This mixed construction tells us that the rustic stacker of wooden beams excels, when occasion offers, in making elegant shell pavements and that it practices rough carpentry and delicate mosaic work indifferently. In the latter instance, the scabbard is made, above all, of Planorbes, selected among the smaller of these pond snails and laid flat. Without being scrupulously regular, the work, at its best, does not lack merit. The pretty, close-whorled spirals, placed one against the other on the same level, have a very pleasing general effect. No pilgrim returning from Santiago de Compostella ever slung handsomer tippet from his shoulders. But only too often the caddis worm dashes ahead, regardless of proportion. The big is joined to the small, the exaggerated suddenly stands out, to the great detriment of order. Side by side with tiny Planorbes, each at most the size of a lentil, others are fixed as large as one's fingernail; and these cannot possibly be fitted in correctly. They overlap the regular parts and spoil their finish. To crown the disorder, the caddis worm adds to the flat spirals any dead shell that comes handy, without distinction of species, provided it be not excessively large. I notice, in its collection of bric-a-brac, the Physa, the Paludina, the Limnaea, the Amber snail [all pond snails] and even the Pisidium [a bivalve], that little twin-valved casket. Land shells, swept into the ditches by the rains after the inmate's death, are accepted quite as readily. In the work made of the Mollusk's cast-off clothing, I find encrusted the spindle shell of the Clausilium, the key shell of the pupa, the spiral of the smaller Helix, the yawning volute of the Vitrina, or glass snail, the turret shell of the Bulimus [all land snails], denizens all of the fields. In short, the caddis worm builds with more or less everything that comes from the plant or the dead mollusk. Among the diversified refuse of the pond, the only materials rejected are those of a gravelly nature. Stone and pebble are excluded from the building with a care that is very rarely absent. This is a question of hydrostatics to which we will return presently. For the moment, let us try to follow the construction of the scabbard.