Bizarre tail weaponry in a transitional ankylosaur from subantarctic Chile

In February 2018, a small team of paleontologists in a very isolated region of Chilean Patagonia faced a unique challenge: with freezing weather and only 5 days of fieldwork left, they needed to extract a block from a steep hill with surfacing bones of a 74 million year-old dinosaur.

Like Comment
Read the paper

(Poster Image: Luis Pérez López)

A series of discoveries

In February 2018, a small team of paleontologists in very isolated region of Chilean Patagonia (the valley of Las Chinas River) faced a unique challenge: with freezing weather and only 5 days of fieldwork left, they needed to extract a block of rock from a steep hill with a few surfacing bones, presumably containing the articulated skeleton of a 74 million year-old dinosaur. After much effort (including near hypothermia, and team members with a sprained ankle and broken rib) they achieved their goal. At the time, slender exposed limb bones suggested these could belong to a small-sized dinosaur (about 1,5 m), perhaps a bipedal herbivore: an "ornithopod". However, once in the lab, as the rock was slowly chipped away, a succession of scientific surprises would come about.  

The first surprise was the greatest of all: the tail was unlike any dinosaur. It possessed a large weapon composed of 7 paired, laterally projected osteoderms (dermal bones) in an arrangement like a fern frond, covering half the tail. Because of these large osteoderms, we realized this was not an ornithopod, but an armored dinosaur. The weapon was the namesake for the new genus, Stegouros, which means "roofed tail"; the species elengassen is the name of a mythical armoured monster in the lore of the native Aonik'enk people.

Image: Francisco Hueichaleo

At the time, no slender-limbed armoured dinosaurs were known to have elaborate tail weapons. Tail weapons seemed to have evolved only in the stout-limbed and broad-footed forms (Eurypoda), specifically in the Stegosauria, and in "advanced" ankylosaurs (Ankylosaurinae). The spectacular Stegosauria are among the most well-known dinosaurs for their upright dorsal plates, and their tail weapon of paired spikes. This weapon is also known as a "thagomizer", a term coined in a joke by cartoonist Gary Larson, and thereafter adopted by scientists.  Rather than upright plates,  the Ankylosauria had broad, heavily armoured backs laid with numerous osteoderms.

Illustration: Luis Pérez López

Early ankylosaurs did not have tail weapons, but the subfamily Ankylosaurinae is famous for evolving a large rounded club at the end of their tails. Clearly, the tail weapon of our dinosaur was "none of the above". Rather than a thagomizer or a rounded club, it reminded us of a macuahuitl, the dreaded war club/sword used by the ancient aztecs.

The fossil of Stegouros was preserved with its posterior half ("waist down") fully articulated and complete, in a deeper position than the anterior half of the animal, which was scattered and missing a few elements. The evidence suggests that the posterior half of the animal was buried quickly at a river bank, while the upper half lay exposed for a while and fell apart before it, too, was buried. It is possible that this dinosaur was stuck in a death trap such as quicksand; its feet were straightened out, which is uncommon (they are folded in most carcasses), and it was also found belly down, unlike carcasses of armoured dinosaurs that have been transported by a river, which tend to be belly up. A detailed discussion of the conditions in which this dinosaur may have been preserved is provided in the supplementary information of our paper.

As preparation continued, the hip of this dinosaur provided yet another surprise: it had a remarkably stegosaur-like shape, even with a sheet of bone covering the sacrum, just like that found in some stegosaurs. The rest of the body (postcranial skeleton) also showed a few stegosaur-like traits, while ankylosaurian traits were almost absent. Most of the anatomy is very close to what would be expected for the last common ancestor of ankylosaurs and stegosaurs. For a long time, we believed our armoured dinosaur could be closer to stegosaurs,  but had split off very early from the rest, before the evolution of vertical dorsal plates and the thagomizer. Eventually, we began doubting the "stegosaur" hypothesis: we realized that skull fragments showed important ankylosaurian characters, such as inset tooth rows, and a maxillary with a palate. We also moved closer towards the "ankylosaur" hypothesis when we realized our dinosaur  had specific resemblances to two of the most interesting ankylosaurs ever discovered: Antarctopelta from Antarctica, and Kunbarrasaurus from Australia.

The mystery of gondwanan ankylosaurs

When Antarctopelta and Kunbarrasaurus were described in the 1980's these were groundbreaking contributions because previously, ankylosaurs were only known from the northern hemisphere (the ancient supercontinent of Laurasia), where their fossil record is abundant and very diverse. To this day, the information about armored dinosaurs from the southern supercontinent of Gondwana is very scarce, but it is has also become clear that they must represent a very important missing chapter of their evolutionary history. For instance, a rib attached to an impressive spike-shaped osteoderm was recently found in Africa, which likely belong to a bizarre early ankylosaur. Antarctopelta is intriguing, with a very unusual anatomy for an ankylosaur; unfortunately, it is very incomplete, with only about 15% of the skeleton preserved. Kunbarrasaurus in turn is largely complete and has a very well-preserved skull that demonstrates unquestionable ankylosaurian affinities. The postcranial skeleton of Kunbarrasaurus is mostly embedded in rock, and remains to be fully studied; however, what has been described so far also suggests it is very different from other ankylosaurs. Indeed, both Kunbarrasaurus and Antarctopelta are placed by most authors in the earliest branches of Ankylosauria. Antarctopelta was found in the antarctic peninsula, and lived around the same time as Stegouros, when both regions where much closer, allowing temporary land bridges and dispersal of organisms between continents. Antarctica, in turn, was also connected to Australia. When we began working at Chilean Patagonia, we were hoping we could find remains of weird ankylosaurus. In South America, only unnamed isolated bones and teeth of ankylosaurs had ever been found. Further, scientists suspected these remains could correspond to forms that originated in North America, that somehow dispersed to South America in the late Cretaceous.  The 80% complete skeleton of Stegouros exceeded our most optimistic expectations, revealing an animal that is clearly distinct from the northern ankylosaurs. 

The Parankylosauria are born

When comparing Stegouros to published information about Kunbarrasaurus, we found that both shared a stegosaur-like covering of the sacrum, which in the end turned out not to be so stegosaur-like: in stegosaurs, it is composed by extensions from the sacral vertebrae, whereas in Kunbarrasaurus and Stegouros, it is formed by a separate layer of dermal bone. Fragments of flat dermal bone that may correspond to a sacral covering are also present in Antarctopelta. All three dinosaurs also share the presence of numerous roughly square-shaped ossicles (small osteoderms), with a specific pattern of orthogonal striations on their underside. The fossils of Antarctopelta include some very large and enigmatic osteoderms, one of which had been suggested to be a skull bone fused to an osteoderm. However, we found that these enigmatic osteoderms are a very close match to those of the tail weapon of Stegouros. Antarctopelta also had especially low tail vertebrae. These are so unusual for an ankylosaur, that some authors had suggested they could belong to a plesiosaur, a marine reptile whose remains could have ended up mixed with those of this dinosaur (Antarctopelta was preserved in a coastal environment). However, these low vertebrae are identical to those within the tail weapon of Stegouros, who presents an especially short and flattened tail associated to the evolution of the macuahuitl. Given the combined evidence of specialized tail vertebrae and large osteoderms, we proposed that Antarctopelta also possessed a macuahuitl. CT scanners allowed us to observe the specialized morphology of the vertebrae encased within the macuahuitl of Stegouros, as well as the flattened sword-like shape of the tail space inside, with no need to mechanically intervene the fossil.

Stegouros may well be considered a sort of “rosetta stone" that allows making sense of gondwanan ankylosaurs: Key resemblances between Stegouros and Kunbarrasaurus are in anatomical regions that are not preserved in Antarctopelta (humerus, pelvis, skull bones), while key resemblances between Antarctopelta and Stegouros are not preserved in Kunbarrasaurus (elements of the distal limbs and tail). Stegouros ties all this information together to provide a new panorama of gondwanan ankylosaurs, and how they may have  differed from northern forms: They tend to be smaller-sized, have slenderer limbs, and be more lightly armoured. At least some forms would also present the distinctive macuahuitl. More definitive support for the hypothesis that Stegouros was an ankylosaur was provided by consistent results of phylogenetic analysis: using modified data matrices from five different previous studies, our dinosaur was always placed closer to Ankylosauria than to Stegosauria, and was always grouped with Antarctopelta and Kunbarrasaurus, conforming a branch of gondwanan ankylosaurus that split off earliest from all other Ankylosauria. Given these results, we formally proposed the name Parankylosauria ("at the side of ankylosauria") for all ankylosaurs closer to Stegouros than to Ankylosaurus, the iconic tail-clubbed form of the north; conversely, we proposed the name Euankylosauria ("true ankylosaurs") for all ankylosaurs closer to Ankylosaurus than to Stegouros.

Evolutionary conclusions

Stegouros and the newly recognized Parankylosauria bring forward several interesting evolutionary conclusions:

1) The broad-footed armoured dinosaurs (named Eurypoda for this sake) were (paradoxically) likely not broad-footed in the beginning; rather, Stegosauria and Euankylosauria independently evolved broader feet, from a last common ancestor with slender feet as in Parankylosauria.

2) Some traits that are considered to be typically stegosaurian are also present in Parankylosauria; therefore, they may well be ancestral traits, that were secondarily lost in Euankylosauria.

3) The evolution of the macuahuitl in Parankylosauria was completely unrelated to the evolution of the thagomizer in Stegosauria, and was also unrelated with the tail club in Ankylosaurinae. While it may be tempting to imagine a macuahuitl-like intermediate leading to the tail club of Ankylosaurinae, the fact remains that early Euankylosauria did not have a tail weapon, much like older lineages of armoured dinosaurs such as Scelidosaurus. Further, as revealed by CT scans, the tail weapon of Stegouros involved a set of anatomical modifications that are very different to those leading to a tail club in Ankylosaurinae. The conclusion is that armoured dinosaurs can now be enthroned as unique among land vertebrates, being the only lineage to have independently evolved three radically different kinds of specialized tail weapons.

Perspectives

Finally, the Parankylosauria are lacking many specialized traits of Euankylosauria that were already present even in the earliest forms of the mid-Jurassic, some 165 million years ago; therefore, the Parankylosauria must have split off from them before that time. Kunbarrasaurus from the late lower Cretaceous (100 million years ago) helps fill in the time gap with Late Cretaceous forms like Stegouros and Antarctopelta, but it is clear that most of the evolutionary history of Parankylosauria still remains to be uncovered.

-Alexander O. Vargas

-Sergio Soto-Acuña

Soto-Acuña, S., Vargas, A.O., Kaluza, J. et al. Bizarre tail weaponry in a transitional ankylosaur from subantarctic Chile. Nature (2021). https://doi.org/10.1038/s41586-021-04147-1

Image: Mauricio Álvarez Abel

Alexander Vargas

Professor, Universidad de Chile