The authors studied 435 men who died from brain cancer in three areas of the USA. These men were matched with controls who died of other causes. Information on occupation and other relevant factors was obtained from interviews with next-of-kin. Occupational exposures were assessed in two ways - by comparison with a known list of exposed occupations, and by individual assessment by an industrial hygienist. There was an excess risk (odds ratio 2.3) of brain tumor death in men ever employed in an electrical or electronic occupation, but not in those exposed to RF radiation who did not work in an electrical job (odds ratio 1.0). It appeared that the increased association in electrical workers was likely to be due to an exposure other than RF radiation.