The effectiveness of permethrin-impregnated (0.5 g/m2) bed-nets and curtains as malaria control measures was evaluated in Uriri, Kenya in 1988. One hundred five families were randomly assigned to 1 of 3 study groups (control, bed-net, or curtain). All participants were cured of parasitemia with pyrimethamine/sulfadoxine. Selective epidemiologic and entomologic parameters were measured weekly, while knowledge, attitude, and practices surveys were conducted at the beginning and end of the 15 week study. Plasmodium falciparum infections per person week at risk were significantly higher in the control group than in either the curtain group (5.42 vs. 2.35 cases/100 person weeks risk) or the bed-net group (5.42 vs. 3.77 cases/100 person weeks risk). The curtain group had fewer infections per person week at risk than the bed-net group (2.35 vs. 3.77 cases/100 person weeks risk). A difference was found in clinical malaria among the groups: 45% of persons in the bed-net and curtain groups vs. 30% of those in the control group reported no episodes of fever and chills (χ2, P < 0.05). Indoor resting Anopheles gambiae or An. funestus were found on 94 occasions in the control houses, but only twice in the treated houses during weekly visits to each house over the study period (χ2P < 0.001). The pyrethrum knockdown method produced similar results with a total of 195, 23, and 3 An. gambiae and An. funestus collected in the control, bed-net, and curtain houses during the same period, respectively. Nets and curtains were readily accepted, although 30% of the bed-net group were not using their nets when visited late in the evening. Widespread community use of the bed-nets and curtains is considered possible to achieve.