The comparative behavioral and neurochemical activities of the muscarinic cholinergic antagonists scopolamine, trihexyphenidyl and pirenzepine, and the nicotinic cholinergic antagonist mecamylamine were evaluated in rats. The three muscarinic antagonists, but not the nicotinic antagonist, impaired memory performance in a spatial alternation task. The minimal effective doses required to disrupt behavior were 0.03, 1.0 and 10 mg/kg for scopolamine, trihexyphenidyl and pirenzepine, respectively. Scopolamine and trihexyphenidyl inhibited ex vivo binding of [H-3] pirenzepine to M1 receptors in cerebral cortex, indicating ready penetration into the brain. In contrast, pirenzepine penetrated into the brain at relatively high doses, suggesting poor penetration into the brain. Scopolamine and trihexyphenidyl, but not pirenzepine, inhibited [H-3]quinuclidinyl benzilate binding to brain stem M2 receptors ex vivo after subcutaneous administration. In addition, scopolamine and trihexyphenidyl, but not pirenzepine, decreased acetylcholine (ACh) levels in striatum and hippocampus, presumably by increasing ACh release by blocking ACh feedback inhibition at M2 receptors. Scopolamine and trihexyphenidyl also produced modest decreases in levels of the dopamine metabolite 3,4-dihydroxyphenylacetic acid in striatum, most likely due to blockade of M1 heteroreceptors on dopamine nerve terminals. The present results are consistent with the interpretation that muscarinic antagonists impair memory performance in rats, at least in part, by blocking M1 muscarinic receptors. The present results do not support a role for blockade Of M2 receptors. Further research is needed to determine the extent to which blockade of other (M3, M4, M5) muscarinic receptor subtypes contributes to the memory-impairing effects of muscarinic cholinergic antagonists.