Environment Life

254-Million-Year-Old fish, nearly mistaken for a frog, uncovers more about ancient Europe

Fahad Redha

September 3, 2026
6 min read

254-Million-Year-Old fish, nearly mistaken for a frog, uncovers more about ancient Europe

When a geologist named Léo Adriano de Oliveira found a flattened shape pressed into reddish siltstone near the town of Alto Garças in 2005, the people living nearby had already decided what it was. A frog, they said, its limbs splayed out where they had been squashed flat by millions of years of rock. They were off by about 254 million years and an entire branch of the animal kingdom. The fossil Oliveira collected has now been formally named Leolepis matogrossensis, after him, in a study led by Valéria Gallo of Rio de Janeiro State University with the Natural History Museum’s Martha Richter as a co-author.

It is not a frog. It is a 15-centimetre ray-finned fish, a member of the group that today includes 96% of all bony fish alive, from tuna to salmon to cod. And it comes from one of the most dangerous places on Earth to go looking for fossils: the ground around Brazil’s Araguainha crater, the largest confirmed meteorite impact site in South America, where the shock of a three-kilometre-wide asteroid folded and warped the surrounding rock layers around 252 million years ago.

The mistake made by the fossil’s original finders is an understandable one. Ray-finned fish from this period are usually recovered in Brazil as scattered teeth, isolated scales or broken fragments, not as a body with fins and armour still discernible in outline. What Oliveira actually found was a fish that, unlike almost all its living relatives, wore its skeleton on the outside as much as within it: scales built from dentine and enamel, the same materials that form human teeth, arranged as tough armour plating rather than the thin, flexible scales carried by most modern bony fish.

Working out what Leolepis actually was required more than a glance at a slab of rock. Gallo’s team compared the fossil’s skull anatomy against 37 other species across 92 separate anatomical features, feeding the results into software built to reconstruct evolutionary family trees. The answer that came back was not the one a Brazilian fossil might be expected to produce. Leolepis matched most closely not to any of its South American contemporaries, but to Paramblypterus, a genus previously known only from lake beds in Europe, thousands of kilometres away and on what was then a different part of the ancient supercontinent Gondwana.

That connection matters because of where and when Leolepis was living. Its home was a “megalake,” the drying remnant of a vast inland sea that had once covered much of what is now the join between South America and Africa, before Gondwana began to break apart. Based on its small teeth, the fish is thought to have fed on invertebrates and plant matter in shallowing, increasingly arid waters. Finding a close relative of a European lake fish in a Brazilian megalake pushes the known range of that fish family across an ocean’s width earlier than palaeontologists had reason to expect.

The timing is what turns this from a naming exercise into something closer to a snapshot of the world moments before catastrophe. Leolepis was alive shortly before the Araguainha impact, and the impact itself struck close enough to the beginning of the End-Permian mass extinction, the deadliest die-off in the planet’s history, that the two events sit almost on top of each other in the geological record. As many as 95% of marine species and 78% of life on land disappeared in that extinction. The prevailing explanation remains a much larger culprit: enormous volcanic eruptions in what are now Siberia and China, which reshaped the climate on a scale a single impact crater could not match on its own. But the Araguainha strike may still have added its own damage to ecosystems in the southern hemisphere at almost exactly the wrong moment.

Dr Martha Richter, the study’s corresponding author and a scientific associate at the Natural History Museum, said the fossil’s survival is itself a small piece of luck. “Most fishes of this age in Brazil are generally known from scattered teeth, scales, and isolated bone fragments,” she said. “Leolepis, on the other hand, is remarkably complete, especially considering a major meteorite strike warped the nearby rocks!” Richter and her co-authors believe the fossil-bearing layer was already buried deep enough underground at the time of the impact to have escaped the worst of the shockwave that folded rock as far away as 1,000 kilometres.

Richter also framed the find as evidence of how much of this record remains unread. “Its completeness and characteristics offer a window into the evolution of ray-finned fish,” she said. “The discovery of Leolepis also shows that there could be many incredible new fossils still to come not just from Brazil, but South America as a whole.”

Fewer than twenty basal ray-finned fish species have ever been formally described from South American Paleozoic rocks, a thin record for a period that, elsewhere in the world, produced far more.

The study, “Leolepis matogrossensis gen. et sp. nov., an Upper Permian actinopterygian fish from the Corumbataí Formation, in the vicinity of the Araguainha impact crater, State of Mato Grosso, mid-western Brazil,” is published in the Journal of South American Earth Sciences.

The next fossil to reshape a family tree this way might already be sitting in a Brazilian drawer. EyeOnLondon will be watching what the Natural History Museum’s South American partnerships turn up next.

For more, visit EyeOnLondon at eye-on-london.com