The jaw nobody believed
Nine fossil lower jawbones, pulled from a dry riverbed in a forest near Brazil’s Amazon rainforest, looked at first like nothing more than breakage, exactly what pressure does to bone after 275 million years. Martha Richter, a paleontologist at the Natural History Museum in London, first suspected the pieces might not even be from a tetrapod (a four-limbed vertebrate) at all, but from a fish. Only once the fossils were cleaned and prepared did their real shape come into focus.
Then came eight more jaws, all showing the identical twist, well-preserved ones included. Lead author Jason Pardo has said the team spent years wondering if they were looking at deformation, until nine jaws in a row showed the same bend. Damage doesn’t repeat with that precision, a pattern does, and that’s the myth this find kills: the twist isn’t crush damage, it’s anatomy. Pardo, Richter, and paleontologist Juan Carlos Cisneros of the Federal University of Piauí concluded the nine jawbones, each about 5.9 inches (15 centimeters) long, belonged to a single new species. They named it Tanyka amnicola, “jaw living next to the river,” after the riverbed at Pastos Bons where the first specimen turned up, in a study published in Proceedings of the Royal Society B in March 2026.
How a jaw learns to point sideways
Tanyka amnicola was a stem tetrapod, a branch of early four-limbed vertebrates that split off before amphibians, reptiles, birds, and mammals diverged. Its lower jaw twists along its own length, back to front, so the front teeth end up angled outward and sideways instead of pointing straight up toward an opposing row, as in a normal bite. The describing paper calls this torsion of the mandibular ramus, the jaw’s horizontal branch, paired with an arched coronoid, a raised ridge along the jaw’s inner edge, carrying an unusually dense set of denticles: small, blunt, tooth-like grinding points, nothing like the blades a predator needs to shear meat.
That’s two odd features on one bone: a structural twist no other known tetrapod carries, fossil or living, and a texture swap, from cutting edge to grinding surface, on the same jaw doing the twisting.
What the twist was actually for
Pardo and Richter’s reading: those inward-facing denticles likely ground against a matching row on the upper jaw, which has never been found, turning the mouth into a mill instead of a pair of scissors. Pardo describes the resulting motion as teeth rasping against each other, calling it a genuinely unusual way to feed among tetrapods.
Cisneros puts a name on what the mill was probably grinding. Based on the teeth, he says, the team thinks Tanyka amnicola ate plants at least part of the time. A grinding surface like this would also work against hard invertebrate shells, so the team isn’t ruling that out either. What stands out either way is the company Tanyka amnicola keeps: most of its relatives on the stem tetrapod branch were built to bite and swallow whole, not chew. If the interpretation holds, this is one of the earliest known specialized chewers on that branch of the family tree, a different diet hiding inside a jaw shape unlike anything else in the record.
The lineage that wasn’t supposed to be here
The standard timeline had stem tetrapods dying out worldwide well before the Permian period, undone when the Carboniferous rainforest collapse dried out the wetlands that lineage needed, across the northern half of the supercontinent Pangaea. Tanyka amnicola breaks that timeline. It’s dated to roughly 275 million years ago, well into the Early Permian, and comes from Gondwana, the southern half of Pangaea, long after the group was supposed to be gone everywhere.
The paper calls Tanyka amnicola the second stem tetrapod group known to have hung on in Gondwana that late, after already dying out to the north in Laurussia, and argues that current ideas about that Carboniferous turnover are too simple. Pardo compares its position on the family tree to a platypus: an older lineage still going about its business long after newer, more modern tetrapods had taken over elsewhere. The region Tanyka amnicola lived in was hot and seasonally dry, closer to the modern southwestern United States than to a swamp, and the team’s hypothesis is that this climate gap let the southern population outlast its northern relatives. The extinction, in other words, wasn’t global, it was regional, and nobody had fossils from the south to notice.
The half of the animal that’s missing
The upper jaw of Tanyka amnicola has never turned up, so the matching denticle row it supposedly ground against is an inference, not a fossil anyone has held. The entire grinding-mill idea currently rests on one half of a mouth. Other fossils found near the same riverbed might belong to this same animal, or might not; the team hasn’t confirmed the association either way.
That leaves the rest of the body an open question. Researchers have sketched Tanyka amnicola as something like a salamander with a much longer snout, and one press estimate puts its full length at around 3 feet, but that number doesn’t come from the paper itself, and there’s no skeleton to check it against. Only more fossils from Gondwana, whether from Brazil, Africa, Australia, or Antarctica, will say whether the climate hypothesis holds, and what the rest of Tanyka amnicola actually looked like.