Six healthy volunteers received 15 mg midazolam, 50 mg ephedrine, or placebo orally before a 50-minute aerobic treadmill exercise and in a control session. Plasma drug concentrations for pharmacokinetic calculations were estimated from samples drawn up to 24 hours after drug intake. Heart rate, blood pressure, critical flicker fusion test, Maddox wing test, and visual analog scales relating to mood and feelings of tiredness were included in the sessions as pharmacodynamic measures. These tests were made at 35, 55, and 75 minutes and at 2, 2 1/2, 3 1/2, and 5 hours after drug intake. Exercise impaired the absorption of midazolam and counteracted the midazolam-induced decrement in flicker fusion threshold. Whether the effect on flicker fusion was caused mainly by the pharmacokinetic changes or by a general alerting effect of exercise cannot be verified by this experiment. The kinetics of ephedrine was not affected by exercise, but exercise enhanced the tachycardic response to ephedrine and abolished its pressor effect.