There is evidence that hormone replacement therapy (HRT) may both stimulate and inhibit breast cancers, giving rise to a spectrum of activities, which are frequently hard to understand. Here we summarise the evidence for these paradoxical effects and, given the current data, attempt to give an indication where it may or may not be appropriate to prescribe HRT.It is clear that administration of oestrogen-progestin (E-P) and oestrogen alone (E) HRT is sufficient to stimulate the growth of overt breast tumours in women since withdrawal of HRT results in reduction of proliferation of primary tumours and withdrawal responses in metastatic tumours. E-P, E including tibolone are associated with increased local and distant relapse when given after surgery for breast cancer. For women given HRT who do not have breast cancer the only large randomised trial (WHI) of E-P or E versus placebo has produced some expected and also paradoxical results. E-P increases breast cancer risk as previously shown in observational studies. Risk is increased, particularly in women known to be compliant. Conversely, E either has no effect or reduces breast cancer risk consistent with some but not all observational studies. Two observational studies report a decrease or at least no increase in risk when E-P or E are given after oophorectomy in young women with BRCA1/2 mutations. Early oophorectomy increases death rates from cardiovascular and other conditions and there is evidence that this may be reversed by the use of E post-oophorectomy. HRT may thus reduce the risk of breast cancer and other diseases (e.g., cardiovascular) in young women and increase or decrease them in older women.