Evaluation of DNA from historical specimens and modern museum samples has enabled an untangling of the complex evolutionary history of four species of rock-wallabies, which are distributed across the Kimberley and Top End. By generating genomic data from museum specimens, we were able to identify episodes of gene flow among these species, despite major differences in their size, ecology and chromosome number. This historic gene flow (introgression) has resulted in unique and previously puzzling patterns of relationship. This study highlights that species boundaries can be fluid and complex; shifting in space and time, with genetic mixing between species more common than previously thought.

Figure 1. The distribution of Wilkins’ Rock-wallaby, Petrogale wilkinsi in the Top End of the Northern Territory, overlaps with that of a smaller species the Nabarlek, Petrogale concinna
Figure 1. The distribution of Wilkins’ Rock-wallaby, (Petrogale wilkinsi) in the Top End of the Northern Territory, overlaps with that of a smaller species the Nabarlek, (Petrogale concinna). Image: Emily Miller
© Emily Miller


An iconic group of Australian marsupials, rock-wallabies (Petrogale) form part of the kangaroo and wallaby family. They comprise 17 known species, that inhabit diverse environments across mainland Australia. They have unique biological features, including a preference for inhabiting complex rocky habitats. This often results in species being isolated to rocky ranges across the continent.

The brachyotis group of rock-wallabies from north-western Australia, are unique in the genus as they are the only species with overlapping (sympatric) distributions. They also display the most size variation with two species, the Monjon (Petrogale burbidgei) and Nabarlek (P. concinna) being 1-2 kg compared to the much larger (4-6 kg) western short-eared rock-wallaby (P. brachyotis) and Wilkins’ rock-wallaby (P. wilkinsi). Most of these species are widespread across the Australian Monsoonal Tropics and have been influenced by changing climatic conditions across deep timescales, causing repeated expansion and contractions of their distributions. This has resulted in genetically divergent populations occurring within each of these species across northern Australia.

Figure 2. Historic specimens of rock-wallabies from north-western Australia, housed in the Australian National Wildlife Collection, Canberra, are sampled for genomic analysis.  Photo by Mark Eldridge. Copyright Australian Museum.
Figure 2. Historic specimens of rock-wallabies from north-western Australia, housed in the Australian National Wildlife Collection, Canberra, are sampled for genomic analysis. Image: Mark Eldridge
© Australian Museum

The mystery

Previous genetic studies have shown puzzling patterns of inter-relationships, where populations of the same species appeared not to be closely related. These studies had used relatively short segments of mitochondrial DNA (mtDNA), which is maternally inherited (passed down from the mother only) and has been commonly used by researchers to identify species through DNA barcoding and to establish relationships among populations and species. But while mtDNA is very useful, it doesn’t always tell the whole story.

The solution

Now with improved genomic technologies, for the first time we’ve been able to re-examine relationships in this group by looking at almost a thousand genes (totally almost 1 million base pairs) from across the nuclear genome (inherited from both parents) and to compare with the patterns detected with the mtDNA. We were also able to fill critical geographic sampling gaps by utilising DNA obtained from historic museum specimens from collections across Australia. Analysis of the nuclear genome (nDNA) resolved the expected relationships amongst these species, grouping species with similar morphology and chromosome numbers together. However, an analysis of the mtDNA continued to show a different and more complex pattern of relationships.

The elusive Nabarlek, Petrogale concinna from John Gould's Mammals of Australia
Figure 3. Although the elusive Nabarlek, (Petrogale concinna) is significantly smaller than the sympatric Short-eared Rock-wallaby, (Petrogale brachyotis) we have found evidence of historical gene flow between them in the Kimberley, Western Australia. Image: John Gould 1863, Mammals of Australia.
© Public Domain

The surprise

Unexpectedly, we found evidence of historic gene flow between species of very different sizes, ecologies and with different chromosome configurations. Some of this gene flow appears to have resulted in mtDNA moving from one species to populations of another species, so examining just the mtDNA gives a different pattern to nuclear DNA (nDNA). This discordance between the patterns of relationships established from mtDNA and nDNA is being increasingly detected in diverse species, indicating the species boundaries are more permeable to gene flow than previously thought. It is also a cautionary tale that simply relying of mtDNA to identify species (i.e., DNA barcoding) to establish relationships can sometimes be fraught.

Dr Sally Potter

Macquarie University and Research Associate, Australian Museum Research Institute

Dr Mark Eldridge

Senior Principal Research Scientist, Terrestrial Vertebrates, Australian Museum Research Institute

More information:

  • Potter, S., Moritz, C., Piggott, M.P., Bragg, J.G., Silva, A.C.A., Bi, K., Turakulov, R., and Eldridge, M.D.B. 2024. Museum skins enable identification of introgression associated with cytonuclear discordance. Systematic Biology https://doi.org/10.1093/sysbio/syae016