Initial velocities of energy-dependent Ca + + uptake were measured by stopped-flow and dual-wavelength techniques in mitochondria isolated from hearts of rats, guinea pigs, squirrels, pigeons, and frogs. The rate of Ca + + uptake by rat heart mitochondria was 0.05 nmol/mg/s at 5 pM Ca + + and increased sigmoidally to 8 nmol/mg/s at 200 uM Ca + + . A Hill plot of the data yields a straight line with slope n of 2, indicating a cooperativity for Ca ++ transport in cardiac mitochondria. Comparable rates of Ca+ + uptake and sigmoidal plots were obtained with mitochondria from other mammalian hearts. On the other hand, the rates of Ca + + uptake by frog heart mitochondria were higher at any Ca + + concentrations. The half-maximal rate of Ca + + transport was observed at 30, 60, 72, 87, 92 MM Ca + + for cardiac mitochondria from frog, squirrel, pigeon, guinea pig, and rat, respectively. The sigmoidicity and the high apparent Km render mitochondrial Ca + + uptake slow below 10 puM. At these concentrations the rate of Ca ++ uptake by cardiac mitochondria in vitro and the amount of mitochondria present in the heart are not consistent with the amount of Ca + + to be sequestered in vivo during heart relaxation. Therefore, it appears that, at least in mammalian hearts, the energy-linked transport of Ca ++ by mitochondria is inadequate for regulating the beat-tobeat Ca + + cycle. The results obtained and the proposed cooperativity for mitochondrial Ca+ + uptake are discussed in terms of physiological regulation of intracellular Ca++ homeostasis in cardiac cells.