The relationship between average indoor levels of radon and lung cance
r mortality in the counties of Cornwall and Devon, England, are invest
igated. The associations of population density, social-class distribut
ion, and regional smoking prevalence with lung cancer mortality in the
local-authority districts of England and Wales were estimated by regr
ession analysis. Low rates of lung cancer in Cornwall and Devon were p
redicted from the relationship. The differences between observed and p
redicted mortality in Cornwall and Devon districts were compared with
average indoor levels of radon, which varied considerably between dist
ricts. Residual variations in lung cancer mortality were not significa
ntly correlated with average indoor radon measurements. The current ad
vice of the National Radiological Protection Board to government is to
concentrate radon measurements, remedial action, and preventive actio
n principally on Cornwall and Devon, but cross-sectional geographical
data do not support the hypothesis that raised levels of radon indoors
in southwest England have an important effect on lung cancer mortalit
y.