The incredible success of living birds (>9000 species) results, in part, from their unique respiratory system, which underpins the key evolutionary innovations of high metabolism and flight. This system comprises the lungs and a complex array of interconnected air sacs. The air sacs allow a unidirectional flow of air through the lungs, permitting exceptionally efficient gas-exchange. Extensions from the air sacs penetrate and pneumatize nearby bones, including vertebrae and limb elements, with the associated effect of reducing skeletal mass. In contrast, the closest living relatives of birds, crocodiles, lack air sacs and corresponding pneumatic features. There is now overwhelming evidence that birds are direct descendents of theropod dinosaurs. Many features previously regarded as uniquely avian appeared first among dinosaurs (e.g. feathers, brooding behaviour). The avian air sac system is another such feature: its presence in theropod and sauropod dinosaurs (and pterosaurs) has been inferred on the basis of pneumatic features in vertebrae that are almost identical to those seen in living birds. However, the origin of the air sac system is poorly understood: there is no consensus on whether air sacs and pneumaticity were present in the common ancestor of theropods, sauropods and pterosaurs, or whether they evolved independently in these three groups. Furthermore, possible evidence of pneumaticity has recently been identified in more distantly related Triassic archosaurs, prompting the controversial hypothesis that pneumaticity (and, by inference, air sacs and some bird-like respiratory capabilities) may have been present in the last common ancestor of birds and crocodiles, and subsequently lost in crocodilians. If true, this would require radical alteration of our understanding of the remarkable biology of birds and crocodiles and how they evolved. Understanding the origin of the avian respiratory system is clearly fundamental to explaining the success and diversity of the various archosaur lineages. However, the main alternative hypotheses have not yet been tested. We propose a pilot study to test alternative hypotheses explaining the origin of bird-like respiration. This work is timely, given recent intensive interest in dinosaur and bird respiratory systems, the availability of the research team and a new micro X-ray Computed Tomography (CT) facility at the Natural History Museum. We will determine the presence/absence of pneumatic structures in the vertebrae of selected Triassic archosaurs that lie close to the common ancestry of crocodiles and birds. The identification of pneumaticity will be based on external and internal vertebral anatomy: the latter data were previously unavailable, but will be obtained using CT scans - an entirely novel approach to this problem. The extent of pneumaticity, both within individual bones and throughout the skeleton, will be documented and the distribution of pneumatic structures will be determined by mapping the presence/absence of these features onto current archosaur evolutionary trees. This will permit us to establish: when pneumaticity appeared in archosaurs; whether the acquisition (or loss) of pneumaticity was a single event or occurred on multiple independent occasions; and the evolutionary sequence in which the different components of the air sac system appeared. Demonstrating the absence of pneumaticity in basal archosaurs would falsify hypotheses that a bird-like respiratory system was present in the ancestral archosaur, and support alternative hypotheses suggesting a later origin of air sacs. However, if pneumaticity is identified in primitive archosaurs this project will demonstrate that evolution of the air sac system is more complex than currently assumed and will facilitate future investigations into the origins of avian and crocodilian respiratory systems. This work will be of fundamental importance to palaeontologists, zoologists and physiologists.