I made this, just for the heck of it.


The critters are, from left to right:

  • OMNH 53062, the holotype of Sauroposeidon proteles, with a reconstructed skeleton grayed in;
  • HM XV2, a fibula of Brachiosaurus brancai, which represents the largest known individual of Brachiosaurus;
  • HM SII, the nearly complete mounted composite skeleton of Brachiosaurus brancai in Berlin;
  • a 20-foot-tall, world record giraffe;
  • a 6’2″ human being, such as myself.

The vertebrae of Sauroposeidon are about a third longer than their counterparts in HM SII, but only about 15% larger in diameter. I have therefore always scaled up the body of Sauroposeidon by only 15% relative to HM SII. It may have been bigger or smaller, I’m just trying to follow what few numbers I have to go on as slavishly, and conservatively, as possible. Sauroposeidon is shown here with a more vertical neck than Brachiosaurus because that’s how I had the necks posed in the  two separate skeleton reconstructions before I decided to combine them, and I’m lazy, and that’s not the point of the post anyway.

The point of the post, or the first point anyway, is that almost everyone, everywhere, at all times underestimates the size of Brachiosaurus. This is because of the immense influence of the HM SII mounted skeleton. Practically every estimate of length or neck length or browsing height or mass or anything else for Brachiosaurus is based on that one skeleton. But we know that there were bigger individuals of Brachiosaurus roaming around, like HM XV2, which was 12-13% larger. Not only that, but we can be pretty certain that HM SII was not fully mature because the scapula and coracoid are unfused, and we know these elements are fused into a single scapulocoracoid in mature brachiosaurids. So between SII being not all grown up and XV2 being considerably bigger, we ought to think of XV2 and not SII when we think about big Brachiosaurus was.

Now, 12-13% might not seem like much, but it’s considerable. It’s the difference between me (6’2″) and someone seven feet tall. HM SII has a neck 8.5 meters long; that of XV2 would have been 9.5 meters long, which is longer than the neck of the holotype of Mamenchisaurus hochuanensis (9 m), but shorter than the estimated neck length of Mamenchisaurus sinocanadorum (~12m).

Crucially, XV2 would have massed 1.4 times as much as SII (1.125^3, because mass depends on volume, which scales with the cube of length). That holds true no matter how much you think SII weighed. If SII had a mass of 40 tons, then XV2 was 56 tons; if SII was 30 tons, XV2 was still 42 tons.

Maybe the most interesting thing about this is that, so far as we can tell, XV2 was almost exactly the same size as the holotype individual of Sauroposeidon. So anything I or anyone else has written about Sauroposeidon being bigger, absolutely, than Brachiosaurus, is bobbins. Sauroposeidon still had a considerably longer neck, 11.5 meters to XV2’s 9.5, but the cervical skeleton weighed about the same thanks to the higher air space proportion in Sauroposeidon. In fact, if the higher ASP of Sauroposeidon applied to the rest of the vertebral column, then the holotype individual of Sauroposeidon might have weighed less than XV2!

The evolutionary upshot is that, as far as we can tell, big brachiosaurids stayed about the same size from the Kimmeridgian-Tithonian (Late Jurassic) to the Aptian-Albian (Early Cretaceous). Maybe they hit some kind of limit, but I doubt it, because Argentinosaurus was probably a lot heavier and Bruhathkayosaurus and Amphicoelias would have knocked any known brachiosaurid right out of the park. I think it is more likely that the debits imposed by large body size finally caught up with the selective advantages of same, within that lineage (but not at the same point within other lineages). Whatever the reason, the biggest known brachiosaurid didn’t get any bigger than Brachiosaurus. Which puts the evolution of the longer, more pneumatic neck in Sauroposeidon into a new light. It might have been a cheat, an evolutionary hack to overcome a limit on whole-body growth, even if that limit was a ‘soft’ one imposed by balanced selection pressures in both directions. That’s sort of assuming that Sauroposeidon was just Brachiosaurus with a redesigned front end, but the weirdness we see in the vertebrae might have extended to the rest of the animal. We won’t know until someone digs up some more specimens. Sigh.

The second point of the post is that, as indicated by the title, Brachiosaurus might have been smaller than we commonly think. Since the 1980s there have been a couple of ~30 ton estimates out there for HM SII, one by Anderson et al. (1985) based on limb bone allometry and one by Paul [1988] based on volumetrics (I have to put 1988 publication dates in brackets rather than parentheses or mrrfin’ frrfin’ WordPress automatically changes the 8 and the ) to a smiley, dammit). I think that by and large people have gotten pretty comfortable with the idea that SII was a 30 ton  critter.

But it might–might–have been quite a bit lighter. Paul (1997) assigned the neck a density of 0.6 g/cm^3 and the torso a density of 0.9 g/cm^3. Those are probably too dense. Some birds have necks as un-dense (sparse?) as 0.3 g/cm^3, and that does not strike me as unreasonable for sauropod necks given the amount of pneumaticity indicated by the skeleton. The lungs and air sacs of birds can account for up to 20% of the volume of the body. Not of the torso, of the whole body. And based on my calculations for derived theropods and sauropods, up to 10% of the whole-body volume was occupied by air in the pneumatic bones. That’s 10% in addition to the 20% for the lungs and air sacs, or 30% of the whole body  volume. That would give a whole-body density of about 0.7 g/cm^3, which is in fact what has been found for some birds.

I got 0.8 g/cm^3 for the whole-body density of Diplodocus in my 2005 paper, and other authors have since used that number for other sauropodomorphs. That’s gratifying, but it’s probably wrong. I erred conservatively at every possible point in that calculation and just flat left out some known air spaces whose volume I could not reliably estimate (e.g., vertebral diverticula outside the vertebrae). I also used 10% rather than 20% for the part of the whole-body volume occupied by the lungs and air sacs, because values as low as 10% have been reported for some birds and I was being conservative. But I don’t think that bird-like densities around 0.7 g/cm^3 are unrealistic for sauropods; in fact, I’d be surprised if the really pneumatic ones–like big brachiosaurids–weren’t about that sparse.

And speaking of big brachiosaurids, Henderson (2004) used computerized volumetrics and a density of 0.8 and got a mass of 25.8 tons for HM SII. If the density was really 0.7 that would shave off an additional 10% and bring the mass down to 22.7 tons. That’s getting crazy light; it’s about half of what Bakker and Alexander were proposing for Brachiosaurus in the mid-80s. And it’s still a quarter lighter than what Anderson (1985) and Paul [1988] got.

So, for the sake of argument, let’s say that HM SII did mass only 22.7 tons. That would give XV2 a mass of 32 tons, and Sauroposeidon a mass of only 34.5 tons without taking any additional pneumaticity into account.

That seems totally nuts. But every step is defensible*, and it might even be true.

* That means if you want to tear me a new one in the comments because teh Brachiosaurus wuz 50 tons!!!1!!111!, please be sure to specify which links in the chain of inference you disagree with, and why.


  • Anderson, J. F., A. Hall-Martin, and D. A. Russell. 1985. Long-bone circumference and weight in mammals, birds and dinosaurs. Journal of Zoology 207:53-61.

  • Henderson, D. M. 2004. Tipsy punters: sauropod dinosaur pneumaticity, buoyancy and aquatic habits. Proceedings: Biological Sciences 271 (Supplement):S180-S183.

  • Paul, G. S. 1988. The brachiosaur giants of the Morrison and Tendaguru with a description of a new subgenus, Giraffatitan, and a comparison of the world’s largest dinosaurs. Hunteria 2(3):1-14.
  • Paul, G. S. 1997. Dinosaur models: the good, the bad, and using them to estimate the mass of dinosaurs; pp. 129-154 in Wolberg, D. L., Stump, E., and Rosenberg, G. (eds.). Dinofest International: Proceedings of a Symposium Sponsored by Arizona State University. Academy of Natural Sciences, Philadelphia, 587 pp.