hen we consider how very few experiments have been made
with natural varieties; while there is good reason for believing that
domestic varieties are exceptionally fertile, partly because one of the
conditions of domestication was fertility under changed conditions, and
also because long continued domestication is believed to have the effect
of increasing fertility and eliminating whatever sterility may exist.
This is shown by the fact that, in many cases, domestic animals are
descended from two or more distinct species. This is almost certainly
the case with the dog, and probably with the hog, the ox, and the sheep;
yet the various breeds are now all perfectly fertile, although we have
every reason to suppose that there would be some degree of infertility
if the several aboriginal species were crossed together for the first
time.

_Parallelism between Crossing and Change of Conditions._

In the whole series of these phenomena, from the beneficial effects of
the crossing of different stocks and the evil effects of close
interbreeding, up to the partial or complete sterility induced by
crosses between species belonging to different genera, we have, as Mr.
Darwin points out, a curious parallelism with the effects produced by
change of physical conditions. It is well known that slight changes in
the conditions of life are beneficial to all living things. Plants, if
constantly grown in one soil and locality from their own seeds, are
greatly benefited by the importation of seed from some other locality.
The same thing happens with animals; and the benefit we ourselves
experience from "change of air" is an illustration of the same
phenomenon. But the amount of the change which is beneficial has its
limits, and then a greater amount is injurious. A change to a climate a
few degrees warmer or colder may be good, while a change to the tropics
or to the arctic regions might be injurious.

Thus we see that, both slight changes of conditions and a slight amount
of crossing, are beneficial; while extreme changes, and crosses between
individuals too far removed in structure or constitution, are injurious.
And there is not only a parallelism but an actual connection between the
two classes of facts, for, as we have already shown, many species of
animals and plants are rendered infertile, or altogether sterile, by the
change from their natural conditions which occurs in confinement or in
cultivation; while, on the other hand, the in