The purpose of this study was to examine the conditions whereby β-blockers cause a pressor response in conscious, unrestrained rats: (1) whether β-blockers cause a pressor response in rats subjected to, or not subjected to, nonselective α-blockade with phentolamine; (2) whether the pressor response to β-blockers is due to the blockade of vasodilator β2-adrenoceptors, and (3) whether it is due to an acute increase in the release of adrenaline (A) and noradrenaline (NA). In the first series of experiments cumulative dose-response curves for propranolol, atenolol and ICI 118,551, nonselective β-, β1 and β2-selective antagonists, respectively, were constructed in rats subjected to a continuous intravenous infusion of phentolamine. The administration of each of the β-antagonists caused a significant dose-dependent increase in mean arterial pressure (MAP). The ED5o values for the increase in MAP were found to be 3.6 ± 0.8, 10 ± 2.6 and 4.6 ± 0.8 μg/kg for propranolol, atenolol and ICI 118,551, respectively. In the second series of experiments, a single bolus injection of a selective or nonselective β-antagonist or saline vehicle was given to rats subjected to a continuous intravenous infusion of phentolamine. Plasma levels of A and NA were determined in the control condition, during the infusion of phentolamine and again after the injection of a β-antagonist. The infusion of phentolamine significantly decreased MAP and increased plasma levels of A and NA. During the infusion of phentolamine, a single bolus injection of propranolol (100 μg/kg), atenolol (100 μg/kg) and ICI 118,551 (30 μg/kg) significantly increased MAP, but did not alter NA and A levels relative to the injection of saline. It is concluded that (1) the pressor response to β-blockers occurs only in rats subjected to α-blockade, (2) pressor response also occurs with a selective β1-antagonist, atenolol, used at doses that do not alter the response of a selective β2-agonist and therefore it is not due to the blockade of vasodilator β2-adrenoceptors, and (3) it is not due to an acute increase in the release of catecholamines.