Crystalline silica has been classified as a human carcinogen, but there is still considerable debate on its variable fibrogenic and carcinogenic potential. We investigated genotoxicity of a panel of four quartz flours in comparison to DQ12 standard quartz with similar size and surface area, using single cell gel electrophoresis (SCGE) or comet assay. A549 human lung epithelial cells were incubated for 4 hours with different concentrations of quartz ranging from 1.6 to 200 mug/cm(2) and cytotoxicity was assessed using leakage of lactate dehydrogenase (LDH), trypan blue exclusion and conversion of a metabolic substrate (MTT). DNA strand breakages were seen with all quartzes at an in vitro concentration of 200 mug/cm(2). At this concentration all tests and quartz samples showed significant cytotoxicity. The most toxic quartz flour (Qz 2/1-C) but not DQ12, showed an increase in strand breaks at 40 mug/cm(2) in cell culture. At this concentration no cytotoxicity was seen with LDH and MTT, but a significant increase in cells with trypan blue uptake was noted. No differences in tail moment percentage were observed at equal concentrations of different quartz flours. Also no correlation between DNA damage and OH-radical generation or surface radicals as measured by electron spin resonance was observed. We conclude that quartzes do not cause strand breaks without concomitant cell toxicity and a sufficient in vitro concentration of > 40 mug/cm(2) can only be reached in vivo with instillation of massive doses (> 100 mg). Therefore, in vitro genotoxicity found here is unlikely to explain the genotoxicity observed in in vivo studies with the same and other quartzes.