The fate of carbamazepine after intravenous injection in rats (n = 24) and the influence of activated charcoal on the kinetics was investigated. After randomization to four groups (n = 6, each), plasma concentration and the quantities of carbamazepine and metabolites excreted into bile, urine and intestine were determined using an in situ perfusion model of the small intestine (Ringer's solution) with or without orally administered activated charcoal (AC+; AC-) and with or without bile duct cannulation (BD+; BD-). The cumulative amount of carbamazepine and metabolites exsorbed into the small intestine within 3.5 hr after intravenous injection was about 15% in BD- animals and about 3% in BD+ animals. About 20% of the dose was detected in the externalized bile. Activated charcoal did not influence the amount exsorbed into the small intestine. Terminal half-life in plasma ranged from 159 min. to 194 min. within the four treatment groups without statistical significant difference (P = 0.751). Correspondingly, the area under the curve did not vary significantly and ranged between 1.13 and 1.41 g/min./l (P = 0.378). Excretion of carbamazepine and metabolites into urine varied between 3% and 6% of dose within all groups and showed close correlation with diuresis. In an identical experimental approach using a 2-fold intestinal perfusion rate (50 ml/hr; n = 8), no fundamental changes compared to the main experiment regarding pharmacokinetics of carbamazepine were observed. The lack of effect of activated charcoal on the elimination of carbamazepine and metabolites must be contributed to the small amount of the drug being exsorbed into the intestine and may be further influenced by reduced intestinal permeability of carbamazepine and metabolites or inadequate luminal stirring.