The lung volume at which airway closure begins during expiration (closing volume, CV) can be measured 1) with a radioactive bolus inspired at residual volume (RV) and 2) with the single-breath N2 elimination test. In previous studies in dogs, we observed that N2 CV was systematically larger than 133Xe bolus CV (Xe CV) [N2 CV %vital capacity (VC) = 35 +/- 2.3 (SE) vs. Xe CV %VC = 24 +/- 2.2, P less than 0.01]. Because the regional RV in the dog is evenly distributed throughout the lung and all airways closed at RV, N2 CV is related to the regional distribution of the tracheal N2; differences between N2 and Xe CV could then be related to the size of the inhaled dead space. Simultaneous measurements of Xe and N2 CV were performed at various sites of Xe bolus injection while the regional distribution of the bolus was measured. Injections at the level of the carina increased Xe CV to a value (30 +/- 1.4%VC) near simultaneous N2 CV (32 +/- 1.5%VC) and increased the unevenness of regional distribution of the Xe bolus. The difference between N2 and Xe CV is then the result of the size of the inspired tracheal dead space. Moreover, comparisons between different values of Xe CV require injections of the boluses at the same distance from the carina.