We have compared the effects of the sarcoplasmic reticulum (SR) Ca2+ release inhibitor, ruthenium red (RR), on single ryanodine receptor (RyR) channels in lipid bilayers, and on Ca2+ sparks in permeabilized rat ventricular myocytes. Ruthenium red at 5 mu M inhibited the open probability (P-o) of RyRs similar to 20-50-fold, without significantly affecting the conductance or mean open time of the channel. At the same concentration, RR inhibited the frequency of Ca2+ sparks in permeabilized myocytes by similar to 10-fold, and reduced the amplitude of large amplitude events (with most probable localization on the line scan) by similar to 3-fold. According to our theoretical simulations, performed with a numerical model of Ca2+ spark formation, this reduction in Ca2+ spark amplitude corresponds to an similar to 4-fold decrease in Ca2+ release flux underlying Ca2+ sparks. Ruthenium red (5 mu M) increased the SR Ca2+ content by similar to 2-fold (from 151 to 312 mu mol/l cytosol). Considering the degree of inhibition of local Ca2+ release events, the increase in SR Ca2+ load by RR, and the lack of effects of RR on single RyR open time and conductance, we have estimated that Ca2+ sparks under normal conditions are generated by openings of at least 10 single RyRs.