The fluorescence decay of the plasma membrane calmodulin-activated Ca2+-ATPase from the erythrocyte was measured for the first time. The availability of a novel procedure for on-line blank subtraction in frequency-domain lifetime data acquisition (G.G. Reinhart, B. Feddersen, D. Jameson and E. Gratton, Biophys. J. 57 (1990) 189a) permitted the elimination of background interference from detergent-solubilized purified plasma membrane ATPase samples. The fluorescence decay of the erythrocyte Ca2+-ATPase was measured in the absence of Ca2+, or in the presence of Ca2+ or Ca2+ plus calmodulin. In the three different experimental conditions the fluorescence decay was very heterogeneous and could be best described by Lorentzian distributions of lifetime values. In the absence of Ca2+ the decay was described by a broad lifetime distribution centered at 4.4 ns with a width of 3.2 ns, indicating heterogeneity of tryptophan microenvironments in the ATPase. Calcium ion binding promoted an 11% increase in the center and a 27% decrease in the width of the distribution. By contrast, addition of calmodulin in the presence of Ca2+ caused a 15% decrease in the center of the distribution, revealing structural difference between calmodulin-activated and Ca2+-activated states of the ATPase. These results indicate the usefulness of on-line blank subtraction in frequency-domain lifetime measurements to investigate conformational changes in detergent-solubilized membrane protein samples.