Atropine, 0.01 mg.kg-1, was given intravenously before the start of surgery to 169 patients who were anaesthetized with one of five different techniques; halothane, enflurane, cervical epidural, lumbar epidural or narcotic anaesthesia in addition to nitrous oxide and oxygen. Atropine produced a significant increase in heart rate (HR) within 1 min in all patients studied; the HR increases in patients anaesthetized with halothane (37 +/- 11 beats.min-1, n = 37) or narcotic (34 +/- 12 beats.min-1, n = 30) were significantly greater than in those anaesthetized with enflurane (25 +/- 10 beats.min-1, n = 35; P less than 0.01) or epidural anaesthesia. Because of the presence of an acute cardiac sympathectomy, the patients who received cervical epidural anaesthesia were expected to have different responses to the atropine. However, there was no significant difference in the HR increases between the patient groups with cervical (19 +/- 12 beats.min-1, n = 32) and lumbar (22 +/- 8 beats.min-1, n = 35) epidural anaesthesia. Atropine also produced a small but significant increase in arterial pressure in all five groups of patients. These results suggest that the cardiac responses to atropine may differ depending on the individual anaesthetic agent used, and are likely dependent upon the agent's effect on autonomic nervous system activity.