Calcium ions are known to play an important role in epileptogenesis. Although there is clear evidence for increased neuronal calcium influx during epileptiform potentials, direct measurements of the corresponding intracellular calcium transients are rare and the origin of calcium influx is not known. Therefore the spatial and temporal distribution of intracellular calcium transients during epileptiform activity in guinea pig hippocampal slices was monitored with the use of the indicator Calcium-Green and a fast optical recording method. Two models of epilepsy (bicuculline and low Mg2+) were compared. In both models, single epileptiform events were evoked by electrical stimulation of the Schaffer collaterals in CA1 or of stratum pyramidale in area CA3. Intracellular calcium transients during epileptiform activity were approximately 5 times larger than during control stimulation. Calcium transients during epileptiform activity were present across at least the entire CA1 area, whereas presynaptic calcium transients from stimulated fibers were only seen at a distance up to 1 mm from the stimulation site. DL-2-amino-5-phosphonovaleric acid (APV), a specific antagonist of the N-methyl-D-aspartate (NMDA) receptor, abolished low-Mg2+ epileptiform activity and reduced bicuculline-induced epileptiform activity; it reduced calcium transients following stimulation of CA1 by only 29% (bicuculline) and 38% (low Mg2+). For comparison, calcium transients during control stimulation were 78% (bicuculline) and 69% (low Mg2+) smaller than epileptiform calcium transients. At a distance from the stimulation site, calcium transients and their NMDA-receptor-dependent components were largest in stratum pyramidale in the bicuculline model and in stratum oriens in the low-Mg2+ model. In both models, minimal onset latencies of calcium influx shifted with increasing distance to the stimulation electrode from stratum radiatum to stratum oriens. APV reduced the extent of spread of calcium transients in the low-Mg2+ model. In the bicuculline model, the spatial extent of spread of epileptiform calcium transients was not affected by application of APV; however, the mean velocity of spread was reduced from 0.20 to 0.12 m/s. In conclusion, the large size of calcium transients and of their NMDA-receptor-dependent components in stratum pyramidale or stratum oriens as well as shortest onset latencies of calcium transients at these sites suggest an important role of cell somata, basal dendrites, and possibly local circuit excitatory interactions for the generation and spread of epileptiform activity.