Data from: Quantifying the gastral mass in Early Cretaceous ornithuromorphs (Aves: Ornithothoraces) from the Jehol avifauna
Data files
Aug 25, 2023 version files 202.24 KB
-
Data_1.xlsx
-
Data_2.xlsx
-
Data_3.xlsx
-
Data_4.xlsx
-
README.md
Abstract
Some birds intentionally ingest stones to facilitate digestion of hard foodstuffs, a behavior inherited from non-avian dinosaurs and present in some of the earliest birds, as evidenced from clusters of gastroliths preserved within the abdominal cavity of a wide range of dinosaurs and Cretaceous birds. For the first time, high-resolution computed laminographic and computed tomographic scans were used to reconstruct the gastral mass in two species of non-neornithine ornithuromorph birds from the Lower Cretaceous Jehol Group. Four specimens of each taxon were analyzed. Preservation of the gastral mass in most of these specimens is in situ and regarded as complete or nearly so. The number of gastroliths, their total volume, and their total mass relative to the estimated body mass were calculated for each specimen. The resultant gastral mass-to-body mass ratios fall within the range observed in extant birds, supporting previous inferences that the digestive system in non-neornithine ornithuromorphs was comparable to that in extant taxa. Compared to available data for non-volant non-avian theropods, the gastral mass is proportionately smaller in birds suggesting that the evolution of flight constrained gastral mass size in the theropod lineage. Currently, available data on gastral mass characteristics suggests that Iteravis ate larger food particles compared to Archaeorhynchus but cannot be utilized to determine diet more precisely. Better understanding of the relationship between gastral mass characteristics and food items across a broader range of extant taxa may provide an indirect but important method through which to infer diet and digestive function in archosaurs.
Methods
Eight ornithuromorph individuals were scanned by computed tomographic (CT) or computed laminographic (CL). After scanning, all data were processed and rendered using the software VG Studio 3.4. Lastly, all related parameters were extracted in VG Studio 3.4 and processed in Microsoft Excel. R was used to demonstrate that all data fell within the confidence interval.