limbs ended with three toes armed with small hoofs, but one of its
relatives of the same time has a vestige of another digit on the hind
foot. By the geological time mentioned, therefore, the earliest true
horses had already lost some of the toes that their progenitors possessed.
In the Miocene the extinct species, obviously descended from the Eocene
forms, had lost more of their toes; still higher, that is, in the rocks
formed during succeeding periods of time, the animals of this division are
much larger and each of their feet has only three toes, of which the
middle one is the largest while the ones on the sides are small and
withdrawn from the ground so as to appear as useless vestiges. To produce
modern horses and zebras from these nearer ancestors, few additional
changes in the structure of the feet are necessary, for the lateral toes
need only to become a little more reduced and the middle one to enlarge
slightly to give the one-toed limb of modern types, with its splint-like
vestiges still in evidence to show that the ancestor's foot comprised more
of these terminal elements. Comparing the animals of successive periods,
these and other skeletal structures demonstrate that the ancestry of each
group of species is to be found in the animals of the preceding epoch, and
that the whole history of horses is one of natural transformation,--in a
word, of evolution.

No less interesting in their own way are the remains of other hoofed forms
that lead down to the elephants of to-day and to the mammoth and mastodon
of relatively recent geologic times. Common sense would lead to the
conclusion that a form like a modern tapir was the prototype from which
these creatures have arisen, and common sense would lead us to expect that
if any fossils of the ancestors of the modern group of elephants occurred
at all they would be like tapirs. Thus a fossil of much significance in
this connection is _Moeritherium_, whose remains have been found in the
rocks exposed in the Libyan desert, for this creature was practically a