slightly larger species of Diodontus which acted in every way like americanus. In spite of all our efforts we were unable to find the holes of this second and larger species. They were present in large numbers and it seemed improbable that they could have nested in territory so familiar to us as the garden came to be without being discovered. After close watching we concluded that these wasps flew over the fence into the woods while americanus settled down close by.

The month of July is evidently the working season of the Diodonti since they were very active from the seventh of the month, when we first saw them, until the first of August. From that time on we saw them less and less, although a few of them were still at work three weeks later.

On July twenty-fifth we first saw the males of this species, which, Mr. Ashmead tells us, had been unknown before. On this and on succeeding days we saw them mating with the females. As the females had been laying their eggs for about three weeks the males had probably been present all the time, but had escaped our attention.

The parasitic Chrysis fly, Omalus corruscans, is always in attendance upon americanus, both on the cherry bushes and on the ground near the nests. Brilliant and beautiful but full of evil intentions, she watches their comings and goings. There is no opportunity for her to lay her egg on the aphis outside of the nest. Holding it closely under her body and not exposing any part to attack, the wasp, without making the least pause, flies into her open doorway. When she comes out again the enemy is still lurking near, but no instinct warns her to cover her treasure. The door stands wide open as she takes her departure, leaving her young exposed to the foe. Perhaps the danger is not so great as it seems. The fly certainly penetrates into the nest, since we have found it in the gallery when excavating; and in one case we found a strange larva feeding on the aphides along with that of americanus; but it may be that the supply of food is so ample as to cover the needs of both. At

any rate it must be confessed that the parasite has not prevented americanus from becoming a flourishing species.

Although as a rule, Diodontus worked for the coming generation, the captured aphis sometimes served not as food for the young but as a dainty morsel for herself. In these cases there was no malaxation, the aphis being held in any position while it was sucked dry of all its juices and then thrown away. This may be a further development of the habit described by Belt of stroking the frog-hoppers to get the drop of honey which they can yield without harm to themselves. It has little in common with the method of Philanthus apivorus, which, according to Fabre, squeezes the honey from its bee because if left, it would prove fatal to the young larva. It is rather like that of Odynerus nidulator which takes nothing from the caterpillars which are destined to feed the young, never storing up those from which it has sucked the juices.

Cerceris ornata offers us still another habit. The neck of Halictus, its prey, is brutally compressed, the skin being broken so that the juices of the body exude, and these juices are licked off by the wasp.* The object of the malaxation, however, seems to be to produce lethargy for the benefit of the wasp larva, since the bee is afterward stored up. The taste that the wasp gets of bee-juice is rather an accident than anything else. How is it that the honey is not fatal to the young of Cerceris, as it is to that of Philanthus?

As has been said, the excavation of the nest of Diodontus is a difficult matter, but in six cases we succeeded in finding the pocket with its contents. In these nests the number of aphides varied from five to forty, the provisioning being only just begun in some cases while in others it had been completed, and the nest closed up.

With a single exception the aphides in these nests were dead. They were usually green when first taken out but turned yellow

*Etude sur l' Instinct du Cerceris ornata, Archives de Zoologie Expérimentale et Générale, Deuxième Série, Tome V., 1887, p. 27.

by the second or third day and by the fourth or fifth were all dry and brown. The exception to the rule was a very lively little aphis found in a closed nest with about forty dead ones. We have no doubt that they are often alive when first taken in, from the fact that while watching the action of the wasps in the bottle we noticed that they not infrequently left the aphides almost uninjured.

In four of the nests we found the egg, which in one instance was on the under side of the body of the aphis, while in the other cases it was placed on the dorsum. Perhaps one of the eggs was laid by a parasite and this was the reason of the difference in position.

Only two of the eggs hatched. The first of these (No. 23) was taken from a nest which had been closed up at three in the afternoon on July twenty-second. At nine in the morning of July twenty-fourth we found that it had hatched, and it seemed to be a few hours old. The egg stage, then, probably lasts about thirty-six hours, although we judge that it may in some instances be less from the case of another wasp, No. 24. This nest was also closed up by the wasp on July twenty-second, and we dug it up on the afternoon of the same day. The larva, however, hatched at nine in the morning of July twenty-third, nearly a day sooner than the one in nest No. 23. This, to be sure, may have been due, not to a more rapid development of the egg, but to its having been laid earlier, perhaps shortly after the provisioning of the nest was begun, while in the other case it may have been laid after the aphides had all been carried in. We know that in certain other wasps (Pelopaeus) there is just such a variation as this in the point of time at which the egg is laid.

These two wasplings lived through the six days of their larval life. They ate the whole aphis, leaving no débris. At the end of that time one of them died and the other spun its cocoon. In the tube with this later larva we discovered, on the seventh day after the nest was taken up, a second smaller larva which was also eating the aphides. This one, which was probably the

young of O. corruscans, disappeared two days later without having exerted any evil influence upon the destiny of the rightful owner of the nest.

Diodontus gracilis and corniger are said to provide aphides for their young, making their nests in holes in posts, while minutus and tristis, like americanus, burrow in the ground. Americanus seems to be the only American species of which anything is known.

CHAPTER XI.

SOME GRAVE DIGGERS.

Cerceris and Philanthus.

Plates I., figs. 3 and 8; VIII., fig. 6; IX., fig. 1; XI., fig. 2.

Dufour, in describing the fearful ravages of Cerceris ornata among the bees, says that the wasps of this genus are among other insects what eagles and hawks are among birds. While this characterization does not seem to fit the American species it is certainly true that the genus stands out as one of those in which the distinctive peculiarities are strongly marked. They might be considered the aristocrats in the world of wasps, their habits of reposeful meditation and their calm, unhurried ways being far removed from the nervous manners of the Pompilidae or the noisy, tumultuous life of Bember. Their intelligence is shown by their reluctance to betray their nests, and by their uneasiness at any slight change in the objects that surround them. It is not necessary to attempt to catch them, or to make threatening gestures in order to arouse their sense of danger. If you are sitting quietly by a nest when the wasp opens her door in the morning she will notice you at once and will probably drop out of sight as though she resented your intrusion into her privacy. After a little she will come up again and will learn to tolerate you, but at the least movement on your part, almost at the winking of an eyelid, she will disappear.

Our three representatives of this genus all prey upon beetles that are injurious to vegetation, and therefore deserve the gratitude of the agriculturalist. They are from one-half to threequarters of an inch in length, clypeata and deserta being banded with bright yellow, while in nigrescens the bands are much paler, being gray with a faint tinge of yellow.