The anatomy and volume of the penguin respiratory system contribute significantly to pulmonary baroprotection, the body O 2 store, buoyancy and hence the overall diving physiology of penguins. Therefore, three-dimensional reconstructions from computerized tomographic (CT) scans of live penguins were utilized to measure lung volumes, air sac volumes, tracheobronchial volumes and total body volumes at different inflation pressures in three species with different dive capacities [Adeĺie (Pygoscelis adeliae), king (Aptenodytes patagonicus) and emperor (A. forsteri) penguins]. Lung volumes scaled to body mass according to published avian allometrics. Air sac volumes at 30 cm H 2 O (2.94 kPa) inflation pressure, the assumed maximum volume possible prior to deep dives, were two to three times allometric air sac predictions and also two to three times previously determined end-of-dive total air volumes. Although it is unknown whether penguins inhale to such high volumes prior to dives, these values were supported by (a) body density/buoyancy calculations, (b) prior air volume measurements in free-diving ducks and (c) previous suggestions that penguins may exhale air prior to the final portions of deep dives. Based upon air capillary volumes, parabronchial volumes and tracheobronchial volumes estimated from the measured lung/airway volumes and the only available morphometry study of a penguin lung, the presumed maximum air sac volumes resulted in air sac volume to air capillary/ parabronchial/tracheobronchial volume ratios that were not large enough to prevent barotrauma to the non-collapsing, rigid air capillaries during the deepest dives of all three species, and during many routine dives of king and emperor penguins. We conclude that volume reduction of airways and lung air spaces, via compression, constriction or blood engorgement, must occur to provide pulmonary baroprotection at depth. It is also possible that relative air capillary and parabronchial volumes are smaller in these deeper-diving species than in the spheniscid penguin of the morphometry study. If penguins do inhale to this maximum air sac volume prior to their deepest dives, the magnitude and distribution of the body O 2 store would change considerably. In emperor penguins, total body O 2 would increase by 75%, and the respiratory fraction would increase from 33% to 61%. We emphasize that the maximum pre-dive respiratory air volume is still unknown in penguins. However, even lesser increases in air sac volume prior to a dive would still significantly increase the O 2 store. More refined evaluations of the respiratory O 2 store and baroprotective mechanisms in penguins await further investigation of species-specific lung morphometry, start-of-dive air volumes and body buoyancy, and the possibility of air exhalation during dives.