Armoured Triassic amphibian may be the first known to live fully at sea

Two 237-million-year-old skeletons from Guizhou, China, and bone strontium values that sit apart from the rock around them suggest Gerrothorax maritimus spent its life in seawater, though one land source could mimic part of that chemical signal.

Peer reviewed · Fossil description plus bone chemistry · Two specimens

No amphibian alive today is known to live its whole life in the sea. A team from the Geological Museum of Guizhou, the University of California, Davis, and Peking University now describes one that may have. Gerrothorax maritimus, a flat, armoured animal, is known from two nearly complete skeletons about 237 million years old. Writing in Nature Ecology & Evolution, Lipin Diao, Ryosuke Motani, Da-yong Jiang and colleagues call it the first fully marine stem amphibian. The case rests on where the fossils lie and a strontium test, and a land source cannot be fully ruled out.

Evidence at a glance

Is this a big deal?

For evolutionary biology, yes, with a caveat. Fully marine amphibians were a blank in both the living world and the fossil record. But the claim rests on two animals, and the chemical evidence has a known alternative reading. Call it a strong first case that needs more specimens.

What scientists found

Gerrothorax belongs to the temnospondyls, an extinct branch on the amphibian side of the family tree. “Stem amphibian” means a relative of frogs and salamanders that branched off before their living group arose. Its close relative G. pulcherrimus, known from East Greenland, is thought to have lived in fresh water.

The new species had a flattened, disc-like body almost entirely covered in bony armour. Nature’s news report gives its length as about 74 cm. Motani said in the UC Davis release that it probably weighed about as much as a giant salamander but was far chunkier. Its sturdy throat bones resemble those of suction-feeding sharks, and the team pictures it waiting on the seabed to suck in passing prey, like a wobbegong carpet shark.

How they found it

The two skeletons, GMG 1119 and GMG 35, are held at the Geological Museum of Guizhou. The second came with a problem: its original collectors had inserted unknown bones and sculpted damaged rock, so the authors left those areas out.

The marine case has three parts. The skeletons lie articulated in marine rock far from any land of the time. They keep their wrists and fingers; carcasses that “bloat and float” tend to lose limbs before sinking, so intact hands suggest the animals died near where they lived.

The third part, added during peer review, is strontium. Seawater’s strontium-87 to strontium-86 ratio is nearly uniform across the oceans at any one time, and growing bone records the ratio of the water and food around it.

SciQuest replotted the published raw data. The 12 bone points span 0.70770 to 0.70794; the 16 rock points span 0.70802 to 0.70924. The sets do not overlap. The authors set the bone values against the seawater value of the time, and the isotope referee called the bone-versus-rock difference convincing.

Why it matters to you

This won’t change anything in your daily life. It answers a wonder question: why amphibians, alone among land vertebrates, seemed never to go to sea. Motani pointed to their skin in the UC Davis release: “That skin is permeable, and that allows the animals to use it for multiple purposes.” Amphibians breathe and take in water through it, which is why frog skin does so much work, and why salt water is a problem.

Think of the armour as a raincoat over that skin. With water, gas and salt blocked, the animal had to use its gills and gut, which can become more salt-tolerant. The authors argue this prepared the lineage for the sea. The analogy’s limit: no skin or gills survive, so how well the armour sealed the body is inferred, not measured.

What the evidence does not show

The isotope referee pointed out that one component of the nearby land would also give values of about 0.707, matching a marine signal. The paper now says so and argues the bone-rock difference makes the sea more likely. The referee also noted the bones are likely influenced by diagenesis, chemical change during burial.

Another reviewer questioned the armour argument, because the freshwater G. pulcherrimus was armoured too. The authors reply that armour was an exaptation: a feature that evolved for other reasons and later proved useful at sea.

The fossils say nothing direct about reproduction, and cannot date the move to sea. After review, the authors changed “in the Late Triassic” to “in the Triassic”, since this may be the first known occurrence, not the first marine generation.

How this fits what we already knew

Of more than 8,000 living amphibian species, fewer than 2% show some salt tolerance, according to the UC Davis release. Even the crab-eating frog, one of the most tolerant, loses weight in brackish water and recovers in fresh. In 2001, Michel Laurin and Rodrigo Soler-Gijón described a 290-million-year-old temnospondyl from Spain that appears to have lived in coastal salt water. G. maritimus goes further: open sea, with no return to fresh water.

Outside views

SciQuest has not interviewed an outside palaeontologist. Steve Brusatte of the University of Edinburgh, who was not involved, told Nature’s news team the find could be a major one if the animal truly was fully marine. The journal’s open peer review file adds views from two anonymous referees with expertise in early amphibians. Reviewer 1 judged the articulated fossils, marine rock, armour and wrist preservation “enough to make a strong case for this animal being marine.” Reviewer 2, writing before the strontium work was added, warned that “a marine depositional environment does not conclusively demonstrate that the taxa would have been marine-adapted.”

How other headlines framed it

• Nature’s news team led with evolutionary significance and Brusatte’s conditional praise, and stressed the strontium match to seawater.
• Gizmodo went for a predator-on-the-seafloor angle, quoted Motani directly, and noted that no soft tissue survives, so digestion and reproduction remain unknown.
• Sci.News gave a plain “first known” summary built on the abstract and release.

Reported lengths differ: 73.7 to 75 cm in the three outlets, but “almost three feet” (about 91 cm) in the UC Davis release, which overstates it. SciQuest could not check the figure in the paywalled paper.

What happens next

Motani told Gizmodo the team wants more specimens; both skeletons lie belly-up, so the back is unseen. The isotope referee suggested finer chemical mapping of bone and rock to test burial alteration, which the authors called beyond this paper’s scope. For a sense of the seafloor these animals lived on, see how deep the ocean is.

In one sentence

An armoured, flat-bodied relative of today’s amphibians may have lived its whole life in a Triassic sea, a first for the amphibian line, though the case rests on two skeletons and one chemical test with a land-based alternative.

Read next

• A Jurassic swimming mammal relative: another animal moving into water.
• Poison dart frog adaptations: what living amphibian skin does.
• Fossil pollen and ancient warmth: reading deep time from chemistry and fossils.

Sources and disclosure

1. Diao L, Motani R, Jiang D-Y, Hao J, Wang S. “A fully marine stem amphibian from the Late Triassic of China.” Nature Ecology & Evolution, published online 26 October 2026. doi:10.1038/s41559-026-03209-9 (abstract, figure captions and references read; full text paywalled).
2. Supplementary Information to the paper, including Supplementary Table 1 (strontium data) and analysis code. PDF.
3. Peer Review File for the paper (CC BY 4.0; reviewers credited as anonymous referees). PDF.
4. University of California, Davis. “Researchers unearth fossil evidence of first fully marine amphibian.” News release, 26 October 2026, via EurekAlert.
5. Laurin M, Soler-Gijón R. “The oldest stegocephalian from the Iberian Peninsula: evidence that temnospondyls were euryhaline.” Comptes Rendus de l’Académie des Sciences III 324: 495–501 (2001). doi:10.1016/S0764-4469(01)01318-X.
6. Jenkins FA, Shubin NH, Gatesy SM, Warren A. “Gerrothorax pulcherrimus from the Upper Triassic Fleming Fjord Formation of East Greenland.” Journal of Vertebrate Paleontology 28: 935–950 (2008). doi:10.1671/0272-4634-28.4.935.
7. Coverage read for framing: Nature news (26 October 2026), Gizmodo, Sci.News.

Disclosure: This article was drafted with AI assistance and checked against the sources above. The full paper is paywalled; SciQuest read its abstract, figure captions, reference list, open Supplementary Information and open peer review file. The graphic replots all 28 published strontium values; ranges and the no-overlap result were computed by SciQuest from Supplementary Table 1. The study was funded by the National Natural Science Foundation of China, according to the UC Davis release. Spotted an error? See our corrections policy.