Back

Ophidian physique: the capacity of middle trunk vertebral shape for quantitative taxonomic delimitation in snakes

Jacisin, J. J.; Lawing, A. M.

2023-08-03 paleontology
10.1101/2023.08.01.551561 bioRxiv
Show abstract

Skeletal specializations in snakes have resulted in incredible locomotive adaptability, including oft-overlooked vertebral complexity. Snake vertebrae are usually identified via qualitative descriptions of morphological traits; however, identifying and describing snake vertebrae in a scientifically replicable way has long hindered fossil snake research, where attempts to identify between or within snake groups can be onerous. Here, we build a framework of extant snake middle trunk vertebral shape using 2D geometric morphometrics (GMM) and quantitative methods to explore the viability of using these tools to assign and identify snake trunk vertebrae taxonomically and ecologically. We use 23 landmarks to evaluate anterior vertebral shape variation in 504 snake trunk vertebrae representing 189 species across 11 families for delimiting taxonomy and primary foraging habitat. We found that snake vertebral shape variation of overall proportions and articular surfaces contained statistically significant taxonomic and ecomorphological information useful for group assignment. Differences in primary foraging habitat also resulted in similar morphological trends within taxonomic groups in shape space. We then used linear discriminant functions to test the reliability of taxonomic assignments based on the shape captured by our landmark scheme. Analysis of the full dataset had high overall accuracy for family and subfamily, but only moderate success for genus, species, and primary foraging habitat. When applied to a single subfamily, overall accuracy greatly increased for genus and primary foraging habitat, implying that iterative application of this method may improve results. This study presents the framework for a new replicable method to supplement qualitative morphological descriptions of taxa. We recommend that GMM is best employed alongside qualitative descriptions for the optimal and reproducible delimitation of snake vertebrae. Finally, this method will allow non-expert diagnosticians to have more confidence in identifying fossil snake vertebrae, helping increase the number of snake fossils identified in museum collections.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.