Screening methods to indicate the genotoxic potential of individual chemicals or environmental mixtures rely mainly on short-term bacterial tests. Differences in the genotoxic response of prokaryotic and eukaryotic cells necessitate the development of nonbacterial screening assays. A promising approach for this purpose could be the comet (single-cell gel electrophoresis) assay performed with fish cells in vitro. In the present study, we evaluated the comet assay with two different fish cell lines from rainbow trout (Oncorhyhnchus mykiss), the fibroblast-like RTG-2 cell line established from gonad tissue, and the epitheloid RTL-W1 cell line established from liver tissue. The cells were exposed in vitro during 2 hr to the genotoxins, 4-nitroquinoline-1-oxide (NQO), and benzo(a)pyrene (BaP), as well as to environmental samples. The LOEC values for NQO were similar in both cell lines, whereas for BaP, the RTL-W1 cells were found to be more sensitive than the RTG-2 cells. The slopes of the concentration-response curves of the two test compounds differed between the two cell lines, with RTG-2 cells showing a steeper slope for NQO, and RTL-W1 cells showing a steeper slope for BaP. When exposed to environmental samples from a remediation site, the RTL-W1 cell line, but not the RTG-2 cell line, indicated a genotoxic potential of the samples. The differences in the genotoxic response pattern of the two cell lines could be only partly explained in relation to metabolic enzymes, cytochrome P4501A, glutathione-S-transferase, and xenobiotic reductase. The findings of this study demonstrate that the comet assay with fish cell lines is suitable as in vitro screening assay in environmental genotoxicity testing, but the choice of test cell line may be critical.