Although changes in intracellular Ca2+ concentration ([Ca2+]i) are spatially heterogeneous during spontaneous contraction in mammalian cardiac muscle, it has not yet been observed how [Ca2+]i changes spatially within cardiac myocytes during delayed (DADs) and early (EADs) afterdepolarizations. The aim of this study is to characterize the spatial features of the increase in [Ca2+]i during such afterdepolarizations and to understand the ionic mechanisms responsible for them. Myocytes were enzymatically isolated from guinea pig ventricles and loaded with fura 2-acetoxymethylester, the Ca2+ fluorescence indicator dye. Membrane potential was recorded with a conventional microelectrode technique, and spatiotemporal changes in fura 2 fluorescence and cell length were recorded using a digital television system. After superfusion with potassium-free Tyrode solution, DADs and EADs were induced. During DADs, fluorescence transients were heterogeneous within myocytes (n = 11). Furthermore, they often propagated within myocytes as if they were "waves." In contrast, during EADs, fluorescence transients showed no waves within myocytes but rather showed synchronous changes throughout the myocytes (n = 15). The results of this study suggest that the spatial features of the increase in [Ca2+]i differ between the DADs and EADs. We concluded from these differing features that the ionic mechanisms responsible for the two triggered activities are different.