Ichthyology

ICHTHYOLOGYICHTHYOLOGY is that branch of zoology which treats of the internal and external structure of fishes, their mode of life, and their distribution in space and time. According to the views generally adopted at present, all those Vertebrate animals are referred to the class of Fishes which combine the following characteristics:—they live in water, and by means of gills or branchiae breathe air dissolved in water; the heart consists of a single ventricle and single atrium; the limbs, if present, are modified into fins, supplemented by unpaired median fins; and the skin is either naked, or covered with scales or with osseous plates or bucklers. With few exceptions fishes are oviparous. There are, however, as we shall see hereafter, not .a few members of this class which show a modification of one or more of these characteristics, and which, nevertheless, cannot be separated from it. The distinction between the class of Fishes and that of Batrachians is very slight indeed.

History and Literature.

The commencement of the history of ichthyology coincides with that of zoology generally. Aristotle (384-322 b.c. ) had a perfect knowledge of the general structure of fishes, which he clearly discriminates both from the aquatic animals with lungs and mammae, i.e., Cetaceans, and from the various groups of aquatic invertebrates. According to him, “the special characteristics of the true fishes consist in the branchiae and fins, the majority having four fins, but those of an elongate form, as the eels, having two only. Some, as the Huraena, lack the fins altogether. The rays swim with their whole body, which is spread out. The branchiae are sometimes furnished with an operculum, sometimes they are without one, as in the cartilaginous fishes. . . . No fish has hairs or feathers; most are covered with scales, but some have only a rough or a smooth skin. The tongue is hard, often toothed, and sometimes so much adherent that it seems to be wanting. The eyes have no lids, nor are any ears or nostrils visible, for what takes the place of nostrils is a blind cavity; nevertheless they have the senses of tasting, smelling, and hearing. All have blood. All scaly fishes are oviparous, but the cartilaginous fishes (with the exception of the sea-devil, which Aristotle places along with themj are viviparous. All have a heart, liver, and gall-bladder; but kidneys and urinary bladder are absent. They vary much in the structure of their intestines: for, whilst the mullet has a fleshy stomach like a bird, others have no stomachic dilatation. Pyloric caeca are close to the stomach, and vary in number; there are even some, like the majority of the cartilaginous fishes, which have none whatever. Two bodies are situated along the spine, which have the function of testicles; they open towards the vent, and are much enlarged in the spawning season. The scales become harder with age. Not being provided with lungs, fishes have no voice, but several can emit grunting sounds. They sleep like other animals. In most cases the females exceed the males in size; and in the rays and sharks the male is distinguished by an appendage on each side of the vent.”

Aristotle’s information on the habits of fishes, their migrations, mode and time of propagation, and economic uses, is, so far as it has been tested, surprisingly correct. Unfortunately, we too often lack the means of recognizing the species of which he gives a description. His ideas of specific distinction were as vague as those of the fishermen whose nomenclature he adopted; it never occurred to him that vernacular names are subject to change, or may be entirely lost in course of time, and the difficulty of identifying his species is further increased by the circumstance that sometimes several popular names are applied by him to the same fish, or different stages of growth are designated by distinct names. The number of fishes known to Aristotle seems to have been about one hundred and fifteen, all of which are inhabitants of the Aegean Sea.

That one man should have discovered so many truths, and laid so sure a basis for future progress in zoology, is less surprising than the fact that for about eighteen centuries a science which seemed to offer particular attractions to men gifted with power of observation was no farther advanced. Yet such is the case. Aristotle’s disciples, as well as his successors, remained satisfied to be his copiers or commentators, and to collect fabulous stories or vague notions. With very few exceptions (such as Ausonius, who wrote a small poem, in which he describes from his own observations the fishes of the Moselle) authors entirely abstained from original research; and it was not until about the middle of the 16 th century that ichthyology made a new step in advance by the appearance of Belon, Rondelet, and Salviani, who almost simultaneously published their great works, by which the idea of species was established definitely and for all time.

P. Belou travelled in the countries bordering on the eastern part of the Mediterranean, in the years 1547-50; he collected rich stores of positive knowledge, which he embodied in several works. The one most important for the progress of ichthyology is that entitled De aquatilibus libri duo, Paris, 1553. Belon knows about one hundred and ten fishes, of which he gives rude but generally recognizable figures. In his descriptions he pays regard to the classical as well as to the vernacular nomenclature, and states the outward characteristics, sometimes even to the number of fin rays; frequently also he gives the most conspicuous anatomical peculiarities. Although Belon but rarely gives definitions of the terms used by him, it is not generally very difficult to ascertain the limits which he intended to assign to each division of aquatic animals. He very properly divides them into such as are provided with blood and those without it,—two divisions corresponding in modern language to vertebrate and invertebrate aquatic animals. The former are classified by him according to size, the further subdivisions being based on the structure of the skeleton, mode of propagation, number of limbs, form of the body, and physical character of the habitat.

The work of the Roman ichthyologist, H. Salviani (1514-72), bears evidence of the high social position which the author held as physician to three popes. Its title is Aquatilium animalium histoma, Rome, 1554-57, fol. It treats exclusively of the fishes of Italy. Ninety-two species are figured on seventy-six plates, which, as regards artistic execution, are masterpieces of that period, although those specific characteristics which nowadays constitute the value of a zoological drawing were entirely overlooked by the author or artist. No attempt is made at a natural classification, but the allied forms are generally placed in close proximity. The descriptions are quite equal to those given by Belon, entering much into the details of the economy and uses of the several species, and were evidently composed with the view of collecting in a readable form all that might prove of interest to the class of society in which the author moved. Salviani’s work is of a high order, very .remarkable considering the age in which he lived. It could not fail to convey valuable instruction, and to render ichthyology popular in the country to the fauna of which it was devoted, but it was not fitted to advance ichthyology as a science generally; in this respect Salviani is not to be compared with Rondelet or Belon.

G. Rondelet (1507-57) had the great advantage over Belon of having received a medical education at Paris, and especially of having gone through a complete course of instruction in anatomy as a pupil of Guentherus of Andernach. This is conspicuous throughout his works— Libri de piscibus marinis, Lyons, 1554; and Universes aquatilium historice pars altera, Lyons, 1555. Nevertheless they cannot be regarded as more than considerably enlarged editions of Belon’s work. For, although he worked independently of the latter, and differs from him in numerous details, of which he had a much more extensive knowledge, the system adopted by him is characterized by the same absence of the true principles of classification. His work is almost entirely limited to European and chiefly to Mediterranean forms, and comprises no less than one hundred and ninety-seven marine and forty-seven freshwater fishes. His descriptions are more complete and his figures much more accurate than those of Belon; and the specific account is preceded by introductory chapters, in which he treats in a general manner of the distinctions, the external and internal parts, and the economy of fishes. Like Belon, he had no conception of the various categories of classification—confounding, for instance, throughout his work the terms “genus’’and “species”; but he had an intuitive notion of what his successors called a “species,” and his principal object was to collect and give as much information as possible regarding such species.

For nearly a century the works of Belon and Rondelet continued to be the standard works on ichthyology; but the science did not remain stationary during that period. The attention of naturalists was now directed to the fauna of-foreign countries, especially of the Spanish and Dutch possessions in the New World; and in Europe the establishment of anatomical schools and academies led to careful investigation of the internal anatomy of the most remarkable European forms. Limited as these efforts were as to their scope, being restricted either to the fauna of some particular district or to the dissection of a single species, they were sufficiently numerous to enlarge the views of naturalists, and to destroy that fatal dependence on preceding authorities which had continued to keep in bonds the minds of such men even as Rondelet and Belon. The most noteworthy of those engaged in these inquiries in tropical countries were W. Piso and G. Margrav, who accompanied as physicians the Dutch governor, Prince Maurice of Nassau, to Brazil (1637-44).

Of the men who left records of their anatomical researches, we may mention Borelli (1608-79), who wrote a work De motu animalium, Rome, 1680, 4to, in which he explained the mechanism of swimming and the function of the air-bladder; Μ. Malpighi (1628-94), who examined the optic nerve of the sword-fish; the celebrated J. Swammerdam (1637-80), who described the intestines of numerous fishes; and J. Duverney (1648-1730), who investigated in detail the organs of respiration.

A new era in the history of ichthyology commences with Ray, Willughby, and Artedi, who were the first to recognize the true principles by which the natural affinities of animals should be determined. Their labours stand in so intimate a connexion with each other that they represent but one great step in the progress of this science.

J. Ray (1628-1705) was the friend and guide of F. Willughby (1635-72). They found that a thorough reform in the method of treating the vegetable and animal kingdoms had become necessary; that the only way of bringing order into the existing chaos was by arranging the various forms according to their structure; that they must cease to be burdened with inapplicable passages and quotations from ancient writers, and to perpetuate the vague and erroneous notions of their predecessors. They therefore substituted facts for speculation, and one of the first results of this change, perhaps the most important, was that, having recognized “species” as such, they defined the term, and fixed it as the starting point of all sound zoological knowledge.

Although they had divided their work so that Ray attended to the plants principally, and Willughby to the animals, the Historia piscium, Oxf., 1686, which bears Willughby’s name on the title page, and was edited by Ray, is clearly their joint production. A great part of the observations contained in it were collected during the journeys they made together in Great Britain and in the various countries of Europe; and it is no exaggeration to say that at that time these two Englishmen knew the fishes of the Continent, and especially those of Germany, better than any native zoologist.

By the definition of fishes as animals with blood, breathing by gills, provided with a single ventricle of the heart, and either covered with scales or naked, the Cetaceans are excluded. The fishes proper are arranged primarily according to the cartilaginous or the osseous nature of the skeleton, and then subdivided according to the general form of the body, the presence or the absence of ventral fins, the soft or the spinous structure of the dorsal rays, the number of dorsal fins, &c. No fewer than four hundred and twenty species are thus arranged and described, of which about one hundred and eighty were known to the authors from personal examination,—a comparatively small proportion, but descriptions and figures still formed in great measure the substitute for our modern collections and museums. With the increasing accumulation of forms, the want of a fixed nomenclature had become more and more felt.

Peter Artedi would have been a great ichthyologist if Ray or Willughby had not preceded him. But he was fully conscious of the fact that both had prepared the way for him, and therefore he did not fail to reap every possible advantage from their labours. Born in 1705 in Sweden, he studied with Linnaeus at Upsala; from an early period he devoted himself entirely to the study of fishes, and was engaged in the arrangement and description of the collection of Seba, a wealthy Dutchman who had formed what was perhaps the richest museum at that time, when he was accidentally drowned in one of the canals of Amsterdam in the year 1734, at the age of twenty-nine. His manuscripts were fortunately secured by an Englishman, Count Clifford, and edited by his early friend Linnaeus. The work is divided into the following parts

(1) In the Bibliotheca Ichthyologien Artedi gives a very complete list of all preceding authors who had written on fishes, with a critical analysis of theiι , works. (2) The Philosophia Ichthyologien is devoted to a description of the external and internal parts of fishes; Artedi fixes a precise terminology for all the various modifications of the organs, distinguishing between those characters which determine a genus and such as indicate a species or merely a variety; in fact he establishes the method and principles which subsequently have guided every systematic ichthyologist. (3) The Genera Piscium contains well-defined diagnoses of forty-five genera, for which he has fixed an unchangeable nomenclature. (4) In the Species Piscium descriptions of seventy-two species, examined by himself, are given,— descriptions which even now are models of exactitude and method. (5) Finally, in the Synonymia Piscium references to all previous authors are arranged for every species, very much in the manner which is adopted in the systematic works of the present day.

Artedi has been justly called the Father of Ichthyology. So admirable was his treatment of the subject, that even Linnaeiis could only modify and add to it. Indeed, so far as. ichthyology is concerned, Linnaeus has scarcely done anything beyond applying binominal terms to the species properly described and classified by Artedi. His classification of the genera appears in the 12th edition of the Systema thus :—

A. Amp>hibia Nantes.—Spiracτdis compositis. — Petromyzon, Eaia, Squalus, Chimaera. Spiraculis solitariis.— Lophius, Acipenser, Cycloptenis, Balistes, Ostracion, Tetrodon, Diodon, Centriscus, Syngnathus, Pegasus.

B. Pisces Apodes. — Muraena, Gymnotus, Trichiurus, Anarrhichas, Ammodytes, Ophidium, Stromateus, Xiphias.

C. Pisces Jugulares.— Callionymus, Uranoscopus, Trachinus, Gadus, Biennius.

D. Pisces Thoraci:i.— Cepola, Echeneis, Coryphaena, Gobius, Cottus, Scorpaena, Zeus, Pleuronectes, Chaetodon, Sparus, Labnis, Sciaena, Perca, Gasterosteus, Scomber, Mullus, Trigla.

E. Pisces Abdominales.— Cobitis, Amia, Silurus, Teuthis, Loricaria, Salmo, Fistularia, Esox, Elops, Argentina, Atherina, Mm il, Mormyrus, Exocoetus, Polynemus, Clupea, Cyprinus.

Two contemporaries of Linnaeus, L. T. Gronow and J. T. Klein, attempted a systematic arrangement of fishes; both had considerable advantages for the study, especially in possessing extensive collections; but neither exercised any influence on the progress of ichthyology.

The works of Artedi and Linnaeus led to an activity of research, especially in Scandinavia, Holland, Germany, and England, such as has never been equalled in the history of biological science. Whilst some of the pupils and followers of Linnaeus devoted themselves to the examination and study of the fauna of their native countries, others proceeded on voyages of discovery to foreign and distant lands. Of these latter the following may be especially mentioned :—O. Fabricius worked out the fauna of Greenland; Kalm collected in North America, Hasselquist in Egypt and Palestine, Brünn ich in the Mediterranean, Osbeck in Java and China, Thunberg in Japan; Forskâl examined and described the fishes of the Red Sea; Steller, Pallas, S. T. Gmelin, and Güldenstedt traversed nearly the whole of the Russian empire in Europe and Asia. Others attached themselves as naturalists to the celebrated circumnavigators of the last century, such as the two Forsters (father and son) and Solander, who accompanied Cook; Commerson, who travelled with Bougainville; and Sonnerat. Numbers of new and remarkable forms were discovered by those men, and the foundation was laid for a knowledge of the geographical distribution of animals.

Of those who studied the fishes of their native countries, the most celebrated were Pennant (Great Britain), O. F. Müller (Denmark), Duhamel (France), Meidinger (Austria), Cornide (Spain), and Parra (Cuba).

The mass of materials brought together by those and other zoologists was so great that, not long after the death of Linnaeus, the necessity made itself felt for collecting them in a compendious form. Several compilers undertook this task; they embodied the recent discoveries in new editions of the classical works of Artedi and Linnaeus, but, not possessing either a knowledge of the subject or any critical discernment, they only succeeded in burying those noble monuments under a chaotic mass of rubbish. For ichthyology it was fortunate that two men at least, Bloch and Lacépède, made it a subject of prolonged original research.

Mark Eliezer Bloch (1723-1799), a physician of Berlin, had reached the age of fifty-six when he commenced to write on ichthyological subjects. To begin at his time of life a work in which he intended not only to give full descriptions of the species known to him from specimens or drawings, but also to illustrate each species in a style truly magnificent for his time, was an undertaking the execution of which most men would have despaired of. Yet he accomplished not only this task, but even more than he at first contemplated.

His work consists of two divisions :—(1) Oeconomische Naturgeschichte der Fische Deutschlands, Berl., 1782-84; (2) Naturgeschichte der ausländischen Fische, Bert, 1785- 95. The first division, which is devoted to a description of the fishes of Germany, is entirely original, and based upon the author’s own observations. His descriptions as well as figures were made from nature, and are, with but few exceptions, still serviceable; indeed many continue to be the best existing in literature. Bloch was less fortunate, and is much less reliable, in his natural history of foreign fishes. For many of the species he had to trust to more or less incorrect drawings and descriptions by travellers; frequently, also, he was deceived as to the origin of specimens which he acquired by purchase. Hence his accounts contain numerous confusing errors, which it would have been difficult to correct had not nearly the whole of the materials on which his work is based been preserved in the collections at Berlin.

After the completion of his great work Bloch occupied himself with systematizing. He prepared a general system of fishes, in which he arranged not only those described in his great work, but also those with which he had afterwards become acquainted from the descriptions of others. The work was ably edited and published after Bloch’s death by a philologist, J. G. Schneider, under the title Μ. E. Βlochii Systema ichthyologioe iconibus CX. illustratum, Berl., 1801. The number of species enumerated in it amounts to 1519. The system is based upon the number of the fins, the various orders being termed Hende- capterygii, Decapterygii, &c. We need not add that an artificial method like this led to the most unnatural combinations and distinctions.

Bloch’s Naturgeschichte remained for many years the standard work, and, with its great number of excellent illustrations, proved a most useful guide to the student. But as regards originality of thought Bloch was far surpassed by his contemporary, B. G. E. de Lacépède, born at Agen, in France, in 1756, a man of great and general erudition, who became professor at the museum of natural history in Paris, where he died in 1826.

Lacépède had to contend with great difficulties in the preparation of his Histoire des Poissons, Paris, 1798-1803, 5 vols., which was written during the most disturbed period of the French Revolution. A great part of it was composed whilst the author was separated from collections and books, and had to rely on his notes and manuscripts only. Even the works of Bloch and other contemporaneous authors remained unknown, or at least inaccessible, to him for a long time. We need not therefore be surprised that his work abounds in the kind of errors into which a compiler is liable to fall. Not only does the same species appear under two or more distinct specific names, but it sometimes happens that the author so little understands the source from which he derives his information that the description is referred to one genus and the accompanying figure to another. The names of genera are unduly multiplied; and the figures with which the work is illustrated are far inferior to those of Bloch. Thus the influence of Lacépède on the progress of ichthyology was vastly less than that of his fellow-labourer; and the labour laid on his successors in correcting the numerous errors into which he had fallen probably outweighed the assistance which they derived from his work.

The work of the principal cultivators of ichthyology in the period between Ray and Lacépède was chiefly systematizing and describing; but the internal organization of fishes also received attention from more than one great anat >mist. Haller, Camper, and Hunter examined the nervous system and the organs of sense; and above all Alexander Monro, secundus, published a classical work, The Structure and Physiology of Fishes explained and compared with those of Man and other Animals, Edin., 1785. The electric organs of fishes {Torpedo and Gymnotus) were examined by Réaumur, Allamand, Bancroft, Walsh, and still more exactly by J. Hunter. The mystery of the propagation of the eel called forth a large number of essays, and even the artificial propagation of Salmonid oe was known and practised by Gleditsch (1764).

Bloch and Lacépède’s works were almost immediately succeeded by the labours of Cuvier, but his early publications were of necessity tentative, preliminary, and fragmentary, so that some little time elapsed before the spirit infused into ichthyology by this great anatomist could exercise its influence on all the workers in this field. Several of such ante-Cuvierian works must be mentioned on account of their importance to our knowledge of certain faunas. The Descriptions and Figures of Two Hundred Fishes collected at Vizagapatam on the Coast of Coromandel, Lond., 1803, 2 vυls., by Patrick Russel, and An Account of the Fishes found in the Fiver Ganges and its Branches, Edin., 1822, 2 vols., by F. Hamilton (formerly Buchanan), were works distinguished by greater accuracy of the drawings (especially the latter) than was ever attained before. A Natural History of British Fishes was published by E. Donovan, Lυnd., 1802-8; and the Mediterranean fauna formed the study of the lifetime of A. Risso {Ichthyologie de Nice, Paris, 1810; and Histoire naturelle de Γ Europe Méridionale, Paris, 1827). A slight beginning in the description of the fishes of the United States was made by S. L. Mitchell, who published, besides various papers, a Memoir on the Ichthyology of New York, in 1815.

G. Cuvier (1769-1832) did not occupy himself with the ( study of fishes merely because the class formed part of the Fegne Animal, but devoted himself to it with particular predilection. The investigation ot their anatomy, and especially of their skeleton, was taken up by him at an early period, and continued until he had succeeded in completing so perfect a framework of the system of the whole class that his immediate successors required only to fill up those details for which their master had had no leisure. Indefatigable in examining all the external and internal characters of the fishes in a rich collection, he ascertained the natural affinities of the infinite variety of forms, and accurately defined the divisions, orders, families, and genera of the class, as they appear in the various editions of the Fègne Animal. His industry equalled his genius: he formed connexions with almost every’accessible part of the globe; not only French travellers and naturalists, but also Germans, Englishmen, Americans, rivalled one another in assisting him with collections; and for many years the museum of the Jardin des Plantes was the centre where all ichthyological treasures were deposited. Thus Cuvier brought together a collection the like of which had never been seen before, and which, as it contains all the materials on which his labours were based, must still be considered as the most important. Soon after the year 1820, Cuvier, assisted by one of his pupils, A. Valenciennes, commenced his great work on fishes, Histoire naturelle des Poissons, of which the first volume appeared in 1828. The earlier volumes, in which Cuvieτ himself took his share, bear evidence of the enthusiasm with which both authors devoted themselves to their task. After Cuvier’s death in 1832, the work was left entirely in the hands of Valenciennes, whose energy and interest gradually slackened, rising to their former pitch in some parts only, as, for instance, in the treatise on the herring. He left the work unfinished with the twenty-second volume (1848), which treats of the Salmonoids. Yet, incomplete as it is, it is indispensable to the student.

The system finally adopted by Cuvier is the following :— [table]

We have only to compare this system with that of Linnaeus if we wish to measure the gigantic stride made by ichthyology during the intervening period of seventy years. The various characters employed for classification have been examined throughout the whole class, and their relative importance has been duly weighed and understood. Though Linnaeus had formed a category of “Amphibia Nantes” for fishes with a cartilaginous skeleton, which should coincide with Cuvier’s “Poissons Cartilagineux,” he had failed to understand the very nature of cartilage, apparently comprising under this term any skeletal framework of less firmness than ordinary bone. Hence he considered Lophius, Cyclopterus, Syngnathus to be cartilaginous fishes. Adopting the position and development of the ventral fins as a highly important character, he had been obliged to associate fishes having rudimentary and inconspicuous ventral fins, like Trichiurus, Xiphias, &c., with the true eels. The important category of “family” appears now in Cuvier’s system fully established as intermediate between genus and order. Important changes in Cuvier’s system have been made and proposed by his successors, but in the main it is still that of the present day.

Cuvier had extended his researches beyond the living forms, into the field of palaeontology; he was the first to observe the close resemblance of the scales of the fossil Palaeoniscus to those of the living Polypterus and Lepidosteus, the prolongation and identity of structure of the upper caudal lobe in Palaeoniscus and the sturgeons, the presence of peculiar “fulcra” on the anterior margin of the dorsal fin in Palaeoniscus and Lepidosteus, and inferred from these facts that the fossil genus was allied either to the sturgeons or to Lepidosteus. But it did not occur to him that there was a close relationship between those recent fishes. Lepidosteus and, with it, the fossil genus remained in his system a member of the order of Halacopterygii abdominales.

It was left to L. Agassiz (1807-73) to point out the importance of the structure of the scales as a characteristic, and to open a path towards the knowledge of a whole new subclass of fishes, the Ganoidei.

Impressed with the fact that the peculiar scales of Polypterus and Lepidosteus are common to all fossil osseous fishes down to the Chalk, he takes the structure of the scales generally as the base for an ichthyological system, and distinguishes four orders :—

1. Ρlαcοids.— Without scales proper, but with scales of enamel, sometimes large, sometimes small, ami reduced to mere points (Rays, Sharks, and Cyclostomi, with the· fossil Hybodontes). 2. Ganoids. —With angular bony scales, covered with a thick stratum of enamel: to this order belong the fossil Lepidoides, Sauroides, Pycnodontes, and Coelacanthi; the recent Polypterus, Lepidosteus, Sclerodermi, Gymnodontes, Lophobranches, and Siluroιdes; also the Sturgeons. 3. Glenoids.— With rough scales, which have their free margins denticulated: Chaetodontidae, Pleuronectidae, Percidae, Polyacantlιi, Sciaenidae, Sparidae, Scorpaenidae, Aulostomi. 4. Cycloids. —With smooth scales, the hind margin of which lacks denticulation: Labridae, Mugilidae, Seombridae, Gadoidei, Gobiidae, Muraenidae, Lucioidei, Salmonidae, Clupeidae, Cyprinidae.

We have no hesitation in affirming that if Agassiz had had an opportunity of acquiring a more extensive and intimate knowledge of existing fishes before his energies were absorbed in the study of fossil remains, he would himself have recognized the artificial character of his classification. The distinctions between cycloid and ctenoid scales, between placoid and ganoid fishes, are vague, and can hardly be maintained. So far as the living and post-Cretacean forms are concerned, he abandoned the vantageground gained by Cuvier; and therefore his system could never supersede that of his predecessor, and finally shared the fate of every classification based on the modifications of one organ only. But Agassiz has the merit of having opened an immense new field of research by his study of the infinite variety of fossil forms. In his principal work, Recherches sur les Poissons fossiles, Neuchâtel, 1833- 43, 4to, atlas in fol., he placed them before the world arranged in a methodical manner, with excellent descriptions and illustrations. His power of discernment and penetration in determining even the most fragmentary remains is truly astonishing; and, if his order of Ganoids is an assemblage of forms very different from what is now understood by that term, he was at any rate the first who recognized that such an order of fishes exists.

The discoverer of the Ganoidei was succeeded by their explorer, Johannes Müller (1801-58). In his classical memoir Geber den Bau und die Grenzen der Gaηoideη, Berl., 1846, he showed that the Ganoids differ from all the other osseous fishes, and agree with the Plagiostomes, in the structure of the heart. By this primary character, all heterogeneous elements as Siluroids, Osteoglossidae, Ac., were eliminated from the order as understood by Agassiz. On the other hand, he did not recognize the affinity of Lepidosiren to the Ganoids, but established for it a distinct subclass, Dipnoi, which he placed at the opposite end of the system. By his researches into the anatomy of the lampreys and Amphioxus, their typical distinctness from other cartilaginous fishes was proved; they became the types of two other subclasses, Cyclostomi and Leplocardii.

Müller proposed several other not unimportant modifications of the Cuvierian system; and, although all cannot be maintained as the most natural arrangements, yet his researches have given us a much more complete knowledge of the organization of the Teleosteous fishes, and later inquiries have shown that, on the whole, the combinations proposed by him require only some further modification and another definition to render them perfectly natural

The discovery (in the year 1871) of a living representative of a genus hitherto believed to be long extinct, Ceratodus, threw a new light on the affinities of fishes. The writer of the present article, who had the good fortune to examine this fish, was enabled to show that, on the one hand, it was a form most closely allied to Lepidosiren, and, on the other, that it could not be separated from the Ganoid fishes, and therefore that Lepidosiren also was a Ganoid,—a relation already indicated by Huxley in a previous paper on “Devonian Fishes.” This discovery led to further consideration of the relative characters of Midler’s subclasses, and to the system which is followed in the present article.

Having followed the development of the ichthyological system down to the most recent date, v>c have to retrace our steps to enumerate the most important contributions to ichthyology which appeared contemporaneously with or subsequently to the publication of the great work of Cuvier and Valenciennes. As in other branches of zoology, almost every year was marked by increased activity. F or the sake of convenience we may arrange these works under three heads.

I. Voyages, containing general accounts of Zoological Collections.

A. French.— 1. Voyage autour du monde εur les Corvettes de S. Μ. l' Uranie et la Physicienne, sous le commandement de Μ. Freycinet, “Zoologie—Poissons,” par Quoyet Gaimard, Paris, 1824. 2. Voyage de la Coquille, “Zoologie,” par Lesson, Paris, 1826-30. 3. Voyage de l'Astrolabe, sous le commandement de Μ. J. Dumont d'Urville, “Poissons,” par Quoy et Gaimard, Paris, 1834. 4. Voyage au Pôle Sud par Μ. J. Dumont d ’ Urville, ‘ ‘ Poissons,” par Hombron et Jacquinot, Paris, 1853-4.

B. English. — 1. Voyage of H. Μ. S. Sulphur, “Fishes,” by J.

Richardson, Lond., 1844-45. 2. Voyage of H.Μ.SS. Erebus and, Terror, “Fishes,” by J. Richardson, Lond., 1846. 3. Voyage of H. Μ. S. Beagle, “Fishes,” by L. Jenyns, Lond., 1842. 4. Voyage of H.Μ.S. Challenger, “Fishes,” by A. Günther.

C. German. — 1. Reise der österreichischen Fregatte Novara, “Fische,” von R. Kner, Vienna, 1865.

II. Faunae.

A. Great Britain. — 1. R. Parnell, The Natural History of the Fishes of the Firth of Forth, Edin., 1838. 2. W. Yarrell, A History of British Fishes, 3d edit., Lond., 1859. 3. J. Couch, History of the Fishes of the British Islands, Lond., 1862-65.

B. Denmark and Scandinavia.— 1. H. Kroyer, Danmark's Fiske, Copenhagen, 1838-53. 2. S. Nilsson, Skandinavisk Fauna, vol. iv. “Fiskarna,” Lund, 1855. 3. Fries och Ekstrom, Skandinaviens Fiskar, Stockh., 1836.

C. Russia.— 1. Nordmann, “Ichthyologie Pontique,” in Demi-doif’s Voyage dans la Russie méridionale, tom. iii., Paris, 1840.

D. Germany.— 1. Heckel und Kner, Die Susswasser-fische der österreichischen Monarchie, Leips., 1858. 2. C. T. E. Siebold, Die Süsswasserfische von Mitteleuropa, Leips., 1863.

E. Italy and Mediterranean. — 1. Bonaparte, Iconografia della Fauna Italica, tom. iii., “Pesci,” Rome, 1832-41. 2. Costa, Fauna del Regno di Napoli, “Pesci,” Naples, about 1850.

F. France.— 1. E. Blanchard, Les Boissons des eaux douces de la France, Paris, 1866.

G. Spanish Peninsula. — The freshwater fish fauna of Spain and Portugal was almost unknown, until F. Steindachner paid some visits to those countries for the purpose of exploring the principal rivers. His discoveries are described in several papers in the Sitzungsberichte der Akademie zu Wien. B. du Bocage and F. Capello made contributions to our knowledge of the marine fishes on the coast of Portugal (Jom. Scienc. Acad. Lisb. ).

H. North America.— 1.J. Richardson, Fauna Boreali-Americana, part iii., “Fishes,” Lond., 1836. The species described in this work are nearly all from the British possessions in the north. 2. Dekay, Zoology of New York, part iv., “Fishes,” New York, 1842. 3. Reports of the Un ited States Commission of Fish and Fisheries, 5 vols., Washington, 1873-79, contain much valuable information. Besides these works, numerous descriptions of North American freshwater fishes have been published in the reports of the various U. S. Government expeditions, and in North American scientific journals, by Storer, Baird, Girard, W. 0. Ayres, Cope, Jordan, Brown Goode, &c., but a good general, and especially critical, account of the fishes of the United States is still a desideratum.

I. Japan.— 1. Fauna Japonica, “Poissons,” par H. Schlegel, Leyden, 1850.

J. East Indies; Tropical parts of the Indian and Pacific Oceans. —1. E. Riippell, Atlas zu der Reise im Nördlichen Afrika, Frankf., 1828. 2. E. Rüppell, Neue Wirbelthiere, f Fische,” Frankf., 1837. 3. R. L. Playfair and A. Günther, The Fishes of Zanzibar, Lon i., 1876. 4. C. B. Klunzinger, Synopsis der Fische des Rothen Meers, Vienna, 1870-71. 5. F. Day, The Fishes of India, Lond., 1865, 4to, contains an account of the freshwater and marine species. 6. A. Günther, Die Fische der Südsee, Hamburg, 4to, from 1873 (in progress). 7. Unsurpassed in activity, as regards the exploration of the fish fauna of the East Indian archipelago, is P. Bleeker (1819-78), a surgeon in the service of the Dutch East Indian Government, who, from the year 1840, for nearly thirty years, amassed immense collections of the fishes of the various islands, and described them in extremely numerous papers, published chiefly in the journals of the Batavian Society. Soon after his return to Europe (1860) Bleeker commenced to collect the final results of his labours in a grand work, illustrated by coloured plates, Atlas Ich-thyologique des Indes Orientales Néerlandaises, Amsterd., fol., 1862; the publication of which was interrupted by the author’s death in 1878.

K. Africa. —1. A. Günther, “The Fishes of the Nile,” in Pethe-

rick’s Travels in Central Africa, Lond., 1869. 2. W. Peters, Naturwissenschaftliche Reise nach Mossambique, iv., “Flussfische,” Berl., 1>68, 4to.

L. West Indies and South America.— 1. L. Agassiz, Selecta genera et species Piscium, quae in itinere per Brasiliam collegit J. B. de Spix, Munich, 1829, fol. 2. F. de Castelnau, Animaux nouveaux ou rares, ί ecueillis pendant! expedition dans les parties centrales de l'Amérique du Sud, “Poissons,” Paris, 1855. 3. L. Vaillant and F. Bocoùrt, Mission scientifique au Mexique et dans VAmérique centrale, “Poissons,” Paris, 1874. 4. F. Poey, the celebrated naturalist of Havana, devoted many years of study to the fishes of Cüba. His papers and memoirs are published partly in two periodicals, issued by himself, under the title of Memorias sobre la Historia natural de la Isla de Cuba (from 1851), and Repertorio Fisico-natural de la Isla de Cuba (from 1865), partly in North American scientific journals. And, finally, F. Steindachner and A. Günther have published many contributions, accompanied by excellent figures, to our knowledge of the fishes of Central and South America.

Μ. New Zealand.— 1. F. W. Hutton and J. Hector, Fishes of New Zealand, Wellington, 1872.

N. Arctic Regions.— 1. C. Liitken, “A Revised Catalogue of the Fishes of Greenland,” in Manual of the Natural History, Geology, and Physics of Greenland, Lond., 1875, 8vo. 2. The fishes of Spitzbergen were examined by A. J. Malmgren (1865).

III. Anatomical Works.

The number of authors who have investigated the anatomy of fisħes is almost as great as that of faunists; and we should go beyond the limits of the present article if we mentioned more than the most prominent ami successful. Μ. H. Rathke, J. Müller, J. Hyrtl, and H. Stannius left scarcely any organ unexamined, and their researches had a direct bearing either on the relation of the class of fishes to the other vertebrates, or on the systematic arrangement of the fishes themselves. E. E. von Baer, F. de Filippi, C. Vogt, W. His, W. K. Parker, and F. Μ. Balfour investigated their embryology; A. Kolliker and G. Pouchet their histology. The osteology was specially treated by G. Bakker, F. C. Rosenthal, L. Agassiz, and C. Gegenbaur; the nervous system by Gottsche, Philipeaux, Stannius, L. de Sanctis, L. Stieda, Baudelot, and Miclucho-Maclay; the organ of hearing by E. H. Weber, C. Hasse, and G. Retzius. The electric fishes were examined by E. Geoffroy, C Matteuci, P. Pacini, T. Bilharz, and Max Schultze. The development and metamorphosis of the lamperns was made the subject of research by H. Millier, Μ. Schultze, and P. Owsjannikow; Müller’s examination of Branchiostoma was continued by J. Marcusen, A. Kovalevsky, L. Stieda, W. Müller, C, Hasse, T. H. Huxley, and F. Μ. Balfour. The most comprehensive accounts of the anatomy of fishes are contained in the subjoined works, which have been chiefly followed in the following anatomical description :—

1. H. Stannius, Zootomie der Fische, 2d edit., Berl., 1854. 2. R. Owen, Anatomy of Vertebrates, vol. i., Lond., 1866. 3. R. Owen, Lectures on the Comparative Anatomy and Physiology of the J "erte>∙ brate Animals, parti., “Fishes,” Lond., 1846. 4. T. H. Huxley, Manual of the Anatomy of Vertebrated Animals, Lond., 1871.

It has been mentioned above that the great work of Cuvier and Valenciennes had been left incomplete. Several authors, therefore, have supplied detailed Recounts of the orders omitted in that work. Müller and Henle published an account of the Plagiostomes, and Kaup of the Muraenidoe and Lophobranchii; while A. Duméril commenced an Histoire naturelle des Poissons ou Ichthyologie générale, of which, however, only two volumes appeared, containing a complete account of the Plagiostomes (Paris, 1865) and of the Ganoids and Lophobranchs (Paris, 1870).

The activity which had prevailed in ichthyology since the publication of the Histoire naturelle by Cuvier and Valenciennes had been so great, and the results of the numerous investigations were scattered over such a multitude of publications, that it ultimately became imperative to collect all these materials in one comprehensive work. This was done in the Catalogue of Fishes, published by the trustees of the British Museum, in eight volumes (Lond., 1859.-70). Besides the species previously described, many new forms were added, the total number of species referred to in those volumes amounting to 8525. As regards the systematic arrangement, Müller’s. system was adopted in the main, but the definition of the families was much modified. This, however, need not be further entered on now, as it will become sufficiently apparent in the systematic portion of this article.

For fuller details than can be given here regarding the structure, classification, and life-history of fishes, the reader is referred to the Introduction to the Study of Fishes, by A. Günther, Edin., 1880.

External Parts.

In the body of a fish tour parts are distinguished,—the head, trunk, tail, and fins; the boundary between the first and second is generally indicated by the gill-opening, and that between the second and third by the vent. The form of the body and the relative proportions of these principal parts are subject to greater variation than is to be found in any other class of vertebrates. In fishes which are endowed with the power of steady and more or less rapid locomotion, a deviation from that form of body which we observe in a perch, carp, or mackerel is never excessive. The body is a simple, equally-formed wedge, compressed or slightly rounded, well fitted for cleaving the water. In fishes which are in the habit of moving on the bottom, the whole body, or at least the head, is vertically depressed and flattened; and the latter may be so enormously enlarged that the trunk and tail appear merely as an appendage. In one family of fishes, the Pleuronectιdce or flat-fishes, the body is compressed into a thin disk; they swim and move on one side only, which remains constantly directed towards the bottom, a peculiarity by which the symmetry of all parts of the body has been affected. In fishes moving comparatively slowly through the water, and able to remain (as it were) suspended in it, a lateral compression of the body, in conjunction with a lengthening of the vertical and a shortening of the longitudinal axis, is found. This deviation from the typical form may proceed so far that the vertical axis greatly exceeds the longitudinal in length; generally all the parts of the body participate in this form, but in one kind of fish (the Crtha-goriscus or sun-fish) it is chiefly the tail that has been shortened, that being reduced so much as to present the appearance of being cut off. An excessive lengthening of the longitudinal axis, with a shortening of the vertical, occurs in eels and eel-like fishes, and in the so-called band-fishes. They are bottom-fish, capable of insinuating themselves into narrow crevices and holes. The form of the body in these long fishes is either cylindrical (snake-like), as in the eels and many codfishes, or strongly compressed, as in the band-fishes {Trichiurus, Regalecus, &c.). It is chiefly the tail that is lengthened, but frequently the head and trunk participate more or less in this form. Every possible variation occurs between these and other principal types of form. The old ichthyologists, even down to Linnaeus, depended in great measure on them for classification; but, although the same form of body often obtains in the same groups of fishes, similarity of form by no means indicates natural affinity; it only indicates similarity of habits and mode of life.

The External Parts of the Head.— The eye divides the head into the ante-orbital and post-orbital portions. In most fishes, especially in those with a compressed head, it is situated on the side and in the anterior half of the length of the head; in many others, chiefly those with a depressed head, it is directed upwards, and sometimes situated quite at the upper side; in a very few, the eyes look obliquely downwards. In the flat-fishes both eyes are on the same side of the head, either the right or the left, always on that which is directed towards the light, and coloured.

Fishes in general, as compared with other Vertebrata, have large eyes. Sometimes these organs are enormously enlarged, indicating either that the fish is nocturnal in its habits, or lives at a depth to which only a part of the sun’s rays penetrate. On the other hand, small eyes occur in fishes inhabiting muddy places, or great depths, to which scarcely any light descends, or in fishes in which the want of an organ of sight is compensated by the development of other organs of sense. In a few fishes, more particularly those inhabiting caves or the greatest depths of the ocean, the eyes have become quite rudimentary and hidden under the skin.

In the ante-orbit ιl portion of the .head, or the Snout, are situated the mouth and nostrils.

The mouth is formed by the intermaxillary and maxillary bones or the intermaxillary only in the upper jaw, and by the mandibular bane in the lower. These bones are either bare or covered by integument, to which frequently labial folds or lips are added. As regards form, the mouth offers as many variations as the body itself, according to the nature of the food, and the mode of feeding. It may be narrow, or extremely wide and cleft nearly to the hind margin of the head; it may be semi-elliptical, semicircular, or straight in a transverse line; it may be quite in front of the snout (anterior), on its upper surface (superior), on its lower (inferior), or extending along each side (lateral); sometimes it is subcircular, organized for sucking. The jaws of some fishes are modified into a special weapon of attack (sword-fish, saw-fish); in fact, throughout the whole class of fishes the jaws are the only organ ever specialized for this purpose, weapons on other parts of the body being purely defensive.

Both jaws may be provided with skinny appendages, barbels, which, if developed and movable, are sensitive organs of touch.

In the majority of fishes the nostrils exhibit a double opening on each side of the upper surface of the snout, the openings of each side being more or less close together. They lead into a shallow groove, and only in one family (the Myxinoids) perforate the palate. In this family, as well as in the lampreys, the nasal aperture is single. In many eels the openings are lateral, the lower perforating; the upper lip. In the sharks and rays (figs. 2, 3, 4) they are at the lower surface of the snout, and more or less confluent. And, finally, in the Dipnoi (fig. 35) and other Ganoids, one at least is within the labial boundary of the mouth.

The space across the forehead, between the orbits, is called the interorbital space; that below the orbits, the infraorbital or suborbital region.

In the postorbital part of the head there are distinguished, at least in most Teleosteous fishes and many Ganoids, the praeoperculum, a sub-semicircular bone, generally with a free and often serrated or variously-armed margin; the operculum, forming the posterior margin of the gill-opening; and the suboperculum and interoperculum along its inferior margin. All these bones, collectively called opercula, form the gill-cover, a thin bony lamella covering the cavity containing the gills.

The gill-opening is a foramen, or slit, behind or below the head, by which the water that has been taken up through the mouth for the purpose of breathing is again expelled. This slit may extend from the upper end of the operculum all round the side of the head to the symphysis of the lower jaw; or it maybe shortened and finally reduced to a small opening on any part of the margin of the gill-cover. Sometimes (Symδrαnchus) the two openings, thus reduced, coalesce, and form what externally appears as a single opening only. The margin of the gillcover is provided with a cutaneous fringe, in order more effectually to close the gill-opening; and this fringe is supported by one or several or many bony rays, the branchiostegals. The space on the chest between the two rami of the lower jaw and between the gill-openings is called the isthmus.

The sharks and rays differ from the Teleosteous and Ganoid fishes in having five branchial slits (six or seven in Hexanchus and Heptanchus), which are lateral in the sharks, and at the· lower surface of the head in the rays {fig. 4). In Myxine only the gill-opening is at a great •distance from the head; in that family lCydostomτ) it is either single or there are six or more on each side {fig. 1), as in the lampreys.

The Trunk and Tail.— In the trunk are distinguished the back, sides, and abdomen. It gradually passes in all fishes into the tail, the termination of the abdominal cavity and the commencement of the tail being generally indicated by the position of the vent. The exceptions are numerous: not only may certain abdominal organs, such as the sexual, extend to between the muscles of the tail, but the intestinal tract itself may pass far backwards, or it may even be reflected forwards, so that the position of the vent may be either close to the extremity of the tail or to the foremost part of the trunk.

In many fishes the greater part of the tail is surrounded by fins, leaving finless only a small portion between the dorsal, caudal, and anal fins; this finless part is called the free portion or the peduncle of the tail.

The Pins. —The fins are divided into vertical or unpaired, and horizontal or paired fins. Any of them may be present'or absent; and their position, number, and form are most important guides in determining the affinities of fishes.

The vertical fins are situated in the median dorsal line, from the head to the extremity of the tail, and in the ventral line of the tail. In fishes in which they are least developed or most embryonic, the vertical fin appears as a simple fold of the skin surrounding the extremity of the tail. In its further progress of developιħent in the series of fishes, it gradually extends further forwards, and it may reach even the head and the vent. In this embryonic condition the fin is generally supported by fine »ays, which are the continuations of, or articulated to, other stronger rays supported by the processes or apophyses of the vertebral column. This form of the vertical fin is very common,—for instance in the eels, and many Gadoid, Blennioid, and Ganoid fishes, in which the rays have ceased, besides, to be simple rods, showing more or less numerous joints (simple articulated rays, fig. 5). Branched rays are dichotomously split, the joints increasing in number towards the extremity.

The continuity of the vertical fin, however, is interrupted in the majority of fishes, and three fins are then distinguished: one in the dorsal line—the dorsal fin; one in the ventral line behind the anus—the anal fin; and one confined to the extremity of the tail—the caudal fin.

The caudal fin is rarely symmetrical so that the upper half is equal to the lower; the greatest degree of asymmetry obtains in fishes with a heterocercal termination of the vertebral column. In fishes in which it is nearly symmetrical it is frequently prolonged into an upper and lower lobe, its hind margin being concave or more or less deeply excised; in others the hind margin is rounded, and, when the middle rays greatly exceed in length the outer ones, the fin assumes a pointed form.

Many and systematically important differences are observed in the dorsal fin, which is either spiny-rayed (spinous) (Acaηthopterygiaη), or soft-rayed {Mαlαcopterygiαn). In the former, a smaller or greater number of the rays are simple and without transverse joints; they may be flexible, or so much osseous matter is deposited in them that they appear hard and truly spinous (fig. 7); these spines form-always the anterior portion of the fin, which is either detached from or continuous with the remaining jointed rays. The spines can be erected or depressed at the will of the fish; if in the depressed position the spines cover one another completely, their points lying in the same line, the fish is called homacanth; but if the spines are asymmetrical, alternately broader on one side than on the other, the fish is called heteracanth. The spinous division, as well as the one consisting of jointed rays, may again be subdivided. In the Malacopterygian type all the rays remain jointed; indeed, sometimes the foremost ray, with its preceding short supports, is likewise ossified, and is a hard spine, but the articulations can nearly always be distinctly traced. Sometimes the dorsal fin of Malacopterygian fishes is very long, extending from the head to the end of the tail; sometimes it is reduced to a few rays only; and in a few cases it is entirely absent. In addition to the rayed dorsal fin, many Malacopterygian fishes (as the Salmonoids, many Siluroids, Scopeloids, &c.) have another of greater or lesser extent, without any rays; and as fat is always deposited within this fold, it is called a fatty fin (pinna adiposa) (figs. 8 and 9).

The anal fin is built on the same plan as the dorsal; it may be single, or there may be more than one; it may also be long or short, or entirely absent; in Acanthopterygians its foremost rays are frequently simple and spinous.

The horizontal or paired fins consist of two pairs, the pectorals and ventrals.

The pectoral fins (with their osseous supports) are the homologues of the anterior limbs of the higher Vertebrata. They are always inserted immediately behind the gill opening,—either symmetrical, with a rounded posterior margin, or asymmetrical, with the upper rays longest and strongest; in Malacopterygians with a dorsal spine the upper pectoral ray is frequently developed into a similar defensive weapon.

The ventral fins are the homologues of the hind-limbs, and are inserted on the abdominal surface, either behind the pectorals (Pisces or Pinnoe abdominales') (fig. 9), or below them (Pisces or Pinnoe thoracicas) (fig. 10), or in advance of them (Pisces or Pinnoe figulares) (fig. 11). They are generally narrow, and composed of a small number of rays, the outer of which is frequently osseous.

For the definition of the smaller systematic groups, and the determination of species, the numbers of the spines and rays are generally of the greatest importance. This holds good especially for the ventral rays, by the number of which the Acanthopterygian affinities of a fish can nearly always be determined. The numbers of the dorsal and anal rays generally correspond to the number of vertebrae in a certain portion of the spinous column, and are therefore constant specific, generic, or even family characters; but when their number is very great, a proportionally wide margin must be allowed for variation, and the taxonomic value of this character becomes uncertain. The numbers of the pectoral and caudal rays are rarely of any account.

The fins are organs of motion, but it is chiefly the tail and the caudal fin by which the fish impels itself forward. To execute energetic locomotion, the tail and caudal fin are strongly bent with rapidity, alternately towards the right and left; whilst a gentle motion forwards is effected by a simple undulating action of the caudal fin, the lobes of which act like the blades of a screw. Retrograde motions can be made by fishes in an imperfect manner only, by forward strokes of the pectoral fins. When the fish wants to turn towards the left, he gives a stroke of the tail towards the right, the right pectoral acting simultaneously, whilst the left remains pressed close to the body. Thus the pectoral fins assist in the progressive motions of the fish, but rather by directing its course than by acting as powerful propellers. The chief function of the paired fins is to maintain the balance of the fish in the water, which is always most unsteady where there is no weight to sink it: when the pectoral of one side, or the pectoral and ventral of the same side, are removed, the fish loses its balance and falls on the side opposite; when both pectorals are removed, the fish’s head sinks; on removal of the dorsal and anal fins the motion of the fish assumes a zigzag course; deprived of all fins, it floats like a dead fish, with the belly upwards, the back being the heavier part of the body.

In numerous groups of fishes which live in mud, or are able to pass a longer or shorter time in soil periodically dried and hardened during the hot season, forms occur entirely devoid of, or with only rudimentary, ventral fins (Cyprinodon, Ophiocephalidoe, Galaxiidoe, Siluridoe). The chief function of these fins being to balance the body of the fish whilst swimming, it is evident that in fishes moving during a great part of their life over swampy ground, or through more or less consistent mud, this function of the ventral fins ceases, and nature can readily dispense with these organs altogether.

In certain fishes the shape and function of the fins are considerably modified: thus, in the rays, locomotion is almost entirely effected and regulated by the broad and expanded pectoral fins acting with an undulatory motion of their margins, similar to the undulations of the long vertical fins of the flat fishes; in many blennies the ventral fins are adapted for walking on the sea-bottom; in some Gobioids (Periophthal- mus), Trigloids, Scorpaenioids, and Pediculati the pectoral fins are perfect organs of walking; in the gobies, Cycloρteri, and Discoboli the ventral fins are transformed into an adhesive disk (see fig. 13); and finally in the flying-fish the pectorals act as a parachute. In the eels and other snakelike fishes, the swimming as well as the gliding motions are effected by several curvatures of the body alternately towards theright and left, resembling the locomotion of snakes. In the Syngnathi (pipe-fishes) and Hippocampi (sea-horses), whose body admits of but a slight degree of lateral curvature, and whose caudal fin is generally small if present at all, locomotion is very limited, and almost wholly dependent on the action of the dorsal fin, which consists of a rapid undulating movement.

The Skin and Scales.— The skin of fishes is either covered with scales, or naked, or provided with more or less numerous scutes of various forms and sizes. Some parts, like the head and fins, are more frequently naked than scaly. All fishes provided with electric organs, the majority of eels, and the lampreys are naked. Scales of fishes are very different from those of reptiles,—the latter being merely folds of the cutis, whilst the former are distinct horny elements, developed in grooves or pockets of the skin, like hairs, nails, or feathers.

Very small or rudimentary scales are extremely thin, homogeneous in structure, and more or less imbedded in the skin, and do not cover each other. When more developed, they are imbricated (arranged in the manner of tiles), with the posterior part extruded and free, the surface of the anterior portion being usually covered by the skin to a greater or less extent. On their surface (fig. 14) may be observed a very fine striation concentric with and parallel to the margin, and coarser striae radiating from a central point towards the hind margin.

Scales without a covering of enamel, with an entire (not denticulated) posterior margin, and with a concentric striation, are called cycloid scales. - Ctenoid scales (figs. 15,16) are generally thicker, and provided with spinous teeth on the posterior edges of the layers of which the scale consists. In some species, only the layer nearest to the margin is provided with denticulations (fig. 15). Scales the free surface of which is spiny, and which have no denticulation on the margin, have been termed sparoid scales; but their distinction from ctenoid scales is by no means sharp, and there are even intermediate forms between the cycloid and ctenoid types. Both kinds of scales may occur, not only in species of the same genus of fishes, but in the same fish.

Ganoid scales are hard and bony, covered with a layer of enamel; they are generally rhombic or quadrangular, rarely rounded and imbricate, and are arranged in oblique rows, those of one row being linked together by an articulary process. This type of scales, common in fossil Ganoid fishes, occurs among recent fishes in Lepidosteus and Polypterus only.

Finally, in sharks, the Balistidce t and others, true scales are absent, and are replaced by the ossified papillae of the cutis (fig. 18), which give the surface the appearance of finegrained shagreen. These generally small bodies, as well as the large osseous scutes (figs. 6 and 20) of the rays, sturgeons, <fcc., have been comprised under the common name placoid scales, a term which is being deservedly abandoned.

Along the side of the body of osseous fishes runs a series of perforated scales, which is called the lateral line (fig. 23). The perforating duct is simple at its base, and may be simple also at its outer opening (fig. 21), or (a frequent case) the portion on the free surface of the scale is ramified (fig. 22). The lateral line runs from the head to the tail, sometimes reaching the caudal fin, sometimes stopping short of it, sometimes advancing over its rays. Some species have several lateral lines, an upper one following the dorsal, a lower the abdominal outline, while a third runs along the middle as usual. The scales of the lateral line are sometimes larger than the others, sometimes smaller; sometimes they are modified into scutes; sometimes there are no other scales besides these, the rest of the body being naked. The foramina of the lateral line are the outlets of a muciferous duct which is continued on to the head, running along the infraorbital bones, and sending off a branch into the praeopercular margin and mandible. The muciferous system is abundantly provided with nerves, and has therefore been considered to be the seat of a sense peculiar to fishes, but there cannot be any doubt that its function is the excretion of mucus, although probably mucus is excreted also from the entire surface of the fish.

The scales, their structure, number, and arrangement, constitute an important character for the determination of fishes. In most scaly fishes they are arranged in oblique transverse series; and, as the number of scales in the lateral line generally corresponds to the number of transverse series, it is usual to count the scales in that line. To ascertain the number of longitudinal series of scales, the scales are counted in one of the transverse series, generally in that running from the commencement of the dorsal fin, or the middle of the back, to the lateral line, and from the lateral line down to the vent or ventral fin, or the middle of the abdomen.

Osteology.

In order readily to comprehend the following account of the modifications of the skeleton in the various subclasses and groups of fishes, the student should acquaint himself with the terms used for the numerous bones of the fish skeleton, as well as with their relative position. For this purpose we commence this section with an account of the skeleton of the Teleostei, which is composed of the greatest number of specialized bones, and is most readily accessible. The skeleton of any of the more common kinds of osseous fish may serve for this purpose; that of the perch was chosen by Cuvier, and is employed here .(fig. 24), as it was in the last edition of the present work.

In the Teleosteous fishes the spinous column consists of completely ossified amphicoelous vertebrae; its termination is homocercal—that is, the caudal fin appears to be more or less symmetrical, the last vertebra occupying a central position in the base of the fin, and being united to a fiat fau-like bone, the hypural (70 in fig. 24), on the hind margin of which the fin-rays are fixed. The hypural is but a union of modified haemapophyses which are directed backwards, and the actual termination of the notochord is bent upwards, and lies along the upper edge of the hypural, hidden below the last rudimentary neural elements. In some Teleosteans, as the Salmonidoe, the last vertebrae are conspicuously bent upwards: in fact, strictly speaking, this homocercal condition is but one of the various degrees of heterocercy, different from that of many Ganoids in this respect only, that the caudal fin itself has assumed a higher degree of symmetry.

The neural and haemal arches generally coalesce with the centrum, but there are many exceptions, inasmuch as some portion of the arches of a species, or all of them, may show the original division.

The vertebrae are generally united with one another by zygapophyses, and frequently similar additional articulations exist at the lower parts of the centra. Parapophyses and ribs are very general, but the latter are inserted on the centra and the base of the processes, and never on their extremities.

The spinal column consists of abdominal and caudal vertebrae, the coalescence of the parapophyses into a complete haemal ring and the suspension of the anal fin generally forming a sufficiently well-marked boundary between the abdominal and caudal regions (fig. 24). In the perch there are twenty-one abdominal and as many caudal vertebrae. The centrum of the first vertebra or atlas is very short, with the apophyses scarcely indicated; neither the first nor the second vertebra has ribs. All the other abdominal vertebrae, with the exception of the last or the two last, are provided with ribs, many of which are bifid (72). A series of flat spines (74), called interneurals, to which the spines and rays of the dorsal fins are articulated, are supported by the neural spines, the strength of the neurals and ihter-neurals corresponding to that of the dermal spines (75). The caudal vertebrae differ from the abdominal in having the haemapophyseal elements converted into spines similar to the neurals, the anterior being likewise destined to support a series of interhaemals (79), to which the anal rays are articulated.

There is a great amount of variation as regards the degree in which the primordial cartilaginous cranium persists; it is always more or less replaced by bone; frequently it disappears entirely, but in some fishes like the Salmonidoe or Esocidce i the cartilage persists to the same or even to a greater extent than we shall find in the Ganoidei holostei. Besides the bones preformed in cartilage there are a great number of membrane-bones, that is, bones originating in membranous or tegumentary tissue. The different kinds of these membrane-bones occur with greater or less constancy throughout the subclass Teleostei ; they often coalesce with and are no longer separable from the neighbouring or underlying cartilage-bones.

1. Cartilage Bones of the Primordial Skull.— The basioccipital (5 in figs. 25-27) has retained the form of a vertebral centrum; it is generally concave behind, the concavity containing remains of the notochord. The exoccipitals (10) are situated on the side of the basioccipital, and contribute the greater portion of the periphery of the foramen magnum. The supraoccipital (8) is intercalated between the exoccipitals, and forms a most prominent part by its median crest, which sometimes extends far forward on the upper side of the skull, and offers attachment to the dorsal portion of the large lateral muscle of the trunk. A transverse supraoccipital ridge, coming from each side of the base of this crest, runs outward to the external angles of the bone. When the interior portion of this bone remains cartilaginous, some part of the semicircular canals may be lodged in it.

The region of the skull which succeeds these bones encloses at least the greater portion of the labyrinth, and its component parts have been named with reference to it by some anatomists. The alisphenoids (11) (prooticum) form sutures posteriorly with the basi-and ex-occipitals, and meet each other in the median line at the bottom of the cerebral cavity; they contribute to the formation of a hollow in which the hypophysis cerebri and the saccus vasculosus are received; in conjunction with the exoccipital they form another hollow for the reception of the vestibulum; generally they are perforated by the trigeminal and facial nerves. The paroccipitals (9) (epioticum) lodge a portion of the posterior vertical semicircular canal, and form a projection of the skull on each side of the occipital crest, to which a terminal branch of the scapular arch is attached. The mastoid (12 + 13) (ppisthoticum) occupies the posteroexternal projection of the head; it encloses a part of the external semicircular canal, is generally united to a membrane-bone, the superficial squamosal, which emits a process for the suspension of the scapular arch, and is frequently, as in the perch, divided into two separate bones.

The anterior portion of the skull varies greatly as regards form, which is chiefly dependent on the extent of the cere bral cavity; if the latter is advanced far forwards, the lateral walls of the primordial cranium are protected by more developed ossifications than is the case if the cerebral cavity is shortened by the presence of a wide and deep orbit. In the latter case parts which normally form the side of the skull are situated in front of the brain-case, between it and the orbit, and, being generally reduced in extent, are often replaced by membranes; the interorbital septum especially may thus be reduced to a membrane. The most constant ossifications of this part of the skull are the orbitosphenoids (14), which join the upper anterior margin of the alisphenoids; superiorly the olfactory and inferiorly the optic nerves pass between them out of the cranium. They vary much with regard to their development: they are small in Gadoids; larger in the perch, pike, Salmonoids, ΛLacrodon, and the Clupeoids; and very large in Cyprinoids and Siluroids, in which they contribute to the formation of the side of the braincase. The single Y-shaped sphenoideum anterius (15) (ethmoid of Owen and basisphenoid of Huxley) is as frequently absent as present; each lateral branch is connected with an orbitosphenoid, whilst the lower branch rests upon the long basal bone; it forms the anterior margin of the fossa for the hypophysis. Finally, to this group of cartilage-bones belongs also the postfrontal (4), a small bone from which the infraorbital ring is suspended.

The centre of the foremost part of the skull is occupied by the ethmoid (3), which shows great variations as regards its extent and the degree of ossification; it may extend backwards into the interorbital septum, and reach the orbitosphenoids, or it may be confined to the extremity of the skull; it may remain entirely cartilaginous, or it may ossify into a lamina which separates the two orbits and encloses an anterior prolongation of the brain-case, along which the olfactory nerves pass,—modifications occurring again in higher vertebrates. A paired ossification attached to the forepart of the ethmoid is the prefrontals (2), which form the base of the nasal fossa.

2. Membrane-Bones attached to the Primordial Skull.— To this group belong the parietals (7) and frontals (1). The latter form the upper margin of the orbits, and extend from the nasal cavities to the occipital. They are enlarged at the expense of the parietals, which are of much smaller extent than in higher vertebrates, and are separated from each other by the anterior prolongation of the supraoccipital. The squamosal (12) has been mentioned above in connexion with the mastoid. The prefrontal (2) and supraorbital are always small, and the latter is frequently absent. The lower surface of the skull is protected by the basisphenoid (parasphenoid) (6), a very long and narrow bone extending from the basioccipital beyond the brain-capsule to between the orbits, where it forms the support of the interorbital septum. Anteriorly it is connate with another long hammer-shaped bone (16), the vomer. Both these bones, especially the latter, may be armed with teeth.

3. Cartilage-Bones of the Alimentary Portion of the Visceral Skeleton of the Skull.— The suspensorium consists of three cartilage-bones, and affords a base for the opercular apparatus as well as a point of attachment to the hyoid, whilst in front it is connected with the pterygo-palatine arch (fig. 27). These are the hyomandibular (23), symplectic (31), and quadrate (26); they are connected by means of the metapterygoid (27) with the pterygoid (24) and entopterygoid (25), the foremost bone of the arch being the palatine (22). Of these bones the uppermost, the epitympanic or hyomandibular (23), is articulated by a double articulary head with the mastoid and posterior frontal. Another articulary head is destined for the opercular joint. The mesotympanic or symplectic (31) appears as a styliform prolongation of the lower part of the hyomandibular; it is cartilaginous in the young, but almost completely ossified in the adult. The position of this bone is noteworthy, because, directly inwards from its cartilaginous junction with the hyomandibular, there is situated the uppermost piece of the hyoid arch, the stylohyal. The next bone of the series is the pretympanic or metapterygoid (27), a flat bone forming a bridge towards the pterygoid, and not rarely absent in the subclass. Finally, the large triangular hypotympanic or quadrate (26) has a large condyle for the mandibular joint. The palatine arch proper is formed by three bones: the entopterygoid (25) is an oblong and thin bone attached to the inner border of the palatine and pterygoid, and increasing the surface of the bony roof of the mouth towards the median line; it constitutes also the floor of the orbit; the pterygoid (or os transversum) (24) starts from the quadrate, and is joined by suture to the palatine, which is toothed in the perch and many other fishes, and reaches to the vomer and anterior frontal. The piece of the mandible which articulates with the quadrate is the articulary bone (35), distinctly part of Meckel’s cartilage; it sends upwards a coronoid process to which the masticatory muscles, as well as a ligament from the maxillary, are attached; it also sends forwards a long-pointed process to be sheathed in a deep notch of the dentary piece. Frequently another portion of cartilage below the articulary remains persistent, or is replaced by a separate membrane-bone, the angular.

4. Membrane-Bones of the Alimentary Portion of the Visceral Skeleton of the Skull.— The suspensorium has one tegumentary bone attached to it, viz., the praeoperculum (30); it is but rarely absent, as for instance in Muroenophis·, as it is quite a superficial bone, and frequently armed with spines (as in the perch), its form and configuration constitute an important item in the description of many fishes. The premaxillary (17) and maxillary (18) of the Teleostei appear to be also membrane-bones, although they are clearly analogous to the upper labial cartilages of the sharks. The premaxillaries sometimes coalesce into a single piece (as in Diodon, Mormyrus), or they are firmly united with the maxillaries (as in all Gymnodonts, Serrasalmo, &c.). The relative position and connexion of these two bones differ much, and form a valuable character in the discrimination of the various families. In some, the front margin of the jaw is formed by the premaxillary only, the two bones having a parallel position, as in the perch, in which case the maxillary is constantly toothless; in others, the premaxillary is shortened, allowing the maxillary to enter, and to complete, the margin of the upper jaw; and finally, in many, no part of the maxillary is situated behind the premaxillary, but the entire bone is attached to the end of the premaxillary, forming its continuation. In the last case the maxillary may be quite abortive. The mobility of the upper jaw is greatest in those fishes in which the premaxillary alone forms its margin. The form of the premaxillary is subject to great variation: the beak of Belone and Xiphias is formed by the prolonged and coalesced premaxillaries. The maxillary consists sometimes of one piece, sometimes of two or three. The principal membrane-bone of the mandible is the dentary (34), to which is added the angular (36) and rarely a smaller one, the splenial or os operculare, which is situated at the inner side of the articulary.

5. Cartilage-Bones of the Respiratory Portion of the Visceral Skeleton of the Skull.— With few exceptions all the ossifications of the hyoid and branchial arches belong to this group.

The hyoid arch is suspended by a slender styliform bone, the stylohyal (29), from the hyomandibulars; it consists of three segments, the epihyal (37), the ceratohyal (38), which is the longest and strongest piece, and the basihyal, which is formed by two juxtaposed pieces (39, 40). Between the latter there is a median styliform ossicle (41), extending forwards into the substance of the tongue, called the glossohyal or os linguale.

The branchial arches (figs. 26 and 28) are enclosed within the hyoid arch, with which they are closely connected at the base. In the perch and in the majority of Teleosteans they are five in number, of which four bear gills, whilst the fifth (56) remains dwarfed, and is beset with teeth; it is called the lower pharyngeal bone. The arches adhere by their lower extremities to a chain of ossicles (53, 54, 55), the basibranchials, and, curving as they ascend, nearly meet at the base of the cranium, to which they are attached by a layer of ligamentous and cellular tissue. Each of the first three branchial arches consists of four pieces movably connected with one another. The lowest is the hypobranchial (57), the next, a much longer one, the ceratobranchial (58), and, above this, a slender and short irregularly-shaped epibranchial. In the fourth arch the hypobranchial is absent. The uppermost of these segments, especially of the fourth arch, are dilated, and more or less confluent; they are beset with fine teeth, and generally distinguished as the upper pharyngeal bones. Only the ceratobranchial is represented in the fifth arch or lower pharyngeal. On their outer convex side the branchial segments are grooved for the reception of large bloodvessels and nerves; on the inner side they support horny processes, called the gill-rakers, which do not form part of the skeleton.

6. Membrane-Bones of the Respiratory Portion of the Visceral Skeleton of the Skull.— These include the opercular pieces, viz., the operculum (28), suboperculum (32), and interoperculum (33). The last of these is the least constant; it may be entirely absent, and represented by a ligament extending from the mandible to the hyoid. The operculum is the principal piece, situated behind, and movably united with, the vertical limb of the praeoperculum. There is an articulary cavity at its upper anterior angle for its junction with the hyomandibular. The interoperculum is connected by ligament with the angular piece of the mandible, and is also attached to the outer face of the hyoid, so that the gill-covers cannot open or shut without the hyoid apparatus executing a corresponding movement. The remaining membrane-bones are the urohyal (42), a single bone, which is connected by ligament with the anterior extremity of the humeral arch, and separates the musculi sternohyoidei, serving as an increased surface for their insertion; and, finally, the branchiostegals (43), which vary greatly in number, but are always fixed to the ceratohyals and epihyals by ligaments; the branchiostegal membrane is extended between them.

7. Dermal Bones of the Skull.— To this category are referred some bones which are ossifications of, and belong to, the cutis. They are the turbinals (20), the suborbitals (19), and the supratemporals. They vary much with regard to the degree in which they are developed, and are seldom entirely absent. Nearly always they are wholly or partly transformed into tubes or hollows, in which the muciferous canals with their numerous nerves are lodged. Those in the temporal and scapulary regions are not always developed; on the other hand, the series of those ossicles may be continued on to the trunk, accompanying the lateral line. The foremost suborbital is termed the praeorbital.

The pectoral arch or shoulder-girdle of the Teleosteous fishes exhibits but a remnant of a primordial cartilage, which is replaced by two ossifications, the caracoid (51) and scapula (52); posteriorly they offer attachment to two series of short rods (53), of which the proximal are nearly always ossified, whilst the distal frequently remain small cartilaginous nodules hidden in the base of the pectoral rays; they have been termed carpals and metacarpals. The bones by which this portion is connected with the skull are membrane-bones, viz., the clavicle (48), with the postclavicle (49 + 50), the supraclavicle (47), and the post-temporal (46). By this last bone the shoulder-girdle is suspended from the skull; it is attached, in the perch, by a triple prong to the occipital and mastoid bones. The clavicle completes the arch below by the symphysis or natural connexion of the bones of each side. Many Teleosteous fishes lack pectoral fins, and in them the pectoral arch is frequently more or less reduced or rudimentary, as in many species of Muroenidae. In others the membranebones are exceedingly strong, contributing to the outer protective armour of the fish, and then the clavicles are generally suturally connected in the median line. The postclavicula and the supraclavicula may be absent. It is only exceptionally that the shoulder-girdle is not suspended from the skull, but from the anterior portion of the spinous column (Symbranchidoe, Muroenidae, Notacanthidoe). The number of elements in each of the two series of basalia never exceeds five, but may be smaller; and the distal series is absent in Siluroids.

The pubic bones (80) of the Teleosteous fishes undergc many modifications of form in the various families, but they are essentially of the same simple type as in the perch, viz., a pair of flat or styliform simple bones, to which the ventral fins are articulated.

The bones of the skull of the fish have received so many different interpretations that no two accounts agree in their nomenclature, so that their study is a matter of considerable difficulty. The table of synonyms given on p. 644 will tend to overcome difficulties arising from this cause; it contains the terms used for the different bones of the skeleton by Cuvier, those introduced by Owen, and finally the nomenclature of Stannius, Huxley, and Parker. Those adopted here are printed in italics. The numbers in the table are those used in the preceding pages and in figs. 24-28.

Modifications of the Skeleton.

We now proceed to pass briefly in review the modifications of the skeleton in the principal types of fishes, commencing at the lowest, and pointing out its gradual development in the other three subclasses.

The lowermost subclass of fishes, which comprises one form only, the Lancelet (Branchiostomα [s. Amphioxus^∖ lanceolatum), possesses the skeleton of the most primitive Table of Nomenclature used by Cuvier, Owen, Stannius, Huxley, Ac., to indicate the Bones in the Skeleton of a Fish. [table]

type. The vertebral column is represented by a simple chorda dorsalis or notochord only, which extends from one extremity of the fish to the other, and, so far from being expanded into a cranial cavity, is pointed at its anterior end as well as at its posterior. It is enveloped in a simple membrane like the spinal cord and the abdominal organs, and there is no trace of vertebral segments or ribs; a series of short cartilaginous rods, however, above the spine evidently represent apophyses. A maxillary or hyoid apparatus, or elements representing limbs, are entirely absent.

The skeleton of the Cyclostomata (or Harsipobranchu) (lampreys and sea-hags) shows a considerable advance of development. It consists of a notochord, the anterior pointed end of which is wedged into the base of a cranial capsule, partly membranous, partly cartilaginous. This skull, therefore, is not movable upon the spinal column. No vertebral segmentation can be observed in the notochord, but neural arches are represented by a series of cartilages on each side of the spinal chord. In Petromyzon (fig. 31) the basis cranii emits two prolongations on each side, an inferior, extending for some distance along the lower side of the spinal column, and a lateral, which is ramified into a skeleton supporting the branchial apparatus. A stylohyal process and a subocular arch with a palato-pterygoid portion may be distinguished. The roof of the cranial capsule is membranous in Myxine and in the larvae of Petromyzon, but more or less cartilaginous in the adult Petromyzon and in Bdellostoma. A cartilaginous capsule on each side of the hinder part of the skull contains the auditory organ, whilst the olfactory capsule occupies the anterior upper part of the roof. A broad cartilaginous lamina, starting from the cranium and overlying part of the snout, has been determined as representing the ethmo-vomerine elements, whilst the oral organs are supported by large, very peculiar cartilages (labials), greatly differing in general configuration and arrangement in the various Cyclostomes. There are three in the sealamprey, of which the middle one is joined to the palate by an intermediate smaller one; the foremost is ring-like, tooth-bearing, emitting on each side a styliform process. The lingual cartilage is large in all Cyclostomes. There is no trace of ribs or limbs.

The Chondropterygians exhibit a most extraordinary diversity in the development of their vertebral column; almost every degree of ossification, from a notochord without a trace of annular structure to a series of completely ossified vertebrae, being found in this order. The sharks in which the notochord is persistent are the Holocephali (if they be reckoned as belonging to this order) and the genera Notidanus and Echinorhinus. In the other sharks the segmentation is complete, each vertebra having a deep conical excavation in front and behind, with a central canal through which the notochord is continued; but the degree in which the primitive cartilage is replaced by concentric or radiating lamellae of bone varies greatly in the various genera, and according to the age of the individuals. In the rays all the vertebrae are completely ossified, and the anterior ones confluent into one continuous mass. In the majority of Chondropterygians the extremity of the vertebral column shows a decidedly heterocercal condition, and only a few, like Squxxtina and some rays, possess a diphycercal tail.

The advance in the development of the skeleton in Chondropterygians beyond the primitive condition of the previous subclasses manifests itself further by the presence of neural and haemal elements, extending to the foremost part of the axial column; but of these the haemal form a closed arch in the caudal region only, whilst on the trunk they appear simply as a lateral longitudinal ridge. The neural and haemal apophyses are either merely attached to the axis, as in Chondropterygians with persistent notochord (the rays and some sharks); or their basal portions penetrate like wedges into the substance of the centrum, so that, in a transverse section, in consequence of the difference in their texture, they appear in the form of an X (figs. 32-34). The interspaces between the neurapophyses of the vertebrae are not filled by fibrous membrane, as in other fishes, but by separate cartilages (laminae or cartilagines intercrurales), to which frequently a series of terminal pieces is superadded, which must be regarded as the first appearance of the interneural spines of the Tdeostei and many Ganoids. Similar terminal pieces are sometimes observed on the haemal arches. Ribs are either absent or but imperfectly represented (Carcharias).

The substance of the skull of the Chondropterygians is cartilage, interrupted especially on its upper surface by more or less extensive fibro-membranous fontanelles. Superficially it is covered by a more or less thick shagreenlike osseous deposit. The articulation with the vertebral column is effected by a pair of lateral condyles. In the sharks, besides, a central conical excavation corresponds to that of the centrum of the foremost vertebral segment, whilst in the rays this central excavation of the skull receives a condyle of the axis of the spinous column. The cranium itself is a continuous undivided cartilage, in which the limits of the orbit are well marked by an anterior and posterior protuberance. The ethmoidal region sends horizontal plates over the nasal sacs, the apertures of which retain their embryonic situation upon the under surface of the skull. In the majority of Chondropterygians these plates are produced conically, forming the base of the soft projecting snout; and in some forms, especially in the long-snouted rays and the saw-fishes (Pristis), this prolongation appears in the form of three or more tubiform rods. As separate cartilages there are appended to the skull a suspensorium, a palatine, a mandible, a hyoid, and rudimentary maxillary elements. The suspensorium is movably attached to the side of the skull. It generally consists of one piece only, but in some rays of two. In the rays it is articulated to the mandible only, their hyoid possessing a distinct point of attachment to the skull. In the sharks the hyoid is suspended from the lower end of the suspensorium together with the mandible. What is generally called the upper jaw of a shark is, as Cuvier has already stated, not the maxillary, but the palatine. It consists of two simple lateral halves, each of which articulates with the corresponding half of the lower jaw, which is formed by the simple representative of Meckel’s cartilage. Some cartilages of various sizes are generally developed on each side of the palatine, and one on each side of the mandible. They are called labial cartilages, and seem to represent maxillary elements. The hyoid consists generally of a pair of long and strong lateral pieces, and a single mesial piece. From the former, cartilaginous filaments (representing branchiostegals) pass directly outwards. Branchial arches, varying in number, and similar to the hyoid, succeed it. They are suspended from the side of the foremost part of the spinous column, and, like the hyoid, bear a number of filaments.

The vertical fins are supported by interneural and inter-haemal cartilages, each of which consists of two or more pieces; to these fins the fin-rays are attached without articulation. The scapular arch of the sharks is formed by a single coracoid cartilage bent-from the dorsal region downwards and forwards. In some genera (Scyllium, Squatina) a small separate scapular cartilage is attached to the dorsal extremities of the coracoid; but in none of the Elasmobranchs is the scapular arch suspended from the skull or vertebral column; it is merely sunk and fixed in the substance of the muscles. Behind, at the point of its greatest curvature, three carpal cartilages are joined to the coracoid, which Gegenbaur has distinguished as propterygium, mesopterygium, and metapterygium, the first occupying the front, the last the hind margin of the fin. Several more or less regular transverse series of styliform cartilages follow. They represent the phalanges, to which the horny filaments which are imbedded in the skin of the fin are attached. In the rays, with the exception of torpedo, the scapular arch is intimately connected with the confluent anterior portion of the vertebral column. The anterior and posterior carpal cartilages are followed by a series of similar pieces, which extend like an arch forwards to the rostral portion of the skull, and backwards to the pubic region. Extremely numerous phalangeal elements, longest in the middle, are supported by the carpals, and form the skeleton of the lateral expansion of the so-called disk of the ray’s body, which thus, in fact, is nothing but the enormously enlarged pectoral fin.

The pubic is represented by a single median transverse cartilage, with which a tarsal cartilage articulates. The latter supports the fin-rays. To the end of this cartilage is also attached, in the male Chondropterygians, a peculiar accessory generative organ or clasper.

The Holocephali differ from the other Chondropterygians in several important points of the structure of their skeleton, and unmistakably approach certain Ganoids. That their spinal column is persistently notochordal has been mentioned already. Their palatal apparatus, with the suspensorium, coalesces with the skull, the mandible articulating, with a short apophysis of the cranial cartilage (autostylic skull). The mandible is simple, without anterior symphysis. The spine with which the dorsal fin is armed articulates with a neural apophysis, and is not immovably attached to it, as in the sharks. The pubic consists of two lateral halves, with a short, rounded tarsal cartilage.

The skeleton of the Ganoid fishes presents extreme variations with regard to the degree in which ossifications replace the primordial cartilage. Whilst some exhibit scarcely any advance beyond the Plagiostomes with persistent cartilage, others, as regards the development and specialization of the several parts of their osseous framework, approach the Teleosteans so closely that their Ganoid nature can be demonstrated by, or inferred from, other considerations only. All Ganoids possess a separate gill-cover.

The diversity in the development of the Ganoid skeleton is well exemplified by the few representatives of the order in the existing fish fauna. Lowest in the scale in this respect are those with a persistent notochord, and an autostylic skull, that is, a skull without separate suspensorium—the fishes constituting the suborder Dipnoi, of which the existing representatives are Lepidosiren, Proto· ρterus, and Ceratodus, and the extinct (so far as demonstrated at present) Dipterus, Chirodus (and Phaneropleuron ?). In these fishes the notochord is persistent, passing uninterruptedly into the cartilaginous base of the skull. Some Dipnoi are diphy-, others hetero-cercal. Neural and haemal elements and ribs are well developed.

The primordial cranium of the Dipnoi is cartilaginous, but with more or less extensive ossifications in its occipital, basal, or lateral portions, and with large tegumentary bones, which, from this suborder upwards in the series, will be found to exist throughout the remaining forms of fishes. A strong process descends from the cranial cartilage, and offers by means of a double condyle (fig. 35, s) attachment to corresponding articulary surfaces of the lower jaw. Maxil-Iary and intermaxillary elements are not developed, but are perhaps represented in Ceratodus by some inconstant rudimentary labial cartilages situated behind the posterior nasal opening. Facial cartilages and an infraorbital ring are developed, at least in Ceratodus. The presence of a pair of small teeth in front indicates the vomerine portion (v) which remains cartilage, whilst the posterior pair of teeth are implanted in a pterygopalatine ossification (Z), which sometimes is paired, sometimes continuous. The base of the skull is constantly covered by a large basal ossification (o). The hyoid is well developed, sometimes reduced to a pair of ceratohyals, sometimes with a basihyal and glossohyal. The skeleton of the branchial apparatus approaches the Teleosteous type, less so in Lepidosiren than in Ceratodus, in which five branchial arches are developed, but with the lateral and mesial pieces reduced in number. A large operculum, and a smaller subor inter-operculum are present.

The scapular arch consists of a single median transverse cartilage, and a pair of lateral cartilages which bear the articular condyle for the pectoral limb. The latter cartilages form the base of a large membrane-bone, and the whole arch is suspended from the skull by means of an osseous supraclavicle. The fore-limb of the Dipnoi (fig. 36) greatly differs externally from the pectoral fin of other Ganoid fishes. It is covered with small scales along the middle, from the root to its extremity, and surrounded by a rayed fringe similar to the vertical fin (crossopterygian type of fin). A muscle split into numerous fascicles extends all the length of the fin, which is flexible in every part and in every direction. The cartilaginous framework supporting it is joined to the scapular arch by an oblong cartilage, followed by a broad basal cartilage (a), generally single, but sometimes showing traces of a triple division. Along the middle of the fin runs a jointed axis (6), the joints gradually becoming smaller and thinner towards the extremity; each joint bears on each side a three-, two-, or one-jointed branch (c, <7). This axial arrangement of the pectoral skeleton, which evidently represents one of its first and lowest conditions, has been termed the archipterygium by Gegenbaur. It is found in Ceratodus and other genera, but in Lepidosiren the jointed axis only has been preserved, while rudimentary rays are added in Protopterus.

The pubic consists of a single flattened subquadrangular cartilage, produced into a long single anterior process. Posteriorly it terminates on each side in a condyle, to which the basal cartilage of the ventral paddle is joined. The endoskeleton of the paddle is almost identical with that of the pectoral.

The Ganoid fishes with persistent notochord, but with a hyostylic skull (that is, a skull with a separate suspensorium), consist of the suborder Choηdrostei, of which the existing representatives are the sturgeons (Acipenser, Scaphirhynclιus, Polyodon), and the extinct the Chon- drosteidoe, Paloeoniscidoe, and (according to Traquair) Platy-somidoe.

Their spinal column does not differ essentially from that of the Dipnoi. Segmentation is represented only so far as the neural and haemal elements are concerned. All are eminently heterocercal. Ribs are present in most, but are replaced by ligaments in Polyodon.

The primordial cranium of the sturgeons consists of persistent cartilage without ossifications in its substance, but superficial bones are still more developed and specialized than in the Dipnoi; so it is, at least, in the true sturgeons, but less so in Polyodon (fig. 37). The upper and lateral parts of the skull are covered by well-developed membranebones. The lower surface of the skull is covered by an extremely large basal bone, which extends from the vomerine region on to the anterior part of the spinal column. The nasal excavation in the skull is rather lateral than inferior. The ethmoidal region is generally much produced, forming the base of the long projecting snout. The suspensorium is movably attached to the side of the skull, and consists of two pieces, a hyomandibular and a symplectic, which now appears for the first time as a separate piece, and to which the hyoid is. attached. The palato maxillary apparatus is more complex than in the sharks and Dipnoi ; a palato-pterygoid consists of two mesially-connected rami in Polyodon, and of a complex cartilaginous disk in Acipenser, being articulated in both to the Meckelian cartilage. In addition, the sturgeons possess one or two pairs of osseous rods, which, in Polyodon at least, represent the maxillary, and therefore must be the representatives of the labial cartilages of the sharks. The Meckelian cartilage is more or less covered by tegumentary bones. In the gill-cover, besides the operculum, a sub-and inter-operculum may be distinguished in Acipenser. The hyoid consists of three pieces, of which the posterior bears a broad branchiostegal in Polyodon.

In the scapulary arch the primordial cartilaginous elements scarcely differ from those of the Dipnoi. The membrane-bones are much expanded, and show a continuous series suspended from the skull. Their division in the median ventral line is complete. The pectoral is supported by a cartilaginous framework (fig. 38) similar to that of Ceratodus, but much more shortened and reduced in its periphery, the branches being absent altogether on one side of the axis. This modification of the fin is analogous to the heterocercal condition of the end of the spinous column. To the inner corner of a basal cartilage (a) a short axis (6) is joined, which on its outer side bears a few branches (d) only, the remaining branches (c) being fixed to the basal cartilage. The dermal fin-rays are opposed to the extremities of the branches, as in the Dipnoi.

The pubic consists of a paired cartilage, to which tarsal pieces supporting the fin-rays are attached.

The other existing Ganoid fishes have the spinous column entirely or almost entirely ossified; these have been comprised under the common name Holostei. They form, however, three very distinct types; several attempts have been made to coordinate with them the fossil forms, but this task is beset with extreme difficulties, of which no satisfactory solution has as yet been advanced.

The Polypteroidei have their spinous column formed by distinct osseous amphicoelous vertebrae, that is, vertebrae with concave anterior and posterior surfaces. It is nearly diphycercal. The neural arches, though ossified, do not coalesce with the centrum, and form one canal only for the myelon. There are no intermediate elements between the neural spines.

Interneurals are developed, but are simple, articulating with the dermoneurals. The abdominal vertebrae have parapophyses developed with epipleural spines. Only the caudal vertebrae have haemal spines, which, like the interhaemals, agree in every essential respect with the opposite neurals. Ribs are inserted, not on the parapophyses, but on the centra, immediately below the parapophyses.

The skull of Polypterus (fig. 39) shows a great advance towards the Teleosteous type, the number of separable bones being greatly increased. They are arranged much in the same fashion as in Teleostei, but a great portion of the primordial cranium remains cartilaginous. The membrane-bones which cover the upper and lower surfaces of the brain-case are so much developed as to cause the underlying cartilage to disappear, so that a large vacuity or fontanelle exists in the substance of the upper as well as of the lower cartilaginous wall. Of ossifications belonging to the primordial skull must be noticed the single occipital with a mastoid on each side. They are separated by persistent cartilage from the sphenoids and postfrontals; the former, which are the largest ossification of the primordial cranium, enclose the anterior half of the brain cavity. Finally, the nasal portion contains a median ethmoid and a pair of praefrontal bones. Only a very small portion of the bones described are visible externally, nearly the whole of the primordial cranium being covered by the membrane-bones. Of these are seen on the upper surface a pair of pariétale, frontale, “nasals,” and turbinals; on the lower surface a large cross-shaped basal, anteriorly bordered on each side by a pterygoid, parallel to a palatine which forms a suture with the double vomer. The suspensorium has in front a metapterygoid and quadrate bone, and an operculum and suboperculum are attached to it behind. Praemaxillaries and maxillaries are now fully developed, but immovably attached to the skull. The lower jaw is ossified, and consists of an articulary, angular, dentary, and splenial. Of labial cartilages a rudiment at the angle of the mouth has remained persistent. The side of the skull, in front of the operculum, is covered by a large irregularly-shaped bone(7 τ ) (corresponding to the “tympanic lamina” of Ceratοdus, fig. 35, <f), held by some to be the praeoperculum; along its upper circumference lies a series of small ossicles, of which two may be distinguished as spiraculars, as they form a valve for the protection of the spiracular orifice of these fishes. An infraorbital ring is represented by a praeorbital and postorbital only.

Each hyoid consists of three pieces, none of which bear branchiostegals, the single median piece being osseous in front and cartilaginous behind. Four branchial arches are developed, the foremost consisting of three, the second and third of two, and the last of a single piece. There is no lower pharyngeal. Between the rami of the lower jaw the throat is protected by a pair of large osseous laminae (gular plates), which have been considered to represent the urohyal of osseous fishes. The scapulary arch is almost entirely formed by the well-developed membrane-bones, which in the ventral line are suturally united. The pectoral fin is supported by three bones, pro-, meso-, and meta-pterygium, of which the dilated middle one alone bears rays, and is excluded from the articulation with the shoulder-girdle.

The pubic consists of a paired bone, to which tarsal bones supporting the fin-rays are attached.

In the Lepidosteoidei the vertebrae are completely ossified and opisthoccelous, having a convexity in front and a concavity behind, as in some amphibians. Though the end of the body externally appears nearly diphycercal, the termination of the vertebral column is, in fact, distinctly heterocercal. The caudal fin is suspended from haemapophyses only, and does not extend to the neural side of the vertebral column. The neural arches coalesce with the centrum; the interneurals are simple. The abdominal vertebrae have parapophyses, to which the ribs are attached. Only the caudal vertebrae have haemal spines.

In the skull of Lepidosteus the cartilage of the endocranium is still more replaced by ossifications than in Polypterus, those ossifications, moreover, being represented by a greater number of discrete bones. The membrane-bones in particular are greatly multiplied: the occipital, for instance, consists of three pieces; the vomer is double, as in Polypterus ; the maxillary consists of a series of pieces firmly united by suture. The symplectic reaches the lower jaw, so that the articulary is provided with a double joint,viz., for the symplectic and quadrate; the component parts of the lower jaw are as numerous as in reptiles, a dentary, splenial, articulary, angular, supra-angular, and coronary being distinct. The sides of the head are covered with numerous bones, and a praeoperculum is developed in front of the gill-cover, which, again, consists of an operculum and suboperculum.

Each hyoid consists of three pieces, of which the middle is the longest, the upper bearing the largest of the three branchiostegals which Lepidosteus possesses; a long and large glossohyal is intercalated between the lower ends of the hyoids. There are five branchial arches, the hindmost of which is modified into a lower pharyngeal; upper pharyngeals are likewise present as in the majority of Teleosteous fishes, but there is no gular plate.

Of the scapulary arch the two halves are separated by a suture in the median line; the membrane-bones are well developed, only a remnant of the primordial cartilage remaining; the supraclavicle is very similar to that of Teleosteous fishes, and in a less degree the post-temporal. The base to which the limb is attached is a single osseous plate, supporting on its posterior margin semi-ossified rods in small number, which bear the pectoral rays. The pubic consists of a paired bone. The elements representing a tarsus are quite rudimentary, and reduced in number (two or three).

The vertebral column of the Amioidei shows unmistakable characters of the Palaeichthyic type. The arrangement of its component parts is extremely simple. The centra of the amphicoelous vertebrae are well ossified, but the neural and haemal arches do not coalesce with the centra, from which they are separated by a thin layer of cartilage. All the vertebrae do not possess apophyses: in the caudal portion of Amia only every alternate vertebra is provided with them. The heterocercal condition of the spinous column is well marked; as in the other Holostei, the hindmost vertebrae are turned upwards, become gradually smaller in size, and lose their neural arches, the haemals remaining developed to the end. Finally, the column terminates in a thin cartilaginous band, which is received between the lateral halves of the fifth or sixth upper caudal ray. The interneurals and interhaemals are simple. Only the abdominal vertebrae have parapophyses, with which the ribs are articulated.

The configuration of the skull, and the development and arrangement of its component parts, approach so much the Teleostean type, that perhaps there are greater differences in skulls of truly Teleostean fishes than between the skulls of Amia and of many Physostοmi. Externally the cranium is entirely ossified; the remains of the cartilaginous primordial cranium can only be seen in a section, and are of much less extent than in many Physo-stomous fishes. The immovable intermaxillary, the double yomer, the double articulary cavity of the mandible for junction with the quadrate and symplectic bones, remind us still of similar conditions in the skull of Lepidosteus ; but the mobility and formation of the maxillary, the arrangement of the gill-covers, the development of a praeoperculum, the suspensorium, the palate, the insertion of a number of branchiostegals on the long middle hyoid piece, the composition of the branchial framework (with upper and lower pharyngeals), are as in the Teleosteous type. A gular plate replaces the urohyal.

The scapular arch is composed entirely of the membranebones found in the Teleostei, and the two sides are loosely united by ligament. The base to which the limb is attached is cartilaginous; short semi-ossified rods are arranged along its hinder margin, and bear the pectoral rays.

Myology.

In the lowest vertebrate, Branchiostoma, the whole of the muscular mass is arranged in a longitudinal band running along each side of the body; it is vertically divided into a number of flakes or segments (myocommas) by aponeurotic septa, which serve as the surfaces of insertion to the muscular fibres. But this muscular band has no connexion with the notochord except in its foremost portion, where some relation has been formed to the visceral skeleton. A very thin muscular layer covers the abdomen. In the Cyclostomes also the greater portion of the muscular system is without direct relation to the skeleton, and, again, it is only on the skull and visceral skeleton that distinct muscles have been differentiated for special functions.

To the development of the skeleton in the more highly organized fishes corresponds a similar development of the muscles: the maxillary and branchial apparatus, the pectoral and ventral fins, the vertical fins and especially the caudal, possess a separate system of muscles. But the most noteworthy is the muscle covering the sides of the trunk and tail (already noticed in Branchiostoma) i which Cuvier described as the “great lateral muscle,” and which, in the higher fishes, is a compound of many smaller segments (myocommas), corresponding in number with the vertebrae. Each lateral muscle is divided by a median longitudinal groove into a dorsal and ventral half; the depression in its middle is filled by an embryonal muscular substance which contains a large quantity of fat and numerous blood-vessels, and therefore differs from ordinary muscle by its softer consistency, and by its colour, which is reddish or greyish. Superficially the lateral muscle appears crossed by a number of white parallel tendinous zigzag stripes, forming generally three angles, of which the upper and lower point backwards, the middle one forwards. These are the outer edges of the aponeurotic septa between the myocommas. Each septum is attached to the middle and to the apophyses of a vertebra, and, in the abdominal region, to its rib; frequently the septa receive additional support by the existence of epipleural spines.

In connexion with the muscles reference has to be made to the electric organs with which certain fishes are provided. That these have been developed out of muscular substance is more than probable, not only from the examination of peculiar muscular organs (the function of which is still conjectural) occurring in the rays, and in Mormyrus and Gymnarchus, but especially from the researches into the development of the electric organ of Torpedo. The fishes possessing fully developed electric organs, with the power of accumulating electric force and communicating it in the form of shocks to other animals, are the electric rays (Torpedinidoe), the electric sheath-fish of tropical Africa {Malapterurus∖ and the electric eel of tropical America (Gymnotus'). The structure and arrangement of the electric organ are very different in these fishes.

In Torpedo the electric organs are large, flat, uniform bodies, lying one on each side of the head, bounded behind by the scapular arch, and laterally by the anterior crescent-shaped tips of the pectoral fins. They consist of an assemblage of vertical hexagonal prisms, whose ends are in contact with the integuments above and below; and each prism is subdivided by delicate transverse septa, forming cells filled with a clear, trembling, jelly-like fluid, and lined within by an epithelium of nucleated corpuscles. Between this epithelium and the transverse septa and walls of the prism there is a layer of tissue on which the terminations of the nerves and vessels ramify. Hunter counted four hundred and seventy prisms in each battery of Torpedo marmorata, and demonstrated the enormous supply of nervous matter which they receive. Each organ receives one branch of the trigeminal nerve and four branches of the vagus, the former, and the three anterior branches of the latter, being each as thick as the spinal chord (electric lobes).

In Malapterurus the electric organ extends over the whole body, but is thickest on the abdomen; it lies between two aponeurotic membranes below the skin, and consists of rhomboidal cells which contain a rather firm gelatinous substance. The electric nerve takes its origin from the spinal chord, does not enter into connexion with ganglia, and consists of a single enormously-strong primitive fibre, which distributes its branches in the electric organ.

The electric eel is the most powerful of electric fishes. Its electric organ consists of two pairs of longitudinal bodies, situated immediately below the skin, above the muscles,—one pair on the back of the tail, and the other pair along the anal fin. Each fasciculus is composed of flat partitions or septa, with transverse divisions between them. The outer edges of the septa appear in nearly parallel lines in the direction of the longitudinal axis of the body, and consist of thin membranes, which are easily torn; they serve the same purpose as the columns in the analogous organ of the torpedo, making the walls or abutments for the perpendicular and transverse dissepiments, which are exceedingly numerous, and so closely aggregated as to seem almost in contact. The minute prismatic cells, intercepted between these two sorts, of plates, contain a gelatinous matter; the septa are about - s ⅛th of an inch from each other, and a length of one inch contains a series of two hundred and forty cells, giving an enormous surface to the electric organs. The whole apparatus is supplied with more than two hundred nerves, which are the continuations of the rami anteriores of the spinal nerves. In their course they give out branches to the muscles of the back, and to the skin of the animal. In the Gymnotes, as in the torpedo, the nerves supplying the electric organs are much larger than those bestowed on any part for the purposes of sensation or movement.

The. phenomena attending the exercise of this extraordinary faculty closely resemble muscular action. The time and strength of the discharge are entirely under the control of the fish. The power is exhausted after some time, and needs repose and nourishment to restore it. If the electric nerves are cut and divided from the brain, the cerebral action is interrupted, and no irritant to the body has any effect in exciting electric discharge; but if their ends be irritated the discharge takes place, just as a muscle is excited to contraction under similar circumstances. Singularly enough, also, the application of strychnine causes simultaneously a tetanic state of the muscles and a rapid succession of involuntary electric discharges. The strength of the discharges depends entirely on the size, health, and energy of the fish, an observation completely agreeing with that made on the efficacy of snake-poison. Like this latter, the property of the electric force serves two ends in the economy of the animals which are endowed with it: it is necessary to them for overpowering, stunning, or killing the creatures on which they feed, whilst incidentally they use it as the means of defending themselves from their enemies.

Neurology.

The most simple condition of the nervous central organ known in vertebrates is found in Branchiostoma. In this fish the spinal cord tapers at both ends, no anterior cerebral swelling, or anything approaching a brain, being present. It is band-like along its middle third, and groups of darker cells mark the origins of the fifty or sixty pairs of nerves which accompany the intermuscular septa, and divide into a dorsal and a ventral branch, as in other fishes. The two anterior pairs pass to the membranous parts above the mouth, and supply with nerve filaments a ciliated depression near the extremity of the fish, which is considered to be an olfactory organ, and two pigment spots, the rudiments of eyes. An auditory organ is absent. The spinal cord, of the Cyclostomes is flattened in its whole extent, bandlike, and elastic; also in Chimaera it is elastic, but flattened in its posterior portion only. In all other fishes it is cylindrical, non-ductile, and generally extending along the whole length of the spinal canal. The Plectognaths offer a singular exception in this respect, that the spinal cord is much shortened, the posterior portion of the canal being occupied by a long cauda equina; this shortening of the spinal cord has become extreme in the sun-fish (f)rthagoriscus), in which it has shrunk into a short and conical appendage of the brain. In the devil-fish {Lophius) also a long cauda equina partly conceals the cord which terminates on the level of about the twelfth vertebra.

The brain of fishes is relatively small; in the burbot (Lota) it has been estimated to be the τ ⅜ σ th part of the weight of the entire fish, in the pike the - r ⅛5th part, and in the large sharks it is relatively still smaller.

The brain of osseous fishes (fig. 40) viewed from above shows three protuberances, respectively termed the prosencephalon, mesencephalon, and metencephalon, the two anterior of which are paired, the hindmost being single. The foremost pair are the hemispheres, which are solid in their interior, and provided with two swellings in front, the olfactory lobes. The second pair are the optic lobes, which generally are larger than the hemispheres, and succeeded by the third single portion, the cerebellum. The optic lobes possess a cavity (ventriculus lobi optici), at the bottom of which some protuberances of variable development represent the corpora quadrigemina of higher animals. On the lower surface of the base of the optic lobes, behind the crura cerebri, two swellings are observed, the lobi inferiores, which slightly diverge in front for the passage of the infundibulum, from which a generally large hypophysis or pituitary gland is suspended. The relative size of the cerebellum varies greatly in the different osseous fishes: in the tunny and silurus it is so large as nearly to cover the optic lobes; sometimes distinct transverse grooves and a median longitudinal groove are visible. The cerebellum possesses in its interior a cavity which communicates with the anterior part of the fourth ventricle. The medulla oblongata is broader than the spinal cord, and contains the fourth ventricle. In most fishes a perfect roof is formed over the fourth ventricle by two longitudinal pads, which meet each other in the median line (lobi posteriores).

The brain of Ganoid fishes shows great similarity to that of the Teleostei∙, there is, however, considerable diversity in the arrangement of its various portions in the different types. In the sturgeons and Polypterus (fig. 41) the hemispheres are more or less remote from the mesencephalon, so that in an upper view the crura cerebri, with the intermediate entrance into the third ventricle (fissura cerebri magna), may be seen. A vascular membranous sac, epiphysis, containing lymphatic fluid, takes its origin from the third ventricle, its base being expanded over the anterior interspace of the optic lobes, and the apex being fixed to the cartilaginous roof of the cranium. This structure is not peculiar to the Ganoids, but is found in various stages of development in Teleosteans, marking, when present, the boundary between the prosencephalon and mesencephalon. The lobi optici are essentially as in Teleosteans. The cerebellum penetrates into the ventriculus lobi optici, and extends thence into the open sinus rhomboidalis. At its upper surface it is crossed by a commissure formed by the corpora restiformia of the medulla.

As regards external configuration, the brain of Lepidosteus and Amia approach still more the Teleosteous type. The prosencephalon, mesencephalon, and metencephalon are contiguous, and the cerebellum lacks the prominent transverse commissure at its upper surface. The sinus rhomboidalis is open.

The brain of Chondropterygians (fig. 42) is more developed than that of other fishes, and is distinguished by well-marked characters. These are— first, the prolongation of the olfactory lobes into pedicles of greater or less length, which dilate into great ganglionic masses, where they come into contact with the olfactory sacs; secondly, the space which generally intervenes between the prosencephalon and mesencephalon, as in some Ganoids; thirdly, the large development of the metencephalon.

The cerebellum is very large, overlying a portion of the optic lobes and of the sinus rhomboidalis, and is often transversely grooved. The side-walls of the fourth ventricle, which are formed by the corpora restiformia, are singularly folded, and appear as two pads, one on each side of the cerebellum (lobi posteriores or lobi nervi trigemini).

The brain of the Cyclostomes represents a type different from that of other fishes, showing at its upper surface three pairs of protuberances in front of the cerebellum; they are all solid. The foremost pair are the large olfactory tubercles, which are extremely large in Petromyzon. They are followed by the hemispheres, with a single body wedged in between their posterior half; in Petromyzon, at least, the vascular tissue leading to an epiphysis seems to be connected with this body. Then follows the lobus ventriculi tertii, distinctly paired in Myxinoids, but less distinctly in Petromyzon. The last pair are the corpora quadrigemina. According to this interpretation, the cerebellum would be absent in Myxinoids, and represented in Petromyzon by a narrow commissure only, stretching over the foremost part of the sinus rhomboidalis. In the Myxinoids the medulla oblongata ends in two divergent swellings, free and obtuse at their extremity, from which most of the cerebral nerves take their origin.

Two very important conditions require mention. The first relates to the optic nerves, viz., to their mutual relation immediately after their origin, which is very characteristic of the subclasses of fishes. In the Cyclostomes they have no further connexion with each other, each going to the eye of its own side. In the Teleostei (fig.

40, n) they simply cross each other (decussate), so that the one starting from the right half of the brain goes to the left eye and vice versa. Finally, in the Palaeichthyes (fig.

41, f) the two nerves are fused together, immediately after their origin, into a chiasma.

The second noteworthy peculiarity occurs in the distribution of the nervus vagus; it emits a strong branch, called nervus lateralis, which accompanies the lateral mucous system of the trunk and tail. This is either a single longitudinal stem, gradually becoming thinner behind, and running superficially below the skin (Sal- monidoe, Cyclopterus) or deeply between the muscles (sharks, Chimoeraf οτ is divided into two parallel branches (most Teleostei) ; thus in the perch there are two branches on each side, the superficial one supplying the lateral line, whilst the deep-seated branch communicates with the spinal nerves and supplies the septa between the myocommas and the skin. In fishes which lack the lateral muciferous system and possess hard integuments, as the Ostracions, the lateral nerve is more or less rudimentary. It is entirely absent in Myxinoids, but the gastric branches of the vagus are continued, united as a single nerve, along the intestine to the anus.

Fishes possess all the spino-cerebral nerves of the higher Vertebrata, with the exception of the nervus accessorius. A separate nervus hypoglossus is also absent, but elements from the first spinal nerve are distributed over the area normally supplied by this nerve in higher vertebrates. The number of spinal nerves corresponds to that of the vertebrae, through or between which they pass out.

A sympathie nervous system appears to be absent in Branchiostoma, and has not yet been clearly made out in the Cyclostomes. It is well developed in the Palaichthyes, but without the cephalic portion. This latter is present in all osseous fishes, in which the communication of sympathy has been found to exist between all the cerebral nerves, except the olfactory, optic, and acoustic. The sympathie trunks run along each side of the aorta and the back of the abdomen into the haemal canal, communicate in their course with the ventral branches of each of the spinal nerves, and, finally, often blend together into a common trunk beneath the tail. At the points of communication with the cerebral and spinal nerves ganglia are frequently developed, from which nerves emerge which are distributed to the various viscera.

Organs of Sense. Organ of Smell.— It is characteristic of the organ of smell in fishes that it has no relation whatever to the respiratory function, with the exception of the Dipnoi, in which possibly part of the water received for respiration passes through the nasal sac.

The olfactory organ is single in Branchiostoma and the Cyclostomes. In the former a small depression on the front end of the body, clothed with a ciliated epithelium, is regarded as a rudimentary organ of smell. In the adult Petromyzon a membranous tube leads from the single opening on the top of the head into the cartilaginous olfactory capsule, the inside of which is clothed by membranes prolonged into a posterior blind tube (fig. 31, s), which penetrates the cartilaginous roof of the palate, but not the mucous membrane of the buccal cavity. In the Myxinoids the outer tube is strengthened by cartilaginous rings like a trachea; the capsule is lined by a longitudinally folded pituitary membrane, and the posterior tube opens backwards on the roof of the mouth, the opening being provided with a valve. In all other fishes the organ of smell is double, there being one on each side; it consists of a sac lined with a pituitary membrane, and may be provided with one or with two openings, or may have none. The position of these openings is very different in the various orders and suborders of fishes.

It is certain that fishes possess the faculty of perceiving odours, and that various scents attract or repel them. A mangled carcase or fresh blood attracts sharks, as well as the voracious Serrasalmonoids of the South American rivers. There is no reason to doubt that the seat of that perception is in the olfactory sac; and it may be reasonably conjectured that its strength depends mainly on the degree of development indicated by the number and extent of the interior folds of the pituitary membrane.

Organ of Sight.— The position, direction, and dimensions of the eyes of fishes vary greatly. In some they have an upward aspect, and are often very close together; in others they are lateral, and in a few they are even directed downwards. In a very few this organ appears to be entirely absent. In some Gobioids and Trachinoids lΡeriophthal-rnus, Boleophthalmus, Uranoscopus, <fcc,) the eyes, which are on the upper side of the head, can be elevated and depressed at the will of the fish. In the range of their vision and in their acuteness of sight, fishes are very inferior to the higher classes of vertebrates, yet at the same time it is evident that they perceive their prey, or approaching danger, from a considerable distance; and it would appear that the visual powers of a Periophthalmus (fig. 12), when hunting insects on the mud-flats of the tropical coasts, are quite equal to those of a frog.

The eye of Braηchiostoma (fig. 30, g} is of the most rudimentary kind. It is simply a minute speck coated with a dark pigment, and receiving the end of a short nerve. In Myxinoids the minute rudiment of the eye is covered by skin and muscles. This is also the case in many of the blind Teleosteous fishes; whilst in the former fishes, however, the organ of sight has not attained to any degree of development, the rudimentary eye of blind Teleostei is a retrogressive formation, in which a lens and other portions of the eye can often be recognized. In fishes with a well-developed eye it is imbedded in a layer of gelatinous and adipose substance, which covers the cavity of the orbit. A lachrymal gland is absent In the Chondropterygians the eyeball is supported by and moves on a cartilaginous peduncle of the orbital wall. In the majority of Teleosteans, and in Acipenser, a fibrous ligament attaches the sclerotic to the wall of the orbit. The proper muscles of the eyeball are always present. In all fishes the general integument of the head passes over the eye, and becomes transparent where it enters the orbit; sometimes it simply passes over the orbit, sometimes it forms a circular fold. The anterior and posterior portions may be specially broad, becoming the seat of an adipose deposit (adipose eyelids), as in Scomber, Caranx, Mugil, &c. In many of these fishes the extent of the eyelids varies with the seasons; during the spawning season they are so much loaded with fat as nearly to hide the whole eye. Many sharks possess a nictitating membrane, developed from the lower part of the palpebral fold, and moved by a proper set of muscles.

The form of the bulhus (fig. 43) is subhemispherical, the cornea (co) being flat. If it were convex, as in higher vertebrates, it would be more liable'to injury; but, as it is level with the side of the head, the chances of injury by friction are diminished. The sclerotica (sc) is cartilaginous in Chondropterygians and Acipensers, fibrous and of varying thickness in Teleosteans, in the majority of which it is supported by a pair of cartilaginous or ossified hemispheroid cups (c). In a few fishes, as in Cerαtοdus, Xiphias, the cups are confluent into one cup, which possesses a foramen behind to allow the passage of the optic nerve (o).

The membranes situated between the sclerotica and retina are collectively called choroidea, and are three in number. The one in immediate contact with the sclerotic, and continued upon the iris, is by no means constantly present; it is the membrana argentea (a), and is composed of microscopical crystals reflecting a silvery or sometimes golden lustre. The middle layer is the membrana vasculosa or Halleri (v), the chief seat of the ramifications of the choroid vessels; the innermost layer is the membrana Ruyscheana or uvea (u), which is composed of hexagonal pigment-cells, usually of a deep-brown or black colour.

In many Τeleοslei a rete mirabile surrounds the entry of the op tie nerve; it is situated between the membrana argentea and vasculosa, and called the choroid gland (cΛ). It receives its arterial blood from the artery issuing from the pseudobranchia, the presence of a choroid gland being always combined with that of a pseudobranchia. Teleosteans without pseudobranchia have no choroid gland. In the Palaeiehthyes, on the other hand, the pseudobranchia is present and a choroid gland absent.

The iris (i) is merely the continuation of the choroid membrane; its capability of contracting and expanding is much more limited than in higher vertebrates. The pupil is generally round, sometimes horizontally or vertically elliptical, sometimes fringed. In the rays and Pleuronectidae a lobe descends from the upper margin of the pupil, and the outer integument overlying this lobe is coloured and non-transparent,—a structure evidently preventing light from entering the eye from above.

In most Teleostei a fold of the choroidea, called the processus falciformis (∕), extends from the vicinity of the entrance of the optic nerve to the lens. It seems to be constantly absent in Ganoids..

The vitreous humour (Λ), which fills the posterior cavity of the eye-ball, is of a firmer consistency than in the higher vertebrates. The lens is spherical, or nearly so, firm, denser towards the centre, and lies in a hollow of the vitreous humour. When a falciform process is present, it has one end attached to the lens, which is thus steadied in its position. It is composed of concentric layers consisting of fibres, which in the nucleus of the body have marginal teeth, by which they are interlocked together. In Petromyzon this serrature is absent, or but faintly indicated.

The anterior cavity of the eye is very small in fishes, in consequence of the small degree of convexity in the cornea; the quantity of the aqueous humour, therefore, is very small, just sufficient to float the free border of the iris; and the lessened refractive power of the aqueous humour is compensated by the greater convexity of the lens.

Organ of Hearing.— No trace of an organ of hearing has been found in Branchiostorna. In the Cyclostomes the labyrinth is enclosed in externally visible cartilaginous capsules laterally attached to the skull; it consists of a single semicircular canal in the Myxinoids, whilst the Petromyzontes possess two semicircular canals with a vestibulum.

In all other fishes the labyrinth consists of a vestibule and three semicircular canals, the vestibule dilating into one or more sacs, which contain the otoliths. A tympanum, tympanic cavity, and external parts are entirely absent in the class of fishes.

In the Chondropterygians and Dipnoi, the labyrinth is enclosed in the cartilaginous substance of the skull. In the former the excavation in the cartilage is larger than the membranous labyrinth, but nearly corresponds to it in form; the part which receives the membranous vestibulum is called vestibulum cartilagineum, from which a canal issues and penetrates to the surface of the skull, where it is closed by the skin in sharks, but opens by a minute foramen in rays. The otolithic contents are soft and chalklike.

In the Teleosteans the sac which contains the otoliths lies on each side of the base of the cranial cavity, and is often divided by a septum into two compartments of unequal size, each containing a firm and solid otolith (fig. 44); these bodies possess indented margins, frequently other impressions and grooves, in which nerves from the N.

acousticus are lodged; they vary much in size and form, but in both respects show a remarkable constancy in the same kind of fishes. Outwards the vestibule is in contact with the osseous side wall of the skull, inwards with the metencephalon and medulla oblongata; it contains another firm concretion, and opens by five foramina into the three semicircular canals. The terminations of the acoustic nerve are distributed over the vestibular concretion and the ampulli-form ends (fig. 45) of the semicircular canals, without being continued into the latter, which are filled with fluid. The semicircular canals (fig. 45) are sometimes lodged in the cranial bones, sometimes partly free in the cranial cavity. Many Teleostei have fontanelles in the roof of the skull, closed only by skin or very thin bone at the place where the auditory organ approaches the surface, by which means sonorous undulations must be conducted with greater ease to the ear.

In many Teleostei a most remarkable relation obtains between the organ of hearing and the air-bladder. In the most simple form this connexion is established in Percoids and the allied families, in which the two anterior horns of the air-bladder are attached to fontanelles of the occipital region of the skull, the vestibulum occupying the opposite side of the membrane by which the fontanelle is closed. . The condition is similar, but more coιhplicated, in many Clupeoids. The anterior narrow end of the air-bladder is produced into a canal at the base of the skull, and divided into two very narro w branches, which again bifurcate and terminate in a globular swelling. An appendage of the vestibulum meets the anterior of these swellings, and comes into close contact with it. In addition, the two vestibules communicate with each other by a transverse canal, crossing the cranial cavity below the brain.

The connexion is effected by means of a chain of ossicles in Siluridoe, Characinidae, Cyprinidae, and Gymnotidoe. A canal issues from the communication between the vestibule and its sac, and meeting that from the other side forms with it a common sinus impar (fig. 45, r), lodged in the substance of the basioccipital; this communicates on each side by a small orifice with two sub-spherical atria, on the body of the atlas, close to the foramen magnum. Each atrium is supported externally by a small bone (∞); a a third and larger bone (k) completes the communication with the anterior part of the air-bladder. From the sinus impar a bifid canal penetrates into the alisphenoids, in which it terminates. In Cobitis and several loach-like Siluroids the small air-bladder consists of two globular portions placed side by side, and wholly included within two bullae formed by the modified parapophyses of the second and third vertebrae. The three ossicles on each side are present, but concealed by the fore part of the osseous bulla.

Organ of Taste.— Some fishes, especially vegetable feeders, or those provided with broad molar-like teeth, masticate their food; and it may be observed in carps and other Cyprinoid fish that this process of mastication frequently takes some time. But the majority of fishes swallow their food rapidly, and without mastication, and therefore we may conclude that the sense of taste cannot be keen. The tongue is often entirely absent, and, even when it exists in its most distinct state, it consists merely of ligamentous or cellular substance, and is never furnished with muscles capable of producing the movements of extension or retraction as in most of the higher vertebrates. A peculiar organ on the roof of the palate of Cyprinoids is perhaps adapted for the perception of this sense; in these fishes the palate between and below the upper pharyngeal bones is cushioned with a thick, soft, contractile substance, richly supplied with nerves from the nervus vagus and nervus glosso-pharyngeus.

Organs of Touch.— The faculty of touch is more developed than that of taste, and there are numerous fishes which possess special organs of touch. Most fishes are very sensitive to external touch, although their body may be protected by hard horny scales. They perceive impressions even on those parts which are covered by osseous scutes, in the same manner as a tortoise perceives the slightest touch of its carapace. The seat of the greatest sensitiveness however, appears to be the snout and the labial folds surrounding the mouth. Many species possess soft and delicate appendages, called barbels, which are almost constantly in action, and clearly used as organs of touch. Among the Triglidae and allied families, there are many species which have one or more rays of the pectoral fin detached from the membrane, and supplied with strong nerves. Such detached rays (also found in the Polynemidoe and in Bathypterois) are used partly for locomotion, partly for exploring the ground over which the fish moves.

Some fishes appear to be much less sensitive than others, or at least lose their sensitiveness under peculiar circumstances. It is well known that a pike whose mouth has been lacerated and torn by the hook continues to yield to the temptation of a bait immediately afterwards. The Greenland shark, when feeding on the carcase of a whale, allows itself to be repeatedly stabbed in the head without abandoning its prey. A pair of congers are so dead to external impression at the time of copulation, and, as it were, so automatically engaged, that they have been taken by the hand together out of the water.

Obgans of Nutbition and Digestion.

Fishes are for the most part either exclusively carnivorous or herbivorous, but not a few feed on vegetable as well as animal substances, or on mud containing alimentary matter in a living or decomposing state. Generally they are very voracious, especially the carnivorous kinds, and the rule of “eat or be eaten” applies to them with unusual force. They are almost constantly engaged in the pursuit and capture of their prey, the degree of their power in these respects depending on the dimensions of the mouth and gullet and the strength of the teeth and jaws. If the teeth are sharp and hooked, they are capable of securing the most slender and agile animals; if with teeth of this kind are combined a wide gullet and distensible stomach, the fish is able to overpower and swallow others larger than itself; if the teeth are broad, strong molars, they are able to crush the hardest alimentary substances; if they are feeble, they are only serviceable in procuring some small or inert and unresisting prey. Teeth may be wanting altogether. Whatever the prey, in the majority of cases it is swallowed whole; but some of the most voracious fishes, like some sharks and Characinidoe, are provided with cutting teeth, which enable them to tear their prey to pieces if too large to be swallowed whole. Auxiliary organs, similar to the claws of some carnivorous mammals and birds, for the purpose of seizing and overpowering their prey before it is torn by the teeth, are not found in this class; but in a few fishes the jaws themselves are modified for that purpose. In the sword-fishes the bones of the upper jaw form a long dagger-shaped weapon, with which they not only attack large animals, but also frequently kill fishes on which they feed. The saw-fishes are armed with a similar but still more complex weapon, the saw, which is armed on each side with large teeth implanted in deep sockets, specially adapted for killing and tearing the prey before it is seized and masticated by the small teeth within the mouth. Fishes show but little choice in the selection of their food, and some devour their own offspring indiscriminately with other fishes. Their digestive powers are strong and rapid, but are affected in some degree by the temperature, which, when it sinks below a certain point, lowers the vital powers of these cold-blooded animals. On the whole, marine fishes are more voracious than those inhabiting fresh waters; and, whilst the latter may survive total abstinence from food, for weeks or months, the marine species succumb to hunger within a few days.

The organs of nutrition, manducation, and deglutition are lodged in two large cavities—an anterior (the mouth or buccal cavity), and a posterior (the abdominal cavity). In the former the alimentary organs are associated with those fulfilling the respiratory functions, the transmission of food to the stomach and of water to the gills being performed by similar acts of deglutition. The abdominal cavity commences immediately behind the head, so, however, that an extremely short thoracic cavity for the heart is partitioned off in front. Besides the alimentary organs it contains also those of the urogenital system and the air-bladder. The abdominal cavity is generally situated in the trunk only, but in numerous fishes it extends into the tail, being continued for some distance along each side of the haemal apophyses. In numerous fishes the abdominal cavity opens outwards by one or two openings. A single porus abdominalis in front of the vent is found in Lepidosiren and some sturgeons; a paired one, opening on each side of the vent, in Cerαtοdus, some species of sturgeon, Lepidosteus, Polypterus, Amia, and all Chondropterygians. As in these fishes semen and ova are discharged by their proper ducts, the abdominal openings may serve for the expulsion of semen, and of those ova only which, having lost their way to the abdominal aperture of the oviduct, would be retained in the abdominal cavity. In those Teleosteans which lack an oviduct a single porus genitalis opens behind the vent.

Mouth.— The mouth of fishes shows extreme variation with regard to form, size, and position. Generally opening in front, it may be turned upwards, or it may lie at the lower side of the snout, as in most Chondropterygians, sturgeons, and some Teleosteans. In most fishes the jaws are covered by the skin, which, before passing over the jaws, is often folded, forming more or less fleshy lips. In the sharks the skin retains its external character even within the teeth, but in other fishes it changes into a mucous membrane. A tongue may exist as a more or less free and short projection, formed by the glossohyal and a soft covering, or may be entirely absent. Salivary glands and a velum palati are absent in fishes.

Teeth.— With regard to the dentition, the class of fishes offers an amount of variation such as is not found in any of the other classes of vertebrates. As the teeth form one of the most important elements in the classification of fishes, their special arrangement and form will be referred to in the account of the various families and genera. Whilst not a few fishes are entirely edentulous, in others most of the bones of the buccal cavity, or some of them, may be toothed, as the bones of the jaws, the palatines, pterygoids, vomers, basi-sphenoid, glossohyal, branchial arches, upper and lower pharyngeals. In others teeth may be found fixed in some portion of the buccal membrane without being supported by underlying bone or cartilage; or the teeth have been developed in membrane overlying one of the dentigerous bones mentioned, without having become anchylosed to the bone. When the tooth is fixed to the bone the attachment has generally been effected by the ossification of the bone of the tooth, but in some fishes a process of the bone projects into the cavity of the tooth; in others the teeth are implanted in alveoli. In these, again, frequently a process of bone rises from the bottom, on which the tooth rests.

Many of the class, especially predatory fishes, with long, lancet-shaped teeth, have all or some of these capable of being bent towards the interior of the mouth. Such “hinged” teeth resume at once the upright position when pressure is removed from them. They are, however, depressible in one direction only, thus offering no obstacle to the ingress, while they oppose the egress of prey. Mr C. S. Tomes has shown that the means by which this mechanism is worked are different in different fishes; for, whilst in the Pediculati and Gadoids (hake) the elasticity resides solely in the tissue of the hinge (the tooth being as resilient as ever after everything else is severed), in the pike the hinge is not in the least endowed with elasticity, but the bundles of fibres proceeding from the interior of the dentine cap are exceedingly elastic.

The teeth may be, and generally are, very different as regards size or form in the different parts of the mouth; they may be also different according to the age or sex of the fish (Raia). The teeth may be few in number and isolated, or placed in a single, double, or triple series, distant from one another or closely set; they may form narrow or broad bands, or patches of various forms. As regards form, they may be cylindrical or conical, pointed, straight, or curved, with or without an angular bend near their base; some are compressed laterally or from the front backwards (the latter may be triangular in shape, or truncated at the top like the incisors of mammals); they may have one apex (cusp) only, or be bi-or tri-lobate (bi-or tri-cuspid), or may have the margins denticulated or serrated. Compressed teeth may be confluent, and form a cutting edge in both jaws, which assume the shape of a parrot’s beak. In some the apex is hooked or provided with barbs. Again, some teeth are broad, with flat or convex surface, like molar teeth. With regard to size, the finest teeth are like fine flexible bristles, ciliiform or setiform; or, if very short and anchylosed to the bone, they appear only as inconspicuous asperities of the bone. Very fine conical teeth arranged in a band are termed villiform teeth; when they are coarser, or mixed with coarser teeth, they are cardlike (dents en râpe or en cardes); molar-like teeth of very small size are termed granular.

In all fishes the teeth are shed and renewed during the whole course of their life. In fishes which have compound teeth, as the Dipnoi, Chimaeroids, Scari, Gymnodonts, as well as in those which have apparently permanent teeth, as in the saw of Pristis, the detrition of the surface is made up by a constant growth of the tooth from its base. When the teeth are implanted in alveoli, they are generally succeeded by others in the vertical direction; but in other forms they succeed one another side by side. In the majority of fishes the new tooth is not developed (as in reptiles and mammals) in a diverticulum of the sac of its predecessor, but, like it, from the free surface of the buccal membrane. Generally there are more than one tooth growing which are in various stages of development, destined each to replace the others in function. This is very conspicuous in sharks, in which the whole phalanx of their numerous teeth is ever marching slowly forwards (or in some backwards), in rotatory progress, over the alveolar border of the jaw, the teeth being successively cast off after having reached the outer margin and fulfilled for a longer or shorter period their special function. Intestines.— The intestinal tract is divided into four portions,—the oesophagus, the stomach, the small and the large intestine; two or more of these divisions may coalesce in fishes and become indistinguishable. But it is characteristic of the class that the urinary apertures are constantly situated behind the termination of the intestinal tract.

In Branchiostoma the whole intestinal tract is straight, and coated with a ciliated mucous membrane. The liver is represented by a green-coloured coecal diverticulum of the stomachic dilatation. In the Cyclostomi the intestinal tract is likewise straight, and without clearly defined divisions.

The Palaeichthyes show differences in the structure of their intestinal tract as considerable as are found among the Teleostei, but they have this in common that the absorbent surface of their intestine is enlarged by the development of a spiral valve, evidence of the presence of which in extinct Palaeichthyes is still preserved in the fossilized faeces or coprolites, so abundant in some of the older strata. Ï In Chondropterygians (fig. 46) the stomach is divided into a cardiac and a pyloric portion, the former frequently terminating in a blind sac, and the latter varying in length. The pyloric portion is bent both at its origin and its end, and is separated from the short duodenum (called Bursa eηtiaηα in these fishes) by a valve; the ductus hepaticus and ductus pancreaticus enter the duodenum. This is succeeded by the straight intestine, provided with the spiral valve, the coils of which may either be longitudinal and wound vertically about the axis of the intestine, as in Carcharias, Galeocerdο, Thalassorhinus, and Zygcena, or they may be transverse to that axis, as in the other genera. The number of gyrations in the latter case varies; there may be as many as forty. The short rectum passes into a cloaca, which contains also the orifices of the urogenital ducts. Only the beginning and end of the intestinal tract are fixed by mesenterial folds.

The structure of the intestinal tract of Teleosteous fishes is subject to so numerous modifications that we should go beyond the limits of the present article were we to attempt to enter into details. Great differences in this respect may be found even in groups of the same natural families. Frequently the intestinal tract remains of nearly the same width throughout its course, and only the entrance of the various ducts serves as a guide for the distinction of its divisions. An intestine of such uniform width may be straight and short, as in Scombresocidae, Symbranchidoe, or it may be more or less convoluted and long, as in many Cyprinidoe, Doradina, &c. On the whole, carnivorous fishes have a much shorter and simpler intestinal tract than the herbivorous.

In the majority of Teleosteans, however, the oesophagus, stomach, duodenum, small intestine, and rectum can be more or less clearly distinguished, even externally.

There are two predominant forms of the stomach, intermediate forms, however, being numerous. In the first, the siphonal, it presents the form of a bent tube or canal, one-half of the horse-shoe being the cardiac, the other the pyloric portion. In the second, the caecal, the cardiac division is prolonged into a long descending blind sac, the cardiac and pyloric openings of the stomach lying close together {Clupea, Scomber, Thynnus, &c.).

The duodenum always receives the hepatic and pancreatic secretions, and also those of the appendices pyloricae, which, in varying numbers (from 1 to 200), are of very common occurrence in Teleosteans. They vary also in length and width, and whilst the narrowest serve only as secretory organs, the widest are frequently found filled with the same contents as the intestine.

Glands.— The liver of fishes is distinguished by the great quantity of fluid fat (oil) which it contains. The gall-bladder is but rarely absent; it is attached to the right lobe, or towards the centre; in some fishes, however, it is detached from the liver and connected with it by the cystic duct only. The bile may be conveyed by one or more hepatic ducts into a common duct which is continued towards the gall-bladder as ductus cysticus, and towards the duodenum as ductus choledochus; or some of the hepatic ducts enter the gall-bladder directly, or the duodenum directly, without communicating with the common duct. Individual variations in this respect are of common occurrence.

A pancreas has been found hitherto in all Chondropterygians, in Acipenser, and in many Teleosteans.

The spleen, which is substantially a lymphatic gland, may be mentioned here, as it is constantly situated in the immediate vicinity of the stomach, generally near its cardiac portion. With the exception of Branchiostoma, it is found in all fishes, and appears as a rounded or oblong organ of dark red colour.

Organs of Respiration.

Fishes breathe the air dissolved in water by means of gills or branchiae. The oxygen consumed by them is not that which forms the chemical constituent of the water, but that contained in the air which is dissolved in the water. Hence fishes transferred to water from which the air has been driven out by a high temperature, or in which the air absorbed by them is not replaced, are speedily suffocated. The absorption of oxygen by fishes is comparatively small; it has been calculated that a man consumes fifty thousand times more than is required by a tench. Some fishes, however, evidently require a much larger supply of oxygen than others: eels and carps, and other fishes of similar low vitality, can survive removal from their element for days, the small quantity of moisture retained in their gillcavity being sufficient to sustain life, whilst other fishes, especially such as have very wide gill-openings, are immediately suffocated after being taken out of the water. In some fishes noted for their muscular activity, like the Scombridoe, the respiratory process is so energetic as to raise the temperature of their blood far beyond that of the medium in which they live. A few fishes, especially such as are periodically compelled to live in water thickened into mud by desiccation and vitiated by decomposing substances, breathe atmospheric air, and generally have special contrivances for this purpose. These are so much habituated to breathing air that many of them, even when brought into pure water of normal condition, are obliged to rise to the surface at frequent intervals to take in a quantity of air, and, if they are kept beneath the surface by means of a gauze net, they perish from suffocation. The special contrivances consist of additional respiratory organs, lodged in cavities either adjoining the gill-cavity or communicating with the ventral side of the oesophagus, or of the air-bladder which enters upon respiratory functions {Dipnoi, Lepidοsteus, Amia).

The water used by fishes for respiration is received by the mouth, driven to the gills by an action similar to that of swallowing, and expelled by the gill-openings, of which there may be one or several on each side behind the head, or rarely one only in the median line of the ventral surface.

The gills or branchiae consist essentially of folds of the mucous membrane of the gill-cavity (laminae branchiales), in which the capillary vessels are distributed. In all fishes the gills are lodged in a cavity, but during the embryonic stage the Chondropterygians have the gill-laminae extended into long filaments projecting beyond the gill-cavity, and in a few young Ganoids external gills are superadded to the internal.

In Branchiostoma the dilated pharynx is perforated by numerous clefts, supported by cartilaginous rods (fig. 30, Ä). The water passes between these clefts into the peritoneal cavity, and makes its exit by the porus abdominalis, situated considerably in advance of the vent. The water is propelled by cilia.

In the Cyclostomes the gills of each side are lodged in a series of six or more antero-posteriorly compressed sacs, separated from each other by intervening septa. Each sac communicates by an inner duct with the oesophagus, the water being expelled by an outer duct. In Bdellostomα each outer duct has a separate opening, but in Myxine all the outer ducts pass outwards by one common gill-opening on each side. In the lampreys the ducts are short, the outer ones having separate openings (fig. 1). The inner ducts lead into a single diverticulum or bronchus, situated below the oesophagus, blind behind, and communicating in front with the pharynx, where it is provided with two valves by which the regurgitation of the water into the buccal cavity is prevented.

The same type of branchial organs persists in Chondropterygians, which possess five, rarely six or seven, flattened pouches with transversely plaited walls. The septa between them are supported by cartilaginous filaments rising from the hyoidean and branchial arches. Each pouch opens by a cleft outwards, and by an aperture into the pharynx, without intervening ducts. The anterior wall of the first pouch is supported by the hyoidean arch. Between the posterior wall of the first and the anterior of the second sac, and between the adjacent walls of the succeeding, a branchial arch with its two series of radiating cartilaginous filaments is interposed. Consequently the first and last pouches have one set of gill-laminae only, viz., the first on its posterior and the last on its anterior wall. The so-called spiracles on the upper surface of the head of Chondropterygians must be referred to in connexion with the respiratory organs. They are the external openings of a canal leading on each side into the pharynx, and situated generally close to and behind the orbit. They frequently possess valves or an irregularly indented margin, and are found in all species during the embryonic stage, but it is only in some that they remain persistent. The spiracles arc tiιe remains of the first visceral cleft of the embryo, and in the foetal state long branchial filaments have been observed to protrude as from the other branchial clefts.

The Holocephali and Ganοidei show numerous deviations from the Chondropterygian type, all leading towards the Teleosteans. As a whole they take an intermediate position between the preceding types and the Teleosteans, but they show a great variation among themselves, and have in common only the imperfect separation of the branchial sacs and the presence of a single outer branchial aperture. ·

In the Teleostei the gills with their supporting branchial arches lie in one undivided cavity; more or less wide clefts between the arches lead from the pharynx to the gills, and a more or less wide opening gives exit to the water after it has washed the gills. The interbranchial clefts have sometimes nearly the same extent as the branchial arches; sometimes they are reduced to small openings, the integuments stretching from one arch to the other. Sometimes there is no cleft behind the fourth arch, in which case this arch has only a uniserial gill developed. The gill-opening likewise varies much in its extent, and when reduced to a foramen may be situated at any part of the posterior boundary of the head. In the Symbranchidoe the gillopenings coalesce into a single narrow slit in the median line of the isthmus. In the majority of Teleosteans the integument of the concave side of the branchial arches develops a series of horny protuberances of various form, the so-called gill-rakers. These serve to catch any solid corpuscles or substances which would be carried into the gill-cavity with the water. In some fishes they are setiform, and make a complete sieve, whilst in others they are merely rough tubercles, the action of which must be very incomplete if they have any function at all.

The majority of Teleosteans possess four complete gills.

The gills of the Teleosteans, as well as of the Ganoids, are supported by a series of solid cartilaginous or horny pointed rods, arranged along the convex edges of the branchial arches. Arches bearing a complete gill have two series of those rods, one along each edge; those with uniserial gills bear one row of rods only. The rods are not part of the arch, but fixed in its integument, the several rods of the one row corresponding to those of the other, forming pairs (feuillet, Cuvier) (fig. 47). Each rod is covered by a loose mucous membrane passing from one rod to its fellow opposite, which again is finely plaited transversely, the extent of surface being greatly increased by these plaits. In most Teleostei the branchial lamellae are compressed, and taper towards their free end, but in the Lophobranchs their base is attenuated and the end enlarged. The mucous membrane contains the finest terminations of the vessels, which, being very superficial, impart a blood-red colour to living gills. The arteria branchialis, the course of which lies in the open canal in the convexity of the branchial arch, emits a branch (a) for every pair of lamellae, which ascends (δ) along the inner edge of the lamella, and supplies every one of the transverse plaits with a branchlet. The latter break up into a fine network of capillaries, from which the oxygenized blood is collected into venous branchlets, returning by the venous branch (<Z), which occupies the outer edge of the lamella.

The so-called pseudobranchiae (fig. 48) are the remains of an anterior gill which had respiratory functions during the embryonic life of the individuals. By ⅞ change in the circulatory system these organs have lost those functions, and appear in the adult fish as retia mirabilia, as they receive oxygenized blood, which, after having passed through their capillary system, is carried to other parts of the head. In Paloeichthyes the pseudobranchia is a rete mirabile caroticum for the brain and eye; in Teleosteans a rete mirabile ophthalmicum only. Pseudobranchiae are as frequently absent as present in' Chondropterygians as well as Teleosteans. Among the Ganoids, the organs occur in Ceratod/us, Acipenser, Polyodon, aenàLepidosteus, and are absent in Lepidosiren, Protopterus, Scaphirhynchus, Poly-ρterus, and Amia.

In Chondropterygians and sturgeons the pseudobranchiae are situated within the spiracles; in those in which spiracles have become obliterated, the pseudobranchiae lie on the suspensorium, hidden below cellular tissue; but pseudobranchiae are not necessarily coexistent with spiracles. In the other Ganoids and Teleosteans the pseudobranchiae (fig. 48, h) are within the gill-cavity, near the base of the gill-cover; in Ceratodus even rudiments of the gillrakers (a∕, xh) belonging to this embryonic gill are preserved, part of them (x") being attached to the hyoid arch. Pseudobranchiae are frequently hidden below the integuments of the gill-cavity, and have the appearance of a glandular body rather than of a gill.

Accessory respiratory organs for retaining water or breathing air are found in the Labyrinthici, Ophioce- phalidoe, certain Siluridoe, and L·itodira. Air-bladder.— The air-

bladder, one of the most characteristic organs of fishes, is a hollow sac, formed of several tunics, containing gas, situated in the abdominal cavity, but outside the peritoneal sac, entirely closed or communicating by a duct with the intestinal tract. Being compressible, its special functions consist in altering the specific gravity of the fish or in changing the centre of gravity. In a few fishes it assumes the function of the organ of higher vertebrates of which it is the homologue —viz., of a lung. The gas contained in the air-bladder is secreted from its inner surface. In most freshwater fishes it consists of nitrogen, with a very small quantity of oxygen and a trace of carbonic acid; in sea-fishes, especially those living at some depth, oxygen predominates, as much as 87 per cent, having been found. Davy found in the air-bladder of a fresh-run salmon a trace of carbonic acid and 10 per cent, of oxygen, the remainder of the gas being nitrogen.

An air-bladder is absent in Leptocardii i Cyclostomi i Chondropterygii, and Holocephali, but occurs in all Ganoids, in which, besides, its respiratory functions more or less clearly manifest themselves. Its occurrence in Teleosteans is very irregular, closely allied species sometimes differing from each other in this respect; it shows in this subclass the most extraordinary modifications, but has no respiratory function whatever.

Constantly situated within the abdominal cavity, below the vertebral column, but outside the sac of the peritoneum which covers only its ventral portion, the air-bladder is frequently prolonged into the tail, the prolongation being either single and lodged between the non-united parapophyses, or double and penetrating between the muscles and haemapophyses of each side. In the opposite direction processes of the air-bladder may penetrate into the skull, as has been mentioned above (p. 653). In some fishes the air-bladder is almost loose in the abdominal cavity, whilst in others it adheres most intimately by firm and short tissue to the vertebral column, the walls of the abdomen, and the intestines. In the Cobitina and many Siluroids it is more or less completely enclosed in osseous capsules formed by the vertebrae.

There are two tunics in the greater number of air-bladders,—an extremely thin internal one, frequently shining with a silvery lustre, containing crystalline corpuscles, sometimes covered with a pavement-epithelium, and a thicker outer one of a fibrous texture, which sometimes attains to considerable thickness and yields isinglass. The outer wall is strengthened in many fishes by muscular layers for the compression of the whole organ or of some portion of it.

A distinction has been made between air-bladders which communicate by a duct with the intestinal tract and those which are entirely closed. It is to be remembered, however, that at an early stage of development all air-bladders are provided with such a duct, which in some fishes is more or less completely obliterated, being then represented by a fine ligament only. Air-bladders without duct are found in Acanthopterygians, Pharyngognaths, Anacanths, and Lophobranchs. They may consist of a single cavity, or may be divided by constrictions into two or three chambers situated behind one another; they may consist of two lateral divisions, assuming a horseshoe-like form, or of a single sac with a pair of simple or bifid processes in front or behind. The families of Scioenidoe and Polynemidoe possess air-bladders with a most extraordinary development of appendages rising from each side. Air-bladders with a pneumatic duct are found in Ganoids and Physostomes, the duct entering the dorsal side of the intestinal tract, with the exception of Polypterus and the Dipnoi, in which it enters on the ventral side of the oesophagus. In most cases the orifice is in the oesophagus, but in some it is in the cardiac portion of the stomach, as in Acipenser, or in its blind sac, as in many Clupeoids. The air-bladder may be single, or may consist of two divisions situated one behind the other (fig. 45); its inner surface may be perfectly smooth, or it may form manifold pouches and cells. If two divisions are present the anterior possesses a middle elastic membrane which is absent in the posterior; each division has a muscular layer, by which it can be separately compressed, so that part of the contents of the posterior may be driven into the elastic anterior division, and vice versa. The posterior division being provided with the ductus pneumaticus does not require the elasticity of the anterior.

Some Siluroids possess a peculiar apparatus for voluntarily exercising a pressure upon the air-bladder. From the first vertebra a process takes its origin on each side, expanding at its end into a large round plate; this is applied to the side of the air-bladder, and by pressing upon it expels the air through the duct; the small muscle moving the plate rises from the skull. The connexion of the air-bladder with the organ of hearing in some Physostomes has been described above, p.

653.

In the modifications of the air-bladder hitherto mentioned, the chief and most general function is a mechanical one: this organ serves to regulate the specific gravity of the fish, to aid it in maintaining a particular level in the water, in rising or sinking, in raising the front part of its body or depressing it, as occasion may require. Yet a secretion of gas from the blood into its cavity must take place; and if this be so, it is not at all impossible that an exchange of gases between the two kinds of blood is also effected by means of the extraordinary development of retia mirabilia in many air-bladders.

In all fishes the arteries of the air-bladder take their origin from the aorta or the system of the aorta, and its veins return either to the portal, the vertebral, or the hepatic veins; like the other organs of the abdominal cavity, it receives arterial blood and returns venous blood.

Whilst the air-bladders of some Ganoids, anatomically as well as functionally, closely adhere to the Teleosteous type, that of Λmia is more cellular and lunglike in its interior than the Teleosteous air-bladder, and Polypterus approaches the Dipnoi, not only in having a laterally divided air-bladder, but also in its pneumatic duct entering the ventral side of the oesophagus. The air-bladder of the Dipnoi possesses still further the anatomical characteristics of a lung and assumes its functions, though, as it coexists with gills, only periodically or in an auxiliary manner. The ductus pneumaticus is a membranous bronchus, entering the ventral side of the oesophagus, and provided at its entrance with a glottis. In Ceratodus (fig. 49) the lung is still a single cavity, but with a symmetrical arrangement of its internal pouches; it has no pulmonary artery, but receives branches from the arteria coeliaca. Finally, in Lepidosiren and Protopterus the lung is completely divided into lateral halves, and by its cellular structure approaches most nearly that of a reptile; it is supplied with venous blood by a true pulmonary artery.

Organs of Circulation.

The blood-corpuscles of fishes, with one exception, are of an elliptic shape; this exception is Petromyzoη, which possesses circular, flat, or slightly biconvex blood-corpuscles. They vary much in size; they are smallest in Teleosteans and Cyclostomes, those of Acerina cernua measuring of an inch in their longitudinal, and in their transverse diameter. So far as it is known at present the Salmonidae have the largest blood corpuscles among Teleosteans, those of the salmon measuring r ⅛ τ by - 2 -⅛ σ in., approaching those of the sturgeon. Those of the Chondropterygians are still larger; and finally, Lepidosiren has blood-corpuscles not much, smaller than those of Perennibranchiates, viz., ^5τu by - s -∣ τ in. Dranchiostoma is the only fish which does not possess red blood-corpuscles.

Fishes, in common with the other vertebrates, are provided with a complete circulation for the body, with another equally complete for the organs of respiration, and with a particular abdominal circulation, terminating at the liver by means of the vena portae; but the peculiar characteristic consists in this, that the branchial circulation alone is provided at its base with a muscular apparatus or heart, corresponding to the right half of the heart of mammalia and birds.

The heart is situated between the branchial and abdominal cavities, between the two halves of the scapulary arch,—rarely farther behind, as in Symbranchidoe. It is enclosed in a pericardium, generally separated entirely from the abdominal cavity by a diaphragma, which is, in fact, the anterior portion of the peritoneum, strengthened by aponeurotic fibres. In some fishes, however, there is a communication between the pericardial and peritoneal sacs, viz., in the Chondropterygians and Acipenser, whilst in the Myxinoids the pericardial sac is merely a continuation of the peritoneum.

Relatively to the size of the body, the heart is very small. It consists of three divisions :—the atrium, with a large sinus venosus into which the veins enter; the ventricle; and a conical hollow swelling at the beginning of the arterial system, the structure of which forms one of the most important characters used in the classification of fishes. In all Paloeichthyes (figs. 50 and 51) this swelling is still a division of the pulsating heart, being provided with a thick muscular stratum; it is not separated from the ventricle by two valves opposite to each other, but its interior is fitted with a plurality of valves, arranged in transverse series more or less numerous in the various groups of the subclass. Lepidosiren and Protopterus offer an example of a modification of this valvular arrangement, their valves being longitudinal, each valve in fact being formed by the confluence of several smaller ones situated behind one another. This Palaeichthyan type is called conus arteriosus.

In Cyclostomes and Teleosteans (fig. 52) the enlargement is a swelling of the artery, without muscular stratum and. without contractility; with the exception of the Myxinoids, its walls are thick and fibrous with many trabeculae and pouches, but it has no valves in its interior, and is separated from the ventricle by two valves opposite to each other. This Teleostean type is called bulbus aortae.

The sinus venosus sends the whole of the venous blood by a single orifice of its anterior convexity into the atrium; two thin membranous valvules turned towards the atrium prevent the blood from re-entering the sinus. A pair of other valves between the atrium and ventricle have the same function. The walls of the ventricle are strong, and, internally, it is furnished with powerful fleshy trabeculae.

The bulbus or conus arteriosus is prolonged into the branchial artery, which soon divides, sending off a branch to each branchial arch. On returning from the respiratory organ the branchial veins assume the structure and functions of arteries. Several branches are sent off to different portions of the head and to the heart, but the main trunks unite to form the great artery which carries the blood to the viscera and to 1 all the parts of the trunk and tail, and which, therefore, represents the aorta of higher animals.

The circulatory system of Branchiostoma and of the Dipnoi shows essential differences from that of other fishes. Branchiostoma is the only fish which does not possess a muscular heart, several cardinal portions of its vascular system being contractile. A great vein extends forwards along the caudal region below the notochord, and exhibits contractility in a forward direction; it is bent anteriorly, passing into another tube-like pulsatory trunk, the branchial heart, which runs along the middle of the base of the pharynx, sending off branches on each side to the branchiae; each of these branches has a small contractile dilatation (bulbillus) at its base. The two anterior branches pass directly into the aorta; the others are branchial arteries, the blood of which returns by branchial veins emptying into the aorta. The blood of the intestinal veins is collected in a contractile tube, the portal vein, below the intestine, and distributed over the rudimentary liver. Of all other fishes, the portal vein is contractile in Myxinoids only.

In Dipnoi a rudimentary partition of the heart into a right and a left division has been observed; this is limited to the ventricle in Ceratodus i but in Lepidosiren and Pro- topterus an incomplete septum has been observed in the atrium also. All Dipnoi have a pulmonal vein, which enters the atrium by a separate opening, provided with a valve. The pulmonal artery rises in Lepidosiren and Pro- topterus from an arch of the aorta, but in Ceratodus it is merely a subordinate branch, rising from the arteria coeliaca.

Urinary Organs.

In Branchiostoma no urinary organs have been found.

In Myxinoids these organs are of very primitive structure; they consist of a pair of ducts, extending from the urogenital porus through the abdominal cavity. Each duct sends off at regular intervals from its outer side a short wide branch (the uriniferous tube), which communicates by a narrow opening with a blind sac. At the bottom of this sac there is a small vaso-ganglion (Malpighian corpuscle), by which the urine is secreted.

In the lampreys the kidneys form a continuous glandlike body, with irregular detached small portions. The ureters coalesce before they terminate in the urogenital papilla.

In Chondropterygians the kidneys occupy the posterior half or tworthirds of the back, of the abdominal cavity, outside the sac of the peritoneum (as in all fishes), which forms a firm tendinous horizontal septum. The kidneys of the two sides are never confluent, and generally show a convoluted or lobulated surface. The ureters are short; each is dilated into a pouch, and communicating with its fellow terminates by a single urethra (which also receives the vasa deferentia) behind the end of the rectum in the large common cloaca.

In Ganoids the kidneys occupy a similar position as in Chondropterygians, but these fishes differ considerably with regard to the termination and the arrangement of the ends of the urogenital ducts.

The kidneys of Teleosteans are likewise situated outside the peritoneal cavity, immediately below some part of the vertebral column, and vary.exceedingly with regard to form and extent. Sometimes they reach from the skull to between the muscles of the tail, sometimes they are limited to the foremost part of the abdominal cavity (in advance of the diaphragm), but generally their extent corresponds to that of the abdominal portion of the vertebral column. The ureters terminate, either separate or united, in a urinary bladder, varying in shape, which opens by a short urethra behind the vent. The urinary opening may be separate from or confluent with that of the genital ducts, and is frequently placed on a more or less prominent papilla (papilla urogenitalis). If separate, the urinary opening is behind the genital; and if a papilla is developed, its extremity is perforated by the urethra, the genital opening being situated nearer the base. A few Teleosteans show an arrangement similar to that of Chondropterygians and Dipnoi, the urogenital openings being in the posterior wall of the rectum [Symbranchidoe, Pediculati, and some Plectogηathι). Organs of Reproduction.

All fishes are dioecious, or of distinct sex. Instances of so-called hermaphroditism are, with the exception of Serranus, abnormal individual peculiarities; such have been observed in the cod-fish, in some Pleuronectidoe, and in the herring. Either the generative organ of one side was found to be male and that of the other female, or the organ of one or both sides was observed to have been developed partly into an ovary, partly into a testicle. In the European species of Serranus a testicle-like body is attached to the lower part of the ovary; but many specimens of this genus are undoubtedly males, having normally developed testicles only.

The majority of fishes are oviparous (comparatively few viviparous), the embryos being developed either in the ovarium or in some dilated portion of the oviduct. In viviparous fishes actual copulation takes place, and the males of most of them are provided with copulatory or intromittent organs. In oviparous fishes the generative products are, during sexual excitement, discharged into the water, a very small quantity of semen being sufficient for effectual impregnation of a number of ova dispersed in a considerable quantity of water,—circumstances which render artificial impregnation more practicable than in any other class of animals.

In Branchiostoma the generative organs occupy the ventral side of the abdominal cavity, into which they discharge their contents. No ducts are developed in either sex.

In the Cyclostomes the generative organ is single, and fixed to or suspended from the median line of the back of the visceral cavity by a duplicature of the peritoneum (mesoarium), the testicle and ovary being distinguishable by their contents only. These escape by dehiscence of the cells or capsules and rupture of the peritoneal covering into the abdominal cavity, and are expelled, by reciprocal pressure of the intertwined sexes, through the porus genitalis.

The ova of the lampreys are small and globular, like those of Teleosteans. Those of Myxine have a very peculiar shape when mature; they are of an oval form, about 15 millimetres long and 8 millimetres broad, enveloped in a horny case, which at each end is provided with a bundle of short threads, each thread ending in a triple hook. Whilst in the mesoarial fold, the eggs are attached to one another by means of these hooks, and after being expelled they probably fix themselves by the same means to other objects. As in all fishes producing ova of large size, the number of ova matured in one season is but small.

In Teleosteans the generative organs are comparatively large. In some families the ovaries are without a closed covering and without oviducts, as in Salmonidce, Galaxiida?, Notopteridae, Muroenidoe, and others. The surface of such an open ovary—as, for instance, that of the salmon—is transversely plaited, the ova being developed in capsules in the stroma of the laminae; after rupture of the capsules the mature ova drop into the abdominal cavity, and are expelled by the porus genitalis. The ovaries of the other Teleosteans are closed sacs, continued into oviducts. Frequently such ovaries coalesce into a single body, or one in which the division is effected internally only by a more or less complete septum. In the viviparous Teleosteans the embryos are developed within the ovary, notably in the Embiotocidoe, many Blenniidoe, and Cyprinodontid<χ l Sebastes viviparus, &c.

The ova of Teleosteous fishes are extremely variable in size, quite independently of the size of the parent species. The ova of large and small individuals of the same species, of course, do not differ in size; but, on the whole, larger individuals produce a greater number of ova than smaller ones of the same species. The larger the size of the ova is in a species, the smaller is the number produced during one season. The ova of the eel are almost microscopic. The numbers of ova in the small-sized roe of the herring, lump-fish, halibut, and codfish have been estimated respectively at 25,000, 155,000, 3,500,000, and 9,344,000. Larger in size and fewer in number are those of Antem narius, Salmo, Aspredo, Lophobranchs, &c. Those of Gastrosteus are comparatively the largest; and the Siluroid genus Arius, the males of which take care of their progeny, produces ova from 5 to 10 millimetres in diameter. The ova of all Teleosteans are perfectly globular and soft-shelled. Teleosteans without oviduct deposit them separated from one another; whilst in many Teleosteans with an oviduct, the ova are enveloped in a glutinous substance, secreted by its glands, swelling in the water and forming lumps or chords, in which the ova are aggregated.

Instances of the female taking care of her progeny are extremely rare in fishes. At present only two examples are known, that of the Siluroid genus Aspredo, and the Soleno-. stoma. In theformer (fig. 55), during the time of propagation, the integuments of the lower side of the flat trunk of the female assume a soft and spongy texture. After having deposited the eggs, the female attaches them to and presses them into the spongy integument, by merely lying over them. She carries them on her belly, as the Surinam toad (Pipa) carries her ova on her back. When the eggs are hatched the excrescence on the skin disappears, and the abdomen becomes as smooth as before. In Solenostoma the inner side of the long and broad ventral fins coalesces with the integuments of the body, a large pouch being formed for the reception of the eggs. There is a peculiar provision for the retention of the eggs in the sac, and probably for the attachment of the embryo. The inner walls of the sac are lined with long filaments, arranged in series along the ventral rays, and more numerous and longer at the base ot the rays than in the middle of their length, behind which they disappear entirely.

The testicles of the Teleosteans are always paired, and occupy the same position as the ovaries. Their size varies extraordinarily at the different seasons of the year. Vasa deferentia are constant. In the males of viviparous Teleosteans the urogenital papilla is frequently enlarged, and clearly serves as an intromittent organ.

Many Teleostei take care of their progeny, but with the exception of Aspredo and Solenostoma, as mentioned above, it is the male on which this duty devolves. In some, as in Cottus, Gastrosteus, Cycloptemis, Antennarius, Ophiocephalus, Cαllichthys, the male constructs with more or less skill a nest, and jealously guards the ova deposited in it by the female. The male of some species of Arius carries the ova (fig. 54) about with him in his capacious pharynx. The species of Chrοmis inhabiting the sea of Galilee are said to take care of their ova in the same manner. And, finally, in the Lophobranchs, nature has aided this instinct by the development of a pouch or the abdomen or lower side of the tail. In the Syngnathidoe this pouch is formed by a fold of the skin developed from each side of the trunk and tail, the free margins of the fold being firmly united in the median line, whilst the eggs are being hatched in the inside of the pouch. In Hippocampus the pouch is completely closed, with a narrow anterior opening.

e The genital organs of Ganoids show diversity of structure similar to that found in Teleosteans, but on the whole they approach the Batrachian type. The ovaries are not closed, except in Lepidosiren ; all Ganoids possess oviducts. In the sturgeons the oviduct as well as the vas deferens is represented by a funnel-shaped prolongation of the peritoneum, which communicates with the wide ureter. The inner aperture of the funnel is on a level with the middle of the testicle or ovary, while the outer is within the ureter; and it is a noteworthy fact that only at certain periods of the life of the fish is this outer aperture found to be open,—at other times the peritoneal funnel appears as a closed blind sac within the ureter. The mode of passage of the semen into the funnel is not known.

In Polypterus and Amia proper oviducts, with abdominal apertures in about the middle of the abdominal cavity, are developed; they coalesce with the ureters close to the common urogenital aperture.

In Cerαtodus a long convoluted oviduct extends to the foremost limit of the abdominal cavity, where it opens by a slit at a considerable distance from the front end of the long ovary; this aperture is closed in sexually immature specimens. The oviducts unite close to their common opening in the cloaca. During their passage through the oviduct the ova receive a gelatinous covering secreted by its mucous membrane. This is probably also the case in Lepidosiren, which possesses a convoluted oviduct with secretory glands in the middle of its length.

The ova of Ganoids, so far as they are known at present, are small, and enveloped in a gelatinous substance. In the sturgeon as many as 7,635,200 have been counted. Those of Lepidosteus seem to be the largest, measuring 5 millimetres in diameter with their envelope, and 3 millimetres without it. They are deposited singly, like those of newts.

In Chondropterygians (and HolocephaH) the organs of reproduction assume a more compact form, and are more free owing to a lengthened attachment to the back of the abdominal cavity. The ovaries of the majority are paired (single in the Carchariidae and Scylliidoe, one remaining undeveloped). But the oviducts are always paired, with a common aperture beginning immediately behind the diaphragma. They consist of two divisions, separated by a circular valve; the upper is narrow, and is provided within its coats with a gland which secretes the leathery envelope in which most of the Chondropterygian ova are enclosed; the lower forms the uterine dilatation in which the embryos of the viviparous species are developed. Generally the vitelline sac of the embryos is free, and has no connexion with the uterus, which in these cases has merely the function of a protecting pouch; but in Carcharias and Must elms lαvis a placenta uterina is formed, the vascular walls of the vitelline sac forming plaits fitting into those of the membrane of the uterus. The ends of the uteri open into the cloaca by a common ap√rture behind the ureter.

The testicles are always paired, rounded, and situated in the anterior part of the abdominal cavity, covered by the liver. The vas deferens opens with the urethra in a papilla within the cloaca.

The so-called claspers of Chondropterygians (fig. 56) are characteristic of all male individuals. They are semiossified appendages of the pubic, with which they are movably joined, and special muscles serve to regulate their movements. Sometimes they are armed with hook-like osseous excrescences (Selache). They are irregularly convoluted longitudinally, and, when closely pressed to each other, form a canal open at their extremity. A gland, which discharges a secretion abundantly during the season of propagation, is situated at the base of the canal, and opens into it. It is still doubtful whether the generally adopted opinion that their function consists in holding the female during copulation is correct, or whether they are not rather an intromittent organ, the canal of which conducts, not only the secretion of their proper gland, but also the impregnating fluid.

The ova of the oviparous Chondropterygians are large, and few in number; they are successively impregnated, and the impregnation must take place before they are invested with a tough leathery envelope, which would be impenetrable to the semen, that is, before they •enter the uterus; therefore, copulation must take place in all these fishes. The form of the egg-shell differs in the various genera; usually (fig. 57) they are flattened and quadrangular, with each of the four corners produced, and frequently prolonged into, lengthy filaments which serve for the attachment of the ova to other fixed objects. In N^otidanus the surfaces are crossed by numerous ridges. In Cestracion the egg is pyriform, with two broad ridges or plates wound edgewise round it, the two ridges forming five spires. The eggs of Callοrhynchus have received a protective resemblance to a broad-leaved fucus, forming a long depressed ellipse, with a plicated and fringed margin.

Growth and Variation of Fishes.

Changes of form normally accompanying growth (after absorption of the vitelline sac) are observed in all fishes, but in the majority these affect only the proportional size of the various parts of the body. Relatively to the size of the head, the eyes in young fishes are always larger than in the adult; and again, the head is relatively larger than the body. Changes amounting to metamorphosis have been hitherto observed in Petromyzon only. In the larval condition (Ammocoetes) the head is very small, and the toothless buccal cavity is surrounded by a semicircular upper lip. The eyes are extremely small, hidden in a shallow groove; and the vertical fins form a continuous fringe. In the course of three or four years the teeth are developed, and the mouth changes into a perfect suctorial organ; the eyes grow; and the dorsal fin is separated into two divisions. In Malacopterygians and Anacanths the embryonal fringe from which the vertical fins are developed is much longer persistent than in Acanthopterygians. A metamorphosis relating to the respiratory organs, as in Eatra-chians, is indicated in the class of fishes by the external gills with which foetal Plagiostomes and the young of some Ganoids, viz., the Protopterus and Polypterus, are provided.

One of the most extraordinary changes by which, during growth, the form and position of several important organs are affected, occurs in flat-fishes {Pdeuronectidοd) ; their young are symmetrically formed, with a symmetrical mouth, and with one eye on each side, and therefore keep their body in a vertical position when swimming. As they grow they live more on the bottom, and their body, during rest, assumes a horizontal position; in consequence, the eye of the lower side moves towards the upper, which alone is coloured; and in many genera the mouth is twisted in the opposite direction, so that the bones, muscles, and teeth are much more developed on the blind side than on the coloured. In a great number of other Teleostei certain bones of the head show a very different form in the young state. Ossification proceeds in those bones in the direction of lines or radii which project in the form of spines or processes; as the interspaces between these processes are filled with bone, the processes disappear entirely, or at least project much less in the older than in the younger individuals. The young of some fishes may be armed with a long powerful praeopercular or scapular spine, or may show a serrature of which nothing remains in the adult fish except some ridges or radiating lines. These processes seem to serve as weapons of defence during a period in the life of the fish in which it needs them most. In not a few instances a portion of this armature is so much developed that the disappearance of its most projecting parts with the growth of the fish is not only due to its being surrounded by other bone, but partially, at least, caused by absorption. The Carangidoe, Cyttidoe, Squamipinηes, Xiphiidoe, offer instances of such remarkable changes. A fish described, as Tholichthys osseus (fig. 58) is probably the young of a Cyttoid, the suprascapula, humerus, and praeoperculum forming enormously enlarged plates. In another fish (fig. 59) these bones appear still enlarged, and the frontals develop a remarkably long and curved horn above the orbit. In the Tholichthys-stags of Pomacanthus (specimens 10 millimetres long), the frontal bone is prolonged into a straight lancet-shaped process, nearly half as long as the body; the suprascapular and praeopercular processes cover and hide the dorsal and ventral fins. The plates attached to the shoulder-girdle remain persistent until the young fish has assumed the form of the adult; thus they are still visible in young Choeto- dοη citrinellus, 30 millimetres long, in which the specific characters are already fully developed (fig. 60). The swordfishes with ventral fins (Histiophorus) belong to the Teleosteans of the largest size; in young individuals, 9 millimetres long (fig. 61), both jaws are produced, and armed with pointed teeth; the supraorbital margin is ciliated; the parietal and praeoperculum are prolonged into long spines; the dorsal and anal fins are a low fringe; and the rentrais make their appearance as a pair of short buds. When 14 millimetres long the young fish has still the same armature on the head, but the dorsal fin has become much higher, and the ventral filaments have grown to a great length. At a third stage, when the fish has attained to a length of 60 millimetres, the upper jaw is considerably prolonged beyond the lower, losing its teeth; the spines of the head are shortened, and the fins assume nearly the shape which they retain in mature individuals. Young sword-fishes without ventral fins (Xiphias) undergo similar changes; and, besides, their skin is covered with small rough excrescences longitudinally arranged, which continue to be visible after the young fish has in other respects assumed the form of the mature (fig- 63).

The Plectognaths show changes no less extraordinary: a remarkable form caught in the South Atlantic, and named Ostracion hoops, is considered by Lütken to be the young of a sun-fish (Orthagoriscus). In still very young but more advanced sun-fishes (18 to 32

millimetres) the vertical diameter of the body is not much less than the longitudinal, and may even exceed it; and small conical spines are scattered over its various parts.

The caudal fin is developed long after the other vertical fins.

Similar changes take place in a number of other fishes, and in many cases the young are so different that they have been described as belonging to distinct genera: thus PriMantichthys has proved to be the young of Serranus, Rhynchichihys that of Holocertrum, Cephalacαnthus of Dactylopterus, Dicrotus of Thyrsites, Nauclerus of Naucrates, Porthmeus of Chorinemusi Lampugus of Coτyphcena, Acronurus of Acanthurus, Keris of Naseus, Porobronchus of Fierasfer, Cοuchia of Hotella, Stomiasunculus of Stomias, &c.

The fins are most frequently subject to changes during growth; but, whilst in some fishes parts of them are prolonged into filaments with age, in others the filaments exist during the early life-periods only; whilst in some a part of the dorsal or of the ventral fins is normally developed in the young only, in others those very parts are peculiar to the mature age. The integuments are similarly altered: in some species the young only have asperities on the skin, in others the young are smooth and the old have a tubercular skin; in some the young only have a hard bony head, in others (some Siluroids) the osseous carapace of the head and neck, as it appears in the adult, is more or less covered with soft skin whilst the fish is young.

In not a few fishes the external changes bear a relation to the sexual development (Callionymus, many Labyrinthici Cyprinodonts). These secondary sexual differences do not show themselves in the male individual till it commences to enter upon its sexual functions, and it may require two or more seasons before its external characteristics are fully developed. Immature males do not differ externally from the old female. The secondary sexual characters of the male consist principally in the prolongation of some of the fin-rays, or of entire fins, and in Sal-monidοe in the greater development of the jaw-bones. The coloration of the male is in many fishes much brighter and more variegated than that of the female, but is permanent in comparatively few (as in some Callionymus, Labmιs mirtus); generally it is acquired immediately before and during the season of propagation only, and lost afterwards. Another periodical change in the integuments, also due to sexual influence and peculiar to the male, is the excrescence of wart-like tubercles on the skin of many Cyprinoids; they are developed chiefly on the head, but sometimes extend over the whole body and all the fins.

With regard to size, it appears that in all Teleosteous fishes the female is larger than the male; in many Cyprinodonts the male may be only one-sixth of the bulk of the female or even less. In Palaeichthyes we possess few observations on the relative size of the sexes, but such as have been made tend to show that, if a difference exists at all, the male is generally the larger (Lepidosteus). In the rays (Raid) the sexes, after they have attained maturity, differ in the development of dermal spines and the form of the teeth, the female being frequently much rougher than the male. There is much variation in this respect in the different species; but the males are constantly distinguished by an oblong patch of erectile clawlike spines on each pectoral fin, and by having the teeth (all, or only a portion) pointed, and not obtuse, like those of the females. In sharks no secondary sexual differences have been observed; the male Chirnoeridoe possess a singular comb-like cartilaginous appendage on the top of the head, which can be erected or depressed into a groove, both the appendage and the anterior part of the groove being armed with booklets. The use of this singular organ is not known.

The majority of Teleostei are mixogamous—that is, the males and females congregate on the spawning-beds, and, the number of the former being in excess, several males attend to the same female, frequently changing from one to another. The same habit has been observed in Lepidosteus. Gastrosteus is truly polygamous, several females depositing their ova in the same nest, guarded by one male only. Some Teleostei (f)phiocephαlus'), and probably all Chondropterygians, are monogamous; and it is asserted that the connexion between the pair is not merely temporary, but lasts until they are separated by accident. All those Teleosteans also are probably monogamous which bring forth living young.

Hybridism is another source of changes and variations within the limits of a species, and is by no means so rare as has been hitherto believed; it is apparently of exceptional occurrence, merely because the life of fishes is more withdrawn from our direct observation than that of terrestrial animals. It has been observed among species of Serranus, Pleuronectidoe, Cyprinidoe, Clupeidoe, and especially Salmonidoe. As with other animals, the more certain kinds of fishes are brought under domestication, the more readily do they interbreed with other allied species. It is characteristic of hybrids that their characters are very variable, the degrees of affinity to the one or the other of the parents being inconstant; and, as these hybrids are known readily to breed with either of the parent race, the variations of form, structure, and colour are infinite. Of internal organs the teeth, the gill-rakers, and the pyloric appendages are those particularly affected by such mixture of species.

Some fishes are known to grow rapidly (in the course of from one to three years) and regularly to a certain size, growth being definitely arrested after the standard has been attained. Such fishes may be called “full-grown,” in the sense in which the term is applied to warm-blooded vertebrates; the sticklebacks, most Cyprinodonts, and many Clupeoids (herring, sprat, pilchard) are examples of this regular kind of growth. But in the majority of fishes the rate of growth is extremely irregular, and it is hardly possible to know when growth is actually and definitely arrested. All seems to depend on the amount of food and the more or less favourable circumstances under which the individual grows up. Fishes which rapidly grow to a definite size are short-lived, whilst those, Teleosteans as well as Chondropterygians, which steadily and slowly increase in size attain to a great age. Carp and pike have been ascertained to live beyond a hundred years.

Abundance or scarcity of food, and other circumstances connected with the localities inhabited by fishes, affect considerably the colour of their muscles and integuments; the periodical changes of colour in connexion with their sexual functions have been referred to above. The flesh of many Teleostei is colourless, or but slightly tinged by the blood; that of Scombridoe, and most Ganoids and Chondropterygians, is more or less red; but in badly-fed fishes, as well as in very young ones, the flesh is invariably white (anaemic). Many fishes, like the Salmonidoe, feed at times exclusively on crustaceans, and the colouring substance of these invertebrates, which by boiling and by the stomachic secretion turns red, seems to pass into their flesh, imparting to it the well-known “salmon” colour. The colora Attempts to acclimatize particularly useful species in countries in which they are not indigenous have been made from time to time, but have been permanently successful in a few instances only, the failures being due partly to the choice of a species which did not yield the profitable return expected, partly to utter disregard of the difference of climatic and other physical conditions between the original and the new homes of the fish. The first successful attempts at acclimatization were made with domestic species, viz., the carp and goldfish, which were transferred from Eastern Asia to Europe. Then, in the early part of the present century, the Javanese goramy was acclimatized in Mauritius and Guiana, but no care seems to have been taken to insure permanent advantages from the successful execution of the experiment. In these cases fully developed individuals were transported to the country in which they were to be acclimatized. The most successful attempt of recent years is the acclimatization of the trout and sea-trout, and probably also of the salmon, in Tasmania and New Zealand, and of the Californian salmon (Salmo quinnat ?), in Victoria, by means of artificially-impregnated ova. In transporting these ice was employed, in order to retard their development generally, and thus to preserve them from destruction during the passage across the tropical zone.

Artificial impregnation of fish-ova was first practised by J. L. Jacobi, a native of Westphalia, in the years 1757-63, who employed exactly the same method which is followed now; and there is no doubt that this able observer of nature conceived and carried out his idea with the distinct object of advantageously restocking water-courses that had become unproductive, and increasing production by fecundating and preserving all ova which, in the ordinary course of propagation, would be left unfecundated or might accidentally perish. Physiology soon turned to account Jacobi’s discovery, and artificial impregnation has proved to be one of the greatest helps to the student of embryology.

Fishes differ in an extraordinary degree with regard to tenacity of life. Some will bear suspension of respiration —caused by removal from water, or by exposure to cold or heat—for a long time, whilst others succumb at once. Nearly all marine fishes are very sensitive to changes in the temperature of the water, and will not bear transportation from one climate to another. This seems to be much less the case with some freshwater fishes of the temperate zones; the carp may survive after being frozen in a solid block of ice, and will thrive in the warmest parts of the temperate zones. On the other hand, some freshwater fishes are s) sensitive to a change in the water that they perish when transplanted from their native river into another apparently offering the same physical conditions (grayling, Salmo hncho). Some marine fishes may be transferred at once from salt into fresh water, like sticklebacks, some blennies, and Cottιιs, &c.; others survive the change when gradually effected, as many migratory fishes; whilst others, again, cannot bear the least alteration in the composition of the salt water (all pelagic fishes). On the whole, instances of marine fishes voluntarily entering brackish or fresh water are very numerous, whilst freshwater fishes proper but rarely descend into salt water.

Abstinence from food affects different fishes in a similarly different degree. Marine fishes are less able to endure hunger than freshwater fishes,—at least in the temperate zones, no observations having been made in this respect on tropical fishes. Goldfishes, carps, and eels are known to be able to subsist without food for months, without showing a visible decrease in bulk; whilst the Trigloids, Sparoids, and other marine fishes survive abstinence from food for a few days only. In freshwater fishes the temperature of the water has great influence on their vital functions generally, and consequently on their appetite. Many cease to feed altogether in the course of the winter; a few, like the pike, are less inclined to feed during the heat of the summer than when the temperature is lowered.

Captivity is easily borne by most fishes, and the appliances introduced in modern aquaria have rendered it possible to keep in confinement fishes which formerly were considered to be intolerant of captivity, and even to induce them to propagate.

Wounds affect fishes generally much less than higher vertebrates. A Greenland shark continues to feed though its head is pierced by a harpoon or by the knife, so long as the nervous centre is not touched; a pike will survive the loss of its tail, or a sea-perch that of a portion of it, and a carp that of half its snout. Some fishes, however, are much more sensitive, and perish even from the superficial abrasion caused by the meshes of the net during capture (Hullus).

The power of reproduction of lost parts in Teleosteous fishes is limited to the delicate terminations of their fin-rays and the various tegumentary filaments with which some are provided. These filaments are sometimes developed in an extraordinary degree, imitating the waving fronds of the seaweed in which the fish hides. The ends of the fin-rays and also the filaments are frequently lost, not only by accident, but also merely by wear and tear; and, as these organs are essential for the preservation of the fish, their reproduction is necessary. In Dipnoi, Ceratodus, and Protopterus, the terminal portion of the tail has been found to have been reproduced, but without the notochord.

Hibernation has been observed in many Cyprinoids and Muraenoids of the temperate zones. They do not fall into a condition of complete torpidity, as reptiles and mammals do, but their vital functions are simply lowered, and they hide in sheltered holes, and cease to go abroad in search of their food. Between the tropics a great number of fishes (especially Siluroids, Labyrinthici, Ophiocephaloids, the Dipnoi) are known to survive long-continued droughts by passing the dry season in a perfectly torpid state, imbedded in the hardened mud. Protopterus, and probably many of the other fishes mentioned, prepare for themselves a cavity large enough to hold them, and coated on the inside with a layer of hardened mucus, which preserves them from complete desiccation.. It has been stated that in India fishes may survive in this condition for more than one season, and that ponds, known to have been dry for several years and to a depth of many feet, have swarmed with fishes as soon as the accumulation of water released them from their hardened bed.

The principal benefit derived by man from the class of fishes consists in the abundance of wholesome and nourishing food which they yield. In the polar regions especially, whole tribes are entirely dependent on this class for subsistence; and in almost all nations fishes form a more or less essential part of food, many, in· a preserved condition, being most important articles of trade. Their use in other respects is of but secondary importance. Codliver oil is prepared from the liver of some of the Gadoids of the northern hemisphere and of sharks, isinglass from the swim-bladder of sturgeons, Sciaenoids, and Polynemoids, and shagreen from the skin of sharks and rays.

The flesh of some fishes is constantly or occasionally poisonous. When eaten, it causes symptoms of more or less intense irritation in the stomach and intestines, inflammation of the mucous membranes, and not rarely death. The fishes which appear always to have poisonous properties are— Clupea thrissa, Clupea venenosa, and some species of Scarus, Tetrodon, and Diodon. There are many others which have occasionally or frequently caused symptoms of poisoning. Poey enumerates no less than seventy-two different kinds from Cuba; and various species of Sphyroena, BaUstes, Ostracion, Caranx, Lachnolcemus, Tetragoηurus, Thynnus, have been found to be poisonous in all seas between the tropics. All or nearly all these fishes acquire their poisonous properties from their food, which consists of poisonous Hednsoe and corals, or of decomposing substances. Frequently the fishes are found to be eatable if the head and intestines are removed immediately after capture. In the West Indies it has been ascertained that all the fishes living and feeding on certain coral banks are poisonous. In other fishes the poisonous properties are developed at certain seasons of the year only, especially the season of propagation; as the barbel, pike, and burbot, whose roe causes violent diarrhoea when eaten during the season of spawning.

Poison-organs are more common in the class of fishes than was formerly believed, but they seem to have exclusively the function of defence, and are not auxiliary in procuring food, as in venomous snakes. Such organs are found in the sting-rays, the tail of which is armed with one or more powerful barbed spines. Although they have no special organ secreting poison, or canal in or on the spine by which the venomous fluid is conducted, the symptoms caused by a wound from the spine of a sting-ray are such as cannot be accounted for merely by the mechanical laceration, the pain being intense, and the subsequent inflammation and swelling of the wounded part terminating not rarely in gangrene. The mucus secreted from the surface of the fish and inoculated by the jagged spine evidently possesses venomous properties. This is also the case in many Scorpaenoids, and in the weever ^Trachinus), in which the dorsal and opercular spines have the same function as the caudal spines of the sting-rays; in the weevers, however, the spines are deeply grooved, the groove being charged with a fluid mucus. In Synanceia the poison-organ (fig. 67) is still more developed: each dorsal spine is in its terminal half provided with a deep 1 groove on each side, at the lower end of which lies a pear - shaped bag containing the milky poison; it is prolonged into a membranous duct, lying in the groove of the spine, and open at its point. The native fishermen, well acquainted with the dangerous nature of these fishes carefully avoid handling them; but it often happens that persons wading with naked feet in the sea step upon the fish, which generally lies hidden in the sand. One or more of the erected spines penetrate the skin, and the poison is injected into the wound by the pressure of the foot on the poison-bags. Death has frequently been the result.

The most perfect poison-organs hitherto discovered in fishes are those of Thalassophryηe, a Batrachoid genus of fishes from the coasts of Central America. In these fishes again the operculum and the two dorsal spines are the weapons. The former (fig. 69) is very narrow,, vertically styliform, and very mobile; it is armed behind with a spine, eight lines long, and of the same form as the hollow venom-fang of a snake, being perforated at its base and at its extremity. A sac covering the base of the spine discharges its contents through the apertures and the canal in the interior of the spine. The structure of the dorsal spines is similar. There are no secretory glands imbedded in the membranes of the sacs, and the fluid must be secreted by their mucous membrane. The sacs have no external muscular layer, and are situated immediately below the thick loose skin which envelops the spines to their extremity; the ejection of the poison into a living animal, therefore, can only be effected, as in Synanceia, by the pressure to which the sac is subjected the moment the spine enters another body

GEOLOGICAL DISTRIBUTION.

Of what kind the fishes were which were the first to make their appearance on the globe, whether or not they were identical with or similar to any of the principal types existing at present, are questions which probably will for ever remain hidden in mystery and uncertainty. The supposition that the Pharyngobranchs and Cyclostomes, the lowest of the vertebrate series, must have preceded the other subclasses, is an idea which has been held by many zoologists, and, as the horny teeth of the Cyclostomes are the only parts which under favourable circumstances could have been preserved, palaeontologists have ever been searching for this evidence.

Indeed, in deposits belonging to the Lower Silurian and Devonian, in Russia, England, and North America, minute, slender, pointed horny bodies, bent like a hook, with sharp opposite margins, have been found and described under the name of “conodonts.” More frequently they possess an elongated basal portion, in which there is generally a larger tooth with rows of similar but smaller denticles on one or both sides of the larger tooth, according as this is central or at one end of the base. In other examples there is no prominent central tooth, but a series of more or less similar teeth is implanted on a straight or curved base. Modifications of these arrangements are very numerous, and many paleontologists still entertain doubts whether the origin of these remains is not rather from annelids and mollusks than from fishes.

The first undeniable evidence of a fish, or, indeed, of a vertebrate animal, occurs in the Upper Silurian rocks, in a bone-bed of the Downton Sandstone, near Ludlow. It consists of compressed, slightly curved, ribbed spines, of less than 2 inches in length {0nchus) ; small shagreenscales {Thelodus); the fragment of a jaw-like bar with pluricuspid teeth (Plectrοdιis); the cephalic bucklers of what seems to be a species of Pteraspis; and, finally, coprolitic bodies of phosphate and carbonate of lime, including recognizable remains of the mollusks and crinoids inhabiting the same waters. But no vertebra or other part of the skeleton has been found. The spines and scales seem to have belonged to the same kind of fish, which probably was a Plagiostome. It is quite uncertain whether or not the jaw (if it be the jaw of a fish) belonged to the bucklerbearing Pteraspis, the position of which among Ganoids, with which it is generally associated, is open to doubt.

No detached undoubted tooth of a Plagiostome or Ganoid scale has been discovered in the Ludlow deposits; but this much is certain, that those earliest remains in Palaeozoic rocks belonged to fishes closely allied to forms occurring in greater abundance in the succeeding formation, the Devonian, where they are associated with undoubted Paloeichthyes, Plagiostomes as well as Ganoids.

These fish-remains of the Devonian or Old Red Sandstone can be determined with greater certainty. They consist of spines, the so-called “ichthyodorulites,” which show sufficiently distinctive characters to be referred to several genera, one of them, Onchus, still surviving from the Silurian epoch. All these spines are believed to be those of Choπdropterygians, to which order some pluricuspid teeth (Cladοdus) from the Old Red Sandstone in the vicinity of St Petersburg have been likewise referred.

The remains of the Ganoid fishes are in a much more perfect state of preservation, so that it is even possible to obtain a tolerably certain idea of the general appearance and habits of some of them, especially of such as were provided with hard carapaces, solid scales, and ordinary or bony fin-rays. A certain proportion of them, as might have been expected, remind us, as regards their external form, of Teleosteous fishes rather than of any of the few still existing Ganoid types; but it is contrary to all analogy and to all palaeontological evidence to suppose that those fishes were, in their internal structure, more nearly allied to Teleosteans than to Ganoids. If they were not true Ganoids, they may justly be supposed to have had the essential characters of Paloeichthyes. Other forms even at that remote geological epoch exhibit so unmistakably the characteristics of existing Ganoids that no one can entertain any doubt with regard to their place in the system. In none of these fishes is there any trace of vertebral segmentation.

The Paloeichthyes of the Old Red Sandstone, the systematic position of which is still obscure, are the Cephalaspidoe from the Lower Old Red Sandstone of Great Britain and Eastern Canada; Ptemchthys, Cοccοsteus, Dinichthys, and Asterolepis (genera which have been combined in one group, Placodermoe) ; and Acanthodes and allied genera, which combined numerous branchiostegals with Chondropterygian spines and a shagreen-like dermal covering.

Among the other Devonian fishes (and they form the majority) two types may be recognized, both of which are unmistakably Ganoids. The first approaches the still living Polypterιιs, with which some of the genera like Diplοpterus singularly agree in the form and armature of the head, the lepιdosιs of the body, the lobate pectoral fins, and the termination of the vertebral column. Other genera, as Holoptychius, have cycloid scales; many have two dorsal fins {Hοlοptychius), and, instead of branchiostegals, jugular scutes; others have one long dorsal confluent with the caudal {Phαηerοpleurοri). In the second type the principal characters of the Dipnoi are manifest; and some of them—for example, Dipterus, Paloedaphus, Holodus— approach so closely the Dipnoi which still survive that the differences existing between them warrant a separation into families only.

Devonian fishes are frequently found under peculiar circumstances, enclosed in the so-called nodules. These bodies are elliptical flattened pebbles, which have resisted the action of water in consequence of their greater hardness, whilst the surrounding rock has been reduced to detritus by that agency. Their greater density is due to the dispersion in their substance of the fat of the animal which decomposed in them. Frequently, on cleaving one of these nodules with the stroke of the hammer, a fish is found embedded in the centre. At certain localities of the Devonian, fossil fishes are so abundant that the whole of the stratum is affected by the decomposing remains, emitting a peculiar smell when newly opened, and acquiring a density and durability not possessed by strata without fishes. The flagstones of Caithness are a remarkable instance of this.

The fish-remains of the Carboniferous formation show a great similarity to those of the preceding. They occur throughout the series, but are very irregularly distributed, being extremely rare in some countries, whilst in others entire beds (the so-called bone-beds) are composed of ichthyolites. In the ironstones they frequently form the nuclei of nodules, as in the Devonian.

Of Chondropterygians the spines of Onchus and others still occur, with the addition of teeth indicative of the existence of fishes allied to the Cestracion type (Cochliodus, Psamrnοdus), a type which henceforth plays an important part in the composition of the extinct marine fish faunae. Another extinct Selachian family, that of the Hybodonts, makes its appearance, but is known from the teeth only.

Of the Ganoid fishes, the family Paloeoniscidoe (Traquair) is numerously represented; others are Coelacanths {Coel- acaηthus, Uhizodus), and Saurodipterini (Megαlichthys). None of thèse fishes have an ossified vertebral column, but in some (Megalichthys) the outer surface of the vertebrae is ossified into a ring; the termination of their tail is heterocercal. The Carboniferous Urοηemus and the Devonian Phanerοpleuron are probably generically the same; and the Devonian Dipnoi are continued as, and well represented by, Gtenodus.

The fishes of the Permian group are very similar to those of the Carboniferous. A type which in the latter was but very scantily represented, namely the Platysomidoe, is much developed. They were deep-bodied fish, covered with hard rhomboid scales possessing a strong anterior rib, and provided with a heterocercal caudal, long dorsal and anal, short non-lobate paired fins (when present), and branchiostegals. Paloeoniscus appears in many species; the Sauridoe are represented by Pygopterus and Acrolepis, and Cestracionts by Jana8sa and Strophodus.

The passage from the Palaeozoic into the Mesozoic era is not indicated by any marked change so far as fishes are concerned. The more remarkable forms of the Trias are shark-like fishes represented by ichthyodorulites, like Nemacaηthus, Liacanthus, and Hybodns ; and Cestracionts represented by species of Acτodus and Strophodus. Of the Ganoids, Coelacanthus, Amblypterηs {Paloeoniscidoe), and Saurichthy8 persist from the Carboniferous epoch. Cera-todu8 appears for the first time (Muschel Kalk of Germany).

Thanks to the researches of Agassiz, and especially of Sir P. Egerton, the ichthyological fauna of the Lias is perhaps the best knownof the Mesozoic era, one hundred andfifty-two species having been described. Of the variou s locali ties, Lyme Regis has yielded more than any other, nearly all the Liassic genera being represented there by no fewer than seventy-nine species. The Hybodonts and Cestracionts continue in their fullest development. Holocephales (Ischyodus), true sharks (Paloeoscyllium), rays (Squaloraia, Arthropterus), and sturgeons (Chondr osteus) make their first appearance; but they are sufficiently distinct from living types to be classed in separate genera, or even families. The Ganoids, especially Lepidosteoids, predominate over all the other fishes: Lepidotus, Semionοtus, Pholidophorus, Pachycoτmus, Eugnathus, Tetragonolepis, are represented by numerous species; other remarkable genera are Aspidorhynchus, Beloηostomus, Saurostomus, Sauropsis, Thrissonotus, Coηodus, Ptycholepis, Endactis, Centrolβpis, Legnonotus, Oxygnathus, Heterolepi-dotus, Isοcolum, Osteorhachis, Mesodon. These genera afford evidence of a great change since the preceding period, the majority not being represented in older strata, whilst, on the other hand, many are continued into the succeeding Oolitic formations. The homocercal termination of the vertebral column begins to supersede the heterocercal, and many of the genera have well-ossified and distinctly segmented spinal columns. The cycloid form of scales also becomes more common,—one genus (Leptolepis) being, with regard to the preserved hard portions of its organization, so similar to the Teleosteous type that some palaeontologists refer it (with much reason) to that subclass.

As already mentioned, the Oolitic formations show a great similarity of their fish fauna to that of the Lias; but still more apparent is its approach to the existing fauna. Teeth have been found which cannot even generically be distinguished from Notidanus. The rays are represented by genera like Spathobatis, Belemnobatis, Thaumas ; the Holocephali are more numerous than in the Lias (Ischyodus, Ganodus). The most common Ganoid genera are Caturus, Pycηodus, Pholidophorus, Lepidotus, Leptolepis, all of which had been more or less fully represented in the Lias. Ceratodus also is continued into it.

The Cretaceous group gives clear evidence of the further advance towards the existing fauna. Teeth of sharks of existing genera, Carcharias (Corax), Scyllium, Notidanus, and Galeocerdo, are common in some of the marine strata, whilst Hybodonts and Cestracionts are represented by a small number of species only; of the latter one new genus, Ptychodus, appears and disappears. A very characteristic Ganoid genus, Macropoma, comprises homocercal fishes with rounded ganoid scales sculptured externally and pierced by prominent mucous tubes. Caturus becomes extinct. Teeth and scales of Lepidotus (with Sρh<xrodus as subgenus), clearly a freshwater fish, are widely distributed in the Wealden, and finally disappear in the Chalk; its body was covered with large rhomboidal ganoid scales. Gyrodus and Aspidorhynchus occur in the beds of Voirons, Cοelodus and Amiopsis (allied to Amia) in those of Comen, in Istria. But the Pal<xichthyes are now in the minority; undoubted Teleosteans have appeared for the first time on the stage of life in numerous genera, many of which are identical with still existing fishes. The most of these are Acanthopterygians, but Physostomes and Plectognaths are likewise well represented, most of them being marine. Of Acanthopterygian families the first to appear are the Berycidae, represented by several very distinct genera: — Beryx∖ Pseudoberyx, with abdominal ventral fins; Berycoρ-sis, with cycloid scales; Hornonotus, Stenostoma, Spheno-cej)halus, Acaηus, Hoplopteryx, Platycornus, with granular scales; Podocys, with a dorsal extending to the neck; Acrogaster, Macrolepis t Rhacolepis, from the Chalk of Brazil.

The position of Pycηosteτyηx is uncertain; it approaches certain Pharyngognaths. True Percidae are abseut, whilst the Carangidae, Sphyroenidce, Cataphracti, Gobiidoe, Cob tidoe, and Sparidoe are represented by one or more genera. Somewhat less diversified are the Physostomes, which belong principally to the Clupeidoe and Dercetidoe, most of the genera being extinct; Cluρea is abundant in some localities. Scopelidae (Hemisaurida and Saurocephalus) occur in the Chalk of Comen in Istria, and of Maestricht. Of all Cretaceous deposits none surpass those of the Lebanon for the number of genera, species, and individuals; the forms are exclusively marine, and the remains in the most perfect condition.

In the Tertiary epoch the Teleosteans have almost entirely replaced the Ganoids; a few species only of the latter make their appearance, and they belong to existing genera (Lepidosteus, Acipenser). The Chondropterygians merge more and more into recent forms; Holocephali continue, and still are better represented than in the present fauna. The Teleosteans show even in the Eocene a large proportion of existing genera, and the fauna of some localities of the Miocene (Oeningen) is almost wholly composed of them. Of the whole more than one-half have already been found to belong to existing genera, and there is no doubt that the number of distinct genera now seemingly extinct will be lessened as the fossils come to be examined with a better knowledge of the living forms. The distribution of the fishes differed widely from that of our period, many of our tropical genera occurring in localities which are now included within our temperate zone, and being mixed with others that nowadays are restricted to a colder climate,—a combination which continues throughout the Pliocene.

A few families of fishes, like the freshwater Salmonidoe, seem to have put in their appearance in Post-Pliocene times; not much attention, however, has been paid to fishremains of these deposits; and such as have been incidentally examined furnish evidence of the fact that the distribution of fishes has not undergone any further essential change down to the present period.

GEOGRAPHICAL DISTRIBUTION.

In an account of the geographical distribution of fishes the freshwater forms are to be kept separate from the marine. When we attempt, however, to draw a line between these two kinds of fishes, we meet with a great number of species and of facts which would seem to render that distinction very vague. There are not only species which can gradually accommodate themselves to a sojourn in either salt or fresh water, but there are also some that seem to be quite indifferent to a rapid change from the one into the other; so that individuals of one and the same species (Gastrosteus, Gobius, Biennius, Osmeτus, Retropinna, Clupea, Syngna- thus, &c.) may be found some distance out at sea, whilst others live in rivers far beyond the influence of the tide, or even in inland fresh waters without outlet to the sea. The majority of these fishes belong to forms of brackish-water fauna; and, as they are not an insignificant portion of the fauna of almost every coast, we shall have to treat of them in a separate chapter.

Almost every large riyer offers instances of truly marine fishes ascending for hundreds of miles of their course, and not periodically, or from any apparent physiological necessity, but sporadically throughout the year. This is evidently the commencement of a change in a fish’s habits; and, indeed, not a few of such fishes have actually taken up their permanent residence in fresh waters, as species of Ambassis, Apogon, Dules, Therapon, Sciοeηa, Biennius, Gobius, Atlιerina, Mugil, Myxus, Hemirhamphus, Clupea, Anguilla, Tetrodon, Trygon, — all forms originally marine.

On the other hand, we find fishes belonging to freshwater genera descending rivers and sojourning in the sea for a more or less limited period; but these instances are much less in number than those in which the reverse obtains. We may mention species of Salmo (the common trout, the northern charr), and Siluroids (as Arius, Plotosus'). Careg onus, a genus so characteristic of the inland lakes of Europe, Northern Asia, and North America, nevertheless offers some instances of species wandering by the effluents into the sea, and taking up their residence in salt water, apparently by preference, as Coregonus oxy-rhynchus. But of all the freshwater families none exhibit so great a capability of surviving the change from fresh into salt water as the Gastrosteidoe (sticklebacks) of the northern hemisphere, and the equally diminutive Cypτino- dontidoe of the tropics; not only do they enter into, and live freely in, the sea, but many species of the latter family inhabit inland waters, which, not having an outlet, have become briny, or impregnated with a larger proportion of salts than pure sea water. During the voyage of the “Challenger” a species of Fundulus (F. nigrofasciatus) which inhabits the fresh and the brackish waters of the Atlantic States of North America was obtained, with Scopelids and other pelagic forms, in the tow-net, midway between St Thomas and Teneriffe.

Some fishes annually or periodically ascend rivers for the purpose of spawning, passing the rest of the year in the sea, as sturgeons, many Salmonoids, some Clupeoids, lampreys, &c. The first two evidently belonged originally to the freshwater series, and it was only in the course of their existence that they acquired the habit of descending to the sea, perhaps because their freshwater home did not furnish a sufficient supply of food. These migrations of freshwater fishes have been compared with the migrations of birds; but they are much more limited in extent, and do not, as is the case with birds, impart an additional element to the fauna of the place to which the fishes migrate.

The distinction between freshwater and marine fishes is further obscured by geological changes, in consequence of which the salt water is gradually being changed into fresh, or vice versa. These changes are so gradual, and spread over so long a time, that many of the fishes inhabiting such localities accommodate themselves to the new conditions. One of the most remarkable and best studied instances of such an alteration is the Baltic, which, during the second half of the Glacial period, was in open and wide communication with the Arctic Ocean, and evidently had the same marine fauna as the White Sea. Since then, by the rising of the land of Northern Scandinavia and Finland, this great gulf of the Arctic Ocean has become an inland sea, with a narrow outlet into the North Sea, and, in consequence of the excess of fresh water pouring into it over the loss by evaporation, it has been so much diluted as to be nearly fresh at its northern extremities; and yet nine species, the origin of which from the Arctic Ocean can be proved, have survived the changes, propagating their species, and agreeing with their brethren in the Arctic Ocean in every point, but remaining comparatively smaller. On the other hand, fishes which we must regard as true freshwater fishes, like the rudd, roach, pike, perch, enter freely the brackish water of the Baltic. Instances of marine fishes being permanently retained in fresh water in consequence of geological changes are well known: as Cottus quadricarms iu the large lakes of Scandinavia; species of Gobius, Biennius, and Atherina in the lakes of northern Italy; Comepharus, which seems to be a dwarfed Gadoid, in the depths of Lake Baikal. Carcharias gangeticus in inland lakes of the Fiji Islands, is another instance of a marine fish which has permanently established itself in fresh water.

Thus there is a constant interchange of species in progress between the freshwater and marine faunae, and in not a few cases it would seem almost arbitrary to refer a genus or even a larger group of fishes to the one or the other; yet there are certain groups of fishes which entirely, or with but few exceptions, are,, and apparently during the whole period of their existence have been, inhabitants either of the sea or of fresh water; and, as the agencies operating upon the distribution of marine fishes differ greatly from those influencing the dispersal of freshwater fishes, the two series must be treated separately. The most obvious fact that dry land, which intervenes between river systems, presents to the rapid spreading of a freshwater fish an obstacle which can be surmounted only exceptionally or by a most circuitous route, whilst marine fishes may readily and voluntarily extend their original limits, could be illustrated by a great number of instances. Without entering into details, it may suffice to state, as the general result, that no species or genus of freshwater fishes has anything like the immense range of the corresponding categories of marine fishes, and that, with the exception of the Siluroids, no freshwater family is so widely spread as the families of marine fishes. Surface temperature or climate, which is, if not the most, one of the most important physical factors in the limitation of freshwater fishes, similarly .affects the distribution of marine fishes, but in a less degree, and only in the case of those which live near the shore or the surface of the ocean; it ceases to exercise its influence in propor∙ tion to the depth, the true deep-sea forms being entirely exempt from its operation. Light, which is pretty equally distributed over the localities inhabited by freshwater fishes, cannot be considered as an important factor in their distribution, but it contributes to the formation of the impassable barrier between the surface and abyssal forms of marine fishes Altitude has stamped the fishes of the various alpine provinces of the globe with a certain character, and limited their distribution; but the number of these alpine forms is comparatively small, ichthyic life being extinguished at great elevations even before the mean temperature equals that of the high latitudes of the Arctic region, in which some freshwater fishes flourish, On the otner hand, the depths of the ocean, far exceeding the altitude of the highest mountains, still swarm with forms specially adapted for abyssal life. That other physical conditions of minor and local importance, under which freshwater fishes live, and by which their dispersal is regulated, are more complicated than similar ones of the ocean, is probable, though perhaps less so than is generally supposed; for the fact is that the former are more accessible to observation than the latter, and are therefore more generally and more readily comprehended and acknowledged. It will thus be necessary to treat of the two series separately, not only because m⅛ny of the most characteristic forms of the marine and the freshwater series are found, on taking a broader view of the subject, to be sufficiently distinct, but also because their distribution depends on causes different in their nature as well as in the degree of their action. Whether the oceanic areas correspond in any way to the terrestrial will be seen in the sequel.

Freshwater Fishes.

Having shown above that numerous marine fishes enter fresh waters, and that some of them have permanently established themselves therein, we have to eliminate from the category of freshwater fishes all such adventitious elements. They are derived from forms the distribution of which is regulated by other agencies, and which therefore would obscure the relations of the faunae of terrestrial regions if they were included in them. They will be mentioned with greater propriety along with the fishes constituting the fauna of brackish water.

The true freshwater fishes are all embraced in the following families and groups :—

Dipnoi, with 4 species; Acipenseridae and Polyodontidae, 26; Amiidae, 1; Polypteridae, 2; Lepidosteidae, 3; Percina, 46; Grystina, 11; Aphredoderidae, 1; Centrarchina, 26; Doles, 10; Nandidae, 7; Polycentridae, 3; Labyrinthici, 30; Luciocephalidae, 1; Gastrosteus, 10; Ophiocephalidae, 31; Mastacembelidae, 13; Chrornides, 105; Comephoridae, 1; Gadopsidae, 1; Siluridae, 572; Characinidae, 261; Haplochitonidae, 3; Salmonidae (3 genera excepted), 135; Percopsidae, 1; Galaxiidae, 15; Morrnyridae (and Gymnarchidae), 52; Esocidae, 8; Umbridae, 2; Cyprinodontidae, 112; Heteropygii, 2; Cyprinidae, 724; Kneriidae, 2; Hyodontidae, 1; Pantodontidae, 1; Osteoglossidae, 5; Notopterîdae, 5; Gymnotidae, 20; Symbranchidae, 5; Petromyzontidae, 12. Total, 2270 species.

As in every other class of animals, these freshwater genera and families vary exceedingly with regard to the extent of their geographical range,—some extending over more than the half of the continental areas, whilst others are limited to one continent only, or even to a very small portion of it. As a general rule, a genus or family of freshwater fishes is regularly dispersed and most developed within a certain district, the species and individuals becoming fewer towards the periphery as the type recedes more from its central home, some outposts, however, frequently being pushed far beyond the outskirts of the area occupied by it. At the same time remarkable instances exist of closely allied forms occurring, almost isolated, at most distant points, without being connected by allied species in the intervening space, and of members of the same family, genus, or species inhabiting the opposite shores of an ocean, and separated by many degrees of abyssal depths.

The dispersal of freshwater fishes has been effected in various ways; probably all the causes are still in operation, most of them working so slowly and imperceptibly as to escape direct observation. From the great number of freshwater forms which we see at the present day already acclimatized or gradually becoming acclimatized in the sea, or periodically or sporadically migrating to it, we must conclude that, under certain circumstances, salt water may cease to be an impassable barrier at some period of the existence of freshwater species, and that many of them have passed from one river through salt water into another. Secondly, the headwaters of some of the largest rivers, the mouths of which are at opposite ends of the continents which they drain, are sometimes distant from each other a few miles only; the intervening space may easily have been bridged over for the passage of fishes by a slight geological change affecting the level of the watershed, or even by temporary floods; and a communication of this kind, if existing for a limited period only, would afford the ready means for an exchange of a number of species previously peculiar to one or the other of those river or lake systems. Some fishes provided with gill-openings so narrow that the water moistening the gills cannot readily evaporate, and endowed, besides, with an extraordinary degree of vitality, like many Siluroids (Clarias, Callichthys∖ eels, &c., are enabled to wander for some distance over land, and thus may reach a watercourse leading them thousands of miles from their original home. Finally, fishes or their ova may be accidentally carried by waterspouts, or by aquatic birds or insects, to considerable distances.

Freshwater fishes of the present fauna were already in existence when the great changes in the distribution of land and water took place in the Tertiary epoch; and, having seen that salt water is not an absolute barrier to the spreading of freshwater fishes, we can now more easily account for those instances of singular disconnexion of certain families or genera. It is not necessary to assume that there was a continuity of land stretching from the present coast of Africa to South America, or from South America to New Zealand and Australia, to explain the presence of identical forms in localities so distant; it suffices to assume that the distances were lessened by intervening archipelagoes, or that an alteration has taken place in the level of the land area.

Dispersal of a type over several distant continental areas may be evidence of its great antiquity, but does not prove that it is of greater antiquity than another limited to one region only. Geological evidence is the only proof of the antiquity of a type. Thus, although the Dipnoi occur in the continents of Africa, South America, and Australia, and their present distribution is evidently the consequence of their wide range in Palaeozoic and Secondary epochs, the proof of their high antiquity can be found in their fossil remains only. The Siluroids, for example, have a still greater range, but their wide distribution is of comparatively recent date, as the few fossil remains that have been found belong to the Tertiary epoch. The rapidity of dispersal of a type depends entirely on its power to accommodate itself to a variety of physical conditions, and on the degree of vitality by which it is enabled to survive more or less sudden changes under unfavourable conditions; proof of this is afforded by the family of Siluroids, many of which can suspend for some time the energy of their respiratory functions, and readily survive a change of water.

To trace the geological sequence of the distribution of an ichthyic type, and to recognize the various laws which have governed and are still governing its dispersal, is one of the ultimate tasks of ichthyology. But the endeavour to establish by means of our present fragmentary geological knowledge the divisions of the fauna of the globe leads us into a maze of conflicting evidence; as Mr Wallace truly observes, “any attempt to exhibit the regions of former geological ages in combination with those of our own period must lead to confusion.” Nevertheless, as the different types of animals found at the present day within a particular area have made their appearance therein at distant periods, we should endeavour, in giving an account of the several zoo-geographical divisions, to decide, so far as we can, the following questions :—

1. Which of the fishes of an area should be considered to be the remnants of ancient types, probably spread over much larger areas in preceding epochs?

2. Which are to be considered to be autochthonous species, that is, forms which in the Tertiary epoch or later came into existence within the area to which they are still limited, or from which they have since spread 1

3. Which are the forms which must be considered to be immigrants from some other region?

It is the aim of every philosophical classification to indicate the degrees of affinity which obtain between the various categories. In dividing the earth’s surface into zoological regions, the two families, Cyprinidae and Siluridae, the former of which yields a contingent of one-third and the latter of one-fourth of all the known freshwater species of our period, afford most important guidance for the estimation of those degrees of affinity. The Cyprinoids may be assumed to have originated in the alpine region dividing the temperate and tropical parts of Asia; endowed with a greater capability than any other family of freshwater fishes of acclimatizing themselves in a temperate as well as in a tropical region, they spread north and south as well as east and west; in the Pre-Glacial epoch they reached North America, but they have not had time to penetrate into South America, Australia, or the islands of the Pacific. The Siluroids, principally fishes of the sluggish waters of the plains, well adapted for surviving changes of the water in which they live, and for living either in mud or in seawater, flourish most in the tropical climate in which this type evidently had its origin. They came into existence after the Cyprinoids, their fossil remains being found only in Tertiary deposits in India, none in Europe. They rapidly spread over the areas of land within the tropical zone, reaching Northern Australia from India, and one species migrating even into the Sandwich Islands, probably from South America. The coral islands of the Pacific still remain untenanted by them. Their progress into temperate regions was evidently slow, only very few species having penetrated into the temperate parts of Asia and Europe, and the North American species, although more numerous, showing no great variety of structure, all belonging to the same group (Amiurina). Towards the south their progress was still slower, Tasmania, New Zealand, and Patagonia being without any representative, whilst the streams of the Andes of Chili are inhabited by a few dwarfed forms identical with such as are characteristic of similar localities in the more northern and warmer parts of the South American continent.

These remarks may serve to introduce the following division of the fauna of freshwater fishes :—

I. The Northern Zone. —Characterized by Acipenseridae. Few

Siluridae. Numerous Cyprinidae. Salmonidae, -Esocidae.

1. Eu,roρo-Asiatic or Paloêarctic Region.— Characterized by absence of osseous Ganoidei; Cobitidae and Barbus numerous.

2. North American Region.— Characterized by osseous Ganoi dei, Amiurina, and Catostomina; but no Cobitidae or Barbus.

II. The Equatorial Zone. —Characterized by the development of Siluridae.

A. Cyprinoid Division.— Characterized by presence of Cyprinidae and Labyrinthici.

1. Indian Region.— Characterized by [absence of

Dipnoi] Ophiocephalidae, Mastacembelidae. Cobitidae numerous.

2. African Region.— Characterized by presence ot

Dipnoi and Polypteridae. Chromides and Characinidae numerous. Mormyridae. Cobitidae absent.

A. Acyprinoid Division. — Characterized by absence of Cyprinidae and Labyrinthici.

1. Tropical American or Neotropical Region.— Char acterized by presence of Dipnoi. Chromides and Characinidae numerous. Gymnotidae.

2. Tropical Pacific Region.— Characterized by pre sence of Dipnoi. Chromides and Characinidae absent.

III. The Southern Zone. —Characterized by absence of Cypri nidae, and scarcity of Siluridae. Haplochitonidae and Galaxiidae represent the Salmonoids and Esoces of the northern zone. One region only.

1. Antarctic Region.— Characterized by the small number of species; the fishes of—

a . The Tasmanian sub-region, b . The New Zealand sub-region, and c. The Patagonian sub-region, being almost identical.

In the following account we begin with a description of the equatorial zone, this being the one from which the two principal families of freshwater fishes seem to have spread.

Equatorial Zone. —Roughly speaking, the borders of this zoological zone coincide with the geographical limits of the Tropics of Cancer and Capricorn; its characteristic forms, however, extend in undulating lines several degrees both northwards and southwards. Commencing from the west coast of Africa, the desert of the Sahara forms a well-marked boundary between the equatorial and northern zones; as the boundary approaches the Nile, it makes a sudden sweep towards the north as far as northern Syria (Afastacembel·us, near Aleppo and in the Tigris; Clamas and Chromides, in the Lake of Galilee), crosses through

Persia and Afghanistan (f)phiocephalus') to the southern ranges of the Himalayas, and follows the course of the Yang-tse-Keang, which receives its contingent of equatorial fishes through its southern tributaries. Its continuation through the North Pacific may be considered as indicated by the tropic, which strikes the coast of Mexico at the southern end of the Gulf of California. Equatorial types of South America are known to extend so far northwards; and, by following the same line, the West India Islands are naturally included in this zone.

Towards the south the equatorial zone embraces the whole of Africa and Madagascar, and seems to extend still farther south in Australia, its boundary probably following the southern coast of that continent; the detailed distribution of the freshwater fishes of south-western Australia has been but little studied, but the few facts which we know show that the tropical fishes of that region follow the principal water-course, the Murray river, far towards the south and probably to its mouth. The boundary line then stretches to the north of Tasmania and New Zealand, coinciding with the tropic until it strikes the western slope of the Andes, on the South American continent, where it again bends southward to embrace the system of the Rio de la Plata.

The four regions into which the equatorial zone is divided arrange themselves into two well-marked divisions, one of which is characterized by the presence of Cyprinoid fishes, combined with the development of Labyrinthici, whilst in the other both these types are absent. The boundary between the Cyprinoid and Acyprinoid division seems to follow Wallace’s line,—a line drawn from the south of the Philippines between Borneo and Celebes, and farther south between Bali and Lombok. Borneo abounds in Cyprinoids; from the Philippine Islands a few only are known at present, and in Bali two species have been found; but none are known from Celebes or Lombok, or from islands situated farther east.

Taking into consideration the manner in which Cyprinoids and Siluroids have been dispersed, we are obliged to place the Indian region as the first in the order of our treatment; and indeed the number of freshwater fishes which appear to have spread from it into the neighbouring regions far exceeds that of the species which it has received from them.

The Indian Region comprises the whole continent of Asia south of the Himalayas and the Yang-tse-Keang, and includes the islands to the west of Wallace’s line. Towards the north-east the island of Formosa, which also by other parts of its fauna shows the characters of the equatorial zone, has received some characteristic Japanese freshwater fishes, for instance, the singular Salmonoid Plecoglossus. Within the geographical boundaries of China the freshwater fishes of the tropics pass gradually into those of the northern zone, both being separated by a broad debateable ground. The affluents of the great river traversing this district are more numerous from the south than from the north, and carry the southern fishes far into the temperate zone. Scarcely better defined is the boundary of this region towards the north-west. Before Persia passed through the geological changes by which its waters were converted into brine and finally dried up, it seems to have been inhabited by many characteristic Indian forms, of which a few still survive in the tract intervening between Afghanistan and Syria; Ophiocephalus and Discogηathus have each at least one representative, Macrones has survived in the Tigris, and Mastacembelus has penetrated as far as Aleppo. Thus freshwater fishes belonging to India, Africa, and Europe are intermingled in a district which forms the connecting link between the three continents. Of the freshwater fishes of Arabia we are almost entirely ignorant; we only know that the Indian Discognathus lamta occurs in the reservoirs of Aden (having also found its way to the opposite African coast), and that the ubiquitous Cyprinodonts thrive in the brackish pools of northern Arabia.

In analysing the list of Indian fishes, we find that out of 40 families or groups of freshwater fishes 12 are represented in this region, and that 625 species are known to occur in it, or two-sevenths of the entire number of freshwater fishes known. This large proportion is principally due to the development of numerous local forms of Siluroids and Cyprinoids, of which the former show a contingent of about 200, and the latter of about 330 species. The combined development of those two families, therefore, and their undue preponderance over the other freshwater types, is the principal characteristic of the Indian region. The second important character of its fauna is the apparently total absence of Ganoid and Cyclostomous fishes. Every other region has representatives of either Ganoids or Cyclostomes, some of both.

Of the autochthonous freshwater fishes of the Indian region, some are still limited to it, viz., the Nandiηa, the Luciocephalidae (of which one species only exists in the archipelago), of Siluroids the Chacina and Bagariina, of Cyprinoids the Semiplotina and Homaloptemna ; others are very nearly so, such as the Labyrinthici, Ophiocephalidoe, Mastacembelidoe, of Siluroids the Silurina, of Cyprinoids the Iiasbomna and Danionvna, and the Symbranchidoe.

The regions to which the Indian has least similarity are the North American and the Antarctic, as they are the most distant. Its affinity to the other regions is of very different degrees :—

1. Its affinity to the Europo-Asiatic region is very slight, and is indicated almost solely by three groups of Cyprinoids, viz., the Cypriηiηα, Abramidina, and Cοbitidina. The development of these groups north and south of the Himalayas is due to their common origin in the highlands of Asia; but the forms which descended into the tropical climate of the south are now so different from their northern brethren that most of them are referred to distinct genera. The only genera which are still common to both regions are (1) the true barbels (Barbus), a genus which of all Cyprinoids has the largest range over the Old World, and of which some one hundred and sixty species have been described, and (2) the mountain barbels (Schizothorax, &c. ), which, peculiar to the alpine waters of Central Asia, descend a short distance only towards the tropical plains, but extend farther into rivers within the northern temperate districts. The origin and the laws of the distribution of the Cobitidina appear to have been identical with those of Barbus, but they have not spread into Africa.

2. There exists a great affinity between the Indian and African regions; seventeen out of the twenty-six families or groups found in the former are represented by one or more species in Africa, and many of the African species are not even generically different from the Indian. As the majority of these groups have many more representatives in India than in Africa, we may reasonably assume that the African species have been derived from the Indian stock; but probably this is not the case with the Siluroid group of Clariina, which with regard to species is nearly equally distributed between the two regions, the African species being referable to three genera (Clarias, Ηeterobranchus, Gymnallabes, with the sub-genus Chan-nallabes), whilst the Indian species belong to two genera only, viz., Clarias and Heterobranchus. On the other hand, the Indian region has derived from Africa one freshwater form only, viz., Ebroρlus, a member of the family of Chromides, so well represented in tropical Africa and South America. Etroplus inhabits southern and western India and Ceylon, and has its nearest ally in a Madagascar freshwater fish, Paretroplus. Considering that other African Chromides have acclimatized themselves at the present day in saline water, we think it more probable that Ebroplus should have found its way to India through the ocean than over the connecting land area, where, besides, it does not occur.

3. No closer affinity exists between the Indian and Tropical American regions than is indicated by the character of the equatorial zone generally. With two exceptions, no genus of freshwater fishes occurs in India and South America without being found in the intermediate African region. Four small Indian Siluroids (Sisor, Erethistes, Pseτιdecheneis, and Exostoma) have been referred to the South .American Hypostomatina ; but it remains to be seen whether this combination is based upon a sufficient agreement of their internal structure, or whether it is not rather artificial. On the other hand, the occurrence and wide distribution in tropical America of a fish of the Indian family Symbranchidoe (Symbranchus marmoratus), which is not only congeneric with, but also most closely allied to, the Indian Symbranchτιs bengaleusis, furnishes one of those extraordinary anomalies in the distribution of animals of which no satisfactory explanation can at present be given.

4. The relation of the Indian to the Tropical Pacific region consists only in its having contributed a few species to the poor fauna of the latter. This immigration must have taken place within a recent period, because some species now inhabit the fresh waters of tropical Australia and the South Sea Islands without having in any way changed their specific characters, as Lates calcarifer, species of Dules, Plotosus anguillaris ; others (species of A rius) differ but little from their Indian congeners. All these fishes must have migrated by the sea, a supposition which is supported by what we know of their habits. We need not add that India has not received a single addition to its freshwater fish fauna from the Pacific region.

It may be mentioned, before concluding these remarks oh the Indian region, that peculiar genera of Cyprinoids and Siluroids inhabit the streams and lakes of its alpine ranges in the north. Some of them, like the Siluroid genera Clyptosternum, Euglyptosternum, Pseudecheneis, have a folded disk on the thorax between their horizontally spread pectoral fins; by means of this they adhere to stones at the bottom of the mountain torrents, and without it they would be swept away into the lower courses of the rivers. The Cyprinoid genera inhabiting similar localities and the lakes into which alpine rivers pass,—such as Oreinus, Schizothorax, Ptychobarbus, Schizopygopsis, Diptychus, Gymnocypηs,— are distinguished by peculiarly enlarged scales near the vent, the physiological use of which has not yet been ascertained. These alpine genera extend far into the Europo-Asiatic region, where the climate is similar to that of their southern home. No observations have been made by which the altitudinal limits of fish life in the Himalayas can be fixed, but it is probable that it reaches the line of perpetual snow, as in the European Alps, which at that height are inhabited by Salmonoids. Griffith found an Oreinus and a loach, the former in abundance, in the Helmund at Gridun Dewar, altitude 10,500 feet, and another loach at Kaloo at 11,000 feet.

The African Region comprises the whole of the African continent south of the Atlas and the Sahara. It might have been conjectured that the more temperate climate of its southern extremity would have been accompanied by a conspicuous difference in the fish fauna. But this is not the case; the difference between the tropical and southern parts of Africa consists simply in the gradual disappearance of specifically tropical forms, whilst Siluroids, Cyprinoids, and even Labyrinthici penetrate to its southern coast; no new form has entered to impart to South Africa a character distinct from the central portion of the continent. In the north-east the African fauna passes the Isthmus of Suez and penetrates into Syria; the system of the Jordan presents so many African types that it has to be included in a description of the African region as well as of the Europo-Asiatic. This river is inhabited by three species of Chromis, one of Hemichromis, and Clarias macracanthus, a common fish of the upper Nile. Madagascar clearly belongs to this region. Besides some gobies and Dules, which are not true freshwater fishes, four Chromides are known. To judge from general accounts, its freshwater fauna is poorer than might be expected; but, singular as it may appear, collectors have hitherto paid but little attention to the freshwater fishes of this island. The fishes found in the fresh waters of the Seychelles and Mascarenes are brackish-water fishes, such as Fundulus, Haplochilus, Elops, Mugil, &c.

Out of the 40 families or groups of freshwater fishes 15 are represented in the African region, or three more than in the Indian region; of two of them, however, viz., the Ophiocephalidoe and Mastacembelidoe, a few species only have found their way into Africa. On the other hand, the number of species is much less, viz., 255, being only two-fifths of the known Indian species. The small degree of specialization and localization is principally due to the greater uniformity of the physical conditions of this continent, and to the almost perfect continuity of the great river systems, which take their origin from the lakes in its centre. This is best shown by a comparison of the fauna of the upper Nile with that of the West African rivers. The number of species known from the upper Nile amounts to 56, and of these not less than 25 are absolutely identical with West African species. There is an uninterrupted continuity of the fish fauna from the west to the north-east, and the species known to be common to both extremities may be reasonably assumed to inhabit also the great reservoirs of water in the centre of the continent. A greater dissimilarity is noticeable between the west and north-east fauna on the one hand and that of the Zambezi on the other; the affinity between them is merely generic, and all the fishes hitherto collected in Lake Nyassa have proved to be distinct from those of the Nile, and even from those of other parts of the system of the Zambezi.

Unlike India, Africa does not possess either alpine ranges or outlying archipelagoes, the fresh waters of which would swell the number of its indigenous species; but, when its fauna becomes better known than at present, the great difference in the number of species between this and the Indian regions may possibly be somewhat lessened.

The most numerously represented families are the Siluroids, with 61 species; the Cyprinoids, with 52; the Hormyridae, with 51; the Characinidoe, with 35; and the Chromides, with 29. There is not, therefore, that great preponderance of the first two families over the rest which we noticed in the Indian region; in Africa there is a comparatively greater variety of distinct freshwater types, making the study of its fauna an unflagging pleasure such as is scarcely reached in the study of the other region. With the forms peculiar to it there are combined those of India as well as of South America.

In tropical Africa there are still remnants of Ganoids,— Protopterus (Lepidosiren) annectens and Ρolypterus bichir, with the singularly modified Calαmoichthys. The first two range from east to west, and are accompanied by an Osteoglossoid (Heterotis), which has hitherto been found in the Nile and on the west coast only. Autochthonous and limited to this region are the Mormyridae, Paη- todontidoe, and Kneriidoe, a singular type somewhat akin to the loaches. Of Siluroid genera the most characteristic are Synodontis, Rhinoglanis, and the electric Hαlapterurus; of Characinoids, Cithamnus, Alestes, Xenο-chαrax, Hydrocyon, Distichodoη, Lchthyborns.

The regions to which Africa (like India) has least similarity are again the North American and Antarctic. Its affinity with the Europo-Asiatic region consists only in its having received, like the latter, a branch of the Cyprinoids, the African carps and barbels, which on the whole resemble Indian more than Europo-Asiatic forms. Its similarity to Australia is limited to the two regions possessing Dipnoous and Osteoglossoid types. But its relations to the other two regions of the equatorial zone are near and of great interest.

1. Africa has in common with India the Siluroid groups of the Clαriinα, Silurina, and Bagrina, and more especially the small but very natural family of Notopteridae, represented by three species in India, and by two on the west coast of Africa. It would be hazardous to state at present in which of the two regions these fishes first made their appearance, but the discovery of remains of Noto· pteridoe and Silurina in Tertiary deposits of Sumatra points to the Indian region as their original home. We are in less doubt about the other, fishes common to the two regions; they are clearly immigrants into Africa from the east, and it is a remarkable fact that these immigrants have penetrated to the most distant limits of Africa in the west as well as in the south, viz., the Labyrinthici, represented by two genera closely allied to the Indian Anabas ; the Ophiocephalidoe and Mastaccmbclidoe, a few species of which have penetrated to the west coast, while, singularly enough, they are absent from the eastern rivers; the Ariina, represented by several species, of which one or two are identical with Indian, having extended their range along the intervening coasts to the east coast of Africa. The Cyprinoids also afford an instance of an Indian species ranging into Africa, viz., Discognathus lamta, which seems to have crossed at the southern extremity of the Red Sea, as it is found in the reservoirs at Aden and in the hill-streams of the opposite coastregion of Abyssinia.

2. No such direct influx of species and genera has occurred from South America into Africa, yet the affinity of their freshwater fishes is striking. Two of the most natural'families of fishes, the Chromides and Characinidoe, are peculiar and (with the exception of Etroplus) restricted to them. The African and South American Dipnoi are closely allied to each other. The Ρirnclodina, so characteristic of tropical America, have three representatives in Africa, viz., Pimelodτcsplatychir, Γ. balayi, and Auchenoglanis biscutatus; the Doradina are another Siluroid group restricted to these two continents. Yet, with all these points of close resemblance, the African and South American series are, with the exception of the two species of Pimelodus, generically distinct,—which shows that the separation of the continents must have been of old date. On the other hand, the existence of so many similar forms on both sides of the Atlantic affords much support to the supposition that at a former period the distance between the present Atlantic continents was much less, and that the fishes which have diverged towards the east and west are descendants of a common stock which had its home in a region now submerged under some intervening part of that ocean. Be this as it may, it is evident that the physical conditions of Africa and South America have remained unchanged for a considerable period, and are still sufficiently alike to preserve the identity of a number of peculiar freshwater forms on both sides of the Atlantic. Africa and South America are, moreover, the only continents which have produced in freshwater fishes, though in very different families, one of the most extraordinary modifications of an organ—the conversion, that is, of muscle into an apparatus creating electric force.

The boundaries of the Neotropical or Tropical American Region have been sufficiently indicated in the definition of the equatorial zone. A broad and most irregular band of country, in which the South and North American forms are mixed, exists in the north, offering some peculiarities which deserve fuller attention in the subsequent description of the relations between the South and North American faunae.

Out of the 40 families or groups of freshwater fishes, 9 only are represented in the tropical American region. This may be accounted for by the fact that South America is too much isolated from the other regions of the equatorial zone to have received recent additions to its fauna. On the other hand, the number of species (672) exceeds that of every other region, even of the Indian, with which, in regard to the comparative development of families, the neotropical region shows a close analogy, as will be seen fron? the following table :—

Indian.— Siluridae, 200 species; Cyprinidae, 330; Labyrinthici, 2t>; Üphiocephalidae, 30; Mastacembelidae, 10.

Neotropical. — Siluridae, 276 species; Cliaracinidae, 226; Chromides, 80; Cyprinodontidae, 60; Gymnotidae, 20.

In both regions the great number of species is due to the development of numerous local forms of two families, the Characinidoe in the New World taking the place of the Cyprvnidoe of the Old. To these are added a few smaller families with a moderately large number of species, which, however, is only a fraction of that of the leading families, the remainder being represented by a few species only. The number of genera within each of the two regions belonging to the two principal families is also singularly alike; the Indian region having produced about 45 Siluroid and as many Cyprinoid genera, whilst the Neotropical region is tenanted by 54 Siluroid and 40 Chara-cinoid genera. These points of similarity between the two regions cannot be accidental; they indicate that agreement in their physical and hydrographical features which in reality exists. Of Ganoids, we find in tropical America one species only, Lepidosiren paradoxa, accompanied by two Osteoglossoids (Osteoglossum bicirrhosum and Arapaima gigash Autochthonous and limited to this region are the Pvlycentridae∙, all the non-African genera of Chromides and Characinidoe ; of Siluroids, the Hypophthalmina, Aspredi-nina, and Stegophilina, and the majority of Pimelodina, Hypostomatina, and Doradina ; the herbivorous Cyprinodonts or Limnophagae, and numerous insectivorous Cyprinodonts or Carnivoroe ; and the Gymnotidoe (electric eel).

The relations to the other regions are as follows :—

1. The resemblances to the Indian and Tropical Pacific regions partly date from remote geological epochs, or are partly due to that similarity of physical conditions to which we have already referred. We have again to draw attention to the unexplained presence in South America of a representative of a truly Indian type (not found in Africa), viz., Symbrαmchus marmoratus. On the other hand, a direct genetic affinity exists between the Neotropical and African regions, as has been noticed in the description of the latter, a great part of their freshwater fauna consisting of descendants from a common stock.

2. A comparison of the specifically Neotropical with the specifically North American types shows that no two regions can be more dissimilar. It is only in the intervening borderland, and in the large West Indian islands, that the two faunae mix with each other. We need not enter into the details of the physical features of Central America and Mexico; the broken ground, the diversity of climate (produced by different altitudes) within limited districts, the hot and moist alluvial plains surrounding the Mexican Gulf, offer a variety of conditions most favourable to the intermixture of the types from the north and the south. Still the interchange of peculiar forms appears to be only beginning; none have yet penetrated beyond the debateable ground, and it is evident that the land connexion between the two continents is of comparatively recent date,—a view which is confirmed by the identity of the marine fishes on both sides of Central America.

Cuba—which is the only island in the West Indies that has a number of freshwater fishes sufficient for the determination of its zoo-geographical relations—is inhabited by several kinds of a perch (Ceηtropomus), freshwater mullets, Cyprinodonts, one species of Chromid (an Acara), and Symbranchus marmoratus. All these fishes are found in Central America, and, as they belong to forms known to enter brackish water more or less freely, it is evident that they have crossed either from that region or from the mainland of South America. But with them there came a remarkable North American type, Lepidosteus. Lepidosteus viridis, which is found in the United States, has penetrated on the mainland to the Pacific coast of Guatemala, where it is common at the mouth of the rivers and in brackish-water likes along the coast; it probably crossed into Cuba from Florida. A perfectly isolated type of fishes inhabits the subterranean waters in the caves of Cuba (two species of Lucifuga). The eyes are absent, or quite rudimentary, as in most other cave animals. Singularly, it belongs to a family (Ophidiidoe) the members of which are strictly marine; and its nearest ally is a genus, Brotula, the species of which are distributed over the Indo-Pacific Ocean, one only occurring in the Caribbean Sea. This type must have witnessed all the geological changes which have taken place since Cuba rose above the surface of the sea. A similar mixture of forms of the tropical and temperate types of freshwater fishes takes place in the south of South America; its details have not yet been so well studied as in the north, but this much is evident, that, whilst in the east tropical forms follow the Plate river far into the temperate region, in the west the temperate fauna finds still a congenial climate in ranges of the Andes, situated close to, or even north of, the tropic.

Like the Indian region, the Tropical American has a peculiar alpine fauna, the freshwater fishes of which, however, belong to the Siluroids and Cyprinodonts. The former are small, dwarfed forms (Arges, Stygogenes, Brontes, Astroblepus, Trichomycterus, Hremορhilus), and have a perfectly naked body, whilst the representatives in the lowlands of at least the first four genera are mailed. The alpine Cyprinodonts, on the other hand (Orestiash exceed the usual small size of the other members of this family, and are covered with thick scales, but have lost their ventral fins. Some of these alpine forms, like Trichomycterus, follow the range of the Andes far into the Southern Temperate region. The majority are found at a height of 15,000 feet above the level of the sea, and a few still higher.

The Tropical Pacific Pegion includes all the islands east of Wallace’s line, New Guinea, Australia (with the exception of its south-eastern portion), and all the islands of the tropical Pacific to the Sandwich group. Comparing the area of this region with that of the others, we find it to be the poorest, not only in point of the number of its species generally (36), but also in the possession of peculiar forms. The paucity of freshwater fishes is due, in the first place, to the arid climate and the deficiency of water in the Australian continent, as well as to the insignificant size of the freshwater courses in the smaller islands. Still this cannot be the only cause; the large island of Celebes, which, by its mountainous portions, as well as by its extensive plains and lowlands, would seem to offer a favourable variety of conditions for the development of a freshwater fauna, is, so far as has been ascertained, tenanted by seven freshwater fishes only, viz., 2 Arius, 2 Ρlotosus, 1 Anabas, 1 Ophiocephalus, 1 Monopterus, all of which are the commonest species of the Indian region. New Guinea has not yet been explored, but, from the faunae nearest to this island, we expect its freshwater fishes will prove to be equally few in number, and identical with those of Celebes and North Australia,—a supposition confirmed by the few small collections which have reached Europe. Finding, then, that even those parts of this region which are favourable to the development of freshwater fishes have not produced any distinct forms, and that the few species which inhabit them are unchanged or but slightly modified Indian species, we must conclude that the whole of this area has remained geologically isolated from the other regions of this zone since the commencement of the existence of Teleostei, and that, with the exception of Ceratodus and Osteoglossum, the immigration of the other species is of very recent date.

Fossil remains of Ceratodus have been found in the Liassic and Triassic formations of North America, England, Germany, and India; it is, therefore, a type which was widely spread in the Mesozoic epoch. Although it would be rash to conclude that its occupation of Australia dates equally far back, for it may have reached that continent long afterwards, yet it is evident that, as it is one of the most ancient of the existing types, so it is certainly the first of the freshwater fishes which appeared in Australia. Osteoglossum, of which no fossil remains have yet been found, is proved by its distribution to be one of the oldest Teleosteous types. There must have been a long gap of time before these ancient types were joined by the other Teleostei. All of them migrated through the intervening parts of the ocean from India. Most of the Ρlotosiηa, some of the Arii, Dules, and Atherinichthys, also Nanno-perca (allied to Apogon), were among the earliest arrivals, being sufficiently differentiated to be specifically or even generically (Cηidoglanis, Nanηoperca) distinguished; but some others, like Anabas scandens, Lates calcarifer, Dules marginatus, must have reached the Australian continent quite recently, for they are undistinguishable from Indian specimens.

In south-western Australia a mingling of the scanty fauna with that of the southern temperate parts takes place. Oligorus macquariensis (the Murray cod), which has a congener on the coast of New Zealand, ascends high up the Murray river, so that we cannot decide whether this Percoid should be located in the tropical or the temperate part of Australia. Several Galaxias also extend to the confines of Queensland, and will probably some day be found members of this region.

In the smaller Pacific islands the freshwater fishes exhibit a remarkable sameness;· they comprise two or three species of Dales, several eels, an atherine, and some gobies, mullets, and other fishes which with equal readiness exchange fresh for salt water, and which would at once reach and occupy any streams or freshwater lakes that might be formed on an island.

The Sandwich Islands are the only group among the smaller islands which are tenanted by a Siluroid, a species of Arius, which is closely allied to Central American species, and therefore υrobably migrated from tropical America.

Northern Zone. —The boundaries of the northern zone coincide in the main with the northern limit of the equatorial zone; but, as has been already indicated, they overlap the latter at three different points. This happens in Syria, as well as east of it, where the mixed faunae of the Jordan and the rivers of Mesopotamia demand the inclusion of this territory in the northern zone as well as in the equatorial; in the island of Formosa, where a Salmonoid and several Japanese Cyprinoids flourish; and in Central America, where a Lepidosteus, a Cyprinoid (Sclera-gnathus meridionalis), and an Amiurus (A. meridionalis) represent the North American fauna in the midst of a host of tropical forms.

There is no separate arctic zone for freshwater fishes; ichthyic life becomes extinct towards the pole wherever the fresh water remains frozen throughout the year, or thaws for a few weeks only; and the few fishes which extend into high latitudes, in which lakes are open for two or three months in the year, belong to types in no wise differing from those of the more temperate south. The highest latitude at which fishes have been obtained is 82° N. lat., whence the late Arctic Expedition brought back specimens of charr (Salmo arcturιιs and Salmo ηaresii).

The ichthyological features of this zone are well marked. The Chondrosteous Ganoids or sturgeons, and the families of Salmonidoe and Esocidoe, are limited to and characteristic of it; Cyprinoids flourish with the Salmonoids, both families preponderating in numbers over the others, whilst the Siluroids are few in number and in variety.

The two regions into which this zone is divided are very closely related to one another, and their affinity is not unlike that which obtains between the sub-regions of the southern zone. Several species are common to both, viz., Acipenser sturio, A. maculatus, Perca fluviatilis, Gastrosteus ρungitius, Salmo salar, Esox lucius, Lota vulgaris, Petromyzon marinus, P. fluviatilis, and P. branchialis ; and ali recent investigations have resulted in giving additional evidence of the affinity and not of the diversity of the two regions.

In Europe and temperate Asia, as well as in North America, mountain ranges elevated above the line of perpetual snow would seem to offer physical conditions favourable for the development of a distinct alpine fauna. But this is not the case, because the difference of climate between the mountain districts and the lowlands is much less in this zone than in the equatorial. Consequently the alpine freshwater fishes do not essentially differ from those of the plains; they are principally Salmonoids, and in Asia there are also mountain-barbels and loaches. Salmo orientalis was found by Griffith to abound in the tributaries of the Bamian river at an altitude of about 11 000 feet.

The Pdloearctic or Europo-Asιatic Region.— The western and southern boundaries of this region coincide with those of the northern zone, so that only those which divide it from North America need to be indicated. Behring’s Strait and the Kamtchatka Sea have been conventionally taken as the boundary, but this is shown to be artificial by the fact that the animals of both coasts, so far as they are known at present, are not sufficiently distinct to be referred to two different regions. As to the freshwater fishes, those of north-western America and of Kamtchatka are but imperfectly known, but there can be little doubt that the same agreement exists between them as is the case -with other classes of animals. The Japanese islands exhibit a decided Palaearctic fish fauna, which includes Barbus and Cobitioids, farms strange to the North American fauna. A slight influx of tropical forms is perceived in the south of Japan, where two Bagrνna (PseudobagrU8 aurantiacus and Liocassis longirostms) have established themselves for a considerable .period, for both are peculiar to the island, and have not been found else where.

In the east, as well as in the west, the distinction between the Europo-Asiatic and the North American regions disappears almost entirely as we advance farther towards the north. Of four species of the genus Salmo known from Iceland, one (N. salar) is common to both regions, two are European (S.famo and S. alpinus), and one is a peculiarly Icelandic race (S. nivalis). So far as we know the Salmonoids of Greenland and the tract adjoining Baffin’s Bay, they are all very closely allied to European species, though they may be distinguished as local races.

Finally, as we have seen above, the Europo-Asiatic fauna mingles with African and Indian forms in Syria, Persia, and Afghanistan. Capoeta, a Cyprinoid genus, is characteristic of this district, and well represented in the Jordan and the rivers of Mesopotomia. Out of the 40 families of freshwater fishes 13 are represented in this region; the number of species is comparatively small, viz., 360.

Assuming that the distribution of Cyprinoids has taken its origin from the alpine tract of country dividing the Indian and Palaearctic regions, we find that this type has found in the temperate region as favourable conditions for its development as in the tropical. Out of the 360 species no less than 215 are Cyprinoids. In the countries and on the plateaus immediately adjoining the Himalayan ranges those mountain forms which we mentioned as peculiar to the Indian Alps abound, and extend for a considerable distance towards the west and east, mixed with other Cyρrinina and Cobitidina. The representatives of these two groups are more numerous in Central and Eastern Asia than in Europe and the northern parts of Asia, where the Leuciscina predominate. Abramidina or breams are more numerous in the south and east of Asia, but they spread to the extreme north-western and northern limits to which the Cyprinoid type readies. The Rhodeina are a small family especially characteristic of the East, but with one or two offshoots in Central Europe. Very significant is the appearance in China of a species of the Catostomiηa, a group otherwise limited to North America.

The Cyprinoids, in their dispersal northwards from the south, are met from the opposite direction by the freshwater Salmonoids. These fishes are, without doubt, one of the youngest families of Téleostei, for they did not appear before the Pliocene era; they flourished at any rate during the Glacial period, and, as is testified by the survivors which we find in isolated elevated positions, like the trout of the Atl is, of the mountains of Asia. Minor, and of the Hindu Kush, they spread to the extreme south of this region. At the present day they are most numerously represented in its northern temperate parts; towards the south they become fewer, but increase again in numbers and species where-ever a great elevation offers them the snow-fed waters which they affect. In the rivers of the Mediterranean, Salmonoids are by no means rare, but they prefer the upper courses of those rivers, and do not migrate to the sea, with the exception, perhaps, of some species in the rivers of the North Adriatic.

The pike, Umbra, and several species of perch and stickleback are also clearly autochthonous species of this region. Others belong to marine types, and seem to have been retained in fresh water at various epochs,—as the freshwater Cottas (miller’s thumb); Coitus quadricornis, which inhabits lakes of Scandinavia, whilst other individuals of the same species are strictly marine; the burbot (Lota vulgaris) ; and the singular Comephorus, a dwarrect and much changed Gadoid which inhabits the greatest depths of Lake Baikal.

Remnants of the Palaeichthyic fauna exist in the sturgeons and lampreys. The former inhabit in abundance the great rivers of eastern Europe and Asia, periodically ascending them from the sea; their southernmost limits are the Yang-tse-Keang in the east, and towards the centre of this region the rivers flowing into the Adriatic, Black, and Caspian Seas, and Lake Aral. None are known to have gone beyond the boundaries of the northern zone. If the lampreys are justly reckoned among freshwater fishes, their distribution is unique and exceptional. In the Palaearctic region some of the species descend periodically to the sea, whilst others remain stationary in the rivers; the same has been observed in the lampreys of North America. They are entirely absent in the equatorial zone, but reappear in the temperate zone of the southern hemisphere. Many points in the organization of the Cyclostomes indicate that they are a type of great antiquity.

The remaining Palaearctic fishes are clearly immigrants from neighbouring regions: thus Silurus, Afacrones, and Pseudobagrus have migrated from the Indian region, Amiurus and, as mentioned above, Catostomus from North Am rica. The Cyprinodonts are restricted to the southern and warmer parts, and all belong to the carnivorous division. The facility with which these fishes accommodate themselves to a sojourn in fresh, brackish, or salt water, and even in thermal springs, renders their general distribution easily comprehensible, but it is impossible to decide to which region they originally belonged; their remains in Tertiary deposits round the Mediterranean are not rare.

The boundaries of the North American or Nearctic Region have been sufficiently indicated. The main features and the distribution of this fauna are identical with those of the preceding region. Out of the 40 families of freshwater fishes 19 are found in this region. The proportion of Cyprinoid species to the total number of North American fishes (135: 339) appears to be considerably less than in the Palaearctic region, but we cannot admit that these figures approach the truth, as the Cyprinoids of North America have been much less studied than those of Europe; of many scarcely more than the name is known. This also applies in a great measure to the Salmonoids, of which only half as many as are found in the Palaearctic region have been sufficiently described to be worthy of consideration. North America will, without doubt, in the end show as many distinct races as Europe and Asia.

Cyprinoids belonging to living as well as extinct genera existed in North America in the Tertiary period. At present the Cyprinina, Leuciscina, and Abramidina are well represented, but there is no representative of the Old World genus Barbus, or of the Cobitidiηa; Rhodeina are also absent. On the other hand, a well marked Cyprinoid type is developed:—the Catostomina, of which one species has, as it were, returned into Asia. Very characteristic is the group of Centrarchina, allied to the perch, of which there are some thirty species; there are two Grystina. Of the sticklebacks there are as many species as in Europe, and of pike not less than seven species have been distinguished. Umbra appears to be as local as in Europe. Some very remarkable forms, types of distinct families, though represented by one or two species only, complete the number of North American autochthonous fishes, viz., Aphredoderus, Percopsis, Hyodon, and the Heteropygii (Amblyopsis and Chologaster). The last ar∙3 allied to the Cyprinodonts, differing from them in some points of the structure of their intestines. The two genera are extremely similar, but Chologaster, which is found in ditches in the rice-fields of South Carolina, is provided with eyes, and wants the ventral fins. Amblyopsis is the celebrated blind fish of the Mammoth Cave of Kentucky·, it is colourless and eyeless, and has rudimentary ventral fins, which occasionally may be entirely absent.

A peculiar feature of the North American fish fauna is that it has retained, besides the sturgeons and lampreys, representatives of two Ganoid families, Lepidosteus and Amia. Both these genera occur in Tertiary formations; whilst the former is represented in Europe as well as in North America, fossil remains of Amia have been found in the western hemisphere only.

It is difficult to account for the presence of the Amiu-ιiηα in North America. They form a well-marked division of the Bagrina, which are well represented in Africa and the East Indies, but are absent in South America; it is evident, therefore, that they should not be regarded as immigrants from the south, as is the case with the Palaearctic Siluroids. Nor again, has the connexion between South and North America been established sufficiently long to admit of the supposition that these Siluroids could have spread in the interval from the south to the northern parts of the continent, for some of the species are found as far north as Pine Islands Lake (54° N. lat.). Southern Zone. —The boundaries of this zone have been indicated in the description of the equatorial zone; they overlap the southern boundaries of the latter in South Australia and South America, but we have not at present the means of exactly defining the limits to which southern types extend northwards. This zone includes Tasmania, with at least a portion of south-eastern Australia (Tasmanian subregion), New Zealand and the Auckland Islands (New Zealand sub-region), and Chili, Patagonia, Tierra del Fuego, and the Falkland Islands (Fuegian sub-region). No freshwater fishes are known from Kerguelen’s Land, or from islands beyond 55° S. lat. The southern extremity of Africa has to be excluded from this zone so far as freshwater fishes are concerned.

With regard to its extent as well as to the number of species, this zone is the smallest of the three, the number of species known being 11 in the Tasmanian, 8 in the New Zealand, and 18 in the Fuegian sub-region. Yet the ichthyological features of this zone are well marked; they consist in the presence of two peculiar families, each of which is analogous to a northern type, viz., the Haplochi- to nidoe, which represent the Salmonidoe (Haplochitoη being the analogue of Salmo, and Prototroctes that of Coregonus), and the Galaxiidoe, which are the pikes of the southern hemisphere.

Although geographically widely separate from each other, the freshwater fishes of the three divisions are nevertheless so closely allied that conclusions drawn from this group of animals alone would hardly justify us in regarding these divisions as sub-regions. One species of Galaxias (G. attenuatus) and three of lampreys are found in all three, or at least in two of the sub regions. Percichthys is in Chili the autochthonous form of the cosmopolitan group of Perciηa. Diplomystax, an Arioid fish of Chili, and Hematogenys seem to have crossed the Andes from tropical America at a comparatively early period, as these genera are not represented on the eastern side of South America; Tmchomyc-teriηα occur on both sides of the Andes, which they ascend to a considerable height. Retropinna is a true Salmonoid, allied to the northern smelt (Osmerus), and representing it in the southern hemisphere. In both these genera part of the specimens live in the sea, and ascend rivers periodically to spawn; another part remain in rivers and lakes, where they propagate, never descending to the sea, this freshwater race being constantly smaller than their marine brethren. That this small Teleostean of the northern hemisphere should reappear, though in a generically modified form, in New Zealand, without having spread over other parts of the southern zone, is one of the most remarkable and at present inexplicable facts of the geographical distribution of freshwater fishes.

Brackish-Water Fishes.

On those parts of a coast at which there is a mixture of fresh and salt water, either in consequence of some river emptying itself into the sea, or from accumulations on the land-surface forming lagoons which are in uninterrupted or temporary communication with the sea, there flourishes a peculiar brackish-water fauna characterized by the presence of fishes found sometimes in the sea, and sometimes in pure fresh water.

This fauna can be somewhat sharply defined if a limited district only is taken into consideration; thus, the species of the brackish-water fauna of Great Britain, of the Pacific co ist of Central America, of the larger East India islands, <fcc., can be enumerated without much hesitation. But difficulties occur when we attempt to generalize in the enumeration of the forms referable to the brackish-water fauna, because the genera and families enumerated include certain species and genera which have habituated themselves exclusively either to a freshwater or a marine existence, and also because a species of fish may be at one locality an inhabitant of brackish water, at another of the sea, and at a third of fresh water. The circumstance that these fishes can live either in the sea or in fresh water has enabled them to spread readily over the globe, a few only being limited to particular regions; in dividing the earth’s surface into natural zoological regions, therefore, the taxonomist receives no assistance from the brackish-water f >rms. The following fishes may be referred to this fauna :—

1. Species oîRaiidoe (Raia, Trygon). 2. Arnbassis. 3. Τherapon. 4. Numerous Sciaenidoe of the equatorial zone. 5. Polynemidae. 6. Numerous species of Caranx (or horse mackerels) of the equatorial · zone. 7. Species of Gastrosteus. 8. The most important genera of the gobies (Gobiιna); Gobius (nearly cosmopolitan), Sicydium, Boleophthalmus, Pcriophlhalmus, Z‰√ns (equatorial). 9. Amblyo-V na 1θ∙ Tryρauchenina. 11. Many species of Blcnnius. 12. The majority of Athcrinidce. 13. Most Mugilidoe. 14. Many Pleuro- ncclidae. 15. Several Siluridoe, as especially the genera Plofosus, C ∣ ιidoglanis, Arius. 16. Many Cyprinodontidoe. 17. Species of Clupea. 18. Chatoessus. 19. Mejalops. 20. Anguilla (eels). 21. Numerous Synqnathidce.

This list could be considerably increased if an enumeration of species, especially of certain localities, were attempted; but this is more a subject of local interest, and would carry us beyond the scope of a general account of the distribution of fishes.

Marine Fishes.

Marine fishes fall, with regard to their mode of life and distribution, into three distinct categories :—

1. Shore Fishes— that is, fishes which chiefly inhabit parts of the sea in the immediate neighbourhood of land either actually raised above, or at least but little submerged below, the surface of the water. They do not descend to any great depth,—very few to 300 fathoms, and the majority live close to the surface. The distribution of these fishes is determined, not only by the temperature of the surface water, but also by the nature of the adjacent land and its animal and vegetable products,— some being confined to flat coasts with soft or sandy bottoms, others to rocky and fissured coasts, others to living coral formations. If it were not for the frequent mechanical and involuntary removals to which these fishes are exposed, their distribution within certain limits, as it no doubt originally existed, would resemble still more that of freshwater fishes than we find it actually does at the present period.

2. Pelagic Fishes- —that is, fishes which inhabit the surface and uppermost strata of the open ocean, and approach the shores only accidentally, or occasionally (insearch of prey), or periodically (for the purpose of spawning). The majority spawn in the opsn sea, their ova and young being always found at a great distance from the shore. With regard to their distribution, they are still subject to the influences of light and the temperature of the surface water; but they are independent of the variable local conditions which tie the shore fish to its original home, and therefore roam freely over a space which would take a freshwater or shore fish thousands of years to cover in its gradual dispersal. Such as are devoid of rapidity of motion are dispersed over similarly large areas by the oceanic currents, more slowly than the strong swimmers, but not less surely. An accurate definition, therefore, of their distribution within certain areas equivalent to the terrestrial regions is much less feasible than in the case of shore fishes.

3. Deep-Sea Fishes— that is, fishes which inhabit such depths of the ocean that they are but little or not at all influenced by light or the surface temperature, and which, by their organization, are prevented from reaching the surface stratum in a healthy condition. Living almost under identical tellurian conditions, the same type, the same species, may inhabit an abyssal depth under the equator as well as one near the arctic or antarctic circle; and all that we know of these fishes points to the conclusion that no separate horizontal regions can be distinguished in the abyssal fauna, and that no division into bathymetrical strata can be attempted on the base of generic much less of family characters.

Chondropterygii, Acanthoρterygiι, Anacanths, Myxi∙ noids, and Pharyngol)ranchii furnish the principal contingents to the marine fauna; whilst the majority of Physostomes, the Ganoids, and Cyclυstomes are freshwater fishe⅛

Shore Fishes.

The principal types of shore fishes are the following :— Chondropterygii.— Holocephαlα, 4 species. Ρlagiostornata— Carchariidae (part), 12; Scylliidae, 30; Cestraeiontidae, 4; Spinacid® (part), 8; Rhinidae, 1; Pristiophoridae, 4; Pristidae, 5; Rhinobatidae, 14; Torpedinidae, 15; Raiidre, 34; Trygonidae, 47.

Acanthoptebygii. —Percidae (part, includ. Pristipomatidae), 625 species; Mullidae, 35; Sparidae, 130; Squamipinnes, 130; Cir-rhitidae, 40; Heterolepidina, 12; Scorpaenidae, 120; Cottidae (part), 100; Cataphracti (part), 20; Trachinidae, 100; Sciaenidae, 100; Sphyraenidae, 15; Triehiuridae, 17; Elacate, 1; Nomeidae (part), 5; Cyttidae, 8; Stromateus, 9; Mene, 1; Carangidae (part), 130; Kurtidae, 7 U Gobiodon, 7; Callionymina, 30; Discoboli, 11; Batrachidae, 14; Pediculati (part), 11; Blenniidae, 90; Acantho-clinidae, 1; Teuthididae, 30; Acronuridae, 60; Hoplognathidae, 3; Malacanthidae, 3; Plesiopina, 4; Trichonotidae, 2; Cepolidae, 7; Gobiesocidae, 21; Psychrolutidae, 2; Centriscidae, 7; Fistulariidae, 4.

Acanthopterygii Pharyngognathi. — Pomacentridae, 150 species; Labridae, 400; Embiotocidae, 17.

Anacanthini.— Gadopsidae, 1 species; Lycodidae, 15; Gadidae (part), 50; Ophidiidae (part), 40; Pleuronectidae, 160.

Physostomi. —Saurina (part), 16 species; Salmonidae (part), 7; Clupeidae (part), 130; Chirocentridae, 1; Chilobranchus, 1; Muraenidae (part), 200; Pegasidae, 4.

Lophobranchii, 120 species.

Plectognathi, 178 species.

Cyclostomata. — Myxinidoe, 5 species.

Leptocardπ, 2 species.

Total number, 3587 species.

These types of shore fishes are distributed over the following oceanic areas :—

1. The Arctic Ocean.

II. The Northern Temperate Zone.

A. The Temperate North Atlantic.

1. The British district.

2. The Mediterranean district.

3. The North American district.

B. The Temperate North Pacific.

1. The Kamtchatkan district.

2. The Japanese district.

3. The Californian district.

III. The Equatorial Zone.

A. The Tropical Atlantic.

B. The Tropical Indo-Pacific.

C The Pacific coast of Tropical America.

1. The Central American district

2. The Galapagos district.

3. The Peruvian district.

IV The Southern Temperate Zone.

1. The Cape of Good Hope district.

2. The South Australian district.

3. The Chilian district.

4. The Patagonian district.

V. The Antarctic Ocean.

As with freshwater fishes, the main divisions of the shore-fish faunae are determined by their distance from the equator, the equatorial zone of the freshwater series corresponding entirely to that of the shore-fish series. But as marine fishes extend farther towards the poles than freshwater fishes, and as the polar types are more specialized, a distinct arctic and antarctic fauna may be separated from the faunae of the temperate zones. The two subdivisions of the northern temperate zone in the freshwater series are quite analogous to the corresponding divisions in the coast s ries. Tn the southern hemisphere the shore fishes of the extremity of Africa form a separate district of the temperate zone, whilst the freshwater fishes of South Africa were found to be tropical types. The marine series of the southern temperate zone is also much more diversified than the freshwater series, and admits of further subdivision, which, although in some degree indicated in the freshwater series, does not entirely correspond to that proposed for the latter.

Arctic Ocean. —The shore fishes clearly prove a continuity of the arctic circumpolar fauna, as the southern limit of which we may indicate the southern extremity of Greenland and the Aleutian Archipelago, or 60° N. lat.

Towards the north, fishes become less in variety of species and fewer in number of individuals, and only very few genera are restricted to this fauna.

The highest latitude at which shore fishes have been observed is 83° N. The late Arctic Expedition collected at and near that latitude specimens of Cottus quadricarms, Icelus hamatus, Cyclopterus spinosus, Liparis fabmcii, Gymnelis viridis, and Gadus fabricü. The number would probably have been larger were it not that the difficulties of collecting fishes in these high latitudes are almost insuperable for the greater part of the year.

So far as we know, the fishes north and south of Behring’s Straits belong to the same generic or family types as those of the corresponding latitudes of the eastern hemisphere, though the majority are specifically distinct. But the information we possess of the fishes of the northernmost extremity of the Pacific is extremely scanty and vague. Farther south, whence now and then a collection reaches Europe, we meet with some European species, as the herring, holibut, and hake.

Chondropterygians are very rare. Of Acanthopterygians the families of Cottidae, Cataphracti, Discoboli, and Blenniidae are well represented, and several of the genera are characteristic of the arctic fauna. Characteristic also is the development of Gadoid fishes, of which some thirteen species, belonging to Gadus, Merluccius, and Hοlva, form one of the principal articles of food for the inhabitants of the coasts of the Arctic Ocean. The Blen-noid Anacanthini or Lycodidae are limited to the Arctic and Antarctic coasts. Ammodytes and a few flat-fishes (Hippoglossoides and Pleuronectes) are common in the more temperate parts. Labroids only exceptionally penetrate so far towards the north. Physostomes are very rare, and are represented only by a few species of Clupea and by Mallotus. The arctic climate is still less favourable to the existence of Lophobranchs, only a few of Syηgηa-thus and Nerophis being present in the more southern latitudes, to which they have been carried by oceanic currents from their more congenial home in the south. Scleroderms and Plectognaths are entirely absent. The Gadoids are accompanied by Myxine, which thrives in them as a parasite.

Northern Temperate Zone.— Temperate North Atlantic.— This part of the fauna may be subdivided into three districts :—

1. The fishes of the north-eastern shores, viz., of the British Islands, of Scandinavia so far as it is not included in the arctic fauna, and of the continent of Europe southwards to about 40° N. lat.: British district.

2. The fishes of the Mediterranean shores and of the adjoining shores of the Atlantic, including the Azores, Madeira, and the Canary Islands: Mediterranean district.

3. The fishes of the western shores, from 60° to about 30° N. lat.: North American district.

1. The British district shows scarcely any marked distinctive features; the character of its fauna is simply intermediate between that of the Arctic Ocean and the Mediterranean district; truly arctic forms disappear, while such as are also found in the Mediterranean make their appearance. With regard to the abundance of individuals and variety of fishes also, this district forms a transition from the north towards the south.

Besides the few arctic Chondropterygians, all of which extend into this district, the small shore dog-fishes are well represented (Mustelus, Gαleus, Scyllium, Pristiurus); the ubiquitous Rhina or monk-fish is common; of rays, Raia predominates in a variety of species over Torpedo and Trygon, which are still rare.

Of Acanthopterygians, Centridermichthys, Icelus, Triglops, and Aspidophoroides do not extend from the north into this district; and Cottus, Anarrhichas, Centronotus, Stichceus, and lhe Discoboli disappear within its limits. Nearly all the remainder are genera which are also found in the Mediterranean district. The following are the principal forms known to propagate on these shores: Ldbrax; Serranus, Ρolyprion, Dentex ; Mullus ; Cantharus, Bagrus, Pagellus; Sebastcs; Cottus, Trigla, Agonus·, Trachinus-, Scicenaf); Zeus ; Trachurns, Capros; Callionymus; Discoboli; Lophius; Anarrhichas, Centronotus, Stichaeus ; Blenniops, Zoarces (not in the Mediterranean); Ceρola ; Lepadogaster.

Of the Anacanthini the Gadoids are. as numerous as in the Arctic Ocean, most being common to both districts; but, whilst the majority show their northern origin by not extending into the Mediterranean, Ammodytes and most Pleiironectidoe prove themselves to be the more southern representatives of this order. In the British district we find Ηippoglossus, Hippoglossoides, Rhombus, Phrynorhombus, Pleuronectes, Solea, and only the first two are not met with in the Mediterranean.

Labroids are common; with the exception of the North American Tautoga, all the other genera are met with.

Physostomes are not well represented, viz., by one species of Qsmerus, one of Engraulis, one of Conger, and about five of Clupea. Syngnathus and Nerophis become more common as we proceed southwards; but the existence of Scleroderms and Plectognaths is indicated by single individuals only, stragglers from their southern home, unable to establish themselves in a climate ungenial to them.

The Gadoids are accompanied by Myxine ; and Branchiostoma may be found in all suitable localities.

2. The Mediterranean district is distinguished by a great variety of forms; yet, with the exception of a few genera established for single species, none of the forms can be considered peculiar to it; ana even that small number of peculiar genera is more and more diminished as our knowledge of the distribution of fishes advances. Some genera are identical with those found on the western coasts of the Atlantic and in the West Indies; but a most remarkable and unexpected affinity obtains with another very distant fauna, viz., that of Japan. The number of genera common to the Mediterranean district and the Japanese coasts is larger than that of the genera common to the Mediterranean and the opposite American coasts.

The Chondropterygians found in the British district continue in the Mediterranean, their number being increased by Centrina, Spinax, Pteroplatea, and some species of Rhinobatus, a genus more numerously represented in the tropics. Torpedo and Trygon are common.

The greatest variety belong to the Acanthopterygians, which are represented by fifty-seven genera.

The Labridae are as common as in the British district, or even more so, and are represented by the same genera. But, besides these, some other Pharyngognaths, properly belonging to the tropical Atlantic, have fully established themselves, though only by a few species, viz., Glyphidodon and HeUastes, Cossyphus, Novacula, Julis, Coris, and Scarus.

The Gadoids show a marked decrease of development; and the species of Gadus, Gadiculus, Mora, Slrinsia, Phycis, and Molva which are peculiar to the Mediterranean seem to inhabit rather the tolder water of moderate depths than the surface near the shore. Motella, however, proves also to be a true shore fish in the Mediterranean, at least in its adult state. Ophidium and Fierasfer appear now besides Ammodytes. As the Gadoids decrease, so the Pleuro~ nectidoe increase, 9 genera being found in this district.

The variety of Physostomes is small, the following only having to be added to those of the British district: Sauras (a tropical genus), Aulopus; Congromuroena, Heteroconger, Myrus, Ophichthys, Maraena.

The Lophobranchs are more numerous in species and individuals than in the British district; and, besides Syngnathus and Nerophis, several species of Hippocampus are common. A few species of Balistes also occur.

Myxine is lost in this district, whilst Branchiostoma is abundant.

3. The shore fishes of the North American district consist, as on the eastern coasts of the North Atlantic, of northern (about 50 general and southern (about 30) elements; but they are still more mixe<l with each other than on the European coasts, so that a boundary line cannot be drawn between them. The affinity to the fauna of the eastern shores is great, but almost entirely limited to the genera composing the fauna of the British district. British genera not found on the American coasts are— Galeus, Scyllium, Chimaera, Mullus, Pagellus, Trigla, Trachinus, Zeus, Callionymus. The southern elements of North America are rather derived from the West Indies, and have no special affinity to Mediterranean forms; very few of the non-British Mediterranean forms extend across the Atlantic; instead of a Mediterranean we find a West Indian element. Many of the British species range across the Atlantic, and inhabit in an unchanged condition the northern parts of this district; and from the frequent occurrence of isolated specimens of other British species on the North American coast, we may presume that many more occasionally cross the Atlantic, but without being able to obtain a permanent footing.

The genera peculiar to this district are few in number, and composed of very few species, viz., Hemitripterus. Pammclas, Chasmodes, Cryptacanthodes, and Tautoga. Temperate North Pacific.— This fauna shows a great affinity to that of the Temperate North Atlantic, not only in including a considerable proportion of identical genera, and even of species, but also in having its constituent parts similarly distributed. Our knowledge of the ichthyology of this fauna, however, is by no means complete, and the details of the distribution of the fishes of these shores have •still to be worked out; nevertheless, three divisions may be recognized which, for the present, may be defined as follows :—

1. The fishes of the north-western shores, to about 37®

N. lat., including the corresponding northern parts of Japan: Kamtchatkan district. This corresponds to the British district of the Atlantic. ♦

2. The fishes of southern Japan and the corresponding shores of the continent of Asia, between 37° and 30 δ N. lat.: Japanese district, which corresponds to the Mediterranean.

3. The fishes of the eastern shores southwards to the latitude of San Francisco: Californian district. This corresponds to the North American district of the Atlantic.

Too little is known of the shore fishes of the coasts between San Francisco and the tropic to enable us to treat of it as a separate division.

The shore fishes of the North Pacific generally are composed of the following elements :—

a. Arctic forms which extend into the Arctic Ocean, and the majority of which are also found in the British district.

l>. Peculiar forms limited to the North Pacific, like the Heteroleρidina, Ernbiotocidoe, and certain Cottoid and Bien-nioid genera.

c. Forms identical with fishes of the Mediterranean.

<Z. Peculiar forms limited to the southern parts of Japan.

e. Tropical forms which have entered the North Pacific from the south.

1. To the Kamtchatkan district we can assign but a small list of fishes, probably because of the imperfect manner in which its fauna has been explored. At present we have positive knowledge of the occurrence of only two Chondropterygians, viz., Chimaera and Raia; the species of the latter genus seem to be much less numerous than in the Atlantic. Of Acanthopterygians 15 genera are known. Labroids are absent; they are clearly a type unable to endure great cold; of the Embiotocoids which represent them in the Pacific, one species only (a species of Ditrema) is known from this district.

The Gadoids are, so far as we know at present, sparsely represented, viz., by isolated species of Gadus, Motella, and Lotella, the last being an inhabitant of moderate depths rather than of the surface. Hippoglossus, Pleuronectes, and Parophrys seem to occur everywhere in suitable localities.

The Physostomes are nearly the same as in the British district, viz., a smelt {Hypomesus), probably also the arctic Mallotus, an anchovy, several species of Clupea, and the conger-eel. A very singular Salmonoid fish, Salanx, which is limited to the northwestern Pacific, occurs in great abundance. The Lophobranchs also correspond in their development to those of the British district, Nerophis being replaced by Urocampus. Neither Myxinoids nor Branchiostoma have as yet been found.

2. The Japanese district is, like the Mediterranean, distinguished by a great variety of forms. Of 102 genera known to inhabit these coasts, 13 are peculiar to it, 53 occur in the Mediterranean, though also in other districts. This resemblance to the Mediterranean is even greater than would appear from a comparison of the genera, inasmuch as a considerable number of species are identical in both districts. Three of the Berycoid genera have hitherto been found in the Japanese and Mediterranean districts and nowhere else. Another very singular fact is that some of the most characteristic genera, like Mullus, Zeus, Callionymus, Centrismes, inhabit the Mediterranean and Japanese districts, but have never reached the opposite American coasts, either in the Atlantic or the Pacific; although, at least in the latter, the oceanic currents would rather favour than obstruct their dispersal in the direction of America. Bold as the hypothesis may appear, we can only account for the singular distribution of these shore fishes by assuming that the Mediterranean and Japanese seas were in direct and open communication with each other within the period of the existence of the present Teleosteous fauna.

Gadoids have disappeared, or are represented by forms inhabiting moderate depths. Neither Myxine nor Branchiostoma are as yet known to have been found.

3. The Californian district includes a marked northern element, the principal constituents of which are identical with types occurring in the corresponding district of the Atlantic, viz., the North American, as exemplified by Discoboli, Anarrhichas, Ccntronotus t Coitus, Hippoglossus, Clupea (harengus), &c. But it possesses also, in the greatest degree of development, some types almost peculiar to itself, as the Heterolejridina, some remarkable Cottoid and Blennioid genera, and more especially the Embiotocoids—viviparous Pharyngognaths—which replace the Labroids of the other herni-sphere. Gadoids are much less numerous than in the North American district. The southern forms are but little known, but it may be anticipated that, owing to the partial identity of the faunae of the two coasts of the isthmus of Panama, a fair proportion of West Indian forms will be found to have entered this district from the south.

Equatorial Zone. —As we approach the tropic from the north, the types characteristic of the arctic and temperate zones become rarer, and disappear altogether, to be replaced by the greater variety of tropical types. Of Chondropterygians, the Chimoeridoe, Spinacidoe, Alustelus, and Raia do not pass the tropic, or appear in single species only; and of Teleosteans, the Berycidoe, Pagrus, lh,eHetero-lepidiηa, Cott us and allied genera, Lophius, Anarrhichas, Stichoeus, Lepadogaster, Psychrolutes, Centriscus, Hotacan-ihus, the Labridoe and Embiotocidoe, the Lycodidoe, Gadidoe, and marine Salmonidoe either entirely disappear, or retire from the shores and surface into the depths of the ocean.

With regard to variety of forms, as well as to number of individuals, this zone far surpasses either of the temperate zones; in this respect, the life in the sea is like that on the land. Shore fishes are not confined to the actual coast line, but abound on the coral reefs with which some parts of the Atlantic and Pacific are studded, and many of which are below the surface of the water. The abundance of animal and vegetable life which flourishes on these renders them the favourite pasture grounds for the endless va∣iety of coralfishes (Squamipinnes, Acronuridoe, Pomacentridoe, Julidoe, Plectognathi, <kc.) and for the larger predatory kinds. The colours and grotesque forms of the fishes of the tropics justly excited the admiration of the earliest observers. Scarlet, black, blue, pink, red, yellow, &c., are arranged in patterns of the most bizarre fashion, mingling in spots, lines, or bands, and reminding us of the words of Captain Cook when describing the coral-reefs of Palmerston Island: “The glowing appearance of the mollusks was still inferior to that of the multitude of fishes that glided gently along, seemingly with the most perfect security. The colours of the different sorts were the most beautiful that can be imagined—the yellow, blue, red, black, &c., far exceeding anything that art can produce. Their various forms, also, contributed to increase the richness of this submarine grotto, which could not be surveyed without a pleasing transport.”

Of Chondropterygians the Scylliidoe, Pristis (saw-fishes), Rhinobatidoe, and Trygonidoe attain to the greatest development. Of Acanthopterygians Centropristis, Serranus, Plectropoma, Alesoprion, Priacaηthus, Apogon, Pristipoma, Haemulon, Diagramma, Gerres, Scolopsis, Syιιagris, Casio, Mullidα, Lethrinus, Squamipinnes, Cirrhites, some genera of Scorpoenidoe, Platycephalus, Scioenidoe, Sphyroema, Caranx, Equula, Callionymms, Teuthis, Acanthurus, Nα-seus, are represented by numerous species; and the majority of these genera and families are limited to this zone. Of Pharyngognaths the Pomacentridoe, Julidina, and Scarina are met with near every coral formation in a living condition. Of Gadoids, a singular minute form, Bregmaceros, is almost the only representative, the other forms belonging to deep water, and rarely ascending to the surface. Flatfishes (Pleuronectidae) are common on sandy coasts, and the majority of the genera are peculiar to the tropics. Of Physostomi only the Saurina, Clupeidoe, and Muroenidoe are represented, the Clupeidoe being exceedingly numerous in individuals, whilst the Muroenidoe live more isolated, but show a still greater variety of species Lophobranchii and Sclerodermi are generally distributed Branchiostoma has been found on several coasts.

Geographically it is convenient to describe the coast fauna of the tropical Atlantic separately from that of the 1ndυ-Pacific Ocean. The differences between them, however, are far less numerous and important than between the freshwater or terrestrial faunae of continental regions. The majority of the principal types are found in both, many of the species being even identical; but the species are far more abundant in the Indo-Pacific than in the Atlantic, owing to the greater extent of the archipelagoes in the former. But for the broken and varied character of the coasts of the West Indies, the shores of the tropical Atlantic would, by their general uniformity, afford but a limited variety of conditions for the development of specific and generic forms, whereas the deep inlets of the Indian Ocean, with the varying configuration of their coasts and the different nature of the bottom, its long peninsulas, and its archipelagoes, and the scattered islands of the tropical Pacific, render this part of the globe the most perfect for the development of fish life. The fishes of the Indian and Pacific Oceans (between the tropics) are almost identical, and the number of species ranging from the Red Sea and the east coast of Africa to Polynesia, even to its most westerly islands, is very great indeed. This Indo-Pacific fauna, however, does not reach the Pacific coast of South America. The wide space devoid of islands east of the Sandwich Islands and the Marquesas group, together with the current of cold water which sweeps northwards along the South American coast, has proved to be a very effectual barrier to the eastward extension of the Indo-Pacific fauna of coast fishes; and, consequently, we find an assemblage of fishes on the American coast and at the Galapagos Islands, sufficiently marked to constitute a distinct zoological division.

Tropical Atlantic.— The boundaries of the tropical Atlantic extend zoologically a few degrees beyond the northern and southern tropics, but, as the mixture with the types of the temperate zone is very gradual, no distinct boundary line can be drawn between the tropical and temperate faunae.

Types almost exclusively limited to it and not found in the Indo-Pacific are few in number, as Centropristis, Rhypticus, Haemulon, Malthe. A few others preponderate with regard to the number of species, as Plectropoma, Sargus, Τrachyηοtus, Batrachidoe, and Gobiesocidoe. The Sciaenoids are equally represented in both oceans. All the remainder are found in both, but are in a minority in the Atlantic, where they are sometimes represented by one or two species only (for instance, Lethrinus'). Tropical Indo-Pacific Ocean.— The ichthyological boundaries of this part of the tropical zone may be approximately given at 30° N. and S. lat.; on the Australian coasts it should probably be placed still farther south, viz., at 34°; it includes, as mentioned above, the Sandwich Islands, and all the islands of the South Sea, but not the American coasts.

Some eighty genera of shore fishes are peculiar to the Indo-Pacific, but the greater number consist of one or a few species only; comparatively few have a plurality of species, as Diagramma, Lethrinus, Equulα, Teuthis, Amjriiprion, Dascyllus, Chαroρs, Chilinus, A nampses, Stethojulis, Coris, Coilia.

The sea-perches, large and small, which feed on crustaceans and on other small fishes,and the coral feeding Pharyngognaths, are the types which show the greatest generic and specific variety in the Indo-Pacific. Then follow the Squamipinnes and Muroenidoe, the Clupeidoe and Caran- gidoe,· —families in which the variety is more that of species than of genus. The Scorpoenidoe, Pleiironectidoe, Acronuridoe, Scioenidoe, Syngnathidoe, and Teuthyes are those which contribute the next largest contingents. Of shoreloving Chondropterygians the Scylliidoe and Trygonidae only are represented in moderate numbers, though they are more numerous in this ocean than in any other.

Pacific Coasts of Tropical America.— As boundaries within which this fauna is comprised may be indicated 30° N. and S. lat., as in the Indo-Pacific. Its distinction from the Indo-Pacific lies in the almost entire absence of coralfeeding fishes. There are scarcely any Squamipinnes, Pharyngognaths, or Acronurid<je, and the Teuthyes are entirely absent. The genera that remain are such as are found in the tropical zone generally, but the species are entirely different from those of the Indo-Pacific. They are mixed with a sprinkling of peculiar genera, consisting of one or two species, like Discopyge, Hoplopagrus, Doydixodon, but they are too few in number to give a strikingly peculiar character to this fauna.

Three districts are distinguishable, viz., the Central American, the Galapagos, and Peruvian. We add a few remarks on the first only. That district shows so close an affinity to the Tropical Atlantic that, if the two were not separated by the neck of land uniting North and South America, they would most assuredly be regarded as embracing a single fauna. With scarcely any exceptions the genera are identical, and of the species found on the Pacific side nearly one-half have proved to be’the same as those of the Atlantic. The explanation of this fact has been found in the existence of communications between the two oceans by channels and straits which must have been open till within a recent period. The isthmus of Central America was then partially submerged, and appeared as a chain of islands similar to that of the Antilles; but as the reef-building corals flourished chiefly north and east of those islands, and were absent to the south and west of them, reef fishes were excluded from the Pacific shores when the communications were destroyed by the upheaval of the land.

Southern Temperate Zone. —This zone includes the coasts of the southern extremity of Africa from about 30° S. lat, of the south of Australia, with Tasmania, and of New Zealand, and the Pacific and Atlantic coasts of South America between 30° and 50° S lat.

The most striking character of this fauna is the reappearance of types inhabiting the corresponding latitudes of the northern hemisphere, and not found in the intervening tropical zone. This interruption of the continuity in the geographical distribution of shore fishes is exemplified by species as well as genera, for instance— Chimoera monstrosa, Galeus canis, Acaηthias vulgaris, Acanthιas blainvillii, Rhina squatina, Zeus faber, Lophhus piscatorius, Centmscus scolopax, Engraulis encrasiclιolus, Clupea sprattus, Conger vulgaris. Instances of genera are still more numerous :— Cestrαcion, Sρinax, Pristiophorus, Raia; Callanthias, Polyprion, Histiopterus, Canthams, Box, Girella, Pagellas, Chilodactylus, Sebastβs, Aploactis, Agonus, Lepidopus, Cyttus, Psychrolutidoe, Notacanthus; Lycodes, Merluccius, Lotella, Phycis, Motella · A ulopus; Urocampus, Solenognathus , Myxine.

Naturally, where the coasts of the tropical zone are continuous with those of the temperate, a number of tropical genera enter the latter, and genera which we have found between the tropics, as well as in the temperate zone of the northern hemisphere, extend in a similar manner towards the south. But the truly tropical forms are absent; there are no Squamipinnes, scarcely any Mullidce, no Acronuri, no Teuthyes, no Pomacentridoe (with a single exception on the coast of Chili), only one genus of Julidiηa, no Scarina, which are replaced by another group of Pharyngognaths, the Odacina. The Labrina, so characteristic of the temperate zone of the northern hemisphere, reappear in a distinct genus {Malαcopterus) on the coast of Juan Fernandez.

The family of Berycidοe, equally interesting with regard to their distribution in time and in space, consists of temperate and tropical genera. The genus by which this family is represented in the southern temperate zone (Trachichthys) is much more nearly allied to the northern than to the tropical genera.

The true Cottina and Heterolepidina (forms with a bony stay of the praeoperculum, which is generally armed) have not crossed the tropical zone; they are replaced by fishes extremely similar in general form, and having the same habits, but lacking that osteolυgical peculiarity. Their southern analogues belong chiefly to the family Trachinidοe, and are types of genera peculiar to the southern hemisphere.

The Discoboli of the northern hemisphere likewise have not penetrated to the south, where they are represented by Gobiesocidoe. These two families replace each other in their distribution over the globe.

Nearly all the Pleuronectidoe (but they are not numerous) belong to distinct genera, some, however, being remarkably similar in general form to the northern Pleuro-ηectes.

With Gadoids Myxinidoe reappear, one species being extremely similar to the European Myxine. Bdellostorna is a genus peculiar to the southern temperate zone.

As in the northern temperate zone, so in the southern the number of individuals and the variety of forms is much less than between the tropics. This is especially apparent on comparing the numbers of species constituting a genus. In this zone genera composed of more than ten species are the exception, the majority having only from one to five.

The proportion of genera limited to this zone is very high, about 65 out of a total number of 170 being peculiar to it

1. The Cape of Good Hope district. Many of the genera found in this district are northern forms {Chimoera, Galeus, Scylliurn, Acaηthias, Τοrρedο, Raia, Dentex, Cantharus, Box, Sagmis, Ρagrus, Pagellus, Chrysophrys; Sebastes, Sphyroena ; Lepidopus, Thyrsites; Zeus; Lophius; Motella), which in conjunction with the peculiarly southern types {Callorhynehus, Chilodactylus, Agriopus, Clinus, Genypterus, Bdellostorna) leave no doubt that this district belongs to the southern temperate zone, whilst the freshwater fishes of South Africa are members of the tropical fauna. Only a few {Rhinobatus, Narcine, Astrape, and Sphyroena) have entered from the neighbouring tropical coasts. The development of Sparoids is greater than in any of the other districts of this zone, and may be regarded as one of its distinguishing features.

2. The South Australian district comprises the southern coasts of Australia (northwards to about the latitude of Sydney), Tasmania, and New Zealand. It is the richest in the southern temperate zone, partly in consequence of a considerable influx of tropical forms on the eastern coast of Australia, where they penetrate farther southwards than would be expected from merely geographical considerations, partly in consequence of the thorough manner in which the ichthyology of New South Wales and New Zealand has been explored. Of the 120 genera hitherto found in this district 42 are peculiar.

The shore fishes of New Zealand are not so distinct from those of south-eastern Australia as to deserve to be placed in a separate district. With the exception of the genera which enter this zone from the tropics, and which are more numerous on the Australian coast than on that of New Zealand,, and a few very local genera in addition, the remainder are identical. Many of the South Australian species, too, are found also on the coasts of New Zealand. The principal points of difference are the extraordinary development of Monacanthus on the coast of South Australia, and the apparently total absence in Australia of Gadoids, which in the New Zealand fauna are represented by 6 genera.

3. The Chilian district extends over 20 degrees of latitude only, and is nearly straight. In its northern and warmer parts it is of a very uniform character; it is there exnosed to high and irregular tides, and to remarkable and sudden changes in the levels of land and water, which must seriously interfere with fishes living and propagating near the shore. No river of considerable size interrupts the monotony of the physical conditions, to offer an additional element in favour of the development of littoral animals. In the southern parts, where the coast is lined with archipelagoes, the climate is too severe for most fishes. All these conditions combine to render this district comparatively poor as regards variety of shop fishes; they belong to 45 genera, of which 6 only are not found in Other districts of this zone. Three are peculiar, viz., Mendosoma, Myxodes, and Malacopterus; Porichthys and Agonusha,v<i penetrated thus far southwards from the Peruvian and Californian districts; and Polyprion is one of those extraordinary instances in which a very specialized form occurs at almost opposite points of the globe, without having left a trace of its previous existence in, or of its passage through, the intermediate space.

4. Our knowledge of the fauna of the Patagonian district is, with the exception of the neighbourhood of the mouth of the Rio de la Plata, in too fragmentary a state to deserve further notice here.

Antarctic Ocean. —To this fauna we refer the shore fishes of the southernmost extremity of South America, from 50° S. lat., with Tierra del Fuego and the Falkland Islands, and those of Kerguelen’s Land, with Prince Edward’s Island. No fishes are known from the other Oceanic islands of these latitudes.

In the southern hemisphere surface fishes do not extend so far towards the pole as in the northern; none are known beyond 60° S. lat., and the Antarctic fauna which is analogous to the Arctic inhabits coasts more than ten degrees nearer to the equator. It is very probable that the shores between 60° and the Antarctic Circle are inhabited by fishes sufficiently numerous to supply part of the means of subsistence for the large seals which there pass at least some portion of the year, but hitherto none have been obtained by naturalists; all that the present state of our knowledge justifies us in saying is, that the general character of the fauna of Magellan’s Straits and Kerguelen’s Land is extremely similar to that of Iceland and Greenland.

As in the Arctic fauna, Chondropterygians are rare, and are represented by Acanthias vulgaris and species of Raia. Holocephali have not yet been found so far south, but Hallorhynchus, which is not uncommon near the northern boundary of this fauna, may prove to extend into it.

As to Acanthopterygians, Cataphracti and Scorpoenidae are represented as in the Arctic fauna, two of the genera (Sebastes and Agonus) being identical. The Cottidoe are replaced by six genera of Trachinidoe, remarkably similar in form to Arctic types; but Discoboli and the characteristic Arctic Blennioids are absent.

Gadoid fishes reappear, but are less developed; as usual they are accompanied by Myxine. The reappearance of so specialized a genus as Lycodes is most remarkable. Flatfishes are few as in the north, and belong to peculiar genera.

Physostomes are probably not entirely absent, but hitherto none have been met with so far south. Lophobranchs are rare, as in the Arctic zone; it is noteworthy, however, that a peculiar genus, with persistent embryonic characters {Protocampus), is rather common on the shores -of the Falldand Islands.

Pelagic Fishes.

Pelagic fishes,—that is, fishes inhabiting the surface of mid-ocean, —belong to various orders, viz., Chondropterygians, Acanthopterygians, Physostomes, Lophobranchs, and Plectognaths. Neither Anacanths nor Pharyngognaths contribute to this series of the marine fauna. The following genera and families are included in it :—

Chondroρterygii.— Carcharias, Galeocerdo, Thalassorhinus, Zygaena, Triaenodon, Lamnidae, Rhinodon, Notida-nidae, Laemargus, Euprotomicrus, Echinorhinus, Isistius; Myliobatidae.

Acanthopterygii.— Dactylopterus, Micropteryx, Scom-brina, Gastrochisma, Nom eus, Centrolophus, Coryphaenina, Seriola, Temnodon, Naucrates, Psenes, Xiphiidae, Antennarius.

Physostomi.— Stemoptychidae, Scopelus, Astronesthes, Scombresocidae (majority).

Lophobranchii,— Hippocampus.

Plectognathi.— Orthagoriscus, and some other Gymnodonts.

Pelagic fishes differ much from one another in their mode of life. The majority are excellent swimmers, which not only can move with great rapidity, but are also possessed of great powers of endurance, and are thus enabled to continue their course for weeks, apparently without the necessity of rest; such are many sharks, scombroids, dolphins, pilot-fish, sword-fishes. In some, as in Dactylopterus and Exoccetus, the ability to take flying leaps out of the water is superadded to the power of swimming (flying-fishes). But in others the power of swimming is greatly reduced, as in Antennarius, Hippocampus, and Gymnodonts; they frequent places in the ocean covered with floating seaweed, or drift on the surface without resistance, at the mercy of wind and current. The Echeneis or sucking-fishes attach themselves to other large fishes, ships, or floating objects, and allow themselves to be carried about, unless change of climate or want of food obliges them to abandon their temporary carrier. Finally, another class of pelagic fishes come to the surface of the ocean during the night only; in the day time they descend to some depth, where they are undisturbed by the rays of the sun or the agitation of the surface-water; such are Brama, the Stemoptychidae, Scopelus, Astronesthes,— fishes the majority of which are provided with those extraordinary visual organs that we find so much developed in the true deep-sea fishes. Indeed, this last kind of pelagic fishes constitutes a connecting link with the deep-sea forms.

Pelagic fishes, like shore fishéS, are most numerous in the tropical zone; and, with few exceptions {Echinorhinus, Psenes, Sternoptychidoe, Astronesthes), the same genera are represented in the tropical Atlantic as well as in the Indo-Pacific. The number of identical species occurring in both these oceans is great, and probably still greater than would appear from systematic lists, in which there are retained many specific names that were given at a time when species were believed to have a very limited range. The pelagic fauna of the tropics gradually passes into that of the temperate zones, only a few genera, like Cybium, Psenes, Antennarius, being almost entirely confined to the tropics. All the other tropical genera range into the temperate zones, but their representatives become fewer with the increasing distance from the equator. North of 40° N. lat. many genera have disappeared, or are met with in isolated examples only, as Carcharias, Zygaena, Notidanus, Myliobatidae, Dactylopterus, Echeneis, Nomeus, Coryphaena, Schedophilus, Seriola, Temnodon, Antennarius, Sternoptychidoe, Astronesthes, Exoccetus, Tetrodοn, Diodon; and only one genus of sharks, Galeocerdo, approaches the Arctic Circle. Some few species, like Antennarius, Scopelus, are carried by currents near to the farther confines of the temperate zones; but such occurrences are accidental, and these fishes must be regarded as entirely foreign to the fauna of those latitudes. On the other hand, some pelagic fishes inhabit the temperate zones, whilst their occurrence within the tropics is very problematical; thus, in the Atlantic, Thalassorhiηus, Selache, Laemargus, Centrolophus, Diana, Ausonia, Lampris (all genera composed of one or two species only). Besides the shark mentioned, no other pelagic fishes are known from the Arctic Ocean.

We possess very little information about the pelagic fishfauna of the southern oceans. This much only is certain, that the tropical forms gradually disappear; but it would be hazardous, in the present state of our knowledge, to state even approximately the limits of the southward range of a single genus. Scarcely more is known about the appearance of types peculiar to the southern temperate zone,—for instance, the gigantic shark Rhinodon representing the northern Selache, near the coasts of South Africa, and the Scombroid genus Gastrochisma, in the South Pacific.

The largest of marine fishes, Rhinodon, Selaehe, Car- charodon, Myliobatùloe, Thynnus, Xiphiidoe, 0 rthagoriscus, belong to the pelagic fauna. Young fishes are frequently found in mid-ocean, which are the offspring of shore fishes normally depositing their spawn near the coast. The manner in which this fry passes into the open sea is unknown; for it has not yet been ascertained whether it is carried by currents from the place where it was deposited originally, or whether shore fishes sometimes spawn at a -distance from the coast. We may remember that shore fishes inhabit not only coasts but also submerged banks with some depth of water above, and that, by the action of the water, spawn deposited on these latter localities is very liable to be dispersed over wide areas of the ocean. Embryos of at least some shore fishes hatched under abnormal conditions seem to have an abnormal growth up <to a certain period of their life, when they perish. The Leptocephali must be regarded as such abnormally developed forms. Fishes of a similar condition are the so-called pelagic Plagusiae, young Pleuronectoids, the origin of which is still unknown. As already mentioned, flat-fishes, like all the other Anacanths, are not otherwise represented in the pelagic fauna.

Deep-Sea Fishes.

The knowledge of the existence of deep-sea fishes is one of the recent discoveries of ichthyology. It was only about twenty years ago that, from the evidence afforded by the anatomical structure of a few singular fishes obtained in the North Atlantic, an opinion was expressed that these fishes inhabited great depths of the ocean, and that their organization was specially adapted for living under the physical abyssal conditions. These fishes agreed in the character of their connective tissue, which was so extremely weak as to yield to, and to break under, the slightest pressure, so that the greatest difficulty was experienced in preserving their body in its continuity. Another singular circumstance was that some of the examples were picked up floating on the surface of the water, having met their death whilst engaged in swallowing or digesting other fishes not much smaller in size if not actually larger than themselves.

The first peculiarity was accounted for by the fact that, if those fishes really inhabited the great depths supposed, their removal from the enormous pressure under which they lived would be accompanied by such an expansion of gases within their tissues as to rupture them, and to cause a separation of the parts which had been held together by the pressure. The second circumstance was explained thus. A raptorial fish organized to live at a depth of between 1500 and 800 fathoms seizes another usually inhabiting a depth of between 300 and 500 fathoms. In its struggles to escape, the fish seized, being nearly as large or strong as the attacking fish, carries the latter out of its depth into a higher stratum, where the diminished pressure causes such an expansion of gases as to make the destroyer with its victim rise with increasing rapidity towards the surface, which they reach dead or in a dying condition. Specimens in this state are not rarely picked up; and as, of course, it is but comparatively few that can by accident fall into the hands of naturalists, occurrences of this kind must happen very often.

The existence of fishes peculiarly adapted for the deep sea has thus been a fact maintained and admitted for some time in ichthyology; and as the same genera and species were found at very distant parts of the ocean, it was further •stated that those deep-sea fishes were not limited in their range, and that, consequently, the physical conditions of the depths of the ocean must be the same or nearly the same ever the whole globe. That deep-sea fishes were not of a peculiar order, but chiefly modified forms of surface types, was another conclusion arrived at from the sporadic evidence collected during the period which preceded systematic deep-sea dredging.

Nothing, however, was positively known as to the exact depths inhabited by those fishes until observations were made during the voyage of H.M.S. “Challenger.” The results obtained by this expedition afforded a surer and more extended basis for our knowledge of deep-sea fishes.

The physical conditions of the deep sea, which must affect the organization and distribution of fishes, are the folkw-ing

1. Absence of sunlight. Probably the rays of the sun do not penetrate to, and certainly do not extend beyond, a depth of 200 fathoms, therefore we may consider this to be the depth where the deep-sea fauna commences. Absence of light is, of necessity, accompanied by modifications of the organs of vision and by simplification of colours.

2. The absence of sunlight is in some measure compensated by the presence of phosphorescent light, produced by many marine animals, and also by numerous deep-sea fishes.

3. Depression and equality of the temperature. At a depth of 500 fathoms the temperature of the water is already as low as 40° Fahr., and perfectly independent of the temperature of the surface-water; and from the greatest depth to about 1000 fathoms beneath the surface the temperature is uniformly but a few degrees above the freezing point. Temperature, therefore, ceases to offer an obstacle to the unlimited dispersal of the deep-sea fishes.

4. The increase of pressure by the water. The pressure of the atmosphere on the body of an animal at the level of the sea is 15 δ) per square inch of surface; but under water the pressure amounts to a ton weight for every 1000 fathoms of depth.

5. With the sunlight, vegetable life ceases in the depths of the sea. All deep-sea fishes are therefore carnivorous,— the most voracious feeding frequently on their own offspring, and the toothless kinds being nourished by the animalcules which live on the bottom, or which, “like a constant rain,” settle down from the upper strata towards the bottom of the sea.

6. The perfect quiet of the water at great depths. The agitation of the water caused by the disturbances of the air does not extend beyond the depth of a few fathoms; below this surface-stratum there is no movement except the quiet flow of ocean-currents, and near the bottom of the deep sea the water is probably in a state of almost entire quiescence.

The effect upon fishes of the physical conditions described is clearly testified by the modification of one or more parts of their organization, so that every deep-sea fish may be recognized as such without the accompanying positive evidence that it has been caught at a great depth; and, vice versa, fishes reputed to have been obtained at a great depth, and not having any of the characteristics of the dwellers of the deep sea, must be regarded as surface fishes.

The most striking characteristic found in many deep-sea fishes is in relation to the tremendous pressure under which they live. Their osseous and muscular systems are, as compared with the same parts of surface fishes, very feebly developed. The bones have a fibrous, fissured, and cavernous texture; they are light, with scarcely any calcareous matter, so that the point of a needle will readily penetrate them without breaking. The bones, especially the vertebrae, appear to be very loosely connected with one another; and it requires the most careful handling to avoid the breaking of the connective ligaments. The muscles, especially the great lateral muscles of the trunk and tail, are thin, the fascicles being readily separated from one another or torn, and the connective tissue being extremely loose, feeble, or apparently absent. This peculiarity has been observed in the Trachypteridae, Plagyodus, Chiasmodus, Melanocetus, Saccopharynx. But we cannot assume that it actually obtains whilst those fishes exist under their natural conditions. Some of them are most rapacious creatures, which must be able to execute rapid and powerful movements to catch and overpower their prey; and for that object their muscular system, thin as its layers may be, must be as firm, and the chain of the segments of their vertebral column as firmly linked together as in surface fishes. It is evident, therefore, that the change which the body of those fishes has undergone on their withdrawal from the pressure under which they live is a much aggravated form of the affection that is experienced by persons reaching great altitudes in their ascent of a mountain or in a balloon. In every living organism with an intestinal tract there are accumulations of free gases; and, moreover, the blood and other fluids, which penetrate every part of the body, contain gases in solution. Under greatly diminished pressure these gases expand, so that, if the withdrawal from a depth is not an extremely slow and gradual process, the various tissues must be distended, loosened, ruptured; and what is a vigorous fish at a depth of 500 fathoms or more appears at the surface as a loosely-jointed body which, if the skin is not of sufficient toughness, can only be kept together with difficulty. At great depths a fibrous osseous structure and a thin layer of muscles suffice to obtain the same results for which, at the surface, thickness of muscle and firm osseous or cartilaginous tissue are necessary.

The muciferous system of many deep-sea fishes is developed in an extraordinary degree. We find in fishes which are comparatively little removed from the surface (that is, to depths of from 100 to 200 fathoms) the lateral line much wider than in their congeners or nearest allies which live on the surface, as in Trαchichthys, Hoplostethus, many Scorpaenidoe. Butin fishes inhabiting depths of 1000 fathoms and more the whole muciferous system is dilated; it is especially the surface of the skull which is occupied by large cavities {Macriimdae, deep-sea Ophidiidae), and the whole body seems to be covered with a layer of mucus. These cavities collapse and shrink in specimens which have been preserved in spirit for some time, but a brief reimmersion in water generally suffices to show the immense quantity of mucus secreted by them. The physiological use of this secretion is unknown; it has been observed to have phosphorescent properties in perfectly fresh specimens.

The colours of deep-sea fishes are extremely simple, their bodies being either black or silvery; in a few only are some filaments or the fin-rays of a bright scarlet colour. Among the black forms albinoes are not rare.

The organ of sight is the first to be affected by a sojourn in deep water. Even in fishes which habitually live at a depth of only 80 fathoms, we find the eye of a proportionally larger size than in their representatives at the surface. In such fishes the eyes increase in size with the depth inhabited by them, down to the depth of 200 fathoms, the large organs being necessary to collect as many rays of light as possible. Beyond that depth small-eyed as well as large-eyed fishes occur, the former having their want of vision compensated by tentacular organs of touch, whilst the latter have no such accessory organs, and evidently see only by the aid of phosphorescence. In the greatest depths blind fishes occur, with rudimentary eyes and without special organs of touch.

Many fishes of the deep sea are provided with more or less numerous, round, shining, mother-of-pearl-coloured bodies, imbedded in the skin. These so-called phosphorescent or luminous organs are either larger bodies of an oval or irregularly elliptical shape placed on the head, in the vicinity of the eye, or smaller round globular bodies arranged symmetrically in series along the side of the body and tail, especially near the abdominal profile, less frequently along the back. The former kind of organs possess in the interior a lenticular body, like the lens of an eye, and are considered by some naturalists true organs of vision (accessory eyes), the function of the latter, which have a glandular structure, being left unexplained by them.

There is no doubt that the functions of these organs have some relation to the peculiar conditions of light under which the fishes provided with them live, these fishes being either deep-sea forms or nocturnal pelagic kinds. And it is highly probable that all produce and emit phosphorescent light, enabling the fishes to see in the darkness of the night or of the depths of the sea.

Whenever we find in a fish long delicate filaments developed in connexion with the fins or the extremity of the tail, we may conclude that it is an inhabitant of still water and uf quiet habits. Many deep-sea fishes {Trachypteridae, Macruridoe, Ophidiidae, Bathyptevois) are provided with such filamentous prolongations, the development of which is perfectly in accordance with their sojourn in the absolutely quiet waters of abyssal depths.

Some of the raptatorial deep-sea fishes have à stomach so distensible and capacious that it can receive a fish of twice or thrice the bulk of the destroyer (Melanocetus, Chiasmodus, Saccopharyηx). Deglutition is performed in them, not by means of the muscles of the pharynx, as in other fishes, but by the independent and alternate action of the jaws, as in snakes. These fishes cannot be said to swallow their food; they rather draw themselves over their victim, after the fashion of an Actinia.

Before the voyage of H.M.S. “Challenger,” scarcely thirty deep-sea fishes were known. This number is now much increased, six times as many new species and genera having been discovered. Modifications of certain organs, perfectly novel, and of the greatest inteιest, were found; but, singularly, no new types of families were discovered, —nothing but what might have been expected from our previous knowledge of this group of fishes.

The fish fauna of the deep sea is chiefly composed of forms or modifications of forms which we find represented at the surface in the cold and temperate zones, or which belong to the class of nocturnal pelagic fishes. The Chondropterygians are few in number, not descending to a greater depth than 600 fathoms. The Acanthopterygians, which form the majority of the coast and surface faunas, are also scantily represented; genera identical with surface types are confined to the same inconsiderable depth as the Chondropterygians, whilst those Acanthopterygians which are so much specialized for a life in the deep sea as to deserve generic separation range from 200 to 2400 fathoms. Three distinct families of Acanthopterygians belong to the deep-sea fauna, viz., Trachypteridae, Lophotidae, and Nota- canthidoe ; they consist of three, one, and two genera respectively.

Gadidae, Ophidiidae, and Macruridoe are very numerous, ranging through all depths; they constitute about onefourth of the whole deep-sea fauna.

Of Physostomi, the families of Sternoptychülae, Scopelidoe^ Storniatidae, Salmonidoe, Bathythrissidae, Alepoceph didoe, Halosauridoe, and Muraenidae are represented. Of these the Scopeloids are the most numerous, constituting nearly another fourth of the fauna. Salmonidoe are only represented by three small genera. Bathythmsmdoe include one species only, which is probably confined in its vertical as well as its horizontal range; it occurs at a depth of about 350 fathoms in the sea of Japan. The Alepoc-ephalidoe and Halosauridoe, known before the “Challenger” expedition from isolated examples only, prove to be true, widely-spread, deep-sea types. Eels are well represented, and seem to descend to the greatest depths. Myxine has been ob tained from a depth of 345 fathoms.

The greatest depth hitherto reached by a dredge in which fishes were enclosed is 2900 fathoms. But the specimens thus obtained belong to a species (Gonostoma microdon) which seems to be extremely abundant in upper strata of the Atlantic and Pacific, and were therefore most likely caught by the dredge in its ascent. The next greatest depth, viz., 2750 fathoms, must be accepted as one at which fishes do undoubtedly live,—the fish obtained from this depth of the Atlantic, Bathyophis ferox, showing by its whole habit that it is a form living on the bottom of the ocean.

CLASSIFICATION.

The class of fishes is divided into four subclasses :—

I. Paloeichthyes. — Heart with a contractile conus arteriosus; intestine with a spiral valve; optic nerves nondecussating, or only partially decussating.

II. Teleostei.— Heart with a non-contractile bulbus arteriosus; intestine without spiral valve; optic nerves decussating. Skeleton ossified, with completely separated vertebrae.

III. Cyclostomata.— Heart without bulbus arteriosus; intestine simple. Skeleton cartilaginous and notochordal. One nasal aperture only. No jaws; mouth surrounded by a circular lip.

IV. Leptocardii.— Heart replaced by pulsating sinuses; intestine simple. Skeleton membrano-cartilaginous and notochordal. Nobrain; no skull.

Subclass I.— Paloeichthyes.

This subclass comprises the sharks and rays and the Ganoid fishes. Though it is based upon a singular concurrence of most important characters, its members exhibit as great a diversity of form, and as manifold modifications in the remainder of their organization, as the Teleostei. The Paloeichthyes stand to the Teleostei in the same relation as the Marsupials to the Placentalia. Geologically, as a subclass, they were the predecessors of Teleosteous fishes; and it is a remarkable fact that all those modifications which show an approach of the ichthyic type to the Batrachians are found in this subclass. It is divided into two orders, —the Chondropterygii and the Gaηoidei. Order I.— Chondropterygii.

Skeleton cartilaginous. Body with medial and paired fins, the hinder pair abdominal. Vertebral column generally heterocercal, the upper lobe of the caudal fin produced. Gills attached to the skin by the outer margin, with several intervening gill-openings; re rely one external gill-opening only. No gill-cover. No air- 1 ladder. Two, three, or more series of valves in the conus arteriosus. Ova large and few in number, impregnated and, in some species, developed within a uterine cavity. Embryo with deciduous external gills. Males with intromittent organs attached to the ventral fins.

This order, for which, also, the name Elαsmobrαnchii has been proposed (by Bonaparte), comprises the sharks, rays, and chimaeras. It is divided into two suborders,— Ρlagiostomata and Holocephala. Suborder I. Plagiostomata.— From five to seven gill-openings. Skull with a suspensorium and the palatal apparatus detached. Teeth numerous. .

The Plagiostomes differ greatly from each other with regard to the general form of the body. In the sharks, or Selachoidei, the body is elongate, more or less cylindrical, gradually passing into the tail; the gill-openings arc lateral. In the rays, or Batoidei, the gill-openings are always placed on the abdominal aspect of the fish; the body is depressed, and the trunk, whi ’ is surrounded by the immensely developed pectoral fins, forms abroad flat disk, of which the tail appears as a thin and slender appendage. Spiracles are always present; the number of gill-openings is constantly five; there is no anal fin; the dorsal fins, if present, are situated on the tail. Some of the rays, however, approach the sharks in having th J 'audal portion less abruptly contracted behind the trunk.

Fossil Plagiostomes are very numerous in all formations. Some of the earliest determinable fish remains are, or are believed to be, derived from Plagiostomes. Those which can be referred to any of the families specified below will be mentioned in due course; but there are others, especially fin-spines, which leave us in doubt to which group of Plagiostomes their owners had affinity: as Onchus, from the Upper Silurian, continuing to Carboniferous formations; Dimeracanthus, Homacanthus, from the Devonian; Oracanthus, Gyracanthus, Τriskychius, Astroptychius, Ptychacankhus, Sphena∙ canthus, &c., from Carboniferous formations; Leptacanthus, from the Coal to the Oolite; Cladacanthus, Cricacanthus, Gyropristis, and Lepracanthxis, from the Coal-measures; Nemacaιdhus, Liacanthus, from the Trias; Astracanthus, Myriacankhus, Pristacanthus from the Jurassic group.

A. Selachoidei, Sharks.

Family 1. Carchariidae. — Eye with a nictitating membrane. Mouth crescent-shaped, inferior. Anal fin present. Two dorsal fins, the first opposite to the space between the pectoral and ventral fins, without spine in front.

Genera: Garcharias (Blue Shark), Galeocerdo, Galeus (Tope), Zygoena (Hammerhead), Mustelus (Hound), Hernigaleus, Loxodon, Thalassorhinus, Tricenodon, Leptocarcharias, and Triads. Fossil: Corax and Hemipristis.

Family 2. Lamnidae. —Eye without nictitating membrane. Anal fin present. Two dorsal fins, the first opposite to the space between the pectoral and ventral fins, without spine in front. Nostrils not confluent with the mouth, which is inferior. Spiracles absent or minute.

Genera: Lamna (Porbeagle), Carcharodon, Odontaspis, Alopecias (Fox-Shark), and Selache (Basking Shark). Fossil: Carcharopsis, Oxytes, Sphenodus, Gomphodus, and Ancistrodon.

Family 3. Rhinodonkidae. — No nictitating membrane. Anal fin present. Two dorsal fins, the first nearly opposite to the ventrals, without spine in front. Mouth and nostril near the extremity of the snout.

This small family comprises one species only, Rhinodon typicus, a gigantic shark, which is known to exceed 50 feet in length.

Family 4. Notidanidoe. — No nictitating membrane. One dorsal fin only, without spine, opposite to-the anal. Dentition unequal in the jaws; in the upper jaw one or two pairs of awl-shaped teeth, the following six being broader, and provided with several cusps, one of which is much stronger than the others. Lower jaw with six large comb-like teeth on each side, besides the smaller posterior teeth. Gill-openings wide, six in number in Hexanchus, seven in Heptanchus. Fossil teeth belonging to this type have been found in Jurassic and later formations {Notidanus and Aellopos).

Family 5. Scylliidae. — Two dorsal fins, without spine, the first above or behind the ventrals; anal fin present. No nictitating membrane. Spiracle always distinct. Mouth inferior. Teeth small, several series generally being in function.

Genera: Scyllium (Dog-Fishes), Pristiurus, Parascyllium, Gin· glymostoma, Stegostoma, Chiloscyllium, Crossorhinus. Fossil: Scyl-liodιιsj Paloeoscyllium, Thyellina.

Family 6. Hybodonticloe. — Two dorsal fins, each with a serrated spine. Teeth rounded, longitudinally striated, with one larger and from two to four smaller lateral cusps. Skin covered with shagreen.

Extinct. From Carboniferous, Liassic, and Triassic formations. Several genera have been distinguished; and, if Cladodus belongs to this family, it was represented even in the Devonian.

Family 7. Cestraciontidoe. — No nictitating membrane. Two dorsal fins, the first opposite to the space between the pectoral and ventral fins; anal fin present. Nasal and buccal cavities confluent. Teeth obtuse, several series being in function.

This family is one of particular interest, because representatives of it occur in numerous modifications in Primary and Secondary strata. Their dentition is uniformly adapted for the prehension and mastication of crustaceous and hard-shelled animals. The fossil forms far exceeded in size the species of the only surviving genus, Cestracion (the Port Jackson Shark); they make their appearance with Ctenoptychius in the Devonian; this is succeeded in the Coal-measures by Psammodus, Chornatodus, Petrodus, Cochliodus, Poly-rhizodus, &c., and in the Trias and Chalk by Strophodus, A crodus, Thectodus, an∣l Ptychodus. Of the 25 genera known, 22 have lived in the periods preceding the Oolitic.

Family 8. Spinaddae. — No membrana nictitans. Two dorsal fins; no anal. Mouth but slightly arched; a long, deep, straight, oblique groove on each side of the mouth. Spiracles present; gillopenings narrow. Pectoral fins not notched at their origin.

Genera: Centrina, Acanthias (Spiny Dog-Fish), Centrophorus, Spinax, Centroscyllium, Scymnus, L<κmargus (Greenland Shark), Echinorhinus, Euprotornicrus and Isistius. The largest of these fishes is the Greenland shark, which attains to a length of 15 feet, and is common in the Arctic regions. Fossil genera: Paloeospinax and Drepanephorus, from Cretaceous and Jurassic formations.

Family 9. Rhinidoe. — No anal fin; two dorsal fins. Spiracles present. Pectoral fins large, with the basal portion prolonged forwards, but not grown to the head. Gill-openings rather wide, lateral, partly covered by the base of the pectoral. Spiracles wide, behind the eyes. Teeth conical.

One genus only: the “Angel-Fish,” or “Monk-Fish” {Rhina squatina), which approaches the rays as regards general form and habits. Extinct forms, closely allied to the “Angel-Fish,” are found in the Oolite, and have been described as Thaumas. The Carboniferous genus Orthacanthus may have been allied to this family, but it was armed with a spine immediately behind the head.

Family 10. Pristiophoridoe. — The rostral cartilage is produced into an exceedingly long, flat lamina, armed along each edge with a series of teeth (saw).

These sharks so greatly resemble the common saw-fishes as to be easily confounded with them, but their gill-openings are lateral, and not inferior. Only one genus is known, Pristiophomιs, which occurs in the Australian and Japanese seas. Squaloraia, from the Lias, is supposed to have its nearest affinities to this family.

B. Batoidei, Rays.

Family 1. Pristidae. — The snout is produced into an exceedingly long flat lamina, armed with a series of strong teeth along each edge (saw).

One genus only: Pristis (Saw-Fishes).

Family 2. Rhinobatidae. — Tail strong and long, with two well-developed dorsal fins, and a longitudinal fold on each side; caudal developed. Disk not excessively dilated, the rayed portion of the pectoral fins not being continued to the snout. Teeth obtuse, granular, the dental surfaces of the jaws being undulated.

Genera: Rhynchobatus, Rhinobatus, and Trygonorhina. Fossil: Spathobalis and Trigorhina.

Family 3. Torpedinidae ((Electric Rays).—Trunk a broad, smooth disk. Tail with a longitudinal fold on each side; a rayed dorsal fin is generally, and a caudal always, present. Anterior nasal valves confluent into a quadrangular lobe. An electric organ composed of vertical hexagonal prisms between the pectoral fins and the head.

Genera: Torpedo, Narcine, Hypnos, DiscoρygetSxoτa, Peru), Astrape, and Temera. A large fish, with the general appearance of a torpedo, has been found at Monte Bolca; and Gyclobatis, from the Upper Cretaceous Limestone of Lebanon, is probably another extinct representative of this family.

Family 4. Raiidae. — Disk broad, rhombic, generally with asperities or spines; tail with a longitudinal fold on each side. The pectoral fins extend to the snout. No electric organ; no serrated caudal spine.

Genera: Raia (Rays and Skates), Psammobatis, Sympierygia, Platyrhina. Fossil: Arthropterus.

Family 5. Trygonidae. — The pectoral fins are uninterruptedly continued to, and are confluent at, the extremity of the snout. Tail long and slender, without lateral longitudinal folds; vertical fins none, or imperfectly developed, often replaced by a strong serrated spine.

Genera: Urogymnus, Trygon (Sting-Rays), Urolophus, Ptero~ platea.

Family 6. MyliobatujlaeC t Devil-Fishes,” “Sea-Devils,”or “Eagle-Rays”).—The disk is very broad, in consequence of the great development of the pectoral fins, which, however, leave the sides of the head free, and reappear at the extremity of the snout as a pair of detached (cephalic) fins. Viviparous.

Genera: Myliobatis, Aetobatis, Rhinoptera, Dicerobatis, Cerato-ptera. Fragmentary portions of their tessellated dentition are common in Tertiary formations.

Suborder II. Holocephala.— One external gill-opening only, covered by a fold of the skin, which encloses a rudimentary cartilaginous gill-cover; four branchial clefts within the gill-cavity. The maxillary and palatal apparatus coalescent with the skull.

This suborder is represented in the living fauna by one family only. Ghimaeridae ; it forms a connecting link with the following order of fishes, the Ganoida In external appearance, and as regards the structure of their organs of propagation, the chimaeras are sharks. The females are provided with “claspers” in connexion with the ventral fins, and the ova are large, encased in a horny capsule, and few in number; there is no doubt that they are impregnated within the oviduct, as in sharks. The males are provided with a singular erectile appendage, spiny at its extremity, and received in a groove on the top of the head. On the other hand, the relations of the chimaeras to the Ganoid, and, more especially, to the Dipnoous type become manifest in their notochordal skeleton and continuity of cranial cartilage. The spine in front of the first dorsal fin is articulated to the neural apophysis, and not merely implanted in the soft parts; it is immovable as in sharks. A cartilaginous operculum makes its appearance, and the external gill-opening is single. The dentition is that of a Dipnoid, each “jaw” being armed with a pair of broad dental plates, with the addition of a pair of smaller cutting teeth in the upper “jaw.” Fossils of similar dental combination are not rare in strata commencing with the Lias and the bottom of the Oolitic series; but it is impossible to decide in every case whether the fossil should be referred to the Holoceρhalous or the Dipnoous type. According to Newberry, Chimaeroid fishes commence in the Devonian with Rhynchodus f the remains of which were discovered by him in Ohio. Undoubted Chimaeroids are Elasmodus, Psaliodus, Ganodus, Ischyodus, Edaphodon, and Elasmognathus, principally from Mesozoic and Tertiary formations. Very similar fossils occur in the corresponding strata of North America. A single species of Callorhynchus has been discovered by Hector in the Lower Greensand of New Zealand.

The living chimaeras are few in number, and remain within very moderate dimensions, probably not exceeding a length of 5 feet, inclusive of their long filamentous, diphycercal tail. They are referred to two genera, Chimaera and Callorhynchus. Order II. —Ganoidri.

Skeleton cartilaginous or ossified. Body with medial and paired fins, the hinder pair abdominal. Gills free, rarely partially attached to the walls of the gill-cavity. One external gill-opening only on each side; a gill-cover. Air-bladder with a pneumatic duct. Ova small, impregnated after exclusion. Embryo sometimes with external gills.

To this order belong the majority of the fossil fish remains of Palaeozoic and Mesozoic age, whilst it is very scantily represented in the recent fauna, and evidently verging towards total extinction. The knowledge of the fossil forms, based on mere fragments of the hard parts of the body only, is very incomplete, and therefore their classification is in a most unsatisfactory state. In the following list only the most important groups will be mentioned.

Eight suborders may be distinguished at present.

Suborder I. Placodermi. — Extinct. The head and pectoral region of the body encased in great bony sculptured plates, with dots of enamel; the remainder of the body naked, or with ganoid scales; skeleton notochordal.

Comprises the oldest vertebrate remains, from Devonian and Carboniferous formations.

Genera: Pterichthys, Coccosteus, Dinichthys, Cephalaspis, Auchenaspis, Didymaspis, Pteraspis, Scaphaspis, Cyathasρis, Astrolepis. Suborder II. Acanthodini.— Extinct. Body oblong, compressed, covered with shagreen; skull not ossified; caudal fin heterocercal. Large spines, similar to those of Chondropterygians, in front of some of the median and paired fins. The spines are imbedded between the muscles, and not provided with a proximal joint.

Genera: Acanthodes, Chiracanthus, from Devonian and Carboniferous formations.

Suborder III. Dipnoi.— Nostrils two pairs, more or less within the mouth; limbs with an axial skeleton. Lungs and gills. Skeleton notochordal. No branchiostegals.

Family 1. Sirenidoe. — Caudal fin diphycercal; no gular plates; scales cycloid. A pair of molars, above and below, and one pair of vomerine teeth.

Genera: Lepidosiren and Ceratodus. Lepidosiren (including Protoptenιs) has the body eel-shaped, with one continuous vertical fin. The limbs are reduced to filaments. Vomerine teeth conical, pointed. Each dental lamina or molar with strong cusps, supported by vertical ridges. Conus arteriosus with two longitudinal valves. Ovaries closed sacs. Two species are known: — L. paradoxa, from the system of the river Amazon, and L. (Protopferus) annectens, which abounds in many localities of the west coast of Africa, is spread over the whole of tropical Africa, and in many districts of the central parts forms a regular article of food.

Ceratodus has the body elongate, compressed, with one continuous vertical fin. The limbs are paddle-shaped, with broad, rayed fringe. Vomerine teeth incisor-like; molars with flat, undulated surface, and lateral prongs. Conus arteriosus with transverse series of valves. Ovaries transversely lamellated.

Two species, C. forsteri and C. mioleρis, are known from fresh waters of Queensland. The specimens obtained hitherto have come from the Burnett, Dawson, and Mary rivers, some from the fresh waters of the upper parts, others from the lower brackish portions. The fish is said to grow to a weight of 20 tb and to a length of 6 feet. Locally, the settlers call it “Flat-head,” “Burnett-Salmon,” or “Dawson-Salmon,” and the aborigines “Barramunda,” a name which they appear to apply also to other large-scaled freshwater fishes, as the Osteoglossurn leichardU. In the stomach there is generally found an enormous quantity of the leaves of plants growing on the banks of rivers, evidently eaten after they had fallen into the water and when in a decomposing condition. The flesh of the fish is salmon-coloured, and is much esteemed as food. The barramunda is said to be in the habit of going on land, or at least on mud-flats; and this assertion appears to be borne out by the fact that it is provided, with a lung. If is much more probable, however, that it rises now and then to the surface of the water in order to fill its lung with air, and descends again until the air is so much deoxygenized as to render a renewal of it necessary. As the barramunda has perfectly developed gills, as well as the lung, we can hardly doubt that, when it is in water of normal composition, and 'pure enough to yield the necessary supply of oxygen, these organs are sufficient for the purpose of breathing, and that the respiratory function rests with them alone. But when the fish is compelled to sojourn in thick muddy water charged with gases which are the products of decomposing organic matter (and this must very frequently be the case during the droughts which annually exhaust the creeks of tropical Australia), it commences to breathe air with its lung in the way indicated above. If the medium in which it happens to be is altogether unfit for breathing, the gills cease to have any function; if it is irrespirable in a less degree the gills may still continue to assist the lung. The barramunda, in fact, can breathe by either gills or lung alone, or by both simultaneously.

The discovery of Ceratodus does not date farther back than the year 1870. It proved to be of the greatest interest, not only on account of the relation of this creature to the other living Dipnoi and Ganoidei, but also because it threw fresh light on those singular fossil teeth which are found in strata of the 'Γriassic and Jurassic formations in various parts of Europe, India, and America. These teeth vary greatly in general shape and size; they are sometimes 2 inches long, much longer than broad, depressed, with a flat or slightly undulated always punctuated, crown, with one margin convex, and with from three to seven prongs projecting on the opposite margin.

Family 2. Ctenododiρteridce.— Caudal fin heterocercal; gular plates; scales cycloid. Two pairs of molars and one pair of vomerine teeth.

Extinct. Diρterus (Ctenodus), Heliodus, from Devonian strata.

Family 3. Phaneropleuridoe. — Caudal fin diphycercal; vertical fin continuous; gular plates; scales cycloid. Jaws with a series of minute conical teeth on the margin.

Extinct. Phaneroρleuron, from Devonian and Carboniferous formations.

Suborder IV. Chondrostei. —Skeleton notochordal; skull cartilaginous, with dermal ossifications; branchiostegals few in number or absent. Teeth minute or absent. Integuments naked or with bucklers. Caudal fin heterocercal, with fulcra. Nostrils double, in front of the eyes.

Family 1. Acipensericloe. — Body elongate, sub-cylindrical, with five rows of osseous bucklers. Snout produced, subspatulate or conical, with the mouth at its lower surface, small, transverse, protractile, toothless. Four barbels in a transverse series on the lower side of the snout. Vertical fins with a single series of fulcra in front. Dorsal and anal tins approximate to the caudal. Branchiostegals none. Air-bladder large, simple, communicating with the dorsal wall of the oesophagus.

Genera: Aciρenscr and Scaphirhynchus (Sturgeons, Hausen, Sterlet).

Family 2. Polyodontidae. — Body naked, or with minute stellate ossifications. Mouth lateral, very wide, with minute teeth in both jaws. Barbels none. Caudal fin with fulcra. Dorsal and anal fins approximate to the caudal.

Genera: Polyodon and Psephurus. Fossil: Ghondrosteus. Suborder V. Polypteroidek— Paired fins with axial skeleton, fringed; dorsal fins two or more. Branchiostegals absent, but generally gular plates. Vertebral column diphycercal or heterocercal. Body scaly.

Family 1. Polypteridae. — Scales ganoid; fins without fulcra. A series of dorsal spines, to each of which an articulated finlet is attached; anal placed close to the caudal fin, the vent being near the end of the tail. Abdominal portion of the vertebral column much longer than the caudal.

Genera: Polypterus and Calamoichthys. Polypterus is confined to tropical Africa, occurring in abundance in the rivers of the west coast and in the upper Nile; it is rare in the middle and lower Nile. There is only oħe species known, Polypterusbichir (“Bichir” being its vernacular name in Egypt), which varies in the number of the dorsal finlets, the lowest being eight, the highest eighteen. It attains to a length of 4 feet. Nothing is known of its mode of life, observations on which are very desirable. Calamoichthys (from Old Calabar) is distinguished from Polypterus by its greatly elongated form and the absence of ventral fins.

Family 2. Saurodipteridoe. — Seales ganoid, smooth like the surface of the skull. Two dorsal fins; paired fins obtusely lobate. Teeth conical. Caudal fin heterocercal.

Extinct. Diρloρterus, Megalichthys, and Osteolepis, from Devonian and Carboniferous formations.

Family 3. Coelacanthidoe. — Scales cycloid. Two dorsal fins, each supported by a single two-pronged interspinous bone; paired fins obtusely lobate. Air-bladder ossified; notochord persistent, diphycercal.

Extinct. Coelacanthus, from Carboniferous strata. Several other genera, from the Coal formations to the Chalk, have been associated with it:— Undina, Graphiurus, Macropoma, Holophagus, Hoplo-pygus, Rhizodus.

Family 4. Holoptychidoe. — Scales cycloid or ganoid, sculptured. Two dorsal fins; pectorals narrow, acutely lobate; dentition den-drodont.

Extinct. Genera: Holoptychius, Saurichthys, Glyptolepis, Den-drodus, Glyρtolcemus, Glyptopomus, Tristichopterus, Gyroptychius, Strcpsodus, from Devonian and Carboniferous strata.

Suborder VI. Pycnodontoidei.— Body compressed, high and short or oval, covered with rhombic scales arranged in decussating pleuro-lepi<ial lines. Notochord persistent. Paired fins without axial skeleton. Teeth on the palate and hinder part of the lower jaw molar-like. Branchiostegals, but no gular plates. Extinct.

Family 1. Pleurolepidoe. — Homocercal. Body less high. Fins with fulcra.

Genera: Pleurolepis and Homoeolepis, from the Lias.

Family 2. Pycnodontidae. — Homocercal. The neural arches and ribs are ossified; the roots of the ribs are but little expanded in the older genera, but are enlarged in the Tertiary forms, so as to simulate vertebrae. Paired fins not lobate. Obtuse teeth on the palate and the sides of the mandible; maxilla toothless; incisor-like teeth in the intermaxillary and front of the mandible. Fulcra absent in all the fins.

Genera: Gyrodus, Mestunιs, Microdon, Coelodus, Pycnodus, Mesodon, from Mesozoic and Tertiary formations.

Suborder VIL Lepidosteoidei.— Scales ganoid, rhombic; fins generally with fulcra; paired fins not lobate. Praeoperculum and interoperculum developed; generally numerous branchiostegals, but no gular plate.

Family 1. Lepidosteidae. — Scales ganoid, lozenge-shaped. Skeleton completely ossified; vertebrae convex in front and concave behind. Fins with fulcra; dorsal and anal composed of articulated rays only, placed far backwards, close to the caudal. Abdominal part of the vertebral column much longer than caudal. Branchiostegals not numerous, without enamelled surface. Heterocercal.

Lep>idosteus.~ This genus existed as early as Tertiary times; their remains have been found in Europe as well as in North America. In our period they are limited to the temperate parts of North America, Central America, and Cuba. Three species can be distinguished, which attain to a length of about 6 feet. They feed on other fishes; and their general resemblance to a pike has gained for them the vernacular names of “Gar-Pike” or “Bony Pike.”

Family 2. Sauridae. — Body oblong, with ganoid scales; vertebrae not completely ossified; termination of the vertebral column homocercal; fins generally with fulcra. Maxillary composed of a single piece; jaws with a single series of conical pointed teeth. Branchiostegals numerous, enamelled, the anterior forming broad gular plates.

Extinct. Genera: Semionotus, Eugnathus, Cephenoplosus, Ma-crosemius, Propter ιιs, Ophiopsis, Pholidophorus, Pleuropholis, Pachycormus, Ptycholcpis, Conodus, Eulepidotus, Lophiostomus, &c.

Family 3. Stylodontidae. — Body rhombic or ovate, with ganoid scales; vertebrae not completely ossified; termination of the vertebral column homocercal; fins with fulcra. Maxillary composed of a single piece; jaws with several series of teeth, the outer ones equal, styliform. Dorsal fin very long, extending to the caudal. Branchiostegals numerous.

Extinct. Tetragonolepis, from the Lias.

Family 4. Sphaerodontidoe. — Body oblong, with rhombic ganoid scales; vertebrae ossified, but not completely closed; homocercal; fins with fulcra. Maxillary composed of a single piece; teeth in several series, obtuse; those on the palate globular. Dorsal and anal fins short. Branchiostegals.

Extinct. The type genus of this family is Lepidotus.

Family 5. Aspidor!ιynchida∙.— Body elongate, with ganoid scales; jaws prolonged into a beak; termination of the vertebral column homocercal. Fins with fulcra; a series of enlarged scales along the side of the body. Dorsal fin opposite to the anal.

Extinct: Mesozoic. Genera: Aspidorhynchus and Belonostomus.

Family 6. Palaeoniscidoe. — Body fusiform, with rhombic ganoid scales. Notochord persistent, with the vertebral arches ossified. Heterocercal. All the fins with fulcra; dorsal short. Branchiostegals numerous, the foremost pair forming broad gulars. Teeth small, conical, or cylindrical.

Extinct. Genera: from the Old Red Sandstone— Chirolepis and Acroleρis ; from Carboniferous rocks— Cosmoptychius, Elonichthys, Nematoptychius, Cyclopiychius, Microconodus, Gonatodιιs, Rhadinichthys, Myriolcpis, Urosthenes ; from the Permian— Rhabdolepis, Paloeoniscus, Amblyptcrus, and Pygopteτus; from the Lias— Centrolepis, Oxygnathus, Cosrnolepis, and Thrissonotus.

Family 7. Platysomidce.— Body generally high, compressed, covered with rhombic ganoid scales arranged in dorso-ventral bands. Notochord persistent, with the vertebral arches ossified. Heterocercal; fins with fulcra; dorsal fin long, occupying the posterior half of the back. Branchiostegals numerous. Teeth tubercular or obtuse.

Extinct. Genera: from Carboniferous and Permian formations —Eurynotus, Benedenius, Mesolepis, Ewrysomus, Wardichtlvys, Chirodus (M*Coy), Platysomus. Suborder VIII. Amioidei. — Vertebral column more or less completely ossified, heterocercal. Body covered with cycloid scales. Branchiostegals present.

Family 1. Caturidae. — Notochord persistent, with partially ossified vertebrae; homocercal; fins with fulcra. Teeth in a single series, small, pointed.

Extinct. Catums, from the Oolite to the Chalk.

Family 2. Leplolepidae. — Scales cycloid. Vertebrae ossified; homocercal; fins without fulcra; dorsal short. Teeth minute, in bands, with canines in front.

Extinct, but leading to the living representative of this suborder. Genera: Thrissoρs, Leptolepis, from the Lias and Oolite. These fishes, so far as the preserved parts are concerned, cannot be distinguished from Teleosteous fishes, to which they are referred by some palaeontologists.

Family 3. Amiidoe. — Skeleton entirely ossified; a single large fular plate; homocercal; fins without fulcra; a long soft dorsal n. Abdominal and caudal parts of the vertebral column subequal in extent. Branchiostegals numerous.

Amia.— The “Bow-Fin” or “Mud-Fish” (A. calva) is notuncommon in many of the fresh waters of the United States; it grows to a length of 2 feet. Little is known about its habits; small fishes, crustaceans, and aquatic insects have been found in its stomach. Wilder has observed its respiratory actions; it rises to the surface, and, without emitting any air-bubble whatever, opens the jaws widely, and apparently gulps in a large quantity of air; these acts of respiration are more frequently performed when the water is foul or has not been changed; and there is no doubt that a conversion of oxygen into carbonic acid is effected, as in the lungs of aerial vertebrates.

Fossil remains occur in Tertiary deposits of North America, which have been distinguished as Protamia and Hypamia. Subclass II.—Teleostei.

Heart with a non-contractile bulbus arteriosus; intestine without spiral valve; optic nerves decussating; skeleton ossified, with completely formed vertebrae; vertebral column diphycercal or homocercal; branchiae free.

The Teleostei form the majority of the fishes of the present fauna, and are the geological successors of the Palaeichthyes, undoubted Teleostei not ranging farther back than the Chalk period. This subclass comprises an infinite variety of forms; and as, naturally, many Ganoid fishes lived under external conditions similar to those of certain Teleostei, and led a similar mode of life, we find not a few analogous forms in both series,—some Ganoids resembling externally the Teleosteous Siluroids, others the Clupeoids, others the Chaetodonts, others the Scombresoces, <fcc. But there is no direct genetic relation between these fishes, as some naturalists have been inclined to believe.

The Teleostei are divided into six orders :—

I. Acanthopterygii.— Part of the rays of the dorsal, anal, and ventral fins non-articulated spines. The lower pharyngeale separate. Air-bladder, if present, without pneumatic duct in the adult.

II. Acanthopterygii Pharyngognathi.— Part of the rays of the dorsal, anal, and ventral fins non-articulated spines. The lower pharyngeals coalesced Air-bladder without pneumatic duct.

III. Anacanthini.— Vertical and ventral fins without spinous rays. Ventral fins, if present, jugular or thoracic. Air-bladder, if present, without pneumatic duct. Lower pharyngeals separate.

IV. Physostomi.— All the fin-rays articulated; only the first of the dorsal and pectoral fins is sometimes ossified. Ventral fins, if present, abdominal, without spine. Air-bladder, if present, with a pneumatic duct.

V. Lophobranchii.— Gills not laminated, but composed of small rounded lobes, attached to the branchial arches. Gillcover reduced to a large simple plate. A dermal skeleton replaces more or less soft integuments.

VI. Plectognαthi.— A soft dorsal fin opposite to the anal; sometimes elements of a spinous dorsal. Ventral fins none, or reduced to spines. Gills pectinate; air-bladder without pneumatic duct. Skin with rough scutes, or with spines, or naked.

Order I.— AcanthοpterygH.

Part of the rays of the dorsal, anal, and ventral fins are nonarticulated, more or less sharp-pointed spines. The lower pharyngeals are generally separate. Air-bladder, if present, without pneumatic duct in the adult.

Division 1. Acanthοpterygii Pcrciformes.— Body more or less compressed, elevated or oblong, but not elongate; the vent is remote from the extremity of the tail, behind the ventral fins if they are present. No prominent anal papilla. No superbranchial organ. Dorsal fin or fins occupying the greater portion of the back; spinous dorsal well developed, generally with stiff spines, of moderate extent, rather longer than, or as long as, the soft; the soft anal similar to the soft dorsal, of moderate extent or rather short. Ventrals thoracic, with one spine and with four or five rays.

Family 1. Percidae. — The scales extend but rarely over the vertical fins, and the lateral line is generally present, continuous from the head to the caudal fin. All the teeth simple and conical; no barbels. No bony stay for the praeoperculum.

A large family, represented by numerous genera and species in fresh waters, and on all the coasts of the temperate and tropical regions. Carnivorous.

Fossil Percoids abound in some formations, for instance, at Monte Bolca, where species of Labrax, Lates, Smerdis and Cyclopoma (both extinct), Dules, Serranus, Apogon, Τherapon, and Pristipoma have been recognized. Paraperca is a genus recently discovered in the Marls of Aix-en-Provence. A species of Peτca is known from the freshwater deposits of Oeningen.

The living genera are— Perea (Perch), Siniperca, Percichthys, Labrax (Bass), Lates, Psammoperca, Percalabrax, Acerina (Pope), Lucioperca, Pilcorna, Boleosoma, Aspro, Centropomus, Enoplosus, Gentropristis, Anthias, Gallanthias, Serranus (Sea-Perch), Anyper∙ odon, Prionodes, Plectropoma, Trachypoma, Polyprion, Grammistes, Rhypticus, Aulacocephalus, Myriodon, Diploprion, Mesoprion, Glaucosoma, Bules, Therapon, Helotes, Pristipoma, Gonodon, Haemulon, Hapalogenys, Diagramma, Hyρeroglyρhe, Lobotes, Histio∙ pterus, Gerres, Scolopsis, Heterognathodon, Dentex, Symρhorus, Synagris, Moena, Smaris, Coesio, Erythrichthys, Oligorus (Murray-Cod), Grystes, Arripis, Huro, Ambassis, Apogon, Chilodipterus, Lanioperca, Acropoma, Scombrops, Pomatomus, Priacanthus, Cen∙ trarchus, Bryttus, Pomotis, Etelis, Niphon, Aprion, Apsilus, Penta· ceros, Velifer, Datnioides, Percilia.

Family 2. Squamipinnes.— Body compressed and elevated, covered with scales, either finely ctenoid or smooth. Lateral line continuous, not continued over the caudal fin. Mouth in front of the snout, generally small, with lateral cleft. Teeth villiform or setiform, in bands, without canines or incisors. Dorsal fin consisting of a spinous and soft portion of nearly equal development; anal with three or four spines, similarly developed as the soft dorsal, both being many-rayed. The vertical fins more or less densely covered with small scales. The lower rays of the pectoral fin branched, not enlarged; ventrals thoracic, with one spine and five soft rays.

Genera: Ghcetodon, Ghelmo, Heniochus, Holacanthus, Pornacan-thus, Scatophagus, Ephippus, Drepane, Hypsinotus, Scorpis, Atypichthys, Toxotes. Some of these genera occur also in Tertiary formations. The majority of these fishes are inhabitants of the tropical seas, and abound chiefly in the neighbourhood of coral-reefs (Coral-Fishes). The beauty and singularity of distribution of the colours in some of the genera, as Ghoetodon, Heniochus, Holacanthus, is scarcely surpassed in any other group of fishes. They are of small size, and are carnivorous, feeding on small invertebrates.

Family 3. Mullidoe. — Body rather low and slightly compressed, covered with large thin scales, with or without an extremely fine serrature. Two long erectile barbels are suspended from the hyoid, and are received between the rami of the lower jaw and the opercula. Mouth in front of the snout, with the cleft lateral and rather short; teeth very feeble. Eye lateral, of moderate size. Two short dorsal fins remote from each other, the first with feeble spines; anal similar to the second dorsal. Ventrals with one spine and five rays. Pectorals short.

Genera: Mullus and Upeneus (Bed Mullets).

Family 4. Sparidae (Sea-Breams).—Body compressed, oblong, covered with scales, the serrature of which is very minute, and sometimes altogether absent. Mouth in front of the snout, with cleft lateral. Either cutting teeth in front of the jaws, or molar teeth on the side; palate generally toothless. One dorsal fin, formed by a spinous and soft portion of nearly equal development. Anal fin with three spines. The lower rays of the pectoral fin are generally branched, but in one genus simple. Ventrals thoracic, with one spine and five rays.

Genera: Cantharus. Box. Scathams Oblata, Crenidens, Tri∙ pterodon, Pachymetopon, Dipterodon, Gymnocrotaphus, Girella, Tephroεops, Doydixodon, Haρlodactylus, Sargus (Sheep’s Head), Lethrinus, Sphaerodon, Pagrus, Pagellus, Chrysophrys, and Pime-leptcmιs.

The extinct forms found hitherto are rather numerous; the oldest come from the Cretaceous formation of Mount Lebanon; some belong to living genera, as Sarqus, Pagellus ; of others from Eocene and Miocene formations no living representative is known— Sparn∙ odus, Sargodon, Capitodus, Soricidens, Asima.

The Hoplognathidoe aie a very small family distinguished by confluent teeth, and allied to the sea-breams.

Family 5. Girrhitidae. — Body oblong, compressed, covered with cycloid scales; lateral line continuous. Mouth in front of the snout, with lateral cleft. Cheeks without a bony stay for the praeoperculum. Dentition more or less complete, composed of small pointed teeth, sometimes with the addition of canines. One dorsal fin formed by a spinous and soft portion, of nearly equal development. Anal with three spines, generally less developed than the soft dorsal. The lower rays of the pectoral fins simple and generally enlarged; ventrals thoracic, but remote from the root of the pectorals, with one spine and five rays.

Genera: Cirrhites, Chironem/us, Chilodactylus, Mendosoma, Nema-dactylus, and Latris.

Family 6. Scorpaenidae. — Body oblong, more or less compressed, covered with ordinary scales or naked. Cleft of the mouth lateral or subvertical. Dentition feeble, consisting of villiform teeth, and generally without canines. Some bones of the head armed, especially the angle of the praeoperculum, its armature receiving additional support by a bony stay, connecting it with the infraorbital ring. The spinous portion of the dorsal fin equally or more developed than the soft and the anal. Ventrals thoracic, generally with one spine and five soft rays, sometimes rudimentary.

Genera: Sebastes, Scorpaena, Glyptauchen, Lioscorpius, Setarches, Pterois, Aρistus, Agriopus, Synanceia, Micropus, Ghorisrnodactylus, Toenianotus, Centropogon, Pentaroge, Tetraroge, Prosopodasys, Aplo∙ actis, Trichopleura, Hemitripterus, Minous, and Pelor.

Family 7. Nandidae. — Body oblong, compressed, covered with scales. Lateral line interrupted. Dorsal fin formed by a spinous and soft portion, the number of spines and rays being nearly equal; anal fin with three spines, and with the soft portion similar to the soft dorsal. Ventral fins thoracic, with one spine, and five or four rays. Dentition more or less complete, but feeble.

Genera: Plesiops, Trachinops, Badis, Nandus, and Catopra.

Family 8. Polyecntridae. — Body'compressed, deep, scaly. Lateral line none. Dorsal and anal fins long, both with numerous spines, the spinous portion being the more developed. Ventrals thoracic, with one spine and five soft rays. Teeth feeble. Pseudobranchiae hidden.

Genera: Polyccntrus and Monociτrhιιs.

Family 9. Teuthididoe. — Body oblong, strongly compressed, covered with very small scales. Lateral line continuous. A single series of cutting incisors in each jaw; palate toothless. One dorsal fin, the spinous portion being the more developed; anal with seven spines. Ventral fins thoracic, with an outer and an inner spine, and with three soft rays between.

Genus: Teuthis. Division II. Acanthopterygii Beryciformes.— Body compressed, oblong, or elevated; head with large muciferous cavities, which are covered with a thin skin. Ventral fins thoracic, with one spine and more than five soft rays (in Monocentris with two only).

The family of Berycidae has a very remarkable geographical distribution, which has been noticed atpp. 679, 681. Members of this family belong to the oldest Teleosteous fishes, the majority of the Acanthopterygians found in the Chalk being Berycoids. Beryx, Holocentrum, and Myripristis have been found in several species, with other genera now extinct:— Pseudoberyx, Berycopsis, Homonotus, Stenostoma, Sphcnoccphalus, Acanus, Hoploptcryx, Platycornus, Podocys, Acrogaster, Macrolepis, and Rhacolepis.

Living genera: Monocentris, Hoplostethus, Trachichthys, Anoplo-gaster, Beryx, Melamphacs, Polymixia, Myripristis, Αnd Holocentrum. Division III. Acanthopterygii KurtiJ ormes. — One dorsal fin only, much shorter than the anal, which is long and many-rayed. No superbranchial organ.

One family, Kurtidae.

Genera: Pemphcris and Kurtus. Division IV. Acanthopterygii Polynerniformes.— Two rather short dorsal fins, somewhat remote from each other; free filaments at the humeral arch, below the pectoral fins; muciferous canals of the head well developed.

Family. Polynemidoe. — Body oblong, rather compressed, covered with smooth or very feebly ciliated scales. Lateral line continuous. Snout projecting beyond the mouth, which is inferior, with lateral cleft. Eye lateral, large. Villiform teeth in the jaws and on the palate. Ventrals thoracic, with one spine and five rays.

Genera: Polynemus, Pentancmus, and Galeoides. Division V. Acanthopterygii Scioeniformes — The soft dorsal is more, generally much more, developed than the spinous and the anal. No pectoral filaments; head with the muciferous canalswell developed.

Family. Seiasnidce.— Body rather elongate, compressed, covered with ctenoid scales. Lateral line continuous, and frequently extending over the caudal fin. Mouth in front of the snout. Eye lateral, of moderate size. Teeth in villiform bands, sometimes . with the addition of canines; no molars or incisor-like teeth in the jaws; palate toothless. Praeoperculum unarmed, and without bony stay. Ventrals thoracic, with one spine and five soft rays. Bones of the head with wide muciferous channels. Air-bladder frequently with numerous appendages.

Genera: Laτirnus, Pogonias (Drum), Micropogon, Umbrina, Eques, Nebris, Lonchurus, Scioena (Meagre), Pachyurus, Otolithus, Ancylodon, and Collichthys. Division VI. Acanthopterygii XiphHformes.- The upper jaw is produced into a cuneiform weapon.

One family, Xiphiidae (Sword-Fishes), with the genera Xiphias and Histiophorus. Division VII. AcanthopterygiiTrichiuriformcs.— Body elongate, compressed or band-like; cleft of the mouth wide, with several strong teeth in the jaws or on the palate. The spinous and soft portions of the dorsal fin and the anal are of nearly equal extent, long, many-rayed, sometimes terminating in finlets; caudal fin forked, if present.

The family of Trichiuridae is composed of the following living genera: Nealotus, Nesiarchus, Aphanopus Euoxymetopon, Lepidopus (Scabbard-Fish), Trichiurus (Hair-Tail), Epinnula, Thyrsites, and. Gempylus. It was well represented in the Chalk and later formations by Enchodus, Anenchelum, Nemopteryx, Xiphopterus, Hemi-thyrsites, and Trichiurichthys. Two other genera, Palaeorhynchv.s and Hemirhynchus, belong to a distinct though allied family.

Division VIII. Acanthopterygii Cotto·scombriformes.— Spines developed in one of the fins at least. Dorsal fins either continuous or close together; the spinous dorsal, if present, always short; sometimes modified into tentacles, or into a suctorial disk; soft dorsal always long, if the spinous is absent; anal similarly developed as the soft dorsal, and both generally much longer than the spinous, sometimes terminating in finlets. Ventrals, thoracic or jugular, if present, never modified into an adhesive apparatus. No prominent anal papilla.

Family 1. Acronuridoe. — Body compressed, oblong or elevated, covered with minute scales. Tail generally armed with one or more bony plates or spines, which are developed with age, but absent in very young individuals. Eye lateral, of moderate size. Mouth small; a single series of more or less compressed, sometimes denticulated, sometimes pointed incisors in each jaw; palate toothless. One dorsal fin, the spinous portion being less developed than the soft; anal with two or three spines; ventral fins thoracic. Nine abdominal and thirteen caudal vertebrae.

Genera: Acanthurus (Surgeon), Naseus, and Prionurus.

Extinct species of Acanthurus and Naseus have been discovered in the Monte Bolca formation.

Family 2. Carangidae. — Body more or less compressed, oblong or elevated, covered with small scales or naked; eye lateral. Teeth, if present^ conical. No bony stay for the praeoperculum. The spinous dorsal is less developed than the soft or the anal, either continuous with, or separated from, the soft portion; sometimes rudimentary. Ventrals thoracic, sometimes rudimentary or entirely absent. No prominent papilla near the vent. Gillopening wide. Ten abdominal and fourteen caudal vertebrae.

Genera: Caranx (Horse-Mackerel), Argyriosus, Micropteryx, Seriola, Seriolella, Seriolichthys (Yellow-Tails), Naucrates (Pilot-Fish), Chorinemus, Lichia, Temnodon (Blue-Fish), Lactarius, Paropsis, Trachynotus, Pammelas, Psettus, Platax, Zanclus, Anomalops, Capros (Boar-Fish), Antigonia, Diretmus, Equula, and Gazza.— Members of this family appear first in Cretaceous formations, where they are represented by Platax and some caranx-like genera (Vomer and Aipichthys). They are more numerous in Tertiary formations, especially in the strata of Monte Bolca, where some still existing genera occur, as Zanclus, Platax, Caranx (Carangopsis), Argyriosus (Vomer), Lichia, Trachynotus. Of extinct genera the following belong to this family :— Pseudovomer (Licata), Amphistium, Archceus, Ductor, Plioncmus (f), and Semio∙ phorus.

Family 3. Cyttidoe. — Body elevated, compressed, covered with small scales, or with bucklers, or naked; eye lateral. Teeth conical, small. No bony stay for the praeoperculum. Dorsal fin composed of two distinct portions. Ventrals thoracic. No prominent papilla near the vent. Gill-opening w ide. More than ten abdominal and more than fourteen caudal vertebrae.

Genera: Zeus (John Dory) and Cyttus.

Family 4. Slromateidoe. — Body more or less oblong and compressed, covered with very small scales; eye lateral. Dentition very feeble; oesophagus armed with numerous horny, barbed processes. No bony stay for the praeoperculum. Dorsal fin single, long, without distinct spinous division. More than ten abdδminal and more than fourteen caudal vertebrae.

Genera: Stromateus and Centrolophus.

Family 5. Coryphaenidae. — Body compressed; eye lateral. Teeth small, conical, if present; oesophagus smooth. No bony stay for the praeoperculum. Dorsal fin single, long, without distinct spinous division. More than ten abdominal and more than fourteen caudal vertebrae.

Genera: Coryphoena (Dolphin), Brama, Taractes, Lamρris (Sun-Fish), Pteraclis, Schedophilus, Diana, Ausonia, Mene. Fossil: Goniognathus.

Family 6. Nomeidoe.— Body oblong, more or less compressed, covered with cycloid scales of moderate size; eye lateral. No bony stay for the praeoperculum. Dorsal fin with a distinct spinous portion separated from the soft; sometimes finlets; caudal forked. More than ten abdominal and more than fourteen caudal vertebrae.

Genera: Gastrochisma, Nomeus, Psenes, and Cubiceps.

Family 7. Scombridae. — Body oblong, scarcely compressed, naked or covered with small scales; eye lateral. Dentition well developed. No bony stay for the praeoperculum. Two dorsal fins; generally finlets. Ventrals thoracic, with one spine and five rays. More than ten abdominal and more than fourteen caudal vertebrae.

Genera: Scomber (Mackerel), Thynnus (Tunny, Bonito, Albacore), Pelamys, Auxis, Cybium, Elacate, and Echeneis (Sucking-Fish).

Family 8. Trachinidce.— Body elongate, low, naked or covered with scales. Teeth small, conical. No bony stay for the praeoperculum. One or two dorsal fins, the spinous portion being always shorter and much less developed than the soft; the anal similarly developed as the soft dorsal; no finlets. Ventrals with one spine and five rays. Gill-opening more or less wide. Ten or more than ten abdominal and more than fourteen caudal vertebrae.

Genera: Uranoscopus, Leptoscopus, Agnus, Anema, Kathetostoma, Trachinus (Weever), Champsodon, Percis, Sillago, Bovichthys, Bathydraco, Choenichthys, Aphritis, Acanthaphritis, Eleginus, Chi-marrhichthys, Cottoperca, Pcrcophis, Trichodon, Pinguipes, Latilus, Opislhognathus, Pseudochromis, Cichlops, Pseudoplesiops, Notothenia, an<VHarpagifer. Fossil: Callipteryx, Trachinopsis, and Pseudoele∙ ginus. Malacanthus is the type of a family allied to the Trachinidce.

Family 9. Batrachidoe. — Head broad and thick; body elongate, compressed behind; skin naked or with small scales. No bony stay for the praeoperculum. Teeth conical, small or of moderate size. The spinous dorsal consists of two or three spines only; the soft and the anal long. Ventrals jugular, with two soft rays; pectorals not pediculated. Gill-opening a more or less vertical slit before the pectoral, rather narrow.

Genera: Batrachus, Thalassophryne, and Porichthys.— Psychrolutes and Neophrynichthys are allied forms.

Family 10. Pediculati.— Head and anterior part of the body very large, without scales. No bony stay for the praeoperculum. Teeth villiform or rasp-like. The spinous dorsal is advanced forwards, composed of a few more or less isolated spines, often transformed into tentacles,or entirely absent. Ventral fins jugular, withfour or five soft rays, sometimes absent. The carpal bones are prolonged, forming a sort of arm, terminating in the pectoral. Gill-opening reduced to a small foramen, situated in or near the axil. Gills two and a half, three, or three and a half; pseudobranchiae generally absent.

Genera: Lophius (Fishing-Frog, Angler), Ceratias, Hi'manto-lophus, Melanocetus, Oneirodes, Antennarius, Brachionichthys, Saccarius, Chaunax, Malthe, Halieutoea, Halieutichihys, Dibranchus, and Aegaeonichthys.

Family 11. Cotlidae. — Form of the body oblong, sub-cylindrical. Cleft of the mouth lateral. Dentition feeble, generally in villiform bands. Some bones of the head are armed; and a bony stay con. nects the praeopercular spine with the infraorbital ring. Two dorsal fins (rarely one), the spinous being less developed than the soft and the anal. Ventrals thoracic, with five or less soft rays.

Genera: Cottns (Bull-head, Miller’s Thumb), Centridermichthys, Icelus, Platy cephalus, Hoplichthys, Bunocottus, Rhamphocottus, Triglops, Podabrus, Blepsias, Nautichthys, Scorpoenichthys, Hemi-lepidotus, Artedius, Ptyonotus, Polycaulus, Bembras, and Trigla (Gurnards).

Family 12. Cataphracti. — Form of the body elongate, sub-cylindrical. Dentition feeble. Body completely cuirassed with osseous keeled scales or plates. A bony stay connects the angle of the praeoperculum with the infraorbital ring. Ventrals thoracic.

Genera: Agonus, Aspidophoroides, Siphagonus, Peristethus, Dactylopterus (Flying Gurnard).

Family 13. Pegasidae. — Body entirely covered with bony plates, anchylosed on the trunk and movable on the tail. Barbels none. The margin of the upper jaw is formed by the intermaxillaries and their cutaneous prolongation, which extends downwards to the extremity of the maxillaries. Gill-cover formed by a large plate, homologous to the operculum, praeoperculum, and suboperculum; interoperculum a long fine bone, hidden below the gill-plate. One rudimentary branchiostegal. The gill-plate is united to the isthmus by a narrow membrane; gill-openings narrow, in front of the base of the pectoral fin. Gills four, lamellated. Pseudobranchiae and air-bladder absent. One short dorsal and a similar anal fin, opposite to each other. Ventral fin present. Ovarian sacs closed.

One genus only is known, Pegasus. Division IX. Acanthopteryii GobUformes.— The spinous dorsal, or spinous portion of the dorsal, is always present, short, either composed of flexible spines, or much less developed than the soft; the soft dorsal and anal of equal extent. No bony stay for the angle of the praeoperculum. Ventrals thoracic or jugular, if present, composed of one sj>ine and five, rarely four, soft rays. A prominent anal papilla.

Family 1. Discoboli (Lump-Suckers).—Body thick or oblong, naked or tubercular. Teeth small. Ventral fins with one spine and five rays, all being rudimentary and forming the osseous support of a round disk, which is surrounded by a cutaneous fringe. Gill-openings narrow, the gill-membranes being attached to the isthmus.

Genera: Cyclopterus and Liparis.

Family 2. Gobiidae (Gobies).—Body elongate, naked or scaly. Teeth generally small, sometimes with canines. The spinous dorsal fin, or portion of the dorsal fin, is the less developed, and composed of flexible spines; anal similarly developed as the soft dorsal. Sometimes the ventrals are united into a disk. Gill-opening more or less narrow, the gill-membranes being attached to the isthmus.

Genera: Gobius, Latrunculus, Euctenogobius, Lophiogobius, Do· liichthys, Apocryptes, Evorthodus, Gobiosoma, Gobiodon, Trioeno- phorichthys, Sicydium, Periophthalmus, Bolcophthalmus, Eleotris, Lentipes, Trypauchen, Callionymus (Dragonet), Benthophilus, Am· blyopus, Orthostomus, Platyptera, Luciogobius, Oxymetopon, and, perhaps, Oxuderces. Division X. Acanthopterygii BlennUforrnes.— Body low, subcylindrical or compressed, elongate. Dorsal fin very long; the spinous portion of the dorsal, if distinct, is very long, as well developed as the soft, or much more; sometimes the entire fin is composed of spines only; anal more or less long; caudal fin sub-tnιncated or rounded, if present. Ventral fins thoracic or jugular, if present.

Family 1. Cepolidoe. — Body very elongate, compressed, covered with very small cycloid scales; eyes rather large, lateral. Teeth of moderate size. No bony stay for the angle of the praeoperculum. One very long dorsal fin, which, like the anal, is composed of soft rays. Ventrals thoracic, composed of one spine and five rays. Gill-opening wide. Caudal vertebrae exceedingly numerous.

Genus: Cepola (Band-Fishes). An allied family are the Tricho∙ notidee, with Trichonotus and Hemerocoetes.

Family 2. Heterolepidotidoe.— Body oblong, compressed, scaly; eyes lateral; cleft of the mouth lateral; dentition feeble. The angle of the praeoperculum connected by a bony stay with the infraorbital ring. Dorsal long, with the spinous and soft portions equally developed; anal elongate, Ventrals thoracic, with one spine and five rays.

Genera: Chirus, Ophiodon, Agrammus, Zaniolepis.

Family3. Blenniidae. — Body elongate, low, more or less cylindrical, naked or covered with scales, which generally are small. One, two, or three dorsal fins occupying nearly the whole length of the back, the spinous portion, if distinct, being as much developed as the soft, or more; sometimes the entire fin is composed of spines; anal fin long. Ventrals jugular, composed of a few rays, and sometimes rudimentary or entirely absent. Pseudobranchiae generally present.

Genera: Anarrhichas (Wolf-Fish), Blennius (Blenny), Chasmodes, Petroscirtes, Salarias, Clinus, Gristiccps, Cremnobates, Tripterygium, Stichceus, Blenniops, Centronotus, Xiphidion, Cryptacanthodes, Patoecus, Zoarces, Blennophis, Ncmophis, Plagiotremus, Neoclinus, Cebidichthys, Myxodes, Heterostichus, Dictyosoma, Lepidoblenniv», Dactyloscopus, Gunellichthys, Urocentrus, Siichceopsis, Stichariurn, Notograptus, Pholidichthys, and Pseudoblennius.— Closely allied is Acanthoclinus.

Family 4. Mastacembelidce.— Body elongate, eel-like, covered with very small scales. Mandible long, but little movable. Dorsal fin very long, the anterior portion composed of numerous short isolated spines; anal fin with spines anteriorly. Ventrals none. The humeral arch is not suspended from the skull. Gillopenings reduced to a slit at the lower part of the side of the head.

Genera: Rhynchobdella and Mastacembelus. Division XI. Acanthopterygii Mugiliformes.— Two dorsal fins more or less remote from each other; the anterior either short, like the posterior, or composed of feeble spines. Ventral fins with one spine and five rays, abdominal.

Family 1. Sphyroenidce.— Body elongate, sub-cylindrical, covered with small cycloid scales; lateral line continuous. Cleft of the mouth wide, armed with strong teeth. Eye lateral, of moderate size. Vertebrae twenty-four.

Genera: Sphyrcena (Barracuda), Fossil; Sphyroenodus, Hypsodon, Portheus, Saurocephalus.

Family 2. Athcrinidoe. — Body more or less elongate, sub-cylindrical, covered with scales of moderate size; lateral line indistinct. Cleft of the mouth of moderate width, with the dentition feeble.

Eye lateral, large or of moderate size. Gill-openings wide. Vertebrae very numerous.

Genera: Athcrina, Atherinichlhys, and Tctragonurus.

Family 3. Mugilidce (Grey Mullets).—Body more or less oblong and compressed, covered with cycloid scales of moderate size; lateral line none. Cleft of the mouth narrow or of moderate width, with or without feeble teeth. Eye lateral, of moderate size. Gill-opening wide. The anterior dorsal fin composed of four stiff spines. Vertebrae twenty-four.

Genera: Mugil, Agonostoma, and Myxus. Division XII. Acanthopterygii Gastrosteiformes.— The spinous dorsal is composed of isolated spines, if present; the ventrals are either thoracic or have an abdominal position in consequence of the prolongation of the pubic bones which are attached to the humeral arch. Mouth small, at the end of the snout, which is generally more or less produced.

Family 1. Gastrostcidas.~ Body elongate, compressed. Cleft of the mouth oblique; villiform teeth in the jaws. Opercular bones not armed; infraorbitals covering the cheek; parts of the skeleton forming incomplete external mail. Scales none, but generally large scutes along the side. Isolated spines in front of the soft dorsal fin. Ventral fins abdominal, joined to the pubic bone, compose 0 ! of a spine and a small ray. Branchiostegals three.

One genus only: Gastrosteus (Sticklebacks).

Family 2. Fistular iidae.— Fishes of greatly elongated form; the anterior bones of the skull are much produced, and form a long tube, terminating in a narrow mouth. Teeth small; scales none, or small. The spinous dorsal fin is either formed by feeble isolated spines or entirely absent; the soft dorsal and anal of moderate length; ventral fins thoracic or abdominal, composed of five or six rays, without spine; if abdominal, they are separate from the pubic bones, which remain attached to the humeral arch. Branchiostegals five.

Genera: Fistularia, Aulostorna, Auliscoρs, and Aulorhynchus. The first three occur also in Eocene formations. Other fossil genera are Urosphen and Rhamphosus. Division XIII. Acanthopterygii Centrisciformes.— Two dorsal fins; the spinous short, the soft and the anal of moderate extent. Ventral fins truly abdominal, imperfectly developed.

One family, Gentriscidoe, with two genera, Ccntriscus (Trumpet-Fish, Bellows-Fish) and Amphisile. Division XIV. Acanthopterygii Gobiesociformes.— No spinous dorsal; the soft and the anal short or of moderate length, situated on the tail; ventral fins subjugular, with an adhesive apparatus between them. Body naked.

Genera: Chorisochismus, Cotylis, Sicyases, Gobiesox, Diρlocrcpis, Crepidogaster, Trachelochismus, Lepadogaster, and Leptopterygius. Division XV. Acanthopterygii Channiformes.— Body elongate, covered with scales of moderate size; no spine in any of the fins; dorsal and anal long. No superbranchial organ, only a bony prominence on the anterior surface of the hyomandibular.

Genera: Ophiocephalus and Channa. Division XVI. Acanthopterygii Labyrinthibranchii.— Body compressed, oblong or elevated, with scales of moderate size. A superbranchial organ in a cavity accessory to that of the gills.

Family 1. Labyrinthici. — Dorsal and anal spines present, but in variable number; ventrals thoracic. Lateral line absent, or more or less distinctly interrupted. Gill-opening rather narrow, the gill-membranes of both sides coalescent below the isthmus, and scaly; gills four; pseudobranchiae rudimentary or absent.

Genera: A nabas, Spirobranchus, Ctenoρoma, Polyacanthus, (Paradise-Fish), Osρhromemιs (Goramy), Trichogaster, Belta, and Micracanthus.— Allied to this family is Luciocephalus. Division XVII. Acanthopterygii Lophotiformes.— Body ribbonshaped, with the vent near its extremity; a short anal behind the vent; dorsal fin as long as the body.

Only one genus is known of this division or family, Lophotes. Division XVIII. Acanthopterygii Taeniiformes (Ribbon-Fishes). —Body ribbon-shaped; dorsal fin as long as the body; anal absent; caudal rudimentary, or not in the longitudinal axis of the fish.

Genera: Trachypterus, Skyloρhorus, and Regalecus. Division XIX. Acanthopterygii Notacanthiformcs.— Dorsal fin short, composed of short, isolated spines, without a soft portion. Anal fin very long, anteriorly with many spines; ventrals abdominal, with more than five soft and several unarticulated rays.

One genus only: Notacanthus. Order II.— Acanthopterygii Pharyngognatlιi.

Part of the rays of the dorsal, anal, and ventral fins are nonarticulated spines. The lower pharyngeals coalesced. Air-bladder without pneumatic duct.

Family 1. Pomacentridae. — Body short, compressed, covered with ctenoid scales. Dentition feeble; palate smooth. The lateral line does not extend to the caudal fin, or is interrupted. One dorsal fin, with the spinous portion as well developed as the soft, or more. Two, sometimes three, anal spines; the soft anal similar to the soft dorsal. Ventral fins thoracic, with one spine and five soft rays. Gills three and a half; pseudobranchiae and air-bladder present. Vertebrae, twelve abdominal and fourteen caudal.

Genera: Amphiprion, Premnas, Dascyllus, Lepidozygus, Pom· acentrus, Glyphidodon, Parma, and Heliastes. Fossil: Odonteus.

Family 2. Labridce.- Body oblong or elongate, covered with cycloid scales. The lateral line extends to the caudal, or is interrupted. One dorsal fin, with the spinous portion as well developed as the soft, or more so. The soft anal similar to the soft dorsal. Ventral fins thoracic, with one spine and five soft rays. Γalate without teeth. Branchiostegals five or six; gills three and a half; pseudobranchiae and air-bladder present. Pyloric appendages none; stomach without caecal sac.

Genera: Labrusι ∖ nd Crenilabrus (Wrasses), Tautoga, Ctenolabrus, Acantholabrus, Centrolabrus, Lachnolaemus, Malacoρterus,Cossyρhus, Chilinus, Eρibulus, Anampses, Platyglossus, Novacula, Julis, Coris, Choerops, Xiρhochilus, Semicossyphus, Trochocoρus, Decodon, Ptcr-agogus, Clepticus, Labrichthys, Labτoides, Duymoeria, Cirrhilabrus, Doratθ)iotus, Pseudochilimιs, Hemigymnus, Gomphosus, Cheilio, Cymolutes, Pseudodax, Scarus, Scarichthys, Callyodon, Pseudoscarus, Odax, Coridodax, Olistherops, and Siphonognathus. Fossil: Nummopalatus, Phyllodus, Taurinichthys, and Egertonia.

Family 3. Embiotocidae. — Body compressed, elevated or oblong, covered with cycloid scales; lateral line continuous. One dorsal fin, with a spinous portion, and with a scaly sheath along the base, which is separated by a groove from the other scales; anal with three spines and numerous rays; ventral fin thoracic, with one spine and five rays. Small teeth in the jaws, none on the palate. Pseudobranchiae present. Stomach siphon al; pyloric appendages none. Viviparous.

Genera: Ditrema and Hysterocarpus.

Family 4. Chromides.— Body elevated, oblong orelongate⅛ scaly, the scales being generally ctenoid. Lateral line interrupted or nearly so. One dorsal fin, with a spinous portion; three or more anal spines; the soft anal similar to the soft dorsal. Ventral fins thoracic, with one spine and five rays. Teeth in the jaws small, palate smooth. Pseudobranchiae none. Stomach caecal; pyloric appendages none.

Genera: Etroplus, Chromis, Hemichromis, Paretroplus, Acara, Heros, Neetroplus, Mesonauta, Petenia, Uaru, Hygrogonus, Cichla, Crenicichla, Chaetobranchus, Mesops, Satanoperca, Geophagus, Sym-physodon, and Pterophyllwm. Order III.— Anacanthini.

Vertical and ventral fins without spinous rays. The ventral fins, if present, are jugular or thoracic. Air-bladder, if present, without pneumatic duct.

Division I. Anacanthini Gadoidti.— Head and body symmetrically· formed.

Family 1. Lycodidae. — Vertical fins confluent. Ventral fin, if present, small, attached to the humeral arch, jugular. Gill-opening narrow, the gill-membrane being attached to the isthmus.

Genera: By codes, Gymnelis, Uronectes, Microdesmus, Blen-nodesmus, and Maynea.

Family 2. Gadidae. — Body more or less elongate, covered with small smooth scales. One, two, or three dorsal fins, occupying nearly the whole of the back; rays of the posterior dorsal well developed; one or two anal fins. Caudal free from dorsal and anal, or, if they are united, the dorsal with a separate anterior portion. Ventrals jugular, composed of several rays, or, if they are reduced to a filament, the dorsal is divided into two. Gillopening wide; the gill-membranes generally not attached to the isthmus., Pseudobranchiae none, or glandular, rudimentary. An air-bladder and pyloric appendages generally present.

Genera: Gadus (Cod-Fish, Haddock, Whiting, Pollack, Coal-Fish), • Gadiculιιs, Mora, Strinsia, Halargyreus, Melanonus, Mcrluccius (Hake), Pseudophycis, Lotella, Physiculus, Uraleptus, Loemonema, Phycis, Haloporphyrus, Lota (Burbot), Molνa (Ling), Motella (Rockling), Raniceρs, Bregmaceros, Muraenolepis, Chiasmodus, and Brosmius. Fossil remains are ' rare: Nernoptcryx and Paloeogadvs from the schists of Glarus, a formation believed to have been the bottom of a very deep sea. In the clay of Sheppey species occur allied to Gadus, Merluccius, and Phycis.

Family 3. Ophidiidae. — Body more or less elongate, naked or scaly. Vertical fins generally united; no separate anterior dorsal or anal; dorsal occupying the greater portion of the back. Ventral fins rudimentary or absent, jugular. Gill-openings wide, the gill-membranes not attached to the isthmus.

1. Brotulina∙. Ventral fins present, attached to the humeral arch. Genera: Brotula, Lucifuga, Bathynectcs, Acanthonus, Typhlonus, Aρhyonus, Rhinonus, Sirembo, Pteridium, Brotulophis, Halidesmus, Dinematichthys, and Bythites.

2. Ophidiina: Ventral fins replaced by a pair of bifid filaments (barbels) inserted below the glossohyal. Genera: Ophidùvm and Genypterus.

3. Ficrasferina : No ventral fins whatever; vent at the throat. Genera: Ficrasfcr and Encheliophis.

4. Ammodytina∙. No ventral fins whatever; vent remote from the head; gill-openings very wide, the gill-membranes not being united. Genera: Ammodyics (Sand-Eels) and Bleekeria. 5 . Congrogadina : No ventral fins whatever; vent remote from the head; gill-openings of moderate width, the gill-membranes being united below the throat, not attached to the isthmus.. Genera: Congrogadus and Haliophis.

Family 4. Macmridae. — Body terminating in a long, compressed, tapering tail, covered with spiny, keeled, or striated scales. One short anterior dorsal; the second very long, continued to the end of the tail, and composed of very feeble rays; anal of an extent similar to that of the-second dorsal; no caudal. Ventral fins thoracic or jugular, composed of several rays. Deep-sea Gadoids.

Genera: Λfacrurus, Coryphaenoides, Macruromιs, Malacocephahιs, and Bathygadus. Division II. Anacanthini Pleu,ronecloidei.— Head and part of the body unsymmetrically formed. This division consists of one family only: Pleuronectidae (Flat-Fishes).

Genera: Psettodes, Hippoglossus (Holibut), Hippoglossoides, Tephritis, Rhombus (Turbot, Brill, Whiff), Phrynorhombus (TopKnot), Arnoglossus, Pseudorhomirus, Rhomboidichthys, Citharus, Anticitharus, Brachypleura, Samaris, Psettichthys, Githarichthys, Hemirhombus, Paralichthys j Liopsetta, Lophonectes, Lepidopsetla, Thysanopselta, Pleuronectes (Plaice, Dab, Flounder), Rhombosolea, Parophrys, Psammodiscus, Amrnotretis, Pellorhamphus, Nemalops, Laeops, Poedlopsetla, Solea (Sole), Synaptura, Aesopia, Gymnachirus, Cynoglossus, Soleotalpa, Apionichthys, Ammopleurops, Aρhoristia, and Plagusia Order IV.— Physoslomi.

All the fin-rays articulated, only the first of the dorsal and pectoral fins is sometimes ossified. Ventral fins, if present, abdominal, without spine. Air-bladder, if present, with a pneumatic duct (except in Scombresoddae).

Family 1. Siluridae. — Skin naked or with osseous scutes, but without scales. Barbels always present; maxillary bone rudimentary, almost always forming a support to a maxillary barbel. Margin of the upper jaw formed by the intermaxillaries only. Suboperculum absent. Air-bladder generally present, communicating with the organ of hearing by means of the auditory ossicles. Adipose fin present or absent.

A large family of freshwater fishes, represented by numerous genera, which exhibit a great variety of form and structure of the fins. Their first appearance is indicated by some fossil remains in Tertiary deposits of the highlands of Padang in Sumatra, where Pseudeutropius and Bagarius, types well represented in the living Indian fauna, have been found. In North America also spines referable to “Cat-Fishes” have been found in Tertiary formations.

From the great number of different generic types this family has 6een arranged under eight subdivisions :—

1. Siluridae Homalopteroe.

A. Glariina : Clarias, Heterobranchus.

B. Plotosina : Plotosus, Cnidoglanis, Copidoglanis, Chaca.

2. Siluridae Heteropterae.

A. Silurina : Saccobranchus, Silurus, Schilbe, Eutropius, Silurichthys, Wallago, Belodontichthys, Eutropiichthys, Cryptopterus, Callichrous, Hemisilurus, Siluranodon, Ailia, Schilbichthys, Lais, Pseudeutropius, Pangasius, Helicophagus, Silondia.

3. Siluridae Anomalopteroe.

A. Hypophthalmina : Hypophthalmus, Helogenes.

4. Siluridae Proteroplerae.

A. Bagrina : Bagrus, Chrysichthys, Clarotes, Macrones, Pseudobagrus, Liocassis, Bagroides, Bagrichthys, Rita, Acroehordonichthys, Akysis.

B. Amiurina : Amiurus, Hopladelus, Noturus.

C. Pimelodina : Platystoma, Sorubim, Hemisorubim, Platystomatichthys, Phractocephalus, Piramutana, Platy - nematichthys, Piratinga, Bagropsis, Sciades, Pimelodus, Pirinampus, Conorhynchus, Notoglanis, Callophysus, Lophiosilurus, Auchenoglanis.

D. Ariina: Arius, Galeichthys, Genidens, Paradiplomystax, Diplomystax, Aelurichthys, Hemipimelodus, Ketengus, Osteogeniosus, Batrachocephalus, Atopochilus.

E. Bagariina∙. Bagarius, Euglyptosternum, Glyptosternum.

5. Siluridae Stenobranchiae.

A. Borodina : Ageniosus, Tetranematichthys, Euanemus, Auchenipterus, Glanidium, Centromochlus, Traehelyo-pterus, Cetopsis, Astrophysus, Doras, Synodontis.

B. Rhinoglanina : Rhinoglanis, Callomystax.

C. Malapternrina∙. Malapterurus.

6. Siluridae Proterop<>des. A . HypostomaHna : Stygogenes, Arges, Brontes, Astrople-bus, Callichthys, Chaetostomus, Plecostomus, Hypopto-poma, Loricaria, Acestra, Sisor, Erethistes, Exostoma, Pseudecheneis.

B. Asprβdinima : Aspredo, Bunocephalus, Bunocephal-ichthys, Harttia.

7. Siluridae Opisthopterae.

A. Nematogenyina∙. Heptapterus, Nematogenys.

B. Trichomycterina : Trichomycterus, Eremophilus, Pari-odon.

8. Siluridae Branchicoloe : Stegoph.ilus, Vandellia.

Family 2. Scopelidae. — Body naked or scaly. Margin of the upper jaw formed by the intermaxillary only; opercular apparatus sometimes incompletely developed. Barbels none. Gill-opening very wide; pseudobranchiae well developed. Air-bladder none. Adipose fin present. The eggs are enclosed in the sacs of the ovary, and excluded by oviducts. Pyloric appendages few in number or absent. Intestinal tract very short.

Genera: Saums, Bathysaums, Bathypterois, Harpodon, Scopelus, Ipnops, Paralepis, Sudis, Plagyodus, Aulopus, Chlorophthalmus, Scopelosaurns, Odontostomus, and Nannobrachium. Fossil: Osrne-roides, Hemisawida, Parascopelus, and Anapterus.

Family 3. Cyprinidae. — Body generally covered with scales; head naked. Margin of the upper jaw formed by the intermaxillaries. Belly rounded, or, if trenchant, without ossifications. No adipose fin. Stomach without blind sac. Pyloric appendages none. Mouth toothless; lower pharyngeal bones well developed, falciform, subparallel to the branchial arches, provided with teeth, which are arranged in one, two, or three series. Air-bladder large, divided into an anterior and posterior portion by a constriction, or intoa rightor a left portion, enclosed in an osseous capsule. Ovarian sacs closed.

The family of ‘ ‘ Carps” is the one most numerously represented in the fresh waters of the Old World and of North America. Numerous fossil remains are also found in Tertiary freshwater formations; the majority can be referred to existing genera: Barbus, Thynn∙ ichthys, Gobio, Leuciscus, Tinea, Amblypharyngodon, Rhodeτιs, Cobitis, Acanihopsis ; only a few showing characters different from those of living genera: Cyclurns, Hexapsephus, Mylocyprinus.

There is much less diversity of forms and habits in this family than in the Siluroids; but the genera are sufficiently numerous to demand a further subdivision z 'f ⅛e family into groups.

1. Catosiomina : Catostomus, Moxostoma, Sclerognathus, Car-piodes.

2. Cyprinina: Cyprinus (Carp), Carassius, Catla, Cirrhina, Dangila, Osteochilus, Labeo, Barynotus, Tylognathus, Abrostomus, Discognathus, Crossochilus, Gymnostomus, Epalzeorhynchus, Capoeta, Barbus (Barbel), Thynnichthys, Barbichthys, Ambly-rhynchichthys, Albulichthys, Oreinus, Schizothorax, Ptychobarbus, Gymnocypris, Schizopygopsis, Diptychus, Aulopyge, Gobio (Gudgeon), Ceratichthys, Bungia, Pimephales, Hyborhynchus, Carn-postoma, Hybognathus, Ericymba, l ’ seu,dorasbora, Cochlognathus, Exoglossum, Rhinichthys.

3. Rhoteichthyina : Rhoteichthys.

4. Leptobarbina : Leptobarbus.

δ. Rasborina : Rasbora, Luciosoma, Nuria, Aphyocypris, Amblypharyngodon.

6. Semiplotina : Cyprinion, Semiplotus.

7. Xenocypridina: Xenocypris, Paracanthobrama, Mystaco∙ leucus.

8. Leuciscina : Leuciscus (White-Fish, Roach, Chub, Dace, Rudd, Minnow), Myloleucus, Ctenopharyngodon, Mylopharodon, Para-ρhoxinus, Meda, Graodus, Tinea (Tench), Leucosomus, Chondrostoma, Orthodon, Acrochilus.

9. Rhodeina : Achilognathus, Rhodeus, Pseudoperilampus.

10. Banionina∙. Danio, Pteropsarion, Aspidoparia, Barilius, Bola, Schacra, Opsariichthys, Squaliobarbus, Ochetobius.

11. Hypophihalmichthyina : Hypophthalmichthys.

12. Abramidina : Abramis (Bream), Aspius, Alburnus (Bleak), Leucaspius, Rasborichthys, Elopiehthys, Pelotrophus, Acantho-brama, Osteobrama, Chanodichthys, Smiliogaster, Culter, Pelecus, Eustira, Chela, Pseudolabuca, Cachius.

13. Hornalopterina : Homaloptera, Psilorhynchus.

14. Cobitidina (Loaches): Misgurnus, Nemachilus, Cobitis, Lepidocephalichthys, Acanthopsis, Botia, Oreonectes, Lepido-cephalus, Acanthophthalmus, Apua, Paramisgurnus.

Family 4. Kneriidae. — Body scaly, head naked. Margin of the upper jaw formed by the intermaxillaries. Dorsal and anal fins short, the former belonging to the abdominal portion of the vertebral column. Teeth none, either in the mouth or pharynx. Barbels none. Stomach siphonal; no pyloric appendages. Pseudobranchiae none. Branchiostegals three; air-bladder long, not divided. Ovaries closed.

One genus: Kneria. ,

Family 5. Characinidoe. — Body covered with scales, head naked; barbels none. Margin of the upper jaw formed by the intermaxillaries in the middle and by the maxillaries laterally. Generally a small adipose fin behind the dorsal. Pyloric appendages more or less numerous; air-bladder transversely divided into two portions, and communicating with the organ of hearing by means of the auditory ossicles. Pseudobranchiae none.

1. Erythrinina : Macrodon, Erythrinus, Lebiasina, Nannostoraus, Pyrrhulina, and Corynopoma.

2. Curimatina : Curimatus, Prochilodus, Caenotropus, Hemiodus, Saccodon, Parodon.

3. Citharinina : Citlιarinus.

4. Anaetomatina : Leporinus, Anostomus, Rhytiodus.

5. Nannocharacina : Nannocharax.

6. Tetragonopterina : Alestes, Tetragonopterus, Chirodon, Megalo-brycon, Gastropelecus, Piabucina, Scissor, Pseudochalceus, Aphyo-charax, Chalceus, Brycon, Chalcinopsis, Bryconops, Creagrutus, Chalcinus, Piabuca, Paragoniates, Agoniates, Nannaethiops, and Bryconaethiops.

7. Hydrocyonina : Hydrocyon, Cynodon, Anacyrtus, Hystri-co Ion, Salminus, Oligosarcus, Xiphorhamphus, Xiphostoma, and Sarcodaces.

8. Distichodontina : Distichodus.

9. Ichthyborina : Ichthyborus and Phago.

10. Crenuchina : Crenuchus and Xenocharax.

11. Serrasalmonina : Mylesinus, Serrasalmo, My letes, and Cato-prion.

Family 6. Cyprinodontidae. — Head and body covered with scales; barbels none. Margin of the upper jaw formed by the intermaxillaries only. Teeth in both jaws; upper and lower pharyngeals with cardiform teeth. Adipose fin none; dorsal fin situated on the hinder half of the body. Stomach without blind sac; pyloric appendages none. Pseudobranchiae none; air-bladder simple, without ossicula auditus.

1. C. Carnivores: Cyprinodon, Characodon, Fitzroyia, Haplochilus, Fundulus, Limnurgus, Lucania, Rivulus, Cynolebias, Orestias, Jenynsia, Gambusia, Pseudoxiphophorus, Belonesox, Anableps.

2. C. Limnophagae: Poecilia, Mollienesia, Platypoecilus, Girardinus. Fossil remains are referable to Cyprinodon and Poecilia.

Family 7. HeteroρygU.— Head naked; body covered with very small scales; barbels none. Margin of the upper jaw formed by the intermaxillaries. Villiform teeth in the jaws and on the palate. Adipose fin none. Dorsal fin belonging to the caudal portion of the vertebral column, opposite to the anal. Ventral fins rudimentary or absent. Vent situated before the pectorals. Stomach caecal; pyloric appendages present. Pseudobranchiae none; air-bladder deeply notched anteriorly.

Genera: Amblyopsis (Blind-Fish of the Mammoth Cave) and Chologaster.

Family 8. Umbridae. — Head and body covered with scales; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Adipose fin none; the dorsal fin belongs partly to the abdominal portion of the vertebral column. Stomach siphonal; pyloric appendages none; pseudobranchiae glandular, hidden; air-bladder simple.

One genus: Umbra.

Family 9. Scombresocidce.— Body covered with scales; a series of keeled scales along each side of the belly. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Lower pharyngeals united into a single bone. Dorsal fin opposite the anal, belonging to the caudal portion of the vertebral column. Adipose fin none. Air-bladder generally present, simple, sometimes cellular, without pneumatic duct. Pseudobranchiae hidden, glandular. Stomach not distinct from the intestine, which is quite straight, without appendages.

Genera: Belonc (Gar-Pike), Scombrcsox, Hemirhamρhus, Ar-rh,amρhus, Exocoetus (Flying-Fish). Fossil: Holostcus.

Family 10. Esocidoe.— Body covered with scales; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Adipose fin none; the dorsal fin belongs to the caudal portion of the vertebral column. Stomach without blind sac; pyloric appendages none. Pseudobranchiae glandular, hidden; air-bladder simple; gill-opening very wide.

One genus only: Esox (Pike).

Family 11. Galaxüdae — Body naked; barbels none. Margin of the upper jaw chiefly formed by the intermaxillaries, which are short, and continued by a thick lip, behind which are the maxillaries. Belly rounded; adipose fin none; dorsal opposite to anal. Pyloric appendages in small number. Air-bladder large, simple; pseudobranchiae none. The ova fall into the cavity of the abdomen before exclusion.

Genera: Galaxias and Ncochanna.

Family 12. Mormyridae. Body and tail scaly; head scaleless; barbels none. The margin of the upper jaw is formed in the middle by the intermaxillaries, which coalesce into a single bone, and laterally by the maxillaries. Suboperculum and interoperculum present, the latter very small. On each side of the single parietal bone a cavity leading into the interior of the skull, and covered with a thin bony lamella. All the fins are well developed (Mormyras) ; or caudal, anal, and ventral fins are absent (Gym· narchus). No adipose fin. Pseudobranchiae none; gill-openings reduced to a short slit. Air-bladder simple. Two caeca pylorica behind the stomach.

Genera: Mormyrus and Gyrnnarchus.

Family 13. Stcrnoptychidce.— Body naked, or with very thin •deciduous scales; barbels none. Margin of the upper jaw formed by the maxillary and intermaxillary, both of which are toothed; opercular apparatus not completely developed. Gill-opening very wide; pseudobranchiae present or absent; air-bladder simple, if present. Adipose fin present, but generally rudimentary 7 . Series of phosphorescent bodies along the lower parts. The eggs are enclosed in the sacs of the ovarium, and excluded by oviducts.

Genera: Sternoptyx, Argyroρelccus, Polyiρnus, Coccia, Mauro· lieux, Chauliodus, Gonostoma, Photichthys, and Diplophos.

Family 14. Stomiatidae. — Skin naked, or with exceedingly delicate scales; a hyoid barbel. Margin of the upper jaw formed by the intermaxillary and maxillary, which are both toothed; opercular apparatus but little developed. Gill-opening very wide; pseudobranchiae none. The eggs are enclosed in the sacs of the ovarium, and excluded by oviducts.

Genera: Astronesthes, Stomias, Echiostorna, Malacostcus, and Bathyophis.

Family 15. Salmonidae. — Body generally covered with scales; head naked; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Belly rounded. A small adipose fin behind the dorsal. Pyloric appendages generally numerous, rarely absent. Air-bladder large, simple; pseudobranchiae present. The ova fall into the cavity of the abdomen before exclusion.

Genera: Salmo (Salmon, Trout, Charr), Oncorhynchus, Brachy∙ mystax, Luciotrutta, Plecoglossus, 0smerux (Smelt), Retropinna, Hypornesus, Thalcichthys, Mallotus (Capelin), Coregonus (Gwyniad, Pollan), Thymallus (Grayling), Salanx, Argentina, Microstoma, Bathylagus. Fossil: Osmeroides, Acrognathus, and Aulolepis.

Family 16. Percopsidoe. — Body covered with ctenoid scales; head naked. Margin of the upper jaw formed by the intermaxillaries only; opercular apparatus complete. Barbels none. Gill-openings wide. Adipose fin present.

One genus only: Percopsis.

Family 17. Haplochitonidae. — Body naked or scaly (cycloid). Margin of the upper jaw formed by the intermaxillary; opercular apparatus complete. Barbels none. Gill-opening wide; pseudobranchiae. Air-bladder simple. Adipose fin present. Ovaries laminated; the eggs fall into the cavity of the abdomen, there being no oviduct. Pyloric appendages none.

Genera: Haplochiton and Prototroctes.

Family 18. Gonorhynchidoe. — Head and body entirely covered with spiny scales; mouth with barbels. Margin of the upper jaw formed by the intermaxillary, which, although short, is continued downwards as a thick lip, situated in front of the maxillary. Adipose fin none; the dorsal fin is opposite to the ventrals, and short, like the anal. Stomach simple, without blind sac; pyloric appendages in small number. Pseudobranchiae; air-bladder absent. Gill-openings narrow.

One genus only: Gonorhynchus.

Family 19. Hyodontidas. — Body covered with cycloid scales; head naked; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally, the latter being articulated to the end of the former. Opercular apparatus complete.. Adipose fin none; the dorsal fin belongs to the caudal portion of the vertebral column. Stomach horseshoe-shaped, without blind sac; intestine short; one pyloric appendage. Pseudobranchiae none; air-bladder simple. Gill-openings wide. The ova fall into the abdominal cavity before exclusion.

One genus only: Hyodon ( Moon-Eye).

Family 20. Pantodontidoe. — Body covered with large cycloid scales; sides of the head osseous. Margin of the upper jaw formed by the single intermaxillary mesially and by the maxillaries laterally. The dorsal fin belongs to the caudal portion of the vertebral column, is short, and opposite and similar to the anal. Gillopenings wide; gill-covers consisting of a praeoperculum and operculum only. Branchiostegals numerous. Pseudobranchiae none; air-bladder simple. Stomach without caecal sac; one pyloric appendage. Sexual organs with a duct.

One genus: Pantodon.

Family 21. Osteoglossidoe. — Body covered with large hard scales, composed of pieces like mosaic. Head scaleless; its integuments almost entirely replaced by bone; lateral line composed of wide openings of the mucous duct. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. The dorsal fin belongs to the caudal portion of the vertebral column, is opposite and very similar to the anal fin; both approximate to the rounded caudal (with which they arc abnormally confluent). Gillopenings wide; pseudobranchiae none; air-bladder simple or cellular. Stomach without caecal sac; pyloric appendages two.

Genera: Osteoglossum, Arapaima, and Heteτotis.

Family 22. Clupeidae. — Body covered with scales; head naked; barbels none. Abdomen frequently compressed into a serrated edge. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally; maxillaries composed of at least three movable pieces. Opercular apparatus complete. Adipose fin none; dorsal not elongate; anal sometimes very long. Stomach with a blind sac; pyloric appendages numerous. Gill-apparatus much developed, the gill-openings being generally very wide. Pseudobranchiae generally present. Air-bladder more or less simple.

Genera: Engraulis (Anchovies), Coilia, Chatoessus, Clupea (Herring. Sprat, Shad, Mossbanker, Menhaden, Ale-Wife, Pilchard, Sar- , dine), Cluρeoides, Pellonula, Clupeichthys, Bellona, Bristigaster, Albida, Elops, Megalops, Chanos, Dussumieria, Etrumeus. Several *,of these genera have been found in Tertiary formations; other fossils are: Thrissopater, Leptosomus, Opisthopteryx, Spaniodon, Halec, Platinx, Coelogaster, Rhinellus, Scombroclupea, Crossognathus, Sρathodactylus, Chirocentrites, and Hemitrichas.

Family 23. Bathythrisidae. — Allied to the herrings— Balhythrissa.

Family 24. Chirocenkridae. — Body covered with thin, deciduous scales; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally, both bones being firmly united in juxtaposition. Opercular apparatus complete. Adipose fin none: the dorsal fin belongs to the caudal portion of the. vertebral column. Stomach with a blind sac; intestine short, the mucous membrane forming a spiral fold; pyloric appendages none. Pseudobranchiae none; air-bladder incompletely divided into cells; gill-opening wide.

One genus only: Chirocentrus.

Family 25. Alepocephalidoe. — Body with or without scales; head naked; barbels none. Margin of the upper jaw formed by the intermaxillaries and maxillaries, the former being placed along the upper anterior edge of the latter. Opercular apparatus complete. Adipose fin none; the dorsal fin belongs to the caudal portion of the vertebral column. Stomach curved, without blind sac; pyloric appendages in moderate number. Pseudobranchiae; air-bladder absent. Gill-openings very wide.

Genera: Alepocephalus, Bathytroctes, Platytroctes, and Xenoderm-ichthys.

Family 26. Notopteridoe. — Head and body scaly; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Opercular apparatus incomplete. Tail prolonged, tapering. Adipose fin none. Dorsal short, belonging to the caudal portion of the vertebral column; anal very long. Stomach without blind sac; two pyloric appendages. Pseudobranchiae none; air-bladder present, divided in the interior. The ova fall into the cavity of the abdomen before exclusion. On each side a parieto-mastoid cavity leading into the interior of the skull.

One genus only: Notopterus.

Family 27. Halosauridce. — Body covered with cycloid scales; head scaly; barbels none. Margin of the upper jaw formed by the intermaxillaries mesially and by the maxillaries laterally. Opercular apparatus incomplete. Adipose fin none. The short dorsal belongs to the abdominal part of the vertebral column; anal very long. Stomach with a blind sac; intestine short; pyloric appendages in moderate number. Pseudobranchiae none. Air-bladder large, simple; gill-openings wide. Ovaries closed.

One genus only: Halosaurus.

Family 28. Hoplopleuridoe.— Body generally with four series of subtriangular scutes, and with intermediate scale-like smaller ones. One (?) dorsal only; head long, with the jaws produced.

Extinct. Genera: Dercetis, Leptotrachelus, Pelargorhynchus, Plinthophorus, Saurorhamphus, Eurypholis, Ischyrocephalus.

Family 29. Gyrnnotidoe. — H ead scaleless; barbels none. Body elongate, eel-shaped. Margin ofthe upper jaw formed in themiddle bythe intermaxillaries and laterally by the maxillaries. Dorsal fin absent or reduced to an adipose strip; caudal generally absent, the tail terminating in a point. Anal fin exceedingly long. Ventrals none. Extremity of the tapering tail capable of being reproduced. Vent situated at, or at a short distance behind, the throat. Humeral arch attached to the skull. Ribs well developed. Gill-openings rather narrow. Air-bladder present, double. Stomach with a caecal sac and pyloric appendages. Ovaries with oviducts.

Genera: Sternarchus, Ramphichthys, Sternopygus, Carapus, Gymιwtus (Electric Eel).

Family 30. Symbranchidcκ.— Body elongate, naked or covered with minute scales; barbels none. Margin of the upper jaw formed by the intermaxillaries only, the well÷developed maxillaries lying behind and parallel to them. Paired fins none. Vertical tins rudimentary, reduced to more or less distinct cutaneous folds. Vent situated at a great distance behind the head. Ribs present. Gillopenings confluent into one slit situated on the ventral surface. Air-bladder none. Stomach without caecal sac or pyloric appendages. Ovaries with oviducts.

Genera: Amphipnous, Monopterus, Symbranchus, Chilobranchus.

Family 31. Muraenidae (Eels).-—Body elongate, cylindrical or bandshaped, naked or with rudimentary scales. Vent situated at a great distance from the head. Ventral fins none. Vertical fins, if present, confluent, or separated by the projecting tip of the tail. Sides of the upper jaw formed by the tooth-bearing maxillaries, the fore part by the intermaxillary, which is more or less coalescent with the vomer and ethmoid. Humeral arch not attached to the skull. Stomach with a blind sac; no pyloric appendages. Organs vf reproduction without efferent ducts.

Genera: Nemichthys, Cyema, Saccopharynx, Synaphobranchus, Anguilla, Conger, Congromuraena, Muraenesox, Nettastoma, Sauren- chelys, Oxyconger, Hoplunnis, Myrus, Myrophis, Paramyrus, Chilo- rhinus, Muraenichthys, Ophichthys, Moringua, Muraena, Enchelycore. Order V.— Lophobranchii. .

The gills are not laminated, but composed of small rounded lobes attached to the branchial arches. Gill-cover reduced to a large simple plate. Air-bladder simple, without pneumatic duct. A dermal skeleton, composed of numerous pieces arranged in segments, replaces more or less soft integuments. Muscular system not much developed. Snout prolonged. Mouth terminal, small, toothless, formed as in Acanthopterygians.

Family 1. Solenostomidoe. — Gill-openings wide. Two dorsal fins, the rays of the anterior not articulated. All the other fins well developed.

One genus only: Solenostoma, which was preceded in the Tertiary epoch by Solenorhynchus (Monte Postale).

Family 2. Syngnathidae. — Gill-openings reduced to a very small opening near the upper posterior angle of the gill-cover. One soft dorsal fin; no ventrals, and sometimes one or more of the other fins are also absent.

Pipe-Fishes and Seahorses. Fossil remains occur at Monte Bolca. Besides species of Biphonostoma and Syngncdhus (Pseudosyngnathus) r remains of an extinct genus, Calamostoma, allied to Hippocampus, but with a distinct caudal fin, have been found.

Genera: Siphonostoma, Syngncdhus, Doryichthys, Ncrophis r Protocampus, Ichthyocampus, Nannocampus, Urocampus, Lepto-ichthys, Coelonotus, Stigmatophora, Gastrotokeus, Solenognathus, Phylloptcryx, Hippocampus. Order VI. — PlectognatM.

Teleosteous fishes with rough scales, or with ossifications of the-cutis in the form of scutes or spines; skin sometimes entirely naked. Skeleton incompletely ossified, with the vertebrae in small number. Gills pectinate; a narrow gill-opening in front of the pectoral fins. Mouth narrow; the bones of the upper jaw generally firmly united. A soft dorsal fin, belonging to the caudal portion of the vertebral column, opposite to the anal; sometimes elements-of a spinous dorsal besides. Ventral fin none, or reduced to spines. Air-bladder without pneumatic duct.

Family 1. Sclerodermi.— Snout somewhat produced; jaws armed with distinct teeth in small number. Skin with scutes, or rough. The elements of a spinous dorsal and ventral fin generally present.

Genera: Triacanthus, Batistes (File-Fish), Monacanthus, Anacan-thus, Ostracion (C'ofler-Fish). Fossil: Acanthoderma, Acantho· pleums, Glyptocephalus.

Family 2. Gymnodontes.— Body more or less shortened. The bones of the upper and lower jaw are confluent, forming a beak with a trenchant edge, without teeth, with or without median suture. A soft dorsal, caudal, and anal are developed, approximate. No spinous dorsal. Pectoral fins; no ventrals.

Genera: Trio don, Tetrodon (Globe-Fish), Diodon (Sea-Hedgehog), Orthagoriscus (Sun-Fish)

Subclass III.— Cyclostomata.

Skeleton cartilaginous and notochordal, without ribs and without real jaws. Skull not separate from the vertebral column. No limbs. Gills in the form of fixed sacs, without branchial arches, six or seven in number on each side. One nasal aperture only. Heart without bulbus arteriosus. Mouth anterior, surrounded by a circular or subcircular lip, suctorial. Alimentary canal straight, simple, without caecal appendages, pancreas, or spleen. Generative outlet peritoneal. Vertical fins rayed.

The Cyclostomes are most probably a very ancient type.. Unfortunately the organs of these creatures are too soft to be preserved, with the exception of the horny denticles with which the mouth of some of them is armed.

Family 1. Petromyzontidoe (Lampreys).—Body eel-shaped, naked. Subject to a metamorphosis; in the perfect stage with a suctorial mouth armed with teeth, simple or multicuspid, horny, sitting on a soft papilla. Maxillary, mandibulary, lingual, and suctorial teeth may be distinguished. Eyes present (in mature animals). External nasal aperture in the middle of the upper side of the head. The nasal duct terminates without perforating the palate. Seven branchial sacs and apertures on each side behind the head; the inner branchial ducts terminate in a separate common tube. Intestine with a spiral valve. Eggs small. The larvae without teeth, and with a single continuous vertical fin.

Genera: Petromyzon, Mordαciα, and Geotria.

Family 2. Myxinidoe. — Body eel-shaped, naked. The singlenasal aperture is above the mouth, quite at the extremity of the head, which is provided with four pairs of barbels. Mouth without lips. Nasal duct without cartilaginous rings, penetrating the palate. One median tooth on the palate, and two comb-like series of teeth on the tongue. Branchial apertures at a great distance from the head; the inner branchial ducts lead into the oesophagus. A series of mucous sacs along each side of the abdomen. Intestine without spiral valve. Eggs large, with a horny case provided with threads for adhesion.

Genera: Myxine and Bdcllostoma (Hag-Fish).

Subclass IV.— Leptocardii.

Skeleton mernbrano-cartilaginous and notochordal, ribless. No brain. Pulsating sinuses in place of a heart. Blood colourless. Respiratory cavity confluent with the abdominal cavity; branchial clefts in great number, the water being expelled by an opening in front of the vent. Jaws none.

This subclass is represented by a single family (Cirro- stomi) and by one or two genera (Βrαnchiostornα and Epigionichthys); it is the lowest in the scale of fishes, and lacks so many characteristics, not only of this class, but of the vertebrata generally, that Haeckel, with good reason, separates it into a distinct class, that of Acrania. The various parts of its organization have been duly noticed in the former parts of this article. (a. c. g.)