Somewhere in the dense, wind-battered scrub of Western Australia’s southern coast, a small marsupial the size of a young rabbit is quietly running out of room. Gilbert’s potoroo, considered the world’s most endangered marsupial, survives in roughly 150 individuals scattered across just three sites. Now, researchers at Edith Cowan University have developed a way to scout new habitat for the species without ever needing to spot one. Their tool: the droppings of other animals.
In a study published in May 2026 in Biodiversity and Conservation, the team used a technique called environmental DNA (eDNA) metabarcoding to analyze scat from quokkas, quendas, and bush rats. These more common mammals share the potoroo’s taste for underground fungi, known as truffles. By cataloging fungal DNA signatures in their droppings, the researchers could determine whether a patch of bush contained enough of the right food to sustain potoroos, all without trapping or even glimpsing one of the notoriously elusive animals.
A marsupial that eats almost nothing but truffles
Gilbert’s potoroo (Potorous gilbertii) is a small, nocturnal marsupial with dark fur, large hind feet, and a pointed snout it uses to dig for hypogeous fungi, the fruiting bodies of fungi that grow entirely underground. Unlike most small mammals, which supplement fungi with insects, seeds, or plant material, the potoroo’s diet is almost exclusively fungal. That extreme specialization makes it acutely vulnerable: if the right fungi disappear from a site, the potoroo cannot simply switch to another food source.
The species was presumed extinct for over a century before a small colony was rediscovered in 1994 at Two Peoples Bay Nature Reserve, near Albany. That find stunned biologists and triggered an intensive recovery effort that continues today.
Australia’s federal government lists Gilbert’s potoroo as Critically Endangered under the Environment Protection and Biodiversity Conservation (EPBC) Act. Genetic studies have confirmed a severe bottleneck in the surviving population, meaning the gene pool is so narrow that the animals have limited capacity to adapt to new diseases, shifting climate conditions, or habitat disturbance.
Three sites, no margin for error
Since the 1994 rediscovery, conservation managers have tried to spread the risk by establishing insurance populations away from Two Peoples Bay. A group of potoroos was translocated to Bald Island, a predator-free island off the southern coast, where the first pouch young bred in that new setting was confirmed in 2006. A second insurance colony was placed inside a fenced enclosure at Waychinicup National Park.
These three locations represent the species’ entire known range. That concentration is the core problem. A single wildfire tore through potoroo habitat in 2015, damaging populations and reinforcing how quickly one event could undo decades of recovery work. In dense coastal heath where fire is a natural force, having every surviving animal within a few dozen kilometers of one another is a gamble conservationists can no longer afford.
Turning other animals’ droppings into a habitat map
This is where the Edith Cowan University study offers a practical breakthrough. Traditional surveys for Gilbert’s potoroo are slow and expensive. The animals are nocturnal, trap-shy, and live in thick vegetation that makes camera trapping unreliable. Searching for the fungi they eat is equally difficult because the target species fruit underground and leave no visible trace on the surface.
The researchers sidestepped both problems by focusing on proxy species. Quokkas, quendas (southern brown bandicoots), and bush rats are far more abundant and easier to sample. They also dig up and eat many of the same truffle species the potoroo depends on. When these animals defecate, fragments of fungal DNA pass through intact.
By collecting scat from these common mammals across multiple sites and running eDNA metabarcoding, a process that amplifies and sequences short DNA fragments to identify which organisms are present, the team built a picture of fungal diversity across the landscape. Sites where scat samples contained a rich variety of the fungi potoroos are known to eat were flagged as strong candidates for future translocation.
The method does not detect potoroos directly. It detects their pantry. But for a species whose survival hinges on a single food group, knowing where that food exists in sufficient diversity may be just as valuable as knowing where the animals themselves are hiding.
What the study does not settle
The research opens a promising new avenue, but several questions remain. The published paper focuses on fungal diet signals as habitat indicators. Whether any scat sample in the study also yielded potoroo DNA directly, or whether the method was validated against sites where potoroos are already known to live, is not fully detailed in available summaries of the work.
The frequently cited figure of approximately 150 surviving individuals appears in conservation materials but lacks a published, methodologically transparent census behind it. Federal recovery documents describe the population as extremely small and geographically restricted, and the genetic bottleneck evidence supports that characterization. The number should be read as an order-of-magnitude estimate rather than a precise headcount.
Raw eDNA sequence data from the study have not yet been released publicly, which is standard for a newly published paper but means independent verification of the specific fungal detection claims will take time. None of this undermines the study’s core contribution. It does mean that the technique’s reliability as a habitat-selection tool will need to be tested further before managers commit scarce potoroo individuals to new sites based on its results alone.
Why faster habitat screening matters now
Climate projections for southwestern Australia point toward hotter, drier conditions and more frequent high-intensity fires over the coming decades. For a species locked into a handful of coastal sites, each of those trends compounds the risk. Translocation to new, well-chosen habitat is widely regarded as the most viable path to long-term survival, but choosing the wrong site wastes animals the species cannot afford to lose.
If eDNA mapping can reliably flag patches of bush rich in the underground fungi potoroos need, it gives managers a screening tool that is faster, cheaper, and less invasive than traditional survey methods. Instead of spending months trapping and soil-coring at a candidate site, a team could collect scat from resident quokkas and quendas, run the lab analysis, and have a preliminary answer within weeks.
Gilbert’s potoroo is not out of danger. No single study can resolve the combined pressures of fire, genetic fragility, habitat loss, and climate change bearing down on a population this small. But the Edith Cowan University research demonstrates something conservation managers rarely get: a practical shortcut in a process where time is the scarcest resource of all. For one of Australia’s most precarious mammals, that shortcut could determine whether the next translocation succeeds or whether the species edges closer to a second disappearance.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
