.) No other noticeable multiplicity or defect occurred. Experiment 3. Five females, A, B, C, D and E were selected and the eggs of each divided into five portions A,-A; through E,—E;, and each lot placed in a separate aquarium. Lots A,-E, were fertilized with sperm from male F Lots A,-E, were fertilized with sperm from male G Lots A;-E; were fertilized with sperm from male H Lots A.-E, were fertilized with sperm from male K Lots A;—-E; were fertilized with sperm from male M Temperature at fertilization 28°. During development 27°—28°. After the 3rd day all of the A series, 7.e., AiF, AG, A;sH, A.K, A;M were characterized by the presence of a high percentage of larvae which under favorable conditions would have completed their metamorphosis. With a slight deviation either way, of a hundred plutei counted from each of the five lots, 21 per cent showed these characters. Series B was weak from the first. At the end of the 3d day all of this lot with the exception of B,M were dead. VARIATION IN ECHINOID PLUTEI 677 Series C was weak. C,M living longest. Series D showed 15 per cent normal larvae of the type described for series A; 4 per cent showed the uncleft condition of the pre- oral region. This variation occurring throughout the series. Series E. Six per cent normal larvae (type described for ) re ri Von] [ove 5S o>) a TR se Tl = | | 7 1 oO > ae a ee = [o-e) mre N Yes) 3 = Li = © > 8 te a MSH NN OH (fo) 7, N ~~ — — “A Q = Fie | aha ica] | s (Yo) re oO rH oD [o 6) Z ~ ae > z A 8 eormoguaasesseer| | Flog) ae ae ae Se BA shog g ‘wsy pouy ifoT “IT St |—a- 36 ee 3 = | | e — | gee lee Raa ae! Ne = SRP = Re Sn es te ee! eee gr ae ee ee | Lae & oc) aaMe ON ie Yor os i | = = “usa SE was ee re E 4 > = | — ee | = RS aoe : = a ar 2 Ee . = = A i 3 | a pe Ee | = | a = | — = | ee s a = HANAMIMAMAAN | 8 sal ye Dis ae | 2 : Gael ae tS ee E: il Maia — b= ~ = i: os = eS or = ie : S : ane Jomo AMHBONBASHABH | ole == i Ce == Ee Oe eee = i Be ee shog § ‘usp youy 15 "JT Ss is mi N re Pes SS al rH te ~ ~ _ Son 692 ~ 6 7 = 8 39 aa E12 ay age S16 693 BOA OAM re MOO OD ON & OH © 1D OOO N ri rei re a i — es | ie a — oa | rei r= R | = oe | _ oo a (e.6) | —— Sen oe Oe | N = — —_ i fi a rr ee lor) aoa a ee SS a Sse a: — | ase re re ba) Yo) | romeeyes ON wo | - Ro) — — = “a = —s = aes 4 noes: be 4 | — — oD Se Se re 8 ee Nw — re re N — So | 1 1 2 1 1 2 1 1 IT. Left Oral Arm, 3 Days | 5 1 1 1 2 VARIATION IN ECHINOID PLUTEI iH 1 t 1] 2 1 1 | sin — — = — on HH |r mormon aRtmwonNHaonaARso Sse Se rs mr BI ANN NAN AINE shoq ¢ ‘Wwsy jouy 1faT ‘JI he 4 5 694 DAVID H. TENNENT IT. Right Oral Arm, 4 Days x _ a a 8} 9} 10) 11) 12) 13) 14) 15) 16, 7 18, 19) 20) 21 —————— | | — —_—— 10° 1 | J — lor) ee bo A jot 10) j= — — our - bo me Rw bd bd . Righ bo iS) KH eoOne w oo —_ FOCOMNNWONNWDOWAHOHOH ee — — — _ —- | | of of 4 4 2 3 4 1214 1819 8 17] 1 695 VARIATION IN ECHINOID PLUTEI IT. Left Oral Arm, 4 Days 18 19 20 11) 12; 13 14 15) 16 17 | ) AotonrrtHoomnowoaonmort O&O a © bo | re rt ori et RN Se ev) [= = hae — — AN no — ee: aay" UE Wl — | 3. — ee wt NA HO 1D at re z fos | — —T 4 Ry a a eh? + Se — a ae Eis | ee ee | ie bene — — — a “= eT i hae ey — b. ae — |° a i Saree — a cS OmrMN OHA Oe ODO TN OD 12 oh w | oN os Bos en NNNN ‘ 2 shog 7 ‘wip youy fay “TI 696 DAVID H. TENNENT II. Right Oral Arm, & Days 10 i 12 13 15 16 17| 18| 19| 20 21| 22 23 24 | 4 4 13 0 14 1 1 15 1 1 > 16 1 1 2 Say Bp) 9) ial a i 8 1 | pl 2 3 j = 19 1 a 20 1] 1 tl ig 4 = 21 1 t.ho ot 2 7 S99 } 1 4 4) 2 16 "S 23 Ly a). Beh oe 11 fe 24 3 2 4 ag 12 é:.25 1 1) 4 5 6 ah a6 "6 2 2 27 1, Ear ie ae 28 1 1 29 1 i 3 0 7; 3| 4) 3] 3} 10) 14! 10] 15| 10, 121 5) 1 lt 9 10 11 VARIATION IN ECHINOID PLUTEI iw) oO IT. Left Oral Arm, 5 Days 1 12 13 14 15 — j= se a | We eet ee 16 17 18 19 20 21) 22, 23 24 SoS | ‘ee ee | 0 ae 0 = 1 Fe 0 Py | | 1 = | | 2 | ie = ou ee | pat ose | Rae ae Pe o.):2 re 342 | | | a7 3 3 2 | eae: a 9 gga.” |. | 33 7344 | | 19 ae wie oe hata ae A.D | Si aoe Se | i ed | 12 10 18 gui 41 ee ie eee 697 III. Right Anal Arm, 2 Days III, Left Anal Arm, 2 Days DAVID H. TENNENT Ill. Right Oral Arm, 2 Doys s| ol 10 11] 121 13 14 15 16 17 eS — ——————_— 10| | 2 11 12 i, 1 13 14 1} 1 15.) 30-2 1) 6 ae 17 | et 18 | 5 19 ne 20 1 21 2 1 1 a eee | 8} 7 2 2 5 4 1) 3 2 6 3 1 Hy. 5 Ded ee ied 210 14 24 29 10 9 III. Left Oral Arm, 2 Days 8 9/10 11 12 13 14 | ——_ SS 10° 2 Gi 1 | awe 13 14. 1 15 a 4 16 | te dh-2 17 | 1} 3 18 | 4 19 | i) 2 20 teas ——— || 2/10 13 | Re CUNT WwW be | 25 28 12 7 15 16 iy VARIATION IN ECHINOID PLUTEI 699 III. Right Oral Arm, 3 Days | 7 8 9 10 11) 12) 13 14 15 16 17 18 19 20 g| 1 | 1 aa 2 9 | coe. = 10} 1 2 | | 3 Su a 4 | Ng gts dace? las © 12 | 11 2 1 | 6 = 13 Hei} 3 1 = 14 - a 3 5 = 15 22 2 3 1) 1 11 £16; 1 | @ 3 1 | 3 16 4334 2 1 1 13 = 18| 1 | 223 1 8 1 13 19 1421 1 1 10 _. 20 1 2 | 2 i ge S 21 1} 1 | 1 3 22 | ie ie | 3 23 1 fee ae 4 2 18 15, 14 15) 6 9 5) 2 4 3 1 2 III. Left Oral Arm, 3 Days | 7 8 ee es 8} 1 | 1 9 | | 0 210| 1) | 1 1 3 fe 11| 3 2 1 | 6 pb 2 2 1 is | | 5 213 | 2 1 | 4 =i oo oF 1 5 216 1) 1) 5 2 1 10 Si; kh eo ay | i 3s 1 16 mig a: 3 2 1 10 >18| 1 22 3 1 3 1 13 ~ 19 4 i) i} i 1 We. 9 ~ 20 | 1) 3 Ee On Be (| 10 S 21 1 1 1] 3 22 | | 2 4 23 | 1 | 1 SSS Oe OOOEOEeE———EE—EE | } 4 2) 17 7 18,10 6 10 4 1) 5| 2 1 3 TENNENT DAVID i. 700 III. Right Oral Arm, 4 Days 6 oe 8 9 10 11; 12} 13) 14) 15} 16 17) 18 19} 20} 21 1 1 mo OO N OD | | | | | | | | | | | | | | | | | | 1 | 12 rast oo SH OD shog 7 ‘ws NX 10 3) 4 1 Th) Sis 1 1 0 if 1 3 2 —_ |/ -——_—_ | 6 of 5] 4) 3 1 } 14| 11) 21; 4, 4 4 ouy big TIT 0} 0} 16 if | Lit) Ret Oval Lore Pe i or eh ales a a | — Se R | NX gi hg = | S| aN a ine = | aH N exe = | sH N hese = | NN a = | OD rH — 10 = | oD N a iis ee 4 mae 8 eles mH N = | Nw Hn | 3 [=] | me HN OD =. eg Le oe tia rm x i Se ee E ieee = pos _ re | |aamecnwragoraem ss | Set BR TT NNN TNA EN shiogq 7 ‘wily ouy ifeT “TIT VARIATION IN ECHINOID PLUTEI 701 II. Right Oral Arm, & Days 24 25 | 19 Le 12 13 14 15 16| 17 18 19 20 21 22 id 2 1 | | | 1 13 | | | | 0 14 H a} \4 | | | | 3 Soa ft ofa Ht 1 Ses | | | 6 & 16 3} 3} 1) 2 : mee fe 2 Oke ae cg Zari i 1 5} i) 6 ie aa | (|B o18 | 1] 2 3} 1) 1) 2 a Be hs | | | 10 £19] | a ee a | ae | | ul ~ 20 aly et at a) a) 2 | 11 S 21 A ee | 5 ~ 22 1 1) | 1) 1) 2 Ph 6 223) | | | Le 0 = 24 | 1 aff ay 8 a5) | 1 4) 1 | | 6 mi 26) | 1 2 fe ae | | 7 1} 2) 10 28 | 2) 2 29 | | ios 30° | 1 1 E 1615 gliz|oj6)4 4/4|6|1\ 9/2 15 DAVID H. TENNENT 702 III. Left Oral : 5 Days ASS ee TO” ee ae rr = & | Om aa a sH 4 es A | 3 x | | nN a ma) ro) A | 3 | eal — _ A | | R | ae AN La A S | aANAN Ee mm 2 4 | a 2S 2 ia. wer) =< mS re 5 | re = a | = . = | Onrmrnd i oe : = | x NON : aa = | a a = | 3 re rc S| BHANANAAN | 4 re mm S| asa aw & ri | > re re ay = ) ha re SS | re me AN =F; ee re e7e = = ae TNSweSeneese | [Razer eagaNRAaRnAas fing ¢ ‘wip qouy ifeT “TIT 703 VARIATION IN ECHINOID PLUTEI IV. Left Oral Arm, 2 Days IV. Right Oral Arm, 2 Days Cee oe ee NOD hy HH 19 x Fm ANA AAN CO S mi Q 3 | _ as a mm | mo1d oO es Wo) oD Lees. | = ge “ta alae ee om) [RAI NOCONn com S £3 an ne we A oe for) tat ok om N , | on | a a HOD 0 OS) CO CO OD | 2 OT A NR ON “NN S ¢ Pe oe ima.) |= att mo) rt 00 00 oD SH | & | as a | ro an — st RS —_ S of} SHANNA ee a) HN os + Bas io ie : ~~ re N al ~H (oilers: sa } : | | Aawmeonaa |] eS e | Ss hoes i ee ee oe oe | z WAOHMAVGHW WON HAR | be A ce I oe ee ce ce Be Oe Oe Be FO | shvg ¢ ‘wy qouy ifayT “AT shog ¢ ‘Wy youy 1faT “AT HAOOMMOUNANMHAMMAHRRA | MeN TN | MO ORDMDOONAN SO 9 s ——_— nN mON Oo - me iB AaNHA ES Ql «7 i asa oe ie ri . rei -e mn 10 On eS CO N OO rt OD N (or) = | | = “ges ae “i — ate 2 el eer es i a a oN Co Gh rt. fap = | Ho re a et | & = = | / 5 ea | eo OA AS wi ila oe — a ~~ Qe ORK GS +O ~ N S snl oO | a ee aay Oa Oe e¢ueo 2 es | EK = * oe aS ~~ —|— 2B an AN Be a) | & og Oe Fialae, te: = oe Ca apa a = | 3eweonea | i eS sos ort ore ort . == =. 4 ri SEY aa [ween agweonaae | SLOT @ WLY JOU 14%1Y “ALT sing ¢ ‘wily youy qybvy “AT aa 704 DAVID H. TENNENT IV. Right Oral Arm, 4 Days IV. Left Oral Arm, 4 Days 5| 6| 7| 8 9 10] 11) 12 5} 6| 7 8| 9] 10) 11) 12 arts 0 ee Tad 1 oe 9 1 1 2 og 1 0 of ae i) (4 2 Me elu 1 2 wie ay i 4 a £AG bg: 2 Si g2) a) || Sb Bes) 7 a ies £ 11 4 3 7 eae 6 4 111 ps habe 2 4 6 8 1 | at S14 212; 6 1) | 21 era i Sit od 17 ae | 414, 3) 3l | 24 = 14 1} 10) 5| 1 | 17 Z 16 Alogi B 2 15 | 3 : 5} 4 | 24 0) | eee Vi i} td 3 NS 16 3.3 ae 18 TG ae a =~ 19 1 1 aoe | 0 19 1 1 | 1] 3 16 46 22) 9 2 1] 1) 3/16 45, 26 rine. IV. Right Oral Arm, 5 Days IV. Left Oral Arm, 5 Days | z 6 7 8 9) 10, 11 6 7 8 9| 10) 11 Se a ae Sele Nn Se 3 SU 9) amet 5 A aa 1 10) 4 4. 6 25 | 13 yeu eae Ee i Pc | 5 E14 ‘a6 4) Bhelee © 10 1 1) 5 6 1 | 14 So aah a 4 Bb) Mee leo 11, 2 3) 5} 5} 1 | 16 S08 1) 7} 8 2 119 S$ - 12-2 5) 5} 6 38, | 21 1a Wr ae epi =< 13 6 6| 2 1) 15 2 15 2) 6 2 | 10 = 14 2 71) 3) iia = 16 d| alee ee N15 a7 4 | 13 17 1 1 . 16 i) a S Re Sed ond ee ets aoe eae ee A es = | ~ 5 11 27 38 18 1 6 11 26 39, 16 2 VARIATION IN ECHINOID PLUTEI 705 V. Right Oral Arm, 2 Days 8 9 10) 11) 12 13 14 15 i) — > 0 Bs — for) bo Left Anal Arm, 2 Days. 20 bo wWDwntwwreee Aon Pwr V. Right Anal Arm, 2 Days. Ww — ear ) CO Pm Ot r bo 3 8 10 25 26 20 "1 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 9, NO. 4, V. Left Oral Arm, 2 Days 8 9) 10) 11) 12) 13) 14 15 6 1 1 7 0 8 1 1 9 | 1 1 10} 1 1 11 1 1 12 1) 1 2 13 | 0 14 2, 2 1 5 15 1) 2 3 ds | a a Re | 12 17 1 | 5 6 12 18 2| 494 1 20 19 | 13 4 6 1 15 20 | 3 7 4 5 2 21 21 1 3 4 22 | 1 1 2 6 8 21 3416 11 2 706 DAVID H. TENNENT V. Right Oral Arm, 3 Days 8 919 1 12 13 14 15 16 17 _ ee SS eee er it 1 1 Q 12 | 1 1 2 a 0 © 14 1 ‘4 3 <= 15 3 1 4 3 26) 3d) 9a) a Ce gies 18 Sag 3 1 6 3 6 8 27 Res 1 ai al 2 10 > 19 | 2 2) 2 5 8 15 fe 20 Qa Sh aloe ee . 21 1 a 3} he a Oe, es es ae aoe 2 6 710 14 17 20 18 5} 1 V. Left Oral Arm, 3 Days | | 8 9 10 11| 12 13 M15 16| 17 qi. Sa Hal | 1 10 1 | 1 = 11 H jt) Aa | 2 12-1 a 1 es 13 0 eo 14) 1 1] 2 | 4 mG 2 3 1 6 La SAG 3 3 4 1 2 15 = 17 = Oe dae cake 27 ie 1 i) 3-30 2 | 12 > 19 eet le es a Ht ~ 20 9} 2) 5 5) I 1 16 eis act eal 13 22 0 id Cie | ; : 2 6 7 12 17 1120 20) 3 2 VARIATION IN ECHINOID PLUTEI V. Right Oral Arm, 4 Days 9 10 11 12 13 14 15 16 212 1 | 1 2 13 1] 3 | 4 at 14 4 1 | 5 aie} i ura you 2 16 ~< 16 His 2 9 4 22 S17 10 310 4 | 27 = 18 i 11 63 2 | 13 = 19 1 3 4 1 9 coos 20 1 i 1) 3 — 2} 6 11) 26 22 24 7) 2 V. Right Oral Arm, 5 Days ei ek 1 / 10 11 12 13 14 15 16 10 1 any “ee § | 0 s . 0 13 1 1 «i? ere 1 = | = 15 1 | 6 7 x . 3 Wo 32332 / |i ef / 3 5 2 4 1) 15 = 18 wie eB Ob: ‘> 19 | 4 a Fh >| 48 = 20 1 6 2 410 3) 25 2 11 2 = 99 1 i 2 707 V. Left Oral Arm, 4 Days } Fear 9,10 11 12 13,14 15 SMG es - shes Sa oe a > 10 | hi i | | 1 Ql | RA | 0 +12) VY | 1 y £13 | i i | 3 —14) | 4 1 I 6 315) 1] 24 5 3 2 | 17 = 16 | 3 2 7 7 2 | 20 eT 1 8} 8 8| 1) 26 1s | | i) 1) 7} 2 1/12 ree th pa 8 2 10 "20 | PP a as 2 10, 9 2529 20 5 V. Left Oral Arm, 5 Days 10 11 12 13 14 15 16 10 1 1 % ll 0 S 12 0 q tom 1 1 = 14 1 1 = 15) 1 4, 2 7 “ 164 2 2 2 a 3 10 a 35 4 4 16 a 3 5 4 1 13 > 19 | 4 6 6 2 19 a 6 6 9 9 | 30 Etat | 0 22 Hitt 23 ie 4 2) 21) 23) 26 20 DAVID H. TENNENT 708 Summary I-V. Left Oral Arm, 2 Days HAHOHNMOANNKDONOHAWNH mses NOD Or © lt tH OM OD =| a — ot = | = | >) al ae Coe | 15 | 7 a = x) Ta AO E = : el | rt Bl a N HT WMONTROROAA | & 5 mAs Ss St SS N = | N rw % 0) 290 190 > & O SH | # r= Se on | ee) = | NW ANAM H TH SCOMM DH | 2. rt oS. Koon) Rea, NAaATA CO mes > a = NAAN = = | aN a ee J spon wmmaonantReonwnagcnaa | Sosos Ss ss sss ANNAN shog ¢ ‘wip jouy 1faT *A-l fhunwung Summary I-V. Right Oral Arm, 2 Days Ore HORAN WM HH 10 bc es Oe Oe | | OmrAOSo ae BANNAN shod ¢@ ‘wy youy qybyy + A-y] funwunyg SHH DONWMNOnH® | OD 69 10 P= 00 Be 10 =H OD | a a — ss ri = | N ~~ i S . = a ie = | ~ HARTA | = =| = 4 2 ae SPOS es ee re | _ 1D 3 | _ att OOM O Or fF ee 1 m4 a os rt AN | e Hq OORHA OHM OK HS | SS rm re (on i | N a nee me = | mr Hr atte Dare © WwW ow | re Se ee ee | 0.6) S| — ONT AATKReAN [3 b | resort rei ile) ea - maar.) OHO OMO A RR a: rei ra 1 Yon) oe qe NAR HHONA SF | a ~ | a ae NAAN ree a a | ie 709 ECHINOID PLUTEI IN VARIATION Right Oral Arm, 3 Days Summary I-V. | 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | OMOMWDOANAHOAMRANOKRKWMROMN | ms SOON HOO mR OO HOO RS 4S | a obs ene te | 7 — — orn laa | wemibahh * re | — AAA ea esr NN oO din AN | a | Sa NO tH HS Ld 350. 8 Similar section of slightly older stamen, showing one archesporial group of cells at each end of thesection. X 350. . 9 Similar section of still older stamen, showing two archesporial masses in one loculus and none inthe other. X 350. 10 Similar section of slightly older stamen, showing two archesporial masses (microsporangia) in each loculus. X 210. 11 A single microsporangium from the stamen shown in fig. 26. X 350. 12 Longitudinal section of a stamen, from a transverse section of a spike, showing microspores, tapetam, wall and line of dehisence. X 150. 13 Surface view of part of spike, showing overlapping of bracts and the variable number of stamens and stigmas. X 10. 14. Longitudinal section of stamen, from transverse section of spike. X 7d. STUDIES IN THE DEVELOPMENT OF THE PIPERACEAE 741 CIO at a a PRON 4 Ore, v8 if oe ae Oats AS 0 v, eee Ll? Ba: £580) Wea “eae Nose i, 6 y OOO RA (faec.2e- Bi oes =<. LSS = SI WRORRCOY n\ WAS omnOv Og) it RSScseree Johnson*’and Kellner, del. 742 DUNCAN S. JOHNSON EXPLANATION OF FIGURES 15 Section of the stamen shown in the last figure, several sections removed from the one there shown. X 160. 16 Transverse section of anearly mature stamen with four microsporangia. X 75. 17 Section of nearly ripe pollen-grain, showing mitosis of first division. x 1000. 18 Section of older, two-celled pollen grain. X 1000. 19 Ripe pollen-grain, in which the wall has disappeared. X 1000. 20 Transverse section of half-mature stamen with four microsporangia. x 100. 21 Transverse section of stamen with three pollen sacs, of a staminodium and of @bract. X 100. ZZ ‘Transverse section of stamen, showing two sporogenous areas in the upper tuculus and one in the lower. X 100. 23 Another section of the same stamen, showing the continuity of the sporo- genous mass in the upper loculus. X 100. 24 Transverse section of stamen showing one sporogenous mass in each locu- lus. xX 100. 25 Surface view of spike, showing bracts, stigmas and variety in distribution of the sporogenous masses. XX 38. 26 Transverse section of stamen with two sporangia in the lower loculus and none in the upper. X 100. 27 Transverse section of a stamen with one sporogenous mass in the upper locu- lus and none inthe lower. X 100. 28 Transverse section of stamen with one sporogenous mass in the lower locu- lusand noneintheupper. X 100. 29 [ransverse section of stamen with one continuous sporogenous area. X 100. 30 Transverse section of the same anther at a level nearer the base, apparently showing two sporogenous masses. X 100. 31 Transverse section of anther showing continuous sporogenous mass across the anterior face of anther, while one sporangium in each loculus remains distinct. X 75. 32 Longitudinal section of staminodium, of about same age as the stamens shown in figs. 20-30. X 160. 33 Longitudinal section of flower and two bracts, showing relation of these structures and the complete submergence of the ovary inthe axis. X 40. STUDIES IN THE DEVELOPMENT OF THE PIPERACEAE 743 , LX hy Cf ss x4 t) COT LA 4 i oo™N OF Os 4 Johnson and Kellner, del. 744 DUNCAN 8S. JOHNSON EXPLANATION OF FIGURES 34 Longitudinal section of flower with one stamiinodium. X40. 35 Transverse section of tip of carpels showing three styles. X 75. 36 A lower section of the same ovary. X 75. 37 Transverse section of another ovary with four styles. X 75. 38-43 Series of successively lower transverse sections of an ovule and its integu- ments, showing the different height of the integument on different sides of the ovule. X 160. : 44 Longitudinal sections of flower with mature stamens and stigmas and with four-nucleate embryo-sac. XX 35. 45 Longitudinal section of flower, and two bracts, in which the pollen has been discharged from the single stamen. X 35. 46 Longitudinal section of flower with two stamens, well-developed stigmas, but no ovules in the cavity of ovary. X 40. 47 Section of another flower like last but with partially aborted styles and stig- mas. XX 30. 48 Section of similar flower but with no ovarian cavity and a mere spine in place of styles. X 40. 49 Section of similar (but younger) flower with still more reduced styles. X 75. 50 Longitudinal section of flower from male spike, with no trace of carpellary tissues evident. x 40. 51 Longitudinal section of flower, from middle of male spike, in which both tamens and carpels are rudimentary. X 75. 52 Longitudinal section of young ovary, with ovules showing inner integument. xX 350. 745 STUDIES IN THE DEVELOPMENT OF THE PIPERACEAE Johnson and Kellner, del. 746 DUNCAN S. JOHNSON EXPLANATION OF FIGURES 53 Longitudinal section of young ovule with two integuments, parietal cell already formed. X 350. 54 Longitudinal section of older ovule, with integuments completed and the megaspore mother-cell preparing for its first division. X 350. 55 Embryo-sac preparing for the second division, each spindle shows about eighteen chromosomes. X 600. 56 Section of a four-nucleate sac, showing a pair of nucleiateachend. X 750. 57 Section of a four-nucleate sac, showing the nuclei grouped near the middle. x 600. 58 Section of four-nucleate sac with one nucleus at the micropylar end: and three at the antipodal end. X 600: 59 Longitudinal section of nearly mature sac, in which the polars have not yet moved tothe middle. X 600. 60 Longitudinal section of nearly mature, somewhat abnormal embryo-sac. x 600. 61 Part of longitudinal section of integuments of an ovule witha nearly mature embryo-sac. XX 350. 62 Longitudinal section of sac, with fertilized egg, free endosperm nuclei, and 25 antipodals in the single section. X 170. 63 Part of a similar section, showing endosperm nuclei, and numerous anti- podals in a deep pocket at the base of the sac. X 180. 748 DUNCAN 8. JOHNSON EXPLANATION OE FIIURES 64 Part of longitudinal section of seed, showing one-celled embryo, free endo- sperm nuclei, antipodals, tapetum, inner integument and part of the perisperm. LO: 65 Transverse section of spike, showing number and arrangement of ripe fruits. x 10. 66 Longitudinal section of a mature fruit, showing structure of fruit and seed. x 40. 67 Part of transverse section of mature fruit. X 40. 68 Part of longitudinal section of innerintegument, showing structure. X 350. 69 Longitudinal section of a half-grown embryo. X 350. 70 Similar section of matureembryo. X 350. | 71 Part of longitudinal section of ripe seed, showing perisperm and per- sistent antipodal mass. X 170. IN THE DEVELOPMENT OF THE PIPERACEAE 749 STUDIES ~ or psn Ee B-d:, del. A COMPARISON OF THE SENSE-ORGANS IN MEDUSAE OF THE FAMILY PELAGIDAE ROBERT PAYNE BIGELOW Instructor in Biology and Librarian, Massachusetts Institute of Technology THIRTY-EIGHT FIGURES The work to be described in the following pages was done at the suggestion and under the supervision of Professor Brooks at Baltimore during the winters of 1888-89 and 1889-90 and dur- ing the summer of 1889 in the Fish Commission Laboratory at Woods Hole, where that summer Professor Brooks was director of the Laboratory and I the proud holder of the university table —my first appointment. The manuscript and illustrations were completed and handed to Professor Brooks in November, 1890, with the intention that the paper should forma part of the mono- graph on the medusae that he then was planning. That book has not been published. The point of view from which this paper was written is exhibited in the following sentences quoted from the opening paragraph of the manuscript of 1890: ‘‘In his System der Medusen (p. 504) Haeckel says, ‘The genera Ephyra, Palephyra, Zonephyra, Pelagia, and Chrysaora form five steps in a connected phyloge- netic process of development which is repeated at the present time in the ontogeny of Chrysaora according to the fundamental law of biogenesis.’ As indicated in my preliminary paper (1890), Dr. Brooks has pointed out that Dactylometra should be added to the series as the final form, and that, so far as the development 152 ROBERT PAYNE BIGELOW of these genera is known, each genus in the course of its ontogeny recapitulates in each successive stage the condition which remains permanent in each of the lower genera. The lack of a single important break in the series of adult forms is remarkable, and in Palephyra and the yet simpler Ephyra are found structures that show relationship to still more primitive forms.’’ It was my task to discover whether a study of the anatomy and develop- ment of the sense-organs would or would not confirm this general- ization. At the present day that point of view appears rather quaintly old-fashioned. Nevertheless it seems worth while to publish the paper because it contains an account of the anatomy and development of these interesting organs that 1s more complete than any hitherto published, so far as the species herein treated are concerned. And it seems especially appropriate that an article reflecting, as this one does, the thoughts and methods of Professor Brooks at the period of his greatest activity, should appear in the present volume. Since this paper was written the conditions of the problem have changed considerably. On the side of greater complexity a new genus, Kuragea, has been added by Kishinouye (’02), and on the other hand Vanhoffen (’02) suggests that Haeckel’s genera Ephyra and Zonephyra are not phylogenetic but ontogenetic steps and he unites all three into the single genus Palephyra. Mayer! has thrown similar doubt upon the mutual relation- ships of the genera Chrysoara, Dactylometra, and Kuragea. He thinks that the so-called Chrysaora of our coast is nothing more than an imperfectly developed Dactylometra. He retains the genus Chrysaora, however, for three imperfectly separable species, and my study of the sense-organs seems to indicate that the forms on our coast are really two species, and perhaps repre- sent two separate genera. Since 1890 our knowledge of the marginal sense-organs of the Scyphomedusae has been increased by Hesse’s detailed descrip- 1 Tam greatly indebted to Dr. A. G. Mayer for the gift of a set of the proof sheets > of his forthcoming work on the medusae and for permission to quote from them. SENSE-ORGANS IN MEDUSAE 4753 tion (95) of the sense-organs of Rhizostoma, in connection with which he refers to special points in the anatomy of these organs in Cotylorhiza; by Vanhdoffen’s (’02) descriptions and figures of the sense-organs of Atolla, Periphylla, Sanderia, and Dactylometra Africana; and by my account (’00) of the development of the rhopalia by metamorphosis of the larval tentacles in Cassiopea. Vanhoffen’s description of the sense-organs of Dactylometra is very brief, and neither he nor Hesse mentions the peculiar pitted lateral pockets that I find in the sensory niche of Chrysaora and Dactylometra, although Hesse finds in Rhizostoma a thickened area of sensory epithelium in a similar postition, thus confirm- ing Eimer’s earlier observation. In the following pages reference to the work of other investi- gators will be made as occasion may arise. For a general histor- ical review of the subject the reader is referred to the admirable summaries given by the brothers Hertwig (’78) and von Lenden- feld (82) and later by Hesse (’95). The material for this investigation consisted for the most part of preserved specimens furnished by Dr. Brooks includingsome young stages of Chrysaora collected by Professor F. 8. Lee. I am indebted to Mr. Austin Cary for some young material of Dactylometra collected, I think, at Newport. In view of Mayer’s observations it should be noted that when reference is made to Chrysaora the form found in the Chesapeake Bay is meant, and the name Dactylometra is used for the form found in the open sea. In the series Palephyra, Pelagia, Chrysaora, Dactylometra, and Kuragea, as generally understood, we have a series of medusae of similar form and structure. In all the umbrella is relatively flat and is provided at its margin with eight sensory clubs, or rhopalia, four perradial and four interradial. These genera may be distinguished briefly as follows. Pale- phyra has 8 tentacles, 16 marginal lappets and in the stomach are four interradial septal nodes, homologous with the septa of the scyphistoma larva (Bigelow, ’00). An account of the pos- session of these septal nodes Palephrya is taken by Vanhoffen (91) from Haeckel’s order Discomedusae and placed with the 754 ROBERT PAYNE BIGELOW Periphyllidae in a great group of Scyphomedusae, separate from the group containing the following genera, which lack the nodes. The genera Pelagia, Chrysaora, Dactylometra, Kuragea, and Sanderia constitute the family Pelagidae of Gegenbaur (Mayer). Pelagia is like Palephyra in having 8 tentacles and 16 mar- ginal lappets, but, as has been said, lacks the septal nodes. Chrysaora has 24 (3 xX 8) tentacles, 3 between each successive pair of marginal sense-organs, and 32 marginal lappets. Dac- tylometra has 40 (5 x 8) tentacles, 5 between each pair of sense- organs, and 48 marginal lappets. Kuragea ends the series of progressive steps in complexity in this line with 56 (7 x 8) ten- tacles and 64 marginal lappets. Sanderia is an aberant form in which all the marginal parts of Pelagia have been doubled, thus having 16 marginal sense organs, and likewise 16 tentacles sep- rated by 32 left marginal lappets. Unfortunately nothing is known in regard to the structure of the sense-organs of Palephyra, and there is some doubt as to how far this genus should be regarded as representing an ancestral type of the group of Medusae classed by Haeckel in the order Discomedusae. The present account of the sense-organs deals with those of Pelagia cyanella Péron et Lesueur, the Chrysaora-like form of the Chesapeake Bay, and Dactylometra quinquecirrha L. Agassiz. First, three successive stages in the development of the sense-organ of Chrysaora will be described, leading to a description of the adult condition. This will be followed by a comparison of the adult sense-organs of Pelagia with the Pelagia-stage of Chrysaora and with the adult. Finally these structures in the adult Dacty- lometra and in some stages of its development will be compared with the corresponding stages in the other species. SENSE-ORGANS IN MEDUSAE 455 THE EPHYRA STAGE IN CHRYSAORA The youngest specimen of Chrysaora that came under my ob- servation was an ephyra larva in which the tentacles were just beginning to bud. The eight arms of the ephyra are tipped each by two lobes and there is suspended from the under side, a little proximal to the notch between the lobes, a sense organ called “Sinneskolbe”’ or ‘‘Randkorper’’ by most German authors, the name rhopalium given to it by Haeckel is, however, the most con- venient for us and is the one [ shall use. Seen from below this appears to be a short club-shaped structure lying horizontally on the under side of the umbrella, fig. 21. In describing the adult form the Hertwigs very aptly compare it in shape to a bent finger. This applies equally as well here (compare figs. 2 and 22). The part by which the rhopalium is attached is perpendicular to the oral surface of the umbrella. The other part extends at right angles to the first away from the mouth and in its distal end is a conspicuous cluster of crystals, fig. 21. The tip of the sense-organ reaches to the edge of the notch between the lobes. The lobe of the stomach that penetrates each arm has about one-third the width of the latter. At the base of the rhopalium it narrows into a small tube, fig. 22, r.c., which bends downward and extends into the rhopalium as far as or a little beyond the angle, where it ends blindly. I shall speak of this tube as the rhopalial canal. There is no extention of the gastric pouch into the lobes of the arm, as has been shown already in the figures given by Agassiz, Claus, and the Hertwigs, and this seems to be generally true of this larval stage. The endodermal lamella, fig. 25, e./., is plainly visible along the side of the gastric pouch, which is more or less triangular in cross section. The lamella extends from the lower angles of the pouch obliquely downward and soon joins the ectoderm and does not extend to the margin of the arm. | The endoderm gradually becomes thickened towards the sense- organ until where the rhopalial canal bends downward the cells are deeper than broad. The distal end of the rhopalium is com- pletely filled with large thin-walled endoderm cells, each of which 756 ROBERT PAYNE BIGELOW secretes a concretion in its interior, fig. 22. These concretions are soluble in weak acids and vary in size and shape, but are generally prisms not more than twice as broad as long with conical ends and they seem to contain a core of organic matter. There is no sharp line of distinction between the cells which produce these bodies and the other endoderm cells of the rhopalium, and for Von Lendenfeld to call this cluster of cells the ‘‘ visceral mesoderm”? is rather stretching analogies. The general ectoderm of the body is composed of very much flattened cells a little thickened in the position of the nucleus. In spots over the surface, the cells are much thicker so that they appear square in vertical section. These are the young batteries of nettle cells, figs. 22 and 26, b. At the base of the rhopalium the ectoderm is abruptly very much thickened and forms a deep covering to all but the most distal part of it, fig. 22. In this thickened part the cell walls are hardly to be distinguished; the nuclei are thickly crowded in two or more irregular rows and between this layer of cells and the thin supporting membrane there appears a thin layer of nerve fibers. This thick ectoderm thins out at the distal extremity of the rhopalium until in my prepara- tions it cannot be recognized, but it is undoubtedly there as a very thin membrane. The thin supporting membrane which every- where separates the ectoderm from the endoderm is a continua- tion of the general mesogloea, which is structureless except for a few fibers running through it. The sense-organ of Chrysaora at this stage then, is quite simple; there is no trace of an olfactory groove, and there are no other organs accessory to the rhopalium. THE PELAGIA STAGE IN CHRYSAORA After the ephyra the next youngest stage of Chrysaora that I have studied is a larva six millimeters in diameter. The begin- nings of the gonadia have not yet appeared, but the animal has eight tentacles, hollow at least for along distance. The pendant oesophagus is rather short, but there are four oral lobes which when spread out extend more than halfway to the edge of the disk, and there are two endodermal pockets into each of the mar- ; ; SENSE-ORGANS IN MEDUSAE (57 ginal lappets. This then may be regarded as in the Pelagia stage. The rhopalium at this stage is twice as large as in the preceding one and the distal horizontal part is longer in proportion to the proximal vertical part, fig. 26. It lies in a small cavity generally known as the sensory niche (Sinnesnische) s. n., figs. 26 and 28, which is not at all developedinthe ephyra. This is bounded above by the hood, figs. 26, 27, and 28, h, laterally, by the sides of the marginal lappets; and proximally, by that part of the subum- bral wall that rises to pass into the hood. The niche becomes gradually shallower proximally; distally and below, the niche is open, except that the free edges of the marginal lappets are folded together so as to convert the part of the niche immediately sur- rounding the rhopalium into a tube such as is figured by the Hert- wigs. There isa low, flat ridge, figs. 26 and 30, 7.7, extending along the slanting roof of the niche from the proximal wall outward about halfway to the edge of the hood. The rhopalium hangs from this ridge and its base covers more than half of it. The ridge contains the rhopalial canal which sends a little conical diverticulum into a small part of the ridge that extends beyond the base of the rhopalium, fig. 26, r.c. The canal penetrates the rhopalium for about half of its length, considerably farther than in the ephyra. The endoderm at the mouth of the rhopalial canal begins to thicken and farther inward becomes columnar. These columnar cells distally fill the whole lumen of the tube and they grade off into the larger otolith cells of the extremity. These are about the same size as in the ephyra, but are more numerous, and as in the ephyra the endoderm is separated from the ectoderm throughout the rhopalium by only a thin supporting membrane. The general ectoderm of the body, including that lining the sensory niche, is flat as in the ephyra, but the thickened spots (b) are more developed and the nettle cells are clearly differen- tiated in them. The ectoderm on the rhopalium is twice as thick as in the ephyra and grades distally, as in that stage, into a thin membrane covering the outer end. The gradation is more abrupt on the upper than on the lower side. 758 ROBERT PAYNE BIGELOW Around the base of the rhopalium this very much thickened ectoderm passes into the general flat epithelium of the body, except at two points. These are at sides of the rhopalial ridge, fig. 30, and here the rhopalial ectoderm is continuous with two folds of thickened ectoderm which run on each side of the ridge centrally to the proximal wall of the niche, fig. 31. The sensory epithelium of the rhopalium, figs. 26 and 28, is of the same character as that in the ephyra and the cells are pro- vided with comparatively long cilia. In this stage the layer of nerve fibers is very well marked; in the deepest part it is as thick as the cellular layer. This thickened portion of the nerve fiber layer describes a U-shaped figure. The loop of the U lies on the upper side of the rhopalium in the most distal part of the nerve layer, fig. 28, and the limbs of the U extend along the upper part of each side of the rhopalium and each one is continuous with a thin layer of nerve fibers underlying the before-mentioned lat- eral thickenings of the rhopalial ridge. The ectoderm in these latter areas is composed of very small cells with very long cilia. In section the layer of cells seems to be composed of a mass of crowded nuclei. Itis probably a single layer of cuboidal or slightly elongated cells and it is folded so that there is an wnvaginated groove at each side of the rhopalial ridge which extends to the proximal end of this thickened ectoderm on the wall of the niche. The groove runs along the upper edge of the layer and in addi- tion to it there are a number of shallow secondary pits. On the proximal wall of the niche the lateral folds lie in the plane of the endodermal lamella and are apparently in contact with it, fig. 32. All along the rhopalial ridge there is a line opposite the bot- tom of the principal fold and continuous with the endodermal lamella in which the endoderm and ectoderm are in contact. This lamella not only connects the endoderm of adjacent gastric pouches but also connects the endoderm with the ectoderm all around the margin of the umbrella, except where the jelly is too thin to permit it, and it is remarkable that in all forms that I have studied there is some point where this lamella comes into contact with the nerve-fiber layer. fl - ; - “a SENSE-ORGANS IN MEDUSAE 759 THE BEGINNING OF THE ADULT CONDITION With the appearance of the tentacles characteristic of the Chrysaora, or shortly after, the foundations of the other adult structures are established. The description of this initial stage will be taken from a specimen 10 mm. in diameter, in which the second set of tentacles reached to about the tips of the lappets. Although a good deal larger, the rhopalium and adjacent parts are about in the same proportion as in the preceding stage, com- pare figs. 26 and 33. The endoderm cells of the rhopalium are more narrowly colum- nar than in the last stage and the lumen of the canal reaches to the concretion-forming cells. The transition from the colum- nar cells to these is rather abrupt. These cells have increased in size and number and the whole mass of concretions is nearly spherical, fig. 33. The ciliated ectoderm of the rhopalium has increased still more in thickness and the nerve-fiber layer is still more marked than in the last stage. There has been but little change in the folds at the sides of the rhopalial ridge, yet it is in the ectoderm that the most important changes have taken place. In the first place, the epithelium of the lower surface of the rhopalial ridge, which was slightly thickened in the pelagia stage, is now very much more thickened, the cells becoming cuboidal or almost columnar, while the epithelium of the roof of the sensory niche adjoining the rhopalial ridge has suffered a similar change. But the chief step in advance is the formation of the “olfactory groove,” or, as von Lendenfeld prefers to call it, ‘‘the dorsal sen- sory groove.’ This is a shallow saucer-shaped depression in the exumbral surface of the hood just above the base of the rho- palium, figs. 33 and 34, s.g. In this groove the ectoderm is consider- ably thickened, being composed of a single layer of columnar cells. These cells are deepest in the deepest part of the groove. In specimens of about the same size as the one just described but in which the evaginations that are to form the second set of ten- tacles are not longer than they are broad, a shallow dorsal groove occurs, but it is clothed simply with the ordinary flat epithelium. 760 ROBERT PAYNE BIGELOW THE ADULT CHRYSAORA Turning now to the fully formed adult, figs. 2 to 10 (fig. 1 is not fully adult), we find no structure not represented in the stage just described, but there are marked changes in form and propor- tion and there is much greater histological differentiation. The size of the animal and of all of its parts has greatly in- creased, the general ectoderm and endoderm have become more marked, the cells being thicker in proportion to their width, the nettle cells and gland cells are fully developed, and the general topography in the region of the sense-organ is very much height- ened, fig. 2. On looking down upon the upper side of the umbrella, one notices in the hood covering each rhopalium an elliptical area free from the nettle batteries that now form thickly set mounds over the rest of the surface. These elliptical areas are the dorsal sen- sory grooves, an early stage of which has just been described. Each groove on closer examination is seen to be now a funnel- shaped cavity, the apex of the funnel extending deep into the mesogloea to a point opposite the base of the rhopalium, fig. 6. At the edges of the groove the common cuboidal epithelium of the exumbrella grades into a deep columnar epithelium that lines the groove. The cells of the epithelium are many times deeper than they are wide and are ciliated, and at the base of the layer next to the mesogloea there is a thin stratum of nerve fibers. This epithelium is probably of that kind of sensory epithelium com- mon to jelly fish which has been carefully described and figured by Eimer (’77), Claus (77), the Hertwigs (78), and Schewia- koff (89). There are present some mucous gland cells like those found by Wilson (’88) in Manicina. The mucus granules stain so deeply with haematoxylin that the mass of mucus might easily be mistaken for a nucleus, but they do not stain with carmine. The surface of the groove, being perfectly even, shows no trace of such complications as are described by Claus in Aurelia, by von Lendenfeld in Cyanea and Crambessa, and by Hesse (’95) in Rhizostoma. SENSE-ORGANS IN MEDUSAE 761 If the animal now be looked at from the under side, the most conspicuous part of the sensory apparatus is the rhopalium, figs. 1, 2, and 10,7., which has the same shape that it had in the pre- vious stage, that of a thick bent finger, and it has not increased in size in proportion to the surrounding parts. The sensory niche in which the rhopalium lies is very much deeper than in the pre- vious stage and the free edges of the marginal lobes are approxi- mated so as to form a tube which extends from about opposite the base of the rhopalium outward to the edge of the hood. The rhopalium is attached as before to a ridge which extends centrally along the arched roof of the niche from a point just distal to the base of the rhopalium to the proximal wall of the niche. But now the base of the rhopalium covers but a very small part of the ridge near its distal extremity, and the proximal part is relatively, as well as actually, very much larger than before, figs. 1, 2, and 10. The ridge gradually becomes wider and thicker as it recedes from the rhopalium, figs. 1 and 2, and 5 to 8. Its under surface is convex and it is grooved on its sides. These lateral grooves are continued into two pocket-like cavities which lie on each side of the ridge in the upper proximal wall of the niche, figs. 1, 2, and 8. The rhopalial canal becomes gradually narrower from its mouth outward to the base of the rhopalium, it is slightly dilated in the proximal part of the latter and extends some distance into the distal part where it ends as a narrow pocket. In a section of the mass of concretions it appears as a narrow vertical slit, figs. 1 and 3. The canal sends off a very small cone-shaped diverticulum into the hood above the rhopalium. With this heightening of the general topography comes a cor- responding increase in the importance of the various histological features. The thickening of the ectoderm, which in the last stage extended a short distance around the base of the rhopalium, has now spread so as to cover the whole surface of the sensory niche, including all but a small part at the base of the rhopalium of the convex under surface of the rhopalial ridge, and has come to form apparently a sensory epithelium. It extends in all direc- tions from the base of the rhopalium, laterally nearly as far as 762 ROBERT PAYNE BIGELOW the bases of the marginal lappets, distally as far as the extremity of the hood, and proximally to the edge of the subumbral mus- cle layer, a very small part of which comes within the hollow of the niche. The thickness of this layer of cells varies somewhat. The ectoderm layer is thickest on the proximal wall of the niche and two thickened areas extend outward along the sides of the marginal lappets midway between the rhopalial ridge and the free edges of the lappets, gradually thinning out at a point beyond the margin of the hood. At the edge of the muscle layer this epithelium of the niche passes rather abruptly into the cells cov- ering the muscles, and the columnar epithelium extends some dis- tance outward between the edge of the muscle layer and the endo- dermal lamella on the subumbral wall of the marginal lappets, fig. 3. The free edges of the marginal lobes are covered by asim- ple slightly flattened epithelium. ‘The sensory epithelium of the niche grades gradually into this. Surrounding the base of the rhopalium on its distal and lateral sides, the epithelium is only about half as thick as it is in the deepest part of the niche. Where the sensory epithelium of the niche is deepest the cells are of about the same proportion as in the dorsal sensory groove, perhaps somewhat longer. These cells as seen in sections are long and columnar, each with a nucleus in the lower third. Below the nucleus the protoplasm is clear, above it, it is granular, and in the upper third of the cell there is often the characteristic struc- ture of a mucous cell. The mucous cells, which are here very abundant, are also found in the general ectoderm and endoderm, and particularly in the dorsal sensory groove (as already stated) and in the endo- derm of the rhopalial canal. In each place the goblet cell is of the same length as the adjoining cells and everywhere, except in the dorsal groove and the sensory niche, the globule of mucus nearly fills the whole cell. These cells seem, however, to be absent from the epithelium of the under side of the rhopalial ridge and from that lining the niche immediately proximal to it, so that in a longitudinal section through the rhopalium they are not seen, fig. 12. Scattered nettle cells are found also in the epithelium of [ SENSE-ORGANS IN MEDUSAE 763 the niche and rarely I have come across one in a section of the rhopalial canal. Cyanea annaskala presents, according to von Lendenfeld, a precisely similar arrangement of the epithelium of the niche, except that in this species there is a peculiar sensory apparatus on the dilated proximal part of the rhopalial ridge which can be compared to nothing in the Pelagidae. Von Lendenfeld has also found in Crambessa a pair of rounded thickenings containing sub- epithelial ganglion cells which correspond in position with the areas of deep columnar epithelium that I have described as extend- ing outward along the sides of the niche. Von Lendenfeld homol- ogizes these with the cone-shaped thickenings found by Claus in Aurelia and Chrysaora, but they seem to be something entirely different, as will appear later. The cellular covering of the extremity of the rhopalium can now be clearly seen to be a simple, slightly flattened epithe- lium. It grades, more gradually on the lower than on the upper side, into the layer of columnar cells and nerve fibers which cov- ers the main part of the rhopalium, fig. 10. This is similar to the sensory epithelium that has been found by Eimer, Claus, Schafer, and Schewiakoff in Aurelia, by the Hertwigs in Pelagia, by Schewiakoff in Carybdea, by Hesse in Rhizostoma and by Ver- hoffen in several species. It is a deep ciliated epithelium of slen- der cells with the nuclei placed irregularly in several rows. There are numerous straight fibers extending from the cellular layer through the thick felted nerve-fiber layer to the supporting mem- brane. These are processes of cells described in Schewiakoff’s paper (’89) and regarded by him as supporting cells. On the distal side of the base of the rhopalium its epithelium passes into the ordinary epithelium of the niche. On the proximal side it grades into a peculiar epithelium that forms a structure of which we found the rudiments in the Pelagia-stage, fig. 32. This epithelium, like that on the rhopalium, overlies a layer of nerve fibers. It is at this stage clearly a single layer of short cuboidal cells provided with very long cilia, fig. 11. The nuclei are of the same size and appearance as in the sensory epithelium of the rhopalium and nearly fill the cell. The layer is very much = a & | 764 ROBERT PAYNE BIGELOW folded and pitted. The mesogloea, however, takes no part in this folding, but the nerve fibers extend outward between the cells which line adjacent pits. This folded epithelium covers a small part of the lower surface of the rhopalial ridge (fig. 10) and extends along the lateral grooves into the pockets that have been mentioned as being sunk into the mesogloea at the sides of the mouth of the rhopalial canal. The pits are thickly set and in the lateral pockets they are very deep and fill nearly the whole of the pocket. The lumen of each pocket opens into the fundus of the niche on one side of the ridge, figs. 2 and 11. The pocket is much wider horizontally than vertically and is deeper than the proxi- mal wall of the niche, so that its apex occupies a small prominence in the roof of the gastric pouch at the sides of the mouth of the rhopalial canal, figs. 8 and 9. ) This pitted ectoderm is separated from the endoderm by only a thin supporting membrane for the whole distance from the base of the rhopalium to the apex of the pocket. From this point the mesogloea thickens and then the endodermal lamella appears and continues in contact with the pocket around to its adradial ex- tremity,fig.8. I cannot, however, discover any protoplasmic con- nection between the cells of the lamella and those of the pocket. The nerve-fiber layer which lines the pocket and underlies the whole of the pitted epithelium is of the same character as the nerve layer of the rhopalium, being a felted mass of extremely fine fibers, and is directly continuous at the base of the rhopalium (fig. 11) with the limbs of the U-shaped thickening which I have described in the previous stages. This thickening is now very prominent on the rhopalium and the membrane beneath it is thickened for its support, fig. 5. The pitted epithelium of the lateral grooves and pockets is, of course, derived from the lateral folds of ectoderm of the previous stage. Structures of the same kind were probably seen by Eimer and certainly were by Claus in Cyanea and in Aurelia, respec- tively. Eimer (’77) says that in Cyanea the ectoderm surround- ing the rhopalium forms numerous conical ingrowths. Claus (77) in speaking of Aurelia says that ‘‘there is found at the base SENSE-ORGANS IN MEDUSAE 765 of the marginal body in the sensory niche a pair of cone-shaped swollen thickenings of the ectoderm which enclose under the epithelium a thick layer of ganglion cells and nerve fibers.”’ He says, moreover, that the nervous system in Chrysaora has the same general structure as in Aurelia. These statements appar- ently refer to the structures that I have described, but fail to give a correct idea of them. As already stated, von Lendenfeld found in Crambessa two prominences on the sides of the niche which he homologizes with the cone-shaped thickenings of Claus. But as he distinctly says that they are not invaginations, they must be quite different from the organs to which I suppose Claus alludes. Nothing need be added concerning the histology of the endodermal parts of the rhopalium at this stage except that the cells have increased in depth. THE ADULT PELAGIA The brothers Hertwig have given in their work on the nervous system and sense-organs of the medusae (’78 a, p. 109) a very clear account of the position and structure of the rhopalia in Pelagia noctiluca (Pér. Les.) and have described their development from the ephyra stage. This agrees in the main with what I have said of the rhopalium of Chrysaora except that the rhopalial canal in the adult does not penetrate the mass of concretions. The ephyra of Pelagia agrees inevery essential particular with the same stage inChrysaora. As in Chrysaora at this stage, the gastric diverticulum does not pene- trate the rhopalium but its interior is filled with endoderm cells, and the rhopalium is not surrounded by a sensory niche. The most important events which take place in the development from this to the adult stage are, according to the Hertwigs, the hol- lowing out of the rhopalium, an increase in the number of con- cretions, and the formation of the sensory niche by the outgrowth of the hood and free edges of the marginal lobes. I have studied only the adult sense-organ in Pelagia cyanella (Pér. Les.) and can confirm the description of the adult given by 766 ROBERT PAYNE BIGELOW the Hertiwgs, so far as it goes. I have noticed additional feat- ures which, however, may be peculiar to the species studied. On the upper surface of the hood immediately above the base of the rhopalium there is a funnel-shaped dorsal sensory groove, fig. 13, s.g. It differs from the groove in Chrysaora in being much smaller and in sinking into the mesogloea not more than half so far. Hesse (’95) confirms Eimer’s statement that a similar groove is to be found in P. noctiluea, but found no sensory epithelium in it. While, as stated by the Hertwigs, the general surface of the niche is covered by the common ectoderm of the body, I find on the rhopalial ridge, which is short and low, fig. 15, a peculiar epithelium which distally passes into the ectoderm of the rho- palium and proximally extends along the wall of the niche towards the muscle band. My material is not sufficient for me to make out very clearly the structure of this layer. It is twice the height of the ordinary epithelium. The outer part of the layer is made up of the thicker parts of the cells, which stain deeply, while the inner portion seems to be composed of processes running from the cells to the surface of the mesogloea. There seems to be a loose network of nerve fibers intermingled with these processes. At any rate, my preparations show very clearly a number of large ganglion cells scattered through this layer just below the deeply stained part, fig. 16,g. When I speak of these cells as large, I mean that they are many times larger than the ordinary epithelium cells. They are provided with comparatively large nuclei and are appar- ently bipolar, the stout processes running parallel to the rho- palium. Cells of this kind similarly situated have been found by Hesse (’95) in Rhizostoma, and, as he says, probably constitute a nerve center in each sensory niche. In the thick layer of nerve fibers on the basal portion of the rhopalium there is a cluster of nuclei like those in the sensory epithelium above them, fig. 14, n. These probably belong to very small ganglion cells. At each side of the ridge at the base of the rhopalium the layer of nerve fibers comes into contact with the endodermal lamella, and the latter, which otherwise has the same relative position as in Chrysaora, may be traced in this species SENSE-ORGANS IN MEDUSAE 767 along each side of the rhopalial canal nearly to the tip of the diver- ticulum into the hood, fig. 14. Aside from the rhopalium and the band of peculiar epithelium and nerve fibers there is no other sensory apparatus in the niche. The sense-organs in Pelagia are then much simpler than in the adult Chrysaora, but they have many points of resemblance to the Pelagia-stage in the young Chrysaora. In both the sensory niche, while well developed, is not so prominent as in the adult Chrysaora and the greater part of it is lined with the ordinary surface epithelium. It is probable that in the larval as well as the adult form there is a thickening and a differentiation of the epithelium upon the lower surface of the rhopalial ridge. The latter in both is of about the same proportion, being short and but slightly raised, and in both the rhopalial canal extends only to the mass of concretions and does not penetrate it. The shape of the rhopalium in Pelagia differs from its shape in the young Chrysaora, as well as in the adult, in that the part which contains the concretions has a considerably smaller circumference than the part covered by the sensory epithelium, and the concretions them- selves are longer in Pelagia in proportion to their width. But the important points in which Pelagia differs from the corresponding stage in the development of Chrysaora are two: the presence of the dorsal sensory groove, and the absence of any fold of the ecto- derm at the side of the rhopalial ridge. Compare figs. 15 and 31. It will be noticed, however, that the line along which the endo- dermal Jamella touches the ectoderm at the sides of the rhopalial ridge in Pelagia (fig. 14, e.l.) coincides exactly with the position of the deepest part of the lateral folds in the Pelagia-like larva of Chrysaora. THE ADULT DACTYLOMETRA The sense organs of Dactylometra quinquecirrha differ from those in Chrysaora, as we would expect, in the opposite direction from Pelagia. While that genus lacks some of the features of Chrysaora and has a simpler sensory apparatus, the adult Dac- tylometra possesses all the characteristics of Chrysaora in an exaggerated degree, figs. 17 to 20. 768 ROBERT PAYNE BIGELOW The rhopalium is of about the same shape and size as in Chry- saora. It is clothed with the same kind of sensory epithelium overlying a layer of nerve fibers as thick as the layer of cells or thicker in the U-shaped areas. The supporting membrane is thickened under this area and the rhopalial canal extends into the mass of cells that contain the concretions. In short, the rhopalia in the two species are alike in every particular. (It is only inPelagia that I find any trace of ganglion cells in the nerve- fiber layer of the rhopalium.) The parts surrounding the rhopa- lium are, however, much larger than in Chrysaora. The meso- gloea of the hood is very much thicker and the dorsal sensory groove is proportionally deeper. It is a little longer than in Chrysaora but no wider, figs. 17 and 18. The sensory niche is also deeper, the lateral pockets are larger, and the rhopalial ridge is more prominent. . It is in the latter that we find the most characteristic differences between Chrysaora and Dactylometra. Immediately at the base of the rhopalium the ridge is covered by ashallow p:tted epithe- lium with its layer of nerve fibers, as in Chrysaora, fig. 18. Pros- imally this soon becomes confined to the sides of the ridge while the area between is clothed with a single layer of small cuboidal cells. At about half the length of the ridge two small elevations appear, one on each side of this simple epithelium, fig. 19. The concave area between them gradually widens until it joins the proximal wall of the niche. The pitted epithelium covers a much greater part of the surface of the ridge and is much more highly developed than in Chrysaora. It gradually spreads out on to the roof of the niche and its pits become deeper and more numerous as it recedes from the rhopalium. At its outer edges the change into the ordinary epithelium of the niche is quite abrupt. A com- parison of figs. 7 and 19 will show these characteristic differences in the two species. With the increase in size of the lateral pockets there is an increase in the depth and number of the epithelial pits. These are branched and closely crowded so that they fill the whole pocket, obliterating its lumen, and the orifices of the pits open directly to the exterior at its mouth; not well shown in fig. 20. SENSE-ORGANS IN MEDUSAE 769 Except the endodermal lamella, there are no cellular elements imbedded in the mesoglcea in the neighborhood of the sense-organ in either of the three species under consideration. There are many fibers in the mesogloea, especially in Dactylometra, but they are apparently only connective tissue fibers serving to give firmness or elasticity to the jelly and there is no evidence that they have any connection with the nervous system. Schiéfer and Schewiakoff both speak of the connection between the layer of nerve fibers in the dorsal sensory groove and the nerve fibers in the niche, and Hesse finds in Rhizostoma fibers extending from the fundus of the dorsal sensory groove to the endoderm of the rhopalial canal, and he regards these as nerve fibers, in spite of the fact that they would not stain with methylene blue or gold chloride. Jf there is such a connection in the Pelagidae :t is not through the mesogloea. In fact, the mesogloea immediately surrounding the sensory groove in Dactylometra seems to be perfectly structureless, although the fibers are so well developed in other parts of it. SOME LARVAL STAGES OF DACTYLOMETRA The questions that now remain to be answered in this paper concern the stages in the development of Dactylometra. Unfor- tunately my material for this purpoes is very scant. The youngest of some larvae taken near Newport that I sup- pose to be Dactylometra is a little less than 2mm. broad. It has four gastric filaments, the tentacles are just budding and the tentacular pouches of the stomach are about half as long as the rhopalial ones, while from the latter the marginal pockets are beginning to grow into the ephyra lobes. This specimen may, therefore, be regarded as on the border line between the Ephyra- and the Palephyra-stage. Sections of the rhopalium, fig. 35, show little differences between this and the slightly earlier stage of Chrysaora that I have already described. The constriction at the end of the rhopalial canal is less marked and the distal part of the rhopalium may be relatively a little longer. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 9, NO. 4. oS, ee ois + 770 ROBERT PAYNE BIGELOW In a little older one of these specimens the rhopalium. fig. 36, is of the same character as in the Pelagia stage of Chrysaora but there is only the merest rudiment, if any trace at all, of the lateral ectodermal thickenings. It is rather difficult to compare this larva with any of Haeckel’s genera. It is 3 mm. in diameter, has a simple quadrate mouth, eight gastric fila- ments, one in each pair being shorter than the other, and the tentacular pouches are a little shorter than the rhopalial ones. The pockets into the marginal lobes from the rhopalial gastric pouches are hardly large enough for a Palephyra and the pockets from the tentacular pouches show an approach to a Zonephyra- stage. The short tentacles like the buds in the younger stage are solid for the greater part of their length. This is the specimen that I spoke of in my preliminary paper as one just passing into the Pelagia stage. I have no specimens of Dactylometra in the Pelagia-stage; all of my remaining larvae have acquired the rudiments of the second set of tentacles, and are therefore in the Chrysaora-stage. It is evident that, as in Chrysaora, the first trace of the dorsal sensory groove appears with the second set of tentacles. In a specimen 18 mm. in diameter in which the tentacles of the second set are about 3 mm. long, the sensory groove may be seen as a flat disk of cuboidal or columnar cells lying directly above the base of the rhopalium, fig. 37. The sensory niche is broad and shallow and is lined with the ordinary flat epithelium of the body. The rhopalium shows all the characters of that organ in the adult Chrysaora; the rhopalial canal penetrates the distal part in the same way, the layer of nerve fibers and the supporting mem- brane show the same thickenings and in my sections Schewi- akoff’s supporting cells can be seen very clearly. The rho- palial ridge is very short and is completely covered with the thick pitted epithelium, fig. 37; this is continued into the rela- tively very large lateral ectodermal pockets and lines of the floor each pocket and fills its inner half, fig. 38. The roof of the pocket is lined with the ordinary flat epithelium of the niche. At the edge of the pockets the epithelial pits are wide and compara- tively shallow, evidently in the process of formation; the layer SENSE-ORGANS IN MEDUSAE i@ of nerve fibers is, however, most noticeable at this point. A thickening of this layer seems to run along the under lip of the pocket. The tips of the pockets extend each one into a prominence on the dorsal wall of the gastric pouch at the side of the rhopalial canal, just as in the adult Chrysaora. In a specimen about 26 mm. in diameter in which the tenta- cles of the second set are over a centimeter long the dorsal groove is a saucer-shaped depression lined in its deeper parts by a dis- tinctly columnar epithelium. The sensory niche is also deeper, the rhopalial ridge longer, and the lateral pockets more nearly filled with the vitted epithelium, which in turn has a much less embryonic appearance than in the younger specimen. These specimens should be compared with young Chrysaoras of about the same size (fig. 1) rather than with the younger ones (figs. 33 and 34) that I have described as in the beginning of the adult stage. A Chrysaora 25 mm. in diameter has its gonadia formed, although they are quite small, and is otherwise inter- mediate between the beginning of the adult condition and its completion. In the young Chrysaora of 25 mm. (fig. 1) the thick- ened epithelium does not cover the rhopalial ridge but forms a simple band on each side until it reaches nearly to the mouth of the pocket, where the first pits appear. The pitted epithelium lines the angle of the pocket and fills one-half or two-thirds of it. The pockets are smaller and the pits less numerous than in the Dactylometras 18-25 mm. in diameter, described above, and, moreover, there is no extension of the pitted epithelium outward along the floor of the pocket, which is lined by the same kind of epithelium as the roof. Dr. Brooks has figured a young Dactylometra which has its third set of tentacles just budding. This specimen is 70 mm. in diameter and is therefore considerably older than my specimens, which are probably in the first half of the Chrysaora- stage and my comparison of them with the condition in Chry- saora of about the same size is the proper one. It would appear, then, that in the Chrysaora-stage of Dactylometra the structures in the sensory niche are more advanced, while the dorsal groove is, 1f anything, less advanced than in the same stage in Chrysaora. tt2 ROBERT PAYNE BIGELOW SUMMARY 1. The eight-armed ephyra of Chrysaora has between the lap- pets at the end of each arm a rhopalium not enclosed in a sensory niche. The rhopalium is covered by a columnar epithelium, over- lying a layer of nerve fibers. Within it contains atits distal end a mass of otoliths enclosed in endodermal cells that are continu- ous with the columnar epithelium lining the rhopalial canal, which is continuous with the gastric pouch of the arm. From the sides of the gastric pouch an endodermal lamella extends through the mesoglcea to the ectoderm. 2. In the Pelagia stage of Chrysaera, 6 mm. in diameter, the rhopalium is enclosed in a sensory niche, covered by a hood, and is attached to the underside of the hood. From the point of attach- ment a rhopalial ridge runs centrally to the proximal wall of the niche. On each side of the rhopalial ridge there is an invaginated fold of epithelium consisting of small, ciliated, cuboidal cells and covering a layer of nerve fibers that is continuous with the layer of nerve fibers upon the rhopalium. This epithelium contains shallow secondary pits, and at some point beneath it the layer of nerve fibers comes into contact with the endodermal lamella. 3. At the beginning of the Chrysaora-stage of Chrysaora when the animal has reached a diameter of 10 mm. the dorsal sensory groove makes its appearance as a shallow depression in the exum- brella over the base of the rhopalium, but is still lmed by flat epithelium like that of the general surface. 4. Inthe adult Chrysaora the dorsal sensory groove has become a deep conical depression and is lined by columnar, ciliated epi- thelium with an underlying stratum of nerve fibers. The topog- raphy of the sensory niche is heightened and the cuboidal pitted epithelium of the sides of the rhopalial ridge extends on each side into a deep lateral pocket that projects centrally beyond the open- ing of the rhopalial canal so that its fundus occupies a ridge on the roof of the gastric pouch. In this area the layer of nerve fibers comes into close contact with the endodermal epithelium as it goes along the whole extent of the rhopalial canal. These pockets SENSE-ORGANS IN MEDUSAE iio are probably similar to structures found by Eimer and Claus in Cyanea and Aurelia. 5. The adult Pelagia cyanella possesses a shallow dorsal sen- sory groove. The rhopalium is similar in all important respects to that of Chrysaora and is attached to a short and low rhopalial ridge which is covered by a columnar epithelium overlying a layer of nerve fibers, among which are large bipolar ganglion cells. In the nerve-fiber layer of the rhopalium are found nuclei probably belonging to small ganglion cells. There is no pitted epithelium at the sides of the rhopalial canal nor lateral pockets, and there is no thickening of the epithelium of the niche. The endodermal lamella may be traced in contract with the ectoderm of the side of the rhopalial ridge to the base of the rhopalium. 6. A comparison of the Pelagia-stage of Chrysaora with the adult Pelagia shows that both possess in each of the eight principal radii a rhopalium lying in a well developed sensory niche, which is lined for the most part by undifferentiated epithelium. Both have a thickening of the epithelium on the surface of the rhopalial ridge. They differ in that Pelagia has no pitted epithelium while in this stage Chrysaora has not yet acquired a dorsal sensory groove. If Hesse be right in saying that Pelagia likewise has no true dorsal sensory groove, the difference between the two forms becomes just so much less. But if it be not sensory, it is difficult to understand the presence of this persistent dimple in two species of Pelagia. 7. The adult Dactylometra presents in its sensory apparatus all the characters of Chrysaora in an exaggerated degree. The pitted epithelium covers a greater surface and is more highly developed. It fills the whole of the lateral pockets with closely packed branching tubules. 8. Of the larval forms of Dactylometra, the ephyra is essen- tially like the same stage in Chrysaora. The Pelagia stage has not been studied. A larva in the Chrysaora-stage, 18 mm. broad with the secondary tentacles 3 mm. long, shows the rudiment of a dorsal sensory groove consisting of a flat disc of columnar epi- thelium. The rhopalium is like that of the adult Chrysaora. But the rhopalial ridge is very short and completely covered by the 144 ROBERT PAYNE BIGELOW pitted epithelium which extends centrally into very large lateral pockets, where it covers only the floor of the pocket. In the dis- tribution of the pitted epithelium and the form of the lateral pockets the Chrysaora-stage of Dactylometra differs from Chry- saora of the same size. 9. Sections of the rhopalia of all of these forms fail to afford any evidence for Hesse’s theory of the intercellular origin of the otoliths. On the contrary, they appear to be formed within the cells, instead of between them. It is true, as Hesse says, that there is an unbroken gradation between the otolith cells and the columnar, endodermal epithelium of the rhopalial canal. The same gradation is found between the chorda-cells andt he colum- nar epithelium at the base of the solid tentacles from which the rhopalia are developed in the scyphistoma (see Bigelow, ’00, pl. 37, figs. 53-56), but it is not evident what bearing this fact has upon the probable origin of the otoliths. 10. The Pelagidae show no layer of nerve fibers underlying the endodermal epithelium and nerve fibers of the dorsal sensory groove do not appear to penetrate the mesoglceea. In this region ~ the mesogloea of Dactylometra appears to be quite structureless, although fibers are abundant in it elsewhere. There may be a nervous communication between the nerve centers in the sensory niche and the endoderm through the endodermal lamella, but if present its demonstration will require special methods of stain- ing, which were not employed in the present investigation. 11. In considering these results due allowance must be made, of course, for individual variation and for liabliity to error due to shrinkage and distortion of specimens. There is also a possi- bility of confusing larval forms when they are not reared from the egg. Still I think that in all essential particulars the results given above will be confirmed by future investigations. SENSE-ORGANS IN MEDUSAE Wi5 BIBLIOGRAPHY (Arranged chronologically) For a more complete bibliography see O. and R. Hertwig (’78a), von Lendenfeld (88), and Hesse (’95). Ermer, T. 1877 Nervensystem dex Medusen. Arch. f. mikro. Anat., Bd. 14, p. 398. Cuaus, C. 1877 Studien iiber Polypen und Quallen der Adria. Denk. der k. Akad. der Wiss.,. Wien, Bd. 38. Scudrer, E. A. 1878 Nervous system of Aurelia aurita. Phil. Trans., vol. 168, p. 563. Hertwie, O. anp R. 1878(a) Das Nervensystem und die Sinnesorgane der Medusen. Leipzig: Vogel. Hertwie, O. anv R. 1878(b) Der Organismus der Medusen und seine Stellung zur Keimblattertheorie. Jena: Fischer. Denk. Med.-Nat. Gesell. Bd. 2. Heft 1. HarEckeEt, E. 1880 Das System der Medusen. Halfte 2. Denk. Med.-Nat. Gesell. Jena. Bd. 1, Abt. 2. LENDENFELD, E. von. 1882 Cyanea anaskala. Zeit. fiir wiss. Zool., Bd. 37, p. 465-548. GoetTe, A. 1887 Entwicklungsgeschichte der Aurelia und Cotylorhiza. Ham- burg. LENDENFELD, R. von. 1888 Die australischen rhizostomen Medusen. Zeit. f. wiss. Zool., Bd. 48, p. 406. Witson, H. V. 1888 Onthe development of Manicina areolata. Jour. Morph., v. 2, 191. ScHEWIAKOFF, W. 1889. Beitrage zur Kenntniss des Acalephenauges. Morph. Jahrbuch, Bd. 15, p. 25. BigeLtow, R. P. 1890 The marginal sense organs in the Pelagidae. Johns Hop- kins University Circulars, v. 9. No. 80, p. 65-67. VANHOFFEN, E. 1891. Periphylla und Nausithoe. Zool. Anz., Jahrg. 14, p. 38-42. Hesse, R. 1895 Uber das Nervensystem und die Sinnesorgane von Rhizostoma cuvierl. Zeit.f.wiss. Zool., Bd. 60, p. 411-457, pl. 20-22. Mayer, A.G. 1898 On Dactylometra. Bull. Mus. Comp. Zoél., v. 32, no. 1. BicELow,R.P. 1900 Onthe anatomy and development of Cassiopea xamachana. Memoirs Boston Soc. Nat. Hist., v. 5, no. 6. VANHOFFEN, E. 1900 Uber Tiefseemedusen und ihre Sinnesorgane. Zool. Anz., Bd. 23, p. 277-279. 1902 Die acraspeden Medusen der deutschen Tiefsee-Expedition 1898-1899. Jena. KisHINovYE, K. 1902 Some new Scyphomedusae of Japan. Jour. College Sci. Univ. Tokyo, v. 17, art. 7. Maas, O. 1903 Die Scyphomedusen der Siboga- Expedition. Leiden pl. xi. —e 776 ROBERT PAYNE BIGELOW EXPLANATION OF FIGURES All figures are camera drawings, except as otherwise noted below. They were made at the magnification indicated and reduced in reproduction to about one- third the original size. a LETTERING COMMON TO ALL THE FIGURES Battery of nettle cells. p. Lateral pocket lined with pitted ec Concretions, or otoliths. epithelium. ec. Ectoderm. p.e. Pitted epithelium. en. Endoderm. f. Rhopalium. q. Ganglion cell. r.c. Rhopalial canal. h. Hood. r.r. Rhopalial ridge. mM. Muscle fibers. s.e. Sensory epithelium. m.l. Marginal lappet. s.g. Dorsal sensory groove. m.p. Marginal endodermal pocket. s.n. Sensory niche. n. Layer of nerve fibers. Figs. 1-12, 21-34, Chrysaora; 13-16, Pelagia; 17-20 and 35-38, Dactylometra. 1 View of the sensory niche of a Chrysaora as seen from the under side by trans- mitted light. The specimen from which this was taken was a young one 25 mm. in diameter with 24 tentacles. > 200. 2 Portion of the margin of the umbrella of a full grown Chrysaora containing the rhopalium and adjacent organs. The cut surfaces are in tangential and radial planes. X 56. This figure is a reconstruction from sections. 3-9 Diagrams from typical sections of the series used in the construction of fig. 2. X 56. Fig. 3 corresponds to plane I in fig. 10, fig. 4 to II, fig. 5 to III, fig. 6 to IV, fig. 7 to VI, fig. 8 to VII, and fig. 9 to VIII. SENSE-ORGANS IN MEDUSAE 777 778 ROBERT PAYNE BIGELOW a EXPLANATION OF FIGURES 10 Diagrammatic drawing of a radial section of the margin of Chrysaora, cut- ting the rhopalium longitudinally. 56. The lines I, II, etc., show the planes of figs. 3, 4, ete. 11 A vertical tangential section cutting the rhopalium obliquely so as to show the band of nerve fibers running from the rhopalium to the lateral pocket. Slightly diagrammatic. X 120. 12 Radial section through the columnar epithelium at the proximal end of the rhopalial ridge. Thissection happens to contain two nettle cells. XX 550. 13-15 Typical tangential sections through the sensory apparatus of Pelagia cyanella, somewhat diagrammatic. 108. Fig. 13 corresponds to fig. 4 of Chry- saora, fig. 14 to a plane between figs. 5 and 6, and fig. 15 approximately to fig. 8. 16 Aradial section of the rhopalial ridge at its proximal end in Pelagia, showing a large ganglion cell. X 1120. ee en me ee ~~ "79 % “ (cS peat Ge SENSE-ORGANS IN MEDUSAE 780 ROBERT PAYNE BIGELOW EXPLANATION OF FIGURES 17-20 A series of sections from Dactylometra quinquecirrha. X 56. Fig. 17 corresponds to fig. 5 of Chrysaora, fig. 18 is in a plane corresponding to V of fig. 10, fig. 19 corresponds to fig. 7, and fig. 20, which is oblique to the radius, nearly corre- sponds to fig. 8. 21 Theends of one of the arms of an ephyra of Chrysaora viewed from the under side as a transparent object, showing the relations between the rhopalium, the gastric pouch (r.c.), and marginal lappets. X 280. 22 Longitudinal section of a rhopalium from the same specimen. X 600. 781 SENSE-ORGANS IN MEDUSAE rl SO 782 ROBERT PAYNE BIGELOW EXPLANATION OF FIGURES 23-25 Tangential sections of another arm of the same specimen. XX 195. Fig. 23 is about in plane I of fig. 22, fig. 24 is in plane ITI, while fig. 25 is a section from near the base of the arm. 26 Radialsection through the medial line of the rhopalium of a Chrysaora larva that is in the Pelagia stage and is6mm.indiameter. X 490. 27 Asection of the same specimen parallel to fig. 26, just to one side of the rho- palium, showing a cross section of the rudimentary lateral pocket. X 490. 28-31 Tangential sections across another rhopalium of the same individual. x 490. Fig. 28 isin plane I, fig. 29 is in plane II, fig. 30 is III, and fig. 31 in IV of fig. 26. 32 A tangential section oblique to the axis of the rhopalium and through the longest diameter of the rudimentary lateral pocket. x 490. SENSE-ORGANS IN MEDUSAE 783 ‘ t 4 9 \ - ; RPB.del. 31 32 784 ROBERT PAYNE BIGELOW EXPLANATION OF FIGURES 33 Radial section through the rhopalium of a young Chrysaora just passing into the adult form. The specimen was about 10 mm. broad and the second set of tentacles about as long as the marginal lobes. X 300. 34 Asection at right angles to the above through another rhopalium of the same animal. XX 280. 35 A radial section through the rhopalium of an ephyra of Dactylometra. there has evidently been considerable shrinkage of the mesogloea in this specimen. XX 523. 36 A similar section from a somewhat older larva. X 490. 37 A tangential section through the sensory niche just proximal to the base of — the rhopalium in a young Dactylometra about 18 mm. in diameter, which is in the early part of the Chrysaorastage. 190. (Compare figs. 7 and 18). 38 A section parallel to the above about 60» farther inward through the outer half of the lateral pockets. X 190. SENSE-ORGANS IN MEDUSAE 785 RPB del, THE QUTGROWTH OF THE NERVE FIBER AS A MODE OF PROTOPLASMIC MOVEMENT Sal ROSS GRANVILLE HARRISON Bronson Professor of Comparative Anatomy, Yale University THIRTY-TWO FIGURES THREE PLATES CONTENTS EE ee a 787 Early development of nerve fibers in the normal embryo.................. 792 Experiments upon embryonic tissues isolated in clotted lymph.............. 799 IR ee 799 ee 2) 808 Description of the behavior of nervous tissue...................-...... 813 ic ocx bpeenn's dime pd dig nk tine ye see 823 The significance of the experiments in the interpretation of normal ec ane kb elm og Gon omannneeenene 823 The bearing of the experiments upon the theories of nerve development... 826 The elementary factors of nerve development ...................--.-+5: 830 Analysis of the factors which produce the specific arrangement of the Np andy wade) meapadices nares 833 cect cea cc ev vb wana web euseseeees 841 INTRODUCTION The idea that protoplasmic movement is concerned in the activ- ities of the nervous system has appeared in a variety of forms during the past twenty years. Not only has it been supposed that the processes of nerve cells may be extended and withdrawn, making and breaking connections with other cells during func- tional activity, but also that the movement of cells and their processes in the course of development has been the chief factor in bringing about the specific nervous connections found in the adult.1 The latter idea is associated particularly with the name of Ramon y Cajal, who in his memoir on the retina (’92) first put ' Schiefferdecker (’06) has discussed at length and in an admirable way the extensive literature bearing upon this subject. 788 ROSS GRANVILLE HARRISON forth the hypothesis of chemotaxis to account for these supposed movements. The discovery, by the same observer (’90), of the cones d’accroissement, found at the end of embryonic nerve fibers very early in their development, had given a clue as to what this growth mechanism might be, for the resemblance of the minute processes borne upon the terminal enlargement of the growing nerve to pseudopodia, naturally suggested that this structure might owe its peculiarities to amoeboid activity. In his larger work on the structure of the nervous system Cajal (’99) elabo- rates his theory more fully and leaves no doubt as to his meaning regarding the activity of the growth cones. After describing their appearance he says (p. 544-5): ‘‘From the functional point of view, the cone of growth may be regarded as a sort of club or battering ram, endowed with exquisite chemical sensitiveness, with rapid amoeboid movements, and with a certain impulsive force, thanks to which it is able to press forward and overcome obstacles met in its way, forcing cellular interstices until it arrives at its destination.’’ From this it is seen that Ramon y Cajal took a considerable step in advance of His (’86—’90), and placed upon a still firmer basis the concept that the nerve fiber is formed as the outgrowth of a single cell. Although this view has enjoyed wide acceptance, the opposing theory of Hensen (’64—’08), which denies that there is a free outgrowth of protoplasmic substance to form the nerve fibers, has met with increasing support within the past few years, es- pecially in the work of O. Schultze (’04—’08), Braus (04-05), Held (06-’09), Paton (07) and Schaeppi (09); and it seems that we are really very far from a satisfactory solution of the question, which even the invention of new and marvelously refined histological methods has failed to bring to a final settle- ment. Nor has Held’s? compromise theory, which is based upon such methods, and which sees in Hensen’s protoplasmic bridges merely a sort of substratum into which the fibrillar substance ex- tends from the neuroblasts or ganglion cells, succeeded in har- 2 Held’s view appears on the surface to be a modification of Hensen’s theory and it is usually classed as such, but a full examination of his complete work shows that in reality it approaches much more closely to His’s view. OUTGROWTH OF THE NERVE FIBER 789 monizing the two views. The wide discussion of the subject which has taken place reached a certain culmination in the contro- versy between Held and Ramon y Cajal in the years 1906-1909, in which it became clear that the evidence for and against the two theories respectively, rested upon such minute histological details that a decision to which all would subscribe was impos- sible of attainment. These two observers studied to a great extent similar material, often by the same methods, and, in fact, their prepared material was so much alike that Ramon y Cajal, after seeing Held’s specimens, expressed great astonishment at the similarity.* Yet the respective interpretations given by them differ diametrically. Under such conditions a search for evidence of other kinds is indicated. It was with the hope that a study of the problem by entirely different methods might yield such evidence, that the work described in the present paper was undertaken. A crucial experiment was sought that would decide between the two theo- ries. That a decision of this question is of fundamental impor- tance becomes apparent when we consider that the analysis of the factors bearing upon the development of this most intricate system of organs is wholly dependent upon it; for it is obviously impossible to study intelligently the mechanics of development of the nerve paths, unless we know whether we are dealing pri- marily with phenomena of protoplasmic movement or with mere progressive differentiation without movement. An extensive series of experiments, as well as observations upon normal embryos, had led me previously to the adoptionof the view of His and Ramon y Cajal. These experiments (Harrison ’06— 10) showed that the ganglion cells within the nerve centers are the one essential element in the formation of the nerve fiber, in- asmuch as pieces of the embryonic nervous system transplanted to any part of the body may give rise to nerve fibers, while no fibers ever develop in the absence of ganglion cells. It was recog- 3 R. y Cajal, 1908, p. 3, footnote: Tout récemment pendant un voyage en Alle- magne, nous avons eu le plaisir d’examiner A Leipzig, les excellentes préparations de M. Held. Ainsi que nous l|’attendions elles sont trés réussies, mais 4 notre grande surprise elles montrent 4 peu prés les mémes images que les ndétres. a 790 ROSS GRANVILLE HARRISON nized, however, that in all of the first experiments the nerve fibers had developed in surroundings composed of living organized tissues, and that the possibility of the latter contributing organ- ized material to the nerve elements, stood in the way of rigorous proof of the view that the nerve fiber was entirely the product of the nerve center. The really crucial experiment remained to be performed, and that was to test the power of the nerve centers to form nerve fibers within some foreign medium, which could not by any possibility be suspected of contributing organized protoplasm to them. Two lines of experimentation were taken up with this end in view. ‘The one was to introduce small pieces of clotted blood into the embryo, in the path of the developing nerves. This gave positive results, in that nerve fibers were found several days after the operation, extending from the medullary cord intothe blood clot, and the sole possible disturbing factor in these ex- periments was the presence of scattered embryonic cells, which began to organize the clot within two days after its transplanta- tion (Harrison 710). The second line of experimentation, which consisted in the isolation of pieces of living tissue in unorganized media, gave con- siderable difficulty at first, but in the springof 1907 a method was finally devised, which satisfactorily accomplished the purpose. The present paper contains a complete account of these ex- periments, which have been described previously in a preliminary notice. In addition, a brief description of the early develop- ment of the nerve elements in the normal amphibian embryo * The first of these experiments were made in the Anatomical Laboratory of the Johns Hopkins University. After my removal to Yale University they were con- tinued during the seasons of 1908 and 1909in the Sheffield Biological Laboratory. The repetition of the work gave results which not only confirmed those of the first season, but which also met many possible objections that might have been raised against the original experiments. The preparations obtained during the second season’s work were, on the whole, much more convincing than those of the first, and they have been used almost exclusively in making the illustrations for the present paper. The first account of the work was given in a paper before the Society for Experimental Biology and Medicine in May 1907, and later the results were incorporated in a lecture before the Harvey Society of New York, in March 1908. OUTGROWTH OF THE NERVE FIBER 791 will be given here, in order to afford a basis for comparison with the protoplasmic filaments formed by the isolated pieces of ner- vous tissue. Fortunately the part descriptive of normal develop- ment need not occupy very much space, for we now have a large mass of facts available in the recent work of Ramon y Cajal (07-08) and in the exhaustive monograph of Held (’09). The method which I have used is, in a word, as follows: Small pieces of embryonic tissue, taken before the histological differ- entiation of nerve fibers has begun, are placed in hanging drops of lymph, and the sealed preparations kept under observation for a number of days. It is found that the embryonic cells under these conditions manifest striking amoeboid activities, which are especially pronounced in cells taken from the nervous system, and result in such cases in the formation of long threads of hyaline protoplasm. These fibers bear a perfect morphological re- semblance to undoubted nerve fibers found in sections of nor- mal embryos of a corresponding stage of development. So strik- ing is the similarity between these structures that no hesitancy is felt in regarding them as identical with one another. This method, which obviously has many possibilities in the study of the growth and differentiation of tissues, has two very distinct advantages over the methods of investigation usually employed. It not only enables one to study the behavior of cells and tissues in an unorganized medium free from the influ- ences that surround them in the body of the organism, but it also renders it possible to keep them under direct continuous observation, so that all such developmental processes as involve movement and change of form may be seen directly instead of having to be inferred from series of preserved specimens taken at different stages. While these two advantages have not here- tofore been combined in a single mode of procedure, the first named has been attained by Loeb (’02) who has embedded pieces of tissue, chiefly epidermis, in blocks of agar or clotted blood and transplanted them to spaces in the body of living animals. It is interesting to note that under these conditions epithelial cells undergo changes which apparently resemble closely the activities of embryonic cells observed in the present investigation, as a comparison of Loeb’s figures with my own shows. eee rer ee a Phe 792 ROSS GRANVILLE HARRISON EARLY DEVELOPMENT OF NERVE FIBERS IN THE NORMAL EMBRYO Conditions obtaining in the central nervous system antecedent to the differentiation of fibers In the walls of the medullary groove and the medullary tube just after it has been completely folded off from the epidermis, one can distinguish in sections a number of irregular layers of cells, mostly oval in shape, with long axis placed radially with respect to the tube as a whole. Sometimes these cells are seen to be bound by a membrane, but usually they are indistinctly defined except where they are deeply pigmented, in which case the pig- ment granules are thickest around the periphery of the cells. At this period the individual cells do not extend through the whole thickness of the tube from the central canal to the external limiting membrane. In slightly later stages, 2.e., when the tail bud is barely distin- guishable, the epithelial cells begin to stretch out radially and then many of the individual cells are seen to extend from the inner to the outer wall of the tube. The boundaries remain indistinct, unless, as before, the cells are marked off from their neighbors by pigmentation. After the elongation of the epithelial cells constituting the walls of the medullary tube has taken place, it is seen that certain cells, less elongated in form, and containing around nucleus, remain in the outer zone of the wall of the tube. These are the first of the neuroblasts of His, the cells destined to give rise to the nerve fibers. There are as yet no peripheral nerves, nor are there any nerve fibers visibly differentiated within the walls of the medullary tube. The cranial ganglia are marked off and occupy approx- imately their definitive position, and in the anterior part of the trunk region the ganglion crest is beginning to break up, its cells extending to the dorsal border of the muscle plates. In the mid- dle of the trunk the crest is intact and it rests entirely upon the medullary cord, while near the tail bud it can scarcely be distin- guished at all. OUTGROWTH OF THE NERVE FIBER 793 At this point it will be profitable to inquire a little more fully into the supposed syncytial nature of the central nervous sys- tem. When sections alone are studied, there may be an appar- ent justification for regarding the walls of the neural tube as a mass of protoplasm with nuclei embedded in it,* for, as has already ~ been pointed out, the cell boundaries within the medullary cord are difficult to make out unless they happen to be indicated by pig- mentation. When examined in the fresh condition, an entirely different state of affairs is revealed. It is astonishing how easily the cells, which in sections seem to be baked together in a mass, come apart when the medullary cord is dissected out of the living embryo and teased in water or salt solution. The cells appear as round glistening vesicles under the binocular microscope, and un- der the oil immersion they are found to be very clearly defined, each being surrounded by a very delicate, though perfectly dis- tinct, cell membrane. The cells are gorged with yolk granules, and the nucleus appears as a clear space near the center of each cell (fig. 15). There is not very much difference in the appear- ance of the cells in the different media named, though in water and the more dilute salt solution (0.2 per cent) there is some imbibition of water, which may result in the formation of a more or less eccentric clear zone just beneath the cell membrane (fig. 15 c). No sign of protoplasmic bridges can be made out. From these observations the conclusion seems clearly justified that the med- ullary cord of the frog embryo is made up of perfectly distinct cells. It isin no sense a syncytium, and statements to the con- trary based upon the insufficient evidence from stained sections, are to be received with skepticism. The medullary cord is sharply marked off from all surrounding structures except where the ganglion crest is breaking down. The cord is in direct contact with the muscle plates and the noto- chord, but in the angle between the two latter structures, and in the the grooves between successive somites there are small spaces, which at this period are entirely devoid of cells. Just whatis the structure of the material that fills these spaces in the living 5 Cf. for instance Weysse and Burgess (1906) on the histogenesis of the retina. 794 ROSS GRANVILLE HARRISON embryo is not, in my opinion, certain, but in sections of preserved specimens, as Held has described in great detail, a delicate net- work, is visible. The character of this intercellular reticulum varies from specimen to specimen and, as will be seen, varies very greatly according to the mode of preservation. It seems to be beyond doubt that the structures in question are due in part to coagulation, though just to what extent it is not easy to say. In order to test the matter a series of embryos were preserved in osmie acid, which, as Fischer (’01) has shown, fixes protoplasm without bringing about any visible change in structure, and which after prolonged action (24 hours, | per cent, in the case of Amoeba proteus) so fixes it that alcohol causes no further change. Sec- tions of these embryos show plainly that the spaces between the organs described above are almost perfectly clear; only occasion- ally do very delicate filaments appear bridging the spaces. The contrast with specimens which have been preserved in a corrosive sublimate-acetic mixture is very great; and very much more pro- nounced still is the difference shown by embryos preserved in Hermann’s fluid, which is, however, otherwise a very ill adapted preservative for this material.6 It is not intended on the basis of the foregoing observations to deny the existence of protoplasmic bridges in embryos of this stage, but it does seem proper to call attention to the facts just stated, in order to show the necessity for caution in ascribing significance to the connection between such fine structures and the developing nerve fibers. Differentiation of nerve fibers The embryo last described is in the stage which was used for most of the experiments. It is the oldest stage of which it can be said with certainty, without microscopic examination, that there are no nerve fibers present. In the next stage to be con- ® On account of the large amount of yolk, which becomes very brittle after pro- longed treatment with osmic acid, the amphibian embryo is not a favorable object for the study of this question. It was found necessary to impregnate the embryos with celloidin before embedding in paraffin, and even then the sections were not perfectly satisfactory. It would be of great interest to have an exact compari- : ' Hl , : OUTGROWTH OF THE NERVE FIBER 795 sidered, an embryo of R. sylvatica, 4.1 mm. long, the beginnings of the peripheral nerves, and of some of the principal central bundles are plainly visible. Of all the peripheral nerves the r. ophthalmicus of the trigeminal, seems to be furthest advanced. A veryearly phaseof this nerve is shown in fig.2 (nf), drawn from an embryo of R. esculenta,3 mm. long, which is in about the same stage of development as the sylvatica embryo just mentioned. Protoplasmic processes of the cells within the ganglion are seen to extend for a short distance into the mesenchyme, without hav- ing any special relation to the cells of that tissue. The ends of the processes are branched and filamentous. In the sylvatica -embryo under consideration, a considerable number of peripheral nerves in addition to the ophthalmic are alreadylaid down. There are at least four ventral spinal roots, corresponding to the sec- ond, third, fourth and fifth muscle plates, to which they may be traced; several of the dorsal nerves of Rohon-Beard, extending out between the myotomes and the epidermis; and some fibers in the r. lateralis vagi.? The early characteristics of the developing nerve are most clearly shown by the fibers which originate in the dorsal cells of Rohon-Beard. These grow just beneath the epidermis in the space between the muscle plates, where at this period there are no loose mesenchyme cells, and they remain free from sheath cells throughout their growth. The clearest cases of the earliest beginning of these nerves have been found in an embryo of Rana palustris, 3.6 mm. long, which is almost identical in degree of development with the sylvatica embryo just described. The son of the protoplasmic bridges fixed in osmic acid with those seen after fixation in the usual preservatives made upon such vertebrate embryos as those of the selachian, the teleost, or the bird, in which there is little or no yolk in the tissues at the time when the first nerve fibers differentiate. 7 These early nervous connections, which are important for the proper inter- pretation of the relation between structure and function in the neuro-muscular system, have been ignored by a number of investigators. In his histogenetic study of the nervous system O. Schultze (’05) has overlooked these stages of de- velopment completely and has thereby been entirely misled in his views regarding the early development of nerve fibers and the formation of the cutaneous plexuses (Cf. Harrison -’04, ’06). Held (’09) has recently subjected Schultze’s work to a searching criticism, all the main points of which seem to be entirely justified. 796 ROSS GRANVILLE HARRISON eells which give rise to the dorsal nerves form a column in the dorso-lateral part of the wall of the medullary tube just within the external limiting membrane. In this stage certain of the cells are seen to have put forth fine branched processes, which extend for a short distance laterally in the notch between successive muscle plates (fig. 1, nf). The processes end in extremely fine filaments, so fine that their exact delimitation is often very diffi- cult to determine. The cell shown in the figure gives off another process quite as extensive as the one shown, but which is seen only in the section next to the one drawn. The structures in question are segmentally arranged, and correspond in the embryo under consideration to the intervals between the muscle plates from the second to the thirteenth segments. A much more advanced condition is shown in an embryo but very slightly older (3.7 mm. long). The dorsal nerves are here composed of several fibers in a bundle, each fiber connecting with a cell. The nerve shown in fig. 3 is composed of four such fibers (nf) which arise in pear-shaped cells (nbl) and converge toward the point where they leave the medullary cord between the second and third myotomes. The endings are not shown in the section because the fibers bend just beneath the epidermis and run dorso-ventrally. They stain intensely with Congo red, as do the cone-shaped proc- esses of the cells from which they originate, and they show a fairly distinct fibrillation, even when stained merely by this method. The ends of the fibers are best seen in sagittalsections taken just between the epidermis and the underlying muscle plates. In a _ series of sections made from an embryo of Rana pipiens, 4mm. long, they show particularly well. In fig. 4 the end of a bundle of three fibers situated between the ninth and tenth segments is shown. This terminal structure (npl) consists of a mass of hyaline protoplasm having a form suggestive of a rhizopod. The mass extends out into a number of very fine filaments. Such structures are found in each segment. Another, more highly magnified, is shown in fig. 5. Further towards the head of the embryo the fibers are longer and more branched (fig. 6), each branch ending in one of the peculiar enlargements just described. The young fibers of the r. ophthalmicus end similarly, although the ending OUTGROWTH OF THE NERVE FIBER 797 cannot always be made out with such clearness, owing to the exist- ence of the branched mesenchyme cells in their immediate vicinity. In other nerves, as in thecase of the r. /ateralis vagi of Amblystoma, _there is a slight enlargement at the end of the growing fiber, though branched filaments are not clearly shown there. I¢ is a striking fact that in these early stages of development, each nerve fiber, in fact each branch of a nerve fiber, ends in an enlargement of thiskind. The enlarged ends, as well as the fibers throughout their whole length, are attached to surrounding’organs by fine threads, but, as stated previously, I am unable to find any safe criterion to distinguish between natural protoplasmic fila- ments and products of coagulation. Aside from these fine fila- ments, the nerve fibers are found to end free, and anastomoses between different nerves are not present at this stage. This is perfectly clear in the case of the cutaneous nerves formed by the cells of Rohon-Beard. A little later, however, as seen in a R. pipiens embryo, 6 mm. long, the branches of the individual seg- mental nerves are found to have extended so far as to come into contact with those of the next segment, the result being the formation of a beautiful plexus of nerve fibers beneath the skin overlying the muscle plates. This is composed of fibers devoid of sheath cells, and in specimens hardened and stained by vom Rath’s picro-platino-osmo-acetic mixture, the fibrillae are shown very clearly. Plexus formation is thus seen to be secondary, resulting from the accidental coming together of the growing ends of nerve fibers which have origin in different segmental nerves (text fig. 1). In all cases the nerve fibers are found to extend gradually out from the center, and the end of each small twig is characterized by an enlargement made up of hyaline proto- plasm, provided with fine filaments, just as the main stem of the fiber itself is at first. The above observations upon the ends of the developing nerves agree substantially with those of Ramon y Cajal, although they are based upon specimens preserved by entirely different methods. The enlarged ending provided with protoplasmic filaments is in all probability the céne d’accroissement first described by him, the filaments being shown in these cases perhaps more completely ne We) ae 798 ROSS GRANVILLE HARRISON > we ¥ Fig. 1 Diagram illustrating the mode of development of the sensory nerve plexus, derived from the dorsal cells of Rohon-Beard. Each segmental nerve is represented in its simplest terms as a single fiber originating in a single cell of the neural tube. A, early stage in which the nerve fiber has just begun to grow out. B, Later stage in which each segmental nerve has begun to branch. C, Final stage in which neighboring nerves form anastomoses with each other. OUTGROWTH OF THE NERVE FIBER 799 by the ordinary embryological methods.® Cajal has figured in a number of places his growth cones, as seen both in Golgi and in silver nitrate preparations. Those shown in his book on the Struc- ture of the Nervous System, vol. 1, p. 515, are in most striking agreement with the figures here presented. Again, there is no sharp discrepancy between these figures and those of Held, whose figures, like those of Cajal are sharper than the present ones, since they represent the specific coloration of the neuro- fibrillae. The only essential difference shown by those of the former observer is in the relation of the young nerve fibers to the protoplasmic net-work between the cells and this to my mind is wholly a question of interpretation. Considering the uncertain nature of the intercellular net-work, as pointed out above, the unusually positive views of Held regarding its réle in the develop- ment of the nerve fibers seem but very insecurely founded. EXPERIMENTS UPON EMBRYONIC TISSUES ISOLATED IN CLOTTED LYMPH Description of methods The first attempt which I made to study the development of iso- lated bits of embryonic nervous tissue gave entirely negative results. The tissue was dissected out from the embryo and put either into physiological salt or Locke’s solution, but no differen- tiation was observed, before disintegration began. Later a more natural environment for the isolated tissue was sought in the ven- tricles of the brain and in the pharynx of youngembryos. The tissues were transplanted to these cavities and the specimens were killed after from two to seven days and examined in serial sec- 8In salmon embryos preserved and stained by the ordinary embryological _methods, no protoplasmic filaments are shown attached to the growing ends of the nerve fibers within the central nervous system, and for this reason the latter were figured as smooth in my paper on the histogenesis of nerves (Harrison’01). Ramon y Cajal has pointed out that this condition is likely due to the insufficiency of the methods. While I agree that there is some ground for this criticism,it seems never- theless probable that there are actual differences between the growing ends found in different places and in different species. a. he = oe aa an 800 ROSS GRANVILLE HARRISON tions. ‘These experiments likewise resulted negatively. In no case were nerve fibersfound extending from the transplanted piece free into the cavity, although the pieces themselves often showed differentiation of fibers, and in cases where the graft had grown fast to the walls of the medullary tube, fibers passed from the former to the latter. The only conclusions which could be drawn from these results were either that the nerve fibers were built up by the differentiation of formed protoplasmic structures, according to the view of Hensen and Held, or else that the growing fibers were positively stereotropic and hence remained within the solid tissue instead of passing out into the surrounding fluid. Acting upon the latter assumption, the next step was to try a fixed medium. ‘Two such were employed, one of which, gelatine, gave no results at all, the transplanted embryonic tissue remain- ing entirely unchanged after imbedding. The other, clotted frog’s lymph, gave the results that are here recorded. It was rather to be expected that this medium would yield positive re- sults, if indeed such were to be obtained at all, for it would be chemically the most natural medium, and the fine net-work of fibrin threads, bathed by the fluid serum, would in a measure simulate mechanically the protoplasmic net-work, which, ac- cording to Hensen, Held and others, seems to exist in the tissue spaces in which the peripheral nerves undergo their early devel- opment. In the first experiments made with the lymph, the technique employed was comparatively simple. The tissue to be studied was dissected out of the embryo under the binocular microscope in 0.4 per cent sodium chloride or in Locke’s solution without sugar. It was then transferred to a cover-slip by means of a cap- illary pipette, and a drop of lymph drawn from one of the lymph sacs of an anaesthetized frog was quickly dropped upon it. The cover-slip was then inverted over a depression slide and the prep- aration kept ina moist chamber. In order to avoid evaporation’ while the specimens were under examination it was necessary to seal the preparations, which was done most satisfactorily by apply- ing melted paraffine around the cover-slip with the edge of a warm plate. OUTGROWTH OF THE NERVE FIBER 801 Although the first definite results were obtained by the above methods, it was found that bacteria quickly invaded the prepara- tions, often destroying them as soon as the second day after im- plantation. Continued observation over a long period was there- fore impossible, and many otherwise good specimens were spoiled before they had yielded anything but negative results. After experimenting a little with antiseptics such as thymol and ace- tone-chloroform, it became apparent that satisfactory prepara- tions could not be obtained except by working aseptically. The procedure necessary for this involved much tedious detail, though it offered no insuperable diffculties.* All glassware, such as slides, covers, pipettes and dishes, was sterilized by dry heat, either in a hot air sterilizer or by passing them through a flame. For cloths and filter paper an Arnold sterilizer or an autoclave was used, and the needles, scissors and forceps were sterilized by boiling. The sterilization of the embryos and the frogs from which the lymph was to be taken offered greater difficulties, and in fact was accomplished only approximately, though the number of organ- isms seems to have been so reduced as not to interfere with the purpose of the experiments. The embryos were simply cut out of their jelly in water which had been boiled or passed through a Pasteur-Chamberland filter. They were then washed in about six successive changes of this water. The salt solution in which the operations were performed was sterilized in the same way. The frogs were chloroformed and then washed thoroughly in ster- ile water, laid out upon moist filter paper and kept in a covered dish. In some cases they were first washed in mercuric chloride (0.1 per cent) and in others they were kept for 24 hours before chloroforming in a solutionof copper sulphateone part to 500,000, but I am not prepared to say whether these means were sufficiently effective to be of material advantage. The results of these manipulations are altogether satisfac- tory as regards asepsis, although the making ready of the appara- tus consumes so much time, and the constant attention to the °T am greatly indebted to Prof. Leo F. Rettger for valuable suggestions as to this procedure, and for his generosity in putting at my disposal the apparatus in his laboratory. JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 9, NO. 4. 802 ROSS GRANVILLE HARRISON detailsof manipulation during operations is so fatiguing, that only a small number of preparations can be made in one day. Many preparations proved to be absolutely sterile. In some of these the tissues were kept alive for over five weeks, and in a great many for one or two weeks. Some were contaminated, most frequently with Bacillus subtilis, but even in these cases the organisms did not usually appear in sufficient number to injure the living tissue until after it had been kept under observation for four or five days, which was long enough for the present purpose. Several epi- demics of mould (Penicillium) were encountered, but this too grew slowly, usually from a single spore or two, and as it does not seem to kill the embryonic cells, it interfered but little with observa- tions. The tissue to be studied is dissected out from the sterilized em- bryos in a small flat dish containing dilute salt solution. After this is done the next step is to obtain a drop of lymph from the frog, which has already been prepared. ‘The animal is suspended or placed in an upright position, and after cutting into the lymph sac near its upper end a long fine pipette is introduced and a small drop is drawn from the bottom of the sac. This isthen placed upon the cover-slip, and the piece of tissue is quickly transferred to the lymph by means of the same pipette, with care to take with it as little of the salt solution as possible. Then the cover-slip is inverted over a depression slide and the preparation sealed by means of paraffine. It is important to have the depression in the slide deep enough to prevent the drop, which must also be small, from coming into contact with the bottom. The procuring of the lymph is the most difficult part of the whole procedure, and the variability in its quality and in the amount obtainable, introduces into the work an element of in- constancy, which is a serious disturbing factor, preventing, as it does, a sharp clear-cut process of experimentation with exact controls from being carried out. The composition of the lymph varies not only amongst individual frogs but also in the dif- ferent lymph sacs of the same individual, according to the position in which the animal has lain, the time since anaestheti- zation, and other factors of unknown nature. In general it may OUTGROWTH OF THE NERVE FIBER 803 be said that the first lymph drawn is the best; it clots readily and is less hemorrhagic, though this is by no means always the case. Usually a single frog can be used for five or six drops. The first two drops were taken in most cases from the femoral sacs. After opening these sacs, the lymph in the crural sacs becomes so watery that it will not form a sufficiently firm clot for the purpose, but the abdominal, lateral, and dorsal sacs of the trunk, as well as those of the forelimb will usually each yield a small drop which clots firmly. The quantity obtainable from a single sac is often too small to be of use. In fact, whenever any very large amount is to be had, it is very watery in quality, as is especially the case in the sacs which happen to lie lowermost. This oedematous condition is no doubt due to weakening of the heart, action but oddly enough it is more pronounced in frogs which have been pithed than in those chloroformed. Perhapsif the animals were anaesthe- tized by cold, the lymph obtained would be more uniform, and the low temperature would retard the clotting somewhat, which would be a distinct advantage. Even after taking the foregoing cir- cumstances into consideration it is impossible always to get lymph of the proper composition. It may be very thin and fail to clot; or it may be so rich in fibrinogen that it clots immediately, even before it can be got out of the pipette, or in any case before the tissue can be transferred to it upon the cover. During this time, which is variable, some evaporation takes place and thus another factor of uncertainty is introduced. Still another vari- able is the amount of lymph relative to the amount of salt solu- tion taken up with the embryonic tissue. Itis not surprising, therefore, that there should be variations in the results of the ex- periments, which cannot be ascribed to any particular cause. On theother hand it apparently makes no difference from whatspecies of frog lymph is taken, Rana sylvatica, pipiens, palustris and clamitans, al] having yielded satisfactory material. Nor does it seem to be of consequence that the lymph should be of the same species as the embryonic tissue. Embryos of R. sylvatica, R. pipiens, R. palustris, and, in a few experiments, of Bufo lentiginosus, were used, all in very nearly the same stage of development, corresponding to that used in most 804 ROSS GRANVILLE HARRISON of the previous experiments upon the development of the nervous system. The medullary folds are just completely closed and the tail bud is barely visible. The reason for choosing this stage is because it is the latest in which there is no histological differen- tiation in the nervous system or muscle plates. All of the cells are compact and no fibers whatever are present. The tissues are thus got into the lymph before their histogenetic development has begun. The transference of the tissue to the lymph drop cannot be accomplished without a considerable amount of tearing. Often single cells or small groups are torn loose from the main mass and individual cells are fragmented, setting free yolk and pigment granules, but the fibrin holds the main masses together, unless the lymph is too thin, in which case the embryonic cells round off and separate from one another. This same kind of disintegration has been observed also in some cases in which the clot was firm. Even in the absence of bacteria the cells in these specimens may remain entirely unchanged, manifesting none of the peculiar protoplasmic activities seen in successful preparations. It has not been pos- sible to assign any particular cause for this condition, and it must be attributed to slight deviations from the normal in the composi- tion of the medium. All such experiments, and these have formed but a small percentage of the whole, have been rejected as incon- clusive, and have been so indicated in the tabulation of results. While the methods of preparation were practically the same in all cases, the experiments themselves were varied considerably as regards the tissues isolated. The chief object of the work being to test the power of embryonic nerve cells to form fibers by out- growth, the largest number of experiments were made with ner- vous tissue. Insome cases the medullary cord was dissected out entire, though it usually broke when transferred, and in others it was purposely fragmented by teasing. In quite a number of cases portions of the muscle plates were left attached to the med- ullary cord. In other experiments pieces of ectoderm from the branchial region, together with the underlying ganglia were taken. The behavior of this tissue, as regards the formation of fibers, was altogether similar to that of the medullary tube, and has OUTGROWTH OF THE NERVE FIBER 805 thus served to confirm the conclusions which have been drawn from the study of the former. On the other hand, the experiments have been controlled by observing the behavior of other embryonic tissues, such as muscle plates, ectoderm from theabdominalregion, notochord, and yolk endoderm, under the same conditions. The results have shown, that while all tissues have certain features in common, each has nevertheless its specific activities, and these peculiarities coincide, as far as they go, with the activities shown by the respective tissues in the normal embryo. In other experi- ments separate pieces of ectoderm or muscle plates were placed in the lymph close to the nervous tissue, with a view to testing the power oi the former tissues to influence the growth of nerve fibers. For instance in some experiments the medullary cord of the trunk was divided into its dorsal and ventral portions, and each was implanted separately with pieces of epidermis or of myo- tome, in the hope that it might be possible to show in this way that each of these tissues exerted some characteristic influence upon particular kinds of nerve fibers, the epidermis upon the sensory and the muscle tissue upon the motor. The results of the latter experiments were entirely negative; but since they were few in number and since the conditions of experimentation were not ideal, hope that this method may ultimately yield important dis- coveries need not necessarily be abandoned. The total number of preparations made was 211. Permanent records have been kept for 150 of these, the remaining ones having given no promise from the beginning. Of the 150 cases, 35 have been rejected because they were found to bein bad condition before they could be expected to yield positive results. Table 1 shows how the experiments were distributed amongst the various em- bryonic tissues. The specimens were studied almost exclusively under the water immersion lens, D* of Zeiss. In fact this lens is almost indis- pensable for the work. It has such a long working distance that the depths of the preparation can be readily examined without fear of breaking the cover. The magnification obtainable by the combination of this objective with eye-piece No. 4 is about 400 diameters, which is sufficient for all practical purposes. It was ie 806 ROSS GRANVILLE HARRISON TABLE 1 Summarizing the results of the experiments | | a ! ! ' head ve ca |e | ee eae | A 3) me oA mx my < 2 5 2 By z A | B a | BB EE A > 2) 2 Ee 2 | feed) ae | Be ag i e ma S| BBB) SS mt lee me a mS Bi) | weet see ae a eS TISSUE ISOLATED D <3 S| See | Re) ae ge ae a Cn sl dleate aa| 9 2 | o - OF Sie) fe KE fe Bee ee Be fe H = ) ° S) HE) (Ge) See ale O,, ee a8 a | moe) BB Boe) #2 me. | ic) ie} 9 (<2) | Boe a, o | Q = > & S A ica} 4 a § Gs & g24) ge 285 528) 25: | i) p< 5 | paw 4 = | bam) pan] p Ow | Z an on Z Z | 8 Medullary'cord®,)... 2.2 27hee- a ieee a l(a ie il ee: 3, 4 12 2 Branchial ectoderm. ........ peigenly a oa rites t Abdominalectoderm........ bi 18 5 ia sit 5 Axial mesoderm alone....... oe) 8 vip 13 3-54 Notochotd :) 7... S70 ih ages. 3 | } Emdoderm (yeti). L458. ek 2 0 2 | Vawallis: £4. slew. aC eeeeee eee 1684) (26s | 122 os ae ce ee 12 14 { | 1 In many of these cases no attempt was made to exclude all of the axial meso- derm. This accounts for the presence of muscle fibers in some. 2 The excess of this number over the total number of recorded experiments is due to the fact that in some preparations several kinds of tissue were included. ’ This isolated case is one of a series in which the attempt was made to separate the myotomes from the medullary cord along their natural boundary. This is very difficult to do with absolute accuracy and it is supposed that in this case some cells from the medullary cord were left attached to the mesodermic tissue. In subsequent experiments cutting in close proximity to the nervous system was avoided and only the lateral portion of the mesoderm was taken. * The small number of cases recorded as showing striations is due to the fact that the preparations were examined only intoto. Had sections been cut it is believed that the number of positive observations would have been considerably larger. only in certain cases that the oil immersion could be used, and then it was found to have no great advantage over the waterimmersion. A large number of sketches were made, nearly all with the camera lucida. In making these especial care was used to show the length of the fibers, and the form of the end organ correctly. Owing to the extreme fineness of the terminal filaments and the con- stant changes which they undergo, it is not, however, possible always to draw them with absolute accuracy in every detail. Still OUTGROWTH OF THE NERVE FIBER S07 is believed that any deviations which may have crept in have not misrepresented the essential character of the structures. The original sketches were made only in outline. The finished draw- ings, which are reproduced in the plates, were traced from these, details of texture being filled in in accordance with studies made for the purpose. Individual cells, when appearing by themselves, have been in most cases drawn in with the camera, but in indi- cating the larger masses of cells nothing more has been attempted than to give their general character. For instance, the exact arrangement of yolk and pigment granules was not copied because it was felt that this was not essential, and it would have required much time to the exclusion of the study of essential features. Study of the material has been confined almost entirely to the fresh preparations. In fact it must be admitted that one serious defect in the work has been the impossibility of obtaining satis- factory preserved specimens. The ideal procedure would be first to study the growth of a particular fiber, recording the events by frequent sketches, and then to preserve that same specimen, dem- onstrating by suitable histological methods the structural iden- tity between the fibers studied and the nerves found within the embryo.'® Owing to the extreme delicacy of the structures and to the almost fluid consistency of the lymph drops, it has, however, been impossible to do this, since the mere immersion of the prep- aration in any fluid brings about a disarrangement of the tissue, and in many cases the clot with the implanted tissue becomes loosened from the cover, or the tissue falls out of the clot. The method which has given the greatest promise is fixation in osmic acid vapor with subsequent hardening in Tellyesniczky’s bichro- mate acetic mixture, and staining in alcoholic haematoxylin by the method of Oskar Schultze ’04. In some of these preparations 10 Since this was written Dr. M. T. Burrows of the Rockefeller Institute, while working with me has devised a satisfactory method for obtaining permanent preparations. He has shown that embryonic nervous tissue of the chick, when isolated in the proper medium, gives rise to the same long filamentous proc- cesses as does that of the frog; and further, that by staining the preparations in Held’s molybdenum haematoxylin the neurofibrillae in these filaments are brought out very clearly. Anaccount of this work will be published at an early date. SOS ROSS GRANVILLE HARRISON isolated cells of various kinds have been well preserved (fig. 12) but satisfactory preparations of the nerve fibers have not been obtained. Some of the preparations have been cut into serial sections. Nerve fibers were found within them, but in all cases they were broken off at the surface of the tissue. This defect in method has in a measure been offset by the experi- ments described elsewhere (’10) in which the nerve fibers from the medulla oblongata were shown to have grown into a blood clot implanted in their path. General description of materval The developmental processes which have been observed in speci- mens prepared as described in the last section involve only the histological differentiation of the tissues. The gross morpholog- ical changes have no resemblance to those which take place within the embryonic body. Thisis as might be expected even on purely mechanical grounds, for the stresses and strains which are brought to bear upon the developing organs when enveloped in the fibrin must be entirely different from those within the intact embryo. From the time when the tissue is implanted in the lymph it shows a tendency to spread out (fig. 16), and often broad laminae made up of a single layer of cells (1) arefound atthe periphery of the mass, while individual cells may move off entirely by themselves. This is the case with both nervous and axial mesodermic tissue, as well as with pieces of ectoderm, though the latter more often roll themselves into complete spheres. One notable peculiarity that has frequently been observed is the formation of large round - or oval openings in the flattened tissue (fen), which may be sur- rounded by very narrow bands or rings of tissue with cells some- times in single file (cd). This phenomenon may possibly be due to the mechanical action of the fibrin upon the implanted tissue, but the spreading out of the cells into thin sheets seems to result largely from the activities of the cells themselves. ‘These activi- ties, which are common to several tissues, in fact to all except the-very inert yolk-laden endoderm and, perhaps, the notochord, may be referred to a form of protoplasmic movement having its OUTGROWTH OF THE NERVE FIBER 809 seat in the hyaline ectoplasm found at the angles and sometimes at the borders of the cells. The movement cannot be observed clearly in the larger masses of cells on account of their opacity, but it may be seen very clearly in those cells which leave the main masses and wander off by themselves. These cells are irregular in shape, varying from unipolar to multipolar form and having — a varying amount of ectoplasm at their angles (figs. 23 and 27). The movement is amoeboid in character and results either in a change in shape of the cells or in their movement as a whole (text fig. 2). Such cells are found usually in greatest numbers in prep- arations of the medullary cord, and it is here that they are most active, though cells from the mesoderm are often quite similar. However, it is only the protoplasm of cells from the medullary cord and from the cranial ganglia (branchial ectoderm), that gives rise by its movement to long fibers. Cells of the epidermis show their power of movement in somewhat different form. As has frequently been observed, the general tendency of isolated bits of epidermis is to round off into small vesicles, which, when left in water, may move about for days by means of their cilia. Within the lymph the same thing frequently takes place, although there is apparently greater resistance to the process of rolling up, and the cells may often remain together in the form of extensive sheets. Along the free border of these sheets of cells there often appears a fringe of hyaline protoplasm, which undergoes continuous amoe- boid changes (figs.13 and 14pl.fr.) In one ease of this kind it was observed that the sheet of cells gradually spread out toward the side on which this fringe was placed. Since the work of Peters (85-89) it has been generally admitted that wound healing in the epidermis is primarily due to the movement, in part amoeboid, of the epithelial cells, so that it seems quite possible that in this fringe of hyaline protoplasm above described, we have one part of the mechanism by which the movement of cells in wound healing is brought about. The most inert of all the tissues is the endo- derm, which will remain for days in the lymph, practically un- changed, gorged with yolk and devoid of hyaline ectoplasm. The notochord is also very inactive, although large pieces of this structure may show after a time the early stages of normal differ- 810 ROSS GRANVILLE HARRISON entiation, unaccompanied, however, by growth, 7.e., increase in length. The changes which take place through the protoplasmic activity of the embryonic cells can usually be distinguished from those which are due to the action of the clot or the sudden spreading out of the drop of plasma. Likewise the fibrin can readily be dis- tinguished from the hyaline protoplasm of the cells, although even in the fresh specimen it varies considerably in appearance. Some- times the fibrin filaments, in spite of their extreme fineness, are plainly visible, and in other cases there are comparatively few to be seen. ‘They may be found singly or in bundles, and often run for a long distance in a straight line, or sweep around in circles, the individual filaments running from one strand to another. The threads are seen to radiate from the transplanted tissue, and often they may be traced from the hyaline ectoplasm of the embryonic cells, upon which they apparently exert considerable tension. This may result in drawing out the ectoplasm to a narrow fringe (figs. 9, 10, and 28), which differs, however, from the fringe of active protoplasm described above, in that it doesnot continually undergo changes in form. Evidence of still greater tension is found in cells which are drawn out into spindle shape, and which often seem to be pulled along bodily, as may be seen in figs. 9, 10 and 11 which show three successive views of the same cell (cfy.) Sometimes long chains of cells in single file or slightly overlap- ping one another may be formed. Direct evidence of mechanical tension may be had in observations like the following: ie 5 . "ewe Je Oe > eee M ey < 2) ‘got set alee” eo = eo yaad . 4 fora one 7 tin dt id , 4 Te tlagnie Lib ; z pg:s 9 t tu : “yt le ; tnews 7. : i Dee «4 ee i A Ww ae 7 ‘ ug ae | v irs = - PLATE 2 EXPLANATION OF FIGURES All figures except figs. 12 and 15 were drawn from camera lucida sketches of living specimens of embryonic tissue isolated in clotted lymph. 7-11 Five views of the same group of nerve fibers made at different times (ex- periment Is, 137), medullary cord tissue from R. palustris, 3.38 mm. long, lymph from R. pipiens (the interval between the first and last figure represents 34 hours). x 350. 7 Apparently single fiber (nf) growing out from a pointed cell (ct,) which pro- jects from a mass of cells (ms) one day after isolationof tissue. April 28, 1908, 12.25 p.m. 8 Same fiber,2p.m. Fiberis now clearly double. 9 Same group of fibers. 10.25p.m. Four distinct fibers (nf;-nf,) are now vis- ible. The fibrin filaments (thr) shown in this figure were present in the earlier stages but were omitted from the original sketches. 10 Samegroup. April29,lla.m. nf; possibly a branch of nf.. 11 Same group. 10.30 p.m. Continuation of nf; and upper branch of nfo, unfortunately left out of sketch. Note migration of cell (ctz). Identity of other isolated cells in figs. 10 and 11 is uncertain. 12 Three cells from a specimen preserved five days after isolation. Osmic vapor followed by Tellyesniczky’s fluid, stained in O. Schultze’s haematoxylin. © The cells have much branched processes which end indefinitely in the coagulum ~ which pervades wholespecimen. Many isolated cells of this kind are in the speci- men, which is quite typical. Experiment Is, 157, medullary cord from R. palus- tris embryo; R. clamitanslymph. X 350. 13. Row of ectoderm cells from the abdominal region, showing fringe of amoe- boid hyaline protoplasm (pl.fr.). Experiment Is, 75, two days after isolation. Tissue from palustris embryo in palustris lymph. X 350. 14 Similarspecimen. In this case the ectoderm cells, which are taken from the branchial region, show the cuticular seam (cut). The hyaline fringe (pl.fr.) be- longs to cells lying below the main row. Experiment Is, 87, one day after isola- tion. Tissue from palustris embryo in palustris lymph. X 350. 15 Three cells from the medullary cord of frog embryos about 3.3 mm. long, ~ in which the medullary folds had closed and the tail bud was just beginning to appear, prepared from the living specimens; a and b taken from an embryo of R. sylvatica, dissected and examined in 0.2 per cent NaCl; c taken from an embryo of R. palustris, examined in tap water. The latter cell (c) has imbibed water and the cell membrane is very distinct at one side. Nucleus shows as a clear space in each cell. X 350. 16 Whole piece of tissue (medullary cord, with small portions of muscle plates attached) isolated in lymph, two days after preparation. The dark area repre- sents a thick opaque mass of tissue. Thin sheets of cells () and isolated cells are shown on all sides. nf, nerve fibers projecting out into lymph from under the masses of cells. fil, threads of hyaline protoplasm bridging spaces between masses of cells. cd, band of cells in single file. Experiment Is, 124. Tissue from pipiens embryo, lymph undetermined. X 32. OUTGROWTH OF THE NERVE FIBER PLATE 2 ‘ROSS GRANVILLE HARRISON if thr nf 24 hours nol 34 hours 13 , pl fr nfs 58 hours THE JOONNAL OF EXPERIMENTAL ZOOLOGY, VoL. 9, NO. 4. 4. G. Harrison, del. See: ¥ 2 avrsy 1 e 0 7 ITITRDRY POST ~ . i . a Phe r sail ; t 4 v shay emg La 9 PLATE 3 EXPLANATION OF FIGURES All figures were drawn from camera lucida sketches of the living specimens isolated in clotted lymph. . 17 Plexus of nerve fibers growing out from a mass of transplanted medullary | cord. Experiment Is, 124. Two days after operation. Pipiens tissue, lymph undetermined. X 350. 18 Bipolar cell with protoplasmic processes. a, Free end of process; b, process — connecting with mass of cells not shown in figure. Length 300u. Fiber probably formed through movement of cell. Experiment Is, 124, four days after isolation — of tissue. Pipiens tissue in undetermined lymph. X 350. | 19 Bipolar cell and protoplasmic fiber. In this case the fibers were both stretched between two groups of cells and may have been formed by drawing apart. No free ends were visible. Experiment Is, 87, two days after isolation. Tissue from branchial region of embryo. Embryo and lymph R. palustris. X 350. 20 Long nerve fiber arising from unipolar cell (ct) at edge of group of cells (ms). Experiment Is, 124. Tissue from medullary cord, three days after isolation. Pipiens tissue, undetermined lymph. X 350. 21 Isolated unipolar nerve cell with long bifurcated nerve filament. Tissue from R. palustris in lymph from R. pipiens. Experiment Is, 137. Seen at4p.m., — two days after isolation. X 350. 22 Same cell as in fig. 21, as seen at 8.45 p.m. (43 hours later). 23 Two cells from a preparation of medullary cord. Experiment Is, 153, three days after isolation. Tissue from R. palustris, lymph from R. clamitans. These forms are typical of the isolated cells found in. the majority of the prepara- tions. XX 350. 24 Two pigment cells from a preparation of medullary cord, including some mesodermic tissue, thirteen days after isolation. Experiment Is, 133. Tissue and lymph R. pipiens. X 180. 25 Same cell as a of fig. 24, fifteen days afterisolation. X 180. 26 Same cell as a of fig. 24, eighteen days after isolation. X 180. 27 Group of cells from medullary cord showing protoplasmic processes and connecting threads. Experiment I[s,124. Three days after isolation. Tissue R. pipiens, lymph undetermined. X 350. 28 Nerve fiber (nf) extending out from mass of cells (ms) to show contrast with fibrin thread (thr). The fibrin shown is attached to the ectoplasm of sey- eral cells upon which it apparently exerts considerable tension. Experiment Is, 69, three days after isolation. The outgrowth of the nerve fiber was observed the previous day. It now shows signs of incipient degeneration. Tissue and lymph from R. palustris. X 350. 29 Plexus of nerve fibers arising from a mass of cells taken from medullary cord. The anastomoses were not permanent. The one at x was seen to separate, and the day following, nearly all had been resolved. Chr, pigment cell. Experi- ment Is, 200, three days after isolation. Tissue and lymph R. pipiens. X 200. OUTGROWTH OF THE NERVE FIBER PLATE ‘ROSS ORANVILLE HARRISON : 564 hours 52 hours 15 days | 18 days ve § 8.0, Marelaon, dele TIE JOURNAL OF BXPEIUMENTAL ZODL0GF, VOI. ), NO. 4. z +, (os yer ae 4 “ my a: Watt “vai aw yee ther ase Oo. ee at . Pay a“ - . ¥ : wi ‘ 4 4 : i - ANS ad | 3 » ‘J ’ e.4 a q 7 wk > at aa Vay ent os — "y ! * a. r ave RAs? ~ a A, -_ ise ae. M3 ey ‘ ' wr ‘ : = ' » SS Fy i : ~ a i id oe " ’ y 7 “ ‘@ j 1 . 7 THE STRUCTURE AND FUNCTION OF THE ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI GEORGE C. PRICE Professor of Zoélogy, Leland Stanford Jr. University FOUR FIGURES At the anterior end of the kidneys of the myxinoids, in the peri- cardial cavity, is a pair of small bodies, which have been known for some years as the head kidneys or pronephroi. They were first observed by A. Retzius (’26) in Myxine glutinosa, and were interpreted by him with some hesitation as the functional kidneys. Johannes Miller (45), who was che first to give an accurate description of tne functional kidneys or mesonephroi, looked upon the kidneys of Retzius as suprarenal bodies. He worked on both, Myxine glutinosa and Bdellostoma fos ‘eri. Thirty years later Wilhelm Miiller (75), working on Myxine, described these organs more fully aad more accurately than had been done by either of his predecessors, and came to the con- clusion that they represented head kidneys or pronepbroi. This interpretation has been generally accepted, and these organs have have been homologized with organs of the same name in the larvae of amphibia and bony fishes. Since the appearance of Wilhelm Miiller’s article the head kid- ney of Myxine has been made the subject of investigation by Kirkaldy (94), Semon (’96), Spengel (’97), and Maas (’97); while Weldon (’84) has described it for the genus Bdellostoma, his work having been done on Bdellostoma fosteri, from the Cape of Good Hope. All these workers, with the exception of Maas, have labored under the disadvantage of having only adults at their disposal. The latter was particularly fortunate in obtain- . ing three youag specimens, the smallest only 8.5 cm. in length, 850 GEORGE C. PRICE in which the head kidneys were less complicated than in the adult, and he was thus enabled to settle conclusively certain disputed points. His paper contains an excellent review of the literature of the subject. The present work has been done on Bdellostoma stouti. This is the only myxinoid in which the development of the excretory organs is at present known (Price ’97, ’04). and hence it offers a peculiar advantage for the study of the adult organs, as their somewhat complicated structure may be interpreted in the light of embryology. This, together with the discovery of a probable function of the head kidney, is the excuse for the present paper. With the exception of a very few species of bony fishes, the myx- inoids are the only Craniota described as having a persistent head kidney. As was pointed out by Johannes Miller, and as may readily be seen by a glance at one of his figures, which has been widely copied in text-books of comparative anatomy and embry- ology, the functional kidney in the myxinoids is very simple, much simpler than in any other of the Craniota; in fact it is more like the kidney of an embryo than of an adult. From this, as well as from the systematic position of the group, one might expect that the head kidney would likewise be simple, and that a study of the excretory organs of the adult would throw light on the question of the homology of the pronephros and mesonephros. But this has not proved to be the case; the head kidney is less simple than the functional kidney, and the primitive relations between them, which is clear in the embryo, is lost in the adult. A study of their development proves conclusively that these two organs are ho- mologous, but this is the only way in which one can be sure of the fact. In the embryo the excretory organs arise as a series of segmen- tally arranged tubules, opening into the coelon, and extending in the specimens studied from segments 11-13 to segments 79-82, the exact point both of beginning and of ending varying in differ- ent individuals and also on the two sides in the same individual. The excretory duct appears later than the tubules and arises from them. Thus it will be seen that in its origin the entire organ has the characteristics of a pronephros. One pronephric character- ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 851 istic, however, is wanting, for at no time is there a glomus in the coelon opposite the nephrostomes. Later the tubules back of segments 30—33 (this point also vary- ing) lose their connection with the body cavity; or to be more accurate, the small coelomic pocket into which each of tnese tu- bules open become cut off from the general body cavity, and forms the distal end of the tubule. In the end of each tubule thus formed, that is, in a cut off portion of the coelom, a glomerulus appears. This portion of the organ has now the structure of a simple mesonephros, a structure which it retains throughout life. However, a few of the more posterior tubules degenerate, and, as will be mentioned below, two or three of the a 1terior ones become incorporated into the head kidney of the adult. The tubules in front of the ones just described retain their con- nection with the coelom, and for a time they also retain their segmental arrangement. But later the gill slits, which are at first far forward, shift their position backwards, and in so doing they crowd before them the anterior end of the kidneys, so that all of the open tubules (17 or 18 in number) and 2 or 3 of those which have been cut off from the coelom and in which glomeruli have been formed, are crowded together into a compact body occupying the space of one or two segments. Itis this body which, after undergoing some further changes, forms the head kidney of the adult. It will be seen that this is in a sense a composite struec- ture, since it is formed of two kinds of tubules. The first develop into the main body of the gland, while from the second is formed what may be called the glomerulus of the head kidney. This is the only structure of the kind found in the organ, as neither glomus nor glomeruli are formed in connection with the open tubules. In the present work adults ranging in length from 22.8 cm. to 56.5 cm. were used, and also two young individuals, the one 7.5 and the other 7.9 cm. in length, in which the intestine still con- tained an abundance of yolk. While the head kidney in these small specimens has the essential characteristics of the adult they are still in some ways less complicated, and formed a sort of transition between the oldest embryos before studied (Price, 04) and the adult. 852 GEORGE C. PRICE More than two dozen head kidneys, in some cases including also the anterior end of the functional kidney, were sectioned and mounted in complete series. While these agree in essential fea- tures there is still much variation in detail, so much that an exacé description of one would not answer in detail for anyother. Liv- ing as well as preserved material was studied. In the adult, as is well known, the head kidneys are situated a little to either side of the dorsal aorta, and project into the dor- sal part of the pericardial cavity. They are intimately connected with veins returning blood from the anterior part of the body, the left with the anterior cardinal, or rather with a wide diverticulum given off from this vein just before it reaches the heart; the right withasimilar diverticulum given off from a vein which empties into the portal heart. This vein is called by Weldon and bysome other authors the anterior portal, but from its position and distribution it looks to be the fellow of the cardinal of the opposite side, al- though it does not extend nearly so far forward. The relation of the head kidney to the vein is well shown in fig. 2. In Bdellos- toma fosteri, according to Weldon, the head kidneys are likewise connected with veins returniag blood from the anterior part of the body, while in Myxine they are connected with the posterior cardinals returning blood from the posterior part of che body. The size of the head kidney varies with age, although it is relatively small even in large animals. In one specimen 24.7c¢m. long the head kidney measured 3 mm. long by 1 mm. broad, while in one 45.6 cm, long it measured 8mm. by 2mm. The ratio, however, between the length of the individual and the size of the head kidney is not constant; nor are the two organs of the same size in the same individual, sometimes the right being larger and sometimes the lef. As a rule the organ forms a single compact body, although this is not invariably the case. In one instance, for example, a small buach of tubules at the anterior end was separated from the main body by aninterval of more than amillimeter. Development offers a ready explanation for cases of this kind. In some of the older embryos one or two of the anterior tubules were observed to be entirely separated from the rest and from the duct, and it seems ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 853 Fig. 1 Head kidney showing the division into lobes. The nephrostomes which lie close together over the surface could not be represented with the magnifi- cation used. )) it 7, & te LS .) i S ng a\e a UUERT es UG Te Lr YY . aaa Fig. 2 Transverse section through the head kidney, passing through two lobes; in the one on the right the tubules are cut for the most part longitudinally, while in the one on the left they are cut for the most part transversely. v, venous di- verticulum with which the head kidney is connected; d, central duct containing blood corpuscles and showing connection with both the tubules and the venous diverticulum. A few sections both in front of this and back of it the duct forms a complete ring lying freely in the venous diverticulum. neph., nephrostomes; sin., sinusoids. 854 GEORGE C. PRICE certain that the bunch of tubules just mentioned arose from the branching of such an isolated cubule. In another case there was a short but complete break about the middle, so that for five sec- tions there was no trace of any part of the organ. Other examples of the same nature might be given. In the great majority of cases there is no connection whatever between the head kidney and the functional kidney. There isa short but complete break between the posterior end of the one and the anterior end of the other. In one case no break could be ob-, served, and it was thought the two must bein some way connected, but upon sectioning it was found they simply overlapped, and that there was no actual union. In another case, however, the cavity containing the glomerulus of tne head kidney was joined to the duct of the functional kidney by a tubule, and also to the duct of the head kidney. In no instance was the duct of the head kidney connected with the duct of the functional kidney even by arudimens. Butit would not be surprising if such a ease should be found, since we know that the two are parts of one and the same duct in the embryo, and that they become separated by the degen- eration of a small portion of this duct. When only the adults were known, the presence or absence of even a rudimentary con- nection between the head kidney and the functional kidney was of considerable interesi as throwing light on the probable relation between the two; but in view of the solution of this part of the problem furnished by embryology this ceases to be of any particu- lar significance. A study of sections shows that the main body of the gland is made up of sumerous tubules which open through narrow nephros- tomes into the pericardial cavity, and, passing inwards, unite with one another, and finally with the central duct (fig. 2). New tubules are being formed throughout life. These branch out from the old tubules just back of the nephrostome, and while still very short acquire an opening into the pericardial cavity. The new tubules, along with all the rest, grow in length, and may in the course of time give rise by branching to still other tubules. It seems certain that new tubules never grow out from the central duct.- If all the tubules in the adult are derived by branching Or ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI -° 8&5 from the seventeen or eighteen primary tubules of the embryo, one would expect to find about that number of main stems in the adult through which all of the tubules would be connected with the central duct. But this is not the case. Occasionally such a stem is found, but usually instead of a single stem two or three tubules are seen opening into the duct together. This may be brought about in two ways: the main stem (which is never long) may in some way become so shortened that the early secondary tubules appear to open directly into the duct; then again, on account of the crowding together of the tubules in the embryo two primary tubules may come to open into the duct very close together. For these reasons it is difficult in a series of sections to determine what must have been the number of primary tubules. And yet with care this may be done with considerable accuracy. In one case the duct was reconstructed on millimeter paper, and it was found that tubules connected with it at twenty places. This is about the number of primary tubules of both kinds found in the head kidney in the embryo. When the entire organ is viewed with a lens it presents a more or less distinctly lobed appearance (fig. 1), as has been observed by previous workers in other members of the group. Judging from what may be seen of the lobes and of the course of the tubules in a stained and cleared preparation, it seems probable that each lobe is made up of a single primary tubule, together with all of its branches. The number of distinct lobes, however, is usually not so great as one would expect if this were the case. The two chief causes for the increase in the size of the head kid- ney with age are the growth of the tubules in length and their increase in number; the diameter changes little if any. The fol- lowing figures will give an idea of the increase in length; in an individual 7.5 em. inches long, in which only part of the tubules showed any indications of branching, the average length of a few of the primary tubules was 0.165mm. ;in an individual 27.9cem. long the average length of a few of the longest tubules was 0.292 mm.; in one 35.5 em. long it was 0.464 mm.; and in one 45.7 em. long it was 0.90 mm. 856 GEORGE C. PRICE The tubules are so numerous and so closely packed together that it was entirely out of the question to try to count them after sectioning. While an idea of the number in a small individual could be gained by first staining the entire organ and then counting the nephrostomes under a strong lens, this method could not be used with large individuals, in which the number ran into the thousands. But by first staining the entire organ, and then teasing into small pieces in glycerin, and counting under the com- pound microscope, a rough idea of the number of tubules in several individuals of different sizes was obtained. In this way 265 tu- bules were counted in a head kidney from an individual 29.2 cm. long, 791 in one from an individual 34.2 em. long, 6084 in one from an individual 49.5 em. long in which the head kidney was was unusually large, and 2300 in one from an individual 56.5 cm. long. In all cases the actual number must have been greater than the number counted, for it was not possible to get all the pieces so small that every tubule could be seen. In the case where 265 tubules were counted at least 300 nephrostomes were seen before teasing. It will be noticed that the largest and presumably the oldest individual did not have half so many tubules as the one next smaller. Perhaps the first explanation that would sug- gest itself for this would oe that it was due to degeneration and the consequent disappearance of tubules. But this is not likely the case, for so far as possible in counting, every tubule was ob- served, and no indication of degeneration was noticed. Further, this is not so surprising when we take into consideration the great amount of variation in all parts of the organ. The average diameter of a number of tubules from three individ- uals of different sizes was 0.055 mm., and the average diameter in the largest individual was no greater than in the one under three inches long. As a rule there is but slight variation in the size of the tubules in an individual; but in some instances, espe- cially in larger individuals, aportionof atubulemay be so enlarged as to form a vesicle as much as four times the diameter of an ordi- nary tubule. These vesicles were absent in more than half the head kidneys examined. When present they varied in number from two to seven. As has been observed in other members of ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 857 the group, the tubules become slightly smaller at the pericardial end, so that the nephrostome is the opposite of funnel-shaped. The tubules are formed of columnar epithelium which is con- tinuous on the one hand with the epithelium of the central duct and on the other wita the flat pericardial epithelium. The tubules are ciliated, as could be clearly seen in part of the preserved material, and still more clearly in living material. W. Miiller, working with living mas‘erial, was unable to find cilia in Myxinc, but Maas found them in his young specimens. Weldon did not find them in Bdellostoma fosteri. Lying between the tubules are sinusoids, which, as may be seen in fig. 2, are connecied with the large venous diverticulum in which the head kidney is in a way imbedded. These sinusoids, together with some small veins from the glomerulus to be men- tioned later, form the only blood supply found in connection with the main body of the gland. It should be mentioned, however, that no well injected material was studied in sections. In the head kidney of Bdellostoma stouti, in the adult as well as in the embryo, there can be no ques‘ion about the existence of a central duct. This is a point of some interest, since there is a difference of opinion as to whether or not such a duct is present in Myxine, the earlier workers, W. Miiller and Kirkaldy affirming, and the later workers, Semon, Spengel and Masa denying 1s existence. Maas’s work on young individuals seems to prove conclusively the correctness of the latter view. In the embyro of Bdellostoma stouti the duct in some eases fails to develo) between some of the anterior tubules. If this should be carried far enough it would result in the formacion of a ductless head kidney. Perhaps this is the explanation of the condition found in Myxine. Weldon describes a central duct for the head kidney of Bdellostoma fosteri. His figures seem to prove that he was correct, although Semon, working on a single poorly preserved specimen, came to the conclusion that Weldon’s duct was in reality a venous sinus. The duct may be continuous throughout the length of the head kidney, and may even extend alittle beyond it at the poste- rior end; or it may be broken into two or more distinct parts: S58 GEORGE C. PRICE or again, the duct may be continuous and the lumen not, owing to the duct having become solidinone or more places. Different sec- tions through a head kidney show great variation in both the size andshape of the duct. In the youngest specimens studied the duct, like the tubules, was lined by a single layer of columnar epithelium, and this is true for the most part also in the adult, But in places the wall of the duct may become greatly thickened by an increase inthe numberof epithelial cells. These change their shape, and be- come much more loosely arranged, so that the tissue loses entirely the structure of columnar epithelium. The distribution of this sissue along the duct is very variable. It increases somewhat in amount with age. In all specimens studied, even in the youngest, openings were found in the wall of the duct (fig. 2). These place the lumen of the duct in direct communication with the venous diverticulum with which the head kidney is connected. These openings are natural and not artificial. A single opening may extend through asmany as a dozensections, though usually they aremuch smaller and extend through not more than one or two sections. The num- ber in a single duct varies from two or three to ten or twelve. The fact has already been mentioned that the tubules open on the one hand into the pericardial cavity and on the other into the duct. Thus it will readily be seen that the tubules, the duct and the openings in the wall of the duct form a direct connection be- tween the pericardial cavity and the circulatory system. Lying along the median side of the head kidney, toward the posterior end, there is a glomerulus, the origin of which has al- ready been described. This, like other parts of the head kidney, is subject to much variation. Sometimes it is about as long as it is wide and does not extend so far posteriorly as the rest of the organ; and again it may be two or three times as long as it is wide and extend beyond the posterior end. Usually it is a single body; but it may show indications of having been formed by the fusion of two or more glomeruli. Occasionally there may be two or three distinct or almost distinct glomeruli lying close together, one after the other. In sections it preseats the same appearance as the glomeruli of the functional kidney, with which embryology shows it to be strictly homologous. ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 859 The artery supplying the glomerulus, in most cases at least, is not given off directly from the aorta, but is a branch of an intersegmental artery. (This is true also for the arteries supplying the glomeruli of the functional kidney.) The artery may divide into two or three branches before entering the glomerulus. The veins carrying blood away usually pass forward along the central duct, branching as they go into a few smaller veins, which in the material studied soon became lost. There is no necessary agreement between the number of arteries and the number of veins, nor between either of these and the number of elements which have united to form the glomerulus. The glomerulus lies in a cavity formed by the fusion of the cavities of the Malpighian corpuscles of the embryo. In cases where the glomeruli remain distinct this fusion may be only partial, the different cavities, each occupied by a glomerulus, being con- nected with one another by small openings. As a rule the cavity is in communication with the pericardial cavity through a variable number of apertures, though this is not always true. The aper- tures may be very small, or they may be so large that in some sections the glomerulus presents the appearance of hanging freely in the coelom, like the glomus of a pronephros. The cavity is connected with the central duct by short tubules. These resemble the other tubules of the head kidney so closely that in seciions where both occur the two can not be certainly distinguished except by tracing them to their termination. They project into the cavity containing the glomerulus, where they open through nephroscomes (fig. 3); and in some cases they resem- ble the other tubules still more closely by branching and opening by from two to five nephrostomes. In a single case a tubule was seen to divide into two branches, one of which opened into the cavity of the glomerulus and the other into the pericardial cavity. The number of tubules is not always the same as the original number of glomeruli. In one instance where there were three distinct glomeruli there was but a single tubule, and in another case there was no tubule at all. It must be left a ques- tion whether these tubules are ciliated, since none of them were examined in the living state; but in sections some of them had 860 GEORGE C. PRICE the appearance of being ciliated. In asingle case already referred to the cavity of the glomerulus, in addition to being connected with the duct of the head kidney, was connected also by a tubule Fig. 3 Transverse section through the region of the glomerulus. v, venous diverticulum; d, central duct with a tubule just opening into it; gl., glomerulus; t, two tubules; the lower one opening by a nephrostome into the cavity containing the glomerulus, while the upper one is beginning to divide into two branches, both of which open into the cavity afew sections farther on. A tubule is seen lying freely in the venous diverticulum; farther on this divides into two branches, one of which opens into the pericardial cavity and the other into the cavity of the glomerulus. This is the only case of the kind met with. witn the duct of the functional kidney. Occasionally a glomerulus may be found between the head kidney and the functional kidaey, but connected wiih neither. ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 861 In Myxine, thereis a glomerulus connected with the head kidney, which is so much like the one in Bdellostoma stouti that it seems it must have ariseninthesame way. We know from the work of Semon that in some cases pronephric tubules open into the cay- ity of the glomerulus and in some they do not, and also that the cavity may be connected with the duct of the functional kidney. Spengel likewise found tubules opening into this cavity, and in addition found the cavity communicating with the pericardial cavity. Weldon mentions a glomerulus at the posterior end of the neph ‘Cu Fig. 4 Diagrammatic longitudinal section. cv., cardinal vein; v, venous divertic- ulum; sin., sinusoids; d, central duct with openings into the venous diverticulum; neph., nephrostomes; gl., glomerulus; a., artery of the glomerulus; vg., vein of the glomerulus. The cavity in which the glomerulus lies opens directly into the pericardial cavity and is connected with the central duct by two tubules, one of which has two nephrostomes. ° head kidney of Bdellostoma fosteri, but says nothing about a connection with either the pericardial cavity or the central duct. In a diagrammatic figure he represents a rudimentary connection with the duct of the functional kidney, and this he calls an atro- phied portion of the segmental duct, although it is more ae an atrophied tubule. Lying in front of the glomerulus in Myxine, and connected with the inner ends of the tubules, is a mass of peculiar tissue, in which arteries from the aorta ramify. This was looked upon by both Kirkaldy and Semon as being a mass of glomeruli. Spengel, on THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 9, NO. 4. 862 GEORGE C. PRICE the other hand, was unable to find that it possessed any of the characterisiics of a glomerulus, and concluded that it had arisen from the modified epithelium of the inner ends of the tubules. Maas’ work oa young individuals seems to demonstrate the correciness of Spengel’s view. This peculiar tissue is not found in Bdellostoma stouti, unless it be represented by the above- mentioned modified epithelium of part of the central duct. But this is never large in amount, nor is it supplied with arteries. The distribution of arteries in the head kidney of Myxine, well shown in the young individuals studied by Maas, is of interest as suggesting the possibility of glomeruli having once been present throughout the entire length of the excretory organ. In examining sections of the head kidney blood corpuscles were observed in both the tubules and the duct; and since the duct communicated directly with the vein, and since the tubules were ciliated and opened into the pericardial cavity, and since further, the pericardial cavity communicated through a large opening with the peritoneal cavity, the thought naturally presented itself that perhaps a function of the head kidney was to transfer lymph from the body cavity into the blood vessels. In order to test this theory, normal salt solution containing powdered carmine in suspension was injected into the peritoneal cavity through the abdominal pore. Twenty-four hours later blood both from the subcutaneous space and from the veatral aorta was examined, and every drop showed particles of carmine, some free and some within the whice corpuscles. When the body cavity was opened carmine was found in the pericardial cavity as well as ia the peritoneal cavity. From another animal a living head kidney was removed and examined under the microscope in salt solation. Powdered carmine was added, and particles were actually seen being swept into and through the tubules by the action of the large and rapidly moving cilia. While these experiments prove that particles may and actually do pass from the coelom into the blood vessels, they do aot demon- trate the fact that they go by way of the head kidneys; they might possibly go in some other way. Still it is hard to escape the con- clusion that this is one of the ways in which the transfer is made. ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 863 In fact, it is difficult to understand how a fluid in the pericardial cavity could escape being taken into the tubules, on account of the great activity of the cilia liaing them, and so on into the veins. It is useless to speculate as to whether the head kidney forms a passage between the body cavity and the blood system in other species Of the myxinoids. Weldon found clotted blood in she central duct of Bdellostoma fosteri, but thought possibly it might have come by way of the veias from the glomerulus. No real evi- dence was given in support of this hypothesis. He worked on complete series of sections, but says nothing about a connection between the central duct and the vein, although one of his dia- erammatic figures would seem to indicate that such a connection existed. It is quite possible that the head kidney may have some other function or functions than the one here suggested. But however this may be, it is certain that excretion can not be one of them, as there is no connection with the exterior; and since the connec- tion is lost before hatching, the organ can not function in the young animal, as is the case with the pronephros of Petromyzon and the Amphibia. 864 GEORGE C. PRICE BIBLIOGRAPHY KirKaupy, J. W. 1894 On the head kidney of Myxine. Quart. Journ. Micr. Sci., vol. 35. Mass, Otto. 1897 Ueber Entwicklungsstadien der Waren und Urniere bei Myxine. Zool. Jhrb., Bd. 10, Anat. MitLier, Jonannes 1845 Vergleichende Anatomie der Myxinoiden, der Cyclo- stomen mit durchbohrenenGaumen. Schluss: Untersuchungen ueber die Eingeweide der Fische.. Berlin. M@t.uer, WitnELM 1875 Ueber das Urogenitalsystem des Amphioxus und der Cyclostomen. Jena. Zeitschr. Naturw., Bd. 9. Price, G. C. 1897. Development of the excretory organs of a Myxinoid, Bdellostoma stouti Lockington. Zool. Jahrb., Bd. 10, Anat. 1904 A further study of the development of the excretory organs in Bdellostoma stouti. Amer. Jour. Anat. vol. 4. Retzius, A. 1826 Beitrag zuden Ader-und Nervensystemen der Myxine Gluti- inosa. Archiv fiir Anatomie und Physiologie. Semon, RicHarp 1896 Das Exkretionssystem der Myxinoiden in seiner Be- deutung fiir die morphologische Auffassung des Urogenital systems der Wirbelthiere. Festschrift Gegenbaur. Leipzig. 1897 Das Excretionssystem der Myxinoiden. Anat. Anz., Bd. 13. 1897a Vorniere und Urniere. Anat. Anz., Bd. 13. SPENGEL, J. W. 1897 Die Excretionsorgane von Myxine. Anat. Anz., Bd. 13. 1897a Semon’s Schilderung des Mesonephros von Myxine. Anat. Anz., Bd. 13. Wetpon, W. F. R. 1884 On the head kidney of Bdellostoma, with a suggestion as to the origin of the suprarenal bodies. Quart. Journ. Mier. Sci., vol. 24. OBSERVATIONS ON THE PERIOD OF GESTATION IN WHITE MICE J. FRANK DANIEL Instructor in Comparative Anatomy, University of California From the Zoélogical Laboratory, University of Michigan Preliminary to a study which I am making on white mice, I have found it necessary to reéxamine in detail their period of gestation. In this examination I have ascertained facts which, so far as I am aware, have not been previously described. The general statement that the period of gestation in white mice is 21 days,! needs considerable modification. Such may be the case, but the likelihood is against it. In fact it may be said that no unqualified statement can be made as to a definite period of gestation for all mice—since this period depends upon the state of the individual mouse. My plan of study, which has consisted of isolating the female immediately upon an observed copulation, and of keeping her apart from the male until the birth of her young—has been greatly facilitated by the use of observation cases of glass, in which the mice were kept under almost constant attention during the entire period of gestation.2. Under this condition, by observ- ing the moment of copulation and the time of birth, I have been able to determine in many cases the period correct to the hour. 1 Allen, Glover M., 1904. The Heredity of Coat Colorin Mice. Proc. Am. Acad. of Arts and Sciences, 40, no. 2 (p. 70). 2 Strictly speaking, the period of gestation is that interval of time elapsing be- tween the act of fertilization and birth. In this paper it will be noted however, that I have used the term rather broadly to cover the period between the act of copulation and birth. © S66 J. FRANK DANIEL In a consideration of the subject of gestation mice fall natur- ally into two well defined groups: (1) those that suckle and (2) those that do not suckle young during pregnancy. The latter class of non-suckling mice includes (a) all cases of females carrying their first litter, and (b) those cases in which copulation took place after the young were weaned, or those cases in which copulation took place and the young were separated from the mother or were early lost. For all such, a definite gestation period of practically 20 days may be stated; and this holds true irrespective of size of litter born. While 20 days may be considered as the average period for non- suckling mothers, for suckling mothers the period presents no such regularity. This will be made evident from table 1, in which ten consecutive cases are considered. TABLE 1 DESIGNATION OF | NUMBER OF YOUNG : M EXAMPLE FEMALE SUCKLED PERIOD OF GESTATION NUMBER BORN 1 F 10 30 days 9 2 M 6 24 days 6 3 J 8 27 days ‘i 4 L 3 | 25 days 5 5 M 6 | 24 days 19 hours’ 11 6 H 3 | 22 days 5 7 M 10 | 30 days 3 hours. 9 8 H 5 | 25 days 1 9 | 5 | 24 days 14 hours < 10 F 8 27 days 15 hours 8 In table 1 two things stand out prominently: First the marked variability, and second the lengthening, of the period of gesta- tion. From the table it will be seen that not only does one mother differ from another in period of gestation, but the same mother at different times shows a similar variability. Asan example of the two cases: H (example 6) went but 22 days from the time of copulation to the birth of her young, while M at one time had a period of 30 days + (example 7). Thesame mouse (M) atanother time, however, had a period of only 24 days (example 2). PERIOD OF GESTATION IN WHITE MICE S67 Equally as striking is the lengthening of the period. It is seen that the minimum period of 22 days in the mice under consider- ation is longer than the period (20 days) given for mice of group 1, while the maximum period in this group sometimes exceeds by one- third the period in group 1. It is thus evident that for group (2) with variations running from two to ten days, no definite time as a period of gestation ean be stated. This does not imply, however, that the period of gestation is not orderly. On the contrary, the variability which in table 1 is so evident, gives place, upon a rearrangement of the data, to striking uniformity. What is the factor upon which this rearrangement can be made? From cases of prolonged gestation like 1 and 7, from which large numbers of young resulted, my first impression was that the number born might be a factor in this variability. That this is not the case, however, can be shown conclusively by rearranging the table in order of the number born. In such a rearrangement it is seen that H (example 8), which gave birth to the smallest number, went longer than M (example 5), which gave birth to the greatest number. If on the other hand the nwmber suckled rather than the number born govern the variability in the period of gestation, it would be expected that, given a definite number of suckling young, the period would be practically the same in all cases. This, in fact, is seen to be the case. For mouse M suckling 6 at two different times (examples 2 and 5) had in one case a period of 24 days and in the other of 24daysand19hours. J and F (examples 3 and 10) suckling 8 each, went 27 days, and 27 days 15 hours, respectively. F and M (examples 1 and 7) suckling 10 went 30 days and 30 days 3 hours, respectively. Moreover, this order is so definite that a series increasing from theminimum to the maximum variation may be constructed thus: H suckling 3 went 22 days (example 6), J suckling 5 went 24 days + (example 9), F suckling 8 went 27 days + (example 10), and M suckling 10 went 30 days + (example 7). This sequence may be shown to further advantage by rearrang- ing table 1 in order of the number suckled. Such a rearrangement is made in table 2. S68 J. FRANK DANIEL TABLE 2 DESIGNATION OF NUMBER OF YOUNG PERIOD OF GESTATION . NUMBER BORN EXAMPLE FEMALE SUCKLED 1 H 3 22 days | 5 p L 3 25 days | 5 3 H 5 25 days | 1 4 As 5 24 days 14 hours | z 5 M 7 6 24 days © | 6 6 M 6 24 days 19 hours | 11 Z J 8 27 days 7 8 F 8 27 days 15 hours 8 9 F 10 30 days | 9 10 M 10 30 days 3 hours 9 Indeed, the variation in such cases has been found to follow so definite an order that by knowing the time of copulation and the number suckled, I have been able to predict within a day or so when the litter would be born. This, as seen in the table, can be done by adding to an initial period of 20 days a number equal to the number suckled. The number resulting, barring certain irregularities, which seem to be individual in nature, is usually slightly greater than the number of days in the period of gesta- tion. Thus, 20 + 5, the number suckled by J, will give her period of gestation in case 4, and 20 + 8 will give her period in case 7. In concluding the discussion of this variability, a law may be formulated which for practical purposes is as follows: ‘The period of gestation, in lactating mothers, varies directly with the number of young suckled. The cause, and time of occurrence, of the lengthening in the period of gestation is a problem in itself,a more complete investi- gation of which I shall reserve for a future discussion, but a word concerning it may not be out of place at. this time. If it could be shown, for example, that ovulation were in any way retarded, an effective explanation of the cause of lengthened period would be forthcoming. In this case, however, the length- ening would be previous to the true period of gestation. But es es eae PERIOD OF GESTATION IN WHITE MICE 869 what factor would be capable of thus causing delay in ovulation? It is a belief that in human beings lactation may thus delay or even prevent ovulation. But, if lactation delay ovulation and thus increase the period, it may be asked—why should not the delay be of practically equal duration in all cases? Why should a mouse when suckling few young, go a relatively short time and when suckling more, go a longer period of time? Does this not suggest amount of lac- tation and does not amount suggest nutrition? If it could be shown that lactation, by robbing the developing young of nourishment, causes them to develop more slowly before birth, as malnutrition may retard development after birth, a second explanation for the delay, equally as effective as that of delay in ovulation, would beathand. In this case the delay would be in the period between fertilization and birth and not in the act of fertilization. In either case lactation might be the cause of the delay. But whether the delay is in retarded ovulation, that is, before fertilization, or in nutrition after fertilization, is a problem open to experiment. Reserving a detailed discussion for a future com- munication on the subject, it may be said that two sources of evi- dence seem to show that, at least in mice, lactation does not delay ovulation. By microscopic examination Sobotta, in working over a great deal of material, found that ovulation and parturition occur close to the same time. Lams et Doorme confirm the opinion of Sobotta and add (p. 284), ‘‘En somme, la mis-bas, la rupture des follicules mirs, le rut et l’accouplement se suivent ordinairement trés vite.’’ Long, in a paper now in press, shows that ovulation takes place in from 143 to 285 hours after parturition. From this source the common conclusion is that ovulation is not delayed. From experimental evidence which I now have it is strongly sug- gested that, in no way does lactation lengthen the period before fertilization. A case in point may be given in conclusion. A mouse, designated as ‘‘F’’ which in examples 8 and 9, table 2, had two periods of lengthened gestation, gave birth to young and as is usual copulated the same day. At the end of the fourth 870 J. FRANK DANIEL day the last one of her young died from cold. Lactation, in this case, was installed and continued for four days, but that it did not delay fertilization is shown by the fact that the mouse gave birth to young just 20 days after copulation. BIBLIOGRAPHY Lams, H. et Doorme, J. 1907 Maturations et Fécondation de L’oeuf des Mammifeérs. Arch. de Biol., 23, pp. 259-365. Lone, J. A. 1910 Maturation of the egg of the :nouse. Carnegie Institution Publication. Soporra, J. 1895 Die Befruchtung und Furchung des Eies der Maus. Arch. f. mik. Anat. 45, pp. 15-93. sent to any member of an tly to the Managing Editor a they aresubmitted. Authors ei articles gratis. ble in advance and no dis- A. AS ubseriptions are always umé and not by the year. Re- pads ae postal money ater, or B wisn, R INSTITUTE, OF 6} “4 AND BIOLOGY _ EL PHIA,PA.,U.S.A.° _ OF MORPHOLOGY _ (JouR. MORPH.) \ EY, ManaGine Eprtor OLLEGE, Mass. relating to animal dad on cytology, protozodlogy an i embryolozy of vertebrates and Show 709 Poe in each volume, ext figures and m price $0. 00 per ee postpaid to x arin JOURNAL OF : ene OMy (ant. JOUR. ANAT.) R, Sec’y EpIToRIAat Boar> 'yY OF CINCINNATI - : gations on the anatomy, Seen oe se: vertebrates with special refer- YY. TAboiss 500 pages per Gotaike, Sidece and plates. pron (domestic) ; $5.50 (foreign) ~ NATOMICAL RECORD (ANAT. REC.) RL HUBER, Manactna lipo a ANN: ARBOR, Micu. pur publication of concise rate anatomy, embryology ane eek as may be preliminary reports and ‘hort reviews of noteworthy publi- ergse anatomists, brief laboravory plans and of Alu pages per volume. . # NEUROLOGY AND ICHOLOGY-: ~~ R. COMP. NEUR. PSYCH.) 7K, a eoraers Eprror 55, Do aomese $5.50 (foreign) a, c Zs a . 43 Dike he, ae + Saa ie fea ie” Bes “ae PQs “se a has”. Po a eee cn (foreign) URNAL OF COMPARA ublished by The Wist lar Institute GENERAL INFORMATION THE JOURNAL OF EXPERIMENTAL ZOOLOGY (JOUR. EXP. ZOUI..) ROSS G. HARRISON, Manaaina Epitor Yave University, New Haven, Conn. Publishes the results of original researches of an experimental or analytical nature in the field of zoél-_. ogy, including investigations on growth and develop-" ment, on reproduction, regeneration, cellular phe- nomena, evolution, variation, heredity and ecology, and upon topics in general physiology. Issued eight times a year, in two volumes. About 600 en per volume, with numerous text figures and plates. Subscription price $5.00 (domestic) $5.50 Fea eign) per volume. AxticLes ACCEPTED FOR PUBLICATION By the Journat or MorpHotoer: H. H. New- man and J. Thomas Patterson, The development of the nine-banded armadillo from the primitive streak stage to birth, with especial reference to the question of specific polyembryony; Leland Griggs, Early stages in the development of the central nervous system of Amblystoma punctatum: R. J. Terry, The morphology of the pineal region in Teleosts; Gideon S. Dodds, Segregation of the germ cells of the Teleost lophius; H. S. Jennings and G. T. Hargitt, Characteristics of the diverse races of Paramecium; J. Parsons Schaeffer, The lateral wall of the cavum nasi in man, with especial reference to the various developmental stages. By THe ANatomicaL Recorpb: Geo. S. Hunt- ington, The genetic principles of the development of the systemic lymphatic vessels in the mamma- lian embryo; A. T. Kerr, Complete double ureter in man; R. O. Moody, Some features of the histo- — of the thyroid gland in the pig; F. W. hyng, The anatomy of a 7.8 mm. pig embryo; Alfred Jerome Brewn, A note on_ post-cardinal omphalc=mesenteric communication in the adult mammal; Edwin E. Reinke, Note on the presence of the fifth aortic arch in a 6 mm. pig embryo. By THE AMERICAN JOURNAL OF ANATOMY: Wil- liam F. Allen, Distribution of the lymphatics in the tail region of scorozenichthys marmoratus; Franklin P. Johnson, The development of the mucous membrane of the esophagus; stomach and small intestine in the human embryo; Edwin G. Kirk, Ou the histogenesis of gastric glands; Leonard W. Williams, The somites of the chick; Florence R. Sabin, Description of a model showing the tracts of fibers medullated in 4 new-born baby’s brain. By THe JoURNAL OF COMPARATIVE NEUROLOGY aND PsycHo.oey: L. W. Cole, Reactions of frogs to chlorides of ammonium, potassium, sodium and lithium; J. B. Johnston, The telencephalon of Sela- chians; Sergius Morgulis, The movements of the earthworm; Karl T. Waugh, The rdéle of vision in. the mental life of the mouse. By THE JOURNAL OF EXPERIMENTAL Z@OLOGY. The entire contents of Vol. § (four numbers, the Brooks Memorial, now being issued, will be iousd in the advertising pages. E. A. Andtewe. es, yee Oa | eet Contugation in- “the ‘oie ates Kini om Zv6logical Laboratory, ‘ohns Hicokins University = figures tees <2 ae Racers S. O. Mast — 1 i aoe eee Cea ‘Reactions in Amoeba *o bee From the Biological “Lab atory, Goucher Collexe. Two t higures . ree se H. S. Jennings - : fees eee What conditions ae 2 conjugation in ‘Parame on heures oo8 oS ae. a Se ee M. M. Metcalf | Studies upon Amoeba. From the , Zoblogical Laboratory Ober a 5 lin College. ~ Forty-fe figures.. 0.08.2... ea ol Samt.. R:::enheuse The embryology of Stomotoca apicata. From the Biological Laboratory, Olivet C° ‘lege. Thirty-two figures. . aie a A. M. Reese 2 = The '.teral line a ee of Chimera colliei. Brow the Zob- logical Laboratory, West V*ginia University. Tee Agures kins University. ‘Thirteen 4 leap Edwin G. Conklin Bers The effects of centrifugai force upon the organization and ieee a opment of the eggs vu: fresh water pulmonates. }'rom the 2 ge Biological Laboratorv, ES tao we University. Forty-seven figures, <.pto.5 4% Sc Dg eg ke ae Si. » ee Se ne _ 17 Henry F. Nachtrieb : % 3 . ‘The primitive pores of Polyodon spathula (WalBau) frou the Laboratory of Animal EPS S The Univer: itv 0’ " Minne- ae 7 sta. Twelve figures. . = Wit yr ee 45 will be issued this year. A volume cbuslate of four abe ( taining atout 150 pages each, with numerous illustrations. The is $5.00 per volume to subs«“‘bers in the United States, Suba, Me it ie no ond Panama, and $5.50 tr ~abcribers in other countries - a Sis espe Sas FOR 1910 eS ee e 8, January to Jul. : a eee 9, July 1% Decemle~ } suo. 0 PORE ee 1. 0 Sactat Re! by, ‘tl money order; mand=" 7e, ste’ Pitan visung, orty on New York. : ee 3 All Business Comnanications should be addressed t to 36rTa STREET AND WOODLAN D ) AVENT 1B = PHILADELPHIA, oe . re Ses = z Manuscripts may be sent to any member of an - Editorial Board, or directly to the Managing. Editor of the Journal for which they aresubmitted. Authors "receive fifty copies of their articles gratis. * Ae Subscriptions are payable in advance and no dis- --~—« eounts are ever allowed. Occasionally two volumes are published In one year, therefore, subscriptions are always taken by the volume and not by the year, Remittances should be made by postal money order, or by draft, payable to ‘THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY ee PHILADELPHIA, PA., U.S.A. JOURNAL OF MORPHOLOGY (JOUR. MORPH.) J.S. KINGSLEY, Manacina Epitor Turts CoLueae, Mass. - Publishes researches relating to animal morphology _ especially those on cytology, protozodlogy and the ; po omg and embryology of vertebrates and inverte- ra with numerous text figures a Sx inrt narg price $9.00 per volume, postpaid to THE SiaiiCan JOURNAL OF ANATOMY (AM. JOUR. ANAT.) H. McE. KNOWER, Sec’y Eprtortat Boarp t UNIVERSITY OF CINCINNATI * Publishes investigations on the anatomy, embryol- ogy and histology of vertebrates with special refer-. ence to mammals. Issued quarterly. About 550 pages per volume, with numerous text figures and plates Subscription price $5.00 (Domedio): $5.50 (Foreign) per volume. THE ANATOM ICAL RECORD (ANAT. REC.) G. CARL HUBER, Manaaetna Epitor 1330 Hixu Street, ANN ARBOR, Mica. Is intended for the prompt publication of concise one articles on rate anatomy, embryolo: y an with such illustrations as may printed in text, of brief preliminary re a technical notes, short reviews of noteworthy publi- cations, critical notes of interest to areas. brief aren of courses, of laboratory plans and of ae aietila: About pages per volume. Subscription price $3.00 (Do Domestic); $3.25 (Foreign) per volume. 3 : Atak bles JOURNAL OF COM PARATIVE NEUROLOGY AND PSYCHOLOGY (JOUR. COMP. NEUR. PSYCH.) C.J. HERRICK, Manaaine EpITorR UNIVERSITY OF ‘CuIcaao Publishes wg ti iano 8 on the compara- _ tive anatomy system and vessel hes ee ee ONE: 600 urease per volume with mumero ‘te tion lee wood (Domes $5.60 (Foreign) issued quarterly. About 700 pages in each volume, ates. GENERAL INFORMATION THE JOURNAL OF EXPERIMENTAL ZOOLOGY (JOUR. EXP. ZOOL.) ROSS G. HARRISON, Manaaina EpitTor © Yace University, New Haven, Conn. Publishes the results of original researches of an experimental or analytical nature in the field of zodl- . ogy, including investigations on growth and develop- ment, on reproduction, regeneration, cellular phenom- ena, evolution, variation, heredity and ecology, and upon topics in general physiology. Issued eight times a year, in two volumes. About 600 pages per volume, with numerous text figures and plates. Subscription price $5.00 (Domestic); $5.50 (For eign) per volume, ARTICLES ACCEPTED FOR PUBLICATION By the JourNaL or MorpHotoeyr: J. F. Guder- natch, The thyroid gland of the teleosts; H. H. Newman and J. Thomas Patterson, The develop- ment of the nine-banded armadillo from the primi- tive streak stage to birth, with especial reference to the question of specific polyembryony. Leland Griggs, Early stages in the development of the central nervous system of Amblystoma punctatum: R. J. Terry, The morphology of the pineal region in Teleosts. By THe ANnatTomicaL Recorp: Arthur W. Meyer. The question of applied anatomy; Franz Weiden- _ reich, Die Morphologie der Blutzellen und ihre Bezichungen zu einander; Geo. 8S. Huntington, The genetic principles of the development of the systemic lymphatic vessels in the mammalian embryo; Harvey E. Jordan, A further study of the human umbilical vesicle; Franklin P. Mall, A list of normal human embryos which have been cut into serial sections. By THE AMERICAN JOURNAL OF ANATOMY: William F. Rika, Distribution of the lymphatics in the tail region of scorpzenichthys marmoratus; Franklin P. Johnson; The development of the mucous membrane of the cesophacus. stomach and small intestine in the human embryo; Edwin G. Kirk, On the histo- genesis of gastric glands; Leonard W. Williams,The somites of the chick; Florence R. Sabin, Description of a model showing ‘the tracts of fibers medullated in a new-born baby’s brain. By THe JOURNAL OF CoMPARATIVE NEUROLOGY AND Pstcuotoey: L. W. Cole, Reactions of frogs to chlorides of ammonium, potassium, sodium and lithium; L. M. Day and, Madison Bentley, A note on learning in Paramecium; G. V. Hamilton, A ‘qualitative study of mammalian trial and error re- actions; C.J. Herrick, The morphology of the fore- brain in Amphibia and Reptilia; S. J. Holmes, The reactions of mosquitoes to light in different periods of their life history; J.B. Johnston, The telencepha- lon of Selachians; Sergius Morgulis, The movements of the earthworm; Karl T. Waugh, The role of vision in the mental life of the mouse. By Tue JOURNAL OF EXPERIMENTAL ZO6LOGY. The entire contents of Vol. 9, (four numbers, the Brooks Memorial, to be issued during 1010), will be found in the advertising pages. CONTENTS Biography William Keith Brooks: A sketch of his life by some of his former pupils and associates. Three portraits: O..%... Edmund B. Wilson Studies on chromosomes. From the Zodlogical Laboratory, Columbia University. Five figures...............0.-. 53 George Lefevre and W. C. Curtis Reproduction and parasitism in the Unionidae. From the Zodlogical Laboratory, University of Missouri. Thirty- nine figures, five plates,” ... ces cag-.i2 rim OSs re gue 19 Otto C. Glaser The nematocysts of Eolids. From the Zodél. Lab., Univer- sity of Michigan. Twelve figures..........:.......0 117 Albert H. Tuttle Mitosis ‘in Oedogonium. From the Dept. Biol., University of Virginia. Eighteen figures................eceeees 143 J. Playfair McMurrich The genus Arachnactis. From the Anat. Lab., University of ‘Toronto. Hive! igites 7 sc Bee, 3 ae 159 Francis H. Herrick Life and behavior of the cuckoo. From Adelbert College, Western Reserve U. Twenty-three figures, seven plates. 169 Two volumes of THE JOURNAL OF EXPERIMEN TAL ZOOLOGY will be issued this year. A volume consists of four numbers, con- taining about 150 pages each, with numerous illustrations. The price is $5.00 per volume to subscribers in the United States, Cuba, Mexico and Panama, and $5.50 to subcribers in other countries. SUBSCRIPTION FOR 1910 Volume 8, January to July Me : Volume 9, July to December } s10.00 (Foreign $11.00) Remit by postal money order; mandat de poste; Blase anes or by draft on Net ew York, All Business Communications should be addressed to . THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY | 367TH STREET AND WOODLAND AVENUE PHILADELPHIA, U. §. A. . THE WAVERLY PRESS BALTIMORE ' a i P " ‘ aut 1S : * : wen 222 A Sa ‘ih ie ite Sd oper Pa a 7 a ll a mies ‘ i] re \ i WILLIAM KEITH BROOKS ie 1848-1908 : ‘THE Sona. OF - ae a. 1 eat ; - EDITORIAL BOARD ¢ > < ; = | Wruzras E. Cidtaat whee : «|. Franx R:; Litu - Harvard University _ e . University of Chicago ~ = G. ConkunN <6 8g Jacques Lorn . aS Princeton University oie The Rockefeller Institute ae ae Ey - ; . Cartes B. Davenport ees _ Tuomas H. Morean : > as Carnegie Institution as OGD De - cd 4 Columbia University me Te ¢ _ Horace JAYNE ~ Peas os - . Grorce H. Parker ; | Fa eee The Wistar Institute - Harvard University co EAI Herserr 8. JENNINGS — p - Caries O. WHITMAN =. we > “Johns: Hopkins University VE University of Chicago Pe Ne => | Epmunp B. Wiison, Columbia University Bete “ a a ‘ ‘7 aegeee F 3 and Sa : Sh oe “ak Ross G. HARRISON , Yale University - 38 4g ; acs pe 2 : Managing Editor “+ . P) = ote a : “> oe . : 4 3 : : : . OF La RSet VOL. 9 No. 1 ne ee es 5 te . SN . ; < c sone SEPTEMBER, 1910 ae 3 a : ; 1 t oe A ‘ ie ) pie Ber be a me PUBLISHED EIGHT TIMES A YEAR BY THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY oe 86TH STREET AND WOODLAND AVENUE eee ee ee PHILADELPHIA, PA. . ~~ * ~ . - a, &., . : ‘caf ee natn, ¢ > %E}9 iS ape =? <2 ea” a ot. > rar tie — , + . 4 - eo _ rt, “ed ae! d > ~*~ aw ; 3 i S , FAN =f ». bE §K ; r ‘ ake NOTICE TO CONTRIBUTORS THE JOURNAL OF EXPERIMENTAL ZOOLOGY, appearing poet times a year, will pablgtt 2 papers which embody the results of original research of an experimental or analytical — nature in the field of zodlogy, including investigations on growth and development, both normal and pathological, on reproduction, regeneration, cellular phenomena, — evolution, variation, heredity and ecology, and upon topics in general physiology. — Contributions must not be of unreasonable length in general, it is not desired to publish papers that will occupy more than fifty printed pages. Purely preliminary notices will not be published. No paper that has already appeared elsewhere will be aseetied for pubjleatie Simultaneous publication in another journal will not be agreed to. Manuscripts should be typewritten on one side of paper 8} x 11 inches and should: be packed flat, not rolled or folded. Footnotes and other subsidiary matter should be typewritten on one or more separate 3 ~ sheets. Footnotes should be numbered consecutively throughout the article. The — 3 location of other subsidiary matter should be indicated by notes, letters or other signs. ; Figures should be drawn for reproduction as line or half-tone cuts, so that they may - be printed in the text, either singly orin groups, unless the author is prepared todefray the — additional cost of a more expensive form of illustration. Half-tones are frequently - printed separately to obtain the required details and inserted as plates. All colored plates, lithographs, heliotypes, photogravures, etc., are, of course, inserted, and cost extra. In grouping the drawings it should be borne in mind that, after the reduction has been made, the figures are not to exceed the dimensions of the printed matter on the page, 43 x 64 inches. Single plates, inserted, may be 5.x 74 inches, or less, and _ double plates (folded in the middle) 113 x 73 inches. The lettering of the figures should be made with a view to a neat appearance when printed. Printed letters and numbers will be furnished without cost to authors for their — illustrations. A list of letters and numbers should be given and the reduction to whiehy they are to be subjected should be stated. * ¢ } Galley proofs will be sent to the author and all corrections are to be clearly marked _ thereon. Page proofs will be revised by the editors. Manuscripts and drawings should in every case be submitted in complete and finished form. The Wistar Institute reserves the privilege of returning to the author for revision approved manuscript and illustrations which are not in proper finished form for the printer; or. if the author so desires, the Wistar Institute will prepare the menuscripey ¥ and drawings and charge the author for the cost of this work. A Styte Brier giving all the details of typographic arrangement and methods to ba followed in the prep@ration of manuscripts and drawings for publication in anyof the five journals published by the Wistar Institute, will be sent free to authors upon opens = cation to the Wistar Institute in Philadelphia. The JourNat willfurnish to authors 50 reprints of their contributions gratis. Addi- ~ tional copies may be obtained according to rates accompanying galley proof. The order for reprints should be sent with the galley proof, for which purpose blank forms are supplied. Manuscripts and drawings should be sent by express or registered mail to the Manag- y ing Editor, Ross G. Harrison, 2 Hillhouse Avenue, New Haven, Conn. Ss Fae detiecien —- wittiaM KEITH BROOKS iz a Bains vt - 1848-1908 a arc aA _ THE JOURNAL ag “3 ” a Rs . : a , i = J O F Pies Gh a = “ ts; a a ° 2 PAS “—G ° * ™~ St asec ote., favtdi cc cae | . 4Z Wruuras RUNS? Tale Sra ees \ Frank R. Linus 4 ok Harvard University f University of Chicago WIN G.CONKLUIN | 4 ae Jacques Lors Ho "Princeton University » = _~ eee ren PE, 2 Gaonay H. Panken Spee The Wistar I Tnstitu te e ; Harvard University > z " rs, Je NINGS panty oS CHARLES O. WHITMAN Johns Hopkins University Bi Tie University of Chicago ss a és ' : _ Epmonp B. WIson, Columbia University ey a Pay oS ane y ; “oss = ae Ross G. Scan: Yale. University Sg a % x Beata! 1S - Managing Editor oy ae . Bato he = ? ae ot i > eee % mice fas ae ee Raa aS I aad Th aS pe - Bets “4 to or ss = VOL. - 9 No. 2 = tt = “> nF, 2 eat > oi a eee OCTOBER, 1910 nae Se agent ae pa ee Pe. . *¥. a et 4 wig sae wee Sd Ss +*¢ ig y ee eee “PUBLISHED EIGHT TIMES A YEAR BY Gee o> 0 WISTAR INSTITUTE OF ANATOMY AND BIOLOGY ‘A es ‘a a 3 is — 86rH STREET AND WOODLAND AVENUE “reas i. Yi S EEOPELEIA, PA. os ‘ s = | = ry a : } “NOTICE TO CONTRIBUTORS gaa Tue JOURNAL OF EXPERIMENTAL ZOOLOGY, appearing eight fimes? a year, will ‘oubliehes : pfpers which embody the results of original research of an experimental or analytical — nature in the field of zodlogy, including investigations on growth and development, — oe both normal and pathological, on reproduction, regeneration, cellular phenomena, — evolution, variation, heredity and ecology, and upon topics in general physiology. Contributions must not be of unreasonable length in general, it is not desired to 6 publish papers that will occupy more than fifty printed pages. Purely preliminary notices will not be published. ae ~ No paper that has already appeared elsewhere will be accepted er publication- Simultaneous publication in another journal will not be agreed to. fe wae Manuscripts Should be typewritten on one side of paper 83 x il inches and shoal be packed flat, not rolled or folded. Footnotes and other subsidiary matter should be typewniare on one or more separate sheets. Footnotes should be numbered consecutively throughout the article. The location of other subsidiary matter should be indicated by notes, “letters or other si signs. Figures should be drawn for reproduction as line or half-tone cuts, so that they may be printed in the text, either singly orin groups, unless the authori is prepared: to defray the additional cost of a more expensive form of illustration. Half-tones are frequently printed separately to obtain the required details and inserted as plates.. All colored — plates, lithographs, heliotypes, photogravures, etc., are, of course, inserted, and cost — extra. In grouping the drawings it should be borne in mind that, after the reduction — has been made, the figures are not to exceed the dimensions of the printed matteron ~ the page, 4! x 6} inches. Single plates, inserted, may be’5 x 7} inches, or less, and— double plates (folded in the middle) 11} x 73 inches. eras te SS The lettering of the figures should be made with a view to a neat appearance wher: printed. Printed letters and numbers will be furnished without cost to authors for their ’ illustrations. A list of letters and numbers should be given and the reduction Af. whic - they are to be subjected should be stated. > z Galley proofs will be sent to the author and all corrections are to be diana marked — thereon. Page proofs will be revised by the editors. Manuscripts and. drawings should in every case be submitted in complete and finished form. » KS : Re _—" ete The Wistar Institute reserves the privilege of returning to the author for revision approved manuscript and illustrations which are not in proper finished form for the | printer; or if the author so desires, the Wistar Institute will prepare ane manuscript and drawings and charge the author for the cost of this work. ~~ ak eae A Sryue Brier giving all the details of typographic arrangement and methods pba = followed in the preparation of manuscripts and drawings for publication in any of t] five journals published by the Wistar Institute, will be sent free to authors upap appli- cation to the Wistar Institute in Philadelphia, _ eS y es The JourNAL willfurnishto authors 50 reprints of their contributions gratis. Addi- tional copies may be obtained according to rates accompanying galley proof.- The order for reprints should be sent with the galley proof, for whieh asi Setols blank forms oe supplied. Bees ee Manuscripts and drawings should be sent by express or eRe, mail to the Manag: | ing Editor, Ross G. Harrison, 2 Hillhouse Avenue, New Haven, Conn. G nae se Re eb eat ta i OE #25 gee rebate wats i: ‘ Ca. ; . : ; 5 . ve el = sola Res gu ‘Pemoriam wna KEITH BROOKS = are sat 2 > “Franx R. Linum | aaa? hehe ; < University of Chicago _ - z Pt eo So tS “Tacaues Lors $e SN. Se yy . Rist Rockefeller Institute aparece es -. Tuomas H. Morcan tin weal Columbia University to ae _ GrorGE H. ParKER © =e i ae a ps Harvard University x Se Cuartes O. WHITMAN a University of Chicago xy B. Witson, ‘Columbia University NON, “s —_i * . Peed a _ and» ers © EM: °; , - eet : bs *. Es Recs ere 5 e666, Hannon Yale University _ Managing Editor i? * 2 NOTICE TO CONTRIBUTORS Rete 4 ; a =< >s “ THE JOURNAL OF EXPERIMENTAL ZOOLOGY, appearing eight he a year, will publish papers which embody the results of original research of an experimental or analytic cal nature in the field of zodlogy, including investigations on growth and development, both normal and pathological, on reproduction, regeneration, cellular phenomena, evolution, variation, heredity’ and ecology, and upon topics in general physiology. Contributions must not be of unreasonable length in general, it is not desired to. publish papers that will occupy more than fifty printed pages. Purely prolingndiry notices will not be published. No paper that has already appeared elsewhere will be jecnten for: publication, - Simultanéous publication in another journal will not be agreed to. = Saas Manuscripts should be typewritten on one side of paper 8} x 11 inches and should iE be packed flat, not rolled or folded. - dS he pas Ray Footnotes and other subsidiary matter should be typewritten on one Shanere regarate sheets. Footnotes should be numbered consecutively throughout the article. The | location of other subsidiary matter should be indicated by notes, letters or other signs, so Figures should be drawn for reproduction as line or half-tone cuts, so that they may be printed in the text, either singly or in groups, unless the authoris prepared to defray the — additional cost of a more expensive form of illustration. Half-tones are frequently printed separately to obtain the required details and inserted as plates. All colored — plates, lithographs, heliotypes, photogravures, etc., are, of course, inserted, and cost extra. In grouping the drawings it should be borne in mind that, after the reducti yn h as been made, the figures are not to exceed the dimensions of the printed matter on the page, 4} x 63 inches. . Single plates, inserted, may be 5x73 inches, or less, ting double plates (folded in the middle) 113 x 7} inches. ear ~ The lettering of the figures should be made with a view to a neat appearance when printed. Printed letters and numbers will be furnished without cost to authors for their — illustrations. A list of letters and numbers should be given and the rednebien Bs which they are to be subjected should be stated. eases Galley proofs will be sent to the author and all corrections are to he eats inane thereon. Page proofs will be revised by the editors. Manuscripts and ‘ieee should in every case be submitted in complete and finished form. : Ss: Ie The Wistar Institute reserves the privilege of returning to the author for petit es » approved manuscript and illustrations which are not in proper finished form for the > ; printer; or. if the author so desires, the Wistar Institute will prepare the spenuger es and drawings and charge the author for the cost of this work. Fic A Sty.e Brier giving all the details of typographic arrangement and fiettiods to Foss followed in the preparation of manuscripts and drawings for publication in any of. es five journals published by the Wistar Institute; will be sent free to authors upon appli- ee 4 cation to the Wistar Institute in Philadelphia. ; ete The JourNAL willfurnish to authors 50 reprints of their contributions gratis. Addi- tional copies may be obtained according to rates accompanying galley proof. The order oe for reprints should be sent with the galley proof, for which purpose blank fee are supplied. vas = Manuscripts and drawings should be sent by express or Fecisieted mail to the Manag- “its ing Editor, Ross G. Harrison, 2 Hillhouse Avenue, New Haven, Conn. ee + gee Ret Bent = Cin #emoriam WILLIAM KEITH BROOKS — 1848-1908 = a a . poet | a t ae Re EDITORIAL BOARD rk < - oS a a f . ot Witiiam E. Castie : Frank R. Linuir ery ae: anf Harvard University University of Chicago te ee “wy G. ConxLIN | or Jacques Lors ete “ -_- Princeton University — “6,2 The Rockefeller Institute TaN a -Cuantes B. Davenport __ Tomas H. Morcan ee 3 _ Carnegie Institution - . Columbia University eS, _ Horace JAYNE ; Grorce H. Parker Babs aie The Wistar Institute Harvard University we Herperr S.Jennines- Cuartes O, WHITMAN -P.> eS Bi ohns Hopkins University io cl of Chicago i" AOS 3 ee ee EpMUND. B. Witson, Columbia University and + * = . ’ Ross G. HARRISON, Yale University Managing Editor Ae in 2y 5 Ras is N 7 ‘ e's > 4 Be VOL. 9 No. : DECEMBER, 1910 Yh | a ait Sade: __. PUBLISHED EIGHT TIMES A YEAR BY “THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY es oe 36ra STREET AND WOODLAND AVENUE ; PHILADELPHIA, PA. —¥ og = , _~ “= “i Te es 7 i “ 2 r ms ~ . ~ ~; = 4 ae cir, 34 fait) ~~ > ae ae 4 c sa - *, o:3 ee Ag . ~ = Sinacl ; “ M > P Theme t a Ss : - ith He f ae 2 * “ A - nF ih Sy BO ES ; ey, a, ee teat “asses $ } Sor Se See a A ee } Se — Se a She ee eS <" y ay re S aa c aa < ie Po ee - —” ae < CT > 4 n pee ee: a7 —- - a 2 my 7. NOTICE TO CONTRIBUTORS so g & Tus J OURNAL OF EXPERIMENTAL Zob.ocy, appearing eight times a a A year, nature in the field of zodlogy, comets investigations on oa and. deve Loy both normal and pathological, on pate age poreera cellular ee notices will not be published. ; St2.3 TE No paper that has already appeared sawhebe ae ee chen for Fr Shh iaaners publication in mptpe si japruel es not be seed a Fos: pss ae be packed flat, not rolled or folded. . = Footnotes and other subsidiary matter should be Geaerices on one or ee sheets. Footnotes should be numbered consecutively throughout the article. : y see: of other srbgtaiary matter phoma be pasion ied by notes, letters or rT OLlie additional cost of a more atone form of illustration. Half-tones are ‘fre printed separately to obtain the required details and inserted as plates. All co’ 6 plates, lithographs, heliotypes, photogravures, etc., are, of course, inserted, , and. s extra. In grouping the drawings it should be*borne in mind that, after the xedae has been made, the figures are not to exceed the dimensions of the printed matter the, page, 41 x 64 inches. Single plates, inserted, may be 5 x 73 inches, or less, ¢ double plates (folded in the middle) 11} x 73 inches. ae > cae "rie The lettering of the figures should be made with a view to a ‘ead appearane printed. Printed letters and numbers will be furnished without cost to authors for tl illustrations. A list of letters and numbers should_ be given and. the reduction to wh they are to be subjected should be stated. SN ee Galley proofs will be sent to the author and all corrections are to be clang yu thereon. Page proofs will be revised by the editors. Manuscripts and dr should in every case be submitted in complete and finished formic >: The Wistar Institute reserves the privilege of returning to the author for ited fos manuscript and illustrations which are not in proper finished form { and ea and charge the iarhior for the cost of this work, A Sryue Brier giving all the details of typographic arreseoneaimeet meth od followed in the preparation of manuscripts and drawings for publication i in an Ly 0 five journals published by the Wistar Institute, will be Sent free to authors upon cation to the Wistar Institute in Philadelphia. é 5 The Journat will furnish to authors 50 reprints of their! soninbnsane gratis tional copies may be obtained according to rates : accompanying galley proof. e for reprints should be sent with the ke sia! for which puree: blank ‘forms supplied. Se ee he, {gt es Manuscripts and drawings should be sent express ¢ or Seale mail to » the I ing Editor, Epes G. Harrison, 2 Hillbouse: i ice ex Haven, Conn. 3 Es Ss a r - rt - 4 o . + a 4 & G fs ” i ‘sy f % - 3 , Nel 4 val a , rr » be) * ‘ \ i% ax. > 2 a . > f , * - a ou ow é en i Ai ‘ipts may be sent to any member of an c ‘dltoria f Board, or directly to the Managing Editor __ ofthe Journal for which they aresubmitted. Authors : _ feceive : fifty copies of their articles gratis. _—_ Se 25 fhe ~ A ~, wa are ever allowed. Subscriptionsare always peter by ve Es ee ne by the year Re- _ mittances | made tal : “or by tral, an iy y postal money order “ os — - THE WISTAR INSTITUTE OF * ANATOMY AND BIOLOGY _ PHILADELPHIA, PA., U, 8. A. Py, - 4 . a 4 oa a re . : os . — THE JOURNAL OF — MENTAL ZOOLOGY (JOUR. EXP. ZO6L.) HARRISON, Manaartne EpiTorR *y se é ‘ “™< reese fee $5.00 (domestic); $5.50 (for- per volume. : 4 LS THE AMERICAN JOURNAL OF ope ANATOMY °°. ~2ag nea ~ (AM. JOUR. ANAT.) __H.McE KNOWER, Suc’y Eprrortat Boarp > _ -— UNIVeRsity of CINCINNATI Publishes investigations on the anatomy, embryol- - ogy and histology of vertebrates with special refer- - -erneetomammals, — — * Issued bi-monthly. About 609 pages per volume, with numerous tert figures and places. = Subscription price $5.00 (domestic); $5.50 (foreign) >» pervolume. | . eae er, - if cig a, : p - _ THE ANATOMICAL RECORD oe “et Ses. _ (ANAT. REC.) __ G. CARL HUBER, Manaarne Eprtor 1830 Niue Sareer, Ann Arsor, Micu. _ Isintended for the prompt publication of conctse F saigina) articles on vertebrate anatomy, embryology A an histology with such illustrations as* may printed in the text, of brief preliminary reports and _ technical notes, short reviews of noteworthy publi- cations, eritical notes of interest to anatomists, brief __ statements’ of courses, of laboratory plans and of <— Sa ere ar inert : : _ -Lssued monthly. > * ow, About 400 pages per volume. + or volume. poe F i 4 OURNAL OF COMPARATIVE =>, NEUROLOGY AND Petes.” PSYCHOLOGY= : pepe _ (s0UR. COMP. NEUR. PSYCH.) ©, J, HERRICK, Manaaine Eprror - __—sUUiversity or Cutcaco original contributions on the compara- of the nervous system and researches ‘ing to animal behavior, . About 600 pages per volume with rts, text figures and plates. . n price $5.00 ipomestads yi PRS 60 ec: -—," * bseriptions are payable in advance and no dis- _ Subscription price $3.00 (domestic) ; $3.25 (foreign). -by The Wistar ustitut GENERAL INFORMATION JOURNAL OF MORPHOLOGY ; (JOUR. MORPH.) J.S. KINGSLEY, Manaaine Epitor Turts Co.Lece, Mass. Publishes researches relating to animal morphology, especially those.on cytology, protozodlogy and the anatomy and embryology of vertebrates and inverte- brates. ‘ssued quarterly. About 700 pages in each volume, with numerous tert figures and plates. Subscription price $9.00 per volume, postpaid to all countries. ARTICLES ACCEPTED-FOR PUBLICATION By the JouRNAL or MorpHo.oey: H. L. Wieman, The degenerated cells in the testis of Leptinotarsa signaticollis; Gideon S. Dodds, Segregation of the germ cells of the Teleost lophius; H. S. Jennings and G. T. Hargitt, Characteristics of the diverse ~2 races of Paramecium; J. Parsons Schaeffer, The lateral wall of the cavum nasi in man, with especial reference to the various developmental stages; F.J. Gudernatsch, The thyroid glands of the Teleosts. Volume 22 (four numbers, to be issued during 1911) is to be known as The Whitman Volume, and will contain exclusively papers by former pupils of Dr. C. O. Whitman, the founder of this Journal. A complete list of authors and titles of their papers will shortly be issued." The first number will be ready March 15th. By THe ANATOMICAL ReEcorD: W. M. Baldwin,. Duodenal! diverticula in man; H. Jeidell, A note on the source and character of the early blood vessels of the kidney; T. D. Merrigan, An unusually large per- itoneal fossa; A. T. Kerr, Complete double ureter in man; F.W. Thyng, The anatomy of a7.8mm. pig embryo; B. F. Kingsbury, Theterm ‘‘chromaf- fine system’’ and the nature of ckromaffine reac- tion; William Snow Miller, The distribution of lymphoid tissue in the lung; Samuel T. Orton, Note on an anomaly of the postcentral sulcus simulating the double rolandic of Giacomini. By THe American JourNAL or ANATOMY: J. Gordon Wilsom, The nerves and nerve endings in the membrana tympani of man; W. E, Dandy and Emil Gactsch, The blood supply of the pituitary body; Florence R. Sabin, Description of a model showing the tracts of fibers medullated in a few- born baby’s brain; Charles R. Bardeen, Further studies on the variation in susceptibility of Am- *phibian ova to the X rays at different stages of development; Jeremiah S. Ferguson, The anatomy of the thyroid gland of Elasmobranchs with remarks upon the hypobranchial circulation in these fishes; J. F. Gudernatsch, Hermaphroditismus verus in man; John Warren, The development of the para- physis and pineal region in Reptilia. By THE JouRNAL OF COMPARATIVE NEUROLOGY AND Psycuotocy: J.B. Johnston, The telenceph- alon of Selachians: Shephard Ivory Franz, On the function of the post-central cerebral convolutions; H. H. Donaldson, On the influence of exercise on the weight of the central nervous system of the al- binorat: H.H. Donaldson, The effect of under- feeding on the percentage of water, on the ether- aleohol extract, and on medullation in the central nervous system of the albino-rat; H. H. Donald- son and Shinkishi Hatai, Note on the influence of castration on the weight of the brain and spinal cord in the albino rat, and on the percentage of water in them; H.H. Donaldson, An interpreta- tion of some differences in the percentage of water found in the central nervous system of the albino rat and due to conditions other than age; Helen Dean King, The effeets of pneumonia and of post- mortem changes on the percentage of water in the brain of the albino rat. By Tue JouRNaL OF EXPERIMENTAL ZooLoeyY: Montrose T. Burrows, The growth of the tissues of the chick embryo outside the animal body, with special reference to the nervous system; Sergius -Morgulis, Contributions to the physiology of regen- eration; G. H. Parker, The olfactory reactions of the common killifish, Fundulus heteroclitus (Linn.) ; Ralph E. Sheldon, The sense of smell in Selachians; Marian L. Shorey, A study of the differentiation of neuroblasts in artificial. culture media; G. 5. Spooner, Embryological studies with the centrifuge. > : “ ‘ % hope <5 € ay 5 pie Ay a oe ae ee - as Sie) { a am eee. we ma 13 ; “Sere | CONTENTS peel | are ss f 7 David H. Tennent. = ‘ | ‘Variation in Echinoid Seated From a «Biological Laborator i 7) SE 2 BE 4 @ sat ‘8 ia") og # Ae o Ris a Duncan S. Johnson as Re Re aa Studies in the development of the Peas From the Biot. ; ~ Jogical Laboratory, J ohns Hopians: University. : — one figures..... gin hg oS ns ere pn nee pat one I, OL Robert P. Bigelow z See ae A comparison of the sense organs in Medusee = the family Pelagide. From the Biological Laboratory, Massachusetts Institute of Technology. ee figures: ees +: : La Bh ee a ed 4 COT ap i BE EET ay Ross Granville Harrison The outgrowth of the nerve fibre as a mode é protien University. Three plates and (hinge noe Pee ER ee oe o% George C. Price The structure and fanptiod of the adult “head “Baueee = seu: Bdellostoma stouti. From the Zodlogical Labornberys es ogy ; Leland Stanford Jr. University. ee RGU cas ST ce J. Frank Daniel Observations on the period of pestation:i in white mice. From the Zodlogical Laboratory, University of Michigan. ge ie ; Two volumes of THE J OURNAL OF EXPERIMEN TAL ZOOLOGY Ace will be issued this year. A volume consists of four numbers, con- — oe taining about 150 pages each, with numerous illustrations. The price Se me is $5.00 per volume to subscribers in the United States, Cuba, Mexico ‘ nl and Panama, and $5.50 to subcribers 1 in other countries. ee es SUBSCRIPTION FOR 1910 Volume 8, January to July Volume 9, July to December sae | Koraee Se on New York. All Business Commanicatsone should be addressed to So a THE WISTAR INSTITUTE OF ANATOMY AND BIOI oe xy ‘ 36TH STREET AND WOODLAND AVENUE. yak. PHILADELPHIA, U. 8. A. BS THE WAVERLY PRESS caw BALTIMORE _— PS: 2 pts may be sent to any member of an Board, or directly to the Managing Editor irnal for which they are submitted. Authors ifty copies of their articles gratis. riptions are payable in advance and no dis- are ever allowed. Subscriptionsare always yy tne volume and not by the year. Re- 2s should be made by as money order, Eepeyaite to : TI LE WISTAR INSTITUTE OF a. ‘ed he AND BIOLOGY * Jot URNAL OF MORPHOLOGY . _ (souR. MORPH.) rif e? 18. KINGSLEY, Manacina Eprror Wa’ ae - Turts Cotiecs, Mass. Bebiicercie relating toanimal morphology, clally those on cytology, protozodlogy and the my and embryology of Verveprates and inverte- et gui, About 700 pages in each volume, rous text Ligh and plates. AMERICAN JOURNAL OF oe a ANATOMY : (AM. JOUR. ANAT.) McE. KNOWER, Src’y Eprrortan Boarp : . a3 _ University or CrncINNATT hes Investigations on the anatomy, embryol- tology of vertebrates with special refer- oe et plates. . 2 ANATOM ICAL RECO RD ~ j iBbe “ (ANAT. REC.) G.CARL HUBER, Mawaaina Maric ° Phe Huy Street, ANN Arson, Mica. ended for the prompt publication of concise articles on vertebrate anatomy, Spe vology stology with such illustrations as may d in t. > text, of brief preliminary reports and © al notes, short reviews of noteworthy publi- s, critical notes of interest to anatomists, brief rents ae courses, of laboratory plans and of ntn nen , ily. About 400° pages per volume. ad oa Sieictien price $3. 00 (domestic); $3.25 Figretign) FP er. ime. 4 . a : SE re + i eae THE - URNAL OF COMPARATIVE _ NEUROLOGY AND . . PSYCHOLOGY 35 (JOUR. COMP. NEUR. PSYCH.) Pe. J . HERRICK, Manaaine Epiror ‘University oF Caicaco ’ ¢ Sctnal contributions on the compara- ioaiy of the nervous system and researches ects rears to animal behavior. bi-monthly. About 600 pages per volume with us charts, tent figures and plates rip Pp Jon price $5. 00 (domestic); ‘$5. 50 (foreign) “ ty ~§ . “" Le i mony About 600 pages per ea 1h number of this Journal. ied by The Wistar lnstiate oe GENERAL INFORMATION THE JOURNAL OF EXPERIMENTAL ZOOLOGY : (JOUR. EXP. ZOOL.) ROSS G. HARRISON, Manaatna Epitor Yave University, Naw Haven, Conn, Publishes the results of original researches of an - experimental or analytical nature in the field of zoél- ogy, including investigations on growth and develop- ment,on reproduction, regeneration, cellular phenom- ena, evolution, variation, heredity and ecology, and upon topics in general physiology. Issued eight times a year, in two volumes. About 600 pages per volume, with numerous text figures and plates. Subscription price $5.00 (domestic); $5.50 (for- elgn) per volume. : ArticLes ACCEPTED FOR PUBLICATION By the Journat or Morpso.oay: H. L-. Wieman, The degenerated cells in the testis of Leptinotarsa signaticollis; Gideon S. Dodds, Segregation of the germ oan of. the Teleost lophius; H. S. Jennings and G. T. Hargitt, Characteristics of the diverse Traces of Paramecium; J. Parsons Schaeffer, The lateral wall of the cavum nasi in man, with especial reference to the various developmental stages. By THe ANATOMICAL Recorp: W. M . Baldwin, ‘Duodenal diverticula in man; H. Jeaell, A note on the source and character of the early blood vessels of the kidney; T. D. Merrigan, An unusually large per- itoneal foss1; A. T. Kerr, Complete double ureter in man; R. 0. Moody, Some features of the histo- genesis of the thyroid gland in the pig; F. W. Thyng, The anatomy of a 7.8 mm. pig embryo; Alfred Jerome Brown, A note on post-cardinal omphalo- mesenteric communication in the adult mammal; Edwin E. Reinke, Note on the presence of the fifth aortic arch in a6 mm. pig embryo. By -THr AMBRICAN JOURNAL or ANATOMY: J: Gorden Wilson, The nerves and nerve endings in the membrana tympani of man; W. BE. Dandy and Ernil Goetsech, The blood supply of the pituitaiy body; Florence R. Sabin, Description of a model showing the tracts of fibers medullated in a new- born baby’ s brain. By Tue JouRNAL OF COMPARATIVE NEUROLOGY AND PsycHouoey: L. W. Cole, Reactions of frogs to chlorides of ammonium, potassium, sodium and lithium; J. B. Johnston, The telencephalon of Selachians; Sergius Morgulis, The movements of the earthworm; Karl T. Waugh, The réle of vision in the mental life of the mouse. By THe JouRNAL OF EXPERIMENTAL Zo6Loay: Three numbers of The Brooks Memorial, Volume 9, have been issued. No.4 will be issued in Decem- ber. The entire contents of the volume was pub- lished in the advertising pages of the Ociober Hubert Lyman Clark wie eee es = ae +e ee ‘Henry L. Osborn = Ware eee ok 2 i H. V. Wilson y as Lee i mit crab. a ee the aroclabea Sa “pes ae versity, Tokyo, J apa ‘Twenty-three REE S. & eae rae ek The development. of an apnioee helotharian: ‘(Chiridota ates ae fera). From the Museum ee Comparative Bosley are artes: University. ‘Six figures. . ae os tes ¥ On the structure of Crehiaree aa “tnigw ean Chyli oy sp), an aberrant distome, from Hishes of Michigan ae New Re cho a figures 2 6 webs ey exe 9 WS wig n ee slate eo wy ogre vt eh 6. wre 8) eel ©) IND 6. se ie Ocean Oe A study of some epithelioid sehihrane4 in monaxonid sponges. vs From the Zodlogical Paperatnry, Halve is North Ceneee = lina. Twenty-one figures. . NE ee oS * Vetere tee nae Charles Wilson Greene : ; : An experimental deterthination: of the speed of migrate of salmon in the Columbia River. From the Physiologie Laboratory, University of Missouri. Two figures. . . fovea T..H. Morgan i . . Cytological tution of centrifuged eggs. Fits the: Zoslaas | “% ical Laboratory, Columbia Paes One bundred a pie eae ‘eighteen. figures! Sete. oye Shoat oS te hee bey will be issued this year. “A volinte eee of ee ance taining about 150 pages each, with numerous. illustrations. ae is $5.00 per volume to subscribers in the United States, Cuba, M and Panama, and $5.50 to subcribers i in other connurige: SUBSCRIPTION FOR 1910 ; s e get Totunns: 8, January to J uly Volume 9, July to December } si0. $10. jes Ae oreign $11 00) Remit by postal money order; mandat de poste; Postanneisung, or by on New York, All Business (nidantoaiea. aiguld a addressed to ; = THE WISTAR INSTITUTE OF ANATOMY AND ae ox THE WAVERLY PRESS. : ae! Fie * ~ _ BALTIMO et Pr ee ae oe Soca BA eae Clas £190 P/ELO TINY 97 s( OTs 6°: er eS Te _ . ax * e org Es Ph Ps rar é ‘ 3 wa -¢ 4 a : és 5 * ‘ ‘ su t #74 ’ ' , ; te ¢ @ a8 § § vos ' , : : © ae 4 ‘ Hate 44 , : . ‘ : Fort Ae ; , N e ks Pe Te er ; ‘ em ¢ fee oe ‘ ‘ ‘ ‘ ‘ ied cfs ‘ : wed Ws 5 / , ia ¢ ‘ : ‘ . : . err ACK t z < F eee ~ ‘ +i , ‘ . . ; ; . F 3 %'s . 1 et ee re r ‘ ‘ * 4 Po ? : ; a Oat eke Oe + ‘ ey cathy : ie , ; és : aa rer - pee 4 , ' BS Ic i "y Oa F ‘ ‘ ‘ é * < 4 4 ye 8 Ey i r ey a ena #3 4 ' ‘ ‘ ’ ‘ P ’ 5 i : C44 wd ’ § J ‘ Pore ‘ e J ; : 7 : : . , wee is “ ; 2 t or rr waded atjey) velegs® re 4 ¢ « : ‘ £ f 2 J . . bef ¢ he : ; sae! gadtae r3e 4 3 r t a aels any = 4 : . . . ~ ‘ Fr ‘ a ay ‘ ’ ; ‘ ‘ i ‘ , % veh . eiel ‘ ? :