parts played might be reversed! The butcher of yesterday is the victim of
to-day.

I replace the Pompilus by a Hive-bee. The interview is not protracted. Two
hours later, the Bee is dead, bitten by the Spider. A Drone-fly suffers the
same fate. The Segestria, however, does not touch either of the two
corpses, any more than she touched the corpse of the Pompilus. In these
murders the captive seems to have no other object than to rid herself of a
turbulent neighbour. When appetite awakes, perhaps the victims will be
turned to account. They were not; and the fault was mine. I placed in the
jar a Bumble-bee of average size. A day later the Spider was dead; the rude
sharer of her captivity had done the deed.

Let us say no more of these unequal duels in the glass prison and complete
the story of the Pompilus whom we left at the foot of the wall with the
paralysed Segestria. She abandons her prey on the ground and returns to the
wall. She visits the Spider's funnels one by one, walking on them as freely
as on the stones; she inspects the silken tubes, dipping her antennae into
them, sounding and exploring them; she enters without the least hesitation.
Whence does she now derive the temerity thus to enter the Segestria's
haunts? But a little while ago, she was displaying extreme caution; at this
moment, she seems heedless of danger. The fact is that there is no danger
really. The Wasp is inspecting uninhabited houses. When she dives down a
silken tunnel, she very well knows that there is no one in, for, had the
Segestria been there, she would by this time have appeared on the
threshold. The fact that the householder does not show herself at the first
vibration of the neighbouring threads is a certain proof that the tube is
vacant; and the Pompilus enters in full security. I would recommend future
observers not to take the present investigations for hunting-tactics. I
have already remarked and I repeat: the Pompilus never enters the silken