Muscarinic acetylcholine receptors were identified by the specific binding of [H](-)quinuclidinylbenzilate ([H-3](-)QNB) and [H-3]oxotremorine-M ([H-3]Oxo-M), to membranes isolated from the sino-atrial (SA) node and right atrium (RA) of bovine heart. The density of [H-3](-)QNB binding sites was greater in the SA node compared to the RA. Specific [H-3](-)QNB binding was saturable and occurred to a single population of binding sites in both regions. The binding of antagonists, as assessed by competition with [H-3](-)QNB, also occurred to a single population of sites; the binding affinities of all antagonists were similar in either region. Agonist competition curves, except for McN-A-343, were complex and computer analyses indicated that agonists bound to at least two populations of binding sites that differed in affinity. The proportion of high-affinity agonist binding sites was consistently greater in the SA nodal, relative to the RA membranes, while the affinity of the high-affinity agonist binding sites to a given agonist was essentially similar in either region. The high-affinity binding of [H-3]Oxo-M was saturable and occurred to a single population of sites. The maximal binding of [H-3]Oxo-M in the SA nodal membranes was higher than in the RA membranes. Guanine nucleotides and N-ethylmaleimide (NEM) markedly decreased [H-3]Oxo-M binding; NEM did not appear to influence guanine nucleotide-dependent decrease in [H-3]Oxo-M binding. Phospholipase A2 decreased both [H-3](-)QNB and [H-3]Oxo-M specific binding, the latter being affected to a greater extent. Phospholipase C also decreased [H-3](-)QNB and [H-3]Oxo-M binding, although to a lesser degree compared to phospholipase A2. Either lipase, however, increased the guanine nucleotide-sensitive agonist binding. Analysis of [H-3](-)QNB binding to microsomal subfractions showed that binding sites were enriched in the light plasma membrane fractions that were also enriched in pertussis toxin sensitive guanine nucleotide binding proteins.