Recent research has suggested that nonionizing radiation in the form of power-frequency magnetic fields may play some role in carcinogenesis in general and in acute nonlymphocytic leukemia in particular. Much of the epidemiologic evidence is preliminary in nature and the methods of previous studies have been criticized. In order to further evaluate this hypothesis, a population-based case-control study of adult acute nonlymphocytic leukemia and residential exposure to power-frequency magnetic fields was carried out in western Washington state. Analyses were based on 114 cases who were newly diagnosed from 1981 to 1984 and identified from a population-based cancer registry, and 133 controls who were chosen from the study area by random digit dialing. Magnetic field exposure was estimated from external electrical wiring configurations within 140 ft (42.7 m) of each subject's residence. In addition, magnetic fields were measured inside the subject's residence at the time of interview. Neither the directly measured magnetic fields nor the surrogate values based on the wiring configurations were associated with acute nonlymphocytic leukemia.