During normovolemia, nitrous oxide causes mild sympathetic stimulation and direct mycardial depression; these effects offset each other, resulting in only minimal cardiovascular changes. To test the hypothesis that during hypovolemia this balance would change and depression predominate, 10 swine were made hypovolemic (30% blood loss) and then were given 70% N2O (0.25 MAC in swine) or an equipotent concentration of halothane, an agent that does not cause sympathetic stimulation. The alternate anesthetic was given to the same hypovolemic swine on another day. Five minutes after induction of anesthesia during hypovolemia, both N2O and halothane caused significant, physiologically important deterioration of compensation for hemorrhage. Halothane decreased systemic vascular resistance (SVR); N2O was more variable in its action, and SVR did not decrease significantly. Both agents caused similar decreases in cardiac output, mean aortic blood pressure, stroke volume, oxygen consumption, and left ventricular minute work, despite increases in plasma epinephrine concentration and plasma renin activity. No differences were found between groups for any of these variables (P> 0.05). Plasma norepinephrine concentration increased only in the N2O group and was greater in that group than in the halothane group. The deterioration of cardiovascular compensation for hemorrhage was expressed metabolically by similar decreases in the two groups in partial pressure of oxygen of mixed venous blood and by increases in blood lactate concentration. Thirty minutes after induction of anesthesia, with stable end-tidal anesthetic concentrations, both groups had some cardiovascular, but no metabolic, recovery. Oxygen tension of mixed venous blood and blood lactate concentrations continued to be similar in the two groups. The authors conclude that administration of N2O in hypovolemic swine results in cardiovascular depression that is not different from that caused by halothane, an anesthetic having no sympathetic properties.