ABSTRACT
This paper explores the distributional data of 4,224 Tipulidae (Insecta: Diptera) species to search for endemism patterns in a worldwide scale and to test the extent to which the global patterns of endemism of the group fit into previously proposed regionalization schemes, particularly Wallace's system and recent revisions of it. Large scale areas of endemism are assessed using the grid-based method implemented in VNDM. VNDM depends on the prior definition of the grid size for analysis, but a criterion for choosing beforehand a particular grid size is not clear. The same holds for the choice of the level of similarity in species composition selected for the calculation of consensus areas. In our study, we developed a methodological approach that helped defining objective criteria for choosing suitable values for these critical variables. Large-scale areas of endemism around the globe are identified and ranked according to endemicity levels: 1--West Palaearctic, 2--Nearctic, 3--East Palaearctic-Oriental, 4--West North America, 5--Australia, 6--Neotropical, 7--Sub-Saharan Africa, 8--Palaearctic, and 9--Middle East. Our main conclusion is that there are still some limitations in applying biogeographical classifications proposed mostly on the basis of vertebrate distribution to other taxonomic groups, such as the Tipulidae. While there is a general congruence of the broad-scale areas of endemism of tipulids with previously proposed regionalization schemes, for some areas, the sharpness of boundaries between traditional regions is not so acute, due to a great level of overlap of part of its biotic elements.
