At the top: our old friend BYU 9024 — the cervical vertebra that’s part of the Supersaurus vivianae holotype. At the bottom, C2 (the longest cervical) of Giraffa camelopardalis angolensis FMNH 34426.

The Supersaurus vertebra is 138 cm long. We don’t know which cervical it is, but there’s no reason to think it’s the longest. The giraffe vertebra is 31 cm long. Not only is the Supersaurus vertebra four times as long as that of the giraffe, it’s one of more than twice as many cervicals as the giraffe has.

Did we cheat by using an unusually small giraffe? Not really. When we articulated all seven cervicals as best we could, the sequence measured 171 cm, which is a fairly healthy 71% of the 2.4 m neck of the world-record giraffe. It’s not a monster, but it’s a decent-sized adult.

Bottom line, giraffes are just lame.

Recapture Creek comparo with measurements

If you’re just joining us, this post is a follow-up to this one, in which I considered the possible size and identity of the Recapture Creek femur fragment, which “Dinosaur Jim” Jensen (1987: page 604) said was “the largest bone I have ever seen”.

True to his word, Brooks Britt at BYU got back to me with measurements of the Recapture Creek femur fragment in practically no time at all:

Length 1035 mm, width 665 mm.  However, you cannot trust the measurements because Jensen put a lot of plaster on the proximal half of the bone.

Now, taking plaster off a bone is not going to make it any larger. So the plastered-up specimen is the best case scenario for the RC femur to represent a gigapod. And I know the stated width of 665 mm is the max width of the proximal end, because I sent Brooks a diagram showing the measurements I was requesting. The length is a little less than anticipated, and doesn’t quite jibe with the max proximal width–I suspect a little might have broken off from the distal end where the preservation looks not-so-hot.

Based on those measurements, it looks like Jensen got the scale bar in Figure 8 in his 1987 paper approximately right–if anything, the scale bar is a little undersized, but only by 5% or so, which is actually pretty good as these things go (scale bars without measurements are still dag-nasty evil, though). By overlapping Jensen’s photo with the femur of the Brachiosaurus altithorax holotype (FMNH P25107) to estimate the size of the element when complete, I get a total length of 2.2 meters–exactly the same size as about 8% bigger than the Brachiosaurus holotype (actual length 2.03 m). If the Recapture Creek femur is from a Camarasaurus, which I don’t think we can rule out, it was 2 meters long when complete, or 11% longer and 37% more massive than the big C. supremus AMNH 5761–about 35 tonnes or maybe 40 on the outside. So it’s a big bone to be sure, but it doesn’t extend only slightly extends the known size range of Morrison sauropods. (Updated 2014-05-19–as I related in the first post, I somehow got it fixed in my head that the holotype B.a. femur was 2.19 m when it is actually 2.03 m.)

So, as before, caveat estimator when working from scaled illustrations of single partial bones of possibly immense sauropods.

Now, here’s a weird thing. Let’s assume for the sake of this discussion that the Recapture Creek femur is from a brachiosaur. That gives us three individual Late Jurassic brachiosaurids–the Recapture Creek animal, the Brachiosaurus altithorax holotype, and the mounted Giraffatitan brancai–that are almost exactly the same size in limb bone dimensions (although B.a. had a longer torso). But we know that brachiosaurids got bigger, as evidenced by the XV2 specimen of Giraffatitan, and based on the lack of scapulocoracoid fusion in both FMNH P25107 and the mounted Giraffatitan. So why do we keep finding these (and smaller) subadults, and so few that were XV2-sized? I know that there gets to be a preservation bias against immense animals (it’s hard to bury a 50-tonne animal all in one go), but I would not think the 13% linear difference between these subadults and XV2-class adults would be enough to matter. Your thoughts?


Jensen, J.A. 1987. New brachiosaur material from the Late Jurassic of Utah and Colorado. Great Basin Naturalist 47(4): 592-608.

From Jensen (1987, page 604):

“In 1985 I found the proximal third of an extremely large sauropod femur (Figs. 8A, 12A) in a uranium miner’s front yard in southern Utah.  The head of this femur is 1.67 m (5’6″) in circumference and was collected from the Recapture Creek Member of the the Morrison Formation in Utah near the Arizona border.  It is the largest bone I have ever seen.”

Jensen included not one but two figures of this immense shard of excellence. Here they are:

Jensen 1987 figure 8

Jensen 1987, Figure 8

Jensen 1987, Figure 12

Jensen 1987, Figure 12

The specimen was heavily reconstructed, as you can see from the big wodge of unusually smooth and light-colored material in the photo. So we can’t put much stock in that part of the specimen.

Unfortunately, the only measurement of the specimen that Jensen gives in the paper is that circumference; there are no straight-line linear measurements, and the figures both have the dreaded scale bars. Why dreaded? Check this out:

Recapture Creek figs 8 and 12 comparedAs you can see, when the scale bars are set to the same size, the bones are way off (the scale bar in the drawing is 50 cm). This is not an uncommon problem. I make the Fig 8 version 30% bigger in max mediolateral width of the entire proximal end, and still 17% bigger in minimum diameter across the femoral head, as measured from the slight notch on the dorsal surface (on the right in this view).

Can we figure out which is more accurate based  on the internal evidence of the paper? For starters, the Fig 12 version is a drawing (1), that does not match the outline from the photo (2), and the hand-drawn scale bar (3) does not actually coincide with any landmarks (4), and that’s plenty of reasons for me not to trust it.

What about that circumference Jensen mentioned? Unfortunately, he didn’t say exactly where he took it, just that the head of the femur had a circumference of 1.67 meters. Is that for the entire proximal end, or for the anatomical head that fits in the acetabulum, er wot? I’m afraid the one measurement given in the paper is no help in determining which of the figures is more accurately scaled.

The obvious thing to do would be to see if this bone is in the BYU collections, and just measure the damn thing. More on that at the end of the post.

In the meantime, Jensen said that the shape of the Recapture Creek femur was most similar to the femur of Alamosaurus, or to that of Brachiosaurus among Morrison taxa, and he referred it to Brachiosauridae. So how does this thing–in either version–compare with the complete femur of FMNH P25107, the holotype of Brachiosaurus altithorax?

The Recapture Creek brachiosaur femur fragment compared to the complete femur of the Brachiosaurus altithorax holotype FMNH P25107

The Recapture Creek femur fragment compared to the complete femur of the Brachiosaurus altithorax holotype FMNH P25107

The first thing to notice is that the drawn outline from Figure 12 is a much better match for the Brachiosaurus altithorax femur–enough so that I wonder if Jensen drew it from the Recapture Creek specimen, or just traced the B.a. proximal femur and scaled it accordingly (or maybe not accordingly, since the scale bars don’t match).

But let’s get down to business: how long would the complete femur have been?

Using the scale bar in the photograph from Figure 8 (on the left in above image), I get a total femur length of 2.36 meters. Which is long, but only 7.7% longer than the 2.19-meter femur of FMNH P25107, and therefore only 25% more massive. So, 35 tonnes to Mike’s 28-tonne B.a., or maybe 45 tonnes to a more liberal 36-tonne B.a. Big, yeah, but not world-shattering. Update 2014-05-19: I don’t know where I got the 2.19-meter femur length for Brachiosaurus altithorax, but it’s a mistake. So the rest of that paragraph should read: Which is 16% longer than 2.03-meter femur of FMNH P25107, and therefore 57% more massive. So, 44 tonnes to Mike’s 28-tonne B.a., or maybe 57 tonnes to a more liberal 36-tonne B.a. That’s nowhere near the 2.5-meter femur and estimated 70-tonne mass of the largest Argentinosaurus, but it’s pretty darned good for a brachiosaur.

Using the scale bar in the drawing from Figure 12 (on the right in the above image)–which, remember, is 50 cm, not 1 meter–I get a total femur length of about 1.9 meters, which is considerably smaller than the B.a. holotype. That is very much at odds with Jensen’s description of it as “the largest bone I have ever seen”, and given that we have many reasons for not trusting the scale bar in the drawing, it is tempting to just throw it out as erroneous.

So it would seem that unless Jensen got both scale bars too big, the Recapture Creek brachiosaur was at most only a shade bigger than the holotype specimen of Brachiosaurus altithorax.

But wait–is the Recapture Creek brachiosaur a brachiosaur at all? Jensen didn’t list any characters that pushed him toward a brachiosaurid ID, and I don’t know of any proximal femur characters preserved in the specimen that would separate Brachiosaurus from, say, Camarasaurus. And in fact a camarasaur ID has a lot to recommend it, in that Camarasaurus femora have very offset heads (the ball- or cylinder-like articular surface at the top end sticks out a big more to engage with the hip socket–see Figure 12 up near the top of the post), moreso than in many other Morrison sauropods, and that would make them better matches for the Recapture Creek femur photo. Here’s what the comparo looks like:

Recapture Creek - Camarasaurus comparo

The Recapture Creek femur fragment compared with a complete femur of Camarasaurus.

I make that a 2.07-meter femur using the photo on the left, and a 1.66-meter femur using the drawing on the right. The one decent femur in the AMNH 5761 Camarasaurus supremus collection is 1.8 meters long, so these results are surprisingly similar to those for the B. althithorax comparison–the drawing gives a femur length shorter than the largest known specimens, and the photo gives a length only slightly longer. A camarasaur with a 2.07 meter femur would be 15% larger than the AMNH C. supremus in linear terms, and  assuming isometric scaling, 1.5 times as massive–maybe 38 tonnes to AMNH 5761’s estimated 25. A big sauropod to be sure, but not as big as the largest apatosaurs, and not nearly as big as the largest titanosaurs.

I have always been surprised that the Recapture Creek femur frag has attracted so little attention, given that “Dinosaur Jim” himself called it the biggest bone he had ever seen. But it appears that the lack of attention is justified–whether it was a brachiosaur or a camarasaur, and using the most liberal estimates the scale bars allow, it simply wasn’t that big.

Update about half an hour later: Okay, maybe I was a little harsh here. IF the photo scale bar is right, the Recapture Creek femur might still represent the largest and most massive macronarian from the Morrison Formation (Edit: only if it’s a brachiosaur and not a camarasaur; see this comment), which is something. I suppose I was particularly underwhelmed because I was expecting something up in OMNH 1670-to-Argentinosaurus territory, and so far, this ain’t it. I’ll be interested to see what the actual measurements say (read on).

The Moral of This Story

So, if it wasn’t that big after all, and if no-one has made a stink about it being big before now, why go to all this trouble? Well, mostly just to satisfy my own curiosity. If there was a truly gigantic brachiosaur from the Morrison, it would be relevant to my interests, and it was past time I crunched the numbers to find out.

But along the way something occurred to me: this should be a cautionary tale for anyone who gets all wound up about the possible max size of Amphicoelias fragillimus. As with A. fragillimus, for the Recapture Creek critter we have part of one bone, and at least for this exercise I was working only from published illustrations with scale bars. And as with A. fragillimus, the choice of a reference taxon is not obvious, and the size estimates are all over the place, and some of them just aren’t that big.

It always amuses me when A. fragillimus comes up and people (well, trolls) accuse us of being big ole’ wet blankets that just don’t want to believe in 200-tonne sauropods. It amuses me because it’s wrong on so many levels. Believe me, when we have our sauropod fanboy hats on, we most definitely do want to believe in 200-tonne sauropods. That would rock. But when we put our scientist hats on, wanting and belief go right out the window. We have to take a cold, hard look at the data, and especially at its limitations.

Oh, the other moral is to go buy a tape measure, and use it. Sheesh!


As I said above, the obvious thing to do would be to just track down the bone and measure it. It does still exist, it’s in the BYU collections, and Brooks Britt has kindly offered to send along some measurements when he gets time. So we should have some real answers before long (and here they are). But I wanted to work through this example without them, to illustrate how much uncertainty creeps in when trying to estimate the size of a big sauropod from published images of a single partial bone.


Jensen, J.A. 1987. What I did on my holidaysNew brachiosaur material from the Late Jurassic of Utah and Colorado. Great Basin Naturalist 47(4): 592-608.

Why giraffes have short necks

September 26, 2012

Today sees the publication, on arXiv (more on that choice in a separate post), of Mike and Matt’s new paper on sauropod neck anatomy. In this paper, we try to figure out why it is that sauropods evolved necks six times longer than that of the world-record giraffe — as shown in Figure 3 from the paper (with a small version of Figure 1 included as a cameo to the same scale):

Figure 3. Necks of long-necked sauropods, to the same scale. Diplodocus, modified from elements in Hatcher (1901, plate 3), represents a “typical” long-necked sauropod, familiar from many mounted skeletons in museums. Puertasaurus modified from Wedel (2007a, figure 4-1). Sauroposeidon scaled from Brachiosaurus artwork by Dmitry Bogdanov, via (CC-BY-SA). Mamenchisaurus modified from Young and Zhao (1972, figure 4). Supersaurus scaled from Diplodocus, as above. Alternating pink and blue bars are one meter in width. Inset shows Figure 1 to the same scale.

This paper started life as a late-night discussion over a couple of beers, while Matt was over in England for SVPCA back in (I think) 2008. It was originally going to be a short note in PaleoBios, just noting some of the oddities of sauropod cervical architecture — such as the way that cervical ribs, ventral to the centra, elongate posteriorly but their dorsal counterparts the epipophyses do not.

As so often, the tale grew in the telling, so that a paper we’d initially imagined as a two-or-three-page note became Chapter 5 of my dissertation under the sober title of “Vertebral morphology and the evolution of long necks in sauropod dinosaurs”, weighing in at 41 1.5-spaced pages. By now the manuscript had metastatised into a comparison between the necks of sauropods and other animals and an analysis of the factors that enabled sauropods to achieve so much more than mammals, birds, other theropods and pterosaurs.

(At this point we had one of our less satisfactory reviewing experiences. We sent the manuscript to a respected journal, where it wasn’t even sent out to reviewers until more than a month had passed. We then had to repeatedly prod the editor before anything else happened. Eventually, two reviews came back: one of them careful and detailed; but the other, which we’d waited five months for, dismissed our 53-page manuscript in 108 words. So two words per page, or about 2/3 of a word per day of review time. But let’s not dwell on that.)

Figure 6. Basic cervical vertebral architecture in archosaurs, in posterior and lateral views. 1, seventh cervical vertebra of a turkey, Meleagris gallopavo Linnaeus, 1758, traced from photographs by MPT. 2, fifth cervical vertebra of the abelisaurid theropod Majungasaurus crenatissimus Depéret, 1896,UA 8678, traced from O’Connor (2007, figures 8 and 20). In these taxa, the epipophyses and cervical ribs are aligned with the expected vectors of muscular forces. The epipophyses are both larger and taller than the neural spine, as expected based on their mechanical importance. The posterior surface of the neurapophysis is covered by a large rugosity, which is interpreted as an interspinous ligament scar like that of birds (O’Connor, 2007). Because this scar covers the entire posterior surface of the neurapophysis, it leaves little room for muscle attachments to the spine. 3, fifth cervical vertebra of Alligator mississippiensis Daudin, 1801, MCZ 81457, traced from 3D scans by Leon Claessens, courtesy of MCZ. Epipophyses are absent. 4, eighth cervical vertebra of Giraffatitan brancai (Janensch, 1914) paralectotype HMN SII, traced from Janensch (1950, figures 43 and 46). Abbreviations: cr, cervical rib; e, epipophysis; ns, neural spine; poz, postzygapophysis.

This work made its next appearance as my talk at SVPCA 2010 in Cambridge, under the title Why giraffes have such short necks. For the talk, I radically restructured the material into a form that had a stronger narrative — a process that involved a lot of back and forth with Matt, dry-running the talk, and workshopping the order in which ideas were presented. The talk seemed to go down well, and we liked the new structure so much more than the old that we reworked the manuscript into a form that more closely resembled the talk.

That’s the version of the manuscript that we perfected in New York when we should have been at all-you-can-eat sushi places. It’s the version that we submitted on the train from New York to New Haven as we went to visit the collections of the Yale Peabody Museum. And it’s the version that was cursorily rejected from mid-to-low ranked palaeo journal because a reviewer said “The manuscript reads as a long “story” instead of a scientific manuscript” — which was of course precisely what we’d intended.

Needless to say, it was deeply disheartening to have had what we were convinced was a good paper rejected twice from journals, at a cost of three years’ delay, on the basis of these reviews. One option would have been to put the manuscript back into the conventional “scientific paper” straitjacket for the second journal’s benefit. But no. We were not going to invest more work to make the paper less good. We decided to keep it in its current, more readable, form and to find a journal that likes it on that basis.

At the moment, the plan is to send it to PeerJ when that opens to submissions. (Both Matt and I are already members.) But that three-years-and-rolling delay really rankles, and we both felt that it wasn’t serving science to keep the paper locked up until it finally makes it into a journal — hence the deposition in arXiv which we plan to talk about more next time.

Table 3. Neck-elongation features by taxon.

In the paper, we review seven characteristics of sauropod anatomy that facilitated the evolution of long necks: absolutely large body size; quadrupedal stance; proportionally small, light head; large number of  cervical vertebrae; elongation of cervical vertebrae; air-sac system; and vertebral pneumaticity. And we show that giraffes have only two of these seven features. (Ostriches do the next best, with five, but they are defeated by their feeble absolute size.)

The paper incorporates some material from SV-POW! posts, including Sauropods were corn-on-the-cob, not shish kebabs. In fact, come to think of it, we should have cited that post as a source. Oh well. We do cite one SV-POW! post: Darren’s Invading the postzyg, which at the time of writing is the only published-in-any-sense source for pneumaticity invading cervical postzygapogyses from the medial surface.

As for the non-extended epipophyses that kicked the whole project off: we did illustrate how they could look, and discussed why they would seem to make mechanical sense:

Figure 10. Real and speculative muscle attachments in sauropod cervical vertebrae. 1, the second through seventeenth cervical vertebrae of Euhelopus zdanskyi Wiman, 1929 cotype specimen PMU R233a-δ(“Exemplar a”). 2, cervical 14 as it actually exists, with prominent but very short epipophyses and long cervical ribs. 3, cervical 14 as it would appear with short cervical ribs. The long ventral neck muscles would have to attach close to the centrum. 4, speculative version of cervical 14 with the epipophyses extended posteriorly as long bony processes. Such processes would allow the bulk of both the dorsal and ventral neck muscles to be located more posteriorly in the neck, but they are not present in any known sauropod or other non-avian dinosaur. Modified from Wiman (1929, plate 3).

But we found and explained some good reasons why this apparently appealing arrangement would not work. You’ll need to read the paper for details.

Sadly, we were not able to include this slide from the talk illustrating the consequences:

Anyway, go and read the paper! It’s freely available, of course, like all arXiv depositions, and in particular uses the permissive Creative Commons Attribution (CC BY) licence. We have assembled related information over on this page, including full-resolution versions of all the figures.

In the fields of maths, physics and computer science, where deposition in arXiv is ubiquitous, standard practice is to go right ahead and cite works in arXiv as soon as they’re available, rather than waiting for them to appear in journals. We will be happy for the same to happen with our paper: if it contains information that’s of value to you, then feel free to cite the arXiv version.


  • Taylor, Michael P., and Mathew J. Wedel. 2012. Why sauropods had long necks; and why giraffes have short necks. arXiv:1209.5439. 39 pages, 11 figures, 3 tables. [Full-resolution figures]

In the recent post on OMNH 1670, a dorsal vertebra of a giant Apatosaurus from the Oklahoma panhandle, I half-promised to post the only published figure of this vertebra, from Stovall (1938: fig. 3.3). So here it is:

And in the second comment on that post, I promised a sketch from one of my notebooks, showing how much of the vertebra is reconstructed. Here’s a scan of the relevant page from my notebook. Reconstructed areas of the vert are shaded (confusingly, using strokes going in opposite directions on the spine and centrum, and the dark shaded areas on the front of the transverse processes are pneumatic cavities), and measurements are given in mm.

Next item: is this really a fifth dorsal vertebra?

Apatosaurus louisae CM 3018 D4 and D5, in anterior (top), left lateral, and posterior views, from Gilmore (1936: plate 25).

Here are D4 and D5 of A. louisae CM 3018. They sort of bracket OMNH 1670 in terms of morphology. D4 has a broader spine, and D5 has a narrower one. The spine of D5 lacks the slight racquet-shaped expansion seen in OMNH 1670, but the overall proportions of the spine are more similar. On the other hand, the transverse processes of D4 taper a bit in anterior and posterior view, as in OMNH 1670, and unlike the transverse processes of D5 with their more parallel dorsal and ventral margins. But honestly, neither of these verts is a very good match (and the ones on either side, D3 and D6, are even worse).

Apatosaurus parvus UWGM 15556 (formerly A. excelsus CM 563) D4 (left) and D3 (right) in anterior (top), right lateral, and posterior views, from Gilmore (1936: plate 32).

Here are D3 and D4 of A. parvus UWGM 15556. D3 is clearly a poor match as well–it is really striking how much the vertebral morphology changes through the anterior dorsals in most sauropods, and Apatosaurus is no exception. D3 looks like a dorsal in lateral view, but in anterior or posterior view it could almost pass for a posterior cervical. If I was going to use the term “cervicodorsal”, indicating one of the vertebrae from the neck/trunk transition, I would apply it as far back as D3, but not to D4. That thing is all dorsal.

And it’s a very interesting dorsal from the perspective of identifying OMNH 1670. It has fairly short, tapering transverse processes. The neural spine is a bit shorter and broader, but it has a similar racquet-shaped distal expansion. I’m particularly intrigued by the pneumatic fossae inscribed into the anterior surface of the neural spine–in Gilmore’s plate they make a broken V shapen, like so \ / (or maybe devil eyes). Now, OMNH 1670 doesn’t have devil eyes on its spine, but it does have a couple of somewhat similar pneumatic fossae cut into the spine just below the distal racquet–perhaps a serially modified iteration of the same pair of fossae as in the A. parvus D4. It’s a right sod that D5 from this animal has its spine blown off–but it still has its transverse processes, and they are short and tapering as in OMNH 1670.

Apatosaurus sp. FMNH P25112, dorsal vertebrae 1-10 and sacrals 1 and 2, Riggs (1903: plate 46)

Here are all the dorsals and the first couple sacrals of FMNH P25112, which was originally described as A. excelsus but in the specimen-level analysis of Upchurch et al. 2005) comes out as the sister taxon to the A. ajax/A. parvus/A. excelsus clade. Note the striking similarity of the D5 here with D4 of the A. parvus specimen in Gilmore’s plate (until the careful phylogenetic work up Upchurch et al. 2005, that A. parvus specimen, once CM 563 and now UWGM 15556, was considered to represent A. excelsus as well). But  also notice the striking similarity of D6 to OMNH 1670. It’s not quite a dead ringer–the transverse processes are longer and have weird bent-down “wingtips” (XB-70 Valkyrie, anyone?)–but it’s pretty darned close, especially in the shape of the neural spine.

So what does this all mean? First, that trying to specify the exact serial position of an isolated vertebra is nigh on to impossible, unless it’s something that is one-of-a-kind like an axis. Second, after doing all these comparos I think it’s unlikely that OMNH 1670 is a D4–those are a bit too squat across the board–but it could plausibly be either a D5 or a D6. Third, I’m really happy that it doesn’t seem to match any particular specimen better than all the rest. What I don’t want to happen is for someone to see that this vertebra looks especially like specimen X and therefore decide that it must represent species Y. As I said in the comments of the previous post, what this Oklahoma Apatosaurus material needs is for someone to spend some quality time seeing, measuring, and photographing all of it and then doing a phylogenetic analysis. That sounds like an ambitious master’s thesis or the core of a dissertation, and I hope an OU grad student takes it on someday.

If you were intrigued by my suggestion that the big Oklahoma Apatosaurus rivalled Supersaurus in size, and wanted to see a technical comparison of the two, I am happy to report that Scott Hartman has done the work for you. Here’s one of his beautiful Apatosaurus skeletal reconstructions, scaled to the size of OMNH 1670, next to his Supersaurus silhouette. This is just a small teaser–go check out his post on the subject for a larger version and some interesting (and funny) thoughts on how the two animals compare.


  • Gilmore, C.W. 1936. Osteology of Apatosaurus with special reference to specimens in the Carnegie Museum. Memoirs of the Carnegie Museum 11:175-300.
  • Riggs, E.S. 1903. Structure and relationships of opisthocoelian dinosaurs, part I: Apatosaurus Marsh. Field Columbian Museum Publications, Geological Series 2(4): 165–196.
  • Stovall, J.W. 1938. The Morrison of Oklahoma and its dinosaurs. Journal of Geology 46:583-600.

Left: the Queen of England, 163 cm.  Middle, the Oklahoma apatosaur dorsal, 135 cm.  Right, classic “big Apatosaurus” dorsal, 106 cm.  To scale.

Something I’ve always intended to do but never gotten around to is posting on some of the immense Apatosaurus elements from the Oklahoma panhandle. Here’s one of the most impressive, OMNH 1670, an isolated dorsal. Notice that the tip of the neural spine is ever-so-shallowly bifurcated, which in Apatosaurus indicates a D4, D5, or D6. The low parapophyses and fat transverse processes are similar to D4, but Apatosaurus D4s usually have somewhat broader spines, so I’m guessing this thing is a D5. These things vary and I could easily be off by a position in either direction.

Next to it is D5 of CM 3018, the holotype specimen of Apatosaurus louisae (from Gilmore 1936: plate 25), which has served as the basis for many of the published mass estimates of the genus Apatosaurus. OMNH 1670 is 135 cm tall, compared to 106 cm for D5 of CM 3018. If the rest of the animal scaled the same way, it would have been 1.27^3 = 2 times as massive. Mass estimates for CM 3018 are all over the map, from about 18 tons up to roughly twice that, so the big Oklahoma Apatosaurus was probably in Supersaurus territory, mass-wise, and may have rivaled some of the big titanosaurs (Update: see the two giant diplodocids square off in a cool follow-up post by Supersaurus wrangler Scott Hartman). Here’s a fun rainy-day activity: take any skeletal reconstruction of Apatosaurus, clone it in Photoshop or GIMP, scale it up by 27%, and park it next to the original. It looks a lot bigger. So I’m continually surprised that Apatosaurus is so rarely mentioned in the various roundups of giant sauropods, both in the technical literature and in popular articles online. This vertebra was figured by Stovall (1938)–if I get inspired, I’ll dig up that figure and post it another day (hey, look, I did!).

Fun fact: in Apatosaurus the tallest (most posterior) dorsals are 1.3-1.5 times as tall as D5 (Gilmore 1936: 201). So D10 from this individual was probably between 1.7 and 2 meters tall–not quite in Amphicoelias fragillimus territory but getting closer than I’ll bet most people suspected.

NB: if you try to use the scale bar lying on the centrum of OMNH 1670 to check my numbers, you will get a wonky answer. The problem is that the vertebra is so large that it is almost impossible to get far enough back from it (above it, in this case, since it is lying on a padded pallet) to get a shot free from distortion due to parallax. For this shot, the pallet with the vert was on the floor, and I was standing on top of the tallest ladder in the OMNH collections, leaning out over the vert to get centered over the prezygapophyses, and shooting straight down–in other words, I had done everything possible to minimize the visual distortion. But it still crept in. Anyway, trust the measurements, which I–and presumably Gilmore–made with a good old reliable tape measure.


  • Gilmore, C.W. 1936. Osteology of Apatosaurus with special reference to specimens in the Carnegie Museum. Memoirs of the Carnegie Museum 11:175-300.
  • Stovall, J.W. 1938. The Morrison of Oklahoma and its dinosaurs. Journal of Geology 46:583-600.

Vanessa Graff and I spent yesterday working in the herpetology and ornithology collections at the Natural History Museum of Los Angeles County (LACM). The herpetology collections manager, Neftali Comacho, pointed us to this skull of Alligator mississippiensis. It’s not world’s biggest gator–about which more in a second–but it’s the biggest I’ve seen in person. Normally it lives in a big rubbermaid tub in the collections area, but this Sunday it will be out on display for Reptile and Amphibian Appreciation Day (RAAD) at the LACM. RAAD will include guest talks, tours of the collections, and live animal demonstrations. If you’re in SoCal and you’re into herps–or have kids, grandkids, nephews or nieces that are into herps–it will be well worth checking out. While you’re there, don’t neglect the newly renovated Age of Dinosaurs and Age of Mammals halls, which are frankly phenomenal: spacious, well-lit, loads of actual material on display, skeletons you can walk all the way around, informative but unobtrusive signage, tasteful integration with existing architecture…I could go on. Better if you just go and see for yourself.

About that gator. First the bad news.  It came to the LACM from another collection, and has no data–no locality, no date collected, nothing. The skull is also missing all of its teeth, the left retroarticular process, the back end of the braincase and the occipital condyle. I think the latter losses were probably caused by a foramen of Winchester.*

Now, the awesome news. The length from the snout tip to the end of the articulars was 680mm and from the snout to the end of the quadrates was 590mm. Irritatingly I did not get a dorsal head length, which is the gold standard for comparative croc skull measurements, because I only reread Darren’s giant croc skull post after I got home last night. Going from the photos, I think the dorsal head length was right around 50 cm (beware, the yardstick in the photos is marked off in inches).

Darren’s post led me to this one, which has some very useful measurements (yay!) of giant croc skulls. The table at the end of that post lists alligator skulls with dorsal head lengths of 58, 60, and 64 cm, so the big LACM gator is nowhere near being the world’s largest. In fact, the 64 cm skull would be a quarter again as large, which is a truly horrifying thought. Still, it’s a big damn skull from a big damn gator.

You might get the impression that here in the Wedel lab we are shamelessly obsessed with giant saurians. And that is in fact true. But we also look at tiny ones, too. Here I’m playing with the skull of a little Tomistoma, the false gharial. Tomistoma is notable because another individual of the genus produced the longest skull of any known extant crocodilian–a whopping 84 cm dorsal head length (photos of this monster are in both of the giant croc skull posts linked above).

The moral of the story? If the sign says don’t go swimming, don’t go swimming. Go to RAAD instead, and see the giant alligator skull, and a ton of other cool stuff besides. And if you’re into gator skulls or just like geeking out on awesome anatomy, check out the 3D Alligator Skull site, a joint project of the Holliday lab and Witmer lab. Have fun!

* bullet hole

In a new comment on an oldish post, Peter Adlam asked:

I recently happened upon a picture of the late Jim Jenson standing beside the huge front leg of “Ultrasauros”, which leads me to ask a few questions. Did he really find a complete forelimb? Was the leg from Brachiosaurus altithorax? If that leg is valid at actual size how tall/long was the whole animal? It looks to be about 40% to 50% taller than the berlin Giraffatitan, I am guessing the leg is a constructed representation of how the leg would look rather than a cast of the actual leg because if the whole front leg was found they would probably be the most talked about sauropod bones in the world and the fact is I’ve heard pretty much nothing about these remains for years.

I answered this in a followup comment, but because the answer involved a few nice images, I thought it ought to be promoted into a post of its own.  So here it is, in expanded form.

I believe I know the picture Peter was talking about: it was either the one on the right, of Jensen working on the limb in the lab, or the one below of the same limb, again with Jensen, this time out in the desert.

As an aside: based on a post by ReBecca Hunt-Foster (scroll down to the 12th picture), it looks like this forelimb may have ended up in the New Mexico Museum of Natural History and Stuff (NMMNHS).

Anyway, the bad news is that, no, this is not a complete forelimb fossil. The worse news is that the limb is not even partly cast from real material: it’s a pure sculpture, based presumably on the forelimb of Giraffatitan brancai, but scaled up according to Jensen’s idea of how big “Ultrasauros” was. The only part of the model that probably was cast from real material is not part of the limb proper, but the scapulocoracoid — which is the only real brachiosaur element that Jensen found and described from the Dry Mesa quarry.  In fact, the scap in these photos (and in ReBecca’s) does look very much like BYU 9462, the element that Jensen meant to designate as the “Ultrasauros” holotype, but didn’t, instead plumping for … ah, you all know the story.

"Ultrasauros" scapulocoracoid BYU 9462 (almost certainly a cast), with Graeme Elliott for scale.

But in fact, the scapulocoracoid in the whole-forelimb pictures above looks much too small in comparison with the other elements; or to put it the right way round, since only the scap is based on an actual fossil, all the other elements are too big — which suggests that Jensen exaggerated the sizes of the sculpted limb bones well beyond what the scapulocoracoid warranted.  (In any case, the idea that this scap represents a much larger brachiosaurid than any previously known specimen was shown by Curtice et al. (1996) to be mistaken — it’s from an animal pretty similar in size to, and probably a little smaller than, the largest known Tendaguru specimens.)

But the good news is, Peter’s sense of awe is not misplaced. Real brachiosaur forelimbs are actually not much less impressive than this. See for example me next to the right forelimb of the Berlin Giraffatitan mount, which is real bone — as shown in our classic post Shedloads Of Awesome:

Or here I am again, this time with the Chicago Brachiosaurus mount. (The Chicago mount is a cast, based on a hybrid of real Brachiosaurus elements, some bits of Giraffatitan, and some sculptures, but the scaling is good.)

My rule of thumb, based on a lot of posing for photos around the Chicago mount, is that if I stand next to the forelimb and reach up, I can just rest my hand on the top of the ulna without stretching.  I’m about six feet tall, if that helps.

Jim Jensen was 4% taller than me — 6’3″.  Bearing that in mind, and looking at the second photograph above (the first one is useless because of the forced perspective), Jensen’s inability to reach close to the top of the ulna suggests that his model is inflated by maybe 30%.  Which means that it represents an animal about 1.3^3 = 2.2 times as voluminous and heavy as it should be.  But let’s not forget that among the Giraffatitan material in Berlin is the isolated fibula XV 2, which at 134 cm in length is 12.6% longer than the 119 cm tibia of S II.  So that is from animal about half way between S II and Jensen’s Imaginary Monster in size.

So.  Real brachiosaurs are awesome enough.

I have a new paper out:

Wedel, M.J. 2012. A monument of inefficiency: the presumed course of the recurrent laryngeal nerve in sauropod dinosaurs. Acta Palaeontologica Polonica 57(2):251-256.

Update June 6, 2012: the final version was formally published yesterday, so the rest of this paragraph is of historical interest only. Like Yates et al. on prosauropod pneumaticity, it is “out” in the sense that the manuscript has been through peer review, has been accepted for publication, and is freely available online at Acta Palaeontologica Polonica. Technically it is “in press” and not published yet, but all that formal publication will change is to make a prettier version of the paper available. All of the content is available now, and the paper doesn’t include any of those pesky nomenclatural acts, and so, as with the prosauropod pneumaticity paper, I don’t see any reason to pretend it doesn’t exist. Think of the accepted manuscript as the caterpillar to the published version’s butterfly: different look, but same genome.

This one came about because last summer I read a review of Richard Dawkins’s book, The Greatest Show on Earth: The Evidence for Evolution. The review mentioned that the book includes a lengthy discussion of the recurrent laryngeal nerve (RLN) in the giraffe, which is a spectacularly dumb piece of engineering and therefore great evidence against intelligent design creationism. It wasn’t the first time I’d heard of the RLN, of course. It’s one of the touchstones of both human anatomy and evolutionary biology; anatomy because of its clinical importance in thyroid surgery, known for more than two millennia, and evolutionary biology because it is such a great example of a developmental constraint. (Dawkins’s coverage of all of this is great, BTW, and you should read the book.)

No, the reason the book review inspired me to write the paper was not because the RLN was new to me, but because it was overly familiar. It is a cool piece of anatomy, and its fame is justly deserved–but I am sick and tired of seeing the stinkin’ giraffe trotted out as the ultimate example of dumb design. My beloved sauropods were way dumber, and it’s time they got some credit.

But first, let’s talk about that nerve, and how it got to be there.

No necks for sex? How about no necks for anybody!

Embryos are weird. When you were just a month old (counting from fertilization), you had a set of pharyngeal arches that didn’t look radically different from those of a primitive fish. These started out quite small, tucked up underneath your comparatively immense brain, and each pharyngeal arch was served by a loop of artery called an aortic arch. What we call the arch of the aorta in an adult human is a remnant of just one of these embryonic aortic arches, and as you’ve no doubt noticed, it’s down in your chest, not tucked up next to your brain. When you were in the embryonic stages I’m talking about, you didn’t yet have a neck, so your brain, pharyngeal arches, aortic arches, and the upper parts of your digestive system were all smooshed together at your front end.

One thing you did have at that stage was a reasonably complete peripheral nervous system. The nerve cell bodies in and near your central nervous system sent out axons into the rest of your body, including your extremities. Many of these axons did not persist; they failed to find innervation targets and their parent neurons died. Imagine your embryonic central nervous system sending out a starburst of axons in all directions, and some of those axons finding targets and persisting, and others failing and dying back. So the architecture of your nervous system is the result of a process of selection in which only some cells were successful.

Crucially, this radiation and die-off of axons happened very early in development, when a lot of what would become your guts was still hanging under your proportionally immense brain like the gondola on a blimp. This brings us to the recurrent laryngeal nerve.

Going back the way we came

The fates of your embryonic pharyngeal arches are complex and I’m not going to do a comprehensive review here (go here for more information). Suffice it to say that the first three arches give rise to your jaws and hyoid apparatus, the fourth and sixth form your larynx (voicebox), and fifth is entirely resorbed during development. Update: I made a pharyngeal arch cheat sheet.

There are two major nerves to the larynx, each of which is bilaterally paired. The nerve of the fourth pharyngeal arch becomes the superior laryngeal nerve, and it passes cranial to the fourth aortic arch. The nerve of the sixth pharyngeal arch becomes the inferior or recurrent laryngeal nerve, and it passes caudal to the sixth aortic arch. At the time that they form, both of these nerves take essentially straight courses from the brainstem to their targets, because you’re still in the blimp-gondola stage.

If you were a shark, the story would be over. The more posterior pharyngeal arches would persist as arches instead of forming a larynx, each arch would hold on to its artery, and the nerves would all maintain their direct courses to their targets.

The normal fate of the aortic arches in humans. From

But you’re not a shark, you’re a tetrapod. Which means that you have, among other things, a neck separating your head and your body. And the formation of your neck shoved your heart and its associated great vessels down into your chest, away from the pharyngeal arches. This was no problem for the superior laryngeal nerve, which passed in front of the fourth aortic arch and could therefore stay put. But the inferior laryngeal nerve passed behind the sixth aortic arch, so when the heart and the fourth and sixth aortic arches descended into the chest, the inferior laryngeal nerve went with them. Because it was still hooked up to the brainstem and the larynx, it had to grow in length to compensate.

As you sit reading this, your inferior laryngeal nerves run down your neck into your chest, loop around the vessels derived from the fourth and sixth aortic arches (the subclavian artery on the right, and the arch of the aorta and ductus arteriosus on the left) and run back up your neck to your larynx. Because they do this U-turn in your chest and go back the way they came, the inferior laryngeal nerves are said to ‘recur’ to the larynx and are therefore more commonly referred to as the recurrent laryngeal nerves (RLNs).

An enlightening diversion

The RLN is the poster child for “unintelligent design” because it is pretty dumb. Your RLNs travel a heck of a lot farther to reach your larynx than they ought to, if they’d been designed. Surely an intelligent designer would have them take the same direct course as the superior laryngeal nerve. But evolution didn’t have that option. Tetrapod embryos could not be built from the ground up but had to be modified from the existing “sharkitecture” of ancestral vertebrates. The rules of development could not be rewritten to accommodate a shorter RLN. Hence Dawkins’s love affair with the RLN, which gets 7 pages in The Greatest Show on Earth. He also appeared on the giraffe episode of Inside Nature’s Giants, in which the RLN was dug out of the neck and the continuity of its ridiculous path was demonstrated–probably the most smack-you-in-the-face evidence for evolution that has ever been shown on television (said the rabid fan of large-tetrapod dissections).

Incidentally, the existence and importance of the RLN has been known since classical times. The RLN innervates the muscles responsible for speech, and on either side it passes right behind the thyroid gland, which is subject to goiters and tumors and other grotesque maladies. So a careless thyroidectomy can damage one or both of the RLNs; if one gets snipped, the subject will be hoarse for the rest of his or her life; if both are cut, the subject will be rendered mute. The Roman physician Galen memorably demonstrated this by dissecting the neck of an immobilized but unanesthetized pig and isolating the RLNs (Kaplan et al. 2009). One moment the poor pig was squealing its head off–as any of us would be if someone dug out our RLNs without anesthesia–and the next moment Galen severed the RLNs and the animal abruptly fell silent, still in unbelievable pain but now without a mechanism to vocally express its discomfort.

Galen versus pig. Figure 2 from Kaplan et al. 2009.

The name of the nerve also goes back to Galen, who wrote:

I call these two nerves the recurrent nerves (or reversivi) and those that come upward and backward on account of a special characteristic of theirs which is not shared by any of the other nerves that descend from the brain.

Like at least some modern surgeons, Galen does not seem to have been overly burdened by humility:

All these wonderful things, which have now become common property, I was the first of all to discover, no anatomist before me ever saw one of these nerves, and so all of them before me missed the mark in their anatomical description of the larynx.

Both of those quotes are from Kaplan et al. (2009), which is a fascinating paper that traces the knowledge of the recurrent laryngeal nerve from classical times to the early 20th century. If you’d like a copy and can’t get hold of one any other way, let me know and I’ll hook you up.

Share and share alike

By now you can see where this is going: all tetrapods have larynges, all tetrapods have necks, and all tetrapods have recurrent laryngeal nerves. Including giraffes, much to the delight of Richard Dawkins. And also including sauropods, much to the delight of yours truly.

Now, I cannot show you the RLN in a living sauropod, nor can I imagine a scenario in which such a delicate structure would be recognizably preserved as a fossil. But as tetrapods, sauropods were bound to the same unbreakable rules of development as everything else. The inference that sauropods had really long, really dumb RLNs is as secure as the inference that they had brainstems, hearts, and larynges.

Wedel (2012) Fig. 1. Course of the left vagus nerve and left recurrent laryngeal nerve in a human, a giraffe, and Supersaurus. The right recurrent laryngeal nerve passes caudal to the right subclavian artery rather than the aorta and ductus arteriosus, but otherwise its course is identical to that of the left.

Giraffes have necks up to 2.4 meters long (Toon and Toon 2003), so the neurons that make up their RLNs approach 5 meters in the largest indiividuals. But the longest-necked sauropods had necks 14 meters long, or maybe even longer, so they must have had individual neurons at least 28 meters long. The larynx of even the largest sauropod was probably less than 1 meter away from the brainstem, so the “extra” length imposed on the RLN by its recurrent course was something like 27 meters in a large individual of Supersaurus. Take that, Giraffa.

Inadequate giraffe is inadequate.

One way or another

It is possible to have a nonrecurrent laryngeal nerve–on one side, anyway. If you haven’t had the opportunity to dissect many cadavers, it may come as a surprise to learn that muscles, nerves, and blood vessels are fairly variable. Every fall in Gross Anatomy at WesternU, we have about 40 cadavers, and out of those 40 people we usually have two or three with variant muscles, a handful with unusual branching patterns of nerves, and usually half a dozen or so with some wackiness in their major blood vessels. Variations of this sort are common enough that the better anatomy atlases illustrate not just one layout for, say, the branching of the femoral artery, but 6-10 of the most common patterns. Also, these variations are almost always asymptomatic, meaning that they never cause any problems and the people who have them usually never know (ask Mike about his lonely kidney sometime). You–yes, you, gentle reader!–could be a serious weirdo and have no idea.

Variations in the blood vessels seem to be particularly common, possibly because the vessels develop in situ with apparently very little in the way of genetic control. Most parts of the body are served by more than one artery and vein, so if the usual vessel isn’t there or takes an unusual course, it’s often no big deal, as long as the blood gets there somehow. To wit: occasionally a person does not have a right subclavian artery. This does not mean that their right shoulder and arm receive no blood and wither away; usually it means that one of the segmental arteries branching off the descending aorta–which normally serve the ribs and their associated muscles and other soft tissues–is expanded and elongated to compensate, and looks for all the world like a normal subclavian artery with an abnormal connection to the aorta. But if the major artery that serves the forelimb comes from the descending aorta, and the 4th aortic arch on the right is completely resorbed during development, then there is nothing left on the right side to drag the inferior laryngeal nerve down into the torso. A person with this setup will have an inferior laryngeal nerve on the right that looks intelligently designed, and the usual dumb RLN on the left.

Can people have a nonrecurrent laryngeal nerve on the left? Sure, if they’ve got situs inversus, in which the normal bilateral asymmetry of the internal organs is swapped left to right. Situs inversus is pretty darned rare in the general population, occurring in fewer than 1 in 10,000 people. It is much more prevalent in television shows and movies, where the hero or villain may survive a seemingly mortal wound and then explain that he was born with his heart on the right side. (Pro tip: the heart crosses the midline in folks of both persuasions, so just shoot through the sternum and you’ll be fine. Or, if you’re worried about penetration, remember Rule #2 and put one on either side.) Anyway, take everything I wrote in the preceding paragraph, mirror-image it left to right, and you’ve got a nonrecurrent laryngeal nerve on the left. But just like the normally-sided person who still has an RLN on the left, a person with situs inversus and no remnant 4th aortic arch on the left (double variation alert!) still has an RLN looping around the aorta and ductus arteriosus on the right.

Bottom line: replumb the vessels to your arms, swap your organs around willy-nilly, you just can’t beat the aorta. If you have an aorta, you’ve got at least one RLN; if you don’t have an aorta, you’re dead, and no longer relevant to this discussion.

Nonrecurrent laryngeal nerves–a developmental Hail Mary?

But wait–how do we know that the inferior laryngeal nerve in embryonic sauropods didn’t get rerouted to travel in front of the fourth and sixth aortic arches, so it could be spared the indignity of being dragged into the chest later on?

First of all, such a course would require that the inferior laryngeal nerve take an equally dumb recurrent course in the embryo. Or maybe it should be called a procurrent course. Instead of simply radiating out from the central nervous system to its targets in the sixth pharyngeal arch, the axons that make up the RLN would have to run well forward of their normal course, loop around the fourth and sixth aortic arches from the front, and then run back down to the sixth pharyngeal arch. There is simply no known developmental mechanism that could make this happen.

Even if we postulated some hypothetical incentive that would draw those axons into the forward U-turn, other axons that took a more direct course from the central nervous system would get to the sixth pharyngeal arch first. By the time the forwardly-recurring axons finished their intelligently-routed course and finally arrived at the sixth pharyngeal arch, all of the innervation targets would be taken, and those axons would die off.

Also, at what point in the evolution of long necks would this forwardly-looping course supposedly be called into existence? Ostriches and giraffes have RLNs that take the same recurrent course as those of humans, pigs, and all other tetrapods. The retention of the recurrent course in extant long-necked animals is further evidence that the developmental constraint cannot be broken.

Finally, the idea that a non-recurrent laryngeal nerve would need to evolve in a long-necked animal is based on the perception that long nerve pathways are somehow physiologically taxing or otherwise bad for the animals in which they occur. But almost every tetrapod that has ever lived has had much longer neurons than those in the RLN, and we all get on just fine with them.

In dire extremity

Probably you have seen enough pictures of neurons to know what one looks like: round or star-shaped cell body with lots of short branches (dendrites) and one very long one (the axon), like some cross between an uprooted tree–or better yet, a crinoid–and the Crystalline Entity. When I was growing up, I always imagined these things lined up nose to tail (or, rather, axon to dendrite) all down my spinal cord, arms, and legs, like boxcars in a train. But it ain’t the case. Textbook cartoons of neurons are massively simplified, with stumpy little axons and only a few to a few dozen terminals. In reality, each neuron in your brain is wired up to 7000 other neurons, on average, and you have about a hundred billion neurons in your brain. (Ironically, 100 billion neurons is too many for your 100 billion neurons to visualize, so as a literal proposition, the ancient admonition to “know thyself” is a non-starter.)

Back to the axons. Forget the stumpy little twigs you’ve seen in books and online. Except for the ganglia of your autonomic nervous system (a semi-autonomous neural network that runs your guts), all of the cell bodies of your neurons are located in your central nervous system or in the dorsal root ganglia immediately adjacent to your spinal cord. The nerves that branch out into your arms and legs, across your face and scalp, and into your larynx are not made of daisy chains of neurons. Rather, they are bundles of axons, very long axons that connect muscles, glands, and all kinds of sensory receptors back to the nerve cell bodies in and around your brain and spinal cord.

Indulge me for a second and wiggle your toes. The cell bodies of the motor neurons that caused the toe-wiggling muscles to fire are located in your spinal cord, at the top of your lower back. Those motor neurons got orders transmitted down your spinal cord from your brain, and the signals were carried to the muscles of your feet on axons that are more than half as long as you are tall.

Some of your sensory neurons are even longer. Lift your big toe and then set it down gently, just hard enough to be sure that it’s touching down on the floor or the sole of your shoe, but not hard enough to exert any pressure. When you first felt the pad of your toe touch down, that sensation was carried to your brain by a single neuron (or, rather, by several neurons in parallel) with receptor terminals in the skin of your toe, axon terminals in your brainstem, and a nerve cell body somewhere in the middle (adjacent to your sacrum and just a bit to one side of your butt crack, if you want the gory details). That’s right: you have individual sensory neurons that span the distance from your brainstem to your most distal extremity. And so does every other vertebrate, from hagfish to herons to hippos. Including, presumably, sauropods.

I had you set your toe down gently instead of pushing down hard because the neurons responsible for sensing pressure do not travel all the way from toe-tip to brainstem; they synapse with other neurons in the spinal cord and those signals have been through a two-neuron relay by the time they reach your brainstem. Ditto for sensing temperature. But the neurons responsible for sensing vibration and fine touch go all the way.

If you want to experience everything I’ve discussed in this post in a single action, put your fingertips on your voicebox and hum. You are controlling the hum with signals sent from your brain to your larynx through your recurrent laryngeal nerves, and sensing the vibration through individual neurons that run from your fingertips to your brainstem. Not bad, eh?

Wedel (2012) Fig. 2. The longest cells in the bodies of sauropods were sensory neurons that connected receptors in the skin of the extremities with interneurons in the brainstem, a pattern of neural architecture that is present in all extant vertebrates. The nerve cell bodies would have been located in the dorsal root ganglia adjacent to the spinal cord. The diagram of the neuron is based on Butler and Hodos (1996: fig. 2–1B).

Getting back to big animals: the largest giraffes may have 5-meter neurons in their RLNs, but some of the sensory neurons to their hindfeet must be more like 8 meters long. I don’t think anyone’s ever dissected one out, but blue whales must have sensory neurons to the tips of their flukes that are almost 30 meters (98 feet) long (subtract the length of the skull, but add the lateral distance from body midline to fluke-tip). And Supersaurus, Amphicoelias, and the like must have had neurons that were approximately as long as they were, minus only the distance from the snout-tip to the back of the skull. I could be wrong, and if I am I’d love to be set straight, but I think these must have been the longest cells in the history of life.

Oh, one more thing: up above I said that almost every tetrapod that has ever lived has had much longer neurons than those in the RLN. The exceptions would be animals for which the distance from brainstem to base of neck was longer than the distance from base of neck to tip of limb or tail, so that twice the length of the neck would be longer than the distance from base of skull to most distal extremity. In that case, the neurons that contribute to the RLN would be longer than those running from brainstem to tail-tip or toe-tip. Tanystropheus and some of the elasmosaurs probably qualified; who else?

Parting Thoughts

In this post I’ve tried to explain the courses that these amazingly long cells take in humans and other vertebrates. I haven’t dealt at all with the functional implications of long nerves, for which please see the paper. The upshot is that big extant animals get along just fine with their crazy-long nerves, and there’s no reason to assume that sauropods were any more troubled. So why write the paper, then? Well, it was fun, I learned a lot (dude: axoplasmic streaming!), and most importantly I got to steal a little thunder from those silly poseurs, the giraffes.

Department of Frivolous Nonsense: yes, I titled the paper after those TV ads for Chili’s hamburgers from a few years back. If you never saw them, the ads compared mass-produced, machine-stamped fast-food burgers with restaurant burgers painstakingly built by hand, and concluded with, “Chili’s Big-Mouth Burgers: monuments of inefficiency!”

Update: All of the following is out of date now that the paper has been formally published. Department of Good Karma: since the paper is at the “accepted manuscript” stage, I still have the chance to make (hopefully minor) changes when I get the proofs. As is always, always, always the case, I caught a few dumb errors only after the manuscript had been accepted. Here’s what I’ve got so far, please feel free to add to the list:

  • Page 1, abstract, line 3: pharyngeal, not pharyngial
  • Page 1, abstract, line 8: substitute ‘made up’ for ‘comprised’
  • Page 6, line 12: substitute ‘make up’ for ‘comprise’
  • Page 9, line 5: citation should be of Carpenter (2006:fig. 3), not fig. 2
  • Page 10, line 7: “giant axons of squid are”, not ‘ares’
  • Page 12, entry for Butler and Hodos should have year (1996)
  • Page 12, entry for Carpenter has ‘re-evaluation misspelled
  • Page 16, entry for Woodburne has ‘mammalian’ misspelled

(Notes to self: stop trying to use ‘comprise’, lay off the ‘s’ key when typing bibliography.)


Why we do mass estimates

Mass estimates are a big deal in paleobiology. If you want to know how much an animal needed in terms of food, water, and oxygen, or how fast it could move, or how many offspring it could produce in a season, or something about its heat balance, or its population density, or the size of its brain relative to its body, then at some point you are going to need a mass estimate.

All that is true, but it’s also a bit bogus. The fact is, people like to know how big things are, and paleontologists are not immune to this desire. We have loads of ways to rationalize our basic curiosity about the bigness of extinct critters. And the figuring out part is both very cool and strangely satisfying. So let’s get on with it.

Two roads diverged

There are two basic modes for determining the mass of an extinct animal: allometric, and volumetric. Allometric methods rely on predictable mathematical relationships between body measurements and body mass. You measure a bunch of living critters, plot the results, find your regression line, and use that to estimate the masses of extinct things based on their measurements. Allometric methods have a couple of problems. One is that they are absolutely horrible for extrapolating to animals outside the size range of the modern sample, which ain’t so great for us sauropod workers. The other is that they’re pretty imprecise even within the size range of the modern sample, because real data are messy and there is often substantial scatter around the regression line, which if faithfully carried through the calculations produces large uncertainties in the output. The obvious conclusion is that anyone calculating extinct-animal masses by extrapolating an allometric regression ought to calculate the 95% confidence intervals (e.g. “Argentinosaurus massed 70000 kg, with a 95% confidence interval of 25000-140000 kg), but, oddly, no-one seems to do this.

Volumetric methods rely on creating a physical, digital, or mathematical model of an extinct animal, determining the volume of the model, multiplying by a scale factor to get the volume of the animal in life, and multiplying that by the presumed density of the living animal to get its mass. Volumetric methods have three problems: (1) many extinct vertebrates are known from insufficient material to make a good 3D model of the skeleton; (2) even if you have a complete skeleton, the method is very sensitive to how you articulate the bones–especially the ribcage–and the amount of flesh you decide to pack on, and there are few good guidelines for doing this correctly; and (3) relatively small changes in the scale factor of the model can produce big changes in the output, because mass goes with the cube of the linear measurement. If your scale factor is off by 10%, you mass will be off by 33% (1.1^3=1.33).

On the plus side, volumetric mass estimates are cheap and easy. You don’t need hundreds or thousands of measurements and body masses taken from living animals; you can do the whole thing in your kitchen or on your laptop in the space of an afternoon, or even less. In the old days you’d build a physical model, or buy a toy dinosaur, and use a sandbox or a dunk tank to measure the volume of sand or water that the model displaced, and go from there. Then in the 90s people started building digital 3D models of extinct animals and measuring the volumes of those.

But you don’t need a physical model or a dunk tank or even a laptop to do volumetric modeling. Thanks to a method called graphic double integration or GDI, which is explained in detail in the next section, you can go through the whole process with nothing more than pen and paper, although a computer helps.

Volumetric methods in general, and GDI in particular, have one more huge advantage over allometric methods: they’re more precise and more accurate. In the only published study that compares the accuracy of various methods on extant animals of known mass, Hurlburt (1999) found that GDI estimates were sometimes off by as much as 20%, but that allometric estimates were much worse, with several off by 90-100% and one off by more than 800%. GDI estimates were not only closer to the right answers, they also varied much less than allometric methods. On one hand, this is good news for GDI afficionados, since it is the cheapest and easiest of all the mass estimation methods out there. On the other hand, it should give us pause that on samples of known mass, the best available method can still be off by as much as a fifth even when working with complete bodies, including the flesh. We should account for every source of error that we can, and still treat our results with appropriate skepticism.

Graphic Double Integration

GDI was invented by Jerison (1973) to estimate the volumes of cranial endocasts. Hurlburt (1999) was the first to apply it to whole animals, and since then it has been used by Murray and Vickers-Rich (2004) for mihirungs and other extinct flightless birds, yours truly for small basal saurischians (Wedel 2007), Mike for Brachiosaurus and Giraffatitan (Taylor 2009), and probably many others that I’ve missed.

GDI is conceptually simple, and easy to do. Using orthogonal views of a life restoration of an extinct animal, you divide the body into slices, treat each slice as an ellipse whose dimensions are determined from two perspectives, compute the average cross-sectional area of each body part, multiply that by the length of the body part in question, and add up the results. Here’s a figure from Murray and Vickers-Rich (2004) that should clarify things:

One of the cool things about GDI is that it is not just easy to separate out the relative contributions of each body region (i.e., head, neck, torso, limbs) to the total body volume, it’s usually unavoidable. This not only lets you compare body volume distributions among animals, it also lets you tinker with assigning different densities to different body parts.

An Example: Plateosaurus

Naturally I’m not going to introduce GDI without taking it for a test drive, and given my proclivities, that test drive is naturally going to be on a sauropodomorph. All we need is an accurate reconstruction of the test subject from at least two directions, and preferably three. You could get these images in several ways. You could take photographs of physical models (or toy dinosaurs) from the front, side, and top–that could be a cool science fair project for the dino-obsessed youngster in your life. You could use the white-bones-on-black-silhouette skeletal reconstructions that have become the unofficial industry standard. You could also use orthogonal photographs of mounted skeletons, although you’d have to make sure that they were taken from far enough away to avoid introducing perspective effects.

For this example, I’m going to use the digital skeletal reconstruction of the GPIT1 individual of Plateosaurus published by virtual dino-wrangler and frequent SV-POW! commenter Heinrich Mallison (Mallison et al 2009, fig. 14). I’m using this skeleton for several reasons: it’s almost complete, very little distorted, and I trust that Heinrich has all the bits in the right places. I don’t know if the ribcage articulation is perfect but it looks reasonable, and as we saw last time that is a major consideration. Since Heinrich built the digital skeleton in digital space, he knows precisely how big each piece actually is, so for once we have scale bars we can trust. Finally, this skeleton is well known and has been used in other mass estimate studies, so when I’m done we’ll have some other values to compare with and some grist for discussion. (To avoid accidental bias, I’m not looking at those other estimates until I’ve done mine.)

Of course, this is just a skeleton, and for GDI I need the body outline with the flesh on. So I opened the image in GIMP (still free, still awesome) and drew on some flesh. Here we necessarily enter the realm of speculation and opinion. I stuck pretty close to the skeletal outline, with the only major departures being for the soft tissues ventral to the vertebrae in the neck and for the bulk of the hip muscles. As movie Boromir said, there are other paths we might take, and we’ll get to a couple of alternatives at the end of the post.

This third image is the one I used for actually taking measurements. You need to lop off the arms and legs and tote them up separately from the body axis. I also filled in the body outlines and got rid of the background so I wouldn’t have any distracting visual clutter when I was taking measurements. I took the measurements using the measuring tool in GIMP (compass icon in the toolbar), in orthogonal directions (i.e., straight up/down and left/right), at regular intervals–every 20 pixels in this case.

One thing you’ll have to decide is how many slices to make. Ideally you’d do one slice per pixel, and then your mathematical model would be fairly smooth. There are programs out there that will do this for you; if you have a 3D digital model you can just measure the voxels (= pixels cubed) directly, and even if all you have is 2D images there are programs that will crank the GDI math for you and measure every pixel-width slice (Motani 2001). But if you’re just rolling with GIMP and OpenOffice Calc (or Photoshop and Excel, or calipers and a calculator), you need to have enough slices to capture most of the information in the model without becoming unwieldy to measure and calculate. I usually go with 40-50 slices through the body axis and 9 or 10 per limb.

The area of a circle is pi*r^2, and the area of an ellipse is pi*r*R, where r and R are the radii of the minor and major axes. So enter the widths and heights of the body segments in pixels in two columns (we’ll call them A and B) in your spreadsheet, and create a third column with the function 3.14*A1*B1/4. Divide by four because the pixel counts you measured on the image are diameters and the formula requires radii. If you forget to do that, you are going to get some wacky numbers.

One obvious departure from reality is that the method assumes that all of the body segments of an animal have elliptical cross-sections, when that is often not exactly true. But it’s usually close enough for the coarse level of detail that any mass estimation method is going to provide, and if it’s really eating you, there are ways to deal with it without assuming elliptical cross-sections (Motani 2001).

For each body region, average the resulting areas of the individual slices and multiply the resulting average areas by the lengths of the body regions to get volumes. Remember to measure the lengths at right angles to your diameter measurements, even when the body part in question is curved, as is the tail of Heinrich’s Plateosaurus.

For sauropods you can usually treat the limbs as cylinders and just enter the lateral view diameter twice, unless you are fortunate enough to have fore and aft views. It’s not a perfect solution but it’s probably better than agonizing over the exact cross sectional shape of each limb segment, since that will be highly dependent on how much flesh you (or some other artist) put on the model, and the limbs contribute so little to the final result. For Plateosaurus I made the arm circular, the forearm and hand half as wide as tall, the thigh twice as long as wide, and the leg and foot round. Don’t forget to double the volumes of the limbs since they’re paired!

We’re not done, because so far all our measurements are in pixels (and pixels cubed). But already we know something cool, which is what proportion each part of the body contributes to the total volume. In my model based on Heinrich’s digital skeleton, segmented as shown above, the relative contributions are as follows:

  • Head: 1%
  • Neck: 3%
  • Trunk: 70%
  • Tail: 11%
  • Forelimbs (pair): 3%
  • Hindlimbs (pair): 12%

Already one of the great truths of volumetric mass estimates is revealed: we tend to notice the extremities first, but really it is the dimensions of the trunk that drive everything. You could double the size of any given extremity and the impact on the result would be noticeable, but small. Consequently, modeling the torso accurately is crucial, which is why we get worried about the preservation of ribs and the slop inherent in complex joints.

Scale factor

The 170 cm scale bar in Heinrich’s figure measures 292 pixels, or 0.582 cm per pixel. The volume of each body segment must be multiplied by 0.582 cubed to convert to cubic cm, and then divided by 1000 to convert to liters, which are the lingua franca of volumetric measurement. If you’re a math n00b, your function should look like this: volume in liters = volume in pixels*SF*SF*SF/1000, where SF is the scale factor in units of cm/pixel. Don’t screw up and use pixels/cm, or if you do, remember to divide by the scale factor instead of multiplying. Just keep track of your units and everything will come out right.

If you’re not working from an example as perfect as Heinrich’s digital (and digitally measured) skeleton, you’ll have to find something else to use for a scale bar. Something big and reasonably impervious to error is good. I like the femur, if nothing else is available. Any sort of multi-segment dimension like shoulder height or trunk length is going to be very sensitive to how much gloop someone thought should go between the bones. Total length is especially bad because it depends not only on the intervertebral spacing but also on the number of vertebrae, and even most well-known dinos do not have complete vertebral series.


Finally, multiply the volume in liters by the assumed density to get the mass of each body segment. Lots of people just go with the density of water, 1.0 kg/L, which is the same as saying a specific gravity (SG) of 1. Depending on what kind of animal you’re talking about, that may be a little bit off or it may be fairly calamitous. Colbert (1962) found SGs of 0.81 and 0.89 for an extant lizard and croc, which means an SG of 1.0 is off by between 11% and 19%. Nineteen percent–almost a fifth! For birds, it’s even worse; Hazlehurst and Rayner (1992) found an SG of 0.73.

Now, scroll back up to the diagram of the giant moa, which had a mass of 257.5 kg “assuming a specific gravity of 1”. If the moa was as light as an extant bird–and its skeleton is highly pneumatic–then it might have had a mass of only 188 kg (257.5*0.73). Or perhaps its density was higher, like that of a lizard or a croc. Without a living moa to play with, we may never know. Two points here: first, the common assumption of whole-body densities of 1.0 is demonstrably incorrect* for many animals, and second, since it’s hard to be certain about the densities of extinct animals, maybe the best thing is to try the calculation with several densities and see what results we get. (My thoughts on the plausible densities of sauropods are here.)

* Does anyone know of actual published data indicating a density of 1.0 for a terrestrial vertebrate? Or is the oft-quoted “bodies have the same density as water” basically bunk? (Note: I’m not disputing that flesh has a density close to that of water, but bones are denser and lungs and air spaces are lighter, and I want to know the mean density of the whole organism.)

Back to Plateosaurus. Using the measurements and calculations presented above, the total volume of the restored animal is 636 liters. Here are the whole body masses (in kg) we get using several different densities:

  • SG=1.0 (water), 636 kg
  • SG=0.89 (reptile high), 566 kg
  • SG=0.81 (reptile low), 515 kg
  • SG=0.73 (bird), 464 kg

I got numbers. Now what?

I’m going to describe three possible things you could do with the results once you have them. In my opinion, two of them are the wrong the thing to do and one is the right thing to do.

DON’T mistake the result of your calculation for The Right Answer. You haven’t stumbled on any universal truth. Assuming you measured enough slices and didn’t screw up the math, you know the volume of a mathematical model of an organism. If you crank all the way through the method you will always get a result, but that result is only an estimate of the volume of the real animal the model was based on. There are numerous sources of error that could plague your results, including: incomplete skeletal material, poorly articulated bones, wrong scale factor, wrong density, wrong amount of soft tissue on the skeleton. I saved density and gloop for last because you can’t do much about them; here the strength of your estimate relies on educated guesses that could themselves be wrong. In short, you don’t even know how wrong your estimate might be.

Pretty dismal, eh?

DON’T assume that the results are meaningless because you don’t know the actual fatness or the density of the animal, or because your results don’t match what you expected or what someone else got. I see this a LOT in people that have just run their first phylogenetic analysis. “Why, I could get any result I wanted just by tinkering with the input!” Well, duh! Like I said, the method will always give you an answer, and it won’t tell you whether the answer is right or not. The greatest advantage of explicit methods like cladistics and GDI is that you know what the input is, and so does everyone else if you are honest about reporting it. So if someone disagrees with your character coding or with how much the belly sags on your model sauropod, you can have a constructive discussion and hopefully science as a whole gets closer to the right answer (even if we have no way of knowing if or when we arrive, and even if your pet hypothesis gets trampled along the way).

DO be appropriately skeptical of your own results without either accepting them as gospel or throwing them out as worthless. The fact that the answer changes as you vary the parameters is a feature, not a bug. Investigate a range of possibilities, report all of those results, and feel free to argue why you think some of the results are better than others. Give people enough information to replicate your results, and compare your results to those of other workers. Figure out where yours differ and why.

Try to think of more interesting things you could do with your results. Don Henderson went from digitally slicing critters (Henderson 1999) to investigating floating sauropods (Henderson 2004) to literally putting sauropods through their paces (Henderson 2006)–not to mention working on pterosaur flight and swimming giraffes and other cool stuff. I’m not saying you should run out and do those exact things, but rather that you’re more likely to come up with something interesting if you think about what you could do with your GDI results instead of treating them as an end in themselves.

How massive was GPIT1, really?

Beats me. I’m not the only one who has done a mass estimate based on that skeleton. Gunga et al. (2007) did not one but two volumetric mass estimates based on GPIT1, and Mallison (2010) did a whole series, and they published their models so we can see how they got there. (In fact, many of you have probably been reading this post in slack-jawed horror, wondering why I was ignoring those papers and redoing the mass estimate the hard way. Now you know!) I’m going to discuss the results of Gunga et al. (2007) first, and come back to Mallison (2010) at the end.

Here’s the “slender” model of Gunga et al. 2007 (their fig. 3):

and here’s their “robust” model (Gunga et al. 2007:fig. 4):

(These look a bit…inelegant, let’s say…because they are based on the way the physical skeleton is currently mounted; Heinrich’s model looks much nicer because of his virtual remount.)

For both mass estimates they used a density of 0.8, which I think is probably on the low end of the range for prosauropods but not beyond the bounds of possibility. They got a mass of 630 kg for the slender model and 912 kg for the robust one.

Their 630-kg estimate for the slender model is deceptively close to the upper end of my range; deceptive because their 630-kg estimate assumes a density of 0.8 and my 636-kg one assumes a density of 1.0. The volumes are more directly comparable: 636 L for mine, 790 L for their slender one, and 1140 L for their robust one. I think that’s pretty good correspondence, and the differences are easily explained. My version is even more skinnier than their slender version; I made it about as svelte as it could possibly have been. I did that deliberately, because it’s always possible to pack on more soft tissue but at some point the dimensions of the skeleton establish a lower bound for how voluminous a healthy (i.e., non-starving) animal could have been. The slender model of Gunga et al. (2007) looks healthier than mine, whereas their robust version looks, to my eye, downright corpulent. But not unrealistically so; fat animals are less common than skinny ones but they are out there to be found, at least in some times and places. It pays to remember that the mass of a single individual can fluctuate wildly depending on seasonal food availability and exercise level.

For GPIT1, I think something like 500 kg is probably a realistic lower bound and 900 kg is a realistic upper bound, and the actual mass of an average individual Plateosaurus of that size was somewhere in the middle. That’s a big range–900 kg is almost twice 500 kg. It’s hard to narrow down because I really don’t know how fleshy Plateosaurus was or what it’s density might have been, and I feel less comfortable making guesses because I’ve spent much less time working on prosauropods than on sauropods. If someone put a gun to my head, I’d say that in my opinion, a bulk somewhere between that of my model and the slender model of Gunga et al. is most believable, and a density of perhaps 0.85, for a result in the neighborhood of 600 kg. But those are opinions, not hypotheses, certainly not facts.

I’m happy to see that my results are pretty close to those of Mallison (2010), who got 740 L, which is also not far off from the slender model of Gunga et al. (2007). So we’ve had at least three independent attempts at this and gotten comparable results, which hopefully means we’re at least in the right ballpark (and pessimistically means we’re all making mistakes of equal magnitude!). Heinrich’s paper is a goldmine, with loads of interesting stuff on how the skeleton articulates, what poses the animal might have been capable of, and how varying the density of different body segments affects the estimated mass and center of mass. It’s a model study and I’d happily tell you all about it but you should really read it for yourself. Since it’s freely available (yay open access!), there’s no barrier to you doing so.


So: use GDI with caution, but do use it. It’s easy, it’s cool, it’s explicit, it will give you lots to think about and give us lots to talk about. Stay tuned for related posts in the not-too-distant future.