ary. We are aware that, if we reclaim some
wild animal, and modify its habits and instincts by domestication, it
may, if it escapes, become in a few years nearly as wild and untractable
as ever; and if the same individual be again retaken, it may be reduced
to its former tame state. A plant is sown in a prepared soil, in order
that the petals of its flowers may multiply, and their color be
heightened or changed: if we then withhold our care, the flowers of this
same species become again single. In these, and innumerable other
instances, we must suppose that the species was produced with a certain
number of qualities; and, in the case of animals, with a variety of
instincts, some of which may or may not be developed according to
circumstances, or which, after having been called forth, may again
become latent when the exciting causes are removed.

Now, the formation of races seems the necessary consequence of such a
capability in species to vary, if it be a general law that the offspring
should very closely resemble the parent. But, before we can infer that
there are no limits to the deviation from an original type which may be
brought about in the course of an indefinite number of generations, we
ought to have some proof that, in each successive generation,
individuals may go on acquiring an equal amount of new peculiarities,
under the influence of equal changes of circumstances. The balance of
evidence, however, inclines most decidedly on the opposite side; for in
all cases we find that the quantity of divergence diminishes after a few
generations in a very rapid ratio.

_Species susceptible of modification may be greatly altered in a few
generations._--It cannot be objected, that it is out of our power to go
on varying the circumstances in the same manner as might happen in the
natural course of events during some great geological cycle. For in the
first place, where a capacity is given to individuals to adapt
themselves to new circumstances, it does not generally require a very
long period for its development: if, indeed, such were the case, it is
not easy to see how the modification would answer the ends proposed, for
all the individuals would die before new qualities, habits, or instincts
were conferred.

When we have succeeded in naturalizing some tropical plant in a
temperate climate, nothing prevents us from attempting gradually to
extend its distribution to higher latitudes, or to greater elevations
above the