Conserved ancestral tropical niche but different continental histories explain the latitudinal diversity gradient in brush-footed butterflies

The global increase in species richness toward the tropics across continents and taxonomic groups, referred to as the latitudinal diversity gradient, stimulated the formulation of many hypotheses to explain the underlying mechanisms of this pattern. We evaluate several of these hypotheses to explain spatial diversity patterns in a butterfly family, the Nymphalidae, by assessing the contributions of speciation, extinction, and dispersal, and also the extent to which these processes differ among regions at the same latitude. We generate a time-calibrated phylogeny containing 2,866 nymphalid species (~45% of extant diversity). Neither speciation nor extinction rate variations consistently explain the latitudinal diversity gradient among regions because temporal diversification dynamics differ greatly across longitude. The Neotropical diversity results from low extinction rates, not high speciation rates, and biotic interchanges with other regions are rare. Southeast Asia is also characterized by a low speciation rate but, unlike the Neotropics, is the main source of dispersal events through time. Our results suggest that global climate change throughout the Cenozoic, combined with tropical niche conservatism, played a major role in generating the modern latitudinal diversity gradient of nymphalid butterflies.F.L.C is supported by an “Investissements d’Avenir” grant managed by the Agence Nationale de la Recherche (CEBA, ref. ANR-10-LABX-25-01). D.J.L. was supported by grants DEB-1541557 from NSF and WW-227R-17 from the National Geographic Society. R.I.V.W. was supported by Leverhulme Trust emeritus programme. M.E. was supported by an ATIP grant, a grant from the Human Frontier Science Program (RGP0014/2016) and a grant from the French National Research Agency (ANR CLEARWING ANR-16-CE02-0012). E.O.A. was supported by Sigma-Xi (G20100315153261), Center for Systematic Entomology and the Council of the Linnean Society and the Systematics Association for the Systematics Research Fund. S.N. was supported by the Swedish Research Council (2015-04218 and 2019-03441). P.M.M. was supported by a Czech Science Foundation grant (Junior GAČR, 20-18566Y). R.V. was supported by projects PID2019-107078GB-I00 / AEI / 10.13039/501100011033 (Agencia Estatal de Investigación) and 2017-SGR-991 (Generalitat de Catalunya). N.W. acknowledges support by the Swedish Research Council (2015-04441) and a start-up grant from the Department of Biology, Lund University