Neonatal female mice of the BALB/cC3H/Crgl strain were given daily injections of 17 beta-estradiol, progesterone and prolactin, singly and in some combinations, for 5 days beginning within 36 hr after birth. Mice were killed at tumor age or by 12 months of age. Differential cell counts of the anterior pituitary showed that prolactin cells were more numerous in neomatally estrogen-treated mice and progesterone-treated intact mice than in control mice. Paired analysis of tumor-bearing and non-tumor-bearing mice of all groups revealed that the occurrence of prolactin cells was greater in the former than the latter. Counts of gonadotropes and thyrotropes did not show any significant correlation with mammary tumorigenesis. However, neonatal estrogen and/or progesterone treatment resulted in significantly decreased numbers of gonadotropes in intact mice. In ovariectomized mice, gonadotropes were significantly increased regardless of neonatal treatment. The present results support the suggestion that the stimulatory effects of neonatal steroid treatment of mammary tumorigenesis may be a consequence of increased prolactin secretion, resulting from sustained minimal estrogen secretion by the ovary.