1. Adult rats and rats with a postnatal age of 3-29 days (PN 3-29) were used for the preparation of in vitro slices of the frontal neocortex. Epileptiform activity was induced by bath application of the γ-aminobutyric acid-A (GABA(A)) receptor antagonists bicuculline or picrotoxin. 2. The voltage-sensitive dye RH 414 and a laser scanning microscope were used for multiple-site optical recordings of membrane potential changes associated with epileptiform activity. Optical signals were compared with simultaneously measured extracellular field potentials. 3. Optical signals could be reliably recorded for the duration of the experiments (2-4 h). Extracellular recordings of convulsant-induced paroxysmal depolarizing shifts (PDSs) in slices stained with RH 414 were comparable with those obtained in unstained slices. Changes in dye signals in response to reductions in extracellular calcium, addition of tetrodotoxin (TTX), or application of excitatory amino acid receptor antagonists indicate that the fluorescence changes correlate well with established electrophysiological measures of epileptiform activity. 4. In slices from adult animals, dye signals were observed at all recording sites. The response with the shortest latency occurred invariably at the site of stimulation, and activity spread rapidly in both vertical and horizontal directions. Spread was significantly faster in the vertical than in the horizontal direction. 5. Epileptiform activity was absent or only weakly expressed in slices from PN 3-9 animals. Activity was detectable predominantly in upper cortical layers. 6. Dye signals were observed at all measurement points in slices from PN 10-19 animals. In this age group, peak amplitude increased with spread of activity from lower to upper cortical layers. There was no significant difference between the speed of propagation in the vertical and in the horizontal directions. Spontaneous epileptiform activity occurred at a high rate in the PN 10-19 age group, and signals associated with spontaneous epileptiform events were largest in upper layers. 7. In the PN 10-19 age group, optical signals were characterized by the repetitive occurrence of PDS discharges superimposed on a sustained response. The amplitude of the sustained response decreased with increasing distance from the site of stimulation. Analysis of the latencies revealed that the superimposed PDS-like events were generated at multiple sites within the scanning area. Amplitude and rate of rise were largest in slices from PN 10- 19 animals. These values declined with ongoing development. 8. The N-methyl- D-aspartate (NMDA)-mediated component of optically recorded epileptiform activity was determined by either subtraction of responses recorded in the presence of D-2-amino-5-phosphonovaleric acid (APV) from those taken under control conditions or by application of the non-NMDA receptor antagonist 6- cyano-7-nitroquinoxaline-2,3-dione (CNQX). The NMDA components obtained by these two procedures displayed slow rates of rise and different patterns of spatial distribution. 9. Our results demonstrate that the voltage-sensitive dye RH 414 can be used to reliably monitor the initiation, distribution, and spread of convulsant-induced epileptiform activity in the rat neocortex in vitro. In PN 10-19 animals, epileptiform activity displays the highest degree of synchronization, which declines with ongoing development. The observation that activity can start from multiple sites in the tissue to form long- lasting ictal-like events that are not observed in adult animals suggests that the threshold for the initiation of epileptiform activity may be lower in the immature neocortex.