Geoffroy said that according to Lamarck organisms designed
themselves, {20a} and endorsed this, as to a great extent he did, he
still does not appear to have seen that either he or Lamarck were in
reality reintroducing design into organism; he does not appear to
have seen this more than Lamarck himself had seen it, but, on the
contrary, like Lamarck, remained under the impression that he was
opposing teleology or purposiveness.

Of course in one sense he did oppose it; so do we all, if the word
design be taken to intend a very far-foreseeing of minute details, a
riding out to meet trouble long before it comes, a provision on
academic principles for contingencies that are little likely to
arise. We can see no evidence of any such design as this in nature,
and much everywhere that makes against it. There is no such
improvidence as over providence, and whatever theories we may form
about the origin and development of the universe, we may be sure
that it is not the work of one who is unable to understand how
anything can possibly go right unless he sees to it himself. Nature
works departmentally and by way of leaving details to subordinates.
But though those who see nature thus do indeed deny design of the
prescient-from-all-eternity order, they in no way impugn a method
which is far more in accord with all that we commonly think of as
design. A design which is as incredible as that a ewe should give
birth to a lion becomes of a piece with all that we observe most
frequently if it be regarded rather as an aggregation of many small
steps than as a single large one. This principle is very simple,
but it seems rather difficult to understand. It has taken several
generations before people would admit it as regards organism even
after it was pointed out to them, and those who saw it as regards
organism still failed to understand it as regards design; an