A Guide to the Study of Fishes, Volume 1 (of 2)

A closely related genus is Prionace, its species Prionace glauca, the great blue shark, being slender and swift, with the dorsal farther back than in Carcharias. Of the remaining genera the most important is Scoliodon, small sharks with oblique teeth which have no serrature. One of these, Scoliodon terræ-novæ, is the common sharp-nosed shark of our Carolina coast. Fossil teeth representing nearly all of these genera are common in Tertiary rocks.

Probably allied to the Carchariidæ is the genus Corax, containing large extinct sharks of the Cretaceous with broadtriangular serrate teeth, very massive in substance, and without denticles. As only the teeth are known, the actual relations of the several species of Corax are not certainly known, and they may belong to the Lamnidæ.

Family Sphyrnidæ, or Hammer-head Sharks.—The Sphyrnidæ, or hammer-headed sharks, are exactly like the Carchariidæ except that the sides of the head are produced, so as to give it the shape of a hammer or of a kidney, the eye being on the produced outer edge. The species are few, but mostly widely distributed; rather large, voracious sharks with small sharp teeth.

The true hammer-head, Sphyrna zygæna, Fig. 337, is common from the Mediterranean to Cape Cod, California, Hawaii, and Japan. The singular form of its head is one of the most extraordinary modifications shown among fishes. The bonnet-head (Sphyrna tiburo) has the head kidney-shaped or crescent-shaped. It is a smaller fish, but much the same in distribution and habits. Intermediate forms occur, so that with all the actual differences we must place the Sphyrnidæ all in one genus. Fossil hammer-heads occur in the Miocene, but their teeth are scarcely different from those of Carcharias. Sphyrna prisca, described by Agassiz, is the primeval species.

The Order of Tectospondyli.—The sharks and rays having no anal fin and with the calcareous lamellæ arranged in one or more rings around a central axis constitute a natural group to which, following Woodward, we may apply the name of Tectospondyli. The Cyclospondyli (Squalidæ, etc.) with one ring only of calcareous lamellæ may be included in this order, as also the rays, which have tectospondylous vertebræ and differ from the sharks as a group only in having the gill-openings relegated to the lower side by the expansion of the pectoral fins. The group of rays and Hasse's order of Cyclospondyli we may consider each as a suborder of Tectospondyli. The origin of this group is probably to be found in or near the Cestraciontes, as the strong dorsal spines of the Squalidæ resemble those of the Heterodontidæ.

Suborder Cyclospondyli.—In this group the vertebræ have the calcareous lamellæ arranged in a single ring about the central axis. The anal fin, as in all the tectospondylous sharks and rays, is wanting. In all the asterospondylous sharks, as in the Ichthyotomi, Acanthodei, and Chimæras, this fin is present. It is present in almost all of the bony fishes. All the species have spiracles, and in all are two dorsal fins. None have the nictitating membrane, and in all the eggs are hatched internally. Within the group there is considerable variety of form and structure. As above stated, we have a perfect gradation among Tectospondyli from true sharks, with the gill-openings lateral, to rays, which have the gill-opening on the ventral side, the great expansion of the pectoral fins, a character of relatively recent acquisition, having crowded the gill-openings from their usual position.

Family Squalidæ.—The largest and most primitive family of Cyclospondyli is that of the Squalidæ, collectively known as dogfishes or skittle-dogs. In the Squalidæ each dorsal fin has a stout spine in front, the caudal is bent upward and not keeled, and the teeth are small and varied in form, usually not all alike in the same jaw.

The genus Squalus includes the dogfishes, small, greedy sharks abundant in almost all cool seas and in some tropical waters. They are known by the stout spines in the dorsal fins and by their sharp, squarish cutting teeth. They are largely sought by fishermen for the oil in their livers, which is used to adulterate better oils. Sometimes 20,000 have been taken in one haul of the net. They are very destructive to herrings and other food-fishes. Usually the fishermen cut out the liver, throwing the shark overboard to die or to be cast on the beach. In northern Europe and New England Squalus acanthias is abundant. Squalus sucklii replaces it in the waters about Puget Sound, and Squalus mitsukurii in Japan and Hawaii. Still others are found in Chile and Australia. The species of Squalus live near shore and have the gray color usual among sharks. Allied forms perhaps hardly different from Squalus are found in the Cretaceous rocks and have been described as Centrophoroides. Other genera related to Squalus live in greater depths, from 100 to 600 fathoms, and these are violet-black. Some of the deep-water forms are the smallest of all sharks, scarcely exceeding a foot in length. Etmopterus spinax lives in the Mediterranean, and teeth of a similar species occur in the Italian Pliocene rocks. Etmopterus lucifer,[150] a deep-water species of Japan, has a brilliant luminous glandular area along the sides of the belly. Other small species of deeper waters belong to the genera Centrophorus, Centroscymnus, and Deania. In some of these species the scales are highly specialized, pedunculate, or having the form of serrated leaves. Some species are Arctic, the others are most abundant about Misaki in Japan and the Madeira Islands, two regions especially rich in semi-bathybial types. Allied to the Squalidæ is the small family of Oxynotidæ with short bodies and strong dorsal spine. Oxynotus centrina is found in the Mediterranean, and its teeth occur in the Miocene.

Family Dalatiidæ.—The Dalatiidæ, or scymnoid sharks, differ from the Squalidæ almost solely in the absence of dorsal spines. The smaller species belonging to Dalatias (Scymnorhinus, or Scymnus), Dalatias licha, etc., are very much like the dogfishes.

They are, however, nowhere very common. The teeth of Dalatias major exist in Miocene rocks. In the genus Somniosus the species are of very much greater size, Somniosus microcephalus attaining the length of about twenty-five feet. This species, known as the sleeper-shark or Greenland shark, lives in all cold seas and is an especial enemy of the whale, from which it bites large masses of flesh with a ferocity hardly to be expected from its clumsy appearance. From its habit of feeding on fish-offal, it is known in New England as "gurry-shark." Its small quadrate teeth are very much like those of the dogfish, their tips so turned aside as to form a cutting edge. The species is stout in form and sluggish in movement. It is taken for its liver in the north Atlantic on both coasts in Puget Sound and Bering Sea, and I have seen it in the markets of Tokyo. In Alaska it abounds about the salmon canneries feeding on the refuse.

Family Echinorhinidæ.—The bramble-sharks, Echinorhinidæ, differ in the posterior insertion of the very small dorsal fins, and in the presence of scattered round tubercles, like the thorns of a bramble instead of shagreen. The single species, Echinorhinus spinosus reaches a large size. It is rather scarce on the coasts of Europe, and was once taken on Cape Cod. The teeth of an extinct species, Echinorhinus richardi, are found in the Pliocene.

Suborder Rhinæ.—The suborder Rhinæ includes those sharks having the vertebræ tectospondylous, that is, with two or more series of calcified lamellæ, as on the rays. They are transitional forms, as near the rays as the sharks, although having the gill-openings rather lateral than inferior, the great pectoral fins being separated by a notch from the head.

The principal family is that of the angel-fishes, or monkfishes (Squatinidæ). In this group the body is depressed and flat like that of a ray. The greatly enlarged pectorals form a sort of shoulder in front alongside of the gill-openings, which has suggested the bend of the angel's wing. The dorsals are small and far back, the tail is slender with small fins, all these being characters shared by the rays. But one genus is now extant, widely diffused in warm seas. The species if really distinct are all very close to the European Squatina squatina. This is a moderate-sized shark of sluggish habit feeding on crabs and shells, which it crushes with its small, pointed, nail-shaped teeth. Numerous fossil species of Squatina are found from the Triassic and Cretaceous, Squatina alifera being the best known.

Family Pristiophoridæ, or Saw-sharks.—Another highly aberrant family is that of the sawsharks, Pristiophoridæ. These are small sharks, much like the Dalatiidæ in appearance, but with the snout produced into a long flat blade, on either side of which is a row of rather small sharp enameled teeth. These teeth are smaller and sharper than in the sawfish (Pristis), and the whole animal is much smaller than its analogue among the rays. This saw must be an effective weapon among the schools of herring and anchovies on which the sawsharks feed. The true teeth are small, sharp, and close-set. The few species of sawsharks are marine, inhabiting the shores of eastern Asia and Australia. Pristiophorus japonicus is found rather sparsely along the shores of Japan. The vertebræ in this group are also tectospondylous. Both the Squatina and Pristiophorus represent a perfect transition from the sharks and rays. We regard them as sharks only because the gill-openings are on the side, not crowded downward to the under side of the body-disk. As fossil, Pristiophorus is known only from a few detached vertebræ found in Germany.

Suborder Batoidei, or Rays.—The suborder of Batoidei, Rajæ, or Hypotrema, including the skates and rays, is a direct modern offshoot from the ancestors of tectospondylous sharks, its characters all specialized in the direction of life on the bottom with a food of shells, crabs, and other creatures less active than fishes.

The single tangible distinctive character of the rays as a whole lies in the position of the gill-openings, which are directly below the disk and not on the side of the neck in all the sharks. This difference in position is produced by the anterior encroachment of the large pectoral fins, which are more or less attached to the side of the head. By this arrangement, which aids in giving the body the form of a flat disk, the gill-openings are limited and forced downward. In the Squatinidæ (angel-fishes) and the Pristiophoridæ (sawsharks) the gill-openings have an intermediate position, and these families might well be referred to the Batoidei, with which group they agree in the tectospondylous vertebræ.

Other characters of the rays, appearing progressively, are the widening of the disk, through the greater and greater development of the fins, the reduction of the tail, which in the more specialized forms becomes a long whip, the reduction, more and more posterior insertion, and the final loss of the dorsal fins, which are always without spine, the reduction of the teeth to a tessellated pavement, then finally to flat plates and the retention of the large spiracle. Through this spiracle the rays breathe while lying on the bottom, thus avoiding the danger of introducing sand into their gills, as would be done if they breathed through the mouth. In common with the cyclospondylous sharks, all the rays lack the anal fin. The rays rarely descend to great depths in the sea. The different members have varying relations, but the group most naturally divides into thick-tailed rays or skates (Sarcura) and whip-tailed rays or sting-rays (Masticura). The former are much nearer to the sharks and also appear earliest in geological times.

Pristididæ, or Sawfishes.—The sawfishes, Pristididæ, are long, shark-like rays of large size, having, like the sawsharks, the snout prolonged into a very long and strong flat blade, with a series of strong enameled teeth implanted in sockets along either side of it. These teeth are much larger and much less sharp than in the sawsharks, but they are certainly homologous with these, and the two groups must have a common descent, distinct from that of the other rays. Doubtless when taxonomy is a more refined art they will constitute a small suborder together. This character of enameled teeth on the snout would seem of more importance than the position of the gill-openings or even the flattening and expansion of the body. The true teeth in the sawfishes are blunt and close-set, pavement-like as befitting a ray. (See Fig. 152.)

The sawfishes are found chiefly in river-mouths of tropical America and West Africa: Pristis pectinatus in the West Indies; Pristis zephyreus in western Mexico; and Pristis pectinatus in the Senegal. They reach a length of ten to twenty feet, and with their saws they make great havoc among the schools of mullets and sardines on which they feed. The stories of their attacks on the whale are without foundation. The writer has never found any of the species in the open sea. They live chiefly in the brackish water of estuaries and river-mouths.

Fossil teeth of sawfishes occur in abundance in the Eocene. Still older are vertebræ from the Upper Cretaceous at Maestricht. In Propristis schweinfurthi the tooth-sockets are not yet calcified. In Sclerorhynchus atavus, from the Upper Cretaceous, the teeth are complex in form, with a "crimped" or stellate base and a sharp, backward-directed enameled crown.

Rhinobatidæ, or Guitar-fishes.—The Rhinobatidæ (guitar-fishes) are long-bodied, shovel-nosed rays, with strong tails; they are ovoviviparous, hatching the eggs within the body. The body, like that of the shark or sawfish, is covered with nearly uniform shagreen. The numerous species abound in all warm seas; they are olive-gray in color and feed on small animals of the seabottoms. The length of the snout differs considerably in different species, but in all the body is relatively long and strong. Most of the species belong to Rhinobatus. The best-known American species are Rhinobatus lentiginosus of Florida and Rhinobatus productus of California. The names guitar-fish, fiddler-fish, etc., refer to the form of the body. Numerous fossil species, allied to the recent forms, occur from the Jurassic. Species much like Rhinobatus occur in the Cretaceous and Eocene. Tamiobatis vetustus, lately described by Dr. Eastman from a skull found in the Devonian of eastern Kentucky, the oldest ray-like fish yet known, is doubtless the type of a distinct family, Tamiobatidæ. It is more likely a shark however than a ray, although the skull has a flattened ray-like form.

Closely related to the Rhinobatidæ are the Rhinidæ (Rhamphobatidæ), a small family of large rays shaped like the guitar-fishes and found on the coast of Asia. Rhina ancylostoma extends northward to Japan.

In the extinct family of Astrodermidæ, allied to the Rhinobatidæ, the tail has two smooth spines and the skin is covered with tubercles. In Belemnobatis sismondæ the tubercles are conical; in Astrodermus platypterus they are stellate.

Rajidæ, or Skates.—The Rajidæ, skates, or rays, inhabit the colder waters of the globe and are represented by a large number of living species. In this family the tail is stout, with two-rayed dorsal fins and sometimes a caudal fin. The skin is variously armed with spines, there being always in the male two series of specialized spinous hooks on the outer edge of the pectoral fin. There is no serrated spine or "sting," and in all the species the eggs are laid in leathery cases, which are "wheelbarrow-shaped," with a projecting tube at each of the four angles. The size of this egg-case depends on the size of the species, ranging from three to about eight inches in length. In some species more than one egg is included in the same case.

Most of the species belong to the typical genus Raja, and these are especially numerous on the coasts of all northern regions, where they are largely used as food. The flesh, although rather coarse and not well flavored, can be improved by hot butter, and as "raie au beurre noir" is appreciated by the epicure. The rays of all have small rounded teeth, set in a close pavement.

Some of the species, known on our coasts as "barn-door skates," reach a length of four or five feet. Among these are Raja lævis and Raja ocellata on our Atlantic coast, Raja binoculata in California, and Raja tengu in Japan. The small tobacco-box skate, brown with black spots, abundant on the New England coast, is Raja erinacea. The corresponding species in California is Raja inornata, and in Japan Raja kenojei. Numerous other species, Raja batis, clavata, circularis, fullonica, etc., occur on the coasts of Europe. Some species are variegated in color, with eye-like spots or jet-black marblings. Still others, living in deep waters, are jet-black with the body very soft and limp. For these Garman has proposed the generic name Malacorhinus, a name which may come into general use when the species are better known. In the deep seas rays are found even under the equator. In the south-temperate zone the species are mostly generically distinct, Psammobatis being a typical form, differing from Raja. Discobatus sinensis, common in China and Japan, is a shagreen-covered form, looking like a Rhinobatus. It is, however, a true ray, laying its eggs in egg-cases, and with the pectorals extending on the snout. Fossil Rajidæ, known by the teeth and bony tubercles, are found from the Cretaceous onward. They belong to Raja and to the extinct genera Dynatobatis, Oncobatis, and Acanthobatis. The genus Arthropterus (rileyi) from the Lias, known from a large pectoral fin, with distinct cylindrical-jointed rays, may have been one of the Rajidæ, or perhaps the type of a distinct family, Arthropteridæ.

Narcobatidæ, or Torpedoes.—The torpedoes, or electric rays (Narcobatidæ), are characterized by the soft, perfectly smooth skin, by the stout tail with rayed fins, and by the ovoviviparous habit, the eggs being hatched internally. In all the species is developed an elaborate electric organ, muscular in its origin and composed of many hexagonal cells, each filled with soft fluid. These cells are arranged under the skin about the back of the head and at the base of the pectoral fin, and are capable of benumbing an enemy by means of a severe electric shock. The exercise of this power soon exhausts the animal, and a certain amount of rest is essential to recovery.

The torpedoes, also known as crampfishes or numbfishes, are peculiarly soft to the touch and rather limp, the substance consisting largely of watery or fatty tissues. They are found in all warm seas. They are not often abundant, and as food they have not much value.

Perhaps the largest species is Tetronarce occidentalis, the crampfish of our Atlantic coast, black in color, and said sometimes to weigh 200 pounds. In California Tetronarce californica reaches a length of three feet and is very rarely taken, in warm sandy bays. Tetronarce nobiliana in Europe is much like these two American species. In the European species, Narcobatus torpedo, the spiracles are fringed and the animal is of smaller size. To Narcine belong the smaller numbfish, or "entemedor," of tropical America. These have the spiracles close behind the eyes, not at a distance as in Narcobatus and Tetronarce. Narcine brasiliensis is found throughout the West Indies, and Narcine entemedor in the Gulf of California. Astrape, a genus with but one dorsal fin, is common in southern Japan. Fossil Narcobatus and Astrape occur in the Eocene, one specimen of the former nearly five feet long. Vertebræ of Astrape occur in Prussia in the amber-beds.

Petalodontidæ.—Near the Squatinidæ, between the sharks and the rays, Woodward places the large extinct family of Petalodontidæ, with coarsely paved teeth each of which is elongate with a central ridge and one or more strong roots at base. The best-known genera are Janassa and Petalodus, widely distributed in Carboniferous time. Janassa is a broad flat shark, or, perhaps, a skate, covered with smooth shagreen. The large pectoral fins are grown to the head; the rather large ventral fins are separated from them. The tail is small, and the fins, as in the rays, are without spines. The teeth bear some resemblance to those of Myliobatis. Janassa is found in the coal-measures of Europe and America, and other genera extend upward from the Subcarboniferous limestones, disappearing near the end of Carboniferous time. Petalodus is equally common, but known only from the teeth. Other widely distributed genera are Ctenoptychius and Polyrhizodus.

These forms may be intermediate between the skates and the sting-rays. In dentition they resemble most the latter.

Similar to these is the extinct family of Pristodontidæ with one large tooth in each jaw, the one hollowed out to meet the other. It is supposed that but two teeth existed in life, but that is not certain. Nothing is known of the rest of the body in Pristodus, the only genus of the group.

Dasyatidæ, or Sting-rays.—In the section Masticura the tail is slender, mostly whip-like, without rayed dorsal or caudal fins, and it is usually armed with a very long spine with saw-teeth projecting backward. In the typical forms this is a very effective weapon, being wielded with great force and making a jagged wound which in man rarely heals without danger of blood-poisoning. There is no specific poison, but the slime and the loose cuticle of the spine serve to aggravate the irregular cut. I have seen one sting-ray thrust this spine through the body of another lying near it in a boat. Occasionally two or three of these spines are present. In the more specialized forms of sting-rays this spine loses its importance. It becomes very small and not functional, and is then occasionally or even generally absent in individuals.

The common sting-rays, those in which the caudal spine is most developed, belong to the family of Dasyatidæ. This group is characterized by the small skate-like teeth and by the non-extension of the pectoral rays on the head. The skin is smooth or more or less rough. These animals lie flat on the sandy bottoms in nearly all seas, feeding on crabs and shellfish. All hatch the eggs within the body. The genus Urolophus has a rounded disk, and a stout, short tail with a caudal fin. It has a strong spine, and for its size is the most dangerous of the sting-rays. Urolophus halleri, the California species, was named for a young man who was stung by the species at the time of its first discovery at San Diego in 1863. Urolophus jamaicensis abounds in the West Indies, Urolophus mundus at Panama, and Urolophus fuscus in Japan. None of the species reach Europe. The true sting-ray (stingaree, or clam-cracker), Dasyatis, is more widely diffused and the species are very closely related. In these species the body is angular and the tail whip-like. Some of the species reach a length of ten or twelve feet. None have any economic value, and all are disliked by fishermen. Dasyatis pastinaca is common in Europe, Dasyatis centrura along our Atlantic coast, Dasyatis sabina ascends the rivers of Florida, and Dasyatis dipterura abounds in the bay of San Diego. Other species are found in tropical America, while still others (Dasyatis akajei, kuhlii, zugei, etc.) swarm in Japan and across India to Zanzibar.

Pteroplatea, the butterfly-ray, has the disk very much broader than long, and the trivial tail is very short, its little spine more often lost than present. Different species of this genus circle the globe: Pteroplatea maclura, on our Atlantic coast; Pteroplatea marmorata, in California; Pteroplatea japonica, in Japan; and Pteroplatea altavela, in Europe. They are all very much alike, olive, with the brown upper surface pleasingly mottled and spotted.

Sting-rays of various types, Tæniura, Urolophus, etc., occur as fossils from the Eocene onward. A complete skeleton called Xiphotrygon acutidens, distinguished from Dasyatis by its sharp teeth, is described by Cope from the Eocene of Twin Creek in Wyoming. Vertebræ of Urolophus are found in German Eocene. Cyclobatis (oligodactylus), allied to Urolophus, with a few long pectoral rays greatly produced, extending over the tail and forming a rayed wreath-like projection over the snout, is known from the Lower Cretaceous.

Myliobatidæ.—The eagle-rays, Myliobatidæ, have the pectoral fins extended to the snout, where they form a sort of rayed pad. The teeth are very large, flat, and laid in mosaic. The whip-like tail is much like that in the Dasyatidæ, but the spine is usually smaller. The eagle-like appearance is suggested by the form of the skull. The eyes are on the side of the head with heavy eyebrows above them. The species are destructive to clams and oysters, crushing them with their strong flat teeth.

In Aëtobatus the teeth are very large, forming but one row. The species Aëtobatus narinari is showily colored, brown with yellow spots, the body very angular, with long whip-like tail. It is found from Brazil to Hawaii and is rather common.

In Myliobatis the teeth are in several series. The species are many, and found in all warm seas. Myliobatis aquila is the eagle-ray of Europe, Myliobatis californicus is the batfish of California, and Myliobatis tobijei takes its place in Japan.

In Rhinoptera the snout is notched and cross-notched in front so that it appears as if ending in four lobes at the tip. These "cow-nosed rays," or "whipparees," root up the soft bottoms of shallow bays in their search for clams, much as a drove of hogs would do it. The common American species is Rhinopterus bonasus. Rhinoptera steindachneri lives in the Gulf of California.

Teeth and spines of all these genera are common as fossils from the Eocene onwards, as well as many of the extinct genus, Ptychodus, with cyclospondylous vertebræ. Ptychodus mammilaris, rugosus, and decurrens are characteristic of the Cretaceous of England. Myliobatis dixoni is common in the European Eocene, as is also Myliobatis toliapicus and Aëtobatis irregularis. Apocopodon seriacus is known from the Cretaceous of Brazil.

Family Psammodontidæ.—The Psammodontidæ are known only from the teeth, large, flat, or rounded and finely dotted or roughened on the upper surface, as the name Psammodus (ψάμμος, sand; ὀδούς, tooth) would indicate. The way in which the jaws lie indicates that these teeth belonged to rays rather than sharks. Numerous species have been described, mostly from the Subcarboniferous limestones. Archæobatis gigas, perhaps, as its name would indicate, the primeval skate, is from the Subcarboniferous limestone of Greencastle, Indiana. Teeth of numerous species of Psammodus and Copodus are found in many rocks of Carboniferous age. Psammodus rugosus common in Carboniferous rocks of Europe.

Family Mobulidæ.—The sea-devils, Mobulidæ, are the mightiest of all the rays, characterized by the development of the anterior lobe of the pectorals as a pair of cephalic fins. These stand up like horns or cars on the upper part of the head. The teeth are small and flat, tubercular, and the whip-like tail is with or without spine. The species are few, little known, and inordinately large, reaching a width of more than twenty feet and a weight, according to Risso, of 1250 pounds. When harpooned it is said that they will drag a large boat with great swiftness. The manta, or sea-devil, of tropical America is Manta birostris. It is said to be much dreaded by the pearl-fishers, who fear that it will devour them "after enveloping them in its vast wings." It is not likely, however, that the manta devours anything larger than the pearl-oyster itself. Manta hamiltoni is a name given to a sea-devil of the Gulf of California. The European species Mobula edentula reaches a similarly enormous size, and Mobula hypostoma has been scantily described from Jamaica and Brazil. Mobula japonica occurs in Japan. A fœtus in my possession from a huge specimen taken at Misaki is nearly a foot across. In Mobula (Cephaloptera) there are teeth in both jaws, in Manta (Ceratoptera) in the lower jaw only. In Ceratobatis from Jamaica (C. robertsi) there are teeth in the upper jaw only. Otherwise the species of the three genera are much alike, and from their huge size are little known and rarely seen in collections. Of Mobulidæ no extinct species are known.

CHAPTER XXXI
THE HOLOCEPHALI, OR CHIMÆRAS

The Chimæras.—Very early in geological times, certainly as early as the middle Silurian, the type of Chimæras diverged from that of the sharks. Hasse derives them directly from his hypothetical primitive Polyospondyli, by way of the Acanthodei and Ichthyotomi. In any event the point of divergence must be placed very early in the evolution of sharks, and this suggestion is as likely as any other. The chief character of Chimæras is found in the autostylic skull, which is quite different from the hyostylic skull of the sharks. In the sharks and in all higher fishes the mandible is joined to the skull by a suspensorium of bones or cartilages (quadrate, symplectic, and hyomandibular bones in the Teleost fishes). To this arrangement the name hyostylic is given. In the Chimæra there is no suspensorium, the mandible being directly attached to the cranium, of which the hyomandibular and quadrate elements form an integral part, this arrangement being called autostylic. The palato-quadrate apparatus, of which the upper jaw is the anterior part, is immovably fused with the cranium, instead of being articulated with it. This fact, gives the name to the subclass Holocephali (ὅλος, whole or solid; κεφαλή, head). Other characters are found in the incomplete character of the back-bone, which consists of a scarcely segmented notochord differing from the most primitive condition imagined only in being surrounded by calcareous rings, no lime entering into the composition of the notochord itself. The tail is diphycercal and usually prolonged in a filament (leptocercal). The shoulder-girdle, as in the sharks, is free from the skull. The pectoral fins are short and broad, without segmented axis or archipterygium and without recognizable analogue of the three large cartilages seen in the sharks, the propterygium, mesopterygium, and metapterygium. In the mouth, instead of teeth, are developed flat, bony plates called tritors or grinders, set endwise in the front of the jaws. The gills are fringe-like, free at the tips as in ordinary fishes, and there is a single external opening for them all as in true fishes, and they are covered with a flap of skin. These structures are, however, quite different from those of the true fishes and are doubtless independently developed. There is no spiracle. The skin is smooth or rough. In the living forms and most of the extinct species there is a strong spine in the dorsal fin. The ventral fin in the male has complex, usually trifid, claspers, and an analogous organ, the cephalic holder, is developed on the front of the head, in the adult male. This is a bony hook with a brush of glistening enameled teeth at the end. The eggs are large, and laid in oblong or elliptical egg-cases, provided with silky filaments. The eggs are fertilized after they are extruded. Mucous channels and lateral line are highly developed, being most complex about the head. The brain is essentially shark-like, the optic nerves form a chiasma, and the central hemispheres are large.

The teeth of the Chimæras are thus described by Woodward, vol. 2, pp. 36, 37:

"In all the known families of Chimæroids, the dentition consists of a few large plates of vascular dentine, of which certain areas ('tritors') are specially hardened by the deposition of calcareous salts within and around groups of medullary canals, which rise at right angles to the functional surface. In most cases there is a single pair of such plates in the lower jaw, meeting at the symphysis, while two pairs are arranged to oppose these above. As a whole, the dentition thus closely resembles that of the typical Dipnoi (as has often been pointed out); and the upper teeth may be provisionally named palatine and vomerine until further discoveries shall have revealed their precise homologies. The structures are sometimes described as 'jaws,' and regarded as dentaries, maxillæ, and premaxillæ, but the presence of a permanent pulp under each tooth is conclusive proof of their bearing no relation to the familiar membrane-bones thus named in higher fishes."

Relationship of Chimæras.—As to the origin of the Chimæras and their relation to the sharks, Dr. Dean has this recent ("The Devonian Lamprey") and interesting word:

"The Holocephali have always been a doubtful group, anatomy and palæontology contributing but imperfect evidence as to their position in the gnathostome phylum. Their embryology, however, is still undescribed, except in a brief note by T. J. Parker, and it is reasonably looked to to contribute evidence as to their line of descent. The problem of the relationships of the Chimæroids has long been of especial interest to me, and it has led me to obtain embryonic material of a Pacific species of one of these forms. It may be of interest in this connection to state that the embryology of this form gives the clearest evidence that the wide separation of the Selachii and Holocephali is not tenable. The entire plan of development in Chimæra colliei is clearly like that of a shark. The ovulation is closely like that of certain of the rays and sharks: the eggs are large, the segmentation is distinctly shark-like; the circular blastoderm overgrows the yolk in an elasmobranchian manner. The early embryos are shark-like; and the later ones have, as T. J. Parker has shown, external gills, and I note further that these arise, precisely as in shark-embryos, from the posterior margin of the gill-bar. A spiracle also is present. A further and most interesting developmental feature is the fact that the autostylism in Chimæra is purely of secondary nature and is at the most of ordinal value. It is found that in a larva of Chimæra measuring 45 mm. in length, the palato-quadrate cartilage is still separated from the skull by a wide fissure. This becomes gradually reduced by the confluence of the palato-quadrate cartilage with the skull, the fusion taking place at both the anterior and posterior ends of the mesal rim of the cartilage. The remains of the fissure are still well marked in the young Chimæra, four inches in length; and a rudiment of it is present in the adult skull as a passage-way for a nerve. Regarding the dentition: it may also be noted in the present connection that the growth of the dental plates in Chimæra suggests distinctly elasmobranchian conditions. Thus on the roof of the mouth the palatine plates are early represented by a series of small more or less conical elements which resemble outwardly, at least, the 'anlagen' of the pavement teeth in cestraciont sharks."

Family Chimæridæ.—The existing Chimæras are known also as spookfishes, ratfishes, and elephant-fishes. These are divided by Garman into three families, and in the principal family, the Chimæridæ, the snout is blunt, the skin without plates, and the dorsal fin is provided with a long spine. The flat tritors vary in the different genera. The single genus represented among living fishes is Chimæra, found in cold seas and in the oceanic depths. The best-known species, Chimæra colliei, the elephant-fish, or chimæra of California, abounds in shallow waters of ten to twenty fathoms from Sitka to San Diego. It is a harmless fish, useless except for the oil in its liver, and of special interest to anatomists as the only member of the family to be found when desired for dissection. This species was first found at Monterey by Mr. Collie, naturalist of Captain Beechey's ship, the Blossom. It is brown in color, with whitish spots, and reaches a length of 2½ feet. As a shallow-water form, with certain differences in the claspers and in the tail, Chimæra colliei is sometimes placed in a distinct genus, Hydrolagus. Other species inhabit much greater depths and have the tail produced into a long filament. Of these, Chimæra monstrosa, the sea-cat of the north Atlantic, has been longer known than any other Chimæra. Chimæra affinis has been dredged in the Gulf Stream and off Portugal. Chimæra phantasma and Chimæra mitsukurii are frequently taken in Japan, and the huge jet-black Chimæra purpurascens in Hawaii and Japan. None of these species are valued as food, but all impress the spectator with their curious forms.

The fossil Chimæridæ, although numerous from Triassic times and referred to several genera, are known chiefly by their teeth with occasional fin-spines, frontal holders, or impressions of parts of the skeleton. The earliest of chimæroid remains has been described by Dr. Charles D. Walcott[151] from Ordovician or Lower Silurian rocks at Cañon City, Colorado. Of the species called Dictyorhabdus priscus, only parts supposed to be the sheath of the notochord have been preserved. Dr. Dean thinks this more likely to be part of the axis of a cephalopod shell. The definitely known Chimæridæ are mainly confined to the rocks of the Mesozoic and subsequent eras. Ischyodus priscus (avitus) of the lower Jura resembles a modern chimæra. Granodus oweni is another extinct chimæra, and numerous fin-spines, teeth, and other fragments in the Cretaceous and Eocene of America and Europe are referred to Edaphodon. A species of Chimæra has been recorded from the Pliocene of Tuscany, and one of Callorhynchus from the greensand of New Zealand. Other American Cretaceous genera of chimæroids are Mylognathus, Bryactinus, Isotænia, Leptomylus, and Sphagepœa. Dental plates called Rhynchodus are found in the Devonian.

Rhinochimæridæ.—The most degenerate of existing chimæras belong to the family of Rhinochimæridæ, characterized by the long flat soft blade in which the snout terminates. This structure resembles that seen in the deep-sea shark, Mitsukurina, and in Polyodon. In Rhinochimæra pacifica of Japan the teeth in each jaw form but a single plate. In Harriotta raleighana, of the Gulf Stream, they are more nearly as in Chimæra. Both are bathybial fishes, soft in texture, and found in great depths. The family of Callorhynchidæ, or Antarctic Chimæras, includes the bottle-nosed Chimæra (Callorhynchus callorhynchus) of the Patagonian region. In this species the snout is also produced, a portion being turned backward below in front of the mouth, forming a sensory pad well supplied with nerves.

Extinct Chimæroids.—According to Woodward, three other families are recognizable among the extinct forms.

The Ptyctodontidæ are known from the teeth only, a single pair of large, laterally compressed dental plates in each jaw, with a few hard tritoral areas. These occur in Silurian and Devonian rocks. Ptyctodus obliquus from the Devonian of Russia is the best-known species. Other genera are Rhynchodus and Palæomylus.

The Squalorajidæ have the head depressed and the snout produced in a flat rostrum, as in Harriotta. There is no dorsal spine, and the teeth are a few thin curved plates. The frontal holder of the male is well developed. The few species occur in the Lias. Squaloraja dolichognathos is known from numerous fragments from the Triassic in England and Scotland. Chalcodus permianus is found in German Permian.

The Myriacanthidæ have the body elongate, with dermal plates on the head and a long straight spine in the dorsal fin. The frontal holder is large. The species, few in number, are found in Mesozoic rocks. Myriacanthus paradoxus is the best-known species. Of another species, Chimæropsis paradoxa, a skeleton about three feet long has been found which shows a number of peculiar traits. The skin is covered with ribbed shagreen scales. The dorsal fin has a large spine with retrorse serrations behind. The tail is slim, and the pectoral and ventral fins are very large. Bony plates with conical spines protect the neck. The teeth are large and angular, of peculiar form.

Ichthyodorulites.—The term ichthyodorulite (ἰχθύς, fish; δόρυ, lance; λίθος, stone) is applied to detached fin-spines, dermal spines, and tubercles belonging to unrecognized species of sharks and chimæras. Some of these are serrated, others entire, some straight, some curved, and some with elaborate armature or sculpture. Some doubtless belong to Cestraciontes, others to Pleuracanthidæ; some to Squalidæ, some to chimæras, and others, perhaps, to forms still altogether unknown.