The effects of two naphthoquinones, juglone and plumbagin, and an isocoumarin, hydrangenol, on intestinal carcinogenesis in rats were examined by dietary exposure during the initiation phase. Starting at 5 weeks of age, male F344 rats were fed the diets containing either of the test chemicals at a concentration of 200 ppm or the control diet without the compounds. At 6 weeks of age, all animals were treated with s.c. injections of azoxymethane (AOM) (15 mg/kg body weight, once weekly for 3 weeks) or saline alone. Animals fed experimental diets were changed to the control diet 1 week after the last carcinogen treatment. Animals given plumbagin together with the carcinogen had a lower incidence (41%) and smaller multiplicity (0.48 +/- 0.62) of tumors in the entire intestine compared with those exposed to carcinogen alone (68% and 1.04 +/- 0.62) (P < 0.05 and < 0.01, respectively). The incidence and multiplicity of tumors in the small intestine (7% and 0.07 +/- 0.25) and the multiplicity of tumors in the entire intestine (0.60 +/- 0.76) of animals treated with juglone and the carcinogen were significantly less than those of animals treated with carcinogen alone (P < 0.05 in each). Hydrangenol tended to decrease the incidence and the multiplicity of tumors in the entire intestine induced by AOM, but the effect was not statistically significant. The present data suggest that the naphthoquinones, juglone and plumbagin, could be promising chemopreventive agents for human intestinal neoplasia.