The mutational effects of pentachlorophenol (PCP) on the hypoxanthine phosphoribosyl transf erase (hprt) locus in human T-cell were analysed by T-cell clonal assay in vitro. Cells were exposed for 24 hours at primary culture to 0~100 ppm (W/V) PCP in dimethyl sulfoxide. Treated cells were allowed at the same time to stimulate by phytohemagglutinin (PHA) and T-cell growth factor (TCGF) and then seeded in medium containing 6-thioguanine to select for hprt-negative routants. We have also defined the optimal condition for the determination of mutant frequency. The parameters investigated include survival counting, first and second subculture for clonal efficiency plating and mutant plating. Under the optimal conditions, mutant frequencies of high dose-treated cells were significantly higher than those of non-treated or low dose cells. The results indicated a clear dose-effect relationship and showed that mutant frequency in 50 ppm PCP treated cell was 4.31$ imes$$10^{-5}$ (background, 8.32$ imes$$10^{-6}$). Above data strongly suggest that hprt mutation assay can be used as a biomarker for the environmental risk assessment.