Synchrotron scans find gills in Westlothiana, the fossil long cited as the earliest land-adapted amniote

This whole result rests on one animal. Westlothiana survives as two crushed slabs, deformed and hard to interpret ^[15], and the 1990 description worked mostly from what showed on their surfaces ^[2]. Seen from the surface it looked like a land animal: a nearly complete skeleton, four limbs, and fingers and toes with claw-like tips, at a time when many of its contemporaries had more fin-like limbs ^[3].

The new work did not need a new specimen. The team took the slabs to the European Synchrotron Radiation Facility in Grenoble, where X-rays pass through the rock and let researchers reconstruct structures hidden inside it ^[5]. "But when we were finally able to see inside the fossil, we found an animal that looked very different from what we expected," said Xavier Jenkins, a postdoctoral fellow at the American Museum of Natural History and co-lead author ^[6]. Besides the gills and the primitive skull, the scans showed a fish-like mouth lined with thousands of tiny teeth ^[7].

I'd weigh the anatomy as the firmest part of the paper. Ossified gills and a fish-like mouth are hard to square with a land animal, and the authors conclude it was aquatic or amphibious ^[7][8]. "The surprisingly primitive features of the skull show us that Lizzie was not only not a reptile but belonged to a much more ancient lineage," said Ben Igielman, the other co-lead author, who began the work as a graduate student at Oxford ^[9].

The claws matter more than the reclassification itself. If Westlothiana sits on the tetrapod stem, then claws and five digits on the hands and feet had already evolved in earlier, more aquatic members of that lineage ^[10]. "It shows that, although we think of amphibians as being more 'primitive' than reptiles today, some of the traits of reptiles were already there in the ancestor of both amphibians and reptiles," said co-author Roger Benson, the museum's Macaulay curator of paleontology ^[11].

That changes how footprints can be read. Trackways showing a four-limbed walker with claw-like digits have recently been used to argue that amniotes arose in the Devonian, more than 359 million years ago ^[12]. On that reading, amniotes would have been around at least 14 million years before Westlothiana ^[16]. If a gilled stem animal also had claws, a clawed print alone cannot identify an amniote, and the authors say some tracks credited to early amniotes may have been made by amphibious tetrapod relatives ^[13]. The scans show that a claw is not enough to identify an amniote. Assigning any particular Devonian trackway to a different maker would take separate work on those tracks.

Since 1990, researchers have often used Westlothiana as a landmark in arguments about when amniotes appeared ^[4]. The phys.org account of the Nature paper does not name a fossil to replace it or give a revised origin date ^[5]. "The broader lesson is that terrestrial-looking traits, such as a reptile-like foot, a weight-bearing forelimb and claw-like phalanges, were not unique features of tetrapods, but rather accumulated piecemeal in their close relatives, in animals that were still living in and out of the water," Jenkins said ^[14].