The purpose of this study was to investigate the incidence rate of hepatitis B in the military service and to examine the effect of the asymptomatic hepatitis B surface antigen(HBsAG) carries on the incidence of hepatitis B. The subject were 223,270 men who were conscripted to the Korean Army from 1991 to 1994. According to the conscripted year, four conscription cohort were constructed. At the screening examination for military service no test for hepatitis B were performed in 1991 and 1992. In 1993, a screening test for hepatitis B were performed and those who were confirmed as HBsAg positive o. showed high titers $(geqq100IU)$ of nm glutamic-pyruvic transaminase(SGPT) were excluded from conscription. In 1994, the criteria for conscription was changed and those who were HBsAg positive were not excluded from conscription. Only those who showed $geq$ SGPT 100IU were excluded. The main results were as follows ; 1. The positive rate of HBsAg is 5.5% in the conscripted men. 2. The incidence rates of the hepatitis B in 1991 and 1992 conscription cohort were 9.96 and 8.10 per ten thousand person-year, respectively. The incidence rate of the hepatitis B was 1.34 per ten thousand person-year in 1993 conscription cohort which was confirmed as HBsAg negative at the screning test, and 7.41 per ten thousand person-year in 1994 conscription cohort which included the HBsAg positive. 3. The incidence rate of hepatitis B was 99.98 per ten thousand person-year in HBsAg positive group and 2.25 per ten thousand person-year in HBsAg negative group. The incidence rate of the group with high SGPT and HBsAg positive was 255 times higher than that of normal population. 4. The incidence of hepatitis B in HBsAg negative group did not increase even though the probability of personal contact with HBsAg positive had been increased. from the above results, the men who have high SGPT with HBsAg positive should be excluded from military service, and it can not be said that asymptomatic HBsAg carriers influence on the hepatitis B incidence among the HBsAg negative through personal contact.