The clinical significance of intramuscular premedication with 0.01 mg/kg of atropine in a procedure involving oral benzodiazepine premedication (15 mg midazolam the evening before surgery and on the morning of surgery) was investigated in a double-blind study. As far as sedation, apprehension, excitement, dizziness, emesis, and headache were concerned, there were no significant differences between group 1 (atropine) and group 2 (placebo) patients; however, both during and after anesthesia patients in group 1 had less excessive salivary secretion (especially during extubation). As a result of sympathetic overactivity, patients in group 1 had an increased heart rate and an increased incidence of supraventricular tachycardia. In group 1 intravenous infusion proved more difficult, and in addition, the patients complained more of subjective side effects (dry mouth). There was no significant correlation between the radioimmunologically measured serum concentrations and the clinical effects of atropine measured just before the induction of anesthesia. Substantial interindividual differences were found in these serum levels. From the anesthetist's viewpoint, atropine has both beneficial effects (antisecretory) and unwanted effects (cardiovascular effects). For the patient atropine caused only subjective unwanted effects. Midazolam, a new short-acting, sedative benzodiazepine derivatives, can be used without atropine as an oral premedicant.