How horrifying was the neck of Barosaurus?

September 16, 2016

Suppose that I and Matt were right in our SVPCA talk this year, and the
Supersaurus” cervical BYU 9024 really is the C9 of a gigantic Barosaurus. As we noted in our abstract, its total length of 1370 mm is exactly twice that of the C9 in AMNH 6341, which suggests its neck was twice as long over all — not 8.5 m but 17 m.

How horrifying is that?

I realised one good way to picture it is next to the entire mounted skeleton of Giraffatitan at the Museum für Naturkunde Berlin. That skeleton is 13.27 m tall. At 17 m, the giant barosaur neck would be 28% longer than the total height of Giraffatitan.

Giraffatitan brancai mounted skeleton MB.R.2181 at the Museum für Naturkunde Berlin, with neck of Barosaurus ?lentus BYU 9024 at the same scale. Photo by Axel Mauruszat, from Wikipedia; drawing from Scott Hartman's Supersaurus skeleton reconstruction.

Giraffatitan brancai mounted skeleton MB.R.2181 at the Museum für Naturkunde Berlin, with neck of Barosaurus ?lentus BYU 9024 at the same scale. Photo by Axel Mauruszat, from Wikipedia; drawing from Scott Hartman’s Supersaurus skeleton reconstruction.

Yes, this looks ridiculous. But it’s what the numbers tell us. Measure the skeleton’s height and the neck length off the image yourself if you don’t believe me.

(Note, too, that the size of the C9 in that big neck is about right, compared with a previous scaled image that Matt prepared, showing the “Supersaurus” vertebra in isolation alongside the Chicago Brachiosaurus.)

118 Responses to “How horrifying was the neck of Barosaurus?”

  1. I have to wonder if there would be some allometry, but in any case, that’s absurdly big. Diplodocids don’t get enough credit for their size.

  2. ggg ggg Says:

    Does this mean that Supersaurus is invalid? Wasn’t it supposed to be an apatosaurine?

  3. Illiterate Scholar Says:

    How complete or how much material do we have of Barosaurus’s neck? Not for BYU 9024, just Barosaurus in general.

  4. Mike Taylor Says:

    Illiterate Scholar, there are a couple of pretty good specimens of Barosaurus that preserve complete or nearly complete necks — see the review in McIntosh (2005). The best is probably AMNH 6341, which is sort of on display in the saurischian gallery at the AMNH in New York.

    ggg ggg, we’re not really in a position to say anything about the validity of Supersaurus. All we’re claiming here is that a specimen which is usually referred to that genus is actually Barosaurus; but the identity of Supersaurus is tied to the type specimen, which is a big scapulocoracoid. Looking at briefly it at the North American Museum of Ancient Life in May, neither Matt nor I saw anything to obviously differentiate it from any other diplodocid scapulocoracoid — but then we weren’t really looking closely.

    Is Supersaurus an apatosaurine? Lovelace et al.(2008) concluded that their referred specimen “Jimbo” is, but their reasons for thinking Jimbo is Supersaurus are not really explained in their paper. So as far as I am concerned, all options are still on the table.

    John D’Angelo: yes, it’s possible and maybe likely that the neck of the giant Barosaurus was allometrically larger than the body; or if you prefer, that the torso, tail and limbs were allometrically small. (Matt, on the other hand, claims that the whole darned animal was positively allometric.) Parrish (2006) showed that neck lengths very with positive allometry between sauropod species, but (A) the data-set is small, and (B) to my knowledge no-one has tried to determine whether this is true between individuals within sauropod species.


    Lovelace, David M. Scott A. Hartman and William R. Wahl. 2008. Morphology of a specimen of Supersaurus (Dinosauria, Sauropoda) from the Morrison Formation of Wyoming, and a re-evaluation of diplodocid phylogeny. Arquivos do Museu Nacional, Rio de Janeiro 65(4):527-544.

    McIntosh, John S. 2005. The Genus Barosaurus Marsh (Sauropoda, Diplodocidae). pp. 38-77 in Virginia Tidwell and Ken Carpenter (eds.), Thunder Lizards: the Sauropodomorph Dinosaurs. Indiana University Press, Bloomington, Indiana. 495 pp.

    Parrish, J. Michael. 2006. The origins of high browsing and the effects of phylogeny and scaling on neck length in Sauropodomorpha. pp. 201-224 in M. T. Carrano, T. J. Gaudin, R. W. Blob, and J. R. Wible (eds.), Amniote Paleobiology: Phylogenetic and Functional Perspectives on the Evolution of Mammals, Birds and Reptiles. Chicago: University of Chicago Press.

  5. Marco Says:

    Really 17 m long? This freakin’ sauropods just ignore any space-saving rule!
    A similar animal should be 40 m in lenght or something

  6. Andrea Cau Says:

    A sauropod neck alone is longer than the whole body of any other non-sauropod dinosaur?

  7. Shahen Says:

    How long is AMNH Barosaurus?

  8. Mike Taylor Says:

    Marco: I know, right? One of my favourite things about prehistoric animals is that every time someone comes up with a fundamental limit on animal size, someone goes out and finds an animal that flagrantly defies that limit — like when everyone thought Pteranodon was at the very limit of how big a flying animal could be.

    Andrea: yes, really — at least, that’s our best estimate. It could be a bit less if BYU 9024 is really more posterior than C9 (or indeed a bit more if it’s more anterior!)

    Shahen: I don’t know offhand how long AMNH Barosaurus is, but given that its neck is longer and iits tail shorter than in the Carnegie Diplodocus, it’s going to be in the same ballpark. Somewhere around 25 m total length.

  9. Brad Says:

    Could this cervical and Cope’s lost “Amphicoelias fragillimus” dorsal belong to the same horrifying taxon?

  10. Mike Taylor Says:

    Brad, the thought has occurred to me. I know of no reason why they couldn’t be the same thing; but also no compelling reason to think that they are. If Barosaurus got twice as big as we thought, maybe Diplodocus did as well>

  11. Marco Says:

    Mike Taylor: You are absolutly right!
    Anyway this afternoon I play with GIMP for one hour or so with a BYU vert. photo and Hartman’s Barosaurus lentus skeletal… And I found a really strong similarities between BYU 9024 and 10th/11th Barosaurus cervical verts. as recostructed by Hartman.
    Try it if you have time :)

  12. Shahen Says:

    If all this is True then Barosaurus could be really huge probably the biggest. Some around 50 m and probably close to 100 tons… Few months ago I was wrote about Bruhathkayosaurus – in my opinion of course if was real could be bigger than Argentinosaurus only 4% and Any sauropod was bigger than around 70 tons… but now Barosaurus could be largest.

  13. Mike Taylor Says:

    Yes, it would have been big. If we’re right that BYU 9024 is a Barosaurus C9 that’s twice as big as the AMNH one, and if the whole animal was isometrically scaled so that it was twice as big as the AMNH specimen, then it was 2 x 2 x 2 = 8 times as massive. 12.5 tonnes would be a reasonable mass estimate for regular Barosaurus, which would make the big one 100 tonnes.

  14. Shahen Says:

    here is my post about Bruhathkayosaurus (polish), when I compare femur with other Macronaria, then that size will be reasonable.

    But Barosaurus could be the largest sauropod its really awesome information!

  15. Mike Taylor Says:

    How can you compare the femur of Bruhathkayosaurus with anything when the only element we know from it is the tibia? (Yes, there is an ilium, but it’s much too small to be part of the same animal.)

  16. Shahen Says:

    In orginal paper are – tibia (I’m not shure that really was), femur fragment, caudal centra, ilium, ishium and radius

  17. Kevin Says:

    Mind blowing. It just seems impossible, and yet good evidence of truly immense sauropods keeps emerging.

    And today we have relatively minuscule giraffes with adaptations like tight, “flight-suit” skin on their legs to counter the internal hydrostatic pressure related to their “large” size. Apologies if I’ve missed it, but have there been any posts on how these immense sauropods would have handled the same? Or does that come down to pneumasticity, too?

  18. Mike Taylor Says:

    Kevin, I have often wondered the same thing. Giraffes need so many anatomic innovations just to manage their feeble 2 m necks. How on earth did sauropods cope? I think they must have had much more going on than we have yet realised.

  19. Tschopp et al. found an OTU based on the material from the Supersaurus type locality to clade with ‘Jimbo’ as a basal diplodocine, but obviously that isn’t very useful if the type locality material they included is a mix of multiple taxa, such as this possible Barosaurus.

    Based on Tschopp et al’s codings, I’m not convinced Supersaurus can be distinguished based on the scapulocoracoid alone. Of the characters which differentiate it from any Diplodocus specimen:

    362. Angle between the acromial ridge and scapular blade. This differs between the right and left scapulocoracoids from the type locality, and the range overlaps with the range in other diplodocid species, although it’s closer to the apatosaurine average than the diplodocine average.
    366. Orientation of glenoid. This is variable (perhaps ontogenetically) in apatosaurines, and individually varies in Diplodocus carnegii. Supersaurus has the typical non-apatosaurine state.
    367. Shape of acromial edge of blade. Individually variable in Diplodocus and lacks a clear phylogenetic signal elsewhere.
    369. Ovate muscle scar on lateral face of blade at base. Variable in Brontosaurus parvus and Diplodocus hallorum.
    370. Subtriangular projection at base of blade, posterior to the glenoid. Individually variable in B. parvus and D. hallorum.
    371. Distal expansion of blade. Individually variable in…uh, B. parvus again, with BYU 1252-18531 being the only diplodocid to resemble Supersaurus in this regard.

    The takeaway here, I guess, is that the two specimens Tschopp et al. refer to Brontosaurus parvus, UW 15556 and BYU 1252-18531, differ in suspiciously many scapular characters, and that I doubt that diplodocid scapulae are particularly reliable character sources. ‘Jimbo’ is presumably a distinct taxon, but whether or not it’s Supersaurus depends largely on what Dry Mesa Quarry material can be reliably assigned to it.

  20. toneelphtuk Says:

    So every now and again we get a glimpse at just how big ‘pods could get. Is it possible that these outliers – Amphicoelias included, if it indeed existed – represent the final growth stage of diplodocids, but that most individuals didn’t live long enough to reach that stage? Hadrosaurs seem to have something similar going on.

    That would raise the followup question – why would most individuals die before reaching their full potential size, if they stopped having to worry about natural predators well before then?

  21. Ryan Kearns Says:

    So it’s just one vertebrae found? Could it be possible that the sauropod had really long but fewer vertebrae in the neck? How confidently can we assign this bone to an already discovered species or genus?

  22. bricksmashtv Says:

    Interesting, so Supersaurus may not even be a valid genus. More food for thought I suppose.

  23. Mike Taylor Says:

    Interesting observations, John. To you and bricksmashtv, I say my take is that the validity of Supersaurus is open to question, but that this is a problem for someone else to tackle — ideally someone who’s spent a lot more quality time with the holotype scap than I have, and has also seen the Jimbo material.

    toneelphtuk, your idea that the super-large sauropods we sometimes find evidence for are merely fully-grown individuals of well-known species is a credible one, and something that Matt Wedel has been avocating for a while. See for example this post. I think the slowly building evidence supports this idea.

    Ryan Kearns, we’re about as confident as we can be that BYU 9024 is Barosaurus. As we explain in the talk abstract, it hss a whole bunch of characters that support this assignment, and none that contradict it. That being so, we can be confident that it was part of a sixteen-vertebra neck, just like the AMNH one.

  24. Dan Chure Says:

    I would expect that given their immense size that the more robust elements of these giants, such as the femur, would be found in sauropod rich formations such as the Morrison. One might invoke ecological segregation but that is hard to envision.

  25. Mike Taylor Says:

    Interesting thought, Dan. So why do we have the fragile cervical vertebra BYU 9024 and the dorsal vertebra of Amphicoelias fragillimus (admittedly now only as a drawing) but none of the robust long-bones? Hmm.

  26. Kevin Says:

    Obviously, the long-bones of full-sized sauropods are never found because they never existed. Sauropod life history:

    1) Highly active hatchlings capable of bipedal running, feeding on vegetation and the occasional invertebrate.

    2) Large juveniles/sub-adults who were reproductively active, before biomechanical considerations became an obstacle.

    3) Post-reproductive adults, whose constant slow browsing creates productive edge habitat that supports the earlier two life stages.

    4) Immense post-adults, which are stationary and lay prone upon the ground. Their metabolism greatly slows, limbs other than the neck are gradually resorbed, and their only feeding is done in the arc reachable with their necks. The protein and other nutrients accumulated over their long lifetimes are slowly secreted in the form of thickened skin, which is fed upon by dozens or even hundreds of the youngest of the species, in the manner of caecilians. This nutrient source is critical to fueling the robust growth of young sauropods, which is impossible on only their nutrient-poor diet of leafy vegetation. Eventually they become nearly inert, and pass almost unnoticeably into death. This life stage leaves plenty of vertebrae for palentologists, although many are already brittle, and almost no limb bones.

    Brb, I’m going to write a dino-fiction novel that includes this.

  27. Andrew Stuck Says:

    So basically, if your average adult sauropod is “scary” in an “awesomebro” Godzilla or xenomorph sense, then these mysterious, only-ever-hinted-at super adults like your Barosaurus, Amphicoelias(?), and the Broome Sandstone trackmaker are the true Lovecraftian horrors of the dinosaur world…

  28. Nima Says:

    Yikes, yet another diplodocid turns out to be ridiculously oversized compared to the once-biggest dinosaur known! A sad day for Brachiosaur fans everywhere… :(

  29. bricksmashtv Says:

    Lol NOOOOO!!!! Poor Brachiosaurids! What will do when the neck of Barosaurus is almost as long as Sauroposeidon is tall?!

  30. Mike Taylor Says:

    Find a bigger brachiosaur, of course! :-)

    (I bet they’re out there.)

  31. Illiterate Scholar Says:

    I don’t know about the 50m estimate. More often than not, when we guess the size of fragmented dinosaurs, they tend to be too big. Heck, even for complete specimens, we get outrageous numbers until the fossil is mounted. Spinosaurus anybody? An even more hilarious example is the recent Das Monster Von Minden.

    How does it compare with the current champion “The Titanosaur” in the AMNH? That beast is about 34 meters in length. I’m guessing this Barosaurus is around that size, perhaps maybe even reaching 40 meters being really optimistic?

  32. Mike Taylor Says:

    I don’t know about the 50m estimate. More often than not, when we guess the size of fragmented dinosaurs, they tend to be too big.

    Really? What is your evidence for that? It’s not as though we have subsequently recovered any complete skeletons of super-giant sauropods and discovered them to be smaller than we thought. Are you sure that “They were probably smaller really” isn’t just an argument from personal incredulity?

    Speaking only of my own incredulity here, the idea of a 34m Barosaurus that is literally half neck seems way more unlikely to me than the 50m isometrically-scaled version. Like you I suspect that ontogenetic positive allometry of the neck may mean that the torso and tail are somewhat shorter proportionally than isometry would suggest. But the torso-and-tail of the AMNH specimen comes in at about 16.5m. For your 34m estimate of the overall length of the giant specimen to be correct, the torso-and-tail would have grown by only 50cm (3%) while the neck length doubled.

  33. Chase Says:

    I wonder if diplodocid necks grew longer every time they lied.

    Its truly awe-inspiring to think of how these animals once walked across the landscape. 40 meters is just insane.

  34. Illiterate Scholar Says:

    Mike Taylor – I forgot to mention I pulled that number out of my behind. A lot of it is from cynicism. It’s after many years of huge specimens (150 feet/40 meters for then “Seismosaurus) turning out to be a lot smaller, it made me into a skeptic.

    There is no science behind my 34 meters estimate, just a gut feeling. I’m more than happy to be proven wrong when we get a 50 meter Barosaurus.

  35. Shahen Says:

    How you know that BYU 9024 cervical is really C9 or not for example C10 or C11?

  36. Mike Taylor Says:

    We don’t. Based primarily on neural spine bifurcation, compared with what we see in the AMNH specimen, we think C9 is the best match. But it could easily be out by one or two positions in either direction.

  37. Niels Says:

    Suppose for the sake of argument we are being conservative and assume it’s two positions closer to the base. What order of mangitude would it be in then?

  38. Mike Taylor Says:

    C11 of the AMNH specimen has a centrum length of 775 mm, which makes BYU 9024 (at 1380 mm) 1.78 times as long. That would give a neck length of 15m and an overall length of perhaps 45 m.

    On the other hand, BYU 9024 might be a C7 or C8. In the AMNH specimen, C7 is not preserved, but C8 is 618 mm long. BYU 9024 is 2.23 times that long, so if it’s a C8 it’s from a neck 19 m long, in an animal perhaps 56 m long.

  39. L Says:

    What would be solid mass estimates assuming the scale up is accurate (ie te giant isn’t proportionally more robust)? This is either for the C9 main estimate, the minimal C11 one or the maximal C7 one? That would be fascinating to figure out. AMNH is iirc in the range of 12-20+ tonnes roughly, but it could vary. C9 suggests exactly twice the size and thus 8 times the mass, but the others are different. A C11 based estimate would come as (1.78×1.78×1.78=) 5.64 times as massive, while a C7 based one would come out at (2.23×2.23×2.23=) 11.09 times as massive. Whatever way you go, it’s pretty fascinating.

  40. Mike Taylor Says:

    What would be solid mass estimates assuming the scale up is accurate

    There are no solid mass estimates for sauropods.

    That said, your ratio calculations are right if the big Barosaurus was isometrically similar to the AMNH one. And if the AMNH specimen massed 12.5 tonnes, which is not an unreasonable guess, then that gives us a range of 70.5-138.6 tonnes for the big one.

  41. L Says:

    Thanks for the information regarding the Barosaurus mass debate. Oh and I realised I made the mistake of mixing the lost C7 with the present C8 (a C7 scaling is impossible currently but would probably have an even more spectacular result), so that was foolish on my part. One question I can’t seem to find the answer to is the height estimates for Barosaurus type specimen. The shoulder and hip heights appear a complete mystery, while it’s rearing up height is debatable. I saw one DA user (hardly reliable on a peer reviewed basis) claim the type specimen could reach a maximum of 14.5m tall at the right angle, but other than that, not much. Shoulder comparisons would definitely be useful on these scales.

  42. Mike Taylor Says:

    I don’t know if there have ever been shoulder- or hip-height estimates for Barosaurus. I don’t think that’s been done for many sauropods. Usually you get length and maybe mass, that’s it. As for a 14.5 m rearing height: that seems to be pushing it for an animal with only an 8.5 m neck. That requires 6 m from ground level to the neck-torso joint. Is it possible? Looking at this image, maybe just barely.

  43. L Says:

    Some have speculative estimates. I’ve only seen 3 proper size charts with AMNH type specimen on them, one on prehistoric and another on on your website. All suggested heights slightly over 4m at the shoulder and a little taller at the hip. Not conclusive my any means, but perhaps they give some advantage for height estimates? I agree that 14.5m could only be reached as an absolute maximum, but does 11-13m sound reasonable when rearing up for AMNH?

    Also, do you have an opinion on the ‘anti-shrink wrapping’ trend within paleontology reconstructions, such as this one: for Barosaurus itself or this for dinosaurs in general? I’m a fan of the movement as it makes the animals for more realistic and life-like than most old portraits. As skilled as Zach Armstrong’s skeletals are, his animals look like skeletons with skin and the absolute bare minimum soft tissue- to an extent no living tetrapod- not even a starving one replicates. Soft tissue that looks healthy, like adequate amounts of muscle, fat, cartilage, integument, keratin etc do help make more plausible reconstructions after all.

  44. Mike Taylor Says:

    Without having worked it through in detail, a rearing height of 11-13 m does seem reasonable for AMNH Barosaurus.

    But how high, in fact, is the mount in the entrace hall? Sauropedia says 15 m, but it doesn’t give a reference for that. the AMNH itself says only that it is “the tallest freestanding dinosaur mount in the world”, which I guess means they claim it’s taller than the 13.27 m Giraffatitan mount. Unless they mean something special by “freestanding” that rules the Giraffatitan out.

    Can anyone find a sourced figure for the height of the AMNH mount? (That’s height above its plinth, of course, not above ground level.)

    I am a big fan of anti-shrink-wrapping in particular, and the whole All Yesterdays approach in general. See my very positive review of the Conway et al. book, and also Matt’s comments (half way down the linked post).

  45. jerrold12 Says:

    The AMNH Activity for Grades K – 4, “How big were the dinosaurs?” gives a height of 50 feet, without indicating whether the plinth is included. file:///Users/CharlesDarwin/Downloads/dino_05_big.pdf . Not definitive perhaps, but at least sourced! (I’m often there as an AMNH Tour Guide.)

  46. Mike Taylor Says:

    Well, that is some information, I guess! Isn’t it odd, though, that this impressive figure (whatever its true value) is not widely publicised? It’s surprisingly hard to find for the Berlin brachiosaur, too.

  47. Mike Habib Says:

    This is just epic.

    In terms of the mechanical support issues (re: giraffes and their suite of adaptations for comparative puny necks), the orientation of the neck has enormous effects on energy expenditure, required cardiovascular pressures, and support (both static and dynamic). So one problem for giraffes is that they keep a highly elevated neck posture (and yes, many sauropods likely had elevated necks, too, but “Supersaurus” might have kept its neck and head at a relatively low inclination).

    Perhaps even more important, though, is that giraffes are stuck with a bunch of mammalian features that just make huge necks difficult to sustain without extra adaptations. They lack pneumaticity, of course, so maximum stiffness:weight ratios for the skeleton are comparatively mediocre. Giraffes also maintain their neck position largely through “brute force” mechanisms (large muscles, thick ligaments, and tight vertebrae) that are not particularly mass efficient.

    Sauropods had a tendency to use more efficient ligamentous attachments, high stiffness:weight bones, and (if I’m right about the cervical tendons – see SVP 2015) spring systems to help maintain neck posture and dampen oscillations at very low cost. There are also potential integument adaptations available to archosaurs that mammals are basically excluded from using. These might have been used by sauropods to improve neck stability and blood flow (but that’s completely speculative at present).

  48. Mike Taylor Says:

    It does astonish me to think that even puny giraffes have this suite of special anatomical adaptations just to handle a 2.2 m neck and the correspondingly long legs that go with them. It was horrifying enough back when we thought that Diplodocus could go to 6 m. But 17 m? Its a whole different ball game.

  49. biologyinmotion Says:

    Yeah, I keep staring at that 17 m neck reconstruction, and it’s just surreal. This sort of thing makes a biomechanist drool.

    I suppose there’s also a perspective component: since the only large, terrestrial, long-necked animal alive today is a giraffe, we see its mechanical support and pressure-modulating adaptations as this enormous, specialist suite. If we had dozens or hundreds of giraffe-like animals around, though, we probably wouldn’t think much of it.

  50. Ben Says:

    Norrell et al. 1991 also puts the Barosaurus mount at 50 feet. Perhaps “freestanding” refers to the fact that it’s a cast with an internal armature?

    Norrell, M.A., Dingus, L.W. and Gaffney, E.S. (1991). Barosaurus on Central Park West. Natural History 100:12:36-41

  51. Ronald Says:

    Mike Taylor Says:
    September 20, 2016 at 7:31 am
    Find a bigger brachiosaur, of course! :-)
    (I bet they’re out there.)

    ‘Brachiosaurus’ nougaredi perhap? Breviparopus?

  52. Mike Taylor Says:

    IIRC, there is is no compelling reason to think that “Brachiosaurusnougaredi is Brachiosaurus, or indeed brachiosaurid. If it’s the specimen I am thinking of, it have a very long, but proportionally very narrow sacrum; and published figures are not good enough to determine whether that is due to crushing.

  53. Michael Habib Says:

    Mike Taylor Says:
    September 20, 2016 at 7:31 am
    Find a bigger brachiosaur, of course! :-)
    (I bet they’re out there.)

    Perhaps a bit bigger, but I wouldn’t be surprised if the largest brachiosaurs known are near the max size. Maximum height should be more constrained than maximum length in sauropods. So taxa using elevated neck postures would be necessarily be more constrained overall than others.

  54. dale Says:

    A 17 M neck is far from outlandish if you take an even fleeting glance at Mamenchisaurus sinocanadorum. A complete skeleton. Compare it to other similar specimens of that species. I’m sure you’ll get more defined answers that don’t require such wild speculation. However, that said, you can only shake your head in awe. Sauropod majesty just keeps climbing.

  55. William Miller Says:

    Why would maximum height be more constrained? Blood pressure issues?

  56. Yes indeed. Blood pressure and heart energetics become an issue for height, but are much less affected by length.

  57. anon Says:

    I just noticed that 17m is a little longer than an entire Elasmosaurus.

  58. I have heard from some people that sauropod necks increase significantly in length proportionately as they age, one of whom posted this reconstruction of the giant Barosaurus specimen ( to show what they think a Barosaurus of this size should look like. However, I have also heard from someone who is usually a very reliable source that this is in fact not the case. I was wondering, could you shed some light on this matter? Should a Barosaurus of this size look more like the picture I linked, or have similar proportions to those of smaller Barosaurus specimens?

  59. Mike Taylor Says:

    Hi, Christian. I’d like to see your references (for pro and con ontogenetic neck allometry) before expressing a very firm opinion myself. For now, here’s I’m prepared to say: we have a very poor record of ontogenetic series of sauropod necks: there might be one for Shunosaurus but if so it’s not been documented; and I don’t know of any others. So that means, right off the bat, we don’t really know anything about neck allometry through life — though we can infer that juveniles likely had proprtionally short necks, as that’s the case in giraffes. Beyond that, it’s speculation, though hopefully at least informed speculation. Individual cervicals seem proportionally short in juveniles — Matt may have more to say about what that can tell us statistically.

    What we do have is Parrish’s (2006) study showing positive allometry in neck length across species of sauropod. But that’s a decade old and starting to show its age. It works from a smallish sample of 20 or so specimens — but does show very strong correlations: log (neck length) =~ log(torso length)^1.35 with r^2 = 0.94. It would be good to see something similar done in more exhaustive detail with a larger data-set, and using phylogenetically indepedent contrasts; and better still, using multiple individuals of a single species. But I don’t know of anyone who’s working on it.

    So my guess is that the giant Barosaurus neck probably had a proportionally small body. But it really is pretty much guesswork at this stage.


    Parrish, J. Michael. 2006. The origins of high browsing and the effects of phylogeny and scaling on neck length in Sauropodomorpha. pp. 201-224 in: M. T. Carrano, T. J. Gaudin, R. W. Blob, and J. R. Wible (eds.), Amniote Paleobiology: Phylogenetic and Functional Perspectives on the Evolution of Mammals, Birds and Reptiles. Chicago: University of Chicago Press.

  60. […] 2016, an SVPCA abstract was released by Mike Taylor & Matt Wedel of SVPOW! In this abstract,  they refer (rather convincingly) a pair of three cervicals (from BYU quarry […]

  61. […] sauropod cervicals got longer through ontogeny came up in the comment thread on Mike’s “How horrifying was the neck of Barosaurus?” post, and rather than bury this as a comment, I’m promoting it to a post of its […]

  62. Matt Wedel Says:

    There’s already a trackback here, but just in case: the question of whether sauropod cervical vertebrae got longer over ontogeny is answered (in the affirmative) in this post.

  63. Shahen Says:

    Parrish, J. Michael. 2006. The origins of high browsing and the effects of phylogeny and scaling on neck length in Sauropodomorpha. pp. 201-224 in: M. T. Carrano, T. J. Gaudin, R. W. Blob, and J. R. Wible (eds.), Amniote Paleobiology: Phylogenetic and Functional Perspectives on the Evolution of Mammals, Birds and Reptiles. Chicago: University of Chicago Press.

    Do you have this paper? Is possible to send me pdf in my mail? very thanks

  64. Mike Taylor Says:

    It’s on the way. Shahen!

  65. […] A correspondent — I will not divulge his or her name unless the person in question chooses to reveal it — had looked over the slides for our 2016 SVPCA talk on new Barosaurus specimens, which claims that Jensen’s Dry Mesa “Supersaurus” cervical BYU 9024 actually belongs to Barosaurus. […]

  66. […] to more evidence of rather extreme gigantism in some other sauropods (Apatosaurus & Barosaurus) can push the limits even […]

  67. […] Gigantický krk barosaura(?) dlouhý 17 metrů, tedy o poznání vyšší než nejvyšší smontovaná kostra dinosaura na světě (Giraffatitan brancai v Berlínském přírodovědeckém muzeu s výškou 13,2 metru). Kredit: Taylor a Wedel, web SVPOW. […]

  68. Cosmin Macasoi Says:

    I think the sauropods were meat eating dinosaurs or piscivore and erroneously most people believe they were herbivorous. They were slow animals, that is right, but they used their long necks to catch the prey – especially fish- with a sudden and fast motion of the neck. That s why they developed longer and longer necks. It s a evolutive convergence with plesiosaurs, another fish eating animals. It s true, they don t have specialized dentition, but that is because these animals didn t rip the prey apart, they swallowed it whole, without chewing it. They were in essence giant aspirators of fish and small prey- small theropods, small ornithopods, lizards, birds, pterosaurs. Or at least omnivorous like brown bear. They could also hunt by hitting the prey with their tails like komodo dragons. What do you think about this hypothesis ? Is there solid evidence for that?
    Kind regards,

  69. Mike Taylor Says:

    It’s a neat idea, but I’m afraid not. Two lines of evidence (at least) rule it out: one is that fossilised gut contents in prosauropods and sauropods have all been vegetation; the other is that the lack of specialised dentition would make it more difficult to catch prey. Even among those that swallow their prey whole, piscivores invariably have sharp teeth for prey capture.

  70. Dale McInnes Says:

    If BYU 9024 really is a C9, larger than anything yet discovered, does one really think that this is a trophy specimen or for that matter even the largest. After
    all, we have only one.

  71. Mike Taylor Says:

    It’s from a very seriously big animal, yes — Amphicoelias fragillimus size, which is part of why I am still inclined to think that was real, and not the result of a typo.

  72. Matt Wedel Says:

    Maybe I’m mistaken – and if so, Dale, please set me straight – but I read Dale’s comment as pointing in the other direction: given that our sampling of these gigapods is so pathetic, why do we think that this particular vert is as big as they got? Wouldn’t that be absurdly lucky, that with a comparative handful of specimens, we just happened to get the literally one-in-a-million or one-in-a-billion individuals?

    Like I said, I don’t know if that’s the question you intended, but it’s a pretty great question nonetheless. As I wrote back in 2008:

    The biggest known whales really are probably close to being the biggest representatives of their species. The individuals listed above are the largest known from a sample of more than 300,000 blue whales killed in the early 20th century. That’s a big pool. Supersaurus and Argentinosaurus are both known from two specimens, and Bruhathkayosaurus and A. fragillimus from one specimen each. The chances that these largest-known sauropods are really the largest sauropods that ever lived is vanishingly small.

  73. Mike Taylor Says:

    Oohhhhh … Yes, of course, you’re right, that is what Dale meant. Sorry for completely misreading your interesting observation, Dale, as a rather commonplace one.

    And I agree wholeheartedly with Matt about the probability of these being the biggest individuals is vanishingly small. It’s part of why I still have a sneaking feeling that the biggest sauropods might have been heavier than the biggest whales — as crazy as that sounds. (And, Matt, you can quote me on that in your forthcoming talk.)

  74. Mike Habib Says:

    For what it’s worth, I don’t think it sounds crazy at all that the largest sauropods may have been heavier than the biggest whales. If weight support is not the limiting factor (and I don’t think it is/was), then there’s no particular reason that the largest animals must be aquatic. Marine animals can get bigger than terrestrial animals within the context of certain morphotypes, but that’s a different issue.

  75. Mike Taylor Says:

    Well, plenty of people have thought that weight-support was the limiting factor, going back at least to Economos (1981) who argues that 20 tonnes(!) was a fundamental limit. But of course gravity affects a lot more than the mere ability to stand up — for example, pumping blood up and down.

    This reminds me that I really, really must get on with a long-running project I’ve had in mind entitled “Why elephants are so small”.

  76. biologyinmotion Says:

    True, that’s been a common thought; I just don’t think it’s true. Large terrestrial animals can use axially loaded limb skeletons (among other tricks) for weight support, and in that configuration, bone is roughly 5x stronger than concrete. Nerve lengths and vascular effects are more likely to be size limiting for most gigantic animals, I suspect. I love the project title!

  77. Mike Taylor Says:

    Sure; but bone is not the only material involved (as though you need to be told that). See for example my very very very preliminary 2005 presentation Upper limits on the mass of land animals estimated through the articular area of limb-bone cartilage.

    (The project title is of course a followup to “Why giraffes have short necks”.)

  78. biologyinmotion Says:

    Good point, but I suspect (based, in part, on your preliminary work on the articular cartilage, and in part on my own numbers) that weight support was still not as limiting as things like cardiovascular systems. You make a good argument that joint structure and cartilage may have limited athletic performance, but the ceiling on pure weight support, assuming limited athletic performance, was probably quite a bit higher.

  79. […] post started out as a comment on this thread, kicked off by Dale McInnes, in which Mike Habib got into a discussion with Mike Taylor about the […]

  80. Mike Taylor Says:

    I agree; all I’m saying right now is that when we’re looking at raw weight-support limits, bone is the wrong material to be thinking about. But there is a lot of work to be done on this. Who knows, the limiting factor might turn out of be bladder size :-)

  81. Dale McInnes Says:

    Yes. That is the question I meant. You got it right. Not only do we have vanishingly small sample numbers but we also inherited a vanishingly small time scale in trying to figure out how it was even possible for such behemoths to acquire such size. Nature has had over 150 million years to perfect her creations.It shouldn’t be any surprise that we were caught so off guard in the last 200 years dealing with “Sauoropodology”. It reminds me of an article I read some time ago (no reference unfortunately) of why certain drugs don’t work very well when compared with “Natural” ones despite Medicine’s ability to copy them precisely as well as creating identical folding of the proteins.Nature can fold them much tighter thus making them vastly more effective (something we apparently haven’t learned to do yet ??) Nothing is impossible with sauropods unless Nature says so. I still shake my head over a 17 M neck but obviously we are going to have to brace ourselves for longer ones.
    “Life Always Finds A Way!!!”

  82. William Miller Says:

    What about thermal effects? Could a 500 ton warm-blooded sauropod keep from cooking itself through metabolic heat and digestive fermentation?

  83. Mike Taylor Says:

    I doubt that would be a limiting factor: for such bulky animals, sauropods had pretty high surface-area-to-volume ratios, due to the long necks and tails and even legs. Add in the possible cooling effects of the air-sac system, and I think other factors would have become limiting before thermal overload. AFAIK this has never been addressed quantitatively, though — you’re just getting the benefit of my fallible intuition here.

  84. Matt Wedel Says:

    I don’t know if there has ever been an “all up” thermoregulatory model of a sauropod that was realistic.

    What we’ve had are models of sauropod heat retention that are horribly unrealistic, assuming that a sauropod was a solid bullet of meat with no internal air spaces at all, such as Paladino et al. (1997). This is especially painful because simple thermodynamic models like that one are governed by a sort of “largest effective cylinder” logic. In these models, the internal air spaces don’t have to be heat sinks necessarily, they just have to be something other than heat-generating meat. I doubt that in a real live Apatosaurus there was any tissue more than 1 meter from an internal or external air space*, so a realistic Apato thermal model should be a bunch of small cylinders and slabs of meat stitched together, not one big fat meat balloon.

    * The biggest chunk of non-aerated meat was probably either the intestines or the thigh muscles – but even there, there were probably diverticula. The amount of air in a living ostrich is kinda insane. The viscera are have extensive soft-tissue pneumatization by visceral diverticula that pass among the loops of intestine, not to mention the thoraco-abdominal air sacs jacketing them on either side. Even the big blocks of thigh muscles have intermuscular diverticula running between and among them.

    We’ve also had models of sauropod respiration that suggest that the long trachea could not help but function as a pretty efficient heat exchanger, such as Sverdlova et al. (2013).

    Given that the models that predict that sauropods would have overheated are pathetically easy to falsify, and the models that suggest that sauropods were pretty good at dumping heat are not yet falsified, I lean pretty strongly toward heat not being a limiting factor, even for fully endothermic sauropods.

  85. William Miller Says:

    OK, cool, I didn’t know that had been studied.

  86. […] diplodocines (Supersaurus, maybe Barosaurus) […]

  87. Gray Stanback Says:

    I’m not fully convinced that this specimen is a Barosaurus and not a Supersaurus. The neural spine seems to be broken, and if it were complete it seems as though it would have been a different shape than it is in Barosaurus. Also, Barosaurus is not known from the Dry Mesa Quarry, where these fossils were collected, but Supersaurus is. Finally, it should be noted that Supersaurus itself was considered to be a close relative of Barosaurus when it was first discovered, and only later was determined to be closer to Apatosaurus.

  88. Mike Taylor Says:

    Hi, Gray, thanks for your thoughts. I don’t recall thinking, when Matt and I were looking at BYU 9024, that the neural spine was broken. Perhaps you can sketch where you think the “complete” spine would be, and how you think it resembles that of Supersaurus.

    For that matter, I’d be interested in where you got your notion of what a Supersaurus cervical ought to look like — I can’t think of much published material. One of the cervicals illustrated in Lovelace et al.’s 2005 poster on “Jimbo” looks pretty complete in lateral view, but it also looks less similar to BYU 9024 than the AMNH Barosaurus cervicals do.

    Also, though Barosaurus has not been reported from the Dry Mesa Quarry, that most certainly does not mean it’s not there. It turns up in nearby quarries, and it’s perfectly possible that some of the known material belongs to that genus and has been misassigned (or indeed that material assigned to Barosaurus has been accessioned, but not published).

    If you want to make a solid case that BYU 9024 is not Barosaurus, a good place to start would be to show why the many shared characters illustrated in the slides of Taylor and Wedel (2016) are incorrect.

  89. Gray Stanback Says:

    These aren’t my thoughts; I’m just repeating what I heard Nima say here:

  90. Mike Taylor Says:

    Ah, right. Nima is a bright guy and a very clean artist, but does have a tendency to make pronouncements that sound much more authoritative than he has any right to be. Unless someone can tell me different, I don’t believe he’s ever seen the specimen in question; and he certainly hasn’t engaged with our SVPCA slides. Let’s hope he does so at some point.

  91. Matt Wedel Says:

    I’ll go further: the neural spine is not broken. I’ve heard this idea from more than one source and its flatly wrong. I’ve spent hours with that specimen over more than a decade and the spinoprezyg and spinopostzyg laminae are intact all the way to the metapophyses, which are themselves intact. Looks like I need to do a post showing that this is true so we can abandon this myth once and for all.

  92. Gray Stanback Says:

    Is it possible that, rather than being directly to scale with smaller and more complete Barosaurus skeletons, this individual simply had a normal-sized body and an outrageously long neck, Mamenchisaurus-style?

  93. Mike Taylor Says:

    Not quite, but it’s certainly possible that, while the neck was twice the size of the the AMNH specimen, its torso exceeded the AMNH torso’s length by a factor of less than two. There’s some evidence that neck length in sauropods was positively allometric relative to torso length. See Parrish (2006) — but note that it’s badly in need of updating.


    Parrish, J. Michael. 2006. The origins of high browsing and the effects of phylogeny and scaling on neck length in Sauropodomorpha. pp. 201-224 in M. T. Carrano, T. J. Gaudin, R. W. Blob, and J. R. Wible (eds.), Amniote Paleobiology: Phylogenetic and Functional Perspectives on the Evolution of Mammals, Birds and Reptiles. Chicago: University of Chicago Press.

  94. Gray Stanback Says:

    So if that’s the case, might we be looking at a total length of perhaps 115 feet and a weight of around 75 tons?

  95. Mike Taylor Says:

    I don’t want to speculate before I’ve had time to run numbers.

  96. Ronald Says:

    Slightly off-topic, but has anybody mentioned the recent study “The real Bigfoot: a pes from Wyoming, USA is the largest sauropod pes ever reported and the northern-most occurrence of brachiosaurids in the Upper Jurassic Morrison Formation”?
    The title says it all.
    In particular, KUVP 129724 is a large brachiosaur, the largest brachiosaur pes found.

  97. Mike Taylor Says:

    Yep: Matt was the handling editor for that paper, and I was Reviewer 1. So we both read it in detail, and we’re both happy with it (and happy about it).

  98. Matt Wedel Says:

    So if that’s the case, might we be looking at a total length of perhaps 115 feet and a weight of around 75 tons?

    Length of 115 ft sounds ballpark plausible. I dunno about the mass. IIRC Scott Hartman estimated Supersaurus at around 40 tons for a ~120ft animal, using methods that looked sound to me. I doubt that a barosaur of equivalent length would have weighed much more (despite the etymology of the name!) – BUT it’s worth keeping in mind how little we know about these things and how large are the error bars on our estimation methods.

  99. Mike Taylor Says:

    Here’s my back-of-the-envelope mass estimate for BYU 9024.

    1. Wedel (2005) estimated the mass of the Carnegie Diplodocus as 12.5 tonnes, and that’s probably as good an estimate as we have for that individual.
    2. The AMNH Barosaurus is a similar animal of a very similar size, so probably weighed about 12.5 tonnes, too.
    3. The BYU 9024 cervical is twice as long as the analogous cervical in the AMNH specimen, so its neck was probably twice as long.
    4. The rest of its body was probably roughly in proportion with the neck (ignoring Parrish for the moment), so twice as long as the AMNH Barosaurus.
    5. The animal’s height and width probably varied linearly with its length, i.e. were double in the BYU specimen what they were in the AMNH specimen, yielding a total body volume 2x2x2 time that of the AMNH specimen.
    6. … and eight times 12.5 tonnes is 100 tonnes.

    I you were following closely, you’ll have noticed that every step, except the trivial multiplication of the the last, contained the weasly disclaimer “probably”. What this means is that my estimate of 100 tonnes contains at least five degrees of uncertainty. The true number could easily be anything between 50 and 150 tonnes, maybe even something outside of that range. But 100 tonnes is my best guess, given the evidence we have.

    BTW., I find the length estimate of 115 feet pretty unlikely. Remember, the neck is the one part of BYU 9024 we know about, and our best data says it was twice as long as the 8.5 m neck of the AMNH specimen. That’s 17 m = 56 feet for the neck alone. That leaves only 56 feet for the torso and tail, which is a bit less than the torso-and-tail of the Carnegie Diplodocus. That would require some frankly ludicrous allometry.

  100. Matt Wedel Says:

    I can’t fault any step in that chain. It just scares me.

  101. Steve O'C Says:

    Count this as a somewhat throwaway comment but looking at skeletal reconstructions of the AMNH Barosaurus, with its long legs and shallow dorsal vertebra, it certainly ”looks” somewhat juvenile/subadult in proportions. (for example, check out Woodruff et. al. 2017, figure 19; although I’m not sure how literal to take it) I’d imagine, assuming you’re correct about BYU 9024, that it would look deeper in the torso then the AMNH specimen.

    Whilst there seems to be evidence of positive neck allometry in sauropods, like what’s seen in the juvenile Camarasaurus vs adult specimens. Is there much evidence whether neck length allometry is linear or whether at some age it starts to plateaus?

    For example, a 5 year old sauropod might have a proportionally shorter neck than a 30 year old, but would a 20 year old have a proportionally shorter neck than a 30 year old?

    I found some examples in a paper (which I need to dig up) of how neck growth might be non-linear; in female giraffes increasing neck mass plateaus at a certain age. In males neck mass scales linearly with age, (although, the paper said it’s debatable whether this equates to a longer neck.)

  102. Mike Taylor Says:

    Fascinating thoughts, Steve, thanks for bringing them up. If you have the reference for that giraffe paper, I’d really like to see it.

  103. Matt Wedel Says:

    For example, a 5 year old sauropod might have a proportionally shorter neck than a 30 year old, but would a 20 year old have a proportionally shorter neck than a 30 year old?

    That’s an excellent question. I doubt that we have the data to answer it. There are few enough sauropods for which we have even one reasonably complete neck for a single ‘adult’ individual. One more on the very tall stack of things we’d like to know if possible.

  104. Steve O'C Says:

    I’ve managed to locate it; Simmons and Scheepers 1996. ‘Winning by a neck….. ‘

    In searching for that paper I’ve also found Mitchell et. al. 2013 ‘Growth patterns and masses of the heads and necks of male and female giraffes.’ Which does a similar analysis and doesn’t find a drastic difference between male and female neck mass allometry… that probably rules out that example.

  105. Mike Taylor Says:

    Ah, that old favourite — just shows that it’s been too long since I re-read it. Thanks for the pointer.

  106. Paes Says:

    On the topic of giant necks, Gillette (1996) gives ”Seismosaurus” a neck that’s over 11 meters long going by the size of the rest of the body; could that still hold up? Nowadays most reconstructions give it a D.carnegii-like neck (in Scott Hartman’s it’s ~8 m), and the website says ”Positions and anatomy of the four neck vertebrae that we collected are conjectural; these bones were isolated in the quarry and heavily eroded.” but knowing that the animal was closely related to D.carnegii, they wouldn’t have given it such different neck proportions without a good reason, I think.

  107. Mike Taylor Says:

    Sadly, the consensus on Seismosaurus is in, and it’s almost certainly “only” a very big Diplodocus. If I’m remembering this rightly, the original work was mistaken in two important aspects. First, it concluded Seismosaurus was distinct on the basis of the odd shape of the distal ischium; but further preparation showed that the protrusion was a piece of vertebra that had adhered to the end of it. And second, a series of proximal caudal vertebrae were interpreted as being from a more distal position, implying they were from a larger animal. Put it all together, and 8 m for the neck is probably not far off.

  108. Paes Says:

    Oh, I knew about the sinking of the genus; it’s still mostly thought to be a different species, either D.longus or D.hallorum, though, right?

    What I was wondering is if the proportionally longer neck they gave it (scaling the torso of the old skeletal to be the same size as in Scott’s, since the dorsal interpretation doesn’t seem to have changed, the cervicals they included as preserved ended up way too long to fit on an 8-meter neck) might be a feature of that species, or if the huge neck was an artistic decision to make the whole thing look larger. I think there aren’t any measurements for those cervicals anywhere, assumedly because of them being in bad conditions as Gillette commented, so maybe it’s impossible to tell.

    Referring to the neck on this one.

  109. Mike Taylor Says:

    Diplodocus species: now that hayi has been removed to its own genus Galeamopus, we’re basically left with the type species longus and the more widely known carnegii. There are no real differences between them and IIRC even Hatcher at some point more or less admitted he only raised carnegii because his sponsor (Carnegie, natch) would have been miffed if he hadn’t. What separates hallorum from the longus/carnegii complex? Not much, that I know of anyway. Although I tend to err on the side of splitting sauropods, I’m inclined to think we probably only know of one species of Diplodocus.

    Not 100% sure about that Gillette reconstructions, but I seem to recall that the curious proportions are because they scales the limbs correctly (i.e. according to the preserved appendicular elements) but the vertebral column too big (based on those misassigned caudals). So much as I love that reconstruction, I’m afraid it’s not supported by our best science.

  110. Paes Says:

    Alright, thanks!

  111. […] that the neck alone of BYU 9024 would have been 17 m long: longer than most complete sauropods and substantially taller than the mounted Giraffatitan skeleton in Berlin. The length of the whole animal is harder to predict, even if we assume isometry, but if […]

  112. Krzysztof Stuchlik Stuchlik Says:

    Is possible that giant BYU 17386 humerus 171 cm and BYU 13744 128 cm ulna also are Barosaurus? the same specimen with BYU 9024 ? If checked Hartman Barosaurus is so possible and close measurements

    Lovelance et al 2008 Morphology of a specimen of Supersaurus (Dinosauria, Sauropoda) from the Morrison Formation of Wyoming, and a re-evaluation of diplodocid phylogeny

    What do you think ?

  113. Mike Taylor Says:

    It’s a good thought. But I’ve never seen BYU 17386 or 13744, and I know nothing about them beyond the very brief mentions in Lovelace et al. (2008) — so I really couldn’t say. Ray Wilhite might know, and he does sometimes drop in here, so let’s hope he does!

  114. Krzysztof Stuchlik Says:

    Ok then just I’m hoply waiting for new Barosaurus post :)

  115. Mike Taylor Says:

    Well, I hope Matt is able to get back out to BYU some time and look at the appendicular material assigned to Supersaurus; but the chances of my doing so are remote for the foreseeable future.

  116. […] cervical originally referred to Supersaurus, but which more likely belongs to Barosaurus, and an ungodly huge one. That critter might–might–have had a 17-meter […]

  117. […] up from the 178.5cm femur of CM 3018). I’m deliberately not dealing with Maraapunisaurus or horrifying hypothetical barosaurs […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: