Fig 1.
Sampling of each extant bat family within this repository.
On the left, all twenty extant families of bats are displayed [24] along the phylogenetic backbone of the order [31]. On the right, estimated total richness (background, white bars) and the repository richness (filled, dark grey sections) are depicted for each family. Note that the axis for species richness is broken between 200 and 400 due to the high richness of extant vespertilionids.
Fig 2.
Phylogeny of species included in this repository.
The phylogeny of bats included in this repository, with the most well-sampled families labeled. We include some examples of cranial surfaces from this repository for these labeled clades, to showcase the breadth of morphological disparity contained in this database, and its link to trophic diversity. Skulls are not to scale, as these species are of considerably different sizes. Species are listed as follows from top to bottom, with their associated specimen identification information from the University of Michigan’s Museum of Zoology (UMMZ) or the American Museum of Natural History (AMNH): Pteropus hypomelanus (UMMZ 130417; frugivore), Hipposideros abae (AMNH 49120; insectivore), Desmodus rotundus (UMMZ 116246; sanguivore), Glossophaga leachii (AMNH 185970; nectarivore), Molossus molossus (AMNH 78895; insectivore), Myotis macrotarsus (UMMZ 160308; insectivore).
Fig 3.
Examples of TIFF slices, a representative PLY surface, and a landmarked sample.
For a specimen of Artibeus aztecus (UMMZ 110526), we have included two examples of individual TIFF files (left), and two analogous views of the PLY surface file (right) with labels for the fixed landmarking scheme of Santana & Lofgren [26] that we used to test for convergence in our example analysis. Full details on these landmarks can be found in that original publication. This species is generally considered a frugivore, with opportunistic consumption of insects [36].
Fig 4.
Relationships between measurements taken from digital and physical specimens.
For each of 9 linear measurements, we present the relationship between measurements taken using calipers on physical specimens versus measurements taken on the surface of digital specimens, across 20 different bat species. A dashed 1:1 line is included for each measurement. Differences between the two methods are minimal and appear random with respect to species, with all R2 values rounding to 0.99 and 1.00 depending on precision (all p < 0.05).
Table 1.
C1 statistic for skull shape among major trophic categories of noctilionoid bats.