The in vivo distribution of [3H]scopolamine in rat brain following establishment of constant, saturating arterial tracer concentrations was examined with the use of quantitative autoradiography. The equilibrium drug distribution, studied 240 min after initiation of tracer infusion, was highly correlated with the regional density of muscarinic receptor sites determined in vitro in the same animals by autoradiographic analysis of [3H]quinuclidinyl benzilate binding. Brain regions of highest receptor density were generally correlated with known terminal fields of cholinergic neurons, and they demonstrated a protracted time course of in vivo labeling. An exception was noted in the basal pons, where a receptor population of high density without documented cholinergic innervation was rapidly labeled. It is suggested that synaptic muscarinic receptors are labeled slowly, as a consequence of either restricted tracer accessibility or competition between tracer and endogenous acetylcholine for available binding sites, and that the pontine receptors may be functionally distinct from those in other brain regions. The in vivo equilibrium binding technique used in the present study results in regional tissue radioligand concentrations directly proportional to receptor density and may, thus, provide a basis for receptor imaging in the human brain by means of positron emission tomography.