Lung cancer risk estimation in relation to residential radon exposure remains uncertain, partly as a result of imprecision in air-based retrospective radon-exposure assessment in epidemiological studies. A recently developed methodology provides estimates for past radon concentrations and involves measurement of the surface activity of a glass object that has been in a subject's dwellings through the period for exposure assessment. Such glass measurements were performed for 110 lung cancer subjects, diagnosed 1985 to 1995, and for 231 control subjects, recruited in a case-control study of residential radon and lung cancer among never-smokers in Sweden. The relative risks (with 95% confidence intervals) of lung cancer in relation to categories of surface-based average domestic radon concentration during three decades, delimited by cutpoints at 50, 80, and 140 Bq m(-3), were 1.60 (0.8 to 3.4), 1.96 (0.9 to 4.2), and 2.20 (0.9 to 5.6), respectively, with average radon concentrations below 50 Bq m(-3) used as reference category, and with adjustment for other risk factors. These relative risks, and the excess relative risk (ERR) of 75% (-4% to 430%) per 100 Bq m(-3) obtained when using a continuous variable for surface-based average radon concentration estimates, were about twice the size of the corresponding relative risks obtained among these subjects when using air-based average radon concentration estimates. This suggests that surface-based estimates may provide a more relevant exposure proxy than air-based estimates for relating past radon exposure to lung cancer risk.