Some implications Adalatherium brings to the table

Just yesterday a remarkable new fossil was anounced. Literally “mad beast”, Adalatherium is both a relatively large Mesozoic mammal (hare sized) as well as the most complete gondwanathere fossil known. For those not in the know, gondwanatheres are a clade of mysterious herbivorous synapsids that lasted from the Late Cretaceous to possibly the Miocene (Chimento 2015), with putative Jurassic specimens further stretching their time on earth (Chimento 2016). Once considered to be xenarthrans, they are now unambigously considered non-therian mammaliaforms, though debate rages on if they are true mammals closely related to multituberculates or haramiyidans (Huttenlocker 2018). This study favours a fully monophyletic Allotheria more closely related to therian mammals than to monotremes, but it’s pretty obvious this debate is far from settled.

Gondwanatheres, like either of their putative relatives, are superficially rodent like, but as far back as Vintana that we know they had weird traits like weird facial flanges (ironically similar to those of sloths), a palinal (front-to-back) chewing style not seen in any modern mammal aside from possibly dugongs and large nasal openings. But now that we have an actual post-cranial skeleton we can see how truly bizarre these animals truly were. Adalatherium has a mosaic of features both primitive and advanced for mammal standards, some of which it acquired independently from therian mammals. Most notable are its limb bones, which are for all intents and purposes “modern” and allowed it a more gracile gait not seen in basal mammals like monotremes. Also, what’s up with that gigantic heel bone?

Up to this point gondwanatherians were depicted as somewhat robust gopher-like animals, basically what happens when you glue a beaver head to a platypus. In retrospect, gracile multituberculates and haramiyidans should have informed us otherwise, but we can now see that gondwies attained not only a higher degree of diversity than previously speculated but actually produced forms compared to modern cursorial herbivores like rabbits and caviomorphs. This should be interesting when discussing the extinction of these animals as rodents are often cited as being the cause of multituberculate decline due to, among other things, supposed higher agility (Adams 2019).

The putative gondwanathere Patagonia alongside the late dryolestoid Necrolestes. While Patagonia was likely a tuco-tuco-like animal, the same can’t be said for the more agile Adalatherium.

In fact, Adalatherium may also offer a new way to scale gondwanatheres. Previously, it was assumed their heads were proportionally huge compared to the body, but Adalatherium has a relatively small head, suggesting that the bodies of other gondwanatheres might have been larger. This might be particularly true for the contemporaneous Vintana, which has a much larger skull.

Diagram by Joschua Knüppe, scaling up Vintana via comparison with Adalatherium. See scale with the contemporary herbivorous crocodylomorph Simosuchus.

Adalatherium is recovered as its own new family, Adalatheriidae, but in practise it is simply the outgroup to the well-established gondwanathere family Sudamericidae, so it can be in a way considered a good analogue for this clade’s currently speculative bauplan. Sudamericids are considered “advanced” in comparison to other gondwanatheres, having hypsodont teeth like those of modern grass eating mammals. They are present in the Late Cretaceous in Africa, Madagascar and India, but show up in the South American and Antarctic Cenozoic, implying that they probably dispersed southwards and westwards after the KT event.

This is in part because they are the largest and most diverse mammals in this region, suggesting an origin in either Africa, India-Madagascar or in Gondwana prior to the break up of these regions. During the Cretaceous their ecological niche was occupied by mesungulatid dryolestoids in South America, which in turn are absent from India-Madagascar. This suggests that both groups likely did not co-exist, and that the extinction of mesungulatids might have prompted the brief expansion of gondwanatheres during the Cenozoic (or might have even been outcompeted by them).

Likewise, with the possible exception of stegosaurs there are no Late Cretaceous ornithischians in India-Madagascar, suggesting that gondwanatheres, zhelestid eutherians and herbivorous crocodylomorphs supplanted these dinosaurs in the small and mid-sized herbivore range. This has some precedent; omnivorous crocodylomorphs apparently supplanted mammals and small herbivorous dinosaurs in the Kem Kem Formation (Ibrahim 2020), so it goes to show how the Mesozoic wasn’t just a Jurassic Park.

Finally, Adalatherium again has a pretty large nasal cavity. I previously speculated that gondwanatheres could potentially evolve trunks, and that certainly reinforces that idea.

So in conclusion:

  • Gondwanathere phylogeny is wacky
  • Gondwanatheres had a mosaic anatomy
  • Gondwanatheres might be larger than you think they are
  • Sudamericid gondwanatheres might have crossed the sea during the Cenozoic
  • Its possible that the absence of small herbivorous dinosaurs in Late Cretaceous India-Madagascar isn’t a coincidence


Nicolás R. Chimento, Federico L. Agnolin and Fernando E. Novas (2015). “The bizarre ‘metatherians’ Groeberia and Patagonia, late surviving members of gondwanatherian mammals”. Historical Biology: An International Journal of Paleobiology. 27 (5): 603–623. doi:10.1080/08912963.2014.903945.

Nicholas Chimento, Frederico Agnolin, Agustin Martinelli, Mesozoic Mammals from South America: Implications for understanding early mammalian faunas from Gondwana, May 2016

Neil F. Adams, Emily J. Rayfield, Philip G. Cox, Samuel N. Cobb, and Ian J. Corfe, Functional tests of the competitive exclusion hypothesis for multituberculate extinction, R Soc Open Sci. 2019 Mar; 6(3): 181536. Published online 2019 Mar 27. doi: 10.1098/rsos.181536

Nizar Ibrahim, Paul C. Sereno, David J. Varricchio, David M. Martill, Didier B. Dutheil, David M. Unwin, Lahssen Baidder, Hans C. E. Larsson, Samir Zouhri, Abdelhadi Kaoukaya, Geology and paleontology of the Upper Cretaceous Kem Kem Group of eastern Morocco, ZooKeys 928: 1–216 doi: 10.3897/zookeys.928.47517