The relationship between diet and colorectal cancer mortality was analyzed in a prospective study of 45,181 men and 62,643 women aged 40-79 yr enrolled in the Japan Collaborative Cohort Study. Between 1988 and 1990, subjects completed a self-administered questionnaire on their sociodemographic characteristics, diet, and other lifestyle habits. During the follow-up period (average 9.9 yr), 284 colon cancer deaths (138 men and 146 women) and 173 rectal cancer deaths (116 men and 57 women) were confirmed. The only significant association of colorectal cancer mortality with vegetable intake was observed between male rectal cancer mortality and green leafy vegetable consumption [hazard ratio (HR) using Cox proportional hazard models = 0.6; 95% confidence interval (CI) = 0.3-0.9; P for trend = 0.02]. Yogurt intake was also inversely associated with male rectal cancer mortality (HR = 0.5; 95% CI = 0.2-1.0; P for trend = 0.04). Egg consumption was positively associated with male colon cancer mortality (P for trend = 0.04). Women with high fruit consumption had increased colon cancer mortality (HR = 1.6; 95% CI = 1.0-2.6; P for trend = 0.04). It should be noted that this study lacked statistical power due to small sample size and measurement error in the food-frequency questionnaire. Further investigation is therefore necessary to confirm the association between diet and colorectal cancer, especially by subsites and gender.