The occurrence of second primary cancers in patients with lip cancer was evaluated in order to test certain etiologic hypotheses. All cases of lip cancer reported to the Finnish Cancer Registry in 1953-74 (3303 men, 320 women) were followed up for a second (or third) primary cancer through the files of the Registry either to death or to 31 December 1974 (a total of 25 510 person-years). The expected numbers of cases were calculated on the basis of the incidence rates specific for sex, age, time and residence (urban or rural) in Finland. A higher than expected risk of cancer was found among both urban and rural male lip cancer patients; among women the observed number of new primary cancers did not differ from that expected. A significant excess risk among males was noted for cancers of the lung and larynx (rural patients) and for non-melanomatous skin cancer in locations other than the head and neck (urban patients). The association of cancers of the lip, lung and larynx found earlier on a geographic level supports the hypothesis that tobacco smoking is a common risk factor in these cancers. The differences in the relative risks between urban and rural patients, however, suggest that the risk factors in lip cancer in urban areas might be partially different from those prevalent in the rural population. The results do not support the hypothesis that sunlight is an important risk factor in lip cancer.