Those larvae must be rejected which are nourished upon one single corpulent insect, as is that of the Scolia.
The grub attacks its prey at a determined point, plunges its head and neck into the body of the insect, skilfully divides the entrails in order to keep the remains fresh until its meal is ended, and does not emerge from the opening until all is consumed but the empty skin.
To interrupt the larva with the object of smearing the interior of its prey with honey is doubly objectionable; I might extinguish the lingering vitality which keeps putrefaction at bay in the victim, and I might confuse the delicate art of the larva, which might not be able to recover the lode at which it was working or to distinguish between those parts which are lawfully and properly eaten and those which must not be consumed until a later period.

As I have shown in a previous volume, the grub of the Scolia has taught me much in this respect.

The only larvae acceptable for this experiment are those which are fed on a number of small insects, which are attacked without any special art, dismembered at random, and quickly consumed.

Among such larvae I have experimented with those provided by chance--those of various Bembeces, fed on Diptera; those of the Palaris, whose diet consists of a large variety of Hymenoptera; those of the Tachytus, provided with young crickets; those of the Odynerus, fed upon larvae of the Chrysomela; those of the sand-dwelling Cerceris, endowed with a hecatomb of weevils.

As will be seen, both consumers and consumed offer plenty of variety.