Dental measures of extant lizards and fossil reptiles
Data files
Nov 12, 2025 version files 57.16 KB
-
code_LDA.R
7.94 KB
-
code_phylogenetic_MANOVA.R
1.09 KB
-
extantlizard_avgs.csv
5.69 KB
-
procolophonid_avgs.csv
1.11 KB
-
README.md
6.57 KB
-
Supplemental_Table_1_lizard_and_proco_SVLs_headlengthdata.csv
30.59 KB
-
ultrametric_lizardtree.newick
4.18 KB
Abstract
Procolophonoids are hypothesized to be some of the first highly specialized herbivores to evolve in the aftermath of the end-Permian extinction. Hypotheses of procolophonoid diet are based on their unique dentitions: bulbous, multicuspid, differentiated teeth, specialized for feeding on fibrous plant matter. However, these conclusions are largely based on qualitative descriptions, and there has yet to be a broadscale, quantitative comparison of this group of fossil reptiles to modern ones. In our study, we use relatively simple measures of tooth shape and linear discriminant analysis (LDA) to predict dietary ecology from tooth dimensions and phylogenetic MANOVA to test for significant differences in tooth shape for different diets. We find that procolophonids are largely predicted as herbivorous but occupy different ecological niches from modern lizards and that simple methods return relatively similar results as complex methods. We hypothesize that the different mechanical challenges of Triassic-age flora produced a different 'kind' of herbivorous tooth from those observed in extant lizards, like those in procolophonids.
Dataset DOI: 10.5061/dryad.9kd51c5x8
Description of the data and file structure
This dataset contains the data and analytical code underlying a comparative study of dental and cranial morphology in a group of fossil reptiles (procolophonids and close kin) and extant lizards. Our dataset includes averaged datasets of dental measures for fossil procolophonids (procolophonid_avgs.csv) and living lizards (extantlizard_avgs.csv), an ultrametric phylogenetic tree of sampled lizard species (ultrametric_lizardtree.newick), two annotated R scripts implementing linear discriminant analyses and phylogenetic MANOVAs (code_LDA.R, code_phylogenetic_MANOVA.R), and a supplementary figure visualizing the relationship between head length and body size (MartinezandJenkins_SuppFigure1.tiff).
We build a predictive model based on living lizards to quantitatively predict the diets of procolophonids using relatively simple measures of tooth shape and size. In our study, we use linear discriminant analysis (LDA) to separate differences in our extant data and use that to predict diet in fossils taxa. We also use phylogenetic multivariate analysis of variance (MANOVA) to understand the extent to which shared evolutionary history guides the differences between groups.
For 91 species of extant lizards, we measured and calculated species averages for the following dental measures using the last tooth in the tooth row: labiolingual width (W), mesiodistal length (L), tooth area (A), and length/width ratio (R). A well-known proxy for body size in reptiles, snout-vent length (SVL), was also included for extant lizards (obtained from the literature). Head length (HL) was also collected (obtained from specimens measured). Measures of size (L, W, A, HL, and SVL) were log-transformed before being incorporated into analyses. The same dental measures were collected for 24 fossil specimens (23 procolophonoids + 1 nyctiphruretid). Because complete specimens are rare, SVL was not feasible to collect for fossil taxa. HL was collected instead (obtained from specimens measured or literature). A regression analysis of log-transformed HL on log-transformed SVL was performed to ensure that HL is a good predictor of overall body size for our sample.
Files and variables
File: extantlizard_avgs.csv
Description: Contains mean linear and area-based measurements for extant lizard species. Each row represents an individual species average derived from multiple specimens.
Variables
- Species: taxonomic identification (genus and species).
- Diet: categorical descriptor of predominant dietary type (e.g., carnivorous, omnivorous, herbivorous); (based on Meiri, 2024).
- Length: average mesiodistal length of the last tooth in the tooth row (millimeters, mm).
- Width: average labiolingual width of the last tooth in the tooth row (mm).
- Ratio: ratio of mean tooth area to head length.
- Area: calculated average surface area of individual teeth (mm²).
- SVL: snout–vent length (mm), measured from snout tip to anterior margin of the cloaca.
- Head_length: total head length measured from snout to occiput (mm).
File: procolophonid_avgs.csv
Description: Contains averaged cranial and dental measurements for fossil* *specimens. Each row represents the mean values for a species or individual specimen. Measurements include head length, tooth dimensions, and derived ratios.
Variables
- Species: taxonomic identification of the specimen (genus and species, when available).
- Length: average mesiodistal length of the last tooth in the tooth row (millimeters, mm).
- Width: average labiolingual width of the last tooth in the tooth row (mm).
- Ratio: ratio of mean tooth area to head length.
- Area: calculated average surface area of individual teeth (mm²).
- Head_length: total head length measured from snout to occiput (mm).
File: ultrametric_lizardtree.newick
Description: Time-calibrated ultrametric phylogenetic tree of extant lizard species included in the dataset extantlizard_avgs.csv. The tree topology is based on Pyron et al. (2013) and has been pruned to include only taxa sampled in this study.
Purpose: Phylogeny for the phylogenetic MANOVA described in the manuscript.
Associated files:
extantlizard_avgs.csv— extant lizard morphometric dataset.code_phylogenetic_MANOVA.R— R script implementing the phylogenetic MANOVA using this tree.
File: MartinezandJenkins_SuppFigure1.tiff
Description: Scatterplot of log-transformed head length (HL) versus snout–vent length (SVL) for extant lizard species, with fitted regression line, 95 % confidence intervals, and taxonomic group labels.
Purpose: To validate head length as a robust predictor of body size (SVL) and justify its use in cross-taxonomic comparisons involving fossil taxa.
Associated datasets: extantlizard_avgs.csv and procolophonid_avgs.csv.
Usage: This figure serves as a methodological reference for interpreting head-length-based scaling analyses. It may be cited alongside the associated datasets to illustrate the allometric relationship between cranial and body size variables.
File: Supplemental_Table_1_lizard_and_proco_SVLs_headlengthdata.csv
Description: Snout–vent length (SVL) and head length (HL) data for both extant lizard species and fossil specimens. Each row represents a specimen or species mean, with accompanying identifiers and measurement metadata. Data were compiled to assess scaling relationships between head length and overall body size across living and extinct reptiles. This supplemental table includes the sources for SVL and HL data for each taxon included in the dataset.
Code/software
code_phylogenetic_MANOVA.R
This file conducts the phylogenetic multivariate analysis of variance described in the manuscript. This file uses ultrametric_lizardtree.newick as the phylogenetic tree in analysis.
Language and Environment
R Environment for Statistical Computing
Version
4.4.3
Dependencies
- geiger
- ape
- phytools
- maps
code_LDA.R
This file performs the linear discriminant analyses (LDAs) described in the manuscript. This file uses extantlizards_avgs.csv and procolophonid_avgs.csv as the datasets for the LDAs.
Language and Environment
R Environment for Statistical Computing
Version
4.4.3
Dependencies
- ggplot2
- MASS
- cowplot
- dplyr