In recent years a strong case has been made in support of a viral aetiology for at least some primary hepatocellular carcinoma (PHC) in areas of low incidence. By pooling routinely collected cancer registration and infection data, study of the relationship between hepatitis B virus (HBV) infection and the incidence of PHC in Scotland over the period 1972-1985 has confirmed this view. Over this period the incidence of PHC in men increased, there was a relationship between the incidence of notification of HBV infection and that of hepatocellular carcinoma in different parts of the country, and an increased risk attached to those chronically infected with the virus. Given the recent introduction of lower cost yeast derived vaccines, there may now be more scope for prevention both of primary liver cancer and of other liver disease.