Associations between individual foods and nutrients and colorectal cancer have been inconsistent, and few studies have examined associations between food, nutrients, dietary patterns, and rectal cancer. We examined the relationship between food groups and dietary patterns and risk for rectal cancer in non-Hispanic Whites and African-Americans. Data were from the North Carolina Colon Cancer Study-Phase II and included 1,520 Whites (720 cases, 800 controls) and 384 African-Americans (225 cases, 159 controls). Diet was assessed using the Diet History Questionnaire. Multivariate logistic regression models were used to estimate odds ratios and 95% confidence intervals. Among Whites, non-whole grains and white potatoes were associated with elevated risk for rectal cancer whereas fruit, vegetables, dairy, fish, and poultry were associated with reduced risk. In African-Americans, high consumption of other fruit and added sugar suggested elevated risk. We identified three major dietary patterns in Whites and African-Americans. The high fat/meat/potatoes pattern was observed in both race groups but was only positively associated with risk in Whites (odds ratio, 1.84; 95% confidence interval, 1.03-3.15). The vegetable/fish/poultry and fruit/whole grain/dairy patterns in Whites had significant inverse associations with risk. In African-Americans, there was a positive dose-response for the fruit/vegetables pattern (P(trend) < 0.0001) and an inverse linear trend for the legumes/dairy pattern (P(trend) < 0.0001). Our findings indicate that associations of certain food groups and overall dietary patterns with rectal cancer risk differ between Whites and African-Americans, highlighting the importance of examining diet and cancer relationships in racially diverse populations.