Postoperative recovery from muscle weakness was investigated by handgrip strength after general anaesthesia using short- and long-acting muscle relaxants. Handgrip strength was measured on four occasions, preoperatively, before and after premedication, and postoperatively, both as soon as the patient could sustain a five-second headlift and one hour later. Twenty patients received either atracurium 0.6 mg/kg or pancuronium 0.1 mg/kg as part of a general anaesthetic technique with neostigmine reversal. Compared with their initial preoperative hand strengths, the two relaxant groups showed only an insignificant decrease in grip strength to 92% after premedication. Postoperatively when a five-second headlift was achieved, the patients in both groups were weak. The group given atracurium had a mean of 60% of their initial grip strength and those given pancuronium had a mean of 44%. This difference in strength was statistically significant from the initial level and from each other. However, there was no difference between the groups in the time interval from administering reversal to achieving a sustained headlift. One hour later in both groups grip strength was much better, within 80% of that seen after premedication in most patients, and no significant difference in grip strength was seen between the groups. This study demonstrates that patients who receive relaxant anaesthesia have substantial muscle weakness in the recovery phase. When neostigmine reversal is used, pancuronium appears to have a similar return of voluntary muscle power to that seen with atracurium. The marked difference in elimination kinetics of these two relaxants seems to be of little importance when neostigmine reversal is used.