With this letter, we wish to raise caution about interpretations of results owing to artifacts that may occur when cell lines are derived by culturing on cell-derived feeder layers. We recently reported the apparent derivation of rat liver progenitor cell lines by co-culture of liver cells enriched in periportal oval cells after allyl alcohol injury with ␥-irradiated mouse STO (SIM [Sandoz inbred Swiss mouse]) fibroblast feeder layers. 1 These reported lines are now undergoing intensive reinvestigation in our laboratories and the results indicate that the cells were actually derived from the STO feeder layers. In a series of first stage RT-PCR and PCR, aimed at extending the key original findings of constitutive and inducible differentiation, we were unable to detect rat genes or rat gene products in any of the 11 reported cell strains and lines, including the clonal "rat" line 3(8)#21, and its clonal derivative 3(8)#21-EGFP, a line transduced with enhanced green fluorescent protein. Instead, only mouse genes were detected in all instances. These results were obtained both with purified and cloned putative "rat liver progenitor" cells grown on or off STO feeder layers (␥-irradiated at UCSD), and with the original frozen "rat" cells obtained from untampered-with vials derived from very early passages at the Albany Medical College. These and other unexpected observations, which tend to eliminate cell fusion as the source of this artifact, are described in detail elsewhere. 2 Ongoing observations of hepatocytic functions in these lines suggest that some aspects of the original report may yet be correct. For example, the line best characterized, 3(8)#21-EGFP, expresses several hepatocyte properties in vitro and following transplantation (manuscript in preparation). In conclusion, this surprising and previously unreported type of cell culture artifact should be considered by all investigators of cultured stem cell systems which require co-culture with mitotically "inhibited" feeder layer cell lines.