A few were brought for me to see. Their little chambers showed no efflorescence of rime, with which all the surrounding earth was coated. The waterproof varnish had been wonderfully efficacious. As for the anchorites, roused from their torpor by the warmth of the room, they began to wander about my bed, where I followed them vaguely with my fading eyes. May came, as eagerly awaited by the sick man as by the Halicti. I left Orange for Serignan, my last stage, I expect. While I was moving, the Bees resumed their building. I gave them a regretful glance, for I had still much to learn in their company. I have never since met with such a mighty colony. These old observations on the habits of the Cylindrical Halictus may now be followed by a general summary which will incorporate the recent data supplied by the Zebra Halictus and the Early Halictus. The females of the Cylindrical Halictus whom I unearth from November onwards are evidently fecundated, as is proved by the assiduity of the males during the preceding two months and most positively confirmed by the couples discovered in the course of my excavations. These females spend the winter in their cells, as do many of the early-hatching melliferous insects, such as Anthophorae and Mason-bees, who build their nests in the spring, the larvae reaching the perfect state in the summer and yet remaining shut up in their cells until the following May. But there is this great difference in the case of the Cylindrical Halictus, that in the autumn the females leave their cells for a time to receive the males under ground. The couples pair and the males perish. Left alone, the females return to their cells, where they spend the inclement season. The Zebra Halicti, studied first at Orange and then, under better conditions, at Serignan, in my own enclosure, have not these subterranean customs: they celebrate their weddings amid the joys of the light, the sun and the flowers. I see the first males appear in the middle of September, on the centauries. Generally there are several of them courting the same bride. Now one, then another, they swoop upon her suddenly, clasp her, leave her, seize hold of her again. Fierce brawls decide who shall possess her. One is accepted and the others decamp. With a swift and angular flight, they go from flower to flower, without alighting. They hover on the wing, looking about them, more intent on pairing than on eating. The Early Halictus did not supply me with any definite information, partly through my own fault, partly through the difficulty of excavation in a stony soil, which calls for the pick-axe rather than the spade. I suspect her of having the nuptial customs of the Cylindrical Halictus.