The evidence from epidemiological studies, studies of biochemical markers of exposure and toxicological studies, confirm that there is a causal association between the risk of lung cancer and exposure to environmental tobacco smoke. Nonsmokers can inhale and metabolize carcinogens in tobacco smoke and other markers of environmental tobacco smoke inhalation (nicotine and cotinine) are raised in nonsmokers exposed to environmental tobacco smoke. In epidemiological studies of women who are lifelong nonsmokers, there is a statistically significant excess risk of lung cancer (24%, 95% confidence interval 13-36%) from exposure to environmental tobacco smoke from the spouse and this increases with the number of cigarettes smoked and duration of the marriage. Misclassification bias (women who smoke but claim to be lifelong nonsmokers) and dietary confounding are unlikely to explain the association; after adjustment for both, the risk of lung cancer from environmental tobacco smoke exposure was still statistically significant. In any event, their effects on the risk of lung cancer in the epidemiological studies are balanced out by allowing for background exposure to environmental tobacco smoke (that is, other than from the spouse) in the reference group; the excess risk after allowing for all three is an estimated 26% (95% confidence interval 7-47%), similar to the adjusted figure of 24%. In Britain, about one in every six nonsmokers are exposed to tobacco smoke from smokers in the home. Passive smoking is an avoidable cause of mortality and morbidity. Prevention strategies to reduce the amount of cigarette smoking in public places should be part of public health policy.