mountains. A sudd- covered or sedgy-banked river or lake shore or sedgy swamp they do not frequent. They are most numerous along the water's edge; they become scarcer and scarcer as one advances inland, and (except in the case of certain species) during the rainy season disappear entirely within a few miles of the water. The places they occupy are sharply defined and, as a rule, permanently established. These places or stations are called "fly belts," and the natives know the limits of these belts precisely. The fly belts vary greatly in disposition and extent. Not infrequently they occur on one side of a stream but not on the other. These fly patches are usually confined to strips of jungle, to banana grounds coming down to the water's edge, or to areas of mosani or mimosa forest. In short, the essential condition of a tse-tse station are—— the presence of open water, a wooded district, and a loose soil. As a rule, the fly patches are in sandy ridges or where there are overhanging or jungle-shaded banks.
The limitation of the tse-tse to definite tracts or " belts " has given rise to much speculation. The prevalent opinion is that the fly waits near water to feed on the animals that come to drink. Austen ascribes it to a characteristic social tendency which is exhibited very frequently amongst Diptera. Sambon suggests that it may be related to some food habit ——possibly to a connection with air-breathing fish, of which there are several genera with numerous species in the rivers and lakes of Africa; the fly either feeding directly on the fish or on some mammal or bird which feeds on these fish. Such an association with air-breathing fish might explain the peculiar patchy distribution of certain species of tse-tse fly, their limitation to the neighbourhood of water, and the sandy and thickly wooded nature of their haunts. In the dry season air-breathing fish are obliged to bury themselves in the mud or to excavate burrows from which they come out towards evening in quest of food; they must therefore necessarily congregate in such places as offer conditions suitable to these habits.*[1] It is very important to ascertain the exact reason for the singular topical limitation of the fly, for it may be that through knowledge of this the prophylaxis of fly-transmitted disease could get its opportunity.
Reproduction.—— The Glossinœ do not lay eggs as do the majority of the Diptera, but, as in the case of forest-flies (Hippoboscidœ), the eggs hatch, and the larvæ feed, develop, and moult within the body of the parent, so that when extruded they have practically reached the pupa stage. In fact, the extruded larva becomes almost immediately a pupa, the larva skin becoming a dark, rigid puparium. When extruded, the fully grown larva is nearly as large as the abdomen of the
- ↑ * It is interesting to note in connection with this suggestion that trypanosomes of the mammalian type have been found by Dutton, Todd, and Toby in Clarias angolensis at Leopoldville, and by Montel in another air-breathing fish of undetermined species.
