Members Creativity Thread

[theropod]

It might be, if there wasn’t the phylogenetic bracketing issue already outlined, and what you’ll have to admit is a fair bit of uncertainty as regards the actual skull/femur ratio of Carcharodontosaurus, which depends on the skull lenght you restore for the holotype (while the skull lenght of SGM DIN 1 can be constrained rather well, being almost certainly between 1.55 and 1.60m).

You are getting such an extreme result because the holotype skull turned out even smaller than my earlier estimate (which I now think was actually too low, I find 120-130cm to be a better fit for all the elements, that includes the braincase fragment).

[Deleted]

Apr 21, 2015 1:54:34 GMT 5 theropod said:

It might be, if there wasn’t the phylogenetic bracketing issue already outlined, and what you’ll have to admit is a fair bit of uncertainty as regards the actual skull/femur ratio of Carcharodontosaurus, which depends on the skull lenght you restore for the holotype (while the skull lenght of SGM DIN 1 can be constrained rather well, being almost certainly between 1.55 and 1.60m).

You are getting such an extreme result because the holotype skull turned out even smaller than my earlier estimate (which I now think was actually too low, I find 120-130cm to be a better fit for all the elements, that includes the braincase fragment).

Using SGM-Din 1 to reconstruct the holotype' missing bits, I can't really seem to get it to ~120-130 centimeters without giving it a premaxilla and/or a rear portion proportionally longer than they are for any other carcharodontosaur, or probably even carnosaur for that matter.

Well, the uncertainty in the actual skull size of the holotype specimen is one of the reasons I chose to scale the postcranials using the femur, as the femur's ~126-centimeter length is an established measurement for a mostly-complete bone while the skull length estimates, despite how well they fit the elements, are only estimates for a rather fragmentary element after all...

Right now most of the skeletal is scaled using the femur so changes in skull length won't change the overall size of the holotype by much(it has more effect on SGM-Din 1's size). Between the choice of getting a lower estimate but doing so via the use of extrapolating from an extrapolation(scaling body size using skull length), and the choice of extrapolating from an actual measurement even though it can create extreme-looking results(scaling using femur length), I think I would go with the latter.

I'll still make that alternate version though, for whose who prefer.

 

[theropod]

I see what you are getting at, but still, that's only for the holotype, but has the opposite effect on SGM Din-1.

Extrapolations from extrapolations are what you get when first extrapolating holotype size and thw using it tto estimate neeotype size, not when you use the skulls of both holotype and neotype individually to estimate their respective sizes. The latter method doesn't add up errrors: if the holotype is estimated incorrectly, that has no bearing on the neotype estimate. On the other hand, the method you used depends entirely on the holotype estimate.

[Deleted]

Apr 21, 2015 10:05:57 GMT 5 theropod said:

I see what you are getting at, but still, that's only for the holotype, but has the opposite effect on SGM Din-1.

Extrapolations from extrapolations are what you get when first extrapolating holotype size and thw using it tto estimate neeotype size, not when you use the skulls of both holotype and neotype individually to estimate their respective sizes. The latter method doesn't add up errrors: if the holotype is estimated incorrectly, that has no bearing on the neotype estimate. On the other hand, the method you used depends entirely on the holotype estimate.

I see your point, but IMO there isn't really a way to really estimate the neotype's size without dependence on the holotype specimen, and vice versa too, since the holotype's skull is quite fragmentary.

Carcharodontosaurus is so fragmentary that extrapolating from extrapolations is a must, but has enough elements that instances of extrapolating from extrapolations can be reduced a bit. Scaling the body size using the femur rather than the skull reduces the instances of extrapolating from extrapolations by one, so I did just that but at the cost of creating more extreme estimates and odd proportions.

Still better than shoving the Carcharodontosaurus neotype's head unto a Giganotosaurus' body with no consideration for the holotype at all.

[theropod]

What I’m saying is that I can’t see where you are reducing the instances of extrapolating from extrapolations, it appears you are doing an extra step.

[Deleted]

Apr 21, 2015 14:35:33 GMT 5 theropod said:

What I’m saying is that I can’t see where you are reducing the instances of extrapolating from extrapolations, it appears you are doing an extra step.

In the estimation of the postcranial body size.

Using femur to scale: Bone measurement(femur) -> Extrapolation(postcranial size)
Using skull to scale: Bone measurement(skull fragments) -> Extrapolation(whole skull size) -> Extrapolation(postcranial size)

You can see that I just removed a single step in estimating the postcranial size.

[theropod]

Only in the estimation of the holotype, but on the whole you have introduced more steps, and ones that lead to consequential errors:
Bone measurement of holotype femur->extrapolation of holotype postcranial size->Bone measurement of holotype fragmentary skull->extrapolation of holotype skull length->bone measurement of neotype skull->extrapolation of neotype skull lenght->extrapolation of neotype total lenght

versus

Bone measurement of partial skull->extrapolation of neotype postcranial size
Bone measurement of fragmentary skull->extrapolation of holotype postcranial size

In the first method, if at any point there is an inaccuracy the whole thing comes tumbling down on you. If the holotype’s skull lenght is wrong, it leads to inaccurate proportions and consequently an inaccurate estimate for the neotype. If the holotype’s lenght correlates more closely with skull length than with femur lenght, it leads to inaccurate size and proportions. And of course both can occur at the same time, then you add up both errors.

In the second method, there is less room for error. The potential error is that body lenght correlates more closely with femur length than with skull lenght. But you eliminate the ability of an inaccurate skull-femur ratio messing up both your estimates, and the possibility of using the wrong measure for estimating size as well as using the wrong proportions simulataneously.

As I wrote, the skull of SGM DIN 1 can be constrained to a rather narrow size interval (at least compared to the orders of deviation in which we are talking here), and we can estimate its TL without any steps involving the holotype. You can then incorporate the holotype’s elements into that reconstruction as an addition, for example telling you it should have longer legs.If the assumed proportions are off, then only that one thing will in the end be off too, not the entire estimate.

Click to enlarge→

[spinodontosaurus]

I need to start creating skeletals myself, this post is rather long.



Back when Scott Hartman first released his Giganotosaurus skeletal, there was a scaling error in the hindlimbs. He later corrected it, saying he had accidentally used the hindlimb proportions of Mapusaurus. I still have a fairly low-resolution version (1px = 2cm) of that skeletal from before he corrected the error.

So I decided to check the difference, and it seems that the incorrect 'Mapusaurus' version (I'm calling it that from here on) has a femur ~3.2% larger than in his new, corrected version. Seeing as his new version already has a femur length of 143 cm, the 'Mapusaurus' version must of had one around 147.5 cm long.

I would have to agree with Theropod - Carcharodontosaurus could well have had a proportionally longer femur than in Giganotosaurus, and it seems also that Mapusaurus did too.

Some minor talking points about the skulls: Broly's SGM-Din 1 skull needing to be shrinked a whopping 1% to match the fossil itself, and the very first maxillary tooth shouldn't be there. The premaxilla is probably a little bit too long and has 1 too many teeth, although it isn't actually preserved so your mileage may vary I guess. Doing this shrinks the skull down to a 'mere' 157 cm or so, virtually the same length as Theropod's reconstruction. I also seem to have found a reconstruction saved on my Laptop that appears to have been done by Blaze. SGM-Din 1 is 155 cm, the type specimen 135 cm.

It's the holotype skulls that I have the most problems with. The type maxilla only has 10 alveoli preserved (in the main chunk of it that is) but both Theropod's 'parsimonious' version (this comes across really passive-aggressive with the quote marks - that wasn't intended) and Broly's skulls have the preserved portion with 11 of them instead. This results in the skulls being scaled too small. I consider Theropod's 'generous' reconstruction to in fact not be generous, but pretty much on the money, at least in terms of length. In terms of placing, the big maxillary chunk should be much closer to the smaller chunk, in fact the way it positioned in the 'parsimonious' one is pretty much how I would do it do it myself, or simply take how it is in Figure C.

So if I reposition the maxilla, then rescale the skull so that the correct number of teeth fit, the skull ends up about 10% larger than Theropod's 'parsimonious' version, for a skull length of around 132 cm. We can then estimate SGM-Din 1 to have a femur around 150 cm long based on these modifications. This is not that much larger than the type specimen of Giganotosaurus, ~9.7% at most. I therefore do not think a femur-based scaling any higher than about 13.6 meters to be realistic. Of course, if you instead go by Blaze's scaling of the two specimens, and take the Giganotosaurus type specimen to have a femur length of 143 cm, then you could be looking at a total length of only 12.5 meters for SGM-Din 1.

Skull-based scaling isn't going to give you estimates higher than about 13 meters at most, which to me either suggests that Carcharodontosaurus has a shorter skull, or has a longer femur than Giganotosaurus. Given that the Carcharodontosaurus type specimen appears to have had a fairly gracile femur, and given that SGM-Din 1 has quite an elongate skull relative to Giganotosaurus, I am leaning towards the "Carcharodontosaurus has a proportionally longer femur" option.



And needless to say, I find Broly's 16 meter behemoth to be wholly unrealistic. The skeletal has some weird proportions.
The torso for starters is absurdly long. It's longer than the Giganotosaurus type specimen despite said specimen being over a metre longer in total, hell it's torso is as long as that of 'Sue'. Seeing as the skull and tail end up at around the same length as in Acrocanthosaurus 'Fran', the torso probably was too, some 9% shorter than in the original skeletal.
The size of the tail is fine based on the size of the single vertebrae, although it is missing some 6 vertebrae compared to how Hartman restores both Giganotosaurus and Acrocanthosaurus, if these are placed in the distal regions of the tail then the tail should become about as long as Acrocanthosaurus specimen 'Fran'. You've done about the best you could with the neck given how bizarrely small those vertebrae are, and of course I believe the skull should be about 18% larger.

I also think that both the articulation and placement of the shoulder girdle is off, it should be rotated so the scapula is more vertical and the whole thing should sit about the length of a vertebrae farther back. Fixing the articulation will also bring the shape of the torso more in line with what we see in both Giganotosaurus and Acrocanthosaurus, with the shoulder region being noticeably shallower than the pelvic region.

After having made these rough modifications to your skeletal, the axial length comes out at ~11.4 meters for the type specimen, and ~13.6 meters for SGM-Din 1. It is slightly longer than MUCPv-95, but the torso is about the same size and shape. They would probably weigh about the same after these modifications.
The purpose of modifying Broly's skeletal was to see how close to "reasonable" I could get it just by correcting what I considered to be scaling errors. The fact that it ended up the same length as my femur-based estimate from earlier is a nice coincidence. If Broly doesn't mind, I could brush up the rough edges on this quick edit and post it here so people can actually see what I did.

[Deleted]

theropod: Well I went for the tightest fit possible in this one. Yours has quite more space between the anterior maxilla edge and the preserved part's anterior edge then mine does. This can even be seen by using the nostril position as a guide.

---

Here's the Carcharodontosaurus skeletals with the tweaked scaling(longer legs as opposed to smaller head):

IPHG 1922


SGM-Din 1


Axial lengths for the holotype and SGM-Din 1 are ~10.18 meters and ~14.49 meters respectively.

references/sources

Stromer, 1931, "Wirbeltier-Reste der Baharijestufe(unterstes Cenornan). Ein skelett-rest von Carcharodontosaurus nov. gen"
Sereno, 1996, "Predatory dinosaurs from the Sahara and Late Cretaceous faunal differentiation"
J. W. Stovall, 1950, "Acrocanthosaurus atokensis, a new genus and species of Lower Cretaceous Theropoda from Oklahoma"
Carpenter, 2000, "A new specimen of Acrocanthosaurus atokensis (Theropoda, Dinosauria) from the Lower Cretaceous Antlers Formation (Lower Cretaceous, Aptian) of Oklahoma, USA"
Drew R. Eddy, Julia A. Clarke, 2011, "New information on the cranial anatomy of Acrocanthosaurus atokensis and its implications for the phylogeny of Allosauroidea (Dinosauria: Theropoda)"

---

Some shorter torso versions for Spinodontosaurus:

IPHG 1912
SGM-Din 1

[drone]

Here's my depiction of how the full skull of the 36,000 year old wolf/dog skull found in Goyet Cave looks:

[Deleted]

Updated Yongjinglong skeletal:




References/sources:

Li, L. G.; Li, D. Q.; You, H. L.; Dodson, P. (2014). "A New Titanosaurian Sauropod from the Hekou Group (Lower Cretaceous) of the Lanzhou-Minhe Basin, Gansu Province, China"
M. Borsuk-Bialynicka, 1977, "A new camarasaurid sauropod Opisthocoelicaudia skarzynskii gen. n. sp. n. from the Upper Cretaceous of Mongolia"
Palaeocritti - Nemegtosaurus
Lehman and Coulson, 2002, Alamosaurus skeletal
Some photos of a mounted Opisthocoelicaudia skeleton

[Deleted]

Another Yongjinglong skeletal update:



References/sources:

Li, L. G.; Li, D. Q.; You, H. L.; Dodson, P. (2014). "A New Titanosaurian Sauropod from the Hekou Group (Lower Cretaceous) of the Lanzhou-Minhe Basin, Gansu Province, China"
M. Borsuk-Bialynicka, 1977, "A new camarasaurid sauropod Opisthocoelicaudia skarzynskii gen. n. sp. n. from the Upper Cretaceous of Mongolia"
Palaeocritti - Nemegtosaurus
Lehman and Coulson, 2002, Alamosaurus skeletal
Scott Hartman, 2013, Alamosaurus skeletal
Some photos of a mounted Opisthocoelicaudia skeleton

[Deleted]

Finally finished that Tapinocephalus skeletal reconstruction:



This reconstruction is a composite of the holotype (vertebrae), SAM 2344 (skull) and SAM 3355 (scapula and humerus), and assumes that the three are similarly-sized specimens. This is also the first ever full-body skeletal reconstruction of Tapinocephalus AFAIK, and it's mine.

Bones in light gray are those that are mentioned but not illustrated. Ilium is based on Phocosaurus, while the femur is based on the indeterminate tapinocephalid specimen SAM 9097, which could possibly be a Tapinocephalus. Moschops' proportions was used as a guide to scale the hips and hindlimbs.

The interclavicle is based on the indeterminate tapinocephalid specimen SAM 9153. The rest of the skeleton is filled in based on Moschops.

AMNH 5611, the specimen composed of an arm and two shoulder girdles, which a photograph of can be found in Wikipedia, is excluded, as according to Boonstra(1955), it appears that AMNH 5611 isn't a Tapinocephalus at all, but rather a titanosuchid. Somebody better change that Wikipedia image...

references/sources
Owen, 1876, "Descriptive and illustrated catalogue of the fossil reptilia of South Africa in the collection of the British Museum"
L. D. Boonstra, 1955, "The girdles and limbs of the South African Dinocephalia"
L. D. Boonstra, 1956, "The skull of Tapinocephalus and it's near relatives"
Gregory, William, 1926, "The skeleton of Moschops capensis, a dinocephalian reptile from the Permian of South Africa"

[Deleted]

Updated my Dreadnoughtus skeletal:

It's ~26.35 meters long in axial length, has a Tapuiasaurus-type head based on it's position in the phylogenetic tree from the Notocolossus description paper, and has a longer tail (50 caudals, previous one had only 33). That's about it as far as current changes go.

references/sources

K. J. Lacovara, 2014, "A gigantic, exceptionally complete titanosaurian sauropod dinosaur from southern Patagonia, Argentina" + supplementary material
Calvo, 2007, "A new Cretaceous terrestrial ecosystem from Gondwana with the description of a new sauropod dinosaur"
Calvo, 2007, "Anatomy of Futalognkosaurus dukei[Calvo, Porfiri, Gonzales Riga & Kellner, 2007](Dinosauria, Titanosauridae) from the Neuquen Group(Late Cretaceous), Patagonia, Argentina"
Zaher, 2011, "A complete skull of an Early Cretaceous sauropod and the evolution of advanced titanosaurians"
B. J. Gonzales, 2016, "A gigantic new dinosaur from Argentina and the evolution of the sauropod hind foot"

[theropod]

www.nature.com/articles/srep20252