er in which it holds equally good upon the vastest and
the smallest scale, and the completeness of its accord with our ideas of
what must inevitably happen when a like combination is placed in
circumstances like those in which it was placed before--when we bear in
mind all this, is it possible not to connect the facts together, and to
refer cycles of living generations to the same unalterableness in the
action of like matter under like circumstances which makes Jupiter and
Saturn revolve round the sun, or the piston of a steam-engine move up and
down as long as the steam acts upon it?

But who will attribute memory to the hands of a clock, to a piston-rod,
to air or water in a storm or in course of evaporation, to the earth and
planets in their circuits round the sun, or to the atoms of the universe,
if they too be moving in a cycle vaster than we can take account of?
{198a} And if not, why introduce it into the embryonic development of
living beings, when there is not a particle of evidence in support of its
actual presence, when regularity of action can be ensured just as well
without it as with it, and when at the best it is considered as existing
under circumstances which it baffles us to conceive, inasmuch as it is
supposed to be exercised without any conscious recollection? Surely a
memory which is exercised without any consciousness of recollecting is
only a periphrasis for the absence of any memory at all. {198b}

REPUTATION--MEMORY AT ONCE A PROMOTER AND A DISTURBER OF UNIFORMITY OF
ACTION AND STRUCTURE. (CHAPTER XII. OF UNCONSCIOUS MEMORY.)

To meet the objections in the two foregoing chapters, I need do little
more than show that the fact of certain often inherited diseases and
developments, whether of youth or old age, being obviously not due to a
memory on the part of offspring of like diseases and developments in the
parents, does not militate against supposing that embryonic and youthful
development generally is due to memory.

This is the main part of the objection; the rest resolves itself into an
assertion that there is no evidence in support of instinct and embryonic
development being due to memory, and a contention that the necessity of
each particular moment in each particular case is sufficient to account
for the facts without the introduction of memory.

I will deal with these two last points briefly first. As regards the
evidence in support of the theory that instinct and growth are