We analyzed preparative and analytical aspects of the dynamic localization of Ca2+ during cell stimulation, using a combination of quenched how and energy-dispersive X-ray microanalysis (EDX). Calcium (or Sr, as a substitute) was retained as fluorides during freeze-substitution, followed by epoxide embedding. The quenched-how used allowed analyses, during stimulation, in the subsecond time range. Sections of 500 nm were analyzed and no artificial Ca or Sr leakage was recognizable, We calculated a primary beam spread from 63 to 72 nm that roughly indicated the resolution of EDX/structure correlation, These values are quite compatible with the size of potential structures of interest, e.g., Ca stores (similar to 100-nm thickness) or cilia (similar to 250-nm diameter). We used widely different standards to calibrate the ratio of CaKalpha net counts in relation to actual [Ca]. Calibration curves showed a linear relationship and a detection limit of [Ca] = 2 mM, while [Ca] in cytosol was 3 mM and in stores was 43 mM, both in nonactivated cells. Eventually Sr2+ can rapidly be substituted for Ca2+ in the medium before and during stimulation, thus allowing one to determine Me2+ fluxes. With our "model" cell, Paramecium, we showed that, upon stimulation (causing rapid Ca2+ mobilization from subplasmalemmal stores), Ca was immediately exchanged for Sr in stores,