The neurochemical identity of preoptic neurons containing oestrogen receptors was investigated in the male and female rat using a sequential double-staining immunocytochemistry procedure. Single-immunostaining revealed large populations of cells with nuclear immunoreactivity to the oestrogen receptor in the medial preoptic area of the male and female rat. Optimal double-staining of sections for the oestrogen receptor and one of several neuropeptides or tyrosine hydroxylase, was achieved with short-term (two- to four-day) gonadectomized rats treated with colchicine where necessary. Neurotensin-immunoreactive cells were distributed in a sexually dimorphic manner in the region of the anteroventral preoptic nucleus and exhibited oestrogen receptor immunoreactivity in both sexes. Double-labelled cells in this area of the female rat comprised 50% and 11% of the total neurotensin- and oestrogen receptor-containing cell populations, respectively, compared with 25% and 4% in the male (P less than 0.01). The numbers of neurotensin-immunoreactive cells in the region of the medial preoptic nucleus were similar in male and female rats with double-labelled cells making up 20-38% and 3-5% of the total numbers of cells containing neurotensin and oestrogen receptors, respectively, in both sexes. Neurons immunoreactive for tyrosine hydroxylase were distributed in a gender-specific manner within the anterior periventricular area but were not immunoreactive for the oestrogen receptor in either sex. Following colchicine treatment, cholecystokinin-immunoreactive cells were identified predominantly within periventricular regions of the preoptic area and similarly, did not possess immunoreactivity to the oestrogen receptor in either the male or the female rat. Neurons containing luteinizing hormone-releasing hormone were found immediately lateral to the cell populations containing oestrogen receptors and immunoreactivity to the oestrogen receptor was not identified within any neurons containing luteinizing hormone-releasing hormone in either the male or female rat. The absence of oestrogen receptor immunoreactivity in neurons containing tyrosine hydroxylase, cholecystokinin or luteinizing hormone-releasing hormone suggests that gonadal steroids acting through this receptor do not influence these cells directly in either sex. In particular, it appears that gender-specific patterns of luteinizing hormone secretion cannot be attributed to sex differences in oestrogen receptor localization within luteinizing hormone-releasing hormone neurons. These experiments also show that the sexually dimorphic neurotensin neurons in the preoptic area possess oestrogen receptors and that female rats have larger number of neurons co-localizing neurotensin and oestrogen receptors. As such, these neurons may be involved in mediating sex-specific actions of the gonadal steroids in the preoptic area.