The objective of this study was to investigate the risk of leukemia from magnetic field exposure among Finnish adults living close to high-voltage power lines. The cohort study included 383700 Finnish people having lived in 1970-1989 within 500 m of overhead power lines of 110-400 kV in a magnetic field calculated to be > or = 0.01 microT. The nested case-cohort study was conducted to investigate dose-response and time-related factors in further detail. Data collection was based on several subsequent record linkages of nationwide registers. The subjects were followed for cancer in 1974-1990, providing over 2.5 million person-years. The outcome measures included standardized incidence ratios, incidence rate ratios, and odds ratios for several exposure indices. The total number of leukemia cases was 203. Magnetic fields were not associated with the overall occurrence of leukemia among adults. The risk estimates were adjusted for age, gender, and municipality; the other covariates had no effect on the risk estimates. However, an almost fivefold increase with statistical significance was observed for the risk of chronic lymphatic leukemia in relation to earlier, or long-lasting, exposure to magnetic fields of > or = 0.1 microT. This finding was based on very small numbers. No risk increases were observed for other types of leukemia. While the possibility of an increase in risk at higher magnetic field levels, or in relation to earlier exposures, cannot be excluded on the basis of this study, the results suggest that typical magnetic fields of high-voltage power lines are not an important cause of leukemia in adults.