No studies have specifically reported the association of lung adenocarcinoma with environmental tobacco smoke (ETS) exposure among nonsmoking males. The objective of this study was to examine the exposure-response relation between ETS exposure and lung cancer among nonsmoking males. In particular, the association with adenocarcinoma of the lung was studied. This is a population-based, case-referent study in Hong Kong during 2004-2006. A total of 132 Chinese male nonsmokers with newly diagnosed primary lung cancer and 536 nonsmoking community referents were interviewed about ETS exposures from the household and/or workplace, including ever ETS exposure, sources of exposure, number of smoking cohabitants/coworkers, and smoker-years. Univariate logistic regression analyses showed a weak association between all lung cancers and ever ETS exposure from the household and/or workplace (odds ratio (OR) = 1.11, 95% confidence interval (CI): 0.74, 1.67), but an increased risk was restricted to adenocarcinoma (OR = 1.68, 95% CI: 1.00, 2.38). After adjustment for family cancer history and other confounders, excess risk (OR = 1.62, 95% CI: 0.91, 2.88) still persisted for adenocarcinoma, although it was no longer statistically significant. Exposure-response relations for adenocarcinoma were found with increasing levels of all ETS indices when exposures from the household and workplaces were combined. The consistent exposure-response relations between ETS exposures and adenocarcinoma suggested a probable causal link, which would have to be confirmed by future larger studies.