Many species, including humans, have emerged via complex reticulate processes involving hybridisation. Under certain circumstances, hybridisation can cause distinct lineages to collapse into a single lineage with an admixed mosaic genome. Most known cases of such ‘speciation reversal’ or ‘lineage fusion’ involve recently diverged lineages and anthropogenic perturbation. Here, we show that in western North America, Common Ravens (Corvus corax) have admixed mosaic genomes formed by the fusion of non-sister lineages (‘California’ and ‘Holarctic’) that diverged ~1.5 million years ago. Phylogenomic analyses and concordant patterns of geographic structuring in mtDNA, genome-wide SNPs and nuclear introns demonstrate long-term admixture and random interbreeding between the non-sister lineages. In contrast, our genomic data support reproductive isolation between Common Ravens and Chihuahuan Ravens (C. cryptoleucus) despite extensive geographic overlap and a sister relationship between Chihuahuan Ravens and the California lineage. These data suggest that the Common Raven genome was formed by secondary lineage fusion and most likely represents a case of ancient speciation reversal that occurred without anthropogenic causes.

A model of range expansions during glacial maxima (GM) for cold-adapted species is generally accepted for the Northern Hemisphere. Given that GM in Australia largely resulted in the expansion of arid zones, rather than glaciation, it could be expected that arid-adapted species might have had expanded ranges at GM, as cold-adapted species did in the Northern Hemisphere. For Australian biota, however, it remains paradigmatic that arid-adapted species contracted to refugia at GM. Here we use multilocus data and ecological niche models (ENMs) to test alternative GM models for butcherbirds. ENMs, mtDNA and estimates of nuclear introgression and past population sizes support a model of GM expansion in the arid-tolerant Grey Butcherbird that resulted in secondary contact with its close relative--the savanna-inhabiting Silver-backed Butcherbird--whose contemporary distribution is widely separated. Together, these data reject the universal use of a GM contraction model for Australia's dry woodland and arid biota.

Changes in climate and sea level are hypothesized to have promoted the diversification of biota in monsoonal Australia and New Guinea by causing repeated range disjunctions and restricting gene flow between isolated populations. Using a multilocus (one mtDNA locus, five nuclear introns) phylogeographic approach, we test whether populations of the mangrove and rainforest restricted Black Butcherbird (Cracticus quoyi) have diverged across several geographic barriers defined a priori for this region. Phylogeographic structure and estimates of divergence times revealed Plio-Pleistocene divergences and long-term restricted gene flow of populations on either side of four major geographic barriers between and within Australia and New Guinea. Overall, our data are consistent with the hypothesis that mesic-adapted species did not disperse across the open dry woodlands and grasslands that dominated the transient palaeo-landbridges during the Plio-Pleistocene despite the presence of mangrove forests that might have acted as dispersal corridors for mesic-adapted species. Our study offers one of the first multilocus perspectives on the impact of changes in climate and sea level on the population history of widespread species with disjunct ranges in Australia and New Guinea.

The phylogeographic structure of the widely distributed arid and semi‐arid Australian splendid fairy‐wren Malurus splendens was investigated by using variation in plumage characters and mitochondrial DNA (mtDNA). We examined sequences of the mtDNA ND2 gene and used spectrophotometry to quantify chromatic variation in plumage in order to test the current morphology‐based intraspecific taxonomy of M. splendens and to discriminate between hypotheses invoking allopatric and parapatric processes in the origin of diversity in the complex. Genetic diversity of M. splendens fell into three divergent geographically structured clades. One represents populations ascribed to the western subspecies M. s. splendens, the other populations of central M. s. musgravi and the third all eastern populations currently ascribed to M. s. emmottorum and M. s. melanotus. Plumage patterns clearly differentiate M. s. splendens and M. s. musgravi, and spectrophotometry identified a step‐wise transition in spectra between M. s. melanotus and M. s. emmottorum. Congruence of patterns of phenotypic and genetic variation among western, central and eastern populations of M. splendens strongly suggests that these populations have diverged in allopatry on either side of historical biogeographic barriers in this region. Decoupled patterns of phenotypic and genetic diversity suggest that the divergence of M. s. melanotus and M. s. emmottorum may have occurred without periods of isolation perhaps in response to differences in local environmental conditions, or alternatively, mtDNA and plumage may have different rates of evolution. Critically, we encountered issues with the placement of the root of the M. splendens complex. The root was placed within the subspecies M. s. splendens separating its northern and southern populations and rendering the subspecies paraphyletic.

No abstract

No abstract

No abstract

Previous studies based on single mitochondrial markers have shown that the common raven (Corvus corax) consists of two highly diverged lineages that are hypothesised to have undergone speciation reversal upon secondary contact. Furthermore, common ravens are paraphyletic with respect to the Chihuahuan raven (C. cryptoleucus) based on mitochondrial DNA (mtDNA). Here we explore the causes of mtDNA paraphyly by sequencing whole mitochondrial genomes of 12 common ravens from across the Northern Hemisphere, in addition to three Chihuahuan ravens and one closely related brown-necked raven (C. ruficollis) using a long-range PCR protocol. Our raven mitogenomes ranged between 16925–16928 bp in length. GC content varied from 43.3% to 43.8% and the 13 protein coding genes, two rRNAs and 22 tRNAs followed a standard avian mitochondrial arrangement. The overall divergence between the two common raven clades was 3% (range 0.3–5.8% in 16 regions including the protein coding genes, rRNAs and the control region). Phylogenies constructed from whole mitogenomes recovered the previously found mitochondrial sister relationship between the common raven California clade and the Chihuahuan raven (overall divergence 1.1%), which strengthens the hypothesis that mtDNA paraphyly in the common raven results from speciation reversal of previously distinct Holarctic and California lineages.