Gorals and serows belong to the subfamily Rupicaprinae which is an early mountain-living offshoot of the Bovidae; it also includes the chamois, takin, and the so-called Rocky Mountain goat of America. The animals are commonly referred to as "goat-antelopes" in order to express the intermediate position which they apparently hold between the goats and antelopes. They are also sometimes called the Rupicaprine antelopes from the scientific name of the chamois ( Rupicapra).

The horns of all members of the group are finely ridged, subcylindrical and are present in both sexes, being almost as long in the female as in the male. Although no one would suspect that the gorals are more closely related to the takins than to the serows, which they resemble superficially, such seems to be the case, but the cranial differences between the two genera are to a certain extent bridged over by the skull of the small Japanese serow (Capricornulus crispus ). This species is most interesting because of its intermediate position. In size it is larger than a goral but smaller than a serow; its long coat and its horns resemble those of a goral but it has the face gland and short tail of a serow. It is found in Japan, Manchuria and southern Siberia.

The principal external difference between the gorals and serows, besides that of size, is in the fact that the serows have a short tail and a well developed face gland, which opens in front of the eyes by a small orifice, while the gorals have a long tail and no such gland.

In the cylindrical form of their horns the serows are similar to some of the antelopes but in their clumsy build, heavy limbs and stout hoofs as well as in habits they resemble goats. The serow has a long, melancholy-looking face and because of its enormous ears the Chinese in Fukien Province refer to it as the "wild donkey" but in Yuen-nan it is called "wild cow."

The specific relationships of the serows are by no means satisfactorily determined. Mr. Pocock, Superintendent of the London Zooelogical Society's Gardens, has recently devoted considerable study to the serows of British India and considers them all to be races of the single species Capricornis sumatrensis. With this opinion I am inclined to agree, although I have not yet had sufficient time in which to thoroughly study the subject in the light of our new material.

These animals differ most strikingly in external coloration, and fall into three groups all of which partake more or less of the characters of each other. Chinese serows usually have the lower legs rusty red, while in Indian races they are whitish, and black in the southern Burma and Malayan forms.

The serows which we killed upon the Snow Mountain can probably be referred to Capricornis sumatrensis milne-edwardsi, those of Fukien obtained by Mr. Caldwell represent the white-maned serow Capricornis sumatrensis argyrochaetes and one which I shot in May, 1917, near Teng-yueh, not far from the Burma frontier, is apparently an undescribed form.

Our specimens have brought out the fact that a remarkable individual variation exists in the color of the legs of these animals; this character was considered to be of diagnostic value, and probably is in some degree, but it is by no means as reliable as it was formerly supposed to be.

Two of the serows killed on the Snow Mountain have the lower legs rusty red, while in two others these parts are buff colored. The animals, all males of nearly the same age, were taken on the same mountain, and virtually at the same time. Their skulls exhibit no important differences and there is no reason to believe that they represent anything but an extreme individual variation.

The two specimens obtained by Mr. Caldwell at Yen-ping are even more surprising. The old female is coal black, but the young male is distinctly brownish-black with a chestnut stripe from the mane to the tail along the mid-dorsal line where the hairs of the back form a ridge. The horns of the female are nearly parallel for half their extent and approach each other at the tips; their surfaces are remarkably smooth. The horns of the young male diverge like a V from the skull and are very heavily ridged. The latter character is undoubtedly due to youth.

These serows are an excellent example of the necessity for collecting a large number of specimens from the same locality. Only by this means is it possible to learn how the species is affected by age, sex and individual variation and what are its really important characters. In the case of the gorals, our Expedition obtained at Hui-yao such a splendid series of all ages that we have an unequaled opportunity for intelligent study. Serows are entirely Asian and found in China, Japan, India, Sumatra and the Malay Peninsula.

On the Snow Mountain we found them living singly at altitudes of from 9,000 to 13,000 feet in dense spruce forests, among the cliffs. The animals seemed to be fond of sleeping under overhanging rocks, and we were constantly finding beds which gave evidence of very extensive use. Apparently serows seldom come out into the open, but feed on leaves and grass while in the thickest cover, so that it is almost impossible to kill them without the aid of dogs or beaters.

Sometimes a serow will lead the dogs for three or four miles, and eventually lose them or it may turn at bay and fight the pack after only a short chase; a large serow is almost certain to kill several of the hounds if in a favorable position with a rock wall at its back. The animal can use its strong curved horns with deadly effect for it is remarkably agile for a beast of its size.