National data from 1943 to 1987 on the two most frequent tobacco-related cancers in Denmark, lung and bladder cancer, were analyzed with multiplicative Poisson models. The temporal trends in the cohort-specific risks for both sites and sexes were similar: the risks increased in the beginning of the period covered by the analysis, but then levelled off; and there was no increase among cohorts born after circa 1930. Women experienced a smaller increase during the period covered by the analysis in the cohort-specific risk for bladder cancer than men (3.7 cf 6.1 times), whereas the overall increase in lung-cancer cohort-specific risk was the same for both sexes. The difference could not be explained by trends in tobacco consumption, types of tobacco consumed, or occupational exposures. On the basis of these findings, it is suggested that women may be less susceptible than men to developing bladder cancer from tobacco smoking.