Let us return to the purple varnish which covers the interior of the cocoon. At first, I thought it should be attributed to the silk glands, which, after serving to spin the double wrapper of silk and the scaffolding, must finally have secreted it. To convince myself, I opened larvæ which had just completed their task of weaving, and had not yet begun to lay on the lacquer. At that period I found no trace of violet fluid in their glands. It is only seen in the digestive canal, which is swelled with a purple pulp, and later in the stercorous load sent down to the lower end of the cocoon. Elsewhere all is white, or faintly tinged with yellow. I am far from suggesting that the larva plasters its cocoon with excrement, yet I am convinced that this wash is produced by the digestive organs, and I suspect—though I cannot positively assert it, having several times missed the moment to ascertain it—that the larva disgorges and applies with its mouth the quintessence of the purple pulp in its stomach to make the wash of lacquer. Only after this last piece of work would it eject the remains of digestion in a single mass, and thus is explained the disgusting necessity of storing the excrement within the larva’s habitation. At all events the usefulness of this layer is clear; its absolute impermeability protects the larva from the damp which would certainly penetrate the poor shelter hollowed for it by its mother. Recollect that it is buried but a few inches deep in sandy, open ground. To judge how far cocoons thus varnished are capable of resisting damp, I have plunged them in water for several days, yet never found any trace of moisture within them. Let us compare the Sphex cocoon, with manifold coverings to protect the larva in a burrow itself unprotected, with that of Cerceris tuberculata, sheltered by a layer of sandstone, more than half a yard down in the ground. This cocoon has the form of a very long pear, with the small end cut off. It is composed of a single silken wrap, so fine that the larva is seen through it. In my many entomological researches I have always found the labour of larva and mother supplement each other. In a deep well-sheltered dwelling the cocoon is of light materials; for a surface abode, exposed to wind and weather, it is strongly constructed.