had not the
patience to see for themselves--the conclusion at which I arrive is
positive: to inject the germs, the Microgaster never attacks the
caterpillars.

The invasion, therefore, is necessarily effected through the
Butterfly's eggs themselves, as experiment will prove. My broad jar
would tell against the inspection of the troop, kept at too great a
distance by the glass enclosure, and I therefore select a tube an inch
wide. I place in this a shred of cabbage-leaf, bearing a slab of eggs,
as laid by the Butterfly. I next introduce the inmates of one of my
spare vessels. A strip of paper smeared with honey accompanies the new
arrivals.

This happens early in July. Soon, the females are there, fussing about,
sometimes to the extent of blackening the whole slab of yellow eggs.
They inspect the treasure, flutter their wings and brush their
hind-legs against each other, a sign of keen satisfaction. They sound
the heap, probe the interstices with their antennae and tap the
individual eggs with their palpi; then, this one here, that one there,
they quickly apply the tip of their abdomen to the egg selected. Each
time, we see a slender, horny prickle darting from the ventral surface,
close to the end. This is the instrument that deposits the germ under
the film of the egg; it is the inoculation-needle. The operation is
performed calmly and methodically, even when several mothers are
working at one and the same time. Where one has been, a second goes,
followed by a third, a fourth and others yet, nor am I able definitely
to see the end of the visits paid to the same egg. Each time, the
needle enters and inserts a germ.

It is impossible, in such a crowd, for the eye to follow the successive
mothers who hasten to lay in each; but there is one quite practicable
method by which we can estimate the number of germs introduced into a
single egg, which is, later, to open the ravaged caterpillars and count
the grubs which they contain. A less repugnant means is to number the
little cocoons heaped up around each dead caterpillar. The total will
tell us how many germs were injected, some by the same mother returning
several times to the egg already treated, others by different mothers.
Well, the number of these cocoons varies greatly. Generally, it
fluctuates in the neighbourhood of twenty, but I have come across as
many as sixty-five; and nothing tells me that this is the extreme
limit. What hideous industry for the exterminati