A community-based randomized, controlled trial of permethrin impregnated bednets was carried out in a rural area of northern Ghana, between July 1993 and June 1995, to assess the impact on the mortality of young children in an area of intense transmission of malaria and no tradition of bednet use. The district around Navrongo was divided into 96 geographical areas and in 48 randomly selected areas households were provided with permethrin impregnated bednets which were re-impregnated every 6 months. A longitudinal demographic surveillance system was used to record births, deaths and migrations, to evaluate compliance and to measure child mortality. The use of permethrin impregnated bednets was associated with 17% reduction in all-cause mortality in children aged 6 months to 4 years (RR = 0.83; 95% CI 0.69-1.00; P = 0.05). The reduction in mortality was confined to children aged 2 years of younger, and was greater in July-December, during the wet season and immediately after (RR = 0.79; 95% CI 0.63-1.00), a period when malaria mortality is likely to be increased, than in the dry season (RR = 0.92, 95% CI 0.73-1.14). The ready acceptance of bednets, the high level of compliance in their use and the subsequent impact on all-cause mortality in this study has important implications for programmes to control malaria in sub-Saharan Africa.