no less
impossible successors! Though the snow-ball, slowly rolling, at
last becomes an enormous sphere, it is still necessary that the
starting-point shall not have been NIL. The big ball implies the little
ball, as small as you please. Now, in harking back to the origin of
these acquired habits, if I interrogate the possibilities I obtain zero
as the only answer. If the animal does not know its trade thoroughly,
if it has to acquire something, all the more if it has to acquire
everything, it perishes: that is inevitable; without the little
snow-ball the big snow-ball cannot be rolled. If it has nothing to
acquire, if it knows all that it needs to know, it flourishes and
leaves descendants behind it. But then it possesses innate instinct, the
instinct which learns nothing and forgets nothing, the instinct which is
steadfast throughout time.

The building up of theories has never appealed to me: I suspect them one
and all. To argue nebulously upon dubious premises likes me no better. I
observe, I experiment and I let the facts speak for themselves. We have
just heard these facts. Let each now decide for himself whether instinct
is an innate faculty or an acquired habit.

CHAPTER 4. THE CETONIA-LARVA.

The Scolia's feeding-period lasts, on the average, for a dozen days or
so. By then the victuals are no more than a crumpled bag, a skin emptied
of the last scrap of nutriment. A little earlier, the russet-yellow tint
announces the extinction of the last spark of life in the creature that
is being devoured. The empty skin is pushed back to make space; the
dining-room, a shapeless cavity with crumbling walls, is tidied up a
little; and the Scolia-grub sets to work on its cocoon without further
delay.

The first courses form a general scaffolding, which finds a support
here and there on the earthen walls, and consist of a rough, blood-red
fabric. When the larva is merely laid, as required by my investigations,
in a hollow made with the finger-tip in the bed of mould, it is not able
to spin its cocoon, for want of a ceiling to which to fasten the upper
threads of its network. To weave its cocoon, every spinning larva
is compelled to isolate itself in a hammock slung in an open-work
enclosure, which enables it to distribute its thread uniformly in all
directions. If there be no ceiling, the upper part of the cocoon cannot
be fashioned, because the worker lacks the necessary points of support.
Under these conditions my