We compared the mammary carcinogenicity in female CD rats of three fjord region diol epoxides to test our hypothesis that such sterically hindered molecules would be potent carcinogens. The diol epoxides tested were racemic anti-3,4-dihydroxy-l,2-epoxy-l,2,3,4-tetrahydrobenzo[c]phenanthrene (BcPDE), anti-11,12-dihydroxy-13,14-epoxy-11,12,13,14-tetrahydrobenzo[g]chry-sene (BgCDE) and anti-11,12-dihydroxy-13,14-epoxy-11,12,13,14-tetrahydrodibenzo[a,l]pyrene (DB[a,l]PDE). Each diol epoxide was dissolved in dimethylsulfoxide (DMSO) and injected under the six nipples on the left side of the rat, with DMSO only being injected under the nipples on the right side. The total dose of each diol epoxide was 1.2 μmol/rat and there were 20 rats/group. The experiment was terminated 41 weeks after treatment. All three diol epoxides were potent mammary carcinogens, with activity greater than previously observed for a bay region diol epoxide, anti-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE). DB[a,l]PDE induced tumors most rapidly, followed by BcPDE and BgCDE. However, different types of tumors were induced. For induction of adenomas and adenocarcinomas, BcPDE and BgCDE had comparable potency; both were more active than DB[a,l]PDE. In contrast, for induction of sarcomas, DB[a,l]PDE was significantly more active than BcPDE and BgCDE. The results of this study support our hypothesis that sterically hindered fjord region diol epoxides are potent mammary carcinogens in the rat.