Previous clinical studies establishing the efficacy of midazolam maleate (RO 21–3981), a new water-soluble benzodiazepine for induction of anesthesia, have not critically evaluated the effects of this agent on the cardiovascular system. The present study compares the cardiovascular effects of midazolam maleate and diazepam in conscious dogs. Systemic arterial, pulmonary arterial and central venous pressures, cardiac output, LVmux dP/dt, heart rate and regional coronary blood flow were measured 3 min following intravenous administration of diazepam (0.5, 1.0, and 2.5 mg/kg) or midazolam maleate (0.25, 1.0, and 10.0 ing/kg). Midazolam maleate increased heart rate 10–20 per cent with all three doses and decreased mean arterial blood pressure approximately 10–20 per cent at 1.0 and 10 mg/kg. Cardiac output was increased 10–12 per cent with all three doses of midazolam maleate, and LVmax dP/dt was decreased 13–16 per cent at the two higher doses. Diazepam at all three doses did not alter heart rate or mean arterial blood pressure. Diazepam, 1.0 and 2.5 mg/kg, produced significant (17 per cent) decreases in LVmax dP/dt, and 2.5 mg/kg produced a significant (10 per cent) increase in cardiac output. Neither drug in any dosage altered regional coronary blood flow, systemic or coronary vascular resistance, stroke volume, or stroke work. Maximum alterations in cardiovascular variables occurred with doses of midazolam maleate that are 10–15 times the recommended clinical induction dosage. It is concluded that in concentrations necessary for induction of anesthesia midazolam maleate has minimal effects on cardiovascular function.