Reports on the duration of action of atracurium (Atr) and vecuronium (Vec) in patients with renal failure are contradictory. It is either stated that the duration is equal for both relaxants [1, 8], or that Atr acts for a longer duration [9]. Because of these conflicting results. we measured the times for latency (tL), onset (tA), duration of action (tW), and recovery (tE) for both agents. Methods. Fourty patients with end-stage renal failure on chronic haemodialysis were randomly assigned to receive either Atr (0.4 mg.kg-1 BW) or Vec (0.08 mg.kg-1 BW). After induction with thiopentone and 0.1 mg fentanyl, anaesthesia was maintained with nitrous oxide and 1 vol.% ethrane and expiratory CO2 partial pressure was kept between 4.6 and 4.9 kPa. If the twitch height of T1 regained 25% of the pre-relaxation value, 20% of the initial relaxant bolus was injected. Relaxation was monitored with a relaxograph [4] after calibration of the device. After testing for a normal distribution, statistical analysis was done by Student's t-test. A value of P less-than-or-equal-to 0.05 was chosen for statistical significance. Results. There were no significant differences regarding tL (Atr: 1.0+/-0.5 min; Vec: 1.1+/-0.5 min) and tA (Atr: 5.5+/-2.1 min; Vec 4.1+/-2.4 min); tW (Atr: 21.3+/-13.1 min; Vec 31.7+/-15.6 min) and tE (Atr: 19.0+/-9.0 min: Vec 30.1+/-19.0) were significantly different. Discussion. Our results are not in accordance with those authors who found in comparison with Atr an equal [1, 8] or shorter [9] duration of action for Vec in patients with renal failure. If the duration of action is equal [5] in subjects with no renal insufficiency, our measurements are in accordance with kinetic evaluations showing the same clearance and half-life for Atr [12, 13] in patients with and without renal insufficiency, but 40% diminished clearance and 60% prolonged half-life for Vec in renal insufficiency [10].