It is pointed out that induction of lung cancer by exposure to Rn daughters, applied at high doses to miners and at low doses to exposures in homes, provides a very stringent and sensitive test of the linear, no-threshold dose-response relationship for high-LET radiation, because this relationship predicts that a substantial fraction of lung cancer among non-smokers is due to average Rn levels. Therefore, it predicts an easily observable elevation of lung cancer rates in areas where Rn levels are many times greater than the average, especially at times before cigarette smoking began to have important effects on lung cancer statistics. While more data are needed (and will be forthcoming), some of the early indications of these studies are reviewed here. Several cases are now known where average Rn levels are very high, and in all of these cases lung cancer rates are well below average. Methods for analyzing these data are discussed, and it is concluded that, based on current evidence, they indicate at least a factor of 4 disagreement with linear, no-threshold predictions.