1. Inactivation of Ca2+ current (I-Ca) induced by Ca2+ release from sarcoplasmic reticulum (SR) was studied in single rat ventricular myocytes using whole-cell patch-clamp and indo-1 fluorescence measurement techniques. 2. Depolarizing pulses to 0 mV elicited large Ca2+ transients and I-Ca with biexponential inactivation kinetics. Varying SR Ca2+ loading by a 20 s pulse of caffeine showed that the fast component of I-Ca inactivation was dependent on the magnitude of Ca2+ release. 3. Inactivation of I-Ca induced by Ca2+ release was quantified, independently of voltage and Ca2+ entry, using a function termed fractional inhibition of I-Ca (FICa). The voltage relation of FICa had a negative slope, resembling that of single-channel Ca2+ current (i(Ca)) rather than the bell-shaped current-voltage (I-V) relation of macroscopic I-Ca and Ca2+ transients. 4. Intracellular dialysis of myocytes with 10 mM EGTA (150 nM free [Ca2+]) had no effect on I-Ca inactivation induced by Ca2+ release, despite abolition of Ca2+ transients and cell contraction. Dialysis with 3 or 10 mM BAPTA (180 nM free [Ca2+]) attenuated FICa in a concentration-dependent manner, with greater inhibition at positive than at negative potentials, consistent With more effective buffering of Ca2+ microdomains of smaller i(Ca). 5. Spatial profiles of [Ca2+] near an opened Ca2+ channel were simulated. [Ca2+] reached submillimolar levels at the mouth of the channel, and dropped steeply as radial distance increased. At any given distance from the channel, [Ca2+] was higher at negative than at positive potentials. The radii of Ca2+ microdomains were significantly reduced by 3 or 10 mM BAPTA, but not by 10 mM EGTA. 6. In conclusion, the distinctive voltage dependence and susceptibility of Ca2+ release-induced I-Ca inactivation to fast and slow Ca2+ buffers suggests that the process is mediated through local changes of [Ca2+] in the vicinity of closely associated Ca2+ channels and ryanodine receptors.