1 The effects of glibenclamide (a blocker of adenosine triphosphate [ATP]-dependent K+-channels) on the inhibition of uterine contractions by relaxin, salbutamol and cromakalim were compared in vivo.2 Glibenclamide (20mgkg-1) did not antagonize salbutamol. Glibenclamide produced a parallel rightward shift in the dose-response curve to cromakalim with a 5.5 fold decrease in uterine sensitivity (postvehicle log ID50, -0.87mgkg'1; post-glibenclamide log ID50, -0.07mgkg-1). Glibenclamide produced a non-parallel rightward shift in the dose-response curve to relaxin (post-vehicle log ID50, 0.99 ug kg1; post-glibenclamide log ID50, 2.28 jug kg-1).3 Glibenclamide reversed established inhibition of uterine contractions by cromakalim or relaxin but not that by salbutamol. 4 Insulin produced no antagonism of relaxin on isolated uterus of the rat, demonstrating that glibenclamide antagonism of relaxin in vivo is not by released insulin. Apamin did not antagonize relaxin in vivo, suggesting that small calcium-activated K+-channels are not involved in the action of relaxin. 5 Comparison of the lack of antagonism of salbutamol with the non-competitive-like antagonism of relaxin by glibenclamide suggests that relaxin does not relax uterine smooth muscle predominantly by increasing intracellular adenosine 3', 5'-cyclic monophosphate concentrations. Comparison of the noncompetitive-like antagonism of relaxin and the competitive-like antagonism of cromakalim by glibenclamide suggests that the two relaxants may share, in part, a common mechanism of action and that additional mechanism(s) may also be involved in the inhibitory action of relaxin.