Abstract
(1) The background of the Bantu and “Cape Colored” races in South Africa is briefly reviewed. The organization of a cancer survey among these people is described in detail and the inherent errors and difficulties are outlined. However, it is believed that with due precautions reasonably accurate incidence rates for cancer have been obtained. (2) The cancer rates determined for these populations in Johannesburg are presented. In the Bantu, cancer of all sites is about half as common as in the United States population. (3) In the Bantu, carcinomas of the esophagus and liver are significantly more frequent than in the populations of the United States and Denmark, whereas tumors of the remainder of the gastrointestinal tract are much rarer. Carcinomas of the paranasal sinuses are common and may be related to snuff. Carcinoma of the lung is nearly as frequent as in Denmark, and the possibility that factors other than cigarette smoking may be of significance is discussed. Carcinoma of the cervix is common but carcinoma of the body of the uterus is almost unknown. Breast carcinoma is much less frequent than in the United States. Tumors of the bones and soft tissues, lymphomas, and leukemias are less frequent than in the United States. Neoplasia in childhood is less frequent than in the United States but the incidence of retinoblastomas and nephroblastomas is not significantly different. Differences are also observed in the incidence of cancer of the skin, bladder, kidney, prostate, and eye. The possible etiological factors involved in these various sites are discussed briefly. (4) Ratio figures from a rural area indicate that as yet, the cancer pattern in urban Johannesburg has not altered greatly as compared to the former. (5) The cancer pattern in the Cape Colored population is similar to that in the United States Negro and differs considerably from that seen in the Bantu.