A comparison was made of temporal changes in lung cancer risk for the U.S. white and Swedish populations from 1950-51 to 1981-82 with the use of both life-table- and cohort-specific proportional hazard models to illustrate different features of the time series. Marked differences were noted for the two populations; Swedish lung cancer risks were generally lower than those in the United States. Sex-specific analyses were conducted that showed that Swedish males had a mortality rate roughly half that for U.S. males over the entire period. In contrast, U.S. white female mortality rates, though roughly equal to those for Swedish females in 1950-51, experienced a relatively more rapid increase; their rates were double the Swedish rates by 1980. The factors contributing to these patterns, such as marked cohort differences in risk, are identified by the various stages of analysis.