Experiments were performed to determine the pulmonary vascular responses to exogenous or endogenous arginine vasopressin (AVP) in rats. Both in vitro and in vivo approaches were used to examine the direct pulmonary vasoactive properties of AVP and how those properties affect pulmonary hemodynamics in the intact animal. In conscious, unrestrained rats, constant infusion of AVP (4.0 mU.kg-1.min-1 iv) resulted in a fall in mean pulmonary artery pressure (PAP), although systemic pressure was increased. Coincident with the fall in PAP were similar reductions in cardiac output and heart rate. Similarly, bolus administration of AVP reduced PAP, and this effect was augmented during hypoxia. Another series of experiments examined the effect of endogenous AVP released by arterial hypoxemia on pulmonary hemodynamics in conscious rats. Administration of a specific V1-vasopressinergic antagonist had no effect on the PAP response to hypoxia; however, systemic resistance tended to fall following V1-antagonism. To determine the vasoactive properties of AVP independent of these changes in blood flow, a series of experiments were performed on isolated, perfused rat lungs. Injection of 25, 200, or 2,000 mU of AVP into the circulation of the isolated lung was without effect under normoxic conditions. In contrast, 25 mU AVP elicited reproducible pulmonary vasodilation when injected during ongoing hypoxic pulmonary vasoconstriction. This vasodilatory response was unaffected by meclofenamate or by the platelet-activating factor receptor antagonist SRI 63-441, but was blocked by a specific V1-vasopressinergic antagonist. We conclude that although AVP exerts profound systemic vasoconstriction, the pulmonary circulation appears relatively unaffected by exogenous or endogenous AVP in vivo.(ABSTRACT TRUNCATED AT 250 WORDS)