1. Single twitch fibres were isolated from frog muscle, then mounted in a chamber which was positioned on an optical bench. The fibres were immobilized by high stretch (sarcomere spacing 3.9-4.3 mum) and by placement on a pedestal. Their optical properties were determined by illuminating a 35-65 mum diameter spot with quasimonochromatic light of intensity I(0) and measuring the intensity I of the transmitted light. Since the main purpose of the experiments was to draw inferences from the absorbance spectra of different indicator dyes injected into the fibres, all results were expressed in terms of absorbance A calculated from the equation [Formula: see text]. Changes in absorbance DeltaA were calculated from the differential form of the equation [Formula: see text].2. The absorbance of a normal, non-injected fibre was, on average, equal to 0.03 at 570 nm and varied approximately inversely with wavelength between 450 and 750 nm.3. The earliest change in absorbance following an action potential was a small, transient increase which was followed by a larger decrease. The decrease in fibre absorbance varied from 0.5 x 10(-4) to 3 x 10(-4) units.4. Resting myoplasmic pH was determined by comparing the absorbance spectrum from fibres injected with Phenol Red with that obtained from calibrating solutions in cuvettes. The muscle measurements were corrected for the intrinsic absorbance of fibre without dye. The average value of pH in two fibres was 6.9. The change in absorbance following an action potential in these highly stretched fibres was small. In one experiment the change, if due to pH alone, corresponded to an increase in pH of 0.004 peak and 0.002 maintained (relative to a resting level of 6.9). The maintained signal can be satisfactorily explained by the known amount of phosphocreatine hydrolysis.5. Estimates of myoplasmic free [Mg(2+)] were made using three metallochromic indicator dyes. A different estimate was obtained with each dye as indicated below. Since these dyes are sensitive to pH, as well as [Mg(2+)], the estimate depends on the assumed value of intracellular pH. [List: see text] This variability probably means that at least two, and possibly all three dyes behave differently inside muscle fibres than they do in calibrating solutions. The most likely explanation is that dye, once injected, can bind to cellular contents and that this alters its properties.6. Changes in absorbance of fibres injected with Arsenazo I, a dye three times more sensitive to Mg(2+) than to Ca(2+), were used to determine whether changes in free [Mg(2+)] occur following an action potential. The observed changes were small and could be due to a small increase in pH, of the magnitude measured with Phenol Red, and/or free [Mg(2+)]. In terms of a change in free [Mg(2+)], the results set an upper limit of 2%.7. The conclusion from the action potential experiments is that neither intracellular pH nor free [Mg(2+)] changes appreciably in highly stretched fibres. Changes in these two quantities can therefore be neglected in analysing the relatively large 650-660 nm Ca(2+) signal in fibres injected with the Ca(2+) (but also pH and Mg(2+)) sensitive indicator dye Arsenazo III.