Testing three hypotheses of rarity among the Buprestidae species of a tropical dry forest

Many insect communities have a higher proportion of rare species than predicted by community-structure models. Rarity has been associated with increased extinction risk, therefore, it is essential to understand which factors cause rarity patterns. The importance of three rarity hypotheses (under-sampling, diffusive rarity, and body size) in a Buprestidae community sampled with active and passive methods are tested in a tropical dry forest in central Mexico. In total, 222 species were recorded. Doubletons and duplicates decreased with the increase of sampled locations, while singletons and uniques increased. At the local level, 67% of the species were no longer rare when all collection methods were considered (n = 60 species), as well as 41% at the regional level (n = 41 species). The diffusive rarity hypothesis was tested in 97 species that appeared in more than two localities; 38% of the species changed their abundance status between localities. The body size hypothesis was tested considering 179 species, and rare species were smaller only in the species of the genus Chrysobothris. More than one rarity hypothesis operated simultaneously in a community, but their contribution was different. In the Buprestidae community studied, the under-sampling hypothesis had the greatest support, affecting a higher proportion of species. Subsequently, the second most important was the diffusive rarity hypothesis, and the least support showed the body size hypothesis, as the size was determined mainly by the taxonomic genus.