A strange result indeed! Honey, the nectar of the flowers, the sole diet of the Bee-tribe in both its forms and the sole resource of the Wasp in her a adult form, is to the larvae of the latter an object of insurmountable repugnance and probably a toxic dish. Even the transformation of the nymphosis surprises me less than this inversion of the appetite. What happens in the insect's stomach to make the adult seek passionately what the youngster refused lest it should die? This is not a question of organic debility unable to endure a too substantial, too hard, too highly spiced dish. The grub that gnaws the Cetonia-larva, that generous piece of butcher's meat; the glutton that crunches its batch of tough Locusts; the one that battens on nitrobenzine-flavoured game: they certainly own unfastidious gullets and accommodating stomachs. And these robust eaters allow themselves to die of hunger or digestive troubles because of a drop of syrup, the lightest food imaginable, suited to the weakness of extreme youth and a feast for the adult besides! What a gulf of obscurity in the stomach of a wretched grub! These gastronomical researches called for a counterexperiment. The carnivorous larva is killed by honey. Conversely, is the mellivorous larva killed by animal food? Reservations are needful here, as in the previous tests. We should be courting a flat refusal if we offered a pinch of Locusts to the larvae of the Anthophora or the Osmia, for instance. (For both these Wild Bees cf. "Bramble-bees and Others": passim.--Translator's Note.) The honey-fed insect would not bite into it. There would be no use whatever in trying. We must find the equivalent of the jam-sandwich aforesaid; in other words, we must give the larva its natural fare with a mixture of animal food. The addition made by my artifices shall be albumen, as found in the egg of the Hen, albumen the isomer of fibrin, which is the essential factor in any form of prey. On the other hand, the Three-horned Osmia lends herself most admirably to my plans, because of her dry honey, consisting for the greater part of floury pollen. I therefore knead this honey with albumen, graduating the dose until its weight largely exceeds that of the flour. In this way I obtain pastes of different degrees of consistency, but all firm enough to bear the larva without danger of immersion. With too fluid a mixture there would be a risk of death by drowning. Lastly I install a moderately-developed larva on each of my albuminous cakes.