Earliest theropod abdominal air sacs?

Skeletal pneumaticity is the presence of air within bones of animals. This is typically in the presence of sinuses (think of your face and your achey sinuses during a cold, caused by a build up of pressure in the air spaces), or in birds, when the respiratory system projects part of itself into the bones to invade and hollow them, typically seen in in many avian vertebrae and wing bones. In birds, their respiratory system is more advanced than those in mammals, with air flow being separated between oxygenated (the air breathed in), and de-oxygenated (used air being breathed out), while mammal respiration is less efficient mixing both oxygenated and de-oxygenated air.  For more background on pneumaticity and post cranial pneumaticity, check out my previous post on pterosaur pneumaticity (and the lightweight skeleton of birds.

In addition to birds, postcranial pneumaticity is commonly found in some animals in the fossil record, including pterosaurs, and non-avian dinosaurs. Sauropods often have highly pneumatised vertebrae, thought the help keep them light and facilitate movement of their massive necks, while some theropods have pneumatic vertebrae and even postcranial elements in some species. Traditional studies on pneumaticity have used just visual methods to identify pneumatic foramina and determine if elements are pneumatic, but more recently, scientists have started using CT scans to look inside the bones and determine if they are pneumatic. This allows us to see through any matrix present, and see where the foremen leads to, without destroying the specimen.

The presence of pneumaticity in theropod dinosaurs was originally thought to be something leading towards birds, as the efficient respiratory system is believed to be what allows birds to be so successful, allowing for better breathing during flight. However, the exact timing of the bird-like respiratory system has been unclear and controversial. A new study, lead by Akinobu Watanabe from the American Museum of Natural History, and published in PLOS ONE, looked at the presence of postcranial pneumaticity in Archaeornithomimus and other ornithomimosaur dinosaurs, a group of theropods not directly on the branch to modern birds. Using CT scans, they were able to show that Archaeornithomimus had pneumatic cervical (neck), dorsal (back), and caudal (tail) vertebrae, but there was no unequivocal evidence of pneumatic sacral vertebrae, although there were some possible pneumatic fossae. Watanabe et al. (2015) also looked at other ornithomimosaurs to look at the evolution of pneumaticity in this group, but unfortunately these specimens were studied without CT scans. They found that the cervical vertebrae of Nqwebasaurus (basal ornithomimosaur), Pelecanimimus, Gallimimus and Ornithomimus showed evidence of pneumaticity, while the dorsal vertebrae of Gallimimus are also pneumatic. They suggest that the sacrum of Gallimimus is also pneumatic, but without CT scans showing precisely where these foramina are going, it's hard to be sure.

Cervical vertebrae and CT images taken at specific points of Archaeornithomimus (Watanabe et al. 2015)

Now the important part of the paper - what does is mean for the evolution of pneumaticity in ornithomimosaurs? Compared to several other groups of non-avian theropods, ornithomimosaurs are less pneumatic. Basal members are less pneumatic (with just their cervical and possibly dorsal vertebrae showing evidence), while more derived members Archaeornithomimus, Gallimimus, and Deinocheirus may have independently evolved higher levels of pneumaticity, which is especially evident in Deinocheirus. Additionally, the presence of a pneumatic hiatus, or an area where the vertebrae appear not to be pneumatised between two sections that are, suggest the presence of distinct air sacs. In the case of Archaeornithomimus, the dorsal and caudal vertebrae are pneumatised, while the sacral are not, suggesting that ornithomimosaurs may have had distinct abdominal air sacs, the evolution of which has been contentious in theropods. If this is the case, this represents the earliest appearance of abdominal air sacs in coelurosaurian dinosaurs. The authors suggest that this may mean that pneumatic hiatuses have been missed before, without the use of CT revealing other pneumatic features.

This paper highlights the need for CT scans in fossil data, and the numerous questions that still exist in understanding the evolution of post cranial pneumaticity in birds, dinosaurs, and of course in my favourites - pterosaurs. As postcranial pneumaticity evolved in all of these groups, several questions about their evolution exist. Derived pterodactyloid pterosaurs appear to have had abdominal air sacs as well, so did they evolve first in a common ornithodiran ancestor, and were subsequently lost by ornithischian dinosaurs and other pterosaurs? Or did they evolve in a basal saurischian ancestor and pterosaurs separately? Or possibly they evolved several different times? We still don't know the answer.

Reference:
Watanabe A, et al. (2015) Vertebral Pneumaticity in the Ornithomimosaur Archaeornithomimus (Dinosauria: Theropoda) Revealed by Computed Tomography Imaging and Reappraisal of Axial Pneumaticity in Ornithomimosauria. PLoS ONE 10(12): e0145168. doi:10.1371/journal.pone.0145168