In 1985, the authors studied the epidemiology of hepatitis B virus (HBV) in a healthy Middle Eastern population. Residents of three remote villages and urban areas of Jordan were assessed for seroprevalence of hepatitis B surface antigen (HBsAg) and HBV infection. Infection was defined as the presence of antibody to hepatitis B core antigen (total) and/or antibody to hepatitis B surface antigen, with or without HBsAg. The overall HBsAg prevalence was 9.9% and was not age-related, although significant differences were noted between the villages (range, 5.7%-12.8%). The prevalence of HBV infection was 36% and increased with age. In addition, there were differences between the villages in patterns of age-specific infection. A newly constructed socioeconomic index showed greater HBsAg prevalence in lower (14.4%) than in upper (2.4%) socioeconomic groups. A history of contact with a jaundiced person and socioeconomic status were independent risk factors for HBsAg-positive status, while contact with a jaundiced person, rural background, and age were independently related to HBV infection. There was evidence of familial clustering of HBV in two of the villages, with HBV carriers and infected children particularly aggregating around HBsAg-positive siblings. There was also a trend toward an association of HBsAg-positive children with HbsAg-positive mothers. HBV carrier prevalence correlated with family size, and HBV infection in the household increased proportionately with the number of carriers in the family. Hepatitis B e antigen was detected most frequently in children and antibody to hepatitis B e antigen in adults. Postnatal early childhood transmission through contact among children of poorer and larger families probably accounts for the high endemicity of HBV in this region.