after a lapse of time which varies considerably in the different
entomological orders.

This explains the long hesitation of the Tarantula, so wearisome to the
experimenter when he presents to her, at the entrance to the burrow, a
rich, but dangerous prey. The majority refuse to fling themselves upon
the Carpenter-bee. The fact is that a quarry of this kind cannot be
seized recklessly: the huntress who missed her stroke by biting at random
would do so at the risk of her life. The nape of the neck alone
possesses the desired vulnerability. The adversary must be nipped there
and no elsewhere. Not to floor her at once would mean to irritate her
and make her more dangerous than ever. The Spider is well aware of this.
In the safe shelter of her threshold, therefore, prepared to beat a quick
retreat if necessary, she watches for the favourable moment; she waits
for the big Bee to face her, when the neck is easily grabbed. If this
condition of success offer, she leaps out and acts; if not, weary of the
violent evolutions of the quarry, she retires indoors. And that, no
doubt, is why it took me two sittings of four hours apiece to witness
three assassinations.

Formerly, instructed by the paralysing Wasps, I had myself tried to
produce paralysis by injecting a drop of ammonia into the thorax of those
insects, such as Weevils, Buprestes, {13} and Dung-beetles, whose compact
nervous system assists this physiological operation. I showed myself a
ready pupil to my masters' teaching and used to paralyze a Buprestis or a
Weevil almost as well as a Cerceris {14} could have done. Why should I
not to-day imitate that expert butcher, the Tarantula? With the point of
a fine needle, I inject a tiny drop of ammonia at the base of the skull
of a Carpenter-bee or a Grasshopper. The insect succumbs then and there,
without any other movement than wild convulsions. When attacked by the