In order to test the possibility that magnetic fields (MF) act as a tumour promoter, a long-term skin carcinogenicity study of 50 Hz sinusoidal MF with flux densities of 50 microT and 0.5 mT was performed in female NMRI mice. 7,12-dimethylbenz[a]anthracene (DMBA) in acetone was applied to the dorsal skin, as an initiator, and exposure to MF was performed for 19 (weekdays) or 21 h/day (weekends and holidays) for 103 weeks starting one week after the initiator treatment. The phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) was used as a positive control for skin tumour promoting activity. MF was also evaluated for complete carcinogenic action in groups of mice that were treated with acetone only. Six animals from each group were taken for skin hyperplasia analysis and were killed after 9, 26 and 52 weeks. The appearance of skin lesions were carefully followed and histopathological diagnosis was made for all neoplasms present at death. The statistical analyses on occurrence of skin tumour bearing animals and cumulated skin tumours, with corrections for survival did not reveal a difference between the controls and the MF exposed groups. The epithelial thickness of DMBA + MF-treated animals was of the same magnitude as for DMBA-treated animals. Leukaemia was a little more frequent among animals exposed to 0.5 mT MF compared to the control animals. However this difference was not statistically significant.