A double-blind controlled trial was undertaken from August 1990 to February 1991 among Karen children on the Thai-Burmese border to evaluate the effects on malaria incidence and prevalence of permethrin-treated bed nets. Three hundred and fifty schoolchildren, aged 4 to 15 years, were allocated at random to receive either a permethrin-impregnated net or a non-treated net. The incidence of malaria infections, confirmed by a blood film, was assessed during 6 months. Three surveys were conducted, on admission and 3 and 6 months later, to measure the prevalence of infections and spleen rates. Compliance was assessed by monthly home visiting. The use of permethrin-treated bed nets reduced the number of parasitaemic Plasmodium falciparum infections by 38% and the number of symptomatic episodes by 42%. The number of P. vivax malaria attacks was similar in each group. The prevalence of positive blood films in the 2 groups did not change significantly during the study. A reduction in spleen rate by 50% in both groups at the end of the study period could not be related to the overall use of nets. Compliance was high and no side-effect was reported. The long-term effects on morbidity and mortality need to be assessed after distribution of permethrin treated bed nets at the village level.