th new
requirements. It was evidently the first time in insect memory in
which so surprising a phenomenon had been seen as a victim at the last
moment again taking the field. We cannot make instinct intervene here.
If the _Sphex's_ acts are so automatic as we are sometimes led to
believe, in accordance with facts which are perfectly accurate, we
ought always to observe the following succession of acts: first,
hollowing of the burrow; second, the chase; third, the blows of the
dart; fourth, the different manoeuvres for placing the victim in the
sarcophagus. Now in the present case the insect had accomplished the
first three series of actions, and had even begun the fourth; it ought
next to drag the cricket into the burrow without listening to the
recriminations which the latter had no business to make, since it was
to be regarded as having received the two routine doses of poison. But
the _Sphex_ sees its victim come to life, understands this fact, and
without seeking to fathom the cause judges that a new struggle and new
blows of the sting are necessary; he understands that it is necessary
to begin afresh, since the usual result has not been attained. He is
then capable of reflection, and the series of acts which he
accomplishes are not ordained with such inflexibility that it is
impossible for him to modify them in order to conform them to varying
circumstances.

The _Sphex occitanica_ acts in the same manner as its relative in this
complicated art of laying up provisions for the family. The
differences are only in detail. Instead of hollowing the burrow first
and then setting out on the chase to fill it, it does not devote
itself to the labour of digging until a successful expedition has
already assured the victim. (Fig. 17.) Instead of attacking crickets
it seeks a larger orthopterous insect, the _Ephippigera_. The struggle
is no doubt more difficult, but the result is proportionately greater,
and the pursuit does not need to be so often renewed; a single captive
is sufficient for its larva.[73]

[73] For some remarks on the action of the _Sphex_, and for
Darwin's opinion on the matter, see Romanes' _Mental
Evolution in Animals_, pp. 299-303.

[Illustration: FIG. 17.]

_The sureness of instinct._--It is not doubtful that a sure inherited
instinct conducts the _Sphex_ to prick its victim in the situation of
the nervous ganglia, which will be wounded in the act. It may be said
that the lesion r