Several chemical compounds have been proven to have a tumorigenic effect in a number of different animal species. Estrogens, on the other hand, do not produce comparable effects in several different species, or a response of the magnitude that is characteristic of a carcinogen; indeed a positive effect of estrogen is restricted to certain situations. It is often stated, for example, that estrogens cause breast cancer, but it is not stated that the effect is confined to male mice from specific inbred strains that possess a mammary tumor virus. The effect of estrogens in the female rat is variable; it may be without effect, or it may produce mammary cancer in as many as 27% of the treated animals. Many clinical studies do not demonstrate a relationship between exogenous estrogens and breast cancer. Similarly, the effect of estrogen on the uterine cervix or the endometrium of different animal species does not demonstrate a carcinogenic profile. The effect of estrogen is strain- and species-dependent, and when a positive response is obtained only a small proportion of the treated animals are affected, which is not typical of a carcinogen. Thus, definite conclusions on the carcinogenic effects of estrogens are not possible.