Lifestyle is a convenient portmanteau term which, in relation to the causes of cancer, has come to mean all aspects of the way people behave, whether determined voluntarily or imposed by economic, cultural, or geographic circumstances, including reproduction but arbitrarily excluding occupation. Modification of lifestyle seems to be able to reduce the age-specific risk of cancer by about four fifths in many communities. This does not imply that the control of other factors could not reduce the risk by more than a fifth, as the chain of causation can be broken in many places and the occurrence of any one case of cancer could be avoided in several different ways. Examples are given of the way different aspects of lifestyle interact, and particular attention is paid to the consumption of ethanol which has a major impact on the incidence of human cancer, although it does not appear to be a carcinogen in animal experiments.