Skull-dimensions of an adult male (of subspecies _H. e. bakeri_):—Basal length 16 inches, greatest breadth 6·75, muzzle to orbit 10·3. Horns stout and strong, cylindrical, heavily ridged, evenly divergent, curved backward; comparatively short for the size of the animal, good specimens being only from 26 to 30 inches in length, and the longest recorded only 33. _Female_ similar to the male, but the horns more slender, smoother, less heavily ridged, and less strongly curved backward, and neck and body less robust. _Hab._ Africa south of the Sahara, except in the Congo wood-region. It is a well-known and generally accepted fact amongst naturalists that animals which have a wide distribution have also a special tendency to vary, and that if specimens of them from different parts of their ranges are compared, such specimens are usually found not to agree exactly, but to be distinguishable by differential characters more or less evident. When these characters are easily observable and definable their possessors are usually referred to different species, which are supposed to “represent” one another in their respective areas, and are hence often called “representative species.” When the distinguishing characters are slight and less easily recognizable it has recently become the practice, especially among American naturalists, to designate their possessors as “subspecies,” and, in order to indicate this, to add a third “subspecific” name to the ordinary generic and specific terms. This plan we have already adopted in some cases in the present work. But there are many cases in which, either from imperfect evidence or from an insufficient supply of specimens, it is very difficult to decide whether a “local form,” as it may be termed, is better treated of as a species or as a subspecies. And in the present instance we have one of these cases before us. The Roan Antelope is very widely distributed in Africa. From the Cape Colony it extends all up the eastern side of the continent to British East Africa and Sennaar, and is also found on the west coast in Senegal, Togoland, Nigeria, and Angola. Specimens from all these countries present a very general resemblance, and have been considered by most authorities to be identical. On the contrary, other writers have regarded the local forms as distinct, and have separated them under different specific names. We confess that we have not been able (mainly, no doubt, from lack of sufficient specimens to consult) to come to a satisfactory conclusion on this subject; but, for the present, we think it a more prudent course to treat the local forms of this species found in the different districts of Africa as only of subspecific rank, and to class them all under the one specific head as _Hippotragus equinus_.