A newly identified order of Australian marsupials survived in Queensland rainforests for more than 35 million years, and it vanished so completely that no descendants of it are alive today

An entire order of marsupials lived in Queensland’s rainforests for more than 35 million years, then disappeared so completely that nothing alive today is descended from it. That is the claim in a paper published in the Journal of Paleontology on 14 June 2026, built from fossil teeth and jaw fragments recovered from the Riversleigh World Heritage Area in north west Queensland. The newly identified order has been named Keeunamorphia.

The lead author is Dr Tim Churchill, a palaeontologist at the University of New South Wales, whose findings are detailed in a paper published in the Journal of Paleontology. The paper describes three new species, including two placed in a newly named genus, Phantasmodon, alongside three previously known species — Keeuna woodburnei and Ankotarinja tirarensis (Archer, M. 1976. “Miocene marsupicarnivores (Marsupialia) from central South Australia, Ankotarinja tirarensis gen. et sp. nov., Keeuna woodburnei gen. et sp. nov., and their significance in terms of early marsupial radiations.” Transactions of the Royal Society of South Australia 100: 53–73), and Djarthia murgonensis — that the authors argue belong to the same order.

All of the animals were small, ranging between 25 and 200 grams, roughly the size of a shrew to a small mouse, and appear to have been insect eating rainforest dwellers.

What the fossils actually establish

The case for a new order rests on teeth and jaw fragments rather than complete skeletons, which is the ordinary state of affairs for Riversleigh’s Miocene deposits, a site better known for producing isolated dental remains than articulated specimens. Dental structure carries real taxonomic weight in marsupial classification, since tooth shape and arrangement track evolutionary relationships closely enough that palaeontologists routinely build classifications from teeth alone. Even so, a classification built from teeth and jaw fragments is a narrower evidentiary base than a fuller skeleton would provide, and the order’s exact position within the marsupial family tree could shift as more complete material is found. The order’s full timespan also hinges on where its single oldest fossil, Djarthia murgonensis, is placed: the authors’ parsimony analysis puts it inside Keeunamorphia, but their own Bayesian analysis does not, placing it instead as a separate, more basal lineage outside the order. The “more than 35 million years” figure assumes the first reading.

The fossils placing the newest species date to around 18 million years ago, in the Early Miocene. The oldest fossil the authors link to the same lineage is closer to 55 million years old, and the paper’s account of the order’s total span, life-of-order estimated at more than 35 million years before it died out roughly 15 million years ago, rests on connecting fossils separated by tens of millions of years and, in some cases, by gaps in the record between them. That is a real limitation. A lineage inferred from fossils at both ends of a long span, with comparatively little in between, is a plausible reconstruction rather than a continuous documented record.

Why it is being called an order rather than a smaller grouping

Order is a fairly high level in the standard taxonomic hierarchy, above family and genus, and marsupial orders are not casually proposed. Churchill’s framing of the find leaned into that scale: “Not only is it a new order, it could also be the most ancient lineage of all Australian marsupials,” he said, adding that Keeunamorphia “may be the early ancestor of all our marsupial carnivores.” Both lines were reported consistently across independent coverage of the paper’s release, including the University of New South Wales’s own account and a separate summary published by Sci.News.

The claim to being marsupial carnivores’ early ancestor is the more speculative half of that framing. Keeunamorphia’s known members were small insect eaters. The paper’s suggestion is about ancestry — a possible position on the family tree — rather than behavioural resemblance to Australia’s modern carnivorous marsupials, such as quolls or the extinct Tasmanian tiger.
Whether Keeunamorphia sits on the branch that eventually produced those animals is a hypothesis the fossil record can test further, not a settled point established by this paper alone.

Where this fits against other marsupial family tree revisions

This is not the first time Riversleigh fossils have complicated the standard account of how Australia’s marsupials are related to one another. The site has produced a steady stream of unusual, hard to classify species over recent years, several of which have prompted researchers to revise assumptions about how many distinct lineages existed early in the group’s history. Today’s marsupial orders increasingly look less like the product of one tidy branching sequence.

Keeunamorphia’s authors point to a possible Gondwanan origin for the lineage, raising the further possibility of a connection to marsupial relatives in South America, a link that would require fossil evidence from outside Australia to test properly and that the current paper does not attempt to establish on its own.

Churchill’s own framing of the broader point was more modest than the order level classification might suggest on its own: evolutionary history, in his account, is “a lot more complex than just one group leading to all of Australia’s marsupials after being left behind when the continent broke off from Antarctica.”

That summary fits a pattern a growing number of Riversleigh finds have been suggesting for some time; this paper adds to it rather than originating it.

What extinction without descendants actually means here

No living marsupial has been identified as a direct descendant of Keeunamorphia. That is different from saying the lineage left no trace at all. If Churchill’s ancestry hypothesis holds up under further testing, some of Keeunamorphia’s genetic and evolutionary legacy could persist indirectly, filtered through whatever lineage did eventually produce Australia’s modern carnivorous marsupials. What is being described as extinct is the order itself as a distinct branch, not necessarily every trace of its influence on what came after it.

The order’s disappearance around 15 million years ago lines up with a broader period of environmental change across northern Australia, as the extensive rainforest that once covered much of Queensland began to contract and fragment. The paper does not present detailed evidence tying Keeunamorphia’s extinction specifically to that environmental shift; the timing lines up, but that alone falls short of a demonstrated cause.

An order known so far only by its teeth

Everything currently known about Keeunamorphia comes from teeth and jaw fragments, which is, in its own way, fitting. An order that persisted for 35 million years and then disappeared without leaving a single living relative was never going to hand over a complete picture easily. Riversleigh remains an active dig site, and whatever comes out of it next, a fuller skeleton or another isolated tooth, is what will actually decide whether Keeunamorphia holds up as its own order or gets folded back into something else.