My pity for the fallen nestlings had brought me this unhoped-for good luck. Having made the circle of spectators draw back and leave free passage for the Sphex, I took away her prey with my pincers, giving her immediately in exchange one of my ephippigers with an ovipositor like that of the one abstracted. Stamping was the only sign of impatience shown by the bereaved Hymenopteron. She ran at the new prey, too corpulent to try to avoid pursuit, seized it with her mandibles by the saddle-shaped corslet, got astride, and curving her abdomen, passed its end under the ephippiger’s thorax. There doubtless the stings are given, but the difficulty of observation prevents me from telling how many. The ephippiger—gentle victim—lets itself be operated on unresistingly, like the dull sheep of our slaughter-houses. The Sphex takes her time and manœuvres her lancet with a deliberation favourable to the observer; but the prey touches the ground with the whole lower part of its body, and what happens there cannot be seen. As for interfering and lifting the ephippiger a little so as to see better, it is not to be thought of; the murderess would sheath her weapon and retire. The next act is easy to observe. After having stabbed the thorax, the end of the abdomen appears under the neck, which she forces widely open by pressing the nape. Here the sting enters with marked persistence, as if more effective than elsewhere. One might suppose that the nerve centre struck was the lower part of the œsophagean collar, but the persistence of movement in the mouthpieces, mandibles, jaws, and palpi, animated by this source of nerve power, shows that this is not so. Through the neck the Sphex simply reaches the thoracic ganglia, or at least the first, more easily attainable through the thin skin of the neck than through the integuments of the chest. All is over. Without one convulsion or sign of pain the ephippiger is rendered henceforward an inert mass. For the second time I deprived the Sphex of the subject operated on, replacing it by the second female at my disposal. The same manœuvres were followed by the same result. Three times, almost without a pause, the Sphex repeated her skilled surgery, first on her own capture, then on those exchanged by me. Will she do so a fourth time on the male which I still have? It is doubtful, not that she is weary, but because the game does not suit her. I have never seen a Sphex with any but female prey, which, filled as they are with eggs, are the favourite food of the larvæ. My suspicion was well founded. Deprived of her third capture, she obstinately refused the male which I offered her. She ran hither and thither with hurried steps, seeking her lost prey. Three or four times she approached the ephippiger, walked round it, cast a disdainful glance at it, and finally flew away. This was not what her larvæ wanted. Experiment reiterated it after twenty years’ interval.