Let us recapitulate the strange metamorphoses which I have sketched. Every Beetle-larva, before attaining the nymphal stage, undergoes a greater or smaller number of moults, of changes of skin; but these moults, which are intended to favour the development of the larva by ridding it of covering that has become too tight for it, in no way alter its external shape. After any moult that it may have undergone, the larva retains the same characteristics. If it begin by being tough, it will not become tender; if it be equipped with legs, it will not be deprived of them later; if it be provided with ocelli, it will not become blind. It is true that the diet of these non-variable larvæ remains the same throughout their duration, as do the conditions under which they are destined to live.

But suppose that this diet varies, that the environment in which they are called upon to live changes, that the circumstances accompanying their development are liable to great changes: it then becomes evident that the moult may and even must adapt the organization of the larva to these new conditions of existence. The primary larva of the Sitaris lives on the body of the Anthophora. Its perilous peregrinations demand agility of movement, long-sighted eyes and masterly balancing-appliances; it has, in fact, a slender shape, ocelli, legs and special organs adapted to averting a fall. Once inside the Bee's cell, it has to destroy the egg; its sharp mandibles, curved into hooks, will fulfil this office. This done, there is a change of diet: after the Anthophora's egg the larva proceeds to consume the ration of honey. The environment in which it has to live also changes: instead of balancing itself on a hair of the Anthophora, it has now to float on a sticky fluid; instead of living in broad daylight, it has to remain plunged in the profoundest darkness. Its sharp mandibles must therefore become hollowed into a spoon that they may scoop up the honey; its legs, its cirri, its balancing-appliances must disappear as useless and even harmful, since all these organs can only involve the larva in serious danger, by causing it to stick in the honey; its slender shape, its horny integuments, its ocelli, being no longer necessary in a dark cell where movement is impossible, where there are no rough encounters to be feared, may likewise give place to complete blindness, to soft integuments, to a heavy, slothful form. This transfiguration, which everything shows to be indispensable to the life of the larva, is effected by a simple moult.

We do not so plainly perceive the necessity of the subsequent forms, which are so abnormal that nothing like them is known in all the rest of the insect class. The larva which is fed on honey first adopts a false chrysalid appearance and afterwards goes back to its earlier form, though the necessity for these transformations escapes us entirely. Here I am obliged to record the facts and to leave the task of interpreting them to the future. The larva of the Meloidæ, therefore, undergo four moults before attaining the nymphal state; and after each moult their characteristics alter most profoundly. During all these external changes, the internal organization remains unchangingly the same; and it is only at the moment of the nymph's appearance that the nervous system becomes concentrated and that the reproductive organs are developed, absolutely as in the other Beetles.

Thus, to the ordinary metamorphoses which make a Beetle pass successively through the stages of larva, nymph and perfect insect, the Meloidæ add others which repeatedly transform the larva's exterior, without introducing any modification of its viscera. This mode of development, which preludes the customary entomological forms by the multiple transfigurations of the larva, certainly deserves a special name: I suggest that of hypermetamorphosis.

Let us now recapitulate the more prominent facts of this essay.

The Sitares, the Meloes, the Zonites and apparently other Meloidæ, possibly all of them, are in their earliest infancy parasites of the harvesting Bees.

The larva of the Meloidæ, before reaching the nymphal state, passes through four forms, which I call the primary larva, the secondary larva, the pseudochrysalis and the tertiary larva. The passage from one of these forms to the next is effected by a simple moult, without any changes in the viscera.

The primary larva is leathery and settles on the Bee's body. Its object is to get itself carried into a cell filled with honey. On reaching the cell, it devours the Bee's egg; and its part is played.

The secondary larva is soft and differs completely from the primary larva in its external characteristics. It feeds upon the honey contained in the usurped cell.

The pseudochrysalis is a body deprived of all movement and clad in horny integuments which may be compared with those of the pupæ and chrysalids. On these integuments we see a cephalic mask without distinct or movable parts, six tubercles indicating the legs and nine pairs of breathing-holes. In the Sitares the pseudochrysalis is enclosed in a sort of sealed pouch and in the Zonites in a tight-fitting bag formed of the skin of the secondary larva. In the Meloes it is simply half-sheathed in the split skin of the secondary larva.