To identify the method of choice for analysis of urine for 2,5-hexanedione (2,5-HD) as an indicator of occupational exposure to n-hexane, the end-of-shift urine samples of 36 n-hexane exposed male workers and 30 non-exposed male workers were analyzed for 2,5-HD under three conditions of hydrolysis, i.e. enzymic hydrolysis at pH 4.8, acid hydrolysis at pH 0.5, and without hydrolysis. The 2,5-HD concentrations thus determined were examined for correlation with 8-h, time-weighted average exposure concentrations of n-hexane measured by diffusive sampling. The regression analysis showed that the 2,5-HD concentrations without any hydrolysis correlated best with the intensity of exposure to n-hexane. No 2,5-HD was detected in the urine of the non-exposed subjects under the analytical conditions with no hydrolysis. Thus, the analysis without hydrolysis was considered to be the method of choice from the viewpoint of simplicity in analytical procedures, sensitive separation of the exposed from the non-exposed, and quantitative increase in the amount of 2,5-HD after n-hexane exposure.