The experiments reported in this study were designed to examine the question of whether a mammary epithelial cell's independence from hormonal requirements is established at the time of carcinogenic initiation, or whether the emergence of hormone independence is associated with the process of tumor progression. A newly developed rat model of mammary carcinogenesis was used in which the latency period to lesion detection is very short and in which the frequencies of both pre-malignant and malignant mammary lesions can be quantified. Two experiments were conducted in Sprague-Dawley rats injected with 50 mg MNU/kg body wt at 21 days of age. In the first experiment 47 animals were ovariectomized after the detection of a mammary tumor of palpable size. Forty-six of the 47 tumors assessed, all of which were subsequently classified as mammary gland adenocarcinomas, regressed to <50% of their initial volume within 14 days of bilateral ovariectomy. However, both pre-malignant and malignant mammary gland lesions were observed when animals were killed. In Experiment 2 a total of 60 rats were ovariectomized 7 days after MNU was injected. At 35 days post carcinogen ovariectomized animals had a higher incidence of intraductal proliferations than sham-operated controls (P = 0.03); there was no effect of ovariectomy on the incidence of ductal carcinoma in situ or carcinoma. The multiplicity of intraductal proliferations was increased by 58% in ovariectomized rats (P = 0.12), but the number of mammary carcinoma per rat was reduced (3.8 vs. 1.57, P = 0.02). These data are consistent with the hypotheses that the progression of pre-malignant to malignant lesions is inhibited in the mammary gland by ovariectomy and that the hormone independent phenotype can be conferred at the time of carcinogenic initiation.