This serves as a follow up from my journal entry: If you ask me about Spinosaurus...
I may have made mistakes on which parts are preserved(white) or inferred(gray), because the color-coded 3d model has it's orange(inferred from holotype) and red(neotype) so similar in color that at times I couldn't even tell them apart!
Seriously, Ibrahim and co. really should have used more contrasting colors.
The tail is based on the notion by Jaime Headden that none of the caudal vertebrae are further anteriorly than the D25.
"As it is, none of the caudal vertebrae seem to be anterior further anterior than about half of the tail’s length, assuming about 50 caudals."
Since neither the paper nor it's supplement actually tries to prove it's placement of the holotype's tall, isolated posterior neural spine(it doesn't even actually mention it), I placed said spine as an anterior caudal. In addition, Cau makes an argument that the tall neural spine is likely an anterior caudal based on the spines' geometry.
The legs are displayed larger than in the Ibrahim et al. paper, due to the Ibrahim et al. model having splayed legs. A quadrupedal Spinosaurus is very unlikely, as Jaime Headden points out:
"First, the forelimbs, if they are anything like normal theropod forelimbs, are completely unsuited for weight-bearing. Rather than arranged vertically, the shoulder blades are aligned across the ribs in such a way that any quadrupedal stance would shove the shoulders deep into the neck and likely behead a Spinosaurus faster than Sean Bean in his next deat–film. Rather, they were suspended as in other theropods free from the ground, to dangle as they please.
Second, the authors even provide a furcula (recovered with the spinosaur Suchomimus tenerensis) for the shoulder, and this would have kept the shoulders from moving independently of one another, meaning it is unlikely the arms were permitted to walk in anything like a natural gait or a “typical” quadruped.Third, the forelimbs while massive have not been described in sufficient detail and appear to possess characteristic typical of other spinosaurs. If they are anything like the forearms of Baryonyx walkeri, they’d lack almost any quadrupedal attributes, especially in the metacarpus, manus, and especially in the phalanges. The authors propose a semi-knuckle-walking locomotion style, but all extant knuckle-walkers have specific adaptations of the arm to permit this (and so it appears in extinct ones, even if you include chalicotheres which may not have been knuckle-walkers), including the presence of thickened, columnar first phalanges and strong curling joints for the other phalanges to bring them out-of-the-way, as well as thickened, solid wrists. Theropods, rather, have flexible wrists due to the semilunate carpal hinge and would likely have been unable to bear weight translating through the upper into the lower arm, much less straighten the limb passively enough to bear weight in such a fashion that a knuckle-walker would require."
I used a dark gray to represent the skin in the sail. Only the topmost 1/3 of the spine has the texture, which, along with sharp edges on the top, according to Ibrahim et al., suggested a sail.
"proximal one-third of dorsal neural spines textured externally by vertical striae"
So you have ~2/3 left for humps and ridges.
Using the chameleon analogue used in the Ibrahim et al. paper, the lowermost 1/3 would be embedded in muscle, likely buried in the back. My flesh silhouette doesn't differentiate between the back and the ridge.
The visible spine in life would be the upper ~2/3, with half of it as a sail and the lower half a transitional ridge.
FSAC-KK 11888, the neotype, is a juvenile specimen estimated at around ~15-19 years of age.
IPHG 1912, the holotype specimen destroyed in WW2, overlaps with FSAC-KK 11888, and seems to be quite small, probably a juvenile or a very small adult. I couldn't get Stromer's original description, luckily the plates are available in the interwebs. I expected this to be a ~15-meter specimen or something, but it turned out much smaller.
MSNM v4047 is a bit trickier though, since it overlaps with neither IPHG 1912 nor FSAC-KK 11888. I had to pull out another scientific source for that one.
This also has huge implications on how we should reconstruct other spinosaurines, such as Oxalaia, Irritator, and Angaturama. Rather than the old Baryonyx/Suchomimus-like models assumed for them, they would very likely have been short-legged "water dragons".
Measurements for IPHG 1912:
Hip height: ~1.75 meters
Total height: ~3.2 meters
Tip-to-tip total length: ~11.09 meters
Axial length: ~11.68 meters
Measurements for FSAC-KK 11888:
Hip height: ~1.81 meters
Total height: ~3.38 meters
Tip-to-tip total length: ~11.5 meters
Axial length: ~12.12 meters
Measurements for MSNM v4047:
Hip height: ~2.55 meters
Total height: ~4.67 meters
Tip-to-tip total length: ~16.19 meters
Axial length: ~17.06 meters
MSNM v4047's humerus turns out smaller than that of NMC 41582, using the postcranial and forelimb proportions implied in the Ibrahim et al. reconstruction. This is probably an indication that NMC 41582 is probably larger than MSNM v4047, but by how much is very uncertain.
Ibrahim, 2014, "Semiaquatic adaptations for a giant predatory dinosaur" + supplementary materials
Headden, 2014, The Bite Stuff - The Outlaw Spinosaurus
Cau, 2016, Theropoda blog - Spinosaurus geometricus
UPDATE(9/22/2014): Tweaked the scale of FSAC-KK 11888. Total length was increased from ~900 centimeters to ~959 centimeters as a result.
UPDATE(9/24/2014): Tweaked the dorsal vertebra of IPHG 1912 and the scale of FSAC-KK 11888. Also remeasured axial lengths using the Google Earth technique, but more closely following the vertebral row this time(the last one was a bit rough).
UPDATE(10/24/2016): Remade the tail and tweaked some of the bones to match the Ibrahim et al. measurements. This time, I kept the same caudal count as the Ibrahim et al. model (55 from my count, although it's hard to count due to the terminal caudals being blurry). The tail is longer due to having to modify the anterior caudals to accommodate the neural spine. I also smoothed out the M-curve since it was actually unjustified, as Jaime Headden pointed out in the comment section.
Previous version for comparison
Measurements (axial lengths) for the previous version:
FSAC-KK 11888: ~9.34 meters, IPHG 1912: ~10.93 meters, MSNM v4047: ~16.48 meters
UPDATE(10/20/2016): Fixed an issue with the feet and some minor scaling tweaks on the head size.
Measurements (tip-to-tip standing lengths) for the previous version:
FSAC-KK 11888: ~9.92 meters, IPHG 1912: ~11.54 meters, MSNM v4047: ~17.33 meters
UPDATE(12/19/2016): Rescaled body : [limb + hip] proportions after a conversation with prompted me to recheck and correct those.
Apparently, the vertebrae of FSAC-KK 11888 were actually longer than those of IPHG 1912, and only had smaller measurement figures due to them not corresponding to the same measurement.
All together, these have the effect of blowing up FSAC-KK 11888's body size by a substantial amount and actually making it larger than IPHG 1912. The hip height stays the same since the legs were already correctly scaled based on the measurements, while the other specimens get reduced heights.
Measurements for the previous version:
FSAC-KK 11888: ~3.02 meters total height, ~9.94 meters standing length, ~10.09 meters axial length
IPHG 1912: ~2.06 meters hip height, ~3.44 meters total height
MSNM v4047: ~3 meters hip height, ~5.02 meters total height
UPDATE(3/22/2017): Detailed the skull and fixed up the body-limb cross-scaling after triple-checking the measurements again. Remade the cervical ribs and adjusted the neck flesh envelope. Also fixed the randomly elongated atlas that I apparently gave IPHG 1912 and MSNM v4047 back in 2014 for some reason.
Measurements for the previous version:
FSAC-KK 11888: ~3.45 meters total height, ~12.6 meters standing length, ~12.79 meters axial length
IPHG 1912: ~1.62 meters hip height, ~3.08 meters total height, ~11.56 meters standing length, ~11.73 meters axial length
MSNM v4047: ~2.37 meters hip height, ~4.52 meters total height, ~16.87 meters standing length, ~17.12 meters axial length
UPDATE(4/9/2017): Reposed limbs to be more in line with my new standards for my skeletal reconstructions, and finally decided to give it the raised neck posture inferred from Sigilmassasaurus.
Measurements for the previous version:
FSAC-KK 11888: ~1.78 meters hip height, ~3.35 meters total height, ~11.94 meters standing length
IPHG 1912: ~1.72 meters hip height, ~3.17 meters total height, ~11.51 meters standing length
MSNM v4047: ~2.5 meters hip height, ~4.62 meters total height, ~16.81 meters standing length
Even if the hinder legs of spinosaurus are shorter than those of other theropods, they should still be much longer than its fore limbs. You can verify that from the first documentary movie of the quadruple spinosaurus, which takes Ibrahim's paper as reference.
On the other hand, from your reconstruction image, I'm not looking at an apex semi-aquatic predator but a funny sailboat mud fish from fairy tale.
Proportional arm size is a tad speculative since the NMC 41852 humerus is an isolated element, the one here is taken from the measurement table from the Ibrahim et al. supplementary.
Amazing skeletal. What do you think of the proposed "mud crawler" hypothesis? As in Spinosaurus would slide/scoot around like seals, grebes, and loons and like what crocs do sometimes. Even Andrea Cau is on board with this notion. I hear a lot about how it's hip could not properly support weight so any form of normal bipedalism is unlikely, and of course you already covered the problem with quadrupedality/knuckle walking. Even the legs and feet themselves are incredibly small, so even they couldn't properly support weight. It's rather front heavy as well, especially since its bones are dense due to an aquatic lifestyle. The posterior (hind limbs and back) are just barely longer than the anterior (hind limbs to front), which is odd for any bipedal animal. All other Theropods and just tailed bipedal animals in general, have tails much longer proportionally. This is especially true for Theropods, because I personally looked at the proportions of various Terrestrial Theropods (Allosaurus fragilis, Tyrannosaurus rex and Velociraptor mongoliensis), and for all three of them the ratio from the anterior to posterior was an even 4:6, which is excellent for a bipedal animal so it can evenly distribute weight...then we have Spinosaurus aegyptiacus...this guy is an underwhelming 5.57:6...that's honestly absurd for not only a bipedal animal, but an animal in the order of several tons to even try and move around upright.
So here we have the largest known "bipedal" animal that is completely ill equipped for terrestrial locomotion; tiny ilium, short and skinny legs, tiny feet equipped with thin toes (that likely supported webbed feet), long, cylindrical torso, long neck, dense bones and worst of all, an anterior to posterior ratio which makes the thing incredibly unbalanced and borderline top heavy. This is the opposite of an adaptation for any form of terrestrial locomotion. Even if it was highly aquatic and barely went on to land, the few times it would have tried would have been incredibly clumsy and likely to be disastrous for such a large bodied animal with comical proportions.
So we have an animal that is physically impossible to be a quadruped and is completely ill equipped to be a biped, so I assume a sliding/mud crawling* form of locomotion.
*when I say crawling, I don't mean actual pulling with the front arms like what a human would do, but something more along the lines of a loon/grebe in which it propels itself forward with its hind limbs, with a little help of the chest by moving around. The arms would help in one way or another, but truly grasping onto things or pushing the entire body by itself isn't what I'm getting at. I'm not completely for sure how it would slide/crawl around but something along these lines makes sense.
Does this all make sense from the lines of evidence I showed?
"Testing the buoyancy of an immersed Spinosaurus (Dinosauria: Theropoda) with a digital model
SVPCA Platform Presentation, 2016 (Liverpool)
A recent interpretation of the fossil remains of the enigmatic, large predatory dinosaur Spinosaurus aegyptiacus proposed that it was specially adapted for an aquatic mode of life – a first for any predatory dinosaur. A detailed, three-dimensional, digital model of the animal was generated and the flotation potential of the model was tested using specially written software. It was found that Spinosaurus would have been able to float with its head clear of the water surface. However, a similarly detailed model of Tyrannosaurus rex was also able to float in a position enabling the animal to breathe freely, showing that there is nothing exceptional about a floating Spinosaurus. The software also showed that the centre of mass of Spinosaurus was much closer to the hips than previously estimated, implying that this dinosaur would still have been a competent walker on land. With regional body densities accounting for pneumatized skeletons and system of air sacs (modelled after birds), both the Spinosaurus and Tyrannosaurus models were found to be unsinkable, even with the air sacs substantially deflated. The conclusion is that Spinosaurus would still have been a competent terrestrial animal."
My reconstruction is actually more rear-heavy than Ibrahim's model, due to it's dorsal ridge extending to the tail, and the longer tail. In addition, like I said in our previous conversation regarding this topic, the neck and torso would be pneumatized to a degree, while the hips, limb, and tail are much denser. The hips, legs, and tail would be packed with muscle, while the neck and torso would have cavities, organs, and airsacs.
About the limbs supporting it's weight, the femur has been noted to actually be quite robust by Ibrahim et al., and that dorsal ridge can fool eyeball guesses on it's apparent mass and weight.
But what of the tiny feet? They look rather comical for such a giant animal. What makes matters worse is that they're most likely to be webbed, so it didn't support any large fleshy pads like various highly terrestrial Theropods, both living and extinct. But I suppose the vast area of muscles around the torso and tail, along with the robust femur could deal with this?
Also, now that we know it's likely Spinosaurus was still a biped, I still find mud crawling/scooting equally as possible. And by this I say that it probably used both methods. After all, animals aren't just monotonous machines that can only perform a certain task. Just look at how modern animals have different modes of locomotion depending on the obstacle in the way. Crocodiles are my favorite example because when on land they can gallop, crawl, slide and mud crawl depending on the obstacle. They of course use mud crawling/sliding when on mud or slippery ground. When looking at Spinosaurus, it lived on a massive expanse of coastal floodplain, marsh and mangrove. This boasts highly slippery ground and quick sand. As anyone who has gone onto a coastal marsh (including myself), you constantly get stuck in sand that's full of water. It's incredibly hard to walk on. This is even worse for an animal that is multiple tons and has relatively small feet. So I say on very slippery/muddy ground and quick sand, it slided and crawled/scooted along, but on more firm land it walked (albeit rather slow).
And how would it have swam? Undulation of the tail like a croc I suppose? I've heard a lot about underwater punting, so I think they probably did both depending on the situation.
Nice job on the overhaul.
This new version is actually mostly the same size as the old ~16.5-meter one, but it has a new, longer tail and a bit shorter legs.
Also do you plan on estimating the size of NMC 41852?
(Sorry for double posting btw)
whats the deal with the 15 meter + specimen i keep hearing about around the web?
is it the same specimen?
also do you have any estimates for NMC 41582
And no, I have no estimates for NMC 41852 currently.
Because it's 12.4 meters long but it doesn't really make the
True size clear.
At best they could've said over 16 meters.
Your Giganotosaurus example doesn't work because Giganotosaurus is much more complete than Spinosaurus is, and thus would have much smaller error bars.
No mass estimate was mentioned in the Ibrahim et al. 2014 paper, that 7-tonne figure was from unofficial internet posts by the authors, but problematic because of the GSP-style model used.
With the exception of the dorsal profile of the neck there's nothing shrink-wrapped in the model, in fact the soft tissue is probably too thick in the chest, belly, pelvis and tail, this is how it should look like in GSP style: i.imgur.com/9LZlKvQ.jpg
Aside from the neck the soft tissues are clearly more substantial in Ibrahim et al.'s reconstruction than in yours.
btw, why is your new specimen that small, going by the known measurements, it's similar in size to the holotype and perhaps a little bigger (dorsals 17-18cm long rim-to-rim vs 17-21cm long maximum length, estimated 15-17cm rim-to-rim).