That estrogen plays a role in the regulation of mood has been postulated since extracts of animal ovarian tissue were administered to oophorectomized women at the end of the last century to alleviate psychological symptoms thought to be related to the removal of the ovaries. The occurrence of depressive symptoms in the perimenopause is associated with a variety of factors. A previous history of either depression and/or premenstrual syndrome as well as cognitive factors explain most of the variance. There are no consistent findings of a correlation between any serum hormone level and severity or presence of mood symptoms. Neurobiological studies show, with regard to an antidepressant effect, promising effects of estradiol on serotonergic, noradrenergic, cholinergic, dopaminergic and GABAergic functions. Progestogens seem to oppose some of these effects. The role of adrenergic hormones and DHEA(S) is less clear. Most clinical trials showed a modest effect on symptoms of depression. However, the predominantly poor methodological quality does not allow generalisation and recommendations. A "tonic" effect on well-being in non- or mild depressed women should not be regarded as true antidepressant effect. Results yielded in studies of surgically menopausal women may not be applicable to women with natural menopause. There is a great potential for exploring various types, doses, and routes of administration of both antidepressants and sex hormones. With regard to the domino theory, future studies should also focus on the mediation of treatment effects through alleviation of vasomotor symptoms or sleep quality.