Your last letter has interested me to an extraordinary degree, and your
truly parsonic advice, "some other wise and discreet person," etc., etc.,
amused us not a little. I will put a concrete case to show what I think A.
Gray believes about crossing and what I believe. If 1,000 pigeons were
bred together in a cage for 10,000 years their number not being allowed to
increase by chance killing, then from mutual intercrossing no varieties
would arise; but, if each pigeon were a self-fertilising hermaphrodite, a
multitude of varieties would arise. This, I believe, is the common effect
of crossing, viz., the obliteration of incipient varieties. I do not deny
that when two marked varieties have been produced, their crossing will
produce a third or more intermediate varieties. Possibly, or probably,
with domestic varieties, with a strong tendency to vary, the act of
crossing tends to give rise to new characters; and thus a third or more
races, not strictly intermediate, may be produced. But there is heavy
evidence against new characters arising from crossing wild forms; only
intermediate races are then produced. Now, do you agree thus far? if not,
it is no use arguing; we must come to swearing, and I am convinced I can
swear harder than you, therefore I am right. Q.E.D.

If the number of 1,000 pigeons were prevented increasing not by chance
killing, but by, say, all the shorter-beaked birds being killed, then the
WHOLE body would come to have longer beaks. Do you agree?

Thirdly, if 1,000 pigeons were kept in a hot country, and another 1,000 in
a cold country, and fed on different food, and confined in different-size
aviary, and kept constant in number by chance killing, then I should expect
as rather probable that after 10,000 years the two bodies would differ
slightly in size, colour, and perhaps other trifling characters; this I
should call the direct action of physical conditions. By this action I
wish to imply that the innate vital forces are somehow led to act rather