Transient changes in intracellular free calcium concentration (delta [Ca2+]) in response to pulse depolarizations were monitored in isolated segments of single frog skeletal muscle fibres cut at both ends and voltage clamped at a holding potential of -90 mV in a double-Vaseline-gap chamber. Calcium transients were monitored optically using the metallochromic indicator dye Antipyrylazo III (APIII), which entered the fibre by diffusion from the solution applied to the cut ends. Optical artifacts due to fibre movement were minimized or eliminated by stretching the fibres to sarcomere lengths at which there was little or no overlap of thick and thin contractile filaments. Remaining movement-independent optical changes intrinsic to the fibre and unrelated to the dye were monitored at 850 nm, where free and dye-bound APIII have no absorbance. These 850 nm signals scaled by lambda -1.2 were used to remove intrinsic components from the signals at 700 or 720 nm, wave-lengths at which the APIII absorbance increases when calcium is bound. The corrected 700 or 720 nm signals were used to calculate delta [Ca2+]. The decay of delta [Ca2+] following fibre repolarization at the termination of a depolarizing pulse was well described by a single exponential plus a constant. The exponential rate constant for the decay of delta [Ca2+] decreased and the final 'steady' level that delta [Ca2+] appeared to be approaching increased with increasing amplitude and/or duration of the depolarizing pulse. Both the decreasing decay rate and the build up of the 'steady' level can be accounted for using a two-component model for the removal of free calcium from the myoplasm. One component consists of a set number of a single type of saturable calcium binding site in the myoplasm. The second component is a non-saturable, first-order uptake mechanism operating in parallel with the saturable binding sites. The removal model parameter values were adjusted to fit simultaneously the decay of delta [Ca2+] after pulses of various amplitudes and durations in a given fibre. The basic procedure was to track delta [Ca2+] during each pulse when an undetermined calcium release was occurring, but to calculate the decay of delta [Ca2+] starting 14 ms after repolarization when release was assumed to be negligible. After appropriate selection of parameter values, the model reproduced most aspects of the decay of delta [Ca2+].(ABSTRACT TRUNCATED AT 400 WORDS)