From the study of movements of $Ca^{++}$ in frog cardiac muscle, Niedergerke (1963) postulated that $Ca^{++}$ necessary for the cardiac contraction is stored in a specific pool. Langer et al (1967) and DeCaro (1967) also found a close relationship between the change of $Ca^{++}$ flux kinetics and the change of contractile force. According to the studies of several investigators, Ca II (Bailey and Dressel 1968) or phase I and II (Langer 1965, Langer et al 1967, 1971) in the $Ca^{++}$ washout curve was associated with cardiac contractility. This investigation was aimed to elucidate the anatomical region of the contractile active $Ca^{++}$ pool. At the same time, it was assumed in this study that $Ca^{++}$ in the sarcoplasmic reticulumn represents one of the major intracellular $Ca^{++}$ pool and cardiac contractility was also dependent on the intracellular $Ca^{++}$ concentration. Consequently, this experiment was performed at different temperatures to activate to activate inhibit the deactivating process of activated $Ca^{++}$ in the intracellular space to see if changes in the contractility decay curve existed at different temperatures. The isolated hearts of rabbits and turtles (Amyda maackii) were attached to the perfusion apparatus according to the method employed by Bailey and Dressel (1968). The isolated hearts were initally perfused with a full Ringer solution containing 2 mg/ml of inulin for 1 hr, and then $Ca^{++}$ and inulin-free Ringer solution was perfused while the isometric tension was recorded and a serial sample of perfusion fluid dripping from the cardiac apex was collected for 10 sec throughout experimental period. The above procedure was performed at $23^{\circ}C$, $30^{\circ}C$ and $38^{\circ}C$ on the rabbit heart and $10{\sim}13^{\circ}C$, $10^{\circ}C$, $25^{\circ}C$, $30^{\circ}C$ and $35^{\circ}C$ on the turtle heart. After determination of $Ca^{++}$ and inulin concentration of the samples, the $Ca^{++}$, inulin washout curve and the contractile tensin decay curve were analysed according to the method of Riggs (1963). The results were summarized as follows; 1. In the rabbit heart, there are 2 inulin compartments, 3 $Ca^{++}$ compartments and sing1e exponential decay of contractile tension. In the turtle heart, there are $1{\sim}2$ inulin compartments, $1{\sim}2$ $Ca^{++}$ compartments and $1{\sim}2$ phases of contractile tension decay. The fact that the inulin space was divided into 3 compartments in the washout curve in these hearts indicates the presence of heterogeneity in cardiac perfusion, i.e., overfused and underperfused area. 2. Ca I a9d Ca II in these hearts were found to have $Ca^{++}$ in the ECF compartments because their half times in the washout curves were far smaller than those of the inulin washout curves in the rabbit heart and similar to those of the inulin washout curves in the turtle heart. Ca III in the rabbit heart may have originated from the intracellular $Ca^{++}$ store. But no Ca III in the turtle heart was found. This may be due to the fact that the iutracellular $Ca^{++}$ pool in the turtle heart was too small to detect using this experimental procedure since sarcoplasmic reticulumn in the turtle heart is poorly developed. 3. In the rabbit heart, there were no chages in the half time of Ca I, Ca II, inulin I and inulin II at different temperatures, but the half time of Ca III was significantly prolonged at lower temperatures, and the half time of the contractile tension decay tended to be prolonged at lower temperatures but this was not significant. In the turtle heart, there were no changes in the half time of Ca I, Ca II, inulin 1, inulin II and phase I of the contractile tension decay at different temperatures, but the half time of phase II of the contractile tension decay was significantly prolonged at lower temperatures. This finding indicates that intracellu!ar $Ca^{++}$ in these hearts was also responsible particulary for maintaining the cardiac contractility at the lower temperatures. 4. The half times of contractile tension decay were shorter than those of Ca II in the $Ca^{++}$ washout curves in both animal hearts. According to the above results it was shown that $Ca^{++}$ in ECF is primarily and $Ca^{++}$ in the intracellular space is partially associated with the cardic contractility.