ultimately to carry them, but before turning round to return, to put
the insects in a circular box with an axle which could be made to
revolve very rapidly first in one direction and then in another, so as
to destroy for a time all sense of direction in the insects. I have
sometimes imagined that animals may feel in which direction they were
at the first start carried.'

This method of experimenting seemed to me very ingeniously conceived.
Before going west, I walk eastwards. In the darkness of their paper
bags, the mere fact that I am moving them gives my prisoners a sense
of the direction in which I am taking them. If nothing happened to
disturb this first impression, the insect would be guided by it in
returning. This would explain the homing of my Mason-bees carried to a
distance of two or three miles amid strange surroundings. But, when
the insects have been sufficiently impressed by their conveyance to
the east, there comes the rapid twirl, first this way round, then
that. Bewildered by all these revolutions first in one direction and
then in another, the insect does not know that I have turned round and
remains under its original impression. I am now taking it to the west,
when it believes itself to be still travelling towards the east. Under
the influence of this impression; the insect is bound to lose its
bearings. When set free, it will fly in the opposite direction to its
home, which it will never find again.

This result seemed to me the more probable inasmuch as the statements
of the country-folk around me were all of a nature to confirm my
hopes. Favier (The author's gardener and factotum. Cf. "The Life of
the Fly": chapter 4.--Translator's Note.), the very man for this sort
of information, was the first to put me on the track. He told me that,
when people want to move a Cat from one farm to another at some