In Japan rubella occurs mostly in kindergarten and middle school students from 5 to 14 years of age. That is why it is presumed that school teachers of these age groups are at potential risk to be infected by this disease because of their close association with the children.Comparing prevalence rates of rubella HI antibody in females who were 21 years old or less and those aged 22 to 29 in 1983, about 87% of the first group (who had received the rubella vaccinations when they were 13 to 14 years old) were seropositive while the prevalence rate was low (about 70%) in the second group.School teachers in Sapporo Municipal schools were studied in October 1983 through questionnaires in order to investigate any special risk of their exposure to rubella. Among a total of 4, 165 school teachers aged under 40 years in 1983 in kindergarten, primary, middle and high schools in Sapporo, replies were obtained from 2, 412 of them (58%). A total of 2, 033 completed replies were analyzed, except those employed in 1983 when the rubella epidemic ceased. Among these 2, 033 school teachers, 932 had a history of rubella before employment, and 126 (6.2%) got rubella after employment. Among these 126 school teachers, 36 answered that they have suffered from rubella during the 1975-77 epidemic and 90 had it during the 1980-82 epidemic. The institute attack rates (%) were as follows: 11.3% (6 cases) in kindergarten, 6.5% (92 cases) in primary school, 5.4% (27 cases) in middle school and 1.5% (1 case) in high school; indicating higher attack rates among school teachers in the institutions for younger children, and a significant difference of the attack rates was obtained between kindergarten and high school. The increasing attack rate among staff in institutes for young school children showed a similar tendency in both epidemics, 1975-77 and 1980-82, compared with the attack rates of school-age children in the 1975-77 epidemic which was 26.4% in primary school, 15.3% in middle school and 2.9% in high shool.In view of these data, it was suggested that rubella infections of school teachers were associated with infected kindergarten and school children.