We tested whether subjects could detect and localize inflation confined to a single lung lobe. A balloon-sealed catheter was placed into a lobar bronchus of unsedated subjects via fiberoptic bronchoscopy. Topical anesthesia (lidocaine) was used to suppress cough and irritation associated with inflation of the sealing balloon. Small (45-60 ml) or large (100-240 ml) stimulus volumes were insufflated via the catheter. In a forced-choice protocol, subjects were readily able to detect large inflations and correctly identify the side on which the stimulus was given, but small inflations were at the threshold of detection and were not correctly localized. Additional lidocaine applied to the bronchus in two subjects did not degrade detection. Circumstantial evidence suggests that the sensation arose in the lung. We conclude that this technique is feasible for the study of pulmonary perception.