ted by a curvilinear triangle with blunted
corners and inwardly convex sides. This is the appearance displayed by
the pseudochrysalis during the winter and spring.

But in June it has lost this withered appearance; it represents a
perfect balloon, an ellipsoid of which the sections perpendicular to
the major axis are circles. Something has also come to pass of greater
importance than this expansion, which may be compared with that which
we obtain by blowing into a wrinkled bladder. The horny integuments of
the pseudochrysalis have become detached from their contents, all of a
piece, without a break, just as happened the year before with the skin
of the secondary larva; and they thus form a fresh vesicular envelope,
free from any adhesion to the contents and itself enclosed in the
pouch formed of the secondary larva's skin. Of these two bags without
outlet, one of which is enclosed within the other, the outer is
transparent, flexible, colourless and extremely delicate; the second
is brittle, almost as delicate as the first, but much less translucent
because of its yellow colouring, which makes it resemble a thin flake
of amber. On this second sac are found the stigmatic warts, the
thoracic studs and so forth, which we noted on the pseudochrysalis.
Lastly, within its cavity we catch a glimpse of something the shape of
which at once recalls to mind the secondary larva.

And indeed, if we tear the double envelope which protects this
mystery, we recognize, not without astonishment, that we have before
our eyes a new larva similar to the secondary. After one of the
strangest transformations, the creature has gone back to its second
form. To describe the new larva is unnecessary, for it differs from
the former in only a few slight details. In both there is the same
head, with its various appendages barely outlined; the same vestiges
of legs, the same stumps transparent as crystal. The tertiary larva
differs from the secondary only by its abdomen, which is less fat,
owing to the absolute emptiness of the digestive apparatus; by a
double chain of fleshy cushions extending along each side; by the rim
of the stigmata, crystalline and slightly projecting, but less so than
in the pseudochrysalis; by the ninth pair of breathing-holes, hitherto
rudimentary but now almost as large as the rest; lastly by the
mandibles ending in a very sharp point. Evicted from its twofold
sheath, the tertiary larva makes only very lazy movements of
co