From June 1982 to May 1986 in a small village in Kwale, Kenya, we studied seasonal fluctuations in populations of Bulinus globosus, prevalence of Schistosoma haematobium infection in this snail, and effects of chemotherapy and piped water supply on infection rate of snails. In the perennially-flowing Pemba River, relatively small numbers of snails were collected; they were found only during the hot dry season (December to March). In a tributary stream, the Kadingo River, whose flow ceased at the end of both the cool and hot dry seasons, snail numbers peaked at the end of the cool dry season (October to November) and at the beginning of the hot dry season (January). Large numbers of infected snails were found in the Kadingo River from November to January (short rainy season and beginning of dry season). Selective mass chemotherapy with metrifonate and provision of piped water were begun in February and March 1984. These control measures achieved a significant reduction in the infection rate of snails (P < 0.001); the annual infection rate for the 2 years before treatment was 9.3% and 13.1%, and for the 2 years after treatment was 3.5% and 3.4%.