In a 7-week prospective multicentre study, the efficacy, tolerability and safety of moclobemide were compared to those of amitriptyline and placebo in parallel groups of out-patients (n = 173) fulfilling the DSM III-R criteria for a major depressive episode. Participants were required to have a minimum baseline total score of 18 on the 17-item Hamilton Depression Rating Scale (HAMD). After a 1-week placebo washout, patients were randomly allocated to the three treatment groups. Assessment of efficacy, as judged by the number of responders achieving a 50% reduction in HAMD score by the end of treatment, showed that both moclobemide and amitriptyline were significantly superior to placebo, but that they were not significantly different from each other. Both treatments differed significantly from placebo with respect to the Physician's Global Assessment of Efficacy ('very good' or 'good' response: moclobemide 57%, amitriptyline 60% and placebo 35%). Assessment of tolerance as judged by the spontaneous reporting of adverse events showed a significant superiority of moclobemide over amitriptyline, but there was no significant difference between moclobemide and placebo. At termination of the study, amitriptyline patients showed a significant elevation of heart rate both supine (10.8 beats/min) and standing (15.5 beats/min), as well as significant weight gain (1.7 kg), but no changes were seen in the moclobemide or placebo groups. In conclusion, both moclobemide and amitriptyline were found to be more effective than placebo in the treatment of depression, while moclobemide had fewer side effects.