Physiological and pathophysiological roles of endothelins are still unclear. One reason is that circulating endothelin levels in normal and pathological states are much lower than the concentrations necessary to elicit contractions in vitro. It is usually assumed that endothelin accumulates in diseased tissues and that, because of its degradation, only a small fraction of it reaches the systemic circulation. Such a hypothesis does not fit with recent observations showing that low circulating endothelin levels may be active. We show here that most of the current inferences about the actions of endothelin assume that the peptide acts in the vessel wall under conditions known as non-stoichiometric binding conditions, that is, under conditions in which the receptor concentration in tissues ([Ro]) is smaller than the equilibrium dissociation constant of endothelin receptor complexes (Kd). Under stoichiometric binding conditions (defined by the condition [Ro] > Kd), most ligand molecules are bound to receptors and cannot be present in a free form. Estimates of [Ro] and Kd from the literature suggests that in vivo endothelin probably binds stoichiometrically to its receptors. Under this condition, most of tissue endothelin is probably bound to receptors. It is therefore suggested that plasma endothelin levels are low probably because tissue free endothelin levels are low, and this is not inconsistent with the presence of high tissue levels of active (that is, bound) endothelin. When the topology of the vessels with respect to the site of production (or of delivery) of endothelin is considered, stoichiometric binding may also account for the higher sensitivity to Et-1 of in vivo preparations. It also suggests that autocrine and paracrine actions of Et-1 are favoured at low and high secretory rates respectively, thus providing an explanation for the dual (vasodilator and vasoconstricting) actions of endothelin. Finally, the stoichiometric binding model predicts that functional receptors also act as clearance receptors and provides an explanation for the observation that antagonists of endothelin receptors are also clearance antagonists.