The purpose of this study is to test the long-standing hypothesis that endogenous agents found in human breast fluid and in plasma are potential initiators of breast cancer. Therefore, we evaluated the tumorigenicity in the mammary glands of female CD rats of cholestan-5 alpha,6 alpha-epoxy-3 beta-ol (cholesterol-alpha-epoxide), cholestan-5 beta,6 beta-epoxy-3 beta-ol (cholesterol-beta-epoxide), and 1,5(10)estradiene-3,14,17-trione (estrone-3,4-quinone). As a positive control, trans-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydrobenzo[c]phenanthren e (BcPDE) was used. Rats were fed a high-fat AIN-76A diet (23.5% corn oil) to mimic the Western dietary composition. Because literature data suggest that the endogenous agents tested in this study are weak electrophiles, the total doses of cholesterol epoxides (12.3 mumol/rat) and of estrone-3,4-quinone (30 mumol/rat)were 10- and 25- fold higher, respectively, than that of BcPDE (1.2 mumol/rat). Each agent was dissolved in DMSO, and one-sixth of the total dose was injected under each of six nipples on the right side. The thoracic glands of the rat were treated at 30 days of age, and those located in the inguinal area were treated on the following day. The experiment was terminated at 44 weeks after treatment. Consistent with our previous study, BcPDE was a strong mammary carcinogen. However, there were no differences between rats treated with DMSO alone or those receiving DMSO containing cholesterol-alpha-epoxide, cholesterol-beta-epoxide, or estrone-3,4-quinone. The results of this study clearly indicate, for the first time, that metabolites derived from cholesterol and estrone lack tumorigenic activity in the rat mammary gland, at least under the conditions of the present protocol.