1. The interactions between angiotensin II (AII), two non-peptide antagonists DuP 753 and IMI, and eight peptide analogues of AII were investigated on the rabbit isolated aorta assay. DuP 753 and IMI behaved as simple competitive antagonists (pKB values 8.4 and 6.8, respectively). To different degrees, all the AII-peptide analogue interactions failed to meet the basic criteria for simple competition. In addition to rightward shift, the most significant feature was a concentration-dependent saturable depression of the upper asymptote of the AII concentration-effect curves. 2. 'Washout' and combined dose-ratio analysis experiments, in which DuP 753 was used as a reference antagonist, indicated that the profile of peptide antagonism was solely due to a reversible and syntopic action at the AII receptor. 3. By use of an operational model of agonism (Black & Leff, 1983) as a starting point, it was possible to account for the data with a new model which describes reversible receptor occupancy and occupied receptor-determined, saturable reduction in the efficacy of AII. Model-fitting gave estimates of pKB values for the peptide analogues and agonist affinity and efficacy parameters for AII. 4. The model was successfully tested by applying it to qualitatively similar results obtained in a cross-tissue analysis on guinea-pig aorta, ileum and stomach. 5. A 'molecular' interpretation of the efficacy changes, based on the concepts of receptor internalisation and expression, is offered.