In order to further examine the mechanism of the increase in the plasma propranolol concentration versus time curve (AUC) caused by ingestion of propranolol with food, we administered R, S-propranolol tablets (0.5 mg kg-l) orally to healthy human volunteers and dogs in the presence and absence of sensory exposure to food without ingestion (teasing). Six healthy human volunteers were fasted on one occasion and on the other they were presented with an appetising meal, without eating it (teasing protocol). There was a strong trend to a greater propranolol AUC in the teasing protocol (139f54mgrnL-'h-' fasting, 178* 105mgmL-'h-l teasing; p=O.l), and time of peak concentration (tma) was significantly prolonged (80 f 22 min and 120 f 32 min, respectively; p < 0.03). Further studies were carried out in dogs who received R-propranolol(2 mg kg-') as an oral solution by gavage tube on four different occasions: fasting, following intragastric administration of a high-value liquid meal, following teasing with food in the animal house at normal feeding time (high-intensity teasing), and following teasing with food at a time and place not associated with feeding (low-intensity teasing). There were no significant differences in pharmacokinetic parameters between the fasting and intragastric food protocols. Low-intensity teasing resulted in significantly lower AUC and peak concentration (Cmm) compared with fasting (p<O.O5), confirming food effect patterns known to occur in dogs. Highintensity teasing resulted in significantly greater AUC and C,, compared with fasting (p < 0.05), reproducing in dogs the increase in propranolol AUC known to occur with food ingestion in humans. These findings suggest that the mechanism of the 'food effect' may involve physiological responses to the sight and smell of food additional to mechanisms activated by ingestion.