Knowing the resting free calcium is important in understanding the role of calcium as an intracellular second messenger. We used a bracketing (null) technique with a luminescent calcium indicator, aequorin, microinjection and image-intensification to measure free calcium in single muscle fibres from the barnacle, Balanus nubilus. We injected modified aequorins (recombinant, and hch-) which after a 30 min diffusion gave reasonable resting glows. Subsequent injection of calcium (strongly buffered with either EGTA or BAPTA, 10 mM) increased or decreased the resting glow depending on the free calcium level in the injected buffer solution. This bracketing (null) method is inherently accurate, but mechanical artifacts on calcium injection reduce the accuracy when total light emission is measured. We therefore used image-intensified video-microscopy of the injected region and video processing (Image-1) of artifact-free regions, to greatly improve the consistency. The luminescence in a pre-selected region of the muscle fibre was measured as a function of time during the injection. Solution calciums were chosen so that if the first injection decreased the resting glow, the second increased it, or vice versa, thus bracketing the true resting value. We used two methods to determine the true value bracketed by our injections: (1) a linear interpolation using the fractional changes in luminescence or (2) a power law interpolation assuming a 2.2 or 2.5 power relationship between luminescence and free calcium. Using these methods, we estimated the free calcium level in the lateral depressor fibres of freshly dredged barnacles to be 279 +/- 36 nM (+/- SD), 339 +/- 42 nM, or 352 +/- 45 nM for the linear, 2.2 and 2.5 powers respectively under the conditions of hch-aequorin and BAPTA buffers (using a K'Ca for BAPTA of 3.0 x 10(6) M-1 for our conditions). Recombinant-aequorin gave essentially the same result while EGTA buffers yielded a somewhat higher value but because of influences of pH on the K'Ca for EGTA (taken as 6.7 x 10(6) M-1 for our conditions) was considered less reliable. Minor changes in [Mg2+] upon buffer injection can lead to underestimates of the true resting [Ca2+] by at most 10%. Thus, we estimate the resting free calcium in barnacle muscle fibres to be 300-380 nM.