Museum genetics solves 88 year-old tree-kangaroo puzzle

An examination of DNA extracted from tree-kangaroo specimens in the Australian Museum collection has confirmed that the mysterious Dendrolagus deltae, described as a new species from southern New Guinea in 1936, is not a valid species but the result of some erroneous locality information.

The wonderfully improbable tree-kangaroos (Dendrolagus) are a unique genus of arboreal macropodids that are confined to the wet tropical forests of New Guinea and north-east Queensland (Fig. 1). They are amongst the most poorly known large (8-20 kg) marsupials, and their taxonomy has remained largely unresolved until recently. Currently two living species are recognised from Australia and ten in New Guinea.

One species, Dendrolagus deltae, has endured ongoing uncertainty ever since it was described 88 years ago under curious circumstances. In 1936, the Australian Museum’s Mammal Curator, Ellis Troughton published a paper describing two new species of tree-kangaroo from New Guinea, with co-author A.S. Le Souef, from Taronga Zoological Gardens, Sydney. The first was Dendrolagus spadix (Lowland Tree-kangaroo) based on a wild-collected flat skin from the lowlands of Papua New Guinea. The second, Dendrolagus deltae was based on the skins and skulls of two captive animals from Taronga Zoological Gardens which were reported to have come from Mt Pratt in south-western New Guinea (Fig. 2).

Subsequently, most mammalogists, noting the similarity between the two D. deltae specimens and specimens of Dendrolagus matschiei (Huon Tree-kangaroo) from the mountains of the Huon Peninsula, north-east New Guinea, thought that the collection locality (Mt Pratt) was erroneous and that D. deltae was not a distinct species. Indeed, Troughton himself also came to harbor doubts about the validity of D. deltae as a taxon and the published collection locality. In a type-written note (undated) added to the first page of a reprint of his 1936 paper (held in the Mammal section at the Australian Museum) Troughton states “Note: Dendrolagus spadix is recognized as valid, but D. deltae remains uncertain owing to the doubtful locality provided for Zoo specimens by co-author Le Souef”. To further muddy the waters, the Mt Pratt area, like many regions of remote and rugged New Guinea, has not had its fauna well surveyed so firm conclusions may be premature.

Fortunately, modern genetic techniques now allow us to examine DNA from historical museum specimens, enabling the exploration of questions associated with species relationships and boundaries, within groups like tree-kangaroos, where only limited numbers of high quality tissue samples are available. Fortunately, the Australian Museum has the most comprehensive collection of tree-kangaroo specimens in the world, including representatives of all currently recognised taxa. Although DNA extracted from museum specimens is often degraded, we were able to successfully obtain DNA sequences of three mitochondrial DNA genes from seven museum specimens, including both specimens of D. deltae and the previously unsampled and very rare Dendrolagus goodfellowi goodfellowi (Goodfellow’s Tree-kangaroo).

Combining these new data, with DNA sequences previously derived from high quality tissue samples, we were able for the first time, to resolve relationships within the distinct Goodfellow’s group of tree-kangaroos. Interestingly, the two specimens of D. deltae had identical DNA sequences and were virtually indistinguishable from our three sampled specimens of D. matschiei from the Huon Peninsula (Fig. 2). This provides strong evidence that D. deltae is not a distinct species separate from D. matschiei, and that the Mt Pratt collection locality associated with them is incorrect. The specimens of D. deltae are also very similar morphologically to specimens of D. matschiei (Fig. 3).

However, as one 88 year-old tree-kangaroo puzzle is laid to rest another arises. Our genetic analysis included, for the first time, samples from both subspecies of Goodfellow’s Tree-kangaroo: D. g. goodfellowi from south-eastern PNG, and D. goodfellowi buergersi from the central highlands of PNG. While D. g. goodfellowi is very poorly known, D. g. buergersi is more familiar, being held in zoos around the world. Although these two subspecies are morphologically quite similar, they were found to be genetically more divergent than some other well accepted Dendrolagus species. But since our sample sizes are quite small, a conclusive decision on whether they actually represent different species, will require additional sampling and an examination of further genetic data.

As well as reviewing the taxonomy of the Goodfellow’s group for the first time since 1982, we were able to bring together and summarise new distributional and biological data for all the currently recognised species and subspecies in this elusive group. We hope our findings will further aid the conservation and management of these unique and threatened species and bring greater awareness to the need for ongoing biodiversity research.

Dr Mark Eldridge

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

Dr Sally Potter

Macquarie University and Research Associate, Australian Museum Research Institute

Prof Kris Helgen

Chief Scientist and Director, Australian Museum Research Institute