Chronic hepatitis B virus infection is far less common in urban born than in rural born southern African blacks, who also have a high incidence of hepatocellular carcinoma. A case-control study was carried out to determine the relative frequency of hepatocellular carcinoma and its relation to hepatitis B virus infection in urban born blacks. Three hundred and ninety two black patients with hepatocellular carcinoma and matched controls seen at two city hospitals were classified by questioning as urban born or rural born. The ratio of rural born to urban born blacks among the controls was 1.1:1.0 (207/185), whereas in the patients with cancer the ratio was 4.8:1.0 (324/68) (p less than 0.0001). Analysis of the prevalence of hepatitis B markers in 62 urban born and matched rural born blacks with hepatocellular carcinoma showed no differences in the frequency of current or past hepatitis B virus infection. It is concluded that urban born blacks are less likely than rural born blacks to develop hepatocellular carcinoma, but when they do the tumour is equally likely to be related to infection with hepatitis B virus. The findings lend further support to an important role for chronic hepatitis B virus infection in the aetiology of hepatocellular carcinoma.