Deviation Actions

Paleo-King's avatar

Argyrosaurus superbus

Forgotten Giants species #4: Argyrosaurus superbus

Location of holotype site: Left bank of the Río Chico, near the Pampa Pelada, to the northeast of Lake Colhué-Huapi, Chubut, Argentina.
Length: ~90+ ft.
Mass: ~50 tons
Time: Late Cretaceous, but beyond that, estimates vary. Its formation, the Bajo Barreal, has been dated using some methods to the Santonian and Campanian epochs, but by others to the earlier Cenomanian. Its contemporaries include dinosaur species which sometimes overlap into the Allen formation and other faunas considered to be Campanian in age, so the former estimate is what I find more likely.


After a LONG time of studying and cross-scaling, here is the FIRST EVER scientific restoration of the mysterious "silver lizard" of Argentina, Argyrosaurus superbus. Named for its proximity to silver mining towns, this is one of the largest and strangest giant dinosaurs ever found - and ironically it was also one of the first to be named.

The obscure discovery of this massive plant-eater over a century ago was barely reported in the papers, but for a while Argyrosaurus was actually the largest dinosaur known to science. Described a full 10 years before Brachiosaurus, and over three decades before its record was beaten by Antarctosaurus giganteus, this monster is still known from only a handful of remains, yet they indicate a creature unusually massive and large even by the standards of other South American titanosaurs.

Argyrosaurus is characterized by very long, wide humeri proportionally similar to a giant snowboard, and massive femora wider than those of almost any other known sauropod, which rival those of Antarctosaurus giganteus for sheer size. The lower arm is relatively short, the shoulder blade large and deep, and the vertebrae massive and roughly similar in shape to those of Epachthosaurus.

Several likely specimens of Argyrosaurus are known, mostly just isolated limb bones from individuals of spanning a wide range of sizes and ages. The most complete specimen is the referred teenager PVL 4628, which includes both limb elements and vertebrae, and is the only specimen complete enough to give an idea of the overall proportions of Argyrosaurus. Thus it is the critical "keystone" specimen for unlocking the secrets of this mysterious dinosaur and its relatives. The isolated scapula indicates that PVL 4628 was not a mature animal, as its coracoid and scapula were not yet fused. It would have been roughly 65-70 ft. in length at the time of its death, while the largest individuals, known only from colossal limb bones, could have topped 90 ft. And given the extremely robust proportions of this dinosaur, it would have been several times as massive as most dinosaurs of similar length.

Argyrosaurus is the founding member of a family which currently contains only one other genus, the even more incompletely known Paralititan, which was similarly robust and heavy-bodied for its length even by titanosaur standards. Their closest relative outside the family appears to be the far smaller Epachthosaurus, which puts the Argyrosauridae in a relatively derived position - more primitive than Antarctosaurids and Saltasauroids, but more advanced than Lognkosaurians.

This animal is also remarkable in having a highly weird and RECURRING tendency to leave behind its right femur as fossil evidence, and not much else.* :XD:



Lydekker, R., 1893. "Contributions to the study of the fossil vertebrates of Argentina. I. The dinosaurs of Patagonia", Anales del Museo de la Plata, Seccion de Paleontologia 2: 1-14

Powell, J.E., 2003, "Revision of South American titanosaurid dinosaurs: palaeobiological, palaeobiogeographical and phylogenetic aspects", Records of the Queen Victoria Museum 111: 1-173

von Huene, F., 1929. Los saurisquios y ornitisquios del Cretacéo Argentino. Anales del Museo de La Plata (series 3) 3: 1–196. [In Spanish]


Some referred remains:

The assigned material comes from 8 localities, some of which do not correspond to the provinces of Chubut and Neuquén. For this reason, they are listed with geographic, stratigraphic, and chronologic provenance. The old geo-terminology "pre-Maastrichtian Senonian" refers to two consecutive Late Cretaceous epochs, the Santonian and Campanian, which were respectively the third-last and second-last epochs of the entire Mesozoic.

1. An anterior caudal vertebra (Huene, 1929:79). West bank of Lake Colhué-Huapi, Chubut Province. Chubut Group, Castillo Formation or Bajo Barreal Formation. Pre- Maastrichtian Senonian.
2. Two caudal vertebrae “…of the region of Neuquén” (Huene op. cit.:7). Geographic and stratigraphic provenance unknown.
3. An incomplete femur (Lydekker, 1893, pl. 5, 2; Huene, 1929:80, pl. 38, 1). To the south of the bend of the Río Sengerr, Chubut Province, probably Bajo Barreal Formation. Pre-Maastrichtian Senonian.
4. An incomplete right femur (Huene, op. cit.:80, pl. 38, 2). Near the railroad bridge over the Río Neuquén, Neuquén Province. Neuquén Group, Río Colorado Formation, perhaps Bajo de la Carpa Member. Pre-Maastrichtian Senonian.
5. A complete right femur (Huene, op. cit.:80-81, pl. 38, 3). Sierra San Bernardino, 45 km west of Colonia Sarmiento, Chubut Province. Chubut Group. Bajo Barreal Formation. Pre-Maastrichtian Senonian.
6. A humerus of a juvenile individual, referred with doubts to Argyrosaurus (Huene, op. cit.:81, pl. 37, 6). Probably from Neuquén. Uncertain geographic and stratigraphic provenance.
7. A left humerus, referred with doubts to Argyrosaurus superbus Huene, op. cit.:81, pl. 37, 4). Left margin of the Río Uruguay, near Colón, Entre Ríos Province. ?Asencio Formation. Upper Cretaceous.
8. Caudal vertebrae (Huene, op. cit.:79). To the east of the Río Leona, between lakes Viedma and Argentino, Santa Cruz Province, Upper Cretaceous (Dibenedetto, pers. comm.).

Source: Dinodata

*NOTE: the unnamed and uncatalogued "Argyrosauridae" femur from Powell (2003) is most likely a lognkosaur judging from its proportions, not an Argyrosaur.
Image details
Image size
4080x4384px 2.08 MB
© 2011 - 2022 Paleo-King
Join the community to add your comment. Already a deviant? Log In
Randomwormo's avatar

absolute u n i t

bricksmashtv's avatar
Hooray for the 100th comment:) (Smile) ! On a more serious note, do you recognize Elaltitan as a unique genus now, or do you still consider it a specimen of Argyrosaurus?
Paleo-King's avatar
It's a unique genus. Probably somewhere near either Argyrosauridae or Antarctosauridae. But it is a unique genus. This image and the old comments are outdated.

That said I'm admittedly a bit bitter that pretty soon after I put up this final version of these specimens, the "juvenile Aryrosaurus" was renamed Elaltitan and got a full description. All those years I was working on this thing on and off in my college years, trying to figure out proportions and hunt down crappy old photos, all that time Mannion and Calvo didn't even TOUCH this critter. All that work cross-scaling proportions for Argyrosaurus off that smaller specimen and a short while later it wasn't even the same animal... :X :X :X  I had my doubts at first but the paper's conclusions about this creature are pretty solid. Elaltitan is legit. Still, Argyrosaurus is a much better sounding name, Elaltitan just sounds annoying, there are better names they could have invented.

However the Elaltitan paper does confirm my overall view of dinosaur taxonomy - contrary to the Hornerites, most dinosaur taxonomy is far from oversplit into "too many genera", in fact it is NOT SPLIT ENOUGH! The same is clear from the official recent separation/redescriptions of Giraffatitan and Lusotitan from Brachiosaurus. Isisaurus from Titanosaurus. Atlasaurus from Cetiosaurus. And so on. Just like in modern terrestrial mammals and birds, which can show FAR less skeletal variation between two genera than in many dinosaur families, yet still be totally genetically separate genera incapable of reproducing even a sterile hybrid, often even having different chromosome numbers! Just try to force the Gemsbok and the Springbok into a single genus. Or the Emu and the Rhea. You can't! Yet they look more sksletally identical to each other than any of Horner's "Toroceratops" or "PachyDracoLoch" constructs. Elaltitan is just more proof that Hornerism is a red herring, "arm-waving" or rather bending over backwards to force data into a pre-cooked hypothesis, and is definitely NOT "maverick".
bricksmashtv's avatar
Exciting. Will you be revising Argyrosaurus anytime in the near future, or do you wanna get some more Forgotten Giants first? Gotta love those Hornerites and the lumping. I'm still holding Seismosaurus, Morosaurus and Nanotyrannus as separate genera, with multiple species of Triceratops, and Tyrannosaurus. And of course Apatosaurus ajax and "Apatosaurus" louisiae.
Paleo-King's avatar
Yeah, I figured they should have split off louisae into its own genus before excelsus! It's so much stumpier and thicker and the sacrum looks nothing like A. ajax. Nano is clearly NOT the same species as Jane. I don't have a problem with calling Jane a young T. rex because its skull shape matches immature T. rexes at the Los Angeles museum. But it's NOT a Nano, Nano has very different snout stress points and lacks Jane's upturned snout and oversized teeth. The Dueling Dinosaurs Nano specimen is what a REAL adult Nanotyrannus looks like. And its proportions and features (small teeth, "roman" snout, notch between the maxilla and premaxilla, even LONGER snout, etc.) do NOT match Jane or any juvenile T. rex.  It's like comparing Deinonychus with Utahraptor, sure they look similar as young animals but are clearly in two different size niches and looked different as adults.

I don't know much on Morosaurus or where it goes, haven't seen much material there... but Seismosaurus for sure is not Diplodocus. Too many differences in the skeleton. Scott Hartman has had to FORCE the odd banana-like ischia and the kinked anterior caudals into some very artificial shapes and postures which they don't really have, to even make his "D. hallorum" idea work. No offense intended to Scott, but I just don't see that much shape alteration being believable, given that the bones in question are not very crushed at all. Even "D. hayi" is now Galeamopus based on far less variation, so I maintain Seimosaurus as its own genus... and Kaatedocus is not Diplodocus either though Hornerites may bristle at that (they try to lump Haplocanthosaurus as a juvenile Diplodocus for crying out loud, really???? :O go figure).

Triceratops in my view has 3, maybe 4 species. T. horridus in the lower hell creek, T. prorsus in the upper hell creek, an intermediate species in the middle hell creek layers which has some features of both, and of course T. eurycephalus which is known from at least two skulls and is probably the most basal form. Nedoceratops is its own genus, and it probably clusters in a "tribe" with Eotriceratops judging by the odd horn angle. These two probably shared a common Arrihnoceratops-like ancestor with T. eurycephalus or whatever the first Triceratops species was.

What's even stranger is that Torosaurus is probably 3 species. T. latus for the Yale, Milwaukee and ANSP material, and perhaps a few others, T. utahensis which has deeper squamosals, and the huge (and yet still immature!) "T. magnus" as I call it, which is represented by the big MOR skull that has smaller fenestrae and more immature epoccipitals than the others despite being a bigger animal. Oh, and it has a HUGE boss-like nasal cone instead of the tiny stub-horn of the other Torosaurus forms.
bricksmashtv's avatar
Gotta love the Hornerites and the excessive "over-lumping" culture we've gotten into in recent years. I'm tired of Toroceratops & Pachydracomoloch & Nanotyrannosaurus & Seismodocus. Then of course, if these guys were clearly unique genera, than who else has been haphazardly thrown into a known genus when it's clearly deserving of it's own unique genus, and who we've subsequently forgotten about for the past 50 or so years? I mean, it's clear Tyrannosaurus has at least three unique species (or subspecies, but nobody uses that term anymore). Just in the Morrison alone, we've got Apatosaurus, "Apatosaurus" louisiae, Brontosaurus, Elosaurus, Eobrontosaurus, Diplodocus carnegii, Diplodocus longus, Seismosaurus, Supersaurus, Barosaurus lentus, Barosaurus sp., Galeamopus hayi, Galeamopus sp. Kaatedocus, Amphicoelias altus, "Amphicoelias" fragillimus, Haplocanthosaurus priscus, Haplocanthosaurus delfsi, Haplocanthosaurus utterbacki (I think that's how you spell it?), Brachiosaurus altithorax, Brachiosaurus sp., "Jensensaurus", Camarasaurus supremus, Camarasaurus lentus, Cathetosaurus, Morosaurus, and who knows what else. If one span of a few million years has THAT much diversity in a single area (just for the Sauropods!), then there is NO WAY the Dinosauria cannot have more than at least 3,000 species (roughly 2,200 genera) over it's entire timespan, and even than, I'm still probably underestimating, just because of the amount of Dinosaurs whose remains will have never fossilized because of their location at the time of death, or those whose remains have been destroyed over time. Who knows how many giant Sauropods there actually were whose remains have never fossilized, because of the circumstances required to bury a 50 ton+ animal.
Paleo-King's avatar
Good point, there may have been as many as fifty sauropod species in any given formation! We just don't know how extensive the plant life was to support so much fauna, but it's definitely within the realm of possibility.

The fact is, whereas Hornerites want only a couple hundred dinosaur species across the mesozoic, REALITY demands that there were tens of thousands, and many of which we don't even know about yet. If you tally up all the remains, my guess is we have at least about 1,000 to 1,500 generally accepted dinosaur taxa so far. The real number based on the fossils we have should be at least triple that, given how much stuff is languishing undescribed in museums (providing for congeneric overlap with previous discoveries of course). But even this is only a tiny fraction of what once was. You're definitely correct on that, the vast majority of dinosaurs never got fossilized, and so there's a high probability that most species literally disappeared without leaving a trace. Such is true of all terrestrial ecosystems.

So imagine something like 70,000 species in around 25,000 genera on average for all the non-avian dinosauria that ever lived. Yes, crazy, I know, but ENTIRELY plausible. Just look at all the new ceratopsians being dug up and described every other week, it's going to be REALLY hard for Horner (or Greg Paul) to shoehorn Medusaceratops, Mercuriceratops, Coahuilaceratops, Kosmoceratops, Utahceratops, etc. all into Chasmosaurus!
bricksmashtv's avatar
Of course. How can we possibly live without GSP lumping all of Brachiosauridae into Brachiosaurus, or Tyrannosauridae into Tyrannosaurus, or Diplodocidae into Diplodocus. At this rate, we'll go back to 10 species of Brachiosaurus, and 15 or so species of Cetiosaurus Roll-Eyes Smiley 
Paleo-King's avatar

funny thing is there may be as many as 10 species as Brachiosaurus... but they are just the ones labeled as "Brachiosaurus sp." and lying largely undescribed in museum vaults. Or these may be other genera. The "Brachiosaurus" humerus in LACM is certainly NOT Brachiosaurus itself, but I don't know if there's any other material from that specimen to reconstruct it.
View all replies
Algoroth's avatar
Just a small, tiny question. AHEM! Should the limb elements of upright quadrupeds be restored in terms of what we can see of sprawlers, like crocs and lizards, or upright animals, like birds and mammals? Just asking.

Frankly, I think most all of us have them wrong, including me, you, Zack, Jurassic Park, and most of the dino documentaries I have seen. I think many artists have theropod hindlimbs correctly, but few correct (IMO!) front limbs.
Paleo-King's avatar
What do you mean? The articulation of sauropod from limbs clearly only makes sense in a vertical upright posture. Not that it had to be PERFECTLY vertical (there was a slight 10-15 degree bowing at the elbow when either arms was in rear stride - this is normal in elephants and such today). But they were definitely not build for sprawling. Also their rib cages are usually so deep that if they had sprawled, they would have needed to find giant ruts to run in, because their bellies would have been well below the ground level of the hands and feet (and all their joints would have to be dislocated).

Plus it's also important to note that sauropods evolved from bipedal (and warm-blooded) ancestors. Crocs and lizards did not. To date there has never been a group of quadrupedal animals whose forelimbs have a completely different posture from their hindlimbs - it's literally impossible to coordinate movements in such an animal especially with spines as rigid as most large dinosaurs. And dinosaur footprints all point to erect forelimbs in quadrupeds.
Algoroth's avatar
I was talking about lower limb robustness compared to upper limb robustness, not the vertical positioning, about which you are absolutely correct. Most upright quadrupedal animal and bipedal animals (birds and the occasional human ;)) have lower limbs that can be downright thin. But the upper limbs are often so well muscled they can scarcely be seen, as in horses, canines rhinos, etc. Even elephants' legs are well integrated with the upper bodies (skin-wise, yes, but not separate almost all the way to the shoulder), and are not thin.

When we get to sprawlers, we have a different picture. The upper limbs are mostly separate and the lower limbs, especially in the large species....crocs and large varanids...are close to being as thick as the upper limbs.

None of this is proof, of course, but, since dinosaurs and birds are so similar, why do often make the upper and lower limbs so alike in muscularity...whether thick or saran-wrap style?
SpinoInWonderland's avatar
Argyrosaurus was never the largest dinosaur known to science...

Amphicoelias fragillimus was discovered in 1878, which was about 15 years before Argyrosaurus was discovered.
Paleo-King's avatar
Yeah but A. fragillimus wasn't really "known" at the time aside from a handful of field workers and Cope himself. He published a brief paper on it which included no size estimates, it was soon lost or destroyed, and even at the same time many people considered it a hoax.

Most scientific publications of the 20th century listed Brachiosaurus as the biggest dinosaur (in fact up until the 80s even most paleontologists said so in their books; they forgot about Argyrosaurus and "Antarctosaurus" giganteus even though these were never lost or destroyed.) So while they were saying Brachiosaurus was the biggest, there was something larger (at least in terms of mass) discovered 10 years earlier, with published photos rather than sketches, more than a single bone, assigned an actual museum catalog number, and the material is still in existence.
Teratophoneus's avatar
Hey Nima, I amplanning to use this as a reference for a scientific reconstructuion of elaltitan-or better Argyrosaurus itself as you said the bone of elaltitan is that of a argyrosaurus. For this I have to know how big the specific specimen (length and height) of the elaltitan/argyrosaurus was. I hope you can help me :)
Paleo-King's avatar
Use the scale bar, it's there for a reason LOL! :D

Both the adult and the teenage Argyrosaurus (aka "Elaltitan") are to scale with the scale bar.

The teenager is roughly 70 ft. long if I remember from last time I re-scaled it. Of course the live animal may have looked nothing like this. This is just my interpretation of how it looked based on the few bones found, and comparing it with other large titanosaurs like Alamosaurus.
palaeozoologist's avatar
The dorsal vertebrae, anterior caudal vertebra, scapula and the pelvic and forelimb elements of PVL 4628 have now been referred to a new genus and species, Elaltitan lilloi, so you should probably change the name on your image for the smaller individual.


Mannion, P. D.; Otero, A. (2012). "A reappraisal of the Late Cretaceous Argentinean sauropod dinosaur Argyrosaurus superbus, with a description of a new titanosaur genus". Journal of Vertebrate Paleontology 32 (3): 614-638.
Paleo-King's avatar
Thanks Zach. But given what I've seen of this animal, I don't see a basis for the reclassification. Mannion previously also made some statements about titanosaur taxonomy that I consider unwarranted and flat-out wrong (such as claiming Ruyangosaurus wasn't even a titanosaur in his redescription of Andesaurus).

Based on Powell 2003 (the better scans of it anyway) PVL 4628 has basically the same arm morphology as the Argyrosaurus holotype. So much so that there is little point in referring it to "basal titanosauria" while putting PVL 4628 in derived lithostrotia.

The differences that Mannion does cite in the abstract are a bit obscure as he only says: "This specimen
is distinct from *Argyrosaurus* and can also be differentiated from other
sauropods based on an unusual character combination (including plesiomorphic
tarsus), plus one autapomorphy."

VAGUE VAGUE VAGUE. I will not endorse this conclusion until I read the paper (which is locked up behind a paywall... I wonder how much of the paleosphere is actually paying attention to SV-POW and the point they're making). One autapomorphy, plus plesiomorphic tarsus (?) and an unusual character combination.... Well it's too premature to separate it from Argyrosaurus based on ANYTHING other than a comparison of the arms, since that's all the Argyrosaurus holotype is. And the tarsus is not in the arm. A single autapomorphy (which may or may not overlap with the Argyrosaurus holotype skeletally) is pretty weak proof. It would be more parsimonious to classify PVL 4628 as an immature Argyrosaurus sigen overall morphology of the arm, and the matching formation/time horizon.

For my part I suspect any differences in the arms of the two individuals are minor enough that they can be put down to ontogeny (PVL 4628 is clearly immature based on the scapula not being fused to a coracoid), the arm elements of both individuals are very similar in overall shape and proportions. The ulnae are SO similar it's kind of freaky. So YES I am lumping. Go sue me Dr. Mannion LOL 8-)

BTW I'm curious, how did name "Elaltitan" come about? I've heard some weird rumors. Is it really named after an airline?
palaeozoologist's avatar
Just because Phil Mannion has made statements in the past that you consider wrong on titanosaur phylogenetics, doesn't mean he is wrong here. I think we can both agree that Ruyangosaurus needs to be redescribed before judgement can be passed on its phylogenetic relationships. I'm aware you think it is a lognkosaur, but until it is properly scored, put in a comprehensive phylogenetic analysis, I think that is a untested hypothesis/assumption (especially since not all lognksaurs have super-wide diapophyses on their dorsal vertebrae, i.e. Mendozasaurus and Drusilasaurus). But I digress.

I'm also having a hard time seeing why you think Mannion is being vague when he says (and I quote), "Argyrosaurus can be diagnosed by five autapomorphies (marked by an asterisk), as well as a unique combination of character states: (1) medial margin of humerus forms a transversely wide ridge that projects prominently anteriorly∗; (2) midshaft of humerus extremely compressed anteroposteriorly (transverse to anteroposterior width ratio = 2.6)∗; (3) transverse width of distal end of radius only slightly greater than midshaft width (ratio = 1.3)∗; (4) radius is subtriangular in distal end view∗; (5) presence of ossified carpals; (6) extreme elongation of metacarpals (longest metacarpal to radius length ratio = 0.6)∗;(7) metacarpals II and III longest elements in metacarpus." (p. 615-616)

On Elaltitan he says, "Elaltitan can be diagnosed by an unique combination of character states, as well as one autapomorphy (marked by an asterisk): (1) spinopostzygapophyseal laminae in middle–posterior dorsal vertebrae bifurcate into medial and lateral branches; (2) dorsoventrally tall neural arch restricted to anterior half of centrum (excluding condylar ball) in anterior-most caudal vertebrae∗; (3) astragalar ascending process does not extend to the posterior margin of the astragalus; (4) presence of a calcaneum." (p. 623)

I'm sorry, but I'm having a hard time seeing how this is "vague".

Also, the arms don't have the same proportions:

Humerus: Elaltitan: 1300 mm, Argyrosaurus: 1370 mm
Radius: Elaltitan: 730 mm, Argyrosaurus: 858 mm
Ulna: (lft)Elaltitan: 827 mm, Argyrosaurus: 965 mm
Ulna: (rght)Elaltitan: 880 mm, Argyrosaurus: not preserved

H-R ratio: Elaltitan: 1.780821918, Argrosaurus: 1.596736597
H-U (both lft) ratio: Elaltitan: 1.571946796, Argrosaurus: 1.419689119
R-U ratio (both lft): Elaltitan: 0.882708585, Argrosaurus: 0.889119171
R-U (left,right)(Elaltitan only): 0.829545
H-U (left,right)(Elaltitan only): 1.477273

Only the Radius-Ulna ratios are similar, but the Humerus-radius and Humerus-ulna ratios are quite a bit different.

I also find it odd that you think the holotype Argyrosaurus specimen and the Elaltitan specimens are of differenet ontogenetic stages. Based on what? The scapula is not preserved with the holotype specimen so you can't compare the ontogeny of the two specimens without a histological analysis (which has not been done AFAIK). Size is very misleading when it comes to ontogeny (as Triceratops specimens show), and smaller specimens of the same species can actually be more mature than larger specimens.

Their arm proportions are actually different (see my comparisons above). Another interesting difference between them, in Elaltitan, "At midshaft,the humerus has a transverse to anteroposterior width ratio of just under 2.0" (p. 629) whereas in Argyrosaurus it is "At midshaft, the humerus is strongly compressed anteroposteriorly, and it has the highest transverse to anteroposterior width ratio of any known sauropod (2.64 [measured on the anteroposteriorly thicker medial side]; see Table 1)[...] we regard the extreme condition in Argyrosaurus as an autapomorphy of the genus."

BTW, the tarsus autapomorphy for Argyrosaurus is not used to distinguish it from Elaltitan, but from other titanosaur sauropods (i.e., it is currently diagnostic when compared to other titanosaurs for which it can be compared).

There are other differences between Elaltitan and Argyrosaurus in the morphology of the radius and ulna such as:

* distal end of radius is mediolaterally expanded in comparison
to the shaft - ratio of 1.7 in Elaltitan compared to 1.3 in Argyrosaurus

*proximal end transverse width to ulna length ratio of 0.45,and differing from the more gracile morphology seen in Argyrosaurus ratio of 0.34

Mannion lists other differences that I will not go into here, but I hardly think they and the ones I have mentioned are "vague".

One point that I might concede is that it is possible for arms to change proportions through ontogeny, this is known to be the case in Alamosaurus for instance, where the humerus gets proportionally longer. However, the other differences in the ratios are not so easily chalked up to ontogeny. In any case, I think the case for taxonomic distinction has been proved beyond a reasonable doubt with the current amount of evidence available.

I also think it is interesting that you think that Elaltitan could morph into Argyrosaurus gaining an autapomorphic medial ridge on the humerus, but you feel that fenestration in Triceratops-Torosaurus is impossible. In other words, Triceratops and Torosaurus share every diagnostic feature except their frills, whereas Elaltitan and Argyrosaurus differ in every diagnostic feature that we can compare them on.

As for the name, the paper says, "EtymologyElal (ee-lal), the creator god of the Tehuelche people of Chubut Province; titan, giant in Greek mythology. Specific name in honor of Miguel Lillo, for his contribution and legacy to natural sciences in Tucuman." So no airline, AFAIK.
TyrannosaurusPrime's avatar
"As for the name, the paper says, "Etymology—Elal (ee-lal), the creator god of the Tehuelche people of Chubut Province; titan, giant in Greek mythology. Specific name in honor of Miguel Lillo, for his contribution and legacy to natural sciences in Tucuman." So no airline, AFAIK."

Not sure about you, but I still find Elaltitan to be a lame name (admittedly mainly it's personal bias, but partly because it sounds like eLOLtitan :XD:). They should have used Argentinotitan as the genus name IMO.;)
Paleo-King's avatar
If someone could prove that the muscle crests on the arms of both animals are radically different, lets say as equidistant phylogenically as the Argyrosaurus holotype is from Epachthosaurus (or for that matter any other titanosaur known from a complete arm), then I'd be ready to buy the argument that PVL 4628 is a different genus. This would be more conclusive than bone length ratios.

However the fact remains that arm proportions can and do change in ontogeny, and titanosaur ontogeny is still poorly understood (as is the anatomy of Argyrosaurus). And there are no axial elements known from the holotype, which makes comparisons tricky.
palaeozoologist's avatar
You are completely missing the point. We don't need the muscles crests to be radically different for two reasons. (1)"Radically different" is not a scientific concept. (2) All we need them to be is diagnostic and phylogenetically distinct or informative. The autapomorphies of Argyrosaurus are diagnostic and are phylogenetically informative (at least, at present they are, new discoveries could render them nondiagnostic in the future, but that is neither here nor there).

Argyrosaurus possesses 4 autapomorphies in elements that overlap with Elaltitan which Elaltitan does not possess (and which no other titanosaur possesses, by definition). If there were an analagous concept in science for "radically different" in taxonomic phylogeny, it would mean being phylogenetically diagnostic and distinct (autapomorphic). In that sense then, the autapomorphies of Argyrosaurus are "radically different" from Elaltitan. BTW, an autapomorphy basically means "as equidistant phylogenically as the Argyrosaurus holotype is from Epachthosaurus (or for that matter any other titanosaur known from a complete arm)" as it is from Elaltitan. Wikipedia says, "an autapomorphy is a distinctive anatomical feature, known as a derived trait, that is unique to a given terminal group. That is, it is found only in one member of a clade, but not found in any others or outgroup taxa, not even those most closely related to the group (which may be a species, family or in general any clade)." The 4 autapomorphies that are on comparable elements for Argyrosaurus and Elaltitan are just that: they are not found anywhere else in the Titanosauria--by definition being an autapomorphy. Only if these traits were found to vary ontogenetically and are subsequently found in various other taxa could an argument for synonymy be made. As it stands, the synonymy argument fails all scientific tests.

And as I said in my other comment, arm proportions are observed to change in specimens demonstrably on opposite ends of the ontogenetic spectrum and of radically different sizes! Neither of which are demonstrably true for Elaltitan and Argyrosaurus. So again, your argument for synonymy fails when tested.
Paleo-King's avatar
Telling "Elaltitan" from Argyrosaurus is indeed a lot more vague than you make it out to be. For starters, most of Mannion's autapomorphies cannot even be compared to the Argyrosaurus holotype due to the lack of overlapping material.

On Elaltitan he says, "Elaltitan can be diagnosed by an unique combination of character states, as well as one autapomorphy (marked by an asterisk): (1) spinopostzygapophyseal laminae in middle–posterior dorsal vertebrae bifurcate into medial and lateral branches;
This can not be compared to the Argyrosaurus holotype as none of its vertebrae were found!

(2) dorsoventrally tall neural arch restricted to anterior half of centrum (excluding condylar ball) in anterior-most caudal vertebrae∗;
Same problem as before. No vertebrae in the holotype to compare this to.

(3) astragalar ascending process does not extend to the posterior margin of the astragalus;
There's no astragalus in the holotype to compare it to

(4) presence of a calcaneum." (p. 623)
Once again no overlap with the holotype; BTW the presence of a calcaneum doesn't prove much of anything. Being so small and often only connected to other bones by cartilage, the calcaneum in sauropods tends to get washed away. Most if not all sauropods probably had a calcaneum. Most artists tent to omit the calcaneum from species for which it is known, including Greg Paul.

Now while the rations of the arm bones do differ in PVL 4628 and the holotype, there's no conclusive way to prove that it isn't ontogeny. Like you said, the arms of Alamosaurus do change proportions over time. However you MAY be able to prove they are different species by locating some muscle crest or lamina in the arm bones of one that is not present in the other. For example there are small rugose laminae on the porimal part of the humerus of Brachiosaurus which are not present in Giraffatitan, aside from the shapes of the humeri being different. However you haven't pointed out any significant morphological differences between the arm elements of the Argyrosaurus holotype and PVL 4628. Does Mannion ever actually compare the morphology of the individual arm bones of the two animals aside from that "medial ridge"? Or does he simply settle for length ratios and leave it at that? One single autapomorphy that can actually be compared between both specimens is a pretty weak basis for splitting. By contrast, Triceratops and Torosaurus can be split on the basis of many more observable differences, many of which have nothing to do with their frills (for example beak shape, nostril elongation, beak/snout ratio and radically different patterns of horn ontogeny - plus, Horner & Co. have never entered all known Torosaurus postcrania into a data matrix along with a good sample of Triceratops postcrania from different growth stages - they claim they have data backing their conclusions, well let them actually crunch it and honestly report the result).

In the best scans I can find of Powell (2003) the two specimens have very similar morphology of the arm bones, the humeri for example have the same wide-shafted, flattened "snowboard" shape which is rare in most titanosaurs (for example Alamosaurus, Neuquensaurus, Saltasaurus, Uberabatitan and even apparently some immature Futalognkosaurus, if field photos can be believed, have a twisted curvature to the humerus and very forward-projecting deltopectoral crests). Aside from Argyrosaurus, only Paralititan, Epachthosaurus, and a couple of australian titanosaurs have flattened wide-shafted humeri with little forward projection of the deltopectoral crest - and none of them resemble the holotype Argyrosaurus in overall poportions as much as PVL 4628 does.

Furthermore, the lack of a scapula with the holotype actually does not support your view: simply because there's no scapula to compare with PVL 4628, that does not mean that PVL 4628 is more mature or somehow violates lumper ontogeny the way Torosaurus does. Indeed, there would be only two possible outcomes if the Argyrosaurus holotype were found with a well-preserved scapula: (1) a coracoid is fused to it, which means the holotype is adult or close to adult, which agrees with the ontogeny theory; or (2) a coracoid is absent, which means that both the holotype AND PVL 4638 are immature animals, meaning there is no way to tell which is more mature based on morphology alone, and a histology study may be needed to determine age and relative maturity for both animals. This second possibility also does not conflict in any way with the ontogeny theory - both specimens clearly had some growing left to do.

The only way to DISprove ontogeny between these two Argyrosaurus specimens is to either find radically different posterior dorsal vertebrae, anterior caudals, femur, astragalus, or a pubis for the holotype or a specimen overlapping with it (which doesn't seem likely any time soon) or to do a detailed morphological inspection (like Borsuk-Bialynicka did on Opisthocoelicaudia) the both animals, which yields very different results for muscle attachment configurations.

Now that said, I don't think ontogeny is a certain fact in the case of PVL 4628. It MAY just be another species of Argyrosaurus or another genus in argyrosauridae (in fact I simply call it Argyrosaurus sp. for now). But it's the most complete specimen of any argyrosaur so far known, so it's the best there is for scaling reference for Argyrosaurus superbus, and may be the same animal based on stratigraphy and location.

As I said before, I can't endorse any part of Mannion's theory until I see the paper for myself. And BTW, Mendozasaurus and Drusilasaura DO have very wide dorsal diapophyses. Especially Drusilasaura, its anterior dorsal is basically a scaled-down copy of Puertasaurus proportion-wise, the diapophyses on that thing are huge (and wedge-shaped with deep buttressing laminae beneath, just like in Puertasaurus and by the look of things, Ruyangosaurus).
palaeozoologist's avatar
No, it is not "vague". Vague means "not clearly or explicitly stated or expressed." Mannion's hypothesis is clearly and explicitly expressed. You could maybe say that you think his hypothesis is "unsound" or "unfounded" or "weak" (characterizations which I still would disagree with—see below), but it is definitely not vague.

You also appear not to have read what I said, because when I quote Mannion’s autapomorphies I do so in the context of showing that Mannion is not vague at all, but it is very explicit and clear in his hypothesis. You appear to have conflated the different aspects of my argument. So your first four bolded statements are useless in supporting your argument because I already implicitly acknowledged that those autapomorphies do not overlap.

Also, you incorrectly focus on the autopamorphies of Elaltitan instead of Argyrosaurus. Of the 5 autapomorphies listed for Argyrosaurus (marked with an asterisk), 4 can be compared to Elaltitan - and Elaltitan shares none of them!

As for the arm proportions, yes, you are correct in saying there is no way to conclusively prove that the changes are not ontogenetic. However, the reverse is also true (that is, there is no way to conclusively prove that the changes are ontogenetic). This means you are assuming a priori that they are ontogenetic in order to support your hypothesis. You are going about that completely backward. You must show that ontogeny is a factor before synonymizing, not after.

Also, as for arm proportions changing ontogenetically in Alamosaurus: the difference in proportions is seen between an animal with a 605 mm humerus compared to a one with a 1360 mm humerus, meaning there is a factor of 2.24 in size. Yet there is only a size factor of 1.05 between the humeri of Elaltitan and Argyrosaurus (a factor of only 1.17 between the radii, and a factor of only 1.16 between the ulnae). This means you are proposing significant proportional and morphological features occurring in animals that are already nearly the same size humerus-wise. (Huh…this is beginning to sound awfully like Horner and Scanella’s late ontogenetic trajectory hypothesis (LOTH) for Torosaurus v. Triceratops-which you reject-even though Horner and Scanella’s hypothesis is backed up by histology whereas your hypothesis Elaltitan v. Argyrosaurus is not. Therefore, you actually support LOTH if it conveniently supports your favored taxonomic hypothesis, but are against LOTH if it doesn’t support your hypothesis. This sounds an awful lot like the pot calling the kettle black.)

Also, in Triceratops and in Torosaurus the beak shape, nostril elongation and beak/snout ratio are not diagnostic and are individually variable between specimens (whereas the autapomorphies for Argyrosaurus actually are diagnostic and different from Elaltitan). We don’t have young Torosaurus specimens to compare horn morphology for (at least not yet described).

You said: “In the best scans I can find of Powell (2003) the two specimens have very similar morphology of the arm bones.”

This is not the case (as I already have said, they have different proportions and Argyrosaurus has 4 autapopmorhies not seen in Elaltitan for which there are overlapping elements to compare).

You said: “ the lack of a scapula with the holotype actually does not support your view: simply because there's no scapula to compare with PVL 4628, that does not mean that PVL 4628 is more mature or somehow violates lumper ontogeny the way Torosaurus does.”

…and you’ll notice that I never said that PVL4628 is more mature or violates lumper ontogeny, all I said is that there is no way to know, based on the external morphology, what their relative ontogenetic stages are. You put the cart before the horse here: assuming ontogenetic differences when you have yet to prove they are actually ontogenetically different! You assume ontogeny is the factor, based on size differences, which we know are an unreliable indicator and they are actually not that different in size (see humeri ratios above: 1.05 vs. 2.24)! Even if they were of different ontogenetic stages, there still are 4 of 5 autapomorphies listed for Argyrosaurus that Elaltitan does not have(and for which they can be compared). I know of no precedence for this occurring in ontogeny.

You said: “ The only way to DISprove ontogeny between these two Argyrosaurus specimens is to either find radically different posterior dorsal vertebrae, anterior caudals, femur, astragalus, or a pubis for the holotype or a specimen overlapping with it (which doesn't seem likely any time soon) or to do a detailed morphological inspection (like Borsuk-Bialynicka did on Opisthocoelicaudia) the both animals, which yields very different results for muscle attachment configurations.

Wrong. They do not need to be radically different (which is not a scientific concept—I know of no meaningful or useful scientific definition for this). All they need to be is diagnostically different (meaning phylogenetically informative). I’m sorry that a medial ridge is not “radical” enough or sexy enough for you, but it is diagnostic and phylogenetically informative as well as distinct. Being diagnostic and phylogenetically informative is what a scientist should care about, not if it is “radical” enough (which is a subjective and unscientific concept).

Also, Mannion did do a detailed morphological inspection and found that PVL 4628 and the Argyrosaurus holotype share no diagnostic characters.

It is unlikely that Elaltitan is closely related to the holotype specimen of Argyrosaurus since Argyrosaurus predominantly possesses characters that are present in basal (plesiomorphic) titanosaurs (e.g., “possesses carpal bones, which are unknown in all other titanosaurs. Secondly, the longest metacarpals are Mc. II and III, a feature Argyrosaurus shares with the putative basal titanosaur Janenschia") whereas Elaltitan possesses a suite of characters that indicate a derived position within Titanosauria (e.g., “the absence of postzygodiapophyseal laminae in dorsal vertebrae; diapophyses positioned dorsal to the prezygapophyses/ parapophyses in posterior dorsal vertebrae; middle–posterior dorsal neural spines that are sub-triangular in anterior view; strongly procoelous anterior caudal centra; and the presence of a prominent muscle scar on the posterolateral surface of the humeral deltopectoral crest.”) It is true that these basal features and derived features cannot be directly compared between the two taxa, but it is also true there is nothing to indicate that Elaltitan and Argyrosaurus are more closely related to each other than to any other titanosaur taxon. There is no shared feature between Elaltitan and Argyrosaurus that is not also found in other titanosaur taxa, therefore referring to the Elaltitan as an argyrosaur is completely unfounded (and actually, vague—what is an “argyrosaur”? What features diagnose that clade? If you want to include Elaltitan in the argyrosaur clade by definition, then that clade may be very large, including most, if not all of, the Titanosauria).

As for stratigraphy, Mannion also covers this, and surprise, surprise, they are differerent: “Additional support for the generic separation of Argyrosaurus and Elaltitan comes from recent stratigraphic revision of the Bajo Barreal Formation. Elaltitan comes from the lower member of the formation in the area of the Rio Senguerr, which is dated as middle Cenomanian–Turonian (Archangelsky et al., 1994; Bridge et al., 2000; Lamanna et al., 2002), whereas Argyrosaurus is probably from the upper member in the area of the Rio Chico, tentatively dated as Campanian–?Maastrichtian (Casal et al., 2007).”

Therefore, there is no justification for referring Elaltitan to Argyrosaurus. All your arguments for synonymy are wrong, ambiguous or hypothetical. On the other hand, arguments for their separation pass all standard taxonomic tests: they have different autapomorphies on the overlapping material, belong to different stratigraphic groups, and share no diagnostic characters that are not shared by other non-argyrosaur titanosaur taxa. In other words, synonymy of the two would require a considerable amount of new evidence. Thus, their taxonomic distinction appears to be justifiably warranted. Separating them requires special pleading.

BTW, I can send you the paper, or you could ask Phil, as he responded quite quickly when I asked for his papers (within a day). The best way to get a paper if it is not open access is to email the corresponding author (which is neither time consuming or terribly inconvenient). They are usually ecstatic that you would request there paper and I have had authors send me related papers that I didn’t even ask for!

I'll go into the Drusilasaura and Mendozasaurus argument at another time. I think this comment is already sufficiently long.
Join the community to add your comment. Already a deviant? Log In