Much uncertainty remains about the type of sunlight exposure that most increases risk of cutaneous melanoma and the role of diet. The authors conducted a population-based case-control study (1986-1992) of Caucasians living on Oahu, Hawaii; included were 278 cases and age- and sex-matched controls. Plasma samples were measured for retinol, carotenoids, and alpha-tocopherol by high-pressure liquid chromatography. Selenium was measured in blood and toenails by neutron activation. Celtic and English ancestries, migration to Hawaii before age 20 years, fair complexion, inability to tan, and family history of skin cancer, as well as height, hours spent in the summer sun, blistering sunburns during adolescence, and moles, all increased melanoma risk. With regard to diet and biomarkers, only alcohol was associated with risk. The odds ratios for increasing tertiles of lifetime ethanol intake were 1.0, 1.2 (95% confidence interval (CI): 0.6, 2.2), and 2.3 (95% CI: 1.2, 4.4) (p for trend = 0.01) for men and 1.0, 1.1 (95% CI: 0.5, 2.4), and 1.7 (95% CI: 0.7, 3.8) (p for trend = 0.19) for women. Dietary lipids were unrelated to risk, but polyunsaturated fat intake appeared to modify the effects of alcohol and toenail selenium on melanoma risk. These data provide additional evidence for the association of constitutional susceptibility factors, intense sun exposure (particularly before age 20 years), and alcohol consumption with melanoma risk.