The objective of our study was to investigate the possible interactive effect of occupational exposure to extremely low-frequency magnetic field (ELFMF) and to known or suspected carcinogenic chemicals on the incidence of the two main histological types of brain cancer, gliomas and meningiomas, in a cohort of male Swedish workers. The historical cohort of all Swedish men gainfully employed in 1970 were followed 19 years (1971-1989). Exposure to ELFMF and to nine chemicals were assessed using two Swedish job exposure matrices based on occupational codes and industrial activity. Relative risks adjusted for age, period, geographical area, and town size were computed using log-linear Poisson models. The main finding was the absence of ELFMF effect on glioma risk in the absence of a simultaneous exposure to chemical products. The effect of petroleum products was independent of the intensity of ELFMF exposure whereas solvents, lead, and pesticides/herbicides were only associated with glioma in workers also exposed to moderate or high levels of ELFMF. On the other hand, whereas ELFMF seemed to enhance the effect of specific chemicals in the causation of gliomas, we did not find a relationship between ELFMF exposure and meningiomas. The potential for ELFMF to act as an effect modifier of the association of chemical agents and glioma is an interesting new finding. It would be worthwhile to evaluate this hypothesis for other tumors. Also, it is necessary to confirm these results in epidemiological studies with individual exposure assessments, and in experimental studies that may elucidate whether there is a true causal mechanism for the results we observed.