Predicting body length and assessing the shape of tail-propelled Mesozoic marine reptiles

Body length is a crucial ecological predictor in vertebrates, yet total body length proxies have seldom been assessed for ancient marine top predators. Here, we test the strength of phylogenetic imputation and 23 linear measurements, sampling both broad skeletal regions and frequently fossilised elements (such as vertebral centra), in predicting the body length of the main clades of tail-propelled Mesozoic marine reptiles (Ichthyosauria, Mosasauridae, and pelagic thalattosuchians). We embed this marine reptile sample within a comparative framework with raptorial cetaceans, and analyse the evolution of body proportions in these clades. We find that trunk length and centrum dimensions are strong predictors of body length, opening up the possibility to build vast datasets of body length estimations for Mesozoic marine reptiles from minimal preserved remains. We provide body length calculation equations for all traits and all clades. Proxies fared much better and often had distinct slopes w..., , # Predicting body length and assessing the shape of tail-propelled Mesozoic marine reptiles

Dataset DOI: [10.5061/dryad.4xgxd25nh](10.5061/dryad.4xgxd25nh)

## Description of the data and file structure

The .zip file (SuppInfo.zip) is a compressed folder which can be uncompressed with software like winzip, winrar, the unarchiver. It contains:

* The raw data (linear measurements in mm of total bodies and body parts, per species, in an Excel file: `SuppData1.xslx`). This is an Excel file that can be opened with Microsoft Excel, Open Office, Google Sheets, or R.
* The phylogenetic data trees (one dataset for each focal clade, in the .tre format: `SuppData2_Cetacea.tre` [Cetacea], `SuppData3_Ichthyosauria.tre` [Ichthyosauria], `SuppData4_Mosasauridae.tre` [Mosasauridae], `SuppData5_Thalattosuchia.tre` [Thalattosuchia]). These are files containing information on the phylogenetic relationships of the species of interest. Such a file can be opened with Mesquite, R. These data are dimensionl...,