y be
transmitted to offspring in that perfection which is called "instinct,"
till the habit or experience has been repeated in several generations
with more or less uniformity; for otherwise the impression made will not
be strong enough to endure through the busy and difficult task of
reproduction. This of course involves that the habit shall have
attained, as it were, equilibrium with the creature's sense of its own
needs, so that it shall have long seemed the best course possible,
leaving upon the whole and under ordinary circumstances little further to
be desired, and hence that it should have been little varied during many
generations. We should expect that it would be transmitted in a more or
less partial, varying, imperfect, and intelligent condition before
equilibrium had been attained; it would, however, continually tend
towards equilibrium.

When this stage has been reached, as regards any habit, the creature will
cease trying to improve; on which the repetition of the habit will become
stable, and hence capable of more unerring transmission--but at the same
time improvement will cease; the habit will become fixed, and be perhaps
transmitted at an earlier and earlier age, till it has reached that date
of manifestation which shall be found most agreeable to the other habits
of the creature. It will also be manifested, as a matter of course,
without further consciousness or reflection, for people cannot be always
opening up settled questions; if they thought a matter all over yesterday
they cannot think it all over again to-day, what they thought then they
will think now, and will act upon their opinion; and this, too, even in
spite sometimes of misgiving, that if they were to think still further
they could find a still better course. It is not, therefore, to be
expected that "instinct" should show signs of that hesitating and
tentative action which results from knowledge that is still so imperfect
as to be actively self-conscious; nor yet that it should grow or vary
perceptibly unless under such changed conditions as shall baffle memory,
and present the alternative of either invention--that is to say,
variation--or death.

But every instinct must have passed through the laboriously intelligent
stages through which human civilisations _and mechanical inventions_ are
now passing; and he who would study the origin of an instinct with its
development, partial transmission, further growth, further transmission,