The effects of low (1.6 per cent end-tidal) and high (3.2 per cent end-tidal) concentrations of isoflurane were compared in a closed-chest dog preparation. Hemodynamics were evaluated using cardiac catheterization for measuring mean right atrial (RAP) and aortic (MAP) pressures; left ventricular end-diastolic (LVED) and peak systolic (LVSP) pressures and maximum rate of pressure rise (LVdP/dt); and the cardiac output (CO) by the dye dilution technique. Myocardial blood flow (MBF) and metabolism were estimated by 133Xe washout from the great cardiac vein and measurement of oxygen and various substrates in aortic and coronary venous blood. Ventricular functional responses to altering preload (LVEDP) by blood infusion or withdrawal, and afterload (systemic vascular resistance, SVR) by balloon occlusion of the aorta were measured at low and high isoflurane concentrations. Body temperature, Pao2 and Pao2 were maintained constant. High concentrations of isoflurane (as compared to low concentrations) produced 40–60 per cent decreases in MAP, CO, LVSV, LVdP/dt, and SVR without changing heart rate or LVEDP. MBF and myocardial O2 uptake were decreased to the same extent without change in O2 or lactate extraction by the heart. Increased preload resulted in small increases in cardiac output at low LVEDP (13–15 torr), but little or no change at higher values (18–23 torr) during both high and low isoflurane concentrations. Likewise, increased afterload decreased ventricular function at both concentrations of isoflurane. When the effects of halothane and enflurane were compared in identical, acute and chronic preparations, the results were similar except the filling pressures (LVEDP) were significantly increased by high concentrations (more than two times MAC) of the other two anesthetics. In addition, there was no change in systemic vascular resistance produced by halothane or enflurane. Thus, although isoflurane produces a dose-related depression of ventricular function in the intact dog, the degree appears to be somewhat less than that seen with halothane and enflurane. This difference may be related to the decreased afterload resulting from the administration of isoflurane and perhaps a minor degree of cardiac sympathetic stimulation. Myocardial perfusion was decreased to the same degree as function and oxygenation was well maintained.