A double-blind clinical trial was carried out involving 26 hospitalized, depressed patients who were randomized into two groups. Patients in each group received chlorimipramine (50 mg/day), combined with L-5-HTP (300 mg/day) in Group A, and with placebo in Group B. The trial lasted 28 days. Two Group B patients dropped out. All the patients were evaluated by HRSD each week, and by ZDSI and CGI at the beginning and end of treatment. Several statistical analyses of the mean HRSD scores of each item presenting initial positive response and of ZDSI scores for both reactive and endogenous depression were performed, using Student's t-test. To evaluate the efficacy of the chlorimipramine +L-5-HTP combination versus control therapy the Mann-Whitney test was applied to the reduction in HRSD scores, showing 0.05 significance. The results for both types of pathology were quantitatively and qualitatively more positive for Group A than for Group B. Cluster analysis of HRSD item scores alone and between this scale and ZDSI, according to Carrol's correlation method for the symptoms of the cluster (mood, anxiety and somatic symptoms), was carried out to confirm the above results.