The aquatic Majungasaurus, or not

WOW! Thanks for this, Darren! Seeing it from the side, i was under the ipression that the tail was more heavily muscled than it really was. You know, my suggesting that it was amphibious was mostly an attempted explaination at why the beast is so freakin' weird in its proportions. Other than the amphibious thing, which you did a pretty good job at knocking down, what else could explain it's stupidly bizarre proportions, wonder?

Sorry for all the typos in the last comment. Zach Miler: I'll get started on the drawing. When it's done I'll send it to ya. Might take a couple days, but it'll get there. How do you want to see it? Swimming, like I had said? (Like all theropods, Majungee* could probably swim pretty well, even if wasn't really amphibious.) Just standing around?

*I wonder if my nickname for the reptile will catch on...

One other big, big problem with this is that northwestern Madagascar during the latest Cretaceous was dry as a bone. The best evidence indicates a highly seasonal environment, with pretty arid conditions most of the time. Not the best place for a large aquatic animal.

I'm still not sure how a long thread on identifying a rodent became subverted by a discussion on Cretaceous theropods

Oh, that's normal. It's called "thread drift".

Nice comparison to Mahajangasuchus in Andrea's picture. That's another animal that deserves being blogged about... :-)

what else could explain it's stupidly bizarre proportions, wonder?

Perhaps you should compare it to Xenosmilus, another animal built for biting at big prey that doesn't run.

The tail, BTW, is not tall enough for swimming, and tapers too soon; it looks much different in phytosaurs.

I think that the highly pneumatic bones is the best argument against a semi-aquatic Majungasaurus. Did Ceratosaurus have highly pneumatic bones? It's toes are, indeed, a little longer than normal, and that tail is crocodile- or phytosaur-like, much more so than that of Majungasaurus. It's also very flexible (from side to side), even more so than is normal for a non-tetanuran theropod. It's shed teeth are found more often among the remains of aquatic animals (turtles, crocs, and HUGE lungfish) than they are among terrestrial dinosaur remains. So, even if we're not going to fool around with an amphibious Ceratosaurus, we at least have to admit that it was very, very fond of aquatic prey. In short, if we are going to have a semi-aquatic theropod (and I can't but see *some* evidence for that notion), Ceratosaurus now appears much better a pick than Majungasaurus. And no, I'm not just agreeing with Bakker for the heck of it, or to stick out from the crowd, it's because I honestly can't see where he's wrong about the Ceratosaurus thing. Unless, of course, I find out that the beast had highly pneumatic bones, or Bakker starts saying that it's anologous to seals and basilosaurids. :-(

Dang, I didn't think about Xenosmilus. Good anology.

"It (the tail) looks much different in phytosaurs."

What about Parasuchus?

Will check Parasuchus and Ceratosaurus.

Pneumaticity is not an argument against aquatic locomotion, it's only an argument against deep sub-surface feeding. This parallels other remarks noted by Andrea about dentition.

Hey, Nathan Myers. You're right! Didn't think about that. That's why ducks and such have reduced pneumaticity. Amphibious Majungasaurus lives??

Okay, I'm not quite ready to go back and start arguing for a semi-aquatic Majungee again. But Nathan is right about pneumaticity. And I still don't know about any fully terrestrial reptiles with vertical caudal neural spines.

Other than people apparently not liking the idea of semi-aquatic theropods, why is it that Bakker's Ceratosaurus paper has not been accepted at all?

Zach Miller: You there, buddy? I can't start the Mujungee drawing 'till you tell me what you want it to look like. What's a pose or activity that you really, really want to see a Majungasaurus in? Please let me know, and rest assured that it'll be finished and sent your way. Eventually...

Is it just the photo, or is that thing extremely front heavy?

The perspective in the picture is weird. But the skeleton is, indeed, very front heavy, or at least it looks that way to us humans. Weather or not the live animal really was is a different matter.

I'm really not sure what Majungasaurus was doing. Perhaps, as David Marjanovic said, it was anologous to Xenosmilus. It was a slow, clumsy land animal, that snuck up on sauropods and gave huge, crippling wounds to their sides. But then again, the reptile stands less than 2 meters above the ground. A sauropod could probably kick it's face in.
I don't know if the amphibious hypothesis has died or not. Aquatic birds always have reduced pneumaticity, but they have to dive deep for food. What if Majungasaurus actually increased its level of pneumaticity in order to be a better floater? Maybe it waited at the surface, dead still, for animals to come down to water to drink, then attacked them croc-style? The toes not being particularly long is a problem for the semi-aquatic argument. On to the tail. Sorry, but it just looks deep, well-muscled, and crocodile-like to me. It's probably just huge and heavy to support the front end of the animal, but why are the caudal neural spines vertical? I know of no fully terrestrial reptile that has this feature.
In short, I think that the semi-aquatic hypothesis for Majungee is injured, but not quite dead yet. Now I'll shut up. :-)

Polar Bears are terrestrial Carnivores that are nonetheless highly aquatic.

Who is to say that this Therapod didn't live in the same way?

Hi everyone!
Thank you Darren for citing my posts!
I wrote in details on this topic also here:
http://theropoda.blogspot.com/2008/11/carnotauri-anfibi-versione-30.html. (A tranlator in the side bar).
I never intended Majungasaurus as a deep diver! In my opinion, it was relatively more adapted for semi-aquatic lifestyle than other ceratosaurs, but not fully aquatic. Most of its features are ceratosaurian apomorphies exapted for a (sligtly?) more aquatic lifestyle:
Tail differing from other abelisaurids in:
1-More mobile, lacking accessory hyposfene-hypantrum articulations. Aquatic adaptation?
2-More large, having sub-horizontal transverse processes instead of dorsolaterally directed. Aquatic adaptation?
Hindlimb:
1-The shape of the cnemial crest, hatched-shaped, is similiar to those of some foot propelled birds (For example, Gansus). Ceratosaur exaptation.
2-Its IV metatarsal diverges laterally more than in other ceratosaurs (it happens also in the sympatric Masiakasaurus too: convergence?). Incipient aquatic adaptation?
3-Its III metatarsal sligthly diverge laterally (in most theropods its straigth). Incipient aquatic adaptation?
Body shape:
Shorter and stouter limbs, recalling semi-aquatic mammals. Aquatic adaptation?
Palaeoecology:
In Majungasaurus's ecosystem fully terrestrial vertebrates are few (two sauropods and ?two abelisauroids), contrasting with the number of semi-aquatic forms (at least five crocodyles!).
Paleogeograpy: At the end of the Cretaceous, Madagascar was probably isolated from Gondwana: a semi-aquatic habit for some abelisaurs could explain better their Gondwanan distribution.

I KNOW THAT IT IS ONLY AN HYPOTHESIS, I'm not saying I'm strongly supporting it.

If I'm wrong in some speculations, please, correct me!
Thank you.

Andrea Cau

Couldn't it specialize in say, young sauropods?

I forgot mentioning other possible semi-aquatic exaptations present in Majungasaurus (indicative of an incipient aptation toward a more amphibious lifestyle than in other theropods*):

1- Pedal unguals ventrally flat, lacking flexor tubercle.
2- Pedal digit IV dorsoventrally compressed.
3- Skull short and telescopied (relatively tall orbits and naris).
4- Broader mouth (mandibular symphysis more "U-shaped") than in other ceratosaurs.
5- Espanded "hypotarsus-like" caudal surface of the proximal end of metatarsal III (antarctometatarsus) for strong pedal phalanges flexor muscles.

Am I "narrow-minded" in this hypothesis? Are there other good explanations of this combination of characters (added to those I listed in the comment 17)?
It is probable that these features may be explained in other equally plausible ways: I hope someone could give me a different interpretation.
I'm not dogmatic in my hypotheses! Thank you.

*I mean "amphibious" (equally adapted for both lifestyles), not "more aquatic and less terrestrial"!

Michael Erickson on Ceratosaurus said:

It's shed teeth are found more often among the remains of aquatic animals (turtles, crocs, and HUGE lungfish) than they are among terrestrial dinosaur remains.

Reference?

That doesn't necessarily means it was aquatic, it might just be the result of collecting bias or, keep reading:

The paleoenvironment of the Morrison Formation, where Ceratosaurus nasicornis is found, seems to have been that of fluctuating wet/dry seasons, with taphonomic evidence pointing towards very long dry periods (Dodson et al., 1980). With scarce aquatic environments, why would Ceratosaurus compete with Goniopholis sp. and Macelognathus vagans, who, as crocodylians were probably better adapted for aquatic environments? If remains of Ceratosaurus are found associated with remains of aquatic animals, it might just have been exploiting the low water levels during the dry season, thus finding an easy meal in the remaining shallow ponds.

Dodson, P., A. K. Behrensmeyer, R. T. Bakker & J. S. McIntosh. 1980. Taphonomy and paleoecology of the dinosaur beds of the Jurassic Morrison Formation. Paleobiology 6(2):208-232.

I'm still not sure how a long thread on identifying a rodent became subverted by a discussion on Cretaceous theropods ...

Many rodents are known to go into hiding when they spot certain modern predaceous theropods. You can only imagine how they might react to a cretaceous theropod.

Ops, I forgot the refs for my comments:
-Carrano M. T. & Sampson S. C., 2008 - The phylogeny of Ceratosauria (Dinosauria: Theropoda). Journal of Systematic Palaeontology, 6(2): 183 - 236.
-Carrano M. T., Sampson S. D. & Forster C.A., 2002 - The osteology of Masiakasaurus knopfleri, a small abelisauroid (Dinosauria: Theropoda) from the Late Cretaceous of Madagascar. Journal of Vertebrate Paleontology, 22: 510 - 534.
-Gilmore, C. W. 1920 - Osteology of the carnivorous Dinosauria in the United States National Museum, with special reference to the genera Antrodemus (Allosaurus) and Ceratosaurus. Bulletin of the United States National Museum, 110: 1 - 154.
-Madsen J. R. Jr. & Welles S. P., 2000 - Ceratosaurus (Dinosauria, Theropoda) a revised osteology. Miscellaneous Publication Utah Geological Survey, 00-2: 1-80.
-Novas F. E., Agnolin F. L., & Bandyopadhyay S., 2004 - Cretaceous theropods from India: A review of specimens described by Huene and Matley (1933). Revista Museo Argentino de Ciencias Naturales, n.s., 6 (1): 67 - 103.
-Sampson S. D. & Krause D. W., 2007 - Majungasaurus crenatissimus (Theropoda: Abelisauridae) from the Late Cretaceous of Madagascar. Society of Vertebrate Paleontology, Memoir 8: 1 -184.
-Sereno P. C., Wilson J. A., & Conrad J. L., 2004 - New dinosaurs link southern landmasses in the Mid-Cretaceous. Proceedings of the Royal Society: Biological Sciences, 271: 1325 - 1330.
-Tykoski R. S., 2005 - Anatomy, Ontogeny, and Phylogeny of Coelophysoid Theropods. Unpublished PhD Thesis, The University of Texas at Austin: 1 - 553.
-Wilson J. A., Sereno P. C., Srivastava S., Bhatt D. K., Kholsa A. & Sahni A., 2003 - A new abelisaurid (Dinosauria, Theropoda) from the Lameta Formation (Cretaceous, MAstrichtian) of India. Contributions from the Museum of Paleontology, University of Michigan, 31: 1 - 42.

I don't see any hallmarks for an aquatic or amphibious lifestyle.

a) Tail isn't deep, transverse processes aren't well developed. So not a strong caudal swimmer.
b) Toes aren't longer proportionally than those of modern terrestrial cursorial fowl. So not a wader.
c) Limb proportions are consistent with those of modern cursorial fowl. Amphibious or aquatic species have reduced femur/tibia/fibula and increased length in the metatarsals and toes.
d) I can't think of any amphibious or aquatic species that have reduced their forelimbs to such a ridiculous degree, but maintained the length of their hindlimbs.

Parasuchus⦠I doubt that it was a tail-propelled swimmer in the first place. Its tail is pathetic -- shaped like in Majungasaurus, only shorter (!) --, and its hindlegs are gigantic. Frankly, I think it was a foot-propelled swimmer. Unless maybe if it had huge dermal scutes, like those of crocodiles, only much bigger.

â¦Actually, that could be it. The thalattosuchians have very short neural spines and not much longer haemal spines, though they do have long tails.

(That it was a swimmer is made obvious by its skull.)

Ceratosaurus: Same as Majungasaurus -- long, rapidly tapering tail. The neural spines at the base are tall, but that's it.

And I still don't know about any fully terrestrial reptiles with vertical caudal neural spines.

If both the neural and the haemal spines were vertical, like in a "nectridean", you'd have an argument.

And what does it matter whether a vertebrate is a "reptile" or not?

A sauropod could probably kick it's face in.

How so, if it comes from the side?

What if Majungasaurus actually increased its level of pneumaticity in order to be a better floater? Maybe it waited at the surface, dead still, for animals to come down to water to drink, then attacked them croc-style?

How so, when it floats on top of the water like a duck? It does not float just below the surface like a crocodile. It's way too pneumatic for that.

contrasting with the number of semi-aquatic forms (at least five crocodyles!).

What makes you so sure they weren't terrestrial? All ziphosuchians are supposed to have been.

Paleogeograpy: At the end of the Cretaceous, Madagascar was probably isolated from Gondwana: a semi-aquatic habit for some abelisaurs could explain better their Gondwanan distribution.

Then what about the sauropods and the ziphosuchians? Isn't it much simpler to assume that carnotaurines were in Madagascar before it broke off the Seychelles and their associated basaltic Iceland-like platforms?

(Unfair advantage: I have the Geological Map of the World just outside this room. Sometimes it's good to be in the geology building!)

1- Pedal unguals ventrally flat, lacking flexor tubercle.
2- Pedal digit IV dorsoventrally compressed.

Don't these simply mean that it didn't use its feet as weapons?

3- Skull short and telescopied (relatively tall orbits and naris).

In an amphibious or aquatic limbed vertebrate I'd expect either a flat head (crocodiles, most temnospondyls, most salamanders and frogs, diplocaulids) or, surprisingly, a carcharodontosaurid-like long and tall one (anthracosaurs, Crassigyrinus, Whatcheeria and the like). Carnotaurines have neither.

4- Broader mouth (mandibular symphysis more "U-shaped") than in other ceratosaurs.

Direct consequence of the shorter skull: if the skull gets shorter while maintaining its width, the symphysis has to become rounder.

5- E[x]panded "hypotarsus-like" caudal surface of the proximal end of metatarsal III (antarctometatarsus) for strong pedal phalanges flexor muscles.

Could be connected to the shorter legs needing longer levers for the same effect.

> It's shed teeth are found more often among the remains of
> aquatic animals (turtles, crocs, and HUGE lungfish) than
> they are among terrestrial dinosaur remains.

Many terrestrial predators - bears, spinosaurs, cats, to name but a few - eat fish when they can get it, this doesn't make them aquatic in the strict sense.

> Goniopholis sp. and Macelognathus vagans, who, as
> crocodylians were probably
> better adapted for aquatic environments?

*Goniopholis*, yes, but but *Macelognathus* was a sphenosuchian, a clade that tends to be terrestrial. AFAIK, *Marcelognathus* was also strongly convergent on dinosaurs.

why would Ceratosaurus compete with Goniopholis sp. and Macelognathus vagans, who, as crocodylians were probably better adapted for aquatic environments?

Worse yet: wasn't Macelognathus terrestrial, and far outside the freshwater clade (Goniopholididae + Neosuchia)?

Sorry for my not always excellent English.

David, thank you for your interesting comments.
In part, I've also considered some of these as plausible objections to the amphibious hypothesis.

"What makes you so sure they weren't terrestrial? All ziphosuchians are supposed to have been."

At least Mahajangasuchus has a skull anatomy (in particular, orbit and naris positions) suggesting an aquatic lifestyle.

Yes, I agree with you that abelisaurids could arrive in Madagascar prior to Gondwana break-up. Both hypothesis are plausible pending a better stratigraphic record of abelisaurid lineages.

"1- Pedal unguals ventrally flat, lacking flexor tubercle.
2- Pedal digit IV dorsoventrally compressed.

Don't these simply mean that it didn't use its feet as weapons?"

The distal end is acuminate, not blunt. I've studied pedal phalanges and unguals of Majungasaurus from Madagascar (Maganuco et al., 2008): in my opinion, they seem more adapted for soft substrate than "typical" theropod's ones. It is interesting the asymmetric shape seen in both pedal unguals II and IV of many abelisauroids. Perhaps, it could be related to a different orientation of the phalanges or spreading of the digits when touching the substrate. These features may indicate a incipient adaptation to walk in softer substrate.

" 4- Broader mouth (mandibular symphysis more "U-shaped") than in other ceratosaurs.
Direct consequence of the shorter skull: if the skull gets shorter while maintaining its width, the symphysis has to become rounder."

I disagree. Carnotaurus and Rugops skull are relatively shorter an deeper than Majungasaurus, but their snouts are more "V-shaped" in dorsal view (based on Carrano & Sampson, 2008 and Sereno et al., 2004)

"5- E[x]panded "hypotarsus-like" caudal surface of the proximal end of metatarsal III (antarctometatarsus) for strong pedal phalanges flexor muscles.
Could be connected to the shorter legs needing longer levers for the same effect."

The "hypotarsus" is present also in other ceratosaurs (Aucasaurus, Elaphrosaurus, both more gracile limbed forms than Majungasaurus), so I consider it a co-opted exaptation and not a feature selected by the evolution of stout limbs.

Ref: Maganuco S., Cau A. & Pasini G., 2008 - New information on the abelisaurid pedal elements from the Late Cretaceous of NW Madagascar (Mahajanga Basin). Atti della Società Italiana di Scienze Naturali e del Museo Civico di Storia Naturale in Milano, 149 (II): 239-252.

Sorry for my not always excellent English.

David, thank you for your interesting comments.
In part, I've also considered some of these as plausible objections to the amphibious hypothesis.

"What makes you so sure they weren't terrestrial? All ziphosuchians are supposed to have been."

At least Mahajangasuchus has a skull anatomy (in particular, orbit and naris positions) suggesting an aquatic lifestyle.

Yes, I agree with you that abelisaurids could arrive in Madagascar prior to Gondwana break-up. Both hypothesis are plausible pending a better stratigraphic record of abelisaurid lineages.

"1- Pedal unguals ventrally flat, lacking flexor tubercle.
2- Pedal digit IV dorsoventrally compressed.

Don't these simply mean that it didn't use its feet as weapons?"

The distal end is acuminate, not blunt. I've studied pedal phalanges and unguals of Majungasaurus from Madagascar (Maganuco et al., 2008): in my opinion, they seem more adapted for soft substrate than "typical" theropod's ones. It is interesting the asymmetric shape seen in both pedal unguals II and IV of many abelisauroids. Perhaps, it could be related to a different orientation of the phalanges or spreading of the digits when touching the substrate. These features may indicate a incipient adaptation to walk in softer substrate.

" 4- Broader mouth (mandibular symphysis more "U-shaped") than in other ceratosaurs.
Direct consequence of the shorter skull: if the skull gets shorter while maintaining its width, the symphysis has to become rounder."

I disagree. Carnotaurus and Rugops skull are relatively shorter an deeper than Majungasaurus, but their snouts are more "V-shaped" in dorsal view (based on Carrano & Sampson, 2008 and Sereno et al., 2004)

"5- E[x]panded "hypotarsus-like" caudal surface of the proximal end of metatarsal III (antarctometatarsus) for strong pedal phalanges flexor muscles.
Could be connected to the shorter legs needing longer levers for the same effect."

The "hypotarsus" is present also in other ceratosaurs (Aucasaurus, Elaphrosaurus, both more gracile limbed forms than Majungasaurus), so I consider it a co-opted exaptation and not a feature selected by the evolution of stout limbs.

Ref: Maganuco S., Cau A. & Pasini G., 2008 - New information on the abelisaurid pedal elements from the Late Cretaceous of NW Madagascar (Mahajanga Basin). Atti della Società Italiana di Scienze Naturali e del Museo Civico di Storia Naturale in Milano, 149 (II): 239-252.

I've been following the comments with great interest. As discussed in my article, I think there are problems with the amphibious hypothesis, but then, those who are proposing the hypothesis happily admit this too. I agree with Andrea that at least a few of the features present in Majungasaurus seem weird compared to those of other ceratosaurs and other theropods.

As others have noted before (Greg Paul mentions it in PDW), it's odd that non-avian theropods never (apparently) took to amphibious or aquatic habits: this looks all the stranger when you note that, even in strongly terrestrial modern groups, like carnivorans, artiodactyls, perissodactyls, passerines and so on, there are always at least a few species that deviate in just minor respects from their fully terrestrial relatives, and hence practise amphibious/aquatic habits (e.g., fishing and flat-headed cats, Congo water civet, sitatunga, water buffalo, dippers). This suggests to me that we should at least consider the possibility that really subtle features - like, say, a more spreading metatarsus, or different shaped pedal unguals - might be indicative of amphibious/aquatic habits in Mesozoic dinosaurs. While the evidence for terrestrial habits in sauropods, hadrosaurs and so on looks good (contra the old idea that some or all of these creatures were amphibious or aquatic), itâs quite plausible that the pendulum has swung too far in one direction, and that at least a few species were semi-aquatic after all.

Back to Majungasaurus: how to resolve this? As mentioned earlier, the high level of skeletal pneumaticity in Majungasaurus seems contrary to swimming or diving habits. We'd have to do some Real Science on the subject to be sure, but remember how Don Henderson showed that sauropod buoyancy probably made it difficult for sauropods to swim in a careful and controlled fashion. Then again, a high level of pneumaticity is not necessarily at odds with the hypothesis, as the animal might still be able to forage in water successfully, even with its high pneumaticity. I wonder if bone histology or Air Space Percentage data might be informative? To answer this properly, weâd need data from Majungasaurus, other abelisaurs, other theropods, a diversity of extant archosaurs where the lifestyles are known⦠yikes, a lot of work.

Finally â what about isotope geochemistry? Could isotopic data from Majungasaurus teeth or bones give you some idea as to whether it subsisted on aquatic or terrestrial prey? This would be a neat study, but it would require a Phd-sized project to be thorough and convincing. And, of course, this testing of the hypothesis would only be initiated if the amphibious hypothesis was in the technical literature to begin with â as youâll know if youâve been following SV-POW!, there is a big âDo Blogs Count?â debate going on right now.

Finally â yes, there is a paper in which amphibious habits are suggested for Ceratosaurus: Bakker & Bir (2004). The authors associate a few morphological characters (like a deep, flexible tail) with data from shed teeth. They contend that Ceratosaurus exploited lungfish. Even if this were so, it still seems a stretch to me to imagine Ceratosaurus as a semi-aquatic, swimming predator, which is what Bakker & Bir seem to imagine. Bakker has previously suggested that carcharodontosaurs and spinosaurs were aquatic, marine predators that âmay be a radiation of marine predators analogous to that of the early seals and toothed whales of the Tertiaryâ (Bakker et al. 1992, p. 23), so weâre less inclined to take him seriously on this than if it were someone else saying this.

Refs - -

Bakker, R. T. & Bir, G. 2004. Dinosaur crime-scene investigations: theropod behavior at Como Bluff, Wyoming, and the evolution of birdness. In Currie, P. J., Koppelhus, E. B., Shugar, M. A. & Wright, J. L. (eds) Feathered Dragons. Studies on the Transition from Dinosaurs to Birds. Indiana University Press (Bloomington & Indianapolis), pp. 301-342.

- ., Kralis, D., Siegwarth, J. & Filla, J. 1992. Edmarka rex, a new, gigantic theropod dinosaur from the middle Morrison Formation, Late Jurassic of the Como Bluff outcrop region. Hunteria 2 (9), 1-24.

Thank you Darren for this post!
You wrote:
"As discussed in my article, I think there are problems with the amphibious hypothesis, but then, those who are proposing the hypothesis happily admit this too."

Science is not Religion: I'm very very very happy if someone could show me that I was (in part or totally) wrong.

Your idea on isotopic analysis is very interesting...

But... who is concerned about a hypothetical amphibian ceratosaur, WHEN WE HAVE AN ERBIVOROUS TOOTHLESS ONE?

Yes, Darren! I do think Ceratosaurus was semi-aquatic, though not quite as much a Bakker and Bir were thinking. On comment 5, I even made fun of the "Spinosaurs were dinosaurian basilosaurids" garbage.

Michael just made me aware of this thread; I think Darren makes a good case against an aquatic Majungasaurus. I do think some of the people above making comments on the tail may be looking at the one posted in the article rather than the one he linked to on my DA account. The tail of Majungasaurus is certainly quite long and relatively deep for a theropod, although the nerual spines do taper down in the normal tetanuran fashion.

BTW, I rechecked the proportions on that skeletal reconstruction 3 or 4 times (I was unhappy at how freakish it kept coming out) and I can't find any errors. There is the possibility, however, of the 3 specimens I used not actually being from the same species. I don't know if it's a good possibility, but there were some significant differences in the anterior caudal neural spine morphology between specimens. If the two animals had differing proportions that could explain the apparently odd shape.

That might also cast an interesting light on the issue of cannibalism...

Michael: I'd still love the drawing. Swimming along, transparant view of the water below. I'd like to see how you envision the beastie's swimming posture.

Thanks, all, for the comments about this hypothesis (especially Darren for writing a whole article about it)! Just to be sure: I'm not exactly throwing down the gamut here in suggesting aquatic habits for Majungasaurus. Indeed, I see plenty of reasons why an aquatic habit could NOT work for him, including pneumaticity, a non-retracted external nares, and lack of basically functional forelimbs. However, I also think that theropods (hell, dinosaurs) were probably more generalized in their habits than we tend to think. You wouldn't look at the bones of a moose (Alces alces) and think it could swim, but they're surprisingly good at it, and the ones up here can swim across Turnagain Arm (quite a distance). I'm with Darren here: subtle variation in limb morphology or proportion may be indicative of semi-aquatic habits. I don't know if Majungasaurus fits that particular bill, but I'll betcha there are theropods that do.

Now, MY question is this: if Scott Hartman's skeletal is a good one, and those hindlimb proportions are right, how did Majungasaurus manage to walk around without looking like a drinking bird?

David Marjanovic: I bet you're right about Parasuchus being a foot-propelled swimmer. Ceratosaurus's tail doen't really taper ALL THAT rapidly. The whole of the first third of the tail is the same depth, and the appendage really doesn't start to taper till a little before halfway down it's legnth. Oh, and alligators have backswept haemal spines and vertical caudal neural spines, exactly what we see in Ceratosaurus. Also, the tail skeleton of Alligator actually looks MORE wimpy than that of Ceratosaurus, despite gators (and crocs) being some of the very best tail-propelled swimmers there has ever been.

"What does it matter if the vertebrate is a 'reptile' or not?"

Dinosaurs(including birds) ARE reptiles. I don't like making anologies that compare dinosaurs to mammals (unless, of course, the anology is good plain good, like your Majungasaurus/Xenosmilus thing.) Besides, all of the best side-to-side tail scullers are reptiles.

"How so, if it (the Majungee) comes from the side?"

Whip around before it gets in a good bite, and kick it? Oh, I don't know.

"How so, when it floats at the top of the water like a duck?"

GIANT SAUROPOD-KILLING DUCKS!!!!! No, really, I think you may have got me there, too. Surely the sauropods weren't dumb enough to drink in peace, thinking that the theropod is a giant rubber ducky with teeth.

Anyway just think about what I've said in this comment about the tails, David. And don't have ducky nightmares.

Zach Miller: Will get started on the drawing when I return from the bookstore today. I bet It'll be done in less than a week (it's gonna be more of a detailed drawing than a quickie sketch). Oh, and I've been lovin' them horns and spikes!

>>>Now, MY question is this: if Scott Hartman's skeletal is a good one...<<<

If? IF???!!

Just kidding, but as I noted above, I spent a ridiculous amount of time hunched over a calculator (which is hard to do when you are using Window's default calculator!) checking and rechecking that specimen. The only significant possibility for error I can see is if there are multiple species currently mixed in to Majungasaurus.

Scott: Really? That's bizarre.

And no, seriously, I love your work, brother.

Yes, Scott. IF???!!!! Your skeletals are really the only ones I trust for complete and total accuracy. They're also the coolest-looking ones around.

Scott Hartman said:
"The only significant possibility for error I can see is if there are multiple species currently mixed in to Majungasaurus."

Scott,
in Maganuco et al. (2008), on the basis of pedal elements, we suggested that there could be at least two abelisaurid species in the Maastricthian of Madagascar.
Ref:
Maganuco S., Cau A. & Pasini G., 2008 - New information on the abelisaurid pedal elements from the Late Cretaceous of NW Madagascar (Mahajanga Basin). Atti della Società Italiana di Scienze Naturali e del Museo Civico di Storia Naturale in Milano, 149 (II): 239-252.

As mentioned earlier, the high level of skeletal pneumaticity in Majungasaurus seems contrary to swimming or diving habits. We'd have to do some Real Science on the subject to be sure, but remember how Don Henderson showed that sauropod buoyancy probably made it difficult for sauropods to swim in a careful and controlled fashion. Then again, a high level of pneumaticity is not necessarily at odds with the hypothesis, as the animal might still be able to forage in water successfully, even with its high pneumaticity.

To clarify, skeletal pneumaticity is reduced in divers, not in things that swim at the water's surface. Swans are _highly_ pneumatic. I would imagine that when it went into the water, Majungasaurus was walking on the bottom or swimming, rather than diving. In which case, axial pneumaticity is no problem.

I have mad respect for Don Henderson, but his floating sauropod results are more than a little counterintuitive. I would think with the apneumatic limbs and very dense limb bones, and very light vertebrae, sauropods would be pretty stable in the water. But I really know squat about buoyancy, and his model gives good results for crocs and elephants, so I don't have anything resembling an actual counterargument. It just seems weird.

@ Andrea: Very interesting... As I mentioned, there is some evidence in the two caudal series that they may not be the same. I may have to do a version that breaks the specimens down by color.

On [Ceratosaurus]: Before I got into Pleistocene mammals, I worked on a few sites in the Morrison Formation. In fact, my master's thesis is titled: "An Analysis of the Morrison Formationâs Terrestrial Faunal Diversity
Across Disparate Environments of Deposition,
Including the Aaron Scott Site Dinosaur Quarry in Central Utah." I used taxa occurance lists of 200+ sites (from the PBDB) and conducted cluster analysis, polar ordination, and PCA of the whole Morrison. My results were limited by crappy 19th century ancillary data collection (ex: Locality = Utah???), but I got a few interesting results. Herd structure in sauropods, theropod prey preferences, etc. I did not find any close association of [Ceratosaurus] and [Ceratodus], though. Indeed [Ceratosaurus] seemed to show up around sauropods, and [Ceratodus] co-occured with [Goniopholis], [Glyptops], and a few docodonts, just like you'd expect. I'm not sure what data set Bakker used, but the association isn't supported by the data in the PBDB.

In the spirit of the discussion, I whipped this up: http://skeletaldrawing.com/ext_photos/aquatic_majungasaurus_small.jpg

It's not an endorsement of the idea, but at least provides a more accurate visual aid for the discussion.

The distal end is acuminate, not blunt. I've studied pedal phalanges and unguals of Majungasaurus from Madagascar (Maganuco et al., 2008): in my opinion, they seem more adapted for soft substrate than "typical" theropod's ones. It is interesting the asymmetric shape seen in both pedal unguals II and IV of many abelisauroids. Perhaps, it could be related to a different orientation of the phalanges or spreading of the digits when touching the substrate. These features may indicate a incipient adaptation to walk in softer substrate.

Interesting.

I disagree. Carnotaurus and Rugops skull are relatively shorter an deeper than Majungasaurus, but their snouts are more "V-shaped" in dorsal view (based on Carrano & Sampson, 2008 and Sereno et al., 2004)

Good point.

The "hypotarsus" is present also in other ceratosaurs (Aucasaurus, Elaphrosaurus, both more gracile limbed forms than Majungasaurus), so I consider it a co-opted exaptation and not a feature selected by the evolution of stout limbs.

Then what has lifestyle got to do with it? Do you think all of these were semiaquatic?

You wouldn't look at the bones of a moose (Alces alces) and think it could swim, but they're surprisingly good at it

Does anyone know what happens if you look into the bones? Are moose pachyostotic to any degree?

Now, MY question is this: if Scott Hartman's skeletal is a good one, and those hindlimb proportions are right, how did Majungasaurus manage to walk around without looking like a drinking bird?

Oh, and alligators have backswept haemal spines and vertical caudal neural spines, exactly what we see in Ceratosaurus.

Oopsie. :-]

Also, the tail skeleton of Alligator actually looks MORE wimpy than that of Ceratosaurus, despite gators (and crocs) being some of the very best tail-propelled swimmers there has ever been.

So it's like the thalattosuchians I mentioned and relies on its large, almost unfossilizable scales. Unfair!!!

"What does it matter if the vertebrate is a 'reptile' or not?"

Dinosaurs(including birds) ARE reptiles. I don't like making anologies that compare dinosaurs to mammals

Oh, that's your point... but then, why not say "sauropsids" or "diapsids" or "archosaurs"?

Besides, all of the best side-to-side tail scullers are reptiles.

Anthracosaurs, keraterpetontid and especially urocordylid "nectrideans"... never mind Acanthostega, it's got a real tail fin. (And in fact, so might the anthracosaurs, according to half-published evidence. But plenty of urocordylid tails are known.)

in Maganuco et al. (2008), on the basis of pedal elements, we suggested that there could be at least two abelisaurid species in the Maastricthian of Madagascar.

I need to read that journal much, much more often.

I did not find any close association of [Ceratosaurus] and [Ceratodus], though. Indeed [Ceratosaurus] seemed to show up around sauropods, and [Ceratodus] co-occured with [Goniopholis], [Glyptops], and a few docodonts, just like you'd expect. I'm not sure what data set Bakker used, but the association isn't supported by the data in the PBDB.

Highly interesting. Have you already published?

Interesting also because the only docodonts known from skeletons as opposed to "the tooth, the whole tooth, and nothing but the tooth", Haldanodon and Castorocauda, were semiaquatic. Apparently Docodon itself, too (known from entire lower jaw halves!!!1!), liked hanging out with lungfishes, freshwater crocodiles, and turtles.

Yeah, I really wasn't thinking about amphibians when I said that the best tail-scullers are reptiles, I don't know why.

"But then, why not say 'sauropsids' or 'diapsids' or 'archosaurs'?"

What's wrong with "reptile"? Is it just because the word is too much assostiated with the old Linnaean "Class Reptila"?

And to clear things up for everyone, I never said that shed Ceratosaurus teeth found among the remains aquatic animals *proves* that it was amphibious. As I've already said, we don't have to fool with an amphibious Ceratosaurus to admit that it was fond of aquatic prey. Regardless of anyone's feelings about Bakker, the evidence is in. Amphibious of not, Ceratosaurus's diet was saturated with "seafood".

Now there's a few points I need to make about Ceratosaurus. Fist off, the toes. They are long. Maybe not crazy, stupidly long, but still long when compared to other theropods (including Majungasaurus). If Cerato (as increasingly lazy typists call it) was, indeed, semi-aquatic, I would imagine it swimming with both strong sculling motions of the tail, AND powerful kicks of the hind legs and feet. If this were true, we would expect the toes to be fairly long. This prediction is met. Now, about the tail (again). One thing everyone keeps glancing over (unintentionally, I'm sure) is that the tail of Cerato is UNUSALLY FLEXIBLE FROM SIDE TO SIDE - MORE FLEXIBLE THAN IS NORMAL FOR EVEN A NON-TETANURAN THEROPOD. I'm talkin' croc-like flexibility. Why could this be? Also, the tail skeleton of Cerato is very broad, as well as deep. This bredth is present almost all the way down the tail. In crocs and gators, this sort of bredth is confined to about the first third of the tail. This may mean that the tail of Cerato was even more powerfully muscled than that of a croc. This in a fully terrestrial animal?
I also noticed that Jorge Velez-Joarbe suggested that the shed teeth found among aquatic life could mean that Cerato was taking advantage of a easy meal in the low water levels of each dry season. If this were true, ne would expect to see a 50/50 situation: About half of shed ceratosaur teeth and/or tooth marks found among terrestrial prey (other dinosaurs), the other half among aquatic prey. That's not what we see. Cerato teeth are most common among the remains of aquatic life. There's simply not much evidence that this thing was making a habit of taking terrestrial prey. In science, we should go where the evidence leads, and so far the evidence has led me in the direction of a semi-aquatic Ceratosaurus (although it has left me on the fence regarding the same for Majungasaurus).

Player devil's advocate here: couldn't the shed teeth thing just mean that stuff in marine strata fossilizes better?

You know, Zach buddy, I really should be getting started on that drawing rather than chatting, but who cares, I'll start on it tomarrow.

"Couldn't the shed teeth thing just mean that stuff in marine strata fossilizes better?"

Sure. Of couse, this is freshwater we're talking about here, but yeaaaah..

Scott: Thanks for the sketch. I wonder, though: would the shoulder articulation permit the arms to be carried pointing dorsoposteriorally? I have no reason to think they were, but that would make the sketch more fun to look at.

Ups, I forgot to check Göhlich et al. (2005) beforehand. Johannes and David are right! Macelognathus vagans was indeed a sphenosuchian, and as such was better adapted for terrestrial locomotion.

Göhlich, U. B., L. M. Chiappe, J. M. Clark & H.-D. Sues. 2005. The systematic position of the Late Jurassic alleged dinosaur Macelognathus (Crocodylomorpha: Sphenosuchia). Canadian Journal of Earth Sciences 42(3):307-321.

Bruce J. Mohn (Comment 23): (d) Hesperornis? I have also read somewhere (and I wish i knew where) that cassowaries are terrific swimmers - obviously not habitually, but when they need to, they can.

This has been (and will no doubt continue to be) an interesting discussion. Looking at Scott Hartman's illustration, it does look as if Majungasaurus would have had trouble walking aroudn without going a*** over t**s... IF it walked with its back in such a horizontal orientation...

A question: how far was Majungasaurus able to tilt its body vertically? It almost seems that it would have been best served by raising its body upright to attack a sauropod where it was most vulnerable - its head. Of course this is highly speculative and not based on anything other than my own (probably fantastical) interpretation of Scott's reconstruction, though I suppose some knowledge of the flexibility of the hip region would support / disprove my speculation.

From what I've read, theropods are thought to have been more than adequate swimmers anyway. However, the case against diving and / or semi-aquatic existence made here seems strong. Could Majungasaurus have countered its pneumaticity to any degree by swallowing pebbles to use as ballast as crocodiles do? Is there any evidence of (as I'd imagine you'd still call them) gastroliths in Majungasaurus?

Cheers, and excellent stuff Darren - always a pleasure to read.

Steve P: I'm sorry, but it would have been impossible for Majungasaurus, or any theropod, to tilt it's body too far vertically. From Gregory S. Paul's Predatory Dinosaurs of the World:

"In theropods, important hind limb retractors were anchored on the base of the ischium, the back one of the two lower hip bones. When the hips were horivontal, these muscles were set behind the hind limb, and pull back on the femur while staying within their stretch limits. When the hips tilted up, the ischial retractors were between the hind limbs. Not only would they end up being badly overstretched if they tried to work from this position, but they would not even be able to pull back on the femur."

I'll answer your stone question in a moment, I must leave for a second.

Steve P: As far as I know, no stones that could be interpreted as gastroliths have been found in assosiation with Majungasaurus, but this means nothing. Not all crocodile and bird skeletons, for example, are found with them. Majungee (my nickname for the beast at hand) could still have used "stomach stones", although I really don't know how well that would have coutered its pneumaticity. By the way, pneumaticity is NOT an argument against aquatic and/or amphibious habits, it's an argument against deep diving. Majungasaurus could still have walked along the bottom, or swam at the surface, with little or notrouble. Let's remember that those beautiful amphibious birds we call swans are highly pneumatic.

Yeah, I really wasn't thinking about amphibians when I said that the best tail-scullers are reptiles, I don't know why.

Of my three examples, only the "nectrideans" are probably amphibians. The other two are most definitely not.

(Amphibia = everything closer to the extant amphibians than to the amniotes.)

What's wrong with "reptile"?

Why pick that particular node in the tree? You can't decide between several, so you shouldn't.

Regardless of anyone's feelings about Bakker, the evidence is in. Amphibious of not, Ceratosaurus's diet was saturated with "seafood".

That's what Bakker & Bir say.

Comment 42 says the opposite.

Celebrity deathmatch!!!1!!

If Cerato [â¦] was, indeed, semi-aquatic, I would imagine it swimming with both strong sculling motions of the tail, AND powerful kicks of the hind legs and feet.

When would it use with method? Both at the same time doesn't work.

the tail of Cerato is UNUSALLY FLEXIBLE FROM SIDE TO SIDE - MORE FLEXIBLE THAN IS NORMAL FOR EVEN A NON-TETANURAN THEROPOD. I'm talkin' croc-like flexibility.

What's the evidence for that again?

Also, the tail skeleton of Cerato is very broad, as well as deep. This bredth is present almost all the way down the tail. In crocs and gators, this sort of bredth is confined to about the first third of the tail. This may mean that the tail of Cerato was even more powerfully muscled than that of a croc. This in a fully terrestrial animal?

If we find an excuse to postulate an extremely long musculus caudifemoralis longus, which moves the legs more than the tail, then yes.

I also noticed that Jorge Velez-Joarbe suggested that the shed teeth found among aquatic life could mean that Cerato was taking advantage of a easy meal in the low water levels of each dry season. If this were true, ne would expect to see a 50/50 situation

â¦if the wet and the dry season had the same length. That's nowhere near a safe assumption.

As far as I know, no stones that could be interpreted as gastroliths have been found in assosiation with Majungasaurus, but this means nothing. Not all crocodile and bird skeletons, for example, are found with them.

That's a statistical question. Aren't there insane amounts of M. material out there, all without associated pebbles?

Let's remember that those beautiful amphibious birds we call swans are highly pneumatic.

So much so that they can't stand as deep in the water as the illustration at the top of this page. They basically lie on top of the water and let their feet hang in.

Thanks Michael, though I would just say: Predatory Dinosaurs of the World was written in 1988, Majungasaurus was (originally) described well before but the beautifully preserved remains which are now most often referred to when Majungasaurus is discussed were not discovered (or at least described) until well after this book was published, thus I wonder if Paul's generalisation holds true for Majungasaurus specifically.

Also, should clarify: crocodiles swallowing pebbles as ballast reduces their buoyancy, not their pneumaticity.

Could open a can of worms here, but since we're talking about "alternative" theropod behaviours... ever since I first laid eyes on them, I assumed that Greg Paul's restorations of Ceratosaurus and Dilophosaurus bouncing on their tails were nothing more flights of fancy. Is there any way that this type of "posture" could have been adopted by any of the large theropods?

To end my brain dump, and since we're on the topic of aquatic adaptations: Richard Ellis quotes Mike Everhart in his book Sea Dragons as having suggested that elasmosaurs would not have been able to lift their necks out of the water "unless the laws of physics were suspended" on their behalf - does this argument hold any water? Apologies if this has previously been covered in Tet Zoo, I have yet to find time to completely explore the archives...

Ellis, R. 2003. Sea Dragons: predators of the prehistoric oceans. University Press of Kansas, USA, 313pp.

Everhart, M. J. 2000. Mosasaurs: last of the great marine reptiles. Prehistoric Times 44: 29-31.

Steve P: What Paul says holds true fror ALL theropods (except for segnosaus, which are built differently). Majungasaurus has no special anything in it's hip design that would allow it to break Paul's "rule". Now, Paul's restorations of "kangaroo bouncing theropods" may seem crazy, but they're perfectly sound. There's simply no reason why Ceratosaurus and Dilophosaurus, at least, couldn't do this. Oh, and yes, it's very true there is no possible way that elasmosaurs could have lifted their necks above the water "Nessie-style".

David Marjanovic: When I said amphibians, I WAS only really refering to the "nectrideans". I never really meant to say that the other examples were amphibians, even though it seemed that way.

"That's what Bakker and Bir say."

Not that we should beleive everything we read, or that he's right all the time, but is there any GOOD reason we should doubt Bakker here? The man is a legend. A true vetran in the field of palaeontology. I don't think he's just making stuff up.

"Why would it use this method? Both at the same time doesn't work."

Says who? An emu couldn't swim anymore if it had a croc's tail?

"What's the evidence for that (unusual flexibility in the tail) again?"

I'm not sure if it's appeared in the literature, or where at, but according to both Bakker and Greg Paul, that's how the tail was. We should think that Greg Paul made this up why?

"If we find an exuse to postulate an extremly long musculus caudifemoralis longus, which moves the legs more than the tail, then yes."

No, not really. The spines that stick out sideways from the tail also support ligiments that help to move the tail from side to side. The bigger those spines, the more powerful the ligiments.

"...if the wet season and the dry season had the same legnth. That's nowhere near a safe assumption."

I guess you're right. How long were they, and do we even know?

"Aren't there insane amounts of M. material out there, all without assosiated pebbles?"

I knew there was a lot, but I wasn't aware that there was that much Majungee material out there! Cool.

Oh, and not that this has anything to do with anything, but Bakker isn't the only one who looked at Cerato's tail and thought "swimmer". Marsh (1877) thought as much, as did Greg Paul (1988).

David Marjanovic, on how highly pneumatic swans are, said:

"So much so that they (the swans) can't stand as deep in the water as the illustration at the top of the page. They basically lie on the top of the water and let their feet hang in."

And Majungasaurus couldn't do the same (with the addition of sculling motions of the tail), because???

As regards to Majungasaurus falling on its head, I'm a little confused why people would be pointing at mine; look again at the (mistakenly) scrawny tail on the illustration at the top of this page; even ignoring the two hyper-extended knees, the difference in mass distribution is much worse in terms of being front-heavy.

Remember that the head and neck don't weigh much, and the torso is less dense than the tail. The center of gravity should be pretty directly beneath the forward knee in my skeletal.

A Nathan: I'm not sure, I'd have to take a closer look at the proximal end of the humerus. It would make for a more "floaty" looking scene.

Yes, Scott, yes! I want to let everyone here know that skeletal at the top is TOTALLY screwed up, in almost every way. And ignoring the other problems, it would fall flat on its face long before yours would. Why anyone would think that animal is fine, but that your animal would have had a hard time balancing, is beyond me.

Yeah, I was referring to the photo; especially when using the knee as a reference point your drawing looks much more feasible.

Now, an area I'm not so well-versed in... is that the case for all large theropods, or is there any kind of noticeable variation?

I will only add that all these M. reconstructions have distressingly straight necks. A deeply S-shaped neck posture would move the CG back noticeably. Heavy cranial/cervical pneumaticity and tiny arms seem to imply a need for that.

My neck is pathetic, I know'tDespite ev'ry effort to grow'tWith neckbones pneumaticI could swan operaticWith an 'S' for the shape of my throat

Sorry that this has nothing to do with the discussion at hand, but it does have something to do with ceratosaurs: Does a skeletal recontruction of Limusaurus exist yet? Where can I find it?

Dude: it was only published today! Some of us (hint hint) haven't even seen the pdf yet...

YOU haven't seen the PDF* yet????!!!! That's makes me feel better, 'cause I didn't know the animal existed 'till about a half hour ago...

There's a PDF! There's a PDF!

It was published in Nature today...

Xu, X., Clark, J. M., Mo, J., Choiniere, J., Forster, C. A., Erickson, G. M., Hone, D. W. E., Sullivan, C., Eberth, D. A., Nesbitt, S., Zhao, Q., Hernandez, R., Jia. C.-k., Han, F.-l. & Guo, Y. 2009. A Jurassic ceratosaur from China helps clarify avian digit homologies. Nature 459, 940-944.

As always with brand-new, sexy dinosaurs, I'm relucant to cover it because so many others have already done so (have you seen Dave Hone's Archosaur Musings, Carl Zimmer's The Loom, or Panda's Thumb today?). Tet Zoo likes not to go with the flow: any idiot can summarise the contents of a neat new paper.

Darren: You might want to be just a little more careful of what you call Carl, PZ, and Dave.

Why?

Seriously: I am not, of course, referring to any individuals in particular in this instance. And I'm a hypocrite anyway, as I've changed my mind and will blog about Limusaurus tomorrow. Sigh.

Since my blog is just about theropods, I hope to be justified if I follow the flow. ;-)
I posted my criticism in two posts:
Today: Post 1: In my opinion, this new guy is not an "elaphrosaurid". I apologize for the pedantic taxonomy, but the blog is called "Theropoda".
Tomorrow: Post 2: I think the hand of Limusaurus is not particularly relevant in understanding the evolution of the hand of birds.

Nathan said: "I will only add that all these M. reconstructions have distressingly straight necks. A deeply S-shaped neck posture would move the CG back noticeably. Heavy cranial/cervical pneumaticity and tiny arms seem to imply a need for that."

While a nice strong S-curve would certainly help, it just wasn't in the cards for Majungasaurus (or ceratosaurs in general, which seem to have reduced neck curvature). Granted I've illustrated it in it's neutral position, and as a certain recent paper argues that doesn't mean it was a habitual position. Still, there are large articular facets on the bizzarely massive cervical ribs where they contact, so I can't seem Majungasaurus retracticing its neck into an allosaur or tyrannosaur-like S-curve.

I think the hand of Limusaurus is not particularly relevant in understanding the evolution of the hand of birds.

You know... I agree... it's the best evidence for the frameshift hypothesis so far, but then the reduction of the thumb could just be part of the usual ceratosaurian arm/hand reduction...

"That's what Bakker and Bir say."

Not that we should beleive everything we read, or that he's right all the time, but is there any GOOD reason we should doubt Bakker here? The man is a legend. A true vetran in the field of palaeontology. I don't think he's just making stuff up.

I most certainly don't think he's making stuff up. He might, however, have misinterpreted something. As long as "D. Esker" above hasn't published (what looks like a huge dataset), I don't think we can tell either way.

How math-heavy is the paper by Bakker & Bir (2000)? I don't have it here.

(And of course the "legend" and "veteran" part go both ways, unfortunately. You mentioned his frankly silly idea that carcharodontosaurids and spinosaurids were ocean-going.)

"Why would it use this method? Both at the same time doesn't work."

Says who? An emu couldn't swim anymore if it had a croc's tail?

Sure it could, just not using tail and hindlimbs at the same time.

I'm not sure if it's appeared in the literature, or where at, but according to both Bakker and Greg Paul, that's how the tail was. We should think that Greg Paul made this up why?

What makes you think making stuff up even enters the question? All I'm saying is that functional anatomy can be difficult to interpret (...that's what this whole thread is about, after all). I merely want to know what the actual evidence is that the tail of Ceratosaurus was more flexible than that of, say, Coelophysis; I'm not familiar with those tails.

"If we find an exuse to postulate an extremly long musculus caudifemoralis longus, which moves the legs more than the tail, then yes."

No, not really. The spines that stick out sideways from the tail also support ligiments that help to move the tail from side to side. The bigger those spines, the more powerful the ligiments.

This requires that a muscle attached to the ligaments. Based on the shape of the transverse processes, which one(s) would that be? (That's not a rhetorical question; I'm not familiar with musculature in general.)

"...if the wet season and the dry season had the same legnth. That's nowhere near a safe assumption."

I guess you're right. How long were they, and do we even know?

We might know. There's work on paleosols from the Morrison Fm out there, for example.

I knew there was a lot, but I wasn't aware that there was that much Majungee material out there! Cool.

There's lots. I'll have to check out the SVP Memoir tomorrow.

"So much so that they (the swans) can't stand as deep in the water as the illustration at the top of the page. They basically lie on the top of the water and let their feet hang in."

And Majungasaurus couldn't do the same (with the addition of sculling motions of the tail), because???

Because, unlike birds, it's not flat enough dorsoventrally. It would be highly unstable. Imagine it lying on the water, with the water reaching up to its knees...

Darren: Because while you could take any pair of them to the cleaners for sure, at three-to-one there's a fair chance you'd get scrubbed.

Does it really make any sense to talk about "the length of the dry season" over millions of years? That stuff varies widely within a 10ky window, doesn't it?

David Marjanovic: By "making stuff up", I didn't really mean that anyone thought that they were actually sitting down, thinking up and concocting lies. For example, when I said, "We should think Greg Paul is making this up why?" what I meant to say (and should have said) is "We should think Greg Paul is misinterpreting the tail anatomy why?" In other words, I just think Greg Paul is very credible, and not as likley to misinterpret the evidence as some others (and no, I'm not thinking of anyone in particular) are. Ditto for Bakker. (I really, really wish he had never said that junk about basilosaurid-convergent carcharodontosaurs and spinosaurs; it damaged his otherwise [in my eyes] very credible argument for an amphibious Ceratosaurus.)

"Sure it could, just not using tail and hindlimbs at the same time."

Me reellie gots to bee mor cleer. I never meant that it was using both methods at the same time.

"I merely want to know what the actual evidence is that the tail was more flexible than that of, say, Coelophysis."

In all honesty, I'm not quite sure what the actual evidence is for the tail's amazing flexibility; I just heard Greg Paul say (somewhere) that it was that way, and, not having any real reason to think he's wrong, beleived him. I should have mentioned that.

"This requires that a muscle attached to the ligiments. Based on the shape of the transverse processes, which one(s) whould that be?"

I do know a decent amount about musculature; unfortuanately, it mostly concerns the hindlimb and forelimb musculature of crocs, lizards, birds, and other dinosaurs. From what little bit I *do* know about tail muscles, I really can't say which one (or ones) attached there. All I can really say is that Ceratosaurus had one kick-butt powerful tail (not that this automatically means it was a swimmer, but it's a bit too powerful for an animal that wasn't moving it from side to side a lot - in my eyes anyway).

"Imagine it (Majungasaurus) lying on the water, with the water reaching up to it's knees..."

Yikes! But Matt Wedel said that a Majungee could still walk on the bottom, even with its high levels of pneumaticity. Is this really possible, ya think?

Oh yeah,

"How math-heavy is the paper by Bakker & Bir (2000)?"

It's actually 2004, and it's really very math-heavy.

It's also a VERY interesting, thought-provoking peice of work. You should get it. I highly reccomend it, regardless of weather or not you agree with the conclusion. Just in case you didn't know, it's in Feathered Dragons: Studies on the Transition From Dinosaurs to Birds, edited by the great Phil Currie.

Oh, forget about what I said about not knowing the exact evidence for why the tail of Cerato is unusualy flexible, I know now. From the Bakker & Bir paper (2004):

"...and longer and more flexible torsos with weaker development of interspinous ligaments - a flexibility continued through to the distal tail, where ceratosaurids lack the elongated prezygapophyses that could stiffen the distal half of the caudal appendage."

I have the paper sitting about two inches away from me; I can tell you anything you would like to know from it.

Zach Miller: I'm so sorry, but I just can't get the Majungasaurus drawing right. I've never drawn a Majungee before, and all my attepts are coming out as crap. Hope that's okay. (P.S. I have drawn Ceratosaurus before, and dang I'm good. =;-) Want me to draw a swimming Cerato for you instead?)

Sure, go nuts. :-)

YEAH! NUUUUUUUUUUUUUUUTS!

@ Michael: "...and longer and more flexible torsos with weaker development of interspinous ligaments - a flexibility continued through to the distal tail, where ceratosaurids lack the elongated prezygapophyses that could stiffen the distal half of the caudal appendage."

I'm afraid those are weak arguments when viewed against phylogeny; those are primitive traits for ceratosaurs. So while the anatomy _could_ be used that way, there is no positive evidence of specialization for such a behavior.

Yeah, Scott. That came from the Bakker & Bir (2004) Ceratosaurus paper. I didn't think it up. Thanks for pointing out that those arguments ain't all that strong. The real specialisations that *could* be used in favor of an amphibious Ceratosaurus are how deep the tail was, and how (apparently) well-muscled it was.

If Ceratosaurus was amphibious, wouldn't it have gotten rid of the the three horns located on it's head. They look like they could have caused a lot of drag. Plus the skull seems very unsuited for talking fish and by that I mean that the skull is heavily built on the top and weak on the bottom. Spinosaurs were preying on huge armored fish and they had relatively weak jaws. So spinosaurs prove that a dinosaur wouldn't need such heavily built jaws, as are present in ceratosaurus, to hunt large fish. There is a chance that ceratosaurus could have descended on fish during the dry season when they became trapped by the lowered water level. But I doubt that they were swimming around charging after fish.

So all we really have is a heavily overbuilt, flexible tail. Seems like a clever M. must have come up with some other use for it. Maybe they played polo, or beat one another with it.

Eric: The horns on its head mean nothing. There was an aquatic crocodile called Ceratosuchus that lived during the late Paleocene. It had two big, flaring horns, one above each eye. This proves that horns can be present in amphibious animals. Also, I really doubt that the horns would cause much drag during fast swimming. The skull really isn't that unsuited for taking fish. It could use its powerful jaws and huge knife-shaped teeth to sink through a giant lungfish's armour and into its flesh, and hold on. Remember that the lungfish it would have been preying on were HUGE, nearly as big as the dinosaur itself. The very powerful jaws would have allowed it to hold on the giant fish much better. Spinosaurs had jaw kinks and big thumb claws for that. Spinosaurs also hunted fish that were MUCH, MUCH smaller than the armoured lungfish Ceratosaurus was hunting.

Nathan Myers:

"So all we really have is a heavily overbuilt, flexible tail."

And the (albeit disputable) evidence from shed teeth, which usually appear to be found amoung the remains of aquatic turtles, crocs, and giant lungfish.

That could be why the bite was weaker in spinosaurs, Eric. They were taking much smaller fish, and therefore didn't need an overly powerful bite.

Michael, I wasn't blaming you; I worked with Bob for a few seasons in '90s, so I know most of his arguments for a pisciverous Ceratosaurus. Some ceratosaur teeth seem to have longitudinal grooves, which are often found in fish eaters to help grip them when held in the mouth; I think that is the primary reason he first developed the hypothesis. The other (primitive) characters of the tail were simply consistent with his idea.

I don't personally find the data very convincing, but I haven't ruled it out either.

Michael:

The horns on its head mean nothing.

That conclusion is a bit rash, isn't it? Or has someone actually tested the hydrodynamic properties of a Majungasaurus skull yet?

I really doubt that the horns would cause much drag during fast swimming.

That may be the case, but they'd surely cause some drag; what should be done is to try and demonstrate how much that was before we can say that it wasn't any significant amount (see above).

There was an aquatic crocodile called Ceratosuchus that lived during the late Paleocene.

Yes, but has it not been suggested* that Ceratosuchus fed more on slow-swimming prey (e.g., invertebrates), rather than on fish?

* By Bartels (1984)? Could someone look this up? I don't have access to J. Paleont. at the moment.

Reference:

Bartels, W.S. 1984. Osteology and systematic affinities of the horned alligator Ceratosuchus (Reptilia, Crocodylia). Journal of Paleontology 58, 1347â1353.

Oops, Eric and Michael were talking about the horns of Ceratosaurus, not Majungasaurus. Sorry, my bad. (But the rest of my comment stands.)

Was Ceratosaurus really amphibious? Short answer: no.

Long answer: Bakker has a really annoying habit of proposing novel or extreme ideas (e.g., that Ceratosaurus was amphibious and spent most of its time swimming after lungfish), but then forgetting about them when he needs to discuss the same animal in the context of some other idea. The hindlimb shape of Ceratosaurus, its deep, narrow snout, its big nasal and orbital horns*, and very long, laterally compressed, recurved maxillary teeth all indicate that it was a terrestrial predator that preyed on large dinosaurs (and its tail does not appear specially suited for aquatic locomotion). The skull is absolutely unlike that of an amphibious or aquatic predator, and youâd certainly get this impression from what work has been done on Ceratosaurus skull mechanics (Bakker 2000, Henderson 2000, 2002). Bakker & Bir's primary argument comes from taphonomy, but the one published criticism I've seen of this paper (it was in a book review by, I think, David Fastovsky) said that the sedimentological/stratigraphical interpretation was completely unconvincing. Moreover, it is not so clear that Ceratosaurus is consistently associated with wetlands, or rivers, or lungfish: quite a few studies on Morrison Fm vertebrate taphonomy have reported an association of Ceratosaurus with such dinosaurs as Diplodocus and Allosaurus (google 'Ceratosaurus taphonomy' if you don't believe me). Also, even if Bakker & Bir are right, and there is a nice correlation between shed Ceratosaurus teeth and lungfish remains, this does not mean that Ceratosaurus was amphibious.

* The 'horns' of 'horned' crocodilians (like Ceratosuchus) are laterodorsal projections of the skull deck, and not at all like the big, dorsally projecting structures seen in Ceratosaurus.

As with Majungasaurus, I think it's conceivable that Ceratosaurus foraged at or in the water for food on occasion, and Iâve no doubt that it was a highly capable swimmer. It may be that some individuals, or members of some populations, routinely caught fish. But, at the moment, the notion that this theropod was a dedicated piscivorous swimmer â which is what Bakker & Bir argued â does not ring true. How to resolve this? I don't think anyone has taken Bakker & Bir's argument that seriously actually, but, again, I wonder if isotope geochemistry could tell us what Ceratosaurus was regularly eating (not that I know anything about isotope geochemistry, you understand)...

Refs - -

Bakker, R. T. 2000. Brontosaur killers: Late Jurassic allosaurids as sabre-tooth cat analogues. Gaia 15, 145-158.

Henderson, D. M. 2000. Skull and tooth morphology as indicators of niche partitioning in sympatric Morrison Formation theropods. Gaia 15, 219-226.

- . 2002. The eyes have it: the sizes, shapes, and orientations of theropod orbits as indicators of skull strength and bite force. Journal of Vertebrate Paleontology 22, 766-778.

Michael: you're saying spinosaurs have rostral kinks and hooked claws on their hands for catching fish significantly smaller than themselves, yet you're then proposing that Ceratosaurus, which has proportionally small forelimbs and no rostral kink, to be diving after fish as large as itself? This strikes me as bizarre. I'd be more inclined to think that Dilophosaurus was more suited to a piscivorous way of life (not that I'm suggesting that... or am I?) based on this line of reasoning. Also - aquatic crocodiles have dorsoventrally flattened skulls, whilst that of Ceratosaurus is quite tall (same in Majungasaurus) - surely there is a reason for this? It reminds me of mygalomorph and araneomorph spiders (sorry to bring up octopods on a tetrapod blog...) - mygs have downward pointing fangs because they pursue their prey on the ground which provides upward resistance for their bite, whilst web-building araneomorphs do not have the ground surface to assist them, so their fangs point inwards. Surely there'd be a similar line of reasoning behind relative height of aquatic vs. terrestrial predator skulls? Would the pressure of the water in which aquatic predators are dwelling reduce their need for a tall skull?

Another thing: the Morrison Formation is an ancient floodplain - finding ceratosaur teeth, which are doubtless very robust, in association with lungfish remains may just be a result of their transportation downstream, might it not? And, as suggested above (not by me), environmental preservational bias?

Also, Michael, when answering questions (post 56), is it possible for you to present some actual evidence so that I can actually see why (and where) my thinking or line of questioning is flawed / validated?
How do you know Majungasaurus was not any more pelvically flexible than any other theropod? All swans (discussion coming full circle in anticipation of impending Limusaurus post...) were thought to be white till black ones were found in my island home, thus I'd like to see why Majungasaurus was the same as all other theroppods in this respect.
Why could ceratosaurs bounce on their tails, and is there any reason why ceratosaurs would've, yet other theropod groups would not (obviously anything with a pygostyle is out of the question)? Also - if they could bounce on their tails, why would they have been semi-aquatic (and vice-versa)?

It reminds me of mygalomorph and araneomorph spiders (sorry to bring up octopods on a tetrapod blog...) - mygs have downward pointing fangs because they pursue their prey on the ground which provides upward resistance for their bite, whilst web-building araneomorphs do not have the ground surface to assist them, so their fangs point inwards.

Bad analogy, sorry. The mygalomorph downward-pointing fangs are the basal state for spiders (liphistiomorphs also have downward-pointing fangs), so it's more correct to say that mygalomorphs are more limited in their predation method options because they haven't developed the cross-fangs of araneomorphs. By they way, not all araneomorphs build webs, and a significant proportion of them are ground hunters.

While I agree with several of the objections posted above, I must find myself agreeing that the cranial horns of Ceratosaurus probably don't mean much either way in this argument; worrying about drag implies that swimming speed and/or efficiency were the motivating selective forces, and it's not at all clear that would be the case. Clearly a hypothetically aquatic ceratosaur would not be a pursuit predator - swimming would likely only get it from foraging location to foraging location, etc. It would probably rule out lateral sweeps of the head through the water (ala spinosaurids) but they wouldn't really interfere with a quick dorsoventral strike into the water.

Still, there's no morphology that suggests a specialization for aquatic living, so the argument rises and falls on the taphonomy IMO, which so far is not that convincing, although like David M I wouldn't be surprised if Ceratosaurus (or any other theropod) sometimes took fish in their diet if they were available.

Just a few short comments:

-- Thanks for the complete reference for Bakker & Bir (2004). I have the book, and read that chapter years ago; I just don't have it here with me in Paris.
-- So, despite their length, the teeth of Ceratosaurus are laterally compressed as usual? I would not use knife-shaped teeth to bite on a lungfish. Rather use teeth that are circular in cross-section (like a spinosaur's) or Deadly Bananas Of Death⢠(seen in tyrannosaurids).
-- The spinosaurids of Africa coexisted with hu-u-u-u-uge lungfish (Neoceratodus africanus). Look up my DML post on that sucker from a few months ago.
-- The spinosaurid snout is specifically adapted to resist torsion, as in "twisting, wriggling fish". Ordinary big-theropod snouts, including that of Ceratosaurus, are more like pairs of hacksaw blades used for hacking at humongous prey, and are tall to resist the resulting vertical stresses.

Steve P.:

"...yet you're then proposing that Ceratosaurus, which has proportionally small forelimbs and no rostral kink, to be diving after fish larger than itself?"

Yep. I meant that because the ceratosaur lacks the big forearms and rostral kink of spinosaurs, it made up for it with the huge, powerful skull. IF it was semi-aquatic. IF. That's what we're discussing here. I never said "It was aquatic! It never even walked on land! It was a dinosaurian dolphin!"

"Surely there's a reason for this (the skull of Ceratosaurus being tall rather than flattened)?"

There's something interesting I'd like to say here. I've noticed that the skull is SO tall that, when swimming at the surface, the eyes, nostrils, and ears would have been above the waterline. Like a crocodile, but with everything still on the sides of the head rather than on top of the head. Just an observation, that may or may not have anything to do with anything.

Oh, and I never said that the shed tooth thing COULDN'T be envornmental preservation bias. I've even admited that this could, indeed, be the case.

I'm really not trying to destroy your ideas about Majungasaurus. They're good ideas. It's just that, in order to say that Majungee was more pelvically flexible than other theropods, we should have evidence for this. We don't. The pelvis of Majungasaurus is no different in design than that of any other theropod. Therefore, there's not much of a reason to think that it would have worked differently.

The reason why ceratosaurs and dilophosaurs could bounce on their tails while other theropods couldn't is pretty simple. Ya see, their tails are more flexible than those of most other theropods (which are in the clade Tetanurae - meaning "stiff tails"), and they're more powerfully muscled.
Why, if Cerato was semi-aquatic, why would it bounce on its tail? Maybe it wasn't semi-aquatic after all. I'm perfectly open to that. Also, semi-aquatic means like a 50/50 situation. Half the time on land, half in the water. It doesn't mean that the beast in question spent MORE time in the water than it did on land.

Thanks Michael and Christopher! Sorry that I wasn't 100% clear in my statements - I guess I did set up a bit of a straw man, and I over-generalised with my (inappropriate) analogy to spider fang morphology - thank you for clearing that up! I have a lot to learn...

Steve P: No problem, buddy. If you don't ask, how will you learn? If you don't inquire about stupid junk, how will you ever get smarter*? You don't, that's how. And I think you are smart. You really don't have that much to learn :-)

*That's how I got so smart. Now, exuse me, I must go to the barn and milk the chicken...

There's something interesting I'd like to say here. I've noticed that the skull is SO tall that, when swimming at the surface, the eyes, nostrils, and ears would have been above the waterline.

The entire head, most of the neck, and half of the back and tail would have been above the waterline, I suspect.

"Spinosaurs were preying on huge armored fish and they had relatively weak jaws."

As an aside... I've heard this a lot, but is there an actual study supporting spinosaur jaws being weaker than most other theropods? I know they were obviously more susceptible to torque effects being long and narrow, and their bite force was relatively weak compared to say, tyrannosaurids, but that could be like saying Mike Tyson is weak compared to Magnus ver Magnusson. You'd think a predator relying on quick, snapping motions of the jaw to catch fast-moving, wriggling prey would need a fairly strong bite.

I know in studies done on dogs, skull size and jaw width were found to effect bite strength quite substantially... but as with crocodiles vs. alligators, I know that's not always the case.

I'd be very curious to see a study done on the bite strength of the Gharial and how it compares to both crocs and alligators of similar sizes.

Erm, affect.

Get those two mixed up constantly!

Matt: There was a study done on the bite force of Suchosaurus (=Baryonyx) walkeri. It determined the bite force to be about that of a gharial.

Sorry, thats Suchosaurus (Baryonyx) cultridens.

"Spinosaurs had jaw kinks and big thumb claws for that."

Crap, I can't believe I forgot about that.

Still I think that it is odd that Ceratosaurus would have long toes for swimming and a tail designed for swimming. I may not know much about anatomy, but I know enough that there are two methods of swimming, by the feet or the tail, and I think that using both would use up a lot of energy.

Plus I doubt that Ceratosaurs/Dilophosaurus were hoping on their tails. The simple reason is that the tails would have to be really fat and yet extraordinarily light. Plus, how could they control their movement? I don't think that bouncing is very easy to control, unless the Ceratosaurs/Dilophosaurs had really big tails that they used like bouncy balls. Third, this method of transportation looks really inefficient. Ceratosaurus was 20 feet long so it couldn't have bounced high enough, or far enough to make such an adaptation worth while. And why would dinosaurs choose tail bouncing over simple walking on their feet? Animals usually choose simple lifestyles and don't develop crazy adaptations unless they need them to exploit a niche that other organisms can't use. And since Ceratosaurus and Dilophosaurs appeared to exploit the same role that most other large theropods were exploiting, it seems unlikely that they would develop such a crazy adaptation to accomplish a relatively simple niche (this makes seems oddest in the Dilophosaurs, who were the only carnivores in their ecosystem, at least in the beginning of their evolutionary history, so why would they develop such a crazy method of transportation when they could have much easily just ran).

So there, my "questions" on the proposed oddities of the two adaptations discussed above. Note that I am not an anatomist(in fact I am only 15!) so don't consider this post the a proper scientific rebuttal. Still, it's a free website.

Eric: Ceratosaurus and Dilophosaurus were not actualy jumping around on their tails like Tigger! They were STANDING on their tails and kicking with their hind legs, just like a kangaroo. They would have done this to fight with each other, perhaps over females or territory. I didn't mean "bouncing" literally. About Ceratosaurus' feet and tail, I never said it was using both methods at the same time. Perhaps it swam slowly with kicks of its hind legs (like an emu does), and when it wanted to go faster, used its tail.

"In fact I am only 15!"

Wonderful! It's very nice to see young people excited about science.

>>>Eric: Ceratosaurus and Dilophosaurus were not actualy jumping around on their tails like Tigger! They were STANDING on their tails and kicking with their hind legs, just like a kangaroo. They would have done this to fight with each other, perhaps over females or territory. I didn't mean "bouncing" literally."

Oh, I see. That does actually make a lot of sense.

Yes it does, Eric. Have you ever seen Greg Paul's drawings and paintings of ceratosaurs and dilophosaurs doing this? They're very cool pictures. The ceratosaur drawing is in Predatory Dinosaurs of the World, and the dilophosaur painting is in The Scientific American Book of Dinosaurs.

Posted by: Michel Erickson

Somewhere, your browser has auto-fill-in. You might like to turn it on. :-)

Thanks, David! I'll certainly do that. :-)

Somehow I think saying "spinosaurid basilosaur analogue garbage" is abit harsh.

First, we dont know the leg length of Spinosaurus. Second, I could imagine the sail on spino might have been a sexually dimorphic feature, like lizards. Thirdly, spinosaur tails were like those of perhaps a crocodile (generally), and thirdly, bakker has done more morphological stuff than most students, or at least attended dissections. All I can say as an illustrator is, I'm not convinced. Before you jump on me, I know you would say that they still had airsacs, but so do birds, and they can dive very deep.

@ Tim Morris: I agree that I was too harsh, and I no longer beleive that Bakker's proposal is "garbage". I am now in pretty good agreement with it having gone basck over the evidence.

And to everybody: Feel free to pay no attention to the comments I posted on this blog prior to mid July of this year. I am young, I had never commented on a blog before and I went totally baserk with retarded comments. Basically, I talked a lot of crap. So everyone can just go ahead ignore them, it does not really matter. I didn't "get it" then, I do now.