The aim of this study was to verify the relationship between the cobalt oxide concentration in air and the cobalt concentration in urine. In the first study, we measured the cobalt concentration in the urine of 16 workers exposed to cobalt oxide in a Digital Video Cassette manufacturing plant at the beginning and end of their workshift during 2 working cycles. In the second study, we measured the exposure level to cobalt oxide and the cobalt concentration in urine at the end of the workshift of other 16 workers on 3 out of 5 consecutive work days (Monday, Wednesday and Friday). Four workers among the 16 participated twice. In the first study, the cobalt concentration in the urine increased about 1.5-3 fold during each workshift and then tended to decrease rapidly to the control value before the next workshift. In the second study, we examined the relationship between air and urine concentration using 22 samples after excluding data from workers with dust respirators. Comparing the air and urine concentrations for samples under 30 microg/m3 (n = 20), a significant correlation was found (r = 0.76). These results indicate that most of the inhaled cobalt oxide was excreted rapidly, and a good relationship was found between air and urine concentrations at the end of the workshift when the exposure level was under 30 microg/m3. Therefore, the cobalt concentration in urine at the end of a workshift can be used as an indicator of one-day exposure to cobalt oxide when the exposure is low level.