1 beta-adrenoceptor-mediated relaxation was investigated in ring preparations of rat isolated thoracic aorta. Rings were pre-constricted with a sub-maximal concentration of noradrenaline (1 mu M) and relaxant responses to cumulative concentrations of beta-adrenoceptor agonists obtained. 2 The concentration-response curve (CRC) to isoprenaline was shifted to the right by propranolol (0.3 mu M) with a steepening of the slope. Estimation of the magnitude of the shift from EC50 values gave a pA(2) of 7.6. Selective beta(1)- and beta(2)-adrenoceptor antagonists, CGP 20712A (0.1 mu M) and ICI 118551 (0.1 mu M), respectively, produced 4 and 14 fold shifts of the isoprenaline CRC. 3 Atypical beta-adrenoceptor agonists also produced concentration-dependent relaxation of aortic rings. The order of potency of the beta-adrenoceptor agonists was (-log EC50): isoprenaline (6.25) > cyanopindolol (5.59) > isoprenaline + propranolol (5.11) > CGP 12177A (4.40) > ZD 2079 (4.24) > ZM 215001 (4.07) > BRL 37344 (3.89). Relaxation to CGP 12177A and ZM 215001 was unaffected by propranolol (0.3 mu M). 4 SR 59230A (less than or equal to 1 mu M) and cyanopindolol (1 mu M), beta(3)-adrenoceptor antagonists, had no effect on the isoprenaline (in the presence of propranolol) or CGP 12177A CRCs. Bupranolol and CGP 20712A, at mu M concentrations (beta(4)-adrenceptor antagonists), inhibited responses to isoprenaline (in the presence of propranolol) and CGP 12177A. 5 In conclusion, atypical beta-adrenoceptors co-exist with beta(1)- and beta(2)-adrenoceptors in rat aorta. Although non-conventional partial agonists and selective beta(3)-adrenoceptor agonist cause relaxation, the vascular atypical beta-adrenoceptor does not appear to correspond to the beta(3)-adrenoceptor. There are, however. similarities with the putative beta(4)-adrenoceptor.