Evolution of the latitudinal diversity gradient in the hyperdiverse ant genus Pheidole

Aim The latitudinal diversity gradient is the dominant geographic pattern of life on Earth, but a consensus understanding of its origins has remained elusive. The analysis of recently diverged, hyper-rich invertebrate groups provides an opportunity to investigate latitudinal patterns with the statistical power of large trees while minimizing potentially confounding variation in ecology and history. Here, we synthesize global phylogenetic and macroecological data on a hyperdiverse (> 1,100 species) ant radiation, Pheidole and test predictions of three general explanations for the latitudinal gradient: variation in diversification rates, tropical conservatism and ecological regulation. Location Global. Time period The past 35 million years. Major taxa studied The hyperdiverse ant genus Pheidole Westwood. Methods We assembled geographic data for 1,499 species and morphospecies, and inferred a dated phylogeny for 449 species of Pheidole, including 167 species newly sequenced for this study. We tested for correlations between diversification rate and latitude with Bayesian analysis of macroevolutionary mixtures (BAMM), hidden state speciation and extinction (HiSSE), geographic state speciation and extinction (GeoSSE), and a non-parametric method (FiSSE), evaluated evidence for richness steady state, and examined patterns of diversification as Pheidole spread around the globe. Results There was no evidence of systematic variation of net diversification rates with latitude across any of the methods. We found that Pheidole diversification occurred in bursts when new continents were colonized, followed by a slowdown in each region, but there is no evidence richness has saturated at an equilibrium in any region. Additionally, we found latitudinal affinity is moderately conserved with a Neotropical ancestor and simulations show that phylogenetic inertia alone is sufficient to produce the gradient pattern. Main conclusions Our results provide no evidence that diversification rates vary systematically with latitude. Richness is far from steady state in each region, contrary to the ecological regulation hypothesis, although there is evidence that ecological opportunity promotes diversification after colonization of new areas. The fact that niche conservatism is strong enough to produce the gradient pattern is in accord with the tropical conservatism hypothesis. Overall, these results shed light on the mechanisms underlying the emergence of the diversity gradient within the past 34 million years, complementing recent work on deeper time-scales, and more generally contribute toward much-needed invertebrate perspective on global biodiversity dynamics.