1. We have investigated whether Ca2+ entry through T-type Ca2+ channels participates in triggering Ca2+ release from the sarcoplasmic reticulum (SR) in single guinea-pig ventricular myocytes (whole-cell voltage clamp, K5fura-2 as [Ca2+]i indicator; all monovalent cations replaced by impermeant ions to record uncontaminated Ca2+ currents; T = 23 or 36 degrees C). 2. T-type Ca2+ currents were elicited from a holding potential of -90 mV during steps to -50 to -20 mV. For steps to -50 mV, very small [Ca2+]i transients could be recorded with high loading of the SR (peak Delta[Ca2+]i, 67 +/- 41 nM; n = 9). 3. For steps to -40, -30 and -20 mV, we compared the amplitude of Ca2+ release for a holding potential of -50 mV with L-type Ca2+ current only to Ca2+ release for a holding potential of -90 mV with both T- and L-type Ca2+ current. Significantly more Ca2+ release was observed with T-type current present, and both the T-type current and the additional Ca2+ release were suppressed by 50 microM NiCl2. 4. Ca2+ influx through T-type Ca2+ channels triggered less Ca2+ release than a comparable Ca2+ influx through L-type Ca2+ channels. 5. Rapid block of T-type Ca2+ current during the action potential (50 microM NiCl2 during steady-state stimulation at 1 or 2 Hz) did not immediately reduce Ca2+ release, although a small decrease was observed after longer application. 6. We conclude that T-type Ca2+ current can trigger Ca2+ release from the SR albeit less efficiently than L-type Ca2+ current. T-type current is most likely to provide only a small contribution to the trigger for Ca2+ release in normal conditions. These results support the hypothesis that L-type Ca2+ channels have a privileged role in excitation-contraction coupling.