An experimental prolonged-release dosage form of aspirin was used as a biopharmaceutical probe to study dissolution rate limited intestinal absorption in man. This dosage form released drug <i>in vitro</i> by an exponential process which was highly sensitive to changes in agitation intensity, but relatively insensitive to pH and pancreatin. Plasma salicylate levels and urinary excretion of salicylate and metabolites were determined in humans after oral administration of 1 Gm. of aspirin in rapidly absorbable form (compressed tablets) and in prolonged-release form. Absorption from the prolonged-release dosage form was determined on two separate occasions in each subject to assess both inter- and intrasubject variations in absorption kinetics. The individual absorption data were found to be describable in most instances as first-order processes preceded by a lag time. In some cases, an initial slow absorption phase preceding the more rapid phase could be noted. It was found also that the averaged data yield a zero-order plot (which is indicative of absorption at a constant rate). This shows that interpretations based on averaged data of the type discussed may lead to erroneous conclusions. The experimental results suggest that the intersubject variation in drug absorption rates from the experimental prolonged-release dosage form studied was due mainly to individual differences in gastric emptying rate and intestinal peristaltic activity and to the pronounced sensitivity (with respect to dissolution rate) of the dosage form to variations in agitation intensity. Intrasubject variation of drug absorption rates appeared to be due primarily to variations in gastric emptying rates.