The avian respiratory system is composed of a unidirectionally ventilated, volume‐constant gas‐exchanging lung, and a series of compliant, flexible air sacs. Diverticula from the lungs and air sacs invade adjacent bone and create air‐filled cavities within the skeleton through the process of pneumatization. While previous studies have addressed the presence or absence of pneumatic bones in multiple species, the pattern of pneumatization has been vaguely generalized with respect to which lung elements are responsible for pneumatizing different skeleton regions. Here, we aim to address which components of the avian respiratory system are pneumatizing each component of the postcranial skeleton and how the patterns vary between different taxa. Computed tomography (CT) and microCT data are used to visualize the pneumatized bones and to segment 3D digital surface models of the skeletal and respiratory systems. Specimens used in this study include the African grey parrot (Psittacus erithacus), ostrich (Struthio camelus), red‐tailed hawk. (Buteo jamaicensis), and tundra swan (Cygnus columbianus). We found substantial differences in the pneumatization patterns between the birds. In all taxa examined, there is an extensive supramedullary diverticulum that travels through the vertebral canal which directly pneumatizes each vertebra in the parrots but does not contribute to any pneumatization in the swan or hawk. The sacral, pelvic, and femoral elements are pneumatized solely by pelvic diverticula in the ostrich, but predominantly by the abdominal sacs in parrots. These data indicate that the pneumatization patterns in birds are highly variable and warrant further study.

Innovations in three‐dimensional (3D) imaging and segmentation have facilitated unprecedented levels of anatomical investigation into the detailed structures of the respiratory system that are often difficult to study in situ. Recent hypotheses of homology between crocodilians and birds have facilitated quantitative comparative analyses of bronchial trees and in situ models reveal new complexities in the relationship between the respiratory and skeletal systems. Here we quantitatively compare the bronchial trees of two crocodilians, the American alligator (Alligator mississippiensis) and Cuvier’s dwarf caiman (Paleosuchus palpebrosus) with select birds, including the ostrich (Struthio camelus), the African gray parrot (Psittacus erithacus), and the red‐tailed hawk (Buteo jamaicensis). Notably, the relative distances from the carina to the secondary bronchi measured are conserved, indicating a possible ancestral or constrained trait. With respect to interspecific avian comparisons, we found grossly observable variation within a single taxon in air sac morphology (e.g., P. erithacus), as well as substantial differences between the individual taxa via segmented surface models – particularly in the expansions of the interclavicular sacs, the extent of the diverticula, and the size of the abdominal sacs. Furthermore, we found that specific sac contribution to postcranial pneumatization varies substantially across our dataset. Individual specimens imaged for this study also revealed multiple pathologies, including scoliosis, foreign objects inside the animals, and broken bones, which have been incorporated into the anatomical models for clinical surgical atlases that are under development. While these data are preliminary, they provide a framework for larger scale comparisons and hypotheses of the ancestral archosaurian pulmonary system.