interlaced threads which fix it to the walls
of the cell. I have compared this bag, because of its shape, with a
fishtrap. Without leaving this hammock, stretching its neck through the
orifice, the worker gathers from without a little heap of sand, which
it stores inside its workshop. Then, selecting the grains one by one,
it encrusts them all around itself in the fabric of the bag and cements
them with the fluid from its spinnerets, which hardens at once. When
this task is finished, the house has still to be closed, for it has been
wide open all this time to permit of the renewal of the store of sand
as the heap inside becomes exhausted. For this purpose a cap of silk is
woven across the opening and finally encrusted with the materials which
the larva has retained at its disposal.

The Tachytes builds in quite another fashion, although its work, once
finished, does not differ from that of the Bembex. The larva surrounds
itself, to begin with, about the middle of its body with a silken girdle
which a number of threads, very irregularly distributed, hold in place
and connect with the walls of the cell. Sand is collected, within reach
of the worker, on this general scaffolding. Then begins the work of
minor masonry, with grains of sand for rubble and the secretion of the
spinnerets for cement. The first course is laid upon the fore-edge of
the suspensory ring. When the circle is completed, a second course of
grains of sand, stuck together by the fluid silk, is raised upon the
hardened edge of what has just been done. Thus the work proceeds,
by ring-shaped courses, laid edge to edge, until the cocoon, having
acquired half of its proper length, is rounded into a cap and finally is
closed. The building-methods of the Tachytes-larva remind me of a mason
constructing a round chimney, a narrow tower of which he occupies the
centre. Turning on his own axis and using the materials placed to his
hand, he encloses himself little by little in his sheath of masonry.
In the same way the worker encloses itself in its mosaic. To build the
second half of the cocoon, the larva turns round and builds in the same
way on the other edge of the original ring. In about thirty-six hours
the solid shell is completed.

I am rather interested to see the Bembex and the Tachytes, two workers
in the same guild, employ such different methods to achieve the same
result. The first begins by weaving an eel-trap of pure silk and next
encrusts the grains o