The efficacy of fluoxetine was evaluated in 32 patients with major depression in double-blind amitriptyline-controlled clinical trials. Patients were randomly assigned to 6 weeks of treatment with 20mg/day of fluoxetine and 25-100mg/day of amitriptyline. We used the Hamilton rating scale for depression(HAM-D) and the Clinical Global Improvement(CGI) to evaluate the improvement of depression. In addition, we also used the Covi Anxiety Scale and the Anxiety/Somatization subscale of HAM-D to investigate the relieving effect of anxiety. The improvement by fluoxetine in mean total score of HAM-D and CGI was comparable to amitriptyline. Fluoxetine also reduced anxiety significantly, but there was no difference between fluoxetine and amitriptyline in relieving anxiety symptoms. Fluoxetine showed considerably less adverse effects, especially anticholinergic effect and weight gain than amitriptyline. In conclusion, 20mg/day of fluoxetine was sufficiently effective in the treatment of depressive patients and was better tolerated than amitriptyline.