Experiments were conducted to determine whether the natural estrogen and an environmental compound with estrogenic action, 4-n-octylphenol (4nOP), could modify tumor development in human c-Ha-ras proto-oncogene transgenic (Tg) rats which are highly susceptible to mammary and skin carcinogens. Female and male Tg and non-transgenic (non-Tg) rats were given a single oral dose of 7,12-dimethylbenz[a]anthracene (DMBA) (25 mg/kg body weight) at 50 days of age and thereafter subcutaneously implanted with cholesterol pellets containing 0.01, 0.1 or 1.0 mg beta-estradiol 3-benzoate (E2) per rat or received diets containing 1000 or 100 p.p.m. 4nOP for 12 weeks in females or for 20 weeks in males. E2 reduced the mammary tumor incidence and multiplicity in a dose dependent manner, especially in female Tg rats. In contrast, E2 increased mammary tumor incidence and multiplicity at the lowest dose (0.01 mg), however it reduced skin tumor induction in male Tg rats. 4nOP at a dose of 100 p.p.m. decreased mammary tumor multiplicity in female Tg rats (P < 0.001). No effects were observed in males. In separate in vitro studies, E2 at low doses (10(-11)-10(-8) M) enhanced the growth of both MCF-7 and T47D cells and this was similarly the case for 4nOP at high doses (10(-7)-10(-5) M) in T47D cells. The finding that E2 and 4nOP at high doses caused reduction in mammary tumor development in female Tg and possibly non-Tg rats, may indicate that excess estrogen can exert a paradoxical inhibitory influence. E2 also appears to have bipotential effects in males, promoting mammary, but inhibiting skin carcinogenesis. These contrasting observations may be caused by differences in background physiological estrogen levels. In addition, the results suggest that Tg rats can be used in medium-term bioassay models to test for the modifying effects of estrogenic environmental compounds on mammary tumor development.