1. We studied the transient changes in myoplasmic Ca2+ concentration under current- and voltage-clamp (double Vaseline-gap technique) in cut fibres of rat extensor digitorum longus muscle using mag-fura-2 (furaptra) as Ca2+ indicator, at 3.6-3.8 microns sarcomere length and 17 degrees C. Mag-fura-5 and fura-2 were also used in order to characterize some aspects of the Ca2+ transients. 2. The peak [Ca2+] in response to a single action potential was 4.6 +/- 0.4 microM (n = 5). The time to peak of the Ca2+ transient was 4.6 +/- 0.42 ms, with half-width of 8.2 +/- 1.5 ms, time constant of the rising phase 1.15 +/- 0.25 ms, time constant of the decaying phase 3.26 +/- 0.65 ms, and delay between action potential and Ca2+ transient 2.0 +/- 0.2 ms. 3. Ca2+ transients were studied under voltage-clamp conditions at different voltages and pulse durations. The rising phase showed a complex temporal course with a fast initial increase and a second component. Both components were separated by a plateau or a brief decrease of the Ca2+ concentration. The peak Ca2+ transient was 10.5 +/- 1.3 microM (n = 22). 4. After interrupting the pulse, Ca2+ concentration decayed exponentially. The time constant of decay of the Ca2+ transient increased with the pulse voltage and duration, reaching a maximum value at potentials more positive than +10 mV and pulses longer than 200 ms. An analysis of the decaying phases of the Ca2+ transients suggests that only the removal process operates after fibre repolarization. 5. The rate of Ca2+ release from the sarcoplasmic reticulum was calculated using the Melzer, Ríos & Schneider model. The value of 17.2 +/- 3.1 micronM ms-1 (n = 10) estimated in these calculations was intermediate between those obtained by other authors from cut frog muscles (10 microM ms-1) and intact frog fibres (100 microM ms-1) using antipyrylazo III (AP III) as the Ca2+ indicator.