Every moment a Bembex entered like a sudden flash, and darted up to the silken ceiling, which resounded with a dull thud. A tumult went on aloft, in which one could not distinguish attacker from attacked, so lively was the mêlée. The struggle was very brief; almost at once the Hymenopteron retired with a captive between its feet. The dull band of gadflies drew a little back all round on this sudden irruption, which decimated them, but without leaving the treacherous shelter. It was so hot outside; wherefore move? Plainly, such swift attack and prompt departure with the prey does not allow the Bembex to use a poignard according to rules. The sting no doubt fulfils its office, but is directed with no precision towards such spots as are exposed by the chances of the combat. To slay outright the half-murdered gadfly, still struggling between the feet of its assassin, I have seen the Bembex chew the head and thorax of her victim. This habit, peculiar to the Bembecids, shows that the Bembex desires death, not paralysis, since she ends the life of the Diptera with so little ceremony. Everything considered, I think that on the one side the nature of the prey, so quickly dried up, and on the other, the difficulties of so vehement an attack, are the reasons why the Bembecids serve up dead prey to their larvæ, and consequently provide it daily. Let us follow the Hymenopteron when it returns with its captive closely clasped to the burrow. Here is one—B. tarsata—coming loaded with a Bombylius. The nest is placed at the sandy foot of a vertical slope, and the approach of the Bembex is announced by a sharp humming, somewhat plaintive, and only ceasing when the insect has alighted. One sees her hover above the bank, then descend, following the vertical line slowly and cautiously, still emitting the sharp hum. If her keen gaze should discover anything unusual, she delays her descent, hovers a moment, ascends again, redescends, then flies away, swift as an arrow. In a few moments she returns. Hovering at a certain height she appears to inspect the locality, as if from the top of an observatory. The vertical descent is resumed with most circumspect deliberation; finally, she alights without hesitation at a spot which to my eye has nothing to distinguish it from the rest of the sandy surface. The plaintive note ceases at once. She must have alighted somewhat by chance, since the most practised eye could not distinguish one spot from another on the sandy tract. She will have dropped down somewhere near her hole, whose entrance she will now seek, marked since her last exit not only by the natural falling in of materials, but by her scrupulous sweeping. No! she does not hesitate in the least—does not feel about—does not seek. All have agreed that the organs fitted to direct insects in their researches reside in the antennæ. At the moment of returning to the nest I see nothing special in their play. Without