[This is part 4 in an ongoing series on our recent PLOS ONE paper on sauropod neck cartilage. See also part 1, part 2, and part 3.]

Big Bend Vanessa 182 small

Weird stuff on the ground, Big Bend, 2007.

Here’s a frequently-reproduced quote from Darwin:

About thirty years ago there was much talk that geologists ought only to observe and not theorise; and I well remember some one saying that at this rate a man might as well go into a gravel-pit and count the pebbles and describe the colours. How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service!

It’s from a letter to Henry Fawcett, dated September 18, 1861, and you can read the whole thing here.

I’ve known this quote for ages, having been introduced to it at Berkeley–a copy used to be taped to the door of the Padian Lab, and may still be. It’s come back to haunt me recently, though. An even stronger version would run something like, “If you don’t know what you’re looking for, you won’t make the observation in the first place!”

OLYMPUS DIGITAL CAMERA

Kent Sanders looking at scans of BYU 12613, a posterior cervical of either Kaatedocus or an anomalously small Diplodocus, at the University of Utah in May, 2008.

For example: I started CT scanning sauropod vertebrae with Rich Cifelli and Kent Sanders back in January, 1998. Back then, I was interested in pneumaticity, so that’s what I looked for, and that’s what I found–work which culminated in Wedel et al. (2000) and Wedel (2003). It wasn’t until earlier this year that I wondered if it would be possible to determine the spacing of articulated vertebrae from CT scans. So everything I’m going to show you, I technically saw 15 years ago, but only in the sense of “it crossed my visual field.” None of it registered at the time, because I wasn’t looking for it.

A corollary I can’t help noting in passing: one of the under-appreciated benefits of expanding your knowledge base is that it allows you to actually make more observations. Many aspects of nature only appear noteworthy once you have a framework in which to see them.

OLYMPUS DIGITAL CAMERA

BYI 12613 going through a CT scanner at the University of Utah medical center. We were filming for the “Megasaurus” episode of Jurassic CSI. That shoot was crazy fun.

So anyway, the very first specimen we scanned way back when was the most anterior of the three plaster jackets that contain the four cervical vertebrae that make up OMNH 53062, which was destined to become the holotype of Sauroposeidon. I’ve written about the taphonomy of that specimen here, and you can read more about how it was excavated in Wedel and Cifelli (2005). We scanned that jacket first because, although the partial vertebrae it contains are by far the most incomplete of the four, the jacket is a lot smaller and lighter than the other two (which weigh hundreds of pounds apiece). Right away we saw internal chambers in the vertebrae, and that led to all of the pneumaticity work mentioned above.

Sauroposeidon C5 cross section Wedel 2007b fig 14

Internal structure of a cervical vertebra of Sauroposeidon, OMNH 53062. A, parts of two vertebrae from the middle of the neck. The field crew that dug up the bones cut though one of them to divide the specimen into manageable pieces. B, cross section of C6 in posterior view at the level of the break, traced from a CT image and photographs of the broken end. The left side of the specimen was facing up in the field and the bone on that side is badly weathered. Over most of the broken surface the internal structure is covered by plaster or too damaged to trace, but it is cleanly exposed on the upper right side (outlined). C, the internal structure of that part of the vertebra, traced from a photograph. The arrows indicate the thickness of the bone at several points, as measured with a pair of digital calipers. The camellae are filled with sandstone. Wedel (2007: fig. 14).

Happily for me, that first jacket contains not only the posterior two-thirds of the first vertebra (possibly C5), but also the front end of the second vertebra. Whoever decided to plow through the second vertebra to divide the specimen into manageable chunks in the field made a savvy choice. Way back in 2004 I realized that the cut edge of the second vertebra was not obscured by plaster, and therefore the internal structure could be seen and measured directly, which is a lot cleaner than relying on the artifact-heavy CT scans. (The CT scans are noisy because the hospital machines we had access to start to pant a bit when asked to punch x-rays through specimens this large and dense.) A figure derived from that work made it into a couple of papers and this post, and appears again above.

But that’s pneumaticity, which this post is allegedly not about. The cut through the second vertebra was also smart because it left the intervertebral joint intact.

Figure 11. Fifth and partial sixth cervical vertebrae of Sauroposeidon. Photograph and x-ray scout image of C5 and the anterior portion of C6 of Sauroposeidon OMNH 53062 in right lateral view. The anterior third of C5 eroded away before the vertebra was collected. C6 was deliberately cut through in the field to break the multi-meter specimen into manageable pieces for jacketing (see [37] for details). Note that the silhouettes of the cotyle of C5 and the condyle of C6 are visible in the x-ray.

Fifth and partial sixth cervical vertebrae of Sauroposeidon.
Photograph and x-ray scout image of C5 and the anterior portion of C6 of Sauroposeidon OMNH 53062 in right lateral view. The anterior third of C5 eroded away before the vertebra was collected. C6 was deliberately cut through in the field to break the multi-meter specimen into manageable pieces for jacketing (see Wedel and Cifelli 2005 for details). Note that the silhouettes of the cotyle of C5 and the condyle of C6 are visible in the x-ray. Taylor and Wedel (2013: figure 11).

Here are a photo of the jacket and a lateral scout x-ray. The weird rectangles toward the left and right ends of the x-ray are boards built into the bottom of the jacket to strengthen it.

Figure 12. CT slices from fifth cervical vertebrae of Sauroposeidon. X-ray scout image and three posterior-view CT slices through the C5/C6 intervertebral joint in Sauroposeidon OMNH 53062. In the bottom half of figure, structures from C6 are traced in red and those from C5 are traced in blue. Note that the condyle of C6 is centered in the cotyle of C5 and that the right zygapophyses are in articulation.

CT slices from fifth cervical vertebrae of Sauroposeidon.
X-ray scout image and three posterior-view CT slices through the C5/C6 intervertebral joint in Sauroposeidon OMNH 53062. In the bottom half of figure, structures from C6 are traced in red and those from C5 are traced in blue. Note that the condyle of C6 is centered in the cotyle of C5 and that the right zygapophyses are in articulation. Taylor and Wedel (2013: figure 12).

And here’s a closeup of the C5/C6 joint, with the relevant radiographs and tracing. The exciting thing here is that the condyle is centered almost perfectly in the cotyle, and the zygapophyses are in articulation. Together with the lack of disarticulation in the cervical rib bundle (read more about that here and in Wedel et al. 2000), these things suggest to us that the vertebrae are spaced pretty much as they were in life. If so, then the spacing between the vertebrae now tells us the thickness of the soft tissue that separated the vertebrae in life.

I should point out here that we can’t prove that the spacing between the vertebrae is still the same as it was in life. But if some mysterious force moved them closer together or farther apart, it did so (1) without  decentering the condyle of C6 within the cotyle of C5, (2) without moving the one surviving zygapophyseal joint out of contact, and (3) without disarticulating the cervical ribs. The cervical ribs were each over 3 meters long in life and they formed vertically-stacked bundles on either side below the vertebrae; that’s a lot of stuff to move just through any hypothetical contraction or expansion of the intervertebral soft tissues after death. In fact, I would not be surprised if the intervertebral soft tissues did contract or expand after death–but I don’t think they moved the vertebrae, which are comparatively immense. The cartilage probably pulled away from the bone as it rotted, allowing sediment in. Certainly every nook and cranny of the specimen is packed with fine-grained sandstone now.

Anyway, barring actual preserved cartilage, this is a best-case scenario for trying to infer intervertebral spacing in a fossil. If articulation of the centra, zygs, and cervical ribs doesn’t indicate legitimate geometry, nothing ever will. So if we’re going to use the fossils to help settle this at all, we’re never going to have a better place to start.

Figure 14. Geometry of opisthocoelous intervertebral joints. Hypothetical models of the geometry of an opisthocoelous intervertebral joint compared with the actual morphology of the C5/C6 joint in Sauroposeidon OMNH 53062. A. Model in which the condyle and cotyle are concentric and the radial thickness of the intervertebral cartilage is constant. B. Model in which the condyle and cotyle have the same geometry, but the condyle is displaced posteriorly so the anteroposterior thickness of the intervertebral cartilage is constant. C. the C5/C6 joint in Sauroposeidon in right lateral view, traced from the x-ray scout image (see Figure 12); dorsal is to the left. Except for one area in the ventral half of the cotyle, the anteroposterior separation between the C5 cotyle and C6 condyle is remarkably uniform. All of the arrows in part C are 52 mm long.

Geometry of opisthocoelous intervertebral joints.
Hypothetical models of the geometry of an opisthocoelous intervertebral joint compared with the actual morphology of the C5/C6 joint in Sauroposeidon OMNH 53062. A. Model in which the condyle and cotyle are concentric and the radial thickness of the intervertebral cartilage is constant. B. Model in which the condyle and cotyle have the same geometry, but the condyle is displaced posteriorly so the anteroposterior thickness of the intervertebral cartilage is constant. C. the C5/C6 joint in Sauroposeidon in right lateral view, traced from the x-ray scout image (see Figure 12); dorsal is to the left. Except for one area in the ventral half of the cotyle, the anteroposterior separation between the C5 cotyle and C6 condyle is remarkably uniform. All of the arrows in part C are 52 mm long. Taylor and Wedel (2013: figure 14).

So, by now, you know I’m a doofus. I have been thinking about this problem literally for years and the data I needed to address it was sitting on my hard drive the entire time. One of the things I pondered during those lost years is what the best shape for a concave-to-convex intervertebral joint might be. Would the best spacing be radially constant (A in the figure above), or antero-posteriorly constant (B), or some other, more complicated arrangement? The answer in this case surprised me–although the condyle is a lot smaller in diameter than the cotyle, the anteroposterior separation between them in almost constant, as you can see in part C of the above figure.

Figure 13. Joint between sixth and seventh cervicals vertebrae of Sauroposeidon. X-ray scout image of the C6/C7 intervertebral joint in Sauroposeidon OMNH 53062, in right lateral view. The silhouette of the condyle is traced in blue and the cotyle in red. The scale on the right is marked off in centimeters, although the numbers next to each mark are in millimeters.

Joint between sixth and seventh cervicals vertebrae of Sauroposeidon.
X-ray scout image of the C6/C7 intervertebral joint in Sauroposeidon OMNH 53062, in right lateral view. The silhouette of the condyle is traced in blue and the cotyle in red. The scale on the right is marked off in centimeters, although the numbers next to each mark are in millimeters. Taylor and Wedel (2013: figure 13).

Don’t get too worked up about that, though, because the next joint is very different! Here’s the C6/C7 joint, again in a lateral scout x-ray, with the ends of the bones highlighted. Here the condyle is almost as big in diameter as the cotyle, but it is weirdly flat. This isn’t a result of overzealous prep–most of the condyle is still covered in matrix, and I only found its actual extent by looking at the x-ray. This is flatter than most anterior dorsal vertebrae of Apatosaurus–I’ve never seen a sauropod cervical with such a flat condyle. Has anyone else?

The condyle of C6 is a bit flatter than expected, too–certainly a lot flatter than the cervical condyles in Giraffatitan and the BYU Brachiosaurus vertebrae. As we said in the paper,

It is tempting to speculate that the flattened condyles and nearly constant thickness of the intervertebral cartilage are adaptations to bearing weight, which must have been an important consideration in a cervical series more than 11 meters long, no matter how lightly built.

Anyway, obviously here the anteroposterior distance between condyle and cotyle could not have been uniform because they are such different shapes. Wacky. The zygs are missing, so they’re no help, and clearly the condyle is not centered in the cotyle. Whether this posture was attainable in life is debatable; I’ve seen some pretty weird stuff. In any case, we didn’t use this joint for estimating cartilage thickness because we had no reason to trust the results.

Figure 15. First and second dorsal vertebrae of Apatosaurus CM 3390. Articulated first and second dorsal vertebrae of Apatosaurus CM 3390. A. Digital model showing the two vertebrae in articulation, in left lateral (top) and ventral (bottom) views. B-G. Representative slices illustrating the cross-sectional anatomy of the specimen, all in posterior view. B. Slice 25. C. Slice 31. D. Slice 33. E. Slice 37. F. Slice 46. G. Slice 61. Orthogonal gaps are highlighted where the margins of the condyle and cotyle are parallel to each other and at right angles to the plane of the CT slice. 'Zygs' is short for 'zygapophyses', and NCS denotes the neurocentral synchondroses.

First and second dorsal vertebrae of Apatosaurus CM 3390.
Articulated first and second dorsal vertebrae of Apatosaurus CM 3390. A. Digital model showing the two vertebrae in articulation, in left lateral (top) and ventral (bottom) views. B-G. Representative slices illustrating the cross-sectional anatomy of the specimen, all in posterior view. B. Slice 25. C. Slice 31. D. Slice 33. E. Slice 37. F. Slice 46. G. Slice 61. Orthogonal gaps are highlighted where the margins of the condyle and cotyle are parallel to each other and at right angles to the plane of the CT slice. ‘Zygs’ is short for ‘zygapophyses’, and NCS denotes the neurocentral synchondroses. Taylor and Wedel (2013: figure 15).

Kent Sanders and I had also scanned several of the smaller sauropod vertebrae from the Carnegie collection (basically, the ones that would fit in the trunk of my car for the drive back to Oklahoma). Crucially, we’d scanned a couple of sets of articulated vertebrae, CM 3390 and CM 11339, both from juvenile individuals of Apatosaurus. In both cases, the condyles and cotyles are concentric (that’s what the ‘orthogonal gaps’ are all about in the above figure) and the zygs are in articulation, just as in Sauroposeidon. These are dorsals, so we don’t have any cervical ribs here to provide a third line of evidence that the articulation is legit, but all of the evidence that we do have is at least consistent with that interpretation.

So, here’s an interesting thing: in CM 3390, above, the first dorsal is cranked up pretty sharply compared to the next one, but the condyle is still centered in the cotyle and the zygs are in articulation. Now, the vertebrae have obviously been sheared by taphonomic deformation, but that seems to have affected both vertebrae to the same extent, and it’s hard to imagine some kind of taphonomic pressure moving one vertebra around relative to the next. So I think it’s at least plausible that this range of motion was achievable in life. Using various views and landmarks, we estimate the degree of extension here somewhere between 31 and 36 degrees. That’s a lot more than the ~6 degrees estimated by Stevens and Parrish (1999, 2005). And, as we mentioned in the paper, it nicely reinforces the point made by Upchurch (2000), that flexibility in the anterior dorsals should be taken into account in estimating neck posture and ROM.

Figure 16. Dorsal vertebrae of Apatosaurus CM 11339. Articulated middle or posterior dorsal vertebrae of Apatosaurus CM 11339. A. X-ray scout image showing the two vertebrae in articulation, in left lateral view. B–D. Slices 39, 43 and and 70 in posterior view, showing the most anterior appearance of the condyles and cotyles.

Dorsal vertebrae of Apatosaurus CM 11339.
Articulated middle or posterior dorsal vertebrae of Apatosaurus CM 11339. A. X-ray scout image showing the two vertebrae in articulation, in left lateral view. B–D. Slices 39, 43 and and 70 in posterior view, showing the most anterior appearance of the condyles and cotyles. Taylor and Wedel (2013: figure 16).

Here’s our last specimen, CM 11339. No big surprises here, although if you ever had a hard time visualizing how hyposphenes and hypantra fit together, you can see them in articulation in parts C and D (near the top of the specimen). Once again, by paging through slices we were able to estimate the separation between the vertebrae. Incidentally, the condyle IS centered in the cotyle here, it just doesn’t look that way because the CT slice is at an angle to the joint–see the lateral scout in part A of the figure to see what I mean.

So, what did we find? In Sauroposeidon the spacing between C5 and C6 is 52mm. That’s pretty darn thick in absolute terms–a shade over two inches–but really thin in relative terms–only a little over 4% of the length of each vertebra. In both of the juvenile Apatosaurus specimens, the spacing between the vertebrae was about 14mm (give or take a few because of the inherent thickness of the slices; see the paper for details on these uncertainties).

Now, here’s an interesting thing: we can try to estimate the intervertebral spacing in an adult Apatosaurus in two ways–by scaling up from the juvenile apatosaurus, or by scaling sideways from Sauroposeidon (since a big Apatosaurus was in the same ballpark, size-wise)–and we get similar answers either way.

Scaling sideways from Sauroposeidon (I’m too lazy to write anymore so I’m just copying and pasting from  the paper):

Centrum shape is conventionally quantified by Elongation Index (EI), which is defined as the total centrum length divided by the dorsoventral height of the posterior articular surface. Sauroposeidon has proportionally very long vertebrae: the EI of C6 is 6.1. If instead it were 3, as in the mid-cervicals of Apatosaurus, the centrum length would be 600 mm. That 600 mm minus 67 mm for the cotyle would give a functional length of 533 mm, not 1153, and 52 mm of cartilage would account for 9.8% of the length of that segment.

Scaling up from the juveniles: juvenile sauropods have proportionally short cervicals (Wedel et al. 2000). The scanned vertebrae are anterior dorsals with an EI of about 1.5. Mid-cervical vertebrae of this specimen would have EIs about 2, so the same thickness of cartilage would give 12mm of cartilage and 80mm of bone per segment, or 15% cartilage per segment. Over ontogeny the mid-cervicals telescoped to achieve EIs of 2.3–3.3. Assuming the cartilage did not also telescope in length (i.e., didn’t get any thicker than it got taller or wider), the ratio of cartilage to bone would be 12:120 (120 from 80*1.5), so the cartilage would account for 10% of the length of the segment–almost exactly what we got from the based-on-Sauroposeidon estimate. So either we got lucky here with our tiny sample size and truckloads of assumptions, or–just maybe–we discovered a Thing. At least we can say that the intervertebral spacing in the Apatosaurus and Sauroposeidon vertebrae is about the same, once the effects of scaling and EI are removed.

Finally, we’re aware that our sample size here is tiny and heavily skewed toward juveniles. That’s because we were just collecting targets of opportunity. Finding sauropod vertebrae that will fit through a medical-grade CT scanner is not easy, and it’s just pure dumb luck that Kent Sanders and I had gotten scans of even this many articulated vertebrae way back when, since at the time we were on the hunt for pneumaticity, not intervertebral joints or their soft tissues. As Mike has said before, we don’t think of this paper as the last word on anything. It is, explicitly, exploratory. Hopefully in a few years we’ll be buried in new data on in-vivo intervertebral spacing in both extant and extinct animals. If and when that avalanche comes, we’ll just be happy to have tossed a snowball.

References

One aspect of sauropod neck cartilage that’s been overlooked — and this applies to all non-avian dinosaurs, not just sauropods — is the configuration of the cartilage in their necks. It’s not widely appreciated that birds’ necks differ from those of all other animals in this respect, and we don’t yet know whether sauropods resembled birds or mammals.

Here’s a classic sagittal view of a mammal neck — in this case a human — from The Basics of MRI (Joseph P. Hornak, 1996-2013):

sagittal-neck

You can see two distinct kinds of structure alternating along the neck: the big, square ones are vertebral centra (slightly hollow at each end), and the narrower lens-shaped ones are the intervertebral discs.

In mammals, and most animals, we find this distinct fibrocartilaginous element, the disc, between the centra of consecutive vertebrae. These discs have a complex structure of their own, consisting of an annulus fibrosus (fibrous ring), made of several layers of fibrocartilage, surrounding a nucleus pulposus (pulpy centre) with the consistency of jelly.

IntervertebralDisc

But in birds, uniquely among extant animals, there is no separate cartilaginous element. Instead, the articular surfaces of the bones are covered with layers of hyaline cartilage which articulate directly with one another, and are free to slide across each other. The adjacent articular surfaces are enclosed in synovial capsules similar to those that enclose the zygapophyseal joints. You can see this in the hemisected Rhea neck from last time:

Figure 18. Cartilage in the neck of a rhea. Joint between cervicals 11 (left) and 10 (right) of a rhea, sagittally bisected. Left half of neck in medial view. The thin layers of cartilage lining the C11 condyle and C10 cotyle are clearly visible.

Taylor and Wedel (2013c: Figure 18). Cartilage in the neck of a rhea. Joint between cervicals 11 (left) and 10 (right) of a rhea, sagittally bisected. Left half of neck in medial view. The thin layers of cartilage lining the C11 condyle and C10 cotyle are clearly visible.

The difference between these two constructions is very apparent in dissection: in birds, adjacent vertebrae come apart easily once the surrounding soft tissue is removed; but in mammals, it is very difficult to separate consecutive vertebrae, as they are firmly attached to the intervening intervertebral disc.

Figure 19. Alligator head and neck. Sagittally bisected head and neck of American alligator, with the nine cervical vertebrae indicated. Inset: schematic drawing of these nine vertebrae, from ([62]: figure 1), reversed.

Taylor and Wedel (2013c: Figure 19). Alligator head and neck. Sagittally bisected head and neck of American alligator, with the nine cervical vertebrae indicated. Inset: schematic drawing of these nine vertebrae, from ([62]: figure 1), reversed.

To complicate matters further, thin articular discs occur in the necks of some birds — for example, the ostrich (see illustration below), the swan, and the king penguin. But these discs do not occur in all birds — for example, they are absent in the turkey and the rhea. When they are present, these articular discs divide the synovial cavity and prevent the (cartilage-covered) bones on either side from ever articulating directly with each other, just like the articular discs in the human temporomandibular and sternoclavicular joints. These discs are thinner than the true intervertebral discs of mammals and crocodilians; and they are different in composition, lacking the annulus/nucleus structure and consisting of a simple sheet of fibrocartilage.

Taylor and Wedel (2013: Figure 4). Intervertebral articular discs of an ostrich (not to scale). Left: first sacral vertebra in anterior view, showing articular disc of joint with the last thoracic vertebra. Right: posterior view view of a cervical vertebra, with probe inserted behind posterior articular disc. The cervical vertebra is most relevant to the present study, but the the sacral vertebra is also included as it shows the morphology more clearly. These fibrocartilaginous articular discs divide the synovial cavity, like the articular discs in the human temporomandibular and sternoclavicular joints, and should not be confused with the true intervertebral discs of mammals and other animals, which consist of a nucleus pulposus and an annulus fibrosus.

Taylor and Wedel (2013: Figure 4). Intervertebral articular discs of an ostrich (not to scale). Left: first sacral vertebra in anterior view, showing articular disc of joint with the last thoracic vertebra. Right: posterior view view of a cervical vertebra, with probe inserted behind posterior articular disc. The cervical vertebra is most relevant to the present study, but the the sacral vertebra is also included as it shows the morphology more clearly. These fibrocartilaginous articular discs divide the synovial cavity, like the articular discs in the human temporomandibular and sternoclavicular joints, and should not be confused with the true intervertebral discs of mammals and other animals, which consist of a nucleus pulposus and an annulus fibrosus.

Crucially, the extant phylogenetic bracket (EPB) does not help us to establish the nature of the intervertebral articulations in sauropods, as the two extant groups most closely related to them have different articulations. As noted, birds have synovial joints; but crocodilians, like mammals, have fibrocartilaginous intervertebral discs. So their most recent common ancestor, the ur-archosaur, could equally have had either condition, and so could its various descendants.

vertebral-joint-type-cladogram

This seems like a mystery well worth solving. For one thing,  in the wholly inadequate database that we assembled for the paper, the birds had much thinner cartilage than the other animals. Since they are also the only animals with synovial neck joints, thin cartilage correlates with this kind of joint — at least across that tiny database. Is that correlation reliable? Does it hold out across a bigger sample? Is there a causation? If so, then finding out what kind of intervertebral joints sauropods had would help us to determine how thick their cartilage was, and so what their actual neutral posture was.

But we can’t tell this directly unless we find sensationally well preserved specimens that let us see the structure of the cartilage. We might speculate that since birds have unique saddle-shaped joints and sauropods have ball-and-socket joints like those of mammals and crocs, they’d be more likely to resemble the latter in this respect, too, but that’s rather hand-wavey.

Can we do better?

If we can, it will be through osteological correlates: that is, features of the bones (which are preserved in fossils) that are consistently correlated with features of the soft tissues (which are not). We’d want to find out from analysis of extant animals what correlates might exist, then go looking for them in the bones of extinct animals.

A couple of times now, I’ve pitched this as an abstract for a Masters project, hoping someone at Bristol will work on it with me as co-supervisor, but so far no-one’s bitten. Maybe next year. It would be a very specimen-based project, which I’d think would be a plus in most people’s eyes.

Figure 8. Cervical vertebra 7 from a turkey. Anterior view on the left; dorsal, left lateral and ventral views in the middle row; and posterior on the right.

Taylor and Wedel (2013: Figure 8). Cervical vertebra 7 from a turkey. Anterior view on the left; dorsal, left lateral and ventral views in the middle row; and posterior on the right.

Anyway, the awful truth is that at the moment we know spectacularly little about the cartilage in the necks of sauropods. We don’t know whether they had true intervertebral discs. If not, we don’t know whether they had articular discs like those of ostriches. We don’t know how thick these elements, if present, were. We don’t know how thick the hyaline cartilage on the bones’ articular surfaces was, or how evenly it covered its those surfaces.

And until we know those things, we don’t really know anything about neck posture or range of movement.

There’s lots of work to be done here!

Last time, we looked at how including intervertebral cartilage changes the neutral pose of a neck — or, more specifically, of the sequence of cervical vertebrae. The key finding (which is inexplicably missing from the actual paper, Taylor and Wedel 2013c) is that adding cartilage of thickness x between vertebrae whose zygapophyses are height y above the mid-height of the centra elevates the joint’s neutral posture by x/y radians.

Figure 14. Geometry of opisthocoelous intervertebral joints. Hypothetical models of the geometry of an opisthocoelous intervertebral joint compared with the actual morphology of the C5/C6 joint in Sauroposeidon OMNH 53062. A. Model in which the condyle and cotyle are concentric and the radial thickness of the intervertebral cartilage is constant. B. Model in which the condyle and cotyle have the same geometry, but the condyle is displaced posteriorly so the anteropos- terior thickness of the intervertebral cartilage is constant. C. the C5/C6 joint in Sauroposeidon in right lateral view, traced from the x-ray scout image (see Figure 12); dorsal is to the left. Except for one area in the ventral half of the cotyle, the anteroposterior separation between the C5 cotyle and C6 condyle is remarkably uniform. All of the arrows in part C are 52 mm long.

Figure 14. Geometry of opisthocoelous intervertebral joints. Hypothetical models of the geometry of an opisthocoelous intervertebral joint compared with the actual morphology of the C5/C6 joint in Sauroposeidon OMNH 53062. A. Model in which the condyle and cotyle are concentric and the radial thickness of the intervertebral cartilage is constant. B. Model in which the condyle and cotyle have the same geometry, but the condyle is displaced posteriorly so the anteroposterior thickness of the intervertebral cartilage is constant. C. the C5/C6 joint in Sauroposeidon in right lateral view, traced from the x-ray scout image (see Figure 12); dorsal is to the left. Except for one area in the ventral half of the cotyle, the anteroposterior separation between the C5 cotyle and C6 condyle is remarkably uniform. All of the arrows in part C are 52 mm long.

But how thick was the intervertebral cartilage in sauropods?

At the moment, no-one really knows. As Kent Stevens (2013) points out in his contribution to the PLOS ONE sauropod gigantism collection:

Determining the ONP of a sauropod’s cervical vertebral column given only its bones requires is necessarily speculative since the cartilage, and thus the intervertebral spacing, is unknown.

Part of the our goal in our own PLOS collection paper (Taylor and Wedel 2013c) was to take some very tentative first steps towards estimating the cartilage thickness. To do this, we used two approaches. First, we looked at CT scans of articulated vertebrae; and second, we measured the cartilage thickness in a selection of extant animals and thought about what we could extrapolate.

Since the CT scans were Matt’s domain, I’m going to pass over those for now, in the hope that he’ll blog about that part of the paper. Here, I want to look at the extant-animal survey.

Figure 18. Cartilage in the neck of a rhea. Joint between cervicals 11 (left) and 10 (right) of a rhea, sagittally bisected. Left half of neck in medial view. The thin layers of cartilage lining the C11 condyle and C10 cotyle are clearly visible.

Figure 18. Cartilage in the neck of a rhea. Joint between cervicals 11 (left) and 10 (right) of a rhea, sagittally bisected. Left half of neck in medial view. The thin layers of cartilage lining the C11 condyle and C10 cotyle are clearly visible.

The first thing to say is that our survey is inadequate in many ways. We worked with the specimens we could get hold of, in the state we had them. This means that:

  • we have a very arbitrary selection of different animals,
  • they are at different ontogenetic stages, and
  • their cartilage thickness was measured by a variety of methods.

Our goal was not at all to reach anything like a definitive answer, but just to get the question properly asked, and so hopefully to catalyse much a more detailed survey.

With that proviso out of the way, here are our main results (from Table 4 of the paper, though here I have removed the sauropod CT-scan rows since we’ll be writing about those separately).

Taxon Thickness Reference Notes
Turkey 4.56% This study Difference in measurements of intact neck and articulated sequence of cleaned, degreased and dried vertebrae.
Ostrich 6.30% Cobley et al. (2013) Difference in measurements of individual vertebrae with and without cartilage.
Rhea 2.59% This study Measurement of in situ cartilage in bisected neck.
Alligator 14.90% This study Measurement of in situ cartilage from photograph of cross section.
Horse 6.90% This study Measurement of in situ cartilage from photograph of cross section.
Camel 13.00% This study Crude measurement from condyle margin to cotyle lip of lateral-view X-ray. This is an interim measurement, which we hope to improve on when we obtain better images.
Dog 17.00% This study Measurement of intervertebral gaps in lateral-view X-ray, uncorrected for likely concavity of cotyles.
Giraffe 24.00% This study Difference in measurement of intact neck and closely articulated sequence of cleaned vertebrae. Young juvenile specimen.
Muraenosaurus 14.00% Evans (1993) Measurement of in situ cartilage in fossils.
Cryptoclidus 20.00% Evans (1993) Measurement of in situ cartilage in fossils.

We’ve expressed the measurements as a ratio between cartilage thickness and the length of the bone itself — that is, cartilage/bone. Another way to think of this is that the percentage is a correction factor which you need to add onto bone length to get whole-segment length. Note that this is not the same ratio as the proportion of total segment length that consists of cartilage: that would be (cartilage thickness + bone length) / bone length.

(We also tossed in some measurements of plesiosaur neck cartilage that Mark Evans made way back when. Get that thing properly published, Mark!)

Even this small survey throws up some interesting points.

First, there is a huge range of proportional cartilage thicknesses: almost an order of magnitude from the 2.59% of the Rhea up to the 24% of the juvenile giraffe — or, even if you discard that because of its ontogenetic stage, up to 17% for the dog. And note that the 17% for the dog is probably an under-estimate, since we were working from an X-ray that doesn’t show the concavity of the vertebral cotyles.

Figure 22. Dog neck in X-ray. Neck of a dog (dachsund), in X-ray, with the seven cervical vertebrae indicated. This photo has been used with permission from the Cuyahoga Falls Veterinary Clinic.

Figure 22. Dog neck in X-ray. Neck of a dog (dachsund), in X-ray, with the seven cervical vertebrae indicated. This photo has been used with permission from the Cuyahoga Falls Veterinary Clinic.

(Two reviewers expressed scepticism that this is the usual condition for dogs, but this X-ray is consistent with those of other dogs illustrated in the veterinary literature.)

The second thing to note is that the cartilage measurements for birds (average 4.5%) are are much lower than those of crocodilians (14.9%) or mammals (15.2%). What does this mean? Among these groups, sauropods are most closely related to birds; but birds and crocs form the extant phylogenetic bracket, so we can’t tell from phylogeny alone whether to expect them to more closely approach the avian or crocodilian condition. Furthermore, in being opisthocoelous (condyle in front, cotyle at the back) sauropod cervicals most closely resemble those of mammals in gross structure — and they have the thickest cartilage of all.

The third thing to note is that there is considerable variation within groups. Although the cartilage is proportionally thin for all three birds, it’s more than twice as thick in the ostrich as in the rhea (although some of this could be due to the different measurement methods used for these two birds). More interestingly, among mammals the cartilage is twice as thick in camels as in horses. In the horse, the condyles are deeply inserted into the cotyles of the preceding vertebrae; but in camels, they don’t reach even the lip of the cotyle. This should worry us, as horse and camel cervicals are grossly similar, and no osteological correlates have been identified that would allow us to determine from the bones alone how very different the cartilage is between these two mammals. So it seems possible that there were similarly dramatic differences in the neck-cartilage thickness of different sauropods.

Note: I said that no osteological correlates have been identified. That doesn’t mean they don’t exist. One thing I would love to see is a serious attempt to analyse cartilage thickness across a broad range of mammals, and to examine the corresponding dry bones to see whether in fact there are correlates that could be informative in this respect. One lesson that Matt and I have learned over and over again is that there’s often plenty of data in places that are out in the open, but where no-one’s thought to look.

Next time: more on searching for osteological correlates of cartilage. Then, measurements of sauropod-neck cartilage from CT scans, and likely implications for cartilage thickness in life.

References

As I mentioned a few days ago, Matt and I have a couple of papers in the new PLOS ONE Sauropod Gigantism collection. We were each lead author on one and second author on the other, so for convenience’s sake we’ll refer to them as my paper (Taylor and Wedel 2013c on neck cartilage) and Matt’s paper (Wedel and Taylor 2013b on caudal pneumaticity.)

Mine is very simple in concept (although it ended up at 17 pages and 23 figures). It’s all about addressing one of the overlooked variables in reconstructing the postures of the necks of sauropods (and indeed of all tetrapods). That is, the spacing between consecutive vertebrae, and the effect this has on “neutral pose”.

The concept of “neutral pose” goes back to the DinoMorph work of Stevens and Parrish (1999). They defined it (p. 799) as follows: “We determined the neutral poses for each animal, wherein the paired articular facets of the postzygapophyses of each cervical vertebra were centered over the facets of the prezygapophyses of its caudally adjacent counterpart.”

x

Taylor and Wedel (2013c: Figure 3). Articulated sauropod vertebrae. Representative mid-cervical vertebra of Giraffatitan brancai, articulating with its neighbours. The condyle (ball) on the front of each vertebra’s centrum fits into the cotyle (socket) at the back of the preceding one, and the prezygapophyses articulate with the preceding vertebra’s postzygapophyses. These vertebrae are in Osteological Neutral Pose, because the pre- and postzygapophyseal facets overlap fully.

One of the more fundamental flaws in Stevens and Parrish (1999) is the assumption that animals habitually rest their necks in neutral pose — an assumption that is unsupported by evidence and, as it turns out, false (Vidal et al. 1986, Taylor et al. 2009). But let’s leave that aside for the moment, and consider what neutral pose actually represents.

The fact that there is even such a thing as neutral articulation between two consecutive vertebrae is due to there being three points of contact between those vertebra: as with the legs of a tripod, three points is the minimum number you need to fix an object in three-dimensional space. Two of these points are at the zygapophyses, as noted in the original definition above. The third point is the articulation between the centra.

The centrum has been curiously overlooked in discussions of neutral pose, but needless to say its length is crucial in establishing what is neutral. In the image above, if the centrum was longer, then the angle between the consecutive vertebrae would need to be raised in order to keep the zygapophyses articulated.

And of course it was longer in life, because of the cartilage in between the consecutive centra. (The use of the more specific term “osteological neutral pose” goes some way to recognising that tissues other than bone have been overlooked, but the problem has not really been addressed or even properly acknowledged in published works before our paper.)

xx

Taylor and Wedel (2013c: Figure 5). Intervertebral gaps in camel necks. Head and neck of dromedary camels. Top: UMZC H.14191, in right lateral view, posed well below habitual posture, with apparently disarticulated C3/C4 and C4/C5 joints. Photograph taken of a public exhibit at University Museum of Zoology, Cambridge, UK. Bottom: OUMNH 17427, in left lateral view, reversed for consistency with Cambridge specimen. Photograph taken of a public exhibit at Oxford University Museum of Natural History, UK. Inset: detail of C4 of the Oxford specimen, showing articulations with C3 and C5. The centra are separated by thick pads of artificial ‘‘cartilage’’ to preserve spacing as in life.

You simply can’t ignore cartilage when modelling neck postures and expect to get anything resembling a meaningful result. That is, presumably, the reason why the habitual posture of rabbits in life exceeds the most extended posture we were able to obtain when manipulating dry vertebrae of a hare: compare Vidal et al. (1986: fig. 4) with Taylor et al. (2009: fig. 1).

How big is the effect? That depends on the thickness of the cartilage and the height of the zygapophyses above the center of rotation. Here is an illustration that we should have put in the paper, but which inexplicably neither of us thought of:

figNEW-angle-at-zygs

Influence of intervertebral cartilage on vertebral articulation angle. Consider the posterior vertebra (black) as fixed. The blue vertebra represents neutral pose of the preceding vertebra with centra abutting and zygapophyseal facets maximally overlapped. The red vertebra indicates neutral pose once intervertebral cartilage is added between the vertebra (where else?) The green lines show the angle by which the more anterior vertebra must be inclined in order to accommodate the cartilage, and the magenta line shows the height of the zygapophyseal articulation above the center of rotation between the two vertebrae.

Here’s some elementary trigonometry. Suppose the intervertebral cartilage is x distance thick at mid-height of the centra, and that the height of the zygs above this mid-height point (the magenta line) is y. The triangle between the middle of the condyle of the posterior vertebra, the middle of the cotyle of the anterior one and the zygapophyseal articulation is near enough a right-angled triangle as makes no odds.

Consider the angle θ between the green lines. Sin(θ) = opposite/hypotenuse = x/y, and by similarity, the additional angle of inclination of the anterior vertebra is also θ.

But for small angles (and this is generally a small angle), sin(θ) ≈ θ. So the additional inclination in radians = cartilage thickness divided by zygapophyseal height. For example, in vertebrae where the zygs are 23 cm above the mid-height of the centra, adding 4 cm of intervertebral cartilage adds about 4/23 = 0.174 radians = 10 degrees of extra inclination. (That’s pretty similar to the angle in the illustration above. Eyeballing the cartilage thickness and zyg height in the illustration suggests that 23:4 ratio is about right, which is a nice sanity-check of this method.)

millionaire-stupid-contestant4

At this point, I am cursing my own stupidity for not putting this diagram, and the very simple calculation, into the paper. I guess that can happen when something is written in a hurry (which to be honest this paper was). The formula is so simple — and accurate enough within tolerances of inevitable measurement error — that we really should have used it all over the place. I guess that will have to go in a followup now.

Anyway — next time, we’ll address this important related question: how thick, in fact, was the cartilage between the cervicals of sauropods?

References

This is an exciting day: the new PLOS Collection on sauropod gigantism is published to coincide with the start of this year’s SVP meeting! Like all PLOS papers, the contents are free to the world: free to read and to re-use.  (What is a Collection? It’s like an edited volume, but free online instead of printed on paper.)

There are fourteen papers in the new Collection, encompassing neck posture (yay!), nutrition (finally putting to bed the Nourishing Vomit Of Eucamerotus hypothesis), locomotion, physiology and evolutionary ecology. Lots every sauropod-lover to enjoy.

x

Taylor and Wedel (2013c: Figure 12). CT slices from fifth cervical vertebrae of Sauroposeidon. X-ray scout image and three posterior-view CT slices through the C5/C6 intervertebral joint in Sauroposeidon OMNH 53062. In the bottom half of figure, structures from C6 are traced in red and those from C5 are traced in blue. Note that the condyle of C6 is centered in the cotyle of C5 and that the right zygapophyses are in articulation.

Matt and I are particularly excited that we have two papers in this collection: Taylor and Wedel (2013c) on intervertebral cartilage in necks, and Wedel and Taylor (2013b) on pneumaticity in the tails of (particularly) Giraffatitan and Apatosaurus. So we have both ends of the animal covered. It also represents a long-overdue notch on our bed-post: for all our pro-PLOS rhetoric, this is the first time either of has had a paper published in a PLOS journal.

Wedel and Taylor (2013b: Figure 4). Giraffatitan brancai tail MB.R.5000 (‘Fund no’) in right lateral view. Dark blue vertebrae have pneumatic fossae on both sides, light blue vertebrae have pneumatic fossae only on the right side, and white vertebrae have no pneumatic fossae on either side. The first caudal vertebra (hatched) was not recovered and is reconstructed in plaster.

It’s a bit of a statistical anomaly that after a decade of collaboration in which there was never a Taylor & Wedel or Wedel & Taylor paper, suddenly we have five of them out in a single year (including the Barosaurus preprint, which we expect to eventually wind up as Taylor and Wedel 2014). Sorry about the alphabet soup.

Since Matt is away at SVP this week, I’ll be blogging mostly about the Taylor and Wedel paper this week. When Matt returns to civilian life, the stage should be clear for him to blog about pneumatic caudals.

Happy days!

References

In a comment on the previous post, Dean asked: “What was the difference in length between the neck with its cartilage and the bones flush together?”

I’m glad you asked me that.  You’ll recall from last time that the fully fleshed neck — intact apart from the removal of the skin and maybe some superficial muscle — was 51 cm in length from the front of the atlas to the back of the centrum of the seventh cervical vertebra.  When I pose the cleaned and cartilage-free bones together, the total length of the series is only 41 cm — 10 cm shorter, coming in at just over 80% of the live length.  Don’t believe me?  Here are the photos!

I’m sure I need hardly say, but the top image is the neck as we got it, the second is the cleaned bones posed in more or less the arrangement they must have been in life (both of these taken from the previous post) and the bottom image the bones fully abutting.

So!  The neck of Wallace the baby giraffe was very nearly a quarter as long again as the bones alone suggest.  Does this mean that the neck of Giraffatitan was really 10.6 m long instead of 8.5 m?

It’s an exciting prospect, but I’m afraid the answer is no.  As I hinted last time, while it’s perfectly acceptable, indeed obligatory, to recognise the important role of cartilage in sauropod necks qualitatively, we can’t blindly apply the numbers from Wallace the baby giraffe to adult sauropods for two reasons: 1, Wallace is a baby; and 2, Wallace is a giraffe.

The first of these reasons is part of why I am keen to do this all over again with an adult giraffe when I get the opportunity; but there’s not much we can do about the second.  One might think that a more closely related extant animal such as an ostrich might have a neck that is more homologous with those of sauropods; and that’s true, but my feeling is that the giraffe is more analogous.  That is, although the birds share more recent common ancestry with sauropods, giraffes’ more similar size seem to have encouraged them to evolve cervicals that are in some ways more similar to those of sauropods, most notably in the possession of ball-and-socket intervertebral joints rather than the saddle-shaped joints that are ubiquitous in birds.

How big a deal is Wallace’s juvenile status?  Well, take a look at his fifth cervical vertebra in posterior view:

If this bone were found in 150 million years by competent palaeontologists, in a world where there were no extant artiodactyls to compare with, what would they make of it?  Most of the articular area of the centrum is very obviously damaged, exposing the internal spongy texture of cancellous bone — presumably the bone surface was attached more firmly to the cartilaginous posterior end of the element than to the inner part of the bone, so it came away with the cartilage during simmering.  So it would be obvious to our future palaeontologists that the articular surface was missing, and that the complete vertebra would have been somewhat longer — but it would be hard to judge by how much.

But the state of this bone is particularly interesting because the middle part of the centrum does have a preserved bone surface.  It would be easy to extrapolate that out across the whole area of the posterior end of the centrum, and assume that this was the maximum posterior extent of the element’s functional length in life — an assumption that we know, having taking the neck apart ourselves, would be completely wrong.

Are we making similar incorrect assumptions with our sauropod vertebrae?

An even more interesting case is the postzygapophyses.  The posterodorsal surface of the left postzyg is slightly damaged, but the bone of the right postzyg has a nice, perfectly preserved surface.  But I can tell you that the functional articular surface of the postzyg was totally different from this: different size, different shape, different position, different orientation.  If we tried to calculate range of movement from these zygapophyseal facets, the results we got would be literally meaningless.

The good news is, there’s a clue that would prevent us from making this mistake — a really nice, obvious one.  The texture of the bone on the postzyg is irregularly crenellated in a way that strongly indicates a cartilaginous extension: it’s the same texture you see on the ends of the long-bones of (even mature) birds if you peel off the cartilage caps.  (It’s also what you see, at a much bigger scale, on the ends of the sauropod long-bones.)

But while the presence of this texture indicates the presence of cartilage, I don’t know whether the converse is true.  In the absence of such a texture, can we assume the absence of cartilage?  I just don’t know.  Anyone?

Isn’t it funny how often an idea seems to pop up all over the place at about the same time?  The classic example is the independent and more or less simultaneous invention of calculus by both Isaac Newton and Wilhelm Leibniz, but similar kinds of things seem to happen quite often.

And there’s something similar going on right now.  After a century of everyone ignoring the role of cartilage in dinosaur anatomy, suddenly everyone’s up and running all at once:

  • Here at SV-POW!, Matt, Darren and I have been running the series on camel necks (which by the way isn’t over yet — stay tuned!)  In that series we have repeatedly made the point that “it is useless to try to reach conclusions about neck posture based on osteology alone. We need to understand the soft-tissue systems — especially the articular cartilage — as well”.
  • Meanwhile, over on his blog Jurassic Journeys, Matt Bonnan has been writing about “long bones and the space between“, emphasising how we can’t really understand sauropod locomotion when we don’t know the true sizes and shapes that the long-bones had in life.
  • Independently of that, Heinrich Mallison, on the Palaeontologia Electronica blog, wrote about the importance of cartilage in his Plateosaurus digital modelling projects.  I highly recommend reading this very relevant article if only for its section headings, which sum up the state of play perfectly: Ask your doctor for advice // Palaeontology is an interdisciplinary science — we just tend to forget // Have you ever read the Journal of International Orthopaedics? // How do these go together? Where’s the manual? HELP!
  • The next thing we know, Casey Holliday and his colleagues wrote about the same issue — not merely blogging, but producing a long-awaited peer-reviewed article in PLoS ONE, “Cartilaginous Epiphyses in Extant Archosaurs and Their Implications for Reconstructing Limb Function in Dinosaurs“.  Casey and his group have gone much further than the rest of us: rather than just whining about the problem of cartilage, they’ve taken steps to solve it — see below for details.
  • Finally, it turns out that Dave Hone has had a blog entry on this subject in the works at Archosaur Musings for a year or more.

It’s a pretty amazing confluence of thought, and the Holliday et al. paper really couldn’t have come at a better time.  It gives us, for the first time, qualitative estimates of the thickness of articular cartilage in limb-bones.  They dissected birds and alligators, measured their limb bones before and after the removal of their cartilage caps, compared the measurements, and determined what they called cartilage correction factors (CCFs) that quantify the increase in limb length when cartilage is included.  They also examined the osteological correlates of extensive articular cartilage, and drew conclusions about the likely form and function of these structures in sauropods (and, yes, I suppose, other dinosaurs as well).

This all ties in nicely with a long-running background project of mine, first presented at Progressive Palaeontology in 2005, and then again at the German sauropod-fest in 2008.  While Holliday et al. were investigating the thickness of articular cartilage, I was thinking in a very naive way about its area as part of a study tentatively entitled Upper limits on the mass of land animals estimated through the articular area of limb-bone cartilage.  The slides for the talk are available, and contain a Godzilla joke that will be hauntingly familiar to anyone who saw my talk on neck elongation at SVPCA this year.

Poorly executed slide from my 2005 Progressive Palaeontology talk. Despite the clumsy graphics, the point should be clear: that the area of articular cartilage available to withstand static and locomotory forces depends hugely on how extensive the cartilage caps are, and on their shape.

I ought to be clear that my work on this was very preliminary and that I am, as usual, years behind where I wanted to be in terms of getting this written up rigorously.  In fact the talk ended with a slide in which I pointed that I was pretty confident that “my figures are correct within a factor of 756″.  And I stand by that :-)

My point is just this: suddenly there’s a visible swell of palaeontologists all saying the same thing: that we can’t expect to understand how the skeletons of extinct animals worked by looking only at their bones, which is a bit of a shame when their bones are usually all we have.  The Holliday et al. paper (2010) is a very welcome first step towards wrasslin’ with this problem as it deserves.

Oh, and it’s open access — go read it!

References

Follow

Get every new post delivered to your Inbox.

Join 381 other followers