1. The effects of the muscarinic M2 receptor antagonist methoctramine, on contractions of airway smooth muscle induced by cholinergic nerve stimulation and by exogenously applied acetylcholine (ACh), have been investigated in vivo and in vitro in guinea-pigs. 2. Stimulation of the preganglionic cervical vagus nerve in anaesthetized guinea-pigs, caused bronchoconstriction and bradycardia which were mimicked by an intravenous dose of ACh. The muscarinic M2 antagonist, methoctramine (7-240 nmol kg-1), inhibited the bradycardia induced by both vagal stimulation and ACh (ED50: 38 +/- 5 and 38 +/- 9 nmol kg-1, respectively). In this dose-range, methoctramine facilitated vagally-induced bronchoconstriction (ED50: 58 +/- 5 nmol kg-1), despite some inhibition of ACh-induced bronchoconstriction (ED50: 81 +/- 11 nmol kg-1). The inhibition of ACh-induced bronchoconstriction and hypotension was dose-dependent, but was not statistically significant until doses of 120 nmol kg-1 and 240 nmol kg-1 respectively. 3. In the guinea-pig isolated, innervated tracheal tube preparation, methoctramine (0.01-1 microM) caused facilitation of contractions induced by both pre- and postganglionic nerve stimulation, whereas contractions induced by exogenously applied ACh were unaffected. Higher concentrations of methoctramine (greater than or equal to 10 microM), reduced responses to both nerve stimulation and exogenous ACh, indicating blockade of post-junctional muscarinic M3 receptors. 4 ACh caused a slow maintained increase in tone of the tracheal tube and at the same time reduced the contractions induced by nerve stimulation. This inhibitory effect of ACh on neuronally mediated responses was antagonized by methoctramine (0.01-1 microM) in a concentration-dependent manner. However, the ACh-induced tone change was unaffected by methoctramine in this concentration-range, indicating a lack of muscarinic M3 receptor antagonist activity in this concentration-range.5. The effect of methoctramine on responses induced by pre- and postganglionic nerve stimulation was not identical. At concentrations of methoctramine of 1 ,microM and greater, preganglionic stimulation-induced contractions were reduced when compared to those induced by postganglionic stimulation, suggesting an inhibitory effect of methoctramine on ganglionic transmission. This ganglion blocking action of methoctramine was not due to its reported M1 receptor antagonist activity (blocking facilitatory Ml receptors in the ganglia) since pirenzepine was without effect in this preparation. We believe that the ganglionic blocking action of metoctramine is due to its nicotinic receptor antagonist properties, since the concentration of methoctramine inhibiting ganglionic transmission in the tube preparation (1 microM) was shown to inhibit contractions induced by the nicotinic agonist, 1,1-dimethyl-4-phenyl-piperazine in tracheal strips.6. These results show that methoctramine is able to demonstrate adequately the presence of autoinhibitory receptors functionally both in vivo and in vitro and confirms their pre-junctional location on pulmonary cholinergic nerve terminals and their classification as muscarinic M2 subtypes. These results also indicate that while methoctramine is a potent muscarinic M2 receptor antagonist, it does not possess the required selectivity to discriminate between cholinoceptor subtypes in preparations, such as the airways, where mixed populations of muscarinic and nicotinic cholinoceptors exist.