It is difficult to make sense of a river system that has existed for hundreds of millions of years. But that is the extraordinary claim surrounding one Australian river, whose origins may stretch back between 300 and 400 million years.
The river is the Finke in Central Australia, associated with the Arrernte name Lhere Pirnte and commonly rendered in English as Larapinta. The neat version of the story calls it the oldest river on Earth, older than the dinosaurs and already following its course before the first reptile walked on land.
I went looking for the geology behind that sentence. The river’s history is genuinely interesting, but the confident record claim runs ahead of the evidence. The Finke has not been directly dated to 300 or 400 million years old. What may be that ancient is part of its drainage pattern, and researchers who study it describe the extreme age as speculative.
An old river is not one permanent channel
The first thing I had to get straight was what it means for a river to be hundreds of millions of years old. It cannot mean the same water has followed the same banks without interruption since the Paleozoic Era.
Rivers are restless. Bends migrate, floods open shortcuts, sediment blocks channels and tectonic movement tilts the land. Tributaries can be captured by neighboring systems. A river can even change which basin receives its water.
Individual grains of sand and gravel do not stay put either. They move, are buried, return to the surface and move again.
When geologists discuss the Finke’s age, they are talking about inherited drainage. The upper river may still cross the Central Ranges along part of a route established in the Paleozoic, even though the channels, gorges and sediments along that route have been reworked many times.
That distinction also tells us what evidence to look for. No one needs to find a 350-million-year-old riverbank where today’s bank sits. The case depends on the river’s relationship to the landscape.
The river cuts across ridges it might be expected to avoid
Rivers generally find easier routes downhill. They follow valleys, exploit softer rock and run around resistant high ground where possible. The upper Finke does something that needs explaining: it cuts across a series of folded east-west ridges in the MacDonnell, Krichauff and James ranges.
The river forms deeply incised bends through hard rock. From above, some look like ordinary meanders from a flat plain lowered into a mountain range.
One explanation is a process geologists call antecedence. A river establishes its course across relatively low terrain, then keeps cutting down as the land rises or folds beneath it. If erosion keeps pace with uplift, the channel can preserve a route that looks almost perverse in the modern topography.
The Central Ranges were shaped during the long Alice Springs Orogeny, a mountain-building episode that deformed Central Australia through the Paleozoic Era. The argument for an ancient Finke is that its cross-range course predates at least some of those structures, including deformation dating to roughly 300 million years ago.
This is the foundation of the 300 to 400 million-year estimate. It is an inference from drainage geometry and tectonic history, not the result of putting a piece of today’s riverbed through a dating machine.
The 2022 study did not date the Finke’s birth
A 2022 paper in Earth and Planetary Science Letters, led by geomorphologist John Jansen, begins by saying the Finke is counted among the world’s oldest drainage systems. In the paper itself, the authors also call such antiquity speculative.
I appreciated that sentence because it shows the difference between a plausible geological history and a measured birthday.
Jansen and his colleagues were trying to explain a strange set of intertwined active and abandoned gorges in the upper Finke. Over a roughly 30-kilometer reach through the Krichauff and James ranges, the present gorge crosses and overlaps an older channel that now sits above it in places.
The team analyzed the shape of the gorges and measured the cosmogenic isotopes beryllium-10 and aluminum-26 in old river gravels. These isotopes are produced while minerals are exposed near Earth’s surface. Their relative amounts change during deep burial, so they can reveal aspects of a sediment’s exposure and storage history.
The isotope measurements pointed to gravel being stored for several million years, long enough to erase much of its earlier exposure signal. Combined with gravity data and computer modeling, the results supported a history of repeated, gentle tilting. The river incised a gorge, the valley filled with sediment, the channel shifted and erosion began again.
This is one paper, not settled consensus about when the Finke began. More importantly, it did not produce a 300-million-year date for the river. It investigated how an old drainage system responded to landscape movement over a much more recent, multi-million-year interval.
The modern Finke spends much of its time dry
The Finke begins in the MacDonnell Ranges and runs southeast across the dry interior toward the Lake Eyre Basin. The broader course is about 800 kilometers in the 2022 paper, though quoted lengths vary depending on which connected channels and endpoint are counted.
It is an intermittent river. Long reaches may be dry for much of the year, while permanent and semi-permanent waterholes persist in parts of the system. After heavy rain, separated channels can reconnect and carry major flows through country that recently looked waterless.
That behavior is another reason to be careful with the mental picture created by the word “river.” The Finke is a drainage system that switches between dry sand, isolated water and flood. Continuity belongs to the landscape pattern, not to an unbroken ribbon of flowing water.
The Finke Gorge National Park joint management plan uses the Arrernte name Lhere Pirnte, pronounced roughly “lara pinta,” and translates it as “salty river.” Larapinta is the familiar English rendering.
The plan describes the river and its waterholes as part of a living cultural landscape for Arrernte Traditional Owners. I think that matters in an article drawn toward a geological superlative. “Oldest river” is an outside ranking. It does not contain what the place means to the people whose country it crosses.
Older than dinosaurs is clear, but the reptile claim is not
Dinosaurs appear in the fossil record roughly 230 million years ago. Even the low end of the Finke estimate, 300 million years, would put the inherited drainage about 70 million years earlier.
The claim about reptiles is now much harder to sustain.
In May 2025, John Long and colleagues published a study in Nature describing probable amniote footprints from the Snowy Plains Formation in Victoria. Amniotes are the lineage that includes reptiles, birds and mammals. The prints have long toes and claw marks, and the authors argue that a primitive reptile-line animal was the most likely track-maker.
The rocks are from the earliest Carboniferous, about 356 million years ago. Before this report, the oldest widely accepted reptile body fossils and trackways were tens of millions of years younger. The Victorian tracks pushed the likely origin of the amniote group much further back.
The identification is based on footprints rather than a skeleton, so it remains an interpretation to be tested against future finds. Still, it changes the comparison with the Finke. A drainage history close to 400 million years would predate the tracks. One close to 300 million years would be younger.
The full 300 to 400 million-year range therefore cannot support the categorical line that the Finke had carved its course before the first reptile walked on land. The lower part of the range misses that mark by more than 50 million years.
“One of the oldest” is the more defensible description
Geoscience Australia calls the Finke “one of the oldest rivers in the world.” After reading the sources, that is the description I would keep.
There is no global register capable of certifying the oldest surviving river. Ages are reconstructed from incomplete landforms, sediment, erosion rates and tectonic histories. Definitions differ too. One estimate may refer to a modern valley, another to the direction of an older drainage network and another to the age of uplift that a river appears to predate.
Other ancient landscapes may have lost equally old evidence through erosion, burial, glaciation or later tectonic change. Australia’s slow erosion and long geological stability make traces easier to preserve, but preservation is not the same thing as exclusive possession of the oldest river.
I began with a number so large that it barely registered as time. Having a young child does that to me occasionally. A year can feel enormous at home, then a landscape makes a million years look brief.
The Finke does not need a certified world record to be interesting. Its upper drainage appears to preserve a Paleozoic inheritance while the river itself has shifted, filled, dried and cut new gorges. The ancient part is the persistence of a pattern, not an untouched channel.
