Rats fed either fish oil (n = 16) or corn oil (n = 16) in calorically and nutritionally balanced diets were injected with 1,2-dimethylhydrazine, which is a colorectal specific carcinogen; differential colorectal tumor induction was then measured. In addition, plasma peroxide concentrations were measured in rats in each dietary group as well as in a group receiving a low-fat diet, either with or without prior carcinogen treatment (n = 3 for each of the 6 groups). Tumor incidence did not differ between groups fed corn oil and fish oil. Tumor yield in the left colon was significantly lower in rats fed fish oil (p = 0.0499). Total colorectal tumors induced were also fewer in the rats fed fish oil (p = 0.065). Plasma peroxide concentrations were difficult to interpret because of the wide variation within groups. The data on tumor yield in the left colon support the hypothesis that a diet rich in n-3 fatty acids, which are found in fish oil, may be less supportive of colorectal tumor development than a diet rich in n-6 fatty acids, which is found in corn oil. However, the mechanism by which fish oil decreases tumor induction is still unknown.