Nature and nurture interact and in most cases it is impossible to specify quantitatively the contribution of either to the causation of a disease. Only rarely is either neutral in the sense that the other accounts for all the variation in risk. It seems unlikely that more than a few per cent of all cancers will be accounted for by inherited susceptibility with a high penetrance and a neutral environment. Small variations in susceptibility may, however, be associated with differentgenetic alleles that will facilitate focused measures of prevention and perhaps provide a lead to causation. For the practical control of most cancers we must intervene by treatment or prevention. Knowledge of the variation in the incidence of cancers suggests that age-specific incidence rates could be reduced by 80-90%, half by the application of existing knowledge. The possibilities of control by prevention are discussed under the headings of tobacco smoke, alcohol, infection, diet, physical activity, reproduction, medicines and medical procedures, and occupation and pollution. In conclusion, attention is drawn to eight types of cancer that have become more common in the UK in the last 25 years, some of which we do not know how to prevent and which require urgent research.