Bramatherium and Hydaspuherium of India the horns of the males were complex, those of the former including an occipital air, while those of the latter arise from a common base. In both enera, as in the okapi, there is a vacuity in front of the orbit. Largest of all is Swatherzum, typically from the Lower Pliocene of Northern India, but also recorded from Adrianople, in which the skull of the male IS short and wide, with a pair of simple conical horns above the eye, and a huge branching pair at the vertex. Libythenum is an allied form from North Africa. Whether the Gzrajdae were originally an African or a Euro-Asiatic group there is not yet sufficient evidence to decide. The family is unrepresented 1n the western hemisphere.

Cervzdae.-In the deer-tribe, or Cerwdae, the lower canine, as in the two following families, IS simple and similar to the incisors. The frontal appendages, when present, are confined (except in the case of the reindeer) to the males, and take the form of antlers, that is to say of type No. II. in the foregoing description. As a general rule, the molars, and more especially the first, are partially brachyodont (short-crowned), although they are taller in the chital (C67'1/MS axzs). In the skull there are two orifices to the lachrymal duct, situated on or inside the rim of the orbit. A preorbital vacuity of such dimensions as to exclude the lachrymal bone from articulation with the nasal. Upper canines usually present in both sexes, and sometimes attaining a very great size in the male (see fig. 3).

F IG. 3.-Skull of Chinese Water-Deer, Hydrelaphus inermis (adult male), a Deer without Antlers, but with largely developed upper canine teeth.

Lateral digits of both fore and hind feet almost always present, and frequently the lower ends of the metacarpals and the metatarsals as well. Placenta with few cotyledons. Gall-bladder absent (except in the musk-deer, Moschus). This family contains numerous species, having a wide geographical distribution, ranging in the New World from the Arctic circle as far south as Patagonia, and in the Old World throughout the whole of Europe and Asia, but absent in Africa south of the Sahara, and, of course, Australasia. Evidently the family originated in the northern continent of the Old World, from which an entrance was effected by way of Bering Strait into America. Some of the more northern American deer, such as the wapiti, reindeer and elk (moose), are closely allied to Old World species, but there is also a group of exclusively American deer (Mazama)-the onl one found in Central and South America »-the members of which are unlike any living Old World deer, and these must be regarded as having reached the western hemisphere at an earlier date than the wapiti, reindeer and elk (see EER, ELK, FALLOW-DEER, MUNTJAC, llUSK-DEER, PERE DAVID'S DEER, REINDEER, ROEBUCK, WATER-DEER, &c.).

Remains of deer more or less nearly allied to species inhabiting the same districts are found over the greater part of the present habitat of the family. It is noteworthy, however, that certain Pliocene European deer (Anoglochzs) appear to be closely allied to the modern American deer (Mamma). As we descend in the geological series the deer have simpler antlers, as in the European Miocene Dfcrocerus; while in the Oligocene Amphztragulus, Dremolherzum and Palaeameryx, constituting the family Palaeomeryczdae, antlers were absent. and the crowns of the molars so low that the whole depth of the hollows between the crestentic columns is completely visible Most of these animals were of small size, and many ad long upper canines, like those of the existing Hydrelaphus; while in all there was no depression for a gland in front of the eye From North America l1ave been obtained remains of certail ruminants which seem in some degree intermediate between deer and the prong buck. Of one of these a complete skeleton was obtained in 1901 from the Middle Miocene deposits of north-eastern Colorado, and as mounted stands 19 in in height at the withers. With the ecept1on that the right antler is malformed and partially aborted, and that the bones of the lateral toes have been lost, the skeleton is practically complete. The one complete antler has a well-marked burr and a long undivided beam, which eventually forks. After this there is a bifurcation of the hinder branch, thus producing three tines. From the presence of these well-marked antlers the skeleton would at first sight be set down as that of a small and primitive deer, conforming in regard to the structure of these appendages to the American type of the group. Mr W. D. I

1

Matthew shows, however, that the skeleton of Merycodus, as the extinct ruminant is called, differs markedly from that of all deer. The most noteworthy point of distinction is in the skull, in which the facial portion is sharply bent down on the posterior basal axis in the fashion characteristic of the hollow-horned ruminants (oxen, antelopes, &c.), and the American prong buck, instead of running more or less nearly parallel to the same, as in deer. Again, the cheek-teeth have the tall crowns characteristic of a large number of representatives of the first group and of the prong buck, thereby showing that Merycodus can scarcely be regarded as a primitive type. As regards the general structure of the rest of the skeleton, it must suffice to say that this agrees closely with that of the antelopes and the prong buck, and differs markedly from the cervine type. In the absence of any trace of the lower extremities of the metacarpal and inetatarsal bones of the lateral toes the skeleton differs from the American deer, and resembles those hollow-horned ruminants in which these toes persist.

As a whole Merycodus presents a curious mixture of cervine and antilopine character. To explain these, two alternatives are offered by the describer. Either we must regard Merycodus as a deer which parallels the antelopes and the prong buck in every detail of skeletal structure, or else, like the prong buck, an antelope separated from the main stock at a date sufficiently early to have permitted the development of a distinct type of cranial appendages, namely, antlers in place of true horns. The former alternative, it is urged, involves a parallelism too close and too uniform between unrelated types to have been probable. On the latter view Merycodus, the prong buck (Annlocapra) and the antelopes must be regarded as representing three branches from an original common stock, divergent as regards the structure of their cranial appendages, but parallel in other respects. If, therefore, Antzlocapra deserves to be separated as a family from the Bawdae, the same can scarcely be refused for Merycodus. But American extinct types appear to indicate signs of intimate relationship between antelopes, prong buck and deer, and it may be necessary eventually to amend the current classification. As a temporary measure it seems preferable to regard Merycodus either as representing 9. distinct subfamily of Antzlocapndae or a family by itself, the latter course being adopted by Mr Matthew.

Whatever be the ultimate verdict, the association of antler sand these, be it noticed, conforming almost exactly with the forked type characteristic of American deer-with an antilopine type of skull, skeleton and teeth 1n Merycodus is a most interesting and unexpected feature. Merycodus was named many years ago by Professor J. Leidy on the evidence of imperfect materials, and other remains now known to belon to the same type were subsequently described as Cosoryx, to wiich Blastomer x seems to be alliecli Not till the discovery of the skeleton of the species described by Mr Matthew was it possible to arrive at an adequate conception of the affinities of this remarkable ruminant.

Antzlocaprzdaef-By man modern writers the American prong buck, pronghorn or “ antelope, " alone forming the genus Antzlocapra., is regarded as representing merely a sub-family of the Bomdae, to which latter group the animal is structurally akin. In view of what has been stated in the preceding paragraph with regard to the extinct American genus Meryrodus, it seems, howe er, at least provisionally advisable to allow the prong buck to remain as the type of a family—Antzlocaprzdae. The characteristic of this family —as represented by the prong buck-is that the sheath of the horns is forked, and shed annually, or every few years. The cheek teeth are tall-crowned (hypsodontly and lateral hoofs are wanting (see PRONGBUCK).

Bowdae.-Lastly, we have the great family of hollow-horned ruminants or Bowdae, in which the horns (present in the males at least of all the existing species) take the form of simple non-deciduous hollow sheaths growing upon bon cores. As a rule the molars are tall-crowned (hypsodont). lfsually only one orifice to the lachrymal canal, situated inside the rim of the orbit. Lachrymal bone almost always articulating with the nasal. Canines absent in both sexes. The lateral toes may be completely absent, but more often are represented by the hoofs alone, supported son etirres by a wery rudimentary skeleton, consisting of mere irre ular nodules of bone. Lower ends of the lateral metacarpals ang metatarsals never present. Gall-bladder almost always present, Placenta with many cotyledons.

The Bowdae form a most extensive family, with membeis widely distributed throughout the Old World, with the exception of the Australian region; but in America they are less numerous, and confined to the Arctic and noithern temperate regions, no species being indigenous either to South or Central America. The home of the family was evidently the Old World, whence a small number of forms made their way into North America by way of what is now Bering Strait. It has already been pointed out that the Cervidae originated in the northern continent of the Old World; and it has been suggested that the Bovidae were developed in Africa. Unfortunately, we know at present tactically nothing as to the ast history of the group, all the fossil) species at present discovered) approximating more or less closely to existing types. While admittin, therefore, that there are several facts in favour

of the theory otli an African origin of the Bomdae, final judgment