Recent reports have thrown doubt on both of the major mechanisms which have been suggested for the regression of breast cancer under high-dosage oestrogen therapy—a direct effect on binding sites and an indirect inhibition of prolactin release. A searching review on the clinical response of 407 patients with advanced breast cancer treated by oestrogen therapy under my direction showed certain anomalies, including age determined tumour inhibition and stimulation, dose dependence of tumour inhibition and stimulation, differential site sensitivity, and tumour regression with oestrogen withdrawal. The hypothesis usually postulated to explain these anomalies is that the tumour may in the same individual be composed of multiple genetically distinct clones of cells. It is suggested instead that the effect of high-dosage oestrogen therapy in breast cancer may depend critically on the absolute and relative concentrations of prolactin and oestrogen actively available at the tumour. On this basis the clinical manifestation of tumour stimulation in response to oestrogen administration suggests that the oestrogen concentration at the target is inadequate. Differences in site sensitivity in the same patient may depend on tumoral factors such as the level of oestradiol and prolactin binding receptors in the tissue.