A hypometabolic response during acute exposure to hypoxia has been measured in both endothermic and ectothermic vertebrates. In the turtle, we determined the metabolic response to normocapnic hypoxia and hypercapnic hypoxia. In addition, we tested the hypothesis that hypoxic hypometabolism was a regulated response that did not depend on O(2) availability. Metabolic, cardiovascular, and blood gas measurements were collected in anesthetized turtles under two conditions: during normocapnic hypoxia [fractional inspired O(2) FI(O(2)) = 0.1 and 0.05] and during hypercapnic hypoxia [FI(O(2)) = 0.1 and 0.05 plus fractional inspired CO(2) (FI(CO(2))) = 0.05]. During normoxia, rate of O(2) consumption (VO(2)) was 0.82 ml. min(-1). kg(-1) and was reduced by nearly 30% at the lowest FI(O(2)). Normocapnic hypoxia of FI(O(2)) = 0.1 had no significant effect on VO(2). The addition of 5% CO(2) to the inspired air did not enhance the effects of hypoxia. Injections of 2,4-dinitrophenol increased VO(2) during hypercapnic hypoxia in some animals to levels greater than those measured during normoxia. We conclude that hypoxia produces a hypometabolic state in anesthetized turtles, and the pharmacological stimulation of VO(2) counteracts the effects of hypoxia on metabolism. The hypoxic hypometabolism in turtles most likely represents a regulated response and does not reflect limited O(2) availability at the cellular level. Finally, we hypothesize that hypoxemia induced by the right-to-left cardiac shunt often associated with diving may trigger the development of a hypometabolic state and therefore contribute to the prolongation of aerobic dive times.