Binding of (-)-[3H]quinuclidinyl benzilate (QNB) to muscarinic sites in guinea-pig atrial and ileal longitudinal muscle homogenates showed the presence of a single population of binding sites in atria (KD = 41 (32-53) (95% confidence limits) pM; Bmax = 0.225 +/- 0.02 pmol/mg protein (3)) and two binding sites in the ileum (KD = 20.9 (8.8-49) pM and 11.3 nM; Bmax = 0.436 +/- 0.09 and 11.85 +/- 2.63 pmol/mg protein (4), respectively). Atropine, gallamine, and pancuronium displaced (-)-[3H]QNB binding from the high affinity binding sites in the two tissues in a dose-dependent manner with -log Ki values of 8.6, 6.4, and 6.9, respectively, in atria and 8.7, 6.8, and 6.9, respectively, in ileal longitudinal muscle. The lack of selectivity of gallamine and pancuronium in binding experiments differed from results obtained in isolated tissue experiments where these antagonists showed a marked difference in their ability to antagonize cholinomimetics in the two tissues. In addition, the Ki values for gallamine and pancuronium in ileal homogenates were ca. 130- and 16-fold lower, respectively, than their KB values determined from isolated tissue experiments. Attempts to correlate data from binding experiments and isolated tissue experiments using combinations of antagonists led to variable results attributed to differences in the rates of dissociation of the antagonists from muscarinic receptors. It is concluded that the interaction of gallamine or pancuronium with agonists or antagonists at muscarinic receptors is not a simple bimolecular interaction.