Let us return to the violet varnish that lines the inside of the cocoon. I thought at first that I must attribute it to the silk-glands, which, after giving a glossy coat to the double wrapper of silk and the scaffolding, have still a secret store of the fluid. To convince myself, I opened some larvæ which had just finished their work as weavers and had not yet begun to apply their lacquer. At that period I saw no trace of violet fluid in the silk-glands. This shade is found only in the digestive canal, which bulges with a purple-coloured pulp; we find it also, but later, in the stercoral plug relegated to the lower end of the cocoon. With this exception, everything is white, or faintly tinged with yellow. Far be it from me to suggest that the larva plasters its cocoon with its excreta; and yet I am convinced that this plaster is a product of the digestive organs and I suspect, though I cannot say for certain—having been clumsy enough several times to miss a favourable opportunity of making sure—that the larva disgorges and applies with its mouth the quintessence of the purple pulp from its stomach in order to form the shellac glaze. Only after this last performance would it reject its digestive residuum in a single lump; and this would explain the unpleasant necessity in which the larva finds itself of making room for its excreta inside its actual habitation. Be this as it may, there can be no doubt about the usefulness of the coating of shellac; its complete impermeability must protect the larva against the damp which would certainly attack it in the precarious refuge dug for it by the mother. Remember that the larva is buried only a few inches down in uncovered, sandy ground. To judge to what extent the cocoons thus varnished are able to resist the damp, I kept some steeped in water for several days on end, without afterwards finding a trace of moisture inside them. Compare the Sphex’ cocoon, with its manifold linings, which are so well adapted for the protection of the larva in an unprotected burrow, with the cocoon of the Great Cerceris, lying under the dry shelter of a slab of sandstone and at a distance of eighteen or twenty inches underground: this cocoon has the shape of a very long pear, with the narrow end lopped off. It consists of a single silken wrapper, so thin and fine that the larva shows through it. In my numerous entomological investigations I have always seen the larva’s industry and the mother’s thus making good each other’s deficiencies. In a deep, well-sheltered abode, the cocoon is of a light material; in a surface dwelling, exposed to the inclemencies of the weather, the cocoon is stoutly built.