Who will explain to me this predilection for the Orthopteron in a tribe whose chief, the Oil-beetle, accepts nothing but the mess of honey? Why do insects which appear close together in all our classifications possess such opposite tastes? If they spring from a common stock, how did the consumption of flesh supplant the consumption of honey? How did the Lamb become a Wolf? This is the great problem which was once set us, in an inverse form, by the Spotted Sapyga, a honey-eating relative of the flesh-eating Scolia.5 I submit the question to whom it may concern.
5 The essays on these will appear in the volume, entitled The Hunting Wasps, aforementioned.—Translator's Note.
The following year, at the beginning of June, some of my pseudochrysalids split open transversely behind the head and lengthwise down the whole of the median line of the back, except the last two or three segments. From it emerges the tertiary larva, which, from a simple examination with the pocket-lens, appears to me, in its general features, identical with the secondary larva, the one which eats the Tachytes' provisions. It is naked and pale-yellow, the colour of butter. It is active and wriggles with awkward movements. Ordinarily it lies upon its side, but it can also stand in the normal position. The creature is then trying to use its legs, without finding sufficient purchase to enable it to walk. A few days later, it relapses into complete repose.
Thirteen segments, including the head, which is large, with a quadrilateral cranium, rounded at the sides. Short antennæ, consisting of three knotted joints. Powerful curved mandibles, with two or three little teeth at the end, of a fairly bright red. Labial palpi rather bulky, short and with three joints, like the antennæ. The mouth-parts, labrum, mandibles and palpi are movable and stir slightly, as though seeking food. A small brown speck near the base of each antenna, marking the place of the future eyes. Prothorax wider than the segments that come after it. These are all of one width and are distinctly divided by a furrow and a slight lateral rim. Legs short, transparent, without a terminal claw. They are three-jointed stumps. Pale stigmata, eight pairs of them, placed as in the pseudochrysalis, that is, the first and largest pair on the line dividing the first two segments of the thorax and the seven others on the first seven abdominal segments. The secondary larva and the pseudochrysalis also have a very small stigma on the penultimate segment of the abdomen. This stigma has disappeared in the tertiary larva; at least I cannot detect it with the aid of a good magnifying-glass.
Lastly, we find the same strong mandibles as in the secondary larva, the same feeble legs, the same appearance of a Weevil-grub. The movements return, but are less clearly marked than in the primary form. The passage through the pseudochrysalid state has led to no change that is really worth describing. The creature, after this singular phase, is what it was before. The Meloes and Sitares, for that matter, behave similarly.
Then what can be the meaning of this pseudochrysalid stage, which, when passed, leads precisely to the point of departure? The Meloid seems to be revolving in a circle: it undoes what it has just done, it draws back after advancing. The idea sometimes occurs to me to look upon the pseudochrysalis as a sort of egg of a superior organization, starting from which the insect follows the ordinary law of entomological phases and passes through the successive stages of larva, nymph and perfect insect. The first hatching, that of the normal egg, makes the Meloid go through the larval dimorphism of the Anthrax and the Leucospis. The primary larva finds its way to the victuals; the secondary larva consumes them. The second hatching, that of the pseudochrysalis, reverts to the usual course, so that the insect passes through the three customary forms: larva, nymph, adult.
The tertiary larval stage is of brief duration, lasting about a fortnight. The larva then sheds its skin by a longitudinal rent along the back, as did the secondary larva, uncovering the nymph, in which we recognize the Beetle, the genus and species being almost determinable by the antennæ.
The second year's development turned out badly. The few nymphs which I obtained about the middle of June shrivelled up without attaining the perfect form. Some pseudochrysalids remained on my hands without showing any sign of approaching transformation. I attributed this delay to lack of warmth. I was in fact keeping them in the shade, on a what-not, in my study, whereas under natural conditions they are exposed to the hottest sun, beneath a layer of sand a few inches deep. To imitate these conditions without burying my charges, whose progress I wished to follow comfortably, I placed the pseudochrysalids that remained on a layer of fresh sand at the bottom of a glass receiver. Direct exposure to the sun was impracticable: it would have been fatal at a period when life is subterranean. To avoid it, I tied over the mouth of the receiver a few thicknesses of black cloth, to represent the natural screen of sand; and the apparatus thus prepared was exposed for some weeks to the most brilliant sunshine in my window. Under the cloth cover, which, owing to its colour, favours the absorption of heat, the temperature, during the day-time, became that of an oven; and yet the pseudochrysalids persisted in remaining stationary. The end of July was near and nothing indicated a speedy hatching. Convinced that my attempts at heating would be fruitless, I replaced the pseudochrysalids in the shade, on the shelves, in glass tubes. Here they passed a second year, still in the same condition.
June returned once more and with it the appearance of the tertiary larva, followed by the nymph. For the second time this stage of development was not exceeded; the one and only nymph that I succeeded in obtaining shrivelled, like those of the year before. Will these two failures, arising no doubt from the overdry atmosphere of my receivers, conceal from us the genus and the species of the Mantis-eating Meloid? Fortunately, no. The riddle is easily solved by deduction and comparison.
The only Melodiæ in my part of the country which, though their habits are still unknown, might correspond in size with either the larva or the pseudochrysalis in question are the Twelve-pointed Mylabris and Schaeffer's Cerocoma. I find the first in July on the flowers of the sea scabious; I find the second at the end of May and in June on the heads of the Îles d'Hyères everlasting. This last date is best-suited to explain the presence of the parasitic larva and its pseudochrysalis in the Tachytes' burrows from July onwards. Moreover, the Cerocoma is very abundant in the neighbourhood of the sand-heaps haunted by the Tachytes, while the Mylabris does not occur there. Nor is this all: the few nymphs obtained have curious antennæ, ending in a full, irregular tuft, the like of which is found only in the antennæ of the male Cerocoma. The Mylabris, therefore, must be eliminated; the antennæ, in the nymph, must be regularly jointed, as they are in the perfect insect. There remains the Cerocoma.