Height and weight and derivations thereof are positively associated with a number of cancers. While several authors have reported an increased risk of melanoma among people at the higher extremes of these measures, the association has not been fully explored. New cases of primary cutaneous melanoma in 1997 in western Washington State (n = 386) were compared to controls selected by random-digit dialing (n = 727). Each study participant completed a telephone survey, and data were collected on height, weight, sun-related melanoma risk factors, demographic characteristics, as well as habits such as diet and exercise. Risk of melanoma was analyzed by logistic regression with adjustment for age, hair color, lifetime sun exposure, and fruit and vegetable intake. An excess risk of melanoma was identified in men in the upper quartiles of height (OR = 2.4, 95% confidence interval (CI) = 1.3-4.5), weight (OR = 2.8, CI = 1.5-5.2), and body surface area (OR = 2.8, CI = 1.5-5.1) vs. the lowest quartiles. In women, no association was present for any anthropometric measure. In addition, we found that men and women exercising five to seven days per week were at a decreased risk of melanoma (OR = 0.7, CI = 0.5-1.0). The anthropometric findings are largely consistent with previous studies, while this is the first report of an association of exercise with melanoma risk. The mechanisms for the effect of exercise and for the difference between men and women in the effect of anthropometric factors are unknown. Future research in basic and epidemiologic science should focus on biochemical or behavioral explanations for these observations.