D1 receptor mechanisms in the median eminence have been studied by means of immunocytochemistry using antisera against dopamine and cyclic AMP-regulated phosphoprotein-32 (DARPP-32) and tyrosine hydroxylase (TH) and by autoradiography using the iodinated analogue of the D1 receptor antagonist SCH-23390. The co-distribution of DARPP-32 and TH immunoreactivity (IR) and of DARPP-32 and luteinizing hormone releasing hormone (LHRH) IR was analysed in the median eminence by means of computer-assisted morphometry and microdensitometry. Functional analysis involved studies on the role of D1 receptors in the regulation of LH serum levels in rats treated with nicotine in the absence and presence of the D1 receptor antagonist. LH serum levels were measured by means of radioimmunoassay procedures. The results on the co-distribution of TH and DARPP-32 IR in the median eminence which were obtained both by analysis of adjacent sections and by two-colour immunocytochemistry on the same section, demonstrated a high degree of overlap of TH and DARPP-32 IR nerve terminals and tanycytes within the medial and lateral palisade zone. Furthermore, studies on LHRH and DARPP-32 IR nerve terminals and tanycytes in the median eminence with the same methodologies demonstrated preferential overlaps within the lateral palisade zone. The overlap area was about 50% of the LHRH or DARPP-32 immunoreactive area in this region. Density maps were also obtained on the distribution of LHRH and DARPP-32 immunoreactive profiles at various rostrocaudal levels. Correlation studies demonstrated a significant and positive co-distribution of LHRH and DARPP-32 immunoreactive terminals and tanycytes within the lateral palisade zone and the subependymal layer (when all DARPP-32 positive squares were considered) of the median eminence. Instead within the medial palisade zone a significant negative correlation coefficient was found, when all the LHRH positive squares were considered. In the receptor autoradiographical analysis a weak-to-moderate labelling was obtained of the part outside the mediobasal hypothalamus using the D1 receptor radioligand [(125)I]SCH-23982, while hardly any labelling was found within the median eminence and the arcuate nucleus. SCH-23390 was found to counteract, in a dose-related way, the inhibitory effects of intermittent nicotine treatment on serum LH levels. The D2 receptor antagonist raclopride in a dose of 1 mg/kg did not counteract the inhibitory effects of nicotine on serum LH levels. The present immunocytochemical, autoradiographic and functional studies suggest the existence of a D1 receptor in the median eminence which can be blocked by the D1 receptor antagonist SCH-23390 in behaviourally relevant doses and which is masked under basal conditions in the male rat. It is proposed that one type of median eminence D1 receptor is located on the axon terminals, not linked to DARPP-32, and which may make possible a rapid regulation of hypothalamic hormone release, e.g. LHRH release from the nerve terminals in the lateral palisade zone as indicated in the present morphological and functional experiments. The other type of median eminence D1 receptor may be located on the tanycytes and linked to DARPP-32. It is suggested that this D1 receptor is responsible for a long-term regulation of hypothalamic hormone release inter alia LHRH release from the terminal and preterminal parts of the LHRH axons in the lateral palisade zone and subependymal layer, respectively.