More out than in
November 24, 2009
I drew a couple of these a while back, and I’m posting them now both to fire discussion and because I’m too lazy to write anything new.
Here’s the neck of Apatosaurus, my own reconstruction based on Gilmore (1936), showing the possible paths and dimensions of continuous airways (diverticula) outside the vertebrae.
Here’s figure 4 from Lovelace et al. (2007), which first got me thinking about pneumatic traces on the ventral surfaces of the centra and what they might imply. You can see pneumatic spaces between the parapophyses in Supersaurus (A) and Apatosaurus (C) but not in Barosaurus (B).
This is another of my moldy oldies, again based on one of Gilmore’s pretty pictures, showing how I think the soft tissues were probably arranged. The muscles are basically the technicolor version of Wedel and Sanders (2002). Two points:
- How bulky you make the neck depends mainly on how much muscle you think was present (which of course depends on how heavy you think the neck was…). Here I was just trying to get the relationships right without worrying about bulk, but it’s worth considering.
- The volume of air inside the vertebra was dinky compared to the probable volume of air outside. In Apatosaurus, either of the canals formed by the transverse foramina has almost twice the cross-sectional area of the centrum.
A fair amount of this has been superseded with better data and prettier pictures by Schwarz et al. (2007), so don’t neglect that work in any ensuing discussion (it’s free, fer cryin’ out loud). And have a happy Thanksgiving!
References
- Gilmore, C.W. 1936. Osteology of Apatosaurus with special reference to specimens in the Carnegie Museum. Memoirs of the Carnegie Museum 11: 175-300.
- Lovelace, D.M., Hartman, S.A., and Wahl, W.R. 2007. Morphology of a specimen of Supersaurus (Dinosauria, Sauropoda) from the Morrison Formation of Wyoming, and a re-evaluation of diplodocid phylogeny. Arquivos do Museu Nacional, Rio de Janeiro 65(4):527-544.
- Schwarz, D., Frey, E., and Meyer, C.A. 2007. Pneumaticity and soft−tissue reconstructions in the neck of diplodocid and dicraeosaurid sauropods. Acta Palaeontologica Polonica 52 (1): 167–188.
- Wedel, M.J., and Sanders, R.K. 2002. Osteological correlates of cervical musculature in Aves and Sauropoda (Dinosauria: Saurischia), with comments on the cervical ribs of Apatosaurus. PaleoBios 22(3):1-6.
Postscript
Mike asked me to add the labeled version of Nima’s brachiosaur parade, so here you go. Click to embiggen.
Finite Element Analysis of sauropod vertebrae
October 27, 2009
Figure 3 from Schwarz-Wings et al. 2009. A is Diplodocus, B-D are Giraffatitan.
Earlier this month Daniela Schwarz-Wings and colleagues published the first finite element analysis (FEA) of sauropod vertebrae (Schwarz-Wings et al. 2009). Above is one of the figures showing some of their results. Following standard convention, stresses are shown on a gradient with cooler colors indicating lower stresses and hotter colors indicating higher stresses. I’m not going to dwell on the on the nuts-n-bolts of FEA in general or of this study in particular. Instead, I want to talk about how sauropod vertebrae are built.
CT cross sections of BYU 12866, a mid-cervical of Brachiosaurus sp.
In cross-section, sauropod vertebrae often have thick bone at the outer edges of the laminae and in the walls and especially the floor of the centrum, as shown in this Brachiosaurus cervical. The bone everywhere else is pretty thin. If you hit one of these vertebrae with some magical forumula that would dissolve away all the bone thinner than, say, 1 cm, all that would be left would be the various apophyses, the outer margins of the laminae connecting them, and probably the bottom half of the centrum. It would be like the outline of a vertebra constructed from tent poles, or tinkertoys.
This is weird because most pneumatic sauropod vertebrae have at least something approaching an I-beam shape in cross-section. You might think that the median septum would be mechanically important, but it’s usually very thin, sometimes perforated (see Hatcher’s [1901] Diplodocus cervicals, for example), and often asymmetrically deviated to one side or the other. Not what you would expect for a piece of bone that was doing any work.
And indeed, Schwarz-Wings et al. (2009) found that:
Comparative stresses are distributed evenly around the vertebrae and mainly on the bone cortex. Peak stresses occur only at points where the tendons and muscles are inserting because the insertion areas used were small resulting in extreme localized stresses. The interior of both vertebrae is nearly stress free. Almost no stresses occur around the cavities and in their bony walls (figure 3).
This reminds me not of I-beams but of the long bones of the limbs of terrestrial vertebrates. There’s a reason why you’ve got a big honkin’ marrow cavity running through the middle of your femur: the stresses are being borne by the walls of the bone. It makes sense that vertebrae would function similarly, especially sauropod cervicals which sometimes approximate limb bones in their proportions.
So how about that median septum? Why aren’t sauropod vertebrae just hollow tubes? My guess–and it is a guess–is that they got as close to being hollow tubes as their evolutionary and developmental origins allowed. The pneumatic diverticula invaded the centra from either side and pushed in lateral-to-medial, and I think the median septum is just the wimpy little bit of bone left in between the two sets of diverticula when they almost meet up in the middle.
Even if that’s correct, there’s another mystery: why don’t the diverticula just go ahead and erode away the median septum? I can think of two possible reasons. One is that, for reasons I don’t know and I’m not sure if anyone else does either, pneumatic diverticula are good at getting into bones but pretty lousy at getting back out. There are comparatively few cases of diverticula inside bones making foramina to get out into the surrounding tissue. It does happen–in humans, the mastoid air cells sometimes bust out and make subcutaneous pneumatocoels, basically bubbles of air under the skin (Anorbe et al. 2000)–but it seems to be rare. Maybe median septa fall under the same inscrutable rule.
(Incidentally, this makes the perforate laminae in Giraffatitan all the weirder.)
Another, more mundane possibility is that the median septa (and other oddly thin bits of bone) are not never loaded, just infrequently loaded. Not enough to make them straight, thick, or normal-lookin’, but enough to make sure they don’t get resorbed entirely.
Sauropod vertebrae are just loaded with these growth-and-form-related mysteries. Kudos to Schwarz-Wings et al. for pushing us a little farther down the road toward solving them.
References
- Anorbe, E., Aisa, P. and Saenz de Ormijana, J. 2000. Spontaneous pneumatocele and pneumocephalus associated with mastoid hyperpneumatization. European Journal of Radiology 36:158–160. [abstract only for free]
- Hatcher, J.B. 1901. Diplodocus (Marsh): its osteology, taxonomy and probable habits, with a restoration of the skeleton. Memoirs of the Carnegie Museum 1: 1-63 and plates I-XIII.
- Schwarz-Wings, D., Meyer, C.A., Frey, E., Manz-Steiner, H.-R., and Schumacher, R. 2009. Mechanical implications of pneumatic neck vertebrae in sauropod dinosaurs. Proceedings of the Royal Society B. doi: 10.1098/rspb.2009.1275
To B.b. or not to B.b — or — So what is a “genus” anyway?
September 12, 2009
Introduction
Back when the Xenoposeidon paper came out, we suggested that Xeno could be the first repesentative of a new sauropod “family”, and then discussed at some length: what is a “family” anyway? Now that the Brachiosaurus paper is out, and I’ve argued that the species “Brachiosaurus” brancai is generically distinct from Brachiosaurus altithorax, it’s time to talk about what a genus is (and so what “generically distinct” means).
In an unnecessarily snarky aside at the end of the last entry, I implied that Randy Irmis would be the one to say that, because my phylogeny recovered “Brachiosaurus” brancai as the sister taxon to Brachiosaurus altithorax, it could and should remain in the genus Brachiosaurus, irrespective of the morphological differences between the genera. He didn’t quite do that — although Daniel Madzia very nearly did — but Jaime Headden certainly did over on the Dinosaur Mailing List, and even ended up asking: “So my question is this: Why do we need Giraffatitan, and cannot have a Brachiosaurus proteles etc.?”
There are plenty of possible responses to this, but before we plough into that, here is a pretty picture:
Ligament rugosities on the neural spines of Brachiosaurus dorsals
Brachiosaurus altithorax holotype FMNH P25107, presacral vertebrae 5-7, neural spines in right posterolateral view
The more retentive among you SV-POW! veterans might remember way back in the very first month of this blog when I showed you what I said were the last four presacral vertebrae of the Brachiosaurus altithorax holotype FMNH P25107. Actually, I don’t know what I was thinking — they were presacrals 4-7, not 1-4, but that’s not the point. The point is that Mike From Ottawa (whatever happened to him?) asked about the very rugose anterior surfaces of the neural spines, and I replied:
What the photo doesn’t show (but if you stay tuned long enough you’ll probably see one that does) is that the posterior faces of the neural spines have very similar rugosities. In life, these would have been the anchor points for epaxial muscles and ligaments. In Brachiosaurus altithorax (but not B. brancai) these have a distinctive inverted-triangle shape. In the most posterior pair of B. brancai dorsals, which are co-ossified, the ligament joining their neural spines is itself ossified. Picture to follow some time, I guess :-)
I am finally following up on the first half of that promise: the picture above shows the three most anterior of those same four presacral vertebrae from the Brachiosaurus altithorax holotype, but this time in right posterolateral view, so you can see the posterior faces of the neural spines. And you’ll notice that on the back of each spine, as well as on the front, there’s a large and extremely rough inverted triangle. I’ve yet to see anything at all like that in any other sauropod — Giraffatitan and the Archbishop included. Very distinctive.
Right then — back to genera!
Rampant genera on the loose!
Here’s a practical reason to reject the idea that if two taxa are sisters, then they should be regarded as congeneric: Jaime wants to retain the species brancai within Brachiosaurus because it is (in the current analysis) the sister to the type species Brachiosaurus altithorax. He then wants to put the species proteles into Brachiosaurus because the species we all know as Sauroposeidon proteles is (presumably) the sister to the Brachiosaurus-altithorax-and-brancai clade. But by induction, if we accept Jaime’s policy, whatever is sister to that clade must also be subsumed into Brachiosaurus, so that we end up losing Titanosauria, Camarasauridae, Diplodocoidea, etc. — Diplodocus carnegii becomes a species of Brachiosaurus. The good side of this scheme is that eventually, we’ll work our way up to the base of Amniota, at which point I become a member of the species Brachiosaurus sapiens. That, I could get on board with. But in other respects, this classification would not be so hot.
I’m assuming that no-one really wants this, and so that advocates of the Sister-Taxa-Are-Congeneric school (hereafter STAC) recognise that you have to draw a line somewhere. But where? And how do you choose where? [Only time will tell whether I just coined an AHATWNUABPANTA.]
How to choose between specific and generic separation
At this point, I am reminded of when I used to be on a mailing list for wannabe writers. Lots of dogma on that list — people saying “don’t overdo adverbs” and “make sure you have enough incidental detail” and so. People trying to nail down an algorithm for good writing. But the best advice I saw on that list was from Jane MacDonald: ”My personal advice is don’t overdo, or underdo, anything in your writing. Do it exactly right.”(*) That’s my attitude to drawing genus boundaries. It is, frankly, an art; and there are no substitutes for taste, experience, judgement, familiarity with the group in question and all those other touchy-feely qualities that uber-cladists would love to find a way to abolish if they could. But they can’t. There is no algorithm for this. I also think of an observation by computer scientist Bjarne Stroustrup, the inventor of the C++ programming language: “Design and programming are human activities; forget that and all is lost.” The same is true of palaeontology. (And of, well, everything.)
Here’s the thing, folks: a genus, just like any other taxon, is there to be useful. Its purpose is not to conform to a dogma, but to inform and enlighten. In the new paper, I wrote that “generic separation is warranted since the two species are more different from each other than, for example, Diplodocus and Barosaurus Marsh, 1890″ (Taylor 2009:798). I stand by that as a great way to figure out when the morphological differences between two species merit generic separation: it’s all about conveying degrees of difference. And, yes, of course I know that the morphological extent of a genus in sauropods is completely different from its extent in, say, botany, where the genus Quercus (oaks) has 700 species or something stupid. Yes, I fully accept that there is no rigorous and absolute standard by which we can determine The Right Place to drop a genus boundary. Sure. But that doesn’t let us out from the responsibility of making the best judgements that we can, based on relevant prior art, recognised conventions, congruence with similar decisions and — there it is again — good taste.
(*) Actually, that is only the second best advice I saw on the wannabe writers’ mailing list. The best advice of all came second-hand, and was passed on by Greg Gunther: “I was on an [email] list with Tom Clancy once. Mr. Clancy’s contribution to the list was, ‘Write the damn book’.” Top advice.
Nomenclatural stability
And so finally I come to Randy’s comment. In response to his question, I guessed that when I put them all in a matrix together, the Archbishop will form a clade with Brachiosaurus, and Sauroposeidon with Giraffatitan. Like this: ((Brachiosaurus altithorax, “The Archbishop”), (Giraffatitan brancai, Sauroposeidon proteles)).
And Randy said:
For the sake of discussion, if the topology is as you say, then I do support the generic separation of altithorax and brancai. Now, of course, as you might surmise, if the two sub-clades are well-supported, I would also advocate putting altithorax and the NHM Tendaguru taxon in the same genus, and brancai and Sauroposeidon in the same genus.
Now I yield to no man in my respect for Randy, whose work exceeds my own humble output by a truly humiliating factor, and who makes it even worse by being such a nice guy. But I hope he will not take it the wrong way if I say that here, he is talking the purest arsegravy. Suppose the topology came out the way I guessed, and we adopted his suggested nomenclature. Then five minutes later Paul Upchurch comes along with a new analysis that finds the Archbishop closer to Giraffatitan after all: and suddenly Brachiosaurus archbishopus becomes Giraffatitan archbishopus. Five more minutes pass and Jeff Wilson publishes his new phylogeny, in which “Sauroposeidon” proteles is sister to Brachiosaurus altithorax, and so what was briefly Giraffatitan proteles becomes Brachiosaurus proteles. Later that afternoon Jerry Harris shows that Cedarosaurus is more closely related to Brachiosaurus altithorax than the species proteles is: at this point, presumably, either Cedarosaurus gets sunk into Brachiosaurus, as B. weiskopfae, or My Big Fat Brachiosaurus Genus gets smashed up and suddenly, woah, proteles needs its own genus after all and Sauroposeidon is back!
Hands up who wants to deal with tracking all that nomenclatural shifting back and forth? Hmm, thought not. Folks, when we name a new species of an existing genus we are betting the nomenclature on the phylogenetic hypothesis. This is just a dumb thing to do in this day and age — especially if you work on dinosaurs which (A) are big and usually very incomplete and so their positions can’t be known with certainty; (B) are trendy enough to be subject to a stream of new phylogenetic analyses; and (C) are in a field where pretty much everyone seems to be hot for mandatory monophyly of genera.
So I end with a plea: unless you know for certain that your new taxon is super-closely related to the type species of an existing genus, and unless you are sure that this isn’t going to change with subsequent discoveries, please put your new species in its own monospecific genus. That way, nomenclature is independent from phylogeny, which is surely how we all want it. A new monospecific genus is essentially a uninomial that happens to be spelled with a space in the middle. And uninomials are nice: they rescue us from Linnaeus’s dumb mistake in lumbering nomenclature with binomials.
This has been an Unwelcome Education Product.
Acknowledgements
Many thanks to Jim Farlow for suggesting the title of this post, which I have cheerfully stolen. I have no idea whether he agrees with the arguments presented in this article.
References
Sauropods were corn-on-the-cob, not shish kebabs
June 25, 2009
This is corn on the cob:
Corn on the cob, in cross section. Stolen from http://www.istockphoto.com/file_thumbview_approve/214165/2/istockphoto_214165-co rn-cob-cross-section.jpg
This is a shish kebab:
Shish kebab. Stolen from http://www.mediterraneancafe-flatiron.com/images/shish.jpg
Most tetrapods are like shish kebabs: a whole lot of meat stuck on a proportionally tiny skeleton. If you don’t believe me, you can look at the human and cow neck torso cross-sections in Matt’s last post, or check out this ostrich-neck cross-section from his 2003 Paleobiology paper:
Ostrich neck in cross section, CT scan. From Wedel (2003a: fig. 2)
Remember that this is a freakin’ ostrich — of all extant animals, one of the ones with a most extreme long, skinny neck. And yet, if sauropods were muscled like ostriches, then their necks would have looked like this in cross section:
Putative shish kebab-style sauropod neck in cross section. Ostrich soft-tissue from Wedel (2003a: fig. 2), Diplodocus vertebra cross-section from Paul (1997: fig. 4) scaled to match size of ostrich vertebra
And soft-tissue reconstructions would have to look like this:
Diplodocus with its neck as fat as an ostrich's. Modified from Paul (1998: fig. 1F)
Which, happily, no-one is suggesting. Instead, published reconstructions of sauropod neck soft-tissue are startlingly emaciated. As exhibit A, I call this pair of Greg Paul cross-sections:
Diplodocus and Brachiosaurus neck cross-sections, showing very light musculature. From Paul (1997: fig. 4)
(Yes, the Diplodocus on the left is the one I used in the photoshopped ostrich cross-section above. It’s instructive to compare Paul’s original with the What If It Was Like A Big Ostrich version.)
Paul’s reconstructions seem to be widely considered too lightly muscled. But even the very careful and rigorous more recent reconstructions of Daniela Schwarz and her colleague show a neck much, much thinner than that of the ostrich:
Diplodocus neck cross-sections. From Schwarz et al. (2007: fig. 7a)
Although Schwarz has put a lot more soft tissue onto the neck vertebrae than Paul did, it is still a tiny proportion of what we see in extant animals — even the ostrich, remember, which has a super-thin neck compared with pretty much anything else alive today. If sauropod necks were muscled as heavily as those of, say, cows, then the soft tissue would pretty much reach down to the ground. But they weren’t: they were more like corn on the cob, with a broad core of skeleton and relatively little in the way of delicious edibles festooned about it.
So why is this? Why does everyone agree that sauropod necks were much less heavily muscled than those of any extant animal?
It’s a simple matter of scaling. A really big ostrich might have a neck 1 m long. (Actually, ostriches don’t get that big, but let’s pretend they do because it makes the maths easier). If the x meter-long neck of a sauropod was just a scaled-up ostrich neck, then it would be x times longer, x times taller and x times wider, for a total of x^3 times as voluminous and therefore x^3 times as heavy. But the cross-sectional area of the tension members that support it is only x times taller and x times wider, for a total of x^2 times the strength. In total, then, the neck’s mass/strength is x^3/x^2 = x times as great as in the ostrich. (The sauropod neck’s mass also acts further out from the fulcrum by an additional factor of x, but that is cancelled by the fact that the tension in the neck also acts x times higher above the fulcrum.)
It seems intuitively obvious (which is is code for “I have no way to prove”) that you can’t reasonably expect the neck muscles of a giant ostrich to work ten times as hard as they do in their lesser cousins, which is what you’d need to do for the 10 m neck of, say, Sauroposeidon. So simple isometric scaling won’t get the job done, and you need to restructure the neck.
But how? Surely just reducing all the muscle around the vertebrae can’t help? No indeed — but that is not really what sauropods were doing. If you look at the typical sauropod-neck life restoration, you’ll see that the proportional thickness of the neck is actually not too dissimilar to that of an ostrich — rather thicker, in fact. If you scaled an ostrich neck up to sauropod size and compared it with a real sauropod neck, you would find not that the soft tissue was too fat, but that the vertebrae were too thin.
And so we come to it at last: rather than thinking of sauropods as having reduced the amount of soft-tissue hanging on the cervical vertebrae, we do better to think of them as having kept a roughly similar soft-tissue profile to that of an an ostrich, but enlarging the vertebrae within the soft-tissue envelope. Of course if you just blindly made the vertebrae taller and wider, they would become heavier in proportion, which would defeat the whole purpose of the exercise — but as everyone who reads this blog surely knows by now, sauropod cervicals were extensively lightened by pneumaticity. By bringing air into the center of the neck, they were effectively able to displace bone, muscle and ligament away from the centre, so that they acted with greater mechanical advantage: higher epaxial tension members, lower hypaxial compression members, and more laterally positioned paraxials.
It’s a rather brilliant system — using the same volume of bone to achieve greater strength by displacing it outwards and filling the center with air (and, in doing so, also displacing soft tissue outwards). And it will be hauntingly familiar to anyone who loves birds, because it is of course exactly what birds (and pterosaurus) have done in their long bones: the hollow humeri of flying vertebrates famously allow them to attain greater strength — specifically, resistance to bending — for the same volume and mass of bone. It’s a neat trick when done with long bones, but it takes a truly awesome taxon to do it with the neck.
So maybe sauropods were not corn on the cob after all. Maybe they were Hostess Twinkies.
Hostess Twinkie. Not truly pneumatic, as the internal cavity is filled with adipose tissue rather than air, but do you have any idea how difficult it is to find good images of hollow junk food? Stolen from http://dixiedining.files.wordpress.com/2008/07/twinkie_070918_ms1.jpg
And now for something completely different
Now that I’ve finished my Ph.D at the University of Portsmouth, what am I going to do with the rest of my scientific life? I’ve always said that I have no intention of going into palaeo full time: my knowledge is far too narrow for that, so that even if paid jobs were not in insanely short supply, I wouldn’t stand much chance of getting one. And in any case, I’d hate to get into the all-too-common situation of being up against a friend for a position we both wanted. Throw in the fact that I really enjoy my computer-programming day-job and it seems pretty clear that what I need is an unpaid affiliation that lets me get on with lovely research.
Well: I am absolutely delighted to announce that, as of last month, I am an Honorary Research Associate in the Department of Earth Sciences at UCL. It’s not just that UCL is such a well-respected institution — see that Wikipedia article for some details — more importantly, it’s where Paul Upchurch hangs out, as Senior Lecturer in Palaeobiology. Sauropod fans will be familiar with Paul’s characteristically detailed and careful work, from his pioneering work on sauropod phylogeny (Upchurch 1995, 1998), through his and John Martin’s indispensible Cetiosaurus makeovers (Upchurch and Martin 2002, 2003) to the state-of-the art review that he lead-authored for Dinosauria II (Upchurch et al. 2004) and the Tokyo Apatosaurus monograph (Upchurch et al. 2005). What many of you won’t know is what an excellent collaborator he is — quick, conscientious, insightful and diplomatic. We’ve already collaborated on a few short papers (Upchurch et al. 2009 and a couple of Phylocode companion-volume chapters that are in press), and I hope there will be more in the future.
References
- Paul, Gregory S. 1997. Dinosaur models: the good, the bad, and using them to estimate the mass of dinosaurs. pp. 129-154 in: D. L. Wolberg, E. Stump, and G. D. Rosenberg (eds.), DinoFest International: Proceedings of a Symposium Sponsored by Arizona State University. Academy of Natural Sciences, Philadelphia.
- Paul, Gregory S. 1998. Terramegathermy and Cope’s Rule in the land of titans. Modern Geology 23: 179-217.
- Schwarz, Daniela, Eberhard Frey and Christian A. Meyer. 2007. Pneumaticity and soft-tissue reconstructions in the neck of diplodocid and dicraeosaurid sauropods. Acta Palaeontologica Polonica 52(1): 167-188.
- Upchurch, Paul. 1995. The evolutionary history of sauropod dinosaurs. Philosophical Transactions of the Royal Society of London Series B, 349: 365-390.
- Upchurch, Paul. 1998. The phylogenetic relationships of sauropod dinosaurs. Zoological Journal of the Linnean Society, 124: 43-103.
- Upchurch, Paul and John Martin. 2002. The Rutland Cetiosaurus: the anatomy and relationships of a Middle Jurassic British sauropod dinosaur. Palaeontology, 45(6): 1049-1074.
- Upchurch, Paul and John Martin. 2003. The anatomy and taxonomy of Cetiosaurus (Saurischia, Suaropoda) from the Middle Jurassic of England. Journal of Vertebrate Paleontology 23(1): 208-231.
- Upchurch, Paul, Paul M. Barrett and Peter Dodson. 2004. Sauropoda. pp. 259-322 in D. B. Weishampel, P. Dodson and H. Osmólska (eds.), The Dinosauria, 2nd edition. University of California Press, Berkeley and Los Angeles. 861 pp.
- Upchurch, Paul, Yukimitsu Tomida, and Paul M. Barrett. 2005. A new specimen of Apatosaurus ajax (Sauropoda: Diplodocidae) from the Morrison Formation (Upper Jurassic) of Wyoming, USA. National Science Museum Monographs No. 26. Tokyo. ISSN 1342-9574.
- Upchurch, Paul, John Martin, and Michael P. Taylor. 2009. Case 3472: Cetiosaurus Owen, 1841 (Dinosauria, Sauropoda): proposed conservation of usage by designation of Cetiosaurus oxoniensis Phillips, 1871 as the type species. Bulletin of Zoological Nomenclature 66(1): 51-55.
- Wedel, Mathew J. 2003. Vertebral pneumaticity, air sacs, and the physiology of sauropod dinosaurs. Paleobiology 29(2): 243-255.
Sauropods were tacos, not corn dogs
June 22, 2009
This is a taco.
This is a corn dog.
Vertebra outlined in green. Click for unmarked original.
Here’s a cross-section of a human. In the terms of fast food, people are corndogs. Most of us even have an outer ring of yellow adipose ‘breading’.
Vertebra oulined in red. Click for unmarked original.
Here’s a cross-section of a cow. In an example of function following form, cows are, and often become, corndogs.
Note that in both the human and the cow the spaces between the neural spine and transverse processes are completely filled with back muscles, which in fact bulge out beyond the tips of the neural spine, as we also saw here. This despite the common paleoart convention of presenting dinosaurs as thin layers of skin conforming perfectly to the underlying skeleton. Just Say No to shrink-wrapped sauropods!
Here is Figure 17 from Holland (1910), one of the most badass scientific smackdowns ever published, in which Holland wiped the floor with Hay, Tornier, and the idea of sprawling sauropods. On the left are torso skeletons of three lizards and a croc; on the right is an anterior dorsal with articulated ribs from Diplodocus. As you can see, it’s a taco, and its taconic form would be perfected if it could roll supine.
The point of the post is not that sauropods had deep, slab-sided bodies. We’ve covered that before. The point is that sauropod torsos are seriously weird. In mammals, the dorsal ribs arch up and out, away from the vertebra, before sweeping around to define the anterior body wall. In lizards, the proximal part of each rib sticks out sideways. In sauropods, the ribs point down. This is mainly because the vertebrae are FREAKIN’ HUGE compared to the size of the body. Whereas in the mammals and lizards the dorsal vertebrae are titchy little things that span a small fraction of the width of the torso, in Diplodocus and other sauropods the dorsal vertebrae account for about half. (The cow cross-section missed the transverse processes, so that vert looks narrower than it actually is.)
This is relevant when we think about the function of pneumaticity. When I write that pneumaticity lightened vertebrae, I usually mean relative to that same vertebra if it wasn’t pneumatized. But we could also ask if the pneumatic vertebra is lighter than a vertebra from a similar-sized animal that lacks pneumaticity–except that, for big sauropods, there are no similar-sized terrestrial animals without pneumaticity to compare.
Imagine that in a big sauropod the dorsal vertebrae are three times as wide and three times as tall as they would be in a similar-sized mammal. They should weigh nine times more. But let’s also assume that the vertebrae of the sauropod are 85% air by volume, which is in fact pretty typical for Early Cretaceous brachiosaurids. The mass of the dorsal column relative to that of the mammal is then 9 x 0.15 = 1.35, a little heavier, but not much (I’m assuming the length of the torso is the same in the two animals). Bigger bones mean better lever arms for the muscles and lower bending stresses on the ribs, which can function more like curtains and less like cantilevered beams.
I can’t think of much published discussion of this stuff as it relates to sauropods, but it seems like it might be important.
Reference
Holland, W.J. 1910. A review of some recent criticisms of the restorations of sauropod dinosaurs existing in the museums of the United States, with special reference to that of Diplodocus carnegiei [sic] in the Carnegie Museum. American Naturalist 44:259-283.
First off, thanks to everyone for reading, commenting on, and discussing the previous post. Seeing the diversity of opinions expressed has been interesting and gratifying for us, and we’ve learned a lot from you about how the blogosphere is changing science already. My own thoughts follow, Mike chimes in at the end, and Darren will probably have something to add soon, too.
The Intolerable Problem
Sometimes people push back on posts of mine they don’t like by telling me I’m out of bounds. Somehow, they say, I’ve crossed the boundary of what I’m allowed to write about. They are angry that I’m now writing about something outside my defined area.
I’m usually taken aback by this, because I didn’t realize I’d actually agreed to any boundaries.
Seth Godin, 2009, “Out of Bounds”
Several commenters have brought up what I call the Intolerable Problem, which is that people online can critique papers and present new evidence and arguments in a format that is impermanent and not peer-reviewed. It’s intolerable because on one hand such material is not currently (operative word) citable in most outlets, and on the other hand repeating it sans citation in peer-reviewed literature smacks of plagiarism (to some, but not to all). Although this material is potentially valuable it “doesn’t count” professionally (see exceptions below), which some professionals (not necessarily those who have commented here) regard as a fatal argument against posting it in the first place. But–and this is crucial–it’s only a problem for the tiny fraction of the audience who might want to cite the freely exchanged material. If you’re in that fraction, we value your attention and comments, but don’t assume we’re writing only for you, or to further our professional standing. We blog because we love this stuff, and even at a technical niche blog like SV-POW! the majority of readers probably don’t care at all whether the information is peer-reviewed or “counts” for professionals; they mostly care whether it’s right or not.
One obvious solution to the Intolerable Problem is to simply let people cite anything they want, including blog posts and DML posts. This is already starting to be implemented–see examples here and here and more discussion here. This runs into two problems: one is permanence (there is no guarantee that the cited post will be up forever, or that the author won’t revise it later in response to criticism [as I have done with this very post!]), which can already be solved using tools such as WebCite (thanks to Cameron Neylon for bringing this to our attention in a comment on the previous post).
The other problem is that citations serve two functions, which are to establish priority and to lend authority to an argument. Citing a blog post may establish priority, but some researchers will cavil at the idea that a blog post is an authoritative source (for varying combinations of researchers and blog posts). Whether they would be right to cavil I don’t know; in the end the market will decide. The market–that is, the desire to attain professional respect and avoid censure–will also dissuade authors from larding up their papers with citations to trivial or worthless online sources.
Those who are troubled by the free discussion of papers, evidence, and hypotheses online need to realize that:
- it’s been going on for a long time (15 years for the Dinosaur Mailing List);
- it’s only going to accelerate in the future;
- it’s not a problem for the vast majority of people participating in the discussions;
- any solution must involve accommodation to the reality of how people exchange information online (immediately, freely, globally, without prior filtering).
These discussions are not going to stop, and ignoring the output of such discussions (because they “don’t count”) will eventually become prohibitively expensive as those workers who insist on playing only by the old rules are outmaneuvered by others who find ways to use all available information regardless of its provenience or “respectability”.
Paper journals will die when online journals stop sucking
Most online publications are hampered by having to be identical to the dead-tree versions (no links, no embedded video, no rotating 3D PDF images, etc.). Eventually people will realize that it is counterproductive to keep hobbling the new medium to make it as slow, flat, and inefficient as the old medium. Once one journal takes the hobbles off, others will do the same rather than lose contributors to cutting-edge outlets. A few boutique journals may still produce flattened, gutted versions of the online publications on paper. People still fly biplanes, too. Paper-based journals will never be popular again and their existence will not stop people from doing whatever technology allows them to in the online venues.
Note that this does not even refer to the economic argument against dead-tree publishing, which has already relocated encyclopedias and newspapers from ubiquity to marginality or extinction.
I’m surprised that the revolution isn’t farther along already. The cage is open.
Whither peer review and editing?
This is all part of the Big Flip in publishing generally, where the old notion of “filter, then publish” is giving way to “publish, then filter.” There is no need for Slashdot’s or Kuro5hin’s owners to sort the good posts from the bad in advance, no need for Blogdex or Daypop to pressure people not to post drivel, because lightweight filters applied after the fact work better at large scale than paying editors to enforce minimum quality in advance.
Clay Shirky, 2003, “The Music Business and the Big Flip”
PLoS ONE is already going gangbusters, without peer-review prior to publication in many cases. The only holdup there is that the post-hoc review by commenters is not working out quite like they’d hoped, because few people are commenting. Not everyone agrees that there is a dearth of commenting at PLoS ONE; the larger point is that people publish there a lot and the community treats those pubs like they count, even though in many cases they are essentially un-reviewed.
[Update: I misunderstood peer review at PLoS ONE. Papers may be reviewed externally by people unconnected to PLoS, or by one or more unpaid Academic Editors, or by a combination. I had thought of the review by Academic Editors only, which accounts for 13% of papers, as a form of internal review, but according to Bora (down in the comments) it should count as external review. If you're happy with that--and the system is not without its critics--then all papers at PLoS ONE are externally reviewed prior to publication; even if you're not, pre-publication review by someone is still in place across the board at PLoS ONE, and 87% of papers are externally reviewed by people unaffiliated with PLoS. Post-publication commenting supplements rather than replaces pre-publication review.]
People do comment on blogs, all the time. Post-hoc review will work, in fact already does work, just fine on blogs. I predict that PLoS ONE clones of the future (PLoS TWO?) will emulate whatever features of blogs make people willing to comment on them but not on PLoS ONE v1.0.
Alternatively, the paucity of post-hoc commenting at PLoS ONE could be taken as further evidence that journal-mediated peer review, whether before or after publication, is dying just off to a slow start. I think that editorial control is not far behind. Both are locally extinct in some parts of the science publishing ecosystem, since people are already citing blogs.
Q: But–but–but? What about protecting the sanctity of the process? What about about guaranteeing respectability? What about prestige?
A: Hey, those questions would make a terrific opinion piece for your local newspaper–oops, too late.
I don’t deny that editors and peer reviewers often make significant contributions to the quality of published work. I just think that people will learn to get along without them if doing so allows faster and easier exchange of information. That was never possible on paper; it’s long been possible here.
A priori peer review and editorial control were invented because publications were scarce (in the Econ 101 sense of being limited) and there needed to be a barrier to entry. Now publication is instant, free, and global. Error correction and the assignment of value will still happen, but they’ll happen after publication rather than before, and they’ll be distributed rather than centralized.
Creeping blogification
Clay Shirky described the problem for newspapers and the recording industry as the existence of “cheap perfect copies”. An expanded but by no means exhaustive list for science publication includes:
- cheap perfect copies
- editable (but also archivable)
- sharable
- linkable (both incoming and outgoing)
- globally distributed
- instantly
- for free
- without pre-publication filtering
- with multimedia embeds (as opposed to including video etc. separately in the suppl. info.)
Online open-access journals currently take advantage of all of those capabilities except the last two. Newsgroup posts cover all the bases except the last one (so do tweets, despite the severe length limitations).
What covers everything? Blog posts. Which have the added advantage that people will comment on them without being asked.
But that’s not the whole simple story.
The center cannot hold–or can it?
So we’re looking at total chaos, right–a world where anyone posts anything they want, no one has any control, and no one knows how to find the good stuff? Well, two out of three, at least. I’m not worried about that last point, for two reasons.
First, thanks to search engines, aggregators, tags, tweets, links, etc., we already have pretty good tools for finding the good stuff. Those direction finders will get better even as the map gets more complicated.
Second, prestige will always be a motivator, so people will always compete to get into exclusive venues. Nature is not going away, although I think that in the near future they will decouple their online and print publications so that the former can take advantage of all the possibilities the web offers.
If I have a really good idea backed up with lots of data, I’ll keep trying to get it into the most prestigious outlet I can. I won’t put my best stuff on a blog just because it’s faster and less encumbered. Blogs probably won’t replace journals, at least not anytime soon. Rather, the spectrum of publishing possibilities will expand; below the category of Least Publishable Unit we’ll add Most Bloggable Unit and so on down to Least Tweetable Unit, and the new categories will interpenetrate with the old over time.
How nice for me
Well, what a striking coincidence that Mr. Paleo Blogger looks into the ole digital crystal ball and sees “bloggy with a 90% chance of exactly-what-he’s-already-doing”.
I can’t claim to be either uninterested or unbiased in all of this. But I am new to actually thinking about the implications. I hadn’t been to most of the above links or had any of these thoughts as of a week ago. When Casey first e-mailed me six days ago, I replied:
If you’re curious, here’s the short short version of my thoughts: science bloggers critique published papers and blog about unpublished observations all the time. Our post-paper run of posts might be an extreme or even vulgar example, and it might fire more discussion about “what counts?”, but I don’t see it as being different in kind from what many science bloggers do. Papers are papers and blogs are blogs, and I never intended to blur the lines. If people feel that all the blog posts only count as “crap some guys wrote on the internet” and that they can be safely ignored, that’s fine with me. If they think the blog posts deserve some higher level of recognition a la “what counts?”, then I’m honored, but that’s extra value that others are investing in our blog, and not anything that we’ve knowingly sought. I suppose you could turn around and say that I’m trying to have my cake and eat it, too, first with all the pro-paper blogging and now with this “I’m innocent” schtick. I don’t know what the answer is, but I know that I’m too tired to figure it out tonight. All the more reason to have an open conversation about this stuff.
Now I realize that the lines between papers and blog posts are blurring, and whether we mean to or not, we SV-POW!sketeers are contributing (Darren’s doing double duty thanks to Tet Zoo). I still think that the investment of blog posts with respectability, value, citability, or whatever rests entirely with readers, and always will. Options range from treating posts like papers to treating them like bar conversations to treating them like spam. You decide.
Also, I tried to keep the writing above value-neutral but probably failed. It’s hard not to get a bit evangelistic about the potential advantages of online publication and online everything else, a tendency I call DISSUADE: Da Internet Shall Save Us All Dead-trees Excepted. Getting published in science hasn’t always been easy up until now, but the process has been relatively clear and familiar. And stable, on decadal and even centennial timescales. Everything about scientific publication is about to get much more fluid and much less clear, and it will probably stay that way for a long time, and it may stay that way forever. Not all of the changes will be for the better, and it may be hard to decide what’s better and what’s worse until we look back with some perspective. Mechanical looms were bad for weavers but good for everyone else. I think many of the changes discussed in this post and the previous comment thread are likely, and some are inevitable.
Set against the shiny digital future is the inertia of the academy and those of us who roost there. I’m not going to stop publishing papers in dead-tree journals (although I will never publish in a journal that doesn’t provide PDFs to authors). Heck, I’m not even going to stop publishing in closed-access journals, some of which are run by societies I admire and want to participate in (after all, everything is open anyway). At the same time I will keep blogging, and while I will frequently bring up technical stuff I don’t want to publish more formally (at least not yet), I will try not to deliberately blur the lines any more than I already have. I don’t need to; the web is already blurring them faster than most of us can keep up.
Hang on.
Oh, about that mystery vert…
Metapophyses, I haz them
…at the end of the post Necks Lie. Nima called it–good spot on the split neural spine. It’s a mid-cervical of Barosaurus, AMNH 6341, in the big bone room (well, one of many big bone rooms) at the American Museum of Natural History in New York. A cast of this vertebra makes up part of the neck in the awesome mounted skeleton in the museum rotunda. Here’s that skeleton, with Mike for scale.
Thanks for slogging through all this. We’ll get back to perforated postcentrodiapophyseal laminae, sacralized caudal transverse processes, and the air space proportions of pneumatic vertebrae soon.
Addendum (from Mike)
Matt is much more ready than I am to throw away peer-review, editorial control, and journals in general. Sometimes, the reasons that things are the way they are, are good ones; it’s not in the interests of professional iconoclasts like Clay Shirky and Cory Doctorow to point that out or to discuss the strengths of how things are today, but that doesn’t mean we have to accept their arguments as uncritically as (say, to pick a name out of the air completely at random) Matt.
Anyway, happily, G. K. Chesterton foresaw the abolition of journals in favour of blogs, and commented thus:
Suppose that a great commotion arises in the street about something, let us say a lamp-post, which many influential persons desire to pull down. A grey-clad monk, who is the spirit of the Middle Ages, is approached upon the matter, and begins to say, in the arid manner of the Schoolmen, “Let us first of all consider, my brethren, the value of Light. If Light be in itself good–” At this point he is somewhat excusably knocked down. All the people make a rush for the lamp-post, the lamp-post is down in ten minutes, and they go about congratulating each other on their unmediaeval practicality. But as things go on they do not work out so easily. Some people have pulled the lamp-post down because they wanted the electric light; some because they wanted old iron; some because they wanted darkness, because their deeds were evil. Some thought it not enough of a lamp-post, some too much; some acted because they wanted to smash municipal machinery; some because they wanted to smash something. And there is war in the night, no man knowing whom he strikes. So, gradually and inevitably, to-day, to-morrow, or the next day, there comes back the conviction that the monk was right after all, and that all depends on what is the philosophy of Light. Only what we might have discussed under the gas-lamp, we now must discuss in the dark.
- Heretics (1905).
A new perspective, or the same old thing?
Brachiosaurus and friends from here (hat tip to Ville Sinkkonen).
In an e-mail with explicit permission to quote, our colleague Casey Holliday sent the following thoughts about our new paper and the subsequent ten days of related blogging:
I don’t know guys. I like your blogs, and your papers are fine. And I liked this paper. And I’m a fan. But it looks to me that you blogged about far more data, in- or not in support of your paper than you actually presented in your paper. So,…wtf? The posts on Dinomorph far exceeded your (or any) published rebuke. Your explanation (and honorable erred parts) of the semicircular canal data also exceeded that actual published part too, with extra photos, description etc. (is that error going to be OA published too?) Also additional pix of necks (e.g., Nigersaurus), and not only from sauropods that would have
potentially bettered the original pub. So what’s fair? Why weren’t
these data also included in the publication? Maybe it’s not my business and was taken up in review…I don’t know. Frankly, none of this blog stuff really counts in the peer-reviewed world of “real” publications. Its not like this blogging and comments all count as Supplementary Data either. But also, I’m obviously here commenting on it, so also crossing into the fray…But who really cares about all this discussion? Its no different than the DML or any other noise in the internet world (or is it). Similar to what Paul Barrett was posting on Tet Zoo…what counts? Why take up arguments here, when they should (maybe?likely?) be taken up more formally and privately.If you’re going to air all this additional data and unreviewed
opinion, then I think this discussion is important.I think this phenomenon of the sauropod neck paper is really
interesting. We have 3 scientists that published a paper, and then, thanks to their current blogosphere cred, basically unleashed a hype not seen in this way previously that I can remember. Maybe that’s the interesting part? and kudos. But interestingly…we’re seeing this intersection of traditional publication (OA or not), blogosphere description, and perhaps, almost certainly, excellent self-promotion.I’m still a fan. I think this paper is generally solid. But I’m
particularly interested in this phenomenon and hope this is a fair
place to raise it.
The comment field is open, and we SV-POW!sketeers are going to refrain from commenting for a couple of days to let the conversation develop unfettered.
We are genuinely curious to know what you think.
What heads tell us about necks, redux
June 5, 2009
I Cannot Brain Today, I Have the Dumb
Man, I hate making mistakes. The only thing worse than making mistakes is making them in public, and the only thing worse than that is finding them in published papers when it’s too late to do anything about them. About the only consolation left–if you’re lucky–is getting to be the one to rat yourself out (we have to do this a lot). So here goes.
Neck angle FAIL
In our figure 4 (from Taylor et al. 2009) we showed the skulls of three sauropodomorphs, Massospondylus, Camarasaurus, and Diplodocus, posed with horizontal semicircular canals (HSCCs) level, angled 30 degrees above horizontal, and angled 20 degrees below horizontal, as it is written (by Duijm 1951). We also showed the angle of the occipital condyle when the HSCCs are level; if the craniocervical joint was in osteologically neutral pose (ONP), that line would indicate the angle of the anterior cervicals.
Trouble is, we put the neck lines for Diplodocus and Camarasaurus in the wrong places.
As any idiot can see from Sereno et al. (2008: fig 1), the brain, brainstem, and occipital condyle form a line that runs from roughly the upper part of the orbit (in lateral see-through view) out the back of the head. Now if you look at our fig. 4 you’ll see that the ONP lines for Camarasaurus and Diplodocus are much too inclined, so that if the brain was in line with the anterior neck–which it should be, in ONP–it would be sticking out the back of the head.
If that doesn’t make sense, just look at the above illustration, imagine the brain and spinal cord in a straight line parallel to the black neck line but also dorsal to it, and you’ll see that the brain would be outside the skull. Those incorrect neck lines don’t represent impossible postures, but they don’t represent ONP, either.
Taxonomic variation WIN!
Here’s a corrected up version of the figure to show what I mean. The black lines are still the ONP neck lines, and now I’ve put in shadowy necks at +30 and -20 to go with the shadowy heads. The 50 degree spans marked out by the shadowy necks are the ranges within which the neck could articulate in ONP with skulls stuck in the 50-degree “Duijm window”.
Caution: it is very easy to misread the shadowy necks as showing a range of movement within an individual; in fact, the neck lines are ‘anchored’ to the skulls in ONP as the skulls rotate through the 50 degrees allowed by the HSCCs. They are not individual movement but the possible range of taxonomic variation in HSCC orientation according to Duijm (1951).
Worth noting here is the likelihood that Massospondylus had a more elevated neck than any of the neosauropods studied so far–certainly a finding at odds with the traditional depictions of basal sauropodomorphs. (It is just a likelihood, though, since the top, neck-wise, of Massospondylus’s Duijm window overlaps with the windows of the other taxa a bit.)
Nigersaurus, buddy, why so down?
In this version I’ve gone one step farther and included Nigersaurus (modified from Sereno et al. (2008: fig 1). Nigersaurus differs from Diplodocus in the angle of the face from the HSCCs and occipital condyle, not in the angle between the HSCCs and the occipital condyle, which is remarkably similar in Camarasaurus, Diplodocus, and Nigersaurus. This suggests that Nigersaurus held its head differently than other sauropods, but not necessarily its neck.
Keep in mind, though, that the difference in facial angle between Diplodocus and Nigersaurus is less than 50 degrees, and that some of the head postures in the respective Duijm windows of the two taxa are identical. So we can’t say for certain that Nigersaurus held its head differently than Diplodocus; it is possible that they held their heads at the same angle and that Nigersaurus just carried its HSCCs at a different angle. If that were the case, the neck of Nigersaurus would have been more inclined than that of Diplodocus. I’m not arguing that that’s likely–it seems perfectly plausible that the two taxa might have held their necks similarly and their heads differently, as suggested above–I’m just pointing out the very wide range of possibilities allowed by the data. To reiterate one of the points of the paper, HSCCs aren’t useless for determining habitual head posture, they just can’t narrow things down very far on their own.
Also note that some of the neck postures allowed by the Duijm window have the anterior cervicals running down, below horizontal, not up. And many of the allowed neck postures for the neosauropods are close to horizontal. So, we were wrong and HSCCs + occipital condyles show that most sauropods held their necks close to level and not strongly elevated after all, right?
Onward and Upward, or Down in Flames?
Not so fast. Remember that all of the neck lines in the above figures show the angle of the anterior neck if the neck was in ONP with the skull. But Vidal et al. (1986) found that the skull is habitually flexed on the neck, even in lizards, and we have since verified this for salamanders, turtles, and more. And sometimes the flexion is dramatic.
Our figure 1 (from Taylor et al. 2009) shows the cranium, cervicals, and first few dorsals from a hare in ONP and in the posture shown by Vidal et al. (1986: fig. 4b). The difference between the anteriorly-directed ONP pose and the backward-leaning Vidal-compliant pose is striking. I measured the angle between the cervical column and the maxillary toothrow to be ~110 degrees in the ONP pose and ~70 degrees in the Vidal-compliant pose (try it yourself with Paint or Photoshop, or download some free image manipulation software). That means the head is flexed on the neck by 40 degrees! That is a big angle. If sauropods did the same, you could take the neck lines shown above and crank them down by 40 degrees (remember that the heads are “fixed” into the 50-degree Duijm windows allowed by the HSCCs), which would make Mike’s elevated Diplodocus look not just achievable, but perhaps even conservative.
Where does all that leave us? In sauropods for which HSCC orientation is known, putting the HSCCs level the anterior neck is still inclined, and even with the HSCCs angled 20 degrees down the ONP neck would only be slightly below horizontal, and if the head was Vidal-compliant (strongly flexed on the neck), the neck would have to be above horizontal. So heads still tell us about necks, and in particular they tell us that the necks angled up. Our neck lines for Camarasaurus and Diplodocus are not correct for ONP, but probably represent attainable postures. My first head ‘n necks post has the angles too exaggeraged for ONP, too, but again all of those poses are not just possible but likely if the head was flexed on the neck.
Miscellanea
We owe mad props to Brian Engh, a.k.a. The Historian, who burst on the paleo-rap scene with a rap video about crocodilian predation and almost certainly the first ever kung-fu rap video to name-check titanosaurs. Brian stumbled across Mike’s extra goodies page for the new paper about week before the paper was due out, and kindly suppressed the information until after D-Day. You can and should download his entire album, Earth Beasts Awaken (open access, yo), and kick it old school.
Congratulations to Francisco “Paco” Gasco, who just got funding for a PhD to do a complete morphological and paleobiological workup on the giant Spanish sauropod Turiasaurus. You’ll be hearing more about Paco in the not-too-distant future, we promise.
Finally, here’s that video of an elephant grabbing an ostrich by the neck that you ordered.
The End of the Beginning?
This brings us to the end of ten solid days of new posts, which is a new record for us and one not likely to be broken for a long time, if ever. We never planned to do all this; in the beginning we each were going to contribute one post and that would have been that. But we kept finding things that we felt needed to be discussed.
As all of us have been saying in every available medium, this is not the end of anything. The sauropod neck posture debate is not over; in a few years we may look back and see that in 2009 we were still stumbling to the real starting line. We don’t think this stuff is unimportant or unknowable, and we’re going to keep working on it, and we hope lots of others do as well.
We’ll see you out there.
Up, boy, up! Heyaaah!!
References
- Duijm, M. 1951. On the head posture in birds and its relation to some anatomical features. II. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Series C 54: 260–271.
- Sereno, Paul C., Jeffrey A. Wilson, Lawrence M. Witmer, John A. Whitlock, Abdoulaye Maga, Oumarou Ide and Timothy A. Rowe. 2007. Structural Extremes in a Cretaceous Dinosaur. PLoS ONE 2 (11): e1230 (9 pages). doi:10.1371/journal.pone.0001230
- Taylor, M.P., Wedel, M.J., and Naish, D. 2009. Head and neck posture in sauropod dinosaurs inferred from extant animals. Acta Palaeontologica Polonica 54 (2): 213–220.
Unstated precision and undemonstrated accuracy: two more reasons why we don’t trust DinoMorph
June 2, 2009
Because the appearance of accuracy has an irresistible allure, non-specialists frequently treat these estimates as factual.
–Graur and Martin (2004: p.80)
Prologue: Why We Hatin’?
Between the first DinoMorph post and this one, it may seem like we have it in for DinoMorph, like we’re trying to discredit the method or bury it. We’re not anti-DinoMorph at all. We really want it to work, because 3D modeling is probably going to be the only way to explore some problems we care about (like the breathing mechanics of an articulated sauropod torso), and so far DinoMorph seems to be farther along than any of the alternatives. It is also worth remembering that building 3D digital dinos for scientific purposes is still in its infancy, and that the VP community has barely gotten started exploring the possibilities. The field has great promise. But we also have to be realistic about limitations in the source data (see Mike’s post) and about the accuracy and precision of the results (this post). We hope that these posts will start constructive conversations and inspire more work to improve the science.
Intro: Accuracy and Precision
Accuracy is how close to the real value a measurement is, and precision is how close repeated measurements are to each other. Say it’s 100 degrees F outside, which it may be for some of you. If you have four thermometers and they read 90, 95, 105, and 110, then the mean is 100. The accuracy of the aggregate setup is high, but the precision is low (big error bars). If, on the other hand, your thermometers read 94.2, 93.8, 94.6, and 93.4, then they are precise (tight grouping) but inaccurate (not centered on the real value)
Oh Error Bars, Where Art Thou?
Here’s what 2 degrees (angular, not temperature) looks like:
It’s not a big measurement. If I was measuring the range of movement (ROM) of a single joint in one individual, like an elbow or shoulder, and I got a precision of plus or minus 2 degrees over repeated movements, I’d be pretty happy. If I got that level of precision on, say, the left knee, in ten different people, I’d start worrying that I was in the Matrix.
All eusauropods have at least 12 cervical vertebrae, and diplodocids have at least 15 (Barosaurus probably has 16, but there are no complete necks so it’s hard to be sure). What happens if we propagate an error of plus or minus 2 degrees down the neck of Diplodocus?
None of these are supposed to correspond to any particular pose in life. I just lined up all the cervicals as straight as I could get them, and then rotated each joint between C3 and C15 by 2 degrees. I left the occipital condyle and C1-C3 in a straight line because I felt the point was made, but the head could be rotated up or down by another 6 degrees if one so chose. Again, this is not an ROM, this is just an error of plus or minus 2 degrees across each of 12 intervertebral joints.
Now let’s look back at the neutral pose and estimated ROM of the neck in the CM 84/94 composite skeleton of Diplodocus (Stevens 2002: fig. 6a):
Notice that the model poses are shown with perfect precision, and no allowance for error. Now, look back up at the first picture to get an idea of what 2 degrees of error looks like, and then try to mentally apply it to each of those three poses. It’s not easy to picture, but in my mind’s eye the three neck poses dissolve into a fuzz of probabilities, like the electron cloud around the nucleus of an atom.
How precise is DinoMorph? Or rather, given that the guts of the program probably allow for Jupiter flyby levels of precision, how precise is any given result, based on the interaction of raw data, necessary but unverified controlling assumptions (see below), and the algorithm itself? Can we really rule out an error of plus or minus 2 degrees per joint? What about 1 degree per joint? What about 5? This is a problem of precision, and it would still exist even with an absolutely perfect neck that was 100% complete and entirely undistorted (which we ain’t got).
It’s possible that the current version of the program doesn’t allow these kinds of error calculations. That’s fine–I realize that DinoMorph, like all of science, is a work in progress. But I’d like to know up front that there is no provision for determining the precision, so I could delay asking the question. And at some point, it will have to be answered.
Maybe it would be better to shift gears and ask: when DinoMorph is applied to extant animals, does it accurately predict the neutral pose and ROM?
Ground Truthiness
It might be better to ask that question, but there are no published answers. From the first DinoMorph paper, where the method is justified (Stevens and Parrish 1999: p. 798):
Our manipulation of muscle and ligament preparations of extant bird necks indicated that synovial capsules constrain movement such that paired pre- and postzygapophyses could only be displaced to the point where the margin of one facet reaches roughly the midpoint of the other facet, at which point the capsule is stretched taut (20). In other words, one facet could slip upon the other until their overlap was reduced to about 50%. In vivo, muscles, ligaments, and fascia may have further limited movement (20); thus, the digital manipulations reported here represent a “best case” scenario for neck mobility.
The reference supporting all this is number 20 (remember how much I like numbered references?), and here’s the full text (Stevens and Parrish 1999: p. 800):
20. J. M. Parrish and K. Stevens, unpublished data.
Those data are still unpublished. But at least one of the basic assumptions–the 50% zyg overlap bit–is contradicted by Stevens and Parrish (2005b: p. 191 [not to mention by Taylor et al. 2009]).
It’s been a decade. There have been three subsequent papers on this stuff (Stevens 2002, Stevens and Parrish 2005a, b). The DinoMorph results have been the foundation for sauropod depictions in the biggest dinosaur documentary ever made and for an exhibit at the biggest natural history museum in the world. And we have no idea if the method is accurate, because the supporting data have never been published.
Sadly, this is not that uncommon in paleontology, particularly when it comes to sauropods, and especially when it comes to necks. Someone comes up with a totally new method, and right out of the gate it gets applied to a thorny paleontological problem, before it’s been demonstrated to work on extant animals. It’s exciting, it’s seductive, and it’s hard to screw up, because when you apply an unproven method to an unsolved problem, it’s impossible to get the wrong answer. In fact, the results are “not even wrong“; it’s impossible to get an answer of any value whatsoever, because there is no way of judging its correctness.
In contrast, the work of Christian and Dzemski (2007) on neck posture in Brachiosaurus warrants serious consideration, not because of the particular answer they got for Brachiosaurus, but because they got the right answers when they applied their method to extant long-necked animals (ostriches and camels; Dzemski and Christian 2007). Don Henderson and Ryosuke Motani, among others, have also been religious about ground-truthing their methods on extant animals before applying them to fossil taxa. That shouldn’t be exceptional. It should be expected. It should be the minimum requirement for being included in the discussion.
Conclusion: Let’s move forward
I can’t accuse the makers of Walking With Dinosaurs or the designers of Dinosaurs: Ancient Fossils, New Discoveries of drinking the DinoMorph Kool-Aid. I don’t know that it is Kool-Aid. It might be fine wine. There’s red stuff in the cup, but no one has tasted it.
If you get nothing else from this post, please understand that I’m not saying the results of DinoMorph are either good or bad. I’m saying that there is currently no objective way of knowing. I want DinoMorph to work, but I want a DinoMorph made rigorous by the publication of supporting data from extant animals demonstrating its accuracy, and ranges of error demonstrating its precision.
If someone has a novel method they want to apply to dinosaurs or any other extinct animal, the burden of proof is on them to show that the method works. And if that evidence is not forthcoming, you–reviewers, editors, readers, science journalists, museum exhibit designers, documentary producers, netizens, laypeople–have the right to ask for it. And until you get that supporting evidence, you don’t have to take the results of the method seriously. Asking “how do you know that?” is the basis of science; it ought to be reflexive.
In the immortal words of Tom Holtz, “Sorry if that makes some people feel bad, but I’m not in the ‘make people feel good business’; I’m a scientist.”
References
- Christian, A. and Dzemski, G. 2007. Reconstruction of the cervical skeleton posture of Brachiosaurus brancai Janensch, 1914 by an analysis of the intervertebral stress along the neck and a comparison with the results of different approaches. Fossil Record 10: 38–49. (subscription required)
- Dzemski, G., and Christian, A. (2007) Flexibility along the neck of the ostrich (Struthio camelus) and consequences for the reconstruction of dinosaurs with extreme neck length. Journal of Morphology 268(8):701-714. (subscription required)
- Graur, D., and Martin, W. 2004. Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision. Trends in Genetics 20:80-86.
- Stevens, K.A. 2002. DinoMorph: Parametric modeling of skeletal structures. Senckenbergiana Lethaea 82(1): 23-34.
- Stevens, K.A. and Parrish, J.M. 1999. Neck posture and feeding habits of two Jurassic sauropod dinosaurs. Science 284: 798–800. (subscription required)
- Stevens, K.A. and Parrish, J.M. 2005a. Neck posture, dentition, and feeding strategies in Jurassic sauropod dinosaurs. In: V. Tidwell and K. Carpenter (eds.), Thunder−Lizards: The Sauropodomorph Dinosaurs, 212–232. Indiana University Press, Bloomington, Indiana.
- Stevens, K.A. and Parrish, J.M. 2005b. Digital reconstructions of sauropod dinosaurs and implications for feeding. In: K.A. Curry Rogers and J.A. Wilson (eds.), The Sauropods: Evolution and Paleobiology, 178–200. University of California Press, Berkeley, California.
- Taylor, M.P., Wedel, M.J. and Naish, D. 2009. Head and neck posture in sauropod dinosaurs inferred from extant animals. Acta Palaeontologica Polonica 54(2): 213-220.
