1 The purpose of the study was to assess the bronchorelaxant effects of the beta3-adrenoceptor agonist BRL 35135 in normal human airways. 2 Eight healthy male subjects were studied, having previously demonstrated airways responsiveness to inhaled salbutamol 200 microg, with a group mean (+/- s.e. mean) fall in airways resistance (Raw), from baseline, of 37 +/- 5%. 3 Subjects attended the laboratory on 3 separate days, having fasted and taken placebo (PL) or nadolol 20 mg (N20), 2 h previously. 4 After 30 min rest, baseline measurements of Raw, serum potassium, glucose and free fatty acid were performed before subjects were given single oral doses of BRL 35135 8 mg (BRL) or placebo BRL. Measurements were repeated 60 min after the BRL or placebo BRL were given. 5 There was a significant (P < 0.05) fall in Raw (% change from baseline, as means and 95% CI) with PL/BRL: -32(-18, -46), compared with either PL/PL: -8(5, -21), or N20/BRL: -11(2, -24). There was no significant difference between PL/PL and N20/BRL. 6 A similar pattern to Raw was observed for both of the beta2-mediated metabolic responses which were also antagonised by nadolol. For the potassium response (mmol l(-1)), there was a significant (P < 0.05) difference between PL/BRL: -0.50(-0.31, -0.69), in comparison with either PL/PL: 0.08(-0.11, 0.27) or N20/BRL: 0.09(-0.10, 0.28), but values for PL/PL and N20/BRL were not significantly different. In contrast, with the free fatty acid response (nmol 1(-1)), the increase seen with N20/BRL: 85(1.0, 171.0) was significantly (P < 0.05) different from PL/PL: 3.7(-82.3, 89.8), but was not different from PL/BRL: 132.5(46.5, 218.5). 7 In conclusion, BRL 35135 produced airways, potassium and glucose responses which were antagonised by nadolol, whereas the lipolysis response was not. This suggests that there are not functional beta3-adrenoceptors in human airways.