a of
quitting its scabbard with all speed, its tactics are certainly most
appropriate; but, in that case, an exceptional condition becomes
obligatory, namely, the capacity for recommencing the work. This most
unusual gift of recommencing it possesses in a high measure. I am ready
to see its origin in the persecutions of the Dytiscus and other pirates.
Necessity is the mother of industry.

Certain caddis worms, of the Sericostoma and Leptocerus species, clothe
themselves in grains of sand and do not leave the bed of the stream. On
a clear bottom, swept by the current, they walk about from one bank of
verdure to the other and do not think of coming to the surface to float
and sail in the sunlight. The collectors of sticks and shells are
more highly privileged. They can remain on the level of the water
indefinitely, with no other support than their skiff, can rest in
unsubmersible flotillas and can even shift their place by working the
rudder.

To what do they owe this privilege? Are we to look upon the bundle of
sticks as a sort of raft whose density is less than that of the water?
Can the shells, which are always empty and able to contain a few bubbles
of air in their spiral, he floats? Can the big joists, which break in so
ugly a fashion the none too great regularity of the work, serve to buoy
up the over-heavy raft? In short, is the caddis worm versed in the
laws of equilibrium and does it choose its pieces, now lighter and now
heavier as the case may be, so as to constitute a whole that is
capable of floating? The following facts are a refutation of any such
hydrostatic calculations in the animal.

I remove a number of caddis worms from their sheaths and submit these,
as they are, to the test of water. Whether formed wholly of fibrous
remnants or of mixed materials, not one of them floats. The scabbards
made of shells go to the bottom with the swiftness of a bit of gravel;
the others sink gently. I experiment with the separate materials one
by one. No shell remains on the surface, not even among the Planorbes,
which a many-whorled spiral ought, one would think, to keep afloat.
The fibrous remnants must be divided into two categories. The first,
darkened by time and soaked with moisture, sink to the bottom. These are
the most plentiful. The second, considerably fewer in number, of more
recent date and less saturated with water, float very well. The general
result is immersion, as in the case of the intact scabbards.