This study was designed to assess the effects of hydralazine administered p.o. on regional circulation, including the carotid, coronary and renal arteries. Hydralazine in doses of 0.5, 1 and 5 mg/kg (p.o., n = 5) increased coronary blood flow by 9 +/- 4, 15 +/- 4 and 18 +/- 3 ml/min (P less than .05), respectively, and renal blood flow by 17 +/- 7, 26 +/- 12, 36 +/- 10 ml/min (P less than .05), respectively, but had no effects on carotid blood flow in conscious, normotensive dogs. The increase in coronary blood flow was correlated linearly with the rate pressure product (r = 0.581, P less than .05). After ganglionic blockade using chlorisondamine (2 mg/kg i.v.), hydralazine in a dose of 1 mg/kg (p.o., n = 4) increased renal blood flow by 45 +/- 6 ml/min (P less than .05), whereas mean arterial pressure, cardiac output, heart rate and coronary blood flow remained essentially unchanged. The increase in renal blood flow produced by hydralazine (1 mg/kg p.o., n = 6) was prevented by pretreatment with a cyclooxygenase inhibitor, indomethacin (2 mg/kg, i.v.). In vitro, hydralazine failed to relax segments of carotid, coronary and renal arteries. Our data indicate that hydralazine is a direct and preferential renal vasodilator. It produces an indirect coronary vasodilation related to an increase in myocardial oxygen demand. Its effect on renal circulation appears to be dependent on the prostaglandin system and is related mainly to a relaxation of arteriolar resistant vessels.