Carbachol was injected into the hypothalamic paraventricular nuclei (PVN) of conscious, unrestrained Long Evans rats, chronically instrumented with intravascular catheters and pulsed Doppler probes to assess changes in regional haemodynamics. Bilateral microinjections of carbachol (1 ng-1 microgram) produced increases in blood pressure, bradycardias and vasoconstrictions in renal, superior mesenteric and hindquarters vascular beds. In the presence of phentolamine, the bradycardic and hindquarters vasoconstrictor responses to carbachol were unchanged while the pressor response was smaller due to a reduction in the renal and the superior mesenteric vasoconstriction. In the presence of propranolol, the bradycardic response was reduced, but the pressor and renal vasoconstrictor responses were potentiated, whereas the superior mesenteric and hindquarter vasoconstrictions were not changed significantly. In the presence of phentolamine and propranolol, the heart rate and pressor responses, as well as the renal vasoconstriction, were unchanged, whereas the superior mesenteric vasoconstriction was reduced and the hindquarters vasoconstriction was potentiated. Together these results are consistent with an involvement of the sympathoadrenal system in the pressor response to carbachol injected into the PVN of untreated animals. They indicate that alpha-adrenoceptor-mediated vasoconstriction in the superior mesenteric vascular bed is a particularly important component in that regard. In the presence of the vasopressin antagonist, d(CH2)5(Tyr(Et))DAVP, alone or in combination with phentolamine and propranolol, the pressor response to carbachol was substantially reduced, while the renal and superior mesenteric vasoconstrictor effects were completely abolished; the bradycardia was not significantly affected by this treatment. These results indicate an important involvement of vasopressin in the cardiovascular responses to carbachol injected into the PVN of untreated animals. Moreover, in the presence of the vasopressin antagonist the hindquarters vascular bed showed a vasodilatation following PVN injection of carbachol; this effect was reversed to a vasoconstriction following combined i.v. pretreatment with the vasopressin antagonist, phentolamine and propranolol and hence was possibly due to circulating adrenaline acting on vasodilator beta 2-adrenoceptors. However, there was a residual hindquarters vasoconstriction raising the possibility that non-adrenergic, non-vasopressinergic vasoconstrictor mechanisms were influencing that vascular bed.