Move over Theropoda, Sebecosuchia rules

i-af14d3295ec68284a592369629f66742-Baurusuchus de Silva.jpg

Most of us have grown up with the idea that the Mesozoic Era was, excepting the Early Triassic, a time when dinosaurs dominated life on land. Or, put another way, a time when dinosaurs were the most ecologically significant and most obvious of all land animals. The familiar generalization, recounted in every book on Mesozoic life, is that dinosaurs were the only diverse big-bodied land vertebrates during the Jurassic and Cretaceous and, for as long as this was the case, other tetrapod groups were unable to achieve big body size. But in the same way that the modern world isn't really 'dominated' by any single taxonomic group, it turns out that this wasn't the case in the Mesozoic either. In fact dinosaurs weren't the only tetrapod group that evolved big, scary, terrestrial species during the Mesozoic. Furthermore, non-dinosaurian terrestrial reptiles may have been significant apex predators in varied Cretaceous ecosystems....

In characteristically childish fashion, I enjoy comparing this with the way the role of The Galactic Empire is portrayed in the Star Wars films and their spin-off stories. Star Wars novels and computer games sometimes state or imply that The Empire rules the entire galaxy with an iron fist, suppressing all resistance and imposing a consistent Imperial presence across all galactic systems. In fact, as evidenced by the role of the Galactic Senate, by Bespin, by what occurs on Tatooine and elsewhere in the Outer Rim, and by the fact that the galaxy is still crawling with such things as bounty hunters, mon calamari, dresselians, ugnaughts and even (shudder) ewoks, there is still plenty of room in the Star Wars universe for all manner of diversity to survive and thrive alongside the significant presence of The Empire. Substitute 'The Empire' for 'dinosaurs' and you might see what I'm getting at. Dinosaurs were big during the Mesozoic in every sense of the word, but it turns out that their presence did not prevent or suppress the evolution of everything else, even among terrestrial non-dinosaurian reptiles [excellent restoration of Baurusuchus above by Deverson da Silva, from Carvalho et al. (2005)].


Crocodilians (I use this word as a vernacular term for the clade Crocodyliformes) have an amazingly rich and diverse Mesozoic evolutionary history, involving small terrestrial generalist predators, long-skulled amphibious and fully aquatic forms, box-headed herbivores and omnivores, as well as the ancestors of modern alligators, crocodiles and gharials (Naish 2001: pdf available, please ask). Most Mesozoic crocodilians were 'small' - less than 3 m long - and giants were rare. The existence during the Cretaceous of gigantic amphibious crocodilians [namely the alligatoroid Deinosuchus from Campanian North America, the bizarre suspension-feeding giant Stomatosuchus from Egypt, and the immense pholidosaurid Sarcosuchus from the Saharan Cretaceous], is well known, but it's usually argued that, because these animals were amphibious, they were clearly occupying a quite different ecological niche from predatory dinosaurs, and hence weren't competing with them [adjacent Sarcosuchus image from wikipedia].

But what if there were big terrestrial predatory crocodilians during the Mesozoic? Surely such animals would have been interacting with predatory dinosaurs: living in the same places, preying on the same prey, and scavenging from the same carcasses. Big, deep-skulled terrestrial crocodilians of this sort are well known from the Cenozoic, mostly thanks to Sebecus icaeorhinus, a South American Eocene crocodilian first described by the famous evolutionary scientist George Simpson (1902-1984) in 1937 [more on Sebecus and other Cenozoic sebecosuchians in a later post].

Sebecus had a laterally compressed, tall snout, and a concave palate: in cross-section its snout was virtually parallel-sided, this forming the skull shape termed either oreinisrostral or altirostral. Skulls of this sort - present widely among predatory terrestrial archosaurs - are best for dealing with compressive vertical loading of the sort encountered during the slashing and slicing of prey (Busbey 1986, 1995), and are quite different from the dorsoventrally compressed platyrostral skulls typical of modern crocodilians (though - note - not universally present even in these animals).

i-f945f87b0acfcaf0877d7a69055ac412-allosaur tooth.jpg

While there is no doubt today that Sebecus and its relatives are crocodilians, little known is that Simpson (1937) initially regarded Sebecus as an unusual crocodilian-like archosaur of uncertain and 'doubtfully crocodilian' affinities. His resultant naming of the group Sebecosuchia* was not, therefore, the recognition of a new and distinctive group of crocodilians, but instead the result of a frustrated effort to provide this Eocene reptile with its own group. The discovery of good Sebecus material allowed Simpson to realize that the post-Cretaceous 'theropod' teeth regarded by Florentino Ameghino as representing late-surviving dinosaurs actually belonged to sebecosuchians: Simpson had discussed the teeth earlier (Simpson 1932) and, though unable to show that they were non-dinosaurian, had concluded that they belonged to some other taxon that possessed theropod-like teeth. These teeth are of the sort known as ziphodont: laterally compressed, gently recurved, and with serrated keels along their edges [adjacent image shows a typical ziphodont tooth, though from a theropod and not a crocodilian]. Like the oreinirostral skull shape, they occur throughout predatory archosaurs and aren't unique to either theropods or sebecosuchians.

* It was not coined by Colbert (1946) as stated in one recent paper (Turner & Calvo 2005).

Though mostly associated with the Cenozoic, we've known since the 1890s that sebecosuchians were also present in the Cretaceous as 1896 saw the description of the Argentinean taxon Cynodontosuchus rothi Woodward, 1896. It's from the Coniacian-Santonian Pichi Picun Leufu Formation. Known only from an incomplete snout and articulated lower jaw, Cynodontosuchus exhibited a tooth pattern common to a number of sebecosuchian taxa: the premaxillary teeth were small and subcircular in cross-section, the maxillary teeth were few in number (there are just five) and the first maxillary tooth was small while the second was particularly large, and a diastema between the premaxilla and maxilla housed an enlarged mandibular tooth when the jaws were closed. The big second maxillary tooth could truly be described as a 'saber tooth' as its tip extended to the ventral edge of the lower jaw. This strongly differentiated dentition is rather different from that of Sebecus and its relatives, and consequently Cynodontosuchus and its relatives are grouped together in the sebecosuchian group Baurusuchidae.

i-489da037c3cb4873166160a0b18d49c7-Baurusuchus Rio model.jpeg

Highly similar to Cynodontosuchus was another Cretaceous form, Baurusuchus. Baurusuchus was notably robust-skulled and scary-looking. Two species are known: B. pachecoi Price, 1955 and B. salgadoensis Carvalho et al., 2005 [adjacent image, showing bones and model of B. salgadoensis, from here]. Incidentally, the specific name of the latter is nothing to do with Leonardo Salgado*, but is a reference to General Salgado County in Brazil's Sao Paulo State. Authors have disagreed on how Baurusuchus compares with the earlier-named Cynodontosuchus. Gasparini (1972), Gasprini et al. (1993) and Carvalho et al. (2005) argued that the two are so similar that they might turn out to be synonymous while Bonaparte (1996) argued that Cynodontosuchus had a rather 'weaker' dentition than Baurusuchus, and thereby implied that it was quite different. Regardless, Cynodontosuchus is much smaller than either of the Baurusuchus species, and hence probably quite different in ecology (Bonaparte 1996).

* Just as well given that it means 'from Salgado'.

Similar in some details to Baurusuchus and Cynodontosuchus was Stratiotosuchus maxhechti Campos et al., 2001 from the Turonian-Santonian Adamantina Formation of Brazil, another baurusuchid. Its skull bones are reported to be particularly thick and its snout was shallower in side view than that of the other baurusuchids. It was also alarmingly big, with a skull over 50 cm long. The two Baurusuchus species were similar in size, with skulls exceeding 40 cm in length.

All the sebecosuchians we've seen so far have been South American. But we now know that these animals weren't restricted to this continent, as once thought. Pabwehshi pakistanensis Wilson et al., 2001 is from the Maastrichtian Pab Formation of Pakistan. Though known only from an incomplete snout and lower jaw tip, enough is preserved to show us that Pabwehshi shares with Baurusuchus a reduced premaxillary dentition, a third premaxillary tooth that overhangs the dentary, a reduced first maxillary tooth, and an enlarged caniniform second maxillary tooth (Wilson et al. 2001). Accordingly, it again seems to be a member of Baurusuchidae, showing us that sebecosuchians of this group were present in Upper Cretaceous India as well.

i-65b6652d05c1701884404cdfcba120ed-Baurusuchus life.jpg

How did these animals live? We can't be sure of course, but their anatomy indicates that they were predominantly, if not entirely, terrestrial. Several palaeoherpetologists have argued at times that sebecosuchians and similar crocodilians were amphibious and in the habit of ambushing prey from the water's edge, but there doesn't seem to be any good reason to assume this. Their skull and tooth morphology suggests that they behaved in a similar manner to theropods and big monitor lizards, though the recent discovery of the ziphodont thalattosuchian Dakosaurus andiniensis shows that such features weren't restricted to terrestrial archosaurs (go here for more on metriorhynchids). Unfortunately the post-cranial anatomy of these animals remains poorly known. Near-complete, articulated baurusuchid skeletons were reported in 2004 [free pdf here] but as yet very little information has appeared on them. In the absence of further information, what we know of related forms (including the notosuchians and peirosaurids [image below shows skull of peirosaurid Uberabasuchus, from here) is consistent with a mostly terrestrial lifestyle [life restoration of two baurusuchids from here].

The remains we have of such sebecosuchians as Baurusuchus shows that they were large animals, with body lengths exceeding 3.5 m and approaching 4 m in Stratiotosuchus. If the idea of a 4-m-long terrestrial predatory Cretaceous non-dinosaurian reptile doesn't grab your imagination, shame on you. It is veeery tempting to wonder whether, and how, sebecosuchians and theropods interacted: surely they must have. I wanted to mention this idea in my 2001 review article on crocodilians, but my reviewer urged me to remove it. Candeiro et al. (2006) however, have hinted at this idea, noting that, in some Late Cretaceous South American faunas, terrestrial crocodilians 'had a more important ecological role to play as the main carnivorous group ... than did theropods' (p. 937).


One last thing. Baurusuchids weren't unique to the Southern Hemisphere, as Iberosuchus macrodon Antunes, 1975 from Spain and Bergisuchus dietrichbergi Kuhn, 1968 from Messel in Germany are also thought to be members of the group. Atacisaurus crassiproratus Astre, 1931 from France has been reidentified as cf. Iberosuchus and Eremosuchus elkoholicus* Buffetaut, 1989 from Algeria has also recently been identified as a baurusuchid. But here we come to something else really interesting: Iberosuchus, Bergisuchus and Eremosuchus are all Eocene in age, which means that baurusuchids survived the end-Cretaceous event and were apparently doing fine many millions of years after the Cretaceous.

* Its specific name refers to the locality of El Kohol and is nothing to do with beers, wines or spirits.

What was it about these terrestrial predators that made them different from the Cretaceous dinosaurs that they were formerly contemporaneous with? Was it something to do with their metabolism, or maybe their habits? We don't know, but their survival suggests that they were significantly different from dinosaurs in some way.

The Triassic might be regarded as the heyday of crurotarsan archosaurs, but it is simply unfair and incorrect to imply that these non-dinosaurian ruling reptiles failed to continue ruling throughout the Age of Dinosaurs and beyond. And you don't need me to remind you that crocodilians remain big, ecologically important predators today.

More on sebecosuchians in future. Still to come soon (honest): vampire pterosaurs (or not), beluwhals and proto-narwhals, amazing social life of the Green iguana, and everything else I've been promising for the past year and two months. And I just saw a movie trailer for a certain film, which reminds me: war rhinos. And rhinogradentians.

Refs - -

Bonaparte, J. F. 1996. Cretaceous tetrapods of Argentina. Muncher Geowissenschaften, Abhandlungen 30, 73-130.

Busbey, A. B. 1986. New material of Sebecus cf. huilensis (Crocodilia: Sebecosuchidae) from the Miocene La Venta Formation of Colombia. Journal of Vertebrate Paleontology 6, 20-27.

- . 1995. The structural consequences of skull flattening in crocodilians. In Thomason, J. J. (ed) Functional Morphology in Vertebrate Paleontology. Cambridge University Press (Cambridge), pp. 173-192.

Candeiro, C. R. A., Martinelli, A. G., Avilla, L. S. & Rich, T. H. 2006. Tetrapods from the Upper Cretaceous (Turonian-Maastrichtian) Bauru Group of Brazil: a reappraisal. Cretaceous Research 27, 923-946.

Carvalho, I. de S., Ribeiro, L. C. B. & Avilla, L. dos S. 2004. Uberabasuchus terrificus sp. nov., a new Crocodylomorpha from the Bauru Basin (Upper Cretaceous), Brazil. Gondwana Research 7, 975-1002.

Colbert, E. H. 1946. Sebecus, representative of a peculiar suborder of fossil Crocodilia from Patagonia. Bulletin of the American Museum of Natural History 87, 217-270.

Gasparini, Z. 1972. Los Sebecosuchia (Crocodilia) del territorio Argentino. Consideracones sobre su "status" taxonomico. Ameghiniana 9, 23-34.

- ., Fernandez, M. & Powell, J. 1993. New Tertiary sebecosuchians (Crocodylomorpha) from South America: phylogenetic implications. Historical Biology 7, 1-19.

Naish, D. 2001. Fossils explained 34: Crocodilians. Geology Today 17 (2), 71-77.

Simpson, G. G. 1932. The supposed association of dinosaurs with mammals of Tertiary type in Patagonia. American Museum Novitates 566, 1-21.

- . 1937. New reptiles from the Eocene of South America. American Museum Novitates 927, 1-3.

Turner, A. H. & Calvo, J. O. 2005. A new sebecosuchian crocodyliform from the Late Cretaceous of Patagonia. Journal of Vertebrate Paleontology 25, 87-98.

Wilson, J. A., Malkani, M. S. & Gingerich, P. D. 2001. New crocodyliform (Reptilia, Mesoeucrocodylia) from the Upper Cretaceous Pab Formation of Vitakri, Balochistan (Pakistan). Contributions from the Museum of Paleontology, The University of Michigan 30, 321-336.


More like this

"bizarre suspension-feeding giant Stromatosuchus" ?!?!?!
That's a REALLY fascinating idea! As far as I know, none of the Mesozoic marine, umm, suropsids (Ichthyosaurs, Pesiosaurs, Mosasaurs, Metriorhynchids...) ever evolved a suspension feeder analogous to baleen whales. Somewhere there is an article by Christine Janis (possibly with a co-author) developing a speculation that there is something fundamentally differentt about mammalian and "reptilian" bauplans that made suspension feeding easier for mammals to evolve. (Sorry, can't remember the details.)
Oh, by the way: another VERY, very nice post!

By Allen Hazen (not verified) on 11 Mar 2007 #permalink

Rhinogradentians! Yes! With many technical illustrations! And a cladistic counteranalysis!

[As always, a lovely and informative post Darren.]

Great post, as usual. I find it particularly interesting that dinosaurs kept mammals pretty small* throughout the Mesozoic, but terrestrial crocs were able to evolve to much larger sizes. There has got to be an interesting and as-yet-untold ecological story in there somewhere. It is tempting to speculate that endothermic dinos were harder on endothermic mammals than on ectothermic crocs (in terms of ecological interference, not necessarily predation or other more direct interactions). Also, do we know how large the K/T-crossing sebecosuchians were? They might be an exception to the widely-circulated "fact" that nothing on land larger than 1m survived the extinction.

* Even Repenomamus was no longer or heavier than a large badger, and it appears to have been an outlier in terms of size.

Incidentally, TZers, many of the references listed above are available for free online, notably the AMNH and UMich publications. Links to these and other sources of free literature are available on the sidebar here.

I've only discovered you since you've come to sciblogs, but I'm very impressed with your posts! I love them! Keep up the good work!

For the article on filter-feeding in reptiles:
Collin, R., and C.M. Janis, 1997: "Morphological constraints on tetrapod feeding mechanisms: why were there no suspension feeding marine reptiles?" in Callaway and Nicholls, eds., "Ancient Marine Reptiles," pp. 451-466.
And the library at my university doesn't have the book. Sniff.

By Allen Hazen (not verified) on 12 Mar 2007 #permalink

Full details another time, but the thing about reptiles being unable to evolve suspension-feeding (note: different from filter-feeding) is not true, as evidenced by placodonts and the morphology of turtles. In other words, Collin & Janis's (1997) primary assertion was incorrect.

I find it particularly interesting that dinosaurs kept mammals pretty small* throughout the Mesozoic, but terrestrial crocs were able to evolve to much larger sizes. There has got to be an interesting and as-yet-untold ecological story in there somewhere.

Isn't this just because mammals are crap?

There was also another giant filter-feeding crocodilian, Mourasuchus from South America, which lived only some million years ago. It hat a huge flat and wide skull, with very fragile jaws which were undoubtly not suited to catch big prey.

No Terry Pratchett fans among palaeolontologists? I am still waiting for a sebecosuchian (or any other ziphodont croc) named Offler ...

Great post! I've tried to find information on Stromatosuchus for years without much success, but suspension feeding sounds awesome. I liked reading about my favorite crocodilians, though. As soon as I saw Baurusuchus' description I thought "dinosaur competition!"
You know, another big crocodilian that likely competed with theropods may have been Sarcosuchus, what with Suchomimus around...

As for crocodiles surviving the K/T crisis I think their ability to do without food for long periods and their probably much lower energy requirements compared to theropods of similar size may have been important.
Crocodiles can occur at absolutely incredible densities in undisturbed habitats (e g Llanos or Pantanal) so it seems reasonable that a few might scrape through even in a devastated post-Chicxulub landscape.

By Tommy Tyrberg (not verified) on 12 Mar 2007 #permalink

Isn't this just because mammals are crap?

Well, not entirely. Mammals served many important ecological roles during the Mesozoic, like providing food for extremely small dinosaurs, and greasing the foot pads of sauropods.

Oh, and eating all the dinosaur eggs. (Chicxulub is so 1994.)

Only 5 maxillary teeth? How gorgonopsian of them.

The photo of Sarcosuchus was taken in the Muséum National d'Histoire Naturelle in Paris. I've been there. Yes, the beast is big...

(Naish 2001: pdf available, please ask)

If you ask me to ask, I will ask :o)

developing a speculation that there is something fundamentally differentt about mammalian and "reptilian" bauplans that made suspension feeding easier for mammals to evolve.


And the tradition of multicusped teeth.

Incidentally, Henodus has been reinterpreted as a herbivore. The paper was in N. Jb. Geol. Paläont. in 2004, I'll post the ref tomorrow if I don't forget.

I've tried to find information on Stromatosuchus for years without much success

Try googling for Stomatosuchus, without the r, "mouth crocodile". Though that won't help much either.

By David Marjanović (not verified) on 12 Mar 2007 #permalink

David: Reif & Stein interpreted Henodus as a herbivore in 1999 (N. Jb. Monat. 1999, 65-80), but the evidence for filter-feeding seems better: it was most recently stated by Rieppel (2002) in Zool. J. Linn. Soc. 135, 33-63. I must have missed the 2004 paper if there was one: I published a review article on placodonts in that year and wasn't aware of it then. I'm emailing the crocodilian pdf to you.

One interesting element I noticed is that the legs of all these Crocodilians seem to be much more vertical, not splayed out beside the torso as with modern Crocodilians and older Therapsids.

I'm sure there are names for these differing gaits, and I'm sure work has been done into the topic of how Baurusuchus walked compared to modern crocodiles. Do you have any further information on this?

So, it looks like its a Maastrichtian phenomenon? I've read in a few places that the dinosaurs were in decline in the Late Cretaceous. Would this be an example of something trying to fill vacant niches? I've also seen contrary writings to that last too, btw.

Oh, and eating all the dinosaur eggs.

Isn't Walter in your dept? Isn't that dangerous to say then, DV? ;)

Well, looks like Spec needs to revise and expand its Ziphosuchian croc fauna. Thank you. how does Megasuchus naishii sound?

and thank you again for writing about this. I intend to show this to the 'Star Wars' fans in my family.

have nice days.

By Anthony Docimo (not verified) on 12 Mar 2007 #permalink

Was Henodus definite in having a beak? Or could it have had a fleshy platypus-mouth like Jaime Headden once inferred?

By Tim Morris (not verified) on 13 Mar 2007 #permalink

Poseidon: the limbs of sebecosuchians and other terrestrial crocodilians (notosuchians, peirosaurids etc) do seem to have been more erect than those of their amphibious relatives. The presence of an enlarged fourth trochanter and other features indicates that the limbs of these animals well were suited for active terrestrial locomotion.

Will: no, Cretaceous sebecosuchians were not a Maastrichtian phenomenon given that some of the forms discussed here are as old as Turonian. What I didn't mention is that there must have been even more Upper Cretaceous sebecosuchians than discussed here: I only really wrote about baurusuchids, but phylogenetic studies show that the lineages that include Sebecus and other Cenozoic sebecosuchian taxa diverged prior to the evolution of baurusuchids, so Sebecus-like forms must also have been around in the Upper Cretaceous too. I'm going to cover Sebecus and other Cenozoic taxa in a later post.

Anthony: yes please, I always wanted a patronym, even if it is only a hypothetical taxon :)

Tim: nope, no beak in Henodus. Its mouth anatomy is remarkably odd: the premaxillary and dentary margins are broader than the rest of the skull, making the head shaped like a very squat 'T'. The premaxillary margins are lined with small subrectangular denticles while the edges of the maxillae and dentaries have gutter-like grooves occupied by a vertical baleen-like material. I plan to blog on placodonts soon and will cover all of this stuff then (with pictures).

One more thing on Cretaceous sebecosuchians: the existence of all of these moderately big terrestrial reptiles means we have to revise that old sound-bite about 'no land animals above [insert size] survived the end-Cretaceous event'. We don't have a complete record of sebecosuchians across the K-Pg boudary of course, but the Cretaceous, Palaeocene and Eocene forms that we have indicate that the forms that survived across the boundary were of the same size: viz, pretty big for a terrestrial non-dinosaur, 2-3 m long. Mostly ignored to my knowledge, this fact might mean something important for the metabolic status of dinosaurs, as some commenters here have already noted.

Reif & Stein interpreted Henodus as a herbivore in 1999 (N. Jb. Monat. 1999, 65-80)

Oh! That's probably the paper I was thinking of. I'll look, but probably there is no 2004 paper. Thanks for the other ref!

On the subject of Spec, keep in mind we haven't yet uploaded the entire crocofauna we've created so far.

By David Marjanović (not verified) on 13 Mar 2007 #permalink

Oh, and eating all the dinosaur eggs. (Chicxulub is so 1994.)

Isn't Walter in your dept? Isn't that dangerous to say then, DV? ;)

Good question. Actually, I'm in Integrative Biology and Walter is over in Geology. And there is a long history of impact dissent at Berkeley. My co-advisor, Bill Clemens, has never been a believer. But he and Walter get along just fine.

But also, I was making a bit of a meta-joke about the long-running comment argument on mammals eating dino eggs from the 2nd feathers and filaments post. And ironically implying that just because the impact hypothesis has a long history it is no longer to be taken seriously (like many pop gen people seem to think about punctuated equilibrium). Perhaps I was too clever for my own good.

FWIW, I'm a confirmed Chicxulubist, and I will be until someone comes up with a better mechanism for taking out big terrestrial critters, big and little marine critters, plankton, trees, and so on all at once.

Yes, it was 1999. 2004 was a false memory.

I've found the 2002 paper, thanks! It's in the last issue of Zool. J. Linn. Soc. that is here in the lab :-)

By David Marjanovi? (not verified) on 13 Mar 2007 #permalink

One thing that comes to mind: is this exclusively a croc trend, or was the terrestrial-erect / amphibious-sprawl a feature which could happen to *any* archosaur?

Dr Vector - they were on my diet.

By Anthony Docimo (not verified) on 13 Mar 2007 #permalink

Isn't this just because mammals are crap?

If synapsids are crap, why do ziphodont crocs try so hard to become ersatz gorgonopsians?

If synapsids are crap, why do ziphodont crocs try so hard to become ersatz gorgonopsians?

Gorgonopsians are not crap -- but then they aren't mammals!*

* Well... under... most definitions of that term.

By David Marjanović (not verified) on 13 Mar 2007 #permalink

Gorgonopsians are not crap -- but then they aren't mammals!*

Not even cynodonts - I just thought proper synapsid self-hatred would include non-mammalian synapsids as well

I like to know what kind of ecosystem the two titanosaur genera Mendozasaurus and Muyelensaurus lived in, and their lengths and growth rates (if possible). Also, how large was a Cynodontosuchus? Could you specify when Stratiosuchus actually lived, instead of just putting in the Turonian-Santonian part?

Can I translate this article from English into Russian?