is striking. The Osmiae, which mature
earlier, emerge; and the Solenius-cocoons, as well as their inhabitants,
which by this time have reached the perfect stage, are reduced to
shreds, to dust, wherein it is impossible for me to recognize a vestige,
save perhaps here and there a head, of the exterminated unfortunates.
The Osmia, therefore, has not respected the live cocoons of a foreign
species: she has passed out over the bodies of the intervening Solenii.
Did I say passed over their bodies? She has passed through them,
crunched the laggards between her jaws, treated them as cavalierly as
she treats my disks. And yet those barricades were alive. No matter:
when her hour came, the Osmia went ahead, destroying everything upon
her road. Here, at any rate, is a law on which we can rely: the supreme
indifference of the animal to all that does not form part of itself and
its race.

And what of the sense of smell, distinguishing the dead from the living?
Here, all are alive; and the Bee pierces her way as through a row of
corpses. If I am told that the smell of the Solenii may differ from that
of the Osmiae, I shall reply that such extreme subtlety in the insect's
olfactory apparatus seems to me a rather far-fetched supposition. Then
what is my explanation of the two facts? The explanation? I have none
to give! I am quite content to know that I do not know, which at least
spares me many vain lucubrations. And so I do not know how the Osmia,
in the dense darkness of her tunnel, distinguishes between a live cocoon
and a dead cocoon of the same species; and I know just as little how
she succeeds in recognizing a strange cocoon. Ah, how clearly this
confession of ignorance proves that I am behind the times! I am
deliberately missing a glorious opportunity of stringing big words
together and arriving at nothing.

The bramble-stump is perpendicular, or nearly so; its opening is at the
top. This is the rule under natural conditions. My artifices are able
to alter that state of things; I can place the tube vertically or
horizontally; I can turn its one orifice either up or down; lastly, I
can leave the channel open at both ends, which will give two outlets.
What will happen under these several conditions? That is what we shall
examine with the Three-pronged Osmia.

The tube is hung perpendicularly, but closed at the top and open at the
bottom; in fact, it represents a bramble-stump turned upside down. To
vary and complicate the e