A novel high-performance liquid chromatographic/ultraviolet method was developed to detect lapachol (LP) and deoxylapachol (DLP) in wood dust as chemical markers of teak wood (a suspected human carcinogen). The specificity of this analysis was determined by noting the absence of LP and DLP in 12 other specimens of different woods belonging to the angiosperm family. The consistency was examined by analyzing teak from three different sources, where the percentages (wt/wt) of the chemicals ranged from 0.006 to 0.261 for LP and from 0.038 to 0.497 for DLP, respectively. Although the LP and DLP components of teak varied according to source, a very high correlation coefficient (r (2) > 0.98 always) was found between the content of the two markers in the bulk specimens and in bulk dust derived from them. The method was then applied to teak dust collected on polyvinylchloride filters from aerosol in an exposure chamber in the range of mass loadings between 0.03 and 3.65 mg, which corresponds to a dust exposure between 0.124 and 8.703 mg m(-3) for a sampling time of 2h. A field test was also carried out in a small factory where teak was used. A good correlation was confirmed between LP and DLP versus the dust collected on the filter in both cases. LP and DLP can be markers to estimate the true quantities of teak dust inhaled in a workplace with mixed wood dust, provided the results are matched to the content of LP and DLP in the bulk wood. LP and DLP have also been proposed as the agents responsible for allergic reaction to teak dust. Therefore, it would be useful to evaluate the exposure to these two substances even without a relationship to teak dust exposure.