Suzhousaurus and its strange relatives

When I was first becoming acquainted with dinosaurs during the latter half of the 1980's, the standard "rule" for theropod dinosaurs was that as they grew bigger through the course of time their heads became more robust and their arms grew smaller. Just comparing the Jurassic predator Allosaurus with the Cretaceous Tyrannosaurus rex (numerous pictures of both filling the books I constantly begged my parents to purchase for me) seemed to confirm this, but there was always one very special set of fossil remains that seemed to contradict the prevailing trend. Hung up in the corner at the American Museum of Natural History in New York were a pair of gigantic forelimbs, about twice as long as I was tall during my early visits, tipped with huge claws. The rest of the skeleton of that animal, known as Deinocheirus, was (and still is) missing, but I couldn't help but imagine a huge carnivorous theropod that could rip its prey to shreds. While such terrifying prospects were once considered for the "terrible hands," Deinocheirus appears to be a huge member of the Ornithomimosauria, although precisely resolving its place in an evolutionary context has been difficult given the lack of skeletal material. Just because we don't have any more skeletal material for the massive theropod at present doesn't mean that we never will, however. In recent years the mystery surrounding another gigantic set of claws from the same geologic formation (the Nemegt or Nemegt Svita Formation) as Deinocheirus has begun to be resolved, revealing some of the most enigmatic dinosaurs yet known; the Therizinosaurids.


A cast of the massive forelimbs of Deinocheirus, on display at the Natural History Museum in London, England.

The taxonomic history of the therizinosaurids is a complicated one, especially since so little was known of them until the latter half of the 20th century. Like Deinocheirus, the earliest known remains of Therizinosaurus cheloniformis were massive claws and elements of the forelimb (as well as some from the hind limb), although the slightly curved claws were initially thought to belong to a turtle of enormous size (hence the species name cheloniformis). By the 1970's, however, it was fully recognized that whatever Therizinosaurus was, it was definitely a theropod dinosaur, but where it fit into the systematic scheme of things was still difficult to discern. The 1970's and 1980's were more productive in terms of therizinosaurid discoveries, though, with relatively more complete skeletal material for Segnosaurus and Erlikosaurus and fragmentary material for Enigmosaurus and Nanshiungosaurus coming out of the ground. All of these discoveries came out of Asia (China and Mongolia), and some were grouped within the Segnosauridae as the connection of these taxa to Therizinosaurus would not be made until the 1990's.

Such historical considerations reinforce how bizarre the members of the Superfamily Therizinosauroidea (containing the families Alxasauridae and Therizinosauridae) actually were. While they were theropod dinosaurs, they looked more like prosauropods than "good" theropods. Members of the Therizinosauroidea typically had long arms with long sharp claws, a long neck, a small head with a toothless premaxilla, four functional and forward-pointing toes on the rear foot, relatively small teeth with minimal serrations, a barrel-shaped torso, and the pubis of the hip rotated backwards towards the ischium like in some dromeosaurs (it should be noted, however, that most of the taxa are known from partial or incomplete skeletons so this group could very well undergo many changes as more complete material is found). One recently discovered variety, Beipiaosaurus inexpectus from the famous Yixian Formation, was even preserved with some feathery "integument" or filamentous feather structures associated with the forelimb and pectoral girdle. Given this odd mix of features and the fragmentary nature of the genus that gives the group its name, it is little wonder that these dinosaurs were so puzzling (one genus was even named Enigmosaurus).

As hinted at earlier, it was initially thought that the group was a some sort of Cretaceous holdover of a more primitive group, perhaps a link between the prosauropods and Ornithischian dinosaurs, and some early illustrations of Segnosaurus reconstructed it as a sort of prosauropod. Today, however, these dinosaurs are recognized as theropods despite their strange characteristics, although their earlier evolution is largely unknown. The ~125 million year old North American genus Falcarius might provide some clues as to the origins of the group (the only other N.A. genus known being the geologically younger Nothronychus at about 90 million years old), and the presence of this group in both North American and Asia during the Cretaceous is just another example of the similarity of fauna across the two continents during this time (another example being the North American Tyrannosaurus and the Asian Tarbosaurus).


"Cranial (1-4) and caudal (5-8) dorsal vertebrae in cranial (1, 5), caudal (2,
6), left lateral (3, 7), and right lateral (4, 8) views; dorsal rib in cranial (9) and caudal (10) views. Scale bar equals 10 cm." From Li, et al. "A Large Therizinosauroid (Dinosauria: Theropoda) from the Early Cretaceous of Northwestern China." ACTA GEOLOGICA SINICA

. Vol. 81 No. 4 pp. 539-549

Just last week, however, a new member of the Therizinosauroidea made the news; Suzhousaurus megatherioides. Discovered in the Lower Cretaceous Xinminpu Group of northwestern China, this new genus seems to be the sister taxon of Nothronychus, being more derived than Falcarius or Beipiaosaurus but not quite as much as Alxasaurus and the Family Therizinosauridae, giving it something of an intermediate status within the superfamily. As mentioned before, however, most of the members of this group are known from partial or fragmentary skeletons, and the remains of Suzhousaurus are no exception. While there was enough to definitely identify the new dinosaur as a member of the Therizinosauroidea outside of the Therizinosauridae, much of the skeleton is missing, the dinosaur currently being known from 10 vertebrae, incomplete dorsal ribs, a right humerus, complete left pubes, and other fragmentary or partial elements. This general lack is even more frustrating as the new genus might actually be from another therizinosaur from the same area named "Nanshiungosaurus" bohlini, but there is so little overlap between the remains from both skeletons that it cannot be ascertained whether they are synonyms or not. This is even more frustrating as N. bohlini doesn't seem to be referable to the type species N. brevispinus, and it is likely that the remains of N. bohlini are either from Suzhousaurus or another therinzinosaur of similar size and stratigraphic age.


Reconstruction of Suzhousaurus by Mark A. Klingler of the Carnegie Museum of Natural History.

Taxonomic reshuffling aside, Suzhousaurus is of further significance because of its large size, or at least the large size of its humerus. Even though Suzhousaurus seems to be an earlier member of the therizinosaurs, it appears to have been among the largest, Segnosaurus being slightly larger and Therizinosaurus being substantially larger than both. As the authors of the paper note, this suggests that therizinosaurs attained large size by about 115 million years ago, with the two larger forms existing between 99 and 68 million years ago.

So how did such animals make a living? It is clear that they are maniraptoran theropods but unlike any other group of theropod known, their skulls and dentition apparently being evolved to a herbivorous diet. Such inferences are reasonable and are likely accurate, but this then begs the question of what the therizinosaurs were doing with their large claws. Some have suggested that they were using them in the fashion of modern-day anteaters or pangolins to break into termite mounds, but this is not likely for the larger forms (a dinosaur as large as Therizinosaurus would likely require a staggering amount of termites, or even staggeringly large termites, to meet its daily requirements). Others have suggested something of a defense role for the claws, and it is not unreasonable to assume that the therizinosaurs used their large claws to protect themselves when faced with danger. Still, the claws might have been adapted to their large size for some other reason, and the most popular idea at presnt is that they used their claws to pull down branches to reach vegetation somewhat like the "classic" reconstruction of giant ground sloths. Indeed, the species name of Suzhousaurus megatherioides is a nod to the giant sloth genus Megatherium and it is possible that the giant dinosaur and mammal genera occupied similar niches despite being separated by a wide gap in time. Such speculation has even been taken as far as suggesting that therizinosaurs "sat" when they ate, resting on their backwards-facing hip bones and reaching up into the trees to bring down vegetation, this mode of foraging recalling the strategy of some modern day gorillas. This type behavior cannot be ruled out for the dinosaurs, although the only way of obtaining positive proof for this would be finding some sort of ichnofossil bearing the marks of the feet and hip bones of a large dinosaur from the same age and area known for therizinosaurs (and even then the connection may still be tenuous).

Regular readers will know how much I can't stand most of the "Chased by..." or "Walking With..." BBC cgi-fests, but the reconstructions of Therizinosaurus and Tarbosaurus in this segment are especially impressive.

While the lifestyle of therizinosaurs is open for speculation, the presence of such gigantic forms in Asia raises some intriguing questions about Cretaceous Asia. Indeed, Deinocheirus, Therizinosaurus, and the recently-discovered Gigantoraptor were all huge theropod dinosaurs, dwarfing their closest known relatives. What led so many different groups of theropods to become so large? Are there more super-size theropod taxa out there awaiting discovery? The ones yet known that did increase dramatically in size all appear to have been at least partly herbivorous (or at least not hypercarnivorous), so could the deviation from the "standard" flesh-eating theropod diet have spurred these changes? Hopefully more fossil material of the existing giants and their ancestors will come to light, but even if we can start to answer some of the questions about these animals, the Cretaceous Asia must have been very strange place to be.


Clark, J.M., et al. "Therizinosauroidea" The Dinosauria, 2nd Ed. (2004) pp. 151-164

Li, et al. "A Large Therizinosauroid (Dinosauria: Theropoda) from the Early Cretaceous of Northwestern China." ACTA GEOLOGICA SINICA. Vol. 81 No. 4 pp. 539-549

Russel, D.A. "Therizinosauria". Encyclopedia of Dinosaurs. (1997) pp. 729-730

CMNH Press Release; "Giant Plant-Eating Dinosaur Waddled Across Ancient China"


More like this

A story of cheeks, beaks, feathers, bizarre theropod dinosaurs, and truly, truly amazing fossils.... Yesterday I made a special visit to the University of Portsmouth's School of Earth and Environmental Sciences in order to attend a talk by, and meet, Professor Altangerel Perle, the famous…
So, on to more ornithomimosaurs, aka ostrich dinosaurs (part I here). This time, the ornithomimids: this is the ornithomimosaur clade that includes only the edentulous arctometatarsalian taxa. Yes, I said arctometatarsalian*. However, note that some authors have incorrectly regarded Ornithomimidae…
First of all, many many thanks to everyone (both here and at SV-POW!) for the congrats regarding baby Emma. I am, shall we say, a little tired right now, but as you'll know if you've visited SV-POW! this morning, the good news as goes blogging is that I've been able to get lots of stuff prepared…
The idea that non-avian dinosaurs might have been able to climb trees is (I assume) not all that familiar to people outside the field of dinosaur research, but within the field of dinosaur research it has become an increasingly familiar idea within recent decades. Thanks to the discovery of such…

Good post, sir, on my favorite theropods. Okay, I'm giving you the spoiler. For years, I have wondered exactly how therizinos walked upright. Every picture I see these days features some massive-bellied monster with 3/4ths of its weight well in front of the hips, struggling to keep itself upright. Even worse, this stance would be nearly impossible given that derived therizinos have "winged" iliums in which the anterior portion of the ilium actually curves out to face the outside of the animal. Literally a 90 degree angle. This bony protrusion would is generally seen as a strange method of dealing with an expansive ribcage. Rather than depress the pelvis against the sacrum, like ornithischians did, therizinosaurs actually made their pelvis a lot bigger, but compacted it.

The "winged" ilium, though, would have made bipedal walking a hassle. Anyway, on to the spoiler. I was reading one of my Cenozoic mammal books one day (awhile back) and came across Chalicotherium, a strange knuckle-walking perissodactyl. It had a giant, seat-like pelvis, long arms and long claws, an expansive gut, and a toothless premaxilla. A lightbulb went off in my head. Therizinosaurs! Like those Cretaceous monsters, chalicotheres were once considered termite-eaters, but the wear patters on the teeth suggested an herbivorous diet. Also, chalicotheres, like therizinosaurs, got really big.

So I'm working on a new reconstruction of therizinosaurs right now which involves quadrupedal knuckle-walking. However, the trouble I'm having comes from frustration with known material. I don't think a single therizinosaur specimen has a full arm. I'm happy to say that, like chalicotheres, the individual phalanges of therizinos were short and broad, but I'd love to know if the arms were equal in length (or longer than) to the legs.

We need a complete skeleton!

Thanks Zach; I knew I had to get back to the dinos soon so you didn't explode. :) The convergence with the Chalicothere idea sounds pretty promising, but like you said we need more material. I had heard that the knuckle-walking in Chalicotheres has been backed up by the morphology of their knuckles, showing a pattern similar to primate knuckle-walkers like gorillas. I'd check for convergence in giant anteaters and giant sloths as well, but like you said we need more material from this group. Given the hypothesis, it would be interesting to see if knuckle-walking became a habit due to large size (i.e. if the smaller forms were bipeds but the larger ones were facultative quadrupeds) or when such a pattern became established. Obviously they arose from bipedal ancestors, so the transition would be an interesting one to try and pin down.

Oh yes, and one more thing. The front limb/hind limb proportions would have to be worked out as well. In the case of the Chalicothere and modern apes, the arms are longer than the legs (giving the animal a downward sloping posture). Would therizinosaur arms be long enough to reach the ground and knuckle walk? It looks like the arms would be a bit short from what we know so far. Either way, more skeletal material will definitely help resolve some of these questions.

tres kewl, Brian.

I have to wonder if there was a major bit of experimentation underway in Laurasia. If the rock hadn't fallen at the end of the Cretaceous, might we have seen some truly odd-ball and interesting ecologies develop with the dinosaurs as the continents separated in the alternate Cenozoic? After all, the isolation might have made for some very interesting evolutionary paths.

Would the Northern Hemisphere have been a therizinosaurian and ceratopsian enclave herbivorously speaking? Would Australia have been an oddball refugium? Even with dicynodonts, sauropods, and uber ampphibians? Or might it have become a home for the mammals to have taken over after some of the minor mass extinctions since the KT?

I'm always interested, Will, in why animals evolve the way they do. Take therizinosaurs, for example. They basically "re-evolved" the prosauropod lifestyle, but their maniraptoran ancestry forced their bodies to take on a different shape than those saurischian forebearers. But their teeth and mandible shapes are extremely similar. So similar, in fact, that Eshanosaurus, an Early Jurassic taxon based solely on a mandible, could be either a therizinosaur or a prosauropod.

(actually, now that Falcarius has been unearthed, the temporal difference would be pretty staggering, so paleontologists are leaning toward prosauropod these days)

Anyway, therizinosaurs seem to have evolved in North America (if Falcarius is any guide) before moving over to Mongolia. And I'll bet that they evolved their herbivory thanks to a lack of sauropod dinosaurs in North America. Therizinos didn't have grinding teeth, so they were probably eating what prosauropods ate (gymnosperms). "But what about ankylosaurs?" you say. "They were eating that sort of stuff." Ah, but their feeding level was very low. Therizinos were able to take advantage of taller feeding levels, thus eliminating competition with contemporary herbivores.

I think it's also telling that therizinosaurs didn't stick around in North America very long. I think they were overtaken by the ceratopsids, who became huge and successful very quickly after crossing over from Asia. Wouldn't it be funny if the first North American ceratopsids passed a group of therizinos on their way to Asia? There were no large ceratopsids in Asia, so the therizinos were able to thrive there. In North America, though, they were unable to deal with the competition from the large ceratopsids.

That's my take, anyway.

Thanks for the comments, guys. I honestly am not very well versed in the biogeography of the Cretaceous beyond recognizing the similarity between North American and Asian fuana (some taxa, like Saurolophus being found in both), but this discussion has inspired me to try and fix that. You can, of course, expect a post when I think I have some idea of what I'm doing.

Like I mentioned in the post, what I'm wondering is why so many Asian theropods grew to such large sizes and what kind of natural history they actually had. I grew up viewing the Cretaceous in the very "traditional" manner (Tyrannosaurus was the apex predator, Triceratops was its favorite prey, hadrosaurs abounded, and Deinonychus made life hell for smaller herbivores), but since China has been re-opened there has been a lot of bizarre fauna coming out. There's definitely related groups across the Pacific, but contingency is definitely in play as it seems that the N.A. animals developed a different kind of overall ecology than the ones in Asia.

Congratulations on the move to science blogs!

Therizinos... they are so strange, no wonder that things that have reached the Kardashev Type IV stage like to play with them ;) (see here:, page 13 and 14).

The video footage was especially awesome, a bit on the conservative side because of the lack of feathers, but secondary loss of integument is perfectly reasonably in a huge animal living under tropical or subtropical anditions.

As for the knuckle walking hypothesis, I actually own a model of a Beipiaosaurus that props itself up on the outer side of its huge claws. I don't know if the sculptor really wanted to suggest knuckle walking, or just wanted to stabilize his model to prevent it from falling over, like models of bipedal dinosaurs tend to do (remember Battat?), but by default or design, the pose looks pretty convincing.

I'm fascinated by the knuckle-walking idea, I guess because it's something I associate more with mammals than with dinosaurs. It sounds like an intriguing puzzle to resolve.

In the case that the link in my last comment doesn't work, just google PANDERAVIS