Absorption of the spasmolytic drug baclofen in three selected intestinal segments of living anaesthetized rats in situ, is shown to be a specialized transport mechanism obeying Michaelis-Menten kinetics. Equation parameters were calculated through different procedures, whose features are discussed. A computer method based on the integrated form of Michaelis-Menten equation which reproduces the entire time course of drug absorption from the data found in three intestinal perfusion series at different initial concentrations, yielded V, and K, values of 12.0 mg h-' and 8.0 mg, respectively, in the mean segment of the small intestine, a rather selective absorption site for baclofen. Lesser but comparable absorption rates were found in the proximal and distal segments of the small intestine, whereas in colon, drug absorption was negligible. Baclofen transport was significantly reduced in the presence of the enzymatic inhibitor sodium azide. If these results were extrapolated to humans, they would explain the excellent bioavailability profiles reported for baclofen at normal doses in spite of its physicochemical properties, which do not favour passive diffusion. Based on the same principle, the administration of usual doses at shorter time intervals could be recommended, instead of high, when higher plasma levels at steady-state are needed. On the other hand, more than 8-h sustainedrelease preparations of baclofen should, probably, be avoided.