I have more than a supposition in another case, equally important. Do the consumers of Orthoptera ever vary their diet? Should their favourite game fail, can they do with another? Does S. occitanica think that except a fat ephippiger, there is nothing in the wide world worth eating. Does S. albisecta admit nothing but crickets to her table, and S. flavipennis only grasshoppers? Or according to time, place, and circumstance, does each replace the favourite food by some equivalent? It would be of the highest importance to bring forward such facts if they exist, as they would tell us whether the suggestions of instinct are absolute and immutable, or if they vary, and within what limits. It is true that in the cell of a Cerceris are buried most varied species of Buprestids or of the Weevil group, which shows that she has a great latitude of choice; but such an extension of hunting ground cannot be supposed for the Sphex, which I have found so faithful to one exclusive prey, invariable for each species, and which, moreover, finds among the Orthoptera kinds of very different shapes. I have, however, had the good fortune to meet with one case—only one—of complete change in the larva’s food, and I mention it the more willingly in the archives of the Sphegidæ because such facts, scrupulously observed, will one day be corner-stones for him who may desire to build the psychology of instinct on solid foundations. This is my fact. The scene is on a jetty by the Rhône. On one side is the great river, with its thunder of waters, on the other, a dense thicket of osiers, willows, and reeds, and between the two a narrow path with a bed of fine sand. A yellow-winged Sphex appears, hopping and dragging its prey along. What do I see! It is no grasshopper, but a common Acridian! And yet the Hymenopteron really is the Sphex so well known to me (S. flavipennis), the energetic huntress of grasshoppers! I can hardly believe my eyes. The burrow is not far off; she enters and stores her booty. I seat myself, determined to await a new expedition—wait hours if need be to see if so extraordinary a capture is repeated. Seated there I occupy the whole width of the path. Two simple conscripts come up, new-clipped, with that incomparable, automaton-look conferred by the first days of barrack life. They are chattering together—no doubt talking of their homes and the girls they left behind them; each is peeling a willow switch with a knife. A fear seizes me; ah! it is not easy to try an experiment on the public way, where, when some fact watched for during long years does present itself, a passer-by may disturb or annihilate chances which may never occur again! I rise anxiously to make way for the conscripts; I withdraw into the osier bed, and leave the narrow way free. To do more was not prudent; to say, “My good fellows, do not go that way,” would have made bad worse. They would have supposed some snare hidden in the