ent is never far distant." Budgett kept many of the fry for several weeks, but "could not get them to feed and eventually they all died."

THE UMBRIDS, OR MUDFISHES.

The pairing and oviposition of Mudfishes or Umbrids has not been observed in America, though I had some for months under observation, but P. Carbonnier appears to have succeeded in mating them and observing the behavior of a parent fish to the eggs and young. The fact, however, has been universally overlooked owing to his misnaming of specimens. He called them "la Fondule (Fundula cyprinodonta, Cuv.)," but no such name appears in any of Cuvier's works and the fish in question was evidently not a Fundulus, and there is little, if any, doubt that it was the common Mudfish (Umbra pygmæa) occurring around New York. The reasons for this identification are given in Science for December 21, 1906. The article by Carbonnier is in the Buletin Mensuel de la Société d'Acclimatation for November, 1874 (pp. 665-674). It seems that the female fish there described. watches over her eggs "with a tender solicitude" during the fortnight of "incubation." "Elle écarte et emporte avec soin tous les sédiments terreux qui pourraient être une cause d'alteration," etc. For full details the original article should be consulted. The identification here made should be confirmed by observations on living fishes. The habits of the species are of unexpected interest.

FIG. 75.-An Ophiocephalid, Ophiocephalus africanus. After Steindachner.

The old Romans had a way of expressing incredulity by reference to nature, but they were not always happy in so doing. Horace took as the acme of improbability the existence of a black swan-rara avis in terris; yet all the swans of Australia are black. Seneca would "go a-fishing in the sea," and contrasted with such a natural course another: "There are many things in this place will come into thy minde, which a man may term after a merry sort, both incredulous and fabulous, that a man should go and fish with his pickaxe and not with nets and hookes."

But with pickax and spade may the Ophiocephaloid fishes be sought in India. The bed of a pond or even a river that appears to

SM 190535

be barren on the surface may harbor a few inches below quite a harvest in the shape of fishes which have found their way there for refuge from the drought. It is this group of Ophiocephalids, too, which furnishes some interesting examples of paternal care.

The Ophiocephalids are elongated fishes with a flattish, declivous head, scales of moderate size on the body and larger ones on the head in oblique rows, a long dorsal and a long but shorter anal (both spineless), and ventral fins behind or under the bases of the pectoral with six soft rays, the outermost being undivided, but articulated distally. These characters are supplemented by a peculiar modification of the branchial apparatus adaptive to aerial respiration, as well as by many cther anatomical characters. The labyrinthiform apparatus is also useful under certain other disturbing conditions, according to Day. "When pollution or poisonous substances find access to rivers or mud is carried down in such quantities as to choke the gills of most forms, these Ophiocephalidæ are almost unaffected, for breathing atmospheric air direct, the presence or absence of fluviatile contamination is not of such material consequence to their existence. They are able to live until the poison has passed downstream and the waters are again purified." The differences from all other fishes, indeed, are so great that there is not universal agreement as to the affinities of the family.

cephalus.

The Ophiocephalids are about 30 in number and found in almost all the tropical and even subtropical regions of FIG. 76.-Ophio- the Old World, 3 occurring in Africa, but all the others in Southern and eastern Asia and the islands to the southward. In the Philippine Islands are found at least 5 species. There is no general English name that can take the place of Ophiocephalids. Snakeheads is a book name that interprets the Latin title, but is rarely used. The word in most popular use is murrul, variously spelled, which is the common vernacular Indian term for species of the family.

When a prolonged drought dries up the ponds or bodies of water in which they have flourished the snakeheads do not necessarily succumb to the consequences, but are provided with means to contend against the adverse conditions. Some burrow into the drying bottom, sometimes to the depth of 3 or 4 feet even, into a moister horizon, and in such places may survive till later rains release them from their entombment and restore the waters to their former beds. The fishes retain moisture enough to keep their gills from dessication and are able to breathe the little air needed direct or even to suspend respiration. Others wander from the receding pond to seek more favorable quarters. Such emigrations are mostly engaged in during night time

and when the ground and air are moist they can travel considerable distances. This capacity is frequently exercised "in the beginning of the rainy season," when they may be "often seen travelling among wet grass; " it is also made use of in the opposite season of incipient drought. It is claimed by some that such emigrations are not entirely pointless, and that instinct guides them to water, but such claims have not been confirmed-at least not fully confirmed.

Peripatetic showmen and jugglers avail themselves of this, for a fish, remarkable aptitude for terrestrial progression, both in India and China. They carry the fishes about, and when a fitting audience can be obtained take the fishes from their receptacle and place them on the ground, when they proceed to "walk" or "crawl about," greatly to the wonder and amusement of the children. "Owing to the breadth of their bodies they are able to progress in a serpentine manner, chiefly by means of their pectoral and caudal fins, first one of the former being advanced and then its fellow." The progression is really of the nature of crawling.

The Ophiocephalids are especially piscivorous, but their voracity

is by no means limited to fishes; indeed they "appear to consider a frog, mouse, or rat as luscious a morsel as a fellow fish. A frog, in fact, proves to be as captivating a bait for some of them at least as it is to a pike. "Near Ganjam," India, a native official told Day "how he had ventured out one night to see how murrul "-the walking fishes or ophiocephals-" were captured. The fisherman was provided with a long flexible bamboo as a rod, and as a bait used a live frog. Hardly had the frog splashed into the water when a moderately sized murrul seized and swallowed it. Desirous of observing what would occur next, the fish was left on the hook as a bait for something else. Before long a large water snake was seen swimming toward it, and soon had the fish inclosed in its capacious jaws, and in this fashion all three were pulled together out of the water." It would from this seem that they are active and feed by night as they also do by day.

Naturally when the species are so numerous, there is considerable diversity in habitat and habits. "Some of them reside in ponds,

others prefer rivers, where they take up their residence in deserted holes which they find in the banks. The pond species delight in lying at the grassy margins, where the water is not deep enough to cover them; and here they are able to respire atmospheric air direct.” So teaches Day (1883). "Some of the ophiocephali prefer dirty to clean water, perhaps for purposes of concealment. When they have stirred up all the sediment and exuded a quantity of mucus they appear to be delighted, their colors become more vivid, and they ascend to their favorite resort, lying amongst the vegetation just beneath the surface of the water. As soon as clean water is given them they become excited, as if they imagined the time had arrived when they should change their abode."

But such places and postures are at times exchanged for others. They may be "frequently seen floating on the surface of the water as if asleep." Such individuals "may be approached very closely.” In some places guns are used and the fish are shot, but "the game usually sinks when killed and has to be dived for or otherwise obtained."

The differences either in size or coloration between the sexes are inconsiderable. Monogamy and parental solicitude are accredited to them. They mostly, it is said, breed twice a year. The eggs are quite large and vary in number in accordance with the size of the female; in one Day estimated there were about 4,700 eggs. Oviposition is provided for in a rude receptacle which is prepared by the male. "The 0. striatus of Mysore is said to construct a nest with its tail among the vegetation near the edges of the tanks, whilst it bites off the ends of the weeds which grow in the water." Over this the parent keeps guard, and his care is exercised not only for the eggs, but for the young until they are developed and strong enough to care for themselves. "Then they are driven away to seek their own subsistence," and those who are "too obstinate to leave," it is believed, are" eaten by their progenitors."

The Ophiocephalids are highly esteemed for table use, "those which inhabit rivers being better flavored than the others which live in sluggish or stagnant water." They form in many places a very important part of the fish supply of the community. They are in prime condition, or at least chiefly demanded, in a perfectly fresh state. To supply such a demand, " in China they are often carried alive in tubs or pails of water, and slices are cut for sale as wanted, the fish selling dear whilst it retains life, while what remains after death is considered as of little value." The entrails are removed before taking the fish around but the muscles remain long responsive to irritation, and "when the irritability of the flesh is so much exhausted that it no longer quivers under the knife, its value is greatly depreciated." This practice, Richardson thinks, "however much it

may shock the feelings when described, is not worse than the crimping of cod on the London fish stalls." It is almost superfluous to add, after this account, that "they are very tenacious of life."

By the natives the adult fishes, as well as their eggs, are "greatly sought after as food."

To insure a continuous supply they are kept to a considerable extent in ponds or tanks and even in aquaria, but their wandering proclivity and climbing capability entail precautions against their emergence. "They are exceedingly difficult to retain in aquaria unless the tops are covered over, as otherwise they manage to escape and proceed on their travels."

STICKLEBACKS.

The Gasterosteids, or sticklebacks, are more or less compressed fusiform or moderately elongated fishes with a partly or wholly naked skin, but generally with a row of high plates along part of the tail at

FIG. 79.-Four-spined Stickleback (Apeltes quadracus). Enlarged.

least; the head is conic and pointed in front or prolonged into a short tube, and the jaws armed with narrow bands of teeth; the ventrals are a short or moderate distance behind the bases of the pectorals, and each has a stout spine and one or two reduced rays; the dorsal furniture consists of two to fifteen free dorsal spines and behind them a dorsal fin with branched rays, and the anal is like the second dorsal.