(4) Bembex Julii (a new species). The egg hatches either on a Sphærophoria or a Pollenia floralis, and the provender is a mixture of Syrphus corollæ, Echinomyia rubescens, Gonia atra, Pollenia floralis, P. ruficollis, Clytia pellucens, Lucilia Cæsar, Dexia rustica, Bombylius. (5) Bembex rostrata. This is above all a captor of gadflies. It lays its egg on a Syrphus corollæ, or a Lucilia Cæsar, but then only brings to the larva large game belonging to the various kinds of Tabanus. (6) Bembex bidentata. Another ardent hunter of gadflies. I have never seen it with other game, and do not know on what the egg is laid. This variety of provisions shows that the Bembecids have no exclusive tastes, and attack one and all of the species of Diptera which are offered by the chances of the chase. They seem, however, to have some favourites—one species especially choosing Bombylius, another Stomoxys, and a third and fourth, Gadflies. XVII HUNTING DIPTERA After this bill of fare for Bembecids in the larva state, we must seek the motive which causes these Hymenoptera to adopt a mode of storage exceptional among Fossors. Why, instead of laying up sufficient food and dropping an egg on it, which would allow the cell to be closed at once without need of returning, does the Hymenopteron oblige itself to come and go perpetually for a fortnight from the fields to the burrow and back again, toiling every time through the sand to issue forth and hunt, or bring back prey? The explanation is that the food must be fresh—an all-important matter, for the grub absolutely refuses game which is at all high, with a hint of decay; like the larvæ of all Fossors, it must have fresh provisions—always fresh provisions. We have seen in the case of the Cerceris, Sphex, and Ammophila how the mother resolves the feeding problem, by placing beforehand in the cell a sufficient quantity of game, and also that of keeping it for weeks perfectly fresh—nay, almost alive, though motionless—in order to secure the safety of the grub which feeds on the prey. This marvel is brought about by the most skilful means known to physiology. The poisoned sting is sent into the nerve centres once or oftener, according to the construction of the nervous system, and the victim retains all which we call life, except power of motion. Let us see if the Bembex practises this deep science of murder. Diptera taken from between the feet of their captor as the latter enters the burrow mostly seem quite dead. They are motionless; only in rare cases are there some slight convulsions of the tarsi—the last vestiges of life soon to be extinct. The same appearance of complete death is found, as a rule, in insects not really killed but paralysed by the skilful stab of a Cerceris or a Sphex. The question as to life or death can, therefore, only be decided by the manner in which the victims keep fresh.