....V. V,
i » » i i § Jrt • rip
•x*:*#x*:
mmm^
•11 111 I
; ' : • :
:3
i: I
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
E. G. CONKLIN, Princeton University CARL R. MOORE, University of Chicago
E. N. HARVEY, Princeton University GEORGE T. MOORE, Missouri Botanical Garden
SELIG HECHT, Columbia University T H MORGAN, California Institute of Technology
LEIGH HOADLEY, Harvard University G R PARKER Harvard Uaiversity
L. IRVING, Swarthmore College
M. H. JACOBS, University of Pennsylvania A' C" R^ELD, Harvard University
H. S. JENNINGS, Johns Hopkins University F- SCHRADER, Columbia University
FRANK R. LILLIE, University of Chicago DOUGLAS WHITAKER, Stanford University
H. B. STEINBACH, Washington University Managing Editor
VOLUME 88
FEBRUARY TO JUNE, 1945
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE 8t LEMON STS.
LANCASTER, PA.
11
THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Penn- sylvania.
Subscriptions and similar matter should be addressed to The Biological Bulletin, Marine Biological Laboratory, Woods Hole, Massachusetts. Agent for Great Britain: Wheldon and Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2. Single numbers, $1.75. Subscription per volume (three issues), $4.50.
Communications relative to manuscripts should be sent to the Managing Editor, Marine Biological Laboratory, Woods Hole, Massachusetts, between July 1 and October 1, and to the Depart- ment of Zoology, Washington University, St. Louis, Missouri, during the remainder of the year.
Entered as second-class matter May 17, 1930, at the post office at Lancaster, Pa., under the Act of August 24, 1912.
\STF.R PRESS, INC., LANCASTER. PA.
CONTENTS
No. 1. FEBRUARY, 1945 PAGE
VON BONDE, CECIL
The external development of the banded dogfish or pofadderhaai Haplo- blepharus edwardsii (M. & H.) 1
BURT, RICHARD L.
Narcosis and cell division in Colpoda steinii 12
CRESCITELLI, FREDERICK
A note on the absorption spectra of the blood of Eudistylia gigantea and of the pigment in the red corpuscles of Cucumaria miniata and Molpadia intermedia 30
FRINGS, HUBERT
Gustatory rejection thresholds for the larvae of the cecropia moth, Samia cecropia 37
GREGG, JAMES H.
Background illumination as a factor in the attachment of barnacle Cyprids 44
MORGAN, T. H.
The conditions that lead to normal or abnormal development of Ciona 50
SCHRADER, FRANZ
Regular occurrence of heteroploidy in a group of Pentatomidae (Hemip- tera) 63
WATTERSON, RAY L.
Asexual reproduction in the colonial tunicate, Botryllus schlosseri (Pallas) Savigny, with special reference to the developmental history of intersiphonal bands of pigment cells 71
SERIAL LIST OF PUBLICATIONS HELD BY THE MARINE BIOLOGICAL LABORA- TORY LIBRARY AND THE WOODS HOLE OCEANOGRAPHIC INSTITUTION ADDITIONAL TITLES 105
No. 2. APRIL, 1945
MAC&NITIE, G. E.
The size of the mesh openings in mucous feeding nets of marine animals. 107
LYNN, W. GARDNER AND THEODOR VON BRAND
Studies on the oxygen consumption and water metabolism of turtle embryos 112
SCOTT, SISTER FLORENCE MARIE
The developmental history of Amaroecium constellatum. I. Early embryonic development 126
RICE, NOLAN E.
Pelomyxa carolinensis (Wilson) or Chaos chaos (Linnaeus)? 139
58812
iv CONTEXTS
PAGE
HIXTOX, TAYLOR
A study of chromosome' ends in salivary gland nurk'i of Drosophila. . . 144
SCHXEIRLA, T. C.
The army-ant behavior patUTii: Nomad-statary relations in the swarmers and the j)rohlem of migration 166
DOUDOROFF, PETER
The resistance and acclimatization of marine fishes to temperature changes. II. Experiments with Fundulus and Atherinops 194
No. 3. JUNE, 1945
\YILMER, CHARLES G.
Origin and function of the protoplasmic constituents in Pelomyxa carolinensis 207
vox BONDE, CECIL
Stages in the development of the picked or spiny dogfish, Squalus acanthias Linn 220
SCRIMSHAW, XEVIN S.
Embryonic development in Poeciliid fishes 233
FRIEDLAND, BEATRICE AND MORRIS HENRY HARNLY
The effect of temperature on the wings of dimorphos/dimorphos ves- tigial-pennant/vestigial in Drosophila melanogaster 247
SPIEGELMAN, S. AND H. B. STEINBACH
Substrate-enzyme orientation during embryonic development 254
STAUBER, LESLIE A.
Pinnotheres ostreum, parasitic on the American oyster, Ostrea (Gry- phaea) virginica 269
Vol. 88, No. 1 February, 1945
THE
BIOLOGICAL BULLETIN
PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY
THE EXTERNAL DEVELOPMENT OF THE BANDED DOGFISH OR POFADDERHAAI HAPLOBLEPHARUS EDWARDSII (M. & H.)
CECIL VON BONDE Government Marine Biologist and Director of Fisheries, Cninu of South Africa
INTRODUCTION
This oviparous species of dogfish, previously known as Scylliorhinus cdi^ardsii (Cuv.), is endemic to South African waters and has been recorded from Saldanha Bay, Table Bay, False Bay and the Agulhas Bank. The largest one on record measured 520 mm. This dogfish is fairly plentiful and specimens are on exhibi- tion in the Sea Point Aquarium.
On November 5, 1942, a female was observed to lay two egg cases in one of the tanks of the Aquarium. One case was immediately removed and dissected to ex- pose the egg, whilst the other was left in the tank as a control in the determination of the duration of incubation.
The egg was completely removed from the egg case and placed under observa- tion in running sea water.
THE FEMALE REPRODUCTIVE ORGANS
The only external sexual characters in the female are the pair of cloacal papillae (Plate I, c.p.), one situated on either side of the median line immediately posterior to the cloacal aperture. The papillae themselves are perforate, and the abdominal pore situated in the center of each papilla connects direct with the coelom. The actual function of these abdominal pores is not known and it is obvious that, in view of the highly specialized structure of the female genital system, they have lost the function of acting as apertures through which the ova leave the body.
A dissection (Plate I) revealed the fact that there was only a single median ovary (ov.) present which contained ova in various stages of development. The ovary is situated in the middle portion of the coelom and is relatively large. The paired oviducts are highly specialized both in structure and function. The oviducts meet medially in the anterior part of the coelom (f.t.o.). Each oviduct may be divided into four distinct portions as follows :
(1) The fallopian tube (Plates I and II, f.t.) is extremely narrow and thin- walled and is about a quarter of the length of the whole oviduct. Its internal surface is slightly convoluted longitudinally. This portion is followed by (2) the albumen gland (a.g.) which is about one-eighth of the total length of the oviduct and is very thick-walled and muscular. This gland secretes the albumen which sur-
CECIL VON BONDE
rounds the ovum, hence its internal surface is richly supplied with glands for this purpose. A transverse section of the albumen gland (Plate II, Fig. 3, l.a.g.) shows that the lumen is extremely narrow, the walls of the gland being completely ap- posed. Smith (1942, p. 703) in describing Bashford Dean's figure of the repro- ductive organs of an adult female Hctcrodontns japonicns states— ''This drawing (my Text-figure 35) is not labelled, nor is it described in Dean's notes, and in the absence of the dissection some features are obscure. In the mid-line near the top of the figure, one readily notes the common abdominal opening of the oviducts. On the extreme right side of the figure the oviduct with its three divisions — oviduct proper, shell gland, and uterine portion — are easily identified." The part of the oviduct designated "shell gland" in the above quotation is very similar in external appearance to that of the present species, but as previously stated, this swollen portion functions as the albumen gland in the present species. There are no in- dications of a laminated structure in this gland as described by Gudger (1940, p. 550) for the nidamental gland of Chlamydoselachus. Immediately following the albumen gland is (3) the nidamental gland (n.g.) which is also about one-eighth of the total length of the oviduct and secretes the shell around the ovum. The interior surface of the nidamental gland is extremely well convoluted and is of a dark yellowish color. Widakowich (1907, p. 527) states— "Das Nidamentalorgan von Acanthias besteht aus einem cranial gelegenen Teile, der, wenn man nach Ana- logic mit Scyllium schliessen darf, Eiweiss erzeugt, aus einem mittleren, der, wie sich zeigen lasst, die Schale liefert, und aus einem untersten, wohl Schleim pro- duzierenden Abschnitte." As the functions of the various parts of the oviduct of the present species, however, are distinctly diverse, it appears to be better to look upon each part as an independently functioning entity and parts two and three as here described may be looked upon as analogous to the cranial portion, and the central and caudal portions of the nidamental gland, respectively, as described by Widakowich. The last part (4) of the oviduct is the vagina (v.), occupying about one-half of the total length. This portion of the oviduct is designated the vagina in analogy to the use of this term in mammals in view of the fact that the intromit- tent organs (claspers) of the male are inserted through the vaginal orifices during copulation.
Each oviduct opens separately by a vaginal orifice (Plate II, l.v.o. and r.v.o.) into the cloaca. The vagina is well convoluted longitudinally to permit of disten- sion both during copulation and during ovulation. In certain of the females dis- sected, parts of the mermaid's purse were present in the upper half of the vagina where it leaves the nidamental gland. These parts consisted principally of the anterior and posterior parts of the mermaid's purse complete with the tendrils (see Plate IV, Fig. 6).
PLATE I
Dissection showing the female reproductive organs of Haploblcphants ahwdssi (M. & H.).
a.g., albumen gland, c.p., cloacal papilla, e.g., external gill-filaments, f.t., fallopian tube, f.t.o., coelomic orifice of same, h., hooks of clasper. h'., hooks normally covered by rhipidion. hy., hypopyle. l.a.g., lumen of albumen gland, l.v.o., left vaginal orifice, n.g., nidamental gland. ov., ovary, r., rhipidion. r.v.o., right vaginal orifice (opened), u., umbilicus \vith remains of cord, u.c., umbilical cord, v., vagina, v.b.v., vitelline blood vessel, y.s., yolk sac.
PLATE I
rr <• - (
\" • ' \
\ I
i i 4 — '
GEGIL VON BOXDE
Tin-: MALI- KKI'ROIUVTIVK ( )KC.ANS
1 he morphology ot the male1 reproductive system follows the general plan of all Elasmobranchs. The testes are paired and occupy the anterior half of the cot-loin. From the anterior end of each testis the vasa efferentia lead to the epi- didymus which is much coiled. The epididymus eventually widens out into the \'as deterens which expands and joins its fellow in the median line posteriorly, opening into the cloaca by a single pore.
The secondary sexual characters of this species are very distinct and it is nceessarv to describe, them in more detail. As in all male Elasmobranchs, the basal element of each pelvic hn (basipterygium) is prolonged to form a stout, back- \\-ardIy-directed rod which is sharply demarcated from the remainder of the hn and specially modified to ionn an injtromittent organ commonly known as the clasper or myxopterygium. In this species the clasper is a very highly differentiated organ which differs in many respects from that found in other Flasmobranchs. In trans- verse section the clasper is almost completely round in its proximal half, the in- ternal cartilagenous skeleton forming a tube which is almost completely tilled with a thick muscular substance through the center of which passes a duct. In the distal part the skeleton only occupies the ventral portion of the clasper, being continued right to the tip. The latero-dorsal wings of this portion of the clasper are formed ot thickened skin, the outer wing being folded over the curved inner wing so as to lorm a canal which is a continuation ot the duct in the proximal portion previously mentioned. This canal opens on the dorsal side of the posterior end of the clasper in the form of a hypopyle ( Plate III, Fig. 4. by. ). Schmidt (1930) first noted the existence of hooks in the clasper of Halddiinis 1(>razania which is a genus closely related to the present one and he noted that these hooks were approximately 100 in number. The clasper of the present species also shows approximately 100 hooks on the outer lateral wall of the clasper ( h. and Fig. 5 ).
Schmidt states— -"Considering now what purpose this arrangement can serve, one comes to the conviction that the hooks can be used only for fastening the clasper on the wall of the vaginal part of the oviduct of the female during the copulation. Probably, when the clasper is introduced, the borders of its gutter are turned out and all the hundred hooks of the row are imbedded in the oviducal wall, holding the clasper as anchors. If the other clasper is introduced simultaneously in the other oviduct, the male is held by 200 little anchors. This arrangement may be necessary, as the female of this shark is larger than the male, which is perhaps car- ried about by the female during a copulation that may continue for a very long time.
"This arrangement of the organs of copulation of the male seems to be unique not only among the fishes, but pei haps in the whole animal kingdom." but be makes no mention of the other series of books ( h'. ) which lie on the inner lateral wall ot the clasper and which are covered by a rhipidioii (r. ), whose function is to spray the spermatozoa in all directions in a radiating manner, under the pressure exerted
PLATE II
The lett oviduct opened medially to sho\v the internal structure.
1. Longitudinal section through the fallopian tuhc and the albumen inland K 1. Longitudinal section through the nidamental inland and the vagina. K ,v Sagittal section throu.sdi ihe albumen eland.
PLATE II
••
6 CECIL VOX I'.OXDK
by tbt' water lovably ejected bv tin- siphon. The reason for this is obvious since fertilization takes place in the fallopian tubes and the sperniato/oa. although motile, have to traverse the complete- length of the vagina, nidamental gland, and the albumen inland to reach the ova in the anterior end of the fallopian tubes.
Leigh-Sharpe ( 1('20) in describing the secondary sexual characters of Scvlliuin Cittnlits mentions the presence of a siphon on the ventral surface of each pelvic fin. The present species also has such a siphon developed in the same position, although it is probably more analogous to the clasper gland mentioned by Leigh-Sharpe (op. cit.. ]>. 260) than to the siphon. The whole of the clasper except the distal end is covered with dermal denticles. A deep dissection of the clasper showed that it is in no way innervated.
DEVELOPMENT
In order to study the external development of this species a mermaid's purse was opened, the ovum removed from the purse and placed in a dish of running sea water. The albumen was completely removed so as to enable one to observe the development of the embryo.
The egg case or mermaid's purse (Plate IV. Fig. 6) is more or less typical of those formed by oviparous dogfishes. The anterior end is broad and almost as wide as the main body and has a straight edge where the two halves have partially coalesced. From each lateral end a long coiled filamentous process, which grad- ually becomes thinner towards the distal end, arises. Near the base of each of these tendrils the purse has a longitudinal slit-like aperture (s.) which leads to the interior. These apertures only appear on one surface of the egg case. Posteriorly the egg case tapers until it is little more than half as wide as the anterior end. At the lateral extremities of this sharper end two processes arise which are the bases of the tendrils. These posterior tendrils are much longer and thicker than the an- terior ones at their points of origin, but. like the latter, they also become thinner towards their distal ends. Tt is this pointed end of the purse which first appears through the vaginal orifice when the egg case is laid. The function of the tendrils is to anchor the egg case securely to rocks or seaweed to prevent its being buffeted about by the waves. The egg case itself is formed from keratin secreted by the nidamental gland. \Yidakowich (1906) stated that the mermaid's purse of Scyl- liiiin canicuhi is formed by a large number of separate elements ("Flatten") which later, on exposure to sea water, adhere closely to form a complete shell (see also Hobson (1930, p. 5X0) ), but in Haploblcphants cdi^anisii there is no evidence of such a formation of the egg case which is here laid down completely in two halves, the dorsal and ventral parts, the edges of which coalesce'.
A dissection of the mermaid's purse shows that it is a structure formed ot two portions which mav be designated the dorsal and ventral halves, each hah being hollow, the edges being apposed and completely fused along their whole length except at the blunt anterior end where the apposition is purely temporary. It is
PLATE III
FH.IKK 4. Dorsal vim of the left clasper with lateral \\in.ys of distal portion distended to -how internal structure. ' 4.
Flf.L'KK 5. A series of 4( I hooks removed from the rlasprr. < 111.
PLATE III
CECIL VON BONDE
through this anteiior end tliat tin- llsli eventually escapes, hence the lack of fusion of these edges. The swollen central portion occupying ahont two-thirds of the internal space of the egg case, is Idled by the relatively enormous yolk. Surround- ing the yolk and occupying the remaining space of the egg case is the thick viscid transparent allminen. From the fact that during the artificial rearing of the embryo a\vav fiom the egg case all the albumen was removed from the yolk, it is obvious that this albumen plays no part in the development or nourishment of the embryo at any stage, and pttrelv serves the purpose of a protective cover around the embryo during its development. Its function could be regarded more or less as that of a shock absorber analogous to the function of the amniotic fluid in higher vertebrates. The egg itself is between 30 and 35 mm. in diameter and is typical of the majority of Selachian eggs in regard to the distribution of yolk, being telolecithal. On one surface of the egg, which mav be looked upon as the dorsal aspect, lies the germinal disc which is slightly lighter in color than the rest of the egg. has a diameter of 1.2 mm., and contains practically no yolk. Segmentation, as in all telolecithal eggs, is meroblastic since the cleavage planes are only restricted to the germinal disc and do not pass through the whole egg. The segmentation of the germinal disc is very similar to that of other Selachians and closely approximates that of the developing chick. The various later stages of gastrulation with the subsequent origin of the germinal layers is also typical of the class and it is not necessary to elaborate this feature. The folding oft" ot the embryo takes place very early during develop- ment, at about 10 days after the egg is laid, the embryo then being about 2 nun. in length and attached to the yolk sac by a very well developed umbilical cord. The vitelline blood vessels at this stage are extremely prominent and spread out over the whole surface of the yolk sac. These vessels or capillaries unite into t\vo median vessels which pass almost completely round the equator of the yolk ( Plate IV, Fig. 9) to a region which lies distal from the umbilical cord. Here these ves- sels break up into two branches running at right angles to the equatorial vessels. These equatorial vessels pass along the sides of the umbilical cord and enter the embryo's heart.
The vitelline circulation is also typically similar to that appearing in the majority ni oviparous Elasmobranchs and closely approximates that of the developing chick embrvo. The umbilical cord is solid and the vitelline circulation plays a very im- portant role in the transference of the nutritive tood material trom the yolk sac to the developing embryo. The absorption ot the yolk material progressively in- creases with the giowth of the embryo, with a concomitant decrease in the size of the yolk sac (cf., Plate IV, Figs. 7. X and (>). At an early stage in the develop- ment, the external gills, typical of most Selachian embryos, make their appearance
PLATE IV
Stages in the external development of //. cdMtnlsii. Fi<;ri<K d. Tin- r.!^ ca^e or mermaid's purse, natural si/e.
FK.I KK 7. A mermaid's purse opened to slio\v a developing embryo 25 days old, natural si/.e.
Fu.rKK S. A 30-day old embryo. < 1.5.
FIGURE (>. A 50-day old embryo. 1.5.
I;H. i KK 10. Ventral view of embryo at time of liatehin.n ( KM days old). < 1.5.
I-'K.I KI 1 1. Dorsal vie\\ < >i same.
All I'hotoizraplis by Autbor.
PLATE IV
CKC1I. VON IK >\I)K
7 and Si. In origin these external gill filaments are totallv unlike the external gills lound in CfOSSOpterygii, l)ipnoi, and anv Amphibia, and over their function and structure many xoologists have- been puxxled.
< iraham Kerr (l'M(>. p. 157) states— 'AYhile exteinal gills occur within three main sub-divisions of the vertebrates, namely, Teleostomatous fishes (Crossoptery- gians — the most archaic of existing Teleostomes ) . Lung-fishes, and Amphibians there are two main groups -Elasmobranchs and Amniotes — in which they are con- spicuous by their absence As it happens, however, there is in the two groups
mentioned a delinite cause which seems quite competent to account for disappear- ance ot exteinal gills, namely, the development of a new organ — a yolk sac with its highly developed vitelline net work of blood vessels — which in addition to its primitive function must necessarily also function as a very efficient organ of respiratory exchange and so render any pre-existing respirator}- organ no longer necessary."
In the present species these external gill filaments are richly supplied with blood vessels and in view of the fact that the embryo studied underwent its complete development in sea water outside of the mermaid's purse it is obvious that their pri- mary function must be a purely respirator}- one. The pair of slits in the mermaid's purse previously mentioned must serve for the conveyance of sea water into the interior of the purse where it probablv mingles with the layer of albumen sur- rounding the developing embryo. Although it has been stated that these gill fila- ments absorb fluids which are milk-like secretions of the uterine mucosa and which serve as food for the growing embryo of all non-placenta! viviparous sharks and rays (Gudger, 1940) in the present species this is not one of their functions. Beard (1890, p. 310) states — "In the Skate-embryo the filaments are said to dis- appear shortly before hatching. It may be expected that their atrophy commences when the purse ruptures sufficiently to allow of the passage of sea water directly to the embryo. Then the ordinary piscine mode of respiration would be initiated, and the external gills would disappear." In the present species, however, they have alreadv disappeared during the volk sac stage when the embryo is 50 days old and the yolk sac has decreased to about half its original sixe ( Plate IV, Fig. {) ) and the internal gills are by this time well developed and functional.
At 104 davs the embryo has absorbed all the volk and the volk sac has shrunk to such an extent that only a vestige of the umbilical cord remains extending through the umbilicus for a length of about 2 mm. ( Fig. 10. u.). At the time of hatching the ventral abdominal surface of the embryo between the umbilicus and the cloaca is richlv supplied with blood vessels which form a reticulation all over this surface (Fig. 10). The umbilicus persists for about 14 days after hatching and then com- pletely disappears and the fish is fully formed ( Fig. 11 ) and able to fend for itself, swimming about activelv. The gestation period of the embrvo which developed inside the egg case was also 104 days and the state of development at the time of hatching was the same.
SCM MA in 1. Haploblepharus <'(/(v</r</.v/Y is an oviparous dogfish endemic to South African
2. The male and female generative systems are dealt with and specialized lea- tures such as the albumen gland, nidamental gland, and the claspers are described.
DEVELOPMENT OF HAPLOBLEPHARUS EDWARDSII 11
3. Development takes place in an egg case or mermaid's purse, the gestation period lasting 104 days.
4. The development of an embryo external to the egg case is described.
LITERATURE CITED
BEARD, J., 1890. On the development of the common skate Raja batis. Fisheries Board Scat- land, 8th Ann.. Kept., Pt. Ill: 300-310.
DANIEL, J. FRANK, 1934. The Etasmobranch fishes. Berkeley, California.
GUDGER, E. W., 1940. The breeding habits, reproductive organs and external embryonic de- velopment of Chlamydoselachus, based on notes and drawings of Bashford Dean. Dam. Mem. Vol. Archaic Fishes, VII.
HOBSON, A. D., 1930. A note on the formation of the egg case of the skate. Jour. Mar. Biol. Assoc., U.K., 16, No. 2: 577-581.
KERR, J. GRAHAM, 1919. Text-book of embryology. II. Vertebrate it'ith the exception of Mammalia. Macmillan, London.
LEIGH-SHARPE, W. HAROLD, 1920. The comparative morphology of the secondary sexual char- acters of Elasmobranch fishes. Jour. Morph., 34, 2, Sept.
SCHMIDT, PETER J., 1930. A selachian clasper with a hundred hooks. Copeia, June 30, 2 : 48.
SMITH, BERTRAM G., 1942. The Heterodontid Sharks : The natural history and external de- velopment of Heterodontus japonicus, based on notes and drawings by Bashford Dean. Dean Mem. Vol. Archaic Fishes, VIII.
WIDAKOWICH, V., 1906. Uber Bau und Funktion des Nidamentalorgans von Scyllium canicula. Zeitschr. zviss. Zoo/., 80: 1.
WIDAKOWICH, V., 1907. Uber den Uterus von Squalus acanthias. Ibid.. 88: 4.
NARCOSIS AND CELL DIVISION IN COLPODA STEINII
RICHARD L. BURT '
Arnold Biological Laboratory, Broiim University, Providence, R. 1.
INTRODUCTION
For many years it has been known that a great number of chemical and physical agents can modify the course of mitosis and induce the development of abnormal division figures. Among the many experimental agents that have' been employed to upset normal cell division are the following: basic dyestuffs (Politzer, 1924), narcotics (Politzer, 1931; Nemec, 1904; van Regemorter, 1926; Shaklevich, 1938; Ludford. 1936; Geiersbach, 1939), ether (Hacker, 1900; Schiller, 1909; Rosen- feld. 1932), alcohol (Krantz, 1938), X-radiation (Albert! and Politzer, 1923; Pfuhl and Kiintz. 1939), radium (Whitman, 1933), ultraviolet radiation (Stevens. 1909), high and low temperatures (Bury, 1913; Bleier, 1930; Vintemberger, 1930; Kemp and Juul. 1931), and hypotonic solutions (Lewis, 1933). The instructive point to be gathered from a survey of such studies is that, irrespective of the agent em- ployed, the induced morphological changes in cell division are quite similar. Thus, within certain limits of concentration or intensity of the experimental agency, dis- orientation of the chromosomes in the division figure, delay in polar movement and scattering of the chromosomes, "amitosis," pyknosis. dissolution of the achromatic figure, inhibition of cytokinesis, and formation of hi- and multinucleate cells are common effects.
The relative lack of specificity in results produced by such diverse experimental treatment makes the formulation of any .complete explanation for these phenomena difficult. However, the suggestion of Ludford (1936) to the effect that metabolic changes may have a bearing on the production of anomalous divisional behavior becomes significant in light of recent observations indicating that special portions of the cell's metabolism are directly associated with certain physiological activity states (For details see Bocline (1934) ; Robbie. Boell and Bodine (1938) ; Deutsch and Raper (1938) ; Goddard and Smith ( 1938) ; Horowitz (1940) ; Pease (1941) ; Allen and Goddard (1938) ; van Schouwenberg (1938) ; Clowes and Krahl (1939) ; MacLeod (1941) ; Ormsbee (1941), and Fisher and Stern ( 1(>42) ). From these studies it becomes apparent that a definite relationship exists between a portion of the over-all metabolism and cell or tissue activity. The use of narcotics as respira- tory poisons and the deductions made by Fisher and his associates from data ob- tained by this type of treatment are of considerable significance in connection with this problem. The present study is an attempt to relate changes in cellular metab- olism during division with the appearance of abnormal mitoses. The results of this study are in essential agreement with those of Fisher et al in that a respira- torv parallelism can be demonstrated in Colpoda slcinii by use of ethyl carbamate or chloral hydrate. The "activity system" of Fisher (that portion of the overall
1 Present Address : I larvard Medical School. Hoston, Mass.
12
NARCOSIS AND CELL DIVISION 13
respiration that is most sensitive to narcotic inhibition) appears to be somehow associated with the maintenance of normal mitosis, for when it is differentially suppressed by appropriate narcotic concentrations, abnormal nuclear figures appear. These aberrant mitoses are believed comparable to those described in the literature that have been induced by many different agents. Barring such experimental intervention, the nuclear complex of Colpoda stcinii is remarkably stable, unlike certain other members of the family Colpodidae (Burt, Kidder and Claff, 1941). Because of this stability it was selected for the study to be described.
I am indebted to Professor George W. Kidder, Arnold Biological Laboratory, Brown University, for his stimulating interest during the course of this investi- gation.
MATERIALS AND METHODS
Colpoda stcinii was employed as experimental material throughout this investi- gation (for species designation see Burt. 1940). Although this ciliate cannot be readily grown in the absence of bacteria (food organisms) it offers several distinct advantages over other forms, notably, its rapid growth to high concentrations and the clarity of the mitotic changes during division (Burt. Kidder and Claff, 1941).
The ciliates were grown in sterile distilled water seeded with the coliform bacterium Acrobactcr cloacae. One liter Erlemneyer flasks were used as culture vessels. The procedure employed in culturing was as follows : 400 mis. of dis- tilled water were placed in the flasks which were then plugged and autoclaved at 15 Ibs. pressure for 20 minutes. Twenty-four hour Kolle flask cultures of A. cloacae on agar were used as food. The bacterial growth of two such Kolle flasks was harvested in 30 mis. of sterile distilled water, and 10 mis. of the resulting sus- pension were added aseptically by means of a pipette to each of two Erlenmeyer flasks containing the sterile distilled water. In this manner fairly uniform suspen- sions of food organisms were obtained. Stock cultures of C. stein ii were carried in tubes with Aerobacter as food. Each flask was inoculated with one ml. of these tube cultures containing 12 hour cultures of ciliates (in the logarithmic growth phase). The Erlenmeyer culture flasks were then incubated 24-30 hours at room tempera- ture before the ciliates were sacrificed. By adhering to this procedure an abundance of logarithmic actively dividing cells could be consistently obtained.
All culture vessels, pipettes and miscellaneous glassware employed in culturing the ciliates were autoclaved 20 minutes at 15 Ibs. pressure to prevent extraneous bac- terial contamination inasmuch as optimum growth of Colpoda stcinii obtains with Aerobacter alone as the food organism (Kidder and Stuart, 1939). Throughout all cultural procedure rigid bacteriological technique was followed to obviate the difficulties arising from contamination of the cultures.
In the cytological studies the cells were removed from the culture flasks and centrifuged at slow speed for four minutes in 50 ml. centrifuge tubes. The super- natant was then removed by means of an aspirator and more organisms added to the tubes from the culture. Following a second centrifugation, giving adequate numbers of organisms, the ciliates were placed in 250 ml. Erlenmeyer flasks con- taining 50-75 ml. of freshly bacterized solution of narcotic whose effect was to be tested. Equal amounts of bacterial suspensions were added to each experimental and control solution in making up the dilution of the reagent employed. When
14 RICHARD L. BURT
the desired time of treatment had elapsed (one hour) the cells were again packed, the supernatant removed, and the cells placed on coverslips by means of capillary pipettes. It was found convenient and satisfactory to employ 25 per cent acetic acid in absolute alcohol for fixation. The nuclear details were found well pre- served after fixation for five minutes in this reagent. The Feulgen technique was used exclusively in making the cytological preparations. This was necessitated by the fact that differentiation of the nuclear details is difficult with haematoxylin or other stains because of the retention of the dye by the division cyst walls. Best staining results were obtained by hydrolyzing in N/l HC1 at 60° C. for 12 minutes and staining in the fuchsin sulfurous acid for six hours. The stained preparations are then washed three times for four-five minutes each in HCl-Na-Bisulfite solution, placed for !/•> hour in running tap water, passed through the lower alcohols to 95 per cent and stained with fast green, dehydrated in absolute alcohol, cleared in xylol and mounted in damar.
It was customary to run these cytological experiments in four sections as follows : by centrifuging in four tubes the contents of the mass culture flasks were roughly quartered. The organisms obtained by centrifugation from three of these tubes were employed experimentally and accordingly treated with various concen- trations of reagent. The cells from the fourth tube were always set up with bac- terized distilled water and thus served as controls for the treated organisms. The control ciliates were always mechanically treated in a manner identical with that received by the experimental organisms. By this method aberrations in the mitotic or divisional processes could always be attributed safely to the effect of the inhibitor used.
For the respiration studies two 24-30 hour cultures set up as described were centrifuged and the cells so concentrated were resuspended in bacterized M/200. phosphate buffer at pH 7.0. One and one half mis. of this cell-buffer suspension was added to the Warburg vessels. Following equilibration (15 minutes) four read- ings were taken at ten minute intervals to determine the normal uninhibited respira- tion. The graded concentrations of inhibitor to be tested (made up in M/200 phosphate buffer at pH 7) were then dumped from the side bulbs and following a second equilibration period of ten minutes readings were again taken at ten minute intervals for one hour to determine the extent of respiratory inhibition. Tempera- ture was controlled at 25° C. ± 0.1° C.
It was considered advisable to set up three vessels with M/200 phosphate buffer in the side bulbs to serve as controls for the experiments. In addition to these normal, controls two more vessels were run with 1.5 ml. suspension of Aerobacter equivalent to the bacterial suspension employed in the experimental vessels. The final computed values for oxygen consumed per hour could therefore be corrected for not only an increase or decrease in normal control respiration but also for the negligible amount of oxygen consumed by the food organisms both before and after inhibition. The fact that organisms of the genus Colpoda encyst when the supply of food bacteria is depleted (Taylor and Strickland, 1938; Kidder and Stuart, 1939) necessitated the addition of bacteria to the experimental sus- pensions. Because of the presence of the bacteria, absolute values of oxygen consumed by the ciliates would be very difficult if not impossible to ascertain. However, where only relative rates of respiration are sought, as in this instance, the food organisms introduce no serious technical difficulty.
NARCOSIS AND CELL DIVISION 15
A simple technique was employed as a means of testing the degree of growth suppression in different concentrations of inhibitor. Although this method lacks the refinement of a cell counting technique, the results are believed to be roughly comparable. In these experiments four tubes were set up for every narcotic solu- tion to be tested. The final volume of the narcotic solution following bacterization was 3 mis. In addition to this series of four tubes for every narcotic concentration tested through the range under investigation, four more tubes were set up with bac- terized distilled water to serve as controls. Following inoculation with one loopful of logarithmic Colpoda, the growth in each of the four inhibited series and controls was read at 12, 24, and 36 hours. Control growth was arbitrarily designated as four plus, and growth in the experimental tubes designated as three plus, two plus, one plus, plus-minus, and minus depending on the degree of inhibition. The range of inhibitor concentration where growth was completely suppressed was termed the zone of complete growth inhibition. In these experiments aseptic procedure was not adhered to in either experimental or control tubes as over the period of time these determinations were made it was felt that extraneous bacteria would not significantly modify the results.
In assaying the effectiveness of the narcotic concentration tested in disrupting mitosis only cells in which the polar migration of daughter micronuclei was obvi- ously retarded were counted as affected (compare Fig. \-H with Fig. l-£). As will presently be pointed out in detail, Colpoda stcinii usually divides twice within a division cyst wall to produce four daughter ciliates. The aberrant nuclear di- visions were accordingly expressed as per cent of all cells counted in mitosis be- tween the first metaphase and the telophase of the second division. This method offered a reliable means of obtaining a quantitative estimate of the damage induced by the agents employed. It is to be emphasized, however, that the delay in polar migration of the daughter micronuclei was not the only observable defect resulting from narcosis. It merely served as a convenient method of evaluating the extent of mitotic derangement. The other changes concomitant with the polar defect will be considered elsewhere.
Potassium cyanide, potassium ferricyanide, sodium arsenite, iodoacetate, vari- ous carbamates and chloral hydrate were tested at varying concentrations for their effect on the division mechanism. Ethyl carbamate and chloral hydrate were most thoroughly studied in this respect and were also selected as inhibitors for the respiration studies.
OBSERVATIONS Normal division
The organisms of the genus Colpoda normally reproduce within division cysts (for details see Kidder and Claff, 1938; Burt/Kidder and Gaff. 1941). The trophic ciliates (Fig. l-A} round up prior to division, dedifferentiate and secrete the cyst wall within which the ensuing fissions occur, C. stcinii usually dividing twice to form four daughter cells. These stages are depicted in Figure 1, A-G. The first changes in the nuclear complex marking the onset of division are to be found in the macronucleus. This organelle loses its trophic ellipsoidal shape, becomes rounded, and the polar chromatin aggregates break up into irregular masses attaining the general configuration shown in Figure \-B. Meanwhile the
16 RICHARD L. BURT
micronucleus proceeds through the prophasie changes first marked by swelling then condensation of chromatin to form striae from which the chromosomes are formed. These micronuclear transformations culminate in the metaphase configuration shown in Figure \-B with the chromosomes oriented parallel to the long axis of the spindle. At this stage the macronuclear chromatin may be considered as divided into two portions, one consisting of the irregular masses centrally disposed ; the other, being peripherally located and in optical section, seems to be plastered to the macronuclear membrane giving rise to a beaded appearance. The anaphase is depicted in Figure \-C. At the termination of metaphase, the division figure ap- pears to break into two parts followed by the rapid movement of the daughter halves to polar positions at opposite sides of the macronucleus (Fig. \-D to E). Upon displacement of the daughter micronuclei by 180° the macronucleus elongates and constricts centrally. Cytokinesis soon follows at the completion of which the daughter micronuclei immediately pass into prophase and the same series of events is repeated with the result that four daughter cells are produced as shown in Figures l-F and l-G. Motor and oral organelles are differentiated in the four daughters and swimming movements are taken up within the cyst wall until finally the cyst membrane is ruptured and the ciliates escape into the surrounding medium. No visible nuclear extrusion has been observed during divisional phases in this species (Burt, Kidder and Claff, 1941).
Experimental modification of mitosis
Striking differences were observed in the effectiveness of the various inhibitors employed in disrupting the divisional mechanism. Ethyl carbamate was found to be quite active in this respect, however, and because of various other advantages offered by this compound, notably its high solubility and effectiveness over a relatively wide concentration range, it was most extensively studied. Potassium cyanide, potassium ferricyanide, sodium arsenite (neutralized solution), and iodoacetate were without effect up to toxic concentrations. Related carbamates produced changes similar to those induced by ethyl carbamate and the results of treatment with chloral hydrate are believed to be entirely comparable to those caused by inhibition with the urethanes. In the following account the cytological changes induced by these substances will be considered.
Ethyl carbamate: Striking changes in the course of mitosis were obtained in concentrations of ethyl urethane ranging from 0.5 to 2.0 per cent. The treat- ment did not appear to affect a particular stage of mitosis, however, but inhibition of any divisional stage in progress at the time of exposure seemed to obtain. A similar observation was made by Ludford ( 1936 ) for this narcotic. In view of
FIGURE 1. Colpoda stciuii '.< 1400. A, trophic organism; B-G, normal division phases; H-L, aberrant divisions produced by 1.3 per cent ethyl carbamate. B, Micronucleus in full metaphase of first division. C, Micronucleus in anaphase. D, micronucleus moving to poles. E, Micronucleus continuing polar movement; macronucleus elongating. F, Micronucleus at metaphase of second division. G, Second division completed. Nuclei of daughter cells return- ing to trophic condition. H, Failure of micronuclear migration ; macronuclear chromatin aggre- grates more diffuse ; reappearance of Binnenkorper. /, Similar to H : arrested polar movement of micronuclei; macronuclear chromatin diffuse; Binnenkorper material reappearing. /. Seg- rrLjation of both daughter micronuclei into one cell following first division. A". Arrestment of division at micronuclear metaphase. L, Both micronuclei at one pole of elongated macronucleus.
NARCOSIS AND CELL DIVISION 17
*«>.,
<7'ij*
B
C
, X
1» . . *, #
&$ i ffj*
D
G
H
'•*•
! •*
!
L
FIGURE 1
K
'•*•"
RICHARD L. P.TRT
this I act, it is not at all surp. is'.rg that a great variety <>f aberrant divisional types \vere observed. P>ecause of this almost infinite variation in the nuclear features associated with narcosis, it is exceedingly difficult to evaluate the activity of the narcotic on any other basis than its effect on polar movement of daughter micro- nuclei.
\\ ith ethyl nrethane this polar defect is exhibited through a concentration range "i 0.5 to 2.0 per cent. Thousands of cells were counted following treatment with various narcotic concentrations (for complete data on this and the following ob- servations see Hurt. 1942). In Figure 2 these data are plotted semilogarithmically. The defective polar movement increases from 0 at 0.5 per cent urethane to a maxi- mum of around 21 per cent at 1.5 per cent urethane concentration. This maximum value is probably determined by the number of ciliates in the first or second divisions whose micronuclei are between metaphase and telophase. During these stages one aspect of inhibition of cell division is expressed by lag in poleward movement.
As pointed out, however, the effect of urethane in appropriate concentrations on dividing C. stcinii is characterized not only by delay of polar movement, but also by changes in the state of aggregation of both macro- and micronuclear chromatin and also by abnormal configurations of the dividing nuclear complexes. Some of these changes are illustrated in Figures \-H to \-L.
In Figure I-// the polar defect is very well shown. At this phase of division, judging by the elongation of the maci onucleus, the daughter micronuclei should have attained polar positions in the cell as is shown in Figure 1-7: , a normal division figure. On the contrary, the daughters are centrally located near the middle of the macronucleus. A somewhat similar situation obtains in Figure I-/, although the appearance is by no means as striking. In Figure 1-L another possible orienta- tion of the nuclear complex is shown. Here the daughter micronuclei are termi- nally located at the same pole. One result of such aberrant behavior on the part of the daughter micronuclei is illustrated in Figure I-/. In this instance cyto- kinesis and division of the macronucleus has occurred, but the micronuclei, due to lailure in assuming polar positions, have both been segregated into one daughter cell. In Figure 1-A" inhibition of division has occurred at the metaphase.
In all of these cases the aggregation of the macronuclear chromatin has been changed to some extent. In general, most of the macronuclear chromatin in the treated cells exhibits a tendency to become plastered to the nuclear membrane rather than maintain a central disposition characteristic of the normal dividing nucleus. Concurrently, the aggregates become more diffuse and varying degrees of fusion are shown ( Fig. 1 -/ ) .
A noteworthy feature of the urethane-treated cells is the behavior of the Binnen- korper or plasmasoinal material. Normally this substance, which stains lightly with last green in tropic ciliates. is not in evidence in preparations immediately after the onset of nuclear division. 1 lowever, in narcotized cells it reappears and is lotind dispersed around the diffuse chromatin aggregates ot the macronucleus.
The morphological behavior of the chromatin and plasmasoinal material in narcoti/ed cells is .suggestive of a partial reversion to a trophic nuclear condition. It is possible that tinder the influence of the narcotic and consequent suppression of normal nuclear division this tendency, although normally expressed at the termi- nation of cell division, becomes dominant. At any rate, the appearance of the inhibited cells in no way suggests an iiuincdiitlc inhibition of all unclear activity.
NARCOSIS AND CELL DIVISION
19
It is more likely that certain changes continue and give rise to the modified division figures observed.
Indicative of the fact that any phase of cell division is susceptible to the effects of narcosis is the diversity of aberrant division figures produced as a result of treatment. A continuous series of aberrant figures representing all stages of division may be constructed.
300
2 50
200
X
o
o
o
I 50
I 00
O 50
00
0
O MASS ACTION
3 ABERRANT DIVISION FIGURES
URETHANE
ZONE OF COMPLETE GROWTH INHIBITION
06 10
LOG (10 X CONCENTRATION)
I 4
FIGURE 2. Summary of complete urethane data. Log (10 X U/I), aberrant division fig- vires (per cent) and zone of complete growth inhibition plotted against Log (10 X concentra- tion of urethane).
Other carbamates: The affects of n-propyl, phenyl, n-amyl, iso-amyl, and ethyl-n-methyl carbamate on cell division were also studied. None of these sub- stances were tested through as wide a range of concentration in small graded in- crements as was ethyl carbamate. However, it was found that these closely related compounds produced divisional changes qualitatively identical with those of ethyl urethane. The ability to inhibit micronuclear migration is common to all members of this group of substances. Similarly, the associated changes in arrangement and aggregation of macronuclear chromatin and the bizarre division figures described for ethyl urethane were also induced.
20 RICHARD L. BURT
Chloral hydrate: The activitv of this compound in disturbing the normal progress of division was assayed in graded concentrations of 0.1 per cent incre- ments between 0.05 and 0.5 per cent. The morphological changes induced were similar to those described for the nrethanes, hut the percentage of chloralized cells in which the polar defect appeared was never high. Moreover, the concentration range for effective induction of the polar defect is quite narrow (0.1-0.3 per cent). The maximum number of deranged mitoses was only 3.9 per cent, a small value compared with the maximum ethvl urethane value of more than 20 per cent.
In excess of 0.3 per cent chloral hydrate the delay in poleward movement and the changes in the nuclei characteristic of cells narcotized in the lower concentra- tions rather abruptly disappear. In these higher concentrations retrogressive changes are induced which are expressed as condensation, diffuse staining and chromatolysis of the macronucleus and pyknosis of the micronucleus. The in- cidence of these changes, although slight to moderate at 0.4 per cent, increases rapidly with still higher concentrations until in 0.5-0.6 per cent chloral hydrate all of the organisms, both trophic and those in division, are frankly moribund. These degenerative changes are physiologically and morphologically reversible in some cells at least after one hour exposure at 0.6 per cent. If the cells are washed free of narcotic, packed by centrifugation and resuspended in freshly bacterized dis- tilled water, growth will ensue. It is impossible to determine, however, whether or not every cell so treated is viable.
In summary, chloral hydrate affects cell division in a manner quite similar to ethyl carbamate and related compounds. However, the effective range of concen- tration for the production of these changes is relatively narrow, and at no concen- tration are the numbers of deranged divisions high. A possible explanation for this difference in activity will he outlined.
Effect of other inhibitors: The following compounds in the concentrations indicated were tested for their effect on the divisional mechanism :
Inhibitor Concentrations
Potassium cyanide 0.005M to 0.000,05 M
Potassium ferricyanide 0.050M to 0.005M
Sodium arsenite 0.001M to 0.000.25M
Monoiodoacetic acid 0.005 M to 0.000.5M
These metabolic poisons throughout the concentrations designated were uniformly ineffective in the production of aberrant divisions of the type described for the carbamates and chloral hydrate. When cells were exposed to the lower concen- trations of these inhibitors no change could be detected cytologically in the progress of division. 1 lowever, at higher concentrations retrogressive changes characterized by extreme condensation and pyknosis of both macronuclei and micronuclei oc- curred. These changes were comparable to those induced at high concentrations of chloral hydrate and are considered as definite signs of cell damage as opposed to tin- milder type of change occurring with moderate narcotic concentrations.
(inrn'/li inhibition
As a correlative procedure in connection with the cytological and metabolic studies, the effect of ethyl urethane and chloral hydrate on suppression of growth in cultures of ('. slcinii was tested.
NARCOSIS AND CELL DIVISION
21
In Figure 3 are tabulated the results from one typical determination with ethyl urethane. Fairly close correspondence in the degree of inhibition among the four experimental series was obtained. It is apparent from these results that growth is only slightly affected by 0.5 per cent urethane (3 plus to 4 plus growth). How- ever, with increasing concentrations of narcotic, suppression of growth becomes more pronounced until at 1.0 per cent urethane no cell division at all can be detected. Similarly, Ormsbee (1941) found that 1.0 per cent urethane would completely inhibit the growth of populations of the ciliate Tetniliyinciia </clcii.
|
CHLORAL HYD. |
EXP. 1 |
EXP.2 |
EX P. 3 |
EXP. 4 |
|||||||||
|
SER- IES |
PERCENT CONC. |
HOURS |
HOURS |
HOURS |
HOURS |
||||||||
|
12 |
24 |
36 |
12 |
24 |
36 |
12 |
24 |
36 |
12 |
24 |
36 |
||
|
A |
.025 |
4 + + 4 |
4-4-4+ |
4444 |
4444 |
+ + +4 |
+ + + 4 |
+ 4+4 |
+ + + + |
44 4-4 |
4444 |
4+44 |
+ 4 4 -t- |
|
B |
.05 |
4-4 + 4- |
4 + 4 |
44 4 |
-t- 4 + |
4+4 |
+ H- ++ |
+ 4 4 |
+• +4 |
444 |
444 |
4 4 |
444 |
|
C |
.1 |
4 |
4- |
•f |
4 |
+ |
+ |
•f |
+ |
+• |
4 |
+ |
4 |
|
D |
.2 |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
— |
- |
|
E |
.3 |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
|
F |
A |
- |
- |
- |
- |
- |
- |
— |
- |
- |
- |
— |
- |
|
G |
CONTROL |
f 4 + + |
+ +44 |
444 + |
+ + + + |
+ 4 + + |
•4-4- + + |
++ + + |
-1- + 4 + |
H-4 44 |
+ 4- 4 4 |
4444 |
44- f 4 |
FIGURE 3. Inhibition of growth in cultures of Colpoda stciuii with various concentrations of ethyl carbamate. Four plus growth represents growth in untreated controls.
|
URETHANE |
EXP 1 |
EXP.2 |
EXP. 3 |
EXR4 |
|||||||||
|
SER- IES |
PERCENT CONC. |
HOURS |
HOURS |
HOURS |
HOURS |
||||||||
|
12 |
24 |
36 |
12 |
24 |
36 |
1 2 |
24 |
36 |
1 2 |
24 |
36 |
||
|
A |
0.5 |
4 4 4 |
4444 |
444 + |
4-4- 4 4 |
4444 |
444 + |
44 + |
44+4 |
444 + |
+ + +4 |
4 + 44 |
4444 |
|
B |
0.8 |
+ |
4 |
4 |
4 |
4 4 |
4 4 |
+ |
44 |
4 4 |
4- |
4 + 4 |
444 |
|
C |
0.9 |
4 |
4- |
4 |
4 |
4 |
4 |
+ |
+ |
4 |
4 |
4 |
4 |
|
D |
1.0 |
4 |
•+• |
+ |
4 |
4 |
+ |
+_ |
4 |
4 |
+ |
+ |
+ |
|
E |
I.I |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
— |
|
F |
1.2 |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
- |
|
G |
CONTROL |
4444 |
4444 |
4444 |
444 4 |
4444 |
4444 |
4444 |
4444 |
+ 44 + |
+ 4 + + |
4444 |
4444 |
FIGURE 4. Inhibition of growth in cultures of Colpoda stciuii with various concentrations of chloral hydrate. Four plus represents growth in untreated control-.
In Figure 4 are represented the results of an analogous determination made with chloral hydrate. Here complete inhibition of growth takes place in 0.2 per cent of the narcotic.
RICHARD L. HURT
Comparison of the two sets of experimental results is instructive in indicating the differences in effective concentration ranges of the two inhibitors. In the case of ethyl carbamate an increment of 0.6 per cent in concentration must be made from the point where its effect is first noted until complete inhibition occurs (from 0.5 to
1.1 per cent). With chloral hydrate this increment is only 0.15 per cent or one fourth that of urethane. This is correlated with the brief range of effectiveness of chloral hydrate in the induction of aberrant division figures.
Respiratory metabolism
As pointed out in a preceding section, overall oxygen consumption in a cell or tissue does not necessarily have uniform significance as regards a specific function. Perhaps the most convincing evidence for this is the interpretation of nacrotic in- hibition data made by Fisher and his colleagues (for review see Fisher, 1942).
Equations expressing the relationship between enzyme and inhibitor may be derived from the Law of Mass Action (Warburg, 1927; Warburg and Negelein, 1928; Fisher and Ohnell, 1940). As required by the Mass Action formulation, when the logarithm of the concentration of inhibitor (narcotic) is plotted against
Uninhibited respiration
the logarithm of . ., . — £— a straight line should result. Non-
Inhibited respiration
linearity of mass action curves so constructed is believed to indicate the presence of at least two parallel respiratory systems in Arbacia eggs and in yeast (Fisher. 1941a, b; Fisher and Henry, 1940; Fisher and Stern. 1942).
The respiration of Colpoda stcinii was tested in graded concentrations of urethane and chloral hydrate with the object of obtaining evidence for such respira- tory discontinuity in this organism. The results of respiratory inhibition as meas- ured by the Warburg technique between concentrations of 0.4 and 2.0 per cent urethane are plotted in Figure 2. At a concentration of urethane between 1.0 and
1.2 per cent a break in the mass action curve occurs at which point respiration is inhibited by 35 per cent. This is taken as evidence for the existence of two separate respiratory mechanisms that differ in sensitivity to narcotic inhibiton.
In Figure 2 is also shown the relationship between this respiratory data, growth inhibition, and aberrant nuclear behavior at various levels of narcotic inhibition. The increase in pathological cell divisions is related to the degree of suppression of the upper limb of the mass action plot representing the inhibition of the more highly narcotic-sensitive fraction of metabolism. It is also interesting to note the correspondence between the concentration of urethane giving complete growth inhibition (1.1 per cent) and the concentration at which the break in the mass action curve occurs (approximately 1.1 per cent).
A discrepancy exists in these relations, however, for the maximum number of aberrant nuclear divisions is not obtained until a concentration of 1.4 to 1.5 per cent urethane is reached. This may be due to experimental error in part, but the theoretically more attractive possibility is that an overlapping in the inhibition of the two respiratory systems occurs. In accordance with this view the resting system would begin to be inhibited before the activity system was completely sup- pressed. This point will be dealt with more fully in a following section. Ap- parently, however, following complete inhibition of the more narcotic sensitive fraction (activity system of Fisher) not only is growth of the culture suppressed, but the incidence of abnormal divisional figures attains a maximum.
NARCOSIS AND CELL DIVISION
23
In Figure 5 is summarized the effect on respiration, growth and the derange- ment of cell division brought about by graded concentrations of chloral hydrate between 0.0125 and 0.6 per cent. These results resemble those obtained with ethyl urethane. Non-linearity of the mass action curve is characteristic. Also a fair correspondence exists between the concentration giving complete growth inhibi- tion (0.2 per cent) and the concentration at which discontinuity of the respiratory inhibition becomes apparent (0.1 per cent). Unlike urethane, however, the per-
2.50
2.00
1.90
x o
(9 O
1.00
0.30
ABERRANT DIVISION FIGURES
ZONE of COMPLETE GROWTH INHIBITION
1.4 1.8 0.2
LOG (10 x CONCENTRATION)
0.6
FIGURE 5. Summary of complete chloral hydrate data. Log (10X £7/7), aberrant divi- sion figures (per cent) and zone of complete growth inhibition plotted against Log (10 X con- centration of chloral hydrate).
centage of aberrant nuclear divisions induced by chloral hydrate is never high, the maximum being in the vicinity of 4 per cent. This maximum is not obtained until the cells are exposed to 0.3 per cent chloral hydrate, or 0.2 per cent in excess of the concentration at which discontinuity in the respiratory data is expressed. Again, a possible explanation for this may be an overlapping in the inhibition of the activity and resting systems whereupon after a certain concentration is exceeded, both are inhibited simultaneously. That the more sensitive fraction indicated by chloral hydrate represents about 45 per cent of the total respiration (taking the break in the curve as the end point) in contrast to 35 per cent with urethane makes this possibility plausible. The higher value of respiratory inhibition down to the
24 RICHARD L. BURT
])dint where discontinuity is displayed in the mass action curve therefore represents the activity system pins the added respiratory decrease caused by the suppression of part of the less sensitive fraction.
DISCUSSION
It is believed that under the experimental conditions adhered to in this study the data indicate the presence of at least two fractions of metabolism in Colpoda stciniL These differ in their sensitivity to inhibition by either chloral hydrate or urethane. One portion of this parallel respiratory mechanism, the activity frac- tion, appears to be associated with the maintenance of normal cell division. In general, this is in agreement with the observations of Fisher and his colleagues (1940-1942). This conception is based on the fact that discontinuity in the relation between concentration of narcotic and its effect in depressing respiration of these ciliates is expressed by mass action treatment of the data. The factors that might invalidate such an interpretation have been pointed out by Fisher and Stern (1942), but it is believed that these factors were not operative in this investigation.
Complete and quantitative separation of activity and resting fractions of the total metabolism by differential narcotic inhibition is considered here as extremely unlikely. Four findings in this report support the view that parallel inhibition of the two systems occurs in an intermediate range of narcotic concentrations :
1. The increase to a maximum in the incidence of nuclear aberrations after mass action discontinuity has been expressed. This is considered most significant and was found with both urethane and chloral hydrate (See Figs. 2 and 5).
2. The difference of 10 per cent between the activity respiration indicated by urethane (35 per cent) and that of chloral hydrate (45 per cent).
3. The delay in obtaining complete growth suppression with chloral hydrate after discontinuity in the mass action curve has been expressed.
4. The brief effective concentration range of chloral hydrate in inducing ab- normal division figures and the low maximum for the polar defect obtained with ibis reagent.
Xuclear divisional activity persists at narcotic concentrations in excess of those at the break in the mass action curves. The maximum number of nuclear ab- normalities with urethane does not occur until a concentration of 1.5 per cent of the narcotic is reached, and 0.3 per cent chloral hydrate is necessary to attain the corresponding maximum. Hence, the relation between the activity fraction, as determined by the respiratory inhibition curves, and divisional activity is not precise. It is believed that this failure of complete correspondence between the production of aberrant mitoses and discontinuity of the respiratory data is due to parallel or simultaneous inhibition of both activity and resting systems in an intermediate range of narcotic concentrations. In effect, this means that con- tinued nuclear activity is possible even after the break in the mass action curves occurs because the activity system is not completely suppressed at this point. Mitosis therefore continues in a'n increasingly abnormal fashion until the activity metabolism is finally completely eliminated.
The conception of overlapping or parallel inhibition of both respiratory mecha- ni>nis also implies that the slopes of the two lines describing the reactions cannot
NARCOSIS AND CELL DIVISION 25
be accurately descriptive of the relation between the narcotic.- on the separate systems even though the existence of at least two respiratory mechanisms is ap- parent. Furthermore, the value of 35 per cent representing the activity respira- tion, as determined by nrethane inhibition, is not an absolute value for the per- centage of the total energy involved in cell multiplication. However, this figure corresponds fairly well with the 28 per cent found for Tctraliyincna (jclcii by Ormsbee (1941) and the 30-40 per cent of the overall respiration associated with the maintenance of cell division in sea urchin eggs reported by Krahl and Clowes (1939).
The value of 45 per cent for activity respiration,- determined by chloral hydrate. can be accounted for by overlapping in the inhibition of the two respiratory mecha- nisms. The additional 10 per cent activity respiration indicated by this drug would thus represent part of the resting respiratory mechanism that was involved at the narcotic concentration where the break in the mass action curve occurred. Failure to completely suppress growth at a concentration of 0.1 per cent chloral hydrate (at break in logarithmic plot) likewise is indicative of this inhibition overlap. With the increase in narcotic to 0.2 per cent cell multiplication ceases because of the more complete elimination of the activity system.
When the progress of mitosis is suppressed by appropriate narcotic concen- trations nuclear changes described earlier in this report as reorganizational are in- duced. Higher concentrations of narcotic inhibit these trophic changes (as 3 per cent urethane or 0.5 per cent chloral hydrate). At these concentrations the nuclei become condensed and pyknotic. The morphology described as resulting from varying degrees of suppression of the activity system obtained at lower narcotic concentrations are not seen at these higher drug levels. Thus in the production of nuclear abnormalities of the type described, two mechanisms appear to operate :
1. The inhibition of the activity system with associated suppression of the progress of mitosis.
2. Reorganization of the nuclei toward the trophic condition at that stage in division where inhibition has occurred.
The trophic changes are considered here to be dependent upon the maintenance of the resting metabolism of the cell (narcotic resistant system). As more of this less sensitive respiratory system is inhibited with increasing narcotic dosage, a point is finally reached when reorganization toward the trophic morphology cannot occur. Concomitantly pyknosis sets in and the nuclear picture is completely altered. Where the resting fraction is inhibited simultaneously with the activity fraction a restriction on the effective production of aberrant divisional figures results. The brief range of effectiveness and low maximum for nuclear abnormali- ties obtained with chloral hydrate can be accounted for on this basis.
The conception of overlapping with respect to narcotic inhibition postulated in the foregoing account is supported by the observation of Fisher (1941 a) that benzoate inhibits the resting system in sea urchin eggs before inhibiting the ac- tivity system. Here it is necessary to inhibit 50 per cent of the respiration before inhibition of cleavage occurs. The condition described for C. stcinii in its be- havior with narcotics appears to represent an intermediate situation between the theoretical, where precise quantitative separation of the two systems would occur, and the juxtaposition of the two systems with respect to sensitivity to benzoate described for sea urchin eggs.
26 RICHARD L. BURT
It is to be emphasized that the narcotic concentration required to inhibit cell multiplication in cultures corresponds closely with that at the point in the mass action curves where discontinuity is expressed, i.e., when the activity system, as determined by this method, is inhibited growth of the culture ceases. To suppress nuclear activity, however, more complete elimination of the activity system appears to be necessary, hence the delay in attaining the maximum number of abnormal division figures.
The cytological effects of narcotics on dividing Colpoda stcinii are considered here as unique, but are comparable to the results obtained with these agents on many other cell types. Furthermore, it is apparent that in addition to narcotics many chemical and physical agents induce quite similar changes in the course of cell division. In general, the interpretation placed on this non-specificity of effect is that regardless of the agent employed there is somehow- brought about a modifica- tion of the intracellular colloid state. This common result of experimental treat- ment would account for the similarities of morphological variation (see Ludford, 1936; von Lehotzky. 1938; Kemp and Juul, 1931 ; M. R. Lewis, 1933, 1934; Rosen- feld, 1932). This point of view, in light of the paucity of information available that is pertinent to the actual mechanics of these defects, appears to be well taken. However, the complexity of the protoplasmic organization in its reactions to vari- ous treatments must not be overlooked.
In the present study it has been shown that with gradual suppression of that portion of the cells' energy yielding reactions designated as activity metabolism, cell multiplication is suppressed and abnormal division figures are induced. It is not surprising that the divisional mechanisms drawing on these energy sources should reflect such reductions in energy production by abnormal behavior. However, it is difficult to visualize the manner in which this metabolic inhibition becomes morphologically expressed.
Possibly by interfering with cellular oxidation (narcosis) physical changes in the protoplasm result (von Lehotzky, 1938) ; this would relate the abnormal di- vision figures obtained by narcotics to those induced by temperature changes, ether, carbon dioxide, hypo- and hypertonic solutions. All of these agencies have been demonstrated to affect the viscosity of protoplasm (for review, see Chamber's discussion in Cowdry's General Cytology). Conceivably, such experimental in- terference might exert its effect by modifying normal cyclical viscosity changes occurring during cell division (Heilbrunn, 1917; 1920; 1921).
No doubt such physical changes are significant for the normal progress of cell division and bear consideration in dealing with experimentally induced changes in mitosis. That such purely physical aspects fail to describe the entire sequence of events encountered in the complex dynamics of cell division is obvious. It may be, however, that they are the links between the biochemical phases involving cellular oxidations and the well known cytological manifestations of activity. Integrating factors for such an association are at present lacking. With regard to the present study, it can only be said that a certain portion of the overall metabolism is prob- ably associated with these mechanisms. To postulate more than this would be the sheerest speculation.
Briefly, the negative results obtained in the- preliminary work with cyanide, ferricyanide, arsenite and iodoacetate may be considered. The reactions of any of these inhibitors in the intact cell, with the possible exception of cyanide, are
NARCOSIS AND CELL DIVISION
complex. At the present time data concerning their effect on the metabolism of Colpoda stehiii are lacking. Apparently, however, with the limited information concerning the morphological expression of treatment by these agents, their effects differ considerably from those of the narcotics tested. To take one example, cyanide does not seem to visibly alter the division mechanism until the occurrence of retrogressive changes resembling those induced by the higher doses of narcotics. Tentatively it may be suggested that cyanide, by virtue of its reaction at the oxygen end of the cellular oxidation chain where a large percentage of respiration is mediated, simultaneously inhibits both activity and resting systems. When a certain percentage of the resting metabolism has been suppressed, the retrogressive alterations are induced.
SUMMARY
1. Ethyl carbamate has been demonstrated to be effective in inducing aberrant nuclear behavior during division of the ciliate Colpoda steiuii. This activity is shared by other carbamates and chloral hydrate. These effects are described in detail.
2. A quantitative means of assaying cytologically the activity of these substances is described. On the basis of defective polar movement of daughter micronuclei ethyl carbamate was found to be more active than chloral hydrate. A possible ex- planation for this observation is postulated.
3. The results of preliminary work involving treatment of dividing cells with potassium cyanide, potassium ferricyanide, arsenite, and iodoacetate are reported. These agents apparently do not share the activity evidenced by the carbamates or chloral hydrate.
4. By means of the Warburg technique the effect of urethane and chloral hy- drate on respiration of C '. stenii was studied.
5. Discontinuity in the quantitative action of these agents on respiration is in- terpreted to indicate the existence of at least two fractions of metabolism. One of these, the activity system, is believed to be associated with the process of cell division.
6. Complete separation of these two systems is not considered probable, how- ever, and caution is urged in the interpretation of discontinuity in Mass Action treatment of this and similar data.
7. The significance of the association of the appearance of aberrant division figures, growth inhibition and discontinuity in the relationship of narcotic concen- tration to respiratory inhibition is pointed out. Discrepancies in these relation- ships are discussed.
8. The bearing of metabolic inhibition on the general problem of the pathology of mitosis is discussed.
LITERATURE CITED
ALBERTI, W., AND G. POLITZER, 1923. Uber den Einfluss der Rontgenstrahlen auf die Zellteil-
ung. Arch. f. mlkrosk. Anat. u. Entwickmech., 100: 83. ALLEN, P. J., AND D. R. GODDARD, 1938. A respiratory study of powdery mildew of wheat.
Amcr. Jour. Bot.. 25 : 613. BLEIER, H., 1930. Experimentell-Cytologische Untersuchungen. I. Einfluss abnormaler Tem-
peratur auf die Reduktionteilung. Zcit. Zellforsch. u. mikr. Anat., 11 : 218.
RICHARD L. BURT
BOOINE, J. II., 1934. The effect of cyanide on the oxygen consumption of normal and blocked
embryonic cells (Orthoptera) . Jour. Cell. Coinp. Pliysioi, 4: 397.
BURT, R. L., 1940. Specific diagnosis of the genus Colpoda with special reference to the stand- ardization of experimental material. Trans. Aincr. Micro. Soc., 59: 414. BURT, R. L., 1942. Narcosis and cell division in Colpoda steinii. Doctorate Thesis. Brown
University. BURT, R. L., G. W. KIDDER, AND C. L. CLAFF, 1941. Nuclear reorganization in the family
Colpodidae. Jour. Morph., 69: 537. BURY, J., 1913. Experimentelle Untersuchungen uber die Einwirkung der Temperatur O Grad
auf die Entvvicklung der Echinidoneier. Arch. f. Entwicknicch., 36: 537. CLOWES, G. H. A., AND M. E. KRAHL, 1939. Studies on cell metabolism and cell division. III.
Oxygen consumption and cell division of fertilized sea urchin eggs in the presence of
respiratory inhibitors. Jour. Gen. Physiol., 23 : 400. COWDRY, E. V., 1924. General Cytology. Chicago. DEUTSCH, \Y., AND H. S. RAPER, 1938. The respiration and metabolism of submaxillary gland
tissue of the cat. Jour. Physiol., 92 : 439. FISHER, K. C., 1941a. The fractionation of cellular respiration by the use of narcotics. Biol.
Bull., 81 : 282. FISHER, K. C., 1941b. A mechanism of narcosis suggested by the effects of narcotics on several
types of cell. Proc. Amcr. Physiol. Soc., 133: 278. FISHER, K. C., 1942. Narcosis. Canad. Mcd. Assoc. Jour., 47 : 414. FISHER, K. C, AND R. J. HENRY, 1940. The use of urethane as an indicator of "activity"
metabolism in the sea urchin egg. Biol. Bull., 79: 371. FISHER, K. C., AND R. OHNELL, 1940. The steady state frequency of the embryonic fish heart
at different concentrations of cyanide. Jour. Cell. Coinp. Pliysioi, 16: 1. FISHER, K. C., AND J. R. STERN, 1942. The separation of an "activity" metabolism from the
total respiration of yeast by the effects of ethyl carbamate. Jour. Cell. Coinp. Phvsiol.,
19: 109. GEIERSBACH, U., 1939. Uber den Einfluss der Narkose (Urethan) auf Gewebekulturen. Arch.
f. expcr. Zcllforsch., 23 : 210. GODDARD, D. R., AND P. E. SMITH, 1938. Respiratory block in the dormant spores of Neuro-
spora tetrasperma. Plant Physiol., 13: 241.
HACKER, V., 1900. Mitose im Gefolge amitosenahnlichen Vorgange. Anat. Ans., 17: 9. HEILBRUNN, L. V., 1917. An experimental study of cell division. Anat. Rcc., 11 : 487. HEILBRUNN, L. V., 1920. An experimental study of cell division. I. The physical conditions
which determine the appearance of the spindle in sea urchin eggs. Jour. E.rper. Zool.,
30: 211. HEILBRUNN, L. V., 1921. Protoplasmic viscosity changes during mitosis. Jour. E.vpcr. Zool.,
34: 417. HOROWITZ, N. H., 1940. Comparison of the oxygen consumption of normal embryos and
dauerblastulae of the sea urchin. Jour. Cell. Comp. Physiol., 15 : 309. KEMP, T., AND J. JUUL, 1931. Der Einfluss der Warme auf Zellteilung untersucht in
Gewebekultur. Arch. f. c.vper. Zcllforsch., 11 : 602. KIDDER, G. W., AND C. L. CLAFF, 1938. Cytological investigations of Colpoda cucullus. Biol.
Bull.. 74: 178. KIDDER, G. W., AND C. A. STUART, 1939. Growth studies on ciliates. II. The food factor in
growth, reproduction, and encystment of Colpoda. Physiol. Zool. 12: 341. KRAHL, M. E., AND G. H. A. CLOWES, 1939. Studies on cell metabolism and cell division. IV.
Combined action of substituted phenols, cyanide, carbon monoxide and other respiratory
inhibitors on respiration and cell division. Jour. Gen. Physio!., 23: 413. KRANTZ, H., 1938. Y.ur Kenntnis der Mitose. V. Einfluss des Athylalkohols auf Beginn und
Alblauf Mitosen in Fibrocytenkulturen. Zeit. f. Zcllforsch. it mikrosk. Anat., 28: 709. VON LEHOTZKY, P., 1938. Die zytologischen Grundlagen der Narkose. Arch, c.rper. Zcllforsch.,
21 : 250. LEWIS, M. R., 1933. Reversible changes in the nature of the mitotic spindle brought about in
living cells by means of heat. Arch. f. c.rpcr. Zcllforsch., 14: 464. LEWIS, M. R., 1934. Reversible solation of the mitotic spindle of living chick embryo cells
studied in vitro. Arch. f. c.rper. Zcllforsch., 16: 159.
NARCOSIS AND CELL DIVISION 29
LUDFORD, R. J., 1936. The action of toxic substances upon the division of normal and malignant
cells in vitro and in vivo. Arch. f. cxpcr. Zcllforsch., 18: 411.
MACLEOD, J., 1941. The effect of glycolysis inhibitors and of certain substrates on the metab- olism and motility of human spermatozoa. Endocrinology, 29: 583. NEMEC, B., 1904. Uber die Einwirkung des Chloralhydrate auf die Kern und Zellteilung.
Jarb. f. zviss. Bot., 39 : 645.
ORMSBEE, R. A., 1941. Normal growth, respiration and the effect of various respiratory in- hibitors on Tetrahymena geleii. Doctorate Thesis. Brown University. PEASE, D. C, 1941. Echinoderm bilateral determination in chemical concentration gradients.
I. The effects of cyanide, ferricyam'de, iodoacetate, picrate, etc. Jonr. E.\-f>cr. /no!., 86:
381. PFUHL, W., AND H. KUNTZ, 1939. Die pathologischen Mitosen in Bindegewebezellen nach
einmaliger Rontgenstrahlen. Zcit. Anat. u. Entwickgesch., 110: 98. POLITZER, G., 1924. Versuch iiber den Einfluss des Neutralrots auf die Zellteilung. Zcit.
Zcllforsch. u. mikrosk. Anat., 1 : 644. POLITZER, G., 1931. Die Zellteilung wahrend und nach Narkose. Ein Beitrag zur Kenntnis
der Storungen des Kernteilungsrhythmus. Zcit. f. Zcllforsch. it. mikrosk. Anat.. 13:
334. VAN REGEMORTER, D., 1926. Les troubles cinetiques dans les racines chloralisees et leur portee
pour 1'interpretation des phenomenes normaux. La Cellule, 37 : 43. ROBBIE, W. A., E. J. BOELL, AND J. H. BODINE, 1938. A study of the mechanism of cyanide
inhibition. I. Effect of concentration on the egg of Melanoplus differentialis. Physiol.
Zool, 11: 54. ROSENFEL'D, M., 1932. The action of ether on cells in mitosis. Arch. f. c.rpcr. Zcllforsch. , 12:
570. SCHAKHLEVICH, M. V., 1938. Izmcuoua adhromatinovoi chast veratena pod vliianien razlich-
nykh agentov. I. Vliianie khloralhidrata na deliashchiesia kletki v koreshkakh pro-
rostkov boba (Vicia faba). Kusskii Arkliir Anutoinii, Gistolotjii i Embriologii, 19: 218.
Arch. f. Entwickincch., 27: 560.
SCHILLER, I., 1909. Uber kunstliche Erzeugung "primitiver" Kernteilungsformen bei Cyclops. VAN SCHOUWENBURG, K. L., 1938. On respiration and light emission in luminous bacteria.
Naainloozc V ermootschap W. D. Mcincina: Delft. STEVENS, N. M., 1909. The effect of ultraviolet light upon the developing eggs of Ascaris
megalocephala. Arch. f. Etitivickiucch.. 27: 171.
TAYLOR, C. V., AND A. G. R. STRICKLAND, 1938. Reaction of Colpoda duodenaria to environ- mental factors. I. Some factors influencing growth and encystment. Arch. Prof., 90:
396. VINTEMBERGER, P., 1930. De 1'action du froid sur les cellules en mitose. C. R. Soc. Bio!., 103:
705. WARBURG, O., 1927. Uber die Wirkung von Kohlenoxyd und Stickoxyd auf Atmung und
Garung. Biochcm. Zcit., 189: 354. WARBURG, O., AND E. NEGELEIN, 1928. Uber die Verteilung des Atmungsferments zwisclien
Kohlenoxyd und Sauerstoff. Biochcm. Zcit., 193: 334. WHITMAN, W. G., 1933. Some observations . on the effects of radium irradiation on t
cultures. Amcr. Jour. Cancer, 17 : 932.
A NOTE ON THE ABSORPTION SPECTRA OF THE BLOOD OF
EUDISTYLIA GIGANTEA AND OF THE PIGMENT IN THE
RED CORPUSCLES OF CUCUMARIA MINI ATA
AND MOLPADIA INTERMEDIA
FREDERICK CRESCITELLI 1
The Oceanographic Laboratories and the Department <>t Znoluvy and Pliysiolouy.
[rnh'crsity of Washington, Seattle
In 1868 Lankester noted the presence of a red pigment (erythrocruorin) in the plasma of certain annelids, in Chironomus larvae and in Planorbis. He also re- ported a green pigment (chlorocruorin) in the plasma of Siphonostoma and Sahella. Subsequent studies have revealed the widespread occurrence of erythrocruorin - (invertebrate hemoglobin) in many worms, in echinoderms (holothuroids), in some molluscs and in a few arthropods (Redfield, 1933; Kobayashi, 1936). :; Chlorocruorin appears to be more limited in its distribution having been found thus far only in certain polychaete worms (Fox, 1925).
The absorption curves of both these heme pigments have been studied in some detail by Fox (1925), Redfield and Florkin (1931) and Kobayashi (1932, 1935, 1936). From a review of this literature it appears that the a-band of oxyerythro- cruorin falls, in most cases, well within the 577-579 m/x region, although figures as low as 574.5 m/x have been reported for some of the worms (Barcroft and Barcroft, 1924; Kobayashi, 1936). The usual position of the a-band agrees well with the position of the comparable band in vertebrate blood. The /3-band of oxyerythro- cruorin occurs, in most cases, between 540-542 m/x although, here again, there are exceptions as in the case of the holothuroid, Caiidina cJiilcnsis, where this band is found at 544.2 niit (Kobayashi, 1932) and in the case of the earthworms, Pherctima coinniitiiisshna and Pherctima hilgendorfi, where the band occurs at 538 m//, and 539 m/x respectively (Kobayashi, 1936). The usual position of the /?-band of oxyerythrocruorin agrees generally with the position of the comparable band in vertebrate blood. Reduced erythrocruorin, in many cases, possesses a single band with maximum at 556 m/x, which is also the case for vertebrate hemoglobin, but in Candina chilcnsis this maximum has been reported to occur at 560 m/x (Kobayashi, 1932) and in Cncmnaria jrancnfcldi at 558 m/x (Hogben and Van Der Lingen, 1938). The situation appears to be distinctly different for some of the worms where the reduced pigment possesses a double peak ; one band being at 566-571 m/x, the other at 549-551 m/x (Kobayashi, 1936; Vies, cited by Kobayashi, N36).
1 On war leave in the Department nf Aviation Medicine, School of Medicine, University of Southern California.
- Many authors, since the original work of Lankester, have used the term hemoglobin for the pigment in invertehrates, but the present writer is following the suggestion of Svedherg and Eriksson (1933) that, since the protein portion of invertebrate hemoglobin is characteristically different from that of vertebrate hemoglobin, the separate name, erythrocruorin, is justified.
3 A report has been published by Sato and Tamiya (1937) indicating bands of hemoglobin in I'in-iiiiieciiiiii caitdtiluin, but the writer has seen only an abstract of this report.
30
ABSORPTION SPECTRA 31
Compared to oxyerythrocruorin, the bands of oxychlorocruorin are shifted toward the red end of the spectrum. In five species of worms, Fox (1925) ob- tained a spectrum with the a-band in the 602.5-605.9 iriju. region, while the /3-hand (in Spirographis) is at 561 m/j. and a third faint band (also in Spirographis) is located at about 517 m/x. Using crystalline chlorocruorin, Roche and Fox (1933) have been able to confirm these results.
Recently the author had occasion to examine spectroscopically the green pig- ment dissolved in the plasma of the tube worm, Eudistylia gigantca, and the red pigment of the corpuscles found in the perivisceral fluid of the holothuroids, Cucn- uiaria miniata and Molpadia intermedia. The results which were obtained agree in many details with previous data, but in some respects the results are so strikingly different that a record of them would be of value, even though the war has pre- vented completion of the investigation.
The author wishes to express his appreciation to Professor Thomas G. Thomp- son for making available the facilities at the laboratories in Friday Harbor and the Oceanographic Laboratory in Seattle. To Professor Trevor Kincaid appreciation is expressed for his aid in identifying the animals used in this work.
MATERIALS AND METHODS
The large tube worm Eudistylia gigantca (family Sabellidae) is found at rela- tively low tidal levels in large colonies attached to rocks or pilings in certain areas of Puget Sound. The pigment, which is dissolved in the plasma, appears red in the concentration occurring in the dorsal blood vessel, but upon removal and dilution it is seen to be green. About 0.1-0.2 ml. of blood was obtained by means of a glass capillary tube inserted into the dorsal blood vessel. After dilution to about 50-100 ml. with either cold distilled water or with cold sea water, the solution was filtered and the filtrate, containing the pigment, was placed in a 10 cm. long specimen tube. The spectrum was then obtained by visual matching, using a Bausch and Lomb spectrophotometer. The holothuroids which were used include the sea cucumber, Cncmnaria miniata (order Dendrochirota) , which is found abundantly at Friday Harbor in between rocks at low tide level, and the apodous sea cucumber, Molpadia intermedia (order Molpadiida), which was ob- tained from the muddy bottom of East Sound at a depth of 12-15 fathoms. The red pigment is located in numerous corpuscles suspended in the perivisceral fluid of both these sea cucumbers. These corpuscles in Molpadia, when examined micro- scopically, are seen to be pale-yellow cells which possess a variety of elongated. multi-lobed shapes. A few are oval or spherical in form and all of them have a single, small, dark and spherical nucleus located at a variety of points in the cell but rarely at the geometric center. When a drop of distilled water is mixed with a drop of the perivisceral fluid the cells assume a perfectly spherical form and the nucleus is seen to occupy an eccentric position. The corpuscles of Cucumaria are also pale-yellow, nucleated cells with a variety of shapes ; some are ovoid, some spherical and others are quite irregular with one or several processes. The elon- gated lobed forms of Molpadia are seldom seen in Cncmnaria miniata. Previous accounts of the holothurian red cell by Dawson (1933) and Ohuye (1936) have already mentioned some of these structural features. By means of a small punc- ture in the body wall, the red perivisceral fluid was collected and the cells cen-
32
FREDERICK CRESCITELLI
trifuged out. After decanting the supernatant fluid, the cells were washed in cold sea water and again centrifuged out. This process was repeated three or four times after which the cells were hemolvzed in cold distilled water. The solution of
j
pigment after filtration and dilution to ahout one hundred ml. with cold distilled water was examined spectroscopically.
Results and discussion
The five curves (Fig. 1) obtained from five different sea cucumbers (Ciicn- iiuirio ininiata) give an indication of the reproducibility of the spectral curve and show that the location of a point of maximum or minimum can be checked within 2 m/t. These curves, which were obtained from well-aerated solutions of the pig- ment, show two maxima ; one 580-581 niju., the other at 544-545 in/x. The point
a.t.
z.o_
O.O D
I
I
51O 5EO SJO 540 SSO S6O 57O 58O S9O 6OO 61O 62O 63O
FIGURE 1. The absorption spectral curves of the pigment from five different sea cucumbers (Cttcitnwria iniiiiata). The density (log In/I) is plotted as ordinates against the wave length, in mM. The individual graphs are plotted on a multiple ordinate scale to avoid crowding. The zero ordinate for each curve is indicated by the corresponding letter. Each plotted point repre- sents the mean of 3-5 spectrometric readings.
of minimum absorption between these two maxima is at 564-565 m^.. A compari- son of the spectral curves of the pigments, in the oxidized state, from all three species is shown in Figure 2. It is evident that all three pigments show two maxima, but the maxima for Eudistylia and for Molpadia are shifted toward the red end of the spectrum when compared with the Cucumaria data, the shift being greater in the case of Molpadia. The position of the maximum and minimum points for all the curves that were obtained from all three organisms are listed in Table I.
Reduction of the pigment by means of sodium hyposulphite results in a radical, though reversible, change in the spectral curve for all three pigments (Fig. 3). Tn the case of Cucumaria the original bands disappear and two new bands, one at 562-563 m/x, the other at 530-532 mp., make their appearance. The point of mini-
ABSORPTION SPECTRA
33
mum between these two maxima is located at 545 niju, (Table I). In Eudistylia, reduction results in the disappearance of the two original bands and in the appear- ance of a band with a peak at 577-580 m/u,. There is also an indication of a second- ary band at about 540 m/A (Fig. 3) but the data on this point are too meagre to merit any degree of confidence. Reduction of the pigment from Molpadia, again with hyposulphite, leads to the replacement of the two original bands by a single band at 588 m^. It is possible that, here too, more data would reveal the existence of a secondary band since an indication of this is visible in the curve for Molpadia
sao s»o wo 010 6eo 6-so 6*0 eso «eo
QO 8 I I I
MO see B.SO
FIGURE 2. The absorption curves of the oxidized pigment from Molpadia intermedia (A), Endistylia gigantca (B), and Cnctunaria miniata (C). The curves are plotted on a multiple ordinate scale. Other details as Figure 1.
S£O SAO 5*0 bad 6«O S7O 560 S»O 6OO 61O 6£O 6-5O MO 65O 66O
FIGURE 3. The absorption curves of the pigment after reduction with sodium hyposulphite from Eudistylia c/iyantea (A), Molpadia intermedia (B), and Cucnmcria miniata (C). The curves are plotted on a multiple ordinate scale. All other details as in Figure 1.
34
FREDERICK CRESCITELLI
TABLE I
|
Species * |
Oxidized state |
Reduced state |
|||
|
a |
0 |
Minimum * |
Main |
Secondary |
|
|
Caudina chilensis |
579.5 |
544.2 |
564.5 |
560 |
|
|
Molpadia roretzii Cucumaria frauenfeldi Spirographis spallanzi (crystals) |
577 579 602.5 |
541.5 543 562.5 |
562 |
557. 558 |
|
|
Eudistylia gigantea 1 2 |
605 604 |
555 555 |
576 576 |
580 577 |
540 |
|
3 |
603 |
558 |
577 |
||
|
4 |
602 |
554 |
575 |
||
|
Cucumaria miniata 1 |
581 |
545 |
564 |
562 |
|
|
2 |
580 |
545 |
564 |
561.5 |
530 |
|
3 |
580 |
544 |
564 |
562 |
531 |
|
4 |
580 |
544 |
565 |
562 |
531 |
|
5 |
580 |
547 |
565 |
562 |
531 |
|
6 |
582 |
545 |
565 |
563 |
532 |
|
7 |
563 |
535 |
|||
|
Molpadia intermedia 1 2 |
615 612 |
570 570 |
592 592 |
588 588 |
|
|
3 |
611 |
568 |
595 |
" The position of the minimum between the a- and /3-bands.
The data from Caudina and Molpadia roretzii were taken from Kobayashi (1932).
The data for Cucumaria frauenfeldi were taken from Hogben and Van Der Lingen (1928).
The data for Spirographis were taken from Roche and Fox (1933).
(Fig. 3). It is clear, then, that reduction of the pigment from all three organisms leads to a hypsochromic shift, but the shift is not one of equal magnitude for all three organisms, since the span from the a-band to the principal band of the reduced pig- ment is about 18 in/*, for Cucumaria, 25 in/A for Molpadia and 26 m/A for Eudistylia.
It seems very likely that the pigment in the red cells of Cucumaria miniata is erythrocruorin. The positions of the a- and /3-bands, as well as the point of mini- mum, agree reasonably well with the previous data obtained with other holothurians (Table I). The reduced pigment from Cucumaria shows two bands; the main band at 562-563 m/A agrees approximately with the 557-560 m/A band previously re- ported for holothurians (Table I ), but the secondary band at 530-532 m/A is previ- ously unreported. although from the plasma pigment of certain worms (Kobayashi, 1936) a spectral curve has been obtained which lias a 2-banded structure with the secondary band in the 549-551 m/A region.
It also seems likely that the green pigment dissolved in the plasma of Eudistylia yigantea is chlorocruorin. The a-band at 602-605 m/A (Table I) agrees completely with the data given by Fox (1925) and by Roche and Fox (1933), although the present data shows the /?-band to be shifted toward the violet end of the spectrum by about 6-7 m/A as compared with the position of the /?-band given by the above-
ABSORPTION SPECTRA 35
mentioned workers. The principal band of the reduced chlorocruorin from Eudistylia has a peak at 577-580 nip. whereas the data of Fox (1925) for Spiro- graphis place it at about 574 nip.
The results obtained from Molpadia intermedia are surprising. At the outset there was no reason to suspect that an absorption curve agreeing approximately with that obtained by Kobayashi (1932) from Molpadia rorctzii, and indicating an erythrocruorin, should not be obtained. Instead, as Figures 2 and 3 indicate, the bands for both the oxidized and reduced pigment are shifted toward the red end to a degree even greater than in the case of chlorocruorin. The a-band of Molpadia intermedia occurs 36 nip. further toward the red than the corresponding band for Molpadia rorctzii, while the band of the reduced pigment of Molpadia intermedia is 31 nip. further toward the red than in Molpadia rorctzii. It is also significant to note that the span between the a- and /?-bands in other sea cucumbers (Table I) is 35- 36 nip. whereas the corresponding span in Molpadia intermedia is about 43 n\p. These differences appear to be too great to be accountable to errors in measure- ment or to the usual species differences that are known to occur. The spectrum of the pigment from Molpadia intermedia does not agree with the spectrum of hem- erythrin, the red pigment found in certain Gephyrean worms as well as in the polychaete, Magelona (Marrian, 1927). It must be concluded that either this represents the true absorption spectrum of a heme pigment (if it is a heme pig- ment) characteristically different from either erythrocruorin or chlorocruorin. or that unrecognized conditions cause a shift of the bands from the typical positions of erythrocruorin. The onset of war resulted in the sudden interruption of the investigation at this point so that this final question must remain unanswered till a later date. A thorough examination of the crystallized pigment by chemical as well as by spectroscopic means should be made before the existence of a new pigment can be accepted.
SUMMARY
A spectrometric examination of the green pigment dissolved in the plasma of the tube worm, Eudistylia gigantea, and of the red pigment in the corpuscles of the sea cucumbers, Cucumaria miniata and Molpadia intermedia, has led to the follow- ing conclusions.
1. The green pigment of Eudistylia appears to be chlorocruorin. In the oxi- dized state it possesses an a-band at 602-605 nip. and a /?-band at 554-555 nip.. In the reduced condition a main band occurs at 577-580 nip, with a second band sug- gested at 540 nip..
2. The red pigment of Cucumaria appears to be erythrocruorin. In the oxidized state an a-band at 580-581 nip. and a /?-band at 544-545 m^ are seen. When the pigment is reduced a band appears at 562-563 nip. as well as one at 530-532 nip..
3. The red pigment of Molpadia, when oxidized, possesses a band at 611-615 nip. and another at 568-570 nip,. In the reduced condition a band at 588 nip. is evident. This spectrum does not agree with the spectrum of either chlorocruorin or of erythrocruorin. The new spectrum may indicate the existence of another pigment with the ability to combine reversibly with oxygen.
36 FREDERICK CRESCITELLI
LITERATURE CITED
BARCROFT, T., AND H. BARCROFT, 1924. The blood pigment of Arenicola. Proc. Royal Soc. London. B. 96: 28-42.
DA\VSOX, A. B., 1933. Supravital studies on the colored corpuscles of several marine inverte- brates. BioL Bull., 64: 233-242.
Fox, H. M., 1925. Chlorocruorin. A pigment allied to hemoglobin. Proc. Roy. Soc. London, B, 99: 199-220.
HOGBEN, L., AND J. VAN DER LiNGEN, 1928. On the occurrence of haemoglobin and of erythro- cytes in the perivisceral fluid of a holothurian. Brit. J. E.vp. BioL. 5: 292-294.
KOBAYASHI, S., 1932. The spectral properties of haemoglobin in the holothurians, Caudina chilensis (J. Miiller) and Molpadia roretzii (V. Marenzeller ) . Science Reports. Tohoku Imperial University, Series IV, 7: 211-227.
KOBAYASHI, S., 1935. The spectral properties of haemoglobin in the mulluscs. Area inflata (Reeve) and Area subcrenata (Lischke). Science Reports, Tohoku Imperial Univer- sity, Series IV, 10 : 257-267.
KOBAYASHI, S., 1936. The spectral properties of haemoglobin in the earthworms, Pheretima communissima (Goto et Hatai) and Pheretima hilgendorfi (Michaelsen). Science Re- ports. Tohoku Imperial University, Series IV, 10: 733-751.
LANKESTER, F. R., 1868. Preliminary notice of some observations with the spectroscope on animal substances. /. Aunt, and Physio!.. 2: 114-116.
MARRIAN, G. F., 1927. A note on haemerythrin. Brit. J. E.rp. BioL. 4: 357-364.
OHUYE; T., 1936. On the coelomic corpuscles in the body fluid of some invertebrates. Science Reports, Tohoku Imperial University, Series IV, 11: 207-222.
REDFIELD, A. C., 1933. The evolution of the respiratory function of the blood. Quart. Rev. BioL, 8: 31-57.
REDFIELD, A. C., AND M. FLORKIX, 1931. The respiratory function of the blood of Urechis caupo. Blol Bull., 61: 185-210.
ROCHE, J., AND H. M. Fox, 1933. Crystalline chlorocruorin. Proc. Roy. Soc. London, B, 114: 161-167.
SATO, T., AND H. TAMIYA, 1937. Uber die Atmungsfarbstoffe von Paramecium. Abst. in BioL Abst., 1941, 15. Abst. No. 3788.
SVEDBERG, T., AND I. B. ERiKssox, 1933. The molecular weight of erythrocruorin. /. Am. Chen,. Soc., 55: 2834-2841.
GUSTATORY REJECTION THRESHOLDS FOR THE LAKVAE THE CECROPIA MOTH, SAMIA CECROPIA (LINN.)
HUBERT FRINGS Biological Laboratories of West J'irc/inia Wesleyau College, Buckhannon, IfV.s'f I'iryinia
INTRODUCTION
Dethier (1937, 1939) and Eger (1937) have so far conducted the only studies on gustation in lepidopterous larvae. Dethier, interested in localizing the gusta- tory receptors and in gustatory acuity and its relation to host selection by cater- pillars, has determined acceptance thresholds for certain sugars and rejection thresholds for hydrochloric acid for a number of species. Eger, interested chiefly in determining the modalities of taste for caterpillars, has determined acceptance thresholds for some sugars and rejection thresholds for hydrochloric acid, sodium chloride, acetic acid, quinine hydrochloride, and saccharine. He concludes that there are probably only two modalities for caterpillars — acceptable and unaccepta- ble, but he admits that his data offer no definite proof for this. The following ex- periments were undertaken to add more information to that already existing in this little known field.
MATERIALS AND METHODS
The larvae of Saniia cccropia used in the following experiments were reared from a single group of eggs laid by a gravid female captured in the early summer, the food plant being the sycamore (Platanus occidcntalis). They were at first confined in a large jar and fresh leaves were added as needed, but later they were put on individual branches and numbered by marking with drops of colored paint. All the animals used in the experiments described below were in the last instar.
Rejection thresholds were determined, without disturbing the larvae, by placing drops of test solutions on the leaves as the animals fed. Since the caterpillars eat leaves in a very orderly fashion, it is easy to place drops so that they are reached by them within a very short time, usually about 10 to 15 seconds. Drops were placed on the leaves from small, new and carefully washed artists' brushes kept in the test solutions. A definite attempt was made to have the drops as nearly the same size as possible, but the importance of this is doubtful, for the larvae ordinarily ate on, undisturbed, below their thresholds and reacted almost immediately upon contact when the solutions were concentrated enough to bring about rejection.
The test solutions were prepared from C.P. chemicals, unless otherwise noted, using volumetric apparatus, and all tests were run within three hours after the solutions were prepared. The following chemicals were used, the concentrations prepared being given in parentheses after each: sucrose (2M, 1M, 0.5M), glucose, U.S.P. (2M, 1M, 0.5M), lactose, U.S. P. (1M. 0.5M), strychnine sulfate, U.S.P. (saturated solution), hydrochloric acid (0.4N, 0.2N, 0.1N, 0.075N, 0.050N. 0.025N), acetic acid (0.8N, 0.4N. 0.2N. 0.1 N, 0.075N, 0.050N), sodium hydroxide
37
38
HUBERT FRINGS
(0.4N, 0.2N, 0.1N, 0.075N, 0.050N), sodium chloride (2N, IN, 0.75N, 0.50N, 0.25N), lithium chloride (4N. 2N, IN, 0.75N, 0.50N), calcium chloride (IN, 0.75N. 0.50N, 0.25N, 0.10N), ammonium chloride (IN. 0.75N, 0.50N, 0.25N. 0.10N), and potassium chloride (IN, 0.75N, 0.50N. 0.25N, 0.10N).
There was no particular order in which these solutions were presented to the individual larvae. Since tests could he carried on only during feeding, it was found convenient to use three or four solutions simultaneously with all the caterpillars, testing the animals, whenever they fed, with the various concentrations. Thus there was no chance for associations to be built up by the caterpillars through con- sistent use of ascending or descending orders of concentrations. Each of the thresholds was determined at least three times for each of the caterpillars, and, in most cases, was determined five or six times, the greatest number for any individ- ual being nine times. Acceptances are easy to determine, because the animals eat three or four "cuts" through the drop in their feeding. Rejections are usually equally sharp, because the caterpillars stop feeding immediately upon touching the drop, withdraw and start feeding elsewhere. It was possible, therefore, although only twelve specimens were used, to make accurate determinations for these solu- tions under these conditions.
The temperature of the laboratory was 23° C. and the relative humidity 70 per cent throughout all the work. Ordinary daylight entering the windows was the source of illumination.
RESULTS
There were no rejection thresholds for sugars or for strychnine sulfate, the larvae eating the most concentrated solutions of these offered to them. The thresholds for acids and salts used in these experiments are given in Table 1.
TABLE I
|
Substance |
Range |
Mean |
i |
|
Threshold |
|||
|
HC1 |
0.04-0.15 N |
0.083 ±. 008 N |
12 |
|
CH3COOH |
0.09-0.6 N |
0.46 ±.05 N |
|
|
NaOH |
0.09-0.15 N |
0.13 ±.008 N |
7.7 |
|
KC1 |
0.2 -0.9 N |
0.45 ±.05 N |
2.2 |
|
NH4C1 |
0.4 -0.6 N |
0.46 ±.03 N |
2.2 |
|
NaCl |
0.4 -1.5 N |
0.89 ±.1 N |
1.1 |
|
CaCl2 |
0.2 -0.9 N |
0.61 ±.05 N |
1.6 |
|
I.iCl |
0.9 -3.0 N |
1.4 ±.15 N |
0.72 |
In the column headed, "Range," are the lowest and highest thresholds dis- covered for the group ; this does not imply a uniform distribution between these two extremes, but it gives some indication of the individual variations. In the column headed, "Mean," are the mean threshold normalities and the standard errors of these means (probable errors would be .6745 times the standard errors). In the column headed, "I/Threshold," are the reciprocals of the thresholds, these figures to be used below as indicative of stimulative efficiencies.
GUSTATORY REJECTION THRESHOLDS
Thresholds were calculated as follows. The two critical concentrations — that is, the pair of concentrations, the lower of which was accepted, the higher rejected— were determined for each larva. Obviously, the true threshold lies between these two, and, for statistical purposes, it was assumed that the threshold lay midway between them. Thus, if a larva accepted 0.25N and rejected 0.50N, the threshold assigned was 0.38N, if between 0.50N and 0.75N, the threshold assigned was 0.63N. Since the second digit cannot be interpreted literally in such a system, these thresh- olds are designated in the table as 0.4N and 0.6N respectively, only the means being given to the second place.
While the validity of using the midpoint betweeen the two critical concentra- tions as the threshold for an individual animal can be questioned, certain supple- mentary observations and considerations made it appear permissible to use this for the calculation of means. Thus, most of the animals readily ate through the drop of lower concentration and stopped clearly on reaching the drop of higher con- centration, indicating a threshold in the middle range between the two. A few, however, hesitated at the lower concentration before continuing with feeding and reacted violently at the higher, drawing back sharply and often ceasing feeding for a time, indicating a threshold near the lower concentration. Others ate readily through the lower concentration and only hesitated at first at the higher, made a few tries at eating it, but finally stopped and usually simply moved a short distance away on the same leaf, indicating a threshold near the higher concentration. Since these latter two cases seemed to be of equal occurrence, with the majority of rejec- tions of the first type described, it seemed fair, for purposes of calculation of means for the group to be used for comparative purposes only, to assign the midpoints between the critical concentrations as the thresholds for the individuals, in the group, without implying that these represent true thresholds for any of the individuals.
DISCUSSION Sugars and strychnine siilfatc
Since there were no rejection thresholds for sugars or for strychnine sulfate, no conclusions can be drawn concerning the ability of the caterpillars to taste these substances. It must be recognized that failure of rejection does not imply that the animals cannot taste these compounds, just as rejection thresholds do not measure the lowest concentrations that can be tasted. To me, sycamore leaves tasted almost as bitter as the strychnine solution, and, since these larvae feed on many species of trees, it would seem logical that their rejection threshold for bitter substances would be high. Dethier (1937) has shown definitely that caterpillars of this species can taste glucose and sucrose, but he used acceptance thresholds in his work.
Acids
The acid thresholds here given are higher than those given by Dethier and Eger, but that is to be expected, for the acids in their experiments were offered in drops of water, while in this case there was some mixing of the acids with the food. The pH of the solution of hydrochloric acid corresponding to the mean rejection thresh- old normality is 1.2, while the pH of the solution of acetic acid at the mean rejec- tion threshold normality is 2.5. Thus, acetic acid, as for man. is much more
40
HUBERT FRINGS
effective a stimulating agent than is hydrochloric acid at the same pH. The ratio for this species is 20: 1. which compares favorably with Eger's (1937) report for another species of a ratio averaging about 23 : 1. This ratio is comparable to that for humans (28: 1), and Eger suggests that it indicates a similarity in the buffer- ing capacities of the salivas of the two species. It further, however, suggests a similarity of action in both, and may indicate a sensitivity to hydronium ion in these caterpillars corresponding to the sour taste in humans.
Salts
The order of relative stimulative efficiencies, as measured by the reciprocals of the normalities of the rejection limens, for the various cations, as chlorides, gives the series : NH4+ : K+ > Ca++ > Na+ > Li+.
This series can be verified further by checking the orders for the twelve in- dividuals to determine in how many cases any specific cation has an equal, greater,
E.5
£.0
l.5
,0
0.5
30 40 50 60 70 60
LIMITING EQUIVALENT IONIC CONDUCTANCES
FIGURE 1. Showing the relationship between the stimulative efficiencies of the cations, as chlorides, and their limiting' equivalent ionic conductances in Mhos at 25° C. Values for the latter are from Gucker and Meldrum (1942).
or lesser stimulative efficiency than any other cation. Using this method, it is found that K> has a greater efficiency than NH4+ (K+ ] > NH4+ ) in 4 individuals, NH4+ has greater than K> (K+ < ; NH4+) in 3 individuals, and K+ is equal to NH4+ (K+"NH4+) in 5 individuals. This indicates that K+ and NHt+ are equal in stimulative efficiency. Comparing K+ and NH4+ with Na+, the following orders are noted : 1C ; > Na+ - 9. K+ : = Na+ -- 3, K+ < ; Na+ - 0, and NH/ ; > Na+ - 8, NH.,+ = Na^ 4, NH4+ Na+ -- 0. These clearly show that K+ and NH4+ have greater stimulative efficiencies than Naf, and further confirm their equality. Comjjaring Na1 with Li', the following are found: Na+ > Li' -9, Na1" - Li+ - 3, Xa+ < ; Li+ -- 0. Obviously. Xa is more stimulating than Lif. K+ and NH4+ are
GUSTATORY REJECTION THRESHOLDS
41
found to be more stimulating than Li+ in all cases, thus verifying their position above Na+. Ca++ shows the greatest variation in position. Comparing with NH4+ or K+, the following are found : K+ or NH4+ ; > Ca++ - 8, K+ or NH4+ = Ca++ - 2, K+ or NH4+ < Ca++ - 2. Thus K+ and NH4+ have greater stimulative effects than Ca++. Comparing Ca++ with Na+ we find : Ca+f ; > Na+ - 9, Ca++ - Na+ - 0, Ca++ < Na+ -- 3. Ca++ is thus more stimulating than Na+. With Li+ we find: Ca++ > Li+ - 11, Ca++ = Li+ -- 1, Ca++ < ; Li+ -- 0; Ca++ is thus more stimulating than Li+. All of these individual orders considered together indicate the same arrange- ment as that found using the mean thresholds, namely : NH4+ - K+ > Ca++ Na+
i-»
m
.-••
o o
NORMAL
— 1 cn
7
ui o
rv ui
100 200 300
LIMITING EQUIVALENT IONIC CONDUCTANCES
FIGURE 2. Showing the relationship between the stimulative efficiencies and the limiting equivalent ionic conductances in Mhos at 25° C. of certain ions. The slope is taken as de- termined by the metallic ions. Values for the conductances are from Gucker and Meldrum (1942).
Hopkins (1932) has tabulated the discoveries of numerous investigators con- cerning stimulative efficiencies of various cations and anions. The order shown in the case of the caterpillars of Samia cccropia here used is similar to that discovered for other species of animals. This suggests some basic chemical or physical rea- son for this order. Hopkins relates the order of cations for the oyster, which he studied, to the atomic weights and atomic mobilities. In the present case, there is no obvious relationship between ionic weights and stimulative efficiences, but Figure 1 shows the close agreement between the stimulative efficiencies of the cations, as determined by the reciprocals of the normalities of the rejection limens, and their limiting equivalent ionic conductances which are directly proportional to the ionic mobilities, in fact, have sometimes been called mobilities. Since the chloride ion is the anion in all cases, it is impossible to determine its relative stimulative effect, and
42 HUBERT FRINGS
it is. therefore, disregarded in this plot, since, whatever its efficiency, it is constant for all substances tested.
These relationships suggest a common mode of action, and possibly a common modality of taste, for all the cations tested. Since, however, this same order is found for man for K+, Na1, and Li+, and KC1 has a taste easily distinguishable from NaCl or LiCl, this conclusion is not warranted without further evidence.
Sodium hydroxide shows a much greater stimulative efficiency than any of the other salts : used in this work, thus indicating that the OH" ion is the critical one in its effectiveness, since some measure of the efficiency of the Na+ ion is given by the stimulative efficiency of NaCl. Comparison of the data for NaOH and HC1 shows that the OH~ ion is much less stimulating than the H3O+ ion.
Figure 2 shows the extrapolation of the graph, stimulative efficiency vs. limiting equivalent ionic conductance to include OH" and H3O+. The value for the stimu- lative efficiency of the OH~ ion is approximated by subtracting from the stimulative efficiency for NaOH the stimulative efficiency for NaCl, assuming thus that most of this is due to the Na+ ion. This shows that there is a possibility that OH~ and H.,O+ fall in the same series. That they are apparently lower in stimulative ef- ficiency than is needed for a perfect "fit" in the graph would be expected from the fact that the solutions containing them were mixed with the protoplasm of the leaves and with the saliva of the caterpillars, both of which are buffered. This might mean that there is only an "unacceptable" modality of taste for all these sub- stances, as Eger (1937) suggests, but this cannot be decided without further evi- dence. The OH" and H3O+ ions, in the case of man, show a high stimulative efficiency when compared with other anions and cations, and stimulation by the H3O+ ion, at least, is admitted to be the important factor in the sour taste, as op- posed to the salt taste of the metallic chlorides.
In all, these results can only be taken as suggestive and indicative of a need for much more extensive and careful work on the gustatory responses of animals to more than the usual NaCl in testing the salt taste.
SUMMARY
Rejection thresholds for HC1, CH3COOH, NaOH, NaCl, NH4C1, KC1, CaCL, and LiCl, presented as drops of solutions on leaves of the food plant, were de- termined for caterpillars of the cecropia moth, Sainia cccropia. Rejection thresh- olds for glucose, sucrose, lactose, and strychnine sulfate either do not exist under these conditions, or are higher than the saturation concentrations of solutions of these substances. The lowest threshold of those tested is that for HC1, but CHr!COOH has greater stimulative efficiency when it is compared with HC1 at the same pH. The threshold for NaOH is higher than that for HC1, indicating that the OH~ ion is less stimulating than the H3O+ ion. The order of stimulative efficiency for the cations, as chlorides, is NH4+ == K+ > Ca+" ; > Na+ > LiH. This is the order of ionic mobilities to which the stimulative efficiencies seem to be related. No conclusions can be drawn with certainty as yet regarding the modali- ties of taste for these animals.
1 Modern Acid-Muse concepts require that NaOH he considered as a salt.
GUSTATORY REJECTION THRESHOLDS ' 43
ACKNOWLEDGEMENTS
It is a pleasure to express my appreciation for the many kindnesses shown me during this work by Dr. Nicholas Hyma, Head of the Department of Chemistry, and Dr. Arthur B. Gould, Professor of Chemistry, at this institution. An especial debt is due my wife who aided me in all phases of the work.
LITERATURE CITED
DETHIER, VINCENT G., 1937. Gustation and olfaction in lepidopterous larvae. Biol. Bull., 72:
7-23.
DETHIER, V. G., 1939. Taste thresholds in lepidopterous larvae. Biol. Bull., 76: 325-329. EGER, HEINZ, 1937. t'ber den Geschmackssinn von Schmetterlingsraupen. Biol. Zentr., 57:
293-308.
GUCKER, F. T., AND MELDRUM, W. B., 1942. Physical chemistry. New York. HOPKINS, A. E., 1932. Chemical stimulation by salts in the oyster, Ostrea virginica. /. £.v/>.
Zool, 61 : 13-28.
BACKGROUND ILLUMINATION AS A FACTOR IN THE ATTACH
MENT OF BARNACLE CYPRIDS
JAMES H. GREGG
L'nircrsity />/ Miami Mtirine Laboratory. Centra Research Laboratories, and the ruirersit
of .-
INTRODUCTION
Various authors, following Visscher (1927), have shown that certain species of barnacles attach in greater numbers to darker surfaces. McDougall (1943), by means of an experiment designed to test the effect of varying degrees of illumina- tion, demonstrated that larvae of Balanus cbuniciis have a tendency to settle most abundantly upon collectors placed in less brightly illuminated areas. Pomerat and Reiner (1942) showed that Balanus cJuinicns larvae attach more readily to black rather than to opal panels when they are exposed in the sea for several days. These investigators also observed that when light is at a minimum, during the hours of darkness, the differences do not occur but that attachments are remarkably similar on light and dark panels. This suggested that attachment to darker panels during daylight is a phototropic response to the contrasting effect of dark surfaces against lighter general surroundings. The purpose of the present study was to investigate whether in fact, contrasting surroundings are effective in promoting attachment. A further objective was the determination of the maximum distance at which the degree of illumination of a background surface is effective in influenc- ing attachment to transparent surfaces.
The experiments were conducted at the Pensacola, Florida laboratory of the U. S. Fish and Wildlife Service by the kind permission of Dr. A. E. Hopkins during the summer months of June, July, and August. 1942. Acknowledgments are also due to Dr. C. M. Pomerat for his generous advice, and to Dr. F. G. Walton Smith, Director of the University of Miami Marine Laboratory, for assistance in preparing the paper and in evaluating the results.
EFFECT OF CONTRASTING BACKGROUNDS
In the experiments designed to determine whether contrasting surroundings influence cyprid attachment, provision was made for base plates of black, trans- parent, and opal Carrara glass plates 10" 12" in size. Upon each of these were mounted 4" : 10" black and opal Carrara glass collecting panels, in such a way as to be surrounded by a 2 inch border of the base panel.
The plates were supported by a wooden frame. The glass bases were arranged one inch apart and the paired black and opal collectors were separated by \\vo inches. Tims, a series of contrasts was offered between the pairs of black and opal plates and the three surrounding borders of black, opal, and transparent glass. The barnacles were counted following attachment upon the collectors, each of which offered an exposed area of 36 sq.in. In the locality where the investigations
44
THE ATTACHMENT OF BARNACLE CYPRIDS Transparent Base Opal Base Black Base
45
\\x\\v
Black Collectors
Opal Collectors
FIGURE 1. Arrangement of black and opal collectors upon black, opal, and transparent ba^< s.
were carried out the adults of Balanus cbnniciis were the harnacles most commonly found, and all attachments were assumed to belong to this species.
The apparatus for the first series of experiments was lowered into the sea with the plates in a vertical position at a mean depth of three feet from the surface. Six successive experiments were conducted with an average duration of 36 hours. At the end of each period the paired plates were removed and washed gently with fresh water to remove salt crystals and silt. Counts were then made of the cyprids and of the metamorphosed harnacles. with the results shown in Table I.
Further experiments were conducted with the plates exposed horizontally, in order to collect attachments on the lower surface. The mean depth of the plates below the surface of the sea was seven feet. Five experiments were conducted,
TABLE I
Barnacles attaching to vertical collectors with borders of varying contrast
|
Experiment |
Date of exposure |
Black collector |
Opal collector |
||||
|
Borders |
Borders |
||||||
|
Transparent |
Opal |
Black |
Transparent |
Opal |
Black |
||
|
I |
7- 2-44 |
68 |
67 |
35 |
15 |
17 |
8 |
|
II |
7- 4-44 |
300 |
293 |
458 |
249 |
183 |
233 |
|
III |
7- 6-44 |
73 |
115 |
59 |
8 |
5 |
7 |
|
IV |
7- 8-44 |
120 |
175 |
121 |
15 |
38 |
35 |
|
V |
7-10-44 |
89 |
97 |
86 |
17 |
7 |
25 |
|
VI |
7-12-44 |
137 |
152 |
162 |
23 |
23 |
18 |
|
Total |
787 |
899 |
921 |
327 |
273 |
326 |
|
|
Average |
131 |
150 |
154 |
55 |
46 |
54 |
46
JAMES H. GREGG
TABLE 11
Barnacles attaching to horizontal collectors with borders of varying contrast
|
Experiment |
Date of exposure |
Black collector |
Opal collector |
||||
|
Borders |
Borders |
||||||
|
Transparent |
Opal |
Black |
Transparent |
Opal |
Black |
||
|
VII |
7-17-44 |
653 |
503 |
703 |
80 |
114 |
171 |
|
VIII |
7-20-44 |
757 |
692 |
711 |
94 |
142 |
194 |
|
IX |
7-23-44 |
582 |
720 |
686 |
142 |
249 |
340 |
|
X |
7-26-44 |
620 |
855 |
701 |
254 |
328 |
437 |
|
XI |
7-31-44 |
526 |
582 |
564 |
194 |
230 |
303 |
|
Total |
3138 |
3352 |
3365 |
764 |
1063 |
1445 |
|
|
Average |
628 |
670 |
673 |
153 |
213 |
289 |
with exposure periods varying from 72 to 120 hours. The results are shown in Table II.
Upon horizontal and vertical black collectors and regardless of the nature of the surroundings, greater attachments occurred upon the black collectors as compared with the opal. The total number of organisms on all black horizontal collectors amounted to 9855 compared with a total of 3272 for opal horizontal collectors, giving a black/opal distribution ratio of 3.0. The total populations on black and opal vertical collectors were 2607 and 926 respectively, giving a ratio of 2.8. These ratios, though greater, qualitatively substantiate the findings of Pomerat and Reiner (1942) whose ratio for attachment numbers upon black and opal collectors was 1.8 (Table III).
TABLK III
Total number of barnacles on black and opal collectors
|
Black collectors |
Opal collectors |
Black collectors |
|
|
Opal collectors |
|||
|
Horizontal |
9855 |
3272 |
3.0 |
|
Vertical |
2607 |
926 |
2.8 |
Attachments to black vertical collectors did not vary greatly with the type of surrounding, but were least in the case of the transparent border (Table I). With black collectors, borders of black, transparent and opal, in this order, afford in- creasing contrasts. I lad the contrast been effective, greater attachments should have occurred upon collectors with the opal and transparent borders rather than with the black border. Attachments upon black horizontal collectors also tailed to
THE ATTACHMENT OF BARNACLE CYPRIDS 47
show significant differences when the three different surroundings were employed, although they were least in the case of a transparent border (Tahle II).
The frequencies of attachment upon vertical opal collectors did not show any correlation with increasing degree of contrast but were similar in all instances (Table I). The horizontal opal collectors gave somewhat different results (Table II). In this series of experiments the greatest attachments were found on the collectors with the black borders. The least numbers, however, were observed on collectors with transparent borders and intermediate attachment frequency occurred with opal borders. In these experiments the least degree of contrast was offered by opal surroundings of similar material to the collectors and the greatest degree by black surroundings which afforded the minimum possibilities of transmitted and reflected light. There was not, therefore, a consistent correlation between the intensity of attachment and the degree of contrast between collector and sur- roundings.
In an attempt to explain the results, the question of variation in the general intensity of illumination beneath the panels may be considered. The amount of transmitted and reflected light was greatest around horizontal opal collectors with transparent surroundings. The reflected light which alone fell below opal collectors with the opal borders was less intense than the light in the former case and it was under this condition that frequency of attachment was found to be increased. The opal collectors with black borders, which reflected the least light of the three types of surrounding were found to have the greatest attachment. Thus, with the de- crease in general illumination beneath opal collectors, there was an increase in attachments, and it appears that attachment frequencies are here related to the degree of shading under the collectors caused by the varying opacity and reflection of the surroundings.
The horizontal black collectors were found to have distinctly greater attach- ments when combined with a black border. They also showed least attachments with a clear border. The black collectors, therefore showed a correlation, not with contrast but rather with the degree of shading which is at a maximum where both collector and surroundings are black.
In the case of both black and opal vertical collectors, the shading effect of the border is less pronounced, due to the fact that much of the light enters the water at a more or less vertical angle. The hypothesis that general shading stimulates attachment would fit in with the observed facts about attachments to these series. The difference between the amount of light reflected by black or opal collectors is more significant than that reflected by the relatively small borders, and hence accounts for the relatively small differences among attachments to black collectors with different surroundings or among attachments to opal collectors with varied surroundings. The differences between attachments to the two types of collector with each surrounding are relatively greater.
DISTANCE AT WHICH BACKGROUND ILLUMINATION Is EFFECTIVE
The equipment designed for the question of determining the maximum distance at which black or opal surfaces are effective in stimulating a cyprid to attach con- sisted of a wooden frame 30" long, 32" wide, and 12" high. This frame supported vertically two transparent glass plates. Behind the first glass plate was mounted a
48
JAMES H. GREGG
black plastic panel in such a manner that it might be moved to varying distances. The second transparent glass plate without any hacking was used as the control. A second series of transparent glass plates at the opposite side of the frame were treated in the same manner with the exception that an opal plastic panel was used instead of a black one. The movable black plastic panel and the movable opal plastic panel were set at increasing distances from the clear panel for successive experiments. Each transparent glass plate from which the barnacles were counted, offered 54 sq.in. of surface area. They were exposed at a mean depth of three feet from the surface for an average duration of 72 hours. In the successive experi- ments the black and opal plastic movable panels were placed at 2", 3", 4", and 6" behind the transparent collectors.
The results (Table IV) show a general decline in the attachments to the glass collector as the black movable plate was moved away. The maximum distance at
TABLE IV
Attachment of barnacles to transparent panels with movable backgrounds
|
Distance of background from collector |
Movable black |
Control transparent |
Ratio Movable |
Movable opal |
Control transparent |
Ratio Movable |
|
Control |
Control |
|||||
|
2 inches |
2998 |
778 |
3.9 |
1258 |
511 |
2.5 |
|
3 inches |
803 |
355 |
2.3 |
835 |
336 |
2.5 |
|
4 inches |
3759 |
1707 |
2.2 |
3686 |
2806 |
1.3 |
|
6 inches |
458 |
453 |
1.0 |
906 |
780 |
1.2 |
which it influenced attachment, as determined by comparison with the control, was between four and six inches. The same general decline in population was noted for the opal plates although the maximum influencing range was greater than six inches. Since attachment is influenced by the movable panels at a distance, it is obvious that the light reflection at the actual collecting surface is not the principal stimulating factor. The facts are more readily explained by assuming that the intensity of general illumination in the vicinity of the collector is the important factor and that where this is reduced by the proximity of black backplates greater attachment occurs. Decreasing distances of the opal backplate, while giving a superficial appearance of increased light intensity, probably succeed in blocking the light passing through the collecting panel. Thus increased attachment under these conditions may still be due to decreased general illumination.
These general conclusions agree with the results of the contrasting background experiments described in the first part of the paper where a relationship appeared to exist between the number of attachments and the amount of shading caused by the type of background employed. The decreased intensity of illumination in the general vicinity of the collector, as opposed to the conditions of illumination im- mediately upon the collector itself may increase attachment by causing a response in the cyprid. This response is in the nature of a conditioning towards subsequent attachment, or the development of a physiological state favoring a subsequent
THE ATTACHMENT OF BARNACLE CYPRIDS 49
attachment response to a contact stimulus. The possibility of a tropism is not supported by the contrast experiments.
In the light of the above conclusions it may be interesting to recall the experi- ments of Schallek (1943) and Whitney (1941) who maintain that directional light is not to any great degree present under natural aquatic conditions but that condi- tions of diffuse light almost invariably predominate. This diffuse light is a natu- rally occurring phenomenon, whereas light from any one direction may not pre- dominate sufficiently for a tropism to occur.
SUMMARY
1 . Experiments were conducted to determine the effect of contrasting surround- ings upon the frequency of attachment of Balanns cbnrncits larvae to opaque black and opal glass collecting surfaces. Further experiments were carried out to determine to what extent the number of attachments of transparent collectors was influenced by black and opal backgrounds placed at varying distances.
2. Greater numbers of attachments occurred upon the under side of horizontal rather than vertical, and upon black rather than opal collectors, thus confirming the observations of previous authors.
3. No correlation was found between the degree of contrast shown in the collector and surroundings, and the frequency of attachments.
4. Both black and opal surfaces were found to increase frequency of attach- ment when placed behind transparent collectors up to distances of six inches.
5. A definite dependence was found to exist between the frequency of attach- ment and a decrease in the intensity of general illumination in the area immediately beneath horizontal opal collectors. Similarly, the influence of movable back- grounds appeared to be in the nature of a shadow effect. It is suggested that "shading" acts as a stimulus which brings about favorable physiological conditions for the subsequent attachment of barnacle larvae and that the amount of light reflected from the collecting surface is only important insofar as it affects the general "shading" in the vicinity.
LITERATURE CITED
McDouGALL, K. D., 1943. Sessile marine invertebrates at Beaufort, North Carolina. Ecoloyicul
Monographs, 13 : 321-374. POMERAT, C. M., AND REINER, E. R., 1942. The influence of surface angle and of light on the
attachment of barnacles and other sedentary organisms. Biol. Bull.. 82 : 14-25. SCHALLEK, W., 1943. The reaction of certain Crustacea to direct and to diffuse light. Biol.
Bull.. 84: 98-105.
VISSCHER, J. P., 1927. Nature and extent of fouling of ships' bottoms. Bull. Bur. Fish: 43, II. WHITNEY, L. V., 1941. The angular distribution of characteristic diffuse daylight in natural
waters. Scars Found. Jour. Mar. Res.. 4: 122-131.
THE CONDITIONS THAT LEAD TO NORMAL OR ABNORMAL
DEVELOPMENT OF CIONA
T. H. MORGAN
irilliani (i. Kerckhoff Laboratories of the Hinl/u/ieal Sciences, California Institute of
Tcclinoloi/y. Pasadena, California
111 earlier papers I have discussed the problem of normal versus abnormal de- velopment of cross-fertilized eggs of Ciona. It was shown that external conditions sometimes influence the result ; at other times it was not apparent that the abnormal development was due to such conditions. However, by a critical series of experi- ments evidence was found that the eggs are very sensitive to what may be called the "cleanliness" of the dishes, i.e. to either chemical or organic contamination. Further experiments have been made to find out whether internal factors may also be concerned with abnormal development of the eggs.
Ciona is a particularly favorable type for study of this problem. The eggs, as soon as they are mature, leave the ovary and accumulate for 24 hours or longer in the oviduct. The germinal vesicle has disappeared and the polar spindle comes to lie at the pole of the egg. The polar bodies are not given off until a spermatozoon enters the egg, i.e. after the eggs have been set free in the sea water. This may happen every 24 hours or be delayed for several days. Spawning occurs early in the morning. If not ejected at the regular time the eggs may accumulate in the oviduct in very large numbers. When the animals are brought into the laboratory and kept in running sea water, or in sea water that is aerated by a stream of air bubbles, the Cionas may hold back their eggs for several days. It has been shown that these delayed eggs may give rise to normal embryos, but it has not been shown clearly that the percentage of abnormals may not be increased. Fresh eggs removed from the oviduct and kept for as long as 24 hours in covered dishes of shallow sea water can be cross fertilized and mav give normal embryos.
-• C3 f
The sperm are matured in the tubes of the testes that ramify over the walls of the intestine in the region of the outlet of the ovary. The ripe sperms pass into the long sperm duct that runs parallel to the oviduct. Only ripe sperms are present in the sperm duct, and some of them are set free when the eggs are ejected. When the eggs are collected by puncturing the oviduct, only mature eggs are set free, and likewise only ripe sperm comes out when the sperm duct is cut across. In most other marine animals, whose eggs are used for experimental purposes, one is apt to get ripe and unripe eggs when the ovary is opened. The uniformity of the eggs trom the- oviduct ol Ciona simplifies the experiment.
In another respect, also, Ciona otters favorable material for a study of normal versus abnormal development. It is well known that polyspermy often plays a significant note in laboratory experiments, unless it is carefully guarded against. In Ciona, polyspermy also occurs after cross-fertilization, but is very rare and will account for only a small percentage of abnormality. It can easily be detected at the two-cell stage when the egg divides into four or more parts if polyspermic.
50
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA
51
Even in cases where much abnormality occurs, I have found, by watching the cleavage, that polyspermy was not the main agent in the results.
Concerning the kind of abnormalities that occur, not much can be said. For convenience I have recognized three classes; (1) Normals, (2) Abnormals, and (3) Bent (or twisted) embryos. The abnormals are those that have not gone beyond an early stage of development. They are usually in the egg membrane and are spherical with one or two black "eye" specks. The twisted or bent are later stages in development, usually out of the egg membrane. The tail is present and crooked or bent, sometimes still coiled loosely or irregularly about the body. There are overlaps between the three different types. All three are generally found in the same dish, but the external or internal conditions that produce each kind are not known. It is safe to assume, however, that polyspermic eggs do not pass beyond the earliest stages of development, but external conditions may also stop the de- velopment in early stages.
Normal and Abnormal Development in Reciprocal Crosses
To test to what extent the ratios of normal to abnormal development are due to intrinsic (genetic?) factors or due to environmental conditions, further experi- ments were made in May 1944. Not only were reciprocal crosses made, but samples of the same eggs were crossed with three different sperm suspensions.
TABLE I
The percentage of normal embryos resulting from reciprocal crosses. The small numbers under the percentages represent, from left to right, the numbers of normal, bent, and abnormal embryos. The bent embryos have been left out of account in calculating the percentages.
|
csp |
b sp |
a sp |
|
|
0 |
93 |
8 |
|
|
0 0 148 |
126 0 8 |
12 42 132 |
D eggs |
|
70 |
89 |
99 |
|
|
67 9 28 |
49 2 6 |
128 0 1 |
E eggs |
|
59 |
96 |
49 |
|
|
73 33 49 |
310 1 12 |
51 7 54 |
F eggs |
|
fsp |
e sp |
d sp |
|
|
0 |
3 |
52 |
|
|
0 0 37 |
1 2 28 |
10 22 9 |
A eggs |
|
14 |
2 |
72 |
|
|
10 12 62 |
1 0 37 |
43 11 16 |
B eggs |
|
16 |
10 |
0 |
|
|
22 29 116 |
11 23 96 |
0 0 177 |
C eggs |
Some of the eggs of A were distributed in three dishes (horizontal lines. Table I), also some of the eggs of B and C, each to three dishes. These were cross fertilized with sperm of d, e and f. Two drops of sperm were added to each dish, and the normal, bent, and abnormal embryos recorded after 24 hours.
The eggs of D, E, and F had been put into three dishes each (horizontal lines) before the sperm of the same animals, d, e, and f, had been removed to cross-
T. H. MORGAN
fertilize the eggs of A, B. and C. Then two drops of the sperm suspension of a, b, and c were added to the D, E, and F eggs. These crosses are the reciprocals of the preceding ones.
In calculating the percentages, the bent embryos were left out of account. Most of them were so nearly fully developed that they might be classified with the normal, rather than with the abnormals. The bent embryos overlapped those abnormal embryos that were fairly well developed. The bent embryos would not have affected the percentages to any great extent if half of them had been combined with the normals and half with the abnormals.
As shown in Table I the percentages of normal embryos resulting from crossing A eggs by d sperm, e sperm and f sperm are very different (52, 3, 0%). This holds also for B eggs and C eggs by the same sperm. In other words, the results are different for the different combinations, and not consistent either for the eggs or for the sperm.
The reciprocal crosses are shown above in the same Table I. Comparing them, there is seen to be no agreement in the percentages. In other words, the reciprocals are no more likely to be alike than other crosses. Before discussing these results, another similar experiment will be described that was made the next clay with Cionas brought to Pasadena (Table II).
This experiment was carried out in the same way as the last, except that in addition to the reciprocals, three more dishes of A, B, and C eggs were set out and crossed with sperm from three other Cionas, viz. g, h, and i. These serve to give further data for the same eggs with different sperm. The reciprocals of g, h, i were not made. Comparing the reciprocals of the first three tests. D, E, and F. each set are more like each other than are the sperm tests a, b, and c, and also more like each other than in the first experiment, but still not the same inter se. The F eggs were much fewer than the others and produced relatively much smaller numbers of normals.
A comparison of the three kinds of eggs. A, B, and C each with the six kinds of sperm, d— i, shows that there are wide differences in the percentages (Table II). Since the environment was as nearly the same as possible, it may seem to follow that the differences are due to internal factors of some kind, possibly genetic in origin.
Why are the reciprocals so often different? The diploid eggs are presumably all genetically alike in each Ciona before fertilization, but after the polar bodies are extruded, the eggs become haploid, and should give as many kinds of eggs as there are combinations of the six pairs of chromosomes ( excluding the possibility of a larger number due to crossing over). If amongst these combinations there are certain combinations that might give abnormal development — lethals, in fact.— it might seem that the chances are about the same in the cross and in its reciprocal. In other words, even with many lethal factor combinations, any differences in reciprocal ratios would only be due to chance, unless the cytoplasm of the diploid egg is involved in the situation.
There are several more or less probable hypotheses that suggest themselves, and some of these may or may not have some connection with the theory of self- .sterilily in Ciona; for, il certain classes should be eliminated, those that remain and produce normal tadpoles might, theoretically, be the ones that carry factors for -elf-sterility.
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA 53
An examination of the data from the dishes of the same eggs (Imri/ontal lines; Table II), each fertilized by different sperm, show very great differences in most cases in the percentages of normals to abnormals. Occasionally, however, a set gives very low values (Table I, F eggs by (a), (b), (c) sperm) or very high rat in > (Table II, E eggs by (a), (b), (c) sperm. Also an examination of the ratios of the same sperm used to fertilize three different lots of eggs shows sometimes high percentages in all three dishes (Table II, (d) sperm by A, B, C eggs) and at other times low percentages in all three dishes (Table II, (g) sperm by A, B, C eggsi. On the other hand there are cases where each of the three sets of eggs, and also each of the three sets of sperm, show great differences in the ratios.
If the data are significant, i.e.. if the differences are not clue to infection in the
TABLE II
The percentages of normal embryos resulting from reciprocal and one way crosses. The small numbers under the percentages represent, from left to right, the numbers of normal, bent, and abnormal embryos. The bent embryos have been left out of account in calculating the percentages.
|
c sp |
b sp |
a sp |
|
97 |
76 |
91 |
|
71 2 1 |
287 8 35 |
152 3 14 D eggs |
|
98 |
95 |
98 |
|
303 12 6 |
256 3 9 |
289 1 1 5 E eggs |
|
3 |
2 |
0 |
|
1 0 32 |
1 0 48 |
0 0 34 F eggs |
|
i sp h sp g sp f sp |
e sp |
d sp |
|
99 37 3 99 |
96 |
82 |
|
319 03 69 34 118 8 17 258 236 1 1 |
225 0 9 |
133 26 29 A eggs |
|
2.8 0 0.2 76 |
58 |
97 |
|
9 65 280 0 8 286 10 403 360 72 112 |
136 24 85 |
400 6 14 B eggs |
|
0 50 12 47 |
91 |
88 |
|
0 0 101 36 32 35 187 30 29 33 |
200 1 24 |
82 7 22 C eggs |
dishes, these results show that the outcome is due neither to the eggs alone nor to the sperm alone, but to the combinations found at the time of fertilization.
In this connection there are some further considerations to be taken into account. The cytoplasm of the eggs has developed under the influence of the diploid set of chromosomes. If the constitution of this cytoplasm determines whether the de- velopment will be normal or abnormal, then the eggs of one individual might be expected to develop in the same way irrespective of what kind of sperm fertilizes the eggs. On the other hand, after the extrusion of the polar bodies, the haploid egg nucleus may have a very different make-up from the original diploid nucleus. But it is not likely that this haploid nucleus can affect or change the constitution <>t the cytoplasm in the very short time before it combines with the sperm nucleus to form a new diploid nucleus, whose products carry on throughout the following development. From the evidence discussed above it might seem, therefore, that whether the development is normal or abnormal will depend on whether the
54 T. H. MORGAN
combination of the two haploid genetic chromosome groups are harmonious or not in their action on the cytoplasm. If the cytoplasm is indifferent, then reciprocal crosses should give the same ratios, hut since they do not do so, it may seem to follow that the cytoplasm, which lias developed under the influence of the diploid egg nucleus, is one of the factors in determining whether normal or abnormal de- velopment takes place. This conclusion would mean that certain kinds of combina- tions are being continually eliminated. But it would not necessarily mean that these have anything to do with cross and self-fertilization of the combinations that survive. However, experiments that will be described later make it probable that the differences in the dishes reviewed in this experimnt are due to external factors, such as toxic infection, and are not due to inherited cytoplasmic differences in the eggs.
Test for Polyspermy and Abnormal Development
Five new sets of experiments were made (Table III) to test whether the abnormals that appear frequently are due to polyspermy, although it had often been recorded that the division of the eggs at once into four (or more) cells only very rarely occurs. Forty, 20 or 2 drops of sperm suspension were added to eggs in Syracuse dishes. After one hour the supernatant fluid was changed to fresh sea water. Reciprocal crosses were made in all cases. As the table shows. Table III, the relative number of normals to abnormals was no lower after 40 drops than after 2 drops of sperm suspension. Evidently polyspermy is not the cause of abnormal development. There are several striking exceptions to the general results. One (June 18-19. 1944), A1-!)1, gave after 2 drops almost all abnormals (328), although after 40 or 20 drops nearly all were normal. Again (June 19-20, A2-b-, in one case after 20 drops all were abnormal (295) ; and in the reciprocal after 40 drops all were abnormal (199), although after 20 and 2 drops nearly all were normal. In another test (June 18-19), A3-b3, normals greatly predominated. In another (June 18-19), A4-b4, all were abnormal after 20 drops, but normal after 40 and after 2 drops sperm suspension. In another lot (June 17-18), Ar'-br' and Br'-ar>, there were two cases where there were more abnormals than normals. These ap- parently contradictory results can only be explained on the assumption that the dishes were in some way responsible for the abnormal developments.
Delay in Fertilization and Abnormal Development
Whether or not a delay in fertilization causes abnormality after eggs and sperm have been kept in sea water was tested. The results are recorded in Table IV. The eggs of three individuals were distributed amongst six dishes. Then the eggs of A were fertilized by 10 drops of the sperm suspension of b and c at once, and after an hour and two hours. Similarly the eggs of B were fertilized by the sperm of a and c ; and for the eggs of C by the sperm of a and b.
In A by b normals were produced by fertilizing at once, and after one hour, but nearly all abnormals after two hours. This is true also for A by c. but B by a and B by c and C by b gave more normals after two hours. C by a gave normals after one hour, but no fertilization after 2 hours. Perhaps the most striking re- sults are A by b and A by c that gave normals at once and after a delay of one hour, but abnormals after two hours, but this did not happen in two other cases.
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA
55
Either the sperm in the two latter cases was less affected by the delay, or there was some kind of difference in the dishes themselves. The irregularities in this experi- ment could only be accounted for on the assumption of infection in some of the dishes. The test was repeated (Oct. 14, 1944) with the eggs of new Cionas at
TABLE III
The effect of density of insemination in reciprocal crosses on the proportions of normal tadpoles, bent embryos, and abnormal embryos.
(June 18-19, 1944)
|
A' by b' |
Tad |
Bent |
Abn. |
BI by a1 |
Tad |
Bent |
Abn. |
|
40 drops |
405 |
0 |
0 |
40 drops |
406 |
6 |
17 |
|
20 drops |
443 |
0 |
2 |
20 drops |
471 |
3 |
13 |
|
2 drops |
3 |
40 |
328 |
2 drops |
364 |
8 |
4 |
(June 19-20, 1944)
|
A" by b= |
Tad |
Bent |
Abn. |
B2 by a- |
Tad |
Bent |
Abn. |
|
40 drops |
122 |
21 |
39 |
40 drops |
0 |
0 |
199 |
|
20 drops |
0 |
0 |
295 |
20 drops |
149 |
12 |
7 |
|
2 drops |
351 |
0 |
0 |
2 drops |
91 |
0 |
0 |
(June 18-19, 1944)
|
A3 by b3 |
Tad |
Bent |
Abn. |
B3 by a3 |
Tad |
Bent |
Abn. |
|
40 drops |
211 |
5 |
0 |
40 drops |
205 |
10 |
21 |
|
20 drops |
268 |
7 |
17 |
20 drops |
101 |
23 |
26 |
|
2 drops |
148 |
14 |
7 |
2 drops |
231 |
1 |
3 |
(June 18-19, 1944)
|
A" by b1 |
Tad |
Bent |
Abn. |
B< by a4 |
Tad |
Bent |
Abn. |
|
40 drops |
230 |
0 |
0 |
40 drops |
225 |
24 |
37 |
|
20 drops |
0 |
0 |
263 |
20 drops |
330 |
0 |
1 |
|
2 drops |
266 |
0 |
0 |
2 drops |
218 |
19 |
38 |
(June 17-18, 1944)
|
A* by b5 |
Tad |
Bent |
Abn. |
BS by a5 |
Tad |
Bent |
Abn. |
|
20 drops |
377 |
6 |
7 |
20 drops |
607 |
13 |
31 |
|
10 drops |
293 |
3 |
1 |
10 drops |
684 |
2 |
12 |
|
2 drops |
45 |
43 |
116 |
2 drops |
191 |
35 |
278 |
Corona Del Mar, and transferred to Pasadena in autoclaved flasks and slenders. Five of the lots fertilized at once, A by b, gave 99.9 per cent normals ; one (B by c) gave 75 per cent normals. After 2 hours A by b gave 99.9 ; A by c, one normal and the rest unfertilized; A by c gave 0.5 per cent normal and the rest unfertilized.
56
T. H. MORGAN
TABLE IV
The effect of delaying insemination in reciprocal crosses on the proportions of normal, bent, and abnormal embryos, and unfertilized eggs.
|
i |
|||||||||||
|
N |
Bent |
Abn. |
Unfert. |
N |
Bent |
Abn. |
Unfert |
||||
|
A by b |
at once |
504 |
0 |
7 |
6 |
B bv a |
at once |
66 |
13 |
232 |
93 |
|
1 hr. |
343 |
0 |
1 |
0 |
1 hr. |
131 |
2 |
84 |
98 |
||
|
2 hrs. |
3 |
5 |
390 |
25 |
2 hrs. |
204 |
0 |
0 |
179 |
||
|
A by c |
at once |
428 |
0 |
20 |
27 |
C bv a |
at once |
317 |
7 |
48 |
11 |
|
1 hr. |
464 |
0 |
9 |
27 |
1 hr. |
304 |
1 |
11 |
12 |
||
|
2 hrs. |
14 |
30 |
364 |
4 |
2 hrs. |
0 |
0 |
0 |
498 |
||
|
B bye |
at once |
763 |
8 |
39 |
20 |
C bv b |
at once |
721 |
0 |
15 |
5 |
|
1 hr. |
483 |
14 |
63 |
4 |
1 hr. |
870 |
0 |
8 |
6 |
||
|
2 hrs. |
888 |
2 |
10 |
7 |
2 hrs. |
910 |
1 |
1 |
157 |
After 3 hours A by b gave 95 per cent normals and 5 per cent unfertilized ; A by c gave 99 per cent normals and 1 per cent unfertilized ; B by c were all unfertilized. Reciprocally B by a at once was, by oversight, not fertilized ; but after 2 hours 99.8 per cent, and after 3 hours 100 per cent were normals. C by a at once gave 99.9 per cent normals ; after 2 hours 99.9 per cent normals ; and after 3 hours 100 per cent normals. C by b at once gave 100 per cent normals; after two hours 99.9; and after three hours none were fertilized.
Another test of the same sort gave much the same kind of result. The dishes had been boiled. The percentages of normals was not so high, but there was no evidence of more abnormals after one or 2 hours.
These tests make it quite clear that delay in fertilization up to 3 hours does not in itself cause abnormal development. The irregularities shown in Table IV may safely be ascribed to infection of the dishes. In fact other tests have shown that eggs may be kept in sea water for 24 hours, and, if fertilized with fresh sperm produce normal tadpoles.
Eggs Fertilized En Masse and Distributed at Tico-Cell Stage to Separate Small Dishes
The exceptional cases, in which most or all of the eggs developed abnormally whereas others treated in the same way developed normally, suggested that some environmental conditions were responsible for the abnormals. The following ex- periments were made to further test this suggestion. Eggs were collected from the oviduct in large clusters, and then drawn up by a pipette and transferred to finger bowls with about 80 cc. sea water. They were cross-fertilized with 10 to 20 drops of sperm suspension. After about one hour, when the two-cell stage was reached, about 100 were picked out with a pipette and put into several small dishes or flasks where they remained for about 24 hours, and were then classified. The eggs in the two-cell stage should be nearly alike on the whole, and any difference observed should be environmental. As shown in Table V,_ there were found some dishes that gave a high percentage of abnormals. It would seem to follow that the differences are due to the environment rather than to internal factors. It is
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA
57
especially to be noted that the eggs left in the finger bowls with more water gen- erally gave nearly 100 per cent normals. The larger volume of water would be expected to dilute any contamination present in the larger dish or introduced with the eggs.
Again, a few eggs (about 100) in the two-cell stage were picked up with a pipette (2 or 3 drops) and each lot put into one of five small flasks, and into five jars with 10 cc. sea water. The flasks were stoppered and the jars had screw tops put on them to prevent evaporation. After twenty hours a few drops of formalin were added to each, and the condition of the embryos examined. Simi- larly the eggs of individual B were crossed by sperm of a. No counts were made
TABLE V
The effect of different glassware on the proportion of normal, bent, and abnormal embryos. Each of the two reciprocal crosses was made in a single finger bowl, and eggs in the two cell stage were pipetted into separate small dishes or flasks, in which they remained for about 24 hours.
|
A by b |
N |
Bent |
Ab |
B by a |
N |
Bent |
Ab |
|
1 |
101 |
0 |
3 |
1 |
0 |
0 |
271 |
|
2 |
103 |
0 |
0 |
2 |
150 |
0 |
0 |
|
3 |
88 |
2 |
6 |
3 |
0 |
0 |
176 |
|
4 |
164 |
2 |
2 |
4 |
125 |
1 |
1 |
|
short |
|||||||
|
5 |
29 |
62 |
4 |
5 |
12 |
56 |
93 |
|
bent |
|||||||
|
6 |
15 |
3 |
123 |
6 |
291 |
2 |
5 |
|
7 |
123 |
7 |
5 |
7 |
75 |
12 |
0 |
|
8 |
0 |
0 |
165 |
8 |
197 |
6 |
4 |
|
short |
|||||||
|
9 |
100 |
1 |
6 |
9 |
4 |
121 |
0 |
|
short |
bent |
||||||
|
10 |
148 |
0 |
4 |
10 |
8 |
68 |
14 |
of the 20 lots since practically all dishes contained either 99 or 100 per cent normal tadpoles, except one flask that had 38 tadpoles, 48 bent tadpoles, and 21 abnormal embryos. There was no evident differences between this dish and the others.
Another experiment, similar to the last, with fresh Cionas was made (Aug. 12, 1944). Five Syracuse dishes (12 cc.) were used for some eggs, 5 Stender dishes for other eggs, and 5 small flasks for others. The dishes were left at Corona del Mar, and formalin added to each after 20 hours. Reciprocal crosses B by a were made. The fertilized eggs, not used, were kept in finger bowls (covered) in about 80 cc. sea water. A by b in the finger bowls gave 95 per cent normals and 5 per cent abnormals. B by a gave 99 per cent normals. The records of the smaller dishes are given in Table VI. It is noticeable that while normals greatly pre- dominated, especially in B by a, there are three striking exceptions where abnormals predominate, and other where a good many abnormals were present. Since the controls — the original larger finger bowls — gave 95 and 98 per cent normals, ih.e exceptional cases in the smaller dishes must have been due to contamination of some kind.
58
T. H. MORGAN
TABLE VI
The effect of different glassware on the proportion of normal, bent, and abnormal embryos resulting from two reciprocal crosses.
|
A by b |
B by a |
||||||
|
N |
Bent |
Ab |
X |
Bent |
Abn |
||
|
1 |
0 |
20 |
213 |
61 |
30 |
93 |
|
|
2 |
121 |
9 |
12 |
115 |
22 |
10 |
|
|
Syracuse |
3 |
112 |
5 |
19 |
52 |
16 |
84 |
|
4 |
100 |
24 |
6 |
157 |
0 |
7 |
|
|
5 |
87 |
2 |
7 |
107 |
15 |
12 |
|
|
1 |
51 |
15 |
44 |
56 |
3 |
5 |
|
|
2 |
80 |
26 |
24 |
188 |
0 |
1 |
|
|
Stender |
3 |
1 |
29 |
152 |
132 |
2 |
4 |
|
4 |
81 |
2 |
15 |
134 |
1 |
3 |
|
|
5 |
142 |
8 |
25 |
128 |
6 |
24 |
|
|
1 |
58 |
7 |
11 |
35 |
0 |
1 |
|
|
2 |
89 |
2 |
14 |
77 |
— |
6 |
|
|
Flask |
3 |
193 |
1 |
12 |
53 |
? |
5 |
|
4 |
72 |
3 |
17 |
86 |
1 |
0 |
|
|
5 |
71 |
11 |
29 |
123 |
1 |
4 |
Another set of 20 tests was made with Syracuse dishes. The Cionas had been collected the day before and brought to Pasadena. They were in excellent condi- tion with many stored eggs and much sperm. The eggs were inseminated with 30 drops of sperm in ringer bowls of 80 cc. sea water. At the two-cell stage about 100 eggs were transferred to Syracuse dishes. Most of the dishes contained a high percentage of normals, but there were six dishes that contained nearly all ab- normals, and one that contained a large excess of bent and short embryos.
In the same lot, there were three dishes of A by b left uncovered that gave 12 bent and 111 abnormal embryos ; one half covered that gave 3 normals, and 1 5 bent ; and one 7/8 covered that gave 153 normals and 1 abnormal. The three similar reciprocals, B by c, gave nearly all normals. Since evaporation must have been greater in those not covered, and partly covered dishes, than in the covered dishes, and since some of these gave normal, it does not seem probable that enough evapora- tion takes place in covered Syracuse dishes to cause abnormal development (see below). The large number of eggs kept in the finger bowl gave nearly all normals.
Another experiment was made in the same way (Aug. 21, 1944). Cross- fertilized in ringer bowls, the eggs in the two-cell stage were distributed in eight Stender dishes. Five lots gave nearly all normals, but one gave all abnormals, and two gave more abnormals than normals. In the large number of eggs left over in the two finger bowls, one gave 100 per cent normals and the other 99.9 per cent normals. Obviously the conditions in three of the Stender dishes were unfavorable.
Effect of Evaporation oj iJie Sea ll'atcr on Development
One of the possible influences that cause abnormal development might be evaporation of the sea water from the covered Syracuse dishes. To test this, 10
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA 59
cc. of sea water was evaporated in the dishes to a slight extent before the eggs, that had been fertilized in a large volume of water, were transferred to the concen- trated sea water (Oct. 16, 1944). As a control, two autoclaved dishes that had been covered, but their volumes reduced only to 9%,, and 9%,, cc. gave 100 per cent normals. Another dish, where the water had evaporated to 7%o cc-- gave all twisted and bent embryos. In another, evaporated to 7 cc., the eggs had remained in 2 cells after 24 hours. In general the water in a covered Syracuse dish is reduced during 24 hours by about %0 to 5/w cc. As shown by these and other tests this amount of reduction does not interfere with normal development. Hence, the irregularities observed in the earlier experiments were not due to concentration of the sea water.
Sterilization of Dishes by HCl
Several tests in duplicate were made. In one (A by b, Sept. 3, 1944) the dishes had been sterilized with weak HCl and washed in distilled water and dried. There were eight that gave 95 to 100 normals, but two gave 100 per cent abnormals. Reciprocally, (B by a) there were six that gave normals (90 to 100 per cent) and four that gave mostly abnormals. The high frequency of dishes with abnormals may seem to be due to the acid, but if any was left after washing, it should have evaporated when the dishes were dried (24 hours). Many eggs were left in the finger bowls, and A by b and B by a each gave 100 per cent normals. Whatever the cause of occasional abnormal development, it appears that it was not removed from all the dishes by the HCl treatment. Either the HCl itself was not entirely removed by washing, or else something else was left behind after the treatment that was later the basis of contamination (infection) of the dishes. The result is in accord with earlier experiments of the same sort (Blol. Bull., 80, 1941, pages 348-349).
Sterilisation of Dishes by Heat
The records of the preceding experiments show almost without exception, one or more similarly treated dishes with abnormal embryos, while the rest contain only, or largely, normals. Such exceptions occur even when the eggs have been transferred to the smaller dishes at the two- or four-cell stage. The result suggests contamination or infection of some lots. Therefore, a new set of 10 duplicated lots and 10 reciprocals were tested in dishes that had been autoclaved (120° C.). The eggs were collected from the oviduct in such a dish and transferred to a large flask (also autoclaved) and there cross-fertilized. At the two-cell stage, 50 to 100 eggs were transferred to five flasks and five Stenders each, 20 in all. After 24 hours, A by b gave practically all normals, i.e. 85 to 100 per cent. The reciprocals all gave 100 per cent, except one lot that had 50 per cent somewrhat bent tadpoles and 50 per cent normals. The eggs left over in the large flasks (A by b) gave 99 per cent normals and one per cent bent, about two thousand in all. In the reciprocal, there were about one thousand normal tadpoles.
Another experiment was made the next day with Cionas brought to Pasadena ; 20 dishes were prepared as above (autoclaved). The eggs were kept in Syracuse dishes. They gave 99.9 per cent normals. The many eggs (about two thousand) left in a large beaker (sterilized) also gave 99.9 per cent normals.
60 T. H. MORGAN
A third experiment of the same kind in which the dishes had been boiled for a short time instead of autoclaved, gave nearly the same results. Most of the dishes, A by b, (Syracuse) gave from 95 to 100 per cent normals; only one had 75 per cent normals and 25 per cent abnormals. The reciprocals (B by a) gave five 90 to 98 per cent normals; four. 80 per cent, and one, 75 per cent normals. One (A by 1)) of the large finger bowls (80 cc.) gave 99.5 per cent normals and 0.5 per cent bent ; the other gave only 5 normals, 1 bent and 94 abnormals. It is evident that this finger bowl (B by a) was contaminated, but the eggs that were removed from it at the two-cell stage and transferred to the ten smaller autoclaved dishes with 10 cc. fresh sea water gave 80 to 95 per cent normals.
These sixty tests are very convincing that the cause of abnormal development is due to some contaminating agent present in the dishes, or that develops there (bacterial action). Washing the dishes in tap water and even rinsing them in distilled water does not remove the source of the contamination, while autoclaving the dishes is effective. It should be recalled that in most of the earlier experiments the embryos after 24 hours were killed in weak formalin and the dishes allowed to stand for several hours or days, before washing them again in tap water. They were then dried for a day or longer. Whether the formalin combined with some organic matter in the dishes, or whether the organic matter alone is responsible for the contamination is. not shown by these experiments.
Another test was made (September 30-October 1, 1944) with fresh Ciona at Corona del Mar. The eggs were brought to Pasadena in autoclaved flasks and Stenders. Eggs had been fertilized in large finger bowls (80 cc.), and transferred in an early cleavage stage to 5 flasks and 5 stenders, 20 in all. Only normal tadpoles developed. Again, a similar experiment was carried out at Pasadena (Oct. 1-2), but the eggs were transferred to 20 Syracuse dishes that had not been autoclaved since the last time they were used. They gave 99 per cent normals in all, but 5 dishes (out of 20) gave 95 per cent. These two tests corroborated the conclusion that in clean dishes nearly 100 per cent are normal tadpoles. The very small percentage of abnormals may be due to polyspermy. or to other defects in the cleavage.
A different test was carried out as follows (Oct. 1, 1944). A set of 20 dishes was made up as above, using Syracuse dishes that had been washed, but not auto- claved. There were 14 that gave 95 per cent to 100 per cent normals; 4 that gave only abnormals and one that gave 80 per cent normals. The 4 that gave abnormals were carried one step further. The water was drawn off and put into an auto- claved Stender. To the original Syracuse dishes, 10 cc. of fresh sea water was added. Eggs of another individual, in the two cell stage were added to each. After 22 hours all of the Stenders had only abnormal embryos. Evidently the water had been fouled in some way. The six original dishes with fresh water gave 95 per cent normals. 5 per cent abnormals; 50 per cent normals; 100 per cent ab- normals; 100 per cent normals; 95 per cent normals; and 80 per cent normals. These eggs did much better in the old dishes with fresh sea water than in the old sea wrater in a fresh dish.
As a control test, six of the original (Syracuse) dishes that had given only normals were treated in the same way. The six original dishes with fresh sea water (10 cc.) gave all normals. The old sea water from 4 dishes transferred
NORMAL OR ABNORMAL DEVELOPMENT OF CIONA 61
to Stenders gave in one dish normals, another gave abnormals, and another 50 normals and 50 abnormals. More normals resulted than when the original dishes that gave only abnormals were tested. These results are consistent with tile- assumption that abnormal development is due to toxic material that develops in some of the dishes. It does not seem to be due to the eggs themselves, but to some foreign material present in some of the dishes. Ordinary washing does not remove the material, but sterilization of the dishes does remove it. If the toxic material is due to bacteria left in the dishes, the protein or other substance on which the bacteria grow may be due either to some substance left in the dishes, or to organic material that goes over with the eggs or the sperm.
SUMMARY
Eggs of Ciona that develop in covered Syracuse dishes produce as a rule normal tadpoles in the course of 24 hours, but occasionally only abnormal embryos develop, or both normal and abnormal in the same dish.
Reciprocal crosses sometimes give very different proportions of normals and abnormals, but the relations are often very inconsistent. The results cannot be ascribed to the eggs alone or to the sperm or to their combinations. Even if an attempt is made to refer the outcome of the reciprocal crosses to the cytoplasm of the egg. that has developed under the influence of the diploid nucleus of the egg, still the differences cannot be satisfactorily accounted for.
Polyspermy can account for only a very small percentage of abnormal develop- ment, probably not more than half of one per cent.
Delay in fertilizing the eggs does not cause abnormal development. Delay in using the sperm suspension does not cause abnormals.
When the eggs of one individual are fertilized by the same amount of sperm suspension of six other individuals, the outcome may be normals in each dish or normals and abnormals, or occasionally, all are abnormals. When the sperm of one individual is used to fertilize the eggs of six others, the same kind of results may happen.
In order to make the samples of eggs as nearly alike as possible, they were first fertilized in a large volume of water, and. when in the two-cell stage, a few were transferred to several small Syracuse dishes. The same kind of irregularities appeared, which cannot be due to chance selection of different eggs, for the original eggs left in the large amount of water generally gave more than 95 per cent normals.
Therefore, several kinds of experiments were made to find out whether differences in the small dishes will account for the occasional, but persistent, appearance of abnormals. ( 1 ) Dirt or impurities in the sea water from the tap was excluded. (2) Evaporation of the sea water in the covered Syracuse dishes was not enough to cause abnormal development. (3) Ordinary washing of the dishes, even rinsing them in distilled water did not remove the cause of abnormal development in some of the dishes. (4) But sterilizing the dishes in an autoclave removed the cause of most of the irregularities.
It follows that the cause of the exceptional cases of abnormals is due to some contamination that remains in the dishes after washing them in fresh tap water. Since the contamination does not affect the early development of eggs (cleavage),
62 T. H. MORGAN
but only later stages, it must be putrefactive in origin ; and since it is removed by autoclaving the dishes, it is probably bacterial or some sort of organic contamination.
The eggs come to rest on the bottom of the dishes in a few minutes where they remain until the tadpole emerges, hence local differences may affect the development and account for those cases where both normal and abnormal development takes place in the same dish.
The discovery of the cause of the occasional abnormal development removes the possibility that it is due to genetic factors, hence is not concerned with the self- sterility of Ciona.
REGULAR OCCURRENCE OF HETEROPLOIDY IN A GROUP OF
PENTATOMIDAE (HEMIPTERA)
FRANZ SCHRADER
Department of Zoology, Coluinlna University, Nciv York
The testis of all pentatomid Hemiptera is composed of lobes which are constant in number for any given species. In some species the size of the spermatocyte cells and sperms may vary in the different lobes, but in any one species there is a char- acteristic size for each lobe. In other words, if the first lobe in one specimen pro- duces unusually large spermatocytes, the corresponding lobe in other specimens of that species will also show such large spermatocytes. This peculiarity was first discovered by Montgomery (1898), who described it in some detail in a later paper (1910). The phenomenon was called "polymegaly" by Bowen (1920, 1922a, 1922b) who in his exact analysis laid special stress on the processes of sperm formation. He found that certain cytoplasmic elements varied in amount propor- tionally to the size of the cells, but that the size of the chromosomes varied little or not at all in the different lobes. Normal sperms were formed from all lobes, and differences in the shape of the head arose only because there was greater elongation in some lobes.
The conditions that are basic to this constant variation are therefore subject to thoroughly regulated processes in any given species. They involve no mitotic irregularities and the occurrence of nondis junction, multipolar spindles and failure of cytoplasmic division, which of course are found here as in other animals and plants, always seem to result from physiological accidents that are sporadic and in no way correlated with polymegaly.
It is therefore of some interest to record a series of cases among the Pentato- midae in which mitotic disturbances occur regularly in a certain lobe of the testis and affect all the cells of that lobe. Moreover such irregularity, although resulting in heteroploidy of an extremely large range, does not affect the basic processes of sperm formation. Hence spermatozoa are produced which vary greatly in volume, but which appear to be otherwise normal in general structure.
MATERIAL AND METHODS
The pentatomids concerned are Loxa flavicollis Drury, of which nine speci- mens were collected in Panama (1940 and 1941) and Costa Rica (1944) ; Lo.va picticornis Horvath, represented by six specimens from Panama (1940) and Costa Rica (1944) ; and Lo.va florida Van Duz, collected in Florida, U. S. A., in 1909. The data on Loxa florida are contained in Bowen's papers (1922a and b) although the main points here at issue are not specifically noted by him. A fourth species, Mayrinia variegata Dist., is represented in my material by only a single male col- lected in Costa Rica in 1944. Though strikingly smaller than the uniformly large species of Loxa, it is nevertheless a close relative and was formerly classified in
63
64 FRANZ SCHRADER
that genus. Its affinity to the species of Loxa is now further attested hy the fact that it shares with them the striking mitotic features that form the suhject of this paper.
Both testes and ovaries were fixed in either Sanfelice or Allen's Bouin and stained with haematoxylin, gentian violet, or the Feulgen method. Although Bowen did not state it, it is more than likely that his material of Lo.ra florida for which he thanks Professor E. B. Wilson, was fixed in strong Flemming solution. In all of my own material the gonads were usually fixed within a short interval of capture, usually less than three hours and sometimes within a few minutes.
My thanks are due to a number of people who have aided me in collecting my material. In this respect I am especially obligated to Dr. T. J. Grant of the U. S. Department of Agriculture in Turrialba, Costa Rica.
STRUCTURE OF THE TESTIS
In all three species of Loxa, the testis is composed of seven lobes, which are arranged side by side in a single series. Following Bowen (1922b) these lobes are designated by consecutive numbers, the first being closest to the side toward which the vas deferens opens (Fig. 1). The lobes show a constant variation in diameter, but only one lobe is strikingly larger than the others. This is the fifth, which may have a diameter two or three times that of any other and which at the same time is characterized by spermatogonial and early spermatocyte cells that are dis- tinctly smaller than the cells of other lobes. It is this lobe which regularly shows heteroploidy in the spermatocytes.
It is this lobe also which shows heteroploidy in Mayrinia. There however, the fifth is not the largest lobe, for it is exceeded in diameter by the sixth. In neither of the two Mayrinia testes at my disposal is a seventh lobe visible, and it may be that in this species the last two lobes have fused to form a single, larger one. Since there are some indications that my specimen is an old male, it is inadvisable to draw any general conclusions on this point. The main point however is fully established. In all four species it is the fifth lobe of the testis which always shows the special development here under discussion.
CYTOLOGY
In all three species of Loxa, the spermatocyte cells are smaller in the fifth than in the other lobes. But as Bowen ( 19221) ) has pointed out in reference to polymegaly in general, this size difference rests in the volume of the cytoplasm rather than that of the chromosomes. In Mayrinia, such polymegaly is not as striking as in Loxa.
The spermatogonial divisions appear to be very much alike in all of the lobes (Fig. 2A and B). The fifth lobe shows the diploid set of 14 chromosomes differing in no perceptible way from that of the remaining lobes, and mitoses are orthodox. The unusual developments do not become evident until the prophase of the first spermatocyte. Since a detailed cytological analysis will be published elsewhere, only a brief outline will be given here.
The most obvious deviation from the normal course first occurs some time before diakinesis. When the cells of a given cyst reach this stage they begin to
REGULAR HETEROPLOIDY
65
f/
/rrf
ip.aonio.
FIGURE 1. Section of the testis of Loxa flavicollis, showing the positions and comparative sizes of the seven lobes. Heteroploidy occurs regularly in the fifth lobe. Actual length of testis is 4 mm. and contents are only partially shown.
fuse with eacli other. The process is irregular in the sense that cellular aggregates resulting from the union of two or very few cells may lie side by side with much larger aggregates comprising many more cells. The nuclei at this time tend to become amoeboid and may divide by constriction, but if fusion occurs among them, it is not general.
66
FRANZ SCHRADER
Later during the prometaphase, cellular boundaries become very vague and the whole cyst may come to simulate a single, giant cell. It is at this time also that the nuclear walls disintegrate, as they do in the normal course of events. The chromosomes, condensing rapidly, then seem to lie scattered irregularly through the cytoplasm. However, immediately prior to the first appearance of spindles, the cytoplasm is segregated into discrete masses which do not seem
FIGURE 2. All from Lo.va fiavicollis (approx. 1.500X).
A. Spermatogonial metaphase from normal lobe ( 14 chromosomes).
B. Spermatogonial metaphase from fifth lobe (14 chromosomes).
C. First spermatocyte metaphase from normal lobe (6 autosomal tetrads + X + Y).
D. First spermatocyte metaphases from fifth lobe (sho\vin,u 3. 12. and upwards of 60 chromosomal bodies).
to correspond exactly to the original fusion aggregates. It is probable that this new step involves the activity of both chromosomes and centrioles in the formation of spindles, and the latter appear very soon thereafter. Simultaneously the chromosomes are arranged into numerous metaphase plates, the number of which corresponds to the rounded masses of cytoplasm just formed. Since the latter seem to have no reference to the original prophase spermatocyte cells, the different metaphases may be composed of as few as two (perhaps even one) and
REGULAR HETEROPLOIDY
67
as many as two hundred or more chromosomes (Fig. 2D}. By the same token, the types of chromosomes included in such metaphase plates vary greatly. All in all, these configurations of chromosomes are the result of rather haphazard processes and it is probably only very rarely that they happen to represent the orthodox set of chromosomes (Fig. 2C] in the first spermatocyte (i.e. six autosomal tetrads -|- X -f- Y), especially since it seems doubtful whether normal pairing occurs in the fifth lobe.
Although the number of multipolar spindles is distinctly higher than in normal lobes, it is not as great as one might expect from the seemingly confused conditions that prevail just before spindle formation. Unlike the chromosomal movements in the preceding period, the maneuvers of the centrioles are evidently not completely
B
FIGURE 3. All from Lo.va flai'icollis (approx. 1,500 X).
A. Chromosomes of telophase of second spermatocyte in normal lobe (6 autosomes + Y).
B. Chromosomes of telophase of second spermatocyte in normal lobe (6 autosomes + X).
C. Chromosomes of telophase of second spermatocyte in fifth lobe (showing 2, 4, 5, and up- wards of 40 chromosomes at each pole) .
at random and their final positions result in bipolar figures in the great majority of cases.
Despite the bewildering range in the composition of these metaphase plates, the actual division of the abnormal first spermatocytes occurs normally. The second spermatocytes therefore carry a correspondingly great variation in the numbers of chromosomes and they too divide successfully to give spermatid cells (Fig. 3).
The nuclei that are formed in the spermatids after the second spermatocyte division of course reflect the variation in the chromosome numbers that are encountered in the meiotic divisions. Hence they may greatly exceed in size the normal spermatid nucleus, and such large nuclei may lie side by side with the tiny nuclei derived from only one or two chromosomes (Fig. 4). However
68
FRANZ SCHRADER
except for this deviation from the normal size, the processes of sperm formation seem to parallel those observed in the normal lobes. The behavior of chondriosomes and (iolgi material seems to lie regular, and the transformations of the nucleus that culminate in the tenuous head of the ripe sperm are undergone just as in normal cells. The number of degenerating sperms is not great and certainly a huge number reach the final stages near the entrance to the vas deferens. Since cells with a normal complement of chromosomes must be very rare indeed, there is no escaping the fact that very many sperms carrying irregular combinations and numbers of chromosomes reach the final stages of development. Whether they then enter the egg and become functional has not been determined.
FH.TKE 4. All from Lo.ra flaricnllis (approx. 1,500 X).
A. Sperms (only heads shown complete) at about Stage n. from normal lobe.
B. Sperms (only heads shown complete) at about Stage n. from fifth lobe, showing great variation in size.
The evidence that Lo.va florida duplicates the conditions here described, is indirect but conclusive. In one of his papers, Bowen (1922a) discusses the occur- rence of abnormally high numbers of chromosomes in that species, and ascribes them to a fusion of spermatocyte cells. He mentions that this takes place in the same lobe of both the Loxa testes available to him. Although not stating which lobe this is, Bowen characterizes it as carrying unusually small spermatogonial and early spermatocyte cells. Tn a somewhat later paper of that year (1922b) he mentions that in Lo.va jlorida it is the fifth lobe that has cells which are distinctly smaller than those of the other lobes. There is thus no reason to doubt that he was dealing with conditions in Lo.va jlorida that are essentially the same as those here described and which there occur also in the fifth lobe of the testis.
REGULAR HETEROPLOIDY 69
EVOLUTIONARY ASPECTS
These then are the astonishing conditions that confront us in the two genera, Loxa and Mayrinia. In four different species, taxonomically distinct, and col- lected from Panama, Costa Rica, and Florida — a latitudinal range of at least 1 ,700 km. (or about 1,050 miles) — one of the lobes of every testis undergoes a very special type of abnormal development. This abnormality results in a great number of unusual distributions of the chromosomes, but is in its basic nature identical in all four species. For in all of them it is always the fifth lobe of the testis that is affected ; in all of them the spermatogonial divisions of this lobe seem normal and cytological indications of abnormality do not appear until the preparatory phases of the meiotic period ; and in all four species the odd assortments of chromosomes ' that result partake nevertheless in a formation of sperms which are normal in appearance except for size.
Whatever the factors may be that bring about this constant and regularly occurring abnormality, its evolutionary aspects pose some interesting questions. It is reasonable to assume that the conditions underlying it are of long standing in the history of these* species and that in fact they were probably present in the parent form from which they arose. Since all of my sixteen specimens carry the orthodox complement of 14 chromosomes, the indications are that the great ma- jority of the sperms with abnormal chromosomes numbers are either nonfunctional or produce nonviable zygotes. The existence of so wasteful a development would seem to be an incongruity which might well be expected to meet with short shrift in the processes of natural selection. Nevertheless it has persisted, either because it involves some benefit to the species that is not immediately obvious or else be- cause the elaboration of useless sperms, even in such huge numbers, entails only an imperceptible drain on the evolutionary welfare of the animal since it is amply compensated by the production of sperms from the normal lobes of the testis.
Be that as it may, the constant production of gametes with irregular chromo- some numbers should greatly enhance the chances for producing zygotes with other than the usual complement of 14 chromosomes. Granted that most of the abnormal sperms will not produce viable zygotes, it is likely that some of the many unusual combinations do at times develop. Certainly the flow of new and heteroploid forms from such a species should be very much greater than in the case of species in which the opportunities for such varietal changes are restricted to the compara- tively rare mitotic accidents that occur in the more orthodox processes of germ cell formation. It may be that a basic conservation that may exist among the Penta- tomidae is not favorable to such evolutionary adventures, but the number of speci- mens at my disposal is surely too small to allow of any final conclusions on this point. Certain it is that a further examination of the 33 so far described species that belong to the three, taxonomically closely knit genera Loxa, Mayrinia, and Chlorocoris may well throw light on these questions (all except Loxa florid a are exclusively neotropical, but their range extends from Florida to Argentina ; see Van Duzee. 1909 and Horvath, 1925).
But if. as already suggested, the conservatism of these Hemiptera does not allow a utilization of such evolutionary opportunities, the case nevertheless presents a mechanism which in some other groups might well be conducive to rapid and great strides in the elaboration of new species. The ready viability of forms with unusual
70 FRANZ SCHRADER
chromosome combinations such as characterizes for instance genera like Datura and Crepis would suggest that other forms might well take advantage of such a range of possibilities. At any rate, the case indicates in what unexpected ways a group of organisms may suddenly become very active in the evolutionary sense.
DIFFERENTIATION OF SPERMS
The case reflects in a rather striking, albeit not entirely novel way, on the problem of differentiation. It is clear that the complex processes which bring about the formation of the highly specialized sperm — including the special trans- formations of Golgi and chondriosome material as well as the change of the spherical spermatid nucleus to the extremely tenuous, chromatic sperm head — all occur regardless of whether only one or two or many dozens of chromosomes are present in the cell. It therefore would seem safe to conclude that the immediate control over these developments is vested in cytoplasmic elements and that, if nuclear genes are primarily responsible they must exert their influence on the cytoplasm at least two mitotic cycles earlier (in the spermatocyte prophase when cell fusion first occurs). A demonstration to the same effect is contained in the work of Dobzhansky and Sturtevant (1931) where it is shown that sperms of Drosophila which have lost various sections of the chromosomes through translocation during the meiotic prophase, nevertheless become functional.
SUMMARY
1. The fifth lobe in the testes of four species of Loxa and Mayrinia regularly produces sperms carrying heteroploid complements of chromosomes.
2. This condition must be of long evolutionary standing and may be of sig- nificance in the formation of new forms.
LITERATURE CITED
BO\YEX, R. H., 1920. Studies on insect spermatogenesis I. Biol. Bull.. 39: 316-363.
BO\VEN, R. H., 1922a. Notes on the occurrence of abnormal mitoses in spermatogenesis. Biol.
Bull, 43 : 184-203. BOWEN, R. H., 1922b. Studies on insect spermatogenesis IV. Proc. Am. Ac. Arts. 5V/., 57:
391-423. DOBZHAXSKY, T., AND A. H. STURTEVANT, 1931. Translocation between the second and third
chromosomes of Drosophila and their bearing on Oenothera problems. Carnegie Jnst.
Publ, 421: 25-29. HORVATH, G., 1925. Do pentatomidarum genere Loxa Am. Serv. et de novo genere ei affini.
Ann. Miis. Not. Hunc/arici, 22: 307-328. MONTGOMERY, T. H., 1898. The spermatogenesis in Pentatoma up to the formation of the
spermatid. Zoo/. Jalirb., Anat., 12: 1-88. MONTGOMERY, T. H., 1910. On the dimegalous sperm and chromosomal variation of Euschistus,
•with reference to chromosomal continuity. Arch. Zcllf., 5: 120-146. VAX DUZEE, E. B., 1909. Observations on some Hemiptera taken in Florida in the spring of
1908. Bull. Buffalo Soc. Nat. Sci., 9: 149-230.
ASEXUAL REPRODUCTION IN THE COLONIAL TUNICATE,
BOTRYLLUS SCHLOSSERI (PALLAS) SAVIGNY, WITH
SPECIAL REFERENCE TO THE DEVELOPMENTAL
HISTORY OF INTERSIPHONAL BANDS OF
PIGMENT CELLS
RAY L. WATTERSON 1 Marine Biological Laboratory and University of California, Berkeley
INTRODUCTION
Botryllus schlosscri (Pallas) Savigny - is a colonial tunicate readily obtained in the Eel Pond at Woods Hole, Massachusetts. In a well developed colony the individual blastozooids (or ascidiozooids) are grouped into one or more systems; each system consists of 2 to 23 blastozooids radiating outwards from a central common cloaca with the separate oral siphons distributed at the periphery. The most striking feature of many Botryllus colonies is the localization of special light- reflecting pigment cells between the oral and atrial siphons of each blastozooid. Because of the association of individual blastozooids into systems within each colony the total aggregation of such reflecting cells within any one system forms an attractive star-shaped pattern (PI. 1, Fig. 3) clearly revealing the spatial dis- tribution of the blastozooids involved. Such patterns will be referred to as inter- siphonal patterns of pigment cells (intersiphonal patterns for short), and each arm of the pattern will be called an intersiphonal band of pigment cells (intersiphonal band for short).
Such intersiphonal bands have been described more or less incidentally by various taxonomists; however, only two of these seemed to realize that the bands are actually aggregations of pigment cells in specific regions. Both of these men realized that these intersiphonal bands are not constant in their appearance. Pizon (1899a) merely states that they undergo changes with time without specifying the nature of these changes. Bancroft (1903) states that in young zooids there are no intersiphonal bands at all and that three to four days elapse before they reach their complete formation. Such brief observations constitute the only information avail- able in the literature concerning the developmental history of intersiphonal bands
1 It is a pleasure to acknowledge the excellent assistance of Miss Juanita Senyard during the accumulation of the data upon which this paper is based.
2 Many species and varieties of Botryllus have been described (Giard, 1872; Herdman, 18! Hartmeyer, 1909-1911; Alder and Hancock, 1912). However, these were distinguished largely on the basis of color differences. Pizon (1899a) early recognized the need to revise the classi- fication of Botryllus since he realized that considerable color variation may occur even within the same colony, and Bancroft (1903) even went so far as to state (p. 161), "ni Bntryllits, as it occurs in Europe and the Atlantic Coast of North America, color characters cannot be used for separating species; . . . therefore, since none of the described species hare been based upon morphological characters, there is no valid reason for recognizing more than the single species, B. schlosseri (Pallas, 1766, pp. 355-356) Savigny (1816)." This view is acceptable to Van Name (1910) and to Herdman (1925).
71 1Z
RAY L. WATTERSON
of pigment cells. This lack of information is rather surprising since the hands are so very striking when fully formed ( I'l. 1. Fig. 5) and in view of the fact that so much has heen written concerning other leatures of Botryllus.
One of the most interesting feature's of these intersiphonal hands has received no attention at all. viz., that they are not permanent additions to the pigment pattern of the colony; instead, as soon as they have formed, they are completely destroyed, and this destructive phase marks the most radical change in the appearance of the colony (PI. II. Fig. 8). This destructive phase is inevitahle since, as is well known (Bcrrill, 1941c). the parent xooids degenerate each time a new generation of zooids arises hy asexual reproduction, and these intersiphonal hands are properties of the individual zooids, not of the colony as a whole. The fact that they give to the colony a characteristic intersiphonal pattern or group of patterns is only secondary.
This paper has a twofold purpose: (1) To describe the visible changes in the intersiphonal hands of pigment cells during the establishment of the colony from the larva. By observing the development of a number of colonies it is possible to correlate rather accurately the steps involved in the formation and destruction of the intersiphonal bands with other known steps in asexual reproduction. (2) To describe the variation in the intersiphonal patterns when different colonies are compared, and to consider possible factors involved in such variations.
MATERIALS AND METHODS
The eggs of Botryllus undergo fertilization and development up to a tadpole stage within the blastozooids. Each day some of these tadpoles escape from the cloaca. In order to obtain them adult colonies were collected from the Eel Pond in the morning and in the laboratory they were distributed in ten inch finger bowls filled with sea water. As the larvae escape from the parent zooids they swim toward the light and toward the surface (see Grave and Woodbridge, 1924, for time of liberation of larvae and for reactions of the larvae to light and gravity) and they can be collected easily with a pipette; three of them were placed into a large drop of sea water in the center of a Syracuse watch glass and were left to attach and metamorphose into the oozooid. The watch glasses were stacked to prevent evaporation. Many of these tadpoles metamorphose without attaching (Grave, 1937) or exhibit certain indications of abnormal metamorphosis (Zhinkin. 1939), but these were discarded. Usually at least one tadpole attached in each watch glass; in some instances two or even all three larvae attached, and if they were sufficiently isolated from one another, they were allowed to continue their develop- ment. After attachment had occurred the watch glasses were immersed in large aquaria through which sea water was circulating. The watch glasses were in- verted so debris would not settle out too thickly and obscure the development of the colonies, and they were held in this position by wooden racks. Each day these watch glasses were removed from the aquaria and the appearance of the colonies developing within them was sketched under the dissecting microscope with re- llected light for illumination, since by such illumination only the distribution of the '•pedal light reflecting pigment cells is revealed and all other pigmentation of the zooids is automatically eliminated from observation (see Plates I and II). The exact time at which each sketch was made was recorded. Whenever a colonv was
INTERSIPHONAL BANDS OF PIGMENT CELLS
73
A.S.
AMP
A P
O.S.
A.S. OS
AMP
AMP
A-
AMP
A.S.
O.S.
AMP
F
OS.
A.S.
G.
AMP
OS.
AS
A.S.
H.
FIGURE 1. The development of intersiphonal bands of pigment -cells in the oozooid. Each dot represents one pigment cell. a. Botryllus tadpole 20 to 30 minutes after escape from parent zooid. b. Tadpole in which the tail is undergoing absorption; one hour after escape from parent zooid. c. Oozooid eight hours after escape from parent zooid. d. Another oozooid 11 hours after escape from parent zooid. e. An optical section of an oozooid eight hours after escape from parent zooid. f. Oozooid No. 8c ; age two days; example of a stnma intersiphonal band of pigment cells, g. Oozooid No. 10; age two days; example of a medium intersiplumal band. h. Oozooid No. ISc ; age two days : example of a i^'cak intersiphonal band. Abbrevia- tions : a.p. = adhesive papillae; amp. = ampullae ; a.s. = atrial siphons; o.s. = oral siphons; 1Z = first generation blastozooids.
74 RAY L. YYATTHRSON
in a crucial stage of transformation it could IK- watched for considerable periods of time without injury. However, because of the large number of colonies studied, continuous observations on any one colony for long periods of time were not possi- ble. although such a procedure would have yielded useful supplementary informa- tion. All sketches made for each colony were then mounted from left to right in a horizontal row on large sheets of paper in the order in which the sketches were made, and they were so arranged on these sheets that sketches of all colonies made on any one day lie in vertical columns. Consequently the changes in the appear- ance of any one colony throughout the period studied can be ascertained quickly by running the eye horizontally across the sheet ; and similarly the appearance of all the colonies on a given day is readily compared by running the eye vertically down the sheet. Sixty-one colonies were started from isolated larvae on July 31, 1942. These were first sketched on August 2, next on August 4, and then at daily in- tervals until August 19 if they still survived. Due to the increased complexity of the colonies by this time it was impossible to continue daily observations on all of them. Ten colonies were therefore selected and were sketched at daily intervals until September 4. Twenty-nine other colonies were allowed to develop until August 29 when sketches were made of them ; however, for these colonies the daily changes between August 19 and August 29 are unknown. In the case of the re- maining 22 colonies no sketches were made after August 19 for various reasons. Consequently, since only incomplete records are available for these 22 colonies, they will not be included in the general observations given below. In order to present these observations the daily changes in the appearance of one colony (No. 38) selected as an example will be described in detail (Fig. 2) ; the observations on the other colonies can then be presented briefly (Table I) by merely emphasizing their similarities and differences when compared with this example. All observa- tions have been made exclusively on living material.
OBSERVATIONS Formation and Variation of Intersiphonal Bands in Oozooids
Since the main series of colonies started on July 31 was first examined t\vo days later it was necessary to start other colonies (on August 31 and September 3) in order to study the establishment of the first intersiphonal band during meta- morphosis of the tadpole into the oozooid. These colonies served only for these early observations on oo/.ooid development and were then discarded. Although the tadpole is rather opaque a few reflecting pigment cells can sometimes be seen in a newly emerged larva (Fig. la). By transmitted light these cells appear yel- lowish and by reflected light they are just barely visible. They are quite definitely localized in the dorsal surface and already outline the siphonal areas. Within an hour after the escape of the larva from the parent these pigment cells have darkened
2. The developmental history of intersiphonal hands of pigment cells in colony No. ,^S. Kach dot represents a pigment cell. Sec text for age of colony at stages illustrated and for description of the progressive changes. The various stages are not drawn accurately to scale. Kach degenerating /ooid is indicated by cross-hatching. Abbreviations: c.c. = common cloaca; o.S. : oral siphons; s. "space free from pigment cells; 1Z to 77. = first to seventh generation blastozooids.
INTERSIPHONAL BANDS OF PIGMENT CELLS
75
as
6Z
FIGURE 2
76
RAY L. WATTERSON
W.
"iGl UK 2— Continual
INTERSIPHONAL BANDS OF PIGMENT CELLS
77
TABLE I
Summary of the relationship between changes in the intersiphonal bands of pigment cells and the major changes in asexual reproduction in all colonies studied. Each number indicates the percentage of colonies undergoing the change indicated. The number in parentheses following each percentage indicates the generation of zooids involved (O = oozooid, 1-7 == first 7 genera- tions of blastozooids). The percentages indicated by the asterisk may be too small since parent zooids obscure the buds in advanced stages of colony formation.
|
Age in days |
Zooids degener- ating |
Pigment cells circulating |
Inter- siphonal bands single |
Inter- siphonal bands double |
Buds enlarging |
Oral siphons present |
Atrial siphons present |
Common cloaru< formed |
Non- growing buds present |
|
5 |
59.0(0) |
30.7 |
33.3(1) |
0.0(1) |
59.0(2) |
0.0(2) |
0.0(2) |
0.0(2) |
0.0(3) |
|
6 |
41.0(0) |
33.3 |
61.5(1) |
5.1(1) |
94.9(2) |
0.0(2) |
0.0(2) |
0.0(2) |
17.9(3) |
|
7 |
2.6(0) |
7.7 |
5.1(1) |
56.4(1) |
100.0(2) |
15.4(2) |
0.0(2) |
0.0(2) |
94.9(3) |
|
8 |
0.0(0) |
7.7 |
7.7(1) |
56.4(1) |
100.0(2) |
51.3(2) |
28.2(2) |
0.0(2) |
100.0(3) |
|
9 |
35.9(1) |
35.9 |
17.9(2) |
2.6(2) |
15.4(3) |
94.9(2) |
94.9(2) |
0.0(2) |
0.0(4) |
|
10 |
41.0(1) |
23.0 |
28.2(2) |
25.6(2) |
82.1(3) |
100.0(2) |
100.0(2) |
23.0(2) |
7.7(4) |
|
11 |
7.7(1) |
2.6 |
20.5(2) |
43.6(2) |
97.4(3) |
100.0(2) |
100.0(2) |
66.7(2) |
74.1(4) |
|
12 |
2.6(1) |
12.8 |
7.7(2) |
56.4(2) |
100.0(3) |
23.0(3) |
0.0(3) |
69.2(2) |
97.4(4) |
|
13 |
23.0(2) |
12.8 |
2.6(2) |
51.3(2) |
5.1(4) |
41.0(3) |
5.1(3) |
0.0(3) |
97.4(4) |
|
14 |
48.7(2) |
38.5 |
30.7(3) |
2.6(3) |
48.7(4) |
41.0(3) |
41.0(3) |
7.7(3) |
2.6(5) |
|
15 |
23.0(2) |
7.7 |
17.9(3) |
28.2(3) |
84.6(4) |
84.6(3) |
84.6(3) |
51.3(3) |
20.5(5) |
|
16 |
5.1(2) |
2.6 |
7.7(3) |
48.7(3) |
100.0(4) |
2.6(4) |
0.0(4) |
66.7(3) |
76.9(5) |
|
17 |
5.1(3) |
2.6 |
0.0(3) |
56.4(3) |
100.0(4) |
25.6(4) |
0.0(4) |
0.0(4) |
89.7(5) |
|
18 |
23.0(3) |
7.7 |
5.1(4) |
5.1(4) |
20.5(5) |
56.4(4) |
15.4(4) |
5.1(4) |
2.6(6) |
|
19 |
23.0(3) |
17.9 |
23.0(4) |
17.9(4) |
87.2(5) |
94.9(4) |
76.9(4) |
38.5(4) |
10.3(6) |
|
20 |
0.0(3) |
0.0 |
10.0(4) |
80.0(4) |
100.0(5) |
100.0(4) |
90.0(4) |
90.0(4) |
20.6(6) |
|
21 |
10.0(3} |
10.0 |
10.0(4) |
80.0(4) |
100.0(5) |
100.0(4) |
90.0(4) |
90.0(4) |
80.0(6) |
|
22 |
10.0(4) |
10.0 |
0.0(5) |
10.0(5) |
90.0(5) |
10.0(5) |
0.0(5) |
0.0(5) |
100.0(6) |
|
23 |
40.0(4) |
40.0 |
10.0(5) |
50.0(5) |
60.0(6) |
60.0(5) |
60.0(5) |
50.0(5) |
0.0(7) |
|
24 |
20.0(4) |
20.0 |
10.0(5) |
80.0(5) |
90.0(6) |
90.0(5) |
90.0(5) |
80.0(5) |
0.0(7) |
|
25 |
0.0(4) |
0.0 |
0.0(5) |
90.0(5) |
100.0(6) |
90.0(5) |
90.0(5) |
90.0(5) |
0.0(7) |
|
26 |
10.0(4) |
20.0 |
0.0(5) |
100.0(5) |
100.0(6) |
100.0(5) |
100.0(5) |
90.0(5) |
30.0(7) |
|
27 |
10.0(5) |
10.0 |
0.0(6) |
10.0(6) |
100.0(6) |
10.0(6) |
10.0(6) |
0.0(6) |
*30.0(7) |
|
28 |
30.0(5) |
10.0 |
0.0(6) |
40.0(6) |
100.0(6) |
40.0(6) |
20.0(6) |
20.0(6) |
*40.0(7) |
|
29 |
10.3(5) |
12.8 |
2.6(6) |
35.9(6) |
100.0(6) |
61.5(6) |
61.5(6) |
51.3(6) |
100.0(7) |
|
30 |
0.0(5) |
0.0 |
0.0(6) |
100.0(6) |
100.0(7) |
100.0(6) |
100.0(6) |
100.0(6) |
0.0(8) |
|
31 |
0.0(6) |
0.0 |
0.0(6) |
100.0(6) |
100.0(7) |
100.0(6) |
100.0(6) |
100.0(6) |
*0.0(8) |
|
32 |
0.0(6) |
0.0 |
0.0(7) |
10.0(7) |
100.0(7) |
10.0(7) |
10.0(7) |
10.0(7) |
*0.0(8) |
|
33 |
20.0(6) |
0.0 |
0.0(7) |
40.0(7) |
100.0(7) |
40.0(7) |
40.0(7) |
10.0(7) |
*0.0(8) |
|
34 |
20.0(6) |
20.0 |
0.0(7) |
80.0(7) |
100.0(7) |
70.0(7) |
60.0(7) |
60.0(7) |
*0.0(8) |
|
35 |
0.0(6) |
0.0 |
0.0(7) |
100.0(7) |
100.0(7) |
100.0(7) |
100.0(7) |
100.0(7) |
*0.0(8) |
until they appear brownish by transmitted light and are more readily visible by reflected light; they now outline the siphonal region rather sharply (Fig. Ib). Tn some instances one or two of these reflecting cells are also visible in the ampullae (amp.. Figs, la-le). No great increase in the number of reflecting cells seems to take place during metamorphosis, at least not during the first day. Seven to eight hours after emergence of the larva the siphons are readily visible and typically the reflecting cells are restricted sharply to the areas immediately surrounding and between the siphons (Fig. Ic) ; however, in a few instances they are more distributed to either side of the dorsal midline as well (Fig. Id). That these cells are very definitely restricted to the dorsal surface of the oozooid can lie seen in optical section (Fig. le).
78 RAY L. WATTERSON
When the oozooids of the colonies started July 31 were first examined on August 2 considerable variation was noted in the appearance of their intersiphonal bands; however, the intersiphonal band of any oozooid could be classified quite readily as one of three types, either : a. stroiuj, in which case there is a broad, almost solid band of rellirtin^ cells between the two siphons (Fig. If) ; 1). medium, in which