I myself at one point gave a mass estimate of 150-250 tonnes and length of about 70 meters.
I think I was probably wrong, and a recent size comparison with Argentinosaurus demonstrates why:
Since my GDI mass estimate for Argentinosaurus is about 64 tonnes, I knew 100+ tonnes estimates for Amphicoelias had to be wrong.
I did a completely new reconstruction of Amphicoelias fragillimus, with missing parts restored after Amphicoelias altus and compared that to a reconstruction of a posterior dorsal of Argentinosaurus. As you can see, in both total width and centrum length, Argentinosaurus just edges out A. fragillimus in size. I should also point out that the posterior dorsals of diplodocids are usually taller than the anterior dorsals, while the opposite is the case in at least some titanosaurs:
Diplodocus dorsal sequence
Malawisaurus dorsal sequence
This means, of course, that we are not comparing apples and oranges and therefore, the absolute size of a vertebra can be misleading, especially when comparing taxa that are not closely related. There are several things one needs to keep in mind, then, when trying to get an idea of overall size of a sauropod based on vertebral size (or indeed, any animal):
1. Serial position of vertebra (i.e., the 1st dorsal, 5th dorsal, 8th dorsal, etc.)
2. Proportions (length to width, etc.)
3. Phylogeny (diplodocid, brachiosaur, etc.)
4. Absolute size
These points may seem obvious, but often times it appears only absolute size is discussed. In fact, this is probably the least important factor in determining the total size of the animal. An example to compare would be Giraffatitan and Diplodocus. Both Giraffatitan and Diplodocus each have a dorsal vertebra that measures 107 cm in total height. If one were to study these vertebra in isolation without respect to points #1, #2 and #3 that I listed, and only compared absolute size (point #4), one might get the impression that both taxa were approximately the same size. In fact, they are not, as you may well be aware. Rigorous reconstructions show that Giraffatitan is consistently estimated to have been about 25-30 tonnes and about 23 meters long, whereas Diplodocus is usually estimated to be about 10-15 tonnes, and about 25 meters long. Clearly, Giraffatitan was bigger in mass than Diplodocus.
Focusing on points #1, #2 and #3 show why, even though both Giraffatitan and Diplodocus have a dorsal vertebra of the same height, one is clearly larger.
First, the serial position of the vertebrae are different. In Giraffatitan, it is the 4th dorsal that is 107 cm tall, while it is the 10th dorsal in Diplodocus. In Giraffatitan the anterior dorsals are taller than the posterior dorsals, while the opposite is the case in Diplodocus.
Giraffatitan dorsal (top right)
Second, consider the proportions. The dorsal of Giraffatitan is considerably wider proportionally (and in absolute terms) than the dorsal in Diplodocus. In Giraffatitan the width of the diapophyses is the same as the height of the entire vertebra, over 107 cm wide (Janensch, 1950). In Diplodocus, the widest dorsal (the 4th) is only 75 cm across the diapophyses, and is only 49.5 cm in the 10th dorsal (Lull, 1919).
Third, consider the phylogeny. We can use phylogenetic bracketing to get a rough idea of the proportions of an animal, even if known from incomplete material. We know, for instance, that diplodicids tend to be long and skinny, whereas brachiosaurs tend to be shorter in total length, but also more robust and therefore heavier than a diplodocid of the same length.
So what does this all mean for Amphicoelias fragillimus, especially in comparison to Argentinosaurus?
First, consider the serial position of the vertebra. The dorsal of Amphicoelias fragillimus appears to be one of the last, if not the last dorsal, and is commonly cited as the 10th (Carpenter, 2006). The vertebra of Argentinosaurus I have chosen to compare it with was originally described as the 2nd (Bonaparte and Coria, 1993), but subsequent analysis suggests it is a posterior dorsal, possibly the last or penultimate dorsal (Novas and Ezcurra, 2006; Salgado and Powell 2010).
Second, consider the proportions. Amphicoelias is clearly much more gracile than Argentinosaurus. In fact, it appears likely that the centrum would have been shorter and skinnier than in Argentinosaurus. Even the diapophyses appear to be not to have been as wide. Obviously, Amphicoelias is considerably incomplete, even the lone dorsal. This is where the third point comes into play.
Phylogenetically A. fragillimus was a diplodocid, and appears to be close to A. altus. This gives us a way to confidently restore the missing bits. The parts in gray were restored after A. altus (Osborn and Mook, 1921). Also, we can compare it to more complete diplodocids, like Diplodocus itself and compare proportions with it.
Restoring the incomplete vertebra of A. fragillimus off of A. altus, gives these estimated measurements:
* Total height: ~2400 mm
* Diapophyses width: ~1070 mm
* Centrum width (posterior face): ~600 mm
* Centrum length (total): ~490 mm
In Diplodocus, these measurements are (10th dorsal; from Lull, 1919):
* Total height: 1070 mm
* Diapophyses width: 495 mm
* Centrum width (posterior face): 325 mm
* Centrum length (total): 290 mm
Here are some ratios of Amphicoelias to Diplodocus:
* Total height ratio: ~2.24
* Diapophyses ratio: ~2.16
* Centrum width: ~1.84
* Centrum length: ~1.69
Taking the average of these ratios, we get that, on average, A. fragillimus was "only" 1.9825 times larger than Diplodocus in linear dimensions, which would suggest being about 7.79 times heavier, or approximately 78 tonnes, assuming Diplodocus was 10 tonnes (a mass estimate I got by doing a GDI of Gregory S. Paul's multi-view Diplodocus reconstruction). My own mass estimate for Argentinosaurus is about 64 tonnes, and for the largest Alamosaurus specimen about 74 tonnes. This means that there are no sauropods yet known that we can confidently say are heavier than 100 tonnes. Amphicoelias fragillimus still was possibly the largest sauropod, although not nearly by the margin often claimed: only by a few tonnes. Since any mass estimates are imprecise, I'd say it is approximately a tie for first place between Alamosaurus and Argentinosaurus for the largest known sauropod.
So, the whales win - by a wide margin*.
*There are at least 4 living whale species for which specimens are known that reach over 100 tonnes: the blue whale, fin whale, bowhead whale, and Northern Pacific right whale.
Bonaparte, J. F., and R. A. Coria. 1993. Un nuevo y gigantesco sauropodo titanosaurio de la Formacion Rıo Limay (Albiano–Cenomaniano) de la Provincia del Neuquen. Ameghiniana 30:271–282.
Carpenter, K. 2006. Biggest of the big: a critical re−evaluation of the mega−sauropod Amphicoelias fragillimus Cope, 1878. New Mexico Museum of Natural History and Science Bulletin 36: 131–137.
Janensch, W. 1950 Die Wirbelsäule von Brachiosaurus brancai ("The vertebral column of Brachiosaurus brancai") Palaeontographica Supplement VII (1), teil 3, leif 2:31-93
Leonardo Salgado & Jaime E. Powell (2010): Reassessment of the vertebral laminae in some South American titanosaurian sauropods, Journal of Vertebrate Paleontology, 30:6, 1760-1772
Lull, R. S. 1919. The sauropod dinosaur Barosaurus Marsh. Memoirs of the Connecticut Academy of Arts and Sciences 6:1–42.
Novas, F. E., and M. Ezcurra. 2006. Reinterpretation of the dorsal vertebrae of Argentinosaurus huinculensis (Sauropoda, Titanosauridae). Ameghiniana 43(4, Supplement):48R–49R.
Osborn, H. F., and C. C. Mook. 1921. Camarasaurus, Amphicoelias and other sauropods of Cope. Memoirs of the American Museum of Natural History, n.s. 3:247–387.
100 tonnes on average.
Whales are living animals and they were weighed.Quite a longtime before i searched in some genuine whale websites.In all those sites,only blue whale was given an average weight of 100 tonnes /100 tonnes plus.Those 3 other Whales where given the 60 - 70 tonne range which is the same i have heard in videos.
Except Cope,how many people have seen the real partial vertebra of Amphicoelias fragillimus before it was lost ?
I'm having a doubt if it was real.
I don't know how many people saw the real vertebra of Amphicoelias fragillimus. Many people have your skepticism about its existence. However, Kenneth Carpenter noted, "The immense size of the measurements given by Cope and the inferred vertebra size have been met with skepticism (several individuals, verbal to Carpenter), with typographical errors in the measurements being the most commonly assumed explanation. There is, however, every reason to accept Cope at his word. First, Cope never made any subsequent corrections in his publications; furthermore, his reputation was at stake. Marsh, who was ever so ready to humiliate Cope, never called into question the measurements. Marsh is known to have employed spies to keep tabs on what Cope was collecting, and it is quite possible that he had independent confirmation for the immense size of A. fragillimus. Osborn and Mook (1921) accept Cope’s measurements without question, as does McIntosh (1998). Thus, there is historical precedence for accepting the measurements as correct." (www.gardenparkdinos.com/wp-con… The particular fact that Marsh and Cope had such a big rivalry, and that Marsh employed spies, and that Marsh never called into question the specimen strongly indicates that it was real and that the measurements were accurate.
Actually, I just found out (after I did this post) that Amphicoelias has been included in two published phylogenetic analysess: Whitlock (2011) & Rauhut et al. (2005). Whitlock (2011) found it to be a basal diplodocoid more derived than Amazonsaurus and Haplocanthosaurus (successively), but less derived than the rebbachisaurids (with the most basal rebbachisaur being Histriasaurus, followed by Rebbachisaurus). I believe only A. altus is scored.
However, it's position is not well resolved, and depending on which other taxa they include it results on an unresolved polytomy, so I'd say the results are very tentative.
If we take these results as valid, then Amphicoelias may have looked more like a cross between Amazonsaurus and Histriasaurus, unfortunately both are terribly incomplete, so it doesn't help much with trying to estimate size.
Rauhut O, Remes K, Fechner R, Cladera G, Puerta P. 2005. Discovery of a short-necked sauropod dinosaur from the Late Jurassic period of Patagonia. Nature 435: 670–672.
Whitlock, J.A. 2011. A phylogenetic analysis of Diplodocoidea (Saurischia: Sauropoda). Zoological Journal of the Linnean Society 161:872–915.
The sole dorsal vertebra (either the 1st or 2nd) is exceptionally wide at 168 cm across the diapophyses, but it is much shorter than the dorsals of Argentinosaurus in both total height and centra length. Also, since the positions of the dorsal vertebrae of Argentinosaurus have been reconsidered, we don't have any anterior Argentinosaurus dorsals, so the width comparisons could be misleading.
The height of the dorsal of Puertasaurus is only 106 cm tall (Novas et al., 2005), compared to at least 145 cm for the known dorsals of Argentinosaurus (see above).
The total centra length for Puertasaurus is not mentioned, but based off of the published drawings, it looks to be about 42 cm, which is significantly shorter than the 50 cm + lengths in Argentinosaurus. Of course, those are posterior dorsals, but the thing to remember is that in many titanosaurs, the anterior dorsals are generally longer than the posterior dorsals, which means the posterior dorsals of Puertasaurus were likely shorter than 42 cm, meaning they were potentially 10 cm shorter than in Argentinosaurus.
Also, consider the width of the diapophyses. In a juvenile specimen of Alamosaurus, the anterior dorsals are 50 cm wide across the diapophyses, and the posterior dorsals are 32 cm (Lehman and Coulson, 2002). Which means the anterior dorsals are more than 50% wider. If this was true in Argentinosaurus, than the anterior dorsals might be ~175 cm wide (based off of 116 cm wide posterior dorsal I reconstructed above), which would be wider than Puertasaurus. This ratio is even more extreme in Opisthocoelicaudia (Borsuk-Bialynicka, 1977), where the 2nd dorsal is 66 cm wide across the diapophyses, and the 11th dorsal is 41 cm wide, for a ratio of more than 60% wider, which would lead to an estimated anterior dorsal width of Argentinosaurus of ~185 cm. So my guess is that the anterior dorsals of Argentinosaurus would actually be wider than Puertasaurus.
Taking these things into consideration, Puertasaurus was probably smaller than Argentinosaurus, Alamosaurus and Amphicoelias.
BORSUK-BIALYNICKA, M. 1977. A new camarasaurid sauropod Opisthocoelicaudia skarzynskii gen. n., sp. n. from the Upper Cretaceous of Mongolia. Palaeontologia Polonica, 37:5–64.
Lehman, T.M. and Coulson, A.B. 2002. A juvenile specimen of the sauropod dinosaur Alamosaurus sanjuanensis from the Upper Cretaceous of Big Bend National Park, Texas. Journal of Paleontology 76: 156–172.
Novas, F.E., Salgado, L., Calvo, J., and Agnolin, F. 2005. Giant titanosaur (Dinosauria, Sauropoda) from the Late Cretaceous of Patagonia. Revista del Museo Argentino de Ciencas Naturales 7: 37–41.
Yes, you are correct, generally size estimates are over-inflated when initially reported (unfortunately). I'm sorry to hear that a number of people have taken my older estimate too seriously Oops. Hopefully that will now be corrected.
As for the validity of A. fragillimus, I'll let Carpenter (2006) explain: "The immense size of the measurements given by Cope and the inferred vertebra size have been met with skepticism (several individuals, verbal to Carpenter), with typographical errors in the measurements being the most commonly assumed explanation. There is, however, every reason to accept Cope at his word. First, Cope never made any subsequent corrections in his publications; furthermore, his reputation was at stake. Marsh, who was ever so ready to humiliate Cope, never called into question the measurements. Marsh is known to have employed spies to keep tabs on what Cope was collecting, and it is quite possible that he had independent confirmation for the immense size of A. fragillimus. Osborn and Mook (1921) accept Cope’s measurements without question, as does McIntosh (1998). Thus, there is historical precedence for accepting the measurements as correct."
Carpenter, K. 2006. Biggest of the big: a critical re−evaluation of the mega−sauropod Amphicoelias fragillimus Cope, 1878. New Mexico Museum of Natural History and Science Bulletin 36: 131–137.
"I've seen several posts where they came up with weights figures over 300 tonnes."
Oh jeez, 300 tonnes?! That's definitely absurd
"There are the dinos enthusiasts who are happy with rigorous estimates hinting on the biggest land animals ever, and there are those who wish absoutely to get an animal bigger than the (boring ?) B. musculus."
Haha, very true...we need more of the former
I believe Hartman stated he felt Puertasaurus and Alamosaurus (he also tentatively put Argentinosaurus into the same category) are in the same size class after he restored both of them, but who is actually larger isn't certain. Nima Hassani's Puertasaurus seems a lot bigger though...
I don't think 4 tonnes here or there in animals this large fragmentary are _that_ significant either, it's pretty much the equivalent of only 400 kg difference in an 8 tonne animal.
Which specimen of Diplodocus is being referred to here? I'm sort of assuming CM 84...
Seems to suggest a less-elongate animal than Diplodocus anyway, and a far more believable weight estimate to boot.
Since only one view of A. fragillimus was published, we can't be sure how long it was, but based on A. altus, the centrum would have been fairly short.
Forgot to add - I think the disparity between the length of the centrum and the width of the centrum (and vertebrae as a whole) relative to Diplodocus suggests that the latter perhaps isn't a great basis for estimating the size to begin with.
I would be interested in seeing how these measurements shape up against other diplodocids, and basal diplodocoids (to align with Cau's analysis).
Unfortunately, most of the basal diplodocoids are not well described (often lacking measurements and good photos/drawings), so it would be difficult to do a comparison other than with the well-known diplodocids: Apatosaurus and Barosaurus.
^Andrea Cau recovers it as a basal diplodocoid IIRC but it could be just because of the lack of codeable characters.
Thank you Zach! people mostly use absolute size and just a single measurement to boot, I had estimated a centrum length as low as 420mm based on measurements of the "seismosaurus" diplodocus compared to CM84 and is good to know that others have looked beyond just the total height of the vertebrae.
btw you made a typo in your first image, it says the centrum width is almost 6m.
Ah, very interesting, thanks for that note. I agree that it is a real possibility. This would therefore make the mass estimates above almost certainly too high, considering the differing proportions of basal diplodocoids.
Good to know you also got a low centrum length estimate. It is true that depending on the Diplodocus specimen you choose to compare, you will get different mass estimates.
Thanks for pointing out the typo! Haha, 6 m - then it would be Godzilla sized....I will go and correct it!
Will that affect the superficial appearance of Amphicoelias much? Is it possible that it looked more like a longer necked Rebbachisaur superficially?
If Dippy is 10-15 tonnes, then Amphicoelias, based on your reconstruction, would be 78-117 tonnes, still humongous.
Now I need to do a rebbachisaur comparison and see where that leads!
If it was closer to a rebbachisaur-grade diplodocoid (which is definitely possible, it is rather incomplete after all ), then I think the mass estimates I gave above are probably over-estimates. Basal diplodocoids tend to have proportionally shorter necks and tails, proportionally tall dorsals, and at least Nigersaurus is extremely pneumatic. All of these factors could chop off 5-6% of the mass estimate easily, which would possibly put it below the largest Alamosaurus specimens.
I think Diplodocus is probably closer to the 10 tonne mark, I did a GDI off of GSP's skeletal, and got 9,999 kilos for an estimated mass. Even being generous in giving it more soft tissue than GSP does, and assuming a higher density, I doubt that the D. carnegii specimen CMNH 84 (which is the specimen I was using the measurements for comparison for above) got much more than 12 tonnes. So we're talking about a 78-94 tonne range as the most likely guess. Keep in mind that A. fragillimus might have even been more pneumatic than other diplodocids, and could therefore had a lower density. But yes, even in that range, it is still humongous...just less humongous than popularly imagined (i.e., <100 tonnes).
Anyway, the mass estimate here is misleading the size enthusiasts a bit. Volume would completely change the result. Assuming that you used an SG of 0.7, your A. fragillimus would end up at more than ~111 kiloliters in volume, larger than the average blue whale*. So you may wanna change that last part, the whales still do not win.
And how wide and long is the centra of Haplocanthosaurus?
*aquatic animals have densities around that of water, mass and volume are practically the same number for them
Hatcher, J.B. 1903. Osteology of Haplocanthosaurus with description of a new species, and remarks on the probable habits of the Sauropoda and the age and origin of the Atlantosaurus beds; additional remarks on Diplodocus. Memoirs of the Carnegie Museum 2:1-75.
This would make A. fragillimus about ~3.051 times wider and ~3.356 times longer than CMNH 572, basing on centrum dimensions, the assumption that the A. fragillimus vertebra is a D13, and a Haplocanthosaurus-like build.
59.8 cm / 19.6 cm = ~3.051
49 cm / 14.6 cm = ~3.356
"This would make A. fragillimus about ~3.051 times wider and ~3.356 times longer than CMNH 572"
Yes, but now you are comparing ratios based on a Diplodocus based reconstruction. You'd first have to do a reconstruction of A. fragillimus based on Haplocanthosaurus for the comparison to ultimately be valid.
Yes, because the neck and tail of derived diplodocids don't count for much in terms of volume or mass. So you could have a shorter overall animal in a basal diplodocid that has the same mass because they have shorter necks and tails, but slightly longer torsos. That is why I am not comparing overall length, but estimated volume/mass.
"Volume would completely change the result."
No it wouldn't, my mass estimates are based off of the volume. My mass for Diplodocus is based off of a multi-view skeletal by Greg Paul. The total volume of it came to be 12.94 m^3 (12.94 kL). I assigned a density for the neck a SG of 0.3, the torso and tail a SG of 0.8, and the limbs a SG of 1.0, resulting in an overall SG of ~0.767 and total mass of ~10 tonnes (I'm rounding...I actually got a specific mass of 9,960.8 kg, but that is misleadingly precise).
Then, if you look at the ratios of how much larger Amphicoelias fragillimus is to Diplodocus, given in my main journal article, you get that it was an average of 1.9825 times larger in linear dimensions. Since volume (and therefore mass) scales to the third power of length: (1.9825)^3=7.7918. Multiply that factor by the volume of Diplodocus and you get: 7.79*12.94 kL=~100.8 kL. Then multiply that by the SG of .767 and you get: 100.8*.767=~77.32 tonnes (this is slightly less than the figure I give in the article because I rounded to two decimal points here).
I'm not sure what the average blue whale size is, but the largest individual accurately measured by scientists massed over 170 tonnes (www.sararegistry.gc.ca/virtual…. A specimen near 190 tonnes is also known. So the whales win by volume and mass by a large margin. Individuals of bowhead whales, north pacific right whales and fin whales all are known to get larger than 100 tonnes, in some cases between 110-130 tonnes. So the whales definitely win.
The average blue whale is estimated at ~92.67 tonnes. I think the average should be used due to A. fragillimus' really small known sample size.
I'm pretty sure ~100.8 kiloliters beats ~92-93 kiloliters. So yeah, it completely changed the result there.
According to your reference, that average is for the Southern Hemisphere population. You may not think that matters, but many whales species have populations that vary in size, sometimes by significant amounts. For instance the pygmy blue whale is one of the smaller subspecies of blue whale, and it inhabits the southern hemisphere.
"I think the average should be used due to A. fragillimus' really small known sample size."
Well, with a sample size of one, you can't have an average, so we don't know if A. fragillimus was "average" or not. You may think it is fairer to compare the average size of a whale species, but it really isn't because in order to fairly compare them, you would need average sizes from both species. Since we don't have a statistically significant population size for A. fragillimus, it doesn't really matter. But, as you can see, the average mass for the Southern Hemisphere populations (which includes the smaller subspecies) still have a mass greater than I estimate for A. fragillimus, by at least 10 tonnes. So yes, the whales do win.
Furthermore, when comparing the size of animals, it is standard to compare masses, not volumes. So I don't understand why you are comparing volumes. Volumes can vary greatly in individuals depending on the activity, for instance. For example, when a blue whale is feeding it can expand the ventral pleats on its lower jaw and greatly expand it's volume temporarily. An animal's volume can simply change with it breathing in or out. That is why no one uses volume to compare animals when trying to determine which one is larger. They are very variable. Mass can also vary, depending on whether an animal has recently fed or not, if they are pregnant, etc. However, mass is the standard when comparing which animal species is larger than another, and I am not aware of who uses volume in lieu of mass when comparing organism size. So again, I am greatly confused as to why you are comparing volumes.
However , it can be shown that an individual is likely to be about 1 standard deviation away from the average (spirospero.net/BELL2.JPG). In fact, about 2/3 of individuals will be within one standard deviation (en.wikipedia.org/wiki/Standard… of the mean (average). What this means is that if the average mass of a species is 50 tonnes, and the standard deviation (which is basically the average amount an individual's size differs from the average) is 5 tonnes, about 68% of individuals will be between 45 to 50 tonnes, with about 34% in the 45-50 tonne group, and 34% in the 50-55 tonne group.
This means the lone specimen of A. fragillimus is likely, but not certainly, within 1 standard deviation of the average size. It is almost certainly not average.
"And I'm using volume because derived sauropods are so pneumatic that mass alone may be misleading when compared against similarly-sized but denser animals."
Except, again, nobody compares volumes when talking size, they talk mass, so you're doing something no one else does, and it is not actually a fair comparison (and I explained why that is earlier). We know sauropods are pneumatic, but so are birds, and nobody compares the volumes of birds to determine which is the largest bird, they use mass. Using volume to compare sizes is misleading, not mass, because volume is much more variable (as I already explained).