t either man or some
other enemies have harassed most of them into migrations; "those whose
nature was sufficiently flexible to lend itself to the new situation
spread far and wide, while others have had no resource but the deserts
in the neighbourhood of their own countries."[123]

Since food and climate, and still less man's empire over them, can have
but little effect upon wild animals, Buffon refers their principal
varieties in great measure to their sexual habits, variations being much
less frequent among animals that pair and breed slowly, than among those
which do not mate and breed more freely. After running rapidly over
several animals, and discussing the flexibility or inflexibility of
their organizations, he declares the elephant to be the only one on
which a state of domestication has produced no effect, inasmuch as "it
refuses to breed under confinement, and cannot therefore transmit the
badges of its servitude to its descendants."[124]

Here is an example of Buffon's covert manner, in the way he maintains
that descent with modification may account not only for specific but for
generic differences.

"But after having taken a rapid survey of the varieties which indicate
to us the alterations that each species has undergone, there arises a
broader and more important question, how far, namely, species themselves
can change--how far there has been an older degeneration, immemorial
from all antiquity, which has taken place in every family, or, if the
term is preferred, _in all the genera_ under which those species are
comprehended which neighbour one another without presenting points of
any very profound dissimilarity? We have only a few isolated species,
such as man, which form at once the species and the whole genus; the
elephant, the rhinoceros, the hippopotamus, and the giraffe form genera,
or simple species, which go down in a single line, with no collateral
branches. All other races appear to form families, in which we may
perceive a common source or stock from which the different branches seem
to have sprung in greater or less numbers according as the individuals
of each species are smaller and more fecund."[125]

I can see no explanation of the introduction of this passage unless that
it is intended to raise the question whether modification may be not
only specific but generic, the point of the paragraph lying in the
words "dans chaque famille, _ou si l'on veut, dans chacun des genres_."
We are told