Tuesday 11 February 2020

Horn function in Arsinoitherium OR... the ArSUMOitherium Hypothesis™

Hello, 2017 painting of sheathed-horn Arsinoitherium zitteli. Time to see if we can figure out what those horns were for, other than looking regal in artwork.
Recently, I was in the Teylers Museum, Haarlem, for the opening of their sensational palaeoart-themed Dinomakers exhibit*, where a fine cast of the skull of the Palaeogene Afro-Arabian embrithopod Arsinoitherium zitteli is perched in the main fossil gallery. Arsinoitherium is a pretty darn fascinating mammal that we're all familiar with, yet we rarely give exclusive focus too. I assume this reflects its scientific vintage. Arsinoitherium was discovered well over a century ago and is now featured in so many books and museums that it's part of the popular palaeo furniture. Its size, fantastic cranial horns and status as the last of the embrithopods make it a remarkable and charismatic fossil species, but it just doesn't seem to be cutting it with the kids.

*If you're a regular reader and are in the Netherlands before June, you really want to check this out. It has heaps of original palaeoartworks, including many by classic artists - Hawkins, Knight, Burian, and some exceptional modern work from the Kennis brothers. I have stuff there too.

Today, Arsinoitherium is mostly discussed by palaeontologists documenting new Palaeogene faunas in Eocene-Oligocene sites of Africa and the Arabian Peninsula, but several fascinating behavioural and ecological revelations about this sometimes controversial mammal have also been published in recent years. It's increasingly apparent, for instance, that Arsinoitherium was anatomically conservative, the oldest members of the lineage being little different, morphologically speaking, to the youngest. Despite this, it was a very long-lived and widespread genus which must have been highly adaptable to demonstrate sustain such a broad geographic and stratigraphic range (Jacobs et al. 2005). It evidently lived in a range of habitats, the most surprising of which are upland regions well away from the coastlines and estuaries thought to be traditionally Arsinoitherium country (Kappelman et al. 2003). This adaptability occurs despite the unusual multifunctional dentition of Arsinoitherium being adapted to specialised browsing (Court 1992): perhaps it was more of a generalist than we realised.

More specific insights into Arsinoitherium palaeobiology have been proposed too. Studies of their ear anatomy, principally for their phylogenetic signature, have found adaptations to hear infrasound in the same manner as modern elephants (Benoit et al. 2013), and a compelling case is being built against the popular idea that Arsinoitherium was a hippo-like semi-aquatic animal. This proposal was founded on both functional and taphonomic grounds (e.g. Court 1993) but a suite of opposing data, including tooth wear, occurrences in relatively dry palaeoenvironments, the presence of graviportal limbs and details of bone chemistry, are now pointing to more terrestrial habits (e.g. Clementz et al. 2008; Sanders et al. 2010). Revisions to Arsinoitherium taxonomy are also shedding insights on behaviour. Two Arsinoitherium species are now recognised - the well known A. zitteli and the larger, relatively longer-legged A. giganteum - and prior taxonomic distinctions accounting for a third species are now interpreted as evidence of probable exaggerated sexual dimorphism in A. zitteli (Sanders et al. 2004, 2010). Behind the scenes, we're building a developed picture of what Arsinoitherium was like as a real animal, and not just a long-standing museum fixture.

Image result for arsinoitherium knight
One of my favourite images of Arsinoitherium is this 1907 piece by Charles Knight, published in Osborn (1907). Despite that awesome headgear, Arsinoitherium isn't often illustrated doing much else than standing around, so it's nice to see it having something to do. Hat-tip to Chris Manias for posting this image and making me aware of it.
But one aspect of Arsinoitherium palaeobiology that does not seem to have been discussed at length is horn function. Long-term readers may recall that this is not the first time I've mentioned these structures, as the life appearance of Arsinoitherium horns was the exclusive subject of a 2017 blog post. The take-home of that article is that we artists have probably been incorrect in generally restoring Arsinoitherium horns with facial skin. Rather, their horn surface texture, structure and growth mechanic is consistent with a bovid-like horny sheath. This is not a new idea, with sheathed horns being proposed by several authors (e.g. Andrews 1906; Sanders et al. 2010), but contradicted by others (e.g. Prothero and Schoch 2002; Rose 2006).

Assessing life appearance already tells us something about horn function as a covering of tough, insert tissue has some major biomechanical implications. Arsinoitherium horn cores are deceptively delicate on account of their hollow construction. Despite the skulls of these animals reaching over 80 cm long and their owners attaining masses of around two tonnes (Sanders et al. 2010), Arsinoitherium horn cores were constructed from bones just 5-10 mm thick. Without additional protection, such delicacy might prohibit antagonistic use and a more passive function would seem likely, such as visual communication or acoustic augmentation (sensu Benoit et al. 2013). But modern species show that a horny sheath over a hollow horn core creates an amazingly strong, impact-absorbing and bending resistant organ that can be used to bludgeon, wrestle and lance other animals or to forcefully modify the surrounding environment. The physics of this is pretty simple: the hollow bone core provides great bending resistance and reduces weight, while the horny sheath absorbs and dissipates shocks and impacts (Drake et al. 2016). We've seen this exact configuration evolving time and again across Tetrapoda, and its presence in Embrithopoda shouldn't be viewed as weird or improbable.

Partly restored left horn of the Teylers' Arsinoitherium zittelli skull showing the characteristic epidermal correlates (numerous oblique foramina and anastomosing blood vessels) for a bovid-like horn sheath. These horns are identical in texture to what you might see under a cow horn.
But while the horns themselves look formidable enough, their use would be limited without a substantial neck to support and wield the head. It's for this reason that I was pleased to see the Teylers' skull without any pesky postcrania obscuring details of its posterior face. The rearward aspects of animal skulls are often overlooked in favour of more spectacular anatomy but, if you're seriously interested in the functional morphology of fossil animal crania, you need to look at the occiput and other aspects of the posterior skull surface to assess the head/neck soft-tissues. These regions reveal much about neck muscle size and distribution, as well as something of head mobility via the shape of the occipital condyle. Even at a glance, you can often say something intelligent about how animals were wielding their heads by looking at the posterior skull.

I was thus greatly interested to see that the Teylers' posterior Arsinoitherium skull bore several features unfamiliar to me from other large animals. Here's what you can see of the back the skull as it stands in the Teylers gallery. I think there's some reconstruction in places but the skull of Arsinoitherium is completely known from several specimens, and any sculpting seems to be a faithful recreation of real anatomy.

Posterior view of the Teylers Museum Arinoitherium zitteli skull cast, as seen in January 2020.
And here's the same thing, more or less, as illustrated by Andrews (1906):

Image from Andrews (1906, courtesy Wikimedia), public domain.
The exciting parts here are not the unsurprisingly robust nature of the skull-neck junction or the general indications of expansive neck musculature. Nor even is it the substantially-sized occipital condyle that is almost as wide as the skull itself, and has a shape seemingly permitting more dorsoventral motion than lateral (a thought posited previously by Andrews, 1906). Rather, the interesting aspect is the unusual configuration of the bones surrounding the occipital condyle. Most animals, even species with large, heavy heads, have relatively flat occipital faces, but the medial dorsal region of the Arsinoitherium occiput is deeply recessed between two large protuberances which extend posteriorly almost as far as the occipital condyle. The dished medial region extends forward quite some way, projecting far over the braincase to form a deep depression in the skull roof (below). The neighbouring protuberances are prominent, posteriorly-directed outgrowths of the dorsal occipital margin (the superior nuchal line) which curve somewhat towards the skull midline, and are supported below by thick bony buttresses. I've looked for similar anatomy in a number of other large mammal skulls and, while my research isn't exhaustive enough to claim Arsinoitherium has an entirely unique posterior skull configuration, I'm happy to declare it unusual.

Dorsolateral view of the Arsinoiherium zittelli occiput, showing the large basin formed by the dorsal region and the two neighbouring projections. Note the complex surface and texturing, indicating scars and attachment sites of neck musculature.
Since seeing this, I've been wondering what it tells us about how Arsinoitherium neck tissues were arranged and what that might mean for horn function. I decided that a good place to start was a stab at reconstructing the muscles of the occipital region, which you can see below. A word of caution about this image: this is not a watertight study of Arsinoitherium specimens based on days and days of work, but more an attempt to get a basic understanding of what that peculiar anatomy represents if we assume conventional mammal occiput myology. I like to think it's not total garbage, but don't treat it as gospel either. I included the classic Gray's Anatomy human occiput illustration in there, scaled to the size of an average human adult (≈ 30 mm wide foramen magnum, apparently), to ram home how large the skulls of Arsinoitherium are.

My attempt to figure out what's going on at the back end of the Arsinoitherium skull. Skull outline after Andrews (1906), with some minor modification (including removal of the restored horn tips). That's an 'average' human occiput on the right, taken from Wikipedia (public domain). Does anyone else feel weirdly inadequate when looking at this image? I mean, I know it's not all about size and all, but still...
If my noodling on this is correct, then dorsomedically-anchoring muscles typically involved with elevating the neck (e.g. Semispinalis capitis, Trapezius) are now anchored partially on the skull dorsal surface, with the anteriormost located some distance forward of the occipital condyle. Such a configuration surely means that their contraction would not only elevate the neck (as expected) but also tip the head upwards to an unusual extent, and the increased distance between the occipital condyle and these muscles signifies a longer lever arm, and thus greater torque, on the head-neck joint. The lateral protuberances aren't quite in the right place for neck elevators however, and I initially wondered they were something to do with jaw musculature. Expanded temporalis muscles often create extended crests at the back of animal skulls but this is not the case in Arsinoitherium, where the temporal muscle housing clearly terminates well anterior to the occipital face. It seems more likely that these protuberances are something to do with laterally-placed skull-neck muscles - perhaps a set of considerably expanded obliquus capitis superior. These are muscles which run between the atlas (the first neck vertebra, a structure which is also huge in Arsinotherium) and the posterior skull to deliver fine control to head elevation and lateral rotation. But because the protuberances have migrated to somewhat overhang the atlas vertebra, the vertical action of these muscles was likely enhanced relative to other mammals. As with the muscles of the dished medial occipital region, this realignment of the oliquus capitis superior would likely see the head pitching up during contraction. Further large muscles are indicated by the broad mastoid and jugular processes, regions which anchor muscles that variably elevate, rotate and laterally flex the head and neck.

All being equal, it seems that the posterior Arsinoitherium skull wasn't just about supporting the head with a series of big, powerful muscles, but also specifically configured to enhance the extension of the craniocervical joint - in other words, to forcibly swinging the head upwards relative to the neck. Much of the rearrangement of the posterior skull seems to be geared towards this, both in terms of expanding muscle attachment area and also reorienting muscle vectors to better serve vertical head motion. This, of course, also fits well with observations that the occipital condyle is structured to facilitate more dorsoventral movement than lateral. I suspect we do not see an equivalent configuration in other mammal skulls because most heavy-headed mammals also possess expanded head-neck muscles anchored to withers (tall vertebral spines of the shoulder region). This configuration allows them to lift their heads and necks around using drawbridge-like actions whereas Arsinoitherium, which lacked significant withers (Andrews 1906), probably had to rely more on muscles localised around neck vertebrae to support and move its head.

Skull variation in Arsinoitherium, as figured in my previous blog post on Arsinoitherium. Note the changing posterior shape from the smallest to the largest skull.
What might all this mean for horn function? It goes without saying that enhanced adaptations for swinging a skull upwards could have a lot of functional implications when that skull is covered in horns. It seems reasonable to assume that Arsinoitherium could use this for a number of practical purposes, such as taking forceful swipes at predators or using its headgear to knock over trees and other vegetation to access certain food sources. However, it's noteworthy that the only the biggest (potentially male?) Arsinoitherium that have the most developed version of the complex occiput morphology outlined here (see image above), suggesting that enhanced head and horn motion was of principal use for big, mature animals concerned with territories, mates and other resource competition. Might this indicate that intraspecific combat, such as horn-locked wrestling matches with rival individuals, was an adaptive pressure here? I'm not aware of tests to see how well Arsinoitherium horns interlocked (as has been done for horned dinosaurs - see Farke 2004) but they certainly look like they'd slot between each other in a way that would allow for intraspecific wrestling, and powerful neck and head elevators would be useful to shove and unbalance opponents, deliver pointed jabs and parry incoming blows. If Arsinoitherium sheathed horns were as strong as those of modern mammals I suspect they could easily withstand the strain of such bouts, and their wide occipital condyle and cervical series would do well to resist the torsion incurred by wrestling activity. Of further significance is that Court (1993) noted that the limbs of Arsinoitherium were adapted for forceful retraction, a feature he assumed was useful for swimming. But a terrestrially wrestling Arsinoitherium would find that useful too, as powerful limb actions would push the body forward against a rival. I'm not sure these retractors would indicate running and charging behaviours however because, even with strong limb muscles, Arsinoitherium has stumpy distal bones ill-suited to rapid locomotion. In my mind, I'm visualising this hypothesis as four-horned sumo wrestling over jousting or fencing.

The idea of Arsinoitherium using its horns aggressively is not, of course, a radical or special insight - any three-year-old could make the same suggestion based on the observation that sharp, pointy bits of animals tend to be used in such ways. But I think it's neat that there might be an overlooked functional signature of this behaviour in the predicted tissues and structure of the horns as well as the morphology of the posterior skull, and suggest this might warrant further research. It seems to fit multiple aspects of Arsinoitherium functional morphology and chimes well with behaviour in large living herbivorous mammals: it's actually difficult to think of large mammals with substantial horns or tusks that don't use them for intraspecific fights. It'd be cool to see this investigated further, but that's beyond the scope of this article. For now, I'll leave you with my artistic take on what's clearly got to be called the Arsumoitherium Hypothesis - with a name like that, this idea has to be correct, right?

Two Arsinoitherium zitteli engaged in a wrestling bout - does antagonistic behaviour explain those powerful head extensors? Those neck humps aren't muscle, by the way, but rhino-like pads of neck tissue. Also, has anyone else rendered Arsinoitherium in this way? I can't find any other examples, but also refuse to believe that no-one has illustrated something similar since Arsinoitherium was described in 1903.

Enjoy monthly insights into palaeoart, fossil animal biology and occasional reviews of palaeo media? Support this blog for $1 a month and get free stuff!

This blog is sponsored through Patreon, the site where you can help online content creators make a living. If you enjoy my content, please consider donating $1 a month to help fund my work. $1 might seem a meaningless amount, but if every reader pitched that amount I could work on these articles and their artwork full time. In return, you'll get access to my exclusive Patreon content: regular updates on upcoming books, papers, paintings and exhibitions. Plus, you get free stuff - prints, high-quality images for printing, books, competitions - as my way of thanking you for your support. As always, huge thanks to everyone who already sponsors my work!


  • Andrews, C. W. (1906). A descriptive catalogue of the Tertiary Vertebrata of the Fayum. Publ. Brit. Mus. Nat. Hist. Land. XXXVII.
  • Benoit, J., Merigeaud, S., & Tabuce, R. (2013). Homoplasy in the ear region of Tethytheria and the systematic position of Embrithopoda (Mammalia, Afrotheria). Geobios, 46(5), 357-370.
  • Clementz, M. T., Holroyd, P. A., & Koch, P. L. (2008). Identifying aquatic habits of herbivorous mammals through stable isotope analysis. Palaios, 23(9), 574-585.
  • Court, N. (1992). A unique form of dental bilophodonty and a functional interpretation of peculiarities in the masticatory system of Arsinoitherium (Mammalia, Embrithopoda). Historical Biology, 6(2), 91-111.
  • Court, N. (1993). Morphology and functional anatomy of the postcranial skeleton in Arsinoitherium (Mammalia, Embrithopoda). Palaeontographica Abhandlungen A, 226, 125-169.
  • Drake, A., Donahue, T. L. H., Stansloski, M., Fox, K., Wheatley, B. B., & Donahue, S. W. (2016). Horn and horn core trabecular bone of bighorn sheep rams absorbs impact energy and reduces brain cavity accelerations during high impact ramming of the skull. Acta Biomaterialia, 44, 41-50.
  • Farke, A. A. (2004). Horn use in Triceratops (Dinosauria: Ceratopsidae): testing behavioral hypotheses using scale models. Palaeontologia Electronica, 7(1), 10p.
  • Jacobs, B. F., Tabor, N., Feseha, M., Pan, A., Kappelman, J., Rasmussen, T., ... & Massini, J. L. G. (2005). Oligocene terrestrial strata of northwestern Ethiopia: a preliminary report on paleoenvironments and paleontology. Palaeontologia electronica [electronic resource]. Vol. 8, no. 1 (2005): 19 p.
  • Kappelman, J., Rasmussen, D. T., Sanders, W. J., Feseha, M., Bown, T., Copeland, P., ... & Jacobs, B. (2003). Oligocene mammals from Ethiopia and faunal exchange between Afro-Arabia and Eurasia. Nature, 426(6966), 549-552.
  • Osborn, H. F. (1907). Hunting the Ancestral Elephant in the FayĆ»m Desert: Discoveries of the Recent African Expeditions of the American Museum of Natural History. Century Company, October 1907, 815-835.
  • Prothero, D. R., & Schoch, R. M. (2002). Horns, tusks, and flippers: the evolution of hoofed mammals. JHU Press.
  • Rose, K. D. (2006). The beginning of the age of mammals. JHU Press.
  • Sanders, W. J., Kappelman, J., & Rasmussen, D. T. (2004). New large-bodied mammals from the late Oligocene site of Chilga, Ethiopia. Acta Palaeontologica Polonica, 49(3), 365-392.
  • Sanders, W.J., Rasmussen, D.T., & Kappelman, J. (2010). Embrithopoda. In: Werdelin, L., Sanders, W.J. (Eds.), Cenozoic mammals of Africa. The University of California Press, Berkeley, Los Angeles, London, pp. 115–122.