Sixty-eight depressed out-patients were allocated to treatment with either oral amitriptyline (75-225 mg/day) or intramuscular flupenthixol decanoate (10-30 mg every 14 days) in flexible dosage for 12 weeks under double-blind procedures. Various observer- and self-rating scales were applied before and after 2, 4, 8 and 12 weeks of treatment. Twenty-four patients completed the course of amitriptyline and 20 the course of flupenthixol. All variables improved over time, but there were no significant differences between the two drugs. The Newcastle scores pre-treatment were not related to drug response suggesting that both drugs were similarly effective across a wide spectrum of depressive disorders. Patients on amitriptyline tended to complain of dry mouth; those on flupenthixol had a higher incidence of extrapyramidal signs, the majority receiving anti-parkinsonian drugs at some time during the treatment. Flupenthixol decanoate in low dose is a useful anti-depressant, but should be restricted to short courses of treatment, to patients refractory to other treatments, and to patients suspected of poor compliance.