The results were as follows;

1. The percentage of inactive genotype of ALDH2 in exposed workers was lower than that of exposed (P=0.081).

2. The percentages of exposed workers with inactive genotype did not have any significant difference by the increase of toluene exposure level or work duration.

3. The frequency of drinking, monthly and maximum amount of alcohol intake in workers with normal genotype were significantly higher than those with inactive genotype.

4. The urinary hippuric acid concentration of nonexposed workers ,with inactive genotype was significantly lower than that with normal genotype. Under 100 ppm of toluene, similar but statistically insignificant trends were found, while above that concentration of toluene, reverse but statistically insignificant trends were found.

5. The number of acute and chronic subjective symptoms were increased positively with the concentration of toluene in workers with normal genotype, but ho such trends were found in workers with inactive genotype.

6. The result of simple linear regression between toluene and urinary hippuric acid concentrations showed a very significant positive linear relation-ship. The mean hippuric acid concentration of nonoccupational exposure was 0.289+/-0.227 (0.062-0.516) g/l. Toluene exposure level unable to discriminate with nonoccupational exposure estimated from regression equation, it range from 7.29 to 9.87 ppm. Considering above all things, it was useful to estimate the exposure level of toluene by means of analysing urinary hippuric acid concentration in both genotype workers, but the biological exposure indices (BEI) of both genotypes were different from each other. The BEI of the total exposed workers was 2.76 g/ I, which was lower than current criteria 3.0g/ I (2.5 g/g Cr), and it also suggest that the BEI for the exposed workers in our country be lowered to the appropriate level after further study.