High intake of red meat has been associated with increased risk of colorectal cancer in Western countries. There has been much interest in the role of n-3 polyunsaturated fatty acids (PUFA) in colorectal cancer prevention, but epidemiological findings are limited and inconsistent. The objective of our study was to examine associations of meat, fish and fat intake with risk of colorectal cancer, paying particular attention to the subsite within the colorectum. Data were from the Fukuoka Colorectal Cancer Study, a population-based case-control study, covering 782 cases and 793 controls. Diet was assessed by interview, using newly developed personal-computer software for registering semiquantitative food frequencies. The intake of beef/pork, processed meat, total fat, saturated fat or n-6 PUFA showed no clear association with the overall or subsite-specific risk of colorectal cancer. There was an almost significant inverse association between n-3 PUFA and the risk of colorectal cancer; the covariate-adjusted odds ratio for the highest (median 3.94 g/day) versus lowest (median 1.99 g/day) quintile of energy-adjusted intake was 0.74 (95% confidence interval 0.52-1.06, trend P=0.050). The consumption of fish and fish products was similarly inversely related to the risk although the association was not statistically significant. These associations were more evident for distal colon cancer; adjusted odds ratio for the highest versus lowest quintile of n-3 PUFA intake was 0.56 (95% confidence interval 0.34-0.92, trend P=0.02). Our findings do not support the hypothesis that consumption of red meat increases colorectal cancer risk but do suggest that high intake of fish may decrease the risk, particularly of distal colon cancer.