Evolution of the Bizarre Arms of Oviraptorosaurs
I don’t need to tell you how awe-inspiring dinosaurs are, but I would like to tell you about how weird they were. There is no shortage of bizarre dinosaurs, from the hotly debated Spinosaurus with a billboard for a sail to the almost ridiculously ornamented skulls of ceratopsians like Kosmoceratops and Diabloceratops. Among the weirdest are the oviraptorosaurs. Lasting up until the cataclysmic end of the Mesozoic, oviraptorosaurs diversified into numerous forms that have fascinated and dumbfounded scientists. The discovery of their palaeobiology is an ongoing adventure, with researchers in several locations across the globe uncovering and describing new oviraptorosaur fossils and others revisiting already published fossils with complex analytical methods to find new perspectives. Among the latter is Milly Mead, a PhD student at the University of Edinburgh, who’s recently published study changes the way we think about the evolution of arms and hands in oviraptorosaurs, complicating our understanding of dinosaur evolution. The evolution of this group is a fascinating tale told by many exceptionally well-preserved fossils, and it’s now more interesting than ever.
The rise of oviraptorosaurs is imperfectly understood, as genealogical analyses suggest they originated sometime in the Jurassic, but the oldest definite fossils of them are known from the Early Cretaceous, some 125 million years ago. Until their demise at the hands of an Everest-sized extraterrestrial projectile 66 million years ago, they diversified into numerous forms that were important parts of Asian and North American environments. Among the group’s oldest representatives is Incisivosaurus, who’s buck-teeth were likely adapted for an omnivorous diet, indicating that the shift away from the carnivorous diet typical of theropods occurred early on in the evolution of oviraptorosaurs. Exceptionally well-preserved fossils of other early forms like Caudipteryx directly inform on the life appearance of these animals, as contour feathers are preserved along their arms, meaning that these animals were winged yet flightless. The majority of oviraptorosaur species we know off don’t belong to these early forms though, but instead to the two main lineages within the group
From a quick glance, both these lineages – the Oviraptoridae and the Caegnathidae – are somewhat similar. A cassowary-like body plan, parrot-like heads, and toothless beaks are features uniting them, and several members of both groups possess ornate head-crests, which are connected to the respiratory system via air sacs. One way these two lineages did differ is the evolution of their hands. Ancestrally, oviraptorosaur hands consist of three digits (the equivalents of all but our pinkie and ring fingers), and while this is retained throughout the caegnathids, there is evidence for shrinking of the third digit (our equivalent is the middle finger) relative to the size of overall hand in the oviraptorids. Oksoko avarsan, a species from the very end of the age of dinosaurs, represents the culmination of this evolutionary trajectory with its simple and scrawny third digit that forced it to only have two usable digits. The loss and simplification of digits in theropod dinosaurs is not restricted to oviraptorids though, and is better known in larger, more charismatic forms like Carnotaurus and Tyrannosaurus. Repetition of digit loss in all these distantly related theropods raises questions about their underlying biology that Milly mead and colleagues set out to answer.
Using oviraptorosaurs as their study system, the researchers compiled a dataset of measurements of various elements of the forelimb (like the humerus and claws) of these theropods as well as for the femur, which is the largest leg bone and has dimensions that directly reflect overall body size (as it is one of the main weight-bearing elements). The total dataset consisted of over 50 measurements for 25 different oviraptorosaurs, ranging from primitive forms like the wonderfully-preserved Caudipteryx to the highly derived Oksoko. Afterwards, these measurements were subjected to various sorts of complex mathematical modelling that could then be plotted on graphs to visualise if and how the evolution of some elements of the skeleton is linked to the evolution of others. I won’t bog you down in the intricacy behind the mathematics, and I would encourage any curious readers to read the research article to learn more. The results of their computing revealed the evolution of oviraptorosaur arms, and likely those of theropods more widely, was far from straightforward.
One of their major discoveries is that the reduction of the size of the arms and the loss of digits do not seem to be associated, but not in the way that has been historically thought. More specifically, only the evolution of the claw (and the nearest bone) of the third digit are not linked to the evolution of the rest of the arm and hand, meaning that while almost the entire forelimb collectively reduced as a single unit, most of the third digit did not follow. Historically, it’s been considered the evolution of the main arm bones and those of the hand and fingers were entirely decoupled, but here it instead seems like the main arm bones, and even most of the hand and fingers are coupled, with only one specific digit following a separate evolutionary course. This phenomenon of specific digits coursing a distinct path to the rest of the arm (including other digits) might be behind the unusual forelimbs of some other theropods like the Triassic Herrerasaurus and the early tyrannosaur Guanlong, both of which have greatly reduced their fourth finger while retaining the ancestral long arms of theropods.
Fascinatingly, they also discovered that oviraptorosaur forelimbs generally fall into 1 of 4 groups depending on their shape. One is that of Caudipteryx and closely related early oviraptorosaur species, another includes various members of both major oviraptorosaur lineages (the oviraptorids and the caegnathids), a third includes a specialised lineage termed the Heyuanninae (which is a sublineage of the oviraptorids), and the final is comprised solely of the bizarre Oksoko (which is a derived member of the Heyuanninae). While this could be interpreted that each distinct group had arms and claws specialised for a certain function, the reality appears to be much more complex. For instance, one of the main ways oviraptorids and caegnathids differ is in the form and function of their lower jaws, so why do so many have forelimbs that look and likely function similarly? At the moment there is no clear answer to this question, but hopefully future research may reveal one. Interestingly, the emergence of a unique forelimb shape in the Heyuanninae seems linked to the rise of this group of oviraptorosaurs in the Gobi Desert, suggesting that their unique forelimb was an adaptation for living in an extreme, arid environment. Oksoko would further modify this unique forelimb shape, resulting in one of the weirdest arms and hands of all theropods, that has even been suggested to be specialised for scratch-digging, a behaviour common among burrowing animals.
To learn more about the evolution of oviraptorosaurs and other theropods, I interviewed the lead author of the research paper this is based on (you can read it at https://doi.org/10.1098/rsos.242114): Milly Mead, PhD student at the University of Edinburgh.
Q1. What motivated you to do this project? Have you always been interested in theropods, and oviraptorosaurs in particular?
A1. I’ve always loved dinosaurs, but what really got me interested in oviraptorosaurs was this amazing fossil of Oksoko. The holotype fossil (the specimen which we use to name a new species) preserves three subadults all huddled around each other in resting postures. It’s rare that we get complete, articulated skeletons – it’s even rarer that these skeletons are preserved in life positions! The three dinosaurs in this fossil would have been sleeping cuddled up to each other, maybe for warmth or for safety in numbers, before they were rapidly buried. It’s very cute but it’s also kind of tragic. So, it was the fossil that got me interested to start with, and then the strange hands and forelimbs made them even more exciting.
Q2. What has been the best part about researching oviraptorosaurs?
A2. Finally getting to share my research! It’s so rewarding to have spent (literally) years working on a project and then to finally have something exciting to show people. I love chatting to people about my work – there’s been a lot of interesting discussions come up and ideas for new angles we could take, which I never would have thought of or considered. It’s the best part about science. There’s such a great community of palaeontologists out there, and it’s really fun when people get excited about research you’ve done!
Q3. Which other methods would you like to use to research the evolution of oviraptorosaurs?
A3. It would be very cool to do some kind of functional analysis of the forelimbs oviraptorosaurs. There’s some really interesting research by a palaeontologist called Sara Burch, which looks at muscle attachment scars on bones to reconstruct the musculature of the forelimb. Applying that technique to the forelimbs of Oksoko or some other oviraptorosaurs might give us a better idea of how they were using their forelimbs – or at least how strong they might have been! I’d also like to look at something called ‘finite element analysis’, which basically uses maths to work out how a structure (like a claw) would respond to different forces. Maybe we could use this to test whether Oksoko’s forelimb really were adapted for scratch digging.
Q4. What other theropods would you like to research? Are you also interested in working on any other types of animals?
A4. The other theropods I’m really interested in are birds. The current project I’m working on is about brain evolution, and birds do some crazy things with their brains. Somewhere in the theropod evolutionary lineage, dinosaurs hugely inflated their cerebrum, gained 20 times more neurons, and became incredibly smart. Some birds are capable of tool use and complex problem solving – that’s crazy! I would love to know why. Big brains cost a lot of energy to grow and maintain, so the trade-off has got to be worth it. What drove these dinosaurs to get smarter?
Q5. What were some of the major challenges you faced while working on this project? Can you tell us about the difficulties you had to overcome in the mathematical modelling and computing part of the project.
A5. The biggest challenge we faced was the missing parts of different fossils. Like I said before, most fossils are far from complete, so we had to deal with a lot of missing data. One of the techniques we used is called ‘Principal Component Analysis’ (usually shortened to ‘PCA’: this is what we used to separate the oviraptorosaurs into groups) and to make this analysis work, the computer needed data for EVERY bone in every species of oviraptorosaur we wanted to look at. This got very annoying, very quickly. Imagine you had a nearly perfectly preserved forelimb: almost everything is there, the humerus, all the claws, the radius, and the ulna, but it’s missing a single bone from one of the fingers. Suddenly, you can’t use it! To make the PCA work, that nearly perfect specimen would need to be left out and we would lose all that nice data. We found a bit of a sneaky way to get around this problem. There’s a piece of software you can use to come up with estimates for missing data based on the data you do have. For example, if two closely related dinosaurs had a humerus, a radius, and multiple hand bones which were very similar lengths, chances are the length of that missing finger bone is also pretty similar. Obviously, we had to use this technique very sparingly – if you use it to estimate too many bone lengths, it gets much less accurate – but it was really helpful for keeping those dinosaurs which were missing just a few parts of the forelimb in the mix for our PCA.
Q6. As so many aspiring palaeontologists are fascinated with theropods, is there any advice you would give to anyone looking into doing research on them?
A6. Make connections and talk to people about what you’re doing/would like to do! Many minds are better than one, and some of the best ideas our research group had for studying oviraptorosaurs came from discussions with people outside our team and even outside the field of palaeontology. Depending on their research backgrounds, everyone will have different perspectives and ideas about how to best approach a problem. Just talking with others about your research will get you thinking outside the box. The best palaeontology is almost always multidisciplinary. My work combines statistics and computer modelling, I know others have used techniques from engineering, microscopy, ecology, and a whole bunch of other fields. Talk to other researchers!
Q7. If I could go back to the Cretaceous, would I be able to shake Oksoko’s hand?
A7. Absolutely. Oviraptorosaurs love to make business deals.
Hady George is a palaeontology PhD student at the University of Bristol researching the jaws of the earliest tetrapods among other things, and seemingly always has a pop science book somewhere in his bag.
Cover image: Cast of a skeleton of the oviraptorosaur Conchoraptor. Image credit to Kabbachi, and used with permission of the following license https://creativecommons.org/licenses/by/2.0
Related Posts
Episode 163: Ecosystem Engineers →
A new species of notosuchian expands our understanding of stem-croc diversity: Epoidesuchus tavaresae →
Episode 161: Notosuchians →
