Abstract Consumption of fish rich in n-3 polyunsaturated fatty acids (PUFAs), such as docosahexaenoic acid, is suggested to reduce colorectal cancer risk through inhibition of the arachidonic acid (AA) cascade related to tumorigenesis and cell proliferation. High intake of saturated fatty acids (SFAs) may increase the risk. To examine associations between colorectal cancer risk and fatty acid compositions in erythrocyte membranes, as biomarkers for dietary intakes of fish, fat, and fatty acids, we conducted a case-control study with 74 incident cases and 221 noncancer controls (matched by age, sex, and season of sample collection). Erythrocyte fatty acids were measured using an accelerated solvent extraction and a gas-liquid chromatography. Colorectal cancer had no association with dietary intakes of meat, fish, fat, and fatty acids. However, the risk was inversely associated with erythrocyte compositions of docosahexaenoic acid, AA, and PUFAs [the highest to the lowest tertile, odds ratios, 0.36, 0.42, and 0.15; 95% confidence intervals, 0.14-0.93, 0.18-0.95, and 0.05-0.46; Ptrend < 0.05, respectively] and positively with those of palmitic acid, SFAs, and the ratio of SFAs/PUFAs (odds ratios, 6.46, 8.20, and 9.45; 95% confidence intervals, 2.41-17.26, 2.86-23.52, and 2.84-31.43; Ptrend < 0.005, respectively). In conclusion, we could clearly show decreased and increased risks for colorectal cancer related to PUFAs and SFAs compositions in erythrocyte membranes, respectively, but further research is needed to investigate the discrepancy between our findings and the generally accepted role of the AA cascade. (Cancer Epidemiol Biomarkers Prev 2006;15(10):1791–8)