x? at va mo be, yt oe: ee Ste lena ced eee stcty Ms On; ! arene Ne a. rot ee ne Che ries ee, phate Seite ae ve ee, Ge SSE Oe 4 7 stint $. pet ~ 3 . > se ray rari et iH is oe peace 4 ys Aish “eth : a? ee q THE JOURNAL OF EXPERIMENTAL ZOOLOGY EDITED BY JACQuES LoEB The Rockefeller Institute Wiiu1aMm E. CastLe Harvard University ,. 7 4 7 \ > , me faye , 7S : $ R - Te ‘ y ‘eo ie bAtee Air Jf G aE VAS Titeaaaan - COMPOSED AND PRINTED Al WAVERLY PRESS By THe Witiiams & W Bautrmore, Mp., CONTENTS NO. 1 JANUARY S.O. Masr. The process of orientation in the colonial organism, Gonium pectorale, and a study of the structure and function of the eyespot. Six figures............ 1 eyo. LASHLEY. Results of continued selection im Elydra.>....5s2..9.4...2.0- SAustees siete 19 NO. 2 FEBRUARY HaArRoup Saxton Burr. The effects of the removal of the nasal pits in Amblystoma em- bryos. Fourteen figures (three plates) ............. Bisestee BE MPa taesy Sree tere CRI Hee AC Ruopa ERpDMANN AND LORANDE Loss Wooprurr. ‘The periodic reorganization process in Paramaecium caudatum. Thirty-five figures (seven plates)..................... 59 Lispie H. Hyman. An analysis of the process of regeneration in certain microdrilous oligochaetes. Twenty-four figures............. eye COLSON Me ey StS EBT eccre ne Ramer ice eh) S. R. Derwiter. The effect of light on the retina of the tortoise and of the lizard. Eleven figures.......... ile one erat a steee a Day che aaa eve ee eo ae oete PPM oe veya aeeteyetts SH Tae he oe OO Reynoup A. SparrH. Evidence proving the melanophore to be a disguised type of smooth muscle cell. ‘Two figures......... Ue Eat teu Seouensmereietatsre ake Slaves Le a eeebarepetefaretar. arama OG Jacques Lors AND HAarpoLpH WastTENEYsS. The relative efficiency of various parts of the spectrum for the heliotropie reactions of animals and plants. Six figures....... 217 Henry Laurens. The reactions of the melanophores of Amblystoma larvae—the sup- [OOSNG! TMA Oe TS poUMEKAl GIFU, SIDS IKAVORAS) SooodcaaocaccousansbucousGacboou‘cs ZU Davip Day Wuitney. The control of sex by food in five species of rotifers. Six figures 263 w=’ NO. 3 APRIL W.J.Crozimr. The rhythmic pulsation of the cloaca cf Holothurians. Thirty-one LIN OTSYS} os & ee Oeste REINER 4 Cera Cede RR rot, ata Aah ees CA Raia Sine Fle Ruahaetaratetas She 5 2 Wituram L. Dotiey, Jr. Reactions to light in Vanessa antiopa, with special reference CORCINCUSHMOVEMeENtSs Mawelity—-ONe) im UNES leis eee cence sila aielstaisl= sleyelerehanenslalc) = Ce OOK H. D. Goopate. A feminized cockerel. Seven. figures............... DBS Sk ate teres aie so 2 PAIL Setic Hecut. The water current produced by Ascidia atra Lesueur. One figure...... 429 E. A. Anprews. Color changes in the rhinoceros beetle, Dynestes tityrus. Four APTOS ys To EME neo ei skige cl died seam sore epiane te Sy lokeaela, si Re eA ree cet ae A oe . 435 1V CONTENTS NO. 4 MAY R. T. Youna. Some experiments on protective coloration. Sixty-nine figures........ 457 THEOPHILUS S. PatnTtER. Contributions to the study of cell mechanics. I. Spiral asters. Seven textiigures andktworplates: ie «0.0.24 dc ean ete bo) ore See ee 509 Asa A. ScHAEFFER. On the feeding habits of Ameba. Six plates..................... 529 BrapLtey M. Parren. The changes of the blowfly larva’s photosensitivity with age. HOW GigE s es eae esi suse na coe cs oe CE tec eh Bc ee cree Ba a chs ee 585 THE PROCESS OF ORIENTATION IN THE COLONIAL ORGANISM, GONIUM PECTORALE, AND A SrUDY OF THE STRUCTURE AND FUNCTION OF THE EYE-SPOT S. O. MAST From the Zoélogical Laboratory of the Johns Hopkins University SIX FIGURES CONTENTS PPMOTEROUIECHION. 5 tea eee Meet. noe ee Cl) Sa es RI a 1 PSEC hUITe: OF GOniiNaT 5.7 aa Mee sce el. cites kas (eae Sas Oe a 3 3. Structure of the eye-spot in Gonium, Eudornia and other forms......... 5 4. PPO GOSS Olt CGN onal, mol Cxonmitonan | Jy sik Sega ao anaadovdudadsaeneoseau cle. 8 MDS CUSSION cc. cagrecs lester ein ae ee ka et Cie co tes Mos) Maal igs abet eRe me 13 Ts PSL OTSOW eh OA anc ae eee Ma teaee 0S) OER aN mb Sab Aa 16 1. INTRODUCTION Among the most machine-like of the activities in organisms is the process of orientation. It is consequently not surprising that this process, which is common to so many different species, has received much attent on in the investigations on behavior, with the result that a mass of highly interesting and important facts regarding it has been col ected. These facts seem to show that the process of orientat on differs fundamentally in different organisms and that it is far more complicated than has been assumed by some investigators, but that in general it facilitates the life processes in the individuals possessing it, and conse- quently tends to perpetuate the species. As to the reduction of the process to mechanical principles even in its simplest form, and as to its relation to conscious phenomena, little more can be said than that the field here is still wide open, anon prospects are not altogether discouraging. Among the questions associated with the process of orienta- tion concerning which there is at present much contention is 1 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VoL. 20, No. 1 JANUARY, 1916 ot Ss. O. MAST that referring to the nature of the orienting stimulus. Loeb and some of his followers hold that this stimulus is, in all organisms, animals as well as plants, dependent, in a specific way, upon the amount of stimulating energy received by the sensitive tissue, in accord with the Bunsen-Roscoe law. That is, that a given amount of stimulating energy (which is the product of the in- tensity of the agent and the time it acts) always produces the same effect no matter how these two factors may vary. Thus according to this idea a weak agent acting a long time should cause the same response as a strong one acting a short time. This is one of the essential characteristics of a theory of ori- entation which will be referred to as the ‘continuous-action’ theory. Darwin and others maintain that, in some cases at least, the orienting stimulus is dependent upon the time-rate of change of stimulating energy. That is, that if there is no change in such energy there will be no response, no matter how much energy may be received. Response in accord with this idea constitutes the most important feature of a theory of orientation which will be called the ‘change-of-intensity’ theory. Our observations on the process of orientation in Gonium strongly support the latter theory. They also support the con- tention that the eye-spots function as direction eyes essentially as do the eyes in some of the flat-worms. The reactions of the colonial organisms have not received much attention. Volvox is the only member of this group that has been extensively studied, and in this form only the responses to light have been thoroughly investigated (Mast ’07). It will be expedient to present, in this connection, the essential features of these responses since we desire later to compare them with those observed in Gonium. Volvox, like most of the simple green organisms, responds very definitely to light. It orients fairly accurately, and is usually positive in light of moderate intensity and negative in that of high intensity. Orientation is direct, that is, if the position of the source of light is changed after the colonies are oriented they always turn at once toward the light again (provided they are ORIENTATION IN COLONIAL ORGANISM 3 positive) never in the opposite direction, as frequently: occurs in Euglena, Stentor and the like. The turning of the colonies is due to an increase in the effective stroke of the flagella on the shaded side. So much has been definitely established. As to the cause of the increase in the activity of the flagella on this side we are, however, not in a position to speak with so much assurance, but our evidence seems to indicate that it is depend- ent upon the time-rate of change of light intensity on the photo- sensitive tissue in the individual zodids. Let us briefly consider this evidence. During the process of locomotion the colonies continuously rotate on the longitudinal axis; consequently, when they are not oriented and opposite sides are unequally illuminated, the zoids are continuously transferred from a region of higher to one of lower light intensity and vice versa. This results in a decrease of intensity on each zo6did, when it reaches the shaded side of the colony; but there is another factor involved in producing changes of intensity, in all probability of more significance than this. In unoriented colonies the zoéids are not only continuously subjected to a transfer from one intensity of light to another, but during the transfer different surfaces become exposed; for when they are on the more highly illuminated side of the colony the outer surface, and when they are on the opposite side the inner surface is directed toward the source of light. Since the zooids contain numerous translucent and refractive bodies this change in the surface exposed necessarily results in numerous changes of intensity on the different parts of each zodid. The eye-spots are the most prominent of the translucent bodies men- tioned and they are consequently of greatest importance in pro- ducing changes of intensity within the zodids. The orienting stimuli, in my opinion, depend upon the time-rate of change of intensity thus produced. This opinion is strongly supported by the reactions to light observed in Gonium. These reactions are essentially like those in Volvox, but before describing them it will be necessary to re- fer briefly to the structure of the organisms under consideration. 4 Ss. O. MAST 2. STRUCTURE OF GONIUM Gonium is a thin flat rectangular structure somewhat over 0.1 mm. wide. It consists of 16 cells or zodids loosely united with protoplasmic strands which penetrate a gelatinous substance found in the intercellular spaces (fig. 1). Each zoéid contains among other things, a relatively large chloroplast, a prominent 7 OLS) 1 Fig. 1 Camera lucida drawing of Gonium as seen from the posterior surface. Each colony contains 16 zodids, all situated in the same plane. 2z, zodids; e. eye- spot; p, pyrinoid; s, strands connecting the zodids; f, flagella; mm, projected scale. The arrow indicates the usual direction of rotation. The eye-spots are located at the outer surface near the anterior end of the zodids. Note that they are a little to one side of the middle of this surface. ORIENTATION IN COLONIAL ORGANISM 5) eye-spot and two flagella which are about as long as the colonies are wide (fig. 2). The zodids are slightly elongated and so situ- ated that the longitudinal axes of all are nearly parallel to each other and approximately perpendicular to the flat surfaces of the colonies. The flagella extend from the anterior surface of the colony and the eye-spots are located near their base on the outer surface of the zodids near the anterior end. Fig. 2 Free hand sketch of Gonium as seen from one side. 2z, zodids; e, eye- spot; p, pyrinoid; f, flagella. The eye-spots are situated on the outer surface of the zodids near the anterior end. For details regarding the structure of the eye- spot, see figure 3. 3. STRUCTURE OF THE EYE-SPOT IN GONIUM, EUDORINA AND OTHER FORMS! Ever since the days of Ehrenberg (’31) the eye-spots or stig- mata, as they are frequently called, have been looked upon by many as the most primitive eyes. They have consequently been of great interest especially to those concerned with the evo- lution of the visual apparatus in the higher forms. They have been described in many different organisms by various investi- gators. Among these Franzé (’93) probably made the most ex- tensive studies. He investigated them in 31 different species. Practically all of those who have worked on the eye-spots main- — tain that they consist of two essentially different substances, a 1 This section is the result of histological studies made by Caswell Grave. It is a pleasure to acknowledge my great indebtedness to him for his generous assistance. 6 Ss. O. MAST hyaline.substance, globular or lenticular in form and a brownish opaque substance frequently somewhat cup-shaped. The latter, it is held, usually surrounds the former more or less completely. Thus it appears that these structures resemble, somewhat, the eyes in turbellaria, rotifera and copepoda, and this is largely re- sponsible for the conclusion frequently stated that the former are homologous with the latter. Franzé (’93, p. 162), however, op- poses this contention. He says: ‘“‘Die Augen der Turbellarien und Rotatorien sind keine Homologa der Stigmata, sondern die fiusserliche Ahnlichkeit beider Differenzirungen wird durch die gleichen Funktionen bedingt.’’ He, in common with a large pro- portion of other investigators, holds that the eye-spots function as light recipient organs. While much of the work on the structure of the stigmata has been thorough, it was our opinion that with the application of modern histological technic it might be possible to discover ele-- ments in them that would throw light on their nature and func- tion. With this in view colonies of Gonium and Eudorina were fixed in Bouin’s and Fleming’s fluids. Some were embedded in paraffin and cut into sections 2u and 3y in thickness and stained with iron haematoxylin and safranin. Others were mounted whole, some stained and some not. These preparations were thoroughly studied with a combination of No. 6 Comp. ocular and 2 mm. Apoch. Homog. Immersion objective and briefly with more efficient combinations. It was found that the eye-spots both in Gonium and in Eudorina consist of two parts, an opaque cup-like structure and a lens shaped hyaline structure (figs. 3 and 4) but no further details could be seen in them although the best lens systems made were used. This, of course, does not prove that there is no finer structure present. It merely indi- cates that if there is, it is ultra microscopic. By referring to figures 3 and 4 it will be seen that the eye- spots in Eudorina are considerably larger than those in Gonium. These figures indicate that they are situated at the surface of the zodids with the hyaline portion outside. Careful observa- tions seem to indicate that there is a thin protoplasmic layer, not represented in the figures, which is outside of this structure and ORIENTATION IN COLONIAL ORGANISM 7 extends entirely around the zodid and is continuous with the strands which connect them with each other (fig. 1). This layer of substance is more distinct in living colonies than in sections. Thus it is highly probably that all of the eye-spots in a colony have protoplasmic interconnections. Superficially these eye-spots are very much like the primitive eyes in turbellaria and Amphioxus as will readily be seen by com- paring figures 3 and 4 with figures 5aandb. In the latter the C .. ———$— 003MM a a Fig. 3 Camera sketch of a section of Gonium taken perpendicular to the plane of the colony showing three zodids. Sections 3u thick. No. 6 compensat- ing ocular and ;'; homo. oil immersion objective. Enlarged 4 diameters with pentograph. e, eye-spot; p, pyrinoid; n, nucleus; mm, projected scale. The eye- spot consists of an opaque saucer shaped structure and,a hyaline lens-shaped body. It is less than Iu in diameter. In the zoéid to the left the razor passed nearly through the middle of the eye-spot; in the other two zodids it passed a little to one side of the middle, consequently the hyaline part appears relatively smaller in these. Drawn by Caswell Grave. opaque part appears to function in restricting to certain areas, the field from which the sensitive hyaline position received light. Thus they seem to function as direction eyes. The eye-spots probably function in the same way. At any rate, these bodies in many species are so well differentiated and so similar in their structure and position in different individuals that they can not be looked upon merely as accumulation of waste products as is maintained by a considerable number of investigators. 8 S. O. MAST 4. PROCESS OF ORIENTATION IN GONIUM The observations on orientation in Gonium were made in es- sentially the same way as those described in earlier works (Mast "11, pp. 92-96) it will consequently not be necessary to discuss methods here. ‘The results of these observations follow: Gonium swims in a fairly direct course with the flat surface perpendicular to the direction of motion. The surface with the a —_— es 77103 MM —WH—_5 Fig. 4 Camera lucida sketch of a longitudinal section of Eudorina nearly through the middle showing two of the four anterior zodids. No.6 comp. ocular; iz homo. oil immersion objective. a, anterior end of colony; s, outer surface; n, nucleus; p, pyrinoid; e, eye-spot; mm, projected scale. The eye-spot in this form is essentially like that in Gonium but it is much larger and the two parts can be much more distinctly seen. The best lenses available fail, however, to reveal any differentiation in these two parts in either form. Drawn by Caswell Grave. flagella is always ahead. As it proceeds it continuously rotates, usually counter-clock-wise as seen from the rear, although it reverses frequently and rotates in the opposite direction for short periods of time. It orients fairly accurately in light, being ordi- narily positive in moderate and negative in strong illumination. .RIENTATION IN COLONIAL ORGANISM 0) If the position of the source of light is changed after a colony is oriented so that the rays strike the anterior surface obliquely, it turns at once until the rays are again approximately perpendicu- lar to this surface. The colony as a whole never turns in the wrong direction. Orientation is direct. The process is essen- tially the same as in Volvox. There is no indication of random movements or trial reactions in the colony as a whole. The turning of the colony in the process of orientation is due to an increase in the activity of the flagellae of the zodids farthest from the source of light. The following evidence indicates that Fig. 5 A, Sketch of the eye of the turbellarian, Ut. vulgaris prepared by crushing a living specimen. After Wilhelmi (Taf. 15, fig. 4). ; B, Sketch of a cross-section of the eye of Amphioxus lanceolatus. After Hesse (Taf. 24, fig. 8). C, Sketch of a cross-section of the eye of the turbellarian, Sabussowia diocia. After Bohmig (Taf. 12, fig. 15). It can readily be seen that in all of these animals the eyes consist of two essentially different parts, just as do the eye-spots in Gonium and Eudorina, but that there is considerable differentiation in each part in the former while in the latter there is none that can be seen. 10 ; S. O. MAST this increase in activity is due to a reduction of light energy on the sensitive tissue in the zoéids and that it is dependent upon the time-rate of reduction, not upon the absolute amount of reduc- tion. If the light intensity in a beam in which positive colonies are oriented is suddenly decreased without in any way changing the direction of the rays, the rate of movement for a short period of time suddenly increases, but if the intensity is suddenly increased there is no response. In negative colonies, however, just the opposite is true. They respond in precisely the same way to a sudden increase but not to a sudden decrease of intensity. This response of the colonies is very striking. It gives one the im- pression of a very marked forward spring, and seems to be in all essentials like the shock-reactions in Euglena. And just as — in Euglena it does not occur if the light-energy is gradually changed. Obviously then, this response is dependent upon the time-rate of change of energy and not upon the absolute change. The increase in activity in the zodids farthest from the source of light, during the process of orientation, appears to be of pre- cisely the same nature as the increase in activity of all the zo- oids, due to a sudden decrease of the light-intensity in the entire field; consequently it would seem reasonable to conclude that it also is due to a sudden decrease of intensity on the sensitive tis- sue in the zodids involved. How can this occur? In unoriented colonies the light strikes the anterior face obliquely (fig. 6), and as these colonies rotate it is evident, just as in Volvox, that in each zodid the surface exposed to the light continuously changes. This necessarily causes changes of in- tensity owing to the movement of the shadows cast by the trans- lucent bodies in the zodids, particularly the opaque portion of the eye-spots. By referring to figure 6 it will be seen that in the zooids on the side of the colony nearest the sources of light the hyaline portion of the eye-spot is fully exposed, while in those on the opposite side this structure is shaded by the opaque portion. There is consequently a great reduction in the intensity of the light on it as the zodids, owing to the rotation of the colony, are transferred from the former to the latter position and an equally ORIENTATION IN COLONIAL ORGANISM bl great increase as they are brought back to the original position again. If, then, the photo-sensitive tissue is largely confined to this hvaline substance as it seems to be in Euglena, we should Vases: Fig. 6 Diagrammatic sketch representing the process of orientation in a colony of Gonium as seen from the side., Each circle represents a zodid; the small arcs represent the eye-spots, the arrows the direction of the rays of light, and a-b, ai-b1, a2-be, etc., different positions assumed by the colony during the pro- cess of orientation. Only 4 of the 16 zodids in the colony are shown. The colony rotates on its anterio-posterior axis as it proceeds. This causes the hyaline por- tion of the eye-spot to become alternately fully exposed to the light and shaded. The turning of the colony is due to an increase in activity of the zodids as they are transferred to a position in which the hyaline part of the eye-spots is shaded, that is, the positions represented by a, b1, a2, b3,etc. The hyaline part is probably highly sensitive to light, and the increase in activity mentioned is probably de- pendent upon the time-rate of reduction in illumination on this part. 2 Ss. O. MAST expect in positive specimens a shock-reaction in the zodids on the side of the colony farthest from the source of light, owing to the shading of this substance, but none on the opposite side where this substance becomes exposed to the light. This is precisely what is observed in the process of orientation. In colonies in the negative state, on the other hand, we should expect just the reverse and this is also in accord with our observations. More- over, as the colony turns toward the light the change of intensity on the hyaline portion of the eye-spot becomes gradually less and when it has turned enough so that it directly faces the light, that is, when it is oriented, there is no longer any change, and consequently on the basis of our assumption no more shock-re- actions and no further turning would be expected. This is again in accord with our observations. If, then, the photo-sensitive substance is confined to the hyaline portion of the eye-spot, or largely so, and if the orienting stimulus is dependent upon the time-rate of change of light-intensity on this substance, we can account for the observed reactions in the process of orientation, in Gonium, and this is in full accord with our explanation of ori- entation in Stentor, Euglena, and a number of other organisms (Mast, 711, pp. 80-135). Moreover, if our assumptions are cor- rect, it is no longer necessary to hold, as in earlier publications (Mast ’11, p. 133), that the function of the eye-spot is not the same in all organisms. But how on the basis of these assump- tions is it possible to explain the fact that after the colony is ori- ented it continues in a fairly direct course toward the light? This question has already been answered, in part, in the state- ment that after the colonies are oriented changes of light inten- sity on the hyaline substance in the eye-spots cease, and conse- quently, if our explanation of orientation holds, no further turn- ing would be expected, and the colonies should therefore remain oriented, unless some factors other than the light in which they are oriented cause them to turn. And if this should occur the orienting stimulus would, in accord with our explanation, again act, and result in reorientation. This explanation is based upon — the well known principle that organisms not subjected to lateral stimulation tend to move in direct paths. To account for con- ORIENTATION IN COLONIAL ORGANISM 13 tinued orientation; it is consequently not necessary to assume that the orienting stimulus continues to act after orientation as well as during the process of orientation, as is demanded by the continuous-action theory. Organisms are partially isolated dynamic systems and much that they do is dependent upon changes within, quite independent of immediate environmental factors. 5. DISCUSSION We have in this and in previous publications presented a con- siderable amount of evidence in favor of the change-of-intensity theory of orientation. Let us now briefly consider the evidence that favors the continuous-action theory. In the reactions of the unicellular and the colonial forms very little has been discovered that supports this theory. In fact practically all of the favorable evidence is found in Bancroft’s work on Euglena (13). Bancroft maintains that he has demon- strated that in this form orientation occurs in accord with Loeb’s continuous-action theory; at any rate that the change-of-inten- sity theory does not hold. I need not here enter upon a discus- sion of Bancroft’s interesting observations, for I have elsewhere shown (’14) that his results must be confirmed under conditions more thoroughly controlled before much dependence can be placed upon them and that if his contentions are valid they ac- tually oppose the theory that he substitutes for the one he claims to have overthrown; that is, they oppose the continuous-action theory. Thus we see that the evidence in support of the continuous- action theory found in the reactions of the forms mentioned is exceedingly weak. In the reactions of some of the more com- plex organisms there is, however, some evidence indicating that this theory holds at least in part. Blaauw (’08), Fréschel (’08 and ’10), Arisz (11), and Clark (’13) have demonstrated that photic orientations, in a number of different seedlings, is within ¢certain limits, dependent upon the amount of light energy received ; that is, that long exposure in weak light produces the same effect as short exposure in strong light. Mast (11, p. 163) reached , 14 Ss. O. MAST conclusions regarding the orientation of Eudendrium which are in harmony with the work just mentioned, and Loeb and Ewald (14) support these conclusions. Ewald (714) also maintains that certain responses in Daphnia are proportional to the amount of stimulating energy; at any rate that they are not dependent upon the time-rate-of-change of such energy. Patten (’14) comes to the same conclusion regarding orientation in blow-fly larvae. He says (p. 272): “‘Orientation in the blow-fly larva depends to a large extent on the stimulating effect of constant intensity. The reaction to light of constant intensity follows the Bunsen- Roscoe law.” Both Ewald and Patten base their conclusions upon the fact that the reactions observed at the intersection of two beams of light were the same when the light in one beam was intermittent, as they were when it was constant in both, provided the relative amount of energy in the two beams was the same under both conditions, and provided that the intermission was relatively fre- quent and continuous. It seems to me that these results do not show that the reac- tions referred to in Daphnia and the process of orientation in the blow-fly larvae are necessarily dependent upon the continuous action of the light as maintained by Ewald and Patten. All that they actually demonstrate is that if the intermission is of suf- ficient frequency, periodic illumination acts the same as continu- ous illumination. This is true for the human eye and yet no one holds that this in itself precludes the possibility that stimu- lation is dependent upon time-rate of change of energy. More- over, Patten in his work on the fly larvae did not eliminate changes of intensity on the sensitive tissues in the larvae due to the alternate extension and retraction of the anterior end, con- sequently the reactions observed in the process of orientation in these animals may have been due to these changes without ref- erence to the amount of light energy received. Patten’s conclusion (p. 272) that ‘‘orientation to light from two sources depends on the relative amount of stimulation re-* ceived by symmetrically located sensitive areas” is equally pre- carious. Asa matter of fact, all of the responses which he main- 1 ~ ORIENTATION IN COLONIAL ORGANISM 15 tains favor this conclusion could be accounted for on the basis of the change-of-intensity theory, even if the photo-sensitive tis- sue were confined to a single median spot such that there could be no balancing of effects on symmetrically located tissues. This author, moreover, comes to another conclusion that seems to be supported by neither logic nor fact. He says (p. 271): ‘“‘Bancroft’s (13) work, in which he showed not only that there was a distinct reaction to constant intensity present in Euglena but that it was largely the reaction to constant intensity which determined its orientation, shows the untenability of Mast’s sweeping statement in one of the forms on which Mast himself worked.”’ What is this sweeping statement? Our author quotes it as follows -(p. 270): Mast (11, p. 234) says: There is no conclusive evidence, except per- haps in animals with image forming eyes, showing that light acts con- tinuously as a directive stimulus, that symmetrically located sides are continuously stimulated . . . . (p. 235). Light no doubt acts on organisms without a change of intensity much as constant tempera- ture does, making them more or less active and inducing changes in the sense of orientation; but there is no conclusive evidence showing that light acting thus ever functions in the process of orientation. Is it not perfectly obvious that the results of Bancroft’s in- vestigations presented in 1913, assuming that they are as quoted above, do not have the slightest bearing on the validity of this statement, made in 1911? Does the discovery of a certain re- sponse at a given time make untenable the statement that it had not previously been discovered? Whatever the final conclusion may be regarding the two theo- ries of orientation in question the fact that many reactions in animate systems depend upon the time-rate-of-change of stimu- lating energy is well established. These reactions are of great interest, partly because they are exceedingly rare in inanimate systems, and a thorough study of them cannot fail to yield in- teresting results. 16 Ss. O. MAST 6. SUMMARY 1) The eye-spots in both Gonium and Eudorina consist of an opaque cup-shaped part and a hyaline lens-shaped part. The latter is partially surrounded by the former and it is probably relatively very sensitive to changes in light-intensity. These changes are probably largely due to shadows produced by the opaque part. 2) Orientation in Gonium is direct. The colonies never turn in the wrong direction, as often occurs in Euglena, Stentor, and many other forms. The turning which results in orientation is due to an increase in the activity of the flagella on the zodids farthest from the sources of light. In these zodids the hyaline part of the eye-spot is shaded by the opaque part at the time the activity of the flagella increases. 3) If the light-intensity of the field is suddenly decreased, the rate of locomotion, in positive colonies, suddenly increases. But if it is slowly decreased or if it is increased there is no response. In negative colonies, however, just the reverse is true. They respond to a sudden increase in rate of locomotion if the illu- mination is decreased but not if it is increased. 4) The increase in the activity of the zodids on one side of the colony during the process of orientationis apparently of the same nature as the increase in the activity of the whole colony when the illumination is changed. This indicates that orientation in these organisms is dependent upon the time-rate of change of light energy on the photo-sensitive substance, probably the hyaline portion of the eye-spots, and not upon the absolute change or the continuous-action of light. ORIENTATION IN COLONIAL ORGANISM i BIBLIOGRAPHY Arisz, W. H.° 1911 On the connection between stimulus and effect in photo- tropic curvatures of seedlings of Avena sativa. Kon. Ak. Wet. Am- sterdam. Proc., pp. 1022-1031. Bancrort, F. W. 1913 Heliotropism, differential sensibility, and galvano- tropism in Euglena. Jour. Exp. Zo6l., vol. 15, pp. 383-428. Biaauw, A. H. 1908 The intensity of light and the length of illumination in the phototropie curvature in seedlings of Avena sativa (oats). Kon. Ak. Wet. Amsterdam. Proc. 1909 Review in Bot. Cent., vol. 110, p. 655. 1909 Die Perzeption des Lichtes. Rec. d. Trav. bot. Neerl., vol. 5, pp. 209-377. 1910 Review in Bot. Cent., vol. 113, pp. 353-356. Boéumic, Lupwic. 1906 Tricladenstudien. Zeitsch. f. Wiss. Zool., Bd. 81,5. 344-504. Cruark, O. L. 1913 Uber negativen Phototropismus bei Avena sativa. Botan. Zeitschr., Bd. 5, S. 737-770. Ewatp, W. F. 1914 Versuche zur Analyse der Licht und Farbenreaktionen eines Wirbellosen (Daphnia pulex). Zeit. f. Sinnesphys., Bd. 48, 8. 285-324. Franz, R. 1893 Zur Morphologie und Physiologie der Stigmata der Mastigo- phoren. Zeitschr. f. Wiss. Zool., Bd. 56, S. 138-164. Froscuer, P. 1908 Untersuchungen iiber die heliotropische Prisentations- zeit. Sb. mat-naturw. Kl. Akad. Wiss., Wien Bd. 117, Abt. 1, S. 235- 256. Hesse, R. 1897 Untersuchungen iiber die Organe der Lichtempfindung bei niederen Thieren. IV. Die Sehorgane des Amphioxus. Zeitsch. f. Wiss. Zool., Bd. 63, S. 456-464. JenNiINGS, H.S. 1906 Behavior of the lower organisms. New York, pp. 366. Lorn, J. and Ewautp, W. F. 1914 Uber die Giiltigkeit des Bunsen-Roscoeschen Gesetzes fiir die heliotropische Erscheinung bei Tieren. Zentb. f. Physiol., Bd. 27, S. 1165-1168. Mast, S. 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. 1914 Orientation in Euglena with some remarks on tropisms. Bio- logisches Centralblatt, Bd. 34, S. 641-674. Patten, B. M. 1914 Quantitative determination of the orienting reaction of the blow-fly larva (Calliphora erythrocephala meigen). Jour. Exp. Zo6l., vol. 17, pp. 213-280. WitHermi, J. 1909 Tricladen. Fauna und Flora des Golfes von Neapel., 32 Monographie, Berlin, 405 pp. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, No. 1 RESULTS OF CONTINUED SELECTION IN HYDRA K. S. LASHLEY From the Zoélogical Laboratory of the Johns Hopkins University . In a recent paper! I reported an attempt to obtain a modified clone of Hydra viridis by the continued selection of variates dif- fering in the number of tentacles. In that experiment a clone was bred from a single wild polyp and two groups of its descend- ants, each composed of 25 lines, were selected for variations in tentacle number in opposite directions from the mean of the clone. Selection was continued for several generations, then the number of tentacles of all buds produced by the last selected gen- eration of the 50 lines was recorded and the averages of the two groups were compared. A slight difference, in the direction of selection, was found in the averages of the earlier buds of the two groups but this difference did not persist in the buds pro- duced later by the same parents; regression was complete in a single generation. The chief criticism of such a negative result in a selection ex- periment is based upon the supposition that in a single character of a species some continuous variations are, others are not in- herited. If this is true, any selected individual may be either a germinal variate, in which case it will contribute to the progres- sive change in the racial character; a somatic variate, in which case its selection will not alter the racial type; or a variate in which both somatic and germinal variations are combined but active in opposite directions, thus leading to an effect of selection the reverse of that expected from the somatic character. It is further assumed that somatic variation follows more frequently than it contradicts germinal variation, giving natural selection an opportunity to produce evolution when applied to a large 1 Inheritance in the Asexual Reproduction of Hydra. Jour. Exp. Zodl., vol. 19, pp. 157-210, 1915. 19 20 K. §. LASHLEY number of individuals. When selection involves only a few in- dividuals the chance selection of a somatic variate or a composite variate of the third type above may counteract the effect of pre- vious selection. Only when enough individuals are selected to average out the composite variates can the results be relied upon. It should be pointed out that if this criticism is carried out consistently it results in practical experimental indeterminism, since it may be said of any experiment that it is much less ex- tensive than the course of evolution and is therefore not con- clusive. Perhaps the only answer that can be given to the criti- cism is a selection experiment extensive enough to make the chance reversal of selection improbable and throw the burden of proof upon those who advance the criticism. The selection ex- periment with Hydra has been prolonged until it now seems to meet this requirement. EXPERIMENTAL DATA Essentially the same technique was used as in the earlier ex- periment. Each polyp was kept in a separate Stender dish; the food supply, Cyclops, was distributed as uniformly as possible to all polyps; the dishes were cleaned and sterilized and the cul- ture fluid renewed every second day, except during the winter months when the polyps were not budding. New buds were removed from the cultures and recorded every second day. The first experiment, that previously reported, was brought to an end by the freezing of the food pond and for the ensuing three months very. little food could be obtained for the polyps. Asa result only 6 of the original 50 lines survived the winter, these being represented by members of the last selected generation. When food was again available all descendants of these polyps were kept until 24 members of each of the groups, varying in the required direction, were obtained. From these, 48 lines were established, half of which were selected in each direction from the mean number of tentacles of the entire clone. During the heat of the summer many lines died and were replaced from others of the same group so that at the end of the experiment only 20 CONTINUED SELECTION IN HYDRA 21 TABLE 1. Relationships and initial numbers of tentacles of all selected Hydras in the lines which survived throughout the experiment. The successive figures from left to right represent the numbers of tentacles of the parents in successive generations. The braces show where two or more offspring of a single parent were retained. The number of tentacles of the final generation from which the unselected buds were obtained is given at the extreme right. ses se sess 6-7-7-7-7-7 4 t 7 | “I | ~] a ! ~J —Ht= ih te ~ iva} 7-5-5-5-5-6-4-6-6-6-6-6-6-5-655. ccc eee cence eeenceeeees 5 { : EG BaGot 668-6 = ee... Go nabs alee 6 apres eS Sat t See elie ie. ae 6 ((G#8=6=6-6262626-65 3 a:b .cewy qin quaess don as Seekoactons 6 6-6-6-6-5-6-............. Be ae a G26=G=G=G=6=0—SR en cin: alsa « geass es ee 6 (5-62 6=5-6-626-6- 5-0-5 Guten Macon te = es Agee nh. ahaa ee Pe 6 LG eRe. 5 EM ee ee cote roe 6 mgm agi ak ase Geb eGe C= Ore © Meee. artis late Se ere ace 6 - Ele PAs eB AG 550-026 $8 Ae Re ee ame eee 6 Oe aeae ro Saree al g [FSB 56666. ee 6 5-6-5-4-6-6-6-6-6........ ((GBRi ete aac x: is tat taNn eis ee Way Rent PAI 6 6-5-6-6-6 -5-6-6-6-6 4 (DSO 5626)... neers aap? tree nes se ae 6 | 6-6-6 ee » okt WAAR oko Behr oy 6 GEORG. she om ah Wear een: ae ete 6 | eee a bogeeeaghy i>) ge Meee CCR aa ments ib esa ded 6 6-5-5-6-5-4-4-6-6-6-6-6-6-6-6 eee Pett are ee Me EE Bis Ao 6 6-6-6-4-6-6.............. 4 5-6-6-6-5-5-6-6-5-6 | 5-6-6-6-6-6-6-6. eee 6 (6-626-55655.02 Sa ea rece cate Sosy) eae eee RD 6 [ 5-6-6 4 f poe ee soe Roe Ras Bee a ered 6 [6-6-6-5-5 4 (eax eC RR 9 a ane cee tae 5 22 K. S. LASHLEY distinct lines of descent were represented. The complete his- tory of selection is given in table 1. When approaching winter again threatened a reduction of the food supply, selection was stopped and the initial numbers of tentacles of all buds produced by the last selected generation were recorded, until at least ten buds had been obtained from each of the 48 lines. These were bred under conditions as nearly uniform for the two groups as it is possible to make them for animals with such complex food habits as Hydra. The data ob- tained from these buds furnished the basis for a trial of the ef- fects of the preceding selection. The entire experiment extended over somewhat more than a year. After the test previously reported, selection was con- tinued for an average of 17.13 generations in the group selected for 6 or less tentacles and for an average of 9.38 generations in the group selected for 7 or more, an average of 13.25 generations for the two groups. The greater number of generations in the group selected for few tentacles is due to the fact that buds with few tentacles are produced early in each fraternity while buds with many tentacles appear only after the parents have fully matured. Before the first test an average of 6 selected genera- tions was obtained and this, with the 13.25 later selections, gives an average of 19.25 generations rigidly selected during the ex- periment. The average number of tentacles of all selected par- ents of the minus-selected group is 5.567, that of the plus-se- lected group is 6.888, giving an average difference of 1.321 ten- tacles per generation as the extent of selection. A total of 366 buds was obtained from the last selected generation of the group selected for 6 or less tentacles, and of 358 in the group selected for 7 or more. The distribution of variations in the initial num- ber of tentacles of these unselected buds is given in table 2. The constants determined from the table are: For the group selected for Mean o GYOTALESSH IS la een co ate 6.584 = 0.022 0.6408 tentacles For the group selected for PROT MOTE se eee ie 6.544 + 0.023 0.6569 tentacles ID iETENGEs nasser a. 0.040 = 0.081 CONTINUED SELECTION IN HYDRA 23 TABLE 2. Distribution of variations in the number of tentacles of buds produced by the last selected generation NUMBER OF TENTACLES NUMBER OF POLYPS 4 5 6 7 8 9 Ancestry with 6 or less tentacles 6 UCP See I) AO 1 366 Ancestry with 7 or more tentacles 1 12 WA AN ALG 358 The difference in the average number of tentacles (0.040) is not significant but the little that appears is in a direction the reverse of that to be expected if variations within the clone are inherited. In a consideration of diverse races the range and extent of vari- ation may be of equal importance with the mean. The standard deviations (co), measures of the amount of variability, are not significantly different for the two groups. The range of varia- tion must be considered in two ways; first, with respect to indi- vididual polyps, second, with respect to variation in the mean number of tentacles of different lines. As appears in table 2, the polyp with the fewest tentacles (4) appeared in the group se- lected for a large number, that with the most (9) in the other group. The range of variation of individual polyps, like the difference between the averages, is the reverse of that which should appear if selection were effective. The mean number of tentacles of all buds produced by each parent in the last selected generation is given in table 3, where the constants are arranged in the order of magnitude. Here again the line with the highest average belongs to the group selected for few tentacles and that with the lowest to the group selected for many, although the differences are not significant. Most theories of evolution by continuous variation assume that the range of variation increases as the mean of the selected group diverges from the racial mean, so that the average of the selected group may be brought beyond the original range of the race by the continued selection of extremes. In this experiment no wid- ening of the range was found: the extent of selection, the differ- 24 K. Si DASHEBY: TABLE 3. Mean number of tentacles of all buds produced during the time of cultivation by each parent of the last selected generation (last column, table 1). Not less than ten buds were obtained from each parent Se oe | 0S | arn a |? a (| 6.00 6.18 6.44 6.47 6.69 6.66 6.05 6.28 6.46 6.50 6.72 6.66 6.08 6.33 6.61 6.50 6.75 6.67 Mean number of 6.20 6.43 6.65 6.50 6.75 6.68 tentacles of buds. 6.29 Ge43 | Gs66 6260 ease: (Serie) 6.37 6.43 6.66 6.61 6.85 6.83 6.40 6.46 6.66 6.63 6.90 6.86 6.42 6.46 6.68 6.65 6.92) e693 ence between the selected members of the two groups, remained fairly constant throughout the experiment. DISCUSSION Clearly this experiment offers no positive evidence for evolu- tion by the accumulation of continuous variations. It remains to determine its value as evidence against such a method of evo- lution. This may best be done by a consideration of the sort of hereditary changes which such an experiment should reveal. The figure, 1.321, determined as the average extent of selection in each generation, makes it possible to compute roughly the theoretical effect of selection through any number of generations for any given strength of heredity, provided that selection of somatic variates does not interfere during the experiment. The number of buds obtained from the last selected generation is great enough to give a probable error for the difference between the groups of only 0.031 tentacles. From this, the probability that any expected difference between the selected groups would be revealed by the experiment may be computed. If we assume that the offspring inherit one-tenth of the variation of their par- ents, the theoretical result of 19 generations of selection at a difference of 1.321 tentacles per generation is a difference of 1.176 tentacles between the offspring of the last selected generations of the two groups. This is 38 times the probable error of the dif- CONTINUED SELECTION IN HYDRA 25 ference found, which means that the true value of the differ- ence could fall so far from the value given by the sample only once in some billions of samples; a certainty that no such differ- ence was produced by selection in this experiment. Even with a strength of heredity of only 0.01, nineteen generations of se- lection result in a diversity of more than 0.2 tentacle which is almost 7 times the probable error so that the chances are more than 400,000 to 1 that no such diversity would be obtained in any given experiment of the same magnitude. If all variations are inherited to the same extent, if selection of every variation con- tributes to evolution in proportion to the extent of the variation from the mean of the race, then the strength of heredity in Hydra must be considerably less than that represented by a coefficient of ancestral inheritance of 0.01. If not all variations are inherited, the present experiment may have failed to reveal a greater strength of heredity than this, but if even three-fourths of the variates selected were strictly soma- tic, a regression of 0.05 would almost certainly have been revealed: The inclusion of so many somatic variates would reduce the num- ber of effective selections to one-fourth and a regression of 0.05 should result in a diversity of 0.29 tentacle after five selections. If it is assumed that regression is as great as 0.30, which is approx- imately that ascribed to Hydra by Pearson, the percentage of variations which are inherited to this extent must sink below 10 per cent to escape detection by this experiment. These figures are of interest chiefly for a comparison with the theory of homotyposis and are sufficient to disprove, for Hydra, the theory as advanced by Pearson in 1901 and 1910,? and make it possible to state with assurance that if inheritance of varia- tions in number of tentacles occurs within the clone of Hydra it does not follow biometrical laws. For other theories, which hold that continuous variations are inherited but do not define the method or extent of such heredity, it can only be pointed out that if any hereditary variations occur in Hydra they must 2 On the Principle of Homotyposis. Phil. Trans. Roy. Soc. London, vol. 197, pp. 443-459, 1901. Darwinism, Biometry, and Some Recent Biology. Biomet- rica, vol. 7, pp. 868-385, 1910. 26 K. S. LASHUEN amount to less than one one-hundredth of the total variation. The difficulty of accounting physiologically for such a condition, upon other than the mutation theory, and the lack of any posi- tive evidence of such heredity suggest that the most tenable position is one which assumes that the continuous variations of Hydra are not inherited in asexual reproduction but are wholly the result of the interaction of the constant reaction-norm of the clone with a fluctuating environment. THE EFFECTS OF THE REMOVAL OF THE NASAL PITS IN AMBLYSTOMA EMBRYOS HAROLD SAXTON BURR From the Osborn Zoological Laboratory, Yale University FOURTEEN FIGURES (THREE PLATES) INTRODUCTION Many of the difficulties encountered in the study of the nervous system by the usual methods are obviated by experiments upon embryonic material in which nerve fibers are not yet developed and the blood Has not begun to circulate. In this way it is possible to observe the effect of the removal of an end organ upon the central nervous system without undue complication of the factors involved, and furthermore such a method makes possible the study of the formative influence one part may exert on another during ontogeny. The number of investigators who have applied this method is relatively small. In 1906 Braus extirpated the forelimbs of the larvae of Bombi- nator before the outgrowth of the brachial: plexus, with a view to determining the effect of the absence of the limb on the ventral horn of the spinal cord at the level of the brachial plexus. The experiment showed that there was at first no observable effect on the cord. The brachial nerves grew out into the surrounding tissue and ended more or less blindly. The size and number of the motor cells of the ventral horn was in no distinguishable manner changed. But this investigator found that when the operated larvae were kept alive until just before metamorphosis and then killed, a distinct reduction in the size of the cells of the ventral horn was discernible. Hence, he concluded, as a corollary to the theory proposed by Roux in 1885, that the devel- opment of the central nervous system was readily divisible into Cod Poll THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 FEBRUARY, 1916 28 HAROLD SAXTON BURR two periods, in the first of which growth and differentiation was independent of functional activity, but in the second of which functional activity determined whether or not growth would con- tinue. But as Forel had previously pointed out, the atrophy which made the distinction in size of the cells evident was an exceedingly slow process as evidenced by the fact that this dis- tinction did not occur until just before metamorphosis. Two other investigators may be mentioned here, who have worked along this line. In 1909 Shorey performed a similar series of experiments on chick and amphibian embryos. She found in the chick an almost immediate effect of the destruction of the limb bud on the size of the ventral horn at the limb level. But whereas Braus concluded that the neurones were self-differ- entiating and that the reduction in size was a secondary one due to the absence of function, Shorey argued that the neurones did not differentiate without the stimulus of function and environ- ment. 128 So far as it has been possible to ascertain, Dirken (11) is the only investigator who has attempted to carry this question further to a consideration of the effect of the absence of an end organ on the gross morphology of the brain. He found very extensive changes in the shape and organization of the brain of Rana fusca and R. esculenta when the fore or hind limbs were extirpated at an early age. The abnormalities affected not only the somatic motor areas, but also parts of the diencephalon and telencephalon, notably in the roof. As will be seen, the experiments with which the present report is concerned show no such fundamental upheavals. The last two named investigators, while using embryonic material, worked with individuals in which the peripheral nervous system had already partly developed, and it is suggested that the fundamental discrepancy between their results and those that follow may be due to this factor. From the experiments of Braus it is evident that there is a possibility that Gudden’s atrophy, when it can be induced, would make possible the tracing of a given system of nerve fibers from its peripheral ending to its cortical origin with all its collateral REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS 29 connections. The problem lies in making a lesion that will cause Gudden’s atrophy, and then in keeping the animal alive suff- ciently long for the atrophy to reach its height. In the spring of 1911, Dr. R. G. Harrison suggested that the removal of the nasal placode of Amphibia would be a practical method to attack such problems. Frog larvae were first used for the experiments but as Bell showed in 1907 the placodes often regenerated, a fact which would destroy the value of the material for this study. Bell found this regeneration occurred in a large number of cases, even where it was believed the placode had been entirely removed. This regeneration may have been due to the fact that the regenerating area about the placode was not removed. In Amblystoma, the placode can be easily extirpated. The nasal Anlage is readily distinguished from the surrounding tissue, and hence complete removal is a relatively simple matter. In the course of the experiments it was found that the extir- pation of the nasal placode made possible not only a study of the neurological problems involved, but also a part, at least, of the developmental mechanics of the skull. Since the material for this morphological work was kept alive for some months, a study of the reactions to food stimuli of the operated and normal forms was also undertaken. These three aspects of the problem form the body of the following report. Owing to the difficulty encountered in keeping operated forms alive through metamor- phosis, the report of the histological changes in the telencephalon as a result of the operations must be deferred until later. MATERIAL AND METHODS Amblystoma larvae 5-6 mm. long, were used for all of the oper- ations, of which two series were performed. In the first the nasal placode of the right side only was removed; in the second both were extirpated. The material obtained from the first series served as a basis for the morphological investigation—that from the second for the physiological experiments and as a check for the morphological study. 30 HAROLD SAXTON BURR The operations were performed with a pair of iridectomy scissors under the binocular microscope. During the operation, and for the subsequent twenty-four hours, or until healing was accomplished, the embryos were kept in a 0.2—0.4 per cent salt solution. They were then removed to individual dishes in which the water was kept fresh. When the larvae were ready to feed they were removed to aquaria, balanced largely with Lemna and Ceratophyllum. Records were kept of the history of each oper- ated specimen. Care was taken in all of the operations not to injure the under- lying forebrain, since any injury causes noticeable defects in the telencephalon. In the series of experiments under consideration a large num- ber of unilateral operations were performed. The material thus obtained was killed at frequent intervals, the oldest larvae being about six months old dating from the time of operation, and the youngest only a few hours old. The killing fluid used was subli- mate acetic. Ehrlich’s Haematoxylin and Congo-red were used to stain the 10» sections. Normal material was subjected to the same technique for controls. From the sections of the oldest larvae a wax reconstraction was made of the rostral part of the skull and brain by the Born method. The experimental investigations that have up to the present time been reported on Amphibia, have tended to show that the Anlage of the nose readily regenerates when parts are removed. Bell (07) goes so far as to assert that the complete extirpation of the nasal anlage in the frog does not prevent the regeneration of a nasal sac. In the few cases he reports, what be believed was complete extirpation resulted in every case in a regenerated sac, sometimes perfectly normal in size and structure, and in other instances, smaller than its fellow. The present series of experiments started with the extirpation of the nasal placode of Amblystoma. It became evident at once that careful and complete extirpation of the rudiment was not followed by regeneration. Of the two hundred and thirty-two operations performed, only four showed any external evidences REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS 31 of a regenerating capsule. Later study by means of sections added two more to the number. A number of experiments were then performed on the frog, em- bryo to see if there was any difference in the behavior of the nasal anlage in two such closely allied forms. Great care was taken to remove all of the nasal placode. It was apparent from the out- set that the placode of the frog is not nearly as compact nor so clearly differentiated from the surrounding ectoderm and not so easily separated from the underlying mesenchyme as it is in Amblystoma. Hence the complete extirpation is a much more difficult and delicate matter. External inspection showed twenty- two regenerating capsules out of seventy-six operations. Sections of a number of the remaining fifty-four taken at random showed an occasional abortive pit. It is evident then, that complete extirpation in the frog does preclude regeneration quite as com- pletely as in Amblystoma. It is possible that here, as in the regeneration of the limbs, lens and gills, there is a circumscribed area about the anlagen which on the removal of the'anlagen may regenerate it. Apparently in the frog this regenerating area is less restricted than in Amblystoma, if any such is present in the latter. REACTIONS OF NORMAL AND OPERATED LARVAE Turning now to a consideration of the experimental study of Amblystoma, it is obvious that by removing the nasal placodes of both sides previous to the formation of nerves, it is possible to obtain larvae which have never possessed functional olfactory organs. Thus is afforded excellent material for a comparative study of the reactions of noseless larvae and normal larvae to olfactory stimuli without the introduction of secondary factors due to shock or discomfort as a result of the operations. Parker (10) was the first to show conclusively that aquatic animals react positively to the olfactory stimulus of food. By suspending two bags in an aquarium, the one containing cheese cloth, the other bits of worm, he was able to detect distinct posi- tive reactions of Ameiurus to the bag containing the food mate- oo HAROLD SAXTON BURR rial. When on the other hand, the olfactory nerves were severed, no positive reaction occurred. Since then Copeland (12) has confirmed Parker’s results with other fish. Reese (712) was the first to attack the problem in Amphibia. Working with Diemyctylus, he showed that the adults would follow and snap at moving bits, whether food or not. They would also make characteristic snapping movements when beef juice was squirted over the external nares. No attempt was made to control the sense of sight. He tried to control the sense of taste, however, by introducing a bit of cotton soaked in cocaine into the mouth. The results were conflicting and inconclusive. Animals in which the olfactory nerves were cut failed to respond to stimulation. Copeland (’13) repeated the experiments of Reese, using more exact methods. Control of the visual sense was accomplished by stimulating the olfactory epithelium from a motionless source. By dividing the total reaction into two periods, during the first of which an approach was made to the source of the stimulus, and during the second, the object was snapped at or taken into the mouth, he came to the conclusion that the approaching re- action was due entirely to the sense of sight, the seizing alone to the sense of smell. The object of the following experiments was to test this prob- lem in Amblystoma, comparing the reactions of the normal larvae to food with those of the noseless. There are in general three groups of sensory organs which conceivably may receive stimuli from the source of olfactory stimulus. These are, in the order of relative importance, the eyes, the taste buds and the lateral line and general cutaneous systems. Stimulation of the latter systems can be eliminated by using a motionless source, since such stimulation is brought about by currents in the water. Such a source would at the same time give relatively little stimulus to the visual sense. Control of the stimuli of the taste buds is hardly practicable experimentally without rather serious operations on adult forms. Fortunately it is known that the sense of taste is operative over REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS 33 relatively limited areas and for concentrated fluids only, and hence the danger of stimulation may be mimimized by keeping the source of the stimulus at a distance from the mouth (Herrick 08) and Sherrington (’06). The most important sense, then that must be controlled, is the visual. This may be accomplished in two ways, (1) as stated above by keeping the source of the stimulus motionless and (2) by removing the optic vesieles at an early stage. In the study of the reactions of Amblystoma to olfactory stimuli, more particularly to the stimulus of food, three sources of stimuli were used, two of which were olfactory and the third purely optic. Pieces of freshly killed Amblystoma larvae and live entomostraca were used to stimulate the olfactory centers; grains of sand, which possess no powers of olfactory stimulus, for the optic centers. It was necessary to test as exactly as possible the visual reactions of the larvae since, as is evident to the most casual observer, under ordinary conditions, that sense is the most active in the capture of food. Under normal conditions where the food supply is abundant, individual larvae rarely move .about, but remain motionless until a small crustacean comes within strik- ing distance. The reactions of larvae under such conditions are quite characteristic. Resting motionless on the bottom of the aquarium, with body held above the debris and head elevated, it will, with a sudden contraction of the trunk muscles followed by a quick forward lurch, snap and engulf some particular crus- tacean whose movements carry it within striking distance. When on the other hand, the supply of moving food is reduced, the larvae will forage the aquarium for food. Under these conditions the reactions are quite as characteristic as in the well stocked aquarium. The young Amblystomas crawl slowly around the aquarium, nosing here and there. The attitude is strikingly like that of a dog following a scent, and suggests that now in the absence of food that is moving, the sense of smell is actively used. Support is given to this suggestion by the fact that the larvae may often be seen to snap up some bit of the debris. 34 HAROLD SAXTON BURR In the experimental study of this problem, sixteen normal and twenty-four operated larvae were used. Sixteen of the latter were noseless and eight were eyeless. These last were kindly loaned by Dr. Henry Laurens who performed -the opera- tions. As soon as the operated individuals as well as the nor- mal ones isolated at the time of the operation began to feed regularly, experimentation was begun. At this time the larvae are about 1 cm. in length. The yolk has completely disappeared, the gills are plume-like arching forward over the head, the fore- limbs are tridigitate, the rudiments of the fourth digit just beginning to appear, and the hind limbs are just noticeable as a slight elevation on either side of the cloaca. Experimentation with these young larvae was difficult because of the very great activity which they exhibited whenever there was any disturb- ance in the water.. For that reason the following report deals with somewhat older larvae, the age varying from one and a half to six months. The first problem was to determine the relative importance in the obtaining of food of the visual and olfactory sense. For this purpose grains of sand were used, dropped from a capillary pipette so as to fall within striking distance of the larva. A single individual was tested at a time. Four noseless and four normal larvae were subjected to these tests. The first were made when the larvae were five months old, dating from the operation. Another set of exactly similar tests were performed one month, later. The forty tests on the four normal larvae resulted in twenty reactions in which the sand grain was snapped at and engulfed, such reactions being designated as positive, and twenty in which no attention at all was paid to the sand. The percentage of positive reactions then, was fifty. Fifty-four tests on four nose- less larvae resulted in thirty-nine positive reactions, a percentage of seventy-two. The same eight larvae tested one month later gave for the normal larvae eleven positive reactions out of forty tests, or 28 per cent and for the operated twenty-two out of forty-four, or 50 per cent (table 1). REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS 35 TABLE 1 Reactions to moving sand grains CONDITION OF LARVAE REACTION NO REACTION eateries mel Normialommontisene os est ce ere ae 20 20 50 Wornndall GmoronGhsia = sh o8 2-3 o2)5 8. 2ean so: il 29 275 Waselesaimontig onsen. sn. ave shame tet 39 15 CALE PMGSGIESS OM OTB Lee ste ts acne. cee esses 23, 22 GO) It is quite evident then, that larvae, whether normal or oper- ated, will snap at almost any object which in any way simulates the movements of the food, no matter whether it stimulates the olfactory sense or merely the visual. An interesting fact was noted in connection with the reaction of the noseless and normal larvae to the sand grains. As will be seen in table 1, there is a marked diminution in the total number of positive reactions of the older larvae as compared with the younger. While the number of reactions is far from sufficient to serve as a basis for any definite conclusion, the difference in behavior suggests that the larvae gradually adapt themselves to the new situation. The second problem was to determine experimentally whether the olfactory sense played any part at all in the quest for food. Bits of beef were placed in the bottom of aquaria containing normal and operated individuals, The normal specimens would soon nose out the beef and engulf it; the noseless ones never. Exact records were not kept of these tests because the larvae did not thrive on the beef. They would often gorge themselves, much to their own detriment. Since entomostraca make up the bulk of their natural food, these were used in a series of tests to determine the reaction to olfac- tory stimulus. By careful handling with a capillary pipette, it was possible to deposit individual Daphnids two or three milli- meters from the head of the larvae, so that they would remain motionless on the withdrawal of the pipette. Ninety-five tests were’ performed on eight normal larvae; eighty-seven of these resulted in positive reactions, a percentage of ninety-two. One hundred and nineteen such tests performed on eight noseless 36 HAROLD SAXTON BURR larvae resulted in not a single positive reaction, the individuals would invariably swim off without paying any attention to the Daphnid, or after a time the entomostracan would move away. The above experiments were performed on larvae varying from one and a half to four months old, dating from the time of oper- ation, and indicate that these larvae do make use of the sense of smell in obtaining their food. This was further corroborated by the following tests on the eyeless forms (table 2). TABLE. 2 Reactions to motionless entomostraca r . Ty! r = ‘ PER CENT CONDITION OF LARV AE REACTION NO REACTION POSITIVE IN OT Stan oe lepers te pe rn ke OR! eee 87 8 91.6 INIOSe@1 OSS 4: Go tec ee ee te cree) ene 0 119 0.0 BiVi@leSSe iss cae Le Riy Pec aise tana Satan Sater 119 10 91.8 The eight eyeless larvae were subjected to the same tests. Out of one hundred and twenty-nine, one hundred and nineteen were positive, giving a percentage of ninety-two, a very close correspondence with the normal. Here the olfactory sense was the only one held in common by the normal and eyeless larvae that could receive stimulation. A curious phenomenon, to be investigated later, was observed in connection with the eyeless individuals. They were very much more sensitive to currents in the water than the normal larvae. GL ier RS 100. 10 Cross section 690 microns anterior to figure 9 showing bony plate connect- ing maxilla and premaxilla, the caudal end of the naso-lacrymal duct, the tectum and solum nasi and the ethmoid column. X 100. 11 Cross section 410 microns anterior to figure 9 showing the left telencephalon with its glomeruli and no sign of the right, the distal end of the naso-lacrymal duct and on the operated side the much thickened trabecula. X 100. ABBREVIATIONS ap, anterior trabecular plate nl, naso-lacrymal duct apm, ascending process of premaxilla ns, nasal sac bp, bony plate connecting premaxilla sn, solum nasi and maxilla t, trabecula ct, erista trabeculae : tc, tectum nasi ec, ethmoid column tel, telencephalon m, maxilla REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS PLATE 2 HAROLD SAXTON BURR PLATE 3 EXPLANATION OF FIGURES 12 Cross section of the forebrain of a unilaterally operated larva through the olfactory bulb showing the glomeruli and the absence of the olfactory bulb on the operated side. X 160. 13 Cross section of the forebrain of a normal larva at same level as figure 12 X 160. 14. Cross section of the same forebrain that is figured in figure 12, 17 microns caudally showing the first signs of the right telencelphalon. 160. PLATE 3 REMOVAL OF NASAL PITS AMBLYSTOMA EMBRYOS HAROLD SAXTON BURR o ec? & ee ee De®, ea, Od ee e '@ Ge, 4 Ge (7 I~ i's) ry THE PERIODIC REORGANIZATION PROCESS IN PARAMAECIUM CAUDATUM RHODA ERDMANN AND LORANDE LOSS WOODRUFF Osborn Zoélogical Laboratory, Yale University THIRTY-FIVE FIGURES (SEVEN PLATES) CONTENTS Via, Lava yO lier ee ae aan a eater Atta ee een ee i eee ee aa gt a 59 ee Matertalgan dem etliGusee sae mementos ors cen neneis caret tes haps ney - ena 60 III. The cytological phenomena of the reorganization process in Paramae- CUUMBCAUGUAL Uae peter eat role rarest tee ear 62 AGM DESCEMGIN OAD NASC! ete Ba-iae eee Meh et ae en i os coals aOR TN 62 Irs 2ap © littrvatee Stet ds Se ee se Merny hs a Pee Be eT Bere, Pen erScee Tee Tete 69 @ArAScending pase: 25 4! hc aepee eens ee 0) ott nice RIGO Weve Ste ee ee 72 IV. Comparison of the cytological phenomena of the reorganization proc- ess n Paramaecium caudatum and Paramaecium aurelia.......... 76 W., IDIsewEsIGmS Hine) GoMOlWMOMS., os ccucaccnasvocouceuocHeceoosuasbouEeue 78 Wale GeratUnercibed Assan eee eee cee Cac GLe Sets? Klaas AOS Ensaio ae ee 83 WATE xpolanartioneo fl atestas see cewere cits aeesctcihs ote etersioe Gatto actions aici rece 84 I. INTRODUCTION In previous papers! there have been presented the morphologi- eal and physiological details of a new reorganization process in the life of Paramaecium aurelia, and the statement that the phenomenon occurs also in Paramaecium caudatum. The present paper is a study of this reorganization process, which we term endomixis, in the latter species of Paramaecium. We ‘Woodruff and Erdmann (’14, I): Complete periodic nuclear reorganization without cell fusion in a pedigreed race of Paramaecium. Proc. Society for Experimental Biology and Medicine, vol ii, Feb. 18. (Preliminary paper.) Erdmann and Woodruff (714, II): Vollstindige periodische Erneuerung des Kernapparates ohne Zellverschmelzung bei reinlinigen Paramaecien. Biol. Centr. Bd. 34. (Preliminary paper). Woodruff and Erdmann (’14, III): A normal periodic reorganization process without cell fusion in Paramaecium. Journal of Exper. Zoélogy, vol. 17, no. 4, Nov. (Complete paper.) 59 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 60 RHODA ERDMANN AND LORANDE L. WOODRUFF shall in this paper confine ourselves chiefly to a description of the facets observed and to a very brief discussion of the theoretical bearings, leaving for consideration at another time, when the details of this reorganization process in other species of Protozoa have been discovered, the more or less academic discussion as to the exact classification of this process among phenomena of Entwicklungserregung. However, we may state at once, in view of the somewhat premature criticism by R. Hertwig of our work as presented by us in an avowedly preliminary paper (714, II), that we are convinced, after such a careful consideration as any paper by Hertwig demands, that not a single objection which he raises is well founded. We are also convinced that the theoretical interpretations which Calkins (715) advances in regard to our results are not warranted by the facts so far at hand. We take, then, in their entirety the facts and discussion as presented in our complete paper ('14, HII) as the starting point for our present communication. Il. MATERIAL AND METHODS The material used in the present study has been derived from pedigreed races of Paramaecium caudatum which have been bred in exactly the same way as described in our work on Para- maecium aurelia, and the reader is again referred to that paper for details (14, UT, p. 432). From our study of several races of Paramaecium caudatum it seems clear that this species is less well adapted than Para- maecium aurelia to withstand the conditions necessary for pedigreed culture work. Whereas all the races of Paramaecium aurelia which we have studied have survived indefinitely under the daily isolation slide method of pedigreed culture manipu- lation, most of the Paramaecium caudatum races have sooner or later refused to divide under these conditions. As will be discussed in detail later, the races on the slides can undergo the reorganization, perhaps two or three times and then at the next onset of the phenomenon, about 90 generations later, are unable to carry it to proper completion and die. If, however, the PERIODIC REORGANIZATION OF P. CAUDATUM 61 animals are placed in tiny tubes of culture medium, instead of on slides, so that the volume of culture medium is somewhat greater, but not large enough to render it impossible to control them daily and so detect conjugation if it occurs, then the ani- mals apparently can live indefinitely. This result is, in general, in accord with the earher observations of Woodruff (’11¢, p. 64) though at that time he used somewhat larger volumes of medium to save the Paramaecium caudatum and so was not able to positively prove that conjugation did not occur. It has seemed best to consider this point in justification of our resorting to material from tiny tube cultures for certain stages of the reorganization process. But we would emphasize that many of the stages of this process in Paramaecium caudatum, as all of the stages of the process in Paramaecium aurelia, have been secured from daily isolated pedigreed animals so that it is absolutely positive that the phenomenon under discussion occurs in caudatum as in aurelia at clearly defined periods. To repeat— we have shown, by the isolation slide method, the physio- logical characteristics of the reorganization process in the re- current rhythms of the division rate, and the accompanying cytological changes of this process to the climax at the low points of the rhythms. The completion of the process in isolated slide animals is evident from the completion of the rhythms in many cases. But owing to the longer rhythmic periods in Paramaecium caudatum as compared with Paramaecium aurelia, and owing to the lesser viability of the former species and the smaller number of generations over which its reorganization process extends, we did not secure all of the later cytological changes from isolated slide animals but resorted to carefully controlled ‘tube’ animals as already described. We will grant that the original discovery of the reorganization process in Paramaecium caudatum would have been more difficult than it was in Para- maecium aurelia because of the longer rhythmic period in the former species and its less marked viability under the arti- ficial slide pedigreed method during the crucial stages of the phenomenon. 62 RHODA ERDMANN AND LORANDE L. WOODRUFF Ill. THE CYTOLOGICAL PHENOMENA OF THE REORGANIZATION PROCESS IN PARAMAECIUM CAUDATUM The complete reorganization process without cell fusion which occurs in Paramaecium caudatum, as in Paramaecium aurelia, synchronously with the rhythms, presents essentially the same fundamental cytological features as those we have described in the latter species (714, III, pp. 436-473). We shall, therefore, present merely a general summary of the phenomenon in Paramaecium caudatum and point out the differences which could be traced in the morphological features. In accordance with our previous investigation we term the three Rhythm Ryhthm Process Text fig. 1 Diagram illustrating the relation of the reorganization process to rhythms. (Woodruff and Erdmann, ’14, ITT.) phases into which the reorganization process (endomixis) natu- rally resolves itself as descending phase, climax and ascending phase (text figure 1). A. The descending phase. The micronucleus of a_ typical Paramaecium caudatum which is not undergoing the reorgani- zation process or normal vegetative division is situated in a slight depression of the macronucleus. Text-figure 2 (p. 63) shows an animal shortly after a typical vegetative division in which the micronucleus has not yet reached its accustomed posi- tion. The structure of the resting micronucleus between two vegetative cell divisions is shown in figure 1, p. 85. The posi- tion of the micronucleus in figure 2 has not changed but its PERIODIC REORGANIZATION OF P. CAUDATUM 63 morphological structure is characteristic.of beginning reorgani- zation. As this phenomenon approaches the micronucleus fails to take up its accustomed place, and remains free in the cy- toplasm, while the chromatin of the micronucleus seems to assume the form of short bands. Figures 3 and 4 show two animals in the 208d generation from different sub-lines of Cul- ture Y just when the process of reorganization begins—the formation of bands of chromatin, and the wandering of the Text fig. 2. Culture Z, Line I, 38th generation, October 12, 1914 micronucleus in the cytoplasm, indicating the definitive onset of the process. This wandering of the micronucleus has been described by us in the reorganization of Paramaecium aurelia Cala sie ties 23, 5) Projections of the macronucleus which extrude at a very early stage of the process of reorganization in Paramaecium aurelia have not been seen in our races of Paramaecium caudatum. 64 RHODA ERDMANN AND LORANDE L. WOODRUFF Here the macronuclear surface loses its smooth appearance while clefts and wrinkles appear in the stained preparations thus indicating that in the living cell the distribution of the chromatin is not homogeneous in the macronucleus (figs. 4 and 7). The micronucleus begins to swell shortly after emergence from its depression in the macronucleus. Figure 5, shows the chro- matin arranged at the equator of the micronucleus in a band of granules while one pole is nearly devoid of chromatin. (Compare Calkins and Cull ’07, fig. 1). The swelling continues until the so-called sickle stage appears. Figure 8 shows a more elong- ated micronucleus than the one shown in figure 5. The for- mation of the ‘sickle’ is completed in an animal from the 202d generation, Culture Y, Line III g (fig.°7). The paucity of the chromatin in the portion of the micronucleus which protrudes from under the macronucleus is remarkable; no distinct granules or threads being recognizable as the chromatin is apparently homogeneously distributed throughout the dividing micronucleus at this stage. The formation of chromatin bodies has begun in the macronucleus. Maupas, Hertwig and Hamburger de- scribed, in the different species of Paramaecium which they investigated, the occurrence of the ‘sickle’ stage before the onset of the reduction divisions preceding conjugation. The ‘sickle’ stage changes in conjugation into the first reduction spindle as described in detail by Calkins and Cull, ’07, p. 383, in Paramae- cium caudatum. In the reorganization process of Paramaecium caudatum we secured an animal from the 184th generation with the very characteristic dumb-bell form of the micronuclear divi- sion spindle (fig. 6). The ‘dumb-bell’ micronuclear division stage in the reorganization process resembles closely the stage which Calkins and Cull (’07, fig. 15) interpret as the telo- phase of the second maturation division in conjugation. In the ~ reorganization process of Paramaecium caudatum there is a slight difference in size of the arising micronuclei as is indicated in the cell shown in figure 6. The amount of chromatin at the two ends is apparently different. It is possible that the arising micronucleus which is nearly devoid of chromatin is the one whose products after the next division are doomed to degeneration. PERIODIC REORGANIZATION OF P. CAUDATUM . 65 Calkins and Cull do not describe a ‘dumb-bell’ formation in the first reduction division in conjugation, but we must interpret our figure as the first ‘reduction’ division of the reorganization proe- ess because no trace of other micronuclei are in the cell. An animal in the second ‘reduction’ division is shown in figure 10. Two micronuclei are dividing. In one the equatorial band of chromatin is intact while in the other it is divided. The macro- nucleus, fragmentated into two parts, is rapidly losing its chro- matin. The breaking up of the macronucleus begins relatively late in Paramaecium caudatum and generally does not extend through several generations as we found to occur in the descending phase of the process in Paramaecium aurelia (Compare ’14, IIT, pl. 1). The expulsion of chromatin bodies (figs. 14, 15 and 16) or the breaking of the macronucleus into two or more pieces (figs. 9, 10 and 11) seem to be a rapid process in the life of the caudatum cell. Culture Z, 91st generation (fig. 13) does not show the slightest trace of chromatin bodies, but in the 93d generation (fig. 14) there are many of them in the cytoplasm. These chro- matin bodies, however, do not have the definite condensed ap- pearance so characteristic of those in Paramaecium aurelia, but are relatively pale and indistinct and seem to suddenly appear in nearly maximum numbers. This possibly may be accounted for by the fact already mentioned that in Paramaecium caudatum finely divided chromatin material, in certain cases at least, streams from the macronucleus into the cytoplasm. This leaves a pau- city of chromatin for the chromatin bodies and may change the tension within the macronuclear membrane so that the chromatin bodies are formed rapidly and thus appear suddenly. Figure 13 gives a good idea of this phenomenon, as does also figure 6 in which the macronucleus is partly devoid of chromatin while no chromatin bodies are visible. This breaking up of the macronucleus into large pieces was observed in animals from small tube cultures (fig. 10) as well as in isolated animals (figs. 9 and 11) as already described. These animals cannot be dividing specimens because in normal vege- tative divisions the macronucleus is greatly elongated and breaks 66 RHODA ERDMANN AND LORANDE L. WOODRUFF only at one point when the cytoplasm itself has nearly completed division (14, ITI, fig. 7). To summarize: The loss of chromatin occurs either by the extrusion of chromatin bodies (Paramaecium caudatum and Paramaecium aurelia); by extrusion of small granules from the macronucleus (Paramaecium caudatum); by breaking up of the old macronucleus into two or more large pieces (Paramaecium caudatum; Paramaecium aurelia, Hertwig ’89, p. 74). The result—the total destruction of the individuality of the macro- nucleus—is the same in each ease. Extruded chromatin bodies were figured by Calkins ’04 (fig. 16) in Paramaecium caudatum and these we have compared with certain stages of Paramaecium aurelia at the beginning of the ascending phase of the process of reorganization ('14, ITT, p. 484, text figs. 20 from our own preparations, and 19, copied from Calkins). Calkins does not mention how the nuclear fragmen- tation has been effected but this probably has occurred in the same way as we have described in our cultures of Paramaecium aurelia (Woodruff’s main culture I, and Erdmann’s culture B) and in some animals from our culture Y of Paramaecium cauda- tum (714, II, fig. 34), because the morphological features are identical. Those cultures of Paramaecium caudatum, in which the macronucleus breaks into large pieces, present in the further stages of macronuclear degeneration a different cytological appearance as 1s shown in an animal from about the 82d gene- ration, culture Y (fig. 8). Here the chromatin is distributed in the cell without first being condensed into spherical chromatin bodies. The macronuclear membrane is torn and its contents are intermingling with the cytoplasm. This general type of fragmentation of the macronucleus has been described with different interpretations by several authors, for example, Kasan- zeft 01, working under R. Hertwig; Popoff ’07, figure 18, plate 4; Popoff ’09, figure 26, plate 2; Calkins ’04, figure 8, plate 1 and figure 11, right animal, plate 2; and Hertwig 714, p. 568. All these authors considered these changes as depression phenomena which, according to Popoff, have close resemblance to phenomena characteristic of the onset of conjugation. PERIODIC REORGANIZATION OF P. CAUDATUM 67 Now Hertwig 714, reviewing and commenting on our prelimi- nary paper (14, IT) on the reorganization process of Paramaecium aurelia, in the light of his theoretical beliefs in regard to depres- sion in Protozoa, shows in his text figure 3, animals 4 and 5 (copied as text fig. 3 in this paper) the breaking up of the macro- nucleus together with the formation of two micronuclei. These stages correspond somewhat to our stages (figs. 8, 9, 10 and 11) at the end of the descending phase of the reorganization process when the first or second ‘reduction’ division has occurred. But it is clear that Hertwig has confused two different series of phenomena: animals with hypertrophied macronucleus and not Text figs. 3 and 4 Paramaecium caudatum. From Hertwig, ’14, after Kasanzeff. clearly discernible changes in the micronucleus, with those of the reorganization process which we have described as endomixis (ef. text fig.3, copied from Hertwig). We have never found in the thousands of isolated animals studied that the macronucleus was enlarged by any means to such a degree as Hertwig figures in his text figure 3, animals 1, 2, 3, from -vhich we have copied the second animal as our text figure 4. The crucial stage, i.e., the total reorganization of the nuclear apparatus with the formation of the macronuclear anlagen, could not have been determined as a phase of the reorganization 68 RHODA ERDMANN AND LORANDE L. WOODRUFF process either by Hertwig, Kasanzeff or Popoff in their mass cultures. However, Hertwig’s suggestion that ‘‘ Woodruff und Erdmann k6énnten somit die Stadien als Beweise fiir eine vor- ausgegangene Parthenogenese in Anspruch nehmen ”’ is a partially right prediction, provided one substitutes for ‘parthenogenesis,’ reorganization process—endomixis! Since Kasanzeff did not find stages of the ascending phase, that is the formation of the macronuclear anlagen, in Paramaecium caudatum mass cultures, Hertwig still beheves that the described phenomena are depres- sion conditions and not stages of the reorganization process, though he himself described in Paramaecium aurelia isolated stages of a process which he called ‘parthenogenesis. ”” We wish to emphasize that in a mass culture it is impossible to know the ancestry of the individual cell. It is clear that from a mass culture so-called depression stages may be either animals which are about to begin the reorganization process, or animals which have just undergone conjugation, or abnormal ani- mals due to some exigencies of the environment. From material of this sort, in view of the fact that the reorganization was not known previous to our work, there has been drawn into the litera- ture on infusorian life histories the long series of atypical animals as evidence of depression. However, by pedigreed culture methods we have resolved this heterogeneous material into its component parts and have shown for both Paramaecium aurelia and Paramaecium caudatum that many of the so-called depression phases are normal stages in a complex reorganization phenomenon. B. Climax. Culture Y of Paramaecium caudatum we had under observation from February 18, 1914, to June 9, 1914, and > We have not called the reorganization process ‘parthenogenesis’ and have not introduced even our new term ‘endomixis’ in our preliminary papers (14, I and ’14, II) because we wished to wait until our complete data was presented (14, III) so that they could be discussed in their entirety. However, in our brief paper (14, IT) an error is present on page 495, owing to the fact that the printer misunderstood our directions and that we were unable to revise the proof on ac- count of the war. The sentence ‘‘ Diese Parthenogenese hier ist ein Sexualakt.”’ is a remnant of a paragraph which appeared in the manuscript but was entirely deleted in the proof, but the printer in some way left this one isolated sentence for the complete paragraph! PERIODIC REORGANIZATION OF P. CAUDATUM 69 during this time the culture attained the 207th generation. The process of reorganization occurred at about the 18th, 136th and 203d generation (table 1). Line IIT which had successfully undergone the process at the 136th generation lived to the 192d generation, just at the onset of the next reorganization period. That the phenomenon was imminent is evident also from Sub- line ITT fa, in which it appeared at the 202d generation. From TABLE 1 Showing the occurrence of the reorganization process in Culture Y PERIODS = pea ests a | REMARKS I 1=5O)36.0.0¢ Died INE, ISNO72o56c 18 Died IIT. 1-192..... 136 191 Died ira 20=130 25-75. it i039 Killed in descending phase NOE Io, EB EIGR ows Died JUL @; ISIN 3 i356 Killed JNO Gh, WGIAUG ES o5 3 | Killed JME @5 WGC 3c: Died JOE ti, WG coe 137 Kulled Miia lG9—203 55s. 202 Killed in descending phase IIT fb: 169-206: .... ¢ 203 Died MEL ge 192-2072... | 204 Died LV): 1-89 ..... 26 | Killed Eira. © 85-105... | | Killed EVi-aa: 93-143... ... 139 | Killed in ascending phase IV ab. 108-114..... | Died IN fe, WAV SB Gee a. 1Sy «| | Killed in descending phase *Designated Line 2 in our 1914 (IIT) paper. a close analysis of the division rate, and of the longevity of the component lines of Culture Y (table 2) it is apparent that of 4 lines and 13 sublines, 13 did not undergo a division for at least twenty-four hours, and in some cases for forty-eight hours, be- fore their death. Death occurred in these lines either ‘naturally’ or as a result of the animals being killed for study since they were dividing slowly. Two lines or sublines died naturally without a lowering of the division rate, ten were purposely killed and four died after a lowering of the division rate. Sub-line III a, 70 RHODA ERDMANN AND LORANDE L. WOODRUFF TABLE 2 Culture Y LINE GENERATIONS DIVISIONS FATE JER en SR eeA Ree rE S eets niga Bc 59 None in 48 hours Died 11 eee ee a Reet Oe rh ee ie od 107 Two in 24 hours Died 1G rine aol medal carn apt) hit v6 192 None in 24 hours Died ] Gt ET er ea Meee ren oh A tery Ute ee Pratt 7 139 None in 24 hours Killed |B) Bo Seer epee er oR oe Pe me let 161 One in 24 hours Died 10 bl Wraps Raa hoe rele Oa es eos 163 None in 24 hours Killed | Bl Uc Naa, eA ett ek eum ee Ae aN ae 165 None in 24 hours Killed ) Gd it ACen te en ease coset ata 169 Two in 24 hours Died NE) ES a aenee, cee erence ee tenet ree Sent 184 None in 24 hours Killed TTB El es es ee eee ALT Gal ry dati ces 202 None in 24 hours Killed efile: Bate ete endothe 206 None in 24 hours Died ISTE oe eee saranda 8 ey Nae he a Pe 207 None in 24 hours Died Lice cert eens a rae eee Ho 89 | None in 24 hours Killed WEE EON oe ee en ore a Re kei 115 None in 24 hours Killed 1 EERE den Ie echt a Pike et een Seen eta 143 One in 24 hours Killed JENVESEN| Ove A AOTC os teen ee AN A 114 None in 24 hours Killed IVa Che Aan eos ee ee wa Ow A eee Mi | 137 None in 24 hours Killed Ill fa and IV ac, were killed in the descending phase of the process. Sub-line IV aa was killed after the climax. Thus although cytological indications of the reorganization process were not observed in certain of the lines and sublines of Culture Y at the time of extinction, nevertheless this extinction was synchronous with the process as actually observed in other lines and sublines from the 136th to 148d generations, 184th to 191st and 202d to 209th generations. These periods clearly are critical ones in the life of this culture (table 2). However our culture of Paramaecium aurelia was able to undergo the reorganization process frequently and with apparent facility under daily isolation methods, and therefore the number of lines which were eliminated by death during the phenomenon was relatively small (cf. 714, III, table 1, p. 462). From all our work the conclusion evidently follows that before cytological signs of reorganization are discernible in the cell the physiological conditions for its onset are in evidence. Culture Z was bred from September 20, 1914, and Culture M from January 7, 1915, to the end of our study. Culture Z under- PERIODIC REORGANIZATION OF P. CAUDATUM (il went reorganization successfully in the 11th, 91st, 184th, 282d and 398th generations, and Culture M at the 89th generation, that is at intervals of about 90 generations, thus confirming our observation on Culture Y that the reorganization process occurs in Paramaecium caudatum at intervals of from S80 to 100 generations. The difficulty of securing stages of the climax and of the as- cending phase compelled us, as previously stated, to amplify our material of isolated pedigreed animals with small tube cultures seeded from our pedigreed cultures, just before reorganization was due to take place on the basis of our computation of the TABLE 3 Showing the occurrence of the reorganization process in Culture Z PERIODS= aa Banca 180—270 270—360 360—450 REMARKS I 1-378 11 90 188 282 Kulled Ila. 93-165 Tulled Ib. 144-187 184 Killed Te. 142-201 187 Kulled I ea. 171-260 Died Id. 192-201 Killed Ie. 275-454 Discon’d If. 394-455 | 398 Discon’d rae 1-92 91 | | | Killed number of generations since its last occurrence and by the divi- sion rate. The most critical stage for the vitality of the race is when the individuality of the macronucleus is lost and the chro- matin bodies are undergoing degeneration. The results of Calkins’ extended experience with Paramaecium caudatum all indicate that the culture of Paramaecium caudatum in isolated lines on depression slides is sooner or later fatal. Woodruff’s comparison, (’1l¢, p. 60,) of pedigreed lines of Paramaecium caudatum with those of his main culture of Paramaecium aurelia showed clearly that it was impossible to breed indefinitely Paramaecium caudatum by the daily isolation method which was so highly favorable for Paramaecium aurelia. 24 RHODA ERDMANN AND LORANDE L. WOODRUFF The following tube cultures therefore were started from the respective isolated slide cultures as indicated above: | Sab e eu G ais cata ee Nr tes ota oi. March 26, 1914 Coles a Tubecculiturejbit rece yo caee ic ood tats otra opal ile 1914 y 4 Cube GUltuneces.aewaee rs ee etre el oe Ate eee Mabe cultures ceaep a wmere ws A eae tere ae April 20, 1914 (Mules culliuneser. Aol ah. alos ert cian saacel April 22, 1914 Culture’ = (ube cultunerastae ce. sence: oe ane March 12, 1915 ZL, Robeaeullitimedy aa ti.aeva ts eres ok ee oe ele March 13, 1915 Continuing the discussion of the cytological changes of the reorganization process, fig. 18 shows an animal in the climax, from Culture Y about the 187th generation, without a complete macronucleus and with degenerating chromatin bodies as is indicated by their elongated form. This shape of the degenerat- ing bodies we have frequently mentioned in our description of the climax and ascending phase of Paramaecium aurelia (714, IIT, pl. 2). The definitive micronucleus (fig. 18), which by three subsequent divisions will form eight micronuclei from which the new nuclear apparatus will be formed, is characterized by its lucid appearance and is lying above the main remnant of the macronucleus. This cell has a perfectly healthy appearance and should be contrasted with the animal shown in figure 17, which is an animal from Line UIT g, 202d generation, which several generations later died in the process of reorganization. Paramaecium caudatum cells with one micronucleus and no macronucleus have been observed by Calkins, Kasanzeff and Hertwig. Calkins ’04, fig. 14, pl. 2, illustrates such an animal. Abnormal animals of this character must not be confused with animals undergoing the reorganization process. The cell with- out a complete macronucleus (fig. 18), which we figure, is an en- tirely normal animal. Uts sister cells carried on the race for nearly 70 generations more, some of them showing the crucial formation of macronuclear anlagen as will be described in detail in the next section. C. Ascending phase. Before the onset of the ascending phase the same general cytological changes are to be observed as those PERIODIC REORGANIZATION OF P. CAUDATUM he we have described for Paramaecium aurelia, 1.e., animals with only one micronucleus and a completely destroyed macro- nucleus, the remnants of which are scattered in the cell. This one micronucleus, as in Paramaecium aurelia, is the carrier of the life of the race through the new rhythm. Maupas and Calkins in the conjugation of Paramaecium cau- datum figure eight micronuclei, the products of the synearyon, which are formed immediately after the separation of the con- jugants. This we have verified in conjugation (fig. 25). In the reorganization process we have discovered animals at a slightly later stage with four micronuclei and four macronuclear anlagen, the latter representing four transformed micronuclei. The micronucleus which persists and carries the life of the race there- Ma. Text fig. 5 fore must have undergone three divisions producing the four micronuclei and four macronuclear anlagen in the stage under discussion. The micronuclei (fig. 19) are characterized by their granular structure and somewhat glistening appearance. In the preparations the few granules appear pure blue without any reddish tinge. These micronuclei resemble, as far as one may judge from figures, one of the four micronuclei which Klitzke (14, text fig. C, copied in the present paper as text fig. 5) shows in the conjugation of Paramaecium caudatum. This micronucleus (Klitzke’s ‘Micronucleusanlage,’ d.k. 3) he believes undergoes degeneration together with two other micronuclei (d.k. 1, d.k. 2). The marked differences he points out are not observable in our specimen in the reorganization process, but this may be due to the fact that the animal has not yet reached this point. Or it may be that no micronuclear degeneration 74 RHODA ERDMANN AND LORANDE L. WOODRUFF occurs, as is the case in the description of this phase in the conjugation of Paramaecium caudatum as worked out by Cal- kins and Cull (’07, p. 396). We summarized in our former paper (14, III, pp. 453, 454) the various opinions in regard to the for- mation of the macronuclear anlagen in Paramaecium, and showed that the problem is not settled simce Maupas and Klitzke believe that four micronuclei and four macronuclear anlagen arise, three of the micronuclei undergoing degeneration. Maupas states definitely that four typical micronuclei are transformed into macronuclear anlagen. In the animal in the process of re- organization under discussion (fig. 19) this transformation of micronuclei into macronuclear anlagen has been completed. The anlagen are quite homogeneous, with the exception of a few granules frequently dispersed in a circle. In _ preparations stained with Delafield the anlagen show a diagnostic reddish tinge which cannot be expressed without colored plates. (Com- pare our present plates with those reproduced in colors in our paper on Paramaecium aurelia). The chromatin bodies are scattered throughout the cell in various stages of disintegration. The next two figures (20 and 21) show only two micronuclei and four macronuclear anlagen in each animal. This stage may be interpreted as an animal in which two micronuclei have degenerated, or as an animal after the degeneration of three micronuclei and after the first somatic micronuclear division. In our work we found it necessary to study the cytology of conju- gation of our pedigreed races of Paramaecium caudatum with particular reference to the fate of the four micronuclei which do not form macronuclear anlagen. In our races it appears clear that all four of the micronuclei do not persist and become dis- tributed by the following two cell divisions as the definitive micronuclei of the four completely reconstructed Paramaecium cells. Figure 26 shows an exconjugant with clearly four macro- nuclear anlagen and four micronuclei. Figure 27 gives a pedi- greed animal, after the first cell division subsequent to conju- gation, in which only one micronculeus is present. Figure 28 shows a pedigreed animal, after the second cell division subse- quent to conjugation, with a macronucleus and micronucleus. PERIODIC REORGANIZATION OF P. CAUDATUM I This is in harmony with the stages in the reorganization process in our races having four anlagen and two micronuclei (figs. 20 and 21) which follow without a cell division the stage with four macronuclear anlagen and four micronuclei (fig. 19). The reconstruction in the reorganization process of the typical vegetative Paramaecium cell is now effected by two cells divisions. Figure 22 shows an animal which has undergone both of these divisions. The cell has a single well developed anlage which has attained more typical macronuclear characteristics. The chromatin bodies are rapidly degenerating. The reorganized cell at the very beginning of the new rhythmical period is shown in figure 23. A eritical survey of the ascending phase shows that even in our small tube cultures, which might be compared with Kasan- zeff’s culture methods (except that he starved his animals while we attempted to supply ideal conditions), relatively few Para- maecium caudatum were able to accomplish the reorganization process. This is in agreement with Popoff’s results because he found, according to our interpretation, only the early stages of the reorganization process. Calkins, though studying care- fully his pedigreed cells, figures but one animal which we would interpret as showing the completion of the reorganization proc- ess (14, ITI, text figs. 19 and 20), Hertwig has never observed what he considered a reorganization process in Paramaecium caudatum (’14, pp. 568, 569). We ourselves, using the same methods which proved successful in Paramaecium aurelia and also new ones adapted to the peculiarities of Paramaecium cau- datum in culture, could not work out in such detail the uninter- rupted sequence of endomictic events from pedigreed series of cells as in the case of Paramaecium aurelia. Nevertheless we have proved that the process of reorganization can be success- fully accomplished in certain cases under pedigreed slide con- ditions though it is clear that these artificial slide cultures afford obstacles which the average caudatum cell finds it difficult to overcome when in the critical climax of the reorganization proc- ess or of conjugation. We believe the data presented establish beyond doubt that the reorganization process is anormal periodic event in the life history of Paramaecium caudatum. THE JOURNAL OF EXP£RIMENTAL ZOOLOGY, VOL. 20, No. 2 76 RHODA ERDMANN AND LORANDE L. WOODRUFF IV. COMPARISON OF THE CYTOLOGICAL PHENOMENA OF THE REORGANIZATION PROCESS IN PARAMAECIUM CAUDATUM AND PARAMAECIUM AURELIA This short outline of the cytological changes of the reorgani- zation process in Paramaecium caudatum makes it clear, we believe, that there is no fundamental difference between the morphological features of this process in Paramaecium caudatum and Paramaecium aurelia, further than that incidental to the fact that the former species has one micronucleus and the latter two micronuclei in its typical vegetative stages.? The destruction of the old macronucleus and the formation of a new macronuclear apparatus of micronuclear origin is effected in both species. However, there are some interesting minor variations which it may be well to contrast. The ‘reduction’ division in the re- organization process of Paramaecium caudatum with its ‘dumb- bell’ formation resembles more closely the phenomenon in the conjugation of this species than do the features of the ‘reduction’ micronuclear phenomena in the reorganization process of Para- maecium aurelia resembles those of conjugation in that species. We were unable to discover the same features in the ‘reduction’ division in Paramaecium aurelia during the reorganization proc- ess that were described by Hertwig for the comparable stages in conjugation. Hertwig ('14) figures a stage (text fig. 2, animal 3) which he interprets, together with the condition in the two pre- vious animals (animals 1 and 2), as ‘‘ Depressionserscheinungen von Paramaecium aurelia.’”’” We would interpret Hertwig’s animal 3, on the basis of the animals figured in our earlier paper (fig. 12, pl. 1, ete.) as a cell in a stage of the reorganization proc- ess after the formation of four ‘reduction’ micronuclei two of which are already preparing for the second ‘reduction’ division. Animals with this morphological structure will actually complete the reorganization process, as we have proved (’14, [II), and therefore such animals as figured by Hertwig cannot be inter- preted as depression stages. Thus Hertwig indirectly and we * For a discussion of the specific characters of Paramaecium aurelia and Para- maecium caudatum and the literature on the subject, ef. Woodruff, Journal of Morphology, vol. 22, p. 223, 1911. PERIODIC REORGANIZATION OF P. CAUDATUM Ch directly prove that the ‘reduction’ micronuclei in Paramaecium aurelia in the reorganization process have somewhat different morphological characteristics from those of the same species in conjugation. The destruction of the macronucleus before the formation of macronuclear anlagen in Paramaecium aurelia occurs, according to our observations, only by the extrusion of chromatin bodies from the macronucleus. But in Paramaecium caudatum there are clearly two methods of macronuclear dissolution in the de- scending phase of the reorganization process. One of these involves the breaking up of the macronucleus into one or more large pieces which finally degenerate in the cell (figs. 8, 9 and 10); the other involves the extrusion of chromatin bodies from the more or less intact macronuclear membrane (figs. 14, 15 and 16). This would seem to indicate that there may be also two distinct methods of accomplishing the ascending phase of the reorganiza- tion process in Paramaecium caudatum. One closely resembles the process in Paramaecium aurelia, being characterized by the presence of chromatin bodies, the absence of the old macro- nucleus in a cell with one micronucleus, this single micronucleus being the one which is to form the new nuclear apparatus. The other method is characterized by the breaking of the macronu- cleus into relatively large pieces and the streaming of chromatin out into the cytoplasm (figs. 10 and 13). Thus when the macro- nuclear destruction takes this form, there are few, if any, chro- matin bodies to be found in the ascending phase. It may be mentioned incidentally that we have some data which suggest that under certain conditions merely a partial reorganization, not involving the formation of macronuclear anlagen, may lead, at least temporarily, to the continuance of the life of the line. Our material did not allow us to prove whether the third so- ealled reduction division occurs or not, but we lean to the view, on the basis of our detailed study of this point in Paramaecium aurelia (14, U1, pp. 446-450 and p. 495) that this division actu- ally isabsent in the reorganization process in Paramaecium cauda- tum as in Paramaecium aurelia. Further, we have no indication 78 RHODA ERDMANN AND LORANDE L. WOODRUFF of a cell division in the climax, and this point must be definitely settled for Paramaecium caudatum by further investigation. Thus, although we have presented sufficient data to establish the occurrence of the reorganization process—-endomixis—in Paramaecium caudatum, we have found no new fundamental facts to modify our brief theoretical suggestions as given in our earlier study of Paramaecium aurelia. We believe the sug- gestions—as there stated—must stand or fall on the basis of further study of endomictic phenomena in other Protista. V. DISCUSSION AND CONCLUSIONS We proved in our previous paper that a periodic reorgani- zation process, to which we gave the name endomixis, occurred periodically throughout the seven years of the life of the main culture of Paramaecium aurelia. We showed in subcultures, from this main culture, in which conjugation was allowed to occur that lines derived from exconjugants underwent endomixis at the regular intervals. We thus proved that endomixis and conjugation are phenomena common to the same race of Para- maecium aurelia. We showed further that endomixis occurred in a race of Para- maecium aurelia (culture B of our former paper) isolated in Germany. On the basis of this we stated (’14, U1, p. 494, and 14, III, p. 474): ‘‘ Therefore, the data justify the conclusion that this reorganization process is a normal phenomenon and proba- bly occurs in all races of the species Paramaecium aurelia.”’ But since Hertwig intimates that endomixis is probably a peculiarity of Woodruff’s main culture, we may cite further evidence to substantiate our former conclusion. We have had oceasion, for certain experiments, to secure other races of Para- maecium aurelia. One of these was obtained from material sent to us by Prof. R. A. Budington of Oberlin, Ohio, and the other from material sent by Dr. Florence Peebles from Bryn Mawr, Pennsylvania. These two races, taken at random from material collected at widely separated localities, immediately showed endomixis at periods similar to those of the races a!veady studied. PERIODIC REORGANIZATION OF P. CAUDATUM 79 Thus endomixis has now been demonstrated in each of the four races which we have studied. Further, the idea of Hertwig that endomixis occurs only after long cultivation of a race of Paramaecium was shown not to be true in Woodruff’s main culture. We stated (14, II, p. 492) that animals preserved during the first year of its cultivation showed stages of the process, and further we stated that the race from Berlin showed stages of endomixis very early in its history (14, II, p. 493). Now, with this point in mind, we have found endomixis in each of the new aurelia races within the first thirty days of their life in culture. Therefore, we have proved that endomixis is a phenomenon common to all four races of Paramaecium aurelia which we have studied and thus it is highly probable that it occurs in all races of this species. Further, we have proved that endomixis is not a phenomenon which is gradually acquired after long pedigreed culture but is completely developed in animals at the time of isolation from wild cultures, and still further we have proved that endomixis is a potentiality of lines which have the power of conjugation (714, III, p. 473). Having, we believe, disposed of these questions which have been raisedin regard to our work on Paramaecium aurelia, we are 1D a position to return to endomixis in Paramaecium cau- datum. The cytological phenomena of this process have been presented in the previous sections, and we believe that we estab- lished beyond peradventure the truth of our statement (14, I, p. 475) that endomixis ‘‘occurs at least with essentially similar features in Paramaecium caudatum also’’ and therefore that endomixis is a regular normal periodic process in the life of Paramaecium caudatum (text fig. 6). Early work on Woodruff’s main culture of Paramaecium aurelia (09, 711) showed that there are periodic fluctuations (rhythms) in the rate of reproduction which are not the results of environ- mental variation, but which are due to some periodic internal phenomena of unknown character (Woodruff and Baitsell, ’11). We have shown (’14) that endomixis is the underlying internal process whose physiological effect had been observed but whose 80 RHODA ERDMANN AND LORANDE L. WOODRUFF nature had only been suspected (714, III, p. 480), and stated (14, III, p. 481) ‘‘Therefore, it is evident not only that the reorganization process is coincident with the low points between two rhythms, but also that there is a causal relation between the reorganization process and the rhythms.” We have now established this same conclusion for Paramaecium caudatum. In our races of this species, however, the endomictic periods appear at intervals of about 50 to 60 days, or from 80 to 100 generations, instead of from 25 to 30 days, or 40 to 50 generations as in Paramaecium aurelia. But they are funda- mentally the same morphologically and physiologically in both species. Le ol i Text fig. 6 Graph of the rate of division of Culture Z, Line I, averaged for five-day periods. The periods during which the reorganization occurred are indicated by a X. Cf. page 71, table 3; and also Woodruff and Erdmann, ’14, III, text figures 16 and 17. Now a critical examination, in the light of our present knowl- edge of endomixis, of the division rate of Calkins’ pedigreed culture of Paramaecium caudatum, the study of which led him to his well known conception of protoplasmic old age in Protozoa, shows clearly that his periods of degeneration can be perfectly interpreted as endomixis (text fig. 7, copied from Calkins). Woodruff in an early paper (’09, p. 300) wrote: I have previously interpreted as rhythms the tri-monthly depres- sions In vitality, which Calkins and earlier workers on Paramaecium have noted, and the results obtained from my culture of Paramaecium seem to indicate that the semi-annual cycles of Calkins are also actually rhythms, recovery from which was not autonomous under the con- ditions of a constant environment. The general occurrence of rhythms in the life history of infusoria is established, I believe, but to what they are due is still awaiting discovery. PERIODIC REORGANIZATION OF P. CAUDATUM Sl In other words the present studies on Paramaecium caudatum show that most if not all the depression periods of Calkins are un- doubtedly rhythms and the ‘cycle’ is non-existent—it is merely, as stated above, a rhythm at which the organisms are unable to recover autonomously by endomixis owing to the more or less artificial culture methods imposed in daily isolation pedigreed cultivation (text fig. 7, copied from Calkins, ’04). It may be well, in view of the recent comments by Hertwig (14) and by Calkins (15) in their discussion of our paper on endomixis, to state the position of the problem of depression and 1902 A APR) HAY | JUNE y JULY » AUO 4 SEPT. | OcT | NOV, Dec -ae ere i u rH RN OFHaAXTESonnusue dy eooreoere oe Text fig. 7 ‘‘History of the A Series from start (Feb. 1, 1901) to finish (Dee. 19, 1902) by ten-day periods (three periods toeach month). The ordinates rep- resent the average daily rate of division. The heavy dotted lines indicate the limits of the several cycles, and the lines of the curve carried to the base indicate that the individuals that were not stimulated by change of diet died out. The points at which such lines leave the curve indicate the time of the successfully changed diet.’’ (Calkins, ’04, p. 426.) the significance of conjugation in Paramaecium in 1907 when Woodruff began his work on Paramaecium. The consensus of opinion of biologists, chiefly on the basis of the work of Maupas, Calkins and Hertwig, was that infusoria are able to reproduce by division for only a limited number of generations, after which protoplasmic old age, depression, and physiological death ensue. For this the sole panacea was conjugation. But Woodruff found that by supplying proper environmental conditions it was possible to breed a pedigreed race of Paramaecium aurelia indefinitely (so far, April 1915, more than 5000 generations) without recourse to conjugation. Therefore, he concluded, in direct opposition to Maupas, Calkins 82 RHODA ERDMANN AND LORANDE L. WOODRUFF and Hertwig, that conjugation is unnecessary for the indefi- nite life of Paramaecium under favorable environmental con- ditions. To Hertwig’s and Calkins’ recent contentions that their con- clusion was correct and Woodruff’s was wrong, we would reply, that the reorganization process (endomixis) is not conjugation and no one had any other phenomenon than conjugation—in- volving synearyon formation—in mind until Woodruff and Erd- mann discovered endomixis—in which a synearyon is not formed. To say that endomixis fills essentially the same roéle in the life history of the infusorian as conjugation is to beg the entire question. Had Hertwig worked out the stages which he found over twenty-five years ago, or realized their general significance, and had he related these with rhythms ten years ago when they were discovered, the problem would have been cleared up then. The aspect of the problem has changed with the discovery of endomixis. The question is now not whether conjugation is necessary—for Woodruff has shown that it is not—but whether endomixis, a new phenomenon which Woodruff and Erdmann have shown to exist, 1s necessary. Whatever the answer afforded by future work to this new question may be, it is clear that conjugation—and by this always was meant, and always is meant, the formation of a synearyon— is not necessary, since an individual Paramaecium is self-suffi- cient to reproduce indefinitely without recourse to conjugation. VI. LITERATURE CITED (For a more extensive bibliography, ef. Woodruff and Erdmann, 1914 (IIT), pp. 498-502. ) Cauxins, G. N. 1904 Death of the A-Series of Paramaecium caudatum. Conclusions. Jour. Exp. Zodél., vol. 1. 1915 Cycles and rhythms and the problem of ‘immortality’ in Paramaecium. Amer. Naturalist, vol. 49. Cauxins, G. N. anp Cutt, 8S. W. 1907 The conjugation of Paramaecium cau- datum. Arch. f. Protistenk., Bd. 10. ErpMANN, Ru. 1908 Kern- und Plasmawachstum in ihren Beziehungen zu einander, Ergeb. d. Anat. u. Ent., Bd. 18. ERDMANN, Ru. AND Wooprurr, L. L. 1914 Vollstiindige periodische Erneuerung des Kernapparates ohne Zellverschmelzung bei reinlinigen Paramae- cien. Biol. Cent., Bd. 34, 8. PERIODIC REORGANIZATION OF P. CAUDATUM 83 HamBureGemr, C. 1904 Die Konjugation von Paramaecium bursaria. Archiv. f. Protistenk., Bd. 4. Hertwic, R. 1889 Ueber die Conjugation der Infusorien. Abh. der k. bayer. Akad. d. Wiss., Cl. II, Bd. 17. 1914 Uber Parthenogenesis der Infusorien und die Depressionszu- stiinde der Protozoen. Biol. Cent., Bd., 34, 9. KASANZEFF, W. 1901 Experimentelle Untersuchungen iiber Paramaecium cau- datum. Inaug.-Diss. Ziirich. Kuirzke, M. 1914 Uber Widerconjuganten bei Paramaecium caudatum. Archiv. f. Protistenk., Bd. 33. Maupas, E. 1888 Recherches expérimentales sur la multiplication des Infu- soires ciliés. Arch. d. Zool. expér. et gén., 28., T. 6. 1889 La rajeunissement karyogamique chez les Ciliés. Arch. d. Zool. expér. et gén., 2S. T. 7. Poporr, M. 1909 Experimentelle Zellstudien, III. Arch. f. Zellforsch., Bd. 4. Wooprurr, L. L. 1909 Further studies on the life cycle of Paramaecium. Biol. Bull., vol. 17. 1911 (a) Two thousand generations of Paramaecium. Arch. f. Protis- tenk., Bd. 21. 1911 (b) Paramaecium aurelia and Paramaecium caudatum. Jour. Morph., vol. 22. 1911 (c) Evidence on the adaptation of paramaecia to different environments. Biol. Bull., vol. 22. 1912 A summary of the results of certain physiological studies on a pedigreed race of Paramaecium. Biochem. Bull., vol. 1. 1914 So-called conjugating and non-conjugating races of Paramaecium. Jour. Exp. Zo6l., vol. 16. Wooprurr, L. L. anp Barrsett, G. A. 1911 Rhythms in the reproductive activity of Infusoria. Jour. Exp. Zo6l., vol. 11. Wooprurr, L. L. anp ERpMANN, Ru. 1914 Complete periodic nuclear re- organization without cell fusion in a pedigreed race of Paramaecium. Proc. Soc. for Exp. Biol. and Med., vol. 11, Feb. 18, 1914. 1914 A normal periodic reorganization process without cell fusion in Paramaecium. Jour. Exp. Zodél., vol. 17. VII. EXPLANATION OF PLATES All the figures represent specimens of Paramaecium caudatum. The animals were fixed in Schaudinn’s sublimat-alcohol and stained with Delafield’s hematoxy- lin. The drawings were made from total preparations, with Abbe camera lucida, Zeiss homogeneous immersion 2 mm. and compensating ocular 12, with drawing board level with stage of microscope. Magnification, about 1500 diameters. Reduction in reproducing plates 1, 2, 3, 4, 5 is one third; plates 6 and 7 is one half. PLATE 1 EXPLANATION OF FIGURES 1 Culture Y, Line IV, 85th generation, April 3, 1914. Normal animal in a period when endomixis is not in progress. 2 Culture Y, Line IV ae, 136th generation, April 28, 1914. Typical animal just at the start of endomixis. 3 Culture Y, Line III fa, 203d generation, June 8, 1914. Animal just in a very early stage of endomixis. Macronucleus wrinkled. Micronucleus has shifted from its typical position close to the macronucleus. 4 Culture Y, Line IIT fb, 203d generation, June 8, 1914. Animal ina slightly more advanced stage than the one shown in figure 3. 5 Culture Y, small tube culture, about 137th generation, April 22, 1914. Start of the first ‘reduction’ division. 6 Culture Z, Line IV, 184th generation, December 24, 1914. Macronucleus partly devoid of chromatin. First ‘reduction’ division nearly completed. 84 PERIODIC REORGANIZATION OF P. CAUDATUM PLATE 1 RHODA ERDMANN AND LORANDE L. WOODRUFF L. Krause del. PLATE 2 EXPLANATION OF FIGURES 7 Culture Y, Line III g, 202d generation, June 7, 1914. Markedly wrinkled macronucleus and one chromatin body. Part of crescentic stage of micronucleus shown protruding from under the macronucleus. 8 Culture Y, small tube culture, about the 82d generation, April 1, 1914. Macronucleus broken into several large pieces. Elongated micronucleus, one pole partly devoid of chromatin. 9 Culture M, Line II, 89th generation, February 19, 1915. Completed fragmentation of macronucleus into two parts. 10 Culture Y, small tube culture, about the 137th generation, April 22, 1914. Animal with macronucleus breaking into two parts. First ‘reduction’ micro- nuclear division completed. 11 Culture M, Line I, 90th generation, February 19, 1915. Macronucleus broken into two pieces from which chromatin is streaming and forming chromatin bodies. Two micronuclei are present, one of which is drawn. The presence of numerous food vacuoles is due to transferrence of this animal to rich hay infusion medium previous to killing. 12 Culture Y, Line IV, 26th generation, March 2, 1914. Two ‘reduction’ micronuclei are present. S6 PERIODIC REORGANIZATION OF P. CAUDATUM PLATE 2 RHODA ERDMANN AND LORANDE L. WOODRUFF = es. Re e L. Krause del. PLATE 3 EXPLANATION OF FIGURES 18 Culture Z, Line I, 91st generation, November 5, 1914. Illustrating the method of macronuclear dissolution by streaming out of chromatin. Micro- nucleus shows the characteristic band formation of early endomixis. 14 Culture Z, Line I, 93d generation, November 6, 1914. Macronuclear dis- solution by elimination of chromatin bodies. Single micronucleus close to macronucleus. live food vacuoles shown in the cell. 388 PERIODIC REORGANIZATION OF P. CAUDATUM PLATE 3 RHODA ERDMANN AND LORANDE L. WOODRUFF 14 L. Krause del. 89 PLATE 4 EXPLANATION OF FIGURES 15 Culture Y, Line IV, 18th generation, February 26, 1914. Descending phase. One micronucleus and some chromatin bodies visible. 16 Culture Y, small tube culture, about 137th generation, April 22, 1914. Descending phase. One chromatin body is in the macronucleus, others have left it. 90: PERIODIC REORGANIZATION OF P. CAUDATUM PLATE RHODA ERDMANN AND LORANDE L. WOODRUFF 16 L. Krause del. 91 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, No. 2 PLATE 5 EXPLANATION OF FIGURES 17 Culture Y, Line III g, 202d generation, June 6, 1914. To illustrate that essentially the same stage as that shown in figure 14 from Culture Z occurred in another Culture Y in isolated animals. 18 Culture Y, small tube culture, about 187th generation, April 22, 1914. Animals in the climax of endomixis. Macronucleus partially resolved into de- generating chromatin bodies. Micronucleus after ‘reduction.’ PERIODIC REORGANIZATION OF P. CAUDATUM RHODA ERDMANN AND LORANDE L. WOODRUIF PLATE 5 wee nt deers ope ae ad L. Krause del. 93 PLATE 6 EXPLANATION OF FIGURES 19 Culture Y, small tube culture, about 137th generation, April 22, 1914 Animal in ascending phase of endomixis. Four micronuclei and four macro- nuclear anlagen present. Numerous chromatin bodies in all stages of degene- ration. 20 and 21 Culture Y, small tube culture, about 82d generation, April 1, 1914. Animals in ascending phase. Two micronuclei and four anlagen are present. 22 Culture Y, small tube culture, about 82d generation, April 1, 1914. Animal in the ascending phase, showing one anlage. 23 Culture Y, small tube culture, about 82d generation, April 1, 1914. Para- maecium cell immediately after undergoing endomixis. PERIODIC REORGANIZATION OF P. CAUDATUM PLATE 6 RHODA ERDMANN AND LORANDE L. WOODRUFF L. Krause del. PLATE 7 EXPLANATION OF FIGURES 24 Culture Z, March 27, 1915. Exconjugant. Division of synearyon. 25 Culture Z, March 27, 1915. Exconjugant. Macronucleus show typical ribbon formation. Micronuclear divisions preliminary to formation of macro- nuclear anlagen. 26 Culture Z, March 27, 1915. Exconjugant. Four macronuclear anlagen and four micronuclei. 27 Culture Z, April 4, 1915. Individual from the first division of exconjugant. 28 Culture Z, April 5, 1915. Individual from the second division after conju- gation. PLATE 7 PERIODIC REORGANIZATION OF P. CAUDATUM WOODRUFF RHODA ERDMANN AND LORANDE L. L. Krause del. 7 > a 4 @ «af 7 % oe 7°49 ee ne : / Pe Ss © A , . - : a os 7 3 C ~~ ‘ ‘ ayo aah ne : ; ‘i, sie as oo i hee Q x ay AN ANALYSIS OF THE PROCESS OF REGENERATION IN CERTAIN MICRODRILOUS OLIGOCHAETES LIBBIE H. HYMAN Hull Zcélogical Laboratory TWENTY-FOUR FIGURES CONTENTS em Venterce lee yc ete om tan prs Rte Ne Ye se MDE Bas Lut sic, Aas in lacey 100 Mee Mireraxdtaloradve mln ters gost ais mek so oe ounce ie eres, oe ee eS 101 ite hexevanide amet hOdis se sseste se SNe ac cero oe ieee so eee 101 Qe eaprimlaryaChAdiemtys antics Seite cele cin hae tee eh ee 105 Se hevoradient sol they al ase ders oper ee hice oc ead claps Be whe eek 112 f the sradient. of iLumpbriculusimconstans oo... ac. 2c 4. son Sessa « 119 ormlihe,eradienteor thettubiic1ds: . 2:5. cane se oe ou ae eae ee 120 Galihereradientginkemibryos se Nene: ae ae. Be Jods dle ccc s doe 120 TES ASIULATNETED Ne Bes OM ec IE AD as ct eS I OPES nS eh ee 123 tleeieheroatarOrmeseneratlOnnet ee gi. te. tct. 50 eso td, ree a Ler eda, ee 124 feherhead region: of oligocttitetes=n:..c:. 202s dso 5 acne Mesos 124 2. General considerations on regeneration in oligochaetes............ 126 Sve Senerallonnn. WErO tMGs ade sk tke, eis eigcis Sawin nie etry ee elke 130 4. Regeneration in Lumbriculus inconstans.......................-<- 133 Je LeMeLatlOm Ui COE GUDTICIOS er 28. 5 's.55 3 oe bon ce Wea eve 141 GMS UIE AT Vee eet tic eee at ate rv ale sti Sasa sac ts os Wed ana casey a Bee eh ae 142 Vie Aitahysis, Of the re reneratLye PROCESS... hc. 6 cons 6G. lee Sana ces eae aoe 148 iPahe-cime (of head. G eterminatiONh. 6.0.9.4. sees eis 2 lee ae 143 PAM VE LOMO ys SE CLIO Ca Maye: cea n s.2 oe ay Sere Bckner aa snetoion esa eee 145 3. General conception of the process of regeneration................ 150 WAS: PSNUIOT TAA eA RON a a Ne SA a a LA a Rm eRe eee Sr a 158 Wileee sto lioomerp hive ce: an tae cet ere ns etre ob a iO Fo 4 yc ea eee soo MG The importance of regeneration as an experimental method has not been sufficiently recognized. The numerous investi- gations which have been carried on in this field have given us a vast amount of data regarding the capacity for regeneration of all groups of animals, the axial differences in rate and amount of regeneration, the histology of the process, and so on, but little progress has been made in the interpretation and analysis of the facts observed. Child (11d) has pointed out the fundamental biological significance of regulatory phenomena, and has shown 99 100 LIBBIE H. HYMAN that important problems may be attacked by means of this relatively simple experimental method. Under Professor Child’s direction, I have been carrying out experiments along similar lines on several species of microdrilous oligochaetes; this paper is a partial report of the results of these investigations. I. MATERIAL The following species have been used in these experiments: Aeolosomatidae Aeolosoma hemprichii Ehrenberg. Naididae Dero limosa Leidy. Dero furcata Oken. Stylaria lacustris Linnaeus. Chaetogaster diaphanus Gruithuisen. Nais elinguis Miller. Lumbriculidae Lumbriculus inconstans Smith. Tubificidae Tubifex tubifex Lamarck. Limnodrilus claparedianus Ratzel. Aeolosoma was obtained from ordinary Protozoan cultures such as are made up for class use. The naids were collected from old ponds and streams, especially Wolf Lake, Indiana, and the Des Plaines River near Lyons, Illinois; quantities of mud and vegetation were brought in from such places, and put into large crystallizing dishes. I have not found it practicable to keep any naids but Dero for any length of time in the laboratory, but this form is readily cultivated, and thrives and reproduces rapidly if a small amount of fermenting grain is added to the culture. The tubificids were collected in the same places and in the same way as the naids; they can be kept for months under laboratory conditions, and are also benefited by the addition of grain to the culture. Lumbriculus inconstans, unlike the other fresh-water oligochaetes, is very restricted as to habitat, occurring, in the Chicago vicinity at least, only in temporary PROCESS OF REGENERATION 101 forest pools of a characteristic ecological type (see Shelford, 14, pp. 179, 185). This species is active, therefore, only during the spring months when the pools contain water, since it passes into an encysted condition when the water evaporates with the approach of summer.t Mrazek (713) states that such pools form also the typical habitat of Lumbriculus variegatus, and his further remarks on the ecology of this species apply equally well to Lumbriculus inconstans. Il. THE AXIAL GRADIENT 1. The cyanide method The most important general fact which has come out of all the work on regeneration is this;—that the result differs according to the level along the antero-posterior axis at which section is made. This difference may be one of rate or of amount of new tissue formed, or, in the most interesting cases, the structure which appears at the cut surface varies according to the level of section. Such an axial difference is obviously the expression of a preexisting internal gradient of some sort,—a_ protoplasmic gradient independent of the more obvious morphological features of the organism. This gradient must undoubtedly have both structural and functional components, for structure and func- tion always interact; in the living organism they can have no separate existence. But it is the functional component alone with which I propose to deal in this discussion of the axial gradient. What means have we for demonstrating a functional, dynamic gradient—or, to put it more simply, a gradient in metabolic processes—along the axis of the living organism? Child (13 a) has devised and extensively employed a simple method depending on the use of lethal concentrations of anaesthetics and cyanides. In the presence of such substances—and of many other poisons also—metabolism cannot continue, and if there exists a meta- bolic gradient in the organism, its parts will show a different ‘A complete account of the life cycle of Lumbriculus inconstans in relation to its peculiar habitat will be published elsewhere. 102 LIBBIE H. HYMAN degree of susceptibility, or, conversely, a different degree of re- sistance to the action of these substances. This will be especial- ly true if the metabolic difference is one of rate, for those parts in which metabolism is going on most rapidly must be most susceptible to substances which stop metabolism; moreover, we have no reason to assume that the fundamental metabolic processes are different in kind along the axis of a simple organism. The method is, therefore, within certain limits, specifically a method for demonstrating differences in metabolic rate between the parts of an organism, or between comparable organisms. Regarding the exact way in which substances containing the eyanogen radical and anaesthetics act on protoplasm, there is at present much difference of opinion. It would be fruitless to review here the numerous theories which have been advanced with regard to the action of anaesthetics (Gwathmey,’14, Chap. IT) among them probably the Verworn conception of anaesthesia as an asphyxiation has been received with the most favor. The most recent evidence does not, however, support this view. Thus Loeb and Wasteneys (13 a, 713 b) and Winterstein (’14) have measured the oxygen consumption under anaesthesia, and have found that narcosis may occur without any decrease in oxygen consumption, or usually only a shght one, or even a slight increase. Winterstein (713, 714) also calls attention to many other data leading to the conclusion that the depression of oxidation is not the primary factor in the production of anaesthe- sia, as Verworn maintains. On the other hand, Tashiro and Adams (14) have reported a decided decrease in CO, production in the anaesthetized nerve of the spider crab. Further experi- mental work is much needed before any conclusion regarding . the nature of anaesthesia can be drawn. More agreement exists as to the effect of the eyanides on liv- ing matter. Since the work of Geppert, it has been generally accepted that the cyanides act by diminishing or inhibiting the oxidation processes. Geppert (’89) showed that an animal poisoned by hydrocyanic acid was consuming less than the usual amount of oxygen, even though such an animal goes into violent convulsions; and further, that the oxygen content of the venous PROCESS OF REGENERATION 103 blood in cyanide poisoning is abnormally high. Geppert there- fore concluded that cyanogen reduces the capacity of the cells to use oxygen. This conclusion is supported by the experiments of Warburg (710), of Loeb (06 a, ’06 b, ’10 a, 710 b), and of Loeb and Wasteneys (710, 11, 718 a, 13 b), who found that the oxygen consumption of eggs and other cells is decreased in the presence of cyanides; of Schroeder (’07) who obtained the same result with Aspergillus; and of Vernon (06) who showed that the oxy- gen consumption of a perfused kidney is diminished when cyan- ides are present in the perfused fluid; and I myself, working with sponges, have recently demonstrated to my entire satisfaction that the oxygen consumption of these animals is invariably lowered when potassium cyanide is present in the sea water, N even in concentrations as low aS 35, 9. —The cyanides also have a general depressing effect on enzymatic processes, and on many chemical actions, oxidative and otherwise. On the other hand, there have not been wanting objections to the idea that the cyanides inhibit oxidations; principal among these is the state- ment that the cyanides are equally poisonous to tissues and organisms which are not affected by the absence of atmospheric oxygen ,as the nerve cord of the Limulus heart (Carlson, ’07), and anaerobic bacteria. In reply to this, it may be suggested that the series of chemical processes which we call oxidation is probably much the same up to a certain point in both aerobic and anaerobic organisms, and it isthese initial reactions which the cyanide affects (Matthews, 705). Direct evidence that the susceptibility to cyanide runs parallel with the rate of metabolism has been furnished by Child (138 a), who has experimentally determined the following facts: 1. Animals stimulated to motor activity are more susceptible than quiescent ones. 2. The susceptibility increases with rising temperature; and the temperature coefficient of susceptibility is the same as the usual temperature coefficient for chemical reactions in general. 3. Other forms of stimulation (as injury, cutting the animal into pieces) increase the susceptibility to cyanide. 2 These results will be published shortly. 104 LIBBIE H. HYMAN 4. Young animals are invariably more susceptible to cyanide than old ones. 5. The CO, production of pieces or animals runs parallel with their susceptibility to cyanide, 7.e., those more susceptible to cyanide show also a more rapid CO, production. The CO: production was determined by Dr. Tashiro with his very ingeni- ous apparatus for measuring minute amounts of carbon dioxide (Tashiro, 13): According, then, to the facts here presented, the time of death in cyanide bears a direct relation to the previous rate of metab- olism. Individuals or parts with the highest rate of metabolism die first, those with the lowest rate last, and the others at inter- mediate times; the cause of this, as already stated, is to be sought in the asphyxiating action of the cyanides, as a result of which the time of death of each part is proportional to its rate of oxy- gen consumption. It is only necessary that the death point should be clearly indicated to the observer; this is brought about in the lower invertebrates through the disintegration which promptly follows death. As a check on one’s observations, one may remove the animal or piece at any stage of the disintegration to water, whereupon recovery of the intact parts takes place. In this way, I have satisfied myself that disintegration follows death almost instantly. The cyanide method is, of course, not applicable to forms in which, owing to resistant outer structures, disintegration cannot occur; but even in such forms, the death of the animal as a whole may usually be determined by employing some other criterion of the death point. The technique of the cyanide method is simple. A concen- tration of potassium cyanide sufficient to kill the animals within one to three hours is used. This concentration must be deter- mined for each species by preliminary experiments. For the oligochaetes it varies from ;!, to 7,)!)) normal. The cyanide solu- tion is made up fresh by weight for each experiment. Animals or pieces which are to be compared as to susceptibility must have been kept under the same conditions of food, temperature, etc., previous to the experiment, and must be of approximately the same size, unless size differences are the object of the experiment. PROCESS OF REGENERATION 105 Wherever susceptibilities are compared, I have always done so at the same time, and with the same cyanide solution, thus avoiding sources of error arising from differences in the solution, external conditions, ete. I have found it most convenient to carry out the experiments in watch glasses. The animals are placed in these, freed as much as possible from water, and a cover put on in such a way as to exclude all air bubbles. In such covered watch glasses evaporation of the cyanide is reduced to a minimum, and the whole can be placed under the low power of the compound microscope, and the progress of the disinte- gration followed very exactly. The following changes take place in cyanide. The worms at first move about vigorously but eventually pass into a state of anaesthesia. The peristalsis of the intestine and especially of the dorsal blood vessel keeps up as a rule until the time of death. Frequently there is a swelling of the animal, due to the intake of fluid into the coelomic spaces, so that the body wall is distended, except at the septa, and the animal then resembles a string of beads. This condition appears to be a regular ante-mortem change since it also occurs in pieces dying in water. The death point is characterized by an abrupt change from the normal yellowish-red color of the oligochaetes to an opaque white; this change of color is quite apparent to the naked eye, and under the microscope can be followed from segment to segment. Simul- taneously with or immediately following this alteration of color, the body wall breaks (it may previously have shown blister-like elevations), and all the structures disintegrate into a shapeless mass of granules. This disintegration, as already indicated, does not occur simultaneously throughout the animal, but pro- ceeds in a perfectly definite manner along the antero-posterior axis. This disintegration gradient of the various oligochaetes will now be described in detail. 2. The primary gradient The kind of gradient which Child has described for Protozoa, Coelenterates and flatworms (Child, ’13 ¢, 14 a, 14 b) constitutes what I call here the primary gradient. In these forms, disinte- 106 LIBBIE H. HYMAN gration begins at the anterior end, and proceeds backwards along the axis to the posterior end. This is interpreted to mean that the head, or what represents the head, of the organism is carry- ing on metabolic processes at the highest rate, and that the rate of metabolism decreases along the antero-posterior axis. This kind of gradient exists in the egg, and Child (’13 ¢) has suggested that it is the physiological basis for the law of antero-posterior development. Child further suggests that this gradient, exist- ing in the protoplasm of the egg is carried over to the nervous system when the latter develops. The cephalic end of the nerv- ous system thus exercises from the very first physiological dominance over the rest of the organism, and retains this domi- nance by virtue of its functional relations to other parts, even though its actual metabolic rate may, and does, fall throughout ontogeny. The gradient of the adult animal may therefore vary markedly from the original gradient; various parts may attain a higher rate of metabolism than the head itself; but the nervous system, owing to long-established antero-posterior conduction paths, can maintain control, for some time at least, over regions of higher metabolic activity than its own. Is the gradient of the adult oligochaete of the primary type? I have found it so in but one species examined, namely, Aeolosoma hemprichii, a member of the most primitive family of oligochaetes. This worm is very small, only 1-2 mm. in length, has a rounded prostomium, a ciliated funnel-shaped pharynx leading into the intestine, and numerous red oil globules in the body wall. Each animal nearly always consists of two or more zooids which arise in connection with fission planes in the typical annelid manner; before they appear morphologically, they are present physiologically, as is readily demonstrated by the cyanide method. For the disintegration experiments, a concentration of > KCN is used, and the animals are placed in covered watch glasses as already described. In an individual without zooids, the dis- integration begins at the tip of the prostomium and proceeds at first slowly, then more rapidly along the axis. The integumental oil globules remain intact longer than the parts in which they were imbedded, but eventually they vanish by sudden extrusion lod PROCESS OF REGENERATION 107 of their colored contents. The gradient of such an individual is graphically depicted in text figure 2. In these graphs the ab- scissae represent the number of segments and the ordinates, the time of death in minutes. The dots along the curve are the points actually determined experimentally. In text figure 2, the flatness of the curve between the third and the fourth seg- ments indicates that the head of a zooid is forming there; this graph should therefore be compared with figure 1, which is the disintegration gradient of a fully developed posterior zooid of Aeolosoma, and which illustrates a typical primary gradient— o 10 X 15 18 1 2 Fig. 1 The axial gradient of a mature zooid of Aeolosoma hemprichii, illus- trating an ideal primary gradient. Fig. 2 The axial gradient of an individual Aeolosoma in which physiological isolation of a zooid is beginning. i.e., one that is steepest at the anterior end, and gradually falls off posteriorly. As the individual Aeolosoma grows, the posterior zooid con- tinues to differentiate physiologically. When such an individual is allowed to disintegrate in cyanide, the first change that occurs is the appearance of a constriction near the posterior end. This constriction marks the position of the head of the posterior zooid. Disintegration then proceeds independently in both zooids, from the anterior to the posterior end of each. Four stages in the disintegration of such an individual are illustrated in figure 3, THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, No. 2 108 LIBBIE H. HYMAN and a graph of the same in figure 4. The relative time of disinte- gration of the second zooid as compared with the first depends Fig. 3. Three stages in the disintegration of an Aeolosoma in which physiologi- cal isolation of a zooid is well advanced. I 2 3 4 5 6 7 Fig. 4 Graph of the axial gradient of an individual similar to figure 3. on the degree of development of the former; at the beginning of its existence, when it is present physiologically only, its time PROCESS OF REGENERATION 109 of death about coincides with that of the principal zooid; but, owing to the processes of dedifferentiation and growth involved in its formation, its susceptibility to cyanide continually increases so that its death comes to preceded by a considerable time inter- val that of the principal zooid. The rate of metabolism of the zooid thus continuously rises during its development. Eventually the presence of the posterior zooid is made known through the appearance of a fission plane, and its structural 4 6 8 Me Fig. 5 Graph of an Aeolosoma with a zooid well differentiated morphologi- cally, showing the two independent gradients. that of the zooid being at a higher level. differentiation proceeds. If examined in cyanide at this time, the two independent gradients show very clearly, that of the posterior zooid being at a higher level (fig. 5). The axial gradient of the posterior zooid is of the primary type, and remains so, indeed, until it has separated from the parent animal and begun to produce zooids of its own. The gradient of the anterior zooid is at the beginning of the development of the posterior zooid also of the primary kind but later becomes altered. Disinte- gration starts at its head and proceeds posteriorly, but soon the 110 LIBBIE H. HYMAN posterior end begins to disintegrate, and this disintegration pro- ceeds anteriorly. The two waves of disintegration meet some- where about the middle of the zooid. The interpretation of this sort of gradient is simple. Annelids grow characteristically by the formation of new segments in front of the anal segment; Fig. 6 Two stages in the disintegration of an Aeolosoma with a zooid, show- ing the secondary posterior rise in the principal zooid. the new segments thus formed would be expected to have a high rate of metabolism, and therefore an increased susceptibility to cyanide. As soon as the Aeolosoma individual has formed a posterior zooid, it begins to grow in this fashion at its new posterior end, and this growing region shows increased suscepti- bility to cyanide (fig. 6). In fact, this process may go to such an PROCESS OF REGENERATION 1 extent that the axial gradient of the anterior zooid is reversed (fig. 7); here disintegration begins at the posterior end and pro- ceeds to the anterior end. This condition is found in individuals in which the posterior end has grown considerably without be- coming physiologically isolated as a zooid; in such cases, the increasing youth of the segments posteriorly, as well as the fall in rate of metabolism of the anterior end through senescence con- tribute to cause the reversal of the gradient. As soon as the posterior end of such an individual becomes isolated as a zooid, the primary gradient reappears in it. 20 3O $15) 7, 5} S if 9 Fig. 7 Graph of an Aeolosoma with a zooid, showing reversal of the gradient in the principal zooid. As a further interesting detail, I may add that although the disintegration of the head usually begins at the tip of the prosto- mium, yet sometimes, and especially in old individuals, the first region of the head to disintegrate is the ciliated pharynx. Such a specialized organ by virtue of its sensory functions, ciliary activity, etc, must possess a relatively high rate of metabolism which it retains when the rate of surrounding parts has dimin- ished through senescence. Such specialized parts with high susceptibility to cyanide are frequently met with in disintegration 2 LIBBIE H. HYMAN experiments; as examples may be mentioned the sensory auricles of Planaria, and the lower lip of the oligochaetes. Summarizing these results, it is shown that in Aeolosoma the primary gradient is present, 1.e., the rate of metabolism—as measured by degree of susceptibility to cyanide—is highest at the head, and decreases along the antero-posterior axis. This gradient is present in the zooid from the very beginning of its existence, and continues during its differentiation, and after it has separated. Its rate of metabolism continually increases during this period, but after it has separated and begun to form zooids of its own, the rate falls, and may eventually be exceeded by that of its growing posterior end. 3. The gradient of the naids The primary gradient which exists in Aeolosoma is not re- tained in any other oligochaetes which I have examined except in the zooid stages. I have already spoken in the case of Aeo- losoma of the rise in susceptibility at the posterior end owing to the formation of new segments there. Aeolosoma, however, gives rise to zooids so rapidly that this growing posterior region does not attain to any considerable size because it is always being cut off by zooid formation. But in other Oligochaetes, especially those which do not reproduce asexually at all, this posterior growth is extensive, and has a marked effect on the axial gradient. Among, the naids, I have worked for the most part with Dero limosa. A concentration of KCN of ;5 or = was used. In individuals without fission planes the process of disintegration is as follows. Disintegration begins at the anterior end, involv- ing the tip of the prostomium and the sensory region about the mouth first, and passes posteriorly along the axis; after it has progressed some distance, which varies with individuals, disinte- gration begins at the posterior end and proceeds forwards; the two waves of disintegration meet about the middle or behind the middle of the worm, the exact point also varying with individuals. PROCESS OF REGENERATION Rs This is the typical annelid gradient,—i.e., one in which the rate of metabolism decreases from the head backwards and rises again at the posterior end.* This gradient, as already stated, is due to the characteristic method of growth of annelids by formation of new segments posteriorly. If the posterior end has been growing very rapidly, as it does when there is abundant food supply, its rate of metabolism may be higher than that of the head, and it may, therefore, disintegrate first, but the disinte- gration of the head always follows shortly after. Three stages in the disintegration of Dero limosa are illustrated in figure 8, and a graph of the same in figure 9. In this graph, and in the succeeding ones, | have attempted to compensate for the continual decrease in size of the posterior segments by gradu- ally increasing the number of segments per unit of the cross- section paper. In this way, a truer picture of the gradient is obtained. The posterior disintegration may begin with the anal segment, or in the region just anterior to this where the youngest segments are forming. In the latter case, the anal segment, which in the > As a matter of fact, Morgan (’04), if he had only known it, long ago demon- strated the typical annelid gradient in the earthworm by means of a galvanometer. He found that in the earthworm the anterior and posterior ends are electronega- tive to the middle part. It is a familar physiological fact that stimulated regions (1.e., regions of increased metabolism) are electronegative to non-stimulated ones (current of injury, negative variation, electrical variation during the heart beat, etc). The fact that cutting is a stimulation also accounts for the general result of Morgan that cut surfaces of the earth worm are electronegative to intact ones. The results of Morgan also indicate that the clitellum and the fifteenth segment are local regions of high metabolic activity, probably owing to their secretory nature. The data which seemed inexplicable to Morgan there- fore find easy explanation when the nature of the annelid gradient and the stimu- lating effect of cutting are known. There can be no possible doubt that the difference of potential between the intact and the cut surfaces has a different value when the cut surface is to form a head than when it is to form a tail, but there is no reason whatever for assuming that a reversal of potential should occur in the tail forming region of the earthworm. In experiments of this kind there are many important factors to be considered, such as the length of time after cutting, length of the piece, part of the animal from which the pieces are cut, part of the intact surface to which the other electrode is applied, etc. The obser- vations of Czwiklitzer (Arch. f. Entw’mech, Bd. 19) on the disintegration and death of the polychaete Ophyotrocha also constitute a demonstration of the annelid gradient. 114 LIBBIE H. HYMAN genus Dero is enlarged to form a gill pavilion, remains alive longer than any other part of the body. The anal segment is, of course, one of the oldest parts of the body in Annelids, but in forms which divide by fission, each anterior zooid has to form a new anal segment, while the posterior zooid retains the old anal — (eep ees n Shee a5 ee Ks : ms x i Ge ce? ie Tas ee aa, ) 5 bros Fig. 8 Three stages in the disintegration of Dero limosa. segment. It thus happens that in any individual Dero, the anal segment may be very old, or it may be relatively young, and this accounts for differences in the time of disintegration of this part as compared with the rest of the body. The axial gradient of Dero is modified by the presence of zooids. When the worms have reached a length of 10-15 mm. with 50-80 PROCESS OF REGENERATION 115 segments, a fission plane in the form of a constriction appears posterior to the middle of the body. This constriction is coinci- dent with a septum; the anterior end of the new zooid arises be- tween this septum and the succeeding bundles of setae, and the region between the septum and the preceding bundles of setae develops a new posterior end for the old animal. All structures develop completely before separation of the zooids takes place. Usually but two zooids are present. In Dero, the region where the fission plane is to appear is not detectable by an increased D 4 28 32 5 10 15 20 25 30 35. 45 5S Fig. 9 Graph of the axial gradient of Dero limosa, showing the secondary posterior rise typical of oligochaetes. susceptibility to cyanide, as is the case in Aeolosoma. In fact, even after the fission plane is visible, there is no increased sus- ceptibility at that point. It is only after differentiation has begun on the two sides of the fission plane that any alteration of suscepti- bility is noticeable. At first this consists in a slight increase in susceptibility at the anterior end of the second zooid—that is, disintegration begins at the anterior end of the first zooid, pro- ceeds back some distance, then attacks the anterior end of the second zooid, then the posterior end, and finally the remaining 116 LIBBIE H. HYMAN parts of both. In terms of metabolism, the metabolic rate de- creases from the head of the anterior zooid backwards, rises at the head of the second zooid, falls again, and finally rises steadily to the posterior end. Figure 10 illustrates a stage in the disinte- Fig. 10 A stage in the disintegration of a Dero with a well developed posterior zooid. gration of the two zooids, and figure 11 is the corresponding graph. As the development of the head of the posterior zooid proceeds, its susceptibility to cyanide continually increases, ap- proaches that of the head of the first zooid, and eventually exceeds it. Meantime, processes of reorganization have been going on PROCESS OF REGENERATION 117 in the posterior zooid, so that a new gradient is established; this gradient is of the primary type. If, therefore, a worm with a well-developed posterior zooid is allowed to disintegrate in cyan- ide, it is found to consist of two independent gradients; dis- integration begins at the anterior end of each zooid, and proceeds to the posterior end of each (fig. 12). There may be a slight increased susceptibility at the posterior end of the first zooid. After separation of the zooids, posterior growth sets in, producing 24 32 38 5 9; =| 7 + = 2 TWO OPERATIONS NORMAL HY POPROSTOMIC APROSTOMIC BiAXIAL TAILS DEAD ————-— — — = — No hours......- | 20 42 i 26 ES INOWIFS. oe ae eo 60 30 6 2 2 D0) \VOUE Ss sano o ns 70 26 2 4 z S10) loos eee oe 88 4 8 the posterior ends of worms of the same size and from the same stock; fifty short pieces ab were cut from the anterior ends of these immediately; and fifty more after elapse of fifteen, twenty, and thirty hours respectively. The results are given in percent- ages; multiple outgrowths are classified under the types of heads. While I regret that I have not a closer series of time intervals, yet I think that it is evident that the head of Lumbriculus is deter- mined as normal in the majority of cases within twenty hours after section, and in practically all cases within thirty hours. 2. Stimulation by section As a second step in the analysis of the process of anterior regeneration, one must know what the metabolic condition of the pieces is at the time when the head is determined as normal. 146 LIBBIE H. HYMAN The rate of metabolism of pieces is not the same as that of cor- responding parts of the intact. worm because, as Child has shown (14d), the rate of metabolism is increased by the operation of cutting; this increase is greater the shorter the piece, and the lower its previous rate as part of the axial gradient. These facts have been demonstrated not only by the increased sus- ceptibility of the pieces to cyanide, but also by their increased CQO, production in Tashiro’s biometer. Similarly, in all the oligochaetes which I have tested Dero, Lumbriculus, Tubifex, and Limnodrilus—stimulation results from section. In long pieces there is little stimulation, and the disintegration gradient of the intact worm is preserved in the pieces, except for increased susceptibility at the cut surfaces. Short pieces are stimulated to a much greater extent, and their disintegration in cyanide takes place without regard to the previous gradient; it begins at the cut surfaces and proceeds towards the middle. It is there- fore the region of the wound which is the seat of the stimulation, and in short pieces the wound regions practically include all of the piece. Stimulation as a result of injury is undoubtedly a general phenomenon exhibited by living matter (Tashiro, 713), and gives us a simple explanation of such facts as the current of injury of nerve and muscle. The degree of stimulation of pieces is a function of their previous axial posit on, and this is particularly noticeable in short pieces. The isolated head shows no stimulation; pieces from anterior regions where the metabolic rate is high are stimulated to some extent; pieces from middle regions where the rate is ‘owest are stimulated most; pieces from the posterior region of high rate are stimulated to a considerable extent, and since they already possessed a high rate before section, their rate after section is usually higher than that of the middle pieces, although the rate of the latter has been increased more relative to their previous rate. The metabolic condition of the axial series of pieces is not, therefore, the reverse of the metabolic gradient of the intact animal, as is the case in Planaria, because of the existence in oligochaetes of the posterior region of high rate. That this posterior region is stimulated by section indicates, as I have al- PROCESS OF REGENERATION 147 ready pointed out, that it is not a truly independent part. The reversal of the gradient after section appears most clearly in a form like Dero, where the posterior region is of small extent. Table 4 is the record of such an experiment on Dero limosa; ten worms were cut up into sixth pieces, and put immediately into 537 KCN. The experiment was begun at 12.45, and the number of pieces completely disintegrated recorded at fifteen minute intervals. In general, then, the regions which in the intact animal have lowest susceptibility have highest suceptibility after section, while the regions having high susceptibility in the whole are not much altered by section, with the exception of the posterior end TABLE 4 TIME 1 2 3 4 5 6 TROO Rec leae 0 0 0 0 0 0 he sera as Sees en & 0 1 2 2 2 1 Thal sg Ae eee 0 4 5 3 4 2 BAe ers e- 0 a 8 5 4 6 75 OO setae eter ene 0 8 8 8 6 8 DG ean eae 1 8 Ss 9 7 9 PA cd DoS eerie 4 9 10 9 7 9 DA 055 i ae 10 10 9 8 10 3) UU & octet 10 9 SOUS) Aer che eee etenirae 10 which although already having a high rate is somewhat simu- lated by section. In the case of Lumbriculus and the tubificids, where over half of the body is involved in the posterior rise of metabolic rate, the susceptibility of all pieces except the most anterior is increased by section, that of the middle pieces relatively most but not enough to raise it above that of posterior pieces which had a much higher rate before section. The gradient is therefore not reversed in these forms; it simply is raised to a higher level, and becomes less steep. The st mulation following section may be explained as due to the severance of conduction and correlation paths. The fact that the head and anterior regions in general are little stimulated by isolation indicates that they are relat vely independent of 148 LIBBIE H. HYMAN other parts; posterior regions are, on the other hand, dependent on anterior parts and subordinate to them, since the severance of conduction paths between them results in stimulation. The increase of metabolism after section is only temporary. If the pieces are tested in cyanide at various intervals after cut- ting, it is found that the susceptibility to cyanide gradually decreases to far below normal, and then begins to rise again as regeneration sets in. For an analysis of the process of head formation, it is necessary to know what the metabolic condition TABLE 5 iy erp : : 5 r y+ N ° RO : : Susceptibility of short anterior piece de to KCN 34 at various times after cutting SAME IMMEDIATE- TIME REGION OF | LY AFTER 5 HRS. 17 HRS. 24 HRS. 48 HRS. 4 DAYS | CONTROL SECTION ee a Oo” | y) One| 0 0 ) 0 DIOOL weet | os 1 1 oO] 0 0 0 Del oi, (Oem 2 1 oO. 0) 0) 0) 30h | O° 4 5 1 0 0 eet 0 MABE Ro) 1 8 2 0 O | Our 0 SOR SAeEe| | 10 3 0 0 a al] 0 Seal Senet: 8 3 0 0 1 1 SOR eat i) 7 ) i 1 3 SPAR tare ea 9 7 0 1 1 3 LOO Seed mt) 9 0 3 4 8 AWA EA Pees Ferd] 10 | i) 1 4 (pe i 10 ASS Ot eter a. | 10 4 6 7 LAS Sereno | 6 7 he 410 SOO ee ae | 9 9 5) 115) 9 9 SOE aan 10 10 of the pieces is during the period in which the head is determined. I have therefore tested the susceptibility to cyanide of short anterior and posterior pieces of Lumbriculus. In table 5 is given the susceptibility to cyanide of the anterior piece de in figure 23, immediately, 5, 17, 24, 48, and 96 hours after section; and in table 6, the susceptibility of the posterior piece ab in figure 23 immediately, 5, 14, 19, 24, 48, and 96 hours after cutting. Ten pieces are used in each case, and the time of death of cor- responding regions of whole worms also noted. The concen- tration of cyanide used was =+,; observations were taken every PROCESS OF REGENERATION 149 fifteen minutes, and the number of pieces completely disintegrated recorded. The general result of experiments of this kind is that the rate of metabolism of anterior pieces falls rapidly after cutting, and rises again, so that within four days, it has reached the normal rate. In posterior pieces, the rate stays up longer after cutting, then falls to a greater extent, and remains low for a longer period of time. Further experiments show that it begins to rise by the TABLE 6 “7p "7° ane . - , ~ N . . “ . Susceptibility of short posterior piece ab to KCN =o y at various times after section are IMMEDI- TIME REGION OF eae 5 HRS. 14 HRS. 19 HRS. 24 HRS. 48 HRS. 4 DAYS CONTROL | spcTION RAE entices. 0) 0 0 0) 0 0 0 0 DOOR eet oe 0 0 1 0 1) 0 0 0 Dena 0 1 tyeeylt 2X0 0 ms 0 253) eae 0 4 Py 1 0) 0 0 0 Tats yt os aoe 0 a) 3 1 0 0 0 0 S200; 4553 0 7 4 Died a0) ip ae Ballas de: 0 9 6 4 2 ea eal 0 BeoU eas. 1 9 oe ea 4 So el 0 See iis Bigeiok 3 10 9 8 HY 3 1 | 0 AAO OR seats ac 4 9 10 6 3 1 0 4.15. 5) !) 6 4 1 0 ASS (Oe terse 10 10 a 5 3 ] AAD se: 9 6 5 1 HEOVs se e0- 10 ii 7 3 aye aye 8 8 4 HW eaves | 10 10 4 Dolleoedve 6 U0), so6oe 7 sixth day. Eventually the rate of metabolism of all regener- ating pieces becomes much higher than their original rate as parts of the organism; rejuvenation thus occurs as a result of regeneration. The rate of metabolism, then, of anterior pieces of Lumbriculus is low during the period that the head is determined as normal; and these pieces give rise to a high percentage of normal heads. On the contrary, the rate of metabolism of posterior pieces is 150 LIBBIE H. HYMAN high during the time of head determination; and it is these pieces which produce inhibited structures. 3. General conception of the process of regeneration This relation between frequency of normal head formation, and rate of metabolism at the critical period of head determi- nation in Lumbriculus has lead me to accept the conclusions to which Child has come from the consideration of similar facts in Planaria dorotocephala. Since he has already presented his conclusions (Child, ’14e), it is not necessary for me to discuss them in any very great detail. For the following brief present- ation, it will be convenient to employ a figure similar to that frequently used by Child (fig. 24). The cells at the cut surface x Fig. 24 Diagram for theory of head formation. as a result of the wound and altered conditions begin to produce new tissue from which the new head is to arise. These cells grow out with a certain rate of metabolism which is relatively high as shown by disintegration experiments. Now if the rate of metabolism of the old tissue y is low, there is nothing to hinder x from continuing its development; it becomes dominant over y, uses up the material of y for regeneration, and produces a normal head. This is the case in anterior pieces of Lumbriculus where, as was shown in the preceding section, the rate of y is low during the time that the head is determined as normal. If, on the con- trary, the rate of y is high, then x will not be able to dominate over y, nor to attain sufficient independence to produce a normal head; the development of x will be inhibited, and in proportion to the metabolism of y. Thus are produced the various types of inhib- PROCESS OF REGENERATION 151 ited heads; the higher the rate of y the more will it inhibit the anterior structure which arises, and if the rate of y is sufficiently high, then region x will not be able to become independent at all but will be dominated by y, and give rise to a subordinate part, a tail. Under such conditions biaxial tails result. It is also conceivable that « might begin head development but later be dominated by y to such an extent that head formation can no longer continue but tail formation sets in; this in my opinion is the explanation of the cephaluran outgrowths. In the case of multiple outgrowths, the outgrowth or outgrowths which have the highest rate become heads, and dominate over the others which then give rise to tails. In Planaria, when the region y dominates the region x, the acephalic condition results; but this is not the case in Lumbriculus, probably because the new tissue begins to grow out before its fate has been decided, and it then forms a head or a tail according as x or y dominate. It is not necessary that the region y should have a metabolic rate which is actually higher than that of x; in fact it probably never has, for disintegration experiments show that x usually disintegrates before y. But it is the ratio of the rate of x to the rate of y which is important, and in all probability the rate of x must exceed the rate of y by a considerable amount before normal head formation can occur. Re atively slight alterations of the ratio ECOL are sufficient to affect the process of morpho- rate of y genesis. Thus is a set of pieces of the same size, and from the same level of the body, the value of y cannot be very different in the different individual pieces; the rate of x probably differs even less; yet from such a set of pieces all types of nterior structures are obtained. If this conception of the process of head formation is correct, an experimental control of morphogenesis should be possible. By depressing the rate of y as compared with x, one ought to be able to increase the percentage of normal heads; and, conversely, by increasing the rate of y as compared with 2, a decrease in the percentage of normal heads should result. The first possibility is readily realized experimentally; the rate of y can be easily 152 LIBBIE H. HYMAN lowered by placing the pieces in dilute cyanide solutions immedi- ately after section. While this treatment must depress the rate of x also, yet the depression is less than in the case of y, because x is young, growing tissue, and regulates better to the depressing rate of x rate of y is thus attained, and as postulated, the percentage of normal heads is decidedly increased. This beautiful experiment was first performed by Child on Planaria dorotocephala, and is readily repeated in the case of Lumbriculus inconstans. In Table 7 are given the results of such an experiment. One hundred short posterior pieces (ab in fig. 23) were cut; fifty were put into water as control and fifty put immediately into cyanide solution of concentration 95 99) for four days, then removed to water. condition. The desired increase in the value of the ratio TABLE 7 NORMAL Pee eat APROSTOMIC ACEPHALIC |BIAXIAL TAIL DEAD CoMim@less skoe 18 30 24 6 22 Experiment... . 46 16 22 2 14 As the ab pieces never in my experience yield more than 30 per cent of normal heads, the increase here is due to the depres- sing action of the cyanide. The converse experiment, that of decreasing the ratio rate ORs rate of y is very difficult to carry out, for it is almost impossible to increase the rate of y, without at the same time increasing the rate of x more. Therefore stimulating conditions, such as increased tem- perature, increased motor activity produce the same effect as in the preceding experiment, as Child has shown; the rate of x is increased more than that of y, and a higher percentage of normal heads results. In Lumbriculus inconstans, however, the desired condition is realized in an unexpected way; in my earliest experiments upon this species I soon found that the ‘vitality’ of the worms decreases greatly if they are kept for any length of time in the laboratory. The capacity for regeneration diminishes to such an extent that the worms can no longer be used for ¢ PROCESS OF REGENERATION 153 experimental purposes. The cause of the decrease in regenerative power appears to be that the cells at the cut surface fail to grow out with their usual vigor. The rate of x is thus lowered, as required in the experiment, and the expected decrease in the percentage of normal heads occurs. In table 8, a comparison is made between the regenerative capacity of ab pieces cut on April 24, and two lots cut on May 30, from the same stock. Fifty pieces were taken in each case, and the results are expressed in percentages. TABLE 8 ~ = = a, a HY POPROS- = a= BIAXIAL TIME OF CUTTING NORMAL TOMIC APROSTOMIC ACEPHALIC TAILS DEAD April 24.5.5... 18 30 24 6 22 Min S08. see 6 34 20 2 6 32 Miawy 20h Ss. o.. 4 20 28 6 6 36 The decrease in normal heads is very characteristic, also the appearance of biaxial tails. The number of normal heads may be considerably increased in these pieces by putting them in dilute cyanide. The mortality in these pieces cut late in the season is unfortunately always high, but as the decrease in normal heads has been noticed in every experiment performed with worms kept some time in the laboratory, it cannot be accounted for on the basis of the increased mortality. The role of the axial gradient in morphogenesis has not perhaps been sufficiently emphasized. In the case of short pieces, the dynamic conditions developed after section are the principal factors, yet it should be obvious from what has been said that these conditions are dependent on the previous position occupied by the pieces in the axial gradient of the intact animal. In long pieces, the axial gradient is more directly concerned; these pieces are only slightly stimulated by section, and the axial gradient is preserved in them in practically its original state. Therefore the anterior end of long pieces always has the highest rate in the piece, and the region x is never threatened with inhibition from the parts behind it. For this reason, long pieces of Planaria and Lumbriculus always give rise to normal heads. It is the 154 LIBBIE H. HYMAN axial gradient also which determines that the head shall arise at the anterior end of the piece (Child, 714 e, p. 73). The objection may be raised to these statements that in some of the oligochaetes which I have been considering, the axial gradient is such as to be higher at the posterior end than at the anterior end of long posterior pieces. Now there appears to be a correlation between the extent of the characteristic posterior rise in metabolic rate, and the capacity for head formation. Thus in Dero, where the posterior rise is of small extent, head formation occurs at all levels; in Lumbriculus, where the region of high rate occupies more than the posterior half of the body, head formation tends to be inhibited in this region; and in the tubificids where nearly all of the body is involved in the ascending gradient, head formation is impossible except at extreme anterior levels. In my opinion, however, the ascending gradient plays only a minor rdle in morphogenesis. I! wish to point out that a similar series with regard to head formation can be found in the Turbellaria; Planaria maculata regenerates normal heads at all levels in reason- ably short pieces; P. dorotocephala in pieces of similar size forms normal heads at anterior levels only; Dendrocoelum lacteum will not produce heads at all behind the anterior third of the body; while in many polyclads head formation ceases posterior to the cephalic ganglia. In these forms, there is no posterior region of high rate to account for the facts, but the explanation lies in all probability in the character of the primary gradient, and in the rate at which new tissue grows out. The posterior rise in rate in oligochaetes is the expression of the increasing youth of cells in the posterior direction; it must be regarded as a secondary gradient superposed on the primary integrative gradi- ent in the nervous system. Evidence for this point of view is found in the experiments on stimulation after section, where it was shown that these posterior regions of high rate are affected in the same way although not to the same extent as regions of low metabolic rate, and are, therefore, like the latter, subordinate parts dependent on correlation with more anterior regions. The fact, however, that they are stimulated to a less degree than the regions of low rate indicates that they possess a certain slight PROCESS OF REGENERATION 155 amount of independence. One may say, then, that there is an antagonism between the primary and the secondary gradient, the one making for subordination of the posterior region, the other for independence. The primary gradient, however, al- ways has the upper hand, for it is there from the very beginning, and each new segment which arises is forced, despite its high metabolic rate, to become part of the system of conduction, which has already been long established in the antero-posterior direction. The secondary gradient probably aids in the inhibi- tion of head formation, but is unable otherwise to influence the morphogenetic effect of the primary gradient, unless the latter is eliminated by other factors. If this were not so, it would be impossible to understand why heads do not arise at the poste- rior end of long pieces of Lumbriculus and Tubifex. I might further point out that the zooids of the naids arise without regard to the secondary gradient, and often within it, and that the ‘break- ing’ region of Lumbriculus is always within the secondary rise; both of which facts indicate that the secondary gradient offers very ineffective opposition to the primary gradient. For zooid formation is, as Child has shown (’11 a), a matter of physiological isolation from the primary gradient, usually as a result of de- creased intensity of correlative stimulation, and would be inhibited by the presence of a gradient running in the other direction. i In how far do these conceptions, developed from the data on Lumbriculus, explain the facts of regeneration in the other forms? Why does Dero form normal heads at all levels? Experiments on the rateof metabolism of pieces of Dero at various times after section have shown that the rate is very high immediately after section, but begins to fall at once. Within three hours it has fallen below that of the corresponding part of the whole worm, in five hours it is still lower, and continues to fall until about twenty hours, after which a permanent and increasing rise in ratesets in. While I have not been able to devise any method for determining the time of head determination in Dero, yet it. certainly is most improbable that it could occur during the very brief period of stimulation. One may therefore safely say that THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 156 LIBBIE H. HYMAN the head of Dero is determined at a time when the rate of me- tabolism of the pieces is low, and that therefore normal heads are always produced. Why does head formation cease at an extreme anterior level in the tubificids? To answer this question is one of the most difficult problems with which we have to deal in the field of morphogenesis. The dynamic factors which have thus far sufficed to explain the experimental results cannot be called upon here, for this result is independent of size of piece. Numerous attempts - to alter the regenerative capacity of Tubifex and Limnodrilus have yielded negative results. The following suggestions are offered as to the cause of failure of head formation in these forms. Tn the first place, the new tissue grows out very slowly, and with a relatively low rate of metabolism. It therefore is very easily inhibited. Secondly, the secondary gradient runs far forward in these forms, and may serve as the inhibiting factor. The cells at the cut surface of a piece of Tubifex taken back of the fifteenth segment are in contact with a region of high rate, which is itself in contact with regions of higher rate, and so on. Unless, there- fore, the cells at the surface grow out rapidly and with a very high rate, and this is contrary to fact, they could not dominate the regions behind. Heteromorphic tails would be expected under such conditions; they actually occur in the earthworms, but here the cells do not grow out fast enough, and they are inhibited before they have an opportunity to produce anything. It is obvious that this explanation can be tested experimentally, and I intend to continue my experiments with Tubifex in the hope of obtaining positive results. As to why the posterior end of such pieces does not give rise to heads has already been dis- cussed; the secondary gradient cannot eliminate the primary gradient to that extent. Regarding the formation of the posterior end, a word should be . said. The cells at the posterior end grow out with a high rate of metabolism, but cannot become independent as long as the pri- mary gradient persists. Owing to this gradient they are in contact with subordinate parts, and must develop in corre'ation with more anterior regions. If, however, the gradient is elimi- i Lar f PROCESS OF REGENERATION 157 nated, as can be done by taking very short pieces, or putting the pieces in depressing agents, then the cells at the posterior end can become independent, and develop a head. Biaxial heads result under these conditions. On the other hand, if the old piece has a high rate of metabolism, it may be able to dominate the new cells growing out at the cut surfaces to such an extent that neither one can produce a head but each develops a subordinate part, resulting in biaxial tails. In most organisms these occur in short pieces only, where the piece attains a high rate of metab- olism through the stimulation from section. But in some oligochaetes, particularly the earthworm, biaxial tails are pro- duced by long pieces. Dynamic factors alone cannot account for this result, but it is probably due to the presence of the secondary gradient. The conception of the process of form regulation to which Child has come as the result of his extensive experiments with Coe- lenterates and flatworms can then be satisfactorily applied to the microdrilous oligochaetes. This conception may be briefly summarized as follows. The head can only arise if the cells which are to form it attain physiological isolation and inde- pendence from the rest of the piece. The head is a self-differenti- ating system; it does not develop in correlation with other parts but is the starting point for a new system of correlations, in other words, a new individual. In long pieces the head forms at the anterior end of the piece because the axial gradient determines that at this end alone can sufficient physiological isolation be attained. In short pieces, head formation depends on certain dynamic relations between the head-forming cells, and the old piece, and these dynamic relations are, in turn, dependent, al- though indirectly, on the axial gradient. All other parts are subordinate, and arise in correlation with the head, or with an- terior regions, which themselves develop in correlation with the head. This statement is proven conclusively by the fact that no piece ever regenerates structures anterior to its level unless a head is formed first at its anterior end. The response of cells or groups of cells to isolation, physical or physiological, is the production of an apical region or head; this response is the 158 LIBBIE H. HYMAN ‘fundamental reaction of the species,’ to use the phraseology of Child. The apical region, or head region, or, in animals which develop a morphologically differentiated nervous system, the cephalic region of the nervous system which is the dominant part of the head is a closer approach than any other part of the organism to a morphological expression of this fundamental reaction system. The fundamental reaction system, dominance of the apical region and the axial gradient are all merely different aspects of the same general idea, viz., that the specific protoplasm of any organism consists fundamentally of a single physico-chemical reaction system. This system is the basis of inherit- ance and its dynamic capacities, the foundation of hereditary char- acters. The first step in organization and in embryonic development results from the establishment in one way or another, of some regicn or portion of this protoplasmic reaction system as a region of higher rate of dynamic activity. This region dominates development, becomes the apical or head region and determines the axial gradient or gradients which constitute the dynamic basis of polarity and of individuation. The organization and development of various parts of the organism rests upon a similar basis of fundamental reaction system and domi- nance and subordination of parts resulting from differences in rate of reaction (Child, 714 e). V. SUMMARY 1. A gradient in rate of metabolism is demonstrated in the oligochaetes. 2. In the primary form of the gradient, the rate of metabolism is highest at the head and decreases along the antero-posterior axis. Among the oligochaetes this primary gradient is found only in Aeolosoma, and the zooids of the naids. The primary gradient is an integrative gradient. 3. In the other oligochaetes examined a posterior region of increased metabolic rate exists, and constitutes a secondary gradient superposed upon the primary gradient. The secondary gradient runs in the reversed direction from the primary; it results from the characteristic method of growth of annelids by continuous formation of new segments posteriorly, and is not integrative in character. 4. In Dero limosa, the secondary gradient involves the pos- terior third of the body; in Lumbriculus inconstans, it includes the posterior half of the body or more;\and in the tubifigids, it includes all of the body except the first five to fifteen segments. ae PROCESS OF REGENERATION 159 5. In zooid formation, the gradient of the zooid gradually becomes independent of the gradient of the parent animal and is of the primary form. Owing to the processes of growth and dedifferentiation involved in zooid formation, the rate of metab- olism of the fully developed zooid is higher than that of the parent; i.e., rejuvenescence results from asexual reproduction. 6. In oligochaetes a certain number of the most anter’or seg- ments are differentiated as a head. 7. In regeneration, the head and tail are replaced by out- growth, the other parts by reorganization of the old tissue. No matter how many anterior segments are removed, only the typical number of head segments is, in general, replaced. 8. The head of oligochaetes will not regenerate a tail unless a certain number of trunk segments are included with it; nor will the end of the tail regenerate a head unless of a certain minimum size. Explanations of these facts are suggested. 9. In‘Dero limosa, any part of the body, whether long or short, regenerates a normal worm (with exceptions noted in 8). 10. In Lumbriculus inconstans, any part of the body, if of sufficient length, regenerates a normal worm. Short pieces show progressive inhibition of head formation along the axis; they give rise to anterior structures showing all gradations between a normal head and a normal tail. Normal posterior regenera- tion occurs at any level, and with any size of piece, but the number of segments regenerated decreases along the . antero- posterior axis. 11. In Tubifes, head fonMLION ceases at about the level of the fifteénth anterior segment, and, in Limnodrilus, at the level of the seventh segment, regardless of size of piece. Tail for- mation occurs at any level. , 12. In Lumbriculus inconstans, it is determined, whether or not the head shall be normal within twenty to twenty-five hours after the pieces are cut. 13. The gradient of an axial series of pieces is not the same as that of a whole worm, because cutting stimulates. This stimu- lation is greater the shorter the piece and the lower its previous rate of metabolism. This stimulation is temporary, the time of 160 LIBBIE H. HYMAN its duration varying with the different species, and is followed by a depression. 15. In Lumbriculus inconstans, short pieces from anterior regions are not much stimulated by section, and depression sets in within the time required for head determination. These pieces produce a high percentage of normal heads. Short pos- terior pieces are stimulated much more, and the stimulation lasts for a longer period of time, as long as the time required for head determination; these pieces produce a low percentage of normal heads, and a high percentage of inhibited structures. 16. The rate of metabolism of the piece during the time when the head is determined is, therefore, the important factor -in anterior regeneration in short pieces. If the rate be high, then the region of new tissue which is to form the head is prevented from attaining the degree of independence and isolation neces- sary fornormalhead formation. Head formation will be inhibited in proportion to the metabolic rate of the old piece. On the other hand, if the metabolic rate of the old piece is low, then the new tissue suffers no inhibition and gives rise to a normal head. The dynamic relations set up between the old and the new tissue after cutting determine the character of the head, or whether a head shall form at all, or whether a tail shall form. . 17. In long pieces, the dynamic factors are unimportant as the primary gradient determines that the cells at any anterior level are always more independent than those of a more poste- rior level. Therefore, normal heads are always formed on long pieces. 18. Experimental proof of the above conception is presented. If the rate of metabolism of the piece be depressed by means of cyanide, then the percentage of normal heads is increased; if the rate of the new tissue be decreased, then the percentage of normal heads decreases. This work was carried on at the University of Chicago under the direction of Prof. C. M. Child during the years 1911-1914. It is a pleasure to me to acknowledge my indebtedness to Pro- fessor Child, and to express my sincere thanks for his continual kind and helpful criticisms and suggestions, and inspiring com- PROCESS OF REGENERATION ; 161 ments during the progress of the work. My thanks are also due to Dr. V. E. Shelford, now of the University of Illinois, for suggestions regarding the collection of material. March 20, 1915. BIBLIOGRAPHY ApeL, Max 1902 Beitrige zur Kenntnis der Regenerationsvorgiinge bei den limicolen Oligochiten. Zeitsch. f. wiss. Zool., Bd. 73, pp. 1-74. Bepparp, F. HE. 1895 A monograph of the order of the oligochaeta. Oxford, Clarendon Press. Bonnet, C. 1745 Traité d’Insectologie, ou Observations sur quelques espéces de Vers d’Eau Douce, qui, coupés par morceaux, deviennent autant d’animaux complets. Paris. Bitow, C. 1883 Ueber Theilungs- und Regenerationsvorgiinge bei Wiirmern (Lumbriculus variegatus Gr.). Arch. f. Naturg., Bd. 49, pp. 1-96. Caruson, A. J. 1907 On the action of cyanides on the heart. Am. Jour. Phy- siol., vol. 19, pp. 223-232. Cuitp, C. M. 1911 a Die physiologische Isolation von Teilen des Organismus. Vort. u. Aufs. u. Entw’ mech., Heft 11. 1911 b Studies on the dynamics of morphogenesis and inheritance in experimental reproduction. J. The axial gradient in Planaria dorotocephala as a limiting factor in regulation. Jour. Exp. Zodl., vol. 10, pp. 265-320. 1911c A study of senescence and rejuvenescence based on experi- ments with Planaria dorotocephala. Arch. f. Entw’ mech., Bd. 31, pp. 537-616. 1911 d The regulatory processes in organisms. Jour. Morph., vol. 22, pp. 171-222. 191le Studies, ete. II. Physiological dominance of anterior over posterior regions in the regulation of Planaria dosatocep ea, Jour. Exp. Zo6l., vol. 11, pp. 187-220. - 1911 f epadicst etc. III. The formation of new zooids in Planaria and other forms. Jour. Exp. Zodél., vol. 11, pp. 221-280. 1913 a Studies, ete. V. The relation between resistance to depress- ing agents and rate of metabolism in Planaria dorotocephala and its value as a method of investigation. Jour. Exp. Zodél., vol. 14, pp. 153- 206. 1913 b Certain dynamic factors in experimental reproduction and their significance for the problems of reproduction and development, Arch. f. Entw’ mech., Bd. 35, pp. 598-641. 1913 c¢ Studies, ete. VI. The nature of the axial gradients in Plan- aria and their relation to antero-posterior dominance, polarity, and symmetry. Arch. f. Entw’ mech., Bd. 37, pp. 108-158. 1913 d The asexual cycle of Planaria velata in relation to senescence and rejuvenescence. Biol. Bull., vol. 25, pp. 181-203. 1914a Susceptibility gradients in animals. Sci., N.8., vol. 39, no. 993, pp. 73-76. 162 ; LIBBIE H. HYMAN Cuitp, C. M. 1914 b The axial gradient in ciliate Infusoria. Biol. Bull., vol. 26, pp. 36-54. 1914c Asexual breeding and prevention of senescence in Planaria velata. Biol. Bull., vol. 26, pp. 286-298. 1914d Studies, etc. VII. The stimulation of pieces by section in Planaria dorotocephala. Jour. Exp. Zoél., vol. 16, pp. 413-441. 1914e Studies, ete. VIII. Dynamic factors in head determination in Planaria. Jour. Exp. Zoél., vol. 17, pp. 61-79. DimFFENBACH, O. 1886 Anatomische und systematische Studien an Oligo- chaetae limicolae. 24 Ber. Oberhess. Gesell., pp. 65-109. GEpPERT, J. 1889 Uber das Wesen der Blausiiurevergiftung. Zeitsch. f. klin. Med., Bd. 15. GwatHmMeEy, J. T. 1914 Anesthesia. New York. Haase, H. 1898 Uber Regenerationsvorginge bei Tubifex rivulorum. Zeitsch. f. wiss. Zool., Bd. 65, pp. 211-256. Hazen, ANNAH P. 1898 Regeneration of a head in place of a tail in an earth- worm. Anat. Anz., Bd. 16, pp. 536-541. HEScHELER, K. 1896 Uber Regenerationsvorgiinge bei Lumbriciden. I. Theil. Jen. Zeitsch. Naturw., Bd. 30. Ivanow, P. (Ivanoff) 1903 Die Regeneration von Rumpf- und Kopfsegmenten bei Lumbriculus variegatus, Gr. “Zeitsch. f. wiss. Zool., Bd. 75, pp. 327-390. JANDA, V. 1912 Die Regeneration der Geschlechtsorgane bei Criodrilus lacuum, Hofm. IJ. Arch. f. Entw’ mech., Bd. 34, pp. 557-587. Krecker, F. H. 1910 Some phenomena of regeneration in Limnodrilus and related forms. Zeitsch. f. wiss. Zool., Bd. 95, pp. 383-450. Lreypoutpt, H. 1912 Transplantationsversuche an Lumbriciden. Zur Beein- flussung der Regeneration eines kleinen Pfropfstiickes durch einen grésseren Komponenten. Arch. f. Entw’mech., Bd. 31, pp. 1-79. Logs, J. 1906a Versuche iitber den chemischen Charakter des Befruchtungs- vorganges. Biochem. Zeitsch., Bd. 1, pp. 183-206. 1906 b Uber die Hemmung der toxischen Wirkung hypertonischer Lésungen auf das Seeigelei durch Sauerstoffmangel und Cyankalium. Arch. f. d. ges. Physiol., Bd. 113, pp. 487-511. 1910 a Die Hemmung verschiedener Giftwirkungen auf das befruch- tete Seeigelei durch Hemmung der Oxidationen in demselben. Bio- chem. Zeitsch., Bd., 29, pp. 80-95. 1910 b The prevention of the toxic action of various agencies upon the fertilized egg through the suppression of oxidation in the cell. Sci., N. S., vol. 32, p. 411. Lors, J. AnD WasTENEYs, H. 1910 Warum hemmt NaCN die Giftwirkung einer Chlornatriumlésung fiir das Seeigelei? Biochem. Zeitsch., Bd. 28, pp. 340-349. 1911 Sind die Oxydationsvorgiinge die unabhiingige Variabel in den Lebenserscheinungen? Biochem. Zeitsch., Bd. 36, pp. 345-356. 1913 a Narkose und Sauerstoffverbrauch. Biochem. Zeitsch., Bd. 56, p. 295. 1913 b Is narcosis due to asphyxiation? Jour. of Biol. Chem., vol. 14, p. 517. PROCESS OF REGENERATION 163 MatuEws, A. P. 1905 A theory of the nature of protoplasmic respiration. Biol. Bull., vol. 8, pp. 331-347. MicHAELSEN, W. 1900 Oligochaeta. Das Thierreich. 10 Lief. Morean, T. H. 1897 Regeneration in Allolobophora foetida. Arch. f. Entw’mech., Bd. 5, pp. 570-587. Moraan, T. H., anp Dimon, A. C. 1904 An examination of the problems of physiological polarity and of electrical polarity in the earthworm. Jour. Exp. Zodél., vol. 1, pp. 331-347. Moreuttis, 8. 1907 Observations and experiments on regeneration in Lumbri- culus. Jour. Exp. Zodl., vol. 4, pp. 549-574. MrAzex, A. 1906 Die Geschlechtsverhiltnisse und die Geschlechtsorgane von Lumbriculus variegatus Gr. Zool. Jahr., Bd. 23, Anat. Abt., pp. 381-462. 1913 Beitrige zur Naturgeschichte von Lumbriculus. Sitz. d. K. Bohm. Gesell. d. Wiss. in Prag., Juli, pp. 1-54. Miurer, O. F. 1771 Von Wiirmern des siissen und salzigen Wassers. Kopen- hagen. _ Rurtorr, C. 1908 Transplantationsversuche an Lumbriciden. Vereinigung inversgelagerter Teilstiicke unter Uberwindung der Polaritit. Arch. f. Entw’mech., Bd. 25, pp. 451-491. ScuroepER, H. 1907 Uber den Einfluss des Cyankaliums auf die Atmung von Aspergillus niger nebst Bemerkungen zur Mechanik der Blausiure- wirkung. Zeitsch. f. wiss. Bot., Bd. 44, pp. 409-481. Semper, C. 1876 Die Verwandschaftsbeziehungen der gegliederten Tiere. Arb. Zool. Inst. Warzburg, Bd. 3. SHetrorp, V. E. 1913 Animal communities in temperate America. Bull. No. 5, Geographic Soe. of Chicago. Tasuriro, 8. 1913 a A new method and apparatus for the estimation of exceed- ingly minute quantities of carbon dioxide. Am. Jour. Physiol., vol. 32, pp. 137-145. 1913 b A chemical sign of life. Biol. Bull., vol. 25, pp. 282-286. TasuHtRo, 8., AND Apams, H. 8. 1914 Studies on narcosis. I. The effect of ethylurethane and chloral hydrate on the CO: production of the nerve fiber. Internat. Zeitsch. f. physik. chem. Biol., Bd. I, pp. 450-462. Trrata, L. G. Th. 1912 Regeneration und Transplantation bei Criodrilus. Arch. f. Entw’mech., Bd. 35, pp. 523-554. Vernon, H. M. 1906 The condition of tissue respiration. Jour. of Physiol., vol. 35, pp. 53-87. Waener, F. von 1900 Beitriige zur Kenntnis der Reparationsprozesse bei Lumbriculus variegatus Gr. I. Zool. Jahr., Bd. 13, Anat. Abt., pp. 603-682. 1905 Beitrage, etc. II. Zool. Jahr., Bd. 22, pp. 41-156. Warsoure, O. 1910 Uber die Oxydationen in lebenden Zellen nach Versuchen am Seeigelei. Zeitsch. f. Physiol. Chem., Bd. 66, pp. 305-346. WINTERSTEIN, H. 1913 Beitrige zur Kenntnis der Narkose. I. Biochem. Zeitsch., Bd. 51, pp. 143-170. 1914 Beitrige, ete. II. Biochem. Zeitsch., Bd. 61, p. 81-102. THE EFFECT OF LIGHT ON THE RETINA OF THE TORTOISE AND THE LIZARD S. R. DETWILER From the Osborn Zoological Laboratory, Yale University ELEVEN FIGURES INTRODUCTION Since the discovery of the migration of pigment by Boll (’77) and by Kihne (’77) and of the contraction of the cones by van Genderen Stort (’87,) (see Engelmann, ’85) in the retina of the frog, this subject has been carefully investigated by many authors in many animals. Garten (’07) has brought together the main and important results of this work on the changes induced in the retina by light. From this we see that light produces a variety of effects on the form and staining reactions of the different parts of the retina. Of these effects three are particu- larly interesting to us. 1) Migration of the pigment in the epi- thelial cells of the retina. 2) Changes in form and position of the visual cells. 3) Changes in form, position and ability to stain of the ganglion cells and of the nuclei of the inner and outer granular layers. _ As far as the reptiles are concerned these questions seem far from settled, and therefore worthy of further investigation. In the first place concerning the migration of pigment, Angelucci (78, p. 372) was not able to say from the results of a few experi- ments on the turtle, Testudo graeca, and on lizards (L. agilis, L. muralis and L. viridis), which have no rods, whether pigment migration took place or not. If it does, he remarks, it is much less marked than in the amphibian eye. Boll (81, pp. 20 and 21) in an incompleted work, also considered this matter and in a theoretical consideration of the physiological properties of the 165 166 S. R. DETWILER pigment epithelium, in which he regards the migration of the pigment to be bound up with the using up and regeneration of visual red in the rods, he concludes that pigment migration should not take place in a rodless retina, such as the lizards have. Boll, however, cannot say whether it does or not. Angelucci (’94) (see Garten, p. 68), however, found that in Testudo marina the pigment does migrate, though less strongly than in the frog. And Chiariii (’06) was also able to clearly demonstrate pigment migration in the retina of the lizard (L. agilis). He figures, rather diagrammatically, side by side a dark and light retina, and, although he gives no measurements of the extent of the migration, it is clear that the pigment in the illumi- nated eye is nearer the external limiting membrane than it is in the dark eye, covering the paraboloids and drawing away from the bases of the pigment cells so that their nuclei are entirely uncovered. Garten (’07, p. 68), however, was unable to obtain prepara- tions of the retina of Emys, Chameleon or of Lacerta which showed constant differences in the position of the pigment accord- ing as to whether the animal had been kept in darkness or in bright light. His results will be referred to again. And finally Hess, (10, p. 281), was no more successful than Garten with Emys europaea, the position of the pigment in eyes of indi- viduals that had been kept for 22 hours in darkness, 2 hours in sunlight and several hours in light of weaker intensity being in all not markedly different, the outer segments being always covered by a mantle of pigment. Concerning the contraction of the cones in light, Engelmann (85, p. 500) found that in the eye of the snake Tropidonotus natrix, which contains no rods in the retina, the cones contracted but little. Also that in Testudo graeca it is doubtful whether any contraction takes place. Angelucci (’94) (see Garten ’07, p. 25) however, claims that in Testudo marina contraction of the cones does take place, though less in extent than in the frog. Chiarini (’06) also reports that in the eye of L. agilis the cones shorten when the eye is brightly illuminated, but only slightly, for the cones measure in dark eyes 25-35y, in light eyes EFFECT OF LIGHT ON THE RETINA 167 23-30u. Finally Garten (07, p. 25) found also a very slight contraction (not more than 1.14) in the eye of Chameleon. From this brief review of the few papers on the subject con- cerning reptiles we see that pigment migration and cone con- traction are very slight if they occur at all. No work of this nature has been carried out on American species, and since it is desired to carry out a series of further experiments on tortoises and lizards with particular reference to vision, it was thought that something should be known concerning the reactions of the various parts of their retinae to light. Three species of tortoises and one of lizards were used in the present investiga- tion, viz., Chelopus guttatus, Chelopus insculptus, Chrysemys picta, and the common southern fence lizard, Sceloporus undu- latus. Most of the work was carried out on Chrysemys and Sceloporus. This investigation was taken up at the suggestion of Dr. Henry Laurens. It gives me pleasure to express here my thanks to Dr. Laurens for the assistance that he has given me during its completion. METHODS : The methods of exposing the animals to light and to darkness were as follows: Two active animals were selected and placed in darkness for 24 hours. At the end of that time one of them was taken from the dark room and placed in direct sunlight for at least 6 hours after which it was killed. The other animal was either killed after it had been in darkness for 24 hours or after it had remained there at least 6 hours more. The eyes were removed as quickly as possible after the animals had been killed by decapitation—the dark eyes under red light, the light eyes in sunlight—and immediately dropped into the fixing fluid. . The time consumed between decapitation and fixation was 5 minutes or less. The fixation and subsequent procedure which gave the best results was the following: Fixation in Kleinenberg’s strong picro- sulphuric for 4 to 5 hours, followed by 70 per cent alcohol, which was frequently changed and in which the eyes were allowed to 168 S. R. DETWILER remain for several days. Further dehydration, consuming at least 2 days, the lens being removed after the eyes had been in 95 per cent alcohol for several hours. For infiltration the chloro- form parafin method was employed, paraffin melting at 52°C. being used for imbedding. Sections were cut 8 to 10, thick, stained in Ehrlich’s haematoxylin, followed by eosin. In a few cases, in an attempt to secure a quicker and perhaps more per- fect fixation, the lens was removed before the eyes were dropped into the fixing fluid. It was found, however, that this method caused shrinkage and folding of the retina to such an extent, at the same time producing no better fixation, that it had to be given up. ANATOMICAL Before we proceed to consider the results of the comparison of light and dark eyes it will be best to give a brief account of the anatomical relations of the species with which we are work- ing. Considerably more has been done by previous investiga- tors on the morphology of the reptile retina than on its physi- ology, and it will be well to review brigiy al the results of this work. Schultze (’66 and 767) noted that in Emys europaea, L. viridis, L. agilis and L. muralis there were no rods. From Hulke’s (’67, p. 94) incomplete description one might assume that rods were to be found in the retinae of, among others, Testudo graeca, Emys europaea, Chelone midas, Lacerta viridis and Anguis fra- gilis, though, as Krause pointed out later, these ‘rods’ are more correctly to be considered as cones without oil drops. Heine- mann (’77, p. 423) who examined the retinae of several Mexican species of tortoises concluded that, if the form of the outer seg- ment be taken as the criterion, then rods as well as cones can be distinguished in the retina of Chelonians. That in lizards, how- ever (p. 431) no elements with rod-like outer segments can be distinguished. In the Geckos, however, (p. 4384) it is doubtful whether the visual elements are rods or cones. Angelucci (’78, p. 371) found that rods are entirely lacking in Testudo graeca, L. agilis, L. muralis and in L. viridis. Boll EFFECT OF LIGHT ON THE RETINA 169 (81) briefly states (pp. 21 and 35) that there are no rods in the retina of Testudo or of Lacerta, while Cheivitz (89, p. 143) adds that in Emys europaea and L. viridis there is but one form of visual cell which, from its form, is a cone. Krause (’93) finds that in Chelonians there are no rods, and that the elements which had been earlier described as rods were nothing more than cones without oil drops. In L. agilis, how- ~ ever (’76 and ’93), he describes rods as being present, but scarce, though in certain places they are found thick together. Anguis fragilis and L. viridis have only cones. According to Angelucci, (’94, see Garten, p. 25) Testudo marina has no rods and Greeff (’00, p. 123) makes the brief statement that the reptilian retina (lizards, snakes and tortoises) has only cones. Chiarini (’06) says that the neuro-epithelium of lizards is formed exclusively of cones of various sizes. Garten (07, p. 24) points out that in Tropidonotus there are no rods, that in-Testudo graeca there are probably none, and that in the Chameleon there are certainly no rods. Pitter (’09, p. 103), concludes from the mode of centripetal connection (dendritic) of the visual cells of the reptiles that all of them must be cones, although the form of the single elements can be very different, e.g.,in Anguis fragilis and in the Gecko there are found cylindrical or rod-like outer segments. But he adds that all Chelonians have conical outer segments. Hess (710, p. 281) gives a review of the literature and states, that, in addition to the anatomical features of the retina indi- cating that there are no rods, the futile endeavors of several in- vestigators to obtain evidences of visual purple in the tortoise retina indicates that there can be no rods present. And finally Franz (13, p. 52) in a very incomplete and, in some particu- lars, incorrect review makes the statement that tortoises possess both rods and cones. From these papers the conclusion may be drawn that cones are by far the principal visual element of the retina of tortoises and of lizards. Further that rods may occur in a few, but that they are searce. It is to be regretted that good figures of the visual cells are not given in any of the articles reviewed. More- 170 S. R. DETWILER over except in the work of Heinemann and Krause, American species have not been investigated. Of those mentioned by these authors only one genus is represented, viz., Sceloporus, that has been studied by me, and this is a Mexican species. The need therefore for some account of the retina of the ordinary American species of tortoises and lizards is still urgent, and since the anatomical features of the visual cells have not yet been described, a short account of these will not be out of place. To begin with the tortoises we find that in the retina of the three species examined there are no rods. The cones are of two sorts, single and double. The single cone is the more numerous type of visual cell, and they are all similar in the possession of an outer and inner segment, in the latter of which is found in all cases an oil drop, an ellipsoid and a paraboloid. In form and size, however, these single cones present individual variations, on the basis of which we may say that there are two kinds, the first of which is considerably broader than the other, but only a little longer. With this increase in size there is found a slightly larger paraboloid and oil drop (fig. 6). The double:cones, of which there is only one kind, are much fewer in number than are the single cones. They are composed of a principal and of an accessory part, there being no twin cones. The principal cone has a very long narrow myoid, a long ellipsoid and an oil drop, there being no paraboloid. The accessory, which is much broader and shorter than the principal cone, has the typical short myoid of a single cone, a paraboloid, a granular ellipsoid but no oil drop (fig. 6). From preparations of fresh retinae it was found that the colors of the oil drops were those which have been usually described for the tortoise retina, namely red, orange, pale yellow and blue green. The red are the largest and the most numerous. Krause (93) describes single and double cones in the retina of Emys europaea, there being two varieties of each. The first kind of single cone is similar to that which we have just de- scribed, the second is much broader and has only a coarse granu- lar ellipsoid, but with no oil drop or paraboloid. Concerning the double cones of Emys the first variety, which is extremely EFFECT OF LIGHT ON THE RETINA eal numerous, is similar to those just described, differing only in that both the principal and the accessory cones may have oil drops. The principal cone has a plano-convex ellipsoid, while the accessory possesses a plano-concave ellipsoid and a homo- geneous paraboloid. In the second variety of double cone the principal cone is similar to the first, but the accessory is similar to the second variety of single cone, that is, thicker, with no oil drop but with an ellipsoid. Heinemann (’77, p. 423) from his study of the retinae of sev- eral Chelonians distinguishes two kinds of cones. 1) those with an oil drop, and 2) those without. Of the first of these the inner segments are much thinner than the others, so that they approximate the form of rods, containing an ellipsoid and a paraboloid. Heinemann subdivides these cones with oil drops into four varieties. a) those with bellied out inner segments and large lens-shaped bodies, b) those which are narrower and with a smaller body, c) those which are pointed on the inside and contain here either a body of appropriate size or none at all, d) cones with strongly bellied out outer portion of the inner segment and with irregularly formed, and always much narrower, inner portion of the same. Seldom there is to be found here a small lens-shaped body, but usually this part is structureless or filled with a finely granular mass. The last two kinds, c) and d), of cones with oil drops unite with the cones without oil drops to form double cones. But in Testudo gray, double cones, both parts of which contain an oil drop are also to be found. Concerning the lizard retina here again it may be said that there are no rods. The cones, however, in addition to showing several varieties among themselves are different from those of the tortoises examined (compare figs. 6 and 7). We again find in the first place that there are single cones and double cones. The single cones are of two varieties, a) with a very long myoid, no paraboloid, but with an ellipsoid and an oil drop. And b) much thicker than the first, with a shorter myoid and with an ellipsoid, paraboloid and oil drop (fig. 7). Similar to the tor- toise retina there is one kind of double cone. The principal cone is similar to the long narrow single cone while the accessory THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 172 S. R. DETWILER is much thicker, with a large paraboloid, a granular ellipsoid, and no oil drop (fig. 7). Krause (93) holds that Lacerta agilis has four kinds of visual elements, one of which is a rod, the other three being cones. The second kind of single cone is distinguished from the first by having a thicker inner segment and lacking an oil drop, although there is present an ellipsoid and a paraboloid. Double cones come principally from a combination of a principal cone with an oil drop and an accessory cone without an oil drop, though he also found double cones, both parts of which contained an oil drop. Heinemann (’77) finds two kinds of cones, those with oil drops and those without. The cones with oil drops show all gradations from voluminous cones with large lens- shaped bodies to very thin ones with very narrow inner segments in which there is no lens-shaped body. These latter unite with cones without oil drops to form double cones. Greeff (00, p. 117) figures a single and a double cone for the retina of a lizard. The single cone is quite similar to the narrow type found in Sceloporus (fig. 7), the onlv difference being that the myoid is comparatively shorter and thicker. The second or broad type of single cone, which is the predominating type of single cone in Sceloporus, he does not figure at all. The double cone is entirely similar to the one found by me. In passing, it 1s interesting to note that in the tortoise retina the external nuclear layer consists of two rows of nuclei. Of these only those nuclei in the row immediately internal to the external limiting membrane are cone nuclei, the other row being bipolar nuclei which connect the cone nuclei with those of the inner nuclear layer. The cone nuclei are in general larger and more oval, with the long axis in the same line as that of the cones. In Sceloporus, however and Cnemidophorus (the sand lizard) the external nuclear layer consists of only one row. Chievitz (’89, p. 146) considers the second row, which he found in both Emys and Lacerta, to be the nuclei of supporting cells. Another point of interest is the presence of a fovea and large papilla in the lizard retina, but neither of these in the tortoises, though there is a small area centralis. Heinemann (’77, p. 425) EFFECT OF LIGHT ON THE RETINA 173 could find no fovea in the retinae of any of the Chelonians which he examined, although in several there was a small papilla. Chievitz (’89, p. 143) describes an area for Emys but no fovea, while in Lacerta viridis (p. 147) he finds a fovea. EXPERIMENTAL When sections of eyes taken from animals which had been placed in sunlight are compared with sections of eyes from ani- mals kept in darkness it is seen, in the first place, that pigment migration does take place (figs. 1 and 2). In the dark eye (fig. 1) the pigment occupies more of the body of the epithelial cell, so that the nuclei are for the most part covered, and extends forward just beyond the oil drop, which in a few cases can be seen through the pigment. In the light eye (fig. 2) it is seen that the pigment has migrated forward so that the pigment cell nuclei are almost entirely uncovered while the pigment extends further toward the external limiting membrane, in many cases as far as the paraboloid. It was also noted that in the lght eye the pigment epithelium adheres more closely than in the dark eye and is not so easily torn away. The results of a series of measurements to determine the extent of migration are shown in table 1 (column 3). It will be seen that the average distance from the external limiting mem- brane to the nearest pigment needle in the dark eye is 8.8u while that for the ight eye is 5.2u. The difference of 3.6u repre- sents the extent of the migration. TABLE 1 5 ore er teen aad DIST. FROM EYT. DIST. FROM EXT. ANIMAL eas ONE ro LIM. MEMB. TO OUTER LIM. MEMB. TO NEAREST = SEG. OF CONE PIGMENT NEEDLE PIGMENT EPITHELIUM See CEN, Chrysemys € . . Ley J >) O Danke eae 26 .Ou 20.7 8.8u iohitiess sachs: 21 Ou 18 .4u 5.2m 1 Based on 10 measurements taken about 1 mm. from the entrance of the optic nerve. 174 S. R. DETWILER oe ee Dy : b- “i " oe 0 tf a Y 4 * ih Cae ate “Ny Noes ine fais etl ti‘ Lae ie « vis i Us at i Sa ma Hoh hoe i 1 itt a) flat ive La a Tr A “wi fee My " tee Ph bite ne is Kea ‘ ‘ an Mg th 4 heat Mh a7 gi oe pas pr) “ he ota touoy tage iy a) it oA bE, fs tee a ‘ "4 ie "i t %, par oi (CMY \ ul - = mde’ pi & B 4,2 , iy rT i ~ vind Ril Pant iui yi PMY iH RAN ne i pe sith Ai atin) nie pani WA ati Way tte Teer EA UM ats HEYA aac RY Sane \ ah ister Act i a ie ' Ca AYA wii oh tals y Vy Hi 5 Mt) iy toybegty ys) He NRT LALA hat ae wR ye nae un aye et Og wh eas ey MOEN thy Ay, yh DV) No ) , Fig. 1 A portion of the retina of Chrysemys picta. Animal kept in darkness for 24 hours. Fig. 2. A portion of the retina of Chrysemys picta. Animal, after having been kept in darkness for 24 hours, placed in sunlight for 6 hours. ke Fig. 3. A portion of the retina of Chrysemys. From an individual with cut optic nerve, kept in darkness for 24 hours, followed by a 7 hour exposure to sunlight. Fig. 4. A portion of the retina of Chrysemys. From an individual with cut optic nerve and kept in darkness for 24 hours. All of the figures were drawn with the aid of the camera lucida, and with Leitz oil immersion +; objective and ocular 4, giving an approximate magnification of 1340 diameters. The figures were then reduced 4 for publication so that as they appear they have a magnification of about 890 diameters. All of the drawings were made at about 1 mm. distance from the entrance of the optic nerve. EFFECT OF LIGHT ON THE RETINA 175 The comparisons of light and dark eyes of Sceloporus show that a pigment migration also takes place here. The average of measurements indicate that the extent of this migration is 3. bp (table. 2). TABLE 21 DIST. FROWN XT. LIMIT. = we ioe DIST. FROM EXT. DIST. FROM EXT. MEMB. TO CHOROIDAL ANIMAL EDGE OF LIM. MEMB. TO OUTER LIM. MEMB. TO NEAREST SEG. OF COD i DD =DLE PIGMENT EPITHELIUM SEG. OF CONE PIGMENT NEEDLE x Sceloporus ID aie caccan os 22 5 4. 9.0 a LUANG soo gece c< 19.1u 2, ayn 5 .9u 1Based on 10 measurements as in table 1. Not only does the pigment migrate but the cones also show a contraction in light. In table 1 the results of measurements are shown. In the first place in the dark eye the distance between the external limiting membrane and the choroidal edge of the pigment epithelium is 54 more than in the light eye. But this does not represent the amount of contraction of the cones, for as will be seen (column 2) the average length of the cones in the dark eye is only 2.3u longer than in the ight eye. From this it is evident that the light not only causes a shortening of the cones but a flattening of the pigment epithelial cells of, on the average, 2.7. This flattening of the epithelial cells in the tortoise retina is in line with the results of others. Angelucci (’84 and ’94) pointed out that in the illuminated eye of the frog a shortening of the epithelial cells in the direction of the axis of the rods was to be observed. Chiarini (04 and ’06) also saw this change in the retinal epithelial cells of the representatives of the five classes of Vertebrates which he examined. And Pergens (’96) noted the same thing in Leuciscus. In addition to this change in form of the epithelial cell there has also been observed a change in the position of the nuclei of these cells. Pergens (96) found that after illumination the nuclei of the pigment cells of Leuciscus were further forward than in dark eyes. Angelucci (’94) also observed this change of position in the frog. Garten (07 b) on the other hand, al- 176 S. R. DETWILER though he observed the differences in position of the nuclei of the epithelial cells of Abramis could find no constant differences due to the effeet of light and darkness. Concerning the third of the effects of ight on the retina in which we are interested, namely changes in form, position and the ability of the ganglion cells and of the nuceli of the inner and outer granular layers to stain there has grown up a com- paratively large literature. This has been excellently reviewed by Garten, who has in addition given a table in which the results of different investigators are shown. We will first take up the effects of ight on the nuclei of the cones. In Garten (’07 b, pp. 18-23) will be found a very good summary of what has been done. Czerny (’67), Gradenigo (’85), Angelueei (’94), and Chiarini (04) in the frog; Pergens (96) in Leuciscus rutilis; and Chiarini (’06) in the lizard found that light caused the cone nuclei to become longer and narrower. In addition, Birch-Hirschfeld ('06) found that light caused a difference in the volume of the cone nuclei of the pigeon, in that they are smaller and narrower. Birch-Hirschfeld further found in the pigeon that ight caused the cone nuclei to approach nearer to the external limiting membrane. Light was found to decrease the power of the outer granules to stain by Pergens (’96, ’97, and ’99) in Leuciscus rutilis, Mann (94) in the dog; Birch-Hirschfeld (’00) in the dog and cat, and (06) in the pigeon (very slight), Sgrosso (05) in the frog, and Garten (’07, p. 23) in Cercopithecus, Macacus rhesus, fishes (Abramis and Leuciscus), Salamandra, frog and the owl. Chiarini (04 and 706), however, found that in Leuciscus any difference between the ability of the dark and the light eye to stain was very uncertain, and that in the frog, lizard, crow and dog there was absolutely no difference. We may pass now in the same way to a brief review of the results that have been obtained concerning the effect of ight on the form and stainability of the inner granules and of the gan- glion cells. Mann (’95) found that in the dog, illumination of the eye for 12 hours caused a decrease ja the stainability of the inner nuclear layer and a decrease in the Niss] substance of the EFFECT OF LIGHT ON THE RETINA Lie protoplasm of the ganglion cells. Pach (95) however, in the rabbit could find no differences in the stainability of the inner nuclear layer or of the ganglion cells in light and dark eyes. Birch-Hirschfeld (’00) however, did find differences between the light and dark eyes of rabbits and of dogs. The nuclei of the inner layer in the dark eye are rounder, in the light eye more oval. In the ganglion cells the Nissl bodies in the ight eye have indistinct boundaries and with the protoplasmic background very diffuse. In the dark eye, on the other hand, the Nissi bodies possess sharp, distinct outlines. Chiarini (’04) found in Leuciscus, no decrease in chromatin in the inner nuclear layer after illumination, and in the ganglion cells hardly noticeable changes. Also later (06) in the inner layer of reptiles, birds and mammals he found no differences between light and dark retinae. In Lacerta, however, he observed a slight decrease in the Nissl bodies of the ganglion cells, in Corvus a decided de- crease, and in the dog again a slight chromatolysis. Schiipbach (’05) found no differences, neither in the inner layer nor in the ganglion cells, between light and dark eyes of pigeons. But Birch-Hirschfeld (06) was able to demonstrate clearly that the ganglion cells showed a distinct decrease in the number of the Nissl bodies and an indistinctness of their boun- daries in the light eye. Carlson (04) has shown the same for another bird—Phalacrocorax penecillatus, and finally Sgrosso (05) found in the frog that the inner nuclear layer showed differences in stainability in the light and dark eye. In order to make observations on these matters, particularly on the stainability of the nuclei, it was necessary to use some other methods of fixing, ete., than had been employed for the study of pigment migration, and of cone contraction. For this purpose the following method was finally decided upon, and gave excellent results. Fixation in warm concentrated subli- mate for 5 hours. Removal of sublimate with iodine in 70 per cent alcohol, further dehydration, and infiltration by the chloro- form paraffin method. Sections were cut 8» thick, stained in eosin and toluidin blue, rapidly dehydrated, cleared in xylol and mounted in damar. In order to insure the same amount 178 S. R. DETWILER of fixation, staining, decolorizing, etc., light and dark eyes were fixed in the same vessel, care being taken that they were dis- tinguishable. Further, after sectioning, alternate rows of dark and light eyes (two of each) were placed on the same slide so that there could be no doubt of their similarity of treatment. In the first place, the cone nuclei of both the tortoise and of the lizard are, for the most part, lengthened and narrowed by illumination. This change is, however, not very great and there is room for doubt owing to the great variability in the shape of these nuclei both in the light and in the dark eye. As far as their ability to stain is concerned the nuclei of the outer granular layer in the dark eye seem to be slightly more deeply stained than those of the light eye, but as far as the nuclei of the inner granu- lar layer is concerned, no differences can be found. Neither could any changes in form of the inner nuclei be noticed after long illumination nor could any changes in the form and volume of the ganglion cells be observed. These latter cells, however, do show marked and clear differences between the light and dark eyes (figs. 8 and 9). Figure 8 represents three ganglion cells from a dark eye. By comparing it with figure 9 which is from a light eye, the differences can be observed. Not only is the amount of chromatin reduced but the Nissl substance has de- creased in amount. Under the microscope these differences can be seen very distinctly, and it is possible to pick out the light and dark eyes by the comparative amount of chromatin and of Nissl substance which they contain. These results show that light causes a migration of pigment and a contraction of the cones, as slight as these may be, in the retina of the tortoises and lizard. Moreover, that the cone nuclei of the tortoises are probably narrowed and lengthened and that furthermore a diminution in the amount of chromatin and Nissl substance in the ganglion cells is brought about, so that they stain less darkly and more diffusely than in dark eyes. It was considered sufficiently interesting to attempt to find out whether some or all of these changes could be brought about in eyes which had either been enucleated or had had the optic nerve cut. EFFECT OF LIGHT ON THE RETINA 179 Angelucei (’78, p. 367) observed that when the optic nerve of a frog is cut the physiological changes of the pigment took place, thirty days after the operation, as in normal eyes. Hamburger (88), Arcoleo (90) and Fick (’01, p. 4) also found that when the optic nerve is cut the pigment changes induced by light and darkness took place as in the normal eye. In addition Engel- mann (’85, p. 505) observed that when the brain of a frog is destroyed that the effect of hght on the migration of pigment is still present. Movements of the cones have also been observed in eyes, the optic nerves of which have been cut, or which have been removed from the body. Hamburger (’89) found that, when he cut the optic nerve of a frog or removed the eye, contrac- tion of the cones will take place when the eye is illuminated. And Dittler (07) found in the isolated frog retina placed in salt solution that upon illumination the contraction of the cones takes place. Experiments on the enucleated bulbus of the tortoise were without results. The same is true of retinae which were isolated in the manner described by Dittler for the frog. The experi- ments on eyes with the optic nerve cut, however, did yield rather interesting results. All the experiments were carried out on Chrysemys picta. The method of cutting the optic nerve is briefly as follows: Under deep ether anaesthesia a small wedge of bone was removed from the left side of the mouth beneath the eye. This was done by means of a long, very narrow saw. The small amount of bleeding consequent to the removal of the bone being stopped, the muscles were pulled to one side and cut, until the optic nerve could be seen. This was then cut by means of a fine pair of scissors. The total loss of blood was small and the animals quickly recovered. Out of eleven individuals operated upon only the first three died. The others after a few hours were active and seemed quite as fit for experimentation as nor- mal animals. One effect of cutting the optic nerve, which was not always observed, was a slight enlargement of the pupil. Fick (91, p. 3) states that after cutting the optic nerve of the frog the pupil is temporarily somewhat narrowed. 180 S. R. DETWILER The tortoises were allowed sufficient time to recover from the effects of the operation before they were placed in light or dark- ness. Two operated animals were always used together for experiments and as controls two normal animals. It is certainly not without interest in this connection that the chromato- phores of cephalopods have been found to develop from single smooth muscle cells by a complicated metamorphosis (Chun ’02). MELANOPHORE A TYPE OF SMOOTH MUSCLE CELL 197 Sphincterfasern und mesodermalen Chromatophoren aufstel- len; and again: ‘Denn es scheint in mehr als einer Beziehung, dass die Chromatophoren nichts anderes als verkappte Muskel- zellen sind.” Many of the older investigations upon the responses of the melanophores to various stimuli were carried out upon living animals. Here the situation was exceedingly complex since so many physiological factors had to be controlled. In a previous paper (13 b) I have called attention to this objection and have described a method whereby the melanophores of teleosts, es- pecially Fundulus heteroclitus, may be removed from the fish and stimulated in a variety of ways without the slightest me- chanical injury in manipulation. The technique is exceedingly simple, consisting of the careful removal of the scales with their superficial sheets of dermal melanophores. These scales are readily transferred from one solution to another and, by selecting adjacent scales from the same fish, we obtain very satisfactory physiological units whereby it is possible to test the effects of a series of solutions—the degree of expansion or contraction of the pigment serving as an indicator of relative stimulation. In preparations of this sort in which the circulation and nervous control have obviously been eliminated, I have found that chemi- cal stimuli such as 0.1 N KCl, heat (80° C.), ultra-violet light and induction currents all bring about a more or less rapid contrac- tion of the pigment granules. Furthermore, in indifferent media like Ringer’s solution or olive oil, the melanophores remain ex- panded for long periods. It therefore seems justifiable to con- sider the contracted phase that of stimulation, since the melano- phores respond by contracting to the above familiar series of physiological stimuli. Throughout the following discusssion [ shall consider the contracted phase of the melanophores as cor- responding, physiologically, to the contraction in smooth muscle. The objections to the application of the term ‘contraction’ in the case of the melanophores, I shall take up at the close of this section. 198 REYNOLD A. SPAETH I. Innervation The activity of vertebrate smooth muscle is normally con- trolled through fibers of the sympathetic nervous system. Vol- untary motor connections do not ordinarily occur. The innervation of the melanophores has been satisfactorily demonstrated both histologically (Ballowitz ’93) and_ physio- logically (Pouchet ’72 and ’76, Von Frisch ’11, Spaeth loc. cit.) in several species of teleosts. v. Frisch has recently corroborated and amplified the original observations of Pouchet, who first claimed the innervation of the melanophores to be sympathetic. I have repeated the striking experiments of v. Frisch upon Phoxi- nus with Fundulus. In this experiment one of the two branches of the sympathetic system is severed immediately behind the body cavity in the region of the haemal arch. Fish so treated lose their power of color adaptation posterior to the point of incision on the operated side, but continue to show normal motor responses. Reciprocally, severing of the spinal cord eliminates motor responses but, provided the operation has been carefully performed, the sympathetic adaptations to different colored bot- toms remains normal. In Fundulus I have also found at the base of the medulla a ‘contraction center’ corresponding with that found by Lode (90) in the trout and by v. Frisch (11) in Phoxinus. A light- ening of the entire body of the fish follows the electrical stimula- tion of this center. No satisfactory histological demonstrations of the nerve end- ings in the melanophores of amphibians and reptiles have been recorded. Bimmerman (’78), Biedermann, loc. cit. and more recently Hooker (12) have however,all demonstrated, physio- logically, the sympathetic innervation in the melanophores of the frog and Carlton (’04) has made similar observations in Ano- lis. There is thus a very satisfactory unanimity of opinion on this question. MELANOPHORE A TYPE OF SMOOTH MUSCLE CELL 199 Il. Effect of Light Through the investigations of Arnold (41), Brown-Sequard (’47), Steinach (’92), Magnus (’99), and Franz (loc. cit.) we know that the excised iris of certain elasmobranchs (Acanthias), tele- osts (Anguilla) and amphibians (Rana) responds to illumination by a contraction of the sphincter pupillae. Hertel (’07), first sueceeded in demonstrating a direct response to stimulation by light in the chromatophores of the cephalopods Sepiola, Octopus and especially Loligo. In this case the reac- tions of the chromatophores are due to contractions and relaxa- tions of the radially arranged smooth muscles. The active phase is here the expanded one, elicited by the contracting radial mus- cles i.e., it is the reciprocal of the condition in the melanophores of lower vertebrates. Hertel found that ultra-violet light of 280 uu produced an almost instantaneous local expansion of all chromatophores. Blue rays of 440 wu and yellow rays of 558 wu of equal intensity also gave a distinct, but somewhat slower expansion. IT have shown (loc. cit.) that the expanded melanophores of Fundulus respond to ultra-violet light (280 uu) by a rapid and reversible contraction. I was unable however, in a series of trials with different regions and intensities of the visible spectrum to obtain contractions of Fundulus melanophores. Following practically the same technique as in my experiments with Fun- dulus, Laurens (’15) has recently reported contractions by ultra- violet light in expanded melanophores in pieces of the skin of Amblystoma larvae. Hertel (loc. cit.) had previously found that the melanophores of Triton larvae responded to ultra-violet, blue and yellow rays by contracting. He used the same wave lengths and intensities as in his experiments with cephalopods (vide supra). In this case, as in so many of the older experi- ments, the presence of a complete nerve mechanism and blood supply presents the possibility of secondary complications. In the case of Fundulus and Amblystoma, however, I believe the results to be free from this objection. 200. REYNOLD A. SPAETH A direct response to light has not thus far been demonstrated in the melanophores of reptiles, though as Fuchs (loc. cit.) has emphasized, it will in all probability be found to exist. We may now summarize the foregoing observations as fol- lows; 1) in certain species of fish and amphibians the sphincter pupillae contracts in response to direct stimulation by light; 2) in certain other species of fish and amphibians the melanophores also respond to direct stimulation by light by contracting. ITl. Effect of Electrical Stumulation Induction currents of sufficient intensity and duration produce contractions in smooth muscle. Such contractions may be seen in strips of frog’s stomach prepared according to Meigs (712), or in preparations of the digestive tube of several species of tele- osts.. Beer (94, ’98) observed that the sphincter pupillae of the eel and the frog (Rana) contracted when stimulated electrically. The radial muscles of the chromatophores of cephalopods also respond to electrical stimulation by contracting (Briicke °52, Keller ’73, Fredericq ’78, Klemensiewicz ’78, Pouchet ’76, Kruken- berg ’80, Phisalix ’92, Steinach ’01, Hofman ’07, 710, and Fuchs 10). Klemensiewicz (loc. cit.) records similar contractions in isolated pieces of the skin of Loligo. Lode (loc. cit.) first showed that in excised pieces of the skin of the trout the melanophores contracted upon being stimulated by an induction current. [n my own experiments with Fundu- lus the melanophores invariably contracted, reversibly, when proper strength and duration of the current and salt concentra- tion of the mounting medium were selected. Winkler (10) found that the melanophores of Rana esculenta and Hyla arborea contracted when directly stimulated by an induction current. Laurens (loc. cit.) has recently verified this observation in large larvae of Amblystoma opacum. He finds also that: ‘‘When various portions of the body are cut out and directly stimulated either with the central nervous system in- 3 Unpublished observations made upon the stomach muscle of Stenotomus, Tautoga, Centropristes and Fundulus. MELANOPHORE A TYPE OF SMOOTH MUSCLE CELL 201 tact or destroyed, a slight contraction of the melanophores is usually induced’”’ (p. 610). In the ease of the reptilian melanophores Bert (’75) and Kruk- enberg (80) were able to corroborate the original observation of Briicke (52) who found that direct faradic stimulation of ex- cised bits of dark skin in the chameleon produced a lightening, i.e., a contraction of the melanophores. Thus the responses of several types of smooth muscle, as well as of the melanophores, in representatives of all three groups of lower vertebrates show a satisfactory agreement in their con- traction to faradic stimulation. IV. Effect of Mechanical Stimulation By gently pinching or stretching excised pieces of frog or fish* stomach and oesophagus, powerful contractions may be induced, which are reversible provided the stimulus has not been too vio- lent. Precisely the same reaction follows a similar treatment of portions of the skin of Loligo; the chromatophores expand widely. I have verified this observation by Klemensiewicz loc. cit. and all of the more recent investigations upon cephalopod chromato- phores record a similar phenomenon in Loligo and other species. Fuchs (loc. cit) has called attention to the objection against the expansion observed by many of the older investigators after the surfaces of varios teleosts had been more or less violently ‘stroked’ with a needle. He says (p. 1432): ‘‘Aus allen Beo- bachtungen geht unstreitig hervor, dass einwandfreie Beobach- tungen tuber die direkte mechanische Reizbarkeit der Fischchro- matophoren nicht vorliegen.’”’ I have shown (loc. cit.) that by selecting scales from the lateral portion of Fundulus where the melanophores are relatively far apart, it is possible to stimulate a single melanophore repeatedly by exerting a gentle pressure with a fine, fire-polished glass needle. Great care must be exer- cised not to rupture the delicate cells, for by so doing, the melanin granules are scattered and produce the effect of an ‘expansion’ recorded by the older observers. In this case it is certain that D2 REYNOLD A. SPAETH mechanical stimulation (gentle pressure) produces a reversible contraction of the melanophores. Asa result of local pressure or pinching with forceps, and along the margins of an incision, the melanophores in the frog contract (we Wittich 754,771, Herme ly ea iister “58 a, 53:b; Fuchs 06; Etérnod and Robert ’08). The darkening of the skin in Poly- pedates Reinwardtii, ‘‘nach leichtem Kratzen mit einer Nadel,”’ observed by Siedlecki (’09), is doubtless another case of ruptured melanophores.‘ The above observations warrant the conclusion that in the radial smooth muscles of the chromatophores of cephalopods, in the smooth muscle of the digestive tract in certain teleosts and amphibia, as well as in the melanophores of these two vertebrate groups, mechanical stimulation (gentle pressure) is followed by a reversible contraction. V. Effects of Chemical Stimulation A great many experiments have been carried out upon the effects of various widely differing chemical stimuli upon the melanophores of vertebrates. The method of procedure has been, in most cases, as follows; 1) the substances were brought directly upon the skin of the normal or operated (pithed, etc.) animal, the color-change being considered the criterion of the action of the chemicals; 2) the substances were injected into the circulation, the body cavity, or subcutaneously, the color-change again serving as an indicator of chemical stimulation; 3) the sub- stances were added to the water of the environment, in aquatic forms. In relatively few instances have excised pieces of skin been immersed in the fluid to be tested. As I have repeatedly emphasized, this is the only satisfactory method of determining the direct effect of any stimulus upon the melanophores, for it is only in this way that the circulation and all central nervous 4 Observations upon the response to pressure in reptilian melanophores are contradictory and unsatisfactory. Most of the experiments have been made with living animals and the few trials with excised bits of skin were all carried out without regard for the possibility of a darkening resulting from the destruc- tion of the melanophores (Milne-Edwards 734 a, ’34 b, Briicke 752, Carlton ’04). MELANOPHORE A TYPE OF SMOOTH MUSCLE CELL 203 control may be simultaneously and effectively eliminated. The objection may be raised that even in excised pieces of the animal it is impossible to destroy the ultimate nerve terminations, hence we are unable to state with certainty that the effect is a direct one and not transmitted through the cut stumps of the sympa- thetic nerves which remain in such a preparation. I have else- where (loc. cit.) adduced evidence to show that this objection is certainly invalid in the case of the isolated melanophores of the scales of Fundulus. 1 shall limit the following discussion of the reactions of melanophores to chemical stimuli to cases which are as nearly as possible comparable to the chemical stimulation of isolated smooth muscle. A. Inorganic Substances. Schultz (97) observed that dis- tilled water acted as a weak contracting stimulus in opened ring preparations of the stomach of the frog. Meigs (10) has shown that distilled water gives a typical curve of contraction in prepa- rations of longitudinal strips of the frog’s stomach. I have found (loe. cit.) that the melanophores of Fundulus slowly contract when brought from the living dark fish or from 0.1 N NaCl to distilled water (fig. 1). Schultz (1. ec.) noted a relaxation and swelling of the smooth muscle of the frog’s stomach in 10 per cent NaCl solution. Meigs (loe. cit.) has shown that in solutions of KCl, the stomach muscle of the frog (R. pipiens) slowly contracts and loses weight while in NaCl it elongates and aborbs water.’ I have observed that the chromatophores in pieces of the mantle of Loligo expand im- mediately upon being immersed in 0.1 N KCl solution. This is obviously due to a contraction of the radial smooth muscles. Furthermore, provided the exposure to KCl has not been too long, when such chromatophores are returned to 0.1 N NaCl or better, Ringer solution, they contract again, the radial muscles are relaxed. The melanophores of all the species of teleosts with which I have experimented, showed a contraction in 0.1 N KCl solution and an expansion or relaxation in 0.1 N NaCl, dilute sea-water, ® Zoethout (’02) found that KCl produced a contraction in the gastrocnemius of the frog and NaCl a relaxation. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 204 REYNOLD A. SPAETH or Ringer’s solution. The effects of other neutral salts of the alkalies were also studied in considerable detail but, since, so far as [ am aware, no parallel experiments have been performed upon smooth muscle, a discussion of these results would be irrel- evant. It is, however, of interest that both cations and anions of the neutral salts show the same order of physiological effect, the so-called ‘lyotropie order,’ as in experiments upon striated muscle (Overton ’04 and Sewartz ’07) and certain colloids ( Hof- meister 791, Pauli 99, Héber 14, Chapter 7). Minutes — Fig. 1 The slow contraction of a melanophore in distilled water. The upper line represents the movements of the terminal pigment granules in one process of the cell. The lower line indicates time in minutes. At D. W. thecellwas immersed in distilled water (from 0.1 N NaCl). The melanophore was completely contracted in twenty minutes. In this case the migration of the pigment granules was followed by means of an ocular micrometer carrying a moveable scale. An empirically selected line of the micrometer scale was kept tangent to the termi- nal pigment granules by turning the adjusting screw of the ocular. The motion of the serew was transmitted to a set of pulleys and a heart lever and the curve was recorded on.a kymograph in the usual way. The actual path of the pigment migration was 0.104 mm. which makes the magnification of the reproduced figure approximately 534. A detailed description of this apparatus will appear shortly in the American Journal of Physiology. MELANOPHORE A TYPE OF SMOOTH MUSCLE CELL 205 Recently® I have found that the alkaline earths produce a contraction of the melanophores in Fundulus. The time for this contraction in isotonic solutions of the neutral chlorides varies in the order Baer Paul Bert: Arch. de Physiol., 1869, 2, 547. * Graber. Grundlinien zur Erforschung des Helligkeits- und Farbensinnes der Tiere. Prag, 1884. 218 JACQUES LOEB AND HARDOLPH WASTENEYS darkness to light gather under the red; and animals which prefer light to darkness gather under the blue glass. This led him to enunciate the law that animals which are ‘fond’ of light are also ‘fond’ of the blue; and animals which are ‘fond’ of the darkness are also ‘fond’ of the red. We see again the tacit assumption that animals which collect under blue glass do so because they are ‘fond’ of this type of light, while animals which collect under the red light do so because they are ‘fond’ of this type of light. The field of animal reactions received a different interpretation by Loeb,‘ who showed that these results can be explained on a purely objective basis without our ascribing to lower organisms sensations the existence of which we can neither prove nor disprove. Loeb showed that the phenomena observed by Bert, Graber, and others can be explained on the assumption that the light automatically orients the animals or determines the direction in which they move, there being two classes of animals, one class being automatically compelled to move to the source of light, the other being compelled to move in the reverse direc- tion; and he pointed out that this phenomenon is the same as the heliotropic reaction in plants, the stem of plants bending to the source of light, the roots bending away from it; or the swarm- spores of algae moving to or from the light. Accordingly he designated the animals going to the light as positively heliotropic, those going away from the light as negatively helotropic. As this bending effect in the plant is a purely automatic orien- tation of the plant, brought about through the influence of the light, so in the animals we are, according to Loeb’s theory, deal- ing only with an orienting effect of the light for the explana- tion of which merely physicochemical conditions are adequate; without our being compelled to introduce hypothetical sensations as a necessary link in the mechanism. This purely mechanistic conception of the motions of animals to or from the light has recently received a new support by the invention of heliotropic machines by Mr. John Hays Hammond, Jr., in which the two Loeb, J. Der Heliotropismus der Tiere und seine Uebereinstimmung mit dem Heliotropismus der Pflanzen. Wiirzburg, 1890. Sitzungsber. d. Wiirzburger physikal-med. Gesellsch., Jan. 1888. HELIOTROPIC REACTIONS—ANIMALS AND PLANTS 219 retinas are replaced by selenium wire, these machines following a lantern in the dark in the same way as a positively heliotropic animal. Tt was easy to interpret the phenomena found by Graber from this hehotropic viewpoint. The botanists had long ago shown that positively heliotropic plants bend readily to the light when behind a blue screen, while they do not do so or only very slowly when behind a red screen; from which they concluded that the rays going through a blue screen had a higher heliotropic efficiency than the rays passing through a red screen. If the animals were, as Loeb stated, merely positively or nega- tively heliotropic the light going through blue glass should act like more intense light than that going through red glass, and hence negatively heliotropic animals should gather in the red, positively helotropic animals in the blue. He could show by a series of experiments that this statement was correct. In this way, purely objective methods and explanations were given for the arbitrary assumption of Graber and the other anthropo- morphic biologists that animals moved to or from the light because they were ‘fond’ of light. The advantage of this change in viewpoint lies in the fact that it opened this field to the methods of exact experiments and meas- urements without which no progress is possible; while the at- tempt to explain reactions by a hypothetical ‘fondness’ of animals for ight or by hypothetical hght sensations barred the way to the exact type of investigation. Recently, however, the old anthropomorphie viewpoint has been resumed by the ophthalmologist Hess,® who has tried to show that all the animals from fish downward suffer from a visual deficiency, namely total color blindness. As a criterion for the presence or absence of color sensations Hess uses (very arbi- trarily in our opinion) the heliotropic reactions of animals. Thus in 1909 he confirmed Bert’s observation that Daphnia col- lect in the yellow-green part of the spectrum but gave it a dif- ferent interpretation. By calling attention to the fact that the vellowish-green, which is heliotropically most efficient for Daph- ° Hess. Gesichtssinn. Handb. d. vergleich. Physiol., 1913, 4, 555. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 220 JACQUES LOEB AND HARDOLPH WASTENEYS nia, appears brightest to the human inflicted with total color blindness, he thinks he has proved that Daphnia is also totally eolor blind. The number of objections to this kind of reasoning is considerable.* Nobody has yet proved that the heliotropic reactions of ani- mals are determined or even accompanied by any sensations of brightness and it is difficult to see how such a proof can ever be furnished. It is plainly unwarranted to assume that every motion of animals induced by light is accompanied by or is the expression of sensations of brightness or of color. The excised iris of the shark (and of other animals) contracts under the in- fluence of illumination and Magnus has shown that the yellowish- green part of the spectrum is most efficient in this case. It would be arbitrary, to say the least, to state that the excised iris has sensations of brightness and that these sensations make it contract; and yet it is difficult to see why such an assumption should be more arbitrary than a similar assumption in the case of the flagellate Chlamydomonas or the heliotropic larvae of Balanus. One wonders also whether we are supposed to assume that Hammond’s heliotropic machines are guided by sensations of brightness or of color. The assumption of Hess might be given some consideration if it could be shown that totally color blind human beings are posi- tively heliotropic, 1. e., are irresistibly drawn to the source of light; but nobody has ever heard of such a case. The human being is the only one about whose sensations we have definite knowledge and as long as we are unable to prove for the human a connection between positive heliotropism and the sensations of brightness we have no right to take such a connection for granted in the lower animals. One wonders also what interpretation is to be put on other tropisms, such as galvanotropism or geotropism, if we accept the validity of Hess’s viewpoint, since it is only logical to treat all the ®° An excellent criticism of Hess’s ideas and experiments has been given by W. F. Ewald, Arch. f. Entweklngsmech., 1915, 37, 581. We are using some of his arguments in this paper. HELIOTROPIC REACTIONS—ANIMALS AND PLANTS ye | tropisms from the same general viewpoint. What sensations are aroused in a Paramaecium which is forced to swim to the cathode under the influence of the galvanic current, or in a Palaemonetes which is forced to swim or walk to the anode? Since Hess starts with an arbitrary assumption, namely that the heliotropism of lower animals is due to their sensations of brightness and that they are totally color blind, it is not unex- pected to see him come into conflict with facts in more than one direction. v. Frisch’ has shown by experiments which appear to us conclusive that bees (which are also positively heliotropic and which according to Hess are totally color blind) can be trained to go to yellow or blue eardboards distributed among similar cardboards of different shades of gray; while they can not be trained to go to definite shades of gray under similar conditions. Even in Daphnia v. Frisch and Kupelwieser,’ and Ewald? have been able to demonstrate selective effects of wave lengths different from those found in the totally color blind human. A second conflict between Hess’s view and reality is due to the fact that the most efficient part of the visible spectrum is not the same for all heliotropic organisms. It is known through Blaauw’s experiments that the heliotropic curvatures of the seedling of oats are produced most rapidly in the blue part of the carbon are spectrum. This should force Hess either to the conclusion that the seedlings of oats do not suffer from total color blindness, since the most efficient part of the spectrum for the totally color blind is in the yellowish-green; or to the assumption that only plants are heliotropic, but that animals which show the same reactions to ight are not heliotropic. Hess chooses the second alternative by stating that plants are heliotropic, while animals are ‘lamprotropic’ (Aapmpés = (bright),!’ 1.e., in plants the helio- tropic curvature occurs purely automatically, while animals bend or move to the source of light because it is ‘bright.’ It 7y. Frisch. Der Farbensinn und Formensinn der Biene, Zool. Jahrb., 1914, 35, 1, Abt. f. allg. Zool. u. Physiol. Sv. Frisch and Kupelwieser. Biol. Centralbl., 1913, 33, 517. : 9Ewald. Ztschr. f. Psychol. u. Physiol. d. Sinnesorg., 1914, 48, Abt. 2, 285. ‘0 Hess, loc. cit., pp. 708 and 709. DoD, JACQUES LOEB AND HARDOLPH WASTENEYS would be difficult to invent a nicer example of reasoning in a circle; since Hess’s assumption that animals have the sensation of brightness is based upon the fact that they move to the light. Yet we will try to follow Hess even into this circle and select a case already mentioned in a previous note!! and to which we shall return in this paper, namely the case of two green flagel- lates, Euglena viridis and Chlamydomonas pisiformis, which are strongly heliotropic, but, being unicellular organisms, of course have no eyes. For Chlamydomonas the place of greatest efficiency in the spectrum is in the region of yellowish-green, for Euglena it is in the blue. If we follow Hess we must logically conclude from this that Chlamydomonas suffers from total color blindness (although it has no eyes), that it is not heliotropic but ‘lamprotropic,’ and that it is an animal; while its cousin Euglena has either a highly developed color sense or is heliotropic and is a plant. Hess! thinks it is inconsistent for Loeb to deny the justifica- tion of the assumption that all heliotropic animals are totally color blind and at the same time to state that phenomena of heliotropism are identical in animals and plants. Hess over- looks the fact that the two statements rest on an entirely different basis. The statement that positively heliotropic animals go to the light because they are totally color blind is as we have seen not a fact but an unnecessary and arbitrary assumption which is in conflict with the facts and not even justifiable on the basis of mere analogy, since totally color blind humans are not posi- tively heliotropic. The fact that the region which is brightest to the totally color blind human is at the same time most efficient in certain heliotropic animals admits or demands, as we shall see, an entirely different interpretation. On the other hand, the statement that heliotropic reactions in animals and plants are identical is merely the expression of the actual observations. Thus Loeb has been able to show that sessile heliotropic animals react to one-sided illumination Just like sessile plants, namely by bending towards the source of 11 Science, 1915, 41, 328. Loc cit., p. 709 HELIOTROPIC REACTIONS—ANIMALS AND PLANTS 223 light until the axis of symmetry of their photosensitive organs goes through the source of light (provided only one source of light is given); while movable plant organs, e.g., the swarmspores of algae move to or from the source of light and collect on the side of the hght (or on the opposite side) just as do motile helio- tropic animals. For the heliotropie reactions of both animals and plants the validity of the law of Bunsen and Roscoe has been proved" and the sense of heliotropic reactions in both groups can be reversed by similar means." It would be artificial to state that because the ones are termed animals and the others plants the identical phenomena in both must be different. Such a view might have been considered at the time of Linné, but today we know that the mechanism of life phenomena in animals and plants is essentially the same. While modern biology, especially since Claude Bernard and Hoppe-Seyler, has tried to establish the essential identity of life phenomena in plants and animals, Hess apparently expects biologists to overlook the progress made in biology and return to the Linnéan viewpoint. In order to maintain his artificial barrier between animals and plants he in- sists that the wave length which is most efficient in the helio- tropic reactions in plants is different from the one most efficient in animals. But even if this were a fact, it would not justify his assumption, since the theory of heliotropism only states that organisms are automatically oriented by the light so that sym- metrical elements of their photosensitive surface are struck at the same angle by the light (or that symmetrical elements re- eelve an equal amount of illumination during a properly chosen unit of time). Whether in one case the yellowish-green, in another the blue light is more efficient is secondary. As a matter of fact, there are heliotropic animals for which the blue rays are as efficient as they are for plants; and there are unicellular organisms, for which the optimum les in different parts of the spectrum. 13 Loeb and Ewald. Centralbl. f. Physiol., 1914, 27, 1165; Ewald, Ztschr. f Psychol. u. Physiol. d. Sinnesorg., 1914, 48, Abt. 2, 285. 14 Loeb. Arch. f. d. ges. Physiol., 1906, 115, 564. 224 JACQUES LOEB AND HARDOLPH WASTENEYS From the viewpoint of objective science we accept the fact that in some heliotropic organisms the place of highest efficiency is in that region of the spectrum which for the totally color blind is the brightest, but on this we put a different interpretation, namely the following. The sensations of brightness in the totally color blind human are determined by the rapidity with which visual purple is bleached by the light. The region in the yellow- ish-green in the carbon are spectrum appears brightest to the totally color blind human because this region \ = 526 up has according to Trendelenburg the greatest bleaching effect. As- suming that heliotropic reactions are also due to a photochemical effect, the fact that in certain organisms the region not far from ) = 526 wu is the most efficient in calling forth heliotropism means simply that the photosensitive substance responsible for the heliotropic reaction in these organisms has one peculiarity in common with the visual purple, namely that it is also most sensitive to a region not too remote from \ = 526 uu; the two substances may possibly be identical, but this would require a definite proof. The fact that the optimal effect for other organ- isms lies in the region of blue would indicate that the photo- sensitive substance in these animals is in all probability different from visual purple. If the effect of light in causing heliotropic reactions were other than chemical we still should be compelled to find a physicochemical and not a psychological explanation for the different heliotropic efficiency of different wave lengths. The question to which we intend to confine ourselves in this paper is a very simple one, namely: Is it true that a sharp line of demarcation exists between animals and plants in that sense that for the heliotropic reactions of plants the blue is most effective, while for the heliotropic reactions of all animals a region in the yellowish-green is the most efficient, as Hess claims? In this paper we shall deal with motile organisms, namely first the two unicellular green organisms Euglena and Chalmydomonas and the larvae of two animal forms, of the annelid Arenicola and of the crustacean Balanus eburneus. i) i) I | HELIOTROPIC REACTIONS ANIMALS AND PLANTS Il. METHODS When we wish to determine where the most efficient spot of the spectrum for freely moving animals lies, we must realize that in order to get reliable results we must work with organisms which are both very small and very sensitive to light. The organisms must be small so that a large number can be crowded into a narrow region of the spectrum; if this condition is not fulfilled it is extremely difficult if not impossible to make statements con- cerning the relative efficiency of the different parts of the spec- trum which are of sufficient accuracy. Thus attempts to deter- mine the most efficient spot for the heliotropic efficiency of the spectrum for young fish or larger insects can only yield crude approximations. The second condition is that the animals must be very sensitive to light; since Loeb has shown in former papers that only in the case of extreme sensitiveness will the animals go directly to the source of light, while if the sensitiveness is small the animals may go in very irregular paths although the sum total of the motions towards the source of light will prevail. It would be impossible to get a definite result with animals of this kind. Thus experiments on the relative efficiency of dif- ferent parts of the spectrum with young fish or other animals which are only moderately sensitive are very unreliable. When we are dealing with positively heliotropic animals dis- tributed in an oblong trough exposed to a carbon are spectrum the animals will move towards the source of light independently of the nature of the rays by which they are struck; provided in- tensity and frequency of the waves are above the threshold of heliotropic efficiency. If the animals are evenly distributed in the trough at the beginning of the experiment they will all move towards the source of light and the result should be that at the end of the experiment the animals should all gather equally at the front wall of the trough and their density should be the same on this front wall in the violet-blue and green provided that the rays are sufficiently effective. On this basis it should be difficult to tell whether the blue or the green is more efficient. Since, however, some scattering of light occurs from the surface of the 226 JACQUES LOEB AND HARDOLPH WASTENEYS animals, some blue light will also reach the animals in the green and vice versa. In this way a comparatively denser gathering of animals may occur in that part of the spectrum which is more effective. It is obvious, however, that this method is not very exact since in an aquarium with animals the scattered light from one part of the spectrum can chiefly reach only those individuals which are not too far from this spot. nS The second method consisted in the comparison of the relative efficiency of two narrow parts of the spectrum. It was described briefly in a former note. A earbon are spectrum, about from 18 to 23 em. wide, was thrown on a black screen SS (see fig. 1) with two slits a and b in the two different parts of the spectrum which were to be com- pared in regard to their heliotropic efficiency. The two beams of light passing through the slits are reflected by the two mirrors M and M, into the square glass trough in such a way as to strike the same region g of the back wall of the trough. The glass trough is surrounded by black paper except at R and R,, where the two beams of light enter from the mirrors. Before the ex- HELIOTROPIC REACTIONS-——ANIMALS AND PLANTS DOG periment begins, all the organisms are collected in the spot g with the aid of an incandescent lamp. As soon as the spectrum is turned on, these organisms are simultaneously exposed to two different beams of light which come from the two mirrors VM and M,. When one type of light, e.g., that from M, is much more efficient than the other coming from M,, practically all the or- ganisms are oriented by the light from M and move toward this mirror, collecting in the region R. When the relative efficiency of the two types of light is almost equal the organisms move in almost equal numbers to R and R;. By using as a standard of comparison the same region of the spectrum and _ successively altering the position of the other slit in the spectrum we were able to ascertain with accuracy the relative efficiency of the dif- ferent parts of the spectrum for the two forms of organisms. When the two parts of the spectrum which are to be compared are very close to each other it is necessary to deflect the beams with the aid of deflecting prisms, before they reach the two mirrors. Ill. THE DISTRIBUTION OF ORGANISMS IN THE CARBON ARC SPECTRUM The spectrum used was a carbon are spectrum and its visible part had a width varying from 18 to 23 em. in different experi- ments. We used very dense cultures of Huglena viridis and filled the trough with this greenish suspension of Huglena. After an exposure varying in length between 30 and 180 minutes the results were ascertained, and in some cases the trough was photographed. Figure 2 gives the photograph of the trough after 30 minutes’ exposure. A very dense mass of Euglena was gathered at the bottom of the trough in front between violet (410 pu) and green (515 wu). In the photograph this mass is visible as a thick, dark, horizontal streak at the bottom. In addition some vertical streaks of organisms are visible in the blue and indigo. The reader will recognize how difficult it is to ascertain the most efficient wave length by this method. We can only say the blue is the most efficient light and the wave lengths > 515 wy are practically without orienting effect. 228 JACQUES LOEB AND HARDOLPH WASTENEYS Figure 3 gives a photograph of the distribution of the organisms in another experiment after 3 hours in the same spectrum. This longer exposure gives a slightly better result. The dense gather- ing at the bottom indicated by a horizontal streak is in the blue VIOLET INDIGO, BLUE ! GREEN YELLOW ORANGE 410 450 493 SI5 VIOLET INDIGO | BLUE GREEN | YELLOW ORANGE WAVE LENGTH pup 452 458 506 533 Ua) rs between 458 and 506 wu. It ends at about 533 and becomes very faint at 452 uu. The vertical streaks on the wall, indicat- ing the sticking of the organisms to the front wall, occur again in the blue. These experiments permit us to draw only the con- HELIOTROPIC REACTIONS—ANIMALS AND PLANTS 229 clusion that the blue is more effective than the green, yellow, red, indigo, and violet; but they do not permit a more definite statement. Engelmann states that he found a strong gathering of Euglena in the blue between 470 and 490 uy in a spectrum. When we made similar experiments with Chlamydomonas pisi- formis, which is also a chlorophyll-bearing unicellular organism like Euglena, we noticed that the gathering went much farther towards the yellow ending at about \ = 560 or 570 uy. The region of maximal gathering seemed to be at about \} = 520 wu. A similar result had previously been obtained with Chlamydo- monas by Loeb and Maxwell.! The method is still less definite with larger and rapidly moving animals, and yet it is mainly by such experiments that Hess tried to prove that the most efficient part of the spectrum for heliotropic animals is identical with that which appears brightest to the totally color blind. IV. EXPERIMENTS WITH THE TWO-BEAMS METHOD With the aid of two slits (fig. 1) two narrow strips of the spec- trum were cut out. Their width was such that in the green part of the spectrum the difference in the wave length of the extreme rays that passed through was about 10 wu. With the aid of prisms and mirrors these two parts of the spectrum were made to converge to one spot in the trough where the organisms had previously been collected. It was ascertained which of the two beams of light was more powerful. In detail the experiments were as follows. The trough was surrounded with black cardboard in which there was one open- ing at that spot where it was intended the animals should col- lect. Then an incandescent lamp was turned on in front of this opening which caused all the organisms to collect at that spot of the trough. When this happened, the spectrum was turned on and the incandescent lamp turned off and the animals were exposed to the two beams of light a and 6 selected for com- 15 Hngelmann. Arch. f. d. ges. Physiol., 1882, 29, 387. 16 Loeb and Maxwell. Univ. Cal. Pub., 1910, Physiol., 3, 195. 230 JACQUES LOEB AND HARDOLPH WASTENEYS parison, after another black cardboard with openings at R and R,, to let the two beams of light pass, had been put over the box, and the first cardboard enclosing the animals had been removed. The organisms collected at g were now under the influence of Euglena viridis Wave lengths in wy these two beams coming from different directions. As stated before, they moved towards R and FR, according to the selective efficiency of the two beams of light. The readings were taken after from 15 minutes to 3 hours or more to make sure that the results were permanent. Before a new experiment was made HELIOTROPIC REACTIONS—ANIMALS AND PLANTS Zo the organisms were all scattered equally again in the trough. Fresh organisms were used every day. As stated before, one of the two parts of the spectrum was the same in a group of experiments while the other changed in suc- cessive experiments throughout the spectrum. In figure 4 the results of 21 experiments with Euglena are plotted. The part of the spectrum which was stationary was situated at about 470 uu which previous experiments had led us to believe was the most efficient wave length. The different degrees of black- ness indicate the denseness of the gathering; the more animals gathered in one spot the darker the oblong representing the ex- periment. We notice that the oblongs at the region 470 up are with two exceptions much darker than all those at other wave lengths. These results indicate that the greatest efficiency is possessed by the rays between 460 and 490 wu for Euglena viridis. The total of all experiments is represented in diagram IT, figure 7, where the distribution of this efficiency through the carbon arc spectrum is plotted for this form. The greatest efficiency is in- dicated by the greatest blackness, the greater blackness indicat- ing the denser gathering. The results with Chlamydomonas were entirely different. In figure 5, the region between 470 and 480 wu was again constant in each determination, while the region compared with this var- ied in each experiment. It is obvious that in contradistinction to the experiments on Euglena the region between 460 and 480 uu Was less efficient than the region from 490 to almost 560 uz. We, therefore, started another series of determinations in which the region about 534 wu was constant (fig. 6). We now found that this region was more efficient than any other region in the visible spectrum. The experiments show that the maxi- mal efficiency lies for Chlamydomonas, approximately in that part of the spectrum which appears brightest to the totally color blind human. In the same way long series of experiments were made with the newly hatched larvae of Arenicola, an annelid. The experi- ments in this form suffer from the difficulty that the larvae have a tendency to stick to the glass walls of the trough. We found Dae JACQUES LOEB AND HARDOLPH WASTENEYS that if we keep them in the dark before using them they are more sensitive to ight and less liable to stick so soon, and such animals gave clearer results. The most efficient part of the spec- trum was situated in the bluish-green in the region of about ) = 495 wu. And finally, experiments with this method on the larvae of Balanus eburneus yielded the result that the most ef- ficient part of the spectrum lies between \ = 560 and \ = 578 pp.!? Chlamydomonas pisiformis. Dill. Wave lengths in wu Sit The relative efficiency of the different parts of the carbon are spectrum for different organisms is plotted in figure 7. The up- per line gives the wave length and below are found the relative efficiency of the various wave lengths as revealed by the two- beams method for Eudendrium ramosum, Euglena viridis, larvae ‘7 These results agree with previous observations by Loeb and Maxwell, Univ. Cal. Pub., 1910, Physiol., 3, 195. HELIOTROPIC REACTIONS—-ANIMALS AND PLANTS Dae of Arenicola, Chlamydomonas pisiformis, and larvae of Balanus; the greater darkness indicating the greater efficiency. The two most striking facts to us are, first, that there are animals for which the most efficient part of the spectrum is in the blue, Chlamydomonas pisiformis. Dill. Wave lengths in wu namely Eudendrium and the larvae of Arenicola; and that among the flagellates, Chlamydomonas is most sensitive to yellowish- green, while the closely related Euglena is most sensitive to the blue. The second striking fact is that the place of greatest efficiency does not seem absolutely identical in the organisms of the same 234 JACQUES LOEB AND HARDOLPH WASTENEYS group. In the line VI of figure 7 is represented the relative brightness of the various parts of the spectrum for the totally color blind human after Helmholtz. A comparison with the rela- tive efficiency of the various wave lengths for Balanus larvae and Wave lengths in wu 592 400 4075 416 4244325441 449 4575 460 474 486 495 10 525 543 500 578 596 615 631 puoempynnmsie | ||| TT | | | | | | EUGLENA VIRIDIS IARENICOLA LARVAE | sheet sal | | j lethal CoLok Bi BLIND HUMAN VIOLET INDIGO BLUE GREEN YELLOW ORANGE lh Chlamydomonas (IV and V, fig. 7) shows that the maximal ef- ficiency is not entirely identical in all three cases. We are not prepared to state whether this is entirely due to the inadequacy of the methods. HELIOTROPIC REACTIONS —ANIMALS AND PLANTS 239 DISCUSSION AND SUMMARY OF RESULTS 1. As stated in the previous papers, the validity of the Bun- sen-Roscoe law for the heliotropic reactions of some (and _ possi- bly many or all) organisms suggests that these reactions are due to a chemical action of the light. There seem to exist two types of heliotropie substances (or elements), one with a maximum of sensitiveness in the yellow-green region, and the second with a maximum of sensitiveness in the blue. 2. It would be wrong to state that the one type of photosensi- tive substances is found exclusively in plants and the other ex- clusively in animals. As a matter of fact, our experiments have shown that the animals Hudendrium ramosum and (the larvae of) Arenicola are most sensitive to blue light, which is also most effi- TABLE I NAME OF ORGANISM REGION OF GREATEST HELIOTROPIC EFFICIENCY we EHudendrium ramosum............-. 460-480 Bed ene NAN ENG NiSSag meee alee ore mia ee toe | 460-490 Wanvae oroAmrenicolan. > esse see eee lee about 495 Chlamydomonas pisiformis............ about 535 Larvae of Balanus eburneus........... 560-578 cient for the seedlings of the plant Avena (according to Blaauw) ; while the larvae of Balanus, Daphnia, and probably many other animals are most sensitive to the yellow-green or yellow part of the spectrum. Of the two green flagellates, Euglena viri- dis and Chlamydomonas pisiformis, the former is most sensitive to blue, the latter to greenish-yellow. The two groups of photo- sensitive substances (or elements) are, therefore, distributed in- dependently of the boundaries between animals and plants. It is quite possible, however, that plants are more generally sensi- tive to the blue rays of the spectrum, while among animals those may prevail that are more sensitive to yellowish-green or yellow. 3. Table I states the wave lengths in the carbon are spectrum for which the different organisms investigated by us are most sensitive. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 236 JACQUES LOEB AND HARDOLPH WASTENEYS Visual purple is bleached most rapidly by light of the wave length of about 530 uu (according to Trendelenburg). Neither Chlamydomonas nor the larvae of Balanus show their maximal sensitiveness exactly at this point; whether the slight deviation is only due to the imperfections of the experimental method or due to the fact that the photosensitive substance is not identical with visual purple can not be decided on the basis of the material which is at present available. We find likewise that those organisms which react best to the blue part of the spectrum do not all have their greatest sensi- tiveness in the same spot of the blue. Thus the seedlings of oats are most sensitive to a region of } = 466 wu, the animal Eudendrium and the flagellate Euglena for a region near \ = 460 to 490 uu, while the larvae of Arenicola are most sensitive to X = 495 wp. THE REACTIONS OF THE MELANOPHORES OF AMBLYSTOMA LARVAE—THE SUPPOSED INFLUENCE OF THE PINEAL ORGAN HENRY LAURENS Osborn Zoological Laboratory, Yale University SIX FIGURES INTRODUCTION Although the literature dealing with the chromatophores is very voluminous, nevertheless there are still many points con- cerning which our knowledge is far from complete. One of these, and a fundamental one, because it lies at the foundation of our comprehension of the physiology of the pigment. cells, is the relation between them and the nervous system. We know much about the reactions to various stimuli of the echroma- tophores of many animals but when we come to compare them it is found that they are so diverse that it is almost impossible to lay down any general rule which will cover all cases. In some animals light causes a contraction of the pigment cells, in others it has no noticeable effect, and in still others it pro- duces expansion. Attempts have been made, of course, to ex- plain these different results, and in the excellent review by Fuchs (14) this is often done, with, it must be admitted, not always marked success. The particular case in which we are interested concerns the reactions to light and to darkness of the melano- phores of Amblystoma larvae. Babak (10) obtained very interesting results regarding the melanophores of these larvae. He found that there was a dif- ference between the reactions of the pigment cells of normal and blinded Axolotl (A. mexicanum Cope, A. tigrinum Laurenti) larvae to light and darkness. In diffuse light, according to Babak, the melanophores of normal seeing larvae contract, those of 237° 238 HENRY LAURENS blinded larvae expand. In darkness the melanophores of normal larvae expand, while those of blinded larvae contract. This opposite reaction of the melanophores of normal and blinded larvae he found, however, not to occur until the larvae had attained a certain stage of development, about 17.0 mm. long. Babak believes, (Laurens 15, p. 592) that before this period the retina has not acquired the pigment motor function which it later has, so that the melanophores simply respond to direct stimulation, which is therefore the same in both normal and blinded individuals. After this period, by means of the control which the eyes have gained through the central nervous system, the sense of the reaction of the melanophores is reversed, the effect of indirect stimulation through the eyes being opposite to that of direct. Babak’s explanation of why there should be this difference between the reactions of normal and blinded lar- rae, or in other words, why the effect of indirect stimulation of the melanophores should be opposite to that of direct is briefly as follows (Laurens, 715, p. 623): the chromatophores of normal Axolotl larvae in both phases of their movement—expanding and contracting—are governed by the central nervous system, and this double innervation is conditioned upon the retinae which have opposite influences upon the nervous system accord- ing as to whether they are illuminated or darkened. The darkened retinae exert a positive influence on the chromato- phores through the nervous system, just as the illuminated retinae do, but in the reverse direction. The destruction of the retinae has an entirely different result from that obtained by darkening them. In other words, the retinde in complete darkness are active and exert a positive influence which is di- rectly opposite to that caused by illumination. Babak, however, does not believe that these two opposite effects of the retinae upon the chromatophores are either of them inhibitory, but that they are two kinds of tonic influences. The impulse bringing about the expansion of the chromatophores originates in the darkened retinae, and is so strong that it over- comes the tendency of the darkened chromatophores to contract and brings about their expansion. On the other hand, the 4 REACTIONS OF MELANOPHORES OF.AMBLYSTOMA 239 impulse for the contraction of the chromatophores originates in the illuminated retinae and is in turn so strong that it overcomes the tendency of the illuminated chromatophores to expand and brings about their contraction. The results which were obtained from a study of the reactions of the melanophores of larvae of A. punctatum and of A. opacum (Laurens ’15) were such that this explanation of Babak’s could not be applied to them. They threw no doubt, however, on the assumption that both phases of the movement of the melano- phores are normally under the control of the nervous system by means of the eyes, although one of the influences of the retinae must be admitted to be inhibitory, and opposite in effect to an impulse which causes the pigment cell to contract. The results of my work showed that the melanophores of nermal and eyeless larvae react primarily in identically the same way, expanding in light and contracting in darkness, the only difference being that the reactions come about more quickly in the normal than in the eyeless, larvae (p. 585 and table 2). Secondarily, however, the melanophores of the normal larvae are found to be in the opposite conditions in both light and darkness to what they were in before, for after having been kept for from three to five days in light the melanophores are contracted, and after having been for five days or more in dark- ness, the melanophores are expanded. The melanophores of the eyeless larvae do not show these secondary reactions. _ It was assumed (pp. 624-625), to explain these secondary reactions of the melanophores of normal seeing larvae, that, although the primary effects of indirect stimulation of the melanophores through the eyes were the same as those of direct stimulation, in the case of light the constant illumination or stimulation of the retinae had the result of causing impulses to be started, the end effects of which were opposite to those of direct stimulation and in this way the secondary contraction of the melanophores was brought about. ‘These impulses were supposed to have their immediate cause in certain photo-chemical changes taking place in the retinae. In the case of darkness the same thing was supposed to take place, in that, due to the 240 HENRY LAURENS long continued absence of light, chemical changes in the retinae started impulses which reaching the melanophores caused them to secondarily expand. Briefly, the seat of the causes of the secondary changes were assumed to be in the retinae, for which there was abundant experimental proof, just as Babak assumed that the reactions of the melanophores of the normal larvae were due to the influence of the eyes, which were opposite in effect to direct stimulation of the melanophores themselves. Now Fuchs (14, p. 1545) considers that Babak’s explanation of the differences between the reactions of the melanophores of normal and blinded Axolotl larvae is unsatisfactory and seeks to explain it by advancing a theory based on the results of von Frisch’s work on the minnow Phoxinus (von Frisch ’11, p. 374). As his chief objection to Babak’s explanation, he points out that the latter’s contention that the expansion of the pigment cell is as much an active process as its contraction, and that there- fore the condition of rest is one of medium pigment contraction, is untenable. For all that is known concerning the distribution of pigment in the chromatophore makes it necessary to regard the expanded condition as that of rest, while a condition of medium contraction must be considered as the result of a tonic condition of excitation, no matter where the tonus arises. Fuchs therefore offers the following explanation of Babak’s re- sults, seeking in the first place to show why the melanophores of young Axolotl larvae expand in the light. Substances, he says, which are perhaps products of inner secretions, but which at any rate are the results of the ordinary processes of life, arise and cause the melanophores to contract. Now if the young larvae are removed from all possibility of external stimulation, then under the influence of these metabolic products the melanophores contract. This is the reason why the melanophores are con- tracted in darkness. But if a light stimulus, at this time, be allowed to act on the parietal organ, then through the stimulation of this organ an impulse is started which inhibits the endogen- ously produced contraction and the melanophores expand. Grad- ually, as the larvae grow older, the eyes develop and gain an influence (i.e., a pigment motor function) over the pigment cells. REACTIONS OF MELANOPHORES OF AMBLYSTOMA 2a The stimulation of the retinae (illumination) produces a con- traction of the melanophores, and as the eyes continue to develop and finally gain the upper hand in the sense life of the animal, so the impulses started by illumination of the retinae become stronger and finally overcome entirely the inhibitory influence of the parietal organ, so that in light the melanophores are con- tracted. Now, if the influence of the eyes is removed, by blind- ing the larvae, then the parietal organ is again in complete con- trol, and the melanophores therefore expand in the light. The remainder of this theory of Fuchs is as interesting as what has gone before. He goes on to say, that whether all larvae have a functional parietal organ, and whether the functioning power decreases as the development of the animal proceeds— which he considers probable—can be learned only experimentally. If this is found to be true then, he believes, it will be easy to understand why in adult animals the eyes have no particular influence over the color changes. For when the eyes have over- come the opposite influence of the parietal organ then after the extinction of its influence the function of the eyes must naturally in this respect also be less, because their antagonist, the stimu- lation of which causes the pigment cells to expand, is lacking. In other words, Fuchs would claim that the pigment motor func- tion of the eyes, by means of which contraction of the pigment cells is brought about, is developed to offset an opposite effect of the parietal organ, and after this has been overcome it dis- appears. Moreover in his theory, nothing is said concerning the direct stimulation of the pigment cells by light, and no indi- cation is given as to just how the chromatophores are stimulated after the influence of the eyes as well as that of the parietal organ has been extinguished. EXPERIMENTAL The work, the results of which will be given in the succeeding pages, was undertaken to put to experimental test thistheory of Fuchs. More particularly to see whether the parietal organ in young normal seeing larvae can be shown to exert any in- hibiting influence upon the melanophores, which later disappears 242 HENRY LAURENS after the pigment motor function of the eyes has developed, but which in eyeless larvae is still present, and remains so until the larvae reach a certain stage of development when, according to the letter of Fuchs’ theory, it would be expected to decrease and finally disappear. It should be mentioned, of course, that this theory of Fuchs was put forward to explain the results obtained by Babak con- cerning the reactions of the melanophores of the Axolotl larvae. As has been pointed out, these are different in certain points from those obtained by me with the larvae of A. punctatum and opacum. But the main fact with which Fuchs’ explanation is concerned is the same, viz. that in larvae deprived of their eyes the melanophores expand when the larvae are illuminated. If his explanation, that this is so because of the stimulation of the parietal organ, by virtue of the fact that impulses going out from it inhibit an endogenously produced contraction of the melanophores, holds for larvae of the Axolotl, it must hold also for larvae of A. punctatum and opacum. The same applies to his belief that when the eyes are present their stimulation starts impulses which are opposite in effect to those sent out by the parietal organ and stronger, so that these are rendered of no avail. To test this hypothesis of Fuchs several methods of experi- mentation were employed and these will be taken up in order. But before proceeding to describe and discuss their results it seems desirable that some idea be had concarning the develop- ment of the so-called parietal organ, and its position at this time and later. To do this it is hardly necessary to more than call attention to the accompanying figures. In figure 1, which isa view of the extreme anterior portion of a sagittal section of a 5.3mm. larva of A. punctatum there is seen the beginnings of the development of the epiphysis and paraphysis. In figure 2, of a 6.6 mm. larva, development has proceeded only a little further. In figure 3, which is of an 8.0 mm. larva, the epiphysis and para- physis are both seen as evaginations, the former of the dienceph- alic wall, the latter of the telencephalic, the two cavities of the brain being partially divided by the velum transversum. In REACTIONS OF MELANOPHORES OF AMBLYSTOMA 243 figures 4 and 5 (of larvae 14.5 and 34.0 mm. long respectively), the later development of these parts of the roof of the brain are shown. ‘The condition seen in figure 5 is typical of older larvae and of adult Amblystoma, although it must be said that the open- Fig. 1 The anterior end of asagittal section of a5.3mm. larvaof A. punctatum. Ep.A., epiphysal arch; P.A., paraphysal arch; P.V.A., post-velar arch; V., velum. Fig. 2 The anterior end of a sagittal section of a 6.6 mm. larva. ing into the stalk of the epiphysis cannot always be made out with great distinctness, so that there is a question as to whether it may not later be solid instead of hollow as here represented. In no case, however, was the epiphysis found not attached to the brain. Its cavity is more or less incompletely divided by septa. 244 HENRY LAURENS The paraphysis, which is much larger and more conspicuous than the epiphysis, has a wall of a single layer of cells and a large irregular cavity with branching tubules with which the blood vessels of the choroid plexus are in intimate relation. Fig. 3 The anterior endof asagittal section of an8.0mm. larva. H.,epiphysis; P., paraphysis; V., velum. These figures also serve the purpose of showing that the epiphy- sis, as in other Urodeles, is relatively poorly developed in Ambly- stoma and moreover that there is no pineal or parietal organ or eye. The paraphysis on the other hand reaches a high degree of development and with the surrounding blood vessels can be seen through the skin and brain case of the larvae, particularly REACTIONS OF MELANOPHORES OF AMBLYSTOMA 245 when the melanophores are contracted or when they are scarce. Figure 6 is given to show this, where, it will be admitted, it has very much the appearance of a brow-spot or ‘‘Scheitelfleck”’ and might be taken for such. From reading over the literature one is soon forced to the con- clusion that it seems doubtful whether a parietal organ in any Urodele could have an influence on the reactions of the melano- rast = Wet ~~ a ~ Cs 68 e FEEX Lass je eo Boerpeyy) Fig. 4 The anterior end of a sagittal section of the headof a 14.5 mm. larva. phores, and this particularly for the reason that a parietal organ as such does not exist. There seems to be in all a small epiphysis (Studniéka ’05, and Warren ’05). Nevertheless, although there is no parietal organ and only a small epiphysis, there was still the chance, considering the results obtained by von Frisch (11), that in this region of the brain there might be a ‘center’ the stimulation of which would have an inhibitory influence on the melanophores. 246 HENRY LAURENS The first of the methods of experimentation and the one which seemed at first thought to be the most promising, but which failed for reasons which will be given later, was to remove that portion of the undifferentiated nervous system from which the epiphysis arises. This was accordingly carried out on larvae measuring between 5.5 and 6.1 mm. The operation is a simple one requiring but slight practice. The embryo to be operated on was placed in 0.02 per cent NaCl in a watch glass, the bottom Fig.5 A sagittal section of a portion of the head of a 34.0 mm. larva to show the position and relative development of the epiphysis and paraphysis. of which was covered with paraffin. In this a little pocket was dug near the middle of the dish and into this the embryo was slipped so that the anterior end was directly upwards. Under the binocular microscope, with a small pair of iridectomy scissors, the epidermis and then the desired portion of the un- differentiated nervous system was removed. The extent of the portion removed is indicated approximately in figure 1 between the dotted lines x—x. This operation was performed on 10 embryos between 5.5 and 6.1 mm. long. On 10 others in addi- REACTIONS OF MELANOPHORES OF AMBLYSTOMA 247 tion the optic vesicles were also removed by the method de- scribed in former papers (Laurens ’14, p. 196 and 715, p. 579). A complete series of experiments were carried out on these 20 larvae together with control normal seeing and eyeless larvae, Fig. 6 The anterior end of a 42.0 mm. larva to show the appearance of the paraphysis (with the surrounding blood vessels) as seen from the outside. for the purpose of seeing whether there was any difference in their reactions to darkness, light and backgrounds and in the time when these reactions first made their appearance (Laur- ens 715, pp. 583-595). The results were in complete agreement with my earlier ones. But unfortunately upon sectioning the 248 HENRY LAURENS larvae and examining the brains it was found that in nearly every case either complete or partial regeneration of the parts concerned had taken place. In some cases only a small epiphysis was present, while in others no difference between this region in the operated larvae and in the normal larvae could be detected. In only 2 larvae sectioned was there no sign of an epiphysis. Therefore another method of getting rid of the epiphysis, etc. had to be resorted to. This was to remove it from older larvae after differentiation of the nervous system had taken place. This operation was suc- cessfully carried out on 28 normal larvae ranging in length from 12.5 to 40 mm., and in 8 eyeless larvae ranging in length from 12.0 to 30.0 mm., from which the optic vesicles had been removed when they were about 5.5 mm. long. In addition to removing the epiphysis the roof of the diencephalon was also cut out in 3 normal and 3 eyeless larvae when they were about 12.5 mm. long. After the experiments to test the various reactions of these larvae to darkness, light and background had been finished sev- eral of them were killed, sectioned and the brain examined for indications of regeneration of the parts involved. Some of the others were allowed to live until they had metamorphosed, others not quite as long. Eventually they were all of them sectioned and studied. In only one was there any sign of an epiphysis. The carrying out of the operation necessary to remove the epiphysis is not a particularly difficult one. The larvae were anaesthetised by placing them in 0.02 per cent chloretone. With a pair of iridectomy scissors a flap of skin was cut and left at- tached at one end so that it could be folded back, after which the roof of the brain case was removed in small pieces until the desired portions of the brain were sufficiently exposed to be removed. Out of a total of 43 operations 28 normal and 8 eye- less larvae survived and grew. Again the experiments carried out on larvae operated on in this way showed no differences between those which have al- ready been reported in such detail for normal and eyeless larvae. In addition to simply placing the larvae in darkness and in light on various backgrounds there was carried out a series of experi- REACTIONS OF MELANOPHORES OF AMBLYSTOMA 249 ments in which a narrow beam of light, reflected from a Nernst glower by a mirror and concentrated with a lens, was thrown on the region of the brain from which the epiphysis arises. The results of these experiments give additional evidence concern- ing the ability of the melanophores to respond to direct stimula- tion by light. When Amblystoma larvae are placed in darkness and observed by means of a faint red light, they remain motionless for the greater part of the time if there are no other larvae present or other animals which may serve as possible food. It is therefore a simple matter to throw such a beam of light upon any por- tion of the dorsal or lateral surface of the body and keep it there. Naturally, at times the larvae move by alternately crawling and swimming with the snout close to the bottom of the vessel in the characteristic nosing fashion, nevertheless, the animals can usually be followed with the narrow beam of light until they again come to rest. It was found that in no case when the beam of ight was thrown upon the region of the epiphysis of normal seeing or eyeless larvae did an expansion of the melanophores over the whole body take place. On the other hand the result of such local illumina- tion at any place on the body results in an expansion of the melanophores stimulated. These experiments are rather tedious because of the time consumed in carrying them out. In table 2 (Laurens, 715, p. 585) it will be seen that it takes from 1} to 2 hours for the melanophores of normal larvae with eyes to ex- pand when the larvae are placed in the light, and 2 to 3 hours for the same thing to happen in eyeless larvae. In these experi- ments with local illumination no expansion was observed tak- ing place in less than 2 hours and in most cases an illumination . lasting 3 hours or more was necessary. These experiments on local illumination were carried out on both normal and eyeless larvae with and without the epiphysis and the roof of the diencephalon. In some of them not only was the beam of light thrown on the region of the head under which the epiphysis is, but the brain was exposed and directly illumi- nated. In addition it was also carried out on larvae in which 250 HENRY LAURENS the central nervous system had been completely destroyed by boring it out. In all, the results were the same, expansion of the melanophores illuminated by the beam of light, and no ef- fect on the pigment cells of the remainder of the body. These results show conclusively that the expansion of the melan- ophores of the larvae of A. punctatum caused by light are due primarily to the direct stimulation of the pigment cells them- selves and not to the inhibitory action of the nervous system. The nervous system certainly helps to bring about the expan- sion, Just as it helps to bring about the contraction of the me- lanophores when the larvae are placed in darkness, for, when the reaction times of normal and eyeless larvae are compared it is seen that those of the former are shorter (Laurens, p. 585, table 2). The experiments on larvae in which the central nerv- ous system was cut through, and partially or totally destroyed (pp. 614-616) could be considered sufficient proof against the assumption of the action of an inhibiting center in Amblystoma, such as Fuchs suggests. But there seemed sufficient reason for carrying out experiments to test just this point. These have afforded us additional evidence concerning several things mentioned in my former paper and have added information not there contained. One interesting fact which was brought out and which seems worthy of remark is that although the roof of the diencephalon is cut out with the epiphysis, nevertheless the secondary reactions of the melanophores, and the reactions to background, which are certainly dependent on the presence of the eyes, still come about. As we know, the median por- tion of the diencephalon has no nervous elements in it, these being in the lateral walls. In a few experiments which were carried out on larvae after cutting the lateral walls, the second- ary reactions of the pigment cells, as well as the reactions to black and white backgrounds did not take place. DISCUSSION AND CONCLUSIONS Fuchs’ idea that the parietal organ, or, as we may how say, the epiphysis, and the surrounding region of the roof of the brain has an inhibitory influence on the melanophores does not hold REACTIONS OF MELANOPHORES OF AMBLYSTOMA 2hit for the larvae of Amblystoma punctatum. It has already been pointed out that Babak’s explanation of the reactions of the melanophores of the Axolotl larvae also cannot be applied to those of the melanophores of A. punctatum. When one considers the relative anatomical insignificance of the epiphysis in the Urodeles (Studniéka, ’05 and Warren, ’05) it is hardly surprising that this organ should be found to have no influence upon the melanophores. In the teleosts and particu- larly in the reptiles, the epiphysis reaches a high degree of development with a distinct parietal organ, so that in these animals it is possible that it may have an important function in connection with the reactions of the melanophores (Fuchs, pp. 1442 and 1651, and von Frisch, ’11, p. 374). Nevertheless, it cannot be assumed from the results of von Frisch’s experiments, —which are the only ones that have been carried out previously to specifically test this point—that even in the minnow Phoxi- nus, the parietal organ is responsible for the inhibiting effect produced by stimulation with light. For he found even after the parietal organ had been extirpated (controlled by micro- scopic sections) that stimulation with light of this region still produced an expansion of the melanophores, while shading it caused the melanophores to contract. Von Frisch further found that when the portion of the roof of the brain which extends from the point of junction of the epiphysis with the brain to the posterior commissure is also removed, that illumination may still sometimes result in an expansion of the melanophores, although sometimes there is no change at all. From these results von Frisch is finally forced to conclude that in the re- gion of the diencephalon there must be light perceiving cells which function as an inhibiting center, and from which nerve fibers run to the deeper portions of the brain which are in this way connected with the pigment motor apparatus so as to bring about this reaction of the melanophores. Perhaps, he thinks, these cells are particularly numerous in the parietal organ, perhaps they are identical with the sense cells and per- _ haps the connecting nerve fibers constitute the ‘Tractus pinealis.’ But they are not limited to the parietal organ, else, when it was THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 252 HENRY LAURENS removed, the reactions of the melanophores would cease to take place. As an interesting fact von Frisch found this region of the roof of the diencephalon to be also particularly easily stimulated by other means than light, e.g., electric currents (pp. 335, 336, and 376) in this particular fish Phoxinus. Von Frisch (p. 377) is led to make the suggestion, later taken up by Fuchs, that perhaps in other animals the parietal organ may be found to have an influence on the pigment cells and that thereby an explanation may be found for some of the perplexing diverse results, e.g., in tadpoles, where darkness causes the melanophores to contract and light causes them to expand, while the opposite reactions take place in the adult frog, and in Axolotl larvae, citing Babak’s results on normal and blinded individuals. Also for the reactions of the melanophores of Salamandra maculosa larvae and of Triton cristatus adults which, however, show an expanded condition of the melanophores in light and in both normal and blinded individuals. However, he has himself (p. 378) carried out experiments with Salamandra larvae and adult Tritons with different results than he hoped for. For in these animals illumination of the head region did not give as clear results as in Phoxinus. Further- more he received the impression that local illumination of any portion of the body results in an expansion of the melanophores of the whole body. In other words the illumination of the portion of the body under which the epiphysis is has no different effect from illumination of any portion of the body. Experiments such as von Frisch suggests should be carried out on tadpoles and frogs, in some of which the parietal or- gan has been destroyed, in others of which the influence of the eyes has been removed, etc. In these animals the pineal organ attains a comparatively high degree of development (Studnicka, p. 110 ff.) Furthermore these results should be compared with those obtained with such an animal as Diemyctylus, where the reactions of the larval melanophores are also opposite to those of the adult, but where there is no pineal organ and where the epiphysis is not highly developed. The results of such experi- . REACTIONS OF MELANOPHORES OF AMBLYSTOMA 253 ments would certainly afford a great deal of evidence con- cerning the influence of the parietal organ on the melanophores. Such experiments should also be carried out on reptiles, where the parietal organ reaches a high degree of development, and in some animal where the color changes of the skin are well marked. The nervous system must be admitted to exert a very impor- tant effect upon the chromatophores of any animal, and this effect is for the most part conditioned upon the eyes, which have been shown to play such an important part in the color changes of many animals (see, for example, the reactions which have been described for the melanophores of Amblystoma, or the effects observed by von Frisch, 711 and ’12, of blinding fish). Take away the eyes and the major part of the controlling influ- ence of the nervous system over the pigment cells of the skin is lost. This of course does not apply in full to those cases where the eyes have, or are believed to have, lost control over the chro- matophores and where other things than light, such as tempera- ture, tactile stimuli, etc., are believed to be of greater importance. But even in such a case, as Biedermann has shown, the re- actions of the melanophores to light are controlled by the nerv- ous system in that this exerts a tonic influence upon the pig- ment cell. Biedermann demonstrated that the pigment cells of the frog as soon as they are released from the influence of the nervous system have the tendency to expand. When he cut through the connection between the thalamus and the mid- brain in Rana esculenta and R. fusea, as well as in Hyla, he found that the melanophores expanded, and that when the frogs (par- ticularly Hyla) operated on in this way are kept for weeks in _ darkness as well as in diffuse light no change takes place in the melanophores. Briefly, that light has under these conditions little, if any, effect on the melanophores, unless direct sunlight is used. It is my belief that the general function of the chromatophores is to expand when illuminated and to contract again when in darkness. But when the melanophores are under the influence 254 HENRY LAURENS of the nervous system, which they normally are, it may be, ow- ing to the nature of things, environmental conditions (back- ground) or psychic factors, ete., that the sense of the reaction to a light stimulus is reversed and contraction is brought about. What causes the melanophores to contract in darkness is another matter, it happens whatever its cause may be. From the conflicting evidence that we have we are forced to specu- late regarding its cause. It may be due to chemical substances, such as Fuchs suggests, which arise as a result of the ordinary processes of life (inner secretions, or what not). It may be due to chemical changes in the skin, brought about by darkness, (or the absence of light). It may be due, on the other hand, in some cases at least, to what may be called for want of a better term, a simple relaxation, for the reason that when pieces of the skin are removed from some animals (the frog, Biedermann; Anolis, .Carlton) the melanophores contract. (Laurens, 715, p. 599). But that this contraction may also be due to chemi- cal changes is perfectly possible. The expanded condition of the melanophores in darkness which comes about in animals in which, due to the control of the nerv- ous system, the pigment cells contract in light, is probably not a specific reaction to darkness but one which is due to the lack of the nervous impulse which light sets up. The pigment cells have become subordinated to the influence of the nervous sys- tem, and when the condition of excitation ceases (in darkness) the melanophores expand. It is interesting, in this connection, that Hargitt could observe, in tree frogs, no definite effect of darkness (and of low temperature) on the pigment cells. The contracted condition of the pigment cell is claimed by many to be its active state. Arguments are put forward to sup- port this view from the fact that muscle contracting stimuli also cause the chromatophores to contract. But light (and — under certain conditions high temperature) is the adequate stim- ulus for the chromatophores of animals and light causes the melanophores in many animals to expand, both when under the control of the nervous system and when released from this control, perhaps in all when the influence of the nervous sys- REACTIONS OF MELANOPHORES OF AMBLYSTOMA 255 tem is removed. The fact that Hertel (07) obtained contrac- tion of the chromatophores when he locally stimulated the skin of Triton larvae with ultra violet rays is another matter, since these rays probably always cause the pigment cells to con- tract (Laurens, ’15, p. 599). That he also obtained contraction when he locally stimulated with yellow and red light, may be due to one or more of several things. First, that the intensity of the light was high; second, since the light was focussed, that the pigment cells were subjected to a high temperature; and third, that the contraction of the pigment cells stimulated was a reflex action. This last possibility has less in its favor than the others, since there was no indication of spreading. Ballo- witz’s (14) results of contraction are probably due to the effects of the high intensity of the light that he used, about 1000 candles, hardly to a heat effect (see p. 200). Moreover the melanophores that he experimented with were taken not from the skin but from the “Hirnhaut.’’ Finally the results of Hooker (12) are also, it seems to me, capable of explanation.!. Hooker found, when the melanophores of the frog (R. fusca) were completely deprived of their nervous connections, either in the body or when bits of skin were placed in hanging drop cultures, that for a day the reactions of the melanophores were the same as when they were under normal conditions. But after this period of time had elapsed the reactions of the pigment cells to light and darkness were reversed. Hooker offers no explanation of this curious fact. An explanation for the primary reactions, those lasting for the first day, may be found in the fact that sunlight was used as the source of light. It is therefore highly probable that the contraction of the melanophores was a heat effect and not due to the light at all. When the melanophores are placed in dark- 1 Opportunity is taken to call attention to the fact that Hooker is misquoted on p. 598 of my former paper (’15) though rightly so on p. 628. Also that Ballo- witz is incorrectly quoted on p. 598. The statement concerning his results should read that when the melanophores are removed from the body they contract, but when placed in salt solution they partially expand. When they are now illuminated they contract. The opportunity is also taken to point out a mistake in table 4, p. 589 where under ‘black background’ a II. reaction is indicated as follows: ‘‘contraction (3-4 hrs.)’’ This should be struck out. 256 HENRY LAURENS ness the pigment expands again, not because of a specific stimu- lating effect of darkness, but simply because the heat stimulus ceases to act. Why, after a day, these isolated melanophores show exactly the reverse reactions to a heat stimulus that they did before is not entirely clear to me. That it may be due to an increase in the acidity of the cells, both of the pigment and surrounding tissue cells, in other words to a change in the H ion content, 1s not absolutely impossible. Steinach’s results, judging from the results of Biedermann’s work which show the dependence of the chromatophores upon the central nervous system, were probably also obtained by illuminating the animals with sunlight. The effects therefore are to be referred to the action of heat and not to light. The chief function of the chromatophores is probably to enable the animals possessing them to adapt themselves to the intensity and color of their background. To make this possible the pig- ment cells must be under the control of the eyes which regulate through the nervous system, the movements of the pigment suspended in their cytoplasm. It is very possible that in some animals the eyes may have no (or very little) influence over the chromatophores as Biedermann has pointed out to be the case in the frog where according to him, light plays a small part in the color changes. In some animals conditions are of course different from those in others. Von Frisch (’11 and ’12) has shown this to be the case in fishes. The minnow Phoxinus does not possess in any marked degree the power of adaptation to background, but the melanophores respond more to the relative intensity of the light. In Salmo the response of the melano- phores to different backgrounds is very marked. In Crenila- brus pavo this is also the case. But the control of the nervous system is in each of a different nature. In Phoxinus when the eyes are removed the melanophores expand in the light and contract in the dark, which is opposite to the conditions seen in the normal fish. In Salmo, when blinded, the melanophores:are under all conditions expanded, and in Crenilabrus the melano- REACTIONS OF MELANOPHORES OF AMBLYSTOMA ZOU phores expand in the light and contract in the dark just as they do in the normal fish, only more extensively. The reactions of Amblystoma larvae to different backgrounds show that this is an important function of the melanophores. Amblystoma larvae are nearly always found in pools in the woods in which the bottom is covered with leaves which are either dark brown or black. Occasionally they may be found in ditches in a field, but here the bottom is of black mud. Over such a background, as my experiments have shown, the larvae are very dark, almost black, due to the complete expansion of the melanophores. But over a white background the larvae are pale, exhibiting, to be sure, not such a marked adaptation, but at any rate the melanophores are contracted. Also over an indifferent background, when the illumination is constant and bright, the melanophores contract (4% to + expansion), so that the larvae are more nearly of the color of the bottom. The fact that the time required for the changes in the melano- phores to come about is long and that therefore the usefulness to the individual in the way of protection, of the ability of the melanophores to change according to the background, is of doubtful value, is an argument that can be answered by the statement that probably such conditions in change of background are not very likely to occur in nature. But the fact that if they do occur the melanophores can react to the change shows that the ability to change is in the nature of an adaptive one. I| have carried out experiments in a large aquarium, directly in front of a window in my room, to further test this matter. In one series of experiments the bottom of one half of the aquarium was of black mud or of black leaves, the other half of whitish sand, the larvae being free to swim from one to the other. Under these conditions the larvae are of course not at all adapted to the background for they do not remain in any one portion of the ‘aquarium long enough for the melanophores to react adaptively. But if the aquarium is divided by a glass partition placed at the dividing line between the two different colored bottoms then the adaptive changes of the melanophores which have been described, do come about and the coloration of the larvae is strikingly 258 HENRY LAURENS similar to the bottom, particularly in the case of the black con- dition, so that they are very difficult to see. If the partition is now again removed so that the larvae can swim through the whole aquarium freely then the difference in coloration between the larvae that have been over the one or the other background is very apparent. 2 The theory that the melanophores in an aquatic animal can have any function in regulating the temperature has been shown by Bauer (714) to be highly improbable. Also it does not seem possible to suppose that the contraction or expansion of the pig- ment cells in the skin has anything to do with the reception of a light stimulus by sensory nerve endings. It is well known that phototactic reactions can be induced in animals, which have been deprived of their eyes, by illuminating the skin. The amphib- ians show this faculty very well (Parker ’03, Laurens ’11). But as Parker (09) has shown only a very few fish show this ability on the part of the integumentary nerves stimulated by light to result in phototactic responses. These reactions are supposed to be started by the action of light on receptors in the skin, either directly as light energy (heat) or indirectly as chemi- cal energy set free by the photochemical changes started by the action of the light on the skin, which stimulate the sensory nerve endings. The fact that in most fishes no such reactions take place, although the pigment motor function is well developed, argues against the pigment cells having any such universal func- tion in aiding or retarding the rapidity of perception. Experi- ments were carried out by me (’14) to test just this point as to whether the condition of the pigment cell—contracted or ex- panded—had anything to do with the sensitiveness of eyeless Amblystoma larvae to light with negative results, and the con- clusion was reached that the.condition of the pigment in the skin melanophores had ‘nothing to do with the sensitiveness of the larvae to light, although it was dependent upon the fact as to whether they had been kept in light or in darkness before their phototactic reactions were tested. To conclude then this brief and fragmentary discussion it seems to me that the reactions of the pigment cells in the skin REACTIONS OF MELANOPHORES OF AMBLYSTOMA 259 to light are in most cases adaptive. The primary reaction of the melanophore is to expand when illuminated which it always does when directly stimulated with light of ordinary intensity and sometimes when the stimulus is indirect and through the eyes. Fuchs (p. 1651) closes his review of the color changes of the reptiles with the statement that in those animals which do not show expansion of the pigment in the light, either the parietal organ has lost its function in the course of phylogeny or ontogeny, or the eyes have gained a regulatory influence over the light reactions which results in changing the original light reaction (expansion) into the directly opposite reaction (contraction). I should change this statement and applying it to the chromato- phores of all animals say, that in those animals in which the pigment cells do not expand in light of ordinary intensity and under normal conditions of temperature but contract, this is due to the controlling regulatory influence of the eyes, by means of which the reactions of the melanophores are made adaptive. In other words, when the pigment cells contract in the light, this reaction is one that is bound up with the nervous system in that an impulse started in the retinae is sent out which is opposite to that produced by direct stimulation. Such is the case in frogs, in Diemyctylus, and in Phoxinus, Salmo, ete. Usually, however, the melanophores expand in the light, due in most part, to their direct stimulation, although the eyes still show an influence in that in many cases the expansion is not maximal as it is when the eyes are removed, and that when the eyes are present the reactions come about more quickly. SUMMARY 1. The epiphysis of Amblystoma punctatum larvae has no influence on the reactions of the melanophores to light and darkness. There is no parietal organ. 2. The view is expressed that the reaction to light of ordinary intensity of the pigment cells in the skin of animals is to expand. When this does not take place it is due to the controlling regula- tory influence of the eyes. 260 HENRY LAURENS 3. The reactions of chromatophores are believed to be adap- tive in that they either respond to the relative intensity of the light or to the color and intensity of the background. 4. The control over the melanophores which the eyes possess is of course most important, for by this means the reactions of the melanophores are able to be adaptive. 5. Indirect stimulation of the melanophores through the eyes is not by any means always opposite in effect to direct stimu- lation, but only so when the conditions are such that it is neces- sary that the chromatophores contract in the light in order that the reaction be adaptive. LITERATURE CITED Bapak, E. 1910 Zur chromatischen Hautfunktion der Amphibien. Pfliiger’s Archiv, Bd. 131, S. 87-118. BattowiTz, E. 1914 Uber die Pigmentstrémung in den Farbstoffzellen und die Kanalchenstruktur des Chromatophoren-Protoplasmas. Pfliiger’s Archiv, Bd. 157, S. 165-210. Baugr, VY. 1914. Zur Hypothese der physikalischen Wirmereguherung durch Chromatophoren. Zeit. f. allg. Physiol., Bd. 16, S. 191-212. BreDERMANN, W. 1892 Uber den Farbenwechsel der Frésche. Pfliiger’s Archiv, Bd. 51, S. 455-508. Caruton, F. C. 1903 The color changes in the skin of the so-called Florida chameleon Anolis carolinensis Cuv. Proc. Amer. Acad. Arts and Sciences, vol. 39, pp. 257-276. von Friscu, K. 1911 Beitrige zur Physiologie der Pigmentzellen in der Fischhaut. Pfliiger’s Archiv, Bd. 138, 8S. 319-387. 1912 Uber farbige Anpassung bei Fischen. Zoolog. Jahrb., Abt. f. allg. Zool. u. Physiol., Bd. 32, S. 171-230. Fucus, R. F. 1914 Der Farbenwechsel und die chromatische Hautfunktion der Tiere. Winterstein’s Handb. d. vergl. Physiol., Bd. 3, 1. Halfte, 2. Teil, S. 1189-1657. Harairr, C. W. 1912 Behavior and color changes of tree frogs. Jour. Animal Behav., vol. 2, pp. 51-78. Herter, E. 1907 Einiges ititber die Deutung des Pigmentes fiir die physio- logische Wirkung der Lichtstrahlen. Zeit. f. allg. Physiol., Bd. 6, S. 44-70. Hooker, D. 1912 The reactions of the melanophores in Rana fusca. Zeit. f. allg. Physiol., Bd. 14, 8. 98-104. REACTIONS OF MELANOPHORES OF AMBLYSTOMA 261 LAuRENS, H. 1911 The reactions of amphibians to monochromatic lights of equal intensity. Bull. Mus. Comp. Zool., vol. 43, pp. .253-302. 1914 The reactions of normal and eyeless amphibian larvae to light. Jour. Exp. Zool., vol. 16, pp. 195-210. 1915 The reactions of the melanophores of Amblystoma larvae. Jour. Exp. Zool., vol. 18, pp. 577-638. ParkKeER, G. H. 1903 The skin and the eyes as receptive organs in the reactions of frogs to hght. Amer. Jour. Physiol., vol. 10, pp. 28-36. 1909 The integumentary nerves of fishes as photoreceptors and their significance for the origin of the vertebrate eyes. Amer. Jour. Phys- iol., vol. 25, pp. 77-80. Srernacu, E. 1891 Uber Farbenwechsel bei niederen Wirbeltieren bedingt durch direkte Wirkung des Lichtes auf die Pigmentzellen. Centralbl. f. Physiol., Bd. 5, 8. 326-330. StupnicKa, F. K. 1905 Die Parietalorgane. Oppel’s Lehrb. d. vergl. Anat. d. Wirbeltiere, 5 Teil, vi. + 2548. WaRREN, J. 1905 The development of the paraphysis and the pineal region in Necturus maculatus. Amer. Jour. Anat., vol. 5, pp. 1-27. an Le vi m 5 . Mj - , j 7 May ; ¥ ty wheats hae “ie We rwelerae a o ry ii Lf ia ee na EWG be! : hoy ae My: wt ae ot 1 RS aT Pe : ; Vikan Aoki { . Poh ai Badia ; iy | Ae | mt as 7 in by tee. ¥ a Wynne) y] , ; i 2 ws THE CONTROL OF SEX BY FOOD IN FIVE SPECIES OF ROTIFERS DAVID DAY WHITNEY Biological Laboratory, Wesleyan University, Middletown, Conn. SIX FIGURES It has been shown in the American and English rotifer, Hy- datina senta, that food conditions are the controlling factors in regulating the parthenogenetic production of the two sexes. When the parthenogenetic females were fed upon a diet of the colorless flagellate, Polytoma, they produced female-producing daughters exclusively even through a period of twenty-two months and extending through many scores of generations. However, when the parthenogenetic females were suddenly transferred from the Polytoma diet to a diet of the green flagellate, Chlamydo- monas pulvisculus, they produced in many instances as high as 80 per cent, or higher, of male-producing daughters within a few hours. In a few selected experiments the percentage of male- producing daughters reached 100 per cent when the diet of color- less Polytoma was suddenly changed to a diet of the green Chlamydomonas. . If the production of the sexes can be regulated in this rotifer by the diet it is of considerable interest to know whether the diet can regulate the production of the sexes in other species of rotifers. Furthermore, it is quite important to determine whether it was the stimulus produced by the change of food from the Polytoma to the Chlamydomonas diet that caused the male- producing daughters to be suddenly produced or whether it was a sufficient quantity of more easily assimilated food that changed the mechanism of the daughters from female to male- producers, or perhaps some other factor. 263 264 DAVID ‘DAY WHITNEY The colorless Polytoma was reared in stable tea (horse manure) solution in a subdued light while the green Chlamydomonas was reared in bouillon solution in direct sunlight. It is very probable that their values as foods would be considerably dif- ferent, In as much as Polytoma lacks chlorophyl and con- sequently cannot carry on photosynthesis while Chlamydomonas possesses chlorophyl and manufactures and stores starch in its cell. Thus the colorless flagellate would contain neither starch nor sugars while the green flagellate might have more or less of each. In order to get new data on these two problems the former experiments upon Hydatina senta were carefully reviewed and new experiments carried out upon the following rotifers: from the order of Ploima the species, Branchionus pala, Diaschiza sterea, Diglena catellina, and from the order of Scirtopoda the species Pedalion mirum. These species were identified by Harry K. Harring, Custodian of the Rotatoria in the United States National Museum, to whom I am greatly indebted for the favor. Hydatina senta The results obtained from numerous experiments carried out upon this species from New Jersey and from England have already been published in detail and need not be repeated again. However, as no diagrams have been published of the American form, diagrams have been drawn to the same scale of a female and a male individual, a parthenogenetic female-producing egg, a parthenogenetic male-producing egg, and of a fertilized egg. These are shown in figure 1. A general review of the results of the former experiments have been put into a plotting in dia- gram 1, showing that a continuous diet of Polytoma caused only female-producing females to be produced for nearly two years but when the food was suddenly changed to the green Chlamy- domonas a high percentage of male-producing females appeared ' within a few days. An important feature of these experiments, which does not show in the diagram, is the place where the stimulus is effective in changing the females from female-pro- CONTROL OF SEX BY FOOD IN ROTIFERS 265 ducers to male-producers. The mother of the female-producing or male-producing daughters is the individual influenced by the diet. If the mother is fed the colorless diet all of her daughters C ya; Fig. 1 Hydatina senta. A, female; B, male; C, parthenogenetic egg which develops into a female; D, smaller parthogenetic egg which develops into a male; E, fertilized egg, which is often called resting or winter egg, and always develops into afemale. This egg if unfertilized would have formed a small parthenoge- netie egg which would have developed into a male. will produce females but if the diet of the mother is changed from the colorless flagellates to the green flagellates she will begin soon to produce daughters which will produce males. i) o> (op) DAVID DAY WHITNEY Brachionus pala In August of 1908 a small pond was found at Cold Spring Harbor, Long Island, New York that contained great numbers of this species of rotifer. The pond was located in a small sunny pig pasture in which there were kept a few pigs. These animals lay in the pond a large part of the time during the hot days and consequently the water became somewhat foul and furnished an ideal culture medium for the rotifers and the va- PBRCENTAGE OF MALE-PRODUCING FEMALES LT a Ht TF | ft ime Tia i /128 y | ogi ae Diagram 1 Hydatina senta. Showing that a continuous diet of Polytoma through a long period of time yielded only female-producing females but when the diet was suddenly changed to Chlamydomonas male-producing females appeared at once. P indicates a Polytoma diet, C indicates a Chlamydomonas diet. rious micro-organisms on which they fed. No experiments were carried out at that time but some mud from the bottom of the pond was taken, dried, put into a paper bag, and stored in an ordinary laboratory room. In February of 1915 some of this dried mud was put into a solution of 10 ce. of bouillon and 140 cc. of tap water and placed near a window. Within a few days various protozoa were nu- merous in the jar and a few females of the rotifer, Brachionus pala, were found. These females undoubtedly had hatched CONTROL OF SEX BY FOOD IN ROTIFERS 267 from the winter eggs which are the fertilized eggs. Such eggs of Hydatina senta have been found to live six years in old culture water and three years in a dried state. When dried they have withstood for several days the extreme low temperature of liquid air which is about —191°C. and also they have withstood as high a temperature as +110°C. for a few hours. Moreover, it has been found that species of rotifers that produce fertilized eggs cannot themselves be dried and later be revived by plac- ing in water as is true of those rotifers which do not produce fertilized eggs. These females of Brachionus pala continued to live and to reproduce rather slowly in the culture jar of bouillon and water. From this stock jar females were taken and many preliminary experiments were carried out in attempts to find optimum food conditions. Finally it was found that when bouillon cultures, inoculated with the miscellaneous green flagellates that developed from the same mud from which the rotifers developed, were placed in direct sunlight they developed very fine food cultures for the rotifers. In order to prevent the temperature of the culture jars from rising too high in the direct sunlight and there- by killing all the flagellates as well as the rotifers that might be in them, the jars were placed in a large pan through which tap water flowed. In this way the temperature could be regu- lated at will and was not allowed to rise above 35°C. to 37°C. The green flagellates seem to be the most active at this temper- ature and as the rotifers could eat them only in an active state all the cultures were usually kept in this pan. At night and on cloudy days the temperature would be the same as the room temperature of about 20°C. The stock solution of bouillon was made by boiling one Ar- mour’s beef bouillon cube in 400 ce. of tap water. The bouillon culture of green flagellates was allowed usually to stand from several days to several weeks in direct sunlight with occasional additions of fresh bouillon. During this time several inoculations with the rotifers were made and usually at the end of a few days there would be a good culture of both the green flagellates and the rotifers. After some time if no THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 268 DAVID DAY WHITNEY fresh bouillon was added the whole culture of the flagellates and rotifers became balanced, that is the flagellates reproduced at such a rate as to keep the rotifers moderately supplied with food. When they were thus supplied moderately with food only fe- males were found in the culture jar but when fresh bouillon was added to the general culture it furnished additional food for the flagellates causing them to multiply very rapidly and to produce enormous numbers of themselves in the jar. With this great increase of food the rotifers also began to eat raven- ously and soon produced a great number of females which to a large proportion produced male offspring. In this manner of increasing the food by adding fresh bouillon males could be produced in enormous numbers within a short time. These males copulated with many of the young females and within a few days the majority of the adult females were carrying fer- tilized eggs. Soon the cultures were depleted of food by the great increase of rotifers and then nearly all the rotifers died of starvation but left thousands of fertilized resting eggs. It has been shown by Lauterborn that if the male egg is fer- tilized in Asplanchna it produces the winter or resting egg. The same fact has been shown by Whitney and Shull to be true of Hydatina senta and recently it has been shown by Moro to be true also of Brachionus pala. These experiments were done in mass cultures many of which contained many thousands of individuals and countings were made of the female-producing and male-producing females at various intervals. These two kinds of females were easily dis- tinguished from each other because after they had extruded the eggs they carried around the eggs attached to their bodies. The male-producing females carried small eggs and the female- producing females carried large eggs. The details of the experi- ments are in table 1. Drawings of the females carrying the different eggs and also of a male are in figure 2. In diagram 2 a plotting is seen of experiment 1, and in diagram 3 a plotting is seen of a part of experiment 3. CONTROL OF SEX BY FOOD IN ROTIFERS 269 Fig. 2 Brachionus pala. A, female with attached parthenogenetic female eggs; B, female with attached parthenogenetic male eggs; C, female with at- tached fertilized eggs; D, male. 270 DAVID DAY WHITNEY TABLE 1 Experiments with Brachionus pala and a mixed diet of miscellaneous green flagel- lates showing that when the flagellates were caused to be very abundant by the addition of fresh culture media the percentage of the male-producing female roti- fers was very high and the percentage of the female-producing female rotifers was very low but when no fresh culture media was added and the flagellates occurred only in moderate numbers the percentage of male-producing females was very low and the percentage of the female-producing females was very high. 1915 CULTURE WATER 29 ove % SP { 15 ce. bouillon April 6 ie ec. tap water mixed green protozoa April 14 15 ce. bouillon added..... April FAM Mee PR tego sie tlc. co eae: Res ee 20 0 0 April 18 15 ce. bouillon added..... 60 0 0 April I). "1h a at ea area a ae Fs Rict 18 2 10 April 20 15 ec. bouillon added..... 28 12 30 |e Aiprall PAW | MEERA Presi Arse cate Ae 70 30 30 + | April DON AN «avs, s¢e 2 oh eee 95 5 5 S | April et < cb Lue ean ane 95 5 5 £ | April OR vl EL ec Ee ieee ie 200 1 0.5 5 | May 1 15 ce. bouillon added..... 200 1 0.5 SI May 3 15 ec. bouillon added..... May FC. OM Pelee, coe tec eee RSE ate eee APN en 65 35 35 May i Sia Wa ees AC sek eae ee ba Pat k8 97 3 3 May 19 15 ce. bouillon added..... 100 0 0 May 21 15 ee. bouillon added..... May DOB Riecet cee en hire tne eye: 70 30 30 May DA Pal Ae Hk em ar gaacig eEy Dore Sa OF Au 100 100 50 June 1 A fi OR Eee ee 100 0 0 AROualee eect 1218 224 15 ec. bouillon March 8 135 cc. tap water mixed green protozoa April 19 15 ec. bouillon added..... 200 0 0 NG) Aral 24 15 ce. bouillon added a | April DOC ssc. eee tee eee 200 200 50 = | April S08 i os ae ae 200 2 1 ‘—& | May 3 15 cc. bouillon added..... & | May et Renee EY Lae Maen 80 20 20 aa May 6 15 ec. bouillon added..... May A NB See Sos Get 83) hc PORE ER oe 70 30 30 May ENGI Le fe aaa ep Sekcah eee Pe a 99 1 il Totals voi con es 849 253 CONTROL OF SEX BY FOOD IN ROTIFERS 271 TABLE 1—Continued 1915 CULTURE WATER [ome) fof) % SQ 15 ec. bouillon April 19 % cc. tap water mixed green protozoa o | April DD, 15 ec. bouillon added..... = | April 24 15 ec. bouillon added..... 5 April 26 15 ec. bouillon added..... - May VT? lj ketch Ror Ve RO hey ae Oe Rar See 95 5 ) & | May 28 15 ce. bouillon added.... . = | June 1 15 ec. bouillon added..... June PHS il one tare pete cpg e se Mee scar hehehe 5 95 95 June Oy Partegere ciate: cat ueet tue scrat ah a: 50 0 0 Mo Gall Peer tower ae : 150 100 | 20 ce. bouillon | | April 19 130 cc. tap water “Fy mixed green protozoa a || A\jovell 22 20 cc. bouillon added..... § | April 24 20 cc. bouillon added.... 5 April 26 20 cc. bouillon added..... 2 June 1 20 cc. bouillon added..... 100 0 0 SI June 3 20 ec. bouillon added..... June Weal) Ans eh Ree evamn tel gis > Siete aia 50 50 50 BROtallA es Pein ae sce 150 50 Experiment 5 ( 20 ce. bouillon April 19 es cc. tap water mixed green protozoa April 22 20 ce. bouillon added..... April 24 20 ec. bouillon added..... April 26 20 ce. bouillon added..... April 28 20 ce. bouillon added..... June 1 20 ce. bouillon added..... August PROVE Nh Ric eh Aan 2, oe 20, 9A Tk RS 200 0 0 August 21 10 ec. bouillon added August Tn || Paes Rete Pile NR casa oieaderahos 100 100 50 TUOUEIBS ¥ obo> Beene 300 100 Be DAVID DAY WHITNEY PERUENTAGE OF MALE-PRODUCING FEMALES Diagram 2 Brachionus pala. Experiment 1 of table 1. Showing the pro- duction of a high percentage of male-producing females when the food conditions were changed by the additions of bouillon. B indicates the addition of bouillon. PERCENTAGE OF MALE-PRODUCING FEMALES Diagram 3 Brachionus pala. Experiment 3 of table 1. Showing the pro- duction of a high percentage of male-producing females when the food condi; tions were changed by the additions of bouillon. B indicates the addition of bouillon. CONTROL OF SEX BY FOOD IN ROTIFERS 273 Diaschiza sterea , In March of 1915 four fertilized eggs of Diaschiza sterea were isolated from a battery jar containing miscellaneous varieties of animals growing in a stable tea (horse manure) solution. This jar was made in September of 1914 and had been left stand- ing since that time without a renewal of stable tea. These fer- tilized eggs were put into fresh tap water for a few hours and then later a little stable tea was added to the water. Within a short time the eggs hatched and produced females. These females were fed upon the colorless flagellate, Polytoma, and reproducing readily produced a culture of many females but no males. The Polytoma was reared in a solution of stable tea as follows: 800 grams of fresh horse manure mixed with 1200 cc. of tap water, was cooked in an autoclave at 15 lbs. steam pressure for one hour and then 1000 ec. of the liquid was pressed out of the cooked manure. This solution, hereafter called stable tea, was used as a stock supply. One part of this stable tea and three parts of sterilized water were mixed and inoculated with Polytoma. Every day one-half of this culture was poured off and replaced by a fresh solution of one part stable tea to three parts water. In this manner a vigorous growing pure culture of Polytoma was daily maintained. The Diaschiza sterea is a small rotifer, no larger than a small paramoecium, and on account of their small size it was found impossible to make individual pedigreed cultures of them. Consequently all the experiments were made in mass cultures either in small watch crystals or in depression slides. In a weak solution of the Polytoma culture (1:2) the female rotifers grew very readily but produced entirely female offspring. How- ever, when these females were transferred to a weak bouillon solution (2 ce. or 5 ce. of the stock bouillon solution to 148 ce. sterilized water) to which pure cultures of Chlorogonium! had been added males appeared within a few days varying in pro- 1This green flagellate formally has been known as Chlorogonium euchlorum but recently Wille has reclassified it as Chlamydomonas euchlorum. Engler u. Prantl, Pflanzenfam. Nachtrage I Teil, 2d Abt., 1911, p. 18. 274 DAVID DAY WHITNEY portion to the females from 10 per cent to 60 per cent. Dia- grams of a male and a female and the different eggs are in figure 3. The details of the experiments are in table 2 and a plotting of experiments 10 to 12, and 16 and 17, are in diagram 4. The Diaschiza sterea are bottom feeders eating only micro- organisms that settle and remain on the bottom of the culture dishes. As both Polytoma and Chlorogonium did this to a consid- erable extent they furnished very good food material. Chlamy- domonas was tried as a food but the rotifers soon died, appar- ently from starvation. At this time another series of experiments was made to deter- mine whether the males were caused to appear by the stimulus of the new food or by the stimulus of the bouillon in which the new food and the rotifers were placed. In order to determine this point some culture water from the Polytoma cultures in which the rotifers were thriving was filtered and the female rotifers put back into it. Then the Chlorogonium which had been cultivated in a bouillon solution (1 part of the stock bouillon solution to 1 part water) were thoroughly washed by placing them in test tubes filled with sterilized water and centrifuging on a large centrifuging machine. This process would, of course, collect the Chlorogonium in a small mass at the end of the tube. This washing process was repeated three or four times until the Euglena were undoubtedly thoroughly freed from any bouillon on their external parts. These washed Chlorogonium were then fed to the rotifers that had been put back into the filtered Polytoma culture water. Males soon appeared and in some experiments as many as 40 per cent were found. These experi- ments demonstrate that the stimulus which caused males to be produced was in the Chlorogonium itself and not in the bouillon. The details of the experiments are in table 3. All the experi- ments with this species of rotifer were done at room temperature on a table in subdued light. Diglena catellina In July of 1915 a few females of the rotifer, Diglena catellina, were found in some general mixed culture jars which were prob- ably inoculated by a collection of various organisms collected CONTROL OF SEX BY FOOD IN ROTIFERS in a towing from a small pond in Middletown. 279 These rotifers are also very small and only mass culture experiments were made, TABLE 2 Experiments with Diaschiza sterea and two diets of pure cultures of Polytoma and Chlorogonium showing that only females were produced in the colorless diet of Polytoma but when the female rotifers were transferred to a diet of Chloro- gonium males appeared soon in the cultures. CONTROL FEMALES REARED IN STABLE TEA nie | PEARED AND CONTINUED IN STA-| TOu, wHANSERRED YO BOUIL: face aa CONTAINING LON CULTURE CONTAINING CHLOROGONIUM EXPERIMENTS Fe- Estimated number of Wee Estimated number of Hours | males Sepa males pape isolated isolated g fou ToS 9 fof % oS 15, CURR he Ree 60 40 400 0 0 20 140 60 | 30 DY Sa oes eee ein 60 40 400 0 0 20 140 60 30 Bs db BG cee AR 60 40 400 0 0 20 160 40 | 20 4b. Solr heen Warn och 60 40 400 0 0 20 160 40 20 ep ees ee hoon nike 60 40 400 0 0 20 180 20 10 (Ccrce been pie tine oe 90 40 500 0 0 20 180 120 40 Use eNO ee ae 90 40 500 0 0 20 210 90 30 See rast eee 90 40 500 0 0 20 225 75 25 Oe ttoekees fe ae 90 40 500 0 0 20 210 90 | 30 LC ceca Re apart 90 40 500 0 0 20 150 150 | 50 ILD pe ARN ale ra ee oe 90 40 400 0 0 20 90 60 40 HDs, TE ene eet eee 90 - 40 400 0 0 20 120 80 40 IS cake eee eee: 90 40 400 0 0 20 90 60 40 NAPE irs. Sie kL ot kone 90 40 400 0 0 20 130 70 35 1U5).09 A ei a RN Se 90 40 400 0 0 20 140 60 30 Grapes Sait onto: 90 40 400 0 0 20 80 120 | 60 eco tics ae a ae 90 40 400 0 0 20 80 120 60 US Roemer chistes) nes ae 90 40 400 0 0 20 120 30 20 1S) oleh tates cee ee eam e cae 90 40 396 4 1 20 105 45 30 PAD ito AC RE ee 90 40 396 4 1 20 96 54 36 Zn bare cic Rca ses Oe PRE EG 90 40 396 4 1 20 135 15 10 Dey PsP Se 90 40 396 4 1 20 90 60 40 DSR aT Lan oy platy, 90 40 396 4 il 20 111 39 26 A ede ead Ciena Re Met 90 40 396 4 1 20 $1 69 46 PENS So oe tae ee 90 40 396 4 1 20 105 45 30 DORIA ES nia ir 90 40 396 4 1 20 iil 39 26 PN (eel ear Te at ERE 90 40 396 | a 1 20 126 24 16 Total 160 1696 | 540 3565 | 1735 32 276 DAVID DAY WHITNEY os rt ae Fig. 3 Diaschiza sterea. A, female; B, male; C, parthenogenetic female egg; D, parthenogenetic male egg; FE, fertilized egg. Ea BERBER RER RAY ae eee eee eee (eae Raho oeee Ba fF Ol Sbiatsel St stet Clete ata ERE CU EESERE SY ACRE ptt tt tt Tt tt SI Fae iz Mar 24) |26l fog 31 Diagram 4 Diaschiza sterea. Experiments 10, 11, 12, 16, and 17 of table 2 Showing the percentage of males produced when the diets were changed. P indicates the feeding of a Polytoma diet, G@ indicates the feeding of a Chloro- gonium diet. PERCENTAGE OF MALES CONTROL OF SEX BY FOOD IN ROTIFERS 207 The female rotifers were placed in a weak solution of Poly- toma culture water which contained many Polytoma. Here TABLE 3 Experiments with Diaschiza sterea showing that it was the food and not the culture media that caused the males to be produced TIME EXPERIMENTS Days eae pete, 8 Als ot PASAT OER 4 Oe keene: 4 BR Wes aeatin es 4 ROR RT 4 Giss eras mats: 4 BRO eR eNe 4 Saeigereren ttle 4 Che bce eee 4 TO ae ticeniene 14 NP era 2 see ie 14 RP cept torsten 14 WS Bes tae ciray ere sy ore 14 Arahat c8 pars Ais 14 LD eysracieltevs, icbieke 14 Gc omere aoe 14 UM ipacrctes. cero) ye 14 Seer eeeiorert eae 14 WO eters assusies x12 14 DOM ert inctoe ais 14 7A Weve ie vere RS Oe 14 Dale wterayave tsa hh 14 GIG HO Ha piece 14 7 ag Se rae 14 7A MEE eae 14 AAD ral ets Shes 14 PA (oR AE Ea 14 DSi icerervere nes 14 Oa ate tee iss aye 14 REARED AND CONTINUED IN STA- BLE TEA CULTURE CONTAINING POLYTOMA Fe males isolated Estimated number of offspring SS Sooaeoeeeooaogqge eo ees Hues 4 ea pa pae ISI IoOosaeeeoeoogege ee Si eae es sae TIME Days PPP PR PPE PPP ER RAED R AR ERA BER DE DP DE DP FEMALES REARED IN STABLE TEA CULTURE CONTAINING POLY- TOMA TRANSFERRED TO FIL- TERED STABLE TEA CULTURE WATER TO WHICH WAS ADDED BOUILLON FED CHLC ROGONIUM THAT HAD BEEN WASHED Estimated number of Fe- offspring males isolated 278 DAVID DAY WHITNEY they lived and produced many daughters which produced in their turn daughters. This was continued for several weeks and many hundred females were produced but not a single male “TABLE 4 Experiments with Diglena catellina showing that when the rotifers were fed the colorless Polytoma diet no males were produced but when the females were trans- ferred to a diet of miscellaneous green flagellates males soon appeared. FEMALES REARED IN STABLE TEA CONTROL CULTURE CONTAINING POLYTOMA wimp | REARED AND CONTINUED IN STABLE TRANSFERRED TO A MIXED STABLE TEA CULTURE CONTAINING POLY- TEA AND BOUILLON CULTURE CON- TOMA TAINING MISCELLANEOUS GREEN EXPERIMENTS See ane Estimated number of Estimated number of Hours | Females offspring Females offspring isolated isolated ce) rot HS Q fof % oS aM mater od 12 20-380 200 0 0 20-30 160 40 20 Die 72 20-30 200 0 0 20-30 170 30 15 Bo oe 02 20-380 200 0 0 20-30 180 20 10 AB aC tks ts Sea ae bs 2 20-30 200 0 0 20-30 160 | 40 20 Dove 72 20-30 200 0 0 20-30 170 30 15 624 SO eee 12 20-30 200 0 0) 20-30 180 | 20 10 Vier Te rae oo We 20-30 200 0 0 20-30 170 | 30 15 8 72 20-80 200 0 0 20-30 160 | 40 20 9 48 15-20 150 0 0 15-20 105 45 30 Otte teres anes 48 15-20 150 0 0 15-20 105 45 30 1 Pe nats ca 48 15-20 150 0 0 15-20 120 | 30 20 OT aAb emeercuctee 2 48 15-20 150 0 0 15-20 135 15 10 13 48 15-20 150 0 0 15-20 120 30 20 UL ee EN res 48 15-20 150 0 0 15-20 120 30 20 IIa ae eer ewe rin 48 15-20 150 0 0 15-20 120 30 20 68s 2.5. he oe 90 50 500 0 0 15-20 160 40 20 TSP ere ee: 90 50 500 0 0 15-20 170 30 15 18 90 50 500 0 0 15-20 180 20 10 AS a Arte cs Phy 90 50 500 0 0 15-20 160 40 20 PA eee Rae Pa 90 50 500 i) 0 15-20 140 60 30 DAs 6 Ce 90 50 500 0 0 15-20 170 30 15 individual appeared during this time. However, when some of these females were taken from the Polytoma culture and placed in a weak culture water of bouillon and stable tea (10 CONTROL OF SEX BY FOOD IN ROTIFERS 279 PERCENTAGE OF MALES Diagram 5. Diglena catellina. Experiment 20 of table 4. Showing the percentage of males produced when the diets were changed. P indicates the feeding of a Polytoma diet, G indicates the feeding of a mixed green flagellate diet. c Fig. 4 Diglena catellina. A, female; B, male; C, parthenogenetic female egg; D, parthenogenetic male egg; H, fertilized egg. 280 _DAVID DAY WHITNEY ee. bouillon, 2 ec. stable tea, and 138 ec. tap water) in which there were growing a mixed culture of various green flagellates many male individuals appeared within a few days. The average percentage of males produced in this species is lower than it is in the preceding species. This may have been due to the inferior quality of these flagellates as a food for this particular species of rotifer, or as these rotifers are also bottom feeders perhaps they could not obtain a superabundance of food as many of the flagellates were more or less free-swimming. These experiments were all done at room temperature on a table in subdued light. Diagrams of a female and a male rotifer and the different eggs are in figure 4, a plotting of experiment 20 is in diagram 5, and the details and the results of the experi- ments are in table 4. Pedalion mirum In February of 1915 some of the same dried mud that was collected at Cold Spring Harbor in 1908 and produced Brachionus pala in February of 1915 was put into weak stable tea water (3 cc. stable tea added to 147 cc. water) and after a few days several females of the jumping rotifer, Pedallion mirum were found. After considerable experimenting a suitable method was found by which large numbers of these rotifers could be readily reared. In some experiments weak solutions of stable tea alone were used and in others bouillon was added to the stable tea solu- tion. All the cultures were inoculated with a miscellaneous collection of green flagellates and then kept in direct sunlight as much as possible. The temperature was not allowed to rise above 37°C. This was rendered possible by placing the cul- ture jars in a large pan through which water flowed from the tap. At night, however, the temperature was that of the room, 20°C. After the cultures had been progressing for a few days the flagellates, mostly Chlamydomonas, and the rotifers became more or less balanced and only female rotifers occurred. How- CONTROL OF SEX BY FOOD IN ROTIFERS 281 ever, when new stable tea or bouillon was added to the bal- anced jars a rapid increase in the number of flagellates took place. This furnished a superabundance of food for the rotifers and they became loaded with eggs. This superabundance of food could be maintained for several days and in the mean- time a great number of the rotifers were seen to be carrying small male eggs. Some females had as many as eighteen of these eggs attached to the outside of their bodies. A little Tne elele ain anmiela ps eT Ce miniaianael 2). Dee ERE ne ee soe eee ones =|) (Se Abts oo Rese oeoS Se | (eae ne al he ahs ih a ao ead Dep anton a Ste alcatel hast in| el atael tabebalest satel Te Wott 2) (oan Sen De Tae w eens aoe SOS 200 Se ese eo See ens Ie eoonn 2.0 (RES Sea Seo esses eee 2) \ Lens U OPS o ese oe Wee 2. SOC OeSREee eae ee Coe Bee ee =) |) ESCO EE REeos Sosa ee ee | tS a NY ST Se pe ee alc fete fon [a a fed hel cade OT S ies Sado foie fete] ma ise a | = Z 1S) i a w El PLETE Ad PY Hl Diagram 6 Pedalion mirum. Experiment 1 of table 5. Showing the pro- duction of a high percentage of male-producing females when the food conditions were changed by the additions of stable tea. 7 indicates the additions of stable tea. later there would be found a very few females carrying female eggs, a few carrying male eggs, and a large number carrying one fertilized egg on the inside of the body. In some Jars as many as 90 per cent of the females bore male or fertilized eggs. After a time nearly all the flagellates were eaten and nearly all the rotifers would die but usually a few survived and as the flagellates would begin to increase again in a few days there would be produced again a balanced culture. In such cul- 282 DAVID DAY WHITNEY tures only female rotifers were usually found. If new stable tea or bouillon was added another epidemic of males was pro- duced. Diagrams of the different females carrying eggs and also a diagram of a male are in figure 5. A plotting of experi- ment | is in diagram 6 and the details of the experiments are in table 5. TABLE 5 Experiments with Pedalion mirum and a mixed diet of miscellaneous green flagel- lates showing that when the flagellates were caused to be very abundant by the addi- tion of fresh culture media the percentage of the male-producing female rotifers was very high and the percertage of the female-producing female rotifers was very low, but when no fresh culture media was added, thus causing the flagel- lates to occur in moderate numbers, the percentage of male-producing females was very low and the percentage of the female-producing females was very high. 1915 CULTURE WATER 99 [oped %S? 2 cc. stable tea 150 ce. tap water July. a Miscellaneous green flag- | ellates July 24 2 ee. stable tea added.... July 26 2 ce. stable tea added.... July 28 2 ec. stable tea added.... — | July 30 2 ce. stable tea added.... = | August Dalle as ees cet Rad So ee i 200 0 0 = August DIGB | Rite ene ia matey acces ket meron sary a 100 100 50 ‘& | August CM an ae tela ial PR 120 80 40 & | August SIRaN Ping ee Sa eT Ee ak ee 160 40 20 Ee | PANT UES tee LOB eh ane, eek mt Oe ee eee 270 30 10 August 1 ites Ps aA Seek A PRE cs 398 2 0.5 August 13 2 ec. stable tea added.... August Pasar | es LS ey Se Ao eo 70 30 30 August 1 USN IPE EAR TS oh MORRO ea 5 BE 10 90 90 August TESS lepabeiee Ceneie ola Ik Pam a Lat) of 60 40 40 August 7 Oak BN ier at a ee Eas 3 90 10 10 RO tale tremcrc stir ee ee 1558 342 DAVID DAY WHITNEY TABLE 5—Continued 283 1915 CULTURE WATER 99 foun) Jo SR 2 cc. stable tea ay 9 150 cc. tap water Miscellaneous green flag- | ellates July 24 2 cc. stable tea added ... July 26 2 ec. stable tea added ... July 28 2 ec. stable tea added ... July 30 2 ec. stable tea added.... ~ | August 2 5 ee. bouillon added...... 200 0 0 = August 5 5 ce. bouillon added...... 200 0 0 = August Ch ls eee rey ies SRS aha Sot aRe st 200 0 0 ‘= | August SBIR ohare Se ete et) Wet Sea 160 40 20 e. August OES ROU ese MARL, Semin e OM! 291 9 3 = | August 11 2 cc. stable tea added... . August 13 2 cc. stable tea added.... August UA aes |More Sosa ve cach toes as ae we ee 90 10 10 August IL Tan hose aren Sek erp, Ps) eae 60 40 40 August LSS |g syst so eon ee eee Me ey 50 50 50 August DAU )SBiE | oe Aa per tia teeas 90 10 10 Total eee. ot 1341 159 (75 ce. old stable tea cul- ture take 98 75 ce. tap water : 5 ce. bouillon Miscellaneous green flag- a | ellates 2 August 5 2 cc. stable tea added... . 2 | August ii 2 ec. stable tea added... . 70 30 30 £ August Ym ce CDS eV Fae es 100 0 0 S| August LON SIS sae, Cp aera nomen etek el Sy Ay 100 0 ) SI August 11 2 ce. stable tea added... .| August 13 2 cc. stable tea added.... August Ar alin oe. ea OS tpl os Sue 50 50 50 August LILO i Mates otis? apeits a es lc a Pe aga 60 40 40 August LS et |e AEC preted 1 te 400 1 0.25 3 dG Calaee eats ay cet 2 780 121 THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 2 284 DAVID DAY WHITNEY TABLE 5—Continued 1915 CULTURE WATER 29 lone) % SP? (75 cc. old stable tea cul- | ture 75 cc. tap water 5 ec. bouillon Miscellaneous green flag- ellates August 2 5 ce. bouillon added...... July 27 ES August 3 5 ec. bouillon added...... Fy 2 cc. stable tea added.... 4 | August 5 2 ec. stable tea added.... 3 August i 2 cc. stable tea added... . 40 160 80 g August LK Obetll 92 -clichsi Ste a RR ERE ey Mee 396 4 1 August 11 2 cc. stable tea added.... August 13 2 ec. stable tea added.... August GE She ch MEME SOR CMS tee storecl 140 60 30 August 15 2 cc. stable tea added.... : August 1G) haley arabe tetanic caNtar 3. cibaces 100 100 50 August TSP Whee. yt ore Ce eek, Meee eae oe 200 0 er ) ANG tial dee ay ens 876 320 40 ec. old stable tea cul- ture F 110 cc. tap water els: ee 5 ec. bouillon Miscellaneous green flag- ellates a August 5 2 ec. stable tea added.... & | August 7 2 cc. stable tea added.... 96 4 4 = | August YE Me es ae ee oe es 285 15 5 3 August 11 2 cc. stable tea added.... A August 13 2 cc. stable tea added.... August TTR Leer tNere ae Seon ee ee 160 40 20 August 15 2 ce. stable tea added... . August Gr WES, bees 2h ae oR OA Te 150 50 25 August LS Ral hiss vce Pee ecm tock ee eee 90 10 10 August DO Lee eae CREE Sep ots al ee 200 0 0 CONTROL OF SEX BY FOOD IN ROTIFERS 285 SS Mey, Ui S a =) eS Ks Ss Me IN\\\ A vy Le LY 5 S LZ TTT NN\ ul WW, a j Loy 2! Fig. 5 Pedalion mirum. A, female with attached parthenogenetic female eggs; B, female with attached parthenogenetic male eggs; C, female with the fertilized egg inside her body; D, male. 286 DAVID DAY WHITNEY THE INFLUENCE OF A SINGLE DIET In all of the preceding experiments there were used either two diets, one of colorless flagellates alternated with one of green flagellates or a mixed diet containing miscellaneous species of protozoa, principally green flagellates. During the experiments with the mixed diets when at times they consisted of nearly all of one species of flagellate, Chlamydomonas, it was suspected that one favorable diet could be so manipulated as to yield either female-producing or male-producing rotifers. In order to test this hypothesis pure cultures of the green Chlamydomonas were made in mixtures of bouillon and stable tea solutions. These cultures were inoculated with the rotifer, Brachionus pala, and several experiments were started and carried through to completion. The flagellate, Chlamydomonas, passes through several stages in its life cycle, some of which are to the advantage of the roti- fers and others of which are distinctly to the disadvantage of the rotifers as far as being of food value. In favorable cul- ture solutions the individuals of Chlamydomonas are very active during the sunny part of the day but do not reproduce very much. However, at night, having reached their full size, many of them become motionless and begin to divide inside of their outer envelope. Here each divides three times resulting in eight small individuals. Sometime in the morning hours the parent envelope breaks and the eight young come out and soon become separated from each other. These young individuals are active during the day while increasing in size and at night having reached their full size they become quiescent, begin to divide, and before morning break into eight small individuals. There is considerable variation in this life cycle. Some may be in the eight-cell stage and break out in the early evening while others may be in such stages in the late morning but the large majority go through the cycle as indicated. When the Chlamydomonas are small in size they are the most readily eaten by the rotifers. As they reach full size, although they may be very active, they are not eaten at ail by the rotifers. CONTROL OF SEX BY FOOD IN ROTIFERS 287 In fact in some old cultures where there were great numbers of these full sized flagellates in an active state but not dividing the rotifers died from starvation. The rotifers are transparent and one can see the food in their stomachs after they have eaten. By adding fresh stable tea or bouillon or both to the cultures enormous numbers of the small sized Chlamydomonas were produced and in a short time the jars were swarming with fe- male rotifers each of which was carrying many small male eggs. Then if no fresh stable tea or bouillon was added nearly all the flagellates were soon eaten, and consequently nearly all the rotifers died. After a short time the few flagellates that had escaped would increase their numbers considerably and the few rotifers that had escaped starvation could thus get a mod- erate amount of food. Soon the flagellates and the rotifers would form a balanced culture and continue thus for many days. In this balanced condition nearly all female-producing females were produced. However, if fresh stable tea or bouillon were added again to the culture an enormous increase in the num- bers of the small sized Chlamydomonas took place accompanied by a rapid increase of the female rotifers. Soon a very high percentage of these females would be carrying male eggs. Thus it is shown that a moderate amount of the diet Chlamydomonas, will cause only female-producing females to be produced but that an abundant diet or a superabundant diet of the same Chlamydomonas will produce in some experiments as high as 90 per cent of male-producing females. The details of these experiments with a pure culture of Chlamy- domonas are seen in table 6 and a semi-diagrammatic plotting of the average results of the nine experiments are seen in diagram 7. Diagrams of the different stages in the life cycle of Chlamy- domonas is seen in figure 6. 288 DAVID DAY WHITNEY TABLE 6 Experiments with Brachionus pala and a pure diet of Chlamydomonas showing that when the Chlamydomonas were caused to be very abundant by the addition of fresh culture media the percentage of the male-producing female rotifers was very high and the percentage of the female-producing female rotifers was very low but when no fresh culture media was added, and thereby causing the Chlamy- domonas to occur in moderate numbers the percentage of male-producing females was very low and the percentage of the female-producing females was very high. 1915 CULTURE WATER 29 lose) % SH 10 ce. bouillon September 22 140 cc. tap water Chlamydomonas September 23 10 cc. bouillon added..... September 25 10 cc. bouillon added..... _ | September 27 10 ce. bouillon added..... = September 28 3 cc. stable tea added.... © | October 9; 10 ec. bouillon added..... PE MaOe coer pemly ir. oust ee eet 20 80 80 aha OCtopermeetar || Pace. 5.5 senha cee 20 180 90 4 | October 20 10 ce. bouillon added..... 95 5 5 October 28 10 ec. bouillon added..... October Ptol. vl) Geese aoe eee 80 120 60 INowerlben ale. \|Puke co. penta. ah mee a ee haearier, ee ean 50 150 75 November "SiN |b 5 s5) ain. 4a en eae nee eae 200 0) 0 TROGIR Nia nO, eee ae. 465 535 { 3c. stable tea | : 10 ee. bouillon Cale: 2 137 cc. tap water Chlamydomonas “ | October 4 10 ce. bouillon added..... a October 6 10 ec. bouillon added..... S | October 1 10 ec. bouillon added..... Pena October. .2tan| Mik coe ess ee eee 40 160 80 & | October 28 10 ec. bouillon added..... 95 5 5 | November 1 10 ce. bouillon added... .. 200 0 0 INiGivemallo er: 48 | Pager cee Uae aah cise ep ee 20 80 80 Movember iG |) oe ee ser eee ee 98 2 2 otalbeerr ces oc. err eee 453 247 TABLE 6—Continued 1915 CULTURE WATER 22 lope) NS? | 3 ce. stable tea 10 ec. bouillon ee 137 cc. tap water ee October 4 10 ce. bouillon added..... October 6 10 ce. bouillon added..... October 11 10 ec. bouillon added..... October OU ee asek tesa ems. eas oats. 40 160 80 OchObera mr Zeallt mane) Sordteet A similar observation was made by Hérouard (’89, p. 689). 306 W. J. CROZIER anal sphincter.° Closure of the anus was also a preliminary operation when Stichopus was beginning to engage in locomotion. The control of anal rhythm by the animal as a whole includes, then, (1) the cessation of pulsation by causing the anus to constrict (either in response to stimuli, or in locomotion), and (2) the great restriction of the amplitude of pulsation (in “spout- ing’) due to the contraction of the radiating muscles and the muscles of the body. Of these modes of control only one, con- traction in response to stimulation from the outside, persists in the isolated cloaca. As regards the function of cloacal pumping in holothurians, it is clear from the work of Bordas (99) and Winterstein (’09) that the water drawn in and: out of the respiratory trees serves to supply the coelomic fluid with oxygen, to remove excretory products, and at the same time to control the amount of fluid contained within the animal, upon which its locomotor move- ments depend.? In connection with the respiratory function of pulsation it is of interest to note that if by repeated mechanical stimulation, pulsation is prevented from occurring for some min- utes, the movements which occur when pulsation is resumed are not increased in rapidity, but are of greater amplitude than those normally seen in undisturbed specimens. ‘This was also noted by Pearse (08, p. 272) in the case of Thyone, and he drew the conclusion that the animal was in this way making up for its preceding oxygen deficiency. This deduction by no means follows, however, because when repeatedly stimulated holothu- rians decrease greatly in volume (cf. Crozier, ’15) by expelling water from the anus, and when left undisturbed after such stimulation they tend to return to their original size. In doing this the first pulsations of the new series are more vigorous than usual, as is also found after each normal water expiration. The increase in amplitude observed after forced cessation of pulsa- tion is therefore only remotely connected with respiration; this 6 These reactions were similar in every respect to those of the previously described H. surinamensis (Crozier, 715 [?]). 7 The movements of holothurians have recently been analyzed by Jordan (’14) from the standpoint of his conception of a ‘Hohlorganartig’ animal. RHYTHMIC PULSATION 307 view is supported by the low oxygen requirements of echino- derms, to be considered subsequently, and the known presence’ of other modes of respiration than that associated with the respira- tory trees. The phenomenon of increased amplitude of pulsa- tion subsequent to forced inactivity in a pulsating structure is also well known in the vertebrate heart, and may be seen, for example, in such curves as those plotted by Vernon (’10) to show the recovery of the heart-beat after perfusion with protoplasmic poisons. The occurrence of anal rhythmic pulsations in holothurians is not an isolated phenomenon, for such movements (with a sug- gested similar function) occur among Enteropneusta (Willey, ’99, p. 244), both in the adult® and in Tornaria (Willey, 99, p. 306), and in decapod crustacea (Miller, ’10), etc. In the lobster and crayfish the muscular arrangements for producing anal rhythm are in a general way similar to those in holothurians, since in the former case circular muscles and radiating muscles running to the body wall also occur; furthermore, the codrdinating mechanism is a local one (Miller, ’10, ’12). c) Correlation with size. It has long been a matter of general knowledge that the activity of animals varies with their size, smaller animals being more active than larger ones. This rule holds conspicuously for the execution of rhythmic movements by animals of the same species. The rapidity of breathing move- ments, ete., are not, however, simply proportional to the recipro- cal of length or any other body measurement, but the empirical curve expressing the relation between size, or weight, and activity is almost invariably of a rather complex hyperbolic type. Poli- manti (713) has recently reviewed the literature of this subject, and has supplied an additional example in the respiratory move- ments of Octopus. The equation derived from his data is of the form: y =at ba + ca? + where y = weight, and x = respiration rate. 81 have observed that the amputated posterior end (4-5 mm. long) of the Bermudan Ptychodera sp. will pulsate in sea water. A sphincter ani is present. 308 W. J. CROZIER The rate of rhythmic pulsation of the anal sphincter for ani- mals of different sizes was observed in Stichopus, Holothuria surinamensis, H. captiva, and Cucumaria punctata. The results are given in figures 6 to 9. The data in the case of Stichopus (fig. 9) are more irregularly distributed than with the other. Relation between size and pulsation-rate S vee : NS < S xa 70 N O Q oY. © X 50 V § 40 Leng lf Fig. 6 Cucumaria punctata. Temp. = 24.5° 8 Ss ‘ Jitne Jor 10 puksal1aprs. a oS 50 B Se & Sy uot te tet ae kenglh. Fig. 7 Holothuria surinamensis. Temp. 25.0° forms. This is due to the fact that in the instances measured the individuals of the other species were of more homogeneous history, in that their small size made it possible to keep numbers of them in a single aquarium, so that their pulsation rates were RHYTHMIC PULSATION 309 all measured at one time. Because of their bulk the specimens of Stichopus had to be studied separately, often at intervals of some days, so that various factors, including temperature differ- ences, probably entered, tending to lack of uniformity in the results. Relation between size and pulsation-rate SO SEC. 8 8 8 Mine fe 10 pulsalorrs. Nes 7 8 cM. aN 5 € Leng. Fig. 8 Holothuria captiva. Temp. = 26.0° SEC. ® Ss Ny is) 9) S 10 20 30 40 50 Om. Fig. 9 Stichopus mobii. Temp. = 24°-26° 310 W. J. CROZIER The smooth curves in figures 6 to 9 are in a general way similar to that of Polimanti (’13) for the rate of respiratory movements in Octopus. The weight of Stichopus is a simple function of length, such that Weight in grams = (0.034 =) xX (length in cms.)* 10 20 3O cm. Length. Fig. 10 Relation between weight and length in Stichopus: —O—O—,, weight including that of the body fluids; —@——®—, weight of the integument alone. (fig. 10); this type of relation between weight and length is also true for the other species. If pulsation-rate were plotted against weight the resemblance to Polimanti’s curve would be increased. As shown in figure 10, the weight of the integument alone is too variable to give smoother results; the variation is produced mainly by the fact that the skin seems to absorb water to a vary- ing extent, depending upon its rigidity. The curves relating size RHYTHMIC PULSATION 311 to pulsation-frequency are of the same type as that found by plotting Mayer’s (’06, p. 8) data on diameter vs. pulsation-rate in Cassiopea. In attempting to account for the relation of pulsation-rate to size, it is possible to adopt the view that the rapidity with which an animal executes a given act is a measure of the amount of energy available for the performance of that type of work, and that when operating under similar conditions the relative energy content of different individuals in the same species may in this way be compared. According to this interpretation, larger ani- mals must contain less motor energy, proportionately, than do smaller ones of the same kind. The observations of Tashiro and Adams (’14), that the eardiac ganglia of large Limuli have a lower output of CO, per gram of nerve-substance than do the (smaller) corresponding ganglia of smaller Limul, and those of Child (13, p. 140), who found by the KCN method that young (small) individuals of Planaria had a higher rate of metabolism than larger ones, lend support to this general idea. III. COORDINATION OF THE PULSATING COMPLEX The pulsation of the cloaca exhibits a high degree of coérdi- nation in the action of a number of individual effectors. The questions arise, From what center, if any, does the stimulus to pulsation proceed, and by what meaus are the various muscles brought to act in appropriate sequence? The following observations and experiments bear upon the answers to these questions. The points to be considered are (1) the cycle of pumping movements in the cloaca, and (2) the cessation of these movements, with the exception of those of the anal sphinctér, during the expulsion of an expiratory stream. It may be mentioned here that during defecation the sphincter movements are not interrupted, and as a result the faecal masses present beaded constrictions at regular intervals along their length, the constrictions (fig. 11) being formed by the pressure of the sphincter as it attempts to close. This applies to Stichopus and Holothuria; but in Cucumaria there are no 312 W. J. CROZIER definite castings, because the ejected material is discharged in fragments during the course of an expiratory act. a) Effects of autoevisceration. When Stichopus undergoes auto- evisceration the cloaca ruptures in the region of the termination of the intestine, and the gut is passed out, pulling with it more or less of the respiratory trees; the whole then being autotomised, Fig. 11 A casting of Stichopus (X 72;), showing the constrictions, somewhat exaggerated, due to the continuance of anal rhythm during defecation. Fig. 12 Stages in wound closure, drawn from one preparation, at the periods indicated. the animal which remains consists of the dermo-muscular tube and the cloaca. Such an animal continues to exhibit cloacal pulsation, the water, however, being now pumped into the body- cavity directly. The rate of pulsation in eviscerated animals was normal, up to at least twelve hours after autotomy, and the characteristic interruption of the inspiratory movements after every sixth to tenth one, by the expulsion of water, continued as RHYTHMIC PULSATION 313 before evisceration. The general tonal depression of the animal involved some decrease in pulsation amplitude. The stimulus to spouting, then, does not necessarily originate from a state of tension in the full respiratory trees. b) Effects of amputation. The cloacal end of Stichopus when amputated at the level a——a (fig. 1) contracted firmly. But if it was then put into sea water, or into one of a variety of salt solutions subsequently to be described, it very shortly opened up and continued to pulsate rhythmically, though with a gradually decreasing frequency and amplitude, for as much as thirty hours. The duration of pulsation depended upon the size of the excised piece, the volume and composition of the surrounding solution, and the existing temperature. About one minute after excision the cut edges of the body wall began to bend inward (fig. 12 8). The cut edge of the cloaca itself was flared outward, while its lumen was closed by the con- traction of a powerful circular muscle about midway between respiratory trees and anus. The radiating muscle-strands near the plane of the cut were relaxed, but they contracted when pinched. The flared anterior end of the cloaca was then pulled toward the anus, while the cut edge of the body wall was bend- ing inward. These processes, tending to close the remnant of the coelom, had a highly protective appearance. Some three to four hours after the inbending of the edge of the cut, the body wall in this region became flabby and relaxed, and a progressive degeneration, which involved swelling and mucoid disintegration, began at the cut edge. After about five to six hours the inter- radii became sunken inward, so that a cross section of the pul- sating piece had the appearance shown in figure 13; in addition, the previously inturned edges of the body wall were now relaxed. At this time, that is after about six hours’ isolation in sea water, the excised cloacae looked like the one sketched in figure 14. The history of individual preparations, made in the way de- scribed in the preceding paragraph, was followed under various conditions until they ceased to live. I shall refer here merely to the performance and fate of cloacal pieces contained in one liter of sea water. The time required for the execution of 10 rhythmic THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 3 314 W. J. CROZIER 13 Fig. 13 Cross-section of a cloacal piece, showing the sunken condition of the inter-radi. Fig. 14 Sketch of a pulsating piece, from above (X zs). tbrl., one of the characteristic wart-like tubercles. tbrl.’, the delicate tactile papillae which they bear (not noted by previous writers). movements of the anal sphincter was measured at intervals of time subsequent to the isolation of the pieces. Exhaustion curves obtained in this way are illustrated by figure 15. A typi- cal example is given in detail: Experiment 33.8. June 21. Stichopus moebir; 24 cms. long; normal pulsation, 58 seconds for 10 movements. ‘Temperature varied between 26° and 24°C. SECONDS REQUIRED MINUTES ELAPSED FOR TEN MOVEMENTS N@GTES 0 Cloacal end amputated; put in 1000 cc. sea water 2 Sphincter opening; beginning to pulsate 5 Pulsation becoming more regular 40 81.0 Amplitude fuller 85 68.5 Normal 220 91.5 290 133.0 350 176.0 Closure no longer complete 755 210.0 1145 21255 1165 212.0 Barely pulsating. Soon stopped There were in general two modes of exhaustion; one, a rapid type (marked ‘A’ in fig. 15), the other showing a sort of ‘fatigue level.’ The first type of exhaustion was exhibited by pieces of RHYTHMIC PULSATION 315 shorter length than that of the other ‘normal’ pieces, i.e., cloacae cut off below the level a a (fig. 1). It is possibly of signifi- cance that the earlier part of these exhaustion curves bears a superficial similarity to those of autocatalytic reactions, which might be taken to indicate that there is here an autocatalyst of exhaustion, represented by fatigue products. The sharp upward 35O 8 8 S _Jiine Sor Jer PEISALIONS — SECONaAs =~ Q ‘sok (6) 100 200 G00 400 $500 10900 1500 Duration of pulsations in minutes bend of the curves near the time of cessation of pulsation indi- cates that pulsation continues until some critical condition is arrived at within the pulsating mechanism, beyond which spon- taneous rhythm is not possible. The exhausted condition was 316 W. J. CROZIER not reversible experimentally, though sphincters which had ceased to beat rhythmically in sea water would pulsate once in response to a touch or to the local application from a pipette of a small volume of.a stimulating solution, and they could momentarily be revived by immersion in sea water free of Ca; possibly the mechanical stimulation involved in the last experiment was responsible for part of the pulsation. A very significant fact about the pulsation exhibited by the isolated cloacal ends, is that the rhythm was not interrupted at intervals corresponding to the expulsion of the cloacal stream in the intact Stichopus, but was indeed perfectly continuous, unless complete contraction was induced by some especially applied stimulus. This condition is intelligible in view of the fact above noted that in the normal animal the edge of the > sphincter was observed to pulsate with faint amplitude during spouting. It follows that the stimulus to spouting has its origin outside the cloaca. The early parts of the exhaustion curves show a peculiarity which must be noted. Immediately after amputation and im- mersion in sea water, the contracted musculature of the cloaca began to relax, and by two minutes, at most, had begun to open. Very commonly it then remained wide open for some seconds, and when pulsation was resumed the rate of sphincter movement was very slow and closure incomplete. The rate of movement was soon improved, however, coincident with the institution of contractions and relaxations of the cloaca, which were of maximal amplitude. It would seem that the codrdination of the members of the pulsating complex is disturbed by cutting this complex away from the rest of the animal, and that some little time must elapse before harmonious interaction can again be established. The full amplitude of the pulsation was usually revived very suddenly. - Rhythmic pulsation in the isolated posterior ends was similar in all essential respects to that in the intact animal. The time relations of the several phases of movement are given in table 3, which may be compared with table 2, showing the normal condition in the intact Stichopus. In the isolated pieces it could RHYTHMIC PULSATION 317 be observed that the cycle of movements constituting a com- plete pulsation began at the cut end of the cloaca. TABLE 3 Time relations, in seconds, of the phases of pulsation for the isolated cloaca in sea water EXAMPLE 1 2 OVER DIOR aes Nas ale ceot.c ae HO mre Ose anne an are 5.6-7.3 7 .0-8 .0 ClNseUey WAG. are Me wet ens sate Soe bie ge oe 3.0-5.7 5.2-5.6 The direction of the current of water produced by the isolated cloaca was studied with the aid of carmine suspended in sea water, and also by small ‘flags’ of mucous or bits of thread at- tached to the inner edge of the sphincter. The current was directed anteriorly, as in the intact animal. The maximum fluid pressure developed in the amputated pieces was ascertained by inserting a glass tube of appropriate size into the anus, which .- then contracted tightly about the tube. The pressures were never more than 1 cm. of sea water. In the intact animal the cloacal pressure is much greater, since the body muscles then play a greater part in the pumping. The pulsating sphincter exhibited a refractory period, such that if the sphincter were touched at the very beginning of an opening movement, it continued to open to its normal extent; but if touched at the edge when more than one-half open, the brim closed down promptly, though some local retraction was evident at the point stimulated (fig. 16). It was pointed out in a previous section that the rate of cloacal pulsation was correlated with the size of the holothurian. It was important to learn if this correlation persisted in the excised pos- terior ends. The rate of pulsation in cloacal pieces of equivalent size (1.e., amputated at the level a——a in fig. 1) derived from 10 Stichopi of increasing lengths is tabulated in table 4, from which it will be seen that, though some trace of this general effect may be maintained, it is by no means clear cut; the rate of pulsation in the isolated posterior ends was very rapidly reduced to an approximately uniform level soon after excision. 318 W. J. CROZIER TABLE 4 Relation of pulsation rate in the excised cloacae to the size of the animals from which they were obtained. Temperature = 26.0° ’ TIME FOR TEN PULSATIONS NO. LENGTH Before amputation After amputation cms. seconds seconds 1 Pye 55.3 88 .6 2 24.0 58 .0 Aeil 3 26.0 64.3 88.0 4 27.0 80.0 90.2 5 28.0 84.1 100.2 6 28.3 85.0 89.5 7 29.0 65.4 85.0 8 30.0 70.0 84.3 9 31.0 18.9 86 .4 10 31.0 83.0 80.8 16 ; yy ae Fig. 16 Reaction of the opening sphincter to tactile stimulation at the point ip (OX 12). Fig. 17 For explanation, see text. Fig. 18 For explanation, see text. RHYTHMIC PULSATION 319 The rate of rhythmic movement is, however, very much con- ditioned by the length of the cloaca included in the cut off piece. In table 5 is given a summary of the history of three posterior ends of different lengths cut from animals of the same size, which illustrates this point. TABLE 5 Dependence of pulsation rate and duration upon the length of the excised piece. A = complete cloaca; B = about three-fourths of cloaca; C = one-half of cloaca; each from a Stichopus 26 cm. long TIME FOR TEN PULSATIONS TIME ELAPSED SINCE AMPUTATION A | B CG minutes seconds seconds seconds 10 71 93 115 70 81 90 123} 250 99 125 foe) 1 Trregular movements. This suggests that the stimulus to pulsation originates in the anterior part of the wall of the cloaca with its associating radiat- ing muscles, in the sense that the structure which of itself pul- sates in sea water faster than any other part of the cloaca is located there, and that its activity carries with it that of the other contractile parts. A somewhat parallel case is found in the vertebrate intestine, where the transmission of pulsation depends upon the myenteric plexus (Cannon and Burket, ’13); Alvarez (’14) found that the rate of pulsation of strips from the small intestine varied inversely with the distance from the pylorus. The localization of the origin of pulsation in the cloaca of Stichopus will be further considered later in this paper. Implied in the above description is the view that the fastest beating member of a pulsating complex determines the rate of movement of the whole. This idea was put forward by Loeb (00 p. 29) in his explanation of the reversal of pulsation in the tunicate heart; the neatest demonstration of the operation of this principle is probably found in an experiment of Mayer (11, p. 7), who grafted together a large and a small Cassiopea and found that the faster beating medusa determined the pulsa- tion rate of the whole mass. 320 W. J. CROZIER The possibility presents itself that the stimulus to pulsation normally arises outside the cloacal region, and that the cut-off pieces continue to pulsate from a sort of ‘habit’ or ‘organic memory.’ It, indeed, is possible to impress rhythms upon holo- thurian structures, as I have discovered in certain experiments upon the physiology of the ‘shading reflex.’ Shadows were cast upon the anterior end of a Holothuria surinamensis at 0.25 minute intervals, to which the animal reacted (cf. Crozier, ’14, 15) by more or less complete retraction of the tentacles and contraction of the buccal sphincter; after the 115th successive stimulation and reaction, the rhythmic shading was discontinued, but the animal continued to retract the tentacles, etc., at very nearly 0.25 minute intervals for the next succeeding 3 minutes (1.e., 11 times), after which the ‘reactions’ became of irregular occurrence. This observation was repeatedly confirmed.® But the fact that the rhythm of the isolated anal parts could be caused to stop, by appropriate sensory stimulation, and then to resume again in perfectly reversible fashion, argues against this interpretation, as does indeed the whole behavior of the pieces in different salt solutions. c) Mutilation experiments. Stichopus having the oral end, including the nerve ring, amputated, continued to exhibit rhyth- mic anal movements like those of the whole animal. As long as the new anterior end remained closed by the close approximation of the inturned edges of the cut, so that some internal fluid ° Quantitative investigation of this phenomenon is contemplated, and should provide important data relative to the physico-chemical nature of ‘protoplasmic memory.’ Somewhat comparable rhythms of short period, impressed by experi- ment, are not unknown among plants. It is especially to be noted that the condition here described in Holothuria is one of “positive memory,’ in contrast to the ‘negative memory’ [the terms are my own] studied by Piéron (’09, and subsequent papers), who investigated the law according to which the sensitivity of snails to rhythmic shading was abolished. Piéron was, I believe, dealing with a condition primarily of sensory exhaustion. Both ‘kinds of memory’ are capa- ble of mechanistic analysis, but that exhibited by Holothuria is more advan- tageous for experimental work. I venture to predict that the study of the impression of short-period rhythms upon animals, rather than the investigation of ‘tidal memory’ (of convoluta, etc., cf. Kafka, ’14, Chap. 8), will afford the clue to the dynamies of primitive associative memory. ; RHYTHMIC PULSATION 321 pressure could be produced, the typical spouting movements also occurred. Animals with the anterior end excised, however, rapidly lost tone and, in the case of Stichopus, died within a day or so. With Holothuria surinamensis, as appears also to be true of Thyone (Scott, 714, p. 289), anal rhythm becomes slow and weak soon after the amputation of the anterior end. This is associated with a general loss of tone and a totally quiescent condition, which is only removed upon the regeneration of a new anterior end (Crozier ’15 [?]). The stimulus to ‘spouting’ is therefore probably derived from a condition of general body tension, resulting from the pumping of water into the interior of the body when the muscular integument presents a volume of definite size. The stimulus is not, necessarily at least, con- ditioned by a state of tension in the respiratory trees alone, since the eviscerated animals behaved in this respect as did the com- plete ones. . The muscles concerned in pulsation of the amputated cloacal end have previously been enumerated. Experiments were carried out to determine the significance of each of these. The results may be briefly stated thus: Cutting out the cloaca, so that only the dermo-muscular tube and sphincter remained, resulted in complete cessation of move- ment. Scraping away the radiating muscles and connective- tissue strands had the same effect. Cutting the radiating mus- cles on only one side gave preparations which pulsated at normal rates; but in these preparations the side of the cloaca and brim on which the muscles were still intact closed and opened before the other side, the rest of the cloaca lagging behind in such a way as to give the impression of being ‘‘dragged along” with the intact half. The anal brim when cut out by itself remained open in sea water and did not pulsate, though it reacted, by a single closure, to delicate mechanical stimulations and to small vol- umes of various stimulating solutions. If preparations of this sort, i.e., isolated sphincters—pieces deprived of the cloaca—or posterior ends in which all the radi- ating muscles had been cut, were placed in solutions of unusual Na/Ca content, they did exhibit rhythmic movement for some 322 W. J. CROZIER minutes. These effects were secured in solutions 6f the fol- lowing composition : 1) 95 ce. sea water + 5 ec. 4} sodium citrate. Cloacae with the radiating muscles cut pulsated for about fifteen minutes, but irregu- larly; parts of the anal sphincter closed before others, so that the anus presented a ragged outline. 2) Van’t Hoff solution without Ca. Effect the same as in (1), but less marked. In both solutions isolated sphincters pulsated for several minutes. 3) ® NaCl. Cloacal ends with the radiating muscles cut pulsated slowly, but the sphincter did not open at all. 4) 100 cc. & NaCl + 2 cc. 3 NH.OH. Slow pulsations, with in- complete closure of the anus. In none of these tests did the pulsation last for more than a few minutes. These experiments are not, of course, decisive as regards an answer to the question of the relation of Ca to pulsation. Continuous stimulation resulting from the action of an exciting solution upon ectodermal sense organs, or upon muscles, might, in connection with the refractory period, produce the same result. The integumentary nerve-net does not play a necessary part in the transmission of the wave of pulsation, since cloacal pieces having the integument cut in various ways (figs. 17 and 18) pulsated with complete codrdination after some preliminary re- adjustment following the disturbances of the operation. To give an example: Experiment 84.2. July 22, 3.44 p.m. a cloacal piece, pulsating in sea water at the rate of 105 seconds for 10 pulsations, was cut as shown in figure 18, so that two separate rings of the integument were each connected with corresponding parts of the cloaca by the radiating muscles. 3.52 Posterior part pulsating feebly. 3.57 Both parts beating. Pulsations of normal amplitude; rate, 96.9 seconds for 10 movements. 4.10 103.9 seconds for 10 movements. 4.30 115 seconds for 10 movements. 8.45 Pulsating in irregular fashion. The cloaca itself is therefore able to control the coérdination of the pulsating system. | RHYTHMIC PULSATION Bla That the integumentary part of the apparatus does normally enter into transmission is indicated by tests made under the following conditions: 1) with the skin completely cut through on one side (fig. 19), and 2) with cloacal sphincter cut into lateral halves (fig. 20). In the first case the sphincter and cloaca on the cut side lagged behind the opposite side, both in systole and in diastole. 19 Figs. 19, 20, 21 For explanation see text In the second instance the brim no longer pulsated in coordi- nated manner and ceased to move after a few minutes. If one of the halves was gently pinched, it reacted alone, by the con- striction of its circular muscles; whereas if stimulated more vigorously, or if a nearby point on the skin was stimulated, both halves of the brim contracted, the one nearer to. the irritated place contracting sooner. The effects of stimulation are thus conducted in a radiating fashion, such as would result if an ectodermal nerve-net were operating, and I conclude that such a net is present. 324 W. J. CROZIER The role of the cloaca in pulsation, and especially of its anterior end, could be demonsrated very clearly. Isolated cloacal ends immersed in sea water containing 5%, caffeine presented this condition: During the early stages of the history of a piece in such a solution the anal sphincter pulsated at a normal rate of 10 movements in 70 to 80 seconds, but the cloaca was tightly contracted excepting immediately in the region of the anus (fig. 21). If now in such a preparation the contracted anterior end of the cloaca was exercised, all rhythm very promptly ceased. This points to the conclusion that the stimulus to pulsation originates at the anterior end of the cloaca with its associated radial muscles and is conducted posteriorly along the cloaca even when the cloaca itself is not pulsating. The fact that in certain salt solutions (e.g., NaCl m 5/8) the cloaca continued to beat after the anus had ceased, serves to strengthen this idea. The excised cloaca alone in sea water pulsated, though in a somewhat irregular manner. In recovery from the immediate effects of amputation, the cloaca and anal sphincter, and especially the former, began to pulsate before the body-wall portion of the complex. It is taken for granted in this discussion that the codrdinating mechanism is essentially nervous in character, and this view is supported by the experimental results. It has been suggested, however, that there is a chemical basis of coérdinated pulsation ‘ in the case of the vertebrate intestine; Weiland (’12) claims to have extracted from the mammalian digestive tube a substance which, acting on Auerbach’s plexus, leads to codrdinated rhyth- mic movement. It must be admitted that the analysis of peri- stalsis upon the basis of the tonus idea (ef. Cannon, ’11) does not account for the inception of the stimulus to contraction, and that therefore some further link is needed in the chain of pulsation processes. d) Summary. The course of events in a pulsation cycle may be pictured as beginning with the opening of the anterior end of the cloaca, due to the relaxation of its circular muscles and the contraction of the associated radiating fibers. The stimulus RHYTHMIC PULSATION a2 derived from their contraction is transmitted posteriorly’? by the integumentary nerve-net, while the wave of opening travels posteriorly along the cloaca, until finally the anal sphincter opens. Upon the contraction of the sphincter a wave of constriction travels anteriorly on the cloaca, forcing out the contained water. IV. RELATION OF PULSATION TO TEMPERATURE The large size of Stichopus precluded any attempt to employ the entire animal in a series of temperature experiments, both because there were no large thermostats available and because the thickness of its integument is prejudicial to the rapid estab- lishment of uniform temperature conditions throughout the ani- mal. For these tests therefore I employed Holothuria suri- namensis and H. captiva; they were subjected to temperature changes in beakers of thin glass contained in a heating or cooling bath. Temperatures were measured by an enclosed-scale ther- mometer reading to 0.01°, placed close to the cloacal end of the animal. The instrument was calibrated. The results of one experiment, typical of all others, are given in figure 22. This curve is entirely characteristic of the temperature curves found in connection with many other biological phenomena, in that it is of an exponential character, with a temperature coeffi- cient (12.5°-22.5°) of about 2.4=. In attempting to obtain temperature coefficients of pulsation rate in the case of the amputated cloacal ends of Stichopus, there entered a very considerable time factor. The method of procedure consisted in (1) obtaining records of rhythm in pieces which had recovered from operative shock and were subjected to fairly slow temperature changes (table 6), and (2) in subjecting similar pieces to sudden changes of temperature and estimating rhythm-rate after 5 to 10 minutes had passed and thermal equilibrium had presumably been established (table 7). The 10 Experiments on Stichopus confirm the results obtained with H. surinamensis (Crozier, 15 [?]), to the effect that the responses to sensory stimuli in the pos- terior region of the animal tend to be conducted posteriorly toward the anus, which often reacts before there is visible any local response to the stimulating agent. 326 W. J. CROZIER time factor enters in two ways, for in addition to the slow exhaus- tion which occurs even at the normal temperature and is itself probably affected by temperature, extreme thermal conditions also exercised a characteristic effect of their own which was mani- By a condition of hysteresis I mean that subjecting a holothurian or its,isolated cloacal end to fested in a kind of hysteresis. Stichopus moebii, 31 cms. long; 83 seconds for ten NOTES Open phase prolonged. Parts dissociated. TABLE 6 Experiment 33.2. June 21. pulsations TIME MINUTES TEMPERATURE sxEc./10 Pe DeguaGs. 8.40 Cut off end. 8.46 6 24.5 80.8 9.04 14 25.0 122.0 9.06 16 26.5 118.0 9.07 17 PH AD 105.0 9.08 18 28.7 98 .0 9.10 20 29.4 100.0 9.11 21 29.7 135.0 9.12 22 29.8 143.0 Oats 25 30.0 150.0 9.19 29° 32.0 164.0 9.25 34 34.0 co No movement. TABLE 7 Experiment 39.4. pulsations, at 24.3°. Jiine 24. Stichopus moebii 29 cms. long; 74 seconds for 10 water at different temperatures Fifteen minutes after excision, plunged successively into TEMPERATURE Deg. C. 25.0 14.0 16.0 17.0 OES 21.2 19.6 20.0 15.0 TIME FOR TEN PULSATIONS seconds 94 198 187 186 140 127 144 155 192 RHYTHMIC PULSATION BA & UuNsQAlOrIS. @ < ol tine for 10 Fermperat ure. Fig. 22 Holothuria captiva. Relation of pulsation rate, in the intact animal, to temperature. The numbers on the curve indicate the succession in which the observations were taken. a temperature of, say, 10°, produced an anaesthetised-like con- dition which persisted for a considerable time after its removal to some higher temperature and thermal equilibrium had in all probability been reached. By way of illustration: Experiment 39.2. June 24. Stichopus moebii, 27 cms. long; 78 sec- onds for 10 pulsations at 25.0°; slowly cooled. TIME eee: TEMPERATURE icicle eet ane tala NOTES P.M. Deg. C. seconds 3.25 0 25.0 64 Six minutes after amputation 380 8 25.0 88 3.37 12 22.0 109 3.45 20 12.0 215 4.00 35 10.0 fos) Closed; no pulsation 4.20 55 17.0 co Beginning to open slightly 4.30 65 18.2 oe) Opening slowly 4.45 80 18.5 216 Not opening completely 4.50 85 18.9 188 Fuller amplitude Experiment 34.1 June 21. Stichopus moebii, 29 cms. long; 76 sec- onds for 10 pulsations at 25.0°; cloaca amputated; during the following 18 minutes, slowly heated up to 30°; held at 30° for 10 minutes, with the following results. 328 W. J. CROZIER TIME REQUIRED FOR TEN ELAPSED TIME PULSATIONS NOTES minutes s conds 2 135 i) 146 8 198 10 fo) Wide open; no pulsation + ’ Then cooled down slowly (for 19 minutes) to 20.0°; pulsations (slowly revived) required 150 seconds for 10 movements (the usual rate 60 minutes after amputa- tion being 60 to 75 seconds for 10 pulsations). All movements ceased in about 10 minutes more. These experiments were checked by many others of similar | type which gave essentially the same result. The main object of these tests being to ascertain the magni- tude of complications possibly introduced into the experiments by temperature effects, there is given in figure 23 some of the curves derived from different tests under several of the possible conditions of temperature change,.and in fgure 24 are collected the data from all the temperature experiments with isolated cloacae. . The discussion of these results with regard to the significance of the temperature coefficient would be unprofitable," since (in view of the source of complication above noted) decisive evidence could be obtained only by the study of isothermal exhaustion curves. But it may be pointed out that the hysteresis effect to which attention has been called indicates that some physical alteration of the substance of the pulsating tissues has occurred at the temperatures which produce this hysteresis effect. It is difficult to believe that this influence is discontinuous, and, as Adrian (’14) has most recently pointed out, the van’t Hoff equa- tion holds only in homogeneous systems in which no alteration of the components is induced by temperature changes. In other words, there is a time-factor to be considered when working with temperatures removed from the normal. It is the neglect of this 4 Pitter (14) has recently given an exhaustive treatment of the temperature- variability curve from the standpoint of the linked reactions occurring in living systems. RHYTHMIC PULSATION 329 factor which renders valueless for exact purposes much of the experimental evidence concerning the influence of temperature upon marine animals (e.g., as in the paper of Mayer, ’14 4); it is Simply incorrect to say that an animal dies at such and such a temperature, for it dies at T° after being heated thereto at a certain rate and kept at T° for a certain length of time. 200 /80 7ime for /0 Puls aliorrs “ SCCONAS. 15° 20° 25° 350° 35° C. 23 Temperature. The curves (figs. 23 and 24) show that there is a rather sharp maximum of pulsation rate in the neighborhood of 26°. No pulsation was apparent below 12.5°, nor above 32.0°. These limiting temperatures hold for the rates of heat change, ete., used in these experiments; there is no more a definite upper THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 3 330 W. J. CROZIER limit to the physiological temperature scale than there is a lower one (Cameron and Brownlee, 713). On either side of the mini- mum in ‘‘time for 10 pulsations” the curves are rather steep and of exponential character, though this is obscured in the collected line for 10 P/SAlLONS ~ SECOMAS. 24 Zermpera lure. data (fig. 24). The range of laboratory temperatures, obtained from temperature measurements in connection with each experi- ment, was from 24° to 27°, hence very little, if any, complica- tion due to a temperature effect entered into the subsequent experiments. RHYTHMIC PULSATION 331 It is interesting that the temperature of maximum pulsation rate should be so near the temperature, at this time of year, of the sea water in which the animals were living; this was 24° to 27° in the day time. V. ACTION OF DEPRESSING AGENTS a) Anaesthetics. The effects of typical anaesthetics upon rhythm in the isolated cloaca of Stichopus are illustrated by the following: Chloroform Experiment 46.1. July 29. Stichopus moebii, 21.5 ems. long; before excision of cloaca, 61 seconds for 10 pulsations. ae cee ce xores = minutes seconds 3 Began to pulsate in sea water 11 56 15 Transferred? to sea water half-saturated with CHCl, 20 113 21 ; 120 22 Pulsation stopped © 24 Washed, and put in normal sea water foo) 40 Recovering amplitude 105 57 Amplitude fully recovered 130 90 Normal | 145 > Previous tests had shown that transferring the pieces from one solution to another had no effect on pulsation rate, even though during handling the sphincter was closed. Ether, ethyl alcohol and ethyl carbamate gave similar results, i.e., the depression of rhythm was, within limits, reversible. Magnesium sulphate to the extent of 0.25 M added to sea water stopped pulsation in about 10 to 12 minutes. Urethane and chloretone were compared quantitatively, with the result that when freshly isolated cloacal pieces were immersed in xt chlo- retone in sea water they ceased to pulsate in less than 3 minutes, whereas in x99 urethane pulsation lasted for as much as 80 minutes. ave W. J. CROZIER In all cases of anaesthetisation the quiescent condition showed the anal sphincter in diastole and the cloaca itself more or less closed, that is, with the radiating muscles relaxed. Traces of anaesthetics commonly induce a temporary acceler- ation in the rate of rhythmic movements. This is also the case with the cloaca of Stichopus (fig. 25). The effect on pulsation rate was less evident than the concomitant increase in ampli- tude and vigor of contraction. The effect lasted only a few minutes. TABLE 8 Volume of solution in each case = 250 ce. DURATION, IN MINUTES, OF : ; GOSSEIISS NOTES: SUBSTANCE CONDITION OF ANAL SPHINCTER TRATION AFTER STOPPAGE | Pulsation | Irritability @urare?.../)0228a... + | 11000 130 170 Circular body muscles con- tracted, longitudinal ones relaxed; = open Caffeine. opener. snot M/200 25 160 Cloaca contracted; sphinc- ter open INiicGObiner ee ease. cae M/200 8 12 Contracted Cocaine hydrochloride} M/200 90 100 All muscles relaxed Atropine sulphate..... M/200 28 60 Wide open; radiating mus- cles strongly contracted Morphine sulphate... .| M/200 140 165 Relaxed Sea water control.....|......... 540+ | 1100+ | Sphincter + open, all mus- cles relaxed b) Alkaloids. The influence of depressing alkaloids was also studied; in each instance the history of five individual isolated cloacal ends being followed in detail and the general result checked by less detailed observations upon other pieces. The results with regard to pulsation and irritability to mechanical stimulation are summarised in table 8. Representative indi- vidual curves of exhaustion in these solutions are plotted in figure 26. The powerful effects of nicotine and atropine are comparable to their influence on many other types of smooth muscle. RHYTHMIC PULSATION 333 0120 > SS 5 00 g S < 90 S 'S 60 Se 4 S 70 lead = f t-adailion of lrace N eo SeHer & 30°40 50 60 %O 80 S09 100 Ilo Time after amputation ~ min 10 290 PoE Sec 260 Fe wa ? Van ag il y BY Sie ! Ge 1 = | oe 220 | 1 — LZ. Atri pine suleha re, Yeo I. Micotine , t/200 II Cocaine hydrochloride , (YZ00 IE. /70/ pine Stifphare, (4/200. EK .Curere, /:/000. Jime for 1o pulsations x lay PI NOM TZOVISOMESO! 50'/Gi0! 470: ¢ BOM oo MoO MD 120 130 Min. HiME DP MNIN C651 O72. 334 W. J. CROZIER VI. RELATION OF PULSATION TO OXYGEN AND METABOLIC PRODUCTS a) Oxygen. It has previously been shown that holothurians would live for relatively long periods in sealed jars of boiled sea water (Crozier 715). This was in conformity with the observa- tions of Moore and his colaborators (’12, ’14), showing the small amounts of dissolved oxygen used up in the activity of some marine animals, including echinoderms. The isolated cloacal ends of Stichopus continued to pulsate for over 700 minutes in jars of sea water which had been boiled, sealed and subsequently cooled to room temperature out of contact with the air. In pre- paring this water the evaporation was very slight, but possibly some faint increase in alkalinity was induced. The course of a typical experiment is illustrated in figure 27 (6), plotted from the record of experiment 78.3, with which may be compared the control (A), begun at the same time. Experiment 78.3. July 10-11. Stichopus moebii, 20 ems. long; 60 seconds for 10 pulsations at 23.5°; volume of boiled sea water, 550 ec. MINUTES ELAPSED Sa See NOTES seconds 0 Cloacal end amputated and put in sealed jar of boiled sea water 10 78 20 78 35 78 50 91 65 88 95 92 350 109 Amplitude decreasing 740 170 Pulsations feeble and irregular 752 Taken out of jar and transferred to fresh sea water. Not pulsating 764 150 Closure nearly complete but more or less ; irregular 778 180 Irregular movements; no wide-open phase 800 Twitching irregularly; barely alive The dependence of pulsation upon oxidations is probably indi- cated, however, by the high toxicity of KCN, as, for example, in RHYTHMIC PULSATION 335 Experiment 71.2. July 25. Stichopus moebti 27 ems. long; cloacal end amputated and immersed in 200 ce. sea water + 5 drops *& KCN. TIME FOR | NOTES TIME OF IMMERSION TEN PULSATIONS minutes seconds 5 80 Not closing completely 10 147 Closed phase very brief 17 84 Full amplitude 23 111 Closure not complete 38 foo) Irregular movements 53 Dead eo 300 - Ue Jer len pulsations — SECON LS 0 100 200 300 400 500 600 700 800 900 1000 100 1200 1300 1400 Hime CL2)SEA ~ NLU CES. Fig. 27 A = control, in sea water. B = in sealed jar of boiled sea water. C = sea water (300 cc.) + 5 per cent urea (5 cc.). Note: At the point marked |, ‘B’ had ceased to pulsate, and was transferred to normal sea water, in which it temporarily revived. 336 W. J. CROZIER This effect was uniformly obtained in other experiments of similar nature. b) Carbon dioxide. Carbon dioxide was a powerful agent in suppressing pulsations. Tests were made by adding to sea water small volumes of rain water charged with CO,. The car- bonated water was slightly acid aside from its H2COs. Experiment 79.2. July 11. Three freshly amputated cloacal ends were immersed each in 200 ec. of sea water to which 2 ec. of ‘carbonated water’ had been added. They did not pulsate in this mixture. After ten minutes immersion the pieces were transferred to normal sea water. Pulsations revived in 5 to 7 minutes. Concordant results were obtained by adding CO,.-water to solu- tions containing pulsating pieces, the pulsations being rapidly reduced in amplitude and usually stopped within a few minutes when from 1 to 2 per cent of the ‘carbonated water’ had been added. | c) Urea. Urea likewise had a depressing effect upon pulsa- tion. Fosse (13) identified urea in echinoderms and their excre- tory products. Experiment 38.2. June 24. Stichopus moebii, 27 ems. long; 68 sec- onds for 10 pulsations at 25.1°; cloacal end amputated and placed in 300 cc. sea water + 5 cc. * urea. IMMERSION FOR TEN PULSATIONS NOTES minutes seconds 153 151 Very long open phase 184 170 Irregular amplitude 214 103 Somewhat improved 242 164 Open phase long 360 © Sphincter half open; reacts only slowly to tactile irritation This particular experiment is plotted at C in figure 27; it was checked by four other experiments, which yielded the same result. d) Light. With Holothuria surinamensis and H. captiva it was previously found (Crozier, 714, ’15) that light exerted a distinctly toxic influence on the animals. It was therefore ex- pected that sunlight would affect in some way the pulsation rate RHYTHMIC PULSATION 337 of the isolated cloacal ends. Stichopus is much less sensitive to photic irritation than is Holothuria, and no influence of light upon either rate or amplitude of pulsation could be detected. In Holothuria it has been shown with some degree of probability that the green fluorescent integumentary pigment acts as a photo- sensitizer (Crozier, 714); hence a photic effect upon cloacal pulsa- tion would be much more probable in this case. Isolated cloacal extremities of Stichopus which had ceased to pulsate regularly in sea water, though otherwise in apparently good condition, were placed in moderately bright sunlight. This did not stimulate to pulsations. When-such pieces were shaded, the sphincter reacted by closing fairly tightly; the reaction time was about 1.2 seconds at 25.0°. Pieces which were pulsating slowly (130 seconds + for 10 movements) gave the same reac- tion, the sphincter closing promptly when the shadow was cast during the open phase, but failing to react during a period of closure. Light did not accelerate the rate of movement in nor- mally pulsating cloacal ends. The isolated cloacal extremity of H. surinamensis, however, gave the following result: Experiment 53.2. Seven specimens of Holothuria surinamensis had the posterior ends (1.5 ems.+) removed. After remaining about 30 minutes in diffuse daylight, their pulsation rates were observed (column a). Four of the pieces were then placed in bright sunlight (b’), the other three remaining in the diffuse daylight (b). After 15 minutes the pulsation rates were again determined (0, b’): Time for ten pulsations, seconds IN DIFFUSE LIGHT AFTER FIFTEEN MINUTES IN BRIGHT SUNLIGHT a b b’ Notes 80 137 Very irregular 90 54 Brim only beating 84 To) 130 fo) 70 72 81 80 338 W. J. CROZIER The cloacae exposed to light did not recover on return to the shade. Light exerted its typical effect upon the excised cloacal ends, and at a rate which is comparable to that at which it causes the disappearance of sensitivity to shading in the intact animal (Crozier 714, 715). e) Summary. Lack of oxygen would appear to be much less powerful as a depressant of rhythmic activity than the presence of urea,’ COs, and other typical metabolic products (the effects of acids will be discussed subsequently ina separate section), though the suppression of oxidations by KCN is rapidly effective in stopping movement. In those cases in which light produces toxic modifications in the animal (H. surinamensis, H. captiva), this agent also causes rapid cessation of rhythm. VII. INFLUENCE OF OSMOTIC PRESSURE a) Dilutions of sea water. According to Henri et Lalou (’03), the membranes of Stichopus regalis which are exposed to contact with water are almost perfectly semi-permeable, the salt con- centration in the ambulacral!* and perivisceral fluids being nor- mally slightly less than that in the sea water. They found that S. regalis readily accommodated itself to moderate dilutions of the sea water in which it was placed. In the experiments about to be described it was found that diluted and concentrated sea water exerted characteristic effects upon cloacal pulsation. It should be remembered that in these experiments with the isolated cloacal ends the medium had access to the internal surfaces of the body wall and cloaca, and further that the cut surfaces of the integument and cloaca at the exposed end were freely open to its action. The pulsation-rate of amputated cloacal ends was observed in sea water at the following volume percentage concentrations: 0, 25,.50, 75, 90, 95, 100,110, 120.14 The history of typical indi- ’ According to Henri et Lalou (’03) the membranes of Stichopus regalis are impermeable to urea, but in my experiments a cut surface was exposed to the action of the fluid. Furthermore, the remnant of the body cavity was freely open. : 13 From the Polian vesicle. 14 The dilutions from 100 per cent were made by the addition of rainwater; the 110 per cent and 120 per cent solutions were made by evaporation of sea water to the required volume. Normal Bermuda sea water contains about 36.8 p.p.m. salt (Mark, 713). RHYTHMIC PULSATION 339 vidual cases is plotted in figure 28. The tests indicated that above 50 per cent concentration, mixtures of sea water with rain water tended to preserve about the normal pulsation-rate, but that in these mixtures the duration of pulsation was less than that of the control (100 per cent); below 50 per cent sea water Lae So | 280 3 Lf fects f several concerurations re} SECQWALES- 260 Lolarie of SOl¢l 270/72 = 500 ee > , 0 75 95 a / Ze 100 a / / Ke 4 220 / Zp oe S ve 120 (es § Zz yet Se ¥ 90 « ? ee UF > 2 I Uyee an eas e a ye .e G, 5 eet ; a NS 160 : ye S ; Q Ve ve *——°/20 % S40 Yoyo o——» /00 % x —- 95% me 120 X——»~ 9O% Q $— 4 75 % Nel Of. H—-H 50% Se C—O 25% 80 } y is) 50 O00 200 300 400 500 28 Time of immersion ~Min. the pulsation-rate and duration fell off rapidly. Above 100 per cent concentration the duration of pulsation was also curtailed. The general nature of these effects did not differ whether (a) the cloacal pieces were immersed in the solution immediately after amputation, or (b) after pulsation in the amputated end had been allowed to start in sea water; procedure ‘a’ was fol- lowed in the cases recorded.!® In these experiments the pieces 15 With this method of immersion, the pieces were in the condition shown at A in figure 12, and the solution therefore had free access to the interior parts of the cloacal end, 340 W. J. CROZIER to be immersed in a given volume of diluted or evaporated sea water were first washed (30 to 60 seconds) in a stream of sea water of the corresponding concentration. The volume of solu- tion containing a single isolated cloacal end was in each case 300 ce. 500 Duration ofirriabiil, Duration oulsaliors Oo 25 50 75 100 125 Terceniage Ailution of sea@waler Fig. 29 Based on experiments with 45 specimens of Stichopus Five individual pieces were carefully studied at each concen- tration of sea water and their behavior checked by less detailed observations upon a number of other examples. The average times for the continuation of pulsation and of irritability to mechanical stimulation under the conditions just specified (at 25° to 27°) are given graphically in figure 29. As regards the condition of the cloacal sphincter after the cessation of pulsa- tion, this difference was noted between the action of rain water e RHYTHMIC PULSATION 341 and the various dilutions of sea water, namely that in the former case the sphincters remained wide open, whereas in the latter instances the sphincters (as in ‘normal’ exhaustion in 100 per cent sea water) were more or less contracted, though with the rain-water effect it was not a matter of the contraction of the anal dilators, but rather of the more complete relaxation of the circular constrictors. With rain water it was found that the brownish skin pigment rapidly made its appearance in the water surrounding a beating cloacal end even before pulsation ceased. This effect became apparent after immersion of about 5 minutes. It was, under certain circumstances, found with other concentrations of sea water, but (as will be shown subsequently) it could be inhibited by the presence of non-electrolytes. This reaction may indicate either that the permeability of the superficial cells had been very Highly modified, or that the cells surrendering their coloring matter were dead; inasmuch as the pigment loss did not occur more at the region of the cut end of the pieces than at any other place, and could be inhibited by the addition of sea water, and further inasmuch as the region of the cut end did not give off any visible amount of pigment into normal sea water, it might rea- sonably be held that the latter alternative is not necessarily the correct one. It is possible that at concentrations much removed from that of normal sea water the cells of Stichopus became rather highly permeable for salts while still alive. This possi- bility was suggested by the very evident amounts of chlorine found in rain water in which pulsating pieces had lain. If this permeability for salts could be proved over the lower range of concentrations here dealt with, the form of the lower parts of the curves in figure 29 might readily be accounted for; between 0 per cent and 95 per cént these graphs might result if the critical internal conditions (salt concentrations?), beyond which no spontaneous pulsation is possible and irritability ceases, were approached (1) by the intake of water and at the same time (2) by the exit of salts. It is stated by Mayer (’14 » p. 40) that the rate of nerve con- duction in operated Cassiopea is accelerated by slight dilution 342 W. J. CROZIER of the sea water (down to 80 per cent of the original concentra- tion), whereas the rate of pulsation in intact medusae (p. 28) is not increased in this way; but in his tables no records are given of the pulsations of entire medusae in sea water of the critical concentrations (95 per cent, 90 per cent). It might appear from the records in my figure 28 that there is some tendency on the part of diluted sea water to preserve in the excised cloacal ends of Stichopus a higher rate of pulsation during the later stages of their history, than is the case with the ‘control’ pieces in 100 per cent sea water. But, as a matter of fact, this is not so, since there was enough variation in these records and in the controls to prohibit a conclusion of this sort. The matter was studied more carefully with 90 per cent and 95 per cent sea water; no increase in pulsation frequency was discoverable upon diluting to these concentrations sea water in which amputated cloacal ends were pulsating. - b) Non-electrolyte solutions. It was found by Loeb (’00 ®) . that in solutions of non-conductors the isolated center of Gonio- nemus did not exhibit rhythmic contractions. This has in general been the experience of others employing a variety of pulsating mechanisms, namely, that in a medium free from salts -but with its osmotic pressure made up to normal by dissolved non-electrolytes, rhythmic contractions are either not initiated or do not continue for any length of time. According to Mayer (14) this physiological inefficiency, of sugars and the like, is also apparent in nerve-conduction in Cassiopea. The results of experiments in which the isolated cloacal end of Stichopus was placed in solutions of non-electrolytes may be given very briefly. Pieces immersed in solutions of sucrose, lactose, maltose, or glycerin theoretically isosmotic with sea water (i.e., 0.9 to 1.0 M) did not pulsate for more than about 15 minutes, the actual time for the duration of pulsation varying from 2 to 25 minutes. Irritability to mechanical stimulation disappeared after about half an hour. Such pieces did not recover on return to sea water. In these tests the amputated cloacal end was allowed to begin its pulsation in sea water and to continue there for about 10 minutes; the pieces were then RHYTHMIC PULSATION 343 washed (inside and out) with a stream of fresh water! before being put in the non-electrolyte solution. No evidence was had of an increase in muscular tone in sugar solutions, though pulsa- tion usually ceased with the sphincter in the contracted condi- tion. I had previously found (Crozier 715) that glycerin and maltose would stimulate the skin of Holothuria surinamensis in a sensory way; this effect probably had nothing to do with the action of sugar solutions upon pulsation, since the glycerin and maltose solutions did not behave differently from those of sucrose, etc., which had been found not to stimulate. Pulsations, where they did occur, were always of lower frequency in sugar solu- tions than in the controls in sea water. VIII. ION EFFECTS , 2) Single salts. The relation of the salts of sea water to cloacal pulsation and general irritability was studied by observing the action of single electrolytes at equivalent concentrations (1.e., isotonic with sea water), and the effects of various combinations of these substances in the proportions at which they oceur in sea water. The method of procedure consisted in placing in the desired salt solution freshly amputated cloacal ends of which the pulsation rate had been determined previous to removal from the animal. Before immersion the pieces were rapidly washed by a stream of rain water from a wash bottle. Check experi- ments showed that washing the outside and inside of the cloacal end in this way did not have any effect upon the rate or duration of subsequent pulsation of pieces reimmersed in sea water. Since the excised pieces were immersed while in the condition shown in A, figure 12, that is, before the inturning of the cut edges had closed the cavity containing the radiating muscle strands, the solution had abundant opportunity to gain access to the ele- ments concerned in pulsation. Care was taken that the cavity became filled with the solution surrounding the immersed piece and that no air remained in it. Since the pumping movements maintained a current of fluid through the cloaca, the solution 16 Check experiments showed that the washing had no effect on pulsation, at least on its duration. 344 W. J. CROZIER was by this means (as well as by the movements of the body wall) efficiently stirred as long as the piece continued to pulsate; local diffusion changes were thus avoided. One disadvantage of this whole method les in the fact that nothing could be done to prevent the action of the solutions upon the cut surfaces along the plane of amputation. The magnitude of this effect could not be estimated. 26 2 / —— /YaCl, 48/1. *—«L/Cl;, 48/7. o------0CAC/> FB. === 1GCh; Val. Tine for 10 PUlMSALIONS ~— SCCOP AS. O° 10 20 3O 40 50 60 770 80 90 x Time Of 1NEL S102 — TU UTeS, Individual records illustrating the action of the single salt solutions in which cloacal pulsation occurred are plotted in figure 30. Table 9 contains a summary of the results of these experi- ments. With NH,Cl and KCl, although the anal sphincters were tightly closed, some irregular pulsation of the body wall usually was evident for about the length of time indicated in table 9. The order of the disappearance of tactile irritability parallels that of the stoppage of pulsation in these salt solutions. The completeness of the correspondence argues for a close similarity in the mechanisms of stimulation by internally and externally generated agencies. RHYTHMIC PULSATION 345 TABLE 9 Effects of single salts DURATION OF SALT M. CONCENTRATION Pulsation Irritability minutes minutes NBO]. USSR esas ene: 5/8 75 105 LGiK CLARE ee mle eere cee 5/8 12 15 INTEC © li" cr aceon cee 5/8 (10) 0 12 Oe eS ck AAN hae: 5/8 (3) 0 10 CaS ea: va ee ee 3/8 5 7 Ole, aeons anes ee 3/8 20 30 RIGSOMRA yo. cashes 9/10 17 24 Averages from five series of tests. The figures in parenthesis opposite NH,Cl and KCl mean that, while the sphincter of cloacal pieces immersed in these solu- tions did not pulsate, the body part did so, irregularly, for the number of min- utes noted. TABLE 10 Summarizing the results of tests made upon five isolated cloacal ends of Stichopus moebit with each of the salts indicated SALT NaCl LiCl NH:Cl KCl Concentr., molecular......} 5/8 | 5/16} 5/8 | 5/16} 5/8 | 5/16 | 5/8 |.5/16 (10) (3) Duration of pulsation.....) 75 | 50 12 35 0 0 0 0 Duration of irritability...| 105 | 80 15 38 12 41 10 15 Time—Minutes. For the chlorides of the alkali metals and radical (table 10), the order Nate hr > NH, > Kk was obtained with reference to their ability to preserve pulsa- tion and irritability. The significance of this series, in terms of the aggregation state of protoplasmic materials, has been treated in a comprehensive manner by Hoéber (’14, pp. 471 et seq.). A beautiful demonstration of the action of the cations of neutral salts in this series has recently been given by Spaeth (13) for the chromatophores of Fundulus. In solutions of NaCl and LiCl the pieces came to rest with the anal sphincter in an expanded condition but with the radiating cloacal muscles con- THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 3 346 W. J. CROZIER tracted. This was unlike the condition in solutions of NH,Cl and KCl, where both cloaca and anal sphincter were tightly con- tracted. The absence of any pulsation in 5/8 M NH,Cl and in 5/8 M KCI is partly a secondary result, since solutions of these salts are powerful sensory stimulants for the skin of holothurians (Crozier, 715), which produced contractions from which the cloacal ends did not recover before the toxic action of the solu- tions led to death. That the single electrolytes LiCl, NH,Cl, and KCl are toxic, is also indicated by the results of tests in which several concen- trations of these substances were compared (table 10). These experiments showed that with LiCl the duration of rhythmic movement and of irritability was greater at 5/16 m than at 5/8 M concentration, and that with NH,Cl and KCl] the duration of irri- tability was also greater at the lower concentration, even though in the case of the more dilute solutions an osmotic factor (see previous section) was also working to produce death. NaCl is thus the only single constituent of sea water which will maintain pulsation and irritability for any considerable length of time, and in this respect it is only partially imitated by LiCl —a, fact further proven by the physiological incompleteness of a van’t Hoff solution made up with LiCl in place of NaCl, in which the pulsation of a cloacal end endured for about 45 minutes. The comparative effects of MgCl, and MgSO, were of particu- lar interest. Spaeth (138, p. 553) found that MgSO, exerted a more powerful action upon the melanophores of Fundulus than did an iso-ionic MgCl, solution. Table 11 contains the data from experiments in which MgCl, and MgSO, were compared at several concentrations by means of their action on cloacal pulsation in Stichopus. The result of these tests was that MgSO, appeared to have a distinctly higher anaesthetic power than MgCl. b) Mixed salts. As the basis for the experiments made with mixed salts there was employed a van’t Hoff solution of the composition: 5/8 m [100 NaCl + 7.8 M,Cl. + 3.8 MgSO, + 2.2 KCl + 2.5 CaCl] (ef. Mayer, ’11, 714°). The solutions were made up in rain water. The averages obtained for the time of RHYTHMIC PULSATION 347 TABLE 11 SALT MgCl: MgSOs Concentrated, molecular................ 5/16 3/8 5/16 9/10 Duration of pulsation, minutes..........|, 14 20 14 17 Duration of irritability, minutes......... 37 30 26 24 duration (a) of pulsation and (b) of irritability to mechanical stimulation, in the several salt mixtures studied, are given in table 11. By way of comparison, data taken from Bethe’s account of the effects of sea water constituents upon pulsation in medusae (Bethe, ’08) are given in the last column of table 12. The cloacal ends of holothurians are in some ways a better indicator for use in studies of this nature than are medusae, since the former pulsate continuously and not, as the jelly fishes do, in more or less interrupted groups of movements. TABLE 12 Summarizing the results of tests with salt mixtures. For details, see text DURATION OF SOLUTION R. Pulsation | Irritability minutes minutes Th dl ane Gils Wee) Recent waste cee 2 ree 75 100 27 Panic leata@ a @ lows weer | MU eee ul 60 280 95 SEleiNa Cle EKG ese booeee. - . 5 WAMe eec a heos as 50 90 47 Ae NaCl == KCli-— Ca@ises.....veeee old: ke 120 260 150* Op PEN aC le=l=s Mio Clow ayant sees. Gey Ae foes Sota aia 42 60 1 6 | NaCl + KCl + CaCl. + MgCle............ 104 300+ (alvAnvinELOtiesSOluGlone.... aeons as 360 480 1200+ Sal eeeemnacdevallica lines.) eet ce 600 ORI SeamWcule ane red cus << PRET StDe Ae cde tie', ': 540 (10 days) In column ‘R’ are given the corresponding data for the pulsation of the medusa Rhizostoma (Bethe, ’08, p. 573). * This solution contained CaCQs. A notable peculiarity of the pulsation curves of pieces immersed in NaCl solution (compare previous section, and figure 30) is the irregularity of the pulsation rate. During the early history of pieces in NaCl, the pulsation rate was also very low. By reference to figure 31 it will be seen that, in the experiments CROZIER J. Wis 348 ‘SATII —~ WOlSIMUIU) 4272[0 FUL O9€ Ore O2F OOF O82 O9c Ore OZe OOS O8!) O99! OF! 02! OO! 08 09 hd 0? Od 40 9b /+ I DOL/+/ 4 YOO +) OM OW +I + 9990 +/QON (D/O / 0) =a YO +/ DOKI SLPSOIBLS — SUOsY Dsy77A O/ LOL EY RHYTHMIC PULSATION 349 plotted, only in the presence of MgSO, is the sodium effect com- pletely abolished. This was quite uniformly observed. MgS0O,, in combination with NaCl, KCl, and CaCh, led to pulsations of a normal character, whereas MgCl, did not do so, not completely at least, even in mixtures where the quantity of Mg normally derived from MgSO, + MgCl. was ‘‘made up” by a calculated amount of MgCl. The deficiency of solutions lacking magne- sium lay in the fact that, as Loeb (’06) observed with the medusa Polyorchis, in these mixtures the sphincters tended to contract permanently, in more or less tetanic fashion. MgCl, and espe- cially MgSO,, and more particularly both together, acting with the other salts of sea water, led to normal relaxations after each systole, and thus tended to preserve a normal rate and duration of pulsation. Solutions containing NaCl and MgCl, or MgSO,, or both, did not maintain rhythmic movements as long as con- trols in NaCl; CaCl. and KCl were necessary for the complete balance of the solution, as regards pulsation. This applies to artificial salt mixtures. They were practi- cally neutral in reaction. An attempt was made to approach the problem from the other side, by precipitating the SO,” out of sea water by BaCl.. Solutions prepared in this way preserved pulsa- tion for not quite so long (about 4 hours) as did the neutral van’t Hoff mixture. The effect noted in the absence, of MgSO, in artificial salt combinations may therefore have been due, in part, to the particular C,. of those mixtures (cf. Loeb, 710). It is unnecessary to go into the analysis of many of the results obtained with salt mixtures, since they are so similar to those which have been found for other pulsating structures (Robertson, 10). Certain findings with reference to the significance of Ca- in pulsation may however be mentioned. From table 12 it will be seen that the antagonism between Na and Ca~ was a very imperfect one as regards the duration of pulsation, though good as determined by the preservation of irritability. Between Na and K’ there was no antagonism. Na: + K’ + Ca-~ formed a nearly completely balanced solution, aside from the K* and Ca~ tetanising effects, which interfered with normal diastolic relaxa- tion. The preservation of irritability was not very much less 350 W. J. CROZIER efficient in NaCl + KCl + CaCl, than in the same + MgCl. The depressant action of salts which diminish the active number of Ca ions has frequently been assumed to be due to their effect on the Ca: concentration, but the results of Salant and Hecht (715) indicate that this conclusion is not well founded. Experi- ments of the following type showed that these well known rela- tions obtained also with the cloaca of Stichopus: Experiment 45.2. June 29. Five preparations from Stichopus moe- bii were allowed to begin pulsating in sea water. They were then trans- ferred to a 0.1 per cent solution of oxalic acid in sea water. Pulsations of the sphincter ceased in 3 minutes, though the cloaca and body wall contracted rhythmically for about 4 minutes longer. The sphincter remained open; it contracted once in response to each light touch, for 5 minutes more. Experiment 75.2. July 20. Three cloacal preparations immersed in a solution of the following proportional composition: 100 cc. sea water + 3 ec. * tartaric acid. Pulsation lasted 15 to 20 minutes. Ceased with the cloaca and sphincter in the open phase. Experiment 74.2. July 20. Five cloacal preparations immersed in a solution of the following composition: 100 cc. sea water+ 5 ec. mM 5/8 sodium citrate. Pulsations continued for a little over two hours, ceas- ing in the open phase. The movements, while they lasted, were of almost abnormal amplitude and vigor. Cessation found the sphincter in the open phase.’ Experiment 74.1. August 3. Three cloacal preparations immersed in a solution of the composition 100 ce. sea water + 2 cc. M 5/8 CaCl. Pulsations continued for about 5 hours, at the end of which time the sphincters were all tightly closed, though they responded to a touch by opening and closing once. Further increase in the amount of CaCl. in sea water stopped pulsa- tion (with the characteristic calcium tetanus) in a shorter time. The efficiency of salt mixtures in preserving pulsation was notably improved by making them alkaline, especially with NH;,OH. NH,Cl was found to be very toxic, but solutions of ‘the composition: 100 cc. sea water + 2 cc.4, NH, OH preserved pulsation for about 20 hours, when the volume of solution for each piece was 500 ce. NaOH or KOH substituted for the ‘7 In this solution the tube feet moved incessantly, in sharp contrast to their usual quiescent condition in sea water. RHYTHMIC PULSATION 351 NH,OH did this for about 15 hours. The amplitude of pulsa- tion in these mixtures was greater than in controls run in sea water. It is suggested that alkalis favored the continuance of rhythm by assisting in ionic exchanges at the surface of the con- tractile elements; probably this was accomplished in part in a secondary way, namely, by the neutralization of acid metabolic products. c) Hydrogen-ion concentration. The hydrogen-ion concentra- tion of sea water is given by various authors as lying between 0.5 X 10-® and 1.5 X 10-§ (Hober, 714, p. 195). The sea water used in experiments with Stichopus was faintly alkaline to neu- tral red and neutral to tropéolin ‘‘000.”’ Its C,. was thus about 10-8. Increasing the Cu. by the addition of acid (HCl)!8 led to the rapid cessation of pulsation movements. Experiment 95.2. July 24. Sea water to which 4 HCl had been added until just neutral to Congo red (i.e., Cy. = ca. 10-*°) was tested with five fresh cloacal preparations. Pulsations of an irregular char- acter were manifested for about 15 minutes in two of these tests. The anal sphincter did not pulsate, but remained about one-third closed. In transferring the cloacal preparations to a desired solution, they were lifted out of sea water; they thereupon ceased beat- ing (contracting as the result of mechanical stimulation)!® and if returned to normal sea water about two minutes usually elapsed before pulsation began again. It was sought to utilize this fact and control the hydrogen-ion effect more closely (as was done by Dale and Thacker, 714, in analyzing the automaticity of different regions of the frog heart) by discovering the concen- tration which would just permit pulsation to begin and continue for about a minute. For the sphincter this concentration was attained by mixing equal volumes of sea water and sea water made just barely acid to Congo red: the mixture had therefore a C,. of ca. 10-*°. This limit is very close to that found by Bethe (09, p. 261) for the pulsations of medusae. 18 The effect of certain organic acids was also studied, but not with sufficient completeness to warrant discussion in this place. 1° No pulsations were ever observed out of sea water. 352 W. J. CROZIER Since it was not practicable to employ a perfusion method, experiments dealing with the action of the C,,. in solutions which permitted the continuance of pulsation for some time were not attempted, as it was found that such solutions were modified in the direction of neutrality by contact with the tissue for about half an hour. . It may be of interest to note that, although the function of cloacal pumping is in part respiratory, no increase in pulsation rate was induced by an increase in C,,.; cloacal rhythm in Sticho- . pus thus resembles the (partly) respiratory movements of the arms of barnacles (Roaf, 712). . IX. SUMMARY The rhythmic pulsation of the cloacal chamber and anal sphincter of Stichopus moebil is dependent upon the continuous generation of stimuli within the cloaca, and particularly at its anterior end. The mechanism whereby the radiating muscles of the cloaca, the circular muscles of the cloaca and anal sphine- ter, the anal dilators, and the muscles of the body wall are brought to act in orderly sequence in the pumping of water into the respiratory trees is likewise locally contained. ‘The aboral ends of Stichopus moebiu, Holothuria surinamensis, H. captiva, and Cucumaria punctata continue to pulsate for many hours after they have been amputated at the level of origin of the respiratory trees. In such amputated parts a complete pulsation movement begins with the opening of the anterior end of the cloaca; a wave of opening runs aborally along the cloaca; the anal sphincter then opens and afterward closes; a progressive constriction of the cloacal chamber begins at the closed anal sphincter and runs forward; at the termination of the pulsation the whole cloaca is closed.2° Normally the pulsation of the cloaca is interrupted by two means: (1) by complete constriction of the anal sphincter, in response to sensory stimulation and during locomotion, and (2) . by holding the cloaca and anal sphincter open, during ‘spouting.’ 20 Doubtless the nervous arrangements for the production of this kind of sequence in movements is similar to that involved in the use of the lanternin the locomotion of Echinus (Gemmill, 712). RHYTHMIC PULSATION 30a In the latter case the edge of the sphincter continues to pulsate. Pulsation continues during defecation. The isolated cloacal ends do not ‘spout;’ they contract in response to mechanical, chemical, and photic (shading) stimuli. The stimulus to spouting in the intact holothurian is probably derived from a condition of ten- sion in the body wall. The radiating muscles and connective-tissue strands of the cloaca are necessary for the performance of pulsation. They appear to act in connection with an integumentary nerve net. The results of experiments upon the cloacal termination of Stichopus are in essential agreement with the data derived from many previous studies of pulsating structures, such as those of medusae, ctenophores, the arthropod heart, and the vertebrate heart and intestine. The rhythm has a temperature coefficient of the order of magnitude of that for chemical processes. The rate of pulsation is related to the size of the animal (in Stichopus, Holothuria surinamensis, H. captiva, and Cucumaria punctata) in such a way as to suggest that the larger animals, which pulsate more slowly, possess relatively less energy than do smaller ones of the same species. Pulsation of the amputated aboral end is readily depressed by urea, carbon dioxide, acids, and KCN; it is resistant to lack of dissolved oxygen. Either dilution or concentration of sea water curtails the duration of pulsation. Rhythmic movements do not continue for more than a few min- utes in non-electrolyte solutions. The relation of pulsation to the salts of sea water is essentially like that in other well known pulsating systems; NaCl + Call. + KCl (in the proportions found in sea water) enables pulsations and irritability to con- tinue longer than with NaCl or with NaCl + CaCh; but mag- nesium, and particularly MgSO, (at least in neutral salt mixtures), must be present to insure normal diastole. Calcium is intimately concerned in contraction. In the series NaCl > LiCl > NH.Cl > KCl (5/8 M) the preservation of pulsation and irritability is successively less. The pulsating mechanism is extremely susceptible to increase in the hydrogen-ion concentration. The addition of NH,OH, or other alkalies, to normal sea water assists in the preservation of pulsation and irritability. 354 W. J. CROZIER X. BIBLIOGRAPHY Aprfan, E. D. 1914 The temperature coefficient of the refractory period in nerve. Jour. Physiol., vol. 48, p. 453-464. ALEXANDROWICZ, J. S. 1913 Zur Kenntnis des sympathischen Nervensystems einiger Wirbellosen. Zeit. f. allg. Physiol., Bd. 14, p. 358-376., Taf. 12-13. Atvarnez, W. C. 1914 Functional variations in contractions of different parts of the small intestine. Amer. Jour. Physiol., vol. 35, p. 177-193. Betrue, A. 1903 Allgemeine Anatomie und Physiologie des Nervensystems. vill + 487 p., 95 Fig., 2 Taf., Leipzig. 1908 Die Bedeutung der Elektrolyten fiir dierhythmischen Bewegung- en der Medusen. Teil I. Die Wirkung der im Seewasser enthaltenen Salze auf die normale Meduse. Arch. ges. Physiol., Bd. 124, p. 541- Dil. 1909 Die Bedeutung der Elektolyten fiir die rhythmischen Bewe- gungen der Medusen. Teil II. Angriffspunkt der Salze, Einfluss der Anionen und Wirkung der OH- und H-ionen. Arch. ges. Physiol., Bde l2Gepa2lo—2 73: Borpas, L. 1899 Etude sur l’anatomie et les functions physiologiques des poumons aquatiques des Holothuries. Ann. Mus. Hist. Nat. Marseille, Zool., Tom. 5, Mém. 3, 15 p., 1 pl. CAMERON, A. T., AND BRowNLEE, T.I. 1913 The effect of low temperatures on cold-blooded animals. Quart. Jour. Exp. Physiol., vol. 7, p. 115-130. Cannon, W. B. 1911 The importance of tonus for the movements of the ali- mentary canal. Arch. Int. Med., vol. 8, p. 417-426. Cannon, W. B., AND Burket, I. R. 1913 The endurance of anemia by nerve cells in the myenteric plexus. Amer. Jour. Physiol., vol. 32, p. 347- 357. Cuitp, C. M. 1913 Studies on the dynamics of morphogenesis, etc. : V. The relation between resistance to depressing agents and rate of metabolism in Planaria dorotocephala and its value as a method of investigation. Jour. Exp. Zodél., vol. 14, p. 153-206. Criark, H.L. 1901 Bermudan Echinoderms. A report on observations and col- lections made in 1899. Proc. Bost. Soc. Nat. Hist., vol. 29, p. 339-345. Crozipnr, W. J. 1914 The orientation of a holothurian by light. Amer. Jour. Physiol., vol. 36, p. 8-20. 1915 The sensory reactions of Holothuria surinamensis. Ludw. Zool. Jahrb., Abt. f. Physiol. Dane, Dorotuy, AND THacker, C. R. A. 1914 Hydrogen-ion concentrations limiting automaticity in different regions of the frog’s heart. Jour. Physiol., vol. 47, p. 493-508. Kpwarps, C. L. 1909 The development of Holothuria floridana Pourtalés with especial reference to the ambulacral appendages. Jour. Morph., vol. 20, p. 211-280. Fosse, R. 1913 Présence de l’urée chez les Invertébrés et dans leurs produits d’excrétion. C. R. Acad. Sci., Paris, Tom. 157, p. 151-154. RHYTHMIC PULSATION 350 GemMiLL, J. F. 1912 The locomotor function of the lantern in Echinus, with observations on other allied lantern activities. Proc. Roy. Soc., London, Ser. B, vol. 85, p. 84-109. HAANEN, W. 1914 Anatomische und histologische Studien an Mesothuria in- testinalis (Ascanius und Rathke). Zeit. wiss. Zool., Bd. 109, p. 185- 250, Dat. 5; 6: Henri, V. 1903 Etude des contractions rythmiques des vaisseaux et du poumon aqueux chez les holothuries. C.R. Soc. Biol., Paris, Tom. 55, p. 1314- 1316. Henri, V., nT Latov,S. 1903 Régulation osmotique des liquides internes chez les holothuries. C. R. Soe. Biol., Paris, Tom. 55, p. 1244-1245. Hérovarp, &. 1889 Recherches sur les holothuries des cétes de France. Arch. Zool. Exp. et Gen., sér. 2, Tom. 7, p. 535-704. =x H6ser, R. 1914 Physikalische Chemie der Zelle und der Gewebe. Leipzig und Berlin. 4te Aufl., xvii + 808 p. : Jorpan, H. 1913 Eine neue Art von Muskeln. Proc. IX. Cong. Int. de Physiol. p. 1914 Uber ‘reflexarme Tiere.’ IV. Die Holothurien, Mitteil. I. Die Holothurien als hohlorganartige Tiere und die Tonusfunktion ihrer Muskulatur. Zool. Jahrb., Abt. allg. Zool. u. Physiol., Bd. 34, p. 365-436. Karka, G. 1914 Hinfithrung in die Tierpsychologie. Bd. I. Die Sinne der Wirbellosen. Leipzig. x1i + 593 p. Loss, J. 1900* Comparative physiology of the brain and comparative psy- chology. New York. x + 309 p. 1900» On the different effect of ions upon myogenic and neurogenic rhythmical contractions and upon embryonic and muscular tissue. Amer. Jour. Physiol., vol. 3, p. 383-396. 1906 The dynamics of living matter. (Columbia Univ. Biol. Series, 8.) New York. xi + 233 p. 1910 The rdéle of the concentration of hydroxyl-ions in the antagonistic action of K and Ca upon Na. Proc. Soc. Exp. Biol. Med., vol. 8, p. HOSE Lupwie, H. 1889-1892 Echinodermen. Bronn’s Thierreich, Bd. 2, Abt. 3, BuchI. Die Seewalzen. Leipzig u. Heidelberg. vi + 460 p., 17 Taf. Mark, K. L. 1913 Preliminary study of the salinity of sea water in the Ber- mudas. Proc. Amer. Acad. Arts and Sci., vol. 48, p. 669-678. Mayer, A.G. 1906 Rhythmical pulsation in Seyphomedusae. Publ. Carnegie Instn. Wash., No. 47, 62 p. 1911 The converse relation between ciliary and neuro-muscular move- ments. Publ. Carnegie Instn. Wash., No. 132 (Papers from the Tortugas Lab., vol. 3), p. 3-25. 1914* The effects of temperature upon tropical marine animals. Publ. Carnegie Instn. Wash., No. 183 (Papers from the Tortugas Lab., vol. 6), p. 1-24. 1914 The relation between the degree of concentration of the elec- trolytes of sea water and the rate of nerve-conduction in Cassiopea. Publ. Carnegie Instn. Wash., No. 183 (Papers from the Tortugas Lab., vol. 6), p. 25-54. 356 W. J. CROZIER Miuter, F.R. 1910 On the rhythmical contractility of the anal musculature of the crayfish and lobster. Jour. Physiol., vol. 40, p. 431-444. 1912 Contributions to the physiology of the Americanlobster. The physiology of the intestine. Contrib. to Canadian Biol., 1906-1910, p. 277-280, pls. 30-35. Moors, B., Epin, E. 8., Wairuey, E., anp Dakin, W. J. 1912 The nutrition and metabolism of marine animals in relationship to (a) dissolved or- ganic matter and (b) particulate organic matter in sea-water. Bio- chem. Jour., vol. 6, p. 255-296. Moors, B., Epis, E. 8., anp WuitLEy, E. 1914 The nutrition and metabolism of marine animals: the rate of oxidation and output of carbon-dioxide in marine animals in relation to the available supply of food in sea- | water. Rept. Lancashire Sea-Fish. Lab., No. 22, p. 297-320. Pearse, A. S. 1908 Observations on the behavior of the holothurian Thyone briareus (Le S[uJeur). Biol. Bull., vol. 15, p. 259-288. Pisron, H. 1909 La loi d’évanouissement des traces mnémoniques en fonction du temps chez la Limnée. C. R. Acad. Sci., Paris, Tom. 149, p. 513- 516. Pouimanti, O. 1913 Sui rapporti fra peso del corpo e ritmo respiratorio in Octopus vulgaris Lam. Zeit. f. allg. Physiol., Bd. 15, p. 449-455. Puirrer, A. 1914 Temperaturkoeffizienten. Zeit. f. allg. Physiol., Bd. 16, p. 574-627. Roar, H. E. 1912 Contributions to the physiology of marine organisms. II. The influence of the carbon dioxide and oxygen tensions on rhythmical movements. Jour. Physiol., vol. 48, p. 449-454. Rosertson, T. B. 1910 Uber die Verbindungen der Proteine mit anorganischen Substanzen und ihre Bedeutung fiir die Lebensvorginge. Ergeb. Physiol., Jahrg. 10, p. 216-361. SaLant, W., AND Hecut, S. 1915 The influence of oxalates, citrates and tar- tarates on the isolated heart. Amer. Jour. Physiol., vol. 36. p. 126- 144. Scort, J. W. 1914 Regeneration, variation and correlation in Thyone. Amer. Nat., vol. 48, p. 280-307. Spantu,.R. A. 1913 The physiology of the chromatophores of fishes. Jour. Exp. Zodl., vol. 15, p. 527-579. Tasurro, 8., AnD ApAms, H. 8. 1914 Comparison of the carbon-dioxide output of nerve fibers and ganglia in Limulus. Jour. Biol. Chem., vol. 18, p. 329-334. Vernon, H.M: 1910 The mode of union of certain poisons with cardiac muscle. Jour. Physiol., vol. 41, p. 194-282. Wertanp, W. 1912 Zur Kenntnis der Entstehung der Darmbewegung. Arch. ges. Physiol., Bd. 147, p. 171-196. Wittey, A. 1899 Enteropneusta from the South Pacific, with notes on the West Indian species. Zodl. Results of A. Willey, Part III, p. 223-334, Cambridge. WINTERSTEIN, H. 1909 Uber die Atmung der Holothurien. Arch. Fisiol., vol. 7, p. 33-40. REACTIONS TO LIGHT IN VANESSA ANTIOPA, WITH SPECIAL REFERENCE TO CIRCUS MOVEMENTS WILLIAM L. DOLLEY, Jr. Professor of Biology, Randolph-Macon College From the Zoélogical Laboratory of The Johns Hopkins University TWENTY-ONE FIGURES CONTENTS PRUE ROCC MON eet Rh ee eke ne idee so 5 Rec Noes Wis aac ot Ae cee SOE Wailer avoye l-5% Shacks eee ae hd OUR Rie ttt, ys ae oo be Ban ae 367 Belaviomormnornalispecimienste- ie. 5/45) eeneen ceineins rele. o1a/oc) dee ete 370 Behavior of specimens with but one functional eye....................... 371 A. Behavior in normal conditions of illumination........................ 371 Erebenaviorin a peam of light... seer ee ccs f.. c's oka eds 5 eee 371 1. Description of reactions—deflection, circus movements and orien- [AEN HIKO) Oleg aortic tea Ee Ea PREM oo cI Cc oHEKG SIRT RCA oP ee) 371 2. Relation between the degree of curvature in circus movements and Chemumimousmmtensihyamerese fee. on. cae eeerecer ania: eae 382 3. Relation between the angle of deflection and the luminous inten- SUV et CSE Te «ty. SemNet ih 3 oar pe MR ace ae EL 383 a. Effect of beginning the trials in different intensities........ 383 b. Effect of sudden changes of intensity on the angle of deflec- [AKO Sohne ean o. 5"5 Me A ie RMI aE a APA Re Ae er ee 386 4. Reorientation after changing the direction of the beam of light.. 389 ©. Effect of the covering of the eye owing to confact................... 394 Drab chaviorninenon-cirectivermnamtnern i400 9. set eisier edict ieee re 399 I. Relation between the degree of curvature in circus movements and the luminous: intensity of non-directive light... 020.0. 005. Ja. 400026: 404 I. Bitectofalluminating:onlyione eye. . 2 &).\206. 2/08 oe he ate steers sae a eeueee 410 1. Effect of illuminating the entire surface of one eye.............. 410 2. Effect of illuminating different areas of one eye................. 413 Generalesummanyvaandeconclislonseeer rere a-oc ation ace eer aero 415 Bibliographys....... - Jupp coc nto vb uge pa om ee aele Ws 6 ch Sota CAL Coe A AA enn owe 419 INTRODUCTION One of the most thorough pieces of work, which have been done on the reactions to light in butterflies, is that reported by Parker (’03) on the mourning-cloak butterfly, Vanessa antiopa. 357 308 WILLIAM L. DOLLEY, JR. This investigator found that these butterflies are highly positive in their reactions to light, but that when they come to rest in bright sunlight they ordinarily orient with the head directed away from the source of light. He found, however, that when one eye is painted black they do not orient, but continuously creep or fly in curves with the functional eye toward the center. Such reactions are usually called circus movements. This be- havior, the author asserts (p. 463), is in accord with the view ‘‘that the orientation of an organism in light is dependent upon the equal stimulation of symmetrical points on its body.”’ A number of other investigators have, also, recorded experi- ments with other organisms in which circus movements have . been observed. Reactions of this nature have been reported in experiments of three sorts: those in which one eye has been pre- vented from functioning, either by being blackened, or by being injured; those in which one antenna has been removed; and those in which certain parts of the brain or of the inner ear have been destroyed. . In these experiments it has been found that photo-positive animals, usually turn continuously toward the functional eye, while photo-negative animals usually turn in the opposite direc- tion. This is especially true in those cases in which one eye has been covered. Holmes (’01 and ’05) and his students, McGraw (713) and Brundin (713), maintain that they have observed this behavior in the following organisms: Hyalella den- tata, Talorchestia longicornis, Orchestia agilis, two species of bees, the robber fly, Asilus, Tabinus, a Syrphid, Ranatra, Noto- necta, several beetles, Stenopelmatus, three species of flies, a number of species of butterflies, and the amphipod, Orchestia pugettensis. In all these cases, positive animals turned toward the functional eye, while negative animals turned toward the covered eye. This, however, was not found to be true in all of the species investigated. Holmes and McGraw (11, p. 370) state that several species of butterflies, among them Vanessa antiopa, frequently went in circles toward the covered eye, while Brundin (713 p. 346) maintains that in positive specimens of the amphipod, Orchestia traskiana, ‘‘cireus movements will occur REACTIONS TO LIGHT IN VANESSA ANTIOPA 359 as often toward the blackened eye as toward the normal eye.” Similar results have also been obtained with animals in which one eye was injured. Radl (’01, p. 458) extirpated one eye of the water scavenger beetle, Hydrophilus, and found that it de- flected toward the side of the injured eye. Hadley (’08, pp. 180-199) seared with a hot needle the surface of one eye of larval lobsters in all stages of development, and maintains (p. 198): “The immediate results following this destruction of photo- reception in one eye are: (1) The production of rapid rotations, often at the rate of 156 per minute on the longitudinal axis of the body, which are invariably in a determined direction. (2) A type of progression in which the larva continually performs ‘circus movements’ or turns toward the side of the injured eye.” Since these animals vary in the sign of their reaction to light at different stages of development, it is interesting to note that Hadley maintains that the circus movements made by animals of all ages were all in the direction of the blinded eye. Mast (10, p. 132) found that ‘‘Planaria with one eye removed, either by gouging it out or by cutting off one side of the anterior end obliquely, turn continuously from the wounded side for some time, evidently owing to the stimulation of the wound, since, after this is healed, they tend to turn in the opposite direction.”’ The destruction of the function of one eye is however not always followed by circus movements. Rdadl (03, pp. 58-64) states that Calliphora vomitoria is apparently not affected in its behavior by having one eye covered, while Musca domestica, although performing circus movements at times, can also “‘run rather long distances in one direction.’ Carpenter (08, pp. 483-491) blackened one eye of Drosophila ampelophila, and reported that now and then one performed circus movements, but he says (p. 486), “This conduct was exceptional, and was never persisted in except in the case of a single sect which had long been active and showed signs of fatigue.” They usually, however, deflected somewhat toward the functional eye as they proceeded toward the light. To quote further (p.‘486), ‘‘They crept in a fairly direct path toward the light, although a ten- deney to deviate toward the side of the normal eye regularly 360 WILLIAM L. DOLLEY, JR. occurred. The insects generally moved in a peculiar, jerky manner. The tendency to diverge from the direct path toward the side of the uncovered eye was overcome by a series of short, quick turns in the opposite direction, which kept them headed toward the light.””’ Mast (11, p. 222) found that the toad, Bufo americanus, with the lens removed from one eye, hops or walks toward a source of light, usually deflecting slightly toward the injured eye. Some individuals, however orient nearly, if not quite, as accurately after the operation as before. Thus, it is evident that there are numerous exceptions to the idea that the destruction of one eye is followed by circus movements. Moreover, it has been found that some animals which make circus movements modify their behavior after having had a cer- tain amount of experience, and move directly toward the light. Holmes (’05), in a detailed description of the behavior of one specimen of Ranatra with the right eye blackened, says that in the first ten trials before an electric light, it made many circus movements, and showed a ‘‘marked tendency to turn to the left.’ In the next four trials it turned directly toward the source of light and in the succeeding ten trials it reached the light by a nearly straight path. After an interval of fifty min- utes, eleven more trials were made, ‘‘and it had not forgotten in the meantime how to reach the light by the most direct means,’’ for it went to the light in every case in a nearly straight course. The author also states that other specimens of Ranatra and Notonecta showed this same modification. Brundin- (713, pp. 334-352) observed similar reactions in the amphipods, Orchestia traskiana and Orchestia pugettensis, except that being negative the animals turned toward the blackened eye. Mast tested on two successive days a toad with one eye destroyed. He says (11, p. 222): ‘‘The following day this toad was again exposed: it now went toward the source of light even more nearly directly than on the preceding day.” Thus, it is clear that the reactions of at least some of these mutilated organisms may become modi- fied as the result of repeated trials. This is apparently not true of some animals. Radl cut out one eye of Hydrophilus, and states that, though it lived for REACTIONS TO LIGHT IN VANESSA ANTIOPA 361 several weeks in an aquarium, it never moved in a straight line, but always in a course curved toward the side of the injured eye. He says (01, p. 458): ‘‘Es hat darnach noch mehrere Wochen in meinem Aquarium gelebt, bewegte sich aber niemals gerade sondern immer nur in einem Bogen concav nach der Seite des extirpirten Auges.”’ This investigator (’03, p. 62) also observed a fly, Dexia carinifrons, on the second day after its eye was blackened and found its behavior was similar to that exhibited immediately after the eye was covered, that is, it moved con- tinually toward the functional eye. The second group of experiments, as previously stated, refer to insects with one antenna removed. V. L. Kellogg (’07, pp. 152-154) removed the left antenna from a male silk worm moth, and found that when such an animal was placed three or four inches from a female it ‘‘moved energetically around in repeated circles to the right, or, rather, in a flat spiral, thus getting (usually) gradually nearer and nearer to the female.’”’ Males with the right antenna removed turned continually to the left. In the same year, Barrows (07, pp. 515-537) removed the terminal segment from one antenna of some fruit flies, Drosophila ampelo- phila, and then, after twenty-four hours without food exposed them to the odor of fermenting banana. He maintains that they moved in circles toward the uninjured antenna in all but a few cases in which they deflected in the opposite direction. The third group of experiments mentioned comprises those in which parts of the brain and inner ear have been injured or removed. In these cases it is also maintained that the animals make circus movements. It can thus be seen that great diversity exists among the results obtained by the various investigators in their experiments on animals with the sense organs on one side destroyed. Among these, those which refer to the eyes are of greatest immediate interest to us. In these experiments it was found that while photo-positive animals usually turn toward the functional eye and photo-negative animals toward the non-functional eye, some turn in the opposite direction and others orient fairly accurately, THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, No. 3 362 WILLIAM L. DOLLEY, JR. while still others make circus movements for a period and then orient fairly accurately. This marked lack of harmony between the results obtained may in some measure, at least, be due to the fact that the num- ber of sources of light was not the same in all of the experiments: Parker does not state the conditions under which the specimen of Vanessa antiopa used by him made circus movements. Radl presumably performed his experiments before a window, Le., under conditions in which the animals received some light from many different directions. The same probably held also for the work of Holmes on amphipods and several insects. In some experiments, however, as in those performed with Ranatra and Notonecta, he worked in a ‘darkened room,’ and used for a source of light a sixteen candle-power incandescent lamp. Brundin and Carpenter also used a similar source of light. It is significant indeed that in every case where a single source of light on the same horizontal plane with the organism was used, at least some trials are described in which no circus movements were made, the animals moving in a fairly straight course toward the light. This was true of Ranatra, Notonecta, Drosophila, Bufo ameri- canus, Orchestia traskiana, and Orchestia pugettensis. On the contrary, in none of the experiments but one, where the light conditions were not sharply defined, have the investigators re- corded any other behavior than movements in circles. This single exception is that described by Radl, in which Calliphora vomitoria and Musca domestica with one eye blackened ran for some distance directly toward a window. The experiments described in the present paper show that in the case of Vanessa antiopa, at least, a knowledge of the number of sources of stimulation is of great importance in a discussion of circus movements; for the same animals, which, in a hori- zontal beam, moved toward the source of light in a fairly straight course, performed circus movements continuously when placed before a window, or when the single source of light was placed above the animal so that the light was non-directive.. The ‘The term ‘non-directive light,’ as used in this paper, denotes diffuse illu- mination. REACTIONS TO LIGHT IN VANESSA ANTIOPA 363 reactions under the former conditions seem to indicate that both eyes are necessary for orientation; those under the latter, that only one eye is necessary. Consequently, if the butterflies had been studied only in front of a window, the conclusions would necessarily have been erroneous. Circus movements have been held by many to have a very important bearing on the question as to the nature of the process of orientation. Holmes discusses this question rather fully. He takes the position that the performance of circus movements indicates a direct or indirect connection between the impulses set up by light in the two retinas and the tension of the muscles of the legs or appendages on the two sides of the body, and that this is a ‘‘sort of mechanical reflex process.’”’ 'To him the pleasure- pain theory explains those cases in which orientation occurs in these asymmetrical animals. He says (11, p. 54): ‘In most crustacea, as in most insects, orientation is effected through the unequal action of the appendages on the two sides of the body. In a form which is positively phototactic, light entering one eye sets up impulses, which, passing through the brain and nerve cord, cause, directly or indirectly, movements predominantly of flexion of the legs of the same side and of extension of the append- ages of the opposite side of the body. If this is a sort of mechani- cal process, we should expect that, in a positively phototactic form, if one eye were destroyed or blackened over, the animal would move continuously toward the normal side.’ Mindful of the fact that the Ranatras and Notonectas in time straightened their courses, and followed the light nearly as precisely as if they had the use of both eyes, he also concludes that ‘‘ Phototaxis may fall, to a certain extent, under the pleasure-pain type of behavior. . . . . Light, in some animals, is followed much as an object of interest is pursued by a higher animal” (’11, p. 55). To these conclusions Brundin (’13) assents. According to Carpenter, the local action theory of tropisms would explain circus movements, were it not that some animals with one eye blackened can orient as accurately as if both eyes were functional. The pleasure-pain theory, he holds, explains 364 WILLIAM L. DOLLEY, JR. this behavior. He says (08, p. 486): “‘It is clear that the tropism theory, with its assumption of local action of stimulus on the side exposed to its effect, does not furnish a complete explanation of these reactions. . . . . A ‘pleasure-pain’ reaction appears to inhibit and dominate a ‘tropic’ reaction.”’ To Radl, cireus movements are an evidence of inequality in the tension of the muscles on opposite sides of the body, pro- duced by the blackening of one eye. He says (’03, p. 63): ‘‘ Bei einem Tier, dem ein Auge geschwarzt wurde, erschlaffen etwas die Muskeln an der Korperseite, wo das Auge nicht sieht; da sich nun die Muskeln der anderen Seite kraftiger bewegen, so erfolgt eine Bewegung in eier nach der Seite dieser starker rect ae Muskeln gekriimmten Bahn.’’ Parker (’03, p. 463), as has been previously stated, maintains that the circus movements he observed in Vanessa antiopa are in accordance with the view ‘‘that the orientation of an organism in light is dependent upon the equal stimulation of symmetrical points on its body.” He says further: ‘‘Should the eyes be the parts stimulated, any interference with one of these ought to result in a disturbance of the direction of the butterfly’s loco- motion. Thus, if the cornea.of one eye were blackened, the insect in locomotion, being positively phototropic, ought to move as though that eye were in shade, namely in a circle, with the unaffected eye toward the center.” To Barrows, who worked on the reactions of Drosophila to odors, circus movements can only be explained by the ‘tropism theory.’ He says (07, p. 535): “It seems impossible to explain the movements under these conditions in any other way than on the basis of the tropism theory. This theory has been stated in several ways. As applied to chemical stimulation, Verworn (99, p. 429) declares: ‘The word chemotaxis is applied to that property of organisms that are endowed with the capacity of active movement by which, when under the influence of chemi- cal stimuli acting unilaterally, they move toward or away from the source of the stimulus.’ ” V. L. Kellogg (07) and Bohn (11) agree with Loeb, whose views are given in the next paragraph, and Bohn even cites circus movements as one of his criteria for tropisms. REACTIONS TO LIGHT IN VANESSA ANTIOPA 365 Loeb (06, p. 140) attempts to refute any notion of a pleasure- pain type of behavior in lower organisms, and accepts the phe- nomenon of circus movements as a fact in support of his theory in explanation of the orientation of animals. This is discussed fully in the Mechanistic Conception of Life. (12, p.: 35-62.) He holds that the orientation of animals is controlled unequivo- cally by external agents, and that in orientation to light, there are two essential factors, the continuous action of light and the symmetrical structure of the organisms. According to his view, which may be called the ‘continuous action theory,’ the tension of the muscles of the appendages on the two sides of the body is controlled through direct reflex ares by the photochemical changes produced by light in the two retinas. He says (12, p. 39): ‘““When two retinae (or other points of symmetry) are illuminated with unequal intensity, chemical processes, also of unequal in- tensity, take place in the two optic nerves (or in the sensory nerves of the two illuminated points). This inequality of chemical processes passes from the sensory to the motor nerves and even- tually to the muscles connected with them. We conclude from this that with equal illumination of both retinae the symmetrical groups of muscles on both halves of the body will receive equal chemical stimuli and thus reach equal states of contraction, while when the rate of reaction is unequal, the symmetrical muscles on one side of the body come into stronger action than those on the other side. Theresult of such an inequality of the action of symmetrical muscles of the two sides of the body is a change in the direction of movement on the part of the animal.”’ It is clear that in this theory it is assumed that light is effective in orientation through its continuous action, that after orienta- tion has occurred, light continwes to stimulate the photosensitive areas, and through direct reflex arcs, continues to affect the muscles of the appendages on the two sides of the body. These assumptions, as stated above, are, according to Loeb, supported by the behavior of animals with the sense organs functional only on one side. He quotes Parker as follows (06, p. 140): ‘Loeb has pointed out that the orientation of an organism in light is dependent upon the equal stimulation of symmetrical points on 366 WILLIAM L. DOLLEY, JR. its body. Should the eyes be the parts stimulated, any inter- ference with one of these ought to result in a disturbance of the direction of the butterfly’s locomotion. Thus, if the cornea of one eye were blackened, the insect in locomotion, being positively phototropic, ought to move as though that eye were in shade; namely, in a circle, with the unaffected eye toward the center.”’ Mast holds that the precision with which some organisms with but one functional eye perform circus movements appears to add support to the ‘continuous action theory,’ but he also says (’11,.p. 222), as a result of his work on the toad, ‘‘ These results show that, in this form and in all other forms which orient after -one eye is destroyed, difference of effective intensity on opposite sides does not regulate orientation.” A glance at these various views shows that the movement of animals in circles when one eye is blackened, or when one antenna is removed, has been held by most of the investigators to support the view that the orientation of animals is in accord with the ‘continuous action theory’ described above. This theory is op- posed by one that may be called the ‘change of intensity theory,’ the adherents of which hold that in some organisms, at least, light does not produce orientation through its continuous action, but by stimuli dependent upon the time rate of change of intensity. According to this theory, an organism going toward a source of light, may turn to one side; but when this occurs, then, imme- diately the photosensitive surfaces are exposed to a change of intensity, and this causes a reaction which results in reorienta- tion, after which the orienting stimulus ceases. The chief points at issue between the two theories concern the following questions: (1) Does light function in orientation through its continuous action, or through a change of intensity? (2) Does an animal, when oriented, continue to be affected by the same stimulus that is effective in producing orientation? and (3) Is bilateral symmetry essential in the process? In view of the bearing that circus movements have on the theories as to the mechanism of normal orientation in animals, and in view of the conflicting results recorded by previous workers it seemed desirable to make a more thorough and a more extended REACTIONS TO LIGHT IN VANESSA ANTIOPA 367 study of this phenomenon than has been done previously. More- over, such a study should throw light on the question as to whether or not the path of nerve impulses resulting in a given reaction can be altered, as well as on the very important prob- lem of modifiability in behavior in general. The mourning cloak butterfly, Vanessa antiopa, was chosen to begin with because the results secured with this. animal by Parker are widely known and frequently quoted. This work is to be followed by a more general study of the phenomenon in question. Before entering upon a discussion of these experiments I wish to express my very sincere appreciation of the kindness of Pro- fessor S. O. Mast in suggesting this problem to me and in so unselfishly aiding me throughout the course of the work. METHODS The butterflies used were all reared in the laboratory from larvae secured from both the June and the August broods in Massachusetts, New York, and Pennsylvania. No difficulty was experienced in keeping them in excellent condition for long periods. They were kept in the laboratory in a large glass case, and fed on honey and a weak solution of maple syrup in water. At frequent intervals the insects were picked up and dropped on filter paper soaked in the latter sweet mixture. If the proboscis was not extended at once, it was uncoiled with a pin, and when once the tip touched the liquid, the animal continued to feed until its abdomen was swollen to an extent which seemed dan- gerous. Since these butterflies pass the winter in the imago state, it is not surprising that six specimens lived from August until the latter part of February. ‘These were the survivors of a lot of about thirty which were received at the same time. Had proper care been taken, it is likely that nearly all would have lived through the winter in the laboratory. The wings of the butterflies were usually clipped to prevent their escape. This was in no wise injurious, for animals with clipped wings lived and thrived at well as those whose wings were intact, and they behaved in the same manner. 368 WILLIAM L. DOLLEY, JR. As already stated, three methods have heretofore been used to prevent the functioning of one eye; extirpation, searing with a hot needle, and covering with asphalt varnish. The latter method was used exclusively in the present work, because it was believed that fewer disturbing factors would be introduced thereby. In the early part of the work, one eye was covered with one or two coats of the asphalt varnish. After having made some experiments with animals treated thus, it was found, to my sur- prise, that insects with both eyes covered in this way still oriented fairly precisely, and went toward the source of light. Thus it is evident that the varnish as used did not exclude all of the light. The eyes were then painted repeatedly until the coats were so thick as to be distinctly evident when the observer was several feet from the butterflies. Under these conditions the animals were indifferent to light. Warned by this experience, the blinded insects used in all future experiments were so treated that it seemed certain that the eye was in every case effectively covered. Moreover frequent examinations were made to make sure that the varnish had not cracked or fallen off; and new coats were from time to time applied to make assurance doubly sure. In work of this sort it is important that the varnish be of such nature that it does not injure the eye in any way. The effect of the covering was consequently repeatedly tested by removing it from the eyes after it had been on for some time. Insects thus tested behaved as did those whose eyes had not been blackened, showing no effect from the varnish. The supply stock of butterflies was ordinarily kept in a large cage which was four feet high, four feet long and two feet wide. This was fastened against a south window in such a way that the window formed one side of the enclosure. The opposite side was also of glass and faced a small laboratory room. ‘The other two sides, the top, and the bottom, were of wood. Careful observations were made on the insects in this enclosure, from time to time, throughout the whole period over which the experi- ments extended. But a much more thorough investigation of REACTIONS TO LIGHT IN VANESSA ANTIOPA 369 the behavior in light was made in a dark room under accurately controlled environmental conditions. In these experiments the animals were exposed in a horizontal beam produced by means of a 110 volt Nernst glower. The glower was mounted in front of a small opening in a light-proof box that was painted dead black inside, so as to form a non-reflecting background. It was placed 10 em. from and at the same level with the top of a table on which the animals were tested. By means of screens the light from the glower was so cut down as to produce a sharply defined beam of the size desired. The edges of this beam could be clearly seen on the black top of the table. This beam was the only light in the room, and this was in large part absorbed by means of dull black paper hung over the exposed walls. There was consequently very little hght in the room aside from that in the beam. Under these conditions therefore, the animals were exposed in a beam of light from a single, small and con- centrated source. The limits of this beam were very apparent in the dark room in which the experiments were made. The nature of the source of light and the sharply defined character of the beam are im- portant, for experiments described later demonstrate that the behavior of animals with one eye blackened depends to a marked extent upon whether there are one or more sources of light present. Not only was the behavior of the animals described by the observer, but the butterflies, themselves, were forced to make permanent records of their own behavior. This was done by allowing them to walk on sheets of paper which had been covered with soot from an oil lamp. These sheets measured 20 x 25 em., but in some experiments, a number of them were placed side by side until the area was as large as desired. The tracings made by the insects were made permanent by means of a coat of shellac. The butterflies were frequently allowed to walk over the sheets of paper covered with soot, and then the same experi- ment was repeated without the use of the blackened paper. The same results were secured in both cases. This shows that the behavior was not affected by the soot. This method of hav- 370 WILLIAM L. DOLLEY, JR. ing the animals make permanent records of their own behavior is most valuable, for the records can be kept indefinitely and studied, thus giving opportunity to recognize many significant features which otherwise might have been overlooked at the time the experiments were performed. It would be of value, no doubt, to the keenest observer. BEHAVIOR OF NORMAL SPECIMENS In the study of normal animals in the cage referred to above, Parker’s observations were confirmed. It was found that the insects were highly positive in their reactions to ight. During the day, in the absence of direct sunlight, they were usually in active movement, flying against the window. Occasionally an animal would fly around the cage, but this was exceptional. When at rest the butterflies were usually grouped on the window side of the cage, -where they assumed various positions on the bottom of the window sash, some facing the light, others in a horizontal position at right angles with the rays, some hanging on the sash in a vertical position with their heads up, and others hanging with their heads down. When the sun was so situated that the butterflies were exposed directly to its rays, and they were undisturbed, they usually ceased their active movements and oriented very definitely. They turned so as to face directly away from the sun and spread their wings to their fullest extent, exhibiting behavior similar to that described by Parker. This position was retained indefi- nitely unless the insects were disturbed. In a beam of light in the dark room the responses were quite different. In making observations under these conditions the animals were placed in the beam at various distances from the glower so that they faced the source of light. As soon as they were released they usually darted directly toward the glower and continued until they reached the edge of the table. The insects were always found to be highly positive in all intensities in which they were tested. They never exhibited the slightest indication of negative reactions. They never came to rest with the head directed from the light and the wings spread, as they usually did in direct sunlight. REACTIONS TO LIGHT IN VANESSA ANTIOPA OL BEHAVIOR OF SPECIMENS WITH BUT ONE FUNCTIONAL EYE A. BEHAVIOR IN NORMAL CONDITIONS OF ILLUMINATION In normal conditions of illumination the behavior of butter- flies with but one functional eye was very different from that described above. Such specimens were tested on the floor of the cage referred to previously, on a table before a window, and in a beam of light in the dark room. Before a window and on the floor of the cage it was found that whenever they moved they turned continuously toward the functional eye, exhibiting behavior similar to that described by Parker. The periods of activity, which in some cases lasted for several minutes, alter- nated with periods of rest in which the animals remained practi- ‘cally motionless, as if recovering from fatigue. But the point that is of especial interest is that they continued to make circus movements from day to day, and that they did not learn to orient. Two insects with one eye blackened were kept for twenty-three days, and although they were observed many times each day no modification in their behavior was detected. In this respect, however, the reactions in a beam of light differed greatly. B. BEHAVIOR IN A BEAM OF LIGHT 1. Description of reactions—deflection, circus movements, and orventation Under the conditions of illumination described in the preced- ing paragraph, the animal receives light from all sides, and all the large areas of the functional eye are approximately equally illuminated in every position assumed by the insect. When exposed to the light in a beam the animal receives light from only one direction, and consequently every movement that is not directed toward the glower produces a change in the illumi- nation of different large areas of the uncovered eye.. This may account for the difference in behavior observed under the two conditions of illumination. The behavior in a beam of light of Vanessa with one eye blackened was studied in 46 different individuals and many of ala WILLIAM L. .DOLLEY, JR. these were tested on several successive days. In nearly all cases the animals were forced to record their reactions on carbon paper, as previously described. In all tests the butterflies were placed in the beam so that they faced the light directly. The results obtained varied considerably in different individuals and also in the same individual under different conditions. In some respects, however, there was but little variation. Nearly all of the butterflies tested turned toward the functional eye immediately after they were exposed, regardless of the luminous intensity or the axial position with reference to the d h 7 cf e ih ie : pik 9 Fig. 1 Reproduction of various trails made by different specimens of Vanessa with the left eye blackened when exposed in a beam froma Nernst glower. The diverging straight lines represent the limits of the horizontal beam. The arrows indicate the direction of motion. Their trails show that there is great variation in the reactions of different individuals under the same conditions. direction of the rays of light. Some of them continued to turn in this direction making repeated circus movements? (fig. 1, a and b) until they became fatigued and stopped, or until they reached the edge of the beam, where many turned sharply toward the glower and traveled along the edge of the beam toward the source of light, as is shown in figure 1, c. A few, however, did not turn toward the light when they reached the edge of the beam, but passed into the shaded region, continuing to make circus movements (fig. 1, e). Others did not make circus move- ments, but turned until the longitudinal axis made a certain 2 In the present paper the term ‘circus movements’ with no further explanation means continuous movement toward the functional eye. REACTIONS TO LIGHT IN VANESSA ANTIOPA 373 angle with the rays of light, and then continued until they reached the edge of the beam. Here they usually turned sharply toward the glower and moved along the edge of the beam toward the source of light (fig. 1, f) but occasionally they continued to turn here and made circus movements (fig. 1, 7), and sometimes they did not respond at all when they reached the edge of the beam, but continued until they had passed into the shaded region from 2 to 5 cm. when they usually turned and proceeded directly toward the glower, remaining in this region (fig. 1, g). On afew occasions, however, they did not turn when they reached the edge of the beam, but proceeded on in the shaded region indefi- nitely (fig. 1, h). A few animals did not turn toward the func- tional eye, but oriented fairly accurately and walked toward the glower in a nearly straight course (fig. 1, &). Several specimens in some trials turned toward the blackened eye, crossed the beam, and on reaching the edge turned and walked along it toward the source of light (fig. 1, 7). Many insects, as the trials proceeded, showed an increase in accuracy of orientation. This was evident in three respects: (1) in the number of circus movements made, (2) in the angle of deflection, and (3) in the promptness with which they oriented at the edge of the beam. The above general description may perhaps be made clearer if the reactions of one organism are described in detail. This animal designated as butterfly 10/25-a (left eye blackened) was tested on three successive days. On the first day this butterfly was given twenty trials (fig. 2). In every one it turned toward the unblackened eye immediately upon being placed in the beam. In the first trial it crossed the beam at an angle of approximately 95 degrees with the rays of light, and passed into the shaded region. After it had gone 6 em. in this region it turned to the left (the blinded eye) and walked toward the glower in a slightly zig-zag course, remaining, however, in the comparative darkness to the right of the beam. In the second trial, after crossing the beam at an angle of nearly 80 degrees, it again went to a point 6 cm. beyond the edge of the beam, but then it turned sharply to the right (toward the func- 374 WILLIAM L. DOLLEY, JR. tional eye) and performed a circus movement. ‘This was fol- lowed by a fairly straight course for 7.5 cm. At this pomt the organism turned again to the right as if to make a circus move- ment but did not complete it, turning instead to the left toward the source of light. In the third trial the insect made a circus movement as soon as it was placed in the beam and then crossed the beam at an angle of about 95 degiees with the rays of light, and went 3 cm. into the shaded region where it turned toward the blackened eye and moved in a course nearly parallel with the edge of the beam. In the fourth the behavior was like that in the preceding trial except that after the organism passed the Fig. 2 Reproduction of 20 successive trails made by butterfly 10/25-a (left eye blackened) on the first day of the tests. a and b, limits of horizontal beam of light; 1-20, trailsmade in successive trials; small arrows, direction of movement of animal; large arrows, direction of rays of light; illumination at x, 624 me.;3 at y, 250 me. edge of the beam it did not turn toward the glower, but con- tinued on in a fairly direct course until it reached the edge of the table. In the fifth trial the butterfly continued across the beam at about the same angle as in the previous trials until it had gone 2.5 cm. beyond the edge. At this point it turned toward the blackened eye and moved fairly directly toward the glower. In the sixth the organism again made a circus movement imme- diately upon being placed in the beam. It then crossed the beam at right angles with the rays, and on reaching the right 3 Throughout this paper the abbreviation ‘mc.’ will be used to indicate meter- candles. REACTIONS TO LIGHT IN VANESSA ANTIOPA 375 edge, immediately turned toward the blackened eye and moved along the edge of the beam toward the glower. In the seventh, eighth, and in the twelfth to the nineteenth trials the behavior of the butterfly was essentially the same as in the fifth, but it usually went further in the shaded region before turning toward the glower, this distance varying from 2.5 to 14cm. After orien- tation, however, it continued to move in all cases fairly directly toward the glower. In the tenth and third trials, the behavior was essentially similar. In the ninth, eleventh, and twentieth the reactions were also very much alike, the organism in each trial curving gradually toward the functional eye, in this way passing beyond the edge of the beam into the shaded region outside, and then coming back to the edge again. On reaching the edge of the beam the second time the butterfly turned much more sharply toward the functional eye, thus completing a circus movement and at the same time arriving at the edge of the beam a third time. When this occurred, the insect turned toward the glower and moved along the edge of the beam toward the source of light. | These reactions in the trials on the first day of the tests show: (1) that Vanessa with but one functional eye tends to turn toward this eye when placed in a beam of light; (2) that it can orient; (3) that orientation does not usually occur in the beam, but does occur either at the edge of the beam or several centimeters beyond it; (4) that after cireus movements have been performed in a given trial the animal often orients and moves directly toward the source of light; and (5) that a change in illumination seems to favor the performance of circus movements, since, out of 8 ’ circus movements, 4 were made almost immediately after the insect was placed in the beam and before it had reached the edge of the beam, 3 were made at the edge of the beam, and only 1 was made elsewhere. On the second day in all of the first eight trials, except the fifth, the butterfly assumed an angle of about 90 degrees with . the rays, and then traveled across the beam and into the shaded region for a distance of from 1.5 to 9 em. where orientation occurred (fig. 3). In the fifth it continued on to the right in 376 WILLIAM L. DOLLEY, JR. a moderately straight course until it fell off the table. The behavior in the next three trials was very much alike, the organ- ism performing a circus movement upon first being placed in the beam, and then, after having gone a few centimeters beyond the edge, it turned and went toward the glower. The eleventh trial is interesting in that, although the organism was started very much nearer to the glower, and consequently in much stronger te ut a3) y Fig. 3 Reproduction of 34 successive trails made by butterfly 10/25-a (left eye blackened) on the second day of thetests. a and 6, limits of horizontal beam of light; 1-384, trails made in successive trials; small arrows, direction of movement of animal; large arrows, direction of rays of light; illumination at x, 624 me.; at y, 250 mc. light, it, after having performed a circus mcvement, deflected at an angle of only 40 degrees with the rays of light, while in several of the previous trials in which it had started further away from the source of light it deflected at a much greater angle. In the twelfth trial the butterfly made a circus move- ment when first started and then after having gone 1.5 em. beyond the edge of the beam it again performed a circus move- REACTIONS TO LIGHT IN VANESSA ANTIOPA Buh ment. This was followed by a zig-zag course nearly parallel with the edge of the beam. This circus movement is worthy of notice for it was made in the shaded region outside the beam, when the animal was in very weak light. It should also be noted that the diameter of the curve made is very nearly the same as the diameter of the curve made in the beam in comparatively strong light, when the insect was first started in this trial. This peculiarity will be correlated later with the results of other experi- ments. In the thirteenth trial after performing a circus move- ment in the beam the organism continued to the right in a fairly straight course to the edge of the table. In the fourteenth a circus movement in the beam was made, and then the animal went 7 cm. beyond the edge and oriented, moving toward the glower. In the fifteenth it crossed the rays of light and made a circus movement to the right of the beam. It then went toward the glower in a fairly straight line, but before reaching the source of light it made another circus movement. In the sixteenth a circus movement was made to the right of the beam. This was followed by a zig-zag course toward the glower. The behavior in the succeeding eighteen trials was essentially similar to that described above. It should be noted, however, that in the twenty-fourth trial the butterfly after moving to the right until the edge of the beam was reached turned more sharply toward the functional eye at this point. It did not, however, perform. a circus movement, but gradually turned to the left. This sharp turn toward the functional eye on reaching the edge of the beam seems to support the conclusion arrived at from the trials on the* previous day, namely, that change in illumination tends to favor the performance of circus movements. These trials on the second day thus confirm strongly the con- clusions drawn from the reactions on the first day, and they show moreover that after a certain amount of experience the angle of deflection tends to decrease, for on the first day the average angle between the path of the butterflies and the rays of light was 100 degrees while on the second day it was only 89.5 degrees. The reactions on the third day (fig. 4) differed very markedly from those described for the first two days in several respects. THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 20, No. 3 378 WILLIAM L. DOLLEY, JR. In all but the fourth and fifth trials the organism turned toward the functional eye, crossed the beam at a definite angle which was smaller than on the preceding days, and, on reaching the edge of the beam, turned at once to the left and walked along this edge toward the glower. In the fourth trial it responded very much like a normal specimen, walking down the center of the beam in a fairly straight line. In the fifth it deflected toward the blackened eye. No circus movements were made in any of the trials on this day. Fig. 4 Reproduction of 23 successive trails made by butterfly 10/25-a (left eye blackened) on the third day of the tests. a and b, limits of horizontal beam of light; 1-23, trails made in successive trials; small arrows, direction of movement of animal; large arrows, direction of rays of light; illumination at x, 906 mc.; at y, 266 mec. Compare figures 2, 3 and 4 and note that the insect on the third day made no circus movements, while on the two preceding days, it made numerous ones. Note also that the angle at which it deflected with the rays of light decreased. e By comparing all of the reactions observed during the three days it will be seen that modification occurred in three different respects, as follows: (1) On the first two days there were numer- ous circus movements; on the third day there were none what- ever; (2) On the first two days the butterfly usually passed into the shaded region a considerable distance before it turned and went toward the glower; on the third day it turned toward the glower promptly on reaching the edge of the beam; (3) The angle of deflection was greatest on the first day and least on the third, the average angle at the edge of the beam for the three days being respectively 100, 89.5, and 41.5 degrees. The reac- REACTIONS TO LIGHT IN VANESSA ANTIOPA 379 tions of three other insects showing similar behavior are pre- sented in figures 5, 6 and 7. The results presented in these figures as well as in the preced- ing ones seem to show that butterflies with but one functional eye improve in the accuracy of orientation with experience. This conclusion and others are strongly supported by the results 8b ‘a yi0 DI lal» Fig.5 Reproduction of 15 successive trails made by butterfly 7/29-c (right eye blackened). a and b, limits of horizontal beam of light; 1-15, paths made in successive trials; small arrows, direction of movement of animal; large arrows, direction of rays of light; illumination at x, 4892 mc.; at y, 544 me. Note that this insect made three circus movements in the first four trials, while in the next eleven trials it made none. obtained in all of the tests made. These are briefly summarized in table 1. This table will be clearer if a brief explanation of some of the data is given. In the columns headed ‘Direction turned’ is stated the direction toward which the butterflies turned immedi- ately after they were placed in the beam. The average angle of deflection was ascertained in the following way. The angle 380 WILLIAM L. DOLLEY, JR. between the rays of light and the trail of the insect at the edge of the beam in each of the trials was measured. This angle is termed the ‘angle of deflection.’ The average then was com- puted for a number of the first trials on each day, this number being equal to the number of trials on that day on which fewest trials were given. The columns marked ‘Place where orientation Fig.6 Reproduction of 40 trails made by butterfly 10/1-b (left eye blackened). A, 1-20, trails made in successive trials on the first day of the tests; B, 1-20, trails made on the second day of the tests; a and 6, limits of horizontal beam of light; small arrows, direction of movement of animal; large arrows, direction of rays of light; illumination at x, 925 mc.; at y, 266 me. Note that this insect modified its reactions in that it made numerous circus movements in the trials on the first day, but made none in the trials on the second day. occurred’ also demand some explanation. By ‘Orientation’ is meant the assumption of an axial position with the head pointed directly toward the glower followed by movement in this direc- tion. If the animal turned and moved directly toward the source of light before it reached the edge of the beam it is said to have . oriented ‘in the beam.’ If it, however, went more than one centi- meter beyond the edge before it turned toward the glower it is | wer 0 0 6 & 1 0 o 0 0 ’ 1 9g Dual 0 08 ie ey sing % ) a we % of te tiuld ’ 9 a Dial " oe te np ol 8 0 Dial 1 oe 1 aioe ’ 0 oT wt 0% oF or sing ’ i L wt ot of te sn # 1 1 wr oO of be Ane 0 1 u Dual a oe or King 0 o ft Ua 8% oe ie sine 9 0 n wer ” oe or sine 0 o or ut on of ve | o 1 6 Ww o1 08 sz o 0 or art oF oe w one /L 0 z sr wi at 08 % sing t 0 le wen of oF oy z 0 oe went zt on 1 y 0 we WL % 08 a “LTE 1 z 8F want 9 08 1 sine 0 0 1 wt u oe % 0 ° 8 wt iy or % 0 0 " Dial i ov % 1/4 0 0 u wer " ov & sing 0 0 0% Day 0% clad & i 0 08 Dial Ww 99 9% 0 t or wr 7 00-98 ba MhB/8 a BB oe wel 99 0-08 vz | ee eae SSE Peg STS. 0 0 it of & 0 0 u of 9% 0 0 1 or % re/e 0 0 or 08 % Aye L ' 0 0 ong 0 0 1 t) or Url it 8% od ) 3 e ry 1 o19 0 0 1 0 ar UL " i % £ z 8 0 909 0 0 0 0 a Dal et 6-05 % or 9 " r our 0 0 t a on wT ry %% 8 T/L t ® wiz 099 0 0 1 ut &% Wr se St-0r Ea , 9 % | 998 0 0 t 1 ze U1 % on 1% stor ‘ 0 i 0 ont 0 0 0 9 ot w 8-08 t i) £ 1 mn 9 one 0 0 t 9 m w Beet 9 1 ' st 1 080 0 0 0 0 i ie oro 9 £ t oh 0 os 0 0 0 1 w C3 or 8I ' ot OL 6 2 bi 0 one cy 1 0 0 0G 0% Craig £ ’ a a 0 ou 0 0 0 0 cd z ore z q0 |} o 0 o1 0 0 0 0 8 8 or Ht | (Oe 0 0 0% 0 ony 0 0 0 0 0G 0G S81 oT 1 0 ts 0 on 0 1 0 1 te pod rat u 0 0 0 0 0 0 0G 0 0 0% ad % 0 o om | 0 08 0 1 0 0 07 went 0% tr-07 % pat/6 0 0 o | 0 08 0 0 0 0 0G awe 0G op-0P % 1 0 § ut oor 0 0 & a st ar 0G 08 % 0 " " 0 om 0 1 0 0 st WANE % 08 % doquioydos 0 oO or 0 009 0 0 0 0 or wor 0G 9c st 0 6 u 0 0°89 1 0 0 0 0% wor 0% oF % 0 s Z 0 sus ) 0 0 0 0% U9 0G oe 8 0 aa/6 8 a 0 0 o'00 £ 0 0 1 % wer % oe wz soquioydog 0 0 oo | 2 £ or £ 0 4 al WL et 08 or Ww * 6 , or " ry £ 1s wor 89 of 6 a we | oF a iit $98 o1 £ 9 a9 WL tL 08 8 Aine 1 0 w | 0 eee 0 0 0 1 % au 08-06 2 0 0 w | 0 008 0 0 0 0 % % O8-ts 9 0 c or 0 ca) 0 0 0 o ity 1 ote 9 0 t 8 0 Leo 0 0 0 0 0 0 gots 5 Roars : 1 L at 0 99 0 0 cy 0 07 0% 8c & 0 ' a 0 oW 0 0 0 1 6r 0G st-8a & | z 6 6 0 rat) t 0 0 0 % 08 tte I | ; 49qov20 | | \ 6 1 1 org 0 0 1 1 % WL ve ost of e , er 0 ou 0 t 0 0 0% Dat 07 Sb-8z 6 0 0 cy , OF 0 0 ’ 0 we wr ve o0r & L or 6 z £00 0 1 z 0 ie we te 90-88 gtr) lI eceae aye t r ol $ LM 0 0 Cy z nt Dial w oF 9% 3 L 1 a 0 Oe 8 1 0 0 0% Wr 0% oF % ow | at 1 0 vse G 1 0 0 oY wr » x iz soquindes: 0 G 7 1 wu 0 0 0 0 ze a ort of } 0 0 mw | oo ow 0 0 0 0 ze we shee « | o 0 wm | 0 eg 0 0 0 0 % % sh9% 3 8 0 a 0 oon £ L 0 0 ie ie shee x 0 0 oz | 0 Or) 0 0 0 0 0% 0% sae % pon 0-86/6 t 1 ar t 9'9% 0 1 t 6 6 0% st % z 6 " 4 900 0 0 0 1 0% te 0% &% 9 1 ® 0 o29 0 0 0 0 %% wean sz 08 Ey aaquaay dog: 0 ' 9 8 908 t 0 8 0 2 wer 1 oe 65 simp 0 1 " & £00 t ) 0 0 a Dual a or oT O-D1/L Sy 1 0 ol 0 9h Ue 0 0 0 t 81 Dial 08-85 1 £ 8 8 ou 1 o i 1 ae we Lorne wa /01 0 ) wz | o 0 o 0 0 8 HEME i ) 0 or 0 oes 0 0 0 0 or o ose eae 0 £ t o o°01 0 1 0 0 s fy ¥e-st > ’ 0 or u 0 1 cy o 0 it ie oral or ’ ’ 0 " £ on t A 1 0 or 0% Qr-RT ot » || ou | ot } 0 #08 nt eel |p ce ie want we OFS 1 fr 29g J i 4 6 0 988 £ so | 0 0 ie Dial te 90°9% st ’ 0 ot 0 Lor 0 ni ' 0 or WoL 0% ae-90 % w4e/0l IHOWO 1 0 or £ ov o 0 & I or Dal cd Base 6 | t ae oi o ew t 0 o 0 ics yer w Si-at 3 o-8/01 49qm00 0 0 a pit wer or on sl 0 0 0 APT 1s oy tL 1 0 0 or wr ita oe 1 Laie t 1 0 0 , ~s/or 0 404090 #1 0 " 1 ooy a 0 1 1 rd hal oF cla c i} ¥e ‘ 0 z 08 oe | 9 | Oo 0 1 wr " pid 1 wot 49190) 9 0 " 0 on ) 0 1 6 or wer 0% Owe 5 ) 0 ' a fF oor 0 o 0 g 1 wT 0% ORs ~ { -°/1 s 9 or 0 tor o Ty 0 o OF w1 0% ote 9 1 & a 1 vr 0 0 1 0 4 trot ] 9 % ’ 0 8°68 ' " 0 0 ” Pid paren 1 cl ? 0 oO 00r r , i) o oF chat 0 0 08 0 oe 0 o ) 0 oc uel ) 0 1 0% o ow 0 0 0 te hal \ sss My best thanks are due to Dr. E. L. Mark, and to the trustees of the Hum- boldt Fund, for making it possible for me to stay at the Bermuda Station during the summer of 1915. 429 430 SELIG HECHT a row of cilia on each side of these slits, there are in all about 384,000 short rows (ca. 0.2 mm. long) of cilia lashing water through the pharynx of the animal. Such water currents are not uncommon among marine animals. They have been studied elaborately in Lamellibranchs by Wal- lengren (’05); and the work of Orton and of Andrews (’93) has shown how similar the mechanism of their production is in Lamellibranchs, Ascidians, and Amphioxus. The details of the maintenance of such currents in the gastrovascular system of Aurelia (Widmark ’13) have brought to light the long suspected fact that, unlike Amphioxus, Tunicates, and Sponges (Parker "10 and ’14), muscular movements are an important factor in reinforcing the ciliary current produced by medusae. 2. Respiration (cf. however, Dakin, ’09, p. 52; Pitter ’07, Tab. V) and the gathering of plankton organisms (Orton, p. 20) are the two functions commonly ascribed to the incoming cur- rent. To these, Pitter (07) has proposed the addition of a third. On the basis of metabolism studies, he believes that plankton food is insufficient for the carbon requirement of a variety of marine animals, ‘“‘and that sea-water constitutes a very dilute nutrient solution, from which they resorb their food entirely or in part” (Piitter, 714, p. 98). To the water current, then, falls the task of bringing this ‘nutrient solution’ to the place where it may be resorbed.’ I mention Piitter’s conclusions because he, and others who have criticised him (Lipschiitz, Moore et al.) make frequent assumptions as to the volume of the current produced by various animals, having, however, almost no accurate quantitative data on which to base their statements (Pitter ’07, p. 293). The velocity measurements made by Wallengren (p. 23) on the anal current in mussels, and the roughly accurate estima- tion of the volume of water passing through the Gehduse of an appendicularian, made by Lohmann (’09, p. 220), show that the quantities are rather large. The only values actually determined are those found by Parker (14) for a Bermuda sponge. A finger 3 An actual case of such a resorption (really a synthesis, too) has just been demonstrated for fats by Churchill (115) with the fresh water mussel. WATER CURRENT PRODUCED BY ASCIDIA ATRA 431 of Spinosella whose volume was ca. 95 ce. discharged 54 ce. per minute through its osculum, or about 78 liters a day. II 1. For determining the volume of the current of water gener- ated by Ascidia atra, I adopted a simple indirect method. ? Exact time not noted. 478 Re LYOUNG resemblance between it and the bellies of the mice was correspondingly lessened. Within ten minutes one mouse was taken from the clay. The mouse taken in this experiment was nearer the baseboard of the cage than the other two on the clay. On this baseboard the owl frequently alighted after flying across the cage from its perch. Experiment 5. Same as Experiment 4, except that only one mouse was placed on the earth. Within thirteen minutes it was taken from the latter. Experiment 6. Figures 8 and 18. Six mice were placed on gypsum and six on moist clay mingled with leaves, straw, etc.2* Almost im- mediately the owl flew directly to the gypsum and took a mouse therefrom. Experiment 7. Same as Experiment 6, with same result. Experiment 8. Same as Experiment 6, except that only four mice were placed on each background. In six minutes, the owl flew directly to the gypsum and took a mouse from it. Experiment 9. Figure 14. Four mice were placed venter upper- most on moist earth, and four on clay, the backgrounds being suffi- ciently excavated to bring the bellies of the mice flush with their sur- faces. The relative positions of the backgrounds in this experiment was again reversed to the original position. Within twenty-four min- utes one mouse was taken from the earth. Experiment 10. The same as 9, but position of backgrounds re- versed. The following morning I found four mice removed from the earth and one from the clay. Experiment 11. Figures 8 and 11. One mouse was placed on gyp- sum and one on leaves mixed with moist earth and clay. Within five minutes, the mouse was taken from the former background. Experiment 12. Same as Experiment 11, except that the back- grounds were reversed in position. The owl flew back and forth across the cage several times, twice alighting on baseboard of cage nearest the background of earth and leaves. In about ten minutes it alighted on the baseboard a third time, remaining there for ten or twelve minutes more. Soon after (exact time not recorded) the mouse was taken from the gypsum. Sometime during the night the mouse was taken from the leaves. Summary In eleven out of twelve experiments or 92 per cent the combi- nation of greater contrast was chosen, and in the other experiment (4) the resemblance between the mouse chosen and its back- ground was not great, and the owl probably alighted nearer to it than the other background before feeding. °*S In this and the two succeeding experiments, the position of the light and dark backgrounds was the reverse of that in the preceding experiments. EXPERIMENTS ON PROTECTIVE COLORATION 479 SERIES VII AND VIII Series VII and VIII were conducted simultaneously and the birds (kingbird—Tyrannus tyrannus and grackle—Quiscalus quiscula anaeus) confined in the same cage.*4 The kingbird was taken from the nest before learning to fly while the grackle was an adult which flew into a building and was caught there. In Series VII the latter bird was employed as the preyer, and in Series VIII the former. In each series several different species of insects (as recorded for the individual experiments) were used as prey. Different backgrounds (the character of which and the relative contrasts between them and the insects placed thereon being noted in each experiment) were prepared at one end of the cage, while the birds were perched at the opposite end, care being taken in each case to avoid attracting the attention of the birds more to one background than to the other while the prey was being placed upon them. The insects in each ease were killed before being placed in position. Kxperiment 1. Figures 33 and 37. Five Melanoplus’ were placed on hay and five on sand. In fifteen minutes the grackle dropped to the ground and walked to the hay from which four grasshoppers were taken followed by four from the sand. The contrast between the grasshoppers and the sand was greater than that between them and the hay, but their resemblance to the latter background was not very great. The bird seemed afraid of those on the sand, drawing back in apparent alarm after seizing one. Experiment 2. Figures 25 and 44. One Gryllus pennsylvanicus was placed on sand and one on moist earth. The former combination presented the greater contrast, but a little sand accidentally mixed with the earth reduced the resemblance between the cricket and the latter background. After three minutes the grackle flew over the back- grounds twice, apparently taking no notice of either cricket. Five minutes later it alighted nearer the earth background, from which it took the cricket. Experiment 3. Same as Experiment 2. In one minute the grackle flew to the ground alighting nearer the cricket on the earth, which was taken one minute later. ‘ Experiment 4. Figures 387 and 50. One Melanoplus was placed on grass and one on sand. The contrast here was greater in the latter than in the former combination. The grackle immediately dropped 74S GExe lesiexl6) mo im size. 480 R. T. YOUNG to the ground and then walked to the backgrounds, somewhat nearer the grass, but took the insect from the sand. Experiment 5. Figures 25 and 58. One Oecanthus quadripunctatus was placed on grass and one on moist earth. The grackle at once dropped to the ground, approached the grass first and took the insect therefrom. The contrast here was greater in the insect-earth than in the insect-grass combination. Experiment 6. Same as Experiment.4. In one minute the grackle flew across the cage to a point nearer the grass than the sand, but walked past the former to take the insect from the latter background, although it was smaller than that on the former. Experiment 7. Figures 35 and 58. One Oecanthus was placed on grass and one on sand, the latter combination presenting the greater contrast. The grackle flew at once to a point nearer the grass back- ground from which the prey was taken. Experiment 8. Same as Experiment 4. The grackle at once dropped to the ground and in two minutes passed the grass to take the insect on the sand background. Experiment 9. Same as Experiment 7. The contrast here was only slightly greater between the insect and the sand than between the former and the grass. The grackle almost immediately dropped to the ground, passed the grass and then turned back and took the insect from it. It then picked some crumbs from the sand. Evi- dently the insects were not seen by the bird before it reached the backgrounds. Experiment 10. Same as Experiment 5. After twenty minutes the grackle approached the grass from which it took the insect, then taking that on the earth. Experiment 11. Figures 25 and 37. Three Melanoplus were placed on hay mixed with moist earth, and three on sand. The latter com- bination showed the greater contrast. The grackle passed first to the hay from which it took the insects and then to the sand from which they were next taken. Experiment 12. Figures 25 and 58. Four Oecanthus were placed on moist earth and four on grass, the arrangement being as shown in the accompanying diagram (fig. B). The grackle’s line of approach and the order of seizure of the insects are also shown in the diagram. The grackle approached the backgrounds and began feeding in one minute after the start of the experiment. The contrast in this experi- ment was greater between the insects and the earth than between the former and the grass. Experiment 13. Figures 40 and 55. One Gryllus was placed on a mixture of burnt paper and moist earth and one on ashes. For some time (time not noted) the grackle remained on its perch. It then flew to the ground and began picking up various small scraps of food, fiaally approaching the backgrounds, but stopping to peck at some- thing when but a few centimeters distant. It then took the cricket from the earth and next that from the ashes. The contrast between EXPERIMENTS ON PROTECTIVE COLORATION 481 the cricket and the grayish-white ashes was greater than that between the former and the dark background. The accompanying sketch (fig. C) shows the arrangement of the backgrounds and the line of the birds’ approach (xy) and the point (x) where it stopped to pick up a particle of food. Grass Humus Grass Humus Grass Line of Approach Fig. B- The figures indicate the positions of the insects on the backgrounds, and the order of their seizure. Experiment 14. Figures 40 and 43. One Gryllus was placed on scraps of burnt paper and one on ashes. The resemblance between the cricket and the paper was close and the contrast between the former Ashes Fig. C and the ashes was strong. After five minutes the grackle flew to the ground close to the latter background, from which it took the cricket. It then took that on the paper. Although the bird flew directly to the backgrounds in this experiment it apparently did not see the insects until after alighting, for it turned from its first position in order to seize its prey. Experiment 15. Figures 24 and 30. Two Melanoplus were placed on ashes and two on a mixture of dead leaves, straw, and earth, the 482 R. T. YOUNG former combination presenting the greater contrast. The grackle at first flew to a perch about 2 m. from the backgrounds, from which, three minutes after the start of the experiment it flew to the ground near the latter background, but passed it by and attempted to seize an insect on the former when I interrupted it. Experiment 16. Same as Experiment 15. In about seven minutes, the grackle flew over the backgrounds alighting nearer the ashes, but seizing an insect from the leaves and straw. Experiment 17. Figures 30 and 46. Two Silpha surinamensis were placed on burnt hay and two on ashes. In twenty minutes the grackle flew to the ground at the middle of the cage and began feeding. It soon approached the backgrounds on the side of the burnt hay back- ground, from which it took a beetle. The contrast here was greater between the beetles and the ashes, but the resemblance between the hay and the beetles was not very close as the latter were a glossy black and the hay dull black. Experiment 18. Figures 27 and 30. Two Silpha were placed on light ashes and two on charred wood, the former combination presenting much the greater contrast. In one minute the grackle dropped to the ground and began feeding. In two minutes it passed the beetles on the black background and took both from the white, taking no apparent notice of the former. Experiment 19. Same as Experiment 18, except that only one beetle was placed on each background. In one minute the grackle flew to the ground beside the white ashes from which it took the beetle. It then passed the black background once or twice taking no apparent notice of the insect upon it. After fourteen minutes the latter was taken, but I am uncertain whether by the kingbird or the grackle. Experiment 20. Same as Experiment 19. The grackle immediately flew to the ground and approaching the white background first took the insect from it. It then took the insect from the black background. Experiment 21. Figures 27 and 45. One Silpha on charred wood and one on flour, the latter combination presenting the greater con- trast. The grackle at once approached the backgrounds in a fairly direct line, passing nearer the wood, but taking the beetle from the flour. It then took the beetle from the wood. Experiment 22. Figures 836 and 60. One moth (Noctuid sp.) was placed on a piece of bark which was partly covered with damp ashes producing a background closely resembling the moth, and one on a strip of very light colored wood.2? The body of the insect was in- serted in a crack in the bark so as to bring its partly expanded wings close to the surface of the latter, but not in anyway concealing it. The grackle flew from one perch to another and then returned. It then flew to the ground and fed for a few minutes. In ten minutes it ap- proached the backgrounds, passed close to the moth on the bark, and took that from the wood. It was seemingly a little suspicious of the 25 A piece of an ordinary berry box was used. EXPERIMENTS ON PROTECTIVE COLORATION 483 latter, as it dropped it once before eating it. It then turned back and took the moth from the bark. Experiment 23. Figures 40 and 42. One Gryllus was placed on ashes and one on a mixed background of charred and uncharred wood, the latter combination presenting a close resemblance, and the former a good contrast. The grackle immediately dropped to the ground and began feeding. It soon went to the backgrounds, passing nearer the charred wood, but taking the cricket on the ashes. It then turned “and walked over the charred wood, passing directly over the cricket upon it, but apparently not seeing it. Experiment 24. Same as Experiment 23, except that in the former both crickets were probably in shadow,”* while in this experiment they were in the sun. The grackle immediately dropped from its perch to the ground and approached the backgrounds about midway be- tween them, paused a moment and seized the cricket on the ashes, after which it turned and took that on the wood background. The prompt approach of the grackle to the backgrounds in this experiment apparently indicates that it realized that food had been prepared for it there. Its attention however was not I believe attracted to one back- ground more than to the other in the preparation of the experiment. Its pause for a moment after reaching the backgrounds, and the prompt seizure of both insects suggests that both were seen as it approached, and that it was a matter of chance, or possibly of some individual preference on the part of the bird as to which was taken. Experiment 25. Figures 51 and 53. One green Melanoplus was placed on grass and one on charred wood. For three-quarters of an hour the insects were untouched, although the grackle several times went to within a short distance of the backgrounds. For seven min- utes the observations were discontinued. Soon after resuming them, the grackle once again approached to within about 3 cm. of the back- grounds but did not feed. Then it again approached, coming nearer the charred wood, from which it seized the insect, and then immediately took one from the grass. The charred wood combination presented the greater contrast. That the insects were left untouched for so — long a time in this-experiment, because of the bird not being hungry, is improbable, as it was pecking at objects on the bottom of the cage during this time, and when/’one insect was finally taken the other was immediately taken also. Further, in the following experiment (26), the grackle took the insects very soon after the experiment was started and within about ten minutes of the last feeding Gn Experiment 25). Experiment 26. Same as Experiment 25. The grackle immediately dropped to the ground and approached the backgrounds. It turned back for a moment and then re-approaching between the two back- grounds, took the insect on the charred wood and immediately after, that on the grass. Experiment 27. Figures 49 and 51. One green Melanoplus was placed on a mixture of grass and straw (lengthwise on a straw so as 26 On this point my notes are uncertain. 484 R. T. YOUNG to more closely resemble its surroundings), and one on charred wood, the latter combination showing the greater contrast. Before I had left the cage after placing the insects, the grackle crossed from the opposite end of the cage to the backgrounds and took the insect from the charred wood, leaving that on the grass and straw untcuched. This was left in position and a few minutes later it too was taken. Comparing Experiment 25 above with Experiment 27, one is impressed with the influence which the attention of the bird exercised on the rapidity with which the results were obtained. In the former experi- ment, the insects were apparently unseen for over fifty minutes, al- though during this time the bird several times came near them; while in the latter they were taken immediately, due in all probability, to the fact that the bird realized that food was being prepared for it on the backgrounds. Why its attention was attracted more readily in one experiment than in another is uncertain. I shall refer to this question later (p. 493). Experiment 28. Figures 27 and 61. One moth (Noctuid sp.) was placed on brown leaves and bits of bark and one on charred wood within about 7 em. of each other. The former moth closely resembled its background, while the latter combination presented a good contrast. Ia three minutes the grackle dropped from its perch to the ground and walked directly to the backgrounds, passing by the moth on the leaves. Ii then paused for a few seconds to inspect the moths, before seiz- ing that on the charred wood, immediately followed by that on the leaves. Experiment 29. Figures 26 and 27. Same as Experiment 28, except that a background of straw was substituted for the leaves and bark, the moth-wood combination presenting the greater contrast. The grackle immediately dropped to the ground and walked directly towards the backgrounds, but its attention being apparently diverted by some object outside of the cage, it ran past, returning on the side of the straw, from which it seized the moth, and then that on the wood. Experiment 30. Figures 27 and 59. One moth (Noctuid sp.) was placed on charred wood and one in an angle of a dead leaf so that the wings overlay it, with the head and thorax projecting over the ground, thereby reducing the relief and enhancing the resemblance of the insect to its background The insects were placed 6 or 7 cm. apart. In ten minutes the grackle dropped to the ground and approached the back- grounds on the side of the latter moth, but passed it by and seized the former, then turning, it apparently was about to seize the latter when I interfered. Experiment 31. Figures 27 and 40. One Gryllus was placed on charred wood mixed with a little earth, and one on ashes. The former combination presented to my eye a close resemblance. The grackle was about 1.6 m. distant during the arrangement of the experiment. As I was leaving the cage it went directly to the backgrounds and approaching the charred wood first, seized the insect upon it, immedi- ately followed by that on the ashes. EXPERIMENTS ON PROTECTIVE COLORATION 485 Experiment 32. Same as Experiment 31, except that the grackle was further distant from the backgrounds at the beginning of the experi- ment. The result was the same as in Experiment 31. Experiment 33. Figures 27 and 60. One moth (Noctuid sp.) was placed on gray bark and one on charred wood, the former presenting, to my eye, a fairly close resemblance to its background, and the latter a good contrast. After flying across the cage a few times the grackle dropped to the ground and a minute later went directly to the back- grounds, passing the charred wood and taking the insect from the bark, followed by that on the charred wood. Summary In 15 out of 33 experiments, or 45 per cent, the combination of less contrast was chosen and in 18 out of 33, or 55 per cent, that of greater contrast. In 15, or 45 per cent, the prey was taken from that background nearest which the bird happened to alight. A further analysis of these apparently inconclusive results will be reserved for later discussion. SERIES VIII Experiment 1. Figures 25 and 37. Five Melanoplus were placed on sand and five on a mixed background of hay and earth, the former combination presenting the greatest contrast. Almost immediately the kingbird flew direct to the sand from which it took one insect. Experiment 2. Same as Experiment 1. In three minutes the same result was obtained. Experiment 3. Same as Experiment 1. In one minute the king- bird flew over the sand alighting on a box about 2.5 em. from the back- grounds. Here it remained a few moments when the same result was obtained. Experiment 4. Same as Experiment 1. The kingbird immediately flew to the sand from which it took one insect, and then alit at a point nearer the hay than the sand. It quickly returned and first seized one insect on the hay and then the remaining four on the sand, leaving four on the hay. The results in the three preceding experiments may have been modi- fied by the memory of the bird’s experience in Experiment 1, in which it found insects on the sand. In Experiment 4 however even after it fed from the hay, it left four insects on the latter and took four from the sand, tending to show that this was not the case. Experiment 5. Figures 35 and 58. One Ceresa bubalus was placed on grass and one on sand, the latter combination showing the greater contrast. Before I had time to leave the cage the kingbird flew to the sand, from which it took the insect. 486 R. T. YOUNG Experiment 6. Same as Experiment 5. In twelve minutes the king- bird flew direct to the sand from which it took the insect. It.then looked closely at the insect on the grass for a few seconds and then took it also. Experiment 7. Figures 25 and 30. Three Melanoplus were placed on backgrounds of hay mingled with earth and three on ashes, as shown in the accompanying diagram (fig. D). The latter combination showed the greater contrast. The kingbird flew to the backgrounds, while I was standing nearby, going first to 5, but not feeding. It then took the i 2 3 4 5 Fig. D 1, 3, and 5—ashes; 2, 4, and 6—hay—earth. 1 ff, Wi pea 7TH Bae tah F Eat xX Burnt Paper Leaves and Straw Fig. E g—Gryllus, z—Melanoplus. Experiment 8. Same as Experiment 7. The kingbird flew direct to 5, which it took followed by 17. The result in Experiment 8 may have been modified by that in Experiment 7, as both were alike. Experiment 9. Figures 24, 30, 40 and 43. In this experiment three backgrounds were prepared, the arrangement of which, with the in- sects placed on each is shown in the accompanying sketch (fig. E). The leaf and straw background was in weak sunlight, while the ashes were in shadow. The insects on the ashes presented the greatest contrast to their background in this experiment. Within two minutes, and while I was still standing near the backgrounds, the kingbird flew direct to the ashes from which it took both Melanoplus. Experiment 10. Same as Experiment 9, except that all the back- grounds were in shadow at the time of experiment. At the com- mencement of this experiment the kingbird was perched on the side of EXPERIMENTS ON PROTECTIVE COLORATION 487 the cage.2’ In thirty seconds it flew direct to the ashes from which it took one Melanoplus, returning to its usual perch.