High levels of dietary fish oils have been shown to reduce tumor incidence in human epidemiological studies and animal studies. Natural killer (NK) cells are leukocytes that can lyse tumors without prior sensitization and without major histocompatibility complex restriction. The purposes of these studies were to determine the effects of different dietary levels (0, 10 or 18.6%) of menhaden fish oil (experiment 1) and different sources (com oil, beef tallow or fish oil) of dietary lipid (experiment 2) on NK cell activity of C57/B mice. Animals fed 18.6% fish oil had suppressed (P <.05) NK cytotoxicity compared with animals fed no dietary fish oil. Average daily gains and final weights were lower (P <.01) for animals fed 18.6% fish oil compared to animals fed 0 or 10% fish oil. In experiment 2, animals fed fish oil had lower (P <.05) NK cytotoxicity at a high effector to target (E: T) ratio than animals fed beef tallow, com oil or the control diet. Therefore, the effect of high dietary fish oil on tumor growth does not appear to be mediated through NK activity. Instead, fish oils may cause a direct reduction in tumor growth or suppressed tumor growth among rodents fed diets containing high concentrations of fish oil and this tumor-reducing effect may be mediated by a non-NK immune mechanism.