We have recently shown that the Ca2+ response in endothelial cells evoked by readdition of Ca2+ to the medium after store depletion caused by a submaximal concentration of agonist can involve Ca2+ release from Ca2+ stores sensitive to both inositol 1,4, 5-trisphosphate and ryanodine. The present experiments were performed to determine whether this mechanism might also exist in other types of cell. For this purpose, we used the human carcinoma cell line A431, which has a varied resting [Ca2+]i. We found that the amplitude of the Ca2+ response evoked by Ca2+ readdition did not correlate with the amplitude of the preceding UTP-evoked Ca2+ release, but did positively correlate with the initial [Ca2+]i. An inspection of the two patterns of response seen in this study (the large biphasic and small plateau-shaped Ca2+ responses) revealed that there is an accelerating rise in [Ca2+]i during the biphasic response. Application of ryanodine during the plateau-shaped Ca2+ response reversibly transformed it into the biphasic type. Unlike ryanodine, caffeine did not itself evoke Ca2+ release, but it caused a further [Ca2+]i rise when [Ca2+]i had already been elevated by thapsigargin. These data suggest that in A431 cells, as in endothelial cells, the readdition of Ca2+ after agonist-evoked store depletion can evoke Ca2+-induced Ca2+ release. This indicates that Ca2+ entry may be overestimated by this widely used protocol.