The soluble organic fraction (SOF) of particulate matter from diesel exhaust (from point sources, ambient air, etc.) contains hundreds of organic constituents. Norman-phase high pressure liquid chromatography (HPLC) has been used to separate the SOF into subfractions suitable for subsequent chemical analysis and bioassays. These fractions consist of non-polar(PAH), moderately polar (transition) and highly polar constituents. The non-polar fractions have been well characterized and consist of PAH and aliphatic hydrocarbons. The specific compounds present in the transition and polar fractions are for the most part unknown. This analytical information has been difficult to obtain since these compounds are highly labile, polar, of low volatility and in very low concentrations when compared to the bulk of material found in the SOF. Mutagenicity tests using the Ames Salmonella typhimurium assay indicate that the transition fraction accounts for most of the mutagenicity when compared to the non-polar (PAH) and polar fractions. A variety of chromatographic and mass spectrometric techniques are described that have been used to determine the composition of the HPLC fractions. More than one hundred species have been identified in the transition fraction of diesel particulate matter using high resolution gas chromatography (HRGC)/high resolution mass spectrometry (HRMS), HPLC and direct-probe high resolution mass spectrometry. It has been found that the transition fraction contains mostly PAH derivatives consisting of hydroxy, ketone, quinone, carboxaldehyde, acid anhydride and dihydroxy derivatives of PAH. Three nitro-PAH species have been tentatively identified and 1-nitropyrene positively identified in the transition fraction. The 1-nitropyrene was found to account for approximately 45% and 30% of the direct-acting mutagenicity observed for the transition fraction and total extract, respectively. The HPLC separation procedure was shown to give better than 95% recovery of the mass and mutagenic activity. The problem of PAH oxidation during the analytical procedures and possible effect on bioassay results are discussed.