out
indiscriminately by the right or left exit. As the number of cells to be
traversed is the same on both sides, her expenditure of energy does not
vary with the direction of the exit; and the principle of least action
is still observed.

It was important to discover if the Three-pronged Osmia shared her
capacity, in the first place, with the other bramble-dwellers and, in
the second, with Bees differently housed, but also destined laboriously
to cut a new road for themselves when the hour comes to quit the nest.
Well, apart from a few irregularities, due either to cocoons whose
larva perished in my tubes before developing, or to those inexperienced
workers, the males, the result was the same in the case of Anthidium
scapulare. The insects divided themselves into two equal batches, one
going to the right, the other to the left. Tripoxylon figulus left
me undecided. This feeble insect is not capable of perforating my
partitions; it nibbles at them a little; and I had to judge the
direction from the marks of its mandibles. These marks, which are not
always very plain, do not yet allow me to pronounce an opinion. Solenius
vagus, who is a skilful borer, behaved differently from the Osmia. In a
column of ten, the whole exodus was made in one direction.

On the other hand, I tested the Mason-bee of the Sheds, who, when
emerging under natural conditions, has only to pierce her cement ceiling
and is not confronted with a series of cells. Though a stranger to the
environment which I created for her, she gave me a most positive answer.
Of a column of ten laid in a horizontal tube open at both ends, five
made their way to the right and five to the left. Dioxys cincta, a
parasite in the buildings of both species of Mason-bees, the Chalicodoma
of the Sheds and the Chalicodoma of the Walls (Cf. "The Mason-bees"
by J. Henri Fabre, translated by Alexander Teixeira de Mattos:
passim.--Translator's Note.), provided me with no precise result.
The Leaf-cutting Bee (Megachile apicalis, SPIN. (Cf. Chapter 8 of the
present volume.--Translator's Note.)), who builds her leafy cups in the
old cells of the Chalicodoma of the Walls, acts like the Solenius and
directs her whole column towards the same outlet.

Incomplete as it is, this symmetry shows us how unwise it were to
generalize from the conclusions to which the Three-pronged Osmia leads
us. Whereas some Bees, such as the Anthidium and the Chalicodoma, share
the Osmia's talent for using