To measure the actual percentage of intracellular free creatine participating in the process of energy transport, the incorporation of [1-14C]creatine into the "free" creatine and phosphocreatine (PCr) pools in spontaneously beating isolated rat atria, under various conditions, was examined. The atria were subjected to three consecutive periods, control, anoxia, and postanoxic recovery, in medium containing tracers of [1-14C]creatine. The tissue content and specific activity of creatine and PCr were determined at the end of each period. The higher specific activity found for tissue PCr (1.87 times) than creatine, independent of the percentage of total intracellular creatine that was present as free creatine, provides evidence for the existence of two separate pools of free creatine. Analysis of the data shows that in the normal oxygenated state approximately equal to 9% of the total intracellular creatine is actually free to participate in the process of energy transport (shuttle pool). About 36% of the total creatine is bound to unknown intracellular components and the rest exists as PCr. The creatine that was taken up and the creatine that was released from the breakdown of PCr have much greater access to the site of phosphorylation than the rest of the intracellular creatine. A sharp increase in the specific activity of residual PCr on prolongation of anoxic time was also observed. This provides evidence for a nonhomogeneous pool of PCr, for the most recently formed (radioactive) PCr appeared to be hydrolyzed last.