In the first case considered, inflation of excised rat lungs to 50% of TLC and subsequent deflation to an end-expiratory pressure of -5cm H2O produces a new, higher residual volume due to the "trapping" of gas by previously inflated lung units. Re-inflating the lung to the previous end-expiratory volume plus the volume of the previously trapped gas produces a deflation curve very similar to the preceding one, except for a parallel upward shift. Successive re-inflations and deflations produce a family of volume-pressure curves with these characteristics (1) the volume change upon inflation stays fixed as the end-inspiratory volume is always incremented by the gas volume trapped in the previous cycle, (2) re-inflation starts from ever higher end-expiratory volumes as the trapped gas volume cumulatively rises, and (3) deflation curves of nearly identical shape are shifted upward in a parallel manner as they start from an ever higher end-inspiratory volume. In the second case considered, the lungs are inflated to TLC and then deflated to an end-expiratory pressure of -5cm H2O. Successive re-inflation/deflation cycles generate cumulative additions to trapped gas volume. While the end-inspiratory volume stays at about TLC, end-expiratory volume rises as progressively smaller increments of trapped gas volume are revealed by successive deflations. The preceding phenomena are compatible with these notions (1) gas trapping occurs during inflation, (2) the gas volume trapped in a lung unit is close to that unit's maximal volume, (3) peripheral units experience trapping before central units, and (4) gas trapping in some units makes it easier for other units to undergo gas trapping on a subsequent re-inflation.