tudinally; more or less obtuse at one end, but pointed at the
other, and sometimes ending in a short curved tail.

In all cases the upper face is convex and regular. In it we can
distinguish three well-marked and longitudinal zones. The middle zone,
which is narrower than the others, is composed of thin plates arranged
in couples, and overlapping like the tiles of a roof. The edges of these
plates are free, leaving two parallel series of fissures by which the
young can issue when the eggs are hatched. In a nest recently abandoned
this zone is covered with fine cast-off skins which shiver at the least
breath, and soon disappear when exposed to the open air. I will call
this zone the zone of issue, as it is only along this bell that the
young can escape, being set free by those that have preceded them.

In all other directions the cradle of this numerous family presents an
unbroken wall. The two lateral zones, which occupy the greater part of
the demi-ellipsoid, have a perfect continuity of surface. The little
Mantes, which are very feeble when first hatched, could not possibly
make their way through the tenacious substance of the walls. On the
interior of these walls are a number of fine transverse furrows, signs
of the various layers in which the mass of eggs is disposed.

Let us cut the nest in half transversely. We shall then see that the
mass of eggs constitutes an elongated core, of very firm consistency,
surrounded as to the bottom and sides by a thick porous rind, like
solidified foam. Above the eggs are the curved plates, which are set
very closely and have little freedom; their edges constituting the zone
of issue, where they form a double series of small overlapping scales.

The eggs are set in a yellowish medium of horny appearance. They are
arranged in layers, in lines forming arcs of a circle, with the cephalic
extremities converging towards the zone of issue. This orientation tells
us of the method of delivery. The newly-born larvae will slip into the
interval between two adjacent flaps or leaves, which form a prolongation
of the core; they will then find a narrow passage, none too easy to
effect, but sufficient, having regard to the curious provision which we
shall deal with directly; they will then reach the zone of issue. There,
under the overlapping scales, two passages of exit open for each layer
of eggs. Half the larvae will issue by the right-hand passage, half by
that on the left hand. This proces