Since the end of the Second World War, cancer of the bronchus of the lung had become the commonest cancer to cause death in men in Britain and the United States. Doll and Hill in the United Kingdom and Hammond, Horn, Wynder and Graham in the United States have shown that the risk of developing lung cancer was proportionate to the number of cigarettes smoked, and that the non-smoker has a very small risk. In South Africa, cigarettes were, and still are, cheap. Most white South African men were heavier cigarette smokers than men in Britain. Yet I was seeing more patients with stomach cancer than with lung cancer. Lung cancer was still relatively uncommon in women because in the past they had smoked much less than men.

Because stomach cancer appeared to me to be more common than lung cancer among white South African males, in 1957 I looked at the national mortality statistics for cancer among white men and found that my suspicions were right: death from stomach cancer was considerably more common than from lung cancer. Because my research into multiple sclerosis and into the porphyrias had shown big differences in prevalence between the immigrants and the South African-born, I wondered whether British immigrants might have a different lung cancer mortality than white South African-born.

When I was next in Pretoria, I went to see Dr H.M. Stoker, director of the South African Census Bureau, and asked him if he would obtain for me an analysis of lung cancer deaths by South African-born, immigrants from the United Kingdom, and immigrants from the rest of Europe. He agreed to do this for the previous year. We found that the death rate from lung cancer among British immigrant men was twice as high as among the white South African-born in the same age-groups. Among the British male immigrants, it was the commonest cancer to cause death.