New research on fossil trackways from Australia suggest an earlier divergence of land-living tetrapods from amphibians.

Andrew Smart, Head of Science and Research

We spend all our efforts working to protect and provide new habitats and environments for our living reptiles and amphibians, but I thought it would be interesting to take a moment to consider a new discovery from Australia and its implications.

Two evolutionary moments of great significance, the separation of lungfish and ancestral tetrapods, and the separation of amphibious tetrapods and land-living (and reproducing) tetrapods are the focus of much study, particularly with novel methods used to assess the footprints and trackways left in the fossil record (paleoichnology).  

A recent paper in the journal Nature analyses a slab of rock from the Snowy Plains Formation of Victoria, Taungurung Country (358 to 354 million years old).  This slab of rock (picture below from Fig. 2 of Long et al (2025)) shows three lines of footprints made by an amniote tetrapod (the tracks show the impressions of claws – a characteristic of amniotes not shared in amphibians in the fossil record of that period). The animal left no body or tail drag marks and based on the dimensions of the modern water monitor, Varanus salvator, the authors estimate a lizard-like animal of some 80 cm in length.

Amniotes (so called because of the evolution of the amniotic membrane in the egg) were the first vertebrate group that could live on land and lay shelled eggs. Shelled eggs removed the need to return to water bodies to reproduce or to lay their eggs in humid conditions and so made more terrestrial habitat available for colonisation.

Prior to this recent find, reptilian (amniote tetrapod) divergence from amphibious tetrapods was estimated, from molecular studies (DNA analysis), at some 352 million years ago and the first likely reptile was Casineria (340 million years ago) found near Edinburgh in 1999.  (Casineria is the subject of some debate because the fossil has no skull).  In 1989 another early amniote skeleton was found from West Lothian carboniferous rocks and identified as 338 million years old (Smithson 1989).

Before this find, the first ‘reptile’ trackmaker was Notalacerta and the first clear reptile body fossil was Hylonomus (c. 310 million years ago). Hylonomus was a lizard-like species, 20 cm in length, that is believed to have used club moss stumps as shelter or nests sites. (Marchetti et al 2021).

The new trackway from Australia moves the divergence of amphibians and amniotes back by between 35 and 40 million years, suggesting tetrapod and amniote lineages must have their origins 380 million years ago.

The separation of tetrapod ancestral ‘fish’ from lungfish is supported by discovery in 2012 of Tungsenia, a Lower Devonian ‘ancestral tetrapod fish’ from China (Lu et al , 2012), suggesting that the earliest ancestor of terrestrial tetrapods diverged from lungfish around 409 million years ago. 

The earliest amphibious tetrapod records are from tracks made in Poland (Niedźwiedzki et al, 2020) that indicate tetrapods of 2m or more lived in intertidal / lagoon environments some 398 million years ago (eighteen million years before the earliest tetrapod fossils were deposited). The authors suggested that the evolution of amphibious tetrapods occurred in intertidal zones rather than in freshwater, with tides providing an available food resource of stranded animals twice a day.

Sumida (2025) discusses the implications of Long et al’s new discovery and the evidence suggests that if:

• ancestral tetrapods diverged from lungfish c. 410 million years ago.
• amphibious tetrapod trackways demonstrate evidence from 390 million years ago
• fossil reptiles were present 318 million years ago (Hylonomus 318 million years ago and tracks of Notalacerta 320 to 330 million years ago)

then the new trackway evidence of reptiles from around 355 million years ago indicates that the evolutionary gap during which amniotic tetrapods separated from amphibious tetrapods is much shorter in time than previously suggested. Previously estimated as 85 to 90 million years, this has now become a gap of ‘only’ fifty million years, suggesting that tetrapod evolution proceeded more quickly, and the Devonian tetrapod record is much less complete, than previously believed.

References