The possible confounding effect of smoking on radon-associated risk for lung cancer mortality was investigated in a case-control study nested in the cohort of German uranium miners. The study included 704 miners who died of lung cancer and 1,398 controls matched individually for birth year and attained age. Smoking status was reconstructed from questionnaires and records from the mining company's health archives for 421 cases and 620 controls. Data on radon exposure were taken from a job-exposure matrix. Smoking adjusted odds ratios for lung cancer in relation to cumulative radon exposure have been calculated with conditional logistic regression. The increase in risk per Working Level Month (WLM) was assessed with a linear excess relative risk (ERR) model taking smoking into account as a multiplicative factor. In addition, the potential impact of temporal factors on the ERR per WLM was examined. Lung cancer mortality risk increased with increasing radon exposure, yielding a crude ERR per WLM of 0.25% (95% CI: 0.13-0.46%). Adjustment for smoking led only to marginal changes of the radon-associated lung cancer risks. The adjusted ERR per WLM was very similar (0.23%, 95%-CI: 0.11-0.46%) to the crude risk and to the risk found in the Wismut cohort study. This stability of the radon-related lung cancer risks with and without adjustment for smoking suggests that smoking does not act as a major confounder in this study and presumably also not in the cohort study.