The possibility that vasopressin plays a role in cardiovascular control arouses increasing interest. We studied in unanesthetized dogs the hemodynamic consequences of 1-hour vasopressin infusions that modified plasma concentrations over a range similar to that found in physiological situations. We also examined the cardiovascular events following the stimulation of endogenous vasopressin release by an increase in plasma osmolality. In dogs with baroreceptor reflexes intact, vasopressin infusions which increased plasma vasopressin concentration by 2-20 fmol/ml did not affect mean arterial pressure. However, they significantly decreased cardiac output (measured by an electro-magnetic flowmeter) and increased total peripheral resistance. After baroreceptor denervation, vaso-pressin infusion rates as low as 40 fmol/kg per min (0.017 mU/kg per min) led to an increase in mean arterial pressure. Cardiac output was unaffected until much higher infusion rates were used. Changes in total peripheral resistance were very similar to those calculated in dogs with intact baroreceptors. The release of vasopressin following infusions of hypertonic solutions either intravenously or into a carotid artery induced detectable hemodynamic changes which appeared in many respects similar to those following low infusion rates of vasopressin. We conclude that physiological plasma concentra-tions of vasopressin have hemodynamic effects even though they do not normally modify arterial pressure, presumably because of some particular interaction of vasopressin with the baroreceptor reflex. Circ Res 47: 346-355, 1980