Limited information is available on the association between the risk of cutaneous malignant melanoma (CMM) and dietary factors. The issue was investigated using data from a case-control study conducted in Italy between 1992 and 1994, including 542 patients with incident, histologically confirmed CMM and 538 controls, admitted to the same hospitals as cases for non-dermatologic and non-neoplastic diseases. We found a significant inverse association between vitamin A intake and CMM risk. The multivariate odds ratio, after allowance for phenotypic characteristics and history of sunburns, for the highest, compared with the lowest, quartile of intake was 0.71 (95% confidence interval (CI) 0.50-1.02) for beta-carotene, 0.57 (95% CI 0.39-0.83) for retinol, and 0.51 (95% CI 0.35-0.74) for total vitamin A. We found no appreciable association of CMM risk with selected food items, including fish, meat, vegetables, fruit, dairy products, wholemeal bread, alcohol, coffee and tea drinking. Consumption of tea appeared to have a protective effect on CMM risk. The relations with measures of dietary vitamin A are, however, moderate compared with the associations between phenotypic characteristics, sun exposure and number of naevi and CMM risk.