Back in 2013, when we were in the last stages of preparing our paper Caudal pneumaticity and pneumatic hiatuses in the sauropod dinosaurs Giraffatitan and Apatosaurus (Wedel and Taylor 2013b), I noticed that, purely by chance, all ten of the illustrations shared much the same limited colour palette: pale brows and blues (and of course black and white). I’ve always found this strangely appealing. Here’s a composite:

wedel-taylor-2013b-all-figures

I’m really happy with this coincidence. In fact I think I might get it printed up as a poster for my office.

(Thought: if I did, would anyone else be interested in buying it?)

Update (a couple of hours later)

At Matt’s suggestion, I switched the order of figures 7 and 8 (the last two on the third row) to get the following version of the image. It break the canonical order of the figures, but it’s visually more pleasing.

wedel-taylor-2013b-all-figures-v2

Now we should write an updated version of the paper that reverses the order in which we refer to figures 7 and 8 :-)

References

  • Wedel, Mathew J., and Michael P. Taylor. 2013. Caudal pneumaticity and pneumatic hiatuses in the sauropod dinosaurs Giraffatitan and Apatosaurus. PLOS ONE 8(10):e78213. 14 pages. doi:10.1371/journal.pone.0078213

photo1

I was at the Natural History Museum of Los Angeles County yesterday to do some research in the ornithology collection. After lunch I was working on this pelican skeleton and I thought, “Geez, there is just no way to do this thing justice with still photos. I should make a video.” Here it is. You’ll want to see it full-screen–this being my first time out making a video, I didn’t realize that I was holding the phone the wrong way for efficient viewing on other devices.

The specimen is LACM Ornithology 86262. I’m posting this video with the knowledge and kind permission of the ornithology collection staff.

For previous things in this vein, please see:

If you like it that stuff like this exists, please support your local natural history museum, especially the LACM, which has some really fantastic education and outreach programs.

This came out two months ago, and I should have blogged about it then, but as usual I am behind. I’m blogging about it now because it deals with a question that has been on my mind for about 10 years now. If you want to skip my blatherations and get on to the good stuff, here’s the paper (Martin and Palmer 2014).

An Unsolved Problem

Back in 2004 I realized that if one had CTs or other cross-sections of a pneumatic bone, it was possible to quantify how much of the cross-sectional space was bone, and how much was air, a ratio I called the Air Space Proportion (ASP). That was the subject of my 2004 SVP talk, and a big part–arguably the most important part–of my chapter in The Sauropods in 2005. Of course the same calculation works for marrow-filled bones as well, where you would refer to it as an MSP rather than an ASP. If you can quantify the areas of bone, air, and marrow, you can figure out how dense the element was. One-stop shopping for all the relevant (simple) math is in this post.

(From Wedel 2005)

(From Wedel 2005)

Sometimes in science you end up with data that you don’t know what to do with, and that was my situation in 2004. Since I had CTs and other cross-sectional images of sauropod vertebrae, I could calculate ASPs for them, but I didn’t know what those results meant, because I didn’t have anything to compare them to. But I knew where to get I could get comparative data: from limb bone cross-sections. John Currey and R. McNeill Alexander had published a paper in 1985 titled, “The thickness of the walls of tubular bones”. I knew about that paper because I’d become something of an R. McNeill Alexander junkie after reading his book, Dynamics of Dinosaurs and Other Extinct Giants (Alexander 1989). And I knew that it had data on the cross-sectional properties of the limb bones in a host of animals, including crocs, birds, mammals, and–prophetically–pterosaurs.

If you know the inner and outer radii of a tubular bone, it is trivial to convert that to an ASP. So I could take the data from Currey and Alexander (1985) and calculate ASPs for the pneumatic bird and pterosaur bones in their study. Cubo and Casinos (2000) had a much larger sample of bird limb bones, and those got fed into my 2005 paper as well.

I was alert to the possibility that a mid-shaft cross-section might not be representative of the whole bone, and I hedged a bit in describing the bird ASPs (Wedel 2005: p. 212):

For the avian long bones described above, data were only presented for a single cross sec- tion located at midshaft. Therefore, the ASP values I am about to discuss may not be representative of the entire bones, but they probably approximate the volumes (total and air) of the diaphyses. For tubular bones, ASP may be determined by squaring K (if r is the inner diameter and R the outer, then K is r/R, ASP is πr^2/πR^2 or simply r^2/R^2, and ASP = K^2). For the K of pneumatic bones, Currey and Alexander (1985) report lower and upper bounds of 0.69 and 0.86, and I calculate a mean of 0.80 from the data presented in their table 1. Using a larger sample size, Cubo and Casinos (2000) found a slightly lower mean K of 0.77. The equivalent values of ASP are 0.48 and 0.74, with a mean of 0.64, or 0.59 for the mean of Cubo and Casinos (2000). This means that, on average, the diaphysis of a pneumatic avian long bone is 59%–64% air, by volume.

Now, even though I hedged and talked about diaphyses (shafts of long bones) rather than whole bones, I honestly expected that the ASP of any given slice would not change much along the length of a bone. Long bones tend to be tubular near the middle, with a thick bony cortex surrounding the marrow or air space, and honeycombed near the ends, with much thinner cortices and lots of bony septa or trabeculae (for marrow-filled bones, this is called spongy or trabecular bone, and for air-filled bones it is best referred to as camellate pneumatic bone). I figured that the decrease in cortical bone thickness near the ends of the bone would be offset by the increase in internal bony septa, and that the bone-to-air ratio through the whole element would be under some kind of holistic control that would keep it about even between the middle of the bone and the ends.

It is fair to ask why I didn’t just go check. The answer is that research is to some extent a zero-sum game, in that every project you take on means another that gets left waiting in the wings or abandoned completely. I was mainly interested in what ASP had to say about sauropods, not birds, and I had other fish to fry.

So that’s me from 2004-2012: aware that mid-shaft cross-sections of bird and pterosaur long bones might not be representative of whole elements, but not sufficiently motivated to go check. Then at SVPCA in Oxford that fall, Liz Martin rocked my world.

journal.pone.0097159.g001

Figure 1. CT scan images from two different regions of pterosaur first wing phalanx. A and B show the unmodified CT scans from A) the distal end of UP WP1 and B) the mid-shaft of UP WP1, while C and D show the modified and corrected images used in the calculation. Air space proportion (ASP) is calculated by determining the cross-sectional area of the internal, air filled cavity (the black centre of D) and dividing that by the total cross-sectional area, including the white cortical tissue and the black cavity. In areas with trabeculae, like C, the calculation of the air space includes the air found in individual trabeculae around the edges. Scale = 10 mm. doi:10.1371/journal.pone.0097159.g001 (From Martin and Palmer 2014)

A Paper in the Can

At SVPCA 2012, Liz Martin gave a talk titled, “A novel approach to estimating pterosaur bone mass using CT scans”, the result of her MS research with Colin Palmer at the University of Bristol. In that talk–the paper for which has been submitted to JVP–Liz and Colin were interested in using CT scans of pterosaur bones to quantify the volume of bone, in order to refine pterosaur mass estimates. I was fully on board, since estimating the masses of extinct animals is a minor obsession of mine. But what really caught my attention is that Liz and Colin had full stacks of slices spanning the length of each element–and therefore everything they needed to see how or if ASPs of pterosaur wing bones changed along their lengths.

At the next available break I dashed up to Liz, opened up my notebook, and started scribbling and gesticulating and in general carrying on like a crazy person. It’s a wonder she didn’t flee in terror. The substance of my raving was that (1) there was this outstanding problem in the nascent field of ASP research, and (2) she had everything she needed to address it, all that was required was a little math using the data she already had (I say this as if running the analyses and writing the paper were trivial tasks–they weren’t). Fortunately Liz and Colin were sufficiently interested to pursue it. Their paper on ASPs of pterosaur wing bones was submitted to PLOS ONE this February, and published on May 9 (while their earlier paper continues to grind its way through JVP).

And I’m blogging about it because the results were not what I expected.

Pterosaur wing bone ASPs - Martin and Palmer 2014

Figure 2. Plot of air space proportion over the length in six pterosaur wing bones. These plots show a polynomial line fit for each bone to show the general shape distribution. Exact measurements can be seen in Table S1. (From Martin and Palmer 2014).

Here’s the graph that tells the tale. Each line traces the ASP per slice along the length of a single pterosaur wing bone. A few things jump out:

  • Almost all of the lines drop near the left end. This is expected–if you’re cutting slices of a bone and measuring the not-bone space inside, then as you approach the end of the bone, you’re cutting through progressively more bone and less space. A few of the lines also drop near the right. I’m puzzled by that–if my explanation is correct, the ASP should plunge about equally at both ends. And the humerus USNM 11925 doesn’t follow the same pattern as the rest. As Martin and Palmer write, “It is unknown if this is a general feature of humeri, or this single taxon and more investigation is needed.”
  • Almost all of the bones have MUCH lower ASPs at mid-shaft than near the ends, on the order of 10% or more. So mid-shaft cross-sections of pterosaur wing bones tend to significantly underestimate how pneumatic they were. It would be interesting to know if the same holds true for bird long bones, or for the vertebrae of pterosaurs, birds, and sauropods. As Martin and Palmer point out, more work is needed.
  • The variation in ASP along the length of a single bone is in some cases greater than the variation between elements and individuals. That’s pretty cool. On the happy side, it means that getting into the nitty-gritty of ASP is not just stamp-collecting; you really need to know what is going on along the length of a bone before you can say anything intelligent about ASP or the density of the element. On the less happy side, that’s going to be a righteous pain in the butt for sauropod workers, because vertebrae are tough to get good scans of, assuming they will fit through a CT scanner at all (most don’t).
  • Finally, pterosaurs turn out to be even more pneumatic than you would think from looking at the already-freakishly-thin-walled shafts of their long bones. That’s pretty awesome, and it dovetails nicely with the emerging picture that pneumaticity in ornithodirans was more prevalent and more interesting than even I had suspected–it’s in prosauropods (Yates et al. 2012) and brachiosaur tails (Wedel and Taylor 2013) and rebbachisaur hips (Fanti et al. 2013) and saltasaur shoulders (Cerda et al. 2012) and, er, a couple of places that I can’t mention just yet. So life is good.

A few last odds and ends:

You can read more of this story at Liz Martin’s blog, scattered over several recent posts.

If you have CTs of bones and you want to follow in the footsteps of Martin and Palmer, you can do a lot of the work, and maybe all of it, in BoneJ, a free plug-in for ImageJ, which is also free.

A final note: this is Liz Martin’s first published paper, so congratulations are in order. Well done, Liz!

Almost Immediate Update: As soon as I posted this, I sent the link to Liz to see if I’d missed anything important. She writes, “It may be worth mentioning that it’s a question that I am actively following up on in my PhD, and looking into it with birds too hopefully. And it is indeed all possible using ImageJ, as that’s how I did the whole thing!”

References

OLYMPUS DIGITAL CAMERA

Now considered a junior synonym of Supersaurus, on very solid grounds.

Incidentally, unlike the neural spines of most non-titanosaurian sauropods, the neural spine of this vertebra is not simply a set of intersecting plates of bone. It is hollow and has a central chamber, presumably pneumatic. Evidence:

OLYMPUS DIGITAL CAMERA

You know the drill: lotsa pretty pix, not much yap.

IMG_5024

Our first stop of the day was the Fruita Paleontological Area, which has a fanstastic diversity of Morrison animals, including the mammal Fruitafossor and the tiny ornithopod Fruitadens.

IMG_5027

Plus it’s a pretty epic landscape, especially with the clouds and broken light we had this morning.

IMG_5039

I found a bone! Several bits, actually, a few meters away from the Fruitadens type quarry. I’d like to think that this proximal femur might be Fruitadens, but I don’t know the diagnostic characters and haven’t had time to look them up. Anyone know how diagnostic this honorary shard of excellence might be?

IMG_5052

After lunch, John Foster took us on a short hike to the quarry where Elmer Riggs got the back half of the Field Museum Apatosaurus. The front half came from a site in southern Utah, several decades later.

IMG_5056

The locals brought Riggs out in the 1930s for the dedication of two monuments–this one at the Apatosaurus quarry, and another like it at the Brachiosaurus quarry some miles away. Tragically, both monuments have the names of the dinosaurs misspelled!

IMG_5088

In the afternoon we visited the Mygatt-Moore Quarry and the Camarasaurus site in Rabbit Valley. Can you see the articulated Camarasaurus neck in this photo?

IMG_5076

Here’s a hint: the neural arches of two posterior cervical vertebrae in transverse horizontal cross-section.

IMG_5092

This Camarasaurus is apparently a permanent feature. If you’re wondering why no-one has excavated it, it’s because it’s buried in sandstone that is stupid-dense. The expenditure of time and resources just isn’t worth it, when right down the hill dinosaurs are pouring out of the much softer sediments of the Mygatt-Moore Quarry like water from a hydrant. This is the lesson I am learning about the Morrison: finding dinosaurs is easy. Finding dinosaurs you can get out of the ground and prepare–that’s something else.

IMG_5094

Our last stop of the day was Gaston Design, where Rob Gaston showed us how he molds, casts, and mounts everything from tiny teeth to good-sized skeletons.

IMG_5110

Like this Deinosuchus that is about to chomp on Jim Kirkland. Jim doesn’t look too worried.

IMG_5117

Here’s a nice cast of a busted sauropod dorsal, probably from Apatosaurus or Diplodocus, showing the pneumatic internal structure. Compare to similar views of dorsals in this post and this one. This is actually one half of a matched set that includes both halves of the centrum. I left with one of those sets of my own, a few dollars poorer and a whole lot happier.

IMG_5120

The end–for now.

Illustration talk slide 47

Illustration talk slide 48

Illustration talk slide 49

Illustration talk slide 50

That last one really hurts. Here’s the original image, which should have gone in the paper with the interpretive trace next to it rather than on top of it:

Sauroposeidon C6-C7 scout

The rest of the series.

Papers referenced in these slides:

“Look at all the things you’ve done for me
Opened up my eyes,
Taught me how to see,
Notice every tree.”

So sings Dot in Move On, the climactic number of Stephen Sondheim’s Pulitzer Prize-winning music Sunday in the Park with George, which on the surface is about the post-impressionist painter Georges Seurat, but turns out to be a study of obsession and creativity.

xx

Un dimanche après-midi à l’Île de la Grande Jatte – 1884 [A Sunday Afternoon on the Island of La Grande Jatte – 1884]

“Taught me how to see”? What kind of talk is that? One the surface, it seems silly — we all know how to see. We do it constantly, without thinking. Yet it’s something that artists talk about all the time. And anyone who’s sat down and seriously tried to paint or draw something will have some understanding of what the phrase means. We have such strong implicit ideas of what things look like that we tend to reproduce what we “know” is there rather than what’s actually there. Like I said, we see without thinking.

In fact, the psychology of perception is complicated and sophisticated, and the brain does an extraordinary amount of filtering of the visual signals we get, to save us the bother of having to consciously process way too much data. This is a whole scientific field of its own, and I’m going to avoid saying very much about it for fear of making a fool of myself — as scientists so often do when wandering outside their own field. But I think it’s fair to say that we all have a tendency to see what we expect to see.

xx

Phylogeny of Sauropoda, strict consensus of most parsimonious trees according to Wilson (2002:fig. 13a)

In the case of sauropods, this tendency has meant that we’ve all been startlingly bad at seeing pneumaticity in the caudal vertebrae of sauropods. Because the literature has trained us to assume it’s not there. For example, in the two competing sauropod phylogenies that dominated the 2000s, both Wilson (2002) and Upchurch et al. (2004) scored caudal pneumaticity as very rare: Wilson’s character 119, “Anterior caudal centra, pneumatopores (pleurocoels)”, was scored 1 only for Diplodocus and Barosaurus; and  Upchurch et al. (2004:286) wrote that “A few taxa (Barosaurus, Diplodocus, and Neuquensaurus) have pleurocoel-like openings in the lateral surfaces of the cranial [caudal] centra that lead into complex internal chambers”. That’s all.

And that’s part of the reason that every year since World War II, a million people have walked right past the awesome mounted brachiosaur in the Museum Für Naturkunde Berlin without noticing that it has pneumatic caudals. After all, we all knew that brachiosaur caudals were apneumatic.

But in my 2005 Progressive Palaeontology talk about upper limits on the mass of land animals estimated through the articular area of limb-bone cartilage, I included this slide that shows how much bigger the acetabulum of Giraffatitan is than the femoral head that it houses:

Screenshot from 2014-01-24 17:30:30

And looking at that picture made me wonder: those dark areas on the sides of the first few caudals (other than the first, which is a very obvious plaster model) certainly look pneumatic.

Then a few years later, I was invited to give a talk at the Museum Für Naturkunde Berlin itself, on the subject “Brachiosaurus brancai is not Brachiosaurus“. (This of course was drawn from the work that became my subsequent paper on that subject, Taylor 2009) And as I was going through my photos to prepare the slides of that talk, I thought to myself: darn it, yes, it does have pneumatic caudals!

So I threw this slide into the talk, just in passing:

Screenshot from 2014-01-24 17:32:06

Those photos were pretty persuasive; and a closer examination of the specimen on that same trip was to prove conclusive.

Meanwhile …

Earlier in 2009, I’d been in Providence, Rhode Island, with my Index Data colleagues. I’d managed to carve a day out of the schedule to hope along the coast to the Yale Peabody Museum in New Haven, Connecticut. My main goal was to examine the cervicals of the mounted Apatosaurus (= “Brontosaurus“) excelsus holotype (although it was also on that same trip that I first saw the Barosaurus holotype material that we’ve subsequently published a preprint on).

The Brontosaurus cervicals turned out to be useless, being completely encased in plaster “improvements” so that you can’t tell what’s real and what’s not. hopefully one day they’ll get the funding they want to take that baby down off its scaffold and re-prep the material.

But since I had the privilege of spending quality time with such an iconic specimen, it would have been churlish not to look at the rest of it. And lo and behold, what did I see when I looked at the tail but more pneumaticity that we thought we knew wasn’t there!

Wedel and Taylor (2013b: Figure 10).

An isolated pneumatic fossa is present on the right side of caudal vertebra 13 in Apatosaurus excelsus holotype YPM 1980. The front of the vertebra and the fossa are reconstructed, but enough of the original fossil is visible to show that the feature is genuine. (Wedel and Taylor 2013b: Figure 10).

What does this mean? Do other Giraffatitan and Apatosaurus specimens have pneumatic tails? How pervasive is the pneumaticity? What are the palaeobiological implications?

Stay tuned! All will be revealed in Matt’s next post (or, if you can’t wait, in our recent PLOS ONE paper, Wedel and Taylor 2013b)!

References

Follow

Get every new post delivered to your Inbox.

Join 421 other followers