The contributions of angiotensin II type 1 (AT1) and type 2 (AT2) receptors to the control of regional kidney blood flow were determined in pentobarbital-anesthetized rabbits. Intravenous candesartan (AT1 antagonist; 10 microg/kg plus 10 microg x kg(-1) x h(-1)) reduced mean arterial pressure (12%) and increased total renal blood flow (29%) and cortical laser-Doppler flux (18%) but not medullary laser-Doppler flux. Neither intravenous PD123319 (AT2 antagonist; 1 mg/kg plus 1 mg x kg(-1) x h(-1)) nor saline vehicle significantly affected these variables, and the responses to candesartan plus PD123319 were indistinguishable from those of candesartan alone. In vehicle-treated rabbits, renal-arterial infusions of angiotensin II (1 to 25 ng x kg(-1) x min(-1)) and angiotensin III (5 to 125 ng x kg(-1) x min(-1)) dose-dependently reduced renal blood flow (up to 51%) and cortical laser-Doppler flux (up to 50%) but did not significantly affect medullary laser-Doppler flux or arterial pressure. Angiotensin(1-7) (20 to 500 ng x kg(-1) x min(-1)) had similar effects but of lesser magnitude. CGP42112A (20 to 500 ng x kg(-1) x min(-1)) did not significantly affect these variables. After PD123319 administration, angiotensin II and angiotensin III dose-dependently increased medullary laser-Doppler flux (up to 84%), and reductions in renal blood flow in response to angiotensin II were enhanced. Candesartan abolished renal hemodynamic responses to the angiotensin peptides, even when given in combination with PD123319. We conclude that AT2 receptor activation counteracts AT1-mediated vasoconstriction in the renal cortex but also counteracts AT1-mediated vasodilatation in vascular elements controlling medullary perfusion. These mechanisms might have an important effect on the control of medullary perfusion under conditions of activation of the renin-angiotensin system.