To assess the cardiovascular actions of magnesium in neurally intact animals, magnesium chloride (1–4 mM/min) administered i.v., producing a peak arterial magnesium level between 4.7 and 7.2 mg/dl, was given to α-chloralose-anesthetized, open-chest dogs. Magnesium lowered heart rate by 36 ± 11 beats/min (P < .05), cardiac output by 0.7 ± 0.2 liters/min (P < .05), left ventricular (LV) peak dP/dt by 410 ± 96 mm Hg/sec (P < .05) and aortic and pulmonary artery pressures, but it did not change LV end-diastolic pressure, systemic resistance or pulmonary resistance. Coronary blood flow also decreased by 39 ± 11% (P < .05), myocardial oxygen consumption by 88 ± 22% (P < .05) and myocardial oxygen extraction by 53 ± 16% (P < .05). When heart rate was held constant, magnesium still decreased LV systolic pressure, LV peak dP/dt and coronary blood flow. The increase in serum magnesium was accompanied by an increase in serum calcium (by 1.4 ± 0.2 mg/dl; P < .05) and a fall in serum potassium (by 0.21 ± 0.1 mEq/dl; P < .05), but not by a change in serum sodium, myocardial electrolyte arteriovenous differences or arterial pH. Thus, at blood concentrations that are observed in humans after therapeutic dosages of magnesium, a depression of cardiac performance is observed int he anesthetized dog. Although magnesium produces a fall in coronary blood flow, this appears to be due at least in part to a decrease in myocardial oxygen requirements because myocardial oxygen extraction also decreases. Rapid changes in serum electrolytes accompany these hemodynamic effects.