The chemopreventive effect of 40% and 80% maximum tolerated dose (MTD) levels of 1,4-phenylenebis(methylene)selenocyanate (p-XSC) administered in the diet during the initiation phase (2 weeks before, during, and up to 3 days after carcinogen administration) and the post-initiation phase (3 days after carcinogen treatment until termination) of azoxymethane (AOM)-induced colon carcinogenesis was studied in male F344 rats. The MTD of p-XSC was determined in male F344 rats and found to be 50 ppm. Beginning at 5 weeks of age, all animals were divided into various experimental groups (42 rats/group) and fed the high-fat semipurified diet or diets containing 20 (40% MTD) and 40 (80% MTD) ppm p-XSC. At 7 weeks of age, all animals (30 rats/group) except the vehicle-treated groups (12 rats/group) were administered s.c. injections of AOM (15 mg/kg body weight/week for 2 weeks). Three days after the second injection of AOM or vehicle (normal saline), groups of animals fed the p-XSC diets and control diet were transferred, respectively, to control diet and p-XSC diets and continued on these diets until the termination of the study. All animals were necropsied during the 36th week after AOM treatment. Colonic mucosal prostaglandin E2 and selenium-dependent glutathione peroxidase were measured in animals fed the control and p-XSC diets at the termination of the study. The results indicate that 40 ppm p-XSC administered during the initiation phase significantly inhibited the colon tumor incidence (percentage of animals with tumors). Dietary p-XSC administered at 20 and 40 ppm levels during the initiation phase significantly inhibited colon tumor multiplicity (tumors/animal and tumors/tumor-bearing animal). Colon tumor incidence and multiplicity were significantly reduced in groups fed 20 and 40 ppm p-XSC diets at the postinitiation phase of carcinogenesis. Colonic mucosal selenium-dependent glutathione peroxidase activity was increased, and prostaglandin E2 was reduced in animals fed the p-XSC diet compared to animals fed the control diet. Whereas the precise mechanisms of p-XSC-induced inhibition of colon carcinogenesis remain to be elucidated, it is likely that the effect during the initiation and postinitiation phases may be due to alteration in carcinogen metabolism and to modulation of prostaglandin synthesis and selenium-dependent glutathione peroxidase activity.