A seroepidemiological study was carried out in a geographically well-defined area in rural Crete in order to determine the prevalence of A, B and C hepatitis markers in the local population. Serum samples were obtained from 257 subjects (94 males, 163 females), aged 15 years and over, who visited the primary health care services of the Spili Health Centre between July 1993 and March 1994, and from 164 subjects (83 males, 81 females) randomly selected from households in three neighbouring villages of the study area. In samples obtained from the Spili Health Centre, antibodies to hepatitis A virus (anti-HAV) were detected in 234/244 (95.9%) subjects, antibodies to hepatitis B virus core antigen (HBcAb) were detected in 63/257 (24.5%) subjects and antibodies to hepatitis C virus (anti-HCV) were detected in 28/257 (10.9%) subjects. The corresponding figures for those randomly selected from the villages were 135/154 (87.7%), 16/164 (9.8%) and 5/164 (3%) respectively. Hepatitis B surface antigen (HBsAg) was positive in three (1.2%) subjects from the first group, while none of those recruited from the villages were positive for HBsAg. Interestingly, hepatitis markers were closely associated with age. No subjects under the age of 15 years showed evidence of prior hepatitis A infection and approximately 20% of those between 15 and 44 years of age were also negative. By contrast, practically all subjects older than 44 years were anti-HAV positive. Similarly, the majority of all those who were anti-HCV positive were older subjects. Seroepidemiology of hepatitis in this well-defined population seems to be different from other parts of Greece, at least for hepatitis B and C viruses. There is a very low prevalence of HBsAg and a very high incidence of anti-HCV. Low exposure to HAV, as found in other parts of the country, was also found in the younger generation in this rural area of Crete.