M^\ i i i m M m Jiiiiiir liiiMf 0 HHiP PH i!';''iH ii5* w'.m mm I ! mm m\\m m m m CO — ^— ^ ^ m^m^^ r-R ^^^ss O m ^= a =^ D 1 — — A HISTORY OF FISHES ANGEL-FISH {Pterophylhwi eiwekei). Photograph by Mr W. S. Pitt. A HISTORY OF FISHES by J. R. NORMAN F.L.S., F.Z.S. A ssistant Keeper, Department of Zoology, British Mtiseum {Natural History) ILLUSTRATED BY LiEUT.-CoL. W. P. C. TEN I SON D.S.O., F.Z.S. WITH 9 PLATES AND 1 47 TEXT-FIGURES History. — A written statement of what is known ; an account of that which exists or has existed ; a record ; a description." — English Dictionary. ERNEST BENN LIMITED London : Bouverie House, E.C.4 FIRST PUBLISHED IN I 93 I PRINTED IN GREAT BRITAIN TO MY FRIEND AND COLLEAGUE C. D. SHERBORN ^m%,^ CONTENTS CHAPTER PAGE List of Plates ....... xi Preface ....... xiii I. Introductory ....... i Definition of a fish. Position in the animal kingdom. Difference between fish and Cetacean. Classes of fishes. Numbers of species and individuals. The science of ichthyo- . II. Form and Locomotion ..... 9 Shape of a typical fish. Fins and their functions. Other animals with a fish-like form. Departures from the ideal form, and compensating factors. Depressed and compressed fishes. Flat-fishes. Fishes with rounded bodies : Globe-fishes, PuflTers, Sun - fishes, etc. Elongate fishes. Sea Horses. Methods of locomotion. Muscular movements. Swimming of Mackerel, Eel, and Trunk-fish. Locomotion by means of fin movements: caudal fin, dorsal and anal fins, pectoral fins. Jet propulsion. Speed. Swimming positions. Leaping. Burrowing. III. Respiration ...•••• 33 How fishes breathe. Structure of gills in Selachians, in Marsipobranchs, in Bony Fishes. Lophobranchs. Process of respiration in a t^^Dical fish, in Lampreys and Hag-fishes, in Rays, in Trunk-fishes. Does a fish drink? Gill-rakers and their uses. Rate of breathing. Fish that can live out of water. Accessory organs of respiration: breathing through the skin, intestinal respiration, labyrinthic organs, air- breathing sacs. Air-bladder and its functions. Origin and evolution of air-bladder. IV. Fins 54 Different kinds of fins. Their origin and evolution. Struc- ture of fins: dorsal and anal, pectorals and pelvics, caudal. Types of tail. Development of tail. Modifications of dorsal and anal in different fishes, of caudal, pectoral and pelvic fins. V. Skin, Scales, and Spines ..... 84 Structure of skin. Dermal denticles of Selachians. Tail- spine of Sting Rays. Saw-fishes. Scales of Bony Fishes : ganoid scales, cycloid and ctenoid scales. Tubercles. Bony scutes. Armoured fishes. Bony plates, rings, spine.?. Scales 42093 ii CONTENTS CHAPTEK PAGE of Lung-fishes. Arrangement of scales, scale counts. Axil- lary scales. Lateral line. Scale-reading and age-determina- tion. Replacement scales. VL Mouths and Jaws . . . . . .103 Form of mouth in Cyclostomes, in Selachians. Jaws of Selachians, of Bony Fishes. Form and position of mouth in Bony Fishes. Modifications of lips. Weevers and Star- gazers. Gar Pikes, Gar-fishes, and Half-beaks. "Sword- fishes." Tube-mouths. Fishes with protractile mouths. Flat-fishes. Vn. Teeth and Food . . . .120 Teeth of Cyclostomes and Selachians. Succession of teeth in Sharks and Rays. Diflferent kinds of Shark dentition. Man-eating Sharks. Thresher Shark. Nurse Sharks and Bull-headed Sharks. Teeth of Rays. Dentition of Chimaeras. Teeth of Bony Fishes. Pharyngeal teeth. Depressible teeth. Feeding habits of Pike, Caribe or Piraya, Blue-fish, Lancet- fish, Barracuda. Canine teeth: Chauliodus, Cynodon. Denti- tion in fishes with mixed diet. Archer-fish. Plankton feeders. Wrasses. Parrot-fishes and Plectognaths. Cyprinids. Stro- mateoids. Unusual meals. A remarkable " Shark story." VI n. Venom, Electricity, Light, and Sound . .140 Poison glands in Selachians. In Bony Fishes: Cat-fishes, Weevers, Scorpion-fishes, Toad-fishes. Eflfects of venom. Treatment. Fishes with poisonous flesh. Electric organs: Torpedo, Electric Eel, Electric Cat-fish, Skates and Rays, Mormyrids, Star-gazers. Origin of electric organs. Nature of the discharge. Uses of electric organs. Luminous fishes. Photophores. Other light organs. Production of light. Purpose of luminous organs. Production of sound : by the air-bladder, by stridulation, by elastic spring mechanism. Sound-producing fishes. IX. Internal Organs . . . . . -159 Skeleton: skull, vertebral column. Muscles. Alimentary canal : mouth, tongue, stomach, intestine, rectum. Vascular system : heart, arteries and veins, lymphatic system. Kidneys. Air-bladder. X. Nervous System, Senses, and Sense Organs . 177 Nervous system: brain, spinal cord, nerves. Sensory and motor nerves. Olfactory organs. Sense of smell. Eyes: structure, modifications, position. Sense of sight. Auditory organs. Otoliths. Connection of air-bladder with internal ear. Weberian mechanism. Sense of hearing. Internal ear and equilibrium. Sense of taste. Of touch. Barbels. Other tactile organs. Lateral line: structure, functions. Sense of pain. Schooling or shoaling sense. Sleep. Reflex and conscious actions. CONTENTS ix CHAPTER PAGE XI. Coloration ....... 207 Colours of some tropical fishes. Variations. Meanings of coloration. Obliterative shading. Oceanic fishes. Shore fishes. Fishes of coral reefs. Protective resemblance. Mimicry. Coloration and environment. Colour changes. Sexual differences. Warning colours. Mechanism of colora- tion: chromatophores, iridocytes, etc. Mechanism of colour changes. Xanthochroism. Influence of light on pigment. Coloration of Flat-fishes. Ambicoloration. Albinism. XII. Conditions of Life ...... 230 Deep-sea fishes. Cave-dwelling fishes: Kentucky Blind-fish, Cuban Blind-fishes. Evolution of cave faunas. Californian Blind Goby. Sponge-inhabiting Gobies. Hill-stream fishes and their modifications. Effects of temperature on fishes. Catastrophe of Tile-fish. Hibernation and aestivation. Association of Pilot-fish and Shark. Commensalism : Remora and Shark, Pomacentrid and anemone, Rudder-fish and jelly-fish, Fierasfer and echinoderm or oyster. Symbiosis. Parasitism: Candiru. XIII. Distribution and Migrations .... 252 Science of Zoogeography. Marine and fresh-water fishes. Oceanic fishes. Coastal fishes. Zones of distribution. Tropical zone : Panama and Suez Canals. South Temperate and Antarctic Zones. North Temperate and Arctic Zones. Migrations of marine fishes: Tunny, Mackerel, Pilchard, Herring, etc. Races of Herring. Origin of fresh-water fishes. Catadromous and anadromous fishes. Distribution of Salmonidae. Distribution of Ostariophysi. Zoogeographical regions. Australian Region: Wallace's Line. Madagascar. Neotropical Region. Fishes of South America and Africa compared. Ethiopian and Indian regions. Palaearctic region. British fresh-water fishes. Nearctic region. XIV. Breeding ........ 279 Reproductive organs. Fertilisation. Ancient theories of spawning. Spawning of Cod, Plaice, Herring, etc. Number of eggs produced. Spawning of Salmon. Of Sea Lamprey. Breeding habits of Fresh-water Eel. Of Cyprinids. XV. Pairing, Courtship, and Parental Care . . 294 Intromittent organs: of Selachians, of Cyprinodonts. Breed- ing habits of Cyprinodonts. Secondary sexual characters. Courtship of Fighting-fish: of Dragonet. Pugnacity of males at breeding season. Parental care: primitive nests. Nests of Mud-fishes, Bow-fins, Cat-fishes, etc. Breeding habits of Three-spined Stickleback, of Fighting-fish and Paradise-fish, of Bitterling. Cichlids and Cat-fishes carrying eggs in the mouth and attached to the body. Care of eggs in marine fishes: Lump Sucker, Gunnel, Kurtus. Egg-pouches in Pipe-fishes. Parasitic males in oceanic Angler-fishes. X CONTENTS CHAl'TEK PAGE XVI. Development . . . • • 3^7 Gametes. Eggs of Cyclostomcs and Selachians. Eggs of Bony Fishes: pelagic and demersal eggs. Segmentation. Yolk. Embryonic development. Viviparous fishes. Con- nection between embryo and parent. Embryonic develop- ment of a Shark. Development of Salmon. Larvae and larval organs: Ammocoetes ; external gills. Larval Sun-fishcs, Deal-fishes, "Sword-fishes," Gar-fishes, etc. Larval develop- ment and metamorphosis of Fresh-water Eel. Metamor- phosis of Flat-fishes. Hybrids. XVIL Fossils and Pedigrees . . . • • 34^ Taxonomy and the science of palaeontology. Stratified rocks. Geological record. Fossil Marsipobranchs : Ostraco- derms. Palaeospondylus. Placoderms: Arthrodira and Anti- archa. Fossil Selachians : Cladoselache, Climatius, Pleuracanthus, etc. Origin of Bony Fishes. Palaeopterygii : Palaeoniscids, Platysomids, Catopterids, Chondrosteids, etc. Crossoptery- gii: Rhipidistia, Actinistia, and Dipneusti. Neopterygii: Semionotids, Pycnodonts, Eugnathids, Pachycormids, etc. Primitive modern Bony Fishes. XVIII. Classification 3^3 Phylogenetic trees. Species and their origin. Races, sub- species, and varieties. Genera and sub-genera. Nomen- clature. Classification of Marsipobranchs, of Selachians, of Bony Fishes. XIX. Fishes and Mankind . . . • . -3^1 Fish as food. Composition of flesh. Value of fresh-water fishes. Diflferent kinds of edible fishes. Commercial value of deep-sea fisheries. Fishing industry of Great Britain. Fishing methods: trawling, seining, drifting, lining. Pre- servation of fresh fish. Fish curing, etc. By-products of fishes: oils, meals, fertilisers, glue, isinglass, leather, etc. XX. Fishes and Mankind [continued) .... 402 Harvest of the sea. Dangers of over-fishing and possible remedies. Fishery investigations. Methods of research. Pisciculture. Artificial propagation. Introduction of fishes into new countries. Angling. Improvement of Salmon and Trout fishing. Pollution and hydro-electric schemes. Fish diseases: Furunculosis, Saprolegnia, Salmon Disease. Fish parasites: Sea Lice, Gill Maggots, Worms. Monstrosities. Varieties of Goldfish. Fishes used in controlling disease. Aquaria. Fishes in museums. XXI. Myths and Legends, etc. . . . • • 424 Size: largest and smallest fishes. Longevity. Medicinal uses of Tench and other fishes. Myth of the "ship-holder." Monk-fish, Bishop-fish, Mermaid and Sea Serpent. Fishes from the clouds. The Miraculous Draught. Crucifix-fish. List of Books .....•• 435 Index 437 LIST OF PLATES Angel-fish {Pterophyllum eimekei) ..... Frontispiece NUMBER OF PLATE To face page I. Scales of the Salmon (vSa/mo i-^/ar) . . . .102 II. (a-b) Colour changes in a Mediterranean Flat-fish {Bothiis podas) ..... 220-221 III. Cocoon of African Lung-fish or Mud-fish {Protopterus annectens) embedded in mud .... 244 IV. Stages in the metamorphosis of Leptocephalus brevi- rostris, the larva of the European Fresh- water Eel 337 V. Stages in the metamorphosis of the Plaice {Pleuronectes platessa) . ...... 338 VI. Slab of chalk (Sussex) with remains of Hoplopteryx superbus ....... 362 VII. (a) A downpour of fishes in Scandinavia; (b) Fishing with the Remora; (c) The great Sea Serpent . 428 PREFACE In the course of my work at the British Museum I am called upon from time to time to supply answers to all kinds of strange questions, some of them but remotely connected with fishes themselves. How fast does a fish swim? How many fishes are there in the sea? Why does a fish die when taken from the water? Where did fishes first come from? To what age does the average fish live? Can a fish think or feel pain? (A favourite query from the angler!) What is Rock Salmon? Are we depleting the stocks of fishes in the sea by over-fishing? It is in the hope that it will provide solutions to these and other problems that the present work has been written, and, be- lieving that it has been planned on more or less original lines, I feel that no apology is needed for its publication. At the same time, it is hoped that it will serve as more than a mere book of reference — a storehouse of facts — and will prove of suflficient interest to provide general reading, not only for the student of fishes and the angler, but for all those who take an intelligent interest in wild life. The customary method of dealing with any group of animals is to begin with a recognised scheme of classification, and to take up each of the smaller groups in turn, describing the main distinguishing features of some of the better-known members of each group, their mode of life, food, distribution, and so on. Sometimes one or two chapters devoted to the anatomy, development, etc., of the animals precede the more general part, but, as a rule, these subjects are omitted altogether or dismissed in a few lines. In the following pages I have tried to give some idea of the story of fish life in all its varied aspects, to show how the fishes "live and move and have their being." In one chapter the manner in which they swim is considered; in another their food; in another their breeding habits, their development, and so on. Many diflferent kinds of fishes are mentioned in illustration of one point or another, and some inevitably figure in more than one chapter. Special stress has been laid throughout on the evolutionary aspect of fish life, the fishes themselves being regarded, not as museum specimens or corpses on the fishmonger's slab, but as living organisms xiv PREFACE which ha\'c been modified in a multitude of different ways in accordance with the nature of their surroundings, in order to fit them for the particular conditions under which they are compelled to live. The importance of the part played by the "struggle for existence" in moulding the bodies of fishes will be apparent, and I have endeavoured to show how many of the remarkable modifications of the various organs which go to make up the body of a fish, although sometimes meaningless at first sight, may be readily interpreted in terms of environ- ment, animate or inanimate. The relation of fishes to the life of mankind has not been neglected, and chapters dealing with the fisheries, fishing methods, fishery research and so on have also been included. The enormous development of our own sea fisheries towards the close of the last century led to a great interest being taken in the habits, and particularly in the feeding and spawning habits, of the edible species. Much important research has been carried out on these problems during recent years, but the results are mostly buried away in scientific journals not readily accessible to the public, who remain largely in ignorance of the work which is being continuously done in order to main- tain or improve the harvest of the sea. In preparing this work I have drawn on my knowledge of the vast literature of the various branches of the science of ich- thyology, and have consequently consulted a large number of works of a technical nature, some of them in foreign languages, not available to the general reader. It would, of course, be of little value to include a bibhography of such works here, but a short list of the more important and accessible books of reference on fishes and kindred subjects in the EngHsh tongue is appended for the convenience of those who may wish to pursue the subject further. The use of technicalities has been avoided as far as possible, and scientific terms have been included only where their omission would be at the expense of clarity. It has seemed to me convenient, however, to refer to each fish by its scientific name (usually only the generic name, but occasionally the specific name as well) in addition to that by which it is popu- larly known, except in the case of lesser-known species for which there are no vernacular appellations. In the legends below the figures the name of the species is nearly always given in full. Regarding the illustrations, the figures in the text are, with PREFACE XV very few exceptions, new, and have all been drawn specially for this work by my friend Lieut. -Col. W. P. C. Tenison. I take this opportunity of offering him my sincere thanks, not only for the great care that he has taken in their preparation, but also for the kindly interest he has shown in the book since its inception. We have been content to make the drawings as simple as possible, believing that it is better to show the salient and characteristic features of the fishes rather than to produce an artistic effect. Those illustrations copied from other works are duly acknowledged in their place, and I am especially indebted to Mr. Arthur Hutton, Professor F. B. Sumner, and to Professor Johannes Schmidt, for permission to reproduce the photographs appearing in plates 1, l.i and IV respectively. It only remains for me to tender my grateful thanks to my colleague Mr. M. Burton for the trouble he has taken in reading through the greater part of the manuscript, and for many helpful suggestions and criticisms; to Dr. E. I. White, for reading and criticising Chapter XVII; to Dr. E. S. Russell,iO.B.E., for performing a like service in connection with parts of Chapters XIX and XX; and to Mrs. Tenison for assistance in the task of passing the proofs for press. Finally, I find it impossible to allow this opportunity to pass without recording the great debt which I owe to Dr. C. Tate Regan, F.R.S., the Director of the British Museum (Natural History) ; his very great know- ledge of matters ichthyological has always been at my disposal, and the many valuable hints and suggestions that he has given me from time to time since my appointment to the museum ha\'e proved of the greatest assistance to me in my work there, and without them it is certain that the writing of this book would have proved a very difficult task. J. R. NORMAN. London, 1931. Certain sections of Chapters II, III, and XI have already appeared in the Salmon and Trout Magazine, and are reproduced here by permission of the Editor, CHAPTER I INTRODUCTORY Definition of a fish. Position in the animal kingdom. Difference between fish and Cetacean. Classes of fishes. Numbers of species and indi- viduals. The science of ichthyology. " These {the fishes) were made out of the most entirely ignorant and senseless beings, whom the transformers did not think any longer worthy of pure respiration, because they possessed a soul which was made impure by all sorts of transgression ; and instead of allowing them to respire the subtle and pure element of air, they thrust them into the water, and gave them a deep and muddy medium of respiration ; and herwe arose the race of fishes and oysters, and other aquatic animals, which have received the most remote habitations as a punishment for their extreme ignorarwe.'* Plato. It is of primary importance in a work of this nature to make it clear from the outset exactly what is meant by a fish, for in popular parlance the word "fish" is often used to include any animal living in the water, a definition which appears in all the older dictionaries. Although convenient, this can hardly be described as scientifically accurate, including, as it does, such diverse organisms as the Whales, Seals, Salmon, Oysters, Cuttle-fishes, Star-fishes, Jelly-fishes, and Sponges, creatures that differ from each other even more widely than do reptiles from birds or birds from mammals. The aquatic animals just mentioned, however, all fall naturally into two main categories in respect of one important bodily feature — those with a vertebral column or backbone and those with none. Man has a backbone, and so have all the mammals, birds, reptiles, amphibians, and fishes; all the others have no backbone. The backboned animals or vertebrates are better known to most people than the majority of the lower animals; indeed, with the exception of a few Hke the oyster and lobster which are eaten as delicacies, the invertebrate animals are regarded for the most part with lukewarm interest, in some cases with actual disgust. This attitude is partly explained by the superior size of the vertebrates, by the greater ease with which they can be observed and studied, and by the beauty of form and colour displayed by many of the birds and mammals. It has been 2 A HISTORY OF FISHES further fostered by the editors of popular works on natural history, who devote three-quarters of the available pages to the mammals and birds, crowding the unfortunate lower animals — every bit as interesting and quite often of extreme beauty — into a few short chapters at the end. A fish, therefore, is a vertebrate, and one specially adapted for a purely aquatic life. But this definition is still inadequate, Fig. I. CETACEAN AND FISH COMPARED. A. Common Dolphin {Delphinus^delphis) ; B.^Mackerel Shark [Isurus oxyrhyn- chus). Both much reduced. for all the vertebrates living in the water are not fishes. What of the Whales, Seals, Otters, Newts, and Frogs? In my official capacity I am sometimes asked to settle arguments, occasionally backed by substantial stakes, as to whether or no a Whale is a fish. Here there is the same fish-like body, the fin-like fore limbs or paddles, and often a fin in the middle of the back (Fig. I a). Nevertheless, a Whale is not a fish, but a mammal. A close examination of its skin reveals the presence of a few vestigial hairs in the region of the muzzle, the structure of the paddle is quite unlike that of the fish's fin (Fig. 2), being in all INTRODUCTORY 3 its essential parts just like that of the human hand, and the so-called dorsal fin is nothing more than a ridge of fatty tissue. Furthermore, although a Whale is able to remain under water for considerable periods of time, it is forced to come to the surface at intervals to empty its lungs of air and to inhale a fresh supply of oxygen — the familiar process of spouting or blowing. Whales also bring forth their young alive, and after birth suckle them just like any other mammal. In short, a Whale is a mammal which has left its kindred and exchanged a terrestrial life for one passed entirely in the water, a change Fig. 2. PECTORAL LIMB OF CETACEAN AND FISH COMPARED. A. Skeleton of paddle of the Common Dolphin {Delphinus delphis) ; b. Skeleton of pectoral fin of the Comb-toothed Shark (Heptranchias perlo). Both much reduced. which has led to the fore limbs being converted into paddles for swimming, while the hind limbs have completely dis- appeared. The Seals give us some idea how this change has come about, representing, as it were, a half-way stage between a typical walking mammal and a specialised swimming one. A Seal is amphibious; that is to say, it is equally at home on land or in the water; but the hind limbs have lost a great deal of their power of supporting the body on terra Jirma, and the fore limbs are becoming more and more paddle-like, the shape of the body tapering and fish-like, and the external ears have more or less disappeared. The form of the tail provides a rough-and-ready means of distinguishing at a glance any 4 A HISTORY OF FISHES member of the Whale tribe (Cetacea) from a large fish such as a Shark; in the Cetaceans the flukes or lobes of the tail are horizontal, in the fishes they are vertical (Fig. i). It is of some interest to note that Aristotle (384-322 B.C.) was well aware of the differences between fishes and aquatic mammals, whereas many of the writers in historical times classed them all together as fishes. The distinctions between the two groups do not appear to have been generally understood until the later part of the seventeenth century, and ignorance as to the real nature of the Cetaceans must often have led our pious ancestors to break Lent, since they enjoyed steaks and cutlets of Whale, Porpoise, or Seal on fast days under the fond delusion that they were consuming fish ! The Ichthyosaurus, an extinct aquatic reptile, exhibits the same general fish-like form and paddle-like limbs, but these have clearly been acquired independently, as in the Whales, as a result of the adoption of a life in the water (Fig. 3). There is yet another creature, common in all our ponds and streams during the spring months, often confused with the fishes in the popular mind, namely the tadpole, which is, of course, the young stage of a Frog or Toad. The Newts, Sala- manders, Frogs and Toads belong to a class of vertebrates known as Batrachians or Amphibians, the latter name referring to the fact that they are not only amphibious in the popular sense, living partly in the water and partly on dry land, but are also actually adapted during the early part of their life to breathe under water by means of gills like the fishes, and at a later period to breathe air by means of lungs like the reptiles. But some amphibians never breathe under water at any stage of their existence, not even when immature, and others retain their gills throughout life. How, then, is it possible to distinguish any amphibian from any fish? By the organs of locomotion. In all amphibians the paired limbs are legs in the adult state, in fishes they are fins. To summarise, a fish may be defined as a vertebrate adapted for a purely aquatic life, propelling and balancing itself by means of fins, and obtaining oxygen from the water for breathing purposes by means of gills. Fishes, thus defined, were formerly regarded as representing a single class of the great sub-kingdom of vertebrates, a class equivalent to the birds {Aves) or the reptiles [Reptilia) ; but a more thorough knowledge of their anatomy and evolutionary history has led to a different conclusion. The Lampreys and their allies INTRODUCTORY 5 (Cyclostomes) , with their pouch-like gills and mouths devoid of biting jaws, resemble some of the true fishes to a certain super- ficial extent in outward form, in habits, and in their general manner of breathing, and may well be regarded as "fishes" Fig. 3. — EXTINCT REPTILE OF FISH-LIKE FORM. Restoration of the Ichthyosaurus. Much reduced. in the popular sense. Actually, the two groups of animals are separated by characters just as fundamental as those which divide all the other fishes from the amphibians, and the Cyclo- stomes must, therefore, rank as a separate class {cf. p. 344). The same is true to a lesser extent of the Selachians, a group including the Sharks, Rays, and Chimaeras, which have been 6 A HISTORY OF FISHES separated from the Bony Fishes for a very long period of the earth's history [cf. p. 350). In speaking of fishes, therefore, it must be remembered that we are referring to three very distinct vertebrate classes, here grouped together merely for the sake of convenience. Certainly in number of individuals, and probably also in number of species, fishes are at the present time superior to mammals, birds, reptiles, or amphibians. Recollecting that three-quarters of the earth's surface is covered by the seas, and that many of the fresh waters of the land teem with fish life, this superiority of numbers is easier to understand. The sur- faces of the great oceans, their middle layers, the abyssal depths and shore regions; the estuaries, mighty rivers, swiftly flowing brooks, turbulent mountain torrents and placid lakes and ponds; each of these possesses its peculiar forms of fish life, variously modified according to circumstances. There are probably more than 20,000 different species of fish in existence to-day, and 100 or more new forms seem to be discovered every year. Aristotle seems to have been familiar with only about 115 species, all of them found in the iEgean Sea. Pliny {circa a.d. 200), whose list included as many as 176 species, triumphantly exclaims: "In the sea and in the ocean, vast as it is, there exists, by Hercules! nothing that is unknown to us, and a truly marvellous feat it is that we are best acquainted with those things which Nature has concealed in the deep." Sancta simplicitas! Regarding the number of individuals of any particular species, it is wellnigh impossible to give an adequate idea of their abundance. It has been estimated that nearly 400,000,000 Cod and more than 3,000,000,000 Herring are caught each year in the Atlantic and adjacent seas alone, and these numbers must represent but a minute proportion of the individuals in existence at a given time. The particular branch of zoology which treats of the structure of fishes, both external and internal, their mode of life, their distribution in space and time, etc., is known as "ichthyology," a word derived from the Greek ichthys, a fish, and logos^ a discourse. The scope of ichthyology is enormous, and it is almost impossible to deal with its many branches within the compass of a single volume. The anatomist, in- vestigating and comparing the internal structure of the various kinds of fishes ; the embryologist, concerned with the develop- ment of the individual from the Ggg to the adult; the evolu- tionist, studying the past history of the group; the systematist INTRODUCTORY 7 or taxonomist, classifying the fishes, and arranging them in larger or smaller groups according to their differences and resemblances; the statistician, dealing with minute variations and huge numbers of individuals; the physiologist, studying living activities and the function of organs and tissues ; and the "field" naturalist, observing the relation of living fishes to their environment; each of these is continually adding his quota to the sum total of our knowledge of the science of fishes. Moreover, the story of fish life does not begin and end with the fishes living at the present time, but had its commencement ages and ages ago, long before man made his appearance on this planet, and when there were no reptiles or amphibians, no birds, and no mammals. For facts concerning the past history of fishes we are indebted to the geologist, who studies the formation of the rocks wherein the records lie buried, and to the palaeontologist, who spends his time searching the dried- up basins of ancient seas and lakes, and describing the fossilised remains which may be found there. As will be shown later on, the story of the rocks — the geological record, as it is called — is necessarily fragmentary and very imperfect, but has already provided a mass of evidence which has confirmed or modified the conclusions drawn from the study of anatomy, embryology, etc. Nor is this all. The body of a fish, as well as its inanimate environment, is continually subject to physical and chemical laws, so that, in order to arrive at a full understanding of fish life, it is necessary to go beyond the realms of pure biology and draw upon the researches of the chemist, physicist, meteorol- ogist, and even the mathematician. The history of ichthyology, like that of zoology itself, may be said to have begun with Aristotle, who recorded a vast array of facts concerning the fishes of Greece. His information relative to their structure, habits, migrations, spawning seasons, etc., is, so far as it has been tested, extraordinarily accurate, but his ideas of species were exceedingly vague, being simply those of the local fishermen from whom he obtained the names of his specimens. As Dr. Giinther has observed: "It is less surprising that Aristotle should have found so many truths as that none of his followers should have added to them." Pliny, Aelianus, Athenaeus, and others certainly recorded some original observations, but the majority of scholars from the time of Aristotle until some eighteen centuries later were content to copy from his works, merely adding a number of fabulous stories and foolish myths. In the middle of the sixteenth 8 A HISTORY OF FISHES century the publication of the mighty works of Belon (1518- 1564), Rondelet (1507-1557), Salviani (1513-1572) and others, gave a fresh impetus to the study of the science, and estabUshed the idea of a species once and for all. From this time onwards the progress of ichthyology was rapid and continuous, and its history includes the names of Linnaeus, Risso, Rafinesque, Bloch, Lacepede, and Cuvier, men whose pioneer work, often carried out in the face of great difficulties, with small material and inadequate apparatus and instruments, has laid the firm foundations of the science upon which modern ichthyologists are still building. CHAPTER II FORM AND LOCOMOTION Shape of a typical fish. Fins and their functions. Other animals with a fish-like form. Departures from the ideal form, and compensating factors. Depressed and compressed fishes. Flat-fishes. Fishes with rounded bodies : Globe-fishes, Puffers, Sun-fishes, etc. Elongate fishes. Sea Horses. Methods of locomotion. Muscular movements. Swim- ming of Mackerel, Eel, and Trunk-fish. Locomotion by means of fin-movements: caudal fin, dorsal and anal fins, pectoral fins. Jet propulsion. Speed. Swimming positions. Leaping. Burrowing. Of the many and varied forms of animal life found in the seas and in the fresh waters few are more perfectly adapted for dwelling in a liquid medium than the fishes. Many of the invertebrates spend the greater part of their lives attached to the sea bottom, or crawl sluggishly over a small area of its surface; others float more or less passively at the surface or in the middle layers of the water, their movements dependent to a great extent on the action of the tides and currents; the Squids and Cuttle-fishes alone approach the fishes in rapidity of motion and grace of form, but lack their agility in the water, and are generally far inferior to them in mastery of their medium. The water in which a fish lives and 'moves is a comparatively dense medium, and in order to attain the most efficient move- ment with the greatest economy of energy a certain form of body is essential, varying somewhat in detail with the actual speed required. The shape of the body, therefore, is not an arbitrary one, but conforms to a number of definite mechanical conditions induced by its environment. These mechanical principles cannot be dealt with here, since a study of this subject would involve the consideration of such theoretical problems as "curves and displacement," "streamlines," "en- tering angles," "runs" and the like, the proper understanding of which entails some knowledge of higher mathematics. It must suffice to point out that the fine form of a typical swift- swimming fish such as the Mackerel {Scomber) (Fig. 5A) or 9 10 A HISTORY OF FISHES Bonito {Gymnosarda) (Fig. 4) is one that is admirably adapted from a mechanical point of view for cleaving the water, and is that which is clearly best suited for progression in that medium. The mechanical conditions which led man to construct his submarine to a certain pattern, in order to have a vessel that would move freely in all directions under water, are precisely the same as those which have determined the shape of the fish's body, so that it is not surprising to find that the form of the animate fish corresponds closely with that of the man-made submarine. The shape of the body of a Mackerel is fusiform ; that is, it is shaped somewhat like a cigar, circular or elliptical in cross- Fig. 4. — A SWIFT PELAGIC FISH. Oceanic Bonito {Gymnosarda pelamis), X g. section and thicker in front than behind. It has been described by one authority as resembling a double wedge, the thick part of which is represented* by the head and anterior part of the body, and one of the thin edges by the free hinder margin of the tail. Every line of its smooth, rounded contour is sug- gestive of swift motion, there being an almost complete absence of irregularities or projections calculated to hinder progression. There is no distinct neck as in the land vertebrates, the head merging insensibly into the trunk and the trunk into the tail, the boundaries between these regions of the body being denoted by the gill-opening and vent respectively. Viewed from the front, the outline of the fish appears as a perfect ellipse of comparatively small size (Fig. 5A). The beautifully moulded, bullet-shaped head, with its pointed snout forming an efficient cutwater; the jaws fitting so close together that it is scarcely possible to insert the blade of a penknife between them; the FORM AND LOCOMOTION II firm, smooth eyes, carefully adjusted so that their surfaces are level with the adjoining surfaces of the head; and the closely Fig. 5. DIFFERENCES IN FORM. A. Mackerel (Scomber scombrus),X l ; B. Trunk-fish (Ostracion gibbosus),X I ; C. Sun-fish (Mola mola),x4^\ d. Globe-fish {Chilomycterus antennatus),x\ \ E. Sea Horse {Hippocampus punctulatus), x ^ ; f. Common Eel {Avgidlla anguilla). fitting gill-covers ; all these are features whose meaning becomes clear when interpreted in terms of rapid progression. The small scales with which the body is covered offer practically A HISTORY OF FISHES no resistance to forward motion, presenting a comparatively smooth surface, which is still further improved by the presence of a copious supply of slime. This mucous covering is designed to reduce friction with the surrounding water to a minimum, Fig. 6. — " SWORD-FISHES." A. Sail-fish (Istiophorus americanus) ; b. Spear-fish {Makaira mitsukurii) ; c. Sword-fish (Xiphias gladius). All much reduced. not only on account of its inherent slipperiness, but also because it fills up any small irregularities in the surface of the body. Finally, the smooth hollow curves of the hinder end of the body, extending from the region of greatest thickness back- wards to the tail, are admirably adapted to permit of the passage of the water displaced during forward motion. FORM AND LOCOMOTION 13 The fins, which form so characteristic a part of the fish, may well be considered here. Within recent years a French scientist has carried out some very interesting experiments, during which bodies composed of more or less plastic material were drawn through the water at varying speeds. As this took place these artificial bodies tended to become more or less closely moulded to the typical fish-like form described above. At the same time, however, they were inclined to roll over to one side when poised motionless in the water, and to wobble to an alarming extent when moving at any speed, and it was only by the appropriate placing of artificial keels that he was able to stabilise the flight of his models. There can be little doubt Spinous Dorsal ^ ^ , "'"'" ttitT/Ty^ Soft Dorsal J^lililtifilll^ .v^^^K'' Caudal Pectoral Petvics Fig. 7. — TOPOGRAPHY OF FINS. that it was in order to obtain stability that wedge-shaped body keels have been evolved along the back and belly of a fish, as well as paired balancing and steering organs projecting from the sides. These keels and balancers are the fins, and, although they are discussed in greater detail in another chapter, it will be necessary to study their arrangement and distinctive names in the accompanying figure (Fig. 7) in order to understand the part which they play in locomotion. The fins are of two kinds, median or unpaired, and paired. The median fins or keels consist of a dorsal in the middle line of the back, an anal on the belly behind the vent, and a caudal or tail-fin at the hinder end of the fish, which, in addition to assisting stability, also acts as a rudder and plays an important part in forward movements. In the fast-swimming Mackerel the dorsal and anal fins form 14 A HISTORY OF FISHES sharp, thin keels, and, although these appear prominent when viewed from the side they are much less so in the front aspect of the fish (Fig. 5A). The paired fins are of two kinds only, the pectorals and pelvicSy corresponding to the fore and hind limbs of land vertebrates. Further evidence that the characteristic form of a typical fish has been determined by its environment is provided by the study of the evolution of other aquatic vertebrates. The Cetaceans, mammals which have forsaken the land and re- turned to a life spent entirely in the water, have revived the fish-like shape, a result which has been brought about by important anatomical and physiological changes. At the same time, their mammalian ancestry is reflected in the diflferent arrangement of the muscles used in swimming, as well as in the swimming movements themselves. Again, in the extinct reptiles, the Ichthyosaurs, we find the same fusiform shape, powerful tail, wedge-like dorsal fin and paddle-like limbs (Fig. 3), clear evidence of the adoption of an aquatic life. The same external conditions acting over an immense period of time on countless generations has produced much the same result in three totally different groups of vertebrate animals, an excellent example of what is known as convergence in evolution. So much for the form of a typical pelagic fish. Departures from this ideal shape of body are both numerous and varied, but it must be remembered that many which at first sight would appear to be anything but streamline shapes will prove to be excellent ones when the particular mode of locomotion of their possessors is considered. One fact is quite obvious: any radical departure from the ideal form must inevitably lead to a loss in swimming efficiency, or, at least, to a marked re- striction of speed, and this becomes more and more evident the further the fish departs from the typical shape. A fish like the Mackerel depends on its speed, not only to obtain its food, but as a means of escape from enemies, and any marked restriction of its activities would leave it liable to the danger of extinction. It is only where rapid mobility ceases to be of primary im- portance to the life of the species, and is replaced by some other compensating factor, such as heavy armour, that a fish is able to dispense with the fusiform shape and survive in the struggle for existence. Three examples selected from the class of Selachians will serve to illustrate this point. The Blue Sharks and their allies FORM AND LOCOMOTION 15 (Carcharinus) possess slender, perfectly streamlined bodies, conical heads, pointed snouts, and powerful muscular tails (Figs. 23c; 82 a) ; the Carpet Sharks {Orectolobus) have stout thick-set bodies, considerably flattened from above downwards, massive heads, broadly rounded snouts, wide mouths, and much reduced tails with comparatively small dorsal fins (Fig. 82b) ; the Rays (Raiidae) have very broad, flat bodies, the head, trunk and enormously expanded pectoral fins being completely welded together to form a circular or quadrangular disc, from which the feeble tail with its tiny dorsal fins projects as a slender appendage (Figs. 8a; 14B). The Blue Shark is an inhabitant of the open sea, feeding almost exclusively on other fishes which it chases with great vigour; it is essentially a strong, speedy fish, every line of its body intended for rapid progress in pursuit of prey. The Carpet Shark, on the other hand, relies on cunning rather than speed to obtain a meal, lying in wait on the sea floor until the prey comes within reach of its jaws. The loss of swimming power is here compensated for by the remarkable manner in which the Shark resembles its surroundings, its appearance when at rest being that of a weed-covered rock (cf. p. 216). The uniform steely blue coloration of the Blue Shark is replaced by a beautiful variegated pattern which harmonises closely with the sea bottom. The Ray is another sluggish, ground-living fish, and also depends to a large extent on its general resemblance to the surroundings to escape observation by enemies. Its flattened form is ad- mirably adapted for this particular mode of life, but, as will be seen later, its unusual method of locomotion enables this fish to move with much greater rapidity than would appear possible from its appearance. Some still more specialised members of this order have acquired other protective devices in addition to their coloration, as, for example, the Torpedo or Cramp-fish (Torpedo) with its powerful electric organs (Fig. 61), and the Sting-ray (Trygon) with one or more strong, saw-edged and poisonous spines on its tail (Figs. 32B; 36E). A body flattened from above downward is generally spoken of as "depressed," while that which is flattened from side to side is "compressed." Among Bony Fishes the former type is very rare, but the well-known Angler-fish or Fishing-frog (Lophius), in which mimetic resemblance and cunning in obtaining a meal has been brought to a pitch of perfection (Fig. 89) and the Httle Bat-fish (Ogcocephalus) , with the upper surface of its body protected by a covering of hard bony warts i6 A HISTORY OF FISHES (Fig. 40D), provide excellent examples. The compressed body, on the other hand, although unknown in Selachians, is much more common in Bony Fishes. Often it is shortened as well, and, flexibility of the body being no longer an absolute neces- sity, many of these forms are able to afford heavy protective armour of some kind. The brilliant little Butterfly-fishes {Chaetodontidae) of tropical coral reefs are excessively quick in their movements, in spite of their short, deep, flattened bodies, and ihey rely largely on their agility to escape being eaten, coupled with the fact that their deep bodies and strong, spiny fins make them awkward mouthfuls to swallow (Fig. 83G, d). The beautiful Angel-fish (Pterophyllum) of the rivers of South America, a familiar object in aquaria, has a very much com- pressed and almost circular body, and the large fins have some of the rays drawn out into lengthy filaments (Fig. 8c). It is a very slow swimmer, spending most of its time suspended almost motionless in mid-water, and relies on its remarkable resemblance to the water plants among which it lives to escape detection. The Flat-fishes [Heterosomata) , a group which includes such well-known edible forms as the Halibut, Turbot, Plaice, and Sole, all have very much flattened bodies, and, like the Rays, spend much of their time on the sea floor, where their mottled coloration harmonises with the ground on which they lie and renders them inconspicuous (Figs. 8b; 40A-G). In the minds of many people the Plaice and the Skate are lumped together as "Flat-fishes," but it is obvious that the resemblance between the two fishes is a purely superficial one. Both have taken to a life on the bottom, where a flattened body is a decided advantage, but the Skate has become flattened from above downwards, whereas the Plaice is compressed from side to side. In other words, the colourless lower surface of the former which rests on the bottom is the true lower or ventral side, whereas in the Flat-fish this surface represents the right or left side. The Globe-fishes or Puffers (Tetrodontidae) and their relatives the Porcupine-fishes {Diodontidae) provide examples of fishes with shortened, rounded bodies, in which the consequent loss of swimming power is compensated for by the development of some sort of bodily armour in the form of spines or small prickles [cf. p. 98). In addition to their spiny covering, these fishes possess the power of swallowing water or air and thereby inflating the body like a balloon. When thus inflated they are fond of floating passively with the currents, more often than FORM AND LOCOMOTION 17 not upside down. When taken from the water a Puffer will generally inflate at once, but if slow to begin can be persuaded to swell up by gentle tickling. In this condition, and with the spines of the skin all standing erect, the fish is adequately protected against most predatory enemies, who would find it a difficult morsel to bite, much less to swallow. Recent obser- vations made by Dr. Beebe, however, tend to show that such Fig. -FLATTENED FISHES. A. Female Thornback Ray (Raia davata), X yV \ B- Flounder {Flesus flesus), X^; c. Angel-fish {Pterophyllum scalare), X 5 ; d. Oar-fish or Ribbon-fish {Regalecus gles7ie),x about oV- protection is not always complete. He watched a number of little Porcupine-fishes, and saw that when they were threatened by a large Gar-fish, four feet in length, they bunched together for protection, giving the appearance of one large, round and prickly fish; occasionally, however, a single individual would become detached from the mass, when it was promptly seized and devoured. The allied Trunk-fishes {Ostraciontidae) are also slow-swimming creatures, living at or near the bottom of the i8 A HISTORY OF FISHES sea, and rely on their armour, which here takes the form of a rigid bony case, for protection (Figs. 5B; 42F). Also related to the Porcupine-fishes and Puffers are the gigantic and grotesque Sun-fishes (Molidae), of which there are three species, all widely distributed in warm seas. The Round-tailed Sun-fish (Mola) has a very remarkable shape, the deep, circular and somewhat compressed body appearing as though the tail end had been amputated just behind the high dorsal and anal fins (Fig. 5c), a feature to which the popular name of "Head-fish" refers. Such a body is probably well adapted for a more or less passive drift in ocean currents, and it has been suggested that its curious shape is in some way associated with the peculiar diving habits of this fish. The presence of small deep-sea fishes in the stomachs of captured Sun-fishes demonstrates that they must descend to considerable depths at times. The Round-tailed Sun-fish attains to a length of eight feet or more and a weight estimated at more than a ton. It is a sluggish, and, from all accounts, singularly stupid fish, often to be observed basking or swimming lazily at the surface of the sea. Underlying the skin, which is very tough and leathery, is a layer of hard, gristly material some two or three inches thick — ample compensation for any loss of locomotive power ! At the other extreme are the fishes with long bodies, which may be rounded as in the Eels (Apodes) or very much com- pressed, as in the Ribbon-fishes (Trachypteridae) or Cutlass- fishes (Trichiuridae) . From their shape one would hardly expect such fishes to be other than slow swimmers, but, as will be shown below, the adoption of a particular method of loco- motion gives them a greater speed than the short-bodied forms mentioned above. The peculiar shape of the Eel's body (Fig. 5f) is almost certainly associated with its habit of living in soft river bottoms, wriggling in and out of the mud, creeping through reeds, or insinuating itself into holes and crevices as do its relatives in the coral reefs. Some of the Eels carry the elongation of the body to such an extreme that they have the appearance of a piece of slender whipcord, and the fins are often much reduced. Such a filiform type of body is charac- teristic of the curious Snipe-eels [Nemichthyidae) , oceanic forms which sometimes descend to considerable depths (Fig. 320). When observed swimming at or near the surface, these Eels are not infrequently mistaken for snakes. It may be noted here that similarity in eel-like form is not necessarily indicative of close relationship, but may be due to that parallelism in FORM AND LOCOMOTION 19 evolution which has been ah'eady mentioned [cf. p. 14). The so-called Symbranchoid Eels (Fig. 2ib), for example, are in no way closely related to the true Eels, and the same type of body in the two groups has been evolved in response to the needs of similar environments, or as the result of the adoption of the same mode of life. In the Sea Horses {Hippocampus) the form of the body is unique, the head being bent at right angles to the trunk in a manner suggestive of that of a horse, and the trunk itself is definitely curved (Fig. 5E). The possession of a distinct neck is not the only remarkable feature, and the tail is also unique in that it is prehensile and can be used by the fish to anchor it to moving or fixed objects. The body is protected by a series of bony, ring-like plates, and the spines or membranous pro- cesses with which these are ornamented serve to break up the outline and so render the fish inconspicuous when swaying to and fro among aquatic vegetation. The Sea Horses are defence- less creatures, and depend largely on this mimetic resemblance to escape from predatory fishes. The locomotion of fishes provides the biologist and physicist with a number of interesting problems, and has also attracted the attention of the marine engineer, some of whose mechanical inventions owe their inception, at least in part, to observations made upon living animals. Owing to the difficulty of studying fishes in their natural habitat, and to the fact that when trans- ferred to the unnatural surroundings of an aquarium they tend to behave in a manner somewhat diflferent from the normal, our knowledge of the different methods of locomotion is still incomplete. Much progress has been made within recent years, however, and the researches of an American investigator, Mr. Breder, have proved very illuminating. Good results may be expected from the use of the cinematograph, and with the improvement of under-water photography it should be possible to take good films of swimming fishes, and to analyse the movements in detail by the use of the slow-motion picture. Although actual swimming and its associated movements forms the main subject for consideration in this chapter, it must be remembered that this is by no means the only locomotor method in use, and walking or creeping over the sea floor, skipping about on sand or mud, burrowing, wriggling on dry land, leaping, flying, and so on, are also indulged in by some fishes for purposes of progression. These are rather in the nature of specialised developments, however — secondary 20 A HISTORY OF FISHES adaptations, cxolvcd in response to a change in habits or environment, and, since they are often accompanied by a modification of certain organs for the particular end, most of them may be conveniently considered in later pages. Flying, for example, which involves the modification of the pectoral fins, may be discussed in the chapter devoted to those organs. The three classes of vertebrates here grouped together as fishes include a very diversified assemblage of forms, but there is, nevertheless, a basic similarity in their swimming movements, however diflferent these may appear to be at first sight. The earliest fishes probably swam by means of simple rhythmic contractions of the muscles of the trunk and tail, designed to produce certain definite contortions of the body; by the pressure of diflferent parts of the body in succession against the sur- rounding water the animal was driven forward. Most fishes have retained the primitive arrangement of the great body muscles, the myomeres, as they are called {cf. p. 167), which form a series of blocks or segments, arranged in pairs one behind the other and separated by partitions. In this respect fishes diflfer from all land vertebrates, in which the main muscle masses are more or less concentrated on the fore and hind limbs, these being the normal organs of locomotion, whereas the corresponding pectoral and pelvic fins of fishes more often than not perform quite subsidiary functions, such as balancing and steering. We have already noticed the essential similarity in the shape of the body in the cetaceans and fishes, but, owing to its diflferent ancestry, the arrangement of the body muscles in a whale is quite unlike that found in a fish, and the swimming movements themselves are in a diflferent plane, being up and down instead of from side to side. This explains the dissimilarity in the position of the tail-fin in the two groups {cf. p. 4) ; the fish generally swims more or less parallel with the surface of the water, and the vertical caudal fin is designed to assist in driving it forward and to act as a rudder; the whale is under the necessity of coming to the surface from time to time, and swims by alternately rising and diving in a sort of wave- like curve, the horizontal tail assisting to drive the animal upwards or downwards as the case may be. Three primary methods are employed by fishes to produce forward movements while suspended in a fluid medium: (i) body movements due to alternate expansion and con- traction of the myomeres; (2) movements of the appendages (fins) ; and (3) movements caused by the action of jets of water FORM AND LOCOMOTION 21 expelled from the gill-openings during the process of respira- tion. The first method is the most common and of the greatest importance, the others being, for the most part, auxiliary to it. It must be borne in mind, however, that in the majority of fishes the three are inter-related, and may all be used at different times, or even at the same time, for the common end of driving the fish forward according to its requirements. Locomotion by means of fin movements, for example, may be employed when Fig. 9. — BODY MOVEMENTS OF FISHES USED IN SWIMMING. Shark (above) and Eel. (After Marey.) slow progress only is wanted, but, should danger threaten or prey appear in sight, body movements quickly come into play, and at the same time the increased rate of breathing due to the emotion of fright or greed assists in the general acceleration of the forward thrust. The Mackerel {Scomber), which depends almost entirely on body movements for forward progression, will serve as an excellent example of the first of these methods. We have 22 A HISTORY OF FISHES compared the body of this swift fish to the man-made and screw-propelled submarine, but here the resemblance ends, since no revolving motion of one part on another is possible in the living animal. It has been stated that the Mackerel is able to twist its tail to such an extent as to be "very much after the manner of a screw in a steamship, and thus to drill the water," but this is, to say the least, very unlikely, and is in no way supported by actual observation. Most standard text-books on fishes contain the statement that they swim by means of lateral flexions of the hinder part of the body, aided by movements of the caudal fin. This is, in the main, correct, but recent investigations tend to show that the part played by the tail in producing a forward thrust is far less important than was previously supposed. In this connection the following experi- ment carried out by Mr. Breder is of some interest. Two Rudd (Scardinius) of equal size were selected, and, the tail of one of them having been carefully amputated, the two fishes were placed in a tank eight feet long, where they rested side by side at one of its ends. The investigator carefully approached this end and gave the glass a smart blow with his hand. Imme- diately the two Rudd scurried away to the opposite end of the tank, where they came to rest in similar positions. This experi- ment was carried out several times, and each time it was repeated both fish arrived at the opposite end at the same time, having traversed the intervening space side by side. The only noticeable difTerence between the two specimens was that the fish without a tail-fin "waggled" the hinder part of its body faster and through a wider arc. When the Mackerel wishes to move forward the first action which takes place is the contraction of the first few myomeres at the front end of the body on one side only, resulting in the throwing of the head sharply to one side. The successive segments then alternately contract and relax from the head towards the tail, and the curve or flexure of the body is, so to speak, passed backwards (Fig. 9). The effort culminates in bringing the tail to the axis of the head with a powerful sweep. Since, at the commencement of the stroke the pivotal point of the swing lies just where the backbone joins the skull, a com- paratively short swing of the head is all that is necessary to bring the tail into position for a long and strong sweep; as this is carried out the pivot must necessarily move backwards, until at the end of the stroke it lies nearer to the tail than to the head. The accompanying illustrations of a swimming fish FORM AND LOCOMOTION 23 give a good idea of the manner in which the body is undulated, and show how the flexure may be traced backwards from the head towards the tail (Fig. 9). The actual forward thrust is effected by the pressure of the fish's body against the surrounding water, aided to a certain extent by the tail-fin stroke, due to the action of the muscles. In order to understand the action of the body movements on the surrounding medium it will be convenient to study an elongated type offish such as the Eel (Anguilla), and compare its locomotor methods with those of the Mackerel. The move- ment is again initiated by the contraction of the first few myomeres on one side. The anterior part of the body is thus thrown into a curve, and this curve is passed backwards in a series of waves by the alternate contractions and expansions of the serial muscle segments. The movement is mechanically the same as that of a long rope held at one end and given a smart jerk with the hand at right angles to its axis. This results in a wave passing down the rope, the curves gradually de- creasing in size and eventually dying out because the initial action of the hand was the sole agent of propulsion ; in the living fish each successive muscle segment gives an added impetus to the wave, and, as soon as the first wave has started backwards, a second follows, but on the opposite side, and so on. Here the forward thrust is attained almost entirely by the pressure of the fish's body against the water contained in the spaces betv/een the curves. With this elongate form the fish gains much greater pressure areas from its sides than does the Mackerel, but, at the same time, it naturally loses the terminal effect of the tail-fin. Indeed, we find that in all eel-like fishes the caudal fin is either very much reduced or wanting altogether, a good example of the inevitable disappearance of a useless organ. In most Eels the anterior part of the body is cylindrical in cross-section, whereas the hinder part is distinctly compressed ; this feature has a mechanical advantage, since a blade-like structure which presents its surface more effectively to the water naturally provides a greater amount of thrust than a rounded one. The elongate Ribbon-fishes {Regalecus) , and other fishes with long bodies greatly flattened from side to side (Fig. 8d), undulate them into curves which are even more ample than those of the Eel, the extreme ribbon shape making this excessive bending comparatively easy to perform. It is of interest to note here that fishes with the body rounded can move over soHd surfaces out of water by applying the same 24 A HISTORY OF FISHES locomotor methods normally used in swimming, whereas those with ribbon-like bodies are unable to progress at all outside the liquid medium. While considering the question of the movements of fishes out of the water, it may be of interest to see why a species with the type of body similar to the Mackerel flops from side to side when taken from its native habitat. In contracting the body muscles on one side in the normal manner, the tail is brought smartly down with a sharp smack, and the accompanying reaction throws the fish upwards. Experiments conducted on living fishes have shown that they are quite unable to direct these movements, progressing indefinitely in any direction, and that only a favourable wind or a slope in the right direction enables them to find their way back to the water. If the locomotion of the Eel be regarded as one extreme type of body movement, that of the Trunk-fish (Ostracion) undoubtedly represents the opposite extreme, the Mackerel and other generalised fishes being intermediate between the two types. In the Trunk-fish (Figs. 5B; 42F), with its head and body enclosed in a hard and inflexible bony case, from which the fleshy tail with a large fan-like caudal fin at the extremity projects freely backwards, undulations of the body are clearly impossible. Normally, the dorsal and anal fins form the chief propefling agents, and the tail acts as a rudder, but where greater speed is required the fish lashes the tail vigorously from side to side, the movements being brought about by the alter- nate contraction of the muscles on either side of the fleshy part of the tail. A Trunk-fish swimming in this way may be likened to a small boat propelled by means of a single oar from the stern. The three types of body movements here described, and exemplified by the Eel, Mackerel, and Trunk-fish respectively, are not to be looked upon as other than purely arbitrarily chosen examples. Among fishes we find such a complete gradation from one extreme to the other that it is not easy to say where one begins and the other ends. The extremes are methods employed by comparatively slow-swimming forms, mostly living close to the shore, whereas those of the Mackerel and its kind are of the highest efficiency and pre-eminently suited for high speed. With the sole exception of fishes such as the Sea Horses, in which the locomotor emphasis is placed entirely on the fins, all existing forms fall somewhere within the series described above. . FORM AND LOCOMOTION 25 Turning to the second of the primary methods of locomotion, it may be noted that movements essentially the same as those of the body may be localised in one or more of the fins, and the Fig. 10. — FIN MOVEMENTS USED BY FISHES IN SWIMMING. A. Bow-fin {Amia calvd),X^^; b. Electric Eel (Electrophorus electncus),Xi ; c. File-fish {Monacanthus sp.), X J ; d. Ray (Raia sp.), X i. (After Breder.) same kind of series occurs, ranging from a serpentine, un- dulating motion like that of the Eel to a fan-like waggle which recalls the tail movements of the Trunk-fish. It has been 26 A HISTORY OF FISHES remarked that the caudal fin is operated primarily by the action of the muscles of the body and tail, but many fishes are capable of moving slowly forward by means of wave-like movements of the fin itself, the waves travelling at right angles to the longitudinal axis of the body. In most fishes the shape of the fins, and more especially that of the caudal, provides a very good index of speed and agility, the same type of fin occurring in quite unrelated groups of fishes whose swimming habits are similar. It is impossible to enter into the mechanical possibilities of the different shapes of caudal fin, but it may be laid down as a general rule that fishes with large tails, the hinder margins of which are square-cut (truncate) or rounded, are comparatively slow swimmers, and, although able to accomplish sudden short bursts of speed, they are incapable of swimming for long periods at a rapid velocity, as are those species provided with deeply forked or lunate tails (Fig. 33). Such fishes have the upper and lower lobes of the fin long and pointed, and the fleshy part of the tail, known as the caudal peduncle, is nearly always very narrow and not infrequently strengthened by one or two fleshy keels on either side as in the "Sword-fishes" (Fig. 6). The Bonito (Gymnosarda) , reckoned to be among the swiftest of all fishes, provides an excellent example of this type of caudal fin — crescent-shaped, without flesh, almost without scales, composed of bundles of rays, flexible, yet as hard as ivory (Fig. 4). Professor Goode writes: "A single sweep of this powerful oar doubtless suffices to propel the Bonito a hundred yards, for the polished surfaces of its body can offer little resistance to the water. I have seen a Common Dolphin swimming round and round a steamship, advancing at the rate of twelve knots an hour, the effort being hardly perceptible. . . . Who can calculate the speed of a Bonito?" The other median fins, the dorsal and anal, may also be used by certain fishes as swimming organs, especially in those forms whose bodies have radically departed from the stream- line shape. They may act in conjunction with the caudal fin or as a substitute for it. By appropriate use of the muscles controlling the fin-rays and their supports {cf. p. 58) a series of wave-like movements can be produced in the fin, similar to the wriggUng contortions of the body of the Eel. In the Bow-fin (Amia) of North America, for example, undulating movements of the long dorsal fin are often used to propel the body slowly forward (Fig. ioa), and the Electric Eel {Electro- FORM AND LOCOMOTION 27 phorus), in which the dorsal fin is wanting, employs the long anal fin in a similar way (Fig. iob). The File-fish or Leatherjacket (Monacanthus) has both dorsal and anal fins placed a little obliquely, and makes use of both simultaneously for forward progression (Fig. loc). Other fishes, such as the Globe-fishes or Puffers (Tetrodontidae) and Porcupine-fishes (Diodontidae) , move by flapping the short dorsal and anal fins in a fan-like manner. The Sea Horse (Hippocampus) , which characteristically swims in an upright position, glides slowly through the water by means of rapid wave-like movements of the dorsal fin, which has the appearance of a tiny propeller revolving in the middle of the fish's back (Fig. 5E). The related Pipe-fishes {Syngnathidae) swim in a similar way, but their bodies being elongate and more flexible they are able to make more rapid progress at times by lashing themselves into curves (Fig. 32E). Flat-fishes, when moving about on the sea floor, often make use of the long dorsal and anal fins which fringe the upper and lower edges of the body (Fig. 8b) to obtain a grip of the ground, and by undulating these fins are able to progress at a fair speed. Turning to the paired fins, the pelvics may be dismissed at once, as these merely assist the dorsal and anal fins in main- taining stability, and rarely, if ever, serve as organs of pro- pulsion. The pectorals, on the other hand, are often used partly, or in some cases almost exclusively, for locomotor purposes, particularly in those fishes of slow or moderate speed. In slow-moving fishes these fins are generally spatulate in shape, and may produce forward movements of the body by a simple synchronised flapping, as in some of the Wrasses (Labridae). In others, of which the File-fish (Monacanthus) and the Porcupine-fish (Diodon) are good examples, wave-like motion similar to that described in connection with the caudal fin, is employed. This type of motion is particularly well marked in the Rays (Raia) and their allies, in. which the pectoral fins are very much enlarged and constitute practically the sole organs of locomotion. It will be noticed, however, that the waves travel in a vertical plane instead of a horizontal one — up and down instead of from side to side (Fig. iod). In a few species, notably among the Damsel-fishes (Pomacentridae) , the pectoral fins seem to be operated after the manner of oars, being brought forward almost edgewise and pulled back broadside on. In fishes of high velocity the shape of the fins is generally long and falcate (i.e. sickle-shaped), and these are probably used mainly for changing course or for slowing down, 28 A HISTORY OF FISHES scarcely ever for propulsion (Fig. 4). In fishes, turning when in motion would appear to be effected largely by appropriate movements of the fins, and particularly of the pectorals, but body movements as well as jets of water from the gill-openings also play their part. Stops are nearly always made by using the pectorals in the manner of brakes, but some forms pull up by reversing their primary locomotor apparatus. Finally, there remains the third primary method of loco- motion, namely, by means of jets of water squirted from the gill-openings during respiration. Recent experiments have shown that these exhalations may play an important part in driving the body forward, but the effect varies with different fishes, being of considerable importance to some and of little or none to others. This method is probably always brought into play for high-speed travelling, and assists the muscular activities of the body. The jets reach their maximum strength between the flexures of the body when the fish is straight for- ward, and when they would clearly produce the best effect. When used in conjunction with movements of the pectoral fins, the motions of the fins are timed so that they do not get in the way of the jets. A particularly powerful jet is usually expelled when a fish commences any swimming movement, thus giving an added impetus to the initial muscular efforts of the body in getting under way. By holding down tightly the gill-opening on one side, and forcing all the water out of the opposite one, the fish may make use of this locomotor method to perform turning movements in any given direction. Unless a fish is actually resting on the bottom, it is by no means as easy as it would appear to maintain a stationary position in the water. Breathing cannot be suspended for a moment, and although this respiration may be comparatively slow as compared with that taking place when the fish is swimming, the exhalant jets of water are of sufficient strength to cause the body to move forward, and some sort of action is necessary to counteract their force. Observation of a fish resting in mid-water in an aquarium shows that the pectoral fins are in more or less constant motion, and it has been freely stated that they are employed in balancing the body. Experiments, consisting of removing one or both of these fins from a living fish, have shown that they play very little part in maintaining stability, and it seems that they are engaged in constantly backing water, to counteract the forward thrust engendered by the respiratory jets. FORM AND LOCOMOTION 29 The speed attained by fishes has always been the subject of much speculation, but, unfortunately, very little accurate data exists as to the relative speeds of different forms. The average rate of progress of the Salmon [Salmo) has been estimated at about seven miles per hour, and that of the Pike {Esox) as from eight to ten miles per hour, but these come nowhere near the speed attained by some of the large oceanic fishes. Remarking that the speed of a Bonito might be reckoned by the aid of the electrical contrivances by which the initial velocity of a pro- jectile is calculated, Professor Goode adds: "The Bonitos in our sounds to-day may have been passing Cape Colony or the Fig. II. — A FISH WHICH SWIMS UPSIDE DOWN. Cat-fish (Synodontis hatensoda), X |. Land of Fire the day before yesterday!" Of all fishes, the Sword-fish {Xiphias) and its aUies (Fig. 6) are perhaps the most rapid swimmers. A number of cases are on record in which they have struck ships, and Professor Owen, who was once called upon to testify in court as to the power of the Sword-fish, stated that it "strikes with the accumulated force of fifteen double-headed hammers; its velocity is equal to that of a swivel-shot, and is as dangerous in its eflfects as an artillery projectile." If the ship struck be a wooden one it is not un- common for the sword to be driven in with such force that it cannot be withdrawn, and the fish frees itself by breaking it oflf short! How great must be the speed of these fishes to produce such results ! Practically all fishes adopt a horizontal position when 30 A HISTORY OF FISHES swimming, but one or two species depart from this normal attitude. The vertical position of the Sea Horse [Hippocampus) has been already mentioned. The little Shrimp-fishes or Needle-fishes [Centriscidae) are curious creatures, with the long compressed body encased in a thin bony cuirass with a knife- like lower edge. One species found in the Indian Ocean lives in small shoals of about half a dozen individuals, and normally swims about in a vertical position with the long tube- like snout pointing upwards (Fig. 12). On occasions, how- ever, it has been observed to move in the normal hori- zontal attitude, and even vertically, but upside down ! A Cat-fish from the Nile and other African rivers (Syno- dontis batensoda) has adopted the remarkable habit of floating or swimming leis- urely at the surface of the water with the belly upwards^ an attitude taken up by no other fish unless it be sick or dead. This habit must have been well known to the ancient Egyptians, as it is frequently depicted in their sculptures and wall paintings in this anomalous position. Among the methods of locomotion other than swim- ming, leaping and burrowing may be considered briefly here, as they result from body rather than fin-movements. A fish may leap out of the water for one of several reasons : to escape from an enemy, to clear a weir or other obstacle, or from pure joie de vivre. The strength and agility displayed by the Salmon (Salmo) (Fig. 13c) in leaping falls in its journey to the spawning ground is well known, and it has been observed to make repeated efforts to clear an obstacle which was too high for it, and to fall back at last through sheer exhaustion. It is this habit which has given the Salmon its name, the Latin Salmo being from the same root as salire^ to leap. The Tarpon (Megalops), a favourite with Fig. 12. — A FISH WHICH SWIMS UPRIGHT. A small shoal of Shrimp-fishes (AeoHscus strigatus), X ^. (After Willey.) FORM AND LOCOMOTION 31 American sea-anglers, is another fish famed for its leaping powers, and its indulgence in this habit makes it necessary to employ some skill and perseverance in its capture (Fig. 13B). As Dr. Gill remarks: "Its frequent leaps into the air . . . seem to be mostly in sportive manifestation of its intense vitality, and not for food or entirely from fear." Opinions differ as to the height to which these fishes are able to jump, but it is generally agreed that seven or eight feet probably represents the Tarpon's Fig. 13. FISHES THAT LEAP. A. Devil-fish {Manta birostris), X about -V ; b. Tarpon (Megalops atlanticus), X about ^V ; c. Salmon {Salmo salar), X yV ; D- Grey Mullet (Miigil sp.) X yV- limit, while the Salmon is only able to better this by one or two feet. How are the jumps accomplished? Generally by the fish swimming rapidly upwards through the surface of the water into the air, giving a sharp flick with its tail as it leaves the liquid medium. All the active propulsion is provided by the muscular actions of the body while in the water, but the passing into the relatively less dense air accelerates the speed considerably and makes powerful leaps possible with a fairly slight muscular effort. Both Tarpon and Salmon hold the body in a curve while out of the water, and naturally fall to 32 A HISTORY OF FISHES the concave side. Others, hke the Grey Mullet (Mugil), keep the body rigidly in a straight line, so that the course in the air is determined solely by such external factors as the velocity of the wind, the angle of its direction to the fish, and so on. The giant Devil-fish {Manta), figuring in so many romances of tropical seas, is another fish which can leave the water on occasion and, by means of an awkward, wheeUng, edgewise leap, sail through the air to a height of more than five feet from the surface of the water. A full-grown specimen is somewhere in the neighbourhood of twenty feet long, and weighs more than 1000 lb., so that its sudden jump from the sea is an awe- inspiring sight. The noise made by its body as it returns to the water resembles the discharge of a cannon, being audible at a distance of several miles. The strength of this fish is prodigious, and when harpooned it will drag a boat through the water at great speed, so that it is sometimes necessary to cut the line at once to avoid disaster. The habit of burrowing is generally associated with the eel- like type of body, but some of the Wrasses {Labridae) and the Httle Mud Minnow {Umbra) of North America and Central Europe are adepts at this art. The latter is said to be perfectly at home in the mud, and one author claims that it can "pass through soft mud with as much ease as other fishes do through clear waters." The same writer states that "if suddenly dis- turbed, they generally dart oflf by swimming only, and bury themselves tail foremost in the mud." The method by means of which burrowing is effected is quite simple, the fish merely employing active swimming movements with the snout pointed into the sand or mud. This is continued until a sufficient portion of the body is covered to enable its surface to obtain an effective grip, when progress is more rapid. Bottom-living forms Hke the Rays and Flat-fishes, instead of actually bur- rowing in the sea floor, wriggle their flattened bodies and throw sand over the upper surface until they are completely covered. CHAPTER III RESPIRATION How fishes breathe. Structure of gills in Selachians, in Marsipobranchs, in Bony Fishes. Lophobranchs. Process of respiration in a typical fish, in Lampreys and Hag-fishes, in Rays, in Trunk-fishes. Does a fish drink? Gill-rakers and their uses. Rate of breathing. Fish that can live out of water. Accessory organs of respiration : breathing through the skin, intestinal respiration, labyrinthic organs, air-breathing sacs. Air-bladder and its functions. Origin and evolution of air- bladder. Paradoxical as it may appear, a constant supply of fresh air is as important to a fish as to ourselves, the air being required for its contained oxygen. Respiration may be defined as a physiological process resulting in the aeration of the blood, a process consisting of taking in oxygen and giving off impurities in the form of a gas known as carbon dioxide. This exchange of gases is essentially the same in a fish as in any higher vertebrate, the difference in the respiratory process being in the manner in which the life-giving substance is obtained. Whereas the land animals extract the oxygen from the atmospheric air by means of lungs, fishes make use of the oxygen contained in the air dissolved in the water by the use of special organs known as gills. The supreme importance of this air to the life of a fish may be demonstrated by placing it in a vessel containing water from which all air has been driven out by intense heating, when it will be speedily suffocated. Similarly, if a bowl of Gold-fish be covered over so that it is impossible for any air to reach the water, the fish will succumb as soon as they have used up the supply of oxygen contained therein. The exact amount of oxygen consumed by a fish in a given time varies greatly in different species, and is also dependent on such factors as the amount of muscular activity displayed, with its consequent greater or lesser consumption of energy. On the whole, it seems to be comparatively small, and Dr. Gunther has estimated that a man uses up 50,000 times more oxygen than a Tench [Tinea), a fish averaging about fifteen inches in length. A careful study of the nature and origin of gills and lungs c 33 34 A HISTORY OF FISHES leaves no doubt that gill-breathing represents the more primi- tive method of respiration, and must have gradually given place to lung-breathing as the ancestors of modern terrestrial vertebrates left the water for the dry land. All the higher vertebrates, whether amphibians, reptiles, birds or mammals, provide us with indubitable proof of their fish ancestry by the possession of gill-like structures of one kind or another at some stage of their lives. In a human embryo, say of about three weeks, the sides of the throat are provided with four pairs of clefts, which not only correspond in position to the gill-slits of a fish, but their supporting skeleton and associated blood-vessels provide further resemblances. This fish-like apparatus is, of course, never used for breathing, and as development proceeds it becomes modified out of all recog- nition. A proper understanding of the functions of the gills is im- possible without some idea of their anatomy, and at the risk of introducing what may appear to be needless technicalities, it will be necessary to give a brief account of the salient features of their structure. The principles of respiration are essentially the same in all fishes, but a marked difference in the type of gills is found in the three main classes, and it will be con- venient to describe first of all the conditions found in the Selachians, and then to compare them with those existing in the Cyclos tomes and Bony Fishes. All breathing organs, no matter what their form, are closely associated with the upper part of the food channel or alimentary canal {cf. p. i68). Such a connection between the nutritive and respiratory apparatus is highly characteristic of backboned animals, and persists even in man himself, but in fishes the two remain more or less intimately connected throughout life, whereas in the higher animals the association is generally a temporary embryonic phase. In a typical shark the side walls of the pharynx, that is to say, of that portion of the alimentary canal at the back of the mouth immediately in front of the commencement of the narrow gullet, are perforated by a series of narrow openings (Fig. 15A, ph.). Each of these pharyngeal openings leads into a kind of flattened pouch, which in turn communicates with the exterior by a comparatively narrow slit, the external gill-cleft, lying on the side of the head between the eye and the pectoral fin (Fig. 14A, g.c.l.). As a rule these clefts are not very long, but in the huge Basking Shark {Cetorhinus) they extend from the RESPIRATION 35 upper to the lower surfaces of the body. They are normally five in number (excluding a small circular opening known as the spiracle, which may lie in front of the first gill-cleft), but in the Frilled Shark {Chlamydoselachus) (Fig. 32c), the Comb- toothed or Cow Sharks (Hexanchidae) , and one of the Saw Sharks (Pliotrema), there may be as many as six or seven. In the Frilled Shark each of the partitions between the successive ^"fi Fig. 14. — EXTERNAL GILL-OPENINGS. A. Spotted Dog-fish (Scyltorhinus sp.),X^; b.i, b.2. Thornback Ray (Rata clavata),X\\ c. Rabbit-fish {Chimaera monstrosa),X\ \ D. Trout {Salmo trutta), X i. g.c, external gill-clefts of Selachians ; g.o., external gill-opening of Chimaeras and Bony Fishes ; w., nostril ; sp., spiracle. clefts is produced backwards as a curious fold of skin covering the cleft immediately behind (Fig. 32c). The partitions, or interbranchial septa, between the separate gill-pouches are fairly thick, and are reinforced by sheets of tough fibre-like substance. Further support is provided by a series of bars of cartilage known as the gill-arches, which lie at the inner edges of the septa and between the pharyngeal openings (Fig. 15 a, g.a.). Each arch has the form of a half- hoop, and is broken up into several segments movably con- 36 A HISTORY OF FISHES nected with one another, the lowest of which is nearly always joined by a coupling piece with its fellow of the opposite side {cf. Fig. 46A). In this way, the inside of the pharynx is sup- ported by a series of encircling jointed girders, the outer convex faces of which are fringed by a number of slender rods of cartilage, which project outwards into the septa and help to strengthen them. All the parts of the gill-arches are provided Fig. 15. GILLS OF SHARK AND BONY FISH COMPARED. A. Dissection of head of Spotted Dog-fish {Scyliorhinus sp.) seen from beloWj X ^ ; B. The same of Salmon (Salmo salar), x -}. g.a., gill-arch ; g.c, external gill-cleft ; g.f. gill-filaments ; g.o., external gill- opening ; ph., pharyngeal opening ; sp., spiracle. with their special muscles, which by their appropriate ex- pansion or contraction bring the hoops closer together or move them wider apart, and thus diminish or enlarge the size of the intervening openings. The opposing walls of each gill-pouch bear a number of closely pleated folds of skin, the branchial lamellae or gill- filaments ig.f'), whose free edges project into the cavity of the pouch. These are richly supplied with fine blood-vessels, and present the appearance of a series of thin red straps or plates, RESPIRATION 37 a feature from which the earlier name of Elasmobranchs (strap gills) given to the class of Selachians was derived. Reference to the accompanying diagram will show how the gills are arranged in a typical Selachian (Fig. 15A), and it will be observed that the anterior wall of the first pouch has its row of filaments, but the posterior wall of the last pouch is not provided with these structures. Mention may be made here of the organ known as the spiracle, which is, in reality, nothing more than the vestige of a gill-cleft, and, indeed, in the early embryonic stages of a Shark this differs little from the clefts that lie behind it, although it subsequently degenerates. Even in the adult, however, the spiracle frequently retains a number of branchial lamellae, and probably aids in aerating the blood going to the eye and brain. It varies greatly in size in the different families, being small or absent in some of the larger Sharks, and com- paratively large in the Torpedoes, Rays, and Sting Rays, where it has acquired a special function to be described later on (Figs. 14A, b; 15A, sp.). The class of Marsipobranchs (Lampreys and Hag-fishes), a name meaning "purse gills," exhibits a type of respiratory organ which, although in some respects more primitive than that of the Selachians, presents several special and peculiar features. The respiratory lamellae are lodged in a series of muscular pouches, well separated from each other, differing in number in the various species. In the Lamprey {Petromyzon) there are seven on either side, each of which opens directly to the exterior by a small rounded opening on the outside of the head, and communicates by a similar orifice internally, not directly with the pharynx as in the Selachians, but with a special canal; this ends blindly behind, but in front opens into the mouth. There are, thus, seven external openings and one internal opening. In the Hag-fish {Myxine) each pouch opens directly into the pharynx, but on the outside it is drawn out into a tubular canal running posteriorly; further back all the canals unite and open together by a single external aperture (Fig. 16). The gill-pouches are large in the Lamprey, and are supported by an elaborate cartilaginous structure, the branchial basket, which in the Hag-fish is greatly reduced. In many respects this basket presents a superficial resemblance to the gill-arches of the Selachians, but lies outside the gill-pouches instead of between them and the pharynx. The general appearance of the gills in a typical Bony Fish 38 A HISTORY OF FISHES such as the Salmon [Salmo salar) must be famiUar to everybody, and may be readily seen by lifting up the bony plate lying on either side of the head behind the eye. How do these gills differ from those already described? In the first place, although A^sa/ CccSe. Sra-in. Gill -openi/iq Fig. l6. — GILLS OF MARSIPOBRANCHS. Dissection of anterior part of Hag-fish (Myxine glutinosa), X i . the same internal pharyngeal openings (ph.) are present, these do not open separately to the exterior, but into a common chamber, the branchial chamber, with a single external aperture behind (Fig. 15B). The outer wall of the chamber is provided by the movable flap already mentioned, which is 7. — CROSS-SECTIONS OF GILL-ARCHES IN DIFFERENT FISHES. A. Typical Selachian ; b. Chimaera ; c. Sturgeon (Acipenser) ; d, E. Bony Fishes. Gill-arch (dots) ; interbranchial septum (white) ; gill-filaments (cross lines). (After Boas.) known technically as the operculum or gill-cover (Fig. 140), and is supported by a series of broad, flat scale-like bones, with or without the addition of a number of slender bony rods below (Fig. 65) . The hinder and lower edges of the operculum are nearly always free, so that the external opening is com- paratively spacious, but sometimes these margins become more RESPIRATION 39 or less joined to the body of the fish and the outer opening is correspondingly reduced to a narrow slit or even to a minute upwardly directed pore. The same hoop-like gill-arches support the walls of the pharynx between the internal openings as in the Selachians, but are here composed of bone instead of cartilage. The more or less extensive interbranchial septa of the Shark have been reduced to minute proportions, and the deUcate red filaments form a double row of processes attached by their bases to the convex outer edge of each gill-arch (Figs. 17, 18). The half gill formed by the filaments on the anterior wall of the first cleft in the Selachians has either disappeared GtU-rakcrs GUI -arch. Fig. 18. — GILL-RAKERS. A Gill-arch of Allis Shad {Alosa alosa) ; b. The same of Twaite Shad {Alosa finta) ; c.The sanneofFerch (Per cafluviatilis) ; d. Isolated gill-rakers of Basking Shark {Cetorhinus maximus). All about f . or is represented by a mere rudiment (pseudobranch) , and the fifth (last) branchial arch is gill-less (Fig. 15B). The spiracle is almost invariably wanting, at least in the adult fish. Certain of the Selachians, namely the Chimaeras and their alUes (Holocephali), present a type of gill arrangement roughly midway between that of a Shark on the one hand and a Bony Fish on the other. The gills lie in a common branchial chamber, bordered on the outside by a skinny flap, foreshadowing the operculum of the Bony Fishes, and opening to the exterior by a single sUt-like aperture (Fig. 140). The interbranchial septa are somewhat shorter, so that the filaments project a httle beyond their outer margins. Passing to some of the more primitive Bony Fishes {e.g. Sturgeons), the septa become pro- 40 A HISTORY OF FISHES gressively shorter, until the condition described in the Salmon is finally attained (Fig. 17). The type of gill structure described here is that found in almost all Bony Fishes, but in the Sea Horses and Pipe-fishes {Syngnathidae), sometimes spoken of as Lophobranchs (tuft gills), the filaments are reduced to small rosette-Hke tufts attached to quite rudimentary arches. When a fish breathes, the initial movement consists of an expansion of the hoop-like gill-arches, with a consequent en- largement of the cavity of the pharynx; at the same time the mouth is opened a little, and a stream of water is drawn into the pharynx, the external gill-opening or openings being kept tightly closed. This is known as the inspiratory phase. It will be noticed that as a general rule the fish does not use the nostrils for breathing purposes, and, although water is inhaled through these apertures in a few forms hke the Chimaeras and Lung-fishes, they usually serve only as organs of smell {cf. p. 182). The expiratory phase, which follows immediately, consists in closing the mouth tight, and at the same time contracting the pharynx, thus driving the water outwards over the gills and through the external openings. The actual exchange of gases takes place as the v/ater passes over the gills. The walls of the fine blood-vessels with which the deHcate filaments are suppHed are exceedingly thin, so that the blood circulating through them is separated from the water by nothing but an infinitesi- mally fragile and permeable membrane. Fish blood, just hke our own, consists of a fluid substance, the plasma, in which float multitudes of red and white corpuscles. The red cor- puscles, forming flat, circular discs, contain that remarkable substance known as haemoglobin, which has the power of taking up oxygen under certain conditions and, later, of giving up this oxygen to the body cells, receiving in exchange their waste products in the form of carbon dioxide. Most fishes breathe in the manner described, but there are sonie which, owing to their pecuHar mode of life, have been obliged to modify this process in accordance with their change in habits.^ The Lamprey {Petromyzon) , for example, has adopted a parasitic habit, and spends a good deal of its time attached to other fishes by means of its sucker-like mouth. It is quite obvious that while in this position it would be impossible to inhale water through the mouth without losing its hold, and, instead, water is often taken into and expelled from the branchial sacs by their external openings, through the alternate expansion RESPIRATION 41 and contraction of their muscular walls. The related Hag- fish (Myxine) possesses still more singular habits, and bores right into the fishes it attacks. Under these conditions the current of water is inhaled through the single nostril situated on top of the head, and after entering the pharynx passes out through the gill-sacs. It has already been pointed out that a* Skate {Raia) is essen- tially adapted for a life on the sea floor, and it is of interest to note that the method of breathing has also been modified for this end. While swimming or crawling about it is able to breathe in the normal manner, but when resting on the bottom there is a grave danger of taking in sand with the stream of water and thus clogging up the delicate gill-filaments. In all the members of this group of Selachians the mouth and external gill-openings are on the under side of the head, but the spiracle remains on the upper surface, and is represented by a com- paratively large opening situated immediately behind the eye and pro\ided with a movable valve (Fig. 14, b.i). To avoid the danger of introducing foreign particles into the gills, the Skate inhales water by way of the spiracles, expelling it through the gill-openings in the usual manner. The respiratory modifications found in fishes normally inhabiting rapid streams or mountain torrents, where they are in the habit of fixing themselves to stones and other objects to avoid being swept away by the current of water, are dealt with in a subsequent chapter {cf. p. 239). The Trunk-fishes {Ostraciontidae), in which the head and body form a firm bony box, are obliged to keep up the flow of water over the gills by a series of rapid panting movements, as many as 180 per minute having been counted in a resting fish. Here the pectoral fins are used to assist respiration, and by their constant motion fan a current of water through the gill-openings. Professor Goode writes: "when taken from the water one of these fishes will live for two or three hours, all the time solemnly fanning its gills, and when restored to its native element seems none the worse for its experience, except that, on account of the air absorbed, it cannot at once sink to the bottom." The famihar phrase "to drink like a fish" is based on a complete misconception; it is assumed that the constant and regular opening and shutting of the mouth is a proof that the fish is drinking, whereas, as has been explained, this is really nothing more than an outward sign of the act of breathing. It is very doubtful whether the fish drinks at all, and, in any 42 A HISTORY OF FISHES case, during respiration the gullet is so tightly constricted behind the last pair of pharyngeal clefts that little if any water is able to find its way into the stomach. This closing of the gullet is brought about by the action of special muscles en- circUng the throat, which function in exactly the same way as does a double string running round the neck of a bag. The pres- sure of any food agairist the closed gullet, however, causes these muscles to relax somewhat, and the solid nourishment is pressed down into the stomach without the entrance of any water. In those fishes whose normal food consists of more or less minute creatures swimming about in the water there is ob- viously a danger of some of these escaping by way of the pharyngeal openings and perhaps clogging or injuring in some way the dehcate filaments. To lessen this danger special structures known as gill-rakers have been evolved, taking the form of a double row of stiff appendages on the inner margin of each hoop-like gill-arch (Fig. i8). These, by projecting across the pharyngeal openings, serve to strain the water which is to bathe the gills, and to prevent any solid particles from passing over with it. Generally the front row of rakers on each arch interlocks with the hinder row on the adjoining arch, and the two together form an eflTective sieve. In the Pike {Esox), feeding almost entirely on other fishes, the gill-rakers are represented merely by bony knobs, which may serve to block the passage of larger food particles. In the fishes of the Herring kind [Clupeidae), on the other hand, whose food consists of minute shrimp-like creatures, the rakers are very numerous and take the form of long, slender, and close-set bristles. In some of the members of this tribe the filtering mechanism is even more perfect, each primary gill-raker giving off secondary and tertiary branches, the whole apparatus having the appear- ance of the finest gauze. The form and number of the gill- rakers may diflfer considerably even in two closely related species, but this diflference may generally be correlated with a difference in the normal diet. The two species of Shad (Alosa) found in our own seas and rivers provide an excellent example of this, the AUis Shad {A. alosa), having about eighty rakers on the lower Umb of each arch in the adult fish, whereas the Twaite Shad [A.jinta) has only thirty (Figs. i8a, b). The AUis Shad feeds largely on small crustaceans, although it takes a certain number of larval fishes. The Twaite Shad does not eat the crustaceans to nearly, the same extent, but is much more destructive to the young fry of other fishes. RESPIRATION 43 As a general rule, gill-rakers are wanting in the Sharks, most of which feed on other fishes, but the huge Basking Shark {Cetorhinus) and Whale Shark (Rhineodon) are both provided with many close-set, flattened and tapering gill-rakers, each perhaps four or five inches long (Fig. 18). In appearance they recall the baleen plates of the Whalebone Whales, which have exactly the same function, namely to act as a filter to strain off the most minute forms of animal life. When feeding, the Basking Shark merely opens its mouth and takes in a mass of water containing myriads of the minute crustaceans forming its usual food. The water rushes out over the gills, and the animals are left sticking to the inner walls of the throat and to the filtering mechanism, where they can be conveniently swallowed. The rate of breathing seems to vary greatly in different fishes, ranging from about twelve to fifteen respirations per minute in the Wrasse (Labrus) and the Rockling [Motella) to as many as 150 in the Minnow {Phoxinus) and Stickleback (Gasterosteus) . Professor Bashford Dean noticed that in the Australian Lung- fish [Epiceratodus) the normal rate was about twelve per minute on a cool day, but rose to as many as thirty-one as the water warmed up. If the water is at all deficient in oxygen the rate of breathing is naturally accelerated and the fish appears to "pant" or respire hurriedly, while other factors which lead to an increased rate are excessive activity or the stress of some emotion such as greed or fear. All the breathing organs so far described have been what are known as internal gills, but in certain fishes external organs are developed similar to those found in larval newts and frogs. In the young Selachians they take the form of long, delicate filaments, protruding for some distance through the external gill-clefts. These seem to aid in some way the breathing of the embryo while enclosed in its egg-case, for they completely disappear when it is hatched {cf. p. 332). Similar structures are developed in the young of certain Bony Fishes, but are soon discarded in favour of the more adequately protected internal gills. In the young Bichir [Polypterus) there is a leaf-like external gill projecting backwards from each side of the head above the ordinary gill-opening. Here they are retained for a somewhat longer period, but gradually become reduced in size as growth proceeds (Fig. 19). The vast majority of fishes are quite incapable of living for any length of time out of the water, and, as a general rule, those with wide external gill-openings expire more rapidly than 44 A HISTORY OF FISHES those in which the apertures are reduced. This is due to the fact that as the filaments dry up they tend to stick together and are thus mechanically prevented from functioning. The im- portance of moisture in enabling a fish to cling to life is em- phasised by the length of time some fishes will remain alive if packed in damp grass or other vegetation. The author has more than once received a parcel containing Roach or Bream Fig. 19. EXTERNAL GILLS. A.I. Young South American Lung-fish (LepidGsircn paradoxa), 30 days after hatching, X 3 ; A.2. The same 40 days after hatching, X 2 ; B. Young Bichir {Polypterus endlicheri), X i^. thus packed, which have come to life successfully several hours after capture when tipped into a bowl of water. Fresh-water Eels (Anguilla) are able to survive for a considerable time away from their native element, and to wriggle across damp grassy meadows at night in order to get from one piece of water to another. It had always been supposed that these land excursions were made possible by the small size of the external gill- openings, but a French investigator has shown that Eels from which the gill-covers have been cut off were able to live out i RESPIRATION 45 of water just as successfully as normal specimens. He suggested that breathing was carried on through the skin, a statement which is not nearly so improbable as it sounds. Since respiration is nothing more than an exchange of gases by the blood, there is no reason why these gases should not penetrate through thin membranes like the skin, and reach the blood contained in the fine vessels situated therein. The remarkable Httle Mud- skipper [Periophthalmus) of tropical countries spends a great part of its time walking or skipping about on the mud-flats of mangrove swamps at low tide in search of food, and is also fond of climbing on to the mangrove roots themselves or of basking in the sun perched on a stone in a pool (Fig. 34f). While out of the water the large gill-chambers are kept filled with air, and the tail is, more often than not, left hanging down in the water, serving as an additional organ of respiration. These little fishes have become so accustomed to a life out of the water that they are said to be unable to live in what should be their native element for any length of time. Another remarkable method of breathing is adopted by some of the Loaches {Cobitidae) and the Mailed Cat-fishes {Loricariidae) , which use the intestine for this purpose at times. The Giant Loach {Cobitis) of Europe, known in Germany as the "Wetter- fisch" (weather fish) on account of its susceptibility to atmo- spheric changes, has been specially studied, and the process of breathing found to take place in the following manner. The fish rises to the surface, and by thrusting its mouth above the water, swallows a certain amount of air, which is passed down into the intestine. There is a bulge in the intestine just behind the stomach which serves as a reservoir; the fine blood-vessels lining the walls of this chamber extract the oxygen from the air, and the latter is finally voided by the vent. It is well known that when the air dissolved in the water fails a fish is obliged to ascend to the surface to avoid suffocation. The habit of seeking oxygen by swallowing bubbles of air is found in many different kinds of Bony Fishes, but in certain forms living in shallow ponds and streams which dry up periodically, or in pools rendered foul by decaying vegetation, this gulping of air becomes a necessity if the fishes are to survive at all. As a result, it is found that the intensification of the air- breathing habit over a very long period of time has led to the development of special accessory breathing organs in addition to the gills, thus enabling the fishes to survive for a compara- tively long time out of water. These organs take the form of 46 A HISTORY OF FISHES reservoirs for the storage of air, which are outgrowths either from the pharynx itself or from the branchial chamber, and contain certain special structures richly supplied with blood- vessels for the aeration of the blood. The so-called Labyrinthic Fishes (Anabantoidea) of the fresh waters of tropical Asia and Africa derive their name from the possession of a labyrinth-like accessory breathing organ on either side of the head. This group of fishes includes a number of species familiar in aquaria, such as the Climbing Perch {Anabas)y Gourami {Osphronemus) , Paradise-fish (Macropodus) , Fighting-fish {Bettaj, and so on, but it will only be necessary to mention the accessory breathing organs of the first of these. The CUmbing Perch {Anabas) was first made known in a memoir printed in 1797 by one Daldorf, a heutenant in the service of the Danish East India Company at Tranquebar. The fish derives its name from a legend current in the East that it climbs palm trees and sucks their juice, and Daldorf stated that he had taken one in a sUt in the bark of a palm which grew near a pond. The recent researches of an Indian naturalist. Dr. Das, have shown that, although the stories of the Climbing Perch being found in trees are quite well founded, the explanation of the facts is an erroneous one. The fish is in the habit of migrating from pond to pond at night, and after a shower of rain it comes out of the water and invades gardens in search of earthworms. During its overland travels it is not infrequently seized by crows or kites, and deposited high up in the forks of branches of the trees to be devoured at leisure. Hence the origin of the story of its tree-climbing activities ! The method of progression adopted on land is of interest, the gill- covers as well as the fins assisting in locomotion. The gill-covers are alternately spread out and fixed firmly to the ground by the sharp spines with which they are armed, while a vigorous push is given by the pectoral fins and the tail. When in the water, the Climbing Perch frequently comes to the surface to breathe air, and so vital has this method of respiration become to the fish that it will suffocate even in water saturated with oxygen if deprived of access to atmospheric air. The ease with which these fishes are able to survive out of water is taken advantage of by the natives of India and the Malay Peninsula, who carry them about alive for days on end in moistened clay pots, thus ensuring a regular supply of fresh fish. The jars must, of course, be kept tightly covered, or the intended meal will climb out and walk away! i RESPIRATION 47 The air inhaled is taken into two chambers situated one on each side above the gills, forming outgrowths from the ordinary branchial chambers. Each contains a more or less rosette-like structure, made up of a number of concentrically arranged, shell-like plates with wavy edges, all richly supplied with fine blood-vessels. Each air reservoir is in communication, not only with the branchial chamber, but also with the pharynx, the entrance from the throat being controlled by a special valve. Fig. 20. — ACCESSORY BREATHING ORGANS. A. Dissection of head of Climbing Perch (Anabas scandens) ; b. The same of an Indian Cat-fish (Saccobranchus fossilis) ; c. The same of an African Cat-fish (Clarias lazera). All about natural size. Air enters by this aperture and passes out through the external gill-opening (Fig. 20a). The related Snake-heads (Ophiocephalus) , long, cylindrical fishes with slightly flattened and somewhat serpent-like heads (Fig. 2 1 a), inhabit rivers and ponds as well as stagnant pools in the marshes. The larger species grow to a length of three or four feet. Their habit of "walking" over land by the aid of rowing movements of the pectoral fins is well known, and Snake-heads are frequently exhibited as curiosities by Indian jugglers. They are extremely tenacious of life, and are carried 48 A HISTORY OF FISHES alive by the Chinese to San Francisco and to Hawaii, where they are now naturaHsed and known as "China-fishes." They are able to survive prolonged drought, burying themselves in the mud and remaining in a torpid state during hot, dry weather. The various species differ in the extent to which they have developed the air-breathing habit, as well as in their power of living out of water. The accessory respiratory organs are of a much simpler character than those of the CUmbing Perch, consisting of a pair of simple cavities Hned with a thickened and puckered membrane supplied with blood-vessels, and do not contain any special structures. These lung-like reservoirs Fig. 21. FISHES THAT CAN LIVE OUT OF WATER. A. Snake-head (Ophiocephalus striatus), X ^ ; b. Cuchia (Amphipnous cuchia), X }. are not developments of the branchial chamber, but are pouches of the pharynx. Other fishes provided with accessory breathing organs are certain Cat-fishes of the rivers and swamps of Africa and Asia, and the famous Cuchia (Amphipnous) of India and Burma, a member of the group of Symbranchoids (Fig. 2ib). In one of these Cat-fishes (Clarias) the organs take the form of elaborate tree-like structures, growing from the upper ends of the gill- arches and contained in a pair of air-chambers situated above the gills (Fig. 20c). Another form {Saccobranchus) has the organs of a simpler nature, but the air-chambers bear a marked resemblance to lungs, extending backwards as far as the tail as long tubular sacs growing out from the branchial cavity, RESPIRATION 49 and situated close to the backbone (Fig. 2ob). The Cuchia, which grows to a length of about two feet, bears a superficial resemblance to some of the Eels, to which, however, it is not at all related. Like the other fishes just mentioned, it is an inhabitant of the fresh and brackish waters of India and Burma, and spends much of its time in the grass on the banks of ponds like a snake. The air-breathing organs consist of a pair of sacs growing out from the pharynx above the gills. This curious fish seems to have lost practically all its power of aquatic respiration, for when in the water it is forced to come to the surface at frequent intervals to gulp air, and the true gills are very much reduced, being represented by a few rudimentary filaments attached to the second of the three remaining gill-arches. The accessory respiratory organs just described are not the only structures used for air-breathing among fishes, and before concluding this chapter some more lung-like organs must be considered. The air-bladder, or, as it is sometimes called, the swim-bladder, must be a familiar object to those who have had occasion to examine the inside of a fish. Situated within the body cavity, and immediately below the backbone, it generally has the appearance of a long, cylindrical bag with glistening silvery walls. It is very variable both in size and form in different fishes, and may be present in one species and entirely absent in a closely related form. That it normally contains air or gas of some kind may be readily demonstrated by punc- turing it with a needle, when the walls promptly collapse. There is no other single organ in any group of vertebrates which performs such a variety of functions as does the air- bladder of Bony Fishes. In the majority it serves as a hydro- static organ or float, enabling its possessor to accommodate itself to the varying pressure encountered at different depths [cf. p. 174), in others it is an organ for the production of sound {cf. p. 156), and in others, again, it is more or less connected with the sense of hearing {cf. p. 193). For the present it must suffice to consider its relation to air-breathing and its con- nection with the lungs of higher vertebrates. Like the lungs of a man, the air-bladder is intimately asso- ciated with the alimentary canal, and a study of the develop- ment of this organ shows that it begins as a minute pouch budded off from the gullet. This gets larger and larger, until it is finally separated off from the gullet, remaining connected only by a narrow tube known as the pneumatic duct. In some fishes this duct remains open throughout life, but in others it 50 A HISTORY OF FISHES closes up or disappears altogether. It nearly always opens into the pharynx by a small aperture in the roof, but in the Lung- fishes {Dipneusti) and Bichirs (Polypterus) y in which the air- bladder is a true breathing organ, the opening takes the form of a small sHt, the glottis, with well-defined lips, situated in the STURGEON AND MANY OTHER bony-fishcs. GAR -PIKE ANQ BOW-FIN. CHARACIN. AUSTRALIAN LUNG -FISH. BICHIR. AFRICAN if S.AM£/?lC^M LUNO-FISH. REPTILES, BIRDS, MAMMALS. Fig. 22. AIR-BLADDER AND LUNG. A series of diagrams showing the relation of the air-bladder or lung to the gullet in different fishes, as seen in cross-section (left-hand column) and from the side. (After Dean.) floor of the gullet (Fig. 22). The lungs of higher vertebrates arise in exactly the same way, as an outgrowth from the gullet, and the glottis occupies the same position as in the Lung-fishes. Furthermore, whereas in the generaHty of fishes the air-bladder is a simple sac filled with a mixture of gases, in the Bow-fin {Amia)^SLnd Gar Pike (Lepidosteus) , two other air-breathing RESPIRATION 51 forms, and to a more marked extent in the Lung-fishes, the inner walls are richly supplied with blood-vessels, and the area of their surface is greatly increased by being produced into recesses or alveoli, each of which is further subdivided much in the same way as in a true lung (Fig. 22). Another important point is the fact that, whereas in nearly all fishes the bladder is a single structure, in the Bow-fin, Gar Pike, Bichir and Lung- fishes it tends to become divided, giving rise to a structure resembhng the paired lungs of higher vertebrates. In the Bichirs, for example, it is divided into two unequal parts, a long right-hand portion and a much shorter left-hand one; the two unite in front and open by a single aperture in the floor of the gullet. The air-bladder of the African (Protopterus) and South American [Lepidosiren) Lung-fishes is divided into two except for a small portion in front, and thus has the form of a pair of lungs. Finally, it may be noticed that the arrange- ment of the vessels taking the blood to and from the bladder in the Bichirs and Lung-fishes is essentially similar to that of the vessels connected with the lungs in Amphibians and reptiles. There is good reason for supposing that the air-bladder originally had a purely respiratory function, and that its use as a hydrostatic organ is secondary. The essential similarity of the mode of development of the air-bladder and the lungs is a striking argument in favour of this view; and further, the bladder is connected with the gullet by a pneumatic duct during the early embryonic or larval stages of most, if not all, Bony Fishes, and this connection is retained throughout life in nearly all the more primitive forms. The air-breathing fishes of to-day, such as the Lung-fishes, Bichirs, Bow-fins and Gar Pikes, are without exception survivors of very ancient groups which flourished at a very early stage of geological time, whereas, all those in which the air-bladder serves as a hydro- static organ are of comparatively modern origin. We may note here that no trace of an air-bladder is found in any Selachian, nor is one developed in the Lampreys and their allies. The statement so often made that the lungs of higher animals have developed from the air-bladder of fishes is quite untrue, and this idea rests on a misconception similar to that which leads many people to speak of man having descended from monkeys. It would be more correct to speak of the air- bladder as a modified and degenerate lung, but actually both bladder and lung seem to have been derived from some sort of 52 A HISTORY OF FISHES primitive respiratory sac, which itself originally arose as a pouch growing out from the gullet. What, then, were the causes that led to the origin of this air-breathing sac? To answer this question it is necessary to go back to the early Silurian and Devonian epochs when the evolution of this organ probably took place. It may be assumed that the earhest fishes swarming in the sea close to the shore found a sufficiency of oxygen dissolved in the water to make air-breathing unnecessary, but in course of time stress of competition must have led many of them to ascend the rivers just as certain Sharks and Rays do to-day, and to become permanent inhabitants of fresh water. In the course of further penetration a number of species would find their way into the smaller streams, ponds, marshes, and so on. Here the water might be expected to be thick with sediment, or, owing to the heat or to the decay of vegetation, poorly aerated, thus making gill-breathing a matter of difficulty. These conditions would necessitate the fishes coming to the surface at intervals to gulp air, just as the Bichirs and Lung- fishes do to-day when the water in which they are hving becomes foul. At first, the air-breathing would perhaps be performed by the walls of the gullet, but as the habit persisted the process would be improved by the development of a more or less definite pouch or reservoir. This would gradually assume the appearance of a lung, and eventually divide down the middle to form the double structure characteristic of the higher vertebrates. In the early fishes which took up their abode in the rivers and marshes the air-bladder probably served merely as an accessory respiratory organ, the bulk of the work of breathing falling on the gills. Conditions being favourable to their progress, these fishes must have multiplied rapidly. Later on, however, owing to a number of factors, among which the arrival of fish-eating reptiles may be men- tioned, a large number of them appear to have been driven back to the sea, where, with an adequate supply of oxygen for gill-breathing, the air-bladder gradually lost its respiratory function to a greater or lesser extent, and either became modified for totally different ends or disappeared altogether. That this invasion of fresh water by the early fishes was a very important step in the evolution of higher vertebrates there can be no doubt, for once the habit of breathing air had become established, the abundance of food and the complete absence of enemies would be an inducement for some species RESPIRATION 53 to venture further on to dry land. Here the fins would be gradually converted into limbs for terrestrial locomotion, and the first lung and gill-breathing amphibians would come into being, to be followed in their turn by the reptiles breathing by means of lungs alone. CHAPTER IV FINS Different kinds of fins. Their origin and evolution. Structure of fins: dorsal and anal, pectorals and pelvics, caudal. Types of tail. Develop- ment of tail. Modifications of dorsal and anal in different fishes, of caudal, pectoral and pelvic fins. The nature and functions of the fins have been indicated in an earUer chapter [cf. p. 13). Briefly, they are of two kinds: (i) median or unpaired, sometimes described as vertical fins; and (2) paired fins. The median fins include a dorsal in the middle line of the back, an anal along the belly behind the vent, and a caudal at the hinder end of the fish. The paired fins are of two kinds only, pectorals and pelvics^ corresponding respec- tively to the fore and hind limbs of land vertebrates (Fig. 7). The pectorals, sometimes referred to as the breast-fins, are always placed close behind the head, but the position of the pelvics varies in different groups of fishes. Before considering the structure of the fins, it will be as well to discuss the manner in which they have arisen in course of evolution. It is generally agreed that the earliest fishes possessed no true fins, but swam entirely by undulations of the body, and the fins were probably first developed as stabilising keels to counteract the tendency of the body to roll over sideways when in motion. In this connection some ingenious experiments carried out by Professor Cunningham may be of interest. Having coated an ordinary penholder thinly and evenly with wax, he held this firmly at one end, and, keeping it in a hori- zontal position, moved it rapidly to and fro in a vessel con- taining warm water. He found that very soon a vertical ridge made its appearance above and below the pen in exactly the same position as that occupied by the dorsal and anal fins of a fish, and that this gradually increased in height until, after five minutes' movement, an upper and lower ridge about half an inch high had developed. Experiments made by a French in- vestigator on plastic bodies {cf. p. 13) produced similar results. During the embryonic or larval stages of almost any fish the development of the ordinary fins is preceded by a stage at 54 FINS 55 which there is a continuous fold of skin extending along the back, round the tail, and forward along the belly as far as the vent. We have already seen that the development of a par- ticular organ in the individual fish frequently repeats to a greater or lesser extent the evolutionary history of that organ, and the development of the fins provides yet another example of the light thrown on the past by the study of embryology. There can be httle doubt that the median fins arose as such a continuous fold of skin, which in course of time ^ became strengthened by the development of slender supporting rods or fin-rays. Still later, certain parts of this fold, not definitely required, tended to degenerate and finally to disappear alto- gether, leaving the separate dorsal, anal, and caudal fins.^ Since the median fins are the only ones present in the primitive Lampreys and their aUies, and since they appear first in em- bryonic development, we may be justified in assuming that their evolution preceded that of the paired fins, an assumption which is largely confirmed by a study of fossil forms (Fig. 23). The origin of the paired fins is not quite so clear, and the subject has been, and still is, a matter for some controversy. It must be understood, therefore, that the following explanation, while appearing to be that which best fits the facts provided by a study of anatomy, embryology, and palaeontology, does not meet with universal support. It seems probable that at some stage after the median fin- fold had made its appearance a further advance was made by the continuation of this fold forward towards the head. Since the vent lay in its path, the original fold spHt into two portions, each of which ran forwards and somewhat upwards along the lower part of the side and ended just behind the gill-opening. This hypothetical condition is illustrated in the accompanying figure (Fig. 23A). No fossil fish has yet been discovered ex- hibiting such complete lateral folds, nor is such a stage found during the development of any living form. At the same time, two or three very primitive Sharks are known in which the condition of the fins may be described as providing an inter- mediate stage. In a form known as Cladoselache, for example, the paired fins are lappet-shaped, like the dorsal and anal, with broad bases, and their position strongly suggests that they have originated from once continuous lateral folds. In Climatias, another archaic species, both pectoral and pelvic fins are preceded by a strong spine, and there is also a row of similar spines along the sides of the body between these firis. As in the 56 A HISTORY OF FISHES case of the median fins, the appearance of the continuous folds must have been followed by the development of a series of rods of cartilage designed to strengthen these structures. Shortly afterwards, the middle portions of these folds degenerated and disappeared, and the parts at either end became enlarged and A, B. Fig. 23. EVOLUTION OF FINS. Hypothetical primitive Shark-like fishes ; c. Blue Shark {CarcJiarimis lamia) ; D. Cladoselache fyleri ; e. Climatius macnicoli. gradually took on the definitive form of the pectoral and pelvic fins. The cause of these gaps is not known, but may have been in some way connected with the undulating movements of the fish. As pointed out abo\'e, the primitive fin-folds very soon became strengthened by the appearance of a series of parallel rods of cartilage set at right angles to the body, and at the line FINS 57 where the folds joined the body each rod later split into two portions. The lower pieces, known as the basals, were situated within the body, and the upper pieces or radials were in the fins themselves. Something very like this primitive condition is found to-day in the Lampreys and their allies, where the median fins are supported solely by such a series of cartilaginous rods. In the Selachians the radials may be further subdivided into two or more portions, and form the main supports for the fins, sometimes extending almost to their margins (Fig. 24A) . The fins, however, are considerably larger than those of the Lampreys, and further strengthening is provided by the presence of numerous •^^ii.sp. Fig. 24. — STRUCTURE OF DORSAL AND ANAL FINS. A. First dorsal fin of Mackerel Shark {hurm punctatus), dissected to show cartilaginous supports ; B. Anal fin of Chinese Sturgeon (Psephurus gladius), similarly dissected ; c. Skeleton of dorsal and anal fins, and portion of vertebral column of Gar Pike {Lepidosteus platystomus). b., basal cartilages ; f.r., fin-rays ; i.sp., interspinous bones ; r., radial cartilages. hair-Uke, horny rays situated outside the radials, each ray tapering oflf to a fine point near the edge of the fin. In these fishes the horny fin-rays, as well as the cartilaginous radials, are completely enveloped by the skin and muscles associated with the fins, and are, therefore, quite invisible externally. Among the Bony Fishes the Sturgeons [Acipenseridae) , living representatives of a very archaic group of fishes, present fins of a distinctly primitive type. The dorsal and anal fins are each provided with a fleshy lobe at the base, composed of fin-muscles surrounding a series of rod-like structures — basals within the body and radials in the muscular lobe of the fin (Fig. 24B). The whole structure is not unHke that of the Selachians, but 58 A HISTORY OF FISHES instead of horny fibres the outer part is supported by bony fin-rays. The conversion of the fibres into bony supports obviously lends additional strength to the fin, and in order to retain the necessary flexibility, these are segmented into a number of sections. In all the higher Bony Fishes the lobe at the base of the fin disappears, the radials are reduced to mere nodules of bone or cartilage sunk within the adjacent muscles of the body, and the sole support of the fin is provided by the bony fin-rays. The basals, which may be joined to the reduced radials, persist as a series of fine rod-like structures alternating with the spines of the backbone, and are, therefore, known as interspinous bones (Fig. 24c). The bony fin-rays being de- veloped in the skin on either side of the fin are necessarily of a double character, and, as will be shown later on, are of two kinds, spines and soft-rays. It may be noted here that in the Lung-fishes {Dipneusti) and Sturgeons (Acipenseridae) , just as in the Selachians, the fin-rays are much more numerous than the supporting radials ; in all the higher Bony Fishes the former are reduced in number and equal to the radials. The paired fins, owing to the need for free movement in various directions, have naturally been more modified than the median fins, and in living fishes little trace remains of the primitive parallel arrangement of the supporting radials and basals. It would be out of place to follow all the changes in detail in a work of this nature, and reference must be made to the zoological text-books for further information on this matter {cf. p. 435). It must suffice to point out that the basals and radials have been variously crowded together and fused, to provide not only a strong supporting axis for the remainder of the fin, but also fin-girdles designed to connect the fins with the body (Fig. 65). As might be expected from their more important position, evolution has proceeded considerably further in the pectoral fins than in the pelvics, and its course has been mainly in the direction of a progressive shortening of the base, in order to permit of free movement in diflferent planes. In the Selachians this shortening has been effected by an outward rotation of the main basal supporting cartilage (Fig. 2b), and in the Bony Fishes by a simple crowding together and reduction of the supporting elements. Some of the more interesting types of pectoral fin structure are shown in the accompanying diagrams (Fig. 25). It will be observed that in the Australian Lung-fish (Epiceratodus) there is a definite central axis to the fin formed of basals, with a series of projecting radials on either side. FINS 59 So far no mention has been made of the supporting skeleton of the caudal or tail-fin. This is of a somewhat different nature to that of the dorsal and anal, and involves a special modifica- tion of the hinder end of the vertebral column. The tails of adult fishes may assume one of three different forms, known respectively as "diphycercal," "heterocercal," and "homo- cereal," terms which may be freely translated as "twofold tail," "unequal tail," and "equal tail" respectively. The diphycercal is undoubtedly the most primitive type, and here the hinder end of the vertebral column is quite straight Fig. 25. STRUCTURE OF PECTORAL FINS. A. Cladoselachefyleri. (After Dean) ; B. Pleuracanthus decheni. (After Fritsch) ; C. Australian Lung-fish {Epiceratodus forsteri) ; D. Cod (Gadtis callarias). and divides the fin into two equal lobes. Around the central axis are arranged a series of rods which support the membrane of the fin. Although characteristic of many of the early fishes, it is doubtful whether any forms living to-day, however straight and symmetrical their tails may appear to be, have a truly primitive diphycercal tail. We find such a tail, however, in almost all larval or embryonic fishes at a stage just before or just after they leave the &gg, but merely as a transitory stage. The heterocercal tail, characteristic of adult Selachians and some of the more primitive Bony Fishes, may best be studied in such a form as the Dog-fish [Scyliorhinus) or Sturgeon {Acipenser). Instead of being continued straight backwards, the hinder end 6o A HISTORY OF FISHES of the vertebral column is bent upwards, and the two lobes of the fin, which still retain their continuity around the tip of the tail, become differentiated into a small upper lobe and a much larger lower lobe, the latter taking its origin entirely from the lower side of the upturned backbone (Fig. 26a). In some of the more generalised Bony Fishes such as the Gar Pikes {Lepi- dosteus), Bow-fin (Amia), and Bichirs [Polypterus) , the upturned part is so much shortened that the lower lobe of the tail-fin Fig. 26. STRUCTURE OF CAUDAL FINS. A. Sturgeon (Acipemer sp.) ; b. Ten-pounder (Elops saurus) ; c. Haddock {Gadus aeglifinus) ; d. Sun-fish (Mola mola). comes to he at the end of the fish, and externally has the appearance of a symmetrical structure. In most of the higher Bony Fishes {i.e. in the Salmon, Perch Plaice, etc.) the tail is of the homocercal type. Outwardly such a tail seems perfectly symmetrical, the prolongation of the axis of the body appearing to divide the fin into equal-sized and continuous upper and lower lobes. Dissection, however, reveals that this superficial appearance is misleading and that this type of tail is nothing more than a modified heterocercal con- dition. The hinder end of the \ ertebral column turns upwards FINS in exactly the same manner, but, whilst the upper lobe has been reduced to a mere rudiment, the lower lobe is enor- mously developed and forms the greater part of the fin. The upturned part of the backbone is generally converted into a single bone, known as the urostyle, and the rays of the caudal fin are attached to the lower spines of the hinder vertebrae, which are greatly enlarged and at the same time in- clined backwards so as to be more or less parallel to the axis of the body (Fig. 26b). In some of the more specialised forms certain modifications of the homocercal tail occur. The term "leptocercal" (leaf tail) is appHed to those tails of an attenuate or whip- like form, in which the vertebral column tapers to a fine point, as in the Grenadiers or Rat-tails {Macruri- dae) (Fig. 62a), and in some of the Blennies [Blenniidae) . In other fishes the upturned part of the backbone is scarcely recognisable, and the fin- rays seem to be equally derived from the upper and lower lobe, resulting in a superficially diphycercal tail. In others, again, an apparent diphycercy is produced by the atrophy of the hinder part of the vertebral column in the adult, the upper and lower lobes of the caudal fin coalescing round the end of the abbreviated tail (Fig. 26d). In the Cods and their allies [Gadidae) occurs yet another modification, known as the isocercal (equal tail) type. This is actually, and not only apparently symmetrical, consisting of equal numbers of fin-rays separated by the axis of the vertebral column, \vhich is continued straight backwards, the separate vertebrae growing progressively smaller behind (Fig. 26c) . This symmetry has clearly been secondarily ac- quired, and has resulted from the loss of the original hetero- Fig. 27. DEVELOPMENT OF CAUDAL FIX. Six Stages in the develop- ment of the tail region in the American Flounder {Pseudoplciironectes avierican- us). (After Agassiz.) 62 A HISTORY OF FISHES cereal tail and the meeting of the dorsal and anal fins round the hinder end of the fish. In other words, the tail-fin of the Cod is not a true caudal fin at all, being composed in the main of dorsal and anal rays. A study of the embryonic development of the tail provides another interesting example of that recapitulation which has been already mentioned {cf. p. 55). In the tail of a larval American Flounder (Pseudopleuronectes) , for example, in the very young fish this is without fin-rays and is truly diphycercal (Fig. 27 1). The hinder end of the vertebral column soon commences to turn upwards, numerous hair-like rods make their appearance along the lower edge of this axis, and the tail becomes definitely heterocercal (II, III). Later, the terminal portion becomes more bent, the upper lobe degenerates, and the lower lobe increases enormously in size, is directed back- wards, and finally forms a superficially symmetrical tail (IV, V, VI). Thus, during the lifetime of a single individual the tail is transformed from a perfectly symmetrical organ through a distinctly asymmetrical stage into a superficially symmetrical one, all three types being represented in the order in which they were originally evolved. A study of the evolutionary history as revealed by the record of the rocks provides a similar story. The earliest known forms had a diphycercal tail, but very soon the heterocercal type appeared, with a very slight upward bend of the backbone and a small dorsal and large ventral lobe. Such a type of tail is found in all the ancient Selachians and Sturgeon-like fishes, and persists to-day in their living descendants. The homocercal tail, however, also made its appearance at an early stage, and is found in a fish {Dapedius) known to have flourished during the epoch known as the Lower Lias. Other curious forms of the Triassic and Liassic periods actually had two tails at one time. Of these, Diplurus and Undina (Fig. 129) are of special interest, exhibiting the true tail very much reduced in size and appearing as a tiny ap- pendage bearing a minute caudal fin. In front of this is another fin, in all probability the functional tail-fin, made up entirely of rays derived from the hinder parts of the dorsal and anal. If these enlarged dorsal and anal rays increased still further in size and eventually superseded the true caudal fin the secon- darily symmetrical tail of the Cods would come into being. Having outlined the main facts concerning the origin and structure of the fins, it will be convenient to examine some of the more interesting modifications which these organs have i FINS 63 undergone in certain fishes in order to fit them for the per- formance of new functions for which they were not originally intended. It will be convenient to deal first with the median fins, commencing with the dorsal. In the Sharks {Pleurotremata) the dorsal fins retain their primitive function of acting as stabiHsing keels, but in the Rays (Hypotremata), fishes adapted for a life on the sea floor, the need for such keels has disappeared and the dorsal fins are pro- gressively reduced (Fig. 8a), until in the more speciaHsed forms such as the Sting Rays (Trygonidae) and Eagle Rays {Mylio- batidae) they are altogether wanting (Figs. 8a; 14B; 32B). In some of the Sharks, notably in the Bull-headed Sharks {Hetero- dontidae) of the Pacific and the so-called SquaUd Sharks {Squalidae) , each of the two dorsal fins is preceded by a^ stout, sharp spine (Figs. 53B; 6oa). The origin of these spines is somewhat obscure, but they are believed to owe their existence to the fusion of some of the shagreen denticles covering the front parts of the fins {cf. p. 85). Such spines provide for- midable defensive weapons, especially when associated with poison glands, as in our own Spiny Dog-fish [Squalus) [cf. p. 140). Many of the Sharks now extinct possessed similar spines, and, not infrequently, where the remainder of the fish's body has been destroyed, these spines, which are grouped together under the name of " Ichthyodorulites," are the only record left. Many have been discovered in Devonian and Carboniferous strata, some saw-edged, some smooth, some straight, some curved, and some with elaborate sculpturing. The owners of some of these spines may never be discovered, but must have been of gigantic build, for a fin-spine found in the carboniferous limestone of Bristol measured no less than three feet in length. Among the Bony Fishes the dorsal fin exhibits great diversity both in size and form, and sometimes becomes specially modi- fied for the performance of peculiar functions. It is rarely wanting altogether, but in the group of South American fresh- water fishes known as Gymnotids, to which the Electric Eel (Electrophorus) belongs, it is either absent or reduced to a mere fleshy filament (Fig. 6og). In the more primitive forms the fin (or fins) is supported entirely by flexible and articulated rays, those at the front end generally being simple, while the majority are branched at their tips. Such fishes were grouped together by the older naturaHsts as Malacopterygians (soft fins) to distinguish them from the Acanthopterygians (spiny fins), in which the rays supporting the front parts of the dorsal and anal 64 A HISTORY OF FISHES Fig. 28. DORSAL AND ANAL FINS. a. Sea Lamprey {Petromyzon marinus),X^(i \ b. Cod {Gadus callanas),X^ff; c. Bow-fin (Amia calva),X'}( ; d. Wels or Sheat-fish (Silurus glanis),X^^ ; e. Sea Perch or Grouper (Epinephelus drinn7nond-hayi),y,^Q ; f. Stickleback (Gasterosteus aculeatus), X ^ ; g. Cardinal-fish {Apogcn frenatus), x\ \h.. Moorish Idol (Zanclus ccrnutus), x\ ; j. Sail-bearer {Velifer\hypselopterns), X i ; k. Flat- fish (Engyprosopon grandisquama), X ^ ; m. Bichir {Polypterus hichir), X \. FINS 65 fins, as well as the outer rays of the pelvics, are converted into stiff-pointed spines. Occasionally these spines are slender and flexible, but they may always be readily distinguished from true soft rays by the absence of joints or branches. Among the more generalised Bony Fishes the dorsal fin may retain its primitive form as a long fringe along the middle line of the back, especially in those species which use it as an organ of locomotion (Fig. 28c). In all the Eels (Apodes) the dorsal and anal fins are united with the caudal when this is present, and in the Morays or Muraenas {Muraenidae) the skin covering the fins is so thick that no trace of the rays is visible externally (Fig. 83A). A peculiar conger-like Eel from the West Indies {Acanthenchelys) is worthy of mention, in which the greater part of the dorsal fin is supported by soft rays in the usual manner, but in a small section near the tail these have been converted into stiff spines. Examples of spines following soft rays are very rare indeed, but the Viviparous Blenny or Eel Pout {Zoarces) of our own shores provides another case. The elongate dorsal fin occurs again in the Ribbon-fishes (Trachypteridae) , which swim by means of wave-like movements of the body, aided by similar undulations of the fin. In the Oar-fish {Regalecus), a large oceanic species attaining a length of more than twenty feet, this fin extends all the way along the upper edge of the com- pressed body, and the first few rays are prolonged into rather long filaments, each of which ends in a membranous flap (Fig. 8d) . The fins are bright scarlet in colour, and the general appearance of the head is not unlike that of a horse {cf. p. 429). In the closely related Deal-fish ( Trachypterus) the form of the dorsal fin in the young fish is remarkable, the first six or so of the rays being produced into fine filaments more than four times as long as the fish itself. These streamers are ornamented with little membranous tags placed at intervals along their length (Fig. 12 ib). Another member of the same tribe, known as Velifer, derives its name from the relatively huge size of both dorsal and anal fins (Fig. 28J). In the majority of Malacopterygians the primitive elongate fin is either considerably reduced in size or is split up into two or three separate fins. Thus, in the Herring {Clupea) and Carp (Cyprinus) there is a single rather small fin in the middle of the back composed entirely of soft rays, of which the first three or four are simple and graduated in size, the remainder being branched at their tips (Figs. 320; 43). Two extreme sizes of single dorsal fin are encountered in the Feather-back [Notop- 66 A HISTORY OF FISHES terus) or the Wels {Silurus), with a tiny flag-Hke fin in the middle of the back (Fig. 28d), and one of the fancy varieties of the Gold-fish {Carassius), in which it takes the form of a huge sail-hke structure. In some fishes allied to the Carp, one of the simple rays of the fin is stiff and spinous, and not infre- quently saw-edged behind, but no true spinous fin is developed. In the Cat-fishes {Siluroidea) a strong spine nearly always precedes the remainder of the fin, and this has resulted from the modification of one or more soft rays. It is often serrated on one or both of its edges, or is provided with formidable barbs, forming a powerful defensive weapon capable of inflicting a nasty wound (Figs. 34K; 4 id). Sometimes this spine is attached to the body by means of an elaborate ball-and-socket joint at its base, enabhng the fish to keep it erect when alarmed. When thus fixed, the spine cannot be depressed without breaking it, but a rotary movement upwards and towards the body serves to lower it again. A curious modification of the dorsal fin is found in the Tarpon (Megalops), as well as in some of the members of the Herring family (Clupeidae), the last ray being drawn out into a long filament which is concave on its hinder edge and tapers to a fine point (Fig. 13B). The purpose of this seems to be connected with the leaping habits of this fish. According to Mr. Mowbray, the Tarpon, preparatory to making a leap, "lashes this whip around to one side of the body and clamps it tight to its side." This adheres by suction to the body, and, by keeping the fin rigidly to one side, aids the fish in deter- mining the course of the jump, the turn naturally being made towards the side to which the ray is pressed. Among those soft-rayed fishes with more than one dorsal fin, mention may be made of the Gadoids, which belong to the order of fishes known as Anacanthini (without spines) , a group including such well-known food-fishes as the Cod, Pollack, Whiting, Haddock, Hake, and Ling. In the first four of these fishes there are three dorsal fins (Fig. 28b) ; in the others only two. As their name implies, the fins of these fishes are sup- ported entirely by soft rays. The little Rocklings {Motella), members of the same order, have a series of free rays just in front of the ordinary dorsal fin, and these may be continuously and rapidly vibrated for long periods. The function of such rays appears to be that of a sensory organ for locating and detecting food. The Salmon {Salmo) also possesses two dorsal fins, but while FINS 67 the first resembles that of the Herring and is supported by soft rays, the second takes the form of a small flap, without any supporting structures, composed entirely of fatty tissue and covered with skin (Fig. 13c). This is known as the adipose fin, and is found in all the members of the Salmon family, as well as in many Characins, and in the majority of Cat-fishes. In some of the latter the adipose fin is comparatively large (Fig. 11), and in certain species may develop a few soft rays. It is of interest to note that in the Mailed Cat-fishes {Loricariidae) this fin is a triangular flap of skin, the front edge of which is supported by a stout, movable spine (Figs. 32F; 42c), but in some related naked forms {Cyclopium) from mountain streams the spine has disappeared and the adipose fin has reacquired the typical Cat-fish form. This pro\ddes one of the very rare cases of reversible evolution, where an organ has become changed and afterwards reverted to its original structure. The position of the dorsal fin or fins also exhibits a fair amount of variation in diflferent fishes. In the Gar Pikes (Lepidosteus) of North America (Fig. 48), and in the quite unrelated Pike {Esox), both dorsal and anal fins are placed well back towards the hinder end of the fish; in the Herring [Clupea] and Carp {Cyprinus) the dorsal occupies a position more or less in the middle of the back (Fig. 43) ; and in some of the Cat-fishes the rayed fin is considerably nearer to the head than to the tail (Fig. sSd). During growth the vertical fins tend to undergo some change in form, those of young fishes being generally higher than those of adults. The position may also alter during the life of the individual fish, as in the Herring [Clupea). In the larval stages the dorsal at first lies close to the tail, but moves gradually forwards as growth proceeds. In the grotesque Shrimp-fish (Centriscus) the arrange- ment of the vertical fins is unique. The thin, bony cuirass encasing the body ends behind in a long, stout spine, and the two dorsal fins, crowded together at the hinder end of the fish, are placed below the spine, the second actually pointing down- wards. The tail has been deflected at an obtuse angle from the trunk, and terminates in a small caudal fin, also pointing downwards (Figs. 12; 42H). Among the spiny-rayed fishes, the rays at the anterior end of the dorsal fin may be transformed into spines as in the Sea Perch [Epinephelus) (Fig. 28e) or Fresh-water Sun-fish (Lepomis) (Fig. 34e), or the spinous portion may be separated off as a distinct fin, as in the Mackerel {Scomber) (Fig. 5A) or Cardinal- 68 A HISTORY OF FISHES fish {Apogon) (Fig. 28g). The evolution of spinous rays added a new function to the fins, that of attack and defence, and pugnacious fishes Hke the httle Sticklebacks (Gasterosteus) know how to use their formidable dorsal and pelvic spines to the best advantage (Fig. 28f). The spines vary greatly in different fishes, both in height and thickness, and may even be soft and flexible as in some of the Gobies and Blennies (Fig. 34f). The Flat-fishes (Heterosomata) provide an example of the secondary trans- formation of spines into soft rays. It is known that these fishes have evolved from spiny- rayed forms not unlike the Sea Perches, but have become very much modified for a life on the sea bottom. In a primitive form from tropical seas {Psettodes) the dorsal fin commences well behind the head, and the front part is still supported by stiff spines, but in all other Flat-fishes the fin has grown forward on to the head, and the spines have been reconverted into flexible articulated rays, thus allowing of the wave-like movements essential for progression {cf. p. 27). In some of the more specialised Soles (Synaptura), and in the Tongue- soles (Cynoglossus) , both dorsal and anal fins are united with the much reduced caudal, so that the three, together with the pelvic, form a complete fringe round the body. In many of the Sea Perches and their alHes the spines are unequal in size and strength, a somewhat delicate spine alter- nating with a stout one. In the Pine-cone Fish {Monocentris) of Japan the spines are particularly formidable and are curiously Fig. 29. LOCKING MECHANISM OF FIN-SPINES. A. Shoulder girdle and spines of pectoral fins of a South American Cat-fish {Doras sp.), ventral view, X f ; b. Dorsal fin-spines and associated bones of the Trigger-fish (Balistes sp.), lateral view, X f , FINS 69 arranged, being alternately directed to the right or left, none of them being truly vertical as in other fishes (Fig. 42E). The closely related Soldier-fishes (Holocentrum) of the coral reefs of tropical seas derive their name from the stout and sharply pointed spines with which the fins are provided. In a few fishes, notably in the Weever ( Trachinus) (Fig. 58A) and in the Poison-fish {Synanceia) (Fig. 58B), the spines of the dorsal fin are associated with poison glands, thus adding greatly to their efficiency as defensive weapons {cf, p. 142). In the majority of fishes the dorsal and anal fins are capable of being erected or depressed at will, the separate spines or soft-rays being provided with special muscles for this purpose. When the fish is moving at any speed, the fins are always lowered, and in a fast-swimming form such as the Mackerel {Scomber) both the dorsal and anal can be folded away into more or less deep grooves in the body, the object being to prevent them from breaking the streamline form and so im- peding progress. In the allied Sail-fish {Istiophorus) the spinous dorsal fin is of enormous size, and is believed to be projected from the surface of the sea and used like a sail to assist progres- sion, but the whole structure can be tucked away into a deep groove when not required (Fig. 6a). After a rapid burst of speed, most fishes erect the dorsal and anal fins to their fullest extent, and these act as brakes and assist in slowing down. Among other fishes with more or less modified dorsal fins, the Trigger-fishes and Bichirs are worthy of mention. The Trigger-fishes (Balistidae) owe their name to the structure of the spinous dorsal, this being supported by three spines, the first very strong and hollowed out behind to receive a bony knob at the base of the second; by this mechanism the first spine remains immovably erect until the second, which acts as a trigger, is depressed (Fig. 29B). The Bichirs {Polypterus) are not really Acanthopterygian fishes, but the anterior part of the dorsal fin takes the form of a number of separate, flag-like finlets, each consisting of a stout spine supporting a sail-like membranous flap (Fig. 28m), hence the name Polypterus (many fins)! In the Mackerel [Scomber), Tunny (Thunnus), Bonito {Gym- nosardd) , and allied forms the soft dorsal fin is followed by a row of separate finlets, each of which is made up of a single much branched ray (Figs. 4, 5A). Their function is a little obscure, but they may act as subsidiary rudders. The Remoras or Sucking-fishes [Echeneididae) are remarkable 70 A HISTORY OF FISHES for the possession of an oval adhesive disc of complicated struc- ture placed on the broad and flat upper surface of the head. It is provided with a varying number of transverse plates with free hinder edges, the whole being surrounded by a mem- branous fringe (Fig. 30). By means of this disc the fish can attach itself to any flat surface, a slight erection of the plates creating a series of vacuum chambers. The adhesion is so strong that a Remora can only be dislodged with difficulty, unless it is pushed forward by a sliding movement. Some naturaUsts state that when attached these fishes seem to become quite insensitive, and show no sign of life however roughly ^M^3 Fig. 30. treated. Feeding as they do on other fish, the Remoras are in the habit of attaching themselves to Sharks, Whales, Por- poises, Turtles, and even occasionally to ships (cf. p. 427), and in this way they are not only protected from their enemies, but are carried without effort to fresh feeding grounds. Once among a shoal of fish they soon detach themselves, dart oflT and swim actively about in pursuit of prey, seeking a fresh anchorage when their appetite has been satisfied. The point of special interest about the sucker is that it is nothing more than a very much modified spinous dorsal fin, whose rays have become divided into two halves, bent outwards in opposite directions, and have been transformed into the trans\erse plates. The peculiar mode of life of the Remoras, and the FINS 7i structural adaptation associated with it, probably arose from the habit found in many fishes of accompanying larger creatures either for protection or gain, and of concealing themselves under the shelter of any large object in the sea. "It is evident," writes Dr. Regan, "that the Sucker-fishes must have been derived from fishes which had, as the Pilot-fish has to-day, the habit of associating with Sharks. The spinous dorsal fin of the Pilot- fish {Naucrates) and of other oceanic fishes of the same type, with the spines folded back within a groove, might possibly have some power of adhesion if the edges of the groove were applied to another object and the spines were then slightly raised." Another remarkable modification is found in the Angler- fishes or Pediculates (Pediculati) , in which the first ray of the spinous dorsal fin is placed on the snout and transformed into a line and bait. In the Common Angler or Fishing-frog [Lophius) of our own shores, for example, this ray is quite flexible and bears a membranous flag-like appendage at its tip, its function being to attract the attention of small fishes when waved about in the water in front of the Angler's formidable jaws (Fig. 89). In the related Frog-fish [Antennarius) and Bat-fish {0 gcocephalus) (Fig. 40D) the line and bait is much reduced in size, and is sometimes represented merely by a short tentacle lodged in a cavity above the mouth. Among the Ceratioids, oceanic Angler-fishes spending their lives in a region of more or less perpetual darkness, the bait generally takes the form of a bulb of varying size which can be made luminous at will, and acts as a lamp to attract other fishes to destruction (Fig. 31). In one species {Lasiognathus saccostoma) the basal part of the dorsal fin-ray has been converted into a stout rod, followed by a slender line, which is provided, not only with the usual luminous bulb, but also with a series of curved horny hooks — a complete angler indeed (Fig. 31G)! The anal fin, placed on the lower edge of the body between the tail and the vent, may be briefly dismissed. Like the dorsal it exhibits some variation both in size and form in diflferent fishes. In the Gymnotids, Eels, and other forms in which it functions as a locomotor organ, the anal fin is very long (Figs. 5f; iob); in other fishes, where it acts mainly as a balancing keel, it is considerably shorter (Fig. 28); in the Ribbon-fishes {Trachypterus, Regalecus) it is absent altogether (Fig. 8d). It may be supported entirely by soft rays, or the first few rays may be converted into stiff spines, often of some size (Fig. 28e, g). 72 A HISTORY OF FISHES In some fishes, of which the John Dory {Zeus) and the Horse Mackerel {Trachurus) may be mentioned, the spinous portion IS separated off from the remainder as a distinct fin (Fig. 4 IB). In the Cod {Gadus) and related species the anal fin IS dnided into two portions, each being composed entirely of soft rays (Fig. 28b). In many of the South American Cyprino- donts [Poeciliinae), tiny fishes inhabiting fresh and brackish Fig. 31. — CERATIOID ANGLER-FISHES. A. Linophryne arborifer,X i ; b. Melanocetus johnsoni^X^ ; c. Lasiognathus saccostoma, X i . waters, the males are much smaller than the females, and the anal fin is specially modified to form an organ of elaborate structure used for purposes of copulation (cf. p. 296). The last of the median fins, the caudal, has already been mentioned in discussing the tail itself, and Httle need be added here. Like the dorsal and anal, it is composed of simple or branched rays supporting a thin membrane; true spines are never developed in this fin, but rudimentary or procurrent rays FINS 73 Fig. 32. — CAUDAL FINS. A. Fox Shark or Thresher (Alopios vulpes), X ^ ; B. Sting Ray (Trygon latus), X ^\ ; C. Frilled Shark (Chlamydoselachus anguineus), X ^V; D. Allis Shad (Alosa alosa), xi ; E. Pipe-fish {Microphis boaja),Xi ; F. Mailed Cat-fish (Loricaria apelto- goster),xi; G. Snipe Eel (Nemichthys scolopaceus),xl ; h. Black-fish {Dallia pectoralis)y X f. 74 A HISTORY OF FISHES resembling spines may be found at the base of the lobes (Fig. 8 1 a). The Sea Horse {Hippocampus), which is unique in using the tail as a prehensile organ, shares with some of the Eels (Apodes) and a few other fishes the distinction of being without a caudal fin (Fig. 5E). The caudal fin of the Sharks varies somewhat in form, but is rarely outwardly symmetrical, and the supporting rays are never visible externally. The function of the curious notch found in the upper lobe of the tail-fins of these fishes has never been satisfactorily explained, and it is possible that it may have possessed some special use in the past which no longer exists to-day (Figs, ib; 32 a). The Thresher or Fox Shark (Alopias) is remarkable for the great length of the upper lobe, which forms half the entire length of the fish (Fig. 32A). This Shark is said to swim round a shoal of fishes, splashing the water with its tail, and thus driving them into a compact mass, when they form an easy prey. Among the bottom-living Rays (Raiidae) the caudal fin tends to be much reduced in size, whilst in the more specialised Sting Rays (Trygonidae) and their allies it is wanting altogether, the long whip-like tail simply tapering to a fine point (Fig. 32B). Among Bony Fishes with outwardly symmetrical tails, the shape and size of the caudal fin exhibits a good deal of varia- tion (Fig. 33). Six main types of fin may be recognised, described respectively as lunate or crescentic (Tunny), forked (Herring, Mackerel), emarginate (Trout, Carp, Perch), trun- cate (Flounder), rounded (Turbot and Lemon Sole) and pointed (Goby). The shape of the tail generally provides a good index of speed and agility. As a general rule, fishes with lunate or deeply forked tails are capable of swimming for long periods at high speed, whereas those with squarish or rounded tails, although capable of sudden, short bursts of speed, are on the whole comparatively slow swimmers. The Deal-fish [Trachypterus) and Sun-fish (Mola) may be selected as examples of fishes with unusual caudal fins. In the former this fin is unique in being directed upwards at right angles to the axis of the body. In the young fish the rays of the lower lobe are prolonged into lengthy filaments like those of the dorsal and anal fins, but these become progressively shorter as growth proceeds and finally the lower lobe of the fin dis- appears (Fig. 12 ib). It will be recalled that in the Sun-fishes the body ends abruptly behind the short, high dorsal and anal fins, and this is margined by a low, rounded caudal with a sHghtly wavy edge (Figs. 50; 26d). Such a tail, to which the FINS 75 term " gephyrocercal " (bridge tail) has been applied, is found only in the Sun-fishes and the Pearl-fishes (Fierasfer), and represents a very specialised condition. So much for the median or unpaired fins. The paired fins, corresponding respectively to the arms and legs of the land vertebrates, are absent in the Lampreys and Hag-fishes (Cyclo- stomes), but, with few exceptions, one or both pairs are developed in other fishes. The pectoral fins vary very little in position, being situated just behind the gill-opening or openings, and placed near the lower edge of the body in some fishes and higher up on the sides in others. The pectorals of the Sharks are considerably larger than those of the generahty of Bony Fishes, being used Fig. 33. — ^SHAPES OF CAUDAL FIN. A. Lunate or crescentic ; b. forked ; c. emarginate ; d. truncate ; e. rounded ; F. pointed ; G. double emarginate. almost entirely for steering purposes (Fig. 34b). A Shark seems to be quite incapable of making a sudden stop, and never uses the pectorals as brakes, being compelled to swerve to one side of an obstacle which lies in its path. The enormous, flattened, lobe-like pectoral fins of the Rays and their relations {Hypo- tremata), joined to the sides of the head and body and forming the principal organs of locomotion, have been already described (Fig. 34a). They may also be used for steering, especially in those forms in which the tail has degenerated to a mere fila- m.ent. As in the case of the median fins, the paired fins of the Selachians are completely covered by skin and muscles, no trace of the rays being visible externally. In the Bony Fishes these fins are nearly always relatively small, paddle-shaped organs, and only that part of the fin 76 A HISTORY OF FISHES Fig. 34. PECTORAL FINS. a. Eagle Ray {Myliobatis freminvillii),xi^ \ b. Spotted Dog-fish {Scyliorhinus ca7iiailus),xi; c. Tunny (Thunnus thynmis) , X ^\ ; d. Thread-fin (Polrnemus paradiseus),x\ \ e. Fresh-water Sun-fish (L^^ow/^ 7riegalotis),X I ; f.' Mud- skipper {Periophthalmus koelreuteri), X ]-; g. Scorpion-fish {Pteroisvolitans), X tV ; h. Cirrhitid (Paracirrhites forsteri), x i ; j. Flying-fish (Exocoetiis volttans),X^ ; k. South American Cat-fish (Doras sp.),x^ ; m. Gurnard {Trigla pini),x^^. FINS 77 which is supported by the fin-rays is visible without dissection. These rays may be simple or branched, and in some of the Cat-fishes there is a stout spine along the outer edge of the fin, which may be saw-edged along one or both of its margins and attached to the body by an elaborate joint (Figs. 29A; 34k). In the Mad Toms or Stone Cats [Noturus, Schilbeodes) of the United States, each of the spines is provided with a special poison gland at the base. A South American Cat-fish {Doras) uses the spines of the pectorals for progression overland, pro- jecting itself along on the tips by the elastic spring of the tail (Fig. 34k). ^ , . i7- u The shape of the pectorals also varies to some extent. Irishes of moderate or slov/ speed, which may use these fins for pro- pulsion or for backing water, have broad, rounded or spatulate pectorals (Fig. 34e). In speedy fishes, on the other hand, where they are employed in executing wheeUng or turning move- ments, the fins are always longer and frequently sickle-shaped (falcate) (Fig. 34c). The pecuhar leaf-like paired fins of the Australian Lung-fish {Epiceratodiis) have been described on an earher page (Fig. 99c). In the other Lung-fishes from Africa [Protopterus) and South America (Lepidosiren) the central lobe becomes long and narrow, the marginal fringe is reduced or entirely suppressed, and the fins take the form of long, tapermg filaments (Fig. 99A, b). Rarely are the pectoral fins wanting, although they may be much reduced in size and efficiency. In some of the Pipe-fishes, however, and in certain of the Eels, they are absent. The pectoral fins, therefore, are concerned mainly with propulsion or steering. The older authors laid considerable stress on their use as organs of balance, but more recent experi- ments tend to show that neither of the paired fins plays any important part in the maintenance of equilibrium. A number of fishes have been experimented upon by removing some or all of these fins, and in no case was the horizontal position seriously affected, provided the individuals remained at rest and attempted no turning movements. True, sick or dead fishes generally float belly upwards, but this is probably due to physiological causes. As in the case of the dorsal and anal, the pectorals have become variously modified in certain fishes for the performance of new functions, and the more interesting of these adaptations may now be examined. The Flying-fishes [Exocoetidae) have greatly enlarged pec- torals, often extending backwards as far as the tail (Fig. 34J), 78 A HISTORY OF FISHES and use these to make the flights through the air for which they are famed. In order properly to understand these flights, it is necessary to look at some more generaHsed members of the same order, the Skippers {Scombresox) , Gar-fishes {Belone), and Half-beaks {Hemirhamphus). These fishes, especially the Half- beaks, are experts at leaping or "skittering" over the surface of the sea, but the pectoral fins, being comparatively small, are only able to raise the head and forepart of the body out of the water, the tail remaining submerged and vibrating rapidly. There can be Httle doubt that the prolonged aerial excursions of the Flying-fishes are improvements upon the spasmodic jumps of the Gar-fishes and their allies. They are undertaken primarily in order to escape from enemies, or when the fishes are alarmed by approaching ships, but sometimes without any apparent cause. The actual flight seems to be carried out as follows: the fish accelerates its speed, rushing along near the surface of the water with its tail moving very rapidly from side to side; it then makes a sudden leap out of the sea and is borne along through the air with the pectoral fins outstretched and practically motionless. The chief motive power of this soaring flight is supphed by the tail, there being Httle, if any, actual flapping of the "wings" as in birds or bats; the pectoral fins act merely as parachutes which enable the fish to glide through the air. ^ The flight appears to be checked by movements of the pelvic fins, which are often enlarged, and the fish returns to the water tail first, although, according to Dr. Hankin, in one species "they plunge head foremost into the water without any visible attempt to check their speed." He estimates the longer flights as from 200 to 400 metres in length, and the average speed under favourable conditions as from 10 to 20 metres a second. The fish seems to be quite incapable of steering itself in the air, but during the flight the hinder part of the body may become re-immicrsed in the water, and by a vigorous flip of the tail the fish changes its direction to the right or left and, at the same time, gams increased speed. As a rule the flights are close to the surface of the sea, but the fishes are not infrequently carried upwards to a height of 15 or 20 feet by a current of air, and iii this way often land on the decks of ships in stormy weather. In the Flying Gurnards [Dactylopteridae) the pectoral fins are even more enlarged, but here they seem to be actually vibrated during flight, moving up and down Hke the wings of a butterfly. These flights, however, are more clumsy and less successful than those of the true FINS 79 Flying-fishes. Among other fishes capable of short and generally erratic flights, the little Chisel-jaw {Pantodon) of African rivers and swamps, and a peculiar deep-bodied Characin fish from South America [Gasteropelecus) may be mentioned. The pectoral fins of the Chisel-jaw are not particularly long, but are joined to the body by flaps of skin. The little Mud-skipper {Periophthalmus) , found on the coasts of tropical Africa, Asia, and AustraHa, is renowned for its habit of leaving the water and walking or skipping about on the sand or mud in search of food. It chases its insect prey among weeds and rocks, and on land is quite as agile as many lizards. The pectoral fins are specially modified in relation to this habit, each being attached at the end of a kind of muscular arm, which can be moved backwards and forwards and is used exactly like a limb. Among other structural peculiarities designed to assist its progression on land, the low anal fin and the stout lower rays of the caudal may be noticed (Fig. 34f). Dr. Regan writes: "When walking on the mud each step is accomplished by a forward movement of both pectoral fins, which are then put on the ground and draw the rest of the body after them; these steps are repeated rapidly, and as each results in an advance of about half an inch, very fair progress is made; the pelvic fins support the body during the turning forward of the pectorals. But, as their name implies, the Mud- skippers often leap along the mud, or from one stone to another; short jumps may be accomplished by the action of the pectoral fins alone, but longer ones, which may be as much as a yard long, are made by a stroke of the tail. This is their way of getting along when they are in a hurry, and they may often be seen playing on the mud, jumping about in chase of each other." In the Sea Toads (Chaunacidae) and Frog-fishes [Antennariidae) these fins again take the form of arms, ending in many fingered "hands," by means of which they are able to crawl slowly about on the sea floor or to hang on to rocks or weeds (Fig. 85). In the related Bat-fishes {Ogcocephalus) the "arms" are even more muscular (Fig. 40D). In certain fishes some or all of the rays of the pectoral fins may be drawn out into delicate filaments which serve as organs of touch. In the Thread-fins {Polynemus) , for exam.ple, some four to eight of the lower rays are detached from the rest of the fin, and take the form of hair-like structures which may be longer than the fish itself (Fig. 34d). In a deep-sea fish known as Bathypterois the eyes are very much reduced in size, 8o A HISTORY OF FISHES but the loss of vision is amply compensated for by the sensitive feelers formed by the rays of the pectoral and pelvic fins In the Perch-like Cirrhitids or Firm-fins (Cirrhitidae) , of which the Australian Trumpeter {Latris) is perhaps best known, the lower rays of the pectorals are simple, thickened, free at their tips, and sometimes more or less prolonged; here, again, they act as sensory organs, and probably aid the fishes in their search for food (Fig. 34h). The Gurnards (Trigla) and Sea Robins [Prionotus) have two or three of the lower rays of the pectoral fins detached from the remainder and modified to form stout finger-Hke appendages (Fig. 34m). These are used for turning over sand or stones, exploring shells, and otherwise searching for food, but also serve a locomotor purpose, the appendages acting as limbs, forward movement being produced by placing the tips of the rays in contact with the sand and pushing backwards. In the Flying Gurnard [Dactylopterus) the upper wing-like portion of the pectoral is used for parachuting, and the lower part, as well as the long thin pelvic fin, for creeping about on the sea floor. According to Dr. Beebe, the pelvic leg-like fins work alternately, one after the other. Coming, finally, to the pelvic or ventral fins, corresponding to the legs of land vertebrates, it may be noted that, unhke the pectorals, their position varies considerably in the different groups of fishes, and is of some importance in classification. In all the Selachians, and in the lower kinds of Bony Fishes, such as the Fierring (Clupea), Salmon {Salmo)^ and Carp (Cyprinus), the pelvics are placed in the middle of the belly between the pectorals and the anal, and are said to be abdominal in position (Figs. 13B, c). In other Bony Fishes, of which the Perch (Perca), Bass (Morone), and Mackerel (Scomber) will serve as examples, they are thoracic in position ; that is to say, they lie farther forward in the region of the chest and more or less below the pectorals (Figs. 5A; 34c, e). In others, again, such as the Cods [Gadidae) and certain of the Blennies (Blennioidea), they are described as jugular in position, and actually lie in front of the pectorals in the region of the throat (Fig. 28b). In a number of Bony Fishes, and particularly in those forms which spend most of their time in burrowing, these fins are either very much reduced in size or are wanting altogether. All the living members of the order of Eeis (Apodes) are without pelvics, but these have undoubtedly been derived from fishes which possessed a full set of fins, confirmation of this view being FINS 8i pro\ided by a fossil Eel ( Urenchelys) from the Chalk of Mount Lebanon which shows distinct traces of having possessed both paired fins as well as a separate caudal fin. Pelvics are absent in all the Pipe-fishes (Syngnathidae) , Symbranchoid Eels {Sym- branchii), Gymnotids {Gymnotiformes) , Globe-fishes {Tetrodontidae) and Porcupine-fishes {Diodontidae) , and are also suppressed in many of the Blennies and Cusk-eels {Blennioidea, Ophidioidea) . E\en when developed, they may be reduced to mere filaments, as in some of the members of the Cod tribe [Gadidae), and in the Oar-fish (Regalecus), where they are represented by a pair of long rays, each expanded into a blade-like structure at the tip (Fig, 8d). In the Sticklebacks [Gasterosteus) each pelvic is composed of a sharply pointed spine and one soft ray (Fig. 28f), and in the Pine-cone Fish {Monocentris) (Fig. 42E) and in some of the Trigger-fishes, etc. (Plectognathi) , is reduced to a spine alone (Fig. 42D) . In certain species of File-fishes {Mona- canthidae) the pelvic bone with" its spine is freely movable, and is connected with the body by a wide flap of skin (Fig. loc); this is said to be used by the fish for fixing itself into crevices in the rocks or coral reefs. It is very rarely that the pelvic fins have any connection with locomotion, and they mainly function as "bilge keels" or as accessory manoeuvring organs. During rapid swimming they are generally drawn in close to the body. As in the case of the pectorals, some or all of the rays may be drawn out into lengthy filaments, as in the Dwarf Cod-fish [Bregmaceros) and Gourami ( Osphronemus) . The most important modification of the pelvics is to form a sucking disc to enable the fish to cling to rocks, stones, and other fixed objects. The little Bornean Sucker {Gastromyzon) , found only in the mountain torrents of Borneo {cf. p. 239), has the whole of the lower surface of the body modified to form a large sucker, in which the long and horizontally placed pectoral and pelvic fins play an important part (Fig. 350). The Gobies (Gobioidea), a large and varied sub-order of fishes, mostly of small size, found mainly am.ong the rocks between tide-marks, have the pelvic fins united to form a rather deep cup-like sucker (Fig. 350). In the Lump-sucker [Cyclopterus) and Sea- snail (Liparis) a somewhat similar sucking disc is developed, but the pelvics have been so much modified as to have com- pletely lost their fin-like appearance (Fig. 35A). This disc is very powerful, and some difficulty is experienced in remo\'ing a fish from an object to which it has attached itself. The Cling- 82 A HISTORY OF FISHES fishes [Gobiesocidae) are curious little creatures found between tide-marks among loose stones and shells, to which they adhere firmly by means of their adhesive discs. The disc is relatively large and of a complicated structure ; it is composed largely of pads of thickened skin, but the widely separated pelvic fins and even the much modified bones of the pectoral girdle may contribute to its formation (Fig. 35B). Among the Flat-fishes [Heterosomatd) , the asymmetry so Fig- 35- — FISHES WITH VENTRAL SUCKERS. A. Lump-sucker (Cyclopterus lumpus), X |^ ; b. Cling -fish (Lepadogaster gouani), X\ ', c. Bornean Sucker {Gastromyzon bornunsis)yX^ ; d. Lower surface of Black Goby (Gobius niger), X ^. characteristic of the group extends to the pelvic fins in a large number of species. In the Scald-fish {Arnoglossus) , for example, the pelvic of the left side, that is to say, of the upper or coloured side, is large and placed along the lower edge of the body like a fringe, whereas that of the lower side is quite small and placed at some distance from the edge. In some specialised Australian Flat-fishes {Rhombosolea) the pelvic fin of the blind side, in this case the left side, has disappeared altogether, and the other elongate fin has become joined to the FINS 83 anal, thus completing a more or less continuous fin round the edge of the fish's body. In all living Sharks and Rays the hinder parts of the pelvic fins in the male are specially modified to form elaborate organs known as "claspers" or mixopterygia, which not only serve to grasp the female during copulation, but also assist in the process of fertilisation {cf. p. 294). A few members of the group of Bony Fishes known as Cyprinodonts [Microcyprini) have the fins in the male modified for a similar purpose {cf. p. 296). CHAPTER V SKIN, SCALES, AND SPINES Structure of skin. Dermal denticles of Selachians. Tail-spine of Sting Rays. Saw-fishes. Scales of Bony Fishes: ganoid scales, cycloid and ctenoid scales. Tubercles. Bony scutes. Armoured fishes. Bony plates, rings, spines. Scales of Lung-fishes. Arrangement of scales, scale counts. Axillary scales. Lateral line. Scale-reading and age-determination. Replacement scales. The skin of a fish, like that of any other vertebrate, is composed of two layers, a thin outer epidermis and an inner dermis. The epidermis is made up of several layers of simple cells, of which the outer are being constantly worn away by wear and tear, and replaced by new ones budded off at its base. The dermis has a more complicated structure, being made up of a thick layer of what is known as connective tissue, with which are mingled muscle fibres, clusters of fine blood-vessels, and nerves. The inherent slipperiness of a fish's body is due to the presence of a slimy mucous which is constantly being poured out in large quantities by special glands situated in the epidermis, its function being to minimise friction with the surrounding water and to enable the fish to glide easily along. The slime excreted varies greatly both in quantity, and probably also in com- position, in different species. In some of the Lampreys the glands are especially numerous, while a single Hag-fish (Myxine) placed in a bucket of water will soon convert the fluid into a thick mass of whitish jelly. In addition to the skin, there is generally an outer covering of scales of one kind or another, generally spoken of as the exo- skeleton, to distinguish it from the endoskeleton (skull, back- bone, etc.). Where it is overlaid by scales the skin itself is nearly always thin and delicate, but in those fishes without scales, plates or spines it is strengthened in some way. Thus, in the naked Cat-fishes {Siluroidea) it is thick and leathery, and in the Sun-fish (Mola) the tough roughened skin is further reinforced by an underlying layer of cartilaginous material two or three inches in thickness. The curious Horse-fish (Agriopus) of South Africa is unique in being able to cast off its skin 84 SKIN, SCALES, AND SPINES 85 in patches like a snake, a new and brightly coloured skin developing below the old one. The scales of a fish are products of the activity of the skin, and owe their existence to the presence of lime salts in the tissues of the body, which become deposited in the dermis. The form of the scales, spines, or other related structures varies considerably in the different groups of fishes, and provides an important character for their classification. In the Cyclostomes scales are altogether wanting, but a study of their fossil ancestors suggests that this is a feature of de- FlG. 36. — DERMAL DENTICLES. a. Isolated denticles of the Spotted Dog-fish (Scyliorhimis caniculus), greatly enlarged ; b. Diagrammatic cross-section of a denticle of a Selachian, showing the enamel covering and the central pulp cavity, greatly enlarged ; c. Portion of skin with dermal denticles of the Bramble Shark {Echinorhinus spinosus), X | ; d. "Buckler" of Thornback Ray {Rata clavata), lateral and dorsal view, xf ; e. Tail-spine of Sting Ray {Trygon sp.), X \. generation rather than a primitive character. The dermal denticles of Sharks and Rays, sometimes known as odontoids or placoid scales, almost certainly represent the most primitive type of scale, and may conveniently be described first. The surface of a shark's body is generally prickly to the touch, due to the presence of innumerable tooth-like structures arranged in regular oblique rows, covering the whole of the head, body, and part of the fins. Each of these denticles consists of two portions, a bone-like base which is embedded in the skin and therefore invisible during life, and a superficial enamel-covered spine projecting freely outwards and backwards (Fig. 36a, b). Such denticles provide the familiar "shagreen," and the 86 A HISTORY OF FISHES .-«i pro\ed durability of shark-leather is largely due to the rein- forcement provided by these structures (cf. p. 401). In embryonic development the denticles make their appear- ance as minute conical or papilliform outgrowths of the skin; the outer epidermal layer of each of these cones becomes transformed into a coat of hard enamel by the deposition of lime salts; the dermal portion gives rise to dentine or ivory— the characteristic substance of teeth — with a central pulp cavity containing the blood-vessels and nerve (Fig. 37A). The base of the cone spreads out to form the bony plate, in the centre of which is a hole for the passage of the blood- vessels and ner\'e. The im- portant point to notice about these denticles is that, both in their structure and in the manner of their development, they are strictly comparable to the teeth, which have quite clearly been derived from them {cf. p. 120). The arrangement of the denticles already described is that found in the familiar Dog - fishes [Scyliorhinus, Squalus, Mustelus)^ as well as in most other Sharks, but the denticles themselves pre- sent considerable differences in form and size, being some- times flat, sometimes spine- like, and sometimes taking the form of rounded knobs. In the Bramble Shark {Echinorhinus) , however, they are distributed irregularly over the body, and appear as large rounded tubercles of varying size, each sur- mounted by a tuft of fine spines (Fig. 36c). In the Rays {Raia) they are generally scattered sparsely and unevenly over the upper surface of the disc formed by the head, body, and pec- toral fins; they are usually most prominent along the middle line of the back and on the upper part of the tail, and may be sharply pointed, flattened, or reduced to mere knobs (Figs. 14B; 107). In the Thornback Ray (Raia clavata) the greatly enlarged denticles are known as "bucklers" (Fig. 36d), and in ?^^^: B Fig. 37. — DEVELOPMENT OF DENTICLES AND SCALES. A. Vertical section through the skin of an embryo shark. (After Gegenbaur); B. Diagrammatic longitudinal section through the skin of a Bony Fish, to show position of scales. (After Boas.) d.y dermis ; e., epidermis. SKIN, SCALES, AND SPINES 87 Other species of the same genus each principal spine has smaller accessory spines developed round its base. In the Torpedoes {Torpedinidae) , on the other hand, and in some of the Sting Rays (Trygonidae) and Eagle Rays (Myliobatidae) , the denticles have been discarded and the skin is quite smooth. Mention may be made here of the tail-spine or "sting" of the Sting Rays and related forms, which in these fishes takes the place of the dorsal fins. Its origin is somewhat obscure, but it may have arisen through the enlargement or fusion of certain denticles in the tail region. It is generally serrated along both margins, and may be as much as from eight to fifteen inches in length. It provides a formidable weapon, and when the tail is lashed from side to side or curled round the intended prey, Fig. 38. Saw-fish {Pristis pectinatus), X ijV- inflicts painful jagged wounds. The "stings" are shed from time to time, and replaced by new ones growmg from under- neath; sometimes two or three may be present in one fish at the same time (Fig. 36e). . In the Saw-fishes [Pristis), ray-Hke fishes found m all warm seas, the snout is produced to form a long flat blade, armed on either margin with a series of strong tooth-like structures (Fig. 38). These are modified dermal denticles, and are not only very much enlarged, but are firmly implanted in sockets m the cartilage of the rostrum. Saw-fishes grow to a large size, specimens twenty feet in length being quite common and "saws" six feet in length and a foot across the base are by no means rare objects in the windows of curio or natural history dealers. Normally, this effective weapon is brought into use 88 A HISTORY OF FISHES among a shoal of fish, and, wielded with a side-to-side move- ment, inflicts great execution. When attacking larger prey, the Saw-fish is said to use the saw to tear off lumps of flesh from the body of the victim, the detached fragments being seized by the mouth and swallowed at leisure. It is of interest to note that in the Chimaeras and their allies {Holocephali) , although the skin is naked in the adult, small patches of denticles, essentially similar in structure to those of the Sharks and Rays, still remain on the claspers (Fig. 80), and in the young there may be a double row of denticles along the back. The scales of all the Bony Fishes differ from the denticles of the Selachians, not only in their structure, but also in being derived entirely from the dermal layer of the skin (Fig. 37B). Since the epidermis plays no part in their development, enamel no longer enters into their make-up. The ancestors of prac- tically all living Bony Fishes, the Palaeoniscids, ranging from the Lower Devonian period to the end of the Jurassic, had the body completely invested in an armour of shining bony plates, arranged in regular parallel, oblique and longitudinal series (Fig. 127). These "ganoid" scales represent the most primitive type known in Bony Fishes, and persist to-day in the Bichirs (Polypterus) and in the Sturgeons {Acipenser) and their allies. Those of the Bichir take the form of juxtaposed plates, roughly rhomboid in shape, articulated with one another by a kind of peg-and-socket joint between the upper and lower edges of adjacent plates. Each scale is made up of three distinct layers; on the surface is a shining, enamel-like substance called ganoine (from which the term ganoid is derived) ; within is a thick layer of bone; and between the two another substance known as cosmine, containing minute blood-vqssels. Where flexibility of the body is a consideration the advantage of jointed plates over solid armour is obvious, and the actual shape of the first scales was probably determined mainly by purely mechanical factors. The scales are in close connection with the underlying muscles, and when the flexures of the trunk and tail in swimming cause these muscles to contract the skin tends to be wrinkled into definite circumscribed areas. In the existing Sturgeons [Acipenser), clearly descended from the Palaeoniscids mentioned above, the sole remains of the elaborate armour plating is a patch of small ganoid scales on the upturned part of the tail (Fig. 26a). There are, however, five widely separated rows of bony scutes or bucklers running SKIN, SCALES, AND SPINES 89 along the body from the head to the tail, which have the same structure as the rhomboid scales. In young fishes the scutes are all touching one another, and each is armed with a strong knife-like spine, but as growth proceeds they become separated and the spine disappears (Fig. 41 a). Between the scutes are obhque series of small denticles. The related Paddle-fishes, of which the curious Spoon-bill {Polyodon) of North America is best known (Fig. 470), are even more degenerate in their bodily armour, the skin being naked and the rudimentary scales of the tail still further reduced. The Gar Pikes (Lepidosteus) of the rivers and lakes of North America (Fig. 48), not to be confused with the marine Gar- d N\o Fig. 39. SCALES. a Portion of skin and isolated scales of the Gar Pike (Lepidosteus sp.), X about i ; b The same of the Bichir (Polypterus bichir), X i ; c. Isolated scute of large Sturgeon (Acipenser sp.), X i ; d. Cycloid scale of Tarpon (Megahps atlanticus), X i ; e. Ctenoid scale of Soldier-fish {Holocentrum ascensioms), X 8. fishes (Belonidae), have an armour of shining diamond-shaped plates similar in appearance to those of the Bichirs, but they differ in lacking the middle layer of cosmine (Fig. 39A). They are very much the same in other respects, and articulate with one another by peg-and-socket joints. Body armour composed of articulated rhomboid plates, although more efficient than soUd mail, has certain disad- vantages, and must restrict to some extent the flexures of the body m swimming. As fishes came to rely more and more upon their speed and agility rather than upon the strength of their armour for protection, the ganoid plates were gradually super- seded by thinner and more flexible structures. Mechanical factors may also have played some part in the change, for an 90 A HISTORY OF FISHES increase in the speed might be expected to lead to a change in the shape of the areas into which the skin is wrinkled. A study of extinct fishes provides a tolerably clear story, first of the gradual development and perfection of heavy armour, followed later by its gradual but no less definite decline when greater freedom of movement became all - important. Interesting transitional forms have been discovered, of which Aetheolepis from the Jurassic of Australia has retained the articulated ganoid plates on its relatively immobile trunk, whereas they have been transformed into thin, overlapping scales on the flexible tail. In other fossil forms all the scales are of the ganoid type, but whereas those on the body are articulated with one another, those in the tail region are quite free. The typical scaly covering in a modern Bony Fish may best be studied in such fishes as the Carp [Cyprinus) or the Perch (Perca). In the Carp (Fig. 43) the whole body with the ex- ception of the head and fins is protected by a number of regu- larly arranged, thin, flexible, bony plates known as cycloid scales, overlapping one another like the tiles on the roof of a house. Each scale, which is shaped roughly like the human finger-nail, has the front end inserted deep into a pouch in the dermal layer of the skin, the hinder portion being quite free. The overlapping (imbrication) of the scales is, from a mechani- cal point of view, important, and may be explained in the following manner. The muscles attached to the dermis tend to exert a somewhat unequal pull, and, therefore, to depress the scale areas, particularly at their front margins; in this way the growing scale is forced to lie obliquely, and at a later stage its hinder end appears through the skin (Fig. 37B). Only this free portion is covered by an epidermal membrane. In the Perch the scales are very much smaller, but have exactly the same arrangement. A closer study of a scale under the hand lens shows that its hinder end is provided with a row of small tooth-like spines instead of being smooth as in the Carp (Fig. 39e). Such a scale is known as ctenoid (comb-like). The arrangement of scales just described may be taken as typical of most Bony Fishes, but a large number of deviations from this occur, either in the direction of degeneration or of further specialisation. Cycloid (smooth) and ctenoid (spinate) scales are not so widely different as would appear from the above descriptions, as the one type is Hnked up with the other by an almost complete series of intermediate stages. For example, the posterior edge of a cycloid scale may be wavy SKIN, SCALES, AND SPINES 91 (crenulated), or the spines of a ctenoid scale may be soft and scarcely noticeable, in which case the scale is spoken of as ciliated. In some fishes the spines may extend on to the hinder free portion of the scale, giving it a roughened appearance. As a general rule, fishes with soft-rayed fins [e.g. Herring, Salmon, Roach) have cycloid scales, whereas the scales are ctenoid in the majority of Acanthopterygians {e.g. Perch, Bass) , but exceptions to this rule are numerous. Both types of scale may be developed on different parts of the body in the same fish. Thus in many of the Sea Perches [Epinephelus) the scales above the lateral line are mostly ctenoid and those below it cycloid, and in the Dab {Limanda) the spiny ones occur on the upper or coloured side, those on the blind or white side being quite smooth. The scales exhibit great diversity in shape in the different species, ranging from the roughly circular to the long oval. They also vary greatly in size. In the Tarpon [Megalops], for example, each scale is more than two inches in diameter, and these large structures are in some demand for ornamental work (Fig. 39d) . Those of the Mahseer {Barbus) , the famous game-fish of the rivers of India, are even larger, each being of the same size as the human palm. At the other extreme we have the minute cycloid scales of the Tunny {Thynnus) and Mackerel {Scomber), and the microscopic scales of the Common Eel {Anguilla) . In fishes of the Herring family (Herring, Sprat, Pilchard, Shad, etc.), the outer epidermal covering is very thin indeed, and the scales, which are placed in shallow pockets, appear to be lying on the surface of the body. Such scales are known as deciduous, because of the ease with which they are rubbed off when the fish is handled. In other fishes, of which the Plaice {Pleuronectes) will serve as an example, they are more or less deeply embedded in the skin. They are often also reduced in size, and instead of overlapping, remain quite separate from each other. The Common Eel {Anguilla) has a very slimy skin, which is, to all appearances, quite naked, but if a piece be examined under a microscope the presence of numerous minute scales embedded therein is revealed. This is clearly the result of degeneration as in the Plaice, and we are justified in assuming that these scales are the remnants of once much larger structures, which have gradually deteriorated as they ceased to be of importance to the fishes. It may be noted here that the ancient Hebrews, misled by the naked appearance of the Eel's skin, included this species among the fishes forbidden to them by Moses. His classification of the fishes into "all that 92 A HISTORY OF FISHES have fins and scales ye shall eat; and whatsoever hath not fins and scales ye shall not eat; it is unclean unto you" had a certain practical object, for it excluded all the Cat-fishes, which, although pleasant to the palate, were known to be unwhole- some and to cause diarrhoea and skin eruptions. These strict laws, forbidding as they did many plentiful and tasty species, naturally became gradually modified; fish with "at least two scales and one fin" were soon permitted, and, finally, any part of any fish on which traces of scales were visible ! The Dab (Limanda), with its ctenoid scales on the upper surface and cycloid scales below, has already been described. In other Flat-fishes the scales on the middle of the upper side are smooth, those on the head and near the edges of the body being spinate. In the Flounder [Flesus) most of the scales on the head, in the region of the lateral line, and also a series along the bases of the dorsal and anal fins, have been transformed into little thorny tubercles, generally stronger on the coloured side of the fish (Fig. 8b) ; the remainder of the body is covered with embedded cycloid scales, but in an allied species, the Diamond Flounder {Platichthys) of the Pacific coast of North America, these have been almost entirely suppressed, and the whole of the head and body is armed with irregularly scattered spiny tubercles (Fig. 40A). In another related form from Japan (Kareius) the tubercles are aggregated into clusters, and take the form of bony patches of varying size (Fig. 40c) . In another group of Flat-fishes {Bothidae) the closely related Turbot {Rhombus maximus) and Brill {R. laevis) have quite diflferent forms of scaly covering. In the Brill the body is armed with small cycloid scales, which are more or less overlapping, whereas the Turbot has a naked skin, but the coloured side bears a number of small, scattered, bony tubercles. The Black Sea Turbot {R. maeoticus), a distinct species, has very much larger tubercles, and these are developed on the lower as well as on the upper surface (Fig. 40B). Among other fishes with a somewhat similar armature, mention may be made of the Lump-sucker (Oyclopierus) , whose thick skin is studded with bony warts, some of which are enlarged to form a series of cone-like projections along the back and three rows on either side of the body (Fig. 35A). Two domesticated varieties of the Common Carp {Cyprinus carpio) produced by continental fish-culturists may be briefly described, since both these artificial forms exhibit modifications of the normal scaling. In the Mirror Carp [Spiegelkarpfen) there are one or two series of relatively enormous scales along the SKIN, SCALES, AND SPINES 93 middle of each side, and generally some smaller ones near the bases of the fins; all these are more or less widely separated from one another, and the rest of the body is naked (Fig. 143D). The Leather Car^D {Lederkarpfen) has a thick, roughened skin, which is entirely devoid of scales. In some of the Cat-fishes (6'z7wrozWffl) , Frog-fishes {Antennariidae) , Fig. 40. FISHES WITH TUBERCLES. A. Diamond Flounder {Platichthys stellatus), X i ; b. Black Sea Turbot {Rhombus maeoticus),X\\ C.Japanese Flounder (Kareius btcoloratus),X i ; D. Bat-fish (Ogcocephalus vesper tilio), X i- and in the Cling-fishes {Gobiesocidae) , the scales are reduced to soft papillae of the dermis. In many of the more speciahsed Flat- fishes (Soles, Tongue Soles, etc.) some of those on the under side of the head become transformed into tiny membranous filaments, which are very sensitive and act as organs of touch. Among other instances of specialisation may be mentioned the presence of bony scutes along the middle of each side in the region of the lateral fine. These may result from the modifica- tion of ordinary scales or may develop as entirely new struc- 94 A HISTORY OF FISHES tures. In the Scad or Horse Mackerel (Trachurus), a member of a large tribe of fishes known as Pampanos or Carangids (Caringidae), the lateral line is armed with a row of numerous keeled, bony shields, which in the tail region arc armed with sharp, knife-like spines (Fig. 41B). Somewhat similar spinous structures in this region occur in many of the Gurnards ( Trigla) . The members of a family of Cat-fishes found in the rivers of South America [Doradidae) possess a row of strong, bony scutes Fig. 41. — FISHES WITH SCUTES. A. Sturgeon (Acipenser sturio),x\ ; b. Pampano or Carangid (Caranx chrysos), xi; c. Unicorn-fish (Naseus brevirostris),X^ \ d. South American Cat-fish (Megalodoras irzvini), X | ; e. Three-spined Stickleback (Gasterosteus aculeatus), xi. along the middle of either side, each of which is armed with a sharp spine and bears a superficial resemblance to the bucklers of the Sturgeons (Fig. 41D). In the little "Tiddler" or Three-spined Stickleback {Gastero- steus) scales are wanting, but there is a series of large plates along each side, which in some specimens extends from the head to the tail, but in others is reduced to two or three plates behind the gill-opening (Figs. 28f; 41E). The Stickleback is equally at home in salt or in fresh water, and there seems to be some definite connection between the salinity of the water and SKIN, SCALES, AND SPINES 95 the development of the plates. As far as the British Isles are concerned, indi\'iduals from inland waters nearly always exhibit the reduced number of plates, whereas examples from Fig. 42. ARMOURED FISHES. A. Hassar or Cascadura {Callichthys littoralis), X ^ ; b. Pogge or Armoured Bullhead (Agonus cataphr actus), Y.l\ c. Mailed Cat-fish (Plecostomus garmani), xi; D. Trigger-fish {Balistes capnscus),xi ; e. Pine-cone Fish {Monocentris japonicus), X^ ; F. Trunk-fish (Lactophrys trigoniis), X ^ ; g. Porcupine-fish {Diodon hystrix), X i ; H. Shrimp-fish or Needle-fish {Aeoliscus strigatus), X ^. the sea are fully armed, intermediate types occurring in estuarine waters. Individuals from still more southerly localities may lack plates altogether. It has been seen that the past history of fishes reveals a story of gradual improvement in heavy armour, followed by a definite 96 A HISTORY OF FISHES decrease in its strength, and those Bony Fishes which have returned to the mail of their ancestors (although of a totally different structure) may now be considered. As a general rule, these are sluggish creatures, which have sacrificed speed and agility and have come to depend on their armour for protection from their enemies. The South American Cat-fishes, known as Hassars or Cascaduras (Callichthys) , have the body completely encased in mail, made up of a double row of broad over- lapping shields on each side. The arrangement of the shields is metameric, there being one pair to each vertebra or muscle segment. AUied to the Hassars are the Mailed Cat-fishes or Loricariids {Loricariidae) , a large and varied family confined to the rivers of Central and South America. The body is pro- tected above and on the sides by series of bony plates (Fig. 42c), the chest and abdomen being either naked or covered with much smaller plates. The plates on the sides have a metameric arrangement, and may be more or less sharply keeled or variously armed with small spines set in sockets (Fig. 42c). These smaller spines are of special interest, for when examined microscopically they are seen to be formed of dentine capped with enamel; thus, they are essentially similar in structure to the dermal denticles of the Selachians, and this represents another of the rare examples of the course of evolution having been reversed. The Mailed Cat-fishes are sluggish creatures, spending most of their time attached to stones or other objects at the bottom of a stream, and the bony armour provides them with an efficient protection against enemies. In some very closely related fishes from the mountain streams of the Andes (Cyclopiiim), where the absence of carnivorous fishes renders armour superfluous, the scutes have disappeared and the skin is quite naked (Fig. 93). Among other Bony Fishes with the body more or less completely cuirassed with bony shields, the Sea Robins of deep water {Peristedion), and the curious little Pogge or Bullhead (Agonus) found round our own coasts (Fig. 42b), may be mentioned. Among the members 'of the large and diverse order of fishes known as Tube-mouths {Solenichthyes) there occur several interesting modifications of the scaly covering. In the Snipe- fishes (Macrorhamphosus) , for example, each scale consists of a bony basal plate, rhomboid in shape, which is produced into a curved and backwardly directed spine, containing a definite pulp-cavity, reminiscent of that of the Selachian denticle. In the remarkable Shrimp-fishes [Centriscidae) the SKIN, SCALES, AND SPINES 97 whole head and body is strongly compressed, and is com- pletely encased in a transparent bony cuirass with a knife-like lower edge. This is made up of a number of thin plates, fused with the underlying ribs in much the same way as the carapace of a tortoise (Fig. 42H). In the Pipe-fishes {Syngnathidae) the scales are replaced by a series of jointed bony rings, encircling the body from behind the head to the tip of the tail. Similar rings surround the prehensile tail of the Sea Horse {Hippocampus), but in the immobile trunk region they take the form of plates which are roughly cruciform in shape, and are interlaced with one another to form a complete outer skeleton. The edges of these plates are not infrequently produced into pointed spines or rounded knobs (Fig. 5E). The bottom-Hxing Bat-fishes {0 gcocephalus) have the upper surface of the body studded with spines or tubercles, not unlike the scales of the Snipe-fishes in structure, but lacking the pulp cavity, the projecting spine being soHd (Fig. 400). Similar tubercles are sometimes found in the deep-sea Angler- fishes or Ceratioids, but these fishes mostly have a naked skin. In the Trigger-fishes (Balistes) the rough scales covering the body are Hke those of the Bat-fishes, the basal plate often being rhomboid in shape, with the outer surface roughened or armed with one or more small spines (Fig. 420). In the File-fishes (Monacanthus) the spines are more numerous, and are set so close together as to give the skin the appearance of velvet. In the allied family of Trunk-fishes (Box-fishes, Coflfer-fishes, Cuckolds) a complete and sohd coat of mail again occurs. The scales are represented by large six-sided plates, united with one another to form a strong, bony box, from one end of which projects the mouth and from the other the naked tail. This box may be three, four, or even five-sided, and one or more of its edges may be armed with strong spines (Figs. 5B; 42F). A West Indian species {Lactophrys tricornis) , with two long spines projecting forward from the forehead, is appropriately named the Cow- fish. The little Pine-cone fish (Mo/zoc^wto), although pertaining to a totally different order of fishes, is another form in which the thick scales unite to enclose the body in a sort of box (Fig. 42E). The Surgeon-fishes ( Teuthidae) of tropical seas derive their name from the presence of a lancet-like spine on either side of the fleshy part of the tail. When not required this is retracted into a sheath in the skin, but can be quickly turned outwards and forms an effective weapon when the fish lashes its tail from side to side. In the Puflfers or Globe-fishes {Tetrodontidae) the scales are 98 A HISTORY OF FISHES replaced by small, movable spines, which stand erect when the body is inflated with air. The Porcupine-fishes and Burr-fishes [Diodontidae) have an even stronger protection, the roots of the long, stout spines coming into contact with one another and providing a more or less continuous coat of mail. In some species these spines are two-rooted and movable, so that they can be laid back flat or erected at will; in other forms they are three-rooted and fixed (Fig. 420). The Lung-fishes {Dipneusti) are the sole living representatives of a once large and important group of fishes, in which the scales have a structure different to those of all other Bony Fishes. The primitive members of this group had a covering of ganoid scales, which must have been derived from plates similar to those of the Palaeoniscids, but the cosmine layer has a special regular structure and the ganoine is reduced to a thin superficial covering. Of the existing forms, the Australian genus (Epiceratodus) has large, thin, overlapping cycloid scales, whereas in the African {Protopterus) and South American [Lepidosiren) forms the scales are very much smaller, completely embedded in the skin, and more or less separated from one another (Fig. 99). With few exceptions, the scales of Bony Fishes have a regular arrangement, and within certain limits both their size and dis- position is constant for any given species. For this reason, a count of the number of scales is often of some importance in identifying any particular fish. Generally, the scales are arranged in obliquely transverse series, and the number of these series is counted along the middle of the side from behind the gill-opening to the base of the caudal fin. In estimating the number of scales across the body {i.e. the number of longitudinal rows) they are usually counted in one of the transverse series, as a rule that which runs from the commencement of the dorsal fin downwards and forwards to the lateral line, and from thence downwards and backwards to the pelvic fin (Fig. 43). Thus the scale formula for a particular species may be 6-7 written : 44-47 — '— . This means that there are from 44 to 47 9-10 scales in a longitudinal series from the head to the tail, 6 or 7 between the origin of the dorsal fin and the lateral line, and 9 or 10 between the latter and the base of the pelvic fin. The Salmon {Salmo salar) and Trout [S. trutta) of our own country provide an example of the importance of scale-counts in dis- tinguishing closely related species. These fishes are extremely similar in most characters, and although there are a number of SKIN, SCALES, AND SPINES 99 points which enables an expert to tell them apart, by far the most reliable method consists in counting the number of scales on the tail, a sure means of identifying almost any specimen. In the Salmon the number in an oblique series from the hinder edge of the adipose fin downwards and forwards to the lateral line varies from 10 to 13, but in the Trout this number ranges from 13 to 16. Mention may be made here of the so-called hybrid between the Pilchard (Sardina) and the Herring (Clupea), specimens of which turn up from time to time. At first sight this fish appears to have about 30 rows of scales along one side of the body, and more than 50 on the other. This is not, of course, a genuine hybrid, and the explanation of the abnormality is that the scales of the Pilchard are unequal in size, the oblique rows being Fig- 43- Carp (Cyprinus carpio), X^. To show arrangement of scales. alternately of larger and smaller scales, the latter being quite con- cealed by the former in normal fish. In the so-called hybrid all the scales of one side of the body are equal in size and regularly arranged, while those of the other are large and small as usual. In many Bony Fishes there is an enlarged and somewhat modified scale in the angle where the upper edge of the pectoral fin joins the body, and often a similar axillary or accessory scale in the outer angle of the pelvic fin. It usually takes the form of a pointed dagger-like process, sometimes stifi:^ and hard, sometimes quite soft and flexible. As a rule, it is present in the more generalised forms, but is lacking in the more specialised fishes. Again, it is often well developed in actively swimming forms, and absent or much reduced in those living at or near the bottom, a fact which suggests that it is in some way con- loo A HISTORY OF FISHES nected with the locomotor activities of the fish. In the Salmon and Trout it is surrounded by definite fatty tissue, and is supported at its base by a splint of bone connected with the outermost ray of the pelvic fin. The lateral line, a conspicuous feature of most Bony Fishes, will be dealt with in detail in the chapter devoted to sense organs {cf. p. 198), and it will suffice to point out here that it consists of continuous grooves or canals in the head and body containing special sensory organs; at intervals these grooves communicate with the exterior through pores or by little tubes which run outwards through the scales (Fig. 79d). It is these pores or modified tube-bearing scales which form the charac- teristic external line, generally running from behind the head to the base of the caudal fin, and not infrequently continued on to the fin itself It may run more or less straight along the side as in the Trout (Salmo) or Carp {Cyprinus) ; it may be curved upwards to follow the line of the back as in the Perch (Perca) , or downwards and parallel with the line of the belly as in the Roach (Rutilus) or Bream [Abramis). In the Parrot-fishes (Scaridae) and others (Fig. 44a) it is disconnected, the upper portion ending abruptly below the soft dorsal fin, and the lower portion commencing again below it and running back- wards to the tail in the usual manner. In the Greenlings (Hexagrammidae) of the North Pacific (Fig. 44b) there may be several lines on each side of the upper part of the body. In the Tongue Soles (Cynoglossidae) there may be one, two or three lateral lines on the upper surface of the body and one, two or none on the lower, in addition to a complicated system of lines on the head (Fig. 44f). The continuation of the lateral line system on to the head occurs in most fishes, but where it is formed of deep-seated canals running along the surface of, or even perforating the bones, it is externally quite inconspicu- ous in this region. In many fishes, notably in the Flat-fishes {Heterosomata) , the line runs straight from the tail to the tip of the pectoral fin, and then forms a more or less well-defined arch above the fin itself (Fig. 44e). In certain groups (Gobies, Cyprinodonts) the surface structures of the lateral line are entirely wanting. In the Sharks it is represented by a simple groove protected by overlapping shagreen denticles. It is sometimes of considerable importance to be able to determine the age of a particular fish, and especially of those fishes which form the national food supply; this can be carried out, in some fishes at least, by what is known as scale-reading. SKIN, SCALES, AND SPINES [OI The discovery, made some years ago by Mr. H. W. Johnston, that every Salmon carries its own hfe-history clearly written on each one of its scales, has proved to be of incalculable value to scientific men, and this method of age-det