Using an in vitro clonogenic assay system, we examined the relationship between concentration, duration and schedule of exposure to etoposide and cytotoxicity in three cell lines. Two cell lines (SK-N-SH and IMR32) were derived from human neuroblastomas, and one (L1210) was the original murine leukaemia cell line used to define schedule dependency of etoposide in vivo. Cytotoxicity was found to be determined by the product of concentration and duration of exposure over a 120-fold range of durations and a 100-fold range of concentrations. No difference in cytotoxicity was seen following equivalent exposure either continuously or in two separate intervals. In one cell line, exposure to etoposide when at confluence led to a highly resistant subpopulation comprising 10-15% of the entire cell number. This did not seem to be associated with any difference in the rate of etoposide efflux from cells preloaded with 3H-etoposide. We conclude that etoposide does not show schedule dependency in vitro, but cytotoxicity is related to total exposure to etoposide. The schedule dependency seen in vivo may possibly arise from host pharmacokinetic factors.