As soon as captured the victim receives the poisonous sting of the hunter. With the pompilids, as with the other paralyzing Hymenoptera, the employment of the sting, according to Fabre, cannot be accomplished with half-knowledge ; it should be perfect from the start, for an incomplete inertia would be dangerous for the egg or the young larva fixed upon the victim, and death following the sting would not be less so. The victim must remain living but inert for as long a period as the larval life.

Among other examples in support of this statement we may cite the observations of Fabre upon a spider-hunter, the buffoon Pompilus ( Calicurgus scurra). After many struggles "the Epeira lies turned over on its back. The Pompilus is on top. . . . With its legs it controls the legs of the spider; with its mandibles it holds the cephalo-thorax. . . . The dart of the wasp, directed from behind forward, is plunged into the mouth of the spider with minute precautions and with great persistence. Almost instantly the venomous fangs are rendered inert and the formidable prey is defenseless. The abdomen of the wasp then extends its arc and the sting is plunged in behind the fourth pair of legs, on the median line near the juncture of the abdomen and the cephalo-thorax. At this point the skin is finer and more penetrable than elsewhere. . . . The nervous centers controlling the movement of the legs are situated a little higher than the point of the wound, but the direction of the sting, from behind forward, allows it to reach them. From this last sting the eight legs are at once paralyzed." The spider is from that moment inert. Its lethargy must last six or seven weeks, and then only "comes true death and its accompanying decomposition." Less time than this is necessary for the larva of the pompilid to devour its victim.

Many pompilids employ this method of the Cali-curgus, but certain others depart from it, as the observations of Ferton have shown. While the banded Pompilus contents itself with stinging the mouth of its Lycosa, the eight-spotted Pompilus renews the puncture under the thorax of its already inert victim, and when we take from Pompilus pulcher the lycosid which it carries, paralyzed, to its nest, it does not regain possession of its victim without stinging it again. One would say that it has the idea that the work is incomplete and that it is necessary to finish it.

Is it in obedience to a similar fear that certain pompilids nibble off or break off the legs of their paralyzed victims t This removal of the legs has been observed by Ferton; and by the Peckhams with numerous species (P. vagans, P. scelestus, etc.)y and with these the complete or partial removal of the legs results. This last phenomenon was noticed in 1839 by Colonel Goureau. It is common with Pseudagenia, as Ferton and Eabaud have shown (1909). Ferton noticed it also with Salius hisdecoratus, and the Peckhams with Pompilus fuscipennis and Agenia bombgcina. Having observed that the palpi and fangs are always respected, and that the removal of the legs is accomplished at the moment when the prey is going to be put into the burrow, the Peckhams supposed that this habit is for the purpose of rendering easier the placing in the burrow ; but I think, with Ferton, that it acts only to render the inertia more complete, for Pompilus vagans lodges its victim in the hole which the latter has dug. Eabaud argues against this interpretation, which presupposes with the Pompilus a rather extraordinary degree of intellect, and he brings forward in support the fact that the victim is rendered inert before the mutilation, but he forgets that the inertia produced by the sting is often fugitive. In any case, if the manœuver of the Pompilus has not for its end a perfect inertia., it surely accomplishes this result. Is this result really necessary? Fabre is convinced of it; but observation shows that, aside from the buffoon Calicurgus and some other pompilids, the inertia is more or less fugitive and incomplete. Ferton tells of an epeirid (Meta segmentata) captured by the hyaline Pompilus (Salius hyalinatus) that remained inert several days, then revived little by little, and completely recovered on the twenty-fifth day. A wandering Pompilus had just stung the large Nemesia badia. The same biologist observes:

At the end of some minutes the paralysis ceased, and almost to the end of the feeding of the larva the Nemesia remained living. ... I have found several half eaten by the Pompilus which were capable of running in the box in which they were enclosed.

In other instances the victims of the same Pompilus have not been completely paralyzed. Of eleven Epeira strix noticed by the Peckhams with Pompilus quinquenotatus, three had been immediately killed, two lived four days, and five others from five to forty days. It is then right to think, contrary to Fabre's opinion, that the inertia of the victim is not .a necessary condition to the perfect development of the larvae of the pompilids, that the females of this group of insects often show a very imperfect knowledge in the art of killing, and that in the course of time they have undergone an evolution in this art.

How does this evolution proceed! The primitive manceuvers, without doubt, were much like those which Ferton observed with a pompilid closely allied to 8alius opacus, where the stings "were given by chance in all parts of the body, from the mouth to the extremity of the abdomen. ' ' These multiple stings exhausted the reservoir of venom without always paralyzing the victim. The author adds :

One thinks, then, that it is of advantage to a species to acquire more precision in the use of the sting, since the dangers which it runs in the struggle will be diminished in this way. It would not be astonishing if time and natural selection had brought certain species to the efficiency of Calicurgus scurra.

Between these two extremes, other pompilids show all the intermediary stages, and as to the nibbling off or amputation of the legs, it is necessary to regard this as a return to the ancestral method, a return which has as its effect, if not for its object, to bring about a more perfect inertia.

But, since the more or less rapid revival of the spiders does not have an injurious influence on the development of the larvae of the pompilids, one can understand how certain species have been able to abandon some of these practices. The mining Pompilus (P. effocliens) seems to have arrived at the stage where this practice is on the point of disappearing. In Corsica, Ferton has seen it strike the Nemesias with two thrusts of the sting, like Calicurgus, but in Algeria, where the same species hunts lycosids, the egg is fixed without a sting upon the spider, which, remains perfectly active. In the course of his. studies Ferton caught one of these pompilids just as it was coming out of a burrow. He exhumed the lyco-sid, which carried its side the egg of the wasp, and left it near the entrance of the burrow.

When touched by the wasp, which returned to finish its work, it fled rapidly, but the Pompilus had seen it, and pursued it and threw itself upon it. I distinctly saw the abdomen of the wasp turn around the spider as if to sting it. The spider stopped, on being assaulted by the wasp, and remained motionless, its legs folded, while the wasp carried and dropped upon it some bits of earth. A few minutes afterward the spider jumped up as lively as before, and ran to hide itself in the grass.

It was captured, put in the post-office, crossed the Mediterranean, and came to Bordeaux, where Professor Perez received it alive with the egg attached. '1 It had not been stung, " concluded Ferton, "and I willingly give the explanation of atavism to the act of the wasp which curved its abdomen under the body of the spider.' 1

We are thus led to pompilids which seem to have lost completely the habit of stinging. In 1872 the zoologist Karsch found a very active lycosid (Tarantula inquilino) which carried an egg on the right side of the abdomen. He watched it and soon saw a legless larva coming from the egg and fixing itself at that point. Placed in a glass filled with earth, the spider dug a cylindrical hole, covered the orifice with silk, and stayed there. Soon there remained only fragments of the legs of the spider, and the larva transformed to a nymph from which four weeks later a Salius fuscus emerged triumphantly. In 1870 Lichtenstein did the same thing with Salius affinis, and -spiders carrying an egg of this kind are not rare. Although the pompilid has not been observed at the moment of egg-laying, it seems true that certain species of the family lay their eggs without stinging their prey.

So the art of paralyzing, with the pompilids, will have begun with punctures made by chance, and will have continued with these Hymenoptera up to the wise use of two stings, but, paralysis not being necessary for the development of their larvae, this habit has disappeared, and reappears only as an act of racial memory (an atavism).

The possibility of this evolution and the differences in the inertia of the victims is explained by the fact that the stings can produce a permanent injury by destroying the nerve centers or a transitory action by means of the secreted poison. The observations of the Peckhams upon a lycosid stung by Pompilus scelestus throws a light on these two kinds of phenomena. Tjhe victim seemed dead at first, then revived somewhat, but always with a noticeable lethargy, but it could not see, and when flies were given to it it was necessary for it to touch them; the cervical ganglia had without doubt been seriously injured. The Peckhams kept this spider for three months. It ran about its cage, but until the end remained perfectly blind.