generally diffused, but they are most abundant in the streams of the Mississippi valley and the Southern States. They are small, perch-like fishes, and may even perhaps belong to that family. A singular variation is exemplified in the development of scales and of the lateral lines; while most of the species have the body more or less completely covered with scales like those of the perches, a few are nearly scaleless. In Amocrypta, for example, the body is almost entirely naked, the scales being confined to the caudal peduncle and lateral line.

A New North American Family.

In a collection of fishes sent to Professor Jordan by Professor H. S. Reynolds from the Little Red River, White County, Ark., were found two specimens of the little fish which Professor Jordan has considered as the representative of not only a new genus, but a new family, or at first a subfamily, of Centrarchida, whose nearest relations are to be found in some extinct forms recently described by Professor Cope. In form and external aspect it is said to have some resemblance to Aphredoderus, but is more compressed. The dorsal fin has five spines, the anal three; the ventrals are thoracic and normal in situation; the branchiostegals are five, and the membrane is broadly united across the pectoral region; the lateral line is absent; the pharyngeal apparatus is unknown. It is suggested by Professor Jordan that the type is most nearly related to the Centrarchida, or perhaps the Cichlidæ. Its position must, however, of course be provisional, and it remains to be determined from the examination of good specimens what are the true relations. The single species has been named Elassoma zonata.

Gigantic Tortoises.

An interesting and noteworthy peculiarity in geographical zoology is the distribution of the gigantic species of land tortoises of the genus Testudo. These are now entirely confined to three archipelagoes-(1) the Galapagos Islands westward of the coast of South America; (2) the Mascarene islands Mauritius and Rodriguez; and (3) the Aldabra group, small islands lying northwest of Madagascar, in lat. 9° 25' S., long. 46° 20′ E. It has long been known that these several archipelagoes were the abodes of large land tortoises,

but the numbers and distinctions of the species were involved in considerable doubt. With a view to settle the doubtful questions, Dr. Günther, of the British Museum, availed himself of the chance afforded him by considerable collections amassed from time to time in London, and chiefly at the British Museum, and has investigated the species and their distinctions. In a recently published richly illustrated volume he has given the results of his final examinations. Not less than fifteen species have been recognized either in a living condition or recently exterminated; these can be segregated into several groups coincident with their geographical range and distinguished by characters derived from the shell, so that hitherto indeterminable shells in museums may be at least referred almost certainly to the archipelago from which they may have been derived, even if no other information can be obtained. The tortoises of the Aldabra Islands have a small anterior unpaired "nuchal" plate in the upper shell, or carapace, and a pair of anterior gular plates in the lower or plastron; those of the Mascarenes have no nuchal, and the gular is single; and those of the Galapagos are also destitute of the nuchal plate, but have a pair of gular ones. So localized are the species that each island almost has its own peculiar form. On the Aldabra Islands still survive at least three, if not four species, and from the Galapagos Islands five living and one extinct species have been obtained; but all of the Mascarene species (five are known from their remains) have become extinct, and since their discovery by the Europeans. The several groups of islands inhabited by these tortoises, as will be recognized, are quite widely distant, and, in the case of the Galapagos, on the one hand, and the remaining islands, on the other, almost as widely separate as could be. They are all situated in the intertropical zone, but have in common otherwise only the negative characteristic of the absence of large terrestrial mammals and that of human inhabitants until recent times. These are doubtless the conditions which favored their development and increase. Historical evidence shows that species existed on all the islands, and were very abundant in individuals. They were exterminated in the Mascarene Islands after their settlement, and but for the absence of permanent settlements on the others would probably have entirely disappeared from the existing

fauna. These tortoises attained a varying size; some even reached a weight of nearly, if not over, 500 pounds, and had shells over six feet long, but most were much smaller. Living to a very great age and with no redoubtable enemies to contend with, they were found by the early visitors to the islands in question in great numbers. According to Leguat, for example (in 1691), "There are such plenty of land turtles in this isle [Rodriguez] that sometimes you see two or three thousand of them in a flock, so that you may go above a hundred paces on their backs." These numbers, however, were soon diminished; the animals afforded savory and nutritive meat which formed a most agreeable variety for the mariner, and consequently they were taken in quantity and stored on shipboard for future consumption. Only the difficulty of access to the islands prevented their complete annihilation.

The question arises, What is the significance of the occurrence of these animals in such widely remote regions without any representatives in intermediate ones? "The naturalists," says Dr. Günther, “who maintain a common origin for allied species, however distant in their habitats, will have to assume a former continuity of land. . . between the Mascarenes and Africa, between Africa and South America, and, finally, between South America and the Galapagos. A continuity of land in this direction is more probable than one in the opposite hemisphere, which would extend over 210°. Indeed, the terrestrial and fresh-water fauna of Tropical America and Africa offer so many points of intimate relationship [see, e. g., Annual Record for 1876, p. clxvii] as very strongly to support such a theory. The tortoises, then, would be assumed to have been spread across the whole of this large area, without being able long to survive the arrival of man or large carnivorous animals." In the face of any formidable enemies, great size in such animals would be disadvantageous, inasmuch as it would render them conspicuous and prevent them from obtaining shelter, while their defensive ability would not be correspondingly increased; small size would be advantageous in relation to their environments, and hence small species of the same genus have survived and still exist over large continental areas.

Birds.

In ornithology there has been apparently an average degree of activity. Many catalogues of species of different regions and descriptions of a number of new species have been published, but none, so far as we are aware, of sufficient interest to be here particularized. The third volume of Sharpe's "Catalogue of the Birds in the British Museum;" parts of Gould's great illustrated works on "The Birds of Asia" and "The Birds of New Guinea;" Sharpe's edition of "The Birds of South Africa" by E. L. Layard; Hartlaub's "Birds of Madagascar and the Neighboring Islands" ("Die Vögel Madagascars und die benachbarten Inselgruppen "); and the first of" A Monograph of the Bucerotidæ, or Family of the Hornbills," by D. G. Elliot, are among the most noteworthy, either on account of extent or as good monographs.

Avifauna of Madagascar.

For many reasons one of the most interesting faunas of earth is that of Madagascar and the Mascarene Islands. In 1861 Dr. Hartlaub had published a small ornithological contribution to the fauna of Madagascar; and last year he gave the results of his own continued studies, and embodied as well those of his numerous fellow-ornithologists, in a volume of over 400 pages. It seems that there are now known as inhabitants of the region in question 284 species: 220 occur in Madagascar and 104 are peculiar to that island; 44 are to be found in the Comoro Islands, about 60 in each Mauritius and Bourbon, and 25 in Rodriguez. "The individuality of the fauna of Madagascar," says Dr. Hartlaub, "is so unique that even that of New Zealand can hardly be compared with it. Wallace's attempted parallel between Madagascar and Africa and the Antilles and South America is, in our eyes, sufficiently disproved by the occurrence in the Antilles of Trochilidae, one of the most characteristic forms of South America. But in Madagascar not a single one of the genera most characteristic of Africa occurs. The originality of the fauna is much too pronounced to allow Madagascar to be treated of only as a 'sub-region' or an aberrant part' of the Ethiopian region." From this conclusion (reproduced in Nature) Professor Newton has, however, dissented, and al

though he does not "wish that its extraordinary peculiarities should be undervalued," he does "not want them to be unduly magnified at the expense of those of the fauna of New Zealand."

The Wild Camel.

Little is known of the camel in a truly untamed condition —that is, not in simply a feral state; but during the past year observations have been recorded respecting the wild animals occurring in Central Asia. The camel is still found. in an aboriginal condition in the Desert of Gobi. It is twohumped, and, according to Mr. Harkloff, "the size is nearly that of the tame; but it is larger and higher on the legs. It is of a darker color than the tame; and the white around the nose is much clearer and paler. In the spring they pair, and the time of gestation is the same as that with the tame camel. The Tanguts and Kirgizes hunt the wild camel and eat its flesh; also they use the hair. The wild camel is said not to be shy, and accordingly not difficult to obtain." According to Major Tichannoff, it cannot easily, if at all, be tamed. The voice is not so strong as that of the tamed camel.

Deer's Antlers.

The morphology of the antlers of the deer has been investigated by Messrs. A. H. Garrod and Theodore Gill. Professor Garrod has suggested that the typical antler is composed of a primarily bifurcate beam and a brow antler springing from its base anteriorly, and that the differences between the species result from the greater or less development, or the atrophy, of one or other of these elements. Several cases, however, are not explicable by this hypothesis. Professor Gill considers the antlers "either as simple spikes or with a tendency to bifurcation, especially (but not exclusively) in the direction of the varying greatest or axial growth," and has applied a new terminology to indicate at once the order of development and as a convenient device for descriptive zoology. (1) The simple spikes of the first year and their after-growths are designated protoceres; (2) the anterior offshoots of the second year deuteroceres; and the succeeding (3) third, (4) fourth, and (5) fifth anterior offshoots, respectively, (3) tritoceres, (4) tetartoceres, and (5) pemptoceres. The chief differences in the several genera of deer with com