OBJECTIVES: To evaluate the personal exposure of members of the general public to atmospheric benzene, toluene, and the xylenes, excluding exposure from active smoking. METHOD: 50 volunteers were equipped with active air samplers for direct measurement of personal exposure to monoaromatic hydrocarbons (MAH) and an activity diary was completed during each sampling period. Exposures were also estimated indirectly by combining activity data with independent measurements of hydrocarbon concentrations in several microenvironments. RESULTS: Personal exposure were generally well in excess of those which would be inferred from outdoor measurements from an urban background monitoring station. A wide range of sources contribute to exposure, with indoor and in car concentrations generally exceeding those measured at background outdoor locations. Environments contaminated with tobacco smoke were among those exhibiting the highest concentrations. Personal exposures determined indirectly from activity diaries/microenvironment measurements were well correlated with those determined directly with personal samplers. Personal 12 hour daytime exposures to benzene ranged from 0.23-88.6 ppb (mean 3.81 ppb), with 12 hour night time exposures of 0.61-5.67 ppb (mean 1.94 ppb) compared with an annual average concentration of 1.18 ppb at the nearest suburban fixed site monitoring station. The excess of personal exposure over fixed site concentrations was greater for benzene and toluene than for the xylenes. CONCLUSION: A wide range of sources contribute to personal exposures to monoaromatic hydrocarbons with exposure duration being as important a determinant of total exposure as concentrations. Exposures generally exceed those estimated from concentrations measured by background fixed point monitors. Microenvironment sampling combined with activity diary information can provide satisfactory estimates of personal exposure to these compounds.