particular in which plants will not vary from each other; it is quite
possible that one of our imaginary plants may vary in such a character
as the thickness of the integument of its seeds; it might happen that
one of the plants might produce seeds having a thinner integument, and
that would enable the seeds of that plant to germinate a little quicker
than those of any of the others, and those seeds would most inevitably
extinguish the forty-nine times as many that were struggling with them.

I have put it in this way, but you see the practical result of the
process is the same as if some person had nurtured the one and
destroyed the other seeds. It does not matter how the variation is
produced, so long as it is once allowed to occur. The variation in the
plant once fairly started tends to become hereditary and reproduce
itself; the seeds would spread themselves in the same way and take part
in the struggle with the forty-nine hundred, or forty-nine thousand,
with which they might be exposed. Thus, by degrees, this variety, with
some slight organic change or modification, must spread itself over the
whole surface of the habitable globe, and extirpate or replace the
other kinds. That is what is meant by NATURAL SELECTION; that is the
kind of argument by which it is perfectly demonstrable that the
conditions of existence may play exactly the same part for natural
varieties as man does for domesticated varieties. No one doubts at all
that particular circumstances may be more favourable for one plant and
less so for another, and the moment you admit that, you admit the
selective power of nature. Now, although I have been putting a
hypothetical case, you must not suppose that I have been reasoning
hypothetically. There are plenty of direct experiments which bear out
what we may call the theory of natural selection; there is extremely
good authority for the statement that if you take the seed of mixed
varieties of wheat and sow it, collecting the seed next year and sowing