Fish Physiology: Reproduction and Growth Bioluminescence, Pigments, and Poisons. Volume III (3)

FISH PHYSIOLOGY Volume III Reproduction and Growth Biohinescence, Pigments, and Poisons

CONTRIBUTORS J. H. S. BLAXTER RYOZO FUJI1

WILLIAM S. HOAR N. R. LILEY

J. A. C. NICOL M. C. QUIMBY FINDLAY E. RUSSELL KEN WOLF TOKI-0 YAMAMOTO

FISH PHYSIOLOGY Edited by W . S . HOAR DEPARTMENT OF ZOOLOGY UNIVERSITY OF BRITISH COLUMBIA VANCOUVER, CANADA

and

D . J. R A N D A L L DEPARTMENT OF ZOOLOGY UNIVERSITY OF BRITISH COLUMBIA VANCOUVER, CANADA

Volume 111

Reproduction and Growth Bioluminescence, Pigments, and Poisons

(23

Academic Press New York and London

1969

COPYRIGHT@ 1969,

BY

ACADEMIC PRESS,INC.

ALL RIGHTS RESERVED, NO PART OF THIS BOOK MAY BE REPRODUCED I N ANY FORM,

BY PHOTOSTAT, MICROFILM, RETRIEVAL SYSTEM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS.

ACADEMIC PRESS, INC. 111 Fifth Avenue, New York, New York 10003

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. Berkeley Square House, London W1X 6BA

LIBRARY OF CONGRESS CATALOG CARDNUMBER: 76-84233

PRINTED IN THE UNITED STATES OF AMERICA

LIST OF CONTRIBUTORS Numbers in parentheses indicate the pages on which the authors' contributions begin.

J. H. S. BLAXTER" (177), Department of Zoology, University of Aberdeen, Aberdeen, Scotland R ~ o z oFUJ I I ~( 3071, Department of Biological Sciences, Northwestern University, Evanston, Illinois

WILLIAM S. HOAR ( l ) ,Department of Zoology, University of British Columbia, Vancouver, Canada

N . R. LILEY(73), Department of Zoology, University of British Columbia, Vancouver, Canada J. A. C. NICOL(355), University of Texas, Marine Science Institute at Port Aransas, Port Aransas, Texas

M. C. QUIMBY (253), Bureau of Sport Fisheries and Wildlife, Eastern Fish Disease Labordo y,Kearneysville, West Virginia

FINDLAY E . RUSSELL(401), University of Southern California School of Medicine, Los Angebs, California KEN WOLF(253),Bureau of Sport Fisheries and Wildlife, Eastern Fish Disease Laboratory, Kearneysville, West Virginia Tom-o YAMAMOTO~( 117), Nagoya University, Biological Institute, Fuculty of Science, Chikusa-ku, Nagoya, Japan

* Present address: N.E.R.C. Marine Research Laboratory, Oban, Argyll, Scotland, and University of Stirling, Stirling, Scotland. t Present address: Division of Biology, National Institute of Radiological Sciences, Chiba City, Japan. $ Present address: Biological Laboratory, Meijo University, Yagoto-Urayama, Showa-ku, Nagoya, Japan. V

This Page Intentionally Left Blank

PREFACE The topics discussed in the third volume of this treatise are extremely diverse. The first volume was devoted to problems of electrolyte and water balance, excretion, and metabolism; the second included much of the endocrinology of fishes. In contrast, Volume 111 is heterogeneous. The first four chapters are devoted to problems of reproduction, differentiation, and development; the last four are quite unrelated, dealing with tissue culture, physiological color changes, venoms, poisons, and those structures which produce light. Thus, the subjects range from the cellular level in fish genetics, development, and tissue culture to organ physiology and interacting organisms at the social level, with descriptions of many curious devices for protection and communication. Admittedly, this arrangement is one of convenience, for it is manifestly impossible to group all the physiological adaptations of fishes into closely integrated book-sized packages. Although lacking in homogeneity, this volume does emphasize the scope of evolutionary adaptation. Fishes have exploited the gamut of devices associated with sexual reproduction and provide examples of highly successful alternatives to sex; their reproductive behavior is frequently complex, and its various phases are neatly timed and associated with environmental cycles. The fishes are unique among the vertebrates in their ability to create light; the production of venoms and poisons and the use of protective coloration play important roles in the defense and protection of many species. Quite apart from the physiology of fishes, it is evident from the chapters on genetics, development, tissue culture, and chromatophores how important the use of fish tissues can b e in fundamental biological research. This volume brings together many scattered observations as well as the results of recent investigations. We feel confident that the general biologist, the zoologist, and the comparative physiologist, as well as the fish physiologist, will find a wealth of exciting information in it.

W. S. HOAR D. J. RANDALL

August, 1969 vii

This Page Intentionally Left Blank

CONTENTS LISTOF CONTRIBUTORS

V

PREFACE

Vii

CONTENTS OF OTHERVOLUMES

xiii

1. Reproduction WillkZTn s. HOUT I. Introduction 11. The Gonads and Their Ducts 111. Viviparity and Gestation IV. The Endocrinology of Reproduction References

1 3 20 40 59

2. Hormones and Reproductive Behavior in Fishes N . R . Liley I. Introduction 11. Gonadal and Thyroid Hormones 111. Pituitary Hormones IV. External Factors and the Endocrine System V. Summary and Discussion VI. Conclusion References

73

75 94 102 104 109 110

3. Sex Differentiation Toki-o Yamamoto I. 11. 111. IV. V.

Introduction: Sexuality in Fishes Hermaphroditism Gonochorism Genetic Basis of Sex Determination Control of Sex Differentiation ix

117 118 127 131 142

CONTENTS

X

VI. Nature of Natural Sex Inducers VII. Differentiation of Secondary Sexual Characters VIII. Summary References

150 153 157 158

4. Development: Eggs and Larvae I. H. S . BlaxCer I. XI. 111. IV. V. VI. VII. VIII. IX. X. XI.

Introduction The Parental Contribution Events in Development Metabolism and Growth Feeding, Digestion, and Starvation Sense Organs Activity and Distribution Mortality, Tolerance, and Optima Meristic Characters Rearing and Farming Conclusions References

178 178 184 191 213 220 221 229 235 238 241 241

5. Fish Cell and Tissue Culture Ken Wolf and M. C. Quimby I. Introduction 11. Physiological Salines

111. IV. V. VI. VII. VIII. IX.

Media Methods Choice of Tissues for Culture Storage and Preservation Fish Cell Lines Shipment of Cell Cultures Needed Developments References

253 260 205 273 286 287 289 294 295 301

6. Chromatophores and Pigments

Ryozo Fujii 1. Introduction 11. Classification and Terminology of Chromatophores

111. Morphology and Chromatophores IV. Chromatophore Pigments V. Physiological Color Changes VI. Morphological Color Changes VII. Other Topics References

307 308 309 313 317 338 341 344

xi

7. Bioluminescence 1. A. C. Nicol I. Introduction 11. 111. IV. V. VI. VII. VIII.

Occurrence Light Organs Biochemistry Regulation of Light Emission Physical Characteristics Significance and Employment of Luminescence Conclusions and Summary References

355 356 357 375 379 384 388 393 394

8. Poisons and Venoms Findlay E . Russell I. Introduction 11. Venomous Fishes 111. Poisonous Fishes References

AUTHOR INDEX SYSTEMATIC INDEX SUBJECTINDEX

401 404 423 440 451 467 477

This Page Intentionally Left Blank

CONTENTS OF OTHER VOLUMES Volume I The Body Compartments and the Distribution of Electrolytes W.N . Holmes and Edward M . Donaldson The Kidney Ckueland P . Hickman,

JT.,

and Benjamin F . Trump

Salt Secretion Frank P . Conte The Effects of Salinity on the Eggs and Larvae of Teleosts F . G. T . Holliday Formation of Excretory Products Roy P . Forster and Leon Coldstein Intermediary Metabolism in Fishes P . W . Hochachka Nutrition, Digestion, and Energy Utilization Arthur M. Phillips, I T .

AUTHOR INDEX-SYSTEMATICINDEX-SUBj ~ c rINDEX

Volume I1 The Pituitary Gland: Anatomy and Histophysiology 1. N . Ball and Bridget I . Baker The Neurohypophysis A. M . Perks

Prolactin (Fish Prolactin or Paralactin) and Growth Hormone 1. N . Ball Thyroid Function and Its Control in Fishes Aubrey GOT^ xiii

xiv

CONTENTS OF OTHER VOLUMES

The Endocrine Pancreas August Epple The Adrenocortical Steroids, Adrenocorticotropin and the Corpuscles of Stannius I . Chester Jones,D . K. 0. Chan, I . W. Henderson, and J . N . Ball The Ultimobranchial Glands and Calcium Regulation D. Harold Copp Urophysis and Caudal Neurosecretory System Howard A. Bern AUTHORINDEX-SYSTEMATICINDEX-SUBj ~ c rINDEX

Volume IV Anatomy and Physiology of the Central Nervous System JerakE 3. Bemstein Pineal Organ lames C. Fenurick The Mauthner Neuron

3. Diamond Autonomic Nervous System Graeme Campbell The Circulatory System

D. 3. Randall Acid-Base Balance C. Albers Properties of Fish Hemoglobins Austen Riggs Gas Exchange in Fish D. J. Randall The Regulation of Breathing G. Shelton Air Breathing in Fishes Kiell Iohansen The Swim Bladder as a Hydrostatic Organ J o h n B. Steen

CONTENTS OF OTHER VOLUMES

xv

Hydrostatic Pressure Malcolm S. Gordon Immunology of Fish john E . Cushing AUTHORINDEX-SYSTEMATIC INDEX-SUBJECT INDEX Volume V (Tentative)

Vision: Visual Pigments F. W. Munz Vision: Electrophysiology of the Retina Tsuneo Tomita Vision: The Experimental Analysis of Visual Behavior D. Ingb Chemoreception Toshiaki I . Hara Temperature Receptors R. W. Murray

Sound Production and Detection W. N . Tavolga The Labyrinth 0. L0U;emtein Mechanoreceptors: The Lateral Line Organ Receptors Ake Flock Electroreception M . V. L. Bennett Electric Organs M . V. L. Bennett Volume VI (Tentative)

The Effect of Environmental Factors on the Physiology of Fish: An Examination of the Different Categories of Physiological Adaptation F . E. 1. Fry Action of the Environment on Biochemical Systems P . W. Hochachka and G. N . Somero

mi Learning and Memory Paul Rozin and Henry Gleitmun The Ethological Analysis of Fish Behavior G. P. Baerends Locomotion R. Bainbridge Biological Rhythms H. 0. Schwassmann Orientation and Fish Migration A. D. Hasbr Special Techniques D. 1. Randull and W.S . Hoar

CONTENTS OF OTHER VOLUMES

1 REPRODUCTION WlLLlAM S . HOAR

.

. . . . . . . . . .

I. Introduction . . . . . . 11. The Gonads and Their Ducts . A. Embryology and Phylogeny . . . B. The Male . . . . . . . C. The Female . . . . 111. Viviparity and Gestation . . . . A. Evolutionary Considerations . . . , B. Viviparity among the Chondrichthyes . . C. Viviparity among the Teleosts . IV. The Endocrinology of Reproduction . A. The Pituitary Gonadotropins . . . . B. The Gonadal Steroids . . . C. Reproductive Cycles and Their Coordination References . . . . . . . .

. .

.

.

.

.

. .

. . . .

.

.

.

.

. . . . . . . . .

. .

.

. . . . . . . . .

. .

.

. . . . . . .

. .

.

. . . . . . .

. .

. . . .

1 3 3 8

15 20 20

22 30 40 41 50 56 59

I. INTRODUCTION

The fishes, like all the other vertebrates, reproduce sexually. In the vast majority of species, eggs and spermatozoa are formed in separate individuals (dioecious), and the gametes are expelled into the surrounding water where fertilization takes place immediately and is promptly followed by the development of a new generation. Within this broad pattern there is an amazing array of curious modifications so that the fishes as a group exemplify almost every device known among sexually reproducing animals; indeed, they display some variations which may be unique in the animal kingdom. 1

2

WILLIAM S . HOAR

Although most fishes are dioecious, hermaphroditism does occurparticularly among the cyclostomes and teleosts; in some species it is the normal way of life ( Atz, 1964). Parthenogenesis has not been observed among fishes in nature but has been produced experimentally (Austin and Walton, 1960). Gynogenesis occurs in Poecilia formosa and perhaps also in some populations of Carmsius auratus (Hoar, 1965a). These modes of reproduction are, however, unusual; and most of the 20,000 or more known species of fish have male and female organs in separate individuals. At one extreme, the two sexes are externally indistinguishable. They swim together during the breeding season to discharge their genital products into the water without specialized mating behavior. At the other extreme, marked differences in morphology and coloration are characteristic of the males and females and these play distinctive roles in elaborate presexual behavior, courtship, mating, and parental care. A synchronized spawning without copulation is usual but copulation occurs at all levels in fish phylogeny; it sometimes merely insures close proximity of eggs and sperm discharged into the water but more frequently involves insemination of the females. Although internal fertilization may be followed by the laying of newly fertilized eggs or the release of primitive larvae, viviparity is highly specialized in some elasmobranchs and teleosts with young born in advanced stages of development; in Cymatogaster aggregata the males may even be sexually mature at birth. Sexual activities are normally followed by the prompt fertilization of eggs but sometimes sperm are stored in the female for prolonged periods prior to fertilization; occasionally the fertilized eggs develop only after a period of diapause ( Wourms, 1967); sometimes development occurs but hatching is delayed during adverse environmental conditions ( Harrington, 1959). Breder and Rosen (1966) have summarized these and many other curious specializations in an extended chart; this tabulates information on secondary sex characters, mating, breeding, and parental behavior for each of the many families of fishes. There are many specializations associated with reproduction in fishes but only a few of them are considered in this chapter. Breeding behavior, fertilization, and early development are considered in other chapters. The present account is confined to a discussion of the gonads and their associated ducts, the special physiology of viviparity, and the endocrinology of reproduction. The latter topic has been repeatedly reviewed (Pickford and Atz, 1957; Dodd, 196Oa; Hoar, 1965a, 1966), but the other two aspects of fish reproduction have not been recently summarized (see Hoar, 1957; GBrard, 1958; Bertin, 1958a).

1.

REPRODUCXION

3

11. THE GONADS AND THEIR DUCTS

A. Embryology and Phylogeny

A knowledge of early embryology, as well as adult morphology, is essential to an understanding of several features of the comparative physiology of the reproductive system of fishes. In particular, the two very different patterns of gonadal ontogeny may account for the variable frequency of intersexuality in different groups while comparative studies of spawning, fertilization, and gestation must be based on an understanding of the diversified anatomy of the gonads and their ducts. The gonads of all vertebrates arise in the dorsolateral lining of the peritoneal cavity-one gonad on each side of the dorsal mesentery. Their development is intimately associated with that of the nephric system. In most of the vertebrates each gonad has a double origin, developing from two distinct but closely associated cellular proliferations. The more laterally located cortex or cortical portion arises as an elongated ridge of peritoneal wall and is destined to become an ovary. The medulla or medullary portion which is destined to form the testis arises from a more medial cellular proliferation which also produces the adrenocortical tissue ( interrenal or mesonephric blastema) . Usually, one of these portions grows rapidly while the other fails to develop and the sex of the individual is thus determined at a very early stage. A section through the undifferentiated gonad of a 22-mm dogfish embryo is depicted in Fig. 1.

Fig. 1. Cross section through a 22-mm embryo dogfish, Scyliorhinus canicu~us,to show the origins of the cortex and medulla of the gonad. From Chieffi (1952).

4

WILLIAM S. HOAR

This pattern of gonad differentiation from two different components is characteristic of the elasmobranchs (Chieffi, 1949, 1952, 1967) and all of the tetrapods. In contrast, the gonads of cyclostomes and teleosts develop from single primordia. In cyclostomes and teleosts the entire gonad develops directly in the peritoneal epithelium and corresponds to only the cortex of other vertebrates. There is evidently no contribution from the interrenal (mesonephric) blastema. Atz ( 1964) has summarized the literature and provided an extensive bibliography; D’Ancona (1950, 1956, 1960) did much of the early work on teleosts. It has been suggested that these differences in embryology may account for the more widespread occurrence of intersexuality among the cyclostomes and teleosts. Whether the ontogeny of the gonad is by way of a single or a double primordium, distinctive cells (the primordial germ cells) which are destined to form gametes first appear within or migrate into the cortical portion of the gonad. They can usually be identified as conspicuously larger cells within this modified layer of proliferating mesothelium (germinal epithelium). There is a voluminous literature on the origin of the germ cells with a considerable body of evidence for a widespread development of these cells and a subsequent migration of them into the germinal epithelium. This literature will not be reviewed here; it can be readily reached through standard embryology texts (Nelsen, 1953) and reviews (Everett, 1945; Brambell, 1956; Franchi et al., 1962). From whatever source, these distinctive cells can be identified at an early stage within the thickening layer of germinal epithelium. In genetic male elasmobranchs there is an early migration of germ cells from the cortex, where they first appear, into the medulla (Chieffi, 1949, 1967); in the genetic females there may be a transient migration of some of the germ cells into the medulla but many of them retain their cortical location to form the basis of ovarian differentiation. The origin and segregation of germ cells within the single primordium of the cyclostome and teleost gonad has been frequently described and reviewed (Nelsen, 1953; Brambell, 1956; Franchi et al., 1962). Under normal conditions, genetic factors probably determine whether the developing fish will be a male or a female (Dodd, 1960b); genetic aspects of sex determination are considered in the chapter by Yamamoto, this volume. Earlier workers postulated male and female inductor substances or hormones which controlled the course of development. Witschi (1942, 1950) who carried out the pioneer work on amphibians referred to them as “medullarin” and “corticin,” for male and female, respectively; DAncona (1945, 1950) working with the teleosts called these theoretical substances “androgenin” and “gynogenin,” Reinboth ( 1962) does not

1. REPRODUCTION

5

consider evidence for the existence of these factors to be at all convincing; Atz (1964) agrees with Reinboth's conclusions. Whether or not there are special embryonic hormones concerned with the determination of sex, it is well established that the differentiation of the gonadal primordium can be readily modified with gonadal steroids similar to those found in adults. Androgens stimulate the development of testes, and estrogens promote ovarian differentiation. Findings are consistent and the literature has been frequently reviewed (Dodd, 1960a,b; Atz, 1964; Chieffi, 1967; and the chapter by Yamamoto, this volume). The gonads of vertebrates always originate from biIateral primordia, but many species as adults possess only one reproductive gland. In some of the fishes there is a fusion of the two primordia during development (as in the ovaries of lampreys), while in other cases one of the gonads fails to develop (as in the myxinoid ovary). The literature reveals a range of specializations in all groups of fishes from complete fusions to partial fusions involving only the posterior portion of the gonads or just the gonoducts; sometimes one of the gonads is rudimentary or merely smaller but still present. Franchi et al. (1962) summarize the pertinent literature with specific examples. The comparative anatomist has found some of his most interesting problems in the phylogeny of the gonoducts and their relationships to the mesonephric tubules and ducts. Gonoducts are absent in the cyclostomes. Spermatozoa or ova are discharged from the surface of the gonad directly into the body cavity and then pass through pores into the urinary sinus or urinary duct; they are expelled through the cloaca or urinogenital papilla-depending on the anatomy of the urinogenital opening in the particular sex and species (Dodd, 1960a; Walvig, 1963). Gonoducts are present in all groups of the gnathostomes although they may be secondarily lost in some fishes (for example, in the Greenland shark, Laemurgus borealis, and in the Salmonidae among the Teleostei, Balfour, 1875). Kerr ( 1901) provided the classic description of the origin of the male gonoducts from the mesonephric system (Fig. 2). The sturgeon and garpike are thought to represent a primitive situation where some of the renal tubules throughout the length of the mesonephros have been conscripted into the service of the testis and form vasa efferentia which drain into the mesonephric duct or vas deferens. In the Chondrichthyes (and Amphibia), the testis is thought to have taken over a group of anterior mesonephric tubules which cease to have any relationship with the excretory system; in Lepidosiren, the vasa efferentia are formed from some of the posterior mesonephric tubules. In Polypterus and the Teleostei there is no connection between the mesonephros and the gonad at maturity; and the vas deferens is

6

WILLIAM S. HOAR

Fig. 2. Representative types of urinogenital systems in male fishes. Upper figures, redrawn from Portman (1948) with permission of Dr. A. Portman and B. Schwabe (Basel); lower figures, originals by Margaret Jensen.

quite separate from the ureter or mesonephric duct (Fig. 2 ) . It is generally assumed, however, that the main gonoduct has been derived from the mesonephric duct during phylogeny. More complete descriptions of this system will be found in Goodrich (1930), GCrard ( 1958) and van den Broek (1967).

1. REPRODUCTION

7

In all of the vertebrates, except some of the more specialized fishes, the ova are discharged into the peritoneal cavity and find their way to the outside through oviducts (Mullerian ducts), which pass from open anterior funnels to the cloaca (Fig. 3 ) . In these groups with naked ovaries ( gymnovuriun condition) and open ovarian funnels, the genital ducts are derived as in the male from the mesonephric ducts although

Fig. 3. Representative types of urinogenital systems in female fishes. Upper figures, redrawn from Portman (1948) with permission of Dr. A. Portman and B. Schwabe (Basel); lower figures, originals by Margaret Jensen.

8

WILLIAM S. HOAR

the evidence for this origin is completely lost in the land vertebrates ( Romer, 1955). In the Teleostei, the oviducts are posterior continuations of the ovarian tunic. The embryology of the ovary and its duct varies so that the ovary in some teleosts has a central ovarian cavity continuous with the oviduct while in others the oviducts are para-ovarian ( Goodrich, 1930; GBrard, 1958; Franchi, 1962; van den Broek, 1967); in any case the oviducts are formed by the backward growth of the same peritoneal folds which enclose the ovary during its development ( cystovarian condition). The gonoducts are also continuous with the ovaries in the holostean Lepidosteus, thus providing the exception to the rule that only teleosts fail to discharge their ova into the peritoneal cavity. Some of the teleosts are themselves exceptional in that they do release their eggs directly into the body cavity. In the Salmonidae, Galaxiidae, Hyodontidae, Notopteridae, Osteoglassidae, and the cyprinid Misgurnus, the oviducts degenerate in whole or in part so that the ova pass into the peritoneal cavity and thence through pores or funnels, depending on the degree of degeneration, to the exterior. In the Anguillidae the loss of gonoducts occurs in both males and females ( Goodrich, 1930).

B. The Male 1. THETESTIS AND SPERMATOGENESIS Spermatozoa are formed from the sperm mother cells or spermatogonia through a series of cytological stages collectively referred to as “spermatogenesis.” This process involves a proliferation of spermatogonia through repeated mitotic divisions and growth to form primary spermatocytes; these then undergo reduction division to form secondary spermatocytes; the division of the secondary spermatocytes produces the spennatids which then metamorphose into the motile and potentially functional gametes-spermatozoa, spermia or sperm. This process of spermatid metamorphosis is often called “spermiogenesis,”Details of the cytological changes are similar in all vertebrates as described in standard textbooks of histology and embryology. Physiologists are interested in the factorsboth environmental and hormonal-which trigger waves of spermatogenesis at different seasons and control the essential steps of meiosis (division of primary to secondary spermatocytes) and the metamorphosis of the spermatid with eventual release of mature sperm. In some species -particularly the elasmobranchs and viviparous teleosts-sperm production involves the packaging of sperm into sperm balls or spermatophores which are transferred to the female.

1. REPRODUCTION

9

Spermatogenesis occurs within testicular units which may take the form of small sacs, ampullae, lobules, or tubules; in many groups of fishes these differ radically from the familiar seminiferous tubules of the mammalian testis. In the cyclostome, spermatogenesis occurs within small bladders, follicles, or ampullae. These are separated by a delicate connective tissue; a number of units may be grouped together and bounded by somewhat thicker connective tissue to form lobules (Dodd et al., 1960; Walvig, 1963). Spermatogenesis is almost synchronous throughout the many follicles and just prior to spawning the follicles filled with mature sperm rupture to release their contents into the body cavity. Walvig ( 1963) summarizes the cytological details of spermatogenesis in Myxine. In elasmobranchs, spermatogenesis occurs within a mass of ampullae arranged in a manner which seems to be unique among the vertebrates. The testis of the basking shark, Cetorhinus maximus, carefully described by L. H. Matthews (1950), is divided by connective tissue trabeculae into many lobules each of which corresponds to the entire testis of the dogfish, Scylliorhinus cunicula, as described by Fratini (1953) and Mellinger (1965). The structure of the dogfish testis is shown diagrammatically in Fig. 4. The spermatogenetic units, usually called “ampullae,” are proliferated from a mesoventral area of the testis referred to as the “tubulogenic zone.” Within this zone, nests of cells-somewhat like primary ovarian follicles-arise and proliferate to form small tubules or ampullae which gradually shift toward the dorsal side of the organ while spermatogenesis occurs within them, By the time the ampullae reach the dorsal surface of the testis, the sperm ( a constant number in each ampulla: Stanley, 1962; Mellinger, 1965) are ready for discharge into the efferent ducts which emerge from the testis at this point. At this stage, the Sertoli cells surround the ampullae and are clearly associated with packets of sperm. According to L. H. Matthews (1950), the ampullae shrink after sperm are discharged into these collecting tubules and the Sertoli cells are resorbed. It is of interest that all of the gonocytes within any one ampulla are in the same stage of spermatogenesis and that within the testis, distinct zones are evident from ventral to dorsal surface with all the tubules of a particular zone in a similar stage of development. Thus, in studies of the pituitary regulation of spermatogenesis, Dodd and his colleagues (1960) were readily able to spot a distinct zone of degeneration in the primary spermatocytes when it appeared following hypophysectomy. The cytology of the elasmobranch testis, including spermatogenesis, the development of the Sertoli cells, and the formation of spermatophores, has been detailed by L. H. Matthews (1950) and Fratini ( 1953).

10

WILLIAM S. HOAR

Fig. 4. Structure of the chondrichthyan testis. Redrawn from van den Broek (1967) with the permission of Asher & Co., Amsterdam, and from Fratini (1953) with the permission of Stazione Zoologica di Napoli.

There are now good descriptions of the testicular histology of several species of teleosts. Among the early papers, the following are particularly helpful: C. L. Turner’s description (1919) of the spermary of the perch, Craig-Bennett’s account ( 1931) of the stickleback, S. A. Matthews’

1. REPRODUCTION

11

report (1938) on Fundulus, and Cooper's study (1952) of the crappies. Many other investigations are cited in the bibliographies of these papers and in the reviews by Hoar (1957) and Dodd (1960a). More recent descriptions are available for the minnow Couesius ( Ahsan, 1966a,b), the rockfish Sebastodes ( Moser, 1967a), the sea perch Cymatogaster (Wiebe, 1968b), and the guppy Poecilia (Pandey, 1969a,c). Testes of dBerent species vary in complexity; the brief description which follows is a generalized one. The main sperm duct (vas deferens) arises from the posterior mesodorsal surface of the elongated testis and leads to the urinogenital papilla. It may be traced anteriorly for a variable distance in a connective tissue groove of the testis along with the spermatic blood vessels and nerves. In many teleosts, the paired testes fuse posteriorly and the vasa deferentia are combined into a single sperm duct. Within the body of the testis, the

Fig. 5. Structure of testis of guppy, Poecilia reticulatu. ( A ) Diagrammatic section to show relation of gametogenetic tissue (acini or cysts) to the system of ducts, and ( B ) series of cysts to show differentiation of gametes from spennatogonia ( a t right) to mature spermatophores surrounded by Sertoli cells ( a t left). Sertoli cells become confluent with the epithelia1 cells of the efferent ducts when the spermatophores enter the ducts. Based on Pandey (1969a).

12

WILLIAM S. HOAR

main sperm ducts give rise to smaller ducts (vasa efferentia) which penetrate ventrally and laterally to form a drainage system of variable complexity. In some species these tubules are extremely short (poeciliids, for example), while in others they form an extensive system of seminiferous tubules which can be followed almost to the periphery of the organ (Fundulus, the rockfishes, and the cottids). Testes of the poeciliid type are sometimes referred to as “acinar” (Fig. 5 ) while those with the extensive duct systems are called “tubular.” This difference is one of degree rather than kind. It is to be noted that the seminiferous tubules of the teleost-in contrast to those of the higher vertebrates-lack a permanent germinal epithelium. Whether the testis is acinar or tubular, nests of spermatogonia proliferate from the resting germ cells near the margin of the organ. In the acinar type, these nests of cells or cysts undergo the various stages of maturation as they are displaced toward the sperm ducts into which they eventually discharge their contents (Fig. 5 ) . In the tubular testis, the resting germ cells are particularly evident and packed together at the blind ends of the tubules near the periphery, but many of them migrate or are displaced along the walls of the tubules. In active spermatogenesis, nests of spermatogonia proliferate both from the ends of the tubules and from the resting germ cells along their walls. Thus, at the end of spermiogenesis, the seminiferous tubules are packed with sperm as the masses of gametes from a multitude of matured cysts combine within the tubules. During maturation all of the cells within one of the cysts are in approximately the same stage of development; the degree of synchrony among the many cysts varies in different species.

2. THE ENDOCRINE AND SUPPORTING TISSUES OF

THE

TESTIS

Current histological and histochemical techniques have now resolved a long-standing uncertainty about the tissues responsible for the production of the testicular androgens ( Hoar, 1965,a). Well-vascularized clusters of cells similar to those described by Leydig many years ago in the mammalian testis have been identified between the seminiferous alveoli and tubules of many different fishes. In routine H and E sections, these large cells with spherical or oval nuclei often appear vacuolated because of the removal of lipoidal substances. Their endocrine nature was first postulated from the marked seasonal proliferation which occurs just prior to the breeding season in the stickleback Gasterosteus (Courrier, 1921; Craig-Bennett, 1931); in this same species Gottfried and van Mullem (1967) have recently shown a convincing correlation between the histological development of the tissue and its biochemical

1. REPRODUCI’ION

13

content of androgen. Nests of typical Leydig cells or interstitial tissue have now been identified in the cyclostomes (Chieffi and Botte, 1962; Hardisty et al., 1967), in all groups of elasmobranchs (Marshall, 1960; Chieffi et al., 1961; Chieffi, 1962, 1967), in the lungfishes and the coelacanth Latimeria (Marshall, 1960) as well as the testes of many teleosts ( Gasterosteus Tilapia, Tinca, Solea, Lebistes, Cymatogaster, Oncorhynchus, and others). Staining of the cytoplasmic droplets with sudanophilic coloring agents and the demonstration of steroid dehydrogenases leaves little doubt of their role in steroidogenesis (Marshall, 1960; Follenius and Porte, 1960; Collenot and Ozon, 1964; Delrio et al., 1965; Stanley et al., 1965; Yaron, 1966; Wiebe, 1969a). In some species of teleosts, however, the hormone-producing cells are located in the walls (basement membrane) of the seminiferous tubules. Marshall and Lofts (1956) who first noted this difference in their studies on the pike, Esox lucius, and the char, Salvelinus willughbii, referred to this tissue as the “lobule boundary cells.” The difference between interstitial and lobule boundary cells is largely one of distribution since both tissues arise from the same source and are similar histochemically (reviews by Marshall, 1960; Dodd, 1960a; Hoar, 1965a). It may be of interest to note that the Urodela also possess lobule boundary cells, in contrast to the Anura and all other groups of tetrapods (Marshall, 1960). Sertoli cells (Figs. 4 and 5 ) are also prominent in the testes of all groups of fishes (L. €3. Matthews, 1950; Fratini, 1953; Chieffi and Botte, 1962; Lagios, 1965; Wiebe, 196813, 1969a,b; Pandey, 1969a). The spermatogenetic units-whether cysts, ampullae, or tubules-are bounded by a thin layer of connective tissue (the basement membrane) and contain two types of cells; one of these is the gonocyte giving rise to the several generations of spermatogenetic cells, while the other is the Sertoli or supporting cell believed to play a nutritive role during spermiogenesis. The process whereby spermatids become embedded in the centripetal end of the Sertoli cells to undergo metamorphosis is well described in the higher vertebrates ( Nelsen, 1953; Patten, 1953; Ham, 1965). Cytological details probably vary in the different species, but in all cases the spermatids (perhaps earlier stages in some fishes: Stanley, 1962) become intimately associated with these nurse cells and presumably draw nourishment from them during transformation. The fully developed sperm are attached to the Sertoli cells prior to spermiation; this association can be very nicely seen in fishes such as the sharks, the poeciliids, and the embiotocids which form spermatophores. In these groups, where all of the sperm in a unit mature at the same time, the Sertoli cells form an almost complete layer just inside the basement membrane (Figs. 4 and 5); in many other fishes the situation is similar to that of the higher vertebrates where

14

WILLIAM S. HOAR

Sertoli cells are interspersed at intervals along the basement membrane with the groups of spermatogonia between them. L. H. Matthews (1950) and Fratini (1953) have detailed the cytogenesis of the Sertoli cell system in the basking shark and the dogfish. Pandey (1969a) describes Sertoli cell development in the guppy. In addition to their nutritive role, the Sertoli cells may be concerned with the phagocytosis of unused sperm (Vaupel, 1929; Nelsen, 1953; Lagios, 1965) and involved in hormone production. The presence of hydroxysteroid dehydrogenases has been demonstrated in the Sertoli cells of dogfishes (Collenot and Ozon, 1964; Simpson and Wardle, 1967) and in the surfperch (Wiebe, 1969a). In summary, the cytology of the Sertoli cell suggests three functions : nutritive, phagocytic, and hormonal; in this and in their embryology (L. H. Matthews, 1950; Lofts et al., 1966) they seem very similar to the granulosa cells of the ovarian follicle which are described later.

3. SECRETIONSOF THE SPERMDucrs AND MALE SECONDARY SEX CHARACXXRS Following spermatogenesis, the mature sperm are made ready for discharge or spermiation. The glandular epithelial lining of the sperm duct probably always contributes to the seminal discharge but there seems to be no systematic study of this. However, a wide variation is recognized-at one extreme, the formation of specialized packets of sperm (spermatophores) and, at the other, a mere thinning of the semen through the hydration of the testis and accumulation of fluid in the testicular passages. L. H. Matthews (1950) has detailed the formation of the complex spermatophores of the basking shark, Cetorhinus maximus, where these sperm packets range in size from a few millimeters up to 25 or 30 mm and have a cortex of translucent hyaline material surrounding a central mass of opaque white sperm; the sperm mass may be 10 mm in diameter. The general process, as described by L. H. Matthews, seems to be representative of the elasmobranchs although the structure of the spermatophores varies in digerent species from simple sperm aggregations to the hyaline packets of the basking shark. In the elasmobranchs, sperm released from the ampullae into the efferent canals pass through a mass of coiled glandular tubules (gland of Leydig) which are derived from the anterior nonurinary portion of the mesonephros. Sperm contained in the secretions of the Leydig gland then pass into an expansion of the vas deferens known as the ampulla. As they move through the complex system of septae in this structure they are consolidated and

15

1. REPRODUCXION

receive additional secretions such as the hyaline cortex of the basking shark spermatophore. Borcea’s monograph ( 1906) should be consulted for anatomical details of the gonoducts of elasmobranchs. Spermatophores are also regularly formed in the viviparous teleosts. They have been described in the embiotocids (Lagios, 1965; Wiebe, 1969b) and in the poeciliids ( Phillipi, 1908; Henn, 1916; Pandey, 1969a). In these groups, the aggregations of sperm formed within the seminiferous acini or tubules become arranged with their tails directed centrally and the heads oriented peripherally to form the sperm balls. As these pass through the efferent ducts, they seem to receive a gelatinous secretion which binds them together so that they remain intact during transfer to the female. More complex spermatophores have also been described in teleosts, but there are very few studies of either the cytogenesis or the physiology of fish spermatophores (see review by Bertin, 1958a). Spenniation in the goldfish is typical of the much simpler process involving only a thinning of the semen. Clemens and his associates (1964; Clemens and Grant, 1964, 1965) have described the weight changes in the testis and established a pituitary regulation of the spermiation process. Yamazaki and Donaldson (1968a) have used the spermiation of the goldfish in the bioassay of salmon pituitary gonadotropin. In some teleosts ( Ariidae, Gobeiidae, and Blennoidae ), large structures often referred to as “seminal vesicles” are found as glandular developments from the sperm ducts-occasionally from the testis as in Tachicorystes (von Ihering, 1937). These “seminal vesicles” do not store sperm and are not comparable to the structures of the same name in the higher vertebrates. They provide secretions which are of importance in sperm transfer or other breeding activities. Descriptions of these glands and the many other interesting secondary sex characters of fishes are beyond the scope of this review. Good general accounts with bibliographies have been given by Bertin (1958a) and Breder and Rosen ( 1966). Secretory activities of the sperm ducts, accessory glands, and secondary sex characters show a marked seasonal development and are under the control of the androgenic secretions of the testis (Section IV, C and chapter by Liley, this volume).

C. The Female 1. THEHISTORY OF

THE

OVARIANFOLLICLE

In structure, the fish ovary ranges from an expanded mesentery (mesovarium) which dehisces mature ova from its ventral margin in the hagfish Myxine (Lyngnes, 1936; Walvig, 1963) to a complex hollow

16

WILLIAM S . HOAR

organ in the viviparous teleosts where the gonad produces eggs, stores sperm, serves as a site for fertilization, and provides nourishment for the development of young to an advanced stage. Comprehensive reviews of the earlier literature are available (Hoar, 1955, 1957, 1965a; Dodd, 1960a, Franchi, 1962). The description which follows is a generalized one. Ovarian follicles develop from or in association with the germinal epithelium which covers the surface of the ovary as an extension of the peritoneum ( mesovarium ) . As described earlier, this germinal epithelium also lines the cavity of the hollow teleost (cystovarian ) ovary. The ovary of the basking shark, Cetorhinus maximus-single organ-is evidently exceptional among the elasmobranchs in that the germinal epithelium invaginates to form a series of tubular ramifications within a gonad which is superficially similar to that of the teleosts (L. H. Matthews, 1950). It differs, however, both in its embryology and in its gonoduct. The cavities of the basking shark ovary open into a pocket on the right side of the organ and ova discharged into this pocket pass via the peritoneum into the open end of the Miillerian duct. These hollow ovaries of the teleosts and the basking shark are unique among vertebrates and quite different from the hollow ovaries of some other elasmobranchs and the amphibians, where the lining is not germinal epithelium and where the cavities develop as large lymph spaces within the stroma (medulla) of the gland. The numerous ovarian follicles of the fish ovary are supported by a richly vascular connective tissue stroma which extends into the gland from the somewhat denser connective tissue layer (tunica albuginea) just under the germinal epithelium. The internal lining of the hollow teleost ovary is thrown into a complex series of folds (ovigerous folds) which may almost obliterate the cavity (Fig. 6). Fish eggs are discharged from mature ovarian follicles into the peritoneum or into the cavity of the ovary; it may be significant that the stroma of the ovary is rich in elastic tissue and smooth muscle. At an early stage, the oogonia which arise from primordial sex cells either in or near the germinal epithelium become surrounded by a layer of small epithelial cells to form the ovarian follicle. In cyclostomes and teleosts this follicular epithelium is single-layered while in elasmobranchs and amniotes it is usually composed of several layers (Franchi, 1962). The connective tissue near this nest of cells forms a distinct theca, which in some species assumes a very active role during the later history of the follicle. As the follicle differentiates and the ovum becomes mature, the epithelial cells increase in size and number to form a glandular granulosa while the theca becomes more distinct and may be divided into an interna and an externa. The maturing ovum is separated from the granulosa by a noncellular membrane usually called the “zona pellucida”;

Fig. 6. Structure of the ovary in Cymutogaster. Upper left, section of ovary to show oocytes developing in ovigerous folds: ( A ) nest of oocytes in very early stages of development, ( B ) Class I oocyte, ( C ) Class I11 oocyte, ( D ) Stage I1 atretic oocyte, and ( E ) atretic oocyte almost at Stage 111. Bottom, Cymutogaster before birth to show specidized dorsal fin, Upper left and bottom two diagrams redrawn from C. L. Turner (1938a); other diagrams, courtesy of John P. Wiebe.

17

18

WILLIAM S . HOAR

the terminology of the egg membranes is not always consistent (Nelsen, 1953; Brambell, 1956). An early follicle and an almost mature follicle of Cymatogaster are depicted in Fig. 6. The functions of the follicular epithelium in fishes are still problematic. The granulosa has a recognized responsibility for the deposition of yolk in the developing ovum and for its removal in ova which degenerate before ovulation. Yolk deposition in some elasmobranchs and reptiles apparently takes place through specialized protoplasmic processes which can be seen to penetrate the zona pellucida from particular granulosa cells; nutritive transfer is not microscopically evident in other fishes ( Brambell, 1956; Bertin, 1958a). The phagocytic activities of the granulosa cells have been described repeatedly by histologists since the last century. In all groups of vertebrates, the number of ovarian follicles started on the road to development is greatly in excess of the number of eggs which are eventually produced; some ova are normally resorbed in different stages of development ( Lyngnes, 1936; Busson-Mabillot, 1967; and reviews cited). In addition to its nutritive and phagocytic functions, the granulosa may also be concerned with the elaboration of the ovarian hormones. The fish ovary does not contain interstitial tissue comparable in development and histochemistry to the Leydig cells of the testis; the theca of the follicle, which in some mammals participates in the formation of the corpus luteum (Brambell, 1956; Ham, 1965) and probably secretes progesterone-perhaps also estrogen-never shows a sudanophilia in fishes and is evidently a simple fibroelastic connective tissue. It is evident that the ovarian hormones must be synthesized by the ovum or the granulosa or the corpus luteum (corpus atreticum) which develops from the granulosa. Major interest has centered around the corpora, which in this review are termed “corpora lutea” even though they may not be physiologically equivalent to the corpora lutea of mammals which are known to produce progesterone and are under the control of the pituitary gland. Hisaw and Hisaw (1959) and Chieffi (1962) agree with this terminology. Corpora lutea have now been observed in all groups of fishes. Their histogenesis has been described repeatedly since the beginning of the century and a rich literature is cited in several reviews of fish endocrinology (Brambell, 1956; Ball, 1960; Hoar, 1955, 1965a; Dodd, 1960a). Although there is a remarkable variation in the extent of these proliferations, they regularly appear when yolky ova become atretic (preovulatory corpora lutea or corpora atretica) or during the postovulatory history of the follicle (postovulatory or corpora lutea of ovulation). The two opposing views concerning their physiology have been ably maintained from the studies of elasmobranchs by Hisaw (1959, 1963) who

1.

19

REX’RODUCITON

argues that they are concerned with yolk phagocytosis in the preovulatory follicles or the removal of tissue fragments and blood cells in the postovulatory ones, while Chieffi (1961, 1962, 1967) finds substantial histochemical evidence for steroidogenesis and attributes an endocrine function to them. Their function is just as problematic in other groups of fishes (Ball, 1960). The controversy was detailed recently and will not be repeated here; the conclusion reached at that time (Hoar, 1965a) still seems valid; it is suggested “that estrogen synthesis was one of the responsibilities of the granulosa from the earliest stages of vertebrate phylogeny and that this capacity developed in association with the synthesis of lipid materials present in the yolk; the high estrogen content of yolk in the ova of some species (Gottfried, 1964) may represent their entire reserve of this hormone. With the variety and complexity of reproductive processes and controls found among fishes, it seems entirely reasonable that the same granulosa cells (in some fishes) may have specialized in hormone production to the point where corpora atretica become functional corpora lutea-even though they may synthesize estrogen rather than progesterone as their hormone.” Chieffis studies (1961, 1967) indicate how different the history of the pre and postovulatory bodies may be in closely related groups of fishes.

2. ACCESSORYREPRODUCTIVE STRUCTURES AND PHYSIOLOGY OF THE OVIDUCTS

THE

In contrast to the males, secondary sex characters are inconspicuous in female fishes (Breder and Rosen, 1966). The females are frequently larger, occasionally have specialized ventral fins ( Ariidae) , distinctive ventral surfaces or folds of unknown function on the abdominal wall (Bunocephalidae and Bagridae), tubercles on the head (Osteoglossidae), or an enlarged urinogenital papilla ( Notopteridae, Cyprinidae) . The genital papilla may be only slightly enlarged but in the cyprinid Rhodeus amarus-the European bitterling-it often extends well beyond the caudal fin as a specialized ovipositor (Bretschneider and de Wit, 1947; Shirai, 1964). Studies of the physiology of the oviduct have been almost completely confined to the Chondrichthyes. In this group, the oviduct (Miillerian duct) not only serves as an open tube for the collection of eggs from the abdominal cavity and their transport to the cloaca but also provides the secretions concerned with the formation of horny shelled eggs in the oviparous species and a site for the development of the young in the viviparous forms. In addition, this tube also serves for the reception of sperm and, at least in some species, for sperm storage and the dissolution of the hyaline cortex of the spermatophores (Metten, 1941; L. H.

20

WILLIAM S . HOAR

Matthews, 1950). The oviduct of viviparous teleosts is also conccrned with some of these sperm functions, but there are no systematic studies of its physiology. The four regions of the Mullerian duct are shown diagrammatically in Fig. 3; the ostium or funnel, the oviducal or nidamental gland, the connecting isthmus, and the expanded posterior uterus. In some species the ostium is closely applied to an ovarian pocket from which the ova emerge to enter it directly; in other species the eggs are discharged at many points on the surface of the ovary and carried into the funnel by continuously beating cilia which line the abdominal cavity. The ciliation of the abdominal cavity has been described and illustrated by Metten ( 1941). The absence of cilia in males and immature females (Metten, 1941) and changes in the size of the ostium during the breeding season (L. H. Matthews, 1950) suggest a hormonal regulation of these structures; estrogens have been shown to stimulate the development of nidamental glands and other areas of the Mullerian tubes (Thiebold, 1954; Dodd and Goddard, 1961). In all chondrichthyans, the oviducal gland secretes albumen and mucus; it is relatively larger in the oviparous species where it is also responsible for the formation of the shell. Two or three distinct glandular zones may be distinguished. In the oviparous species there is an anterior albumen-secreting area and a posterior shell-secreting zone; an intermediate mucus secreting area may be present, as in the ratfish, Hydrolagus colliei, or absent, as in the ovoviviparous species Rhinobatus granulatus or located caudad as in Raja rhinu (Prasad, 1948). The shellsecreting area of the nidamental gland also serves as a seminal vesicle in the dogfish (Metten, 1941) where sperm are stored to fertilize eggs before or at the time of shell formation, The wall of the uterus is smooth and covered with a 3attened epithelium in the oviparous elasmobranchs where it serves only as a passageway for the eggs. In viviparous species, this portion of the Miillerian duct is variously modified through the formation of villuslike appendages which nourish the developing young. These specializations are considered in the next section.

111. VIVIPARITY AND GESTATION

A. Evolutionary Considerations

If an animal species is to survive, each adult member of the population must on the average produce one reproductively active adult.

1. REPRODUCTION

21

Darwinian evolution rarely operates through one channel, and the fishes as a group have achieved reproductive success in many different ways. At one extreme, eggs and sperm are broadcast in sufficient numbers to balance the unusual pressures of the environment and satisfy the predators while, at another extreme, fertilization is internal and the developing young are housed within the parent's body until ready for an independent existence. The provision of millions of unprotected gametes represents the primitive condition among fishes, Along the road to specialized vviparity many curious and successful devices have evolved to provide a measure of protection during incubation. Parental care must confer great biological advantages in reducing energy demands for production of eggs or by spreading this load over a longer period while the young are developing within the parent. Thus, it is perhaps not surprising that viviparity has evolved independently several times in both the vertebrates and the invertebrates. Between those fishes which represent the primitive situation-scattering millions of unprotected gametes-and the highly specialized viviparous forms, there are numerous species which build nests and exercise parental care (Salmonidae and Gasterosteidae), other species in which incubation takes place either in the buccal (Cichlidae and Bagridae) or branchial chambers ( Amblyopsidae ) , fishes which provide special devices for the attachment of their young to their bodies (cutaneous incubation in the male Kurtus gulliveri or the females of Aspredo cotylephorus and Sobnostomus laciniatus), and the fascinating sea horses and pipefishes (Syngnathidae) in which the males have a ventral marsupial pouch where the young are incubated until ready for an independent existence. Bertin (1958b) has provided a summary with many more examples and illustrations of these interesting and curious devices. Although there is a rich literature on the natural history and functional morphology, there are very few physiological studies beyond those which have shown a strong dependence of secondary sex characters such as the marsupium on the gonadal steroids (Oguro, 1958; Noumura, 1959; Wai and Hoar, 1963). Among the fishes, only elasmobranchs and teleosts have achieved a true viviparity. In each of these groups there is an array of species from the ovoviviparous-where the eggs have sufficient yolk for the nourishment of the young and the female provides only protection-to the truly viviparous species-where the yolk content of the egg is greatly reduced and the developing young establish a connection with the maternal tissues at an early stage to draw nourishment from them and to satisfy the respiratory and excretory demands. The literature on viviparity in fishes has been reviewed several times

22

WILLIAM S . HOAR

(Hoar, 1957; Budker, 1958; Bertin, 1958c; L. H. Matthews, 1955; Amoroso, 1960). The following is a brief summary; since the phenomena are so very different in elasmobranchs and teleosts these two groups will be dealt with separately.

B. Viviparity among the Chondrichthyes The Chondrichthyes produce relatively few eggs which vary greatly in their content of organic matter but are all of the yolky telolecithal type (see Table 8 in Needham, 1942). In the truly oviparous groups (Scyllidae, Heterodontidae, Raiidae, and Chimaeridae) , ova are housed in horny protective shells which provide protection throughout a long period of incubation. In the ovoviviparous and viviparous species, the young develop within the Mullerian duct (uterus) on which they usually depend for nourishment as well as protection. Ranzi (1932, 1934) did much of the pioneer work and his papers should be consulted for details; Needham (1942) gives a valuable English summary and bibliography. Budker (1958) summarizes these important papers in French. 1. INTERNAL FERTILIZATION Fertilization is always internal among the Chondrichthyes. LeighSharpe in a series of classic papers (1920, 1921, 1922, 1924, 1925, 1926) described the modifications of the male pelvic fins which form the copulatory organs or claspers. These posterior extensions of the fins are stiffened not only by the cartilages of the metapterygia (Romer, 1955) but also during copulation by erectiIe tissue. The reIative contributions of cartilage and erectile tissues vary in different species. In all cases, an essential portion of the organ is the clasper groove formed by skin folds, the edges of which overlap to form a scroll-like tube along which the sperm are transported from the cloaca. Another constant feature of this apparatus is the presence of a clasper syphon or gland which contributes to the seminal discharge or provides a pumping mechanism for its release. In the sharks, the syphon takes the form of a blind muscular sac situated just under the skin anterior and lateral to the cloaca1 region; in the skates and rays a clasper gland takes the place of the hollow syphon. LeighSharpe concluded from his anatomical studies that the dilute fluid found in the syphon was mostly seawater and that during copulation the contractions of the muscular syphon wall pumped sperm through the clasper groove in a jet of seawater. It is by no means certain, however, that this is its main function. Botte et d. (1963) and La Marca (1964) among others have reported more recently on the physiology and histochemistry.

1.

REPRODUCTION

23

Mann (1960, 1964; Mann and Lutwak-Mann, 1963) has shown that the syphon of the spiny dogfish, Squalus acanthias, secretes an abundance of 5-hydroxytryptamine (serotonin) ; a pair of syphons in a mature male may contain as much as 20 mg of serotonin or 0.3% (Mann, 1960). The further demonstration of a marked stimulatory effect of serotonin on the isolated uterus of the spiny dogfish is strongly suggestive of a role for the syphon in the transport of sperm after transfer to the female (Mann and Prosser, 1963). At this stage, generalizations are not justified since only traces of serotonin have been found in the syphon sacs of the smooth dogfish, Mustelus canis, and none was demonstrated in the clasper glands of Torpedo and Raja. The clasper glands of skates and rays produce a milky white semiviscous fluid which coagulates on contact with seawater; its function remains speculative and, indeed, there are still many questions concerning the physiological mechanisms of internal fertilization in these fishes. 2. OVOVIVIPARITY The ovoviviparous species are far more numerous than the truly placental or viviparous ones; a true viviparity with yolk sac placenta is confined to certain species of two families of sharks-the Carcharhinidae and the Sphyrnidae, both belonging to the Galeiformes (Tortonese, 1950; Budker, 1958). Actually, the distinction between ovoviviparity and viviparity is a rather artificial one for the physiologist since the maternal contribution of the placental shark to the nourishment of the fetus is an intermediate one in a series ranging from almost zero in the primitive ovoviviparous species to an almost complete dependence in the highly complex ones (Fig. 7 ) . Moreover, in some sharks (Sphyrna tiburo and Mustelus canis) the placenta develops only after several months of an ovoviviparous existence ( Te Winkel, 1950; Schlernitzauer and Gilbert, 1986). These facts are recognized by Budker (1958) who divides the Chondrichthyes into only two major groups-the oviparous and the viviparous-with the latter subdivided into the aplacentals and the placentals. The rectangles to the left of the ordinate in Fig. 7 show the extent to which the newborn fish receives nourishment from its mother while in utero; the rectangles to the right show the content of organic substances in the uterine fluids; the black bars show the relative reductions in size of maternal liver during development. Although the presence of organic material in the uterus is universal, the organic content of the egg may be actually greater by 20-404: than the organic content of the animal at birth (Group I in Fig. 7); in short,

Ia . .

I b Rhinobotur pandumtur)

I

------

I

- - - - - - - - - -tt

Ptsmplotea micruru

?

’------

- - - - - - - - -F,-

Myliobafis bovine

I

Trygon violaceu I

r

I

I

I

I

100

10

1

0.1

0.01

0

I

I

1

1

0

1

1

10 20%

1

orgonic subsfonce

R-

in uterine

milk

I

i

10

100%

1. REPRODUCTION

25

the mother has provided protection but probably does not provide organic nutrients to the fetus. At the other extreme, eggs with very little organic material (only 200 mg in Pteroplatea micrura) depend almost entirely on the maternal secretions for growth; at birth, Pteroplatea contains about 10 g of organic material, an increase of almost 5000% (Needham, 1942). In contrast, the change in organic substance from egg to newborn pup is only 840 and 1050%in two placental speciesCarcharias glaucus and Mustelus laevis, respectively. Selected tables summarizing the biochemical data, largely the work of Ranzi, will be found in Needham (1942) and Amoroso (19csO). A host of research problems await the interested physiologist. It has now been almost 40 years since Ranzi’s classic work on the morphology and biochemistry; most of the endocrinological work does not go beyond correlations between the cytology of the endocrine glands and the cycle of morphological events during gestation. Histologically, the uterine lining varies from a mucus-secreting layer of cuboidal epithelial cells in species which depend on the egg yolk for nourishment (Type Ia in Fig. 7) through forms with moderately folded, serous-secreting linings, as in Torpedo, to those with uterine linings beset with villi or trophonemata of varying lengths and complexities and glandular surfaces which secrete an abundance of fat (8%of the organic substance in Trygon violacea) . Amoroso ( 1960) refers to all these uterine secretions as “uterine milk” and provides a summary of the composition in various species. Budker (1958) describes them as mucous, serous, and lipid. The yolk is mainly digested within the intestine of embryos which depend on it for nourishment. Te Winkel (1943) has described the process in the Squalidae. In Squalus acanthias, she found that the relatively enormous yolk sac established at an early stage in ontogeny decreased rapidly in size while an internaI yolk sac (expansion of the yolk stalk) became relatively larger until the external one remained as only a remnant. Yolk granules are moved by cilia from the external sac through the yolk stalk into the internal sac; then they pass on into the intestine where digestion occurs (Fig. 8). All of these structures, including the Fig. 7. Summary of the maternal-fetal relationships in the elasmobranch fishes (Ranzi, 1934). Rectangles to left of ordinate, extent to which the embryo receives organic substance from mother (negative sign indicates that this does not occur); rectangles to right of ordinate, content of organic substance in uterine fluid (dotted lines show probable values); black bars show degree to which maternal liver is reduced during development as measured by factor R ( Needham, 1942). Sketches of the uterine lining are shown at the right. Based on Ranzi (1934) and Needham (1942).

26

WILLIAM S. HOAR

Fig. 8. Dissection of a 220-mm embryo of Squalus suckleyi.

intestine, are ciliated. Enzymic activity is established in the gut when the embryo is about 60-70 mm long (size at birth about 250-300 mm). A small reserve of yolk still remains in the internal sac at birth. The skates and rays display a variety of specializations from those which are oviparous through a series of curious adaptations in ovovivi-

1.

REPRODUCTION

Fig. 9. Structures concerned with embryonic nutrition in Dasyatis violacea ( =Trygon uiohcea). ( A ) Villi of uterine wall, ( B ) diagram of circulation in uterine villus, and ( C ) and ( D ) cross sections of uterine villi (23 g embryo) taken through the apex ( C ) and base ( D ) of a villus. Bottom series of cells shows progressive stages in development of uterine gland cells with accumulation of fat droplets at right. Redrawn from Ranzi (1934) with the permission of the Stazione Zoologica di Napoli.

28

WILLIAM S. HOAR

parity. In the most highly specialized situations, nutrition by uterine milk or embryotrophe becomes more efficient than placentation in terms of organic material transferred from mother to fetus ( Amoroso, 1960). In the electric ray Torpedo the uterine secretions are taken directly into the digestive tract through the mouth and spiracles; while yolk digestion is going on in the intestine, the uterine secretions are being digested in the stomach, In one of the stingrays Pteroplatea, the highly glandular trophonemata or villi enter through the spiracles of the embryo and extend down into the esophagus to release the secretions directly into the gut. In another stingray (Trygon or Dasyatis) and in the eagleray Myliobatis the villi are shorter but extremely numerous; they produce a secretion rich in fat which is aspirated into the gut mostly through the spiracles (Alcock, 1892; Amoroso, 1960). The structure of the trophonemata of Trygon is shown in Fig. 9. Shann (1923) describes a very different mode of embryonic nutrition in the shark, Lamna cornubica. In this animal, the egg yolk is absorbed early in development and thereafter the developing pups depend almost entirely on the swallowing of immature eggs and degenerating ovarian tissues which pass down the oviduct packaged in a delicate secretion of the shell gland. As L. H, Matthews (1955) notes, some of these ovoviviparous arrangements are not very different from the oviparity found in the Holocephali, where only a small part of the yolk is enclosed in the yolk sac; the remainder breaks up into a thick milky fluid which is first absorbed by the filamentous external gills and later ingested through the mouth. 3. VIVIPARITY

Descriptions of the placentae of the silky shark, C u r c ~ r h i ~ u fa&s formis ( Gilbert and Schlernitzauer, 1966), and the bonnethead shark,

Sphyrna tiburo ( Schlernitzauer and Gilbert, 1966), have recently been added to the literature of true viviparity cited in Amoroso’s review ( 1960). The ova of Mustelus canis, on arriving in the uterus, become isolated into separate uterine compartments. These are formed by the growth of ridges on the uterine wall which eventually fuse so that each embryo is enclosed in a separate chamber (Fig. 10); although uterine compartments are not formed in most of the nonplacental sharks there are at least two exceptions ( Gabus canis and Mustelus vulgaris) cited by Ranzi (1934). Mustelus canis (also Sphyrna tiburo) leads an ovoviviparous existence during this early period and may not establish a placenta for several months. During this time, while nutrients are obtained from the yolk, these embryos develop elaborate circulatory networks in the walls

Fig. 10. Yolk sac placenta of Mustelus cunis ( =be&). ( A ) Diagram of ventral view of right oviduct and uterus to show orientation of egg cases and embryos, ( B ) section through placenta with detailed portion at lower right, and ( C ) transverse section through umbilical cord of 18 g embryo. Upper diagram after Te Winkel (1950); others redrawn from Ranzi (1934) with the permission of the Stazione Zoologica di Napoli.

30

WILLIAM S. HOAR

of their yolk sacs and the yolk sacs themselves become greatly folded. While these developments are taking place in the yolk sac, the uterine mucosa which is initially smooth, loose, edematous, and covered with a columnar or cuboidal epithelium also becomes folded. A placenta is established through the interdigitation of these two series of folds with a thinning of their epithelia to bring the maternal and fetal circulations into close proximity ( Fig. 10). The umbilical cord or stalk which attaches the embryo to the placenta is a modified yolk stalk. The interdigitations characteristic of the placentae of the two fishes just described do not seem to develop in C~rcharhinus( Mahadeven, 1940; Schlernitzauer and Gilbert, 1966). The specialized yolk sac sits on a modified and extremely vascular discoidal patch of the uterus. Detailed descriptions of the umbilical cords will be found in the literature cited. An interesting specialization found in some forms ( Paragaleus, Scoliodon, and Sphyma) is the presence of numerous villi (appendiculae) which appear to be absorptive (see Fig. 1285 in Budker, 1958); these suggest that the dependence on uterine milk was not completely lost in the evolution of viviparity among the sharks. In this and in several other features, it is clearly indicated that the phylogeny of viviparity in these fishes was via a highly specialized ovoviviparity and that the steps involved were not very long ones. Both conditions may be markedly developed within a single genus; Mustelus vulgaris is ovoviviparous while Mustelus canis is viviparous.

C. Viviparity among the Teleosts

Development of the young within the female occurs in only two orders of teleost fishes, the Cyprinodontiformes and the Perciformes. The nine families involved represent only a small proportion of the Teleostei. However, the diversity of mechanisms is as great as that of any other group of vertebrates which range from the oviparous, through several stages of ovoviviparity to highly successful viviparous forms. This, as well as a lack of correlation with particular habitats or geographical regions, argues for a separate evolution on more than one occasion. The viviparous teleosts include such geographically remote groups as the comephorids unique to Lake Baikal in Siberia and the goodeid fishes of the Mexican plateau (C. L. Turner, 1947); both marine and freshwater species are represented. Amoroso’s review (1960) is the most informative of the recent ones and includes citations to the papers of C. L. Turner which are still the richest source of information on these fishes. Amoroso’s summary (1960) is reproduced in Table I and lists the known families of viviparous

31

1. REPRODUCTION

teleosts (see also Breder and Rosen, 1966). It will be noted that internal development in the teleosts always takes place in the ovary; the Mullerian duct, which houses the developing elasmobranch, is absent in the teleosts (Fig. 3 ) . Sometimes the young develop within the ovarian follicle to a very advanced stage before ovulation (Poeciliidae and Anablepidae) ; more frequently, embryogenesis occurs within the cavity of the ovary after fertilization within the follicle. Although one might speculate that the ovarian type of gestation is more primitive, C. L. Turner (1947) believes that with one probable exception (Zoarces) fertilization precedes ovulation in present-day viviparous teleosts. Bertin ( 1 9 5 8 ~ )summarizes the three general situations as follows where “0” represents ovulation, “ F is fertilization, “H” is hatching, and “ P is parturition: Type Zoarces Type Jenynsia Type Gambusia

0-F . . . . . . . . . . . . . . . . .H ...................... P F . . . . . . . .O. ..................... H . . . . . . . . . . P F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0- H-P

Both follicular and ovarian gestation are characterized by ovoviviparous and viviparous types of development. As in the elasmobranchs, the total dry weight of the larva at birth may be less than that of the egg, indicating the complete lack of a maternal contribution to the nourishment of the embryo while in the more specialized viviparous types the egg is endowed with very little yolk and the developing embryo obtains almost all of its nourishment from its parent (Fig. 11). Belonesox be1iza nus

0.018

0.016 0.014

; g 0.012

.,

Aulophallus elongatus

c

-

0.01 ’

g

0.008.

.-

r

0

0.006

0.002

Mollienesia velifera Gornbusia nobilis Sebastes marinus

Fig. 11. Changes in the dry weight (including yolk) of several viviparous fishes during their development. From Scrimshaw (1945).

Table I Summary of Modes of Development in Teleosta with Internal Fertilizationa Order, family, or species Siluridae Trachychoristes Perciformesb Scorpaenidae Sebastodes

Zoarcidae (Zoarcea

Type

Gestation

Fetal nourishment and placental structures

Corpora lutes ?

Oviparous

Ovarian cavity Eggs immediately discharged Yolk in yolk sac and from or gonaduct into water aquatic environment

Oviparous

Follicle zygote Brood remain in ovarian retained for cavity for short period; long period thereafter development oviparous

Yolk in yolk sac and from aquatic environment

Present

Viviparous

Ovarian cavity Retained in ovarian cavity until birth

Very little yolk; ovary furnishes greater part of food; some embryotrophe from dead embryos; no supplementary structures; absorption through skin

Present

Viviparous

Follicle

Secretions; embryotrophe Present from desquamated cells; sperm, dead embryos; absorption through hypertrophied yolk sac. S u p plementary structures for respiration in form of V ~ S cular spatulate processev on fins

viviparus)

Embiotocidae (Cymatogaster aggregatus)

Fertilization

Hatched into ovarian cavity before segmentation mm. pleted. Retained for very long period; embryos born in very advanced state. Males sexually mature a t birth

Cyprinodontes Poeciliidae (Xiphophoriis helleri)

Oviparous

Follicle

Follicular for entire gestation; high degree of superfetsr tion; embryos very immature at birth; gonads in indifferent stage

Yolk provides most of the intraovarian food, not completely absorbed a t birth; “neck” strap forms s u p plementrtry structure for absorption of secretions

Em

I

Poeciliidae (Heferandria formosa)

Viviparous

Follicle

Follicular for entire gestation; high degree of superfetatioii

Ovarian secretions; “neck” strap and follicular pseudoplacen ta

Anablepidae ( A nahleps anahleps)

Viviparous

Follicle

Follicular for most of gestation; short time in ovarian cavity

Ovarian secretions; yolk sac hypertrophied and is principal organ of absorption; enlargement of gut form intestinal pseudoplacenta

Goodeidae Ovoviviparous with Follicle (Ataenohius superimposed tuweri) viviparity (Goodea hilineata) (Zoogonecticus cuitzooensis) (Lermicthys multiradiatus)

Ovarian cavity for entire gestation ; brood in advanced state of development a t birth; high degree of mortality

Ovarian secretions; embryoPresent trophe from stored sperm and dead embryos; Trophotaeniae (fetal) absent in Ataenobius toweri, poorly developed and voluminous in Zoogonecticus cuitzooensis and Lermichthys multiradiatus

Jeny nsiidae

Ovarian cavity for entire gestation; brood in advanced state of develop ment at birth; high degree of mortality

As for Goodeidae but ovarian Present fluids imbibed directly by mouth; trophonemata (maternal) in the form of ovarian flaps constitute a branchial pseudoplacenta

a

b

Ovoviviparous with Follicle superimposed viviparity

From Amoroso (1960). To this list of perciform fishes Breder and Rosen (1966) add certain species of the Clinidae and the Comephoridae.

?

Present

z“

34

WILLIAM S. HOAR

Actually, most of the teleosts which develop within the ovary seem to draw nourishment from the mother and in this sense are not truly ovoviviparous. As indicated in Fig. 11, only Sebastes marinus shows a decline in dry weight between fertilization of the egg and parturition; most of the species show little change in weight indicating that the mother has met the metabolic costs of respiration in the developing tissues so that the total weight of the larva at birth is just balanced by the stored material in the original egg. Only two of the poeciliids (Heterandriu formosa and Aulophallus elongatus) studied by Scrimshaw (1945) are truly viviparous, depending on the mother for both their respiratory needs and the anabolic demands for growth. The weight decrease of 34% in Sebastes marinus is essentially the same as the value 37%reported for the oviparous trout Salmo fario ( Scrimshaw, 1945). 1. INTERNAL FERTILIZATION As in the Chondrichthyes, internal fertilization in the teleosts is not necessarily followed by a development of the young in the female. One of the most complex intromittent organs described in male teleosts (the priapium located on the throat of the Phallostethidae) serves only to fertilize the eggs as they are being laid (Bailey, 1936; Breder and Rosen, 1966). In teleosts, the copulatory organ is usually an enlarged genital papilla or a specialized anal fin. Unlike the Chondrichthyes, the pelvic fins are rarely used directly in the transfer of sperm (Bertin, 1958~);the priapium of the phallostethid fishes is again an exception with skeletal elements derived from pelvic fins (Bailey, 1936). An enlarged genital papilla is sometimes referred to as a “pseudopenis.” Its size bears little relation to viviparity. In the Embiotocidae, a highly successful viviparous group, it appears only seasonally as a modest fleshy papilla; in the Cottidae, an oviparous group, it is often greatly enlarged and may contain erectile tissue (Weisel, 1949). A pseudopenis is found in several other families both oviparous and viviparous (Bertin, 1 9 5 8 ~ ) There . are many strange modifications of the genital papillae of both sexes. In female Orthonopias triacis the protrusible oviduct is smeared with sperm during copulation and then withdrawn to effect fertiIization of the eggs which are laid one at a time ( BoIin, 1941). In Apogon imberbis the urogenital papilla of the female is enlarged and introduced into the male for the reception of sperm (Garnaud, 1950). As noted earlier, the elongated genital papilla of Rhodeus amurus transfers eggs to the gill spaces of a freshwater mussel. These enlarged genital papillae-both male and female-appear as secondary sex characters

1. REPRODUCTION

35

during the breeding season, and their development is probably always regulated by the gonadal steroids. The copulatory organs of the cyprinodonts are specialized anal fins. The fins are only slightly altered in the Goodeidae with a foreshortening of the anterior rays to form a spermatopodium (Langer, 1913; Mohsen, 1961). In contrast, the anal fin of the Poeciliidae, referred to as the “gonopodium,” is profoundly modified through an elongation and specialization of several of its rays; the genital aperture is at its base, but during sexual activity the elongated bony fin rays and the web of tissue connecting these rays interact to form a transitory groove presumably associated with the transfer of sperm bundles to the tip of the gonopodium ( Rosen and Gordon, 1953; Rosen and Bailey, 1963) . The Jenynsiidae and the Anablepidae show a third stage in anal fin specialization to form a large penis containing a permanent tube opening at its tip; at times the organ may be greatly extended; it is a naked structure in the Jenynsiidae but covered with scales in the Anablepidae (Bertin, 1958~). 2. GESTATION WITHIN THE OVARIAN CAVITY

Sebastodes paucispinis (family Scorpaenidae ) undergoes an entirely ovoviviparous development within the ovarian cavity. Moser ( 1967a), who has recently studied this species, found numerous sperm singly or in clumps embedded among the epithelial cells of the maturing follicles or on their outer surfaces but never within the follicles; he believes that fertilization occurs immediately after ovulation and this would place Sebastodes in the “Zoarces type” of Bertin’s classification. Sebastodes embryos lack placentalike connections and depend entirely on their egg yolk for nourishment; they hatch just prior to spawning when the yolk has been largely absorbed and they are about 4-5 mm long; life in the ocean commences as a planktonic larva. Large females may produce broods of as many as two million larvae; the respiratory demands of this large mass of developing tissue must create one of the major problems in maternal physiology. A specialized dual vascular system to the ovary seems to be unique to this group of fishes and is evidently related to these special needs (Moser, 196%). The Embiotocidae show an intermediate situation between the strict ovoviviparity of Sebastodes and the goodeid or jenynsiid fishes which usually form elaborate placentalike connections. Cymtogaster aggregata has been carefully investigated by C. L. Turner (1938a) and Wiebe (1968b,c). Copulation occurs during the spring or early summer and sperm remain dormant in the ovary until fertilization occurs about 6 months later. Eggs, fertilized within the follicle, are shed into the cavity

36

WILLIAM S. HOAR

of the ovary during early segmentation stages and develop there for 10-12 months to an advanced stage (sexual maturity in some males: C. L. Turner, 1938a). The Cymatogaster egg contains relatively little yolk and depends mainly on the secretions of the highly glandular ovigerous folds for nourishment (Fig. 6 ) . By the time the larvae have reached a length of 15 mm, spatulate vascular extensions have developed on the vertical fins (Fig, 6 ) to assist in nutrition and respiration; these are resorbed just before birth. The alimentary canal begins to function later in gestation and long villi located in the gut are evidently concerned with digestion and absorption of foods entering through the mouth (C. L. Turner, 1938a); relationships are also established between the ovarian tissues and the embryonic gills (C. L. Turner, 1952). More elaborate types of placental connections are characteristic of both the Goodeidae (C. L. Turner, 1937a, 1940a) and the Jenynsiidae (C. L. Turner, 1940b). Turner's investigations of the goodeids have revealed a spectrum of species with progressively diminishing yolk stores, a less significant role for the yolk sac and pericardial sac (mentioned below) and an increasing importance of highly vascular rectal trophotaenia of many diverse forms (Fig. 12 and C. L. Turner, 1937a). These are bathed in the copious secretions of the ovarian cavity; the ovarian

Fig. 12. The trophotaeniae of Zoogoneticus quitzeoensis during late gestation with a segment of a longitudinal section through one of the nutritive processes at the lower left. Redrawn from C. L. Turner ( 1933, 1940a).

1. REPRODUCTION

37

secretions are supplemented by degenerating embryos. The brotulid fishes have similar absorptive surfaces (C. L. Turner, 1936). In the jenynsiids there is also an early ovulation following fertilization, an initial dependence on the small supply of yolk, the development of a yolk sac and a pericardial sac. For the major portion of intraovarian development, however, an intimate association is established between the ovarian tissues and the pharyngeal and mouth cavities of the embryo; vascular folds of the ovarian epithelium are in close contact with the gills-an arrangement which Amoroso (1960) refers to as a “potential branchial placenta.” These fishes also imbibe ovarian fluids as a source of nourishment (C. L. Turner, 1940b). In several of the papers cited above, Turner discusses the progressive specialization in ovarian gestation from the strictly ovoviviparous to these several curious placental connections and notes that this trend is evident in a single family-the Goodeidae.

3. FOLLICULAR GESTATION

A prolonged follicular gestation is found only in the Poeciliidae and the Anablepidae. Again there is a series of forms from the ovoviviparous ( Poecilia ) to species with a specialized follicular pseudoplacenta; the complete array is evident within the poeciliids which remain in the follicle for the entire period of gestation (C. L. Turner, 1 9 4 0 ~ ) . The yolk supply is adequate for embryonic nutrition in several of the familiar poeciliids ( guppy, black molly, and swordtail) and respiratory exchanges are effected through an expanded yolk sac which extends over the head as a “neck strap.” Although it is generally assumed that the intimate association of the portal blood system on the yolk sac and the vascular wall of the ovarian follicles serves only for the exchange of gases and nitrogenous wastes, a transfer of nutrients has not been ruled out. In Heterandria formosa the yolk supply is meager and the yolk sac very small; the fetal-maternal exchanges are effected through a greatly expanded and vascular pericardial sac which is wrapped around the anterior part of the embryo; the wall of the follicle is smooth and makes intimate contact with the portal system of the embryo. A further stage of specialization found in this group is an enlarged ventral expansion of the coelomic cavity to form a “belly sac” (only modestly developed in Heterandria) and the formation of finger-like villi on the follicular wall (Fig. 13). Turner applies the term “follicular pseudo-placenta” to this complex of follicular wall, follicular space, and the portal system of the belly sac.

38

WILLIAM S. HOAR

Fig. 13. The embryonic-maternal connections in several poeciliids. ( A ) The portal circulation of the yolk sac of 7.5-mm embryo of Lebistes (=Poecilia) reticuhtus, ( B ) Poeciliopsis embryo showing the portal network of the belly sac (pericardial sac anteriorly and expanded coelom posteriorly), ( C ) low branched villi on the internal surface of the follicle of Poeciliopsis, and ( D ) section of follicle and adjacent Poeciliopsis embryo. Redrawn from C. L. Turner ( 1 9 4 0 ~ ) .

The anablepid fishes show still another stage in the specialization of the follicular pseudoplacenta. Not only is the follicular wall clothed with complex villi but also numerous vascular bulbs appear in the portal circulation of the belly sac (Fig. 14). In addition, there is a spectacular enlargement of the gut in Anabkps (midgut in A. anableps and posterior gut in A. doweri), and this is evidently concerned with the digestion of follicular fluids taken into the alimentary canal (Fig. 14); an involution of this specialized area of the gut occurs before birth. The enlargement of the gut and the presence of vascular bulbs is unique to the Anablepidae (C. L. Turner, 1940d). A multitude of fascinating problems awaits the physiologist who becomes interested in the viviparous fishes. Although the functional morphology has been systematically investigated there are relatively few

1.

REPRODUCTION

39

studies of the physiology. Nutritional demands of elasmobranch embryos were investigated by Ranzi (1934) many years ago and there is a small amount of work on their nitrogen metabolism (Read, 1968); there are also a few studies of the nutrition of developing teleosts (Scrimshaw, 1944, 1945). However, to my knowIedge, there are no carefuI investiga-

Fig. 14. Expanded midgut and maternal-fetal connections in Anableps anableps. ( A ) Dissection of a 21-mm embryo. Redrawn from C. L. Turner (194Od). ( B ) Section of the wall of a follicfe containing a 21-mm embryo. Redrawn from C. L. Turner ( 193813).

40

WILLIAM S. HOAR

tions of the gaseous exchanges, the properties of maternal and fetal blood, or the ultrastructure of the maternal-fetal connections. 4. SUPERFETATION

In some of the poeciliids, several broods of young may be developing simultaneously within the ovary ( superfetation) . C. L. Turner ( 1947) suggests that the condition has evolved through a shortening of the period required for the development and maturation of the ova; a second group of eggs reaches maturity and is fertilized before the previous group completes development. It must also be related to storage of sperm within the ovary. In many fishes the ovarian epithelium assumes a nurse-cell function and immobilized sperm, embedded in its epithelium, remain viable for many months. Superfetation seems to reach a climax in Heterandria formosa where live sperm have been found 10 months after a single contact with the male and as many as nine broods may be developing in the ovary simultaneously with births at intervals of about 10 days (C. L. Turner, 1937b).

IV. THE ENDOCRINOLOGY OF REPRODUCTION

The timing of reproduction, regulation of the associated morphological changes, the mobilization of energy reserves for gonadal development, and intricate breeding behavior are very largely dependent on the glands of internal secretion. Most of the endocrine organs are either directly or indirectly involved since this complexity requires profound metabolic adjustments, associated not only with gonadal development but sometimes also with changes in habitat during the breeding season. Only the pituitary gonadotropic hormones and the gonadal steroids are considered in this chapter. Chapters associated with the other endocrine glands contain pertinent references to their role in reproduction; the endocrinology of reproductive behavior is considered separately in the chapter by Liley, this volume, The literature on fish endocrinology has been reviewed several times during the past decade (Pickford and Atz, 1957; Vivien, 1958; Bern and Nandi, 1964; Hoar, 1966; Matty, 1966), while the reproductive hormones of fishes have been separately considered by Dodd (1960a) and Hoar (1965a). The bibliographies of these reviews are a rich source of the earlier literature; the citations included here have been kept to a minimum.

1. REPRODUCTION

41

A. The Pituitary Gonadotropins 1. PHYLOGENY OF GONADOTROPIC REGULATION-STUDIES IN THE AGNATHA It is now evident that the pituitary gland regulates certain aspects of reproduction in all vertebrate animals. It is equally true that this control is sometimes less precise and embraces fewer elements of reproduction among the fishes and, further, that there may be marked variations among the different groups of fishes. Pickford and Atz (1957) have tabulated the early experiments involving hypophysectomy of fishes. By the midthirties it was evident that gonadal development and the maturation of the sexual products were dependent on the pituitary gland in all gnathostomes. Investigation of the gonadotropic activities of the pituitary in the Agnatha awaited the investigations of J. M. Dodd and his associates in the early sixties. Their studies and investigations of Larsen (1965, 1969) have clearly shown pituitary gonadotropic controls in the cyclostome Lampetra fluuiatilis, one of the representatives of the most primitive living vertebrates. It would seem that the pituitary has dominated vertebrate reproductive processes since the Ordovician, a period of almost 500 million years. Dodds studies, however, went further and revealed a less complete dominance among the Agnatha suggesting a certain phylogeny in the establishment of this physiological regulation. Lampetra fluuiatilis, hypophysectomized in the late autumn and winter, were followed for a period of 5 months and the changes in the gonads carefully described (Dodd et al., 1960; Evennett and Dodd, 1963). At the beginning of the experiment, secondary sex characters were absent, the ovaries contained only small eggs, and the testicular ampullae a high percentage of spermatocytes. During the 5-month period the normal and sham-operated animals matured rapidly, secondary sex characters appeared, the ova increased in size, the testicular ampullae became filled with spermatozoa, and spawning took place in early April. On the contrary, the hypophysectomized animals remained immature in appearance while gonadal development was completely arrested in the females and markedly delayed in the males. Larsen (1965,1969) finds continued relatively slow growth of the ovaries after hypophysectomy but confirms Dodd's findings in the male. The effects on the gonads are particularly interesting. In the gnathostomes, hypophysectomy is followed not only by an arrest of ovarian development but also by a general atresia of the follicles (corpora lutea formation) while spermatogenesis is completely blocked in the testis.

42

WILLIAM S. HOAR

In the lamprey, spermatogenesis, spermiogenesis, and ovarian growth seem to be autonomous processes; they are retarded but not suppressed in the absence of a pituitary. The endocrine tissues of the gonads are evidently pituitary-regulated since the secondary sex characters which are thought to depend on them (Matty, 1966) fail to develop in the operated animals. Dodd et al. (1960) and Larsen’s investigations (1969) of the adult Lampetra fluuiatilis are the only long-term studies of the effects of hypophysectomy in the Agnatha. They have established the indispensable nature of the pituitary for reproduction. The Myxinidae have not yet been critically investigated in this connection (Adam, 1963); these primitive animals should prove particularly interesting and may well increase our understanding of the phylogeny of gonadotropic controls among the vertebrates. On the basis of the lamprey work, it has been suggested that in phylogeny the metabolic controls involving yolk mobilization and gonadal growth preceded those concerned with gametogenesis ( Hoar, 196513). 2. CHEMICAL INHIBITION OF THE ACTIONOF GONADOTROPINS Investigations of the pituitary functions of fishes have sometimes been hampered by the technical dSculties of hypophysectomy. This is true of some of the teleosts which are particularly popular with fish physiologists (Gasterosteus and Cymatogaster, for example); the mouth and opercular openings may be very small or the skull extremely deep dorsoventrally with much vascular tissue between the roof of the mouth and the brain or the fish may be difficult to handle postoperatively. Hoar et at. (1967) and Wiebe (1968a) have used Methallibure as a chemical blocking agent of gonadotropic functions. This agent, prepared by Imperial Chemical Industries ( 33,828), is l~-methylallylthiocarbamoyl-2methylthiocarbamyl hydrazine. It has been carefully tested in the homeotherms and is now finding a use in agricultural practice. We have examined the reactions of four fishes treated with Methallibure either by injection or by addition to the ambient water. It appears to be particularly effective in Cymatogaster aggregata ( Wiebe, 1968a), and Poecilia reticulata (Pandey, 1969d), but much less active in Gasterosteus aculeatus; the work with the goldfish, Carassius auratus, is preliminary. At present it is not known whether the variable response is because of species differences in reaction to the substance or species differences in the physiological actions of the gonadotropins; it is also possible that the effect varies with the environment-sea water vs. fresh water. It is of interest that some species of domestic animals are proving

1. REPRODUCTION

43

more responsive than others; the substance has been particularly effective in regulating the farrowing of pigs. In spite of the variations in response, all species of fish studied show clear evidence of the blocking of gonadotropic action in a manner comparable to that which follows surgical hypophysectomy. Spermatogenesis is either completely suspended or greatly depressed; yolk deposition ceases, and, in the more responsive species, steroidogenesis comes to an end. The side effects of the compound appear to be minimal involving a slight stimulation of the pituitary thyrotrophs (because of a mild thiourea action) and some depression of the somatotrophs (Wiebe, 1967; Leatherland and Pandey, 1969) . In histochemical studies of Foecilia, Leatherland and Pandey (1969) have shown that the most likely locus for the block is in the synthesis of the gonadotropins; there was a sharp and significant decrease in both number and size of these cells and no evidence of suppression of secretory activity in the neurosecretory system of the hypothalamus. Ovine LH was found to partially counteract the inhibitory effects of Methallibure in the gonads of Cymutogastet aggregata ( Wiebe, 1969b). 3. THE PITUITARY-GONADAL RELATIONS IN GNATHOSTOME FISHES

Pituitary-gonadal relations have now been investigated in several elasmobranchs and many teleosts. In general, the findings are consistent; the pituitary regulates both gametogenesis and steroidogenesis. In the hypophysectomized male dogfish, Sc yliorhinus caniculus, Dodd and his associates (1960) found that the effects of hypophysectomy were localized in the transitional zone between spermatogonia and spermatocytes. The zonated structure of the dogfish testis (Fig. 4) was particularly helpful in localizing these effects. It is of considerable interest that the seasonal cycle-presumably regulated by the pituitarycreates an identical picture ( Fig. 15). Spermatocyte production ceases during the springtime or following hypophysectomy, but in both cases spermatocytes already formed continue to differentiate and become mature spermatozoa (Simpson and Wardle, 1967). Dodd has reached the tentative conclusion that the gonadotropin( s ) are essential for the normal transformation which occurs when the spermatogonia (which divide by mitosis) become primary spermatocytes (which divide by meiosis). The testicular changes are well marked in dogfishes 6 weeks after h ypophysectomy. Removal of the ventral lobe of the female dogfish pituitary led to cessation of egg laying within 10 days to be followed somewhat more

44

WILLIAM S. HOAR

Fig. 15. Cycle of spermatogenesis in the dogfish, Squalus acanthius. The stages of gamete development are pictured in the circles representing ampullae. The four columns dividing the testis indicate four periods of the year. The filled circles indicate the position of the degenerate band, and they coincide with the absence of one stage and reduction of the neighboring stages of ampullae. Note the accumulation of evacuated ampullae to a maximum at the mating time ( December-February) followed by a resting phase when breakdown of spermatogonia results in a new degenerate band ( March-May). The rate of development from spermatocyte to spermatozoon is clearly seen using the degenerate band as a marker (7-9 months). 1, Epigonal tissue; 2, ampulla with sperm bundles and no Sertoli bodies; 3, origin of the degenerate band from breakdown of spermatogonia; 4, evacuated ampulla with Sertoli cell nuclei; 5, ampulla with ripe sperm and Sertoli bodies; 6, remains of degenerate band; 7, spermatid metamorphosis; 8, spermatids; 9, spermatocytes; 10, Sertoli cell nucleus; 11, spermatogonia; and 12, germ ridge. From Simpson and Wardle ( 1967 ) .

1. REPRODUCTION

45

slowly by distinct histological changes in the ovaries. Vitellogenesis ceases and all eggs larger than 4 mm become atretic; these changes are evident 3 weeks after hypothysectomy, but it requires about 14 months for the disappearance of all yolk and the reduction of the gonad to the juvenile condition. This atresia which follows hypophysectomy was first noted many years ago (Pickford and Atz, 1957); Hisaw and Hisaw (1959) and Chieffi (1962) have described it in some detail. Dodd and his associates have not observed any changes in the oviducts or secondary sex characters of their hypophysectomized animals. It is assumed but not yet proved that the elasmobranch pituitary like that of the higher vertebrates regulates the steroid-producing endocrine tissues of the gonads. In 1957, Pickford and Atz tabulated 17 species of teleosts and one lungfish known to have been successfully hypophysectomized at that time. This numbcr has increased substantially during the past decade; following are more recent additional species hypophysectomized in studies of pituitary-gonadal relationships: Mollienesia latipinna ( Ball, 1962), Pleuronectes platessa ( Barr, 1963a,b,c), Ophicephalus punctatus ( Belsare, 1965), Couesius plumbeus ( Ahsan, 1966b), Heteropneustes fossilis ( Sundararaj and Nayyar, 1967; Sundararaj and Goswami, 1968), and Poecilia reticulata-both adults and juveniles ( Liley, 1968; Pandey, 1969a,c). Of the species listed earlier by Pickford and Atz, more recent studies of Fundulus heteroclitus (Lofts et al., 1966) and, particularly those of Yamazaki on Carassiw auratus are of special significance in the literature on fish gonadotropins (Yamazaki, 1961, 1962, 1965; Yamazaki and Donaldson, 1968a,b). Wiebe’s work on the pharmacologically hypophysectomized surfperch, Cymutogaster aggregnta, also forms a part of this literature. All investigators agree that the pituitary is required for gonadal maturation; in its absence vitellogenesis is suppressed with atresia of the larger developing oocytes, spermatogenesis is blocked a t the spermatogonia-spermatocyte stage, and steroidogenesis does not occur in the gonadal endocrine tissues. Investigations of the endocrine tissues have been almost entirely confined to the testis. Although these general findings appear to be consistent, there are still many unresolved details. As yet, it is not certain whether gonadotropin is required for ( a ) multiplication of the spermatogonia, ( b ) specifically triggering the reduction divisions, and ( c ) spermiation or ovulation, nor is it certain to what degree the Sertoli cells and other supporting tissues in the gonads are dependent on the pituitary. Barr’s work (1963~)suggests that some of the variations recorded by different workers may have a seasonal basis and depend on the variable stages of gonadal maturation at the time of hypophysectomy. The later stages

46

WILLIAM S. HOAR

of spermatogenesis and spermiogenesis, if well under way, seem to continue with production of mature sperm but spermiation is usually not observed in the absence of the pituitary (Yamazaki and Donaldson, 1968a; Pandey, 1969a) and the mature sperm gradually disappear through phagocytosis or resorption. Spermiation has been reported in the hypophysectomized plaice ( Barr, 1963c) and lake chub ( Ahsan, 1966b); these differences may depend on the stage of sexual maturity at the time of surgery. It has already been noted (Section 11, B, 3 ) that the pituitary normally triggers the secretions which are essential in thinning the seminal fluid prior to its discharge. Pandey’s work (1969a) shows that in the absence of the pituitary spermatophores form from the later spermatocytes, but they rupture and are resorbed without a discharge of the sperm. Pandey’s work ( 1 9 6 9 ~ on ) the juvenile guppy has also emphasized the interplay of both pituitary gonadotropins and androgens in the maintenance of spermatogenesis. The changes in ovarian histology which follow hypophysectomy have been much less frequently examined. However, the findings are consistent with those recorded for the elasmobranchs and indicate that after hypophysectomy a slow but inevitable resorption of ova may be expected; vitellogenesis comes to an end; new ova fail to form; and ovulation and spawning do not occur unless yolk deposition was complete prior to the surgery. Yamazaki (1965) found that relatively few goldfish ovulated even though the operation was performed only a few hours prior to the normal ovulation time. A major component of the literature on fish endocrinology is concerned with the practical matter of regulating ovulation in fish culture operations. The work primarily involves the use of mammalian pituitary or chorionic gonadotropic preparations. Many citations will be found in reviews by Pickford and Atz (1957), Ball (1960), and Dodd (1960a). This topic is also discussed in the chapter by Liley, this volume. The activity of some of the preparations commonly used will be considered in the next section. The endocrine-producing tissues of the gonads seem to be pituitary regulated in all of the vertebrates, and regressive changes may be expected in them after hypophysectomy. Atrophy of the testicular interstitial tissues has now been recorded in several teleost fishes (BuserLahaye, 1953; Lofts et al., 1966; Sundararaj and Nayyar, 1967; Pandey, 1969a) and in lobule boundary cells (Ahsan, 1966b). Regressive changes in the Sertoli cells have also been noted (Pandey, 1969a). The gonadotropic inhibitor Methallibure causes atrophy of both the interstitial cells and the Sertoli cells in the surfperch Cymatogaster; both of these groups of cells are actively concerned with steroidogenesis in this species

1. REPRODUCTION

47

(Wicbe, 1969a). All thcse data are consistent with the concept of a pituitary control of steroidogenesis in the gonadal endocrine tissues. Regressive changes noted in the efferent duct system (Lofts et al., 1966; Pandey, 1969a) are probably secondary to the degeneration of the steroidproducing tissues ( Wiebe, 196813). 4. EFFECTSOF HYPOPHYSECTOMY ON GESTATION Several workers have associated cytological cycles in pituitary secretory activity with the pregnancy cycle of viviparous fishes. Ranzi (1936, 1937) examined the endocrine glands in a part of his extensive studies of selachian viviparity and described pituitary hypertrophy and hyperemia with a decreased acidophilia in pregnant females. Chieffi (1961, 1967) and Della Corte and Chieffi (1961) cite confirmatory recent evidence for elasmobranchs. Likewise, pituitary changes have been associated with the reproductive cycles of viviparous teleosts (Stolk, 1951a). However, in both groups of fishes these pituitary cycles may only regulate the timing of reproduction and the development of the gametes as they do in the oviparous species. From present studies it must be concluded that the pituitary plays little or no part in the maintenance of pregnancy in the elasmobranchs but may be involved in some teleosts (Chester Jones and Ball, 1962; Hoar, 1965a, 1966). However, there is still very little experimental work directed to relatively few species. In the late thirties, Hisaw and Abramowitz (1938, 1939) reported that pregnancy was uninterrupted by hypophysectomy in the viviparous dogfish, Mustelus canis; animals hypophysectomized in early stages of pregnancy showed normal development of young for a period of 3% months (Hisaw, 1959). These seem to be the only experiments of this nature recorded for the elasmobranchs, although Dodd et d. (1960) report that removal of the ventral lobe of the pituitary of Scyliorhinus has no effect on the oviducts which retain their normal secretory capacities. Studies of the effects of hypophysectomy on gestation in the viviparous fishes are likewise preliminary. Ball (1962) found that 12 MoZZienesia latipinna, hypophysectomized during pregnancy, gave birth to normal off spring; the embryos completed their development in the usual time and were as lively and survived as well as the controls. Chambolle (1964)) however, working with another viviparous cyrinodont Gambusia, recorded a highly significant mortality of embryos when females were hypophysectomized during the first days of pregnancy; in part, the mortality was thought to be related to a disturbance in water and electrolyte metabolism-perhaps associated with the loss of the pituitary-adrenal

48

WILLIAM S. HOAR

control, Thus, the evidence from two different species of viviparous cyprinodonts is contradictory but there are further indications from other studies that the pituitary is essential to successful gestation and parturition in some teleosts. Wiebe (1967) found that treatment with the gonadotropin-blocking agent Methallibure leads to an atrophy of the secretory ovigerous epithelium in Cymatogaster. Associated with this, Wiebe noted reduced growth and a 3% mortality of the intraovarian embryos. These results are preliminary and based on only three pregnant females; the effects may also be indirect ones mediated through the gonadal steroids. They are, however, indicative. In addition, there are several reports (reviewed by Chester Jones and Ball, 1962) of premature release of young by pregnant cyprinodonts following injection of mammalian pituitary preparations. It is also pertinent that incubation and parturition of developing young in the male sea horse, Hippocampus hippocampus, is disturbed by hypophysectomy although gonadectomy is without effect (Boisseau, 1964). Finally, there is some suggestive evidence of a role for prolactin in the gestation of certain cyprinodonts and an intimation that the neurohypophysial hormones are active in parturition (Egami and Ishii, 1962; Ishii, 1963). Prolactin is considered in the chapter by Ball, Volume 11, and the neurohypophysial factors in the chapter by Perks, Volume 11. It is obvious that further research is required in this area of endocrinology. It may yet be necessary to modify Hisaw’s conclusions (1959)-based on only one species-that the responsibilities of the fish pituitary end with ovulation and that an involvement in the maintenance of pregnancy developed at a later stage in vertebrate phylogeny,

5. BIOCHEMICALNATUREOF

THE

GONADOTROPIN( s)

Two separate gonadotropins, the follicle-stimulating hormone ( FSH ) and the luteinizing hormone ( L H ) , are physiologically distinct in all tetrapods (Bern and Nandi, 1964; Hoar, 1965a). Although the nature of the fish gonadotropins is still not resolved, the data for a single factor seem to be accumulating steadily. In addition to the physiological Iiterature summarized below, the biochemical evidence for a single protein increases as the preparations (salmon or carp) have been purified more carefully ( Burzawa-GBrard and Fontaine, 1965, 1966; Schmidt et al., 1965; Yamazaki and Donaldson, 1968a; Donaldson et al., 1969). Most of the studies-both physiological and biochemical-have been concerned with teleosts. The excellent discussion by Pickford and Atz (1957) is still valuable. When fish pituitary extracts are tested in tetrapods, both the FSH-

1. REPRODUCl'ION

49

and the LH-like effects are frequently elicited. Witschi (1955) assayed pituitaries of sharks, garpikes and salmon-representatives of the three major groups of jawed fishes. Positive evidence for FSH was obtained by using the vaginal cornification test in the rat and for LH by using the weaver-finch feather reaction. The FSH content was very low; most of the gonadotropic activity appeared to be associated with LH. Ball ( 1960) has summarized these and other investigations-most of which provide suggestive evidence for the presence of the FSH-like effect and strong evidence for the LH action when fish pituitaries are injected into mammals. A series of papers by Otuska (1956a,b,c, 1957) using partially purified salmon pituitaries injected into newts and mice are in agreement with evidence for two gonadotropic proteins with distinct physiological effects. However, not all of this evidence for an action of fish gonadotropin( s ) in higher vertebrates is positive. Burzawa-Gdrard and Fontaine (1965) prepared a carefully purified extract of carp pituitaries which appeared to be a single protein and was entirely inactive in mammals. The histophysiology of the pituitary and the question of two different gonadotrophs is discussed in the chapter by Ball and Baker, Volume I1 (see also van Overbeeke and McBride, 1967). Evidence concerning the nature of fish gonadotropin has also been sought in tests with the pursed mammalian fractions injected into fishes. The preparations of mammalian FSH have consistently given negative results; LH, on the contrary, usually elicits both the gametogenetic and the steroidogenetic actions commonly associated with two different fractions in the higher vertebrates. The earlier literature is summarized in reviews previously cited. Among the more recent studies are those of Sundararaj and Goswami (1966) who report that LH alone, of several pituitary factors tested, would stimulate ovulation in Heteropneustes; Ahsan (1966b) who found a partially purified salmon gonadotropin to have an almost identical physiological action to mammalian LH when injected into hypophysectomized Couesius plumbeus; Fontaine and Chauvel (1961), Fontaine and Gdrard (1963), and Schmidt et al. ( 1965) who have reported strongly positive Galli-Mainini reactions (sperm release in frogs) with partially purified extracts of carp and salmon pituitaries; Ramaswami ( 1962) who has used enzymes which selectively digest or inactivate the gonadotropic factors in mammalian pituitaries (trypsin or pepsin for LH and ptyalin for FSH) and found convincing evidence for the LH-like factor but no support for FSH in Heteropneustes fossilis. Sundararaj and Nayyar's work (1967) with the latter species is confirmatory as is also that of Yamazaki (1965) and Yamamoto and Yamazaki ( 1967) for the acceleration of ovulation in hypophysectomized goldfish. However, it cannot yet be concluded that the fish gonadotropin

50

WILLIAM S. HOAR

is always physiologically similar to mammalian LH. By using the spermiation test in goldfish, Yamazaki and Donaldson (1968a) have obtained strong reactions with a partially purified salmon gonadotropin and with human chorionic gonadotropin (HCG) but no reaction with either FSH or LH. The mammalian chorionic gonadotropins are often used for experimental purposes and in fish culture work because of their ready availability. These factors, produced by the mammalian placenta during later stages of pregnancy, mimic many of the activities of the pituitary gonadotropins although they are definitely known to be different biochemically from the pituitary factors (C. D. Turner, 1966). Two chorionic gonadotropins are in common use: one, prepared from the serum of pregnant mares (PMS), usually has biological properties similar to a combination of FSH and LH; the other, prepared from human pregnancy urine ( HCG), acts more like LH. In mammals, the precise effects depend somewhat on the dosage used (C. D. Turner, 1966). When injected into fish, HCG has often been found to be quite effective (Yamazaki, 1965; Sundararaj and Nayyar, 1967; Yamazaki and Donaldson, 1968a) and this is in line with its LH-like activity. Pregnant mare serum is sometimes active (Ahsan and Hoar, 1963) but has usually been less consistently effective than HCG (reviews by Pickford and Atz, 1957; Dodd, 1960a). Sundararaj and Goswami (1966) report that 100 IU/fish of HCG will induce ovulation in Heteropneustes while 250 IU/fish of PMS are required to elicit the same reaction. In summary the evidence, both biochemical and physiological, now indicates a single proteinaceous gonadotropic factor in the pituitaries of teleost fishes. Although this may sometimes mimic FSH and frequently mimics LH when injected into tetrapods, it is clearly not identical with either of these factors. It more closely resembles LH but has some unique physiological properties and may be expected to differ from its counterpart in the tetrapods and even differ in its effects in different groups of fish. Investigations of some of the more primitive groups of fish should be rewarding. The lactogenic hormone (prolactin, luteotropin) is also gonadotropic in some mammals (the ovary of the rat), but no evidence has yet been presented for a comparable action in the lower vertebrates. The physiology of fish prolactin is considered in the chapter by Ball, Volume 11. B. The Gonadal Steroids

In each of the major groups of fishes, development of the secondary sex characters prior to breeding has been shown to depend on gonadal

1.

REX'RODUCXION

51

steroids (reviews cited and the chapter by Liley, this volume). This broad generalization is now backed by considerable experimental evidence and there seem to be no exceptions to it. Following gonadectomy, secondary sex characters fail to develop or, if seasonal in occurrence, regress. Their differentiation is initiated or stimulated by a wide variety of androgens and estrogens-both natural and synthetic. This may be the only broad generalization possible; in all the later events associated with reproduction ( ovulation, spermiation, spawning, and breeding behavior) the division of regulatory responsibilities between pituitary and gonads appears to be rather variable in different species. The physiologically most active gonadal hormones of higher vertebrates are testosterone from the interstitial cells of the testis, estradiol17p and its derivatives from the ovarian follicle, and progesterone from the corpus luteum. The biogenesis of all these compounds is from acetate via cholesterol, as the parent sterol, to testosterone through intermediate steps involving progesterone; the estrogens are derivatives of testosterone or closely related molecules and form the terminal part of the biosynthetic chain (see Fig. 14 in the chapter by Yamamoto, this volume). Thus, the biologically active gonadal steroids are linked through common pathways and one may expect to find small amounts of any or all of them in tissues which are primarily concerned with the synthesis of a single important hormone. The biogenesis of the adrenocortical steroids is linked to the same chain. These metabolic pathways are very ancient phylogenetically. Progesterone, estradiol-17/3, and some other estrogens have been identified in the ovaries of invertebrates (echinoderms and mollusks) as well as the lower vertebrates ( Botticelli et aE., 1960, 1961; Hisaw, 1963). Estrogens are also well known in plant tissues (Bickoff, 1963). It is not then surprising that both androgens and estrogens of several biochemical sorts have been regularly found in the gonadal tissues and blood of many different fishes. However, the presence of a gonadal steroid is no evidence of its significance as a hormone and the much more difficult task is to identify steroids which are physiologically important in regulating the reproductive activities of fishes. At present, it is by no means certain that all the steroids recognized as physiologically active in higher vertebrates are biologically significant in the fishes. Different groups of fishes seem to react somewhat differently to some of the same steroids (Egami and Arai, 1965). Moreover, it seems certain that there are important biosynthetic pathways of steroidogenesis in fishes which have not been recognized in the higher vertebrates (see below) although the general scheme of biosynthesis is probably similar in both groups (Breuer and Ozon, 1965; Idler and MacNab, 1967). In fishes, as in the higher verte-

52

WILLIAM S. HOAR

brates, some of the synthetic compounds are more active than the naturally occurring ones (Egami and Arai, 1965). The localization, identification, and analysis of synthetic pathways of fish steroids are being investigated in many places using a variety of techniques-physiological, biochemical, and histochemical. The specific histochemical techniques now available for the detection of enzymes involved in steroid biosynthesis ( e.g., 3P-hydroxysteroid dehydrogenase: Wattenberg, 1958; Galil and Deane, 1966) have been particularly useful in localizing the tissue sites of hormone synthesis (Delrio et al., 1965; di Prisco et al., 1965; Bara, 1965; Yaron, 1966; Simpson and Wardle, 1967; Wiebe, 1969a). Bern and Chieffi (1968) have provided a useful and comprehensive bibliography of the steroid hormones of fishes while Nandi (1967) summarizes the comparative endocrinology for the nonmammalian vertebrates. 1. THEANDROGENS The tissues of several different elasmobranchs have now been investigated, and there is good evidence that pathways recognized for biosynthesis of androgens in higher forms are operating in the elasmobranchs and that testosterone is an important end product in the testis. Chieffi and Lupo ( 1961), using chemical methods, investigated the testicular tissues of mature dogfish, Scylliorhinus stellaris, and recorded progesterone (100 pglkg), testosterone (50 pglkg), androstenedione (70 pg/kg), and estradioL17p (20 pglkg). Idler and Truscott (1966) were the first to isolate testosterone from the blood of a male elasmobranch. They found values which were relatively high (7.4 pg/lOO ml in Raja radiata) when compared with human males (about 0.56 pg/lOO ml). Testosterone was also found in the blood of female R . radiata, but the mean values were much lower. Progesterone, androstenedione, androsterone, and other steroids have been isolated from the semen of the dogfish, Squalus acanthias ( Simpson et al., 1963a), and testosterone biosynthesis was demonstrated by incubation techniques in the testis of this fish (Simpson et al., 1964). A number of different androgens have been isolated from the tissues and blood of teleost fishes. The studies of salmon tissues by Idler and his associates (1960, 1961a,b, 1964; Idler and Tsuyuki, 1959; Idler and Truscott, 1963; Idler and MacNab, 1967) are particularly significant. Steroid values have also been recorded for testicular tissues of the teleosts Morone labrax (Chieffi, 1962), Mugil cephulus (Eckstein and Eylath, 1968), and Serranus scriba (di Prisco and Chieffi, 1965); the latter is a hermaphroditic form.

1. REPRODUCTION

53

Idler’s meticulous chemical analyses have included gonadal tissues and blood from both the Atlantic salmon, Salmo salar, and the Pacific salmon-particularly the sockeye, Oncorhynchus nerka. Testosterone has been found in significant amounts together with several other steroids such as 17a-hydroxyprogesterone, recognized in higher forms as a step in testosterone biosynthesis, and androsterone-a product of testosterone synthesis. As might be expected from the known relationships of these substances, androgens have also been found in the tissues of female fish. Testosterone values are extremely variable (ranging up to 17 pg/lOO ml of male blood) ; since tissue values must represent a balance between synthesis and utilization it is always difficult to relate the actual amounts to the physiology. Three particular facets of Idler’s studies require special comment: ( a ) the presence of ll-ketotestosterone, ( b ) the occurrence of conjugated testosterone, and ( c ) the marked variations associated with the life cycle of the salmon. Idler et al. (1960, 1961a,b) were the first to identify ll-ketotestosterone as a natural product and subsequently to show its biological activity as an androgen in stimulating the development of secondary sex characters in salmon and chickens. The isolation of ll-ketotestosterone from both males and females of Atlantic and Pacific salmon suggests that the synthetic pathways of the androgens in these fishes are somewhat different from the usually recognized ones. This steroid is present in amounts up to 17 pg/lOO ml of blood and occurs along with several of the more familiar androgenic compounds including testosterone. Idler and Truscott ( 1963) found that testosterone and l7a-hydroxyprogesterone will serve as precursors for ll-ketotestosterone in the sockeye salmon and, subsequently Idler and MacNab (1967) demonstrated its in vitro synthesis from adrenosterone and testosterone in Atlantic salmon gonads and sperm; they suggest the probable pathways of biosynthesis. Arai and Tamaoki (1967) also studied the in vitro synthesis in Atlantic salmon gonads and sperm from adrenosterone and testosterone; they too suggest probable pathways of biosynthesis. Arai and Tamaoki (1967) have studied the in vitro biosynthesis of ll-ketotestosterone in rainbow trout, Salmo gairdneri. Conjugated testosterone was first reported in fish blood by Grajcer and Idler (1961, 19fX3). In higher vertebrates, the steroids are transported, at least in part, as conjugates with serum proteins and glucuronic acid. Grajcer and Idler obtained the release of testosterone in amounts of 13.7 pg/lOO ml of male sockeye salmon blood after treatment with the enzyme p-glucuronidase. The “free” testosterone in these samples of blood was 1.7 pg/100 ml. The blood of female salmon also contained conjugated as well as “free” testosterone with relatively less of the con-

54

WILLIAM S. HOAR

jugated form (7.6 pg of conjugated to 7.8 p g per 100 ml of free) even though the total amounts were about the same. Since these initial studies of testosterone glucuronoside, Idler and Truscott ( 1966) have reported on conjugated testosterone in the skates, Raia radiata and R. ocellata; these conjugates are probably as common in fish blood as they apparently are in the higher vertebrates (previous references and di Prisco and Chieffi, 1965). Schmidt and Idler (1962) studied the quantitative changes in several plasma steroids ( including 11-ketotestosterone and testosterone) during the migration of the sockeye salmon. Definite differences were recorded and the shift in ratio of 11-ketotestosterone and testosterone may be significant in the physiology and behavior of the spawning migration. Changing levels of androgens have also been reported in females of the ovoviviparous elasmobranch, Torpedo murmrata, during the reproductive cycle (Buonanno et aZ., 1964); there was a moderate increase in androsterone at the end of the gestation period and a marked steady rise in dehydroxyepiandrosterone from pregestation to the midgestation period, These variations may indicate physiological changes in the demands for estrogens rather than imply a role for androgen in the gestating ray. In the stickleback, Gasterosteus aculeatus, where the development of sexual behavior is cyclical and associated with strong agonistic behavior in the males, Gottfried and van Mullem (1967) report that the dominant males have testicular androgen levels which are five to seven times higher than the nondominant individuals; testosterone could not be detected in the testes of the nondominant fish but was present in the dominants. These cyclical and seasonal changes in gonadal steroidogenesis are presumably triggered by variations in the gonadotropic activity of the pituitary; seasonal changes in the latter have been traced in both plaice Pleuronectes (Barr and Hobson, 1964) and the perch Perca (Swift and Pickford, 1965).

2. THEESTROGENS AND PROGESTERONE Estradiol-17P has been found in the ovaries of many fishes at all levels in phylogeny: the lampreys, Petromyzon marinus (Botticelli et al., 1963), dogfishes, Squalus suckbyi (Wotiz et al., 1958, 1960) and Scyliorhinus caniculus (Simpson et al., 1963b), the ray, Torpedo marmorata (Chieffi, 1962; Chieffi and Lupo, 1963), the ratfish, Hydrolugus colliei (Hisaw, 1963), the lungfish, Protopterus annectes ( Dean and Chester Jones, 1959), and many different teleosts (Cedard et al., 1961; Gottfried et al., 1962; Hisaw, 1963; Lupo and Chieffi, 1963). This wide dktributiontogether with its presence in several invertebrate phyla-suggests that it

1. REPRODUCTION

55

will be found in at least small amounts wherever there is active steroid synthesis. Estradiol-17P has also been identified in the blood of elasmobranchs (Simpson et al., 1963b) and teleosts (Cedard et aZ., 1961) and is probably the physiologically most important estrogen. Gottfried ( 1964 ) tabulates estradiol-17,B values for ovarian tissues ranging from a trace to 120 pg/kg, with the most usual amounts varying from 10 to 20 ,ug/kg. Estradiol-l7P is recognized as the parent substance for several other estrogens known to occur widely in the animal world. Particularly frequent in occurrence are estrone (an oxidation product of estradiol-17P) and estriol which is derived from estrone by 16-hydroxylation and reduction at 17. Both estrone and estriol were found in the ovaries of many of the species listed above, but the proportions are quite variable and occasionally one of the compounds is absent (summary by Hoar, 1965a). Again, this is not surprising with a series of related substances in which the active utilization of at least one of them probably varies with the stage of maturity. Cedard et al. (1961) traced seasonal changes in the estrogens of the blood of both male and female Atlantic salmon. The total estrogen content in both sexes showed a five- to sixfold increase at the time of spawning, reaching 5-6 pg/lOO ml of blood. Estrone was found at all seasons; estriol appeared in significant amounts only at spawning; and estradiol was only present in small amounts and seemed to disappear completely in the females at spawning and in the postspawning males. Again, it is difficult to interpret the findings in terms of physiological demands but results such as these emphasize the hazards of conclusions based on a few estimations of the gonadal steroids at only one season. Progesterone has been found in the tissues of all vertebrates and many of the invertebrates (Gottfried, 1964; Botticelli et al., 1960, 1961). It is probably ubiquitous as a link in steroid biogenesis. As yet, however, there is no definitive evidence that progesterone is a physiologically active hormone with distinct endocrine responsibilities. Hisaw ( 1959, 1963) has discussed this problem at length and concludes that the role of progesterone as a hormone was established during the evolution of placentation in the mammals although it probably had its beginning among reptilian ancestors. Chieffi ( 1962, and elsewhere) has convincingly correlated the development of corpora lutea with the stages of gestation in the electric ray Torpedo. He has also demonstrated histochemically the biosynthesis of a variety of steroids-including progesterone-in the corpora lutea ( Chieffi, 1961; di Prisco et aE., 196S), while chemical methods have identified progesterone in Torpedo blood during pregestation (di Prisco et al., 1967) but again the question of its presence as hormone or precursor has )

56

WILLIAM S. HOAR

'O!

oocytes

9 days after parturition

Fig. 16. Variations in the number of oocytes and embryos during the 28-day gestation period of Poeciliu. The number was cdculated as a percentage of the total number of structures present in each ovary per 24 hr. From Stolk (1951b).

not been settled, It appears, however, that active steroidogenesis occurs in the corpora lutea of some fishes; it seems equally evident that this is not the case in all fishes (Lambert and van Oordt, 1965).Further work is required to clarify the endocrinological status of these structures.

C. Reproductive Cycles and Their Coordination Reproduction is almost always a seasonal or cyclical phenomenon. An annual cycle of temperatures and photoperiods characterizes the tem-

1.

57

RFPRODUCTION

perate and frigid zones; the rainy seasons may markedly alter freshwater habitats of the tropics. In these seasonally unstable environments, reproduction is geared to take advantage of seasons which offer the greatest opportunities for survivaI and development of the new generation. Even where conditions are relatively stable and the eggs or young are produced regularly throughout the year, there may still be a cycle of gonadal maturation imposed by the energy demands of maturing a batch of eggs or young (Fig. 16). These cycles of gonadal development frequently alter many aspects of metabolism as well as behavior and reproductive physiology. Cycles of active feeding with storage of fat and long periods of starvation are characteristic of many species (Greene, 1926; Idler and Bitners, 1960, and earlier; Idler and Clemens, 1959; Bentley and Follett, 1965; Tomlinson et d.,1967; Fig. 17). The electrolyte metabolism may change seasonally both in species which inhabit the ocean waters of relativeIy constant salinity (Woodhead, 1968) and in the euryhaline and anadromous forms (see chapters by Holmes and Donaldson, and Hickman and Trump, Volume I ) . The basic cycle is probably the one imposed by the seasonal nature of reproduction, and these regular changes in metabolism are secondary to it. The endocrine system forms the major link between the environment and the organs concerned with reproduction. Changing environmental

80-

c

60-

0

\

\

/

Feeding

\\

/

/ \

/

Y L?

40

-

-

20

Gonad development

'.J

L z L

n "

( 1

I

I

I

I

'

Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jui Aug

Fig. 17. Relation between feeding cycle (percent sample with food in stomach) and reproductive cycle (gonad development, percent fish with ripening gonads; spawning, percent ripe fish) of the haddock, Melanogrummus aeglifinus. Data from Homans and Vladykov ( 1954 ) .

58

WILLIAM S . HOAR

conditions, operating through the sensory system and specific centers in the brain, trigger neurosecretions which in turn regulate the activities of the pituitary gland. The pituitary hormones have direct effects on gametogenesis, metabolism, and behavior; these hormones also regulate the development of the gonadal endocrine tissues. Gonadal hormones take over some of these pituitary responsibilities and carry on the coordination of events in the production of gametes, sexual behavior, fertilization, and sometimes parental care. There are, in fact, several cycles within the gonads (Figs. 16 and 18), but many of the details of regulation at the cellular level remain to be unraveled. Studies of the cyclical changes in the structure of the gonads and endocrine organs, as well as variations in actual secretion of hormones (Barr and Hobson, 1964; Swift and Pickford, 1965; Gottfried and van Mullem, 19671, form one of the most voluminous components of the literature concerned with reproduction. No attempt will be made to summarize it here; the reader is referred to the many existing reviews (L. H. Matthews and Marshall, 1956; Pickford and Atz, 1957; Hoar, 1959, 1965a)b; see chapter by Liley, this volume). Although the pattern of. regulation via the peripheral sense organs, the neural centers, pituitary, and gonads is general throughout the vertebrates, the details vary considerably even in closely related species. A definite breeding season is absent in the female of the spotted dogfish, Scyliorhinus caniculus, and in the male spermatogenesis is continuous with a regular progression in the ampullae from spermatogonia to sperm (Fig. 4 ) ; in comparison there is a limited breeding period of about 2 months in the spiny dogfish, Squulus acanthias, and during this period Female Oocytes

Yo1 k formation

.15

Maturation Male Spermatogenesis Sex characters

10

5

Interstitial ti s u e OC

Fig. 18. Seasonal cycle of reproduction in Gasterosteus. Ordinate, months; curve, temperature in "C; horizontal black bar, spawning period; horizontal arrows, periods when more than 50%of samples showed characters listed on the left. Data from Craig-Bennett (1931); reproduced from Hoar ( 1955) by permission of the Cambridge University Press.

59

1. REPRODUCTION

there is a maximum accumulation of semen in the efferent ducts which is followed by a pause in sperm production (Fig. 15).This suspension of sperm production is the result of a cessation in the proliferation of spermatogonia several months earlier ( Simpson and Wardle, 1967). Since a band of degenerating ampullae appears in the testis of the spotted dogfish following hypophysectomy (Dodd et ul., 1960), there seems little doubt that these cyclical changes in the sperm production of the spiny dogfish are regulated by the pituitary gonadotropins. Similar examples could be drawn from many different groups of fishes. Occasionally, the details of regulation may even be somewhat different in the males and females (Wiebe, 1968b,c, 1969b). There are diurnal as well as seasonal cycles. In the cyprinodont, Oryzius lutipes, there is a daily cycle of ovulation (between 1 AM and 5 A M ) , mating behavior ( 4 AM to 7 A M ) , and the laying of fertilized eggs which follows soon after mating. Ovulation and oviposition are independent phenomena; the former depends on the temperature and the light cycle while the latter depends on contact stimuli associated with the sexual embrace (Egami, 1959a,b; Egami and Nambu, 1961). In another cyprinodont, RivuZus murmorutus, there is an internal self-fertilization which is also timed by the daily light cycle. Rivulus murmorutus is hermaphroditic with functional ovotestes. The timing of events within these organs is so correlated that there is a peak frequency in ovulation; fertilization occurs at dawn with a peak in oviposition at noon. Harrington (1963) also found evidence of seasonal changes in the cycles in accordance with light conditions. The modifications in physiological controls are fully as numerous and diverse as those of an anatomical nature. In a sense, the reproductive system of an animal is independent of the other organ systems and can, perhaps, respond to evolutionary pressures more freely; although reproduction is indispensable to the survival of the species, it is not a matter of life or death for the individual. Perhaps, for this reason, the adaptive variations in anatomy and physiology are particularly diverse and curious in the biology of reproduction. At any rate they emphasize the opportunism, the compromise, and the adaptiveness of Darwinian evolution. ACKNOWLEDGMENT

It is a pleasure to thank Margaret Jensen for her careful preparation of Figures 1 through 14. REFERENCES Adam, H. (1963). The pituitary gland. In “Biology of Myxine” (A. Brodal and R. Fange, eds.), pp. 459476. Oslo Univ. Press, Oslo.

60

WILLIAM S. HOAR

Ahsan, S. N. (1966a). Cyclical changes in the testicular activity of the lake chub, Couesius plumbeus ( Agassiz). Con. J. Zool. 44, 149-171. Ahsan, S. N. (1966b). Effects of gonadotropic hormones on male hypophysectomized lake chub, Cousesius plumbeus. Can. 1. 2001.44, 703-717. &an, S. N., and Hoar, W. S. (1963). Some effects of gonadotropic hormones on the threespine stickleback, Gasterosteus aculeatus. Can. J . 2001.41, 1045-1053. Alcock, A. (1892). On utero-gestation in Trygon Bleekeri. Ann. Mag. Not. Hist. 9, [el, 417-427. Amoroso, E. C. (1960). Viviparity in fishes. Symp. Zool. SOC.London 1, 153-181. Arai, R., and Tamaoki, B. ( 1967). Steroid biosynthesis in uitro by testes of rainbow trout, Salmo gairdneri. Gen. Comp. Endocrinol. 8, 305-313. Atz, J. W. (1964). Intersexuality in Fishes, In “Intersexuality in Vertebrates Including Man” (C. N. Armstrong and A. J. Marshall, eds.), pp. 14S232. Academic Press, London. Austin, C. R., and Walton, A. (1960). Fertilisation. In “Marshall’s Physiology of Reproduction” (A. S. Parkes, ed.), Vol. 1, Part 2, pp. 310-416. Longmans, Green, New York. Bailey, R. J. (1936). The osteology and telationships of the phallostethid fishes. J. Morphol. 59, 453-183. Balfour, F. M. (1875). On the origin and history of the urogenital organs of vertebrates. J . Anat. Physiol. 10, 1 7 4 8 . Ball, J. N. (1960). Reproduction in female bony fishes. Symp. 2001.SOC. London 1, 105-135. Ball, J. N. ( 1962). Brood-production after hypophysectomy in the viviparous teleost Mollienesia latipinnu Le Sueur. Noture 194, 787. Bara, G. ( 1965). Histochemical localization of A5-3,8-hydroxysteroid dehydrogenase in the ovaries of a teleost fish, Scomber scomber L. Gen. Cornp. Endocrinol. 5, 284-296. Ban, W. A. (1963a). The endocrine control of the sexual cycle in the plaice, Pleuronectes platessa (L.). I. Cyclical changes in the normal ovary. Gen. Comp. Endocrinol. 3, 197-204. Barr, W. A. (196313). The endocrine control of the sexual cycle in the plaice, PZeuronectes platessa (L.). 11. The endocrine control of oogenesis. Gen. Comp. Endocrinol. 3, 205-215. Barr, W. A. (1963~).The endocrine control of the sexual cycle in the plaice, Pleuronectes platessa (L.). 111. The endocrine control of spermatogenesis. Gen. Comp. Endocrinol. 3, 216-225. Barr, W. A., and Hobson, B. M. (1964). Endocrine control of the sexual cycle in the plaice, Pkuronectes platessa L. IV. Gonadotropic activity of the pituitary gland. Gen. Comp. Endocrinol. 4, 608-613. Belsare, D. K. (1965). Changes in gonads and the thyroid gland after hypophysectomy in OphicephaZus punctatus Bloch. J. Exptl. 2001.158, 1-8. Bentley, P. J., and Follett, B. K. (1965). Fat and carbohydrate reserves in the river lamprey during spawning migration. Life Sci. 4, 2003-2007. Bern, H. A., and Chieffi, G. (1968). Bibliography on the steroid hormones of fishes. Pubbl. Staz. 2001. Napoli 36, 287-320. Bern, H. A., and Nandi, J. (1964). Endocrinology of poikilothermic vertebrates. In “The Hormones” ( G . Pincus, K. V. Thimann, and E. B. Astwood, eds.), Vol. 4, pp. 199-298. Academic Press, New York. Bertin, L. (1958a). Sexualit6 et Gcondation. In “Trait6 de zoologie” P.-P. Grass&, ed.), Vol. XIII, Part 2 pp. 1584-1652. Masson, Paris.

1.

REPRODUCTION

61

Bertin, L. ( 1958b). Nidification. In “Trait6 de zoologie” (P.-P. Grass&,ed.), Vol. 13, Part 2, pp. 1653-1684. Masson, Paris. . de T61CostBens. In “Trait6 de zoologie” (P.-P. Grass&, Bertin, L. ( 1 9 5 8 ~ )Viviparit6 ed.), Vol. 13, Part 2, pp. 1791-1812. Masson, Paris. Bickoff, E. M. (1963). Estrogen-like substances in plants. Proc. Ann. Biol. Colloq. (Oregon State Univ.) 22, 93-118. Boisseau, J. P. (1964). Effects de la castration et de I’hypophysectomie sur I’incubation de 1’Hippocarnpe mile ( Hippocampus hippocampus L.). Compt. Rend. 259, 48394840. Bolin, R. L. (1941). Embryonic and early larval stages of the cottid fish, Orthonopias triacis Starks and Mann. Stanford Ichthyol. Bull. 2, 73-82. Borcea, I. ( 1906). Recherches sur le systkme uro-ghnital des elasmobranches. Arch. Zool. Exptl. Gen. 4, 199484. Botte, V., Chieffi, G., and Stanley, H. P. (1963). Histological and histochemical observations on the male reproductive tract of Scylwrhinus stelluris, Torpedo mrmorata and T . torpedo. Pubbl. Staz. Zool. Napoli 33. 224242. Botticelli, C. R., Hisaw, F. L., Jr., and Roth, W. D. (1963). Estradiol-l7P, estrone, and progesterone in the ovaries of lamprey (Petromyzon marinus). Proc. SOC. Exptl. Biol. Med. 114, 255-257. Botticelli, C. R., Hisaw, F. Y., Jr., and Wotiz, H. H. (1960). Estradiol-17P and progesterone in ovaries of starfish ( Pisaster ochraceous). Proc. SOC. Exptl. Biol. Med. 103, 875-877. Botticelli, C. R., Hisaw, F. L., Jr., and Wotiz, H. H. (1961). Estrogens and progesterone in the sea urchin (Strongylocentrotus franciscanus) and pecten (Pecten hericius). Proc. SOC. Exptl. Biol. Med. 1106, 887-889. Brambell, F. W. R. (1956). Ovarian changes. In “Marshall’s Physiology of Reproduction’’ ( A . S. Parkes, ed.), 3rd ed., Vol. 1, Part 1, pp. 397-542. Longmans, Green, New York. Breder, C. M., Jr., and Rosen, D. E. (1966). “Modes of Reproduction in Fishes.” Natural History Press, Garden City, New York. Bretschneider, L. H., and de Wit, J. J. (1947). “Sexual Endocrinology of NonMammalian Vertebrates.” Elsevier, Amsterdam. Breuer, H., and &on, R. (1965). MCtabolisme des hormones stbriodes androgines et oestrogenes chez les vert6brbs infbrieurs. Arch. Anat. Microscop. Morphol. Exptl. 54, 17-33. Budker, P. (1958). La viviparitb chez les silaciens. In “Trait6 de zoologie” (P.-P. Grass&, ed.), Vol. 13, Part 2, pp. 1755-1790. Masson, Paris. Buonanno, C., Chieffi, G., and Imparato, E. (1964). Variazioni del tasso plasmatic0 di deidroepiandrosterone e androsterone nel corso del ciclo riproduttivo di Torpedo marmorata (Selacio ovoviviparo). Pubbl. Staz. Zool. Napoli 34, 66-74. Burzawa-GCrard, E., and Fontaine, Y. A. (1965). ActivitCs biologiques d’un facteur _ _ hypophysaire gonadotrope purifC de Poisson t616ostCen. Gen. Comp. Endocrinol. 5. 87-95. Burzawa-Cirard, E., and Fontaine, Y. A. (1966). Sur le probkme de l’unicit6 ou de la dualit6 de l’horrnone gonadotrope hypophysaire d’un Tilbost6en, la Carpe. Etude du poids molCculaire de la ou des substances actives, Ann. Endocrinol. (Paris) 27, 305-309. Buser-Lahaye, J. ( 1953). Btude exp&rimentale d u dbterminisme de la rhgbnbration des nageoires chez les Poissons TClbstiens. Ann. Inst. Odanogr. Monaco. 28, 1-61.

82

WILLIAM S. HOAR

Busson-Mabillot, S. (1967). Structure ovarienne chez la Lamproie de Planer adulte Lampetra planeri (Bloch), Arch. Zool. Exptl. Cen. 108, 413446. Cedard, L., Fontaine, M., and Nomura, T. ( 1961). Sur la teneur oestrogknes du sang du Saumon adulte (Salmo salar L.), en eau douce. Compt. Rend. 252, 265@ 2657. Chambolle, P. (1964). Influence d l’hypophysectomie sur la gestation de Gambusia sp. (Poisson TClCostCen). Compt. Rend. 259, 3855-3857. Chester Jones, I., and Ball, J. N. (1962). Ovarian-pituitary relationships. In “The Ovary” ( S . Zuckerman, ed.), Vol. 1, pp. 361434. Academic Press, New York. Chieffi, G. (1949). Ricerche sul differenziamento dei sessi negli embrioni di Torpedo ocellata. Pubbl. Staz. Zool. Napoli 22, 57-78. Chieffi, G. ( 1952). Sull’organogenesi dell’interrenale e della gonade in Torpedo ocelkzta e in Scylliorhinus canicukz. Pubbl. Staz. Zool. Napoli 23, 186-200. Chieffi, G. (1961). La luteogenesi nei selaci ovovivpari. Ricerche istologiche e istochimiche in Torpedo marmorata e Torpedo ocellata. Pubbl. Staz. Zool. Napoli 32, 145-166. Chieffi, G. (1962). Integration of reproductive functions. 11. Endocrine aspects of reproduction in elasmobranch fishes. Cen. Comp. Endocrinol. Suppl. 1, 275-285. Chie5, G. (1967). The reproductive system of elasmobranchs: development and endocrinological aspects. In “Sharks, Skates and Rays” (P. W. Gilbert, R. F. Mathewson, and D. P. Rall, eds.), pp. 553-580. Johns Hopkins Press, Baltimore, Maryland. Chie5, G., and Botte, V. (1962). I1 Tessuto interstiziale del TesticoIo dei ciclostomi. Ricerche istologische ed istochimiche in Lampetra fluviatilis e Lampetra planeri. Rend. 1st. Schi. Univ. Camerim 3, 90-95. Chieffi, G., and Lupo, C. (1961). Identification of oestradiol-17p testosterone and its precursors from Scylliorhinus stellaris testes. Nature 190, 169-170. Chie5, G., and Lupo, C. (1963). Identification of sex hormones in the ovarian extracts of Torpedo marmomta and Bufo vulgaris. Cen. Comp. Endocrinol. 3, 149-152. Chie5, G., Corte, F. D., and Botte, V. (1961). Osservazioni sul tessuto interstiziale del testiculo dei Selaci. Boll. Zool. 28, 211-217. Clemens, H. P., and Grant, F. B. (1964). Gonadal hydration of carp (Cyprinus carpio) and goldfish ( Carassius auratus) after injections of pituitary extracts. Zoologica 49, 193-210. Clemens, H. P., and Grant, F. B. (1965). The seminal thinning response of carp ( Cyprinus carpio ) and rainbow trout (Salmo gairdnerii) after injections of pituitary extracts. Copeia 1965, No. 2, 174177. Clemens, H. P., Ciereszko, L. S., Shoemaker, J. D., and Grant, F. B. (1964). Partial characterization of the gonadal hydration principle in the pituitaries of carp (Cyprinus carpio). Cen. Comp. Endocrinol. 4, 503-507. Collenot, G., and &on, R. (1964). Mises en kvidence biochemique et histochemique d’une A6-3p-hydroxystCroide dhhydrogenase dans le testicule de Scyliorhinus canicula L. Bull. Soc. Zool. France 89, 577687. Cooper, L. J. (1952). A histological study of the reproductive organs of crappies (PommD nigro-Maculatus and Pomoxis annulah). Trans Am. Microscop. SOC. 71, 393404. Courrier, R. (1921). Sur l’existence d u n e glande interstitielle dans le testicule des Poissons. Compt. Rend. SOC. Biol. 85, 939-941. Craig-Bennett, A. ( 1931 ). The reproductive cycle of the Three-spined stickleback, Gasterosteus aculeatus, Linn. Phil. Trans. Roy. SOC. London B219, 197-279.

1. REPRODUCTION

63

D’Ancona, U. ( 1945). Sexual differentiation of the gonad and the sexualization of the germ cells in teleosts. Nature 156, 603-604. D’Ancona, U. (1950). Dhtermination et diffbrenciation du sexe chez les poissons. Arch. Anat. Microscop. Morphol. Exptl. 39, 274-294. D’Ancona, U. ( 1956). Morphoghn6se et diffkrenciation sexuelle chez les poissons tkkosthens. Bull. SOC. 2001.France 81, 219-229. D’Ancona, U. ( 1960). Induzione genetica e suscettibilita territoriale nel differenziamento sessuale della gonade ne Teleostomi. Riu. Biol. (Perugia) 53, 11-25. Dean, F. D., and Chester Jones, I. (1959). Sex steroids in the lungfish (ProtopterUs unnectens Owen), J . Endocrinol. 18, 366-371. Della Corte, F., and Chieffi, G. (1961). Modificazioni dell’ipofisi di Torpedo marmorata Risso, durante la gravidanza. Boll. Zool. 28, 219-225. Delrio, G., Botte, V., and Chieffi, G. (1965). Ulteriori osservazioni sul tessuto interstiziale del testicolo dei Teleostei. Boll. Zool. 32, 199-205. di Prisco, C. L., and Chief%, G. (1965). Identification of steroid hormones in the gonadal extract of the synchronous hermaphrodite teleost Serrunus scriba. Gen. Comp. Endocrinol. 5, 698-699. di Prisco, C. L., Botte, V., and Chieffi, G. (1965). Differenze nella capacita di biosintesi degli ormoni steroidi tra follicoli post-ovulatori ed atresici in Torpedo marmorata e Scyliorhinus stelkris. Boll. Zool. 32, 185-191. di Prisco, C. L., Vellano, C., and Chieffi, G . (1967). Steroid hormones in the plasma of the elasmobranch Torpedo murmorata at various stages of the life cycle. Gen. Comp. Endocrinol. 8, 325-331. Dodd, J. M. ( 1960a). Gonadal and gonadotrophic hormones in lower vertebrates. In “Marshall’s Physiology of Reproduction” (A. S. Parkes, ed.), 3rd ed., Vol. 1, Part 5 Chapter 11, pp. 417-582. Longmans, Green, New York. Dodd, J. M. ( 1960b). Genetic and Environmental aspects of sex determination in cold-blooded vertebrates. Mem. SOC. Endocrinol. 7, 1 7 4 4 . Dodd, J. M., and Goddard, C. K. (1961). Some effects of oestradiol benzoate on the reproductive ducts of the female dogfish, Scyliorhinus caniculw. Proc. 2001. Soc. London 137, 325-331. Dodd, J. M., Evennett, P. J., and Goddard, C. K. ( 1960). Reproductive endocrinology in cyclostomes and elasmobranchs. Symp. 2001.SOC. London 1, 77-103. Donaldson, E. M., Yamazaki, F., and Dye, H. (1969). Preparation of gonadotropin from salmon ( Oncorhynchus tshawytscha) pituitary glands. In preparation. Eckstein, B., and Eylath, U. (1968). Steroidogenesis in testicular tissue of a teleost, Mugil cephalus. Comp. Biochem. Physiol. 25, 207-212. Egami, N. (1959a). Record of the number of eggs obtained from a single pair of Oryzias latipes kept in laboratory aquarium. J . Fac. Sci., Uniu. Tokyo, Sect. IV 8, 521-538. Egami, N. (1959b). Effects of exposure to low temperature on the time of oviposition and the growth of the oocytes in the fish, Oryzias lutipes. J . Fac. Sci., Uniu. Tokyo, Sect. IV 8, 539-548. Egami, N., and Arai, R. ( 1965). Male reproductive organs of Teleostei and their reactions to androgens, with note on androgens in Cyclostomata and Teleostei. PTOC.2nd Intern. Congr. Endocrinol., London, 1964 pp. 146-152. Excerpta Med. Found., Amsterdam. Egami, N., and Ishii, S. ( 1962). Hypophysial control of reproductive functions in teleost fishes. Gen. Comp. Endocrinol. Suppl. 1, 248-253.

64

WILLIAM S. HOAR

Egami, N., and Nambu, M. (1961). Factors initiating mating behavior and oviposition in the fish, Oryzias lutipes. J . Fm. Sci., Uniu. Tokyo., Sect. IV 9, 263-278. Evennett, P. J., and Dodd, J. M. (1963). Endocrinology of reproduction in the river lamprey. Nature 197, 715-716. Everett, N. B. (1945). The present status of the germ-cell problem in vertebrates. Biol. Reu. 20, 45-55. Follenius, E., and Porte, A. (1960). Cytologie fine des cellules interstitielles du testicule du Poisson Lebistes reticulatus P. Experientiu 16, 190. Fontaine, M., and Chauvel, M. (1961). Evaluation du pouvoir gonadotrope de I‘hypophyse des Poissons TBltbstBens, et en particulier du S a l m sulur L. P diverses &apes de son dhveloppement et de ses migrations. Compt. Rend. 252, 822824. Fontaine, Y. A,, and G r a r d , E. (1963). Purification d’un facteur gonadotrope de l’hypophyse d‘un TklhostBen, la Carpe (Cyprinus carpio L.). Compt. Rend. 256, 5634-5637. Franchi, L. L. (1962). The structure of the ovary. In “The Ovary” ( S . Zuckerman, ed.), Vol. 1, pp. 121-142. Academic Press, New York. Franchi, L. L., Mandl, A. M., and Zuckerman, S. (1962). The development of the ovary and the process of oogenesis. In “The Ovary” (S. Zuckerman, ed.), Vol. 1, pp. 1-88. Academic Press, New York. Fratini, L. ( 1953). Osservazioni sulla spermatogenesi di Scylliorrhinus cunicula. Pubbl. Staz. Zool. Napoli 24, 201-216. Galil, A. K. A,, and Deane, H. W. (1966). A6-3P-hydroxysteroid dehydrogenase activity in the steroid-hormone producing organs of the ferret (Mustela puturius furo).J . Reprod. Fertility 11, 33%338. Garnaud, J. (1950). La reproduction et l’incubation branchiale chez Apogon imberbis, G. et L. Bull. Inst. Oceanog. 977, 1-10. Grard, P. ( 1958). Organes reproducteurs. In ‘‘Trait4 de zoologie” (P.-P. GrassB, ed.), Vol. 13, Part 2, pp. 15651583. Masson, Paris. Gilbert, P. W., and Schlernitzauer, D. A. (1966). The placenta and gravid uterus of Curcharhinus falciformis.Copeia 1966, No. 3, 451-457. Goodrich, E. S. (1930). “Studies on the Structure and Development of the Vertebrates.” Macmillan, New York (Paperback ed., Dover, New York, 1958). Gottfried, H. (1964). The occurrence and biological significance of steroids in lower vertebrates. A review. Steroids 3, 219-242. Gottfried, H., and van Mullem, P. J. (1967). On the histology of the interstitium and the occurrence of steroids in the stickleback (Gusterosteus muleatus L.) testis. Actu Endocrinol. 56, 1-15. Gottfried, H., Hunt, S. V., Simpson, T. H., and Wright, R. S. (1962). Sex hormones in fish. The oestrogens of cod ( Gadus callarim). J . Endocrinol. 24, 425430. Grajcer, D., and Idler, D. R. (1961). Testosterone, conjugated and “free” in the blood of spawned Fraser river sockeye salmon (Oncorhynchus nerka). Can. J . Biochem. Physiol. 39, 1585-1593. Grajcer, D., and Idler, D. R. (1963). Conjugated testosterone in the blood and testes of spawned Fraser river sockeye salmon ( Oncorhynchus nerku). Can. J . Biochem. Physiol. 41, 23-30, Greene, C. W. (1926). The physiology of the spawning migration. Physiol. Rev. 6, 201-241. Ham, A. W. ( 1965). “Histology,” 5th ed. Lippincott, Philadelphia, Pennsylvania.

1.

REPRODUCI’ION

65

Hardisty, M. W., Rothwell, B., and Steele, K. (1967). The interstitial tissue of the testis of the river lamprey, Lampetra fluuiatilis. j . Zool. 152, $18. Harrington, R. W., Jr. (1959). Delayed hatching in stranded eggs of marsh killifish, Fundulus confluentus. Ecology 40, 430-437. Harrington, R. W., Jr. ( 1966). Twenty-four-hour rhythms of internal self-fertilization and of oviposition by hermaphrodites of Rivulus manoratus. Physiol. 2001.36, 325341. Henn, A. W. (1916). On various South American poeciliid &shes. Ann. Carnegie Museum 10, 93-142. Hisaw, F. L. (1959). Endocrine adaptations of the mammalian estrous cycle and gestation. In “Comparative Endocrinology” (A. Gorbman, ed. ), pp. 533-552. Wiley, New York. Hisaw, F. L. (1983). Endocrines and evolution of viviparity among the vertebrates. Proc. Ann. BioE. Colloq. (Oregon State Uniu.) 22, 119-138. Hisaw, F. L., and Abramowitz, A. A. (1938). The physiology of reproduction in the dogfish, Mustelus canis. Rept. Woods Hole Oceanog. Inst. 1937, 21-22. Hisaw, F. L., and Abramowitz, A . A. ( 1939). Physiology of reproduction in the dogfishes, Mustelus canis and Squalus acanthias. Rept. Woods H o b Oceanog. Inst. 1938, 22. Hisaw, F. L., Jr., and Hisaw, F. L. (1959). Corpora lutea of elasmobranch fishes. Anat. Record 135, 269-277. Hoar, W. S. (1955). Reproduction in teleost fish. Mem. SOC. Endocrinol. 4, 5-24. Hoar, W. S. (1957). The gonads and reproduction. In “The Physiology of Fishes” ( M . E. Brown, ed.), Vol. 1, pp. 287321. Academic Press, New York. Hoar, W. S. (1959). Endocrine factors in the ecological adaptation of fishes. In “Comparative Endocrinology” ( A . Gorbman, ed.), pp. 1-23. Wiley, New York. Hoar, W. S. (1965a). Comparative physiology: Hormones and reproduction in fishes. Ann. Reu. Physiol. 27, 51-70. Hoar, W. S. (196513). The endocrine system as a chemical link between the organism and its environment. Trans. Roy. SOC. Can., Sect. I l l , Ser. IV, 3, 175-200. Hoar, W. S. (1966). Hormonal activities of the pars distalis in cyclostomes, fish and amphibia. In “The Pituitary Gland” (G. W. Harris and B. T. Donovan, eds.), Vol. 1, pp. 242-294. Butterworth, London and Washington, D. C . Hoar, W. S., Wiebe, J., and Wai, E. H. (1967). Inhibition of the pituitary gonadotropic activity of fishes by a dithiocarbamoylhydrazine derivative (I.C.I. 33, 828). Gen. Comp. Endocrinol. 8, 101-109. Homans, R. E. S., and Vladykov, V. D. (1954). Relation between feeding and the sexual cycle of the haddock. J. Fisheries Res. Board Can. 11, 535-542. Idler, D. R., and Bitners, I. (1960). Biochemical studies on sockeye salmon during spawning migration. IX. Fat, protein and water in the major internal organs and cholesterol in the liver and gonads of the standard fish. j . Fisheries Res. Board Can. 17, 113-122. Idler, D. R., and Clemens, W. A. (1959). The energy expenditures of Fraser river sockeye salmon during the spawning migration to Chilko and Stuart Lake. Prog. Rept. Intern. Pacific Salmon Fisheries Comm., pp. 1-80. New Westminster, B.C., Canada. Idler, D. R., and MacNab, H. C. (1967). The biosynthesis of 11-ketotestosterone and 1lp-hydroxytestosterone by Atlantic salmon tissues in vitro. Can. j . Biochem. 45, 581-589.

66

WILLIAM S. HOAR

Idler, D. R., and Truscott, B. (1963). In vivo metabolism of steroid hormones by sockeye salmon. Can. J. Biochem. Physiol. 41, 875-887. Idler, D. R., and T,ruscott, B. ( 1966), Identification and quantification of testosterone in peripheral plasma of skate. Gen. Comp. Endocrinol. 7, 375-383. Idler, D. R., and Tsuyuki, H. (1959). Biochemical studies on sockeye salmon during spawning migration VIII. Androgen content of testes. J. Fisheries Res. Board Can. 16, 559-560. Idler, D. R., Schmidt, P. J., and Ronald, A. P. (1960). Isolation and identification of 11-ketotestosterone in salmon plasma. Can. J. Biochem. Physiol. 38, 105S1057. Idler, D. R., Bitners, I. I,, and Schmidt, P. J. (1961a). 11-ketotestosterone: An androgen for sockeye salmon. Can. J. Biochem. Physiol. 39, 1737-1742. Idler, D. R., Schmidt, P. J., and Biely, J. (1961b). The androgenic activity of 11ketotestosterone: A steroid in salmon plasma. Can. J. Biochem. Physiol. 39, 317-320. Idler, D. R., Freeman, H. C., and Truscott, B. (1964). Steroid hormones in the plasma of spawned Atlantic salmon (Su2mo salar) and a comparison of their determination by chemical and biological methods. Can. J. Biochem. 42, 211218. Ishii, S. (1963). Some factors involved in the delivery of the young in the topminnow, Gambusia afinis. J. Fac. Sci., Uniu. Tokyo, Sect. N 10, 181-187. Kerr, J. G. (1901). On the male genito-urinary organs of the Lepidosiren and Protopterus. Proc. Zool. SOC. London 1901( 2 ) , 484-498. Lagios, M. D., (1965). Seasonal changes in the cytology of the adenohypophysis, testes, and ovaries of the black surfperch, Embiotocu jacksmi, a viviparous percomorph fish. Gen. Comp. Endocrinol. $, 207-221. La Marca, M. J. (1964). The functional anatomy of the clasper and clasper gland of the yellow stingray, Urolophus Jumaicensis (Cuvier). J . Morpho2. 114, 303324. Lambert, J. G. D., and van Oordt, P. G. W. J. (1965). Preovulatory corpora lutea or corpora atretica in the guppy, Poecilia reticulatu. A histological and histochemical study. Gen. Comp. Endocrinol. 5, 693-694. Langer, W. F. (1913). Beitrage zur Morphologie der viviparen Cyprinodontiden. Morphol. Jahrb. 47, 193-307. Larsen, L. 0. (1965). Effects of hypophysectomy in the cyclostome, Lampetra fluviatilis (L.) Gray. Gen. Comp. Endocrinol. 5, 16-30. Larsen, L. 0. (1969). Effects of hypophysectomy before and during sexual maturation in the cyclostome, Lampetru fluviatilis (L.) Gray. Gen. Comp. Endocrinol. 12, 200-20&. Leatherland, J. F., and Pandey, S. ( l W 9 ) . Changes in the histology of the pituitary of the guppy Poecilia reticulatu Peters after treatment with Methallibure. Can. J. Zool. Submitted for publication. Leigh-Sharpe, W. H. (1920). The comparative morphology of the secondary sexual characters of elasmobranch fishes. J. Morphol. 34, 245-265. Leigh-Sharpe, W. H. ( 1921). The comparative morphology of the secondary sexual characters of efasmobranch fishes. J. Morphol. 35, 359-380. Leigh-Sharpe, W. H. (1922). The comparative morphology of the secondary sexual characters of elasmobranch fishes. 1. Morphol. 36, 191-243. Leigh-Sharpe, W. H. (1924). The comparative morphology of the secondary sexual characters of elasmobranch fishes. J. Morphol. 39, 55%577.

1.

REPRODUmON

67

Leigh-Sharpe, W. H. (1926). The comparative morphology of the secondary sexual characters of elasmobranch fishes. J . Morphol. 42, 307-358. Liley, N. R. (1968). The endocrine control of reproductive behaviour in the female guppy, Poecilia reticuhta Peters. Animal Behuuiour 16, 318-331. Lofts, B., Pickford, G. E., and Atz, J. W. (19643). Effects of methyl testosterone on the testes of a hypophysectomized cyprinodont fish, Fundulus hetmoclitus. Cen. Comp. Endocrinol. 6, 74-88. Lupo, C., and Chieffi, G. (1963). Oestrogens and progesterone in ovaries of the marine teleost Conger conger. Nature 197, 596. Lyngnes, R. (1936). Riickbildung der ovulierten und nicht ovulierten Follikel in Ovarium der Myxine glutinosa L. Skrifter Norske Vindenskaps-Akad. Oslo, I: Mat.-Naturv. KI. No. 4, 1-116. Mahadevan, G. (1940). Preliminary observations on the structure of the uterus and the placenta of a few Indian elasmobranchs. Proc. Indian Acad. Sci. B11, 1-44. Mann, T. ( 1960). Serotonin (5-hydroxytryptamine) in the male reproductive tract of the spiny dogfish. Nature 188, 941-942. Mann, T. ( 1964). “The Biochemistry of Semen and of the Male Reproductive Tract.” Methuen, London. Mann, T., and Lutwak-Mann, C. ( 1963). Comparative biochemical aspects of animal reproduction. Bull. A c d . Roy. Med. Belg. [7] 3, 563-593. Mann, T., and Prosser, C. L. (1963). Uterine response to 5-hydroxytryptamine in the clasper-syphon secretion of the spiny dogfish, Squalus acanthias. Biol. Bull. 125, 384-385. Marshall, A. J. (1960). Reproduction in male bony fish. Symp. Zool. SOC. London 1, 137-151. Marshall, A. J., and Lofts, B. (1956). The Leydig-cell homologue in certain Teleost fishes. Nature 177, 704-705. Matthews, L. H. (1950). Reproduction in the basking shark, Cetorhinus rnaximus (Gunner). Phil. Trans. Roy. SOC. London B234, 247-316. Matthews, L. H. (1955). The evolution of viviparity in vertebrates. Mem. SOC. Endocrinol. 4, 129-148. Matthews, L. H., and Marshall, F. H. A. (1956). Cyclical changes in the reproductive organs of the lower vertebrates. In “Marshall’s Physiology of Reproduction” (A. S. Parkes, ed.), 3rd ed., Vol. 1, Part 1, pp. 156-225. Longmans, Green, New York. Matthews, S. A. (1938). The seasonal cycle in the gonads of Fundulus. Biol. Bull. 74, 66-74. Matty, A. J. (1966). Endocrine glands in lower vertebrates. Intern. Rev. Cen. Exptl. Zool. 2, 43-138. Mellinger, J. (1965). Stades de la spermatogenbe chez Scyliorhinus caniculus ( L . ) : Description, donnkes histochimiques, variations normales et expkrimentals. Z . Zellforsch. Mikroskop. Anut. 67, 653-673. Metten, H. ( 1941). Studies on the reproduction of the dogfish. Phil. Truns. Roy. SOC. London B230, 217-238. Mohsen, M. ( 1961). Sur la prbence d’un copulateur interne, trks kvoluk chez Skiffia lermae ( Cyprinodonte Goodeidae) . Compt. Rend. 252, 3327-3329. Moser, H. G. (1967a). Seasonal histological changes in the gonads of Sebastodes paucispinis Ayers, an ovoviviparous teleost (Family Scorpaenidea) . J. Morphol. 123, 329-353. Moser, H. G. (1967b). Reproduction and development of Sebustodes paucispinis and

68

WILLIAM S. HOAR

comparison with other rockfishes off Southern California. Copeia 1967, No. 4, 773-797. Nandi, J. ( 1967). Comparative endocrinology of steroid hormones in vertebrates. Am. Zool. 7 , 115-133. Needham, J. ( 1942). “Biochemistry and Morphogenesis.” Cambridge Univ. Press, London and New York. Nelsen, 0. E. ( 1953). “Comparative Embryology of the Vertebrates.” McGrawHill, New York. Noumura, T. ( 1959). Induction of marsupium-like structure by testosterone in females of the pipefish, Syngnathus schlegeli. J. Fac. Sci., Univ. Tokyo, Sect. IV 8, 515520. Oguro, C. ( 1958). Effects of sex-hormones on the kidneys of the three-spined stickleback, Gasterosteus aculeatus ( L ,). J . Fac. Sci., Hokkaido Univ., Ser. VI 14,

45-50. Otuska, S. (1956a). Comparative studies on the pituitary hormones of vertebrates. 111. Effects of the pituitary hormones of the tunny, Makaira marlina on the gonads of mice. Endocrinol. lapon. 3, 59-60. Otuska, S. (1956b). Comparative studies on the pituitary hormones of vertebrates. IV. Effects of crude LH and FSH on the ovulation of hypophysectomized newts. Endocrinol. ]upon. 3, 181-185. Otuska, S. ( 1 9 5 6 ~ ) .On the extraction and bioassay of the follicle stimulating and luteinizing substances of the salmon. Endocrinol. Japon. 3, 272-279. Otuska, S. (1957). Comparative studies on the pituitary hormones in vertebrates. VI. Extraction of the follicle stimulating and luteinizing substances in domestic fowls and frogs. Endocrinol. Japon. 4, 219-226. Pandey, S. (1969a). Effects of hypophysectomy on the testis and secondary sex characters of the adult guppy Poecilia reticulata Peters. Can. 1. Zool. 47, (in press ) Pandey, S. (1969b). Effects of methyl testosterone on the testis and secondary sex characters of the hypophysectomized adult guppy Poecilia reticulata Peters. Can. J . Zool. 47, (in press ) . Pandey, S. (196%). The role of pituitary and gonadal hormones in the differentiation of testis and secondary sex characters of the juvenile guppy Poecilia reticuluta Peters. Biol. Reproduction ( in press ). Pandey, S. (1989d). Effects of Methallibure on the testes and secondary sex characters of the adult and juvenile guppy Poecilia reticuluta Peters, Submitted for publication. Patten, B. M. (1%3). “Human Embryology,” 2nd ed. McGraw-Hill, New York. Philippi, E. ( 1908). Fortpflanzungsgeschichte der viviparen Teleostier Gluridichthys januarius und G . decem-maculatus in ihrem Einfluss auf Lebenweise, makroskopische und mikroskopische Anatomie. Zool. Jahrb., Abt. Anat. Ontog. Tiere 27, 1-94. Pickford, G. E., and Atz, J. W. (1957). “The Physiology of the Pituitary Gland.” N.Y. Zool. SOC.,New York. Portman, A. ( 1948). “Emfiihrung in die vergleichende Morphologie der Wirbeltier.” Benno Schwabe, Basel. Prasad, R. R. (1948). Observations on the nidamental glands of Hydrolugus colliei, Raja rhina and Platyrhinoidis triseriatus. Copeia 1948, No. 1, 5457. Ramaswami, L. S . (1962). Endocrinology of reproduction in fish and frog. Gen. comp. Endocrinol. suppl. 1, 286-299.

.

1.

REPRODUmION

69

Ranzi, S. (1932). Le basi fisio morfologiche dello sviluppo embrionale dei Selaci. Parte 1. Pubbl. Staz. Zool. Napoli 12, 209-290. Ranzi, S. (1934). Le basi fisio morfologiche dello sviluppo embrionale dei Selaci. Parti I1 e 111. Pubbl. Staz. 2001.Napoli 13, 331J37. Ranzi, S. (1936). Ipofisi e gestazione nei Selaci. Atti Accad. Nazl. Lincei, Rend., Classe Sci. Fis., Mat. Nut. [6] 23, 365368. Ranzi, S. (1937). Ghiandole endocrine, maturita sessuale e gestazione nei Selaci. Atti Accad. Nazl. Lincei, Rend. Classe Sci. Fis., Mat. Nut. 161 24, 528-530. Read, L. J. (1968). Ornithine-urea cycle enzymes in early embryos of the dogfish Squalus suckleyi and the skate Raja binoculata. Comp. Biochem. Physiol. 24, 669-674. Reinboth, R. ( 1962). Morphologische und funktonelle Zweigeschlechtlichkeit bei marinen Teleostiern ( Serranidae, Sparidae, Centracanthidae, Labridae ) . 2001. Juhrb., Abt. Allgem. 2001.Physiol. Tiere 69, 405480. Romer, A. S. (1955). “The Vertebrate Body,” 2nd ed. Saunders, Philadelphia, Pennsylvania. Rosen, D. E., and Bailey, R. M. ( 1963). The poeciliid fishes (Cyprinodontiformes), their structure, zoogeography, and systematics. Bull. Am. Museum Nat. Hist. 126, 1-176. Rosen, D. E., and Gordon, M, (1953). Functional anatomy and evolution of male genitalia in poeciliid fishes. Zoologica 38, 1 4 7 . Schlernitzauer, D. A., and Gilbert, P. W. (1966). Placentation and associated aspects of gestation in the bonnethead shark, Sphyrna tiburo. 1. Morphol. 120, 219-231. Schmidt, P. J., and Idler, D. R. (1962). Steroid hormones in the plasma of salmon at various states of maturation. Gen. Comp. Endocrinol. 2, 204-214. Schmidt, P. J., Mitchell, B. S., Smith, M., and Tsuyuki, H. (1965). Pituitary hormones of the Pacific salmon. I. Response of gonads in immature trout (Salmo gairdnerii) to extracts of pituitary glands from adult Pacific salmon (Oncorhynchus). Gen. Comp. Endocrinol. 5, 197-206. Scrimshaw, N. S. (1944). Embryonic growth in the viviparous poeciliid, Heterandria formosa. Biol. BuU. 87, 37-51. Scrimshaw, N. S. (1945). Embryonic development in poeciliid fishes. B i d . Bull. 88, 233-246. Shann, E. W. (1923). The embryonic development of the porbeagle shark, Lamnu comubica. Proc. Zool. SOC. London 1923( l ) , 161-171. Shirai, K. (1964). Histological study on the ovipositor of the rose bitterling, Rhodezrs ocellutus. Bull. Fac. Fisheries, Hokkaido Univ. 14, 193-197. Simpson, T. H., and Wardle, C. S. (1967). A seasonal cycle in the testis of the spurdog, Squalus acanthias, and the sites of 3P-hydroxysteroid dehydrogenase activity. J. Marine Biol. Assoc. U.K. 47, 699-708. Simpson, T. H., Wright, R. S., and Gottfried, H. (1983a). Steroids in the semen of dogfish (squalus acanthias). 1. Endocrinol. 26, 48-98. Simpson, T. H., Wright, R. S., and Hunt, S. V. (1963b). Sex hormones in fish. Part 11. The oestrogens of Scyliorhinus caniculus. J. Endocriml. 26, 499-507. Simpson, T. H., Wright, R. S., anad Hunt, S. V. (1964). Steroid biosynthesis in the testis of dogfish (Squalus acanthias). J . Endocrinol. 31, 2 9 3 8 . Stanley, H. P. ( 1962). Morphological relationships between Sertoli cells and germinal cells in the testes of chondrichthyan fishes. Am. Zoologist 2, 561. Stanley, H., Chieffi, G., and Botte, V. ( 1965). Histological and histochemical obsenra-

70

WILLIAM S. HOAR

tions on the testis of Gobius paganetlus. Z. Zellforsch. Mikroskop. Anat. 65, 350-362. Stolk, A. ( 1951a). Histo-endocrinological analysis of gestation phenomena in the Cyprinodont Lebistes reticulatus Peters, 111. Changes in the pituitary gland during pregnancy. Koninkl. Ned. Akad. Wetenschap., PTOC. C54, 566-573. Stolk, A. ( 1951b). Histo-endocrinological analysis of gestation phenomena in the cyprinodont Lebistes reticulatus Peters. IV. The oocyte-cycle during pregnancy. Koninkl. Ned. Akad. Wetenshcap., PTOC. C54, 574-578. Sundararaj, B. I,, and Goswami, S. V. (1966). Effects of mammalian hypophysial hormones, placental gonadotropins, gonadal hormones, and adrenal corticosteroids on ovulation and spawning in hypophysectomized catfish, Heteropneustes fossilis (€%loch.). J. Exptl. ZOO^. 161, 287-296. Sundararaj, B. I., and Goswami, S . V. (1968). Effect of short- and long-term hypophysectomy on the ovary and interrenal of catfish, Heteropneustes fossilis ( Bloch) . J. Exptl. Zool. 168, 85-104. Sundararaj, B. I., and Nayyar, S. K. (1967). Effects of exogenous gonadotrophins and gonadal hormones on the testes and seminal vesicles of hypophysectomized catfish, Heteropneustes fossilis (Bloch. ). Gen. Comp. Endocrinol. 8, 403-416. Swift, D. R., and Pickford, G. E. (1965). Seasonal variations in the hormone content of the pituitary gland of the perch, Perca fluoiatilis L. Gen. Comp. Endocrinol. 5, 354-365. Te Winkel, L. E. (1943). Observations on later phases of embryonic nutrition in Squalus acanthias. J. Morphol. 73, 177-205. Te Winkel, L. E. (1950). Notes on ovulation, ova, and early development in the smooth dogfish, Mustelus canis. Biol. Bull. 99, 474486. Thiebold, J. J. (1954). dtude preliminaire de I’action des hormones sexuelles sur la morphogenese des voies gbnitales chez Scylliorhinus canicula L. Bull. Biol. France Belg. 88, 13&145. Tomlinson, N., McBride, J. R., and Geiger, S. E. (1967). The sodium, potassium and water content of the flesh of sockeye salmon (Oncorhynchus nerka) in relation to sexual development and starvation. J. Fisheries Res. Board Can. 24, 243248. Tortonese, E. ( 1950). Studi sui Plagiostomi. 111. La viviparita: Un fondamentale carattere biologic0 degli Squali. Arch. Zool. Ital. Torino 35, 101-155. Turner, C. D. ( 1966). “General Endocrinology,” 4th ed. Saunders, Philadelphia, Pennsylvania. Turner, C. L. (1919). The seasonal cycle in the spermary of the perch. J. Morphol. 32, 681-711. Turner, C. L. ( 1933). Viviparity superimposed upon ovo-viviparity in the Coodeidae, a family of cyprinodont teleost fishes of the Mexican plateau. J. Morphol. 55, 207-251. Turner, C. L. (1936). The absorptive processes in the embryos of Parabrotuh dentiens, a viviparous, deep-sea brotulid fish. I. Morphol. 59, 313-325. Turner, C. L. (1937a). The trophotaeniae of the Goodeidae, a family of viviparous cyprinodont fishes. 1. Morphol. 61, 495-523. Turner, C. L. ( 1937b). Reproductive cycles and superfetation in poeciliid fishes. Biol. Bull. 72, 145-164. Turner, C. L. (1938a). HistoIogical and cytological changes in the ovary of Cymatogaster aggregatus during gestation. J. Morphol. 62, 351373.

1. REPRODUCTION

71

Turner, C. L. (1938b). Adaptations for viviparity in embryos and ovary of Anableps anableps. 1. Morphol. 62, 323-349. Turner, C. L. (1940a). Pericardial sac, trophotaeniae, and alimentary tract in embryos of goodeid fishes. J. Morphol. 67, 271-289. Turner, C. L. (1940b). Adaptations for viviparity in jenynsiid fishes, J . Morphol. 67, 291-297. . pseudochorion, and follicular pseudoplacenta Turner, C . L. ( 1 9 4 0 ~ ) Pseudoamnion, in poeciliid fishes. J. Morphol. 67, 59-89. Turner, C. L. ( 1940d). Follicular pseudoplacenta and gut modifications in anablepid fishes. J . Morphol. 67, 91-105. Turner, C. L. (1947). Viviparity in teleost fishes. Sci. Monthly 65, 508-518. Turner, C. L. (1952). An accessory respiratory device in embryos of the embiotocid fish, Cymatogaster aggregatu, during gestation. Copeia 1952, No. 3, 146-147. van den Broek, A. J. P. (1967). Gonaden und Ausfuhrungsgange. In “Handbuch . pp. der vergleichenden Anatomie der Wirbeltiere” (L. Bok et al., eds. ), V O ~6, 1-154. Asher & Co., Amsterdam. van Overbeeke, A. P., and McBride, J. R. ( 1967). The pituitary gland of the sockeye ( Oncorhynchus nerka) during sexual maturation and spawning. J . Fisheries Res. Board Can. 24, 1791-1810. Vaupel, J. (1929). The spermatogenesis of Lebistes reticulatus. 1. Morphol. 47, 555-587. Vivien, J. ( 1958). Les glandes endocrines, In ‘‘Trait6 de zoologie” (P.-P. Grass& ed.), Vol. 13, Part 2, pp. 147s-1544. Masson, Paris. von Ihering, R. (1937). Oviducal fertilization in the South American catfish, Trachycorystes. Copeiu 1937, No. 4, 201-205. Wai, E. H., and Hoar, W. S. (1963). The secondary sex characters and reproductive behavior of gonadectomized sticklebacks treated with methyl testosterone. Can. J. Zool. 41, 611-628. Walvig, F. (1963). The gonads and the formation of the sexual cells. In “Biology of Myxine” ( A . Brodal and R. Fange, eds.), pp. 530-580. Oslo Press, Univ. Oslo. Wattenberg, L. W. ( 1958). Microscopic histochemical demonstration of steroid-3P01 dehydrogenase in tissue sections. J . Histochem. Cytochem. 6, 225-232. Weisel, G. F. (1949). The seminal vesicles and testes of Gillichthys, a marine teleost. Copeia 1949, No. 2, 101-110. Wiebe, J. P. (1967). The reproductive physiology of the viviparous sea perch, Cymatogaster aggregata Gibbons. Ph.D. thesis, University of British Columbia, Vancouver, Canada. Wiebe, J. P. (1968a). Inhibition of pituitary gonadotropic activity in the viviparous seaperch Cymatogaster uggregata Gibbons by a dithiocarbamoylhydrazine derivative (I.C.I. 33 828). Can. J . 2001.46, 751-758. Wiebe, J. P. (1968b). The reproductive cycle of the viviparous seaperch, Cymutogaster aggregatu Gibbons. Can. J . 2001.46, 1221-1234. . effects of temperature and daylength on the reproductive Wiebe, J. P. ( 1 9 6 8 ~ )The physiology of the viviparous seaperch, Cymatogaster aggregata Gibbons, Can. J. ZOOZ.46, 1207-1219. Wiebe, J. P. (1969a). Steroid dehydrogenases and steroids in gonads of the viviparous seaperch, Cymatogaster aggregata Gibbons. Gen. Cokp. Endocrinol. 12, 256-266. Wiebe, J. P. (1969b). Endocrine control of spermatogenesis and oogenesis in the

72

WILLIAM S. HOAR

viviparous seaperch Cymutogaster aggregata Gibbons. Gen. Comp. Endocrinol. 12, 267-275. Witschi, E. (1942). Hormonal regulation of development in lower vertebrates. Cold Spring Harbor Symp. Quant. B w l . 10, 145-151. Witschi, E. (1950). Gbndtique et physiologie de la differenciation du sexe. Arch. Anat. Microscop. Morphol. Exptl. 39, 215-246. Witschi, E. ( 1955). Vertebrate gonadotrophins. Mem. SOC. Endocrinol. 4, 149-165. Woodhead, P. M. J. (1968). Seasonal changes in the calcium content of the blood of Arctic cod. J. Marine Biol. Assoc. U.K.48, 81-91. Wotiz, H. H., Botticelli, C. R., Hisaw, F. L., Jr., and Ringler, I. (1958). Identification of estradiol-17p from dogfish ova (Squalus suckleyi). I . Biol. Chem. 231, 589592. W o k , H. H., Botticelli, C. R., Hisaw, F. L., Jr., and Olsen, A. G. (1960). Estradiol17p, estrone, and progesterone in the ovaries of dogfish (Squalus suckleyi). Proc. Natl. Acad. Sci. U.S. 46, 580583. Wourms, J. P. ( 1967). Annual fishes. In “Methods in Developmental Biology” (F. H. Wilt and N. K. Wessels, eds.), pp. 123-137. Crowell, New York. Yamamoto, K.,and Yamazaki, F. (1967). Hormonal control of ovulation and spermiation in goldfish. Gunma Symp. Endocrinol. 4, 131-145. Yamazaki, F. (1961). The effects of hypophysectomy on the ovary of the goldfish, Carossius auratus. Bull. Fac. Fisheries, Hokkaido Uniu. 12, 167-180. Yamazaki, F. (1962). Effects of hypophysectomy on the ovulation, oviposition and sexual behavior in the goldfish, Carmsius auratus. Bull. Fac. Fisheries, Hokkaido Uniu. 13, 39-46. Yamazaki, F. (1965). Endocrinological studies on the reproduction of the female goldfish, Carassius auratus L., with special reference to the function of the pituitary gland. Mem. Fac. Fisheries, Hokkaido Uniu. 13, 1-64. Yamazaki, F., and Donaldson, E. M. ( 1968a). The spermiation of goldfish (Carassius auratus) as a bioassay for salmon (Oncorhynchus tshawytscha) gonadotropin. Gen. Comp. Endocrinal. 10, 383-391. Yamazaki, F., and Donaldson, E. M. (1968b). The effects of partially purified salmon pituitary gonadotropin on spermatogenesis, vitellogenesis, and ovulation in hypophysectomized goldfish (Carassius auratus). Gen. Comp. Endocdnol. 11, 292-299. Yaron, Z. ( 1966). Demonstration of 3P-hydroxysteroid dehydrogenase in the testis of Tilapia mossambica ( Cichlidae, Teleostei). J . Endocrinol. 34, 127-128.

2 HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES N . R. LZLEY I. Introduction . . . . . . . . . . 11. Gonadal and Thyroid Hormones A. Gonadal Hormones in Male Fish . . . . . B. Gonadal Hormones in Female Fish . . . . C. Nonspecific Effects of Gonadal and Thyroid Hormones D. Thyroid Hormones and Migratory Behavior . . 111. Pituitary Hormones . . . . . . . . A. Prespawning Behavior . . . . . . . B. Spawning Behavior . . . . . . . C. Prolactin and Parental Behavior . . . . . IV. External Factors and the Endocrine System . . . V. Summary and Discussion . . . . . . . VI. Conclusion . . . . . . . . . . References . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

.

. . . .

. . .

. . .

. . .

73 75 76 88 90 93 94 94 97 100 102 104 109 110

I. INTRODUCTION

The significance of hormones in the regulation of reproduction presumably derives from the fact that in most animal species breeding occurs during restricted periods of the life history. The endocrine system provides a relatively slow acting link between the external environment and the internal state. The result is that reproductive behavior is nicely synchronized with the maturation of the gonads and the environmental conditions appropriate to breeding. The term “reproductive behavior” as used here encompasses a diverse range of activities involved in a propagative function. It includes sexual, parental, and nestbuilding behaviors. Agonistic and migratory behavior are also considered “reproductive” insofar as they are essential preliminaries to, or are involved in, reproduction. 73

74

N. R. LILEY

In comparison with studies involving birds and mammals the investigation of the hormonal regulation of reproductive behavior in fish is at a relatively primitive stage. Consequently it is frequently necessary to draw upon avian and mammalian examples to illustrate some of the principal features of the relationship between hormones and behavior. It is well established that in all classes of vertebrates the pituitary gonadotropins and gonadal steroids are the hormones most directly involved in the regulation of reproductive behavior. There is a number of ways in which the CNS-pituitary-gonad axis may regulate reproductive behavior; pituitary hormones may act directly on behavioral control mechanisms; pituitary hormones may stimulate the secretion of gonadal hormones which in turn regulate behavioral activity; both pituitary and gonadal hormones may act synergistically to regulate behavior. In addition, these hormones may in%uence behavior indirectly by their role in the development of secondary sexual characteristics. In its turn the CNS-pituitary-gonad axis is influenced by internal and external factors acting through the sense organs and brain. Studies involving birds and mammals, particularly those of Lehrman (1965) and Hinde (1965) with ring doves and canaries, respectively, have emphasized that the integration and coordination of the components of reproductive behavior depend upon a continuous interplay of external and internal factors; not only do hormones affect behavior but also behavior affects the endocrine state. Although the hormones of the gonads and anterior pituitary seem to play the major role in controlling reproductive processes, there is some evidence that neurohypophysial and thyroid hormones are also directly involved in the regulation of certain components of reproductive behavior in fish. Other hormones may affect reproductive behavior indirectly by their effects on general metabolism and growth and will not be considered here. Little is known of the details of the mechanisms by which hormones exert their effects upon behavior. This aspect will be considered in Section V. However, it is relevant to point out that hormone-induced changes in behavior differ widely in the time span involved, suggesting that the same or different hormones may act in a number of different ways. In general, hormones serve as relatively slow acting chemical links between the environment and an effector system. Thus there may be long-term seasonal changes over weeks or months, as in the annual cycles of very many animals. Changes may take place over hours or days, e.g., the switch from sexual to parental phases in birds (Hinde, 1965; Lehrman, 1965). In addition, there are some situations in which it appears as if hormones may be involved in short-term changes over periods of minutes

2.

HORMONES AND REPRODUCI’IVE BEHAVIOR IN FISHES

75

or seconds, e.g., the spawning of the female in response to courtship in the medaka, Oryzias Zatipes (Egami and Nambu, 196l), or the postcopulatory “after reaction” in the rabbit ( Kawakami and Sawyer, 1959). There is a danger of assuming that all reproductive behavior must be regulated by hormones. It is possible that only certain components of the behavior repertoire are under hormonal control, while other activities which are usually associated with reproduction may be causally linked to hormone-regulated activities without being themselves under the influence of the hormone state. Thus, there may be situations comparable to the causation of maternal responsiveness in rodents, reviewed by Richards ( 1967). In rodents, initiation of maternal responsiveness may depend on the endocrine state associated with parturition, while its maintenance is dependent on a nonendocrine mechanism stimulated by the presence of pups, Alternatively, certain behavior patterns associated with reproduction may occur outside the breeding season in other functional contexts. For example, agonistic behavior may be associated with reproduction, but it may also occur at times and places unrelated to breeding. Such activities are unlikely to be under the same hormonal control as seasonally changing behavior and may not be regulated by hormones at all. This emphasizes the need for care in the use of terms included under the general heading of reproductive behavior. Clearly, it would be misleading to refer to a behavior pattern obviously associated with reproduction during the breeding season as “reproductive” when it occurs out of the season in a different functional context. Beach (1967) makes a similar point with respect to studies of mammalian behavior. Lastly it is important to be aware of the possibility of a process in fish analogous to Beach‘s “corticalization of function” in the control of sexual activities in mammals ( 1964). Although hormones may be essential to the development of coordinated reproductive behavior, these activities may become less dependent on gonadal function after maturation and experience of breeding.

11. GONADAL AND THYROID HORMONES

It is well established that fish possess a CNS-pituitary-gonad axis similar in general pattern to that established for other classes of vertebrates (see chapter by Hoar, this volume). Furthermore, a wide range of steroids has been identified from male and female fish (see reviews by Gottfried, 1964; Hoar, 1965a; and chapter by Hoar). Surprisingly, there

76

N. R. LILEY

have been few unequivocal demonstrations for any fish that reproductive behavior is in fact induced or regulated by gonadal steroids, Many workers have demonstrated effects of exogenous steroids on sex determination and the development of secondary sexual characteristics in fish, but in most cases there is little mention of the effects of these treatments on reproductive behavior (see chapter by Yamamoto, this volume).

A. Gonadal Hormones in Male Fish The three-spined stickleback, Gasterosteus aculeatus, has been the subject of more intensive ethological and behavior-oriented endocrine investigations than any other species of fish and therefore provides a particularly suitable starting point. Reproductive behavior and the annual cycle are described by van Iersel (1953) and Baggerman (1957). There are two distinct races of the stickleback: the leiurus form which spends its entire life cycle in freshwater, the trachurus form which breeds in freshwater in spring and summer but migrates to the sea in the fall where it remains until a return migration to freshwater in the spring. There are a number of consistent morphological and ecological differences between the two races. Hagen (1967) found that the two races are almost completely reproductively isolated and questions their inclusion in a single species. In spring the males, which were previously associated in loose bisexual schools, become spaced out and acquire territories which they defend against other males. At the same time males acquire a distinctive appearance: red on throat and belly, dark back, and a blue iris. Internally, the kidney tubules begin to secrete mucus which is used in gluing materials used in nestbuilding. From this stage the male may go through one or more reproductive cycles lasting 20-30 days. Each cycle involves a series of overlapping phases or subcycles: nestbuilding, sexual, and parental (van Iersel, 1953; Baggerman, 1968; van Mullem, 1967). Throughout these cycles the male defends his territory against males and other intruders. Baggerman (1966) also traced cyclical changes in displacement fanning and frequency of comfort movements. Castration of the male in full breeding condition brings about a reversion to the nonbreeding condition-loss of nuptial coloration, a reduction in kidney tubuIes and cessation of sexual and nestbuilding behavior ( Ikeda, 1933; Baggerman, 1957; Hoar, 1962a,b). However, Hoar (1962a,b) working with the leiurus form noted that agonistic behavior persisted in fish castrated shortly before nestbuilding had commenced and held under long photoperiod (16L:SD). In contrast, agonistic be-

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

77

havior decreased along with other reproductive behavior in males placed under short photoperiod (8L:l6D) after castration. Baggerman ( 1966), working with the truchurus form, carried out a more detailed study of the effects of castration at different stages of the reproductive cycle. Castration of animals in full breeding condition, 9-19 days after the onset of breeding, resulted in a decrease in the frequency of swimming movements and a reduction in agonistic, sexual, nestbuilding, and displacement fanning behaviors to the level exhibited by intact animals after the end of the breeding period. In this study the onset of breeding is taken as the day a male built its first nest of the season in the presence of a male separated by a glass partition. Males castrated about 1 week before they were expected to build their first nest showed a decline in swimming activity and sexual behavior to normal postbreeding levels; nestbuilding and displacement fanning disappeared completely. In contrast, agonistic behavior increased to just below the level achieved by males in breeding condition and remained high for at least 3-4 weeks. This finding supports the suggestion by Hoar (1962a,b) that there is a change in the internal causation of agonistic behavior. Hoar (1965b) and Baggerman (1966) hypothesize that in the period before the onset of breeding the level of aggressive behavior is regulated by an increasing level of pituitary gonadotropic hormone. Gradually the mechanism underlying this behavior becomes less sensitive to gonadotropin and increasingly under the control of gonadal hormone ( s ) , Recent work by Wootton (1968) suggests that it is important to maintain the distinction between the leiurus and trachurus forms of the three-spined stickleback. Working with leiurus, Wootton finds that males castrated in the prenestbuilding stage and held under long photoperiod show a high level of aggressive behavior when presented stimulus males or models in a test situation. Fish castrated after they have built nests also maintain a level of aggressive behavior, not less than that of prebreeding fish but lower than control fish with nests present. This result contrasts with Baggerman’s finding (1966) that in trachurus castration of fish which have built nests results in a marked reduction in aggressive behavior. In spite of important differences in testing procedures, which make direct comparison difficult, it may be tentatively suggested that the two findings reflect real differences in the endocrine mechanisms underlying the reproductive cycle which presumably are related differences in the biology of the two forms. These findings concerning the control of agonistic behavior raise the semantic difficulties referred to on p. 75. Each worker has used different testing procedures which tend to emphasize different aspects of

78

N. R. LILEY

the agonistic response: Hoar (1962a,b) tested small groups of males and females in a situation in which few males succeeded in acquiring territories and building nests. Baggerman ( 1966) and Wootton ( 1968), using different procedures, tested isolated males which in many cases were able to build nests. Quite clearly, it is important to distinguish between agonistic behavior involved in defence of a nest and territory, and agonistic behavior without reference to a territory. In the latter case agonistic behavior may provide a mechanism of general dispersion and perhaps simply represent the reaction of a fish to an intruder within its “individual distance” (Hediger, 1950). Defense of a territory is undoubtedly an important component of reproductive behavior, whereas defense of “individual distance” is a much more general phenomenon and cannot be considered as exclusively reproductive in function. Therefore, although the agonistic motor patterns involved in both contexts are similar, it is not surpising that the underlying causal mechanisms appear to be somewhat different. It is evident from the preceding paragraph that in the stickleback investigations, and also in other similar studies, a final evaluation of the role of endocrine mechanisms in the control of agonistic behavior requires a considerable understanding of the functional significance of the behavior in different situations. This will in turn influence the choice of behavioral testing procedures. The acquisition of nuptial coloration and the onset of breeding behavior has been correlated with evidence of increased steroidogenesis. Gottfried and van Mullem ( 1967) identified testosterone, androstenedione, dehydroepiandrosterone, and progesterone in the testes of male sticklebacks. They estimated that “dominant” males, which had acquired territories, had 5 7 times the steroid level of fish caught at the same time but which had not been allowed to pass into full breeding condition. Replacement therapy has confirmed the role of androgens in inducing secondary sexual characters and reproductive behavior in the stickleback (Hoar, 1962a,b; Wai and Hoar, 1963). Methyltesterone treatment stimulates kidney tubule development and male nuptial coloration in gonadectomized adult males and females as well as intact juveniles of both sexes. However, juvenile fish failed to perform any nestbuilding behavior, and only 5 out of 86 gonadectomized females held under long photoperiod built nests. In these five fish most of the nestbuilding movements were present but occurred sporadically and resulted in poorly constructed nests, Masculinized females never showed sexual behavior toward introduced females. In the case of castrated males treated with methyltestosterone, Hoar (1962a) found that in those maintained under long photoperiod (16L:BD) 87.5% built nests, whereas only 57% of

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

79

males under short photoperiod (8L: 16D) built nests and in addition took more than twice as long to commence building. The apparent refractoriness of androgen-treated fish under short photoperiod leads Hoar to suggest that although reproductive behavior requires the gonadal hormone its full expression occurs only when gonadotropic activity of the pituitary is maintained at a high level by long photoperiod. Smith and Hoar (1967) investigated the endocrine involvement in displacement and parental fanning behavior. They found that low doses of methyltestosterone restored only the earlier stages of reproductive behavior in castrate males, i.e., digging, collecting and gluing. Pushing and displacement fanning appear only after treatment with higher doses of androgen. Castration early in the parental phase resulted in a decline in fanning even though live eggs were present in the nest. Smith and Hoar conclude that displacement and parental fanning are regulated by a testis hormone. Baggerman (1966) makes the important point that in interpreting the effects of gonadectomy and hormone treatment it may be difficult to distinguish the direct from the indirect effects of a hormone. It appears that sexual, nestbuilding, and parental behaviors are correlated with a high level of gonadal hormone, suggesting that the hormone is playing a causal role. The same could be true of displacement fanning, changes in swimming activities, and frequency of comfort movements. Alternatively, these latter activities may occur as a result of the arousal and perhaps interaction of the mechanisms underlying sexual, agonistic, and parental behavior. Baggerman (1957, 1959) considered the possibility that gonadal hormones might be involved in the prespawning migration into freshwater of the trachurus form of the stickleback. Using salinity preference as a measure of migration disposition, Baggerman found that gonadal maturity coincides with a change in preference from saltwater to freshwater, whereas gonad regression was correlated with a change in preference from freshwater to saltwater, But, since similar changes in salinity preference can be induced in gonadectomized fish by manipulation of the photoperiod and temperature, it is clear that gonadal hormones are not directly involved. Baggerman (1962) did, however, find the testosterone treatment resulted in an increase in locomotory activity of a type associated with migrating fish (further discussion in Section 11, C). Poeciliids have been used in numerous studies in which attention has centered on the effects of steroid hormones on morphological secondary sexual characters and on sex determination (see Pickford and Atz, 1957, and chapter by Yamamoto, this volume, for full reviews). In a few cases the effects of these treatments on behavior are mentioned.

80

N. R. LILEY

Cohen (1946) and Tavolga (1949) found that treatment with pregnenolone induced male characters in maturing, genetically female platyfish, Xiphophorus ( =Platypoecilus ) muculatus. These sex-reversed “males” showed male courtship patterns and attempted to copulate but, according to Tavolga ( 1949), courted less vigorously than genetic males under the same treatment. In contrast, genetic males treated with estradiol benzoate are said to have behaved like females and were pursued by normal males (Cohen, 1946). Hildemann (1954) was able to induce male courtship behavior in female platyfish by treatment with methyltestosterone. Noble and Borne (1940) report without giving details that females of X. helbri treated with testosterone propionate rose in the pecking order until a reversal in sexual behavior occurred. As the sword and gonopodium developed, the sex-reversed “male” made attempts to copulate. Treatment of females with estradiol dipropionate failed to induce any change in social rank. These workers found that gonadectomized males and females maintained rank position for several months. They concluded that male sex hormone brought about a rise in rank of the female only by reversal in sex. Laskowski (1954) treated both immature and adult genetic females of Xiphophorus ( =Platypoecilus ) variatus with testosterone. The females performed what is referred to as phase 1 courtship which is characterized by a lateral display. Even after several weeks of treatment during which gonopodia developed in the young fish, the genetic females failed to perform the zigzag dance (phase 2) and the mating attempt (phase 3). The question arises as to whether phase 1 behavior should be regarded as a courtship movement. Males perform similar activities in establishing dominance hierarchies. Laskowski ( 1954) points out that in a community of untreated females a definite hierarchy is set up. Females introduced into such a community show considerable phase 1 type of interaction with other females over a period of several hours until dominance relations become established. Androgen-treated adult females differed from untreated fish chiefly in that phase 1 behavior persisted for several days. Evidently the movements which initiate a courtship sequence are similar to those involved in agonistic behavior, and perhaps sex recognition depends upon the response of the female to an initial threat by the male. Thus, it appears that phase 1 activities should be regarded as agonistic behavior common to both male and female. Under normal circumstances agonistic behavior occurs more frequently among males than among females. Androgen treatment of genetic females increases the

2.

HORMONES AND REPRODUcIlvE BEHAVIOR IN FISHES

81

frequency of agonistic behavior and in this respect has a masculinizing effect but is relatively ineffective in inducing movements which are functionally restricted to the sexual context. Gonadectomy of the male platyfish, Xiphophorus muculatus, resulted in a decrease in frequency or duration of copulation thrusts, swings, and sidling, whereas pecks, backs, and approaches remained the same (Chizinsky and Aronson in Aronson, 1959). An interpretation suggested by the previous discussion of Laskowski’s work is that those activities which persist after castration are also components of agonistic behavior common to both sexes. Clemens et al. (1966) treated guppies with testosterone from birth to 60 days. After treatment they found a very marked increase in the proportion of males-in some cases the males outnumbered females 9: 1. However, when paired with females only 14%of these males, including sex-reversed genetic females, sired young, even though in many cases they showed full male coloration and yielded viable sperm on stripping. Clemens et at. suggest that the failure to breed resulted mainly from a behavioral deficiency which was not simply the result of a lack of endogenous androgen. A possibility, suggested by the present author’s work with the guppy (Liley, 1966, p. 160), is that males failed to respond to females because of the prolonged isolation from responsive females. Working with the medaka, Oryzias latipes, Okada and Yamashita (1944) confirmed by castration and testosterone treatment that the male secondary sexual characteristics are under hormonal control. Testosterone treatment of females or implantation of a testis results in masculinization and the performance of male behavior including pursuit of normal females, Yamamoto (1962, and chapter, this volume) has confirmed that complete functional sex reversal may be achieved with androgen treatment of genetic female medakas. Unfortunately, there has been no detailed camparison of the behavior of normal males and sex-reversed genetic females. Tavolga (1955) found that castration of male gobies, Bathygobius soporator, abolished aggressive behavior toward an introduced male. Instead the operated males failed to discriminate between males and females, and between gravid and nongravid females, and courted all equally. Spawning behavior of castrated males with gravid females appeared to be normal. The male brooded the infertile eggs which resulted from the spawning. Tavolga (1956) also investigated the stimulus situation involved in prespawning behavior of goby males. Intact males readily distinguished gravid females from nongravid females and males. However, if stimulus fish were presented in glass containers males failed to discriminate and courted all females and nonaggressive males equally. This suggests that

82

N. R. LILEY

normally the males recognize the gravid female on the basis of chemical cues. Tavolga (1956) demonstrated that a substance present in the ovary of gravid females, and presumably released through the genital pore, will elicit a vigorous courtship response in a male, even in the absence of a stimulus fish. Tavolga (1956) suggests that in the male goby a gonadal hormone affects the sensitivity of the olfactory organs perhaps by making the male differentially sensitive to chemical factors released by the gravid female. Castration may lower the sensitivity threshold and permit chemicals other than the ovarian fluid to stimulate courtship behavior. Cichlids and anabantids have been the subjects of many intensive ethological studies ( e.g., Baerends and Baerends-van Roon, 1950; Forselius, 1957; see also chapter by Baerends, Volume VI). A few species have been investigated from the point of view of the endocrine mechanisms involved in their complex reproductive processes. Noble and Kumpf (1936) reported that males of Hemichrornis bimaculatus performed typical courtship, fertilization, and brooding movements for 202 days after castration. The males developed nuptial coloration and genital tubes at each spawning. After a normal spawning both parents are inhibited from eating their young and perform brooding behavior which consists of collecting and guarding the young. Noble et at. (1938) found that brooding in response to donated young could be induced by treating nonbrooding fish, both male and female, with a variety of hormones. Positive responses were only obtained with fish which had had previous experience of brooding and/or spawning. Thirteen fish, including four castrates, began to brood normally after treatment with corpus luteum extracts ( source not given: presumably rich in progesterone). Proluton and prolactin were also highly effective in inducing brooding. Anterior pituitary extract, fresh fish pituitary, thyroxin, desiccated thyroid, and testosterone were also effective in a smaller proportion of fish treated. Nearly half the fish given control phenol injection also began brooding. These results are difficult to interpret but serve as a useful warning of ( a ) the role of experience in determining the outcome of hormone therapy, and ( b ) the possibility that a wide variety of agents may produce the same behavioral effects: In this case it seems likely that some or all the agents were exerting their effects indirectly, perhaps by stimulating pituitary secretory activity. Aronson (1951) found that castration of male Tilapia macrocephala resulted in a reduction of the genital tubes and loss of the yellow coloration of the operculum. Castrated males continued to dig nests as frequently as intact males. The effects of castration were reversed by testosterone treatment (Levy and Aronson, 1955). Previously, Aronson

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

83

and Holz-Tucker (1947) had shown that testosterone treatment of ovariectomized and intact females resulted in the growth of the genital papilla and the acquisition of male coloration on the operculum. Working with another species, Tilapiu mossambica, Clemens and Inslee ( 1968) obtained functional sex reversal in genetic females by treating them with methyltestosterone for the first 69 days of life. The sex-reversed fish exhibited male mating coloration and nestbuilding behavior when placed with ripe females. Aronson ( 1959) compared the reproductive behavior of three males of Aequidens latifrom before and after gonadectomy. Up to 6 weeks after castration all elements in the mating pattern were still present, most of these showing little change in frequency of occurrence or duration from their own preoperative levels. There was an increase in nest passing and several other items of behavior after castration. The increase was particularly marked for the period 1 day after spawning. On the other hand, there was a noticeable decline in nestdigging after castration. Noble and Kumpf report (1936), without giving details, that gonadectomized males of the anabantid, Betta splendens, perform courtship movements, whereas gonadectomized females do not. These authors also found (1937) that a small number of ovariectomized B. splendens females developed a testis and in such cases acquired male coloration and behavior. Also working with anabantids, Forselius (1957) reports that sterile Colisa hbiosa x lalia hybrids acquired male secondary sex characteristics and exhibited “migratory” and reproductive behavior. Histological examination of a large number of the hybrids failed to reveal any trace of gonadal tissue. However, evidence that androgens are involved in the control of male characteristics is provided by the finding (Forselius, 1957) that testosterone propionate treatment of females of C. lalia induced male coloration and, in some cases, nestbuilding and defense, Johns et al. (1969) found that some males of the blue gourami, Trichogaster trichopterus, performed reproductive behavior after castration. Of a total of 16 males castrated, 11 failed to build bubble nests or show any sexual behavior. The other five built nests and “spawned within 7 days of being paired with females. Apart from a relatively long delay between introduction of females and actual spawning, the reproductive behavior of these five did not differ from that of control fish. Castrate males which had spawned (but not fertilized the eggs) readily accepted fertile eggs and performed parental behavior for several days. Fertile eggs given to nonspawning castrates were eaten. Castrate males placed with intact or other castrate males performed agonistic behavior until a dominance relationship was established. Agonistic behavior of castrates

84

N. R. LILEY

did not obviously differ qualitatively or quantitatively from that of intact males. The dorsal fin of the male gourami is longer and more pointed than that of the female. The dorsal fin of castrated males became shorter and rounded, more like that of the female. Significantly, however, the dorsal fins of the castrates which spawned were more like those of intact males than were the fins of nonspawning castrates. Castrated males treated with methyltestosterone developed the long pointed dorsal fin and in all cases spawned when paired with ripe females. These results indicate that sexual behavior and the characteristic male dorsal fin are under gonadal hormone control. Of the five castrates which succeeded in spawning before testosterone treatment only one showed identifiable regenerating testis. The similarity in behavior and length of dorsal fin in the testosteronetreated fish and those which spawned after castration suggests that in the latter there was either undetected testicular tissue or an extragonadal source of androgenic steroid. Johns et al. (1969) conclude that their results emphasize the need for considerable caution in the interpretation of castration experiments and demonstrate the importance of independent checks on the presence of androgen before concluding that reproductive behavior is not dependent on the direct action of testicular hormone. Machemer and Fiedler ( 1965) investigated the hormonal involvement in nestbuilding of another anabantid, the paradise fish, Macropodus opercularis. Two out of eight intact males given only methyltestosterone showed an increase in building tendency but, apparently because of inadequate mucus production, building was incomplete. Testosterone in combination with prolactin ( which increases mucus production ) resulted in nestbuilding in three males and two females. Machemer and Fiedler do not appear to have run a control series, and it is important to note that, in at least some species of anabantid, isolated males begin nestbuilding apparently spontaneously ( Forselius, 1957; own observations of Betta splendens and Trichogaster trichopterus) . Greenberg (1947) and Hale (1956) note that in the green sunfish, Lepomis cyanellus, immature and mature fish of both sexes perform agonistic behavior, although defense of a nest is limited to sexually mature males. There was no decrease in agonistic behavior after gonadectomy in a small number of males and females placed in a test situation (Hale, 1956). Smith (1967) found that castrated males of Lepomis megalotis and L. gibbows maintained high levels of aggressive behavior but failed to build nests. Treatment with testosterone restored nestbuilding but, in the case of L . megalotis placed in large experimental pools, the males did not cluster together to the same extent as sham-operated fish.

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

85

Wiebe (1967) demonstrated that in the male of the viviparous seaperch, Cymatogaster aggreegata, the occurrence of full reproductive behavior coincides with maximum gonadal development and histochemically demonstrated steroidogenesis. The gonads of fish in winter condition (October) matured in fish placed under long photoperiod ( 16L:8D) and high temperature (20.C). Such fish also acquired the secondary sexual structures and performed sexual behavior, including chasing, heading-off the female, courtship dance, lateral quiver, and mating attempt. Castrated fish under the same conditions did not acquire the sexual structures. However, they still performed some of the elements of sexual behavior: chasing and mating attempt, but without the darkening characteristic of full reproductive behavior. Castrated males treated with methyltestosterone acquired the secondary sexual characteristics and showed an increase in sexual behavior, although the level of sexual behavior did not reach that of intact males under the same environmental conditions. Significantly, chasing and mating attempts which persisted after castration occur in intact fish with reduced vigor and frequency throughout the year. Thus, it appears that certain components of the sexual behavior occur only at times at which insemination is known to occur. These activities disappear after gonadectomy and reappear as a result of testosterone therapy. In contrast, chasing and mating attempt appear to be to some extent independent of gonadal control and for much of the year are not obviously associated with a reproductive function. Surprisingly little is known of the role of gonadal hormones in the reproductive behavior of salmonids. Jones and King ( 1952) castrated four adult males of Salmo salar. In three castrated males treated with testosterone propionate sexual behavior was partially restored: males followed females and one male quivered. The one castrate which was not treated with androgen showed no interest at all in the females. In all species of fish studied, male secondary sexual characteristics and at least some components of reproductive behavior are under gonadal hormone control. In a number of cases it has been claimed that complete reproductive behavior persists in castrated fish. The most striking example of this is provided by Tavolga’s study (1955) of Bathygobius soporator. Others involve cichlids and anabantids. However, Johns e t al. (1969) have reason to suggest that in the gourami the persistence of reproductive behavior may result from the presence of unidentified regenerated testis or the existence of an extragonadal source of androgen. In addition to the above there are several examples of situations in which behavior usually classified as “reproductive” persists after castration. Most of these examples apply to behavior involved in a nonreproductive context out of the breeding season and thus, not surpris-

86

N. R. LILEY

ingly, these activities turn out to be more or less independent of gonadal control. This applies particularly to agonistic behavior which occurs in very many species in both reproductive and nonreproductive situations and frequently plays a prominent role in the preliminary encounters between the two sexes.

B. Gonadal Hormones in Female Fish The hormonal regulation of female behavior has received far less attention than that of the male. This is at least partly because the female is in most cases a more passive partner. Even where nestbuilding and parental care occur, in many species it is the male which is responsible. Thus the behavior of the female is usually difficult to detect or quantify, with the result that it may be difficult to assess the effects of gonadectomy or hormone treatment. Numerous investigators have applied estrogen treatment to male and female poeciliids. Most of this work has been reviewed by Pickford and Atz (1957) and need not be detailed here. Indeed most investigations have been concerned with the effects of estrogens on the gonads and morphological structures. In general, estrogens suppress the testis and male characteristics in developing genetic males. When treatment is withdrawn the testis and male gonopodium and coloration develop. None of these investigations have given any clear indication that estrogenlike hormones are involved in the regulation of female behavior. Ovariectomy of poeciliid females has been performed in a small number of investigations. Noble and Kumpf (1936) and Ball (1960) found that females of Xiphophorous helleri and Poecilia ( =Mollinesia) sp., respectively, remained sexually attractive to males after ovariectomy. However, there is no indication that an attempt was made in either of these studies to assess female responsiveness to male courtship. It is well known that poeciliid males will direct persistent and vigorous courtship toward unresponsive females, Liley (1966) has described the sexual response of the female guppy, Poecilia reticulata, and finds that nonvirgin females go through a cycle of receptivity which can be correlated with the cycle of brood production. Females respond most readily and are more likely to accept copulation in the few days following the birth of a brood of young. Virgin females are more persistently receptive until fertilization of the eggs occurs. Ovariectomy of virgin females results in an initial decline in receptivity over a period of 5-6 days. However, nonvirgin females tested several

2.

HORMONES AND REPRODUCITVE BEHAVIOR I N FISHES

87

weeks after gonadectomy still occasionally responded to male courtship. It was concluded that sexual behavior may still occur in the absence of gonads, but that in intact fish the ovary exerts a regulatory and perhaps stimulating effect upon a more direct pituitary or neural control. The importance of the gonadal contribution is emphasized by the finding ( Liley and Donaldson, 1969) that treatment of hypophysectomized females with the gonadotropin fraction of salmon pituitary extract brought about a restoration of sexual responsiveness only after the gonad recovered from its regressed state resulting from hypophysectomy. In the medaka, Oqzias Eatipes, Okada and Yamashita (1944) found that ovariectomy of females and estriol treatment of adult males had no effect on secondary sexual characteristics. However, Yamamoto (1962), and Yamamoto and Matsuda (1963) have established that functional sex reversal of genetic males may be induced by treatment of young with estrogens. Sex-reversed genetic males are fully functional as females and breed with true males. Unfortunately, there has not been a detailed behavioral comparison of sex-reversed genetic males with normally developing females. Little is known of the role of gonadal hormones in the sexual response of the medaka. Egami (1955) found that treatment of females with estradiol benzoate brought about a reduction in the occurrence of oviposition. This result may have been caused by the suppression of pituitary gonadotropin secretion and the consequent regression of the ovaries. Wai and Hoar in an unpublished study (1968) compared the effects of gonadectomy on the aggressive behavior of male and female threespined sticklebacks maintained under long photoperiod. In one series of experiments involving the leiurus form, aggressive behavior was measured by placing two fish of the same sex in a tank and scoring the number of attacks during a series of 5 min observations over a period of several weeks. The results, Table I, confirm earlier findings (Hoar, 1962a,b) that prenestbuilding agonistic behavior of males is hardly affected by castration. Intact females attacked each other at about onethird of the rate shown by intact males. Ovariectomized females performed twice as many attacks as intact females. These results suggest that ovarian secretions normally suppress aggressive behavior to some extent. Wai and Hoar noted that most of the intact females which showed high levels of aggressive behavior were immature and therefore, presumably, estrogen levels were low. The fact that gonadectomized females were less aggressive than gonadectomized males suggests that there are genetic sex differences in the levels of aggression, or that the presence of gonadal hormones earlier had some persistent effects. Forselius (1957)

88

N. R. LILEY

Table I Mean Number of Attacks per 5 min in Pairs of Intact and Gonadectomized Male and Female SticklebacksoJ' Treatment

No. of pairs of fish

Mean No. of attacks/5 min

Intact male Gonadectomized male Intact female Gonadectomixed female

45 48 48 44

33 28 11 21

Unpublished data from Wai and Hoar (1968). Means are for two or three 5-min observations per week in the fourth and fifth weeks after gonadectomy in four experimental series.

found evidence that estrone injections resulted in a marked reduction in aggressive behavior in maIes of Colisa l a b . In contrast to their findings with castrated males, Noble and Kumpf (1936) found that ovariectomy of females of Hemichromis bimuculutus resulted in total loss of reproductive behavior. Injection of ovarian extract (source not stated) restored most of the sexual behavior. As noted earlier, Noble et al. (1938) found that brooding behavior could be induced in experienced females by treatment with a variety of gonadal and pituitary hormones. Of these agents corpus luteum extract proved to be one of the most effective. Ovariectomy of Tilupia macrocephula brought about a reduction of the genital tube and the reappearance of a silvery immature condition of the operculum ( Aronson and Holz-Tucker, 1947). Estradiol treatment of ovariectomized fish induced the growth of the genital tube but had no effect on the silvery operculum. Aronson found (1951) that the majority of intact females built nests prior to spawning. Ovariectomy resulted in a reduction in nestbuilding to about the level of intact or castrate males, that is, nests were build in about 11%of the pairings. Aronson notes (1957) that in Tilupiu mucrocephala completion of ovulation is marked by the occurrence of nest-passing, an activity closely associated with spawning. The close temporal relationship between ovulation and spawning, which has been noted by several workers, will be considered further in Section 111, B. Noble and Kumpf (1936) mentioned briefly their finding that Bettu splendens females no longer performed sexual behavior after ovariectomy. This contrasts with the results of castration of males. Johns et al. (1969) considered the effects of ovariectomy in the blue gourami, Trichogaster trichopterus. Normally the female initiates the spawning sequence by approaching and butting a dark, nestbuilding male. None of

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

89

the eight ovariectomized females made sexual responses when exposed to males in reproductive condition. Two ovariectomized females treated with estrone also failed to respond. In addition, ovariectomized females were evidently less attractive to males than were intact females: Males directed significantly fewer contacts with their long pelvic rays toward ovariectomized females. There is histological and behavioral evidence (Scharrer et al., 1947) that the pelvic rays are chemosensitive. Thus the effect of ovariectomy on female “attractiveness” suggests that the ovary is responsible directly or indirectly for the elaboration of a specific chemical detected at the female body surface by the male. Such a secretion might provide the means by which males discriminate between males and females and between gravid and nongravid females. It should be noted however that Picciolo (1964) was unable to find behavioral evidence that male anabantids responded to chemical stimuli during reproductive behavior, although he did find that amputation or cauterization of the pelvic fins reduced reproductive success. Ovarian involvement in the production of a chemical attractant has been well established by Tavolga (1956) in his work with Bathygobius soporator. The chemical has not been identified but is evidently only produced in effective quantities in gravid fish ready to spawn. Amouriq (1964, 1965a,b) has concluded that a substance produced in the ovaries of female guppies which elicits increased male activity, is in fact an estrogen (discussed in Section 11, C ) , Ball (1960) reviewed much of the work concerning the hormonal control of ovipositor growth in the bitterling, Rlaodeus amarus. Early work had suggested that ovipositor growth is stimulated by ovarian hormones. However, Bretschneider and de Wit (1947) and others found that ovipositor growth could be induced by a wide variety of chemical and physical stimuli. Ball suggests that although there is no reason to suppose the normal ovipositor growth is not under ovarian steroid control, in many experimental situations it was mainly a response to internal changes involved in a stress reaction. Shirai (1962, 1964) finds a clear correlation between ovipositor growth and ovarian condition in the Japanese bitterling, R. ocellatus. He notes (1964) that there is a period of slow steady growth in the prebreeding season and a periodic fluctuation in length during the breeding season. Shirai (1964) speculates that there may be two hormone factors involved: an estrogenlike hormone, which stimulates the long-term growth, and another factor involved in the short-term cyclical changes during the breeding season. In this last phase fish show maximum length of ovipositor at the time that ripe eggs are extruded into the ovarian lumen, and it is at this time that oviposition occurs. If there

90

N. R. LILEY

is no mussel present, maturation, ovulation, and ovipositor growth are all inhibited. The clear correlation between ovipositor growth, ovulation, and oviposition suggests that gonadal hormones are in some way involved in the regulation of the spawning act. Shirai (1962) makes an alternative suggestion that a physical stimulus caused by the extrusion of mature eggs into the lumen brings ovipositor growth under nervous control. Apart from an obvious correlation between ovarian maturation and spawning readiness, there is no direct evidence that gonadal hormones are involved in the regulation of spawning behavior in female fish. There are relatively few reports of the experimental use of steroid estrogen, none of these has revealed behavioral responses to hormone treatment. Sundararaj and Goswami ( 1966) treated sexually mature but unovulated female catfish, Heteropneustes fossilis, with estradiol benzoate, testosterone propionate, and progesterone 3 days after hypophysectomy. These steroids failed to induce ovulation or spawning behavior. In contrast, three corticosteroids induced ovulation and, in some cases, oviposition in the absence of males. Thus, it appears that corticosteroids can act as ovulating agents and proved almost as effective in this regard as mammalian luteinizing hormone ( LH) . Furthermore, hypophysectomy of the experimental fish ruled out the possibility that the steroids exerted their effect by an action involving the pituitary. Sundararaj and Goswami ( 1966) direct attention to investigations which have shown that in the Pacific salmon the 17-hydroxycorticosterone titer in the blood progressively increases in fish as they migrate from the sea to spawning grounds. They speculate that perhaps the corticosteroids are directly involved in ovulation and spawning in fish. Changes in corticosteroid levels in migrating and spawning fish have usually been interpreted, e.g., Chester Jones and Phillips (1960) and Fagerlund (1967), as a response to activity and stress. However, considering that treatment with gonadal steroids has been so conspicuously unsuccessful in inducing reproductive behavior in female fish it seems worthwhile considering the possibility that corticosteroids may be in some way involved. C. Nonspecific Effects of Gonadal and Thyroid Hormones So far this review has been concerned with hormones which regulate the performance of behavior directly involved in reproduction. There are a number of indications that gonadal and thyroid hormones may have more general effects on behavioral activity. Stanley and Tescher (1931) reported a considerable increase (400%) in the locomotory activity of

2.

HORMONES AND REPRODUCI’IVE BEHAVIOR IN FISHES

91

goldfish fed on ground mammalian testicular substance. This effect became marked within hours of feeding and continued for at least 24 hr. Hoar et al. (1955) found that goldfish immersed in solutions of thyroxine, testosterone, or stilbestrol showed greater locomotory activity and were more responsive to electrical stimulation than untreated fish. Forselius reports (1957), without giving details, “An increase of appetitive behavior and general locomotory activity” in the anabantid, Colisa lalia, after injection of thyroxine, testosterone, and chorionic gonadotropin. Thyroxine treatment resulted in increased locomotory activity and jumping in guppies, Poecilia reticutata ( Sage, 1968). Hoar et al. (1952) showed that chum salmon fry treated with methyltestosterone and synthetic thyroxine became more active and showed less marked schooling than control or thiourea-treated fry. Coho and sockeye salmon yearlings treated with thyroxine, testosterone, or stilbestrol, and yearling sockeye treated with estrogens, showed a decrease in the time required to make a standard response to flowing water (Hoar et al., 1955). Van Iersel ( 1953) found that male three-spined sticklebacks (trachums form) treated in midwinter with testosterone propionate began “fluttering” (persistent swimming against the glass walls of the aquarium) which did not occur in control fish at that time. Iersel regards this fluttering as an expression of migratory behavior. Both androgen and thyroid stimulating hormone ( TSH ) treatments resulted in significant increases in swimming movements and bouts of fluttering in castrated male sticklebacks ( Baggerman, 1962). Baggerman suggests that TSH produced this effect indirectly by stimulating the secretion of thyroid hormones. A different mechanism of hormone action is suggested by Amouriq (1964, 1965a)b) who found that male guppies, Poecilia ( =Lebistes) reticulata, placed in water which had previously held females, exhibited a marked increase in their locomotory activity; water which had previously held males was ineffective. Amouriq added extracts of skin, intestine, and ovary of females to the aquarium water and found that only the latter induced a significant increase in male activity. An estrogen, hexestrol dipropionate, when added to the water, resulted in a marked increase in male activity. More recently Amouriq ( 1967) has established that the range of concentrations of the steroid affecting locomotory activity is quite narrow: 0.025 mg/ml evokes hyperactivity; 0.05 mg/ml results in hypoactivity; concentrations below the first and above the second have no effect, Amouriq (196%) concludes that an ovarian hormone induces hyperactivity in males and maintains female attractiveness to males. Thus, he is suggesting that the hormone act as a pheromone in addition, presumably, to its endocrine role in the female.

92

N. €3. LILEY

Earlier studies, Breder and Coates ( 1935), Clark and Aronson ( 1951) , Baerends et al. ( 1955), and Liley (1966), as well as Amouriq (1965a), have emphasized the role of visual stimuli in guppy courtship. Although not tested experimentally there has been little evidence to indicate that chemical stimuli are involved. On the basis of data provided it is difficult to appreciate the functional significance of the response described by Amouriq. The response was only obtained using water from a 2liter container which had held as many as 10 females for 24 hr ( Amouriq, 1965a). Maximum locomotory effects occurred 5 hr after the start of the addition of ovarian extract (Amouriq, 1965b). The effects of hexestrol remained slight for 6-7 hr from the start of treatment after which there was a marked increase in male activity. This suggests that the hormone may not be acting directly as an external stimulus but is being taken up through the skin and gills of the male and acting on the central nervous system or sense organs, perhaps increasing the susceptibility of the fish to external stimuli and thereby bringing about an increase in locomotory activity. Because of the delay in obtaining a locomotory response, the relatively high density of females required to obtain effective concentrations, the mobility of both male and female, and the lack of obvious contact behavior in courtship (compared with a species such as Poecilia uiuipara, see Liley, 1966) it is difficult to see how the chemical released by the female could play an important role as a sexual attractant or even by simply increasing the searching behavior of the male. It is of course possible that more refined behavioral measures may reveal femaleoriented male responses at lower concentrations and with less delay. This work with several species of fish indicates that hormones can affect behavior by bringing about an increase in general activity. However, it is not clear whether the hormones influence activity by an effect on general metabolism or whether they act directly on neural mechanisms underlying the behavioral responses. Evidence that steroid and thyroid hormones may directly affect CNS and sense organ states is provided by a number of electrophysiological studies. Several investigations involving goldfish have revealed that various steroids and thyroid hormones exert differential facilitatory and inhibitory effects on the excitability of the olfactory bulb and regions of the fore and midbrains (Oshima and Gorbman, 1966a,b, 1968; Hara, 1967; Hara et al., 1965, 1966). Godet and DupC (1965) found that in Protopterus an active thyroid or thyroxine treatment facilitates responsiveness to olfactory stimuli at the level of the forebrain. Although little is known of the behavioral significance of most of the above neurophysiological findings, they appear to indicate that steroid

2.

HORMONES AND REPRODUCTIVE BEHAVIOR I N FISHES

93

and thyroid hormones influence the responsiveness of the CNS to external stimuli. It seems reasonable to suppose that hormone-induced changes in responsiveness to general or specific external stimuli are in part responsible for increased locomotory activity. Such changes in activity and responsiveness may be important components in migratory and other reproductive behavior in fishes.

D. Thyroid Hormones and Migratory Behavior The previous section surveyed a number of investigations in which it was found that treatment with steroid and thyroid hormones resulted in an increase in locomotory activity. In most examples this “activity” was not obviously linked with any specific behavioral system. In a few cases it was suggested that the activity induced by the hormone treatment was reminiscent of the behavior of migrating fish held in captivity ( van Iersel, 1953; Forselius, 1957; Baggerman, 1962). Baggerman (1957) showed that in the stickleback the gonads are not essential to the onset of migration disposition (as measured by salinity preference). Baggerman ( 1962) mentions other investigations which have indicated that gonadal maturation and regression are not factors underlying migration. However, Baggerman ( 1962) suggests that gonadal hormones are involved in timing and perhaps augmenting the action of the thyroid hormone which, she believes, plays an important causative role in the onset of migration in certain anadromous fish. Baggerman (1957, 1959) tested the effects of thyroxine, TSH, and various thyroid blocking agents on the salinity preference of the truchurus form of the three-spined stickleback. The results suggest that an increase in the level of circulating thyroid hormone induces a preference for freshwater. This agrees with the fact that the stickleback migrates to freshwater in spring when the thyroid gland shows signs of heightened activity. Low levels of thyroid hormone, as is believed to occur in the fall, result in a preference for saltwater. As mentioned previously, Baggerman (1962) also found that both thyroid hormone ,and androgen treatments stimulate a form of locomotory activity characteristic of migrating fish. Baggerman (1962) concludes that an increase in thyroid activity brought about by long photoperiod induces two changes which are known to be associated with migration. Baggerman (1960, 1963) carried out similar experiments with juvenile Pacific salmon, Oncorhynchus, and found an increased preference for saltwater associated with the downstream migration. The saltwater preference is in some species correlated with heightened thyroid activity.

94

N. R. LILEY

Juvenile coho saImon treated with TSH showed a change from freshwater to saltwater preference ( Baggerman, 1963). Thiourea treatment induced a freshwater preference in coho which initially showed a preference for saltwater ( Baggerman, 1960). Similarly, underyearling pink salmon treated with thyroid blocking agents showed a change from saltwater to freshwater preference ( Baggerman, 1963). Although the changes in salinity preference induced by thyroid hormone are directly opposite in sticklebacks and salmonids, Baggerman (1962) argues that the thyroid gland is directly involved in the causation of migration of these species. On the other hand, Hoar ( 1 9 6 5 ~ )reviews the available information regarding the role of the thyroid and concludes that the data suggest a generalized function of thyroxine in metabolism and electrolyte balance rather than a specific causative role associated with migration. Hoar suggests that the onset of migration may depend upon general growth and maturation with associated, but not regulating, hormone changes.

111. PITUITARY HORMONES

A. Prespawning Behavior Reviews by Pickford and Atz (1957), Dodd (1960), and Hoar (1965a, 1966; and chapter by Hoar, this volume) survey a considerable body of evidence which indicates that at least one pituitary hormone regulates gonadal development and hormone secretion and thereby exerts an indirect control over reproductive activities. Indeed, treatment of intact fish with fish or mammal pituitary material has proved a highly effective approach to the breeding of commercially important freshwater fishes. Pickford and Atz (1957) provide a thorough review of these procedures. Ramaswami and Lakshman ( 1958), Sneed and Clemens (1959), Chaudhuri (1960), Sneed and Dupree ( 1961), Das and Khan (1962), Clemens and Sneed (1962), Stevens (1966), and Sundararaj and Goswami (1966) have provided additional information on techniques and the effectiveness of various substances of mammalian and piscine origins. In addition to their gonadotropic role there is reason to believe that in a number of cases pituitary hormones may have a more direct influence on behavior, perhaps by acting on the CNS directly. Much of the evidence for such a pituitary control is of an indirect nature. Usually the first indication that the pituitary may have a direct effect on behavior arises from experiments in which it is found that reproductive behavior

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

95

persists after gonadectomy. Of course it does not necessarily follow in such cases that the pituitary is directly involved; other endocrine glands may be important, for example, the thyroid in stickleback migration, or there may be no endocrine involvement at all. There have been relatively few attempts to investigate direct pituitary effects by hypophysectomy and replacement therapy. The work of Baggerman (1966) and Hoar (1962a,b) leaves little doubt that in the stickleback gonadal development and hormone secretion (at least in the male) are under pituitary gonadotropic control. But, in addition to the indirect effects of the pituitary on reproductive behavior, both of these workers have hypothesized that prenestbuilding aggressive behavior is directly controlled by the pituitary. Baggerman (1966) found that castration of males prior to nestbuilding did not affect aggressive behavior, whereas castration after the onset of breeding resulted in a marked decline in aggressive behavior. Hoar's evidence (1962a,b) for a direct pituitary action is based upon a comparison of castrated sticklebacks under regimes of long and short photoperiod. The long photoperiods are assumed to produce a high level of pituitary gonadotropin, while the output of these pituitary factors is greatly depressed or eliminated under short photoperiods. Males castrated prior to nestbuilding show a high level of aggression if they are maintained under long photoperiod. Castrates held under short photoperiod are much less aggressive. In experiments designed to measure the effects of a series of mammalian pituitary hormones on aggressive behavior only treatment with LH consistently produced an increase in aggressive behavior (Hoar, 1962a). Ahsan and Hoar (1963) also found LH to be the most effective mammalian gonadotropic hormone in eliciting gonadal development in immature fish. There are indications that in the stickleback androgens and pituitary hormones act synergistically to maintain full reproductive behavior. Hoar ( 1962a,b) found that methyltestosterone treatment was less successful in inducing nestbuilding in castrate males kept under short photoperiod than in males under long photoperiod. These findings suggest that although the behavior of the nestbuilding phase requires the gonadal hormone its full expression only occurs when the gonadotropic activity of the pituitary is high. Smith (1967) found that in males of two species of Lepomis aggressive behavior remained high after Castration and was not affected by androgen treatment. Aggressive behavior remained high under short photoperiod provided the temperature remained high ( WOC), but it declined at low temperatures ( 13°C). Treatment with human chorionic gonadotropin, although effective in stimulating nestbuilding, did not

96

N. R. LILEY

affect aggressive behavior. Smith suggests that in these species aggression, although inhibited by low temperature, seems more closely related to social conditions than hormonal state. Wiebe (1967) was able to relate the reproductive cycle of Cymatogaster aggregata to the seasonal light and temperature cycle, indicating that pituitary gonadotropins are involved in stimulating gonad development and hormone secretion. Treatment with methallibure (ICI 33,828), a substance shown to have antigonadotropic activity in a number of fish species (Hoar et al., 1967; Wiebe, 1968; Pandey and Leatherland, 1969), effectively blocks gonadal development and results in the complete elimination of reproductive behavior including chasing and mating attempt. As these two activities persist after castration Wiebe concludes that they are normally under direct pituitary control. Burger ( 1941) hypophysectomized males of Fundulus heteroclitus and noted that the later stages of spermatogenesis came to a halt. Implantation of Fundulus pituitary material, five pituitaries at each treatment, at 3 5 day intervals, resulted in vigorous sexual behavior. By the tenth day the fish were in a “frenzy of display.” As the treatment also resulted in the recrudescence of the testes, it is not known whether the behavioral effects were the results of a direct action of the pituitary material on the CNS or a result of the stimulation of gonadal secretory activity. Tavolga ( 1955) hypophysectomized males of Bathygobius soporator and found that courtship and agonistic behavior were eliminated. This result is evidently not simply an effect of a reduction in gonadal hormone secretion since castration does not eliminate courtship behavior. Hypophysectomy of virgin female guppies, Poecilia reticulata, results in an immediate and complete decline in sexual responsiveness (Liley, 1968). Because this effect contrasts with the relatively slow decline in sexual behavior following gonadectomy, and because of the reappearance of sexual responses several weeks after gonadectomy, Liley proposed that there is a pituitary factor directly involved in the control of receptivity in the female. On the other hand, the decline in receptivity after gonadectomy and the cyclical changes in responsiveness of nonvirgin fish suggest that the ovary is also producing a hormone which regulates, or acts synergistically with, the more direct pituitary mechanism. Liley and Donaldson (1969) treated female guppies with a partially purified gonadotropin fraction of salmon pituitary material. Several of a group of hypophysectomized, ovariectomized fish exhibited a low level of sexual responsiveness. Hypophysectomized females with their ovaries intact showed a marked increase in sexual responsiveness after several days of hormone treatment. This increase coincided with the growth of

2.

HORMONES AND REPRODUCI’IVE BEHAVIOR IN FISHES

97

the ovaries which had previously regressed. These results appear to support the suggestion that in the female guppy sexual responsiveness is not completely dependent upon any one hormone, but under normal circumstances the gonadotropic and ovarian hormones act synergistically-the absence of one hormone may be partly compensated for by the other. B. Spawning Behavior

It is convenient to give separate consideration to the role of hormones involved in the actual spawning act involved in the release of gametes as opposed to the more lengthy prespawning procedures. There is some evidence that the two phases may be under separate controls. In general, the distinction between prespawning and spawning behavior is more clearly defined in males than in females. In females there may be little in the way of overt prespawning behavior before a female in “ripe” condition responds to male courtship and proceeds rapidly to actual spawning. Pituitary hormone treatment procedures described by Pickford and Atz ( 1957) and others cited (Section 111, A ) have proved highly effective in the induction of spawning in fish which normally prove difficult to breed under laboratory or artificial pond conditions. Because females generally prove more refractory than males under such conditions most of the investigations involving pituitary treatment have involved females. Pituitary hormone therapy has been used to accelerate maturity in fish in nonreproductive condition (e.g., Combs and Burrows, 1959; Ahsan and Hoar, 1963). More often these techniques have been applied to fish already mature to induce or accelerate ovulation (release of ova into the ovarian lumen) and perhaps actual spawning. In most studies reported treated fish are stripped by hand and the eggs fertilized artificially. However, if stripping is not carried out but the females are placed with mature males, spawning occurs within a few hours of ovulation. Fontenele (1955) reports that in Brazilian fish culture practices a hormone-treated female is separated by a partition from a male until readiness to spawn is indicated by a state of accentuated agitation. The partition is then removed and the fish allowed to spawn. Spawning usually occurs within 24 hr of the first of a series of injections of fish pituitary material. This last report brings to the fore the close temporal relationship between ovulation and the spawning act already referred to in Section 11, B. It appears that spawning only occurs after ovulation of mature eggs. The apparently close dependence of spawning on ovulation gives rise to

98

N. R. LILEY

a number of suggestions as to the endocrine or neural involvement in the control of spawning: the gonadotropic hormone responsible for ovulation may also act directly on the CNS to induce a state of sexual receptivity; the gonadotropin may induce a marked increase in gonadal hormone which in turn affects behavior; the release of ova into the ovarian cavity may also involve a sudden release of gonadal hormones from the follicles; the physical presence of ova in the ovarian cavity may have a direct neural effect on the CNS, this effect may be accentuated by gonadal hydration which takes place at ovulation (Clemens and Grant, 1964); and, lastly, the endocrine events involved in ovulation, or behavioral events which follow, may stimulate the secretion of a neurohypophysial hormone which in turn induces actual spawning. There is some evidence that in male fish readiness to spawn is associated with seminal thinning (Clemens and Grant, 1964, 1965). This increase in fluidity of the semen appears to be the same as the process referred to as “spermiation” by Yamazaki and Donaldson (1968a,b). Both pairs of workers have demonstrated for a number of species of fish that spermiation (or gonadal hydration) occurs in response to treatment with fish and mammalian gonadotropin preparations. More recently, Yamazaki and Donaldson (1968~)have found that androgen alone will induce spermiation in hypophysectomized goldfish. However, it is evident from work discussed in Section 11, A that spawning behavior may occur, or be induced by androgen treatment, in castrated fish. Thus, spawning in males is not dependent upon the presence of intact gonads as appears to be the case in females. At present there is little convincing evidence for or against the above suggestions regarding the significance of ovulation in the timing of oviposition. Yamazaki (1965) has gone further than most workers in an attempt to determine the factors involved in the regulation of spawning. Under natural spawning conditions the male goldfish courts the female for 1 or 2 days before ovulation and spawning occur. Evidently male courtship is important in the induction of ovulation. Spawning will not occur in females from which ovulated eggs have been stripped, suggesting that the presence of ova is an important factor. On the other hand, females hypophysectomized after ovulation may oviposit normally about 3 hr after the operation. A small number of hypophysectomized females in which ovulation had been induced by hormone treatment 1 4 days after hypophysectomy spawned normally when paired with males. Yamazaki (1962) also found that male goldfish would spawn several hours, or in some cases, the day after hypophysectomy. Yamazaki (1965) concludes that, although ovulation is induced by pituitary gonadotropic hormone, actual spawning is not under direct pituitary control. However,

2.

HORMONES AND REPRODUCITVE BEHAVIOR IN FISHES

99

he recognizes the possibility that in his experiments residual pituitary hormone may remain in effective quantities for some time after hypophysectomy. In addition he draws attention to the possibility that neurohypophysial hormones or their precursors may be released directly from the hypothalamus of hypophysectomized fish and could be involved in the regulation of spawning behavior. There have been a number of suggestions that neurohypophysial hormones are directly involved in the causation of spawning behavior. Wilhelmi et al. (1955) found that purified fish and mammalian neurohypophysial preparations and synthetic oxytocin would induce the spawning reflex in Fundulus heteroclitus. The response usually appears within 10 min after injection and may persist for 2030 min. This behavior occurs in hypophysectomized or castrated fish of either sex. However, there is no coordination of the response when males and females are injected and placed together. It is suggested that this response is mediated by direct excitation of a nervous center and that sex hormones play no part in mediating this phase of the sexual behavior pattern. Egami ( 1959) injected mammalian neurohypophysial substances into Oryzias Zutipes and observed spawning movements in both males and females. These responses occurred within 10-60 min of treatment. Females isolated from males prior to treatment oviposited even though there was no male present. Egami and Nambu (1961) consider these findings in relation to the normal breeding activity of Oryzias. During the breeding season the female lays eggs almost every morning. Maturation of oocytes and ovulation may take place in a female which is isolated from a male, but oviposition only occurs naturally after a period of male courtship. Egami and Nambu suggest that stimuli arising from male courtship induce the secretion of neurohypophysial hormone which, when it reaches a certain level, causes oviposition to occur. Females of Gambusia afinis performed behavior resembling a spawning reflex and released their embryos shortly after treatment with oxytocin, or homogenates of frog neurointermediate lobe or rat neural lobe (Ishii, 1963). Egami and Ishii ( 1962) mention unpublished work by Shirai in which it was found that injection of neurohypophysial substances will induce oviposition in ovulated female bitterling, Rhodeus ocelhtus. These authors also refer to reports, without giving details, that loach, salmonids, and some other species fail to exhibit any definite response to the injection of a neurohypophysial extract. Blum (1968) injected reserpine into immature angelfish, PterophyZlum scaZure. One to three hours after the injection the fish darkened; 6-8 hr after injection several of the fish performed typical spawning movements. Blum suggests that reserpine may exert its effect by stimulating the release of melano-

100

N. R. LILEY

phore stimulating hormone, responsible for the darkening, and a neurohypophysial hormone, responsible for the induction of the spawning response. Pickford and Atz (1957) review earlier work in which it had been suggested that pituitary secretions pass directly to the brain and induce the final stages of maturation and spawning activity. Gerbilskii (1938, cited in Pickford and Atz, 1957) named this secretion the “spawning hormone.” Support for the existence of such a hormone was derived from the findings by Gerbilskii and others that intracranial injections of pituitary substances were more effective in inducing final maturation than intraperitoneal or intramuscular injections. However, Pickford and Atz (1957) point out that other Russian workers have not confirmed these results. Vivien (1941) also noted that intracranial implantation of pituitary gland into female Gobius paganellus were far more effective in inducing spawning than similar amounts injected into the body cavity. It is impossible at this stage to draw any firm conclusions regarding the mechanism by which spawning is regulated in fish. In general, gonadotropin-induced ovulation must occur before spawning can proceed-it may turn out that this is not an essential precursor to spawning behavior. Several workers have suggested that spawning is directly evoked by a neurohypophysial hormone secreted in response to appropriate sexual and environmental stimuli. However, it has not been possible to confirm this in several species of fish tested. Also, the interpretation of the effects of neurohypophysial preparations has been questioned on the grounds that the doses required to elicit an effect are extremely high (Dodd et al., 1968). C. Prolactin and Parental Behavior

Several studies have implicated a prolactinlike hormone in the regulation of parental behavior in fish, Noble et al. (1938) found that injections of prolactin, as well as several other hormone preparations, induced parental behavior in both males and females of Hemichromis bimculutus (discussed in Section 11, A ) . Fiedler ( 1962) reports that males of the wrasse, Crenilabrus ocellatus, treated with prolactin would perform parental fanning even though no nest was present. Similarly, prolactin treatment of Symphysodon aequifasciata resulted in a number of changes associated with the parental phase of the reproductive cycle (Blum and Fiedler, 1964). At the lower dose leveIs tested there was a marked increase in the number of mucus-secreting cells at the. body surface. The secretions of these cells normally provide supplementary nutrient for the

2.

HORMONES AND REPRODUCXNE BEHAVIOR I N FISHES

101

young. At high doses the fish began to perform parental-type fanning directed to a particular place in the aquarium. Agonistic behavior in response to a test fish was also affected: At intermediate doses there was a decrease in fighting behavior, at higher doses fighting increased. Blum and Fiedler (1965) examined the effects of pituitary hormone treatments on several species of cichlid including Pterophyllum scalare, Aequidens latifrons, Cichlasoma seuerum, and Astronotus ocelhtus. In all cases prolactin treatment resulted in an increase in mucus cell production. This effect was less marked than in Symphysodon which is the only species in which the mucus serves as a nutrient for the young. Some behavioral effects of prolactin treatment were noted. In the case of Pterophyllum a low dose of hormone induced a parental type of fanning directed toward a fixed point in space; at higher doses fanning decreased. Agonistic activity in response to a stimulus animal was depressed by prolactin treatment in Pterophyllum and Aequidens but unaffected in the other species. Prolactin tended to make fish very calm and to depress the feeding response. Blum and Fiedler regard this latter effect as a component of parental behavior which prevents eggs or young being eaten by the parents. They also note that hormone-induced parental responses occurred in fish which had had no previous experience of breeding. Blum ( 1966) tested progesterone and several adenohypophysial hormones on Pterophyllum and Symphysodon. Only prolactin treatment resulted in the appearance of parental behavior, although both somatotropin and thyrotropin elicited an increase in mucus cell production. Luteinizing and follicle-stimulating hormones resulted in an increase in aggressive behavior. Blum (1966) concludes that prolactin is the only hormone involved in the control of parental behavior in cichlids. Prolactin treatment resulted in an increase in mucus cell production in the paradise fish, Macropodus operculuris ( Machemer and Fiedler, 1965). This effect was more pronounced in males than in females which do not normally build bubble nests or show parental care. However, prolactin treatment alone did not produce an increase in nestbuilding even though the fish produced more mucus. Chorionic gonadotropin alone, or in combination with prolactin, increased nestbuilding behavior. Similarly, methyltestosterone combined with prolactin treatment resulted in extensive nestbuilding in both males and females. Machemer and Fiedler (1965) conclude that for full nestbuilding activity two hormones are necessary: Androgen increases building activity, but this effect is only expressed fully in the presence of a prolactin-induced increase in mucus production. Smith and Hoar (1967) studied the effects of prolactin treatment on parental and displacement fanning behavior in the stickleback. They

102

N. R. LILEY

conclude that there is no evidence of a pituitary regulated control of fanning behavior in the stickleback. In fact, injections of prolactin appeared to depress fanning, although in no case was the difference between the treatment groups statistically significant. Smith and Hoar suggest that this depression effect may have resulted from a feedback which reduced the output of natural gonadal androgen. Blum (1966) also obtained evidence that prolactin treatment suppresses gonadotropin production in cichlids. It is difficult to draw any definite conclusions regarding the role of prolactin in reproductive behavior in fish. It is important to separate two types of effect. First, in a number of fish tested prolactin treatment results in an increase in mucus production (Egami and Ishii, 1962; Blum, 1966; Blum and Fiedler, 1964, 1965; Machemer and Fiedler, 1965; Leatherland and Lam, 1969; see review in chapter by Ball, Volume 11). However, this effect on epithelial tissues is evidently not universal among fish (Egami and Ishii, 1962; Bern, 1967). Second, evidence suggests that prolactin, or rather a prolactinlike hormone, is involved in the regulation of parental behavior. The situation is complicated by the fact that untiI now only mammalian prolactin has been used in treatment of fish. Thus differences among various species of fish might result in part from differences in specificity in responsiveness to tetrapod hormone. Evidence reviewed by Bern (1967) indicates that fish “prolactin” is only partly related to the mammalian hormone in terms of its capacity to produce typical prolactin effects in birds and mammals.

IV. EXTERNAL FACTORS AND THE ENDOCRINE SYSTEM

The role of photoperiod, temperature, and other physical factors in the regulation of seasonal changes in reproductive behavior has been reviewed by Aronson ( 1965), Hoar (1965a), and in the chapter by Schwassman, Volume VI. In general, physical stimuli serve as proximate factors which act through the brain-pituitary-gonad axis to ensure that the fish are ready to carry out the reproductive processes at a time and place most favorable to the production and survival of eggs and young. The synchronization achieved in this manner is relatively crude. A “fine adjustment” is provided by the animals in their responses to the physical environment and by their behavioral interaction with conspecifics. Little is known of the internal mechanisms by which the final syn-

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

103

chronization is achieved. It is possible that there is no endocrine involvement other than that the sex hormones induce a state of responsiveness to certain classes of behavioral stimuli. On the other hand, it seems more likely that, apart from immediate behavioral responses, the performance of reproductive behavior may induce endocrine changes in the performing and responding animals, providing the appropriate conditions for the later phases of reproductive activity. In other words, it seems reasonable to look for a situation in fish comparable to that described most clearly in the canary ( Hinde, 1965) and the ring dove (Lehrman, 1965). Not only do hormones affect behavior, but behavior in turn affects the endocrine state. The interaction of these relationships results in the integration of behavior with other physiological conditions and the smooth progression through successive stages of the reproductive cycle. Relatively little attention has been directed to the details of the breeding requirements of fish, The difficulties encountered by aquarists and commercial fish culturists in obtaining spawning in gravid fish indicates that for the final act certain key stimuli are necessary. Aronson (1965) points out that it is common aquarium practice to lower the temperature of the water to induce cyprinids and characins to breed in captivity. Chaudhuri (1960) remarks that in several species of Indian cyprinid spawning only occurs on overcast days and appears to be inhibited by direct sunlight. Yamazaki (1965) points out that goldfish will only spawn in the presence of green plants. Bitterling require the presence of a freshwater mussel before they will show any reproductive behavior (Shirai, 1962). Courtship plays an important role in the final synchronization of the sexes. It seems likely, although there is virtually no experimental evidence, that courtship affects the endocrine state of one or both partners. Johns et al. (1969) found that 2-3 days of exposure to a nestbuilding male is necessary before a female gourami will spawn. This suggests that the exposure to the male affects the final maturation and preparation of the endocrine state before mating. Egami and Nambu (1961) maintain that in Oyxias male courtship stimulates neurohypophysial activity which in turn is responsible for the induction of oviposition behavior in the female. Exposure to other individuals of the same sex may also be important. Aronson (1951) notes that females of Tilapia macrocephalu maintained in visual isolation from conspecifics could be spawned, on average, three times a year. Females able to see other fish, males or females, went through seven or eight spawnings a year. Aronson (1965) refers to work involving salmon, herring, whitefish, and minnows which suggests that social facilitation may occur in the synchronization of breeding.

104

N. R. LILEY

Van Mullem (1967) provides evidence of a considerable degree of synchronization in the reproduction of the stickleback in natural populations. Van den Assem (1967) found that in the three-spined stickleback, males in winter condition came into breeding more readily if paired, one on each side of a glass partition, than did solitary males. The presence of vegetation also had a stimulating effect upon nestbuilding. On the other hand, the presence of males in breeding condition may have an inhibitory effect upon those without nests. The mechanism of these inhibitory and stimulatory effects is unknown but could involve the endocrine system. In conclusion, although much is known regarding the proximate environmental factors involved in the long-term control of breeding, little is known regarding the short-term synchronization of sexual partners and the regulation of the successive subcycles which constitute reproductive behavior in many species of fish. The little evidence available suggests that we may find that endocrine mechanisms are involved and that, as in other vertebrates, we will find that not only do hormones affect behavior but in turn behavior affects the endocrine system.

V. SUMMARY AND DISCUSSION

The preceding survey of the literature reveals a considerable diversity in the hormonal regulation of reproductive behavior, even though relatively few species have been studied, It proves virtually impossible to draw any general conclusion about the role of any one hormone or the control of any one class of behavior. It is well established that in male fish morphological secondary sex characters are dependent on gonadal hormones. In several species examined certain components of reproductive behavior are evidently under gonadal control, but there are others in which much or all reproductive behavior persists after castration-with the implication that this behavior is independent of gonadal hormones. The effects of ovariectomy have been studied in only a few species, usually without the careful behavioral examination necessary to detect subtle changes resulting from the operation. Ovariectomy abolishes the sexual response in those species examined. In the guppy sexual behavior reappears some time after ovariectomy (Liley, 1968). There has been no unequivocal demonstration that administration of estrogen or progesterone will induce reproductive behavior in fishes. On the basis of salinity preference tests, thyroid hormone has been

2.

HORMONES AND REPRODUCTIVE BEHAVIOR I N FISHES

105

implicated in the migratory behavior of the stickleback (Baggerman, 1957, 1959). However, no consistent picture emerges in studies involving other anadromous fishes, and there has been no exhaustive study of the effects of other factors on the migratory behavior in sticklebacks. A similar lack of uniformity emerges in studies of the pituitary hormones. The pituitary is inevitably involved in the regulation of reproductive behavior by its gonadotropic control of gonad activity. In addition, there have been a number of suggestions that gonadotropic hormones act directly upon behavioral control mechanisms or many act synergistically with gonadal hormones (Hoar, 1962a,b; Liley, 1968). Neurohypophysial hormones appear to play a role in the control of spawning in certain species of fish (Wilhelmi et al., 1955; Egami and Ishii, 1962). Although these findings have not been widely confirmed they do raise the possibility that the act of spawning may be under different control to those activities which precede it. Prolactin is said to induce parental behavior in several species of cichlid and a wrasse (Fiedler, 1962; Blum and Fiedler, 1964, 1965). There has been no widespread confirmation of a relationship between prolactin and parental behavior: One attempt with the stickleback (Smith and Hoar, 1967) resulted in negative evidence. There are several possible explanations for the apparent lack of uniformity in the behavioral results of experiments involving surgical and hormone treatments. First, the apparent diversity may be spurious, arising out of the experimental procedures themselves. Second, fish may show varying degrees of cephalization in the control of reproductive behavior, analogous to Beach‘s “corticalization of function” in mammals ( 1964). Third, the diversity in endocrine mechanisms may be real, arising from a considerable flexibility in the responsibilities of the hormones. These three possibilities will be considered in turn. Experimental procedures. Most experiments involving hormone treatment have used synthetic hormones or naturally occurring hormones from other species of fish or mammal, It would not be surprising to find species or group differences in responsiveness to hormones from such sources. Although there is no absolute biological specificity, important phylogenetic differences have been recognized in the protein and peptide hormones of the pituitary (Pickford and Atz, 1957; Dodd et al., 1966; Bern, 1967). On the other hand, the similarity of most of the steroids identified so far in fish (Gottfried, 1964; see chapter by Hoar, this volume) to those of mammals suggests that phylogenetic specificity is not too serious a problem in applying steroid therapy. But, in the case of steroids a different problem is raised by the apparent similarity of the biosynthetic pathways involved in the production of estrogens, proges-

106

N. R. LILEY

terone, androgens, and corticosteroids (see Fig. 23.5 in Hoar, 1966). This gives rise to the difficulties in the interpretation of the results of hormone therapy or identification discussed by Hoar (1966, p. 699, and chapter given in this volume), e.g., an exogenous hormone may be metabolized to another rather different form which is in turn responsible for the observed physiological effects. Furthermore, there is the possibility that another steroid secreting gland may produce sufficient estrogen or androgen to act on reproductive processes-a situation which occurs under some circumstances in mammals (Julesz, 1967). Aronson (1959) stressed the danger that even very small fragments of gonadal tissue may regenerate and provide sufficient hormone to induce reproductive behavior. Thus, undetected regeneration may have been responsible for the persistence of reproductive behavior in some of the gonadectomy experiments reported earlier. Lastly, work with fish has so far been characterized by the small number of replications within any one study, or a lack of independent replications by different workers investigating the same species. The exception here is work with the stickleback. Significantly, Baggerman (1966) and Wootton ( 1968), studying the stickleback, obtained rather different results which could result from differences in the races used or the testing procedures. Cephalization of the control of reproductive behavior. If we assume that the surgical operations were completely successful then a striking feature of many gonadectomy experiments is the persistence of reproductive behavior after the operation. This has been interpreted as an indication that the regulation of this behavior is not dependent on gonadal hormones. At first sight these findings appear to be similar to the results of gonadectomy experiments with several species of mammal. Beach (1947, 1964) has pointed out that in certain groups of mammals, notably the carnivores, ungulates and primates, we find a tendency toward a “corticalization of function” in the control of sexual activities; that is, although hormones may be essential to the development of reproductive behavior, these activities become less dependent on gonadal function after maturity and experience of breeding. In males, in particular, sexual arousal and performance appear to depend to a considerable extent upon learning and experience. Is there a process in fish analogous to the “corticalization of function” in mammals? Aronson (1959) considers this question and concludes that although the teleost forebrain differs in important respects from that of a mammal “we must recognize the possibility that, incorporated within the complex teleostean forebrain, there may be a mechanism equivalent to that indicated in higher mammals, whereby some elements of the

2.

HORMONES AND REPRODUCTIVE BEHAVIOR IN FISHES

107

reproductive pattern have been released from the functional control of the gonadal hormones.” It is important to recognize that independence of gonadal hormones does not necessarily imply a cephalization of function in the sense outlined by Beach. In particular, Beach (1964) stresses the role of learning and experience in arousal and performance of sexual behavior. Little attention has been given to the role of experience in studies of hormones and behavior in fish. One of the few cases in which experiential factors have been considered is the investigation of Noble et al. (1938). They found that several of their treatments would induce parental behavior in Hemichromis, but this only occurred in fish with previous breeding experience. Although in most gonadectomy experiments there is little direct evidence available, it seems unlikely that experiential factors could have played an important role in the persistence of reproductive behavior after gonadectomy. Furthermore, it is important to stress that in several cases the behavior which persists after gonadectomy is not exclusively reproductive in function. This applies particularly to agonistic behavior which, while it is prominent in reproduction, is also involved in the maintenance of individual distance or dispersal of young and adults not in breeding condition. Therefore, it is not surprising to discover that agonistic behavior as such is more or less independent of gonadal hormone. In a number of cases there are indications that such activities are more directly controlled by the pituitary. But, one might go further and question the existence of any endocrine control at all. Smith (1967) has suggested that in Lepomis agonistic behavior depends more on social conditions rather than on hormonal state. Flexibility in hormone responsibilities. It may be unrealistic to expect uniformity in the endocrine regulation of behavior in such a diverse group as the teleosts. Endocrine evolution has evidently involved the evolution of the uses to which hormones are put as well as the hormones themselves. Even among closely related species we may expect to find considerable flexibility in the responsibilities of certain hormones. Hoar (196513) has pointed out that in the anadromous form of the stickleback there appears to be a shift in hormone responsibilities within the annual cycle. Long photoperiods stimulate the secretory activity of the anterior pituitary; its hormones first induce the physiology associated with the change in tonicity of the environment and the migratory and presexual phases of behavior. Three factors, luteinizing, lactogenic, and thyroidstimulating hormones, are known to be involved. At the same time, the pituitary-presumably the luteinizing hormone-activates the interstitial tissue of the gonads which produce the gonadal steroids; these dominate the sexual phases, gradually taking over complete control of behavior

108

N. R. LILEY

during the parental phase. In other words, there is a shift in the endocrine control of reproductive events during the cycle. It seems reasonable to assume that the same type of shift or flexibility in endocrine responsibility which appears to occur within the life cycle of a single species will also provide the basis for variability in control mechanisms of different species or major groups. The three possible sources of diversity in the results of endocrinebehavior investigations in fish are not mutually exclusive. However, at present few species have been studied with sufficient thoroughness to exclude one or more of the various “explanations.” The manner in which hormones exert their effects on behavior in fish is poorly understood. Studies involving other vertebrate classes have suggested a number of possible mechanisms (Lashley, 1938, cited by Hinde, 1966): ( 1 ) The hormone stimulates the growth of new nervous connections ( 2 ) The hormone acts by inducing specific changes in various organs, and these mediate sensory impulses which influence the central nervous system ( 3 ) The hormone increases the excitability of specific sensory-motor mechanisms in the central nervous system (4) The hormone increases the general excitability of the organism It is possible to point to what seem to be examples of each of the mechanisms listed above. Much of the work with poeciliids and Oryzias suggests that gonadal hormones are involved in sex determination and, it would appear, in the development of the neural mechanisms appropriate to each sex (see chapter by Yamamoto, this volume). In a few cases we have evidence that hormones induce changes in peripheral structures and thereby affect the animals’ sensitivity to certain stimuli. Tavolga ( 1955) believes that castration of Bathygobius soporator results in a change in olfactory sensitivity. Electrophysiological studies involving goldfish (Section 11, C ) reveal that hormones may affect the sensitivity of the olfactory organ. It seems likely that in many cases hormones exert their effects on the excitability of specific sensory-motor mechanisms in the CNS. However, the only direct evidence that hormones affect central mechanisms in the fish brain comes from electrophysiological studies of the goldfish. Hara ( 1967), Hara et al. ( 1965, 1968), and Oshima and Gorbman ( 1966a) recorded from the fore- and midbrain of the goldfish and report changes in electrical activity and responsiveness to stimulation, brought about by treatment with steroids. The behavioral significance of these findings is not known. Lastly, several investigators have shown that hormones may result

2.

HORMONES AND REPRODUCITVE BEHAVIOR IN FISHES

LO9

in changes in locomotory activity. Such changes in behavior may be in part the indirect result of metabolic effects of the hormones concerned. But, in addition, it is suggested that the increased activity reflects changes in general susceptibility to external stimuli. In either case these changes in activity may play an important role in behavior leading up to and including reproduction. In addition to their direct effects on the nervous system, hormones evidently influence reproductive behavior by their role in the development of secondary sexual characteristics which serve as signals or accessories during the breeding process. The most obvious of these are morphological structures and coloration. There is some inconclusive evidence that pheromones are present and under gonadal control in Bathygobius soporator ( Tavolga, 1956), Trichogaster trichopterus (Johns et al., 1969), and possibly in the guppy (Amouriq, 1964). A different form of indirect hormone effect is suggested by Baggerman (1966), who points out that in the stickleback displacement fanning and perhaps changes in activities such as swimming or comfort movements may arise as a result of the interaction of other motivational systems which are in their turn under hormonal control. Clearly these categories of mechanisms of hormone action are not mutually exclusive; the same hormone may act in one or all of the possible ways.

VI. CONCLUSION

The preceding review reveals an apparent diversity and lack of agreement in studies of the role of hormones and behavior in fish. This is perhaps an inevitable outcome of the relatively primitive stage of development of this field of investigation. There is considerable need for more investigations using basic endocrine procedures of gland removal and replacement therapy, with the addition of the more refined techniques of hormone implantation and electrical recording. Such studies must incorporate a thorough analysis and understanding of the normal behavior of the species under investigation. Not only do we require more "in depth" studies but also it is essential to consider a wider variety of species. Comparison of different species and a comparison of fish with other vertebrates will almost certainly provide insight into the significance of the differences between taxa and perhaps shed further light on the evolution of endocrine mechanisms. Undoubtedly the investigation of hormones and behavior in fish can

110

N. R. LILEY

gain great impetus from the ideas and experimental procedures which have emerged from the more sophisticated studies of birds and mammals. However, it is well to bear in mind the warning implicit in Bern’s comments (1967) that “there is an elementary and unjustifiable tendency to view the fishes, in view of their collective ancestral position, as much like their landliving descendents only more primitive.” As Bern points out the fishes reveal a broader range of variation and a longer history of adaption than do the landliving vertebrates. In other words, while benefiting from the more advanced state of our knowledge of other vertebrates, it is important to be prepared to find basic differences in the details of the mechanisms which have evolved. REFERENCES Ahsan, S. N., and Hoar, W. S. (1963). Some effects of gonadotropic hormones on the three-spined stickleback, Gasterosteus aculeatus. Can. J . ZOOZ.41, 1045-1053. Amouriq, L. (1964). L’activitk et le phbnomkne social chez Lebistes reticulatus ( Poeciliidae, Cyprinodontiformes ) . Compt. Rend. 259, 2701-2702. Amouriq, L. (1965a). L’activit6 et le phknomkne social chez Lebistes reticulatus (Poeciliidae, Cyprinodontiformes). Ann. Sci. Nat. Zool. Biol. Animale [12] 7 , 151-172. Amouriq, L. (1965b). Origine de la substance dynamogene 6mise par Lebistes reticulatus femelle ( Poisson Poeciliidae, Cyprinodontiformes) . Compt. Rend. 260, 2334-2335. Amouriq, L. (1967). L’optimum de sensibilit6 de Lebistes reticulatus (Poisson Poeciliidae, Cyprinodontiformes ) a l’hormone synthbtique femelle. Rev. Comp. Animal 3, 57-60. Aronson, L. R. ( 1951). Factors influencing the spawning frequency in the female cichlid fish, Tilupia mucrocephala. Am. Museum Novitates 14.84, 1-26. Aronson, L. R. (1957). Reproductive and parental behavior. In “The Physiology of Fishes” (M. E. Brown, ed.), Vol. 2, Chapter 3, pp. 271-304. Academic Press, New York. Aronson, L. R. ( 1959). Hormones and reproductive behavior: Some phylogenetic considerations. In “Comparative Endocrinology” (A. Gorbman, ed. ), pp. 98-120. Wiley, New York. Aronson, L. R. ( 1965). Environmental stimuli altering the physiological condition of the individual among lower vertebrates. In “Sex and Behavior” ( F . A. Beach, ed.), pp. 290318. Wiley, New York. Aronson, L. R., and Holz-Tucker, M. (1947). Morphological effects of castration and treatment with gonadal hormones on the female cichlid fish, Tilapia macrocephulu. Anat. Record €@, Suppl., 572. Baerends, G. P., and Baerends-van Roon, J. M. (1950). An introduction to the study of the ethology of cichlid fishes. Behaviour Suppl. 1, 1-243. Baerends, G. P., Brouwer, R., and Waterbolk, H. Tj. (1955). Ethological studies on Lebistes reticulatus (Peters). I. An analysis of the male courtship pattern. Behaviour 8, 249-334. Baggerman, B. (1957). An experimental study on the timing of breeding and migration in the three-spined stickleback (Gasterosteus acukatus L,). Arch. Neerl. Zool. 12, 105417.

2.

HORMONES A N D REPRODUCI’IVE BEHAVIOR IN FISHES

111

Baggerman, B. (1959). The role of external factors and hormones in migration of sticklebacks and juvenile salmon. In “Comparative Endocrinology” (A. Gorbman, ed.), pp. 24-37. Wiley, New York. Baggerman, B. (1960). Factors in the diadromous migrations of fish. Symp. ZOO^. SOC. London 1, 33-60. Baggerman, B. (1962). Some endocrine aspects of fish migration. Gen. Comp. Endocrinol. Suppl. 1, 188-205. Baggerman, B. (1963). The effect of TSH and antithyroid substances on salinity preference and thyroid activity in juvenile pacific salmon. Can. J. Zool. 41, 307419. Baggerman, B. ( 1966). On the endocrine control of reproductive behaviour in the male three-spined stickleback (Gasterosteus acukatus L.). Symp. SOC. Exptl. Biol. 20, 427-456. Ball, J. N. (1960). Reproduction in female bony fishes. Symp. Zool. SOC. London 1, 105-135. Beach, F. A. (1947). A review of physiological and psychological studies of sexual behavior in mammals. Physiol. Reu. 27, 240407. Beach, F. A. ( 1964). Biological bases for reproductive behavior. In “Social Behavior and Organization among Vertebrates” (W. Etkin, ed.), pp. 117-142. Univ. of Chicago Press, Chicago, Illinois. Beach, F. A. ( 1967). Cerebral and hormonal control of reflexive mechanisms involved in copulatory behavior. Physiol. Reo. 47, 289-316. Bern, H. A. (1967). Hormones and endocrine glands of fishes. Science 158, 455-462. Blum, V. (1966). Zur hormonalen Steuerung der Brutpflege einiger Cichliden. Zool. Jahrb., Abt. Allgem. Zool. Physiol. Tiere 72, 264290. Blum, V. ( 1968). Die Auslosung des Laichreflexes durch Reserpin bei dem sudamerikanischen Buntbarsch Pterophyllum scalare. Z . Vergleich. Physiol. 60, 79-81. Blum, V., and Fiedler, K. (1964). Der Einfluss von Prolactin auf das Brutpflegerverhalten von S ymphysodon aequifasciata arelrodi L. P. Schultz ( Cichlidae, Teleostei). Naturwissenschaften 51, 149. Blum, V., and Fiedler, K. (1965). Hormonal control of reproductive behavior in some cichlid fish. Gen. Comp. EndocrinoZ. 5, 186-196. Breder, C. M., and Coates, C. W. (1935). Sex recognition in the guppy, Lebistes reticulatus. Zoologica 19, 187-207. Bretschneider, L. H., and d e Wit, J. J. D. (1947). “Sexual Endocrinology of Nonmammalian Vertebrates.” Elsevier, Amsterdam. Burger, J. W. (1941). Some experiments on the effects of hypophysedomy and pituitary implantations on the male Fundulzls heteroclitus. Biol. BUZZ. 80, 31-36. Chaudhuri, H. (1960). Experiments on induced spawning of Indian Carps with pituitary injections. Indian J. Fisheries 7 , 2048. Chester Jones, I., and Phillips, J. G. (1960). Adrenocorticosteroids in fish. Symp. Zool. SOC. London 1, 1 7 3 2 . Clark, E., and Aronson, L. R. (1951). Sexual behavior in the guppy, Lebistes reticulatus (Peters). Zoologica 36, 49-66. Clemens, H. P., and Grant, F. B. (1964). Gonadal hydration of carp (Cyprinus carpio) and goldfish (Carassius auratus) after injections of pituitary extracts. Zoologica 49, 193-210. Clemens, H. P., and Grant, F. B. (1965). The seminal thinning response of carp (Cyprinus carpio) and rainbow trout (Salmo gairdnerii) after injections of pituitary extracts. Copeia, 1965, No. 2, 174-177.

112

N. R. LILEY

Clemens, H. P., and Inslee, T. (1968). The production of unisexual broods by Tilupia mossambica sex-reversed with methyl testosterone. Trans. Am. Fisheries SOC. $7, 18-21. Clemens, H. P., and Sneed, K. E. (1962). Bioassay and use of pituitary materials to spawn warm-water fishes. U.S.Fish Wildlife Serv. Bur. Sport Fisheries Wildlife Res. Rept. 61, 1 3 0 . Clemens, H. P., McDermitt, C., and Inslee, T. (1966). The effects of feeding methyl testosterone to guppies for 60 days after birth. Copeia, 1966, No. 2, 280-2234. Cohen, H. (1946). Effects of sex hormones on the development of the platyfish, Platypoecilus mucukztus. Zoologica 31, 121-128. Combs, B. D., and Burrows, R. E. (1959). Effects of injected gonadotrophins on maturation and spawning of blueback salmon. Progressive Fish Culturist 21, 165-168. Das, S. M., and Khan, H. A. (1962). The pituitary and pisciculture in India with an account of the pituitary of some Indian fishes and a review of techniques and literature on the subject. Ichthyologica 1, 43-58. Dodd, J. M. (1960). Gonadal and gonadotrophic hormone in lower vertebrates. In “Marshall’s Physiology of Reproduction” (A. S. Parkes, ed. ), 3rd ed., Vol. 1, Part 2, Chapter 11, pp. 417-582. Longmans, Green, New York. Dodd, J. M., Perks, A. M., and Dodd, M. H. I. (1966). Physiological functions of neurohypophysial hormones in submammalian vertebrates. In “The Pituitary ButterGland” (G. W. Hams and B. T. Donovan, eds.), Vol. 3, pp. 57-23. worth, London and Washington, D.C. Egami, N. (1955). Effect of estrogen administration on oviposition of the fish, Olyzius kztipes. Endocrinol. Japon. 11, 89-98. Egami, N. (1959). Preliminary note on the induction of the spawning reflex and oviposition in Oryzius latipes by the administration of neurohypophysial substances. Annotattones Zool. Japon. 32, 13-17. Egami, N., and Ishii, S. (1962). Hypophysial control of reproductive functions in teleost fishes. Gen. Comp. Endocrinol. Suppl. 1, 248-253. Egami, N., and Nambu, M. (1961). Factors initiating mating behavior and oviposition in the fish, Oryzius lutipes. J . Fac. Sci., Univ. Tokyo, Sect. IV 9, 263-278. Fagerlund, U. H. M. (1967). Plasma cortisol concentration in relation to stress in adult sockeye salmon during the freshwater stage of their life cycle. Gen. Comp. Endocrinol. 8, 197-207. Fiedler, K. (1962). Die Wirkung von Prolactin auf das Verhalten das Lippfisches Crenilabrus ocellatus ( Forskil ). Zool. Jahrb., Abt. Allgem. Zool. Physiol. Tiere 69, 6-20. Fontenele, 0. ( 1955). Injecting pituitary (hypophysial ) hormones into fish to induce spawning. Progressive Fish Culturist 17, 71-75. Forselius, S. (1957). Studies of anabantid fishes. Zool. Bids. Uppsak 32, 97-597. Godet, R., and Dup-5, M. ( 1965). Quelques aspects die relations neuroendocriniennes chez Protoptesus annectens (Poisson dipneuste ). Arch. Anat. Microscop. Morphol. Exptl. 54, 319-330. Cottfried, H. ( 1964). The occurrence and biological significance of steroids in lower vertebrates. A review. Steroids 3, 219-242. Gottfried, H., and van Mullem, P. J. (1967). On the histology of the interstitium and the occurrence of steroids in the stickleback (Casterosteus aculeatus L.) testis. Acta Endocrinol. 56, 1-15.

118

TOKI-0 YAMAMOTO

maphroditism, protandrous and protogynous hermaphroditism, to gonochorism. At the outset, it is necessary to clarify the concept of sex. The male and female can best be defined as sperm and egg producers, respectively. This definition, although self-evident, is needed to correct statements that appear frequently in literature, with respect to spontaneous sex reversal in gonochorists (Sections IV, B and C). The term “bisexuality” used to denote hermaphroditism is likely to lead to a misunderstanding. This term should be used to describe gonochorist. The term “intersex” (Goldschmidt, 1915, 1927, 1931) is used in the present review to denote either sporadically appearing or experimentally produced hermaphroditic individuals of a species in which all or nearly all individuals are gonochoristic. Whether hermaphroditism is the more primitive condition from which bisexuality or gonochorism may have arisen or a specialization derived from the more usual vertebrate gonochorism is a matter for debate. The solution of this intriguing problem will require a great deal more information in the future. Nevertheless, fishes provide excellent material to approach the problems of sex differentiation and of evolution of sex among animals. In this study, stress is laid on sex differentiation as a process rather than examining sex phenotypes as they appear in adults. For the latter approach, Gordon’s review (1957) may be consulted.

11. HERMAPHRODITISM

Unlike other vertebrates, a number of teleost fishes are hermaphrodites. Atz (1964), among others, defined the types of hermaphroditism. An individual is hermaphroditic if it bears recognizable male and female tissues. If all, or nearly all, individuals possess both ovarian and testicular tissues, that species is hermaphroditic. Synchronous ( balanced) hermaphrodities are those in which the male and female sex cells ripen at the same time, regardless whether or not self-fertilization is possible. In consecutive ( metagonous) hermaphrodites there are two types: protogynous hermaphrodites that function first as females and then transform into males and protandrous hermaphrodites that transform from males into females. Atz lists 13 families of teleosts, belonging to five orders, that include species of these types. The transformation may be accomplished in several ways, depending upon the arrangement of sexual tissues (Reinboth, 1962, 1967; Smith, 1985, cf. Fig, 3).

114

N. R. LILEY

Johns, L. S., Liley, N. R., and Seghers, B. H. (1969). The effects of gonadectomy on the reproductive behaviour of the blue gourami, Trichogaster trichopterus. In preparation. Jones, J. W., and King, G. M. (1952). The spawning of the male salmon parr (Salmo salar Linn. juv.). PTOC.Zool. SOC. London 122, 615-619. Julesz, M. (1967). Recent results concerning the clinicopathology of the function of the pituitary, adrenal cortex and gonad. In “Symposium on Reproduction” ( K . Lissak, ed.), pp. 125-148. Akad. Kiado, Budapest. Kawakami, M., and Sawyer, C. H. (1959). Induction of behavioral and electroencephalographic changes in the rabbit by hormone administration or brain stimulation. Endocrinology 65, 631-643. Laskowski, W. ( 1954). Einige Verhaltensstudien an Phtypoecilus vaht us. Biol. Zentr. 73, 429-438. Leatherland, J. F., and Lam, T. J. (1969). Effect of prolactin on the density of mucus cells on the gill filaments of the marine form (trachurus) of the threespine stickleback, Gasterosteus aculeatus L. Can. J . Zool. In press. Lehrman, D. S. ( 1965). Interaction between internal and external environments in the regulation of the reproductive cycle of the ring dove. In “Sex and Behavior” (F. A. Beach, ed.), pp. 355-380. Wiley, New York. Levy, M., and Aronson, L. R. (1955). Morphological effects of castration and hormone administration in the male cichlid fish Tilapia macrocephah. Anat. Record 122, 450-451. Liley, N. R. (1966). Ethological isolating mechanisms in four sympatric species of poeciliid fishes. Behauiour Suppl. 13, 1-197. Liley, N. R. (1968). The endocrine control of reproductive behaviour in the female guppy. Poecilia reticulata Peters. Animal Behauiour 16, 318-331. Liley, N. R., and Donaldson, E. M. (1969). The effects of fish pituitary material on the behaviour of hypophysectomized female guppies, Poecilia reticulata Peters. Can. J . Zool. 47, 569573. Machemer, L., and Fiedler, L. (1965). Zur hormonalen Steurung des Schaumnestbaues beim Paradiesfisch, Macropodus operculuris L. ( Anabantidae, Teleostei ) . Naturwissenschaften 52, 648-649. Noble, G. K., and Borne, R. (1940). The effect of sex hormones on the social hierarchy of Xiphophorus helleri. Anat. Record 78, Suppl., 147. Noble, G. K., and Kumpf, K. F. (1936). The sexual behavior and secondary sex characters of gonadectomized fish, Anat. Record 67, Suppl., 113. Noble, G. K., and Kumpf, K. F. (1937). Sex reversal in the fighting fish, Betta splendens. Anat. Record 70, Suppl., 97. Noble, G. K., Kumpf, K. F., and Billings, V. N. (1938). The induction of brooding behavior in the jewel fish. Endocrinology 23, 353-359. Okada, Y. K., and Yamashita, H. (1944). Experimental investigation of the manifestation of secondary sexual characters in fish, using the medaka, Oryzias latipes, as material. J. Fac. Sci., Univ. Tokyo, Sect. IV 6, 383437. Oshima, K., and Gorbman, A. (1966a). OIfactory responses in the forebrain of goldfish and their modification by thyroxine treatment. Gen. Comp. Endocrinol. 7 , 39-09. Oshima, K., and Gorbman, A. (1966b). Influence of thyroxine and steroid hormones on spontaneous and evoked unitary activity in the olfactory bulb of goldfish. Gen. Comp. Endocrinol. 7, 482-491. Oshima, K., and Gorbman, A. (1968). Modification by sex hormones of the spon-

2.

HORMONES AND REPRODUCI’IVE BEHAVIOR IN FISHES

115

taneous and evoked bulbar electrical activity in goldfish. J . Endocrinol. 40, 409-420. Pandey, S., and Leatherland, J. F. ( 1989). Effects of a dithiocarbamoylhydrazine derivative on the histology of the thyroid gland and adenohypophysis of the male guppy, Poecilia reticuluta Peters. In preparation. Picciolo, A. R. (1964). Sexual and nest discrimination in anabantid fishes of the genera Colisa and Trichogaster. Ecol. Monographs 34, 53-77. Pickford, G . E., and Atz, J. W. (1957). “The Physiology of the Pituitary Gland of Fishes.” N.Y. Zool. Society, New York. Raniaswami, L. S., and Lakshman, A. B. (1958). Spawning catfish with mammalian hormones. Nature 182, 122-123. Richards, M. P. M. (1967). Maternal behaviour in rodents and lagomorphs. In “Advances in Reproductive Physiology” (A. McLaren, ed.), Vol. 2, pp. 5 g 1 1 0 . Academic Press, New York. Sage, M. (1968). Respiratory and behavioral responses of Poecilia to treatment with thyroxine and thiourea. Gen. Comp. Endocrinol. 10, 304309. Scharrer, E.,Smith, S. W., and Palay, S. L. (1947). Chemical sense and taste in the fishes Prionotus and Trichogaster. 1. Comp. Neurol. 86, 183-198. Shirai, K. (1962). Correlation between the growth of the ovipositor and ovarian conditions in the bitterling, Rhodeus ocellatus. Bull. Fac. Fisheries, Hokkaido Univ. 13, 137-151. Shirai, K. ( 1964). Histological study on the ovipositor of the rose bitterling, Rhodeus ocellatus. Bull. Fac. Fisheries, Hokkaido Univ. 14, 193-197. Smith, R. J. F. (1967). Endocrine control of prespawning behaviour in two species of sunfish, Lepomis megalotis and L. gibbosus. Ph.D. thesis, Lawson Library, University of Western Ontario, London, Ontario. Smith, R. J. F., and Hoar, W. S. (1967). The effects of prolactin and testosterone on the parental behaviour of the male stickleback Gasterosteus muleatus. Animal Behaviour 15, 342452. Sneed, K. E., and Clemens, H. P. (1959). The use of human chorionic gonadotrophin to spawn warm-water fishes. Progressioe Fish Culturist 21, 117-120. Sneed, K. E., and Dupree, H. K. (1961). The effect of thyroid stimulating hormone combined with gonadotropic hormones on the ovulation of goldfish and green sunfish. Progressive Fish Culturist 23, 179-182. Stanley, L. L., and Tescher, G . L. (1931). Activity of goldfish on testicular substance diet. Endocrinology 15, 55-56. Stevens, R. E. (1966). Hormone induced spawning of striped bass for reservoir stocking. Progressive Fish Culturist 28, 19-28. Sundararaj, B. I., and Goswami, S. V. (1966). Effects of mammalian hypophysial hormones, placental gonadotrophins, gonadal hormones, and adrenal corticosteroids on ovulation and spawning in hypophysectomized catfish, Heteropneustes fossilis (Bloch). 1. Ezpptl. Zool. 161, 287-296. Tavolga, M. C. (1949). Differential effects of estradiol, estradiol benzoate and pregnenolone on Platypoecilus maculatus. Zoologica 34, 215-237. Tavolga, W. N. (1955). Effects of gonadectomy and hypophysectomy on prespawning behavior in males of the gobiid fish Bathygobius soporator. Physiol. Zool. 28, 218-233. Tavolga, W. N. (1956). Visual, chemical and sound stimuli as cues in the sex discriminatory behavior of the gobiid fish Bath ygobius soporator. Zoologica 41, 49-64.

116

N. R. LILEY

van den Assem, J. ( 1967). Territory in the three-spined stickleback (Gasterosteus aculeatus L. ) , An experimental study in intraspecific competition. Behaoiour Suppl. 16, 1-164. van Iersel, J. J. A. (1953). An analysis of the parental behavior of the male threespined stickleback ( Gasterosteus aculeatus L. ). Behauiour Suppl. 3, 1-159. van Mullen, P. J. (1967). On synchronization in the reproduction of the stickleback ( Gasterosteus aculeatus L. forma leiura Cuv. ) . Arch. N e e d Zool. 27, 258-274. Vivien, J. H. (1941). Contribution a I'6tude de la physiologie hypophysaire dans ses relations avec l'appareil ghital, la thyroide et les corps suprarknaux chez les poisons sblacians et t616ostkns Scylliorhinus canicula et Gobius paganellus. Bull. Bid. France Belg. 75, 257-309. Wai, E. H., and Hoar, W. S. (1963). The secondary sex characters and reproductive behaviour of gonadectomized sticklebacks treated with methyl testosterone. Can. J. Zool. 41, 611-628. Wai, E. H., and Hoar, W. S. (lSe8). Unpublished study. Wiebe, J. P. (1967). The reproductive physiology of the viviparous seaperch Cymtogaster aggregata Gibbons. Ph.D. thesis, University of British Columbia, Vancouver, Canada. Wiebe, J. P. (1968). Inhibition of pituitary gonadotropic activity in the viviparous seaperch Cymatoguster aggregata Gibbons by a dithiocarbamoylhydrazine derivative (I.C.I. 33,828). Can. J. Zool. 46, 751-758. Wilhelmi, A. E., Pickford, G. E., and Sawyer, W. H. (1955). Initiation of the spawning reflex response in Fundulus by the administration of fish and mammalian neurohypophysial preparations and synthetic oxytocin. Endocrinology 57, 243-252. Wootton, R. J. (1968). A comparison of measures of aggression in the male threespined stickleback. Ph.D. thesis, University of British Columbia, Vancouver, Canada. Yamamoto, T. (1962). Hormonic factors affecting gonadal sex differentiation in fish. Gen. Comp. Endocrinol. Suppl. 1, 341355. Yamamoto, T., and Matsuda, N. (1963). Effects of estradiol, stilbestrol and some alkyl-carbonyl androstanes upon sex differentiation in the medaka, Oryzias kztipes. Cen. Comp. Endocrinol. 3, 101-110. Yamazaki, F. (1962). Effects of hypophysectomy on the ovulation, oviposition and sexual behavior in the goldfish, Carassius auratus. Bull. Fac. Fisheries, Hokkaido Unio. 13, 3946. Yamazaki, F. (1965).Endocrinological studies on the reproduction of the female goldfish Carussius uuratus L., with special reference to the function of the pituitary gland. Mem. Fac. Fisheries, Hokkaido Uniu. 13, 1-64. Yamazaki, F., and Donaldson, E. M. ( 1968a). The spermiation of goldfish (Carassius auratus) as a bioassay for salmon ( Oncorhynchus tshawytscha) gonadotropin. Gen. Comp. Endocrinol. 10, 383-391. Yamazaki, F., and Donaldson, E. M. (1968b). The effects of partially purified salmon pituitary gonadotropin on spermatogenesis, vitellogenesis, and ovulation in hypophysectomized goldfish ( Carussius auratus) , Gen. Comp. Endocrinol. 11, 292-299. Yamazaki, F., and Donaldson, E. M. ( 1 9 6 8 ~ )Personal . communication.

3 SEX DIFFERENTIATION TOKI-0 YAMAMOTO It thus becomes of great interest to discaver the mechanism by which sex is determined. and to find whether by any means we can bring it under our control . Julian Huxley ( 1938) I . Introduction: Sexuality in Fishes . . . . . . . I1. Hermaphroditism . . . . . . . . . . A . Synchronous Hermaphroditism B . Consecutive Hermaphroditism . . . . . . . I11. Gonochorism . . . . . . . . . . . A . Undifferentiated Gonochorists . . . . . . . B. Differentiated Gonochorists . . . . . . . C . All-Female Species . . . . . . . . . IV . Genetic Basis of Sex Determination . . . . . . A . XX-XY and WZ(Y)-ZZ(YY) Types . . . . . . B. Polygenic Sex Determination and So-called . . . . . . . . Genetic Sex Reversal . C. “Spontaneous Sex Reversal” in the Swordtail . . . . V. Control of Sex Differentiation . . . . . . . . A . Surgical Operation . . . . . . . . . B . Modification of Sex Differentiation by Sex Hormones . . C. Complete (Functional) Reversal of Sex Differentiation . . VI . Nature of Natural Sex Inducers . . . . . . . A . Steroid versus Nonsteroid Theories . . . . . . B . Detection of Steroids and Relevant Enzymes in Fish Gonads . . . VII . Differentiation of Secondary Sexual Characters . VIII . Summary . . . . . . . . . . . . References . . . . . . . . . . . . .

. . . . . . .

117 118 119 120 127 127 129 129 131 132 134 139 1 4 142 142 144 150 150 153 153 157 158

.

I INTRODUCTION: SEXUALITY IN FISHES

Members belonging to the class Pisces exemplify an almost complete range of the various types of sexuality from synchronous her117

118

TOKI-0 YAMAMOTO

maphroditism, protandrous and protogynous hermaphroditism, to gonochorism. At the outset, it is necessary to clarify the concept of sex. The male and female can best be defined as sperm and egg producers, respectively. This definition, although self-evident, is needed to correct statements that appear frequently in literature, with respect to spontaneous sex reversal in gonochorists (Sections IV, B and C). The term “bisexuality” used to denote hermaphroditism is likely to lead to a misunderstanding. This term should be used to describe gonochorist. The term “intersex” (Goldschmidt, 1915, 1927, 1931) is used in the present review to denote either sporadically appearing or experimentally produced hermaphroditic individuals of a species in which all or nearly all individuals are gonochoristic. Whether hermaphroditism is the more primitive condition from which bisexuality or gonochorism may have arisen or a specialization derived from the more usual vertebrate gonochorism is a matter for debate. The solution of this intriguing problem will require a great deal more information in the future. Nevertheless, fishes provide excellent material to approach the problems of sex differentiation and of evolution of sex among animals. In this study, stress is laid on sex differentiation as a process rather than examining sex phenotypes as they appear in adults. For the latter approach, Gordon’s review (1957) may be consulted.

11. HERMAPHRODITISM

Unlike other vertebrates, a number of teleost fishes are hermaphrodites. Atz (1964), among others, defined the types of hermaphroditism. An individual is hermaphroditic if it bears recognizable male and female tissues. If all, or nearly all, individuals possess both ovarian and testicular tissues, that species is hermaphroditic. Synchronous ( balanced) hermaphrodities are those in which the male and female sex cells ripen at the same time, regardless whether or not self-fertilization is possible. In consecutive ( metagonous) hermaphrodites there are two types: protogynous hermaphrodites that function first as females and then transform into males and protandrous hermaphrodites that transform from males into females. Atz lists 13 families of teleosts, belonging to five orders, that include species of these types. The transformation may be accomplished in several ways, depending upon the arrangement of sexual tissues (Reinboth, 1962, 1967; Smith, 1985, cf. Fig, 3).

3.

SEX DIFFERENTIATION

119

A, Synchronous Hermaphroditism Dufoss6 (1854, 1856) found that Serranus scriba (Serranidae) is a synchronous, functional hermaphrodite. This was confirmed by van Oordt ( 1929). D’Ancona’s intensive studies ( 1949a,c,d, 1950) revealed that other Mediterranean serranids such as S. cabrilla and Hepatus hepatus belong to this category. The gonad of these fishes is separated into the ovarian and testicular areas (Fig. 1).D’Ancona ( 1949a,c,d, 1950) suggested the possibility of self-fertilization. Clark (1959, 1965) in S . subligerius, a Florida serranid proved that self-fertilization and development are possible. In captivity, an isolated individual can emit sperm and fertilize its own eggs. In both nature and in the aquaria containing two or more fish, the fish may form spawning pairs. Immediately after one fish in the female phase spawns, the partner fertilizes the eggs. At this moment, the color pattern (vertical stripes) in the female phase changes to that of the male phase. Then, the first fish reverses its sexual role and acts as a male. Of serranid fishes from Bermuda studied by Smith (1959), four species (one belonging to the genus Hypoplectrus and three of the genus

Fig. 1. Transverse section of the gonad of the synchronous hermaphrodite, Serranus scribn (Serranidae). Upper portion is the ovary and lower region the testis. After D’Ancona ( 1950).

120

TOKI-0 YAMAMOTO

Prinodus) are synchronous hermaphrodites while at least nine species are protogynous hermaphrodites (see Section 11, B, 2). Harrington (1961, 1965) demonstrated that most Aiuulus marmomtus, an oviparous cyprinodont, are genuine hermaphrodites capable of internal self-fertilization. It is remarkable that over 10 uniparental generations have been propagated and each fish has been kept in lifelong isolation ab ouo. Tissue grafts between the parent and its offspring and among siblings were histocompatible, thus providing tight evidence of self-fertilization. According to Harrington ( 1967), low temperature ( 18"20°C)tends to transform the hermaphrodites to males. The order Perciformes includes four families ( Serranidae, Sparidae, Centracanthidae, and Labridae) in which either synchronous or consecutive hermaphroditism occurs normally. In all these families, however, there are some gonochoristic species. Mead (1960) and Gibbs (1960, cited in Atz, 1904) discovered that several fishes of four families belonging to the order Myctophiformes (also called Iniomi) are hermaphroditic (cf. Atz, 1964), Some of Meads fish gave evidence of being synchronous hermaphrodites. However, at least five of the remaining dozen families in the order are gonochoristic.

B. Consecutive Hermaphroditism 1. PROTANDROUS HERMAPHRODITES Since the time of Syrski (1876) and Brock (1879), it has been known that the Mediterranean bream, Sparus auratus, is a protandrous hermaphrodite. Pasquali (1941) and D'Ancona (1941) described the precise process of gonadal differentiation from male to female phases. They demonstrated that its gonad consists of both testicular and ovarian areas from a very young stage. In smaller fish the lateroventral testicular region predominates over the ovarian zone, and in larger fish the reverse is true (Fig. 2). Similar patterns have been reported by Syrski ( 1876), McLeod (lSSl), Hoeck (1891a,b), Williamson (1910), and van Oordt (1929). The latter author found in Sargus (Diplodus) annularis that there are some individuals in which the sexes are separated, Le Gall (1929) found that Pagellus centrodontus is a protandrous hermaphrodite. D'Ancona (1949a-d, 1950, 1956) showed that some sparids such as Sparus auratus, Sargus (Diplodus) sargus, and Pagellus mormyrus are protandrous hermaphrodites (see Fig. 2), while others such as S. annuluris, S. uulgaris, Puntazzo ( C h a r m ) puntazzo, Boops boops, Obludu

3.

SEX DIFFERENTIATION

121

Fig. 2. Transverse sections of the gonad of the protandrous hermaphrodite, Sparus auratus (Sparidae). ( A ) Male phase, ( B ) transitory phase, and ( C ) female phase.

Here Ov stands for ovary and Tes for testis. After D’Ancona ( 1950).

mlanulu, and Dentex dentex are rudimentary hermaphrodites. Pagellus a c a m is protandrous ( Reinboth, 1962). The classification of Japanese sparids is notoriously unsatisfactory. Here, the writer follows Dr. Abe’s recent personal communication ( 1967). Kinoshita (1936) reported “sex reversal” from male to female in Acanthopagrus schlegeli (syn. Sparus longispinis, Mylio mucrocephalus). He ( 1939) also reported sex reversal in A. ( Sparus, Mylio) latus and Sparus sarba (Sparus arks, Rabdosargus sarba). He remarked that not all individuals transform from male to female, i.e., some males retain maleness even when they become large. Okada’s histological study (1952b) indicates that an ambisexual organization is present in the early stages, so that the situation is similar to the reports of DAncona and others. In the flat-head fish, Inegocia (Cociella) crocodila (Platycephalidae), Aoyama et al. (1963) examined a large number of individuals ( >lOOO) caught by trawls in the East China Sea and the Yellow Sea. Small individuals had testes and medium-sized fish hermaphroditic gonads with functional testes, whereas large individuals had ovaries. Another flat-head fish, Inegocia (Suggrundus) merderuoort, is also a protandrous hermaphrodite (Aoyama and Kitajima, 1966). Okada (1966) postulated that this form repeats the hermaphroditic state. In Gonostomu gracile ( Gonostomatidae ) , a deep-sea luminescent fish, Kawaguchi and Marumo (1967) found that individuals less than 7 cm are mostly males and those more than 9 cm are invariably females. Sex

122

MKI-0 YAMAMOTO

succession takes place in the medium-sized fish (7-9 cm), and hermaphrodites are frequently found in the specimens of 6 7 cm. 2. PROTOGYNOUS HERMAPHRODITES

Certain fishes belonging to the Sparidae, i.e., Pagellus erythrinus ( DAncona, 1950; Larrafieta, 1953, 1964) and Spondyliosomu (Cantharus) canthurus ( D’Ancona, 1950, 1956; Reinboth, 1962) are protogynous hermaphrodites. Among Japanese sparids, Dentex ( Taius) tumifrons seems to be protogynous. Inversion of sex is brought about by development of the testicular region and regression of the ovarian part. In Dentex ( Taius) tumifrons ( now Sparidae), Aoyama ( 1955) found fishes with hermaphroditic gonads. He concluded that some of the females change into males. Of serranids from Bermuda four species were synchronous hermaphrodites, as already stated, while at least nine species belonging to the genera Epinephelus, Mycteroperca, Alphes, Petrometopon, and Cephalopholis were protogynous hermaphrodites ( Smith, 1959). Smith pointed out that there are three patterns of hermaphroditism in serranids: the Serranus, Rypticus, and Epinephelus type (Fig. 3 ) . The patterns are essentially similar to those hermaphrodites reported by Reinboth (1967; cf. Fig. 3 ) . Epinephelus and its allies have gonads in which the male tissue is present throughout the germinal epithelium lining the central lumen of the gonad. This male tissue becomes functional only after the female tissue has ceased to function. The genus Rypticus shows an intermediate type of gonad in which scanty male tissue is present in the lower part but is also found intermixed with the female tissue. The protogynous Chelidoperca hirundinacea is also of this type ( Reinboth, 1967). The eastern Pacific Paralabrax chthratus seems to be secondarily gonochoristic. Here, we can make an inference on the process in evolution of gonochorism from hermaphroditism. Figure 3 illustrates the three types. Kuroda (1931) postulated that larger and red Sacura margaritacea (Serranidae) and smaller and yellow S. pulcher are males and females, respectively, of the same species. He found some intermediate individuals. The protogynous hermaphroditism in this species was demonstrated by Reinboth ( 1963) and Okada ( 1965a,b). Protogynous sex reversal in this species seems to occur by degeneration of ovarian tissue after spawning. The Atlantic sea bass, Centropristes striatus ( Serranidae), is also a protogynous hermaphrodite ( Lavenda, 1949; Reinboth, 1965). In Centracanthidae ( Maenidae ) , three species Spicara smuris, S . chryselis, and S . m e n u , of the genus Spicara (Maena) are known to occur in the

Fig. 3. Gonadal development and phylogenetic relations in serranid fishes, Redrawn after Reinboth (1987).

124

TOKI-0 YAMAMOTO

Mediterranean. Zei (1949) has postulated that the two species S . smuris and S . chryselis are protogynous hermaphrodites. Lozano Cab0 (1951, 1953) confirmed this in S. smaris and concluded that S. smuris and S . akedo are two sexes of the same species. Lepori (1960) also conchded that S . maena and S . chryselis are protogynous species. Reinboth (1962) performed an elaborate study on S. maenu. For a long time the greenish and reddish Japanese wrasses (Labridae) have been regarded as different species; the former was called Julis poecilopterus and the latter I. pyrrhogrammu. Jordan and Snyder (1902) suggested that the two forms might be sex variants of the same species, Halichoeres poecilopterus. Y. Kinoshita ( 1934,1935,1936) has verified this experimentally. He observed that the blue wrasse rapidly loses its secondary sex characters after castration, while the red wrasse is not affected by ovariotomy. Transplantation of a testis into the red wrasse transforms it into a blue wrasse. However, transplantation of an ovary into the blue wrasse causes no change in coloration of the recipients. Kinoshita found testis-ova in some individuals and postulated that this fish may be protogynous. Okada (1962) corroborated this. He (1964b) performed experiments on the effects of androgen and estrogen on sex reversal. The Mediterranean labrid fishes Cork giofredi and C . julis have long been regarded as separate species. The two types differ in both size and color (Fig. 4 ) . Smaller individuals (C. giofredi) are usually females and larger ones (C. julis) are males. Some medium-sized fish show intermediate color and have testes (Bacci and Razzauti, 1957, 1958). The two types are the same species, C. julis, a protogynous hermaphrodite. An interesting fact is that a few large fish with C. giofredi coIor are nevertheless maIes when their gonads are examined. Because each individual is at all times either a male or female, this labrid is termed a “false gonochorist.” The two species of labrids in the waters of the Mediterranean around Livomo, Labrus turdus and L. merula, are also protogynous species Sordi, 1962). In both species, all individuals less than 27 cm have ovaries. In the largest fishes of L. turdus, all individuals have testes. In L. merula, the largest individuals have a balanced sex ratio. This seems to indicate that while 50%of the fish change from female to male, 504: never change but remain females. In both Coris julis and a Caribbean labrid, Thalassomu bifasciatum, Roede (1965) observed 70% 0 o 30%d d in the young and 100%d d in adults. Reinboth (1962) pointed out the existence of two types of morphologically distinct males in labrids (Coris julis and Haliochoeres poecilopterus). The primary male looks like a female but remains a male through-

+

3. SEX

DIFFERENTIATION

Fig. 4. Change of color and sex phase in the labrid, Cork iulis. After Bacci, 1966: “Sex Determination,” Fig. 7.1. Reprinted with permission from the author and Pergamon Press.

out its life, and the secondary male that changes to a male from the female. However, according to Vandini (196!5), there are no primary males that retain the C . giofredi color throughout their life cycle, viz., even the C . gwfredi males eventually develop C. iulis color. In the striped wrasse, Labrus ossifagus, there are some individuals with a red and others with a blue pattern. It was believed that the former is the female and the latter the male. Lonneberg and Gustafson (1937) reported that females greatly outnumber males in red fish and the reverse is true in blue-striped fish. They also found intersexual individuals changing from femaIes to males. This seems to indicate that this form is protogynous and most of these fish, if not all, exhibit sex succession. Liu (1944) discovered that small individuals of the synbranchoid eel, Monopterus albus ( M . jauanensis), are females and the large ones males and offered a good case for protogynous hermaphroditism (cf. also Bullough, 1947). Liem (1963, 1985) confirmed this using nearly 1000 fish, and

126

TOKI-0 YAMAMOTO

concluded that every individual starts its reproductive cycle as a functional female and then becomes a functional male. The sexuality of this fish has the following sequence: juvenile hermaphrodite + functional female + intersex + functional male. In consecutive hermaphrodites such as Sparidae and Serranidae, either protandrous or protogynous, there is a basic feature in common. The juvenile gonad has an ambisexual organization. From the very beginning of differentiation of the gonad, ovarian and testicular rudiments are present in every fish, although the topographical arrangement of the male and female tissues and change of the dominant tissue in time differ from species to species (cf. Figs. 3 and 5 ) . A question arises as to whether all individuals in a consecutive hermaphroditic species inverse sex in sequence. In observations made on natural populations, occasional individuals are found smaller or larger than is usual for this sex. Smith (1959) noted unusually large females in some protogynous serranids, and Reinboth (1965) reported unexpectedly small males of Centropristes striatus, also a protogynous species. In Dentex dentex only some individuals undergo a transitional hermaphroditic stage ( Lissia Frau, 1964). According to Larrafieta ( 1964) about 5% of the fish in a population of Pagellus erythrinus are males throughout their life, while 45%transform from female to male and 50% never transform but remain females. Rijavec and Zuvanovic (1965) obtained comparable results in the same species. Smith (1967) proposed a

Fig. 5. Schematic transverse sections of early developmental stages ( A + E ) of the gonad in Sparidae, where ov stands for ovogonia; sp, spermatogonia; gc, gonocoelom; ef, efferent ducts; lc, lacunae; and bv, blood vessels. Redrawn from D’Ancona ( 1950).

3.

SEX DLFFERENTIATION

127

theory of hermaphroditism in which he postulated that sex reversal takes place in different individuals at different sizes and ages viz., sexual succession is a prolonged continuous process for the population. “During the first time interval following sexual maturity, only a part of the population changes sex, and during each time period a fraction of the remaining individuals transform.” He interpreted Larraiieta’s results as indicating that 5%of the individuals change before sexual maturity and that the life span is such that half the females do not live to change. Researches on consecutive hermaphrodites have been mostly based on observations on natural populations. There are only a few experimental or physiological strides and these are not satisfactorily documented. Hence, we know little about the cause of sex reversal. It seems that sex reversal may be caused by sex hormone imbalances. In the female phase the female hormone might dominate the male hormone and in the male phase the reverse might be true.

111. GONOCHORISM

As Witschi (1914a,b, 19.30)has pointed out for amphibians so also in gonochoristic fishes there are “undifferentiated and “differentiated species. In the former, the indifferent gonad first develops into an ovarylike gonad and then about one-half of individuals become males and the other half females. In the latter, the indifferent gonad directly differentiates into either a testis or an ovary. In both types, sex differentiation seems to be brought about by male- and female-inducing substances (Section VI, A ) . It is natural that undifferentiated species are more unstable than differentiated ones in sexuality. Unfortunately, however, the two types have been studied embryologically in only a few fishes. In the absence of embryological evidence, it is as yet hazardous to correlate the two modes with the occurrence of sporadic intersexes which represent a remnant of the embryological condition manifested late in some adults. However, a few species in which undifferentiated and differentiated conditions are known provide evidence that spontaneous intersexes among gonochorists occur mostly in the “undifferentiated” species while in the “differentiated ones the occurrence of intersex is rarely or never seen. A. Undifferentiated Gonochorists The lampreys and hagfishes ( Agnatha ) are undifferentiated species. Sexuality of the brook lamprey, Lurnpetra lumottei (Entosphenus

128

TOKI-0 YAMAMOTO

wildmi), has been studied by Lubosch (1903) and more extensively by Okkelberg (1921). Ammocoete larvae up to 3.5 cm possess indifferent gonads. After this stage, some of the larval gonads contain both male and female germ cells. Of 15 adult males (sperm producers) undeveloped ova were found in the testis of seven. In ammocoete larvae of L. planeri and L. fluuiatilis, differentiation of the gonads is mostly completed at 5-6 cm in length. Hardisty (1960) recorded the sex ratio in animals longer than 6 cm. In L. planeri ammocoetes there is a small but significant excess of females. There are a number of transitional individuals with disintegrating oocytes and developing stroma, which may eventually become males. Among hagfish, Atlantic species, Myxine glutinosa, seems to be undifferentiated. Conel's ( 1917) and Schreiner's (1904, 1955) reports seem to indicate that this species is an intermediate type between hermaphroditism and gonochorism, in which juvenile hermaphrodites are common. The Japanese hagfish, Eptatretus burgeri, is also an undifferentiated gonochorist. Dean (cited by Conel, 1931) found only one male with an ovotestis out of 569 fish. Turning to the class Pisces, there are many undifferentiated gonochorists which normally develop into either male or females. However, sporadic intersexes are known to occur in these fishes. It is beyond the scope of the present chapter to cite a vast number of instances; Freund (1923) and Atz ( 1964) have listed numerous cases. Grassi (1919) in the eel, Anguilla anguilla, and Mrgid (1923, 1930) in the rainbow trout, Salmo gairdneri irideus, have demonstrated that these species are undifferentiated. In the latter, accidental intersexes have been reported not infrequently. In Gadidae, sporadic intersexes have also been frequently reported. In the herring, Clupea harengus, there are several types of the intersexual gonad (Gabler, 1930; Rudolf, 1931). In the minnow, Phoxinus laeuis, Bullough (1940) found 10 fish with intersexual gonads. In all cases the ovarian portion was suppressed while the testicular region was normal. He thought that these intersexes represented transitory stages changing from female to male. The effects of an androgen and an estrogen supported this postulation. However, a complete sex reversal was not achieved with these hormones. These are three species of the paradise fishes: Macropodus operculark,M . chinensis, and M . concolor; all three are undifferentiated species. Whereas the male index ( o o J o o + d d ) of M. opercularis and M . chinensis is 50$, that of M . concolor varies from 68% to 91%;thus there is a preponderance of males (Schwier, 1939). The presence of multiple sex factors in M . concolor is apparent. While the offspring from M . opercularis X M . chinensis are fertile, males from M . concolor X M . chinensis are sterile.

3.

SEX DIFFERENTIATION

129

The live-bearing cyprinodont Poecilia ( formerly Lebistes) reticulata is an undifferentiated gonochorist ( Goodrich et al., 1934; Dildine, 1936). Blacher (1926a) and Winge (1927) found a few aged intersexes with an XX constitution. Eggert’s report (1931, 1933) on intersexes of the mudskipper, Periophthalmus vulgaris ( Gobiidae) , seems to suggest that there are two geographical races in the same species. In the fish on the south coast of Java, the two sexes differentiate independently from the start, whereas on the north coast the testes of young have basically ovarian structure indicating that the male pass through the female phase. However, number and sizes of specimens examined by him were very limited.

B. Differentiated Gonochorists Only a few fishes have been shown to be differentiated gonochorists. Sexuality of differentiated species is fairly stable among fishes. Wolf ( 1931) demonstrated that the “domesticated platyfish, Xiphophows ( formerly Pktypoecilus ) maculatus, is a differentiated species. Bellamy and Queal (1951) stated that not a single sporadic intersex has ever been found among 50,000 specimens. The situation is the same in the medaka, OTYZ~US latipes (Cyprinotontidae ), No spontaneous, true intersex has ever been found among more than thirty thousand fish studied during 40 years. Yamamoto (1953) showed that this species is a differentiated gonochorist. In this connection, Oka’s report (1931b) on the occurrence of oviform cells in three males of this species may be mentioned. His finding was based on fish which had been left to starve for 3 months. It is unfortunate and even strange that the title of this paper included the words “accidental hermaphroditism” since Oka carefully observed the cytological difference between these enlarged cells and genuine ovocytes, calling them “pseudoovocytes.” It is probable that these cells are enlarged proto- or spermatogonia prior to degeneration, As to the so-called testis-ova in this species, induced by various agents, comments are given in Sections V, C and VI, A.

C. All-Female Species The viviparous toothcarp, Poecilia ( Mollienisia) formosa, which is thought to be a form of hybrid origin, inhabits northeastern Mexico and southwestern Texas. The “all-female’’ phenomenon in this form was first found by C. L. Hubbs and Hubbs (1932, 1946). In natural habitats,

130

TOKI-0 YAMAMOTO

it propagates itself by mating with sympatric, related bisexual species, either P . latipinnu or P . sphenops. However, paternal characters are not transmitted to off spring which are solely females. Subsequently, this fact has been confirmed by Meyer (1938), C. Hubbs et al. (1959), Haskins et al. (lWO), C . Hubbs (1964), Kallman (1962, 1964a,b). C. L. Hubbs and Hubbs, and Kallman concluded that the monosexuality is the result of gynogenesis, the mate providing a stimulus to activate development of the ovum without syngamy. Nevertheless, there is strong evidence that the species is diploid (cf. C. Hubbs et al., 1959; Kallman, 1962). In terms of DNA levels, P. formosu nucleus is the same as in the diploid spaces P. sphenops (Rasch et al., 1965). C . Hubbs et al. (1959) discovered a single wild male in the Brownville population, and Haskins et al. (1960) reported a single male in their laboratory stock. It proved fertile in mating to P . formosu siring all females. The appearance of extremely rare males is still a matter for debate. In this connection, Kallman’s study (1964b) deserves attention. He found ammg thousands of off spring from gynogenetic P . formo~amated with males of P . vituttu or P . sphenops 14 exceptional fish of which 12 exhibited paternal patterns and possessed morphologically intermediate features while the remaining two had mosaic patterns with some areas patriclinous and others matriclinous. This shows that in rare cases syngamy occurs or a single chromosome from the male nucleus, governing pattern formation, may become accidentally incorporated into some or all cells of developing embryos giving rise to mosaic patterns. Gynogenesis has been reported in certain natural populations of the crucian carp, called the “silver goldfish,” Carrussius auratus gibelio, which produce all-female progeny (Lieder, 1955, among others). He regarded this phenomenon as a natural parthenogenesis. Artificial insemination revealed that a gynogenetic stimulus to the ova can be provided by the common goldfish, C. aurutus uuratus, or by the carp, Cyprinus carpio. Spurway ( 1953) claimed the occurrence of “spontaneous parthenogenesis” in two anomalous guppy females. Later ( Spurway, 1957) this postulation was withdrawn and she assumed that the phenomenon results from the self-fertility of “functional” intersexes. Subsequent to Spurway’s study, Stolk ( 1958) reported pathological gynogenesis observed in the guppy and the swordtail. The ovary of each female was infested by a phycomycete fungus, Ichthyophorus hoferi, which apparently was responsible for the pathological activation of the ova. More than 64 broods were produced by these fishes; these without exception were daughters. Stolk criticized Spurway’s finding by saying that her fish might also be infected by the parasite. The mode of reproduction of the two unisexual “species” or strains of the genus Poeciliopsis, viviparous toothcarps of northwestern Mexico,

3.

SEX DIFFERENTIATION

131

is unique to vertebrates (Miller and Schultz, 1959; Schultz, 1961, 1966). At first, it was considered that each of the two undescribed species had both bisexual and unisexual strains. Bisexual strains were tentatively referred to as C and F and unisexual ones as Cx and Fx, respectively. Miller (1960) identified the C and Cx as P . lucida, but F and Fx are still undescribed. Later, Schultz (1966) showed that monosexual Cx is a form different from the bisexual P . lucidu but closely allied to it. In nature, Cx propagates itself by mating with P . lucida. Hybridization experiments between Cx and males of two bisexual species P . lucidu or P . latidens provided evidence that genetic factors of both parent combine to form the F1 offspring. However, the entire male genome appears to be eliminated during ovogenesis, probably at meiosis. Hence, ova have no paternal genome in each generation. Besides Cx, Schultz ( 1967) further reported two additional all-female stocks, Cy and Cz, which were previously thought to be Cx. Poecilia Zucida provides sperm for these monosexual forms. While Cx and Cz are diploid expressing characteristics of both parents in the F,, Cy is a triploid and in mating with males of various bisexual species, e.g., P . latidens, produces all-female, triploid offspring by gynogenesis devoid of paternal characters. Thus, the mode of reproduction of Cy is quite different from that of Cx and Cz in which paternal chromosomes of the F, generation are not transmitted through the ova; only those characteristics of the female germ line pass to the next generation. In gynogenetic diploid forms, the diploid complement might be maintained by suppression of one of the meiotic divisions, reentry of the second polar body, or suppression of the first mitotic cleavage. The means by which the triploid Cy undergoes meiosis and produces fertile triploid eggs is obscure. At this point, it may be mentioned that Rasch et al. (1965) mated the gynogenetic, diploid Poecilia formosa to P. vittata and obtained offspring which were triploids as judged by their nuclear DNA. However, these triploids were sterile. IV. GENETIC BASIS OF SEX DETERMINATION The genetic study of sex is important not merely because sex is instinctively our major interest, but it lies at the root of Mendelian heredity itself and is the major factor in evolution. It provides such admirable material for the study of gene interaction, of phenogenetics, that is, of developmental physiology. Abbreviated from H. J. Muller (1932)

The classic sex factor studies use notations such as FF = ? ,FM = d , heterogametic species, and F M = 0 , MM = d, where M > F in

132

TOKI-0 YAMAMOTO

where F > M in o heterogametic forms, which are oversimplified and are not generally found in nature, if the F and M symbolize single sex determiners (genes), Hartmann ( 1951, 1956) has proposed the formulas ( 8 G ) F F = o and (A@)MF = d , where the M and F represent the sex determiners (realizers), the AG stands for ambisexual potencies and the symbols # and $ represent the inhibiting effects exerted by the combinations of sex realizers. The supposed existence of the double system is also a formalization and has been a matter for debate since experiments on the localization of the AG complex have not been fruitful. Kosswig (1964) is also an opponent to these formulas.

A. XX-XYand WZ(Y ) -ZZ( YY) Types At this point, it may be appropriate to comment on the designation of sex chromosomes. The symbols in male heterogametic forms, that is, XX for female and XY (or XO) for male, are in universal agreement. Fishes which were found to have this type are listed in Table I. Among fishes, those which have this type are more numerous than those with female heterogamety. On the other hand, in organisms with female heterogamety, some Table I Fishes with Male Heterogamety (XX 0 , XY 3 ) Based Mostly on Genetic Evidence Species Oryzias latipes Poecilia" reticulata Poeciliab nigrofasciata Xiphophorus variatus X . ziphidium X . maculatusc X . couchianus X . milleri X . montszumae cortezi Betta splendens Mogrunda obsculad COttllS polluxd Carassius auratus'

Family

Author

Year

Cyprinodontidae Poeciliidae Poeciliidae Poeciliidae Poeciliidae Poeciliidae Poeciliidae Poeciliidae Poeciliidae Anabantidae Gobiidae Cottidae Cyprinidae

Aida Winge Breider Kosswig Kosswig Gordon Gordon and Smith Kallman Kosswig Kaiser and Schmidt Nogusa Nogusa Yamamoto and Kajishima

1921 1922 1935 1935 1935 1946 1938 1965 1959 1951 1955 1957 1969

Formerly Lebistes. Formerly Limia. Mexican populations. d Cytologically established. Cultivated goldfish.

a b 0

(I

3.

133

SEX DIFFERENTIATION

Table I1 Symbols of Sex Chromosomes in the ? Heterogametic Platyfish, Xiphophorus maculatzcs" ~

0

3

Author

Year

Stock

XY

wz

xx zz

1922 1951 1926

Domesticated Domesticated Domesticated

XY WY WY WY

YY YY YY YY

Bellamy Bellamy and Queal Gordon, Kosswig, Breider, Bellamy Castle Gordon Gordon Kallman

1936 1947 1950 1965

Domesticated Domesticated Belize river Rio Hondo

a

From several authors, but chiefly Gordon (1952)

authors use the formula WZ( OZ) P -ZZd, while others, including outstanding geneticists, use the formula XY (or XO) ?-XXd. The latter denotation is not only perplexing but also is based on the misleading assumption that the Z = X and W = Y, which is in contradiction to the experimental facts revealed in fishes. This notation becomes seriously confusing when matings are made between male heterogametic and female heterogametic species or races. In Table 11, symbols used for the female heterogametic platyfish by several authors are listed. As a result of the studies that follow, the symbol XY P -XX$ should be eliminated. In connection with the Bellamy's experiment (1936) of the interspecific matings between P heterogametic Xiphophorus maculatus and heterogametic X . variatus, Castle (1936) suggested use of the terms XY for P and YY for in the 9 heterogametic species, where the Y is equivalent to the Z and the X to the W. In fact, Gordon (1946a,b, 1947a) proved experimentally in xiphophorin fishes that the Z = Y. However, since there was no evidence that W = X, he suggested that the formula WZ 0 -ZZd might have been better written as WY Q -YY d. German authors continue to use the WZ-ZZ system, whereas American investigators, particularly Kallman (1965b), stress that it is best to eliminate the symbol Z. The W of ? heterogametic forms is considered to have strong F gene(s), and the Y of a" heterogametic forms is considered to have strong M gene(s). In this connection, it may be noted that in the interspecific mating between o heterogametic X . maculatus ( WZ-ZZ) and 8 heterogametic X. variatus (XX-XY), Kosswig (1935a) stated that the Z of the former behaved like the Y of the latter, although at that time he considered that Z = X and W = Y. Heterosomal sex-determining mechanisms have often been called

134

TOKI-0 YAMAMOTO

“monogenic.” This term is inadequate at least in fishes since sex is “determined indirectly by the totality of sex genes in heterosomes and autosomes. Polygenic sex determination with or without heterosomes will be discussed later (Section IV, B). Female heterogamety in the “domesticated” platyfish, Xiphophorus (Platypoecilus) muculatus, has been established by Bellamy ( 1922, 1928), Gordon (1927), Fraser and Gordon (1929), and Kosswig (1933). The dual mechanism of sex determination operating in the same species has been found by Gordon (1946a,b,c, 1947a, 1950, 1951a) who showed that, while the Mexican wild populations are male heterogametic, those originally from British Honduras are female heterogametic. The “domesticated” varieties are considered to have derived from British Honduras. Kallman (1965a,b) extended Gordon’s findings by studying the genetic behavior of populations from a vast area of Guatemala, intermediate between Mexico and British Honduras. The populations in this area contain both male and female heterogametic types even in a single pond. They interbreed with each other with the result that WX females were also collected. In view of evolution of the WY-YY and XX-XY types, this study is most interesting. TiEapia mossambica (Cichlidae) is considered to be a superspecies or group of closely related forms. Crosses of Malayan females to African males result in progeny consisting only of males, or nearly so, whereas the reciprocal crosses produce offspring in a sex ratio of 1o :3d (Hickling, 1960). Whether the Malayan form has the XX-XY and the African form the WZ(Y)-ZZ(YY) system, analogous to the dual system in the platyfish, X. maculatus, has not yet been clarified.

B. Polygenic Sex Determination and So-called Genetic Sex Reversal To grasp the real situation of sex determination, we must adopt the broader view originally advanced by Bridges (1922, 1925) for the fruitfly and elaborated by Winge (1934) for fish viz. that “a given property, the sex included, depends upon all the chromosomes, some of which pull in one direction and others in the other direction, some strongIy and others faintly or not demonstrably at all” (Bridges, 1939). In fishes in which the homogamety/ heterogamety is established, exceptional XXd d or XY o o and WZ(Y) d d appear occasionally, although no exceptional ZZ(YY) o o are reported. These exceptions have been misleadingly called genetic “sex reversals” even by distinguished geneticists. This is based on the concept that only sex chromosomes are carriers of the sex determiners. Only a few geneticists-of whom Winge

3.

SEX DIFFERENTIATION

135

(1934) is the most celebrated-have grasped the intrinsic nature of sex differentiation. He has never referred to such exceptions as sex reversals because his theory is based on multiple sex factors with superior sex genes in the allosomes. The genetic evidence indicates that the guppy, Poecilia ( Lebistes ) reticulata, has sex chromosomes, XX for p and XY for d’ (Winge, 1922b). However, minor or polygenic F and M genes are distributed throughout the autosomes. The Y has superior (epistatic) M gene(s) and the X is supposed to possess epistatic F gene(s) (Winge, 1934; Winge and Ditlevsen, 1947, 1948). In the majority of individuals, autosomal sex genes are more or less in balance. Consequently, sex in most individuals is determined by the heterosomal combination. However, by fortuitous combinations of autosomes or recombinations a few exceptions may appear in which E M > EF in spite of an AAXX constitution. In Mutatis mutandis, exceptional XY females are considered to be individuals in which the totality of sex genes becomes SF > EM. These exceptions cannot be regarded as genetic “sex reversals” since they have a genetic basis to develop either into males or females. Aida’s breeding results ( 1930) in the medaka, Oryzias ( Apbcheilus) Zatipes, can be interpreted by this postulation (Winge, 1930). Since polygenic sex genes are numerous, by selective breedings of XX- d d , the sex ratio of offspring can be varied. Aida (1936) established an XX-XX strain of the medaka and adopted a theory of polygenic sex differentiation which is somewhat different from that of Winge. He suggested that XX males may be the result of a lowering of the femaledetermining potency of the X chromosome. It is unfortunate that he referred to XX males as “sex reversals.” In our d-rR strain of the medaka, where normally X’X‘ are females and XrYR males, about 0.5% of the progeny are exceptions (0.2%crossovers and 0.3%of X‘X‘ d plus XrXR o , Yamamoto, 1959a, 1984a,b). These rare X’X‘d d and X‘YR 9 o are regarded as exceptions in which autosomal M genes and F genes, respectively, over-accumulate in such a way that the sum of autosomal sex genes outbalances the superior heterosomal sex factors. As pointed out earlier (Yamamoto, 1963), the number of possible autosomal combinations is 2‘O, an astronomical number. Even if only some of autosomes are assumed to be sufficiently different in respect to M and F genes (or modifiers), the rest being more or less in even balance, the number of combinations would be enormous. Figure 6 is a graphical diagram illustrating the concept of the polyfactorial sex determination with epistatic sex genes in sex chromosomes, as illustrated in the guppy, the platyfish, and the medaka. For the sake of simplicity, it is taken for granted that the ratio of EM of AA:xF of AA of a popula-

136

TOKI-0 YAMAMOTO

c

.

I F> I M

33

I

I M>IF

$8

I

3

Fig. 6. A diagram of polyfactorial “sex determination” with epistatic sex genes in sex chromosomes.

tion (where A stands for a set of autosomes) has a center (mean or mode) around which the individuals are fluctuating, thus forming a normal distribution curve (broken line). In an AAXY constitution the AA curve is shifted in the male direction (right) because of epistatic M gene( s ) in the Y, and in the AAXX constitution the curve is pulled to the female direction (left), perhaps because of the presence of epistatic F gene( s ) in the X or absence of the Y. Exceptional XX males and XY females encountered in normal breeding may be regarded as individuals which fall in the two extreme regions of the AA curve (stippled and oblique hatched), i.e., the z M greatly overweighs the Z F of AA or vice versa. By inbreeding and systematic selections, the AA curve may be shifted in either direction. In fact, the concept of polygenic sex factors has been substantiated by selective breeding. Thus, Aida (1936) obtained a strain of the medaka, in which both female and male have the XX constitution. In Aida’s stock, exceptional XX males were rare at first; by continued selection for XX males a breed was established in which XX males outnumbered XX females. Formerly, Kosswig ( 1932, 1933, 1935a,b) and Breider (1935a) have asserted that sex of the platyfish is determined genotypically, while that of the swordtail is determined phenotypically. Kosswig ( 1939a,b, 1954) has maintained that the X in d” heterogametic species and the Z in ? heterogametic forms are devoid of sex factors and that ? determiners in the former and d determiners in the latter are present in autosomes. Later, Kosswig and dktay (1955) modified their earlier views on mechanisms of sex determinations in xiphophorin fishes (X. mculatus, X. uariutus, X . xiphidium, and X . helleri). In this article, they maintained

3.

SEX DIFFERENTIATION

137

that in d heterogametic (XX ? -XYd ) fish the X contains the F gene which is weaker than the M gene of the Y, and thai the W has a strong F gene which is stronger than the M gene of the Z in P heterogametic fish. The x of X. helleri has a strong masculinizing potency, Kosswig and his colleagues (Breider, 1942, 1949; Oktay, 1959, Anders and Anders, 1963; Peters, 1964; Dzwillo and Zander, 1967) eventually adopted the concept of polygenic sex determination (cf. Kosswig, 1964). This being so, exceptions are either prospective males or females from the very start of development in spite of their opposite sex chromosome formulas. Oktay (1959) and Anders and Anders (1963) correctly attributed the occurrence of exceptional XX males and XY females, appearing in the Mexican platyfish, to the effects of autosomal sex genes. MacIntre (1961), working on the same animal, reported a not infrequent occurrence of phenotypic females with the XY constitution. When mated to normal XY males, “sex-reversed females gave rise to XX daughters and XY and YY sons. It seems that autosomal F genes must have been accumulated in these strains by several generations of inbreeding. In fishes with entirely polygenic sex determination, sex chromosomes are as yet not differentiated. They are characterized by diverse sex ratios. In crossing them with a species with a homogameticIheterogametic system, sex chromosomes of the latter usually become epistatic. The swordtail, X. helleri, is the most famous gonochorist with a polygenic system. Kosswig (1932, 1935b), Breider and Kosswig (1937), and Gordon (1948) symbolized a pair of chromosomes which are counterparts of sex chromosomes as xx or X’XJ in both sexes. Sengiin (1941) reported that the Wx class of F, (Mh) from the interspecific matings between X . muculutus (WZ) and X. helleri ( x x ) and of backcross ( M h h ) to X. heZZeri are all 0 Q , while Xx class of F, (Mh) of the Mexican X . muculatus (XX) Q x X. helleri d (xx) comprised both sexes. This means that the W is stronger than the X and Xx-d d are the result of X. helleri-polygenic M genes, Oktay (1963) performed a hybridization between the swordtail, X . helleri, and d heterogametic platyfish, X . xiphidium. The F, ( Hx) from X. helleri P ( x x ) x X . xiphidium d (XY) were comprised of both 0 0 (xX) and d d ( x Y ) . In the backcross (Hxh) cross, the F, (Xh) from XX Q x xxd consisted of xx 0 and x X d d.This can be interpreted as indicating that the totality of sex genes in X . helleri 0 is XF > EM, while that of X . helleri d is XM > EF. There are three species of the genus Poecilia (Linzia) ( Poeciliidae) in the Caribbean Islands, P. nigrofasciata, P. caudofasciata, and P. uittata. Of these, P. nigrofasciata possesses the XX-XY mechanism while the

138

TOKI-0 YAMAMOTO

other two species show polygenic sex determination (Breider, 1935b, 1936b; Breider and Kosswig, 1937). The F, of XX-P. nigrofasciatu o x xx-P. cuudofasciutu d are all females (Xx). In the reciprocal cross, xx-P. caudofasciata o x XY-P. nigrofasciata 8,the F, are in a ratio of 19 ( xX):1 g (xY). “he Y of P. nigrofasciata contains a strong M factor. In Section 111, A, Schwier’s studies (1939) on the genus Macropodus ( Anabantidae) are cited. Of three species, M . opercularis, M . chinensis, and M . concolor, M . opercularis and M . chinensis seem to have a homogametic/heterogametic mechanism, because the sex ratio of the latter two species is 19 :l d . In the absence of any sex-linked character the heterogametic sex cannot be decided. In M . concolor, males outnumber females. Zander (1965) interpreted this fact as indicating that the X chromosome might be absent in this form, viz., sex formulas, being xx o -xY d type. Some males are XY and the others are xx males produced by autosomal, polygenic EM > zF genes. Females are considered to be invariably xx constitution with autosomal Z F > EM genes. In short, sex determination in gonochorists is polyfactorial with or without epistatic sex gene( s ) in sex chromosomes, in both cases, a zygote in which xM > xF differentiates into a male and that in which Z F > ZM develops into a female. In the experiments of Bellamy and Queal (1951) using the o heterogametic platyfish, X . muculatus, a type of exceptional male appeared, which sired offspring in a ratio of 3 :Id as if it had the genotype WZ(Y). About one-third of the females bred as though they were WW and produced only daughters. These exceptions were regarded as “sex reversals.” If such were the case, we would expect to find some intersexes at the transitional stage. Confronted with the fact that not a single intersex has ever been found among 50,000 fishes handled in the past 28 years, the authors postulated that early sex inversion might have taken place. However, Wolf (1931) demonstrated that this species is a differentiated one. Hence, no juvenile hermaphroditism can occur. In the platyfish, X . muculatus, with opposing sex-determining mechanisms, one XX-XY the other WY(Z)-YY(ZZ), exceptional XX or WY males and XY females appear occasionally. Besides Bellamy and Queal, Gordon (1946b,c, 1947a, 1951a), MacIntre ( 19sl) and Kallman ( 1965b) regarded these as “sex reversals,” notwithstanding the lastmentioned author’s statement that sexuality of this fish is stable and that although more than 100,000 platyfish have been examined during the last 25 years, not a single female has ever changed into a male (fish with a testis). Consequently he defined a sex-reversed fish as one that is func-

3.

SEX DIFFERENTIATION

139

tionally one sex, but genotypically the other. However, if we take autosoma1 sex genes with various potencies into consideration, it is likely that it is genotypically also one sex and not the other. If we accept these exceptions as real sex reversals, we are confronted with an array of paradoxical facts: (1)Both the medaka and the platyfish are sexually stable among fishes; ( 2 ) not a single sporadic intersex has ever been found, however, so-called sex reversals have been reported occasionally; and ( 3 ) no artificial induction of reversal in sex differentiation by any means, including sex hormones, has ever been accomplished in the platyfish, despite the occurrence of “spontaneous sex reversals.” These puzzling facts can be clarified if we accept that they are genotypic males ( z M > Z F ) and genotypic females ( Z F > Z M ) , respectively. On the basis of the concept of multiple autosomal sex genes or modifiers these exceptions are not sex reversals at all. Breider (1942) described a WZd of 9 heterogametic X . maculatus. In mating this exception with a normal WZ P , the offspring were in a ratio of 3 o (1WW P , 2 WZ o ) : l d (ZZ). Breider did not detect the WW female. Bellamy and Queal (1951) detected the genotype by progeny tests. To sum up, in strictly “differentiated gonochorists such as the platyfish and the medaka the occurrence of spontaneous sex reversals is impossible.

C. “Spontaneous Sex Reversal” in the Swordtail A male and female swordtail are illustrated in Fig. 7. Amateur breeders have repeatedly claimed instances of spontaneous sex reversal in some of their fishes, particularly in aged swordtail X . helleri. Biologists have also reported the same phenomenon ( Harms, 1926a,b; Friess, 1933). Hild (1940) regarded sex reversal in this form as a regular phenomenon. To accept a fish as a true functional sex reversal, it is necessary to ascertain that (1) the fish first functioned as a female and ( 2 ) later it turned out to be a functional male. Some reports on the matter have not ascertained these facts. Popoff (1929) and Sacks (1955) doubted that a true functional sex reversal occurred. Pathological masculinization in two guppy females was recorded by Wurmbach ( 1951). The fish developed a gonopodium. Histological study revealed that they were infected by a fungus, Zchthyophorus hoferi, which caused degeneration of peripheral ovocytes, followed by appearance of spermatogonia derived from ovarian-cavity epithelium. He

140

TOKI-0 YAMAMOTO

Fig. 7. ( A ) Male and ( B ) female of the swordtail, Xiphorphorus helleri. Redrawn by Dr. G. Eguchi after Gordon (1956).

pointed out that Fries’ “rest bodies” in the masculinized gonad of the swordtail are not remnants of degenerated ovocytes but are cysts of the parasite. Peters (1964) found that in X. helleri from Honduras there are two types of males: the normal or rapidly growing males ( d F ) and the slowly developing males ( d s ) which reach larger sizes and differentiate sexually more slowly. With this background, his criticism on reported sex-reversal merits special attention. Among some hundreds of females which he observed during several years, not a single sex reversal has ever been found, although he obtained some arrhenoid advanced-aged females (Fig. 8A ) . The outer appearance of d s looks like females in the immature stage, and male secondary sex characters manifest themselves only later (Fig. 8B). It is natural that slowly developing d s can fertilize ova. Peters considered that alleged sex reversals might not be males but arrhenoid females caused by hormonal upsets in old age, and unripe d s are likely to be taken as females. Philippi (1908) described arrhenoid females with male sex characters in Glarichthyes (Poeciliidae). The most famous cases are in Essenberg’s report (1926) on spon-

3.

SEX DIFFERENTIATION

141

taneous, functional sex reversal in the swordtail. He claimed that two fish (B,, and G ) , after producing broods, turned out to be males and sired a few offspring in mating with “virgin females.” This article has been cited in zoology textbooks without qualification. However, Gordon (1956, 1957), a celebrated poeciliid geneticist, expressed doubt as to the validity of Essenberg’s conclusions. He pointed out that a functional swordtail male must have not only a complete testis but a functional gonopodium (modified anal fin) as well as its suspensorial skeleton. If the fish in question were first female and then turned out to be male, the female-type anal fin and suspensoria must have transformed into a

Fig. 8. ( A ) An arrhenoid 0 (above) and a rapidly growing 8 I’. ( B ) A slowly $S (above) and a rapidIy developing $ F (below) of the swordtail, Xiphophorw helleri. Drawn by Dr. G . Eguchi from photos by Peters (1964). growing

142

TOKI-0 YAMAMOTO

perfect gonopodium. For him, it is inconceivable that such a drastic change should have occurred. Incomplete male secondary sexual characters engendered by hormonal upsets appear in aged swordtails. He questioned the virginity of the females used as mates of alleged sex reversals. It is well known that sperm from a single copulation can sire up to eight broods in the subsequent 8 months. Gordon (1956) states that “unconditional evidence is required before the reality of spontaneous, complete and functional sex reversal in a fish can be accepted.” V. CONTROL OF SEX DIFFERENTIATION Science unfolds and controls nature. Toki-o Yaniamoto

A. Surgical Operation Surgical castrations of female Siamese fighting fish, Betta splendens, have in rare instances produced fish in which the regenerated gonad was a functional testis. These experiments were first performed independently in the United States (Noble and Kumpf, 1937) and later in Germany (Kaiser and Schmidt, 1951). The American authors obtained only seven reversals out of 150 spayings, of which only three proved to be fertile, siring both sexes in mating with females. The German authors obtained three reversals, of which only one was fertile and fathered solely a total of 109 daughters. The latter experiment proved that the male is heterogametic.

B. Modification of Sex Differentiation by Sex Hormones In the rainbow trout, Salmo gairdneri irideus, Padoa (1937, 1939a) reported that injection of a follicular hormone produced ova in the testes, while testosterone induced testicular tissue in the ovaries. Ashby (1952, 1956, 1959) reported the paradoxical similarity of action of estradiol and testosterone in the brown trout, Salmo trutta; both hormones retarded gonadal development and produced hypertrophy of the somatopleure. There was no evidence of sexual inversion. The result might be ascribed to the fact that he started hormone administration in alevins in which gonadal sex differentiation has already been established. A number of attempts have been made in viviparous toothcarps to modify sex differentiation by administration of sex hormones, especially androgens, starting with newly born broods or adults. However, andro-

3.

SEX DIFFERENTIATION

143

gens mainly affect the secondary sexual characters and the action on the gonad has been either incomplete, pathological, or negative. In the guppy, Poecilia reticulatn, the effects of sex hormones have been studied by Witschi and Crown (1937), Berkowitz (1937, 1938, 1941), Rkgnier ( 1938), Eversole (1939, 1941), Gallien (1946, 1948), Hildeman (1954), Miyamori (1961), and Querner (1956). Berkowitz caused the production of ovo-testis by administering estrogens to young male guppies. He reported that the optimal dosage resulted in an almost complete reversal of testis to ovary. The same effect was observed by Querner. Miyamori produced ovo-testis by administration of androgen to young females. In the top minnow, Gambusia holbrookii, which shows “transitional intersexuality,” Lepori ( 1942a,b, 1948) produced testis-ova in females by administration of an androgen and in males by an estrogen. Comparable results were obtained in G. afinis by Okada (1944). In the swordtail, X . helleri, Baldwin and Goldin (1940) reported that androgen administration to young females not only simulated the male secondary characters such as formation of a caudal sword and the gonopod but also induced masculinization of the gonad after degeneration of ovocytes. Querner (1956) showed that androgens produce ovo-testis in genetic females. Contrary to the cases in amphibians, corticoids have no effect on sex differentiation. Vivian ( 1952a) reported that 11 incompletely hypophysectomized female swordtails developed an involuted ovary with so-called “rest bodies” and claimed that two or three of them showed partial masculinization. The fact that these rest bodies are not degenerated ovocytes but are cysts of a parasitic mycomycete has already been mentioned (Section 111, C ) . This pathological change might be caused by infection of Ichthyophorus hoferi. Failure to obtain complete reversal of sex differentiation in these studies may be because administration of hormones was started after the onset of gonadal sex differentiation. As stated before, although the guppy and swordtail are “undifferentiated” species, gonadal sex differentiation has already been established when broods are born (Essenberg, 1923, in the swordtail; Goodrich et al., 1934; Dildine, 1936, in the guppy). Dzwillo’s success (1962) in obtaining complete reversal in sex differentiation in the guppy will be discussed later. In the oviparous toothcarp, Oryzias htipes, Okada (194313, 1949, 1952a) claimed that the formation of testis-ova in adult males can be induced by administration of either estrogens or androgens. Treated males had gonads containing large oviform cells which look like ovocytes but which are not surrounded by follicle cells. The nature of these largesized cells will be discussed later (Section VI, A ) . Okada (1964b) obtained a true testis-ova by estrogen administration starting in juveniles.

144

TOKI-0 YAMAMOTO

In this case, ovocytes are genuine since they are accompanied by follicle cells.

C. Complete (Functional) Reversal of Sex Differentiation In the swordtail (undifferentiated species), Dantschakoff ( 1941) claimed to have induced males by the administration of androgens to females. There have been some reservations regarding these observations (cf. Gordon, 193-8, 1957, and Peters, 1964, in Section IV, D). If, however, this is true, it is possible that the induction was accomplished because this “undifferentiated species is very labile in sexuality. As stated before (Section III, A ) this species has no sex chromosomes. By treatment with an androgen, female germ cells including ovocytes might degenerate and protogonia might differentiate ,into male germ cells. In the differentiated species such as the platyfish and the medaka, where neither spontaneous sex reversal nor even sporadic intersexes occurs, induction of complete reversal may be impossible if hormone treatment starts after the onset of gonadal sex differentiation. The rank of stability in sexuality seems to be platyfish 5 medaka > guppy > swordtail. No sex reversals have ever been successful by administration of sex hormones in the platies (Cohen, 1946; Tavolga, 1949; Laskowski, 1953) because heterologous sex hormones have inevitably been administered after birth when the gonadal sex differentiation has already been established. The medaka, Oryzias latipes (Fig. 9 ) , an oviparous toothcarp, is a strict gonochorist. Genetically, the sex-determining mechanism norma1 to this fish was established as XX for female and XY for male (Aida, 1921). Embryologically, it is a differentiated species ( Yamamoto, 1953). Using our genetically analyzed strain (d-rR) of the medaka, in which white is female X‘X‘ and orange-red is male X7YR,where R stands for xanthic pigmentation, functional reversal in sex differentiation has been accomplished in both directions by heterologous sex hormones (Yamamoto, 1953, 1958). Thus, it has become possible to control sex differentiation ad libitum either from genetic males (XY) to functional females or from genetic females (XX) to males. In other words, we are able to inverse the prospective fate of the fish, genetically destined to develop into one sex and direct it to differentiate into the other. Success in reversing sex differentiation in both directions in the medaka rendered it possible to mate estrone-induced XY females with niethyltestosteroneinduced XX males (Yamamoto, 1961, Fig. 10). In contrast to viviparous poeciliids, the newly hatched fry ( g 4 . 8 mm) has an indifferent gonad (gonad primordium) and gonadal sex differentiation takes place between the 6 and 11 mm stages. To accomplish

3. SEX

145

DIFFERENTIATION

Fig. 9. ( A ) Female and ( B ) male of the medaka, Oryzias latipes.

complete reversal in sex differentiation in any differentiated gonochorists, certain conditions should be fulfilled. First, a heterologous sex hormone should be administered starting with the stage of indifferent gonads and passing through the stage of sex differentiation. In our experiments, heterologous hormones were administered by the oral route to newly hatched fry and continued until they reached 12 mm or longer. Therer?

7-?

Fig. 10. A diagram illustrating the progeny of estrone-induced X'YR female mated with methyltestosterone-induced X'X' male (from Fig. 4, Yamamoto, 1961).

146

TOKI-0 YAMAMOTO

Table 111 GDaoDosage Levels of Estrogens Acting as Gynotermone (Gynogenin) to Induce 50% X Y Females in the Medaka, Oryzias latipes Estrogens

GD5o ( r d g diet)

Hexesterol Euves tinn

0.4 0.8

Ethynyl estradiol

1.7

Estradiol-17b Stilbestrol Estrone Estriol 11

5.8 7.5 20 130

Authors Yamamoto and Iwamatsu (unpublished data) Yamamoto, Hishida, and Uwa (unpublished data) Yamamoto, Noma, and Tsuzuki (unpublished data) Yamamoto and Matsuda (1963) Yamamoto and Matsuda (1963) Yamamoto (1959b) Yamamoto (1965)

p,p’-Dicarboethoxyoxy-trans-a#-diethylstilbene.

after, they were raised on a normal diet until they reached maturity. Second, adequate dosage levels of hormones should be used; these differ in potency with different sex hormones (Tables I11 and IV). Estrogen-induced XTYRfemales (egg producers) were first obtained by Table IV AD50 Dosage Levels of Androgens Acting as Androtermone (Androgenin) to Induce 50% X X Males in the Medaka, Oryzias Zatipes Androgens

19-Nor-ethynyltestosterone

1.0

Fluoxymesterone (Halotestin)

1.2

17-Ethynyltestosterone (Pregneninolone, ethisterone) Methylandrostenediol

3.4

Methyltestosterone Androstenedione Testosterone propionate Androsterone Dehydroepiandrosterone

7.8 15 500 560 580 >3200

Yamamoto, Hishida, and Takeuchi (unpublished data) Yamamoto and Hishida (unpublished data) Yamamoto and Uwa (unpublished data) Yamamoto and Onitake (unpublished data) Yamamoto (1958) Yamamoto (1968) Yamamoto et al. (1968) Yamamoto et al. (1968) Yamamoto and Oikawa (unpublished data)

3.

147

SEX DIFFERENTIATION

estrone and stilbestrol ( Yamamoto, 1953, 1959a). The induced X'YR females in matings with normal X7YR males produced offspring in a ratio of 1o : 2 g instead of 1o : 3 g ,indicating that viable YRYR males are rare. However, two YRYR males which sired all-male progeny were actually detected among 57 ( X'YR YRYR)males singly tested by X'X' females (Yamamoto, 1955, 1959a). By using other mating systems (cf. Table V ) , it happened that YRYr males produced by mating estrogeninduced X'YR females with X'Y' males were all viable (Fig. 11). The Y'Y' males yielded by mating induced XRF or X'Y' females with X'Y' males also all survived. By administration of estrone in two consecutive generations, it was possible to invert sex differentiation in YY zygotes (Yamamoto, 1963, 1967). Although it does not relate directly to the main theme of this section, reference should be made to the intriguing problem of viability of YY zygotes. This problem has been fully discussed elsewhere (Yamamoto, 1964a,b). In short, the regular YR chromosome contains an inert section ( - ) which in duplex condition ( YR,-YR*-) renders

+

Table V Production of YrY' (White) Males and Estroneinduced V V Females in the Medaka, Orvrias Zutiveso*b r(X'Xr) 9 0 (12) x R(XRY') $ 3 (12) Normal diet 10 R ( X R F ) 9 9 (44) r(XrY.) $ 3 (40) PI T(X'X') 9 9 (12) x R(XRY') $ 3 (12) Estrone diet I I, ER(XRXr) 9 0 (38) Er(X'Y.) 9 9 (24) Er(X'Y7) 3 8 ( 2 ) ~ Po Er(X'Yr)d 9 9 (24) X R(XRYr) $ 3 (12) Normal diet R(XRX') 9 9 (11) R(XRYr) 3 3 (6) r(XrYr YrYr) 30' (17) 11, .....__...._....__ Detected 5 + 4 Po E r ( P Y r I d 9 9 (24) X R(XRY.) 8 3 (12) Estrone diet I ER(XRXr XRYr) 9 9 (28) Er(X'Yv YrYr) 9 9 (13) IL ..-..-............. ...........-..... Detected 3 + 5 3 + 3 P1

I

I

+

+

+

From Yamamoto (1967). Subscripts c and e denote control and experimental, and I and I1 are the first and the second generations. Numbers enclosed in parentheses represent numbers of parents and offspring obtained. Animals designated by broken lines were submitted to testcrosses. Free figures are numbers of fish detected by testcrosses. Sex chromosome formulas following r or R are presumed sex genotypes. E denotes estronized fish; r, white, and R, orange-red. c Nonreversals. d In the original paper (1967) this formula was misprinted as XrXr. a

b

148

TOKI-0 YAMAMOTO

r 99

R db ,:;:

:~r~yj;?$:;"':~ r = White

Estrone diet

r ??

R ?9

R= Orange-red r dd

Sex reversal

Sex reversal

Sex reversal

Fig. 11. ( A ) A diagram of production of Y'Y" males feeding to offspring from estrone-induced X'Y' females males, ( B ) Induction of YRY' females by estrone feeding induced X'Y' females mated to normal X'Y" males (from 1963a )

.

Sex reversal

(control of B ) by normal mated with normal X'Y' to offspring from estroneFigs. 1 and 2, Yamamoto,

the zygote nonviable, while both regular X' and XR possess the viability section (+). Rare survivaI YRYR zygotes are experimentalIy verified to have the YRi-YcR,+constitution, the ( + )of which is known to have been derived from X'*+by crossing over. This being so, it is quite rational that the Y' is Y',+ in constitution and YRYr is YR,-Yrv+.It is hardly surprising that practically all Y'Y' zygotes (Y",+Y'>+)are viable (Fig. 12). Using white females (X'x') and orange-red (XRYr) males as parents in the first generation, estrone-induced YRY and YrYr females were produced by administration of estrone in two consecutive generations (Yamamoto, 1963, Fig. 11; 1967, TabIe VI).

3.

149

SEX DIFFERENTIATION

AA XX [ZF >ZM]--

Gyminductor

-

Indifferent gonad +Ovary

(0

-

Andr6AA X Y b M >ZF]-inductor (Androterrnone)

Indifferent '\ gonad --Testcs(d)

Fig. 12. A scheme of normal sex differentiation (solid lines) and reversal in sex differentiation b y means of heterologous sex inductors (from Yamamoto, 1962).

Returning to the main theme, adequate dosage levels are very different for various estrogens (Fig. 13). The symbol GD,, was suggested as the designation of the dosage level at which there are 50%induced XY females (Yamamoto and Matsuda, 1963). The writer and his co-workers have performed experiments to estimate GD,, values of both natural and synthetic estrogens (Table IV) . Reversal in sex differentiation in the opposite direction, viz., induction of X'X' males by methyltestosterone, has also been accomplished (Yamamoto, 1958); these males in mating with normal X'X' females sired allfemale progenies. Androsterone and testosterone propionate have also a male-inducing action (Yamamoto et al., 1968). The symbol AD,, may be used as a notation of dosage level at which there are 50% induced XX males. In Table X, AD,, values of various androgens are listed. In the guppy, an undifferentiated species, all attempts to control sex by hormone administrations after birth when the gonadal sex differentiation is already established (cf. Goodrich et al., 1934) failed until Dzwillo (1962) succeeded in getting functional XX males by androgen administration to gravid females containing embryos having indifferent gonads. Clemens (1965) was able to obtain androgen-induced XX males in Tilapia mossambica by administration starting with newly hatched fry having indifferent gonads. In the goldfish, Carassius auratus, Yamamoto and Kajishima (1969) obtained functional reversals in both directions by administering heterologous sex hormones for 2 months to newly hatched fry with indifferent gonads. By testcrosses of the hormone-treated fish, Table V1 Terminology of Sex Inducers Female inducer

Male inducer

Author

Cortexin (corticin) Gynogenin Gynotermone

Medullarin Androgenin Androterrnone

D' Ancona

Witschi Hartmann

Beginning 1931 1949b 1951

150

TOKI-0 YAMAMOTO

1

2

s

10

50

20

DOSAGE

100

1,EVE;L

200

wo

IN 9 g / g

loon

2000

DIET

Fig. 13. Potencies of estradiol, estrone, and estriol in inducing XY females (from Fig. 2, Yamamoto, 1965b).

evidence for male heterogamety ( X U ) in the goldfish was first obtained. In differentiated species, neither spontaneous intersex nor sporadic sex reversals can appear. Nevertheless, induction of complete reversal in sex differentiation can be achieved if the following necessary condition is fulfilled. Sexogens (estrogens and androgens) at suitable dosage levels should be administered consistently, beginning with the stage of the indifferent gonad and passing through the stage of sex differentiation. In summary, by finding the correct dose of a suitable hormone for treating embryos or larvae and by starting medication at the indifferent gonadal stage, it is possible to convert an otherwise potential female into one that functions as a male or vice versa. This principle will hold true not only for “undifferentiated” species but also for “differentiated” ones. VI. NATURE OF NATURAL SEX INDUCERS

A. Steroid versus Nonsteroid Theories

Witschi (1914, 1929) postulated the inducer theory of sex differentiation in amphibians. Although the sex genes are effective sex determiners, their action appears to be mediated by the sex-gene-controlled sex inducers as far as the vertebrate is concerned. Terminology of sex inducers differs among authors (Table VI) . The exact chain of events which leads from sex genes to sex differ-

3.

SEX DIFFERENTIATION

151

entiation has not yet been clarified but is outlined in Fig. 12 (Yamamoto, 1962). The important point is that sex genes are not direct sex determiners, but it is the sex-gene-controlled inducers that determine sex differentiation. There are pros and cons for the steroid theory of sex inducers. In amphibians, Witschi (1942, 1950, 1955, 1957) and his followers are “cons” and postulated a protein nature of the sex inducers without evidence. Chieffi (1965) also stated that “the embryonic sex inducers, . . . , are in all probability different from sex hormones of the adults.” The postulation is based inter alia on (1) the so-called paradoxical feminization of the gonad by high doses of androgens, ( 2 ) the nonspecificity of sex hormones as sex inducers, viz., non-sex-hormonic corticoids are believed to also have the potency of sex inducers, and ( 3 ) the ineffectiveness of androgens on some WZ amphibians. These statements are based on the sex ratio scored by a limited number of gonads of young, the sex genotype of which is unknown. Neither a single case of androgen-induced functional nor a corticoid-induced sex reversal has ever been produced. Furthermore, criteria for ovocytes are questionable. Experiments on the adult male medaka show that not only estrogens but androgens, corticoids, and any noxious treatments such as heat treatment and starvation result in the formation of large-sized cells in the testis (testis-ova), without characteristics of ovocytes such as the presence of follicle cells (cf. Egami, 1955a-e, 1956a,b). These oviform cells, which actually look like auxocytes, may be regarded as degenerating protogonia or spermatogonia enlarged prior to deterioration (Yamamoto, 1958). As to point ( 3 ) above we should not rely on negative results. Reported experiments have been made by immersion in hormone. It may well be that androgen immersion cannot simulate the natural condition surrounding the protogonia and induce them to differentiate into male germ cells. Inducers are thought to be densely produced only by cells surrounding the protogonia. In the medaka the following occurs: (1) Estrogens act as female inducers and androgens function as male inducers; no paradoxical phenomenon are engendered. ( 2 ) Sex hormones act specifically as sex inducers; non-sex-hormonic corticoids have no effect. ( 3) Effective dosage levels of estrone and methyltestosterone are so low that it may be possible for juvenile fish to elaborate a small amount of sexogens ( Yamamoto, 1959b, Hishida, 1965). (4)By administration of radioactive te~tosterone-l6-~T propionate and diethylstilbestrol-( m~noethyl-l-’~C), Hishida (1962, 1W) showed that these are selectively incorporated into the juvenile gonad. Point ( 4 ) seems to indicate that the sex-differentiating gonad requires more sex hormones than other organs. On the basis of

152

TOKI-0 YAMAMOTO

these results, we are inclined to the sex steroid theory of natural inducers. The current view of pathways of biosynthesis of sexogens in adults is illustrated in Fig. 14 based on Meyers (1955a,b), Solomon et al. (1956), and others. Whether or not the pathways are valid for cells surrounding the indifferent gonad is not known. Of these naturally occurring steroids, pregnenolone, progesterone, and 17a-hydroxyprogesterone are found to be ineffective (Yamamoto and Matsuda, 1963; Yamamoto, 1968) while androstenedione and other sex steroids are effective ( Tables I11 and IV).

-

&o

/

no Cholesterol

Pregnenolone

Progesterone

Androstenedione

110-Hydroxyprogesterone

19- Hydroxy testosterone

Androsterone

19-Hydroxyandrostenedione

Estriol

Eetradiol-170

Eetrone

Testosterone

-

HO

Fig. 14. A diagram of the current view of biosynthesis of key sex steroids (based on Meyers, 1955a; Solomon et aZ., 1956; and others).

3. SEX

DIFFERENTIATION

153

B. Detection of Steroids and Relevant Enzymes in Fish Gonads First of all, it is not known whether all the sex hormones of adult fishes are identical with those of mammals. The identity of some hormones has been established by Chester Jones and Phillips (1960), Chieffi and Lupo ( 1962, 1963), Lupo and Chieffi (1963), and Wotiz et al. (1960). Research on the detection of steroids in the larval gonad at the onset of sex differentiation by classic chemical methods is difficult because of the paucity of available material. Therefore, almost all approaches depend on histochemical and enzymic methods. In the elasmobranch Scylwrhinus canicuhs Chieffi (1955) found a few sudanophilic granules in the gonadal medulla before sex differentiation. After the onset of sex differentiation, there appear many sudanophilic as well as Schultzpositive granules. The identification of enzymes involved in biosynthesis of sex steroids is another approach. Techniques have been developed for the histochemical demonstration of three relevant enzymes: A5-3P-dehydroxysteroid dehydrogenase ( A6-3p-HSDH), A5-3P-ketoisomerase ( A53P-KIM ) and 17P-hydroxysteroid dehydrogenase ( 17P-HSDH) which is DPN and TPN dependent. Together with A5-3p-KIM,A5-3p-HSDHis involved in the pathway from pregnenolone to progesterone, one of the first steps of biosynthesis of sex steroids (cf. Fig. 14). Chieffi and coworkers (1961-1963) were unable to detect A6-3p-HSDH in the embryonic gonads of the elasmobranchs, Scyliorhinus caniculas, S. stellaris, and Torpedo murmorata. The enzyme, however, was detected in the interrenal tissue of S . stellaris and T. mumorada. Bara (1966) showed the presence of 3P-HSDH in the testis of Fundubs heteroclitus. Many questions remain unanswered. It is not known, for instance, where the supposed sex inducers are produced and in what way they influence the indifferent gonia ( protogonia ) to develop into either ovocytes or spermatocytes. Finally, the chemical nature of natural sex inducers still remains obscure. Research along these lines should be fruitful.

VII. DIFFERENTIATION OF SECONDARY SEXUAL CHARACTERS

Our primary concern is with gonadal or primary sex differentiation. References to the manifestation of secondary sexual characters are so numerous that only a brief account is presented here. Within the last two decades some excellent reviews have been pub-

154

TOKI-0 YAMAMOTO

lished (Hoar et al., 1951; Hoar, 1955, 1957a,b; Dodd, 1955). Pickford and Atz (1957) presented a monograph on the piscine pituitary gland and its role in reproduction. It is convenient to divide secondary sex characters into two categories: (1) temporary characteristics which normally appear only during the breeding season such as nuptial coloration, pearl organs, and the ovipositor of bitterlings; ( 2 ) permanent organs developed fully at the onset of sexual maturity such as the gonopodium of viviparous cyprinodonts and the papillar processes on the male anal fin and the urinogenital papilla in the female medaka, 0yzias. It is also convenient to define male- and female-positive secondary sex characters (Yamamoto and Suzuki, 1955). The former is a character that is either specific to the male or is more developed in the male than in the female. The latter is the reverse. The mechanism by which these characters manifest themselves may only be elucidated by experiments. Most of secondary sex characters of fishes are male-positive. Kop3 (1918, 1928) was the first to demonstrate in the minnow Phorinus laevis that the nupital coloration is dependent on the testicular hormone. This has been confirmed in the sticklebacks, Gasterosteus pungitius and G. muleatus, by van Oordt (1923, 1924), van Oordt and van der Mass ( 1927), Bock ( 1928), Craig-Bennett ( 1931), and Ikeda ( 1933); in the bitterlings by Tozawa ( 1929), Wunder ( 1931), and Glaser and Haempel (1932); and in the guppy by Blacher (1926b) and Samokhvalova (1933). Similar results were obtained in the medaka (Niwa, 1965a)b). Tozawa (1923) proved that the pearl organs of the goldfish, Carassius auratus, are controlled by testicular hormone. While castration results in the disappearance of these male-positive characters, ovariotomy causes no effect. This means that the absence of these in the female is not because of the inhibitory action of the ovary. On the contrary, in the ganoid, Amia calva, Zahl and Davis (1932) showed that the gray-black pattern (male-positive) is absent in the female because of an inhibitory actiton of the ovary; ovariotomy produced this pattern. The male swordtail and the platyfish have complicated suspensorial skeletons including three hemal spines in the caudal vertebrae; these are necessary for proper functioning of the gonopodium. In the female, the three hemal spines are absent providing an undivided space required by the gravid female for the brood of embryos. Vivian (1952b) reported that spayed swordtails develop the three hemal spines as in the male, indicating that these are absent in the female by an inhibiting action of the ovary. Rubin and Gordon (1953) showed in the platyfish that .westradio1 benzoate induced the dissolution of one or two of these hemal spines in the male and methyltestosterone induced

3.

SEX DIJ?FJiXE%TIATION

155

fusion of basal bony elements (mesonost and baseost) in the female, which are normally separated. The secondary sex characters of the medaka, Oryzias latipes, have been described by Oka (1931a). Nagata ( 1934, 1936) demonstrated that castration results in a regression of both the shape and papillar processes of the male anal fin. Transplantation of a testis into the sprayed fish produces an almost perfect replica of the male characters. Administration of an androgen to females induces the male-type anal fin (Okada and Yamashita, 1944a). Other sexual differences such as teeth, bones, and body shape were reported by Egami (1956b, 1959b) and Egami and Ishii ( 1956). In the topminnow, Gambusia holbrookii, Dulzetto (1933) noted the correlation between testicular development and the gonopod formation. A host of experiments on gonopod formation has been performed by Turner (1942a,b, 1947, 1960) and Okada and Yamashita (1944b) in the topminnow G . ufinis, by Grobstein (1940a,b, 1942a,b, 1948) in the platyfish X . muculatus, by Hopper (1949) and Hildemann (1954) in the guppy, and by Taylor (1948) in the guppy-swordtail hybrid. Some of the earlier studies of hormone-induced characters, aimed at inducing modifications of sex differentiation by administration of steroids in poeciliids, have already been discussed (Section V, B ) . Vivian ( 1950) reported that X-ray irradiation in the females induced formation of a malelike anal fin ( incomplete gonopod). At this point, it may be remarked that not all fins are sensitive to androgens. For example, androgens have a stimulating action on the anal and dorsal fins of the medaka but exert no effect on the caudal fin. We will refer to the pelvic fin later. In the gobiid fish, Pterogobius zonoleucus, Egami ( 1 9 5 9 ~ )stated that while the dorsal fin is elongated by administration of an androgen, other fins are not affected. Another malepositive character is the large tooth in the male medaka. Large teeth can be produced in the female by administration of methyltestosterone (Egami, 1957; Takeuchi, 1967). Turner’s extensive researches ( 1960) demonstrated that regional parts in the anal fin of Gambusia have a differential sensitivity to androgens. The 3-4-5 ray complex is the most sensitive. He among others suggested that the enzyme pattern may be different in each area of the fin. Only a few female-positive secondary sex characters are present in teleosts. The urinogenital papilla (UGP) and the pelvic fin of the medaka and the ovipositor (elongated UGP) of the bitterlings are examples. Although the UGP is stimulated by both estrogen and androgen, it is far more sensitive to the former than the latter (Yamamoto

156

TOKI-0 YAMAMOTO

and Suzuki, 1955). It appears that normally the UGP of the female is produced by an estrogen from the ovary. The UGP of the Indian catfish, Heteropneustes fossils, is also femalepositive. Kar and Ghosh (1950) reported that injections of either an estrogen or an androgen in females resulted in drastic hypoplasia of the ovaries, oviduct, and the UGP. They attributed this seemingly curious effect of the estrogen to the paucity of gonadotropin through suppression of its action by a large amount of exogenous hormone and concluded that the UGP is under the control of the ovarian hormone. The pelvic fin of the female medaka is longer than that of the male (Oka, 1931a). Niwa (196513) showed that this is because of an inhibitory action of the testicular hormone and cannot be attributed to a stimulating action of the ovarian hormone. In the Japanese bitterling Acheilognathus indermedia Tozawa ( 1929), showed that ovariotomy inhibits ovipositor lengthening. Research on the ovipositor formation in the European species, Rhodeus umurus, has provided a very confusing and puzzling problem. Fleishmann and Kann (1932) and Ehrhardt and Kuhn (1933) found that estrogen induces this reaction in the nonbreeding season. In 1935, American and German authors proclaimed that this response can be used for the diagnosis of pregnancy. It has been reported that urine from both men and pregnant women usually gives a positive reaction, while that from nonpregnant women may or may not cause this reaction. A male hormone in the urine was believed to be responsible. Banes et al. (1936) reported, however, that crude extracts of adrenal cortex gave a positive response while androsterone was negative. On the other hand, Kleiner et al. (1937) obtained a positive response by administration of androgens. De Wit (1939) confirmed this and further stated that progesterone and deoxycorticosterone (DOCA) are far more active than an androgen. Bretschneider and de Wit (1940-1941, 1947) and de Wit (1940, 1941, 1955) developed a quantitative method and studied this phenomenon. However, a survey of literature reveals that not only the abovementioned substances but also adrenaline and anesthetics have been reported to be active in lengthening the ovipositer. De Groot and de Wit (1949) reported that alcohol, heat shock, and strong light also causes ovipositor growth. This would seem to indicate that stresses in the sense of Selye also cause the reaction. Furthemore, the presence of mussels and/or males and an exhalent current from mussels have the influence on the cyclic ovipositor growth (de Wit, 1955). It appears that normally environmental factors exert their effects on the pituitary gland through intermediacy of the nervous activity evoked by sense organs. Further work must be done before these questions can be adequately answered.

3.

SEX DIFWRENTIATION

157

VIII. SUMMARY

The need to clarify our thinking about sexuality in fishes is emphasized. First, it is unsound to consider that sexuality in all members of the class Pisces is labile, There are in fact many sex types: synchronous, protandrous, and protogynous hermaphrodites, as well as gonochorists. Among gonochorists, there are “undifferentiated and “differentiated species in the sense of Witschi. The latter are more stable than the former in sexuality. Evidently, there are graded levels in sexuality among fishes. Second, sex is a phenotypic expression. In gonochorists, it is important to clarify the concept of sex to eliminate confusion. A male is a sperm producer and a female an egg producer. This definition, although selfevident, is particularly important in avoiding controversy in the problem of sex reversal. By definition, sex genotypes of zygotes before sex differentiation such as AAXX and AAXY are usually presumptive or prospective females and males, respectively. Although combinations of sex genes are determined at the time of fertilization, sex is actually determined by sex-gene-controlled sex inducers at a certain critical period of development, that is, sex genes act only indirectly upon sex differentiation. Their action is directly mediated by sex inducers. Sex genes are present not only in the sex chromosomes but also distributed over a great number of autosomes. In view of multiple (polygenic) sex factors, exceptional AAXX males or AAXY females and WZ( Y ) males are not “genetic sex reversals.” These exceptions appear when the totality of sex genes become ZM > XF by fortuitous combinations of autosomes in spite of AAXX constitution and zF >XM despite AAXY constitution. This concept may also be applied to exceptional WZ(Y) males in female heterogametic species. They have the genetic basis for either male or female. A diagram illustrates this concept (Fig. 6). In general, sex differentiation is induced by sex inducers controlled by sex genes with or without (in species without sex chromosomes) sex chromosomes. Sex chromosomes contain epistatic (superior) sex genes. Reports on spontaneous sex reversals in gonochorists have been criticized. Such cases seldom, if ever occur in fishes. Most cases are masculinized females (with ovary) brought about by hormonal upsets that accompany advanced age. Artificial control of sex differentiation may be one of the key projects in biology. One approach is to override the AAXX (in which XF > XM) or AAXY (in which XM > zF ) constitution by artificial means. This

158

TOW-0 YAMAMOTO

has been successfully accomplished in both directions in the medaka and the goldfish by administration of heterologous sex hormones, i.e., estrogens to XY zygotes and androgens to XX zygotes. In the medaka, it is possible to invert sex differentiation of even YY zygotes into females by administration of estrone in two consecutive generations. Reversal in one direction, from the presumptive female to the functional male, was achieved in the guppy ( Dzwillo, 1962) and Tilapiu (Clemens, 1965). It is of paramount importance that heterologous sex steroid administration must start at the stage of the indifferent gonad and continue through the stage of gonadal sex differentiation and that a suitable dosage should be used. GD,, of estrogens and AD,, of androgens which indicate relative potencies in inducing XY females and XX females in the medaka, respectively, are listed. The chemical nature of the endogenous sex inducers are discussed. There are pros and cons of the steroid theory of the natural sex inducers. Because of ( 1 ) the specificity of sex steroids as exogenous sex inducers, ( 2 ) the very low effective dosage of natural sex steroids, and ( 3 ) the selective incorporation of sex steroids into the differentiating gonad, it is likely that endogenous sex inducers are allied to or identical with sex steroids. A brief review on the manifestation of secondary sexual characters was given. REFERENCES Abe, T. ( 1967). Personal communication. Aida, T. (1921). On the inheritance of color in a fresh-water fish Aplocheilus Zatipes Temminck and Schlegel, with special reference to the sex-linked inheritance. Genetics 6, 554-573. Aida, T. (1930). Further genetical studies of Aplocheilus Zutipes. Genetics 15, 1-16. Aida, T. ( 1936). Sex reversal in Aplocheilus Zatipes and a new explanation of sex differentiation. Genetics 21, 138.153. Anders, A., and Anders, F. ( 1963). Genetisch bedingte XX- und XY-0 und XYund YY- 8 beim wilden Platvpoecilus maculutw aus Mexico. Z. Verefbungslehre 94, 1-18. Aoyama, T. (1955). O n the hermaphroditism in the yellow sea bream, Taius tumifrom. Japan. I . Ichthyol. 4, 119-129 (in Japanese with English &sun&). Aoyama, T.,and Kitajima, T. (1966). Sex reversal in the flat-head fish, Suggrundus meerderuoort ( Bleeker ) . Oral communication. Aoyama, T., Kitajima, T., and Mizue, K. (1963). Study of the sex reversal of Inegochi, Cociella crocodila (Tisesius). Bull. Seikai Regional Fisheries Res. Lab. 29, 11-33. Ashby, K. R. (1952). Sviluppo del sistema riproduttivo di Salmo truttu L. in condizioni normali e sotto l'influenza di ormoni steroidi. Rio. Biol. (Perugia) 44,3-19. Ashby, K. R. (1956). The effect of steroid hormones on the brown trout (Salmo trutta L.) during the period of gonadal differentiation. J. Embryol. Erptl. Morphol. 5, 225449.

3.

SEX DIFFERENTIATION

159

Ashby, K. R. (1959). Leffetto di basse concentrazioni di estradiolo e di testosterone sul differenziamento gonadico di Salmo trutta L. (Nota preliminare), R i a Biol. (Perugia) 51, N. S . No. 11, 453-468. Ashby, K. R. (1965). The effect of steroid hormones on the development of the reproductive system of Salmo trutta L. when administered at the commencement of spermatogenetic activity in the testis. Rev. Biol. (Perugia) 80, N.S. No. 18, 139-169. Atz, J. W. (1959). The relation of the pituitary to reproduction in fishes. In “The Physiology of the Pituitary Gland of Fishes” (G. E. Pickford and J. W. Atz, eds. ), Part VII, pp. 178-270. N.Y. Zool. SOC.,New York. Atz, J. W. (1964). Intersexuality in fishes. In “Intersexuality in Vertebrates including Man” (C. N. Armstrong and A. J. Marshall, eds.), pp. 145-232. Academic Press, New York. Bacci, G. ( 1966). “Sex Determination.” Pergamon Press, Oxford. Bacci, G., and Razzauti, A. (1957). Fdso gonocorismo in Coris julis (L.). Atti Accod. Nazl. Lincei, Rend., Classe Sci. Fis., Mat. Nat. [8] 23, 181-189. Bacci, G., and Razzauti, A. (1958). Protogynous hermaphroditism in Coris iulis L. Nature 181, 43-33. Baldwin, F. M., and Goldin, H. S. (1940). Effects of testosterone propionate of the female viviparous teleost, Xiphophorus helleri. Proc. SOC. Exptl. Biol. Med. 42, 813-819. Bara, G. (1966). 3P-Hydroxysteroid dehydrogenase activity in the testis of Fundulus heteroclitus. Anat. Record 152, 449. Barnes, B. O., Kanter, A. E., and maw, A. H. (1936). Bitterling ovipositor lengthening produced by adrenal extracts. Science 84, 310. Bellamy, A. W. ( 1922). Sex-linked inheritance in the teleost PlutypoeciZus mucuhtus. Anat. Record 24, 41-20. Bellamy, A. W. (1928). Bionomic studies on certain teleosts (Poeciliinae). 11. &lor pattern inheritance and sex in Phtypoecilus muculutus Gunth. Genetics 13, 226-232. Bellamy, A. W. ( 1936). Interspecific hybrids in Plutypoecilus: One species ZZ-WZ; the other XY-XX. Proc. Natl. Acad. Sci. U.S. 22, 531-536. Bellamy, A. W., and Queal, M. (1951). Heterosomal inheritance and sex determination in Platypoecilus muculatus. Genetics 36, 93-107. Berg, O., and Gordon, M. (1953). The relationship of atypical pigment cell growth to gonadal development in hybrid fishes. In “Pigment Cell Growth” (M. Gordon, ed.), pp. 43-72. Academic Press, New York. Berkowitz, P. ( 1937). Effect of oestrogenic substances in Lebltes reticulatus. Proc. SOC. Exptl. Biol. Med. 36, 416418. Berkowitz, P. ( 1938). The effect of oestrogenic substances in Lebbtes reticulatus. (guppy). Anat. Record 71, 161-175. Berkowitz, P. (1941). The effect of oestrogenic substances in the fish (Lebistes reticulatus). I. Exptl. Zool. 87, 233-243. Blacher, L. J. (1926a). Fall von Hermaphroditism bei Lebistes. Trans. Lab. Exptl. Biol. Zoo-Park Moscow 1, 90-95. Blacher, L. J. ( 1926b). The dependence of secondary sex-characters upon testicular hormones in Lebistes reticulatus. Biol. Bull. So, 374-381. Bock, F. ( 1928). Xastration und sekundare Geschlechtscharactere bei Teleostiern. 2. Wiss. Zool. 130, 455-168. Breider, H. (1935a). Ober Aussenfaktoren, die das Geschlechtsverhiihtnis bei Xiphophorus helleri Heckel kontrollieren sollen. Z. Wiss. Zool. A, 146, 383416.

160

TOKI-0 YAMAMOTO

Breider, H. ( 1935b). Geschlechtsbestimmung und-differenzierung bei Limia nigrofascida, Caudofasciata, vittata und deren Artbastarden. Z. lnduktiue Abstammungs- Vererbungslehre 68, 285-299. Breider, H. ( 193f3a). Eine Allelenserie von Genen verschiedener Arten. Z. Znduktiue Abstammungs- Verebungslehre 72, 80-87. Breider, H. ( 1S6b). Weiteres iiber die Geschlechtsbestimmung und Geschlechtsdifferenzierung bei Limia vittata Poy. Zool. Anz. 114, 113-119. Breider, H. ( 1942). ZW-Miinnchen und WW-Weibchen bei Platypoecilus muculatus. Biol. Zentr. 62, 187-195. Breider, H. ( 1949). Der Geschlechtsbestimmungs-mechanismus in der Artdifferenzierung. Biol. Zentr. 6 4 3249. Breider, H., and Kosswig, C. ( 1937). Neue Ergebnisse der Geschlechtsbestimmunganalyse bei Zahnkarpfen. Verhundl. Deut. Zool. Ges. 1937, pp. 275-284. Bretschneider, L. H., and de Wit, J. J. D. (1940-1941). Histophysidlogische Analyse der sexuell-endokrinen Organisation des Bitterlings-weibchens ( Rhodeus ama w ) . Z . ZeUforsch. Mikroskop. Anat. 31, 227-344. Bretschneider, L. H., and de Wit, J. J. D. (1947). “Sexual Endocrinology of Nonmammalian Vertebrates.” Elsevier, Amsterdam. Bridges, C. B. ( 1922). The origin and variation in sexual and sex-linked characters. Am. Naturalist 56, 51-63. Bridges, C. B. (1925). Sex in relation to chromosomes and genes. Am. Naturalist 59, 127-137. Bridges, C. B. ( 1939). Cytological and genetic basis of sex. In “Sex and Internal Secretions” (E. Allen, ed.) I 2nd ed., pp. 15-83. Williams & Wilkins, Baltimore, Maryland. Brock, J. (1879). Beitrage zur Anatomie und Histologie der Geschlechtsorgane der Knochenfische. Gegenbaurs Jahrb. 4, 505-572. Bullough, W. S. (1940). A study of sex reversal in the minnow (Phoxinus lamis L.). J. Exptl. 2001.85, 475494. Bullough, W. S. (1947). Hermaphroditism in the lower vertebrates, Nature 160, %11. Castle, W. E. (1936). A simplified explanation of Bellamy’s experiments concerning sex determination in tropical fishes. Proc. Natl. Acad. Sci. U.S.22, 679-682. Chavin, W., and Gordon, M. (1951). Sex determination in Platypoecilus maculatus. I. Zoobgica 36, 135-145. Chester Jones, I., and Phillips, J. G. (1960). “Adrenocorticosteroids in fish.” Symp. Zool. SOC. London 1, 17-32. Chieffi, G. ( 1955). Nuove osservazioni sull’organsgenesi della mudulla gonade nei Vertebrati: Ricerche isto chemiche in Ram esculenta, Bufo viridis e Scylliohinus canicula. Pubbl. staz. Zool. Napoli. 27, 62-72. Chieffi, G. (1959). Sex differentiation and experimental sex reversal in elasmobranch fishes. Arch. Anat. Microscop. Morphol. Exptl. 4 4 21-36. Chieffi, G. (1981). La luteogenesi nei Selaci ovo-vivipari. Richerche istologiche e istochimiche in Torpedo mumzorata e Torpedo ocellata. Pubbl. Staz. Zool. Napoli 32, 145-166. Chieffi, G. (1965). Onset of steroidogenesis in the vertebrate embiyonic gonads. In “Organogenesis” (R. L. De Haan and H. Urspring, eds.), pp. 65-71. Holt, New York. Chieffi, G., and Botte, V. ( 1963). Histochemical ,reaction for steroid-3 pol-dehydrogenese in the interrenal and the corpuscles of Stannius of Anguilla anguilla and Conger conger. Nature 200, 793-794.

3.

SEX DIFFERENTIATION

161

Chiefti, G., and Lupo, C. (19sl). Identification of estradiol-17 testosterone and its precursors from Scylliorhinus stellaris testes. Nature 190, 169-170. Chieffi, G., and Lupo, C. (1962). Analisi comparativa degli ormoni sessuali negli Ittiopsidi. Excerptu Med. Intern. Congr. Ser. 51, 146-147. Chieffi, C., and Lupo, C . (1963). Identification of sex hormones in the ovarian extracts of Torpedo m a m r a t a and Bufo vulgaris. Gen. Comp. Endocrinol. 3, 149-152. Chieffi, C., Botte, V., and Visca, T. ( 1963). Attivita steroide-3-deidrogenasicanell’interrende di alkuni Selacei. Acta Med. Romana 1, 108-116. Clark, E. (1959). Functional hermaphroditism in a serranid fish. Science 129, 215-216. Clark, E. (1965). Mating of groupers. New studies detected reversal of stripes in hermaphroditic fish. Nut. Hist., N.Y. 74, 2S25. Clemens, H. P. (1965).Sex direction in presumptive female, Tikzpia Mossambica. “Intersexuality in Fishes.” Abstr. Papers, Conf. Cape Haze Marine Lab., Sarasota, Florida. Cohen, H. (1964). Effects of sex hormones on the development of the platyfish, Platypoecilus maculatus. Zoologica 31, 121-127. Conel, J. L.. ( 1917). The urogenital system of Myxinoids. J . Morphol. 29, 75-138. Conel, J. L. ( 1931). The genital system of the Myxinoidea; a study based on notes and drawings of these organs in Bdellostomu made by Bashford Dean. In “The Bashford Dean Memorial Volume Archaic Fishes” (E. W. Gudger, ed.), pp. 63110. Am. Museum Nat. Hist., New York. Craig-Bennett, A. ( 1931). The reproductive cycle of the three spinned stickleback, Gasterostew aculeatus, Linn. Phil. Trans. Roy. SOC. London, Ser. B, 219, 197278. D’Ancona, U. ( 1941) , Ulteriori osservazione e considerazioni sull’ermafroditismo dell’orata (Sparus auratus I.,,). Pubbl. Staz. Zool. Napoli 18, 314-336. D’Ancona, U. ( 1943). Nuove ricerche sulla determinazione sessuale dell’anguilla. Parte I. Lo sviluppo della gonade e i processi della gametogenesi. Arch. Oceanog. Limnol. (Roma) 3, 159-269. D’Ancona, U. (1945). L’ermafroditismo nel genere Diplodus (sin. Sargus; Teleostei). Boll. Soc. ltal. Biol. Sper. 20, 153-155. DAncona, U. ( 1948). Prime osservazioni sull’azioni deghi ormoni sessuali sulla gonade dell’anguilla. Atti. Accad. Nazl. Lincei, Rend., C h s e Sci. Fis., Mat. Nut. 181 5, 82-87. D’Ancona, U. ( 1949a). Osservazioni sull’organizzazione nella gonade ermafrodita di alcuni Serranidi. Nova Thalmsia 1(S ) , 1-15. D’Ancona, U. (1949b). I1 differenziamento della gonade e l’inversione sessuale degli Sparidi. Arch. Oceanog. Limnol. (Romu) 6, 97-163. . ed intersexualitd nei Teleostei. Proc. 13th D’Ancona, U. ( 1 9 4 9 ~ ) Ermofroditismo Intern. Congr. Zool., Park, 1948, pp. 144-146. D’Ancona, U. ( 1949d). Ermafroditismo ed intersessualita nei Teleostei. Erpetientia 5, 381-389. D’Ancona, U. (1950). DBtermination e differenciation du sexe chez les poissons. Arch. Anat. Microscop. Morphol. Exptl. 39, 274-294. D’Ancona, U. ( 1956). Inversion spontankes et exphrimentales dam les gonades des TeleostBens. Annee Biol. [3] 32, 89-99. Dantschakoff, V. (1941). “Der Aufbau des Geschlechts beim kohren Wirbeltier.” Fischer, Jena. de Groot, B., and de Wit, J. J. D. (1949). On the artificial induction of ovipositor growth in the bitterling (Rhodeus amarus Bl.). 11. Ovipositor growth caused by

162

TOKI-0 YAMAMOTO

different chemical and physical agents. 111. The relation between artificially induced ovipositor growth and the adaptation syndrome of Selye. Acta Endocrinol. 3, 266-297. de Wit, J. J. D. (1939). Ovipositor lengthening of the female bitterling produced by administration of progesterone. Endocrinology 24, 580-582. de Wit, J. J. D. (1940). A quantitative and qualitative test for steroid hormones based on the ovipositor reaction of the female bitterling (Rhodeus amarus Bloch). J . Enducrinol. 2, 141-156. de Wit, J. J. D. (1941 ) , Eine neue Testmethode fiir endokrinologische Zweck (der Legerohrentest am weiblichen Bitterling). Arch. Ges. Physiol. 245, 282-299. de Wit, J. J. D. (1955). Some results of investigations into European bitterling. Rhodeus a m u w Bloc. Japan. 1. Ichthyol. 9, 94-104. Dildine, G. C. ( 1936). Studies on teleostean reproduction. I. Embryonic hermaphroditism in Lebistes reticulatus. J . Morphol. 60,261-277. Dodd, J. M. (1955). The hormones of sex and reproduction and their effects in fish and lower chordates. Mem. SOC. Endocrinol. 4, 166187. DufossB, M. (1854). Sur l'hermaphroditisme de certain Percoides. L'lnstitut 22, 394-395. Dufossb, M. (1856). De l'hermaphroditisme chez le serran. Ann. Sci. Nut. 141 5, 295-332. Dulzetto, F. (1933). La struttura del testicolo di Gambusia holbrookii e la sua evoluzioni in rapport0 con lo sviluppo del gonopodio. Arch. Zool. Ztal. 19, 405437. Dzwillo, M. ( 1962). Wber kiinstliche Erzeugung funktionellee Mannchen weiblichen Genotypus bei Lebistes reticulatus. Bwl. Zentr. 81, 575-584. Dzwillo, M., and Zander, C. D. (1967). Geschlechtsbestimmung und Geschlechtsnmstimmung bei Zahnkarpfen (Pisces). Mitt. Hamburgischen Zool. Museum Inst. 64, 147-162. Egami, N. (1954a). Appearance of the male character in the regenerating and transplanted rays of the anal fin in females of the fish Oryzias latipes, following treatment with methyl dehydroxytestosterone. J . Fac. Sci., Uniu. Tokyo, Sect. IV 7, 271-280. Egami, N. (195413). Effects of hormonic steroids on the formation of male characteristics in female of the fish Oyzias Zatipes, kept in water containing testosterone propionate. Annotationes 2001.Japon. 27, 122-127. Egami, N. (1955a). Production of testis-ova in adult males of Oryzias Iatipes. I. Testis-ova in the fish receiving estrogens. Japan. J . Zool. 11, 21-34. Egami, N. (1955b). Production of testis-ova in adult males of Oyzias latipes. 11. Effect on testis-ova production of non-estrogenic steroids given singly or simultaneously with estradiol. Japan. J. Zool. 11, 35-39. Egami, N. ( 1 9 5 5 ~ ) .Production of testis-ova in adult males of Oryzias latipes. 111. Testis-ova production in starved males. J . Fac. Sci., Uniu. Tokyo, Sect. IV 7, 421-428. Egami, N. (1955d). Production of testis-ova in adult males of Oryzias latipes. 1V. Effect of X-ray irradiation on testis-ovum production. J . Fac. Sci., Uniw. Tokyo, Sect. IV 7, 429-441. Egami, N. (1955e). Production of testis-ova in adult males of Oryzias latipes. V. Note on testis-ovum production in transplanted testes. Annotationes 2001.Japon. 28, 206-u)9. Egami, N. (1956a). Production of testis-ova in adult males of 0ryziu.s Zatipes. VI. Effect on testis-ova production of exposure to high temperature. Annotationes Zool. Japon. 29, 11-18.

3.

SEX DIFFERENTIATION

163

Egami, N. (195613). Note on sexual difference in size of teeth of the fish, Oryzia~ latipes. Japan. J . 2001.12, 66-69. Egami, N. (1957). Production of testis-ova in adult males of Oryzias htipes. VIII. Effect of administration of desiccated frog pituitary on testis-ovum production in males receiving estrone pellet. J. Fac. Sci., Hokkaido Uniu. Ser. VZ 13, 36-72. Egami, N. (1959a). Note on sexual difference in the shape of the body in the fish, Oryzias latipes. Annotationes Zool. ]upon. 32, 59-64. Egami, N. (195913). Note on sexual difference in the shape of the body in the fish, Oryzias latipes. Annotationes 2001.Japon. 32, 59-64. Egami, N. ( 1 9 5 9 ~ )Effect . of testosterone on the sexual characteristics of the gobiid fish, Pterogobius zonoleucus. Annotationes 2001.Japon. 32, 123-128. Egami, N., and Ishii, S. (1956). Sexual differences in the shape of some bones in the fish, Oryzios latipes. J. Fac. Sci., Uniu. Tokyo, Sect. iV 7, 563-571. Eggert, B. (1931). Die geschlechts Organe der Gobiiformes und Blenniformes. Z. Wiss. 2001.189, 249-558. Eggert, B. (1933). Die Intersexualitat bei Knochenfischen. Z. Wiss.2001.144, 402420. Ehrhardt, K., and Kiihn, K. (1933). Uber einen neuen Test fur weibliches Sexual hormon. Monatsschr. Geburtshulfe Gynaekol. 94, 64. Essenberg, J. M. (1923). Sex differentiation in the viviparous teleost Xiphophom helleri. Biol. Bull. 45, 4 6 9 7 . Essenberg, J. M. ( 1926). Complete sex-reversal in the viviparous teleost Xiphophorus helleri. Biol. Bull. 51, 98-111. Eversole, W. J. ( 1939). The effects of androgens upon the fish (Lebistes reticulatus). Endocrinology 25, 328-330. Eversole, W. J. (1941). Effect of pregneninolone and related steroids on sexual development in fish Lebistes reticulatus. Endocrinology 28, 603410. Fleischmann, W., and Kann, S. (1932). Uber eine Funktion des weiblichen Sexualhons bei Fischen (Wachstum der Legerohre des Bitterlings). Arch. Ges. Physiol. 230, 662-667. Fraser, A. C., and Gordon, M. ( 1929). The genetics of Platypoecilus. 11. The linkage of the two sex-linked characters. Genetics 14, 160-179. Freund, I. ( 1923). “Bibliographia Pathologiae Piscium.” Prag. Friedman, B., and Gordon, M. (1934). Chromosome numbers in xiphophorin fishes. Am. Naturalist 68, 446455. Friess, E. ( 1934). Untersuchungen iiber die Ceschlechtsumkehr bei Xiphophorus helleri. Arch. Entwicklungsmech. Organ. 129, 255-355. Gabler, H. ( 1930). Zwei Falle von Zwittergonaden bei Clupea hurangus L. Zool. Anz. 91, 72-75. Gallien, L. ( 1948). Rhactiviti: sp6cifique 6. la prhgnbninolone chez Lebistes reticulatus, des gonades et des caracthres sexuels somatique. Compt. Rend. 223, 5 2 5 3 . Gallien, L. ( 1948). Sur la structure nageoire chez les femelles de Lebistes reticulatus Regan, masculiniskes par pregneninolone. Compt. Rend. 226, 1749-1751. Glaser, E., and Haempel, 0. ( 1932). Das experimentell hervorgerufene Hochzeitskleid des kastrierten Fisches als Stigme einer Test-und Standardisierungs method des mannlichen Sexual hormones. Arch. Ges. Physiol. 229, 1-14. Coldschmidt, R. ( 1915). Vorlaufige Mitteilung iiber weiteren Versuche zur Vererbung und Bestimmung des Geschlechts. Biol. Zentr. 35, 565570. Coldschmidt, R. ( 1927). “Physiologische Theorie der Vererbung.” Springer, Berlin. Goldschmidt, R . ( 1931 ) , “Die sexuelle Zwischenstufen.” Springer, Berlin.

164

TOKI-0 YAMAMOTO

Goldschmidt, R. (1937). A critical review of some recent work in sex determination. I. Fishes. Quart. Rev. Biol. 12, 426-439. Goodrich, H. B., Dee, J. E., Flynn, B. M., and Mercer, R. N. (1934). Germ cells and sex-differentiation in Lebistes reticulatus. Biol. Bull. 67, 83-96. Gordon, M. (1926). Melanophores of Platypoecilus, the top minnow of geneticists. Anat. Record 34, 138 (abstr.). Gordon, M. (1927). The genetics of a viviparous top-minnow Platypoecilus; the inheritance of two kinds of melanophores. Genetics 22, 376-392. Gordon, M. (1946a). Genetics of Xiphophom maculatus. Sex determining in two wild populations of the Mexican platyfish. Genetics 31, 641 (abstr.). Gordon, M. ( 194613) . Interchanging genetic mechanisms for sex-determination. 1. Heredity 37, 307320. . transformation of a genetically constituted female fish Gordon, M. ( 1 9 4 6 ~ ) Sexual (Platypoecilus muculatus) into a functional male. Anat. Record 943, 6 (abstr.). Gordon, M. (1947a). Genetics of Platypoecilus maculatus. IV. The sex determining mechanism in two wild populations of Mexican platyfish. Genetics 32, 8-17. Gordon, M. (194713). Speciation in fishes. Distribution in time and space of seven dominant multiple alleles in Platypoecilus maculatus. Aduan. Genet. 1, 95-132. Gordon, M. (1948). Effects of five primary genes on the site of melanomas in fishes and the influence of two color genes on their pigmentation. Spec. Publ. N.Y. Acad. Sci. 4, 216-268. Gordon, M. (1950). Genetics and speciation in fishes. Am. Phil. SOC., Year Book pp. 158-159. Gordon, M. (195la). Genetics of Plutypoecilus macuhtus. V. Heterogametic sexdetermining mechanism in females of a domesticated stock originally from British Honduras. Zoologica 36, 127-134. Cordon, M. ( 1951b). Platypoecilus now becomes Xiphophorus Aquarium 20, 277279. Gordon, M. ( 1 9 5 1 ~ )Genetics . and correlated studies of normal and atypical pigment cell growth. Growth S y m p . 10, 153-219. Gordon, M. ( 1952). Sex-determination in Xiphophorus (Phtypoecilus) maculatus. 111. Differentiation of gonads in platyfish from broods having a sex ratio of the three females to one male. Zoologica 37,91-100. Gordon, M. (1956). “Swordtail-The Care and Breeding of Swordtails.” T. F. H. Publ. Inc., New Jersey. Gordon, M. (1957). Physiological Genetics of fishes. I n “The Physiology of Fishes” ( M. E. Brown, ed. ), Vol. 2, pp. 431-501. Academic Press, New York. Gordon, M., and Aronowitz, 0. ( 1951) . Sex determination in Pldypoecilus muculatus. 11. History of a male platyfish that sired all-female broods. Zoologica 36, 147-153. Gordon, M.,and Smith, G. M. (1938). The production of a melanotic neoplastic disease in fishes by selective matings. IV. Genetics of geographical species hybrids. Am. 1. Cancer 34, 543-565. Gottfried, H., Hunt, S. V., and Wright, R. S. (1962). Sex hormones in fish. The estrogens of cod Gadus callarias. 1. Endocrinol. 84, 425-430. Grassi, B. (1919). Nuove ricerche sulla storia naturale dell’Anguilla. R. Comitato Tahsografico Itd. M . 67, 1-141. Grobstein, C. ( 1940a). Endocrine and developmental studies of gonopod differentiation in certain poeciliid fishes. I. The structure and development of the gonopod in Plutgpoecilus maculatus. Univ. Calif. (Berkeley) Publ. Zool. 47, 1-22. Grobstein, C. (1940b). Effect of testosterone propionate on regeneration of anal fin

3.

SEX DIFFERENTIATION

165

of adult Platypoecilus maculatus females. PTOC. Sac. Exptl. Biol. Med. 45, 484486. Grobstein, C. (1942a). Effect of various androgens on regenerating anal fin of adult Platypoecilus maculatus females. Proc. Sac. Erptl. Biol. Med. 49, 477478. Grobstein, C. ( 1942b). Endocrine and developmental studies of gonopod differentiation in certain poeciliid fishes. 11. Effect of testosterone propionate on the normal and regenerating anal fin of adult Platypoecilus m u d a t u s females. J . Exptl. Zool. 89, 305428. Grobstein, C. ( 1948). Optimum gonopodial morphogenesis in Platypoecilus moculatus coith constant dosage of methyl testosterone. J . Exptl. Zool. 109, 215-236. Hardisty, M. W. (1960). Sex ratio of ammocoetes. Nature 186, 988-989. Harms, J. W. ( 1926a). Beobachtungen iiber Geschlechtsumwandlungen reifer Tiere und deren Fl-Generation. Zool. Anz. 67, 67-79. Harms, W. (1926b). “Korper und Keimzellen.” Springer, Berlin. Harrington, R. W., Jr. (1961). Oviparous hermaphroditic fish with internd selffertilization. Science 135, 1749-1750. Harrington, R. W., Jr. ( 1965). Intersexuality in Rivulus marmoratus. In “Intersexuality in Fishes,” Abstr. Papers, Conf. Cape Haze Marine Lab., Sarasota, Florida. Harrington, R. W., Jr., ( 1967). Environmentally controlled induction of primary male gonochorists from eggs of the self-fertilizing hermaphroditic fish, Rivulus marmoratus Poey. Biol. Bull. 132, 174-199. Hartmann, M. ( 1951). “Geschecht und Geschlechtsbestimmung in Tier- und Pflanzenreich.” de Gruyter, Berlin. Hartmann, M. ( 1956). “Die Sexualitat.” Fischer, Stuttgart. Haskins, C. P., Haskins, E. F., and Hewitt, R. E. (1960). Pseudogamy as an evolutionary factor in the poeciliid fish Mollienisia formosa. Evolution 14, 473-483. Hickling, C. F. (1960). The Malacca Tilapia hybrids. J. Genet. 57, 1-10, Hild, S. ( 1940). Versuche zur Geschlechtsdifferenzierung von Xiphophows helleri. Jena. 2. Naturw. 73, 135-143. Hildemann, W. H. (1954). Effects of sex hormones on secondary sex characters of Lebistes reticulutus. J . Exptl. Zool. 126, 1-15. Hishida, T. ( 1962). Accumulation of testosterone-4-C14 propionate in larval gonad of the medaka, Oryzias latipes. Embryologia Nagoya 7, 5 5 6 7 . Hishida, T. ( 1965). Accumulation of estrone-16-C” and diethylstilbestrol (Monoethyl-1-C”) in larval gonads of the medaka, Uryzias lutipes, and determination of the minimum dosage of estrogen for sex reversal. Gen. Camp. Endocrinol. 5, 137-144. Hoar, W. S. (1955). Reproduction in teleost fish. Mem. SOC. Endocrinol. 4, Part 1, 5-24. Hoar, W. S. (1957a). Endocrine organs. In “The Physiology of Fishes” (M. E. Brown, ed.), Vol. 1, pp. 254-285. Academic Press, New York. Hoar, W. S. (1957b). The gonads and reproduction. In “The Physiology of Fishes” ( M. E. Brown, ed.), Vol. 1, pp. 287-321. Academic Press, New York. Hoar, W. S. (1965). Comparative physiology: Hormones and reproduction in fishes. Ann. Rev. Physiol. 27, 51-70. Hoar, W. S., Black, V. S., and Black, E. C. (1951). Some aspects of the physiology of fish. I. Hormones in fish. Publ. Ontario Fisheries Res. Lab., 61, 1-52. Hoek, P. P. C. (1891a). Hermaphroditisme bij visschen. Verhandl. 3rd Ned. Nut. Geneesk. Congr. pp. 185-187. Hoek, P. P. C. (1891b). Over het hermaphroditisme van de visschen uit de familien der Percidae en Sparidae. Tiidschr. Ned. Dierk. Ves. [ 2 ] 3, 37-38.

166

TOKI-0 YAMAMOTO

Holly, M. ( 1930). Zur Geschlechtsumbildung bei Lebendgebarern. Bl. Aquar.Terrarienk. 41, 114-119. Hopper, A. F. ( 1943). The early embryology of Platypoecih maculatus. Copeia pp. 218-224. Hopper, A. F. (1949). The effect of ethynyl testosterone on the intact and regenerating anal fins of normal and castrated females and normal males of Lebistes reticulatus. J. Erptl. 2001.111, 393414. Hubbs, C.,Jr. (1964). Interaction between a bisexual fish species and its gynogenetic sexual parasite. Bull. Trans. Mem. Museum 8, 1-72. Hubbs, C., Drewry, G . E., and Warburton, B. (1959). Occurrence and morphology of a phenotypic male of a gynogenetic fish. Science 129, 1227-1229. Hubbs, C. L., and Hubbs, L. C. (1932). Apparent parthenogenesis in nature, in a form of fish of hybrid origin. Science 7’6,628-630. Hubbs, C. L., and Hubbs, L. C. (1946). Breeding experiments with the invariable female, strictly matroclinous fish, Mollienisia formosa. Rec. Genet. Soc. Am. 14, 48. Hudey, J. (1938). In “Essays in Popular Science,” p. 45. Penguin Books, Ltd., Harmondsworth, England. Ikeda, K. ( 1933). Effect of castration on the sexual characters of anadromous threespined sticklehead, Gasterosteus aculeatus aculeatus ( L. ), Japan. J. Zool. 5, 135-157. Jordan, D. S., and Snyder, J. 0. (1902). A review of the labroid fishes and related forms found in the waters of Japan. Proc. US. Natl. Museum 24, 595-662. Kaiser, P., and Schmidt, E. ( 1951 ). Vollkommene Geschlechtsumwandlung beim weiblichen siamesischen Kampffish, Betta splendens. Zool. Anz. 146, 66-73. Kallman, K. D. ( 1962). Gynogenesis in the teleost, Mollienesia formosa (Girard), with a discussion of the detection of parthenogenesis in vertebrates by tissue transplantation. J . Genet. 58, 7-24. Kallman, K. D. (1964). Homozygosity in a gynogenetic fish. Genetics 50, 260-261. Kallman, K. D. (1965a). Sex determination in the teleost Xiphophorus milleri. Am. Zoologist 5, 246-247. Kallman, K. D. ( 1965b ) . Genetics and geography of sex determination in the poeciliid fish, Xiphophorus maculatus. Zoologica 50, 151-190. Kar, A. B., and Ghosh, A. (1950). Responses of the genital system and the urogenital papilla of the female catfish Heteropneustes fossilis (Block) to sex hormones. Arch. Entwicklungsmech. Organ. 144, 257-264. Kawaguchi, K., and Marumo, R. (1967). Biology of Gonostoma gracile (Gonostomatidae). I. Morphology, life history and sex reversal, Inform. BUZZ. PlanJctoZ. Japan., Commen. No. of Dr. Y . Matsui pp. 53-67. Kinoshita, J. (1936). On the conversion of sex in Sparus longispinis (Temminck et Schlegel) (Teleostei). 1. Sci. Hiroshima Uniw. B4, 69-89. Kinoshita, J. (1939). Studies on the sexuality of genus Sparus (Teleostei). J . Sci. Hiroshima Univ. B7, 25-37. Kinoshita, Y. (1934). On the differentiation of the male color pattern, and the sex ratio in Halichoeres poecilopterus ( Temminck and Schlegel). J . Sci. Hiroshimu Uniw. B3, Div. 1, 65-76. Kinoshita, Y. (1935). Effect of gonadectomies on the secondary sexual characters in Halichoeres poecilopterus (Temminck and Schlegel). J . Sci. Hiroshima Uniw. B4, Div. 1, 1-74. Kinoshita, Y. (1936). Testis-ova found in the Japanese wrasse, Hdichoeres poecilopterus, and its sex reversal. Botany & Zool. (Tokyo) 4, 35-38 (in Japanese).

3. SEX

DIFFERENTIATION

167

Kleiner, I. S., Weisman, A,, and Mishkind, D. I. (1937). The similarity of action of male hormones and adrenal extracts on the female bitterling. Science 85, 75. KopG, S. (1918). Contribution to the study of the development of the nuptial colors of fishes. Sprawozdania Posiedzen Towarz. Nauk. Warszaw. 3, 11. Kopt.6, S. (1928). Experiment on depence of nuptial hue in the gonads in fish. Biol. Generalis 3, 259. Kosswig, C. ( 1932). Genotypische und phanotypische Ceschlechtsbestimmung bei Zahnkarpfen und ihren Bastarden. I. 2. lnduktive Abstammungs- Vererbungslehre 62, 23-34. Kosswig, C. ( 1933). Genotypische und phanotypische Geschlechts bestimmung bei Zahnkarpfen und ihren Bastarden. 111. Arch. Entcoicsklungsmech. Organ. 128, 393446. Kosswig, C. ( 1935a). Genotypische und Phanotypische Geschlechts bestimmung bei Zahnkarpfen. V. Ein X (Z)-Chromosom als Y-chromosom in fremdem Erbgut. Arch. Entwicklungsmech. Organ. 133, 118-139. Kosswig, C. ( 1935b). Genotypische und phanotypische Geschlechts bestimmung bei Zahnkarpfen. VI. Arch. Entwicklungsmech. Organ. 133, 140-155. Kosswig, C. ( 1939a). Ceschlechtsbestimmungs analyse bei den Zahnkarpfen. Istanbul Univ. Fen. Fak. Mecmuasi B4, 239-270. Kosswig, C., ( 1939b). Uber einen neuen Farbfaktor des Plutypoecilus maculutus. Istanbul Uniu. Fen, Fak. Mecmuasi B4, 395402. Kosswig, C. ( 1954). Zur Geschlechtsbestimmungs analyse bei den Zahnkarpfen. Istanbul Univ. Fen. Fak. Mecmuasi B19, 187-190. Kosswig, C. ( 1959). Beitrage zur genetischen Analyse xiphophoriner Zahnkarpfen. Biol. Zentr. 78, 711-718. Kosswig, C. ( 1964). Polygenic sex determination. Experientia 20, 1-10, Kosswig, C. ( 1965). Polygene Geschlechtsbestimmung. Naturw. Rundschuu 18, 392401. Kosswig, C., and Oktay, M. (1955). Die Geschlechtsbestimmung bei den Xiphophorini ( Neue Tatsachen und neue Bedeutung ) . IstanbuE Uniu. Fen. Fak. Hidrobiol. 2, 133-156. Kuroda, N. (1931). Male and female of Sakura-dai and their scientific names. Zool. Mag. (Tokyo) 43, 627-628 (in Japanese). Larraiieta, M. G. (1953). Observaciones sobre la sexualidad de Pagellus erythrinus L. Publ. Inst. Biol. Apl. (Barcelona) 13, 83-101. Larraiieta, M. G. ( 1964). Sobre la biologia de Pugellus erythrinus (L.) especialmente del de costas de Castellon. Invest. Pesquera 27, 121-146. Laskowski, W. ( 1953). Reaktionen der primaren und sekundaren Geschlechtsmerkmale von Pkztypoecilus variatus ( 8 Heterogamet) und Platypoecilus m c u h t u s ( 8 Homogamet) auf Sexual hormone. Arch. Entwicklungsmech. Organ. 146, 137-182. Lavenda, N. ( 1949). Sexual differences and normal protogynous hermaphroditism in the Atlantic sea bass, Centropristes striatus. Copeia 3, 185-194. Le Gall, J. ( 1929). Note sur la sexualitbe de la dorade ( Pugellus centrodontus) reported. Rev. Trau. Office Peches Maritimes 2, 31-32. Lepori, N. G . (1942a). Azione della follicolina sul sesso di Gumbusiu holbrookii (GRd.). Atti SOC. Toscanu Sci. Nut. Mem. 51, 1-19. Lepori, N. G. ( 1942b). Inversione sperimentale delle gonadi femminili di Gumbzrsia holbrooki Grd. trattala con testosterone. Monit. Zool. Ital. 53, 117-131. Lepori, N. G. (1945). Osservazioni sul determinism0 dei carrateri sessuali nei pesci. Rio. Biol. (Perugia) 37# 67-89.

168

TOKI-0 YAMAMOTO

Lepori, N. G. (1846). Effecti nulli degli ormone sessuali sulle creste genitali delle ceche. Atti SOC.Toscam Sci. Nut. Mem. 53, 1-9. Lepori, N. G. ( 1948). Osservazioni sulla “Intersessualita transitoria” nei pesci. Boll. Pesca, Piscicolt. ldrobiol. [N.S.] 11, 3-19. Lepori, N. G. (1960). Ermafroditismo Proteroginico in Maena maena ( L . ) ed in Maena chryselis (Cuv. e Val.) (Perciformes, Centracanthidae). Boll. Pesca, Piscicolt. Idrobiol. [N.S.] 14, 155-165. Lieder, U. ( 1955). Mannchenmangel und natiirlische Parthenogenese bei der Silberkarausche Carassius auratus gibeIio ( Vertebrate, Pisces). Natuwissenschaften 42, 590. Liem, K. F. (1963). Sex reversal as a natural process in the Synbranchiform fish Monopterus albus. Copeia No. 2, 303-312. Liem, K. F. ( 1965). Intersexuality in symbranchoid eels. I n “Intersexuality in Fishes,” Abstr. Papers, Conf. Cape Haze Marine Lab., Sarasota, Florida. Lissia Frau, A. M. (1964). Osservazioni sul differenziamento sessuale di Denter dentex (L. ). (Teleostei, Sparidae). Boll. Zool. 31, 783-747. Liu, C. K. ( 1944). Rudimentary hermaphroditism in the symbranchoid eel, Monopterus favanensis. Sinensia 15, 1-18. Liinneberg, E., and Gustafson, G. (1937). Contributions to the life history of the striped wrasse, Labrus ossifagus. Arkiv Zool. 2 9 4 No. 7, 1-16. Lozano Cabo, F. ( 1951) . Nota preliminare sull’ identificazione di Spicaris smdis L. e Spicaris alcedo C. V. come sessi differenti di una medisma con process0 di inversione sessuale. Atti. Accadi. Naz. Lincei, Rend., Clusse Sci. Fis., Mat. Nat. [8] 11, 127-132. Lozano Cabo, F. (1953). Monographia de 10s Centracandidos mediterraneos, con etude especial de la biomebia, biologia y anatomia de Spicarfs s m l i s L. Mem. Acad. Cienc. Madrid 17, No. 2, 1-128. Lubosch, W. ( 1903). Ueber die Geschlechtsdifferenzierungbei Ammocoetes. Anat. Anz. 23, 6574. Lupo, C., and Chieffi, G. ( 1963). Oestrogens and progesterone in ovaries of the marine teleost Conger conger. Nature 197, 598. MacIntre, P. A. (1961). Spontaneous sex reversals of genotypic male in the platyfish Xiphophorus maculutus. Genetics 46, 575-580. McLeod, J. (1881). Recherches sur la structure et le d6veloppment de l’appareil reproducteur femelle des T6l&st&ns. A d . Biol. (Liege) 2, 497-530. Mead, G. W. (1960). Hermaphroditism in archibenthic and pelagic fishes of the Order Iniomi. Deep-sea Res. 6, 234-235. Meyer, H. ( 1938). Invertigation concerning the reproductive behavior of Mollienisia formosa. J. Genet. 36, 329386. Meyers, A. S. ( 1955a). 19-Hydroxylation of Atandrostene-3, 17-dione and dehydroepiandrosterone by bovine adrenals. Experientia 11, 99-102. Meyers, A. S. ( 1955b). Conversion of 19-Hydroxy-A’-androstene-3, 17-dione to estrone by endocrine tissue, Biochim. Biophys. Acta 17, 441-442. Miller, R. R. (1960). Four new species of viviparous fishes, genus Poeciliopsis, from northwestern Mexico. Occasional Papers Museum Zool., Uniu. Mich. 619, 1-11. Miller, R. R., and Schultz, R. J. (1959). All female strains of the teleost fishes the genus Poesiliopsis. Science 130, 16551657. Miyamori, H. ( 1961). Sex modification of Lebistes reticulatus induced by androgen administration. Zool. Mag. (Tokyo) 70, 310332 (in Japanese). Mrliib, W. (1923). Die Spatbefruchtung und deren Einfluss auf Entwicklung und

3.

SEX DIFFERENTIATION

169

Geschlechtsbildung, experimentell nachgefriift an der Regenbogenforelle. Arch. Mikroskop Anut. Entwicklungsmech. 98, 129-209. MrgiC, W. (1930). Uber das Auftreten intermediarer Stadien bei der GeschlechtsdifFerenzierung der Forelle. Arch. Entwiklungsmech. Organ. 123, 301332. Muller, H. J. (1932). Some genetic aspects of sex. Am. Naturalist 66, 118-138. Nagata, Y. (1934). Castration experiments on the medaka, Oryzias kztipes. Zool. Mag. ( T o k y o ) 46, 2fXL294 (in Japanese). Nagata, Y. (1936). Transplantation of testis into ovariotomized Oryzias latipes. 2001. Mag. ( T o k y o ) 48, 102-108 (in Japanese). Niwa (Suzuki), H. ( 1985a). Effects of castration and administration of methy testosterone on the nupital coloration of the medaka (Oryzias latipes). Embryologia (Nagoya) 8, 289-298. Niwa (Suzuki), H. ( 1965b). Inhibition by estradiol of methyltestosterone-induced nupital coloration in the medaka (Oryzia lutipes). Em.blyologiu (Nagoya) '8, 299307. Noble, G. K., and Kumpf, K. F. R. (19.37).Sex reversal in the fighting fish, Betta splendens. Anut. Record 70, 97. Nogusa, S. (1955).Chromosome studies in Pisces. IV. The chromosome of Mogrundu obscura (Gobudae), with evidence of male heteromamety. Cytologia ( T o k y o ) 20, 11-18. Nogusa, S. (1957). Chromosome studies in Pisces. VI. The X-Y chromosomes found in Cottux poElux Giinth. (Cottidae). J. Fac. Sci., Hokkaido Univ., Ser. Vl 13, 289-292. Oka, T. B. (1931a). On the processes on the fin rays of the male of Oryzias htipes and other sex characters of this fish, J . Fac. Sci., Imp. Univ. Tokyo, Sect. IV 2, 209-218. Oka, T . B. (1931b). On the accidental hermaphroditism in Oryzlas lutipes. J. Fac. Sci., Imp. Univ. Tokyo, Sect. IV 2, 219-224. Okada, YB K. ( 1943a). A review of literature on secondary sexual characters in fish and their experimental consideration. Ann. Reu. Exptl. Morphol. 1, 34-64 (in Japanese). Okada, YB K. (1943b). Production of testis-ova in Oryzias latipes by oestrogenic substances. Proc. Imp. Acad. ( T o k y o ) 19, 501-504. . the bipotentiality of the sex cells in poeciliid fish. Proc. Okada, YB K. ( 1 9 4 3 ~ )On Imp. Acad. ( T o k y o ) 20, 244-250. Okada, YB K. (1944). On the bipotentiality of the sex cells in poeciliid fish. Proc. Imp. Acad. ( T o k y o ) u), 244-250. Okada, YB K. (1949). Experimental studies on the testis-ova in fish. Zikkenkeitaigaku (Exptl. Morphol. ) 5, 149-151 (in Japanese). Okada, Yi3 K. (1952a). Experimental production of testis-ova in the fish Olyzias kztipes. Papers Coord. Comm. Res. Genet. 3, 139-142 (in Japanese). Okada, YB K. (1952b). Studies on the sex and sex separation in a fish, ,Sparus longispinis. Papers Coord. Comm. Res. Genet. 3, 147-150 (in Japanese). Okada, YB K. (1962). Sex reversal in the Japanese wrasse, Halichoeres poecilopterus. PTOC.Japan Acad. 38, 508-513. Okada, Y6 K. (1964a). A further note on sex reversal in the wrasse, Halichoeres poecilopterus. Proc. Japan Acad. 409533535. Okada, YB K. ( 1964b). Effects of androgen and estrogen on sex reversal in the wrasse, Halichoeres poecilopterus. Proc. Japan Acad. 40, 541-544.

170

TOKI-0 YAMAMOTO

Okada, YB K. ( 1 9 6 4 ~ ) .A further note on testis-ova in the teleost, Oyzias latipes. Proc. Japan Acad. 40, 753-756. Okada, Y6 K. (1965a). Sex reversal in the serranid fish, Sacura masgaritacea. I. Sex characters and changes in gonads during reversal. Proc. Japan h a d . 41, 737740. Okada, YB K. (1965b). Sex reversal in the serranid fish, Sacura murgaritacea. 11. Seasonal variations in gonads in relation to sex reversal. Proc. Japan Acad. 41, 741-745. Okada, YB K. (1966). Sex reversal in Inegocia meerdervoort with special reference to repitation of hermaphroditic state. Psoc. Japun. Acad. 42, 497-502. Okada, YB K., and Yamashita, H. (1944a). Experimental investigation of the manifestation of secondary sexual characters in fish, using the medaka, Oryzias latipes (Temminck and Schlegel) as material. 1. Fac. Sci., Imp. Univ. Tokyo, Sect. IV 6, 383437. Okada, YB K., and Yamashita, H. (194413). Experimental investigation of the sexual characters of poeciliid fish, J . Fac. Sci., Imp. Uniu. Tokyo, Sect. IV 6, 589-633. Okkelberg, P. ( 1921). The early history of the germ cells in the brook lamprey, Entosphenus wilderi, up to and including the period of sex differentiation. J. Morphol. 35, 1-152. Oktay, M. (1959). Uber Ausnahmemaennchen bei Platypoecilus mculatus und eine neue Sippe mit XX-Maennchen und XX-Weibchea. lstanbul Uniu. Fen Fak. Mecmuasi B24, 75-92. Oktay, M. ( 1963). Die Rolle artfremder Gonosomen bei der Geschlechtsbestimmung von Bastarden mit Platypoecilus riphidim. Istanbul Uniu. Fen Fak. Hidrobiol. B6, 1-13. Padoa, E. ( 1937). Differenziazone e inversione sessuale (feminizzazione) di avanotti di Trota (Salmo irideus) trattati con ormone follicolare. Monit. Zool. Ital. 48, 195-203. Padoa, E. (1939a). Observations ultkrieures sur la daerenciation du sexe, nonnale et modifike par l’administration d’hormone folliculaire, chez la truite i r i d h ( Salmo irideus). Bio-Morphosis 1, 337-354. Padoa, E. (193913). Prime osservazioni sulle gonadi di un telesteo ermafrodita (Hepatus hepatus L.) trattati con ormone femmile. Monit. 2001.Ital. SO, 129132. Pasquali, A. ( 1941). Contributo all0 studio dell’ermafroditismo e del differenziamento della gonade nell’orata ( Sparus auratus L. ) . Pubbl. Staz. Zool. Napoli 18, 282312. Peters, G. ( 1964). Vergleichende Untersuchungen an drei Subspecies von Xiphophorus helleri Heckel (Pisces). Z. 2001.Syst. Eoo1ut.-Forsch. 2, 18Ei271. Philippi, E. ( 1908) . Fortpflanzungsgeschichte der viviparen Teleostier Glaridichthyes januarius und Gbridichthyes decemmaculatus in h e n Einfluss auf Lebensweise, makro- und mikroscopische Anatomie. Zool. Jahrb., Abt. Anut. Ontog. Tiere 27, 1-94. Pickford, G., and Atz, J. (1957). “Physiology of the Pituitary Gland in Fishes.” N.Y. Zool. SOC., New York. Popoff, W. ( 1929). Wber Grosse des Schwerffishmiinnchens (Xiphophorus helleri), 2001.Anz. 86, 159-160. Querner, H. (1956). Der E i d u s s von Steroidhormonen a d die Gonaden juveniler Poeciliiden. Biol. Zentr. 75, 28-51. Rasch, E. M., Darnell, R. M., Kallman, K. D., and Abramoff, P. (1965). Cytophoto-

3.

SEX DlFFERENTIATION

171

metric evidence for triploidy in hybrids of the gynogenetic fish, Poecilia f o m s a . I . Exptl. 2001.160, 155-170. Rbgnier, M. T. (1938). Contribution B I’etude de la sexualit6 des cyprinodontes vivpares (Xiphophorus helleri, Lebistes reticulatus). Bull. Biol. France Belg. 72, 385-493. Reinboth, R. ( 1960). Natiirliche Geschlechtsumwandlung bei adulten Teleosteern. Verhandl. Deut. Zool. Ges. Bonn/Rhein pp. 259-262. Reinboth, R. ( 1962 ) . Morphologische und funktionelle Zweiggeschlechtlichkeit bei marinen Teleostieren ( Serranidae, Sparidae, Centracanthidae, Labridae ) . Zool. Jahrb. Abt. Allgem. 2001.Physiol. Tierre 59, 405-480. Reinboth, R. ( 1983). Natiirliche Geschlechtswechsel bei Sacura murgaritacea ( Hilgendorf) (Serranidae). Annotationes Zool. lapon. 36, 173-178. Reinboth, R. (1965). Sex reversal in the black sea bass Centroprktes striatus. Anat. Rec. 151, 403. Reinboth, R. ( 1967). Protogynie bei Chelidoperca hilrrndinacea (Cuv. et Val.) (Serranidae ) , -Ein Diskussionsbeitrag zur Stammesgeschichte amphisexueller Fische. Annotationes Zool. lapon. 40, 181-193. Rijavec, L., and Zupanovic, S. (1965). A contribution to the knowledge of biology of Pagellus emhrinus L. in the middle Adriatic. Rappt. Proces-Verbaux Reunions, Monaco 18, 195-200. Roede, M. J. (1965). Intersexuality of Labrids. “Intersexuality in Fishes,” Abstr. Papers, Conf. Cape Haze Marine Lab., Sarasota, Florida. Rubin, A. A,, and Gordon, M. (1953). Effects of alpha-estradiol benzoate and methyl testosterone upon the platyfish, Xiphophoms maculatus skeleton. Proc. SOC. Exptl. Biol. Med. 83, 648848. Rudolf, H. (1931). Wieder ein Herringszwitter. 2001.Anz. 95, 37-41. Rust, W. ( 1941 ) . Genetische Untersuchungen iiber die Geschlichtsbestimmungtypen bei Zahnkarpfen unter besonderer Beriicksichtigung von Artkreuzungen mit Platypoecilus variatus. Z. lnduktive Abstammungs Vererbungslehre 79, 338395. Sacks, W. B. ( 1955). Uber Xiphophorus helleri Heckel. Deut. Aqua. Tern. Zeit., pp. 309310. Samokhvalova, G. V. (1933). Correlation in the development of the secondary sexual characters and the sex glands in Lebistes reticulatus. Trans. Dyn. Develop. 7, 65-76. Schreiner, K. E. (1904). Uber das Generations-organ von Myrine glutinosa ( L . ) . 2001.Anz. 24, 91-104. Schreiner, K. E. (1955). Studies on the gonad of Myxine glutinosa L. Univ. Bergen Arbok, Natuwitenskap. Rekke 8, 1-37. Schultz, R. J. ( 1961). Reproductive mechanisms of unisexual and bisexual strains of the viviparous fish Poeciliopsis. Evolution 15, 302-325. Schultz, R. J. (1966). Hybridization experiments with an all-female fish of the genus Poeciliopsis. Biol. Bull. 130, 415-429. Schultz, R. J. (1967). Gynogenesis and triploidy in the viviparous fish Poeciliopsis. Science 157, 1564-1567. Schwier, H. ( 1939). Geschlechtsbestimmung und-differenzierung bei Macropodus opercularis, concolor, Chinensis und deren Artbastarden. Z. Induktive Abstammungs Vererbungslehre 77, 291435. Sengiin, A. ( 1941). Ein Beitrag zur Geschlechtsbestimmung bei Platypoectlus

172

TOKI-0 YAMAMOTO

maculatw und Xiphophorus helleri. Istanbul Uniu. Fen. Fak. Mecmuasi B6, 3348. Smith. C. L. (1959). Hermaphroditism in some serranid fishes from Bermuda. Papers Mich. Acad. Sci. 44, 111-119. Smith, C. L. (1965). The patterns of sexuality and the classification of serranid fishes, Am. Museum Novitates 2207, 1-20. Smith, C. L. ( 1967). Contribution to a theory of hermaphroditism. J . Theoret. Biol. 17, 76-90. Solomon, S., Wiese, R, V., and Lieberman, S . (1956). The in uitro synthesis of 17aHydroxyprogesterone and A'-Androstene-3, 17-dione from progesterone by bovine ovarian tissue. J. Am. Chem. SOC. 78, 5453-5454. Sordi, M. (1962). Ermafroditismo proteroginico in Labrus turdus L. e in L . merula. Monit. Zool. Ital. 69-84. Spurway, H. (1953). Spontaneous parthenogenesis in a fish. Nature 171, 437. Spurway, H. ( 1957). Hermaphroditism with self-fertilization, and the monthly extrusion of unfertilized eggs, in the viviparous fish Lebistes reticulatw. Nature 18Q, 1248-1251. Stolk, A. ( 1958). Pathological parthenogenesis in viviparous toothcarps. Nature 181, 1660. S p k i , S. ( 1876). Ergebnisse von Untersuchungen der Geschlechtsorgane von Knochedschen ( D e piscium osseortm organis genitalibw). Kosmos ( L w o w ) 1, 41-55 Takeuchi, K. (1967). Large tooth formation in female medaka, Oryzias Zatipes, given methyl testosterone. J. Dental Res. 46, 750. Tavolga, M. C. (1949). Differential effects of estradiol and pregneninolone on Platypoecilw maculatus. Zoologica 34, 215-238. Taylor, A. B. (1948). Experimental sex reversal in the red swordtail hybrid Xiphophow-Platypoecilzrs. Trans. Am. Microscop. SOC. 67, 15S164. Tozawa, T. (1923). Studies on the pearl organ of the goldfish. Annotationes Zool. lapon. 10, 253-263. Tozawa, T. (1929). Experiments in the development of the nuptial coloration and pearl organs of the Japanese bitterling. Folia Anat. Japon. 7 , 407417. Turner, C. L. (1942). Morphogenesis of the gonopodial suspensorium in Gambusia afinis and the induction of male suspensorial characters in the female by androgenic hormone. J . Exptl. 2001.91, 167-193. Turner, C. L. (1947). The rate of morphogenesis and regeneration of the gonopodium in normal and castrated males of Gambusia afinis. J. Erptl. Zool. 106, 125-143. Turner, C. L. (1960). The effects of steroid hormones on the development of some secondary sexual characters in cyprinodont fishes. Trans. Am. Microscop. SOC. 79, 320-333. Vandini, R. Z. (1965). I1 problema dei maschi primari di Cork Julis. Monit. 2001. Ital. 83, 102-111. van Oordt, G. J. (1923). Secondary sex characters and testis of the ten-spined stickleback (Gasterosteus pungitius). Koninkl. Ned. Akad. Wetenschap., Proc. Sect. Sci. 26, 309314. van Oordt, G. J. (1924). Die Veriinderung des Hodens wahrend des Auftretens der sekundaren Geschlechtsmerkmale bei Fischen, Gasterostew pungitius L. Arch. Mikroskop. Anat. Entw4cklungsmech. la 379405.

3.

SEX DIFFERENTIATION

173

van Oordt, G. J, (1925). The relation between the development of the secondary sex characters and the structure of the testis in the teleost, Xiphophorus helleri Heckel. Brit. J . Exptl. Biol. 3, 4 3 4 9 . van Oordt, G. J. (1928). The duration of life of the spermatozoa in the fertilized female of Xiphophorus helleri. Tiidschr. Ned. Dierk. Ver. 3, 77-81. van Oordt, G. J. (1929). Zur mikroskopischen Anatomie des Ovariotestes von Serranus und Sargus (Teleostei). Z . Mikroskop.-Anat. Forsch. 19, 1-17. van Oordt, G. J,, and van der Mass, C. J. J. (1927). Castration and implantation of gonads in Xiphophorus Helleri Hechel. Koninkl. Ned. Acad. Wetenschap., Proc. 29, 1172-1175. Vivian, J. (1950). Masculinisation des femelles de Xiphophores par action des r ons X sur la gonade. Compt. Rend. Acad. Sci., Paris 231, 1166-1168. Vivian, J. ( 1952a). Rale de l’hypophyse dans le determinism de l’involution ovarienne et de l’inversion sexuelle chez le Xiphophores. J. Physiol. (Paris) 44, 349-351. Vivian, J. ( 195213). Sur l’existence de charactBres sexuels secondaires femelles et sur le determinisme de leur differenciation chez Xiph. hell. Heckel. Compt. Rend. Acad. Sci., Paris 235, 1697-1699. Williamson, H. C. (1910). Report on the reproductive organs of Sparus centrodontus Delaroche; Sparus canthalus L.; Sebastes marinus ( L . ) and Sebastes dactylopterus (Delaroche); and on the ripe eggs and larvae of Sparus controdontus and Sebastes marinus. Rept. Fishery Board. Scotland Sci. Invest. 1, 1-35. Winge, 0. ( 1922a). A peculiar mode of inheritance and its cytological explanation. J. Genet. 12, 137-144. Winge, 0. ( 1922b). One-sided masculine and sex-linked inheritance in Lebistes reticulatus. J. Genet. 12, 145162. Winge, 0. (1927). The location of eighteen genes in Lebistes reticulatus. J. Genet. 1 4 1-43. Winge, 0. (1930). On the occurrence of XX-males in Lebistes with some remarks on Aida’s so-called “non-disjunctional” males in Aplocheilus. J. Genet. 23, 69-76. Winge, 0.(1934). The experimental alteration of sex chromosomes into autosomes and vice versa, as illustrated by Lebistes. Compt. Rend. Trav. Lab. Carlsberg, Ser. Physiol. 21, 1-49. Winge, U., and Ditlevsen, E. (1947). Colour inheritance and sex determination in Lebistes. Heredity 1, 65-83. Winge, O., and Ditlevsen, E. (1948). Colour inheritance and sex determination in Lebistes. Compt. Rend. Trav. Lab. Carlsberg, Ser. Physiol. 24, 227-248. Witschi, E. ( 1914a ). Experimentelle Untersuchungen uber die Entwicklungsgeschichte der Keimdriisen von Rana temporaria. Arch. Mikroskop. Anat. 85, 9-113. Witschi, E. ( 1914b). Studien uber die Geschlechtsbestimmung bei Froschen. Arch. Mikroskop. Anat. 86, 1 5 0 . Witschi, E. ( 1429). Bestimmung und Vererbung des Geschlechts bei Tieren. Handbuch Vererbungswiss. 2, 1-115. Witschi, E. (1930). Studies on sex differentiation and sex determination in Amphibians. IV. The geographical distribution of the sex races of the Eurupean grass frog ( R a w temporaria L.).J . Ex&. 2001.!56, 144-165. Witschi, E. ( 1931). Studies on sex differentiation and sex determination in amphibians. V. Range of the cortex-medulla antagonism in parabiotic twins of Ranidae and Hyalidae. J. Ex@. ZOO^. 58, 113-145. Witschi, E. (1939). Modification of development of sex. In “Sex and Internal

174

TOKI-0 YAMAMOTO

Secretions” ( E . Allen, ed.), 2nd ed., pp. 145-226. Williams & Wilkins, Baltimore, Maryland. Witschi, E. (1942). Hormonal regulation of development in lower vertebrates. Cold Spring Harbor Symp. Quant. Biol. 10, 14S151. Witschi, E. (1950). GnBtique et physiologie de la diffhenciation du sexe. Arch. Anat. Microscop. Morphol. Exptl. 39, 215-246. Witschi, E. ( 1955). On morphogenic capacities of the estrogens. J. Clin. Endocrinol. Metab. 15, 647-652. Witschi, E. (1957). The inductor theory of sex differentiation. J. Fac. Sci., Hokkaido Uniu., Ser. VI 13, 428-439. Witschi, E., and Crown, E. N. (1937). Hormones and sex determination in fishes and in frogs. Anat. Record 70, 121-122. Wolf, L. E. (1931). The history of the germ cells in the viviparous teleust Platypoecilus maculatus. ]. Morphol. 52, 115-154. Wotiz, H. H., Botticelli, C. R., Hisaw, F. L., Jr., and Olsen, A. C. (1960). Estradiol17p, estrone and progesterone in the ovaries of dogfish (Squalus suckleyi). Proc. Natl. Acad. Sci. U.S. 46, 58k583. Wiinder, W. ( 1931 ). Experimentelle Eneugung des Hochzeitskleides beim Bitterling ( Rhodeus amarw ) durch Einpritzung von Hormonen. Z. Vegleich. Physiol. 14, 676-708. Wurmbach, H. ( 1951). Geschlechtsumkehr bei Weibschen von Lebistes reticulutus bei Befall mit lchthyophonous hoferi Plehn-Mulsow. Arch. Entwicklungsmech. Organ. 145, 109-124. Yamamoto, T. (1953). Artificially induced sex reversal in genotypic males of the medaka (Oryzias latipes). 1. Exptl. Zool. 123, 571594. Yamamoto, T. (1955). Progeny of artificially induced sex reversals of male genotype (XY) in the medaka (Oryzias latipes) with special reference to YY-male. Genetics 40, 4OtM19. Yamamoto, T. ( 1958). Artificial induction of functional sex-reversal in genotypic females of the medaka (Oryzias latipes). J. Exptl. 2002. 137, 227-262. Yamamoto, T. (1959a). A further study of induction of functional sex reversal in genotypic males of the medaka (Oryzias latipes) and progenies of sex reversals. Genetics 44, 739-757. Yamamoto, T. (195W). The effects of estrone dosage level upon the percentage of sex-reversals in genetic male (XY) of the medaka (Oryzias kztipes). J . Exptl. Zool. 141, 13S.154. Yamamoto, T. (1961). Progenies of induced sex-reversal females mated with induced sex-reversal males in the medaka, Oryzias latipes. J . Elrptl. Zool. 146, 163-179. Yamamoto, T. (1962). Horrnonic factors affecting gonadal sex dserentiation in fish. Gen. Comp. Endocrinol. Suppl. 1, 341-345. Yamamoto, T. (1963). Induction of reversal in sex differentiation of YY zygotes in the medaka, Oryzias latipes. Genetics 48, 293-306. Yamamoto, T. (1964a). The problem of viability of YY zygotes in the medaka, Oryzias latipes. Genetics 510, 45-58. Yamamoto, T. (1964b). Linkage map of sex chromosomes in the medaka, Oryzias latipes. Genetics 50, 59-84. Yamamoto, T. (1965). Estriol-induced XY females of the medaka (Oryzlas latipes) and their progenies. Gen. Comp. Endocrinol. 5, 527433. Yamamoto, T. (1967). Estrone-induced white YY females and mass production of white YY males in the medaka, Oryzias latipes. Genetics 55, 329-338.

3.

SEX DIFFERENTIATION

175

Yamamoto, T. ( 1968). Effects of 17~~-hydroxyprogesterone and and,rostenedione upon sex differentiation in the medaka, Oryzias Zatipes. Gen. Comp. Endominol. 10, 8-13. Yamamoto, T., and Kajishima, T. (1969). Sex-hormonic induction of reversal of sex differentiation in the goldfish and evidence for its male heterogamety. J. Exptl. Zool. 168, 215-222. Yamamoto, T., and Matsuda, N. (1963). Effects of estradiol, stilbestrol and some alkyl-mrbonyl androstanes upon sex differentiation in the medaka, Oqzias Zatipes. Gem Comp. EndocrinoZ. 3, 101-110. Yamamoto, T., and Suzuki, H. (1955). The manifestation of the urinogenital papillae of the medaka (Oryzim Zaripes) by sex hormones. Embryologia ( N a g o y a ) 2, 133-144. Yamamoto, T., Takeuchi, K., and Takai, M. (1968). Male-inducing action of androsterone and testosterone propionate upon XX zygotes in the medaka, Oryzias latipes. Embryologia (Nagoya) 10, 142151. Zahl, A. P., and Davis, D. D. (1932). Effects of gonadectomy on the secondary sexual characters in the ganoid fish, Amia calua Linnaeus. 1. Exptl. 2001.63, 219307. Zander, C. D. (1965). Die Geschlechtsbestimmung bei Xiphophorus montezumae cortezi Rosen (Pisces). Z . Vererbungslehre 96, 12S141. Zei, M. (1949). Typical sex-reversal in teleosts. Proc. Zool. SOC. London 119, 917921.

This Page Intentionally Left Blank

DEVELOPMENT: EGGS AND LARVAE J . H . s. BLAXTER

. . . . . . . . . . . . . . . .

1. Introduction I1. The Parental Contribution . . . A . Conditions for Incubation . . . B Fecundity and Egg Size . . . 111. Events in Development . . . . A. Fertilization . . . . . . B . Incubation (Fertilization to Hatching) C Hatching . . . . . . D The Larva E . Metamorphosis . . . . . F.Timing . . . . . . . IV. Metabolism and Growth . . . . A . Rate of Development . . . . B . Yolk Utilization . . . . . C. Viviparity D . Biochemical Aspects . . . . E Respiration . . . . . . F. Growth . . . . . . . G. Endocrines. Growth and Metamorphosis V. Feeding, Digestion, and Starvation . . VI . Sense Organs . . . . . . A. Vision . . . . . . . B. Neuromast Organs . . . . VII . Activity and Distribution . . . . A . Phototaxis and Activity . . . B. Vertical Distribution . . . . C . Buoyancy and Pressure . . . D. Locomotion and Schooling . . . E . Searching Ability . . . . . VIII . Mortality. Tolerance. and Optima . . IX . Meristic Characters . . . . . X . Rearing and Farming . . . . A Techniques . . . . . . B. Sensory Deprivation . . . . XI . Conclusions . . . . . . Acknowledgments . . . . . . References . . . . . . . . 177

.

. .

.

.

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. .

. . .

.

. .

. . .

.

. .

. . .

.

. .

. . .

.

. .

. . .

.

178 178 178 180 184 184 187 189 190 190 191 191 191 194 198 198 203 209 212 213 220 220 221 221 221 223 224 226 229 229 235 238 238 240 241 241 241

178

J. H. S . BLAXTER

I. INTRODUCTION

Fish eggs and larvae provide a relatively untapped source of biological material, increased by the recent improvements in techniques for rearing marine species. Apart from their intrinsic interest, experimentally based information on these early stages is required for further progress in the advancing fields of fish culture and fisheries research. General textbooks on ichthyology such as those of Lagler et al. (196Z), Nikolsky (1963), Norman (1963), and Marshall (1965) and on reproduction in fish by Breder and Rosen (1966) provide both general and some detailed information. Identification of eggs and larvae, apart from specialist papers, is possible from publications of Ehrenbaum ( 1909), DAncona et al. (1931-1933), through the current series of plankton sheets issued by the International Council for the Exploration of the Sea, and with the help of the extensive bibliographies by Dean (1916) and Mansueti ( 1954). Most species of fish pass through a larval stage before assuming the adult form at metamorphosis. Sometimes the newly hatched fish is called a “prolarva” (or alevin in salmonids) until the yolk is resorbed, and then a “postlarva” (or fry), The term “larva” is used here for all stages to metamorphosis in marine fish, although alevin and fry may be used when referring to salmonids or other freshwater groups.

11. THE PARENTAL CONTRIBUTION

Apart from the more obvious genetical effects on differentiation, rate of development, body form, size, and behavior, the parents, and especially the female, have an important influence on the viability of the offspring both on a species and individual level, in terms of ( a ) the conditions for incubation, ( b ) fecundity, and ( c ) egg size. A. Conditions for Incubation Differences of spawning season and time and of spawning sites and substrate mean that incubation can take place in a great variety of conditions which influence the early development and physiology of the off spring.

4. DEVFLOPMENT:

EGGS A N D LARVAE

179

1. EGGSSINGLE, WITH No PARENTAL CARE ( a ) Buoyant, planktonic-most marine fish, e.g., gadids, clupeids, flathh, and deep-sea fish. ( b ) Nonbuoyant, loose or attached to substrate-a common freshwater characteristic, e.g., cyprinids, pike Esox, or in littoral species, e.g., blenny Bbnnius, bullheads Cottus, sand eels Ammodytes; also found in some marine species, e.g., herring Clupea harengus, capelin Mallotus uillosus, cathh Anarhichas, and American flounder Pseudopburonectes americunus. Tendrils for attachment are found in many oviparous elasmobranchs, in the hagfish Myxine, smelt Osmerus, saury Scomberesox, and flying fish Exocoetus. ( c ) Nonbuoyant, buried in sand or gravel-many salmonids, grunion Leuresthes tenuis, and lamprey Petromyzon; in peat or mud Aphyosemion and Cynolebias where the eggs undergo diapause during the dry season. 2. EGGSSINGLE,SPECIAL ENVIRONMENTS The bitterling Rhodeus amarus lays eggs in the gills of the freshwater mussel and the lumpsucker Cureproctus under the carapace of the Kamchatka crab. 3. EGGSSINGLE,WITH PARENTAL CARE

( a ) No nest, but eggs protected-found in many littoral forms, e.g., the bullheads Cottidae, blennies Blenniidae and gobies Gobiidae. ( b ) Nests, often with parental protection and ventilation-also found in littoral species, e.g., blenny Ictalurus, sticklebacks Gasterosteus, and in other freshwater species such as sunfish Centrarcidae, bowfin Amia, lungfish Protopterus and Lepidosiren, and in the Cichlidae. Bubble nests giving good aeration are found in tropical or swamp species, e.g., Siamese fighting fish Betta splendens. ( c ) Parents carrying eggs-sea horses Hippocampus and pipefish Syngnathus have brood pouches and the sheat fish Platystacus a specially modified area of “spongy” skin. Marine catfish Ariidae, cardinal fish Apogonidae and Tilapia are mouth brooders, and Tachysurms incubates the eggs intestinally. ( d ) Ovovipiparity and viviparity (see Section IV, C)-elasmobranchs include picked dogfish Acanthius, smooth hound Mustelus vulgaris, electric ray Torpedo, stingray Trygon, and the nurse hound Mustelus laevis. Teleosts include redfish Sebastes, Heterandria, Anubleps, poeciliids such as Xiphophorus and half beaks Hemirhumpw.

180

J. H. S. BLAXTER

4. EGGSMASSED

Angler fish Lophius and yellow perch Perca flavescens have massed but unprotected eggs; in the lumpsucker Cyclopterus and butterfish Blennius pholb the eggs are protected by the male.

B. Fecundity and Egg Size In higher latitudes the spawning season is often short and the eggs are liberated over a brief period of perhaps hours (clupeids) or over periods of some days, probably at certain times of the day or night (flatfish and gadids). Where the seasons are less marked spawning may occur over a much longer period or be intermittent throughout the year, especially where the time between generations is only a matter of weeks or months. Fecundity may be considered as the number of eggs produced in one year by a female although this may be very difficult to determine where spawning is protracted. Some examples of fecundity, egg size, and length at hatching are given in Table I. In general, fecundity is high where the eggs are liberated into open marine waters; it is lower in freshwater species and where there is parental care. There is also a strong tendency for fecundity and egg size to be inversely related. Apart from enormous interspecific differences, there are also considerable variations of fecundity within a species. Many authors have found that fecundity increases with length, weight, or age (see Fig. 1)) the relationship usually being of the form F = aLb, where F is fecundity, L is length, and a and b are constants. Year-to-year differences resulting almost certainly from environmental effects are also well established. For instance, sea temperature was correlated by Rounsefell (1957) with the fecundity of pink salmon, Oncorhynchus gorbuscha, higher temperatures apparently resulting in lower fecundities. (Here the effect of temperature on growth is a complication.) Bagenal (1966) reported density-dependent factors operating in Scottish flatfish, high densities being correlated with low fecundity. Anokhina (1960) found that fecundity in Baltic herring could be related to feeding conditions, high fat content of the female being reIated to high fecundity. Experimental studies by D. P. Scott (1961) indicated in rainbow trout, Salmo gairdneri, that an insufficient diet caused a reduction in egg number: in the guppy Lebistes fewer offspring were also produced when the females were kept on short rations (Hester, 1963). Extensions of this type of work are badly needed. Differences within a species resulting from latitude, area, race, or

Table I Fecundity (Eggs/Female/Yemr), Egg Diameter, and Length a t Hatching Species Molva molva Gadus morhua Melanogrammus aegleJinus Pleuronectes platessa Solea solea Swmber swmbrus Clupea harengus Clupeonella delicatula Salmo salar Osmerus eperlanus Acipenser sturio Cyprinus carpio Aeanthurus triostegus Oryzias latipes Scyliorhinus caniculus Lebistes reticulatush Zoarces viviparusa Sebastes viviparusb Mustelus mustelusb Squa1u.s amnthiasa a

a

Common name Ling Cod Haddock Plaice

Sole Mackerel Herring Kilka Salmon Smelt Sturgeon Carp Convict surgeon fish Medaka Spotted dogfish GUPPY Blenny Redfish Smooth hound Spur dogfish

Fecundity

20-30 X 10' 20-90 X 106 12 X 103-30 X 16 X 103-35 X 15 x 104 35-45 x 104 50 x 102-20 x 10-60x 103 103-104 50 X lO"50 X 80 X lo4-24X 18-53 X lo4 40 x 1 0 3 ~ 20-40" 2-20 10-50" 20-300 12-30 x 103 10-30 2-7

Number per spawning, which may be repeated often in one year. Viviparous or ovoviviparous.

Diameter (mm)

lo6 lo4

104

1.0-1.1 1.1-1.6 1.2-1.7 1.7-2.2 1.0-1.5 1.0-1.4 0.9-1.7 1.0 5.0-6.0

lo3

lo6

0.9 ?

0.9-1.6 0.7 1 .O-1.3 65.0(length) -

24-32

Length at hatching (mm)

3.0-3.5 4.0 4.O-5.0 6.0-7.0 3.2-3.7 3.0-4.0 5.0-8.0 1.3-1.8 15.0-25.0 4.0-6.0 9.0 4.8-6.2

1.7 4.5-5.0 100.0 6.0-10.0 35.0-40.0 5.0-8.0 250.0 240-310

182

J. H. S . BLAXTER

I06

-

5 1 0

0

x 0" 2 .-

-U C

3 U

u

' Lenqth,

loq scale (crn)

Fig. 1. The relationship between fecundity and length within a species. 1. Acipenser stellatus (Nikolsky, 1963); 2. Salmo salar (Pope et al., 1961); 3. Cyprinw carpio (Nikolsky, 1963); 4 . Pleuronectes platessa (Clyde) (Bagenal, 1966); 5. Clupea harengus (northern North Sea) ( Baxter, 1959); 6. Mekznogrammus aeglefinus (see Parrish, 1956); 7. Osmerus eperlanus ( Lillelund, 1961); 8. Salvelinus fontinalis (Vladykov, 1956); 9. Sardinops caerulea ( MacGregor, 1957); and 10. Sebastes marinus (Faroe Island) (Raitt and Hall, 1967).

season are no doubt interconnected. Considerable differences of this type were reported for plaice, PZeuronectes platessu ( Bagenal, 19fH), for herring (Baxter, 1959; Kandler and Dutt, 1958), for species of Oncorhynchus (Rounsefell, 1957), and for Sulmo sulur (Pope et ul., 1961). An interesting characteristic of certain species is a difFerence in fecundity of the left and right ovary. The left gonad contains more eggs in Oncorhynchus (Rounsefell, 1957), Salmo (Pope et al., 1961), and smelt Osmerus ( Lillelund, 1961) . The significance of this is not clear. Egg size varies at the interspecific level (see Table I ) , with larger eggs being especially associated with freshwater species like the sal-

4. DEVELOPMENT:

183

EGGS AND LARVAE

monids or where fecundity is very low, as in many elasmobranchs. At the intraspecific level, differences of egg size as a result of area or river were noted by Rounsefell (1957) in Oncorliynclzus, Salmo, Cristivomer, and Salvelinus species. In Salmo salar (Pope et al., 1961) there are differences in egg diameter related to length, fecundity, and river. In Tilapia egg weight may increase 2-4 times or even more depending on the size of the female (Peters, 1963); in the flounder, Platichthys flesus, large females, or females from low salinities, have larger eggs (Solemdal, 1967). Larger eggs are also found in larger females of the spur dogfish Acantliias ( Templeman, 1944). There are differences in average dry weight of the order of four times among the various races of herring (Hempel and Blaxter, 1967) and small differences between very young first spawners and repeat spawners. In two darter Etheostoma species with long spawning seasons, the egg diameter tends to be greater in the cooler winter months (Hubbs et al., 1968). The connection between egg size and fecundity in related species may be correlated with the conditions for incubation. For example, Tilapia tholloni, a substrate brooder, has 5003000 eggs depending on length. On the other hand, T . mossambica and T . macrocephala, which are mouth brooders, have less than 500 eggs that are considerably greater in weight (Peters, 1963). Garnaud (quoted by Smith, 1957) reported two species of Apogon, one, A. imberbis, with an egg diameter of 0.5 mm and fecundity of 22,000, and the other, A. conspersus, with an egg diameter of 4.5 mm and fecundity of 150. The relationship between egg size and egg weight within a species is shown for the different races of herring in Fig. 2. The winter-spring spawners have a low fecundity and large eggs,

I .

\

'.._ --

I

/

y'.

! 0.15

' F ' M ' A ' M ' J'J ' A ' S ' O ' N ' D

Spawning month

Fig. 2. Fecundity and average egg dry weight in different races of herring (data from Baxter, 1959; Hempel and Blaxter, 1967; Parrish and Saville, 1965; Kandler and Dutt, 1958).

184

J. H.

S.

BLAXTER

an adaptation to poor food for the young, but a low predator population. In summer-autumn conditions fecundity is high and egg size low, presumably an adaptation to good food supplies and many predators. Intraspecific differences in fecundity and egg size deserve further study in terms of a link between ecological conditions and the physiology of maturation of the ovary. A general pattern seems to emerge of marine fish with many, small buoyant eggs, a short incubation period, and vulnerable larvae. Freshwater fish have larger demersal eggs, a long incubation period and larger, less vulnerable larvae, while the littoral forms exhibit protective devices to prevent losses in this particularly difficult environment. The initial conditions of development determined by the genotype of the parent and reproductive behavior must have a considerable influence on the viability of the young and its physiology, in that environmental conditions such as temperature affect the speed of development, salinity presents problems of osmoregulation, and oxygen must be obtained for respiration. The egg is susceptible and yet cannot make defensive responses to mechanical shock, drift by current, toxins, light, and predators. The larva should live in conditions where food can be obtained and protective behavioral devices can be practiced.

111. EVENTS IN DEVELOPMENT

A. Fertilization The physiology of fertilization in fish, with special reference to the extensive Japanese work, has been fully reviewed by Yamamoto (1961) from which much of the present account is taken. There is some evidence for the action of gamones in Lampetra and in teleosts; these activate the sperm and serve as chemical attractants toward the egg, while other gamones are known to paralyze or agglutinate sperm. In the bitterling species Acheilogwthus and Rhodeus sperm aggregation and activity have been noted in the micropyle region of the chorion. The chorion or egg case is relatively tough with a funnel-shaped micropyle at the animal pole. Within the chorion a plasma or vitelline membrane [also called a pellicle or surface gel layer (Trinkaus, 1951)] surrounds the yolk and cytoplasm (ovoplasm) of the egg. Fertilization, which requires the presence of small concentrations of Ca or Mg ions, is normally monospermic in teleosts, the micropyle being too narrow to

4.

DEVELOPMENT: EGGS AND LARVAE

185

allow more than one sperm to pass at a time. The ovoplasm and chorion separate as the egg is activated by the sperm and a plug forms in the micropyle, further sperm being rejected. Where polyspermy occurs, as in some elasmobranchs, only one sperm fuses with the egg nucleus, the rest probably being resorbed and used as nutrient. Removal of the chorion permits polyspenny in teleost eggs; it seems that polyspermy is usually prevented by rapid changes at the micropyle, rather than over the egg cortex. In salmonids water activation (not to be confused with activation by sperm) takes place (e.g., see Prescott, 1955). When the egg is released into hypotonic solutions like river water the vitelline membrane becomes opaque and there are changes in its permeability. If sperm are not immediately avaiIable these changes may also affect f ertilizability. Following fertilization, the prominent alveoli in the egg cortex of salmonids, acipenserids, and lampreys disappear. In the medaka, Oryxius Zutipes, these alveoli break down progressively from the animal pole. The separation of the cortex from the chorion leads to the appearance of the perivitelline space. The chorion is permeable to water and small molecules, but larger molecules of a colloidal nature are retained in the perivitelline fluid. In Oryzias and Lampetru these colloids maintain an osmotically based tension within the chorion. It seems likely that the colloid is derived from polysaccharide material in the cortical alveoli so that the formation of the perivitelline space is partly owing to a decrease in volume of the ovoplasm, as the alveoli release colloid, and partly due to an osmotic distension of the chorion. According to Ginsburg ( 19Sl), polyspermy is blocked in sturgeon and trout eggs by the discharge of the cortical alveoli in the micropylar region. Chemicals, such as urethane, which cause polyspermy apparently retard the secretions of the alveoli, while removal of the perivitelline fluid in trout eggs allows the penetration of many sperm. The chorion also hardens (see Section VIII) thus protecting the embryo in the early, more vulnerable stages. qrobably the inner layer of glycoprotein is mainly responsible for this and it has been suggested that the alveolar colloid, Ca ions, phospholipids, or hardening enzymes also play a part. In salmonids, Zotin (1958) reported hardening of the chorion because of an enzyme in the perivitelline fluid; Ca ions affect the enzyme rather than the chorion itself. Ohtsuka (1960) considered that a phospholipid was liberated from the cortex (not from the alveoli) in Oryxias eggs. In Fundulus the chorion hardened with oxidizing agents but not with reducing agents. The soft chorion appeared to be impregnated with protein containing SH groups. Hardening resulted from oxidation of SH to SS groups by means of aldehydes produced from

186

J. H. S. BLAXTER

polysaccharides with a-glycol groups. Zotin distinguished between the initial enzyme action and subsequent hardening processes which last much longer and where the enzyme is no longer functioning. Thus the initial enzyme reaction is blocked when Ca ions are bound by citrate or oxalate or by the use of NaCl or other chlorides. Later hardening is not susceptible to many of these factors. The eggs of Oryzias, Gasterosteus, and Lampetra can be activated by pricking. Other activating agents are surface-active chemicals and lipid solvents (perhaps emulsifying the cytoplasm at the animal pole) and thermal shock, electric fields, ultraviolet light, and high frequency vibrations. This type of artificial parthenogenesis usually leads to irregular cleavage, but stringent precautions are needed to prevent contamination with sperm in such experiments. Of interest is the ability of eggs and sperm to retain their fertilizability after leaving the parent. According to Yamamoto (1961) fish eggs lose this capacity after a very short time, but this can be increased if they are retained in isotonic Ringer's solution. While this is true of some freshwater eggs, presumably as a result of water activation, in seawater the capacity for fertilization is retained for much longer-certainly for hours in the herring. Nikolsky (1963) states that sperm motility is short lived where spawning takes place in fast-flowing water, for example, 1015 sec in Oncorhynchus. In slower flows, sturgeon sperm is motile for 230-290 sec, and in the sea herring sperm may be motile for hours or days. Observations of short-lived activity are difficult to make; thus, some of these figures must be considered approximate. Storage of gametes is a useful technique in fish farming to allow controlled fertilizations in the laboratory and to obviate the need for transporting eggs in the susceptible pregastrulation stages. Salmonid gametes are best stored dry below 5°C (e.g., Barrett, 1951, Withler and Morley, 1968); those of the herring may be held in buffered egg-yolk diluents, but are also best kept dry at about 4°C (Blaxter, 1955). While this permits storage for, at most, a few days long-term techniques are also possible. Herring sperm, but not eggs, were kept for some months in a diluent consisting of 12.5% glycerol in 3% salt solution (diluted seawater) at -79°C and crosses made successfully between spring and autumn spawning races (Blaxter, 1953, 1955; Hempel and Blaxter, 1961). Sneed and Clemens (1956) also succeeded in holding out-of-season carp sperm immotile for 30 days at 3'5°C in frog Ringer. Carp sperm could also be frozen and stored at -73°C in isotonic Ringer containing &12%glycerol with some survival when thawed after 60 hr of storage. More recently, Truscott et al. (1968) have shown that salmon sperm can be stored for 1-2 months at temperatures of -3" to -4.5"C using diluents such as 5%

4.

DEVELOPMENT: EGGS AND LARVAE

187

ethylene glycol or 5% dimethyl sulfoxide, retaining 7040%fertility. Horton et al. (1967) obtained alevins from salmon eggs fertilized with sperm frozen in liquid nitrogen with dimethyl sulfoxide as a protecting agent, but the fertility rate was low. Hoyle and Idler (1968) have also obtained fertile salmon sperm after storage in liquid nitrogen using ethylene glycol with added lactose or serine. Slow freezing produced the best results. Mounib et al. (1968) succeeded in storing cod sperm for up to 60 days using 17-24% glycerol and temperatures of -79” and -196°C. Initial experiments suggest that faster cooling rates gave the better results with cod sperm.

B. Incubation ( Fertilization to Hatching ) The progress of cleavage, formation of layers, and morphogenesis have been described in a number of standard textbooks such as Rudnick ( 1955), Waddington ( 1956) and Smith ( 1957), with Oppenheimer (1947) and Devillers (1961) stressing structural changes from the viewpoint of experimental embryology. More detailed information is limited mainly to freshwater species like the trout Salmo trutta, killifish Fundulus, medaka Oryzias, goldfish Carassius, and to the dogfish. New (1966) gives information on the problems of culturing Fundutus, Oryzias, and Salmo eggs for the purpose of experimental embryology. Most fish eggs are round, although in the anchovy Engraulis and bitterling Rhodeus they are ovoid, and in certain gobies pear-shaped. Most species have telolecithal eggs with yolk more concentrated at the vegetative pole; some marine species have oil globules of varying size and number (see Simpson, 1956). Before fertilization the cytoplasm may be mixed with or separate from the yolk. The extent to which polarity exists at this stage has not been described in many species. After fertilization (as the cortical alveoli release colloid, the perivitelline space develops and the chorion hardens) cytoplasm migrates to the future blastodermal region, most of it arriving by the first cleavage. The remaining cytoplasm forms a “halo” or periblast under the blastoderm (see Fig. 3 ) . In lampreys cleavage is holoblastic but with the formation of microand macro-meres. In hagfish, elasmobranchs, and teleosts it is meroblastic. Other groups such as bowfin Amia, gar Lepidosteus, and sturgeon Acipenser have intermediate features. With some variation, the meroblastic group possess a blastodermal cap of cells at the animal pole after the initial stages of cleavage. In Fundulus the surface gel layer overlying the blastoderm, being sticky on its inner surface, serves to hold the outer blastomeres together (Trinkaus, 1951). Usually cleavage is not complete,

188

J. H. S. BLAXI’ER

Fig. 3. ( A ) Transverse section of early blastula showing adhesion of blastomeres to the surface gel layer and attachment of blastoderm to the periblast at the periphery only (Fundulus, after Trinkaus, 1951 ). ( B ) Sagittal section of later blastula showing gastrulation; epiboly shown by arrows. ( C ) and ( D ) Surface view of eggs in later stages of gastrulation. Key: b, blastoderm; bp, blastopore; dl, dorsal lip of blastopore; gr, germ ring; n, notochord; np, neural plate; p, periblast; sgc, subgerminal cavity; sgl, surface gel layer; and y, yolk.

and in the deeper layers the periblast becomes syncytial and is involved in mobilizing the yolk reserves. There are substantial cohesive forces between the developing blastomeres and the surrounding periblast which are important in the subsequent morphogenetic movements. The blastoderm now commences to thin and overgrow the yolk (epiboly) and at the same time invaginate at its periphery (Fig. 3 ) . The periblast seems closely connected with this spreading tendency of the blastoderm, which at the junction may be thickened to form a germ ring. The syncytial periblast seems to have the property of autonomous spreading, and it is likely that cell proliferation in the blastoderm is relatively unimportant. Devillers (1961) suggests that the periblast acts as an intermediary between two “non-wettable” components-the blastoderm and yolk. As epiboly proceeds the blastopore contracts as does the surface gel layer over the yolk. In Fundulus this layer probably solates and passes inward (Trinkaus, 1951). The form of the developing germ then seems to be controlled by a balance of the forces of adhesion of the deeper blastomeres with each other and the syncytium, of tension in the surface (yolk) gel layer and of contractility of the periblast. The embryonic axis is laid down by a process of convergence and concentration in relation to the dorsal lip of the blastopore, the quantity of yolk present having some influence on the time at which this event occurs. Presumptive areas have been mapped in the early gastrulae of

4.

DEVELOPMENT: EGGS AND LARVAE

189

some species and show considerable variation. In earlier stages the eggs seem to be of the regulatory type. In FunduZus the 2-cell and 4-cell stage can withstand a loss of half the number of blastomeres. The embryos of Carussius can be divided at the 8-cell stage, each part sometimes giving a normal embryo. Up to the 1Bcell stage two embryos can be fused resulting either in twinning or an oversized single embryo. Removal of the yolk from the blastoderm before a critical stage is reached ( 8 cell stage in Carassius, 32 cell stage in Fundulus, blastula in Salrrw) brings development to a halt. Incomplete removal of the yolk before the critical stage may, however, not prevent further development. It is likely that organizer substances rather than nutrient material diffusing from the yolk are more important. At later stages even the embryonic shield of Fundulus may be isolated and cultured to a fairly advanced stage, with the development of primitive axial organs, ears and eyes, and even with cardiac contractions and independent movement ( Oppenheimer, 1964 ) .

C. Hatching The time to hatching is both a specifically and environmentally controlled character with temperature and oxygen supply exerting a considerable effect. Hatching results from a softening of the chorion (see Fig. 15) because of enzymic or other chemical substances which are secreted from ectodermal glands usually on the anterior surface or from endodermal glands in the pharynx. In the sturgeon Acipenser the latter are innervated by the palatine nerve (Ignat’eva, quoted by Deuchar, 1965). The activity of the larvae, which may be enhanced by increase of temperature or light intensity or by reduction of oxygen tension, assists in breaking through the chorion. The biochemical aspects of hatching are dealt with by Hayes ( 1949), Smith ( 1957), and Deuchar ( 1965). The enzymes have apparently been identified in a number of species, but certainly in salmonids Hayes’ work shows there is doubt about their mode of action. The chorion, which resists digestion by trypsin and pepsin, appears to be of “pseudo-keratin.” Kaighn (1964) measured the amino acid and carbohydrate components in Fundubs. Cystine comprised only l%,compared with 12%in keratin. It is therefore unlikely that disulfide links play a role in stabilizing chorionic protein as they probably do keratin. The hatching enzyme works best under alkaline conditions, pH 7.2-9.6 and temperatures of 14”-20°C having been reported as optima. Very little hydrolysis takes place and Hayes speculates that the enzyme may be a reducing agent which liquefies the chorion. In Oyzias the enzyme is probably a tryptase. In Ftmdulus, Kaighn (1964) obtained purified chorionase and concluded

190

J. H. S. BLAXTER

that digestion of the chorion was mainly a proteolytic process, although he could not duplicate its action with other proteases. Whatever the mode of action of the enzyme, it is likely that a considerable part of the nutrient material in the chorion can be utilized by the embryo via the perivitelline fluid and the losses at hatching may not be too serious ( Smith, 1957).

D. The Larva At hatching the larva is usually transparent with some pigment spots of unknown function. Notochord and myotomes are clear with usually little development of cartilage or ossification in the skeleton. A full complement of fins is rarely present, but a primordial fin fold is well developed in the sagittal plane. The mouth and jaws may not yet have appeared, and the gut is a straight tube. Although the heart functions for a considerable period before hatching, the blood is colorless in the majority of species and the circulation and respiratory systems poorly developed, The yolk sac is relatively enormous with, presumably, hydrodynamic disadvantages. Pigmentation of the eyes is very variable, but where the eye is not functioning at hatching it very soon develops. The kidney is usually pronephric with very few glomeruli. Very little is known about the endocrine glands, gonads, and other organs of the body cavity at such an early stage. As the yolk is resorbed, the mouth begins to function, the gut and the eyes develop further, and the larva becomes fitted for transfer to sources of external food. One of the earlier systems to develop is that responsible for locomotion and support, the primordial fin being fairly soon replaced by median fins and the skeleton laid down. This is one of the better known aspects of later development because it is a system less easily damaged in such delicate organisms and because of the importance of meristic characters (see Section IX) in racial studies of fish. Branchial replaces cutaneous respiration as the gill arches and filaments appear. The swim bladder may or may not be present during the larval phase. It is possible that this and the eyes, which are potentially dangerous in making the transparent larva visible, are silvered in such a way as to render them inconspicuous.

E. Metamorphosis A clear change or metamorphosis from the larval to adult form is to be found in many species. In others there may be a number of less

4.

DEVELOPMENT: EGGS AND LARVAE

191

marked metamorphoses, e.g., in salmonids and eels. The most obvious signs are the laying down of scales and other pigmentation and often the first appearance of hemoglobin in the circulation. The swim bladder and lateral line may also develop first at this stage. In flatfish there is rotation of the optic region of the skull and the change in the normal orientation of the body so that they eventually come to lie on one side. There are often concomitant changes in distribution and behavior such as schooling. Barrington ( 1961) gives detailed consideration to the physiological changes associated with metamorphosis in salmonids, eels, and the lamprey, especially from the aspect of thyroid activity and osmoregulatory functions. The time to reach metamorphosis may be a matter of days in tropical species, a few weeks or months in the majority of fish from temperate latitudes, or periods of years in the sturgeon Acipenser and eel Anguilla. It is controlled not only genetically but also by temperature and food supply, which may affect the rate of growth, and possibly by social (hierarchical) factors as well.

F. Timing To give some idea of the timing of the events just described, examples of the approximate duration of different stages under natural conditions are given in Table 11. The modification of these times experimentally or by fluctuations in environmental conditions is discussed in the succeeding pages.

IV. METABOLISM AND GROWTH

A. Rate of Development Obvious specific differences in time to hatching may be masked by variations in ambient temperature, which is one of the most potent influences on rate of development. Detailed observations of temperature effects during development are scarce. Fluchter and Rosenthal ( 1965), however, showed between 3.5" and 9°C a doubling of heart rate in the embryos of the blue whiting, Micromesistius poutassou, and more rapid embryonic movements. The effect of temperature on time to hatching, a commonly used criterion, is shown in Fig. 4. Low temperatures retard hatching (see reviews by Battle, 1930; Hayes, 1949; Kinne and

192

J , H. S. BLAXTER

Table I1 Duration of Events in the Development of Some Species Weeks from fertilization to Species

Hatch

First feeding

Metamorphosis

11-13 0.8-1.5 1.3-2.0 Scomber scombrus (mackerel) 4-5 0.75 0.25 Roccus saxatilis (striped bass) 8-10 2.0-4.0 2.5-5.0 Osmerus eperlanus (smelt) ? 0.15 0.7 Acanthurus triostegus (convict surgeon fish) 16-18 2.5 3.5 Clupea harengus (Clyde herring) 10-12 4.0 2.5 Pleuronectes platessa (plaice) 1.5-2 .o 2 . 0 - 2 . 4 Not clear cut Oryrias latipes (medaka)

Temp. range ("C) 9-15

Main author Sette (1943)

17

Mansueti (1958)

4-14

Lillelund (1961)

26

Randall (1961)

7-10

Blaxter and Hempel (1963) Ryland (1966)

7-11

Salmo salar (salmon)

20-22

26-28

Gradual

1-7

Squalus acanthias (spur dogfish) Scyliorhinus caniculus (spotted dogfish)

ca. 104

ca. 104

< 104

4-12

New (1966); T. Iwai, personal communication D. H. A. Marr, personal communication Templeman

24-32

28-36

28-36

4-12

Amoroso (1960)

20-25

(1944)

Kinne, 1962), and at a theoretical low temperature (the biological zero) the incubation period will be infinite. The product of incubation time ( D ) and temperature ( T)-day-degrees-was originally thought to be constant, i.e.,

TD = k (1) This was modified to use the temperature, not from zero, but from the biological zero ( T o ) ,i.e., ( T - T,)D = k (2) There has been increasing criticism of the concept, for example, by Kinne and Kinne (1962) and Garside (1966), on various grounds. Plots of 1/D against T are curvilinear over wide ranges of temperature, simple linearity only applying over a narrow range. This invalidates the formulas given above. Furthermore, there may be inflections even of the curvilinear relationship at extreme temperatures. The biological zero, which

4.

DEVELOPMENT: EGGS AND LARVAE

193

Fig. 4. Time from fertilization to hatching at different temperatures. 1. Saluelinus fontinulk (see Hayes, 1949); 2. Pleuronectes platessa (see Simpson, 1956); 3. Gadus mucrocephulus (Forrester and Alderdice, 1966); 4. Sardinops caerulea ( Lasker, 1964); 5. Clupea harengus (Blaxter and Hempel, 1963); 6. Coregonus clupeaformis (Price, 1940; Braum, 1964); 7. Osmerus eperlunus (Lillelund, 1961); 8. Roccus saxatih (Mansueti, 1958); 9. Cyprinodon macularius (Kinne and Kinne, 1962); 10. Salmo gairdneri (irideus) (Garside, 1966); 11. Salmo trutta (fario) (Gray, 1928b); 12. Enchelyopus cimbrius (Battle, 1930); and 13. Scomber scombrus (Sette, 1943).

is usually given between 0" and -2°C and most often around -1.5"C, may be below the freezing point of water or the body fluids themselves. In addition, abnormalities may occur at less extreme temperatures which are not necessarily lethal in the strict sense. Improvements in describing the mathematical relation between D and T arise in later work. For example, Blaxter (1956) used the equation

(T - To)(D - Do) = k (3) for development of Clupea harengus, where Do is the theoretical time to hatching at infinite temperature, not in itself a very satisfactory additional constant. Lasker (1964), working with the eggs of Sardirwps caerulea, used the equation

D = aTb (4) where a and b are constants, and Braum (1964), using the eggs of whitefish Coregonus and pike Esox lucius, D

=

D, $- 1.26T=-T

(5)

194

J. H. S. BLAXTER

where D, is the minimum possible incubation time at the maximum permissible temperature T,. The van't Hoff values over different temperature ranges reflect the nonlinearity of plots of log D against T (see Fig. 4), the values being higher at lower temperatures. Thus in EncheEyopus cimbrius the Qlo varies between 6.5 and 1.5 over the temperature range 5"-23"C (Battle, 1930) and between 6.5 and 2.0 over the range 3"-18°C in herring (Blaxter, 1956). The value of this type of theoretical consideration may lie in establishing criteria for optimum conditions of development. Thus the optima may be where van't Hoff values lie between certain limits. Certainly the day-degree concept is useful as an approximation for predicting events in normal hatchery practice. Other environmental factors influence the rate of development. Low salinities may accelerate or retard the time to hatching (see Kinne and Kinne, 1962; Forrester and Alderdice, 1966), while oxygen lack has a retarding effect on development, especially at higher temperatures ( see Garside, 1966). Laale and McCallion (1968) found that the development of the zebra fish, Brachydanio rerio, could be arrested before gastrulation by the use of the supernatant of homogenates produced from other zebra fish embryos. This arrest, which could be reversed, appeared to be an effect at the cellular level, the nuclei of the arrested embryos all being in interphase.

B. Yolk Utilization The efficiency with which yolk is transformed to body tissue and the effect of the environment on utilization is important in that larger larvae may be expected to be stronger, better swimmers, less susceptible to damage, and less liable to predation. Efficiency at any time may be expressed as a percentage: dry weight increment of body dry weight decrement of yolk

x

100

More often efficiency is measured from fertilization to final yolk resorption (or to maximum weight attained on the yolk reserves). This is gross efficiency, i.e., dry weight of final body dry weight of original yolk or from fertilization to intermediate stages as

x

100

4. DEVELOPMENT:

195

EGGS AND LARVAE

dry weight of body dry weight of original yolk - dry weight of remaining yolk

x

100

or more precisely dry weight of body

dry weight of body

+ dry weight of yolk used for maintenance x 100

The difficulties of measuring efficiency by dry weight lie in the need for taking samples of an egg population at different stages with the accompanying problems of initial differences in egg weight. Utilization of material from the chorion or losses of excretory products are also difficult to allow for, as are the possibilities of uptake of organic matter from the environment (fitter’s theory, see Section V ) . Another serious problem when comparing, for example, environmental effects such as temperature on efficiency is the question of making dry weight measurements at “equivalent” stages (D. H. A. Marr, 1966). Both hatching and maximum weight (attained on the yolk) can be questioned for staging; hatching at different temperatures can result in larvae of quite different appearance, while full yolk utilization is often not complete when maximum weight is reached, some yolk remaining in the yolk sac (see Fig. 5 ) . Furthermore, the disappearance of the yolk sac is no certain indication that all the yolk has been used as it may be present in storage spaces within the larval body, for example, in the subdermal spaces of cod and plaice larvae (Shelbourne, 1956). D. H. A. Marr ( 1966) adopted the ratio dry weight of body X 100 dry weight of body remaining yolk

+

(as percentage)

as a criterion for equivalent staging, comparing in S . salar the efficiency Herring (a-9”C)

10

30

50

Salmon (IO’C)

1 1 0

130

I50

I70

I90

Days after fertilization

Fig. 5. The relative proportions of yolk ( Y ), embryo ( E ) , and chorion ( C ) during the development of a small egg (herring: Blaxter and Hempel, 1963) and a large egg (salmon: D. H. A. Marr, 1966). The vertical dashed line represents hatching. Note the dihrence in the scale of the ordinates.

196

J. H. S. BLAXTER

of development at different temperatures between the 15% and 80%stages. Ryland and Nichols (1967) used the ratio rate of growth in length rate of yolk disappearance

x loo

for equivalent staging when comparing the efficiency of development at different temperatures during the yolk sac stage of P. platessa. The use of maximum weight as a stage for making comparisons still remains, however, a useful criterion and one with immediate meaning when deciding on the optimum conditions for hatchery practice. Calculations of efficiency by various methods are given in Table 111. Efficiency over the whole process of yolk utilization is mainly between 40% and 70%although clearly cumulative efficiency must decrease as growth proceeds and the maintenance requirements increase ( Gray, 1928a). Experiments with temperature ( see Fig. 6 ) indicate certain optima for maximum efficiency. Other influences on efficiency are the original egg weight at the intraspecific level in C. hurengus (Blaxter and Hempel, 1966), and light conditions, contour of the substrate, and turnover of water in the photonegative alevin of s. salar living within the interstices of the spawning redd (D. H. A. Marr, 1965, 1967). Here highest efficiency is achieved under dark conditions, on a grooved substrate, with a rapid turnover of water. A word of caution is required where larval feeding may occur well

\

\

3

0

5

1 0

15

20

TemflC)

Fig. 6. Efficiency of development (see text and Table 111), at different temperatures. 1. Salmo truttu (furio)-yolk sac period (see D. H. A. Marr, 1988); 2. Sdmo sulur-alevin stage (D. H. A. Marr, 1986); 3. S. sah-early alevin (Hayes and Pelluet, 1945); 4. Pleuronectes plutessu-yolk sac larva (Ryland and Nichols, 1987); 5. Clupea hurengus-small eggs, yolk sac period; 6. Clupea harengus-large eggs, yolk sac period (Blaxter and Hempel, 1968).

Table III Efficiency of Yolk Utilization Species Salmo trutta (trout) Salmo trutta (trout)

Method Dry weights Dry weights

Stage Fertiluation to 50-80 days Fertilization to max. weight

Hatching to 10 days after Calorific values Fertilization to max. weight Special dry weight % 15430% (yolk index (see text) sac stage)

Temp. ("C)

Efficiency

10

63

Gray (1926)

15

56

Gray (1928%)

0-16

Author

( %)

Salmo salar (salmon) Salmo salar (salmon) Salmo salar (salmon)

Dry weights

Salmo gairdneri (rainbow trout)

Dry weight and metabolic criteria

Fertilization to max. weight

10

60

Smith (1957)

Fundulus kteroclitus (killifish) Silurus glanis (sheat fish)

Wet weights

19.4-21.4

62

C. G. Scott and Kellicott (1916)

(?)Wet weights

Fertilization to max. weight During yolk sac stage

?

66

Ivlev, quoted by Lasker (1962)

Sardinops caerulea (California sardine)

Calorific values and respiration

Fertilization to yolk resorption

14

79

Lasker (1962)

Clupea harengus (herring)

Dry weights

Pleuronectes platessa (plaice)

Rate of growth in length + rate of yolk disappearance

Fertilization to hatch Fertilization to max. weight During yolk sac stage

42-59

10

41

7.G14.3

64-70

t312 8-12 2.6-9.8

40-80 5M.5 35-58

Hayes and Pelluet (1945) (Fig. 6) Hayes (1949)

D. H. A. Marr (1966) (Fig. 6)

Blaxter and Hempel (1966) (Fig. 6)

Ryland and Nichols (1967) (Fig. 6) Y

w

-a

198

J. H. S. BLAXTER

before final yolk resorption. High efficiency may result from low activity, a high proportion of yolk being used for growth; if this is reflected in low feeding activity it could be a very undesirable trait.

C. Viviparity Amoroso (1960), who gives a comprehensive review of this subject, points out the rather indistinct barrier between ovoviviparity where the young develop within the female only on their yolk reserves, and viviparity, where the nutrient requirements are obtained from the mother. In the first place, there are a wide variety of methods of obtaining these nutrients: by absorption through simple external surfaces, by swallowing, or by “placental” connections. Perhaps all these may be considered as viviparous. In the second place, initial development may be ovoviviparous (on the yolk supply) with viviparity superimposed later. This is likely to be the case in the very early stages of all viviparous fish, but in the smooth dogfish, Mustelus laevis, and goodeid teleosts, for example, there is a change from one to the other rather later. Details of the functional morphology of viviparity are dealt with in the chapter by Hoar, this volume. The changes of weight found during the development of various species ( Amoroso, 1960) can be very striking. Some of the ovoviviparous ones, where maternal nutrients are scarce, show decreases of organic matter between the fertilized egg and final embryo, for example, of 23434% in Torpedo spp. This gives an efficiency of 6674%which puts this species very much in the same category as oviparous forms (Section IV, B ) although it is rarely certain what nutrient is obtained from the mother. The long gestation period of 4 6 months with this order of efficiency suggests some nutrients are being absorbed from the oviduct. In other “ovoviviparous” species there may be gains in organic matter, e.g., of 35%% in Mustelus uulgaris and 16288 in Trygon. In Mustelus laeuis, sometimes more strictly called viviparous, the gain is 1064%,although Te Winkel (1963) reported a fall in weight of organic matter early in the development of a close relative, M. canis.

D, Biochemical Aspects Much work has been done in the past on the larger salmonid eggs which can provide a greater bulk for analysis (see Hayes, 1949; Smith, 1957, 1958) or on whole ovaries (Lasker, 1962). Some more recent work

4. DEVELOPMENT:

EGGS AND LARVAE

199

which includes the use of isotopes, chromatography, and histochemistry is mentioned by Deuchar (1965) and Williams (1967).

1. WATERRELATIONS The swelling of eggs, with the formation of the perivitelline space as a result of water uptake, is a general phenomenon signifying fertilization. The chorion at fertilization is permeable to water and also to urea, glucose, salts, and certain dyes. It seems likely that the colloidal materia1 liberated from the cortical alveoli cannot escape through the chorion and creates an osmotic pressure which draws in water. This effect can be inhibited by high osmotic pressure in the outside medium, The chorion then hardens, a process taking a matter of a few hours (see Fig. 15) and the osmotic forces become matched by the resistance of the chorion. In hypertonic solutions the eggs of salmonids and of Mullus barbatus lose water only from the perivitelline space and not from the embryo (see Zotin, 1965). Some loss does, however, occur in acipenserids. The use of D,O on water-activated and early fertilized eggs of Oncorhynchus tshawytscha seems to confirm the view that only the perivitelline space is penetrable by water (Prescott, 1955). Subsequent use of 3H20,?,NNaCl, and Na13*Ion the water-activated eggs of Salmo gairdneri has shown a definite but limited permeability of the vitelline membrane to anions, cations, and water (Kalman, 1959). Unfortunately, this was not done on fertilized eggs. Recently, however, Potts and Rudy (1969) have confirmed with 3 H , 0 that the vitelline membrane of fertilized eggs of S. salar has a high permeability before laying and during water hardening. Subsequently permeability is low until the eyed stage. The use of 24Na(Rudy and Potts, 1969) showed that sodium exchange is confined initially to the perivitelline fluid but accumulation within the embryo occurs during the eyed stage. Terner (1968) reported that during the eyed stage of Sabno gairdneri external substrates such as I4C-labeled pyruvate and acetate were apparently taken up and metabolized as judged by the presence of 14C0, in the respiratory CO,. Wedemeyer (1968) also found that s5Zn was taken up by developing eggs of coho salmon, 0. kisutch. Almost all was bound to the chorion, but 26%was found in the perivitelline fluid, 2%in the yolk, and 1% in the embryo. Mounib and Eisan (1969) found that both "C-labeled pyruvate and glyoxylate could be utilized in the form of an exogenous substrate by salmon ( S . salar) eggs. Lactate was produced and any carbon atom in these compounds could be incorporated by the eggs into organic acids, lipids, nucleic acids or proteins. The formation of 14C-aminoacids indicated the presence of an active transaminase system.

200

J. H. S. BLAXTER

The use of freezing point measurements on the yolk of herring and plaice eggs (Holliday and Jones, 1965, 1967) gives an alternative picture. In the herring, which has a deniersal egg, the yolk is not regulated osmotically until after gastrulation when it is covered by cells. However, in the pelagic floating egg of the plaice, regulation occurs from fertilization, indicating the ability of the vitelline membrane to osmoregulate. This seems necessary from the buoyancy aspect. If Gray's (1926) measurements of wet and dry weights of the embryo and yolk of trout are generally true of salmonids, then the yolk as a high density nutrient (41%dry weight) requires considerable quantities of water for transformation to the relatively watery embryo (16% dry weight). Obtaining this water may be a problem if the permeability of the vitelline membrane remains limited throughout much of development. Smith (1957) suggests that growth may be retarded until hatching occurs and water becomes more readily available. It is likely that highly desiccated yolk is only feasible in the demersal egg; in Sardinops caerulea the water content of the larval yolk is about 91%(Lasker, 1962).

2. CHEMICAL COMPOSITION The chemical composition of the eggs of two species are given in Table IV. Further data are given by Phillips and Dumas ( 1959), who showed in particular that there was no difference in the constituents of different sized eggs of Salmo trutta. There is considerable difficulty in deciding on the sequence of utilization of various materials for energy production. Analysis of the change in chemical components may be unreliable where substances like carbohydrates are being synthesized. Heat production and respiratory quotients are difficult to determine accurately in small eggs, and COz Table IV Analyses of Eggs" Percent of wet weight Dry weight, Species

(%)

Protein

Total fat

Oil

PhosCarbopholipid hydrate

S . gairdneri (irideus) (rainbow troot) Sardinops caerzileab (California sardine)

33.8

20.2

-

3.6

3.8

0.2

1.3

29.3

21.0

3.8

-

3.2

<0.3

2.1

Ash

Data for trout from Smith (1957) and for sardine ovary from Lasker (1962).

* Calorific value of dry sardine yolk was 5.4 cal/mg.

4. DEVELOPMENT:

201

EGGS AND LARVAE

liberation may be masked by a buffered external medium. Amberson and Armstrong (1933) found the RQ’s of Fundulus eggs over the first 6 days of development were 0.90, 0.78, 0.77, 0.76, 0.72, and 0.72 suggesting very early carbohydrate metabolism. In Oryxius, Hishida and Nakano ( 1954) found 0.75 after fertilization, 0.92 at gastrulation, and 0.70 later. Another problem to overcome is the retention of nitrogenous excretory matter within the egg which gives values of protein metabolism which are too low. The original concept of Needham (1931) that materials were used up for energy in the sequence carbohydrate-protein-fat

has to some extent been supported by Devillers (1965). However, Hayes (1949) was of the opinion that the sequence in salmon eggs was hatch

1

fat-protein-fat-protein

and Smith (1957, 1958) that it was hatch

phospholipid-protein

1 + carbohydrateprotein-phospholipid-triglyceride

fat

Further work is required in this field, especially of a comparative nature.

3. CARBOHYDRATE METABOLISM Since carbohydrate is present in small quantities and synthesis is continuously occurring, its role is particularly difficult to ascertain by bulk analysis. Hayes (1949) reported a gradual increase in carbohydrate level during salmon development with a temporary fall in glucose level at hatching. Glycogen is probably synthesized near hatching and is present exclusively in the embryo, being stored in the liver near the end of the yolk sac stage. Glucose, in aqueous phase, is present in relatively greater quantities in the embryo than in the yolk due to its higher water content. Smith ( 1957) described falls in total carbohydrate after gastrulation (linked with the establishment of the circulatory system and therefore a higher potential for metabolism) and a further fall in glucose level at hatching, perhaps because of an interruption of glycogen synthesis. He did not report the general increase found by Hayes. Terner (1968) has more recently found that the ova of Sulmo gairdneri have a free store of glucose which increases during development and only falls at hatching. Carbohydrate is certainly being synthesized in development for use in the final stages of energy production and for mechanical and osmoregulatory work.

202

4.

J. H. S. BLAXTER

PROTEIN

Protein is the dominant raw material in the yolk and the main source for tissue formation. The use of labeled amino acids in 0Tyzia.s shows a slow passage from the yolk to a low molecular weight pool, a more rapid uptake occurring when the vitelhe circulation is established (Monroy et al., 1961). The proportion of yolk used for energy requirements is not easy to determine since the egg retains some nitrogenous metabolities (shown by an increase of nonprotein nitrogen with age) and only small amounts of ammonia are excreted. Of particular interest is the finding by Read (1968) of ornithine carbamoyltransferase and arginase in the embryos of Squalus suckkyi and Raja binaculuta, suggesting an early functioning ornithine urea cycle with the retention of urea for osmoregulatory purposes. The use of protein for energy seems to decline after hatching, while the amount used over the whole of early development depends on whether calculations are made when some yolk remains or whether, at a later stage, the larva starts to consume its own tissues. While it has been suggested that 40% of the original protein may be metabolized, more recent detailed studies using isotopes (see Williams, 1967) on 0~yzia.sindicate that all protein being resorbed from the yolk in the 9 days subsequent to gastrulation was being transformed into new tissue and none lost in combustion. Regarding different types of protein it appears that in Oryzias 40% of phosphorus is incorporated as phospho-protein; it falls after gastrulation and the loss is very rapid at hatching, especially in S. gairdneri (irideus). Deuchar (1965), reviewing analyses of amino acids present in the same species, reports that aspartic and glutamic acid are always present, valine and leucine appearing after gastrulation.

5. FAT There is some ambiguity about the role of fat in energy production during different phases of development. Fat is undoubtedly used as a fuel, possibly 7040%being consumed over the whole period of development. Hayes (1949) reported the main loss of fat to be in the fourth week after hatching in the salmon. It seems that the triglyceride fats are the last to be utilized before food is required from external sources. The energy requirements before hatching may also be met by fat, according to Hayes, or by fats in combination with protein (Smith, 1957). According to Phillips and Dumas (1959) fat is synthesized from protein in the eyed stage of the eggs of brown trout SaZmo truttu. Terner et al. (1968) found that I4C-labeled acetate in the incubation medium of

4. DEVELOPMENT:

EGGS AND LARVAE

203

the eggs of S . gairdneri, S . trutta, and Salvelinus fontinulis was incorporated into free fatty acids and other lipids of the egg, which were probably the substrates for endogenous respiration. In addition, labeled acetate was found in the egg phospholipids, presumably as an intermediate stage in the synthesis of complex fats. These authors suggest that the lipids of the embryo are not transferred directly from the yolk but are resynthesized by the embryo after the breakdown of triglyceride fats in the lipid pool of the yolk.

E. Respiration In the sense of obtaining oxygen, respiratory problems vary with the environment of the egg and larva, but they are probably rather rarely limiting with pelagic eggs. Eggs deposited on or in a substrate may act as an oxygen “sink” (Daykin, 1965) the oxygen level at the egg surface always being less than that of the surroundings even in high velocities of current. Hayes et al. (1951) calculated a flow of 1.2 x lo-* pI/cm*/cm thicknesslmin through the chorion of salmon eggs. Sensitivity to low oxygen tensions or anoxia varies with both species and stage of development ( see Devillers, 1965). In salmonids, cleavage may continue in anoxic conditions; perhaps this is possible because requirements are low and oxygen is stored within the egg, especially in the perivitelline fluid. Gastrulation is more easily blocked by lack of oxygen or by respiratory poisons such as cyanide or azide. Development can also be retarded by blocking oxidative phosphorylation with dinitrophenol. During oxygen lack there are often increases in lactic acid production but, in the shortterm, retardation of development to anoxia is reversible. De Ciechomski (1965) succeeded in keeping the demersal eggs of Austroatherina for 17-18 days in “Vaseline oil.” Development was at first normal, but no heart beat or movement was observed. Transfer to water resulted in the heart starting to beat and some movements occurred, but abnormal pigmentation developed, and no embryos hatched. Eggs of three pelagic species were killed by immersion in the oil. Various adaptations are found which assist in obtaining oxygen. The spawning redd of the salmonid fish has a rapid current of water passing through it; parents guarding their eggs may ventilate them (e.g., the stickleback, Gasterosteus aculeatus ) while the lungfish Lepidosiren has special pelvic “gills” for oxygenating the eggs in its nest; viviparous species obtain oxygen as well as nutrients from the female. After hatching functional gill filaments are often lacking; in herring they do not appear for some weeks at a length of about 20 mm (Harder, 1954).

204

J. H. S. BLAXTER

Oxygen is obtained over the body surface and almost certainly via internal body surfaces such as the pharynx and gut. The ultramicroscopic corrugations described by Jones et al. (1966) and Lasker and Threadgold (1968) on the epidermis might possibly be directing the flow of water over the body surface. In less favorable environments (see Nikolsky, 1963) external gills may be found, for example, in the bichir Polyptems, in the lungfish Protopterus and Lepidosiren, in the loach Misgurnus fossilis, and in Gymnarchus nibticus. In newly hatched trout, Salmo trutta, respiratory currents are produced by the pectoral fins which also keep the body clear of silt. After a short time mouth and gill movements replace those of the fins and then clogging of the gills by silt is prevented by a coughing reflex or aggregating the silt with mucus (Stuart, 1953). Intake of oxygen has been measured in a number of species. In the egg stage this may be very low and variable before fertilization. In Oyzias Zutipes there is no sharp rise until a few hours after fertilization, although it seems likely that the respiratory rate becomes steadier at this stage (Nakano, 1953). Before any movements occur within the egg, relative values of oxygen uptake between different stages and species are quite valid, but subsequent to movement being possible the oxygen consumed can be dominated by bursts of activity. Oxygen uptake may be used to express metabolic functions, total or general metabolism consisting of two components, active metabolism (owing to activity) and basal metabolism ( owing to maintenance functions). Standard metabolism applies to practically motionless fish under experimental conditions; in the fry of Salmo salar, for example, the basal metabolism was 6 4 6 % the standard rate (Ivlev, 1960a). For active fish metabolism is proportional to KV”, where V is velocity and K and n are constants. The change in oxygen uptake with varying velocity is shown in Fig. 7A. The function KV” is called the “scope for activity” and has been estimated for a number of species (see Table V ) . Some values of oxygen uptake expressed as Qo2 (pl/mg dry weight/ hr) are given in Table VI, based where possible on subjective criteria of activity and inactivity. Without the type of precise control as demonstrated in Fig. 7A these are the best available data at the present time. The standard Qo2’s show some similarity considering the variety of the material used. Active Qoz’s will naturally fluctuate widely, depending on the type and extent of activity permitted. The use of certain anesthetics to control activity (e.g., Holliday et al., 1964) may also help in establishing at least some of the factors controlling oxygen uptake. Cumulative oxygen uptake has been measured over short periods (Fig. 7B) as well as fairly long periods of development. In Fundulus heteroclitus, the killifish, it was 80 pl/ egg from fertilization to hatching

I

(A'

6

Velocity (c m/sec)

0.23

Time (min)

(0

0 10

20

30

0

,

, 40

I

, 80

,

I I

1

'

I20

O2 Tension Onm Hg)

Days otter fert ilizat Ion

Fig. 7. ( A ) Oxygen uptake ( Qo,) for various sizes of young fish stemming differS . salar (Ivlev, 1960a); ( - - - ) Alburnus ent velocities of current at 20°C. (-) alburnw (Ivlev, 196Ob). Wet weights of original data estimated as four times the dry weight. Estimated dry weight and the length given at the end of each curve. ( B ) Cumulative oxygen consumption in clupeid eggs and larvae. Sardinops caeruba: 1. 20 hr embryo at 14"C, 2. newly hatched active larvae at 14°C (Lasker and Theilacker, 1962); Clupea hurengus: 3. newly hatched inactive larva at 8"C, 4. newly hatched active larva at 8°C (Holliday et al., 1964). ( C ) The effect of ~ on wet weight) of different stages temperature on oxygen consumption ( Q o based of the eggs of Esox kcius; hatching at about 11 days at 10°C (Lindroth, 1942). and left-hand ( D ) The effect of oxygen tension on oxygen consumption. (-) ordinate, eggs and larvae of salmon, S. salar (Hayes, 1949); ( - - - ) and right-hand ordinate, eggs of pike, Esox lucius (Lindroth, 1942). 205

206

J. H. S . BLAXTER

Table V Scope for Activity

Species

Stage

Range of increase of Qo, owing to activity

Salmo salar (salmon) Salmo salar (salmon) Sardinops caerulea (California sardine) Clupea harengus (herring)

Eggs

3

Fry

7-14

Larvae

33

Larvae

9-10

Author Hayes et al. (1951) Ivlev (1960~) Lasker and Theilacker (1962) Holliday et al. (1964)

(Scott and Kellicott, 1916). In salmon eggs (Hayes et al., 1951) the uptake was about 0.2 pl/egg/hr at fertilization rising to 3.4 pl/egg/hr at hatching, the cumulative total between these stages amounting to 1400 &egg. Privolniev, quoted by Devillers (1965), found only 850 &egg from fertilization to an age of 120 days. Such values may be compared with the short yolk sac period of the California sardine, Sardinops caerulea, where only 12 &larva were used from hatching to 180 hr subsequently ( Lasker and Theilacker, 1962). Obviously cumulative uptake and uptake per unit time will vary with size, age, and temperature. Devillers (1965) considers that increases in oxygen uptake with age may be shown by the exponential equation Q

=

aekt

(6)

where t is time and a and k are constants. QO2'swill also vary with environmental conditions (see Fig. 7C). In herring eggs and larvae the Qlo for oxygen uptake (Qop)was 2.0 between 5" and 14°C. Qop also increased with a salinity shock but not with constant rearing salinities. There were only slight effects because of light intensity and starvation (Holliday et al., 1964). In salmonids (see Section VIII) oxygen uptake falls at low ambient oxygen tensions (Fig. 7D), but the young stages of fish will resist oxygen lack quite well for a time. This raises the problem of oxygen debt in such experiments and indeed in any respiration measurements where activity is intense. The accumulation of such a debt can be tested by continuing respiration measurements after activity or anoxia. Hayes et al. (1951) found no oxygen debt after periods at reduced oxygen tensions in salmon eggs, suggesting there had been a reduction in metabolism under these conditions. Salmon fry, however (Ivlev, 1960a), showed an oxygen debt

Table VI

A

Oxygen Uptake as Qo2 (pl/mg dry wt/hr)

E

c Species

Salrno trutta (brown trout) S. salar (salmon)

S. salar (salmon)

Stage Egg, embryo only

Egg, embryo only19 days 50 days Fry (no yolk) 3 . 5 mm 4.6 mm 5.4 mm

Temp.

("C)

Standard rate

10

0.55

10

0.9

10 20 20 20

2.3 2.0 1.6

19-21 19-2 1 5 10 15 20

3.2 0.02-? 0.02-0.36 0.02-0.44 0.02-0.54

Fundulus heteroclitus (killifish) Esox lucius (pike)

Egg, embryo onlyb 6 day larvae (no yo1k)l Whole eggsb (range from fertilization to hatch)

Sardinops caerulea (California sardine)

Whole eggs Whole larvae

14 14

0.8 1.33

Clupea harengus (herring)

Whole eggs

8 12 5 14

1.5 4.0 2.0 3.5

Larvae (less yolk)

a

Dry weight estimated a t 16% of wet weight. Dry weight estimated a t 25% of wet weight.

P

Rate General not (active) defined rate

Author Gray (1926)

%

5

z+I

..

Hayes et a!. (1951)

1 .o 35.0 15.3 12.1

Ivlev (1960s)

C. G. Scott and Kellicott (1916) 3.9

Lindroth (1942)

1.79 2.68

Lasker and Theilacker (1962) Holliday et al. (1964)

2.5 5.0 tQ

0 4

208

J. H. S. BLAXTER

amounting to 48%of general metabolism after a period of activity. Fry of bleak, Albumus dburnus, had a negligible debt when stemming a water velocity of 2 cm/sec but one of about 15%in 6.42 cm/sec (Ivlev, 1960b). An overall review of respiration rates, including that of larval fish, allowed Winberg (1960) to deduce that the relationship Qs

=

0.3 W0.'

(7)

represented the best fit for the data on standard metabolism in microIiter per hour at 20°C for organisms of wet weight W mg. He provides conversion factors for other temperatures but stresses that such relationships are designed to evaluate experimental results rather than replace them. Blaxter (1966), recalculating the data of Holliday et al. (1964) for herring, concluded that in larvae of mean wet weight 1.53mg respiration amounted to 17 pl O,/day at 8°C. Using Winberg's relationship a value of 23 pl O,/day is arrived at, fair agreement considering his relationship is operating at its limit. Deductions can be made about the type of metabolism involved in respiration (Hayes, 1949; Ivlev, 1960b; Winberg, 1960). The consumption of 1000 pl 0, corresponds to the utilization of different materials as follows: 1.05 mg protein 1000 plO2

= 1.23 mg carbohydrate 0.50 mg fat

1 mg of protein yields 4.25 cal, 1 mg of carbohydrate yields 4.15 cal, and 1 mg of fat yields 9.45 cal; therefore, 4.5 cal from protein 1000 p1 0, = 5.1 cal from carbohydrate 4.7 cal from fat 4.77

The value 4.77 cal/lOOO pI 0, is sometimes called the oxycaloric coefficient. This makes it possible to calculate the requirements for growth and metabolism from measurements of growth rate and respiration. On the basis of 1000 pl 0, being equivalent to 4.77 cal, and 1 mg wet weight of growth corresponding to 1 cal, Ivlev (1960b) concluded that young Baltic herring of wet weight 66.5 mg required 16.2 callday for growth and maintenance. These and other data are given in Table VII. Smith (1957, 1958) reviews the work on growth, heat production, and

4. DEVELOPMENT: EGGS

209

AND LARVAE

Table VII Calorific Intake in Some SDecies

Species Engraulis japonicus (anchovy) Alburnus alburnus (bleak) Clupea harengus (herring) C. harengus (herring)

Wet weight (mg)

Daily intake

1500-6300

78426

14-20

38.3

19-22

16.2

(?)15-17

Ivlev (1960b)

8

Blaxter (1966)

250 66.5 1.53

(4

0.18

Temp. ("C)

Author Takahashi and Hatanaka (1960) Ivlev (1960b)

respiration. High respiration and high rates of growth appear to be correlated, at least at some phases of development. In particular, specific growth rate and heat produced show very similar trends at various stages of development of the rainbow trout, Salmo gairdneri (irideus). Up to hatching, plots of heat production on specific growth lie on a straight line with positive slope passing through the origin; after hatching the line passes through an intercept on the ordinate. There are two components in heat production-one resulting from growth and the other from maintenance. Temperature may affect the relative requirements of growth and maintenance thus giving differences in efficiency of development (see Section IVYB) with temperature optima.

F. Growth Growth is influenced in the early stages by the ratio between embryo and yolk weight, especially by the problems of efficient yolk utilization (Section IV, B ) and the need to mobilize the nutrient reserves. Higher maintenance requirements per unit weight in small organisms, surfacevolume ratios, change of diet with age, feeding and searching potential (Section VII, E ) are also important. Growth has often been shown in terms of length but there is a need for uniformity, i.e., whether the caudal fin or fin fold is included, and length gives no information on the condition (fatness) of the organism or of the changes in body proportions with age; this is especially true in the yolk sac stage and at metamorphosis. Similarly, wet weights introduce the error of varying water content; thus, dry weight is the most satisfactory measure. Examples of the relative quantities of yolk and embryo in a large

210

J. H. S. W T E R

and a small egg at different stages of development have already been given in Fig. 5. The increase of weight of the embryo and decrease of the yolk tend to be logarithmic in nature. It is clear (Section IV, B ) that in the earlier stages relatively less yolk is being used for maintenance than for growth, giving a higher efficiency of utilization. Without external feeding the larval weight begins to fall before all the yolk is resorbed. Farris (1959) divided the growth in length of four species of pelagic larvae after hatching into three phases: an early rapid phase after hatching, a slow phase near the compIetion of resorption, and a subsequent negative phase if no food was available. This shrinkage in terms of length was also described by Lasker (1964) in the larvae of Surdimps

caerulea. Growth of salmonid and other freshwater fry in hatcheries has received considerable attention from an economic point of view, to provide maximum growth at minimum cost. Growth in marine conditions, where sampling of the same population is difficult owing to mortality, migration, or changes in net avoidance with age, is less adequately known, although examples are given in Fig. 8. Some estimates of survival of fish larvae depend on a knowledge of growth rate and this is often inadequate or based on flimsy experimental data. Comparisons from tank experiments are dBcult owing to mortalities, size-hierarchy effects, or other possible aquarium artifacts. Examples of size hierarchies in rearing experiments are given in Fig. 9, showing the enormous range of size which may be found from an initially fairly uniform stock. Thus Shelbourne (1964), after a series of rearing experiments on plaice, found the length ranged from 7.5 to 37.5 mm. In dense

40 501

0

20

40

60

BO

100

Days a f t e r hatching

Fig. 8. Length of different species related to age under natural conditions. 1. Melanogrammus aeglefinus; 2. Scomber scombrus; 3. Clupea harengus; 4. Esox lucius (all from Sette, 1943); and 5. Pleuronectes pkztessa ( Ryland, 1986).

4. DEVELOPMENT:

211

EGGS AND LARVAE

80

T I I

60

E l + z

T I I I

rn

-

40

c

0

u

P

20

a 0 0

4

8

12

16

20

24

Weeks a f t e r hatching

Fig. 9. Size hierarchies in rearing experiments: (-) Chpea harengw length (left-hand ordinate) (Blaxter, 1968a), and ( - - - ) Pleuronectes platessa wet weight (right-hand ordinate) (Ryland, 1966).

populations there were also relatively larger numbers of small larvae. Magnuson (1962) tested the effect of food supply on Oryzias kept at high densities. As long as food was adequate, growth was not retarded, social hierarchies did not develop, and aggressive behavior was not manifested. Territorial behavior only occurred when food was limited spatially. Whether size hierarchies occur in natural conditions is not known. It seems they must stem from inherently different growth rates or from some type of dominance hierarchy-the first could occur in the wild, but the second is more likely to occur in crowded tank conditions. Food supply and temperature are the most potent environmental factors in controlling growth in the later stages, and there are few data available on this except for salmonids (see chapter by Phillips, Volume I). I n the egg stage low temperatures often produce longer larvae at hatching (see Table VIII), although Lasker (1964) found the maximum length of Surdinops caeruku larvae at intermediate temperatures. There are then probably temperature optima for yolk utilization which partially give rise to these effects, but often the range of temperature used may be insufficient to show them, On the assumption that high temperatures are nonoptimal other conditions producing smaller larvae, such as low oxygen tensions and current velocities for salmonids (Silver et al., 1963; Shumway et al., 1964), as well as high salinities in the desert minnow, Cyprirwdon mucularius, shown by Sweet and Kinne (1984) and in cod, Gadus mucrocephalus, by Forrester and Alderdice (1966), may also be considered nonoptimal. Combinations of temperature and salinity

212

J. € S. I . BLAXTER

Table VIII Effect of Temperature on Size, Using Yolk Reserves Only

Species Salmo trutta (brown trout) Coregonus clupeaformis (whitefish) Omerus epwlanus (smelt) Cyprinodon macularius (desert minnow)

Method

Temp. Corresponding range size range (“C) (mm)

Maximum wet 5-16 weight Hatching length 0.5-10

135-95 (mg)

Hatching length

12-18

5.24.6

Ldelund (1961)

Hatching length

28-35

5.3-3.7

Kinne and Kinne (1962) Sweet and Kinne (1964) Braum (1964)

13-8.8

4.2-3.6

Coregonus wartmanni (whitefish) Cadus macrocephalus (Pacific cod)

Author

Hatching length

1-7

11-9

Hatching length

2-10

4.1-3.5

Gray (1928b)

Price (1940)

Forrester and Alderdice (1966)

suggest that the optimum conditions for growth in length ouer the ranges used were 26°C for 70%0 salinity, 28°C for 35%0 and 33OC for freshwater in Cyprinodon, and 6°C and 19%o€or Gad- macrocephalus. The mortality, length, and maximum weight of salmon alevins (on the yolk supply) was greatest when they were reared in the dark, on a grooved surface, where presumably activity was minimal (D. H. A. Marr, 1965).

G. Endocrines, Growth and Metamorphosis Russian, Canadian, and other work on the role of the endocrine system in fish larvae is fully reviewed by Pickford and Atz (1957) and the role of the thyroid in metamorphosis by Barrington (1961). The main techniques used were histological, histochemical, or immersion in dilute solutions of thyroid hormone or antithyroid drugs. Evidence from a number of sources suggests that the pituitary is inactive until metamorphosis, for example, in herring Clupea harengw, eels Anguillu anguillu, bream Abramis brama, milkfish Chunos chanos, and sturgeon

Acipenser stellutus. The thyroid varies in its histological “activity” in the early larval stages but at metamorphosis in sturgeon, flatfish, eels, herring, pilchard Sardinu pilchardus, and bone fish Albula uulpes, there are signs of secretion of colloid into the lumen of the thyroid follicles. Whether this

4.

DEVELOPMENT: EGGS AND LARVAE

213

is storage, or release into the circulation is taking place, can be satisfactorily tested only by measuring hormone levels in the blood. Earlier, thyroxine may retard growth and accelerate morphogenesis (e.g., in salmon, sturgeon, Misgurnus fossilis and Lebistes) at a concentration of 1 ppm or less. Rate of oxygen consumption may also be increased. In the developing eggs of ScyZiorhinus canicula (see Dimond, 1963) the functioning of the thyroid seems highly dependent on temperature. At 8°C use of lS1Ishowed the thyroid concentrated and bound iodine, but there was no evidence of thyroid hormone formation. At room temperature thyroid hormone reached a measurable level. Thiourea at concentrations of about 100 ppm may delay hatching, retard yolk resorption, and inhibit growth in salmonids, inhibit morphogenesis in sturgeon larvae, and decrease oxygen consumption in salmon and sturgeon larvae by as much as 15%.Thiourea seems to improve the ability to utilize oxygen at low tensions, although it is not clear whether this is because of its antithyroid action. Thiourea might well improve survival in suboptimal conditions of oxygen concentration that might occur, for example, during transportation. The effect of thyroxine on promoting metamorphosis was tested on the eel (Vilter, 1946). There was an accelerating, but not a sudden, effect with an indication that an increasing threshold controls the sequence of events in metamorphosis. In other words, the later stages of metamorphosis require higher concentrations of thyroxine. Other experiments on lampreys showed no effect of thyroxine, thyroid extract, iodide, or iodine on metamorphosis ( see Barrington, 1961). It may be concluded that there is a suggestion of the thyroid playing a role in growth and differentiation in the larval stages but further work is required, especially at the important metamorphosis stage. If the pituitary is not functioning until metamorphosis the thyroid must be relatively independent of pituitary control in the early stages of life.

V. FEEDING, DIGESTION, AND STARVATION

Often the mouth is not completely formed at hatching, but rapid development in many marine fish larvae leads to the possibility of taking external food before the yolk is finally resorbed. Without success in feeding there is eventually self-metabolism and loss of weight. Yolk may be transported and stored within the body, especially in the base of the primordial fin and other subdermal spaces (Shelbourne, 1956), and fat may be stored in the mesenteries of the larval gut. Unlike demersal

214

J. H. S. BLAXTER

freshwater species there is no vitelline circulation in most pelagic larvae although Scomberesox, Trachypterus, and a few other species are known to be exceptions ( Orton, 1957). Usually there is a yolk sac sinus in connection with the heart and with lateral branches to the subdermal spaces; in the species with a vascularized yolk sac these spaces are much less inflated. After final resorption of the yolk the larvae retain their potential to feed for some days depending on species, egg size and temperature. The concept of a “point of no return” was introduced by Blaxter and Hempel ( 1963) for herring larvae (see Fig. 10) This point, after which the larvae are still living but too weak to feed, may vary from 15 days after fertilization for summer spawners in warmer water to 45 days for winter spawners. In the larval cisco, Leuchichthys artedi, it was 27-38 days after hatching at 3’4°C (John and Hasler, 1956). The activity of larval Soka solea drops from about 70%of the time active at first feeding to 20% some 5-6 days later when, if feeding is unsuccessful, inanition occurs (Rosenthal, 1966). Information on the time to point of no return in relation to spawning and food supply may help to predict the probability of survival in different broods. Most fish larvae are predatory with a large mouth and well-developed I

5-

-

do

days

I-

I3-

0.I

0.2

0.3

0.4

Egg d r y weight (mg)

Fig. 10. The concept of “point of no return.” The time from fertilization to the point where the larvae of Clupea hurengus are too weak to feed is given in the form of a nomogram based on original egg dry weight of different races and the temperature (Blaxter and Hempel, 1963).

4. DEVELOPMENT:

EGGS AND LARVAE

215

eyes. In herring the gape of the jaw increases by 50%during the yolk sac stage: the gape of 0.3-0.4 mm at first feeding depends on the length of the larvae and, therefore to some extent, on the original egg size (Blaxter, 1965). An elastic ligament at the articulation makes it possible for even larger organisms to be taken (Fliichter, 1962). Some species have very small gapes which is partly a function of their small size (e.g., lemon sole Microstomus kitt, sprat Sprattus sprattus, pilchard Sardina pilchardus, sardine Sardinops caerulea, and many gadoids). These must require very small food organisms for their early survival, a factor which is of considerably less importance for many freshwater fish. The forward darting movements of fish larvae in order to engulf the prey are described in a number of species, for instance, in Coregonus and Esox by Braum (1964) and in sole, Soka solea, by Rosenthal (1966). Iwai (1964) found three types of feeding in the larvae of PlecogZossus altivelis: a predatory snapping, a respiratory current with a sieving action by the gill rakers, and a ciliary current associated with the olfactory pits. The extent to which fish larvae are herbivorous and might make use of such currents is not clear. Green food in the gut may be incidental to the swallowing of water or due to fecal material from prey species, and its presence is no evidence of utilization. In the anchovy, Engmulis anchoita, however, phytoplankton in the guts of young stages can be correlated with the formation of gill rakers which presumably act as a sieve ( De Ciechomski, 196%). The utilization of dissolved organic matter (Putter’s theory) has also never been adequately tested. Morris ( 1955), in a reappraisal of the theory, concluded from circumstantial evidence of a low incidence of feeding in larvae collected from nature, that it was likely they were utilizing at least fine particulate matter, which could be collected by mucous cells within the buccal cavity or during swallowing of seawater for osmoregulatory purposes. Morris estimated that the larvae of the night smelt, Spirinchus starksi, might pass about 1.15 ml of water per day across these mucous surfaces, possibly containing 0.011 mg of organic matter, a very small quantity even for a larva of 0.9 mg wet weight. The recent work of Terner (1968), mentioned in Section IV, D, showing that the ova of Salmo gairdneri can probably incorporate exogenous substrates into their endogenous reserves, may also be significant. The guts of many pelagic marine larvae are often empty, a puzzling feature when one considers the fairly high food requirements in the early stages. Apart from a lack of suitable food this may result from the capture by net of the weaker feeders only, rapid digestion, defecation during capture, or diurnal feeding rhythms. Examples of such rhythms, which

216

J. H. S . BLAXTER

71

,_j -,-16

- --

, -4 4

8

Time of sunrise

20

Time of sunset

Fig. 11. The time period over which the gut contents of various species of larvae increase and decrease related to the time of sunrise and sunset. Note the lag at sunset caused, presumably, by the time taken to digest food taken earlier. 1. Pkuronectes plutessu ( Shelbourne, 1953); 2. Pleuronectes platessu; 3. Amnwdytes (Ryland, 1984); 4. Clupea hurengus (see Blaxter, 1965); and 5. Sulmo salar purr (Hoar, 1942).

are given in Fig. 11, do indicate a decrease in feeding at night and an increase at dawn. The role of vision in feeding has been tested experimentally and thresholds of light intensity measured (Table IX). The fall off in rate of feeding corresponds with the dusk and dawn periods. Some species can take food in the dark, e.g., cisco (John and Hasler, 1956), especially when it is present in high concentrations. Some, like sole, take food in the dark for most of the larval life and others, like plaice, only at later stages around metamorphosis ( Blaxter, 1968b). Where vision is important the daily feeding period must vary considerably with season, latitude, clarity of water, and even average cloud cover. Ivlev (19eOb) calculated, from the food requirements of young Baltic herring, their rate of feeding and the density of food, that they needed to feed 15 hr/day in August, in other words nearly all the hours of daylight. Blaxter ( 1966) estimated that much younger stages of herring had 10 hr/day to feed in the southern winter spawning groups and up to 24 hr/day in the more northern summer spawners. Variations in feeding time affect searching power, and therefore survival and growth, and these may be further influenced by the temperatures prevailing in different seasons.

4.

217

DEVELOPMENT: EGGS AND LARVAE

Table IX Main Range of Light Intensity over which Feeding Becomes Reduced Species Onmhyzchus keta (chum salmon) 0. gorbuscha (pink salmon) 0. nerka (sockeye salmon) 0. lcisutch (coho salmon) Coregonus wartmnni (whitefish) Esox lueius (pike) Clupea harengus (herring) Plturonectes plateasa (plaice)

Range of light intensity (me)

Author

10L10-S

Ali (1959)

10L10-'

Ali (1959)

10L10-*

Ali (1959)

100-10-'

Ali (1959), Brett and Groot (1963)

?-lo0

Braum (1964)

1-10-1

Braum (1964)

1w10-1

Blaxter (1966)

10e10-*=

Blaxter (196813)

Older stages feed in the dark.

Success in early feeding, feeding drives, and learning factors have been studied. At very early stages Braum (1964) found only 3-84: of feeding movements in Coregonus were successful, but 30%in Esox. In the yolk sac stage, herring larvae take food successfully in &lo% of their feeding movements but later in 80-90% (Rosenthal and Hempel, 1968; Blaxter and Staines, 1969a). In plaice, feeding is 6040% successful throughout larval development. In herring larvae from 20 to 40 mm in length the feeding drive depends on factors such as satiation and especially on the activity of the prey ( Blaxter, 1965). Live organisms, although more difficult to catch, seem to increase the feeding drive and are probably very important in the young stages, where the main visual process may centre round movement perception (Section VI, A). Larval cisco, however, showed a much higher rate of feeding on dead than live Cyclops (John and Hasler, 1956). The gut is usually a straight or simple tube at first feeding and the food is often digested near the anus. This is an interesting reverse of the normal situation in adult vertebrates. It seems likely that the gut is unspecialized at this stage and digestive enzymes are secreted along its length, although no critical work has been done on this. Harder (1960) gives a detailed account of the changes in the gut during the early growth of clupeids and engraulids; Ryland (1966) gives a rather simpler

218

J. H.

S. BLAXTER

account for the plaice. Movement along the gut is mainly by peristalsis, although Iwai (1964) found cilia in the gut of ayu, Pfecogbssus altivelis, especially in positions posterior to the liver. Backwardly directed ciliary currents and forward peristalsis seemed to cause a circulation of the gut contents, perhaps to aid digestion. In further reports (Iwai, 1967b,c) use of the electron microscope showed intermingling of ciliated cells and columnar cells with microvilli in both Plecoglossus and Hypomesus olidus. Reduction of cilia in later stages suggested a transition from cilia to microvilli. Whether cilia are to be found in many species is still open to investigation. Rates of digestion have been measured by rate of disappearance of the gut contents (the gut wall and body being transparent) or by interposing differently colored food in the normal diet and observing the first signs of colored feces. Kurata (1959), using herring 12 days old, found that the gut clearance rate ranged from 12 to 19 hr or more at 9°C depending on the extent of food intake. The rate of digestion varies from 9 hr at 7°C to 4 hr at 15°C judged by transparency of the gut contents in this species (Blaxter, 1965). Reports of larvae in poor condition or dead in net hauls may be due to selective capture or poor washing and are not certain evidence for mortality. Shelbourne (1957) estimated the condition of plaice larvae in the North Sea and found more signs of inanition early in the year. Measurements of condition factor [weight/length3 x 10001 in herring larvae (Hempel and Blaxter, 1963; Blaxter, 1969a) compared with starving larvae in tanks suggested that larvae in the sea were often very near the point of starvation although, over a number of years, it was not possible to correlate high condition factors with a good food supply. Ivlev (1961) looked at various aspects of starvation in young catfish, Silurus glanis, and bream, Abramis brama. With complete starvation the survival time was 34-46 days corresponding to a weight loss of 3349%.With rations below the maintenance requirement survival was prolonged, but even on a full maintenance diet, the weight remaining constant, the fry died after 126-151 days, indicating that a “need for growth may be characteristic of these young stages and that merely to maintain uniform weight is physiologically or developmentally inadequate. The current velocity stemmed by young carp Cyprinus carpi0 and roach Rutilus mttilus for 5 min was reduced by 10 times, from about 100 to 10 cm/sec, over 52 days of starvation. Age had a considerable influence on survival time without food. In catfish, roach, and bleak Alburnus alburnus, it ranged from 3 to 6 days at yolk resorption and from 109 to more than 180 days when 100 days old. The loss in weight and percentage survival of 12

4.

219

DEVELOPMENT: EGGS AND LARVAE

10

50

30

70

90

Days starved

-

Fig. 12. Weight loss (and left-hand ordinate) and percent sumval ( - and right-hand ordinate) as “envelopes” for 12 species of fish larvae during starvation ( Ivlev, 1961).

species of larvae 25-30 days old during starvation is shown in Fig. 12. Losses of weight of 3045% were possible before death, with survival times of 30-80 days. Starving fish were much more susceptible to unfavorable conditions (see Table X) and also to infection and predation. Table X Lethal Limits of Certain Environmental Factors on Fry Before and After Starvation‘ Species

Normal limit(s) ~~

Starvation limit(s) Days starved ~~

~~

~

Salinity Alburnus alburnus (bleak) Caspialosa volgensis (Volga herring)

4.5

2.5

20

16.5

14.5

20

4.0-10.8

5.7-8.3

50

5.3-8.2

6.3-7.8

30

PH

Tanca tinca (tench) Perca fluviatilis (perch)

0 2

Cyprinus caTpio (carp) Alburnus alburnus (bleak)

From Ivlev (1961).

PPm 0.7

2.1

50

2.8

3.1

20

220

J, H. S. BIAXTER

VI. SENSE ORGANS

A. Vision

The newly hatched larvae of many species have unpigmented, and presumably nonfunctioning, eyes (e.g., sole Soleu soba, mackerel Scomber scombrus, whiting Merhngius merlangus, pilchard Sardinu pilchardus, and sardine Sardinops caerulea) ; others have pigmented eyes ( e.g., plaice Pleuronectes platessa, cod Gadus morhua, herring Clupea hurengus, and salmonids) . Schwassmann ( 1965) examined sections of the eye and brain of larval Sardinops caeruba after hatching. On the first day there was no pigment and little differentiation, but by the second day pigment with a stratified retina and visual cells had developed. At 5 days the optic nerve was myelinated and the optic chiasma showed some interdigitation. Considering there are only about 40 individual fibers in the optic nerve when feeding commences at 3 days, it seems probable that the eye at first feeding is only capable of coarse movement perception. The larval eye is also limited in other respects. In newly hatched Lebistes ( Muller, 1952) and Oncorhynchus ( Ali, 1959), and in Clupea harengus (Blaxter and Jones, 1967) and Pburonectes platessa (Blaxter, 1968b) and in a number of other species (Blaxter and Staines, 1969b) up to metamorphosis, there is a pure-cone retina, although the adult has both cones and rods. Lyall (1957a,b) examined the developing eye of the trout, Salmo trutta, and concluded that the rods might develop from single cones or by an outward migration of bipolar cells, the latter view being supported by Blaxter and Jones (1967). Retinal pigment migration and other retinomotor responses associated with dark and light adaptation only develop when the rods appear and are therefore absent in herring and plaice until the larvae metamorphose. This may be true of many other marine species. Thus the larvae are, visually, poorly equipped, with a single type of visual cell, no ability to dark or light adapt, and with a coarse retinal mosaic. Oncorhynchus species acquire rods and retinomotor responses very rapidly after hatching compared with the marine fish which have been examined. The range of light intensity over which light and dark adaptation takes place is then 101-10-1 meter-candles ( mc), which corresponds fairly well with the thresholds for feeding (Table IX) . The acuity of the larvae is poor compared with the adult. Although the retinal cells are small and fairly closely packed, the eye is also small, as is the focal length of the lens. Baerends et al. (1960) found that young

4. DEVELOPMENT:

EGGS A N D LARVAE

221

cichlids, Aequidens portalegremis, could be trained to distinguish stripes 1.5 mm apart at 3 cm body length; this improved to 0.3 mm at 11 cm body length. Blaxter and Jones (1967) calculated the acuity of the eyes of larval herring (minimum separable angle) to be 200 minutes at 1 cm body length improving to 50 minutes at metamorphosis. Thresholds and spectral sensitivity were measured by behavior techniques in herring, plaice, and sole larvae (Blaxter, 1968c, 1969b). Using a negative phototaxis as criterion the visual thresholds were about mc, the spectral sensitivity curves being plateaulike with peaks which might correspond to different cone populations. Distances of perception and visual fields have been measured by a number of workers. These are given in Table XI.

B. Neuromast Organs It has probably not been fully appreciated that many species of fish larvae have free neuromast organs. Iwai reviews earlier reports and his own work on Tribolodon hakonensis, Tridentiger trigonocephalus, and Oryzias latipes (1967a). These organs are usually situated in lateral rows along the body with the small hump of sensory cells being the main element visible especially in sea-caught or fixed specimens. Rearing, with the use of phase contrast microscopy, makes it possible to retain and see the cupulae which are relatively enormous4.05 mm long in Oryzim but 0.1-0.4 mm long in some other species. In the adult it is likely that they are more often sunk into pits. The ability of larvae to avoid capture before the eyes fully develop is almost certainly a result of these neuromast organs. VII. ACTIVITY AND DISTRIBUTION

A. Phototaxis and Activity Much of the published work on phototaxis in fish larvae is anecdotal. Generalizations that marine fish larvae are photopositive and demersal freshwater larvae photonegative do not stand up to a detailed investigation using different temperatures, light intensities, species, or ages. Changes of the sign of the phototaxis may be related to temperature, especially where a sudden change is applied, e.g., in Coregonus (Braum, 1964). Herring larvae show a positive phototaxis at high light intensities which becomes negative below a certain threshold (Blaxter, 1968~).

Table XI Visual Fields and Perception Distances Distance for locating food (mm)

Feeding distance (mm)

Visual field Single eye

Binocular

Species

Size

Coregonus (whitefish) Esox lucius (pike) Clupea harengus (herring) Clupea hurengus (herring) Pleuronedes vlatessa (plaice)

Early larvae

10

0.5-3

145"

45

Braum (1964)

Early larvae

(?)lo

3-4

150"

80"

Braum (1964)

12-14 mm

5

3

-

-

Blaxter (1966)

10-19 mm

10-26

-

-

-

Rosenthal and Hempel (1968)

5

<5

-

Blaxter and Staines (1969a)

6-8 mm

Author

Y

F cn

!

m m

4. DEVELOPMENT:

EGGS AND LARVAE

223

Activity levels are frequently associated with diurnal light changes and phototaxis. Hoar (1956,1958))Ali (1959), and Heard (1964) review much work on activity, rheotaxis, schooling, and migration of young salmonids. Emergence of the alevins of sockeye salmon, Oncorhynchus nerka, seems to be controlled by incident light on the gravel and can be delayed by artificial light at night. The younger fry of pink salmon, Oncorliynchus gorbuscha, for example, are negatively phototactic and hide among the stones of their home stream by day. They become active at night and may rise to the surface and swim downstream or get displaced by the current. After schooling both pink and chum salmon 0. keta, prefer lighted conditions and presumably substitute schooling for hiding as their protection against predators. They show then a positive phototaxis, stem the current by day displaying a cover reaction only with abrupt changes of light, and again become displaced downstream at night when the visually controlled rheotaxis phases out. Coho salmon fry, 0. kkutch, are rather different in behavior; they are strongly territorial and generally more active, but their responses to light and their diurnal changes in activity are much less marked. The behavior of the fry of the different species of Oncorhynchus can, in fact, be related rather generally to their migratory habits. Pink and chum fry migrate to the sea quite soon after hatching, whereas sockeye fry tend to move into lakes, and coho remain in the home stream for a year or so. Loch trout, Salmo trutta, also show incipient territorial behavior at an early age (Stuart, 1953). Woodhead (1957), in a laboratory study of the larvae of Salmo species, related changes in photokinetic activity levels to ambient light intensity and age. Between 0.005 and 100 mc the photokinetic activity of S. truttu gradually increased above a basal dark level, and this species and S. gairdneri (irideus) and S. salar all showed increased activity with age. A positive photokinesis, together with a negative phototaxis, would work together to keep the larvae within the stones of their home stream. Woodhead and Woodhead (1955) found a similar increase in the activity of herring larvae with light intensity above 0.3 mc, and there was a basal level of activity in very low illumination as there was in the salmonids.

B. Vertical Distribution Diurnal variations in distribution are linked to activity, phototaxis, and brightness discrimination. Some earlier work (see Simpson, 1956) on changes of vertical distribution did not take into account the need for opening and closing the plankton nets at defined depths and for preventing differential net avoidance by day and night. A general lack of larvae

224

J. H. S . BLAXTER

by day, especially larger specimens, partly results from this latter factor, as emphasized by Bridger (1956). With slow tow nets it can be difficult to estimate the relative effects of net avoidance and vertical distribution when comparing day and night catches. Strasburg (1960) used slow tow nets to catch tuna larvae. Some species such as skipjack, Katsuwonus pelamis, were rarely caught by day at all, while catches of all species were always much higher by night. He concluded that there was some net avoidance, but the main diurnal difference was owing to vertical migration. Stevenson (1962) used high speed nets for larvae of Pacific herring, Clupea palbii. The catches at night were far higher and the average size of the larvae was greater, from which he concluded that the larvae were concentrated at the surface by night and spread over a fairly wide stratum by day. There seemed to be net avoidance, even of high speed nets, by the larger larvae in the daytime. Ryland (1964) found a reverse distribution of larval plaice and sand eels Ammodytes, which were spread over a wide depth range from the surface to about 35 m by night and much more concentrated between 5 and 10 m during the day. The use of high speed nets in this study and in that of Colton (1965) on haddock larvae, Melanogrtzmmus aeglefinus, prevented diurnal variation in size and presumably overcame the net avoidance factor.

C. Buoyancy and Pressure There are a number of devices which provide buoyancy in pelagic fish eggs and larvae. In marine species, found in high salinities, maintenance of the body fluid concentration below that of the environment acts as a buoyancy mechanism. In the early stages lack of a skeleton also keeps the specific gravity fairly low. In freshwater the eggs are often demersal so that the much greater problem of maintaining buoyancy does not exist. In fact, in salmonid eggs the specific gravity may be very high in the early stages because of low water content of the yolk. It is sometimes said that a large perivitelline space is a mechanism for buoyancy. However, this normally contains water of the same concentration as the environment and is thus much more likely to act as a shock absorber, particularly in pelagic fish eggs subject to wave action. The embryo and yolk may regulate the body fluids from fertilization (e.g., in pelagic plaice eggs: Holliday and Jones, 1967) by means of the vitelline membrane. In the demersal herring egg there is no regulation until the yolk is covered by cellular tissue at gastrulation ( Holliday and Jones, 1965). Many species such as pilchard Sardina pilchardus, mackerel Scomber scombw, hake Merluccius merluccius and sole Solea solea have oil

4.

DEVELOPMENT: EGGS AND LARVAE

225

droplets, but these are not typical of pelagic eggs for they are absent in the sprat Sprattus sprattus and in Gadus and Pleuronectes species. Tendril-like outgrowths found, for example, in Exocoetus and Scomberesox eggs are as likely for attachment as for buoyancy. There may be a mechanism to adjust specific gravity to the spawning medium. Solemdal (1967) found that variations in the osmotic pressure of the blood of female flounders, Pleuronectes fleas, at the spawning season depended on the salinity. This, in turn, caused changes in the specific gravity of the eggs when spawned. In water of low salinity the eggs were larger and their osmotic pressure lower. In the low saline water of the Baltic (6.5%,), fertilization and development were possible but the eggs rested on the bottom. In the sprat there is a correlation between low salinity and larger eggs (see Nikolsky, 1963) and also in herring (Holliday and Blaxter, 1960) and plaice (Holliday and Jones, 1967). In the larval stages large subdermal spaces as in the plaice and cod (Shelbourne, 1956) will reduce specific gravity as long as the body fluids are maintained at an osmotic pressure below that of the environment. Other species have long processes to the dorsal or lateral fins, e.g., the angler fish Lophius and deal fish Trachypterus. It may be significant that Orton (1957) quotes the latter species as having only moderate fin folds and subdermal spaces, but whether the increased surface area assists buoyancy is open to question; these may be organs to reduce predation. In freshwater, the larvae of salmonids can afford to retain dense yolk reserves as they rest on the bottom. Others like carp and bream have cement organs for attaching themseIves to weed. The first filling of the swim bladder is well known in freshwater fish (see Tait, 1960). In the physostomatous salmonids like S . trutta, S . gairdneri, Cristivomer numuycush, and Coregonus clupsaformis the swim bladder is filled 1 3 months after hatching, depending on temperature or feeding. Air is taken in from the surface via the pneumatic duct and without access to the surface the swim bladder remains empty. The pneumatic duct remains open under such conditions and functions later. Lake trout, Cristivomer numuycush, were able to swim considerable vertical distances ( 280 m ) with unfilled bladders suggesting that development in deep water presented no problems for filling the swim bladder. Physoclists like Gasterosteus, Lebistes (von Ledebur, 1928) and Hippocampus (Jacobs, 1938) also require access to air to fill the swim bladder but if deprived of this at the critical time the pneumatic duct closes and the swim bladder cannot be filled later. Pressure effects on fish larvae were tested briefly by Bishai ( 1 W l ) . Newly hatched herring survived compression to and decompression from 4 atm and young plaice 3 7 5 0 mm long to and from 2 atm. Young sal-

226

J. H. S. BLAX’IZR

monids could live at 5 atm to the end of the yolk sac stage, sometimes showing increased activity during compression or decompression. Older fry, about 20 days after yolk resorption, could withstand compression to 2 atm but had difficulty in dealing with air bubbles inside the body and in the swim bladder during decompression. Older fry still, about 60-140 days after yolk resorption, could not even withstand compression to 1.2 atm. At later stages, the ability to withstand compression increased but decompression was lethal. Qasim et al. (1983) observed changes in the vertical distribution of larval teleosts subjected to pressure changes. Young plaice larvae tended to swim upwards when the pressure was increased from 1 atm to 2 atm and sank when the pressure was returned to normal. Metamorphosing plaice did not show this response. Larvae of the blenny Centronotus gunnellus responded in a similar way to pressure changes equivalent to 25 cm of seawater and larvae of another blenny Blennius pholis to only 5 cm of seawater. Only the last of these species has a swim bladder. It is possible in such experiments that the larvae could have been responding to water currents produced by the pressure changes. If this is not so, one may ask how larvae without swim bladders can respond to pressure.

D. Locomotion and Schooling Most species show signs of activity within the egg, and movement is often an important part of the hatching process. The relatively large yolk sac at hatching must be a hydrodynamic embarrassment to the larvae, but in salmonids, where it is especially large, there is little movement as the larvae grow in a rather passive way within the stream bed. The buoyant oil globule found in many marine pelagic larvae may also make equilibrium difficult. Two sorts of movement are found at yolk resorption-a serpentine eel-like mode of swimming in the long thin-bodied clupeoid type and a more maneuverable movement allowing “backingup” with the use of the pectoral fin and marginal fin folds in the shorter flatfish type of larvae. Swimming performance is shown in Fig. 13 in terms of burst speeds maintained for a few seconds, where the theoretical 10 body lengths/ sec (shown as dashed line) seems to fit quite well. Cruising speeds, of the order of 23 body lengths/sec (ideally the maximum sustainable speed without an oxygen debt accumulating), are shown in Table XII. Improvements in performance can be correlated with development and upturning of the caudal fin. If schooling occurs it usually starts at metamorphosis. Presumably

4.

227

DEVELOPMENT: EGGS AND LARVAE

‘7 E

5 loo

10

I0

0

I5

20

25

0

Length (mm)

Fig. 13. Darting speeds of larvae of different length: (-0) Clupea harengus 1Oo-15”C (Blaxter, 1962), ( X-X ) Pkuronectes platessa 6.5”-7.5” (Ryland, 1963), ( A )Carassius auraatus, no temperature (Radakov, 1964), and ( V )Perca fluuiatilis, no temperature (Radakov, 1964).

the larvae, being transparent, do not provide very good mutual stimuli for keeping schools together visually. Shaw (1960, 1961) analyzed the development of schooling in the silverside Menidia and found responses initiated between two fry at a body length of 8-9 mm with increasing participation as they grew. By 11-12 mm as many as 10 could be seen in parallel orientation. Fish reared in isolation in paraffin-coated bowls took a few hours to become integrated into existing schools. Generally speaking there was a considerable latency in schooling when “isolates” and “quasi-isolates” (isolated S 7 days after hatching) were brought together at a size where schooling was normally well developed. Surprisingly, this initial latency was less among “isolates” than “quasi-isolates.” In the newly emerged fry of Oncorhynchus species, schooling ceased in the dark and following subjection to light they needed 15 min or more to complete the school again, although in all species except coho this time became reduced with age ( Ali, 1959). In herring ( Rosenthal, 1968) schooling started about 2 months after hatching at a body length of 25-30 mm. Starving larvae tended to school more positively, but the school rapidly broke up when food was offered. Rheotropic ( optomotor) responses (orientation to currents or moving backgrounds) are often well developed in the larval stage as shown in some experiments on swimming performance and from the many observations on salmonid larvae by day and night (e.g., Hoar, 1956, 1958). This response seems to be maintained by either visual or tactile cues, and the sign may change under shock treatment. In young salmonids a temperature rise of 12°C changed the rheotaxis from positive to negative, presumably a defensive response ( Keenleyside and Hoar, 1954).

Table XII Cruising Speeds of Some Young Fish Length Species

(-)

Pleuronectes platessa 7.5 9.6 (plaice) 10-12 Coregonus wartmanni (whitefish) 35-45 Rutilus rulilus (roach) 45-55 Abramis brama (bream) 5MO Cyprinus carpi0 (carp) 30-40 Trachurus trachurus (horse mackerel) 35-45 Mullus barbatus (mullet) 45-55 Mugil sp. (mullet) 18-26 Blicca bjorkna (white bream) 6.5-8 Clupea hurengus (herring) 12-14 Clupea harengw (hel-l.hd 4 Solea solea (sole) 0.26 g (wet wt) Albumus alburnwr (bleak) 22 Micropterus dolomieui (smallmouth bass)

Speed Temp. (mm/sec) ("C)

Time maintained

Distance swum (m/hr)

A few sec A few sec

Author

12 27 16 29 138

6.5-7.5 6.5-7.5 4 ?

1 min

126

?

1 min

129

?

1 min

136

?

1 min

128

?

1 min

156

?

1min

330

?

?

Radakov (1964)

(?)14

45 min

Bishai (1960d)

10

9-10

60 rnin

36

6-9

15

Long periods intermittent swimming

17-19

29.8

20

146-312

20-35

5.8

16

? ?

Ryland (1963)

58 103

Braum (1964) Aslanova quoted by Radakov (1964)

107

Blaxter (1966) Rosenthal (1966) Ivlev (1960b) Larimore and Duever (1968)

4.

229

DEVELOPMENT: EGGS AM) LARVAE

E. Searching Ability The ability to search for prey or, in the case f pisual feeders., ptically filter” the environment, .depends on cruising speed and the distance at which food is perceived. Perception distances (see Table XI) depend on the movement of the prey, its orientation in relation to the eye, contrast with the background, and illumination. Activity in terms of cruising speed may well be reduced after a period of feeding and is certainly influenced by feeding drives in older larvae. Starvation may also cause a reduction in activity. The volume searched per hour has been estimated in whitefish and herring larvae (see Table XIII),When multiplied up on a daily basis the volume searched becomes very dependent on day length, that is, on latitudinal and seasonal effects. “

VIII. MORTALITY, TOLERANCE AND OPTIMA

From hatchery, rearing, and experimental studies it is well known that fish eggs and larvae go through periods of mortality. Susceptibility to mechanical shock or nonoptimal temperatures and salinities may vary with age (see Battle, 1944). Marine fish eggs seem especially delicate until the completion of gastrulation although this has not been systematically tested. Perhaps sensitive morphogenetic processes in the early stage or failure to osmoregulate are the cause. Similarly there are stages, such as the eyed stage in salmonids, where eggs may be particularly reTable XIII Volume Searched during Feeding

Species

Size (mm)

Volume searched (liter/hr)

~

Coregonus wartmanni (whitefish) Clupea hrengus (herring) Chpea harengus (herring) Sardina pilchardus (pilchard) Pleuronectes platessa (plaice)

(?)10

Author ~

14.6

8-16

0.3-2.0

10 13-14 5-7

1.5-2 6-8 0.1-0.2

6-10

0.1-1.8

Braum (1964) Blaxter (1966), Blaxter and Staines (1969a) Rosenthal and Hempel (1968) Blaxter and Staines (1969a) Blaxter and Staines (1969a)

230

J. H. S . BLAXTER

sistant to damage. After hatching, further stages of mortality may be experienced which can sometimes by connected with the further development of certain organs and with such physiological changes as the transition from cutaneous to gill respiration or the development of the swim bladder. Rearing studies on marine fish (Section X ) show a high mortality at the end of the yolk sac stage in species such as cod G. morhua, haddock Melunogrammus aeglfinus, pilchard Sardina pilcFcardUs, herring Clupea harengus, and lemon sole Microstomiis kitt. These "critical phases" are probably in part the result of unsuitable food being supplied, especially from the aspect of size. The question of whether critical phases at yolk resorption occur under natural conditions is a much vexed question, examined especially by fishery biologists looking for the factors controlling brood survival. J. C. Marr (1956), in reviewing this hypothesis, concluded in two species (mackerel Scomber scombrus, and Pacific sardine Sardinops caerulea ) that mortality was more likely to be constant or steadily decreasing without the catastrophic periods advocated by earlier workers. Undoubtedly some of the five species mentioned have small mouths and the finding of suitable food in the sea at the end of the yolk sac stage before the point of no return (Section V ) must be a serious problem. The lethal levels of various environmental factors such as salinity, temperature, oxygen tension, pH, radiation, and mechanical stimuli have been assessed for fish eggs and larvae. Salinity tolerance is surprisingly wide, ranging in marine species which have been examined from about S%, up to 4550%0 with even a wider range over short periods of exposure (see Chapter by Holliday, Volume I). Predictably, the lethal temperature varies with level of adaptation. A wide variety of criteria have, unfortunately, been used for defining lethal temperature, ranging from 50% mortality after one day (or longer) to the number of days to 50%mortality. Examples of high and low lethal temperatures where the criteria could be standardized are given in Fig. 14 related to temperature of acclimation. Temperature fluctuations are much damped down in the sea and may be negligible in deep water. In the tropics the seasonal fluctuations may also be slight, and this may lead to a narrow range of tolerance. In 10 species of fish larvae from the Indian Ocean, Kuthalingam (1959) found the range from lower to upper lethal temperature was only 3'4°C (about 28"-32"C). Temperature tolerance may also vary with age. Combs (1965) measured long-term temperature thresholds over the period from fertilization to hatching in Oncorhynchus species. In 0 . nerka, for example, the low temperature threshold was about 5°C and the high temperature threshold about 13.5"C. At the 128cell stage and later both 0. nerka and Chinook 0. tshuwytscha, had a

4.

DEVELOPMENT: EGGS AND LARVAE

231

Acclimation temp("C)

Fig. 14. Lower and upper lethal temperatures at different acclimation temperatures, criterion of lethal temperature being 50%survival after 24 hr; horizontal dashed lines give range of blood freezing point values. 1. Breuoortia tyrannus-in 24%0 salinity, acclimated for 12+ hr (Lewis, 1965). 2. Clupea harengus (spring spawned) and 3. C. harengus (autumn spawned)-in 34%0 salinity, acclimated for long period 28%0 salinity, acclimated for 21 days (Blaxter, 1960). 4. Oncorhynchus keta-in (Brett and Alderdice, 1958). 5. Salmo solar, 6. S. trutta, and 7. S. trutta (fario)-in freshwater, acclimated for 5 days (Bishai, 1960b). 8. Oncorhynchus tshawytscha, and 9. 0.nerka (Brett, 1952).

greater range of tolerances. In freshwater or high latitudes resistance to cold must be adequate or else adaptations are deveIoped which prevent contact with extremely cold water. The flounder, Pseudopleuronectes americanus (Pearcy, 1962), is one of the few species of flatfish with a demersal egg and it spawns inshore in very cold water. Temperature optima may be judged by hatching success. For instance in Coregonus clupeaformis the highest percentage hatching was at 0.5"C with a very

232

J. H. S. BLAXTER

rapid fall off above 6°C (Price, 1940). In herring temperature optima vary with the race (Blaxter, 1956). In medaka Oryzias the best survival was obtained when the rearing temperature was alternated between 22” and 30°C every 12 hr (Lindsey and Ah, 1965). Generally, outside the optima not only does the percentage hatch fall but there are more abnormalities. Oxygen is apparently only limiting at its low level. However, supersaturated solutions may be harmful if air bubbles are swallowed by the larvae. This “gas disease” can be fatal if the larvae cannot eliminate the bubbles from the gut (see Bishai, 1960a) as their buoyancy is disturbed. The eggs of the salmonids S . salar, rainbow trout S . gairdneri, Pacific salmon 0. tshawytscha and 0. kisutch, brook trout Salvelinus fontinalis, and lake trout Salvelinus namaycush, have been tested at low oxygen tensions (Hayes et al., 1951; Silver et al., 1963; Shumway et al., 1964; Garside, 1959, 196f3). Most were able to resist levels of 2.5 ppm or even less (25%saturation) although development was retarded and the larvae smaller at hatching. Reductions in current flow had a similar effect. Alderdice et al. (1958), using chum salmon eggs, measured the retardation in hatching resulting from low dissolved oxygen levels at different developmental stages. The intermediate stages, about one-third of the way from fertilization to hatching, were most susceptible. Hayes et al. defined the limiting tension not from its lethal aspect, but the tension at which normal oxygen consumption started to drop (see Fig. 7D). In S. salar eggs this was 3.0 ppm at 20 days, 7.5 pprn at 45 days, and 4.7 ppm at hatching. Other salmonid data on limiting tensions are summarized by Wicket (1954). In the egg of chum salmon, 0. keta, the calculated limiting tension varied from 0.72 to 3.7 ppm increasing with age (temperatures from 0.1” to 8.2”C). In the eggs of pike, Esox lucius (Lindroth, 1942), it varied from 1.2 to 4.1 ppm depending on age and temperature. This type of finding is of value in assessing the survival chances of fish eggs buried in gravel. Wickett (1954)) for example, has shown that the flow rates and oxygen content of the water may be quite inadequate for chum salmon eggs, being even as low as 2 mmlhr and 0.2 ppm. Marine and other fish larvae were studied by Bishai (1960a). Herring larvae at hatching were “restricted at 27-32% saturation and at 55-64% saturation a few days later. Larvae of the lumpsucker, Cyclopterus lumpus, just tolerated 43%saturation 3 weeks after hatching and young plaice 4-5 cm in length just tolerated 14%saturation. He found that the larvae of the salmonids, S . salur and S . trutta, could survive very low levels of saturation, %lo%,after hatching, but this ability decreased with age, the minimum level tolerated being 1628%.Again, criteria for de-

4. DEVELOPMENT:

EGGS AND LARVAE

233

fining lethal levels vary, and it is difficult to compare the results of different workers. To some extent acclimation to low oxygen increases resistance to hypoxia and there are also specific differences in tolerance. Bishai (1960~)has also summarized work on the effects of pH on larval fish. Compared with seawater ( p H about 8.0) which is fairly well buffered, freshwater has a pH which is more variable and more susceptible to change by effluents or run-off. Bishai found that herring larvae could survive values between 6.5 and 8.5. Lumpsucker larvae had a low tolerance level of about 6.9, plaice (4-5 cm) 6.2-6.5, while salmonid larvae had a range of tolerance from about 5.8 to 6.2 up to 9.0, or even 10.0 in sea trout alevins. To some extent the levels of tolerance depend on the substances used to change the pH. It is possible that CO, for reducing pH has an additional toxic or respiratory effect compared with the use of HC1. Certainly Alderdice and Wickett (1958) found high levels of CO, inhibited oxygen consumption in the eggs of 0. keta. Radiation effects on salmonid eggs and larvae are reviewed by Hamdorf (1960) and Eisler (1961). Strong visible light may cause early hatching, mortality, poor growth, and greater pigmentation. Exposure to light is less serious when the intensity is low or in the later stages, especially after the onset of pigmentation. It is, perhaps, not surprising that light is harmful to eggs and larvae which normally develop in complete or semidarkness among gravel or stones; increased activity probably contributes to its deleterious effect. Other species which develop in sand such as the grunion, Leuresthes tenuis ( McHugh, 1954), or near the sea bed such as herring (Blaxter, 1956) may also show a poorer rate of hatching under lighted conditions. Careful controls are required in experiments since the light may cause other changes such as raising the temperature or increasing growth of phytoplankton in the water. The main effect on many other species is to accelerate hatching. Ultraviolet light has been tested on Fundulus and causes abnormalities of the skeleton, cylopia and twinning (see Eisler, 1961). In sockeye salmon, 0. nerka ( Bell and Hoar, 1950), it causes premature hatching, vertebral abnormalities, high mortality, and delays pigmentation. Marinaro and Bernard (1966) exposed the eggs of pelagic fish such as pilchard Sardina pilchardus, mullet Mullus, horse mackerel Trachurus and “sargue” Diplodm annularis to sunlight with and without ultraviolet filters, finding a lower rate of hatching with ultraviolet. The transparency of many fish eggs and larvae may be an adaptation to prevent absorption of light as well as acting as a means of camouflage. X rays have been tested, for example, by Solberg (1938) on Fundulus embryos. They were most sensitive after fertilization with a decrease of 10 times in sensitivity between early cleavage and 4-5 days later. Other work is reviewed by Eisler (1961)

234

J. € S. I . BLAXTER

especially on salmonids. X rays have a number of effects such as higher mortality, decrease of growth, deformities, decrease in erythrocytes, and destruction of hematopoietic tissue and increased pigmentation. Generally, susceptibility decreases with age and in optimal conditions of temperature. Neyfakh (1959) used X rays for inactivating the nuclei in the developing eggs of the loach, Misgurnus fossilis. Irradiation during certain phases of development was followed by death at a very specific time, suggesting a periodical functioning of the nuclei. Fish eggs in almost any environment are subject to mechanical stimuli which are potentially harmful-movement of gravel within a spawning redd, current borne objects on the sea bed, or wave action at the surface. The resistance of the egg may be measured by the maximum load it will take before the chorion bursts. Some results are shown in Fig. 15. They indicate an increase in the resistance of the chorion as it hardens subsequent to fertilization and a decrease in the strength of the chorion prior to hatching, presumably resulting from hatching enzymes. Changes in the chorion may in part account for periods of varying resistance to external conditions. Other harmful influences may be hatching enzymes, especially for embryos which may be somewhat unhealthy in other respects, or where eggs are very crowded and the enzymes are concentrated in the water after hatching. Stuart (1953) observed sediments attached to the chorion of loch trout. These might well reduce oxygen intake. The alevins, howl0.000,

Days after fertilization, log scale

Fig. 15. The maximum load before bursting in eggs of different species during development. 1. Salmo salur, 2. Coregonus luvaretus, 3. Acipenser sp. (see Nikolsky, 1983); 4. Salvelinus fontinalis ( P ) , 5. Salmo salar (see Hayes, 1949); 6. Pleuronectes platessa (range, varying age) (Shelbourne, 1963); and 7. Engraulis anchoita ( De Ciechomski, 1967a).

4.

DEVELOPMENT: EGGS AND LARVAE

235

ever, seemed to have the ability to disperse such sediments by means of respiratory currents passing over the body surface. With gill respiration sediments were aggregated by mucus and rendered less harmful. A recent trend in research on tolerance is to use combinations of factors such as temperature, salinity, and oxygen concentration to see how they interact and to find out the optima. Kinne and Kinne (1962) used such combinations during the development of the desert minnow, Cyprinodon mucularius (i.e., temperature between 10" and 37"C, salinities from 0 to 85%,, and 70%)lo@%,and 3004: saturation of air in water). Mortality was least at near lethal temperatures when the salinity was held between one-half seawater and 35%0. The lethal temperature was lower in hypoxial conditions. It seemed that of the salinities used 35%, was the optimum for incubation. Lewis (1966))using menhaden larvae, Brevoortia tyrunnus, found that temperature tolerance was greatest at salinities of 1CL15%0, suggesting that isotonic conditions were optimum for survival. Forrester and Alderdice (1966) calculated from experimental data that the optimum salinity and temperature for hatching eggs of cod, Gadus mucrocephulus, was 19.4%, and 53°C.

IX. MERISTIC CHARACTERS

Counts of vertebrae, myotomes, scales, gill rakers or fin rays have been used in racial studies. The variability in number and lability under different environmental conditions are striking examples of rather flexible raw material for evolutionary forces. Usually mean counts from samples of fish are required, the differences from race to race being inadequate for identscation of individuals. While correlations between the counts in adults or young and the environmental conditions on the spawning ground first suggested that meristic characters were labile, confirmation has been obtained by experimental studies. They show that factors such as temperature, salinity, and oxygen content superimpose their effect on the range of meristic counts determined by the genotype. There is also an individual variation even in fish from the same parents reared under the same conditions. Earlier work, mainly on salmonid and inshore or estuarine species, has been reviewed by T h i n g (1952) and Barlow (1961). Since it is necessary to keep the young until the meristic characters form there has been less work on species which require to be fed and reared for a considerable time after hatching. The effects of some factors are given in Table XIV. Most work has

236

J. H. S. BLAXTER

been done on the effect of temperature on vertebral and fin ray counts, Generally a V-type relationship is found where the average vertebral count is minimum at an intermediate temperature. For example, in sea trout, Salmo trutta, THning found this was at about 6°C; in plaice it is at 8°C (Molander and Molander-Swedmark, 1957); at lower and higher temperatures the average counts were higher. With fin rays the V may tend to be asymmetrically inverted with counts highest at intermediate temperatures, In other instances the mean vertebral counts are inversely related to temperature, but the lack of an inflection may result from an insufficiently wide range of experimental temperatures being used. Apart from the factors shown in Table XIV, both intensity and duration of light ( McHugh, 1954; Lindsey, 1958) and CO, ( T h i n g , 1952) may have an effect. Higher intensities in the grunion, Leuresthes tenuis, reduce the number of vertebrae. A 16-hr day caused lower caudal vertebrae and possibly anal fin ray counts in Oncorhynchus nerka, while in sea trout, S a l m trutta, increasing tension of CO, produced lower numbers of vertebrae. The work of Heuts (1949) and Lindsey (1962) using different temperature-salinity combinations in Gasterosteus aculeatus were somewhat inconclusive resulting from the problem of getting sufficient numbers of survivors in an extensive series of experiments, some of which were in unfavorable conditions. Certainly it was shown that temperature effects vary with the salinity level. Using a freshwater and brackish water race, Heuts found the greatest effect of temperature on fin ray counts at the salinity of adaptation. Differential mortality and the unsolved problem of controlling, for example, oxygen and temperature independently have often led to difficulties in interpreting results. It is the underlying reasons for meristic lability which is, perhaps, of most interest to the physiologist. Sensitive periods have been found during which lability of vertebral counts is at its greatest. In sea trout it is around gastrulation, with a further period when the last vertebrae are being preformed ( THning, 1952). It is also before hatching in herring (Hempel and Blaxter, 1961) and killifish (Gabriel, 1944). However, in plaice ( Molander and Molander-Swedmark, 1957) and paradise fish ( Lindsey, 1954) meristic characters are still susceptible to modification after hatching. Generally speaking, fin ray counts are determined later than those of vertebrae. Later studies on Oryzias (Lindsey and Ali, 1965) showed that transfer to a higher temperature 2 days after fertilization produced fewer vertebrae, whereas transfer after 6 days resulted in a higher vertebral count. Regular diurnal alterations of temperature between 22" and 30°C gave vertebral counts intermediate between those resulting from sustained temperatures of 22" and 30°C.

Table XIV Environmental Effects on Meristic Characters bv Emeriment

+i

i3

Effects of increasing Species Lebistes retaculatus (guppy) Salmo trulta (sea trout) Salmo trutta (sea trout) Salmo gairdneri (rainbow trout) Oncorhynchus tshawytscha (Chinook salmon) Fundulus heteroclitus (killifish) Macropodus opercularis (paradise fish) Pleuronectes platessa (plaice) Pleuronectes platessa (plaice) Channa argus (snake-headed fish) Clupeu harengus (herring) Gasterosteus muleatus (stickleback) Oryeias latipes (medaka)

Character

Temperature

Salinity

4

P Oxygen

Author

Fin rays

Increase

Schmidt (1917, 1919)a

Vertebrae

V relation

Schmidt (1921)"

Vertebrae Fin rays Vertebrae Lateral scales Vertebrae Fin rays Vertebrae

V relation A-relation (asymmetrical) (?)Decrease Decrease V relation A relation Decrease

Vertebrae Fin rays Vertebrae

V relation V relation Decrease

Dannevig (1950P

Vertebrae Fin rays Vertebrae

V relation Increase V relation

Molander and Molander-Swedmark (1957) Itazawa (1959)

Myotomes Vertebrae Vertebrae Fin rays Vertebrae

Decrease Decrease V relation Decrease Decrease

References in Thing's (1952) review.

Decrease

T h i n g (1952)

(?)Decrease Mottley (1934, 1937)" Seymour (1959) Gabriel (1944) Lindsey (1954)

Increase

Hempel and Blaxter (1961)

See text

Lindsey (1962) Lindsey and Ali (1965)

%

i5

3.. H

H

z

E

238

J. H. S . BLAXTER

Of the various hypotheses put forward to explain meristic lability those of Gabriel (1944) and Barlow (1961) fit the facts best. Often, but not always, environmental factors which delay hatching produce higher counts, Alternatively, higher counts may be considered to come from nonoptimum conditions. It is likely that the environmental factors change the relationship between growth and differentiation. Hayes et al. (1953), in their studies of morphogenesis in salmon at temperatures from 1" to 15"C, certainly found that the relative appearance of certain anatomical characters varied with temperature. If differentiation is late more tissue is available to be differentiated, leading to a higher count. Thus one may consider an interaction of the Qlo for growth and the Qlo for differentiation of the character; as the temperature is varied so the relationship between growth and differentiation varies. Where V-shaped curves are obtained there are then points of inflection for the temperature coefficients. Garside and Fry ( 1959), using normal and reciprocal hybrid fry of speckled trout, Sahelinus fontinalis, and lake trout, S . namuycush, found that the mean myomere count was lower where the fish developed on the speckled trout yolks, which were smaller. There was also an inverse relationship between myomere count and the degree of twinning in Siamese twins of speckled trout, the counts being lower where there was less shared tissue. X. REARING AND FARMING

A. Techniques Three lines of approach are of interest at the present time: (1) Tropical and temperate fish culture for food ( 2 ) Rearing of salmonids and sturgeon to maintain population size after intensive fishing or hydroelectric schemes (3) Temperate marine fish rearing and farming

A good general review on fish farming by Iversen (1968) is available and on tropical species by Hickling (1962). Some of the most popular species are the common carp Cyprinus carpio, mullets, Tilapia, milkfish Chanos chanos, the carp Puntius jauanicus, and grass carp Ctenopharyngodun idella. Many of the species are reared in still water fertilized artificially or by sewage. The last four species mentioned are herbivorous which gives maximum efficiency of food utilization. Some of the species, e.g., Tilapia and Chams, can be kept in salt or brackish water. In the case of Tilapia and the Israeli carp, breeding takes place in the holding

4.

DEVELOPMENT: EGGS AND LARVAE

239

ponds; in the European carp special breeding ponds are used, On the other hand, the fry of Puntius, Chunos, Ctenopharyngodon, and mullet must be caught in the wild and transferred to rearing ponds. Generally in tropical culture there are no serious problems of feeding the young which are of sufficient size to take the food available. The extensive breeding of salmonids presents no problem of food supply for the young, which are relatively large when the plentiful yolk has been resorbed. Rearing is usually in running freshwater although acclimation of the young of Salmo species to marine conditions is now being developed, with improved growth and freedom from disease. The question of dietetics is well advanced in the group (see papers in The Progressive Fish Culturist). Marine fish rearing (see reviews by Morris, 1956, and Atz, 1964) and fanning (review by Shelbourne, 1964), although attempted from time to time over the last hundred years, have only recently made significant advances. Starting with the liberation of millions of plaice and cod eggs or yolk sac larvae in coastal areas, only the later experiments by Dannevig (1963) seem convincingly exploitable. Very recently, Mugil and Soba have been established as self-perpetuating species in some of the Mediterranean coastal lagoons. Following the earlier rearing in aquaria of some species to an advanced stage by highly empirical methods, a better experimental approach has now been adopted. Success in rearing Pleuronectes platessa, Solea soba, Microstomus kitt ( Shelbourne, 1964; unpublished data) , Clupea harengus (see Blaxter, 1968a; Rosenthal, unpublished data), Sardinops caeruba, Engraulis and Scomber ( Schumann, 1967; unpublished data), and Sardinu pilchardus (Blaxter, 1 9 6 9 ~ )has depended on some or all of the following factors: ( 1) Food supply. The use of Artemia and Balanus nauplii, Mytilus trochophores, young Tigriopus, small nematodes, rotifers and oligochaetes and secondary foods like Dunaliella has improved survival rates, so also has the extensive collection and selection of natural plankton ( 2 ) Antibiotics. Particularly with flatfish, dosages of mixed sodium penicillin (50 IU/ml) and streptomycin sulfate (0.05 mg/ml) have improved survival in the egg stage ( 3) Black-walled tanks and uniform overhead illumination have helped feeding success by providing contrast of the food against the background and even distribution of food in the tank ( 4 ) Tank hygiene

240

J. H. S. BLAXTER

There are interesting physiological and behavioral aspects of rearing where the fish may be artificially crowded and conditions of stress set up (see Shelboume, 1964; Riley and Thacker, 1963). Size hierarchy effects prevail (see Section IVYF), but growth seems to be density dependent only from the beginning of metamorphosis (Bowers, 1966a). Bitten fins are found more often in the smaller members of a tank community but are not always dependent on density. Perhaps more serious is the incidence of abnormal (albino or semi-albino) pigmentation in flatfish. It is generally worse in smaller fish and in very crowded tanks but may be reduced by using spawning stock acclimated to hatchery conditions ( Shelbourne, 1964; unpublished data). Survival to metamorphosis is also better in larvae from acclimated spawning stock ( Bowers, 1966b).

B. Sensory Deprivation Hatchery-reared flatfish do not survive well when transferred to the sea, presumably because of high predation; hatchery-reared salmonids usually have a poorer swimming performance. In fact, the hatchery environment by preventing contact from predation, by often rather uniform physical and chemical conditions and lack of shelter may prevent the development of appropriate defensive responses and muscular systems. The need for high density may also encourage undesirable intraspecific relationships at the expense of the more desirable interspecific responses. The more obvious shortcomings of the aquarium tank may be accompanied by sensory deprivation in a more general and less easily defhed way. The uniformity of the aquarium environment may lead to insufficient sensory input. For instance, Qasim (1959), when rearing the young of Blennius pholis, obtained better survival with a day-night regime than with continuous light. Blaxter ( 1968a), rearing herring, found a hint of better survival when the light conditions over the tank were continually changed and when air jets playing on the surface kept up a continuous series of ripples. Hoar (1942) found that young salmon would feed on chopped earthworm in darkness during the day, but not at night, perhaps because of an activity (or retinomotor) rhythm. Such rhythms might eventually be suppressed by continuous light. Activity rhythms with periods of rest might play as important a part in the development of the sense organs and nervous system as sufficient sensory input during the active phases.

4. DEVELOPMENT: EGGS AND

LARVAE

241

XI. CONCLUSIONS The most substantial advances in recent years have been made in investigations of marine eggs and larvae. Further work is needed on the following:

(1) Factors determining spawning time and fecundity, especially of marine fish ( 2 ) Permeability studies on the egg using isotope techniques, with associated studies on the fine structure of the membranes (3) The order in which different substrates in the yolk are utilized and the interrelations between the organic materials in the yolk and embryo being used for growth and maintenance ( 4 ) Differences in egg quality (of a subtle or less subtle nature) which may influence viability ( 5 ) Feeding mechanisms and nutrition in larvae with special reference to the possible utilization of small particles or dissolved matter ( 6 ) The respiratory and circulatory system in the larval stage where no functional gills or hemoglobin are present ( 7 ) Differences in the sensory systems of larvae and adults (8) The endocrine system in larvae, especially at metamorphosis ( 9 ) Metamorphosis itself, particularly changes in the blood, laying down of scales and pigment, and behavior (10) In rearing studies, so vital for investigations of larval stages and for techniques of fish culture, aspects of nutrition and stress need attention, especially the effect of tank conditions on growth, mortality, and the development of behavior ACKNOWLEDGMENTS The author is most grateful to Professor F. G. T. Holliday, University of Stirling, Dr. P. A. Orkin, University of Aberdeen, Dr. Sydney Smith, University of Cambridge, and to the editors, who read this chapter in d,raft. REFERENCES Alderdice, D. F., and Wickett, W. P. (1958).A note on the response of developing chum salmon eggs to free carbon dioxide in solution. J. Fisheries Res. Board Can.

15, 797-799.

Alderdice, D.F.,Wickett, W. P., and Brett, J. R. (1958).Some effects of temporary exposure to low dissolved oxygen levels on Pacific salmon eggs. J. Fisheries Res. Board Can. 15,229-249.

242

J. H.

S.

BLAXTER

Ali, M. A. ( 1959). The ocular structure, retinomotor and photobehavioural responses of juvenile Pacific salmon. Can. J. Zool. 3'7, 965-996. Amberson, W. R., and Armstrong, P. B. (1933). The respiratory metabolism of Fundulus heteroclitus during embryonic development. J. Cellular Comp. Physiol. 2, 381397, Amoroso, E. C. (1960). Viviparity in fishes, Symp. Zool. SOC. London 1, 153-181. Anokhina, L. E. (1960). Relationship between the fertility, fat content and the variations in the size of ova in the clupeid fishes. Tr. Soveshch. Ikhtiolog. Komis. Akad. Nauk SSSR 13, 290-295 (in Russian). Atz, J. W. (1964). A working bibliography on rearing larval marine fishes in the laboratory. U.S. Fish WiZdlife Sew. Bur. Sport Fishing Wildlife Res. Rept. 63, 95-102. Baerends, G. P., Bennema, B. E., and Vogelzang, A. A. (1960). Wber die h d e r u n g der Sehscharfe mit dem Wachstum bei Aeguidens portalegrensis (Hensel) (Pisces, Cichlidae). 2001.Jahrb. 88, 67-78. Bagenal, T. B. (1968). The ecological and geographical aspects of the fecundity of plaice. I . Marine Biol. Assoc. U.K. 46, 161-186. Barlow, G . W. (1961). Causes and significance of morphological variation in fishes. Systemutic Zool. 10, 105-117. Barrett, I. (1951). Fertility of salmonid eggs and sperm after storage. J . Fisheries Res. Board Can. 8, 125-133. Barrington, E. J. W. (1961). Metamorphic processes in fishes and lampreys. Am. Zoobgist 1, 97-106. Battle, H. I. (1930). Effects of extreme temperatures and salinities on the development of Enchlyopus cimbrius (L.). Contrib. Can. Biol. Fisheries [N.S.] 5, 109-192. Battle, H. I. ( 1944). The embryology of the Atlantic salmon (Salmo salur Linnaeus). Can. J. Res. D22, 105-125. Baxter, I. G. (1959). Fecundity of winter-spring and summer-autumn herring spawners. 1. Conseil Conseil Perm. Intern. Exploration Mer 25, 73-80. Bell, G. M., and Hoar, W. S. (1950). Some effects of ultra-violet radiation on sockeye salmon eggs and alevins. Can. J . Res. D28, 3543. Bishai, H. M. (196Oa). The effect of gas content of water on larval and young fish. z. Wks. Zool. 163, 37-64. Bishai, H. M. (198ob). Upper lethal temperatures for larval salmonids. J. Consei2 Conseil Perm. Intern. Explorution Me? 25, 129-133. Bishai, H. M. ( 1 9 6 0 ~ )The . effect of hydrogen ion concentration on the survival and distribution of larval and young fish. Z . Wiss. Zool. 164, 107-118. Bishai, H. M. (196od). The effect of water currents on the survival and distribution of fish larvae. 1. Conseil Conseil Perm. Intern. Exploration Mer 25, 134-146. Bishai, H. M. (1961). The effect of pressure on the survival and distribution of larval and young fish. J . Conseil Conseil Perm. Intern. Exploration Mer 26, 292-31 1. Blaxter, J. H. S. ( 1953). Sperm storage and cross fertilization of spring and autumn spawning herring. Nature 172, 1189. Blaxter, J. H. S. (1955). Herring rearing. I. The storage of herring gametes. Marine Res. Scotland No. 3, 1-12. Blaxter, J. H. S. (1956). Herring rearing. 11. The effect of temperature and other factors on development. Marine Res. Scotland No. 5, 1-19.

4. DEVELOPMENT:

EGGS AND LARVAE

243

Blaxter, J. H. S. (1960). The effect of extremes of temperature on hemng larvae. J. Marine Biol. Assoc. U . K . 39, 605-608. Blaxter, J. H. S. (1962). Herring rearing. IV. Rearing beyond the yolk-sac stage. Marine Res. Scotland No. 1, 1-18. Blaxter, J. H. S. (1965). The feeding of herring larvae and their ecology in relation to feeding, Rept. Calif. Coop. Oceanog. Fisheries Inoest. 10, 79-88. Blaxter, J. H. S. (1966). The effect of light intensity on the feeding ecology of herring. S y m p . Brit. Ecol. SOC. 6, 393409. Blaxter, J. H. S. (1968a). Rearing herring larvae to metamorphosis and beyond. J. Marine Biol. Assoc. U.K. 48, 17-28. Blaxter, J. H. S. (1968b). Light intensity, vision and feeding in young plaice. J. Exptl. Marine Biol. Ecol. 2, 293-307. . thresholds and spectral sensitivity of herring larvae. Blaxter, J. H. S. ( 1 9 6 8 ~ )Visual J. Exptl. Biol. 48, 39-53. Blaxter, J. H. S. (1969a). Feeding and condition of Clyde herring larvae. Rapp Proces-Verbaux Reunions, Conseil Perm. Intern. Exphation Mer (in press ) . Blaxter, J. H. S. (1969b). Visual thresholds and spectral sensitivity of flatfish larvae. J. Exptl. Biol. 51, 221-230. . rearing of pilchard larvae, Sardina pilchardus. Blaxter, J. H. S. ( 1 9 6 9 ~ )Experimental J. Marine Biol. Assoc. U.K. 49, (in press). Blaxter, J. H. S., and Hempel, G. (1963). The influence of egg size on herring larvae (Clupea hurengus L.). J. Conseil Conseil Perm. Intern. Exploration Mer 28, 211-240. Blaxter, J. H. S., and Hempel, G. (1966). Utilization of yolk by herring larvae. J. Marine Biol. Assoc. U . K . 461, 219-234. Blaxter, J. H. S., and Jones, M. P. (1967). The development of the retina and retinomotor responses in the herring. J. Marine Biol. Assoc. U . K . 47, 677-697. Blaxter, J. H. S., and Staims, M. (1969a). Food searching potential in marine fish larvae. Proc. 4th European Marine Biol. Symp. (in press). Blaxter, J. H. S., and Staines, M. (196913). The pure-cone retina and retinomotor responses in teleost larvae. J . Marine Biol. Assoc. U.K. (submitted for publication), Bowers, A. B. (1966a). Growth in hatchery-reared plaice. Rept. Challenger SOC. 3, No. 18. Bowers, A. B. (1966b). Marine fish culture in Britain. VI. The effect of the acclimatization of adult plaice to pond conditions on the viability of eggs and larvae. J. Conseil Conseil Perm. Intern. Exploration Mer 30, 196-203. Braum, E. ( 1964 ). Experimentelle Untersuchungen zur ersten Nahrungsaufnahme und Biologie an Jungfischen von Blaufelchen (Coregonus wartmunni Bloch), Weissfelchen (Coregonus fera Jurine) und Hechten (Esox lucius L.). Arch. Hydrobiol. 28, 2/3, 183-244. Breder, C. M., Jr., and Rosen, D. R. (1966). “Modes of Reproduction in Fishes.” Natural History Press, Garden City, New York. Brett, J. R. ( 1952). Temperature tolerance in young Pacific salmon Genus Oncorhynchus. 1. Fisheries Res. Board Can. 9, 265-323. Brett, J. R., and Alderdice, D. F. (1958). The resistance of cultured young chum and sockeye salmon to temperatures below 0°C. J. Fisheries Res. Board Can. 15, 805-813. Brett, J. R., and Groot, C. (1963). Some aspects of olfactory and visual responses in Pacific salmon. I. Fisheries Res. Board Can. 20, 287403.

244

J. H. S. BLAXTER

Bridger, J, P. (1956). On day and night variation in catches of fish larvae. 3. Conseil Conseil Perm. Intern. Exploration Mer 22, 42-57. Colton, J. B., Jr. (1965). The distribution and behavior of pelagic and early demersal stages of haddock in ,relation to sampling techniques. Spec. Publ. Intern. Comm. Northwest Atlantic Fisheries 6, 317433. Combs, B. D. (1965). Effect of temperature on development of salmon eggs. Progressive Fish Culturist 27, 134-137. D’Ancona, U., Sanzo, L., Sparta, A,, Bertolini, F., Ranzi, S., and Montalenti, G. (1931-1933). Uova, larve e stadi giovanili di Teleostei. “Fauna & Flora del Golfo di Napoli,” Monograph 38, 4 parts. Publ. Naples Zool. Sta. Dannevig, A. ( 1963). Artificial propagation of cod. Rept. Norwegian Fishery Marine Invest. 13, 73-79. Daykin, P. N. (1965). Application of mass transfer theory to the problem of respiration of fish eggs. 3. Fisheries Res. Board Can. 22, 159-171. Dean, B. (1916). “A Bibliography of Fishes,” 3 vols. New York Museum, Massachusetts Univ. Press, Cambridge, Massachusetts. De Ciechomski, J. Dz. (1965). The development of fish embryos in a non-aqueous medium. Acta Embryo1 Morphol. Exptl. 8, 183-188. De Ciechomski, J. Dz. (1967a). Influence of some environmental factors upon embryonic development of the Argentine anchovy Engraulis anchoita ( Hubbs, Marini). Rept. Calif. Coop. Oceanog. Fisheries Inuest. 11, 67-71. De Ciechomski, J. Dz. (1967b). Investigations of food and feeding habits of larvae and juveniles of the Argentine anchovy Engraulis anchoita. Rept. Calif. Coop. Oceanog. Fisheries Invest. 11, 72-81. Deuchar, E. M. (1965). Biochemical patterns in early developmental stages of vertebrates. In “The Biochemistry of Animal Development” (R. Weber, ed.), Vol. 1, pp. 258-263. Academic Press, New York. Devillers, C. (1961). Structural and dynamic aspects of the development of the teleostean egg. Advan. Morphogenesis 1, 379-428. Devillers, C. (1965). Respiration et morphogenAse dans I’oeuf des T&ost&ns. Ann. Biol. 4, 157-186. Dimond, M. T. (1963). The relation of whole body I‘31 uptake to thyroid activity in the developing dogfish Scyliorhinus canicula (L.). Biol. Bull. 124, 170-181. Ehrenbaum, E. (1909). “Eier und Larven von Fischen des Nordischen Planktons.” Lipsius & Tischer, Kiel and Leipzig. Eider, R. ( 1961). Effects of visible radiation on salmonid embryos and larvae. Growth 25, 281-346. Farris, D. A. (1959). A change in the early growth rates of four larval marine fishes. Limnol. Oceanog. 4, 24-36. Fliichter, J. ( 1962 ). Funktionsanatomische Untersuchungen am Kieferapparat der Heringslarven. Kurze M i f f . Inst. Fischereibiol. Uniu. Hamburg 12, 1-12. Fliichter, J., and Rosenthal, H. (1965). Beobachtungen iiber das Vorkommen and Laichen des Blauen Wittlings ( Micromesistius poutassou Risso) in der Deutschen Bucht. Helgolaender Wiss. Meeresuntersuch. 12, 149-155. Forrester, C. R., and Alderdice, D. F. (1966). Effects of salinity and temperature on embryonic development of the Pacific cod ( Cadus mucrocephalus) , 3. Fisheries Res. Board Can. 23, 319-340. Gabriel, M. L. (1944). Factors affecting the number and form of vertebrae in Fundulus heteroclitus. 3. Exptl. Zool. 95, 105-147.

4. DEVELOPMENT:

EGGS AND LARVAE

245

Garside, E. T. ( 1959). Some effects of oxygen in relation to temperature on the development of lake trout embryos, Can. J. 2001.D37, 689-698. Garside, E. T. (1966). Effects of oxygen in relation to temperature on the development of embryos of brook trout and rainbow trout. J. Fisheries Res. Board Can. 23, 1121-1134. Garside, E. T., and Fry, F. E. J. (1959). A possible relation between yolk size and differentiation in trout embryos. Can. J. Zool. D37, 38-86. Ginsburg, A. A. (1961). The block to polyspermy in sturgeon and trout with special reference to the role of cortical granules (alveoli). J. Embryol. Exptl. Morphol. 9, 173-190. Gray, J. (1926). The growth of fish. I. The relationship between embryo and yolk in Salmo fario. J. Exptl. Biol. 4, 215-225. Gray, J. (1928a). The growth of fish. 11. The growth-rate of the embryo of Salmo fario. J. Exptl. Biot. 6, 110-124. Gray, J. (192813). The growth of fish. 111. The effect of temperature on the development of the eggs of Salmo fario. J . Exptl. B i d . 6, 125-130. Hamdorf, K. ( 1960). Die Beeinflussung der Embryonal-und Larvalentwicklung der Regenbogenforelle ( Salmo irideus Gibb. ) durch Strahlung im sichtbaren Bereich. Z. Vergleich. Physiol. 42, 525-565. Harder, W. ( 1954). Die Entwicklung der Respirationsorgane beim Hering (Clupea harengus L. ), Z. Anat. Entwicklungsgeschichte 118, 102-123. Harder, W. ( 1960). Vergleichende Untersuchungen zur Morphologie des Darmes bei Clupeoidea. Z. Wks. 2001.163, 65-167. Hayes, F. R. (1949). The growth, general chemistry and temperature relations of salmonid eggs. Quart. Rev. Biol. 2 4 281-308. Hayes, F. R., and Pelluet, D. (1945). The effect of temperature on the growth and efficiency of yolk conversion in the salmon embryo. Can. J. Res. D23, 7-15. Hayes, F. R., Pelluet, D., and Gorham, E. (1953). Some effects of temperature on the embryonic development of the salmon (Salmo salar). Can. J . Zool. D31, 42-51. Hayes, F. R., Wilmot, I. R., and Livingstone, D. A. (1951). The oxygen consumption of the salmon egg in relation to development and activity. J. Exptl. 2002.116, 377495. Heard, W. R. (1964). Phototactic behaviour of emerging sockeye salmon fry. Animal Behauiour 12, 382-388. Hempel, G., and Blaxter, J. H. S . (1961). The experimental modification of meristic characters in herring (Clupea harengus L. ) , 1. Consdl Conseil Perm. Intern. Exploration Mer 26, 336-346. Hempel, G., and Blaxter, J. H. S . (1963). On the condition of herring larvae. Rapp Proces-Verbaux Reunions, Conseil Perm. Intern. Exploration Mer 154, 35-40. Hempel, G., and Blaxter, J. H. S. (1967). Egg weight in Atlantic herring (Clupea harengus L.). J . Conseil Conseil Perm. Intern. Exploration Mer 31, 170-195. Hester, F. J. (1963). Effects of food supply on fecundity in the female guppy Lebistes reticulatus (Peters). J. Fisheries Res. Board Can. 21, 757-764. Heuts, M. J. (1949). Racial divergence in fin ray variation patterns in Gasterosteus aculeatus. J. Genet. 49, 183-191. Hickling, C. F. (1962). “Fish Culture.” Faber & Faber, London. Hishida, T., and Nakano, E. (1954). Respiratory metabolism during f%hdevelopment. Embyologia (Nagoya) 2, 67-79.

246

J. H. S. BLAXTER

Hoar, W. S. (1942). Diurnal variations in feeding activity of young salmon and trout. J . Fisheries Res. Board Can. 6, 90-101. Hoar, W. S. (1956). The behaviour of migrating pink and chum salmon fry. J. Fisheries Res. Board Can. 13, 309-325. Hoar, W. S. (1958). The evolution of migratory behaviour among juvenile salmon of the genus Oncorhynchus. I. Fisheries Res. Board Can. 15, 391-428. Holliday, F. G. T., and Blaxter, J. H. S. (1960). The effects of salinity on the developing eggs and larvae of the herring. 1. Marine Biol. Assoc. U.K. 39, 591-603. Holliday, F. G. T., and Jones, M. P. (1965). Osmotic regulation in the embryo of the herring (Clupea harengus). J. Marine Biol. Assoc. U.K. 45, 305-311. Holliday, F. G. T., and Jones, M. P. (1967). Some effects of salinity on the developing eggs and larvae of the plaice (Pleuronectes platessa). J. Marine Biol. ASSOC. U.K.47, 39-48. Holliday, F. G. T., Blaxter, J. H. S., and Lasker, R. (1W).Oxygen uptake of developing eggs and larvae of the herring (Clupeu harengus). J. Marine B i d . ASSOC. U.K. 44, 711-723. Horton, H. F., Graybill, J. R., and Wu, A. S. H. ( 1967). “Cryogenic Preservation of Viable Fish Sperm.” Dept. Fisheries and Wildlife, Oregon State University, Corvallis, Oregon. Hoyle, R. J., and Idler, D. R. (1968). Preliminary results in the fertilization of eggs with frozen sperm of Atlantic salmon (Salmo salar). J. Fisheries Res. Board Can. 25, 1295-1297. Hubbs, C., Stevenson, M. M., and Peden, A. E. (1968). Fecundity and egg size in two central Texas darter populations. southwestern Naturalist 13, 301424. Itazawa, Y. (1959). Influence of temperature on the number of vertebrae in fish. Nature 183, 1408-1409. Iversen, E. S. (1968). “Farming the Edge of the Sea.” Fishing News (Books) Ltd., London, Ivlev, V. S. (1960a). Active metabolic intensity in salmon fry (Salmo salar L.) at various rates of activity. Paper No. 213 (mimeo). Salmon & Trout Committee, Intern. Council for Exploration of the Sea, Copenhagen. Ivlev, V. S. (1960b). On the utilization of food by planktonophage bhes. Bull. Math. Biophys. 22, 371-389. Ivlev, V. S. (1961). “Experimental Ecology of the Feeding of Fishes.” Translation by Yale Univ. Press, New Haven, Connecticut. Iwai, T. ( 1964). Feeding and ciliary conveying mechanisms in larvae of salmonid fish Plecoglossus altivelis. Temminck et Schlegel. Physiol. Ecol. ( K y o t o ) 12, 38-44. Iwai, T. (1967a). Structure and development of lateral line cupulae in teleost larvae. In “Lateral Line Detectors” (P. Cahn, ed.), pp. 2 7 4 . Indiana Univ. Press, Bloomington, Indiana. Iwai, T. (1967b). The comparative study of the digestive tract of teleost larvae. I. Fine structure of the gut epithelium in larvae of Ayu. Bull. Japan. SOC. Sci. Fisheries 33, 480-496. . comparative study of the digestive tract of teleost larvae. 11. Iwai, T. ( 1 9 6 7 ~ ) The Ciliated cells of the gut epithelium in pond smelt larvae. Bull. Japan. SOC. Sci. Fisheries 33, 1116-1119. Jacobs, W. ( 1938). Untersuchungen zur Physiologie der Schwimmblase der Fische. IV. Die erste Gasfiillung der Schwimmblase bei jungen Seepferdchen. Z . Vergleich. Physiol. 25, 379-388.

4.

DEVELOPMENT: EGGS AND LARVAE

247

John, K. R., and Hasler, A. D. (1956). Observations on some factors affecting the hatching of eggs and the survival of young shallow-water cisco Leucichthys artedi Le Sueur, in Lake Mendota, Wisconsin. Limnol. Oceanog. 1, 176-194. Jones, M. P., Holliday, F. G. T., and Dunn, A. E. G. (1966). The ultra-structure of the epidermis of larvae of the herring (CZupea harengus) in relation to the rearing salinity. J. Marine Biol. Assoc. U . K . 46, 235-239. Kaighn, M. E. (1964). A biochemical study of the hatching process in Fundulus heteroclitus. Develop. Biol. 9, 56-80. Kalman, S. M. (1959). Sodium and water exchange in the trout egg. J. Cellular Comp. Physiol. 54, 155-162. Kandler, R.,and Dutt, S . (1958). Fecundity of Baltic herring. Rapp. Proces-Verbaux Reunions, Conseil Perm. Intern. Exploration Mer 143, 99-108. Keenleyside, M. H. A,, and Hoar, W. S. (1954). Effects of temperature on the responses of young salmon to water currents. Behauiour 7, 77-87. Kinne, O., and Kinne, E. M. (1962). Rates of development in embryos of a cyprinodont fish exposed to different temperature-salinity-oxygen combinations. Can. J. ZOO^. W, 231-253. Kurata, H. (1959). Preliminary report on the rearing of the herring larvae. Bull. Hokkaido Reg. Fisheries Res. Lab. 20, 117-138. Kuthalingam, M. D. K. (1959). Temperature tolerance of the larvae of ten species of marine fishes. Current Sci. (India) 28, 75-76. Laale, H. W., and McCallion, D. J. (1968). Reversible developmental arrest in the embryo of the Zebra fish Bruchydunio rerio. J. Exptl. Zool. 167, 117-123. Lagler, K. F., Bardach, J. E., and Miller, R. R. (1962). “Ichthyology.” Wiley, New York. Larimore, R. W., and Duever, M. J. (1968). Effects of temperature acclimation on the swimming ability of smallmouth bass fry. Trans. Am. Fisheries SOC. 97, 175-1 84. Lasker, R. (1962). Efficiency and rate of yolk utilization by developing embryos and larvae of the Pacific sardine Sardinops cuerulea (Girard). J. Fisheries Res. Board Can. 19, 867-875. Lasker, R. ( 1964). An experimental study of the effect of temperature on the incubation time, development and growth of Pacific sardine embryos and larvae. Copeia No. 2, 399-405. Lasker, R., and Theilacker, G. H. (1962). Oxygen consumption and osmoregulation by single Pacific sardine eggs and larvae (Sardinops caewlea Girard). J. Conseil Conseil Perm. Intern. Exploration Mer 27, 25-33. Lasker, R., and Threadgold, L. T. (1968). “Chloride cells” in the skin of the larval sardine. Erptl. Cell. Res. 52, 582-590. Lewis, R. M. (1965). The effect of minimum temperature on the survival of larval Atlantic menhaden Breuoortia tyrannus. Trans. Am. Fisheries SOC.94, 409-412. Lewis, R. M. (1966). Effects of salinity and temperature on survival and development of larval Atlantic menhaden Breuoortia tyrannus. Trans. Am. Fisheries SOC. 95, 423-426. Lillelund, K. ( 1961 ). Untersuchungen iiber die Biologie und Populations-dynamik des Stintes, Osmerus eperlanus eperlanus ( Linnaeus 1758) der Elbe. Arch. Fischereiwissenschaft 12, 1-128. Lindroth, A. ( 1942). Sauerstoffverbrauch der Fische. 11. Verschiedene Entwicklungs-und Alterstadien vom Lachs und Hecht. 2. Vergleich. Physiol. 29, 583594.

248

J. H. S. BLAXTER

Lindsey, C. C. ( 1954 ) . Temperature-controlled meristic variation in the paradise fish. Macropodus opercularis ( L . ) . Can. J. 2001. D32, 87-98. Lindsey, C. C. (1958). Modification of meristic characters by light duration in kokanee Oncorhynchus nerka. Copeia No. 2, 134-136. Lindsey, C. C. ( 1W2). Experimental study of meristic variation in a population of three spine sticklebacks Gmterosteus aculeatus. Can. 1. Zool. D40, 2 7 1 3 1 2 . Lindsey, C. C., and Ali, M. Y. (1965). The effect of alternating temperature on vertebral count in the medaka (Oryzias latipes). Can. J. 2001.D43, 99-104. Lyall, A. H. (1957a). The growth of the trout retina. Quart. J. Microscop. Sci. 98, 101-110. Lyall, A. H. (195713). Cone arrangements in teleost retinae. Quart. J. Microscop. Sci. 98, 189-201. MacGregor, J. S. ( 1957). Fecundity of the Pacific sardine (Sardinops caeruleu). U S . Fish Wildlife Sem., Fishery Bull. 57, 427449. McHugh, J. L. (1954). The influence of light on the number of vertebrae in the grunion Leuresthes tenuis. Copeia No. 1, 23-25. Magnuson, J. J. (1962). An analysis of aggressive behaviour, growth and competition for food and space in medaka (Oryzias lutipes (Pisces, Cyprinodontidae) ). Can. J. ZOO^. D40, 313-363. Mansueti, R. (1954). A partial bibliography of fish eggs, larvae and juveniles with particular reference to migratory and estuarine species of the Atlantic coast and supplemented by a check list and references to the early development of the fishes and fish-like chordates of Maryland waters. Chesapeake Biol. Lab. ( mimeo), Maryland Department of Research and Education. Mansueti, R. (1958). Eggs, larvae and young of the striped bass Roccus smatilis. Contrib. Chesapeake Biol. Lab. 112, 1-35. Marinaro, J. Y., and Bernard, M. (1966). Contribution B I’6tude des oeufs et larves pelagiques de poissons mhditerran6ens. Pelagos 6, 49-55. Marr, D. H. A. (1965). Factors affecting the growth of salmon alevins and their survival and growth during the fry stage. Yearbook Assoc. River Authorities pp. 133-14 1. Marr, D. H. A. (1966). Influence of temperature on the efficiency of growth of salmonid embryos. Nature 212, 957-959. Marr, D. H. A. (1967). Experiments on the artificial rearing of young Atlantic salmon (Salmo salar L.). Anadramous and Catadramous Fish Committee, Paper No. 25 (mimeo).Intern. Council for Exploration of the Sea, Copenhagen. Marr, J. C. (1956). The “critical period” in the early life history of marine fishes, J. Conseil Conseil Perm. Intern. Exploration Mer 21, 16CL170. Marshall, N. B. (1965). “The Life of Fishes.” Weidenfeld & Nicolson, London. Molander, A. R.,and Molander-Swedmark, M. ( 1957). Experimental investigation on variation in plaice (Pleuronectes platessa LinnB). Inst. Marine Res., Lysekll, Ser. Biol. Rept. 7, 1-45. Monroy, A., Ishida, M., and Nakano, E. (1961). The pattern of transfer of the yolk material to the embryo during the development of the teleostean fish Oryzias latipes. Embryologia (Nagoya) 6, 151-158. Morns, R. W. (1955). Some considerations regarding the nutrition of marine fish larvae. J. Conseil Conseil Perm. Intern. Exploration Mer. 20, 255-265. Morris, R. W. (1956). Some aspects of the problem of rearing marine fishes. Bull. Inst. Oceanog. 1082, 1-61.

4.

DEVELOPMENT: EGGS AND LARVAE

249

Mounib, M. S., and Eisan, J. 5. (1969). Metabolism of pyruvate and glyoxylate by eggs of salmon (Salmo salar). Comp. Biochem. Physiol. 29, 259-264. Mounib, M. S., Hwang, P. C., and Idler, D. R. (1968). Cryogenic preservation of Atlantic cod (Gadus morhua) sperm. J . Fisheries Res. Board Can. 25,2623-2632. Miiller, H. (1952). Bau and Wachstum der Netzhaut des Guppy (Lebistes reticulatus). 2001.Jahrb. 63, 275324. Nakano, E. (1953). Respiration during maturation and at fertilization of fish eggs. Embryologia (Nagoya) 9, 21-31. Needham, J. ( 1931), “Chemical Embryology,” 3 vols. Cambridge Univ. Press, London and New York. New, D. A. T. (1966). “The Culture of Vertebrate Embryos.” Lagos P,ress, London. Neyfakh, A. A. (1959). X-ray inactivation of nuclei as a method for studying their function in the early development of fishes. I . Embryol Exptl. Morphol. 7 , 173192.

Nikolsky, G. V. (1963). “The Ecology of Fishes.” Academic Press, New York. Norman, J. R. (1963). “A History of Fishes” (P. H. Greenwood, ed.), 2nd ed. Ernest Benn, London. Ohtsuka, E. (1960). On the hardening of the chorion of the fish egg after fertilization. 111. The mechanism of chorion hardening in Oryzias latipes. Biol. Bull. 118, 120128.

Oppenheimer, J. M. (1947). Organization of the teleost blastoderm. Quart. Reu. Biol. 22, 105-118. Oppenheimer, J. M. (1964). The development of isolated Fundulus embryonic shields in salt solution. Acta Embryol. Morphol. Exptl. 7, 143-154. Orton, G. (1957). Embryology and evolution of the pelagic fish egg. Copeia No. 1, 5R58.

Parrish, B. B. ( 1956). The cod, haddock and hake. I n “Sea Fisheries” (M. Graham, ed.), pp. 251-331. Arnold, London. Parrish, B. B., and Saville, A. (1965). The biology of the north-east Atlantic herring populations. Oceanog. Marine Biol. Ann. Rev. 3, 323-373. Pearcy, W. G. (1962). Distribution and origin of demersal eggs within the order Pleuronectiformes. J. Conseil Conseil Perm. Intern. Exploration Mer 27, 232c235. Peters, H. M. (1963). Eizahl, Eigewicht und Gelegeentwicklung in der Gattung Tilapia ( Cichlidae, Teleostei ). Intern. Reu. Ces. Hydrobiol. 48, 547-576. Phillips, A. M., Jr., and Dumas, R. F. (1959). The chemistry of developing brown trout eyed eggs and sac fry. Progressive Fish Culturist 91, 161-164. Pickford, G. E., and Atz, J. W. (1957). “The Physiology of the Pituitary Gland of Fishes.” N.Y.Zool. SOC., New York. Pope, J. A., Mills, D. H. and Shearer, W. M. (1961). The fecundity of Atlantic salmon (Salmo solar Linn) . Freshwater Salmon Fisheries Res., Scotland 28, 1-12.

Potts, W. T. W., and Rudy, P. P., Jr. (1969). Water balance in the eggs of the Atlantic salmon Salmo salar. J. Exptl. Biol. 50, 223-237. Prescott, D. M. (1955). Effect of activation on the water permeability of salmon eggs. J. Cellular Comp. Physiol. 45, 1-12. Price, J. W. (1940). Time-temperature relations in the incubation of the whitefish Coregonus clupeafonnis (Mitchill). J. Gen. Physiol. 23, 449468. Qasim, S. Z. (1959). Laboratory experiments on some factors affecting the survival of marine teleost larvae. J . Marine Biol. Assoc. India. 1, 13-25. Qasim, S. Z., Rice, A. L., and Knight-Jones, E. W. (1963). Sensitivity to pressure

250

J. H. S. BLAXTER

changes in teleosts lacking swim-bladders. 1. Marine Biol. Assoc. India 5, 28%293. Radakov, D. V. (1964). “Velocities of Fish Swimming,” pp. 4-28. Pamphlet from A. N. Severtsov Institute of Animal Morphology, Moscow, Nauka (in Russian). Raitt, D. F. S., and Hall, W. B. (1967). On the fecundity of the redfish Sebastes marinus (L.). J. Conseil Conseil Perm, Intern. Exploration Mer 31, 237-245. Randall, J. E. (1961). A contribution to the biology of the convict surgeonfish of the Hawaiian Islands, Acanthurus triostegus sanduicensis. Pacific Sci. 15, 215-272. Read, L. J. (1968). Omithine-urea cycle enzymes in early embryos of the dogfish Sqzba1u.s suckleyi and the skate Rufa binoculata. Comp. Biochem. Physbl. 24, 669-674. Riley, J. D., and Thacker, G. T. (1963). Marine fish culture in Britain. 111. Plaice (Pleuronectes platessa ( L . ) ) rearing in closed circulation at Lowestoft 1961. J . Conseil Conseil Perm. Intern. Exploration Mer 28, 80-90. Rosenthal, H. ( 1966). Beobachtungen uber das Verhalten der Seezungenbrut. Helgobender Wiss. Meeresuntersuch. 13, 213-228. Rosenthal, H. ( 1!368). Beobachtungen uber die Entwicklung des Schwarmverhaltens bei den Larven des Herings Clupea harengus. Marine Biol. 2, 73-76. Rosenthal, H., and Hempel, G. ( 1968). Experimental Studies in Feeding and Food Requirements of Herring Larvae. Symposium on Food Chains. Univ. of Aarhus, Denmark. Rounsefell, G. A. (1957). Fecundity of the North American Salmonidae. U.S. Fish Wildlife Seru., Fishery Bull. 57( 122), 451468. Rudnick, D. (1955). Teleosts and birds. In “Analysis of Development” ( B . H. Willier, P. A. Weiss, and V. Hamburger, ed. ), Saunders, Philadelphia, Pennsylvania. Rudy, P. P., Jr., and Potts, W. T. W. (1969). Sodium balance in the eggs of the Atlantic salmon Salmo sahr. 1. Exptl. Biol. 50, 239-246. Ryland, J. S. (1963). The swimming speeds of plaice larvae. J. Exptl. Biol. 40, 285299. Ryland, J. S. (1964). The feeding of plaice and sand-eel larvae in the southern North Sea. J. Marine Biol. Assoc. U . K . 44, 343-364. Ryland, J. S. (1966). Observations on the development of larvae of the plaice, Pleuronectes platessa L. in aquaria. I. Conseil Conseil Perm. Intern. Exploration Mer 30, 177-195. Ryland, J. S., and Nichols, J. H. (1967). Effect of temperature on the efficiency of growth of plaice proiarvae. Nature 214, 529-530. Schumann, G. 0. (1967). La Jolla first to raise mackerel and sardines to advanced juvenile stage. Corn. Fisheries Rev. 29( 2 ) , 24-25. Schwassmann, H. 0. ( 1965). Functional development of visual pathways in larval sardines and anchovies. Rept. Calif. Coop. Oceanog. Fisheries Inuest. 10, 64-70. Scott, C. G., and Kellicott, W. E. (1916). The consumption of oxygen during the development of Fundulus heteroclitus. Anat. Record 11, 531-533. Scott, D. P. (1961). Effect of food quantity on fecundity of rainbow trout Salmo gairdneri. 1. Fisheries Res. Board Can. 19, 715-731. Sette, 0. ( 1943). Biology of the Atlantic mackerel (Scomber scombrus) of North America. Part 1. Early life history, including the growth, drift and mortality of the egg and larval populations. US.Fish Wildlife Seru., Fishery Bull. 50(38), 149-237.

4.

DEVELOPMENT: EGGS AND LARVAE

251

Seymour, A. (1959). Effects of temperature upon the formation of vertebrae and fin rays in young chinook salmon. Trans. Am. Fisheries SOC.88, 58-69. Shaw, E. (1960). The development of schooling behaviour in fishes. Physiol. 2001. 33, 7 W 6 . Shaw, E. (1961). The development of schooling in fishes, 11. Physiol. Zool. 34, 263272. Shelbourne, J. E. (1953). The feeding habits of plaice post-larvae in the Southern Bight, J. Marine Biol. Assoc. U.K. 32, 149L160. Shelboume, J. E. (1956). The effect of water conservation on the structure of marine fish embryos and larvae. 1. Marine Biol. Assoc. U.K.35, 275-286. Shelbourne, J. E. (1957). The feeding and condition of plaice larvae in good and bad plankton patches. J. Marine Biol. Assoc. U.K. 36, 539-552. Shelboume, J. E. ( 1963). Marine fish culture in Britain. 11. A plaice rearing experiment at Port Erin, Isle of Man, during 1960 in open sea water circulation. J. Conseil Conseil Perm. Intern. Exploration Mer 28, 70-79. Shelbourne, J. E. (1964). The artificial propagation of marine fish. Aduan. Marine Biol. 2, 1-83. Shumway, D. L., Warren, C. E., and Doudoroff, P. (1964). Influence of oxygen concentration and water movement on the growth of steelhead trout and coho salmon embryos. Trans. Am. Fisheries SOC.93, 342-356. Silver, S. J., Warren, C. E. and Doudoroff, P. (1963). Dissolved oxygen requirements of developing steelhead trout and chinook salmon embryos at different water velocities. Trans. Am. Fisheries SOC. 92, 327-343. Simpson, A. C. (1956). The pelagic phase. In “Sea Fisheries” (M. Graham, ed.), pp. 207-250. Arnold, London. Smith, S. (1957). Early development and hatching. In “The Physiology of Fishes” ( M . E. Brown, ed.), Vol. 1, pp. 323-359. Academic Press, New York. Smith, S. (1958). 111. Yolk utilization in fishes. In “Embryonic Nutrition” ( D . Rudnick, ed.), pp. 3353. Univ. of Chicago Press, Chicago, Illinois. Sneed, K. E., and Clemens, H. P. (1956). Survival of fish sperm after freezing and storage at low temperatures. Progressiue Fish Culturist I S , 99-103. Solberg, A. N. (1938). The susceptibility of Fundulus heteroclitus embryos to Xradiation, J . Exptl. 2001.78, 441-469. Solemdal, P. (1967). The effect of salinity on buoyancy, size and development of flounder eggs. Sarsia 29, 431442. Stevenson, J. C. ( 1962). Distribution and survival of herring larvae (Clupea palkzsii Valenciennes) in British Columbia waters. J. Fisheries Res. Board Can. 19, 735-810. Strasburg, D. W. (1960). Estimates of larval tuna abundance in the Central Pacific. US.Fish Wildlife Seru. Fishery Bull. 60, 231-255. Stuart, T. A. (1953). Spawning migration, reproduction and young stages of loch trout (Salmo trutta L.). Freshwater Salmon Fisheries Res., Scotland No. 5, 1-39. Sweet, J. G., and Kinne, 0. (1964). The effects of various temperature-salinity combinations on the body form of newly hatched Cyprinodon macularius (Teleostei). Helgolaender Wiss. Meeresuntersuch. 11, 49-69. Tait, J. S. (1960). The first filling of the swim bladder in salmonids. Can. 1. Zool. D38, 179-187. Talbot, G. E., and Sykes, J. E. (1958). Atlantic coast migration of American shad. US.Fish Wildlife Serv., Fishery Bull. 58( 142), 473-490. Takahashi, M., and Hatanaka, M. A. (1960). Experimental study on utilization of

252

J. H. S. BLAXTER

food by young anchovy, Engraulis japonicus Temminck et Schlegel. Tohoku J. Agr. Res. 11, 161-170. Tgning, A. V. (1952). Experimental study of meristic characters in fishes. Btol. Rev. Cambridge Phil. SOC. 27, 169-193. Templeman, W. (1944). The life history of the spur dogfish Squaltls ucanthias and the Vitamin A values of dogfish liver oil. Res. Bull. Div. Fishery Resources, Newfoundland No. 15, 1-102. Terner, C. ( 1968). Studies of metabolism in embryonic development. I. The oxidative metabolism of unfertilized and embryonated eggs of the rainbow trout. Comp. Biochem. Physiol. 24, 933-940. Terner, C., Kumar, L. A., and Choe, T. S. (1968). Studies of metabolism in embryonic development. 11. Biosynthesis of lipids in embryonated trout ova. Comp. Biochem. Physiol. 24, 941-950. Te Winkel, L. E. (1963). Notes on the smooth dogfish Mustelus canis during the first three months of gestation. I and 11. J. Exptl. Zool. 152, 115-137. Trinkaus, J. P. ( 1951). A study of the mechanism of epiboly in the egg of Fundulus heteroclitus. 1. Exptl. Zool. llS, 269-319. Truscott, B., Idler, D. R., Hoyle, R. J., and Freeman, H. C. (1968). Sub-zero preservation of Atlantic salmon sperm. J. Fisheries Res. Board Can. 25, 363-372. Vilter, V. ( 1946). Action de la thyroxine sur la mbtamorphose larvaire de l’Anguille. Compt. Rend. SOC. Biol. 140, 783-785. Vladykov, V. D. (1956). Fecundity of wild speckled trout (SaZvelinus f o n t i d i s ) in Quebec lakes. J. Fisheries Res. Board Can. 13, 799-841. von Ledebur, J. F. ( 1928). Beitrage zur Physiologie der Schwimmblase der Fische I. 2. Vergleich. Physiol. 8, 44-86. Waddington, C. H. (1956). “Principles of Development.” Allen ti Unwin, London. Wedemeyer, G. (1968). Uptake and distribution of ZnU in the coho salmon egg ( Oncorhynchus kisutch). Comp. Biochem. Physiol. 25, 271-279. Wickett, W. P. (1954). The oxygen supply to salmon eggs in spawning beds. J . Fisheries Res. Board Can. 11, 933-953. Williams, J. (1967). Yolk Utilization. In “The Biochemistry of Animal Development” ( R. Weber, ed.), Vol. 2, pp. 372-374. Academic Press, New York. Winberg, G. G . (1960). Rate of metabolism and food requirements of fishes. Fisheries Res. Board Can. ( T r a w l . ) 194. Withler, F. C., and Morley, R. B. (1988). Effects of chilled storage on viability of stored ova and sperm of sockeye and pink salmon. 1. Fisheries Res. Board Can. 25, 2895-2699. Woodhead, P. M. J. (1957). Reactions of salmonid larvae to light, J. Erptl. B i d . 34, 402-416. Woodhead, P. M. J., and Woodhead, A, D. (1955). Reactions of hemng larvae to light: A mechanism of vertical migration. Nature 176, 349-350. Yamamoto, T. ( 1961 1. Physiology of fertilization in fish eggs. Intern. Reo. Cytol. 12, 361-405. Zotin, A. I. (1958). The mechanism of hardening of the salmonid egg membrane after fertilization or spontaneous activation. J. Embryol. Exptl. Morphol. 6, 546568. Zotin, A. I. (1965). The uptake and movement of water in embryos. Symp. SOC. Exptl. Biol. 19, 365-384.

5 FISH CELL AND TISSUE CULTURE K E N WOLF and M . C. QUIMBY I . Introduction . . . . . . . . . . . A . History . . . . . . . . . . . . B. Present Status . . . . . . . . . . I1. Physiological Salines . . . . . . . . . . A . General . . . . . . . . . . . . B. Modifications in Balanced Salt Solutions for . . . . . . . . Use with Various Fishes I11. Media . . . . . . . . . . . . . A . Nutritional Factors . . . . . . . . . B.pH . . . . . . . . . . . . C . Storage of Prepared Medium . . . . . . . D . Antibiotics . . . . . . . . . . . IV. Methods . . . . . . . . . . . . A. Preparation of Fish for Obtaining Tissues . . . . . B. P.reparation of Primary or Original Monolayer Cell Cultures . C. Seeding Density for Primary Monolayer Cultures . . . D. Dispersion of Monolayer Cell Cultures . . . . . E. Seeding Density for Cell Lines . . . . . . . F. Precautions to Be Taken with Fish and Other Poikilotherm Cells . V. Choice of Tissues for Culture . . . . . . . . VI . Storage and Preservation . . . . . . . . . A. Freezing . . . . . . . . . . . B. Low Temperature Incubation . . . . . . . VII Fish Cell Lines . . . . . . . . . . . A . General . B. Sources of Fish Cell Lines . . . . . . . . VIII . Shipment of Cell Cultures . . . . . . . . IX. Needed Developments . . . . . . . . . Acknowledgments . . . . . . . . . . . . References . . . . . . . . . . . . .

. . . . . . . .

.

. . . . . . . . . .

253 254 259 260 260 263 265 265 271 271 271 273 273 276 280 281 283 284 286 287 287 289 289 289 294 294 295 300 301

.

I INTRODUCTION

This chapter is intended to be a comprehensive source of information and a guide to methods of fish cell and tissue culture. Among the subject 253

254

KEN WOLF AND M. C . QUIMBY

texts, Cameron (1950) presents the most information on fish cell and tissue culture, but the book is largely obsolete and inadequate for today’s techniques and needs. There are specific requirements for culturing cells and tissues from fishes, but the basic principles apply to all animals. Parker ( 1961), White (1963), and Paul (1965) are single volume subject texts which are currently in their second or third editions. The newest and largest reference is the three-volume set edited by Willmer (1965). Merchant et al. (1964) have prepared an excellent laboratory guide for both the student and technician or investigator; it has directness and simplicity. All the foregoing works reflect the fact that attention has been concentrated on mammalian and avian cells and that there has been much less work with the lower vertebrates and invertebrates. Because the techniques have such widespread application, tissue culture literature is widely scattered and the interested reader must use several approaches in reviewing and keeping abreast of developments. A bibliography of cell and tissue culture covering 1884-1950 was compiled by Murray and Kopech (1953). The same workers undertook the tremendous task of compiling the subsequent literature; several issues of a serial publication were printed, but the endeavors were soon halted and fate of the project is unknown (Murray and Kopech, 1965). Watson (1966; Watson and Gilford, 1967) has compiled a bibliography which is restricted to invertebrate and poikilothermic vertebrate cell and tissue culture. In addition to the well-known biological indexes and abstracts, some of the current literature is abstracted by biological supply houses and published in serials. Tissue Culture Abstracts is an organ of Grand Island Biological Co., Grand Island, New York, and Tissue Culture Bibliography is published by Microbiological Associates, Inc., Bethesda, Maryland. The newcomer will find the catalogs from cell and tissue culture supply houses to be additional sources of useful information. Those from Grand Island Biological Co.; BioQuest, Cockeysville, Maryland, and Difco Laboratories, Detroit, Michigan, have “how-to-do-it” sections and informative descriptions and explanations.

A. History Animal cell and tissue culture has advanced in a very short span of time from an esoteric art to a workaday tool of many uses. The dominant stimulus for this evolution has been the widespread need and the use of cultures in growing virus; knowledge of both animal virology and animal cell culture has grown apace.

5.

FISH CELL AND TISSUE CULTURE

255

Quite understandably, researchers working with diseases such as cancer, poliomyelitis, and a number of acute human, avian, and mammalian viral infections have concentrated efforts on the development of culture methods for homoiothermic animal cells and tissues. However, the techniques of cell culture had their origin not in mammalian tissue but first in chick tissue culture, next in frog, and soon after that in fish tissue. The early work was quite simple and consisted of short-term study of bits of tissue in slide or watchglass preparations, and this all began in 1885 when Roux maintained chick neural tissue in warm saline. In 1887, Arnold lured frog leukocytes into elder pith soaked in aqueous humor and observed cell movement and behavior. It was also the frog which provided Harrison in 1907 with neural tissue which in clotted lymph showed outgrowth of nerve cell fibers-the first real “tissue culture.” Like its parent techniques, the evolution of fish cell and tissue culture follows a progression in time and complexity; at first, small bits of fully organized tissue were merely kept alive for a matter of hours in saline. Cover slip, embryo dish, and watchglass preparations gave way to flasks and culture tubes, while saline was fortified with inorganic ions and a wide array of sera and other natural fluids. Today’s methods use a wide variety of glass and plastic laboratory ware. Media have become very complex, but the commercial availability has diminished the problems of preparation by the user. Today, media may be purchased from a number of suppliers who offer it in ready-to-use form, in liquid concentrates, or in powdered or lyophilized preparations. Solution or reconstitution of the latter requires the use of high purity water. At the very pinnacle of today’s accomplishments, some mammalian cell lines are grown in completely synthetic media. The first report of fish tissue culture was made by Osowski (1914) who maintained fry and embryonal trout explants for 24 hr in both Ringer’s solution and in frog lymph. In 1916 Lewis described a physiological medium based on seawater and employed bouillon as a source of nutrients. Subsequently, Dederer ( 1921) grew Fundulus embryo explants and kept heart beating for 10 days in Locke’s physiological saline plus glucose and fish bouillon. In 1924, Goodrich also used a similar medium for his work with Fundulus. Chlopin (1925, 1928) grew in cover slip culture a number of different tissues from lamprey, pike, and crucian carp using rabbit plasma diluted with fish spleen extract, and he specifically reported mitotic cells. He also cultured freshwater teleost and amphibian explants compatibly in the same vessel. Except for an occasional paper during the 1930’s and 1940’s, fish were

256

KEN WOLF AND M. C. QUIMBY

not in the mainstream of tissue culture. Pfeiffer (1935) cultured lamprey, Petromyxon fluuiutilis, explants in Chlopin’s medium. Lewis and MacNeal (1935) used a combination of fish and chick plasma for hanging drop cultures of pituitary from three marine teleosts and two marine elasmobranchs; the latter was the first work with this class of fish. They stressed the importance of isotonicity for the various animals’ tissues and were able to maintain ciliary beating for 8-10 days and cultures for 3 weeks. In 1941, Grand, Gordon, and Cameron cultured melanotic tumor tissue from freshwater teleost aquarium species with weekly changes of a medium composed of fish serum, chicken plasma, and embryo extract. They kept cover slip cultures “alive and active for many weeks.” Prior as it was to the introduction of antibiotics, they were confronted with the problems of microbial contamination and found dilute Merthiolate to be a helpful disinfectant for use on the surface of their fish. Grand and Cameron (1948) used the same kind of medium for a similar study with pigmented tumor tissue of FunduZus, but they had the advantage of penicillin. The use of antibiotics was a major advance in animal tissue culture for it helped transform the techniques which at that time were almost a goal in themselves into a tool which could be easily and conveniently used by many. Carrel introduced flask cultures in 1923, but with few exceptions (Cameron, 1949) they were seldom used in early fish tissue culture. In time T-flasks and pIastic flasks replaced the Carrel flask, but the culture tube became the standard vessel. Media too were improved, and semisynthetic preparations based on blood serum analyses were developed. Commercial marketing of such preparations began in 1947, and this too was an important factor in speeding the advance and application of cell and tissue culture. Although fowl pox had been grown in chick tissue culture some 20 years earlier, a major breakthrough occurred in 1949 when Enders, Robbins, and Weller grew poliomyelitis virus in nonneural human tissue in uitro. Reciprocal stimulation had begun between virology and animal tissue culture, and the latter soon gave rise to the more sophisticated cell culture. During this transitional period, Schlumberger ( 1949) used roller tubes and mammalian-type medium in a tissue culture study of a neoplasm from adult goldfish, Carussius uurutus. Some evidence suggested an infectious agent and although the etiology of this tumor has never been established, this was probably the first application of fish tissue culture in virology. Soon afterward, Sanders and Soret (1954) and Soret and Sanders (1954) reported growth of an arthropod-borne virus in fish tissue culture; in this case the “tissue” was actually intact Gumbusiu embryos grown in mammalian-type medium. The same methods

5.

FISH CELL A N D TISSUE CULTURE

257

permitted in uitro development of active young fish from early stages of eggs. Although it was not used for the bulk of the study, their work marked the first report of a synthetic medium designed for mammalian cells (medium 199) being used for teleost tissues. In 1956, Griitzner (1956a,b) published two reports of a series of carefully conducted, detailed studies of fish tissue culture and its applications in fish virology. She cultured explants of embryonal juvenile and adult aquarium teleosts in diluted mammalian-type medium. Heart was kept beating and gut showed peristalsis for 2 weeks, and cultures in general could be kept for 3-6 weeks. Such cultures were employed in a study of viral lymphocystis disease and of carp pox. In the same year, Wolf (1956) used explants of adult trout tissue in mammalian-type medium in an attempt to determine whether virus was involved in bluesac disease of fish. The following year, Wolf and Dunbar (1957) reported the use of mammalian-type media for culture of adult trout and goldfish tissues. They reported no subculturing but kept cilia beating for nearly 2 months and goldfish heart beating for 10 months. Grutzner (1958) made a major contribution by describing the first trypsinization of fish tissue yielding cultivable cells which grew in monolayer. The methods which can be considered “second generation” followed those which had been developed for mammalian use, but temperature during digestion was 20°C or less and solutions were diluted to correspond to the reported freezing point of tench, Tinca tinca, serum. Griitzner’s primary monolayers were successfully subcultured and thus fish cell culture had been established. Although its applications were to decrease, fish tissue culture was extended and used by others, Bargen and Wessing (1960), Ghittino (1961), Kunst (1961), and Wolf et al. ( 1960), and the latter made the first isolation of a fish virus in cultures of trout fin explants. Unaware of Griitzner’s work, Wolf et al. (1960) employed a different procedure for use with cold-water teleosts. Using extended trypsinization at 4 ’ 4 ° C and unmodified mammalian-type medium, they cultured cells of six freshwater teleosts, an amphibian, and a reptile. Subcultures were effected by mechanical dispersion as well as with trypsin or disodium versenate. Significantly, some tissues were stored at 4°C for 24 hr before use. Preparation of monolayer cell cultures from enzymically disaggregated fish tissues has been reported by a number of other workers who anticipated or made immediate application in fish virology (Babini and Ghittino, 1961; Clem et al., 1961; Kunst, 1962; Tec and Jakovleva, 1962; Jensen, 1963; Fryer, 1964; Pfitzner, 1964; Pfitmer and Froehlich, 1966; Kunst and Fijan, 1966). Clem et al. ( 1961) were the first to establish monolayer cell cultures from marine teleosts and understandably

258

KFJi WOLF AND M. C . QUIMBY

obtained their best results in commercial medium modified with about 0.07 M added NaCl. Some generalizations seemed to be evolving. There was uniform agreement that teleost tissues could be trypsinized and grown with methods very much like those used with mammalian cells. Temperatures needed to be lower however (opinions differed as to degree), and all agreed that mammalian serum was at least as good and in most cases better than homologous serum. Embryonic and gonad tissues emerged as dependable sources of cultivable cells from fish and other lower vertebrates. Continuously cultivable cells meeting certain criteria ( Fedoroff, 1966 ) are termed “established or “permanent lines” and might properly be considered a “third generation” in the advance of fish cell and tissue culture. Cell lines are usually derived from monolayer cell culture, but this is certainly not a requirement and the most noteworthy exception was the first-Carrel’s line of chick fibroblasts, which were propagated for 34 years. Cell lines have a number of advantages, and as early as 1958 serial subcultivation of some tissues had been effected at the Eastern Fish Disease Laboratory; however, none of the early efforts resulted in a permanent cell line. With trypsinized cell preparations the likelihood of success was increased. RTG-2 cells of rainbow trout, Salmo gairdneri, gonad origin were one of several lines initiated at our laboratory in January, 1960; they were the first permanent fish cell line to be developed (Wolf et al., 1962). Within a year, Clem et al. (1961) initiated trypsinized blue-striped grunt, Haemulon flavolineatum, fin cultures which provided GF-1 cells, the first line of marine fish origin. The next established line was initiated in 1962; it was an epithelial-like cell from the fathead minnow, Pimephales p r o m e h , and designated FHM by its originators Gravel1 and Malsberger ( 1965). At Oregon State University, Department of Microbiology, Fryer (1964) made a very carefully quantitative study of requirements for dispersing tissues and culturing cells from embryonic Pacific salmon and rainbow trout hepatoma. The work is additionally noteworthy for it reports the establishment in 1963 of five digerent lines of cells from these fishes. Part of the work has been published (Fryer et al., 1965), and equally important is the fact that the lines are still prospering (Table VI ) ( Fryer, 1968). Basic studies in fish cell and tissue culture continue to appear. Townsley et al. (1963) employed mammalian-type culture medium without osmotic adjustment and reported the successful culture at 5°C of explants from a number of cold-water Atlantic teleosts and one elasmobranch. Li and Stewart ( 1965) recently used short-term monolayer

5.

FISH CELL AND TISSUE CULTUFU3

259

cultures of rainbow trout ovary for evaluating the growth promoting qualities of several natural substances. They found that a combination of serum and embryo extract gave maximum proliferation. In the first such study ever to be made, Pilcher et al. (1968) cornpared growth and glycolysis in two established embryonic salmonid cell lines and a line of diploid human embryonic cells. Although incubating temperatures of necessity were different, they found one of the salmon cell lines grew almost as rapidly as the mammalian cell, but that rates of glucose utilization and lactic acid formation were lower in the fish cells. They concluded that glycolysis in all cells was similar but that it differed quantitatively. Understandably, most applications of fish cell and tissue culture have been in virology, but though they may be fewer in number, important applications have been made in other research areas. Cultured fish cells have been shown to produce interferon (Beasley et al., 1966; Beasley and Sigel, 1968) and in the same laboratory, Ortiz-Muniz and Sigel (1968) used organ cultures to demonstrate in vitro synthesis of antibody by fish lymphoid tissue. Roberts (1964, 1966, 1967) effectively used cell culture as a tool for determining the chromosome numbers of fishes. Because the in vitro approach provided a completely controlled endocrine environment, Hu and Chavin (1960) employed goldfish fin explants for a study of hormonal stimulation of melanogenesis. They found only ACTH capable of stimulating in vitro melanogenesis. As with any system, things can go wrong and among those who work with mammalian cells there have been a number of disconcerting incidents of mistaken identity-possibly by mislabeling or “contamination” with another cell. To forestall such mixups among lower vertebrate cell lines, Levan and his co-workers (1966) showed chromosome idiograms to be useful in distinguishing certain cell lines. A simpler and more specific approach has been the cytotoxic antibody test (Greene et al., 1966). Quite recently, there has been a revival of interest in in uitro culture of nonteleost fishes. Stephenson and Potter (1967) worked with a cyclostome while Wolf and Quimby (1968) used both a cyclostome and elasmobranchs.

B. Present Status Many different kinds of tissues and cells from freshwater and marine teleosts can be cultured on a routine basis. As a general index of the state of the art, teleost cells are the second most numerous among the

260

KEN WOLF AND M. C . QUIMBY

animal lines which have been developed-mammalian cells being the most numerous. Although all are heteroploid, there are at least 17 extant lines of fish cells. Thus far, fish cell and tissue culture has been predominantly concerned with material from teleosts. Very little is known about culture of cyclostome cells, and while there is some indication that present methods and media may be satisfactory, there is also some evidence that both cyclostomes and elasmobranchs may have specific requirements different than teleosts. The methodology for in vitro culture of fish cells closely follows that used with homoiotherm material; the major differences being, first, in temperature requirements and tolerances, and, second, in osmolarity of salines and media. For the most part, mammalian-type solutions are entirely satisfactory for many freshwater teleosts and possibly for cyclostomes. For marine fishes, best results are obtained after the osmolarity is increased. Fish cells are comparable to mammalian cells in their response to freezing and storage at ultra-low temperatures, but like all poikilothermic animal cells, their rate of metabolism can be manipulated with temperature control. This attribute provides the researcher with a range of growth or activity rates for his studies, and at the lower end of temperature tolerance it permits long-term storage of cells without freezing. To a certain extent, the present status of fish cell and tissue culture can also be described in negative terms; the reader will find these aspects discussed in Section IX.

11. PHYSIOLOGICAL SALINES

A. General Physiological salines are fundamental to cell and tissue culture for they maintain pH, provide osmotic pressure, essential ions, and glucose as an energy source and are therefore in themselves adequate for handling, washing, manipulating, and short-term holding of living materials. More importantly, they provide the inorganic foundation on which media are elaborated. Ringer’s solution, an empirical mixture of the essential ions Na, K, and Ca, was the prototype saline. Today’s salines, balanced salt solutions (BSS) as they are usually called, are commonly patterned after the inorganic constituents of blood serum. Sera from the

5.

261

FISH CELL AND TISSUE CULTURE

major vertebrate classes have basic similarities in their inorganic chemical composition, accordingly this resemblance is reflected in the physiological salines which are based upon them. In theory one should be able to use with fish the salt solutions which have been formulated for homoiotherms. In practice this has worked very well indeed, the one note of caution being that adjustment in osmotic pressure has usually been found necessary for marine forms. The existing mammalian-type balanced salt solutions then are in effect vertebrate physiological salines. Understandably, the first physiological salt solutions were developed for use by physiologists, and different formulations were prepared for use with the various kinds of fishes. Young (1933), who carried out studies with three classes of fishes, summarizes fish physiological salines to that date. Lockwood (1961) prepared an extensive exposition on physiological salines for animals of different phyla including fishes. Curiously, it includes principal salines in use with mammalian cell and tissue culture, but there are no references on fish cell and tissue culture. The reverse situation is almost equally true; recent methods of fish cell and tissue culture have not employed the solutions used by the fish physiologists (Wolf, 1963). In the earlier literature on fish tissue culture one finds a considerable array of physiological salines (Table I), Reports and micrographs of tissue growth show that these solutions were adequate for their purTable I Physiological Salt Solutions Used in Fish Tissue Culture, 1914-1956 Solution

Holtfreter’s solution

Locke-Lewis solution Ringer’s solution Tyrode’s solution

Reference Freshwater Teleost Grand et al. (1941) Devillers (1947) Grand and Cameron (1948) Greenberg et al. (1956) Grutaner (1956~) Osowski (1914) Schlumberger (1949)

Holtfreter’s solution Locke’s solution Seawater (diluted)

Marine Teleost Oppenheimer (1935) Lewis and MacNeal (1935) Lewis (1916) ; Dederer (1921) ; Goodrich (1924)

Fuhner’s solution

Elasmobranch Lewis (1916); Lewis and MacNeal (1935)

262

KEN WOLF AND M. C. QUIMBY

pose. By present standards, however, they are primitive-being

deficient

in some components or lacking completely certain ions. Few had an

energy source, and most were not adequately buffered. Undoubtedly one reason they were adequate is the fact that the tissues were not kept in saline for long periods. In addition, the culture systems usually employed rather large volumes of tissue for the volume of medium, therefore leaching could occur and the system would move to equilibrium. Most importantly, however, then as now, salines were usually only a part of the culture medium, and many deficiencies were met by serum, tissue extracts, or other natural products. The earlier physiological salt solutions are principally of historic interest. Today, virtually all fish cell and tissue culture employs the “mammalian-type” BSS usually without modification, but in some cases diluted or supplemented with additional NaCl (marine forms) or urea (elasmobranchs) or both. Hanks’ BSS is by far the most commonly used saline in fish cell and tissue culture, as indicated in about 80%of the reports published during the last 15 years. Earle’s BSS has had limited use as a saline per se; it is however the most common base used in media for growing fish cells. Gey’s BSS and the Cortland salt solution which was based on the determinations of Phillips et al. (1957, 1958) each have one report of use with fish. Hanks’, Earle’s, and Gey’s solutions were formulated for mammalian cell use; the first two, and to a lesser degree the third, have won wide acceptance in homoiothermic cell culture. These salines are offered by cell and tissue culture supply houses and are available both in liquid Table I1 Balanced Salt Solutions for Use with Fishes ~~

Earle’s BSS-man Constituents NaCl CaClr2Hn0 KCl NaH2P04.HI0 NanHPO4.2HZO NaHCOa KHnPOr MgClz*6Hz0 MgSOc7HzO Glucose Water (ml) Mean freezing point (“C)

(g)

6.80 0.27 0.40 0.14 2.20

-

0.20 1.00 1000 -0.58

Hanks’ BSS-man (9)

8.00 0.19 0.40 0.045 0.35 0.06 0.10 0.10 1.00 1000 -0.59

Cortland BSS-brown trout (9)

7.25 0.23 0.38 0.41

-

1.00

-

0.23 1.00 1000

-0.58

5.

FISH CELL AND TISSUE CULTURE

263

and in powdered form. The Cortland salt solution was based on serum constituents of a freshwater teleost (Wolf, 1963). A comparison with Hanks' and Earle's BSS (Table 11) shows their basic similarity.

B. Modifications in Balanced Salt Solutions for Use with Various Fishes

1. FRESHWATER TELEOSTS According to Black (1957), freshwater teleost blood has a freezing point of about -0.57"C (the range, -0.38 to -0.89"C). Lockwood (1961) cites somewhat narrower limits (-0.64"C to -0.45OC), but the mean is comparable (-0.54"C). One might conclude from the composition and freezing point of the foregoing salines that they are quite suitable for freshwater teleosts. Indeed, most workers have used them without alteration; others have found it desirable and even necessary to dilute the salines about 20% (Griitzner, 1956a,b, 1958; Pfitzner and Griitzner, 1964; Tomaiec et al., 1964; Rachlin et al., 1967; Babini and Ghittino, 1961). Kunst (1962) initially found it desirable to dilute salines and media about 20%,but in subsequent comparisons found undiluted solutions to be better (Kunst and Fijan, 1966). Osmotic concentration is usually reduced with water, this effects dilution without changing ionic ratios. An alternate method is to dilute with saline from which the NaCl has been omitted. Dilution alters the ionic ratio but maintains levels of other components. It is generally understood that slight hypotonicity is desirable for in vitro cell and tissue growth, but critical comparisons of osmotic concentrations have not been carried out. On the other hand, most freshwater teleost cell and tissue culture, including the establishment of permanent lines has been with salines and media having a freezing point of about -0.6"C.

2. MARINETELEOSTS Clem et al. (1961) used both unmodified Hanks' and preparations which were supplemented with 0.07 M additional NaCl. They reported significantly better results with the latter and established a permanent marine teleost cell line. We have successfully used isotonically adjusted salines and medium (addition of 0.3 g % NaC1) for handling, establishing cultures, and subculturing cells of the spot, Leiostomus xaiathurus, a marine teleost. If the serum freezing point has been properly determined for fish which are normal and healthy, the saline and medium should be

264

KEN WOLF AND M. C. QUIMBY

matched accordingly. As an example, we established primary cultures of frog tongue fibroblasts in medium having a freezing point of about -0.59"C-considerably hypertonic for the frog. Several subcultivations were effected but cell number declined and cells became pyknotic. The medium was then diluted with NaC1-free Earle's BSS to a freezing point of about -0.47"C. Cell quality and vigor immediately improved, and a permanent cell line was established (Wolf and Quimby, 1964). Supporting the thesis that a strongly hypotonic saline and medium is similarly unfavorable are the findings of Roberts (19sS).Although Atlantic herring cells grew in medium having a freezing point of -0.58"C, they were highly vacuolated, and Roberts considered this a possible consequence of hypotonicity. 3. MARINEELASMOBRANCH

A number of physiological salines have been developed for elasmobranchs, but only two have been used in tissue culture. The more complete of the two is Hanks' BSS modified with additional NaCl and urea. Lockwood (1961) gives the composition of five different salines for different elasmobranchs; but none was used in cell or tissue culture. An additional reference is that of Pereira and Sawaya (1957). All the formulations include the very high but completely normal and physiologically necessary levels of urea-the single component which literally characterizes the serum of these fishes. Lewis (1916) proposed a culture medium based on a physiological saline described by Fiihner in 1908. It is noteworthy for the fact that it has the necessary levels of both NaCl and of urea. We have tested salines and media with compositions intended for freshwater teleosts or marine teleosts and found them quite unsuited for marine elasmobranchs. Empirically, we determined the levels of NaCl and urea necessary for the most favorable response of shark cells and tissues. We used Hanks' BSS and 1.75-2.0 g % urea added as a small volume of stock concentrate. The stock solution was prepared by dissolving 40 g of urea in 50 ml of water. This was filter sterilized and had a volume of approximately 80 ml; accordingly each milliliter contained 0.5 g of urea. Sodium chloride alone did not provide proper physiological conditions even though the medium was isotonic with serum from our donor animals (-1.6"C). Hanks' BSS with about 0.5 to 0.6 g % additional NaCl (and the urea levels above) had a freezing point of about -1.5"C and seemed suitable for several species of sharks (Wolf and Quimby, 1968). There has been no culture work on freshwater elasmobranchs, but the animals have high urea levels (Urist and Van de Putte, 1967), and this compound may be needed in the medium.

5.

FISH CELL AND TISSUE CULTURE

265

4. CYCLOSTOMES Unmodified Hanks’ BSS has been found suitable for freshwater larval cyclostomes (Stephenson and Potter, 1967; Wolf and Quimby, 1969). From data presented by Urist and Van de Putte (1967) the inorganic constituents of teleosts and cyclostomes in freshwater are very similar. Conversely their data show that salines for marine cyclostomes would require higher levels of NaCl.

111. MEDIA

Not only do cell culture media provide nutrients, but also they are the reservoir for metabolic products. Considering nutritional requirements to mean knowledge of the qualitative and quantitative needs for specific elements, ions, vitamins, amino acids, etc., virtually nothing has been determined for fish cells. However, much of what is known about homoiotherm cells probably applies, at least broadly, to all poikilothermic vertebrate cells. Thus far, a “shotgun” approach has been effective in growing fish cells; they simply are provided with the inorganic salts, vitamins, and amino acids known to be needed by mammalian cells; for good measure, generous quantities of serum or other nutritive supplements are added. This method has certainly worked; it has permitted widespread application of fish cell and tissue culture as a tool, but it has contributed little to knowledge of nutritional requirements for fish cells in uitro.

A. Nutritional Factors The first animal tissue culture media consisted of very simple salines plus serum, tissue extracts, or other natural supplements. Eventually, the proportion of natural components was reduced and the number of defined substances was increased. Thus the media became semisynthetic; they consisted in part of a defined basal solution of inorganic salts, glucose, vitamins, and amino acids. To sustain growth, such media required the addition of serum or other natural products. Today, practically all fish cell and tissue culture-and the great bulk of homoiotherm cell culture-uses media of this developmental level. The ultimate goal in cell culture has been to grow cells continuously in completely defined media. Today there are several formulations which supply all the requirements of some mammalian cells. These media quite probably will

266

KEN WOLF AND M. C. QUIMBY

meet the requirements of at least some cells from some fishes. As formulations are improved, it is expected that other cell types can be grown in them. On the other hand, it is possible that synthetic media may have to be specifically tailored for certain cells from specific kinds of fishes. As previously mentioned, the inorganic composition of mammalian and teleostean sera are quite similar, but too few data are available from fish to compare the organic constituents. The principal organic components may also be similar, for when the solutions are isotonic, teleost cells readily grow in media developed for homoiotherm cells. The same is true for cells from reptiles and amphibians. In discussing animal cell nutrition Paul (1965) has said: “. . . the most extraordinary thing about known nutritional requirements of different cells is their general similarity” (italics added). There are few published data, but it appears that cyclostome cell requirements could be like those of teleosts. The elasmobranchs also show a basic similarity, but there are significant differences, not only in the urea tolerance but also in the near absence of albumin. Considering the fact that most fish cell and tissue cultures are now grown in semisynthetic media with natural supplements and that these have proved superior to older media, there seems to be little justification in discussing or enumerating the more primitive preparations. The interested reader will find partial reviews in Grutzner (1956a), Wolf and Dunbar (1957), and Fryer (1964). Although their own definitive work employed a different medium, Soret and Sanders (1954) were the first to report use of a semisynthetic solution, medium 199, for fish tissue culture. Wolf and Dunbar (1957) grew tissues from several fishes in a mixture which contained medium 199, but in addition to high levels of serum they also used chick embryo extract. Grutzner (1958) also used medium 199; in part it was mixed with multiple supplements, but more importantly she was the first to use a synthetic medium with serum as the sole supplement-the mode of today’s fish cell culture. Cataloging the details of reports on fish cell culture shows that three synthetic media have been used far more than others; in order of decreasing frequency of use they are: medium 199, Eagle’s minimal essential medium (MEM) and Eagle’s basal medium (BME).* Other synthetic media which have been reported as suitable for fish cells are: CMRL 1066 (Townsley et al., 1963); Leibovitz L-15 (Gravel1 and Malsberger, 1965); McCoy’s 5a ( McFalls et al., 1967); NCTC 109 (Wolf et al., 1960,

* The synthetic media listed here are standard in vertebrate cell and tissue culture; their composition, originators, and pertinent references are given in most subject texts. These media are also available from the major subject supply houses. Accordingly, it is not considered necessary or desirable to list formulation, etc.

5.

267

FISH CELL AND TISSUE CULTURE

1962; McFalls et al., 1967); and Puck's (Hu and Chavin, 1960; Wolf et at., 1960, 1962). Eagle's MEM is the one medium that we suggest for routine fish cell and tissue culture (Table 111). For our own use we prefer the formulation which uses Earle's BSS because it has greater buffering capacity. Admittedly, medium 199 has had a slightly greater frequency of published use, but from our own experiences and discussions with others, including people in the commercial supply houses, MEM has been effective and unquestionably is now used more frequently. Also, the cost of MEM is about 10%less than that of medium 199; in addition, MEM is used for culture of amphibian and reptilian cells and is widely employed in homoiotherm cell culture work. Other things being equal, it is likely that most if not all of the synthetic media originally intended for homoiotherms will prove satisfactory for culture of teleost cells and probably for cells from other fishes too. For exploratory work with new fishes or new techniques, we suggest a variety of media be used in order to determine which is the most satisfactory, 1. SERUMADDITIVES

With the above synthetic preparations, serum or other complex undefined organic materials are usually added-and probably neededTable I11 General Purpose Culture Media for Fish Cells and Tissues" Supplements Culture type Explant or monolayer in a closed system

Explant or monolayer in petri dish or other open system

Synthetic component Eagle's MEM with Earle's or Hanks' BSS, medium 199 Leibovitz' L-15

Required 5-20% serum (fetal bovine or calf serum suggested)b . c As above

Optional 5% whole egg ultrafiltrate

As above

Media are suitable without alteration in osmolarity for many freshwater teleosts ; we have found various lots of Eagle's MEM (Earle's BSS) with 10% fetal bovine serum to have a freezing point of -0.57" to -0.65"C. Eagle's MEM and medium 199 with and without a 10% reduction in osmolarity have been used with larval cyclostomes. About 0.07 M (0.4 g %) NaCl should be added for marine teleost cells and tissues. For marine elasmobranch cells and tissues 0.08-0.1 A4 (0.5-0.6 g %) NaCl a n d 0.29-0.33 M (1.75-2.0 g %) urea should be added. Fetal bovine serum is the current nutritional supplement of choice, but it is not wholly satisfactory.

268

KEN WOLF AND M. C. QUIMBY

for sustained vigorous growth. Disregarding for the moment the quality of response, cost, and inherent risks, and considering all fish cell and tissue culture, homologous fish serum and calf serum have been used most often. These are followed by heterologous fish, human, human cord, and chicken serum. Horse, bovine, rabbit, fractionated calf serum, sheep, and swine serum have also been used. Some generalizations can be made about the results. Teleost cells often fare poorly in rabbit serum, and in 10%chicken serum they accumulate abnormal quantities of cytoplasmic globules which may be lipid. When chicken plasma is used as a clotting matrix and the culture medium contains another serum, the quantity of chicken serum released is too small to result in problems of cytoplasmic globule accumulation. Stephenson and Potter ( 1967) obtained excellent results with larval Iamprey tissues grown in chicken serum. However, it is one of the more expensive sera. Human cord serum has uniformly been found to be excellent for culture of fish materials but limitations because of cost or problems in collecting and processing restrict its use. Human serum generally has been good to excellent. Depending on the various workers, horse, bovine, and sheep sera have given good, fair, or poor results; there are too few data to support a generalization. Fish sera, either homologous or heterologous, have also given mixed results. Among the reports of the more recent workers, some have found the growth response to fish serum to be excellent (Babini and Ghittino, 1961; Kunst, 1961, 1962; Tomaiec et al., 1964; Stephenson and Potter, 1967), but for others, heterologous or homologous fish serum has been inhibitory (Clem et al., 19Sl) or toxic (Fryer, 1964; Fryer et al., 1965). Compared to mammals, fish have a smaller blood volume. This can often be a problem in acquiring adequate amounts of fish serum for cell culture; of course, the problems are virtually insurmountable with the very small fishes. Perhaps the greatest deterrent to advocating use of fish serum for fish cell culture is the risk of introducing latent fish viruses. Calf serum, be it the natural material from fetal or full-term animals, or chemically fractionated, stands out clearly as the serum of choice for growing teleost cells and tissues. It is also suitable for cyclostome cells and has been used for elasmobranch material but not completely satisfactorily. The problem with elasmobranch cells has been massive cytoplasmic accumulations of granular material, possibly albumin, which is almost a foreign protein to sharks, skates, and rays (Wolf and Quimby, 1968). Where budgets can afford it, fetal bovine serum is the serum of choice. It is notable for its lack of toxicity and for its growth stimulating properties. Without doubt, one of the most important reasons for its widespread use in fish cell culture is that much of the work has been in

5.

FISH CELL AND TISSUE CULTURE

269

fish virology and fetal bovine serum is generally free of virus neutralizing activity. In contrast, although the material may have been nonspecific, some lots of calf and human serum have been found to neutralize infectious pancreatic necrosis virus. Cross-reacting antibody may also neutralize virus, and to eliminate this possibility, yglobulins may be selectively precipitated; this results in the so-called agamma calf serum. Where viral neutralizing activity is not a consideration, calf serum should suffice. Of course whether locally or commercially prepared it is always prudent to test any serum for growth promotion, toxicity, and viral neutralizing activity prior to large-scale use. When serum from commercial sources is used and sufficiently large volumes are involved, the biological supply houses will usually provide a sample and reserve a specific lot or portion thereof pending the outcome of such testing. We use this system of pretesting serum and buy a year’s supply at a time. It is stored at -20°C. When used with MEM, medium 199, or comparable synthetic medium, the usual level of serum is 10-15% (Table 111). Some fish cells will grow with as little as 2% serum, but the growth rate is slow. Some have found it necessary to use as much as 20 or even 30%serum levels, the latter however causes one to wonder whether such a high level was necessary, or whether the other components were adequate.

2. OTHERADDITIVES A number of natural and partially processed materials have been used as nutrients for fish cell and tissue culture. In some instances these products are used in lieu of serum, in other work they have been used in addition to serum. The products are human ascitic fluid, bouillon, bovine amnionic fluid, fish, bovine or chick embryo extract, lactalbumin hydrolysate, serum ultrafiltrate, peptone, yeast extract, and whole egg ultrafiltrate. Bouillon is of historical interest and has not been used for about 30 years. Human ascitic fluid has been used with saline and chick embryo extract as a complete medium for very short term work (Hu and Chavin, 1960). Greenberg et al. ( 1956) used 55% ascitic fluid with chick plasma and embryo extract and found it gave better results than a comparable amount of carp serum. Townsley et al. (1963) found that with medium 199, a 10% level of ascitic fluid gave excellent resultscomparable to the same amount of human serum. Li and Stewart (1965) used ascitic fluid in conjunction with 15%serum and 5% embryo extract in medium 199-the mixture gave a very rapid rate of growth. There is no doubt that ascitic fluid is a good source of nutrients; its cost and secondarily its pathological origin are factors which tend to limit its use. Embryo extracts have almost been synonymous with tissue culture. They have been used as a stimulant to growth and as a source of enzymes

270

KEN WOLF AND M. C . QUIMBY

essential to clotting of plasmas, For either use, fish, chick, or bovine embryo extracts can be used interchangeably. As a component of medium, the commercial products are expensive and there now seems to be little real justification or need for embryo extract in that role. Embryo extracts can be made locally quite cheaply, but it seems likely that similar benefit can be had from a comparable amount of good serum. In our own media we discontinued using embryo extract in 1959. Ultrafiltrates, protein-free fractions of serum or of whole chicken eggs, are also used as sources of nutrients and growth factors. Though effective, there have been very few applications in fish cell and tissue culture, and the principal limiting factor has probably been cost. In vitro development of Gambusia sp., from fertile egg to larva, has been carried out in medium composed of 20%ox serum ultrafiltrate and 80% Hanks’ BSS (Sanders and Soret, 1954; Soret and Sanders, 1954). We have used a 10%level of whole egg ultrafiltrate in medium for developing a permanent line of amphibian cells (Wolf and Quimby, 1964) and a 5% level during the development of several fish cell lines when it appeared that vigor was declining and continuation of the culture was threatened. Bovine amnionic fluid ( BAF), a natural product with physiological ions and a low level of protein, has had limited use in vertebrate cell and tissue culture as a saline substitute or serum extender. Under some local conditions it may provide a cheap component for medium, but commercial prices range from about half to twice that of calf serum. It has comparatively few advocates and its applications have been declining. When used at a 5-20% level it had neither significant advantages nor disadvantages (Wolf and Dunbar, 1957; Jensen, 1963; Townsley et al., 1963). Griitzner (1958) obtained growth of tench, Tinca tinca, liver and occasionally of kidney cells in 70%BAF (containing lactalbumin hydrolysate and yeast extract) and 30%calf serum. However, this medium was later reported to be unsuited for tench swimbladder and gonadal cells (Pfitzner and Griitzner, 1964). Kunst (1962) tried five media for growing carp, Cyprinus carpio, kidney cells and concluded that the best results were obtained with 74%BAF plus 26%calf serum when these were isotonically diluted (about 20%).Cells did not grow when the medium was not diluted. In later work with carp ovary cells the use of BAF was abandoned (Kunst and Fijan, 1966). It might be that kidney and liver cells better tolerate the urea levels present in BAF. Lactalbumin hydrolysate (LAH), peptone, and yeast extract or lysate are low-cost sources of nutrients which have had limited use in fish cell culture. These ingredients have a history of use in the so-called maintenance type of media employed in homoiotherm virology. Jensen (1963) and Tomalec et al. (1964) used 0.5%LAH in growth media for fish

5.

FISH CELL AND TISSUE CULTURE

271

virology. However, in several studies, which included an evaluation of media containing either these nutrients or serum, the authors consistently concluded that the best results were obtained with serum (Griitzner, 1958; Pfitzner and Griitzner, 1964; Fryer, 1964; Fryer et al., 1965).

B. pH The pH of medium necessary for good growth of fish cells does not appear to be particularly critical, and most cells seem to fare well in the range of 7.2-7.8. Primary cultures and low densities of cells will usually do better at 7.3-7.4 than at pH 7.8. Routine passage of some cell lines can be made at pH 7.8 or even 8.0, but the lag phase may be extended somewhat. At the other end, old cultures can have a pH as low as 6.8apparently without undue damage to the cells. Such cultures can be dispersed in fresh medium and growth will usually resume. We have often heard of cultures becoming excessively alkaline soon after seeding, and this we believe has been erroneously attributed to an unusual cellular or tissue response. Invariably, these are media with a bicarbonate buffer and the rise in pH has simply been equilibration between the gas and liquid phases in the culture system. This alkaline shift occurs even when medium alone is added to comparable culture vessels. The pH shift is particularly marked if small volumes of medium are involved-as, for example, 1 ml or less per 16 x 125 mm culture tube. The remedy is to use CO, to lower the pH to 6.8 or 7.0. Upon equilibration, the final pH will be 7.4-7.6.

C. Storage of Prepared Medium Since some of the growth factors in medium are labile, the medium should be refrigerated or frozen. We prefer not to store complete medium longer than 1 month at 4"C, but have found no untoward effects in some which were held for several months, We routinely prepare medium and store it at -20°C. Most lots are used within a month, but we have held some media for 4 years and found that they seemed to support growth as well as when prepared. D. Antibiotics

The same antibiotics are used throughout vertebrate animal cell and tissue culture, but little quantitative work has been done with fish tissues or cells. Working concentrations of the more commonly used

272

KEN WOLF AND M. C. QUIMBY

antibiotics have been within the same range and have been used with cells and tissues from a variety of fishes. At present, inhibitory or even toxic concentrations are largely unknown for cells and tissues from most fishes. The use of antibiotics involves selection of the drug and of the concentration that is to be employed. Antibiotics differ in the effects they produce on cells, and it is likely that cells of different fishes will vary in their response to a particular antimicrobial substance. Accordingly, the nature of the work must be considered; e.g., will it be longor short-term, a terminal use in virology, or a metabolic study? And of course the risk of contamination must be assessed. Routine passage of cells presents a much lower risk of contamination than establishing primary cultures of external tissues. For routine purposes, many have used media containing 100 IU of penicillin, 100 pg of streptomycin, and 25 IU of nystatin per milliliter. We have used chlortetracycline at 50 pg/ml in lieu of the mixture of penicillin and streptomycin, but chlortetracycline forms a precipitate in the medium. In our experience kanamycin and amphotericin B are more inhibitory or toxic at lower concentrations than other antibiotics. We use them primarily when contamination risk is high or when it has already occurred and salvage of cells is necessary. If for example a mold colony is found in a bottle, it is carefully removed and the necessary amount of amphotericin B added to the medium. Kanamycin has similarly been used with light bacterial contamination in the presence of penicillin and streptomycin. Such cells are employed in terminal use only and subcultures are not propagated further, Table IV Antibiotics Used in Fish Cell and Tissue Culture" Antibiotic

Routine use

Remarks

Amphotericin B Chlortetracycline HC1 Kanamycin sulfate Neomycin Nystatin Penicillin G, potassium Polymyxin B sulfate Streptomycin sulfate

2-10 pg/ml 50 pg/ml 25-50 p g / d 50-100 pg/ml 25-50 IU/ml 50-100 IU/ml N o data 50-100 IU/ml

May be toxic a t < 5 pg/ml 500 pg toxic to BF-2 cells 100 pg halves growth rate of RTG-2 cells 100 IU inhibitory for some cells 2000 IU/ml has been used 2000 IU/ml as brief bath 2000 IU/ml has been used

Most of the data are derived from work with teleosts, but a very few values are from work with elasmobranchs. In our own limited work with cyclostome tissue culture we found evidence that routine levels of penicillin and streptomycin were inhibitory. When we reduced the concentration to 25 I U and 25 pg, respectively, slightly less than that used by Stephenson and Potter (1967), the inhibitory effect was not evident. (I

5.

FISH CELL AND TISSUE CULTURE

273

Table IV is a suggested guide to the use of antibiotics with fish cells and tissues. It is based upon the values already reported and upon a limited amount of testing which we have not previously reported. For general information on the use of antibiotics in tissue culture the reader is referred to a booklet by E. R. Squibb and Sons, New Brunswick, New Jersey, or to excerpts of the same which appear in the price and reference manual of the Grand Island Biological Company. The catalog of the Industrial Biological Laboratories, Inc., 451 S. Stonestreet Ave., Rockville, Maryland, 20850 contains a valuable table of both recommended and cytotoxic concentrations of antibiotics used in tissue culture.

IV. METHODS

A. Preparation of Fish for Obtaining Tissues In marked contrast to reptiles, birds, and mammals, the outermost layer of the fish integument is living, and if the animal is healthy and comes from a clean environment the microbial flora is very low in number. The skin may support protozoan and higher parasites. The kind of tissue needed, the habits and ecology of the donor fish, the nature and purpose of the work, and whether or not antibiotics are to be used will all be factors in deciding which decontaminating, disinfecting, or sterilizing procedures should be used, No antimicrobial substance is without effect on the host tissue, and the rigors of decontamination should not result in death of the fish cells or tissues to be cultured. Unless proper precautions are taken, the results of the most thorough disinfection or sterilization of fish surfaces can be nullaed quickly by contamination from draining feces or regurgitated stomach contents. This problem can be minimized by withholding food for several days before using the fish, but where this is neither practical nor feasible much of the feces can be manually stripped from the fish and the mouth thoroughly rinsed prior to disinfection. 1. EXTERNAL TISSUES

The use of antibiotics in culture media has greatly reduced-but not eliminated-the problems associated with cell culture in growing external fish tissues with minimal microbial contamination. Fin, skin, barbels, cornea, and even caudad trunk portions of healthy fish may simply be washed in cold (preferably chlorinated) tap water and rinsed in sterile BSS. The BF-2 cell line was initiated from four small bluegills which were

274

KEN WOLF AND M. C. QUIMBY

so treated. As specific tissues are removed they are rinsed in cold BSSeither in vessels or in a stream from a syringe and small gage needleand placed in a sterile covered container. We have used this procedure routinely for years with fin tissue and on occasion with corneal tissue. Gill tissue, which grows well, is often heavily contaminated with gramnegative bacteria, and is difficult to clean for culture. Polymyxin B is one of the few bactericidal antibiotics, and we have had some success with hour-long treatments of gill arches in BSS containing 2000 IU/ml. Grand, Gordon, and Cameron (1941) and Grand and Cameron (1948) used a 1:10,000 solution of sodium Merthiolate for 2-3 min followed by a wash of sterile physiological saline. Griitzner (1956a) held guppies for 1-2 days in water containing 100CL1500 pg streptomycin per milliliter of water, then treated the freshly killed fish for about 5 min in 0 4 % chloramine (sodium p-toluene-sulfonchloramide) in BSS. From 88 to 95% of her cultures were sterile. Babini and Ghittino (1961) washed yolk sac fry in 1:200,000 malachite green for 15 min followed by immersion in 1:10,000 Merthiolate for 10-15 min then a rinse in sterile water containing penicillin and streptomycin. Another method is to soak the tissue in Dakin's solution for 2-3 min with agitation, then wash it in three changes of sterile BSS (Beasley, 1968). The RTF-1 cell line was started from nonfeeding yolk sac rainbow trout fry from which we removed the yolk then simply washed the fry several times in cold sterile phosphate buffered saline (PBS) (Wolf and Quimby, 1968). Fryer (1964) similarly found that rinsing embryonic or yolk sac salmonid fry was adequate decontamination. Such procedures will probably be adequate only if the fish have not yet begun to feed and thereby have yet to acquire an internal flora. Where caudad trunk tissue is used for celI culture, external tissue which may be killed by sterilization comprises only a small percentage of the biomass and therefore its loss will usually be negligible. In obtaining such tissue for establishing the FHM cell line Gravel1 and Malsberger (1965) immersed the donor minnows for 1 min in a filtered solution of 10%calcium hypochlorite and followed that with a rinse in 70!Z ethanol. The tissue was then given three 10-min washes in BSS containing penicillin and streptomycin. 2. EMBRYOS Sterile embryos may be obtained by surface sterilization of either eggs or gravid females. Dederer (1921) immersed eggs for 1 sec in 95% ethanol and then transferred them to sterile water for aseptic removal of the embryo. Soret and Sanders (1954) immersed gravid Gumbusiu

5.

FISH CELL AND TISSUE CULTURE

275

momentarily in Merthiolate ( concentration not given) and followed with two washings in 70%alcohol. They dried the fish with sterile cotton and aseptically removed the embryos. We have used benzalkonium chloride followed by alcohol (see Section IV, A, 3) to disinfect sharks prior to surgical opening and collection of sterile embryos from developing eggs in the oviducts.

3. INTERNAL TISSUES UnIess an animal is infected, and with the exception of the digestive tract, internal tissues of fishes are sterile and their aseptic removal is simple. Prior to opening the fish, the area of incision-or when feasible, the entire fish-is topically disinfected or sterilized. It is advantageous to remove scales from heavily scaled fishes. Wolf and Dunbar (1957) used procedures developed by Dr. S. F. Snieszko for bacteriological examination of fish. The fish is bathed for several minutes in a 1 :1000

Fig. 1. Illustration of a convenient approach used for the aseptic removal of sterile internal tissues from adult donor brook trout. Hemostat is clamped on anterior attachment of left ovary. Note rack for holding sterile instruments in shielding tubes and nearby boiling water and Coplin jar of ethanol for disinfection of contaminated tools.

276

KEN WOLF AND M. C . QUIMBY

solution of benzalkonium chloride, and this is followed by washing in 70% ethanol. Isopropanol (70%)is equally satisfactory and we have since used 1 :2000 household bleach (5.25%sodium hypochlorite) or about 500 ppm available chlorine interchangeably with benzalkonium chloride. Hypochlorite solutions have the advantage of wide availability and low cost. For obtaining sterile internal larval lamprey tissue, Stephenson and Potter (1967) used a disinfecting technique developed for tadpoles (Stephenson, 1967). The ammocoetes were rinsed in tap water, washed in detergent (7X brand) and immersed briefly in 95% alcohol, then passed through two changes of BSS containing 40 IU of penicillin, 50 pg of streptomycin, and 20 IU of nystatin per milliliter. Using sterile instruments, internal tissues are obtained by removing the lateral musculature covering the peritoneum and if the heart is needed, the pericardium (Fig. 1).If necessary, soiled instruments may be wiped clean, boiled briefly in distilled water, then transferred to 70% ethanol. Sterile tubes provide protection for the instruments between use. We have used a series of washes in sterile water to reduce external contamination and then held larval lampreys for 2 hr in sterile water containing 500 IU of polymyxin B, 500 pg of neomycin, and 40 IU of bacitracin per milliliter. The rate of contamination of external tissue explants was less than 5% (Wolf and Quimby, 1968). B. Preparation of Primary or Original Monolayer Cell Cultures

1. FRESHWATER TELEOST Disaggregation or partial "digestion" of tissues to obtain fish cells for primary monolayer culture follows the general procedures used with avian and mammalian tissues. An important exception however is temperature; to maintain viable fish cells throughout ' the procedures temperatures should not exceed 2Oo-25"C. There is, moreover, considerable evidence that lower temperatures are often desirable. Tryptic enzymes are employed most often, being used at a final concentration of 0.25% in a PBS having a pH preferably near 7.2-7.4 but not exceeding 7.6. As with the salines, some workers have diluted the PBS 2030% but many do not-the decision has usually been based upon the serum freezing point of the fish involved. Digestions are either relatively short trypsinizations at temperatures between 15" and 20°C or extended digestion at 4°-60C. The tissues should be fresh, but with adequate precautions to prevent damaging dehydration, surplus tissues may be safely stored at 4°C

5.

FISH CELL AND TISSUE CULTURE

277

for a day. On occasion, Roberts (1968) has had viable tissue after 48 hr storage at 4°C and once kept salmon ovary for 72 hr. Tissue is minced in BSS or PBS until the largest pieces are at most several millimeters in diameter. Fish embryos can also be mechanically disrupted simply by forcing them through a syringe several times. The tissue pieces are usually washed several times to remove debris and blood cells which are undesirable in the cultures. Griitzner (1958) introduced modem methodolgy to fish cell culture; she used a relatively brief digestion period and cultured tench liver and kidney cells in media which were diluted from mammalian concentrations. Trypsin at 0.25%was used in Dulbecco and Vogt’s PBS which had been diluted 20%with water (1954). After a 10-min mixing over a magnetic stirrer, the supernatant was discarded and fresh trypsin solution added. Harvests of separated cells were made at 30 min intervals, cooled with ice, and sedimented by gentle centrifugation. She cautioned that the temperature during digestion should not exceed 20°C. In later work (Pfitzner and Griitzner, 1964) mention was made that 4-5 hr digestion was so damaging to cells that they would not grow, and that overnight digestion at 4°C did not yield cultivable cells. Subsequently, however, tench gonads were trypsinized routinely at 4°C (Pfitzner, 1968). Kunst (1962) used much the same procedure for preparing carp kidney cell monolayers; the temperature was 18°C and he used intermittent agitation instead of continual mixing for the tissue which was held in four volumes of digestant. More recently, Pfitzner and Froehlich (1968) extended the methods to gonadal cells from carp and goldfish. Working principally with salmonids, Wolf et al. ( 1 9 0 ) found that slight modification of the so-called cold trypsinization used by Bodian (1956) for monkey kidney cells consistently gave cultivable cells from these and other fishes and also from an amphibian and a reptile. Dulb e m and Vogt’s PBS was modified by buffering to about p H 7.2, and it constituted 87.5%of the digestion mixture. Trypsin solution (2.5%of 1:250) accounted for 10%and the remaining 2.5%was serum which was added to protect the released cells.* Penicillin and streptomycin were present at 200 IU and 200 pg/ml, respectively. The ratio of tissue to digestion mixture was not critical; up to several volumes of minced tissue are used for each 10 volumes of digestant. The digestion itself is carried out on magnetic stirrers at 4°C. Cultivable cells are released during the first hour’s treatment, but they are a minority and seldom worth culturing; they should be discarded. Digestion is allowed to proceed until *Although an excess of serum will inhibit trypsin, this low level does not prevent digestion. Moreover the survival of cells in parallel digestions with and without serum is decidedly greater in the presence of a low level of serum.

278

KEN WOLF AND M. C. QUIMBY

many cells and cell clusters are in suspension (Fig. 2). This takes several hours. The digestion can proceed overnight, then cells may be harvested and fresh trypsin added for further action. Because earlier work (Wolf and Dunbar, 1957) had shown mammalian osmolarity to be appropriate for trout and goldfish, neither the digestion mixture nor culture medium was altered. Babini and Ghittino ( 1961) carried out cold digestions of carp, tench, and trout tissues in Dulbecco and Vogt’s PBS which was diluted 20%.Others have successfully used the cold digestion in undiluted PBS (Jensen, 1963; Roberts, 1964, 1966; Li and Stewart, 1965; Gravel1 and Malsberger, 1965). Kunst and Fijan (1966) trypsinized carp ovarian tissue at 4°C for both 3 and 20 hr and at 20°C for 0.5 and 2 hr. They found that the prolonged trypsinization at 4OC gave the best results. Fryer ( 1964) and Fryer et al. (1965) successfully used an intermedi-

Fig. 2. An example of a trout kidney “‘digestion”at the time of harvesting. Note the many small fragments of nondispersed tissue including portions of tubules. Cilia in this preparation were highly active, and trypan blue dye exclusion indicated a viability of over 95%. We consider such preparations to be ideal for primary monolayer culture.

5.

FISH CELL AND TISSUE CULTURE

279

ate temperature, 15"C, trypsin in pH 7.4 Hanks' BSS, and 30 min digestion periods for successful disaggregation of salmonid tissues. Trypsinization for primary cultivation of freshwater teIeost cells has become routine and methodology is now omitted from some reports on fish cell culture (Rachlin et al., 1967; McFalls et al., 1967). Some teleost livers, kidneys, swim bladders, and ovaries possess ciliated cells. The student and newcomer who undertakes fish cell culture for the &st time is urged to consider using these tissues for their beginning efforts. Positive assessment of viability can be determined at any step in the procedures simply by examining the released cells and tissue fragments for ciliary activity. Although trypsin was used in the work thus far reported, other enzymes have been tried on fish tissues. Fryer (1964) tried collagenase, hyaluronidase, pancrease, papain, and Vibrio comma receptor destroying enzyme (N-acetylneuraminidase ) . While all were enzymically effective, trypsin had the least deleterious effect upon the cells. We have used pronase for digestions but find no particular advantage over trypsin; instead, it is more expensive. Pancreatin, a crude tryptic enzyme may be used interchangeably with trypsin, but it can be toxic to cells of some fishes (Wolf and Quimby, 1988). 2. h h m TELEOST ~ ~

The only detailed work on preparation of monolayer cell cultures from marine teleosts is that of Clem et al. (1961). Their digestion mixture was prepared with magnesium-free Hanks' BSS. For some work the BSS was modified with about 0.07 M additional NaCl. Trypsinization was carried out at room temperature with harvests at about 1 hr intervals. Roberts (1966) prepared primary monolayer cell cultures from a marine clupeid with procedures used for freshwater teleosts. He noted that the cells were highly vacuolated and considered the abnormal condition possibly a consequence of the hypotonic medium. 3. MARINEELASMOBRANCH There is no published information on preparation of monolayer cell cultures from the cartilaginous fishes. We have done a limited amount of work and have had limited success with spiny dogfish, Squalus acanthias, and the sandbar shark, Carcharinus milberti. Digestion was effected with 0.25%pancreatin in pH 7.3 PBS which was modified by the addition of 1.75-2.0 g % urea and 0.5-0.6 g % NaCI. The freezing point (-1.59OC) closely matched that of spiny dogfish serum (-1.6OC). At different times, digestions were carried out at

280

KEN WOLF AND M. C . QUIMBY

26", 15", and 4"C, but the best results were obtained at the two lower temperatures, At 15OC, harvests were made at 1 and 2 hr intervals and oviduct cell cilia were still beating in the mixture after a total of 12 hr digestion. After the suspended material was decanted, undigested fragments were left in the residual fluid and kept at 4°C where cilia continued to beat for 76 days. The best results were obtained from digestions at 4°C with harvests at 4 hr intervals. In order of productivity the following shark tissues yielded cultivable cells: embryo, oviduct, rectal gland, kidney, spleen, and heart. Ovary and pancreas were dispersed, but cells were not cultivable. Heat-inactivated homologous serum was toxic to the cells. C. Seeding Density for Primary Monolayer Cultures Following digestion of tissue, the cells to be cultured are harvested by centrifugation. It is generally agreed that 200 g for 10 min is both adequate and safe. Cells from many fishes readily tolerate centrifugation at 20°C or even higher, but frictional heating coupled with high ambient temperature may injure cells from cold-water fishes. The risk of heat damage may be avoided by chilling the cell suspension prior to centrifugation or by centrifugation in the cold. It is worth noting that cells from a wide variety of vertebrates are safely centrifuged at 4"-8"C. Primary monolayer cultures of fish cells have most often been seeded with cell densities which have been established by counting, but to a lesser extent volumetric dilution of the harvested cell pack has also been used. There is a considerable range of cell numbers reported to be necessary to establish primary fish cell monolayers. A majority of the reports cite values well above those normally used for primary cultures of homoiotherm cells, but a minority found that 1 to 3 x lo5 cells/ml were adequate. It seems likely that different tissues and even different species of fish could account for some of the differences. We believe however that other factors are more important and that the composition and properties of the medium, and especially the extent and conditions of digestion, temperature, physical forces, and possibly other factors of centrifugation critically affect the minimal threshold of cells necessary for primary monolayer cultures. Griitzner (1958) found 2 to 2.5 x lo5 tench liver or kidney cells to be inadequate and that 2, 3, or even 4 times as many cells were needed to establish monolayers. Her procedures included 30-40 min of centrifugation, but relative centrifugal force and temperature were not given. From marine teleost tissues, Clem et al. (1961) initiated cultures with

5.

FISH CELL AND TISSUE CIJL”uRJ3

281

4 to 5 x lo5 cells/ml when centrifugation was not used. Following 7 min at 5°C ( g not given), they reported it necessary to use 8-10 x lo5 cells/ml to establish primary cultures. Fryer (1964) and Fryer et ul. (1965) considered cell concentration to be an “extremely critical factor” in establishing cultures. Cells were harvested with 20 min centrifugation (temperature and g not given), resuspended and washed in medium, and the minimal number necessary was about 6 x lo5 cells/ml. For routine purposes however, the density was 1 to 1.5 x lo6 cells/ml. Subculturing-which usually requires only one centrifugation-was successful with as few as 3 x lo5 cells/ml though 6 x lo5 cells/ml was preferred. The fewer cells required for subcultures could reflect enhanced survival with less centrifugation, but we recognize too that subcultures are usually more amenable to handling than are primary cells. Jensen (1963) used 10 min centrifugation ( g not given) in co2d tubes (italics added) and cultured trout ovary monolayers with 3 x lo5 cells/ml. Li and Stewart (1965) harvested trout ovary cells with 35 g for 5 min in the cold and found 2 x lo5 cells/ml to be adequate for establishing monolayers. Volumetric dilution of centrifuged cell packs for seeding primary cultures is simpler and takes less time than counting. As determined by trypan blue dye exclusion but especially by ciliary activity we have had consistently high percentages of viable cells in digests of poikilothermic vertebrate tissues which have been stopped while small fragments of undispersed tissue were still present (Fig. 2 ) . The counting of cells in such harvests is tedious and the accuracy is unknown, accordingly we use volumetric dilution and suspend the cells in from 400 to 600 volumes of medium ( 1 to 3 x lo5 cells/ml) (Wolf et ul., 1960). We used two centrifugations but carried them out at 200 g for 20 min at 4°C. We have subsequently eliminated one centrifugation because it was found that washing the cells was not necessary; also, centrifugation time was reduced to 10 min. Babini and Ghittino (1961) also used volumetric dilution but at much lower ratios-1:25 to 1:lOO. Kunst (1962) used a volumetric dilution of 1:100 for carp ovary cells but in later work compared ratios of 1:300, 1:600, and 1:1200 and found the latter to be the the best ( Kunst and Fijan, 1966). Although the relative centrifugal force and temperature were not given, they used two centrifugations each of 10 min duration.

D. Dispersion of Monolayer Cell Cultures Monolayer cell cultures may be dispersed either mechanically or chemically for subculturing. At present, the use of a chelating agent with a tryptic enzyme is superior to other methods.

282

KEN WOLF AND M. C. QUIMBY

Fish cells differ in the tenacity of their attachment to each other and to the substrate. In part this reflects differences between cell lines; but media, culture conditions, and even culture age affect the ease or difEculty with which cells may be dispersed. In our experience old cultures are more cohesive than young cultures, and cells grown in medium with 10-15% serum are more cohesive than cells grown in 5% serum. Dispersed fish cell lines can be easily heat-fixed for stabilization and delayed counting. Preparations held at 57'40°C for 30 min may be counted accurately through several days' storage (Wellborn and Wolf, 1961) . 1. MECHANICALDISPERSION

There is only occasional justification for using mechanical means to disperse cell sheets. The objections to this method are that even under the best conditions many cells are destroyed and the dispersion itself is usually incomplete-the sheet being broken into fragments rather than individual cells. This of course compounds the problems of accuracy in enumeration. The cell sheet is scraped from the surface-with a rubber policeman or glass culture scraper-into a sufficiently small volume of fresh medium to permit vigorous pipetting for additional dispersion. Estimates of cell loss have been as high as 50%(Cerini, 1964; Malsberger and Cerini, 1965). 2. CHEMICAL DISPERSION

Disodium versenate is a chelating agent for divalent cations. A 1:5000 solution is used to chelate calcium which is essential for cell cohesion and adhesion. Cell sheets are drained of medium and covered with ethylenediaminetetraacetate ( EDTA ) prepared in a calcium and magnesium-free salt solution (Merchant et al., 1964).After &lo min, the action is stopped by adding an excess of calcium (the original medium is both effective and economical for this purpose) and the sheet fragments pipetted vigorously for additional dispersion. The cells must be sedimented by centrifugation, the supernate decanted, then the pellet of cells can be resuspended in fresh medium. Temperature should be maintained within the cells' tolerance. A 0.25%solution of trypsin or pancreatin in either BSS or PBS can be used in place of EDTA. An alternate procedure eliminates the need for centrifugation. The cell sheet is covered with trypsin solution and watched carefully. When the very first cells are loosened-a matter of one to several minutes-the vessel is rotated 180" to invert the cell sheet. Most of the trypsin is de-

5.

283

FISH CELL AND TISSUE CULTURE

Table V EDTA-Trypsin Solution for Dispersion of Monolayer Cell Cultures Constituent

Quantity

Water NaCl KHtPO, KC1 NaZHP04 Disodium EDTA

960 ml 8.00 g 0.20g 0.20 g 1.15 g 0.20 g"

Fisher S-311 or equivalent. May be filter sterilized or autoclaved. Add 40 ml of sterile 2.5% trypsin solution, dispense in appropriate quantities and store a t -20°C or lower. Mixture may be thawed and refrozen several times.

canted, but a small residual volume is retained and by gentle rocking the fluid is made to flow back and forth across the sheet until the cells are free. This usually takes 5-10 min and at that time a small amount of fresh medium is added and the cells dispersed by pipetting. The enzymic action is inhibited by serum, and the cells can be diluted as necessary. In our experience, the best dispersion is achieved by a combination of 1:5OOO EDTA (20 mg %) and 0.25%trypsin or pancreatin in calcium and magnesium-free salt solution used as described in the preceding paragraph (Table V). Cells so dispersed have had a high survival. Viability as determined by trypan blue dye exclusion has been about 95%.

E. Seeding Density for Cell Lines The seeding density for subcultures of cell lines will vary with the cell, the medium, and the particular need. Where rapidity of growth is important and other factors are equal the lag phase will be shortest at high population densities. In routine use, fish cell lines are split at ratios of 1:5 to 1:20, but if a need arises well-established lines may be split 1 :100 or even 1 :1000. In terms of cells per milliliter of fresh medium, routine seeding densities would be from lo4 to lo5 cells/ml or more, but as few as lo3 or even lo2 cells/ml can be adequate under good conditions. Of course with the latter low numbers, the lag phase will be considerable. Using initial seeding densities and different incubating temperatures as the variables, quantitative growth studies have been carried out with the RTG-2 cell line (Plumb and Wolf, 1969). Replicate culture tubes were seeded at densities of 1.25 x 104 and at twofold increases to 2 X lo5

284

KEN WOLF AND M. C . QUIMBY

Table VI Effects of Seeding Density and Temperature on Growth of RTG-2 Cells Days required to reach confluency (2.75 X 105 cells) a t incubating temperatures of Seeding density

30°C

25°C

20°C

15°C

10°C

5°C

1.5 3.1 4.8 7.5 10.5

2.4 4.8 7.3 14.0 20.0

3.5 9.3 14.0 35.0 -

8.0 14.0

63.0 70.0

2.0

3.5

7.6

13.2

~

Death 1.8 200,000 3.3 Death 100,000 50,000 Death 6.1 25,000 Death 8.5 12,500 Death 14.0 Mean population doubling time (days) 2.5

-

celIs/ml and incubated at temperature increments of 5" from 5" to 30°C. Confluent cultures were found to have a mean population of 2.75 x lo5 cells/ml. As shown in Table VI, cultures seeded most heavily became confluent in 1.5 days at 20"C, the cells' optimal temperature, while the lowest density required 70 days to become confluent at 5°C. The data clearly show that the cells grow more rapidly at 20°C than at 25°C.

F. Precautions to Be Taken with Fish and Other Poikilothem Cells It is a common practice to grow homoiotherm cells at 37°C in the presence of 5%CO,, and many of the media use buffer systems which equilibrate in the physiological range under such conditions. Because the solubility of CO, is increased at lower temperatures, media which equilibrate at pH 7.4 at 37°C become acid when the temperature of CO, cabinets is lowered. It is a common mistake to use CO, cabinets in handling fish and other poikilotherm cells, and at the very low pH the cells are soon killed. Excessive heat during enzymic disaggregation of poikilotherm tissues -especially those from cold-water species such as salmonids-is likely to damage many cells and may in fact kill all of them. Since digestions are usually stirred with a magnetic unit, the principal source of heat is from the stirring apparatus itself. Stirrers with a separate rheostat in the power line are strongly suggested. Although they are more compact, stirrers with a self-contained rheostat may transfer excessive heat to the cells. To prevent heat transfer, one may use insulation between the stirrer and the flask or keep the flask in a tray with crushed ice or cold water.

5.

FISH CELL AND TISSUE CULTURE

285

Cells in culture lack most of the defense mechanisms of the intact donor and are usually quite vulnerable to virus infection; this property makes them extremely valuable as a tool in virology. As a broad generalization, primary cell cultures are usually susceptible to more viruses than are their derived permanent cell lines, in fact during their development, cell lines often lose susceptibility to some viruses. Primary cultures however can have disadvantages. Removing tissues from the host usually strips away defense mechanisms. If such tissues harbor a latent viruswhich sometimes happens-the virus can result in a fulminating infection which may destroy the culture. Latent IPN virus has been unmasked a number of times in primary trout cell cultures (Malsberger and Cerini, 1963). Thus far, however, less obvious viruses which replicate without cytopathology have not been documented in primary fish cell cultures. The solution to the problem is to attempt to learn beforehand the health of the donor animals. It is routine to carry out sterility checks on media as they are prepared for cell culture. For this purpose, aliquots of complete medium without antibiotics can be incubated as a test procedure. Bacteriological media such as thioglycollate or enrichment broths are usually used for testing salines and other solutions. On a number of occasions we have found bacterial contaminants which grew poorly in bacteriological medium but luxuriantly in cell culture medium; accordingly, we routinely use the latter as sterility test medium. Mycoplasma, formerly called pleuropneumonia-like organisms or simply PPLO’s, are microbial contaminants of significance in mammalian cell culture. Forms which are of human origin prevail, but mammalian serum sources are a secondary source. Thus far, no lower vertebrate cell line has been found contaminated with a mycoplasma. Whether or not this situation will prevail is not known, but maintenance of cell stock cultures by strict aseptic technique and without antibiotics is recommended. Most fish, especially those from the wild, may have protozoan or metazoan parasites located internally or externally which can create problems for the fish cell or tissue culturist. One should be aware of them when planning for primary cultures. As an example, some female largemouth bass may have ovaries which are so heavily infested with cestodes that it is impossible to obtain unadulterated fish tissue. Kidneys, liver, and heart of many centrarchids are commonly riddled with metacercarial trematode cysts, Mincing and trypsinization will release these minute worms and they will live for a week or more in cultures of the fish cells. While tissue from such infested sources may be useful for single usage short-term applications, it should not be used for critical work such as cell line development or other long-term applications.

286

KEN WOLF AND M. C. QUIMBY

V. CHOICE OF TISSUES FOR CULTURE

Critical comparisons have not been made, but cells and tissues from adult fishes seem to be cultured somewhat more easily than tissues from adult mammals, Such an impression is consistent with the biology of fishes for they generally continue to grow throughout life and have comparatively strong regenerative ability-as, for example, when fins are amputated. There is nevertheless a hierarchy among fish tissues; some are grown much more readily than others, and some-such as adult skeletal muscle--seem disinclined to grow in uitro. With little doubt, embryonic fish cells are the most readily cultivated, and their state of active mitosis lends impetus to continued division in culture. Embryonic tissue has the added advantages of usually being bacteriologically sterile, of having a low probability of harboring latent virus, and of having relatively little cell storage products and a small complement of differentiated cells. The disadvantages of embryonic tissue are that the available mass is generally small. In' addition, embryos are often inaccessibleas with pelagic spawning species; moreover, with many fishes embryos are available only during a limited portion of the year. The second most easily cultured fish tissue is gonad-preferably juvenile or immature organs because there is a greater proportion of undifferentiated tissue and more actual germinal tissue. Ovary is generally preferred over testes and may yield abundant cultivable cells at full maturity; the great mass of ovary at that time however is yolk and shell, and while it is acellular and noncultivable, it does not seem to interfere with the culturing of cells and tissue. The cultivability of gonadal tissue is a feature that applies broadly among the lower vertebrates. In our experience, the sole exception has been adult shark ovary which we have not been able to culture successfully. Townsley et al. (1963) failed to grow gonadal tissue from skate and a marine teleost. Swim bladder, fin, mesentery, cornea, gills, heart, and skin generally culture wen. Cornea and swim bladder are usually clean tissues which lend themselves well to neat explants. We have had tro,ut swim bladder explant show persistaltic-like movement for over a year. Although it is usually cluttered with acellular debris, hematopoietic tissue such as spleen and kidneys usually fare well in culture. Strangely enough, circulating leukcoytes from teleosts may be induced to attach to culture vessel surfaces and form monolayers, but there is little if any mitosis. This may be a result of their being differentiated; if so, thymus and

5.

FISH CELL AND TISSUE CULTURE

287

anterior kidney tissue-possibly taken by needle biopsy-may yield active stem cells. Cell culture thus could clarify some of the uncertainty existing in fish leukocyte phylogeny and nomenclature. Liver tissue has responded erratically in culture. Griitzner (1958) apparently found tench liver to grow well, but Pfitzner (1968) noted it could only occasionally be cultivated. Others have also had mixed results with liver tissue (Wolf and Dunbar, 1957; Clem et al., 1961; Townsley et al., 1963). Babini and Ghittino (1961) and Fryer (1964) were unable to grow normal liver tissue. On the other hand, Fryer (1964) found that rainbow trout hepatomatous tissue readily yielded cultivable cells, and he established a permanent line of cells from such tissue. This probably reflects derepression of controls that exist for normal tissue, and tumor tissue almost traditionally has been easily grown in culture. A number of different fish tumors have been cultivated (Grand et aE., 1941; Grand and Cameron, 1948; Schlumberger, 1949; Greenberg et al., 1956). We have been unable to grow tissue from the corpuscles of Stannius, but Lewis and MacNeal (1935) grew pituitary and McLimans ( 1961) grew fish pancreas. As far as we know, no other fish endocrine tissues have been cultivated. In principle, tissue from the alimentary canal should respond favorably. We have explanted stomach and intestinal tissue a number of times but were unable to control the bacterial contamination.

VI. STORAGE AND PRESERVATION Freezing and long-term storage of living cells at ultra-low temperature has been practiced for about 15 years. The method that has been used so successfully for mammalian cells, works equally well with fish cells. Thus far, however, there are few published data on results of freezing and recovering fish cells. Cells from many poikilotherms not only tolerate, but also actually metabolize, at temperatures near freezing. The rate of metabolism is greatly reduced at low temperatures; thus, in the culture of cold-blooded animal cells one also has the advantage of prolonged storage without freezing.

A. Freezing In general, cells are suspended in medium having 10%or more serum and 510% of either glycerol or dimethyl sulfoxide (DMSO) as a pro-

288

KEN WOLF AND M. C . QUIMBY

tective additive. The suspension is sealed in glass to prevent dehydration and to exclude CO, if refrigeration is to be with Dry Ice. After sealing, cells are allowed to equilibrate at 5°C for 30-60 min. For best results freezing should be done slowly by cooling l"C/min to -25°C then transferring to -65°C or lower. For recovery, ampoules are thawed in less than 1 min and diluted for planting. To date, the best single source of published data on freezing fish cells is the American Type Culture Collection, Registry of Animal Cell Lines Certified by the Cell Culture Collection Committee (1964 with first and second supplements). The following tabulation lists fish cell lines that freeze well in the media given. Cell line

Freeze medium

FHM (CCL 42)" Eagle's MEM (Hanks') 85%' calf serum lo%, DMSO 5% RTG-2 (CCL 55) Eagle's BME SO%, fetal bovine serum 15%, DMSO 5% GF (CCL 58) Eagle's BME with nonessential amino acids (Hanks' with 0.196 M NaC1) 75%, calf serum lo%, human serum lo%, DMSO 5% Certified Cell Line.

During the past 4 years we have frozen fish and frog cells with the following procedures. Freeze medium consists of Eagle's MEM ( Earle's BSS) 85%,fetal bovine serum lo%,and either glycerol or DMSO 5%.The medium is cooled to 4"C, and cells are added to a density of 2 to 6 x lo6 cells/ml. One milliliter aliquots are sealed in 2 ml ampoules and cells are allowed to equilibrate at 4°C for about an hour. Ampoules are wrapped in several layers of flexible insulation, placed in an expanded polystyrene bead insulated container and moved to -80°C. Because of the insulation the heat loss is slow and good freezing is assured. For recovery the ampoules are thawed with rapid agitation in water not over 20"C, then the cell suspension is added to 6-8 volumes of growth medium and planted. Most of the viable cells attach within an hour. Following attachment, the freeze medium should be withdrawn and replaced with medium having no preservative. Storage at -80°C gave good recovery. Although their immediate appearance following thawing and planting is not as good as with DMSO, cells frozen with glycerol survive longer in unopened ampoules at 20°C. We have held unopened ampoules of RTG-2 cells for 5 days and recovered live cells-a sufficient time to permit mailing. Using the above procedures we have successfully frozen, thawed, and cultured RTG-2, RTF-1, FHM, BF-2, and frog tongue fibroblasts. Cells from vigorously growing cultures are preferred. Kunst and Fijan (1966) froze primary carp ovary cells in Eagle's

5.

FISH CELL AND TISSUE CULTURE

289

BME with 10%calf serum and 8% glycerol. They used 60 volumes of medium per 1 volume of cells and stored the suspensions at -70°C.

B. Low Temperature Incubation Cells from some but not all fishes have been grown at temperatures of 5"-1OoC. It seems likely that the physiology of the fish and the quality of the medium are two of the factors involved, and there may be additional factors. The RTG-2 cell line (from a "cold-water fish") grows well from 26" to 4°C and possibly lower, and rates of protein synthesis and glucose utilization have been measured through that temperature range (Wolf et al., 1962). Because of the reduced growth at low temperatures we have kept some cell stocks at low levels of activity for many months. Cultures can be incubated at 12"-15"C and subcultured at 3 4 month intervals (Wolf and Quimby, 1967). FHM cells are reported to grow at 14"C, but there is little activity at 4°C (Gravel1 and Malsberger, 1965). We routinely incubate some fish cells at 4°C. By seeding bottles at about 1.5 to 2.5 x 104 cells/ml, cultures of RTG-2 cells have been stored for 2 years at 4°C. The only attention given them is a visual check at 3 month intervals. It has recently been determined (Plumb and Wolf, 1969) that RTG-2 cells at 4°C have a population doubling time of about 13.2 days. In contrast, the doubling time at 2OoC, which is near the optimum, is only 2 days. The medium for this work has uniformly been Eagle's MEM (Earle's) 90%and fetal bovine serum 10%While we have no quantitative data, it is our opinion that our stocks of FHM and BF-2 cells also metabolize and divide for months at 4°C. Stephenson (1967) attempted culture of lamprey tissue at 5°C but reported that no outgrowth occurred. On the other hand, Townsley et al. ( 1963), who were working with cold water marine teleost expIants, chose 5°C as the most desirable temperature for their work.

VII. FISH CELL LINES

A. General The fish cell lines which have been established (see Fedoroff, 1966, for terminology) and those which are in development are all of teleost origin (Table VII). Seven of the lines are from salmonids, six are from

Table VII

Cell& morpholOgY

Mediumb

Embryo Embryo Embryo Hepatoma Embryo

F F F E E

MEM 20% ACS MEM $. 20% ACS MEM 20% ACS MEM 20%ACS MEM 10% FBS

RTG-2 (CCL-55)r 8.gairdneri

Gonad

F

+ + + + MEM + 10% FBS

S. gairdneri RTO FHM (CCG42)o P . promelas

Ovary Caudad trunk

F E

199 5% FBSh MEM $. 10% FBS

MEM FBS (8: 2 HzO) MEM 10% CS or FBS BME 10% CS and 10% HS MEM 10% FBS

Designation CSE 119 CHSE 114 STE 137 RTH 149 RTF-1

Species

0. kisutch 0. tschawytscha S. gairdneri S. gairdneri S. gairdneri

Tissue

+

+

ZDG-1

B. rerio

Gonad

F

BB

I . nebulosus

F

SF

L. griseus

Trunk w/o viscera Fin

F

BF-2

L. macrochirus

Caudad trunk

F

BGL

L. macrochirus

Caudad trunk

F

+ + + MEM + 10% FBS

LBF-1

M . salmoides

Caudad trunk

F

MEM f 10% FBS

GF-1 (CCL-58)S

H . sciurus

Fin

E

GFS

H . pavolineatum Fin

E

ESF-1

P . vetulus

F

BME 10% CS and 10% HSi BME 10% CS and 10% HS' 10% FBSi MEM

Fin

PHC preference 7.4 7.4 7.4 7.4 7.4-7.8 7.3-7.6

7.2 7.2

7.0-7.2 7.4 7.0-7.2 7.4-7.6 7.2-7.4 7.4

+ +

7.0-7.2

+

7.5

7.0-7.2

Here F stands for fibroblastic and E for epithelioid. Abbreviations: minimal essential medium of Eagle (MEM), Eagle's basal medium (BME), fetal bovine serum (FBS), human serum (HS), calf serum (CS), bovine serum (BS), and agamma calf serum (ACS). c Initial pH of newly seeded cultures. Tolerance may extend above and below values shown. d Tolerance may extend below and above the values shown. * Abbreviations: centrarchid myxovirus (CM), chinook salmon virus (CSV), eastern equine encephalomyelitis (EEE), frog virus (FV), grunt fin agent (GFA), infectious pancreatic necrosis (IPN), lymphocystis of freshwater fishes [L(FW)], lymphocystk of marine fishes [L(M)], tadpole edema virus ('JXV), Venezuelan equine encephalomyelitis (VEE), vesicular stomatitis virus a

b

(VSV).

290

Fish Cell Lines: Established and in Development 1968 Temperatured tolerance ~

Low Optimal High Year Current ("C) ("C) ("C) initiated passage 4 4 4

21 21 21 21

4

100 71 65 65 74

Unknown

26

1963 1963 1963 1963 1960

CM, CSV, EEE, Egtved, IPN, SSV, TEV, VEE, refractory to L(FW) and polio type 1 Unknown CSV, ECHO 11. FV-1, IPN, polio type 1, SSV, TEV, VSV, refractory to CM IPN

27 27

n

4

20

26

1960

162

0

18 34

36

1964 1962

20 90

21

25

1967

24

4

34

1962

67

20

25

1967

45

10

30

1964

48

35

1964

65

25

1964

17

25

1961

294

25

1965

90

25

1965

53

4

34

15 15

21

20 10

20

Virus susceptibility.

csv, ssv CSV, IPN, SSV Unknown IPN

Originator and' reference or location Fryer et al. (1965) Fryer et al. (1965) Fryer et al. (1965) Fryer et al. (1965) Wolf and Quimby E.F.D.L. Wolf and Quimby (1962)

Li and Stewart (1965) Gravell and Malsberger (1965)

Filazolla et al. N.Y.U. Cerini and Malsberger IPN (1962) Lehigh U. Horowitz et al. GFA, L(M), IPN U. Miami CSV, CM, IPN, L(FW), Wolf and Quimby E.F.D.L. TEV Gravell and Malsberger FV-1, FV-3, IPN, (1965) Lehigh U. UFW) Wolf and Quimby CM, TEV E.F.D.L. Clem et al. (1961) GFA, IPN, refractory to L(M) 6. Sigel et al. GFA, IPN, L(M) U. Miami R. C. Cooper Unknown U. Calif., Berkeley

Here E.F.D.L. stands for Eastern Fish Disease Laboratory, Kearneysville, W. Va., and N.Y.U. for New York University. Characterized as a Certified Cell Line and available from the cell repository of the American Type Culture Collection. h Best growth in medium additionally supplemented with 5% bovine embryo extract, and either 10% bovine serum or 10% human serum. i Requires 0.196 M NaC1. 2 Requires 0.227 M NaC1.

291

KEN WOLF AND M. C. QUIMBY

Fig. 3. Fixed and stained ( May-Griinwald-Giemsa) preparations of representative fish cell lines. ( A ) FHM, an established certified (CCL-42) freshwater epithelioid line from adult Pimephales promelas. ( B ) BF-2,a freshwater fibroblast line from juvenile Lepomis macrochirus currently in development. ( C ) BB, an established but as yet undescribed freshwater fibroblast line from juvenile ZctaZurus nebulaus. ( D ) RTG-2, an established certified (CCL-55) freshwater fibroblast line from subadult Salmo gairdneri. ( E ) GF-1, an established certified (CCL-58) marine epithelioid line from adult Haemdon sciurus. ( F ) SF, a marine fibroblast line from adult Lutianus griseus currently in development. ( G ) GFS, an established but as yet undescribed marine epithelioid line from adult Haemulon jkauolineatum. ( H ) RTF-1, an established but as yet undescribed freshwater epithelioid line from embryonic Salmo gairdneri.

5.

FISH CELL AND TISSUE CULTURE

293

strictly freshwater species, and four are from marine fishes ( Fig. 3 ) . With the exception of the lone aquarium species and the marine forms, all are from hatchery-propagated fishes of moderate to high economic value. The lines have been developed by people who are interested in the viruses and the viral diseases of poikilotherms, and most of the cells have been used for propagation of one or more fish viruses. Several of the lines support amphibian viruses as well, and FHM and RTG-2 cells are susceptible to some homoiotherm viruses (Gravel1 and Malsberger, 1965; Officer, 1964). Because of the work involved, chromosome numbers have been

294

KEN WOLF AND M. C. QUIMBY

determined for only about half of the extant fish cell lines, and these are all now heteroploid. The fish counterpart of the mammalian diploid cell line may have existed, but it has been lost. An embryonic sockeye salmon line, Se E (Wingfield, 1968), remained diploid for 50 subcultures but could not be continued (Fryer, 1968). Among cell lines from all animals, attainment of the potential for indefinite subculturing is usually accompanied by alteration to a heteroploid chromosome constitution. The modal chromosome number of such fish cell lines apparently is quite stable. The stability of karyotype is another matter and has been determined only for the FHM cell (Levan et al., 1966).

B. Sources of Fish Cell Lines Starter cuItures of fish cell lines may be obtained from the originator and if submitted, from the Cell Repository of the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland, 20852. Use of the latter spares the originator time, effort, and money but more importantly it provides Certified Cell Lines of known identity and of cultural quality possibly even exceeding that routinely available from the originator. Because of their widespread use, RTG-2 and FHM cells are available as starter cultures but also in production quantities from several of the cell culture supply houses. One may of course wish to develop his own cell line. For this we would suggest starting with many culture units (we prefer tubes) and using several different media. When the response of the cells can be evaluated, the efforts should concentrate on the most promising-whether it be vigor, appearance, or uniformity. If feasible, one should consider working without antibiotics, or withdrawing them from at least part of the cultures after several subcultures. Initially at least, subcultures should be made 1:2 or 1:3. If possible, a portion of the primary culture and each tenth passage should be frozen for subsequent chromosome analysis. Once a line is established, the originators should submit it to the American Type Culture Collection for characterization and accession.

VIII. SHIPMENT OF CELL CULTURES

Living fish cells may be shipped almost anywhere either in the frozen state with Dry Ice or as active cultures. Many hundreds of cultures have been sent from our laboratory by air and surface mail. With very few

5.

FISH CELL AND TISSUE CULTURE

295

exceptions, they arrived safely. High temperature-at least for salmonid cells-and drying are probably the greatest hazards in shipping active cultures. Regardless of how a package is marked, one can expect that at times the labels will be ignored, therefore, cultures will probably be inverted and the medium drained from the cells. Cell sheets should be up to 50%confluent and grown either in tubes or small bottles. For shipment the vessels are filled with medium and tightly capped. Insulated containers such as expanded bead polystyrene are preferred for they provide both thermal and physical protection. When high temperatures may be encountered, either ice in plastic bags or commercial canned refrigerant is added for cooling. Dry Ice should not be used. For domestic mailing, packages should be marked “outside mail” and “special handling.”

IX. NEEDED DEVELOPMENTS

The needs of today’s fish cell and tissue culture can be roughly divided into two categories: methodology and biology; separation of the two is sometimes difficult. Perhaps the greatest need is study of the physiology, nutrition, and metabolism of the cells themselves. Of course, one must specify the kind of fish cell. At this time we would suggest that it be of teleost origin because they are the dominant vertebrate class. Before such studies are undertaken, however, it would be most desirable to have the cells growing in a completely defined medium. Although much of any medium in use today for growing fish cells is chemically defined, the serum fractions (5-2M) , ultrafiltrates, embryo extracts, yeast extracts, and protein hydrolysates are almost completely undefined. The first need is to establish cells on completely defined medium. There are a number of defined media available commercially. One approach has been to grow cells in such media with serum then gradually reduce the concentration until none is required. This process of adaptation and selection is not difficult but it is time consuming. A more positive approach is to transfer cell lines to chemically defined medium or to establish primary cultures directly in completely synthetic medium. The next need in the fundamental study is a standardized cell. Ideally, such a cell or cells should be as near normal as possible. Such a cell line does not now exist, therefore development of fish counterparts to the human diploid cell lines is one approach that could be taken. In the

Table VIII W QJ

Fishes Used for Cell and Tissue Culture, 1914-1968 References Order.

Scientific name

Petromyzoniformes

Petromyzon jluviatilis (known as P . marinus in North America) Mordacia m r d a x

Squrtliformes

Species not given

Rajiformes

carcharinus milberti Squalus acanthias Raja sp. Raja rodiata

Explant

Monolayer

Class Petromyzones Chlopin (1925) Pfeiffer (1935) Wolf and Quimby (1968) Stephenson and Potter (1967) Class Elasmobranchii Cameron (1949) Lewis and MacNeal (1935) Wolf and Quimby (1968) Wolf and Quimby (1968) Lewis and MacNeal (1935) Townsley et al. (1963) Class Teleostomi

Clupeiformes

Clupea harengus Onwrhynchus kisutch

0. nerka 0. tshawylscha

Wolf and Quimby (1968)

Salmo gairdneri

Wolf and Dunbar (1957)

Roberts (1966) Fryer (1964) Fryer et al. (1965) Fryer (1964) Fryer (1964) Fryer el al. (1965) Babini and Ghittino (1961) Fryer f 1964) Fryer el al. (1965) Jensen (1963) Li and Stewart (1965) Pfitrner (1965)

S . salar S. trutta

Cypriniformes

Salvelinus fontinalis Esoz. sp. Caraasius auratus

Carassius carassiua Brachydunio rerio Cyprinus carpio

Osowski (1914) Wolf and Dunbar (1957) Wolf and Dunbar (1957) Chlopin (1928) Hu and Chavin (1960) Kim and Tchen (1963) Loud and Mishima (1963) Wolf and Dunbar (1957) Schlumberger (1949) Tchen et al. (1963) Chlopin (1928) Kunst (1961) Tec and Jakovleva (1962) TomaBec et al. (1964)

Pimephales promelas

Ptychocheilus oregonensis Tinca vulgaris

Cyprinodontiformea

letalurus nebulasus I . punetatus Fundulus h-eteroclitus

F. mnjalis Gambusia holbrookii

Dederer (1921) Goodrich (1924) Oppenheimer (1935) Goodrich (1924) Sanders and Soret (1954) Soret and Sanders (1954)

Wolf et al. (1960) Roberts (1968) Babini and Ghittino (1961)

Pfitzner and Froehlich (1966)

.vr

g

1 !i $

Pfitzner (1966) Rachlin et al. (1967) Babini and Ghittino (1961) Kunst (1962) Kunst and Fijan (1966) OsadEaja (1964) Pfitzner and Froehlich (1966) Gravel1 and Malsberger (1965) Fryer (1964) Babini and Ghittino (1961) Griitzner (1958) Pfitzner (1965) Pfitzner and Griitzner (1964) Cerini and Malsberger (1962) Wolf and Quimby (1968)

Table VIU (Continued) References

Order

Scientific name Lebistes reticulatus Xiphophorus hellerii

x.maculatus Gadiformes Perciformes

X . variatus Gadus mrhua Pollachius virens Diplectrum formsum Roccus amaricanus Lutjanus sp. Acuntharchus pomotis Ambloplites rupestris Centrarchus maeropterus Chrwbryttus gulosus Elassoma zonatum Enneacanthus glorwsus Lepomis auritus

Explant Bargen and Wessing (1960) Griitzner (1956a,b) Grand et al. (1941) Greenberg et al. (1956) Grand et al. (1941) Grand and Cameron (1948)

Monolayer Pfitzner (1965) McFalls et al. (1967) McFalls et al. (1967) McFalls et al. (1967)

Townsley et al. (1963) Townsley et al. (1963) Clem et al. (1961) Townsley et al. (1963) Clem et al. (1961) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964)

P n

I'. cyanellus L . gibbosus L. macrochirus L . microlophus Microplerus dolomieui M . salmoides Pomoxis nigromaculatus Anisotremus virginicus Haemulun jlavolinealum Haemulou seiurus

Pleuronectiformes Lophiiformes Syngnathiformes

Leiostomus xanthurus Pomacanthus sp. Calamus sp. cottidac Myoxocephalus scorpius Macropodus opercularis Flounder (species not given) Pseudopleuronectes americanus Lophius americanus L . piscatorius Hippocampus sp.

Orders of fishes according to Berg (1947).

Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Roberts (1964) Clem el al. (1961) Clem et al. (1961) Clem el al. (1965) Moewus and Sigel (1963) Wolf and Quimby (1968) Clem et al. (1961) Clem et al. (1961) Lewis and MacNeal (1935) Townsley et al. (1963) Griitzner (1956a,b) Lewis and MacNeal (1935) Townsley et al. (1963) Townsley el al. (1963) Lewis and MacNeal (1935) Moewus and Sigel (1963)

vr

wF

300

KEN WOLF AND M. C. QUIMBY

interim, another approach would be to select clones with stable karyotypes from existing cell lines. This is not a di5cult feat, but it does highlight still another need-trained researchers who are principally interested in the cells themselves rather than in the applications of the cells to other uses. Fish cells have not yet been grown satisfactorily in suspension culture. Large roller bottles which greatly enhance ability to grow many cells in a limited amount of space are helpful to the virologist, however, the biochemist and molecular biologist could also benefit by growing large batches of standard clones in chemically defined medium. In the way of methodology, fish cell culture is notably lacking in techniques for growing leukocytes. Strange as it may seem, fish ought to provide an easily cultivable leukocyte of some sort and although we and others have attempted this, these efforts have not been successful. The cytogeneticists, of course, would make immediate use of the technique as a nondestructive means of determining the chromosome complement of particular fish. Perhaps use of the thymus or efforts to culture biopsy specimens from spleen and anterior kidney would be productive. Conceivably such studies would shed light on development of the various fish leukocytes, a currently disputed area. There is also a real need for the application of organ culture to fish material. For example, the pseudobranch is a tissue which could be productively studied in uitro, and the corpuscles of Stannius-as yet with unknown function-may yield their secret in uitro. Today’s fish cell culture is largely te2eost cell culture. Petromyzones and elasmobranchs have received litlle more than token attention while Myxini, Holocephali, and Dipnoi apparently have not yet been tried (Table VIII) . Within the teleosts there are representatives of eight orders from which cells or tissues or both have been cultured in uitro, but there are also fishes such as the paddlefish, sturgeons, bowfin, and gar which have not yet been tried. Permanent cell lines have been established from representatives of three orders of fishes, but most of these lines have a fibroblastic morphology (Table VI ). The virologists certainly would welcome additional lines of fish cells with epithelioid morphology. ACKNOWLEDGMENTS We sincerely appreciate the help we have received from the knowledgeable friends and colleagues who reviewed the manuscript for us. They noted errors and omissions, made constructive criticisms, and offered valuable suggestions: Mr. D. F. Amend, Miss Ann Beasley, Dr. J. L. Fryer, Dr. R. G . Malsberger, Dr. I. Pfitzner, Dr. M. M. Sigel, Dr. S. F. Snieszko, and Dr. K. S. Pilcher. Our thanks are due, too, to the editors Dr. W. S. Hoar and Dr. D. J. Randall.

5.

FISH CELL AND TISSUE CULTURE

301

REFERENCES Babini, A., and Ghittino, P, (1961). Coltivazione in &TO in monostrato di tessuti di pesci. (Monolayer fish cell cultures in uitro. Translation by Dr. Carlo Rizzoli.) Atti SOC. Itul. Sci. Vet. 15, 4 5 U 6 2 . Bargen, G., and Wessing, A. ( 1960). Lebendbeobachtungen an in uitro kultivierten Fischzellen. (Observations on living fish cells cultivated in UitTO. Translation by Rikelt, Perrotte and Walker.) 2. Wiss. Mikroskopi 64,356-361. Beasley, A. R. ( 1968). Personal communication. Beasley, A. R., and Sigel, M. M. (1968). Interferon production in cold-blooded vertebrates. In uitro 3, 154-165. Beasley, A. R., Sigel, M. M., and Clem, L. W. (1966). Latent infection in marine fish cell tissue cultures. PTOC.SOC. Exptl. Biol. Med. 121, 1169-1174. Berg, L. S. (1947). “Classification of Fishes Both Recent and Fossil.” Edwards, Ann Arbor, Michigan. Black, V. S. (1957). Excretion and osmoregulation. In “The Physiology of Fishes” (M. E. Brown, ed.), Vol. I, pp. 163-205. Academic Press, New York. Bodian, D. (1956). Simplified method of dispersion of monkey kidney cells with trypsin. Virology 2, 575576. Cameron, G. (1949). Cultivation of tissues from cold-blooded animals. Anut. Record 103,431. Cameron, G. (1950). “Tissue Culture Technique.” Academic Press, New York. Cerini, C. P. (1964). Multiplication and morphology of infectious pancreatic necrosis virus. Doctoral dissertation, Lehigh University. Cerini, C. P., and Malsberger, R. G. (1962). Personal communication. Chlopin, N. G. ( 1925). Studien iiber Gewebeskulturen im artfremden Blutplasma I. Allgemeines. 11. Das Bindegewebe der Wirbeltiere. Z . Mikroskop.-Anut. Forsch. 2, 324365. Chlopin, N. G. ( 1928). Gewebskultur niederer Vertebraten. Arch. Exptl. Zellforsch. Gewebezucht. 6, 97-102. Clem, L. W., Moewus, L., and Sigel, M. M. (1961). Studies with cells from marine fish in tissue culture. PTOC.SOC. Exptl. Biol. Med. 108, 762-766. Clem, L. W., Sigel, M. M., and Friis, R. R. (1965). An orphan virus isolated in marine fish cell tissue culture. Ann. N.Y. Acod. Sci. 126, 34X.374. Dederer, P. H. (1921). The behavior of cells in tissue cultures of Fundulus heteroclitzls with special reference to the ectoderm. Biol. Bull. 41,221-240. Devillers, C. ( 1947 ). Explantations in uitro de blastodermes de Poissons ( Sulmo, Esox). (Explantation in vitro of fish blastoderms. Translation by B. Wolf.) Experientiu 3, 71-72. Dulbecco, R., and Vogt, M. (1954). Plaque formation and isolation of pure lines with poliomyelitis viruses. J. Exptl. Med. 99, 167-182. Fedoroff, S. (1966). Proposed usage of animal tissue culture terms. In uitro 2, 155159.

Fryer, J. L. ( 1964). Methods for the in &TO cultivation of cells from the tissues of salmonid fishes. Doctoral dissertation, Oregon State University. Fryer, J. L. (1968). Personal communication. Fryer, J. L., Yusha, A., and Pilcher, K. S. (1965). The in uitro cultivation of tissue and cells of Pa& salmon and steelhead trout. Ann. N.Y. Acud. Sci. 126, 566586.

302

KEN WOLF AND M. C. QUIMBY

Ghittino, P. (1961). Caratteri morfologici e biologici dei macrofzgi in colture di centri nervosi di pesci. Atti SOC.Ital. Sci. Vet. 15, 462464. Goodrich, H. B. (1924). Cell behavior in tissue cultures. BioZ. Bull. 46, 252-262. Grand, C. G., and Cameron, G. (1948). Tissue culture studies of pigmented melanomas: Fish, mouse, and human. N.Y. Acad. Sci. Spec. Pubt. 4, 171-176. Grand, C. G., Gordon, M., and Cameron, G. (1941). Neoplasm studies VIII. Cell types in tissue culture of b h melanotic tumors compared with mammalian melanomas. Cancer Res. 1, 660-666. Gravell, M., and Malsberger, R. G. (1965). A permanent cell line from the fathead minnow (Pimephales promelas). Ann. N.Y. Acad. Sci. 126, 555565. Greenberg, S. S., Kopac, M. J., and Gordon, M. (1956). Cytology and cytochemistry of melanoma cells. Ann. N.Y. Acad. Sci. 67, 57-121. Greene, A. E., Goldner, H., and Coriell, L. L. (1966). The species identification of poikilothermic tissue culture cells by the cytotoxic antibody test. Growth 30, 30-13. Griitzner, L. (1956a). Versuche zur Ziichtung des gewebes von Macropodus operculuris ( LinnB) und Lebistes reticulatus (Peters) in vitro. (Experiments on the culture of the tissues of Macropodus opercularis (LinnB) and Lebistes reticulatus (Peters) in uitro. Translation by R. L. Herman.) Zentr. Bakteriol., Parasitenk., Abt. I. Orig. 185, 8-24. Griitzner, L. ( 195613) . Wberpriifung einiger Amwendungsmoglichkeiten der Gewebekultur von Lebistes reticulatus (Peters) und Macropodus opercularis ( LinnB) in der Virusforschung. Zentr. Bakteriol., Parasitenk., Abt. I. Orig. 165, 81-96. Griitzner, L. (1958). In vitro-Ziichtung des Leber- und Nierengewebes von Tinca vulgaris Cuv. ( Schleie ) in trypsinierten Einschichtgewebekultren. Zentr. Bakteriol., Parasitenk., Abt. I . Orig. 173, 195-202. Hu, F., and Chavin, W. (1960). Hormonal stimulation of melanogenesis in tissue culture. 1. Invest. Dermatol. 34, 377391. Jensen, M. H. (1963). Preparation of fish tissue cultures for virus research, Bull. W c e Intern. Epizooties 59, 131-134. Kim, K., and Tchen, T. T. (1963). Studies on ACTH-induced melanocyte formation in cultures of goldfish caudal fin. Ann. N.Y. Acad. Sci. 100, 708-717. Kunst, L. ( 1961 ). Kultura bubreinog tkiva iarana. (Culture of kidney tissue of carp. Translation by Dr. N. Fijan.) Vet. Arhiv 31, 241-245. Kunst, L. (1962). Kultura bubreinih stanica Barana. (Culture of carp kidney cells. Translation by Dr. N. Fijan.) Vet. Arhiv 32, 121-126. Kunst, L., and Fijan, N. (1966). Preparation of primary monolayer cell cultures of carp ovary. Vet. Arhiv 36, 235-238. Levan, A., Nichols, W. W., Peluse, M., and Coriell, L. L. (1966). The stemline chromosomes of three cell lines representing different vertebrate classes. Chromosoma 18, 34-58. Lewis, M. R. (1916). Sea water as a medium for tissue culture. Anat. Record 10, 287-299. Lewis, M. R., and MacNeal, P. S . (1935). A study of the pituitary gland of certain fishes by means of tissue cultures. Bull. Mt. Desert Isl. Biol. Lab. 37, 14-16. Li, M. F., and Stewart, J. E. (1965). A quantitative study of the effects of naturally occurring supplements an the growth of rainbow trout ( S a l m gairdneri) gonadal cells. Can. J . Microbiol. 11, 9-14. Lockwood, A. P. M. ( 1961 ), “Ringer” solutions and some notes on the physiological basis of their ionic composition. Comp. Biochem. Phgsiol. 2, 241-289. Loud, A. V., and Mishima, Y. ( 1963). The induction of melanization in goldfish scales

5.

FISH CELL AND TISSUE CULTURE

303

with ACTH, in uitro. J. Cell Bid. 18, 181-194. McFalls, F. D., Markee, J. E., and Wilbanks, G. D. (1967). The culture of adult and embryonic fish tissues. In uitro 3, 196. McLimans, W. F. ( 1961 ). Personal communication. Malsberger, R. C., and Cerini, C. P. (1963). Characteristics of infectious pancreatic necrosis virus. J. Bacteriol. 86, 1283-1287. Malsberger, R. G., and Cerini, C. P. (1965). Multiplication of infectious pancreatic necrosis virus. Ann. N.Y. Acad. Sci. 126, 320-327. Merchant, D. J., Kahn, R. H., and Murphy, W. H., Jr. (1964). “Handbook of Cell and Organ Culture.’’ 2nd ed. Burgess, Minneapolis, Minnesota, Moewus, L., and Sigel, M. M. (1963). In uiuo and in uitro cultivation of a freshwater fish virus in marine host systems. Bacteriol. Proc. V . %, 135. Murray, M. R., and Kopech, G. (1953). “A Bibliography of the Research in Tissue Culture 1884 to 1950.” Academic Press, New York. Murray, M. R., and Kopech, G. (1965). “Current Tissue Culture Literature.” October House, New York. Officer, J. E. (1964). Ability of a fish cell line to support the growth of mammalian viruses. Proc. SOC. Exptl. Biol. Med. 116, 190-194. Oppenheimer, J. M. (1935). The development of isolated blastoderms of Fundulus heteroclitus. J. Exptl. Zool. 72, 247-269. Ortiz-Muniz, G., and Sigel, M. M. (1968). In uitro synthesis of anti-BSA antibodies in fish lymphoid organs. Bacteriol. PTOC.p. 66. OsadEaja, E. F. (1964). Vyldelenie tsitopatogennykh agentov ot karpov pri ostroi forme krasnukhi. (Isolation of cytopathic agents from carp with acute form of rubella. Translation by Dr. N. Fijan.) Veterinariya 41, 29. Osowski, H. ( 1414). Uber aktive Zellbewegung im explantat von Wirbeltierembryonen. Arch. Entwicklungsmech. Organ. 38, 547-583. Parker, R. C. ( 1961). “Methods of Tissue Culture.” Harper (Hoeber), New York. Paul, J. ( 1965). “Cell and Tissue Culture.” Livingstone, Edinburgh and London. Pereira, R. S., and Sawaya, P. (1957). Contribution 21 I’ktude de la composition chimique du sang d e certains sklaciens du brbsil. Boll. Zool. 21, 85-92. Pfeiffer, H. ( 1935). Versuche iiber die Deformation Kleiner Lymphozyten und Retikulumzellen aus der in uitro kultiverten Milz von Petromyzon fluviatilis. Arch. Exptl. Zellforsch. Gewebezucht. 17, 45-62. Pfitzner, I. ( 1964). Fischgewebekulturen und ihre Anwendungsmogiichkeiten bei der Erforschung von Fischviruserkrankungen. Verhandl. Deut. Ges. Zool. Kiel 4, 387493. Pfitzner, I. (1965). Cell and tissue culture of freshwater fish in virus research. Ann. N.Y. Acad. Sci. 126, 547-554. Pfitzner, I. ( 1966). Beitrag zur Xtiologie den “Haemorrhagischen Verusseptikaemie den Regenbogenforellen.” Zentr. Bakteriol., Parasitenk., Abt. I . Orig. 201, 30% 319. Pfitzner, I. ( 1968). Personal communication. Pfitzner, I., and Froehlich, B. ( 1%6). Trypsmierte Einschichtgewebekulturen von Siiswasser-teleosteer-gonadenzellen.Zentr. Bakteriol., Parasitenk., Abt. I . Orig. 201, 186199. Pfitmex, I., and Griitzner, L, (1964). In oitro-Ziichtnng des Gonaden- und Schwimmblasengewebes von Tinca vulgaris Cuv. ( Schleie ) in trypsinierten Einschichtgewebekulturen. Zentr. Bakteriol., Parasitenk., Abt. I . Orig. 191, 474485. Phillips, A. M., Jr., Podoliak, H. A., Brockway, D. R., and Balzer, G. C., Jr. (1957).

304

KEN WOLF AND M. C . QUIMBY

The nutrition of trout. Cortland Hatchery Report 25 for the year 1956. Fisheries Res. Bull. 20. State of New York Conservation Department, Albany. Phillips, A. M., Jr., Podoliak, H. A., Brockway, D. R., and Vaughn, R. R. (1958). The nutrition of trout. Cortland Hatchery Report 26 for the year 1957. Fisheries Res. Bull. 21. State of New York Conservation Department, Albany. Pilcher, K. S., Kleeman, K. T., and Fryer, J. L. (1968). Growth and glycolysis in two cell lines derived from embryos of salmonid fish. Proc. SOC. Exptl. Biol. Med. 127, 659-664. Plumb, J. A., and Wolf, K. (1969). Temperature and seeding density: Effects on growth rates of RTG-2 cells. In uitro 4, Abstr. 2, 125-126. Rachlin, J. W., Perlmutter, A., and Seeley, R. J. (1967). Monolayer culture of gonadal tissue of the zebra danio Brachydanio rerio. Progressive Fish Culturist 29,232-234. Roberts, F. L. (1964). A chromosome study of twenty species of centrarchidae. J. Morphol. 115, 401-417. Roberts, F. L. (1966). Cell culture of fibroblasts from Clupea harengus gonads. Nature 212, 1592-1593. Roberts, F. L. ( 1967). Chromosome cytology of the osteichthyes. Progressive Fish Culturist 29, 75-83. Roberts, F. L. ( 1968). Personal communication. Sanders, M., and Soret, M. G. (1954). Cultivation of animal viruses in embryonic teleost cells. Trans. N.Y. Acad. Sci. [2] 17, 19-25. Schlumberger, H. G. ( 1949). Cutaneous leiomyoma of goldfish. I. Morphology and growth in tissue culture. Am. J. Pathol. 25, 287-299. Soret, M. G., and Sanders, M. (1954). In uitro method for cultivating eastern equine encephalomyelitis virus in teleost embryos. Proc. SOC. Exptl. Biol. Med. 87, 526-529. Stephenson, E. M. (1967). Effects of temperature on tadpole hearts in uitro. 1. Embryol. Exptl. Morphol. 17, 147-159. Stephenson, E. M., and Potter, I. C. (1967). Temperature and other environmental effects on ammocoete organs in culture. J. Embryol. Exptl. Morphol. 17, 4 4 1 4 5 2 . Tchen, T. T., Ammeraal, R. N., Kim, K., Wilson, C. M., Chavin, W., and Hu, F. ( 1963). Studies on the hormone-induced differentiation of melanoblasts into melanocytes in explants from xanthic goldfish tailfin. Natl. Cancer Inst. Monograph 13, 67-80. Tec, V. I., and Jakovleva, G. S. (1962). Polucenie kultury thanej karpa i ee primenenie pri izuEenii etiologii krasnuhi ryb. (Obtaining of carp tissue culture and its application in studying of etiology of rubella of fish. Translation by Dr. N. Fijan. ) Nauchn.-Tekhn. Bju2Z. GosNlORH 15, 73-77. TomaHec, I., Brudnjak, Z., Fijan, N., and Kunst, L. (1964). Weiterer Beitrag zur Aetiologie der infektiosen Bauchwassersucht des Karpfens. Jugoslau. Akad. Znanosti i Umjetnosti, Zagreb, Mat.-Prirodosbuni Ruzred 16, 3-4. Townsley, P. M., Wight, H. G., and Scott, M. A. (1963). Marine fish tissue culture. J. Fisheries Res. Board Can. 20, 679-684. Urist, M. R., and Van de Putte, K. A. (1967). Comparative biochemistry of the blood of fishes: Identification of fishes by the chemical composition of serum. In, “Sharks, Skates and Rays” (P. W. Gilbert, R. F. Mathewson, and D. P. Rall, eds. ) ? pp. 271-285. Johns Hopkins Press, Baltimore, Maryland. Watson, M. S. (1966). “Bibliography on Tissue Cultures: Invertebrates and Cold-

5.

FISH CELL AND TISSUE CULTURE

305

blooded Chordates,” Misc. Publ. 10. Dept. of the Army, Ft. Detrick, Frederick, Maryland. Watson, M. S., and Gilford, J. H. ( 1967). “Bibliography on Tissue Cultures: Invertebrates and Cold-blooded Chordates.” Misc. Publ. 10, Suppl. I. Dept. of the Army, Ft. Detrick, Frederick, Maryland. Wellborn, T. L., and Wolf, K. (1961). Heat fixation in preparing cells for enumeration. Excerpta Med. 15, No. 7, Section 1, 589. White, P. R. ( 1963). “The Cultivation of Animal and Plant Cells.” Ronald Press, New York. Willmer, E. N., ed. (1965). “Cells and Tissues in Culture,” Vols. 1 3 . Academic Press, New York. Wingfield, W. H. (1968). Characterization of the Oregon sockeye salmon virus. Doctoral dissertation, Oregon State University. Wolf, K. (1956). The cause and control of blue-sac disease. Doctoral dissertation, Utah State University. Wolf, K. ( 1963). Physiological salines for fresh-water teleosts. Progressiue Fish Culturist 25, 135-140. Wolf, K., and Dunbar, C. E. (1957). Cultivation of adult teleost tissues in uitro. Proc. SOC. Exptl. Biol. Med. 95, 455-458. Wolf, K., and Quimby, M. C. (1962). Established eurythermic line of fish cells in uitro. Science 135, 1065-1066. Wolf, K., and Quimby, M. C. (1964). Amphibian cell culture: Permanent cell line from the bullfrog (Rana catesbeiana). Science 144, 1578-1580. Wolf, K., and Quimby, M. C. (1967). Low-temperature incubation using a water supply. Appl. Microbiol. 15, 1501. Wolf, K., and Quimby, M. C. (1968). Unpublished research. Wolf, K., and Quimby, M. C. (1969). Progress report on in uitro culture of cyclostome and elasmobranch cells and tissues. In uitro 4, Abstr. 2, 125. Wolf, K., Quimby, M. C., Pyle, E. A., and Dexter, R. P. (1960). Preparation of monolayer cell cultures from tissues of some lower vertebrates. Science 132, 1890-1891. Young, J. 2. (1933). The preparation of isotonic solutions for use in experiments with fish. Publ. Staz. Zool. Napoli 12, 425-431.

This Page Intentionally Left Blank

CHROMATOPHORES AND PIGMENTS* RYOZO FUJZZ

. .

. . . . . . . . . . . .

. . . . . . . . . . . .

I. Introduction . . 11. Classification and Terminology of Chromatophores . 111. Morphology of Chromatophores . . IV. Chromatophore Pigments . . A. Melanins . . . . . B. Carotenoids . . . , C. Pteridines . . . . D. Purines . . . V. Physiological Color Changes . . . . . A. Methods of Measuring Chromatophore Responses B. Mechanism of Pigment Movements . . C. Hormonal Control of Chromatophores D. Nervous Control of Chromatophores . . E. Effects of Chemicals and Drugs on Chromatophores VI. Morphological Color Changes . . . . . VII. Other Topics Acknowledgments . . . References . . . .

. . . . . . . . .

. . . . . . . . .

. . .

. .

. . . .

.

.

. . . . . . . .

. . .

. . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .

. .

*

.

3#7 308 309 313 313 314 315 316 317 317 319 322 326 332 338 341 344 344

I. INTRODUCTION

That fishes change their hues in response to background coloration has long been known. They also display color responses during excitement and courtship. These spectacular changes are mediated through the activities of integumentary pigment-containing cells called “chromatophores.” Twenty years ago Parker (1948) devoted many pages to fish classes in his extensive book on this subject. Also more than 10 years have passed since the two articles in the treatise “The Physiology of Fishes” appeared, one by Odiorne (1957) on chromatophores and the

* This chapter was written during the tenure of a research associateship under the support of NSF grant GB-4956X to Dr. Ronald R. Novales as principal investigator. 307

308

RYOZO FUJII

other by D. L. Fox (1957) on pigments. Meanwhile, several monographs and reviews have been published on animal color changes; some of them are comprehensive ( e.g., Waring, 1963; Fingerman, 1965), and others are rather narrowly directed to some special interest such as the hormonal control of pigmentation ( e.g., Pickford and Atz, 1957; Novales, 1967). In this chapter, therefore, the author attempted to select and organize some of the recent important findings in the field of pigment cell physiology of fishes with the aim of making the article useful to students interested in this particular area of investigation. Owing to the limitation of size, a number of publications of importance had to be left out. These can be located in recent relevant reviews and other articles cited here.

11. CLASSIFICATION AND TERMINOLOGY OF CHROMATOPHORES

According to the color of pigments they contain, the chromatophores of fishes have been commonly classified as melanophores (brown or black), erythrophores (red), xanthophores (yellow), leucophores (white), and iridophores ( reflecting). Both the leucophore and iridophore contain colorless pigments (primarily guanine), but the former contains small crystals which can move back and forth in the cytoplasm, whereas in the latter there are larger crystals incapable of movement and usually stacked in layers. Sometimes more than one type of pigment can be found in the same chromatophore. For instance, Goodrich et al. (1941) described the xantho-erythrophore of the swordtail, showing that the center of the cell contains yellow carotenoid pigments, whereas in the cellular processes there is an accumulation of the red pteridine. Frequently, the pigment-containing bodies are not unicellular chromatophores but combinations of such cells. These have been referred to as “chromatosomes” by Parker ( 1948). The most commonly found chromatosome is probably the melaniridosome, a complex of melanophore and iridophore. In view of the fact that the suffix “-some” has more commonly been used to denote cytoplasmic inclusions such as ribosomes, lysosomes, or melanosomes (see below), this kind of term may be replaced with a more proper expression, e.g., “compound chromatophore.” As far as the melanin-containing cells of vertebrates are concerned, a recommendable terminology has recently been proposed ( Fitzpatrick et al., 1966). In this terminology, the melanocyte is defined as a cell which synthesizes a special melanin-containing organelle, the melanosome. The new definition of the melanophore, with which we are much concerned here, is essentially the same as that familiar to most zoologists, i.e., a type

6.

CHROMATOPHORES AND PIGMENTS

309

of melanocyte which participates with other chromatophores in the rapid color changes of animals by intracellular displacement of melanosomes. In addition, the melanoblast is a cell which serves at all stages of the life cycle as the precursor of the melanocyte and melanophore. The terms “melanosome” and “premelanosome” are more or less unfamiliar to students of zoology. In most cases, the commonly used term “melanin granule” may be replaced with “melanosome,” which is defined as a discrete melanin-containing organelle in which melanization is complete, more or less uniformly “electron dense” by electron microscopy; with tyrosinase activity usually not demonstrable. “Premelanosome” means all distinctive particulate stages in the maturation of melanosomes. Their electron density is variable, and they possess an active tyrosinase system after the onset of melanin synthesis. For describing other kinds of chromatophores than the melanophore, a number of terms have been employed (cf. Parker, 1948; Bagnara, 1968). However, most of the less commonly used words can be grouped into a few generally accepted terms without causing much inconvenience. As for the cellular inclusions carrying pigment, no definite usage of terms has been adapted. This may be largely because of the lack of definitive information about the chemical nature of pigments inside these cells. In recent years, attempts to identify these pigments chemically in various chromatophores have been in progress (D. L. Fox, 1957; Hama, 1963). These studies will surely result in a finalization of the nomenclature of these intracellular granules in near future. For instance, Matsumoto (1965a) has recently designated the pigmented particle in the erythrophore of the swordtail as the “pterinosome,” since it primarily contains pteridines.

111. MORPHOLOGY OF CHROMATOPHORES

Early morphological literature on fish chromatophores by means of light microscopy can be referred to in the work by Franz (1940). In 1957, Falk and Rhodin presented their electron microscope observations of the Lebistes Teticulatus melanophore. They held that there were two cell membranes, an outer limiting membrane and an inner cytoplasmic membrane containing pigment granules and the other usual cell organelles. The space between the two was occupied by a meshwork of fibrils which they supposed to be contractile. On the basis of the electron microscopy of the melanophores of the goby, Chasmichthys gulosus, Fujii (1966a) claimed that they do not have the double-membrane structure

310

RYOZO FUJI1

described by Falk and Rhodin (1957), but rather have only the one thin cell membrane usually present in cells. Bikle et al. (1966) also showed that the melanophores of Fundulus heteroclitus are enclosed by a single cell membrane. Fujii (1966b) then showed that the Lebistes reticulatus melanophore does indeed have only one cell membrane rather than the two claimed by Falk and Rhodin (1957). The cell organelles in the melanophores described by these researchers include a nucleus ( nuclei), melanosomes, centrioles, mitochondria, ribosomes, vesicular smooth-surfaced endoplasmic reticulum, and micropinocytotic vesicles. Besides these, Bikle et at. (1966) found in Fundulus heteroclitus melanophores a number of microtubules about 225 A in diameter aligned parallel to the long axes of the branches of the cell, Novales and Novales (1966a) and Green ( 1968) confirmed this finding. Although microtubules have not yet been described in species other than Fundulus, they probably exist more widely among fishes. For example, unidentified small vesicles about 200 A in diameter in the Chasrnichthys melanophore (Fujii, 1966a) might be cross sections of microtubules. Quite recently, an ultrastructural study of the melanophores of the Atlantic hagfish, Myxine glutinosa, appeared (Holmberg, 1968). These melanophores have a fundamentally similar structure to those of teleost species but contain cytoplasmic tubules 400-700 A in diameter. These elements are apparently different from microtubules and seem to project inward from the cell membrane, being probably elongated micropinocytotic vesicles. The melanosomes are spherical or ellipsoidal bodies with an average diameter of about 0.5 p. Each melanosome is surrounded by a limiting membrane. The enclosed melanin is very dense both to visible light and the electron beam (Bikle et al., 1966; Fujii, 1966a,b). In those which appear to be in the early stage of melanosome development, small vesicles and dense particles are observed inside the limiting membrane (Bikle et al., 1966). Matsumoto (1965a) studied the fine structure of erythrophores of the swordtail, Xiphophorus helbri. The red pigment granules consist of an outer limiting membrane and inner lamellae. Since pteridines are found primarily in these granules, they were designated as “pterinosomes.” Other cellular organelles found in the erythrophore include mitochondria, tubular smooth-surfaced endoplasmic reticulum, ribosomes, and pinocytotic vesicles similar to those found on the melanophores. In addition to the mitochondria and sparse endoplasmic reticulum, the cytoplasm of the goldfish xanthophores contains two characteristic components not found in melanophores: so-called large bodies and small vesicles (Loud and Mishima, 19e3). The former are spherical and 0.40.5 p in diameter, containing either amorphous material or unorganized

6.

CHROMATOPHORES AND PIGMENTS

311

membranous structures. The small vesicles are empty in electron micrographs, and have a very dense membrane, with a diameter between 400 and 500 A. It was concluded that the large bodies are the structural precursors of the melanosomes, the premelanosomes, since these granules can be stained with Masson’s ammoniated silver nitrate method for premelanin, and that the smaller vesicles may be related to the yellow pigment substance, The Loud and Mishima’s premelanosomes in the goldfish xanthophore are very similar to the pterinosomes in the erythrophore of the swordtail ( Matsumoto, 1965a)- Therefore, the possibility that these premelanosomes are also the site of pteridine deposition should be considered. The small vesicles in the goldfish xanthophores might be the storage packets of carotenoid pigments. In any case, pigment granules found in either erythrophores or xanthophores seem to be variable in size, form, and internal structure. In no case does the plasma membrane of these chromatophores make close contact with adjacent cells; desmosomes and tight junctions have not been found. Although there are some presynatic neural elements around the plasma membrane (Section V, D ) , the chromatophores are mainly surrounded by an extracellular matrix, usually containing collagenous fibrils. Iridescent blue spots on the skin of the Japanese porgy, Chrysophrys muior, are composed of an assemblage of small flat iridophores (Kawaguti and Kamishima, 1966a). It was found that such iridophores are filled with several flat vesicles, possibly cisternae of smooth-surfaced endoplasmic reticulum, in each of which a thin reflecting platelet is embedded. These platelets are very large and about 700-1000 A in thickness. These guanine-loaded vesicles are arranged parallel to each other within the cytoplasm of a cell. Iridophores responsible for the dorsal greenish blue coloration of the clupeoid fish, Harengula zunasi, are somewhat larger cells than those of the porgy (Kawaguti and Kamishima, 1966b). Vesicles containing very thin guanine platelets ( ca. 60 A in thickness) are seen in groups, in each of which they are arranged parallel to each other. The distance between these vesicles is about 1700 A. In the dermis, xanthophores overlie the iridophore layer, below which is a layer of melanophores. Such a special layered construction may be responsible for the beautiful and delicate tone of the fish. Kawaguti (1965) also described the fine structure of the iridophores of the blue male wrasse, Haliochoeres poecilopterus. Small mitochondria and vesicular bodies are found around the nucleus or along the margin of the cell. The rest of the cytoplasm is occupied with lamellar structures. They are found in small groups, in each of which five or ten such bodies run parallel to one another. The thickness of a lamellar body is about 1000 A. Neither guanine granules nor platelets was observed. Since no pigment

312

RYOZO FUJI1

has been detected even by spectrophotometric studies of the intact scale skin, Kawaguti concluded that these lamellar structures may be primarily responsible for the bluish iridescent coloring of the wrasse. The sides and ventral surfaces of fishes are often silvery and white. These surfaces are actually very good reflectors, and according to Denton and Nicol (1966) the ventral skin of some species reflects over 80% of the light striking it. In some case the integument has different reflecting properties when it is viewed from different directions, e.g., the upper surface of a horse mackerel Trachurus shows such diEerence well, appearing dark when viewed from above and silvery blue when viewed from the side and below (Denton and Nicol, 1966). Histological studies have shown that these various effects are achieved by oriented reflecting crystals which lie in the iridophores of the skin at different angles with respect to the surface of the fish, Denton and Nicol (1966) also made a comparative survey of reflecting layers in the integument of various silvery teleosts. These fishes have a layer of oriented reflecting platelets lying outside the stratum argenteum. By histological means it was found that these platelets lie more or less vertical to the surface of the water. In the water, the reflecting layers diminish the visibility of the fish from most fields of view because they reflect light approximately equal to the background light against which the fish is seen. Working with the histology of the scales of the juvenile sprat, Clupea spruttus, Denton and Land (1967) showed that the iridophore cytoplasm consists of a stack of flat guanine platelets, each having an optical thickness close to a quarter of a wavelength and separated from its neighbors by layers of cytoplasm of equal optical thickness. Thus, constructive interference occurs effectively, and very high reflectivity is achieved. No published work has appeared on the fine structure of fish leucophores. But the author’s unpublished observations (1965a) of the caudal leucophores of the goby Chasmichthys indicate that guanine particles are granular and not so flat as those of iridophores as described by Kawaguti and Kamishima ( 1966a,b). These granules are rather sparsely distributed in such a cell, being consistent with their mobility through the cytoplasm. Physiological color changes are mostly caused by the activities of dermal chromatophores. These cells are mainly found in the uppermost part of the dermis. Recent electron microscope studies have revealed that between the basal border of the epidermis and the layer of pigment cells there is a layer of rather uniform thickness consisting mainly of collagenous fibrils (Fujii, 1966a,b). In Chasmichthys (Fujii, 1966a, 1968), Fundulus (Bikle et al., 1966), and Carassius (Fujii, 1965b), the collagenous fibrils irl such a space are arranged in lamellae with those in alternat-

6.

CHROMATOPHORES AND PIGMENTS

313

ing layers at right angles to those in the intervening layers. Fujii (1966a) has designated such an orderly fabric as the “subepidermal collagenous lamella.” He further thinks that its remarkable degree of architectural order may be the structural demand for increasing transparency of the tissue which overlies the chromatophores; the more tissue transparency increases, the more effective is the manifestation of the activity of underlying pigment cells. In smaller fishes such as Lebistes the collagenous fibrils in such a layer are not orderly arranged (Fujii, 1966b). Perhaps the thinness of the epidermal and collagenous layers may not require such an elaborate architectural regularity.

IV. CHROMATOPHORE PIGMENTS

Two outstanding monographs have been published dealing with animal pigments (D. L. Fox, 1953; H. M. Fox and Vevers, 1960). Although the earlier article in the present treatise remains useful (D. L. Fox, 1957), some recent findings will be mentioned here. Pigments other than those found in integumentary pigment cells are dealt with in other chapters. A. Melanins The brown or black pigments widely found in fish species are melanins. They are highly polymerized compounds derived from tyrosine. The synthesis of melanins involves first the oxidation of tyrosine to 3,4dihydroxyphenylalanine (Dopa) and then to Dopa quinone. The enzyme concerned is tyrosinase. Dopa quinone is then polymerized to form melanins, which are usually found attached to a protein. Melanin synthesis takes place in melanocytes, young melanophores, and possibly even in mature melanophores. The subcellular site of synthesis is known as a premelanosome, the early stage of melanosome development. Usually melanosomes are only to be expected in adult melanophores wherein malanization has been completed. However, melanogenesis can take place in mature melanophores because morphological color changes may involve an increase in melanin content without a corresponding increase in number of melanin-containing cells (Section VI). Indeed, Bikle et al. ( 1966) found that among melanosomes presumably immature melanin granules, premelanosomes, are present which contain a number of vesicular structures and dense particles inside the limiting membrane.

314

RYOZO FUJII

Recently, there have been some quantitative studies on fish melanogenesis. Thus, Kosto et al. (1959) incubated isolated Fundulus fin samples with Dopa and obtained a significant correlation between 0, consumption and the degree of darkening. Recently, a sensitive radiometric tyrosinase assay has been developed by Chen and Chavin (1965). Working with skin of the black moor goldfish, Chen and Chavin (1966) made a contribution by modifying the classic metabolic pathway of melanogenesis: The incorporation of tyrosine and Dopa into melanin without decarboxylation was demonstrated by using carboxyl-labeled compounds. Chen and Chavin (1967) have further shown that in general the dorsaI integument of fishes contains higher tyrosinase levels than the ventral. In four varieties of goldfish (white, xanthic, gray, and black) the total tyrosinase activity increases with increasing pigmentation, and the activity increase occurs mostly in the particulate fraction. This observation is rather inconsistent with the finding by Tomita and Hishida (1961a) that tyrosinase and Dopa-oxidase activities in adult Oryzias skins are higher in light-colored varieties than in the brown variety. Hishida et al. (1961) showed that the light-colored varieties of Oryzias have amelanotic melanophores and the application of sulfhydryl poisons or some other agents can bring about their melanization. Thus, they concluded that tyrosinase exists in an inhibited state in the arqelanotic melanophore. In recent years, it was found in goldfish skin (Kim et al., 1962), Fundulus embryos (Spitz and Burnett, 1968), and in a variety of other fishes including lamprey, shark, and eel (Chen and Chavin, 1967), that the molecules of the enzyme tyrosinase do not have the same physical and chemical characteristics. Thus, there are multiple forms of the active enzyme. In a few species (the hagfish Eptatretus, the bichir Polypterus, the garpike Lepisosteus, and the African lungfish Protopterus) tyrosinase was found only in the particulate fraction (Chen and Chavin, 1967).

B. Carotenoids Carotenoids which occur in fishes were especially well documented in the article by D. L. Fox (1957). Although some exceptions were known (cf. Section IV, C), it had long been thought that all yellow and red pigments of xanthophores and erythrophores were carotenoids. These are highly unsaturated hydrocarbon compounds and consist of a chain of carbon atoms with a ring structure at one or' both ends. In addition to carbon and hydrogen, xanthophylls or their esters involve oxygen. In many species large amounts of xanthophylls including taraxanthin, lutein, and astaxanthin have been detected, while carotenes, mostly pcarotene, were less frequently found in the integument ( D. L. Fox, 1957).

6.

CHROMATOPHORES AND PIGMENTS

315

All carotenoids are insoluble in water but soluble in organic solvents. Hence, the term “lipophore” has been rather widely used to denote both xanthophores and erythrophores. It is known that animals do not have the ability to synthesize carotenoids. Therefore, those found in fishes must have come either directly from plant sources or from their prey. The incorporated pigments are transferred to xanthophores or erythrophores, where a great amount can be stored. Some species contain a variety of carotenoids, whereas others contain only specific ones (D. L. Fox, 1957). The latter fact can be explained by their ability to convert ingested carotenoids into the type characteristic of the species. For instance, Sumner and Fox (1935) demonstrated that Fundulus parvipinnis is able to convert dietary carotenoids into an integumentary xanthophyll. More information was lately added about the transfer of carotenoids by Takeuchi (1960), who suggested that carotenoids in larval Oryzias xanthophores may have come from the pigments found in yolk. It is also worth noting that during early stage of differentiation, xanthophores do not contain carotenoids, while in later life these pigments begin to accumulate in the cell (Matsumoto, 196513; Hama and Hasegawa, 1967; cf. also Section IV, C). Hama and Hasegawa (1967) found that xanthophores of blue and white color varieties of Oryzias do not contain carotenoids and suggested the use of “acarotenoid xanthophores” to denote such cells. The carotenoid-containing cells of the wild or orange-red varieties may therefore be called “carotenoid xanthophores.” The subcellular site of carotenoid deposition in either xanthophores or erythrophores is not clear. Since larger granules about 0.5 p in diameter are claimed to be the storage parcels solely for pteridines (Matsumoto, 1965a), the small vesicles described by Loud and Mishima (1963) might be the possible organelles. Very recently, Bellamy ( 1966) reported that yellow and orange pigments of the chromatophores of the painted comber, Serranus scriba, are xanthophylls, and they are stably associated with the particulate fraction of tissue homogenates. Although their presence has not been observed in electron micrographs as yet, the fat yellow droplets described by Hama and Hasega.wa (1967) in xanthophores might possibly be storage compartments.

C. Pteridines Much attention has recently been directed to the presence of pteridines in the chromatophores of fishes as well as of amphibians (Hama, 1963; Ziegler, 1965). These compounds are soluble in water and closely related to purines and flavins; they have both a pyrimidine and an asso-

316

RYOZO FUJII

ciated pyrazine rings. Both colored and colorless pteridines are present in the chromatophore (cf. Hama, 1963). Drosopterins including drosopterin, isodrosopterin, and neodrosopterin are red, while sepiapterins, comprising sepiapterin and isosepiapterin, are yellowish. The colorless leucopterins are generally divided into two groups, i.e., blue or violet fluorescent. They include biopterin, rana-chrome 3, xanthopterin, and isoxanthopterin . Hama (1963) and Matsumoto (1965a,b) reported an intimate association of pteridines with yellow or red pigmentation, emphasizing that colored pterdines actually serve as functional pigments of both xanthophores and erythrophores. The colored pteridines are usually accompanied by an array of colorless pteridines. Although the role of these colorless pteridines in pigment cells is still unclear, Hama (1963) recently suggested their possible participation in melanin or carotenoid formation in chromatophores. As mentioned previously, pteridines are primarily contained in the pterinosomes (Matsumoto, 1965a; cf. Section 111). Matsumoto (196513) reported that in early larval stages of cyprinid fish the pigments responsible for integumentary color are exclusively pteridines. He found that both late larval and adult xanthophores contain pteridines as well as carotenoids, and that at least in larval stages sepiapterins act as the actual coloring agent in xanthophores. Similar changes of pigment substances during development have also been observed in Oryzias (Hama and Hasegawa, 1967). Their results indicate that the larval colored pterinosomes, so-called “sepiapterinosomes,” may be transformed gradually into noncolored adult-type pterinosomes. Chemically, the amount of the colored pteridines also decreased during this process becoming almost absent in the adult. That is, in adult life, fat and/or carotenoid pigments are mainly responsible for the color of xanthophores. Melanophores also seem to contain a considerable amount of colorless pteridines comprising xanthopterin, isoxanthopterin, and biopterin (Hama, 1963; Matsumoto, 1965b). On the other hand, Matsumoto (196513) reported that the appearance of leucophores during the development of cyprinid fishes is not accompanied by a modification in pteridine fractions.

D. Purines Among the purines responsible for the white or silvery tone of the integument of fishes, guanine definitely predominates. Its abundant ac-

6.

CHROMATOPHORES AND PIGMENTS

317

cumulation is found in special chromatophores, the leucophores or iridophores. In the cytoplasm of these cells, guanine exists in minute crystalline or granular form, or sometimes as thin platelets (Section 111). Thus, these cells have often been referred to as “guanophores.” Actually, Hitchings and Falco (1944)) and Neckel (1954) formerly demonstrated chemically the abundant presence of guanine in the fish integument. At least in iridophores the molecules of purines seem to be orderly arranged within individual platelets (Kawaguti and Kamishima, 1966a). Recently, however, other kinds of purines than guanine have also been reported. For example, Ziegler-Giinder (1956) showed the occurrence of hypoxanthine in the integument of cyprinid fishes. The presence of both guanine and hypoxanthine has been confirmed in the belly skin of presmolts, smolts, and grilse of the coho salmon, Oncorhynchus kisutch (Markert and Vanstone, 1966)) and in the integument of parrs and smolts of the Atlantic salmon, S . sular (Johnston and Eales, 1967). Ota (1954) detected uric acid in the skin of the hairtail, Trichiurus maUmelU.

V. PHYSIOLOGICAL COLOR CHANGES

Physiological color changes are defined as those changes in external coloration produced by modification in the distribution of pigment in the chromatophores. Such changes usually occur more rapidly than those resulting from an increase or decrease in the amount of pigment, the so-called morphological color changes, dealt with in Section VI. A. Methods of Measuring Chromatophore Responses A number of methods for describing the conditions of chromatophores have been employed. The most simple method is to use the words, “expanded,” “intermediate,” and “contracted” or “reticulate,” “stellate,” and “punctate.” Since most chromatophore activities result from the pigment movements in the cell, “dispersed and “concentrated” or “aggregated have also been used. For some qualitative descriptions these simple expressions are still quite useful and convenient. As an elaborate modification of such methods, the “Melanophore Index” as originally used by Hogben and Slome (1931) for staging amphibian melanophores has been conveniently applied to the cells of fish, and sometimes even to chromatophores other than melanophores. In this system, the range from maximal pigment aggregation to maximal dispersion is divided into five stages.

318

RYOZO FUJI1

Stage 1 means maxima1 aggregation, each increasing number denoting increased pigment dispersion, and finally stage 5 corresponds to the fully dispersed state. Although the figures reflect only indirectly the state of melanophores, Healey (1967) proposed quite recently a method called “Derived Ostwald Index.” This method involves matching of the general tint of the dorsal surface of a fish, as seen with the naked eye, against graded color standards. As standards, he chose 9 points of the Ostwald gray series and gave them the arbitrary numbers 0 to 8 in order to permit graphical presentation. In some experiments, direct measurement of changes in the diameter of a chromatophore has been made (Spaeth, 1916a). Measuring the length of a branch of a pigment cell has been more commonly employed (Spaeth, 1916b; Kinosita, 1953). In a special attempt, even the migration velocity of a pigment granule through the cytoplasm of a melanophore process was measured (Kamada and Kinosita, 1944). The photoelectric method was first introduced by Hill et aZ. (1935). By making use of a photocell they measured the changes in reflected light from the dorsal surface of an intact Fundulus. The reflectance method was also conveniently employed for the study of melanophore responses in dogfish skin in uitro (Novales and Novales, 1966b). Smith (1936) described the pulsation of melanophores by recording the light transmitted through an isolated scale of Tautoga. Instead of the scale, Fujii ( 1959a) utilized the “split preparation” from the tail fin of the goby, Chasmichthys gulosus. This preparation is made by splitting an isolated piece of fin consisting of two fin rays and intervening web into two symmetrical halves by means of fine forceps. The changes in the transmitted light through the web can be recorded either by plotting the photoelectric current every moment (Fujii, 1959a; Etoh and Egami, 1963) or continuously by making use of a paper recorder (Fujii and Novales, 1968a, 1969a). In such a preparation, chemicals or drugs are not required to penetrate the highly impermeable epidermis to reach the pigment cell layer. Therefore, the chromatophore responses to chemical stimuli develop very rapidly and proceed almost simultaneously among the cells all over the preparation. Earlier workers using photocells have measured the activities of a number of melanophores at the same time. In some experiments, however, the recording of the response of a single melanophore is advantageous, since analysis of the data can be more easily carried out. Thus, Fujii and Novales (1969a) could examine the nature of the nervous supply to a given Fundulus melanophore (cf. Section V, D). An argument against the photoelectric measurement of melanophore response is that the activities of other kinds of chromatophores might

6.

CHROMATOPHORES AND PIGMENTS

319

interfere with the recording. By using a proper light filter, however, such interference can be mostly eliminated. For instance, Fujii (1961) employed a red filter, which eliminates light shorter than 610 mp in wavelength. Thus, the activities of xanthophores or erythrophores were almost excluded.

B. Mechanism of Pigment Movements By means of a pressur-temperature analysis on the Fundubs melanophore, Marsland (1944) showed that the aggregation of pigment granules is correlated with gelation of the protoplasm containing them, whereas melanin dispersal involves cytoplasmic solation. Very recently, Marsland and Meisner (1967) extended Marslands earlier results by exposing FunduZus melanophores to solutions containing D,O or to higher temperatures, both such agencies being known to foster protoplasmic gelations generally. Based on their electron microscope observations on the Lebistes melanophores { cf. Section 111))Falk and Rhodin (1957) gave a unique explanation for melanin movements within teleost melanophores. The contraction or the relaxation of the fibrils which are present in the space between the inner and outer membranes of the cell was held to cause the aggregation or dispersion of pigment granules in the inner sac. Until recently this view was rather widely accepted. However, Fujii (19Seb), working with the same species, pointed out that the melanophore is enclosed by a single thin membrane as is usually for cells. The thick outer membrane investigated by Falk and Rhodin was found to be the basal lamina underlying the epidermis, while the inner thin sac was found to be the true cell membrane of the melanophore. The presumed contractile filaments were disclosed to be extracellular collagenous fibrils. Studies on other teleost forms have also shown that the melanophores have fundamentally similar fine structures (cf. Section 111). Thus, the mechanism proposed by Falk and Rhodin ( 1957) should now properly be abandoned. Recently, Kinosita ( 1953, 1963), working with the scale melanophores of Oryzias Zatipes, reported that melanosomes are negatively charged, and that under the influence of K' or epinephrine, which causes the aggregation of pigment in the cell, the membrane potential decreases in the central region and at the same time increases on the processes of the melanophore. The reverse changes are recorded following the application of physiological saline or atropine solution in which melanin dispersal takes place. In both the aggregation and dispersion processes the melanosome movements are always preceded by the changes in electric

320

RYOZO FUJII

potentials. On the basis of these findings, Kinosita presented the idea that melanosomes may migrate electrophoretically through the cytoplasm in the direction opposite that of the intracellular current flow as a result of the potential gradient established along the cellular processes. Leucophores of Oryzias, which respond to various stimuli in an opposite manner to melanophores, were also subjected to microelectrode studies by Kinosita (1953). He found that guanine granules are also negatively charged but that the gradient of membrane polarization along the cellular process of the leucophore either in the physiological or in KC1 solution is opposite that found in a melanophore process. These results led him to conclude that his electrophoresis hypothesis may also be applied to other kinds of chromatophores than the melanophore. By means of electron microscopy, Fujii (1966a,b) reported that the melanophores are mainly surrounded by collagenous extracellular space, where the specific resistance may be comparatively low, and that there exist in the cell numerous elements of vesicular endoplasmic reticulum which are supposed to increase the specific resistance of the cytoplasm. These structural features may be advantageous for increasing the possible intracellular potential gradient along the projections of the melanophore, thus being in favor of the electrophoresis theory. The unpubIished results by Fujii and Novales, on the other hand, show that increase in the external K+ concentration causes membrane depolarization of Fundulus melanophores, and that this depolarization results from the high K+ conductance of the melanophore membrane. Similar K+-produced potential changes are seen even when adrenergic receptors in the melanophores have been completely blocked by dibenamine. After dibenamine treatment, the melanin aggregation due to K+ or epinephrine does not occur. High external [K'] also fails to induce melanosome aggregation in denervated melanophores of the goby ( Fujii, 1959a). In the same melanophores, a rapid pigment aggregation is ob. served in response to epinephrine even in high K+ solutions (Fujii and Taguchi, 1969). These observations suggest that melanin movements may not be directly related to the value of the membrane potential or even to changes in membrane polarization. Thus, the electrophoresis mechanism should now be reevaluated. Recently, it was suggested that the electrogenic Na channels in the membrane are probably not involved in the control of melanin movements since tetrodotoxin has no effects on the response of the melanophore itself ( Fujii and Novales, 1968a). Working on the electron microscopy of Fundulus scale melanophores, Bikle et al. ( 19%) recently pointed out that in a melanophore process melanosomes are arranged in files or rows, and that the microtubules are present between these rows. In no case were the tubules found to be directly at-

6.

CHROMATOPHORES AND PIGMENTS

321

tached to the melanosomes. Microtubules are present in the processes regardless of whether the melanosomes are aggregated in the center of the cell or dispersed. These observations led them to suppose that the microtubules function as cytoskeletal elements which heIp maintain the extended form of the melanophore processes and at the same time define the channels in which the melanosomes move. Similar observations were made by Green (1968) who also studied Fundulus. Such descriptions of orderly arrangement of melanosomes are more or less different from those by Kamada and Kinosita (1944) on the scale melanophores of some freshwater fish, who believe that melanosomes easily alter their location from one branch to another during the course of their redispersion. Melanosomes in the cellular process of a melanophore cut either at the proximal part or at the distal end or at both with a microneedle still move back and forth in response to various stimuli (Kamada and Kinosita, 1944). This indicates that the receptors for both melaninaggregating and dispersing stimuli may be distributed along the cellular processes, and that even though the probable importance of microtubules cannot be denied, their connections with centrioles or with far extremities of processes are not required for the pigment movements. High hydrostatic pressure which is known to cause the reversible breakdown of microtubules brings about the melanosome dispersion in Fundulus ( Marsland, 1944). This suggests that microtubules are important at least for the sequence of pigment aggregation. Since microtubules are also known to be broken down by colchicine, Wikswo and Novales (1969) made use of it on the melanophores of Fundulus scale to see if melanin movements are affected, Melanin aggregation resulting from epinephrine was much slowed down, while the dispersion of aggregated melanophores was somewhat accelerated. These results give further evidence for the importance of microtubules. Based on his observation on embryonic Fundulus melanophores, Rebhun (1967) recently claimed that melanosome movements differ from Brownian movements, being characterized by discontinuous jumping of the granules. To explain this “saltatory” movement, he assumed two types of fiber systems inside the cell, i.e., microtubules and 50 A microfilaments. Actually, the former have recently been described by Bikle et al. (1966), Novales and Novales (1966a), and then by Green (1968). They are known not to be force-generating elements, while his hypothetical thin filaments are structurally similar to smooth muscle filaments and should be contractile. Rebhun supposes that these filaments might be closely associated with or buried in the microtubule wall. Although maintaining that the microtubules may function as cytoskeletal elements, Green (1968) emphasized the active role in pigment motillity of the elements

322

RYOZO FUJII

of the smooth-surfaced endoplasmic reticulum which are abundantly found around melanosomes. Among the substances which affect pigment movements in chromatophores, melanocyte-stimulating hormone ( MSH ) has been most precisely studied on its mode of action, although most of the work has been conducted with amphibian cells (cf. Novales, 1967). According to Novales ( 1959), melanin dispersion by MSH requires the presence of Na ions in the medium. As the most likely explanation for this he presented the view that MSH may increase the permeability of the melanophore membrane to Na'.. Melanocyte-stimulating hormone might also interfere with the Na pump of the melanophore. Either effect would result in the net accumulation of Na ions in the cell. Although no direct evidence has been given for Na+ entry through the melanophore membrane, accumulation might result in the water entry at the same time, which is known to produce melanin dispersal (Novales, 1959). The increase in Na+ level in the cell might also cause the solation of the cytoplasm. Quite recently, Novales and Novales (1966b) tried to extend the Na requirement theory to elasmobranch melanophores. Although their results on isolated Squalus acanthias skin indicate that during MSH action water entry occurs, and that the water entry may cause melanin dispersion, MSH was still effective in the absence of Na+. Na+ could be replaced by other ions, e.g., Li+, K+, choline, and Mg". On the other hand, a short report by Novales and Gratzer (19sO) showed that Na+ is necessary for the melanin-dispersing action of MSH in isolated scale melanophores of the goldfish. The importance of the intracellular Ca'+ level in pigment migration was suggested by Ishibashi (1957), who injected Na oxalate, a Ca precipitant, into an 0y z i a s melanophore and observed the aggregation of melanosomes followed by their rhythmic centrifugal and centripetal movements (pulsation). Hu (1983) claimed that goldfish melanocytes in tissue culture respond by shape changes. The mechanism of cellular activities of melanocytes should thus be separately considered from that of the mature melanophores. Naturally, the presence of a system of contractile elements may be supposed in the ectoplasm of these cells. C. Hormonal Control of Chromatophores The coordinating systems for color changes existing among fishes show great diversity. In some fish, blood-borne hormones are believed to be predominantly responsible for the pigment movements while in

6.

CHROMATOPHORES AND PIGMENTS

323

others, the pigment cells are solely regulated by nerves. Between these extremities, there are many examples where both humoral and neural control mechanisms are actually working. The latter kind of regulation will be discussed later in this section. A chapter in the outstanding volume by Pickford and Atz (1957) is a complete review of the works up to that time about the hormonal control of chromatophores of fishes. In cyclostomes, Young (1935) indicated that the melanophores of Lampetra planeri are aneuronic, suggesting pituitary control of the pigment cells. He aIso found that pinealectomy results in a permanent darkening of the integumentary tone. Although he concluded that the pineal may take part in inhibitory control of pituitary, the possible role of melatonin or related substances as a hormone from the pineal complex antagonizing the pituitary hormone should be examined. Lanzing ( 1954) recently observed that pituitary extracts of Lampetra darken the frog skin, and suggested the actual role of a pituitary principle which resembles MSH. Observations on the pigment cells of the hagfish, Myxine glutinosu, also seem to be in favor of the hormonal control of cyclostome cells (Coonfield, 1940). On the other hand, Holmberg (1968) recently reported that no background responses were detected in the same species of the hagfish. It has been generally accepted that the melanosome dispersion in elasmobranch melanophores is controlled by MSH released from the pituitary gland ( cf. Parker, 1948; Waring, 1963). Pituitary preparations from the barracuda Sphyraenu and from the beef cause melanin dispersion in the stingray Urobatis (Weisel, 1950). Recently, Waring ( 1960) applied MSH-containing extracts of ox pituitary to dogfish melanophores both in vivo and in vitro and found them effective in inducing melanfn dispersion. Novales and Novales ( 1966b) demonstrated the dispersing effects of bovine P-MSH and synthetic a-MSH on Squalus melanophores in vitro. No agreement has yet been reached about the mode of the control of melanin aggregation in elasmobranch melanophores. The postulated neural mechanisms will be discussed later (Section V, D). Lundstrom and Bard (1932) and Abramowitz (1939) regarded the aggregation as a passive process, being simply caused by the decrease in the MSH level in blood. Mellinger (1963) claimed that persistent darkening occurs following the section of the hypothalamo-hypophysial tract of Scyliorhinus, and that at the same time cytological features of MSH-producing hypophyseal cells are in a hyperactive state. These observations led him to conclude that the intermediate lobe receives an inhibitory control from the higher center. Based primarily on the observation that following the removal of the

324

RYOZO FUJII

anterior lobe of the pituitary the fishes became dark and failed to respond to a white background, Hogben (1936) proposed the existence of a hormone from the anterior lobe which accelerates melanin aggregation. As discussed later by Abramowitz (1939) and by Mellinger ( 1963), however, more direct evidence is necessary for the demonstration of the hormonal control of pigment aggregation. Although a vast amount of literature has shown that pituitary extracts from various sources affect fish melanophores, only little is known about the chemistry of the pituitary principles which physiologically control the melanin distribution in the cells. Weisel (1950) reported that acetone-dried preparations from both the barracuda Sphyraena and beef cause melanin dispersion in Ameiurus as well as in an elasmobranch Urobatis, but produce aggregation in some teleost species. Such preparations also bring about the dispersion of xanthophore pigment. It was suggested that different species have different sensitivity to a certain hormone and that there is a differential in reactivity of different types of pigment cells in a species. Weisel also supposed that pituitary material from different sources may have unlike action. On the other hand, Kent (1961) suggested that Pleuronectes pituitaries contain a peptide hormone resembling mammalian MSH’s. It causes melanin aggregation in Phoxinus and dispersion in Ameiurus. Upon alkali treatment it loses such effects, but it produces slight dispersion in Phoxinus. Kent claims that these alterations of effects may result from changes in molecular configuration of a single hormonal substance. By using a zone electrophoresis method, Burgers (1963) found that the pituitary of the cod, Gadus morhuu, contains at least two MSH components similar to the MSH’s described in mammals, indicating that in fish the melanin-dispersing agents of pituitary origin may also be MSH’s or similar peptides. Baker (1968a) showed that both a-MSH and /I-MSH are effective in dispersing eel melanophores in uitro. It is interesting that a large amount of MSH was found in the pituitary of the blind Mexican cavefish, Anoptichthys iordani, which does not display adaptive color changes (Burgers et al., 1963). In the review of Pickford and Atz (1957) several instances are cited where the administration of pituitary extracts does not cause pigment dispersion but rather its aggregation. The aggregation might be attributable to the effect of some other hormonal principles than MSH. Healey (1940) suggested the presence of a whitening hormone from the pituitary of the minnow Phoxinus. Enami (1955) disclosed that extracts from the pituitary of the catfish, Parasilurus asotus, induce significant pigment aggregation in the dermal melanophores of the fish both in viuo and in uitro, and referred to the effective principle as “melanophore-concentrating hormone” or “MCH.” He suggested that MCH has its origin in

6.

CHROMATOPHORES AND PIGMENTS

325

the nucleus lateralis tuberis of the hypothalamus. Some of the physicochemical properties of MCH have been studied by Imai (1958). Although it apparently differs from acetylcholine its chemical structure has not yet been determined. Histochemical studies of the pituitary of mugiloid fishes by Stahl (1958) also suggested that the meta-adenohypophysis contains two cell types which perhaps produce two antagonizing melanophore hormones. Kent (1959, 1960) also showed that both the melanophore-aggregating and the melanophore- and erythrophore-dispersing factors are present in both anterior and posterior portions of the pituitary of the minnow Phoxinus, although the amounts are higher in the latter. He thought that these hormones are produced in the anterior part of the pituitary and passed to the posterior part for storage. Working with the same material, on the other hand, Baker (1968b) found that the pars intermedia cells extend anteriorly as far as the pituitary stalk. She pointed out that Kent’s anterior portion of the pituitary would contain pars intermedia cells as well as pars distalis. The two-hormone hypothesis assuming two antagonistic principles, e.g., MSH or related melanin-dispersing peptide and MCH, however, has not been universally accepted. Some researchers consider that the data which have been thought to indicate the presence of two antagonistic hormones can be explained in terms of only one hormone (Waring, 1963). For instance, the above-mentioned rather complicated results by Weisel (1950) and by Kent (1961) have been neatly interpreted by postulating only one peptide principle. Histological studies on the pituitary of a few species of bony fishes by Baker (1963) revealed that in some species a cell type of the pars intermedia shows hyperactive cytological features when the fishes have been adapted on a black background. She failed, however, to detect any activated cells in the pituitary of the white-adapted specimens, suggesting that the melanin-dispersing hormone or MSH is the sole pituitary principle controlling fish melanophores. Another endocrine organ which might have implication in chromatophore control is the pineal. Recently, however, Rasquin (1958) failed to demonstrate any function of the pineal in pigmentation: Its removal or intraperitoneal implantation had no effects on chromatophores. On the other hand, Lerner and Case (1960) isolated a potent blanching substance for frog skin from bovine pineal glands and called it “melatonin.” Its chemical structure is N-acetyl-5-methoxytryptamine, being closely related to serotonin (5-hydroxytryptamine) , In some fish, it strongly aggregates melanophore pigment, while in others it does not (cf. Section V, E ) . It remains possible, however, that it actually functions as a humoral agent of pineal origin in controlling melanosome aggregation. A considerable number of studies have appeared about the control

326

RYOZO FUJII

mechanisms of xanthophores and erythrophores (cf. Pickford and Atz, 1957). Although in some species neural control has been suggested, humoral control by MSH from the pituitary seems to be the more general type of regulation. In fact, in all 35 species subjected to MSH injections, pigment dispersion in xanthophores or erythrophores was observed ( Rasquin, 1958). Whether or not the responsible hormone is chemically classified as a MSH peptide is still open for further investigations. Kohler ( 1952) reported that adrenocorticotropic hormone ( ACTH) causes pigment dispersion in Phoxinus erythrophores and that the effect is inhibited by a prior injection of monoiodoacetic acid. At least one instance has been reported where xanthophore or erythrophore pigments aggregate in response to a pituitary melanin-aggregating hormone, MCH (cf. Pickford and Atz, 1957). The movements of fine purine granules in the leucophores are thought to be mainly controlled by chromatophores nerves (cf. Section V, D ) .

D. Nervous Control of Chromatophores

No evidence has hitherto been put forward for the nervous control of cyclostome chromatophores (cf. Section V, C). Also, in elasmobranchs the hormonal mechanism seems to be the exclusive or at least dominant means of controlling pigment cells. This is especially true for the control of pigment dispersion (Section V, C), and the few suggestions so far for the presence of regulatory nervous supply apply solely to melanosome aggregation. Based on his observation that the blanched band or splotch was formed peripheral to the incision made on the fin or trunk of the dogfish Mustelus, Parker (1935) postulated the presence of melanin-aggregating fibers. According to him, the neurotransmitter involved is oil-extractable and he called it “selachine” (Parker, 1948). Parker’s conclusion ( 1935) was primarily dependent upon his unusual speculation of the presence of the long-lasting generation of nerve impulses at the cut end of the melanin-aggregating axon. Since such persistent discharges at the stump of an axon disconnected from the cell body is unlikely, the possibility of the neural control of elasmobranch chromatophores should be reexamined through different methods from those employed by Parker. In many teleost species including FunduZus, Ameiurus ( Parker, 1948), Macropodus ( Umrath and Walcher, 1951), Phoxinus ( Healey, 1954; Gray, 1956), 0y x i a s (Ando, 1960), and Chmmichthys ( Fujii, 1959a, 1961 ), melanophores are primarily controlled by the autonomic nervous system, and the nerves which function in aggregating the pigment are sympathetic ( cf. Parker, 1948). Using adrenergic blocking agents, Fujii

6.

CHROMATOPHORES AND PIGMENTS

327

(1961), Scheline (1963) and Scott (1965) have affirmed that the transmitter involved is adrenergic, being probably norepinephrine. The pathway of such fibers was shown by von Frisch (1911) on the minnow Phoxinus. According to him the axons originate from a center in the medulla, pass through the spinal cord as far as vertebra 15 where they emerge through rami into the sympathetic chain, finally reaching the integumentary melanophores through spinal and cranial nerves. The fundamental scheme has been rather widely accepted for other bony fishes (Parker, 1948; Scott, 1965). That periphera1 nerves of fishes actually contain epinephrine as well as norepinephrine was recently demonstrated by von Euler and Fange (1961) in a cyclostome Myxine, an elasmobranch Squalus, and in a bony fish Gadus. Regarding the presence of pigment-dispersing innervation, the results are still controversial. Some researchers are of the opinion that such a set of nerves is actually operating to induce active dispersal of melanin (von Gelei, 1942; Parker, 1948; Umrath and Walcher, 1951; Gray, 1956; Fujii, 1959b; Iwata et al., 1959a; Healey, 1967), while others have expressed their doubt about its existence (Healey, 1954; Pye, 1964a; Scott, 1965; Healey and Ross, 1966). Parker et al. ( 1945) detected acetylcholine in the skin of the catfish. The tracing of the dispersing motor connections from the center in the brain has also been tried, and von Gelei (1942) depicted it in the minnow Phoxinus: Through the first or second rami the fibers flow out of the spinal cord, and pass posteriorly through the sympathetic chain, arriving finally at the melanophores via spinal nerves. This plan, however, has been criticized by some later workers (Healey, 1954; Gray, 1956; Pye, 1964a; Scott, 1965). The major and most widely known argument for the presence of melanin-dispersing axons was presented by Parker ( 1948) who, using mainly Fundulus and Ameiurus, observed the behavior of the melanophores on a caudal band induced by cutting a bundle of radiating nerves. Among the observations he presented as strongly supporting his double innervation theory are the foIlowing: First, a second incision within the previously formed and faded band on a caudal fin produces a secondary darkened area between the cut and the free edge of the fin. Second, a localized cold point applied just distal to the incision limits the spread of the darkened area beyond the block. These results led Parker to insist that melanin dispersion distal to the cut results from the action of the transmitter liberated by repetitive firing of the melanin-dispersing axons at their cut ends. Although it has been supported by some later investigators ( Umrath and Walcher, 1951; Gray, 1956), this explanation has met with considerable opposition from a number of sources (Healey, 1954; Pye, 1964a; Scott, 1965).

328

RYOZO FUJI1

Fujii and Novales (1969b) followed Parker’s procedure on the caudal fin of the goby Chasmichthys, and found that a second incision does not induce a dark zone within the primary band, and that a cold block does not interfere with the generation of the peripheral dark zone. Together with Fujii’s previous observations that in isolated tail fin preparations the dispersed state of melanophores is maintained by spontaneous release of the dispersing transmitter from the presynaptic structures involved, (Fujii, 1959b; cf. also Section V, E ) , they came to the following conclusion about the sequence of the so-called Parker effect: As soon as the cut is made, the distal melanophores become free from nervous control. The axons in the dark band which were disconnected from their cell bodies can survive for some time, but, unless stimulated by some artificial means, they do not fire. In spite of the absence of axonal discharges, however, the spontaneous release or leakage of the dispersing transmitter continues until its exhaustion occurs in the presynaptic structures. The released transmitter, of course, brings about the dispersion of the melanophores within the band. Fujii (1959a), working on the caudal band of the goby, reported that the store of the dispersing transmitter in the remnant axons seems to disappear 3-5 hr after incision at 26°C. On the other hand, the dark band remains visible for days or weeks. One explanation for this secondary phase of the caudal band is that in the denervated area the melanophores are rather free from the influence of either melanin-dispersing or -aggregating transmitter liberated in the surrounding innervated area, since inactivating enzymes for them in the tissue space, possibly acetylcholinesterase and catecholamine-O-methyltransferase,may metabolize the transmitter molecules which invade the denervated region. Thus, the melanosomes in the melanophores in the band distribute passively by Brownian movement without being affected by the transmitters. In such a condition, we can easily envisage the homogeneous distribution of melanosomes throughout the cytoplasm, i.e., the dispersed state of the melanophores. In respect to not assuming the continued firing in the dispersing fibers, the present explanation is a revival of the socalled “paralysis theory,” which was critically reviewed by Parker ( 1948). Although the proposed mechanism for the remarkable effect is quite different from that proposed by Parker, it also supports the presence of dispersing neural control of melanophore pigment. By making use of the Derived Ostwald Index (Section V, A ) , Healey ( 1967) could discriminate the rapid and slow color changes of the minnow Phorinus, these being produced by nervous and humoral mechanism, respectively. His finding of the recovery of the rapid darkening response from chronic denervation also supports the presence of dispersing fibers. Working with the tail of the catfish Ameiurus, Parker and Rosenblueth

6.

CHROMATOPHORES AND PIGMENTS

329

( 1941) claimed that the melanin-aggregating and -dispersing nerve fibers could be independently stimulated by changing the frequency, duration, and intensity of repetitive pulses. Although they regarded their observations as supporting the concept of the double innervation of fish melanophores, no one seems to have ever succeeded in confirming their results. For instance, Fujii and Novales (1969a) observed only aggregation of pigment in Fundulus melanophores in response to nervous stimulation, even though various parameters of the stimulating pulses were changed widely. Although both the melanin-aggregating and -dispersing fibers may have been stimuIated by electrical shocks, the effective amount of the transmitter from the former axons seems to consistently overwhelm that from the latter. Fujii and Novales (1969a) showed that even a single electrical shock to nerves causes discernible melanin aggregation in a Fundulus melanophore. Since this response is not of all-or-none type, they concluded that more than two nerve fibers control pigment aggregation in a melanophore. The antidromic propagation of impulses in chromatophore nerves has been described in a few teleost species (cf. Parker, 1948; Ueda, 1955). By making use of tetrodotoxin, which effectively blocks the conduction of chromatic nerves without affecting melanin movements in melanophores or the secretory activity of presynaptic structures of nervous elements (Fujii and Novales, 1968a), Fujii and Novales ( 1969a) have recently confirmed the presence of such retrograde conduction of nervous excitation. Since in their experiments, the possibility of the spread of electric current to the proximal melanophores was completely excluded, the responses seen among the melanophores were clearly the result of the nervous excitation induced at the peripheral part of controlling nerves. From these observations, they presented an idea that axons passing close to the melanophores release transmitter substance at intervals along their length to control a number of adjacent melanophores. Recently, Fujii (1966a) and Fujii and Fujii (1966) on the goby Chmmichthys, Fujii (1966b) on the guppy Lebistes, and Bikle et al. ( 1966) on the killifish Fundulus have referred to their findings of the nerve endings or presynaptic structures containing synaptic vesicles near the surface of the melanophores. The synaptic vesicles observed are empty in electron micrographs and about 500 A in diameter. These are thought to be the packets of the melanin-dispersing transmitter. Mixed with these usual vesicles, larger granulated vesicles which measure about 1000 A in diameter were found in the nervous elements around Lebistes melanophores (Fujii and Novales, 1969b). Since this kind of vesicles is known to be the storage site of the catecholamine transmitter, these elements may

330

RYOZO FUJII

be the presynaptic structures of the melanin-aggregating nerves. These observations may afford morphological evidence for the concept of double innervation of fish melanophores. Since the 500-A empty synaptic vesicles mentioned above have generally been identified as the quanta1 packets of acetylcholine, we can now suppose that acetylcholine might be involved in inducing melanin dispersal in the fish melanophores. However, in spite of the fact that the transmitter concerned has often been stated to be cholinergic (Parker, 1948; Robertson, 1951; Umrath, 1957), its nature is still a matter of speculation, because acetylcholine has also been often claimed to be ineffective or very weak in inducing pigment dispersion in melanophores (Watanabe et at., 1962a; Healey and Ross, 1966). Substances other than acetylcholine may be taking part as transmitters for melanin dispersal. If so, the small vesicles without core in nerves should be assumed to be the parcels of a new kind of transmitter. Based on their observations on Oryzias melanophores that in some cases epinephrine causes melanin dispersion where normally aggregation is expected, Watanabe et al. (1962b) recently came to the conclusion that epinephrine might be involved both in melanin-aggregating and -dispersing neural mechanisms. Fujii and Fujii (1968) have lately suggested that not all the melanophores of some fish are innervated. Their conclusion was based mainly on the paucity of nervous elements around the melanophores found in electron micrographs and on the presence of cell-to-cell contacts among the melanophores. In such a system of coordination, the nervous excitation might be transmitted at first to the melanophores in proximity with the nerve bundle, and then one after another to the melanophores connected with the activated cells. In Fundulus, on the other hand, every melanophore on the tail seems to be under the direct control of the nervous system, at least by melanin-aggregating fibers (Fujii and Novales, 1969a). All the melanophores of the guppy, so far examined on electron micrographs, are richly innervated ( Fujii, 1986b). Recently, it was found that alkaline earth ions are necessary for the transmission of excitation through the junction between melanin-aggregating nerve and melanophore (Fujii and Fujii, 1965, in Chamnichthys, Fujii and Novales, 1968b, in Fundulus). In contrast to the well-known antagonism between Ca2+and Mgz+at neuromuscular junctions, Mg2+can substitute for Ca2+,and this is also the case with other alkaline earths. This suggests that the neuromelanophoral junction might be phylogenetically more primitive than other various kinds of vertebrate junctions, since, in the former, a discriminating mechanism for Ca2+and Mg2+has not been developed. Iwata et al. (1959a) claimed that after the isolation of a Caradus

6.

CHROMATOPHORES AND PIGMENTS

331

scale the melanin-aggregating fibers lose their function faster than the dispersing ones. This is rather contrary to the conclusion obtained in Chasmichthys by Fujii (1959a) where the function of the dispersing axons seem to degenerate first. The normal time course of the changes in the state of Carassius melanophores after the isolation of a scale was markedly shortened by KCN added to the physiological saline (Watanabe, 1960a). The cyanide may destroy the function of nerves at first, and then that of melanophore itself. Effects of ionizing radiation on the melanophore nervous system were studied in Carassius carmsius ( Etoh, 1963). The melanin-dispersing process is more sensitive to y-rays than the aggregating mechanism. The radiosensitivity of the axons was highest, and then nerve endings. Pigment motility in the melanophore itself is very resistant to radiation. Although hormonal control by the pituitary may be a predominant method of xanthophore and erythrophore regulation (Section V, C ) , there are a few reports suggesting nervous control of these chromatophores. Fries (1943) concluded that in Fundulus both pigment-aggregating and -dispersing nerves are operating to control xanthophores, in addition to hormonal factors. Brantner (1956) claimed that erythrophores of the bitterling Rhodeus are innervated by pigment-aggregating fibers. In many cases the changes in the distribution of pigment granules in the leucophore cause alterations in the whitish coloration or the reflective ability of the skin. These cells are responsive to background changes, i.e., on black bottom they contract, while on white they expand (Fries, 1958). Odiome ( 1933) found that epinephrine causes pigment dispersion in Fundulw leucophores. Miyoshi (1952) reported that K ions have the same effect on leucophores in an isolated O~yxiasscale, while in NaCl or in physiological solution pigment granules are aggregated in the central bodies of the cells. Past experiments of injecting pituitary preparations suggest that the pituitary does not play a significant role in the control of leucophores (Pickford and Atz, 1957). Based on the observation that leucophores of hypophysectomized Bathygobius are still responsive to background changes, Fries ( 1958) concluded that leucophores are mainly controlled by nerves, Nerve cutting causes aggregation of leucophore granules (Fries, 1958). Reviewing these observations, one finds that stimuli which induce pigment aggregation in melanophores usually produce dispersion of leucophore granules, whereas melanin-dispersing stimuli aggregate pigment in the leucophore. It is therefore possible that the chemical transmitter from the melanin-aggregating nerves acts as the leucophore-dispersing transmitter, and the melanin-dispersing transmitter acts in turn as the leucophore-aggregating transmitter. Finally, there is no evidence for nervous control of iridophores.

332

RYOZO FUJII

E. Effects of Chemicals and Drugs on Chromatophores That K and other alkaline ions except Na cause aggregation of melanosomes was frst observed by Spaeth (1913) on the scale melanophores of Fundulus. Later he also described the identical action of alkaline earth ions (Spaeth, 1916a). Kamada and Kinosita ( 1944), Nagahama ( 1953), and Iwata and Yamane ( 1959) have presented quantitative descriptions of the effect of K ions on the scale melanophores of some freshwater forms. Among alkaline ions and alkaline earths, K, Rb, Cs, NH,, Ba, and Sr ions are generally effective enough to induce melanin aggregation (Spaeth, 1913,1916a; Kamada and Kinosita, 1944; Fujii, 1959a; Watanabe et al., 1965). Lithium is less effective. In some fish Ca and Mg ions can aggregate melanin, while in others they are less potent (cf. Watanabe et al., 1965). No response was observed in response to these ions in Chasmichthys ( Fujii, 1959a). Immersion of an isolated scale of Carassius into NH&l solutions induces melanin aggregation ( Watanabe and Mori, 1964). Upon soaking the scale in the physiological saline again, melanin dispersion takes place which is immediately followed by reaggregation of pigment. When the Carasdus scale is alternately exposed to glucose and any one of the alkaline or alkaline earth salts, the latter’s melaninaggregating action is gradually reversed in such a way as to cause dispersion ( Watanabe et al., 1965). The frst interpretation of the mechanism whereby the melanin-aggregating ions act on the chromatophores was given by Spaeth (1913). Based on his observations on scale melanophores of Fundulus, he concluded that these ions act directly on the melanophores without any intervention of nervous mechanisms. An alternative explanation has recently been put forward by Fujii (1959a). Principally based upon the observations that denervated melanophores are unresponsive to K and other melanin-aggregating cations and that the liberation of the aggregating transmitter is involved in their action on innervated cells, Fujii concluded that K and other melanin-aggregating ions do not act directly upon the melanophores but act on the aggregating nerve endings. The liberated transmitter then brings about the aggregation of melanosomes in the cells. Working with normal or denervated melanophores in the scales of Carassius, Iwata et al. (1959a,b) have reached the same conclusion. They further observed that after urethane treatment the axonal stimulation by electric current becomes ineffective, whereas the responses to K ions or to epinephrine remains unchanged. Fujii ( 1961) added the observation in favor of this explanation that the effect of K+ can be completely blocked after the treatment of the tail fin preparation

6.

CHROMATOPHORES AND PIGMENTS

333

with an adrenergic blocking agent, dibenamine, indicating that K action is mediated through the secretion of adrenergic transmitter. Observations that denervation or cocaine treatment causes a loss of K+-induced aggregation in Oryzias scale melanophores are also in agreement with this view (Ishibashi, 1962). Recently, Fujii and Novales (1968a) reported that the K+-induced response was still observable after the application of tetrodotoxin, which effectively blocks the conduction of melanin aggregating nerves, thereby excluding the possibility of axonal involvement in K+ action. Increase in the external K+ concentration causes the depolarization of melanophore membrane according to Fujii and Novales (1969~).Since changes in C1 - concentration had practicaIIy no effect on the resting potential, they concluded that K+ conductance may be the main factor in the generation of the membrane potential of melanophores. These results on Fundulus suggest that, although melanin aggregation resulting from K+ is mediated through the transmitter action (Fujii, 1959a), these ions have some direct effects on the melanophores, at least by lowering its membrane polarization. The melanophores and the leucophores behave quite oppositely (cf. Section V, D ) . In the leucophores of Oryzias dispersion of pigment particles occurs in KC1 solution, while aggregation is seen in the balanced salt solution ( Miyoshi, 1952). Spaeth ( 1913) recorded that melanosomes disperse widely within .the melanophores when the isolated scales of Fundulus are immersed in solutions of neutral salts of Na. Since then, many researchers have conveniently used isotonic NaCl or proper physiological saline solutions for the purpose of obtaining the dispersed state of melanophores in various species of fish. Spaeth concluded that Na+ acts directly on the melanophore. Since it is rather easy to suppose an analogy between chromatophores and muscle cells (cf. Spaeth, 1916), most researchers have been inclined to regard the dispersed state of melanophores as corresponding to the resting or relaxed state of muscle fibers. Thus, little attention has been paid to the analysis of pigment dispersal in Na+-rich solutions until recently, On the other hand, the double innervation theory for teleost melanophores (cf. Parker, 1948; Section V, D ) necessarily requires that pigment dispersion be another active process reversing aggregation. These studies, however, have been primarily concerned with the activation of melanin-dispersing nerves, independent of the ionic action. An attempt was made to prove the dispersal in Na+-rich medium to be an active process (Fujii, 1959b). Under the influence of Na+, it was found that the rate of melanin dispersion is faster in innervated melanophores than in denervated cells and that the previously blanched denervated

334

RYOZO FUJI1

band was marginally invaded by the surrounding darkened area. These observations indicate that in a normally innervated area the difFusibIe melanin-dispersing transmitter is secreted under the influence of Na+. Blocking of nervous conduction did not change the dispersion response characteristic of Na+. Thus, Fujii (1959b) maintained that Na+ may act selectively on the endings of the dispersing nerves to induce the liberation of the transmitter concerned which then causes melanosome dispersion inside the cells. Although the dispersion in Na+-rich solution is perhaps an active process, this explanation should be somewhat modified as later commented upon by Fujii (1961), i.e., the release of the transmitter may be of spontaneous nature, and Na+ may provide a proper environment in which the rate of liberation of the antagonistic melanin-aggregating transmitter is minimized. Spaeth (1913) considered the neutral salts of K+ to be the cause of pigment dispersion in Fundulus xanthophores; the effect is opposite in the melanophores. He also noted that anions including C1-, Br-, NOs-, I-, and SCN- are not primarily concerned with the state of chromatophores but affect the survival of the cells in the isolated scales. Many hormonal substances other than MSH and MCH are known to affect the state of fish chromatophores either in vivo or in vitro. Loud and Mishima (1963) observed that both the melanophores and xanthophores of goldfish respond similarly to ACTH in vitro by assuming a more finely dendritic and stellate shape. Employing the freshwater characin Astyanax, Rasquin ( 1958) considered mammalian anterior pituitary hormones, FSH, LH, TSH, prolactin, and growth hormones to be ineffective in causing melanin movements. She also noted that these hormones induce pigment dispersion in xanthophores of Bathygobius. Robertson (1951) indicated that extracts of mammalian thyroid caused marked aggregation of the melanophores of the rainbow trout, Sa2mo gairdneri. Using the brown trout, Salmo trutta, Woodhead (1966) reported that thyroxine aggregates melanophore pigment, while thiourea disperses it. Cortisone has no effect on the trout melanophores both in vivo and in vitro ( Robertson, 1951). Epinephrine caused melanin aggregation in Mustelus melanophores ( Lundstrom and Bard, 1932). Parker ( 1948), reviewing past observations, however, concluded that adrenergic substances probably play no part in the normal paling reaction in elasmobranchs. The melanosome aggregation resulting from the action of epinephrine is certainly one of the most remarkable phenomena in the chromatic physiology of teleost fishes (cf. Parker, 1948; Fujii, 1961). Several investigators have been making use of its potent action in their studies of chromatophore mechanisms (e.g., Kamada and Kinosita, 1944; Kino-

6.

CHROMATOPHORES AND PIGMENTS

335

sita, 1953; Fujii, 1959b; Iwata et al., 1959a,b; Fujii and Novales, 1968a). On the other hand, Breder and Rasquin (1955) reported that some teleost fishes such as Chaetodipterus faber showed dark markings after injections of epinephrine. Enami (1955) also reported that the catfish Parusilurus darkens after injections of epinephrine. Rasquin ( 1958) surveyed its effects in a variety of fish and noted that melanophores of some fish are unresponsive to epinephrine. Watanabe et al. (1962a) described that Oryzias melanophores sometimes respond to epinephrine by dispersion. The excitement of a fish may cause a sudden pallor. This phenomenon was first recorded by von Frisch ( 1911), and has been ascribed to a sudden secretion of epinephrine from the adrenal medulla. Burgers et al. (1963) observed a similar pallor in a blind Mexican cavefish Anoptichthys in which no adaptive color response is expected. Injections of epinephrine also cause marked melanosome aggregation in melanophores of this species. Reidinger (1952) claimed that epinephrine does not aggregate pigment in xanthophores of Phoxinus. Chavin (1956) has demonstrated that following hypophysectomy or injection of epinephrine the pigment of goldfish xanthophores as well as melanophores become centrally aggregated. An extensive survey by Rasquin (1958) showed that either xanthophores or erythrophores respond to epinephrine by dispersion or aggregation of pigment granules, depending upon the species. Both Fries (1958) and Rasquin (1958) reported that leucophores can respond to epinephrine injection by dispersion of guanine granules. Recently a few reports have appeared which indicate that norepinephrine is effective in arousing melanosome aggregation. These are by Umrath (1957) on Rhodeus, by Fujii (1961) on Chasmichthys, by Fange (1962) on Gadus and Lebistes, by Scheline ( 1963) on Labrus, by Scott (1965) on Scopthalmus, and by Healey and Ross (1966) on Phoxinus. The tissue-cultured melanophores of the goldfish also respond to norepinephrine by aggregation, while the ACTH-induced melanocytes either show little or no response to it (Hu, 1963). Since norepinephrine is currently believed to be the melanin-aggregating transmitter ( Section V, D ) , its potent action is quite natural. Effects of some adrenergic blocking agents have been examined in a few species of fish. Dibenamine (Fujii, 1961) and phenoxybenzamine ( Scheline, 1963), both being P-holoalkylamines, inhibit the action of catecholamines. These results indicate that the melanin-aggregating transmitter is of adrenergic nature, and that the adrenergic a-receptors are responsible for the pigment aggregation in teleost melanophores. Watanabe et al. (1962a) and Scott (1965) stated that dibenamine has melanin-

336

RYOZO FUJII

dispersing action, Ergotamine, another adrenergic blocker, elicits local pigment dispersion in the sand flounder ( Scott, 1965). However, it causes melanin aggregation in trout melanophores ( Robertson, 1951). The melanin aggregation resulting from ergotamine is usually followed by a reversal of catecholamine action ( Fujii, 1961), the observation being later confirmed by Healey and Ross (1966). Ergotamine induces temporary dispersion of Bathygobius leucophores ( Fries, 1958). Lysergic acid diethylamide ( LSD ) produces melanosome dispersion in guppy melanophores (Cerletti and Berde, 1955). They also showed that this effect of LSD is inhibited by 5-hydroxytryptamine (5-HT, serotonin). Hydergine, guanethidine, and piperoxane have a slight aggregating effect, while phentolamine, bretylium, and yohimbine induce melanosome dispersion ( Healey and Ross, 1966). Hydrazines cause local melanin aggregation, when injected into Scopthalmus (Scott, 1965). Their action does not result from monoamine oxidase (MAO) inhibition, since other MA0 inhibitors, iproniazid and isocarboxazid, are without effects. Pretreatment of fishes with pyrogallol markedly potentiates epinephrine or norepinephrine action (Scott, 1965).Since pyrogallol is known to potentiate the effect of catecholamines by competitively inhibiting catecholamine-0methyltransferase, the active participation of the enzyme in the metabolism of catecholamines in chromatophore regulation is suggested. 5-Hydroxytryptamine was reported to have melanin-aggregating action (Scheline, 1963; Scott, 1965). It has no such effect in Chasmichthys (Fujii, 1961). It does not have any significant action upon Phoxinus melanophores in viuo either in aggregating or in dispersing pigment (Healey and Ross, 1966). Preliminary results of Novales (1968), using the isolated skin of the dogfish Squalzrs, indicated that melatonin slightly accelerates the paling effect of elasmobranch saline solution. It also significantly antagonizes the darkening effect of MSH. Working with Chasmichthys, Fujii (1961)has shown remarkable melanin aggregation with melatonin. Since denervated cells are also responsive, he concluded that melatonin acts directly on the melanophores. Healey and Ross (1966) observed that it causes paling of black-adapted Phoxinus. Unpublished data by Boyles (1969) also demonstrate its potent aggregating action on goldfish melanophores in vitro. The tissue-cultured melanophores of goldfish responded to melatonin by aggregation also (Hu, 1963). On the other hand, melanophores of certain species seem to be unresponsive to melatonin. Namely, practically no response was seen in Fundulus (Mori, 1961) and in Carassius carassius (Etoh, 1961). Fain and Hadley (1966) claimed that although melanophores of adult Fundulus are refractory, embryonic or larval cells respond to melatonin by pigment aggregation. They believe that loss of the sensitivity to melatonin may be related to the developmental

6.

CHROMATOPHORES AND PIGMENTS

337

acquisition of innervation. Hu (1963) reported that the ACTH-induced melanocytes of goldfish are unresponsive to melatonin. Tyramine strongly aggregates pigment granules in the chromatophores of the cuckoo wrasse Labrus (Scheline, 1963). A reserpine pretreatment of the scale which may diminish peripheral store of catecholamines results in the ineffectiveness of tyramine in inducing melanin aggregation. The action of tyramine is restored by exposing the scale to catecholamines. These results led Scheline to conclude that tyramine acts to release catecholamine stores in presynaptic elements of the aggregating nerves. Dopamine is also extremely active in inducing pigment aggregation (Scott, 1965). The paling effects of dopamine and other sympathomimetic drugs including ephedrine and tyramine were recently confirmed by injecting them into dark-adapted minnows (Healey and Ross, 1966). Histamine shows a slight aggregating effect, while antihistaminic drugs produce marked dispersion in trout melanophores ( Robertson, 1951). When injected into the minnow, histamine has no definite action on melanophores ( Healey and Ross, 1966). Agreement has not been reached about the effect of acetylcholine. Believing it to be the melanophore-dispersing transmitter, Parker ( 1948) referred to a few reports which detected its effectiveness in inducing melanin dispersion. Such an action has also been reported in some recent papers (Robertson, 1951; Reidinger and Umrath, 1952; Ando, 1960; Green, 1968). Umrath ( 1957) reported that acetylcholine acts to disperse erythrophore pigment in Rhodeus. On the other hand, very high concentrations of acetylcholine were required to produce a detectable antagonizing effect to epinephrine action on Oryzius melanophores (Watanabe et at., 1962b). Acetylcholine and/or eserine do not induce darkening in the injected Phoxinus, while carbachol does (Healey and Ross, 1966). Scott (1965) indicated that when injected into the spinal cord both acetylcholine and eserine induce the excitation of pigment-aggregating fibers, while no local action on melanophores can be detected. Both pilocarpine and eserine disperse pigment in Salmo (Robertson, 1951) and in Oryzias melanophores ( Watanabe et al., 1962b). Atropine’s remarkable action in dispersing melanosomes is rather peculiar ( Fujii, 1960). Similar conclusions were reached by both Fujii (1960) and Watanabe (196Ob) that it acts directly on the cells, since it also disperses melanin in denervated melanophores. Its action is slight when injected into a live Phoxinus (Healey and Ross, 1966). In addition to its marked dispersing action in higher concentrations, Reidinger and Umrath (1952) demonstrated that in the scale melanophores of Macropodus atropine at lower strengths inhibits or weakens the pigment dispersing action of acetylcholine, Using erythrophores of the anal fin of

338

RYOZO FUJI1

Rhodeus, Umrath (1957) also observed the inhibitory effect of atropine on acetylcholine action. Scopolamine also disperses melanin in trout melanophores both in vim and in oitro (Robertson, 1951). Consistent with their blocking action in autonomic ganglia, both hexamethonium and presidal cause some darkening of white-adapted Phoxinus (Healey and Ross, 1966). Nicotine and caffeine are effective in dispersing melanosomes in Oryzias ( Watanabe et al., 1962a). Cocaine shows a potentiating action on the responses of Phoxinus melanophores to epinephrine ( Healey and Ross, 1966). Procaine, on the other hand, causes dispersion of melanosomes in Oryzius (Watanabe et al., 1962a). It blocks conduction of melanin-aggregating fibers in Fundulus (Fujii and Novales, 1968b). The puffer poison, tetrodotoxin, also has the same effect (Fujii and Novales, 1968a). Turner and Carl (1955) reported that some species of fish become intensely colored when they are immersed into reserpine solutions. It was also observed that the reserpine-treated fish does not show excitement pallor. Scott ( 1965) mentioned that reserpine injection causes local darkening of skin. Scheline (1963) concluded that reserpine does not act at the receptor site, but on the neural elements to deplete a catecholamine store. Chlorpromazine has no definite effect of Phoxinus melanophores ( Healey and Ross, 1966). Phenothiazine tranquilizers were found to be potent pigment-dispersing agents (Scott, 1965). Their effective doses are raised after the injection of fish with pyrogallol. The conclusion reached was that these compounds disperse melanin in the same manner as adrenergic blockers do. Hypersensitivity of denervated melanophores to certain substances has been reported several times: Using the goby Chasmichthys, Fujii ( 1961) observed that the effective concentrations of epinephrine, norepinephrine, and melatonin to induce melanin aggregation are significantly lower in denervated cells. Healey and Ross (1966) also noted such a hypersensitivity of Phoxinus melanophores in response to epinephrine, norepinephrine, or to ergotamine. Such a heightened responsiveness was also reported in a dispersing agent, atropine (Fujii, 1960). The process of the hypersensitivity development is still a matter of speculation. However, the elucidation of this problem may surely be helpful for further understanding of receptor mechanisms in chromatophore control. VI. MORPHOLOGICAL COLOR CHANGES

A number of stimuli cause the increase or decrease in either the net content of pigmentary substances or the number of pigment cells. These

6.

CHROMATOPHORES AND PIGMENTS

339

changes have commonly been referred to as morphological color changes. Usually morphological and physiological color changes proceed simultaneously, although the latter process is must faster. For example, fishes kept in a dark container for a long time develop an increase in both the number of melanophores and the melanin content of the skin (Odiorne, 1957). Rearing a goby on white or dark background results in an increase or decrease in both the number and the guanine content of leucophores, respectively ( Fries, 1958). Among the agencies which affect melanogenesis, MSH is most widely known for its ability to increase the melanin content in the integument of fishes. According to Odiorne (1948), the prolonged dispersion of melanosomes in response to MSH provides a favorite condition for the synthesis of melanin in the melanophores. In response to injected MSH, Pickford and Kosto (1957) detected an increase in skin melanogenesis in Fundulus. On the other hand, Reidinger (1952), using the minnow Phoxinus, found that the yellow background adaptation resulting from xanthophore activity is controlled by both epinephrine and MSH, and that epinephrine actually increases the number of xanthophores without causing dispersion of pigment inside the cells. The conclusion derived was that the morphological color change is not initiated by the physiological color change and is directly regulated by hormonal substances. Brantner (1956) and then Umrath (1957) have confirmed this idea on erythrophores and melanophores of the bitterling Rhodeus, respectively. Kosto et al. (1959) actually found that MSH stimulates the proliferation of melanocytes in Fundulus. Hu and Chavin (1960) stated that MSH is inactive in promoting melanogenesis in vitro. However, Tchen et al. (1964) recently indicated that both a- and p-MSH stimulate the formation of melanophores from premelanophores. Chavin (1956) showed that the immersion of xanthic goldfish in a salt solution causes the induction of melanophores in the skin, whereas the same treatment of hypophysectomized fish fails to do so. He claimed that among pituitary hormones only ACTH is capable of initiating melanogenesis in the skin. These observations led him to conclude that in a salt solution the pituitary releases a large amount of ACTH. Hu and Chavin (1960) confirmed the ACTH-induced melanization in a tissue-cultured fin of xanthic goldfish. Using Fundulus, Kosto et al. (1959) also reported that ACTH has melanogenetic action. Loud and Mishima (1963), using Masson's ammoniated silver nitrate staining for premelanin, concluded that precursor cells of ACTH-induced melanized cells are xanthophores or premature xanthophores. Their observations a t the fine structural level have disclosed that so-called large bodies (Section 111) which react positively to the premelanin stain and

340

RYOZO FUJII

have similar sizes to melanosomes are the site of melanization. They believe that in the xanthophore cytoplasm ACTH acts to facilitate the association of tyrosinase with large bodies where the melanization is then triggered. By making use of colchicine, Kim et al. (1961) reported that the formation of melanophores from promelanophores by ACTH involves a nuclear division. Chavin et al. (1963) found that ACTH-induced melanogenesis in xanthic goldfish is accompanied by an increase in both absolute and particulate tyrosinase activity in the skin. After hypophysectomy, tyrosinase activity decreased. This decrease is restored by chronic administration of MSH, ACTH, or prolactin ( Kosto et al., 1959). It was also found that prolactin actively promotes melanin synthesis in Fundulus, possibly by increasing the supply of endogenous precursor. The melanophores induced by ACTH in the skin of xanthic goldfish contain the same kinds of pteridines as those found in melanophores of black goldfish ( Matsumoto, 1965b). Ionizing radiation is also effective in inducing melanogenesis; X rays are found to stimulate the formation of new melanocytes and eventually melanophores in xanthic goldfish (Egami et al., 1962). Egami et al. ( 1962) concluded that the ionizing radiation increases the secretion of ACTH from the pituitary which results in melanin synthesis in the skin. These stress-induced melanophores become supplied with nerves controlling pigment movements (Etoh and Egami, 1963), although during early stages of their appearance, i.e., melanocytes or early melanophores, they are unresponsive to neural stimuli. The treatment with thyroxine of larvae of the brown trout Salmo trutta, significantly increases the number of melanophores, whereas in the thiourea-treated group the number decreases (Woodhead, 1966). On the other hand, the thyroxine-treated fish contains less melanin than the control, and the thiourea-treated contains more. Matsumoto (1965b) showed that experimental albinism, induced by treating with phenylthiourea results in the enhancement of xanthophore pigmentation in goldfish and carp. This is because of the increase in the level of sepiapterins in xanthophores. Umrath (1959) demonstrated in the minnow and the bitterling that nuptial coloration can be induced by ACTH or MSH administration. He also found that sex hormones can also elicit similar color patterns. Arai and Egami (1961) reported that androgenic steroids are effective in increasing the number of leucophores in adult male Oyzias, while cholesterol, progesterone, and estrogens are not. Epinephrine and some parasyrnpathomimetic substances including yohimbine are also able to cause the nupital coloration in some fish (Umrath, 1959). Epinephrine may act through the secretion of ACTH.

6.

CHROMATOPHORES AND PIGMENTS

341

Chavin (1963) found that hydroquinone is effective in destroying melanocytes and melanophores of black moor goldfish. The cytopathological sequence is clumping of melanosomes, cytoplasmic vacuolation, pinching off of cellular processes, and cell lysis. The process of the melanophore destruction following hypophysectomy is different from that observed after hydroquinone. The melanosomes aggregate to the center of the cell, and about 3 days after the operation the cell lyses. From these observations Chavin concluded that the action of hydroquinone is directly on the cell, not through the function of the pituitary. A variety of substances have been categorized as depigmentary agents ( Chavin and Schlesinger, 1967). The cytological alterations leading to cell death are generally similar despite the chemical nature of these compounds. It was further concluded that the mode of action of these agents is not on tyrosinase but on a vital subcellular constituent. It is interesting from the clinical point of view that hydroquinone effectively destroys melanoma cells in hybrid swordtail Xiphophorus (Chavin and Schlesinger, 1967).

VII. OTHER TOPICS

Various physical stimuli induce pigment movements. An exposure to ultraviolet light of isolated scales of Fundulzls causes the aggregation response of the melanophores, although visible light fails to do so (Spaeth, 1913). Robertson (1951) claimed that even visible light elicits trout melanophore aggregation. The action, however, may be through the nervous activities. An increase in temperature accelerates the aggregating effect of K ions (Spaeth, 1913). Robertson ( 1951) stated that high temperature itself causes melanosome aggregation in the trout melanophores both in vivo and in vitro. Ando (1962) observed the same effect in embryonic Oryzias melanophores. A local application of warm water brings about the maximal melanophore dispersion in Phoxinus, whereas a cool jet induces melanin aggregation (Pye, 1964b). Mechanical pressure applied by a needle arouses the aggregation of melanosomes and the dispersion of xanthophore pigment in Fundulus ( Spaeth, 1913). Robertson ( 1951) c o n h e d this effect in the SaZmo melanophore, suggesting that the effect is probably produced by K ions liberated from the underlying muscle fibers. Very high hydrostatic pressure, on the other hand, produces dispersion of Fundulus melanophores, possibly resulting from the isolation of the cytoplasm (Marsland, 1944). Electric shocks are known to cause melanin aggregation in teleost cells (cf. Section V, D ) . Since denervated

342

RYOZO FUJI1

cells are refractory to an electric field, the effect may be solely through the activation of the nervous system (Fujii, 1959a; cf. also Section V, D ) . In low pH solutions the Brownian motion of melanosomes in isolated scales of some freshwater fish ceases (Kamada and Kinosita, 1944). The melanosomes also lose their mobility in alkaline solutions, but in the aggregated state. Robertson (1951), on the other hand, stated that low pH induces melanin aggregation, while high pH causes its dispersion both in uiuo and in uitro. Hypertonicity of the bathing media inhibits the Brownian movement of melanosomes, and at the same time the responses to both melanin-aggregating or -dispersing stimuli ( Kamada and Kinosita, 1944). The well-known concept that chromatophores of fish are, like those of other vertebrates, derived from the neural crest has been confirmed by recent investigators ( Orton, 1953; Gordon, 1959; Shephard, 1961) . Newth (1951) indicated that the melanophores of the lamprey Lampetra are also of neural crest origin. The responses of embryonic chromatophores are often different from those of the adult. Such activities of pigment cells have been designated “primary color responses,” while the adult type responses which enable the animal to adapt to backgrounds are called “secondary color responses” (cf. Parker, 1948). The primary responses are exerted through routes other than the eyes. They may occur through the direct action of light or some other physical factors on pigment cells or through the involvement of receptors other than eyes. They are normally masked with the coordinated secondary color responses in adults. The latter responses disappear in blinded animals where the primary responses can often be seen. Ando (1962) reported that the responsiveness to various stimuli of embryonic Oryxias melanophores develop gradually and that soon after hatching adult-type responses of melanophores are observed. Using embryos of the rainbow trout, Sdmo irideus, Fujii (1962) reached the conclusion that melanophores first appear in the integument without nervous supply and later become innervated. Both tyrosinase and Dopa-oxidase are gradually synthesized in the course of embryonic development of Oryzias (Tomita and Hishida, 1961b). It was also found that embryos cultured in media containing inhibitors such as phenylthiourea lack any activity of these enzymes. In the transparent-scaled goldfish, complete degeneration of iridophores takes place about 10 days after hatching. Neither pituitary nor thyroid gland takes part in the manifestation of gene action (Kajishima, 1960a). During the course of the iridophore degeneration, the amount of free purine N decreases, suggesting that guanine is the most important portion of free purine bases in the goldfish (Kajishima, 1960b). The degeneration is accompanied by an increase in guanine deaminase activity, indicating its participation in decomposing guanine granules in larval iridophores.

6.

CHROMATOPHORES AND PIGMENTS

343

Integumentary silvering is the most remarkable feature during parrsmolt transformation of the Atlantic salmon, Salrno salar. Actually, Johnston and Eales (1967) found that both guanine and hypoxanthine contents increase during this process. They further detected a correlation between rising temperatures and the increased purine levels but failed to determine whether photoperiod changes modify the purine contents appreciably (Johnston and Eales, 1968). By observing the normal development and the process of regeneration of the anal fin of Brachydunio, Goodrich and Greene (1959) studied the mechanism of the formation of a color pattern. The main factor involved is a sequence of emergence of chromatophores in the order of melanophores, xanthophores, and erythrophores. Genetics of pigment cells in relation to color patterns has been reviewed by Gordon (1957). Based mainly on his observations on species of Xiphophorus, he also discussed the genetics of abnormal pigment cells and the etiology of melanomas (Gordon, 1959). The transplanted melanoma survives well in hybrid embryos of swordtail and platyfish, where normal cells do not grow well (Humm and Humm, 1959). The development and the growth of melanomas of Xiphophorus hybrids are enhanced by thiouracil and inhibited by thyroxine, suggesting that both hyperthyroidism and hypothyroidism affect tumor formation (Stolk, 1959). Extending the work of Gordon (1957, 1959), Atz ( 1962) investigated the genetics of both monomorphic and polymorphic pigmentary patterns in a number of species of Xiphophows and their hybrids. He also analyzed the effects of genes influencing melanosis and melanomas in these hybrids. The melanin-bearing cells constituting melanomas induced in Oryzias skin by trematode infection are responsive to a variety of pigment-aggregating stimuli ( Iga, 1965). It was concluded that the melanoma cells are under the control of melanin-aggregating nerves, like normal melanophores. The function of the pineal complex in fish is still unclear (see chapter by Fenwick, Volume V). It might function as an endocrine gland secreting a melanophore-aggregating hormone (cf. Section V, C). The light sensitivity of the diencephalon in blinded minnows was first suggested by von Frisch (1911). Using the cave-dwelling blind characin Anoptichthys, Breder and Rasquin (1947) concluded that the pineal determines the directional sense of the light reaction. Hoar (1955) also reported that blinded smolts of Oncorhynchus show a light reaction which is abolished when the pineal was destroyed. A recent electron microscope study of the pineal of the minnow showed that the organ contains a large number of sensory cells strongly resemble photoreceptors of the ciliary type ( Oksche and Kirschstein, 1967), suggesting the photoreceptive function of the gland. Using a few teleost species, Parry and Holliday (1960) suggested

344

RYOZO FUJI1

that the pseudobranch, which is the remnant of the first gill arch, may produce or activate a hormonal principle causing pigment aggregation in chromatophores. They also supposed that the entry of this hormone into the general circulation is regulated by the choroid gland in the eye. Pulsation, the rhythmic alternate aggregation and dispersion of pigment granules in chromatophores, is one of the most spectacular phenomena seen in isolated scales (Parker, 1948). The first description was given by Spaeth (1916a) who found that foIlowing the treatment of Fundulus scales with BaC1, this remarkable activity takes place. Ishibashi (1957) reported that microinjection of a Ca-precipitant saIt solution induces the melanophore pulsation in Oryzias. Following treatment with BaCl,, Watanabe ( 1 9 6 0 ~ )could not observe this activity in denervated melanophores of Carassius, suggesting the involvement of the rhythmic secretory activity of melanin-aggregating nerve endings. Observing the isolated scales immersed in a variety of solutions, Watanabe (1961) further emphasized that pulsation can be induced by any treatment that keeps the melanophores in intermediate state for a fairly long time. Thus, Ba and Ca precipitants are not necessarily specific agents capable of inducing this peculiar phenomenon. ACKNOWLEDGMENTS The author is deeply indebted to Dr. R. R. Novales for his invaluable advice during the preparation and the critical reading of the manuscript. Many thanks are also due Professor H. Kinosita, University of Tokyo, and Dr. N. Egami, Head of the Division of Biology, National Institute of Radiological Sciences, for their constant encouragement, REFERENCES Abramowitz, A. A. (1939). The pituitary control of chromatophores in the dogfish. Am. Naturalist 73, 208-218. Ando, S. (1960). Note on the type of mechanism of the colour change of the medaka, Oryzias Zatipes. Annototiones Zool. Japon. 33, 33-36. Ando, S. ( 1962). Responses of embryonic melanophores of the wild medaka ( Oryzias latipes) to various stimuli. EmbryoZogia (Nagoya) 7, 169-178. Arai, R., and Egami, N. ( 1961). Occurrence of leucophores on the caudal fin of the fish, Oryzias latipes, following administration of androgenic steroids. Annotationes Zool. Japon. 34, 185-192. Atz, J. W. (1962). Effects of hybridization on pigmentation in fishes of the genus Xiphophorus. Zoologica 47, 153-181. Bagnara, J. T. ( 1966). Cytology and cytophysiology of non-melanophore pigment cells. Intern. Reu. Cytol. 20, 173-205. Baker, B. I. (1963). Effect of adaptation to black and white backgrounds on the teleost pituitary. Nature 198, 404. Baker, B. I. (1968a). Unpublished data.

6.

CHROMATOPHORES AND PIGMENTS

345

Baker, B. I. (1968b). Unpublished data. Bellamy, D. (1966). On the lipochromes in the skin of marine teleost fish with special reference to the painted comber ( Serranus scriba L . ) . Comp. Biochem. Physiol. 17, 1137-1140. Bikle, D., Tilney, L. G., and Porter, K. R. (1966). Microtubules and pigment migration in the melanophores of Fundulus heteroclitus L. Protoplasm 61, 322-345. Boyles, M. (1969). Unpublished data. Brantner, G. ( 1956). Die Unabhangigkeit des morphologischen Farbwechsels vom physiologischen Farbwechsel bei der Entstehung des Hochzeitskleides des mannlichen Bitterlings. Z . Vergleich, Physiol. 38, 324-333. Breder, C. M., Jr., and Rasquin, P. (1947). Comparative studies in the light sensitivity of blind characins from a series of Mexican caves. Bull. Am. Museum Nut. Hist. 89, 319-352. Breder, C. M., Jr., and Rasquin, P. (1955). Further notes on the pigmentary behavior of Chaetodipterus in reference to background and water transparency. Zoologica 40, 85-90. Burgers, A. C. J. (1963). Melanophore stimulating hormones in vertebrates. Ann. N.Y. Acad. Sci. 100, 669-677. Burgers, A. C. J., Bennink, P. J. H., and van Oordt, G. J. (1963). Investigations into the regulation of the pigmentary system of the blind Mexican cave fish, Anoptichthys jorcluni. Koninkl. Ned. Akad. Wetenschap., Proc. CM, 189-195. Cerletti, A., and Berde, B. ( 1955). Die Wirkung von D-Lysergsaure-diathylamid ( LSD-25) und 5-Oxytryptamin auf die Chromatophoren von Poecilia reticuhtus. Experientia 11, 312-313. Chavin, W. ( 1956). Pituitary-adrenal control of melanization in xanthic goldfish, Carassius auratus L. 1. Exptl. Zool. 133, 1 3 6 . Chavin, W. (1963). Effects of hydroquinone and hypophysectomy upon the pigment cells of black goldfish. J. Pharmacol. Exptl. Therap. 142, 275-290. Chavin, W., and Schlesinger, W. (1967). Effects of melanin depigmental agents upon normal pigment cells, melanoma, and tyrosinase activity. In “The Pigmentary System” (W. Montagna and F. Hu, eds.), pp. 421445. Pergamon Press, Oxford. Chavin, W., Kim, K., and Tchen, T. T. ( 1963). Endocrine control of pigmentation. Ann. N.Y. Acad. Sci. 100, 67-85. Chen, Y. M., and Chavin, W. (1965). Radiometric assay of tyrosinase and theoretical considerations of melanin formation. Anal. Biochem. 13, 234-258. Chen, Y. M., and Chavin, W. (1966). Incorporation of carboxyl-groups into melanin by skin tyrosinase. Nature 210, 35-37. Chen, Y. M., and Chavin, W. (1967). Comparative biochemical aspects of integumental and tumor tyrosinase activity in vertebrate melanogenesis. In “The Pigmentary System” ( W . Montagna and F. Hu, eds.), pp. 253-268. Pergamon Press, Oxford. Coonfield, B. R. (1940). The pigment in the skin of Myxine glutinosa Linn. Trans. Am. Microscop. SOC.59, 398-403. Denton, E. J., and Land, M. F. (1967). Optical properties of the lamellae causing interference colours in animal reflectors. I . Physiol. ( London) 191, 23P-24P. Denton, E. J., and Nicol, J, A. C. (1966). A survey of reflectivity in silvery teleosts. J. Marine Biol. Assoc. U.K. 46, 685-722. Egami, N., Etoh, H., Tachi, C., Aoki, K., and Arai, R. (1962). Role of the pituitary gland in melanization in the skin of the goldfish, Carassius auratus, induced by X-ray irradiation. Proc. Japan Acad. 38, 345-347.

346

RYOZO FUJII

Enami, M. ( 1955). Melanophore-concentrating hormone (MCH) of possible hypothalamic origin in the catfish, Parasilurus. Science 121, 36-37. Etoh, H. ( 1961). Personal communication. Etoh, H. (1963). The effect of y-irradiation on the physiological responses of the nerve-melanophore system in Carassius carassius. 2001. Mag. (Tokyo) 72, 277-282. Etoh, H., and Egami, N. (1963). Responses to Na+, K’, atropine and adrenaline of melanophores induced by X-irradiation in the fin of the goldfish CUTUSS~W auratus. Anmtationes ZooE. Japon. 36, 133-139. Fain, W. B., and Hadley, M. E. (1966). In vitro response of melanophores of Fundulus heteroczitus to melatonin, adrenaline and noradrenaline. Am. Zoologist 6, 596. Falk, S,, and Rhodin, J. (1957). Mechanism of pigment migration within teleost melanophores. In “Electron Microscopy: Proc. Stockholm Conf.” ( F. S. Sjostrand and J. Rhodin, eds.), pp. 213-215. Academic Press, New York. Fiinge, R. (1962). Pharmacology of poikilothermic vertebrates. Phannacol. Reu. 14, 281 316. Fingerman, M. (1965). Chromatophores. Physiol. Rev. 45, 296-339. Fitzpatrick, T. B., Quevedo, W. C., Jr., Levene, A. L., McGovern, V. J., Mishima, Y., and Oettle, A. G. (1966). Terminology of vertebrate melanin-containing c e h : 1965. Science 152, 8-9. Fox, D. L. (1953). “Animal Biochromes and Structural Colowrs.” Cambridge Univ. Press, London and New York. Fox, D. L. ( 1957). The pigments of fishes. In “The Physiology of Fishes” (M. E. Brown, ed.), Vol. 2, pp. 367-385. Academic Press, New York. Fox, H. M., and Vevers, G. (1960). “The Nature of Animal Colours.” Sidgwick & Jackson, London. Franz, V. (1940). Struktur und Mechanisms der Melanophoren. Teil 11: Das Endoskelett. Z . Zellforsch. Mikroskop. Anat. 30, 194-234. Fries, E. F. B. (1943). Pituitary and nervous control of pigmentary effectors, especially xanthophores, in killifish ( Fundulus). Physiol. Zool. 16, 199-212. Fries, E. F. B. ( 1958). Iridescent white reflecting chromatophores (antaugophores, iridoleucophores) in certain teleost fishes, particularly in Bathygobius. J . Morphol. 103, 203-254. Fujii, R. (1959a). Mechanism of ionic action in the melanophore system of fish. I. Melanophore-concentrating action of potassium and some other ions. Annotationes Zool. Jupon. 32, 47-58. Fujii, R. (1959b). Mechanism of ionic action in the melanophore system of fish. 11. Melanophore-dispersing action of sodium ions. J. Fac. Sci., Univ. Tokyo, Sect. IV 8, 371-380. Fujii, R. ( 1960). The seat of atropine action in the melanophore-dispersing system of fish. J. Fac. Sci., Univ. Tokyo, Sect. IV 8, 643-657. Fujii, R. (1961). Demonstration of the adrenergic nature of transmission at the junction between melanophore-concentrating nerve and melanophore in bony fish. I. Fac. Sci., Univ. Tokyo, Sect. IV 9, 171-196. Fujii, R. (1962). Changes in the responsiveness of melanophore during embryonic development of Sulmo irideus. Zool. Mug. ( T o k y o ) 71, 187. Fujii, R. ( 1965a). Unpublished observations. Fujii, R. ( 1965b). Unpublished observations. Fujii, R. (1966a). Correlation between fine structure and activity in fish melanophore. In “Structure and Control of the Melanocyte” (G. Della Porta and 0. Miihlbock,

6.

CHROMATOPHORES AND PIGMENTS

347

eds.), pp. 114-123. Springer, Berlin. Fujii, R. (1966b). A functional interpretation of the fine structure in the melanophore of the guppy, Lebistes reticulotus. Annotationes 2001.Japon. 39, 185-192. Fujii, R. (1968). Fine structure of the collagenous lamella underlying the epidermis of the goby, Chmmichthys gulosus. Annotationes 2001.Japon. 41, 95-106. Fujii, R., and Fujii, Y. (1965). Mechanism of nervous control of fish melanophores. 11. Role of divalent ions in the transmission at the melanin-aggregating nerveendings. 2001.Mag. (Tokyo) 74, 351. Fujii, R., and Fujii, Y. (1966). An electron microscope study of the innervation of fish melanophore. Symp. SOC. Cellular Chem. 16, 87-100. Fujii, R., and Novales, R. R. (1968a). Tetrodotoxin: Effects on fish and frog melanophores. Science 160, 1123-1 124. Fujii, R., and Novales, R. R. (1968b). Melanin movements in fish melanophores in response to electrical stimulation of their controlling nerves. Proc. Intern. Union Physwl. Sci. 7, 145. Fujii, R., and Novales, R. R. (1969a). The nervous mechanism controlling pigment aggregation in Funddus melanophores. Comp. Biochem. Physiol. 29, 109-124. Fujii, R.,and Novales, R. R. (1969b). Cellular aspects of the control of physiological color changes in fishes. Am. Zoologist 9, 453-463. Fujii, R., and Novales, R. R. ( 1 9 6 9 ~ )Unpublished . data. Fujii, R., and Taguchi, S. (1969). The responses of fish melanophores to some melanin-aggregating and dispersing agents in potassium-rich medium. Annotationes 2001.Japon. 42, (in press). Goodrich, H. B., and Greene, J. M. (1959). An experimental analysis of the development of a color pattern in the fish Brachydunio albolineatus Blyth. J . Exptl. Zool. 141, 15-45. Goodrich, H. B., Hill, G. A,, and Arrick, M. S. (1941). The chemical identification of gene-controlled pigments in Platypoecilus and Xiphophorus and comparisons with other tropical fish. Genetics 26, 573-586. Gordon, M. (1957). Physiological genetics of fishes. In “The Physiology of Fishes” ( M . E. Brown, ed.), Vol. 2, pp. 431-501. Academic Press, New York. Gordon, M. (1959). The melanoma cell as an incompletely differentiated cell. In “Pigment Cell Biology” (M. Gordon, ed.), pp. 215-236. Academic Press, New York. Gray, E. G. ( 1956). Control of the melanophores of the minnow (Phoxinus phoxinus ( L. ) ). J . Exptl. Biol. 33, 44M59. Green, L. (1968). Mechanism of movements of granules in melanocytes of Fundulus heteroclitus. Proc. Natl. Acad. Sci. U S . 59, 1179-1186. Hama, T. (1963). The relation between the chromatopHores and pterin compounds. Ann. N.Y. Acad. Sci. 100, 977-986. Hama, T., and Hasegawa, H. (1967). Studies on the chromatophores of Oryzias Zatipes (teleostean fish) : Behavior of pteridine, fat and carotenoid during xanthophore differentiation in the color varieties. Proc. Japan Acad. 43, 901-906. Healey, E. G. ( 1940). Uber den Farbwechsel der Elritze (Phoxinus laeuis Ag.). Z. Vergleich. Physiol. 27, 545-586. Healey, E. G. ( 1954). The colour change of the minnow (Phoxinus lueuis Ag.). 11. Effects of spinal section between vertebrae 1 and 15 and of anterior autonomic section on the responses of the melanophores. J . Exptl. Biol. 31, 473490. Healey, E. G. (1967). Experimental evidence for the regeneration of nerve fibres controlling colour changes after anterior spinal section in the minnow (Phoxinus phoxinus L.). Proc. Roy. SOC. BIBS, 57-81.

348

RYOZO FUJI1

Healey, E. G., and Ross, D. M. (1966). The effects of drugs on the background colour response of the minnow Phoxinus phorinus L. Comp. Biochem. Physiol. 19, 545-580. Hill, A. V., Parkinson, J. L., and Solandt, D. Y. (1935). Photoelectric records of the colour change in Fundulus heteroclitus. J . Exptl. Biol. 12, 397-399. Hishida, T., Tomita, H., and Yamamoto, T. (1961). Melanin formation in color varieties of the medaka (Oryzias latipes). Embryologiu (Nagoya) 5, 335-346. Hitchings, G. H., and Falco, E. A. (1944). The identification of guanine in extracts of Cirella nigricans. The specificity of guanase. Proc. Natl. Acad. Sci. U S . 30, 294-297. Hoar, W. S. (1955). Phototactic and pigmentary responses of sockeye salmon smolts following injury to the pineal organ. J . Fisheries Res. Board Can. 12, 178-185. Hogben, L. T. (1936). The pigmentary effector system. VII. The chromatic function in elasmobranch fishes. Proc. Roy. SOC. B120, 142-158. Hogben, L. T., and Slome, D. (1931). The pigmentary effector system. VI. The dual character of endocrine coordination in amphibian colour change. Proc. Roy. SOC. B108, 10-53. Holmberg, K. (1968). Ultrastructure and response to background illumination of the melanophores of the Atlantic hagfish, Myxine glutinosa L. Gen. Comp. Endocrinol. 10, 421428. Ilu, F. (1963). Hormonal influence on goldfish pigment cells in uitro. In “Cinemicrography in Cell Biology” (G. G. Rose, ed.), pp. 339-356. Academic Press, New York. Hu, F., and Chavin, W. (1960). Hormonal stimulation of melanogenesis in tisslle culture. J. Invest. Dermatol. 34, 377-391. Humm, D. G., and Humm, J. H. (1959). The growth and metabolism of normal and abnormal pigment cells in fishes. In “Pigment Cell Biology” ( M . Gordon, ed.), pp. 197-214. Academic Press, New York. Iga, T. (1965). Studies on melanomas induced by trematode infection on the skin of fresh-water teleosts. I. Physiological properties of melanoma cells. Bull. Shimane Uniu., Nat. Sci. 14, 85-91. Imai, K. ( 1958). Extraction of melanophore concentrating hormone ( MCH) from the pituitary of fishes. Endocrinol. Japon. 5, 34-48. Ishibashi, T. (1957). Effect of the intracellular injection of inorganic salts on fish scale melanophore. J. Fac. Sci., Hokkaido Uniu., Ser. VI 13, 449-454. Ishibashi, T. (1962). Further studies on the action of potassium ion upon melanophores in an isolated fish scale, Mem. Fuc. Sci., Kyushu Uniu. E3, 137-142. Iwata, K. S., and Yamane, H. (1959). Response of fish scale melanophore to modification of ionic composition. Biol. 1. Okuyuma Uniu. 5, 1-11. Iwata, K. S., Watanabe, M., and Kurihara, T. (1859a). Changes in state and response of the fish scale melanophore during continuous immersion in Ringer’s solution. BioE. 1. Okuyama Uniu. 5, 185-194. Iwata, K. S., Watanabe, M., and Nagao, K. (195913). The mode of action of pigment concentrating agents on the melanophores in an isolated fish scale. Biol. J . Okayama Univ. 5, 195-206. Johnston, C. E., and Eales, J. G. (1967). Purines in the integument of the Atlantic salmon ( S d m o salar) during parr-smolt transformation. J . Fisheries Res. Board Can. 24, 955-964. Johnston, C. E., and Eales, J. G. (1968). Iduence of temperature and photoperiod

6.

CHROMATOPHORES AND PIGMENTS

349

on guanine and hypoxanthine levels in skin of Atlantic saImon (Salmo salar) during parr-smolt transformation. J . Fisheries Res. Board Can. 25, 1901-1909. Kajishima, T. ( 1960a). Analysis of gene action in the transparent-scaled goldfish, Carossius auratus. 11. The effects of pituitary and thyroid on gene action. Embryologia (Nagoya) 5, 127-138. Kajishima, T. ( 1960b). Analysis of gene action in the transparent-scaled goldfish, Carassius auratus. 111. Guanine metabolism in the developing larvae. Annotationes Zool. Japan. 33, 24S248. Kamada, T., and Kinosita, H. (1944). Movement of granules in fish melanophores. Proc. Japan Acad. 20, 484-492. Kawaguti, S . (1965). Electron microscopy on iridophores in the scale of the blue wrasse. PTOC.Japan Acad. 41, 610-613. Kawaguti, S., and Kamishima, Y. (1966a). A supplementary note on the iridophore of the Japanese porgy. Biol. J . Okayama Uniu. 12, 57-60. Kawaguti, S., and Kamishima, Y. (196613). Electron microscopy on the blue back of a clupeoid fish, Harengula zunasi. Proc. Japan. Acad. 42, 389-393. Kent, A. K. ( 1959). Distribution of melanophore-aggregating hormone in the pituitary of the minnow. Nature 183, 544-545. Kent, A. K. (1960). Distribution of chromactivating hormones in the pituitary of the minnow. Nature 186, 395396. Kent, A. K. (1961). The influence of extraction in sodium hydroxide on the activity of the color change factors of the teleost pituitary. Gen. Comp. Endocrinol. 1, 409-415. Kim, K., Tchen, T. T., and Hu, F. (1961). Studies on ACTH-induced melanocyte formation: Effect of colchicine. Exptl. Cell Res. 25, 454-457. Kim, K., Tchen, T. T., and Chavin, W. ( 1962). Tyrosinase of the goldfish Carassius auratus L. 11. Correlation of tyrosinase activity with pigmentation. Biochim. Biophys. Acta 59, 577480. Kinosita, H. (1953). Studies on the mechanism of pigment migration within fish meIanophores with special reference to their electric potentials. Annotationes 2001.Japon. 26, 115-127. Kinosita, H. (1963). Electrophoretic theory of pigment migration within fish melanophore. Ann. N.Y. Acad. Sci. 100, 992-1003. Kohler, V. (1952). Die Wirkung des Adrenocorticotropins auf die Lipophoren der Pfriue ( Phoxinus laevis) . Naturwissenschaften 39, 554. Kosto, B., Pickford, G . E., and Foster, M. (1959). Further studies of the hormonal induction of melanogenesis in the killifish, Fundulus heteroclitus. Endodnology 65, 869-881. Laming, W. J. R. (1954).The occurrence of a waterbalance-, a melanophoreexpanding and an oxytocic principle in the pituitary gland of the river-lamprey (Lampetra fluviatilis L,). Acta Endocrinol. 16, 277-284. Lerner, A. B., and Case, J. D. (1960). Melatonin. Federation PTOC.19, 590-592. Loud, A. V., and Mishima, Y. (1963). The induction of melanization in goldfish scales with ACTH, in uitro. Cellular and subcellular changes. J. Cell Biol. 18, 181-194. Lundstrom, H. M., and Bard, P. (1932). Hypophysial control of cutaneous pigmentation in an elasmobranch fish. Biol. Bull. 62, 1-9. Markert, J. R., and Vanstone, W. E. (1966). Pigments in the belly skin of coho salmon, Oncorhynchzrs kisutch. J . Fisheries Res. Board Can. 23, 1095-1098.

350

RYOZO FUJII

Marsland, D. A. (1944). Mechanism of pigment displacement in unicellular chromatophores. Biol. Bull. 87, 252-261. Marsland, D. A., and Meisner, D. (1967). Effects of D10 on the mechanism of pigment dispersal in the melanocytes of Fundulus heteroclitus. A pressuretemperature analysis. J . Cellulur Physiol. 70, 209-216. Matsumoto, J. (1965a). Studies on fine structure and cytochemical properties of erythrophores in swordtail, Xiphophorus helleri, with special reference to their pigment granules (pterinosomes). J. Cell Biol. 27, 493-504. Matsumoto, J. (1965b). Role of pteridines in the pigmentation of chromatophores in cyprinid fish. Japan. J . 2001.14, 45-94. Mellinger, J. ( 1963). Etude histophysiologique du systdme hypothalamo-hypophysaire de Scyliorhinus caniculus ( L. ) en &tat de mblanodispersion permanente. Gen. Comp. Endocrinol. 3, 2 6 4 5 . Miyoshi, S. ( 1952). Responses of iridocytes in isolated scale of the medaka (Oryzias latipes) to chlorides. Annotationes 2001.lapon. 25, 21-29. Mori, W. ( 1961). Personal communication. Nagahama, H. (1953). Action of potassium ions on the melanophores in an isolated fish scale. Japan. J . Zool. 11, 75-85. Neckel, I. ( 1954). Chemische und Elektronenmikroskopische Untersuchungen uber das Guanin in der Haut einiger Fische. Mikroskopie 9, 113-119. Newth, D. R. ( 1951). Experiments on the neural crest of the lamprey embryo. J . Erptl. Bid. 28, 247-260. Novales, R. R. (1959). The effects of osmotic pressure and sodium concentration on the response of melanophores to intermedin. Physiol. 2001.32, 15-28. Novales, R. R. ( 1967). Melanocyte-stimulating hormone and the intermediate lobe of the pituitary: Chemistry, effects, and mode of action. In “Neuroendocrinology” (L. Martini and W. F. Ganong, eds.), Vol. 2, pp. 241-259, Academic Press, New York. Novales, R. R. (1968). Personal communication. Novales, R. R., and Gratzer, B. M. (1960). Effect of sodium concentration on the response of goldfish scale melanophores to MSH. Anat. Record 137, 385. Novales, R. R., and Novales, B. J. (1966a). Electron microscopic studies of pigment movements in melanophores. Am. Zoologist 6,576. Novales, R. R., and Novales, B. J. (1966b). Factors influencing the response of isolated dogfish skin melanophores to melanocyte-stimulating hormone. Biol. Bull. 131, 470-478. Odiorne, J. M. (1933). The occurrence of guanophores in Fundulus. Proc. Natl. Acad. Sci. US. 19, 750-754. Odiorne, J. M. (1948). Morphological color changes in vertebrates. In “Biology of Melanomas” (M. Gordon, ed.), Spec. Publ., pp. 288-308. N.Y. Acad. Sci., New York. Odiorne, J. M. (1957). Color changes. In “The Physiology of Fishes” (M. E. Brown, ed.), Vol. 2, pp. 387401. Academic Press, New York. Oksche, A., and Kirschstein, H. (1967). Die Ultrastruktur der Sinnezellen in Pinealorgan von Phoxinus laevis L. 2. Zellforsch. Mikroskop. Anat. 78, 151-166. Orton, G. L. (1953). Development and migration of pigment cells in some teleost fishes. J. Morphol. 93, 69-100. Ota, F. (1954). Chemical studies on lustrous particles in epidermis of the hairtail, Trichirus muumelu. 11. On the presence of uric acid in its raw particles. Bull. Japan. SOC. Sci. Fisheries 19, 106S1069.

6.

CHROMATOPHORES AND PIGMENTS

351

Parker, G. H. (1935). The chromatophoral neurohumors of the dogfish. J. Gen. Physiol. 18, 837-846. Parker, G. H. (1948). “Animal Colour Changes and Their Neurohumours.” Cambridge Univ. Press, London and New York. Parker, G. H., and Rosenblueth, A. (1941). The electric stimulation of the concentrating (adrenergic ) and the dispersing ( cholinergic) nerve-fibres of the melanophores in the catfish. Proc. Natl. Acad. Sci. U.S. 27, 1 9 s 2 0 4 . Parker, G. H., Welsh, J. H., and Hyde, J. E. (1945). The amount of acetylcholine in the dark skin and in the pale skin of the catfish. Proc. Natl. Acad. Sci. U.S. 31, 1-8. Parry, G., and Holliday, F. G. T. (1960). An experimental analysis of the function of the pseudobranch in teleosts. J. Exptl. Biol. 37, 344-354. Pickford, G. E., and Atz, J. W. (1957). “The Physiology of the Pituitary Gland of Fishes.” N.Y. Zool. SOC.,New York. Pickford, G. E., and Kosto, B. (1957). Hormonal induction of melanogenesis in hypophysectomized killifish ( Fundulus heteroclitus ) . Endocrinology 61, 177-196. Pye, J. D. (1964a). Nervous control of chromatophores in teleost fishes. I. Electrical stimulation in the minnow (Phoxinus phoxinus ( L , ) ) , J . Exptl. Biol. 41, 525-534. Pye, J. D. (1964b). Nervous control of chromatophores in teleost fishes. 111. Local temperature ,responses in the minnow (Phoxinus phoxinus (L.)). 1. Exptl. Biol. 41, 543-551. Rasquin, P. (1958). Studies in the control of pigment cells and light reactions in recent teleost fishes. Bull. Am. Museum Nut. Hist. 115, 1-68. Rebhun, L. I. (1967). Structural aspects of saltatory particle movement. J. Gen. Physiol. 510, 223-239. Reidinger, L. ( 1952). Wber den morphologischen und physiologischen Farbwechsel der Elritze. 2. Vergleich. Physiol. 34, 394-406. Reidinger, L., and Umrath, K. ( 1952). Die parasympathikolytische und parasympathikomimetische Wirking des Atropins auf die Chromatophoren. Z. Vergleich. Physiol. 34, 472-478. Robertson, 0. H. (1951). Factors influencing the state of dispersion of dermal melanophores in rainbow trout. Physiol. Zool. 24, 309-323, Scheline, R. R. ( 1963). Adrenergic mechanisms in fish: Chromatophore pigment concentration in cuckoo wrasse, Labrus ossifagus L. Cornp. Biochern. Physiol. 9, 215-227. Scott, G. T. ( 1965). Physiology and pharmacology of color change in the sand flounder Scopthulmus aquosus. Limnol. Oceanog. 10, R236R246. Shephard, D. C. (1961). A cytological study of the origin of melanophores in the teleosts. Biol. Bull. 120, 20&220. Smith, D. C. (1936). A method for recording chromatophore pulsations in isolated fish scales by means of a photo-electric cell. J. Cellular Cornp. Physiol. 8, 83-87. Spaeth, R. A. (1913). The physiology of the chromatophores of fishes. J . Exptl. Zool. 15, 527585. Spaeth, R. A. (1Q16a). Evidence proving the melanophore to be disguised type of smooth muscle cell. 1. Exptl. Zool. 20, 193-215. Spaeth, R. A. (1916b). A device for recording the physiological responses of single melanophores. Am. J. Physiol. 41, 597-602. Spitz, L. M., and Burnett, J. B. (1968). The tyrosinases of Fundulus heteroclitus at different stages of embryonic development. J. Embryol. Exptl. Morphol. 19, 1-8. Stahl, A. (1958). Sur la prksence d‘une dualit6 cellulaire a u niveau du lobe inter-

352

RYOZO FUJI1

mediaire de l’hypophyse de certains poissons. Compt. Rend. SOC. Biol. 152, 1562-1565. Stolk, A. (1959). Effect of thiouracil and thyroxine on development and growth of cutaneous melanoma in killifish hybrids. Nature 184, 562-563. Sumner, F. B., and Fox, D. L. (1935). Studies of carotenoid pigments in fishes. 111. The effects of ingested carotenoids upon the xanthophyll content of Fundulus paroipinnis. Proc. Natl. Acad. Sci. U.S. 21, 330440. Takeuchi, K. (1960). The behavior of carotenoid and distribution of xanthophores during development of the medaka (Oyzias latipes). Embryologia (Nagoya) 5, 170-176. Tchen, T. T., Amnieraal, R. N., Kim, K., Wilson, C. M., Chavin, W., and Hu, F. ( 1964). Studies on the hormone-induced differentiation of melanoblasts in explants from xanthic goldfish tailfin. Natl. Cancer Inst. Monograph 13, 67-80. Tomita, H., and Hishida, T. (1961a). A quantitative study on phenol oxidase of skin in color varieties of the medaka (Oryzias Zatipes). Embryologia (Nagoya) 5, 347456. Tomita, H., and Hishida, T. (1961b). On the phenol oxidase of embryonic and larval stages of the medaka (Oryzias latipes). Embryologia (Nagoya) 5, 423439. Turner, W. J., and Carl, A. (1955). Effect of reserpine on the melanophores of fish. Science 121, 877-878. Ueda, K. ( 1955). Stimulation experiment on fish melanophores. Annotationes Zool. Japon. 28, 194-205. Umrath, K. ( 1957). Wber den physiologischen und den morphologischen Farbwechsel des Bitterlings, Rhodeus amarus. Z. Vergleich. Physiol. 40, 321-328. Umrath, K. (1959). Wber den Einfluss des Adrenocorticotropen Hormons auf die Farbung und iiber die Auslosbarkeit des Hochzeitskleides bei Einigen Fischen. Z. Verguch. Physiol. 42, 181-191. Umrath, K., and Walcher, H. ( 1951). Farbwechselversuche an Macropodus opercularis und ein Vergleich der Geschwindigkeit der Farbanderung bei Macropoden und Elritzen. Z. Vergleich. Physiol. 33, 129-141. von Euler, U. S., and Fange, R. (1961). Catecholamines in nerves and organs of Myxine glutinosa, Squalus acanthias, and Gadus callarias. Gen. Comp. Endocrinol. 1, 191-194. von Frisch, K. (1911). Beitrage zur Physiologie der Pigmentzellen in der Fischhaut. Arch. Ges. Physiol. 138, 319387. von Gelei, G. (1942). Zur Frage der Doppelinnervation der Chromatophoren. Z. VeTgleich. Physiol. 29, 532-540. Waring, H. (1960). The effect of pituitary extracts on melanophores in isolated elasmobranch skin. Australian J. Exptl. Biol. Med. Sci. 38, 187-194. Waring, H. ( 1963 ). “Color Change Mechanisms of Cold-Blooded Vertebrates.” Academic Press, New York. Watanabe, M. (1960a). Effect of potassium cyanide on the melanophores in the isolated scale of crucian carp. Biol. J. Okoyama Uniu. 6, 103-113. Watanabe, M. (1960b). The mode of action of atropin on melanophores in the isolated scale of a crucian carp. Biol. J. Okayama Uniu. 6, 114-123. Watanabe, M. ( 1 9 6 0 ~ )The . relation of nerves to pulsations of melanophores in the isolated scale of crucian carp. Biol. J. Okayama Uniu. 6, 19-26.

6.

CHROMATOPHORES AND PIGMENTS

353

Watanabe, M. (1961). Pulsations of the scale melanophore in the crucian carp. Biol. J . Okayamu Univ. 7, 65-85. Watanabe, M., and Mori, Y. (1964). Action of ammonium chloride to the scale melanophore of crucian carp. Biol. J . Okuyuma Uniu. 10, 65-74. Watanabe, M., Kobayashi, M., and Iwata, K. S. (1962a). The action of certain autonomic drugs of the fish melanophore. Biol. J. Okayama Univ. 8, 103-114. Watanabe, M., Izumi, I., and Iwata, K. S. (1962b). The action of adrenaline on the melanophore of Olyzias, with special reference to its pigment dispersing action. Biol. J . Okayama Uniu. 8, 95-102. Watanabe, M., Naitoh, T., and Tsuchiya, K. (1965). Reversal of action of electrolytes to the melanophore of crucian carp. Biol. J. Okuyuma Uniu. 11, 19-30. Weisel, G. F. (1950). The comparative effects of teleost and beef pituitaries on chromatophores of cold-blooded vertebrates. B i d . Bull. 99, 487496. Wikswo, M. A., and Novales, R. R. (1969). Biol. Bull. (in press). Woodhead, A. D. (1966). Effects of thyroid drugs on the larvae of the brown trout, Salmo trutta. I . Zool. (London) 149, 394-413. Young, J. 2. (1935). The photoreceptors of lampreys. 11. The functions of the pineal complex. J. Ezptl. Biol. 12, 254-270. Ziegler, 1. (1965). Pterine als Wirkstoffe und Pigmente. Ergeb. Physiol., Biol. Chem. Exptl. Pharmakol. 56, 1-66. Ziegler-Giinder, I. ( 1956). Untersuchungen iiber die Purin- and Pterinpigmente in der Haut und in den Augen der Weissfische. Z . VergZeich. Physiol. 39, 163-189.

This Page Intentionally Left Blank

BIOLUMINESCENCE ..lA . C. NZCOL I . Introduction I1. Occurrence . I11. Light Organs

. . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

A Location . . . . . . . B Anatomy . . . . . . . C Fine Structure . . . . . . IV . Biochemistry . . . . . . . V. Regulation of Light Emission . . . . A . Direct Control . . . . . . B . Indirect Control . . . . . . VI . Physical Characteristics . . . . . VII . Significance and Employment of Luminescence A Theories of Luminescence . . . . B . Elaboration and Criticism of Theories . VIII Conclusions and Summary . . . . References . . . . . . . . .

.

.

. .

. .

. .

. .

. . .

. . .

. . .

. . .

. .

. .

. .

. .

.

.

.

.

. . . . . .

. . . . . .

. . . . . .

. . . . . .

.

.

.

.

355 356 357 357 360 369 375 379 380 382 384 388 388 388 393 394

.

I INTRODUCTION

The light of luminescent fishes is produced in well-defined and discrete light organs. distributed in positions and patterns characteristic of each species. The production of light is luminescence. in particular some form of chemiluminescence. It is known to occur only in two groups of fishes. namely. the Elasmobranchii and the Actinopterygii (Teleostei) . To prepare the way for the discussion that follows we may distinguish three modes of light production among fishes. namely. light generated 355

356

J. A. C. NICOL

by the animal within its own tissues ( intracellular luminescence), discharge of luminous secretion ( extracellular luminescence), and light generated by symbiotic bacteria ( bacteria1 luminescence). The literature dealing with the light organs and luminescence of fishes is vast, although very few of the accounts can be regarded as physiology, s e w stricto. Something more than passing reference to structure seems demanded, however, when dealing with the function, performance, and capabilities of light organs, because few readers are well conversant with such an esoteric field. Furthermore, we are still heavily dependent upon the inferences and deductions that can be drawn from structural appearances for some comprehension of the functioning of these organs. Fortunately, thanks to exhaustive accounts by our predecessors and to recent reviews, all the earlier literature dealing with the luminescence of fishes has been collected and discussed, and the bibliographies are exhaustive and practically complete, The late Professor E. N. Harvey published a historial survey (1957a) that terminates in 1900, and a monograph on bioluminescence ( 1952) that deals with modern scientific investigations. In 1957, he summarized the then known features of the luminous organs of fishes (Harvey, 1957b). Recent reviews of luminescence include sections on fishes (Nicol, 1960c, 1962a; Boden and Kampa, 1964; Hastings, 1966; Johnson, 1967). The proceedings of a grand conference on bioluminescence are also available (Johnson and Haneda, 1966). The account that follows considers the luminescence of fishes as a discrete and integral field of physiological inquiry, the scattered strands are drawn together, the gaps are underlined, and more than passing reference is made to the ways that luminescence may be of service to the animal.

11. OCCURRENCE

Luminescence in fishes, so far as it is known, is confined exclusively to marine species. No satisfactory explanation has been advanced to account for the difference in this regard between saltwater and freshwater species. Luminescence is common among oceanic fishes, especially of mesopelagic waters. Some two-thirds of the species of fishes found in this region are luminous (Harvey, 1952). The prevalence of luminescence in oceanic fishes, its absence in lacustrine fishes, may be linked to the abundance and variety of oceanic species, and to the extensive and regular vertical migrations which they carry out.

7 . BIOLUMINESCENCE

357

111. LIGHT ORGANS

A. Location Light organs occur in a bewildering variety of locations: in the skin, a common location, in the ventral musculature (Acropoma), or within the abdomen (Apogon). They are distributed over the entire surface of the body of Idiacanthus, or scattered in large numbers over the lower surface of the body in Etmopterus. In many Isospondyli and Iniomi (Fig. 1) the photophores or light organs are regularly arranged in rows on the lower surface of the head and trunk, e.g., Cyclothone. Photophores lie beneath the eyes of Astronesthes, on the nose of Diaphus, the tongue

Fig. 1. Outlines of three species of lantern fishes occurring in the same geographical area. They differ in the patterns of light organs. ( A ) Larnpnnyctus kucopsatus (male), ( B ) Diuphus rafinesquii (= D.theta), and ( C ) Tarletonbeania crenuhris (female). From Bolin ( 1939).

358

J. A. C. NICOL

Fig. 2. Deep-sea angler fish, Linophryne breoibarbis Pam. Detailed view of the illicium and lateral view of a female fish, Drawing by W. S. Bronson. From Parr (1927).

of Neoscopelus, within the oropharynageal cavities of Chauliodus, on the caudal peduncle of Myctophum (Fig. l ) , and the tail of Saccopharynx. Luminous bulbs occur on tentacles (Fig. 2 ) ) or barbels-on the dorsal fin ray (illicium) of C e r a t h , and the chin barbel of S t o m h . There are luminous glands on the lower jaw of hlonocentris (Fig. 3 ) , in the ventral

Fig. 3. Longitudinal section through the mandibular photogenic organ of the knight-fish, Monocentris faponicus. X22. cap, capillary; chr, chromatophores; ct, connective tissue; dp, dermal papilla; dt, duct; m, muscles; op, gland opening; os, submaxillary bones; rf, reservoir; se. t., secretory tubules; and tr, trabeculae. After Okada (1926).

7.

BIOLUMINESCENCE

359

abdominal wall of Malacocephalus, and in a ring surrounding the esophagus of Gazza (Fig. 4 ) . Some species have but one or two organs, e.g., Melunocetus, whereas others have thousands, e.g., Chauliodus (Haneda, 1951; Harvey, 1952; Tchernavin, 1953; Iwai, 196Oa; Nicol, 1960b). The positions of luminous organs in fishes have been tabulated by McAllister ( 1967). In the great majority of fishes luminescence is intracellular. Others have light organs harboring luminous symbiotic bacteria [ Leiognathidae

Fig. 4. Transverse section through the luminous organ of Leiognathus equulus. It surrounds the esophagus, and its glandular palisades contain bacteria. d, duct; gl, glandlike region; E, esophagus; and p, pigment. After Harms (1928).

360

J. A. C. NICOL

( Fig. 4), Acropomatidae and Monocentridae ( Fig. 3), and Gadidae]. Several species ( Searsia and Ceratias) discharge a luminous secretion; these fishes also have simple light organs concerned with intracellular and bacterial luminescence, respectively. The light organs of fishes, being so diverse in appearance and structure, and occurring in so many different positions, bewilder by their variety. Therefore, it is useful to have some means of classifying them, even if it be completely arbitrary, to force order on chaos. At the present time a classification based upon structure is more satisfactory than any other, because we know a great deal about the microanatomy of the light organs of fishes, and very little about their functioning and usefulness. An anatomical classification has been adopted in the following account (vide von Lendenfeld, 1887; Brauer, 1908; Nicol, 1967).

B. Anatomy Categories recognized are simple organs and complex organs, the latter comprehending integumental photophores, alveolar organs and sacs in the body wall, and visceral organs. 1. SIMPLEORGANS

Simple organs consist of small masses of photocytes, with or without an interrupted mantle of pigment. Simple acquisitions shift them into the next category ( compound organs). All photophores of selachians are of this type; they occur only in bathybenthopelagic squaloid sharks. The photophores are small ( 0.10.3 mm in width) and very numerous. Each has a small accumulation of light-generating cells-the photocytes-arranged in the manner shown in Fig. 5. The distal regions of the cells contain acidophilic granular material, a feature common to photocytes of many animals. A single layer of melanophores invests the photophore; the tendency of the pigmented cell processes to restrict the outer face of the photophore has been repeatedly noticed, and the arrangement has been likened to an iris diaphragm. A few small cells containing a homogeneous secretion are situated external to the photocytes; this group of cells has been likened to a lens because of the histological resemblance to the cells of the lens of the eye. Groupings of similar cells occur in the complex photophores of many teleost fishes and in conformity with usage they will be called “lenses” or “lenticular cells” ( Harvey, 1952; Iwai, 1960a). Simple light organs of stomiatoid fishes, e.g., Stomias and Lamprotoxus, lie superficially in the gelatinous corium. Such an organ consists

7 . BIOLUMINESCENCE

361

Fig. 5. Section of the luminous organ of a shark Spinax (= Etmopterus) niger, showing black pigment cells, pg; luminous cell, 1um.c.; luminous secretion, lum. sec.; basement membrane, b.m.; and lens cells, 1. After Dahlgren and Kepner.

of a small packet of photocytes, sometimes provided with a pigmented mantle but having no other accessory structure (Fig. 6). Tchernavin (1953) estimated that there are several thousand of them in Chauliodus, on all parts of the animaI, and in concerted action they must cause the whole body of the fish to glow. Beebe (1935) observed luminous Chaulwdus at a depth of 500 m, and the body of the fish glowed from a multitude of small lights. Other simple light organs are those occurring on the chin barbels of stomiatoid fishes, on the illicia or fishing tentacles of ceratioids, and the caudal glands of myctophids and the dermal bars of searsiids (Brauer, 1906, 1908; Beebe and Crane, 1939; Anadbn, 1957; Iwai and Okamura, 1960).

362 J- A. C. NICOL

LI

E

m

cd

c

7.

BIOLUMINESCENCE

363

Early suggestions that the escae of ceratioids owe their light to symbiotic bacteria receive support from investigations by electron microscopy. The luminous bulb is surrounded by a reflecting and pigmented layer, except for a distal area, free of pigment cells, which allows light to escape. There are radiating tubules largely filIing the bulb: tubules which contain two kinds of glandular cells. Microorganisms resembling bacteria occur within the glandular cells and in the central cavity, and the center of the bulb communicates to the exterior by a pore (Bassot, 1966; Hulet and Musil, 1968). 2. COMPOUND ORGANS

Compound organs contain, besides the photogenic tissue and pigment mantle, some putative dioptric accessories, lenses and reflectors. Consider the subocular light organs of Astronesthes as an example (Fig. 7 ) . The organ is rather large (about 1 mm in diameter), lying in deep dermal tissue, and separated from the superficial epidermis and dermis by a cell-free space. It contains a dense mass of photocytes arranged in cords and bands. The inner surface of the organ is invested by a thick reflector and a black pigmented mantle. Nerves and blood vessels penetrate the pigment and reffector layers to invade the photogenic tisue, and there is a clear area in the skin over the photophore. A very peculiar mechanism for rotating and concealing the light organ exists. Passing down behind the photophore is a band of muscle, which extends below it and continues upward to become inserted on its external face. Also, there is a black pigmented pocket in the dermis below the light organ. The muscle is so disposed that on contracting it pulls the outer face of the photophore downward, causing its bright surface to be concealed; at the same time some part of the pigmented upper wall of the photophore is moved in front of the patent window (Brauer, 1908; Tchernavin, 1953; Nicol, 1960b). The luminescence of the cheek organs of stomiatoids has been repeatedly observed (Harvey, 1931; Beebe and Crane, 1939; Haneda, 1955), and it is certainly intracellular. The longitudinally arranged photophores found in so many fishesround-mouths, stomiatoids, hatchetfish, etc.-have, in addition to the components just described, lenslike structures, cylindrical discs, and filters on their outer surfaces. There is much variation in the detailed anatomy of these organs brought out in accounts by Brauer (1904, 1908), Nussbaum-Hilarowicz ( 1923), Haneda ( 1952b), and Bassot ( 1966) ( see Section 111, C ) . The dermal photophores (that of Stomias is shown in Fig. 8) have a hemispherical reflector and a pigmented sheath. TWO

384

J. A. C. NICOL

W

P.t

M

Fig. 7. Transverse section through a subocular light organ of Astronesthes richardsoni. M, muscle; P.b., pigment backing; p.t., photogenic tissue; R, reflector; v.pm., ventral pigment; v.pk., ventral pocket; and W, clear window.

kinds of glandular cells ( A and B ) are distinguishable in the photogenic central tissue, and there is a capping plate or lens on the outer surface. Light organs of myctophids generally are shallow cups lying in the corium. There is the usual backing of pigment and a reflector containing fine platelets. The photocytes, however, are generally restricted to a limited area. They are usually elongate, spindle-shaped, and dorsoventrally arranged; they have been likened to category A photocytes of stomiatoids (Nicol, 1958; Iwai and Okamura, 1960; Kier, 1967).

3. ALVEOLARORGANS AND SACSIN

THE

BODYWALL

These organs occur sporadically among teleost fishes. They are glandular sacs embedded in the dermis, and like compound glands they

7.

BIOLUMINESCENCE

365

Fig. 8. Section through the luminous organ of a mesopelagic fish Gonostoma elongaturn. The organ, lying within the body wall of the lower trunk, is organized to direct its light downward. d, skin; ex, exterior; 1, lens; ph.c., photocyte; PA, pigment sheath; r, reflector; and sc, scale.

are subdivided internally into chambers, alveoli, and tubules. They usually have one or more openings to the exterior; some owe their luminescent ability to symbiotic bacteria. Photoblepharon and Anomalops are the oft-described exemplars of this type. The large light organ, lying below the eye, contains a mass of

366

J. A. C . NICOL

secretory tubules; internally there is a reflector and a layer of pigment. Several kinds of glandular cells can be distinguished, and bacteria are abundant in the tubules. The light organ faces a clear window in the skin through which it shines. That of Photoblephuron permits itself to be occluded by a pigmented screen which can be drawn up over its external surface, whereas that of Anornulops is concealed by rotation, the light organ being turned downward until its light-emitting surface becomes concealed within a dark ventral pocket (Harvey, 1952; Haneda, 1953; Bassot, 1966). The reader may notice the essential similarity of the rotatable subocular organs of Anomlops and those of stomiatoids to each other. The latter is described in Section 11, B, 2. The light organs of Monocentris (Monocentridae) are luminous discs situated beneath the mandible (Fig. 3 ) . Within them are tubules and canals lined with glandular epithelium. The organ is invested by a sheath of connective tissue containing chromatophores, but the pigment is less dense in the skin lying over the light organ. Light is produced by luminescent bacteria occurring within the tubules of the organ (Okada, 1928; Haneda, 1968). In Ckidopus (of the same family), the light organs lie on the sides of the mandibles. Light production is continuous, but when the fish closes its mouth, the upper jaws cover the luminous organs and the light fs no longer seen (Haneda, 1968). Some luminous fishes of the families Trachichthyidae, Macrouridae, and Gadidae have a light organ in the form of a glandular bulb lying within the ventral body wall. The bulb lies close to the anus and it is connected to the exterior, sometimes to the rectum, by a long duct. Light is produced by luminous bacteria. The skin over the organ is transparent, and structures regarded as lens and reflector are sometimes present. Luminescence has been observed over the whole lower surface of Paratruchichthys and Physiculus, the result, apparently, of light diffusing through translucent ventral muscle masses ( Haneda, 1951, 1952a, 1957).

ORGANS 4. VISCERAL Luminous organs sometimes are associated with the viscera of fishes, and arrangements of great variety exist. They are glandular structures connected to some part of the alimentary canal: the esophagus in Leiognuthus (Fig. 4 ) , the intestine in Apogon, and the pyloric cecae in Purupriucunthus. In some organs symbiotic bacteria are present, but this is not always so (Kato, 1947; Haneda, 1950; Iwai and Asano, 1958; Haneda and Johnson, 1982). Haneda (1950) classifies these organs as indirect emitting systems because they make use of extensive reflectors and large translucent

7 . BIOLUMINESCENCE

367

bodies, far exceeding the small photogenic gland in size, that cause the light to be spread, scattered, and emitted as a diffuse glow over a large area. In Acropoma, for example, the luminous gland is a U-shaped tube embedded in the ventral body wall. Above it is a white opaque reflector (an intermuscular septum) that separates the lateral from the ventral muscles of the body wall; the ventral muscles are translucent. Light from the gland is reflected, it is scattered within the muscles, and it appears as a diffuse glow over the lower surface of the fish (Haneda, 1950). Again, in Gazza (Fig. 4) the visceral light organ encircles the esophagus, It contains a layer of tubular glands invested by a reflector and pigment. Part of the light escapes through a ventral aperture, and part passes dorsally to be reflected by the silvery coverings of the swim bladder and peritoneum. In either event the light is dispersed by translucent ventral muscles and is emitted over the whole lower surface

Fig. 9. Transverse section through the anterior abdominal region of Parapriacanthus berycifomes. lo, region of light organ; m', muscle of lateral body wall;

m*,

muscle of lower abdominal wall; g, stomach; pc, pyloric cecum; pg, pigment layer; rf, reflector; and cc, coelomic cavity. After Haneda and Johnson (1962).

368

J. A. C. NICOL

of the fish (Haneda, 1950; Iwai, 1958, 1960b; Haneda and Johnson, 1962; Bassot, 1966). Parapriacanthus has two visceral light organs, viz., a thoracic duct and a posterior anal duct. The former opens into pyloric ceca; the latter, to the exterior by a pore near the anus. The thoracic organ lies above a pigmented screen which is situated between the light organ and the

Fig. 10. Anatomy of the light organ and adjacent structures of Opisthoproctzrs soleatus. Both illustrations show the rectal bulb and the posterior part of the reflector, above as seen in a dissection, below in a median section. Light is generated in epithelial folds (11, lu) of the rectal bulb (rb) and is transmitted through a lens (1) to a hyaloid body (h). The latter has a reflecting sheath and light escapes through the ventral body wall. Among other structures shown are the intestine, i; anal opening, an; pigmented tissue, pc; and window, w, in front of the lens. v is the blood supply. From Bertelsen and Munk (1964).

7 . BIOLUMINESCENCE

369

ventral surface of the fish (Fig. 9 ) . Above and on either side are reflectors, seemingly in the intermuscular septa. The arrangement is such as to cause light to be spread outward and downward through the translucent tissues of the lower trunk (Haneda and Johnson, 1962; Haneda et al., 1966). A similar glandular organ of Steindachneria encircles the anus. The ventral muscle mass is translucent and the intermuscular septum delimiting it above is silvery. Actual areas observed to luminesce are the lower and ventral surfaces of the head, throat, trunk, and the sides of the head (behind the eyes). Presumably, light from the small gland is conducted and diffused through the ventral muscle mass and is reflected downward ( Cohen, 1964). Deep-sea fishes of the family Opisthoproctidae have a visceral light organ derived from the rectum. It is a simple glandular structure in Rhyncholagus, emitting light directly to the exterior, and more complicated in Opisthoproctus, emitting light indirectly (Fig. 10). In the latter fish the photogenic part is the rectal bulb lying above and behind the rectum; in front of the bulb is a lenticular body. This region is optically isolated from the exterior on all sides except anteriorly by densely pigmented tissue. In front of the lenticular body is a columnar channel which extends anteriorly through the ventral wall of the fish, and above the channel is a reflector. Light originating in the rectal bulb passes through the lenticular body into the columnar channel and escapes over a large ventral area. The luminescence of living fish has not been observed ( Bertelsen and Munk, 1954; Bertelsen et al., 1965).

C. Fine Structure 1. GLANDULAR CELLS From histochemical and electron microscope studies, Bassot ( 1960a,b, 1963, 1966) has been led to recognize several categories of photocytes in the serial photophores of stomiatoid fishes such as Gonostomatidae, Sternoptychidae, Stomiatidae, and Maurolicidae. Category A photocytes occupy the inner and deeper region of the photophore. Category B usually occur in the outer region of the photophore (see following section). The A photocytes have an abundant basophilic ergastoplasm (i.e., endoplasmic reticulum), and they are rich in secretory granules of a glycoprotein nature. According to their number and arrangement they have been subdivided into several categories. Suffice it to say that in

370

J. A. C. NICOL

some photophores they are small cells densely packed ( e.g., Maurolicus and Argyropebcus), or they are grouped in acini which open into a common canal (e.g., Gonostomu and Cyclothone), or they are tall and radially arranged ( e.g., Chauliodus and Vinciguerria) The B photocytes, either enclosed by the A photocytes (Chauliodus) or lying external to them ( Argypropekcus), are also secretory in appearance. They may be pigmented (Maurolicus); and they contain fine granules, glycogen, glycoprotein, and much ribonucleic acid. Ribosomes are abundant as are vesicles derived from Golgi. There is a recognizabIe transition between A and B photocytes in some species (e.g., Maurolicus). Furthermore, B photocytes become transformed into specialized cells of a lens or a gelatinous body in the outermost region of some photophores, e.g., those of Argyopekcus. B cells are sometimes absent, e.g., Gonostoma. By electron microscopy a secretory cycle has been traced in the photocytes of stomiatoids. The successive stages of the secretory sequence are observed in separate cells spaced proximo-distally along the longitudinal axis of the photophore (Maurolicus),or within one elongate cell along its long axis (Vinciguerria) ( Bassot, 1963, 1966). In Maurolicus, the A cells are fairly small; they contain a great many secretory granules and a strongly basophilic ergastoplasm. Proximal cells (Le., the celIs deepest in the photophore) have large and abundant secretory granules possessing irregular membranes. The ergastoplasm consists of small cisterns bearing numerous ribosomes. Mitochondria are small and have cup-shaped depressions. Intermediately placed A photocytes are heavily charged with ergastoplasm and secretory granules. The ergastoplasm is made up of very long cisterns, fairly regular in size and lying parallel to each other. Secretory granules are large and their membranes are well defined. Mitochondria are much enlarged. Distal A photocytes contain some large secretory granules and elongated cisterns of ergastoplasm, diminished mitochondria, and lamellar concentric figures. The latter are derived from the various double membrane systems, but especially the mitochondria. The A photocytes of Vinciguerria are very tall cells, radially arranged. The innermost region of the cell contains a basophilic ergastoplasm crossed by long mitochondria. Cisterns composing the ergastoplasm are formed of rough ER; they are very long, uniform, and they are aligned along the long axis of the cells. Above the basal region they break up into little vesicles. The long mitochondria exhibit structual changes along their axes; the basal region has cup-shaped depressions containing some secretory material; the middle region is closely invested by ER, and lacks secretory material; and the distal portion gives

.

7.

BIOLUMINESCENCE

371

way to myelinlike figures ( Bassot, 1966). The significance of these various cytoplasmic features in terms of light production is conjectural. A description, based on electron microscopy, of the photophores of lantern fishes is not yet available. There are three kinds of cells within the photophore of Porichthys, viz., the lens cells, the photocytes (supposed photogenic cells), and supportive cells closely associated with the latter (Fig. 11). Electron micrographs (Fig. 12) show that the photocyte, in addition to the usual cytoplasmic components, contains numerous vesicles and lamellar membranous whorls. The latter possess cytoplasmic cores containing mito-

Fig. 11. Diagrammatic representation of a photophore of Porichthys notatus, showing the terminations of nerve fibers. Boutons termineux abut on lens cells ( t l ) , supportive cells ( t 2 ) ,extracellular channels between photocytes ( t 3 ) , channels between lens cells and photocytes ( t 4 ) ,on blood vessels (k),and photocytes ( t e ) . BV, blood vessel; L, lens; Phc, photocyte; R, reflector; and S, supportive cell. After a drawing by Dr. Judy Strum, based on electron micrographs of serial sections.

372

J. A. C. NICOL

Fig. 12. Electron micrograph of a section through a photophore of Porichthys The upper region is occupied by a cell ( P ) presumed to be a photocyte; below and adhering to it is a supportive cell ( S ) . The photocyte contains mitochondria and numerous vesicles; microvilli extend into extracellular spaces. The supportive cell contains many cytoplasmic filaments which extend into its processes. Photogenic and supportive cells make contact at desmosomal adhesion sites ( D ), Nerve ( N ) ; blood vessel ( B V ) . The arrow points to a nerve ending containing many synaptic vesicles having dense cores. Photograph supplied by Dr. Judy Strum.

notatus.

7.

BIOLUMINESCENCE

373

chondria and rough ER, and they exhibit structural changes in sympathy with photophore stimulation and activity. Long microvilli extend from the photocytes into surrounding extracellular channels. The supportive cell extends around the photogenic cell in cup-shaped fashion, adhering closely to it at desmosomal sites (Fig. 12). Many cytoplasmic filaments, extending out into cell processes, are characteristically found in the processes of the supportive cells. Numerous nerve boutons terminate around the photogenic region, not directly on the photocytes, but on the side of the basement membrane opposite that which faces the photogenic tissue. They adjoin lens cells, photogenic and supportive cells, extracellular channels, and connective tissue ( Fig. 11).Their abundance and distribution are such that they could saturate the photogenic region of the photophore with transmitter substance (Strum, 1966, 1968a,b, 1969a,b). 2. LENSAND REFLECTOR Compound photophores frequently contain a mass or plate of clear tissue overlying the photogenic mass. The tissue is epithelial, the cells have a homogeneous cytoplasm devoid of inclusions, the appearance is like that of an ocular lens, and the structures are generally regarded as lenses or accessory dioptric bodies. Such lenticular structures occur in the serial photophores of stomiatoids (Fig. 8), in the light organs of Porichthys (Fig. 11) and Opisthoproctus (Fig. lo), and in many other kinds of fishes. In myctophids the scale over the light organ is thickened and lenticular in shape. The ventral photophores of Maurolicus are capped by a flat plate and a gelatinous body. The lenslike body lying over the photogenic tissue of stomiatoids is formed from B category cells. Derived through transitional stages from A cells, at first they are rich in ribonucleic acid and glycogen and contain glycoprotein and many ribosomes ( Maurolicus) . Further distally the cytoplasm becomes electron dense and devoid of organelles, and the cells become transformed into the specialized cells of the lens (Bassot, 1960a, 1966). The wall of the photophore (and often of light glands) generally contains a reflector together with a backing of dark pigment. In those photophores that have been examined minutely, especially stomiatoids, the reflector is found to contain elongated connective tissue cells, which are regularly rectilinear and which lie tangential to the curvature of the photophore. The cells contain platelets of guanine, arranged in regular piles above one another, parallel to the surface of the reflector. In surface view each pile of superposed platelets is lozenge-shaped or

374

J. A. C. NICOL

hexagonal (Barraud et al., 1959; Bassot, 1959, 1966). The reflecting platelets of Porichthys are small, thin hexagons and within the cell each lies inside a membrane-bounded sac (Fig. 13) (Strum, 1968a). The system resembles that found in tapeta lucida of eyes and silvery skins of fishes (Denton and Nicol, 1965; Land, 1966; Best and Nicol, 1967), where high reflectivity is achieved by platelets suitably arranged to cause constructive interference.

Fig. 13. Reflector of o photophore of Porichthys notatus. Electron micrograph of a transverse section. The elongated clear areas are spaces occupied by crystals of guanine (lost during preparation). The crystals are intracellular, each being surrounded by a membrane. Photograph supplied by Dr. Judy Strum.

7.

BIOLUMINESCENCE

375

Without precise optical data it is difficult to understand the functioning of the accessory structures of photophores. Dark pigmented mantles are so commonly present that there must be some advantage in ensuring that no light diffuses inward into the tissue of the fish. They occur even when there is a well-organized reflector, presumably mopping up light transmitted through the latter. Photophores are generally arranged and organized in such a way as to cause light to be emitted in preferred directions-simple observation of the luminescence of the ventral photophores of Stomius reveals how the angle of emission is restricted-and the pigment hulls are part of the total mechanism for achieving this effect. The function of the so-called lenses of photophores needs clarification. It may be that the problem is one of semantics rather than of physiological optics. If the refractive indices of the lenses prove to be greater than those of the surrounding media, the lens could affect the spread of light, possibly concentrating or collimating the beam. A gelatinous body lies distal to the lens in some photophores. Bassot (1960a) has stated that in Maurolicus it is transparent, homogeneous, and refractive (rbfringente). The shape of the reflector and the orientation of its component platelets may be more important than the action of the lens in controlling direction and spread of the light beam.

IV. BIOCHEMISTRY

Until recently all efforts to determine the chemical basis of the luminescent system in fish made use of a simple reaction discovered by Dubois in 1885, and afterward extended to many animals. It consists, first, of grinding the photogenic tissue in water, producing a luminescent brei in which the light gradually dies away as the substrate becomes exhausted, leaving a crude solution of key enzyme (luciferase); second, it consists of extracting photogenic tissues with hot water, thereby destroying the enzyme but leaving the substrate (luciferin) in solution. When applied to favorable material, the procedure yields separate enzyme and substrate preparations which emit light on mixing. This simple luciferinluciferase test has been tried on various species of luminous fish. Results have been negative with stomiatoids and myctophids (Harvey, 1955). Positive reactions have been obtained with preparations of Malacocephalus ( Macrouridae ) , Parapriacanthus ( Pempheridae) , and Apogon (Apogonidae) (Haneda et al., 1958; Johnson et al., 1961a). It is generally believed that the luminescence of Malacocephalus is caused by luminous

376

J. A.

C. NICOL

symbiotic bacteria (Haneda, 1938). (For a review of fish luminescence consult Johnson, 1967.) The luminescent glands of both Parapriacanthus beryciformes and Apogon ellioti yield crude extracts of luciferin and luciferase that luminesce when combined; the reaction is oxygen-dependent (Johnson and Haneda, 1958; Haneda and Johnson, 1958). The substrates (the luciferins) from both species have been purified, and that from Parapriacanthus has been obtained in crystalline form; the methods are given by Johnson et al. (1961a) and Sie et al. (1961). It is noteworthy that the two luciferins are identical or very closely similar, both with each other and with that of the ostracod crustacean, C ypridina Iiilgendorfii. Common characteristics include closely similar absorption spectra, fluorescence and emission spectra, changes on autoxidation, and behavior to Cypridina luciferase (Sie et al., 1961). The luciferin of Cypridina is an indole derivative having tryptamine, arginine, and isoleucine as constituents; its properties ( absorption spectrum, fluorescence, quantum yield, etc. ) have been well described (Johnson et al., 1961b; Johnson, 1967).

-C

lN H 2 HCI

“H,

H

Structure of the luciferin of Cyplr’dina (Kishi et al., 1966)

In the presence of luciferase it is oxidized to oxyluciferin, with emission of light; visible light is produced by an extremely minute amount of substrate (by 5 ml of a solution containing 2 x l&ll M luciferin with 0.01 mg/ml of luciferase protein) (Johnson, 1967).

7 . BIOLUMINESCENCE

377 lueiferase

2 LH2 + O2 ......--.......---2 cations Nit+. etc.

L + 2 H $0

+ hv

where L is luciferin, hv a photon. There are noteworthy immunological, chromatographic, and other differences between the luciferases of Apogon and Cypridina. Nevertheless, the luciferases from the two sources are each mutually effective on the luciferins. A ground suspension of the light organ of Apogon persistently luminesces for a long time, but if diluted luciferase of Cypridina is added a bright luminescence of short duration results. The luciferase of Apogon does not appear to be less active than that of Cypridina, and the prolonged luminescence of crude extracts may result from the low concentration of luciferase (Johnson et al., 1960; Tsuji and Haneda, 1966). First-order decay of luminescence of closely matched mixtures of substrate and enzymes from two sources has been compared, viz., of Cypridina luciferin plus luciferases of Apogon and Cypridina ( Fig. 14). Immediately after mixing, the initial rate of decay of light intensity is significantly greater for Apogon luciferase than for Cypridina luciferase, thereafter, the rate of Apogon is slightly higher than that for Cypridina.

0

30

60

90

120

Seconds after mixing

Fig. 14. Decay of luminescence in mixtures of luciferase and luciferin of Cypridina, and of luciferase and luciferin of Apogon. From Tsuji and Haneda ( 1966).

378

J.

A. C. NICOL

Wavelength (nm)

Fig. 15. Comparison of the spectrum energy distribution of the luminescence of the luciferins of ( A ) Cypridina and ( B ) Porichthys. Emission is recorded as a function of apparent wavelength, but the latter is uncorrected for nonlinear response of the photodetector. From Cormier et al. (1967).

The whole picture is very complicated because it has been discerned that dead but still luminescing Cypridina occur in the gut of these fishes, and there is a possibility, not yet excluded, that the fish may acquire its luciferin, but not the enzyme (luciferase), from the crustacean (Haneda and Johnson, 1962; Johnson, 1967). Less equivocal in origin is the luciferin of the midshipman, Porichthys porosissimus. Using methods of extraction successfully developed for Cypridinu luciferin, Cormier et uZ. (1967) have isolated and obtained from that fish a highly purified luciferin that in chemical properties is identical with that of C ypridim, namely, R f values, characteristics of autoxidation, solubilities, and emission spectra ( Fig. 15). The oxidation of Porichthys luciferin is catalyzed by Cypridina luciferase, and the kinetics of light production by the luciferins from the two sources are the same (Fig. 16). The luciferase of Porichthys also reacts with Cypridina luciferin to produce light. The level of luciferase in the photophores of Porichthys is very low. This circumstance may account for Haneda’s failure to obtain a positive response on testing for luciferin-luciferase

7 . BIOLUMINESCENCE

379

Fig. 16. Kinetics of the luminescent reaction of the luciferins of ( A ) Cypridina and ( B ) Porichthys when catalyzed by the luciferase of the former. From Cormier et al. (1967).

(Haneda, 1963). Luciferin is widely distributed in tissues of the midshipman, but it occurs in greatest concentration in the photophores (Cormier et al., 1967). Only further work will show how widespread is the Cypridina-type of luciferin among fishes. It may be noted that lights having colors other than blue, in uivo, have been noted by several observers, but even these observations need confirmation. A second type of luminescent system among fishes is indicated by work of Crane (1968) on the batfish, Dibranchus atlanticus. The skin of the fish exhibits yellow fluorescent patches, and it luminesces when adrenaline is injected into the fish. Light emission from pieces of skin is increased by adding hydrogen peroxide and horseradish peroxidase, and the author is inclined to believe that a peroxidase system is involved, such as that which characterizes the luminescence of the hemichordate Balunoglossus biminiensis ( Cormier et al., 1966).

V. REGULATION OF LIGHT EMISSION

Even when the luminescence of fish is continuous, as it is in those species having luminous bacteria, there are devices that regulate or modulate the emission of light. Probably all species having intracellular

380

J. A. C. NICOL

luminescence produce light intermittently and have photophores subject to internal control.

A. Direct Control 1. ELASMOBRANCHS

All known luminous elasmobranchs are of the suborder Squaloidea. Of these, Isistius brasiliensis luminesces spontaneously and continuously for long periods, but luminescence is not a constant feature, for one live specimen failed to show light (Bennett, 1840; Bigelow and Schroeder, 1948) . In another, Etmopterus niger, the light appeared spontaneously but intermittently, and luminescence could be induced by electrical stimulation (Beer, in Johann, 1899; Hickling, 1928). Etmopterus frontimuculatus luminesced when handled, the light appearing after several minutes ( Ohshima, 1911). Whether the controlling mechanism is hormonal or nervous is unknown. Physiological studies depend upon a reasonable supply of experimental animals, and this factor has severely limited the amount of information forthcoming about luminescence of pelagic fishes, Luminescent sharks can be secured alive and in good condition by line fishing, and they offer possibilities for further physiological investigations ( Hickling, 1928; Forster, 1964). 2. TELEOSTS The serial photophores of fishes generally are innervated. Nerves have been traced to the dermal light organs of Cyclothone, Argyropehcus, Polypinus, Myctophum, Porichthys, and others (Fig. 11). The nerve penetrates the sheath of the photophore and enters the aggregate of photogenic cells. Terminations are described in Section 111, C. In the trunk the serially arranged photophores are supplied by nerves coming off medial and ventral branches of spinal nerves. In the head they are innervated by fibers coming from the facial nerve (the mandibular, buccal, and hyoid branches), and by a branch of the superior maxillary nerve (which probably includes facial fibers) (Handrick, 1901; Gierse, 1904; Brauer, 1908; Ohshima, 1911; Ray, 1950; Nicol, 1957). In 1924, Greene and Greene showed that adrenaline when injected could excite luminescence in Porichthys notatus, and thereby a legend of hormonally mediated luminescence was born. This hypothesis, timeworn into respectability by repetition, is hardly credible on the evidence adduced. Adrenaline also evokes luminescence .in Echiostoma barbatum,

7.

BIOLUMINESCENCE

381

causing all the photophores to emit light (the rows of lateral and ventral photophores, scattered minute photophores all over the body and the ventral fins); the cheek organ is unaffected (Harvey, 1931). Similarly, adrenaline causes the hatchetfish, Argyropelecus olfersi, to luminesce (Bertelsen and Grgntved, 1949). The response following injection of adrenaline is immediate in Echiostoma or delayed in Porichthys (where the latency is several minutes). In one trial, injection of pituitrin was found to excite luminescence in Porichthys ( Greene and Greene, 1924). This action has not been confirmed (Nicol, 1957). Electrical stimulation causes the photophores of Porichthys to luminesce. Stimulation of the spinal cord is effective when the circulation is interrupted. There is no doubt that luminescence is under nervous control and that the photocytes are activated by efferent fibers, but the possibility exists that the suprarenal or other glands may activate the photophores as well-at least this possibility is not excluded by the evidence (Nicol, 1957). The general conclusion is that the serial photophores of these teleosts are under nervous control, probably of the peripheral sympathetic system. The outflow of this system from the spinal cord occurs at restricted levels, and the postganglionic efferent fibers are distributed to the periphery via the mixed spinal and certain of the cranial nerves. The nerve fibers concerned may be adrenergic, and there is a clear analogy between the innervation of the serial photophores and that of the dermal melanophores. What is required now is not more flailing of this exhaustively threshed gleaning but some more welldirected work and results (Nicol, 1960c, 1967). Steindachneria, a fish having a visceral light organ, is also caused to luminesce by the injection of adrenaline (Cohen, 1964). Very little is known for certain about the light organ of this fish. Some teleost fishes possess several anatomically distinct kinds of photophores. In melanostomiatids the various organs are the cheek (postorbital) organ, the tentacular bulb, the serial photophore, bands and patches of minute light organs, and strips of luminous tissue. Echiostoma barbatum is one of the few species that have been examined alive. The cheek organ was observed to flash rhythmically when the fish was handled and its periodic flashing seemed to be unaffected when adrenaline was injected into the fish (see Section IV, A, 2 ) (Beebe, 1926; Harvey, 1931; Beebe and Crane, 1939; Morrow and Gibbs, 1964). Beebe and Crane (1939) have stated that the cheek organ of Echiostoma, and also of Tactostomu, is movable, and that when it is rolled downward the light is cut off. Harvey (1931, 1952) has asserted, however, that flashing in the postorbital organ of Echiostoma is not a result of unscreening of a continuously luminous surface and that the light appears

382

J. A. C. NICOL

and disappears in the organ itself. Postorbital and suborbital photophores of stomiatoids are innervated and nerves to the photogenic tissue have been traced in Astronesthes and other forms (Brauer, 1904, 1908; Nicol, 1960b); however, the source of the innervation is unknown. Luminescence has been observed in the barbels of various stomiatoids (Beebe and Crane, 1939). The light bulb of Leptostomius flared momentarily after mechanical stimulation; no other photophores became luminous (Nicol, 1958). Brauer ( 1908) describes the innervation of tentacular bulbs. The caudal glands of myctophids flash independently of the serial photophores, and it has been observed that when the caudal lights are flashing the other photophores are seldom seen (Beebe, 1926; Beebe and Vander Pyl, 1944). The caudal organs are innervated by the sympathetic chain ( Brauer, 1908; Anadh, 1957). It appears from these few observations that different kinds of photophores in the same fish may be independently controlled, possibly by different components of the nervous system ( Nicol, 1980~).

B. Indirect Control There are fishes that have devices for screening or concealing light organs or for changing the brightness of the light emitted. In a preceding section (Section IV, A ) the rotating character of the cheek organs of stomiatoid fishes has been noted. The organ lies in a loose pocket beneath the skin from which it is separated by a narrow space; beneath the light organ there is a narrow cleft which is lined with dark pigment. In various species examined, including Chauliodus, Astronesthes, and Photostomias, there is a small muscle which passes down beneath the light organ, to be inserted on the lower or outer surface of the latter (Fig. 7 ) (Tchernavin, 1953; Nicol, 1960a). The arrangement is such that when the muscle contracts, it rotates the light organ downward, causing its bright surface to be concealed in the ventral pigmented cleft and bringing its upper pigmented surface against the aperture of the window. Since the photocytes are innervated, it is probable that the photocytes themselves are subject to direct nervous control; consequently, the rotating device would seem to be concerned with some function other than control of light emission. Stomiatoid fishes like Chauliodus and Stomias are dark; and the light organ, by virtue of its efficient reflector, is a conspicuous organ on the cheek of the fish when exposed. It is quite possible that the chief virtue of the rotating mechanism is to conceal the brightly reflecting surface when the light organ is not in use

7.

BIOLUMINESCENCE

383

or when the fish is endangered. It is curious that Idiacanthus (Idiacanthidae) appears to have a different kind of mechanism for concealing the cheek organ. The muscle to the light organ is inserted in a pigmented mass above it in a manner that suggests that it pulls the pigmented tissue down in front of the light-emitting surface (Nicol, 1960a). The light organs of the Anomalopidae shine continously; it has been supposed that symbiotic bacteria are involved (Harvey, 1952; Haneda, 1953; Bassot, 1966). In the species Anornulops katoptron the light organ is provided at the anterior edge with a hinge which aIlows it to be rotated downward until its light-emitting surface comes into contact with a layer of black pigmented tissue, as in a pocket; in consequence, the light from the organ is cut off. Photoblepharon palpebratus, on the contrary, has a fold of black tissue, a membranous curtain, lying along the ventral edge of the light organ, and this fold can be drawn up over the organ, thus concealing the light. A third species Kryptophanaron alfredi, known from a single specimen, apparently has a pigmented lower lid like that of Photobbpharon (Harvey, 1922, 1952; Haneda, 1955). Muscle and mechanisms responsible for moving the light organ or screen have been described by Steche ( 1909), but the accounts are not easy to comprehend. Both Photoblepharon and Anomalops are surface dwellers in inshore waters of the western Pacific. Under natural conditions, when swimming near the surface at night, Anomulops reveals its light intermittently, the luminous surface of the light organ appearing and disappearing at intervals of several seconds. The fish shows its light only in darkness; when the fish is suddenly illuminated it conceals its light. Photoblepharon is said to show its light continuously (Haneda, 1953, 1955). The light organs of Anornulops and Photoblepharon are innervated by a branch of the superior maxillary division of the trigeminal nerve; its ultimate destination is unknown, but it may supply the blood vessels of the organ. Descriptions have been presented previously (Sections 111, B, 3 and 111, B, 4) of visceral and other internal light organs, the contents of which shine continuously. Yet there are well-authenticated accounts describing how luminescence from the fish possessing them varies in intensity, or on occasion ceases altogether. Of Acropomu it has been written that the luminosity is continuous, but it can be dimmed by means of chromatophores which, by their contraction and expansion, probably control the amount of light emitted (Haneda, 1950). Two mechanisms have been suggested to account for changes of light intensity of leiognathid fishes. The visceral light organs (Fig. 4 ) , surrounded by an opaque white membrane, possess two fenestrae. Normally the windows are uncovered, but on occasion they can be occluded either

384

J. A. C. N I O L

by movement of pigment or by expansion of the opaque membrane over the openings (Haneda, 1940, 1950). The situation in some deep-sea gadids and macrourids is still obscure. The suggestion that they discharge a luminous secretion has not been confirmed. The visceral luminescent gland lies over a clear spot or spots in the ventral body wall, and these spots are seen to shine in live fish. Haneda (1938, 1951) believes that light emission is influenced by concentration or dispersion of pigment in melanophores lying in the skin over the light organ. A similar explanation has been advanced to account for changes in the intensity of light emitted from the bacterial light organs of Monocentris (Fig. 3 ) (Okada, 1926; Yasaki, 1928).

VI. PHYSICAL CHARACTERISTICS

The luminescence of fishes generally is blue or blue-green. Harvey (1955) has tabulated such information as is available. Some pertinent observations are summarized in Table I. The colors of very dim lights, when no standard of reference is available, are notoriously difficult to evaluate. Several spectrum emission curves are now available. Apogon, Parapriacanthus, and Porichthys have luciferins similar to or identical with that of Cypridina; the spectrum emission curves for the luminescent systems in vitro are identical with that of Cypridina luciferin (Figs. 15 and 17). Peak emission occurs at about 460 nm; half the energy lies between 425 and 520 mm ( Sie et al., 1961). The emission spectrum of Mytcophum punctatum is similar, peak emission occurring at about 470 nm, and half the energy lying between 435 and 530 nm (Nicol, 1960a). Luminous bacteria, cultured from the light organ of Malacocephalus laevis, emit light in the range 430-638 nm, with A,, at 510 nm, whereas bacterial cultures from Coelorhynchus emit light having A,, at 472 mm (Takase and Haneda, in Harvey, 1952). For various reasons the light from bacterial cultures and from reactions in vitro need not have emission spectra identical with the light emitted from living fish. As one instance it may be mentioned that luminescence of the pinecone fish, Cleidopus gloria-maris, is blue-green, whereas the contents of the light organ emit a blue light. An external layer (termed a “filter”) covering the light gland is orange-red in color, and the absorption characteristics of this layer may account for the difference in colors of the luminescence noted above ( Haneda, 1966). With the limited data now available there is little more one can

Table I Colors of Luminescent Lights of Fishes Fish

Color of light

Remarks

Reference"

Elasmobranchs Squalidae Etmopterus spinax Etmopterus spinax Daletiidae Isistius brasiliensis

Green Blue-green

Photophores Photophores

1 2

Green

Photophores

3

Blue-green

Secretion

4

Blue Green-yellow

Photophores Photophores

5 6

Blue

Photophores

7

Blue

Photophores

8

Blue Yellow Blue

Cheek organ Trunk photophores Submental tentacle

9 9 8

Blue

Photophores

10

Green

Luminous gland with bacteria Luminous gland with bacteria Luminous gland

11

Teleosts Searsiidae Searsia schnakenbecki Sternoptychidae Polyipnus stereope Argyropelecus olfersi Chauliodontidae Chauliodus aloanii Stomiatidae Stmnias ferox Melanostomiatidae Echiostoma barbatum Echiostoma barbatum Leptostomias sp. Myctophidae Myctophum watasai Mecrouridae Coelorhynchus hubbsi Malacocephalus laevis Steindachneria argentea Monocentridae Monocentris japonicu Cleidopus gloriamaris Anomalopidae Anamalops katoptron Leiognathidae Gazza minuta and Secutor insidiator Trachich thyidae Paratrachichthys prosthemius Oneirodidae Dolopichthys sp.

Blue or blue-green Blue Blue Blue-green

Luminous gland with bacteria Luminous gland with bacteria

12

13 14 15

Blue-green

Luminous gland

16

Blue or bluewhite

Luminous gland with bacteria

17

Blue-white

Luminous gland with bacteria

18

Blue

Esca

19

References: 1, Beer, in Johann, 1899; 2, Hicking, 1928; 3, Bennett, 1840; 4, Nicol, 1958; 5, Haneda, 1952b; 6, Bertelsen and Grclntved, 1949; 7, Skowron, 1928; 8, original; 9, Harvey, 1931; 10, Ohshima, 1911; 11, Haneda, 1951; 12, Hickling, 1925; 13, Cohen, 1964; 14, Harvey, 1917; 15, Haneda, 1966; 16, Haneda, 1953; 17, Haneda, 1950; 18, Haneda, 1957; 19, Waterman; 1939. 385

386

J. A.

C. NICOL

Wavelength (nm)

Fig. 17. Spectrum emission C U N ~ S of the luminescence of Apogon luciferin catalyzed by Cypridinu luciferase (curve A), and the luminescence of Myctophum punctatum (whole fish, curve B ) . A from Sie et a2. (1961); B from Nicol (1960a).

say beyond the general statement that maximal light emission lies in about the same region of the spectrum as that to which the eyes of fishes are most sensitive. Fishes of epipelagic and inshore waters have visual reds ( rhodopsins) which absorb maximally in the blue-green region of the spectrum (500510 nm), and deep-sea fishes have chrysopsins which absorb maximally in the blue (around 480 nm). These are rod pigments, employed in scotopic vision, when luminescence is operative. Moreover, the broad asymmetric shape of the spectrum emission curves is much like that of the absorption curves of visual pigments. Figure 18 shows two curves superposed for comparison. Such similarities of the emission and absorption spectra allow a considerable degree of luminous efficiency for the luminescent system, i.e.,

where A is the absorption and E the emission. In Myctophum punctatum it is about 75%. The efficiency is one factor that determines the distance at which fish can employ light usefully for many of the purposes delineated in Section VII.

7 . BIOLUMINESCENCE

387

Wavelength (nm)

Fig. 18. Luminous efficiency of the luminescence of the lantern fish, Myctophum punctaturn. Curve A, spectrum emission; B, absorption (difference curve) of the retina; and C, product curve based OR curves A and B (the latter two normalized at A,,,,). A from Nicol (1960a); B from Denton and Warren (1957).

The light intensities of lantern fishes lie in the region of 1 x 10-E pW/cm2 receptor surface at a distance of 1 m (Nicol, 1958; G. L. Clarke et al., 1962). These, the only values available, were obtained from captive fish, artificially stimulated, and it would be desirable to know the radiant output of the lights of fish emitting under more normal conditions and the intensities in known directions with reference to established axes of the fish. However, with these albeit imperfect data for emission, some estimates can be made of the distances at which a mesopelagic fish such as lantern fish should be able to see a luminescent light. For very clear oceanic waters having a transmission of over 90% per meter, the distance theoretically should be 10-16 m under optimal conditions; in actuality it may well be considerably less (Nicol, 1958; G. L. Clarke and Denton, 1962). Observations of schools of lantern fishes, Ceratoscopelus maderensis, from a bathyscaphe revealed that the schools were about 100-200 m apart, and the densities within schools were 10-15 fish per cubic meter. However, these fish were not luminescing (Backus et al., 1968).

388

J. A. C. NICOL

VII. SIGNIFICANCE AND EMPLOYMENT OF LUMINESCENCE

A. Theories of Luminescence Much that has been written about the function and role of luminescence in fishes is conjecture. Various explanations offered for luminescence are: ( 1 ) The light is a lure used to attract other animals, which are captured and consumed. (2) Light is used to deter, or to distract the attention of, attacking animals. ( 3 ) Luminescence provides a source of light for seeing at night or in dark surroundings. ( 4 ) Luminescence is used to make the animal conspicious, advertising it and serving as a warning signal. (5) Animal lights are signs exchanged by members of a species for various purposes, including: ( a ) mutual recognition in schooling; ( b) sexual recognition, attraction, and repulsion; (c) courtship; and ( d ) aggression, defense of territory. ( 6 ) Ventral lighting conceals by obliterating silhouette, causing the animal to melt into the background. Many of these theories have been elaborated repeatedly elsewhere; a particularly good exposition has been presented by Marshall (1954). Deep-sea teleosts are generally very fragile and short-lived in captivity, little is known about their behavior and habits in life, consequently these intriguing hypotheses mostly are still to be proved.

B. Elaboration and Criticism of Theories 1. LUMINESCENT LURES

It has been suggested that some fishes employ light organs as lures to attract prey within striking distance. The function has been ascribed to the luminous bulb on the fishing tentacle or illicium of deep-sea ceratioids, the luminous chin barbels of stomiatoids and Linophyne ( a ceratioid) (Fig. 2 ) , the photophores inside the mouth of Chaulwdus and Dactylostomias, and the lingual photophore of Neoscopelus, to mention some plausible examples. The basis on which this theory is founded, viz., that animals are attracted to small point-sources of light in the darkness of the depths

7.

BIOLUMINESCENCE

389

is, to be sure, hypothetical. It is not unlikely that a luminous flash is one of the clues that pelagic creatures make use of when seeking food, that is, if they have good visual organs. Chauliodus and Ceratias have remarkable mechanical devices for manipulating their angling apparatuses; their tentacles bear luminous lures which obviously are of great importance to the animal (Brauer, 1908; Rauther, 1941; Bertelsen, 1943, 1951; Waterman, 1948; R. Clarke, 1950; Tchernavin, 1953). There are analogies in the behavior of Lophius and Uranoscopus, both of which are fishes of coastal waters and which employ conspicious lures which attract prey by their color, shape, and movement (Wilson, 1937; Cudger, 1945). These various arguments and analogies are very persuasive. But the same chain of arguments leads to the conclusion that the showing of a light can be a hazardous business in the deep sea, to be done at peril and to be kept to a minimum. The intensity of luminous lures has not been measured, but those observed have been very dim and it is unlikely that they are visible beyond a few meters. Since such lights can present sufficient contrast to be visible only in feebly illuminated or dark surroundings, they are effective only at night or at depths below light penetration. 2. REPELLINGOR CONFUSING PREDATORS Theories of this nature suppose that sudden flashing (and quenching) of photophores, or discharge of a luminous secretion, temporarily distract the attention of a predator, thus enabling the fish responsible for the flash or cloud of light to effect an escape. This role, among others, is one ascribed to the caudal glands of lantern fishes (Beebe and Vander Pyl, 1944). Fishes which discharge a luminous secretion are ceratiids (Ceratiidae) and searsiids (Seariidae). Female ceratiids ( Cerutias and Cryptosarm) have cephalic luminous glands. These are sacs, two or three in number, in the surface of the head, each opening by a pore. The sacs are indicated externally by caruncles; when pressed they discharge a secretion that spreads out into numerous points of light ( Bertelsen, 1951 ). The luminous secretion of Seursiu is produced by a large sac in the “shoulder,” and the light is bright when compared with that of other pelagic animals, approaching that of ctenophores (Nicol, 1958). This explanation, that light is used to confuse an enemy, has been elaborated with some plausibility to account for the luminous discharges of copepods, prawns, and squid (Harvey, 1952; David and Conover, 1961), The luminous discharge must have an arresting effect like that produced by discharge of ink by cephalopods (Tompsett, 1939).

390

J. A. C. NICOL

3. LIGHTORGANS AS TORCHES

Fishes having large light organs or light-emitting surfaces that shine steadily may make use of them to illuminate their surroundings that they may see when it is dark. This may well be the function of the suborbital and cheek organs of Anomulops, of stomiatoids, and of the alveolar and visceral organs of other teleosts (Harvey, 1952). A stomiatoid fish was observed by E. R. Gunther (in R. Clarke, 1950) that emitted a beam of light from its cheek organ. The light was directed fonvard and the fish, poised obliquely to the surface, lay beneath a swarm of krill which it attacked from time to time. Certainly, luminescence did illuminate the immediate binocular visual field of the fish. In a similar manner, lantern fish, Gonichthys coccoi, were observed to snap at copepods that swam within the range of the light given off by the ventral photophores (Beebe and Vander Pyl, 1944). Marshall (1954) has commented how light organs often are sited in the blind field of fishes, below the eyes, on the lower jaw, gular surfaces and isthmus, and along the lower and ventral surfaces of the trunk. They cannot be seen by the fish bearing them; the light they emit does not enter the eye directly, if it did it would negate its effect for perception. Nevertheless, the light they emit illuminates the forward and lower visual fields. There remains for notice, however, some peculiar small photophores so situated that they shine directly into the eye. They are found, for example, in Cyclothone braueri and Argyropelecus afinis ( Brauer, 1908). Are these small point-sources photometric standards, enabling the fish to estimate or compare ambient light levels, to adjust its luminescent output, perhaps, or to clock its emission against a response from another fish?

4. LUMINESCENCE AS A WARNING SIGNAL

Cott (1940)deals with warning coloration, that is, conspicuous colors and patterns shown by noxious animals, and he suggests that luminescence may sometimes have the same function. Some experiments in this field have been carried out with the midshipman Porichthys, which possesses a toxic spine. The midshipman flashes when attacked by predatory fish, and it is avoided or rejected (Lane, 1966). No other fish is yet known to use its light for this purpose. 5. INTRASPECIFIC SIGNALS There are several ways in which fishes can employ luminescence for signaling to one another. Shoaling species, by flashing their lights, may

7.

BIOLUMINESCENCE

391

be enabled to keep together. Anornulops swims about in large schools. From the bathysphere during a deep dive, Beebe (1935) observed that lantern fishes were usually grouped together and that their ventral lights shone continuously. Lantern fishes are one of several groups of animals whose light organs have definite and peculiar patterns varying from species to species, and it is possible that a particular pattern may be distinguished by members of a species ( Fig. 1). Inhabiting the pelagic zone between the surface and 1000 m, they execute vertical migrations and, at any one geographical position, they are, to a considerable extent, stratified according to species. Off the coast of Oregon, Pearcy (1964) found three species ( Tarletonbeania crenularis, Diaphus theta, and Lampancytus leucopsarus ) distinctly stratified but with overlap. Each of these species has a distinct pattern of light organs capable of providing a recognizable character. Boden et al. (1961) incline to believe that euphausiids (which are pelagic crustaceans carrying out extensive vertical migrations), by their flashing, arrange to keep together in a community. Because the light organs of the two sexes are sometimes different it has been supposed that they may therefore enable the sexes to recognize each other. Lantern fishes show such differences, the male having one or several light glands above, and the female below, the caudal peduncle (or they may be wanting in the female) (Marshall, 1954). Myctophum afine responds by flashing its caudal glands when it is shown a faint luminous surface (Beebe and Vander Pyl, 1944). Among stomiotoid fishes there are sometimes noteworthy differences in the size of the cheek organ between the two sexes. In many Melanostomiatidae the postorbital light is large in the male and small or absent in females. Males of Zdiacanthus (Idiacanthidae) are very small, about one-sixth the size of females, but have an enormous cheek organ, whereas the female’s cheek organ is minute. These differences suggest that the females seek the males and are guided to them by the light emitted from the postorbital organs (Beebe and Crane, 1939). Likewise, among deep-sea anglers, the small males may be guided to the females by the luminous lures of the latter. The escae, varying in detail from species to species, may serve for species recognition. Luminescent caruncles of female ceratiids may have the same role (Rauther, 1941; R. Clarke, 1950; Marshall, 1954, 1965). Both sexes of the midshipman Porichthys are alike in appearance. They have a courting behavior, they vocalize, luminesce, and the male guards a nest. The gravid female becomes luminous, the male responds by flashing and drives her beneath a stone, where the nest is located, to deposit the eggs (Crane, 1965). Porichthys is unusual among luminescent fishes because, being a batrachoid, it is one of the few fishes outside the groups Isospondyli

392

J. A. C. NICOL

and Iniomi to have serially arranged compound photophores [another is a sciaenid, Collichthys lucidus (Haneda, 196l)l. It is also a fish of shallow water, whereas luminous fishes of the other two groups are pelagic. The role of luminescence in its behavior is an instance of the way in which self-generated light can be employed for signaling between members of a species. No doubt luminescence is employed by pelagic fishes in other and singular ways of which we have no knowledge at present. The luminous displays of fireflies and of spawning polychaetes are analogous activities (Buck, 1937; Huntsman, 1948; Buck and Buck, 1968). Solitary animals in the depths of the ocean may become evenly distributed as a consequence of luminescence. The postulate is that there is mutual repulsion between members of a species, individuals tending to be dispersed at limits at which they can just see the lights of each other. Luminescence may be a clue through which solitary midwater fishes are able to maintain discrete their dimensional territories in a dark and continuous world ( Nicol, 1962b). THAT OBLITERATES SILHOUETTE 6. VENTRALLIGHTING

It has been noted several times that the longitudinal rows of photophores of deep-sea fish (Isospondyli and Iniomi) mostly face downward and that, as a consequence, they match downwelling light in the sea. The ventral surface of a fish, having this arrangement, seems as bright as the background. If it were not self-luminous, by interrupting downwelling light, the fish would appear dark against the background (Fraser, 1962; W. D. Clarke, 1963; McAllister, 1967). Silvery fishes have their reflecting and absorbing surfaces so organized and oriented that they reflect incident light in preferred directions that causes the fish to match background illumination and be practically invisible from most fields of view. The ventral side of a fish presents a special problem, however. Light in the sea is highly directional and the ratio of upward to downward directed light is small (1:200 to 1 : l O O in clear oceanic waters). When the belly of the fish tapers, or when it is laterally compressed as in the hatchetfish, the region lying in shadow below the fish is reduced correspondingly. In deep waters where light intensities are low, light organs on the lower surface of the fish could emit light that matches environmental light passing downward on either side of the fish. Moreover, ventral and lateral photophores are frequently so organized that the light they emit is narrowly restricted to a ventral path (Nicol, 1967). The lights of lantern fishes are within the range of intensities measured during the daytime in mesopelagic waters

7 . BIOLUMINESCENCE

393

inhabited by the fish (within one or two orders of magnitude; simultaneous measurements of both factors for one species-luminescence and environmental illumination-are not yet available) ( W. D. Clarke, 1966). Fishes having visceral light organs give off a diffuse glow from their ventral surfaces, an effect like that achieved by banks of closely spaced photophores ( Haneda and Johnson, 1962; Cohen, 1964). Mesopelagic fishes that execute great vertical migrations spend much time each day swimming upward or downward. Ascent and descent of fishes (not always precisely identified) have been followed by recording changes in depth of sound scattering layers (Taylor, 1968)) Observations from bathyscaphes also show that some fishes are often oriented at rest in attitudes other than horizontal. Supposedly, when the orientation of the fish is not horizontal, the light of the ventral photophores is ineffective to obliterate silhouette. Moreover, lantern fishes, Cerutoscopelus maderensb, observed from a bathyscaphe were at rest, but they were not luminescing (Blaxter and Currie, 1967; Harrison, 1967; Backus et al., 1968). McAllister ( 1967), after reviewing these various theories, suggests that the numerical abundance of predators in epipelagic waters has been a major factor in bringing about the ventral siting of photophores. The light from photophores, so placed, is invisible to fish above them.

VIII. CONCLUSIONS AND SUMMARY

Luminescence occurs exclusively, as far as is known, in marine fishes. Several kinds of photogenic organs are involved: dermal light organs or photophores, luminescent bulbs on tentacles, glands in the skin or ventral musculature, and glands connected to the viscera. Luminescence is intracellular, it is extracellular when a secretion is produced, or it is the result of bacterial activity. Luminescent organs usually have a pigment backing and quite often a reflector. The latter contains piles of guanine crystals, the organization being such as to produce constructive interference of reflected light. Lenslike structures are sometimes present. Photogenic tissue contains several kinds of cells. Progressive changes, involving secretory granules, mitochondria and ergastoplasm, related in some way to light production, have been described in the Iniomi. Cellular organelles in the photocytes of Porichthys alter during luminescent activity. Photophores are usually innervated and are subject to nervous control; the axonic terminals ( boutons termineux ) within the photophore of Porichthys have been described.

394

J. A. C. NICOL

Luminescent materials of three fishes have been extracted, purified and identified, viz., of Apogon, Parapriacanthus, and Porichthys. Luciferins from these sources are identical with that of the ostracod Cypridina, an indole derivative of established composition. Luciferases from these several sources cross-react with the luciferins. Most luminescent lights of fishes are blue or blue-green, Emission spectra approach or overlap spectrum sensitivity for scotopic vision. The many hypotheses advanced to explain use and employment of luminescence serve to emphasize its multifarious role among fishes. Some hypotheses, such as using luminescence to illuminate surroundings or as part of nuptial behavior, are supported by observations; other, highly plausible hypotheses, await confirmation. ACKNOWLEDGMENTS Support from P.H.S. Grant No. 7 R 0 1 NB08128-01 toward preparation of this chapter is acknowledged. REFERENCES Anadbn, E. (1957). Anatomia e histologia de las placas luminosas caudales de Lampodenu nitida (Toaning). Bol. Real. SOC.Espan. Hist. Nat., Secc. Biol. 55, 129-144. Backus, R. H., Craddock, J. E., Haedrich, R. L., Shores, D. L., Teal, J. M., and Wing, A. S. ( 1968).Cemtoscopelw maderensis: Peculiar sound-scattering layer identified with this myctophid fish. Science 160, 991-993. Barraud, J., Bassot, J.-M., and Favard, P. ( 1959). Identification radio-cristallographique et aspects cytologiques de la guanine dans le rkflecteur des photophore; chez “Maurolicvs pennanti” Walbaum ( T6lhstken Maurolicidae ). Compt. Rend. 249, 2633-2635. Bassot, J.-M. ( 1959).Les structures annexes des organes photoghnes de Maurolicus pennanti Walbaum (T6l&stken Maurolicides ) . Compt. Rend. 248, 297-299. Bassot, J.-M. ( 1960a). Donnkes histochimiques et cytologiques sur les photophores du Thl&st6en Maurolicus pennanti. Arch. Anat. Microscop. Morphol. Exptl. 49, 23-72. Bassot, J.-M. ( 1960b). Caractbres cytogiques des cellules lumineuses chez quelques Tkl6ostkens. Compt. Rend. 250, 3878-3880. Bassot, J.-M. (1963).Aspects du cycle s6crktoire des photocytes chez le T6lhstken Maurolicvs mulleri (Gmelin). Compt. Rend. 256, 47324735, Bassot, J.-M. (1966).On the comparative morphology of some luminous organs. In “Bioluminescence in Progress” (F. H. Johnson and Y. Haneda, eds.), pp. 557610. Princeton Univ. Press, Princeton, New Jersey. Beebe, W. (1926). “The Arcturus Adventure: An Account of the New York‘s Zoological Society’s First Oceanographic Expedition.” Putman, New York. Beebe, W. ( 1935).“Half Mile Down.” John Lane, London. Beebe, W., and Crane, J. (1939).Deep-sea fishes of the Bermuda Oceanographic Expeditions. Family Melanostomiatidae. Zoologica 24, 65-238. Beebe, W., and Vander Pyl, M. (1944).Eastern Pacific expeditions of the New York Zoological Society. XXXIII. Pacific Myctophidae (fishes). Zoologica 29, 59-95.

7.

BIOLUMINESCENCE

395

Bennett, F. D. (1840). “Narrative of a Whaling Voyage Around the World from the Year 1833 to 1836,” 2 vols. Richard Bentley, London. Bertelsen, E. (1943). Notes on the deep-sea angler-fish Ceratias holbblli Kr. based on specimens in the Zoological Museum of Copenhagen. Vidensk. Medd. Dansk. Naturh. Foren. Kbh. 107, 185-206. Bertelsen, E. (1951). The ceratioid fishes. Dana Rep. No. 39, 276 pp. Bertelsen, E., and Gr@ntved, J. (1949). The light organs of a bathypelagic fish Argyropelecus olfersi (Cuvier ) photographed by its own light. Vidensk. Medd. Dansk. Naturh. Foren. Kbh. 111, 163-167. Bertelsen, E., and Munk, 0. (1964). Rectal light organs in the argentinoid fishes Opisthoproctus and Winteria. Dana Rept. 62, 17 pp. Bertelsen, E., Theisen, B., and Munk, 0. (1965). On a post-larval specimen, anal light organ, and tubular eyes of the argentinoid fish Rhynchohyalus natalensis (Gilchrist and von Bonde). Vidensk. Medd. Dansk. Naturh. Foren. Kbh. 128, 357-371. Best, A. C. G., and Nicol, J. A. C. (1967). Reflecting cells of the elasmobranch tapetum lucidum. Contrib. Marine Sci. 12, 172-201. Bigelow, H. B., and Schroeder, W. C. (1948). Sharks. In “Fishes of the Western North Atlantic” (J. Tee-Van, ed.), Mem No. 1, pp. 59546. Sears Found. Marine Res. Yale Univ. Press, New Haven, Connecticut. Blaxter, J. H. S., and Currie, R. I. (1967). The effect of artificial lights on acoustic scattering layers in the ocean. Symp. Zool. SOC. London 19, 1-14. Boden, B. P., and Kampa, E. M. (1964). Planktonic bioluminescence. Oceanog. Marine Biol. 2, 341371. Boden, B. P., Kampa, E. M., and Abbott, B. C. (1961). Photoreception of a planktonic crustacean in relation to light penetration in the sea. Proc. 3rd Intern. Congr. Photobiol., Copenhagen, 1960 pp. 18S-196. Elsevier, Amsterdam. Bolin, R. L. ( 1939). A review of the myctophid fishes of the Pacific coast of the United States and lower California. Stanford Ichthyol. Bull. 1, 89-156. Brauer, A. (1904). Ober die Leuchtorgane der Knochenfishe. Verhandl. Deut. Zool. Ges. 14, 16-35. Brauer, A. ( 1906). Die Tiefsee-Fische. I. Systematischer Teil. Wiss. Ergeb. ‘Valdiuia’ 15, 432 pp. Brauer, A. ( 1908). Die Tiefsee-Fische. 11. Anatomischer Teil. Wiss. Ergeb. ‘Valdiuia’ 15, 266 pp. Buck, J. B. (1937). Studies on the firefly. 11. The signaI system and color vision in Phtinus pyralis. Physiol. Zool. 10, 412-419. Buck, J. B., and Buck, E. (1968). Mechanism of rhythmic synchronous flashing of fireflies. Science 159, 1319-1327. Clarke, G. L., and Denton, E. J. (1962). Light and life. In “The Sea” ( M . N. Hill, ed.), Vol. 1, pp. 456-468. Wiley (Interscience), New York. Clarke, G. L., Conover, R. J., David, C. N., and Nicol, J. A. C. (1962). Comparative studies of luminescence in copepods and other pelagic marine animals. J. Marine Biol. Assoc. U.K. 42, 541-564. Clarke, R. (1950). The bathypelagic angler h h Ceratias holbolli Kroyer. ‘Discouery’ Rept. 26, 1-32. Clarke, W. D. (1963). Function of bioluminescence in mesopelagic organisms. Nature 198, 1244-1246. Clarke, W. D. (1966). “Bathyphotometric Studies of the Light Regime of Organisms

396

J. A. C . NICOL

of the Deep Scattering Layers,” Final rept. G. M. Defense Res. Lab., Sea Operations Dept., Santa Barbara, California. Cohen, D. M. (1964). Bioluminescence in the Gulf of Mexico anacanthine fish Steindachneria argentea. Copeia pp. 406409. Cormier, M. J., Kreiss, P., and Prichard, P. (1966). Bioluminescent systems of the peroxidase type. In “Bioluminescence in Progress” (F. H. Johnson and Y. Haneda, eds. ), pp. 363-384. Princeton Univ. Press, Princeton, New Jersey. Cormier, M. J,, Crane, J. M., Jr., and Nakano, Y. (1967). Evidence for the identity of the luminescent systems of Porichthys porosissimus (fish) and Cypridina hilgendorfii (crustacean). Biochem. Biophys. Res. Commun. 29, 747-752. Cott, H. B. (1940). “Adaptive Coloration in Animals.” Methuen, London. Crane, J. M., Jr. (1965). Bioluminescent courtship display in the teleost Porichthys notatus. Copeia pp. 239L241. Crane, J. M., Jr. ( 1968). Bioluminescence in the batfish Dibranchus atlanticus. Copeia pp. 410-411. David, C. N., and Conover, R. J. (1961 ). Preliminary investigation on the physiology and ecology of luminescence in the copepod, Metridia lucens. Biol. Bull. 121, 92107. Denton, E. J., and Nicol, J. A. C. (1965). Reflexion of light by external surfaces of the herring, Clupea harengus. J. Marine Biol. Assoc. U.K.45, 711-738. Denton, E. J., and Warren, F. J. (1957). The photosensitive pigments in the retinae of deep-sea fish. J. Marine Biol. Assoc. U.K. 36, 651-662. Forster, G. K. (1964). Line-fishing on the continental slope. 1. Marine Biol. Assoc. U.K.4& 277-284. Fraser, J. (1962). “Nature Adrift. The Story of Marine Plankton.” G. T. Foulis & Co., London. Gierse, A. (1904). Untersuchungen iiber das Gehirn und die Kopfnerven von Cyclothone acclinidens. Morphol. J . 32, 802688. Greene, C. W., and Greene, H. H. (1924). Phosophorescence of Porichthys notatus, the California singing fish. Am. 1. Physiol. 70, 500-507. Gudger, E. W. (1945). The angler-fishes, Lophius piscatodus et americanus, use the lure in fishing. Am. Naturalist 79, 542548. Handrick, K. ( 1901). Zur Kenntnis des Nervensystems und der Leuchtorgane von Argyropelecus hemigymnus. Zoologica, Stuttgart 13, No. 32, 68 pp. Haneda, Y. (1938). On the luminescence of the deep-sea fish Mahcocephalus laevis (Lowe). Japan. J. Physiol. 3, 318-326. Haneda, Y. (1940). 0x1 the luminescence of the fishes belonging to the family Leiognathidae of the tropical Pacific. Palm Trop. Biol. %dies 2, 29-39. Haneda, Y. (1950). Luminous organs of fish which emit light indirectly. Pacific Sci. 4, 214227. Haneda, Y. (1951). The luminescence of some deep-sea fishes of the families Gadidae and Macrouridae. Pacific Sci. 5, 372-378. Haneda, Y. ( 1952a). Observations on luminous organisms of the Miura Peninsula, Japan. “History of Yokosuka-shi,” No. 5. Yokosuka, Japan (in Japanese), Haneda, Y. (1952b). Some luminous fishes of the genera Yarrella and Polyipnus. Pacific Sci. 6, 13-16. Haneda, Y. (1953). Observation on some marine luminous organisms on Hachijo Island, Japan. Records Oceanog. Works Japan 12, 103-108. Haneda, Y. (1955). Luminous organisms of Japan and the Far East. In “The

7 . BXOLUMIIW,SCENCE

397

Luminescence of Biologics1 Systems,” (F. H. Johnson, ed.), Publ. NO. 41, pp. 335-385. Am, Assoc. Advance. Sci., Washington, D.C. Haneda, Y. ( 1957). Observations on luminescence in the deep sea fish, Paratrachichthys prosthemius. Sn’. Rept. Yokosuka City Museum 2, 15-23. Haneda, Y. (1961). A preliminary report on two new luminous fish from Bombay and Hong Kong. Sci. Rept. Yokosuka City Museum 6, 45-50. Haneda, Y. (1963). Some observations on photophores, especially a transparent check area in the toadfishes, Porichthys. Sci. Rept. Yokosuka City Museum 8, 17-22. Haneda, Y. (1966). On a luminous organ of the Australian pine-cone fish, Cleidopus glorka-muris De Vis. In “Bioluminescence in Progress” (F. H. Johnson and Y. Haneda, eds. ), pp. 547-555. Princeton Univ. Press, Princeton, New Jersey. Haneda, Y., and Johnson, F. H. (1958). The luciferin-luciferase reaction in a fish, Parapriacanthus beryciformes, of newly discovered luminescence. Proc. Natl. Acad. Sci. U S . 44, 127-129. Haneda, Y., and Johnson, F. H. (1962). The photogenic organs of Parapriacanthus berycifomes Franz and other fish with the indirect type of luminescent system. J. Morphol. 110, 187-198. Haneda, Y., Johnson, F. H., and Sie, E. H.-C. (1958). Luciferin and luciferase extracts of a fish, Apogon marginatus, and their luminescent cross-reactions with those of a crustacean, Cypridina hilgendotJii. Biol. Bull. 115, 336. Haneda, Y., Johnson, F. H., and Shimomura, 0. (1966). The origin of luciferin in the luminous ducts of Parapriacanthus ransonneti, Pempheris klunzingeri, and Apogon eZlioti. In “Bioluminescence in Progress’^ (F. H. Johnson and Y. Haneda, eds. ), pp. 533-545. Princeton Univ. Press, Princeton, New Jersey. Harms, J. W. ( 1928). Bau und Entwicklung eines eigenartigen Leuchtorgans bei Equula spec. Z . Wiss. Zool. 131, 157-179. Harrison, C. M. H. (1967). On methods for sampling mesopelagic fishes. Symp. Zool. SOC. London 19, 71-126. Harvey, E. N. ( 1917). The chemistry of light production in luminous organisms. Carnegie Inst. Wash. Publ. 251, 171-234. Harvey, E. N. (1922). The production of light by the fishes Photoblepharon and Anomalops. Papers Tortugas Lab. 18, 43-60. Harvey, E. N. (1931). Stimulation by adrenalin of the luminescence of deep-sea fish. Zoologica 12, 67-69. Harvey, E. N. (1952). “Bioluminescence.” Academic Press, New York. Harvey, E. N. (1955). Survey of luminous organisms: Problems and prospects. In “The Luminescence of Biological Systems,” ( F. H. Johnson, ed.), Publ. NO. 41, pp. 1-24. Am. Assoc. Advance. Sci., Washington, D.C. Harvey, E. N. (1957a). A history of luminescence from the earliest times until 1900. Mem. Am. Phil. SOC. 44, 692 pp. Harvey, E. N. (1957b). The luminous organs of fishes. In “The Physiology of Fishes” ( M . E. Brown, ed.), Vol. 2, pp. 345-366. Academic Press, New York. Hastings, J. W. (1968). The chemistry of bioluminescence. Current Topics Bioenergetics 1, 113-152. Hickling, C. F. ( 1925). A new type of luminescence in fishes. J . Marine Biol. Assoc. U.K.13, 914-937. Hickling, C. F. (1928). The luminescence of the dog-fish Spinax niger Cloquet. Nature 121, 280-281.

398

J. A. C. NICOL

Hulet, W. H., and M u d , G. (1968). Intracellular bacteria in the light organ of the deep sea angler fish, Melanocetus murrayi. Copeiu 506-512. Huntsman, A. G. ( 1948). Odontosyllis at Bermuda and lunar periodicity. J. Fisheries Res. Board Can. 7, 363-369. Iwai, T. (1958). A study of the luminous organ of the apogonid fish Siphamia uersicolor (Smith and Radcliffe). J . Wash. Acad. Sci. 48, 267-270. Iwai, T. ( 1960a). Luminous organs of the deep-sea squaloid shark, Centroscyllium ritteri Jordan and Fowler. Pacific Sci. 14, 51-54. Iwai, T. (1960b). Notes on the luminous organ of the apogonid fish, Siphamia majimai. Ann. Mag. Nat. Hist. [I31 2, 545-550. lwai, T., and Asano, H. (1958). On the luminous cardinal fish, Apogon ellioti Day. Sci. Rept. Yokosuka City Museum 3, 5-13, Iwai, T., and Okamura, 0. (1960). A study of the luminous organs of the lantern fish. Sci. Rept. Yokosuka City Museum 5, 1-5. Johann, L. ( 1899). Ueber eigentiimliche epitheliale Gibilde (Leuchtorgane) bei Spinar niger. Z. Wiss. Zool. 66, 136-160. Johnson, F. H. (1967). Bioluminescence. Comp. Biochem. 27, 79-136. Johnson, F. H., and Haneda, Y. (1958). The luciferin-luciferase reaction in a fish, Parapriacanthus beryciformes, of newly discovered luminescence. Sci. Rept. Yokosuka City Museum 3, 25-30. Johnson, F. H., and Haneda, Y., eds. (1966). “Bioluminescence in Progress.” Princeton Univ. Press, Princeton, New Jersey. Johnson, F. H., Haneda, Y., and Sie, E. H.-C. (1960). An interphylum luciferinluciferase reaction. Science 1352, 422423. Johnson, F. H., Sugiyama, N., Shiniomura, O., Saiga, Y., and Haneda, Y. (1961a). Crystalline luciferin from a luminescent fish, Parapiacanthus beryciformes. Proc. Natl. Acad. Sci. U S . 47, 486-489. Johnson, F. H., Sie, E. H.-C., and Haneda, Y. (1961b). The luciferin-luciferase reaction. In “Light and Life” ( W . D. McElroy and B. Glass, eds.), pp. 206218. Johns Hopkins Press, Baltimore, Maryland. Kato, K. (1947). A new type of luminous organ of fish. Zool. Mag. ( T o k y o ) 57, 195-198 (in Japanese). Kier, A. (1967). Photophore histology in the lantern fish family Myctophidae. M. A. thesis, University of California, Santa Barbara, California. Kishi, Y., Goto, T., Hirata, Y., Shimomura, O., and Johnson, F. H. (1966). The structure of Cypridina luciferin. In “Bioluminescence in Progress” ( F. H. Johnson and Y. Haneda, eds.), pp. 89-113. Princeton Univ. Press, Princeton, New Jersey. Land, M. F. (1966). A multilayer interference reflector in the eye of the scallop Pecten maximus. J . Expptl. Biol. 45, 433-447. Lane, E. D. ( 1966). The biology of midshipman Porichthys porosissimus (Valenciennes) on the Texas coast around Port Aransas. Ph.D. dissertation, University of Texas, Austin, Texas. McAllister, D. E. (1967). The significance of ventral bioluminescence in fishes. J. Fisheries Res. Board Can. 24, 537-554. Marshall, N. B. ( 1954). “Aspects of Deep Sea Biology.” Philosophical Library, New York. Marshall, N. B. (1965). “The Life of Fishes.” Weidenfeld & Nicolson, London. Morrow, J. E., Jr., and Gibbs, R. H., Jr. (1964). Family Melanostomiatidae. In

7.

BIOLUMINESCENCE

399

“Fishes of the Western North Atlantic” (H. B. Bigelow, ed.), Part 4, pp. 351510. Sears Found. Marine Res., Yale Univ., New Haven, Connecticut. Nicol, J. A. C. (1957). Observations on photophores and luminescence in the teleost Porichthys. Quart. J . Microscop. Sci. 98, 179-188. Nicol, J. A. C. (1958). Observations on luminescence in pelagic animals. J. Marine Biol. Assoc. U . K . 37, 705-752. Nicol, J. A. C. (1960a). Spectral composition of the light of the lantern-fish, Myctophum punctatum. J. Marine Biol. Assoc. U.K. 39, 27-32. Nicol, J. A. C. (1960b). Studies on luminescence. On the subocular light-organs of stomiatoid fishes. J. Marine Biol. Assoc. U.K. 39, 529-548. . regulation of light emission in animals. Biol. Reu. CamNicol, J. A. C . ( 1 9 6 0 ~ )The bridge Phil. SOC.35, 1 4 2 . Nicol, J. A. C . (1962a). Animal luminescence. Adwan. Cmnp. Physiol. Biochem. 1, 217-273. Nicol, J. A. C. (1962b). Bioluminescence. PTOC. Roy. SOC. A265, 355-359. Nicol, J. A. C. (1967). The luminescence of fishes. Symp. ZOO^. SOC. London 19, 27-55. Nusbaum-Hilarowicz, J. ( 1923). Etudes d’anatomie comparbe sur les Poissons provenant des campagnes scientifiques de S.A.S. le Prince de Monaco. Rdsult. Camp. scient, Prince Albert I, Part 11, Fasc. 65, pp. 1-100. Ohshima, H. (1911). Some observations on the luminous organs of fishes. I . Coll. Sci., Imp. Uniw. Tokyo 27, Art. 15, 25 pp. Okada, Y. K. (1926). On the photogenic organ of the knight-fish (Monocentris iaponicus Houttuyn). Biol. Bull. 50, 365-373. Parr, A. E. (1927). Scientific results of the Third Oceanographic Expedition of the “Pawnee” 1927. Ceratioidea. Bull. Bingham Oceanog. Collection 3, Art. 1, 1-34. Pearcy, W. G. (1964). Some distributional features of mesopelagic fishes off Oregon. J. Marine Res. (Sears Found. Marine Res.) 22, 83-102. Rauther, M. ( 1941). Einige Beobachtungen uber den Tentakelapparat von Ceratias holbolli. 2001.Anz. 136, 131-140. Ray, D. L. (1950). The peripheral nervous system of Lampanyctus Zeucopsaws. J. MorphoZ. 87, 61-178. Sie, E. H.-C., McElroy, W. D., Johnson, F. H., and Haneda, Y. (1961). Spectroscopy of the Apogon luminescent system, and of its cross-reaction with the Cypridina system. Arch. Biochem. Biophys. 93, 286-291. Skowron, S. ( 1928). Wber das Leuchten des Tiefseefisches Chauliodus Sloanii. Biol. Zentr. 48, 680-684. Steche, 0. ( 1909). Die Leuchtorgane von Anomalops katoptron und Photoblepharon palpebratus, zwei Oberflachenfischen aus dem Malaiischen Archipel. Z. Wiss. 2001.93, 349-408. Strum, J. (1966). Fine structure of the dermal luminescent organs in the fish Porichthys notatus. Anat. Record 154, 429. Strum, J. ( 1968a). Fine structure of the dermal luminescent organs, photophores, in the fish Porichthys notatus. Ph.D. dissertation, University of Washington, Seattle, Washington. Strum, J. ( 1968b). Photophores of Porichthys notatus: Ultrastructure of innervation and changes following adrenalin stimulation. Anat. Record 160, 434-435. Strum, J. ( 1969a). Fine structure of the dermal luminescent organs, photophores, in the fish Porichthys notatus. Anat. Record 164, 433462.

400

J. A. C. NICQL

Strum, J. ( 1969b). Photophores of Porichthys notatus: Ultrastructure of innervation Anat. Record 164, 463-478. Taylor, F. H. C. (1968). The relationship of midwater trawl catches to sound scattering layers off the coast of northern British Columbia. J. Fisheries Res. Board Can. 25, 457-472. Tchernavin, V. V. (1953). “The Feeding Mechanisms of a Deep Sea Fish Chauliodus slonni Schneider.” Brit. Museum (Nat. Hist.), London. Tompsett, D. H. (1939). “Sepia.” Liverpool Marine Biol. Comm. Mem. No. 32. Tsuji, F. I. and Haneda, Y. (1966). Chemistry of the luciferases of Cypridino hilgendorfii and Apogon ellioti. In “Bioluminescence in Progress” (F. H. Johnson and Y. Haneda, eds.), pp. 137-149. Princeton Univ. Press, Princeton, New Jersey. von Lendenfeld, R. (1887). Report on the structure of phosphorescent organs of fishes. Rept. Sci. Results ‘Challenger,’ Zool. 22, 277-335. Waterman, T. H. (1939). Studies on deep-sea angler-fishes (Ceratioidea). I. An historical survey of our present state of knowledge. 11. Three new species. Bull. Museum Comp. Zool. Harvard 85, 65-94. Waterman, T. H. ( 1948). Studies on deep-sea angler-fishes (Ceratioidea). 111. The comparative anatomy of Gigantactis bngicirra Waterman. J . Morphol. 82, 81-149. Wilson, D. P. (1937). The habits of the angler-fish, Lophius piscatorius in the Plymouth aquarium. J. Marine Biol. Assoc. U.K.41, 477496. Yasaki, Y. (1928). On the nature of the luminescence of the kingiish (Monocentris japonicvs) (Houttuyn). J. Erptl. Zool. So, 495-505.

POISONS AND VENOMS FINDLAY E . RUSSELL

. . . . . . . . .

I. Introduction . . . . . . . . . 11. Venomous Fishes . . . . . . . . A. Venom Apparatus . . . . . . . B. Chemistry, Pharmacology, and Toxicology of Fish Venoms 111. Poisonous Fishes . . . . . . . . . . A. Ichthyosarcotoxism . . . . . . . B. Ichthyocrinotoxism . . . . . . . . . Acknowledgments . . . . . . . . . . References . . . . . . . . . . . .

. . . .

. .

401 404 404 414 423 424 438 440 440

I. INTRODUCTION

At least 1000 species of fishes are known or thought to be venomous or poisonous. For the most part, these species are widely distributed throughout the world, and while their numbers may sometimes be quite large in some areas, they do not produce major ecological problems by virtue of their toxicity alone, nor at the present time are they other than a local danger to man’s health and economy. It is generally believed that most of the venomous and poisonous fishes have been identified, although it is conceded that a number of species have not yet been adequately described (Russell, 1965). This review treats of the toxins of most of the important venomous and poisonous piscines. While it is primarily concerned with the chemical and physiopharmacological properties of the toxins, some attention has been given to the structure of the animal’s poisoning apparatus, and the problem of poisoning in man and other animals. Several words and terms peculiar to the study of these toxins (“toxinology”) should be noted. The term “poisonous fishes” is generally applied to those fishes whose tissues, either in part or in their entirety, are toxic. 401

402

FINDLAY E. RUSSELL

The term “venomous fishes” on the other hand, is usually limited to those piscines which are capable of producing a poison in a highly developed secretory organ or group of cells, and which can deliver this toxin during a biting or stinging act. In reality, all venomous animals are poisonous but not all poisonous animals are venomous. Fishes in which a definite venom apparatus is present are sometimes called “phanerotoxic” ( Greek p i v c p i 5 ~ ,evident TO&&, poison), while fishes whose body tissues are toxic are called “cryptotoxic” ( Greek K P U S T T ~ S ,hidden), Although the words “venomous” and “poisonous” are often used synonymously, most toxinologists now try to confine the use of the term “venomous fishes” to those creatures having a gland or group of highly specialized secretory cells, a venom duct (although this is not a constant finding), and a structure for delivering the venom. While in the past there has been a tendency to employ the term “venom apparatus” to denote only the sting or spine used by the fish to deliver its venom, most investigators now use the term in its broader context, that is, to denote the gland and duct in addition to the sting or spine. Poisonous fishes, as distinguished from venomous fishes, have no such apparatus; poisoning by these forms usually takes place through ingestion. Fish poisoning is synonymous with ichthyotoxism. It is meant to exclude that type of poisoning which may occur following the ingestion of fish contaminated by bacterial pathogens. Halstead (1964, 1967) has suggested a further division of the poisonous fishes:

+

( 1 ) Zcthyosarcotoxic fishes-those fishes which contain a toxin within their musculature, viscera, or skin (2) Zcthyootoxic fishes-those fishes which produce a toxin generally confined to the gonads. In these piscines there is a relation between gonadal activity and the production of the toxin. Most members of this group are freshwater species. This subdivision would include those fishes whose roe are poisonous ( 3 ) Zcthyohemotoxic jishes-those fishes which have a toxin in their blood. Some freshwater eels and several marine fishes make up this group (4)Zchthyocrinotoric-those fishes which produce a toxin by glandular secretion, but which lack a true venom apparatus. This group would include such piscines as the boxfishes or trunkfishes, the hagfishes and the lampreys, all of which may produce a toxic substance in their skin and subsequently release this toxin into their environment The toxins of fishes vary remarkably in their chemical and zootoxicological properties. For the most part, they are complex mixtures. Some are proteins, while others are peptides, amines, quaternary ammonium

8.

POISONS AND VENOMS

403

compounds, mucopolysaccharides, lipids, and saponins; however, the structure of most of the marine poisons is still unknown. Some of the marine venoms contain enzymes, but these substances are not nearly as common as they are in the terrestrial animal venoms. The zootoxicological properties of the fish toxins vary as remarkably as do their chemical properties. Some fish venoms provoke rather simple effects, such as transient vasoconstriction or vasodilatation, while others provoke more complex responses, such as parasympathetic dysfunction or multiple concomitant changes in the blood-vascular dynamics. The effects of the separate and combined activities of the fractions of these poisons and of the metabolites formed by their interactions are further complicated by the response of the envenomated organism. The organism may produce and release several autopharmacological substances which may not only complicate the poisoning but also may in themselves produce more serious consequences than the venom. There appear to be some chemical and toxicological similarities in the various fish venoms. This might be expected since the venom apparatus in fishes is primarily a part of the animal's defensive armament, and it would not be inconsistent to propose that there might be a pattern for the development of deterrent substances as part of this armament. All of the fish venoms have one outstanding common biological property: they produce immediate, intense pain. This pain is usually far more severe than that experienced following the bite of a snake or a black widow spider, whose venoms are used primarily in offense-in the gaining of food. The nature of the pain indicates that at least this part of the venom of the stingrays, weeverfishes, scorpionfishes, lionfishes, stonefishes and perhaps certain other fishes, is composed of a common fraction or fractions. The toxins of the ichthyosarcotoxic fishes differ remarkably from the toxins of the venomous fishes. There is no chemical or physiopharmacological relationship between the toxins of the venomous and poisonous fishes. There appears to be some relationship between certain of the toxins of the poisonous fishes, but this is not as clearly evident as the similarities in the venomous piscines. Nevertheless, ciguatera toxin and Gymnothorax toxin appear to be the same; tetrodotoxin is the same poison as that found in the newt Taricha torosa, and while tetrodotoxin and saxitoxin seem to be chemically distinct, they do possess some very similar biological activities; and gonyaulax toxin and saxitoxin, which are now known to be the same, are thought by some to be related to ciguatera toxin. Most of these relationships have been established within the past several years, and it seems likely that with the more definitive studies now being done, the specific chemical structures and modes of action,

404

FINDLAY E. RUSSELL

and possibly the etiology for the toxicity in these fishes, will be found within the next decade. There is no comprehensive classification for the toxins of the fishes. Our knowledge of the properties of these complex substances is not broad enough or consistent enough, at the present time, to permit the adoption of a single working classification. It would seem, in the absence of a useful and reliable method for classifying fish toxins, that it is wisest to use a system based on taxonomy. While such a classification would be somewhat bulky, it might serve our purpose during the interim in which we attempt to organize our data on the chemical and biological properties of these complex substances in a more thorough manner. The study of the fish poisons is further complicated by the fact that qualitative as well as quantitative differences in the toxins may exist not only from species to species but also from individual to individual within the same species. A toxin may even vary in its chemical and zootoxicological properties within the individual animal at different times of the year or under different environmental conditions. Until the fractions of the fish toxins responsible for the deleterious effects have been isolated and studied individually, and in combination, we need to exercise extreme care in systematizing data which are based partly on biological assay methods, partly on biochemical studies, partly on clinical observations, and partly on intuitive hunches.

11. VENOMOUS FISHES

Over 200 species of fishes are known or thought to be venomous. Among the more common of these piscines are the stingrays, weevers, scorpionfishes, zebrafishes, stonefishes, stargazers and certain of the catfishes, sharks, ratfishes, and surgeonfishes. Venomous fishes are most frequently found in the Pacific area, and for the greater part they are shallow-water reef or inshore fishes. Some families, such as the weevers, stargazers, and stingrays, are chiefly benthonic and may sometimes be found in quite deep waters. Most venomous fishes are nonmigratory and slow swimming. They tend to live in a protected habitat in or around corals, rocks or kelp beds, or they spend much of their time buried in the sand. A. Venom Apparatus

While it is not possible within the confines of this chapter to discuss the venom apparatus of every venomous fish, an attempt will be made

8. POISONS

AND VENOMS

405

to describe this structure in the more important, or perhaps better known fishes. In most venomous piscines the venom apparatus is used as a defensive weapon. In some, the use is not evident, although fossil records would seem to indicate that even in these genera the structure is also part of the animal's defensive armament.

1. fh"l?NRAYS The suborder Myliobatoidea includes the families Dasyatidae, whiprays or stingrays; Urolophidae, round stingrays; Myliobatidae, bat or eagle rays; Gymnuridae, buttedyrays; Potamotrygonidae, river rays; Rhinopteridae, cow-nosed rays; and Mobulidae, devil rays or mantas. The sting, or caudal spine, of the Mobulidae, when present, is very rudimentary. In some species of Gymnuridae the spine is also rudimentary, and few venom cells are found. The sting is composed of an inner core of vasodentine and the outer layer of enamel. It is bilaterally serrated with the serrations directed cephalically. There may be one or several stings. In adult Urolophus hahri the sting is approximately 4 cm in length, while in larger species it may reach 40 cm in length. The spine is encased in an integumentary sheath. The venom is contained for the most part in the epidermal tissues within and overlying the two ventrolateral grooves of the spine (Fig. 1). In the stingray U. halkri the integumentary sheath is composed of an inner variable layer of loose areolar connective tissues covered by an outer layer of epithelium. The inner dermal layer is rich in thin-walled blood vessels. Occasionally, it projects into the epidermis, forming delicate papillae which carry nutritive blood vessels. The epidermis is composed of a basal group of flattened, heavily pigmented cells. In places these cells form a distinct pigmented layer. The innermost layer of these basal cells is perpendicularly oriented. As these cells mature, their nuclei become smaller and stain more intensely. A layer of varying thickness of oval, vacuolated (secretory) cells, having a nucleus compressed at one end and eosinophilic cytoplasm, overlies the basal epithelium in many areas. These cells occupy between 20 and 60%of the total tissue area of the epidermis in and over the ventrolateral grooves. The cells have been called glandular by Evans (1916) and Halstead and Bunker (1953). They give positive reactions to several protein tests and are believed to contain the venom. In some areas, narrow columns of modified squamous cells connect the basal layers of the epidermis with the outer layers. An outer layer of partially cornified ceIls with small dark staining nuclei covers the vaculated and squamous cells. On the surface is a thin layer of dense, comified material in which no cell detail remains (Russell and Lewis, 1954, 1956). The above description is similar

406

FINDLAY E. RUSSELL

Fig. 1. Venom apparatus of the stingray, Urobphus h u h % . ( a ) sting; ( b ) cross section through A-B; and ( c ) area C-D showing dentinal spine, areolar connective tissue with blood vessels, and epithelium containing venom-producing cells. From Russell ( 1963).

in most respects to that published by Halstead et al. (1955a) on five genera of marine stingrays. This description, however, is quite different from that detailed by Porta (1905), Evans (1916), and Fleury (1950), who describe the glandular cells as being in the inner or dermal layer of the integumentary sheath, the area we have described as containing the loose areolar connective tissues and blood vessels ( Russell and Lewis, 1956). Fleury states that in Myliobatis aquila the venom is evacuated through one or two “excretory canals or their ramifications in the interdental spaces.” The

8.

POISONS AND VENOMS

407

ducts described by Fleury are very similar to those we have noted as blood vessels. Many of our sections show blood cells in these structures (Russell, 1965). It is not particularly clear from Fleury’s paper where these ducts arise and where they terminate. Castex has contributed a number of valuable studies on the venom apparatus of the freshwater stingrays. His description of the venom secretory cells of Potamotrygon motoro and P . pauckei (Castex and Loza, 1964) is similar to that proposed by Porta (1905) and Fleury (1950). Subsequently, he examined the caudal spine of P . magdalenae (Castex and Suilar, 1965) and presented a description not unlike that presented by Ocampo et al. (1953) and by Russell and Lewis (1956). More recently, he has summarized the findings of his studies and suggests that in the river stingrays the venom cells are probably in the epidermis (Castex, 1967). Pearson ( 1967) has examined the stings of seven species of stingrays found in Australian waters. In all species the venom-producing cells were located in the epidermis of the integumentary sheath. No venomproducing cells were found in the tissues of the dermal layer within the ventrolateral grooves, which Pearson calls the “dermal triangles,” a term this writer feels is preferable to the commonly employed “glandular triangles.” Pearson describes the venom-producing cells as club cells, Iocated for the most part in the epidermis of the ventrolateral grooves of the spine, although, as described by previous workers, some are also found in the dorsolateral and ventrolateral epidermis enveloping the spine. The club cells were found to contain large amounts of proteinaceous materials. Their cytoplasm was either homogeneous or slightly granular, except for a single spherical to rod-shaped cell inclusion and the nucleus. The cytoplasm of the club cells was eosinophilic. Tests for mucous substances were negative; they were weakly positive for lipids, and strongly positive for proteins. Stingings by stingrays occur frequently in humans, but apparently infrequently in marine animals (Russell, 1953; Russell et al., 1958a). Fossil records, however, would seem to indicate that at one time a number of marine animals, particularly sharks, fed upon stingrays and were apparently stung by them. Recently, we found the caudal spines of two small stingrays embedded in the lining of the stomach of a young seal. 2. WEEVERS

The weevers are small marine fishes of the family Trachinidae. The venom apparatus of the weevers consists of the two opercular spines, five to eight dorsal spines, and the tissues contained within the integumentary

408

FINDLAY E. RUSSELL

sheath surrounding each spine. In the greater weever, Trachinus draco, the two dentinal opercular spines extend posteriorly and very slightly downward from near the superior margin of each operculum. Each is firmly attached to the operculum for the proximal one-third of its length, and lies free and superficial to the posterior portion for the distal twothirds. These spines are covered by a thin-walled integumentary sheath which, when removed, exposes the white, soft, spongy glandular tissue along the superior and inferior margins. Figure 2 shows a pin inserted into the midportion of the deep groove along the superior edge of the left opercular spine. A similar groove exists along the inferior margin. Within the superior and inferior grooves, and the two conic cavities into which they pass at the base of the spine, lies most of the spongy glandular tissue that produces the venom (Russell and Emery, 1960). The five to eight dorsal spines are enclosed within individual integumentary sheaths connected by thin interspinous membranes. In a 38-cm T . draco, the first dorsal spine measures approximately 21 mm. Cross sections through the opercular spine near the operculum reveal a peripheral layer of integument continuous with an inner or dentinal layer, and a glandular area between. The integument is composed of an epidermis of stratified squamous epithelium which rests upon a basement

Fig. 2. An opercular spine of Trachinus dram showing pin inserted into superior groove. Note conic cavity at the base of the groove. From Russell and Emery (1960).

8.

POISONS AND VENOMS

409

membrane, which in turn separates the epidermis from the dermis. Large secretory or mucous cells with clear cytoplasm are seen scattered throughout the epidermis. The dermis is composed of a thick layer of dense fibrous connective tissue and is separated from the spine by a space. Projecting into each of the glandular grooves of the spine is a thin layer of dermis and glandular tissue from the adjacent glandular areas, which appears embedded in the surrounding dermis. The glandular parts are made up of masses of large, polygonal, distended cells of variable sizes, staining light pink and filled with finely granular secretions (Halstead and Modglin, 1958; Skeie, 1962a). The integumentary sheath of a dorsal spine is composed of an epidermis of stratified squamous epithelium of five or six cells thick resting on a thick pigmented membrane, a dermis of dense, fibrous connective tissue, and the gland tissue within the anterolateral-glandular grooves. The gland cells are large, polygonal cells of varying sizes, perpendicularly oriented to the axis of the spine, and staining pink. Their cytoplasm is finely granular. In both the opercular and dorsal spines envenomation takes place when the integumentary sheath is ruptured by trauma (Halstead and Modglin, 1958; Russell and Emery, 1960).

3. STARGAZERS The stargazers are small, bottom-dwelling marine fishes of the family Uranoscopidae. The venom apparatus of the stargazer Uranoscopus scaber consists of the two cleithral, or shoulder spines, their venom glands, and the enveloping integumentary sheaths. The spines are sharp, conical, acellular structures of cementumlike material having concentric growth laminae. They protrude from the superior-posterior edge of the operculum above the pectoral fin. The distal portion of each spine is almost round, while the proximal portion tends to be flattened in the medial-lateral plane. Their lengths in a fish measuring 170 mm would be approximately 10 mm. Along the inferior edge is a shallow groove which is lost toward the middle of the spine. A similar, but much less marked groove is found along the superior edge (Fig. 3). The integumentary sheath enveloping the spine has an epidermis composed of a basal layer of columnar cells, stratified squamous epithelium having cuboidal or polyhedral cells interspersed with unicellular mucous glands. The dermis is separated from the epidermis by a prominent basement membrane. It is relatively thick, moderately dense, and contains spaces filled with a gelatinous material. The gelatinous material is believed to contain the venom. Areas of this material are often associated with fibrocytes, although the association is not clearly definable.

410

FINDLAY E. RUSSELL

Fig. 3. Pectoral girdle of the stargazer Uranoscopus scaber showing the cleithral spine and venom gland.

Halstead and Dalgleish (1967) note that in the large spaces within the spine itself they also found a gelatinous material, as well as thinwalled blood vessels. They suggest that “modified fibrocytes either produce the venom continuously, or hypertrophied fibrocytes produce a holocrine secretion which is collected within dermal pockets or cancellous spaces of the bone near the base of the spine.” Passage of the venom to the tip of the spine is accomplished by means of the shallow groove extending along the inferior edge of the spine. When pressure is put upon the sheath, as during the stinging act, the venom-containing cells as well as the sheath rupture, and the venom is forced into the groove of the spine and then to the tip. 4. SCORPIONFISHES

a. Notesthes robusta. This scorpaenid, commonly known as the bullrout, occurs throughout the estuarine and coastal river systems of eastern Australia, particularly in shallow, weed-infested creeks. Its venom apparatus consists of the 15 dorsal spines, three anal spines, two pelvic spines, and the gland and tissues associated with these spines. The first 12 dorsal spines are strongly curved caudally, the remaining ones are only slightly curved. The tips of the dorsal spines are normally exposed when the fin is erect. Both the anal and pelvic spines are directed caudally. Each spine is enclosed in an integumentary sheath which is continuous with the web of the fin.

8.

POISONS AND VENOMS

411

Removal of the integumentary sheath reveals a fusiform venom gland in the paired anterolateral grooves of the spine. These venom glands are straw-colored and show numerous opaque areas when the gland is in the fresh state. They terminate 1-2 mm from the tip of the spine. The mean ratio of gland length to spine length is approximately 1 :3 ( Cameron and Endean, 1966). A cross section through a typical dorsal spine reveals an epidermis composed of a basement membrane, a basal layer of cells, a layer of stratified epithelium five to ten cells thick, and an outer single row of columnar cells possessing fingerlike outgrowths. This epidermis is approximately 60 p in thickness. Within the epidermis, two types of gland cells are found. One is a spherical type with a small peripheral nucleus and a large vacuole, the contents of which are nongranular. These cells are fairly constant in size, averaging 16 p in diameter. When they are found at the outside edge of the epidermis they appear as goblet cells. The second type of gland cell is oval to circular in shape; it possesses a peripheral nucleus and a granular cytoplasm. When circular, these cells average 26 p in diameter, and their granules 0.5 p in diameter. The underlying dermis is four to five times thicker than the epidermis, and is composed chiefly of collagenous connective tissue ( Cameron and Endean, 1966). The two venom glands in the grooves of the spines are closely invested in a capsule of collagenous connective tissue, which also invests the spine itself. The glands are pear-shaped, circular, or oval. When circular they measure approximately 350 p in diameter. A sheet of cells identical in appearance and staining qualities to the stratified cells of the epidermis is found across the center of the gland. On either side of this sheet is a single row of very large, rectangular gland cells (Cameron and Endean, 1966). Between the gland cells, and particularly in the periphery of the gland, occasional granule-filled sacs of approximately 60 p in diameter are found. The granules are spherical and average 1.5 p in diameter. Further study of these cells and sacs by Cameron and Endean (1966) revealed that the gland cells became the granule-filled sacs, and that these sacs tend to push toward the periphery of the gland. There does not appear to be any duct from these glands, either to the tip of the spine or to the base, Expulsion of the venom takes place when the integumentary sheath is torn or retracted along the spine in the direction of its base.

b. Pterois volitans. The zebrafish, lionfish, or turkeyfish is a scorpaenid usually found in shallow waters around coral heads, in underwater caves, and about underwater debris. Because of its brilliant colors and long delicate fins, it is easily seen and rarely contacted accidentally. Its

412

FINDLAY E. RUSSELL

venom apparatus consists of 13 dorsal spines, three anal spines, two pelvis spines, the enveloping integumentary sheaths, and the glandular complex lying within the anterolateral grooves of the spines. The dorsal spines are almost straight, except at the base and the tip where they incline slightly caudally. The anterolateral grooves originate just above the base of the spine and extend the entire length of the structure. In a 45-mm spine they may measure 39 mm. The spine is invested in the integumentary sheath, a thin layer of spectacularly colored, fibrous connective tissue. Within the anterolateral grooves are fusiformshaped “venom glands,” which may occupy three-quarters of the total length of the spine. The glandular structures within the grooves are composed of several types of tissues, including the venom-producing cells, which are large, p l y gonal-shaped cells with pinkish gray, finely granular cytoplasm. According to Halstead et al. (195%) these cells may measure 207 by 75 p or more and are collected together in connective tissue compartments within the anterolateral grooves. They are covered in turn by the integumentary sheath. There does not appear to be any glandular duct through which the venom is secreted or discharged. Envenomation probably occurs through mechanical pressure on the spine and rupture of the sheath.

c. Scorpaenu guttata. The California sculpin, or scorpionfish, is a common inhabitant of the rocky shores and kelp beds of the Southern California coast. It is involved in numerous stingings to fishermen ( Russell, 1965). Its venom apparatus consists of 12 dorsal spines, three anal spines, two pelvic spines, and their enveloping integumentary sheaths. The spines are shorter and heavier than in Pterois, and the sheath is thicker. When the integumentary sheath is removed from a dorsal spine, a slender, elongated, fusiform strand of grayish tissue is found lying within the distal one-half or two-thirds of the glandular grooves on either side of the sting. The venom is contained within this tissue. The “venom glands” are similar to those found in Pterois, although more highly developed. Envenomation occurs through mechanical pressure on the spine, which generally tears the integumentary sheath.

d. Synanceja horridu. There are a number of species of Synanceja, and they vary in length as adults from 22 to 48 cm. They inherit their name from the fact that they so remarkably resemble stone or coral. They have the habit of lying motionless in coral or rock, or partially buried in the sand or mud, and so still can they lie that molluscs and other benthonic marine animals may often crawl over them (Whitley and Boardman, 1929). According to Endean (1961), if approached while they are swirn-

8.

413

POISONS AND VENOMS

Fig. 4. A dorsal spine of the stonefish Synuncefu horridu showing fusiform venom glands. From P. R. Saunders, unpublished data.

ming they will frequently turn themselves so that their dorsal spines point toward the intruder. The venom apparatus of Synanceja horrida consists of 13 dorsal spines, three anal spines, two pelvic spines, the thick, warty integumentary sheaths enveloping these spines, the two glandular grooves in each spine, and the venom glands and their ducts within these grooves. Table I Types of Venom Apparatus in Scorpion Fishes Based on Morphology Structure Fin spines Integumentary sheath Venom glands

Venom ducts

Pterois

Scorpaena

Elongated and Shorter and heavier slender than in Pterois Thin Moderately thick Small and well Larger and more highly developed developed than in Pterois Not evident Not evident

Synanceja Short and stout Very thick Large and highly developed Well developed

414

FINDLAY E. RUSSELL

The dorsal spines are short, stout, and straight. The two anterolateralglandular grooves extend almost the entire length of the spine, The glands appear as two large fusiform masses at the middle and toward the tip of the spine (Fig. 4). The distal end of each gland terminates in a ductlike structure lying within the grooves. These ducts extend from the glands to the tip of the spine. Table I gives the gross differences in the structures of the three types of scorpionfish dorsal spines.

B. Chemistry, Pharmacology, and Toxicology of Fish Venoms As noted in the introductory remarks in Section 11, A, it is not possible within the confines of this chapter to consider all of the venomous fishes. Thus, a few representative species have been selected, and the present knowledge on the venoms of these species will be reviewed.

1. ! h N G R A Y S a. Chemistry. The principal difficulty encountered in the various studies on the chemistry of stingray venoms has been the fact that since the animal has no true venom gland, and since the venom has to be obtained by macerating the venom-containing tissues, it has not been possible to obtain a pure toxin. In most studies, all of the tissues surrounding the spine have been removed, macerated, and put into solution. In some studies, only the integumentary sheath has been used, while in others the spines with the tissues intact have been first lyophilized then macerated and put into solution. Extracts from fresh tissues are more toxic than those from tissues which have been stored, even at very low temperatures. Individual stings should be frozen immediately on dry ice and acetone following their removal from the fish. They should then be stored at -6OOC if they are to be kept for long periods of time. Even at this temperature it appears that some degree of toxicity is lost with the passing of several years. Once the tissues are macerated and put into solution the toxin becomes unstable, and some such extracts loose their lethal properties within hours. Others may remain relatively stable for days, and in the presence of 10-20% glycerol or certain other substances they are more stable (Russell and van Harreveld, 1954; Russell et al., 1958a,b; Russell, 1953, 1965, 1967a, 1969). Freshly prepared water extracts of the tissues from the ventrolateral grooves of the elasmobranch's sting are clear, colorless, or light gray in color and have a pH of 6.7. They have a fishy taste and ammoniacal odor. Total protein was found to be approximately 30%, total nitrogen

8.

POISONS AND VENOMS

415

3%,and carbohydrate 3%.A number of amino acids have been identified by paper chromatography. It has been suggested that the toxin is a protein of average molecular weight ( Russell et al., 195813). Using disc electrophoresis, we have identified 15 fractions in extracts from the venom-containing tissues of Urolophus halleri. Further studies on these extracts, using gel filtration (Sephadex G-100 and G-200), suggested that the toxic protein( s ) may have a molecular weight in excess of 100,000. The fraction showing the greatest lethal activity was found to have 2 or 3 distinct bands when subjected to disc electrophoresis. Further studies have shown that the venom contains serotonin, 5nucleotidase, and phosphodiesterase, but no protease or phosphoIipase activity has been demonstrated (Russell, 1965). The activity of the venom is lost at pH 4.80 but is retained between 5.20 and 8.60. The toxic protein( s ) of stingray venom can be concentrated by pervaporation in a cellophane dialysis membrane and by ultrafiltration in a collodion membrane. Activity is correlated with an increase in the optical density at both 260 and 280 mp. Moreover, the toxic protein( s ) is (are) selectively retained by DEAE Sephadex in 0.05 M tris-HC1 ( pH 8.60) and can be eluted by increasing the molarity of the buffer by the addition of NaCl to a 0.20 M equivalent. This ion exchange procedure, when carried out by a stepwise elution with increasing NaCl molarity, has been used not only to concentrate the toxic principle but also to eliminate large quantities of contaminants. The toxic principle can also be separated from other components of the crude extract by gel filtration on Sephadex G-200. Activity is associated with peak I1 in most runs. b. Pharmacology and Toxicology. Stingray venom is known to exert a deleterious effect on the mammalian cardiovascular system. Low concentrations of the venom give rise to simple vasodilatation or vasoconstriction. The most consistent change seen in the electrocardiographic pattern of cats when small amounts of the venom are injected is bradycardia with an increase in the PR interval giving a first-, second-, or thirddegree atrioventricular block. Reversal of the small dose effect occurs within 30 sec following the end of the injection. Cats receiving larger amounts of the venom show, in addition to the PR interval change, almost immediate ST, T-wave change indicative of ischemia and, in some animals, true muscle injury. High concentrations appear to cause marked vasoconstriction of the large arteries and veins as well as the arterioles. Much more serious is the direct effect on the heart muscle. The venom produces changes in heart rate and amplitude of systole, and it may often cause complete, irreversible cardiac standstill. It is apparent that the venom affects the normal pacemaker. The new rhythm evoked following cardiac standstill is often irregular and appears to be elabor-

416

FINDLAY E. RUSSELL

ated outside the sinoatrial node (Russell and van Harreveld, 1954, 1956; Russell et al., 1957). While small doses of the venom may cause some increase in the respiratory rate, large doses depress respiration. Part of this depression is secondary to the cardiovascular changes, but the venom may also have a direct effect on the respiratory centers. The toxin may provoke changes in behavior. In mammals the venom occasionally produces convulsive seizures. These may be due in part to the cardiovascular hypotension which provokes a cerebral anemia. The direct effect of the venom on the central nervous system is far less deleterious than that produced by snake and arthropod venoms (Russell and Bohr, 1962; Russell et al., 1958a). The venom has little or no effect on neuromuscular transmission ( Russell and Long, 1961) . Mice injected with a lethal dose of stingray venom develop hyperkinesis, prostration, marked dyspnea, blanching of the ears and retina, and exophthalmos. These are followed by complete atonia, cyanosis, gasping respiratory movements, coma, and death. In cats and monkeys a similar pattern has been observed. The LD,, in mice has been calculated as 28.0 mg/kg for crude extracts of all the tissues from the ventrolateral grooves of the sting (Russell et al., 1958a,b). However, we have found that the peak I1 portion of the Sephadex G-200 fraction has an intravenous LD,, in mice of approximately 2.9 mg of protein per kilogram of body weight ( Russell, 1969). 2. WEEVERS

a. Chemistry. Gressin’s thesis (1884) contains many interesting observations on the weeverfishes and their venom. He obtained a relatively pure venom by merely pressing on the base of the opercular spine, working his fingers up the spine and drawing off the “drop of liquid from the tip with a syringe. The product was clear and light blue in color, contained several different types of cells, and was coagulated by strong acids and bases, and by heat. Pohl (1893) found that the toxin was inactivated by 25 and 95%alcohol, and by ether and chloroform. Russell and Emery (1960) analyzed extracts from the venom-containing tissues of Trachinus vipera and T . draco. The freshly prepared extracts had a fishy taste and ammoniacal odor; the pH was 6.78. The protein content of the extracts of macerated tissues was 21.78, nitrogen 6.9%ether-soluble lipids 3.58, and carbohydrates 20.48; sulfur was absent. Assays of the dialyzed and nondialyzable solutions indicated that the lethal portion was nondialyzable. Carlisle (1962) found that the dialyzable fraction of the venom pro-

8.

POISONS AND VENOMS

417

duced the stabbing pain characteristic of the whole venom. The nondialyzable fraction failed to provoke the pain but did produce a rise in pulse rate and some respiratory distress. He concluded that the systemic effects of the toxin were due to the nondialyzable fraction, while the pain was a consequence of some constituent of the dialyzable part of the venom. He also found that the dialyzable fraction contained a large amount of 5-hydroxytryptamine, and that it was associated with a substance of low molecular weight which acted as a histamine releaser. The nondialyzable fraction contained a neutral amino polysaccharide lacking in sulfur, and two separate albumins. Carlisle suggested that these three substances may not exist separately but may “represent a complex muco-substance of combined polysaccharide-protein nature.” Haavaldsen and Fonnum ( 1963) separated three protein fractions by electrophoresis from the venom of T . dram. On a paper chromatogram they obtained two spots, one of which was identified as histamine and the other as a catecholamine. Photofluorinietric studies revealed the presence of adrenaline and noradrenaline in high concentrations. The toxin also showed considerable cholinesterase activity; it did not contain 5-hydroxytryptamine, lecithinase, or phosphodiesterase. Many of the earlier chemical and physiopharmacoIogica1 properties of weever venom were confirmed by Skeie (1965). He found the venom to be nondialyzable, the dry weight to be 13.2 mg/ml of venom, the total nitrogen 2.1, and the protein nitrogen 1.58 mg/ml of venom. He calculated the total protein to be 75%of the dry weight of the crude venom. This last figure would indicate that the protein content of the venom is more than three times greater than that of the protein obtained from the macerated tissues containing the venom. Paper electrophoresis of a partly purified toxin demonstrated three principal fractions. However, toxicity tests were not carried out on these fractions. Subsequent fractionation studies by density gradient zone electrophoresis gave three major peaks and three lesser peaks, absorbing at 254 mp. The most lethal portions were contained in fractions 14-19, the area of the second and third major peaks, where 85%of the total toxicity was found. The most toxic sample, which was associated with peak 2, contained approximately 25%of the protein of the original crude venom ( Skeie, 196213). b. Pliarmacology and Toxicology. In 1874, Schmidt carried out one of the first important studies on the toxicological properties of weeverfish venom. While some of his work was done with preserved stings, several experiments were conducted with fresh materials. Gressin ( 1884) observed that in frogs and rats the venom he obtained by the milking procedure produced inactivity, respiratory changes, prostration, convul-

418

FINDLAY E. RUSSELL

sions, and sometimes death. Autopsy revealed ecchymosis and necrosis at the injection site, and congestion of the heart, kidneys, liver, and brain. He also noted that potassium permanganate had no effect on the action of the venom. Phisalix (1899) found that glycerine extracts of tissues from the opercular and dorsal spines produced paralysis, local swelling and necrosis, and death when injected into guinea pigs. Briot (1902) demonstrated in vitro hemolytic activity for the venom. He found it had no effect on the blood coagulation time. Glycerine extracts were found to be lethal to rabbits. Briot attributed the lethal effect of the toxin to respiratory paralysis and suggested that the action was peripheral rather than central. He also demonstrated the necr3tic effect of the toxin. He succeeded in immunizing rabbits against the lethal effect of the venom and found that the sera of immunized rabbits would protect other rabbits against the venom. In subsequent studies (Briot, 1903, 1904), he prepared glycerine extracts of the opercular and dorsal spines and found the extracts of the former to be much more dangerous. He also studied the kinase activity of the toxin. Using venom extracted by syringe from the opercular spines of freshly caught Trachinus draco, Evans (1907) demonstrated a fall in blood pressure concomitant with an increase in heart contractions and respiratory changes, when the toxin was injected into rabbits and cats. His studies on the hemolytic effects of the venom yielded results quite different from those obtained by Briot. An increase in central nervous system permeability to potassium following injection of weeverfish venom into frogs WBS observed by De Marco (1936). In subsequent experiments, De Marco ( 1938) demonstrated a more rapid exhaustion of the frog’s gastrocnemius in the presence of the toxin, but no significantly deleterious effect that might be attributed to the direct action of the venom on the muscle. Mareti6 (1957) found that guinea pigs stabbed with weeverfish spines became restless and noisy, and subsequently developed a paresis of the hind legs. Other than one animal which developed tachycardia, he did not observe any significant systemic effects as a result of the poisoning. Local swelling occurred at the site of the injury. Russell and Emery ( 1960) studied the physiopharmacological effects of extracts of the venom-containing tissues taken from the spines of Trachinus vipera and T.draco. These extracts had no deleterious effect on mammalian neuromuscular transmission. When large amounts were added to the nervemuscle bath, there was some shortening of the muscle and a very gradual depression of both the directly and indirectly elicited

8.

POISONS AND VENOMS

419

contractions, but at no time was there any evidence of a differential that could be attributed to changes at the neuromuscular junction or at the nerve. In cats, the venom produced a precipitous fall in systemic arterial pressure concomitant with changes in the pulse pressure, cardiac rate, pulmonary arterial pressure, venous and cisternal pressures, respiration, and the electrocardiogram and electroencephalogram. The findings are similar to those produced by stingray venom and certain of the venoms of other fishes. The electrocardiograms demonstrate that the venom can produce both changes in rhythm and injury to the heart muscle. The fall in pulmonary artery pressure indicates either a failure of the heart to maintain an effective stroke volume or a decrease in pulmonary resistance. The latter seems unlikely; studies of pulmonary artery blood flow using a gated sine-wave electromagnetic flowmeter indicate that the blood flow in this vessel is reduced during the period of decreased pressure. The findings of lowered pulmonary arterial pressure and flow, a decrease in heart rate with various degrees of auriculo-ventricular block, and evidence of heart muscle injury may be interpreted to mean that some degree of cardiac failure is probably responsible for the fall in systemic arterial pressure and the rise in venous and cisternal pressures. Smaller amounts of weeverfish venom produced transient vasoconstriction or vasodilatation, depending upon the quantity injected. With these doses there is little or no deleterious effect upon the heart, although there may be some changes in cardiac rate and in respiratory rate. The depression in central nervous system activity seen following the intravenous injection of a lethal dose of the toxin can be attributed to ischemic anemia produced by the lowered systemic arterial pressure. The wave pattern is typical of that which occurs during cardiovascular failure from any of a number of causes. However, this finding by itself does not exclude the possibility that the venom might have a direct effect on central nervous system activity. A technique for extracting the venom from opercular spines has recently been described by Skeie (196213). The method is similar to that which has been employed for extracting venom from certain other fishes. Using this technique Skeie extracted the toxin from 600 weevers into 60 ml of solution. He found the DML,,, (LD,,) in two mice to be 0.0004 ml. On the basis of this formula he calculated that each weever contained sufficient venom to kill 250 mice weighing 17 g each. In studies upon mice with 19 batches of weever toxin he found the average venom con. tent to vary from 6 to 1066 LD,,, per weever, or from 640 to 2560

420

FINDLAY E. RUSSELL

LD,,,/ml venom. He also found that it is possible to quantitate the venom on tissue cultures and by studies on skin reactions in guinea pigs.

3. SCORPIONFISHES a. Chemistry. Freshly prepared water extracts of the integumentary sheaths and underlying glandular tissues from dorsal spines of Pterois volitans were found to be somewhat turbid, reddish orange in color, and had a pH of approximately 7.0. They were unstable at room temperature and were nondialyzable. Mean protein content ( calculated for the original volume of extract prior to the addition of glycerol) was 0.29%. Lyophilized samples reconstituted to the original volume of the extract possessed a mean protein content of 0.55%and a mean nitrogen content of 0.095%(Saunders and Taylor, 1959). The toxic extract from macerated tissues taken from the dorsal spines of Scorpaena guttata was found to be opalescent, of neutral pH, very unstable, and nondialyzable. Treatment with a chelating agent ( EDTA) , an antioxidant (reduced glutathione), and a proteinase inhibitor ( diisopropyl flurophosphate ) did not preserve the extract’s toxicity. Partial purification of the extract was achieved by salting-out with ammonium sulfate, and by DEAE-cellulose column chromatography. With the latter method, purification of the toxin to 10 pg protein per LD,, was sometimes obtained. A toxic fraction was also obtained adjacent to the origin of the small pore polyacrylamide gel in which the venom extract was resolved by disc electrophoresis. The lethal fraction is thought to be a high molecular protein and moderately negatively charged at pH 7.4 ( P. B. Taylor, F. E. Russell, and P. R. Saunders, unpublished data). The fresh venom of Synanceja horrida is clear and colorless or bluish in color, and it may become opalescent and cloudy after the fish dies. Bottard (1889) noted that it could be coagulated by nitric acid, alcohol, ammonium, and heat. On microscopic examination he found the venom to appear like an albuminous liquid containing large refractile round cells with a single, small, centrally placed nucleus. Saunders ( 1959, 1960), Saunders and TOk& (1961), and Saunders et al. (1962) found the venom to have a pH of 6.8, a nitrogen content of 2%,and a protein content of approximately 13%. The lethal fraction was nondialyzable. Lyophilized, glycerol-treated extracts and extracts kept at -20°C were relatively stable, but extracts held at room temperature soon lost their activity. On starch gel electrophoresis the venom gave seven to ten bands. Material lethal to mice was recovered from only one of these bands. Reelectrophoresis of this lethal fraction gave a single band.

8.

POISONS AND VENOMS

421

Wiener (1959) found that the average yield of venom from a functional dorsal spine of Synanceja trachynis was 0.03 ml, or 5.1-9.8 mg of the dried venom. Lyophilized venom stored in a desiccator for several months did not lose its lethal activity. The venon gave all the reactions of a protein; it was destroyed on heating and inactive at a pH below 4, as well as at a high pH. At 4"C, extracts lost their toxicity in 48 hr. More recent studies on the venom of the stonefish have shown that the toxin can be purified by several methods. The most successful procedure was starch gel electrophoresis of the crude material, followed by recovery of the lethal fraction from one of the zones by a second electrophoresis into a buffer solution. Using this technique, a 10-fold increase in purification, as measured by lethality, was obtained. Ultracentrifugation led to the appearance of two major peaks. The lethal material appeared in the more rapidly sedimenting fraction (Deakins and Saunders, 1967).

b. Pharmacology and Toxicology. Some of the biological activities of scorpaenid venoms are shown in Table 11. The data are taken from a review (Russell, 1969), and are based on the studies by Duhig and Jones ( 1928a,b), Duhig ( 1929), Gail and Rageau ( 1956), Saunders ( 1959, 1960), Saunders and Taylor ( 1959), Wiener ( 1959), Saunders and TokikBs (1961), Austin et al. (1961), Saunders et al. (1962), and Russell ( 1965) . Saunders and Taylor (1959) found that the nondialyzable toxin of P. volitans had an intravenous LD,, of 1.1 mg protein/kg test animal body weight. Intravenous doses of the venom in mice produced ataxia, circling movements, and partial or complete paralysis of the legs. This syndrome was followed by a period of inactivity during which there was evidence of muscular weakness. According to the authors, the apparent skeletal muscle weakness was more pronounced following poisoning with the venom of Pterois than that observed following poisoning by the venom of Synanceja. Approximately 2500 LD,, for mice were calculated to be present in each extract prepared from the combined tissues of the dorsal spines of a single Pterois. The physiopharmacological properties of the venom from Scorpaena guttata is similar in many respects to that obtained from Pterosis, as can be seen from Table 11. Extracts of the macerated tissues from the spines of a single fish many yield up to 4230 calculated mouse LD,,. The mouse LD,, of a specific extract ranged from 50 to 100 pg of protein. The painproducing effects of the venom of Scorpaena, as well as the lethal effects, are retained on dialysis. The venom has also been found toxic to the tidepool fish Clinocottus analis australis.

Table I1 Some Properties of Scomaenid Venoms Pterosis

Synanceja

Scorpaena

Small dose

Decreased arterial pressure Minimal ECG changes Increased respiratory rate Muscular weakness in mice

Decreased arterial pressure Minimal ECG changes Increased respiratory rate Tremor

Slight decrease in arterial pressure Increased then decreased venous pressure Minimal ECG changes Increased respiratory rate with decreased respiratory excursions

Medium dose

Marked fall in arterial pressure Myocardial ischemia, injury or conduction defects Increased respiratory rate

Marked fall in arterial pressure Myocardial ischemia, injury or conduction defects Increased respiratory rate

Partial paralysis of legs in mice

Muscular weakness in mice Tremor

Fall in arterial pressure Myocardial ischemia, injury or conduction defects Changes in venous and cerebrospinal fluid preseures Increased respiratory rate with decreased respiratory excursions Muscular weakness in mice

Precipitous, irreversible fall in systemic arterial pressure Extensive ECG changes Markedly decreased respiratory rate -+ cessation Complete paralysis of legs in mice Intravenous LDw mice, 1.1 mg protein/kg body weight

Precipitous, irreversible fall in systemic arterial pressure Extensive ECG changes Markedly decreased respiratory rate -+ cessation Some paralysis of legs in mice Possible neuromuscular junction changes Produces tremors, convulsions, marked muscular weakness, coma; myotoxic Intravenous LDw mice, 200 pg proteinfig body weight

Lethal dose

Precipitous, irreversible fall in arterial pressure Extensive ECG changes Markedly decreased respiratory rate ---t cessation Some paralysis of legs in mice Intravenous LDN mice, in excess 2.0 mg protein/kg body weight

3

3ce +

M

m

s v)

E

8.

POISONS AND VENOMS

423

111. POISONOUS FISHES

A few of the more important works on the poisonous fishes and a discussion of the data on the chemistry and biological effects of these interesting toxins are given below. Excluding those fishes which become toxic because of bacterial pathogens or the products produced by them, or by the process of tissue decay, approximately 700 species of fishes are known to be poisonous. Most of these species are found in the coral reef belt. They tend to occur in greater numbers around islands than along continental shores, although their distribution is spotty even in a particular part of the ocean or around an island. Most poisonous fishes are nonmigratory reef fishes. They may be either herbivores or carnivores. Some species have tissues which are toxic at all times; other species are poisonous only at certain periods, or in certain areas, while still others have only specific tissues or organs which are toxic, and the toxicity of these may vary with time and place (Russell, 1965). Early writers wea-e inclined to attribute fish poisoning to a single common factor. Thus one finds that numerous “causes” for ichthyotoxism are described in the early literature. Among the more often mentioned are : ( 1 ) Feeding on toxic or nontoxic marine plants. In the latter case it was suggested that the metabolic processes of the fish were able to alter certain plant components into a toxic form ( 2 ) Feeding on toxic protistan, particularly dinoflagellates ( 3 ) Feeding on shoreline terrestrial plants whose berries or leaves have been swept into the sea ( 4 ) Pollution of the ocean by industria1 poisons ( 5 ) The dumping of large quantities of metals, particularly copper, into the ocean, such as occurred in the Pacific at the end of World War I1 ( 6 ) Epidemics in fishes, giving rise to toxic endogenous substances ( 7 ) Spawning activities ( 8 ) Feeding on corals ( 9 ) Feeding on jellyfishes (10) Feeding on palolo worms (11) Feeding on toxic molluscs (12) Feeding on zooanthellae in corals (13) Feeding on other toxic fishes ( 1 4 ) The dumping of radioactive materials into the ocean

FINDLAY E. RUSSELL

(15) Climatic changes (16) Leaving the fish exposed to moonlight It is not a purpose of this chapter to review or discuss the factors which contribute to ichthyotoxism. The reader is referred elsewhere for a detailed survey of this problem (Hiyama, 1943; Fish and Cobb, 1954; Halstead, 1958, 1967; Randall, 1958; Russell, 1965; Pigulevsky, 1966; Komalik, 1967).

A. Ichthyosarcotoxism This type of poisoning is caused by the ingestion of fishes containing a poison within their musculature, viscera, skin, or body slime. It is generally identified with the kind of fish involved: cyclostome, elasmobranch, chimaera, ciguatera, tetraodon, scombroid, clupeoid, and gempylid. Hallucinatory fish poisoning is also identified with this type of poisoning. The toxin is distinctly different in some of these fishes (ciguatera and tetraodon), while it is the same or very similar in others (ciguatera, gymnothorax, and clupeoid) .

1. CYCLOSTOME POISONING This vague entity has been described in the early literature (Chevallier and Duchesne, 1851), but it is difficult to separate the clinical cases of possible cyclostome poisoning due to the ingestion of lampreys or hagfishes from gastrointestinal disturbances precipitated by the bacterial contamination of these fishes. Prokhoroff ( 1884) stated that cyclostome poisoning from the eating of lampreys occurred frequentIy in the Yambu District of Russia. Other references to poisoning by these fishes are recorded by Halstead (1967). The slime of the hagfishes is reported to be toxic, and some authors have attributed the toxicity of these fish to slime which has been improperly removed prior to eating (Linstow, 1894; Engelsen, 1922). Nothing is known of the chemical and pharmacological nature of the suspected toxin from cyclostomes. It is thought by some to be an amine, but this seems unlikely. No definitive study on the musculature, skin, or slime of these fishes has been carried out (Kaiser and Michl, 1958). The buccal secretions of certain lampreys contain an anticoagulant ( Gage and Gage-Day, 1927).

2. ELASMOBRANCH POISONING As with cyclostome poisoning, intoxication following the ingestion of shark flesh has been noted in the early literature (de Sauvages, 1758), but again it is dficult to distinguish between true cases of ichthyotoxism

8.

POISONS AND VENOMS

425

and those of food poisoning. Coutaud (1879) described seven cases of poisoning following the eating of the liver of an unidentified shark off New Caledonia in 1873. Cooper (1964) also noted that the liver of the tiger shark, white-tipped lagoon shark, and perhaps other large sharks may be toxic and produce symptoms and signs not unlike those of ciguatera poisoning, although perhaps more severe and of quicker onset. Deaths from eating shark livers have been reported as recently as 1957, 1960, and 1961 in the Gilbert Islands. Jensen (1914, 1948) has reported on intoxications in dogs and man following the eating of fresh shark flesh (Somniosus microcephulus) in Greenland. In lethal poisonings the agonal period was followed by tonic and clonic convulsions and death (Bgje, 1939). B@jesuggested that the toxic material may be related to choline. Halstead and Schall (1958) found the liver and gonads of the gray reef shark Garcharhinus menisorrah of the Line Islands to be toxic on intraperitoneal injection into mice. Li (1965) stated that extracts of the liver of the gray shark provoked similar pharmacological responses to those produced by extracts from the red snapper Lutjanus bohar. He concluded that the toxic material was a cholinesterase inhibitor. This conclusion rests upon questionable deductions and experimental data which are not easy to interpret. Several skates, rays, and ratfishes have also been implicated in poisonings to man (Phisalix, 1922; Halstead, 1967), but again the experimental and cIinical pictures are not clear. It is possible that in the species of skates and rays found in tropical waters, the toxin, which appears to be present in greater quantities in the viscera than elsewhere, may be similar to shark liver toxin, and may be related to ciguatoxin. Aqueous extracts of the oviducts of a large specimen of the chimaera Hydrolugus colliei were found to be toxic by Halstead and Bunker (1952), but nothing is known of the chemical or toxicological properties of chimaera poison. It is obvious that “elasmobranch poisoning” is a complex entity and that the etiological factors which contribute to this intoxication may be quite different. The toxin involved in shark liver poisoning in Greenland appears to be different from that found in the flesh of sharks from certain tropical islands. Skate and ray poisoning in the tropical and subtropical oceans appears to produce similar clinical manifestations as shark liver poisoning, and the two may be related to ciguatera poisoning. Chimaera poisoning seems to be a completely different entity. 3. CIGUATERA POISONING The term “ciguatera poisoning” is applied to a type of intoxication characterized by certain gastrointestinal-neurological manifestations. It

426

FINDLAY E. RUSSELL

may occur following the ingestion of any of a considerable number of tropical reef and semipelagic marine fishes, including the barracudas, groupers, sea basses, snappers, surgeonfishes, parrotfishes, jacks, wrasses, toadfishes, triggerfishes, herrings, mullets, and eels ( Russell, 1965). Halstead (1967) lists over 440 species of marine fishes which have been implicated as being ciguatoxic. This type of poisoning may also be found in certain marine echinoderms, molluscs, and arthropods. Since almost all of the fishes noted above are normally edible, and many are valuable food fishes in some parts of the world, ciguatera poisoning is not only the most common but also the most treacherous form of ichthyotoxism. The majority of the fishes implicated in ciguatera poisoning are reef or shore species; a few are open-water forms. Almost all of the toxic species are found between latitude 35”N and 34”s. They are usually bottom-dwellers, although seldom found below depths of 200 feet. They are more likely to be found in lagoons than along seaward reefs. Most toxic species feed on benthic algae, benthic fishes, or some other benthic organism. According to Randall (1958), none appear to be plankton feeders as adults. Among the carnivorous species there appears to be a positive correlation between the amount of fish in the diet and the degree of toxicity. There is also a tendency for the larger fish of a species to be more toxic than the smaller fish of the same species. In most cases the flesh is less toxic than the viscera. The liver is usually the most toxic part of the animal, although the reproductive organs may be equally as poisonous. It seems certain that ciguatera poisoning is associated with the foodchain relationship of the fish. Chisholm (1808) appears to have been one of the earliest writers to have seriously considered the possibility that fish poisoning might be a result of something in the piscine’s diet. He notes that certain fish became toxic when they fed upon a “sea moss (corallina opuntia) .” R. Matsuo ( 1934) also noted that fishermen in the Marshall Islands believed fish were poisonous because they fed upon “seaweeds or corals.” This belief is held by many fishermen in the tropical Pacific. More recently, Hiyama ( 1943), Halstead ( 1951, 1953, 1967), Russell (1952, 1965), Halstead and Bunker (1954), Dawson et al. (1955), Randall ( 1958), Helfrich and Banner ( 1963), and Cooper ( 1964) have all attributed ciguatera poisoning to the food-chain relationship. The poison probably originates in a benthonic organism. Most investigators feel that the source of toxin is marine algae. Algae which have been implicated are Lyngbya mu~uscula(Dawson et al., 1955; Helfrich et al., 1968) and Plectonema terebrans (Cooper, 1964; Helfrich et al., 1968). Other Cyanophytae may no doubt sometimes be involved, and indeed perhaps other algae (Habekost et al., 1955; Banner, 1967). It is

8.

POISONS AND VENOMS

42'7

thought that the poison in the algae or some product of the algae is transvectored from the algae to herbivorous fish, which in turn is eaten by man, or that the herbivorous fish is eaten by a carnivorous species which is subsequently eaten by man. It may be that the end poison is a complex of toxins, or perhaps there are several toxins that can be present singly or in combinations. Since certain marine invertebrates ( echinoderms, molluscs, and arthropods) are also known to contain the toxin, or a substance very similar to the toxin, it may be that in some cases the poison is transvectored through these animals to carnivorous fishes and hence to man. It is also possible that zooxanthellae may be a source of the poison. It is known that nontoxic fish fed a diet of toxic Lutjanus bohar may become toxic ( Helfrich and Banner, 1963). a. Chemistry. Much of the early work on ciguatera toxin led investigators into rather unfruitful areas. Among the problems on which considerable confusion arose was that of the solubility of the active principle. By 1960, ciguatoxin was described as being heat stable, stable to drying, and soluble in ether, chloroform, acetone, methanol, and 90%ethanol. It was found to be insoluble in water and dilute acids, and it was dialyzable. It was thought to be a lipid (Hashimoto, 1956; Banner and Boroughs, 1958; McFarren and Bartsch, 1960). Banner et al. (1960) attempted various extraction methods using ethanol, Soxhlet extraction with alcohol, and column chromatography. Hessel et al. (1960)found the toxin from Lutjanus bohar to be soluble in ethyl acetate, isopropyl alcohol, methylene chloride, methyl ethyl ketone, and carbon tetrachloride. Further experiments indicated that the active portion probably did not contain acid or basic groups. They prepared an acetone extraction of the toxic musculature, concentrated the residue, extracted with ether, evaporated the ether, and extracted the residue with acetone. The light yellow, mobile oil was then emulsified into frog Ringer's solution and the material assayed on the frog sciatic nerve. Subsequently, Hessel (1961) prepared seven fractions of the toxin by various methods of fractionation. A partially purified product was obtained by dissolving the toxin in warm methanol and precipitating the nontoxic contaminants by cooling to -20°C. The toxin was recovered from the methanol by evaporation, and subsequent fractionation was carried out by silicic acid column chromatography. Four fractions were collected. These were assayed by feeding experiments with cats, by intraperitoneal injections of aqueous emulsions into mice, and by studies on the action potential of excised frog sciatic nerve preparations. The studies indicated that the toxin could be carried through the seven

428

FINDLAY E. RUSSELL

extraction-fractionation processes. However, the last two fractions were the only fractions that showed appreciable toxicity. From this work, Hessel concluded that the toxic component of ciguatera poison probably contains more than one substance, and that these substances possess polar characteristics, and that they are probably not phospholipids. Banner et al. (1963b) also proposed an extraction method which gave them a concentrated product for further toxicological studies. In 1967, a group at the University of Hawaii isolated and described the chemical nature of ciguatoxin. The raw flesh of the toxic fish was treated with acetone at 20°C. The acetone extract was concentrated and distributed between ether and water. The ether-soluble portion was concentrated and taken up in acetone at -20°C. The precipitate was removed, the acetone solution concentrated, and the concentrate taken up in methanol. The insoluble portion was removed and the methanolic solution concentrated, then extracted with n-hexane. The inactive hexane phase was discarded and the methanol solution concentrated. The toxic produce was subjected to column chromatography on silicic acid at 5°C and elution with chloroform to remove the inactive lipid material. Further purification was achieved by repeated preparative thin layer chromatography at 5°C on silica gel. Approximately 1.5 mg of the methanol-soluble toxic oil was obtained from every 2 kg of raw flesh. Ciguatoxin prepared in this manner is a yellowish, viscous oil, relatively unstable and losing its toxicity on contact with air and light, and to some extent even when stored in chloroform in the dark at -20°C. At present, it is not known whether or not ciguatera poison is a single substance with a tendency to decompose or several closely related substances. It would appear that it is composed of more than one substance. Combustion data indicate the probable empirical formula of C,,H,,NO,, if there is only one nitrogen atom. It has infrared bands at 2924, 2849, 1460, and 1379 cm-l, and a large nuclear magnetic resonance peak centered at 1.25 ppm. Ciguatoxin is considered to be a lipid containing a quaternary nitrogen atom, one or more hydroxyl groups, and a cyclopentanone moiety (Scheuer et nl., 1967). b. Pharmacology and Toxicology. The partially purified extract of ciguatera toxin prepared by Banner et al. (1960) was found to be lethal at 200 mg/kg body weight when injected intraperitoneally into mice. When administered intravenously into rabbits, it produced an immediate fall in blood pressure with a simultaneous increase in respiratory rate and depth. Temporary changes were noted in the electrocardiogram during the period of hypotension (Banner et al., 1963a). In cats, similar changes were observed by us following administration of the toxin. In small doses the poison produced a fall in systemic arterial pressure concomitant with

8.

POISONS AND VENOMS

429

an increase in respiratory rate, and transient changes in the electrocardiogram and electroencephalogram. With large doses there was a more precipitous fall in systemic arterial pressure and severe respiratory distress, which sometimes led to cessation of respirations. The electrocardiogram reff ected changes varying from prolongation of the PR interval or ST T segment changes to a third degree block. The changes in the electroencephalogram were indicative of a decreased blood supply to the brain. The toxin did not appear to have a direct, deleterious effect on those areas of the brain where the conventional lead electrodes were placed ( Russell, 1965). Hessel et al. (1960) found that the partially purified product they isolated had an inhibitory effect on the action potential of the frog sciatic nerve preparation, and that this effect could be correlated with the oral toxicity test in the cat. Subsequently, two of the seven fractions isolated by Hessel (1961) were found to have toxic activity when assayed in feeding experiments, on injection into mice, and when tested on the frog sciatic nerve preparation. Further studies by Banner and his colleagues (1963b) indicated that the toxin appeared to have an effect on mammalian neuromuscular transmission, or at least on the junction and/or the nerve. Direct stimulation of the muscle produced contractions which were not significantly altered by the toxin. In the toad sciatic nerve-sartorius muscle preparation, these investigators found that the normal twitch and tetanus response elicited through the nerve were lost following prolonged exposure to the toxin; the muscle retained its contractility when stimulated directly. According to Li ( 1965), ciguatoxin is an anticholinesterase which causes death through asphyxiation. On several occasions I have questioned that the clinical picture of ciguatera poisoning could be accounted for on the sole basis of anticholinesterase activity. Recent experimental data indicate that ciguatoxin is perhaps more than an anticholinesterase (Rayner et al., 1968). Banner (1967) notes that when 6.1 pg of the pure toxin from the eel Gymnothorax jauanicus, which caused death in two mice in less than 4 hr when injected intraperitoneally at the 0.5 mg/kg level, were added to a standard acetylcholine-cholinesterase solution, the normal rate of cholinesterase activity was inhibited approximately 371%.This was equivalent to the anticholinesterase activity of 6.8 pg of eserine sulfate. Banner lists a number of marine animals which are known at times to contain a toxin that inhibits cholinesterase activity, as demonstrated by titrimetric tests. These include the clam Tridacnu maxima, the gray shark Carcharhinus menisorrah, the eel Gymnothorax jauanicus, the barracuda Sphraenu barracuda, several genera of groupers, the amberjack Seriola

430

FINDLAY E. RUSSELL

dumerili, the red snapper Lutjanus bohar, and the mullet Parupeneus chryserydros. He also notes the toxicity of the blue-green alga Schizothrix calcicolu. Mammals which ingest fish containing the ciguatera poison develop muscular weakness, particularly in the hind legs first, muscular incoordination and ataxia, vomiting, diarrhea, and increased parasympathetic activity. If lethal amounts are consumed, marked respiratory distress, vomiting, salivation, cyanosis, and prostration are often seen prior to death. Deep reflexes are usually hypoactive, and in the more severe cases of poisoning the animal loses its righting reflex (Russell, 1965). The number of clinical cases of ichthyotoxism reported during recent years and the numerous findings of toxic fish by biologists and public health workers indicate that ichthyosarcotoxism is an important medical entity within a broad circumglobal area. The clinical significance of the entity has been the subject of numerous reports (von Fraenkel and Krick, 1945; Lee and Pang, 1945; Cohen et al., 1946; Halstead and Lively, 1954; Banner et al., 1963a; Russell, 1965; Banner, 1967; Halstead, 1967).

4. TETRAODON POISONING Tetrodotoxin (TTX), puffer or fugu poison, is found in certain puffers, ocean sunfishes, porcupinefishes, triggerfishes, spikefishes, trunkfishes, and filefishes. Halstead (1967) lists over 75 tetrodotoxic fishes reported to be poisonous. TTX (tarichatoxin) is also found in certain amphibian species of the family Salamandriadae. The puffers or pufferlike fishes appear to be the only piscines universally regarded as poisonous. Some of the more important of the toxic species, the general biology of these animals, and the chemical and physiopharmacological properties of TTX have recently been reviewed by Russell (1965, 1967b), Kao (1966), Cheymol and Bourillet ( 1966), and Halstead ( 1967). Table I11 shows the concentration of TTX in various tissues of tetrodotoxic fishes and for the newt Taricha torosa. The table is from the excellent review by Kao (1966), and the reader is referred to that work for specific references on the data shown. It can be seen from Table I11 that in most cases the toxin is concentrated in the ovaries and liver, with lesser amounts being found in the intestines and skin, and only small amounts in the body musculature and blood. In almost all fish species so far studied, the concentration in the ovaries has been considerably higher than in the corresponding male tissues. The appearance and amount of toxin in the fish is related to the reproductive cycle and appears to be greatest just prior to spawning, which in turn varies with the species involved and the locale.

8.

431

POISONS AND VENOMS

Table I11 Concentrations of Tetrodotoxin in Tetraodontidae Fishes and a Newt" Species

Ovary ~~

Sphaeroides niphobles Sphaeroides alboplumbeus Sphoeroides pardalis Sphaeroides vermicularis Sphaeroides porphyreus Sphaeroides oscellatus Sphaeroides basilewskianus Sphaeroides chrysops Sphaeroides pseudommus Sphaeroides rubripes Sphaeroides xanthopterus Sphaeroides stietonotus Lagocephalus ine-rmis Canthigaster rivulatus Taricha torosa 0 Taricha torosa 3

Liver ~

400 200 200 400 400 1000 100 40

100 100 100 20 0.4 <2

25


Skin

Intestines

Muscle

Blood

4 4

1 1 -

~~

1000 40 1000 20 1000 100 200 100 200 20 40 20 40 4 40 20 4 10 1 100 40 1 <0.2 2 1 <0.2 2 40
400 40 40 40

40 40 40 4

2 2

4 1 0.4

4 (0.l)b (0.5)b

1 4

1 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 0.4

<0.2 2 8

<0.2 <0.2 1 21

Amounts expressed in micrograms of toxin per gram fresh tissue of female specimens. Visceral organ. Testis. From Kao (1966).

a. Chemistry. Tetrodotoxin is an amino perhydroquinazoline compound having a molecular formula of C,,H,,N,O,, and represented by the following structure (Tsuda et al., 1963, 1964; Tomiie et al., 1963; Goto et al., 1963a,b, 1964; Mosher et al., 1964; Woodward, 1964): 0-

H

OH

Unusual structural features of the toxin include an abundance of OH groups, a guanidinium moiety and a hemilactal link ( C5-lo)between two separate rings. The normally high pK, of the guanidine group is masked by the acidic OH group at C, which has a pK, of 8.5. The terminal NH, forms a zwitterion with one of the OH groups. Tetrodotoxin base is only sparingly soluble in water and easily degraded into several quinazoline

432

FINDLAY E. RUSSELL

compounds under alkaline conditions. Commercially available TTX is buffered with approximately five times its weight of sodium citrate so that when dissolved in water it becomes pH 4.8 (Sankyo Drug, personal communication).

b. Pharmacology and Toxicology. Tetrodotoxin is one of the most potent nonprotein poisons known. The species in which the toxin occurs naturally, the tetraodontids and newts, have been found to be resistant to the poison (Ishihara, 1924), as have barnacle muscle fibers (Hagiwara and Nakajima, 1966). Tetrodotoxin has an extremely steep lethal dose-response curve. The intraperitoneal minimal lethal dose in mice is 8 pg/kg, while the LD,, is 12 pg/kg, and the LD,, approximately 10 pg/kg (Kao and Fuhrman, 1963). The oral LD,, in mice is 332 pg/ kg ( Sakai et al., 1961), while that in cats is in excess of 200 pg/kg. Modification of the toxin’s structure has been found to affect lethality (Tsuda et al., 1964; Narahashi et al., 1966, 1967; Deguchi, 1967). It is evident that the C,OH group is of considerable importance, possibly because of complementary steric binding, along with the guanidine group, upon a receptor site. The unique C,_,, oxygen bridge is also essential for activity, since tetrodonic acid, which lacks this group, is inactive. Camougis et al. (1967) found that the cationic forms of TTX were more potent nerve blocking agents than the zwitterion form. Fuhrman et al. (1968) found that a series of guanidine esters related to tetrodotoxin were much less potent. There is no specific chemical method for quantitating TTX. Several reasonably accurate bioassay techniques are available ( Wakely et al., 1966; Hamada, 1957, 1960; Ogura, 1963). Tetrodotoxin appears to be easily and rapidly absorbed from both the mouth and gastrointestinal tract. It is also rapidly absorbed parenterally. Application to the cornea of the rabbit, however, produces anesthesia only when relatively high concentrations are used (Kao and Fuhrman, 1963). There are few data on the distribution and metabolism of TTX. Ogura (1958) found that peak tissue concentrations were reached 20 min following subcutaneous injection. The kidneys and heart were found to have the highest concentrations, while the lowest were found in the brain and blood. Extrapolation from in vitro data on effective concentrations to in vivo lethality suggests such an occurrence (Mosher et al., 1964). The minimal lethal concentration, based on this distribution, would be approximately 100 nanomoles. There has been some disagreement over whether or not certain TTX effects result from changes in the central or in the peripheral nervous

8.

WISONS AND VENOMS

433

systems. Kao et al. (1967) carried out cross-perfusion experiments and concluded that the hypotensive and respiratory depressant effects of the toxin were due primarily to changes in peripheral nerve and muscle. From their studies, Koizumi et al. (1967) concluded that TTX had an initial central blocking effect, since respiratory, vasomotor, and cortical activities, as well as spinal reflexes were abolished at lower concentrations than those required €or a peripheral axon block. Borison et al. (1963) and Hayama and Ogura (1963) have shown that TTX has a central emetic effect. Animals pretreated with barbiturates in doses known to cause central nervous system depression do not appear to be more reactive to lethal or sublethal doses of TTX than those not treated (Kao and Fuhrman, 1963; Frank and Pinsky, 1966; Bernstein, 1968). Tetrodotoxin is known to depress cortical excitability ( Murtha et al., 1958; Frank and Pinsky, 1966; Koizumi et al., 1967). Ochs and Ng ( 1966) and Ochs and Clark (1968) have demonstrated a local effect for TTX on both the presynaptic elements and the uppermost portions of the apical dendrites. The toxin has also been shown to affect the activity of brain cortex slices (Chan and Quastel, 1967). The respiratory system is quite sensitive to TTX. The action has been attributed to both central and peripheral mechanisms. An initial or early respiratory stimulation has been observed by a number of workers (Ishihara, 1924; Nomiyama, 1942; Kawakami, 1952; Borison et al., 1963; Kao et al., 1967). However, the ultimate respiratory effect is one of depression. Some direct effects on the respiratory center have been observed by Takahashi and Inoko (1890), Ishihara (1917), and by Borison et al. (1963). Li (1963) suggested that the cause of death following a IethaI dose of the poison was respiratory arrest from the action on the brain stem respiratory centers. The arguments put forth by Li are not wholly convincing. It is well known that the toxin has a direct effect on the threshold of the phrenic nerve (Ishihara, 1917; Sakai d al., 1961; Cheymol et al., 1962). Cheng and Li (1966) have shown that the diaphragm motor end plate potential is lost before the block in the phrenic nerve. It seems likely that the peripheral elements of the respiratory system are more sensitive than the central elements, although both probably play a part in the production of the respiratory deficit. Following the intravenous injection of 5 pg/kg of the toxin into a cat there is a precipitous fall in systemic arterial blood pressure. Similar changes are seen in the cat and rat bilaterally vagotomized preparations with a transected cervical cord (Murtha et al., 1958; Cheng and Li, 1966). Feinstein and Paimre (1967) attributed the hypotensive effect to a block in peripheral sympathetic neurons and neural impulses to the adrenal medulla. Lipsius et al. (1967) have demonstrated a direct action

434

FINDLAY E. RUSSELL

of TTX on the perfused gracilis-muscle preparation, while Koizumi et aZ. (1967) indicated that the initial and chief effect was on the medullary vasomotor center. Bernstein ( 1968) has shown that phenoxybenzamine completely blocked the normal recovery response from the hypotension produced by TTX, while not affecting the pressure fall. Tetrodotoxin produces sinus bradycardia in a number of different animals (Tsukada, 1957; Murtha et al., 1958; Feinstein and Paimre, 1967; Bernstein, 1968). The toxin inhibits cardiac contractility ( Murtha et al., 1958), while cardiac output has been reported to be unchanged (Li, 1963) or decreased ( Feinstein and Paimre, 1967). Mean carotid arterial and abdominal aortic blood flows decreases simultaneously with the TTX-induced hypotension ( Cheng and Li, 1966). Feinstein and Paimre (1967) have shown that the sinus bradycardia does not occur following cardiac sympathetic denervation, while Bernstein ( 1968) has demonstrated that reserpine and propranolol prevent or decrease TTX-induced bradycardia. The toxin has been shown to decrease or block the rate of rise of the atrial and ventricular muscle-action potentials, without affecting the atrial pacemaker potential or Purkinje system ( Hagiwara and Nakajima, 1965, 1966; Kuriyama et al., 1966; Yamagishi and Sano, 1966; Aceves and Erlij, 1967; Coraboeuf and Vassort, 1967; Dude1 et aZ., 1967). Tetrodotoxin also blocked aconitine-induced fibrillations ( Kuriyama et al., 1966; Peper and Trautwein, 1967), as well as ouabain-induced arrhythmias (Bernstein, 1968). Sano et al. ( 1968) have shown that tetrodotoxin was more effective than propranolol or diphenyl hydration in reversing atrial fibrillation induced by aconitine, acetylcholine, or calcium. Pretreatment with TTX prevented the cardiac slowing usually seen with cardiac glycosides ( Ten Eick and Hoffman, 1967; Bernstein, 1968). Norepinephrine induced the recovery of a TTX-poisoned atrial preparation ( Aceves and Erlij, 1967). Tetrodotoxin has also been shown to block vagal conduction by preventing cardiac vagal inhibition ( Gershon, 1967; Feinstein and Paimre, 1967). It was observed that when the toxin was ingested by mammals, lethargy, muscular weakness, and incoordination developed. Ataxia occurred and paralysis was observed, usually appearing first in the hindlimbs and subsequently in the forelimbs. In cats and dogs, retching and vomiting sometimes occurred. Deep reflexes were lost and respirations became labored. Cyanosis was sometimes seen, as were convulsions, particularly in the mouse. The most striking sign of the poisoning was the rapid and progressive weakness of the voluntary muscles, including those of respiration. A number of studies of the toxin’s effect on the neuromuscular system

8. POISONS

AND VENOMS

435

have been undertaken (see Russell, 1965, 1969; Kao, 1966; Cheymol and Bourillet, 1966; Halstead, 1967). Only a few will be noted here. Two significant studies were those by Kurose (1943) and Furukawa et al. ( 1959). Kurose ( 1943) presented evidence which appeared to indicate that TTX had little effect on the acetylcholine contracture of the perfused gastrocnemius muscle of the toad. He proposed that the toxin had no curarelike activity. Subsequently, Furukawa et al. ( 1959) concluded that while tetrodotoxin has a potent narcotic effect on nerve and muscle, it does not depolarize them. The poison did not suppress the sensitivity of the end plate to acetylcholine, even at concentrations greater than those necessary for the narcosis of the nerve and muscle. These and certain other studies, while they did not clarify the mode of action of TTX and did raise some new questions, led to studies which have, we trust, clarified the mechanism and site of action of this most interesting poison. It was found that when TTX produced a complete nerve-muscle block, increased neural stimulation did not overcome the block nor cause post-tetanic reversal, and that acetylcholine injected intra-arterially had no effect on the block (Cheymol et al., 1961; Kao and Fuhrman, 1963). The block was potentiated by tubocurarine but was not antagonized by neostigmine or by decamethonium (Kuriaki and Wada, 1957; Kuga, 1958; Cheymol et al., 1961). It became apparent from these and other studies that the failure in neuromuscular transmission was not at the junction, but at the motor axon (Cheymol et al., 1962; Kao and Fuhrman, 1963), and also at the muscle membrane (Matsumura and Yamamoto, 1954; Narahashi et at., 1960; Kao and Fuhrman, 1963). When stimulated directly, the excitability of skeletal muscle in the presence of TTX is blocked, although the time required for a block is longer than that required for the nerve block ( Ishihara, 1917). Slow muscles are more easily blocked than fast muscles. It was found that the block produced by TTX occurred without depolarization (Dettbarn et at., 1960; Cheymol et al., 1962; Kao and Fuhrman, 1963), much like that produced by certain anesthetics (Russell et al., 1961; Mosher et al., 1964). However, TTX differs from procaine and cocaine in that it acts selectively to prevent or reduce the usual increase in permeability to the sodium ion without seeming to affect the outward potassium current (Narahashi et al., 1 9 0 ) . At certain dose levels, however, it also blocks the outward movement of the sodium ion (Moore et al., 1966). With the possible exception of saxitoxin, puffer poison is the only substance known to have this highly selective action. Quantitatively, its action on the nerve axon is greater than that of saxitoxin ( Cheymol, 1965). Combining perfusion and voltage clamp techniques in the squid giant

436

FINDLAY E. RUSSELL

axon and detailed studies on the ionic movements across the axon membrane have been carried out (Narahashi et al., 1964; Nakamura et al., 1965; Takata et al., 1966). The molecular events involved in the permeability changes have not yet been determined. It is possible that the “gates” controlling the inward movement of sodium and the outward flow of potassium are in some way altered, or the ion pump is involved in the ionic changes, or perhaps both contribute to the permeability changes. In his comparative studies on saxitoxin and tetrodotoxin, Evans (1967) has shown that the intravenous injection of TTX causes a rather selective block in nerve conduction in sensory fibers, and at lower concentrations than are needed to block conduction in motor fibers. Large myelinated sensory fibers become blocked with doses of 4.5-13 pg/kg; large motor fibers were not blocked until the dose was raised approximately 35%.The conduction block in the sensory fibers first appeared in the region of the dorsal root ganglion. The study of the action of TTX on the neuromuscular junction is made difficult because the poison affects both the nerve ending and the muscle membrane. Intracellular microelectrode studies of the action of the toxin on single end plates in the frog and guinea pig indicate that the toxin does not cause depolarization (Furukawa et al., 1959). The end-plate potential stimulated through the nerve rapidly decreases even with nanomolar concentrations of the poison, although the membrane remains responsive to acetylcholine (Fleisher et al., 1961). This and other evidence indicates that the chemosensitive end-plate receptors are not affected by the toxin, and provides further support for the contention that conduction in nerve and muscle membranes is fundamentally different from that at the synapse. Slightly over 100 persons die each year in Japan from eating tetrodotoxic fishes (Kao, 1966), and additional deaths are reported from other areas of the Orient. The mortality rate has declined from approximately 80%in 1900 to approximately 52%in 1955. In 1957 there were 119 episodes of poisoning involving 176 persons, of which 90 died (Ogura, 1958).

POISONING 5. SCOMBROID Certain of the mackerallike fishes, the tunas, skipjacks, and bonitos, and the Japanese saury, Cololabis sairu, are occasionally involved in poisonings to man. While the symptoms and signs of scombroid poisoning in man are quite different from those produced by ciguatera poisoning, these fishes can also be involved in ciguatera poisoning. If scombroids are inadequately preserved, a toxic substance is said to be formed within the body musculature. This substance was once thought to be histamine,

8.

POISONS AND VENOMS

437

formed by the action of enzymes and bacteria (Markov, 1943) or released by bacterial action on the death of the fish (Geiger et al., 1944). However, more recent evidence would seem to indicate that the toxic component may not be histamine, although the complex does resemble this amine in certain of its properties (Kawabata et al., 1955). Kawabata has given the toxic substance the name “saurine.” Saurine is said to closely resemble acetylcholine and histamine in some of its pharmacological properties, but it is chemically distinct. It has an A, value of 0.1 with n-butanol-acetic acid-water (4:1:2) and is insoluble in ether, acetone, benzene, and chloroform. It is dialyzable, can be precipitated with phosphotunstic acid, and has certain other properties which would indicate that it has a relatively low molecular weight. It appears that the toxic complex of scombroid poisoning may be saurine and histamine, and perhaps a third substance. 6. CLUPEOID POISONING

This form of ichthyosarcotoxism may occur following the ingestion of certain herrings, anchovies, tarpons, bonefishes, and slickheads. Since most of these fishes are valuable food fishes, this entity is one of considerable concern in certain parts of the world. Helfrich (1961) notes that one species of herring is thought to be toxic at all times in Fiji, while a second species is poisonous only during October and November. The biogenesis of the toxicity is not known. It would seem that toxicity is associated with the food chain, but the contributing organism(s) has(have) not been idenaed. It has been suggested that a dinoflagellate ( D’Arras, 1877), or a blue-green alga ( Randall, 1958),or even the palolo worm may be involved. Nothing appears to be known about the chemistry and toxicology of this poison. In 1967, several reportedly toxic Clupea oenenosa were sent to our laboratory for examination. The viscera and some flesh was fed to a cat. The cat smelled but did not touch the viscera. It did eat a small portion of the flesh before abruptly stopping and retreating to a corner, where in about 4 min it assumed a most un-catlike stance, with all extremities extended and its head resting on the floor. It lay on its abdomen, panted with unusual force and retched several times. The pupils were dilated; deep reflexes seemed normal but the cat could not be made to stand; the righting reflex was intact. At approximately 7 min following the meal the cat began to salivate profusely, to vocalize, and it assumed a guarded position. The signs subsided over the next 2 hr, and other than passing two loose stools there were no further significant changes.

438

FINDLAY E. RUSSELL

7. GEMPYLIDPOISONING The flesh of certain fishes, particularly the gempylids, is known to have a high concentration of wax, The chemistry and physiopharmacology of the musculature of the castor oil fishes Lepidocybium fiauobrunneum (Smith) and Ruuettus pretiosus Cocco, both belonging to the family Gempylidae, have been studied by Japanese workers. Kimura (1927) found that cetyl and oleyl alcohol were the chief components of the unsaponifiable fraction from R. petiosus. Cox and Reid (1932) classified the oil of the flesh as a liquid wax consisting mainly of cetyl and oleyl esters of oleic and hydroxy oleic acids. Matsumoto et al. (1955) reported that the oil of the flesh of L. flauobmnneum was also composed of cetyl and oleyl alcohol. Mori et al. (1966) analyzed the flesh of both species using modem chromatographic techniques. The wax, which made up 204: of the acetone soluble oil, contained cetyl and oleyl alcohol and oleic and eicosenol acids. The castor oil fishes are not usually used as food. Their seborrheic and diarrheal effects in mammals are well known (Sahashi, 1933; N. Matsuo et aE., 1963; Mori et al., 1966). Animals maintained on a diet of castor oil fish lose weight and eventually die. One of the cod, Erilepsis zonifer Lockington, also provokes diarrhea when eaten in quantity. Kaneda (1952) has shown that the oil of the flesh is composed chiefly of glycerides high in oleic acid. The toxin appears to be different from that found in the castor oil fishes.

(a%),

B. Ichthyocrinotoxism 1.

OSTRACITOXIN AND PAHUTOHTN

Certain fishes are known to release a toxic substance from the skin which is capable of killing other fishes and perhaps other marine animals (Brock, 1956; Whitley, 1957; Maretzki and del Castillo, 1967). This toxin appears to be part of the animal's defensive armament and is perhaps released as an alarm substance to deter possible predators (Pfeiffer, 1962).

a. Chemistry. Thomson (1964) separated a toxic principle from the

skin secretions of the boxfish or trunkfish Ostracion lentiginosus. The toxin was heat stable, nondialyzable and soluble in water, methanol, ethanol, acetone, and chloroform. It was stable in acid ( p H 2.0) and basic ( p H 11.0) solutions, but it was rapidly detoxified upon excess addition of a strong base. Repeated extractions of residues (obtained

8.

439

POISONS AND VENOMS

from drying the skin secretions) with acetone or chloroform and diethyl ether gave a particulate substance which was toxic to fish at concentrations of 1 : 1,000,000. Approximately 50-100 mg of the crude dried toxin could be obtained at one time from a single adult boxfish. Thomson called the toxin “ostracitoxin.” Maretzki and del Castillo (1967) studied the foamy skin secretions of the tropical AtIantic soapfish Rypticus saponaceus. They found the toxin to be nondialyzable and insoluble in solutions of low ionic strength. Crude solutions of the toxin rapidly became inactive at a neutral pH but were stable at pH 3.4. The toxic principle could not be extracted with ethyl ether, chloroform, or methyl acetate. Activity was retained following heating at 65°C for 2 hr. A number of amino acids were identified in the toxin, which was relatively rich in leucine, isoleucine, and ammonia. The total nitrogen of their toxic preparation was 4.4%. Boylan and Scheuer (1967) have given us considerable data on the skin secretions of the boxfish Ostracion lentiginosus. When a crude solution was extracted into 1-butanol, a 20-fold purification of the toxin was obtained. The toxic butanol solution was chromatographed on a column of silicic acid, and elution carried out with a chloroform-methanol (88:12) mixture. A toxic white amorphous solid was recovered. This material was passed through an anion exchange column treated with picric acid, and yielded a product which crystallized from acetone in the form of long colorless needles. The product was called “pahutoxin.” It had a specific rotation of +3.05” (22”; 2.30 g/100 ml methanol), was soluble in water, ethanol, chloroform, hot acetone, and hot ethyl acetate. Spectroscopic data, hydrolytic degradations and synthesis gave the formula C,,H,,NO,Cl and the structure: H

I r L ~ ~ 2A -~ ( c H~ ~)~1 c~- - - ( ~ ~ a )

CH~- (cH~)

I

OCOCHs

Pahutoxin is thus the choline chloride ester of 3-acetoxyhexadecanoic acid. It and its C,, and C,, homologs have been synthesized as the racemates.

b. Pharmacology and Toxicology. Thomson (1964) demonstrated that when the skin mucus of the boxfish Ostracion Zentiginosus was added to an aquarium containing other reef fishes, these fishes exhibited “irritaability,” gasping, then activity with a decrease in opercular movements, loss of equilibrium and locomotion, and finally, sporadic convulsions and death. He established a bioassay based on the amount of the toxic rinse volume per 100 ml of seawater needed to kill four to six newborn sailfin

440

FINDLAY E. RUSSELL

mollies, 10-12 mm long. When the mucus was injected into the boxfish, the fish immediately lost its balance, and death occurred within a few minutes. The boxfish was also found to be sensitive to its own mucus when exposed to high concentrations of the toxin in an aquarium. When injected into mice, ostracitoxin produced ataxia, labored respirations, coma, and death. The minimal lethal dose was 200 mg/kg body weight ( Thomson, 1964). Maretzki and del Castillo (1967) found that when guppies were placed in seawater in which the soapfish Rypticus saponaceus had been handled, a concentration of one part exudation to 16 parts seawater was sufficient to kill the fish within 15-20 min. Pahutoxin, the toxin isolated by Boylan and Scheuer (1967), was quantitated for its hemolytic property, which correlated with its lethal property. The minimum lethal concentration for fish was found to be 0.176 pg/ml, when death was measured at 1 hr. ACKNOWLEDGMENTS

I wish to thank my colleagues Dr. Jacob W. Dubnoff and Dr. Martin Bernstein for their advice on certain parts of the treatise. I am also indebted to MIS. Bernice Kellar and Mr. Henry Gonzalez for their cheerful assistance in preparing the manuscript. Some data not heretofore reported were taken from studies supported by a contract from the U.S. Office of Naval Research. REFERENCES Aceves, J., and Erlij, D. (1967). Effects of norepinephrine on tissues of the frog heart atrium poisoned by tetrodotoxin. Nature 215, 1178-1179. Austin, L., Cairncross, K. D., and McCollum, I. A. N. ( 1961). Some pharmacological actions of the venom of the stonefish Synanceju hrrida. Arch. Intern. Pharmacodyn. 131, 339-347. Banner, A. H. (1967). Marine toxins from the Pacific, I. Advances in the investigation of fish toxins. In “Animal Toxins” (F. E. Russell and P. R. Saunders, eds.), pp. 157-165. Pergamon Press, Oxford. Banner, A. H., and Boroughs, H. (1958). Observations on toxins of poisonous fishes. Proc. SOC. Exptl. Biol. Med. 98, 776-778. Banner, A. H., Scheuer, P. J., Sasaki, S., Belfrich, P., and Mender, C . B. (1960). Observations on ciguatera-type toxin in fish. Ann. N.Y. Acad. Sci. 90, 770-787. Banner, A. H., Shaw, S. W., Alender, C. B., and Helfrich, P. (1963a). Fish intoxication. South Pacific Comm., Tech. Paper 141, 1-17. Banner, A. H., Helfrich, P., Scheuer, P. J., and Yoshida, T. (196313). Research on ciguatera in the tropical Pacific. Proc. Gulf. Caribbean Fisheries Inst., 16th Ann. pp. 84-98. Bernstein, M. E. ( 1968). Pharmacologic effects of tetrodotoxin: Cardiovascular and anti-arrhythmic activities. Thesis, Indiana University, Indianapolis, Indiana. BZje, 0. (1939). Toxin in the flesh of the Greenland shark. Medd. Greenland 125, 1-16.

8.

POISONS AND VENOMS

441

Borison, H. L., McCarthy, L. E., Clark, W. G., and Radhakrishnan, N. (1963). Vomiting, hypothermia, and respiratory paralysis due to tetrodotoxin in the cat. Toxicol. Appl. Pharmacol. 5, 35&357. Bottard, A. ( 1889). “Les poissons venimeux.” Doin, Paris. Boylan, D. B., and Scheuer, P. J. (1967). Pahutoxin: A fish poison. Science 155, 52-56. Briot, A. (1902). Sur l’action due venin d e la vivre ( Trachinus druco). Compt. Rend. SOC. Biol. 54, 1169-1171. Briot, A. (1903). Diffkrence d’action venimeuse des 6pines dorsales et des &pines opercularies de la vivre. Compt. Rend. SOC. Biol. 55, 623-624. Briot, A. (1904). Sur l’existence d’une kinase dans le venin de la vivre (Truchinus draco). Cumpt. Rend. SOC. Biol. 57, 666-667. Brock, V. E. (1956). Possible production of substances poisonous to fishes by the box fish, Ostracion lentiginosus Schneider. Copeia No. 3, 196196. Cameron, A. M., and Endean, R. ( 1966). The venom apparatus of the scorpion fish Notesthes robusta. Toxicon 4, 111-121, Camougis, G., Takman, B. H., Rene, J., and Tasse, P. (1967). Potency difference between the zwitterion form and the cation forms of tetrodotoxin. Science 156, 1625-1627. Carlisle, D. B. (1962). On the venom of the lesser weeverfish, Trachinus uipera. J. Marine Biol. Assoc. U.K. 42, 155-162. Castex, M. N. (1967). Freshwater venomous rays. In “Animal Toxins” (F. E. Russell and P. R. Saunders, eds.), pp. 167-176. Pergamon Press, Oxford. Castex, M. N., and Loza, F. (1964). Etiologia de la enfermedad Paratrygonica. Rev. Assoc. Med. Arg. 78, 314-324. Castex, M. N., and Suilar, F. (1965). Observactiones sobre un lote de P. magdalenae. Phgsis 25,239-243. Chan, S . L., and Quastel, J. 11. (1967). Tetrodotoxin: Effects on brain metabolism in uitro. Science 156, 1752-1753. Cheng, K. K., and Li, K. M. (1966). The hypotensive action of p d e r fish toxin. 1. Pathol. Bacteriol. 92, 471476. Chevallier, A., and Duchesne, E. A. ( 1851). MBmoire sur les empoisonnements par les huitres, les modes, les crabes, et par certains poissons de mer et de rivihre. Ann. Hyg. Publ. (Paris) 45, 387-437; 46,108-147. Cheymol, J. ( 1965). Do dos substancias biomarinas inhibidoras neuromusculares : Tetrodotoxina y saxitoxina. Arch. Fac. Med. Madrid 8, 151-158. Cheymol, J. and Bourillet, F. (1966). D u n e nouvelle classe de substances biologiques: tBtrodotoxine, saxitoxine, tarichatoxine. Acfualites Phurmacol. 19, 1-61. Cheymol, J., Kobayashi, T., Bourillet, F., and Thtreault, L. (1961). Sur I’action paralysante neuromusculaire de la t6trodotoxine. Arch. Intern. Phamnacodyn. 134, 28-53. Cheymol, J., Foulhoux, P., Bourillet, F., and Simon, P. (1962). Action de la thtrodotoxine sur les phhnomenes electriques de la transmission neuromusculaire. Compt. Rend. SOC. Biol. 158, 602-607. Chisholm, C. (1808). On the poison of fish. Edinburgh Med. Surg. J . 4, 396422. Cohen, S. C., Emert, J. T., and Goss, C. C. (1946). Poisoning by barracuda-like fish in the Marianas. US. Naval Med. Bull. 46, 311-317. Cooper, M. J. (1964). Ciguatera and other marine poisoning in the Gilbert Islands. Pacific Sci. 18, 411-440. Coraboeuf, E., and Vassort, C. ( 1967). Effects d’inhibiteurs des permeabilities

442

FINDLAY E. RUSSELL

ioniques sur les potentiels d’action cardiaques de mammiferes. J . Physiol. (London) 59, 228. Coutaud, H. (1879). Observations sur sept cas d’empoisonnement par le foie de reqiun. M.D. Thhse, No. 8, Fac. Mhd. Paris. Cox, W. M., Jr,, and Reid, E. E. (1932). The chemical composition of oil of Ruuettus pretiosus, the “castor oil” fish. J. Am. Chem. SOC. 54, 22CL229. D’Arras, L. (1877). Essai sur les accidents causes par les poissons. Thhse No. 156, Fac. MBd. Paris. Dawson, E. Y., Aleem, A. A., and Halstead, B. W. (1955). Marine algae from Palmyra Island with special reference to the feeding habits and toxinology of reef fishes. Occasional Papers Allan Hancock Found. 17, 39. Deakins, D. E., and Saunders, P. R. (1967). Purification of the lethal fraction of the venom of the stonefish Synanceja hotrida (Linnaeus). Toxicon 4, 257-262. Deguchi, T. (1967). Structure and activity in tetrodotoxin derivatives. Japan. J. Pharmacol. 17, 267-278. De Marco, R. ( 1936). Ricerche sulla permeabilith dell’asse cerebrospinale di Bufo viridis sotto l’azione del veleno di Trachinus draco e di S c o ~ ~ a e nscropha a Boll. SOC. Ital. Biol. Sper. 11, 767-768. De Marco, R. (1938). Effetti del veleno di Trachinus sulla capacita di lavoro del gastrocnemio di rana. Rio. Biol. (Penrgia) 25, 225-234. de Sauvages, F. B. (1758). De venenatis galliae animalibus. M.D. Thesis, Rothomagensi Academia, Joannis Martel, Monspelii. Dettbam, W. D., Higman, H. B., Rosenberg, P., and Nachmansohn, D. (1960). Rapid and reversible block of electrical activity by powerful marine biotoxins. Science 132, 300-301. Dudel, J., Peper, R., and Trautwein, W. (1967). The effect of tetrodotoxin on the membrane current in cardiac muscle (Purkinge fibers). Arch. Ges. Physiol. 295, 213-226. Duhig, J. V. ( 1929). The nature of the venom of Synanceju horrida (the stonefish). Z . Zmmunitaelsforsch.62, 185-189. Duhig, J. V., and Jones, G. ( 1928a). The venom apparatus of the stonefish (Synanceja horrida). Mem. Queensland Museum 9, 136-150. Duhig, J. V., and Jones, G. (1928b). Haemotoxin of the venom of Synancda horrida. Australian J . Exptl. Biol. Med. Sci. 5, 17S179. Endean, R. (1961). A study of the distribution, habitat, behavior, venom apparatus and venom of the stonefish. Australian J. Marine Freshwater Res. 12, 177-190. Engelsen, H. (1922). Om gift6sk og giftige fisk. N o d . Hyg. Tidskr. 3, 316-325. Evans, H. M. (1907). Observations on the poisoned spines of the weever fish, Trachinus draco. Brit. Med. J . 4 73-76. Evans, H. M. (1916). The poison-organ of the sting-ray (Trygon pastinaceu). Proc. 2001.SOC. London 29, 431440. Evans, M. H. (1967). Block of sensory nerve conduction in the cat by mussel poison and tetrodotoxin. I n “Animal Toxins” (F. E. Russell and P. R. Saunders, eds.), pp. 97-108. Pergamon Press, Oxford. Feinstein, M. B., and Paimre, M. ( 1967). Cardiovascular actions of tetrodotoxin. Pharmucologtst 9, 198. Fish, C. J., and Cobb, M. C. (1954). Noxious marine animals of the central and western Pacific Ocean. US. Fish. Wildlife Serv., Res. Rept. 36, 1-45. Fleisher, J. H., Killos, P. J., and Harrison, C . S. (1961). Effects of puffer poison on neuromuscular transmission. J . Pharmucol. Exptl. Therap. 133, 98-105.

8.

POISONS AM, VENOMS

443

Fleury, R. (1950). L’appareil vbnimeux des salaciens trygoniformes. M h . SOC.2001. France 30, 1-37. Frank,. G. B., and Pinsky, C. (1966). Tetrodotoxin induced central nervous system depression. Brit J . Pharmacol. ‘26, 435443. Fuhrman, F. A,, Ranney, B. K., Schmiegel, J. L., and Mosher, H. (1968). Some guanidine esters and their possible relationship to the action of tetrodotoxin. Proc. Western Pharmacol. SOC. 11, 3 7 4 2 . Furukawa, T., Sasaoka, T., and Hosoya, Y. (1959). Effects of tetrodotoxin on the neuromuscular junction. Japan. J . Physiol. 9, 143-152. Cage, S. H., and Cage-Day, M. (1927). The anti-coagulating action of the secretion of the buccal glands of the lampreys (Petromyzon, Lampetra and Entosphenus). Science 66, 282-284. Gail, R., and Rageau, J. (1956). Premieres observations sur un Poisson marin venimeux de la Nouvelle-Calbdonie: La synancee ( Synanceia verrucosa Bloch) . Bull. Soc. Pathol. Exotigue 49, 846-854. Geiger, E., Courtney, G., and Schnakenberg, G. (1944). The content and formation of histamine in fish muscle. Arch. Biochem. 3, 311-319. Gershon, M. D. (1967). Effects of tetrodotoxin on innervated smooth muscle preparation, Brit. J . Phamacol. 29, 259-279. Coto, T., Kishi, Y., Takahashi, S . , and Hirata, Y. (1963a). The structure of tetrodotoxin. Tetrahedron Letters No. 30, 2105-2113. Goto, T., Takahashi, S., Kishi, Y., and Hirata, Y. (1963b). The structure and stereochemistry of tetrodotoxin. Tetrehedron Letters No. 30, 2115-2118. Goto, T., Takahashi, S., Kishi, Y., and Hirata, Y. (1964). Tetrodotoxin. 111. A revised molecular formula for tetrodotoxin. Bull. Chem. Soc. Japan 37, 283-284. Gressin, L. (1884). Contribution a I‘btude de I’appareil vbnin. Thesis, A. Parent, Paris. Haavaldsen, R., and Fonnum, F. (1963). Weever venom. Nature 199,286-287. Habekost, R. C., Fraser, I. M., and Halstead, B. W. (1955). Toxicology-observations on toxic marine algae. J. Wash. Acad. Sci. 45, 101-103. Hagiwara, S., and Nakajima, S. (1965). Tetrodotoxin and manganese ion: Effects on action potential of the frog heart. Science 149, 1254-1255. Hagiwara, S . , and Nakajima, S. (1966). Differences in Na and Ca spikes as examined by application of tetrodotoxin, procaine, and manganese ions. 1. Gen. Physiol. 49,793-806. Halstead, B. W. ( 1951). Injurious effects from the sting of the scorpionfish Scorpaena guttata. Calif. Med. 74, 395-396. Halstead, B. W. (1953). Some general considerations of the problem of poisonous fishes and ichthyosarcotoxism. Copeia No. 1, 31-33. Halstead, B. W. (1958). Poisonous fishes. Public Health Rept. ( U . S . ) 73, 302-312. Halstead, B. W. ( 1964). Fish poisonings-their diagnosis, pharmacology, and treatment. Clin. Pharmacol. Therap. 5, 615-627. Halstead, B. W. (1967). “Poisonous and Venomous Marine Animals of the World,” Vol. 2. U.S. Govt. Printing Office, Washington, D.C. Halstead, B. W., and Bunker, N. C. (1952). The venom apparatus of the ratfish Hydrolugus colliei. Copeia No. 3, 128-138. Halstead, B. W., and Bunker, N. C. (1953). Stingray attacks and their treatment. Am. J . Trop. Med. Hyg. 2, 115-128. Halstead, B. W., and Bunker, N. C. (1954). A survey of the poisonous fishes of Johnston Island. Zoologica 39, 61-77.

444

FINDLAY E. RUSSELL

Halstead, B. W., and Dalgleish, A. E. (1967). The venom apparatus of the European star-gazer Uranoscopus scaber Linnaeus. In “Animal Toxins” (F. E. Russell and P. R. Saunders, eds.), pp. 177-186. Pergamon Press, Oxford. Halstead, B. W., and Lively, W. M. ( 1954). Poisonous fishes and ichthyosarcotoxism. US. Armed Forces Med. J. 5, 157-175. Halstead, B. W., and Modglin, F. R. (1958). Weeverfish stings and the venom apparatus of weevers. 2. Tropenmed. Parasitol. 9, 129-146. Halstead, B. W., and Schall, D. W. (1958). A report on the poisonous fishes of the Line Islands. Acta Trop. 15, 193-233. Halstead, B. W., Ocampo, R. R., and Modglin, F. R. (1955a). A study on the comparative anatomy of the venom apparatus of certain North American stingrays. J. Mo~phol.87, 1-21. Halstead, B. W., Chitwood, M. J., and Modglin, F. R. (1955b). The anatomy of the venom apparatus of the zebrafish Pterois volitans ( Linnaeus ) . Anat. Record 122, 317-333. Hamada, J. (1957). Etude pharmacologique de la tbtrodotoxine cristallistk sur la fibre musculaire lisse. Ann. Rept. lnst. Food. Microbiol., Chiba Univ. 10, 73-77. Hamada, J. (1960). Effect of crystalline tetrodotoxin on smooth muscle. 11. Effect of tetrodotoxin and ganglionic active drugs on the isolated intestine and bronchial muscles. J . Chiba Med. SOC. 36, 1358-1368 (in Japanese), Hashimoto, Y. (1956). A note on the poison of a barracuda, Sphyraena picuda Bloch and Schneider. Bull. Japan. SOC. Sci. Fisheries 21, 1153-1157. Hayama, T., and Ogura, Y. (1963). Site of emetric action of tetrodotoxin in dog. J. Pharmacol. Exptl. Therap. 139, 94-98. Helfrich, P. (1961). “Fish Poisoning in the Tropical Pacific,” Contrib. No. 155. Hawaii Marine Lab., Honolulu, Hawaii. Helfrich, P., and Banner, A. H. (1963). Experimental induction of ciguatera toxicity in fish through diet. Nature 197, 1025-1026. Helfrich, P., Piyakarnchana, T., and Miles, P. S. (1968). Ciguatera fish poisoning, I. The ecology of ciguateric reef fishes in the Line Islands. Occasional Papers Bishop Museum 23, 305-382. Hessel, D. W. (1961). Marine biotoxins. 11. The extraction and partial purification of ciguatera toxin from Lutjanus bohar. (Forskal). Toxicol. Appl. Pharmacol. 3, 574-583. Hessel, D. W., Halstead, B. W., and Peckman, N. H. (1960). Marine biotoxins. I. Ciguatera poison: Some biological and chemical aspects. Ann. N.Y. Acad. Sci. 90, 788-797. Hiyama, Y. (1943). “Report of an Investigation on Poisonous Fishes of the South Seas.” Nissan Fish Expt. Sta., Odawara, Japan (in Japanese). Ishihara, F. (1917). The physiological study of puffer poison. Tokyo Igaku Zasshi 31, 27&279 (in Japanese). Ishihara, F. (1924). Studien iiber das Fugutoxin. Arch. Exptl. Pathol. Phamakol. 102, 209-222. Jensen, A. S. (1914). The selachians of Greenland. Mindeskr. Steenstrups, Fods., Kbh. 30, 12-16. Jensen, A. S. ( 1948). Contributions to the ichthyofauna of Greenland. Skrifter Univ. Zool. Museum Copenhagen 9, 20-25. Kaiser, E., and Michl, H. (1958). “Die Biochemie der Tierischen Gifte.” Deuticke, Wien. Kaneda, T. (1952). Studies on the composition of fatty acids of the “black cod,”

8.

POISONS AND VENOMS

445

Erilepsis zonifer-body oil and its nutritive value. Bull. Japan. SOC. Fisheries 17, 15-1 9. Kao, C. Y. (1966). Tetrodotoxin, saxitoxin and their significance in the study of excitation phenomena. Pharmacol. Rev. 18, 997-1049. Kao, C. Y., and Fuhrman, F. A. (1963). Pharmacological studies on tarichatoxin, a potent neurotoxin, J. Pharmacol. Exptz. Therap. 140, 3 1 4 0 . Kao, C. Y., Suzuki, T., Kleinhaus, A. L., and Siegmann, M. (1967). Vasomotor and respiratory depressant actions of tetrodotoxin and saxitoxin. Arch. Intern. Pharmacodyn. 16.5, 438450. Kawabata, T., Ishizaka, K., and Muira, T. (1955). Studies on the food poisoning associated with putrefaction of marine products. I1 and 111. Bull. Japan. Soc. Sci. Fisheries 21, 3 4 1 3 4 6 and 3 4 7 4 5 1 (in Japanese). Kawakami, K. ( 1952). Pharmacological studies on fugu toxin. lgaku Kenkyu 22, 1278-1282. Kimura, S. ( 1927). Zur kenntnis der Wirkung des Tetrodongiftes. Tohoku J. Exptl. Med. 9, 41-65. Koizumi, K., Levine, D. G., and Brooks, C. McC. (1967). Effect of tetrodotoxin (puffer fish toxin) on the central nervous system. Neurology 17, 395-404. Kornalik, F. (1967). “Animal Toxins.” State Public Health, Prague. Kuga, T. (1958). The actions of tetrodotoxin on neuromuscular transmission. Folia Pharmacol. Japon. 55, 1257-1267 (in Japanese). Kuriaki, K., and Wada, I. (1957). Effect of tetrodotoxin on mammalian neuromuscular system. Japan. J. Pharmacol. 7 , 35-37. Kuriyama, H., Osa, T., and Toida, N. (1966). Effect of tetrodotoxin on smooth muscle cells of the guinea pig taenia coli. Brit. J. Pharmacol. 27, 366-376. Kurose, T. (1943). Pharmacological effect of mussel toxin on the stretching and contraction of muscle. Folia Phurmacol. Japon. 38, 4 4 1 4 7 8 (in Japanese). Lee, R. K. C., and Pang, H. Q. (1945). Ichthyotoxism-fish poisoning. Am. J . Trop. Med. 25, 281-285. 791. Li, K. M. (1963). Action of puffer fish poison. Nature Li, K. M. (1965). Ciguatera fish poison. A cholinesterase inhibitor. Science 147, 1580-1581. Linstow, 0. V. ( 1894). “Die gifttiere und ihre wirkung auf den menschen.” Hirschwald, Berlin. Lipsius, M., Siegman, M. J., and Kao, C. Y. (1967). Direct action of procaine and tetrodotoxin on vascular muscle. Federation Proc. 96, 736. McFarren, E. F., and Bartsch, A. F. (1960). Application of the paralytic shellfish poison assay to poisonous fishes. J. Assoc. Ofic. Agr. Chemists 43, 548-554. Maretii., Z. (1957). Erfahrungen mit Stichen von Giftfischen. Acta Trop. 14, 157-161. Maretzki, A,, and del Castillo, J. (1967). A toxin secreted by the soapfish, Rypticus saponaceus. A preliminary report. Toxicon 4, 245-250. Markov, S. ( 1943). Seefischvergiftung. Wien. Med. Wochschr. 93, 388. Matsumoto, T., Sone, H., and Niiya, I. (1955). The lipids of a fish, Xenogramma carinaturn. J. Japan Oil Chemists’ Soc. 4, 131-133. Matsumura, M., and Yamamoto, S. (1954). The effect of tetrodotoxin on the neuromuscular junction and peripheral nerve of the toad. Japan. J. Pharmacol. 4, 62-68. Matsuo, N., Tamashita, A., and Takahashi, H. (1963). Biochemical studies on oils and fats. XVII. Relations between carbon numbers of esters and occurrence of seborrhea in rats, Seikagaku 35, 316-321.

446

FINDLAY E. RUSSELL

Matsuo, R. (1934). Study of the poisonous fishes of Jaluit Island. Nanyo Gunto Chibobyo Chosa Ronbunshu 2, 309-326 (in Japanese). Moore, J. W., Anderson, N., and Narahashi, T. (1966). Tetrodotoxin blocking early conductance channel or sodium? Federation PTOC.25, 569. Mori, M., Saito, T., Nakanishi, Y., Miyazawa, K., and Hashimoto, Y. (1966). The composition and toxicity of wax in the flesh of castor oil fishes. Bull. Japan. SOC. Sci. Fisheries 32, 137-145. Mosher, H. S., Fuhrman, F. A., Buchwald, H. D., and Fischer, H. G. (1964). Tarichatoxin-tetrodotoxin: A potent neurotoxin. Science 144, 1100-1110. Murtha, E. F., Stabile, D. E., and Wills, J. H. (1958). Some pharmacological effects of puffer poison. 1. P h u m c o l . Exptl. Therap. 122, 247-254. Nakamura, Y., Nakajima, S., and Grundfest, H. (1965). The action of tetrodotoxin on the electrogenic components of squid giant axons. J. Gen. Physiol. 48, 985996. Narahashi, T., Deguchi, T., Urakawa, N., and Ohkubo, Y. (1960). Stabilization and rectification of muscle fiber membrane by tetrodotoxin. Am. J. Physiol. 198, 934938. Narahashi, T., Moore, J. W., and Scott, W. R. (1964). Tetrodotoxin blockage of sodium conductance increase in excitation. J. Gen. Physiol. 47, 985-974. Narahashi, T., Moore, J. W., and Poston, R. N. (1966). Specific action of tetrodotoxin derivatives on nerve. Science 154, 425. Narahashi, T., Anderson, N. C 3 ,and Moore, J. W. (1967). Comparison of tetrodotin and procaine in internally perfused squid giant axon. 1. Gen. Physiol. 50, 1413. Nomiyama, S. ( 1942,). Pharmakologische Untersuchungen uber Fugugift. Folia Pharmucol. Japon. 35, 458-496. Ocampo, P. R., Halstead, B. W., and Modglin, F. R. (1953). The microscopic anatomy of the caudal appendage of the spotted eagleray, Aetobatus rnarinori (Euphrasen) with reference to the venom apparatus. Anut. Record 115, 87-99. Ochs, S., and Clark, F. J. (1968). Tetrodotoxin analysis of direct cortical responses. Electroencephulog. C h . Neurophysioz. 24, 101-107. Ochs, S., and Ng, M.-H. (1966). Tetrodotoxin on N-wave direct cortical response. Federation Proc. 25, 637. Ogura, Y. (1958). Some recent problems on fugu-toxin, particularly on crystalline tetrodotoxin. Seitai No Kaguku 9, 281-287 (in Japanese). Ogura, Y. (1983). The biological evaluation of crystalline tetrodotoxin. 111. On the isolated stomach vagal nerve preparation of rat. Ann. Rept. Inst. Food Microbiol., Chiba Univ. 1.5, 93-96 (in Japanese), Pearson, R. B. (1967). The venom apparatus of selected Queensland stingrays. Thesis, University of Queensland, Brisbane, Australia. Peper, K., and Trautwein, W. (1967). The effect of aconitine on the membrane current in cardiac muscle. Arch. Ges. Physiol. 2Q6, 328-336. Pfeiffer, W. (1982). The fright reaction of fish. Biol. Reu. 37, 49-11. Phillips, C., and Brady, W. H. (1953). “Sea Pests.” Univ. of Miami Press, Miami, Florida. Phisalix, C. ( 1899). Exp6riences sur le venin des vives ( Trachinus uipera et Trachinus draco). Bull. Museum Nut. Hist. Nut. (Paris) 5, 256-258. Phisalix, M. (1922). “Animaux venimeux et venins,” 2 vols. Masson, Paris. Pigulevsky, S. V. ( 1966). “Poisonous and Venomous Reptiles and Fishes.” Medicin, Leningrad ( in Russian ) .

8.

POISONS AND VENOMS

447

Pohl, J. (1893). Beitrag zur Lehre von den Fischgiften. PTUgeT Med. Wochschr. 18, 31-33. Porta, A. ( 1905 ). Ricerche anatomiche sull’apparecchio velenifera di alcumi pesci. Anat. Anz. 26, 232-247. Prokhoroff, P. (1884). On the poisonous character of certain lampreys. Vrachebnoe S t . Petersb. 5, 54-55 (in Russian). Randall, J. E. (1958). A review of ciguatera, tropical fish poisoning, with a tentative explanation of its cause. Bull. Marine Sci. Gulf Caribbean 8, 236-267. Rayner, M. D., Kosaki, T. I., and Fellmeth, E. L. (1968). Ciguatoxin: More than an anticholinesterase. Science 160, 70. Russell, F. E. (1952). Poisonous fishes. Eng. Sci. 15, 11-14. Russell, F. E. (1953). Stingray injuries: A review and discussion of their treatment. Am. J. Med. Sci. 226, 611-622. Russell, F. E. (1963). Venomous animals and their toxins. London Times Sci. Rev. 49, 10-11. Russell, F. E. (1965). Marine toxins and venomous and poisonous marine animals. Aduanc. Marine Biol. 3,255-384. Russell, F. E. ( 1967a). Pharmacology of animal venoms. Clin. Pharmacol. Therap. 8, 849-873. Russell, F. E. ( 1967b). Comparative pharmacology of some animal toxins. Federation PTOC.26, 1206-1224. Russell, F. E. ( 1969). Marine toxins. In “Encyclopedia of Pharmacology,” Pergamon Press, Oxford (in press). Russell, F. E., and Bohr, V. C. ( 1962). lntraventricular injection of venom. Toxicol. Appl. Pharmucol. 4, 16!5-173. Russell, F. E., and Emexy, J. A. (1960). Venom of the weevers Trachinzls draco and Trachinus uipera. Ann. N.Y. Acad. Sci. 90, 805-819. Russell, F. E., and Lewis, R. D. ( 1954). Evaluation of the current status of therapy for stingray injuries, Abstr. 1st Intern. Conf. Venom, Berkeley, California, 1954, Am. Assoc. Advan. Sci. Russell, F. E., and Lewis, R. D. (1956). Evaluation of the current status of therapy for stingray injuries. In “Venoms,” Publ. No. 44, pp. 43-53. Am. Assoc. Advance. Sci., Washington, D.C. Russell, F. E., and Long, T. E. (1961). Effects of venoms on neuromuscular transmission. In “Myasthenia Cravis” ( H . R. Viets, ed.), Chapter 7, pp. 101-116. Thomas, Springfield, Illinois. Russell, F. E., and van Harreveld, A. (1954). Cardiovascular effects of the venom of the round stingray Urobatis halleri. Arch. Intern. Physiol. 62, 322-333. Russell, F. E., and van Harreveld, A. (1956). Cardiovascular effects of the venom of the round stingray Urobatis halleri. In “Venoms,” Pub]. No. 44, pp. 33-41. Am. Assoc. Advance. Sci., Washington, D.C. Russell, F. E., Barritt, W. C., and Fairchild, M. D. (1957). Electrocardiographic patterns evoked by venoms of the stingray. PTOC.SOC. Exptl. Biol. Med. 96, 634-635. Russell, F. E., Panos, T. C., Kang, L. W., Warner, W. M., and Colbet, T. C. (1958a). Studies on the mechanism of death from stingray venom. A report of two fatal cases. Am. J. Med. Sci. 235, 566584. Russell, F. E., Fairchild, .M. D., and Michaelson, J. (1958b). Some properties of the venom of the stingray. Med. Arts Sci. 12, 78-86.

448

FINDLAY E. RUSSELL

Russell, F. E., OBrien, B. A., and Inaba, D. I. (1961). Venoms and neuromuscular transmission. Proc. loth Pacific Sci. Congr., Honolulu, 1961, pp. 455-456. Sahashi, Y. (1933). Nutritive value of sperm-whale oil and finback-whale oil. Sci. Papers lnst. Phys. Chem. Res. (Tokyo) 20, 245-253. Sakai, F., Sato, A., and Uraguchi, K. (1961). Uber die Atemlahmung durch tetrodotoxin. Arch. Exptl. Pathol. Pharmakol. 240, 313-321. Sano, T., Suzuki, F., Sato, S., and Iida, Y. (1968). Mode of action of new antiarrhythmic agents. Japan. Heart ]. 9, 161-168. Saunders, P. R. (1959). Venom of the stonefish Synanceja verrucosa. Science 129, 272-274. Saunders, P. R. (1960). Pharmacological and chemical studies of the venom of the stonefish (genus Synanceja) and other scarpionfishes. Ann. N.Y. Acad. Sci. 90, 798-804. Saunders, P. R., and Taylor, P. B. ( 1959). Venom of the lionfish Pterois uolitans. Am. 1. Physiol. 197, 437-440. Saunders, P. R., and Tok&, L. (1961). Purification and properties of the lethal fraction of the venom of the stonefish Synonceja horrida (Linnaeus ). Biochim. Biophys. Acta 52, 527-532. Saunders, P. R., Rothman, S., Medrano, V. A., and Chin, H. P. (1962). Cardiovascular actions of venom of the stonefish Synanceja homida. Am. 3. Phvsiol. 203, 429432. Scheuer, P. I., Takahashi, W., Tsutsumi, I., and Yoshida, T. (1967). Ciguatoxin Isolation and chemical nature Science 155, 1267-1268. Schmidt, F. T. (1874). Om fjarsingens stik og giftredskaber. Nord. Med. Arkiu 6, 1-20. Skeie, E. (1962a). The venom organs of the weeverfish (Trachinus draco). Medd. Danmarks Fiskeri-og Hauunders#gelser 3, 327438. Skeie, E. ( 196213). Weeverfish toxin. Extraction methods, toxicity determinations and stability examinations. Acta Pathol. Microbiol. Scand. 55, 168-174. Skeie, E. ( 1965). “Fjaesingstik.” Munksgaard, Copenhagen. Takahashi, D., and Inoka, Y. ( 1890). Experimentelle Untersuchungen uber das Fugugift. Arch. Exptl. Pathol. Pharmakol. 26, 45-57, Takata, M., Moore, J. W., Kao, C. Y., and Fuhrman, F. A. (1966). Blockage of sodium conductance increase in lobster giant axon by tarichatoxin (tetrodotoxin). 1. Gen. Physiol. 49, 977-988. Ten Eick, R. E., and Hoffman, B. F. (1967). The direct action of digitalis on sinus rate. Pharmacologist 9, 237. Thomson, D. A. ( 1964). Ostracitoxin: An ichthyotoxic stress secretion of the boxfish Ostracion lentiginosus. Science 146, 244-245. Tomiie, T., Furusaki, A,, Kasami, K., Yasuoka, N., Miyake, K., Haisa, M., and Nitta, I. (1963). The crystal and molecular structure of bromoanhydrotetrodoic lactone hydrobromide, a derivative of tetrodotoxin. Tetrahedron Letters No. 30, 21012104. Tsuda, K., Tamura, C., Tachikawa, R., Sakai, K., Amakasu, O., Kawamura, M., and Ikuma, S. (1963). Die Konstitution und Konfiguration der Tetrodonaaure. Chem. Phumn. Bull. (Tokyo) 11, 1473-1475. Tsuda, K., Ikuma, S., Kawamura, M., Tachikawa, R., Sakai, K., Tamura, C., and Amakasu, 0. (1964). Tetrodotoxin. VII. The structure of tetrodotoxin and its derivatives. Chem. Phamn. Bull. (Tokyo) 12, 1357-1374.

8. POISONS

AND VENOMS

449

Tsukada, 0.(1957). Sur le mkchanisme d e la bradycardie provoqu6e par le tbtrodotoxine cristallisee. Ann. Rept. Inst. Food Microbiol., Chiba Uniu. 10, 78-83. von Fraenkel, P. H., and Krick, E. S. (1945). Fish poisoning by barracuda in the Marianas. US.Naval Med. Bull. 44, 427-431. Wakely, J. F., Fuhrman, G. J., Fuhrman, F. A., Fischer, H. G., and Mosher, H. S. (1966). The occurrence of tetrodotoxin (tarichatoxin) in amphibia and the distribution of the toxin in the organs of newts (Turicha). Toricon 3, 19S203. Whitley, G. P. (1957). A kennel of frogfishes. Australian Museum Mag. 12, 139-142. Whitley, G. P., and Boardman, W. (1929). Quaint creatures of the Coral Isle. Australian Museum Mag. 3, 366-374. Wiener, S . ( 1959). Observations on the venom of the stonefish (Synanceja truchynis). Med. J . Australia 1, 620-627. Woodward, R. B. (1964). The structure of tetrodotoxin. Pure Appl. Chem. 9, 49-74. Yamagishi, S., and Sano, T. (1966). Effect of tetrodotoxin on the pacemaker action potential of the sinus node. PTOC.Japan Acad. 42, 1194-1196.

This Page Intentionally Left Blank

AUTHOR INDEX Numbers in italics refer to the pages on which complete references are listed.

A Abbott, B. C., 391, 395 Abe, T., 121, 158 Abramoff, P., 130, 131, 170 Abramowitz, A. A., 47, 65, 323, 324, 344

Aceves, J., 434, 440 Adam, H., 42, 59 Ahsan, S. N., 11, 45, 46, 50, 60, 95, 97, 110

Aida, T., 132, 135, 136, 144, 158 Alcock, A,, 28, 60 Alderice, D. F., 193, 194, 211, 212, 231, 232, 233, 235, 241, 243, 244 Aleem, A. A., 426, 442 Alender, C. B., 427, 428, 430, 440 Ali, M. A,, 217, 220, 223, 227, 242 Ali, M. Y., 232, 236, 237, 248 Amakasu, O., 431, 432, 448 Amberson, W. R., 201, 242 Ammeraal, R. N., 297, 304, 339, 352 Amoroso, E. C., 22, 25, 28, 30, 33, 37, 60, 192, 198, 242 Amouriq, L., 89, 91, 92, 109, 110 Anadbn, E., 361, 382, 394 Anders, A., 137, 158 Anders, F., 137, 158 Anderson, N. C., 432, 435, 446 Ando, S., 326, 337, 341, 342, 344 Anokhina, L. E., 180, 242 Aoki, K., 340, 345 Aoyama, T., 121, 122, 158 Arai, R., 51, 52, 53, 60, 63, 340, 344, 345

Annstrong, P. B., 201, 242 Aronson, L. R., 81, 82, 83, 88, 92, 102, 103, 106, 110, 111, 114 Aronwitz, O., 164 Arrick, M. S., 308, 347 Asano, H., 366, 398 Ashby, K. R., 142, 158, 159 Atz, J. W., 2, 4, 5, 14, 40, 41, 45, 40, 47, 48, 50, 58, 60, 67, 68, 79, 86, 94, 97, 100, 105, 115, 118, 120, 128, 154, 159, 170, 212, 239, 242, 249, 308, 323, 324, 326, 331, 343, 344, 351

Austin, C. R., 2, 60 Austin, L., 421, 440

B Babini, A., 263, 288, 274, 278, 281, 287, 296, 297, 301 Bacci, G., 124, 125, 159 Backus, R. H., 387, 393, 394 Baerends, G. P., 82, 92,- 110, 220, 242 Baerends-van Roon, J. M., 82, 110 Bagenal, T. B., 180, 182, 242 Baggerman, B., 76, 77, 78, 79, 91, 93, 94, 95, 105, 108, 110, 111 Bagnara, J. T., 309, 344 Bailey, R. J., 34, 60 Bailey, R. M., 35, 69 Baker, B. I., 324, 325, 344, 345 Baldwin, F. M., 143, 159 Balfour, F. M., 60 Ball, J. N., 18, 19, 45, 46, 47, 48, 49, 86, 89, 60, 62, 111 Balzer, G. C., Jr., 262, 303

4-51

452 Banner, A. H., 426, 427, 428, 429, 430, 440, 444 Bara, G., 52, 60, 153, 159 Bard, P., 323, 334, 349 Bardach, J. E., 178, 247 Bargen, C., 257, 298, 301 Barlow, J. W., 235, 238, 242 Barnes, B. O., 156, 159 Barr, W. A., 45, 46, 54, 58, 60 Barraud, J., 374, 394 Barrett, I., 186, 242 Barrington, E. J. W., 191, 212, 213, 242 Barritt, W. C., 416, 447 Bartsch, A. F., 427, 445 Bassot, J.-M., 363, 366, 368, 369, 370, 371, 373, 374, 375, 383, 394 Battle, H. I., 191, 193, 194, 229, 242 Baxter, I. G., 182, 183, 242 Beach, F. A., 75, 105, 106, 107, 111 Beasley, A. R., 259, 274, 301 Beebe, W., 361, 363, 381, 382, 389, 390, 391, 394, 440 Belfrich, P., 427, 428, 440 Bell, C. M., 233, 242 Bellamy, A. W., 129, 133, 134, 138, 139, 159 Bellamy, D., 315, 345 Belsare, D. K., 45, 60 Bennema, B. E., 220, 242 Bennett, F. D., 380, 385, 395 Bennink, P. J. H., 324, 335, 345 Bentley, P. J.. 57, 60 Berde, B., 336, 345 Berg, L. S., 301 Berg, O., 159 Berkowitz, P., 143, 159 Bern, H. A., 40, 48, 52, 60, 102, 105, 110, 111 Bernard, M., 233, 248 Bernstein, M. E., 433, 434, 440 Bertelsen, E., 368, 369, 381, 385, 389, 395 Berth, L., 2, 15, 18, 21, 22, 34, 35, 61 Bertolini, F., 178, 244 Best, A. C . G., 374, 395 Bickoff, E. M., 51, 61 Biely, J., 52, 53, 66 Bigelow, H. B., 380, 395 Bikle, D., 310, 312, 313, 320, 321, 329, 345

AUTHOR INDEX

Billings, V. N., 82, 88, 100, 107, 114 Bishai, H. M., 225, 228, 231, 232, 233, 242 Bitners, I. I., 52, 53, 57, 65, 66 Blacher, L. J., 129, 154, 159 Black, E. C . , 154, 165 Black, V. S., 154, 165, 263, 301 Blaxter, J. H. S., 183, 186, 192, 193, 194, 195, 196, 197, 204, 205, 206, 207, 208, 209, 211, 214, 215, 216, 217, 218, 220, 221, 222, 225, 227, 228, 229, 231, 232, 233, 236, 237, 239, 242, 243, 245, 246, 393, 395 Blum, V., 99, 100, 101, 102, 105, 111 Boardman, W., 412, 449 Bock, F., 154, 159 Boden, B. P., 356, 391, 395 Bodian, D. 277, 301 Bohr, V. C., 416, 447 Boisseau, J. P., 48, 61 B@je,O., 425, 430 Bolin, R. L., 34, 61, 357, 395 Borcea, I., 15, 61 Borison, H. L., 433, 441 Borne, R., 114 Boroughs, H., 427, 440 Bottard, A., 420, 441 Botte, V., 13, 22, 52, 55, 61, 62, 63, 69, 153, 160, 161 Botticelli, C. R., 51, 54, 55, 61, 72, 153, 174 Bourillet, F., 430, 433, 435, 441 Bowers, A. B., 240, 243 Boylan, D. B., 439, 440, 441 Boyles, M., 336, 345 Brady, W. H., 446 Brambell, F. W. R., 4, 18, 61 Brantner, G., 331, 339, 345 Brauer, A., 360, 361, 362, 363, 380, 382, 389, 390, 395 Braum, E., 193, 212, 215, 217, 221, 222, 228, 229, 243 Breder, C . M., Jr., 15, 19, 31, 33, 34, 61, 92, 111, 178, 243, 335, 343, 345 Breider, H., 132, 136, 137, 138, 139, 159, 160 Bretschneider, L. H., 19, 61, 89, 111, 156, 160 Brett, J. R., 217, 231, 232, 241, 243 Breuer, H., 51, 61

4s3

AUTHOR INDEX

Bridger, J. P., 224, 244 Bridges, C. B., 134, 160 Briot, A., 418, 441 Brock, J., 120, 160 Brock, V. E., 438, 441 Brockway, D. R., 262, 303, 304 Brooks, C. McC., 433, 434, 445 Brouwer, R., 92, 110 Brudnjak, Z., 263, 268, 270, 297, 304 Buchwald, H. D., 431, 432, 435, 446 Buck, E., 392, 395 Buck, J. B., 392, 395 Budker, P., 22, 23, 25, 30, 61 Bullough, W. S., 125, 128, 160 Bunker, N. C., 405, 425, 426, 443 Buonanno, C., 54, 61 Burger, J. W., 96, 111 Burgers, A. C. J., 324, 335, 345 Burnett, J. B., 314, 351 Burrows, R. E., 97, 112 Burzawa-Grard, E., 48, 49, 61 Buser-Lahaye, J., 46, 61 Busson-Mabillot, S., 18, 62

C Cairncross, K. D., 421, 440 Cameron, A. M., 411, 441 Cameron, G., 254, 256, 261, 274, 287, 296, 298, 301, 302 Camougis, G., 432, 441 Carl, A., 338, 352 Carlisle, D. B., 416, 441 Case, J. D., 325, 349 Castex, M. N., 407, 441 Castle, W. E., 133, 160 Cedard, L., 54, 55, 62 Cerini, C. P., 282, 285, 291, 297, 301, 303 Cerletti, A., 336, 345 ChamboIIe, P., 47, 62 Chan, S. L., 433, 441 Chaudhuri, H., 94, 103, 111 Chauvel, M., 49, 64 Chavin, W., 160, 259, 267, 269, 297, 302, 304, 314, 335, 339, 340, 341, 345, 348, 349, 352 Chen, Y. M., 314, 345 Cheng, K. K . , 433, 434, 441 Chester Jones, I., 47, 48, 54, 62, 63, 90, 111, 153, 160

Chevallier, A., 424, 441 Cheymol, J., 433, 435, 441 Chieffi, G., 3, 4, 5, 13, 18, 19, 22, 45, 47, 52, 54, 55, 60, 61, 62, 63, 67, 69, 151, 153, 160, 161, 168 Chin, H. P., 420, 421, 448 Chisholm, C., 42.6, 441 Chitwood, M. J., 412, 444 Chlopin, N. G., 255, 296, 297, 301 Choe, T. S., 202, 252 Ciereszko, L. S., 15, 62 Clark, E., 92, 111, 119, 161 Clark, F. J., 433, 446 Clark, W. G., 433, 441 Clarke, G. L., 387, 395 Clarke, R., 389, 390, 391, 395 Clarke, W. D., 392, 393, 395 Clem, L. W., 257, 258, 259, 263, 268, 279, 280, 287, 291, 298, 299, 301 Clemens, H. P., 15, 62, 81, 83, 94, 98, 111, 112, 115, 149, 158, 161, 186, 251 Clemens, W. A., 57, 65 Coates, C. W., 92, 111 Cobb, M. C., 424, 442 Cohen, D. M., 369, 381, 385, 393, 396 Cohen, H., 80, 112, 144, 161 Cohen, S. C., 430, 441 Colbet, T. C., 407, 414, 416, 447 Collenot, G., 13, 14, 62 Colton, J. B., Jr., 224, 244 Combs, B. D., 97, 112, 230, 244 Conel, J. L., 128, 161 Conover, R. J,, 387, 389, 395, 386 Coonfield, B. R., 323, 345 Cooper, L. J., 11, 62 Cooper, M. J., 425, 426, 441 Coraboeuf, E., 434, 441 Coriell, L. L., 259, 294, 302 Cormier, M. J., 378, 379, 396 Corte, F. D., 62 Cott, H. B., 390, 396 Courrier, R., 12, 62 Courtney, G., 437, 443 Coutaud, H., 425, 442 Cox, W. M., Jr., 438, 442 Craddock, J. E., 387, 393, 394 Craig-Bennett, A., 10, 12, 58, 62, 154, 161 Crane, J., 361, 363, 381, 382, 391, 394

AUTHOR INDEX

Crane, J. M., Jr., 378, 379, 391, 396 Crown, E. N., 143, 174 Currie, R. I., 393, 395

D Dalgleish, A. E., 410, 444 DAncona, U., 4, 63, 119, lu), 121, 122, ,126, 149, 161, 178, 244 Dannevig, A., 237, 239, 244 Dantschakoff, V., 144, 161 Darnell, R. M., 130, 131, 170 D’Arras, L., 437, 442 Das, S. M., 94, 112 David, C . N., 387, 389, 395, 396 Davis, D. D., 154, 175 Dawson, E. Y.,426, 442 Daykin, P. N., 203, 244 Deakins, D. E., 421, 442 Dean, B., 178, 244 Dean, F. D., 54, 63 Deane, H. W., 52, 64 De Ciechomski, J. Dz., 203, 215, 234, 244 Dederer, P. H., 255, 261, 274, 297, 301 Dee, J. E., 129, 143, 149, 164 de Groot, B., 156, 161 Deguchi, T., 432, 435, 442, 446 del Castillo, J., 438, 439, 440, 445 Della Corte, F., 47, 63 Delrio, G., 13, 52, 63 DeMarco, R., 418, 442 Denton, E. J., 312, 345, 374, 387, 396 de Sauvages, F. B., 424, 441 Dettbarn, W. D., 435, 442 Deuchar, E. M., 189, 199, 201, 244 Devillers, C., 187, 188, 201, 203, 206, 244, 261, 301 de Wit, J. J., 19, 61, 89, 111, 156, 160, 161, 162 Dexter, R. P., 257, 266, 267, 277, 281, 297, 305 Dildine, J. C., 129, 143, 162 Dimond, M. T., 213, 244 di Prisco, C . L.,52, 54, 55, 63 Ditlevsen, E., 135, 173 Dodd, J. M., 2, 4, 5, 9, 11, 13, 16, 18, 20, 40, 41, 42, 43, 46, 47, 50, 59, 63, 64, 94, 100, 105, 112, 154, 162 Donaldson, E. M., 15, 45, 46, 48, 50, 63, 72, 87, 96, 98, 114, 116

Doudoroff, P., 211, 232, 251 Drewry, G. E., 130, 166 Duchesne, E. A., 424, 441 Dudel, J., 434, 442 Duever, M. J., 228, 247 DufossB, M., 119, 162 Duhig, J. V., 421, 442 Dulbecco, R., 277, 301 Dulzetto, F., 155, 162 Dumas, R. F., 200, 202, 249 Dunbar, C. E., 257, 256,270, 275, 278, 287, 296, 297, 305 Dunn, A. E. G., 204, 247 D u k , M., 92, 112 Dupree, H. K., 94, 115 Dutt, S., 182, 183, 247 Dye, H., 48, 63 Dzwillo, M., 137, 143, 149, 158, 162

E Eales, J. G., 317, 343, 348 Eckstein, B., 52, 63 Egami, N., 48, 51, 52, 59, s3, 64, 75, 87, 99, 102, 103, 105, 112, 151, 155, 162, 163, 318, 340, 344, 345, 346 Eggert, B., 129, 163 Ehrenbaum, E., 178, 244 Ehrhardt, K., 156, 163 Eisan, J. S., 199, 249 Eider, R., 233, 244 Emert, J. T., 430, 441 Emery, J. A., 408, 409, 416, 418, 447 Enami, M., 324, 335, 346 Endean, R., 411, 412, 441, 442 Engelsen, H., 424, 442 Erlij, D., 434, 440 Essenberg, J. M., 140, 143, 163 Etoh, H., 318, 331, 336, 340, 345, 348 Evans, H. M., 405, 406, 418, 436, 442 Evennett, P. J., 9, 41, 42, 43, 47, 59, 63, 64 Everett, N. B., 4, 64 Eversole, W. J., 143, 163 Eylath, U., 52, 63

F Fagerlund, U. H. M., 90, 112 Fain, W. B., 336, 346 Fairchild, M. D., 414, 415, 416, 447

455

AUTHOR INDEX

Falco, E. A., 317, 348 Falk, S., 309, 310, 319, 346 Fange, R., 327, 335, 346, 352 Farris, D. A., 210, 244 Favard, P., 374, 394 Fedoroff, S., 258, 289, 301 Feinstein, M. B., 433, 434, 442 Fellmeth, E. L., 447 Fiedler, K., 84, 100, 101, 102, 105, 111, 112, 114

Fijan, N., 257, 283, 268, 270, 278, 281, 288, 297, 302, 304 Fingerman, M., 308, 346 Fischer, H. G., 431, 432, 435, 446, 449 Fish, C. J., 424, 442 Fitzpatrick, T. B., 308, 346 Fleischmann, W., 158, 163 Fleisher, J. H., 436, 442 Fleury, R., 400, 407, 443 Flynn, B. M., 129, 143, 149, 164 Fliichter, J., 191, 215, 244 Follenius, E., 13, 64 Follett, B. K., 57, 60 Fonnum, F., 417, 443 Fontaine, M., 49, 54, 55, 62, 64 Fontaine, Y. A., 48, 49, 61, 64 Fontenele, O., 97, 112 Forrester, C. R., 193, 194, 212, 235, 211, 244

Forselius, S., 82, 83, 84, 87, 91, 93, 112 Forster, G. K., 380, 396 Foster, M., 314, 339, 340, 349 Foulhoux, P., 433, 435, 441 Fox, D. L., 308, 309, 313, 314, 315, 346, 352

Fox, H. M., 313, 346 Franchi, L. L., 4, 5, 8, 18, 64 Frank, G. B., 433, 443 Franz, V., 309, 346 Fraser, A. C., 134, 163 Fraser, I. M., 428, 443 Fraser, J., 392, 396 Fratini, L., 9, 10, 13, 14, 64 Freeman, H. C., 52, 66, 188, 252 Freidman, B., 163 Freund, I., 128, 163 Fries, E. F. B., 331, 335, 338, 339, 346 Friess, E., 139, 163 Friis, R. R., 301 Froehlich, B., 257, 277, 297, 303

Fry, F. E. J., 238, 245 Fryer, J, L., 257, 258, 259, 266, 268, 271, 274, 278, 279, 281, 287, 291, 294, 296, 297, 301, 304 Fuhrman, F. A., 431, 432, 433, 435, 438, 443, 445, 446, 448, 449

Fuhrman, G. J., 432, 449 Fujii, R., 309, 310, 313, 318, 319, 320, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 338, 337, 38, 342, 346, 347

Fujii, Y., 329, 330, 347 Furukawa, T., 435, 438, 443 Furusaki, A,, 431, 448

G Gabler, H., 128, 163 Gabriel, M. L., 238, 237, 238, 244 Gage-Day, M., 424, 443 Gage, S. H., 424, 443 Gail, R., 421, 443 Galil, A. K. A,, 52, 64 Gallien, L., 143, 163 Garnaud, J., 34, 64 Garside, E. T., 142, 193, 194, 232, 238, 245

Geiger, E., 437, 443 Geiger, S. E., 57, 70 Gltrard, E., 49, 64 Gbrard, P., 2, 6, 8, 84 Gershon, M. D., 434, 443 Ghittino, P., 257, 263, 288, 274, 278, 281, 287, 298, 297, 301, 302 Ghosh, A., 158, 166 Gibbs, R. H., Jr., 381, 398 Gierse, A., 380, 396 Gilbert, P. W., 23, 28, 64, 69 Gilford, J. H., 254, 305 Ginsburg, A. A., 185, 245 Glaser, E., 154, 163 Goddard, C. K., 9, 20, 41, 42, 43, 47, 59, 63

Godet, R., 92, 112 Goldin, H. S., 143, 159 Goldner, H., 259, 302 Goldschmidt, R., 118, 163, 164 Goodall, R. G., 91, 113 Goodrich, E. S., 0, 8, 64 Goodrich, H. B., 129, 143, 149, 164, 261, 297, 302, 308, 343, 347

456

AUTHOR INDEX

Gorbman, A., 92, 108, 113, 114 Gordon, M., 35, 69, 118, 132, 133, 134, 137, 138, 140, 141, 142, 144, 154, 159, 160, 163, 164, 171, 261, 269, 274, 287, 298, 302, 342, 343, 347 Gorham, E., 238, 245 Goss, C. C., 430, 441 Goswami, S. V., 45, 49, 50, 70, 90, 94, 115 Goto, T., 398, 431, 443 Gottfried, H., 12, 19, 52, 54, 55, 58, 64, 69, 75, 78, 105, 112, 164 Graybill, J. R., 187, 246 Grajcer, D., 53, 64 Grand, C. G., 256, 261, 274, 287, 298, 302 Grant, F. B., 15, 62, 98, 111 Grassi, B., 128, 164 Gratzer, B. M., 322, 350 Gravell, M., 258, 266, 274, 278, 289, 291, 293, 297, 302 Gray, E. G., 328, 327, 347 Gray, J., 193, 198, 197, 200, 207, 212, 245 Green, L., 310, 321, 337, 347 Greenberg, B., 84, 113 Greenberg, S. S., 261, 269, 287, 298, 302 Greene, A. E., 259, 302 Greene, C. W., 57, 64, 380, 381, 396 Greene, H. H., 380, 381, 396 Greene, J. M., 343, 347 Gressin, L., 418, 417, 443 Grobstein, C., 155, 164, 165 Grentved, J., 381, 385, 395 Groot, C., 217, 243 Grundfest, H., 438, 446 Grutzner, L., 257, 261, 283, 286, 270, 271, 274, 277, 280, 287, 297, 298, 299, 302, 303 Gudger, E. W., 389, 396 Gustafson, G., 125, 168

H Haavaldsen, R., 417, 443 Habekost, R. C . , 426, 443 Hadley, M. E., 338, 346 Haedrich, R. L., 387, 393, 394 Haempel, O., 154, 163 Hagen, D. W., 78, 113

Hagiwara, S., 432, 434, 443 Haisa, M., 431, 448 Hale, E. B., 84, 113 Hall, W. B., 182, 250 Halstead, B. W., 402, 405, 406, 407, 409, 410, 412, 424, 425, 426, 427, 429, 430, 435, 442, 443, 444, 446 Ham, A. W., 13, 18, 64 Hama, T., 309, 315, 316, 347 Hamada, J., 432, 444 Hamdorf, K., 233, 245 Handrick, K., 380, 396 Haneda, Y., 356, 359, 363, 366, 367, 368, 369, 375, 376, 377, 378, 379, 383, 384, 385, 388, 392, 393, 396, 397, 398, 399, 400 Hara, T. J., 92, 108, 113 Harder, W., 203, 217, 245 Hardisty, M. W., 13, 65, 128, 165 Harms, J. W., 139, 165, 359, 397 Harms, W., 139, 165 Harrington, R. W., Jr., 2, 59, 65, 120, 165 Harrison, C. M. H., 393, 397 Harrison, C. S . , 436, 442 Hartmann, M., 132, 149, 165 Harvey, E. N., 358, 359, 360, 363, 388, 375, 381, 383, 384, 385, 389, 390, 397 Hasegawa, H., 315, 316, 347 Hashimoto, Y., 427, 438, 444, 446 Haskins, C . P., 130, 165 Haskins, E. F., 130, 165 Hasler, A. D., 214, 216, 217, 247 Hastings, J. W., 356, 397 Hatanaka, M. A,, 209, 251 Hayama, T., 433, 444 Hayes, F. R., 189, 191, 193, 198, 197, 198, 201, 202, 203, 205, 206, 207, 208, 232, 234, 238, 245 Healey, E. G., 318, 324, 326, 327, 328, 330, 335, 336, 337, 338, 347, 348 Heard, W. R., 223, 245 Hediger, H., 78, 113 Helfrich, P., 426, 427, 428, 429, 430, 437, 440, 444 Hempel, G., 183, 188, 192, 193, 195, 198, 197, 214, 217, 218, 222, 229, 236, 237, 243, 245, 250

457

AUTHOR INDEX

Henn, A. W., 15, 65 Hessel, D. W., 427, 429, 444 Hester, M. J., 180, 245 Heuts, M. J., 236, 245 Hewitt, R. E., 130, 165 Hickling, C. F., 134, 165, 238, 245, 380, 385, 397 Higman, H. B., 435, 442 Hild, S., 139, 165 Hildemann, W. H., 80, 113, 143, 155, 165 Hill, A. V., 318, 348 Hill, G. A., 308, 347 Hinde, R. A., 74, 103, 108, 113 Hirata, Y.,398, 431, 443 Hisaw, F. L., 18, 45, 47, 48, 51, 54, 55, 61, 65, 72, 153, 174 Hishida, T., 151, 165, 201, 245, 314, 342, 348, 352 Hitchings, G. H., 317, 348 Hiyama, Y., 424, 426, 444 Hoar, w. s., 2? 11, 12, 13, 1%18, 19, 21, 22, 40, 42, 47, 48, 50, 55, 58, so, 6% 757 76’ 77’ 78y 79’ 87’ 94, 95, 96, 97, 101, 102, 105, 106, 107, 110, 113, 115, 116, 154, 165, 216, 223, 227, 233, 240, 242, 246, 247, 343, 348 Hobson, B. M., 54, 58, 60 Hoek, P. P. C., 120, 165 Hoffman, B. F., 434, 448 Hogben, L. T., 317, 324, 348 Holliday, F. G. T., 200, 204, 205, 206, 207, 208, 224, 225, 246, 247, 343, 351 Holly, M., 166 Holmberg, K., 310, 323, 348 Holz-Tucker, M., 83, 88, 110 Homans, R. E. S., 57, 65 Hopper, A. F., 155, 166 Horton, H. F., 187, 246 Hosoya, Y.,435, 436, 443 Hoyle, R. J., 186, 187, 246, 252 Hu, F., 259, 267, 269, 297, 302, 304, 339, 340, 348, 349, 352 Hubbs, C . L., 129, 166 Hubbs, L. C., 129, 166 Hubbs, C., Jr., 130, 166 Hubbs, C., 130, 166, 183, 246 Hulet, W. H., 363, 398

Humm, D. G., 343, 348 Humm, J. H., 343, 348 Hunt, S. U., 52, 54,55, 64, 69, 164 Huntsman, A. G. 392, 398 Huxley, J., 117, 166 Hwang, P. C., 187, 249 Hyde, J. E., 327, 351

I Idler, D. R., 51, 52, 53, 54, 57, 64, 65, 66, 69, 186, 187, 246, 249, 252 Iga, T., 343, 348 Iida, Y.,434, 448 Ikeda, K., 76, 113, 154, 166 Ikuma, S., 431, 432, 448 Imai, K., 325, 348 Irnparato’E’’ 54y Inaba, D. I., 435, 448 Inoko, Y., 433, 448 Inslee, T., 81, 83, 112 Ishibashi, T., 322, 333, 344, 348 Ishida, M,, 202, 248 Ishihara, F., 432, 433, 435, 444 Ishii, S., 48, 63, 66, 99, 102, 105, 112, 113, 155, 163 K., 437, 445 Itazawa, y., 237, 246 Iverson, E. S., 238, 246 Ivlev, V. S., 204, 205, 206, 207, 208, 209, 216, 218, 219, 228, 246 Iwai, T., 215, 218, 221, 246, 359, 360, 361, 364, 366, 368, 398 Iwata, K. S., 327, 330, 332, 335, 337, 338, 348, 353 Izumi, I., 330, 337, 353

J Jacobs, W., 225, 246 Jakovleva, G. S., 257, 297, 304 Jensen, A. S., 425, 444 Jensen, M. H., 257, 270, 278, 281, 296, 302 Johann, L., 380, 385, 398 John, K. R., 214, 216, 217, 247 Johns, L. S., 83, 84, 85, 88, 103, 109,

114 Johnson, F. H., 356, 366, 367, 368, 369, 375, 376, 377, 378, 384, 386, 393, 397, 398

458

AUTHOR INDEX

Johnston, C. E., 317, 343, 348 Jones, G., 421, 442 Jones, J. W., 85, 114 Jones, M. P., 200, 204, 220, 224, 225, 243, 246, 247

Jordan, D. S., 124, 166 Julesz, M., 106, 114

K Kahn, R. H., 282, 303 Kaighn, M. E., 189, 247 Kaiser, E., 424, 444 Kaiser, P., 132, 142, 166 Kajishima, T., 132, 149, 175, 342, 349 Kallman, K. D., 130, 131, 132, 133, 134, 138, 166, 170 Kalman, S. M., 199, 247 Kamada, T., 318, 321, 332, 334, 342, 349

Kamishima, Y., 311, 312, 317, 349 Kampa, E. M., 356, 391, 395 Kandler, R., 182, 183, 247 Kaneda, T., 438, 444 Kang, L. W., 407, 414, 416, 447 Kann, S., 156, 163 Kanter, A. E., 156, 159 Kao, C. Y.,431, 432, 433, 435, 436, 445, 448

Kar, A. B., 156, 166 Kasami, K., 431, 448 Kato, K., 366, 398 Kawabata, T., 437, 445 Kawaguchi, K., 121, 166 Kawaguti, S., 311, 312, 317, 349 Kawakami, K., 433, 445 Kawakami, M., 75, 114 Kawamura, M., 431, 432, 448 Keenleyside, M. H. A., 91, 113, 227, 247 Kellicott, W. E., 197, 206, 207, 250 Kent, A. K., 324, 325, 349 Ken, J. G., 5, 66 Khan, H. A., 94, 112 Kier, A,, 364, 398 Killos, P. J., 436, 442 Kim, K., 297, 302, 304, 314, 339, 340, 345, 349, 352

Kimura, S., 438, 445 King, G . M., 85, 114

Kinne, E. M., 192, 193, 194, 212, 59.5, 247

Kinne, O., 192, 193, 194, 211, 212, 235, 247, 251

Kinoshita, J., 121, 166 Kinoshita, Y., 124, 166 Kinosita, H., 318, 319, 320, 321, 332, 334, 335, 345 349 Kirschstein, H., 343, 350 Kishi, Y., 398, 443 Kitajima, T., 121, 158 Klaw, A. H., 156, 159 Kleeman, K. T., 259, 304 Kleiner, I. S., 156, 167 Kleinhaus, A. L., 433, 445 Knight-Jones, E. W., 226, 249 Kobayashi, M., 330, 335, 338, 353 Kobayashi, T., 435, 441 Kohler, V., 326, 349 Koizumi, K., 433, 434,445 Kopac, M. J., 281, 269, 287, 298, 302 Kop66 S., 154, 167 Kopech, G., 254, 303 Komalik, F., 424, 445 Kosaki, T. I., 447 Kosswig, C., 132, 133, 134, 130, 137, 138, 160, 167 Kosto, B., 314, 339, 340, 349, 351 Kreiss, P., 379, 396 Krick, E. S., 430, 449 Kuga, T., 435, 445 Kiihn, K., 156, 163 Kumar, L. A., 202, 252 Kumpf, K. F., 82, 83, 86, 88, 100, 107, 114, 142, 169 Kunst, L., 257, 263, 288, 270, 277, 278, 281, 288, 297, 302, 304 Kurata, H., 218, 247 Kuriaki, K., 435, 445 Kurihara, T., 327, 330, 332, 335, 348 Kuriyama, H., 434, 445 Kuroda, N., 122, 167 Kurose, T., 435, 445 Kuthalingam, M. D. K., 230,247

L Laale, H. W., 194, 247 Lagios, M. D., 13, 14, 15, 66 Lagler, K. F., 178, 247

459

AUTHOR INDEX

Lakshman, A. B., 94, 115 Lam, T. J., 102, 114 La Marca, M. J., 22, 66 Lambed, J. G. D., 56, 66 Land, M. F., 312, 345, 374, 398 Lane, E. D., 390, 398 Langer, W. F., 35, 66 Laming, W. J. R., 323, 349 Larimore, R. W., 228, 247 Larraiieta, M. G., 122, 126, 167 Larsen, L. O., 41, 42, 66 Lasker, R., 193, 197, 198, 200, 204, 205, 206, 207, 208, 210, 211, 246, 247 Laskowski, W., 80,114, 144, 167 Lavenda, N., 122, 167 Leatherland, J. F., 43, 66, 96, 102, 114, 115

Lee, R. K. C., 430, 445 Le Gall, J., 120, 167 Lehrman, D. S., 74, 103, 114 Leigh-Sharpe, W. H., 22, 66, 67 Lepori, N. G., 124, 143, 167, 168 Lemer, A. B., 325, 349 Levan, A., 259, 294, 302 Levene, A. L., 308, 346 Levine, D. C., 433, 434, 445 Levy, M., 82, 114 Lewis, M. R., 256, 261, 264, 287, 296, 299, 302 Lewis, R. D., 405, 406, 407, 447 Lewis, R. M., 231, 235, 247 Li, K. M., 425, 429, 433, 434, 441, 445 Li, M. F., 258, 269, 278, 281, 291, 298, 302 Lieder, U., 130, 168 Liem, K. F., 125, 168 Lieberman, S., 152, 172 Liley, N. R., 45, 67, 81, 83, 84, 85, 86, 87, 88, 92, 96, 103, 104, 105, 109, 114 Lillelund, K., 182, 192, 193, 212, 247 Lindroth, A., 205, 207, 232, 247 Lindsey, C. C., 232, 236, 237, 248 Linstow, 0. V., 424, 445 Lipsius, M., 433, 445 Lissia, Frau, A. M., 126, 168 Liu, C. K., 12.5, 168 Lively, W. M., 430, 444 Livingstone, D. A., 203, 206, 207, 232, 245

Lockwood, A. P. M., 261, 264, 302 Lofts, B., 13, 14, 45, 46, 47, 67 Long, T. E., 416, 447 Lhneberg, E., 125, 168 Loud, A. V., 297, 303, 310, 315, 334, 339, 349 Loza, F., 407, 441 Lozano, Cabo, F., 124, 168 Lubosch, W., 128, 168 Lundstrom, H. M., 323, 334, 349 Lupo, C., 13, 52, 54, 62, 67, 153, 160, 168 Lutwak-Mann, C., 23, 67 Lyall, A. H., 220, 248 Lyngnes, R., 15, 18, 67

M McAllister, D. E., 359, 392, 393, 398 McBride, J. R., 49, 57, 70, 71 McCallion, D. J., 194, 247 McCarthy, L. E., 433, 441 McCollum, I. A. N., 421, 440 McDermitt, C., 81, 112 McElroy, W. D., 376, 384, 386,399 McFalls, F. D., 266, 267, 279, 298, 303 McFarren, E. F., 427, 445 McGovern, V. J., 308, 346 MacGregor, J. S . , 182, 248 MacHemer, L., 84, 101, 102, 114 McHugh, J. L., 233, 236, 248 MacIntre, P. A., 137, 138, 168 MacKinnon, D., 91, 113 McLeod, J., 120, 168 McLimans, W. F., 287, 303 MacNab, H. C., 51, 52, 53, 65 MacNeal, P. S., 256, 261, 287, 298, 299, 302 Magnuson, J. J., 211, 248 Mahadevan, G., 67 Malsberger, R. G., 258, 266, 274, 278, 282, 285, 289, 291, 293, 297, 301, 302, 303 Mandl, A. M., 4, 5, 64 Mann, T., 23, 67 Mansueti, R., 178, 192, 193, 248 Maretik, Z., 418, 445 Maretzki, A., 438, 439, 440, 445 Marinaro, J. Y., 233, 248 Markee, J. E., 266, 267, 279, 298, 303 Markert, J. R., 317, 349

460 Markov, S., 437, 445 Marr, D. H. A,, 195, 198, 197, 212, 248 Man, J. C., 230, 248 Marshall, A. J., 13, 67 Marshall, F. H. A., 58, 67 Marshall, N. B., 178, 248, 388, 390, 391, 398 Marsland, D. A., 319, 321, 341, 350 Marumo, R., 121, 166 Matsuda, N., 87, 116, 146, 149, 152, 175 Matsumoto, J., 309, 310, 311, 315, 316, 340, 350, 438, 445 Matsumura, F., 435, 445 Matsuo, N., 438, 445 Matsuo, R., 426, 446 Matthews, L. H., 9, 13, 14, 16, 2.0, 22, 28, 58, 67 Matthews, S. A,, 10, 67 Matty, A. J., 40, 42, 67 Mead, G. W., 120, 168 Mercer, R. N., 129, 143, 149, 164 Medrano, V. A., 420, 421, 448 Meisner, D., 319, 350 Mellinger, J., 9, 67, 323, 324, 350 Merchant, D. J., 282, 303 Metten, H., 19, 20, 67 Meyer, H., 130, 168 Meyers, A. S., 152, 168 Michaelson, J., 414, 415, 416, 447 Michl, H., 424, 444 Miles, P. S., 426, 444 Miller, R. R., 131, 168, 178, 247 Mills, D. H., 182, 183, 249 Mishima, Y., 297, 303, 308, 310, 315, 334, 339, 346, 349 Mishkind, D. I., 156, 167 Mitchell, B. S., 48, 49, 69 Miyake, K., 431, 448 Miyamori, H., 143, 168 Miyazawa, K., 438, 446 Miyoshi, S., 331, 333, 350 Mime, K., 121, 158 Modglin, F. R., 406, 407, 409, 412, 444, 446 Moewus, L., 257, 258, 263, 268, 279, 280, 287, 291, 298, 299, 301, 303 Mohsen, M., 35, 67 Mohnder, A. R., 236, 237, 248 Molander-Swedmark, M., 236, 237, 248 Monroy, A., 202, 248

AUTHOR INDEX

Montalenti, G., 178, 244 Moore, J. W., 432, 435, 436, 446, 448 Mori, M., 438, 446 Mori, W., 336, 350 Mori, Y., 332, 353 Morley, R. B., 186, 252 Morris, R. W., 215, 239, 248 Morrow, J. E., Jr., 381, 398 Moser, H. G., 11, 35, 67 Mosher, H. S., 431, 432, 435, 443, 446, 449 Mounib, M. S., 187, 199, 249 Mr%, W., 128, 168, 169 Muira, T., 437, 445 Muller, H., 220, 249 Muller, H. J., 131, 169 Munk, O., 368, 369, 395 Murphy, W. H,, Jr., 282, 303 Murray, M. R., 254, 303 Murtha, E. F., 433, 434, 446 Mud, G., 363, 398

N Nachmansohn, D., 435, 442 Nagahama, H., 332, 350 Nagao, K., 332, 335, 348 Nagata, Y., 155, 169 Naitoh, T., 332, 353 Nakajima, S., 432, 434, 436, 443, 446 Nakamura, Y., 436, 446 Nakanishi, Y.,438, 446 Nakano, E., 201, 202, 204, 245, 248, 249, 378, 379, 396 Nambu, M., 59, 64, 75, 99, 103, 112 Nandi, J., 40, 48, 52, 60, 68 Narahashi, T., 432, 435, 436, 446 Nayyar, S. K., 45, 46, 49, 50, 70 Neckel, I., 317, 350 Needham, J. 22, 25, 68, 201, 249 Nelsen, 0. E., 4, 13, 14, 18, 68 New, D. A. T., 187, 192, 249 Newth, D. R., 342, 350 Neyfakh, A. A,, 234, 249 Ng, M.-H., 433, 446 Nicholas, J. H., 196, 197, 250 Nichols, W, W., 259, 294, 302 Nicol, J. A. C., 312, 345, 356, 359, 360, 363, 364, 374, 380, 381, 382, 383, 384, 385, 386, 387, 389, 392, 395, 396, 399

461

AUTHOR INDEX

Niiya, I., 438, 445 Nikolsky, G. V., 178, 182, 186, 204, 234, 249 Nitta, I., 431, 448 Niwa (Suzuki), H., 154, 156, 169 Noble, G. K., 82, 83, 86, 88, 100, 114, 142, 169 Nogusa, S., 132, 169 Nomiyama, S., 433, 446 Nomura, T., 54, 55, 62 Norman, J. R., 178, 249 Noumura, T., 21, 68 Novales, B. J., 310, 318, 321, 322, 350 Novales, R. R., 308, 310, 318, 320, 322, 323, 328, 329, 330, 333, 336, 338, 350, 353 Nusbaum-Hilarowicz, J., 363, 399

225,

107,

323,

321, 335,

0 O'Brien, B. A,, 435, 448 Ocampo, R. R., 406, 407, 444, 446 Ochs, S . , 433, 446 Odiorne, J. M., 307, 331, 339, 350 Oettle, A. G., 308, 346 Officer, J. E., 293, 303 Ogura, Y., 433, 444, 446 Oguro, C . , 21, 68 Ohkubo, Y., 435, 446 Ohshima, H., 380, 385, 399 Ohtsuka, E., 185, 249 Oka, T. B., 129, 155, 156, 169 Okada, Yi3 K., 81,87, 114, 121, 122, 124, 143, 155, 169, 170, 358, 366, 384, 399 Okamura, O., 361, 364, 398 Okkelberg, P., 128, 170 Oksche, A., 343, 350 Uktay, M., 136, 137, 167, 170 Olsen, A. G., 54, 72, 153, 174 Oppenheimer, J. M., 187, 189, 249, 261, 297, 303 Ortiz-Muniz, G., 259, 303 Orton, J., 214, 225, 249 Orton, G. L., 342, 350 Osa, T., 434, 445 OsadZaja, E. F., 297, 303 Oshima, K., 92, 108, 114

Osowski, H., 255, 261, 297, 303 Ota, F., 317, 350 Otuska, S., 49, 68 Ozon, R., 13, 14, 51, 61, 62 P

Padoa, E., 142, 170 Paimre, M., 433, 434, 442 Palay, S. L., 89, 115 Pandey, S., 11, 13, 14, 15, 42, 43, 45,46, 47, 66, 68, 96, 115 Pang, H. Q., 430, 445 Panos, T. C . , 407, 414, 416, 447 Parker, G. H., 307, 308, 309, 323, 326, 327, 328, 329, 330, 333, 334, 337, 342, 344, 351 Parker, R. C., 254, 303 Parkinson, J. L., 318, 348 Parr, A. E., 358, 399 Parrish, B. B., 182, 183, 249 Parry, G., 343, 351 Pasquali, A,, 120, 170 Patten, B. M., 13, 68 Paul, J., 254, 266, 303 Pearcy, W. G., 231, 249, 391, 393 Pearson, R. B., 407, 446 Peckman, N. H., 427, 429, 444 Peden, A. E., 183, 246 Pelluet, D., 196, 197, 238, 245 Peluse, M., 259, 294, 302 Peper, K., 434, 442, 446 Pereira, R. S., 264, 303 Perks, A. M., 100, 105, 112 Perlmutter, A,, 263, 279, 297, 304 Peters, G., 137, 140, 141, 144, 170 Peters, H. M., 183, 249 Pfeiffer, H., 256, 296, 303 Pfeiffer, W., 438, 446 Pfitzner, I., 257, 263, 270, 271, 277, 287, 298, 297, 298, 303 Philippi, E., 15, 68, 140, 170 Phillips, A. M., Jr., 200, 202, 249, zS2, 303, 304 Phillips, C . , 446 Phillips, J. G., 90, 111, 153, 160 Phisalix, C., 418, 446 Phisalix, M., 425, 446 Picciolo, A. R., 89, 115

462

AUTHOR INDEX

Radhakrishman, N., 433, 441 Rageau, J., 421, 443 Raitt, D. F. S., 182, 250 Ramaswami, L. S., 49, 68, 94, 115 Randall, J. E., 192, 250, 424, 426, 437, 447 Ranney, B. K., 432, 443 Ranzi, S., 22, 25, 27, 28, 29, 39, 47, 69, 178, 244 Rasch, E. M., 130, 131, 170 Rasquin, P., 325, 325, 334, 335, 343, 345, 351 Rauther, M., 389, 391, 399 Ray, D. L., 380, 399 Rayner, M. D., 427 Razzauti, A., 124, 159 Read, L. J,, 39, 69, 202,250 Rebhun, L. I., 321, 351 Redlich, A., 91, 113 RBgnier, M. T., 143, 171 Reid, E. E., 438, 442 Reidinger, L., 335, 337, 339, 351 Reinboth, R., 4, 69, 118, 121, 122, 123, 124, 126, 171 Rene, J., 432, 441 Rhodin, J., 309, 310, 319, 346 Rice, A. L., 226, 249 Richards, M. P. M., 75, 115 Rijavec, L., 126, 171 Riley, J. D., 240, 250 Ringler, I., 54, 72 Roberts, F. L., 259, 264, 277, 278, 279, 296, 297, 298, 299, 304 Robertson, 0. H., 330, 334, 336, 337, 338, 341, 342, 351 Q Roede, M. J., 124, 171 Qasim, S. Z., 226, 240, 249 Romer, A. S., 8, 22, 69 Quastel, J. H., 433, 441 Ronald, A. P., 52, 53, 66 Queal, M., 129, 133, 138, 139, 159 Rosen, D. E., 15, 19, 31, 33, 34, 35, 61, Quemer, H., 143, 170 69, 178, 243 Quevedo, W. C., Jr., 346 Rosenberg, P., 435, 442 Quimby, M. C., 257, 258, 259, 284, 285, Rosenblueth, A., 329, 351 266, 267, 288, 270, 274, 276, 277, Rosenthal, H., 191, 214, 215, 217, 222, 279, 281, 289, 291, 296, 297, 299, 227, 228, 229, 244, 250 305 Ross, D. M., 327, 330, 335, 336, 337, 338, 348 R Roth, W. D., 54, 61 Rothman, S., 420, 421, 448 Rachlin, J. W., 263, 279, 297, 304 Rothwell, B., 13, 65 Radakov, D. V., 227, 228, 250

Pickford, G. E., 2, 14, 40, 41, 45, 46, 47, 48, 50, 54, 58, 67, 68, 70, 79, 86, 94, 97, 99, 100, 105, 115, 116, 154, 170, 212, 249, 308, 314, 323, 324, 326, 331, 339, 340, 349, 351 Pigulevsky, S. V., 424, 446 Pilcher, K. S., 258, 259, 268, 271, 278, 281, 291, 296, 301, 304 Pinsky, C., 433, 443 Piyakamchana, T., 426, 444 Plumb, J. A., 283, 289, 304 Podoliak, H. A., 262, 303, 304 Pohl, J., 416, 447 Pope, J. A., 182, 183, 249 Popoff, W., 139, 170 Porta, A., 406, 407, 447 Porte, A., 13, 64 Porter, K. R., 310, 312, 313, 320, 321, 329, 345 Portman, A., 6, 7, 68 Poston, R. N., 432, 446 Potter, I. C., 259, 265, 268, 272, 276, 296, 304 Potts, W. T. W., 199, 249, 250 Prasad, R. R., 20, 68 Prescott, D. M., 185, 199, 249 Price, J. W., 193, 212, 232, 249 Prichard, P., 379, 396 Prokhoroff, P., 424, 447 Prosser, C . L., 23, 67 Pye, J. D., 327, 341, 351 Pyle, E. A., 257, 266, 267, 277, 281, 297, 305

463

AUTHOR INDEX Rounsefell, G. A., 180, 182, 183, 250 Rubin, A. A., 154, 171 Rudnick, D.,187, 250 Rudolf, H., 128, 171 Rudy, P. P., Jr., 199, 249, 250 Russell, F. E., 401, 405, 408, 407, 408,

409, 412, 414, 415, 416, 418, 421, 423, 424, 428, 429, 430, 435, 447, 448 Rust, W., 171 Ryland, J. S., 192, 196, 197, 210, 211, 216, 217, 224, 227, 228, 250

5

Schreiner, K. E., 128, 171 Schroeder, W. C., 380, 395 Schultz, R. J., 131, 168, 171 Schumann, G. O., 239, 250 Schwassmann, H.O., 220, 250 Schwier, H., 128, 138, 171 Scott, C. G., 197, 206, 207, 250 Scott, D.P., 180, 250 Scott, G. T., 327, 335, 336, 337, 338, 351 Scott, M. A., 258, 268, 269, 270, 286, 287, 289, 296, 298, 299, 304 Scott, W.R., 438, 446 Scrimshaw, N. S., 31, 34, 39, 69 Seeley, R. J., 263, 279, 297, 304 Seghers, B. H., 83, 84, 85, 88, 103, 109, 114 Sengiin, A,, 137, 171 Sette, O.,192, 193, 210, 250 Seymour, A., 237, 251 Shann, E. W., 28, 69 Shaw, E.,227, 251 Shaw, S. W., 428, 430, 440 Shearer, W. M., 182, 183, 249 Shelbourne, J. E., 195, 210, 213, 216, 218, 225, 234, 239, 240, 251 Shephard, D. C., 342, 351 Shimomura, O.,389, 375, 370, 397, 398 Shirai, K., 19, 69, 89, 90, 103, 115 Shoemaker, J. D., 15, 62 Shores, D. L., 387, 393, 394 Shumway, D. L., 211, 232, 251 Sie, E. H.-C., 375, 376, 377, 384, 388, 397, 398, 399 Sigel, M. M.,257, 258, 259, 283, 288, 279, 280, 287, 291, 298, 299, 301,

Sacks, W. B., 139, 171 Sage, M., 91, 115 Sahashi, Y., 438, 448 Saiga, Y., 375, 376, 398 Saito, T.,438, 446 Sakai, F.,433, 448 Sakai, K., 431, 432, 448 Samokhvalova, G. V., 154, 171 Sanders, M., 258, 268, 270, 274, 297, 304 Sano, T.,434, 448, 449 Sanzo, L., 178, 244 Sasaki, S., 427, 428, 440 Sasaoka, T.,435, 430, 443 Sato, A., 433, 448 Sato, S., 434, 448 Saunders, P. R., 420, 421, 442, 448 Saville, A,, 183, 249 Sawaya, P.,284, 303 Sawyer, C. H., 75, 114 Sawyer, W.H., 99, 105, 116 Schall, D.W., 425, 444 Scharrer, E., 89, 115 303 Scheline, R. R., 327, 335, 336, 337, 338, Siegman, M. J., 433, 445 351 Silver, S. J., 211, 232, 251 Scheuer, P. J., 427, 428, 429, 439, 440, Simon, P., 433, 435, 441 440, 441, 448 Simpson, A. C., 187, 193, 223, 251 Schlernitzauer, D. A., 23, 28, 64, 69 Simpson, T. H., 14, 43, 44,'52, 54, 55, Schlesinger, W., 341, 345 59, 64, 69 Schlumberger, H.G., 258, 261, 287,297, Skeie, E., 409, 417, 419, 448 304 Skowron, S., 385, 399 Schmidt E., 132, 142, 166 Slome, D.,317, 348 Schmidt, F. T., 417, 448 Schmidt, P. J., 48, 49,52, 53, 54, 66, 69 Smith, C. L., 118, 119, 122, 128, 172 Smith, D. C., 318, 351 Schmiegel, J. L., 432, 443 Schnakenberg, G.,437, 443 Smith, G. M., 132, 164

464

AUTHOR INDEX

Smith, M., 48, 49, 69 Smith, R. J. F., 79, 84, 95, 101, 105, 107, 115 Smith, S., 183, 187, 189, 190, 197, 198, 200, 201, 202, 208, 251 Smith, S. W., 89, 115 Sneed, K. E., 94, 112, 115, 186, 251 Snyder, J. O., 124, 166 Solandt, D. Y., 318, 348 Solberg, A. N., 233, 251 Solemdal, P., 183, 225, 251 Solomon, S., 152, 172 Sone, H., 438, 445 Sordi, M., 124, 172 Soret, M. G., 250, 266, 270, 274, 297, 304

Spaeth, R. A,, 318, 332, 333, 334, 341, 344, 351 Sparta, A., 178, 244 Spitz, L. M., 314, 351 Spurway, H., 130, 172 Stabile, D. E., 433, 434, 446 Stahl, A., 325, 351 Staines, M., 217, 220, 222, 229, 243 Stanley, H. P., 9, 13, 22, 61, 69 Stanley, L. L., 90, 115 Steche, O., 383, 399 Steele, K., 13, 65 Stephenson, E. M., 259, 285, 268, 272, 276, 289, 296, 304 Stevens, R. E., 94, 115 Stevenson, J. C., 224, 251 Stevenson, M. M., 183, 246 Stewart, J. E., 258, 269, 278, 281, 291, 296, 302 Stalk, A., 47, 56, 70, 130, 172, 343, 352 Strasburg, D. W., 224, 251 Strum, J., 373, 374, 399, 400 Stuart, T. A., 204, 223, 234, 251 Sugiyama, N., 375, 376, 398 Suilar, F., 407, 441 Sumner, F. B., 315, 352 Sundararaj, B. I., 45, 46, 49, 50, 70, 90, 94, 115 Suzuki, F., 434, 448 Suzuki, H., 154, 156, 175 Suzuki, T., 433, 445 Sweet, J. G., 211, 212, 251 Swift, D. R,, 54, 58, 70

Sykes, J. E., 251 Syrski, S., 120, 172

T Tachi, C., 340, 345 Tachikawa, R., 431, 432, 448 Taguchi, S., 3.20, 347 Tait, J. S., 225, 251 Takahashi, D., 433, 448 Takahashi, H., 438, 445 Takahashi, M., 209, 251 Takahashi, S., 431, 443 Takahashi, W., 428, 448 Takai, M., 146, 149, 175 Takata, M., 436, 448 Takeuchi, K., 146, 149, 155, 172, 175, 315, 352 Takman, B. H., 432, 441 Talbot, G. E., 251 Tamaoki, B., 53, 60 Tamashita, A., 438, 445 Tamura, C., 431, 432, 448 Thing, A. V., 235, 236, 237, 252 Tasse, P., 432, 441 Tavolga, M. C., 80, 115, 144, 172 Tavolga, W. N., 81, 82, 85, 89, 96, 108, 109, 115 Taylor, A. B., 155, 172 Taylor, F. H. C., 393, 400 Taylor, P. B., 420, 421, 448 Tchen, T. T., 297, 302, 304, 314, 339, 340, 345, 349, 352 Tchernavin, V. V., 359, 361, 383, 382, 389, 400 Teal, J. M., 387, 393, 394 Tec, V. I., 257, 297, 304 Templeman, W., 183, 192, 252 Ten Eick, R. E., 434, 448 Terner, C., 199, 201, 202, 215, 252 Tescher, G. L., 90, 115 Tbtreault, L., 435, 441 Te Winkel, L. E., 23, 25, 29, 70 Te Winkel, L. E., 198, 252 Thacker, G. T., 240, 250 Theilacker, G. H., 205, 206, 207, 247 Theisen, B., 369, 395 Thiebold, J. J., 20, 70 Thomson, D. A,, 438, 439, 440, 448 Threadgold, L. T., 204,247

465

AUTHOR INDEX

Tilney, L. G., 310, 312, 313, 320, 321, 329, 345 Toida, N., 434, 445 TokBs, L., 420, 421, 448 Tomagec, I., 263, 268, 270, 297, 304 Tomiie, T., 431, 448 Tomita, H., 314, 342, 348, 352 Tomlinson, N., 57, 70 Tompsett, D. H., 389, 400 Tortonese, E., 23, 70 Townsley, P. M., 258, 286, 269, 270, 286, 287, 289, 296, 298, 299, 304 Tozawa, T., 154, 156, 172 Trautwein, W., 434, 442, 446 Trinkaus, J. P., 184; 187, 188, 252 Truscott, B., 52, 53, 54, 66, 186, 252 Tsuchiya, K., 332, 353 Tsuda, K., 431, 448 Tsuji, F. I., 377, 400 Tsukada, O., 434, 449 Tsutsumi, J., 428, 448 Tsuyuki, H., 48, 49, 52, 66,69 Turner, C. D., 50, 70 Turner, C. L., 10, 17, 30, 31, 35, 36, 37, 38, 39, 40, 70, 71, 155, 172 Turner, W. J., 338, 352

van Overbeeke, A. P., 49, 71 Vanstone, W. E., 317, 349 Vassort, G., 434, 441 Vaughn, R. R., 262, 304 Vaupel, J., 14, 71 Vellano, C., 55, 63 Vevers, G., 313, 346 Vilter, V., 213, 252 Visca, T., 153, 161 Vivien, J., 40, 71, 100, 116, 143, 154, 155, 173 Vladykov, V. D., 182, 252 Vogelzang, A. A,, 220, 242 Vogt, M., 277, 301 von Euler, U. S., 327, 352 von Fraenkel, P. H., 430, 449 von Frisch, K., 327, 335, 343, 359 von Ielei, 327, 352 von Ihering, R., 15, 71 von Ledebur, J. F., 225, 252 von Lendenfeld, R., 360, 400

U

Wakely, J. F., 432, 449 Walcher, H., 326, 327, 352 Walton, A., 2, 60 Walvig, F., 5, 9, 15, 71 Warburton, B., 130, 166 Wardle, C . S., 14, 43, 44, 52, 59, 69 Waring, H., 308, 323, 325, 352 Warner, W. M., 407, 414, 416, 447 Warren, C . E., 211, 232, 251 Warren, F. J., 387, 396 Watanabe, M., 327, 330, 331, 332, 335, 337, 338, 344, 348, 352, 353 Waterbolk, H. Tj., 92, 110 Waterman, T. H., 385, 389, 400 Watson, M. S., 254, 304, 305 Wattenberg, L. W., 52, 71 Wedemeyer, G., 199, 252 Weisel, G . F., 34, 71, 323, 324, 325,

Ueda, K., 92, 108, 113, 329, 352 Uladykov, V. D., 57, 65 Umrath, K., 326, 327, 330, 335, 337, 338, 339, 340, 351, 352 Uraguchi, K., 433, 448 Urakawa, N., 435, 446 Urist, M. R., 204, 165, 304

V van den Assem, J., 104, 116 van den Broek, A. J. P., 6, 8, 10, 71 Van de Putte, K. A., 264, 265,304 van der Mass, C. J. J., 154, 173 Vander Pyl, M., 382, 389, 390, 391, 394 Vandini, R. Z., 125, 172 van Harreveld, A., 414, 416, 447 van Iersel, J. J. A., 76, 91, 93, 116 van Mullem, P. J., 12, 54, 58, 64, 76, 78, 104, 112, 116 van Oordt, G. J., 119, 120, 154, 172, 173, 324, 335, 345 van Oordt, P. G. W. J., 56, 66

W Wada, I., 435, 445 Waddington, C. H., 187, 252 Wai, E. H., 21, 42, 65, 71, 78, 87, Se, 113, 116

353

Weisman, A., 156, 167 Wellborn, T. L., 282, 305 Welsh, J. H., 327, 351 Wessing, A., 257, 298, 301

466 White, P. R., 254, 305 Whitley, G. P., 412, 438, 449 Wickett, W. P.,232, 233,241, 252 Wiebe, J. P., 11, 13, 14, 15, 35, 42, 43, 47, 48, 52, 59, 65, 71, 85, 96, 113, 116 Wiener, S., 421,449 Wiese, R. V., 152, 172 Wight, H. G., 258, 286, 289, 270, 286, 287, 289, 298, 298, 299, 304 Wikswo, M. A., 321, 353 Wilbanks, G. D., 286,207, 279,298,303 Wilhelmi, A. E., 99, 105, 116 Williams, J., 199, 202, 252 Williamson, H. C., 120, 173 Willmer, E. N., 254,305 Wills, J. H., 433,434,446 Wilmot, I. R., 203,u)6,207, 232, 245 Wilson, C. M., 297, 304, 339, 352 Wilson, D. P., 389, 400 Winberg, G. G.,208, 252 Wing, A. S., 387, 393, 394 Winge, 0,129, 132, 134, 135, 173 Winpfield, W. H., 294, 305 Withler, F. C., 186, 252 Witschi, E., 41, 49, 72, 127, 143, 149, 150, 151, 173, 174 Wolf, K., 257, 258, 259, 261, 203, 284,

265, 286, 287, 268, 270, 274, 275, 276, 277, 278, 279, 281, 282, 287, 289, 291, 296, 297, 299, 304, 305 Wolf, L. E., 129, 138, 174 Woodhead, A. D., 223, 252, 334, 340, 353 Woodhead, P. M. J., 57, 72, 223, 252 Woodward, R. B., 431, 449 Wootton, R. J., 77, 78, 106, 116

AUTHOR INDEX

Wotiz, H. H., 51, 54, 55, 61, 72, 153, 174 Wourms, J. P.,2, 72 Wright, R. S., 52, 54, 55, 64, SS, 164 Wu, A. S. H.,187, 246 Wiinder, W., 154, 174 Wurmbach, H.,139, 174

Y Yamagishi, S., 434, 449 Yamamoto, K., 49, 72 Yamamoto, S., 435, 445 Yamamoto, T., 81, 87, 116, 129, 132,

135, 144, 145, 140, 147, 148, 149, 150, 151, 152, 154, 156, 174, 175, 184, 188, 252, 314,348 Yamane, H., 332, 348 Yamashita, H.,81, 87, 114, 155, 170 Yamazaki, F., 15, 45, 40, 48, 49, 50, 63, 72, 98, 103, 116 Yaron, Z., 13, 52, 72 Yasaki, Y., 384, 400 Yasuoka, N., 431, 448 Yoshida, T., 428, 429, 440, 448 Young, J. Z., 261, 305, 323, 353 Yusha, A., 258, 268, 271, 278, 281, 291, 290, 301

Z Zahl, A. P., 154, 175 Zander, C. D., 137, 138, 162, 175 Zei, M., 124. 175 Ziegler, I., 315, 353 Ziegler-Gunder, I., 317, 353 Zotin, A. I., 185, 199, 252 Zuckerman, S.,4,5, 64 Zupanovic, S., 126, 171

SYSTEMATIC INDEX Note: Names listed are those used by the authors of the various chapters. No attempt has been made to provide the current nomenclature where taxonomic changes have occurred.

A Ab7amis brama, 212, 218, 228 Acanthanchus pomotis, 298 Acanthias, 179, 183 Acanthopagrus A. latus, 121 A. schlegeli, 121 Acanthurus tdostegus, 181, 192 Acheilognathus, 184 A. indermedia, 156 Acipenser, 187, 189, 191, 234 A. stellutus, 182, 212 A. sturio, 181 Acropomu, 357, 367, 383 Acropomatidae, 360 Actinopterygii, 355 Aequidens, 101 A. latifrons, 83, 101 A. portakgrensis, 221 Agnatha, 41, 42, 127 Albulu vulpea, 212

labiosa, Colisa lalia, Macropodus opercukzrk Anabantidae, 132, 138 Anablepidae, 31, 33, 37, 38 Anableps, 38, 179 A. anableps, 33, 38, 39 A. doweri, 38 Anarhich, 179 Anchovy, see Engravlis Anglefish, see Pterophyllum scalure Angler fish, see Lophius Anguilh anguillu, 128, 191, 212 Anguillidae, 8 Anisotremus virginicus, 299 Anomalopidae, 383, 385 Anomalops, 365, 366, 383, 390, 391 A. katoptron, 383, 385 Anoptichthys, 335, 343 A. iorduni, 324 Anura, 13 Aphyosemion, 179 Alburnus albumus, 205, 208, 209, 218, Apogon, 183, 357, 366, 375, 377, 384, 386, 394 219, 228 Alga, blue-green, see Schizothrix calcicola A* 183 Alpha, 122 A. eltioti, 376 A. imberbis, 34, 183 Amberjack, see Seriola dumerili Apogonidae, 179, 375 Ambloplites rupestds, 298 Argyropelecus, 370, 380 Amblyopsidae, 21 A. afinis, 390 Ameium, 324, 328, 327, 328 A. olfersi, 381, 385 Amia, 179, 187 Ariidae, 15, 19, 179 A. calva, 154 Artemia, 239 Ammodytes, 179, 216, 22.4 Aspredo cotylephorus, 21 Amphibia, 5 Astronesthes, 357, 363, 382 Anabantid, see Betta spkndens, Colisa A. r-khardsoni, 364 407

468

SYSTEMATIC INDEX

Carassius, 187, 189, 312, 330332, 334 C. auratus, 2, 42, 45, 132, 149, 154, 227, 256, 297 C. auratus auratus, 130 C . auratus gibelio, 130 C . carassius, 297, 331, 336 B Carcharhinidae, 23 Bagridae, 21 Carcharhinus, 30 Balanoglossus biminiensis, 379 C. falcifomis, 28 C. mnisorrah, 425, 429 Baknus, 239 Carcharias gluucus, 25 Barracuda, see Sphyraenu Carcharinus milberti, 279, 296 Bass Atlantic sea, see Centropristes striatus Cardinal fish, see Apogonidae smallmouth, see Micropterus dolomieui Careproctus, 179 striped, see Roccus saxatilis Carp, see Cyprinus carpio, Puntius Batfish, see Dibranchus atlanticus javanicus Bathygobius, 331, 334, 336 Caspiabsa volgensis, 219 B . soporator, 81, 85, 89, 96, 108, 109 Catfish, see Anurhichas, Heteropneustes Betta splendens, 83, 84, 88, 132, 142, fossilis, Parasilurus, Parasilurus asotus 179 Bichir, see Polypterus marine, see Ariidae Bitterling, see Rhodeus amarus Cavefish, mexican, see Anoptichthys Japanese, see Acheilognathus inderjordani media, Rhodeus ocellatus Centracanthidae, 120, 122 Bleak, see Albumus alburnus Centrarcidae, 179 Blenniidae, 179 Centrarchus macropterus, 298 Blennius, 179 Centronotus gunnellus, 226 B. pholis, 180, 226, 240 Centropristes striatus, 122, 126 Blennoidae, 15 Cephalopholis, 122 Blenny, see Blennius, Ictalurus, Zoarces Ceratias, 358,360, 389 viviparus Ceratiidae, 389 Bliccu biorkna, 228 Ceratoscopelus maderenis, 385, 393 Bone fish, see Albula vulpes Cetorhinus maximus, 9, 14, 16 Boops boops, 120 Chuenobryttus gubsus, 298 Bow fin, see Amia Chaetodipterus faber, 335 Boxfish, see Ostracion lentiginosus Chenna argus, 237 Brachydanio, 343 Chunos, 238, 239 B. rerio, 194, 290, 297 C. chanos, 212, 238 Bream, see Abramis brama Char, see Salvelinus willughbii Mediterranean, see Sparus auratus Characin, see Astyanar white, see Blicca bjorkna Chasmichthys, 310, 312, 326, 328332, Breuoorita tyrannus, 231, 235 335, 336, 338 Bullhead, see Cottus C. gulosus, 309, 318 Bullrout, see Notesthes robusta Chaulidus, 358 Bunocephalidae, 19 Chauliodontidae, 385 Butter fish, see Blennius pholis Chauliodus, 359, 361, 370, 382, 388, 389 C. sloanii, 385 C Chelidoperca hirundinacea, 122 Calamus, 299 Chimaeridae, 22 Canthigaster rivulatus, 431 Chondrichthyes, 5, 22, 23, 34 Capelin, see Mallotus villosus Chrysophrys major, 311 Astronotus ocellatus, 101 Astyanux, 334 Ataenobius towed, 33 Aulophallus elonqatus, 34 Austroatherina, 2Q3

469

SYSTEMATIC INDEX

Cichlasoma severum, 101 Cichlidae, 21, 134, 179 Cleidopus, 366 C. gloriamaris, 384, 385 Clinocottus analis austral&, 421 Clupea C. harengus, 128, 179, 181, 182, 192, 193, 196, 197, 205-207, 204-212, 214, 216, 217, 220, 222, 227-231, 2.37, 239, 296 C . palhii, 224 C. sprattus, 312 C. venenosa, 43 Clupeiformes, 296 Clupeonella delicatula, 181 Cod, see Gadus morhuu Coelorhynchus, 384 C . hubbsi, 385 COl&U C. labiosa, 83 C. klia, 83, 88, 91 Collichthys lucidus, 392 Cololabis saira, 436 Coregonus, 193, 215, 217, 221, 222 C . clupeaformis, 193, 212, 225, 231 C . lavaretus, 234 C . wartmanni, 212, 217, 228, 229 coris C . giofredi, 124, 125 C. $is, 124, 125 Cottidae, 34, 132, 179, 299 Cottus, 179 C. pollux, 132 Covesius, 11 C. plumbeus, 45, 49 Crenilabrus ocellatus, 100 Cristiuomer, 183 C. namaycush, 225 Cryptosaras, 389 Ctenopharyngodon, 239 C. idella, 238 Cyanophytae, 426 Cyclopterus, 180 C. lumpw, 232 Cyclops, 217 Cyclothone, 357, 370, 380 C . braueri, 390 Cymatogaster, 11, 13, 17, 36, 42, 46, 48 C. aggregata (aggregatus) 2, 32, 35, 42, 45, 85, 90 Cynolebias, 179

Cypridina, 376-379, 384, 394 C. hilgendorfii, 376 c. luciferase, 386 Cyprinid, see Rhodeus amarus Cyprinidae, 19, 132 Cypriniformes, 297 Cyprinodon, 212 C . macularius, 193, 211, 212, 235 Cyprinodontes, 33 Cyprinodontidae, 132 Cyprinodontiformes, 30, 297 Cyprinw carpio, 130, 181, 182, 218, 219, 228, 238, 270, 297

D Dactylostomias, 388 Dalatiidae, 385 Dasyatidae, 405 Dasyatis, 28 D. violacea, 27 Deal fish, see Trachypterus Dentex D. denter, 121, 126 D. (Taius)tumifrons, 122 Diaphus, 357 D. rafnesquii, 357 D. theta, 391 Dibranchus atlanticus, 379 Diplectrum formosum, 298 Diplodus annularis, 233 Dipnoi, 300 DogEsh, see Acanthias, Mustelus, Scylliorhinus, Squalus smooth, see Mustelus canis spiny, see Squalus acanthias spotted, see scyliorhinus caniculus spur, see Squalus acanthias Dolopichthys, 385 Dunaliella, 239

E Eagleray, see Myliobatis Echiostoma, 381 E. barbatum, 380, 381, 385 Eel, see Anguilla, Gymnothorax javanicus sand, see Ammodytes Elasmobranchii, 296, 355 Elassoma zonutum, 298 Embiotocidae, 32, 34, 35 Enchelyopus cimbrius, 193, 194

470

SYSTEMATIC INDEX

Engraulis, 187, 239 E. anchita, 215, 234 E . japonicus, 209 Enneacanthus gloriosus, 298 Entosphenus wilderi, 127 Epinephelus, 122 Eptatretus, 314 E. burgeri, 128 Erilepis zonifer, 438 Esox, 179, 217, 297 E . lucius, 13, 193, 205, 207, 210, 217, 222, 232 Etheostoma, 183 Etmopterus, 357 E. frontimuculatus, 380 E . niger, 380 E . spinax, 385 Exocoetus, 179, 225

F Fighting fish, Siamese, see Betta splendens Flounder, see Platichthys, Pleuronectes, Pseudopleuronectes Flying fish, see Exocoetus Fundulw, 11, 12, 96, 185, 187-189, 201, 233, 255, 256, 310, 312, 314, 318-321, 326, 327, 329-334, 336, 338-341, 344 F . heteroclitus, 45, 96, 99, 153, 197, 204, 207, 237, 297, 310 F. nuzialis, 297 F . parvipinnis, 315

G Gadidae, 128, 360, 366 Gadiformes, 298 Gadus, 225, 327, 335 G . macrocephalus, 193, 211, 212, 235 G. morhuu, 181, 220, 230, 298, 324 Galaxiidae, 8 Galeiformes, 23 Galeus canis, 28 Gambusia, 31, 47, 155, 256, 270, 274 G. afinis, 99, 143, 155 G. holbrookii, 143, 155, 297 Gar, see Lepidosteus Garpike, see Leipososteus Gasterosteidae, 21 Casterosteus, 12, 13, 42, 58, 179, 186, 225

G. aculeatus, 42, 54, 76, 154, 203, 236, 237 G. pungitius, 154 Gazza, 359, 367 G. minuta, 385 Gempylidae, 438 Glarichthyes, 140 Gobeiidae (Gobiidae), 15, 129, 132, 179 Gobius paganellus, 100 Goby, see Bathygobius, Chmmichthys Goldfish, see Carassius Conichthys cocwi, 390 Gonostoma, 370 G. elongatum, 365 G . gracile, 121 Gonostomatidae, 121, 369 Goodea bilineata, 33 Goodeidae, 33, 35-37 Gourami, blue, see Trichogaster trichopterus Grunion, see Leuresthes tenuis Grunt, blue-striped, see Haemulon flavolinentum Guppy, see Lebistes reticuhtus, Poecilia reticulata Gymnurchus niloticus, 204 Gymnothorax javanicus, 429 Gymnuridae, 405

H Haddock, see Melanogrammus aeglejinus Haemulon H. flavolineatum, 258, 290, 292, 299 H . s c i u w , 290, 292, 299 Hagfish, see Eptatretus, Myxine Hake, see Merluccius metluccius Halichoeres poecilopterus, 124, 311 Hardtail, see Trichiurus muumela Harengula zunasi, 311 Hemichromis, 107 H . bimaculatus, 82, 88, 100 Hemirhampus, 179 Hepatus hepatus, 119 Herring, see Clupea volga, see Cnspialosa volgensis Heterandria, 37, 179 H . formosa, 33, 34, 37, 40 Heterodontidae, 22 Heteropneustes, 49, 50 n. fossilis, 45, 49, go, 156

471

SYSTEMATIC INDEX

Hippocampus, 179, 225, 299 H. hippocampus, 48 Holocephali, 28, 300 Hound, see Mustelus Hydrolugus colliei, 20, 54, 425 Hyodontidae, 8 Hypomesus olidus, 218 Hypoplectrus, 119 I Ichthyophom hoferi, 130, 139, 143 Ictabtus, 179 1. nebulosus, 290, 292, 297 1. punctatw, 297 Idiacanthidae, 383, 391 ldiacanthus, 357, 383, 391 Inegocia 1. crocodib, 121 I. meerdemoort, 121 Iniomi, 120, 357, 392, 393 Islstius brasiliensis, 380, 385 Isospondyli, 357, 391, 392

J Ienynsia, 31 Jenynsiidae, 33, 38 lulls 1. poecilopterus, 124 I. pywhogrammu, 124

K Katsuwonus pelamis, 224 Kilka, see Clupeonella delicatukz Killifish, see Fvndulus Kryptophanaron alfredi, 383 Kurtus gulliveri, 21

L Labridae, 120 Labrus, 335, 337 L. merulu, 1.24 L. ossifagus, 1% L. turdus, 124 Laemurgus borealis, 5 Lagocephalus inermis, 431 Lamna cornubica, 28 Lampanyctus leucopsarus, 357, 391 Lampetra, 184-188, 323, 342 L. fluuiatilis, 41, 42, 128 L. lamottei, 127 L. planeri, 128, 323

Lamprey, see Petromyzon brook, see Lampetra lamotter Lamprotoxus, 360 Lantern fish, see Ceratoscopelus maderenis, Gonichthys coccoi, Myctophum punctatum Latimeria, 13 Lebistes, 13, 180, 213, 220, 225, 313, 319, 329, 335 L. reticulatus, 38, 181, 237, 298, 309, 310 Leiognathidae, 359, 385 Leiognathus, 366 L. equulus, 359 Leiostomus xanthuw, 283, 299 Lepidocybium pavobrunneum, 438 Lepidosiren, 5, 179, 203, 204 Lepidosteus, 8, 187, 314 Lepomis, 107 L. auritus, 298 L. cyanellus, 84, 299 L. gibbosus, 84, 299 L. mucrochirus, 290, 292, 299 L. megalotis, 84 L. microlophus, 299 Leptostomiaa, 382, 385 Lermicthys multiradiatus, 33 Leuchichthys artedi, 214 Leuresthes tenuis, 179, 233, 236 Ling, see Moba Molva Linophryne, 388 L. brevibarbis, 358 Loach, see Misgurnus fossilis Lophiiformes, 299 Lophius, 180, 225, 389 L. americanus, 299 L. piseatortus, 299 Lumpsucker, see Careproctus, Cyclopterus Lungfish, see Protopterw Lutianus, 298 L. bohar, 425, 427, 430 L. griseus, 290, 292 Lyngbya majuscub, 426

M Mackerel, see Scomber scombws horse, see Trachuw Macropodus, 138, 326, 337 M. chinensis, 128, 138 M . concolor, 128, 138

472 M . opercularis, 84, 101, 128, 138, 237, 299 Macrouridae, 366, 375, 385 Maena, 122 Maenidae, 122 Malacocephalus, 359, 375 M . laevis, 384, 385 Mallotus villosus, 179 Maurolicidae, 369 Maurolicus, 370, 373, 375 Medaka, see Oryzias Melunocetus, 359 Melanogrammus aeglifinus, 57, 181, 182, 210, 224, 230 Melanostomiatidae, 385 Menidia, 227 Merlangius merlangus, 220 Merluccius merluccius, 224 Micromesistlus poutassou, 191 Micropterus dolomieui, 228, 299 Microstomus kitt, 215, 2-30,239 Midshipman, see Porichthys porosissimus Milkfish, see Chanos chanos Minnow, see Couesius, Phorinus desert, see Cyprinodon macularius fathead, see Pimephales promelas Misgurnw, 8 M . fossilis, 204, 213, 234 Mobulidae, 405 Mogrunda obsculu, 132 Mollienesia latipinna, 45, 47 Molva molva, 181 Monocentridae, 360, 366, 385 Monocentris, 358, 366, 384 M . japonica (japonicus), 358, 385 Monopterus albus, 125 M . iaoanensis, 125 Mordacia mordux, 296 Morone labrax, 52 Mudskipper, see Periophthalmus vulgaris Mugil, 228, 239 M . cephalus, 52 Mullet, see Mugil, Mullus, Parupeneus Mullus, 233 M . barbatus, 199, 228 Mustelus, 326, 334 M . canis, 23, 28-30, 47, 198 M . laevis, 25, 198 M . mustelus, 181 M . vulgaris, 28, 30, 179, 198 Mutatis mutandis, 135

SYSTEMATIC INDEX

Mycteroperca, 122 Myctophidae, 385 Myctophiformes, 120 Myctophum, 358, 380 M . afine, 391 M . punctatum, 384, 386, 387 M . watasai, 385 Mylio mucrocephalus, 121 Myliobatidae, 405 Myliobatis, 28 M . aguilu, 406 Myliobatoidea, 405 Myoxocephalus scorpius, 299 Mytilus, 239 Myxine, 9, 15, 179, 327 M . glutinosa, 128, 310, 323 Myxini, 300 Myxinidae, 42

N Neoscopelm, 358, 388 Newt, see Taricha torosa Noteathes robusta, 410 Notopteridae, 8, 19

0 Oblada melanula, 120 Oncorhynchus, 13, 93, 182, 183, 186, 220, 223, 227, 230, 343 0. gorbuscha, 180, 217, 223 0. keta, 217, 223, 231-233 0. kisutch, 199, 217, 223, 232, 290, 290, 317 0. nerka, 53, 217, 223, 230, 231, 233, 236, 296 0. tshawytscha, 199, 230, 231, 232, 237, 290, 296 Oneirodidae, 385 Ophicephalus punctatus, 45 Opisthoproctidae, 369 Opisthoproctus, 369, 373 0. soleatus, 368 Orthonopias triacis, 34 Oryzias, 99, 103, 108, 154, 185-187, 189, 201, 202, 211, 221, 232, 236, 314316, 320, 322, 330, 331, 333, 335, 337, 338, 340-344 0. latipes, 59, 75, 81, 87, 99, 129, 132, 135, 143-147, 155, 181, 185, 192, 204, 221, 237, 319, 326

SYSTEMATIC INDEX

473

Platyfish, see Xiphophorus maculatw Phtystaczcs, 179 Plecoglossus, 218 P. altivelis, 215, 218 Plectonemu terebrans, 426 P Pleurowctes, 54, 225, 324 Pagellus P. flesus, 22s P. acarne, 121 P. phtessa, 45, 181, 182, 192, 193, P. centrodontus, 120 196, 197, 210, 211, 216, 217, 220, P. erythrinus, 122, 128 222, 227-229, 234, 237, 239 P. mormyrus, 120 Pleuronectiformes, 299 Poecilia, 11, 37, 38, 43, 56, 137 Paradise fish, see Macropodus Paragaleus, 30 P. caudofascfata, 137, 138 P . f o m s a , 2, 129-131 Paralabrax cluthmtus, 122 Parapriacanthus, 366, 368, 375, 376, 384, P. latidens, 131 P. latipinnu, 130 394 P. beycifomes, 367, 376 P. ~ucida,131 Parasilurus, 335 P. nigrofasciata, 132, 137, 138 P. reticulata, 11, 42, 45, 86, 91, 98, P. asotus, 324 129, 132, 135, 143 Paratrachichthys, 386 P. sphenops, 130 P. prosthemius, 385 P. uetulus, 290 Parupeneus chryserydros, 430 P. vittata, 130, 131, 137 Pempheridae, 375 P. uiuipara, 92 Perca, 54 P. flauescens, 180 Poeciliidae, 31, 33, 35, 37, 132, 137 P. fluuiatilis, 219, 227 Poeciliopsis, 38, 130 Pollachius uirens, 298 Perch, see Cymutogaster, Perca Perciformes, 30, 32, 120, 298 Polypinus, 380 P. stereope, 385 Periophthalmus vulgaris, 129 Polypterus, 5, 204, 314 Petrometopon, 122 Pomacanthus, 299 Petromyzon, 179, 300 Pornxis nigromaculatos, 299 P. fluuiatilis, 256, 296 Porgy, Japanese, see Chysophrys muior P. murinus, 54 Porichthys, 371, 373, 374, 378-381, 384, Petromyzones, 296 390, 391, 393, 394 Petromyzoniformes, 296, 300 P. notatus, 371, 372, 374, 380 Phallostethidae, 34 P. porosissimus, 378 Photoblepharon, 365, 366, 383 Potamotrygon P. palperbratus, 383 P. magdalenae, 407 Photostomias, 382 Phorinus, 324428, 335-339, 341 P. motoro, 407 P. pauckei, 407 P. laeuis, 128, 154 Potamotrygonidae, 405 Physiculus, 366 Prinodus, 1u) Pike, see Esor lucius Protopterus, 92, 179, 204, 314 Pilchard, see Sardina pilchardus P. annectes, 54 Pimephales promelas, 258, 290, 292, 297 Pine cone fish, see Cleidopus gloriamaris Pseudopleuronectes americanus, 179, 231, 299 Pipe fish, see Syngathus Pterogobius zonoleucus, 155 Pisces, 117, 128, 157 Pterois, 412, 413, 421, 422 Plaice, see Pleuronectes platessa P. volitans, 411, 420, 421 PlutichthrJs flesus, 183 Pterophyllum, 101 Platycephalidae, 121

Osmerus, 179, 182 0. eperlanus, 181, 182, 192, 193, 212 Osteoglossidae, 8, 19 Ostracion lentiginosus, 438, 439

474

SYSTEMATIC INDEX

P . scalure, 99, 101 Pteroplutea, 25, 28 P. micrura, 25 Ptychocheilus oregonensis, 297 Puntazzo puntazzo, 120 Puntius, 239 P. jauanicus, 238

R Rabdosargus sarba, 121 Raiidae, 22 Raia (Raia), 23, 290 R. binoculuta, 202 R. radiata, 54, 296 R. rhina, 20 R. ocehta, 54 Rajiformes, 290 Rat fish, see Hydrolugus colliei Ray bat, see Myliobatidae butterfly, see Gymnuridae cow-nosed, see Rhinopteridae devil, see Mobulidae eagle, see Myliobatidae electric, see Torpedo manta, see Mobulidue river, see Potamotrygonidae sting, see Dasyatidae whip, see Dasyatidae Redfish, see Sebmtes

Rhinobatus granukztus, 20 Rhinopteridae, 405 Rhodeus, 184, 187, 331, 335, 337439 R . amurus, 19,34, 89, 156, 179 R. ocellatus, 89,99 Rhyncholagus, 369 Rivulus marmoratus, 59, 120 Roach, see Rutilus nrtllus Roccus R. americanus, 298 R. saxatilk, 192, 193 Rock fish, see Sebastodes Rutilus rutilus, 218, 228 Ruoettus pretiosus, 438 Rypticus, 122 R. saponaceus, 439,440

S Saccopharynx, 358 Sacura S. murgaritacea, 122

S . pulcher, 122 Salamandriadae, 430

Salmo, 183, 187, 189, 223, 239, 337, 341 S. fario, 34 S. gairdneri, 53, 128, 142, 180, 193, 197, 194-203,209, 215, 223, 225,

232, 237, 258, 290, 292, 290, 334 S. irideus, see Salmo gairdneti S. salar, 53, 85, 181-183, 192, 195197, 199, 204-207, 216, 223, 231, 232, 234,297, 317, 343 S. trutta, 142, 187, 193, 196, 197, 200, 202-204, 207, 212, 220, 223, 225, 231, 232, 230, 237, 297, 334, 340 Salmon, see Oncorhynchus, Salmo Atlantic, see S a h sahr chum, see Oncorhynchus keta

coho, see Oncorhynchus kisutch pink, see Oncorhynchus gorbuscha sockeye, see Oncorhynchus nerka Salmonidae, 5, 8, 21

Saluelinus, 183 S . fontinulis, 182, 193, 203, 234, 238, 297 S. namaycush, 232, 238 S. willughbii, 13 Sardina pilchardus, 212, 215, 220, 224, 229, 230, 233, 239 Sardine, California, see Sardtnops caerulea Sardinops caerulea, 182, 193, 197, 200, 205, 206, 207, 210, 211, 215, 220, 230, 239 Sargue, see Diplodus annularis Sargus S. annulatis, 120 S. sargus, 120 S. oulgaris, 120 Saury, see Scomberesox Japanese, see Cololabis saira Schizothrix calcicolu, 430 Scoliodon, 30 Scomber, painted, see Sewanw scriba Scomber scombnrs, 181, 192, 193, 210, 220, 224, 230 Scomberesox, 179, 214, 225 Scopthalmus, 335, 336 Scorpaena, 413, 421, 422 S. guttata, 412, 421 Scorpaenidae, 32, 35

SYSTEMATIC INDEX

Sculpin, California, see Scorpaenu guttata Scyllidae, 22 Scylliorhinus, 47, 323 S. canicuh, 9, 213 S. caniculus, 3, 43, 54, 58, 153, 181, 192 S . stelhris, 52, 153 Sea horse, see Hippocampus Sea perch, see Cymatogaster aggregata Seariidae, 389 Searsia, 360, 389 S. schnakenbecki, 385 Searsiidae, 385 Sebastes, 179 S. marinus, 34, 182 S. viviparus, 181 Sebastodes, 11, 32, 35 S. paucispinis, 35 Secutor insidiator, 385 Seriola dumerili, 430 Serranidae, 119, 120, 122, 126 Seranus, 122 S. cabrilh, 119 S. scriba, 52, 119, 315 S. subligerius, 119 Shark, see Mustelus, Spinax basking, see Cetorhinus maximus bonnethead, see Sphyrna tiburo gray, see Carcharhinus menisorrah Greenland, see Laemargus borealis sandbar, see Carcharinus milberti silky, see Carcharhinus falciformis Sheat fish, see Platystacus, Silvrus glanis Silun'dae, 32 Silverside, see Menidia Siurus ghnis, 197, 218 Skipjack, see Katsuwonus pelamis Smelt, see Osmerus night, see Spirinchus starksi Snapper, red, see Lutjanus bohar Soapfish, Atlantic, see Rypticus sapoMCeUS

Sole, see SoZea solea lemon, see Microstomus kitt Solea, 13, 239 S. solea, 181, 214, 215, 220, 224, 228, 239 Solenostomus laciniatus, 21 Somniosus microcephalus, 425 Sparidae, 120, 122, 128

475 Sparus S. aries, 121 S. auratus, 120, 121 S. bngispinis, 121 S. sarba, 121 Sphaeroides S . alboplumbeus, 431 S. basilewskianus, 431 S. chysops, 431 S . niphobles, 431 S. oscelhtus, 431 S . pardalis, 431 S. porphyreus, 431 S . pseudommrts, 431 S. rubripes, 431 S. stictonotus, 431 S. vermicularis, 431 S. xanthopterus, 431 Sphraena barracuda, 429 Sphyraena, 323, 324 Sphyrna, 30 S. tiburo, 23, 28 Sphyrnidae, 23 Spicara, 122 S. aheak, 124 S. chryselis, 122, 124 S. maena, 122, 124 S . smaris, 122, 124 Spinax niger, 361 Spirinchus starksi, 215 Spondyliosoma (Cantharus) cantharus, 122 Spot, see Leiostomus xanthurus Sprat, see Clupea sprattus and Sprattus sprattus Sprattus sprattus, 215, 225 Squalidae, 25, 385 Squaliformes, 296 Squabidea, 380 Squalus, 323, 327, 336 S. acanthias, 23, 25, 44, 52, 58, 181, 192, 279, 296, 322 S. suckleyi, 26, 54, 202 Steindachneria, 369, 381 S. argentea, 385 Sternoptychidea, 369, 385 Stickleback, see Gasterosteus Stingray, see Dasyatis, Pteroptatea, Tvgon round, see Urolophidae Stomias, 358, 360, 375, 382

476 S . ferox, 385 S . valdivae, 362 Stomiatidae, 369, 385 Stonefish, see Synanceia horrida Sturgeon, see Acipenser convict, see Acanthurus triostegus Sunfish, green, see Lepomis cyanellus Swordtail, see Xiphophorus helleri Symphysodon, 101 S. aequifasciata, 100 Synanceja, 412, 413, 421, 422 S. horridu, 412, 413, 420 S. trachynis, 421 Syngathus, 179 Syngnathidae, 21 Syngnathiformes, 299

SYSTEMATIC INDEX

Tridacna maxima, 429 Tridentiger trigonocephalus, 221 Trout brown, see Salnm trutta lake, see Salvelinus namaycush rainbow, see Salmo gairdneri Trunkfish, see Ostracion lentiginosus Trygon, 28, 198 T . uiolucea, 25, 27 U Uranoseopidae, 409 Uranoscopus, 389 U. scaber, 409, 410 Urobatis, 323, 324 Urodela, 13 Urolophidae, 405 Urobphus hallefi, 405, 406, 415

T Tachicorystes, 15 V Tactostoma, 381 Vibrio comma, 279 Tachysurus, 179 Vinciguerria, 370 Taricha torosa, 403, 430, 431 Tarletonbeania crenularis, 357, 391 W Tautoga, 318 Weever, greater, see Trachinus draco Teleostei, 5, 8, 30, 355 Whitefish, see Coregonus Teleostomi, 296 Whiting, blue, see Micromesistius Tench, see Tinca tinca poutassou Thalassoma bifasciatum, 124 Wrasse, see Crenilabrus ocellatzrs Tigriopus, 239 blue, see Haliochoeres poecilopterus Tilapia, 13, 158, 179, 183, 238 cuckoo, see Labrus T . macrocephala, 82, 88, 103, 183 T. mossambica, 83, 134, 149, 183 X T. tholloni, 183 Xiphophorus, 179, 341, 343 Tinca, 13 X . caudofasciata, 137 T . vulgaris, 297 X . couchianus, 132 T. tinca, 219, 257, 270 X . helkri, 33, 80, 86, 136, 137, 139Toothcarp, see Poecilia formosa 141, 143, 298, 310 Topminnow, see Gambusia holbrookii X. maculatus, 80, 81, 129, 132-134, Torpedo, 23, 2.8, 55, 198 136-139, 155, 298 T. marmorata, 54, 153 X. milhi, 132 Trachichthyidae, 366, 385 X . montezumae cortezi, 132 Trachinidae, 407 X . variatus, 80, 132, 133, 136, 298 Trachinus X. xiphidium, 132, 136, 137 T. draco, 408, 416, 417, 418 T . vipera, 416, 418 Z Trachurus, 233, 312 Zebra fish, see Brachydanio rerio T. trachurus, 228 Zoarces, 31 Trachychoristes, 32 Z. viuiparms, 32, 181 Trachypterus, 214, 225 Zoarcidae, 32 Tribobdon hakonensis, 221 Zoogonecticus cuitzooensis ( quitzeonsls) , Trichiurus maumela, 317 33, 36 Trichogaster trichopterus, 83, 84, 88, 109

SUBJECT INDEX A A cells, see Photocytes Acetylcholine, 327, 330 Acetylcholinesterase, 328

in testes, 5 2 5 4 Androstenedione, 52, 78 Androsterone, 53 Androtemone, 146

ACTH, see Adrenocorticotropic hormone Activity

Antibiotics amphotericin B, 272 chlortetracycline, 272 kanamycin, 272 nystatin, 272 penicillin, 272 streptomycin, 272 in tissue culture, 256, 272, 273 Arginase, 202 Atropine, 337 Autosomes, 135

diurnal light changes and, 223 feeding drives and, 229 larval distribution and, 221-229 searching ability, 229 starvation and, 229 Adrenaline, 379, see also Epinephrine Adrenocorticotropic hormone, 326 Aggressive behavior androgens on, 80, 81 estrone on, 88 follicle-stimulating hormone on, 101 gonadectomy and, 76, 77, 81, 84, 87 gonadotropic hormone and, 77 luteinizing hormone on, 101 photoperiod and, 77 temperature and, 9 5 Agonistic behavior, see Aggressive behavior

B see Photocytes BAF, see Bovine amnionic fluid Balanced salt solutions, 26&265, see also Physiological salines Courtland's, 262 for cyclostomes, 265 Earle's, 262 for freshwater teleosts, 263 Albinism, 340 Alevin, 178 Gey's, 262 Hank's, 262 Alveolar organs, 364 for marine elasmobranchs, 264 Androgenin, 4, 146 for marine teleosts, 263, 264 Androgens, 5, 52-54 Behavior biosynthesis of, 53, 152 androgens and, 76-86 in blood, 5 2 5 4 courtship, 81, 82, 92, 103 dosage levels in male induction, 146 estrogens and, 86-90 effect on fins, 155 gonadectomy and, 76-90 reproductive cycle and, 54 hormones and, 73-110 secondary sex characters and, 78-79 mating, 83, 130, 133, 145-147 in sex reversals, 143, 144, 146 migratory, 79, 91-94 synergistic action with pituitary, 95 parental, 1 W 1 0 2 477

478

SUBJECT INDEX

prespawning, 94-97 spawning, 97-100 thyroid hormone and, 9&94 Bioluminescence, 355-394, see also Luminescence, Photophores Blastoderm, 188 BME, see Eagle’s basal medium Bovine amnionic fluid, 270 Bretylium, 336 Brooding, hormones on, 82, 88 BSS, see Balanced salt solutions Buoyancy, fish eggs and larvae, 226226

C Caffeine, 338 Calf serum, 288 Caloric intake, 209 Carbachol, 337 Carotenoids, 314, 315 Castration, 77, 81-84, 142, 154 Catecholamine, color changes and, 329, 330 Catecholamine-0-methyltransferase,328 Caudal glands, 382 Cell and tissue culture, 253-300 antibiotics in, 272 cell line derivations, 258 dispersion in, 282 fish cell lines, 289-294 fishes used in, 296299 freezing cells, 288, 289 history of, 254-259 media in, 265-273 methods, 273-285 present status of, 259, 260 seeding density, 283, 284 sources of cell lines, 294 in virology, 257 Chicken serum, 268 Chloramine, 274 Chlorpromazine, 338 Chorion, 185, 187, 195 composition of, 189 hardening of 185-186 permeability at fertilization, 199 Choroid gland, 344 Chromatophores, 307453, see also Melanophores chemicals on, 332338 classification and terminology, 308, 309

derivation of, 342 drugs on, 332438 excitement and, 335 hormonal control of, 322-328 impulse propagation, 329 measuring response, 317419 melanin aggregation in, 332-338 morphology, 309-313 nervous control of, 3 2 6 3 3 1 pigments, 313-317 pineal control, 325 two-hormone hypothesis, 325 Chromatosome, 308 Chromosomes, see Sex chromosomes Chrysopsins, 386 Ciguatera poisoning, 426 Ciguatoxin, 4 2 7 4 3 0 Clasper glands, 2 2 2 3 syphon, 2 2 2 3 Cleavage, 187 Club cells, 407 Clupeotoxin, 437 Cocaine, 113 Color changes, see also Chromatophores, Melanophores morphological, 3 3 8 3 4 1 physiological, 312, 3 1 7 3 3 8 primary responses, 342 secondary responses, 342 Condition factor, 218 Copulation after gonadectomy, 81 organ in teleosts, 34, 35 pelvic fin modifications in Chondrichthyes, 22, 23 Corpora lutea, 18, 19, 41, 55, 56 Corpus atreticum, 18 Corticalization of function, 106 Courtship behavior, 81, 82, 92, 103 Cryptotoxic, 402 Cystovarian, 8, 16

D Decamethonium, 435 Dehydroepiandrosterone, 78 Deoxycorticosterone, 156 Derived Ostwald Index, 318 Development, 177-252 dinitrophenol on. 203 in elasmobranchs, 2 3 3 0

479

SUBJECT INDEX

events during, 184-191 fat utilization in, 202, 203 of hatchery fish, 240 mortality during, 23G235 rates of, 191-194 temperatures and, 193, 196 in viviparous teleosts, 32, 33, 37 weight changes in, 24, 25, 31, 198 yolk utilization during, 25, 26, 194198 Dibenamine, 320, 332, 333, 335, see also Phenoxybenzamine Digestion, 218 Dimethyl sulfoxide, 287, 288 Dioecious, 1 Disinfectants, 273, 275, 276 Distribution of fishes, 223-224 DMSO, see Dimethyl sulfoxide DOCA, see Deoxycorticosterone Dopamine, 337 Dopa-oxidase, 314, 342 Drosopterines, 316

E Eagle’s medium, 260 EDTA, see Ethylenediaminetetraacetate Eggs chemical composition, 200 cleavage types, 187 early development of, 187-189 hatching of, 189, 190 incubation of, 187-189 organic content, 23, 24 radiation effects, 233 R Q values, 201 shape, 183, 187 sizes of, 18CL184 storage and fertilizability, 186 weight and female size, 183 Embryonic axis, 188 Embryonic development, see Development Envenomation, 412 Ephedrine, 337 Epiboly, 188 Epinephrine, 327, see ulso Adrenaline Erythrophore, 308, 310, 331, 337 Eserine, 337 Estradiol, 150 17p-Estradiol, 51, 54, 55 Estradiol benzoate, 87, 154, 155

Estradiol dipropionate, 80 Estriol, 55, 150 Estrogens seasonal changes in, 55 sex differentiation and, 5, 87, 145-147, 151 in tissues, 54-56 Estrone, 55, 88, 147, 150 Ethylenediaminetetraacetate, 282 Eyes of larvae, 220-221

F Fanning behavior, 102 Farming of fish, 238-240 Fat, 202, 203 Fecundity, 180-184 Feeding, 213-219 Fertilization, 184-187 internal, 22, 23, 34, 35 Fetal bovine serum, 268 Fibroblasts, culture of, 284 Follicle-stimulating hormone, 48-50 Freezing cells and tissues, 287-289 Fry, 178 FSH, see Follicle-stimulating hormone

G Gamones, 184 Gastrulation, 188, 203 Genital papilla, 34, 35 Germ cells, 4 Gestation, 20-40 in elasmobranchs, 23-30 follicular, 37-40 hypophysectomy on, 47, 48 ovarian, 3 5 3 7 in teleosts, 30-40 Glucose, in ova, 201 P-Glucuronidase, 53 Glycerol, 287 Glycogen storage, 201 Gonad, 3-19 embryology of, 3-5, 123, 126 female, 15-20 hermaphroditic, 118-127 histochemistry of, 13, 18, 153 hypophysectomy and, 41 male, 8-15 rnethallibure on, 46, 96 phy1ogeny, %8

480

SUBJECT INDEX

pituitary relations, 43-47 steroid hormones of, 50-56 Gonadectomy, 78, 79, 81, 83 Gonadotropins, 41-50 Gonochorism, 127-131 Gonochorists differentiated, 129 undifferentiated, 127-129 Gonoducts, 3-20 Gonopodium, 35 Granulosa, 18 Growth, 209-212 Guanethidine, 338 Guanine, 317 Guanophores, 317 Gymnovarian, defined, 7 Gynogenesis, 2, 130 Gynogenin, 4

H Hatching, 189-190 enzymes, 189, 234 radiation on, 233 temperature and, 193 HCG, see Human chorionic gonadotropin Hermaphroditism, 2, 118-127 consecutive, 120-127 protandrous, 118 protogynous, 118, 122-127 sex inversion, 121, 122 synchronous, 118-120 temperature on, 120 Heterogamety, 132 Hexamethonium, 338 Hexestrol dipropionate, 91 Hormones gonadal, 75-94 gonadotropic, 41-50 locomotor activity and, 109 reproductive behavior and, 73-116 Human chorionic gonadotropin, 50, 95 5-HT, see 5-Hydroxytryptamine Hybridization, 131 Hydergine, 338 Hydroquinone, 341 bHydroxytryptamine, 23, 336, 417, see ah0 Serotonin Hypophysectomy behavior and, 98 effect on gestation, 4748

gonadal effects, 4 1 4 7 parturition after, 48 Hypoxanthine, 317

I Ichthyocrinotoxic fishes, 402 Ichthyocrinotoxism, 438-440 Icihyohemotoxic fishes, 402 Icthyootoxic fishes, 402 Icthyosarcotoxic fishes, 402 Ichthyosarcotoxism, 4 2 4 438 Ichthyotoxism, 402, 423 Incubation, 178-180, 187-189 temperature and, 192 in tissue culture work, 289 Integumentary sheath, 408, 409 Intersex, 118 Interstitial cells, 1 2 1 4 , 46 Iridophore, 308, 311, 312, 342 K 11-Ketotestosterone, 53, 54

1 Lactalbumin hydrolysate, 270 LAH, see Lactalbumin hydrolysate Larva, 190, see also Development Lens, see Photophore Leucophores, 308, 320, 333 androgens on, 340 epinephrine on, 331 ergotamine on, 338 nervous control of, 331 potassium ions on, 331 Leydig cells, 13 LH, see Luteinizing hormone Light organs, 357475, see also Luminescence, Photophores Lobule boundary cells, 12-13 Locomotion cruising speeds, 228 darting speeds of larvae, 226 gonadal hormones on, 90, 91 schooling and, 228 thyroid hormones on, 91 LSD, see Lysergic acid diethylamide Luciferase, 375, 378 Luciferin, 375, 376, 378 Luminescence, 355394, see also Photophores

SUBJECX INDEX

adrenaline and, 380 bacterial, 356 biochemistry of, 375379 biological significance of, 388-393 colors of, 385 criticism of theories, 388-393 decay of, 377 direct control of, 380-382 extracellular, 356 horseradish peroxide on, 379 hydrogen peroxide on, 379 indirect control of, 382 intensities in lantern fishes, 387 intracellular, 356 occurrence of, 355 physical characteristics, 384-387 pituitrin and, 381 regulation of, 379384 spectrum emission curves, 384, 386 Luteinizing hormone, 48-50, 95 Lysergic acid diethylamide, 336

M Masculinization, pathological in guppy, 139, 140 Mating, see behavior MCH, see Melanophore-concentrating hormone Media for cell and tissue culture, 265273, see also Cell and tissue culture Melanins, 313-314, see also Melanogenesis, Melanophores Melaniridosome, 308 Melanization, 340 Melanoblast, 309 Melanocyte, 308, 339, 341 Melanocyte-stimulating hormone, 322 Melanogenesis, 313, 314, 339 ACTH on, 339 ionizing radiation on, 340 MSH on, 339 prolactin on, 340 Melanomas, 343 Melanophore-concentrating hormone, 325, 326 Melanophore Index, 317 Melanophores, 308, see also Color changes action of chemicals on, 332438 electric shock on, 341

481 electrophoresis theory, 3u) fine structure of, 310, 320, 321 hormonal control of, 322-326, 340,341 hydrostatic pressure on, 341 hypophysectomy on, 341 mechanisms of pigment movement, 319322 morphology, 309-311, 319 MSH on melanin dispersion, 322, 323 nervous control of, 326-331 pharmacology of, 332, 334-338 radiation on, 331 ultraviolet light on, 341 Melanosomes, see also Melanophores chemicals and drugs on, 332, 335-338, 342 definition, 308 dispersion, 323 mechanical pressure on, 341 morphology, 310 movements, 321 temperature on, 341 Melatonin, 325, 336 MEM, see Eagle’s minimal essential medium Meristic characters, 235-238 Merthiolate, 256 Metabolism during development, 191-209 growth and, 191-213 rates of respiration, 203-209 scope for activity, 204 Metamorphosis, 190, 191 Methallibure, 42, 43, 46, 96 Methods in tissue culture, 273-286 la-Methylallylthiocarbamyl-%methyl thiocarbamyl hydrazine, see Methallibure Methyltestosterone courtship behavior and, 80 kidney tubule development and, 78 nest building and, 78, 79 nuptial coloration and, 78 schooling and, 91 sex reversal and, 83, 146 sexual behavior and, 78, 84 on skeleton of swordfish, 154, 155 Migratory behavior, 79, 91, see also Hormones, gonadal salinity preference and, 93

482

SUBJECT INDEX

thyroid hormones on, 91, 92, 94 Monoiodoacetic acid, 326 Monolayer cultures, 276-283 Mortality during development, 230-235 MSH, see Melanocyte-stimulating hormone Miillerian duct, see oviduct

N Neostigmine, 435 Nest building, 78, 84, 95 Neuromast organs, 221 Neuromelanophoral junction, 330 Nicotine, 338 Nidamental gland, 20 Norepinephrine, 327, 435 Nuptial coloration, 76, 340

0 Oocytes, 16, 17, 56 Oogonia, 16, 18 Opercular spine, 408 Ornithine carbamyltransferase, 202 Ornithine urea cycle, 202 Ostracitoxin chemistry, 438, 449 lethal dose, 440 ovarian follicle, 15-19 Ovariectomy, 86-88, 124, 154 Ovary chemical attractant in, 89, 91 development, 3-8 hypophysectomy on, 46 structure, 15-19 testosterone in, 142 tunica albuginea of, 16 Oviducal gland, 20 Oviducts, 8, 19, 20 Ovigerous folds, 16 Oviposition, 87, 99 Ovipositor growth, 89, 156 hormonal control, 89 ovariotomy on, 156 steroids on, 156 Ovoviviparity, 21, 23-28, 198 Ovulation, 49, 90, 97, 98 Oxycaloric coefficient, 208 Oxygen consumption, 203-208, 232 Oxygen debt, 206

P Pahutoxin, 4 3 8 4 4 0 Parr-Smoit transformation, 343 Parthenogenesis, 2, 186 PBS, see- Phosphate buffered saline Penis, see Pseudopenis Periblast, 187, 188 Perivitelline space, 185 Phaneratoxic, 402 Phenothiazine, 338 Phenoxybenzamine, 335, see also Dibenamine Phentolamine, 336 Pheromone, 91 Phosphate buffered saline, 274 Photocytes, 360473, see also Photophores Category A, 369, 370 Category B, 369, 370 Photophores, see also Luminescence alveolar organs, 364 anatomy, 3-69 compound, 363-364 electrical stimulation of, 381 fine structure, 369-373 glandular cell types, 369-373 lens, 373-375 mantle, 375 melanophores and, 360 nervous control, 380383 reflector, 3 7 3 3 7 5 rotating device, 363, 382 sacs in body wall, 3 6 4 3 6 6 simple, 360 visceral organs, 366 Phototaxis, 221-223 Physiological salines, 260-265, see also Balanced salt solutions Pigments, 307-353, see also Chromatophores, melanophores carotenoids, 314, 315 chrysopsins, 386 melanins, 313, 314 pteridines, 315, 316 purines, 316, 317 rhodopsins, 386 Pineal, 343 Piperoxane, 336 Pituitary, see also various hormones behavior and, 94-103

SUBJECr INDFX

gonadal relations, 43-47 gonadotropins, 41-50 Placenta in sharks, 28-30 in teleosts, 30-39 yolk sac, 29 PMS, see Pregnant mare serum Poisoning ciguatera, 425-430 clupeoid, 437 cyclostome, 424 elasmobranch, 424 gemphylid, 438 scombroid, 436, 437 tetrodon, 4 3 0 4 3 6 Poisonous fishes, 1, 423-440 Poisons, 401-449 Polymyxin B, 274 Polyspermy, 185 Pregnancy in elasmobranchs, 2 3 3 0 hypophysectomy and, 47 in teleosts, 30-39 Pregnenolone, 80 Premelanosome, 309 Presidal, 338 Priapium, 34 Procaine, 338 Progesterone, 51, 54-56, 78, 153 Prolactin brooding behavior and, 82 gonadotropic action of, 50 mucus cell production and, 101 parental behavior and, 100-102 Pseudobranch, 344 Pseudopenis, 34 Pseudoplacenta, 38 Pteridines, 308, 315, 316 Purines, 316, 317 Pyrogallol, 338

Q Qo,. 204 Qio,

194

R Rearing techniques, 238-240 Reflector, see Photophores Reproduction, 1-59 behavior of, 73-109

cephalization of control, 106 cycles and coordination of, 5 8 5 9 endocrinology of, 40-59, 75-102 environmental control of, 57-59 feeding cycles and, 57 Reserpine, 99-100,338 Respiration, 203-209 stingray venom on, 416 Rheotropism, 227 Rhodopsins, 386 RTG-2 cells, 284

S Saline, see Balanced salt solutions Salinity preference, 93, 94 Saurine, 437 Saxitoxin, 403 Schooling development of, 226 gonadal hormones on, 91 thyroid hormones on, 91 Selachine, 328 Seminal vesicles, 15 Seminiferous tubules, 11-12 Sense organs, 220-221 Sepiapterinosome, 316 Sepiapterins, 316 Sertoli cells, 9, 13, 14 Methallibure on, 46, 47 Sex characters accessory female, 19, 20 castration and testosterone on, 51, 81, 155 differentiation of, 153-157 female-positive, 154-156 gonadectomy on, 51 male-positive, 154 male secondary, 14, 15, 154 Sex chromosomes, 131-139 Sex determination genetic basis, 131-142 heterosomal, 133, 134 polyfactorial, 135, 136 polygenic, 134-139 Sex differentiation, 4, 117-175, see also Sex determination androgens and estrogens on reversal, 145-150 corticoids on, 143 functional reversal, 144-150

484

SUBJECT INDEX

inducer theory of, 150, 151 normal, 149 polygenic determination, 134-139 sex hormones and, 142-144 Sex inducers, 149, 150-153 Sexogens, 150, 152 Sex reversal, 122, 126, 127, 134-142 Sexual behavior, see Behavior, Reproduction Silvering, 343 Sodium Merthiolate, 274 Sodium D-toluene-sulfonchloramide,see Chloiamine Spawning behavior, 97-100 Sperm, 6-12, 58-59 storage in uttro, 186 Sperm ducts, 5-8, 11-12 secretions, 14, 15 Spermatid, 8, 11, 44 Spermatocyte, 8 Spermatogenesis, 8-12, 44, 58 hypophysectomy on, 9, 41, 96 Methallibure on, 43 Spermatogonia, 8, 11, 43, 44, 58 Spermatophores, 11, 14 Spermatozoa, see Sperm Spermiation, 14, 15, 98 Spermiogenesis, 8 Starvation, 218 219 Steroidogenesis, 43, 5258, 152 Sting, 405, 406, 414 Stingray venom, 414-416 Superfetation, 40 Syphon, 22, see also Clasper syphon

T Testis development of, 3-8 endocrine tissues of, 12-14 gametogenetic tissue of, 8-12 histology, 10-12 hypophysectomy on, 46 structure in Chondrichthyes, 10, 44 in Poecilia, 11 Testosterone, 51-54, 152, see abo Androgens conjugated, 53 propionate, 80, 146, 149 in tissues, 52, 53 treatment effects, 77-85

Tetrodon poisoning, 430-436 Tetrodotoxin chemistry of, 431, 432 concentrations in tissues, 431 dose-response curve, 432 ingestion by mammals, 434 on melanophores, 338 molecular formula, 431 pharmacology of, 432436 toxicology of, 432436 Thiourea, 93, 213 Thyroid development of, 212-213 hormones and behavior, 90-94 Thyroxine, 91, 213 Tissue culture, see Cell and tissue culture. Toxinology, 401 Toxins, 402404 Trophotaenia, 36 Trypsinization, 279 TTX, see Tetrodotoxin Tubocurarine, 435 Tunica albuginea, 16 Tyrosinase, 314, 340, 342

U UGP, see Urogenital papilla Ultraviolet, 233 Urogenital papilla, 155, 156 Urogenital system, 4-8 Uterus histology of lining, 24-25, 27 secretions of, 24, 28

V Vasa efferentia, 11, 12 Vas deferens, 6, 10, 11 Venom, 401-449, see abo Poisoning apparatus, 404-414 chemistry of, 414-423 pharmacology of, 414-423 toxicology, 414-423 Venom gland, see Venom apparatus Venomous fishes, 402, 404-423 Vitellogenesis, 18 hypophysectomy on, 45 Viviparity, 2, 20-40, 198

485

SUBJECr INDEX

in elasmobranchs, 22-30 in teleosts, 30-40

hormonal control, 331, 334 K' on pigment of, 334

Y

X Yohimbine, 336 Xantho-erythrophore, 308 Xanthophore, 308, 311, 314, 340 cytoplasm of, 310, 311 epinephrine on, 335, 339

Yolk storage, 195, 213 utilization of, 25, 26, 194-198 Yolk sac, 26, 29

This Page Intentionally Left Blank