THE NEAREST ALLIES OF THE HORSE 97 
case, we may surmise that the ancestor in question was a sort 
of dun brown in colour with hog mane, dorsal stripe, and probably 
asinine tail. The lower part of his body and the inner sides of his 
limbs were lighter in colour than the general ground of the skin, 
he probably had zebra marks and castors; was rather small in size 
(say 12 hands), had large head and ears, stout and rather short 
limbs, narrowish mule feet. His habits were those of the wild 
equines in the present day, but his range was much more extensive, 
comprising Hurope, Asia, Africa, and America. In time he became 
extinct in America after undergoing development in several direc- 
tions probably ; in Central Asia he developed caballine characters 
Buch as are seen in the kiang and terpan, and, as they proved use- 
ful to man, were the cause of subjection of this branch of the family 
to man and its enormous modification by art. In Arabia and 
South-East Asia the asinine characters found favourable influences, 
and some of them, as seen in the taniope and onager, also proved 
useful to man, who brought artificial selection to bear in such a way 
that the domesticated ass of the present day differs remarkably from 
the onager, Finally, in Bast Africa to the extreme south, the 
hippotigrine characteristics seem to have found scope for deve- 
lopment. Man found this branch uncontrollable, and it pro- 
bably affords us an example of the least changed descendant 
of equus primigenius. It is generally now admitted that the theory 
of descent can be accepted as a good working hypothesis, 
It certainly receives support from study of the HEquide and, 
on the other hand, would lead us to interesting conclusions with 
regard to them. It would explain to us where the stripes of the 
zebra come from, and why in South Africa the equines differ in 
appearance from those of Asia and North Africa, Taking it into 
consideration, we can to an extent understand that the inborn ten- 
dency of the horses of all kinds to develop donkey stripes and zebra 
marks, a tendency most often manifested in the foal of non-striped 
races which in the South African forms, becomes exaggerated, so that 
we areled through the quagga and Burchell’s zebra to the true zebras, 
which exhibit striation to the full extent. The theory of descent as 
regards zebras is much more satisfactory than that of mimicry. We 
must consider it much more probable that equines spread to South 
Africa and gradually become modified by surroundings than that 
in the South African fauna there were developed animals showing 
full equine characters but differing from the horses and asses in being 
