The level of amitriptyline (AT) and its metabolites was investigated. Three separate experiments were carried out. In 2 of these experiments, rats were treated over 7 days with i.p. doses of AT (10 mg/kg in experiment A and 2 .times. 20 mg/kg in experiment C). The rats were sacrificed 2 (experiment C) or 12 h (experiments A and C) after the last dose. In experiment B, rats were sacrificed 2 or 12 h after a single dose of 20 mg/kg AT. In experiment A only AT was measurable in the brain and in the plasma, in contrast to experiments B and C, where NT and the hydroxylated metabolites AT-OH and NT-OH reached significant levels in the plasma and in the brain. The concentrations of AT-OH, NT-OH and NT (12-h values) that were found in the brain are probably not pharmacologically relevant. The 12-h plasma values of all compounds tested were, even with the highest dose, lower than those expected to be clinically effective in man. Apparently, AT, at higher doses, may induce its own metabolism. The free plasma levels of this drug and its metabolites are higher in man than in the rat. The possible implications of these results in the use of antidepressants in the treatment of depression are discussed.