Historical shifts, geographic biases, and biological constraints shape mammal species discovery

This repository contains the data and code used in the manuscript "Historical shifts, geographic biases, and biological constraints shape mammal species descriptions". In this study, we explore how biological traits, geographic patterns, and taxonomic practices influence the way mammal species have been described from 1990 to 2025. We analyze trends in the comprehensiveness of species descriptions, the use of evidence types (e.g., molecular, morphological), authorship patterns, and potential phylogenetic structure in these patterns. All analyses were conducted in R and are fully reproducible through the materials provided here. File structure The repository contains six files: Rcode_MammDescTrends.R – R script used for running the analyses and creating the figures. The script is organized into 10 main sections, each addressing a key part of the study: Load and understand the dataset Map mammal species described in the last three decades Temporal trends in robustness of publications – based on annual means Publication robustness by mammal order Temporal trends in robustness of publications – based on GLMs Create plots with model coefficients and confidence intervals Check phylogenetic correlation in model residuals Make phylogenetic correlograms Explore temporal trends in the use of molecular data on mammal description Relationship between international description and taxonomy practices phylogeny/output.nex – Phylogenies used to generate the correlograms in step 8. The file represents 100 fully-sampled phylogenies for mammals (Upham et al., 2019 PLoS Biology). wwf_simplified.zip – contains simplified shapefiles of the biogeographical realm limits extracted from https://ecoregions.appspot.com/. You will need a ZIP extractor to extract the files first. world_limit.zip – contains a shapefile used to add the limits around the map in figure 1. You will need a ZIP extractor to extract the files first. Dataset.RData – contains 30 columns, each explained below. mammals_taxonomy.csv – contains the same information as Dataset.RData, but in a more common file (.csv), allowing easy exploration of the data via softwares such as Excel. Column Name Description SpeciesName Binomial name. Genus Taxonomic genus to which a species belongs. Family Taxonomic family to which a species belongs. Order Taxonomic order to which a species belongs ("Rodentia", "Chiroptera", "Primates", etc.). Authority Author(s) involved in the species description. Year Year in which the species was formally described. TaxonomicReview Indicates if the description was based on a taxonomic review (0 = No, 1 = Yes). N_authors Number of authors per description. N.Countries Number of countries (based on authors' affiliations) involved in the description. Log10BodyMass_g Maximum body mass per species (log10-transformed). SppRichPerGenus Per-genus species richness based on the year of each species' description. Morphometrics Number of morphometric measurements used in the description. Osteology Number of osteological measurements used in the description. Dentition Binary variable indicating whether dentition data was provided in the description. InternalAnatomy Binary variable indicating whether data on internal anatomy was provided in the description. ShapeDescription Binary variable indicating whether any aspects of the species' shape were described. Trichology Binary variable indicating whether trichology data was provided in the description. Coloration Binary variable indicating whether color data was provided in the description. Karyotype Binary variable indicating whether karyotype data was provided in the description. Molecular Binary variable indicating whether the authors used molecular data in the description. MolMethod Molecular method used in the description (e.g., mtDNA, nucDNA, multiLoci, SNPs). N.Genes Total number of genes sequenced when molecular data was used. N.Specimens Number of specimens of the new species used in the description. TaxaComparedExamined Number of taxa analyzed/inspected for comparisons with the new species. TaxaCompared Number of taxa mentioned in the text during comparisons with the new species. N.Pages Number of pages (METHODS and RESULTS sections only) of the article divided by the number of described species. One page consists of 4x0.25 parts. N_evidences Number of evidence types used in descriptions. All variables are treated as binary (0 or 1), giving them equal weight. Latitude & Longitude Species type-locality coordinates. CollectionIsAuthor Binary variable indicating whether the collector of the holotype is the author of the description manuscript. Usage notes R 4.4.1 software is required to open a the Rcode_MammDescTrends.R and Dataset.RData files, as well as to read the phylogenetic trees available in the phylogeny/output.nex file. Microsoft Excel (or similar softwares) can be used to explore the csv. file.