We studied dogs with unilateral papain-induced emphysema to answer two questions: (1) Do emphysema lung-apposed hemidiaphragm (Di(E)) and normal lung-apposed hemidiaphragm (Di(N)) have equal capacities for lowering lung surface pressure? and (2) Are side-to-side differences in intrathoracic pressure the result of unequal force outputs by Di(E) and Di(N) or are they caused by differences in their mechanical efficiency as pressure generators? After the airways of the emphysematous and normal lungs were intubated with a dual lumen endotracheal tube, both phrenic nerves were maximally stimulated at rates between 1 and 50 Hz and the changes in airway occlusion pressure (DELTAPao(E,N)) and diaphragm length (sonomicrometry) were recorded. In all animals, DELTAPao(N) exceeded DELTAPao(E). Differences in pressure ranged from 1.2+/-0.6 cm H2O during a twitch to 6.0+/-2.9 cm H2O during a 50-Hz tetanus. Midcostal bundles of Di(E) shortened less than corresponding bundles of Di(N), but both reached the same active length relative to their optimal lengths, which were measured in vitro. There was no significant difference in fiber type distribution, fiber cross-sectional area, or maximal isometric tetanic tensions among midcostal regions of Di(E) and Di(N). We conclude that unilateral hyperinflation impairs the mechanical efficiency of the apposing hemidiaphragm as a pressure generator.