A case-control study of lung cancer was conducted within a cohort of painters and allied tradesmen drawn from the New York State membership of a large international union. Lung cancer cases were identified through the New York State Cancer Registry. Controls were randomly selected from the same union population. Data collected by mailed questionnaire included lifetime work history, work environment, and life-style factors. An almost threefold excess risk (odds ratio, 2.75) of lung cancer was associated with a usual occupation of painting. When a logistic regression model was used to adjust for other risk factors such as cigarette smoking, a fivefold excess lung cancer risk was observed among painters who never wore a mask or respirator; no excess risk was observed among painters who wore a mask or respirator.