Lower lobe origin and histologic diagnosis of adenocarcinoma have been described as useful parameters for attributing lung cancer to prior asbestos exposure. To assess whether these characteristics differed between asbestos-exposed individuals and smokers, we evaluated lobe of origin and histologic type of tumors in 78 asbestos-exposed and 214 nonexposed heavy smokers developing lung cancer during the Carotene and Retinol Efficacy Trial (CARET), a prospective cancer chemoprevention trial. Most tumors in both cohorts, regardless of radiographic fibrosis at baseline, originated in upper lobes, representing 67% in asbestos-exposed and 80% in smokers, respectively (adjusted OR for lower lobe = 1.41; 95% CI = 0.69-2.91). Adenocarcinoma represented 32% of lung tumors in the asbestos cohort, and 30% in the smoking cohort (adjusted OR = 0.78; 95% CI = 0.40-1.55), and was inversely associated with radiographic fibrosis (adjusted OR = 0.19; 95% CI = 0.06-0.62). We conclude that neither anatomic site nor histologic cell type of tumors distinguishes effectively between smoking and asbestos as causal factors in development of lung cancer.