CHAPTER XXVI

GOBIOIDEI, DISCOCEPHALI, AND TÆNIOSOMI

=SUBORDER Gobioidei, the Gobies: Gobiidæ.=—The great family of _Gobiidæ_, having no near relations among the spiny-rayed fishes, may be here treated as forming a distinct suborder.

The chief characteristics of the family are the following: The ventral fins are thoracic in position, each having one spine and five soft rays, in some cases reduced to four, but never wanting. The ventral fins are inserted very close together, the inner rays the longest, and in most cases the two fins are completely joined, forming a single roundish fin, which may be used as a sucking-disk in clinging to rocks. The shoulder-girdle is essentially perch-like in form, the cranium is usually depressed, the bones being without serrature. There is no lateral line, the gill-openings are restricted to the sides, and the spinous dorsal is always small, of feeble spines, and is sometimes altogether wanting. There is no bony stay to the preopercle. The small pharyngeals are separate, and the vertebræ usually in normal number, 10 + 14 = 24.

The species are excessively numerous in the tropics and temperate zones, being found in lakes, brooks, swamps, and bays, never far out in the sea, and usually in shallow water. Many of them burrow in the mud between or below tide-marks. Others live in swift waters like the darters, which they much resemble. A few reach a length of a foot or two, but most of the species rarely exceed three inches, and some of them are mature at half an inch.

The largest species, _Philypnus dormitor_, the guavina de rio, is found in the rivers of Mexico and the West Indies. It reaches a length of nearly two feet and is valued as food. Unlike most of the others, in this species there are teeth on the vomer. Other related forms of the subfamily of _Eleotrinæ_, having the ventral fins separate, are _Eleotris pisonis_, a common river-fish everywhere in tropical America; _Eleotris fusca_, a river-fish abounding from Tahiti and Samoa to Hindostan; _Dormitator maculatus_, the stout-bodied guavina-mapo of the West Indian regions, with the form of a small carp. _Guavina guavina_ of Cuba is another species of this type, and numerous other species having separate ventrals are found in the East Indies, the West Indies, and in the islands of Polynesia. Some species, as _Valenciennesia strigata_ of the East Indies and _Vireosa hanæ_ of Japan, are very gracefully colored. One genus, _Eviota_, is composed of numerous species, all minute, less than an inch in length. These abound in the crevices in coral-heads. _Eviota epiphanes_ is found in Hawaii, the others farther south. _Hypseleotris guntheri_, of the rivers and springs of Polynesia, swims freely in the water, like a minnow, never hugging the bottom as usual among gobies.

Of the typical gobies having the ventrals united we can mention but a few of the myriad forms, different species being abundant alike in fresh and salt waters in all warm regions. In Europe _Gobius jozo_, _Gobius ophiocephalus_, and many others are common species. The typical genus _Gobius_ is known by its united ventrals, and by the presence of silken free rays on the upper part of the pectoral fin. _Mapo soporator_ swarms about coral reefs in both Indies. _Gobionellus oceanicus_, the esmeralda or emerald-fish, is notable for its slender body and the green spot over its tongue. _Gobiosoma alepidotum_ and other species are scaleless. _Barbulifer ceuthœcus_ lives in the cavities of sponges. _Coryphopterus similis_, a small goby, swarms in almost every brook of Japan. The species of _Pterogobius_ are beautifully colored, banded with white or black, or striped with red or blue. _Pterogobius virgo_ and _Pterogobius daimio_ of Japan are the most attractive species. Species of _Cryptocentrus_ are also very prettily colored.

Of the species burrowing in mud the most interesting is the long-jawed goby, _Gillichthys mirabilis_. In this species the upper jaw is greatly prolonged, longer than the head, as in _Opisthognathus_ and _Neoclinus_. In the "American Naturalist" for August, 1877, Mr. W. N. Lockington says of the long-jawed goby:

"I call it the long-jawed goby, as its chief peculiarity consists in its tremendous length of jaw. A garpike has a long jaw, and so has an alligator, and it is not unlikely that the title will call up in the minds of some who read this the idea of a terrible mouth, armed with a bristling row of teeth. This would be a great mistake, for our little fish has no teeth worth bragging about, and does not open his mouth any wider than a well-behaved fish should do. The great difference between his long jaws and those of a garpike is that the latter's project forward, while those of our goby are prolonged backward immensely.

"The long-jawed goby was discovered by Dr. J. G. Cooper in the Bay of San Diego, among seaweed growing on small stones at the wharf, and in such position that it must have been out of the water from three to six hours daily, though kept moist by the seaweed.

"On a recent occasion a single _Gillichthys_, much larger than any of the original types, was presented by a gentleman who said that the fish, which was new to him, was abundant upon his ranch in Richardson's Bay, in the northern part of the Bay of San Francisco; that the Chinamen dug them up and ate them, and that he had had about eleven specimens cooked, and found them good, tasting, he thought, something like eels. The twelfth specimen he had preserved in alcohol, in the interest of natural science. This gentleman had the opportunity of observing something of the mode of life of these fishes, and informed us that their holes, excavated in the muddy banks of tidal creeks, increase in size as they go downward, so that the lower portion is below the water-level, or at least sufficiently low to be kept wet by the percolation from the surrounding mud.

"When the various specimens now acquired were placed side by side, the difference in the relative length of their jaws was very conspicuous, for while in the smallest it was about one-fifth of the total length, in the largest it exceeded one-third.

"As the fish had now been found in two places in the bay, I thought I would try to find it also, and to this end sallied out one morning, armed with a spade, and commenced prospecting in a marsh at Berkeley, not far from the State University. For a long time I was unsuccessful, as I did not know by what outward signs their habitations could be distinguished, and the extent of mud-bank left bare by the retreating tide was, as compared with my powers of delving, practically limitless.

"At last, toward evening, while digging in the bend of a small creek, in a stratum of soft, bluish mud, and at a depth of about a foot below a small puddle, I found five small fishes, which at first I believed to belong to an undescribed species, so little did they resemble the typical _G. mirabilis_, but which proved, upon a closer examination, to be the young of that species. There was the depressed, broad head, the funnel-shaped ventral 'disk' formed by the union of the two ventral fins, and the compressed tail of the long-jawed goby, but where were the long jaws? The jaws were, of course, in their usual place, but their prolongations had only just begun to grow along the sides of the head, and were not noticeable unless looked for. A comparison of the various specimens proved conclusively that the strange-looking appendage is developed during the growth of the fish, as will be seen by the following measurements of four individuals:

"In the smallest specimen the maxillary expansion extends beyond the orbit for a distance about equal to that which intervenes between the anterior margin of the orbit and the tip of the snout; in No. 2 it reaches to the posterior margin of the preoperculum; in No. 3 it ends level with the gill-opening; while in the largest individual it passes the origin of the pectoral and ventral fins.

"What can be the use of this long fold of skin and cartilage, which is not attached to the head except where it joins the mouth, and which, from its gradual development and ultimate large dimensions, must certainly serve some useful purpose?

"Do not understand that I mean that every part of a creature is of use to it in its present mode of life, for, as all naturalists know, there are in structural anatomy, just as in social life, cases of _survival_; remains of organs which were at some former time more developed, parallel in their nature to such survivals in costume as the two buttons on the back of a man's coat, once useful for the attachment of a sword-belt. But in this fish we have no case of survival, but one of unusual development; the family (_Gobiidæ_) to which it belongs presents no similar case, although its members have somewhat similar habits, and the conviction grows upon us, as we consider the subject, that the long jaws serve some useful purpose in the economy of the creature. In view of the half-terrestrial life led by this fish, I am inclined to suspect that the expansion of the upper jaw may serve for the retention of a small quantity of water, which, slowly trickling downward into the mouth and gills, keeps the latter moist when, from an unusually low tide or a dry season, the waters of its native creek fail, perhaps for several hours, to reach the holes in which the fishes dwell. It may be objected to this view that, were such an appendage necessary or even useful, other species of _Gobiidæ_, whose habits are similar, would show traces of a similar adaptation. This, however, by no means follows. Nature has many ways of working out the same end; and it must be remembered that every real species, when thoroughly known, differs somewhat in habits from its congeners, or at least from its family friends. To take an illustration from the mammalia. The chimpanzee and the spider-monkey are both quadrumanous and both arboreal, yet the end which is attained in the former by its more perfect hands is reached in the latter by its prehensile tail.

"Why may not the extremely long channel formed by the jaw of this rather abnormal member of the goby family be another mode of provision for the requirements of respiration?"

Of the Asiatic genera, _Periophthalmus_ and _Boleophthalmus_ are especially notable. In these mud-skippers the eyes are raised on a short stalk, the fins are strong, and the animal has the power of skipping along over the wet sands and mud, even skimming with great speed over the surface of the water. It chases its insect prey among rocks, leaves, and weeds, and out of the water is as agile as a lizard. Several species of these mud-skippers are known on the coasts of Asia and Polynesia, _Periophthalmus barbarus_ and _Boleophthalmus chinensis_ being the best known. _Awaous crassilabris_ is the common oopu, or river goby, of the Hawaiian streams, and _Lentipes stimpsoni_ is the mountain oopu, capable of clinging to the rocks in the rush of torrents. _Paragobiodon echinocephalus_ is a short thick-set goby with very large head, found in crevices of coral reefs of Polynesia.

In numerous interesting species the first dorsal fin is wanting or much reduced. The crystal goby, _Crystallogobius nilssoni_, of Europe is one of this type, with the body translucent. Equally translucent is the little Japanese shiro-uwo, or whitefish, _Leucopsarion petersi_. _Mistichthys luzonius_ of the Philippine Islands, another diaphanous goby, is said to be the smallest of all vertebrates, being mature at half an inch in length. This minute fish is so very abundant as to become an important article of food in Luzon. The rank of "smallest-known vertebrate" has been claimed in turn for the lancelet (_Asymmetron lucayanum_), the top minnow, _Heterandria formosa_, and the dwarf sunfish (_Elassoma zonatum_). _Mistichthys luzonius_ is smaller than any of these, but the diminutive gobies, called _Eviota_, found in interstices of coral rocks are equally small, and there are several brilliant but minute forms in the reefs of Samoa. The snake-like _Eutæniichthys gilli_ of Japanese rivers is scarcely larger, though over an inch long. _Typhlogobius californiensis_, "the blindfish of Point Loma," is a small goby, colorless and blind, found clinging in dark crevices of rock about Point Loma and Dead Man's Island in southern California.

Its eyes are represented by mere rudiments, their loss being evidently associated with the peculiar habit of the species, which clings to the under side of stones in relative darkness, though in very shallow water. The flesh is also colorless, the animal appearing pink in life.

In the Japanese species _Luciogobus guttatus_, common under stones and along the coast, the spinous dorsal, weak in numerous other species, finally vanishes altogether. Other gobies are band-shaped or eel-shaped, the dorsal spines being continuous with the soft rays. Among these are the barreto of Cuba, _Gobioides broussoneti_, and in Japan _Tænioides lacepedei_ and _Trypauchen wakæ_, the latter species remarkable for its strong canines. Fossil gobies are practically unknown. A few fragments, otoliths, and partial skeletons in southern Europe have been referred to _Gobius_, but no other genus is represented.

The family of _Oxudercidæ_ contains one species, _Oxuderces dentatus_, a small goby-like fish from China. It is an elongate fish, without ventral fins, and with very short dorsal and anal.

=Suborder Discocephali, the Shark-suckers: Echeneididæ.=—Next to the gobies, for want of a better place, we may mention the singular group of _Discocephali_ (δίσκος, disk; κεφαλή, head). In this group the first dorsal fin is transformed into a peculiar laminated sucking-disk, which covers the whole top of the head and the nape. In other respects the structure does not diverge very widely from the percoid type, there being a remarkable resemblance in external characters to the Scombroid genus _Rachycentron_. But the skeleton shows no special affinity to _Rachycentron_ or to any perciform fish. The basis of the cranium is simple, and in the depression of the head with associated modifications the _Discocephali_ approach the gobies and blennies rather than the mackerel-like forms.

The _Discocephali_ comprise the single family of shark-suckers or remoras, the _Echeneididæ_. All the species of this group are pelagic fishes, widely diffused in the warm seas. All cling by their cephalic disks to sharks, barracudas, and other free-swimming fishes, and are carried about the seas by these. They do not harm the shark except by slightly impeding its movement. They are carnivorous fishes, feeding on sardines, young herring, and the like. When a shark, taken on the hook, is drawn out of the water the sucking-fish leaves it instantly, and is capable of much speed in swimming on its own account. These fishes are all dusky in color, the belly as dark as the back, so as to form little contrast to the color of the shark.

The commonest species, _Leptecheneis naucrates_, called pegapega or pegador in Cuba, reaches a length of about two feet and is almost cosmopolitan in its range, being found exclusively on the larger sharks, notably on _Carcharias lamia_. It has 20 to 22 plates in its disk, and the sides are marked by a dusky lateral band.

Almost equally widely distributed is the smaller remora, or shark-sucker (_Echeneis remora_), with a stouter body and about 18 plates in the cephalic disk. This species is found in Europe, on the coast of New York, in the West Indies, in California, and in Japan, but is nowhere abundant. Another widely distributed species is _Remorina albescens_ with 13 plates in its disk. _Remoropsis brachyptera_, with 15 plates and a long soft dorsal, is also occasionally taken. _Rhombochirus osteochir_ is a rare species of the Atlantic with 18 plates, having the pectoral rays all enlarged and stiff. The louse-fish (_Phtheirichthys lineatus_) is a small and slender remora having but 10 plates in its disk. It is found attached, not to sharks, but to barracudas and spearfishes.

A fossil remora is described from the Oligocene shales in Glarus, Switzerland, under the name of _Opisthomyzon glaronensis_. It is characterized by the small disk posteriorly inserted. Its vertebræ are 10 + 13 = 24 only. Dr. Storms gives the following account of this species:

"A careful comparison of the proportion of all the parts of the skeleton of the fossil _Echeneis_ with those of the living forms, such as _Echeneis naucrates_ or _Echeneis remora_, shows that the fossil differs nearly equally from both, and that it was a more normally shaped fish than either of these forms. The head was narrower and less flattened, the preoperculum wider, but its two jaws had nearly the same length. The ribs, as also the neural and hæmal spines, were longer, the tail more forked, and the soft dorsal fin much longer. In fact it was a more compressed type, probably a far better swimmer than its living congeners, as might be expected if the smallness of the adhesive disk is taken into account."

Concerning the relations of the _Discocephali_ Dr. Gill has the following pertinent remarks:

"The family of _Scomberoides_ was constituted by Cuvier for certain forms of known organization, among which were fishes evidently related to _Caranx_, but which had free dorsal spines. Dr. Günther conceived the idea of disintegrating this family because, _inter alias_, the typical _Scomberoides_ (family _Scombridæ_) have more than 24 vertebræ and others (family _Carangidæ_) had just 24. The assumption of Cuvier as to the relationship of _Elacate_ (_Rachycentron_) was repeated, but inasmuch as it had 'more than 24 vertebræ' (it had 25 = 12 + 13) it was severed from the free-spined _Carangidæ_ and associated with the _Scombridæ_. _Elacate_ has an elongated body, flattened head, and a longitudinal lateral band; therefore _Echeneis_ was considered to be next allied to _Elacate_ and to belong to the same family. The very numerous differences in structure between the two were entirely ignored, and the reference of the _Echeneis_ to the _Scombridæ_ is simply due to assumption piled on assumption. The collocation need not, therefore, longer detain us. The possession by _Echeneis_ of the anterior oval cephalic disk in place of a spinous dorsal fin would alone necessitate the isolation of the genus as a peculiar family. But that difference is associated with almost innumerable other peculiarities of the skeleton and other parts, and in a logical system it must be removed far from the _Scombridæ_, and probably be endowed with subordinal distinction. In all essential respects it departs greatly from the type of structure manifested in the _Scombridæ_ and rather approximates—but very distantly—the _Gobioidea_ and _Blennioidea_. In those types we have in some a tendency to flattening of the head, of anterior development of the dorsal fin, a simple basis cranii, etc. Nevertheless there is no close affinity, nor even tendency to the extreme modification of the spinous dorsal exhibited by _Echeneis_. In view of all these facts _Echeneis_, with its subdivisions, may be regarded as constituting not only a family but a suborder.... Who can consistently object to the proposition to segregate the _Echeneididæ_ as a suborder of teleocephalous fishes? Not those who consider that the development of three or four inarticulate rays (or even less) in the front of the dorsal fin is sufficient to ordinarily differentiate a given form from another with only one or two such. Certainly the difference between the constituents of a disk and any rays or spines is much greater than the mere development or atrophy of articulations. Not those who consider that the manner of depression of spines, whether directly over the following, or to the right or left alternately, are of cardinal importance; for such differences, again, are manifestly of less morphological significance than the factors of a suctorial disk. Nevertheless there are doubtless many who will passively resist the proposition because of a conservative spirit, and who will vaguely refer to the development of the disk as being a 'teleological modification,' and as if it were not an actual fact and a development correlated with radical modifications of all parts of the skeleton at least. But whatever may be the closest relations of _Echeneis_, or the systematic value of its peculiarities, it is certain that it is not allied to _Elacate_ any more than to hosts of scombroid, percoid, and kindred fishes, and that it differs _in toto_ from it notwithstanding the claims that have been made otherwise. It is true that there is a striking resemblance, especially between the young—almost as great, for example, as that between the placental mouse and the marsupial _Antechinomys_—but the likeness is entirely superficial, and the scientific ichthyologist should be no more misled than would be the scientific therologist by the likeness of the marsupial and placental mammals."

=Suborder Tæniosomi, the Ribbon-fishes.=—The suborder _Tæniosomi_ (ταινία, ribbon; σῶμα, body), or ribbon-fishes, is made up of strange inhabitants of the open seas, perhaps aberrant derivatives of the mackerel stock. The body is greatly elongate, much compressed, extremely fragile, covered with shining silvery skin. The ribbon-fishes live in the open sea, probably at no very great depth, but are almost never taken by collectors except when thrown on shore in storms or when attacked by other fishes and dragged above or below their depth. When found they are usually reported as sea-serpents, and although perfectly harmless, they are usually at once destroyed by their ignorant captors. The whole body is exceedingly fragile; the bones are porous, thin, and light, containing scarcely any calcareous matter. In the _Tæniosomi_ the ventral fins are thoracic, formed of one or a few soft rays. More remarkable is the character of the caudal fin, which is always distorted and usually not in line with the rest of the body. The teeth are small. The general structure is not very different from that of the cutlass-fishes, _Trichiuridæ_, and other degraded offshoots from the scombroid group. The species are few and, from the nature of things, very imperfectly known. Scarcely any specimens are perfectly preserved. When dried the body almost disappears, both flesh and bones being composed chiefly of water.

=The Oarfishes: Regalecidæ.=—The _Regalecidæ_, or oarfishes, have the caudal fin obsolete and the ventrals reduced to long filaments, thickened at the tip. The species reach a length of twenty or thirty feet, and from their great size, slender forms, and sinuous motion have been almost everywhere regarded as sea-serpents. The very long anterior spines of the dorsal fin are tipped with red, and the fish is often and not untruthfully described as a sea-serpent "having a horse's head with a flaming red mane."

The great oarfish, _Regalecus glesne_ (see Fig. 237, Vol. I) was long known to the common people of Norway as king of the herrings, it being thought that to harm it would be to drive the herring to some other coast. The name "king of the herrings" went into science as _Regalecus_, from _rex_, king, and _halec_, herring. The Japanese fancy, which runs in a different line, calls the creature "Dugunonuatatori," which means the "cock of the palace under the sea."

The Atlantic oarfish is named _Regalecus glesne_, from the Norwegian farm of Glesnæs, where the first recorded specimen, described by Ascanius, was taken 130 years ago. Since then the species has been many times found on the shores of Great Britain and Norway, and once at Bermuda.

In this species the body is half-transparent, almost jelly-like, light blue in color, with some darker cross-stripes, and the head has a long jaw and a high forehead, suggesting the head of a horse. The dorsal fin begins on the head, and the first few spines are very long, each having a red tuft on the end. When the animal is alive these spines stand up like a red mane.

The creature is harmless, weak in muscle as well as feeble in mind. It lives in the deep seas, all over the world. After great storms it sometimes comes ashore. Perhaps this is because for some reason it has risen above its depth and so lost control of itself. When a deep-water fish rises to the surface the change of pressure greatly affects it. Reduction of pressure bursts its blood-vessels, its swim-bladder swells, if it has one, and turns its stomach inside out. If a deep-water fish gets above its depth it is lost, just as surely as a surface fish is when it gets sunk to the depth of half a mile.

Sometimes, again, these deep-sea fishes rush to the shore to escape from parasites, crustaceans that torture their soft flesh, or sharks that would tear it.

Numerous specimens have been found in the Pacific, and to these several names have been given, but the species are not at all clearly made out. The oldest name is that of _Regalecus russelli_, for the naturalist Patrick Russell, who took a specimen at Vizagapatam in 1788. I have seen two large examples of _Regalecus_ in the museum at Tokio, and several young ones have recently been stranded on the Island of Santa Catalina in southern California. A specimen twenty-two feet long lately came ashore at Newport in Orange County, California. The story of its capture is thus told by Mr. Horatio J. Forgy, of Santa Ana, California:

"On the 22d of February, 1901, a Mexican Indian reported at Newport Beach that about one mile up the coast he had landed a sea-serpent, and as proof showed four tentacles and a strip of flesh about six feet long. A crowd went up to see it, and they said it was about twenty feet long and like a fish in some respects and like a snake in others. Mr. Remsberg and I, on the following day, went up to see it, and in a short time we gathered a crowd and with the assistance of Mr. Peabody prepared the fish and took the picture you have received.

"It measured twenty-one feet and some inches in length, and weighed about 500 or 600 pounds.

"The Indian, when he reported his discovery, said it was alive and in the shallow water, and that he had landed it himself.

"This I very much doubt, but when it was first landed it was in a fine state of preservation and could have easily been shipped to you, but he had cut it to such an extent that shipment or preservation seemed out of the question when we first saw it.

"At the time it came ashore an unusual number of peculiar fishes and sharks were found. Among others, I found a small oarfish about three feet long in a bad state of preservation in a piece of kelp. One side of it was nearly torn off and the other side was decayed."

Mr. C. F. Holder gives this account of the capture of oarfishes in southern California:

"From a zoological point of view the island of Santa Catalina, which lies eighteen miles off the coast of Los Angeles County, southern California, is very interesting, many rare animals being found there. Every winter the dwellers of the island find numbers of argonaut-shells, and several living specimens have been secured, one for a time living in the aquarium which is maintained here for the benefit of students and the entertainment of visitors. A number of rare and interesting fishes wander inshore from time to time. Several years ago I found various Scopeloid fishes, which up to that time had been considered rare, and during the past few years I have seen one oarfish (_Regalecus russelli_) alive, while another was brought to me dead. From reports I judge that a number of these very rare fishes have been observed here. The first was of small size, not over two feet in length, and was discovered swimming in shallow water along the beach of Avalon Bay. I had an opportunity to observe the radiant creature before it died. Its 'topknot'—it can be compared to nothing else—was a vivid red or scarlet mass of seeming plumes—the dorsal fins, which merged into a long dorsal fin, extending to the tail. The color of the body was a brilliant silver sheen splashed with equally vivid black zebra-like stripes, which gave the fish a most striking appearance.

"The fish was a fragile and delicate creature, a very ghost of a fish, which swam along where the water gently lapped the sands with an undulatory motion, looking like one of its names—the ribbon-fish. The fortunate finder of this specimen could not be persuaded to give it up or sell it, and it was its fate to be pasted upon a piece of board, dried in the sun as a 'curio,' where, as if in retaliation at the desecration of so rare a specimen, it soon disappeared.

"This apparently was the first oarfish ever seen in the United States, so at least Dr. G. Brown Goode wrote me at the time that it had not been reported. In 1899 another oarfish was brought to me, evidently having been washed in after a storm and found within a few yards of the former at Avalon. The discoverer of this specimen also refused to allow it to be properly preserved, or to donate or sell it to any one who would have sent it to some museum, but, believing it valuable as a 'curio,' also impaled it, the delicate creature evaporating under the strong heat of the semitropic sun.

"This, as stated, was the second fish discovered, and during the past winter (1900) a fine large specimen came in at Newport Beach, being reported by H. J. Forgy, of Santa Ana. The newspapers announced that a Mexican had found a young sea-serpent at Newport, and investigation showed that, as in hundreds of similar instances, the man had found a valuable prize without being aware of it. According to the account, the discoverer first saw the fish alive in the surf and hauled it ashore. Being ignorant of its value, he cut it up, bringing in a part of the scarlet fins and a slice of the flesh. This he showed to some men, and led the way to where lay the mutilated remains of one of the finest oar-or ribbon-fishes ever seen. The specimen was twenty-one feet in length, and its weight estimated at five hundred pounds. The finder had so mutilated it that the fish was ruined for almost any purpose. If he had packed it in salt, the specimen would have returned him the equivalent of several months' labor. Apparently the man had cut it up in wanton amusement.

"This recalls a similar incident. I was on one occasion excavating at San Clemente Island, and had remarked that it was a singular fact that all the fine stone ollas were broken. 'Nothing strange about that,' said a half-breed, one of the party. 'I used to herd sheep here, and we smashed mortars and ollas to pass away time.'"

=The Dealfishes: Trachypteridæ.=—The family of _Trachypteridæ_ comprises the dealfishes, creatures of fantastic form and silvery coloration, smaller than the oarfishes and more common, but of similar habit.

Just as in Norway the fantastic oarfish was believed to be the king of the herrings and cherished as such, so among the Indians of Puget Sound another freak fish is held sacred as the king of the salmon. The people about Cape Flattery believe that if one does any harm to this fish the salmon will at once leave the shores. This fable led the naturalists who first discovered this fish to give it its name of _Trachypterus rex-salmonorum_.

In Europe a similar species (_Trachypterus atlanticus_) has long been known by the name of dealfish, or vogmar, neither of these names having any evident propriety.

The dealfish is one of the most singular of all the strange creatures of the sea. It reaches a length of three or four feet. Its body is thin as a knife and would be transparent were it not covered over with a shining white pigment which gives to the animal the luster of burnished silver. On this white surface is a large black blotch or two, but no other colors. The head is something like that of the oarfish, to which animal the dealfish bears a close relationship. Both have small teeth and neither could bite if it would, and neither wants to, for they are creatures of the most inoffensive sort. On the head of the dealfish, where the oarfish has its mane, is a long, streamer-like fin. At the end of the tail, instead of the ordinary caudal fin, is a long, slim fin which projects directly upwards at right angles to the direction of the back-bone. No other fish shows this strange peculiarity.

The dealfish swims in the open sea close to the surface of the water. It does not often come near shore, but it is occasionally blown on the beach by storms. _Trachypterus rex-salmonorum_ has been recorded two or three times from Puget Sound and twice from California. The finest specimen known, the one from which our figure is taken, was secured off the Farallones in 1895 by a fisherman named W. C. Knox, and by him sent to Stanford University. The specimen is perfect in all its parts, a condition rare with these fragile creatures, and its picture gives a good idea of the mysterious king of the salmon.

Four of these fishes have been obtained on the coast of Japan, and have been described and figured by the present writer in the annals of the Imperial University of Tokyo. These are different from the California species and are named _Trachypterus ishikawæ_, but they show the same bright silver color and the same streamers on the head and tail. Probably they, too, in Japan are kings of something or other, or perhaps silver swans from the submarine palace, for along such lines the Japanese fancy is more likely to run.

The young of the dealfish has the caudal symmetrical, and the dorsal spines and ventral rays produced in very long streamers.

According to Goode and Bean, the dealfishes are "true deep-sea fishes, which live at very great depths, and are only found when floating dead on the surface or washed ashore by the waves. Almost nothing is known of their habits except through Nilsson's observations in the far north. This naturalist, as well as Olafson, appears to have had the opportunity of observing them in life. They say that they approach the shore at flood-tide on sandy, shelving bottoms, and are often left by the retreating waves. Nilsson's opinion is that its habits resemble those of the flatfishes, and that they move with one side turned obliquely upward, the other toward the ground; and he says that they have been seen on the bottom in two or three fathoms of water, where the fishermen hook them up with the implements employed to raise dead seals, and that they are slow swimmers. This is not necessarily the case, however, for the removal of pressure and the rough treatment by which they were probably washed ashore would be demoralizing, to say the least. _Trichiurus_, a fish similar in form, is a very strong, swift swimmer, and so is _Regalecus_. Whether or not the habits of _Trachypterus arcticus_, on which these observations were made, are a safe guide in regard to the other forms is a matter of some doubt, but it is certain that they live far from the surface, except near the Arctic Circle, and that they only come ashore accidentally. They have never been taken by the deep-sea dredge or trawl-net, and indeed perfect specimens are very rare, the bodies being very soft and brittle, the bones and fin-rays exceedingly fragile. A considerable number of species have been described, but in most instances each was based on one or two specimens. It is probable that future studies may be as fruitful as that of Emery, who, by means of a series of twenty-three specimens, succeeded in uniting at least three of the Mediterranean species which for half a century or more had been regarded as distinct. The common species of the eastern Atlantic, _Trachypterus atlanticus_, is not rare, one or more specimens, according to Günther, being secured along the coast of northern Europe after almost every severe gale. We desire to quote the recommendation of Dr. Günther, and to strongly urge upon any one who may be so fortunate as to secure one of these fishes that no attempt should be made to keep it entire, but that it should be cut into short lengths and preserved in the strongest spirits, each piece wrapped separately in muslin."

The family of _Stylephoridæ_ is known from a single specimen of the species, _Stylephorus chordatus_, taken off Cuba in 1790. In this form the tail ends in a long, whip-like appendage, twice as long as the head.

No fossil dealfishes or oarfishes are known.