Evidence supporting a cholinergic hypothesis of depression is presented. First, cholinergic overdrive produces behavioral, neuroendocrine, and polysomnographic features of melancholia, and melancholics exhibit state-independent supersensitivity to cholinergic overdrive. Drugs inducing up-regulation and supersensitivity of cholinergic systems produce behavioral, polysomnographic, and neuroendocrine effects of melancholia when withdrawn. These observations also implicate cholinergic system supersensitivity as a factor in the pathophysiology of certain affective disorders. Cholinergic and monoaminergic mechanisms reciprocally regulate drive-reduction, and substances of abuse either activate monoaminergic networks or antagonize cholinergic systems. These points are consistent with the hypothesis that dynamic interaction between cholinergic and monoaminergic systems is involved in the regulation of mood and affect. Finally, antimuscarinic agents have antidepressant effects. Thus, the hypothesis that supersensitivity of cholinergic systems is involved in the pathophysiology of affective disorders is supported by several lines of evidence. This evidence is reviewed; directions for future research and promising methods of investigation are discussed.