How have these relations—whatever in their entirety they may be—come about? My own idea is that these beetles, like some other creatures—amongst them the little white wood-louse that lives with our own _Formica flava_—found ants’ nests very comfortable places of retirement, since, by reason of their peculiar weapon of defence, they could defy any attempt to interfere with them, on the part of the ants. The ants, on their side, would soon have given up molesting them, so that, never requiring to defend themselves against the creatures by whom they were surrounded, the intruders got to associate them with quite other ideas, and, having first lost the habit, at length lost the power of turning their artillery in this direction. Meanwhile _Paussus_, owing to its sweet secretion, which, after relations had once become amicable, the ants would soon have discovered, had got to be a very welcome guest, so much so that, even when it took to eating their larvæ, they retained their love for it, as a species, though resenting such conduct upon the individual. And now the once redoubtable invader could be punished with impunity, for the habit of never discharging against an ant had become a fixed, inherited instinct, not to be got rid of even though life were at stake. Thus, as it appears to me, it may have come about that, though armed with dynamite, and carrying bombs, no living _Paussus_ has ever defended itself against an ant, and no living ant, perhaps, ever seen a _Paussus_ discharge its artillery. Of course these are only conjectures, and the last, especially, may be opposed to fact, since it has been suggested that one way in which _Paussus_ may make itself useful within the nest of its hosts, may be by bombarding certain obnoxious parasites, or other would-be invaders. This does not, however, appear to me to be likely, for how could these explosion take place, under such circumstances, without doing damage to the ants themselves? In one’s own house one would hardly wish a bomb to be thrown, even against one’s greatest enemy—at any rate not in the drawing-room. That the ants should, by this means, be able, or, if able, willing to rid themselves of the mites which infest them, as has been conjectured, seems especially unlikely—indeed, hardly possible. On the whole, it seems to me that the relations at present existing between the two insects could only have grown up through _Paussus_ having ceased to discharge, not only at an ant, but even—owing, probably, to there never being any occasion for it—in an ants’ nest. The experimental tickling with a straw was, of course, an artificial stimulus. In spite of its sweet secretion, I cannot see how a beetle with such a power at its command as _Paussus_ has, can have been originally selected by the ants for domestication, but, on the other hand, an armed invader might easily, by coincidence, possess some property which would make it, in time, of use and value to the population on which it forced itself.