larming accuracy.

If we accept the hypothesis of a special nerve-centre for the mandibles,
the difficulty would be a little less, without detracting from the
operator's talent. The sting would then have to reach a barely visible
speck, an atom in which we should hardly find room for the point of a
needle. This is the difficulty which the various paralysers solve in
ordinary practice. Do they actually wound with their dirks the ganglion
whose influence is to be done away with? It is possible, but I have
tried no test to make sure, the infinitely tiny wound appearing to be
too difficult to detect with the optical instruments at my disposal. Do
they confine themselves to lodging their drop of poison on the ganglion,
or at all events in its immediate neighbourhood? I do not say no.

I declare moreover, that, to provoke lightning paralysis, the poison, if
it is not deposited inside the mass of nervous substance, must act
from somewhere very near. This assertion is merely echoing what the
Two-banded Scolia has just shown us: her Cetonia-grub, stung less than
a millimetre from the regular spot, did not become motionless until next
day. There is no doubt, judging by this instance, that the effect of the
virus spreads in all directions within a radius of some extent; but
this diffusion is not enough for the operator, who requires for her egg,
which is soon to be laid, absolute safety from the very first.

On the other hand, the actions of the paralysers argue a precise search
for the ganglia, at all events for the first thoracic ganglion, the
most important of all. The Hairy Ammophila, among others, affords us an
excellent example of this method. Her three thrusts in the caterpillar's
thorax and especially the last, between the first and second pair of
legs, are more prolonged than the stabs distributed among the abdominal
ganglia. Everything justifies us in believing that, for these decisive
inoculations, the sting seeks out the corresponding ganglion and acts
only when it finds it under its point. On the abdomen this peculiar
insistence ceases; the sting passes swiftly from one segment to another.
For these segments, which are less dangerous, the Ammophila perhaps
relies on the diffusion of her venom; in any case, the injections,
though hastily administered, do not diverge from a close vicinity of the
ganglia, for their field of action is very limited, as is proved by the
number of inoculations necessary to induce complete t