$ —- he es FD of cr, i 4 py cat er A yee rae ae Year bee i #7449 4. ae a x 3 *) an ~ 4 ¢ + PA ore? a sie Hee Behe Be i pes - 73 pase aes oe ae Pays Pe ae - phy not pha hath +. r a + Lethe 7 Sy at, +. a oe hs & ee ? ey ote. ab * * reels + ~ THE JOURNAL OF EXPERIMENTAL ZOOLOGY EDITED BY Wiuuiam E. Caste Jacques LorsB Harvard University The Rockefeller Institute EpmuNnp B. WILSON Columbia University EpwIn G. CoNKLIN Princeton University THomas H. Morcan CuHarLes B. Davenport Columbia University Carnegie Institution : : GEORGE H. PARKER HERBERT S. JENNINGS Pee at ecaiee Johns Hopkins University RaYMOND PEARL FRANK R. LILLIE Maine Agricultural University of Chicago Experiment Station and Ross G. HARRISON, Yale University Managing Editor VOLUME 22 EON 7 THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY PHILADELPHIA, PA. /1150 COMPOSED AND PRINTED AT THE WAVERLY PRESS By THE Wittrams & WILKINS CoMPANY Batrimore, Mp., U.S. A. CONTENTS NO. 1. JANUARY Marcarer Morris. A cytological study of artificial parthenogenesis in Cuminerasss Hourtexteheunres;and elehteplatesseers ser nee ee A. A. SCHAEFFER. On the reactions of ameba to isolated and compound pro- UIE SSIES EOS ooo gun toe 5 oa meee ce I o. E So. cone am tine Sa ictmer ote See G. H. Parker. Nervous transmission in the actinians. Three figures...... G. H. Parker. The movements of the tentacles in actinians. One figure. . G. H. Parker. Pedal locomotion in actinians. One figure................ Raymonp Praru. The experimental modification of germ cells. I. Gen- eral plan of experiments with ethyl alcohol and certain related sub- Stan Cesta hreesiounes a mse eens 5 ks. Ga ORI, ec Le pie eee RaymMonp Praru. The experimental modification of germ cells. II. The effect upon the domestic fowl of the daily inhalation of ethyl alcohol andacertain related substances: Four figures#eagse--0-2-0-6.- one Jacqurs Lorsp AnD HARpOLPH WASTENEYS. A reexamination of the appli- eability of the Bunsen-Roscoe law to the phenomena of animal helio- (THROW ONISHI ier e gS Meee cae eee aoto Sin Seon ott Ra ENeRS cc bo ob. oe. O RISES enol oes mama OB NO. 2. FEBRUARY lamelllibranchstaHouneiounes sacs: : scr eee cate ae eee RayMonpD PEARL. The experimental modification of germ cells. III. The effect of parental alcoholism, and certain other drug intoxications, upon thenpLo Geman e SCVCN tOUnES:.esacicus oo - ae eee cunt Rin eames ate sees 6 Epwin G. Conkuin. Effects of centrifugal force on the structure and de- velopment of the eggs of Crepidula. One hundred and twenty-four fig- Gro. T. Hareirr anp WauTeR W. Fray. The growth of Paramaecium in DUHEVCULGURESHOlmDACbelvaenws techs .clss 5 « cava ROI Ae cae hc kece erste ceere ake Watrer W. MarsHaLL AND HerMANN J. Muir. The effect of long-con- tinued heterozygosis on a variable character in Drosophila. Two figures. NOn3:- APR S. O. Mast. The relation between spectral color and stimulation in the TOME TROLS aI Sins ME OU SOMES i.y,y00e c. «- - oy oemeetee Ste raieea lc gotctcnscesyisos, «a0 Waxtpo Suumway. The effects of a thyroid diet upon Paramaecium. ‘Twelve IIUSAD UH ETS} cated hn Se Soe RR ee Mee SEE, 2 dn Rate k ead See Se een oro il 125 187 193 231 241 lV CONTENTS Mary J. Hocur. The effect of media of different densities on the shape of ame Dac Mee CVT CURE Sit Scers a... < co fc oi Ae ee ee Se) ee WHEELER P. Davny. The effect of X-rays on the length of life of Tripoli Comhusuini er, Hive wh CULES teens cre tce +. cae eae on ces aie Sea RicHARD GoLpscHmip?T. A further contribution to the theory of sex. Fifty- three figures 565 573 aN CYTOLOGICAL STUDY OF ARTIFICIAL PARTHE- ; NOGENESIS IN CUMINGIA MARGARET MORRIS Osborn Zoological Laboratory, Yale University FOUR TEXT FIGURES AND EIGHT PLATES I. INTRODUCTION The following experiments with the egg of the mollusc Cu- mingia tellinoides were begun at Woods Hole in the summer of 1914. At first it was intended simply to find an easy method of inducing artificial parthenogenesis and to make a cytological study of eggs undergoing such development. A method was soon found which gave a fair percentage of cleavage and a few swimming larvae in most experiments. LHarly in the course of the experiments, however, it was observed that while some eggs gave off both polar bodies in the normal manner, others formed only one, and still others passed at once to a 2-cell stage with- out forming polar bodies at all. From the cytological study made in the following winter, it was found that in some of the eggs subjected to the treatment, both nuclei formed by the division of chromosomes in the first polar spindle are retained in the egg, and that these two nuclei fuse. It seemed as if a sort of self-fertilization had taken place in these eggs, and the question whether this process was the beginning of normal de- velopment became the central one of the problem. The later experiments were, therefore, made for the purpose of finding out whether the swimming larvae obtained by the partheno- genetic treatment came from eggs which had formed polar bodies or from those which had not. When it was found that normal larvae do, in fact, come from eggs in which maturation has been suppressed, it was still necessary to determine from a further study of preserved material whether the eggs in which the polar nucleus fused with the egg nucleus were those whic’ if THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 1 JANUARY, 1917 2 MARGARET MORRIS developed to larvae. There might, of course, be some other way in which eggs without polar bodies could develop, and those in which the fusion of nuclei took place might prove to be,. after all, incapable of further development. Finally, besides settling this question, it was thought that a study of the pre- served material would throw light on the question as to which of the two maturation divisions was the true reducing division. II. METHODS a. Experimental. Cumingia is fairly common in the waters around Woods Hole, and each female, when ripe, yields a large number of eggs. The animals are brought in dry, and do not begin to spawn until placed in sea-water. It is therefore easy to. obtain eggs free from contamination by sperm, by placing each individual in a separate dish of water. To guard further against contamination by sperm the usual precautions with regard to clean dishes were observed. Controls of all the ex- periments were kept and carefully watched. Very little tend- ency towards parthenogenetic development such as Morgan (10) observed in the course of his experiments on Cumingia was seen jn these controls. In one case a few eggs were found which had formed the first polar body, but no cleavage stages or larvae were found in this or in any other control. The agents used to obtain parthenogenetic development in these experiments were heat and hypertonic sea-water. Heat alone, hypertonic sea-water alone, and hypertonic sea-water fol- lowed by heat gave some development, but the results were not so satisfactory as when heat was used first and followed by hypertonic sea-water. This is, of course, in accord with Loeb’s method of artificial parthenogenesis, in which a cytolytic agent is followed by a ‘corrective.’ It does not come within the scope of the present study to say whether the effect of the hypertonic sea-water is in fact corrective as Loeb thinks, or additive as R. S. Lillie has maintained more recently. Loeb and Wasteneys | induced normal segmentation and formation of larvae in the eggs of Cumingia by sensitizing them with SrCl, and treating ARTIFICIAL PARTHENOGENESIS IN CUMINGIA a them with ox-serum followed by hypertonic sea-water. No cytological study of these eggs was made, and from the brief notes the authors have published it is impossible to say whether the results of this treatment correspond with those set forth in the following pages. The technique of these experiments is simple. A small flask of sea-water is suspended in a beaker of water and warmed over a flame till the temperature of the water in the flask is slightly above that to which one wishes to expose the eggs. The intro- duction of a little cool sea-water with the eggs lowers the tem- perature somewhat, and it is an easy matter to keep it constant for an hour or more within the rather wide limits necessary for these experiments. The temperatures to which the eggs were subjected varied from 32°C. to 37°C., and the length of the exposures from 13 to 90 minutes. The treatment with hypertonic sea-water was also varied, and the interval between the two treatments, as well as the interval between the spawning of the eggs and the beginning of the experiment. A detailed study of these varia- tions is given in a later section and in the tables. Some eggs were fertilized, in order that a comparison might be made between the normal and the parthenogenetic develop- ment. In fertilizing eggs, care must be taken to avoid poly- spermy by using a dilute sperm suspension. b. Microscopic methods. Material was taken from the ex- perimental cultures at varying intervals, and preserved for cyto- logical study, and corresponding series were made of the nor- mally fertilized eggs. Almost all of this material was preserved in Bouin’s fluid. A few sets were fixed in Mme. Danchakoft’s modification of Zenker’s fluid, in which 8-10 per cent formalin is substituted for the glacial acetic acid. f As the eggs are very small, they were stained in této before embedding. Conklin’s picro-hematoxylin was found to be most useful for this purpose. Some of the material was stained with borax carmine, but this makes the sections opaque and practically valueless unless they are bleached with chlorine gas. AY. MARGARET MORRIS The eggs were carried up through xylol to paraffin in the vials in which they were preserved, and were kept in the vials while the process of enfiltration took place in the embedding oven. When they were ready for embedding each vial was taken from the oven and held against a cool surface till the paraffin at the bottom of it began to solidify. The rest of the paraffin was then poured off, and the solid mass at the bottom of the vial was picked out with a fine forceps and embedded in a dish of hot paraffin, just as if it were a piece of tissue. As all the eggs settle to the bottom of the vial during the process of enfiltration, the solid mass contains them all and none are lost. If the thing is done quickly so that the mass of paraffin is still very soft, there is no difficulty in getting it out of the vial or in embedding it. On account of the large number of eggs cut in a single section it was often difficult to follow one egg through a series of sec- tions. For this reason, large numbers of paramecia were fixed and embedded with the eggs, and their presence served as a guide to the location of an egg in successive sections. Sections were cut 7-9 m. in thickness and stained with iron hematoxylin. Some were counterstained with Orange-G, but this is no im- provement on the plain iron hematoxylin stain. Ill. DEVELOPMENT OF FERTILIZED EGGS The maturation and fertilization of the egg of Cumingia have already been described by Jordan. For the sake of convenience, however, it is best to review the phenomena here, especially as the question of the individuality of the chromosomes and the genetic continuity of the centrosomes which were Jordan’s main interest are not those which concern us. The egg is laid after the first polar spindle has been formed. This spindle (represented in figure 1) is large, and lies near the center of the egg. It might, in fact, be mistaken for a cleavage spindle in the metaphase, if it were not for the form of the chromosomes which is entirely different in the two divisions. If the eggs are not fertilized or subjected to any parthenoge- netic agent they remain in this condition, with the spindle near the middle. After fertilization has taken place, however, the ARTIFICIAL PARTHENOGENESIS IN. CUMINGIA 5 spindle moves nearer to the periphery of the egg, the eighteen chromosomes divide, and a normal anaphase ensues (fig. 3). The cytoplasm of the egg pushes out to form the first polar body, and the first maturation mitosis is completed (fig. 4). The second maturation division follows without the interven- tion of a resting stage (fig.-5).. During the formation of the second polar body, the first is often pulled back into a concavity of the surface of the egg, so that it would be invisible in a sur- face view, but the walls between the two cells remain intact. Such a condition is represented in figure 7, but in figure 6 we see a very different case in which the first polar body and the cytoplasmic bud for the second both stand out prominently be- yond the surface of the egg. The second polar spindle (repre- sented in figures 6, 7, and 9) is rather smaller than the first, but is like it in having eighteen chromosomes which divide and go to the poles of the spindle in a perfectly normal manner (fig. 9). The chromosomes in the two divisions are, however, quite dif- ferent in form. Those of the first maturation are varied in shape and size, but are all larger than those of the second polar spindle. No regular series can be made of them, as has been done with the chromosomes of some other forms. The two clearest equatorial plates of this spindle that were found are represented in figures 2a and 2b. They are alike in having two rings and one U-shaped chromosome apiece, but one could not say whether these forms are constant without much more study of the point than seems worth while in the present connection. The remaining fourteen chromosomes of the first polar spindle are round, oval, or cross-shaped bodies of various sizes. At the beginning of the anaphase, as soon as the arrangement in a regular plate is lost, the chromosomes undergo a marked reduc- tion in size. The daughter-chromosomes of the anaphase (fig. 3) seem less than half as large as the chromosomes of the equa- torial plate. The equatorial plate of the second polar spindle is repre- sented in figure 8. Here the chromosomes are all in the form of short rods with somewhat irregular outlines. They are much more constant in size and shape than those of the first polar spindle. 6 MARGARET MORRIS After the formation of the second polar body, the chromo- somes that have remained in the egg form a large female pro- nucleus. The sperm nucleus has enlarged, in the meantime, and the two nuclei unite (figs. 10 and 11). No attempt has been made to trace the history of the aster which appears before the two nuclei fuse. The first cleavage spindle does not lie in the center of the egg, as the first division is an unequal one. It is about the size of the first polar spindle, but the chromosomes are very different in form as well as in number from those of the matura- tion divisions. In the cleavage spindles they are long, thin rods, or threads, with slight terminal swellings. They are so much bent and intertwined in the equatorial plate that an accurate count is impossible, but the number is presumably thirty-six. Figure 13 shows the anaphase of the first cleavage, in which the chromosomes are small rods. Throughout the early cleavage the chromosomes have, in the metaphase, the form of long threads; but in the later development a gradual change of shape is seen. Figure 15 shows chromosomes from a middle cleavage stage, and figure 16 is from an egg fixed nine hours after ferti- lization. Here the chromosomes have been reduced to the short rods which Jordan says are characteristic of Cumingia, and which are considerably smaller than the chromosomes of the fourth cleavage, for instance, shown in figure 14. The normal cleavage pattern of Cumingia has been described by Browne (710) in her study of the effect of pressure. It is illustrated here for comparison with the cleavage of partheno- genetic eggs in the text-figure 1 (surface views) and figures 17 to 28 (sections). The first cleavage plane passes through the polar bodies and divides the egg into two unequal blastomeres (fig. 17, text-fig. I, 3 and 4). Of these the larger one (lettered CD in the drawings) is usually the first to divide. The spindle forms in the middle of the cell, but moves to an eccentric posi- tion, with the outer end slanted towards the cell AB, while still — in the metaphase (figs. 19 and 20). The result of this division is a 3-cell stage in which the cells are all unequal in size, C being equal to about half of AB (figs. 23 and 24), text-fig. I, 6 and 7). ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 7 The division of AB which follows leads to the 4-cell stage repre- sented in figure 25 and in text-figure I, 8, in which there are three cells, A, B, and C, of about the same size and a larger cell D. The size-relations of these cells are very constant in normally fertilized eggs. ‘The time-relations of the divisions are less constant, as may be seen from the different phases of the spindles in figures 18 to 24. In some cases AB and CD 8 : MARGARET MORRIS divide simultaneously, as in figure 22, so that the egg passes at once from a 2-cell to a 4-cell stage. The normal cleavage is not given here in detail beyond the 4-cell stage. A 5-cell stage (not illustrated) results from the division already begun in the egg illustrated in figure 25. The cell d; which is cut off from D is about the size of A, B or C, so that there are four equal cells forming a cap on the cell D which is still larger than any of them. Four hours after fer- tilization the egg is in the condition shown in figure 26. Thé large cell is a descendant of D, the smaller ones of A, B, C, and d,. Figure 27 shows a late cleavage stage in which the lineage of the cells cannot be traced. The 24-hour larva is sketched in surface view in text-figure 2, 1. The larvae are lively swimmers at this time, and it is diffi- cult to get a drawing of a living specimen which shows more than the general size and shape. The section (Tig. 28) shows, however, that the mid-gut, stomodeum, and shell-gland are already formed, and that scattered mesenchyme cells are pres- ent in the segmentation cavity. IV. DETAILED STUDY OF EXPERIMENTS a. Best method of inducing artificial parthenogenesis. As has been said, many variations were made in the treatment by which artificial parthenogenesis was induced. The most significant variations are those in the exposure to heat, and the experi-— ments may therefore be grouped according to the temperature to which the eggs were exposed. Variations in the hypertonicity of the sea-water used as a second treatment and in the length of the exposure to this agent seem to be of comparatively slight importance, and it makes little difference in the result whether the eggs are used as soon as they are shed or allowed to stand for some time, first. The length of the interval be- tween the heating and the treatment with hypertonic sea-water is also unimportant. For counting the percentage of eggs that develop to larvae the following method was used: Five hundred eggs were picked out at random from the culture and set aside, and the number of ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 9 larvae that developed from them was counted the next day. If there were no larvae in this lot, but some in the main dish, the number is recorded as ‘few.’ If there were only a very few in the main dish, that fact is recorded. Although the eggs vary greatly in their susceptibility to treat- ment, a tabulation of the experiments shows that the highest percentages of larvae result from an exposure of 60 minutes to a temperature of 32-33°C. This treatment was used in eighteen experiments, of which four failed to produce any larvae and three gave less than 0.2 per cent. Of the remaining eleven ex- periments, ten yielded larvae in proportions varying from 0.2 per cent to 4 per cent and in one 18 per cent of the eggs de- veloped to this stage. The unusually successful result of this last-mentioned experiment is an illustration of the variation in the susceptibility of the eggs to the treatment. All attempts to obtain a similar result by repeating the experiment exactly were failures—no one of them gave more than 4 per cent of larvae. One might think that the experiment was contaminated by sperm except for the fact that the control showed no devel- opment, and there was no evidence of fertilization among the eggs preserved from the experiment for cytological study. Although 60 minutes is the optimum length of exposure to 32-33°C., larvae may be obtained by shorter exposures. Even a 30-minute treatment may give 0.2 per cent larvae, and a 45- minute one may give | per cent. Exposures lasting as long as 90 minutes also give fair results. To summarize, we may say that out of forty experiments in which this temperature was used, six gave no larvae at all, nine gave less than 0.2 per cent, and the remaining twenty-five gave 0.2 per cent or over. For further details of these experiments, the reader is referred to table 1. ; The next group of experiments is made up of those in which the eggs were heated to 33-35°C. For this temperature the optimum exposure is 30 minutes. This group includes four experiments which were made at the beginning of the study, when no counts were made to obtain the percentage of larvae. Setting these four aside, however, we have the following re- MARGARET MORRIS 10 ‘JoyBMBOS OIUOQIedAY OY An Suryvur ur pesn SEM [OPN JO JOM. JO uorynyjos pw FZ B syuowTedxo oy} ynoysnory | 1 May AIBA | 100d ‘ulm 0¢ O&MS + OOM g 09 smoy $§ Pooyg may ATA | 100g ‘ulm 0€ OMS + OTIOM f 09 saMoy E Pog ouON | FI %O1 ‘ulm 0g OfMS + OLIOM if 09 sanoy Z p0ojg euoN | 0€ % ‘ulm 0€ O¢MS + OIIOM g 09 sinoy Z p0oyg ST 9% Moy AIO A ‘ulm 0€ OMS + OTIOM ¢ 09 sanoy ¢ pooyg Z'0 ¢ + YC ‘umm 0¢ OMS + OOM z 09 ysory oF 91 Moy AIO A ‘ulm 0€ OGMS + OTIOM G 09 ysory 9°0 100q YET ‘ulm 0g O&MS + OIIOM G 09 ysoayy Mag | Tf | PZ OU 4ST OUIOg ‘ulm ¢ O¢MS + OTIOM cg 09 ysody i 91 ‘say Z OSMS+ SIOM Z 09 Ysod yy vi &% ‘Siu 71 OSMS + SION z 09 ysoly ¢°0 ial %6 ‘ulm 09 C&¢MS+ SION ¢ 09 sdnoy Z poorg € 0Z ‘ulm 09 O¢MS + SIOM z 09 ysor yy (6, Ce K¢ aml0g ‘uM OF OSMS + SIOM ¢ 09 ysoay v0 Dh MOT ‘ulm €T OSMS + SIOM cI 09 ysory £0 61 | PZ Mo} “4ST EUIOg ‘ulm gf Of¢MS + SIOM 0 09 Sod] I Or eulog ‘ulm 6% OGMS + OTIOM iy gg Yysor 9°0 9 eulog ‘ulm 0g OGMS + OIIOM € 0g SINOY FZ P0OIS I 100g %e OL ‘ulm Ze O¢MS + OLIOM 6 cr ysolg QUuON | 100g %G'E ‘ulm 0g OSMS + OOM 61 CF Ysod yy Moy AIOA 0% ete ‘ulm Q9 OGMS + SIOM T CHE smmoy Z poojg (ps MAT | § | puB 4sT) AUB ‘ul OF OSMS + SIOM g 0g ysody 20 G Aurur poos VW ‘ulm 6 OC¢MS+ SIOM cI 0g ysory quad wad quan sad “Ure “Ur @Q@VAUVT sees saiadod uvi1od YUALYM-VGS DINOLYUAdAH TVAUGINI Soars soda AO NOILIANOOD agZ VEL 99 VoL ago vg9 8¢ 69 'V8P O6¢ “X@ AO "ON “DoS&-6& .109 H W ATEVL ial IN CUMINGIA e ARTIFICIAL PARTHENOGENESIS aUuON aa 0 v0 MO J] ST i Mo} AOA Moy AIO A ouON ouoN ST Il GG c& 8G 8G SI ai Or 61 éI &G GG 9T OL cI atlog MO MO J MOT MO JT (qsay) eulog auoN MOT auoN MOOT auoN auoN MOT MOT MOT {ouoN “Uru “UTUL “UTUL “UTUL “UTUL “UTUL “UU “UTUL “ULUL “UTUL “UTUL “UTUL “UT “UTUL “UTUL “UTUr “UTI 06 a OSMS OSMS OSMS OS MS OSMS OSMS OSMS OSMS OSMS OSMS O¢MS + OSMS + OSMS + OSMS + OSMS + OSMS + OSMS + ++++++++4++4 cIOM O1lIOM CIOM clOM clOM cIOM 01OM cIO®N clOM CIOBN clOM cIOM cIOM CIOPN GIOM O1IOM O1OM usd A) SH oH SH HH 2D od CO 1 CO 1) ao Yysod iy] YSot yy] sory ysoly] soli] Ysodyy ysody Ysod] ysod iy YSoliy YSoL {Soto sod yf sod] Yysor yy sunoy ¢Z poo YSsod yf 12 MARGARET MORRIS sults from these experiments, of which there are twenty-two. Hight gave no larvae at all, twelve gave few, or very few, and only two gave over 0.2 per cent. The highest percentage ob- tained was 4 per cent. ‘These experiments are set forth in detail in table 2. Finally, a few experiments were made with a temperature of 35-37°C. If the eggs are subjected to this temperature for ten minutes, fragmentation ensues. The optimum length of ex- posure is two minutes, but there was only one of the eleven experiments with this temperature which resulted in the forma- tion of larvae, and in this case they were very few. Table 3 shows the results of these experiments. A few experiments were made with temperatures below 32°C. but they were entirely unsuccessful. b. Experiments to show that larvae come from eggs which have not formed polar bodies. Up to this point, no mention has been made of the effect that different exposures to heat have on polar body formation. We must consider this question, as it gives the first evidence that it is eggs which have not formed polar bodies that develop to larvae. It has been seen that exposing the eggs to 32—33°C. for an hour gives the highest percentage of larvae, while the two-minute exposure to 36-37°C. gives very few larvae, or none at all. It is, however, the latter group of experiments which gives the highest percentage of polar body formation. Compare, for instance, No. 53 (table 3) and No. 65 B (table 1). In the first, polar bodies were formed in almost all the eggs, but only 2 per cent divided, and none developed to larvae. In the second, the polar bodies were very few, while 26 per cent of the eggs divided and 18 per cent grew to larvae. These experiments represent the extremes, and perhaps the point is more fairly illustrated by an experiment in which one lot of eggs was divided into two portions. One half was ex- posed to a temperature of 36-37°C. for two minutes, the others kept at 32-33°C. for thirty minutes. The first half showed 22 per cent with polar bodies, hardly any cleavage, and no larvae; the second had 10 per cent polar body formation, 25 per cent dividing and 0.2 per cent larvae. 13 CUMINGIA IN ARTIFICIAL PARTHENOGENESIS “ULUL (9 “UTU (9 “UTUL (9 “UIUW GZ “ULUL 0G TOS, TGS STOIOS Se Os SLANOOE Of AOD ENE “UIUL 8 “UIUL “ULUT g “UTUI 8 “UIUE g "sIy Z “UIUL 09 gees (019) “UIUL (9 “UIUL (0 EEN SS “ULUL OE “UlUL ZE “UTUL GT “UTUI 09 OS MS CGMS OfMS CSMS 0SMS OSMS Og MS +++4+++4++ G1IOM MODs S{D 51 SION {TOM STOM STO O¢MS + O1IOM OfMS + OTIOM O¢MS + OL1IOM O¢MS + OLIOM OSMS O¢MS O¢MS OSMS OSMS OSMS OfMS OSMS OSMS OSMS OSMS + 0¢MS + +++++4++4+4++ sIOM slOM sIOM SIOM SIOM SIOM SIOM SIOM SIOM SIOM S1IOM S1OM O€MS + OTIOM O€MS + O11IOM OSMS + SIOM auoN CT Aue A[pav yy MO JT 6 {ouoN MOT 91 {ou0oN MOH SZ MOT CF MO] L¥ MOF Gg AO] ouoN el aulog auoN | 100g aulog Moy | 100g aulog ouoN O1 {ouoN auON |poor) aulog ouON | IIB MO | Moy | puodes puUB 4SILT MOY jeulog AUB ~ Aurvyy joulog aulog UOT} V{UIUISBI,J—JUsUIdoOJoASp ON MO JT 61 awlog MOT aulog ¢°0 GG SutOs MOST eulog 7 gI aULOY ouoN II aulos Moy AlaA | 100g aulog MO JT y aul0g (PZ MO WT Ll puv 4sT) Aueypy quad Jad quaa wad GAVAUVT ae SaIadod UV10d UALVM-VaAS OINOLYAdAH io.8) Se: S (Skotos. (O20 (i = CNS baal isa) NME~e NWI NWO N N ol IVAUGLNI 09 09 cP 8& ste 8& 8& O& O€ O& 0€ O& O& 0& O€ O& 0€ 0& 0€ 0€ 0& 0€ O€ GS OL OL “UU LVaH OL au Osodxa "‘DoGE-6E 109H % WTA “ry FT pooys ysod ysol yso.l iy ysed yy ysoly ysed Peers ysody ysol Ysot] ysol ysol] qsedyy ysoel ysol ysod iy ysol Wr leeP EEL ysol i] ysel yy ysod iT ysol yf ysol Ysol nin “IY, Ysol yy §$99a JO NOILIGNOO 08 daZ¢ OLS avs OF8 avs V8 92 aé8 Ves 09 It 9& GE 0Z aZt 98 alg 68 ass V68 V88 $8 18 V09 VZS “X@ JO ‘ON MORRIS MARGARET 14 auoN au0N aUON auoN auoON MO} AIO A auoN ouoN auoN quaa wad GUVAUVT ouoN eul0g 100g auo0g 100g aulog auoN awlog %E au0g (PZ. MIT | pus ast) AusBy Aue Ay -pivyy AUB], %Z | S88 ][B SOUTY | Zl ueyy Sso'[T quaod wad %GS adv -yuourdopaAop ON—poyuowseay sddo yo uorzs0doid os1e'T O1 ‘ulm OT OSMS + SIOM OF g ‘ult Tg ~OSMS + OTIOM 81 F ‘ula 9T OSMS + OTIOM LY 6 ‘ura 0g OMS + OTIOM if G ‘ulul GT =OSMS + OLIOM L G ‘ulul OT OSMS + SOM OF G ‘ul GT OSMS + SIOM ST G ‘ulut OT OSMS + SIOM 0€ G ‘ulm O§ OMS + OTLOM 16 tl “UU Ure UALVA-VAS OINOLUAMAH IVANGLINI pole -AVa@'IO Salaod UvIod "‘DokE-GE 109] € ATAVAL ysot yy ysol iy] SINOY FZ P0odg Inoy FI poorg Inoy [ pooys simoy +7 poo ySsody] Ysoly] ysol sanoy £g poo4sg spda dO NOILIGNOO ‘Xa JO ‘ON ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 15 In most experiments, no attempt was made to count the per- centage of maturation, as the polar bodies would be hidden be- hind the egg in many cases. In the experiment just described, the counts were made to give the relative numbers in the two parts of the experiment. In one half 22 per cent of the eggs had polar bodies that were in such a position that they could be seen without rolling the eggs around; in the other half the count was only 10 per cent. The percentage of cleavage, on the other hand, represents the actual proportion of the eggs that are undergoing segmentation, as a 2-cell stage is recognizable as such from any point of view. These experiments serve as an illustration to show that, in general, in a case where polar body formation is good, cleavage is poor and few larvae develop. But they do not prove conclu- sively that larvae come from eggs in which maturation has been suppressed. To do this, it was necessary to resort to the method of isolating eggs without polar bodies and observing their development. The highest powers of the binocular micro- scope were used in selecting the eggs, and each one picked out was first rolled around with a fine brush and viewed from all sides to make sure that it had, in fact, no polar bodies. In all, 1215 such eggs were isolated; and from these, five larvae devel- oped. One of these larvae is shown in text-figure 2, 2. It is smaller than the larva from the normally fertilized egg, but is evidently cellular, and is fairly normal in external appearance. This showed, then, that larvae could develop from eggs which had not formed polar bodies, but did not prove that eggs which had undergone maturation could not also form larvae. That such eggs could divide at least once was shown from the 2-cell stages with polar bodies which were occasionally found in the cultures. 313 eggs with one or two polar bodies were iso- lated, and from these a single swimmer developed. This was not, however, a normal larva. It is illustrated in text-figure 2, 3. Evidently it has developed without further cleavage from a 2-cell stage. Abnormal swimmers like this are not uncommon in the ex- periments. Sometimes an egg will be found that has been 16 MARGARET MORRIS differentiated entirely without cytoplasmic cleavage, such as is shown in text-figure 2, 4, or one like text-figure 2, 5, in which one cell has developed to a ciliated swimming structure while the other remains an inert mass. This phenomenon of differ- entiation without cleavage is known in the parthenogenetic de- velopment of other forms. In the egg of Cheetopterus, for instance, Lillie (02) has found it to be of common occurrence, and Scott (’06) has also produced it in the eggs of Amphitrite. In both of these cases it appears to be independent of matura- tion. Eggs without polar bodies may develop in this way as well as those which have matured fully. In Cumingia there is no evidence that the phenomenon is confined to eggs with polar bodies. The preserved material does not give a complete series of stages, so that the cytology of this kind of development could not be worked out at this time. Table 4 gives the details of the experiments in which eggs without polar bodies were separated from those which had un- dergone maturation. The numbers are not very large, to be ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 17 sure, but one may fairly conclude from them that eggs in which maturation has been suppressed are able to develop to cellular larvae; while those in which it has been completed form ab- normal non-cellular swimmers if they develop at all. This con- clusion is supported by the experiments already described, which showed that a high percentage of maturation resulted in a low percentage of cleavage-stages and larvae. It js also sup- ported by the cytological evidence given in the following sections. TABLE 4 Isolated eggs 1 oR 2 Eis |) See ee a Lee Numberudlsolatedsss: tn seen eres on. 100 INimberrotelarvidewey wast va wastes cele 0) Niumibersisolatedheers ts orice tesa roa. 4 296 300 Numbernol lanvacesaemer.. seas: 0 2 P INumbersisolated 2245.5 ee ee ele oe 3 34 179 200 INmMber OlmlanVae. co. clnes Wns cies: 0 1 INUumberaisolatedsrrmer ase asec cote. 25 90 200 Numibertotplanyacseae cence eee a. oe 0 0 0 INUmberisolatedess roe centre. : 15 300 INIMper. OL Manvae# aun cia emests ie) a.- 0) 3 Nimmibensisolatediees jac ccsovcmiekisccs ose. 25 100 400 INumbersotelamvales. santo) os cise oasis. ++ 0) 0 0 INumbersisolatedi tesa... Sissies +c 25 100 400 INtimberiotelarvaet meres eee co - 0 0 2 INtmbenisolaredce seems seein. te: 40 100 300 Niumibertof larvae: seein ee ee 1 0 2 Numiber isolated: aay saeco -...- 4 100 300 INtimiberiotslarviacuyaaeetia pine as oes 0 1 0 INUumbenisolated: sameeren... 100 200 INGIMbDERIOn VanrvaAe. sence ees see ©: 0 1 Niimbenisolated 4: . 420s seamen ene 5 7 100 400 INEM DeTAO LL AT; Vials, ater eR ae tai. 20 0 il 16 INumibensisolatedss cst cates one 9 50 IND eraO le larvae: -y.04, de bo eer. Hee a Sees — - v= / 4 0 \ x! .e Pa ~ Yn Bh . 3 ETA sec ee @ =, INS =, SSS % - / \= r 74 a: 67 i ry ; . 8 49 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 1 PLATE 8 EXPLANATION OF FIGURES All the figures in this plate are from parthenogenetic eggs. 74 Resting nuclei from second polar spindle. 75 Cleavage nucleus of an egg with one polar body. 76 Vesicles in egg with two polar bodies. 7 Cleavage nucleus of an egg with three polar bodies. Cleavage chromosomes, egg with two polar bodies. 50 ARTIFICIAL PARTHENOGENESIS IN CUMINGIA PLATE 8 MARGARET MORRIS _— Re a pane ON THE REACTIONS OF AMEBA TO ISOLATED AND COMPOUND PROTEINS A. A. SCHAEFFER From the Zoological Laboratory of the University of Tennessee g 4 SIX PLATES CONTENTS TIMUPOYSL TSAO NT? Hoe acct pee ers OE Sones Oe RCR En iris 6.5 ciGin Sd CIRM RMI oe aIG VOD) BXperuMents, wit be rlOl Uli cess ies. < 0s, Avcechctey~ 2 cheney faeeeeeets| yaiolciolels oe esieiedeese setae 55 (Giemiahl ene hone) ov ISH we ncena Gece bo Rae Neuron t,o Lin Ce OS aba ees ee ON 56 IREROtGINS Wey ERS FEC OH! CAlOlouillti a gassoeeoaodsd scan mdeeopenobedoaenotn Ot) va Onell amie Dasrarre © set mre. Byock cect res 6 os feed chee chal aes oh ae legend 61 SS UMN ayer ey ERP eee aie PPA Lac tess, 6s oR an Seectay 2 leas ave shane oe 63 IB xperiimentishayliuia Ol Uiteme sersiuerey ladies: Aho 2 Serpe tech-c sum eyecce Secia soils aucun 2 64 SULTON ex Taya, Hr arg Pa ee art Sy Nig i UPA coy oe of os 7s 2 xg op RR Feels ttle eon AMV ch 68 Expenimentsnwibnealeunonatiessemmccs (ste. s- = oem enemas rac eicie steerer 69 Exp erimMemiGs ava bly a CGA UMA Mi oes cero: sisi sh 2 2d = Shera SRNR atetctgieneyahS Sas esetael saya 70 SKOUASTAMEW ALS pie oro. eioreueyatc ec OAT o a c-a.e Scien he NEE Ie 16-5,0 0 ot 6 db COW EID Sins BOSS 74 IBp-qoveunbmaVenn nash allo), “Asi to cen peed cabto p acreSaeericicrae aie © 0.0 cS. didi oo bigest cig. Orc 74 EE XPeriIMengs wilh Vv alo UNAle eco crac 6. osc 25a. = ee eRe ele iso tlogelole eters sine 75 Bp ERINMeN GS) Wil NORA GIT selec c cyto cs Se os ce Sopa ane ee er reece lo he aja ent olevevere 76 Exe TIME Mis nwahlawiil GUIs race cess acaysaps cals vie oha.coleunnetetMehe Mekee> Gre Chcie s) oterees Cis evans 78 (Geinsresy ASU TONE alo Bion oO eS OOOO IG OR Ee nnn och ciclo CoD RSE Oa ODO a OI 7 INTRODUCTION Amebas have been shown to react positively to carmine, uric acid, india ink and old solid egg white (Schaeffer, ’16, Jour. Exp. Zool., vol. 20, On the feeding habits of ameba). All these substances are comparatively easily soluble. Particles of these substances are not only eaten when the amebas come into con- tact with them, but all these substances are sensed at a dis- tance of 100 microns or more and the usual result of such sens- ing at a distance is movement toward them. It may be pre- sumed therefore that since these substances are soluble, the reaction at a distance to the solid particle is explained by the diffusion of the dissolved substance. 53 54 A. A. SCHAEFFER Now all these substances except uric acid are compound sub- stances or mixtures of substances, and nothing is known of the rate of solubility of the various ingredients in them as they exist in the mixtures; it is therefore impossible to say Just what part of the india ink or the egg white or the very attractive car- mine affects the sense organs of the ameba or in what manner. Such reactions can shed but little light upon the problem of the selection of food. It seems probable therefore that the use of isolated chemical compounds is to be preferred to the use of mixtures of substances in experimenting upon the feeding reactions of ameba because in this way the number of unknown factors is reduced to the minimum as far as the test substances are concerned. For this reason a considerable number of experiments were performed using as food substances globulin (crystallin and ovoglobulin) lactalbum n, ovalbumin, zein, keratin, and also a few compound proteins such as gluten, aleuronat and fibrin. From the chemist’s practical point of view the isolated proteins mentioned can be made quite pure and quite insoluble. Three of the compounds, lactalbumin, ovalbumin, zein, were made under the direction of a chemist with long experience in this line of work and these products were made as pure as they can be made at present. The other substances were bought from dealers in chemical supplies. But the insolubility of these as of all substances is doubtless relative, not absolute. The chemist, using his utmost skill, may not be able to detect a minute fraction of a quantity of substance going into solution, just as he is unable to detect odoriferous substance in a hare’s foot-tracks. But these small quantities of material are frequently of the first importance to the student of sense perception. So while we shall apply the term ‘insoluble’ to globulin, lactalbumin, zein, ovalbumin, keratin, it should be remembered that not only the word but the meaning also is borrowed from the chemist. Speaking generally, we may not expect to be able to understand the relation between a sense organ and the substances which stimulate it by experimenting with two or three substances which are said to be pure. It is REACTIONS OF AMEBA TO PROTEINS 55 more than probable that a large number of tests with a considera- ble number of substances will be necessary to detect and eliminate errors arising from incomplete knowledge of the physiological action of the test substances due to the presence of chemically imperceptible impurities. Moreover the insolubility of some of the substances such as globulin, lactalbumin, ovalbumin, ete., is influenced to some extent by substances, organic and inorganic, held in solution in the water in which the amebas live. Since it is experimentally impossible to keep pure the water in which the amebas may be placed for observation, on account of their excretory products diffusing out, the proteins themselves may possibly become slightly soluble when these excretory substances come into con- tact with them. On the other hand it should be pointed out that the age of some protein compounds, especially albumins, affects their solubility—the older they become, the more insoluble. In view of all these disturbing factors attending the use of proteins as test substances, it is evident that the observed re- actions must be interpreted with care. EXPERIMENTS WITH GLOBULIN Merek’s globulin with subtitle ‘crystallin’ was used. When- ever the word globulin is used unqualified in this paper this product is meant. A few trials were also made with a purer ovoglobulin on raptorial! amebas (the only kind of amebas ob- tainable at the time) but no definite results were procured. The solubility of Merck’s globulin was tested by soaking for some hours a few small particles in a drop of distilled water on a Clean glass slide and then evaporating the water. By ex- amination with the microscope no difference could be observed between the slight amount of residue of such a drop of water and one in which no globulin had been soaked. Similar experi- ments were also made with tap water with similar results. If any of the globulin went into solution therefore, the amount was excessively small. ‘See Postscript on p. 79. Or (op) A. A. SCHAEFFER Globulin was used in this work as the ‘standard’ food sub- stance. This accounts for the frequent use of it in many of the experiments. The strength of the stimulus proceeding from any other substance may thus be compared directly with that of globulin. The degree of hunger or satiety may also be de- termined within certain limits in this way. That globulin is a real food substance undergoing solution (digestion?) inside the body of the ameba is shown in figures 532 and 534. The ame- bas were fed in filtered culture solution and occasionally in tap water. Granular amebas An ameba ate a piece of grain gluten (326-336,) then moved past another piece of grain gluten, then moved past this piece of grain gluten a second time. A piece of globulin was then laid in its path (866). The ameba flowed directly into contact with the globulin, formed a food cup over it and ingested it. A second piece of globulin was partly surrounded as if ingestion was about to follow, but the ameba presently withdrew (377). Three and one-half minutes later, when the same grain of glob- ulin was presented, it was promptly eaten (383). In this ex- periment globulin seems to have been somewhat more attrac- tive than grain gluten. A grain of globulin was placed near another ameba that had come to rest. It was moving only very slightly (387). At once two pseudopods were sent out toward the globulin, but they were withdrawn before they came into contact with it, the protoplasmic stream then turning to the left. The ameba con- tinued flowing in this direction until the posterior part was being dragged past the globulin (3899). The protoplasmic cur- rent was then reversed, and also a side pseudopod was thrown out on the right (now the left), which turned strongly toward the globulin until it came into contact with it. The side pseudo- pod was then withdrawn, and a little later another was formed on the left and anterior to the globulin. It also started to turn toward the globulin, but was at this stage retracted, whereupon the ameba moved past the globulin without further reaction. REACTIONS OF AMEBA TO PROTEINS bf This experiment is of interest in that it shows a gradual in- crease in intensity of hunger (the quiet condition of the ameba suggests the absence of hunger) until the side pseudopod came into contact with the globulin; then a decrease in intensity of hunger, leaving the ameba with even less intense hunger than at the beginning of the experiment. A somewhat similar result is observed in the following experiment, but infortunately only the first part of the process, the increase in the intensity of hunger, is recorded. In the experiment just alluded to (393) the ameba moved gradually forward until the left limb (899) came almost into con- tact with the globulin grain placed before the ameba, when it was retracted and the ameba moved away. On the second trial with the same piece of globulin (402) the ameba again broke up into Y-form, the left pseudopod being directed toward the test substance (404). When the left prong came into contact with the globulin the right was retracted. The ameba then moved on without further reaction. Note that in the first trial with this ameba, the prong directed toward the globulin was retracted while the other prong became the main pseudopod; while in the second trial the reactions of the respective pseudopods were reversed. This difference in behavior shows that the intensity of hunger increased during the course of the experiment because of the presence of the globulin. Similar to these two experiments is that recorded in figures 1616 to 1626. The ameba moved nearly into contact with the globulin (1619), then withdrew a short distance and threw out a pseudopod on the right. From this side pseudopod another was sent out which moved into contact with the globulin (1624). But as soon as the pseudopod came into contact with the test object, it was withdrawn and the ameba started to move away to the right; but streaming was soon interrupted and directed into the posterior end where a small pseudopod had been slowly forming. The posterior end now became the anterior. There can be little question that the ameba was not hungry and that the presence of the globulin heightened in this case the effect of surfeit. The experiment is of interest because of the striking 58 A. A. SCHAEFFER changes in protoplasmic streaming in the various pseudopods, which were provoked by the globulin. Action was not unified in 1624; the tendencies to move toward and away from the globulin were equally strong. But after a few seconds of con- tact with the globulin, the tendency to move away rapidly became the stronger. At this same time a small pod was form- ing at the posterior end, and it was doubtless this activity which finally prevented the ameba from flowing away through the right pseudopod, which appeared to become the main one. But the protoplasm of the posterior half would have had to flow toward the globulin to get into the pseudopod on the right. The necessity of this may have caused the ameba to flow away through the posterior end rather than through the pseudopod on the right. The side pseudopods in figure 1623 illustrate graphically an internal condition affecting behavior that is frequently observed in amebas reacting indifferently toward food substances. When an ameba that is not hungry encounters a food particle which stimulates the ameba at the side of a pseudopod so that a new pseudopod is formed toward the food object, there is formed simultaneously or nearly so on the opposite side of the main pseudopod a pseudopod of about the same size and flowing at the same angle and at about the same rate as the one directed toward the food. Opposite pseudopods of this character are not formed by hungry amebas under otherwise similar conditions, nor by any amebas toward indifferent objects (glass, sand, etc.) nor toward objects producing usually negative behavior. ‘There are in these reactions two tendencies present, a positive and a negative, judging from the objective behavior. The formation of pseudopods is not directly determined by the stimulating object as can be readily observed from the record of every ex- periment. Then how is pseudopod formation controlled? This question is of the profoundest interest. An unusual method of ingestion in which two pseudopods were involved is shown in figures 939 to 950. The ameba was in Y-form with the globulin lying nearer the right limb. Nevertheless, the left limb enlarged the more rapidly and came first into contact REACTIONS OF AMEBA TO PROTEINS 59 with the globulin. But when both prongs came into contact with the globulin (944) they started to surround it. The ameba was at this time in the form of a ring (945). The tips of the prongs fused as the globulin was ingested. The globulin was ingested in a food cup formed chiefly by the left prong. After ingestion the ameba rested for about twenty minutes, then moved off. Several similar instances have been observed. Reactions to large pieces of globulin In the experiments recorded above fragments of globulin small enough to be easily eaten were always employed. Let us now consider some experiments in which the globulin grains in every case were too large to be eaten. In the first trial (408) the globulin grain was definitely avoided. The globulin was then shifted (412). As the ameba moved directly toward it, the antero-posterior diameter foreshortened, while the anterior edge broadened out considerably. Before the ameba came quite into contact with the globulin, a huge food cup large enough to enclose the test object was started (416). The ameba moved forward into contact with, and laid the partly formed food cup over the grain of globulin, to which the ameba then adhered with but slight movement for several minutes. After this period of comparative quiet, the ameba became more active in its movements and less of its body was in contact with the globulin. Pseudopods were then sent out successively in various directions (422 to 426) only to be re- tracted later. Finally one was sent out (429) which led the ameba away from the globulin, but only after the globulin had been dragged for a considerable distance. The ameba was in contact with the globulin for an hour and nine minutes. One minute later the globulin was shifted (433). The ameba flowed directly into contact with it, and remained so for eleven min- utes. There was no attempt at the formation of a food cup, nor was the globulin adhered to with avidity. On the fourth trial the ameba avoided the globulin (445). On the fifth trial (450) the ameba moved into contact with the globulin and re- 60 A. A. SCHAEFFER mained in very loose contact, without at any time quieting down, for fourteen minutes. On the sixth trial the ameba moved into contact with the globulin at its side (457). The ameba remained in rather loose contact with the globulin for about twelve minutes. The same piece of globulin was, for the seventh time, placed ahead of the ameba (467). As the ameba moved forward it broke up into several pseudopods, indicating negative behavior; but presently one of them moved forward some distance, then turned to the left and toward the globulin, and finally moved into contact with it. The ameba remained in loose contact with the globulin for about six minutes. A half minute later a new piece of globulin was laid in the ameba’s path (482). At first the ameba reacted decidedly negatively and moved on past the globulin. But a pseudopod was then sent out on the right which was directed toward the globulin and which carried the ameba into contact with it. The globulin attracted the ameba strongly but not so strongly as did the other piece in the first trial. The ameba remained in contact with the globulin for about thirty-seven minutes. For the sake of comparison a small piece of fresh globulin was laid near the ameba, now in Y-form (496). The globulin was ingested promptly without the formation of a food cup. Twenty-three hours later the ameba was found to be in normal condition. EXPERIMENTS WITH. ZEIN By reason of the fact that this protein is soluble in 95 per cent aleohol, it may be obtained in a state of great purity. Zein is insoluble in water. This is one of the purest proteins obtainable. In the path of a granular ameba which had previously eaten two grains of globulin but which was unable to retain lactalbu-- min, although several pieces had been ingested, was placed a grain of zein (833). While the ameba was moving forward toward the zein, a pseudopod was formed on the right through which the ameba moved off. ABW 1d | Hda| OLE 9861 pe} verju yp) OI6T ‘T “qoq “Burary FI6I FI wady |6 | Uda | Ssor pozeorju py) GI6I ‘Z “ydeg ‘pros PIGI FI dy |6 | Hdd] g9 (2 JAY X) FZIZ pezvet}u yy) 9I6T ‘T° 4aq SULATT FIG ‘4g [udy |6 | Uda | SPAT (2 [Ayjour X) [ZZ foyooye [AYA «161 ‘@ = “ydag Pier Zo tady |6 | Waa | Z8PT | 6sZT (© [Ayjou X) EZIZ | pe yBeryuy) QI6I ‘GG URE FIG “26 [ady 6 | Uda | L821 pozverquy) CI6I ‘T “ydeg “prog FI6L 22 [Ady 14 | Uda | SPE 6861 pezeerjzuy) CI6I ‘6% Youryy FIGI ‘FI IMdy |é6 | Uda | 98 (© [AyQou X) OZIZ JoyooTs [AUIOTN xaI6l ‘¢ “4deg FI6GI ‘42 [Mdy 16 | Yada | OSFT | ° O6ZT RAYMOND PEARL 142 S8LT poyeerjuy) CI6l ‘91 oune SI6l ‘2o [ady |6 | Hdd | 6ST 698T peyverqu 9I6T ‘TI “oq “Sutary SI6l ‘72 [dy |6 | Hda | OFST #202 ‘6EST peyveryuy) CI6I ‘Z “3deg ‘pjog eI6r ‘Ol tady |6 | Hda | TIST (© 19499 X) LOIZ | Joy «S161 ‘GT “ady SI6I ‘OL [dy [6 | Uda | S21 | ° SLsT 89ST peyverquy, FIGI ‘41 «adv eI6el ‘e ABN 16 | Hdd | F9M (© 10499 X) SOTZ III axSI6I ‘ST ‘“ady SI6L ‘ec Wady |6 | Yad | ZLST | 20ST pe}vorju py CI6l ‘2 “3deg ‘plog FIG ‘ZI 4=AVIN «616 | Uda | 9IS (© 10439 X) OTIZ pozverju() 9161 ‘IT “qoaq ‘Surary PI6I ‘Ie [dy |6 | Uda | OFLT pe}verqu O16 ch UGE FI6I ‘FI [udy |6 | Hada! TZ9T peyvorguy) CI6I ‘IT “yd9g ‘prog FIGI ‘FI IHdy 1/4 | Hdd | s¢ pe}eoryuy CI6I ‘I “3dog ‘pjog FI6I ‘IT [dy {46 | Uda} 09 (© Udd X) Seer pezvorjuy) 9161 ‘8 ue PI6L ‘IT [Mdy |é6 | Wd | FLT (© 19499 X) 90IZ IOyyA axSI161 ‘ST ‘ady PI6L ‘Ie dy |6 | Waa) 16PL | “TzZZ1 peqyverju yy FI6I ‘Z "AON FI6I ‘IZ INdy |6 | Uda] Sos pezvoerquy CI6I ‘Z “3deg “pos FI6GI ‘Ic dy 16 | Uda | 6¢e (© 19y99 X) 601Z po}vemuy) QI6T ‘IT “eq “Burary FI6I ‘IZ [Mdy |6 | Uda] SPLT (© rayye X) SOTZ I9yyg «x16 ‘GT ‘aidy FI6I ‘26 [dy 6 | Ud | O6FT | FS8ZT poyvorjuy) GI6I ‘IZ ‘dy FIGI ‘8c INdy 1/46 | Uda | OFF poyvortyuy) FI6I 2 99d FI6I ‘1G [dy |d | Uda | SIT (© [wu0u X) TEIZ | pe}ve1j uy) 9161 ‘9 = “4a FI6I ‘Ic [Mdy | | Hdd | 9ELT | poyverju yp) GI6I ‘§ [uudy FI6I ‘Ie IMdy |6 | Hdd | FON (© peyvorjun X) 9ZIZ | Ioyyy OI6L ‘IT “Go “Surary FI6I ‘8c IMdy |6 | Uda | SSPI | SeZt | uaa UaAAaWoNn HAAAON ONILVA NI LNGAWLVaAUL aqida dqdadHoOLvVH X48| daauad -WON DONILVN auid xa ponurwu0oO—~2 ATAVL 143 WITH ALCOHOL EXPERIMENTS Il1Z ‘OIIZ ‘601% ‘S01z ‘L01Z ‘901% ‘SOIZ ‘FOTZ ISCIG GolG, -lGlc.. 0CIS 611 slic —Z1le Oliz IGIIGeePIIe =SLIG ZIG CEG cela. Lele Olona! ‘6Z1Z ‘SZIZ ‘LEIS ‘9ZIZ | | ‘ggT “d “Jo ‘yuaprooe Aq pa[ly » x po}vorj uy) TOI [Oyoo|® [AqQaTY pe7eor4 ul) peyverquy) poyoore [Aya joyoo ® [Aya pezverzu poyvorju yp 916. ‘IT “qoaq ‘durary O16 ‘I “qoq ‘surary OI6L ‘T ‘qoq ‘SUTAVT 4So'T GI6I ‘T “4deg 9161 ‘I “qoq “SuLAT CI6I ‘6 “4dag CI6T ‘9% 98. 9161 ‘T “qa “SUIATT FI6L FI6I FI6I FI6I PIbI Fi6l FI6I FI6L FI61 ABIN AB oune ABN ABIN oune oune ‘qudy Judy fo) ves Or OF Or Ha Ha Ha Ha Ha Ha Hd Had Ha g99 |, 9687 £99 |°°006T 697 SOF 199) |e LOGE LSE IFGIN 999 | S061 144 RAYMOND PEARL animals and the controls belong to two different strains or blood-lines. Because of the writer’s belief in the fundamental importance of the general problem with which this paper has to do it seems desirable to take the space necessary to give complete data as to the breeding of all the foundation animals used in this study, covering a period of four years before the beginning of the ex- periments. These data are given in the form of pedigrees. A pedigree extending through four ancestral generations is given for each one of the matings listed in the first column of table 2, or in other words, for all the stock used in the experiments, since, of course, the pedigrees of all the full brothers and sisters in one family will be identical. In these pedigrees I have used the same conventions to indicate ancestral repetition (inbreeding) that have been employed in my earlier papers on inbreeding (Pearl, 19 and later papers in the same series). A solid black circle against an animal’s number indicated a primary reappear- ance, a cross within a circle denotes a reappearance resulting from the fact that some later individual in the ancestral series has been primarily repeated. We may first consider the pedigrees of the families furnishing the pure Barred Plymouth Rock females to the alcohol experi- ments. These pedigrees are arranged in ascending order of mating numbers. The numbers in the body of the pedigree are bird numbers. The males are designated in nearly all instances by a number alone; the females by a number and a letter. When a bird’s number is followed by the letters ‘O. F. 8.’ it is to be under- stood that this bird formed a part of the original foundation stock with which the writer’s poultry breeding experiments started. Information regarding the sources of this original foundation stock will be found in Pearl 17 (p. 137 and 138). EXPERIMENTS WITH ALCOHOL Pedigree of individuals ex mating 1507 ne 564 No. of No. 621 No. ! cr (Mating 1507) No. 9 No. H9 INO OS Glee No. No. No. No. F255 9? { @No. 563 77 No. F105 9° . 550 ot . £309 9 2D560. FS: co! pD407O. ES: 9 2D31-0. BF. 8.0 eD4lL OE. 8S: 9 . 554 J . E237 2 . 556 . E422 9 @No. 5540 @No. H237 2 Pedigree of individuals ex mating 1536 No. of No. 622 No. (Mating 1536) | ( @No. 606 ....... o No. J54 ray (CoyR Ce eee THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 1 ING S080) tie. cs%10 INIOE 3 Eel MES INOsG2UMON). ocy50:1- @No. 587 o...... INO, dG) tla cpogoce Noe TOSCO. tir... No. 564 o' No. F255 9 @No. 563 7 No. F105 2 No. 566 #7 No. F387 9 No. 574 ¢ No. F191 9 @No. 564 7 @No. F255 9 @No. 563 & @No. F105 9° No. 550 #7 No. E309 @ No. D58 0. F. S. & No. D99 O. F. 8. 9 145 146 (Mating 1568) (Mating 1575) of No. 624 g No. J184 No. 624 Se 2 No. J137 No. (No. No. No. No. RAYMOND PEARL Pedigree of individuals ex mating 1668 { \ No. 68 O. No. C161 @No. 563 ¢ No. F105 9 No. 567 & No. F352 9 @No. 563 7 No. F303 9 @No. 563 ¢ @No. F105 9 @No. D560. F.S.¢ No. D381 0. F. S. 9 @No. 563 7 No. F133 9 Pedigree of individuals ex mating 1575 No. DOSNOUanO cee No No 15 US salo Reet a Meee No No GOStot. eae No No. Je Lilt elope Rees oe No. No. GR) ola ca oo nao No | No EU SROn eee er No. [No GOR No No. 15 BY Flee apace No. see eee wee No. 680. F. 8. 0 No. C161 O. F. S. 9 No. D560. F. 8. #7 No. D407 O. F. S. 9 @No. 563 o No. F105 9? No. 567 # No. F352 9 @No. 563 0 No. F303 9 @No. 563 o No. E303 9 No. 569 # No. F177 9 No. 5740 No. E66 @ (Mating 1737) (Mating 1738) ee No. 634 No. J35 of No. 634 Q No. J34 No. EXPERIMENTS WITH ALCOHOL Pedigree of individuals ex mating 1737 No. 550 & No 563 fotereetts k At, Eee ee E309 Q G24) wee Nes No. 598 of Ae ERIS OFS Cae eee G253 9 @No. 563 # [No DSSiiGt eee hee F303 9 Hilvcpetess, 7 if @No. 563 7 Ete: G250O eee \No. E303 9 ( f @No. 563 #1 @No. 588 7....... \ @No. F303 9 GOS G asec { | [ @No. 563 of (Re: G41 Q dodo nc. o6 \ No. F105 ie) [ No. 569 [ No. QO ONCE ee \No. F177 9 HT Oe es { | f @No. 569 v7 [Ne G546 Lot ereeint peo \No. F273 2 Pedigree of individuals ex mating 1788 { No. 550 ot [ No. 563 Gur sean me E309 fe} G24e Gly ae A J No. 598 37 he: H18 ORE 6 eh Ha in G253 Q J @No. 563 07 Nomassiict... 0.8. No. F303 9? NS Og W(ekeh I eo eae x {f @No. 563 o @No. 563 #1 ane: ss) Geass abc f @No. F803 9 GUST cian eee ae @No. 563 7 WowGAItor. ©..02. re F105 9 [No. 553 3 INGRS7S Cts la ss ee E406 9 Gsroetcts 2 Wo HE cc 4 eee S — "*" | No. E202 9 147 148 (Mating 1771) (Mating 1772) —_— No. No. 637 9 No. K63 rot No. 637 2 K65 Z ° No. RAYMOND PEARL Pedigree of individuals ex mating 1771 No. 587 of (No. 606 DEA cyabobees ts. ea G4 Q BODIE vas bs No. 599 ot No. H29 2 Rael ate ccs fel © ae @2ii Q @No. 587 3 | @No. 606 o....... { @No. G4 2 TSAO esc Me ten 4 @No. 563 o INO GRTE) OS gaceenee fos F105 9 ( No. 588 o No. 610 Sucre Ghote « fot No. G25 fe) ‘OP iTe HO RartY ca No. 593 ot ag: H196 Q aad ihetere ee G74 Q No. 550 o No. 563 o......... ae E309 @ IQIS .O a etge ate No. 564 INOS REIGDINON.. ocee ae G79 9 Pedigree of individuals ex mating 1772 f Bo. 587 ot No. 606 Corie neat \No. G4 fe) CPPX ot oe enh ne co {[No. 599 3 No. 18242) CoM ee ue ne G21 9 @No. 587 o @ No: 60696". { @No. G4 2 Bis. omega te No. 563 INGMG27S Oh ee pe F105 9 No. 588 o INOFsGIONG". Je ae oe No. G25 2 PAiot pg sia ang OF No. 598 o IN@, JEDI Ossccsoce oe G151 2 @No. 598 37 INI@s GPA oie soto coor No. G37 9 VOU O en cea @No. 598 7 INOS CEL SQ ae vecouehtere tire ‘— G110 2 (Mating 1784) See (Mating 1785) 9 K68 No. EXPERIMENTS WITH ALCOHOL Pedigree of individuals ex mating 1784 N a (No. No. z { No. 550 7 No. 563 ot Riciaiatetsleleicle ie E309 Q (No. 598 7 INO. EDISiO Seen. - fe G253 9 No..610. cee a pa No. F159 19, eee a Se f @No. 588 7 @No. 610 #....... | @No. G25 9 No:H196) 9-22... ee ae IND fe ae es ee ie (No. 574 ¢ NOs GO0SOR. cee ee F356 9 Pedigree of individuals ex mating 1785 ( @No. 624 2. | 1 - JI. No.1563 OM..0dls... ee aca 3 No. H18 9........ Re Bee 5 [No. Te ee fe ee [xe ms 9.... {RO ee @No. 563 d....... f ee Soe [ox mi. (oR Be, @No. 610 ....... { oNo. one @INo HI58... 23... f cs Boe 149 150 RAYMOND PEARL Pedigree of individuals ex mating 1788 No. 550 i 563 Olen, tees Be E309 fe) [ % ee 624 o silelle) # «! ellehene No 598 ou | | No. H18 Oise ese: s ie G253 Q 23 te) No. 588 3 © No. 610 ot COS. COREE RRO No G25 Q te | = Or eee a No. 599 & Z z | Be H158 ea a seers oe G64 9 a 4 : ®No. 563 @....... f @No. E309 9 Gen IM@ING.1624 o?... /55.% | @No. 598 ot | t @No. H18 2 sielelo'=! elle @No. G253 Q 3 @No. 588 : @NONGIO iota. @No. G25 2 amo. 3137 9... Aa. No. 595 of No. H27 fe) © ee cc cee ee G188 12) Pedigree of individuals ex mating 1789 Nosn68. 0% 001). ee age 9 es ING@3G624 Clan. ssnen No. 598 & | Nee H18 2 ejoselefeteene ae G253 Q 2 : No. BIOS it ie. ine oe se A INGOs eUZ IO esa a J No. 599 o& ez NOS EIISS 9 osteo. co5 No. G64 2 a0 q : [No. 609 oe... Neen = (Mto1623 os hae | @No. 563 of , | INO RNGBO NO 6. a No. F303 2 om | S | J @No. 588 & ie | @No. 610 o....... @No. G25 9 Z ENGI OS TOF is. a ee | : @No. 599 & @No. H158 @...... ae G64 2 (Mating 1790) ee _ No. K820 9 of No. 639 (Mating 1805) ee No. K375 EXPERIMENTS WITH ALCOHOL 151 Pedigree of individuals ex mating 1790 No. 550 0 No. 563 ot aul /citealio eee ee E309 Q Soop boaeeo No. 598 o' No. Hi8 @......... ia G253 ? | No. 588 <7 No. 610 fof a fellciielotelielie ia a G25 2 JAZ ORE. Res. No. 599 o No. F158: 9... ae G64 9 @No. 588 J @No. 610s". { @No. G25 9 DEOGSoHOOL @No. 598 7 lx H191 OM eS 5 Dic te G151 Q @No. 550 7 @No. 563 o’....... { @No. E309 2 J462 2.2.02... No. 567 o' No. G110 OF? ae na F178 (2) Pedigree of individuals ex mating 1805 No. 606 "......... eC a e No. 599 3 Now HOS 9......... re ie @No. 606 7....... { ae aH . ie ae { @No. 599 7 | @No. H29 Oe \ @No. G21 9 No, G10". Pe ay aoe bee EMON Ov eee ay ae @No. 588 7 @No. 610 c'.......1 Oxo. G5°9 No. 593 of No. H82 DT cee ee ee G101 @ 152 RAYMOND PEARL With these pedigrees in hand it is desirable and possible to answer certain questions which arise in the mind of any experi- enced geneticist in connection with experiments of the sort here under discussion. Some of the more important of these are (a) To what degree is the foundation stock inbred? (b) Does the foundation stock represent a relatively wide or a rela- tively narrow range of lines of descent (blood lines)? (¢) Can the foundation stock be regarded as a fair random sample of the general population from which it was drawn? Let us consider these points in order. Fortunately we are able, by means of the coefficients of inbreeding devised by the writer, to give precise and definite numerical statements regard- ing the degree of inbreeding exhibited by this foundation stock. These coefficients are shown in table 3. TABLE 3 Inbreeding coefficients of stock used in alcohol experiments, with comparison data MATING Z Z2 Zs 1507 0 0 18.75 1536 25.00 25.00 31.25 1568 0 0 37.50 1575 0 0 18.75 1737 0 12.50 37.50 1738 0 12.50 31825 Agfa 0 12.50 SSSA AS 1772 0 12.50 25.00 1784 0 12.50 12.50 1785 50.00 50.00 50.00 1788 25.00 37.50 37.50 1789 0 25.00 31.25 1790 0 25.00 31.25 1805 0 37.50 37.50 Means ae: 18.75 29.41 Mean of random sample American Jer- Sy Cattle... sche incd enn: ee ae ee: 4.10 6.97 12.50 Continued single cousin mating.......... 0 25.00 50.00 Continued parent X offspring mating... . 25.00 50.00 68.75 Continued brother X sister mating....... 50.00 75.00 87.50 EXPERIMENTS WITH ALCOHOL L538 From the data given in this table it is apparent that this stock cannot, on the average, be considered to be closely inbred. The mean coefficients, except in the case of Zi, are very decid- edly smaller than the coefficients for single first cousin mating. In all the matings except 1536, 1785, and 1788 the value of Z, is zero. Two of these are half brother and sister matings and the third is a full brother X sister mating. Omitting these three matings we get for mean values: Z, = O, Z. = 138.63, Ag 26214. That the mean coefficients of inbreeding for the foundation stock jn these experiments should be higher than those for a random sample of the general population of Jersey cows (data from Pearl and Patterson, 25, p. 60) is in no way remarkable when it is recalled that the Barred Plymouth Rock flock of the Maine Station has been line-bred for a long time. It is equally clear that the degree of inbreeding exhibited by the present poultry stock is well below the degree (if there be any such) which, per se, causes a weakness and lack of constitutional vigor. This is evident from many considerations. Data which will be presented later on in this paper demonstrate it. In this connection it is of interest to examine a little more closely the performance of the individuals from the most closely in- bred mating of table 3. Mating 1785, which was of brother x sister, contributed two individuals, 1482 and 1741, to the mat- ing list of the alcohol experiments. It will be shown in table 1 of paper No. III in this series that 1482, the ethyl treated sister, produced eggs giving a 75.0 per cent hatch on the basis of all eggs set, and 81.8 per cent on the basis of fertile eggs. No. 1741, the untreated sister, did nearly as well. Of all of her eggs set 70.0 per cent hatched; of her fertile eggs 77.8 per cent hatched. In neither of these records can one find any evidence of constitutional weakness, induced by inbreeding or in any other way. We may summarize the results of our examination of inbreed- ing in the foundation stock used in these experiments by saying that while it is a sample from a line-bred population and is therefore, to some degree, inbred, the amount or intensity of 154 RAYMOND PEARL this inbreeding is low, on the average, even as compared with single cousin mating. The data from which an answer may be obtained to our sec- ond and third questions are most readily had from an examina- tion of the males which have been used as sires in the Station’s Barred Rock flock for a period of years. Since all the females used as breeders within this period will be descendants or col- lateral relatives of these males, we are able to form an ade- . quate idea of the blood lines involved in our alcohol foundation stock from an examination of the males. In the following résumé the bird number of every male Barred Plymouth Rock used as a breeder in the years 1910-1914 inclusive is given. Each bird number is followed by a figure in brackets. These bracketed figures show the number of times the designated male appears in the first four ancestral generations of the stock used in the alcohol experiments, as shown in the pedigrees exhibited in the preceding pages. B. P. R. males used as breeders In 1914 Birds Nos. 634 (2), 636 (0), 637 (2), 638 (5), 639 (1), 640 (0). In 1913 Birds Nos. 619 (1), 620 (2), 621 (1), 622 (4), 623 (1), 624 (11), 625 (2). In 1912 Birds Nos. 563 (36%), 606 (9), 607 (1), 608 (3), 609 (1), 610 (15), 611 (0), 612 (1), 613 (0). In 1911 Birds Nos. 563 (36), 564 (6), 587 (10), 588 (20), 589 (1), 593 (8), 595 (8), 597 (0), 598 (16), 599 (12). In 1910 Birds Nos. D81 (1), 552 (0), 554 (2), 562 (1), 563 (86), 564 (6), 566 (1), 567 (4), 569 (3), 573 (1), 574 (38). From the above data it will be seen that of the 40 Barred Plymouth Rock males used as breeders in the general popula- tion in the years 1910 to 1914 inclusive, thirty-four, or all but six (Nos. 552, 597, 611, 613, 636, and 640) are represented in the 3In this and other cases where the same male was used in successive years the number in brackets is the total number of occurrences regardless of genera- tions, just as in the cases where the male was used during one year only. EXPERIMENTS WITH ALCOHOL 155 pedigrees of the stock used in the alcohol experiments. Of these six males not represented some were used chiefly in cross-bred matings and, therefore, left but few pure Barred Plymouth Rock descendants: others were poor breeders and left few de- scendants of any sort. It seems evident without further discussion that the Barred Rock stock in the alcohol experiments included practically as many blood lines as the general population from which it was drawn and may be regarded as a representative sample of that population. ; The pedigrees of the Black Hamburgs in the experiments may next be considered. Pedigree of individuals ex mating 1896 No. 6300. F.S. # Mating 1896 No. 631 OR Me S.-c No. 1351 O. F. S. 9 No. K222 9 ( Pedigree of individuals ex mating 1900 (Mating 1900) No. 138500. F. S. 9 No. 6800. F.S. o& Pedigree of individuals ex mating 1901 (Mating 1901) ONoe 12a2.0: Fos), 9 No. 6800. F.8. & Pedigree of individuals ex mating 1903 (Mating 1903) No. 1358 O. F.S.. 9 No. 6800. F.8. o& It is evident from these pedigrees that we are much nearer original foundation stock in the case of the Black Hamburgs than with the Barred Rocks. It is important to notice in this connection, however, that these O. F. S. Hamburgs had been previously bred upon the Station plant and their purity and Mendelian behavior thoroughly tested before they were used to breed the stock which was used in the alcohol experiments. 156 RAYMOND PEARL D. Time of beginning treatment The inhalation treatment was started on the different birds at different dates. This was done in order to determine whether the length of time they had been treated prior to entering upon the breeding season would make any difference in their per- formance as breeders in the breeding season of 1915. The data regarding the beginning of the experimental treatment of the different birds are set forth in table 4, which requires no further explanation. TABLE 4 Showing the dates of beginning of alcohol treatment TREATMENT BEGUN ON INDICATED BIRDS DATE Ethy] aleohol Methyl alcohol Ether October SR OAs oon 2 een ee 9 1574 October MLO. an eee 91575 October Orel QU ee ois lor does 9 1572 October SeelGIAT as... keke 9 1573 November (6), 1914.......6:... 2 1482 2 1486 2 1491 IDecembereeelem | OAR Heese emer oas 9 1488, «664 | 21487, 0663 9 1485, 0665 December 201914... 3 se ee. January il: OSes Bena eoe = 9 1484 © 1492 9 1490 Vemouemnye, 1G), IS Se oa sh cabo Q 21481, 1489 E. Plan of matings in 1916 The alcoholized birds and the untreated controls were mated early in February, 1915. Eggs were saved for incubation from these matings from about February 15. The general methods of handling these matings, incubation, brooding the chicks, etc., were the routine methods followed in the writer’s breeding ex- periments with poultry. Long experience has proved these methods to be excellently adapted to the rearing of normal healthy chicks. They are described by Pearl (22). During incubation and brooding, indeed throughout life, the eggs and chicks from these alcohol experiments were not sepa- rated from the general flock. In other words, the eggs from these experiments were put at the same time in the incubators EXPERIMENTS WITH ALCOHOL 157 with eggs from other experiments, being kept separate by wire partitions at the time of hatching for pedigree purposes only; the hatched and properly tagged chicks were indiscriminately mixed with normal chicks from other experiments in the brooders and later in the adult houses. At no stage in the life history have the offspring of aleoholic matings been kept by themselves. By this indiscriminate mixing with the general flock any possi- bility of conscious or unconscious differential treatment of these birds has been avoided. Any differences which appear between various groups of these F; progeny individuals must be attrib- uted to differences antecedent in point of time to their own contact with the environment. The general plan of the matings in 1915 was to breed a treated male of each of the three classes, ethyl, methyl and ether with (a) untreated control females, and (b) with treated females of his own class (i.e., ethyl & x ethyl ?, methyl ~ X methyl °, ether «& Xether ¢?). In addition to these matings an un- treated control male was mated with (a) untreated control females, (b) ethyl females, (c) methyl females, and (d) ether females. A general conspectus of all the 1915 matings is ex- hibited as table 5. All of the matings were of the type Black Hamburg &# X Barred Plymouth Rock @¢. Each male might have been mated with more females than he was. The matings were purposely kept small in number in order that there could be no possible criticism that the males were overworked in the breeding pens and in consequence pro- duced weak or degenerate offspring. There was nothing in the sexual behavior of the alcoholized males noticeably different from that of normal males. All three of the treated males used proved to be very vigorous breed- ers. A characteristic, but, not especially significant, bit of be- havior was noted in the fact that nearly always the first act of a male bird upon being released from an alcoholic inhalation treatment in the tank was to copulate with one of the females in the pen. It is not entirely certain whether this is to be re- garded as an expression of an increase in the libido sexualis 158 RAYMOND PEARL ~ TABLE 5 Plan of matings in 1915 NATURE OF MATING cael fou NUMBER te) NUMBER Untreated o X Untreated @............ 2131 666 1736 2132 666 1744 2133 666 42 Untreatedict/ S@ibther’ 9.00.04. 2126 666 1485 Bthylic >< Wmbtreated. 9). 222s. sea 2116 664 487 2117 664 1741 2118 664 1738 2119 664 1734 Eitinyvlwct: XqGhwaAr®! .<.... Mele hh Gace 2112 664. 1481 2113 664 1482 2114 664 1574 2115 664 1489 Methyl o >@Untreated, 97.4)>.5. eee. 2123 663 1737 2 2124 663 1742 2125 663 1733 Methylict Methyl O°. ...0..... eee 2120 663 1486 2121 663 1487 2122 663 1575 Eithenici < Untreated Os2....+-.aeeeee 2108 21 142 Extheraict >< Untreated! Osa. eee 2110 22 143 Ethyleciex<, Untreated O)...0>. .-4 7 eee 2117 26 147 Ethyl Gane Untreated. :O xn 4: cn eee 2116 27 148 Ethyl @ X Ethyl oN ae 2115 7 213 Ethyl o& X Ethyl 1.) ie oh ea 2112 10 216 Methyl o& X Methyl Odie, oS pe oe 2121 a 256 Ether # X Ether 0... ec ee 2106 15 296 Methyl & X Methyl OMS ers (ee 2120 27 320 Ethyl o& X Ethyl 0). SF eee Pits 20 336 Ether o X Ether O nox cae one 2107 20 349 Ether o X Ether © siya. ane eee 2108 27 351 Methyl # X Methyl. Os Ele aerate 2122 27 354 EXPERIMENTS WITH ALCOHOL 161 one of the parents only is treated and matings in which both the parents are treated. Table 6 gives in order the germinal dosage index for each of the F, matings which produced offspring in these experiments. The matings are arranged in ascending order of total germ dosage index. From this table it is seen that the total germ dosage index for the F, progeny in these experiments ranges from 130 days to 354 days with the matings for the different substances used well scattered over the range. The facts are represented graphically in figure 1 of paper No. III in this series. G. Scope of present reports It is the purpose of the present paper and the two next fol- lowing in this series to present and discuss the data which have accumulated in this investigation from September, 1914 up to February 1, 1916. This includes the F; generation of progeny only. The experiment is of course being continued and later reports will be given on further generations of progeny and others matters of interest not taken up in the present reports. III. SUMMARY This paper is the first of a series of studies having to do with attempts, in the first place, to modify hereditary factors or determinants in a definite and specific way, and in the second place, to observe and analyze the hereditary behavior following such modification. The results here reported followed attempts to modify the germ cells by treating the individual domestic fowl with one or another of three poisons, viz., ethyl alcohol, methyl alcohol, and ether. Summarily stated the chief points brought out in the paper are: 1. The males used in the experiments were pure bred Black Hamburgs. The females were pure bred Barred Plymouth Rocks. There are shown to be numerous advantages in hay- ing the progeny of treated parents F, crossbreds rather than pure. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 1 162 RAYMOND PEARL 2. A detailed account of the breeding of the stock used, prior to the beginning of these experiments, is given. It is shown to be inbred to only a comparatively low degree. It is shown to be a random sample of the general population from which it came. Full brothers and sisters of treated are used as controls. 3. The poisons used were administered daily by the inhalation method in practically as large doses as could be tolerated when given in this way. An account of the methods used and the precautions taken to ensure critical results will be found on pp. 132-138. 4. The total germ dosage index, defined as the foteil ame of days to which the gametes forming zygotes had been exposed to treatment when the offspring were produced, ranged from 130 to 354 days in these experiments, with a mean of 210.35 days, or approximately 7 months. The results of these experiments will be presented in the two next following papers in this series. ‘Those two papers and the present one (Nos. J, IJ, and III in the series) deal with the re- sults from the beginning of the alcohol experiments in Septem- ber, 1914 up to February 1, 1916. Later reports will deal with the results after the latter date. IV. LITERATURE CITED (1) Bastuz, C. 1908 Influenza della lecitina sulla determinazione del sesso e sui caratteri Mendeliani. Rend. R. Acc. dei Lincei. Cl. Sci. fis. ecc., vol. 17, ser. 5a, pp. 643-652. (2) Cent, C. 1904 Influenza dell’alcoolismo sui potere di procreare e sui discendenti. Riv. sper. di Freniatria, vol. 30, pp. 339-353. (3) Conn, L. J. anp Bacuuuser, L. J. 1914 The effect of lead on the germ cells of the male rabhit and fowl as indicated by their progeny. Proc. Soc. Exp. Biol. Med., vol. 12, pp. 24-29. (4) Cote, L. J. anp Davis, C. L. 1914 The effect of alcohol on the male germ cells, studied by means of double matings. Science, N. S., vol. 39, pp. 476-477. (5) CzapeK, F. 1905 Biochemie der Pflanzen, Zweiter Band. Jena, pp. xii + 1026. (6) East, E. M. anp Hayzs, H. K. 1912 Heterozygosis in evolution and in plant breeding. U.S. Dept. Agr. Pur. Plant Ind. Bulletin 243, pp. 1-58. (7) Exprrton, E. M. anp Pearson, K. 1910 A first study of the influence of parental alcoholism on the physique and ability of the offspring. Eu- genic Lab. Mem. X, (London), pp. 1-46. (Second edit.) EXPERIMENTS WITH ALCOHOL 163 (8) GoopaLE, H. D. 1909 Sex and its relation to the barring factor in poul- try. Science, N.S., vol. 29, pp. 1004, 1005. ; (9) HANSEMANN, D. von. 1913 Ueber den Kampf der Eier in den Ovarien. Arch. f. Entwickl.-mech., Bd. 35, pp. 223-235. (10) Heron, D. 1912 A second study of extreme alcoholism in adults, with special reference to the Home-Office Inebriate Reformatory data. Eugenics Lab. Mem. 17, pp. 1-95. (11) Ivanov, J. 1913 Action de l’alcool sur les spermatozoides des mammi- féres (Premiére communication). Comptes rend. Soc. Biol. Paris, T. lxxiv, pp. 480-482. (12) Ivanov, J. 1913 Expériences sur la fécondation des mammiféres avec le sperme mélangé d’alcool (deuxiéme communication). Ibid., pp. 482-484. (13) Kammerer, P. 1913 Vererbung erzwungener Fatbverinderungen. IV. _ Mitth. Das Farbkleid des Feuersalamanders (Salamandra maculosa Lau- renti) in seiner Abhangigkeit von der Umwelt. Arch. f. Entwickl.- mech., Bd. 36, pp. 4-193. (14) Krankheit und Sterblichkeitsverhiltnisse in der Ortskrankenkasse fir Leipzig und Umgegend. Untersuchungen iiber den Hinflues von Geschlecht, Alter und Beruf. Bearbeitet im kais. Stat. Amte. Bd. I, Teil c, pp. 190-198. (15) MacDouaat, D. T. 1911 Alterations in heredity induced by ovarial treatments. Bot. Gaz., vol. 51, pp. 241-257. (16) Nice, L. B. 1912 Comparative studies on the effect of alcohol, nicotine, tobacco smoke and caffeine on white mice. I. Effects on reproduc- tion and growth. Jour. Exp. Zool., vol. 12, pp. 133-152. (17) Peart, R. 1911 Breeding poultry for egg production. Me. Agr. Expt. Sta. Ann. Rept. for 1911, pp. 113-176. (18) 1912 The mode of inheritance of fecundity in the domestic fowl. Jour. Exp. Zo6l., vol. 13, pp. 153-268. (19) 1913 Studies on inbreeding. I. A contribution towards an analysis of the problem of inbreeding. Amer. Nat., vol. 47, pp. 577-614. (20) 1915 Mendelian inheritance of fecundity in the domestic fowl, and average flock production. Amer. Nat., vol. 49, pp. 306-317. (21) 1915 Modes of Research in Genetics. New York, (Macmillan), pp. vii + 182. (22) 1916 Methods of poultry management at the Maine Agricultural Experiment Station (Revised to January, 1916). Me. Agr. Expt. Stat. Circular 515, pp. 1-98. (23) 1916 Studies on the physiology of reproduction in the domestic fowl. XIV. The effect of feeding pituitary substance and corpus luteum substance on egg production and growth. Jour. Biol. Chem., vol. 24, pp. 123-135. (24) Peart, R. anp Miner, J. R. A table for estimating the probable signifi- cance of statistical constants. Me. Agr. Expt. Sta. Ann. Rept. 1914, pp. 85-88. (25) Peary, R. anp Parrerson, S. W. 1916 On the degree of inbreeding which exists in American Jersey cattle. Proc. Nat. Acad. Sci., vol. 2, pp. 58-61. 164 RAYMOND PEARL (26) Peart, R. anp Surrace, F. M. Appliances and methods for pedigree poul- try breeding. Me. Agr. Expt. Stat. Ann. Rept. for 1908, pp. 239-274. (27) 1908 Selection index numbers and their use in breeding. Amer. Nat. vol. 48, pp. 385-400. (28) 1910 On the inheritance of the barred color pattern in poultry. Arch. f. Entwickl. Bd. 30, pp. 45-61, Taf. II and ITI. (29) 1911 A biometrical study of egg production in the domestic fowl. II. Seasonal distribution of egg production. U. S. Dept. Agr. Bur. An. Ind. Bul. 110, pp. 81-170. (30) Peary, R., Surrace, F. M., anp Curtis, M. R. 1911 Diseases of poul- try. heir etiology, diagnosis, treatment, and prevention. New York (Macmillan), pp. xi + 342. (31) Pearson, K. 1911 The fight against tuberculosis and the death-rate from phthisis. Questions of the Day and Fray, No. 4. (32) Punnett, R. C. 1909 On the alleged influence of lecithin upon the de- termination of sex in rabbits. Proc. Cambridge Phil. Soc., vol. 15, pp. 92-93. (33) Russo, A. 1907 Modificazione sperimentali dell’elemento epitheliale dell’ovaia dei mammiferi. Atti Acc. dei Lincei, Cl. Sci. fis. ece., ser. 5a, vol. 6, pp. 313-384. (34) Srockarp, C. R. 1912 An experimental study of racial degeneration in mammals treated with alcohol. Arch. Intern. Med., vol. 10, pp. 369-398. (35) 1913 The effect on the offspring of intoxicating the male parent and the transmission of the defects to subsequnt generations. Amer. Nat., vol. 47, pp. 641-682. (36) 1914 A study of further generations of mammals from ancestors treated with alcohol. Proc. Soe. Exp. Med. Biol., vol. 11, pp. 136- 139. (37) StockarpD, C. R. anp Craic, D. M. 1912 An experimental study of the influence of alcohol on the germ cells and the developing embryos of mammals. Arch. f. Entwickl., Bd. 35, pp. 569-584. (88) SrockarD, C. R. anp Paranicotaou, G. 1916 A further analysis of the hereditary transmission of degeneracy and deformities by the de- scendants of alcoholized mammals. Amer. Nat., vol. 50, Part I, pp. 65-88, Part II, pp. 144-177. (39) Sumner, F. B. 1915 Some studies of environmental influence, heredity, correlation, and growth in the white mouse. Jour. Exp. Zodél., vol. 18, pp. 325-482. (40) Toppr, C. 1910 L’azione dell’alcool sullo sviluppo e sulla funzione dei testicolli. Riv. sper. Fren. e med., vol. 36, pp. 491-515. (41) Tower, W. L. 1910 The determination of dominance and the modifica- tion of behavior in alternative (Mendelian) inheritance, by condi- tions surrounding or incident upon the germ cells by fertilization. Biol. Bul., vol. 18, pp. 285-352. (42) Wetter, C. V. 1915 The blastophthoric effect of chronic lead poisoning. Jour. Med. Res., vol. 28, pp. 271-293. (43) Wuirney, D. D. 1916 The control of sex by food in five species of roti- fers. Jour. Exp. Zool., vol. 20, p. 253-296. THE EXPERIMENTAL MODIFICATION OF GERM CELLS Il. THE EFFECT UPON THE DOMESTIC FOWL OF THE DAILY IN- HALATION OF ETHYL ALCOHOL AND CERTAIN RELATED SUBSTANCES! RAYMOND PEARL Maine Agricultural Experiment Station FOUR FIGURES CONTENTS 1 LAL ANE 06 HONG MOY OVE REIS Ota Sc Og ee Ee ee cn OO. | ka ee ee ee 165 Melpee lorie latipes Sek ear acelin tas. .3.« « 5-0 hey RNRMES oreke osoee ore rece 166 URES OGL y sWCT OR Mera Sa ee clk t's Sic.. b a SIONS MOE to on oes Gene 171 Veer MTOdUetlonecn eres acto ke: Sale): osc 1 Se oo A SS Re bie 175 YL RSHDRAETCT A? SEIT ely Gr es a PS a eS 182 I. INTRODUCTION Before entering upon any discussion of the effect of the alcohol treatment on the progeny it seems desirable to examine w-th some care into the effects, both structural and physiological, upon the treated individuals themselves of the daily administration by the inhalation method as described -in I,? of ethyl alcohol, methyl alcohol, or ether. In this examination attention will be confined to characters which are capable of quantitative definition and measurement. It seems highly desirable in the .experimental study of a matter so warmly debated to deal chiefly with things which can be measured. A limitation of the present section of this report is found in the fact that the experiment is still in progress and only a small 1 Papers from the Biological Laboratory of the Maine Agricultural Experi- ment Station. No. 101. * This refers to the first paper in this series, which was entitled: ‘‘The experi- mental modification of germ cells. I. General plan of experiments with ethyl alcohol and certain related substances,’”’ Jour. Exp. Zodél., vol. 22, pp. 125-164. Throughout this and later papers in the series cross-references to other papers in the same series will be made simply by the Roman numeral designating the paper referred to, together with the particular page number to which reference is made. 165 166 RAYMOND PEARL amount of exact autopsy material from treated individuals has as yet come into hand. Later it will be possible to deal more exhaustively with organ weight data. It will be recalled that, as stated in I, the present report in- cludes only the data obtained from the beginning of the experi- ment in September, 1914, up to February 1, 1916. II. MORTALITY Inasmuch as many of the birds with which the experiment started are still alive it is obviously impossible at the present time to go into the question of the effect of the aleohol treatment upon the duration of life. What can be done, however, is to examine the facts as to the mortality of the treated and control birds during the first 15 months of the experiment. The data for such an examination have already been given in table 2 of I for the females only, and are summarized in convenient form in table 1 of the present paper. In explanation of the headings of table 1 it should be said that considerations of space on the poultry plant made it necessary to dispose of a part of the control birds (untreated) in the fall of 1915. The individuals so disposed of by sale were nearly a ran- dom sample from every point of view of the respective matings from which they came. They were certainly a random sample so far as concerns general bodily vigor and probable duration of life. From the standpoint of mortality figures the only thing which can be said of them, however, is that they are certainly’ known not to have died within the first 500 days of their lives. It should be further noted in regard to table 1 that in the case TABLE I Showing data regarding mortality to February 1, 1916. Females only NUMBER |. AT SOLD KILLED OR DIED LIVING NET CLASS 7 bana aes DIED BY OTHER- res. 1, | MORTALITY Spentie Ae ACCIDENT WISE 1916 PER CENT MENT Are ate xs acierscoe ovr 15 0 if 0 8 0 Untreated controls... 39 14 1 16 8 41.0 EFFECT OF ALCOHOL ON FOWL 167 of matings 1568, 1536 and 1575 only such untreated control birds are included in any column of the table as were on the plant at the beginning of this alcohol experiment. In other words, no birds from these matings which were sold in the fall of 1914, or died before that time, are here regarded as having been in the experiment, even as controls. For some other purposes these birds may be, and are used as controls. From this table it appears that: 1. Out of the 15 treated birds with which the experiment started the only ones which had died at the end of 15 months were those which were killed by an overdose of the reagents used (cf. I, p. 166). 2. Out of 24 (=39-15) untreated control sisters of the treated birds, which started in at the same time and have been kept on the plant until they died or to the present time, 16 have died, or 41.0 per cent of the whole number which started or 66.7 per cent of the 24 which were given every opportunity to live through the experiment if they were able to do so. It is obvious from these figures, if we take them at their face -value, that the mortality so far has been much heavier among the untreated control birds than among the treated. There is nothing which would gainsay such conclusion to be found from an examination of the causes of death of the 17 control birds which died. The pertinent autopsy data in this connection are given in table 2. The diagnoses are based on symptoms and lesions described in detail by Pearl, Surface and Curtis (80. This and other citation numbers in this paper refer to the bibli- ography at the end of I). From this table it is seen that out of the 16 deaths from non- accidental causes 9 were due to diphtheria or diphtheritic roup, either with or without other complications; 2 were due to visceral gout; 2 had their original causes in derangements of the oviduct; and finally diarrhea, pernicious anemia, and peritonitis, probably not of oviducal origin, each caused one death in the group. Roup has existed in endemic form on the Maine Station Poultry plant for many years, as on most other plants where for experi- mental, or any other purpose, birds are brought in from outside 168 RAYMOND PEARL TABLE 2 Causes of death in control females. Data to February 1, 1916 BIRD NO. AUTOPSY NO. CAUSE OF DEATH M487 1045 Diphtheritic patches in larynx and trachea. Pneumo- nia patches in lungs. 372 938 Visceral gout. 69 984 Peritonitis. Derangement of oviduct as initial cause. 1508 1031 Lungs congested. Diphtheria patches in throat. K42 969 Large concrement in oviduct. M36 991 Choked to death in trapnest. Accidental death. 1737 1177 Diphtheritic roup. 506 977 Digestive disorders. Diarrhea. 1744 1198 Diphtheritic roup. M24 998 Diphtheria. 1736 1203 Diphtheritic roup. 118 949 Pernicious anaemia. 440 1006 Roup—some peritonitis. 365 943 Visceral gout. 1724 1145 Diphtheritic roup. 1671 1121 Diphtheria. 1549 1037 Peritonitis. fairly frequently. Ordinarily it gives very little trouble. Oc- casionally it will break out into an epidemic of greater or less violence, always as a result of a relaxation of some routine sani- tary or hygienic measure. During the course of this alcohol experiment we have passed through a particularly violent epi- demic of the sort mentioned. This fact is reflected in the large proportion of the deaths due to diphtheritic roup or some of its complications. On account of this epidemic the total mortality in the experiment must be regarded as abnormally high. The remarkable thing is that during the 15 months covered in this report, i.e., to February 1, 1916, not a single one of the treated birds succumbed to this disease, though they were exactly as much exposed to contagion as the controls. This is a surprising result. It seems impossible that it can be due to any real in- crease in resisting power in the alcoholic birds. A _ possible explanation is that the daily inhalation treatment acts as a disinfectant of the air passages, and the treated birds do not take the disease because its germs are killed or greatly weakened be- fore thay have an opportunity to get an effective foothold. It EFFECT OF ALCOHOL ON FOWL 169 would be altogether premature to draw any conclusion in regard to the matter until more extended data are at hand. At present I desire merely to put on record the facts now available. During the period covered by the present report none of the male birds, either alcoholic or control, died. The superior mortality record of the treated birds, while a side issue to the main genetic interest of the study, has some interest on its own account in connection with the general prob- lem of the effects of aleohol upon the organism. There is a wide- spread popular opinion that life insurance statistics have ‘proved’ that even the most moderate use of alcohol definitely and meas- urably shortens human life. In common, as I suppose, with most persons who have made no special personal investigation of the original literature on the subject I had supposed this statement to be true. The present results were, however, so clear-cut in the opposite direction that my curiosity was aroused to examine critically the actuarial evidence. The results were somewhat astonishing. The evidence on which the current statements are based would not be accepted by anyone trained — in the critical valuation of statistical and biological evidence as ‘proving’ anything. All of the various actuarial investigations of the question which have been made, including Moore’s analy- sis of the experience of the United Kingdom Temperance and General Provident: Institution, MecClintock’s review of the ex- perience of the Mutual Life of New York, Phelps’ study of the experience of the Northwestern Mutual Life, and the widely quoted Medico-Actuarial Mortality Investigation, based on the mortality experience of 43 American life insurance companies, appear to suffer, in greater or less degree, from the following defects, which entirely invalidate them for the purpose of deter- mining critically and scientifically the effect of aleohol in differ- ent dosages upon human longevity: (1) The numbers dealt with are small. (2) There is no evidence of any sort or kind as to how much alcohol the subjects of the investigations consumed except their own statements on the subject made at the time insurance was applied for. (3) No allowance is, or can be, made for the influence of almost numberless other factors which may differentially influence the mortality in the groups com- 170 RAYMOND PEARL pared. (4) There is no control on the question of whether the drinking habits of the insured changed during the life of the policy. I am informed by competent actuarial experts that they regard the problem of the effect of the moderate use of alcohol (corresponding to the dosages employed in the present investigation) upon human longevity as still an absolutely open question. In this same connection the statements of Heron (10) regard- ing the death rate among extreme alcoholists is of interest. He made a very thorough and critical investigation of the mortality and morbidity of female inebriates, committed under the Ine- briates Act between January 1, 1907, and December 31, 1909, to the inebriate reformatories in England. He first shows in detail what must be evident on general grounds, that any person to come under the operations of the act and be committed to a reformatory must be chronically, extremely, and, as the event shows, practically incurably addicted to the regular and excessive use of aleohol. They represent the upper limit of chronic alco- holism. In regard to morbidity he finds (p. 17) that 77 per cent of these maximally alcoholic persons ‘‘are free from definite organic disease.” Regarding mortality he finds (p. 22) that the death-rate from all causes among inebriates while under sen- tence is only half that of the total female population of England and Wales and is less than a fourth of the death-rate of the class from which they are drawn, if the assumptions made in arriving at the death-rate among this class be accepted; the death-rate among inebri- ates from cancer is slightly less and from phthisis is decidedly less than in this class. The lower death-rate from phthisis is possibly due, to some extent at least, to selection before admission and close medical supervision after admission to the Reformatories. The official German statistics show in general a smaller death rate from tuberculosis among aleoholists than among abstainers. This result, which is similar to that obtained by Elderton and Pearson (7), is attributed by Pearson (31, p. 16) to selection, in the manner that individuals of better physical constitution are more likely to be drinkers. Such a factor as this would, of course, not come into consideration in the experiments with fowls at all. EFFECT OF ALCOHOL ON FOWL al Ill. BODY WEIGHT Stockard and Papanicolaou (38, p. 72) make the following statement regarding their guinea pigs: The general condition of the animals under the fume treatment is very good. They all continue to grow if the treatment is begun before _ reaching their full size, and become fat and vigorous, taking plenty of food and behaving in a typically normal manner. . . . . Aleco- holized animals are usually fat, but there is no fatty accumulation in the parenchyma of any of the organs. They give no exact data regarding body weights. In planning the experiments it was felt to be highly important to collect exact information on the changes in body weight which occurred. Clinical experience has abundantly demon- strated that body weight changes, in spite of their rough and inexact character, furnish an index of general metabolic con- ditions and changes which is by no means to be despised. Its value is greatly enhanced if parallel data from proper controls are also at hand. The treated birds in these experiments have been weighed at intervals of about one month, except that no records are at hand for October 1, 1915. Owing to an unfortu- nate misunderstanding the control birds were not weighed as frequently as the treated birds. The available data on body weight in grams are given in table 3, and figures 1 and 2. From table 3 and the diagrams the ieee points are to be noted: 1. The majority of the birds used in the experiment had plainly not completed their growth at the time the experiment started. This is shown by the fact that even with those birds whose treatment did not begin until December 1 or January 1 the same rise in body weight is shown in the earlier months as for those whose treatment began in November (cf. table 4 in I). That this initial rise in the body weight curves can not be due entirely to growth is proven, however, by the fact that 2 2 1572, 1573, 1574 and 1575 show it just as clearly as do any of the other birds. But these four birds were nearly a year 172 RAYMOND PEARL and a half old at the beginning of the experiment, and their increase in body weight at this time can not be attributed to growth in the ordinary sense of the term. Examination of rather extensive statistics on body weight changes in poultry indicates that there is some tendency for a bird to increase in weight 3000 2800 2600 2400 2200 NOV JAN MAR MAY JULY SEPT NOV JAN Fig. 1 Diagram showing the changes in body weight of females subject to inhalation treatment. The figures plotted are the means for the specified sorts of birds, except where a statement to the contrary is made. Solid line, un- treated controls; dash line, ethyl alcohol birds; dotted line, methyl alcohol birds (here the plottings after September 1915 are based on a single bird); dash-dot line, ether birds. during the autumn and early winter months, quite regardless of her age. I hope later to be able to present exact and compre- hensive data on the normal seasonal fluctuations in the body weight of hens. In the absence, at the present time, of such data, and because of the fact shown by the data given here that this initial rise in body weight in the autumn of 1915 is only slightly less in the controls than in the treated birds, it seems ‘OI6I ‘ez Arenuve UO OpBUl SSUIYSIOM UO Poseq SI UBOTH SIYT, » ‘CT6L ‘6 AleNUes UO OpvU SSUIYSIOM UO posed SI UBT SIU], ¢ se o1z | zg1z | 280 | 280 | OBE | FIST | SLOT | FZLT | OOST | OOST-| OOST | OTST | OL9T | OOFT tS Tey 890 1 170z | 9661 | SOGT | FIST | FZLT | FLT | S8GT-| TOET | TOET | OSET | OZET | OBST | OLGT | OZET “*yoyooye [Ayjour “E99 gsct | 26F1 | 90FT | 26FT | COST | OGFT | OTFT | OGTT | 026 “*Joyooys [Ayya “Zezeo om © &aeez_| zzz | L80z | 1FO% | OSET | OOST | SLOT | ZFST | SST | OE9T | OGOT | OG9T | OST | OT “-oyoore [Aqyo ‘F99 © = s3] 18968 8998 2988 HT On ls ae ee s]oajuoo po}vomun [jw Jo uBayy 9796 | O9LG | 89LG | GE69E | 9996 | LoS | SPAT IOY}e [[B JO Uva] OZTIEROTSESNOGEES |MOS6G MONEE O2Gel\e mus ote ee 1OY}9 “ELCT S OOTE | OFFE | OOSE | OBTE | 0662 | 2ZFS | 19y}9 ‘ZLCT 5 ose | OSES | C&ace | OSecn|Oacar| celal: seem one 1oy}o “T6FT Fe OPIS | OS, | Oles | |GoeG NOccou|ar (Sim |ene ee eee Layo ‘OGFT A ZOE | ZFS | 99zE | GATS | 222% | S8Gz | 6SES | 89ZZ | 89ZZ | OSS | OGEZ | OTE | OOLE | 069S | LATS | 1OY}O “CQFT = 9946 | 6698 | 6¢96 | SIee | S8TS | 8E98 | TI8@ | OS8e | O9Se | GEse | Sve |Spatq [AYJoUT [IB Jo UBITY jan} ee | | | | | S tzze | oete | sez | z1sz | ezzz | 2922 | Te9% | oFge | esez | Ogse | 866z | 026% | OE8% | OZ2z | Esez | “TOYOOTe TAYQoUur “cLcT a gesz | shez | ce2z | s9zz | ZEIZ | OF9Z% | O8S% | OFSS | 0262 | OZLz | OZ2z | “TOYOoTe [AYoUr “CEFT = | z18z | 192 | FOFS | ETE% | Zz2z | 098% | 0008 | 0Z6z | OS6z | OTOE | Ezze | “1OMOoT% [AYIOUT “Z8FT e) te9z | errs | BGez | ceTZ | ETez | OZzz | OsEz | OBEz | 069% | OG8z | BEEZ | “TOYOOTe [AYIouUR ‘OgFT | | | —__}+_ |__| 2° reese | sore | ogee | sase | ¢99¢ | or9e | tere | tore | sase | gers | Lede | 489% | 968s | S6Le | BEF | Sp4rq [AYIA [TB Jo UBoTY | ff es ogte | gere | FS0E | 99zE | EzTE | SO6S | ZZ2z | Z22z | 292% | OTSS | OSTE | OTZE | OGOE | OTE | SF6G | foyooTe [AYO “FLCT 191Z | 89%% | TE9% | GeFZ | TE9% | GFZ | ETES | GSES | 89ZZ | 09ZZ | OFLZ | 062Z | 0622 | OZZZ | FIST | foyooys® [Aqyo “6SFT SEZ | 89ZZ | 280Z | OST | 966T | FIST | FZLT | BOLT | FZLT | OSBT | O2OS | OSZZ | OGFE | OLEs | ETET | [Oyoo Te [AIO “FSFI F9cr | Z8OF | FL9E | BILE | H29E | ZOFE | SLTE | OkTE | Z90E | O8OE | OOZE | OG9T | OZSE | O8ZE | 4946 | JoyooTR [AYIo “ESFT cere | cere | ozTe | eoez | Fe6z | ZI8z | Gore | GFFZ | 6SEz | OZEZ | OFZZ | 098% | OFEZ | OL8Z | FOF |’ ~~ "TOYOOT® TAY 30 ‘ZSFT e6re | ZOE | F8OE | FE6% | ZZLZ | S6FS | LLTS | 966T | 280% | 000% | OLE% | O9ES | OT9S | OFFS | L80G | joyooye [AY}o “TSPT $a7DULA es a > et a > taj S ° re | 6B re 1G wf | ze | 2& | 2 2e =e =e me | 8 of rg ee | Se | se | se | se | SB | Se | oe | Se | SS | SS | Se | SB | SE | SF od < ar ee ae a x = = ag ad ae = INGWLVGUL ONY “ON Guid 2 a § 4 y a SALVA GaIWIoads LV SSVUD NI LHNIGM AGO a ee quawyzna4y panuywos yyw yybram fipog uz sabunya ayz burznoy 174 RAYMOND PEARL reasonable to conclude that the inhalation treatment had very little if anything to do with causing it. 2. The male birds show the same initial rise in body weight as do the females. Unfortunately we have no control figures for the males, but in all probability the same considerations hold for them as for the females just discussed. 2200 2000 1800 1400 JAN MAR MAY. JULY SEPT NOV JAN |Fig. 2 Diagram showing the changes in body weight of males subjected to the inhalation treatment. The lines have the same significance as in figure 1. All plottings are on the basis of single individuals not means. 3. Following the initial rise, which reaches its peak in Janu- ary or February, there is a sharp and prolonged fall in body weight which reaches its lowest point in either May or June, in the case of the females and a month or two months later in the case of the males 664 and 665. The absolute amount of this loss in body weight is large, particularly in the females. In it is probably to be seen the first significant effect of the con- tinued administration of the poisons. Just as before, it is im- EFFECT OF ALCOHOL ON FOWL 175 possible to measure exactly the portion of this effect resulting from the treatment because of the absence of weighings of un- treated controls during this critical period. There is probably normally some decline in body weight between January and June. It is, however, not normally as great as that here ob- served in the treated birds. 4. Beginning in June, 1915 in the case of the ethyl alcohol birds, and in July, 1915 in the case of the methyl birds, there has been a steady increase in the mean weight of the treated birds up to and including February 1, 1916. At the latter date the treated averaged about 300 grams more per bird in body weight than the controls. Put in another way, the alco- hol birds were 9.9 per cent heavier after about 15 months of inhalation treatment than untreated control birds of the same average age. This is a sensible, though not very great differ- ence. It certainly does not indicate that any profound or far- reaching effect upon the general metabolic processes of these birds has as yet been produced by the treatment. 5. In general the changes in the body weight shown by treated Barred Plymouth Rock females are paralleled in the treated Black Hamburg males. The chief difference is that the changes are absolutely somewhat smaller in the males. The further course of the body weight changes in these birds will be watched with great interest. In such birds as have so far come to autopsy there has been no indication of fat infiltra- tion of any of the visceral organs. Apparently the 10 per cent increase in weight is due entirely to deposition of body fat. IV. EGG PRODUCTION In the egg production of fowls we are dealing directly with an easily measurable activity of the very organ whose products we hope to influence, namely the ovary. On this account it is of particular importance to examine carefully the facts regard- Ing this character in the alcoholized birds as compared with their untreated sisters. The present experiments may be re- garded as especially favorable for such a study, inasmuch as the Barred Plymouth Rock females used in the work had been 176 RAYMOND PEARL pedigree bred for a number of years and their hereditary qualifi- cations in respect of egg production were well known (Pearl, 18). I have lately called attention (28, p. 1384) to some of the diffi- culties which are involved in interpreting critically and fairly the results of physiological experiments in which egg production is used as an indicator. One must have proper controls in the first place. The ideal condition in respect of controls is only reached when one has full sisters of the experimentally treated birds. This condition has been met in the present experiments. In the second place the management of the birds in the experi- ment should be such that one can be sure that the control birds are laying normally throughout the experiment. By normal laying in this connection is meant the full somatic expression of the innate inherited capacity of the birds for egg production. Unless this is reached by the controls in any physiological experiment on egg production one never can be quite sure that any difference which may appear between the control and experimentally treated birds is not due to the effect of some overlooked environmental factor upon the controls, which re- duced their production below what it should have been. | The egg production of all the birds in these experiments, by months, is given in table 4. In this table ‘D’ means that the bird in question died in the indicated month after laying the number of eggs in that month shown by the figures preceding the D. From this table it is evident that all of the birds, both treated and untreated, have laid normally throughout the experiment. To get beyond this general impression and make exact compari- sons between controls and treated birds it is necessary to reduce the mass of figures of table 4 to means or averages. This has been done in table 5. In calculating the means in table 5 only such birds have been included as began their first laying year at the beginning of the experiment, namely the autumn of 1914. Those birds which were yearling hens at that time (treated birds numbered in the 1500’s and their sisters) were entering their second laying year. First year and second year laying records are not homogeneously comparable, and therefore should TUE EFFECT OF ALCOHOL ON FOWL 996 9OLT I¥G q TIVLOL &] Spi a) ~ Arenues 9T6I peoqd | 0 0 0 0 |°° °°’ peyverqun ‘spy peed | 2T OT ST I 0 G Gj el Ft fe peyBorqun “o8OeT 010 |0 0 é 4] dl 1é 9 ST 9 0 0 0 Oe ee ae hype “ef LE] PICS | FI ‘4 FI I ST ¥G IZ Go ST 0 Os eee peyBorqun ‘Z6Z PIOS | FI GS CT Mi 1Z 8% 6 0 0 0 Oi eae es peyworjzun ‘967% OF 19" W412.) 61 SI CT II 1Z ee 1Z OT CT Sj 0 0 |'°°°°* payeergun ‘gezT PICS | il OT SI LT GG ZG 0 0 0 0 0 | °° ' paezeerzun ‘oFE PIOS | 12 0z LT 6 0z 1Z SI G 8 OZ 2S | sexes poeyeorjun ‘Tee 7 NE Lee ese a His él 8] I] Dit 8] val 0 ORNS a liek wet toes phiyia “6gtT prog | 0 g AL qe eee “poeyeortqun ‘69 NE eer | 108 1é 1é lé 06 6] Vai II Or 91 8) a RW ca ca phyia “¥EyT Gan y ee |Z, ial Mi LT 61 CT II ST F 61 i 0 Qe eae poyeoryun “TFLT Ganeor 90 9 8 OT 61 a 06 61 61 ST II OR WN Ny! Sees en LEC S| Cd61 ZG CG PI 61 OT OL | eo Mc, ‘popeorjun “2ep PICS | LT 61 0z 61 6 1Z ijl ST SI t% | T |''° °° payvomun ‘FogIy Gaalean 0 9 & 9% 0z 9% GG ¥Z 1Z €% 1Z OT SON eee poyeorjun ‘97 GAO! Nag él 7] 06 1é be 1é 61 0 9 1é 8 DONE! RH II G ABS EARCE | 0 61 6] 61 ee ral &é vine ras Le 8 SA ICED SIH EI Bike Whe) (Sy Merle ell Sel ee A ae ieee Wace alee aamee ile geal g s 4 z a 5 act if ey 5 INGWLVaNL GNV ‘ON aula CI6I FIGI splig jo1juod pun papas fo uoyonpoid bba hyyyuopy % ATAVL THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. l RAYMOND PEARL 178 d2Z1 1G 9 OL 61 1 Q | 0 °** pezeorzun “OfP pod Cc 6 Dae ew “poe7ywoatyzUuNn IL cst | 0 ) Ne | Cs I We SZ II £G 8% 61 OI II 0 0 Z (Oiel iaacee poyworun “QELT ad¢ G OT ST ) Ol | 1% | °° °° (pepeerqzun “FZIN Ost | O OD VO. | o I g OL 8 Sé GT va! II val! G] ae || OOo el Pm 19Y]9 “EST eG 0 WO No 0 yi 9] bik LI 61 g] 6 0 0 0 QO | phyzaue oS GL eT OS 0) OS OGel mec GG ‘all yk ST féil 1G ik ¢ 0 0) “ peyverqzUNn “PPLT PIOS |. 1Z 1G IZ LT SI GI el G 0 (Cr a ie ‘popworqun “6TG PIOS | I ST 8 yl 9 6I SI 0G 61 LE Oe |e 2 = peteort OMneGy. peod | 7] 8 jij fey él 06 Z é Ty (7 | te erg ee jhiyjaw “e6y 1 dtl GT 8 (Ohi 2 anurans « poywotun “QOC PIOS | 61 GG SI (Kb IZ FI 6 ) 0 0 0 | °° (pazeerzun ‘OZ SIG erie | Fe nO 29 CT IZ GI SI ¥G GG CG GG 6 3 1 eo | Oe hae pozyeotzun “ESgT PIOS | 0 0 6 ral ¢ SI OT F1 LT 1 (res aay popBorzun “C9 esI | ¢ ee Oma le, ral CT ST GI ral LT SI pit 1Z OT 0 Oren ees ‘ popvorjUN ‘ZPLT peed | or 8] 6] lé 8] lé 6] jy TT OLA ye eS hy ee piyjou “{97T PEOCEROV a UD eee ee SI GG SI 1Z 61 NG GI 0% I 0 OEM eae poyvorjun “LEL1 PISS | -§ ¢ OT i) G6 8% Il 0 0 0 (Oia ame pozyBorjUN “QPE d6 ST 1T OL Coy ANNA tl eects poyworjzun “OE{y pvod | él 8 él gl dl 1é 06 1é 91 0 QO | phyrew “os7T 2 o > 8 Bs = | 4 o o 5 INANLVGUL GNV ‘ON GUled eg |g g : 5 5 OST6L C161 PI6L ponunuo)—F WTEaVL eS) ‘TjozZ oq ‘Jo ‘szorpnd jo ssoyy YIM ATJOodIp o[qvaedulod you o1e Splodal MOY], ‘FIGL 10qo0JQ Ul IWoOA SutAvy puovos Aoay} uodn SurtojzUd o19M SpaAIG oso} VY} PoloequUloulol OG P[NOYS IT ¢ EFFECT OF ALCOHOL ON FOWL peod | 2 il ral 8 G GI Gl non © pozyvotquN “GFT I6. | 0 OW ese Si 61 FI ST 1G 6I FI " IT ¢ 6 LT | (°° poyvorqun , “OFST PIOS | O 0 eT ) 6 CT FI 6 0 0 eT |" *pegvorqun “7 T¢T 6 SI 8 0 0 0 0) se aaa 19479 6 GLGT do! II 0 0 0 0 (ate ae ae 19Y79 o CLG] PICS | FI [Ie 0G GE 1Z 0G 0% SI] I pera ike | poyvotjun “gTe C1Z Ge We We Oe Te ST 1Z Go 1G a 8G 6 LI SI 8 0 “- paywatqun ‘OFT DES A ON M Ie vee ‘re 8% Ge #G VG &Z vi OT SI Ae alee lter ae poywoljun “TLOT PIOS | T 8 SI GG GG 8% ST 61 SI Ce Gahan a ane po}vorjun “Gg PICS | OT OI at 6I SI SI OL 6 0 E Glee: poyBoun ‘99 pred |0 |0 |6 0% 1G GG ral 9% cS 9G 1Z 1Z 0 QI | & | °° poxvorgun “FZZT : él é | 06 | AT a RE | a |e ea ees 9y79 “T67T PBICT @) oR te Poy Bolu ‘egg PIOS | II FI We ral FI rail ST GI iG 9 On (atest: pozBorqun ‘GCE ST | 0 OM OL al |" OT ) 1G ill OT 0c SI SI 0 G Ola aamee poyworgun “EpLT C6 $8 rae fey g 0 OS SRT AUTOR OOHT 180 RAYMOND PEARL not be lumped together in calculating means or other constants. So few of the yearling hens with which the experiment started lived through that there is not enough material to make any critical detailed study of their production. Accordingly we may confine our attention to the laying activity of the pullets with which the experiment started. In the first column of table 5 are given the monthly mean productions for all treated birds surviving through the indicated TABLE 5 Mean monthly egg production of treated and control birds. MONTH TREATED N Se aoe N Betta N November, 1914...... Biss 11 6.91 33 6.91 33 Decembers-a-- ees. T4.91 Tet 7.91 32 7.91 32 Janurcimyael Oller ee 12.36 11 12.50 32 12.50 32 ebruanynaee a oo. 14.00 11 14.63 30 14.63 30 Mianchinreaeemen isc 19.54 11 19.48 29 19.54 29 ATI lre eee deem ete 16.44 9 18.15 i 7526 19 INT aS Aree aetna case 19.44 9 17.74 Di 16.79 19 Ab Osoidactemrclatae aes ee 16.89 9 17.08 26 16.17 18 Dl Seer re ae ts 13.78 9 16.35 26 16.28 18 MMU ccoodsdandadenll | Lleske2 9 15.50 26 16.06 18 September........... 12-38 6 15.33 12 15.00 4 Octobersecres.o. yas 9.00 6 13.25 12 12.00 4 INovember®..2%....:. 8.67 6 6.58 12 8.00 4 December. 24... .5... Bs seio) 6 1.00 12 1.75 4 JanuanyeelOlG yess... 6.38 6 2/38 9 0 4 MO tales scent exter.) i 183.97 184.74 180.80 month, beginning with November, 1914, and continuing through January, 1916. No bird which died in the course of the experi- ment is included in the calculation of the mean production for the month in which she died. In the third column of the table, headed ‘General Control,’ are given the monthly mean produc- tions for all untreated control birds surviving through the indi- eated month. This column represents the distribution of the production of the general control flock, regardless of whether the treated sisters of any of these control birds had died. In the fifth column of the table, headed ‘Special Control,’ are given EFFECT OF ALCOHOL ON FOWL 181 the mean monthly productions for the full sisters, and only those of the treated birds used to calculate the mean for any month in column 1. In other words, when the ether birds, for example, dropped out by death in April, 1915, all of the untreated control sisters of the ether birds killed were also dropped in calculating the means for the ‘Special Control’ column. This column gives a critically fair comparison with the treated birds MEAN EGG PRODUCTION NOV DEC JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT NOV DEC JAN Fig. 3 Showing the mean monthly egg production during the first 15 months of the experiment. Solid line, untreated controls (data from ‘Special Control’ column of table 5); dash line, all treated birds; dot line, old general flock data (see text). at any stage of the experiments. The columns headed ‘N’ give in each case the number of surviving birds on which the calcula- tion of the mean was based. The data of table 5 are shown graphically in figure 3. In this diagram there has been added the curve of mean monthly pro- duction of the Maine Station Barred Rock stock in earlier years, as given by Pearl and Surface (29, p. 89). This shows clearly enough that the laying of the birds in the experiment as a whole has been excellent. 182 RAYMOND PEARL From these data the following points appear to be clearly established: 1. The egg production of the treated birds and the untreated controls was entirely normal in respect of its seasonal distribu- tion, as well as in regard to its amount. 2. There has been no significant difference in the egg produc- tion of the treated birds and their untreated control sisters, either in the total average number of eggs produced per bird, nor in the seasonal distribution of this production. Taking the whole untreated flock, the mean production per bird in the 15 months was 184.74 eggs, while the mean production for the treated birds was 183.97, making a difference of 0.77 egg in favor of the untreated. Taking the ‘Special Control’ mean of 180.80 eggs there is a difference between this and the treated of 3.17 eggs in favor of the treated. Obviously the only con- clusion which can be drawn from these insignificant differences is that the inhalation treatment has not affected the egg produc- tion of the birds, either favorably or adversely. 3. During the months of July, 1915, to October, 1915, inclu- sive the mean production of the treated birds falls below that of their control sisters. The difference between the two curves in this region is no greater than may at any time occur between two similarly managed groups of sisters, according to the writer’s experience with egg records. There appears to be no reason to attach any significance to this dip of the treated below the control curve. Taking the whole period covered by the diagram it is clear that the two curves wind about one another, now one, now the other being uppermost, just as curves for two random samples of the same material would be expected to do. V. SUMMARY A summary of the numerical data regarding the effect of the treatment upon the treated birds themselves is given in table 6. In this table the superior result is printed in bold faced type. In the last column of each table a plus sign denotes that, with reference to the particular character discussed, the alco- EFFECT OF ALCOHOL ON FOWL 183 TABLE 6 Showing in summary form the effect of continwed administration of alcohol (ethyl and methyl) and ether, by the inhalation method, upon the treated individuals themselves. NET RESULT ON ALCOHOLISTS | TREATED UNTREATED CHARACTER OR QUALITY STUDIED | : : | INDIVIDUALS CONTROLS 1. Mean number, per bird of consecutive | daysior, Greatment-eme ners... ase! 344.2 0) 2. Net percentage mortality (to February| 1, 1916) exclusive of birds acciden-| | eellbsse IMME; Goo aoe koe s Ob ee eee | 0 41.0 sin 3. Mean body weight of females (in gms. | 3266 2953 — 4. Mean egg production per bird, 14 | IaaKOSONG AVS Se Alo caScre Res Oise choke Rake ie me 183.97 180.80 0 De dGrenera lea Chivl ty sass te efi 0 ol ae oer Reduced Normal = OmoemualvactivitivermeuLe seniinc ocee ae Reduced Normal — holists®’ have been favorably affected; a minus sign that they have been unfavorably affected as compared with untreated eontrols. A zero indicates that no effect of the treatment, one way or the other, has been detected. From these summarized data it is possible to gain a tolerably clear comprehension of the objective happenings in these experi- ments so far. The treated animals themselves are not con- spicuously worse or better than their untreated control sisters or brothers. The survivors, i.e., those not killed by accident, after roughly a year and a half’ of daily treatment, are becoming a bit too fat for their best physiological economy, but except for that point, and the reduced activity which goes with it, they are very much like normal fowls. Their apparently much bet- ter mortality record is indeed conspicuous, but in view of the small numbers involved, no great significance can be attached to it at present. It is probable that as the experiments prceceed this superiority in relative mortality will be considerably dimin- ished. However, as has already been pointed out, the effect of 6 T adopt this convenient noun from Pearson to denote individuals subjected to the influence of alcohol. 7It should be noted that the mean of item 1 in Table 6 is greatly reduced by the fact that all birds, including the methyls and ethers killed in the tanks, are used in its calculation. RAYMOND PEARL C D Fig. 4 Photographs of alecoholized Barred Plymouth Rock hens and their untreated control sisters. A, ethyl treated 9 No. 1481; B, untreated control Q No. 1726, sister of 1481; C, ethyl treated 9 No. 1489; D, untreated control 2 No. 1738, sister of 1489. Scale of reduction in the negative the same for all. Photographed February 18, 1916. EFFECT OF ALCOHOL ON FOWL 185 chronic alcoholization upon the duration of life has by no means been well established. There is a widely prevailing popular opinion that even the very moderate use of alcohol shortens life. As is pointed out in the body of this paper, there seems to be no critical evidence as yet that such is in fact the case. The data to this effect which are usually cited are found upon exami- nation not to be critical. As experimental investigations like the present one, and Stockard’s with guinea pigs, go on, some rather definite and critical evidence should accumulate regarding this point. It seems desirable to show by actual photographs that after some 15 months of daily alcohol treatment there is very little visible difference between the treated birds and their untreated sisters. To this end figure 4 has been prepared. This shows two pairs of treated and untreated sisters. A and C are ethyl treated birds 1481 and 1489; B and D are their untreated sisters 1726 and 17388. All pictures are to the same scale and were taken the same day, February 18, 1916. It is evident that the alcoholics are in no wise essentially different in appearance from the untreated. The pose of D (Bird No. 1738) is bad; she is really just as sprighly and active a bird as any of the others. Summarily stated the essential results of this paper are: 1 The mortality among the treated birds was much smaller than among their untreated control sisters After 15 months of treatment the difference was 41 per cent in favor of the treated birds. 2. The body weight changes in the treated birds were as follows: immediately following the starting of treatment, which was in the autumn, there was an increase in mean body weight, probably in no way due to the treatment. Following this initial rise, which reached its peak in January or February, there was a sharp and prolonged fall in mean body weight which reached its lowest point in May or June. Beginning in June or July there was a steady increase in mean body weight continuing without break until the end of the period covered in this report (February 1, 1916). At the date mentioned the treated birds 186 RAYMOND PEARL were on the average 9.9 per cent heavier than their untreated sisters. 3. Neither the total amount nor the distribution of egg pro- duction were significantly different in the treated birds from what they were in the controls. Both treated and control birds laid normally and well. A REEXAMINATION OF THE APPLICABILITY OF THE BUNSEN-ROSCOE LAW TO THE PHENOMENA OF ANIMAL HELIOTROPISM JACQUES LOEB AND HARDOLPH WASTENEYS Rockefeller Institute for Medical Research, New York It has been shown by a number of botanists that the helio- tropic reactions of plants obey the Bunsen-Roscoe law whereby the heliotropic effect is determined by the product of the inten- sity into the duration of illumination. The reactions of free swimming animals to light are generally too quick to permit an examination of the validity of this law, but Ewald! has shown that if the efficiency of intermittent and constant light is com- pared in such forms it is found that both have equal efficiency when the product of duration into intensity of illumination is equal in both eases (Talbot’s law.) This proof is in reality also a proot of the fact that the heliotropic reaction is determined by the product of intensity into duration of illumination. It should also be mentioned that the striking results of Bradley Patten? on the direction of movements of the negatively heliotropic larvae of the blowfly under the influence of two lights of unequal intensity strongly suggest the validity of some such law as that of Bunsen and Roscoe for heliotropic reactions. It is, however, desirable to have a direct proof for the appli- cability of this law to heliotropic reactions of animals. For such a proof we are compelled to turn to sessile animals. Loeb and Ewald* have made some preliminary experiments on the hydroid Eudendrium which is positively heliotropic and their observa- tions agreed with the Bunsen and Roscoe law. The number of 1 Ewald, W. F., Science, 1913, 38, 236. 2 Patten, B., Am. Jour. Physiol., 1915, 38, 313. 3 Loeb, J., and Ewald, W. F., Zentralbl. f. Physiol., 1914, 27, 1165. 187 188 JACQUES LOEB AND HARDOLPH WASTENEYS observations was limited and it seemed desirable to continue these experiments, since it is of fundamental importance to know whether or not apparently purposeful instinctive reactions such as the tropisms can be expressed in terms of purely physico- chemical laws. The method followed in the work of Loeb and Ewald consisted in ascertaining the time required to cause 50 per cent of the polyps of Eudendrium to bend towards the light, and it was found that this time varied inversely with the square of the dis- tance of the light from the animals. We found that this method could not be followed with satisfaction on account of the great variation in the quality of the material from day to day. We therefore selected another method. We confined our ex- periments to three intensities of light by putting the specimens at distances of 25, 37.5, and 50 em. from a Mazda incandescent lamp of about 33 Hefner candles. The times of exposures were adjusted so that on the assumption of the applicability of the Bunsen-Roscoe law the same effect, i.e., the same percentage of polyps bending towards the light should be produced. Thus in some experiments the exposure for the three distances given was 10, 22.5 and 40 minutes respectively, in others, 7, 15.75, and 28 minutes, and so on. The ratios of the percentage of polyps bending towards the light for the three distances should be as 1:1:1. Since the material differed widely in different experi- ments and in different dishes, it was necessary to compute the averages of a large number of experiments. The source of light was, as stated, a Mazda incandescent electric light of about 33 Hefner candles. This was screened with a series of black screens having circular openings of about 7.5 cm. in diameter to prevent reflection of stray light. The colonies, immersed in sea water, were arranged in a row in rectangular glass dishes, the stems being inserted in holes made in a layer of paraffin mixed with lamp black as in the previous experiments. The rear side of the dish was also coated with the paraffin lamp black mixture in order to prevent reflec- tion of light from the slightly uneven back surface of the dish. BUNSEN-ROSCOE LAW—ANIMAL HELIOTROPISM 189 The previous treatment of the colonies was similar to that used in the experiments of Loeb and Ewald. After cutting off the existing polyps the stems were allowed to le for about twenty- eight hours exposed to diffused light. They were then placed in the dark for from eighteen to twenty hours, by which time the hydranths had usually regenerated. The exposure was then made as soon as possible. In all 3873 hydranths were used or an average of 82 to a dish. Of these 1671 were available for the determinations, i.e., they bent to the light as a result of a short or prolonged exposure. The remaining 2202 hydranths were not available for the deter- minations, being either refractory or being originally placed parallel to the direction of the rays of light, facing either to the back or front. At the beginning of an experiment the hydranths bent towards the light were counted, as well as the total number present in the dish. After the exposure the dishes were allowed to remain in the dark for from two to three hours, during which time the heliotropie bending occurred.!' They were then replaced in hight in the same relative position as during the exposure and the polyps turned or bent towards the light were counted. The difference between this number and the number originally turned to the front gave the number of hydranths caused to bend by the exposure. The dish was then allowed to remain in the light for from two to three hours longer and the polyps turned to the light at the end of this period were once more counted. This value less the number turned to the front before the original exposure was taken as 100 per cent in computing the percentage which had been caused to bend by the initial exposure. The following example will indicate the method of calculation and also prove the fact that the number of polys which bend to the light increases with the duration of exposure. 4The very fact that the bending occurred in the dark and not while the organisms were exposed to the light should in itself suffice to prove the untena- bility of the anthropomorphic explanations of heliotropic reactions by “‘trial and error’ or by hypothetical sensations of brightness. 190 JACQUES LOEB AND HARDOLPH WASTENEYS 33 Hefner candle lamp. 50cm. distance. July 15, 1916 20 minutes exposure. Number of total hydranths = 58 At start bent backwards = 8| _ ,, At start bent forwards =8f Hence 42 were apparently available for the experiment. At the end (after 20 minutes exposure and 2 hours in dark) bent forwards 13, i.e. 13 — 8 = 5 actu- ally bent to light; hence =; = 11.9 per cent reacted of apparently available num- ber of polyps. After long exposure® to same light, 41 bent forwards, showing that 33 were actually available for experiment. Hence in 20 minutes =; = 15.1 per cent bent to light. 40 minutes exposure. Total number of hydranths = 87 At start bent forwards =15| _ ,, At start bent backwards = 13/ - Hence 59 were apparently available. At the end of experiment, bent forward 38; hence 3S — 15= 23 bent under influence of light, i.e., 23 = 39 per cent of apparently available. After long exposure to same light, 64 bent forward. 64 — 15= 49 actually available. Hence in 40 minutes 73 = 47 per cent bent to light. Fifty minutes exposure gave the same result. This shows that with the time of exposure the percentage of bending polyps increases until finally all the available polyps bend. The example chosen also shows that it is sometimes impossible to obtain 50 per cent of polyps to bend. The varia- ation in the condition of material makes a large number of experiments necessary. In our experiments we exposed in the majority of cases only 28 minutes at a distance of 50 em., in order to avoid the possi- bility that all the available polyps underwent bending; since in this case the observed ratio of 1:1 :1 for the three distances would have been meaningless. In all the experiments the num- ber of polyps which bent was always smaller, and generally considerably smaller, than the number of available polyps. The following table gives a summary of the results. The first three columns give the times of exposure for the three distances of the source of light, selected, as stated, on the assumption that ® Usually from two to three hours. BUNSEN-ROSCOE LAW—ANIMAL HELIOTROPISM 191 TABLE 1 TIMES OF EXPOSURE IN MINUTES OO ees oui BENDING 25 cm. 37.5 em. 50 em. 25 em. :37.5em.) 25 em. : 50 cm. |37.5 em. : 50cm. 15 60 1.50 20 | 80 1.30 10 22. 40 20) (3.08) (2.56) 10 22.5 40 0.94 1.47 1.55 10 22.5 40 1 OG (2.30) (2.48) 10 22.5 40 1.43 1.04 0.94 10 22).9 40 0.76 1.09 1,47 10 22.5 40 1.05 1.15 0.90 0.96 10 22.5 40 1.15 0.99 7 15.795 28 0.84 0.62 0.74 7 15.75 28 1.70 0.49 0.58 7 G73) 28 0.85 1.25 1.35 a 15.75 28 (2.09)! 0.99 1.08 7 15.75 28 1.14 1.15 0.55 i NG, 0) 28 0.44 0,92 0.44 ai 15.75 28 1.92 0.80 0.61 ri 15.75 28 0.59 0 356 0.70 7 15.75 28 0.48 1.07 0.31 7 15.75 28 1.00 0.48 1.80 G 15.75 28 0.69 1.09 0.81 7 15.75 28 1.26 0.85 1.09 i 15.75 28 0.86 1.38 0.85 7 15.75 28 0.70 LOR 1.59 7 15s 75 28 OR ai 1.24 7 15.75 28 0.60 IN EER TRA ie oO Acs nocd oe aS Re 1.02 0.99 1.02 PROM ADIGVEETOR pete Geet etal. +(0.01 +0.01 +0.01 ' Bracketed values being extreme variates are excluded from calculations of the means and probable errors. the Bunson-Roscoe law holds. On that assumption the ratio of percentage bent in any two or all three dishes on any one day should equal 1.0. These ratios for each pair of distances of the source of light are given in the three other columns of the table. The percentage bending was only compared in dishes contain- ing material regenerated and exposed on any one day, since only 192 JACQUES LOEB AND HARDOLPH WASTENEYS in this case was there any likelihood that the material was in any way uniform and since only in this case the experiments were carried on at the same temperature and the same conditions of regeneration. The result was that the observed ratios were as 1.02 : 0.99 : 1.02 (with a probable error of + 0.01) while the values calculated on the assumption of the validity of the Bunsen-Roscoe law were as 1:1:1; 1e., the results showed as great an approximation between observed and calculated values as one could expect. These experiments carried on by a somewhat different method from those previously published by Loeb and Ewald harmonize with the idea that the Bunsen-Roscoe law is the correct expres- sion of the influence of light upon the heliotropic reactions of Eudendrium. CONTRIBUTIONS FROM THE ZOOLOGICAL LABORATORY OF THE MUSEUM OF COMPARATIVE ZOSLOGY AT HARVARD COLLEGE. NO. 290, ACTINIAN BEHAVIOR G. H. PARKER 1. INTRODUCTION The behavior of actinians has been interpreted in the past in many different ways and the subject even now is open to the ereatest uncertainty. Gosse (’60, p. 81), one of the most enthu- siastic and industrious students of these animals, after watching the creeping of Sagartia pallida, wrote that ‘it was impossible to witness the methodical regularity of the process, and the fit- ness of the mode for attaining the end, without being assured of the existence of both consciousness and will in this low ani- mal form.” But such naturalists as Gosse had been schooled to regard adaptations as necessary evidence of intelligence and it was only gradually that these workers were brought to see in Darwin’s natural selection one means at least of explaining adaptations without recourse to such a factor. So far has this mechanistic movement gone in the explanation of animal reac- tions and so vigorously have such workers as Loeb (’99) applied its principles that Baglioni (’13) in his general account of the activities of actinians felt called upon to argue at length for the presence even of nervous action in these forms. It is not my purpose to discuss the question of ‘consciousness and will’ of the existence of which in these lowly creatures Gosse was so firmly convinced. The futility of such a procedure is too evi- dent. But it is planned to examine some of the more complex activities of these forms with the view of gaining a clearer insight into their elements and into the relation of these elements to the animal as a whole. The activities of almost every species of organism are direcied now into one, now into another of three principal channels; these are, first, the great array of protective measures against unfa- 193 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 2 FEBRUARY, 1917 194 G. H. PARKER vorable features in the environment, next, the maintenance of a normal metabolism, and, last, reproduction. With this final category we shall have nothing in particular to do; under the second we shall take up the matter of feeding, and under the first that of general retraction and expansion. The elements that are combined in most of these processes have been discussed elsewhere. Actinians such as Metridium possess at least four systems of effectors; slime glands, cilia, nematocysts, and muscles, of which only the last gives evi- dence of being under nervous influence, and even among these certain muscles are very probably independent effectors. By means of combinations of these elements, the various acts in the appropriation of food, and in retraction and expansion are accomplished. These general activities will be considered on the following pages in the order named. 2. APPROPRIATION OF FOOD The appropriation of food is an activity with which the oral dise of actinians is principally concerned. The movements of the tentacles, mouth, and other such parts by which food is ingested were ascribed by Nagel (92, 94) to muscular action alone, but Loeb (95) pointed out that cilia also play an impor- tant rdle. The parts that are immediately concerned in the appropriation of food are the five following: the tentacular gland cells, whose secretions render the tentacles adhesive whereby pieces of food become attached to them; the musculature of the tentacles, by which these organs are pointed toward the mouth; the tentacular cilia, which sweep toward the ends of the tenta- cles and thus deliver the food to the mouth when the tentacles are pointed in that direction; the transverse muscles of the com- plete mesenteries, by which the esophagus is opened; and the cilia of the lips and esophagus, which in the presence of food reverse their usual outward stroke and thus transport such materials to the gastro-vascular cavity. Beside these five sets of parts some actinians include in the means by which they appropriate their food a sixth system, namely, the musculature of the oral disc. In Stoichactis, for instance, as described by ACTINIAN BEHAVIOR 195 Jennings (’05, p. 449) and in Cribrina as reported on by Gee (13, p. 814), the mouth during feeding is moved by the oral musculature toward the food-bearing tentacles, a shifting which has also been observed in certain corals (Carpenter, 710). This operation, though it can be seen to occur in Metridium, is rela- tively so insignificant in this form that it may be passed over without comment; the important elements in the feeding of this actinian are the five already mentioned. Much confusion and uncertainty exists in the various accounts of the methods by which actinians obtain their food and more or less of this is due to the failure on the part of writers to desig- nate the particular form of activity that they are for the moment discussing. Thus both ciliary and muscular activity are involved in the appropriation of food and have often been indiscriminately dealt with in accounts of this operation. Their significance for the animal as a whole is, however, very different and it is, there- fore, highly desirable that they should be kept clearly in mind as separate processes in any discussion in which they are in- volved. Of the five principal events that go to make up the act of food appropriation, three exhibit so little variation that they may be regarded as essentially uniform. ‘These are the secre- tion of mucus, the beat of the tentacular cilia, and the opening of the esophagus. In none of these are there during feeding any | important readjustments which are essential to the acquisition of food; the production of mucus is apparently a strictly local | response to a local stimulus; the beat of the tentacular cilia is constant and irreversible; and the opening of the esophagus is as simple and mechanical a reflex as could well be imagined. The idea that the esophagus, as often intimated, exhibits peri- stalsis is probably incorrect. At least a careful inspection of this organ in action in Metridium gives no support to this idea. The two remaining events in the appropriation of food, the responses of the oral cilia and the movements of the tentacles, are both open to significant changes and are of the utmost importance in judging of the relation of this process to the actinian as a whole. — 196 G. H. PARKER Unlike the tentacular cilia, the oral cilia, those of the lips and the esophagus, may reverse the direction of their stroke so that the usual outward current can be converted into an inward one. This reversal is under ordinary circumstances a local response on the part of the cilia to certain dissolved substances in the food. Its relative independence of the other activities of Me- tridium can be shown in a number of ways. Thus, though it is a response to food, excessive feeding has no marked influence on it. Allabach (05, p. 38) caused a Metridium to gorge itself with food, a process which can result finaily in disgorgement, and yet immediately after the animal had emptied itself, its oral cilia were found to reverse to food, which was thus passed down its esophagus. My own observations confirm this state- ment. Further if pieces of meat are fed to the lips of the oral half of a Metridium cut transversely in two, the cilia reverse and the masses of food thus carried through the esophagus are discharged at its open pedal end. By this means in the course of an hour or so I have passed through the esophagus of a Metridium many times the amount of food that its body could have contained, and yet the ciliary reversal was as effective after this period of continuous feeding as before. Other evidence of the relative independence of the oral cilia as compared with other effectors is well seen in specimens of Metridium that have been narcotized with chloretone, by which all nervous activity is abolished. A piece of food placed upon the tentacles of such an animal calls forth no special response and either remains where it was placed or moves sluggishly off to the periphery of the dise under the action of the tentacular cilia. When, however, such a piece is put on the lips, the cilia reverse and the morsel is gradually carried down the esophagus and discharged into the gastrovascular cavity. The swallowing is usually not so rapid as in the normal animal for, under this form of narcotization, the transverse muscles of the mesenteries do not respond to the food by opening the esophagus and conse- quently the cilia are obliged not only to transport the morsel but to force it down a partly closed tube. This, however, they are usually able to do and thus quite independent of neuro- ACTINIAN BEHAVIOR 197 muscular help, they bring about the swallowing of food and the rejection of non-food, for under these circumstances inert mate- rials were found not to reverse the ciliary stroke. Thus, as Allabach (’05, p. 38) has pointed out, the reversal of the effective stroke of the oral cilia is a process which is largely independent of the physiological state of Metridium. In one particular only does this process appear to be related to the general condition of the animal. Ordinarily the reversal of the oral cilia is accomplished by dissolved substances from the food and in my earlier studies on Metridium I was able to get this reversal only by such means. Torrey (’04), however, showed that in Sagartia this reversal could be brought about by mechanical stimuli as well as by chemical means and that it was favored by a starved condition of the animal. Allabach (05) also found that in Metridium a ciliary reversal could be induced by mechanical means and Gee (713) has recently shown that specimens of Cribrina which have been in the laboratory some time do not exhibit a reversal to mechanical stimuli, whereas those still in their native pools give evidence of it. From my own reinvestigation of the question, I am led to agree with Allabach (05, p. 37) that in Metridium marginatum some individuals on mechanical stimulation reverse their ciliary stroke readily, others less readily, and still others not at all, variations largely dependent upon whether the animals have been starved or fed. Two underfed specimens of Metridium which on being tested were found to reverse their cilia to clean filter-paper were vigorously overfed and after three hours were tested again with bits of clean filter-paper. In both instances the paper failed to bring about a reversal of the cilia and conse- quently was ejected. In another test made eighteen hours after feeding, the paper was engulfed showing that the cilia had returned to the state characteristic of animals that had lacked food. I therefore, believe, contrary to my former opinion, that an underfed Metridium will reverse the effective stroke of its oral cilia to mechanical stimulation, though a small supply of food will obliterate this peculiarity and leave these organs inca- pable of such reversal. 198 G. H. PARKER The oceasion of this loss of the power to reverse the stroke of the oral cilia on mechanical stimulation has been ascribed by Allabach (’05, p. 39) to the difference in metabolism between a well fed and an underfed individual. I have tested this by cutting out the esophageal tubes from several specimens of Metridium, laying them open and experimenting with them as ciliated membranes. If they are carefully prepared from ani- mals that have not been recently fed, they will show a well marked ciliary reversal to pieces of clean filter-paper. To frag- ments of mussel they reverse the ciliary stroke in the way char- acteristic for food and after a dozen or more such trials they will no longer reverse to pieces of clean filter-paper. Thus the isolated membrane exhibits all the changes that it does as a part of the whole animal and under conditions where it is quite obvious that the one change that it has suffered is fatigue. I therefore believe that the general metabolism of Metridium is ~ not so much concerned with the change in the character of the response of the cilia to filter-paper as the fatiguing of the recep- tive mechanism of the ciliated surface is. In the undisturbed state this mechanism is at its greatest sensitiveness but on feed- ing its efficiency diminishes and hence filter-paper no longer excites a reversal, a change which is now called forth only by the more vigorous stimulation from the dissolved products of the food. Hence in my opinion the activities of the oral cilia are more independent of the rest of the actinian than even Allabach (’05, p. 38) was inclined to insist upon. The feeding movements of the tentacles in actinians are obvi- ous neuromuscular reactions, as their disappearance on narcoti- zation with chloretone amply shows. ‘The independence of the individual tentacles in their feeding reactions has been demon- strated in a number of forms, in which these responses have been observed after the tentacles have been cut from the polyp. That one tentacle can influence another through connections in the oral disc has been proved for Condylactis and is probably true for Metridium. The muscular responses of the tentacles in feeding, therefore, give much more opportunity for unified action than do the ciliary responses just considered. ACTINIAN. BEHAVIOR 199 That tentacular responses in actinians change with continued activity has long been recognized. Jennings (’05, p. 400) found that the tentacles of Stoichactis after they had been vigorously plied for a while with meat ceased for a time to react to food: Allabach (’05, p. 38) noted that in Metridium the tentacular reactions became gradually slower or even ceased as feeding -progressed, and the same is recorded by Gee (713, p. 320) for Cribrina. I long ago published evidence of this in Metridium and my recent work on this point is entirely confirmatory. Jennings (05) attempted to explain this change as due to loss of hunger,! but (Allabach, ’05) showed that it also occurred when the tentacles were stimulated but the animal was not allowed to swallow the food. Her conclusion is that it is simply the effect of fatigue. Gee (13, p. 324), however, declined to accept this explanation because if an actinian that will ordinarily show this tentacular change after having been fed eight or ten times, is experimented upon when in a fresh condition and is made to contract about the same number of times, its tentacles are found not to have lost their responsiveness. But both Allabach and Gee have failed to recognize that there are several kinds of fatigue. It is perfectly clear, from Gee’s experiment, that mus- cular fatigue is not accountable for the change in the responsive- ness of the tentacles, but it is entirely possible that it may have been caused by sensory fatigue. It is a common observation that if a sensory surface is placed under active stimulation, it is often only a short time before it will fall off very considerably in its receptiveness and it is this form of fatigue, I believe, that is accountable for the change in the tentacular responses of Metridium on continuous feeding. I have had occasion several times to repeat Allabach’s experiment of placing food on the tentacles of Metridium and, after they have responded, of remov- ing it from the lips before it was swallowed, and in all instances 1 Tt is perhaps unfortunate that the term hunger should have been used, for it is somewhat ambiguous. Usually it stands for a well known sensation due to movements of the stomach (Cannon and Washburn, 712); less commonly for in- sufficient bodily nutrition. Pathology has long since demonstrated that these two phenomena are not necessarily connected, but in which sense Jennings in- tended to use the term is not always wholly clear. 200 G. H. PARKER I can confirm her results, namely, the tentacles fall off in re- sponsiveness. In view of what has already been stated I am unable to explain this phenomenon except as a result of sensory fatigue. But there are also changes in the tentacular responses of actin- ians that are by no means so easily explained as are those that have just been considered. Jennings (’05, p. 457) states that when the tentacles on the left side of an Aiptasia were plied with crab meat, they transferred the food to the mouth quickly five times, after which they reacted slowly on the sixth trial and hardly at all on the seventh. On trying the meat on the tentacles of the right side, it was found that the transfer to the mouth was quickly accomplished. Returning now to the left side four sluggish deliveries were effected after which the right side would now take no meat at all. Allabach (’05, p. 39) states that Metridium can be fed from one side of its disc till no more food will be accepted, whereupon / food will likewise not be accepted by the tentacles of the oppo- ~ site side. Gee (13, p. 321) has also recorded essentially the same condition in Cribrina. From these observations it seems clear that changes induced in the muscular responses of the tentacles of one side profoundly influence the reactions of the tentacles on the other side. As Jennings (’05, p. 457) has put it, the animal reacts as a unit, one side influencing the other. I have repeated experiments of this kind on Metridium and though my results are not as striking as those described by the authors already quoted, I am convinced that when a Metridium is fed persistently by means of the tentacles of one side and so as to avoid touching with the food those of the other side, the opposite tentacles are nevertheless eventually influenced in their reactiveness and become less responsive as the feeding proceeds. | Here would seem to be a good instance of some such general effect as that of changed metabolism or the general utilization even of nervous experience. To ascertain whether changes in the tentacular responses of one side of the dise are transmitted nervously to the other side, I fed small pieces of mussel to the tentacles of one side of a ACTINIAN BEHAVIOR 201 Metridium but removed them before they were swallowed and then, after the tentacles of that side began to lose in responsive- ness, I tested those of the other side to see if they too had lost in their capacity to respond. The times in seconds required for the swallowing of each piece of food are recorded in the following table. The rejection of a piece of food is indicated by the sign of infinity. It must be evident from an inspection of table I that the right side of the animal gave no evidence of having been influ- enced by the left and that therefore we are not warranted in TABLE I Time in seconds for the transfer by the tentacles of Metridium of small pieces of mussel to the mouth whereupon they were removed as they were about to be swal- lowed. Sixteen trials were made on the left side and then the same number on the right. © indicates a discharge of the piece of meat at the periphery of the oral disc NUMBER OF THE TRIAL 1 Pa Me Bye | EE NR) Go ZS ORs LON eS PON Sia S14 TS 6 Left side of disc...... ./284/121)107| 86/103; 92| 71} 58} | 97/108} 2} ©} 72) «| o Right side of disc..... 72| 86/306) 63)112| ©} 83/132) 74) 96/103) 97| 86) 62) 78/109 assuming that the experience of one side is transmitted nerv- ously to the other. In other experiments, in which the frag- ments of mussel delivered to the tentacles of the first side were allowed to be swallowed instead of being removed, the tentacles of the opposite side very regularly exhibited a decline in re- sponsiveness. I therefore believe that this change is due to the food introduced into the gastrovascular cavity, and, since the, _ pieces of food were very small, not to the accidental transfer of food juices from the side of the disc stimulated to the other. as suggested by Gee (713). To remove any doubt on this point I adopted a modification of an experiment tried by Gee (’13) and injected by means of a fine glass syringe through the column wall of small specimens of Metridium a considerable amount of mussel juice into their gastrovascular spaces. This operation is easily accomplished Z 7 202 G. H. PARKER especially if the region through which the puncture is made is previously anesthetised with magnesium sulphate. I could not see that the injected juice escaped from the mouths of the ani- mals which, however, took in a considerable amount of sea- water and enlarged much as well fed actinians do. After an hour or so I tested the tentacles of the injected actinians with fragments of mussel and found them very noticeably insensitive to food. It therefore seemed clear that it was the food in the gastrovascular cavity rather than any accidental overflow that had influenced the tentacles. I have already pointed out reasons for believing that the change in the responses of the tentacles after continuous feeding is due to sensory fatigue and not to a general metabolic change and I believe that the same is also true in the particular instance under consideration. Though the meat juice injected into the gastrovascular cavity unquestionably serves as material for metabolism and eventually must have its influence on the ani- mal’s general state, its first condition is that of a component of the fluid mixture which bathes the inner surfaces of the actinian. These surfaces include the cavities of the tentacles. As I have ‘shown elsewhere (Parker, ’17a) substances in solution in the gastrovascular space of such organs as the large tentacles of Condylactis penetrate in a very short time the thin walls of ‘these parts and thus make their way to the exterior. In doing vso they must come in contact with the sensory ectoderm. Since the changes in the reactions of the tentacles produced by food juices injected into the gastrovascular cavity are in the direc- tion of diminished response and since these changes come over the tentacles with considerable rapidity and before a modified metabolism dependent upon new food could have got much headway, I believe that the loss of responsiveness in this instance, like that in the former case, is due to sensory fatigue and not changed metabolism. In the first instance the fatigue was pro- duced by the direct application of stimulating substances to the exterior of the tentacles; in the second to the transfusion of those substances from the cavities of the tentacles to their sensory mechanism. If this explanation is correct, as there is good reason ACTINIAN BEHAVIOR 203 to suppose it is, the responses of the tentacles are like those of the oral cilia in that they are not especially dependent upon the condition of the animal as a whole. As Gee (’18, p. 326) states, “the view that Ne seat of the modi- fied responsiveness lies very largely in the individual tentacles is more clearly in accord with what is known of the structural organization of the sea-anemone than that the animal acts as arunits’”’ The appropriation of food by sea-anemones then is a process which involves factors none of which necessitate the assumption of the action of the animal as a whole. All are most strikingly local and the changes that they exhibit are apparently entirely due to fatigue. In these respects they are in strong contrast with food appropriation in the higher animals, a process which has become so deeply wrought into the make-up of these forms that its relation to the animal as a whole is most profound. While almost every one of the elements involved in actinian food appropriation may be experimentally isolated and made to act for itself in a most remarkably local way, scarcely any such inde- pendence is observable in the parts concerned in the similar operations of higher animals; the jaws and their muscles, buccal glands and so forth in these higher animals exhibit a highly unified action dependent chiefly upon central nervous connec- tions such as is scarcely suggested in actinians, but as isolated elements they have almost no reactive power at all as compared with what is possible in sea-anemones. Food appropriation in actinians then emphasizes rather the relative independence of parts than the action of the organism as a whole. 3. RETRACTION AND EXPANSION As the locomotor activities of Metridium, and in truth of most other actinians, are extremely limited, the chief protective re- sponse of these animals is general retraction whereby they are reduced greatly in bulk, their more delicate parts are brought under cover, and they shrink close to the substratum to which they are attached. In many instances in fact retraction brings about a withdrawal of the body of the actinian into deep, rocky 204 G. H. PARKER recesses and the like whereby very efficient protection is secured. The reverse process, expansion, is one which involves an enlarge- ment and protrusion of the body as a whole and the opening of its folded surfaces and apertures in such a way that the opera- tions of feeding, respiration, and so forth may be resumed. The means by which retraction and expansion are carried out have already been partly described (Parker, 716). Retraction in its initial phases is chiefly the result of the action of the mesenteric muscles, the longitudinal muscles of the non-direc- tive mesenteries depressing the oral disc, those of the directives serving chiefly to fold the siphonoglyphs, and the parietai mus- cles acting on the column wall. After the depression of the oral disc has proceeded somewhat, the contraction of the sphinc- ter muscle completes the process by bringing the oral dise under cover through the puckering effect of this muscle on the column wall. Incidentally the process of general retraction involves the expulsion of almost all the water contained in the gastrovascular cavity of the actinian. The reverse operation, expansion, is dependent first of all upon the relaxation of the sphincter and of the mesenteric muscles; then follows the slow filling of the gastrovascular spaces with sea-water through the ciliary cur- rents in the siphonoglyphs; and probably as a last step the circular muscles of the column contract on the fluid contents of the body whereby the oral disc is forced well up above the pedal attachment. The details involved in the processes of re- traction and expansion allow retraction to be accomplished much more quickly than expansion. This relation has all the appear- ance of an adaptation, for the quickness of a withdrawal may often be the essential part of the protection given by retraction, whereas there is nothing about the economy of an actinian, such as feeding, respiration, and so forth, that makes it vitally impor- tant for the animal to expand quickly. The conditions under which a Metridium remains fully ex- panded are by no means simple but include an aggregate of factors. In the laboratory the fullest expansion was obtained when the animals were in well-oxygenated, cool, running sea- water in the dark. Under such circumstances this sea-anemone ACTINIAN BEHAVIOR 205 will extend itself to as much as six times the diameter of its column, and hold its oral disc fully opened. In no instance have I ever found in nature a degree of expansion greater than that seen in the laboratory under the circumstances Just stated. This maximum degree of expansion under natural circumstances has often been observed in sea-anemones in pools during the night or even during the day in dark situations such as under bridges and so forth. The elements that contribute to this extreme expansion are certainly diverse. Of these I have tested light, temperature, food, oxygen supply, and water currents. The influence of light on actinians is by no means uniform but differs with different species. According to Nagel (94, p. 545; 796, p. 33) Adamsia, Anemonia? and Actinia are not respon- sive to light. Fleure and Walton (07, p. 217) have noted this lack of response in Anthea as well asin Adamsia. Piéron (’06 ¢, p. 44; 08 c, p. 1021) has confirmed Nagel’s statement for Actinia. Although this lack of response may be true of the forms just mentioned, I have not been able to demonstrate it in Metridium marginatum nor in Sagartia luciae, both of which according to Hargitt (07, p. 280) are said to be quite indifferent to light. My observations on these species leave no doubt in my own mind that both close quickly on bright illumination. This is in agreement with Bohn’s observations (’06a, p. 421) as well as with Gosse’s account (’60, p. 15) of the closely allied species, Metridium dianthus. Concerning this form Gosse remarks that “it is under the veil of night that the anemones in general expand most readily and fully. While the glare of day is upon them, they are often chary of displaying their blossomed beau- ties; but an hour of darkness will often suffice to overcome the reluctance of the coyest. The species before us,’’ M. dianthus, “is not particularly shy; it may often be seen opened to the full in broad daylight; but if you would make sure of seeing it in all the gorgeousness of its magnificent bloom, visit your tank with a candle an hour or two after nightfall.’ Retraction under > Bohn (07 ec) states that Anemonia is not entirely without response to light. In weak light it is said to place its tentacles at right angles to the rays and in strong light parallel to them. 9 206 G. H. PARKER bright illumination has also been recorded for a number of actinians among which are the following: Edwardsia* (de Quatre- fages, ’42, p. 76; Fischer, ’88, p. 23), Cerianthus? (Haime, ’54, p. 348; Nagel, 94, p. 545; Hess, ’13, p. 4388), Phillia (Gosse, ’60, p. 350), various species of Sagartia (Gosse, ’60, pp. 81, 111; Fleure and Walton, ’07, p. 217; Hargitt, ’07, p. 275; Piéron, (08 c, p. 1021), Paractis (Jourdan, ’79, p. 28), Cladactis (Hert- wig, ’79-80, p. 56), Aiptasia (Jennings, ’05, p. 459), Tealia (Fleure and Walton, ’07, p. 217), Eloactis (Hargitt, ’07, p. 275), Ceractis (Schmid, ’11, p. 538), and Bunodes (Hess, ’13, p. 438). Although closure in the presence of light is the ordinary form of response for most actinians, there seems to be good evidence that a tew react in the opposite way. Actinia equina according to Bohn (08 a) is expanded in the daytime and retracted at night and the same is true of Cribrina zanthogrammica as observed by Gee (713, p. 309), who also adds that a closed Cribrina in the dark will expand under the influence of a 32 candlepower light. Both Actinia and Cribrina contain symbiotic algae in their tis- sues and it js easy to imagine that their expansion in daylight may be an advantage so far as photosynthesis is concerned, but whether this expansion is a reversal of the usual form of actinian response to light or is due to the effects of some such substance as oxygen which may be given off by symbiotic alga in the light is not known. It thus appears that aside from a few indifferent actinians and a few that open in the light, the majority respond to the stimulus by retraction. In this respect, as already intimated, Metridium is not exceptional. If a fully expanded Metridium in the dark is suddenly exposed to diffuse daylight, it will shorten its column to one-third or one-fourth its former length and with its oral dise fully expanded remain in this state more or less continuously. The shortened state produced in Metridium by general illumination represents the ordinary condition in which many of these sea-anemones are found in nature during the daytime. If on such a partly 3 These instances, Edwardsia and Cerianthus, are often attributed to Bronn: (60, p. 23) who apparently simply repeated the statements made by de Quatre- fages and by Haime without giving references. ACTINIAN BEHAVIOR 207 contracted Metridium a beam of reflected sunlight is thrown, the - animal will after a minute or so almost invariably shorten its column completely and contract its oral disc, thus assuming the condition of complete retraction. This state is commonly met with in nature as a result of direct exposure to sunlight. It occurs in situations where the sea-anemones are subjected dur- _ing a part of the day to shadow and during the rest to full sun- light. Under the latter circumstances they are almost invariably fully retracted; under the former they are more or less expanded. When a fully expanded Metridium in running water in the dark is illuminated either from the side and trom above by a 16 candlepower electric light at a distance of half a meter the animal will shorten considerably but, as a rule, not cover the oral disc. This was occasionally induced by very strong arti- ficial illumination, but it is a reaction by no means easily called forth. It was however often enough met with to warrant the conclusion that so far as light is concerned Metridium will undergo complete contraction of both column and oral dise only in the very brightest illumination; that in weaker light it shortens the column but does not cover the. oral dise and that its fullest expansion is called forth only in complete darkness. The effect of the temperature on actinian response has been little studied. The specimens of Metridium upon which my observations were made were kept in an aquarium with running seawater, the temperature of which was about 23°C. The temper- ature of the outside water from which the supply for the aquarium was obtained was about 21°C. (August). At such temperatures, as was to be expected, the animals remained expanded when the other conditions were appropriate, and normally responsive. When the animals were supplied with running seawater that had been artificially cooled to about 8°C, they remained fully expanded in the dark and would shorten in the hght. They responded to a mechanical stimulus by contraction, and in other respects they reacted as they did under more usual temperatures. If sea-anemones in seawater at 23°C. are.flooded with water at 35°C. even though they are kept in the dark, they invariably contract completely. This response is in agreement with what 208 G. H. PARKER was found by Fleure and Walton (07, p. 217), namely that Actinia and Anthea retract at temperatures above 22°C. If, how- ever, Metridium is subjected to a gradual change of temperature which eventually reached 36°C., it slowly loses its responsiveness to mechanical and chemical stimuli and soon dies. The loss of responsiveness begins at about 34°C., and is complete’ at 36°C. An animal kept a few minutes in seawater at 35°C. may be touched repeatedly on the column near the pedal disc without showing any response and may be eventually killed in alcohol in an expanded condition. Animals which have thus been ren- dered insensitive seldom recover but in the course of a day or so die. So far as Metridium is concerned subnormal temperatures have little influence on its responsiveness except possibly on the rate. Supernormal temperatures, if quickly applied, induce gen- eral contraction; if gradually applied and of sufficient intensity (85°C), they bring about a condition of non-responsiveness that quickly passes over without contraction into one of death. I made no attempt to localize the receptors for differences of temperature (if, in fact, this response is dependent upon recep- tors) and I am, therefore, not in a position to confirm or deny Nagel’s statement (’94, p. 337) that the tentacles are the organs concerned. Many observers in the past have noted that retracted sea- anemones can be induced to expand by placing pieces of meat or other food so near them in the water that dissolved materials from this food are wafted to the animals. Pollock (’83, p. 474) and Romanes, in consequence of such observations, were led to assume the presence of the olfactory sense in these animals. More recently this response to food has been observed in Me- tridium by Allabach (’05, p. 37) and in Actinia by Piéron (’06 b, 706 c). So far as Metridium is concerned, I can fully confirm Allabach’s statement. If into two large glass dishes of fresh seawater many specimens of contracted Metridium are placed and into one of these dishes is poured a small amount of juice from a crushed Mytilus edulis, the sea-anemones in that dish almost without exception will expand their oral discs in a very ACTINIAN BEHAVIOR 209 few minutes, whereas those in the other dish will remain almost to an individual retracted. It was quite clear to me from obser- vations of this kind that the dissolved products from the food of the sea-anemone would induce the expansion of its oral dise though this agent had very little effect on the shortened condi- tion of the column in these animals. The part played by oxygen in the expansion and retraction of sea-anemones has been a matter of recent dispute. Accord- ing to Piéron (06b) Actinia equina opens in seawater with a large oxygen content and closes when there is a deficiency of this gas. Piéron (’08 a, ’08 b, ’09) as a result of further investi- gations was led to believe that not only did oxygen have this effect but that it was one of the most important factors in deter- mining expansion and retraction. Bohn (08 a, ’08c, ’10 a) on the other hand maintains that Actinia equina will remain ex- panded in seawater containing very little oxygen and will close when that water is richly oxygenated. In the opinion of this investigator the states of expansion and retraction are due chiefly to light and darkness and not to the supply of oxygen. In the face of such differences of opinion it is difficult to arrive at any conclusion without further observation. All the experiments that I made to ascertain the importance of oxygen in retraction and expansion were carried out on Metridium marginatum at Woods Hole, Massachusetts. Speci- mens of this species were studied in rock pools which were flooded at high tide and left isolated at low tide. The oxygen content of the seawater from the several situations involved was determined by the Winckler method. I am under obligations to Dr. H. Wasteneys for having made these determinations for me. The outside water on the incoming tide was found to con- tain 7.06 mgm. of oxygen per 1000 cc. The water in a small undisturbed pool just previous to the entrance of the tide con- tained 3.15 mgm. of oxygen per 1000 cc., while that in the undis- turbed end of a pool into which the tide was beginning to flow, contained 2.76 mgm. At the end of the pool into which the tide had entered, the oxygen was found to be 7.02 mgm. per 1000 ce. From these figures it is evident that at the time of THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 210 G. H. PARKER observation the water in the pools contained decidedly less than half as much oxygen as that in the flowing tide and that the entrance of the tide into a pool quickly changed the water there from a condition poor in oxygen to one relatively rich in this gas. Another point of difference in the water of the pool and that of the flowing tide was that the pool water had a tempera- ture of about 27.5°C. and that in the inflowing tide 21.5°C. Experiments to ascertain the effect of the oxygen in the several kinds of water on Metridium were conducted under the follow- ing conditions. The pools, which were on Pine Island, in Woods Hole, Massachusetts, were studied on clear days in August. Work was begun as the tide was rising but before it had reached the pools. In bright daylight almost all the specimens of ‘Metridium were retracted. Large battery Jars were carefully filled with water from the pools and into these jars stones were put having attached to them several specimens of Metridium in the retracted condition. The jars were allowed to stand in the same exposure as the pool to determine whether the act of transfering the sea-anemones would influence their conditions. As a matter of fact the animals remained closed and gave no evidence of being in any other state than that of the actinians that remained in the pools. Careful transfer from pool to jar is therefore not a source of disturbance to Metridium. If, now, pieces of stone on which there are closed actinians are quickly transferred from the pools to the outside tidal water, many of the sea-anemones on them will in a few minutes expand their oral dises though their columns will remain contracted. This response, though not invariable, was of such common occur- rence that it was quite obviously typical of the transfer. It must depend upon some difference between the two bodies of water, for as has already been shown, the act of transfer in itself is without significance. The difference between the two bodies of water are differences of temperature, oxygen content, and current action. To ascertain the effect of temperature, carefully collected pool- water was cooled by being surrounded with ice from its initial temperature of 27.5°C to that of the outside tidal water, 21.5°C. ACTINIAN BEHAVIOR 210 Into this cool pool-water pieces of rock from the pools on which were closed actinians were introduced, and the animals watched. They remained contracted for over an hour and it was con- cluded that the expansion of the sea-anemones when transferred from the pools to the outside tidal water was not due to the difference in temperature. Pool-water,was now collected and thoroughly aerated by being poured back and forth from one jar to another many times, but when placed in this the sea-anemones also failed to expand. Some of this water on being examined proved to contain 7.33 mgm. of oxygen per 1000 cc. It is therefore clear that Metridium does not expand in the running tide because of the increase of oxygen. Finally two jars were so arranged that one conducted water into the other through a large siphon in such a way as to expose the flowing water to air as little as possible. The upper jar being kept full of pool-water, supplied the lower jar from which the water was in continuous overflow. In this way pool-water was given a current without changing in any marked degree its temperature or its oxygen content. When closed specimens of Metridium on bits of rock were introduced into the jar through which the water was flowing, they very commonly expanded their oral discs though their columns remained short. Ithere- fore concluded that the motion of the tide water, rather than its lower temperature or greater oxygen content, was the element responsible for the expansion of Metridium under the cireum- stances noted. As a check on this conclusion several vessels were filled with tidal water and after it had come to rest stones carrying Metridium were introduced into it. Although this quiet tidal water retained its characteristically lower temperature and its higher oxygen content, the sea-anemones remained closed in it, thus confirming the conclusion already expressed that motion is the element in tidal water that induces expansion. The effect of water currents and other forms of agitation were not only observed under natural conditions but were tested likewise in the laboratory. If a Metridium is put in a darkened vessel through which seawater is running, it quickly assumes 212 G. H. PARKER a condition of maximum expansion both as to its oral disc and its column. If, now, the current is shut off, in about a quarter of an hour the oral disc will be found covered but the column will remain more or less elongated. The same was found true of groups of Metridium on stones. Five, in one group, were made to expand fully in running seawater in the dark. The current was then cut off and in eighteen minutes the oral discs of all five specimens were covered and some of the animals a little shortened. An hour and a half after the current had been stopped all were still closed except one which had partly expanded its oral disc. Still an hour later all were retracted, whereupon the current was reestablished and in seven minutes all were expanding, a process completed by all five in about thirteen minutes. These responses were found to occur as well at 8°C. as at the more usual temperature of 21°C. The agitation of the seawater, in a purely mechanical way and without reference to oxygen and the like, appears, there- fore, to be a means of inducing the expansion of Metridium, especially of its oral disc. This form of reaction has already been observed in Actinia by Piéron (’06b, ’08 d), who, however, points out that certain forms of mechanical agitation also induce retraction (Bohn, ’07 a). Since the expansion of the oral disc is dependent chiefly upon the relaxation of the sphincter muscle, it seems probable, as already pointed out (Parker, 716), that the mechanical stimu- lus of the moving water in one way or another has a very spe- cific effect on this muscle. The condition of relaxation thus induced is apparently exactly lke that seen in such sponges as Stylotella, where the oscular sphincter remains. relaxed in running water but contracts when the current ceases (Parker, 110): The foregoing account shows quite clearly that the expansion and retraction of such a sea-anemone as Metridium is dependent upon a variety of factors. Light and high temperature, espe- cially when suddenly applied, produce retraction; food and water currents, expansion; the oxygen supply, in Metridium at least, ACTINIAN BEHAVIOR 213 seems to have very little if any direct influence on retraction and expansion. These operations in sea-anemones have been regarded by some investigators, notably Bohn and Piéron, as occurring in rhythmic fashion, and two types of rhythm have been distinguished; a tidal rhythm and a daily or nychthemeral rhythm. According to Bohn (06.b, 709 b, 710 b) Actinia equina retracts when it is exposed to air by the falling tide and expands when it is again covered by water. This rhythm may be retained for from 3 to 8 days in an aquarium though the animals under such con- ditions are always under water. Piéron (’08 ec) on the contrary questions the presence of a pronounced tidal rhythm in Actinia equina. Metridium marginatum is found commonly either below low- water or in pools that do not empty on the falling of the tide. When exposed to the air it usually retracts though this is not invariable. This species, partly from the situations in which it is found and partly from its irregularity of response, is not a very favorable one in which to seek evidence of tidal rhythm. In this respect Sagartia luciae is very much more promising. This species attaches itself to stones, shells, and other fixed objects that are commonly exposed to air by the falling tide. When thus exposed this species is very regularly retracted, and when covered with water it is expanded though not invariably so. To ascertain whether this rhythm would persist, as main- tained by Bohn for Actinia, I transferred at various times to an aquarium stones covered with Sagartia luciae and kept records of their subsequent conditions. My results were quite uniform and may be well illustrated by a single example. On July 7 at 11.00 in the morning a stone that had been exposed by the tide for some hours and that had upon it twelve contracted Sagartia was transferred to an aquarium. At half past eleven all the sea-anemones had expanded and they remained so for the next thirty hours, after which they began to close irregularly. Similar conditions were repeatedly observed and I am quite sure that in Sagartia luciae there is no persistence of a tidal rhythm. 214 G. H. PARKER In this respect my observations agree with those of Gee (’13, p. 310) on Cribrina, where no trace of the persistence of tidal rhythm could be discovered. Metridium marginatum is almost always under water and is so responsive to light that it might well be suspected to be a species that would exhibit a pronounced daily or nychthemeral rhythm. On August 9 at 10.30 in the morning a large pool in full sunlight was plotted and twenty large specimens of Metri- dium were accurately Jocated. All were fully retracted. At 10 o’clock on the evening of the same day, the sky being overcast with clouds and the night dark, the pool was again visited and by means of a hand hight the twenty sea-anemones were reiden- tified. All were fully expanded. A number of other observa- tions of this kind and many casual records were made of the condition of pool animals in daytime and at might, and always with the same results; the sea-anemones were fully expanded at night and partly or completely retracted in the day. Observa- tions on animals located under bridges and in other dark situ- ations showed that they were more or less continuously expanded, but aside from such exceptions it was clear that Metridium in its natural surroundings exhibited a well marked nychthemeral rhythm. This form of rhythm agrees with what Hargitt (07) has observed in Eloactis, and Piéron (’08 c) in Sagartia troglodytes, and what has been claimed by Bohn (’06b, ’07 b) to occur in Actinia equina, though the nychthemeral rhythm in this species has been questioned by Piéron (08 ¢, ’08 e). That in Metridium it is dependent upon light, as maintained in general by Bohn (08 a, 710 a), and not upon oxygen, as was claimed for other species by Piéron (’08 c, 08 e), has already been shown in an earlier part of this paper. I have never observed anything about the activities a Me- tridium that would lead me to suppose that its nychthemeral rhythm is ever reversed or is ever exchanged for a tidal rhythm as has been claimed for some species by Bohn (’08 b, ’09 b). A persistence of the nychthemeral rhythm in Metridium after its removal from the influence of day and night is apparently ACTINIAN BEHAVIOR Palle as little in evidence as the tidal rhythm in Sagartia luciae. Specimens of Metridium in a retracted condition were removed from a quiet pool at noon and placed in running water in the dark. In less than an hour all were fully expanded and remained so for over 36 hours. These observations agree with those of Gee (713, p. 310), who was unable to find any evidence for the persistence of the nychthemeral rhythm in Cribrina. That sea-anemones may exhibit in the sequence of their states of expansion and retraction a tidal rhythm or a nychthemeral rhythm, as pointed out by Bohn and by Piéron, there can be not the least question, buc that these rhythms may persist even for a few days in the absence of the external stimulus, as main- tained especially by Bohn, js certainly not true for Metridium marginatum nor Sagartia luciae. The fact that a persistence of rhythm in aquarium specimen has not been seen by Appelléf (Retterer, ’07), Gee (13), and others throws great doubt on the occurrence of this phenomenon at all, but a decisive answer to this question can not be given till the species for which these peculiarities have been claimed are reinvestigated. Bohn and Piéron (’06), and especially Piéron (’06b, ’08 a, 10) have claimed that in Actinia equina the tidal rhythm is carried out a little in advance of the actual tidal changes, thus giving evidence of what may be called an anticipatory reaction. This reaction, according to Piéron (’10), may be lost when the animals are placed in an aquarium and may be regained after a week or so when they are again subjected to the tides. I have watched Metridium very closely for signs of this preparatory activity, but I have never seen any conclusive evidence of it. It is astounding how quickly Metridium will begin to expand on the entrance of the tide into a pool in which this sea-anemone is located. With this species expansion often begins within a few minutes after the arrival of the first new tidal water. As already pointed out I believe this expansion to depend upon the move- ment of the water and not upon its temperature or its oxygen content. Since the first water that enters the pool, often by indirect and not easily visible channels, may cause all the water in the pool to move somewhat, a stimulus imperceptible to the 216 G. H. PARKER observer may be given to every actinian there and thus induce expansion in what seems to be an anticipatory manner, whereas in reality it is a response to a direct stimulus. It is in some such way as this that in my opinion Piéron has probably been de- ceived, for from my own observations I am led to concur with Bohn (08a) in questioning the existence of reactions really anticipatory. Such rhythms as have thus far been studied in sea-anemones seem, therefore, to depend upon immediate rhyth- mic stimuli external in origin as the changes of the tides or the change from day to night and the reverse, and not upon rhythmic operations of a more internal nature such as probably control the pulsing of a jellyfish or the beat of the vertebrate heart. 4. PSYCHOLOGY The term psychology as applied to such lowly animals as actinians is commonly used to cover a discussion of those activi- ties which may or may not give grounds for the assumption of primitive psychic conditions in these forms. In this non-com- mital sense it is used here as it already has been used by many who have held most divergent views as to the problems involved. In all instances it implies a fundamental consideration of the more complex nervous processes of a given group of animals; a standpoint which it is quite appropriate to assume concerning the actinians. When we examine the organization of the human body and note the perfection of its voluntary adjustments with their involved psychoses and the equal perfection of such neuro- muscular but non-psychic activities as those exhibited by the heart, we may reasonably ask, Is the actinian an organism that responds as the vertebrate heart does, or does it necessarily in- clude in its activities elements of a psychic order? It is with questions such as these that the psychology of the actinians is concerned. As Baglioni (13) has pointed out, the neuromuscular reac- tions of actinians fall into two general classes, first, responses to beneficial stimuli, and, secondly, responses to noxious influ- ences. These two categories doubtless represent the neural background on which rest those states of so-called pleasure and ACTINIAN BEHAVIOR pA la pain that play so prominent a part in the central nervous activi- ties of the higher animals (Holmes, 711). The best instance of the beneficial responses is that seen in feeding, and the noxious responses are well shown in general retraction and in locomotion, for in all actinians, so far as is known, locomotion is always away from the centre of stimulus. It is from a consideration of activities. such as these that sound conclusions can be drawn concerning the possible presence of higher nervous operations in these animals. From the standpoint of these general reactions the question of organic unity and centralization has already been discussed. The view clearly set forth by Jennings (05) that the feeding actinian acts as a unit and that hunger and satiety are impor- tant elements in explaining the changes that appear in the course of its general responses seems to be quite unsupported by subsequent work. The essentially independent action of the tentacles as well as that of the reversing mechanism in the esophageal cilia and the discovery by Allabach (’05) that the changes in the whole mechanism as feeding proceeds are due to fatigue and not to anything comparable with satiety, a dis- covery with which my own observations are in accord, make clear that the feeding process is an activity which involves many semi-independent parts as such rather than the activity of the animal as a unit. When one contrasts the utter loss of effec- tiveness of the isolated appendages of higher animals with the almost normal activities of the detached tentacles of many actinians, the low degree of unity present in such forms as the sea-anemones becomes at once apparent. In the feeding of actinians each part reacts appropriately to its proper stimulus and the total act is carried out by a sequence of responses that have almost no relation to a central control. As already stated I agree thoroughly with Gee (13) that the feeding reactions of actinians give no real BUBDOF to the idea of organic unity in these animals. General retraction is a response to conditions of an unfavor- able or deleterious kind. It is an exhibition of the excessive tonicity of actinian muscle as pointed out by v. Uexkiull (709) 218 G. H. PARKER and by Jordan (08). It involves the animal as a whole and yet it persists as strikingly in small fragments of an actinian as it does in the whole (Wolff, 04). Hence in this respect it gives no ground for the assumption of a specially unified state. The limp toneless condition of a muscle isolated from its nerve in a higher animal is in strong contrast to the tightly contracted fragment from an actinian’s body. The tidal and nychthemeral rhythms of these animals, as described by Bohn, are much more suggestive of organic unity than the single act of retraction itself. This is especially true when we take into account the retention of this rhythm after the removal of the rhythmic stimulus. Such activities imply the origin of new internal states through past stimulations and their retention in the subsequent modes of response of the ani- mal as a whole. - But it is by no means easy to judge of the value of Bohn’s observations in these directions. His first description (Bohn, ’06 ¢) of the rhythms and their retention was relatively simple, but his subsequent account shows such diversity and complexity that it is difficult for the reader to convince himself that rhythms have really been observed. Jennings (’09), who accepted Bohn’s earlier observations with enthusiasm, was led in this way to entertain grave doubts about the accuracy of much that had been claimed. The fact that Piéron (’08 e) reexamined the question of the persistence of tidal rhythm in Actinia without being able to confirm Bohn’s statement about it, and that Gee (’13) and I have been absolutely unable to find any evidence of retained rhythms, either tidal or nych- themeral, in the actinians that we have studied leads me to conclude that, though actinians may exhibit rhythms in conse- quence of rhythmic stimuli, they do not retain these rhythms on the disappearance of the stimuli. If retained rhythms do not occur in actinians, they can not of course be called upon as evi- dence of complex unified nervous states in these animals. Intimately associated with the retention of rhythm in actin- ians is the retention of characteristic positional responses. Jen- nings (’05, p. 461) has shown that an Aiptasia will assume an irregular and distorted form in consequence of the irregularities ACTINIAN BEHAVIOR 219 of the rocky cavity in which it has taken up its abode and that it will retain this shape for some time after it has been removed from its retreat. Moreover irregular forms of some permanence can be artificially and quickly produced by putting an animal into an artificial, irregularly shaped chamber. Such iregular forms are without much doubt due to differences in the degree to which tonicity is developed on particular parts of the animal’s body by variously disposed stimuli. When it is remembered that the tonicity of general retraction may continue in some actinians even for days, it would not be surprising if an irregu- larly distributed tonicity should also have a lengthy period. Van der Ghinst (’06) has also pointed out an interesting case of the retention of a characteristic positional response in Actinia. Specimens of this sea-anemone are found attached either to the undersides or to the uppersides of rocks. When individuals from both locations are collected and put in an aquarium in which both positions are possible, those that were originally on the underside of rocks reassume this position and those that were above move to the uppersides of objects. The positional relation apparently impressed upon them by their previous en- vironment thus reasserts itself and in this manner gives evidence of modified central activities. The habit is said to be lost in twenty-four to forty eight hours. The species of sea-anemones with which I have worked are not often found in the two positions assumed by Actinia and I have therefore not been able to carry out experiments on the lines worked on by Van der Ghinst. As no one seems to have repeated these observations on Actinia or other sea-anemones, and as Van der Ghinst himself claims for the reten ion of the response only the brief period of a day or so, !t seems to me that they call for confirmation before they can be taken seriously into account in a discussion like the present one. A third form of response which may be taken to involve the actinian as a whole is creeping. Locomotion by means of the pedal disc in these animals has already been rather fully dis- cussed (Parker, 717b) and it has been pointed out that this operation can be successfully carried out by specimens of Sa- 220 G. H. PARKER gartia from which the oral half has been cut away. Such frag- ments not only creep but creep away from the light as normal individuals do. In fact their activities are in no essential par- ticular different from those of whole animals. Creeping, then, is In no sense dependent upon the animal as a unit but is an activity of the pedal dise and adjacent parts. It is, however, an activity of the disc as a whole. I have never been able to observe locomotion in pieces of the pedal dise. When actinians are cut in such a way that the fragments retain only parts of the original pedal disc, they remain attached to the substratum by their pedal surface but they never exhibit locomotion. It is only after regeneration has set in and a new pedal dise has been established that locomotion recommences. Creeping then is a response which calls for a much more unified mechanism than feeding and I agree with Lukas (’05, p. 126) in regarding it as a response which gives evidence of the highest form of nervous activity thus far discovered in actinians. I am, however, not prepared to go as far as he does and see in it evidence of a primi- tive form of desire and the earliest traces of consciousness (Lukas, 05, p. 127), but of its importance as indicative of a certain. amount of unity in actinians there can be not the least doubt. Another line of investigation that is suggestive of more than the simplest form of nervous activity in actinians is the modifia- bility of their responses. This subject has been justly empha- sized by Jennings (’05), who has shown its significance by direct experiment. If a drop of water is allowed to fall on the surface of the water in which an expanded Aiptasia rests, the animal will usually retract. After expansion a second drop often fails to call forth any such response and in fact it is necessary to allow as a rule an interval of five minutes before a second response can be elicited. ‘Thus the earlier stimulus influences the neuro- muscular apparatus of the sea-anemone in such a way that a repetition of the stimulus is not followed by a response. ‘To put the matter as Jennings does, the previous history of an organism has its influence upon its subsequent responses. This feature in actinians and in fact in most other animals has long been familiar to workers in this field, but it is to the credit of ACTINIAN BEHAVIOR aA | Jennings to have insisted on its importance. When an expla- nation of this phenomenon is sought, one naturally turns, as in the case of the dying away of feeding responses, to exhaustion. Does not the initial stimulus, the vibration from the first drop of water, so exhaust the neuromuscular mechanism that it is incapable of receiving in an effective way a second stimulus till after a certain time for recovery? This subject has been quantitatively studied by Kinoshita (11), who has shown by the use of several kinds of weak stimuli that the response to the first stimulus is so considerable as com- pared with that to most subsequent stimuli that it is highly improbable that exhaustion plays any important part in the whole operation. Much more likely is it that the neuromuscular apparatus having responded once, assumes a state rather of adaptation than exhaustion and thus saves the organism from subsequent and useless responses. From this standpoint the condition left by the first stimulus and response seems to be that of inhibition—of the production of a refractory period so to speak—rather than that of exhaustion. At least it is clear that the first response has a relatively profound influence on the organism and that this influence lasts long enough—five or ten minutes—to affect subsequent stimuli. Here, then, in the truly nervous activities of actinians is evidence of the beginning at least of nervous states analogous to the more complex conditions found in higher forms. In attempting to make clear the conditions under which the second or modified response takes place, care must be exercised that confusion does not arise as to the nature of the explanation. To one class of workers, those having a physico-chemical bent, a satisfactory explanation of the modified form of response would be found in an understanding of the interaction of the second stimulus and the receptor together with the chain of events that terminate in the muscular movements. ‘This form of explanation is concerned exclusively with the working mechan- ism as such and has nothing to do with its historical origin. The second form of explanation, the one more likely to be adopted by those of a more biological turn, would seek for an understanding 222 G. H. PARKER of the modified response in the influences that had emanated from the original response and thus brought that response into historical relation with the second and modified one. This form of explanation emphasizes the effect of the history of the animal on its immediate state. Both forms of explanation have their places, but they are sufficiently diverse to require separation, a condition not always observed in discussions of this kind. The opinion that the past history of an individual actinian is a potent factor in understanding its behavior has been expressed not only by Jennings (’05) but also by Piéron (06 ¢, p. 15), who declared that the responses of actinians could not be looked upon as purely mechanical operations, but included traces of those activities characteristic of the central nervous organs of higher animals. But very little work has been done on actin- ians to ascertain the extent to which such central activities as those just indicated may extend. The limited range of response in these animals restricts such experimentation considerably. Heretofore associative processes have never been directly identi- fied in actinians and my own efforts in this direction have always yielded negative results. One of these attempts may be briefly described. When Me- tridium is slightly stimulated mechanically on the pedal edge of its column, it responds by a slight initial retraction. When food is put on the tentacles, the first response is an irregular but very characteristic waving of these organs in the immediate vicinity of the food. An attempt was made to associate the two stimuli mentioned so that the tentacle response might be called forth by a mechanical stimulation of the pedal edge of the col- umn. oS oe j iv’ eet if i i 4 “Tarr slat ¢ : q ea Ae ae on ‘ } one vi Wire ¢ ‘nergaing sa - OGD pga! | ere oe d ‘. ~ CONTRIBUTIONS FROM THE ZOOLOGICAL LABORATORY OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. NO. 291. THE RHYTHMIC CONTRACTIONS IN THE MANTLE OF LAMELLIBRANCHS ELIZABETH S. P. REDFIELD FOUR FIGURES INTRODUCTION The respiratory currents through the mantle chamber of the lamellibranchs have been attributed exclusively to the action of the ciliated epithelium, covering the mantle, gills, and foot. Kellogg (15) has recently made a thorough study of these cur- rents and it appears from his work that they have a dual func- tion; to supply water to the gills and to remove foreign matter from the mantle chamber. Babak (18, p. 197) has called attention to rhythmic move- ments of the shells of lamellibranchs, to which he attributes a cleansing function. Die Schalenbewegungen, welche zuweilen gleichsam rhythmisch und in auffalliger Frequenz erscheinen, haben (wenigstens bei Anodonta und Unio) kaum irgendwelche gréssere respiratorische Bedeutung, da sie gerade im sauerstoffarmen Wasser nicht vermehrt werden, obwohl durch den auf diese Weise erfolgenden Wasserwechsel unzweifelhaft eine miachtige Férderung des Gaswechsels zustande kommen wiirde. Es lasst sich dartun, dass die Schalenbewegungen hauptsiichlich als Reinig- ungs-reflexe aufzufassen sind. This paper is a description of certain rhythmic movements found in the mantle of several species of lamellibranchs, and of experiments to determine the function of these movements. The recent summary of work on the respiration of lamellibranchs by Babak (12) makes no mention of this phenomenon, which apparently thus far has passed undetected. I wish to thank Dr. George Howard Parker, under whom this work has been done, for his assistance, and the United States 231 Poe ELIZABETH S. P. REDFIELD Bureau of Fisheries for the privilege of working in its Woods Hole Laboratory. DESCRIPTION OF THE MOVEMENTS OF THE MANTLE If a small hole is cut in the shell of a clam (Unio complanatus), the mantle will bulge out through it. At regular intervals the mantle may be observed to contract rapidly and then to bulge out slowly. To determine the periods of time between these contractions, a small bent needle was hooked through the mantle and attached by a silk thread to an aluminum lever and records were taken on a kymograph. Several species of lamellibranchs were tested and the rate of the rhythm was found to vary in them. It also varied in different individuals of the same spe- cies. In some the period was about one minute, in others about three. In Mya arenaria it was most uniform. The periods lengthen after a clam has been kept in the laboratory for several weeks. It was necessary therefore to use only clams which had been collected within a short time before the tests were made. The movement of the mantle might be attributed to changes of pressure within the mantle chamber, produced by the rhythmic movements of the shell, such as Babak has described. In order to determine this point the following experiment was carried out. Simultaneous records were taken of the movements of both the shell and the mantle of a Unio complanatus. Figure 1, A, B, shows these movements to be in unison. The parts of the valve of this clam to which the anterior and posterior ad- ductor muscles were attached were now cut free from the rest of the valve, thus allowing the adductor muscles to contract without moving the rest of the valve. Records were again taken simultaneously of the movements of both the shell and mantle (fig. 1, C, D). It is clear from these records that, although the shell was motionless, a slight but distinct contraction of the mantle occurred at regular intervals. During the summer of 1915 the movements of the mantle in a number of marine lamellibranchs were studied at the labora- tory of the United States Bureau of Fisheries at Woods Hole, Mass. CONTRACTIONS IN MANTLE—LAMELLIBRANCHS Zao In Mya arenaria the shell may be held tightly closed without disturbing the flow of water through the mantle chamber, for the siphons can be extended when the valves are in this posi- tion. The pulsations of the mantle in Mya arenaria are much more powerful than those in Unio complanatus. A wave of contraction can be seen to start in the distal end of the extended siphon and move forward ending with the rise and fall of the mantle. This recalls an observation of Dubois (’92) on Pholas yy et ta ot Fig. 1 Kymographic records of the movements of the mantle and shell of a Unio complanatus. The upper line (A) is a record of the movements of the mantle; the second line (B) is a record of the movements of the shell. Both records were made simultaneously. The third line (C) is a record of the movements of the mantle; the fourth line (D) is a simultaneous record from the shell of the same specimen after the shell had been rendered immobile by cutting away from the rest of the valve the part of the shell which was attached to the adductor muscles. The lower line indicates time intervals of thirty seconds each. dactylus; a wave of muscular action in the walls of the siphons served as a means of circulating water when the valves of the animal were tightly closed. Figure 2, A, is a record of the mantle movements of Mya. It shows that the interval between successive contractions is about one minute, with often a longer pause after three con- tractions. In table 1, is shown the rate of mantle and shell contractions of Mya arenaria kept in running water during a period of three days. This table indicates that the rate varies considerably from time to time. Figure 3 is a record of the movements of the mantle lobes of both sides of a Mya arenaria taken simultaneously. It will be 234 ELIZABETH S. P. REDFIELD observed that the contractions of the two sides are in most cases synchronous. By contracting synchronously the lobes would tend to decrease the size of the infra-branchial chamber and consequently drive some of the water out of it. In ten other species of marine lamellibranchs pulsations of the mantle of varying degrees of intensity were noted. In those i oe a OI a ee Fig. 2. Kymographic record of the movements of the mantle of Mya arenaria, the valves of which have been held tightly closed. The upper line (A) is a record taken while the clam was in running water. The second line (B) is a record taken thirty minutes after the water in which the clam was placed had been covered with paraffin oil. It illustrates the accelerated rate which characterizes the early stages of suffocation. The lower line indicates time intervals of thirty seconds each. ' TABLE 1 The rate of contraction of the mantle and shell of a Mya arenaria kept in running water for three days. Averages are based on counts extending over twenty-minute periods AVERAGE NUMBER | AVERAGE NUMBER OF CONTRACTIONS OF CONTRACTIONS es ‘tgs OF MANTLE PER OF SHELL PER MINUTE MINUTE Mains Gad aiygees cic: ERR eee 9.30 a.m. 1-3 0.2 Bins tad averse scat eee ee 10.00 a.m. 1.3 0.35 PUTSCRO Oye ete. - . winies cis eoeeee 11.00 a.m. 1.3 Out pecondudanyss. ():...:.''- 202 5 ste as 5.30 p.m. 0.9 0.15 hind Kdlayeeee. .- xcs. ayovenies see: 6.30 p.m. 0.5 0.3 AP hind Razer ce te gee eee 7 245-p-m. 0.9 0.25 DU a ee Cy aa eae Ser ee EE SS a a ae a ee ee Dee en a Fig. 3 Simultaneous kymographic records of the movements of the right and left mantle lobes of Mya arenaria with the shell tied closed. The top line is a record of the left mantle lobe; the middle line is a record of the right mantle lobe. The bottom line indicates time intervals of thirty seconds each. These records illustrate the fact that the two lobes contract synchronously. CONTRACTIONS IN MANTLE—LAMELLIBRANCHS 23) forms which are sedentary, the rhythmic movements were found to be much as in Mya arenaria. Such forms were Modiolus modiolus, Modiolus plicatula, and Mytilus edulis. The last showed only weak movements. In the forms which were active, such as Solemya velum, Ensis directis, Cummingia tellinoides, and Yoldia sapotilla, there are also movements of the mantle perhaps respiratory in function, but somewhat different from those in Mya. In Pecten gibbus, Venus mercenaria, and Ostrea virginica the mantle is very thin and in close connection with the shell. No movements of the mantle were discovered in these forms. The gills in the last three are more active than those in the others, which may be of significance. THE FUNCTION OF THE MOVEMENTS OF THE MANTLE To determine whether the movements of the mantle were concerned with setting up the respiratory currents in the infra- branchial chamber, the effect of suffocation upon the rate of the rhythm was determined. Specimens of Mya, placed in a bow] of running sea-water, were attached to the recording appa- ratus and the rate of the mantle pulsations determined. The water was then covered with a thin layer of paraffin oil.!' As the oxygen in the water became exhausted the rate of the rhythm increased rapidly (fig. 2), but as the movements proceeded this rate fell off, the movements stopping completely before the clam succumbed. Figure 4 is a graphic representation of the changes in the rate of movements of the mantle of Mya arenaria during suffocation. If, as the foregoing facts suggest, the mantle movements are concerned with setting up respiratory currents, it might be expected that an obstruction of these movements would influ- ence the quantity of oxygen consumed by the clam. Accord- ingly experiments were carried out in which the oxygen consump- 1 Bayliss (715) has pointed out the futility of attempting to preserve solutions from the action of gases in the atmosphere by covering them with oil or hydro- carbons. This observation does not invalidate the results of the experiments in question since the animals died of suffocation, though perhaps no more rapidly than if the surface of the water had not been covered with paraffin oil. 236 ELIZABETH S. P. REDFIELD tion by fresh clams was compared with the oxygen consumption of the same animals in which the movement of the mantle had been checked. Several liters of water drawn from the tap were left stand- ing in a closed vessel for twenty-four hours in order that the oxygen content might come to a uniform condition. The amount of oxygen in a sample of this water was then determined by the Winkler method as used by Birge and Juday (11). A 350 ec. glass jar containing a well cleaned Unio was filled with this x 1 2 3 Fig. 4 The solid line indicates the rate of contractions of the mantle of Mya arenaria during suffocation. Time is indicated in hours along the abscissa; the number of movements of the mantle in twenty minutes is measured along the ordinate. Suffocation was commenced at X. water and closed with a tight-fitting ground-glass cover. The jar was allowed to stand twenty-four hours and then a 250 cc. sample was siphoned off and the amount of its contained oxygen was determined. This value subtracted from the original oxygen content of the same volume of water gave the amount of oxygen consumed by the clam in twenty-four hours from that volume. Next, the mantle of the same clam was made functionless by inserting a small knife between the ventral edges of the valves and slitting the mantle on both sides longitudinally from one ~T CONTRACTIONS IN MANTLE—LAMELLIBRANCHS V3 end to the other. To overcome whatever shock might follow and disturb the normal behavior of the organism, the clam was put in fresh water and left for six to eight hours. This clam was then placed in a freshly filled jar and the oxygen consumed by it in twenty-four hours again determined. ‘Table 2 gives the results of a series of such experiments. From this table it is clear that the oxygen consumption of the clams in which the action of the mantle has been checked is greatly reduced. This fact seems to me to show that the mantle in Unio complanatus is concerned with the respiratory function. TABLE 2 Comparison of the quantity of oxygen consumed by specimens of Unio complanatus before and after the activity of the mantle had been checked CUBIC CENTIMETERS PER LITER OF OXYGEN USED IN 24 HRS. BY same CLAM WITH THE MANTLE ACTIVITY DESTROYED CUBIC CENTIMETERS PER LITER UNIO COMPLANATUS OF OXYGEN USED IN 24 HRS. BY CLAM WITH MANTLE INTACT 1 5.671 4.016 2 5.495 1516. 3 5.584 1.906 4 4.918 1.994 5 5.986 2.316 6 3.950 2.610 It may be objected, however, that the injury caused by the operation may have lowered the vitality of the clam, or caused such a loss of blood as to have produced death. To determine whether this objection had any weight or not, clams were sub- jected to numerous other operations. The distal ends of the siphons were cut off, the foot was mutilated, the mantle cut only on one side, or numerous small holes were made in it. Clams thus operated on lived: some a week, others several weeks, others even months. As table 3 shows, these operations had almost no retarding effect on respiration. If fresh clams (Unio complanatus) are left in a sealed Jar containing 300 cc. of water, death occurs in not less than four days. If the movements of the mantle in clams be checked as described in the foregoing experiment, death occurs in twenty- 238 ELIZABETH S. P. REDFIBLD TABLE 3 Comparison of the quantity of oxygen consumed by specimens of Unio complanatus before and after being mutilated in various ways CUBIC CENTIMETERS PER LITER OF CUBIC CENTIMETERS PER LITER OF UNIO COMPLANATUS OXYGEN USED BY A CLAM WITH OXYGEN USED BY SAME CLAM SIPHONS INTACT WITH SIPHONS CUT OFF 1 3.841 4.206 2 4.985 3.148 CUBIC CENTIMETERS PER LITER OF AUSSI EERIE SLD TEE! Op OXYGEN USED BY A CLAM WITH NS RHEIINT WISIRID thy (etal Gee THE MANTLES INTACT WITH A SMALL HOLE CUT IN MANTLE OF ONE SIDE i 5.688 4.256 2 6.043 3.130 CUBIC CENTIMETERS PER LITER OF VAS PERRO GIESSEN BS OPES) ee NE Ot OXYGEN USED BY SAME CLAM OXYGEN USED BY A CLAM WITH THE MANTLES INTACT WITH MANTLE OF ONE SIDE i DESTROYED 1 4.529 4.158 2 4.128 2.086 four hours. This fact indicates that the clams on which opera- tions have been performed are unable to avail themselves of the oxygen which is in the water surrounding them. SUMMARY 1. A rhythmic movement of the mantle, independent of move- ments of the shell, is described in a number of marine and fresh- water lamellibranchs. 2. Experiments are detailed which indicate that the rate of these movements increases during the early stages of suffoca- tion, that the oxygen consumption of the clams is decreased by checking completely these movements, and that clams suffocate more rapidly under these circumstances than otherwise. 3. It is concluded that the movements of the mantle in lamelli- branchs is an important factor in setting up the respiratory cur- rents. The movements may also be of significance in driving waste materials and foreign bodies from the mantle chamber. CONTRACTIONS IN MANTLE—LAMELLIBRANCHS 239 BIBLIOGRAPHY Basak, E. 1912 Die Mechanik und Innervation der Atmung. In Winterstein, Handbuch der vergleichenden Physiologie, Bd. 1, Hilfte 2, pp. 265- 918. Basak, E. 1913 Zur Regulation des Atemstromes bei den Lamellibranchiaten. Zugleich ein Beitrag zur Physiologie der Flimmerbewegung. Zeit- schrift fiir allgemeine Physiologie, Bd. 15, pp. 184-198. Bayuiss, W. M. 1915 Principles of General Physiology, London, 8vo, xx + 850 pp. Birce, E. A., and Jupay, C. 1911 The Inland Lakes of Wisconsin. Wisconsin Geological and Natural History Survey, Bulletin 22, Series No. 7, pp. 13-21. Dusois, R. 1892 Anatomie et physiologie comparées de la pholade dactyle. Annales de |’Université de Lyon, Tome 2, fase, 2, x + 167 pp., 15 pl. Kettoae, J. L. 1915 Ciliary Mechanisms of Lamellibranchs with Descriptions of Anatomy. Journal of Morphology. Vol. 26, pp. 625-701. poe NG ee Se A ies © . wh . is f de 5 x ‘ * e ' ms . . id ‘ ‘ ce i © Noo OL ae Fi ‘ bare i} } it} ov a) ‘ 1 alidopay’tct - \ abil \ ' j ee + dae ‘ hi sv)" f ! \ ay vile < 4 ‘ ‘ & t 4 y tore “ ‘ 4 ~ q af x ; : - t ‘ s 7 t 1 t ' Z ‘ ve } ‘ . * * 3 +s ~~ oe = ard 7 i + : ; \ f ; 7 = 7 : 7 © > *. _ . 7 THE EXPERIMENTAL MODIFICATION OF GERM CELLS. Il. THE EFFECT OF PARENTAL ALCO- HOLISM, AND CERTAIN OTHER DRUG INTOXI- CATIONS, UPON THE PROGENY! RAYMOND PEARL Maine Agricultural Experiment Station SEVEN FIGURES CONTENTS Pep FEO GUC BUOIT 14 4 t\a erate dm PAN cok hs hoa hy Meas ee, CXes cha eas ttoreueee Sees E 241 Il. The fertility and hatching quality of the eggs from alcoholized par- TMU Sear epee ry AN i ene tee ROP SO RRS io LSE eee ae Tew tke ES Rana 242 tite Rotvallbreedinioycap ach tiv saacscce cece oe ara eee en olen 250 IVE 2 Datessotehatehiney Otel chickse is... .t .i 1. eee eae ee ae 254 WaNiNforuall ity (Ob iy Chicks earns mee o.c5 cis <5 Oe MaeRo RE as ak aires a ceases 256 Vit vibe sex ratio umbohe SH i progemyy. Aa. ... soem eis ce oh cae adits ens 262 Neilevidatening weigh tsobel mprogen yA. 5.05... c.2: ae eee tet oa ce ee es 265 Wihtks Growth ior theyll provenys << 4. oes. 40. 50: EL Ae eh irae NEE 269 Xen Weformitiesmmatheshepho ceniverye cor pene eRe ein ae 277 X. Mendelian characters in the F; progeny................0..eeeeeeeee- 278 EM PDMSCUSSIOM Ole rESUNLGS. axecac oe Caer. ~..o ls «a1 J Eee a sittolad obeee ke 279 PM Rep TBO ARP in ewe eo NERS SLUR UNA 80). dicks ee Rae Chia PAT RE 394 DEITIES, AN 0) (S00 liber cre a ee ee Gr 8 eA MSRP St ed Se ae ee 398 1. INTRODUCTION In this paper it is proposed to discuss the effects, so far as any are observable, of the alecoholization of one or both of the parents upon the progeny in the first generation using the do- mestic fowl as material. Different characters of the progeny will be considered, and as before primary attention will be given to such characters as admit of quantitative expression. A detailed account of the experimental methods used, the specific problems attacked, the breeding of the foundation stock 1 Papers from the Biological Laboratory of the Maine Agricultural Experi- ment Station. No. 102. 241 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 242 RAYMOND PEARL used, ete., has been given in I, the first paper in this series. The effect of the aleoholization by the inhalation method used, upon the treated birds themselves, has been given in II.’ It will be recalled that, as stated in I, the present report covers the results of the work up to February 1, 1916 only. Further data will be given in later reports. II. THE FERTILITY AND HATCHING QUALITY OF THE EGGS FROM ALCOHOLIZED PARENTS One of the surest and most delicate indicators of constitu- tional vigor and vitality in poultry which has yet been dis- covered is the hatching quality of the eggs. Anything which upsets the general metabolic balance or impairs the vitality of either partner in a mating will show its effect in a diminished hatching power of the eggs from that mating. In view of these facts an examination of the data relative to this character in these alcoholic matings becomes of especial interest. Before entering upon such an examination it 1s necessary to consider the question of control data. Unfortunately the un- treated control male No. 666 (ef. I, table 2) proved to be prac- tically completely impotent sexually. He mated regularly and apparently effectively with the females in his pen, but practi- cally all of the eggs proved to be infertile, regardless of whether the female concerned was an alcoholic or an untreated normal control. In consequence of this matings 2131, 21382, and 2133 which were planned to serve as controls, were practically com- plete failures. The results which they gave were so far from normal that it would be entirely misleading to use them as con- 2 This refers to the first paper in this series, which was entitled: ‘‘The experi- mental modification of germ cells. I. General plan of experiments with ethyl alcohol and certain related substances.’’ Jour. Exp. Zool., vol. 22, pp. 125. Throughout this and the later papers in the series cross-references to other papers in the same series will be made simply by the Roman numeral designating the paper referred to, together with the particular page number to which refer- ence Is made. ’ This refers to the second paper in this series, which was entitled: ‘‘II. The effect upon the domestic fowl of the daily inhalation of ethyl alcohol and cer- tain related substances.’’ Jour. Exp. Zool., vol. 22, pp. 165. PARENTAL ALCOHOLISM AND THE PROGENY 243 trols on the aleoholic matings. When o666 was autopsied the testes were found to be very small, much under size for a bird of his age and body weight. He was apparently a naturally impotent male, of a sort that occurs not infrequently in poultry breeding operations. Owing to his deceptively vigorous behavior it was too late in the year to substitute another bird in his place when it was finally proven that he was worthless as a breeder. In consequence of the failure of this #666 we are compelled to resort to other data to furnish proper controls for the fertility and hatching data. Fortunately such data are at hand from other experiments, and are entirely adequate both in kind and amount. For comparison with the results of the alcoholic mat- ings there will be used in this section of the paper the results from a random sample, comprising 22 matings, from all of the 1915 matings of normal untreated birds in which the female partner was a pure Barred Plymouth Rock. A sample of 22 matings is taken because that is the number of matings in which one or both of the animals involved was a treated bird. The data for the treated matings are given in table 1. The arrangement and meaning of the captions of this and the next table need some explanation. In the column headed ‘Hggs set’ ig given for each mating the total number of eggs which went into the incubator from that mating. The percent of infertile eggs (5th column) is calculated on this total. An infertile egg is one in which there has been no union of sperm and ovum, ie., no zygote formation. In the ‘Died in shell’ column is given . the number of embryos which, having started development, died before actually hatching. The ‘Per cent of embryos dying in shell’ is calculated on the basis of the number of fertile eggs, according to the following formula: 100 (Died in shell) (Eggs set) — (Infertile) = Per cent of embryos dying in shell Other captions are self explanatory. The control data for comparison are given in table 2. From these tables the following points are to be noted: 1. All of the treated males were clearly entirely potent. This is proved by the fact that certain of the matings in each sub- 244 TABLE 1 RAYMOND PEARL Data on the fertility and hatching of eggs from treated parents in 1915 E 4 g yw = a 5 5 Be | oS NATURE OF MATING 5 5 2 z : Baa a ahs AR Boles |e \eaol . eles ae ealuene 2116) 233/917 |51.5| 7) 48.8) (99) 56. 3\2723 2 , P 22. % : Bthyl of X untreated @sovs..... PU] 2) 2 |10-0) 4 22.2) 14 | 77.470. ZNO ets eA | 2 2275 Oe dalle 2 Aeall ae Totals: Ethyl sire only............. 111] 23 |20.9| 37] 42.0] 51 | 58.0/45.9 2112) 28 15 |53.6| 2] 15.4] 11 | 84.6/39.3 24) 2 : 4 2 8/75. Eo ween al ian 2113] 24) 2 | 8.3} 18.2) 18 | 81.75. 2115} 27) 19 |70.4) 2) 25.0) 61) 75.0/22.2 Totals: Ethyl sire and dam.......... 104 56 [53.9] 13} 27.1) 35 | 72.9|33.7 Grand totalsestithy) olen sse eee rn 215) °29 36.7| 50) 36.8] 86 | 63.2|40.0 fe 281] 18 |64.3] 5] 50.0} 5 | 50.0/17.9 Methyl o X untreated @......... 2124) 32 | 14 |63.6) 3) 37.5) 5 | 62.5122.7 2125) 381} 9 |23.7| 2] 7.4) 26 | 92.9168.4 Totals: Methyl ¢ only.............. | 88 | 41 [46.6] 10] 21.7] 36 | 78.3/40.9 fiat) ae 50.0| 1 | 50.0) 2.9 Methylect < methyl 9.-22.-.7.52. 2121) 24 | 9 |37.5) 3] 20.0] 12 | 80.0)50.0 2122) 27 | 26 |96.3) 0} 0.0] 1 |100.0) 3.7 Totals: Methyl sire and dam........ 85 | 67 |78.8| 4| 22.2] 14 | 77.8|16.5 Grand totals Methylicises ss. 02-0 20e {| 173/108 |62.4] 14| 21.9] 50 | 78.1|28.9 2104] 32} 0 | 0.0] 15| 46.9| 17 | 53.1]53.1 2 De a ; 46 .2|25 .0 Hither oc < untreated 9... 2). .2u oe a . ae ie ae ss Pape 2111} 29 0} 0.0} 2) 6.9) 27 } 93-1)93"1 Rotalss Bither co? only? i235... 95 ne 115] 15 |13.0} 39] 39.0] 61 | 61.0/53.0 ee 35 | 34 [97.1] 1/100.0/ 0| 0.0) 0.0 2106) 25 7/72. Etherwet SGether” 0 ...4. cn 5. eee ae 35 : a é bie ie She : 2108} 221} 7 31.8} 3] 21.4) 11 | 78.6/52.4 Totals: Ether sire and dam.........-. 104} 51 |49.0} 15} 28.8) 37 | 71.2/35.6 Grand ttotalse Either cit... eee 219) 66 |30:1} 54) 35.3} 98 64.1/44.7 Crandon ei a 314| 79 |25.2| sol 36.6] 148] 68.0147.1 untreatedmovO!< «. tye te ilar Gran tae at ened $7 XY) Vg rn 2 of 0.9 0 | 1QOne egg broken during incubation. PARENTAL ALCOHOLISM AND THE PROGENY 245 TABLE 2 Data on the fertility and hatching of eggs from random sample of untreated BPR 22 X untreated DoS in 1915 ee PER CENT MATING . ERM ICEND prep EN) eee PERTIEN | PLE CENT NuMBER | £248 SET |INFERTILE ee anit, puive rs HATCHED Bac Enea 1956 27 16 59.3 6 54.5 5 45.5 18.5 1957 28 15 53.6 8 61.5 5 38.5 17.9 1958 39 0 0.0 15 38.5 24 61.5 61.5 1959 17} 8 47.1 8 100.0 0 0.0 0.0 1960 7H 1 3.7 5 19.2 21 80.8 77.8 1961 38 2 9.3 3 8.3 33 OG 86.8 1962 41 20 48.8 15 71.4 6 28.6 14.6 1964 36 2 5.6 12 35.3 22 64.7 Glen 1965 31 6 19.4 5 20.0 20 80.0 64.5 1968 35! 0 0.0 15 44.1 19 55.9 55.9 1971 3l 5 16.1 12 46.2 14 53.8 45.2 1972 33 0 0.0 18 46.7 15 53.3 53.3 1973 40 11 27.5 4 13.8 25 86.2 62.5 1974 37 37 100.0 0 0.0 0 0.0 0.0 1983 20 a 35.0 5 38.5 8 61.5 40.0 1984 22 1 4.5 3 14.3 18 85.7 81.8 1985 36 12 33.3 18 75.0 6 25.0 16.7 1986 36 5) 8.3 24 (2.7 9 27.3 25.0 1987 29 0 0.0 12 41.4 IN? 58.6 58.6 1988 45} 3 (ar 14 31.8 27 68 .2 61.4 1989 28 11 39.3 7 41.2 10 58.8 35.7 1991 41 6 14.6 22 62.9 13 37.1 31.7 Motais: .|) 717 166 23.3 231 42.2 317 57.8 44.4 1 One egg was broken during incubation. division of table 1 show a high degree of fertility. Thus in the case of the ethyl # mating 2118 gave no infertile eggs out of a total of 25 eggs set. Out of these 25 eggs were hatched 21 good chickens. In the case of mating 2113, with an ethyl °, this same ethyl & had but 2 infertile eggs out of a total of 24 set. Anyone who has had experience with poultry knows that records of this sort are not to be obtained with impotent or defective males. The methyl & made no record quite so good as these of the ethyl @ in respect to fertility of eggs. Mating 2125, however, with 9 eggs infertile out of 37 set shows that the male was not seriously defective in his ability to fertilize eggs. 246 RAYMOND PEARL > The ether & has two records of perfect fertility with untreated females (matings 2104 and 2111). With an ether ¢@ he gave, in mating 2106, only 4 eggs infertile out of a total of 25 set. 2. Considering totals and averages it is clearly evident that the inhalation treatment of the females seriously reduces the proportion of fertile eggs which they are capable of producing. Taking grand totals 59.2 per cent of all eggs set were infertile when the mating was of the type “‘treated @ X treated 92,” as against 25.2 per cent infertile for matings of the type ‘‘ treated o xX normal, untreated ¢,’’ and 23.2 per cent for matings of the type “‘normal, untreated & X normal, untreated ¢.” It thus appears that the treated females gave rather more than twice as many in fertile eggs as untreated females in the same season and under the same conditions except for the alcoholic treatment. Calculating the correlation coefficient between germ dosage index and percentage of eggs infertile we get r = +0.316 +0.136 This is a relatively large coefficient and almost certainly signifi- cant. It means that the higher the germ dosage the greater the proportionate failure of the germ cells to form zygotes. This defect in the percentage of fertility in treated females’ eggs might, on a priort grounds, conceivably be due to any one of three general sets of causes, viz., (a) that the oviduct of the treated females formed a less favorable environment for the sperm than the oviduct of the normal female, or (b) that the eggs of the treated females were themselves adversely affected by the alcohol, so that in respect either of their chemical or physical condition or both they were less capable of being fer- tilized than the eggs of normal untreated females, and at the same time the sperm of treated males was less capable of fer- tilizing, or (c) that there was an assortative mating operating against treated females, and in favor of normal, untreated females in the same breeding pen. ‘There appears to be no doubt that the principal cause of the reduced fertility is the second one mentioned (b). The correlation between germ dosage and infer- tility indicates that a certain proportion of the germ cells which PARENTAL ALCOHOLISM AND THE PROGENY 247 would form zygotes under normal circumstances are definitely put out of commission by the treatment. While such inactiva- tion of germ cells is undoubtedly the primary factor in reducing the fertility, the other two factors also play some part. All the evidence at hand from observations on the behavior of these birds indicates that factor (c), preferential mating in favor of untreated females, is second in importance to (b) in reducing the percentage of fertility of the eggs of treated females. Observations of the males in the breeding pens indicate that the alcoholized females are not sought by the males with either the eagerness or the frequency that the untreated females are. Just what is the basis of this preference does not yet appear. Factor (a) probably plays a minor part in the diminished fertility of the eggs of the treated females. 4. Comparing the different males. it appears that, taking the data from all matings both with treated and untreated females, the ether & gave the highest fertility (69.9 per cent fertile), the ethyl & the next highest (63.3 per cent fertile), and the methyl 3 considerably lower (37.6 per cent fertile). All of these figures are lower than the mean of the random sample of 22 normal matings of untreated males with untreated B. P. R. females recorded in table 2. Those matings gave 76.8 per cent fertile eggs. There appears to be no doubt that the process of zygote formation (fertilization) in these treated matings was signifi- cantly impaired by the treatment in every case. This impair- ment was most serious in matings of the methyl ~. His best mating (2125) gave a percentage of fertility, 76.3 per cent slightly lower than the average for a random sample of matings of un- treated matings in the same season. All cross-bred matings of untreated males and females in 1915 gave 536 eggs infertile out of 2628 set, or 79.6 per cent fertile. The methyl <’s percentage is significantly below this. 5. The percentage of fertile eggs hatched is higher for the matings of treated males, whether mated with treated or un- treated females, than for the random sample of normal matings in which both parents are untreated. It is also higher in these treated matings than in all cross-bred matings in 1915, which 248 RAYMOND PEARL gave 1122 chickens from 2092 fertile eggs, or 53.6 per cent. The figures for the treated matings, 63.0 per cent for treated 7 xX untreated @ 9, and 72.3 per cent for treated ~o X treated 2 ©, are significantly higher than the normal hatching records of the same year. These figures appear to demonstrate that the alcohol (or ether) treatment had no deleterious effect upon the hatching quality of the eggs of treated individuals, On the contrary the fertile eggs of such individuals hatched distinctly better than eggs of normal individuals. The correlation between total germ dosage index and per cent of fertile eggs hatched gave r = +0.288 + 0.138 On account of the small series the probable error is large and too much stress can not be laid upon the result. On the other hand, small as the series is, the coefficient is more than twice its probable error, and is to be regarded as probably significant. So taken, it means that the larger the germ dosage the larger the proportion of embryos which batched, or, put the other way about, the smaller the prenatal mortality. 6. The last column of each table gives the percentage of chicks hatched to all eggs set. For matings of treated 70 xX untreated 9° 2 the final average percentage is 47.1. This is a higher proportion of chicks hatched to eggs set than is given either by the random sample of normal matings in table 2 (44.4 per cent), or by all cross-bred matings of 1915, in which 1122 chicks, or 42.7 per cent, were hatched from 2628 eggs < 525) Sex differences Treated & X untreated 2 2 series: 9 — o = 0.13+0.39 gram. Treated ~ X treated 92? @ series: 2 — o = 0.20+0.54 gram. Control series: 9 — & = 0.49+0.44 gram. From these figures the following conclusions appear to be warranted: 1. In the present series of experiments there is no significant difference in mean hatching weight between the offspring of treated males and the offspring of normal untreated control males when both are mated to normal untreated females. The slight differences which do appear are of the same order of magnitude as their probable errors. 2. Both the male and the female offspring of matings in which both parents were treated have a larger mean hatching weight (i.e., are heavier when hatched) than the offspring of either completely normal control matings, or of matings in which the father only is treated. The differences here are relatively large and are statistically significant in comparison with their probable errors. The figures in parentheses following the difference lines give the ratio Diff./P. E. diff. These figures range from 3.8 to 6.4. From the table given by Pearl and Miner (24, p. 88) it appears that the odds against the fortuitous occurrence of devia- tions as great as or greater than these range from about 95 to 1 to about 20,000 to 1. They may fairly be considered real and significant differences. PARENTAL ALCOHOLISM AND THE PROGENY 269 3. The mean hatching weights of the females are in all three series slightly greater than those of the males. The differences, however, are entirely insignificant in comparison with their probable errors. Turning to the consideration of relative variability in hatch- ing weight, as measured by the coefficient of variation, it appears that the males’ are more variable than the females in all three series. The differences, however, are not very large in propor- tion to their probable errors, and can be regarded only as some- what doubtfully significant. There appears to be no evidence that the relative variability in respect of hatching weight has been in any way affected by the alcohol treatment. VIII. GROWTH OF THE F, PROGENY Growth, as measured by increase in body weight, is univer- sally recognized by physiologists and by practical animal hus- bandmen as one of the most valuable indices of innate consti- tutional vigor and vitality which it is possible to obtain. On this account it was thought to be of first-class importance to study the growth of the offspring from aleoholized as compared with untreated parents. The frequency distributions eiving the raw data on the growth of the animals in these experiments are exhibited as Appendix Tables I to XV inclusive. The constants deduced from these distributions are given in tables 11, 12, and 13. Regarding the collection of the growth data it may be said that weighings were made at regular intervals according to a fixed schedule. For purposes of analysis weighings for ages differing by but few days were grouped, and regarded as concentrated at the central age of the group. ‘This amounts essentially to a first smooth of the material. It is not our intention in the present paper to enter upon any discussion of general problems of growth. It is desired here merely to make a comparison between the offspring of aleoholized and those of non-alcoholized parents. It will be noted by comparison of tables 11 to 13 that the centered ages for which mean body weights are given are ip RAYMOND PEARL 270 LG" “GE = 0G 6006 13° 8g" 90° PL Le cy OS” &1 c il 0 +76 S¢ 0 0 26+ 90 0806 I€ +79 996T 9T = €8 G89T €1+6I LZSP1 S1+9P SOIT GIF8E GSS 6 22 69S G +09 608 El PLL + 10° €8 +89 IP +=F0 GE v8 8909" Sh 98+ SG" 89 &6*6L° 9€° 8T+0¢" 89 GI +€0° IZ IIT +98 186 16:8 Is 9 GG & So I €8°0 cP 0 66 0 +29 169 == GS +GL I8T +90 78 Og SS +6 OF +16 PE 6FLG OLSG £966 L981 8661 C'06Z G 40% ¢'T61 ¢ 99 GST ¢ O11 C68 ¢'89 G LF c' ee C61 ¢'ZI Gg (SurIyoyV A) O 96° 0+*F0'9 0S 61+ €9 Gol v6 T+08'S 62° 1+*60°6 V8 06+8P OTT 89° 8h += 00 096 9T' T+¥0'6 LO T2716 VS 66*08 9LT 64 GSEFCL GEG 02 OEP OT GL 09801 GL IL-0 S21 VL OL GL Sh v9 0+8S'6 02° 01801 6 6 +19 68T 86 GI +9F COI 88 0+*G0 €1 08 088° IT 1Z'6 +28 OFT 60° TT+*G0 99T KO) Se Si7 SSL 98° 0+ F9 CI IZ 8 +99 I§1 86°8 G0 FI 66 180° LT 96° 088 ' 1 89°9 60°96 o8 6 = SL468 8&° T+ €0 06 OF T9961 Ol F 00°29 9h F +62 49 Ih 168 06 66 TPL 81, GEG =SG, 96 866 10 FE I€ T+*GZ 61 && LGZ 61 G0 T OP 91 80 T +81 9T COMEETGEST 80 [2s 9T of 0 +39'8 8¢'0 +00 6 18 O22 CI 08 O+8T €1 80 +=1e"s c&é 0 *6E SG 9 0+€0 6 89° 0*82'6 SLO) Ze 0G 0 GF € ) Le fo) Le 6 © NOILVIUVA JO LNAIIMIAO9 saidas & 6 paypauyun K papas ayz ur Bursdsffo fo qybram fipog ur yyno0wb sof syunjsuod UwoYynrw A (SNVUD NI) NOILVIAGG GUVAONVLIS (SNVUD NI) NVAW SAVd NI ADV Il HIaAVL 271 PARENTAL ALCOHOLISM AND THE PROGENY G3 1-69 GI 86 8E+ IG 99 PI PS+ES L606 G 062 G9 I+02 6 SS) USS 72 Ol SCE 96I 6h LV+P9 G6C | G8 9F+0G'FC0G | 91 2909 6886 ¢ P06 pe) OSite 0) tes] 28 066 9 18 TE+ TE 8Sc 09° €¢+6E TL | 86 THES C8EI | LE SE+0S FEL ¢ F6r 3 0+99 6 AD [Lh 10! I€ VI+8s POT 1 €G*06 296 | &'OZFEB COLT | GS EE+ZO STEZ% G 900 68 0+ 6F 6 iG Wasi) tel! PI GI+OP 6ET 0O'E6+ce GGc | LI LI*OG 69FT | &9 CE+E6 OL8T ¢ 8&1 88 0+ €0 TI 6 106 FI GO OTOL SIT Ly 8I+6I 106 | ZL PI¥E8S LZPIT | CI 9260 LIFT ¢ O11 SOS ssa [9 T= 62, OF 96°8 *98°COT GL OTT CLT | 29 CT =E8 298 61°66 CF 9Z0T ¢ 68 60 I+¥€I GI 8€ T+*c0 ST 1G 9 ZG Lz 868 29°66 €6°8 E&P L8G OL GI*6L €99 ¢°39 Ig 196 Cl OF I-21 SI 8 & =P IV 067 CE FS he9, +99) 216 669 20 89E Gc LV OF Ta SI LG GBP EG GPG +8896 SLY =ZG Gt. Sy & =P9i 921 669 E&I F6I g's ¢9 T+00 81 6G L=9F ZI 6é_I +68 FI xe 10 See) tei! 98°T +*69°c8 96 T +IP 68 G61 G6 026 OL Z6 1+O1 FI 1g 0 +96°S G0) = 9Is8 GL 0 +09 'FS 60 1 =PF8 Lg COL 16 O88 OL 66 OF TP OL 66°0 ¥f9'F Iv 0 +29 7 G¢o°0 +89 Cr 69°0 +16 HF ca Go 0-6 9 L830 €8 6 060 89% G& 0 +9 € 66 0 ¥ZT LE Sp 0 +26 9€ (suryo38H) 0 6 £ 6 £ é e : SAVd NI ADV NOILVIUVA JO LNAIDIGAIOO (SAVUD NI) NOILVIAGG GUVANVLS (SHVUD NI) NVAW savas & 6 paypay K L paynas ayz ur bursidsff[o fo yybram fipog ur yynosb sof szupjswoo uornriv 4 ol WIAVL RAYMOND PEARL 272 eS ee See SS SSS SS ¥0 0¢° 10° SL 99° 98° 89° 90° G6 68° 92° ceils L8 &v 61 ST 9¢° ols 0€* 69 Og ‘T+#68 OL T+62 OT L027 01 OOF IT O18 6 OFGoecl OCP 6 T+82Z OT 009 GI 0+0E GI T+€6 LT I+¥6 OL Oo 6 T+08 1 T9681 72 ie | [+8 9T I+€o IT T+ tI ‘009 ‘069 2 co) von 66 98° OV" 66° 1S GE 9¢° G0" 9c" FG" OF GL 68° 7 a eo" a 66° LL OFS6 2 0918 0&1 OI I+=16 9I T+9¢ FI T+29 IT I +69 ST I+=II ZI C16 LT T+69 OT T+06 IT 9G ST T=0OP ST T+0¢ FI T+62 ST I+18 TI I+0L €1 O=-ZI ZL 0+69'6 © NOILVIUVA JO LINGIIddAHOO G8" 0G 8S 81 26° F6+ FO 606 06° 8T+ TIT 861 FP EL+8s 661 vO OLE OST 9c OL+89 EST 86°9 PE 18 9L°L 69°62 08'S +99 EZ 88°§ eo LP c8'€ *&0 OF G9"S G0 92 a 9¢°T 96 8cl SI'S *PT 801 02'S *6é 16 WIT 60 TL VET PS 29 68°0 +96 7S vL°0 62 OF 820 86 SE 60 FEL VE é 88°66 LO E9FG 8h LO F9 IETS GP CG* 19 O98T OT 26+ &6 SOP L8°6T +21 T80T LG LI IG SE8 TZ’ TL 60 S¥9 £6°8 OP 9IP €9°9 GG SEs 0G ¢ *G0 6&6 88°S EG LET S°€ +14 LOT VES CE 86 SZ°T +00 29 y3'T +09 $9 60°T +62 TS 98°0 +02 TP GPO *€E CE LEO VC VE © (SWVUD NI) NvaW 0° 982 0° 19% S60 C'PLI 0° ¢9T CLF ¢'6IT 0°&6 CLL 0°29 0's¢ ¢'0F 0°28 G°9Z 0°82 ¢ 61 0°91 CZ 0'6 gg 07% (SurIyoye Hy) 0 SAvd NI GOV sawas & & payayun X OO paywa4jun 10.ywo9 poutou ay? ur Buridsffo fo yybram fipog ur yyno1b LOf SJUDISUOI WOYDULD A el ATAVL PARENTAL ALCOHOLISM AND THE PROGENY 273 many cases not the same for the treated and the untreated series. On this account as well as othefs a comparison between the different series in respect to growth can best be made by graphical means. To this end figures 2 and 3 have been pre- pared. It will be noted from these figures that the weighings were carried into adult life in the case of both sexes. They stop somewhat earlier in the case of the males than the females owing to considerations of space on the plant. It scarcely needs to be said that the utmost care was used to ensure accu- racy in this growth work, and to keep the surrounding condi- tions of housing, feed, etc., in every respect favorable to normal growth. The latter point is, of course, a very important one for critical results. The favorableness or unfavorableness for animal growth of the general environmental conditions, the season,’ etc., is not a thing which can be exactly measured, but, it may be judged by one experienced in the husbandry of the particular animals dealt with. On such general observational grounds one would have judged that the season and general environmental conditions surrounding the chickens in 1913, the year in which the control chickens ex untreated 7 X untreated 2 2 were grown, were more favorable to growth than in 1915, the year in which the offspring of aleoholized parents were grown. If such was in fact the case, and my observations lead me to believe that it was, the results obtained, as will presently appear, take on added significance. It should be particularly noted that no question of artificial selection can enter as a factor in influencing the growth data for the simple reason that no such selection was practiced. Each and every individual chicken was regularly weighed as long as it lived or until it reached maturity and the weighing records came to an end for the season. As the mortality was small it practically means that there was no selection at all between hatching and maturity. In the case of weighings of females at ages over 210 days there has been a selection. The indi- viduals at these higber ages are the ones kept over for winter egg production and to be usedias breeders the next year. Great care was taken, however, to ensure that the samples chosen THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 BODY WEIGHT 274 RAYMOND PEARL 0 25 50 75 100 125 150 175 200 AGE IN DAYS Fig. 2. Diagram showing the growth in body weight of the male offspring of alecoholized parents as compared with the male offspring of untreated parents. Solid line, offspring of untreated 7 X untreated 9 9; dash line, offspring of treated oo’ X untreated 2 @ ; dot line, offspring of treated 7 X treated 2 9. 275 PARENTAL ALCOHOLISM AND THE PROGENY ‘Z HNSY Ul SB SOUT] Jo dULOYIUSIg ‘sj}uorvd payvarjun Jo Sutdsyo sajvurey oy} yJIM posed -W09 sv syuored pozTjoyoo]e Jo Sursmdsyo oyeurey ay} Jo yySIoM Apo UI YYMOIS oY} Surmoys weisviq gE ‘3yqT SAV NI aDV 4 0Sz Szz 002 SLI ost Szl 001 Sc 0S Sz 0 00s yal LHDIAM ACO 276 RAYMOND PEARL should be random ones and there is every reason to believe that they were. If anyone, however, hesitates to accept them as such or feels that a greater accuracy necessarily inheres in unsampled material he should confine his consideration of the growth results to ages under 200 days. From the data in tables 11 to 13 and figures 2 and 3 the fol- lowing points appear to be well established: 1. The offspring of aleoholized parents, whatever the nature of the mating, showed a higher mean adult body weight than offspring of untreated parents of the same breeds mated in the same way. This is true of both sexes. 2. In the case of the male chickens there was no substantial difference in the rate of growth in the three lots until after an age of about 100 days was passed. From that point on the male offspring of treated @ X untreated and treated 9 @ grew at a more rapid rate than the controls. The differences in mean body weight for a given age became increasingly large as the age advanced. At 200 days of age we have, by inter- polation on the curves, the following set of comparative mean body weights. Comparative mean body weights at 200 days of age Absolute weight Relative weight Males ex untreated oo X untreated 99 2392.32 gram 100 Males ex treated oo X untreated 22 2668.97 gram 112 Malesex treated ic X treated 92 2815.25 gram 118 3. In the case of the female chickens there was no substantial difference in the rate of growth in the three lots until after an age of about 150 days was passed. During the next 25 days the controls grew faster than the chicks from treated parents. At and after 200 days of age, however, the offspring of treated parents (one and both) showed a higher mean body weight than the controls. We have the following set: of comparative mean body weights at 250 days of age, obtained by interpolation on the curves. PARENTAL ALCOHOLISM AND THE PROGENY 1) This statement may appear difficult to reconcile with statistics given in tables 1 and 5. It is shown in table 1 that 234 chicks hatched, whereas accord- ing to table 5, but 215 were banded. What happened was that, in order to get more extensive statistics on the hatching quality of the eggs, another hatch was brought off on May 12. This produced 18 normal chicks, which were not banded, but immediately discarded. These chicks hence appear in the hatching records, but not in the subsequent data. We have 215 normal banded chicks + 18 normal, but unbanded chicks + 1 abnormal = 234 hatched. PARENTAL ALCOHOLISM AND THE PROGENY 279 The data from the alcoholic matings of 1915 are as follows: Non-barred Barred (self black) ING G6 i RE ANE, | 5 2 a nee Re DAES oe, RTL Go) 95 0 Hemraleseeeyrise: Uh enc os osc soso chdn Lee eee. 0 104 Sioa role ah, A at gs oo cd 6S RO ED ORE coe ceed eo 8 8 The normality of these results is beyond cavil. The results regarding comb form are equally clear. The Black Hamburg is a rose-combed breed, the Barred Rock a single- combed. Rose X single normally gives rose in F;. Out of 215 chicks from the alcohol matings 215, or all, were rose-combed. Similar data might be given for various other characters. They all are the same, however, in principle, simply showing that the normal regular course of Mendelian inheritance has in no wise been altered or interfered with by the alcoholization of the parents. XI. DISCUSSION OF RESULTS Before attempting to discuss any interpretation of the meaning of our results it is first desirable to do something in the way of summarizing them so that a clear and definite picture of the net result may emerge from the mass of statistical data pre- sented in the preceding pages. Such a summary is given in table 14. The casual reader and the hostile critic are strongly urged, however, not to regard table 14 as the only evidence for the conclusions reached. This table aims only to summarize fairly the net results of the detailed statistical evidence in the body of the paper. The plan of this summary table is as follows. The superior result is printed in bold faced type. In the last column of each table a plus sign denotes that, with reference to the particular character discussed, the progeny of the alcoholists have been favorably affected; a minus sign that they have been unfavorably affected as compared with untreated controls. A zero indicates that no effect of the treatment, one way or the other, has been detected. We see from this table that out of 12 different characters for which we have exact quantitative data, the offspring of treated par- 280 RAYMOND PEARL TABLE 14 Showing in summary form the effect of continued administration of alcohol (ethyl and methyl) and ether, by the inhalation method, upon the progeny TREATED TREATED ALL NET CHARACTER STUDIED oNtaEATED| Tamas | TEATED |"Covmmors | ALCOHOL. 29 ge OFFSPRING 1. Mean germ dosage index..| 137.8 299 .0 210.35 0 2. Percentage of infertile eggs (i.e., in which no embryos were formed).. 25e2 59.2 41.7 25.3 = 3a. Percentage of embryos dy- ine imeshellae ae eae ee 36.6 26.9 Sono 42.2 + 3b. Percentage of fertile eggs which hateched.......... 63 .0 12.3 66.7 57.8 4. Percentage of all eggs which hatched.......... 47.1 29.4 38.6 44.4 - 5. Percentage mortality un- ; der 180 days of age...... Piles 10.6 17.6 36.9 6. Percentage mortality over 180 days of age.......... 5.9 13.6 10.3 15p3 + 7. Sex ratio: per cent c'c’... 48.9 45.5 47.7 50.0 0 8. Mean hatching weight per bird, males.. 34.91 36.97 34.24 + 9. Mean hatching oetiar per birgstemales.,..s. ses: 35.04 Hl sal) 34.73 +4 10. Mean adult weight per pind, smales 4.8 ae ost. 2669 2815 2392 + 11. Mean adult weight per bird, females. . a 2020 2063 1928 + 12. Percentage of ar ae deformed chicks......... 0.7 0 0.4 1.0 + 13. Abnormalities of Mendel- jan mbheritance.......... 0 0 0 0 0 ents taken as a group are superior to the offspring of untreated parents in 8 characters. The offspring of untreated parents are superior to those of the alcoholists in respect of but two char- acters, and these are characters which are quite highly corre- lated with each other. Finally with respect to two. character groups there is no difference between the alcoholists and the non-aleoholists. The character groups which have been dealt with in this study, and for which definite quantitative data are herein presented, seem to cover a much wider range of physio- PARENTAL ALCOHOLISM AND THE PROGENY 281 logical and genetic factors and phenomena than has ever been included, or even approached, in any previous study regarding the effects of parental aleoholism upon the progeny. They have the further advantage of being characters which are measurable (either statistically or otherwise) and on that account greatly reduce, if they do not entirely eliminate the possibility of per- sonal bias or prejudice influencing the results. For example, to take but a single instance, the weighings for growth records were made by my assistant Dr. Maynie R. Curtis, in the case of all progeny of alcoholic parents, and for all ages except hatch- ing where the writer himself did the weighing. The weights made in the field by Dr. Curtis were posted into the permanent record books by another assistant, Miss Hazel F. Mariner, and were reduced to the form of frequency distributions some six months later by the staff computer of the laboratory, Mr. John Rice Miner. Not until after all the constants for these alcohol distributions had been completely worked out were the control data from the untreated matings of 1913, where all the weigh- ings were made by the writer, put into the form of frequency distributions. No one of these workers could possibly have known, at any stage in this process before the final one, what the net result in respect to growth was going to be. Similar considerations obtain in regard to the other characters. The mutual accordance of the results from characters involv- ing such a manifold range of physiological factors is striking. This fact in considerable degree offsets the fact that as yet our series of experimental animals is statistically small, leading to such large probable errors that the individual differences are not in every case significant in comparison with their probable errors. The experiments are of course being continued and expanded, and concurrently the probable errors of differences are being reduced. If results in the same sense as the present ones continue to appear (as seems to be the case) they are bound presently to become very convincing to such persons as are not prevented by prejudice from accepting or appreciating scientific evidence on the problem of the effect of parental aleoholism upon the progeny. 282. RAYMOND PEARL We may evaluate our results in general terms as follows: 1. There is no evidence that specific germinal changes have been induced by the alcoholic treatment, at least in those germ cells which produced zygotes. 2. There is no evidence that the germ cells which produced zygotes have in any respect been injured or adversely affected. 3. The results with poultry are in apparent contradiction to the results of Stockard, Cole and some others with mammals.® I believe, for reasons which will presently appear, that this con- tradiction is only apparent and not real, paradoxical as such a statement may seem. 4. The results with poultry are in a number of important respects in essentially complete agreement with those of Elder- ton and Pearson (7) on parental alcoholism in man, of Nice (16) with mice and of Ivanov (12) with rabbits, guinea pigs, dogs and sheep. Elderton and Pearson (p. 32) summarize their investigation in the following words: To sum up then, no marked relation has been found between the intelligence, physique or disease of the offspring and parental alco- holism in any of the categories investigated. On the whole the bal- ance turns as often in favour of the alcoholic as of the non-alcoholic parentage. * T make no mention of the results of Ceni (2) with fowls simply for the reason that his work on this question seems to me so uncritical that I am unable to con- sider it seriously. His method of administration of the alcohol, the very small number of animals dealt with, the absence of any quantitative data regarding the progeny, the total absence of controls, and the obviously pathological ele- ment in the stock and the experiment generally, are sufficient grounds, it seems to me, for regarding Ceni’s contribution as of no significance, either one way or the other, in the discussion of this problem. If it were worth the space in this journal I could show point by point in detail wherein Ceni’s paper is without real meaning. Furthermore after a careful study of this and other papers by the same author I am confident that I could repeat his alcohol experiments with poultry and get the same results. If one applies some deleterious agent plenti- fully, and then so manages the environmental conditions that the birds have not even a fighting chance for normal, healthy life it is easy to produce a very sorry lot of chickens. Ceni’s investigation obviously started from the point of view that alcohol was going to harm his chickens. From the account given it is diffi- cult to conceive how that happy consummation could have been avoided under the conditions. PARENTAL ALCOHOLISM AND THE PROGENY 283 They further find (p. 31) that: The general health of the children of alcoholic parents appears on the whole slightly better than the health of the children of sober par- ents. There are fewer delicate children and in a most marked way cases of tuberculosis and epilepsy are less frequent than among the children of sober parents. The source of this relation may be sought in two directions; the physically strongest in the community have probably the greatest capacity and taste for alcohol. Further the higher death rate of the children of alcoholic parents probably leaves the fitter to survive. Nice (loc. cit., p. 146) summarizes his studies of the effect of parental alcoholism upon the growth of the progeny in the fol- lowing way: “Although the young of the alcohol mice when given alcohol themselves excelled all the other mice in growth, other young of these same mice [i.e., of alcoholic parentage] when not given alcohol grew even faster.’’ Ivanov, in his experi- ments on artificial insemination, has obtained normal offspring, which lived and made normal growth, from rabbits, guinea pigs, dogs, and sheep, where the spermatozoa used to fertilize the female were actually immersed at the time of fertilization in solutions of ethyl alcohol ranging in strength from 0.5 per cent to 10 per cent. These various results are not to be dismissed in so light ahd cavalier a manner, and without reasons given, as they have been in some recent reviews of the literature. The memoir by Elderton and Pearson is a masterpiece of statistical research, sane and temperate to a degree in its conclusions. Nice’s study of mice seems to be a sound, thorough and careful piece of experimental work, quite the equal in respect of its technique and its logic, of any experimental work which has been done in this field. In attempting the interpretation of these results we are con- fronted with several possibilities. In the first place it might be maintained that there are fundamental physiological differences between birds and mammals, of such extent and degree as to make the action of alcohol and similar substances upon the germ cells totally different in kind in the two cases. While such a possibility can not be categorically denied, it seems to me 284 RAYMOND PEARL to be highly improbable. In the first place my results agree fully with Stockard’s so far as concerns the effects produced upon the treated animals themselves. This would imply that the physiology of reproduction is alone so different in kind in birds and mammals that it is differentially affected. But it seems to me that all we know about the matter agrees in indi- cating that the fundamentals of ovarian and testicular physi- ology are essentially the same in birds and mammals. A second possibility is that while the effect of alcohol upon the germ cells is the same in kind in birds and mammals, it differs markedly in degree in the two cases, the germ cells of birds being much more resistant to injury by alcohol than those of mammals. This seems to offer a valid explanation of the apparent discrepancy in results, and I shall return later to a more detailed discussion of it. A third possibility is that what has here been called the total germinal dosage (cf. I), is too low to produce any effect, and that with higher dosage harmful results would have mani- fested themselves. ‘There are three things bere to consider. One is that the average duration of treatment prior to the birth of offspring (i.e., germinal dosage) is certainly as long, and per- haps even longer than that which has been definitely shown by Stockard to be necessary to produce deleterious results in guinea pigs. While Stockard has not, so far as I am able to find, any- where given definite figures for total germinal dosage in con- nection with particular individual matings, it is clear from the general context of all his papers that a few months continuous treatment of the parents with alcohol prior to conception is amply sufficient to injure the germ ceils to the point where defective offspring are produced. Jn some cases he apparently has got results with very short duration of treatment. Thus he says (35, p. 656): ‘‘A number of experiments in which the treatment of a female was commenced at the beginning of pregnancy have so far given rather indefinite results, although a slight effect may be indicated.” This of course is a great reduction of germinal dosage and no one could expect marked or definite results. Stockard points out in many places in his PARENTAL ALCOHOLISM AND THE PROGENY 285 papers that as the duration of the treatment is prolonged the injurious effects on the germ cells get worse. This is of course what one would expect, but the point is that he was able to show positively harmful effects on the progeny in the early stages of his experiments when the treatment prior to concep- tion had not been greatly prolonged, and the results of these matings early in the course of his experiments are, quite properly of course, included in the latest summaries of the investigations (ef. Stockard and Papanicolaou 38). Furthermore Stockard (cf. 34, p. 381) as well as Féré earlier, showed that the hen’s egg is very easily influenced by alcohol during incubation and caused to develop teratogenetically. The second point is that a careful study of the present results makes it impossible to assert that the treatment of the parents has had no effect upon the progeny, which would be logically necessary if one holds simply that the dosage has been too low to be effective. The offspring of the alcoholists, as a class, are indubitably differentiated from the offspring of the non- alcoholists. The probability that the former group is a ran- dom sample from the latter, when all 12 of the characters dealt with are taken into account, is so small as to amount to practical impossibility. The treated matings, by and large, plainly give better results in a number of respects than the controls. With all the critical precautions which were taken with the experiments this can only mean that the treatment has produced an effect. Altogether it seems impossible to explain the results of these experiments on the supposition that the duration of treatment prior to conception was not long enough to produce any effect whatever.’ It is quite clear that the validity of the present experiments can not be challenged on the 7In this connection it is very difficult to refrain from discussing the 1916 results which are coming to hand as I write. Since in adequately reporting the results of an experiment of this sort it is essential to present the original data in detail, limits of space demand that a data limit be set on progress reports. As set forth earlier, the present paper reports the results up to February 1, 1916, only. It must suffice here merely to say that after 18 months continuous daily alcoholic treatment of both parents we are still getting results in the offspring essentially like those here reported. 286 RAYMOND PEARL ground of too low germ dosages as compared with mammalian experiments until we have definite statistics regarding mam- malian experiments which shall-show for each individual sepa- rate mating, in actual tabulated figures (a) the total germinal dosage prior to conception, using the term ‘germinal dosage’ in the sense defined in this paper, (b) the number and quality of offspring. We may now return to the further consideration of the second possibility mentioned above, namely that the apparent dis- crepancy between the avian and mammalian results is funda- mentally due to a difference in degree of resistance of the germ cells to aleohol. On this basis it is possible, I believe, to frame an hypothesis which will bring together in a satisfactory manner under one point of view the apparently discrepant results of Stockard, Pearson and the writer. At the outstart let us remind ourselves of a point which one is apt to overlook in considering results of this sort, namely that the germ cells which produce the zygotes, which are the progeny of our experiments, are only a very minute fraction of all the germ cells which the parents form. Let X be the total number of germ cells (ova or spermatozoa) which the individual produces, and let a be the number which succeed in taking part in the formation of zygotes, and let A be the number which do not so succeed. Then, of course, A = X — a, or put the other way about, X=A-+a This is the fundamental gametic equation. Starting from this point let us attempt to develop, very briefly, a general theory of the action of deleterious agents upon germ cells, and then compare our experimental results with such a general theory. We know that A is enormously greater than a. There is furthermore a great deal of evidence that a is not a random sample of X, but on the contrary is a highly selected sample. To Roux in his ‘Kampf der Theile’ is to be given the credit for first pointing out what now seems axiomatic, that there is con- stantly going on a struggle for survival among the cells of the organism, the physiologically ‘best’ being the survivors. To the PARENTAL ALCOHOLISM AND THE PROGENY 287 philosophical breeder of animals nothing seems more certainly established than that this process of selection is constantly going on and is of very special importance among the germ cells. Direct and convincing observational and experimental proof of it has been given for ova by von Hansemann (9). The double mating experiments which Cole and Davis and Cole and Bach- huber have carried out, prove the same point. We may represent the general facts about variation in the vitality or vigor or survival value of germ cells with considerable probability by a diagram like figure 4. This diagram represents the whole population X of germ cells (ova and spermatozoa). On the base the degrees of physiological vigor or survival value are laid off in 10 equal intervals. 1 denotes a weak, poor sperm (or ovum) unlikely to survive or take part in the formation of a zygote. 9 denotes a strong highly vigorous sperm, practically sure to survive and enter upon zygote formation, except under the most adverse circumstances. The curve, which is the normal or Gaussian curve of errors, denotes the probable frequency of the different degrees of physiological vigor among the germ cells. Now to return for a moment to our fundamental gametic equation, X=A+4, it is obvious that both of the groups A and a may include germ cells of varying degrees by physiological vigor, a fact which we may express symbolically in this way: An Anes A. BHONO fo (Olek cd Oe G1 = Oo +. Og. ce ok g 8’ The degree of variation within the subgroups A and a will be given, as usual, by the standard deviation of those subgroups, and these standard deviations we may indicate in the usual way aso4 and og, The quantity is theoretically a very important one. It measures the relation between the variabilities of the parts of the total germ cell population which do and which do not produce zygotes. Ordinarily we may suppose that S will be less than 1, but there is no @ priori reason why it should not be 5 1. 288 RAYMOND PEARL Suppose some deleterious agent, such as alcohol, to act upon the germ cell population X with an intensity P, the value of P being less than that required to kill all the germ cells at once. It is a reasonable assumption in accordance with known physi- ological laws that effect produced by the agent will be propor- tioned to the initial distribution of physiological vigor among the cells. The weakest germ cells will be killed, those a little stronger will be severely injured and so on. It is legitimate to assume that for values of the intensity factor P which are below a certain level, say P,, there will be a certain proportion of germ cells, which because of their high innate physiological vigor are very resistant, will not be sensibly affected by the harmful agent at all. The effect of the agent upon our funda- mental gametic equation wil) be to introduce a new term. Using 1 2 3 4 5 6 7 & 9 Fig. 4 For explanation see text. primes throughout to indicate the conditions after the dele- terious agent has acted we shall have this result— X’ =A’+(a’ +0’) (ii) where A’ denotes the number of germ cells which do not take part in forming zygotes, (a’ + 6b’) denotes those germ celJs which do enter into zygote formation, a’ being those physiologically vigorous cells which are highly resistant, and form vigorous and perfect zygotes, and 6b’ denoting the less vigorous germ cells which are injured, but not put entirely out of commission, by the deleterious agent, and because of their injuries produce defective offspring. The relations between these several parts of the germ cell population before and after the action of a harmful agent are of interest. It would seem probable that the following rela- tions must hold finally. PARENTAL ALCOHOLISM AND THE PROGENY 289 A’>A, because a loses to A’ as result of the agent’s action (a’ + 6b’) 0, that is so long as any normal offspring are produced at all, the quality of these normal offspring will be higher the greater the value of P, that is the more intense the action of the deleterious agent. This is obvious on biological grounds, because the more intense the action of the harmful agent, the more intense the selection: hence only the very ‘best’ germ cells will survive and make zygotes.? For any given organism, deleterious agent, and intensity of dosage or application, there should be a stable relation between Gea and: P- Let us now return to our experimental data in the light of these theoretical considerations. ‘The essential point to Stock- ard’s results, as I understand them, is that the value of P for alcohol on guinea pigs is relatively so high that practically speaking a’ = 0 and the fundamental gametic equation has become X’=A'+0’. If a’ has any value it is very small. One gathers that after a sufficiently high degree of alcoholization of the guinea pigs a ® The symbolic proof of the point is simple. Let Ma denote the mean vigor of the a germ cells, and My the mean vigor of the a’ germ cell. Then we have Ma' > Ma because , Im and n'Ma’ = A az where 2 denotes summation between the indicated limits, n and n’ are the number of cases, Z and Z’ are frequencies, and h is any subscript greater than 1 and S m. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 290 RAYMOND PEARL great majority if not, indeed, practically all of the offspring are in some degree defective. This condition of affairs I have tried to represent very roughly diagramatically in figure 5. The cross-hatched portion of the area (A’) represents the sperm (or ova) which are definitely out of commission so far as taking part in zygote formation, either because of their inherent weak- ness or because of the action of the alcohol, or both. The singly ruled portion (b’) represents the germ cells which form zygotes but have been so weakened by the alcohol that these zygotes are in some degree defective. SOLS Fig. 6 For explanation see text. A less extreme relationship between effective dosage of the agent, P, and the resisting power of the germ cells is shown in figure 6. In this diagram it will be seen that a very small part of the total area falls in the a’ class, which represents germ cells of such high resisting power as not to be detectibly affected by the deleterious agent. In actual breeding experiments, in- volving the statistically rather small numbers which experiments with higher vertebrates must always be contented with, it would be very difficult to distinguish critically between such condi- tions as those represented in figures 5 and 6 respectively. It may be the case that figure 6 represents the conditions with Stockard’s guinea pigs actually more truly than does figure 5. It would need larger numbers of animals and more detailed PARENTAL ALCOHOLISM AND THE PROGENY 291 quantitative data than he has yet published in order to reach a critical opinion. Dr. Stockard informs me, since the above was written, that in his opinion figure 6 would more fairly represent the guinea-pig case. Now in the case of the fowls in these experiments we may suppose that the germ cells are more resistant so that the same intensity P produces an effect such as that shown in figure 7. Here relatively many more of the gametes are capable of form- ing normal zygotes (a’), and the b’ band is much narrower, indi- cating that relatively few defectives are formed. The normal offspring produced are of superior quality because the alcohol has acted as a selective agency, putting completely out of com- mission all the poorer grades of germ cells and yet not being sufficiently intense to injure the best grades. 225) © > Se atts sees \y AWS x Bo 05 S aS BxS \S N Fig. 7 For explanation see text. In the case of the guinea pigs the selective agent, on this hypothesis, acts with such great intensity that, having put the poorer grades of germ cells out of physiological commission, it also injured all the best grades to such a degree that they pro- duced abnormal or defective offspring. Elderton and Pearson’s results would appear to resemble the fowl results most closely. Figure 7 might be taken on this hypothesis to represent the conditions in man. Alcohol would appear to be a less intense germ cell selective agent in man than in the guinea pig. In this connection it is interesting and significant to note that Ivanov (11) found, by direct measure- ment of the duration of life as evidenced by movement, that the spermatozoa of the guinea pig and the rabbit possess a relatively low degree of resistance to the action of ethyl alcohol. 292 RAYMOND PEARL On this hypothesis it might be supposed that with larger administration to the fowls (higher germ dosage) or more years of drinking behind them in the case of Elderton and Pearson’s workingmen, the conditions shown in figure 7 would gradually pass over into those shown in figure 5. Possibly this-is so, but there is no evidence as yet that it is. The germ dosage index of this paper is a time index. It takes no account of the inten- sity factor in dosage because the intensity factor is a constant in these experiments. It is represented by the time spent in the tank in the saturated atmosphere each day. This intensity factor is the same in my experiments as in Stockard’s, and in both cases it represents according to his experience and mine practically a maximum value of P. Fowls can not be left much longer than one hour at a time in an atmosphere saturated with alcohol vapor. Now in view of the facts (a) that alcohol is rapidly eliminated from the system and not accumulated therein and (b) that large and repeated doses immediately affect the germ cells very markedly as shown by Todde (40), it seems to me probable that when alcohol is administered by the inhalation method the factor which determines the width of the b’ zone relative to a’ is the length of time the animal stays in the tank per day, and not the number of days it is treated. It may well be that the longer the treatment is continued the greater will A’ become. But all the evidence now available seems -to indicate that this is at the expense of a’ alone and that b’ is simply pushed along or may even be narrowed. On the hypothesis here advanced we see why the percentage of infertile eggs is higher for alcoholic than control matings. This merely is the expression of the transference of the a, dz ye see an germ cells over into the A’ group. They are germ cells which before treatment were of rather low grade but still good enough to take part effectively in zygote formation. Alco- hol treatment put them definitely over the line into the A’ class. In this connection it is of interest to note that we have for the correlation between per cent of infertile eggs and germ dosage index r = + 0.316 = 0.136 PARENTAL ALCOHOLISM AND THE PROGENY 293 This result means that as the duration of treatment in days before hatching increases the proportion of germ cells falling in the A’ class increases... Comparing with controls as a base this further means that this change is at the expense of the cells in the a class. The percentage of fertile eggs hatched, the mortality of the offspring, the weight at hatching and the growth to adult weight are all superior in the progeny of alcoholists. These facts argue very strongly in favor of the present hypothesis in general, and in particular that part of it which postulates a group of germ cells a’ which are of such high physiological vigor as to be effectively beyond the range of the selective agent acting at intensity P. If in respect to one or two only of these charac- ters were the alcoholists’ offspring superior we might attribute the result to accident. But when the whole series shows the same thing such an explanation is out of court. No sensible person would argue that the aleohol benefited the germ cells over so long a period of time. A selective action of the sort here postulated seems the only reasonable explanation of the objec- tive experimental facts. It is known that an immediate, but transitory, stimulating effect may follow the administration of very minute doses of substances which in higher dosages act as poisons. This was shown to be the case by Braconnot, work- ing about the middle of the last century on the effect of various substances upon the sensitive plant Mimosa. Czapek (5, p. 883) in commenting upon Braconnot’s result especially mentions that, in spite of its potentially great importance, no later investigator has systematically followed it up. Prof. C. M. Child informs me that he has obtained exactly the same kind of a result in his studies on the effect of such substances as KCN and alcohol upon planarians. In the case of the present experiments, how- ever, it could hardly be maintained that this primary stimu- lating effect of a dilute poison would continue regularly and constantly to appear after the poison had acted on the same individuals continuously for months! For such a supposition there appears to be no warrant in any known biological facts. 294 RAYMOND PEARL XII. SUMMARY This paper deals with the effects produced upon the progeny of fowls after treatment of the parents with either (a) ethyl alcohol, or (b) methyl alcohol, or (c) ether. The chief results may be summarily stated as follows: 1. The proportion of fertile eggs, i.e., eggs in which a zygote was formed by the union of sperm and ovum, was materially reduced in the matings in which one or both individuals had been treated. The higher the germ dosage index for the mating the smaller was the percentage of fertile eggs found to be. 2. The prenatal mortality, measured by the precentage of embryos (zygotes) which died before hatching to all embryos formed, was materially smaller in the case of offspring from matings in which one or both parent individuals were treated, than in the case of offspring from untreated control parents. 3. The post natal mortality at all ages was materially lower in the case of offspring from matings in which one or both indi- viduals were treated, than the average mortality of individuals from untreated control parents. The only matings of untreated individuals showing as low a rate of mortality as the treated matings were a selected group picked as having the very lowest mortality. 4. The sex ratio of the progeny was not sensibly affected by the treatment of the parents. 5. There was no significant difference in mean hatching weight between the offspring of treated males and the offspring of norma! untreated control males when both were mated to normal untreated females. The slight differences which did appear were of the same order of magnitude as their probable errors. 6. Both the male and female offspring of matings in which both parents were treated showed a higher mean hatching weight (i.e., are heavier when hatched) than the offspring of either completely normal control matings, or of matings in which the father only was treated. 7. The offspring of alcoholized parents, whatever the nature of the mating, showed a higher mean adult body weight than PARENTAL ALCOHOLISM AND THE PROGENY 295 offspring of untreated parents of the same breeds mated in the same way. This is true of both sexes. 8. In the case of the male chickens there was no substantial difference in the rate of growth in the three lots until after an age of about 100 days was passed. From that point on the male offspring of treated #@ X untreated and treated 2 2 grew at a more rapid rete than the controls. The differences in mean body weight for a given age became increasingly large as the age advanced. 9. In the case of the female chickens there was no substantial difference in the rate of growth in the three lots until after an age of about 150 days was passed. During the next 25 days the controls grew faster than the chicks from treated parents. At and after 200 days of age, however, the offspring of treated parents (one and both) showed a higher mean body weight than the controls. 10. At all ages in the case of the male chicks, and in all ages but two (12.5 and 19.5 days) in the case of the female chicks, the mean body weight of the offspring having both parents alco- - holic was higher than that of the offspring having one parent only, the father, alcoholic. 11. The proportion of abnormal chicks produced from treated parents was no greater than that produced from untreated parents. 12. The normal Mendelian inheritance was in no way affected by the treatment of the parents, so far as concerns any of the numerous characters observed and tested. This statement ap- plies only to phenomena of dominance, recessiveness and sex linkage. Other Mendelian phenomena have not as yet been tested in these experiments. 13. There was no evidence from these experiments that the treatment of individual fowls, whether male or female, with either ethyl alcohol, methyl alcohol, or ether, had any dele- terious effect upon those germ cells which formed zygotes. The treatment rendered many germ cells incapable of forming zygotes at all, but those which did form zygotes had plainly not been injured in any way. 296 RAYMOND PEARL 14. There was no evidence that specific germinal changes have been induced by the treatment, at least so far as concerns those germ cells which produced zygotes. 15. It is suggested that these results, as well as the results of earlier workers, may be most satisfactorily accounted for on the hypothesis that alcohol and similar substance act as selec- tive agents upon the germ cells of treated animals. The essen- tial points in such an hypothesis may be put in the following way. a. Assume that the relative vigor, or resisting power of germ cells varies or grades continuously from a low degree to a high degree and further assume that the absolute vigor of the whole population of germ cells, measured by the mean let us say, is different for different species. b. In the intensity of dosage employed in inhalation experi- ments alcohol does not destroy or functionally inactivate all germ cells. The proportionate number of the whole popula- tion of germ cells which will be inactivated by such dosage may fairly be supposed to depend upon the mean absolute vigor or resisting power characteristic of the particular species or strain used. In a species with germ cells of absolutely low mean vigor proportionately more will be functionally inactivated than in a species of high absolute mean vigor of germ cells. c. Besides the germ cells which are wholly inactivated by the deleterious agent, and which we may designate as class (a), we ‘may fairly assume that there is a possibility of two other classes existing, viz., (b) germ cells which, while not completely inacti- vated, are so injured by the agent as to produce zygotes which are measurably defective in some degree, and (c) germ cells which are not measurably affected by the agent at all in the dosage employed, and produce zygotes which are not discern- ibly otherwise than perfectly normal. d. It appears entirely fair to assume that germ cells of the (a) class are of relatively the lowest mean vigor or resisting power, class (b) next, andclass (c) the highest. The proportion- ate number of the two sorts of zygotes corresponding to classes (b) and (c) of germ cells which would be expected to appear in PARENTAL ALCOHOLISM AND THE PROGENY 297 any experiments made to test the point would clearly be a func- tion of the mutual relationship or proportionality between two variables, the dosage of the deleterious agent on the one hand, and the mean absolute resisting power of the germ cells charac- teristic of the strain or species of animal used in the experiments on the other hand. e. If the dosage of the agent be relatively high in proportion to the mean absolute resisting power it would be expected that all the germ cells would fall into classes (a) and (6b), producing no zygotes at all or zygotes in some degree defective. This about represents the condition, so far as can be judged from the data given, with Stockard’s alcoholized guinea pigs and Weller’s lead-poisoned guinea pigs. The dosage is sufficiently high in proportion to the absolute germinal resisting power that allor practically all of the offspring are defective in greater or less degree and in reference to some one or more characters. Stock- ard’s F, and F; results indicate that though the untreated F; animals from alcoholists may appear normal, they really are somewhat defective. f. If, on the other hand, the dosage, though absolutely the same, be relatively lower in proportion to the mean absolute resisting power of the germ cells it would be expected that all three germ cell classes (a), (b) and (c) would be represented. The zygotes actually formed would be chiefly produced by (c) germ cells, and to a much smaller extent by (6) cells. Under these circumstances it would necessarily follow that a random sample of the zygotes produced after the action of the deleterious agent would, on the average, be superior in respect to such qualities as growth, etc., which may be supposed to depend in part at least upon germinal vigor, to a random sample of zygotes formed before the action of the agent, because the germ cells of class (c) are a selected superior portion of the total gametic population. g. Essentially that proportionality between effective dosage of the deleterious agent and absolute resisting power of the germ cells outlined in the preceding paragraph (f) is believed to have obtained in the present experiments with fowls, Nice’s experi- 298 RAYMOND PEARL ments with mice, and nature’s experiments with the working- men’s population studied statistically by Elderton and Pearson. The experiments here reported are being continued. XIII. APPENDIX In this appendix are given the frequency distributions of the growth measurements, on which the constants of tables 11, 12, and 13 in the text are based. TABLE I Untreated iS X untreated 99. Ages: 2 days and 5.5 days AGE = 2 DAYS AGE = 5.5 DAYS WEIGHT IN GRAMS (CENTER OF CLASS) Qy 40 Q, 40 26.5 28.5 30.5 32.5 34.5 36.5 38.5 40.5 42.5 44.5 46.5 48.5 50.5 Re WR DD tO “ICO & Co Pe ON Wr OOo oO Nore NRK Oe KS OK bt Motalstens-e-- 12 22 20 28 EEDA a PARENTAL ALCOHOLISM AND THE PROGENY 299 TABLE II Untreated SS X untreated 99. Ages: 9 days, 12.5 days, 16 days, 19.5 days, 23 days and 26.8 days WEIGHT IN 9 DAYS 12.5 pays 16 pays 19.5 pays 23 DAYS 26.5 DAYS GRAMS (CENTER OF CLASS) 3 Q rou Q of g fou 9 fo 9 rou Q 37 1 42 1 47 2 52 6 3 1 aS eR CO Or Or Bee Dw OF NON OW OR KS Ye (0) BFPwWwWNnNwWwWN Ree Hm HB CO bo Co Jj - © © =) =<] NF NOWORNHrRFFH WN BPWrRNnNNnNON ON, OFF We BHHOOOCOCOONWFOFE WrFrE OOF RF oooocooorrRR RR OF NH Ke WoONnNONF KR KE WORF NOOO Fr KF OWORONF OFF Or Totals....) 14 | 22 | 20 | 28 | 14 | 22 | 19 | 28 | 14 | 22 | 19 | 28 TABLE III Untreated Jo X untreated 92. Ages: 37 days and 40.5 days WEIGHT IN GRAMS 37 DAYS 40.5 DAYS (CENTER OF CLASS) fou fe) oe fo) 134.5 1 l 144.5 154.5 164.5 1 : 174.5 1 3 184.5 1 1 9 204.5 1 1 1 5 214.5 . 2 9 234.5 2 6 254.5 1 3 9 ; 264.5 2 1 9 4 284.5 1 9 294.5 304.5 1 SS Se eee MotalseP Ak. 11 22 19 25 TABLE IV Untreated iS X untreated 99. Age: 53 days 53 DAYS WEIGHT IN GRAMS (CENTER OF CLASS) 229.5 1 249.5 269.5 289.5 309.5 329.5 349.5 369.5 i 389.5 409.5 429.5 449.5 469.5 489.5 509.5 2 529.5 3 — HB bo ht hb OO RH bo — BW Nia So NOU Se RO AIS ae etre Sass 29 45 300 PARENTAL ALCOHOLISM AND THE PROGENY 301 TABLE V Untreated AS X untreated 99. Ages: 67 days, 77.5 days, 95 days, 119.5 days, 147.5 days WEIGHT IN 67 DAYS 77.5 DAYS 95 DAYS 119.5 pays | 147.5 pays GRAMS c(Deageepemiy: Qype || = i ee ee CLASS) ots °) fou °) fou °) J 2 ou WNHONWONY a — (=) ~J He or NOwWNrFN Wwe BPnNnwwWN RR NR — Pwo ow 00D bo w _ eo PP OLD WwW Oe — eS Re Dw OUND OO CO bo lt bo — 102) bo pss or i) Nr DADOWRhHE Ee Pp 302 RAYMOND PEARL TABLE VI Untreated 7 7 X untreated 9Q. Ages: 165 days, 174.5 days, 209.5 days, 267 days, 286 days 165 Days 174.5 pays | 209.5 pays| 267 pays | 286 pays WEIGHT IN GRAMS (CENTER OF CLASS) ———————————— oo ———— 1424.5 1474.5 1524.5 1574.5 1624.5 1674.5 1 1724.5 1774.5 1824.5 1874.5 1924.5 1974.5 2024.5 2074.5 2124.5 2174.5 2224. 2274. 2324. 2374. 2424. 247A. 2024. 2574. 2624. 2674. 2724. 2774. 2824. Worn & WY WwW bo bo — or) | NO A el Nore Nr owe Noe bo me ww oO bb Nore Fw dw Ww WwW Wb Nee CW bo as eee Orv St St Sr Or Or Sr Or Sr Sr Sr Or or e ee Hm bo e CO No oO eR b — — Potals wees. sb 8 21 50 25 35 16 25 PARENTAL ALCOHOLISM AND THE PROGENY 303 TABLE VII Treated & X untreated 2 Q and treated S X treated 92. Age: 5.6 days Fe eee tar a iiis TREATED oO’ X UNTREATED Q 2 TREATED o& X TREATED @ 2 (CENTER OF CLASS) fou g rot Q 26.5 1 28.5 1 0 30.5 2 2 1 32.5 2 1 1 2 34.5 6 1 0 1 36.5 Of 7 2 0 38.5 9 5 1 2 40.5 6 12 0 3 42.5 10 18 4 a 44.5 a 5 10 9 46.5 5 4 3 4 48.5 6 3 3 3 50.5 2 4 2 1 52.5 1 0 3 54.5 0 56.5 1 WOVEN Ss oonosoe 64 59 29 33 TABLE VIII Treated & X untreated 29 and treated 7 X treated 99. Age: 12.5 days a econ renee TREATED o' X UNTREATED @ 9 TREATED Go’ X TREATED 2 9 (CENTER OF CLASS) rofl ) fou 9 39.0 2 39.5 1 0 48.5 9 4 3 1 47.5 6 8 1 7 51.5 of 10 5 5 55.5 10 6 5 9 59.5 5 a 7 5 63.5 9 8 1 3 67-25 3 9 5 0 TAL 5) 2 1 1 1 75.5 2, 1 0 Rotealseeen oe 54 55 29 31 304 RAYMOND PEARL TABLE IX _ Treated 3 X untreated 2 2 and treated S X treated 99. Age: 19.5 days TREATED GO’ X UNTREATED 9 9 TREATED oc X TREATED 92 9 WEIGHT IN GRAMS (CENTER OF CLASS) a 9 a 9 (0) Oomw ff bo Sa hy Nb NRPWATRUORDAADNWO SO W BH PRWADMWAMNOANH OO hb HBOrRWWHWHERY PORE FPCoOOrFWNRWWOHNWOrF a PARENTAL ALCOHOLISM AND THE PROGENY 305 TABLE X Treated io X untreated 9 2 and treated 1H X treated 99. Age: 33.5 days TREATED o'o' X UNTREATED Q 2 TREATED o'o' X TREATED @ 9 WEIGHT IN GRAMS (CENTER OF CLASS) 94.5 1 104.5 114.5 124.5 134.5 144.5 154.5 164.5 174.5 184.5- 194.5 204.5 214.5 224.5 234.5 244.5 254.5 264.5 1 274.5 1 354.5 yi! — NOwWrPorR PR DONW FD KK LO NwWwWwoanw Pr POR ND W dD bo Br wWwWorDy Rt — Mopalssss:. 5. 51 54 27 28 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 RAYMOND PEARL TABLE XI Treated jo X untreated 9 9 and treated JS X treated 2 @. and 68.5 days WEIGHT IN GRAMS (CENTER OF CLASS) 209. 229. 249. 269. 289. 309. 329. 349. 369. 389. 409. 429. 449. 469. 489. 509. 529. 549. 569. 589. 609. 629. 649. 669. 689. 709. 729 749. 769. 789. 809. 829. 5 5 5 5 a) 5 5 5 5 5 47.5 DAYS Ages: 47.5 days 68.5 DAYS Treated o& X untreated 2 a g i) 5 5 5) 5 5 5 5 5 5 5) 5 5 5 5 5 5 5 5 5 5 5 WOONN OWA W He —_ _ Bwortwoont rk We bd w Treated @ treated 9 fot WwwnwnrPrR RE WH — —_ x g Se Dw OR Orr — Treated o X untreated 9 rou eo PNONN PWOND > bo — Qg WOrRFNTRFN rR WRN FWD W bo Treated & X treated 2 fo bw wre Pe [== bo Be Doe 2 FPNrPr NW WN WwW Od Ne Totals. . 48 or iw) PARENTAL ALCOHOLISM AND THE PROGENY 307 TABLE XII Treated oS X untreated 9 2 and treated Jo X treated 92. Ages: 89.5 days, 110.5 days 89.5 DAYS 110.5 pays WEIGHT IN GRAMS Treated X o@ Treated 7 X Treated o X Treated 7 X (CENTER OF untreated Q treated 2 untreated @ treated 9 CLASS) . (of! 2 ron 2) rol g rot g 424.5 Aehel 474.5 524.5 574.5 624.5 674.5 724.5 774.5 824.5 874.5 924.5 974.5 1024.5 1074.5 1124.5 1174.5 1224.5 1274.5 1324.5 1374.5 1424.5 1474.5 1524.5 1574.5 1624.5 1674.5 1724.5 1774.5 1824.5 1874.5 1 —_ _ bo —_ _ BPN RW ODwWe WNONONORON WH bt bet TN) SO) ST ST) OO! HR Eb —_ BKWONonNndarHwrH RE RON Ree DO _ — — NONnNrF Own» ee _ ONMrF AWN oONw WOHhd —& & pb NOnwmwhd ob ww — Totals..| 51 52 26 30 51 52 27 30 308 RAYMOND PEARL TABLE XIII Treated Sh X untreated 9 9 and treated io X treated 99. Ages: 138.5 days, 166.5 days 138.5 DAYS 166.5 DAYS SOU PETE CERI DN, || ga GRAMS Treated & X Treated «| X Treated & X Treated «7 X (CENTER OF untreated 2 treated 9 untreated 9 treated 9 CLASS) of fof g fol 2 fof g +0 — — —_ (2S) Wo) SS) Loy Sat tS _ — 1 BS Gv Or CO-sI WS OU et rr — None © Od bY NwWwwnds orer SE Co. a SS —_ FoNnwomuonwntrr wre me rpow ond & & bd bo —" Totals:.| 50 52 28 30 49 51 29 30 PARENTAL ALCOHOLISM AND THE PROGENY 309 TABLE XIV Treated oh X untreated 9 2 and treated SH X treated 99. Ages: 194.5 days 204.5 days ; 194.5 DAYS 204.5 DAYS WEIGHT IN GRAMS Treated & X Treated J X Treated «7 X Treated «| X (CENTER OF untreated treated 9 untreated 9° treated 9 CLASS) of! g ou g of 2 fot 2 1649.5 1749.5 1849.5 1949.5 2049.5 2149.5 2249.5 2349.5 2449.5 2549.5 2649.5 2749.5 2849.5 2949.5 3049.5 3149.5 3249.5 1 1 Bere Dw e bo bo ae ole Oa Bere De BD dD Ww CO FS SSS — Totals. . 19 14 12 . 15 6 5 5 8 310 RAYMOND PEARL TABLE XV Treated io X untreated 2 @ and treated Oo X treated 99. Age: 290.5 days WEIGHT IN GRAMS (CENTER OF CLASS) Totals 1474. 1524.5 to 1774.5 inclusive 1824. 1874. 1924. 1974. 2024. 2074 2124. 2174 2294. 2274. 2324. 2374. 2424. | 2474. 2524. 2574. NAAN AAaAaaanaana»w«na 5 5 TREATED co’ X UNTREATED 92 Qg TREATED oO’ X TREATED Q © 1 no bo 11 EFFECTS OF CENTRIFUGAL FORCE ON THE STRUC- TURE AND DEVELOPMENT OF THE EGGS OF CREPIDULA EDWIN G. CONKLIN Princeton University ONE HUNDRED AND TWENTY-FOUR FIGURES CONTENTS Tee NETOCUCTIONE EAs kc cee oe Oe se.es | s+ «6 OO REEDS aS ba aa aoe ae 312 TRPELISCOLICA ene) cae ee MeN SIRT Gl a... 2c eats ees eto RRO EES 312 2eaGeneralsaim syan dare sullitsesenenic/a.s >.<. Meets on rare 314 Se vaterialsandpmethodsss «pra s+... eee One oe eae 316 lJ. Results of centrifuging during maturation and fertilization stages..... 317 1. Before and during the first maturation division..... ete ae 317 2. Durinothesecondansturation divisionaeseneee. s.ce- cei ae ae 321 Shonmationomeiantypolar bodies)... eee en inion coer 322 A Wihy- polar bodiessdo mot. develop: . . 0 dmpeiieae vue sc ak indaces he 325 5. Cleavage of eggs centrifuged during maturation stages............ 330 6. The maturation pole does not determine the animal pole of the egg nor the ectodermal pole of the embryo.....:................-.-. 332 7. Results of centrifuging after maturation and before the first CLE BIN AIS CHINE oN oie Oe isco 6-5-0 Se EE a ote rotate Gmc 333 Ill. Results of centrifuging during cleavage stages....................0000- 335 1. Modifications of first and second cleavages: Equatorial cleavages. 335 2. Results of centrifuging during resting stages between first and second and second and third cleavages. ..................02205: 337 3. Later cleavages of eggs centrifuged in the two or four cell stages. 339 4. Results of centrifuging during the third and fourth cleavages.... 340 5. The potency of substances, regions and blastomeres of centrifuged (Yat erences ORES ioic oe chick cht or oa) elk aPC ten ls cc vehi ckG oo SIRS ARS CERI GRS cemacre 343 ivenGeneralconchustonsrerts ricer me ce ai osk 6 alo or Re once biaaras 346 i) the nature’ and causestoficell polarity: ...cseitiseamaeest as sees s- << 346 Porbne Structure Of ProlOpeasMe sc :/2.;.,s0-s: ce tome era hieto c haere 3 csi oe Oe 356 3. Protoplasmic flowing and intracellular movements................ 362 ae theyonentations of deyelopment...:..:. states: aes ene ees «me 364 VER SURLEDA ENS ATICUITIGOX 1... 7 1/ AeVRE RARE oe ai''a 2. aisys bon seC CRE Adah lero ical atleast 367 VelemInT LOE DUNE CULES» or: 5 2). RRP os, is, < cals bia kcl bis wieleyamtsle ob a sekels 372 33119 EDWIN G. CONKLIN 1. INTRODUCTION 1. Historical The use of very strong centrifugal force in the study of the structure and development of the egg was first made, I believe, by Gurwitsch (04) and Lyon (06). Before that time the effects of a rather weak centrifugal force of from 4 to 20 times gravity on the development of eggs had been studied by Rauber (’84), Roux (’84), O.. Hertwig (’99, ’04), Morgan (’02), and Wetzel (04); but Gurwitsch used a very strong centrifugal force by which he injured or destroyed the protoplasm for the purpose of analyzing its structure, while Lyon used a force of from 4500 to 6400 times gravity in order to study its effects on develop- ment. Lyon discovered that by means of this great force the substances in the eggs of Arbacea, Asterias, Chaetopterus, Phascolosoma, and Cynthia could be separated into three or four layers differing in color or refractive index, and he made a brief study of the development of the centrifuged eggs of Arbacea. This work was quickly followed by extensive studies of the de- velopment of centrifuged eggs of Chaetopterus by Lillie (06, ’09) and of the centrifuged eggs of Arbacea, Cumingia, Cerebratulus, Hydatina, the fish and the frog by Morgan (’07, ’09, 710). Bo- veri (10) and Hogue (10) studied the effects of strong centrif- ugal force on the eggs of Ascaris; Conklin (10) on the eggs of Physa, Lymnaea and Planorbis; Konopacki (711) on the eggs of the frog, and Jenkinson (’14) also on the eggs of the frog. In general it has been found that yolk, which is usually the heaviest substance in the egg, is thrown to the centrifugal pole, oil or fatty substance to the centripetal pole, while the trans- parent cytoplasm together with the nucleus occupies the middle zone between the other two. Usually eggs develop normally after this stratification, although the distribution of oil, yolk, and pigment may be very abnormal; and even the cytoplasm may be more abundant in certain cleavage cells than in normal development, or less abundant in others, without Same interfering with typical development. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA Bulle Furthermore a general result of previous work has been to show that the polarity and pattern of organization of an egg are not changed by this dislocation of egg substances; the polarity and pattern of the embryo which develops from such an egg remains unchanged irrespective of the location of these different materials within the egg. This is a surprising fact which invites further study. How is it possible to dislocate in any axis the visible material substances of an egg and yet leave its polarity and pattern of organization undisturbed? Lillie (’06) concludes that polarity is a property of the ‘ground substance’ of the egg, this substance being ‘‘a fluid which has no filar, reticular or alveolar structure,” but yet is ‘firmly organ- ized’ so that it is not affected by centrifuging. He regards the substances which are dislocated by centrifugal force as mere ‘inclusions’ in this ‘ground substance,’ consequently polarity remains unchanged when these inclusions are forced to occupy new positions since polarity inheres in the ‘ground substance’ which is not moved by centrifugal force. It is evident from Lilhe’s use of this term that he means the ‘ground substance’ to include what is commonly called cytoplasm as contrasted with metaplasm or inclusions. However it will be shown in this paper that most of the cytoplasm of an egg can be displaced without permanently changing the polarity of the egg. In eggs which contain relatively little yolk, such as those of echinoderms, Chaetopterus, Cumingia, etc., the yolk may be thrown to any pole without greatly displacing the protoplasm from its normal position, and consequently it is possible in these cases that normal development results because the real formative materials, viz., nucleus and cytoplasm, have not been displaced _to any great extent by centrifugal force. But when the volume of yolk is large, as in the egg of Crepidula, the nucleus and cyto- plasm may be displaced from their normal positions by nearly the whole diameter of the egg, and the subsequent developm ‘nt of such eggs throws light not only upon the specific value of dif- ferent egg substances but also upon the polarity and organi- zation of the egg as a whole. 314 EDWIN G. CONKLIN 2. General aims and results of this work It was with a view to determine more exactly whether there is a ‘ground substance’ which remains unmoved in centrifuged eggs, or whether the morphogenetic substances of the egg are moved with the other substances and later resume their nor- mal positions that the following work was undertaken. The eggs upon which these experiments were performed were those of the marine gasteropod, Crepidula plana. This object was chosen not only because of my familiarity with its normal de- velopment but also because the yolk in this egg is so abundant that any change in its position involves marked changes in the positions of nuclei and cytoplasm, which are presumably parts of the ‘ground substance.’ If the eggs of this gasteropod are subjected, after fertilization and before the first cleavage, to centrifugal force of approximately 600 times gravity, the yolk is thrown to the centrifugal pole, where it occupies about three-quarters of the volume of the whole egg; the middle zone, consisting of nucleus and clear cyto- plasm, comprises a little less than one-quarter, and the oil zone constitutes about one sixty-fourth of the volume of the entire egg, the relative volumes of the three zones being 49: 14:1. In normal eggs of this stage the nucleus, centrosphere and most of the cytoplasm he near the animal pole, but in centrifuged eggs these formative substances may be displaced far from this position, the yolk, for example, being thrown to the animal pole and the protoplasm to the vegetal one, or these displacements may take place in any other axis. Nevertheless such eggs fre- quently develop normally, showing that the polarity and pat- tern of organization of the egg have not been permanently changed by this dislocation of the formative materials. It seems necessary to conclude that there is some material substance, or relation of parts, in these eggs which persists with relatively little change, in spite of the dislocations caused by centrifuging, but if there is a ‘ground substance’ here which is not moved by centrifugal force it must be a relatively small part of the gen- eral protoplasm of the egg. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA ae There is good evidence, which will be presented in the de- scriptive part of this paper, that this is indeed the case and that, while the greater part of the cytoplasm is free to move under the influence of centrifugal force, there is in these eggs a denser, more viscid portion of the protoplasm which forms a framework running through the cell and connecting the nucleus and centro- some, or centrosphere, with a peripheral layer which surrounds the entire egg. This framework may be stretched or distorted and yet may be able to ‘bring back dislocated parts to their normal positions unless partition walls have been formed in the meantime which prevent this return. This framework is the seat of the polarity and pattern of organization of the cell; it holds the cell organs, especially the centrosphere and the nucleus, in a definite relation to one another and to the cell axis, and it prevents the complete stratification of cell substances into sharply marked zones according to their specific weights. The substance of this framework is probably identical with the ‘ground substance’ of Lillie, though in Crepidula it constitutes a relatively small part of the cell contents, and in my opinion it does have a “‘filar, reticular or alveolar structure.” Further- more this substance is affected by centrifuging; it is stretched and distorted if centrifuging is strong enough, but is capable of recovering its normal form afterward. It seems evident that the term ‘ground substance’ is not an appropriate one for this denser protoplasm, which constitutes the achromatic substance of the mitotic figures and of the resting nucleus as well as the astral radiations and strands which con- nect these with the peripheral layer of the cell; this denser pro- toplasm is much less abundant than the more fluid protoplasm which forms the chief part of the middle zone of centrifuged eggs; it is not uniformly distributed throughout the cell, but exists in astral radiations and fine strands which run through the more fluid protoplasm as well as through the yolk. It is probable, however, that it is identical with the ‘spongioplasm’ of Leydig, the ‘kinoplasm’ of Strasburger and in many respects it corresponds to the ‘interalveolar substance’ of Biitschli, and to the ‘archoplasm’ of Boveri. 316 EDWIN G. CONKLIN This introductory account of the aims and chief results of this work will serve perhaps to make more easily intelligible the fol- lowing detailed account of these experiments. 3. Material and methods The experiments here described were begun ten years ago at the Marine Biological Laboratory at Woods Hole, Massachusetts, and have been continued there almost every summer since. More than one hundred and forty different experiments ‘were per- formed and in every instance the results of these experiments were studied by means of carefully stained and permanently mounted preparations. In the earlier years of this work the eggs in their capsules were centrifuged in a machine driven by hand at such a speed as to make the centrifugal pressure about 2000 times gravity; in later years a machine driven by water pressure was used, the centrifugal force being approximately 600 times gravity. There is a slight tendency for eggs to rotate during centrifuging so that the animal pole becomes centripetal and the vegetal pole centrifugal in position, as is shown by the somewhat larger number of eggs in this position than in-any other, and yet the viscosity of the fluid in which the eggs are suspended or the pressure of the thin-walled capsules upon them prevents many of the eggs from rotating. Eggs in different stages of develop- ment were centrifuged for various lengths of time. They were then removed from the centrifuge and either fixed at once or left in finger bowls of fresh sea-water for varying lengths of time, as indicated in the description of figures given at the end of this paper. In most cases the eggs were fixed, stained and mounted entire in the method described by me in previous papers (97, ’02). These permanent preparations were made soon after the experi- ments were performed and they are still in good condition. Many serial sections also were cut, but in general they are less instructive than whole amounts. All drawings were made with a 3 mm. homogeneous immersion lens with which the finer details of nuclei, centrosomes and cytoplasm can be seen with great distinctness. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA B17 Il. RESULTS OF CENTRIFUGING DURING MATURATION AND FER- TILIZATION STAGES In Crepidula, as in most other prosobranchs, the eggs are fer- tilized within the oviduct of the female and are then surrounded by secretions from the nidamental gland; the outermost layer of these secretions hardens into a capsule. By opening the oviduct of females taken in the act of egg-laying or by getting capsules immediately after they have been deposited, it is possible to obtain eggs before the germinal vesicle breaks down and before the spermatozoon enters the egg. Under normal conditions the polarity of the unfertilized egg is marked by the eccentricity of the germinal vesicle toward the animal pole. Generally the spermatozoon enters the egg near its vegetal pole, though there may be exceptions to this rule. Under normal conditions both first and second polar bodies are formed invariably at the animal pole; indeed so general is this rule that the animal pole is frequently defined as that pole of the egg at which the polar bodies are formed, and yet, as we shall see later, the polar bodies may be caused to form at any point on the sur- face of the egg without in any way changing the polarity of development. In Crepidula the polar bodies never change their point of attachment to the egg;.as long as they are present they remain at the point where they were extruded and they there- fore constitute a valuable landmark. 1. Results of centrifuging before and during the first maturation division (figs. 1-18) From the facts Just stated it is evident that before the forma- tion of the first polar body, there is no sure way of distinguishing the original poles of a centrifuged egg of Crepidula, though the incomplete stratification of the cell constituents, and particu- larly the position of the sperm nucleus and of the first matura- tion spindle and the direction of movement of various cell con- stituents after centrifuging, may indicate with a certain degree of probability the location of the original poles. 318 EDWIN G. CONKLIN Before maturation, yolk and cytoplasm intermingle in all parts of the egg and, although the cytoplasm is somewhat more ~ abundant at the animal pole than elsewhere, these two substances are not sharply segregated. When such an egg is strongly cen- trifuged the yolk is driven to the centrifugal pole, while the cyto- plasm is displaced toward the opposite pole, so that these sub- stances come to be partially segregated (figs. 1 and 2), though this segregation is never so complete as it is in later stages. The germinal vesicle goes with most of the cytoplasm to the centrip- etal pole. The entire egg is frequently flattened in the axis of centrifuging, but the germinal vesicle is elongated in that axis, probably owing to the fact that the nucleus is not subjected to external pressure, whereas the egg is. Before the prophase of di- vision, the chromatin is not moved within the germinal vesicle by centrifuging; after the prophase has begun, the chromatin ap- pears to be free to move. This corresponds more or less closely with the conditions found by Kite (13) in his micro-dissections where the nucleus during resting stages was found to be a gel which becomes more fluid during division phases. In the pro- phases shown in figures 1 and 2 the chromosomes are evidently heavier ‘than other constituents of the nucleus since they collect in the centrifugal end of the germinal vesicle while the nucleolus goes to the centripetal end of the vesicle, thus showing that in this egg the nucleolus is lighter than other nuclear constituents; in other cases it is heavier, as for example in the ovarian egg of the lobster (Herrick, ’95) and in the electric-motor nerve cells of torpedo (Dahlgren, ’15). The axis of the spindle in figure 2 is approximately at right-angles to the axis of centrifuging, but as the initial position of the spindle bears no constant relation to the axis of the egg this fact has no particular significance. When the membrane of the germinal vesicle of a normal egg dissolves, the first maturation spindle is left in the egg usually at some distance from the surface and at the same time the sper- matozoon enters the egg usually near the vegetal pole. When eggs are strongly centrifuged at this stage there is a fairly sharp separation of cytoplasm and yolk; the maturation spindle is car- ried along with the cytoplasm thus showing that it is not at this CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 319 stage closely attached to the surface layer of the cell; the nucleo- lus of the germinal vesicle is carried to the centripetal pole of the egg where it lies with the lightest substances of the cell, but the sperm nucleus may be found now in the cytoplasm, now in the yolk and again on the border between the two, thus showing that it is not moved to any great extent by centrifugal force and that yolk or cytoplasm may stream past it without much altering its position (figs. 3-18). If the sperm nucleus were free to move according to its specific weight it would always occupy some constant position either in the cytoplasm, in the yolk, or in the region between the two; and the fact that it is not found in any constant position with regard to cytoplasm or yolk shows that something prevents its free movement. It seems, probable, therefore, that after the sperm nucleus has entered the egg it becomes attached to the cell framework by astral filaments which radiate from the sperm centrosome and which are so fine that it is difficult to see them and yet so strong that, although all the movable substances of the egg may flow past, the sperm nucleus is not torn from its moorings. After the maturation spindle of the normal egg has reached its metaphase it becomes closely attached to the egg surface at the animal pole by one of its asters and thereafter this attachment can not be broken by the strongest centrifuging to which the eggs were subjected. Figures 5 and 6 represent such eggs in which yolk has been driven to the animal pole and cytoplasm to the opposite pole, but the maturation spindles are so firmly anchored to the surface layer of the egg that they can not be pulled away, even though the spindle itself is stretched in length and the egg surface is indented, apparently by the pull of the spindle upon it. Since this attachment of the spindle to the egg surface occurred in this case before centrifuging there is no doubt that this point of attachment represents the true animal pole of the egg. Figures 9-12 are all from the same experiment; the eggs were centrifuged in the prophase before the first maturation spindle had become firmly attached to the surface of the egg, conse- quently the spindles were carried into the interior. The eggs 320 EDWIN G. CONKLIN were then allowed to stand for one and one-half hours before they were fixed, during which time the spindles advanced to the anaphase. In these four figures the spindles lie near the middle of the egg and the egg contents are not regularly stratified but the spindles and some of the cytoplasm project into the yolk in such a manner as to suggest either that the spindles were limited in their movement at the time of centrifuging or that after centri- fuging they were moving back to their original positions. In either case it seems necessary to assume that the connection of the spindles with the animal pole, probably by means of proto- plasmic fibers, was never lost. Finally such spindles probably come to the surface of the egg at the animal pole and form nor- mal polar bodies since in all eggs of this experiment which were allowed to develop further the polar bodies are typical in ap- pearance and position, lying at the center of the ectodermal pole. Figures 13-18 represent eggs which were centrifuged in the late anaphase of the first maturation division after the spindle was closely attached to the animal pole; in figure 13 the egg was fixed immediately after centrifuging; in figures 14 to 18 they were left in normal conditions for one and one-half hours be- fore being fixed. In every case the spindle has remained in its original position and a normal polar body has been formed, though the cytoplasm and yolk are abnormal in position. As in previous figures the sperm nuclei show little indication. of having been moved by the centrifuging. In figures 17 and 18 the lane of cytoplasm leading from the animal pole to the cyto- plasmic field indicates either that the egg substances were not regularly stratified by centrifuging, or that they are beginning to return to their original positions. If the first maturation division occurs normally it is always easy thereafter to identify the original animal pole of the egg, how- ever much the egg substances may have been moved out of their normal positions, by the location of the first polar body. This is an invaluable landmark since as long as the polar body re- mains attached to the egg it does not move from the position at which it was first formed. Before the formation of the first polar CENTRIFUGAL FORCE ON EGGS OF CREPIDULA azil body the identification of the animal pole in centrifuged eggs is more or less a2 matter of conjecture, dependent largely upon the position of the egg nucleus or spindle and upon the subsequent movements of the egg substances. After the first polar body has been formed normally the position of the animal pole is no longer uncertain. However, as we have seen in figures 9 to 12, it is possible to move the first polar spindle before it becomes firmly attached to the periphery of the egg, and if such eggs are kept whirling for a long time the polar body may form at other points on the egg than the original animal pole, as shown in figures 36 to 38. 2. Results of centrifuging during the second maturation division (figs. 19-58) In the late anaphase of the first maturation division the chromosomes in the egg can not be centrifuged away from the animal pole to which they are bound probably by the interzonal fibers (figs. 13 to 18). When the second maturation spindle has reached its early metaphase this attachment is relaxed, for if eggs are centrifuged at this stage the spindle is carried away from the animal pole along with the cytoplasm and may be trans- ported to any part of the cell (figs. 19 to 21). However, if the second maturation spindle has reached its anaphase at the time the centrifuging begins it can not be moved very much since one pole of the spindle becomes more or less closely attached to the surface of the egg by its astral fibers. In some cases such a spindle may be pulled away from the surface for a short dis- tance without breaking this attachment (figs. 23, 24), but in other cases it remains closely attached to the surface at the animal pole (figs. 22, 25, 29). This variation is probably due to slight differences among various eggs in the age of the spindle and in the firmness of its attachment to the surface, as well as to differences in the strength of centrifuging. | If the young maturation spindle is centrifuged away from the animal pole (figs. 19 to 21) it takes a position in the cytoplasm between the yolk on one side and the oil cap on the other. It may be carried all the way through the egg from the animal to THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL, 22, No. 2 Be, EDWIN G. CONKLIN the vegetal pole (fig. 21), but however far it may be from the animal pole it moves back to its normal position to finish its division unless it is prevented by very long or strong centrifu- ging; in the latter case the polar body may be cut off wherever the spindle happens to be. Thus in eggs which are centrifuged from 30 minutes to 2 hours during the maturation divisions the polar bodies may be extruded at any point on the surface. 3. Formation of giant polar bodies (figs. 23 to 27, 31 to 58) Most polar bodies that are formed during centrifuging and all that are formed at a distance from the animal pole are larger than normal ones, though they may vary greatly in size (figs. 32 to 58). The size of a polar body or of any cleavage cell depends upon the position of the mitotic figure at the time of cell constriction, since the partition wall between daughter cells always goes through the equator of the spindle; when one pole of the spindle is pressed against the cell membrane, as in the maturation divisions, the size of the polar body depends upon the length of the spindle. The extremely small size of normal polar bodies is due to the fact that the maturation spindle con- tinually grows shorter during the later stages of mitosis, and whenever giant polar bodies are formed it is due to the median position of the spindle in the cell or to its elongation if it is attached to one pole. When centrifuging occurs during the anaphase of a matura- tion division after the peripheral pole of the spindle is closely attached to the surface of the egg at the animal pole the spindle may become much elongated, especially if yolk is driven to that pole (figs. 6, 22 to 26). And conversely, whenever there is a stretching of the spindle, it is evident that one or both of the poles are attached to the surface layer even though this at- tachment may not be directly visible. Consequently, the elonga- tion of the spindle in the formation of giant polar bodies in- dicates that one pole of the spindle is attached to the surface at the animal pole. The fact that a polar body may be given off at the opposite pole of the spindle, i.e., at the vegetal pole of the egg, and yet the polarity of the egg remain undisturbed and CENTRIFUGAL FORCE ON EGGS OF CREPIDULA a2 normal development result, proves that there is no essential dif- ference in the two poles of the maturation spindle. Also the return of the egg nucleus to the animal pole after centrifuging and after the formation of a polar body at some other pole of the egg is probably due to attachments which connect the nucleus to the animal pole. . Because of the elongation of the spindle in these centrifuged eggs, the polar body which is formed may be extraordinarily large. If centrifuging occurs during the first maturation divi- sion, the first polar body is the giant one; if during the second maturation, it is the second polar body; if during both maturation divisions, both polar bodies are abnormally large. In my ex- periments the giant polar body is usually the second one, since many more eggs were centrifuged during the second maturation division than during the first. The cell constrictions shown in figures 23 to 25 le opposite the equator of the second matura- tion spindle and indicate that giant polar bodies are about to be formed. These giant polar bodies may be formed at the animal pole or at any other point on the surface of the egg; they may con- tain the oil of the light zone, the cytoplasm of the middle zone or the yolk of the heavy zone or they may contain samples of all these substances, depending upon whether they are formed at the centripetal or the centrifugal pole and also upon their size (figs. 32 to 34, 43 to 55, etc.). Normal first polar bodies usually caviae by mitosis, second polar bodies rarely do, but I have never seen a case in aie a giant polar body divides, and this in spite of the fact that the polar body may contain samples of all the egg substances and may be as large as, or even larger than, the rest of the egg. In most cases the chromosomes of the polar bodies never form a resting nucleus nor even chromosomal vesicles but remain as dis- tinct chromosomes up to the latest stage studied (figs. 538, 54, 56). In a few instances these chromosomes form chromosomal vesicles or even a resting nucleus (figs. 49, 50, 51). But in only one instance (fig. 57) have I observed an indication of an ap- proaching division of a giant polar body and this case is a very doubtful one. 324 EDWIN G. CONKLIN These ‘polar bodies’ may be caused to form at any point on the surface of the egg, without changing in the least the position of the ectodermal pole. It may be doubted whether such cells which are formed at some distance from the ectodermal pole can properly be called ‘polar bodies,’ not merely because they do not mark a specific pole but also because they are larger than normal polar bodies, they frequently contain different odplasmic sub- stances such as oil and yolk as well as cytoplasm, and they doubtless sometimes contain that pole of the spindle and its chromosomes which would have remained in the egg if the polar bodies had been formed normally; that is these abnormal cells differ both in position and in constitution from normal polar bodies. On the other hand they show certain resemblances to polar bodies, viz., they rarely divide and never undergo regular cleav- age, and most important of all they are formed by those peculiar nuclear divisions which are known as the maturation divisions. Since the notable work of O. Hertwig (’90) and Boveri (’91) in which they pointed out the parallelism between odgenesis and spermatogenesis it has been universally recognized that the two maturation divisions are homologous in o6genesis and spermato- genesis and that consequently the cells which are formed by these divisions are comparable. The cells formed by the first matura- tion divisions are known as ‘second oécytes’ or ‘spermatocytes,’ those formed by the second maturation division as ‘oétids’ or ‘spermatids.’ In normal eggs three of these odtids are very small and are known as ‘polar bodies,’ while the fourth is large and is called the ‘egg;’ it is evident from the manner of their origin that the polar bodies are rudimentary eggs, a view which was first set forth by Mark (’81). When eggs have been sub- jected to pressure or to centrifugal force, all of the odtids may be of approximately the same size, although only one of these cells develops; the fact that three of these cells do not always lie at the animal pole of the cell which develops indicates that the term ‘polar body’ in such cases is a misnomer. But at least all four cells are o6tids and for the sake of simplicity of expression it seems desirable to call these o6tids, which do not develop and CENTRIFUGAL FORCE ON EGGS OF CREPIDULA BAD, which are usually smaller than the one which does, ‘polar bodies” whether they le at the animal pole or not. 4. Why polar bodies do not develop! Normal polar bodies then are rudimentary eggs which do not develop, though they sometimes divide once or twice. Their failure to develop is usually held to be due to their small size, but even where the polar bodies are quite large, as is sometimes the case in gastropods, polyclads and nematodes, they do not develop. In one case only has the development of a polar body, or rather of two second odcytes, been observed. Francotte (’98) discovered in the polyclad Prostheceraeus that at the first mat- uration division the egg divided into two nearly equal cells; each was then entered by a spermatozoon and normally fertilized and at the second maturation division each formed a small sec- ond polar body and underwent normal cleavage and developed to the gastrula stage. In a few other instances the entrance of a spermatozoon into a polar body has been reported though some of these cases are not entirely convincing and need verification. Thus Platner (’86) described the entrance of a spermatozoon into a polar body of Arion; he maintained that the polar bodies are formed before the entrance of the sperm, which would make this case similar to that of Prostheceraeus, but the evidence offered is by no means conclusive. Sobotta (95) calls special attention to the large size of the polar bodies in the mouse and suggests that they may be capable of being fertilized, but offers no evi- dence in favor of this view. Kostanecki (’91) has observed a spermatozoon with its head penetrating the second polar body of Physa, a thing which he regards merely as a ‘curiosity.’ Lefevre (07) found that the same reagent (HCl) which causes the eggs of Thalassema to develop parthenogenetically also caused the polar bodies to undergo several cleavages. The most striking difference between Prostheceraeus and other animals in which giant polar bodies have been reported is to be 1The substance of this section was summarized in Proceedings National Academy Sciences, 1, pp. 491-496, 1915. 326 EDWIN G. CONKLIN found in the fact that in the former fertilization does not take place until after the first maturation division is completed and then each of the daughter cells is fertilized, whereas in the latter the entrance of the spermatozoon occurs before the completion of the first maturation division, with the result that one of the daughter cells contains a spermatozoon and the other does not. In Crepidula the spermatozoon usually enters the egg at the time when the germinal vesicle dissolves and always before the first polar body is cut off. In many mollusks, annelids and ascid- ians, the first maturation spindle remains in the metaphase until the spermatozoon enters the egg or until the egg is stimu- lated by other means (artificial parthenogenesis) to begin de- velopment. The giant polar bodies of Crepidula behave like unfertilized eggs in these regards: 1) the chromosomes do not usually unite to form a daughter nucleus but remain as if they were in the metaphase, as in Chaetopterus, Ciona, ete., though no distinct spindle is visible (figs. 38, 40 to 46). 2) They also resemble unfertilized eggs in that the whole cell stains a purple color in picro-haematoxylin showing that cytoplasm is diffused throughout the whole cell, whereas after fertilization there is a fairly sharp separation of cytoplasm and yolk, the former alone staining purple. 3) Associated with this lack of segregation of cell substances in giant polar bodies there is a lack of the move- ments which in the fertilized egg lead to the segregation of cyto- plasm at the animal pole and of yolk at the vegetal one. These giant polar bodies contain samples of all the odplasmic substances; they may be larger than the odtid which does develop, but the one thing which they lack is a spermatozoon, whereas that odtid which does develop invariably contains a spermato- zoon; we must conclude therefore that the giant polar bodies of Crepidula do not develop because they are not fertilized, and they are not fertilized because a spermatozoon had entered the egg before their formation, thus rendering the polar bodies as well as the egg impervious to other spermatozoa. In this fact is to be found the explanation of the different behavior of the giant polar bodies of Prostheceraeus and of other animals, e.g., Crepidula, for it is well known that one of the first CENTRIFUGAL FORCE ON EGGS OF CREPIDULA S20 effects of the entrance of a spermatozoon into an egg is the pre- vention of other spermatozoa from entering. If the spermato- zoon enters the egg before the first polar body is cut off that polar body as well as other cells which are formed from the egg are rendered ‘immune’ to other spermatozoa. — But although the influence of the entering spermatozoon spreads so rapidly over the egg that within a few minutes at most it renders all portions of the egg surface ‘immune’ to other spermatozoa and thus prevents the fertilization of polar bodies which are formed after fertilization, this influence does not go so far as to cause the polar bodies to develop, even though such polar bodies may be formed several hours after the spermatozoon enters the egg. In Crepidula the second polar body is formed about three hours after the entrance of the spermatozoon, and during this time the sperm head has grown into a vesicular nu- cleus and the sperm aster has become quite large, but in spite of this the spermatozoon has not sufficiently affected the egg sub- stance to cause the second polar body to develop even though that body may contain the larger part of the egg protoplasm. Only that portion of the egg develops which contains the sperm nucleus and aster. This conclusion is similar in many respects to that reached by Ziegler (98), who found that when eggs of the sea urchin, Echinus microtuberculatus, were constricted by cotton fibers under pres- sure only that portion of the egg which contained the spermato- zoon segmented while the portion containing the egg nucleus never divided, though its nucleus frequently went through the division ‘phases, but without any division resulting. In this case the portion of the egg containing the sperm might remain for some time connected with the other portion by a narrow neck, and yet the influence of the sperm in the one half did not cause the other half to develop. These facts are of interest because of their bearing on the na- ture of one of the processes concerned in fertilization. In a series of important and extensive works on artificial partheno- genesis and fertilization, which he has summarized in a recent book, Loeb (713) has shown that at least two factors are involved 328 EDWIN G. CONKLIN in artificial parthenogenesis, (1) an external factor, such as bu- tyric acid, which causes a cytolysis of the cortical layer of the egg followed by increased oxidation and which leads to the rapid disintegration of the egg at normal temperatures, and (2) an internal factor, such as hypertonic solutions, lack of oxygen, etc., which inhibits this disintegration. Loeb concludes also that in normal fertilization both of these factors are present and that the spermatozoon carries substances into the egg which (1) cause cytolysis of the cortical layer and increased oxidation and (2) other substances which inhibit this cytolysis before it leads to the disintegration of the egg. Godlewski (’11) also finds that the cytolysis which is caused by fertilizing the eggs of sea ur- chins by the sperm of Chaetopterus may be checked and arti- ficial parthenogenesis induced by a brief treatment of such cross fertilized eggs with hypertonic sea water. R. 8. Lillie (11) concludes that the cortical changes consist in increased permeability of the cell membrane which tends ‘‘to de- stroy the normal osmotic equilibrium and allow abnormal dif- fusion of substances into and out of cells.’ The essential re- sult of the after treatment of such eggs with hypertonic sea water is to decrease the permeability of the cell membrane and thus restore normal conditions. F. R. Lillie (12) holds that “the action of the spermatozoon in fertilization involves two distinct phases, the first of which may be effected before penetration and brings about a sudden and marked increase in permeability of the egg membrane; the second, which follows after penetration, consists essentially in the establishment of normal interchange between nucleus and cytoplasm, and consequently normal regulation of all the ac;. tivities of the cell.”’ More recently (’13) he has put forth jg new view based upon the reactions of spermatozoa to substances secreted by the ova. He concludes that the ‘lysin’ which causseg cytolysis is contained in the egg, not in the sperm, as Leeb thought; “‘if cytolysis is involved it is a case of autocytolysis;,” My experiments on the giant polar bodies of Crepidulas show that changes in the cortical layer which prevent the entrarqce of a second spermatozoon take place very rapidly over tme entire CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 329 egg, but that the spermatozoon which enters does not cause any portion of the egg to develop except the cell in which it lies. Although the spermatozoon enters the egg of Crepidula about three hours before the formation of the second polar body the influence of the spermatozoon on the egg protoplasm during this time is not sufficient to start development in the second polar body even though it may contain the greater part of the egg substance.’ This indicates that the second factor concerned in the process of normal fertilization is not to be found in the diffu- sion through the egg of some chemical substance carried in by the spermatozoon, but rather in some non-diffusable substance, prob- ably an organic structure. Long ago Boveri (’87) showed that under certain circumstances the egg of Ascaris may divide at the first cleavage so that half of the egg nucleus passes into each daughter cell while the sperm nucleus does not divide, but goes entire into one of the first two cells. Such a condition he called ‘partial fertilization,’ and in such cases he found that both halves of the egg develop, thus showing that the activating influence of the spermatozoon has affected both halves. Since in this case the centrosome is the only structure derived from the spermatozoon which is known to go into both cleavage cells he reached his well known conclu- sion that the essential thing in fertilization is the addition of a centrosome to the egg cell. It is possible of course that other as yet unrecognized struc- tures are introduced by the spermatozoon and serve to activate the egg. Meves (’11) found that the spermatozoon of Ascaris introduces into the egg a number of coarse granules, the ‘plasto- chondria,’ which he thinks unite with similar granules in the egg and are then distributed to the cleavage cells. However, in one of the Echinids he finds that the large granule or ‘plasto- some’ which is derived from the middle-piece of the spermato- zoon goes into one only of the first two cleavage cells and yet both develop. I have found that the granules in the eggs of gastropods and ascidians which are presumably identical with ‘plastosomes’ or ‘mitochondria’ may be distributed very un- equally to the first two cleavage cells without interfering with 330 EDWIN G. CONKLIN the further division of both cells, and there is no evidence what- ever that the activating influence of the spermatozoon is due to these granules. On the other hand many investigators have held that fertili- zation is essentially a chemical process and that the activation of the egg depends upon the introduction by the spermatozoon of certain chemical substances which diffuse throughout the egg. The observations recorded in this paper indicate that the sec- ond or internal factor in normal fertilization is a non-diffusible substance which is introduced by the spermatozoon, and they strongly suggest, though they do not prove, that this factor is the sperm centrosome, a position which Boveri has long main- tained and which I have hitherto contested. 5. Cleavage of eggs centrifuged during maturation stages (figs. 33 toa, 79; 105) If centrifuging ceases long enough before cleavage begins to allow a return of the egg substances to their usual positions, the cleavage will be absolutely normal, irrespective of where the polar bodies may have been formed. Thus in figures 35 to 44, 47, 49 to 54 the cleavage is proceeding in a wholly normal man- ner although one or both of the polar bodies were formed at a distance from the animal pole, and in some instances even at the vegetal pole (figs. 35, 36, 38, 39, 44, 105). In figure 40 the effects of centrifuging persisted throughout the first cleavage so that one of the daughter cells contains more cytoplasm and less yolk than the other one, but such eggs may develop into normal embryos, as I have shown elsewhere (712). Even in such eggs as those shown in figures 44, 45, 49 in which the vol- ume of the ‘polar body’ may be greater than that of the ‘egg,’ the cleavage of the latter is perfectly normal except for the smaller size of the blastomeres formed and for alterations in the relative quantities of cytoplasm and yolk. In figure 44 the giant polar body was formed at the vegetal ‘pole which is still marked by the protruding ‘yolk lobe;’ in figure 45 the giant polar body lies at the animal pole and contains almost all the CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 331 yolk; and yet in these two cases, in which the polar bodies lie at opposite poles, the cleavage of the egg up to the 4-cell stage is normal except for the relative amounts of yolk and cytoplasm. The same thing is shown in the later stages of cleavage shown in figures 47, 49 to 54; in spite of the fact that one or both of the polar bodies were formed at a distance from the animal pole, every cleavage takes place in a perfectly normal manner except for slight differences in the relative amounts of cytoplasm and yolk. Figure 48 is an abnormal egg in which the ectodermal pole lies about midway between the first and second polar bodies; the abnormal cleavage of this egg is probably due to an unequal dis- tribution of cytoplasm and yolk to the first two blastomeres and to a partial separation of one of the blastomeres from the others. Figures 55 to 57 represent eggs which were centrifuged dur- ing the second maturation division, a giant polar body being formed in each case, and in which the cleavage is more or less abnormal. In figure 55, which corresponds to a 24-cell stage, the cleavage is normal in one half (left) of the egg, but abnormal in the other half (right); the polarity in these two halves dif- fers, the ectodermal pole being above in the left half and at the right margin in the right half. In figure 56 four macromeres may be recognized, each of which has produced one micromere, but owing to the partial separation of these macromeres the chief axes are not parallel in the four quadrants; in all probabil- ities the normal polarity of each macromere is unchanged, but in the process of separation it has been twisted out of its normal relation to the other macromeres. Figure 57 is a 12-cell stage in which there has been a partial separation of two of the macro- meres from the other two and in which there is an abnormal dis- tribution of cytoplasm and yolk; the abnormalities of the cleay- age are referable to these two factors. Figure 105 is a 22 to 24-cell stage of an egg which was cen- trifuged for 23 hours during the maturation divisions and was then allowed to develop for 18 hours longer. Both polar bodies, one of them large and containing a spindle, lie at the vegetal aan EDWIN G. CONKLIN pole at the lower side in the figure. The probable identity of the cleavage cells is indicated by the labeling and the arrows. On the whole then it may be concluded that changes in the maturation pole induced by centrifuging have no lasting influ- ence on the polarity of the egg and do not modify the normal type of cleavage; such modifications of cleavage as do occur may be attributed to, (1) the prevention of the return of nuclei and cytoplasm to the animal pole by long continued centrifug- ing or by the formation of division walls, (2) abnormal distri- bution of cytoplasm and yolk to the first two blastomeres and the permanent separation of these substances by partition walls, or (3) the partial separation of blastomeres of the 2-cell or 4-cell stages. 6. The maturation pole does not determine the animal pole of the egg nor the ectodermal pole of the embryo (figs. 35 to 58) Probably in all animals the polar bodies are formed at the animal pole of the egg and the latter becomes the ectodermal pole of the embryo. However these experiments prove that the polar bodies may be forced to form at any point whatever on the surface of the egg without changing in the least the location of the animal or ectodermal pole. Thus in figures 35 to 41, which represent eggs which were centrifuged for 4 hours and fixed 6 hours later, the original animal pole is clearly indicated by the position of cytoplasm and nuclei and yet one or both of the polar bodies were extruded at a distance from this pole and after- ward cytoplasm and nuclei returned to the original animal pole. In figures 47 to 54 the center of the plate of micromeres (ecto- meres) is the ectodermal pole and yet this lies some distance from one or both of the polar bodies. The evidence that it is the maturation pole and not the ani- mal pole which has been moved in these eggs is the following: 1) The polar bodies are larger than normal, showing that they were formed during centrifuging; many of them contain cytoplasm and oil, thus proving that they were formed at the centripetal pole; it is certain that maturation spindles in the early stages may be CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 333 centrifuged away from the animal pole (figs. 19 to 21). 2) After centrifuging, the eggs were kept in normal condition from 6 to 24 hours, during which time the cytoplasm and nuclei moved to their present positions; the fact that this movement takes place under normal conditions indicates that it is a move- ment back to the normal animal pole. 3) The most important evidence that the maturation pole and not the ectodermal pole is shifted by centrifuging is to be found in those cases in which one polar body lies at the ectodermal pole and the other is more or less distant from it. Such a condition is shown in figures 33, 35, 41, 47, 54, 58. In these cases the polar body lying at the ectodermal pole is quite normal in size and appearance, whereas the other polar body is much larger than normal. The plain import ‘of this is that a normal first polar body was formed at the animal pole before the egg was centrifuged, that an abnor- mally large second polar was formed at the pole to which the cytoplasm and the spindle were displaced and that the cytoplasm and the germ nuclei have moved back to the animal pole after the eggs were removed from the centrifuge. The animal pole is thus marked by the first polar body and though the second polar body may be formed far from this pole it does not change the original polarity of the egg, nor of the embryo which de- velops from it. These facts constitute, I think, indisputable evidence that the polarity of the egg and embryo is not the result of the formation of polar bodies at a particular point, but rather this polarity of the egg antedates the maturation and under normal conditions is the cause of the location of the maturation spindles and polar bodies as well as of the ectodermal pole. The maturation pole does not necessarily coincide with the animal pole of the egg, nor does it determine the ectodermal pole of the embryo. 7. Results of centrifuging after maturation and before the first cleavage (figs. 28 to 30, 59 to 62) In normal eggs cytoplasm continues to segregate at the ani- mal pole and yolk at the vegetal pole throughout the periods of maturation, fertilization and early cleavage. The egg nucleus and centrosphere lie close to the animal pole and consequently 334 EDWIN G. CONKLIN in a cytoplasmic field, while the sperm nucleus and centrosphere approach the animal pole through the yolk in the vegetal hemi- sphere of the egg. Other things being equal, the sizes of these germ nuclei and centrospheres depend upon the volume of cyto- plasm in which they lie; consequently in normal eggs the egg nucleus and centrosphere are larger than those of the sperm (fig. 28). If eggs are centrifuged in the anaphase or telophase of the second maturation division, both egg and sperm nuclei remain unmoved while yolk may be driven to the animal pole and cytoplasm to the vegetal pole. Under these circumstances the egg nucleus comes to lie in a field of yolk and remains cor- respondingly small, while the sperm nucleus lies in the cyto- plasm and grows large (figs. 23 to 25, 29, 30). In the earlier phases of their growth both egg and sperm nuclei are moved but little by centrifugal force (fig. 59); in later stages both nuclei move more freely (figs. 60 to 62). This is probably due to the fact that in the earlier phases mitotic fibers still bind the nuclei to the surface layer, whereas in later stages these relax. Even after the germ nuclei have become quite large they may occasionally be seen to be held by a cytoplasmic framework, which connects the egg nucleus to the animal pole, as in figure 60, and which prevents the free movement of the nuclei so that the latter become stretched and distorted under centrifugal force (figs. 60, 61). Sometimes strands of this framework may be seen running through all the zones of the centrifuged egg (fig. 60). On the other hand, if centrifuging continues for a long time the strands become less evident and the nuclei again as- sume a spherical form (fig. 62), as is also the case when centri- fuging ceases. In all these cases nuclei and cytoplasm come back again to the animal pole after centrifuging. This move- ment takes place especially during mitosis, which probably in- dicates that this orienting framework is stronger or more active during mitosis than during resting stages. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA aon lil. RESULTS OF CENTRIFUGING DURING CLEAVAGE STAGES An extended study has been made of eggs that were centrifuged at various stages during the first four cleavage periods. Some of these results, which bear on the relation of cell-size to nu- clear-size, have appeared in a previous publication (Conklin, 712). In this place we shall consider only the bearings of these experiments on the polarity and pattern of organization of the _ ege. 1. Modifications of first and second cleavages: equatorial cleavages (figs. 68 to 82, 128) In normal eggs of Crepidula the first two cleavages are merid- ional and nearly equal and they are followed by three very unequal cleavages by which three sets of micromeres (ecto- meres) are cut off at the animal pole from the large macromeres at the vegetal pole. The direction of every cleavage and its equality or inequality depend upon the direction and position of the mitotic figure and this is controlled by many factors among which the most important are the axes of nucleus and centrosome before division and the relation of these to the polarity and structure of the cytoplasm. At the close of every cleavage (telophase) the centrosphere, nucleus and cytoplasm rotate in each daughter cell in such a manner as to bring the centrosphere to the free border of the cell and as near as possible to the animal pole (Conklin ’98, ’02). The mitotic spindles of the first and second cleavages can be moved from their normal positions by strong centrifugal force, though they are sometimes bent and distorted as a result of this and the asters or centrospheres are usually elongated toward their original positions, thus indicating that they are still con- nected in some way to those positions (figs. 63 to 65). The entire spindle may be moved from its normal position or one pole may be moved and the other remain relatively station- ary. Consequently the spindle may be turned into any axis. In figures 63 and 64 the entire spindle has been moyed toward the vegetal pole, the centrospheres stretching back toward their 336 EDWIN G. CONKLIN normal positions; in figure 65 one pole of the spindle has been displaced more than the other one and the spindle is therefore oblique to the chief axis of the egg; in figures 66 ef seq. the spindle was moved into the chief axis and the resuuine cleavage is nearly equatorial in position. Equatorial cleavages are of particular interest since they afford an opportunity of studying critically the polarity of the egg and the potency of its different parts. When the first (or second) cleavage is changed from a meridional to an equatorial one there is a totally new collocation ofnuclei and cytoplasmic substances with respect to the original axis and poles of the egg. Never- theless, the original polarity is preserved almost unchanged or if modified at all is subsequently restored. In the cell below the equator, cytoplasm, centrosphere and nucleus move during telokinesis to the free surface and as near as possible to the animal pole; in the cell above the equator a similar movement is limited apparently by some attraction on the part of the proto- plasm of the lower cell so that cytoplasm, centrosphere and nucleus come to occupy a position more or less intermediate be- tween the lower cell and the animal pole (figs. 67 to 73). If an egg in which the first cleavage was equatorial is freed from pressure and permitted to develop, the second cleavage will be meridional (figs. 71, 72, 75 to 78) and each of the four macromeres thus formed will give rise to three sets of micro- meres (ectomeres) precisely as in a normal egg, except that the micromeres of the cells below the equator are not able to reach the animal pole though they move as far as possible in that direction (figs. 77 to 81). Figure 123 represents an egg in which the second cleavage was equatorial in position. The four macromeres have given off the first set of micromeres (/a—1d) and are in process of forming the second set (2a—2d); the posi- tions of the spindles and of the cleavage cells are somewhat abnormal in most of the cells. Whatever the factors are which in normal eggs determine the equality of the first two cleavages and the inequality of succeeding cleavages, as well as the polarity of egg and cleavage cells, these factors are still present in centrifuged eggs in which the first or second cleavage CENTRIFUGAL FORCE ON EGGS OF CREPIDULA Sor was an equatorial one. The internal organization which de- termines the position of spindles and the size and position of daughter cells is still present and active, however much it may have been distorted or disturbed. But although this organization persists in centrifuged eggs, it - does not persist as a constant structure which is unaffected by the manner in which it is cut by successive cleavages. If the organi- zation of the egg were antecedent to the cleavage and were in nowise changed by the way in which it is cut by the cleavage furrows, the cell which is formed below the equator should give rise to no ectomeres and the one above the equator should pro- duce all of the ectomeres, and all of these should lie at the ani- mal pole. But this does not happen, the cell below the equator gives rise to its normal number of ectomeres just as the one above the equator does. The dislocation of the first or second cleavage does not change the differential character of. the follow- ing cleavages. 2. Results of centrifuging during resting stages between first and second and second and third cleavages (figs. 83 to 92, 95, 96) All cell constituents may be dislocated more readily by cen- trifugal force during periods of interkinesis than during mitosis, consequently when eggs are centrifuged in resting stages be- tween cleavages, not only are yolk and cytoplasm displaced, but also nuclei and centrospheres. In figures 83 to 92 are shown eggs which were centrifuged in various axes in the 2-cell stage; figures 95 and 96 were centrifuged in different axes in the 4-cell stage. The polar bodies mark the animal pole and in every egg shown the different cell constituents are more or less displaced from their normal positions. But the manner of this displace- ment shows clearly that these constituents are not free to stratify according to their specific weights. While the major portion of the yolk goes to the centrifugal pole and of cytopalsm to the centripetal pole the boundary bétween these is not a plane surface, but there are ‘lanes’ or projections of cytoplasm which remain connected with the animal pole even though most THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 2 338 EDWIN G. CONKLIN of the yolk may have been driven to this pole and most of the cytoplasm to the opposite pole (figs. 83 to 91). Similarly, resting nuclei and centrospheres do not move freely through the cell under the influence of centrifugal force, but are evidently limited in their movements by attachments similar to, but less strong than, the astral radiations of the mitotic figure. The centrosome is always closely attached to the chromosomes, the centrosphere to the resting nucleus, and it is very difficult to separate the two completely. Furthermore, both centrosome and centrosphere are attached by astral radiations to a free sur- face of the cell and in the telophase of the division this place of attachment is carried as near as possible to the animal pole. Consequently, throughout the resting period the centrospheres are attached on one side to the nucleus and on the other to the free border of the cell lying nearest the animal pole; these at- tachments persist even in centrifuged eggs and although © nucleus and centrosphere may be forced away from the animal pole side of the cell, these attachments are not destroyed, but are merely stretched, as is shown by the lane of cytoplasm lead- ing toward the animal pole, by the elongation of the centro- sphere in this direction and by the return of centrosphere and nucleus to their normal positions after centrifuging (figs. 87 to 90). In figures 84 and 85 the axis of centrifuging was parallel with the first cleavage plane and nearly at right angles to the egg axis; in figures 83, 86, 87, 89, 90, 92 the axis of centrifuging was in the chief axis of the egg, the animal pole being centrifugal and the vegetal pole centripetal; in figures 88, 91, 96 the axis of centrifuging was at right angles to the first cleavage plane and to the egg axis; in figures 95, 97, 98 it was at right angles to the second cleavage plane and to the egg axis. These axes of cen- trifuging are all the principal axes of the egg in which an abnor- mal distribution of cell substances can be brought about, and yet in every case the centrospheres lie between the nucleus and the animal pole and in some instances the strands connecting these can be clearly seen (figs. 87 to 90, 95 to 98). Figure 86 is especially interesting because the mid-body (MB) and spindle remnants are present as well as the nuclei and centrospheres. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 339 The mid-body has remained unmoved by the centrifuging, but the nuclei were carried nearly to the vegetal pole, the centro- spheres lie on the animal pole side of the nuclei and the spindle remnant in each cell still connects the mid-body and nucleus, though in this case the nucle: have been moved from the animal to the vegetal pole and the spindle remnant has been bent through an angle of nearly 180°. Figure 92 is especially interest- ing because one of the two cells contains two nuclei and one centrosphere, while the other contains a centrosphere but no nucleus. Evidently this centrosphere was separated from its nucleus before the division of the cell body began; the cell con- striction began at the vegetal pole rather than at the animal pole because cell constrictions always begin on ‘that side of a cell where the cytoplasm is most abundant. 3. Later cleavages of eggs centrifuged in the two or four cell stages (figs. 104) 107 to 112) In most cases cell substances slowly come back to their nor- mal positions after centrifuging and subsequent cleavages are quite normal; but if the division of the cell body is suppressed or if the macromeres are separated or dislocated, the later cleavages are quite abnormal. Such cases are shown in figures 104, 107 to 112. In figure 104 the second cleavage furrow was suppressed in the right half and at the third cleavage a large protoplasmic cell was formed at the animal pole; the cleavage in the other half of the egg is absolutely normal. In figures 107 and 108 the second cleavage furrow was suppressed in one or both macromeres and as a result some of the later cleavage cells con- tain multiple nuclei or tetrasters, and are more or less abnormal in position and time of division. Nevertheless, every cell may be identified with a corresponding cell of the normal egg. In figures 109 and 111 the second cleavage was rendered quite un- equal, but the following cleavages were very nearly normal ex- cept that some of the cells of figure 111 contain multiple nuclei. In figures 110 and 112 the four macromeres were dislocated, with the result that the micromeres form two separate groups; how- 340 EDWIN G. CONKLIN ever, the micromeres formed from each macromere are normal in number and relative position as is best shown in figure 110. In none of these eggs is there any evidence that the polarity or pattern of organization has been changed in any quadrant, though the relations of the different quadrants to one another is changed. 4. Results of centrifuging during the third and fourth cleavages (figs. 97 to 104, 113 to 122) Centrifuging during the third and fourth cleavages is of especial interest because these cleavages are under normal conditions plainly differential and lead to the formation of the first and sec- ond groups of micromeres (ectomeres). In this case as in every other one, the cytoplasm and yolk may be moved more easily than the mitotic figure or the resting nucleus and centrospheres. In figures 97 to 102 eggs are shown whichwere centrifuged dur- ing the third cleavage mitosis; in figures 97 and 98 the axis of centrifuging was at right angles to the egg axis and the fibers connecting the upper pole of the spindles with the animal pole are clearly shown in those cells in which yolk was forced to the animal pole. A later condition of such an egg is shown in figure 100, in which the micromeres /c and /d were formed some distance from the animal pole owing to the fact that yolk was forced to that pole in the macromeres 1C and 1D. In figure 99 the axis of centrifuging was in the direction of the egg axis, the animal pole being centrifuged in position; consequently yolk was driven to the animal pole and cytoplasm tothe vegetal one. The further development of such an egg is shown in figures 101 and 102 in which the first set of micromeres (/a—1d) are yolk laden and much larger than normal. Figure 102 is especially instructive because it shows the second set of micromeres (2a— 2d) being formed in a manner entirely normal in quadrants B and C although the size and contents of the first set of micro- meres (1b, 1c) was very abnormal in these quadrants. Further- more, the orientation of the spindles in the cells 1b, 1c shows that a small protoplasmic cell will be formed at the upper pole of this spindle, a large yolk cell at the lower pole, which is just CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 341 the reverse of what takes place in normal eggs. This condition may be explained by assuming that the wpper pole of the spindle maintains its normal position by virtue of tts attachment to the cell surface at the animal pole, whereas the lower pole of the spindle is relatively free. This condition obtains in practically all stages of maturation and cleavage and it explains one of the most perplex- ing problems regarding the orientation of the spindle. The at- _tachment of the spindle to the cell surface at the aniaml pole side of the cell is always stronger and more persistent than its attach- ment at any other point. This is probably due to the greater con- centration of spongioplasm at the animal pole. The egg shown in figure 103 was centrifuged after the com- pletion of the third cleavage. ‘The micromeres are entirely nor- mal and since they contain no yolk their contents are not dis- placed, but in the cells 7C and 1D the yolk is thrown to the animal pole and the protoplasm is correspondingly displaced. In every cell it is.plain that the centrospheres have been dis- placed least of all the cell constituents. Figures 113 to 122 represent eggs which were centrifuged for 5 hours in the 4-8 cell stage; the axis of centrifuging was in the chief axis of the egg, the animal pole being centrifuged in position. In all cases the cytoplasm and nuclei or mitotic figures of the macromeres were carried through these cells to the vegetal pole and were kept in this position until one or more sets of proto- plasmic micromeres had been formed at the vegetal pole. Fig- ures 113 to 118 were fixed immediately after centrifuging; fig- ures 119 to 122, 5hours later. Figures 113 to 116 are viewed from the vegetal pole, the polar bodies being shown in dotted outline on the farther side of the egg; figures 117 and 118 repre- sent the same egg, the former showing the cells at the animal pole, the latter those at the vegetal pole. Figures 119 to 122 are viewed from the animal pole, the cells at that pole being shown in heavy outline, while those at the vegetal pole are shown in light or dotted outlines. In figures 113 two of the cells, A and D, have given off near the vegetal pole small cells containing protoplasm and oil, and similar cells are in process of being formed from the cells B and C. Figure 115 shows four 342 EDWIN G. CONKLIN protoplasmic cells at the vegetal pole, each of which is dividing or has just divided, while the four macromeres are giving off a second set of small protoplasmic cells at this pole. It is prob- able that these small cells are micromeres (ectomeres) of the first and second quartets. Figures 114, 116 to 122 were centrifuged after the formation of the first quartet at the animal pole, which is shown in faint or dotted outline in figures 114, 116, 118; in figure 114 four protoplasmic cells, which probably represent the second quartet, ie at the vegetal pole; in figure 116 the second quartet cells are subdividing and a third quartet is being formed; in figure 117 are shown at the animal pole eight cells, four ‘cen- trals’ and four ‘turrets,’ derived from the first quartet, while twelve cells of the second and third quartets are shown in faint outline at the vegetal pole; these twelve cells, which represent eight cells of the second and four of the third quartet, are shown in figure 118 as they are seen through the egg. If the small cells formed at the vegetal pole in the preceding figures are really micromeres (ectomeres) it should be possible to allow the first. quartet to form normally at the animal pole, to then force the second quartet to form at the vegetal pole, and finally to allow the third quartet to come back and form at the animal pole. This is just what has happened in one or more quadrants of the eggs shown in figures 119, 121, 122; in these three eggs the first quartet, which has now subdivided into four ‘central’ and four ‘turret’ cells, lies at the animal pole; the second quartet, each cell of which has subdivided, lies in the furrows between the macromeres on the vegetal side of the egg, while in figures 121 and 122 the third quartet has formed outside the first quartet on the animal side of the egg. There can be little doubt, from the manner and time of division of the macro- meres and micromeres in these eggs, that the micromeres on the vegetal. side of the egg are typical second quartet cells ex- cept in respect to their position. On the other hand in figure 120 the first quartet was formed normally at the animal pole, but centrifuging was not sufficiently strong to carry the nuclei or mitotic figures for the fourth cleavage clear through the macro- meres to the vegetal pole; as a result the fourth cleavage was a CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 343 nearly equal one, the second quartet cells (2a—2d), if they may be called such, being large and full of yolk. 5. The potency of substances, regions and blastomeres of centrifuged eggs From the preceding account it: is evident that the fate of any cell in development is not determined by the amount of yolk or cytoplasm which it contains. Typically these two substances are divided equally to the first four blastomeres, but if all of the yolk is thrown into two of these blastomeres and most of the cytoplasm into the other two the ensuing development may be nearly normal. . There is no evidence that the differentiation of a cell is deter- mined by one daughter nucleus or centrosome. being different from the other one, for by means of centrifugal force the spindle of a differential cleavage, such as the third, may be forced to take a position so that the micromere will be formed at the lower pole of the spindle instead of the upper pole as normally, or it may be forced into an equatorial position so that the cleavage is into two macromeres and is not differential; in short the mi- cromere may be formed at either pole of the spindle or at neither pole, depending upon its position. The fate of a cell is not merely a ‘function of its position,’ as Driesch maintained, but it is in the main a function of its differ- entiation at the time of its formation; this differentiation de- pends primarily upon the stage in development at which the cell has.arrived and only secondarily upon the direction of cleavage and the position of the blastomere. Whatever the position, direction or differential character of the first two cleavages may be, the three following cleavages, if freed from outer force, are very unequal, giving rise to three micromeres (ectomeres) at the animal pole side of each cell. These differential cleavages can not occur before the third nuclear division and if at the first or second cleavage a small cell is forced to form at the animal pole it behaves like a macromere and not like a micromere. On the other hand if a third cleavage plane 344 EDWIN G. CONKLIN is forced into a meridional position so that eight macromeres are formed, each of these will give rise to three micromeres, just as each of the four macromeres does in the normal egg. If con- versely the second cleavage furrow is suppressed, but the nuclear division is not, each of the two daughter nuclei in each macro- mere may divide in such a way as to give off the regular number of micromeres from these two macromeres. All this proves that the formation of micromeres at the third cleavage is not due to the segregation of a peculiar ‘micromere substance’ at the animal pole, for in whatever plane the first two cleavages may divide the egg or in whatever axis the egg substances may be displaced by centrifugal force, each macromere, if freed from external pressure, still give rise to a typical micromere at the third cleavage. The localization and orientation of the mt- totic spindle, which determines the size-and position of the cleavage cells depends upon the viscid spongioplasm or kinoplasm which connects nucleus and centrosphere with one another and with the peripheral layers rather than upon the more movable constituents of the cell; and furthermore, since the orientation of the spindle dif- fers in successive cleavages in a characteristic manner, the orien- tatcon of this viscid protoplasm must also differ. Consequently, we may conclude that there is an inherent col- location of spongioplasm in every cell which determines the for- mation of micromeres at the third cleavage, and this does not occur before the third cleavage, since the differentiation of spon- gioplasm has not proceeded far enough at the first and second cleavages to make possible the formation of micromeres. I have shown elsewhere (’12) that if a blastomere is isolated in the 2-cell stage, the second cleavage is normal and gives rise to two typical macromeres; if one is isolated in the 4-cell stage, the three following divisions of that cell produce micromeres just asin anormal egg. But no normal micromere can ever be gotten from a cell before the period of the third cleavage. In short, the organization of the kinoplasm in a blastomere of the 4-cell stage differs in some essential way from that in an earlier or later stage, —it has reached a certain peculiar stage of differentiation. Di- rect observation of eggs in various stages from the time of fer- CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 345 tilization up to the period when all the micromeres are formed shows conclusively that one feature of this progressive differen- tiation of the blastomeres consists in the continuous segregation of cytoplasm at the animal pole and of yolk at the vegetal pole. Before cleavage begins there is a very small area of pure cyto- plasm at the animal pole and the entire cell stains with cytoplasmic stains, thus indicating that the segregation of cytoplasm and yolk at the two poles is far from complete; at the 4-cell stage and still more at the 8, 12 and 20-cell stages this segregation is much more complete, the area of cytoplasm at the animal pole is increasingly large and the yolk area stains but littlg with cytoplasmic stains. Nevertheless, the differentiation which leads to the formation of micromeres does not depend upon this segregation only, for if the segregation of yolk and cytoplasm is brought about by centrifugal force at the 1 or 2-cell stage it never leads to the for- mation of micromeres. Even if almost all of the cytoplasm is thrown into two of the cells at the second cleavage and all of the yolk into the other two, each of these four cells gives rise at the third cleavage to a typical micromere. It is evident therefore that micromere formation depends on something other than the segregation of cytoplasm at the animal pole and of yolk at the vegetal pole. The size of a micromere is fairly constant and is within limits independent of the size of the macromere from which it comes. This is probably due to the fact that the upper pole of the spindle lies at a constant distance from the animal pole. If this dis- tance is forcibly increased or decreased the size of the resulting micromere may be increased or decreased, and if these spindles are turned into a horizontal position the third cleavage planes may be meridional, thus giving rise to eight macromeres, each of which may then give off three micromeres. This case shows clearly that the differentiation of a blasto- mere is not due to the differentiation of its nucleus, nor is it wholly due to the position of the cleavage plane, but rather it is caused by a progressive change in the spongioplasm, which change is normally associated with certain mitoses. If the mitoses go on, but the division of the cell body is halted the 346 EDWIN G. CONKLIN differentiation of the spongioplasm proceeds at least for a time, but if the mitoses are stopped the differentiation of the spongio- plasm is also stopped. There is no indication that the nucleus is Ac wectis differen- tiation during cleavage; the nucleus in a micromere is evidently of the same character as its sister in the macromere, as is proved by those cases in which the third cleavage spindles are forced into an equatorial position, thus giving rise to eight macromeres; the daughter cells in this case are all macromeres in that-each gives off three micromeres, and the nucleus which would have gone into a micromere ynder normal conditions now goes into a macromere without in any way changing its future differentia- tions. In short the nuclei of macromeres and micromeres are not differentiated at the time of their formation, but may be thrown about, as Driesch has said, ‘like balls in a pile’ without changing the fate of any of the cells into which they go. It is thus possible to show that the differentiation of a blasto- mere does not depend upon the differentiation of its nucleus nor does it depend entirely upon the segregation of cytoplasm and yolk, nor upon the direction or position of the cleavage plane. After differentiation and localization of cytoplasmic substances has already occurred the direction of cleavage is an ae factor, but not before. lV. GENERAL CONCLUSIONS 1. The nature and causes of cell polarity Polar differentiation, or more briefly polarity, may be defined as the condition of having unlike poles, particularly in the chief axis of a body, while symmetry is the condition of having lke poles in certain axes. The polarity of an entire organism, or of any of its parts, is an expression of the relative positions with respect to the chief axis of subordinate parts having different structures and functions; in this sense it is customary to speak of the polarity of organisms, organs, cells, nuclei, ete., but there is no evidence that the polarity of the entire organism is the re- sultant of the polarities of its constituent parts, as is the case with magnetic polarity. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 347 Organic polarity may be viewed from the standpoint of struc- ture or of function; neither of these aspects is complete in itself and neither is at variance with the other but the two are complemen- tary. In a series of important papers (11-15) and in a recent book (’16) Child has announced that the polarity of various adult organisms and of certain eggs and embryos is shown physiologically as a gradient in the rate of metabolism from one pole to the other. But in spite of this important discovery polarity can not be looked upon asa physiological process merely ; there must be a material, structural basis for such a gradient of metabolism; furthermore it has not been demonstrated that physiological differentiations are the causes of morphological ones, for although functional changes are often more readily visible than structural ones, there is every reason to think that structure and function are inseparable in living organisms and that neither is the cause of the other. In this paper attention is devoted largely to the morphological aspects of polarity, but it is not to be assumed therefore that the author considers the physiological aspects as negligible. The polarity of the egg cell is the earliest recognizable and most fundamental differentiation of morphogenesis; it is the chief factor in determining the localization of developmental processes, such as the segregation of different odplasmic sub- stances and of specific physiological activities, the orientation of mitotic figures and cleavage planes, and finally the determina- tion of the polarity and symmetry of the embryo and of the adult. In short the’polarity of the organism in the one-celled stage of development is the chief condition and cause of the polarity of all later stages. In many animals the polar differ- entiation of the egg may be recognized even in the stages of its development in the ovary and it is probable that in all cases such polar differentiation exists at this time. When first recognizable this differentiation usually consists in the eccentricity of the nucleus and centrosome and of the greater part of the cytoplasm toward one pole of the egg and the greater accumulation of yolk at the other pole, the former being known as the animal pole and the latter as the vegetal. Dur- 348 EDWIN G. CONKLIN ing the period of maturation, preceding or accompanying fer- tilization, two minute cells, the polar bodies, are formed at the animal pole of the egg, and in all animals from sponges to mam- mals the animal pole gives rise to the ectoderm and the vegetal pole to the endoderm of the embryo. But while the polazity of the embryo as a whole corresponds directly to the polarity of the egg from which it develops the axes of the constituent cells of the embryo are shifted in successive cell generations so as to form all possible angles with the chief axis of the embryo. The axis of a cell is usually marked out by the line passing through the center of the resting nucleus and centrosome, this being the ‘cell axis’ (Van Benedan ’83, Heiden- hain 794). In cleavage cells the resting centrosome, or centro- sphere, comes to lie at the free border of the cell at a point as near as possible to the animal pole, consequently in cells lying near the animal pole the cell axis is nearly parallel with the chief axis of the egg, but in cells which le near the equator of the egg the cell axis may be nearly at right angles to the egg axis and in cells which lie near the vegetal pole the cell axis may be nearly the reverse of the egg axis. As long as the cleavage cells are rela- tively large it is possible to see that the resting centrosomes always lie at that point of the free surface of the cell which is nearest the animal pole, but when these cells become very small it is no longer possible to see that the centrosomes are turned toward the ani- mal pole although they always lie at the free border of the cell. Similarly in tissue cells it can be seen in many cases that the cen- trosomes lie at the free border of the cell (Heidenhain and Cohn 97), but there is no evidence of their approximation to the origi- nal animal pole. The same is true of the various stages in the formation of the germ cells, whether ova or spermatozoa, the centrosomes always lie on the side of the nucleus toward the free border of the cell and away from its attached side, but no ap- proximation of the centrosome to the original animal pole can be traced. The resting nucleus shows polar differentiation, as was pointed out long ago by Rabl (’85), the pole which les nearest the cen- trosphere being known as the ‘Pol’ or central pole, the opposite CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 349 one as the ‘Gegenpol’ or distal pole. Experiments show that the centrosphere is attached rather firmly to the central pole of the nucleus and however much the relative positions of the different parts of the cell may be changed this attachment of centrosphere to nucleus is rarely broken. Within the nucleus the developing chromosomes are aggregated chiefly at the cen- tral pole, the achromatin at the distal pole. The centrosphere also shows a polarity of its own, its chief axis being that in which the daughter centrosomes move apart and form the initial spindle. This initial spindle, or ‘netrum’ of Boveri (’00), usually lies at right angles to the cell axis and to the preceding spindle axis. But the axis of the fully developed spindle within the cell may differ from that of the initial spindle since there are characteristic movements of the cytoplasm of every cell which transport the spindle into its definitive position. During the last phase of nuclear division (telophase of Heiden- hain ’94) the daughter centrosomes and nuclei turn back toward the original cell axis, the centrospheres remaining attached to the nuclei at their central poles and moving to that point on the free surface of the cell which is nearest the animal pole; the new cell axes which are thus formed are approximately but not exactly parallel with the old cell axis. In egg cells and cleavage cells the cell axis does not change every time the centrosomes separate in division, but this axis remains relatively constant while the mitotic figures may form any angle with it, but at the close of division the centrosomes and nuclei come back once more into the cell axis. Thus the axis passing through nucleus and centrosome coincides with the cell axis only during the rest- ing stage of the cell. Although the cell axis is usually marked out by the position of the resting nucleus and centrosphere, it is not entirely dependent upon that position. By pressure or centrifugal force the posi- tions of nuclei and centrospheres may be changed without per- manently altering the cell axis, as is shown by the fact that these structures usually come back once more to their normal positions as soon as the pressure is removed. Similarly the position and direction of the mitotic figure and of the resulting cell division 350 EDWIN G. CONKLIN may be changed experimentally without changing the real cell axis. In short the cell polarity persists in the organization of the cytoplasm after the positions of centrospheres, nuclei, mi- _ totic figures and cleavage planes have been changed. But while in such cases the polarity of the cell persists in the cytoplasm, there is evidence that the polarity of the cytoplasm has devel- oped in connection with and in definite relation to the polarity of the nucleus and centrosome. With regard to symmetry it is known in a few cases, notably cephalopods and insects, that the egg is bilateral even in the ovary; in other cases such as amphibians and ascidians, bilater- ality first becomes apparent shortly after fertilization; in still other cases bilaterality does not become evident until later stages of the cleavage or even of the blastula or the gastrula. In the case of sinistral gastropods the inverse symmetry may be traced back in development to inversely symmetrical cleavage of the egg, indeed to the very first cleavage, and it is evident that the causes of this inversion must be present in the egg before cleavage begins (Conklin ’03). In addition to these general axial differentiations of polarity and symmetry other more specific differentiations of regions and substances of the egg exist in some animals. In ascidian eggs the substances which give rise to different organs and tissues, such as the nervous system, the chorda, the caudal muscles, the mesenchyme, the ectodermal and endodermal epithelium are definitely localized and may be clearly distinguished as early as the first cleavage (Conklin (05). Although these eggs show an unusual degree of differentiation at a very early stage, there are many others in which the ‘pattern of localization’ is present either before or just after cleavage begins. Among the gastro- pods generally the first cleavage separates an anterior blasto- mere (A B) from a posterior one (C D), the second cleavage divides these into right and left halves. Each of these four blastomeres gives rise to three ectomeres and to a large ento- mere, while the left posterior cell (D) gives rise also to the meso- mere (4d). Each of these cells produces in later stages a definite portion of the embryo so that the development is, as Wilson has CENTRIFUGAL FORCE ON EGGS OF CREPIDULA oo said, a ‘visible mosaic work.’ All of these orientations of de- velopment find their earliest visible expression in the polar dif- ferentiation of the egg; the problem of the causes of these orientations is perhaps the greatest problem of embryogeny. Causes of cell polarity Any satisfactory explanation of the causes of polarity and symmetry of cells must be able to explain the following phenomena : ray a. The typical localization of substances in cells, such as yolk at the vegetal pole and of cytoplasm and nucleus at the animal pole, together with the typical orientation of spindles, centrospheres and other cell constituents during and after mitosis. b. The return of all cell substances to their typical positions after they have been displaced if sufficient time and opportunity for this return are given, and if the injury to the cell is not too severe. It is therefore evident that the cause of polarity in cells is one of the most fundamental problems in the study of the structure and functions of cells and in the processes of differentiation and regulation. The localization of formative substances in eggs determines the localization of the parts of the developing embryo, and the return of these substances to their normal positions when once they have been displaced is a remarkable case of adaptation or regulation in which the organization concerned is merely the polarity of a single cell. Because of the apparent simplicity of this problem of the polarity of the cell, the hope is raised that a thorough analysis of it may throw light on the problems of differentiation and regulation in general. It is conceivable that the causes of this polarity may be due (1) to electric charges on the colloidal particles of protoplasm, or more particularly on centrospheres and cell membrane, or (2) to external or internal surface tension phenomena, or (3) to a framework of viscid protoplasm which is so elastic or contrac- tile that it recovers its normal form after distortion: The oo EDWIN G. CONKLIN evidence for or against these possibilities may be considered briefly. 1. Electric polarity: polarity of fused eggs. There is really little or no evidence that cell polarity is of the nature of electric polarity. Neither the entire egg nor any of its parts orients with respect to a constant current passed through the water in which eggs are placed, and none of the constituent parts of a cell is moved or oriented by an electric current passed through the cell itself unless that current is so strong that phenomena of convection occur (see Conklin, 712). A certain amount of light is thrown upon the nature of the polarity of the egg by the effect on this polarity of the fusion of two or more eggs. If this polarity were the result of electric charges carried by colloidal particles or if it were due to differ- ent properties of the cell membrane at the two poles, the polarity of fused eggs should be the resultant of the polarities of its com- ponents. On the other hand if the polarity is due to an internal framework of protoplasm which is but slightly miscible with other similar frameworks and which therefore preserves to a great extent its identity, the original polarity of two or more eggs would not be much changed by their fusion. The facts show that the latter is the case. When eggs of Crepidula are strongly centrifuged they are pressed closely together and frequently adhere to one another. Very rarely they fuse together so that no boundary can be seen between the two. Thus figure 106 represents two eggs which fused together in the 4-cell stage along the plane indicated by the dotted line. Each egg is now in the 8-cell stage and each preserves its original polarity, as is shown by the polar bodies, micromeres centrospheres and nuclei. Although each egg is in the 8-cell stage, the fusion of three pairs of cells reduces the total number of separate cell bodies to 6 macromeres and 7 micromeres. The nuclei and centrospheres of the fused cells are quite distinct and even the cytoplasmic areas are partially separated by a tongue of yolk. It rarely happens that eggs are fused together in experiments with centrifugal force, and figure 106 represents one of these CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 308 rare cases, but it is very easy to cause the fusion of eggs by other means. For example in experiments where eggs are treated with carbonic acid there are many cases of such fusion. Some- times two eggs are fused as in figure 106, or three, four, or many eggs may be fused into one mass. It is an interesting fact that eggs rarely if ever fuse by their protoplasmic poles, but almost invariably by some portion of the lower hemipshere which contains yolk. There is apparently some peculiarity of the egg surface over the animal hemisphere which prevents its fusing with another egg at this pole. A large number of such eggs which were caused to fuse together at various stages before and during cleavage has been studied, and in every instance the polarity of each constituent egg remains practically unchanged. The ectomeres from different eggs unite in later stages of de- velopment into a continuous layer, but there is no indication that the polarity of one cell is changed by its fusion with an- other cell, as would be true if polarity were due merely to elec- tric charges on colloidal particles or to physical properties of the cell membrane. On the other hand these observations in- dicate that the polarity of an egg inheres in the organization of its more viscid protoplasm which is but slightly miscible with that of other eggs. 2. Surface tension as a cause of polarity. It is possible that the various constituents of cells are oriented and held in place, or brought back to normal positions if displaced, by surface tension. For example if spindles or centrospheres which are attached to the surface layer are centrifuged strongly, that layer may be indented as in figures 5 and 6 and the surface thereby increased; in such a case surface tension would restore the spherical shape of the cell after centrifuging, and if the asters or spheres still remained attached to the surface they would be drawn back to their normal positions. However, I have never seen the egg surface indented as shown in figures 5 and 6 except in the stage of the first maturation spindle and even at this stage it is unusual. In no other stage figured in this paper is there any such indentation of the cell and yet in every case all cell parts come back-if possible to their normal positions THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 2 354 EDWIN G. CONKLIN after centrifuging. External surface tension can not therefore be the cause of the normal positions of cell parts or of their return to these positions after centrifuging. Probably internal surface tension between the different constituents of a cell may play a more important part in localization of these constituents, but even this would not account for the persistence of polarity in centrifuged eggs unless there is some portion of the cell which remains unmoved during centrifuging. Since surface tension increases with decreasing temperature and vice versa an attempt was made to determine whether the return to their original positions of cell substances or parts which have been displaced by centrifuging is hastened by lowering the _ temperature. Eggs in various stages of development were cen- trifuged for 10 minutes and then placed for varying lengths of time on ice where the temperature was about 2°C., while one control was kept at room temperature (about 20°C.) ; in another experiment the temperature was raised to about 35°C. That the surface tension of these eggs is increased by lower tempera- tures is indicated by the fact that eggs are more nearly spheri- cal at lower temperatures than at higher ones, but the results showed conclusively that the return of cell parts to their normal positions took place more rapidly at about 20° than at 2°, thus indicating that this return is not due to internal or external sur- face tension. It is true that viscosity increases at lower tem- peratures as well as surface tension, nevertheless it does not pre- vent the eggs from assuming a spherical form; for these reasons, as well as for those mentioned above, it is evident that neither external nor internal surface tension is the principal cause of the normal location of cell parts nor of their return when once they have been displaced. However it is not denied that internal surface tension may be one of the contributory factors in the return of displaced substances, such as yolk, to their normal positions. 3. Spongioplasmic framework as the cause of polarity. There remains the explanation which has been maintained throughout this paper, namely, that the orientation and localization of cell CENTRIFUGAL FORCE ON EGGS OF CREPIDULA S50 parts is due to a framework of more viscid protoplasm and that the return of displaced parts to normal positions is due in the main to the elasticity or contractility of this framework. The evidence in favor of such a view may be summarized as follows :-— a. The substances of the egg of Crepidula are never com- pletely stratified by centrifugal force of from 600 to 2000 times gravity. Of all substances in the egg the yolk and cytoplasm are most completely stratified and yet the boundary between the two is never a plane, as it would be if the substances were free to move according to their specific weights, but the boundary between these substances is an irregular one with ‘lanes’ or projections of cytoplasm into the yolk. This indicates that while there is a relatively large amount of cytoplasm which is freely movable within the cell, there is a small amount of more viscid substance which penetrates every part of the cell and is especially abundant in nuclei, centrospheres and mitotic figures; this viscid material prevents the complete stratification of cell substances according to their specific weights. b. In many instances strands of this viscid substance may be seen running through various portions of eggs; such strands are seen most plainly in the mitotic spindles and astral radiations of dividing cells and also in the connections between nuclei and centrospheres and between the latter and the cell surface in dividing cells. Thése connections may be stretched or bent, but are rarely broken. The fact that when the animal pole of the egg is centrifugal in position the spindle may be stretched or distorted and the surrounding cytoplasm may be forced away, while yolk comes to be densely packed around the spindle, proves that the spindle is not merely the expression of lines of force, like iron filings in a magnetic field, but that it is a relatively persistent structure of a viscid or gelatinous character. The same is true also of resting nuclei and centrospheres and of the strands which connect these to the cell surface (see fig. 60). c. This viscid material is most abundant in spindles and astral radiations of dividing cells and in nuclei, centrospheres and the connections between these and the cell surface in rest- 356 EDWIN G. CONKLIN ing cells. It serves to hold the fully formed mitotic spindle in a definite position with respect to the cell surface; in early pro- phases and in resting stages these connections are relaxed so that spindles or nuclei may be moved out of their normal positions, but these connections are not easily destroyed and they always hold the nuclei and centrospheres in the same relative position to the cell surface, however much they may be stretched. 2. The structure of protoplasm The identification of a more viscid and a more fluid portion of the protoplasm in centrifuged eggs leads to a consideration of the relation of these two constituents of protoplasm to each other .and to various cell inclusions; it also raises the question of their relation to the polarity and orientation of development. It has long been evident that protoplasm is not a homogenous fluid. Dujardin held that ‘sarcode’ was a “substance gluti- neuse, parfaitment homogéne, elastique, contractile, diaphane. : On n’y distingue absolument aucune trace d’organi- zation, ni fibres, ni membranes, ni apparence de cellulosité.”’ (Quoted from Henneguy ‘La Cellule,’ p. 31.) Max Schultze, Haeckel, Kiihne and many other early observers regarded proto- plasm as a fluid owing to phenomena of protoplasmic flowing and of surface tension. Briicke (’61) first contested the possi- bility of ‘this on a priori grounds, holding that a homogeneous fluid would be unable to perform the functions which proto- plasm performs, and maintaining that it must have a ‘special structure’ or ‘organization’ made up of more liquid or more solid parts among which are thé ‘smallest living parts’ or vital units. All students of protoplasm now agree that it is composed of more fluid and more solid parts, though there is much differ- ence of opinion as to the form of each of these and their rela- tion to each other, as is shown by the various theories on the ‘structure of protoplasm.’ Different names have been given by authors to the more fluid and the more solid parts of protenes as inglicated in the following incomplete list: CENTRIFUGAL FORCE ON EGGS OF CREPIDULA . oot More fluid part More solid part Author Grind abatance) ' Heitzmann, Bele Bae lemina Reticulum Carnoy, 1883 uk VanBeneden, 1883 Alveolar Substance Interalveolar Substance Biitschli, 1873-1892 Paramitome Mitome Flemming, 1882 Hyaloplasm Spongioplasm Leydig, 1885 Trophoplasm Kinoplasm Strasburger, 1892 Ete. Ete. Some of the earlier students of the cell considered that only one of these substances was ‘living,’ though they differed as to whether it was the more fluid or the more solid part. Wilson (00, p. 30) concludes that ‘““we are probably justified in regarding the continuous substance (i.e. spongioplasm, inter-alveolar sub- stance, kinoplasm) as the most constant and active element and that which forms the fundamental basis of the system, transforming itself into granules, drops, fibrillae, or networks in accordance with varying physiological needs.’ With this opin- ion | entirely agree. In addition to these two substances protoplasm contains many other parts, some of which as Wilson suggests are formed prob- ably as differentiations of the spongioplasm, others perhaps as differentiations of the hyaloplasm. Among the substances which are embedded in the protoplasm, but are not a part of it, are the ‘inclusions’ such as oil, water, yolk, ete. Lillie (06, p. 156) says of the protoplasm of the egg of Chaetop- terus, ‘“‘The ground substance is a suitable name for the fluid that contains and suspends all the granules and droplets; if these were imagined removed it would preserve a faithful semblance of the egg. Thus it is regarded as forming the external pellicle and as continuous through the nuclear membrane with the nucleo- plasm.” Speaking of the vibrations of the microsomes and spherules in living protoplasm, Lillie says, ‘‘No one who has studied these movements, as I have done for hours at a time, could believe that the microsomes are nodal points of a network, or are connected by filaments as they appear to be in the best stained sections. One is forced to conclude that they have freedom of movement in all directions, i.e., that they are suspended in a 358 EDWIN G. CONKLIN fluid medium which has no filar, reticular or alveolar structure.” Lillie further concludes that ‘‘microsomes are the primitive formed elements of the cytoplasm” and that in point of origin they are chromatin particles. The appearance of a reticulum in fixed eggs he holds to be an artifact due to coagulation of a colloidal solution. It is unquestionably an extremely difficult task to determine with certainty the ‘ultimate structure’ of a substance which is so changeable in appearance as living protoplasm. There is, however, good reason for believing, as I have attempted to show, that it is composed of a more fluid and a more viscid portion and that while the former may be moved readily by centrifugal force the latter is not so readily moved; also that this more viscid part of the protoplasm holds nuclei and centro- spheres in a definite relation to the periphery of the cell and brings parts back to their normal positions when once they have been displaced; in short that the polarity and morphogenetic organization of the egg reside in this more viscid substance, and not in the more fluid medium as Lillie maintains. Whatever the ultimate structure of this denser portion of the protoplasm of the egg of Crepidula may be, it is certainly not a true fluid, nor is it the more fluid portion of the protoplasm. It does take the form of fibers or strands in the amphiaster of the living egg and these strands anchor the amphiaster to the periph- eral layer; in resting stages similar fibers are present between nucleus and centrosphere and between the latter and the periph- ery of the cell. I suspect that the microsomes of which Lillie speaks are not a part of this denser protoplasm, but that they lie in the more fluid substance, which would account for their free- dom of movement and would also explain the fact that the microsomes aggregate to so large an extent in the middle zone of centrifuged eggs. Mitochondria also, in this respect at least, behave as microsomes; in the eggs of mollusks and ascidians they move through the cell under the influence of centrifugal force almost as freely as do yolk spherules and oil drops. It is customary to explain the polarity or other differentiations of an egg as the result of its organization. But ‘organization’ CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 359 is a general and indefinite term which may include anything from metaphysical entelechies to hard and fast structures. Omitting from consideration all hypothetical causes which are beyond the reach of experimental investigation, we find that polarity, development, regulation or any other vital phenomenon may be regarded from the standpoint of static or of kinetic con- ditions, of morphological or of physiological causes. Ideally these distinctions are sharp and definite, but they are not so in reality. In a living organism static and kinetic, morphological and physiological conditions are really inseparable. However, for the sake of clear thinking, it is necessary to form some sort of a mental picture of what is meant by such a phrase as ‘the organization of the egg.’ On its morphological side the polar organization consists, as I have attempted to show, in a relatively persistent framework of viscid material which is also elastic and contractile so that it tends to resume its normal form when distorted; by this framework nuclei, centrospheres and mitotic figures are bound more or less firmly to the peripheral layer or ‘Hautschicht’ (Strasburger). There is no good evidence that this viscid material exists in the form of fibers which are definite in number and position. On the other hand its behavior during centrifuging would indicate that the appearance of fibers is due to inclusions which are forced into an otherwise continu- ous substance; this substance is more abundant and more uni- formly continuous at the animal pole than elsewhere in the cell. In normal eggs the presence of large yolk spherules at the vegetal pole gives to this substance a coarse sponge-like texture while the smaller spherules of yolk, oil and enchylemma toward the animal pole give to it a finer alveolar character. This appear- ance 1s very evident in good sections of normal eggs, as is shown in the figures of my paper on Karyokinesis and Cytokinesis (Conklin ’02), and that this structure is not an artifact, but is normal, is confirmed by the experimental studies of this and of a former paper (Conklin 712). When yolk is driven to the animal pole by centrifugal force the fluid portion of the cytoplasm and much of the viscid portion are driven away and that which remains is stretched or compressed 360 EDWIN G. CONKLIN into strands or fibers which are not to be regarded as preformed structures. It is clear that the localization of yolk at the vegetal pole in normal eggs and its return to that pole after it has been displaced is due to some differentiation of the spongioplasm at the two poles, for otherwise the yolk might be localized at any pole and would remain wherever it happened to be thrown. In normal eggs the spongioplasm is more abundant at the animal pole than elsewhere, but in centrifuged eggs it may be less abun- dant at this pole and yet normal conditions may be restored after centrifuging. It is therefore necessary to assume that the spongioplasm differs in some way, perhaps in elasticity or vis- cosity, at the two poles. If this material is more elastic or more contractile in the region of the animal pole than elsewhere in the cell the localization of cytoplasm and yolk in normal eggs and the return of dislocated substances to their normal positions would find an explanation. Furthermore the connections of centro- spheres and nuclei, the orientations of mitotic figures and the progressive localizations of cell substances indicate that the spongioplasm must differ in different regions of the egg and at different stages of development. , The question may well be raised whether the spongioplasm, or more viscid portion of the protoplasm, is not the real formative material, while the cytolymph as well as the oil and yolk are mere inclusions. I have already indicated that it is the more im- portant or indispensable part of the protoplasm, as is shown by experiments with centrifugal force and also by the desiccation of protoplasm in seeds and in-certain animals (rotifers, tardigrades, etc.); in both of these cases the fluid within the protoplasm may be largely eliminated without permanently destroying or injur- ing the protoplasm. It is well known that the fluidity or vis- cosity of protoplasm depends upon its water content and that this differs under different external conditions and in different stages of the cell cycle. Evidently the colloids in protoplasm may change from gels to sols and vice versa. It does not seem wise therefore to identify as protoplasm the gels only. ‘The most convincing work which has yet been done on the physical prop- erties of protoplasm in living cells is that of Kite, Chambers, CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 361 et al., by the method of microdissection. Kite (13) found by this method that the living cytoplasm of the egg of Asterias is an apparently homogeneous and a very viscous gel in which micro- somes and globules are suspended. This gel he found to be very elastic so that when portions of it were drawn out with a needle they would at once retract when released. On the other hand he found that the nuclear substance of this egg, with the exception of the nucleolus which is a quite rigid gel, is all in the sol state. In the male germ cells (probably spermatocytes) of insects he found that the cytoplasm and nucleus of the resting cells ‘‘are far too rigid to flow or change shape under such ex- perimental treatment. In the dividing cells the spindle fiber is an elastic concentrated thread of nuclear gel and its absorptive power and refractive index are also different from those of the dilute gel in which the spindle fiber is imbedded and from which it cannot be entirely freed. The homogeneous gel in which a telophase spindle is imbedded is so rigid that all the surrounding cytoplasm can be cut away and the spindle and chro- mosomes show no appreciable change; metaphase, anaphase and telophase spindles can be cut to pieces in Ringer’s fluid and the pieces are so rigid that they undergo no change in'shape.” Fi- nally he concludes ‘‘that cell division results primarily from con- comitant shrinking and swelling or changes in water holding power of different portions of the cell protoplasm. Many of the structural elements of the mitotic figure separate out of the protoplasm and change in rigidity according to their water con- tent. During the prophase the nuclear substance becomes so soft that movement of the components of the nucleus is affected by flowing of the nuclear gel. The mechanism at the basis of this flowing seems to be a change in the water holding power of the nuclear components.” Chambers (15) found that ‘‘the dissection of the germ cells of insects and of the frog reveals an extreme variability in con- sistency of their protoplasm, depending probably upon their water content.’’ He also found that ‘‘in many egg cells and free living unicellular organisms the surface layer of protoplasm may be decidedly more rigid than the interior.” The nucleus of a 362 EDWIN G. CONKLIN living cell consists, according to him, of a gelatinous substance “surprisingly more rigid than the cytoplasm in which it lies.” Mitochondria are not persistent structures but ‘‘they disappear and reappear and must be merely changes in the physical states of the colloids which compose the cytoplasm.” My own observations and experiments on resting and dividing cells, most of which were made before the publication of the work of Kite and Chambers, lead to essentially similar conclusions. The cytoplasm of the eggs of gasteropods and ascidians is com- posed of a viscid, elastic, contractile gel in which are included water, oil, yolk, pigment, microsomes, etc. This gel is more rigid at certain phases of the cell cycle than at others depending probably upon its water content. During the resting stage the nuclear contents are in a state of gel, but in the beginning of the prophase the achromatin becomes more fluid. It is quite evi- dent that the nucleus grows by absorbing substance from the cytoplasm. If the nuclear membrane is really a membrane, and there is much evidence that it is, such absorbed substance must enter as a fluid, though once within the nuclear membrane it. is converted into a gel. On the other hand when the nucleus reaches the prophase of mitosis much of its contents becomes more fluid and flows out toward the centrosomes where it again gels in the form of astral rays and spindle fibers (Conklin, ’02, "10,2 a,b). 3. Protoplasmic flowing and intracellular movements Another general phenomenon which is involved in these con- clusions is that regarding the nature of protoplasmic flowing and intracellular movements. In the eggs of Crepidula more or less extensive movements of the cell substance take place, as shown in the movements of the maturation spindles to the ani- mal pole, the migration of the sperm nucleus and aster through the egg, the segregation of cytoplasm at the animal pole and of yolk at the vegetal pole, and the movements of metakinesis and telokinesis during cleavage. If the protoplasm is a viscid, elas- tic, contractile gel how can such movements be explained? CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 363 In these movements we may distinguish the active movements of protoplasm which occur in the localization of cytoplasm, nuclei and mitotic figures, and the passive movements of inclu- sions such as oil, yolk and pigment. These passive movements may be considered first since they are simpler than the active ones. They are plainly of two kinds: 1) inclusions such as pig- ment, granules, water, etc. may be carried along with proto- plasm in its active movements, as in the case of the yellow pig- ment in the ‘mesoplasm’ of Cynthia (Conklin ’05); 2) by the concentration of protoplasm in certain regions inclusions may be forced out of those regions as in mitotic spindles and asters; similarly the concentration of protoplasm at the animal pole forces yolk to the vegetal pole. Such passive movements are due to active movements of the protoplasm and require no further explanation. The active movements of protoplasm are more difficult to ob- serve and explain. The denser portion of the protoplasm is highly elastic and contractile, as Kite has shown by direct ob- servation, and its capacity for movement is probably due to this property. Thus the flowing of peripheral protoplasm to the point of entrance of a spermatozoon and the formation there of an entrance cone may be regarded as due to the contraction or concentration of this protoplasm to the point stimulated. Prob- ably the collection of spongioplasm around the sperm centrosome or in the aster of any mitosis is likewise due to the contractility of this substance. The movement of the sperm nucleus and aster toward the animal pole and the segregation there of most of the spongioplasm may be explained in the same way. If the spongioplasm is highly contractile in all directions and concen- trates to a point of stimulation these and many other cell activi- ties find a ready explanation. The possibility of such concen- tration depends of course upon the fact that spongioplasm is not uniformly distributed throughout the cell but that it exists in- termingled with other substances and that in its concentration to one point these other substances are displaced. In former papers (Conklin ’02, 712) I have dealt at some length with the movements of metakinesis and telokinesis and since I have 364 EDWIN G. CONKLIN nothing new to add to those conclusions I need only say that the flowing movements which I there described may be interpreted as the result of the contractility of the spongioplasm,—which is indeed the original explanation of these movements (Van Bene- den ’87, Boveri ’87). 4. The orientations of development According to the view here expressed the localizations of spindles and cleavage planes, of nuclei and centrospheres, of ‘eytoplasm and yolk, and indeed the orientation of all develop- mental processes is associated with the structure and activities of the spongioplasm. In eggs generally cytoplasm becomes con- centrated at the animal pole during early stages of development and coincidently yolk is foreed away from that pole, probably by contraction of the spongioplasm to the animal pole; nuclei and centrospheres are bound together and are held in a definite relation to the animal pole by strands of spongioplasm; mitotic figures are oriented by means of the framework of spongioplasm and the planes of cleavage are thereby determined. In all cases the position and direction of the division planes is controlled by the position and direction of the spindle in the later stages of mitosis, the division plane always passing through the equator of the spindle and at right angles to its axis. In nor- mal eggs of Crepidula the first maturation spindle forms in the position previously occupied by the germinal vesicle—a little removed from the surface of the egg. This spindle reaches its maximum length in the metaphase at which time it is about as long as the radius of the egg. In the anaphase the peripheral pole of the spindle comes into close contact with the peripheral layer of protoplasm and at the same time the aster at this pole grows smaller and smaller and is at last completely absorbed into the peripheral layer while coincidently the spindle grows shorter so that when the division wall is formed through the equator of the spindle it cuts off a very small polar body from a relatively enormous egg. On the other hand where both asters are attached to the peripheral layer as in certain cleavages, the CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 365 spindle grows longer during the anaphase. If the maturation spindles are prevented from. shortening, giant polar bodies are formed since the division plane must pass through the equator of the spindle. If the first cleavage spindle is turned into the chief axis of the egg the cleavage plane is equatorial, instead of meridional as it should be, since the division plane must be at right angles to the spindle axis. However the initial position and directionof the spindle may differ from its definitive orientation. In several earlier papers on Crepidula (’97, ’98, ’99, ’02) I have shown that the spindle may form out of its definitive position and subsequently be moved into it by the activity of the cytoplasm. Lobes of cytoplasm may indicate where micromeres will form while the newly formed spindles are yet some distance away from these lobes; ultimately one end of a spindle moves into each of these lobes and then the cell division takes place through the equator of the spindle cutting off a micromere from a macromere. Without doubt the position and direction of a cleavage furrow is determined by the position and direction of the fully devel- oped spindle, but what determines the orientation of the latter? It is sometimes assumed that the orientation of a spindle is de- termined by yolk or other inclusions, for example that micro- meres are formed at the animal pole because yolk is segregated at the vegetal pole and this displaces cytoplasm, nuclei and spindles toward the animal pole,—but this is quite erroneous. In this paper as well as in a former one (712) I have shown that the pattern of cleavage is more or less independent of the amount of yolk, oil, water or other inclusions contained in a particular egg or blastomere, and Lillie (06) and Morgan (710) found this to be true in the eggs which they centrifuged. If the yolk which collects at the centrifugal pole or the oil which collects at the centripetal pole are thrown out of the egg completely the re- mainder of the egg which contains the nucleus and the material of the middle zone may segment like a normal egg, the first: and second cleavages being approximately equal and the subsequent ones unequal, thus giving rise to four macromeres and to three sets of micromeres which form from these. If the yolk is thrown 366 EDWIN G. CONKLIN into one of the first two blastomeres and the lighter substances together with most of the cytoplasm into the other one, each of these blastomeres continues to segment in a normal manner, the second cleavage being approximately equal and the subsequent ones giving rise to macromeres and micromeres as in normal eggs. The formation of macromeres and micromeres therefore does not depend upon the presence of yolk in the former but upon some other factor. But if the pattern of cleavage is not determined by the cell inclusions it is equally clear that it is not determined by any fixed and unalterable localization of the protoplasm with respect to the cell axes, for if the first cleavage plane is forced to take an equatorial position the second cleavage is meridional and equal and from each of the four cells thus formed micromeres are cut off on the animal pole side as in normal eggs. In short the macromeres which lie at the original animal pole are not the only ones which form micromeres, but even those which were cut off below the equator of the egg also form micromeres. This shows that the pattern of cleavage is not predetermined with reference to the original polarity of the egg. On the other hand eggs which have been subjected to pressure in the direc- tion of the chief axis of the egg at the time of the third cleavage may divide so as to form five, six, seven or eight macromeres and in such cases each of these macromeres gives rise later to three sets of micromeres as if it were a normal macromere. Consequently it cannot be said that the character of the cleav- age is determined by an inherited and wholly definite orienta- tion of each succeeding cleavage spindle, for if this were the case, when the third cleavage is rendered equal by pressure, subsequent cleavages should give rise only to second and third sets of micromeres; indeed when the third cleavage is forced to be an equal one the subsequent cleavages occur as if this were an entirely new cleavage which had been intercalated be- tween the typical second and third cleavages. In conclusion, then, the pattern of the cleavage is dependent upon the position and direction of the spindles and this is deter- mined, not by inclusions, but by the spongioplasm which holds — CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 367 nuclei, centrospheres and mitotic figures in a definite relation to the cell axes, and which is so elastic that when it is distorted by pressure or centrifugal force it tends to bring parts back to their normal positions. But such an explanation does not ex- plain the thing which we most wishto know, namely what de- termines the definite, orderly succession of orientations in de- velopment. For example, why do the two maturation spindles usually have the same orientation, while all succeeding divisions of the egg alternate in direction? Why are the first and second cleavages in Crepidula equal while subsequent ones are unequal? Why does every cleavage take place normally in a perfectly definite way, which differs from every other cleavage, and give rise to perfectly definite blastomeres which differ from all other blastomeres? If the orientations of development depend upon the spongioplasm does the structure of this spongioplasm change in a definite way from cleavage to cleavage? These are ques- tions which for the present must be left unanswered. V. SUMMARY AND INDEX 1. If the fertilized but unsegmented eggs of Crepidula plana are subjected to a centrifugal force of approximately 600 times gravity yolk is thrown to the centrifugal pole, oil and other light substances to the centripetal pole, while nucleus, centrosphbere and most of the cytoplasm occupy the middle zone between the other two. The relative volumes of these three zones is about 49: 1:14, or in other words the yolk occupies a little more than ¢ and the protoplasm a little less than } of the volume of the entire egg. This relatively large quantity of heavy yolk makes it possible to displace nuclei and cytoplasm in any direction and to study the effects of this on later development (p. 328). 2. While the greater portion of the cytoplasm may be displaced by the yolk a small residual portion of viscid spongioplasm is left between the yolk spheres and in a peripheral layer around the egg; this spongioplasm also forms a framework throughout the entire cell and connects nucleus and centrosphere of resting stages, or mitotic figure of dividing ones, to the peripheral layer. 368 EDWIN G. CONKLIN Because of this framework the stratification of ege substances in centrifuged eggs is never complete, but strands of spongioplasm prevent the free movement and stratification of substances according to their relative weights (pp. 329, 333). 3. The spongioplasm is highly elastic and contractile and when it is stretched or distorted it tends to come back to its normal form and to bring back to their normal positions displaced con- stituents of the cell (pp. 369, 373). 4, Mitotic figures, especially after the metaphase, are more firmly bound to the peripheral layer than are resting nuclei and centrospheres; the latter are always firmly united and the cen- trospheres are connected to the peripheral layer of the cell at the point nearest to the animal pole (pp. 333-335, 351-353). 5. As a result of these connections mitotic figures as a whole can be displaced only before the metaphase; after that stage they may be stretched or distorted but their astral radiations can rarely be separated from the peripheral layer (pp. 336, 337, 349). ; 6. Centrospheres and nuclei of resting stages may be displaced in any direction, but because of their connections with each other and with the periphery they always maintain a definite axial relation, the centrospheres lying between the nuclei and — that portion of the periphery which is nearest the animal pole. Nucleus, centrosphere, mitotic figure—each has a polarity of its - own, but all are held together in a definite relation to the cell body by the spongioplasm (pp. 352, 362, 363). 7. The persistence of the original polarity in centrifuged eggs in which most of the parts have been displaced and the return of those parts to their normal positions is due to these connections of spongioplasm, which are elastic and contractile (pp. 358, 374). 8. There is no good evidence that the polarity of a cell is a resultant of the electric charges carried on colloidal particles or on cell membranes. When eggs are caused to fuse together each component preserves its own polarity (pp. 366, 368). 9. Neither external nor internal surface tension phenomena are able to explain satisfactorily the persistence of cell polarity in centrifuged eggs (pp. 367, 368). CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 369 10. The spongioplasm of the egg of Crepidula is the interal- veolar or continuous substance within which are found enchy- lemma, microsomes, mitochondria, as well as yolk, oil and other inclusions. The form taken by this otherwise continuous substance depends largely upon its relation to these other in- cluded substances and it does not consist of preformed fibers or other structures which are definite in number and position. It is most abundant at the animal pole of normal eggs from which it radiates as a spongework between the other inclusions growing more and more coarse as it approaches the vegetal pole (pp. 369-376). 11. Protoplasmic flowing and intracellular movements are probably caused by the contractility of the spongioplasm. It contracts to points of stimulation, such as the entrance point of the spermatozoon, the centrosomes of mitotic figures, ete. Very small inclusions, such as pigment, may be carried along with the spongioplasm in its contraction; larger inclusions such as yolk spheres are forced out of the regions where spongioplasm concentrates (pp. 376-378). 12. The orientations of development such as polarity, sym- metry, localization of inclusions, pattern of cleavage, etc. are largely determined by the structure and activities of the spongioplasm, which probably differ in different parts of the egg and at different stages of development (pp. 378-381). 13. The division plane between daughter cells always passes through the equator of the mitotic spindle and at right angles to its axis. If both poles of the spindle are attached to the periphery of the cell it cannot be moved except by very violent centrifuging; if one pole is attached the other may be deflected to one side or the other; if neither pole is firmly attached the en- tire spindle may be moved. In normal maturation divisions one pole of the spindle is attached to the periphery at the ani- mal pole of the egg and during division this aster is absorbed into the peripheral layer, the spindle grows very short and when the division wall forms through the equator of the spindle it cuts off a minute polar body from a relatively enormous egg (p: 336). THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 370 EDWIN G. CONKLIN 14. If yolk is foreed to the animal pole after the maturation spindle has become attached to the periphery the spindle is stretched in length and when the division plane forms through its equator it cuts off a giant polar body or may divide the egg equally. Giant polar bodies do not develop because they do not receive a spermatozoon, and they do not receive a spermato- zoon because they are formed after the fertilization of the egg and after the entire cortical layer has been rendered imper- vious to the entrance of other spermatozoa(pp. 336-344). 15. If yolk is forced to the animal pole before the maturation spindle has become attached to the periphery the spindle may be driven to any point on the egg surface and if held there by continued centrifuging either one or both polar bodies may be formed there. Nevertheless nuclei and cytoplasm move back to the animal pole and yolk to the vegetal pole when centrifug- ing ceases and the polarity of the egg and embryo remains un- changed. Therefore the maturation pole does not determine the animal pole of the egg nor the ectodermal pole of the embryo (pp. 335, 336, 346, 347). 16. By centrifuging during cleavage all the yolk may be driven into one daughter cell and most of the cytoplasm into the other one, or by centrifuging early in mitosis the spindle may be carried out of its normal position so that the first or second cleavage may be equatorial instead of meridional, unequal in- stead of equal. Nevertheless the cells formed by the first two cleavages behave like normal macromeres in that each gives rise to three micromeres (ectomeres) on its animal pole side in the three succeeding cleavages (pp. 349-354). 17. If the first or second sets of micromeres are forced to form at a distance from the animal pole the succeeding set forms at the animal pole if the pressure is removed. If the unequal cleavages by which micromeres are formed normally are rendered equal by centrifuging the subdivisions of these large ‘‘micro- meres”’ are normal only so far as the cells are concerned which lie nearest the animal pole. This is due to the fact that under normal conditions the upper pole of a spindle is attached to the periphery at the animal pole side of the cell more firmly than is the lower pole of the spindle (pp. 354-356). Lar A CENTRIFUGAL FORCE ON EGGS OF CREPIDULA 371 18. The differentiation of daughter cells does not depend upon a differentiation of their centrosomes or nuclei, for the spindles may be turned about without changing the differentiation; nor does it depend upon the segregation of the movable parts of the cytoplasm or of the yolk in one cell or the other, for these segregations may be reversed without changing the differentia- tions; nor does it depend entirely upon the position and direction of the mitotic figure and the cleavage plane with reference to the ege axes, for these may be forcibly changed as in equatorial first or second cleavages without changing the normal course of differentiation in those cells after the force has ceased to act. These may be contributory factors in the differentiation of cells, but the principal factor is evidently to be found in the spongio- plasm which always tends to come back to its normal form if it is stretched or distorted, and which probably differs in strue- ture in different parts of the egg and in different stages of . development (pp. 357-360). BE EDWIN G. CONKLIN LITERATURE CITED Bovert, Tu. 1887 Zellenstudien 1. Jena. Zeitschrift, 21. 1891 Befruchtung. Ergebnisse der Anat. und Entw., 1. 1900 Zellenstudien 1V. Jena. 1910 Ueber die Teilung centrifugierter Eier von Ascaris megalo- cephala. Arch. Entw. Mech., 30. CuamBers, R. 1915 Microdissection studies on the germ cells. Science, 41. Cuitp, C. M. 1911-1914 Studies on the dynamics of morphogenesis and in- heritance in experimental reproduction, I-VIII. Jour. Exp. Zool., TQ), dhl, ae, We Gs alr 1915 Individuality in Organisms, Chicago. Conxuin, E.G. 1897 The embryology of Crepidula. Jour. Morph., 13. 1898 Cleavage and differentiation. Woods Hole Biological Lectures for 1896 and 1897. 1899 Protoplasmic movement as a factor of differentiation. Id., for 1898. 1902 Karyokinesis and cytokinesis, ete. Jour. Acad. Nat. Sei., Phileas 2: 1912 Experimental studies on nuclear and cell division, ete. Id., 15. 1910 Effects of centrifugal force upon the organization and develop- ment of the eggs of fresh water pulmonates. Jour. Exp. Zool., 9. 1912 Cell size and nuclear size. Id., 12. 1915 Why polar bodies do not develop. Proc. Nat. Acad. Sci., 1. 1916 Effects of centrifugal force on the polarity of the eggs of Crepi- loll. Illa, Y. DaunuGrReN, U. 1915 Structure and polarity of the electric motor nerve-cell in Torpedoes. Carnegie Inst. Wash. Publ. No. 212. Francorre, P. 1898 Recherches surla maturation, la fécondation et la segmen- tation chez les Polyelads. Arch. de Zool., 6. GurwitscH, A. 1904 Zerstérbarkeit und Restitutionsfaihigkeit des Protoplasmas des Amphibieneies. Verh. Anat. Ges. HEIDENHAIN, M. 1894 Neue Untersuchungen itieber die Centralkérper, ete. Arch. mik. Anat., 48. HerIDENHAIN UND Coun 1897 Ueber die Mikrocentren in den Geweben, etc. Morph. Arbeit., 7. Hennecuy, L. F. 1896 Lecons sur la Cellule. Paris. Herrick, F. H. 1895 The American lobster. Bull. U. S. Fish Commission. Hertwic, O. 1890 Vergleich der Ei- und Samenbildung bei Nematoden. Arch. mik. Anat., 36. 1899 Beitrige zur experimentellen Morphologie und Entwicklungs- geschichte. Id., 53. 1904 Weitere Versuche tiber den Einfluss der Zentrifugalkraft auf die Entwicklung tierischer Eier. Id., 63. Hoaur, Mary 1910 Ueber die Wirkung der Zentrifugalkraft auf die Eier von Ascaris megalocephala. Arch. Entw. Mech., 29. Kirt, G. L. 1913 Studies on the physical properties of protoplasm. Am. Jour. Physiol., 32. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA oie Konopacki, B. 1908 Die Gestaltungsvorginge der in verschiedenen Entwick- lungsstadien zentrifugierten Froschkeime. Bull. Acad. Sci., Cracovie. KostraNneckl, K. 1897 Ueber die Bedeutung der Polstrahlung, ete. Arch. mik. Anat., 49. Lerevre, G. 1907 Artificial parthenogenesis in Thalassema mellita. Jour. Exp. Zool., 4. Linum, F. R. 1906 Observations and experiments concerning the elementary phenomena of embryonic development in Chaetopterus. Id., 3. 1909 Polarity and bilaterality of the annelid egg. Experiments with centrifugal force. Biol. Bull., 16. 1915 Studies of fertilization I-VI. Jour. Exp. Zool., 12, 14, 16; Jour. Morph., 22; Biol. Bull., 28. Linu, R. 8. 1911 The physiology of cell division. Jour. Morph., 22. Logs, J. 1913 Artificial parthenogenesis and fertilization. Chicago. Lyon, E. P. 1907 Results of centrifugalizing eggs. Arch. Entw. Mech., 23. Mark, E. L. 1881 The maturation, fecundation and segmentation of Limax campestris. Bull. Mus. Comp. Zool., 6. Meves, Fr. 1911 Ueber die Betheilung der Plastochondrien an der Befruch- tung des Eies von Ascaris megalocephala. Arch. mik. Anat., 76. 1912 Verfolgung des sogenanten Mittlestiickes des Echinidensper- miums im befruchteten Ei, etc. Id., 80. Moraan, T. H. 1908 The effects of centrifuging the eggs of the molluse Cumingia, Science, 27. 1910 Cytological studies of centrifuged eggs. Jour. Exp. Zool., 9. Mora@an, T. H. AND Spooner, G. B. 1909 The polarity of the centrifuged egg. Arch. Entw. Mech., 28. PuatNer, G. 1886 Ueber die Befruchtung von Arion empirocorum Arch. mik. Anat., 27. Rasu, C. 1885 Ueber Zellteilung. Morph. Jahrb., 10. Roux, W. 1884 Beitrige zur Entwicklungsmechanik des Embryos. Ges. Abhandl., No. 19. Sosorra, J. 1895 Die Befruchtung und Furchung des Eies der Maus. Arch. mik. Anat., 40. Van BENEDEN, E. 1883 Recherches sur la maturation de l’oeuf, ete. Arch. de Biol., 4. Van BENEDEN ET Neyt 1887 Nouvelles recherches sur la fécondation et la division mitotique. Leipzig. Warasnp, S. 1893 On the nature of cell organization. Woods Hole Biol. Lectures. WerzeL, G. 1904 Zentrifugierversuche an unbefruchteten Eiern von Rana fusca. Arch. Mik. Anat., 63. Witson, E. B. 1900 The cell in development and inheritance. New York. ZIEGLER, H. KE. 1898. Experimentelle Studien tiber die Zelltheilung, I. Asch. Eutw, Mech. 6. DESCRIPTION OF FIGURES All figures represent entire eggs of Crepidula plana, fixed, stained, and mounted on slides in balsam. They were drawn by means of a camera lucida with Zeiss apochromatic oil immersion Obj. 3 mm., Ocular 4, at stage level and are there- fore magnified 333 diameters. In the process of reproduction they have been reduced one-third. The oil droplets, indicated by a coarse aveolar structure, mark the centrip- etal pole; the great mass of yolk, which is left unshaded, les in the centrif- ugal half of the egg; while the cytoplasm of the middle zone is shaded by stipples, the more granular part (spongioplasm) being stippled more densely than the less granular part (hyaloplasm) when these two are separated. The eggs drawn usually represent common types of abnormalities produced by cen- trifuging. Thousands of other kinds of abnormalities are produced, indeed no two are ever identically the same. PLATE 1 EXPLANATION OF FIGURES land 2 (1126, 1) Eggs taken while being laid and centrifuged in the ger- minal vesicle stage for 10 minutes; fixed immediately after centrifuging. The segregation of yolk and cytoplasm is fairly complete. The eggs are flattened in the axis of centrifuging and the nuclei are elongated in that axis; the chromo- somes and centrosomes are at the heavier end of the nucleus, the nucleolus is at the lighter end. The grouping of the chromosomes shows that the axis of the future spindle will be oblique to the long axis of the nucleus in figure 1, trans- verse to it in figure 2. 3 and 4 (1125, 1) Centrifuged 10 minutes (3000 revolutions per minute), fixed at once. The nucleolus is thrown with the fatty substance to the lighter pole of the egg; the sperm nucleus remains in the yolk in figure 3, in the ecyto- plasm in figure 4. The two poles of the spindle are equidistant from the cell surface and are not in contact with it. Figures 3, 4, 7, 8, are from the same slide, figures 9 to 12 are from the same experiment 1} hours after centrifuging. Eggs of this same lot 24 hours after centrifuging show both polar bodies in the middle of the ectodermal plate and everything absolutely normal. 5 and 6 (1038) Centrifuged 10 minutes, fixed 3 hours later. The cyto- plasm has been moved to the vegetative pole. The first maturation spindle has remained attached to the animal pole and its traction on the cell membrane is shown by the indentation of the latter. The sperm nucleus lies near the fatty substance of the lighter pole of the egg in figure 5, and in the yolk in figure 6. In figures 5 and 6 the spindle was attached to the cell membrane at the animal pole before centrifuging began. 374 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA EDWIN G. CONKLIN PLATE 1 375 PLATE 2 EXPLANATION OF FIGURES 7 and 8 (1125, 1) Centrifuged 10 minutes (3000 revolutions per minute); fixed at once. Anaphase of first maturation division; the spindle has been some- what elongated and in figure 8 the central end of the spindle has been thrown out into a lobe containing oil drops; the sperm nucleus lies in the yolk. 9 to 12 are all from the same slide. The eggs were centrifuged for 10 min- utes, in the prophase of the first maturation division and were fixed 14 hours later. The spindles were moved from the animal pole and during the hour and a half which elasped after centrifuging the spindles advanced to the anaphase, but made no progress toward the present protoplasmic pole of the egg. On the other hand there are evidences that they are progressing slowly toward the original animal pole which is now occupied by yolk. In subsequent stages eggs of this lot show both polar bodies at the ectodermal pole and almost all eggs have developed normally. This seems to indicate that in this lot of eggs the first maturation spindle moves back to the original animal pole and that the polar bodies are formed there. 9 (1125, 1) First maturation spindle near the middle of the egg, on the boundary between yolk and cytoplasm. Sperm nucleus in cytoplasm near surface. 10 (1125, 2) Late anaphase of the first maturation mitosis. The spindle lies in the middle of the egg with neither pole near the cell membrane; there is no trace of cell constriction. The spindle is. apparently moving from the cyto- plasm into the yolk. 11 (1125, 2) Late anaphase of first maturation mitosis; the spindle les in the middle of the egg, almost surrounded by yolk. 12 (1125, 2) Anaphase of the first maturation mitosis. The spindle les on the boundary between the yolk and the cytoplasm and is somewhat bent and distorted. CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 2 EDWIN G. CONKLIN pais B 377 PLATE 3 EXPLANATION OF FIGURES 13 (1125, 1) Centrifuged 10 minutes, fixed at once. The first polar body was formed, before centrifuging, at the original animal pole; this was the cen- trifugal pole and consequently the yolk was thrown to this pole while the cyto- plasm was forced to the vegetal pole; the half of the first maturation spindle left in the egg remains attached to the animal pole by the spindle fibers. 14 to 18 All from same slide (1125, 2); centrifuged 10 minutes, fixed 14 hours after. Centripetal pole marked by oil droplets. At time of centrifuging the first maturation spindle was attached to the surface of the egg at the animal pole and therefore was not moved; consequently the first polar body has formed at that pole. The cytoplasm was displaced more or less from the animal pole but is now returning, as is shown by figures 15 and 16 in which the centripetal pole (marked by the oil droplets) does not lie in the middle of the cytoplasmic field. and by figures 17 and 18 in which a narrow lane of cytoplasm is returning to the animal pole (cf. figure 13 in which the eggs of this experiment were fixed at once after centrifuging). CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 3 EDWIN G. CONKLIN -4° PB PLATE 4 EXPLANATION OF FIGURES 19 to 21 (1037) Centrifuged 10 minutes in the prophase of the second mat- uration division; fixed at once. The spindle was not firmly attached to the sur- face layer and was carried with the cytoplasm to the centripetal pole. 22 (1088) Centrifuged 100 turns of hand machine in about 1 minute; fixed 1 hour later. The second maturation spindle in the anaphase remains anchored to the cell membrane at the animal pole, but is stretched in length by the centri- fuging. The sperm nucleus lines near the animal pole. 23 to 27 (1038) Centrifuged 10 minutes; fixed 3 hours later. The cytoplasm has been displaced from the animal pole; the second maturation spindle remains at- tached to the surface at that pole, but is greatly stretched in length. In figures 23 to 25 the cell constriction shows a giant polar body in the process of forming; only the cell containing the & N will develop, in figure 23 the portion nearest the animal pole, in figure 24 that nearest the vegetal pole; in figures 26 and 27 a giant second polar body has been cut off. All nuclei and centrospheres within the yolk are much smaller than those within the cytoplasm. 380 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 4 EDWIN G. CONKLIN 381 PLATE 5 EXPLANATION OF FIGURES Figures 25-27 are explained on p. 394. 28 Normal egg showing the usual distribution of cytoplasm and yolk and the usual relative sizes of egg and sperm nuclei and spheres. 29 (1088) Centrifuged 100 turns of hand machine; fixed 1 hour later. Cy- toplasm lies at vegetal pole, and the sperm nucleus which lies in it is much larger than the egg nucleus. 30 (1125, 2) Centrifuged 10 minutes; fixed 13 hours later. The sperm nucleus is immense owing to its position in the cytoplasm, the egg nucleus is small owing to its position in the yolk. wo 1G) CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 6 EDWIN Ge CONKLIN : i 1" & 24PB 385 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22 NOW? PLATE 7 EXPLANATION OF FIGURES 37 to 40 (1146) The polar bodies are large and protoplasmic and contain oil droplets (first polar body has dropped off in figures 39 and 40), showing that they were extruded at the centripetal pole; after centrifuging ceased the cyto- plasm and nuclei moved back to the present protoplasmic pole. 41 (1145) The first polar body was formed normally before centrifuging began; the second was extruded during centrifuging and is much larger than normal and lies at a distance from the first. Cytoplasm and nuclei moved back to the true animal pole after centrifuging and the egg continued to develop normally. 42 to 54 (1136) Centrifuged 30 minutes; fixed from 6 to 24 hours later. In some cases the first polar body is normal and hes at or near the animal pole (figs. 42, 47, 54) showing that it was formed before centrifuging; in all cases the second polar body is much larger than normal and lies at some distance from the animal pole, showing that it was extruded during centrifuging; and in all cases cytoplasm and nuclei returned more or less completely to the animal pole after centrifuging ceased and the further development is normal in most cases. Chro- mosomes of a second polar body rarely give rise to a resting nucleus, showing that a centrosome is necessary for this process. 42 First polar body is normal and lies near animal pole; second is very large and contains cytoplasm and oil, showing it was formed at centripetal pole. 386 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 5 EDWIN G. CONKLIN eS APR 583 vw PLATE 6 EXPLANATION OF FIGURES 31 (1038) Centrifuged 10 minutes; fixed 3 hours later. Normal first polar body has divided. Giant second polar body forming, as indicated by con- striction opposite middle of second maturation spindle; latter connected to ani- mal pole. Sperm nucleus and sphere in cytoplasmic lobe on right; adjoining this the egg nucleus and sphere. 32 to 34 (1040) Centrifuged 10 minutes; fixed 5 hours later. Figure 32. Normal first polar body has divided; giant second polar body contains spherules and peculiar (telophase) nucleus; first cleavage spindle (metaphase) in egg. Figures 33 and 34. Telophase of first cleavage. First polar body normal; giant second polar body containing cytoplasm and yolk, chromosomes have not formed a resting nucleus; ce scattered chromaten near mid-body of second maturation spindle. 35 to 41 (1145, 1146) Centrifuged 4 hours (2000 revolutions per minute); fixed 6 hours later. All eggs are in the second cleavage stage and most of them continued to develop normally. Most of the eggs figured (figs. 33, 34, 35, 41) were centrifuged after the extrusion of the first polar body at the animal pole, but before the formation of the second polar body. The latter is a giant polar body and is usually extruded at some distance from the animal pole. In figures 36, 37, 38, and possibly figures 39 and 40 also, the eggs were centrifuged during the formation of both polar bodies, and consequently both are displaced from the animal pole. In these eggs the maturation pole was forced away from the ani- mal pole, but the cytoplasm and the nuclei went back to the latter when centrif- uging ceased. 35 (1145) The smaller (first) polar body evidently marks the animal pole, the second was formed some distance from the pole. 36 (1146) Polar bodies larger than normal, but they contain no yolk. The maturation spindles together with the cytoplasm were evidently forced away from the animal pole, and the polar bodies formed at this centripetal pole; after- wards the cytoplasm and nuclei moved back to their present position. Spindles present for the second cleavage. 384 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 7 EDWIN G. CONKLIN PLATE 8 EXPLANATION OF FIGURES 43 to 46 First polar body has dropped off. Enormous second polar body, containing a large amount of yolk, lying near animal pole in figures 43 and 45, at vegetal pole in figures 44 and 46; evidently formed as shown in figures 23 to 27. Only that portion of egg develops which contains a spermatozoon, and any portion whether at animal or vegetable pole which contains a spermatozoon may develop. 388 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 8 EDWIN G. CONKLIN 389 PLATE 9 EXPLANATION OF FIGURES 47 Normal first polar body at animal pole; giant second polar body a little to one side of the animal pole; it contains cytoplasm and oil and was therefore formed at the centripetal pole. 48 Original animal pole probably indicated by the small first polar body; the large second polar body was formed some distance from this. During the first, second and third cleavages the original polarity was not completely re- stored but the positions of the micromeres and of their nuclei and centrospheres indicate that there has been an attempt to restore the original polarity; in the upper, yolk-laden micromeres, cytoplasm, nuclei and centrospheres are turned away from the animal pole and toward those of the lower micromeres. There- fore the micromere pole is somewhat removed from the original animal pole. 49 to 54 The giant first and second polar bodies do not lie at the middle of the plate of micromeres (ectodermal pole), though the normal first polar body in figure 54 does; the ectodermal pole therefore coincides with the original animal pole. All the micromeres are quite normal in number, size and position. 390 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 9 EDWIN G. CONKLIN LZ em 391 PLATE 10 For explanation of figures 51-54 see page 404 392 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 10 EDWIN G. CONKLIN 393 PLATE 11 EXPLANATION OF FIGURES 55-58 Eggs in which the original polarity has been more or less modified. 55 (1147) Centrifuged 4 hours, fixed 18 hours later. This figure may be interpreted as follows:—The original animal pole was probably at the point AP (the first polar body has been lost). The cytoplasm and second polar spindle were carried away from this pole and a giant second polar body was formed near the equator of the egg. Cytoplasm and nuclei did not return to the ani- mal pole before the first cleavage (J) which was consequently equatorial, as in figure 46. The micromeres formed from A and B are normal, though removed from the original animal pole; those from C and D are abnormal though lying at that pole. 56 (1186, 5) Centrifuged 30 minutes, fixed 7 hours later. The first polar body marks the position of the original animal pole, at which point two micro- meres now center. The second polar body was formed at the vegetal pole as in- dicated by the presence on it of the yolk lobe (yl). At the first or second cleav- age the macromeres were partially separated and turned so that the poles at which the micromeres form do not coincide in each quadrant except in quad- rants A and B. Each macromere has formed one micromere which is norma! in size and appearance in quadrants A and C but which contains yolk in quad- rants B and D. 57 (1138) Centrifuged 30 minutes, fixed 12 hours later. At the upper side of the figure is a large lobulated second polar body containing two masses of chromatin and three centrospheres; it contains oil and cytoplasm and was ex- truded at the centripetal pole. The two lower macromeres are partially sepa- rated from the other two and contain but little cytoplasm and much yolk. Each macromere has given off two micromeres which are abnormal in contents and position in the lower quadrants but entirely normal in the upper ones. where the micromeres of the first set are dividing. 58 (1136, 4) Centrifuged 30 minutes, fixed 12 hours later. The first polar body had formed at the animal pole before centrifuging; a giant second polar body was then extruded at the vegetal pole. Afterward most of the cytoplasm returned to the animal pole, but a portion remained at the vegetal pole, probably cut off by an equatorial first cleavage as in figures 65 to 73. 594 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 11 EDWIN G. CONKLIN Ee ae odpR< ee =” \ eS PLATE 12 EXPLANATION OF FIGURES 59 to 82. Centrifuged after both maturation divisions but before the comple- tion of the first cleavage. 59 (1088) Centrifuged 100 turns of hand machine, fixed 1 hour later. Side view of an egg in which the axis of centrifuging was at right angles to the polar axis. Egg nucleus connected by strands of spongioplasm with the animal pole. 60 (1037) Centrifuged 10 minutes, fixed at once. Yolk was thrown to the animal pole, oil and cytoplasm to the vegetal pole. Both germ nuclei are stretched out in the direction of centrifuging and both have been pulled away from the animal pole. The egg nucleus is still connected with that pole by strands of granular cytoplasm (spongioplasm) and there is an aggregation of this immedi- ately under the polar bodies; other strands run through the zone of spongioplasm and of liquid yolk into the zone of yolk spherules. 61 (1037) Centrifuged 10 minutes, fixed at once. The germ nuclei and cytoplasm are displaced somewhat from the animal pole and the nuclei are distorted. 62 (1139, 1) Centrifuged 2) hours, fixed at once. Both germ nuclei to- gether with the cytoplasm have been forced away from the animal pole. 63 to 65 (1037) Centrifuged 10 minutes, fixed at once. The first cleavage spindle in the anaphase has been forced away from the animal pole, the spindle is more or less distorted and the centrosomes and spheres are turned toward the animal pole. The lobes (Z) containing oil at the centripetal pole indicate that the centrifugal force was strong. 396 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 12 EDWIN G. CONKLIN 397 PLATE 13 EXPLANATION OF FIGURES Figure 65 is described on p. 410. 66 (1040) Centrifuged 10 minutes, fixed 5 hours later. The first cleavage spindle has been turned into the polar axis and the cleavage furrow is appear- ing in an equatorial position. The animal pole is marked by the polar bodies, the vegetal pole by the yolk lobe (yl). Development delayed about three hours. 67 to 70, 72 (1035) Centrifuged 30 minutes, fixed 3 hours later. First cleav- age equatorial. After centrifuging, the cytoplasm, nuclei and centrospheres in the lower cells move as near as possible to the animal pole. One end of the cleavage spindle remains near the animal pole, to which it is probably attached. In figure 70 the centrosome which remained near the animal pole was separated from its nucleus and has divided twice; the chromosomes and a portion of the cytoplasm in the upper cell lie near the equator adjoining the cytoplasm and spindle in the lower cell; the polarity is modified to this extent. In figure 72 the second cleavage is beginning in a meridional axis and nearly at right angles to the first (equatorial) cleavage (1). 398 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 13 EDWIN G. CONKLIN 399 PLATE 14 EXPLANATION OF FIGURES For descriptions of figures 69, 70 and 72 see page 412. 71 (1146) Centrifuged 4 hours, fixed 6 hours later. The first cleavage (I) was nearly equatorial; the second cleavage spindle is at right angles to the first in the upper cell, and nearly parallel with the first in the lower cell. 400 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 14 EDWIN G. CONKLIN 401 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 2 PLATE 15 EXPLANATION OF FIGURES 73 (1089) Centrifuged 100 turns of hand machine, fixed 4 and 6 hours later. First cleavage (J) equatorial; second meridional in cell below equator, oblique and belated in cell above the equator. 74 (1139, 1) Centrifuged 2} hours at the close of the first cleavage, fixed at once. In the right half of the egg, cytoplasm and nuclear spindle were dis- placed toward the vegetal pole and were held in that position during the second cleavage; therefore the second cleavage furrow in the right half is nearly equa- torial in position and in both daughter cells the centrospheres lie on the animal pole side of the nucleus. In the left half of the egg division is nearly normal, though somewhat belated. 75 (1136,5) First polar body missing. The second polar body (with two nuclei) contains oil and cytoplasm and was therefore extruded at centripetal pole, which was near vegetal pole of egg. Cytoplasm and nuclei did not return to animal pole before the first cleavage (J-J) which was equatorial, consequently cyto- plasm and nuclei were cut off in cells below the equator. Positions of cytoplasm, nuclei and spheres show that the animal pole (AP) is at top of figure. The posi- tion of the cytoplasmic areas in the different cells, as well as that of the mid- body and the spindle remnants, prove that furrow running from the top to the bottom of the figure is the second cleavage but the spiral direction of this cleav- age is dexiotropic, as shown in the position of the cells and by the polar fur- rows, whereas in normal eggs this cleavage is laeotropic. 76, 77 (1035) Centrifuged 30 minutes, fixed 3 hours later. The first cleavage (1) was equatorial, the second (J7) meridional. Each macromere is giving off in a dexiotropic direction and as near as possible to the animal pole, a micromere of the first set. 78 (1151) Centrifuged 5 hours, fixed 5 hours later. The first cleavage was equatorial, the second meridional. Each macromere has formed a micromere of the first set and the two macromeres below the equator have also formed the second set of micromeres (2c, 2d); the macromeres above the equator are in proc- ess of giving off the second set of micromeres. 402 PLATE 15 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA EDWIN G. CONKLIN PLATE 16 EXPLANATION OF FIGURES 79 (1145) Centrifuged 4 hours, fixed 6 hours later. Cytoplasm and nuclei were displaced toward the vegetal pole and the first cleavage was nearly equa- torial, while the second cleavage was meridional. In the four macromeres thus formed the original polarity is preserved as far as possible and micromeres are being formed on the sides of the cells toward the polar bodies. 80 (1136, 5) Centrifuged 30 minutes, fixed 7 hours later. The first cleavage (I) was oblique as in figure 65; the second was meridional, being at right angles to the first in the lower cell and parallel with it in the upper one. The lower cells contain most of the cytoplasm and give rise to the micromeres of the first set which are reversed in position; the upper cells which contain little cytoplasm have not yet formed micromeres. 81 (1032A) Centrifuged 23 hours, fixed 21 hours later. Cytoplasm and nuclei were displaced toward the vegetal pole, the first cleavage was nearly equatorial, and the second nearly meridional. Each of these macromeres has formed a micromere of the first quartet around a point, about 90° from the polar bodies, which constitutes a new animal pole; the two lower macromeres are bud- ding off micromeres of the second quartet. 82 (1090) Centrifuged 100 turns of hand machine, fixed 16 hours later. The lower half of the egg is developing normally; the upper half is abnormal and is difficult to interpret. 404 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 16 EDWIN G. CONKLIN 405 PLATE 17 EXPLANATION OF FIGURES 83 (1040) Centrifuged 10 minutes in I-cell stage, fixed 53 hours later. De- velopment delayed; cytoplasm and cleavage spindle near vegetal pole. In the right hand cell a karyomere (km) which has not yet fused with remainder of the nucleus. The polar bodies le a little to one side of the cleavage furrow. 84 (1038) Centrifuged 10 minutes at close of first cleavage, fixed 3 hours later. Axis of centrifuging at right angles to the egg axis. Centrospheres have been moved less than the cytoplasm and nuclei, indicating that they are attached to the surface layer near the animal pole. 85 (1139, 1) Centrifuged 2} hours in 2-cell stage, fixed at once. The cyto- plasm and nuclei are displaced toward the vegetal pole; the centrospheres have been displaced but little from their normal positions and lie between the nuclei and the polar bodies. 86 to 89 (1037) Centrifuged 10 minutes in 2-cell stage, fixed at once. Cyto- plasm, nuclei, centrospheres (Cs) and spindle remnants have been displaced toward the vegetal pole in figures 86, 87, 89, toward outside in figure 88; the mid body (fig. 86, 7B) remains unmoved and the centrospheres (Cs) lie on the sides of the nuclei toward the polar bodies. The lane of cytoplasm leading from the centrospheres to the animal pole shows that the substances of the egg are not free to move according to their specific weights. 406 PLATE 17 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA EDWIN G. CONKLIN 407 PLATE 18 EXPLANATION OF FIGURES Figure 89 is described on p. 420. 90 (1071) Centrifuged 100 turns of hand machine at 2-cell stage, fixed at once. Centrospheres lie between nuclei and animal pole, with which they are connected by strands of cytoplasm. 91 (1078) Two-cell stage centrifuged 100 turns of hand machine, fixed at once. Axis of centrifuging oblique to egg axis. Centrospheres lie between nuclei and polar bodies. 92 (1148) Centrifuged 4 hours during first cleavage, fixed 18 hours later. Development has been arrested. Cytoplasm, nuclei and centrospheres were displaced toward the vegetal pole. The cleavage furrow has cut in from this pole, leaving one centrosphere in one cell and both nuclei and the other centro- sphere in the other cell. 93, 94 (1158) Centrifuged 5 minutes during first cleavage, fixed 3 hours later. Both first and second cleavage mitoses have taken place, but neither cleavage furrow has formed in figure 93, and only one of them in figure 94. The daughter nuclei show traces of division into gonomeres (male and female halves). 408 PLATE 18 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA EDWIN G. CONKLIN 409 PLATE 19 EXPLANATION OF FIGURES 95,96 (1150) Centrifuged 15 minutes in 4-cell stage, fixed at once. In figure 95 the axis of centrifuging was parallel with the first cleavage plane, in figure 96 at right angles to it. Cytoplasm, nuclei and centrospheres have been displaced from the animal pole by yolk in two of the other cells, but so far as possible the centrospheres remain between the nuclei and the polar bodies. 97,98 (1038) Centrifuged 100 turns of hand machine during third cleavage, fixed 15 minutes later. The axis of centrifuging was parallel with the first cleav- age plane. The spindles are normal though they have been somewhat displaced from their typical positions. 99 (1145) Centrifuged 4 hours in 4-cell stage, fixed at once. Cytoplasm, nuclei and centrosomes were displaced to the vegetal pole, and the third cleavage spindles which were present in the cells are so placed that large yolk-rich micro- meres will be cut off at the animal pole. 100 Centrifuged 4+ hour, fixed at once. Centrifuging took place during the third cleavage so that two of the micromeres (1c, 1d) were formed at some distance from the animal pole. Lobe (ZL), containing oil, marks the centripetal pole. 410 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 19 EDWIN G. CONKLIN 411 PLATE 20 EXPLANATION OF FIGURES 101 (1038) Centrifuged 10 minutes during third cleavage, fixed 3 hours later. Large yolk-laden micromeres were formed at the animal pole. 102 (1035) Centrifuged + hour during third cleavage, fixed 3 hours later. Egg essentially like the preceding. The micromeres of the first quartet, two of which are dividing, are large and full of yolk, the micromeres of the second quartet contain no yolk and are of normal size and constitution. 103 (1050) Centrifuged 15 minutes, fixed at once. Centrifuged after the formation of the first quartet of micromeres which are not displaced from the animal pole, though the nuclei and centrospheres are displaced in these cells in the same direction as in the macromeres. 104 (1136, 4) Centrifuged 30 minutes during the second cleavage, fixed 12 hours later. The left half of the egg has developed normally, giving rise to two macromeres, and each of these to three micromeres, two of which have subdi- vided; the right half has divided into a large protoplasmic cell at the animal pole and beneath this a yolk-rich cell which contains two large nuclei connected by a chromatic thread, and four or more centrosomes and spheres, but the cell has not divided. 105 (1139, 2) Centrifuged 2} hours during the maturation division (?), fixed 18 hours later. Both polar bodies, one of them large and containing a spindle, lie at the pole opposite the protoplasmic cells, which is probably the vegetal pole. The probable identity of the cell is indicated by the labelling and the arrows, but the entire egg is abnormal and hard to interpret. 106 (1129, 1) Centrifuged 10 minutes in gum arabic, fixed 4 hours later. Two eggs fused in the plane of the broken line, each in the 8-cell stage. The polarity of each egg is unchanged by the fusion, asis shown by the positions of the cells relative to the polar bodies. 412 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 20 EDWIN G. CONKLIN 101 102 413 PLATE 21 EXPLANATION OF FIGURES 107, 108 (1090) Centrifuged 100 turns of hand machine before or during second cleavage, fixed 18 hours later. The second cleavage was suppressed in one blastomere (the lower) of figure 107 and in both of figure 108, and conse- quently the micromeres are more or less abnormal in number and position. 109 (1038) Centrifuged 100 turns of hand machine in 2-cell stage, fixed 4$ hours later. The second cleavage was very unequal, with the result that two macromeres are large and two small. There are 18 micromeres, some of which are difficult to identify. 110 (1032, A) Centrifuged 2} hours during second cleavage, fixed 21 hours later. Egg with four macromeres abnormally joined, and with two separate areas of micromeres, which can not be individually identified with certainty. 111 (1090) Centrifuged 100 turns of hand machine during second cleavage, fixed 16 hours later. Two macromeres are large and two small. There are 24 micromeres, some of which are hard to identify. 112 (1134) Centrifuged 30 minutes in .2-cell stage, fixed 21 hours later. Macromeres arranged in a linear series, micromeres in two groups at opposite ends of the egg. 414 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA EDWIN G. CONKLIN PLATE 21 109 nLe 415 PLATE 22 EXPLANATION OF FIGURES 113 to 118 (1149) Centrifuged 5 hours in the 4-8-cell stage, fixed at once. Figures 113 to 116 are viewed from the vegetal pole, figure 117 from the animal pole. In all these cases the animal pole was centrifugal in position, the vegetal centripetal; consequently the cytoplasm and nuclei of the macromeres was car- ried to the vegetal pol>, where a numbe: of micromeres have been formed, while the first set of micromeres (ectomeres) remains at the animal pole (except in figs. 113, 115, where it had not yet formed). The micromeres at the vegetal pole correspond in number, contents and subdivisions to the second and third sets of micromeres, but it is not certain that they give rise to ectoderm. Figures 117, 118 represent the same egg under a high and low focus; the stippled cells in figure 117 are the first set of micromeres at the animal pole; in figure 118, the second and third sets at the vegetal pole. 416 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 22 EDWIN G. CONKLIN THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 PLATE 23 EXPLANATION OF FIGURES 119 to 122. (1151 and 1152) Centrifuged 5 hours in the 8-cell stage, fixed 5 hours later. In these eggs the animal pole was centrifugal in position, conse- quently the cytoplasm and the nuclei of the macromeres, A, B, C, D, were forced away from the animal pole by the yolk and were held in this new position dur- ing one or more cleavages. As a result the second set of micromeres (2a—2d) are small protoplasmic cells at the vegetal pole (figs. 199, 121, 122) or they are large yolk-rich cells at the upper pole (fig. 120). 123 (1152) Centrifuged 5 hours in the 2-cell stage, fixed 5 hours later. The second cleavage was forced to form in the equator of the egg, and the first set of micromeres (1a—I/d) were cut off on one side of the animal pole and are larger than normal. All the macromeres are dividing to form the second set of micro- meres (2a—2d) which will le on the animal pole side of those cells. Two cells of the first set (1a-1d) are dividing laeotropically as in normal eggs. 124 (1154) Centrifuged 5 hours in the 8-cell stage in the direction shown by the arrow; fixed 24 hours later. The cytoplasm and the nuclei of the macro- meres were thrown to the left and toward the vegetal pole and small cells were cut off in this position. 418 CENTRIFUGAL FORCE ON EGGS OF CREPIDULA PLATE 23 EDWIN G. CONKLIN 419 THE GROWTH OF PARAMECIUM IN PURE CULTURES OF BACTERIA GEO. T. HARGITT AND WALTER W. FRAY From the Zoological Laboratory of Syracuse University INTRODUCTION Paramecium has long been a favorite form for studies of various kinds, and with justice since it is of considerable size, is composed of a single cell, is easily obtained, and is easily maintained in vigorous condition in cultures in the laboratory. Some of the factors which are involved in the cultivation of Paramecium have been carefully studied with the result that we have rather complete and precise data on their significance. This analysis includes not only some of the biological factors involved in the activities and functions of the infusoria, but some of the chemical and physical factors of hay infusions and culture fluids. One of the biological factors of importance is the food of Paramecium. It has long been known that bacteria furnish the chief food supply, and in some of the most careful work attempts have been made to secure a uniformity of food, with some measure of success. It is, however, rather striking that not a single effort has been made by modern methods to analyze the hay infusion bacteriologically. This failure is the more striking in view of the known fact that lack of vigor in cultures and certain ‘depression’ states, are doubtless due to something unfavorable in the food supply. Also th® conclusions of much of the modern work on Paramecium may be significant to the degree that the factor of food is understood, and under control. Recognizing this gap in our knowledge of one of the most essential factors touching the life of Paramecium we have, dur- ing the past year, made a start on a bacteriological analysis of 421 422 GEO. T. HARGITT AND WALTER W. FRAY the hay infusion. More especially we have collected some data on the growth of Paramecium in pure cultures of known bac- teria. The work has been most rigidly controlled by checks and tests at all times, which enables us to say that in our experi- ments the food has been positively known and no foreign bac- teria have gained entrance to the experimental cultures. The data covering the growth of Paramecium are not so extensive as might be desired but certain conclusions seem to be warranted. The technic of the cultures also appears to be of sufficient value to merit attention. Since circumstances have compelled a termination of these experiments for an indefinite time it seems well to set forth the methods employed and the results obtained. Our thanks are due to Prof. Henry N. Jones of the Depart- ment of Bacteriology for the facilities in material and apparatus placed at our disposal in working out the bacteriological part of the investigation. HISTORICAL REVIEW The classic investigations of Maupas (’88) were the most ex- tensive and carefully worked out of the earlier attempts to understand the reproductive activities of infusoria. In his paper an account is given of the methods employed, and while these have been considerably modified by later workers they still are of some importance. Among nore recent workers Calkins (’02 a) was the first to undertake a careful study of the growth of Paramecium under known conditions and with controlled factors. This paper gives a detailed explanation of the method of mak- ing hay infusions, methods of cultivation of the animals on depression slides, and the like. Calkins’ chief modification of the earlier methods of study lay especially in growing Para- mecium in depression slides with small amounts of liquid, and their isolation each day to prevent conjugation. As a criterion of the favorableness of the culture media, and as an indication of the rate of growth and metabolism Calkins (02 b) used the rate of fission. ‘‘The division-rate is taken as the measure of vitality, for it represents the rate of metabolism, PARAMECIUM IN PURE CULTURES OF BACTERIA 493 erowth, and reproduction. A better index of the general vitality could not be found and while fairly constant from day to day, its fluctuations mark out clearly the periods of vigor and depres- sion.” The use of the division rate as an index has been ac- cepted by all who have worked on the infusoria. In the culti- ration of these animals Calkins used a hay infusion, always made in the same way, and from the same kinds of material. This infusion was raised to the boiling point and cooled before using. The composition of the medium was, therefore, fairly constant, indeed results of later work suggest the possibility of its being too constant. Temperature and other physical fac- tors were either controlled or known, but the bacterial content of the media was entirely unknown. Calkins says: “The bac- teria in the hay-infusion constitute the normal food of the Para- moecide. Of these, Bacillus subtilis, is, probably, the only one left alive after the infusion is raised to the boiling point, and this organism, therefore, forms the staple article of diet for Paramecium in culture.” It is not certain that Calkins was entirely correct in his inference for, while B. subtilis is resistant to water at the boiling point there are other bacteria which are nearly as resistant and may be unaffected by that temperature. Furthermore there are numerous bacteria which are morpho- logically like B. subtilis so the exact form could not be known without the use of bacteriological technic. Even if the infer- ence made were correct it would have no significance since air infection of cultures on slides, or in flasks, might radically change the bacterial flora of the media. Hence Calkins was working with an unknown variable so far as food was concerned. Woodruff (’05) used essentially the same methods of cultivation for various infusoria, and was no more certain of the bacterial food of these forms. Peters (07 a and b) by careful chemical analyses of infusions added considerably to our knowledge of the composition of such fluids. Concerning the bacteria he said: ‘“‘I have never deter- mined the specific kinds of bacteria, but I have observed suf- ficiently to see that practically the same forms are characteristic of the same cycles of the cultures.’’ Doubtless this means he 424 GEO. T. HARGITT AND WALTER W. FRAY could tell whether bacilli or cocci were present. His chemical analysis showed some correlation between the acidity of the infusion and the abundance of bacteria. He suggested a prob- able sequence of bacteria (correlated, perhaps, with certain sequences of chemical change) which influenced or governed the protozoan sequence. From this one might infer a certain kind of bacterial food was essential for particular infusoria. There was, however, nothing in the investigation of Peters which would establish that as a fact. Jennings (08) was the first to fully appreciate the importance of carefully considering the bacterial flora of his cultures, though he made no attempt at identification. His method was prob- ably fairly efficient in introducing similar bacteria into all his cultures, though without the use of bacteriological technic this could not be known. Also no attempt was made to prevent infection from the air. He recognized that certain ‘bad con- ditions’ of cultures were probably due to injurious bacteria and under these conditions the protozoa decreased in vigor, or even died. If one attempts to account for this lack of vigor or death of infusoria in cultures he must consider several possible causes. It is probable that death may sometimes be due to toxic excre- tions of the bacteria; there may not be sufficient bacteria to furnish food for the protozoa; or the bacteria present, though sufficient in number, may not be suitable for food so the pro- tozoa die in the midst of apparent plenty. All of these pcussi- bilities must be considered in dealing with cultures. In making the cultural conditions identical in order to com- pare the division-rates, Jennings says: ‘‘To make the conditions of existence the same it zs not sufficient to altend merely to the basic fluid; the bacteria must also be the same.’ This, he insists, is not a mere theoretical consideration but a practical necessity. The introduction of certain organisms may lead to the death of the protozoa in the culture. In order to secure this identity of bacteria in all cultures Jennings washed all his paramecia in two lots of fresh hay infusion, all being washed in the same fluid, for the second washing. This he believed would bring into the second wash fluid all the different kinds of bacteria which were PARAMECIUM IN PURE CULTURES OF BACTERIA 425 present in the original cultures from which the paramecia were taken. Finally he added a definite amount of this second wash fluid to each culture in which he was growing Paramecium. By this method he hoped to have the same bacteria in all his cul- tures. Such a method is more cumbersome than to use exact bacteriological methods, and has other serious drawbacks. Jen- nings could not have any precise knowledge as to whether he had accomplished his aim of securing uniformity of nutritional conditions; he could only assume that he had. A more serious disadvantage is the impossibility of later duplicating the experi- ment exactly, and this means no other culture could fairly be compared with this one on account of ignorance concerning the bacteria present. Woodruff in 1909 introduced some modifications into the culti- vation of Paramecium on depression slides by making his media from different materials, hay, grass, pond weeds, material from swamps, ditches and the like. He believed in this way he could make conditions more nearly normal, than by the constant use of a similar medium. Calkins’ paramecia died out when kept on a constant medium, and Woodruff believed this was due to the uniformity of the medium. Both at this time and later (711 a) in analyzing the effect of excretion products on reproduction of Paramecium, Woodruff depended on the bacteria carried with the animals to inoculate his culture media. He assumed the bacterial content of the cultures was the same, hence believed the differences observed in the division-rates were due to the different volumes of culture fluid, i.e., to the amount of excre- tory products of Paramecium itself. But since it is probable that the bacterial flora was not uniform in all the cultures there is a possibility that the excretory products of the bacteria may have been the occasion for the differences in the rates of fission. Woodruff and Baitsell (11 ¢) found a 0.025 per cent solution of beef extract to be a better culture fluid than hay infusion, since protozoa did as well in a constant medium of this solution as in a varied hay infusion. There was, as they admitted, a variation in the bacterial content in their different slide cul- tures, but they believed this variation was so slight as to be 426 GEO. T. HARGITT AND WALTER W. FRAY negligible as a modifying factor in their experiment. In a further attempt (711 d) to so control the experimental factors as to deter- mine whether rhythms in reproduction were due to environment they still depended upon the chance inoculation of their media by the transfer of the bacteria with the Paramecium at the time of its isolation. Their belief that the variation in the bacterial content was slight may have been true as to numbers of bacteria, but it is probable that the kinds of bacteria may have been, or come to be, decidedly different. In one culture, for example, one kind of bacteria might have got the start and become the predominant kind, while the predominant type in another cul- ture would be different, and such difference might be significant. Whether it was ‘‘unnecessary to attempt to ‘sterilize’ the para- moecia and feed them on pure cultures of bacteria’’ they cer- tainly could not determine without a trial. Other work done since this time on Paramecium has not added anything so far as the nutritional factors are concerned. There- fore, from this brief summary it is clear that not a single worker has had any precise knowledge of the bacterial food of Para- mecium in experimental work. Furthermore all have entirely neglected the precaution insisted on by Jennings of making the ‘bacteria the same,’ save Jennings, himself. While it is not to be expected that each worker would laboriously perform his experiments under strict bacteriological methods, it is compara- tively simple and easy to use a method like that of Jennings to render the bacterial content of media as nearly the same as possible. The data obtained by us will show that it is not necessary to carry on all of this work with a strict bacterio- logical technic. But it will be as clearly shown that there is the necessity of securing a uniformity of bacteria in all cultures which are to be compared, and also the desirability of excluding cer- tain kinds of bacteria from the cultures. METHODS AND OBSERVATIONS Bacteriological analysis. Two main points are involved in a bacteriological study of the hay infusion: 1) what bacteria are normally present in a healthy hay infusion; 2) what forms PARAMECIUM IN PURE CULTURES OF BACTERIA 427 gain the supremacy in fermenting, putrefying, or other abnormal infusions. Under abnormal infusions are included all which for any reason are unfavorable for continued growth or existence of Paramecium and other infusoria. These unfavorable conditions are probably due to bacteria directly or indirectly and the type of bacteria producing such conditions should be known. One of the first things to determine in a study of healthy infu- sions is the source of the bacterial infection and obviously there are three possibilities. A given organism may have been present on the hay and by this means gained entrance to the infusion. It may have been present in the water used for starting the infusion, or it may have settled in from the air. ‘To secure the forms present on hay, in the water or air, cultures were made as follows: The standard hay, infusion used by Jennings and Hargitt (10) was made by allowing 10 grams of chopped hay to macerate in a liter of tap-water. This was then sterilized in an autoclave to kill all the bacteria. By pulling out the cotton plug from the flask, after the fluid was cool, inoculation from the air was made possible. A second flask was prepared in the same way except that 100 cc. less of water was used. After having been sterilized and cooled 100 ce. of tap water was added to furnish the source of inoculation. The addition of this amount of water made the solution standard. The third culture was made by leaving out a little of the hay till after the sterilization of the liquid. When the remaining hay was added the inoculation from hay was accomplished and this medium was brought to the standard. The three flasks contained the same amounts of water and of hay, the water and hay being identical; they dif- fered, therefore, in the source of bacterial inoculation. Since the flasks were plugged with cotton, after having been made and inoculated, any subsequent differences observed would clearly be due to the source of inoculation. These cultures were analyzed at the end of a few days, at the end of a few weeks, and again after four months had elapsed. In the analysis the ordinary technic of making plates was fol- lowed, the infusion being diluted with isotonic salt solution to prevent crowding of the colonies of bacteria on the agar plates. 4928 GEO. T. HARGITT AND WALTER W. FRAY After a few days the cultures inoculated from the air and from hay showed the presence of many chromogenic bacteria. Most such were yellow but some pink and some red forms were pres- ent. The culture inoculated from water showed only a few of these chromogenic bacteria, but they contained forms which imparted a green color to the agar of the plates. This form (Bacillus fluorescens) was also recovered from the hay-infected culture. But very few colonies of a lobose or running type were present on the plates made at this first analysis, a few stellate colonies coming from the hay inoculation. At the end of two weeks the chromogenic colonies were disappearing and white amoeboid colonies were taking their place, Bacillus fluorescens still being abundant. When the analysis was made four months later the chromogenic forms had almost entirely disappeared and many of the colonies were of an amoeboid or arborescent type, this being especially true of infusions inoculated through air and hay. These analyses demonstrate a distinct succession of bacterial life in a hay infusion that is left undisturbed for a long time. It is quite probable that certain of these bacteria, especially those which appear later in the sequence, are not favorable for the continued vigor and growth of Paramecium. Such forms are seen to gain the ascendency in undisturbed cultures, and this ascendency is prevented, more or less, by the frequent addition of fresh hay and water. In selecting the colonies on the plates for the production of pure cultures, it was obviously out of the question to test all and still keep the investigation within reasonable bounds. But such a program is unnecessary, since there are certain types of bacteria which are predominant and it is these which have con- trol over the future of the infusion and which must therefore, furnish the greater part of the food for the protozoa. To attempt to get the minor organisms present would have proved mis- leading in interpreting later results. Particular attention was therefore paid to securing all the predominant types of bacteria in pure cultures. PARAMECIUM IN PURE CULTURES OF BACTERIA 429 The bacteria taken from the plates were secured in pure cul- ture and maintained on agar slants in test tubes. From these slants their morphological and cultural characteristics were de- termined. In every case the examination was carried far enough to secure a good idea of the general character of the organism. Their shape (whether rod, coccoid, or spiral), size and motility were determined in every case. Tests were made for spores by subjecting them to a temperature of 76°C. for twenty minutes. Their reaction to stain was tested with methylen blue. carbo- fuchsin, and Gram’s stain. Their general cultural characteristics were determined as follows; agar was used to determine the kind and amount of growth; beef broth for an examination of motility and the test for indol; litmus milk was used to show whether the organism produced acid, and whether it produced a coagula- tion and saponification; in dextrose fermentation tubes was ob- served the power of fermentation and gas production; gelatin stab cultures were made to determine whether the organisms could liquefy the medium, and to show the general character of growth. In some cases other media were used, such as lactose broth, hay infusion agar, potato. Having secured a good description of each culture they were identified, so far as possible, by using Chester’s ‘“‘Manual of Determinative Bacteriology.”’ In some cases the identification was exceedingly difficult, since the description of these sapro- phytic bacteria is incomplete. One or two of the bacteria iso- lated and used for experimentation could not be identified by name. In the analysis of abnormal hay infusions several protozoan cultures, formerly rich in Paramecium, were found in which fer- mentation and putrefaction were in progress and in which Para- mecium and other infusoria were either absent or few in number and not in vigorous condition. Materials from six such cultures were plated and pure cultures of the bacteria obtained in the same way as for the normal infusions, and their morphological and cultural characteristics determined. A very large number 430 GEO. T. HARGITT AND WALTER W. FRAY of different kinds of bacteria were present and the selection of the forms for further study was largely an arbitrary one. Those bacteria were isolated which were present in the largest numbers in the hope that they were the ones causing the abnormal condi- tions. This is a fairly logical method of choice since only those which are very abundant can be active in producing changes in the infusion. In each of the abnormal hay infusions examined there was a distinctly sour or putrefactive smell at the time of the examination. This odor was used later as a test in deter- mining what bacteria were responsible for the production of the abnormal conditions. Flasks of sterile hay infusion were inocu- lated with different pure cultures of the bacteria, or in some cases inoculated with bacteria of a certain type but not neces- sarily of one kind. At the end of a few days, or only after several weeks in some cases, some of the flasks so inoculated gave odors similar to those of the original infusions. The bacteria present in such flasks were later used in pure cultures for feeding Para- mecium to further test their unfavorableness as food. Not all of the bacteria which were isolated were completely studied and identified. But all which were used for experi- mentation with Paramecium were carefully studied and a com- plete description of their characteristics will be found in a table at the end of the paper. It may be said in general that the bacteria isolated from abnormal infusions are more zymogenic than are those from normal infusions. Preliminary experiments. The paramecia which were used for nutrition experiments with the pure cultures of bacteria were obtained in pure strains. The necessity for having all the pro- tozoa of one and the same line is so obvious as to need little comment. In order that results may be comparable the para- mecia must be alike so far as possible and those of the same pure strain are as nearly identical as can be obtained. Hence all the animals used in our experiments are descendants of a single individual of Paramecium aurelia or of Paramecium cau- datum. In the beginning both P. aurelia and P. caudatum were used but owing to the lower rate of division of the latter, and to its less marked adaptability to growth in depression slides, PARAMECIUM IN PURE CULTURES OF BACTERIA 431 the work was later limited to P. aurelia. The results of both are given but those of P. aurelia cover a longer period of time. For a discussion of the differences of the two species the reader is referred to Jennings and Hargitt (10), and Woodruff (11 b). In the starting of the pure strains of the Paramecium the single individuals were isolated in the usual way with capillary pipettes and grown in depression slides, or in dishes to produce large cultures. Before starting the experiments on growth in pure cultures of bacteria the protozoa were grown in depression slides for a time in the usual manner, depending upon chance inoculation of the media to furnish the bacterial food. The purpose of this was to determine the normal rate of fission, to acclimatize the animals to growth in a limited amount of medium, if such was necessary, and to determine what was the best medium for this growth. The slides were used without cover glasses and were kept in moist chambers to prevent evaporation of the fluid, the usual practice for work of this kind. In the initial experiment the medium used was hay infusion liquid taken from a stock culture and filtered to remove the protozoa. This medium was placed in the depression slides and examined with a binocular microscope to be certain that no strange paramecia, and no protozoa of any kind, were present. The rates of fission for two weeks averaged approximately one per day for P. caudatum and nearly one and a half for P. aurelia. As might have been expected this filtered infusion was found not to be entirely satisfactory. It varied in composition and in age and was sometimes too rich, since the bacteria increased so ‘apidly as to hamper the activities of Paramecium. The stand- ard 1 per cent hay infusion was also unfavorable if used full strength, and it was necessary to dilute it somewhat. It was obvious that if hay infusion was to be used as a culture medium it must be so standardized in its preparation that its composition did not vary, and also it must be of a concentration suitable for use in the depression slides. To determine what concentration was best the standard infusion was diluted 20 times, 10 times, and 5 times with water making standard infusions of 0.05 per 432 GEO. T. HARGITT AND WALTER W. FRAY cent, 0.1 per cent, 0.2 per cent, respectively. At the same time a 0.025 per cent solution of Liebig’s extract of beef in tap- water was made. The sister cells obtained by several divisions of a single Para- mecium were distributed between the various fluids and grown for some time, being changed to fresh solutions daily. The results for growth during two weeks are shown in tables 1 and 2. Tables 1 and 2 show that deaths were frequent in the 0.2 per cent hay infusion, and with both P. aurelia and P. caudatum the results showed the superiority of the 0.1 per cent hay infusion and the 0.025 per cent beef extract solution. From this time on the 0.1 per cent hay infusion was used as the culture fluid for all work. Occasionally the 0.025 per cent beef extract was used and its use will be indicated each time. The hay infusion was used rather than the beef extract solution, in order to have the solutions of the same sort as an ordinary infusion. The objection might be raised that since Woodruff, and Woodruff TABLE 1 Growth of Paramecium aurelia in mixed cultures of bacteria, in media of different concentrations. Figures represent number of divisions. H. I., hay infusion; B.B., beef extract; ' dead replaced from 0.1 per cent hay infusion Paramecium aurelia I MARCH APRIL AVER- oe ee 25 | 27] 28 | 29| 30] 31} 1 | 3 | 4 | 5 | 6 | SIONS loprpay (OO are Cemmedsls Ie ssseoceon| 2 eye | 2 ye ak |) Se ae aye A PAL 1.615 Oil facie Game el, Wocesconcnes nll 2) & | lye | 2 psy iy 2 eh al 23 1.769 (2) foere Gemitelals Nocosoocssoch 2 | | Wo @S) 2 yi | ay Oy wey 16M e230 0.025 per cent B. B......... BP Ze eat Bs Wal yak ees pal al eal] a 20 1.538 Filtered infusion........... BP Ba Th SN Te OLN Bl Ba), Th 18 1.384 Paramecium aurelia IT 0:05spernscent EL. Tc shes QS We ED Qe he es Teal eat 3 le ee eel enctek O:percenthie sl Fa ee: Te eS | WP tl PE a eB |i TL 21 1.615 O:2per cent HoT. eects lee OE e282 ON Ue et a en 5} |) AL 1s} 0.025 per cent B.B.......... || eee | ORO ee ak ee at a Wal 22 1.692 HilteredmmiusiOns eee see PNW eo | eal BB | ek | ©] AO | wast PARAMECIUM IN PURE CULTURES OF BACTERIA 433 TABLE 2 Growth of Paramecium caudatum in mixed cultures of bacteria, in media of differ- ent concentrations. Figures represent number of divisions. H.1., hay infusion; B. B., beef extract; | dead, replaced from 0.025 per cent B. B.; * dead, replaced from 0.1 per cent H. I. . Paramecium caudatum I | MARCH | APRIL TOTAL acl a4 DIVI- |vistons 25 | 27 | 28 | 29) 30) 30) 1 || 3 | 4 | 5 | 6 | StON® pen pay OLOS jaar Gainth las Weosconoeen|| 2ull o> |) TE eae aly) al | at eat sO) yal 13 | 1.000 Osi! joer @eimti Iolo Wshocs oa Fu oll 2 al4| AU aaa | 2 |) ete) al 15), )) 10153 O:2pemcentria lees: sna. 74M GEM OIL Te eth al Te ORM eeh | alas i ah Oe (HOS joer GxiniplB. IBocpoo0 cul | ol) te Meat) 224) ab als) he 15 1.153 Hilteredkimfusionse.-222 oe. PO OAL 2 By BIN) 12 1.000 Paramecium caudatum II OMOhipericent lense We er a CON aL a ab) a) hal 10 | 0.769 OMe percent Hele ees ne. Bas PE MOE ale lal SP Oy |h) 12 | 0.922 Oh npen cent Hy Ieee... OZ NOZ eee tO) | es OM Oe Om nO2Za 02 0.231 O25 per cent Br Bie... (2 oe) OP UN 2) 2 see ete ORO Dy 4) 12153 Filtered infusion.......... PVN Oya aE ah Nh 40). ale | 2) WW). hoa 0.846 and Baitsell had found a solution of beef extract to be a better medium for constant use than hay infusion, we should have used it here in place of the hay infusion. But the experiments just recorded showed little difference for two weeks, and the above mentioned workers found the beef solution superior mainly when used for extended periods. Our cultures were planned to cover periods of about ten days, long enough to show the effects of pure cultures of bacteria on Paramecium, but not of sufficient length to be affected by the constant use of the hay infusion. Technic devised for the growth of Paramecium in pure cultures of bacterva. It has just been stated that when the most desir- able concentration of hay infusion was found (viz. 0.1 per cent, which is the standard solution of Jennings, diluted ten times with water) this one concentration was used for all subsequent work. a |< Mixed culture...............00.. 1 Tea | ik Al Wa 8 VD bole 7) Bacillus subtilis (Bact. XI)..... gl gL 2 1 0 1| 4 | 0.666 Transferred to Bacillus flavescens (Bact. IV) Bacillus fluorescens (Bact. III). 01 ol 1 0 1 0| 2 | 0.3383 : Transferred to Bacillus flavescens (Bact. IV) division being only 0.454 per day; this was only about one half the rate of a similar line in a previous experiment (0.909 per day). Since the paramecia fed on B. subtilis and on B. fluorescens began to appear dark in color as though filled with undigested food (these are marked as abnormal in the table) they were sterilized by washing and both lines placed on a diet of Bacillus flavescens (Bact. V). In the ease of the line transferred from B. subtilis to B. flavescens the rate remained the same and the abnormal appearance changed to normal. The B. fluorescens transferred to B. flavescens dropped in rate of division shghtly (0.454 to 0.333 per day). The rate of division of the same line of paramecia fed on this same B. flavescens in a previous experi- ment had been 0.857 per day, though death occurred within a week in that case. In the series recorded in table 8 all of the protozoa were alive at the end of the experiment. PARAMECIUM IN ‘PURE CULTURES OF BACTERIA 449 From the data recorded in the various tables it seems clear that cultures of mixed bacteria are, as a rule, far superior as a diet for Paramecium to a diet of any one kind of bacteria. Most bacteria in pure cultures, even when isolated from normal, healthy, hay infusions, were quite unfavorable if used alone. This was true even when such bacteria were present in enormous numbers in healthy infusions and were the predominant types. Only a single kind, Bacillus subtilis, approached a mixed diet as a favorable kind of food; this sometimes seemed to be a better and sometimes a less satisfactory food. It would therefore appear, under normal conditions, that Para- mecium thrives by virtue of the use of a diet of different kinds of bacteria. It is possible that Bacillus subtilis is the chief dependence of Paramecium as food, as some have claimed, but the experimental evidence does not show the superiority of this form over a mixture of different kinds. The probability that different bacteria are usually eaten is in accord with the struc- ture and habits of Paramecium. ‘This infusorian is one of many which has a mouth constantly open, and apparently there is no cessation in the beating of the cilia. Under such circum- stances it is hardly conceivable that there is any choice of food, rather all bacteria which are not too large are swept into the bueeal groove in the ciliary current and taken into the body. Doubtless some of the bacteria which get into the body are but slightly, if at all, digested and assimilated. If such forms of bacteria should come to be the predominant type in an infusion one would expect Paramecium to decrease in vigor and many of them to die. There is little doubt that some of the ‘bad’ cultural conditions observed in infusions are due to just this condition of certain bacteria gaining the ascendency and these being so unsatisfactory as food that the animals die. It should be possible by using pure cultures of bacteria, mix- ing these known forms in various combinations in sterile infu- sions and growing Paramecium therein, to secure a mixture which would be better than the ordinary mixed culture obtained by a chance infection of the culture fluid from air, hay, or water. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 450 GEO. T. HARGITT AND WALTER W. FRAY Such an attempt was a part of our program but we were not able to carry it out, on account of lack of time to give to it. Any attempt to correlate the characteristics of the bacteria with their favorableness or unfavorableness as food leads to little. From the table of bacteria it appears that a majority of the bac- teria which were recovered from abnormal infusions were of a spreading or running type of growth. It was bacteria of this type which produced the putrefactive or fermentative odors when grown in hay infusions, and when used in pure cultures as food for Paramecium the latter soon died. Any culture fluid which has any considerable number of this type of bacteria is therefore apt to be a very unfavorable medium in which to grow Paramecium. CONCLUSIONS The principal points brought out by this investigation may be summarized as follows: 1. Bacteria present in hay infusions gain access from the hay, from the water, or from the air. Some forms may be intro- duced from all three sources, otlters only from a single source. 2. Both normal and abnormal (fermenting and putrefying) hay infusions were analyzed and the predominant types of bac- teria present were obtained in pure cultures by bacteriological methods. The characteristics of some of the bacteria isolated are tabulated in a table at the end of the paper. A total of 30 different bacteria were isolated in pure cultures. 3. The bacterial flora of a hay infusion changes when the infusion is allowed to stand without adding fresh hay and water. Analyses made at the end of a few days, at the end of a few weeks, and again after four months showed a different flora each time. In old cultures colonies with amoeboid or lobose type of growth, and capsulated forms are present in great abundance. j 4. Hay infusions and beef extract solutions when sterilized in the autoclave at a temperature of about 130°C. are so modified by the high temperature as to be unsuitable as a culture medium for Paramecium, since in these fluids the animals died in a short PARAMECIUM IN PURE CULTURES OF BACTERIA 451 7 s time. If these fluids are sterilized at a temperature of 100°C. they are not so modified and are satisfactory media for the growth of Paramecium. 5. It was found possible to get the body of Paramecium abso- lutely sterile by washing in sterile water. This washing was done in sterile tap-water in depression slides enclosed within sterile Petri-dishes. A bacteriological examination of the wash waters showed no bacteria present after the third washing, but five wash waters were used in the preparation of every Para- mecium for further study. The Paramecium itself when tested by being placed on an agar plate developed no colonies of bac- teria showing positively that its body had been entirely freed from these by the washing. Paramecium so sterilized was not injured in any way and suffered no loss of vigor. 6. Sterile Petri-dishes, large enough to contain a single depres- sion slide, were used as moist chambers and it was demonstrated that they are as satisfactory as large moist chambers. Further- more the Petri-dishes have certain marked advantages as moist chambers over larger receptacles. 7. Pure cultures of bacteria, with Paramecium growing in them, usually remained uncontaminated by foreign bacteria for periods of at least two weeks (they were not tested out for longer periods). The most extreme case of contamination of such a pure culture was at the rate of 1 foreign bacterium to 350 of the pure culture. It is believed that such a contamination is with- out significance as having effect on the food supply of Para- mecium when the cultures are not carried for long periods. 8. Paramecia were grown in pure cultures of bacteria isolated from normal and abnormal hay infusions. In no case was a single kind of bacteria as satisfactory for food as a mixture of many kinds. This was true whether the bacteria came from normal or from abnormal infusions. The bacteria isolated from abnormal infusions were so unfavorable as to cause death of the protozoa in a few days; probably toxic excretions were set free by these bacteria. The only bacteria in pure culture which approached the control mixed cultures as satisfactory food for Paramecium was Bacillus subtilis; this sometimes seemed better 452 GEO. T. HARGITT AND WALTER W. FRAY and sometimes worse than the mixed cultures. A mixture of different kinds of bacteria, therefore, seems essential as a diet for Paramecium grown in depression slides; probably the same thing is true for large cultures. , SUGGESTIONS From the results obtained in this study it is seen that certain precautions are demanded of those who carry on pedigree cul- tures of Paramecium or other bacteria-eating Protozoa. Some of these are here suggested: 1. Whenever constancy in the matter of food is desired the pipettes used must be sterile. This is not obtained by the inser- tion of the pipette in boiling water for the spores of some bac- teria are resistant to this temperature. The pipette should be sterilized within a closed vessel placed in a hot air sterilizer, or by insertion in the flame of a burner. The pipette must be sterilized each time before being used. 2. Before starting cultures the protozoa should be sterilized. Bacteriological tests show this may be efficiently accomplished with Paramecium by washing through five or six sterile fluids in sterile depression slides enclosed within sterile Petri-dishes. Transfer is to made each time with a sterile pipette. 3. Where different cultures are to be critically compared all the protozoa, after having been sterilized, should be kept in identical culture fluids with identical bacteria present. So far as the present data goes it suggests the inoculation of the cul- ture medium from a normal hay infusion or by chance infection from the air as preferable to the use of pure cultures of bacteria. But the food must be identical and this can only be the case when a single fluid is inoculated and some of this is added to the slides in which the sterile paramecia are to grow. 4. If the demands for uniformity of food are very strict the slide cultures should be kept in a sterile moist atmosphere, pro- tected from the contamination of air bacteria. This condition may easily be met by placing sterile water in sterile Petri-dishes. Transfers of animals or addition of fluid, and the like, should be PARAMECIUM IN PURE CULTURES OF BACTERIA 453 made by raising the cover of the dish only sufficiently to permit the insertion of the capillary portion of the sterile pipette. If these precautions are followed the bacteria will remain practi- cally unchanged for a considerable time, so far as the entrance of new forms are concerned. 5. The degree to which these precautions are to be observed will depend upon the requirements for uniformity of conditions. Paramecium seems to keep in good condition and grow best on a diet of different kinds of bacteria and for most work the usual methods are satisfactory. It is only where the needs of the experiment are for the greatest constancy in all environmental conditions that the above refinements are called for. 6. While a mixed diet is best it must not be forgotten that the number of bacteria on the hay and in the air are very great (30 were obtained in pure culture by us, and these are only a portion of those present), and some of these are injurious to Paramecium. The food is a real factor to be considered and it may be controlled as precisely as desired. For critical work charice infection of media, or cross infection of cultures by the transferral of several animals with a single pipette, is not a satisfactory, nor a scientific method. 454 GEO. T. HARGITT AND WALTER W. FRAY BIBLIOGRAPHY Catkins, Gary N. 1902a Studies on the life history of protozoa. I. The life cycle of Paramaecium caudatum. Arch. f. Ent-mech., vol. 15, pp. 139-186. 1902 b Studies on the life history*of protozoa. III. The six hundred and twentieth generation of Paramaecium caudatum. Biol. Bull., vol. 3, pp. 192-205. Cuester, F. D. 1914 A manual of determinative bacteriology. The Mac- millan Company. JenNINGS, H. S. 1908 Heredity, variation and evolution in protozoa. II. Proc. Amer. Philos. Soc., vol. 47, pp. 393-546. Jennines, H. S. and Harairr, Gro. T. 1910 Characteristics of the diverse races of Paramaecium. Jour. Morph., vol. 21, pp. 495-561. Maupas, E. 1888 Recherches expérimentales sur la multiplication des Infu- soires ciliés. Arch. de Zool. Exp. et Gén., Ser. 2, vol. 6, pp. 165-277. Peters, Amos W. 1907 a Chemical studies on the cell and its medium.—Part 1. Methods for the study of liquid culture media. Amer. Jour. Phys- iol., vol. 17, pp. 448-477. 1907 b Chemical studies on the cell and its medium.—Part ITI. Some chemico-biological relations in liquid culture media. Amer. Jour. Physiol., vol. 18, pp. 321-346. Wooprvurr, L. L. 1905 An experimental study on the life history of Hypo- trichous infusoria. Jour. Exp. Zodél., vol. 2, pp. 585-632. . 1909 Further studies on the life cycle of Paramaecium. Biol. Bull., vol. 17, pp. 287-308. 1911 a The effect of excretion products of Paramaecium on its rate of division. Jour. Exp. Zodél., vol. 10, pp. 557-581. 1911 b Paramaecium aurelia and Paramaecium caudatum. Jour. Morph., vol. 22, pp. 223-237. Wooprurr, L. L. and Barrseti, G. A. 1911 c¢ The reproduction of Paramaecium aurelia in a ‘constant’ culture medium of beef extract. Jour. Exp. Zool., vol. 11, pp. 135-142. 1911 d Rhythms in the reproductive activity of infusoria. Jour. Exp. Zodél., vol. 11, pp. 339-359. lov i=] a ch UIIO}I}VI4S ‘Ty ‘Aqoyyvey ‘5D ulioj1deu “7 ‘opernuryoe ‘5 UIIOJI}8.14S ‘7 Sun0sTy “OD UIIOJI -deu ‘T !popvaq ‘5 uojideu “J ‘unoyg “D wiiojyideu ‘J fuuoyy “5H ,ouou ‘uIoyTTy “DH euou UO; “1 auou SOMIOTTY 4) UIIOJ17B148 ual SUOFTTY ‘5) CIIO}TZB.148 “T suo “p) qeqs UlyRley *auOU [OPUT ‘pozUEUIJej YOU Y4IOIq O801}xOp ‘oATZeSou WIeIS ‘sor0ds OU ‘e[T}OW-UOU = — poziuojded ) 0) ‘poyemnseoD | xX | -Hed ‘prqiny 9pryod peonpel “ploy VY jou ‘prqiny, peyruodes epr{jod Ajeyoydurog | xX yoy} ‘IBeO apet[[od euleyTy | VW | ou ‘prqiny aTqelsj ofo SORES) V | -ted ‘prqiny, (A[su074s) 991] ouTLe YTV _— -Jed ou ‘1veIOM 9a = -Tped ‘prqany, Ser posueyouy — | -jed ou ‘ivelD 9jorjed pesueyoug xe ou ‘Apno[g poytuod eotjed -BS ‘poysesiq x ou ‘Apno[pg peztu0y soled -ded ‘peonpeyy - ou ‘Apnolg o fom) yar snuryyT | 3A | yoIq joog Bg i) SAYOLVAd TVOINGHO-OI18 GNY Tvu¥oL1n9 ses OU !plov = X pepyuna ‘OUIYM ‘pasrey sutpvoids ‘poser ‘UIYL, sutpeoids ‘poster ‘AAGOTT soTuo -[00 punol ‘[[BuUg sutuuni ‘oyTyM = ‘YJOOTIg eyeoTyiquin ‘ayy = ‘pastey MOTIOA ‘poster 4stoul “Gg ‘AwT]s ‘ayy ‘AABO]] ueeIs pouinjy ese FOV M ‘pestey MO] -[oA Fy sty ‘AquBOg ry a 4SIOUL "g {MOT -[eA poster AABOTT IVsy Ivsy uo YJMOIy) + - ‘ourTeyye = Vy + | sn{[loeq osre7 unt _ -10}408q qI0Yyg + SN][loBq osIVT + | snqyoeq ysoqg + | snqpoeq yr04g uInt194 _ -oeq Joys + | snqjroeq ya0yg ‘aAtpisod wed ‘yueseid sorods ‘ayyour = ++ ‘uorjovjonbiy Jo odAy = J 11g pue ABTT uUOISNjur Avy [BulLouqy woIsnyul Avy [euliouqy uoIsnjul Awy [euouqy UOISN jut Avy [euiouqy uoISnjut Avy [pulouqy ‘aoBjIms = g UBULIOUL -wTZ = ‘sny4eqqns sny{[loeg stolaleral slate et ohetaiateestelereistenstets an -sn’'T ‘shoijlipuep sn{p[lovg poyrjuept oN pur| Avy |-YuBsy ‘susosoavy sniplorg SOCIO DOr TPOUOARY, SNy[ORg lel wieveluie) wie ieiwratars suorovyjonbTy -uoU—SUdsd0S010N] Shoe erketaiefiis is: aisfersiauataite ueu0Ur -mrz ‘winyeoyd untieyoeg ‘oDIN[Y ‘Shay snooo0I 01] -+ | snq{[loeq esreT ITV + snjploeqd 404g 1978 wn — |-lueyoeq 41049 Avy sno000 — | -O1orUr ][BUIg IO} Ay d -MOW ezis pus odeyg 901n0g DNINIVLS GNV ADOTOHdUONW GWYN "YQMois = 5 IX x XI IITA TIA IA A AI Ill UudgdWwoNn suoisnfur finy JoUWLoUuGgD puv JDWLOU WoOLf payvj0S7 D149}IDq ay) JO saunqoaf ooLWaYys-019 puM JoLnyyna “Wwaibojoydow ay) Burnoys 4404) 455 ey SAT ae + 8 Welt or a (al, s i) *- s f salah hah 3 THE EFFECT OF LONG-CONTINUED HETEROZYGOSIS ON A VARIABLE CHARACTER IN DROSOPHILA WALTER W. MARSHALL AND HERMANN J. MULLER Rice Institute TWO FIGUKES The belief that a factor may sometimes be contaminated by its allelomorph when the two meet in the hybrid has been upheld by Castle and by some geneticists of the non-Men- delian camp. On a priori grounds there is no reason why this might not occur, but there is no evidence for arriving at such a conclusion. The bulk of Mendelian inheritance seems to show that factors are not affected by their allelomorphs. Bateson supposes that some cases of multiple factors are due to fractionation and that the products of this ‘quantitative disin- tegration’ segregate independently of one another. He explains these supposed fractional degradations as due to irregularities in the segregation of the factors in the germ cells, during cell divisions in which he imagines the qualities to be sorted out each to its place. In such a case a character might become weaker and weaker as a result of continued ‘crossing to other stocks even though it originally differed from these stocks in only a single factor. The aim of the experiment considered in this paper was to contribute evidence in regard to the constancy of factors in a state of heterozygosis; it was believed that the apparent fluctuation in factors, which is thought by certain workers to be contamination or some sort of quantitative disin- tegration, can be accounted for on other and more satisfactory grounds. The subject of the present investigation is a variable wing character in Drosophila ampelophila which is called balloon, and the factor for which lies in the second chromosome. 457 458 WALTER W. MARSHALL AND HERMANN J. MULLER The balloon factor was kept heterozygous for at least fifty generations, covering a period of nearly three years. Hetero- zygous males having the mutant factors streak,' dachs, jaunty, curved, and balloon in one chromosome, and the mutants black, purple, vestigial, arc, and speck in the other, were repeatedly crossed to homozygous females containing black, purple, vestigial, are, and speck. All these mutant factors are recessive except streak, so that the male offspring containing the first named combination of factors were easily picked out. Since there is no crossing over in the male and the females used were pure for b-p-v-a-sp, the factors of each kind of chromosome were kept together, and there were only two kinds of males and females produced in each generation, of the same types as those above described. One kind showed the character streak, a marking on the dorsal side of the thorax, and in the other kind this character was absent but the characters of b p v a and sp were manifested. The males containing streak thus always contained the factors for balloon wing and for its normal allelomorph. The close association of these allelomorphs over such an ex- tended period furnished an excellent opportunity for contami- nation if contamination really occurs. Before the balloon character could be studied it was necessary to separate it from the other mutants and obtain flies homozy- gous for balloon. This process of purification was a somewhat complicated one. The accompanying diagram will serve to illus- trate the method of ‘freeing’ the balloon factor. Males of the heterozygous type just described were crossed with normal fe- males. The F, offspring were heterozygous, some containing the Sdjcba chromosome and the normal chromosome, others con- taining the b pvasp and normal chromosomes. The latter were eliminated. Males and females of the former type were then crossed with each other. Part of the offspring (/2) from this cross received from their mother balloon but none of the other mutant factors (the balloon having crossed over), and from their father they received the chromosome containing Sdj cba. These 1 The factor streak was introduced in the way described only during the later generations. 459 DROSOPHILA EFFECT OF HETEROZYGOSIS ON | INS y I[® Laas sUuol} } (uoojpuq) “4 MT (papvase]) os on Bol}s EO BULIOU 4 I 5 eqatpg ooo Feer ae , fe ie vqio 1p eu SoyOUBr) Pensomae suUor (}0038 pyLA wo4y) ion ayewe,y (uoojyeq years) - Sota rN Wd oT RS 099 ees g eq eq Oo p g ‘eo Sg af pg sajyawen | (pap.nosiq ) conan ea = an [eui0U jeuLi0U ~yeuL0u — . (Bulow) deaadiq Pee ips aula] (3{89.1}8) 2eee qo Tp TN W a eee Se dsvad es Q sayourey (jeurtou) ieee o[BUL9,T (3780498) de eed IVAW VW yeursou eqo [ pg 460 WALTER W. MARSHALL AND HERMANN J. MULLER flies were homozygous for balloon but heterozygous for the other factors and they were identified by the phenotypic appear- ance of the balloon and streaked factors. The other combinations djcb b to wild (normal) females in order to eliminate the Sdjc. Half of the F; flies received from their father the chromosome with Sdjcba and half received the chromosome with ba. Both types of course received from their mother the normal chromo- some. ‘The two sorts of flies could be distinguished by the presence or absence of the character streak. Those flies con- taining streak were discarded. The Ff; males and females that were normal in appearance were bred together and their resulting offspring (F'4) that showed phenotypic balloon were homozygous for this character and contained no other mutant factors. Now that this balloon factor which had been in a state of continued heterozygosis was freed, it remained to be compared to the balloon in flies which had been kept homozygous for a con- siderable length of time. The flies used for this comparison had been kept homozygous for an even longer time than the others had been kept heterozygous, the latter having in fact been derived from this homozygous stock. The first problem that was encountered here was to arrange the variations of the character in a consistent series. After a large number of wings had been observed seven grades were established which approximately accounted for all the varieties that had appeared. The accompanying diagrams (fig. 2) will serve to show the nature of these grades, which are based upon the wing venation. The sorts of variation used as standards for these classes were fairly typical, and although all intermediate conditions were found there was seldom doubt in deciding to which grade a particular wing belonged. The different grades may now be considered in detail. 1. Some few wings showed no modification whatever; these were classed as normal. In all others the balloon character made its appearance near the posterior margin of the wing by an : S were discarded. Next, males of this type ; were crossed EFFECT OF HETEROZYGOSIS ON DROSOPHILA 461 ae 1 normal 2 slight 2 slight 2 slight 3 moderate 3 moderate 4 marked 4 marked 4 marked He » 5 small blister 6 medium blister 7 large blister Figure 2 462 WALTER W. MARSHALL AND HERMANN J. MULLER abnormality in the region of the cross-vein which connects the fourth and fifth longitudinal veins. 2. If the modification was merely a bend in the cross-vein, it was classed as ‘shght.’ Likewise if the disturbance took the form of a slight projection, eather proximal or distal, it was considered in the same class, or if the extra mark was appar- ently disconnected or removed from the cross-vein. 3. If short projections or markings appeared, both proximally and distally, the wing was classed as ‘moderate.’ Also if only one projection occurred and this was more extended than those of the class ‘slight,’ even so much as to become branched, it was recorded in this grade. 4. If fairly extended projections in both directions from the cross-veln appeared, the wings were graded as ‘marked.’ Such projections were frequently branched, especially distally. Wings in which the venation in this region was confused or considerably disturbed were classed in this lot. 5. If a small watery blister appeared in this region of modification which was considerably less than half as long as the entire wing, it was placed in the class ‘small blister.’ 6. When the blister was approximately half as long as the wing, it was classed as ‘medium blister.’ 7. The grade ‘large blister’ was established for a similar con- dition that occupied the entire or nearly all of the wing surface. The type and the range of each class being fixed, the question next occurred as to what their order or relative degree of devia- tion from the normal might be. After a comparison of the amount of modification in each class, the order in which they have been described was adhered to, and experimental support for it was also found and will be described presently. The method of observing and recording the variations was carefully considered so as to avoid bias. The flies of either the stock (of homozygous ancestry) or of the outcrossed type (of heterozygous ancestry) were inconspicuously marked by ampu- tating a small portion of one member of the front pair of legs. The front pair were chosen because these legs usually project backwards beneath the body rendering them less obvious, and EFFECT OF HETEROZYGOSIS ON DROSOPHILA 463 secondly because these were most easily accessible. The ‘branded’ flies were thoroughly mixed with those to be compared with them before the record was taken. Each wing was first graded and then identified as stock or outcrossed type. In this process of classification a diagram ofthe grade types was constantly referred to so as to make the judgment as uniform as possible. Each wing was recorded separately, there being no satisfactory way in which to determine the variation of the fly as a whole. As far as possible the crosses for comparison were started on the same day, using food from the same batch, of the same age and consistency, and the bottles were placed in the same con- ditions of environment. Approximately the same amount of food was used in each case; and with the exception of a few instances the same number of parents were allowed to breed in each bottle and have access to this food. The food and rearing problem was without doubt the most difficult question encountered. It was necessary to start many crosses in order to insure a sufficient number of offspring. Un- fortunately, owing to the infertility of the flies, it was often impossible to obtain counts on the same date from stock and outerossed bottles that had been started simultaneously and with similar food. But the results include observations on a number of bottles and are consistent enough to show that varia- tions in the character due to environmental differences between the bottles were not responsible for the main outcome of the comparison. The flies for each day were recorded in a correlation table, one mark representing both wings, the modification in the right wing being placed in its proper column within the row labeled for the modification of the left. The data obtained in this way gave a correlation between the right and left wings which justified the order in which the grades had previously been classified. For example, that ‘small blister’ is a greater deviation from normal than is ‘marked’ is proved by the fact that ‘small blister’ (in one wing) shows a greater tendency to be associated with ‘marked’ (in the other wing) than with ‘moderate.’ This may be seen by consulting table 1. Similarly each of the other 464 WALTER W. MARSHALL AND HERMANN J. MULLER TABLE 1 RIGHT WING Nor, | stght | Mode starkea| Spall [Medium] Large | ota INonmaleee. 22). Sa 122 28 1 0 0 1 0 152 EAC), ar ee 1A NWed90Us e3Se| V4 1 3 0 | 250 | | Moderate........... DA 53al 130))|. a5 8 4 3). | 2235 Plaid... 25-24. G15 | "23 004" 36 «206 |. 26 an aalmommalll blisters. sasece 0 2, 6 | 30 Hil 36 17 168 - | Medium blister..... 0 0 BY ty 2 28 36 4 91 Large blister........ 0 1 1 12) 21 3 3 41 Motels... .. 002.20 oe 88a 279" 1" 202m B05 ealmtia: | Osha, Areas grades may be proved to have the order in the series which has been assigned to it. Exception must be made in the case of the last grade, ‘large blister,’ but in this case the numbers are not large enough to be significant, andit seems fairly obvious that the ‘large blister’ represents a greater deviation from normal than medium blister (especially since medium blister is shown by the table to be less normal than small blister). In table 2 the results for both right and left wings have been added together, but the counts for the stock and outcrossed types are shown separately. All counts produced by bottles that were started on a certain date (with similar food) are grouped together, with their results for each day shown on a separate line. It should be stated here that the offspring were kept one day after being taken from the bottle, in order to have both types of approximately the same age when graded. It will be seen from these figures that the outcrossed type which has been kept in a state of continued heterozygosts deviates even further from normal than does the stock. The average grade of the former is 4.2, which places this type in ‘marked,’ and the average grade of the stock is 3.0, or ‘moderate.’ The figures show that in twenty-five cases the stock is weaker than the outcrossed type. In only five cases is the outcrossed weaker than the stock, and in one case both are of the same value. EFFECT OF HETEROZYGOSIS ON DROSOPHILA 465 In four of the six latter cases the number of outcrossed flies (2, 4, 1, and 3) was too small to be of significance. In the other two cases the number was somewhat larger than in the above, although still rather small, but here there is very little or no difference between outcrossed and stock (—0.2 and 0). By comparison of the lots that were started on different days in either stock or outcrossed, it is evident that there is variation in the character (presumably due to environment) quite apart from any difference between the stock and the outcrossed. The differences between the latter two types, however, are too con- sistent to be due to chance operations of the same cause. Neither is the difference due to contamination from the nor- mal allelomorph, otherwise the outcrossed would be nearer the normal than the stock is. We find, on the contrary, that the out- crossed type is really better balloon than the stock. Moreover, the standard deviation of the stock is 1.55 grades, and of the outcrossed is 1.37. Hence there has been no increase in_ the variability in the outcrossed. The greater variability of the stock constitutes additional evidence that there has been no contamination or fractionation. The difference, then, must be due either to variation in the balloon factor itself (multiple allelomorphs) or to a difference in modifying factors (multiple factors). Assuming that varia- tion of one type or the other could occur at all, natural selection would tend to cause the races finally to differ in the direction found. For the wings that made for increased balloon, being less adequate for efficient flight, would tend to be eliminated. Also the watery blister, which usually bursts, often causes the fly to adhere to the side of the bottle and perish. The inbred stock has had an excellent opportunity to become changed in the direction of normal in the period of three years, but the out- crossed did not have a chance for such selection, since balloon is recessive and was kept hidden while in the state of hetero- ZYZOSIS. Tests could be made to determine definitely whether the change was due to multiple allelomorphs by the use of linkage experi- ments, since the knowledge which hes elready been obtained of THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 2 DATE STARTED April 30 May May May May May May 138 DATE GRADED es Z S is) > NORMAL SLIGHT MODERATE MARKED SMALL BLISTER May 15 May 16 May 17 May 18 May 19 May 20 Total May 15 May 16 May 17 May 18 May 19 May 20 Total —" Nw He & Oro 1) TAB MEDIUM BLISTER LARGE BLISTER TOTAL AVERAGE “I May 15 May 16 May 17 May 18 May 19 May 20 | Total May 15 May 16 May 18 May 19 Total May 18 May 19 May 20 May 21 May 22 May 23 May 24 Total May 18 May 19 May 20 May 21 May 22 | May 28 May 24 et weoewa|Elonur es | eS Nj/ooco rH -& Hannes|oocuecs — — [=] — May 17 | — CC a on en ie We) bo = * NOUR BBR WR | ro} Re bw eS bb ele — NOCOCOCONONIN I LP OOAWwWO Slaton cee CONN WROD or CONF WANS WH Total p ee bo May 24 May 25 May 26 May 27 May 28 May 29 Total bd ON w oO SD OS] Or > [o,0) wm TN WBWWNWIN|IHONNNNININYNNH NWN DWI WH wWwWh Ww — ee) SS Ss | | ee bo (0) — (Su) bo iN WOrRWFr DW] NA nS Hi Grand Total —" CO; rF RF Or ot) aa 330 99 46 | 1676} 3.0 : iN _ or DIFFER- ENCE (B-A) GHDOVUAAV UaALsi Teh GSuvit UaLsiTa GLVuaaqow 3 ie) w nw S S s 3 So isa) UaALSITa TIVNS qayuvn '~ GLVuaqaqon =r Lapis | IVANUON 468 WALTER W. MARSHALL AND HERMANN J., MULLER linkage groups in Drosophila is adequate for this purpose. It is important to know whether this difference is due to multiple allelomorphs or multiple factors since if it were found to be due to multiple allelomorphs the, further question would be raised as to whether it was due to one or two mutations in the factor for balloon or whether it was due to natural selection work- ing on continual fluctuations in this factor. There is no evidence at present to show whether the latter is true, but this question too could be determined by certain breeding tests. Another experiment was made with the curved wing character in Drosophila, which was treated in the same general way. The factor curved (c), like balloon, was in the chromosome S dj c ba that had been subjected to the long-continued outcrossings. After it was purified by a series of crosses in a Manner similar to those made in ‘freeing’ balloon, it was compared to stock which had been kept homozygous for an extended period. Owing to the infertility of the curved winged flies the number of resulting offspring was small. Nevertheless the results corroborate those already given for the balloon character. The character curved appears as a convexity throughout the general surface of the wing and the grades are based on the degree of variation found. After careful observation 4 grades of intensity were established, which may be called: (1) ‘Shght curvature,’ (2) ‘moderate curva- ture,’ (3) ‘fair curvature’, and (4) ‘marked curvature,’ and in addition there were also found a number of cases which had to be grouped in a fifth grade, (5) ‘spongy wing.’ In this experi- ment there were only two strictly comparable lots of stock and outcrossed flies (made up and counted on the same day), Owing to the small numbers of stock offspring, an extra lot ‘of stock flies started on a different day was counted. Table 3 shows the correlation between the right and left wings in the case of curved, as table 1 did for balloon. Table 4 gives the values obtained for the different lots of flies examined. In these examinations, as in the case of the balloon, the flies of one kind were marked and then mixed with those of the other kind. The grades of the wings were then determined before they were identified as stock or outcrossed. EFFECT OF HETEROZYGOSIS ON DROSOPHILA 469 TABLE 3 SLIGHT MODERATE FAIR MARKED SPONGY TOTALS Shishitepame <5 eee 4 2 0 i) 0 6 Mioderuittensc..sscces. 0 14 8 4 0 26 LGW ae A, Been Seen 0 6 17 3 1 27 IMiagrkedis jae crt n cee 0 3 il i} 1 18 DOM Oye eee 2 0 2 0 1 5 Tho gallstead chee sate 6 25 28 20 3 82 TABLE 4 1 2 3 4 5 DATE a eo . TOTAL AG a Pa Shght Moe. Fair | Marked] Spongy Cae pRovalgstocksasere reer tr May 2 1 4 2 2 1 9 2, Total outcrossed....... May 2 2 5 16 14 1 37 Roll Motalistock. ys. ssseeea: May 5 4 Bll | 28 10 3 33 DP Standard deviation of stock, 0.81 grades; of outcrossed, 0.84. In the totals and averages the number of spongy winged flies were not included, because the position of this grade in the series was uncertain, but the number of these flies was negligible. It will be seen that here, too, the rather unexpected result was obtained, that not only was there no evidence of contami- nation but the outerossed was, if anything, less normal than the stock, and did not show a greater variability. It should be mentioned that the same result had previously been obtained in the case of the character dachs legs, when comparisons were made between stock flies and outcrossed flies Sdjcba bpva sp few flies were examined, and those of the outcrossed race had not been freed from the other mutant factors, which were pres- ent in heterozygous condition. derived from the race In this case, however, only a 2 Muller, H. J. The mechanism of crossing-over. American Naturalist July, 1916. ; 470 WALTER W. MARSHALL AND HERMANN J. MULLER CONCLUSION The results above recorded give further evidence of the fre- quency with which characters are variable and change geneti- cally without any artificial selection. In the case of the charac- ters dealt with in this paper at least the variation that was observed was certainly not due to contamination or fractiona- tion. This was true in spite of the fact that the factors had been. kept in a state of heterozygosis for over fifty generations. It should be noted that if the balloon and the other char- acters observed had had a‘higher survival value than the nor- mal instead of a lower, they might have varied in the opposite direction from that observed and the results might then have been mistakenly attributed to contamination. This fact empha- sizes the importance of not accepting results apparently showing contamination or fractionation of factors at their face value without a thorough factorial analysis. THE RELATION BETWEEN SPECTRAL COLOR AND STIMULATION IN THE LOWER ORGANISMS! S. O. MAST The Johns Hopkins University FOUR FIGURES CONTENTS ] ay{ TROY NEGLI) Ne creees Bigaicex Aa othe tele Olan One RAE eat S055 3.chy a SIAR 38 ong Pee 472 Methodsrandematentals 42 Stasis cers ee cin ches sic, soph pees ca ORT ME oko: 483 Hixperimientt aliObSeEwatlOms jesiac cr cucrs cise 3056 ccs cle Ie oe ioe Secon tee 491 UeiELOCUCHONG rte weiss eee ciel Sais 3.53560 5 os, wea ee te eRe ae noe 491 BA eV eri aeavAL TCLS fee acer Re eae I cok Ss = acs inte Woe RR Pe Sie Toes, Sheaves 492 Pretenaeeraeiisas.: tascpea eaten s fee fed ells wd s Catan RA ete eee ae ets 497 1d aAIGND, HAVES conooomoocuscooueuguaeee io Ey EIR 499 EU lene remit aba. «Pee hee sive aces. Abed. « cos) a peg ae Rd ogre ae 500 Hilo erra eran lana =) esas ee 517 AB Lovwetliyes (laiiyjae) parrot ccs RR Te © oc ls fecs 2) sale cic tern ge 519 IDISGTISSIONS 5 opreepe ame eh tn HAs (Ser tM. 5 ot ME oo 1b Ad ob. Se apn ee aS, Shel 522 SUTIN LTS SES On eM oP SS ocho ikifs [6 oke = 0 ws Oh Dae ae Se ARE ages Nes 524 JB ONG} ag of Oh ater 3 oic'otcs a: Bend orn hee Cea ae a i a 526 1 Contribution from the Nela Research Laboratory, National Lamp Works of the General Electric Company. The literary work connected with this paper was done at the Johns Hop- kins University; the experimental work at the Nela Research Laboratory, Cleveland, Ohio, during the summer of 1915 (July-September). It is a genuine pleasure to express my appreciation of the courtesies extended on every hand by the director of this laboratory and his associates. Much expert advice of the greatest importance, especially that concerning the manipu- lation of electrical apparatus, was received from various members and the re- sults obtained both in quantity and in quality are in no small measure directly related to the generous interest taken in my work and to the unexcelled facilities placed at my command. 471 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 3 APRIL, 1917 472 Ss. O. MAST INTRODUCTION But little interest was manifested in the response of animals to colors until some time after the appearance of the Origin of Species (1849). This work at once greatly stimulated research | in comparative morphology and embryology, with the result that there was soon collected a mass of material in support of the theory of evolution in so far as it pertains to the structure of organisms. In connection with this work on the origin of structure there continually arose problems concerning the evo- lution of functions, reactions, behavior, consciousness, etc. Among these was the question as to the relation between color and stimulation. This question was associated with the prob- lem of evolution of psychic phenomena, and that of the evolu- tion of color-patterns in animals and plants especially those in the flowers. As applied to color-patterns in animals, the theory of evolu- tion demanded color-vision on the part of the animals involved, for the patterns were considered to be associated with concealing phenomena and sexual selection; and as applied to flowers it de- manded color-vision in those insects which have to do with pol- lination, for it was assumed that color is a determining factor in attracting insects and consequently that the reproduction of a plant depends upon the color of its flowers. But these ideas did not result in as much experimental work on the response of animals to colors as did the problem of the evolution of psychic phenomena. It was at this time maintained that if psychic phenomena originated by the process of evolution, one ought to find evi- dence of mental faculties in the lower organisms. And there soon appeared a group of investigators who took up the study of the behavior of animals primarily in search of just such evidence. Prominent among these may be mentioned Bert (’69), Darwin (’80), Lubbock (’81), Romanes (’83), Graber (’83) and Preyer (’86). In general, the results obtained by the men just mentioned led them to conclude that psychic phenomena extend well down in the scale of animal life, if not to the very bottom. SPECTRAL COLOR AND STIMULATION 473 Some of these investigators maintained that all classes of vertebrates, some of the arthropods and even some of the worms have color-vision. They used the so-called preference method in their investigations on the response of animals to colors, and largely owing to this and to the fact that they proceeded from a psychological point of view, their work has been severely criticised. However, a number of them, Lubbock in particular, used monochromatic light of high purity, and made a very thorough objective investigation of the subject; and more- over, a large proportion of their results has been abundantly confirmed. But many of their conclusions, especially those con- cerning subjective sensations have not been generally accepted. Botanists have been interested in the response to colors very largely from a purely physiological point of view, and a num- ber of them have made very extensive and thorough studies of the subject, using spectra of high purity, e.g., Guillemin (’58), Wiesner (’79), Strasburger (’78), and Blaauw (’09). In gen- eral it has been found that the region of maximum stimulation for green plants is in the blue, although some hold that it is in the violet; and that for fungi it is somewhat nearer the red end of the spectrum than it is for green plants. The most extreme shift in this direction that is known was discovered by Engelmann (’82) in his work on Bacterium photometricum in which he found a primary and a secondary maximum, the former in the infra-red and the latter in the orange. A number of investigators have also been interested in the study of responses of animals to colors largely from the point of view mentioned above. Prominent among these may be men- tioned Engelmann (’82), Verworn (’89), Hess (’10) and others. Loeb should probably also be included in this group. His alm was, however, quite different from that of the other inves- tigators mentioned. He was not interested in the evolution of consciousness, and he objected strenuously to the conclusion that psychic phenomena are involved in the reactions of animals. In support of this objection he attempted to prove that the reactions in plants and animals are fundamentally identical, maintaining that such a proof would show that there are no 474 S. O. MAST psychic phenomena involved in the reactions of animals. One of the points of identity that he undertook to establish referred to the relation between color and stimulation in plants and ani- mals. He rejected the results obtained by earlier investigators (Lubbock, Graber and others) which militated against his idea of identity, on the ground that the method employed by these investigators is faulty. They ascertained in which of two or more colors the organisms tend to aggregate, i.e., they used the so-called preference method. Loeb maintained that the results thus obtained have no bearing on the question as to the rela- tion between color and orientation, the phenomenon primarily involved in his idea of identity. He consequently studied directly the effect of color on orientation. His methods in the earlier work (90) were, however, very crude. He used only two different colors, red and blue, and the constitution of neither was known, being produced by colored glass or colored solutions. Observations were made on the fol- lowing animals: museca larvae, plant lice, caterpillars of Por- thesia chrysorrhoea, moths of Sphinx euphorbia and Geometra piniarla, various copepods, the meal worm, Tenebrio molitor, and larvae of the June bug, Melolontha vulgaris, Limulus poly- phemus and Polygordius. Loeb maintains that all of these ani- mals responded in blue just as in white light and in red just as in darkness, and he concluded that in this respect the reactions in plants and animals are identical, in spite of the fact that Kraus (76) had demonstrated that the stalks bearing the perithecal heads of the fungus, Claviceps microcephala, turn toward the light nearly as rapidly in the red as in the blue and Brefeld had obtained similar results for Pilobolus microsporus and Pilobolus erystallinus. In 1910 Loeb for the first time madeé use of spec- tral colors. He and Maxwell tested Daphnia, Balanus larvae and Chlamydomonas using the preference method, so severely criticised in his earlier work, and found the green or yellow region in the spectrum to be the most effective for all, thus confirming the results obtained on Daphnia by Bert (’69) and Lubbock (’81) which had been rejected in his earlier work. SPECTRAL COLOR AND STIMULATION A475 Later Loeb and Wasteneys (1516) found the region of maximum stimulating effect in a carbon-are prismatic spectrum to be in the yellow (560-578 uu), for Balanus larvae; in the blue (460-480 uu), for Eudendrium; in the Blue near 495 uu for Aren- icola larvae; in the blue (460—490 uu), for Euglena viridis and in the green about 535 yu for Chlamydomonas. These results are in full harmony with those of earlier workers in so far as they indicate that the relation between wave-length and stimulation is not the same for all animals. Loeb, however, still holds that it is the same for plants and animals, for he maintains that the fact, that the maximum in Chlamydomonas is in the green, shows that for plants it is not exclusively in the blue, just as has been found to be true for animals. This fact was, however, fairly well established before his first work appeared, as reference to the following table (table 1) containing a summary of the re- sults obtained in previous experiments will show. Other con- clusions reached by Loeb and Wasteneys will be considered later. The more important conclusions that a study of this table warrants may be summarised as follows: 1. A large proportion of the investigations were made with the use of prismatic spectra, insuring fairly pure monochromatic light, but in some gaslight was used as a source of illumination, in others electric light and in still others sunlight. Moreover, in some experiments the effects of the different colors on the process of aggregations were studied, in others the effects on orientation and in still others the effects on activity. The distribution of energy in the different spectra used is however, similar in all and in all cases in which two or more of the three reactions men- tioned were studied on the same organisms, the results were essentially the same, indicating that the relative stimulating effect of the different wave-lengths is the same for all three sorts of response. It is consequently possible to compare directly the results obtained by most of the different investigators as they are given in the table; but these results refer only to the relative stimulating effect of rather large regions in the spectrum and they are of such a nature that they show but little more than 476 Summary of previous experiments on reaction to colors in plants and lower animals INVESTIGATOR |DATE Poggioli Payer Gardner Dutrochet and Pouillet Guillemin Sachs Miller Brefeld 1817 1842 Brassica and Raphanus Seedlings Seedlings Roots of white and black mustard Seedlings of cress and Sinapis alba Seedlings Cress and Sin- apis alba Claviceps (fungus) Pilobolus Ss. O. MAST TABLE 1 Plants ORGANISM METHODS Seedlings of | Spectrum Solar prismatic spectrum and color media Solar prismatic spectrum Strong solar prismatic spectrum Solar prismatic spectrum. 25 tests made Red and blue solutions and glass Solar prismatic spectrum Colored media RESULTS Turned toward light in red and more strongly in violet. Other colors not mentioned Green, yellow, orange and red act like darkness. Turn toward light in violet and more strong- ly in blue Seedlings turn toward light in all colors but most strongly in indigo All rays active including infra-red and_ ultra- violet but blue most active Primary maximum effect in violet or ultra-violet. Secondary between infra-red and_ green; minimum effect near boundary between blue and green (490 uu) In red no reaction, in blue strong curvature Maximum for cress in blue at 490 wy. Maxi- mum for Sinapis, green at about 550 wy. In weaker light maximum nearer red Reaction in red nearly as strong as in blue Reaction in red nearly as strong as in blue eINVESTIGATOR Biaauw Strasburger | ee —————SSSSSSsSsSSSSSSSSSSSSSSSSSSSSSSSMsSFs Collect in blue (470-490 eS es EE EO OOECOO DATE 1865 1878 SPECTRAL COLOR AND STIMULATION TABLE I—Continued ORGANISM Roots plumules of seedlings Seedlings Avena Ply- comyces METHODS and | Solar spectrum and tested absorbing media of | Spectrum. Equal energy Unicellular green forms Swarm spores Swarm spores, mainly Bo- trydium Paramecium bursaria dis Bacterium photometri- cum Colored glass Solar spectrum, quartz prism, colored glass and solutions; orientation Solar and gas microspec- trum Solar and gas microspec- trum Solar and gas microspec- trum Solar and gas 1 ' Red no response. 477 RESULTS Maximum effect in vio- let. Secondary maxi- mum in red. Maxi- mum in green Maximum efficiency for Avena in blue 465 up, for fungi nearer red Blue most effective on orien- tation Violet, indigo and blue only rays cause orien- tation. Maximum in indigo. Other colors cause quivering move- ment Active in red, orange and yellow, not in other colors Collect in red (650-700 pp). Probably reaction to oxygen 9) Maximum collection in infra-red (800-900 pp). Secondary collection in orange (580-610 up) Verworn Loeb and Maxwell Oscillaria Chlamydomo- nas microspec- trum Colored _ solu- tions and glass red, yel- low, green, blue Spectrum pris- matic and normal Oscillaria orient in all colors, diatoms only in blue and violet Maximum aggregation about 520 up TABLE 1—Continued INVESTIGATOR |DATE ORGANISM METHODS RESULTS Loeb and 1910} Chlamydomo- Carbon-are Maximum effect on orien- Maxwell nas spectrum tation about 535 uu Loeb and 1910) Euglena viri- | Carbon-are Maximum aggregation be- Maxwell dis spectrum tween 460-510 uu about at 485 wu. Maximum effect on orientation 460-490 pu Lower animals Bert 1869} Daphnia Electric spec- | Maximum aggregation in trum yellow and green. Ori- entation in all colors most active in yellow and green Merejkow- {1881} Dias and larvae (?) In equal brightness, ef- sky of Balanus fect same in all colors Lubbock 1881} Daphnia pulex | Solar prismatic | Maximum aggregation in : spectrum green. Order of prefer- enee, 1G, Y¥,71G, RVs Wilson 1891} Hydra viridis | Colored glass | Maximum aggregation, and gas spec- blue (430-490 ppm) trum Yerkes 1899} Simocephalus | Welsbach gas | Maximum aggregation prismatic yellow. Order of pref- spectrum erence Y, O, G, R, B, V Harrington /|1900) Ameba Colored media | Maximum effect on activ- and and spectrum ity violet Leaming Mast 1909} Ameba proteus | Solar prismatic | Maximum effect on activ- spectrum ity blue (480-490 uy) Loeb and 1910| Daphnia Prismatic and | Maximum aggregation in Maxwell normal spec- green trum Loeb and 1910} Balanus Prismatic and | Maximum aggregation in Maxwell (larvae) normal spec- green trum Hess 1910} Daphnia Nernst glower | Movement of eye, order prismatic spectrum 478 of effect beginning with maximum G, Y, B, R INVESTIGATOR |DATE SPECTRAL COLOR AND STIMULATION TABLE 1—Continued ORGANISM METHODS 479 RESULTS Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Hess 1910} Porthesia Nernst glower chrysorroea prismatic (larvae) spectrum Hess 1910} Hyponomenta | Nernst glower variabilis prismatic (larvae) spectrum Hess 1910} Dasychira fas- | Nernst glower : celina prismatic (larvae) spectrum Hess ~ 1910} Lasio campa | Nernst glower potatona prismatic (larvae) spectrum Hess 1910} Phrogenatobia | Nernst glower fulignosa prismatic (larvae) spectrum Hess 1910} Culex pipiens | Nernst glower (larvae) prismatic spectrum Hess 1910} Culex pipiens | Nernst glower (adults) prismatic spectrum eS e Hess 1910} Coccinella Nernst glower septempunc- prismatic tata spectrum (adults) Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow 480 Ss. O. MAST TABLE 1—Continued INVESTIGATOR |DATE ORGANISM Home and ich- neumon flies (adults) Podapsis slab- beri (adults) Atylus swam- merdami (adults) Mussels (sev- eral species) Crayfish 1913) Calliphora erythroceph- ala (larvae) METHODS Nernst glower prismatic spectrum Nernst glower prismatic spectrum Nernst glower prismatic spectrum Nernst glower prismatic spectrum Nernst glower prismatic spectrum Nernst glower prismatic spectrum Nernst glower prismatic spectrum Spectral colors equal in en- ergy B-V = 430-490 uu; Y- G = 524-576 pp; R = 625- 665 pp Four — spectral colors equal in energy* RESULTS Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation; order of effect G, Y, B, R. Maximum in green near yellow Rate of movement or re- gion of aggregation: order of effect G, Y, B, R. Maximum in green near yellow Movement of siphon (Y- G); By Ok Reaction of pupil (Y-G), BS VitdEY Maximum activity in (Y- G) Pigment migration in eye about equal in (B-V) and (Y-G). Much less in R Order of efficiency begin- ning with maximum: (Gj 18, Wee dae SPECTRAL COLOR AND STIMULATION TABLE 1—Continued 481 INVESTIGATOR |DATE | ———_—_ | ———_ |] — — |S | Of | ee) es | ad es | ———— | ——— ee | 7 Gross Gross 1913 Gross 1913 Gross 1913 Gross 1913 Gross 1913 Frisch and |1913 Kupel- wieser Ewald 1914 Loeb and 1916 Wasteneys Loeb and 1916 Wasteneys Loeb and 1916 Wasteneys ORGANISM 1913| Zeuzera pyri- na (larvae) Feltia subgo- thica (larvae) Feltia subgo- thica (adults) Calliphora erythroceph- ala (adults) Drosophila ampelophila (adults) Periplaneta americana (adults) Daphnia Eudendrium Balanus (larvae) Arenicola (larvae) METHODS Four _ spectral colors equal in energy* Four spectral colors equal in energy* Four — spectral colors equal in energy* Four — spectral colors equal in energy* Four — spectral colors equal in energy* Four _ spectral colors equal in energy* Colored media and spectrum Carbon-are spectrum Carbon-are spectrum Carbon-are spectrum RESULTS GeBay, Rk GeEy VY, K BAG. YR BuGey, R i335 (Ge gues Equally strongly negative in G and Y. Positive in B, no response in R Positive orientation in R, Y, G. Negative ori- entation in (B-G), B, V Maximum effect on posi- tive orientation in red (650-660 uu) ; on orienta- tion (410-420 uu) Maximum effect on orien- tation in blue (460-480 a) In yellow (560-578 uu) In green about 495 up *Blue (420-480 yy), Green (490-550 pu), Yellow (570-620 uu), Red (630-655 pp). 482 S. O. MAST which of these large regions is most effective. Concerning relative efficiency all that can be ascertained is that all waves longer than those in the region of maximum effect are less efficient than those in this region, for they contain relatively more energy. Regarding the efficiency of the shorter waves which contain less energy than those in the region of maximum effect no definite statement can be made. Only three of the investigators, Blaauw, Day and Gross, at- tempted to ascertain the relative stimulating efficiency of the different wave-lengths. Day and Gross used spectra! colors equal in energy but they tested only four different colors. Blaauw tested more colors but I am not certain as to the method used, having access only to an abstract of his paper. 2. For seedlings of green plants, plumules and radicles, the region in the spectrum of maximum stimulating effect is in the blue or violet. For the fungi it is somewhat nearer the red. For Bactertum photometricum it is in the infra-red and the orange. For Oscillaria and Paramecium bursaria is it question- able, activity and aggregation being probably determined by chemical changes in the solution associated with the colors. For Chlamydomonas it is in the green; for all other unicellular forms tested it is in the blue, as it is also for the ccelenterates and vermes and for a few of the molluscs and arthropods. But for most of the molluses and arthropods it appears to be in the green or yellow. In none of the organisms mentioned in the table are the reac- tions specifically associated with the wave-lengths; they are not entirely independent of intensity. If, e.g., the green were made relatively sufficiently intense, the region of maximum effect could be changed from the blue to the green, etc. In the bees, how- ever, and in many of the vertebrates, the evidence obtained indi- cates that the reactions may be independent of the relative in- tensity. Bees, e.g., can be trained to select any given color, regardless of its intensity in relation to that of other colors; and such reactions to colors are the only ones which are like the reactions associated with color-vision in man. SPECTRAL COLOR AND STIMULATION 483 I have for some time held the opinion that the study of reac- tions to colors, aside from its importance in comparative psy- chology and physiology, ought to yield results which will throw light on the nature of those chemical changes in the organisms, which are associated with the reactions to light (Mast 711, pp. 320, 363). But for this purpose it is necessary to ascertain more in detail the relation between the wave-length and stimu- lation than it has been ascertained in previous work. It is highly essential to use monochromatic light of such a nature that it can be measured directly or indirectly in terms suitable for comparative work, preferably in terms of energy, so as to be able to give the relative stimulating efficiency of the different wave- lengths. It is also highly essential to ascertain the stimulat- ing efficiency for all regions in the spectrum that are at all effective, not merely for those that are most effective. These ideas have been the guiding principles in the following experi- ments. METHODS AND MATERIAL The methods used in the following experiments are based upon those used in observations on orientation of organisms in a field of light consisting of two beams crossing at right angles (07, pp. 132-134; 711, pp. 86-89). In these observations it was found that among the organisms which orient all of those without image forming eyes proceed toward or from a point situated between the two beams and that the location of this point depends upon the relative effectiveness of these two beams. If the illumina- tion in the two beams is the same in quality and quantity, so that the stimulating effect is the same, then the point lies half way between them, provided the organisms in a single beam travel parallel with the rays and do not deflect to the right or the left. This being true, it is obvious that whenever such organisms, exposed to light from two sources, proceed toward or from a point midway between them it may be concluded that the light re- ceived from the two sources is equal in stimulating effect, no matter how much it may differ either in quantity or in quality. 484 Ss. O. MAST Consequently, it is evident that we have here a method by means of which the stimulating effect of ight differing in wave-length can be compared and the relative stimulating effect ascertained. And if this is known it is a simple matter to calculate the rela- tive stimulating efficiency provided the relative energy of the different wave-lengths compared is known. The simplest method of procedure is to keep the quality of the light (white, e.g.) in one beam constant while that in the other is changed and then to adjust the illumination from the white light in the one beam for each change of wave-length in the other until, in each case, the organisms proceed along the same path. The stimulating effect of the different wave-lengths tested will then be directly proportional to the various illuminations from the white light required to make the organisms, under each of the different conditions, proceed in the same direction. For ex- ample, if for the green it requires twice as much light from the white source to make the creatures take a given course as it does for the yellow, then the stimulating effect of the green is twice as great as that of the yellow, and if the yellow has twice as much energy as the green then the stimulating efficiency of the green is four times as great as that of the yellow. This method is ap- plicable to organisms which, in light from a single source, do not tend to travel parallel with the rays, but tend to deflect to the right or the left, e.g., Volvox, as well as to those which do not tend to deflect, provided the extent of the deflection is equal under the various conditions. In the experiments described in the following pages, two 80- watt, gas-filled, street series, tungsten lamps with vertical coiled filaments were connected in series with a 40-volt storage battery and enclosed in light-proof boxes each containing an opening of such a size and so situated as to produce a small horizontal beam of light. One of these beams passed through a Hilger constant deviation spectrometer and a lens which focused the filament on the slit, the other passed through a Lummer-Brod- hun rotating sector. The whole apparatus was so arranged that the two beams of light crossed at right angles in the field of observation as represented in figure 1. SPECTRAL COLOR AND STIMULATION 485 By means of the higher spectrometer the organisms could be subjected in the one beam to any spectral color and the change from one to another could be rapidly made without in any way altering the size or position of the beam of light. Moreover, the wave-length of the light in the beam could be readily ascertained by direct readings on the revolving drum of the spectrometer. Fig. 1 Outline of apparatus used in the following experiments: ¢ and 2’, tungsten street series lamps; 0 and o’, opaque boxes; s, Hilger spectrometer; r, Lummer-Brodhun rotating sector; 6, beams of light, 10 mm. wide, 3 mm. deep; s’, opaque screens; g, glass plate; J, line on glass plate; a, aquarium. By means of the Lummer-Brodhun sector the illumination in the other beam could be instantly changed from zero to the maxi- mum available and the illumination used at any time could be directly ascertained by readings on the sector. All of these features and the constant current from the storage battery were found to be of the greatest importance in the work undertaken. In fact, it would have been quite impossible to have obtained some of the most important results without them. 486 Ss. O. MAST The two slits in the spectrometer were both, unless otherwise stated, 0.75 mm. wide, insuring a fair degree of purity in the spectral colors used. All of the observations on the microscopic forms were made in a small rectangular aquarium under a binocular supported by an extension arm. The aquarium, which was 26 mm. wide, 26 mm. long and 9 mm. deep, was made of the best quality of colorless glass slides accurately cut and ground and glued with Khotinsky cement. An opaque screen containing an opening 3 x 10 mm. was fastened to each of the two adjoining sides of the aquarium in such a position that the long axis of the opening was parallel with the bottom, the lower edge of it just a trifle lower than the upper surface of the bottom and the ends of it equal distance from the ends of the aquarium. Thus each of the two horizontal beams of light crossing each other at right angles in the aquarium was 10 mm. wide. The aqua- rium was supported by a glass plate containing ledges so arranged that the aquarium could be easily removed and returned to pre- cisely the same position. The glass plate contained a line which passed under the center of the aquarium and bisected the angle between the two beams of light. Some distance under the center of the glass plate there was a 2 c. p. ruby electric bulb and above this a plate of ground glass and a petri dish containing water for the purpose of diffusing the light and reducing the heat (fig. 2). In some of the work the dish was replaced by a few ad- ditional glass plates. The light from below made it possible to see the organisms and the line on the glass plate clearly, but being red it did not appreciably influence their direction of movement. For further information regarding apparatus used see page 517. In making the observations, a considerable number of the or- ganisms was put into the aquarium in clear pond or tap water about 4mm. deep. The aquarium was then put in place and the illumination in the beam of white light adjusted by altering the size of the opening in the Lummer-Brodhun sector until the organisms travelled parallel with the line under the aquarium on the glass plate. This adjustment was made in darkness, so as to eliminate the personal factor as a possible error. After the a SPECTRAL COLOR AND STIMULATION 487 reading indicating the extent of the opening of the sector was recorded, the wave-length was changed, the water in the aqua- rium thoroughly stirred up so as to distribute the organisms equally throughout unless otherwise stated, and the illumination in the white beam again adjusted until they proceeded parallel with the line under the aquarium. This was repeated for all effective regions of the spectrum usually differing by 10 wun. Thus the readings indicating the extent of the opening of the sec- Aon eg Fig. 2 Side view of observation aquarium, stand, etc. a, aquarium; §, stand; b, opaque box, 8 x 10.7 x 10.7 em.; 1, ruby light; m, mirror; g, glass plate; g’, ground glass plate; p, petri dish. The binocular was supported by an ex- tension arm vertically over the aquarium. tor, when the course of the organisms bisected the angle between the two beams of light, were ascertained for the different re- gions; and since these readings directly express the relative illumination, they express also, as previously demonstrated, the relative stimulating effect of the regions tested; and from these data the relative stimulating efficiency of the different wave- lengths with reference to the energy can readily be calculated provided the relative distribution of energy in the spectrum THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, NO. 3 488 S. O. MAST - used is known.? It was consequently not necessary to make meas- urements of the absolute illumination used, and this constitutes one of the most advantageous features of the method em- ployed, for it simplifies greatly the work involved. We shall, however, give sufficient data in the following paragraphs and in connection with the description of the individual experiments to make it possible to reduce the readings as presented in the fol- lowing pages to terms of absolute intensity in case anyone should desire to do so. In all but a few preliminary experiments, which were made before August 1 the current used was 5.75 amperes with only very sight variation. With this current and with the Lummer- Brodhun sector set at 100, the candle power in the white beam was 40.6. In all of the experiments except those on earth- worms and blowfly larvae the source of light in the white beam was 50 em. from the center of the observation aquarium. The illumination in this beam with the sector set at 100 was therefore approximately 162.4 meter-candles at the center of the field of observation. It was actually somewhat less owing to reflection and absorption. To calculate the illumination with the sector set at any other point it is only necessary to read the number on the scale opposite the point in question, divide it by 100 and multiply the quotient by 162.4, 1.e., by the illumination in meter- candles with the sector at 100. In this way the illumination can readily be calculated for any of the sector readings in the following tables except those which refer to earthworm; and blowfly larvae. In the experiments on these two forms the distance between the center of the field of observation and the source of light in the white beam was 100 ecm. and the light was also reduced by screens so that with the sector set at 100 the candle-power was in some of the experiments 21.4, in others only 7.4. These numbers must therefore be used’ in place of the — preceding in calculating the illumination. In these experi- ments the distance .between the ocular slit of the spectrometer and the center of the field of observation was 53 em. and both beams of light were 6 cm. wide (see p. 517). 2See figures 3 and 4, and tables 2 to 15. SPECTRAL COLOR AND STIMULATION 489 The relative distribution of energy in the spectrum used in the following experiments was ascertained by Dr. W. E. Forsythe and Mr. Francis E. Cady, members of the staff of the Nela Re- search Laboratory. Mr. Cady also measured the light used. I am greatly indebted to these gentlemen for their generous assistance. At the end of the experiments the lamp used was matched in color with one of the laboratory standard lamps whose distribu- tion was known in terms of a standard radiator. A new lamp precisely like the one used in the experiment was then similarly matched and the results agreed so closely that it was evident that there was no appreciable change in the lamp during the time it was used in my experiments and that the distribution of energy in the spectrum remained practically constant. The results of the above mentioned tests are plotted in figure 3. By referring to the accompanying curve it will be seen that beginning at the violet end of the spectrum and proceeding toward the red end the energy first increased rather gradually but later very rapidly, so that while there was, in the green at 560 pu, 30% times as much energy as in the violet at 400 uu, there was, in the red at 700 uu, more than 138 times as much. This great difference in energy in different regions of the spectrum shows clearly that if one should obtain a response in the longer waves and none in the shorter, it would not necessarily prove that the stimulating efficiency of the former is greater than that of the latter, for the difference in the response might be due solely to the difference in the amount of energy involved. To ascertain the relation between wave-length and stimulating efficiency, it is consequently necessary to make corrections for the unequal distribution of energy in the spectrum. Such correc- tions were made in all of the experiments described in the follow- ing pages. Most of the organisms investigated were collected in temporary clay pools formed in the immediate vicinity of the laboratory at Nela Park, owing to the unprecedented abundance of rainfall during the season. In these pools, many of which were so small that they contained only a few gallons of water, unicellular and 490 S. O. MAST colonial organisms appeared in great abundance, especially the green forms, Chlamydomonas, Euglena, Pandorina and the like. In any given pool a given species ordinarily predominated at any given time, but usually there was a succession of species between successive rain storms, depending apparently upon changes in the Fig. 3 Curve representing the distribution of energy in the spectrum used in the following experiments. The circles represent the points experimentally established, the ordinates the energy and the abscissae the wave-lengths in yy. constitution of the solution. For example, in one particular pool which frequently dried up entirely Chlamydomonas appeared first each time after several rain-storms and became very abun- dant with a few scattered Pandorina colonies. Then after a few SPECTRAL COLOR AND STIMULATION 491 days Pandorina became very abundant while the other organ- isms practicaliy disappeared. A few days later Eudorina de- veloped, while Pandorina gradually disappeared. Thus by watching these pools many of the species studied were secured ‘n any desired numbers and practically pure. Among these were several of which ordinarily only a few scattered individuals are found, e.g., Euglena tripteris, Trachelomonas, Eudorina, and Gonium. This fortunate location of an abundance of excellent material close at hand greatly facilitated the work undertaken. In fact it would otherwise have been quite impossible to have so thoroughly covered such a large field in the few months at my disposal. The following fifteen species were fairly thoroughly investigated with reference to responses to colors, and a number of others were superficially studied. Trachelomonas, Chlamy- domonas, and Phacus each one species; Euglena five species, Gonium, Pandorina, Eudorina, Spondylomorum, Lumbricus, Arenicola (larvae) and blowfly (larvae) each one species. Details concerning these studies and the results obtained are presented in the following pages. EXPERIMENTAL OBSERVATIONS Introduction Success or fajlure in attempting to ascertain the relation be- tween wave-length and stimulation with the method previously described depends largely upon the condition of the organisms used. The more precisely they orient, the more strongly posi- tive or negative and the more active they are, the more accurate the results will be. It is consequently essential, especially for those who may wish to repeat these experiments, to know as much as possible about the treatment the organisms received preced- ing the tests, their habits and habitats and the conditions under which they are likely to respond in a way most favorable to the work. I shall therefore in this section present, in addition to the results obtained, some of the characteristics of the responses of the organisms, their habits, and their environment which are intimately related to the observations made. 492 S. O. MAST Euglena viridis The specimens of Euglena viridis used in the following experi- ments were collected in a pig-yard in Michigan, near Ann Ar- bor, August 17, 8 a.m., taken to Cleveland and placed in a jar in TABLE 2 Euglena viridis (negative). Relation between wave-length and stimulation} RELATIVE STIMULATING EFFECT OF DIFFERENT REGIONS IN PRISMATIC RELATIVE STIMU- WAVE- SPECTRUM IN TERMS OF SECTOR READINGS LATING EFFICIENCY LENGTH IN CALCULATED ON ME BASIS OF EQUAL Results of individual tests Average ENERGY 422.4? 1.5 155 2.50 432.6 2.5 4.0 3.2 | 3.233 4.09 442.8 90. )), 1255 8.2 | 12.8 | 10.625 10.41 452.9 170} 1825 15,7 | S225) eli. 350 13.24 463.1 27.5 | 19.4] 23.0] 28.5 |) 24.0] 32.5 | 24.983 14.86 473.2 38.5.| 36.0] 40.5 | (34:8) 38.9) 38.77) 37:900 18.22 483 .4 53.5 | 50.9] 55.0] 46.3 | 46.7] 61.0 | 52.233 20.40 493.6 Al.2)|- 35.5: |) 5104). 4%e3)| 42.1. |) Sie Ones S35 14.46 503 .7 23.3.\ 2723 | 30.2.) 32:4 | 34.651" 35.57) 30-550 8.23 513.8 16.) 1459 9.9| 16.7] 14.6 | 13.540 3.06 524.0 7.0 5.0 af 5.0 | 5.666 1.08 : August 18 August 19 ae { 5.10-6.02 p.m. 10.43-11.37 a.m. 1 The figures in the columns 2-7 express relatively the stimulating effect of the different wave-lengths tested. The larger the number the greater the effect. Thus in column 3 the numbers show that the effect of wave-length 432.6 uu was very slight and that from this point toward the red end of the spectrum the effect gradually increased until it reached a maximum near 483.4 wy, after which it gradually decreased until it became practically zero at 524 wz. The blank space in the columns indicate either that no readings were made or that the stimulating effect was so small that it could not be successfully measured by the method employed (see pp. 483-487). The numbers in the last column indicate the distribution of stimulating effect in a spectrum having a uniform distribution of energy throughout, i.e., they express the relative stimulating efficiency of the different wave-lengths. These numbers were obtained from those in the preceding column by applying the corrections for unequal distribution of energy indicated in the curve in figure 3. 2 The readings were presumably all made 10 wy» apart as the drum of the spec- trometer indicated, i.e., at 420, 430, 440 ny, etc. It was, however, discovered by Dr. W. E. Forsythe that this series of readings on the drum, owing to imperfect adjustment, produced the wave-lengths given in the first column of this and the following tables. SPECTRAL COLOR AND STIMULATION 493 rather weak diffuse daylight. The following day some speci- mens, together with about 2 cc. of clear solution, were taken from this jar, put into the observation aquarium and tested in the spectrum. They were found to orient fairly accurately and to be strongly negative. : 9°SF9 91'0 aT) 9¢°0 |60°0 |61°0 @°¢79 | +g £9'0 |2z'0 2c19( & 9&°T |09°0 |0F'0 0°09 78's lec°T |T¢°0 6 F6S 26'S |68°S |cO'T pa) K< 66°2 |29°9 |2°T QFLg > & Il O1lcr'6 \er'e ¢79g] 2 C8 FIGS FIITL’S P PSS 8% STIL9' LIITF'S @ PFS ZL OZITS ST\Ss" EI 66'2 |62'0 |89°0 Tres] o PE SIIOT OZ|TZ 9I/10'S JOC ITL2@ Seo GOL 92 TP (08-0 GT © |1¢°0 00°T |80°T lo'Fz¢> cg" 2TI82ST/0¢ 6T\OL ETlFE'z |8S°9 l9z°S leg ILlor's lZF'9 | 86'T ler's \2Z'T |90'T |FO'T |(2)¥FE°0 let’e [90'¢ Iger¢| = 6 9TIGT ST/98' O/88 ETIFS'Z |ZE ITIZE STlps SI\IZ' OL\6e'sI| Zr's |2r'z |00°9 |I8'F [69°F | seh l9F'6 \Ez'8 |Z EOS 60’ OT|TS ZI/6E 61/96’ LT|€Z' STIFE LTGP 6LIIF STIF9' 216002] 2h OTITIS STOO ZT\ze S1\ZF S| TO ZIl0e ST\9F FIO’ SEF = wd F0'S |ST'6 |16 STI€8 FLIPT 0%|FS OZ/OS LT|FS OZ\EO OZIIF 6I| L0 OB|ZF’ 6T|Z0'0Z|00 OZ/OT 02 aCe aa 89°2 |98°2 |ZT'9T ZL°6 |€9' OTIOT OBITS 6I\6S STITT 91] ZO'6TISO OBIFL 6IIST ST\09'6T} 60° OZIOE ST\Zz'STIZ SzF 86'9 |68'F |9F'6 9Z°9 |10°FT|09 9T/98°ST|TS ST/96 ST} GF OT/6F GII9L FIISS LTS 21] EF 9T/99°9TIOS FIIT gor | & 96° |Z¢°¢ log" FS 1106 IT\692T/G¢'S |2z' FI] TZ ST|O9'ST\Z8' ZIPS" ST|90 ‘OT sce eae | 5 ZET_ aCinles 6F'¢ C6 E&I I6'IT| 22 PEL ISL°CLl0L'2 | &F'9 |69°2 \IF OLS ct Gorz Le Coarse Le |9eee 62°F \60'F [9 'zer] oat les r9'T 08°% IP Zar SBS hgh | soe tone licteeee licmaiy lcteoer erage: (cepshee li aaoue mane Root NI omaiceiesancel ads eyso pa OS Te mSNC | caec, a emene wo Q lel la] ; w - a ie a 1 > adh} hanay uri wnuixnyy ‘pawodwoo fizvppas aq wna fiay, 7Y7 Os pabo.van pun apnjubpu aus ay) 0) paonpas‘saroads uaaify sof wn.ijoads oy) U2 suorbas juasiaffip fo houarja Buyvjnwuays aarnjay Cl WIAVL 522 Ss. O. MAST The results taken as a whole consequently support the con- tention that the relation between wave-length and stimulation in the blowfly larvae is more nearly like that in Chlamydomonas than like that in any of the other organisms studied but that the longer waves are somewhat more effective in the former than in the latter (table 15). Gross (’13) made a very thorough study of the blowfly larva with reference to the relative stimulating efficiency of four differ- ent spectral colors, blue (420-480 yu), green (490-550 yu), yel- low (570-620 uu) and red (630-650 uy). In this study he forced the larvae to enter the side of a field of light consisting of two horizontal beams with rays opposite in direction and different in color but equal in energy. Under these conditions the beam from which they turn is, of course, the more efficient of the two in which they are exposed. By thus testing successively various combinations, Gross found the order of efficiency in the four colors tested to be green, blue, yellow, red. These results are in full harmony with those presented above, as can be clearly seen by referring to table 15 and figure 4. * DISCUSSION The results obtained show clearly that the reactions in all of the species studied are dependent upon wave-length, certain colors are much more efficient as stimulating agents than others, but they are not wholly dependent upon wave-length, for while there is clearly a region of maximum stimulating efficiency in the spec- trum, stimulation is not confined to this region and the stimu- lating effect of the wave-lengths on either side of it can be made greater by simply increasing their intensity. There is conse- quently no evidence in the results obtained indicating the pres- ence of color-vision in any of the forms studied, for it is the absence of any such relation between reactions and intensity . that constitutes the chief objective characteristic of color- vision. Moreover, the fact that there is no variation in the distribution of stimulating efficiency in the spectrum dependent upon differ- ent physiological states, it being the same in specimens, e. g., SPECTRAL COLOR AND STIMULATION 523 either in the negative or in the positive state, is not what would be expected in organisms with ecolor-vision. The relation be- tween color and stimulation is of the same order in all of the fif- teen species studied as it is in color-blind human beings, and in one of the former (blowfly larvae) it is in fairly close agreement with that of the latter. The lowest forms in which color-vision has been clearly es- tablished are, in my opinion, honey bees. Both Lubbock (’95) and Frisch (’14) have shown fairly conclusively that these in- sects can be trained, in gathering honey, to select any one of a considerable number of different colors regardless of their lu- minous intensity. The conclusion of Frisch and Kupelwieser (18, p. 552) that Daphnia has a sense of color (Farbensinn) is in my opinion not well founded. ‘These authors demonstrated that under certain conditions Daphnia is positive in red, yellow and green, and negative in blue-green, blue and violet, i.e., that change from any of the former to any of the latter has the same effect as an increase in luminous intensity; and that a change in the opposite direction has the same effect as a decrease in lumi- nous intensity. This seems to indicate merely that these two groups of colors are antagonistic in their effect, just as has been found to be true for similar groups in their effect on certain photochemical reactions and on the reactions of some other organisms (Mast, ’11, pp. 310, 335). Loeb and Wasteneys (’16) conclude that the relation between wave-length and reactions is the same in animals as it is in plants and that there are two different ‘‘types of photosensitive sub- stances” in both. Neither of these conclusions is well founded. The first conclusion is based largely upon the fact that the rela- tion mentioned above is not the same for all plants, just as it is not the same for all animals. But it would be quite as logical to conclude the opposite, for in some plants the shorter wave- - lengths are relatively more efficient as stimulating agents than they are in any animals (see table 15), and in none of the plants is the entire visible spectrum effective, nor are there any in which the reactions depend upon the wave-length to such an extent as they do in those animals which have color-vision. 524 S. O. MAST The second conclusion is based upon the contention that the region in the spectrum of maximum stimulating efficiency is either in the blue or in the green for all organisms. This is prob- ably not true, but even if it were true it would be just as logical to conclude, on the basis of the results obtained by these au- thors, that there are three or four types of photosensitive sub- stances, for they maintain that the maximum is respectively at 460-490, 495, 535, 560-578 wu for Euglena, Arenicola, Chlamy- domonas and Balanus. Our results and those of others also indi- cate that the maximum is located in several different regions and that some of these do not correspond with any of those given; consequently if the argument of Loeb and Wasteneys is valid there are at least four ‘‘types of photochemical substances.” This argument appears to be valid, however, only if the relative absorption of the different colors is the same in those organisms in which the location of the maximum differs. All that can be said, then, regarding the chemical reactions associated with reactions to colors is that they differ in all organ- isms for which the distribution of stimulating efficiency in the spectrum differs, provided there is no difference in selective ab- sorbtion. It can not be said, however, that the chemical re- actions are the same in all organisms in which the spectral distribution of stimulating efficiency is the same, for, judging from our present knowledge of photochemistry, which is ad- mittedly very inadequate, the relation between photochemical reaction and wave-length is in many instances the same for different substances. The fact that the relation between wave- length and reaction is the same for individuals when they are negative as it is when they are positive supports this con- tention, for the chemical reactions associated with these differ- ent responses are, in all probability, not the same. SUMMARY 1. The distribution in the spectrum of the stimulating effect and the stimulating efficiency in reference to energy was ascer- tained for fifteen species, including.unicellular and colonial forms, worms and fly larvae. SPECTRAL COLOR AND STIMULATION 525 2. In all of the species studied stimulation by light was found to depend upon the wave-length, 1.e., certain spectral colors are much more efficient as stimulating agents than others. In the spectrum from the region of the maximum the stimulating effi- ciency decreases rather rapidly in either direction and the effec- tive region is much shorter than the visible spectrum. 3. There is no evidence indicating that stimulation in any of the species studied is independent of luminous intensity, for if the light in the spectrum on either side of the maximum be made sufficiently intense it becomes more effective than that at the maximum. This holds also for the reaction of plants to light and probably for all photochemical reactions. There is conse- quently no evidence indicating the presence of color-vision in any of the forms studied. Bees are the lowest form in which color-vision has been clearly established. 4. The distribution in the spectrum of stimulating efficiency ’ differs in some species that are closely related (Gonium and Pan- dorina) and is essentially the same in some that are not closely related (Euglena and Lumbricus). The region of maximum efficiency is near 483 yu for Euglena, Trachelomonas, Phacus, Gonium, Arenicola, and Lumbricus; near 524 yy for Pando- rina, Eudorina and Spondylomorum; near 5038 uy» for Chlamy- domonas and blowfly larvae; near 465 uu for green plants and somewhat nearer the red for fungi. 5. It consequently differs for plants as well as for animals, but the shorter wave-lengths are relatively more efficient for green plants than they are for any animals and there is nothing in the nature of color-vision in. any of the plants. The conten- tion therefore that the reactions to colors in plants and animals is the same is not well founded, although some of the chromatic reactions in animals may be essentially the same as those in plants. 6. The distribution in the spectrum of stimulating efficiency in any given species is continuously the same, regardless of changes in physiological states, environment and character of response, e.g., 1t is the same in individuals when they are negative as it is when they are positive. 526 S. O. MAST 7. As to the nature of the chemical processes associated with the responses no definite conclusions can be drawn; but if the absorption is the same in the photosensitive tissues of all of the organisms studied these processes differ in all species in which the relation between wave-length and stimulation differs. It can not, however, be concluded, even if this is true, that they are the same in all of the species in which this relation is the same, for the relation between chemical reactions and wave- length is, in all probability, the same for certain substances that differ. Further progress in this analysis depends largely upon further work on the relation between photochemical reactions and wave-length. BIBLIOGRAPHY Bancrort, F. W. 1913 Heliotropism, differential sensibility and galvano- tropism in Euglena. Jour. Exp. Zoél., vol. 15, pp. 383-428. Bert, P. 1869 Sur la question de savoir si tous les animaux voient les mémes rayons que nous. Arch. de Physiol., t. 2, p. 547. Biaauw, A. H. 1908 The intensity of light and the length of illumination in the phototropic curvature in seedlings of Avena sativa (oats). Kon. Ak. Wet. Amsterdam, Proc. Meeth., Sept. 26, 1908. Review in Bot. Cent., 1909, vol. 110, p. 655. Coun, F. 1865 Uber die Gesetze der Bewegung mikroscopischer Thiere und Pflanzen unter Einfluss des Lichtes. Jahresber. d. Schles. Ges. f. Vaterl. Cult., vol. 42, pp. 35-36. Darwin, C. 1881 The formation of vegetable mold through the action of - worms, with observations on their habits. New York. 326 pp. Day, E. C. 1911 The effect of colored light on pigment migration in the eye of the crayfish. Bull. Mus. Comp. Zool., vol. 53, pp. 303-343. DutTrocHET, M. anp PourLLeET 1844 Ann des. sc. nat., 3 ser., t. 2, pp. 96-113. ENGELMANN, T..W. 1882 Uber Licht- und Farbenperception niederster Or- ganismen. Arch. f. d. ges. Physiol., Bd. 29, pp. 387-400. Ewap, W. F. 1914 Versuche zur Analyse der Licht- und Farbenreaktionen eines Wirbellosen (Daphnia pulex). Zeitschr. f. Sinnesphysiol., Bd. 48, pp. 285-824. Friscu, K. v. 1914 Der Farbensinn und Formsinn der Biene. Zool. Jahrb. abt. Phys., Bd. 35, S. 1-182. Frisco AND Kupretwieser 1913 Uber den Einfluss der Lichtfarbe auf die phototaktischen Reaktionen niederer Krebse. Biol. Centralbl., Bd. 33, S. 517-552. GarpNer, D. P. 1844 Surl’action de la lumiére jaune dans la production de la couleur verte des plantes et sur celle de la lumiére indigo dans la pro- duction de leurs mouvements. Bibliothéque universelle de Genéve, Feor. 1844. SPECTRAL COLOR AND STIMULATION 527 GraBER, V. 1883 Fundamentalversuche iiber die Helligkeits- und Farben- empfindlichkeit augenloser und geblendeter Thiere. Sb. d. Akad. Wiss., Wien, Bd. 87, pp. 201-236. Gross, A. O. 1913 The reactions of arthropods to monochromatic lights of equal intensities. Jour. Exp. Zodl., vol. 14, pp. 467-514. GUILLEMIN 1858 Production de la chlorophyll et direction des tiges sous Vinfluence des rayons ultraviolette, calorifiques et lumineux du spectra solaire. Ann. des sc. nat., 4 ser., t. 7, pp. 154-172. Harrineton, N. R. anp Leamine, E. 1900 The reactions of Amoeba to light of different colors. Amer. Jour. Physiol., vol. 3, pp. 9-16. Hess, C. 1910 Neue Untersuchungen iiber den Lichtsinn bei wirbellosen Tieren. Arch. f. d. ges. Physiol., Bd. 136, pp. 282-367. Kraus, G. 1876 Versuche mit Pflanzen im farbigen Licht. Ber. Sitzungs d. Naturf. Ges. Halle. Jahre 1876, pp. 4-8. Loss, J. 1890 Der Heliotropismus der Thiere und seine Ubereinstimmung mit dem Heliotropismus der Pflanzen. Wiirzburg, 118 pp. 1906 The dynamics of living matter. New York, 233 pp. Lozs, J. AND Maxwe tt, 8.8. 1910 Further proof of the identity of heliotro- pism in animals and plants. Univ. of Cal. pub. in Physiol., vol. 3, pp. 195-197. , Lors, J. AnD WasTENEYS, H. 1915 The relative efficiency of various parts of the spectrum for the heliotropic reactions of animals and plants. Jour. Exp. Zoél., vol. 19, pp. 23-35. 1916 The relative efficiency of various parts of the spectrum for the heliotropic reactions of animals and plants. Jour. Exp. Zodl., vol. 20, pp. 217-236. Lussock, Sir J. 1881 On the sense of color among some of the lower animals. Part I. Jour. Linn. Soe. (Zool.), vol. 16, pp. 121-127. 1895 Ants, bees and wasps. New York. 448pp. Preface to original edition dated 1881. Mast, 8. O. 1907 Light reactions in lower organisms. II. Volvox. Jour. Comp. Neur. and Psych., vol. 17, pp. 99-180. 1911 Light and the behavior of organisms. New York. 410 pp. Mier, N. J. C. 1872 Uber die Kriimmung der Pflanzen gegen das Sonnen- licht. Bot. Ztg., vol. 30,.p. 446. Payer, J. 1842 Memoire sur la tendance des tiges vers la lumiére. Compt. rend., t. 15, pp. 1194-1196. Pogatour, 8. 1817 Opuscoli Scientifici. Bologna. Preyer, W. 1886 Uber die Bewegungen der Seesterne. Mit. a. d. zool. Sta. z. Neapel, Bd. 7, pp. 27-127, 191-233. Romanes, G. J. 1883 Animal intelligence. New York. 520 pp. Sacus, J. v. 1864 Wirkungen farbigen Lichtes auf Pflanzen. Bot. Ztg., Bd. 22, pp. 353-358; 361-367; 369-372. (Also in his Gesammelte Abh. uber Pflanzenphysiologie, pp. 261-292.) STRASBURGER, E. 1878 Wirkung des Lichtes und der Wirme auf Schwarm- sporen. Jena. Zeitschr., N. F., Bd. 12, pp. 551-625. Torrey, H. B. 1913 Trials and tropisms. Science, N. S., vol. 37, pp. 873- 876. 528 Ss. O. MAST Vrerworn, M. 1889 Psycho-physiologische Protisten studien. Jena: Fischer. 218 pp. Wiesner, J. 1879 Die heliotropische Erscheinungen im Pflanzenreiche. Eine physiologische Monographie. Theil I. Denkschr. Wien Akad., Bd. 39, pp. 143-209. 1881 Die heliotropischen Erscheinungen u. s. w., Theil II. Jbid., Bd. 43, pp. 1-92. Wiuson, E. B. 1891 The heliotropism of Hydra. Amer. Nat., vol. 25, pp. 413-433. Yerxes, R. M. 1899 Reactions of Entomostraca to stimulation by light. Amer. Jour. Physiol., vol. 3, pp. 157-182. ’ THE EFFECTS OF A THYROID DIET UPON PARAMAECIUM WALDO SHUMWAY From the Department of Biology, Amherst College TWELVE FIGURES CONTENTS Inntroductionaa esse Rat, TT ee NS enc Se ch Ree eRe or 529 IWIGUeLnaYC Es Gene cet ere oie RR ae DA te 6 AS Reet OD RE cr kh Es MI ge Doll IPR LOR VA Om UNC AT ACES. tfaxc cite ie aes APt ae» 20's 3s a Ee ote teats vote 532 HeedingrexperiMents)..5::.ck cea sae ee eee woe RE is\c'o'n Bika oak ee ath 536 ae OMEGHEE ArT ACO: <5. 3< 5 IRN arene ons coos, v= 6 cL ERR Pater, ror ae 536 eeOMP Gey EGEACE (2.5 tah RERUN oo, «se ds ee aS hal conte 541 oy (Osi) (OR Te) a hs. Aen ee > |” Se ann 547 TNE Ghaiexcircobe el oVS)r ob acco) Co lenge eb ee or eek |S Be a 549 a» On therdityision rate. -....94. 4: CES tr WN ns 0 SL eae 549 bz Onvthereastric.-vacuolese rem cctatssc ce Nis ko a.o > «carota nee alana ced 554 ce. On the contractile vacuoles........ Be OE SEE BOIS oiccio.5 Cote poo ree 555 d. On the non-contractile excretory vacuoles.....................000005 557 Ere ONUMOMSEROSIGLES nc: ka ae eer es Ane ely oo «Se AOE | seh ela ice 560 feee COM Cr mime Pies.) ho kent ernie: co ere teen 22 | ne SEE 8 561 DUM May aAN Ce CONCIUSIOIS!,; . 2 ese aetce aad «coo Sse ora kk ee ee oe 561 Literature cited......... Se aps A RRR AOE PS Naito sie 2 SPM RRs ch susie irate gt 562 INTRODUCTION The classical investigations of Maupas (’88—’89) upon the life cycle of Paramaecium, with the subsequent researches of Calkins and of Woodruff, have established the value of the division-rate as an index to the vitality of pedigreed races. Calkins (’15) con- cludes, ‘‘that more or less definite cycles of vigor or depression, ending in natural death unless conjugation or its equivalent su-. pervenes, are characteristic of all pedigreed races of infusoria.”’ It has been shown by many investigators that this life cycle, as indicated by the division-rate is readily influenced by environ- mental changes, i.e., in the temperature, amount or kind of food present, or chemical composition of the culture medium. 529 530 WALDO SHUMWAY Certain substances which produce decreases in the rate of di- vision have been carefully studied by Woodruff (summary and literature list, 1912). While on the other hand several sub- stances have been reported to produce the opposite effect, less evidence has been adduced either as to the effect or its nature. In 1912 I commenced feeding experiments upon Paramaecium, using the substance of the thyroid gland. In 1914 I was able to show that this substance produced great increases in the division rate of a pedigreed race of Paramaecium aurelia. In that paper I reviewed experiments of Nowikoff (’08), who had arrived at somewhat similar conclusions through a different method of in- vestigation. Since that time Budington and Harvey (15) have reported experiments using the thyroid substance from fish, amphibia, reptiles, birds and mammals which confirm the results I obtained from mammalian tissue. Meantime I have carried my investigations further with a view to confirming my results on Paramaecium aurelia by similar ex- periments on P. caudatum; to ascertain whether the effect pro- duced by the thyroid is unique among the internally secreting glands and by what fraction of the thyroid it is produced, what effect continued thyroid feeding might have upon the life cycle, and what other effects on the activities or structures of Para- maecium might be discovered bearing on the nature of the thy- roid effect. While definite answers to all these queries have not yet been obtained, the data collected may be brought together at this time, and certain conclusions obtained. The experiments here reported have been carried on in the Zoological Laboratory of Columbia University, the Marine Bio- logical Laboratory at Woods Hole, and the Biological Labora- tory of Amherst College. For the preparation of desiccated glands I am indebted to the Research Laboratory of Organo- therapeutics of Armour and Company. From my colleagues at Amherst I have received many helpful suggestions in the prepa- ration of this report. I take this opportunity, finally, to ex- press my sincere thanks to Prof. Gary N. Calkins, at whose sug- gestion these experiments were commenced and whose critical advice has assisted their prosecution. EFFECTS OF THYROID ON PARAMAECIUM 55 f METHODS The experiments here reported have been performed on three races of Paramaecia whose life histories are given in full below. These races were in each case derived from a single individual and thereafter maintained in four files by daily isolations following the method described by Calkins (02). Each individual was placed in a solid shallow watch glass in four drops of the culture medium and at the end of twenty-four hours (or rarely a longer period) the number of Paramaecia present was obtained. and the num- ber of divisions calculated. An individual selected at random was then isolated in the same manner, using a pipette reserved for the line in question. Records were also kept of the number of deaths and monstrosities occurring. At the end of every five days the total number of divisions in the four files of each line was obtained and averaged to give the average division rate per day of the line for that period. These five day averages form the units for our discussion. The culture media employed for the control lines were in all cases hay infusions. The formulae for these will be cited under the life histories of the three races. The thyroid and other glandular media were prepared in different experiments as emul- sions of fresh glands from rabbits, rats, and cattle, and as sus- » pensions of commercial gland preparations. The following preparations of Armour and Company were employed: desic- cated thyroids U. S. P. (sheep), thymus desiccated (calf), pitui- tary body desiccated (ox), desiccated suprarenal capsules (?), pancreas desiccated (pig), spleen desiccated (calf), ovarian sub- stance desiccated (pig). Experiments were also conducted with iodothyrin (combined with sugar of milk), a commercial prepa- ration of Friedr. Bayer and Company. Formulae used are cited under each experiment. Experimental lines were in all cases derived by isolating one individual from each file of the control line, thus obtaining two lines of maximum similarity. Where several lines were required as in some experiments, the four files of the control were allowed to multiply until the required number was present in each, and a32 » WALDO SHUMWAY then isolated to establish the several lines required. All the customary precautions against contamination were observed, and every attempt made to keep the control and experimental lines under conditions exactly similar except for the kind of culture medium employed. All the watch glasses were stacked in a single large moist chamber and the number of Paramaecia in each was ascertained at the same time. The cultures were carefully examined for monstrosities and the later history of many of these followed. In cases where a single individual required a closer study than could be obtained with a binocular or the lower power of the microscope, the fol- lowing procedure was adopted. A ring of vaseline was turned on a slide, the individual in question was placed in a minute drop on a coverslip, the drop drained off with a fine capillary tube until there was barely enough left for the individual to move, the coverslip was then inverted upon the ringed slide and the individual could be studied under high power for a consider- able time and finally returned to the watchglass. After this technique was perfected, no ill-effects were produced on the Paramaecia so treated. Permanent preparations were made from time to time by fixing in Bouin’s fluid (picro-aceto-formol) and staining with Delafield’s haematoxylin according to the following formula, for which I am indebted to my friend Professor Haughwout: Delafield’s haematoxylin conc..aq. Sol. -2s jhe) - 2 1: sees 9 ce. Gia craluncetic Cis so-so ni oo ees chase Mme ece aca cr ane ee 1 ce. Distilled hwater tc. ck Ase y cote cee ee eee es tee ee re 90 ce. C@hioralchy drate! volte ts aon ents hee ee eee + gr HISTORY OF THE RACES The life history of the three races of Paramaecia which have been studied are summarized in the following paragraphs. Race A. Paramaecium aurelia. Descended from an indi- vidual isolated March 12, 1913, from a mass culture which had been maintained in the Zodlogical Laboratory at Columbia Uni- versity for several months. This race was kept in pedigreed cul- ture until December 20, 1913, when it died in a typical depres- EFFECTS OF THYROID ON PARAMAECIUM e@ 033 sion period after a life history of 420 generations. The con- trol medium used was one-half strength standard hay infusion, prepared by boiling 1 gram of hay for ten minutes in 100 cc. of tap water and allowing it to stand exposed to the air for twenty- four hours, when it was mixed with an equal quantity of boiled tap water. The life history of this race as shown in the daily division rate averaged by ten day periods is given in figure 1. Race B. Paramaecium caudatum. Descended from an indi- vidual isolated July 2, 1914, from a stock culture of the Marine Biological Laboratory. This race was maintained as a pedigreed culture until October 18, when a pair of conjugants was isolated from an epidemic of conjugation which had been induced in stock at about the 150th generation. From an individual re- sulting from the third division after the ex-conjugants had sepa- rated, or in other words, when reorganization was completed, a new line B’ was established. Line B’. This line was maintained in a pedigreed culture until January 6, 1915, when it died out in a depression period, 81 generations after conjugation and 230 (approximately) after the original isolation. An experiment with boiled thyroid was in progress at this time, and from this line after thirty days of thyroid treatment a new line B” was established. Line B”. This line was maintained similarly until May 31, 1915, when it died out in transportation from New York. It had been maintained for 159 generations after thyroid feeding was discontinued and 460 after the original isolation of the race. The control medium employed for all these lines was prepared by boiling approximately 0.5 gram of hay seed in 50 ce. of spring water, filtering and using it after cooling. Figure 2 shows the life history of this race by ten day periods. Race C. Paramaecium caudatum. On July 27, 1915, a con- jugating pair was isolated from an epidemic of conjugation in- duced in stock from the Zoélogical Laboratory of Smith College. Two divisions after separation one of the quadrants was isolated. From this individual a pedigreed culture was established and has been maintained up to the date of writing when it stands in the 172nd generation. The control medium used for this race WALDO SHUMWAY 534 “‘pooUoWMIUIOD 910M 4X0} OY} UI pozlo syUoUTIedxe OY} YOIYM 4B syurod oy} oyvVUSISep s;eIoUMU UBWIOY ‘sporled Avp uo, AG posvIGAvOI 94¥1 UOISIAIP ATIVp oSvIOAV UL DVI YW oY} Jo ArO04STY OJ] Surmoys ydery J “S1q suols012uUebG oo;¢ 00h wor ‘DOT Aon 120 yaad ony Ayn aun OW ady IO" 535 EFFECTS OF THYROID ON PARAMAECIUM UBUIOY ‘poadoUdUIUIOD 919M 4X9} 9Y} UI pozld syUsuIIIOdXe 9Y} YOIYM 9B syurod oYy oyvUSIsop s[eiowNU ‘spolied Avp uo Aq posBsoAB-a1 04¥1 UOISIAIP ATIVp ISBIOAB UI 9dvI G 94} JO AIOYSIY OJT] OY} Burmoys ydvuin Z% “Sly SuOlfO12UBLB OCO/ 3 NO. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, 536 WALDO SHUMWAY has been one-half strength standard hay infusion used after cooling. The life history as averaged for ten day periods is shown in figure 3. In this connection it may be pointed out that the last two races, B’ and C, were derived from a single reorganized indi- vidual of an endogamous conjugant and present protoplasm as homogeneous as possible. It will also be observed that these conjugations serve as logical starting points for the life histories, a point of considerable importance for experiments such as these, in view of the fact that the protoplasm of Paramaecium is in different conditions at different stages of the life history. Aug. Sept Oct Nov. ioe cwaee 4 /00 generations. Fig. 3 Graph showing life history of the C race in average daily division rate re-averaged by ten day periods. FEEDING EXPERIMENTS a. On the A race. Nine experiments on this race of Paramae- cium aurelia in which parallel lines were maintained on hay in- fusion and thyroid media are here cited (tables 1 to 9) ; a summary (table 10) is added for comparison with the results obtained with Paramaecium caudatum. The relation of the different experi- ments to the life cycle of the race is shown in figure 1, where the points at which the experiments were begun are indicated by roman numerals. Experiment A 1 was initiated by extracting two lines from the control and treating one daily with thyroid, the other with thy- mus. These gland preparations were from three sources, a, EFFECTS OF THYROID ON PARAMAECIUM 53 TABLE 1 Race A. Experiment 1. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS THYROID-CONTROL vf 1.50 2.70 0.55 8 1.05 1.95 1.10 9 1.00 2.90 1.10 10 1.55 3.60 1.90 11 1.45 2.80 1.40 12 2.20 3.15 2.25 13 0.95 2.15 14 1.75 2.60 15 1.00 2.40 Average....... 1.40 2.80 0.90 1.85 from the freshly dissected glands of ten rabbits macerated and mixed one part to two parts boiled tap water. The emulsions were kept in a refrigerator at about 10°C. and used during period 7, in the proportion 2 drops to 2 drops of the control hay infu- sion. A second source, b, was from twenty rats similarly pre- pared and used during period 8, as above and during periods 9-12 in the proportion } drop to 4 drops of hay infusion. The third source, c, was from nine rabbits prepared and used as above during periods 13-15. The thyroid and thymus emulsions were prepared similarly from the same animals and were kept and used under identical conditions. After it was evident that the thyroid line was dividing more rapidly than the control (an ef- fect not produced by the thymus) a line was extracted from the thyroid treated one and returned to control medium in which it returned to the average division rate shown by the control line. TABLE 2 Race A. Experiment 2. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS 10 2.15 3.15 1.00 11 1.20 2.80 12 2.50 3.10 Average.......... 1.65 3.10 538 WALDO SHUMWAY Experiment A 2 was commenced while the previous experi- ment was yet in progress, to discover whether other individuals of the A race were susceptible to thethyroideffect. Anindividual was selected from the A line and allowed to divide until there were twelve cells of the same generation. Four of these were continued as a control line, four treated with thyroid and four with thymus. TABLE 3 Race A. Experiment 3. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS 23 2.65 3.20 2.15 Experiment A 3, performed after a railroad journey to Woods Hole and a change to the local tap water, was continued for only one period owing to the difficulty of keeping fresh tissue, in this case from cattle glands supplied me through the kindness of Dr. Gudernatsch. TABLE 4 Race A. Experiment 4. Daily division rate averaged by five day periods PERIOD CONTROL THYROID 23 1.70 2.90 24 2.10 3.50 PAS CEA CER 6 soit ey 1.90 3.20 Experiment A 4, performed at the same time as A 3 and under the same conditions, had as control a line which had been treated with thyroid for nine periods and thereafter maintained in the control medium for eight periods. TABLE 5 Race A. Experiment 5. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYROID-CONTROL THYMUS 31 0.70 3.70 1.75 32 evil) 1.80 1.80 33 2.25 2.50 2.00 Experiment A 5 was carried on during a depression period at a time when the control line was reduced to two individuals. EFFECTS OF THYROID ON PARAMAECIUM 539 One of these was maintained in the control medium, where it continued four days without division and then slowly returned to the control rate. Another was treated with a suspension of Armour’s thyroids (prepared by shaking up 1.5 mgm. in 2 ce. boiled tap water and mixing with 2 ee. control hay infusion) where it divided twice in the ensuing twenty-four hours. Two of the resulting eells were returned to the control medium and table 5 shows the comparative rates of division of the control, thyroid and thyroid-control during the next few periods. A thymus line started during this experiment using Armour’s thymus (prepared Jike the thyroid suspension) shows the same rate of, division as the control. : TABLE 6 Race A. Experiment 6. Daily division rate averaged by five day periods PERIOD CONTROL THYROID 32 1.70 1.85 33 2.25 3.70 IAVCT AGC. essere faeyaye ses 2.00 2.55 Experiment A 6 was carried on as the race began to emerge from the depression period. Thyroid prepared as in the pre- vious experiment was used and after one period of slight effect produced a very large increase in the division rate. TABLE 7 Race A. Experiment 7. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS 34 2.50 3.10 2.50 35 2.30 3.25 2.00 36 2.40 3.25 2.95 PAVICTAT Es fo. cic eas 2.40 3.20 2.50 Experiment A 7 was performed during a time when the race was at its highest division rate, and is rendered especially sig- nificant by the fact that eight files were maintained for each line instead of four. It will be noticed in this and other experiments where the desiccated thymus was employed that the lines so 540 WALDO SHUMWAY treated did not show the decrease in division rate that is ob- servable in the earlier experiments, a decrease due to the increase in death rate produced at the same time. TABLE 8 Race A. Experiment 8. Daily division rate averaged by five day periods THYROID THYMUS 1.50 2.65 1.30 PERIOD | CONTROL 46 Experiment ‘A 8 carried on after return to New York from Woods Hole and after an interval during which observations were made at irregular intervals (one to four days) resylted in the regular increase in division rate after thyroid feeding. TABLE 9 Race A. Experiment 9. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS 54 0.65 1.00 0.80 55 0.70 1.40 Lats 56 0.95 0.70 1.05 57 0.40 0.20 0.15 AVICTAZ Cece: (ss - 0.70 0.85 0.75 Experiment A 9 was performed during the final depression period which carried off the race. The individual cells at this time gave evidence of their condition not only in lowered di- vision rate, but in decreased size, more sluggish movements and the appearance of monstrosities. Thyroid feeding produced no effects after the first two periods and the thyroid treated line succumbed more rapidly than the control. All attempts to save the race by other food media, chemicals and increasing the amount of culture media, failed. These experiments showed that the thyroid fed individuals divide more rapidly than those kept in hay infusion no matter whether the race was at a high or low point in the life cycle. The only exception observed was during the final depression period which carried off the race. They also show that thyroid indi- ' EFFECTS OF THYROID ON PARAMAECIUM 541 TABLE 10 Race A. Summary of experiments. Average division rate per day EXPERIMENT CONTROL THYROID DAYS I 1.40 2.80 35 II 1.65 3.10 15 III 2.65 3.20 5 IV 1.90 3.20 10 V 0.70 3.70 5 VI 2.00 200 10 VII 2.40 3.20 15 VIII 1.50 2.65 5 IX O70” 0.85 20 Average.......... 1.60 2.65 120 Percentage increase produced by thyroid feeding, 65 per cent. viduals were less liable to death from slight environmental changes. At all times the thyroid fed individuals were more active, smaller and more transparent. Paramaecia treated with thymus showed no such effects. In the course of these experi- ments, some thyroid particles were stained with an alcoholic solution of Congo red, the free color washed out in distilled water. These prepared thyroid particles were then fed to Paramaecia in the manner customary in these experiments. It was possible to observe the formation of gastric vacuoles containing the pre- pared thyroid and later to note the change from an acid to an alkaline reaction shown by the indicator, exactly as described by Métalnikow (’12). b. On the Brace. In the summer of 1914 attempts were made to discover whether the effects produced by thyroid feeding were due to the iodine content of the gland. It-was found that iodine and iodine in combination with potassium iodide re- sulted in depressing rather than raising the division rate even when used in such small doses as 1/20,000,000. The experi- mental data are given in table 11. In the fall of this year experiments were undertaken to dis- cover whether the effect of the thyroid is unique among the in- ternally secreting glands in producing an increased rate of di- vision. In the first experiment cited, four lines were derived from the control B’ and treated for five periods of five days 542 WALDO SHUMWAY TABLE 11 Race B. Experiments with iodine. Average daily division rates re-averaged by five day periods PERIOD CONTROL ‘ IODINE n-2000 5 1°30 0.95 6 1.30 1.10 7 1.40 1.30 AViGEA Ge sees. ch:.. hs ee 1.35 1.10 Experiments with iodine and potassium iodide. Protocol Primary solution iodine, 1, potassium iodéde, 2, water 300 Solution iodine 1—2000 Death immediately 1—20,000 Death in two or three minutes. 1—200,000 Death in 48 hours 1-2,000,000 Death in 72 hours 1—20,000,000 Decrease in division rate Average daily division rates re-averaged by five day periods PERIOD CONTROL IODINE IODINE-CONTROL 26 2.15 11 1b) 27 1.90 1.05 28 1.60 1.30 1.75 DENCE coca eae: 1.90 1.15 each in a medium of 2 drops of hay infusion plus 2 drops of the desired gland suspension (prepared by shaking up 2 mgm. of the desiccated gland substance in 2 cc. of boiled tap water). The glands used were thyroid, suprarenal, thymus, and pituitary. TABLE 12 Race B. Experiment 1. Daily division rate averaged by five day periods PERIOD CONTROL THYROID ADRENAL THYMUS PITUITARY 4 1.15 1.85 1.95 1.55 2.00 5 0.70 2.00 1.75 1.05 1.60 6 0.30 0.65 0.75 0.70 0.75 7 0.25 1.75 1.40 1.70 1.70 8 0.50 2.00 2.10 2.10 2.20 Average...... 0.60 1.65 1.60 1.40 1.65 EFFECTS OF THYROID ON PARAMAECIUM 543 The experiment was unfortunately interrupted during period 6, when I was called out of town. During this period the lines were kept in hay infusion and not isolated daily. Table 12 contains the daily division rate averages by five day periods. It will be seen from these figures that all the lines divided more rapidly than the control, with practically no differences between them. This experiment is the only one giving such a result and by consulting the life history of the B race, figure 2, it will be seen that the race is passing through the low point of a rhythm. TABLE 13 Race B’. Experiment 2. Daily division rate averaged by five day periods PERIOD CONTROL THYROID THYMUS PANCREAS SPLEEN OVARY Baie PITUITARY 10 1.55 1.65 9.50 0.35 0.30 0.70 11 1.30 1.4 0.50 0.25 0.30 0.60 12 0.80 0.90 0.20 0.55 13 0.30 0.85 0.00 Average} 1.00 1.20 0.50 0.25 0.30 0.65 0.20 0.55 In order to secure a more crucial test it was determined to keep the experimental lines in the gland suspension alone, i.e., without the addition of hay infusion. Experiments were per- formed on seven lines, using the following glands: thyroid, thy- mus, pancreas, spleen, ovary, suprarenal and pituitary bodies, with the result shown in table 13. Under these conditions the thyroid line alone survived after a few days treatment, presenting on the contrary active indi- viduals at all times and a division rate consistently a little higher than the control. In order to discover whether this effect was permanent the thyroid line was continued from this time on till January 26, 1915, when it died out in the 101st generation since conjugation, on the same day on which the control line died out. The five day averages of the daily division rate are shown in table 14. Figure 4 shows graphically the life history of the control and thyroid lines during this period. The close similarity of the two curves is worthy of mention. It will be ob- 544 WALDO SHUMWAY TABLE 14 Race B’. Experiment 3. Division rate averaged by five day periods PERIOD CONTROL THYROID 10 1.55 1.65 11 1.30 1.45 12 0.80 0.90 13 0.30 0.85 14 0.30 0.85 15 0.45 1.30 16 0.65 0.95 17 1.00 1.20 18 1.05 1.60 19 . 0.50 1.10 20 0.25 0.20 21 ‘ IAVCT AGC! rete eek cae cae 0.75 1.00 served that though the thyroid line shows a consistently higher division rate than the control it was unable to survive the de- pression period. 3 i} 10 4] 12 /3 14 (hey 16 he? 18 42 20 Fig. 4 Graph showing comparative life histories of the thyroid fed and control lines during Experiment B3 in average daily division rate re-averaged by five day periods. Control ——,, thyroid - - - -. Experiments were now begun in the hope of localizing the factor to which the thyroid effect was due. Solutions of the commercial iodothyrin were prepared in the same manner as the glandular media.’ Table 15 shows the division rates obtained from the control, thyroid, and two iodothyrin lines. It will be EFFECTS OF THYROID ON PARAMAECIUM 545 TABLE 15 Race B’. Experiments with todothyrin. Average daily division rate averaged again by five day periods PERIOD CONTROL THYROID IODOTHYRIN I IODOTHYRIN II 25 1.45 2.00 1.55 1.45 26 1.25 2.00 1.35 1.00 Average::.2.):- 1.35 2.00 . 1.45 1.25 seen that the iodothyrin produced no effect at a time when the thyroid treatment brought about an increase of 0.65 divisions per day. Other experiments were conducted with weaker solutions of the iodothyrin which gave essentially the same results. It hav- ing been suggested that boiling the thyroid would test the ques- : TABLE 16 Race B'. Experiment 4. Daily division rate averaged by five day periods PERIOD CONTROL THYROID BOILED THYROID THYROID BOUILLON 17 1.00 1.20 1.20 1.00 tion as to whether the effects produced were due to an hor- mone, experiments along this line were begun. - A 2 grain tablet of thyroid substance was boiled for an hour in 100 ce. of spring water and filtered. The residue was evaporated by slow drying. Approximately 2 mgm. of the dried substance was mixed in 2 cc. of spring water to make a culture medium in which a line of Paramaecia was maintained for five days by the usual method. Another line was similarly treated with the filtrate (thyroid bouillon). The results are shown in table 16. TABLE 17 Race B’. Experiment 5. Daily division rate averaged by five day periods J Ee eee ee eee PERIOD CONTROL THYROID BOILED THYROID 18 1.05 1.60 1.10 19 0.50 1.10 1.90 20 0.25 0.20 1.75 IAVICN AP Gree asec « 0.60 1.00 1.25 546 WALDO SHUMWAY To test this point still further another experiment was per- formed in which a line was maintained on boiled thyroid prepared by boiling the usual thyroid suspension for ten minutes and adding boiled water to supply that lost by evaporation. The control line and the thyroid line of Experiment B 3 were contin- ued during this experiment until the life cycle of the B’ line came to an end, the thyroid line after sixty days of treatment dying out at the same time. The division rates by five day periods are shown in table 17. TABLE 18 Race B’’. Experiment 6. Daily division rate averaged by five day periods PERIOD ‘ BOILED THYROID TH YROID-CONTROL 21 OS 35 22 1.40 0.70 23 1.35 0.80 24 1.35 0.80 25 2.00 1.45 26 2.00 125 27 1.60 25 28 2.05 Ib slls 29 1.80 0.95 30 1.70 0.95 31 1.95 115 32 170s 1.65 33 1.30 0.40 34 1.70 0.80 35 1.70 1.50 36 2.20 TO 3 2.00 1.10 38 - 1.50 0.95 » 39 2.30 0.80 40 1.75 USO) 4] 1.85 1.10 42 1.70 0.50 43 0.90 0.40 44 iL 1155 0.70 45 0.85 0.10 46 ANVCPAD Gime. Vis Sata etree 1.65 0.95 EFFECTS OF THYROID ON PARAMAECIUM 547 The line which had been drawn from the control only fifteen days before and treated with boiled thyroid continued at a high rate of division while the control and thyroid lines cited above were undergoing their last depression period. From this an- other line B’” was drawn and returned to the control medium. Table 18 gives the division rate of these two lines for the fol- lowing one hundred and twenty-five days, during which the line treated with boiled thyroid divided at a rate of 0.70 of a di- vision per day more rapidly than the line derived from it and subsequently maintained in hay infusion. And in figure 5 are shown graphically the life histories of these two lines, where again the close similarity of the two curves may be observed. TABLE 19 Race B. Summary of experiments. Average division rate per day EXPERIMENT CONTROL THYROID DAYS I 0.60 1.60 25 II 1.00 1.20 25 III 0.75 1.00 60 IV 1.00 1.20 5 Wie. 0.60 1.25 20 VI 0.95 1.65 125 Average.......... Osa: 1.40 260 * Percentage increase produced by thyroid feeding, 64 per cent. Finally in table 19 is given a summary of the average daily division rate of the control and thyroid lines respectively ob- tained in all the experiments performed with race B. A com- parison with table 10 shows that the effect of the thyroid is not at first sight so marked in the caudatum race (an increase of 0.55 division per day) as in the aurelia race (an increase of 1.05 divisions per day). This difference disappears when we com- pare the average division rates of the control lines for the two species (0.85 and 1.60 divisions per day respectively) and we find that the percentage of increase in the two races is practically iden- tical, being 65 per cent for Race A and 64 per cent for Race B. c. On the C race. The third race of Paramaecia, C, was es- tablished in order to observe in greater detail the other effects 548 WALDO SHUMWAY produced in thyroid fed individuals. Records were also kept for the purpose of comparison with the effects produced on the division rate. A summary of these reduced to five day averages is given in table 20 for the control, a thyroid line. The results are similar to those obtained in the other two races and it will be noted that the percentage increase produced by the thyroid is again 65 per cent. A graph (fig. 6) comparing the life his- tories of the thyroid and control line illustrates how closely the control life curve follows that of the control, points of high and low rates of division occurring simultaneously notwithstanding the fact that at these points the thyroid line 1s many generations older. TABLE 20 Race C. Average daily division rate re-averaged by five day periods PERIOD CONTROL THYROID 2 0.80 0.60 3 1.60 1.10 4 1.40 1.40 5 1.30 1.60 6 0.80 1.40 74 1.10 2.00 8 0.70 2.30 ‘ 9 0.60 2.90 10 1.10 1.60 16 0.40 1.00 12 0.30 0.90 13 0.40 1.10 14 0.70 1.40 15 0.70 1.90 16 1.30 2.90 17 1.80 2.70 18 1.20 2.10 19 1.60 2.30 20 1.00 1.60 21 0.70 1.80 22 0.95 1.95 5 1.70 1.85 24 1.85 1.80 25 125 1e25 Average for 125 days .. 1.00 1.65 Percentage increase produced by thyroid feeding, 65 per cent. EFFECTS OF THYROID ON PARAMAECIUM 549 THE EFFECT OF THE THYROID a. On the division rate. In my experiments on Paramaecium aurelia (Race A), a small quantity of thyroid emulsion was added to the hay infusion which is the common laboratory cul- ture medium for this form, and a sharply marked increase in division rate resulted. No such increase was observable in the similarly treated thymus lines. The same method was tried in the first experiments on the caudatum race (B 1) but did not give satisfactory results. Since that time I have conducted my experiments with media prepared by shaking up the gland desiccations in spring water. _Many and long continued experi- ments have demonstrated that the thyroid tissue (together with the inevitable bacterial flora) under these conditions presents all the elements necessary for the maintenance of life in these in- fusoria. I emphasize this statement in view of my demonstra- tion of the ingestion and digestion of thyroid particles by Para- maecia. Other internally secreting glands either do not sup- ply these elements or form in decomposition (as some of them undoubtedly do) substances lethal to the individuals treated with them (Experiment B 2). The thyroid however contains some substance which causes the Paramaecia to divide more rapidly. Experiments have demon- strated that in both species tested the thyroid feeding has pro- duced an increase in the division rate of 65 per cent. In other words a Paramaecium dividing once in twenty-four hours in hay infusion would give rise to 1024 daughter cells in ten days; the same Paramaecium if treated daily with thyroid prouiel produce about 185,000 daughter cells. Attempts have been made to identify this substance in iodo- thyrin (table 15). These have failed, perhaps because the prepa- ration (commercial iodothyrin in sugar of milk) was not suffi- ciently active. I have not yet had the opportunity of testing a pure iodothyrin. Other experiments have been conducted with iodine, pure or combined with potassium iodide (table 11). Far from causing any increase in division rate these preparations proved toxic, causing decreases in the division rate even in the minimal effective dosage. 550 WALDO SHUMWAY Experiments with boiled thyroid (tables 16, 17, and 18) have demonstrated that it produces effects similar to those of the raw and desiccated gland substance. The agent at work is not ex- tracted by boiling as shown by parallel experiments with the filtered bouillon. It may be noted in this connection that Oliver and Schafer (95) reported that boiled thyroid produced the same depressant effect on heart action as the untreated gland. If the results obtained in these experiments are due to a hormone secreted by the thyroid it must be one of remarkable stability. Attention is called to the constancy with which the effect of the thyroid has been observable during extended periods of daily treatment (figs. 4, 5, and 6). In the longest individual experi- ment (one hundred and forty days) the thyroid fed line divided more rapidly than the control in every period. Nor has this rapid rate of division produced any harmful effects. In no case has a thyroid fed line died out before its control. At many different periods lines have been instituted from thyroid lines of long standing and abnormally high division rates, returned to the control medium and continued as parallel lines to the parent thyroid and control lines (Experiments A 1, A 5, B 6). In every case the thyroid-control line slowly returned to the normal (hay infusion) rate of division. No unusual rise in the death-rate marked the transfer. It may be remarked in this connection that the transfer from hay infusion to thyroid media is also unaccompanied by any unusual mortality. It is of particular interest to observe the effects of the thyroid at times in the life history when the race is undergoing a depres- sion period. It has been shown by Calkins (’04) that a number of different factors may be involved in carrying a line through this critical period. He has demonstrated that the mechanical agitation of a railroad journey, a sudden change of diet, a dif- ference in the salt content, or an increase in temperature may be sufficient to restore a weakened line to its normal vitality. According to his interpretation this power of inducing regenera- tion is the essential feature of conjugation. The recent sugges- tive observations of Woodruff and Erdmann (’14) show that the 551 EFFECTS OF THYROID ON PARAMAECIUM ‘- -- - proréy} ‘—— Joryuog ‘sporsed Aep aay Aq poseaaav-o1 0781 WOISIAIp A[Iep oSBIOAB UI Og yUaUT “WodxqY Sulnp seul] [O1jUOD puB pey prlorAYy} oY} JO soTIOySIY oJ1] OAYeAvdUIOD SurMoys ydern ¢ ‘sq 9h Sh hh 8h th th ‘Or ES (CE “LE 98 “SE KE “Ee) ee Ve of 6c (82 we ot Sey ne. fe 22 Ve. OC. 6) ey, = =a esr wea L—— / (aaee =e ae ee = Bre —— | Se ! THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 22, No. 3 SHUMWAY WALDO 552 *---- prorAy} ‘ jomu0g ‘sporsed Avp eay Aq poSvroav-o1 9781 UOISTATP ATep oBvIOAB UI ‘OH JUoUTIIedxyY SulInp seul] [01}U0D puw pey prlor4y} OY} JO solsO}sTY os] eATyvavduod Surmoys ydeipy 9 “SI 9@ SE whe €@ C2 12 OC 6/4 Bs £I GLa SL Ail ae ely, ‘i On & 8 Zz 9 S: 4 & @ U 5 i 5 - wogece 1 ' ' L] Be ee ae | - i] iJ I i i] ‘ a) EFFECTS OF THYROID ON PARAMAECIUM 998 race may in some casés rejuvenate itself by a peculiar process of parthenogenesis (endomixis). The cyclical character of the life histories of my three races is shown by figures 1, 2, and 3. A few cases where experiments with thyroid were in progress at depression periods may now be considered. Experiment A 5 was commenced shortly before the beginning of a depression period (fig. 1) where for four days there was but one individual in the control line. Differ t ( mm MM, FF, FF FF, MM; M ) Time Figure 3 the initial action of FF, t(M) the same for M and t(MM) the same for MM. We see that the male has completed differen- tiation before crossing the line t(FF), and the female before crossing the line t(M). Female intersexuality can now be pro- duced either by having a less concentrated FF combined with the same M, or by combining the same FF with a higher M. In the first case t(FF) would become t(FF,), lower concentra- tion meaning a longer time of reaction. Having combined with it a slower development, the female curve would become the dotted curve I ° 1, which crosses the line t(M), at the point a, that is, being a male development from this point on. In the second case, t(FF) would be left, but t(M) changed into t(M):; THE THEORY OF SEX 607 the female curve crosses t(M,) at the point b and becomes the dotted line, with male determination. In the case of male intersexuality we have two similar possibilities. Either the value of MM is weakened in the hybrid combination, or the value of FF is strengthened. In the first case, the line t( MM) becomes t( MM), and the male curve the dotted line, I 1; it crosses the line t(FF) at the point ec and development is female from there on. In the second case—strengthening of FF—we have the line t(FF,). The male curve cuts it at the point d, resulting in the dotted end of the curve with female determination. This representation does away with all difficulties and has the great advantage of being open to experimental test. The tests being now under way, we shall refrain here from further discussion. But there is one additional point which might, at least, be mentioned. In modern Mendelian discussions, the question of a possible variation of factors plays a conspicuous role. Most of the orthodox Mendelians decline to accept the possibility of such variation, one of them actually using the expression ‘inad- missable.’ We can hardly see why the assumption of the vari- ability of a factor in regard to its quantitative value should be anything but most natural, unless we assume mystical proper- ties for such factors. Accepting our conception of factors, there is no reason why enzymes should not exhibit slight variations in quantity, although their exact concentration seems to be one of the fundamentals of heredity. (The denomination of such variations as mutations is merely a matter of taste.) The facts which we have observed during our work are much in favor of this view. In the crosses which yield intersexual animals two types of variation can be observed. There is, first, a certain variability within a given culture (brothers and sisters). This is probably due to different conditions 1a development, external as well as internal, which influence the relation of the time-factor of the enzyme reaction to the progress of differentiation. The second variation concerns the results of the same cross in dif- ferent individual cases. Generally this result is more or less similar, the mean of the resulting variability differmg only 608 RICHARD GOLDSCHMIDT shightly. But occasionally one cross gives aberrant results. For example, we have made the cross of the two races which gives nothing but males some dozen times with identical results. Once only there were three extremely intersexual females, rep- resenting some minus individuals of a range of variation. This, and similar facts from other crosses, points strongly to a quan- titative variation of the factor-enzymes in the gametes, the oc- casional extreme departures from the typical results being due to the union, in fertilization, of a minus value for one, and a plus value for the other enzyme. VIl One of the important advances which recent genetic research has made is to furnish proof for the fact that the distribution of the chromosomes, especially the sex-chromosomes, and the symbolistic conceptions about the behaviour of Mendelian factors are one and the same thing. Thus the step from a symbolistic representation of a mechanism to the real disclosure of the mech- anism has been made. We have tried in this paper to advance still further, namely, to a realization of what is moved by the mechanism and why. Do these conceptions fit the facts of cytology? We have briefly discussed that question in another place in connection with other questions, but wish again briefly to indicate the chief problem. We cannot conceive the chro- mosomes as built up from chromatin particles, which are them- selves the chemical substratum of heredity. We believe that chromatin is a skeleton substance which works as an adsorbens for the enzymes, which really constitute the chemical basis of heredity. We have now seen how important the quantitative behavior of these enzymes is for the process of heredity. The quantity of adsorption of an enzyme by an adsorbens depends upon the qualities of both and the surface of the adsorbens. The wonderful uniformity of size and shape of the chromosomes of a given animal appears, therefore, as a minute mechanism to guarantee the typical quantity of enzymes of heredity to be as- sembled at the moment of fertilization. And all the strange processes preceding the maturation of the sex-cells appear THE THEORY OF SEX 609 easily understandable, as well as the meaning of the peculiar mechanism of mitosis. The formation of a chromosome means, physically, the same thing as the dropping of a piece of charcoal into a solution containing enzymes. Harrison and Doneaster®? showed some time ago that species- crosses of the moth Biston exhibit phenomena which will prob- ably prove to be of the same type as intersexuality in the gypsy-moth. In one of these crosses, also, exclusively males are produced. (I may take the opportunity to add here that some years ago I, also, studied species-crosses of Biston and that my results, as far as they went, agree with Harrison’s experiments. However I dropped the work because of the difficulties in breed- ing, due to the systematic distance between the forms.) Har- rison and Doneaster studied then the chromosomes of the two parental forms and found that one of the species had very much larger elements than the other. They suggested, therefore, that size of chromosomes or quantity of chromatin might be the real thing underlying our conception of different potencies. As a matter of fact we had not overlooked this possibility, and my assistant, J. Seiler, had studied the chromosomes of European and Japanese gypsies. In his paper® which was in press when Harrison and Doneaster’s was published he gives pictures of the chromosome-sets of those forms and states that the Japanese form has slightly larger chromosomes. The difference was, how- ever, not great enough to appear very important. We have re- cently investigated this point again and compared the chromo- some sizes of many races of known potencies. The result is not very encouraging, as nothing like a parallel between potency and chromosome size could be found. In figure 53 is given a photograph of the equatorial plate of the first maturation di- vision in the spermatocytes of one of the forms with very high potency of the sex-factors (the Japanese race A); figure 54 rep- resents the same stage under the same magnification from the 5 Harrison, 1. W. H. and Doncaster, L. On hybrids between moths of the geometrid subfamily Bistoninae, etc. Journ. Genetics, 11, 1914. 6 Seiler, J. Das Verhalten der Geschlechtschromosomen bei Lepidopteren. Arch. f. Zellforseh., 13, 1914. 610 RICHARD GOLDSCHMIDT very weak Japanese race, H. And the chromosomes of the latter are larger. We do not, therefore, expect much information from a study of the chromosomes in our case. We believe that the adsorption of different quantities of factor-enzymes by the chro- mosome skeleton may sometimes be connected with visible dif- ferences of chromosome surface; but it is not at all necessary that the differences should be actually visible. Vill If the views which, in consequence of our experiments, we feel compelled to adopt, come near the truth, we should expect them to be applicable to other facts in regard to sex-determination (i.e., the case of Bonellia, the hormonic alteration of sex in transplantation and castration experiments in birds and mam- mals or in the free-martin, the case of the frog, etc.) as well as to the general facts of heredity. This is actually the case but we shall refrain from detailed discussion here. The appli- cation is so evident that it may easily be inferred. It is, more- over, by no means a new idea that factors may be regarded as enzymes. Many writers have advanced similar views, as, for example, Bateson, Guyer, Hagedoorn, Loeb, Moore, Woltereck aad the writer. And Woltereck,‘ especially, has worked out the idea in regard to sex-determination, in order to explain his breeding results with Daphnids. He uses the view which we, as well as some other Mendelian writers, also used from the be- ginning of our work, that separate factors exist for both sexes. These factors Woltereck calls concurring sex-substances, which are present in every egg and which he conceives as zymogens. One of them can become dominant either by the action of acti- vators oc by the action of inhibitors of the alternative zymogen. Thus the zymogens are transformed into active enzymes. He then works out this conception in terms of immunochemistry. And since he needs definite cyclical changes of the relative valency and latency for the explanation of the life-cycles, he adds the necessary hypotheses for the explanation of the latter. In gen- eral his hypothesis does not differ from some Mendelian formu- lations which work with inhibitors, activators, changes of domi- THE THEORY OF SEX 611 nance, ete., except that he speaks in terms of definite substances instead of symbolic factors. We believe that we can reach a really good explanation of the life-cycles of Daphnids by omit- ting some of the complications of Woltereck’s hypothesis and adding to it the simple quantitative conception deduced from our work. ‘This seems especially hopeful at present since Banta’ has very recently communicated the discovery of intersexual strains in Daphnia. Further discussion had, therefore, better await the details of Banta’s work. Osborn Zoological Laboratory, Yale University 7 Banta, A. M. Sex intergrades in a species of crustacea. Proc. Nat. Acad. Sc., 2, October, 1916. THE THEORY OF SEX RICHARD GOLDSCHMIDT PLATE 1 3 © . 61 THE THEORY OF SEX RICHARD GOLDSCHMIDT 614 PLATE 2 615 , NO. 3 22 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. THE THEORY OF SEX RICHARD GOLDSCHMIDT 616 PLATE 3 SUBJECT AND AUTHOR INDEX Ko behavaOren serine emits etensrstelsio 193 Actinians. Nervous transmission AM thesasecte Ou Actinians. Pedal LOCOMOLION INE: wc ccic- = see: 111 ‘Actinians. The movements of the tentacles 7" tai lbp ueenndo seco goepoaueSo soc DunOndEor 5 Aleohol and certain related substances. The experimental modification of germ cells. I. General plan of experiments with ethy! 125 Alcohol and certain related substances. The experimental modification of germ cells. II. The effect upon the domestic fow! of the daily inhalation of ethyl...........--- 165 Alcoholism, and certain other drug intoxica- tions, upon the progeny. The experi- mental modification of germ cells. III. The effect of parental............---..--- 241 Ameba to isolated and compound proteins. On the reactions Of..0: .--)c.6-sescse2 +s 53 Amebae. The effect of media of different densities on the shape of...........---+-+ 565 ACTERIA. The growth of Paramae- B cium in pure cultures of...........++++-- 421 Behaviors, ACHMIAI secre ciclereie cjevejeresisteieiecyeicen 193 Bunsen-Roscoe law to the phenomena of ani- mal heliotropism. A re-examination of the applicability of the.........-..-.++--- 187 ELLS. I. General plan of experiments C with ethyl aleohol and certain related substances. The experimental modifi- CamlOnnOle CLMMe ecclissi etter let 125 Cells. II. The effect upon the domestic fowl of the daily inhalation of ethyl alcohol and certain related substances. The ex- perimental modification of germ.......... 165 Cells. III. The effect of parental alcoholism, and certain other drug intoxications, upon the progeny. The experimental modi- FiGATIOMNOL POLI: