Crypsis and convergence: integrative taxonomic revision of the Gehyra australis group (Squamata: Gekkonidae) from northern Australia

PeerJ. 2020 Jan 27:8:e7971. doi: 10.7717/peerj.7971. eCollection 2020.

Abstract

For over two decades, assessments of geographic variation in mtDNA and small numbers of nuclear loci have revealed morphologically similar, but genetically divergent, intraspecific lineages in lizards from around the world. Subsequent morphological analyses often find subtle corresponding diagnostic characters to support the distinctiveness of lineages, but occasionally do not. In recent years it has become increasingly possible to survey geographic variation by sequencing thousands of loci, enabling more rigorous assessment of species boundaries across morphologically similar lineages. Here we take this approach, adding new, geographically extensive SNP data to existing mtDNA and exon capture datasets for the Gehyra australis and G. koira species complexes of gecko from northern Australia. The combination of exon-based phylogenetics with dense spatial sampling of mitochondrial DNA sequencing, SNP-based tests for introgression at lineage boundaries and newly-collected morphological evidence supports the recognition of nine species, six of which are newly described here. Detection of discrete genetic clusters using new SNP data was especially convincing where candidate taxa were continuously sampled across their distributions up to and across geographic boundaries with analyses revealing no admixture. Some species defined herein appear to be truly cryptic, showing little, if any, diagnostic morphological variation. As these SNP-based approaches are progressively applied, and with all due conservatism, we can expect to see a substantial improvement in our ability to delineate and name cryptic species, especially in taxa for which previous approaches have struggled to resolve taxonomic boundaries.

Keywords: Australian monsoonal tropics; Cryptic species; Endemism; Hybridisation; Lizard; New species.

Grants and funding

This research was supported by grants from from the Australian Research Council to Paul Oliver (ARC LP120100081 and ARC DE140100220) and Craig Moritz (ARC FL110100104). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.