Mechanisms and variations in the food-induced increase in the bioavailability of propranolol were assessed by single-dose (80 mg) studies in healthy volunteers who took the drug on an empty stomach, immediately after a protein-rich breakfast, and together with a carbohydrate-rich, protein-poor breakfast. Concomitant intake of the protein-rich, but not the carbohydrate-rich, protein-poor breakfast, increased the bioavailability of propranolol in most, but not all, subjects. The food (protein) effect displayed much inter-individual variation, from a decrease to a 250% increase, which could be explained, at least in part, by a correlation between the oral clearance of propranolol and the food-induced increase in its bioavailability. The food effect was not associated with decreased total availability, but with delayed appearance, of the oxidative metabolites 4-OHP, NLA and PG. Hence the food (protein) effect does not seem to be caused by enzyme inhibition, but rather it is due to reduced hepatic extraction of propranolol, probably consequent to an increase hepatic entry rate. When taken together with the protein-rich breakfast, propranolol usually appeared in systemic blood at least as early as when taken on an empty stomach, implying that gastric absorption of propranolol may be possible in the presence of protein-rich food. Within the individual the food effect was reproducible, but its magnitude showed an intraindividual variation that may reflect its dependence upon the rates of gastrointestinal absorption and splanchnic-hepatic blood flow, and hence upon the rate of hepatic drug entry.