Seven population groups were tested by radioimmunoassay for the presence of hepatitis A antibody (anti-HAV), hepatitis B surface antigen (HBsAg) and antibody to HBsAg (anti-HBs). Detection of anti-HAV was indicative of past hepatitis A infection and presence of HBsAg or anti-HBs indicated past hepatitis B infection. The results of tests of 680 serum specimens from the seven groups were as follows: (1) of 100 newly admitted children to Willowbrook where hepatitis A and B were hyperendemic, 32% had anti-HAV and 4% had anti-HAV and 4% had HBsAg (1%) plus anti-HBs (3%); (2) of 100 Willowbrook residents who lived in the institution about three or more years, 97% had anti-HAV and 90% had HBsAg (32%) plus anti-HBs (58%); (3) of 100 new Willowbrook employees, 50% had anti-HAV and 13% had anti-HBs; (4) of 100 Willowbrook employees who worked in the institution more than three years, 75% had anti-HAV and 30% had HBsAg (3%) plus anti-HBs (27%); (5) of 80 house staff physicians, 27.5% had anti-HAV and 10% had anti-HBs; (6) of 100 student 34% had anti-HAV and 6% had anti-HBs; and (7) of 100 suburban teenagers 4% had anti-HAV and 5% had anti-HBs. This study confirmed the variability of the prevalence of hepatitis A and B markers among different population groups and the effect of socioeconomic status and environmental factors on the incidence of past infection caused by hepatitis A and B viruses.