A comprehensive molecular phylogeny of Afrotropical white-eyes (Aves: Zosteropidae) highlights prior underestimation of mainland diversity and complex colonisation history.

White-eyes (Zosterops) are a hyper-diverse genus of passerine birds that have rapidly radiated across the Afrotropics and Southeast Asia. Despite their broad range, a disproportionately large number of species are currently recognised from islands compared to the mainland. Described species-level diversity of this 'great speciator' from continental Africa-Arabia is strikingly low, despite the vast size and environmental complexity of this region. However, efforts to identify natural groups using traditional approaches have been hindered by the remarkably uniform morphology and plumage of these birds. Here, we investigated the phylogenetic relationships and systematics of Afrotropical Zosterops, including the Gulf of Guinea and western Indian Ocean islands. We included exceptional sampling (~160 individuals) from all except one subspecies of the 55 taxa (32 species, plus 23 additional named sub-species) currently recognized throughout the region, in addition to a subset of extra-Afrotropical taxa, by exploiting blood and archival samples. Employing a multi-locus phylogenetic approach and applying quantitative species delimitation we tested: (1) if there has been a single colonisation event of the Afrotropical realm; (2) if constituent mainland and island birds are monophyletic; and (3) if mainland diversity has been underestimated. Our comprehensive regional phylogeny revealed a single recent colonisation of the Afrotropical realm c.1.30 Ma from Asia, but a subsequent complex colonisation history between constituent island and mainland lineages during their radiation across this vast area. Our findings suggest a significant previous underestimation of continental species diversity and, based on this, we propose a revised taxonomy. Our study highlights the need to densely sample species diversity across ranges, providing key findings for future conservation assessments and establishing a robust framework for evolutionary studies.

Niche divergence promotes rapid diversification of East African sky island white-eyes (Aves: Zosteropidae).

The Eastern Afromontane biodiversity hotspot composed of highly fragmented forested highlands (sky islands) harbours exceptional diversity and endemicity, particularly within birds. To explain their elevated diversity within this region, models founded on niche conservatism have been offered, although detailed phylogeographic studies are limited to a few avian lineages. Here, we focus on the recent songbird genus Zosterops, represented by montane and lowland members, to test the roles of niche conservatism versus niche divergence in the diversification and colonization of East Africa's sky islands. The species-rich white-eyes are a typically homogeneous family with an exceptional colonizing ability, but in contrast to their diversity on oceanic islands, continental diversity is considered depauperate and has been largely neglected. Molecular phylogenetic analysis of ~140 taxa reveals extensive polyphyly among different montane populations of Z. poliogastrus. These larger endemic birds are shown to be more closely related to taxa with divergent habitat types, altitudinal distributions and dispersal abilities than they are to populations of restricted endemics that occur in neighbouring montane forest fragments. This repeated transition between lowland and highland habitats over time demonstrate that diversification of the focal group is explained by niche divergence. Our results also highlight an underestimation of diversity compared to morphological studies that has implications for their taxonomy and conservation. Molecular dating suggests that the spatially extensive African radiation arose exceptionally rapidly (1-2.5 Ma) during the fluctuating Plio-Pleistocene climate, which may have provided the primary driver for lineage diversification.

Divergence times and colonization of the Canary Islands by Gallotia lizards.

The Canary Islands have become a model region for evolutionary studies. We obtained 1.8 Kbp of mtDNA sequence from all known island forms of the endemic lizard genus Gallotia and from its sister taxon Psammodromus in order to reanalyze phylogenetic relationships within the archipelago, estimate lineage divergence times, and reconstruct the colonization history of this group. Well-supported phylogenies were obtained using maximum parsimony and Bayesian inference. Previous studies have been unable to establish the branching pattern at the base of the tree. We found evidence that G. stehlini (Gran Canaria) originated from the most basal Gallotia node and G. atlantica from the subsequent node. Divergence times were estimated under a global clock using Bayesian Markov Chain Monte Carlo methods implemented by three different programs: BEAST, MCMCTREE, MULTIDIVTIME. Node constraints were derived from subaerial island appearance data and were incorporated into the analyses as soft or hard maximal bounds. Posterior node ages differed slightly between programs, possibly due to different priors on divergence times. The most eastern Canary Islands first emerged just over 20 mya and their colonization appears to have taken place relatively quickly, around 17-20 mya. The subsequent node is consistent with cladogenesis due to colonization of Gran Canaria from the eastern islands about 11-13 mya. The western islands appear to have been colonized by a dispersal event from Lanzarote/Fuerteventura in the east to either La Gomera or one of the ancient edifices that subsequently formed Tenerife in the west, about 9-10 mya. Within the western islands, the most recent node that is ancestral to both the G. intermedia/G. gomerana/G. simonyi and the G.galloti/G. caesaris clades is dated at about 5-6 mya. Subsequent dispersal events between ancient Tenerife islands and La Gomera are dated at around 3 mya in both clades, although the direction of dispersal cannot be determined. Finally, we show that G. galloti is likely to have colonized La Palma more than 0.5 Ma after emergence of the island 1.77 mya, while G. caesaris from the same clade may have colonized El Hierro very soon after it emerged 1.12 mya. There are tentative indications that the large-bodied endangered G. simonyi colonized El Hierro around the same time or even later than the smaller-bodied G. caesaris. This study demonstrates the effectiveness of Bayesian dating of a phylogeny in helping reconstruct the historical pattern of dispersal across an oceanic archipelago.