The efficiency of oxytocic drugs in therapeutic abortion was studied in 113 women who were randomly assigned to either a control group (n = 44) receiving no oxytocic drugs, a group (n = 36) receiving 0.2 mg ergometrine by slow intravenous injection, or a group (n = 33) given 5 units oxytocin in a intravenous infusion throughout the procedure. The three groups were comparable with regard to the age of patients, the age of pregnancies and parity. The surgical procedure was the same in all three groups and carried out by the same surgeon. General anaesthesia was obtained with an infusion of 500 mg methohexitone and 500 micrograms fentanyl in 500 ml isotonic dextrose solution at a rate of 3 ml X min-1. The anaesthetic requirements were 2.94 +/- 0.80 micrograms X kg-1 fentanyl and 2.94 +/- 0,80 mg X kg-1 methohexitone. The criteria studied were the blood loss as measured by the volume aspired, the presence or absence of nausea and vomiting after the procedure, the age of pregnancy and the total amount of anaesthetic drugs given. No correlation was found between the amount of anaesthetic drugs given and the frequency of nausea and vomiting, and between the amount of blood lost (r = 0.287; ddl = 111; alpha = 17.322). The study did not, therefore, confirm the reputation of oxytocic drugs in reducing the bleeding. It seemed that, in therapeutic abortion, spontaneous uterine contraction was sufficient to control the bleeding. But a significant correlation was found between the amount of blood lost and the age of the pregnancy (r = 0.399; ddl = 111; alpha less than 1%).