The impact of use of bed-nets treated with alphacypermethrin, at 20 mg (ai)/m2, in comparison to untreated nets or no nets on malaria vectors and malaria incidence was studied in tribal villages of Malkangiri district, Orissa state, India, which are highly endemic for falciparum malaria. Treated or untreated nets were supplied to the villagers in June 1999 and the nets were re-treated in September 1999, just before the rise in vector abundance and malaria incidence. The seasonal pattern of indoor resting Anopheles fluviatilis females was similar in all the three groups of villages before the start of intervention and the indoor resting catches were not significantly different between treatment-groups (two-way ANOVA, F = 1.53; d.f. = (2, 78); P = 0.2). During intervention, the indoor resting catches differed significantly among treatment groups (two-way ANOVA, F = 38.9; d.f.= (2, 66); P < 0.005). There was a 99% reduction in the indoor resting catches of An. fluviatilis in villages with treated nets and 61% reduction in villages with untreated nets compared with no-net villages. Comparison between villages with and without treated nets showed that there was 97% reduction in indoor resting catches in villages with treated nets. Pair-wise comparison showed that the reductions between villages with and without nets as well as between villages with treated and untreated nets were significant (Dunnett's C-test, P < 0.05). The indoor resting catches of Anopheles culicifacies did not differ significantly among the three groups of villages either before (F = 0.99; d.f. = (2, 121); P = 0.4) or during intervention (F = 0.21; d.f. = (2, 66); P = 0.8). Bioassay with 3 min exposure to treated bed nets showed 100% mortality of An. culicifacies for 2 months and with An. fluviatilis for 4.5 months after which tests were not carried out. In villages with treated nets, the Annual Parasite Incidence (API) significantly declined (P < 0.05) by about 65.7% and prevalence of infection among children (< 15 years) declined by 57.1%, whereas in villages with untreated nets, there was only 34% reduction in API and 13% in the prevalence of infection. In villages with treated nets, there was 48% reduction in API and 64% in prevalence of infection compared with villages with untreated nets. The impact of use of treated bed nets on other arthropod pests lasted for at least 1.5 months. After 1.5 months, observations on arthropod pests were not continued. The use rate of treated nets varied from 49.8 to 93.7% in three seasons and about 68.3% of treated bed nets and 60% of untreated nets were in good condition 1 year after distribution. Out of 489 users of treated net, five people complained of a burning sensation on the face (skin irritation) for 5 days following the distribution of treated nets. There were no other complaints of any discomfort in using the treated nets. The use of alphacypermethrin treated bed nets at 20 mg (ai)/m2 can be one of the options for reducing the vector abundance and incidence of malaria in this area.