Shortening of the cardiac action potential during ischemia and anoxia is likely to contribute to the decline in contractility that occurs under such conditions. It has been hypothesized that a decrease in the intracellular ATP concentration ([ATP]1) underlies the changes in the action potential. The recently discovered potassium channel activated at low ATP concentrations might provide the link between action potential shortening and low [ATP]1. However, it has yet to be shown that [ATP]1falls to the range required for channel activation at the time when action potential shortening occurs. We have measured action potentials and [ATP]1simultaneously in isolated ferret hearts during ininbition of both oxidative phosphorylation and anaerobic glycolysis (metabolic blockade). Metabolic blockade caused a rapid decline in cardiac contractility, accompanied by a rapid fall in action potential duration. [ATP]1fell only slightly and remained well above the range where activation of the ATP-sensitive K+channel would be expected to occur. Moreover, rentroduction of glucose to the perfusate led to a substantial recovery in both contraction and in action potential duration, again in the absence of any great change in [ATP]1. These results suggest that the action potential shortening observed in metabolic blockade cannot be explained by the simple hypothesis of K+channel opening as a consequence of a decrease in bulk [ATP]1unless the K±for suppression of channel activity by ATP is very much ingher in intact cells than in any of the patch configurations studied. An alternative explanation is that the channel may be regulated under these conditions by mechanisms other than a change in [ATP]1.