to gain a neighbouring bough and there undergo transformation, I see it
immediately make a prudent retreat, descending to the bottom of its
burrow without the slightest difficulty--a proof that even when about to
be abandoned for ever the refuge is not encumbered with rubbish.

The ascending shaft is not a hurried piece of work, scamped by a
creature impatient to reach the sunlight. It is a true dwelling, in
which the larva may make a long stay. The plastered walls betray as
much. Such precautions would be useless in the case of a simple exit
abandoned as soon as made. We cannot doubt that the burrow is a kind of
meteorological observatory, and that its inhabitant takes note of the
weather without. Buried underground at a depth of twelve or fifteen
inches, the larva, when ripe for escape, could hardly judge whether the
meteorological conditions were favourable. The subterranean climate
varies too little, changes too slowly, and would not afford it the
precise information required for the most important action of its
life--the escape into the sunshine at the time of metamorphosis.

Patiently, for weeks, perhaps for months, it digs, clears, and
strengthens a vertical shaft, leaving only a layer of earth a finger's
breadth in thickness to isolate it from the outer world. At the bottom
it prepares a carefully built recess. This is its refuge, its place of
waiting, where it reposes in peace if its observations decide it to
postpone its final departure. At the least sign of fine weather it
climbs to the top of its burrow, sounds the outer world through the thin
layer of earth which covers the shaft, and informs itself of the
temperature and humidity of the outer air.

If things are not going well--if there are threats of a flood or the
dreaded _bise_--events of mortal gravity when the delicate insect issues