Biological structures with extreme morphologies are puzzling because they often lack obvious functions and stymie comparisons to homologous or analogous features with more typical shapes. An example of such an extreme morphotype is the uniquely modified vertebral column of the hero shrew Scutisorex, which features numerous accessory intervertebral articulations and massively expanded transverse processes. The function of these vertebral structures is unknown, and it is difficult to meaningfully compare them to vertebrae from animals with known behavioral patterns and spinal adaptations. Here we use trabecular bone architecture of vertebral centra and quantitative external vertebral morphology to elucidate the forces that may act on the spine of Scutisorex and that of another large shrew with unmodified vertebrae (Crocidura goliath). µCT scans of thoracolumbar columns show that S. thori is structurally intermediate between C. goliath and S. somereni internally and externally, and both Scutisorex species exhibit trabecular bone characteristics indicative of higher in vivo axial compressive loads than C. goliath. Under compressive load, Scutisorex vertebral morphology is adapted to largely restrict bending to the sagittal plane (flexion). Although these findings do not solve the mystery of how Scutisorex uses its byzantine spine in vivo, our work suggests potentially fruitful new avenues of investigation for learning more about the function of this perplexing structure.

These data were collected by analysis of vertebral trabecular bone regions of interest (ROIs) in the software Quant3D, and measurements of µCT scans of whole vertebrae using Autodesk Meshmixer. For more detail see "Methods" in "Deciphering an extreme morphology: bone microarchitecture of the Hero Shrew backbone (Soricidae: Scutisorex)".

This data set includes two files: "TBA and external measurement data.xls", which is the raw data resulting from the main body of analyses, and "Segmentation sensitivity tests and uncertainty.xls", which is raw data from an analysis of how segmentation protocol affects results of trabecular bone analysis, and raw data used to estimate uncertainty in reported values for TBA metrics. More detailed information on segmentation and uncertainty analyses can be found in the electronic supplementary material associated with this publication.