The concentrations of 2,5-hexanedione (2,5-HD), an n-hexane metabolite, and 2-acetylfuran (2-AF) were measured in urine samples from 123 workers who had predominantly been exposed to n-hexane vapor and 53 workers who had experienced no exposure to solvents. The time-weighted average intensity of exposure to n-hexane vapor was determined by a diffusive sampling method. For biological monitoring of exposure, urine samples were collected late in the afternoon during the second half of a working week and were analyzed in the presence and absence of acid hydrolysis (at pH less than 0.5) for 2,5-HD and 2-AF by gas chromatography on a nonpolar capillary DB-1 column. The urinary 2,5-HD concentration increased as a linear function of the intensity of exposure to n-hexane, showing a correlation coefficient of 0.64-0.77 after acid hydrolysis and that of 0.73-0.83 in the absence of hydrolysis, depending on the correction for urinary density (P less than 0.01 in all cases, with no improvement in the coefficient occurring after the corrections). In contrast, 2-AF levels were independent of n-hexane exposure. The geometric mean 2,5-HD concentration in urine samples from 53 nonexposed men was 0.26 mg/l as observed (i.e., with no correction), 0.19 mg/l after correction for a urinary specific gravity of 1.016, and 0.23 mg/g creatinine after correction for creatinine concentration, and the geometric standard deviation was approximately 2.