The chemical relationships among the 3 major families of endorphins are summarized, and the CNS distribution and possible circuit functions are discussed. The overlapping distributions of enkephalin and dynorphin pathways is intriguing, and their existence in classical hippocampal pathways has implications for limbic function which are not understood. Functionally the predominant effect of these systems is 'disenabling', in contrast with the 'enabling' effects of monoamines, although specific and reversible excitatory responses in hippocampal pyramidal neurons may have important behavioral implications. The complex theoretical and methodological issues that arise in attempting to establish transmitter correlates of psychiatric disorders defy simplistic hypotheses. Endorphin research, where well established agonists and antagonists are available, illustrates the problems facing other areas of peptide research in conceptualizing a peptide related pathophysiology of the major psychoses. The structural, functional and chemical indices of hypothetical forms of endorphin pathology, and the reliability of various radio-immunoassays used now to quantify known endorphins are discussed. The complex range of behaviors produced in animals by endorphin antagonists continue to make research in this area important to psychiatry.