Arctotherium angustidens: Biggest Bear Ever?

I have a bit of an obsession with animal size superlatives, and megabears* are among my favorites - that's right, I have interests beyond testudines. Anyways, in this blog's even more poorly-written past, I discussed purported giant hypercarnivores wherein I argued that the One-Ton(ne)-Hyperpredatory-Arctodus meme is unsubstantiated Godzillafication which somehow managed to infect even some peer-reviewed literature. Brian Switek wrote an article at the old Laelaps covering newer research which further demolished the mythology of the Giant Short-Faced Bear: it didn't have a short face (it did have a deep snout), or particularly long legs (somewhat of an optical illusion caused by a short back), and was probably a generalist omnivore like extant bears (presumably with some differences in niche, of course). Just as it seemed that speculations about One-Ton(ne)-Hyperpredatory-Bears would be a thing of the past, this happened:

* Definition pending.

Figure 2(1) from Soibelzon and Schubert (2011). The bear's height is ~ 3.3 meters (10' 10") and note how the legs are realistically flexed.

Arctotherium is composed of 5 South American species - of which A. angustidens is the earliest, largest, and apparently most predatory - and is the sister clade of Arctodus; the two are in turn part of the clade Tremarctinae which further includes Tremarctos (spectacled bear and kin) and Plionarctos (Soibelzon and Schubert 2011). The Arctotherium angustidens specimen of concern is not a new discovery, as it was found prior to 1935 during construction of a hospital in La Plata, Argentina (Soibelzon and Schubert 2011). It is presently composed of radii, ulnae, and humeri from both forelimbs; metacarpals, phalanges, and a scapula fragment were also recovered but unfortunately lost (Soibelzon and Schubert 2011). The dimensions of the bones* are incredible, the humerus has a length of 62 cm (2' 0.4") and the mid-shaft humerus width is 9 cm (3.5"); comparable maximum measurements of other giants bears are: Arctodus simus - 59.4 cm/6.4 cm; Ursus spelaeus - 44.8 cm/5.6 cm; and Ursus maritimus - 38.5 cm/4.65 cm (Soibelzon and Schubert 2011). For those who would prefer a more graphical comparison:

* The limbs had somewhat different measurements - the left was shorter and wider and the right vice versa. See below for why.

Humerus length in cm:

The number of specimens is in parenthesis. This does not include both limbs from the Arctotherium specimen in question, subsequently the average was re-calculated from Table 3.

Humerus mid-shaft width in cm:

Table 5 figures kept the specimens in question separate (unlike the prior example) so the range and average have been re-calculated. Even without the newly measured specimen, the average for Arctotherium was substantial (5.88 cm).

Using humerus greatest length, humerus mid shaft circumference, humerus greatest distance of distal epiphysis, and radius proximal epiphysis greatest diameter, the estimated weight for the giant Arctotherium angustidens specimen ranged from 983-2042 kg (2,167-4,502 lbs), with the value likely around the mean and median of 1588 and 1749 kg (3501-3856 lbs), respectively (Soibelzon and Schubert 2011). The other known specimens were given the same treatment (when possible), and a couple of them appeared to mass around a tonne (Soibelzon and Schubert 2011). This would seem to suggest that the giant Arctotherium angustidens specimen was not an outsized freak, and could represent a "normal" maximum size for the species. Before too many conclusions can be made, some more discussion of the specimen is in order.

Left humerus in caudal view. The scale bar is 10 cm (~ 4 inches) and the arrow points to a pathology.  Taken from Figure 3(1), Soibelzon and Schubert (2011). Compare with the humerus of Arctodus: here.

The most striking aspect of this specimen are the osteogenic changes to the deltoid crests of both humeri - more apparent in the left humerus pictured above, see arrow - and the distal third of the left radius shaft, which suggest a deep injury followed by infection and then new vascular growth over a long period of time (Soibelzon and Schubert 2011). Judging by the high degree of epiphyseal fusion, the specimen managed to become an old adult (Soibelzon and Schubert 2011). Humeral mid-shaft measurements gave on average larger estimated masses (Soibelzon and Schubert 2011 - Table 3), which makes me wonder if the measurements were artificially inflated by the injury and subsequent pathological growth. However, the limited data on Table 3 shows similar proportions with a somewhat smaller specimen:

	HMSC	HGL	HMSC/HGL
MLP 35-IX-26-5	26.5	61.5	43%
MLP 35-IX-26-6	26.2	62	42%
MACN 5132	22	54	41%
MLP 82-X-22-2	16	49.5	32%

Measurements in cm: HMSC = Humerus mid shaft circumference; HGL = Humerus greatest length. The first two rows are from the specimen in question.

Curiously, specimens of Arctotherium angustidens appear to differ considerably in limb proportions, for instance, two specimens have the same humeral circumference (22 cm) but the greatest diameter of the distal humeral epiphysis differs considerably (20.5 vs. 18 cm), one of which is larger than the giant specimen's maximum measurements (18.5 cm) (Soibelzon and Schubert 2011)... this is certainly confusing, and the value of compiling median and mean figures is readily apparent. While the mass of the giant Arctotherium angustidens cannot be precisely pinned down, the available evidence suggests it exceeded all other bears in size.

... or does it?

Incredibly, one mass regression of Indarctos atticus exceeded 3 metric tonnes, although predictably it was found to be highly improbable (Finarelli and Flynn 2006) and, along with fellow Mio-Pliocene bear Agriotherium, they are not believed to have reached the same size as Arctodus simus or Arctotherium angustidens (Soibelzon and Schubert 2011). Polar bears (U. maritimus) can get extremely large*, with one specimen shot in Alaska in 1960 purportedly standing 3.39 m (11' 1.5") and weighing 1002 kg (2210 lbs); the whereabouts are apparently unknown and the skull was never submitted for measurement (Wood 1981). Assuming the record is genuine, I'm wondering if the height included unnaturally straitened legs or was in fact the length lying down and outstretched, which would certainly be easier to take. Considering the largest Arctotherium angustidens humeral length is about 160% as large as the largest polar bear measurement included in Soibelzon and Schubert (2011), it would take one freakish polar bear to get that tall, and it would probably weigh a lot more than a tonne. Polar bears reached their largest sizes in the late Pleistocene (Soibelzon and Schubert 2011) and Ursus maritimus "tyrannus" apparently had an ulna 44 cm in length (see Markus Bühler's Bestiarium, comment #5), which is of course smaller than that of the specimen in question's (57 cm), but still surprisingly large for a single specimen. Information on these giant polar bears is unfortunately quite hard to come by, but I think it's safe to assume it didn't exceed Arctotherium angustidens in size, at least regularly.

* DeMaster and Stirling (1981) give maximum figures of: mass 800 kg (1764 lbs), nose-tail length 2.5 m (8' 2"), and shoulder height 1.6 m (5'3").

Exactly how the average sizes of the giant bears compare is difficult to determine at the present time - particularly when gender isn't obvious. However, as weights of a tonne or more appear to have been reached by Arctotherium angustidens regularly, the average must have considerably exceeded that of other giant bears, which have only been demonstrated to exceed a tonne in one instance. The authors' assertion that Arctotherium angustidens was the biggest bear ever is well-supported, until something bigger turns up.

---

This post has run long enough, I'll write a followup on the proposed ecology of Arctotherium angustidens shortly.


References:

DeMaster, D. P., and Stirling, I. (1981). Ursus maritimus. Mammalian Species 145, 1-7. Available.

Finarelli, J. A., and Flynn, J. J. (2006). Ancestral State Reconstruction of Body Size in the Caniformia (Carnivora, Mammalia): The Effects of Incorporating Data from the Fossil Record. Systematic Biology 55(2), 301-313. doi: 10.1080/10635150500541698. Available.

Soibelzon, L. H., and Schubert, B. W. (2011). The Largest Known Bear, Arctotherium angustidens, from the Early Pleistocene Pampean Region of Argentina: With a Discussion of Size and Diet Trends in Bears. Journal of Paleontology 85(1), 69-75. doi: 10.1666/10-037.1

Wood, G. L. (1981). The Guinness Book of Animal Records. Guinness Superlatives: Middlesex, Great Britain.

Later This Year, On The Lord Geekington!

For my own good, I've decided to keep track of the posts I have started but never finished, and the ones I have promised but never gotten around to. I've set up a page to keep track of them (see toolbar, to be updated regularly), and here are the contents thus far:

Research blogging. I'll finally get on the bandwagon (if I can get the page to work) and discuss some recent papers on clades I have an unusual fondness for (turtles, vultures, remipedes, loricariids, cephalopods, et cetera) which receive little attention from the science blogosphere at large. I'm counting on you Google Scholar alerts!

'Cadborosaurus' analysis. A sequel to my treatment of Heuvelmans' Many-Finned (reports, analysis), I will utilize anecdotal data from LeBlond and Bousfield's Cadborosaurus: Survivor from the Deep to determine if their proposed cryptid is actually suggested by the reports they included. After reading the book, I suspect that the proposed morphology for 'caddy' was based mostly on the Naden Harbour carcass; if the carcass is assumed to be mundane and/or over/misinterpreted, 'caddy' will either get much simpler (say, elongated body + ungulate-like head + big eyes), or no well-supported patterns will emerge at all. One controversial aspect is lumping different eyewitness traits (i.e. horse-like head, cow-like head, giraffe-like head, et cetera), so I'd like to see how big a difference different degrees of lumping will make. This will all inevitable tie in with the controversy over whether the very concept of a cryptid taken from numerous eyewitness reports has any validity. I'd also like to record the number of characteristics per report and see if there is a trend with distance or possibility of a hoax. Obviously this is going to be a whole serious of posts!

The Hagelund specimen. There's something in the works about this one and I'm planning on giving some additional background and commentary upon publication.

'Caddy' Reports. Some of them are really quite interesting!

The Canvey Island Monsters. Cryptids that are definitely known fish, although this does not appear to be widely acknowledged.

Clade-by-clade turtles. I'm thinking of reviewing turtles as a whole by major clade (mostly 'families', perhaps some well-supported sub-families). It just seems inevitable.

Brackish and Saltwater turtles. There are a few non-seaturtles that can venture out to sea (previously mentioned here) and the ability for 'freshwater' turtles to survive in brackish water is downright common.

Teeny Turtles.Turtles on average are rather large creatures (previously discussed here), and it would seem that at small sizes the shell would offer little protection and become an unnecessary burden. However, very small turtles (< 15 cm strait shell length) can thrive in areas with crocodilians and other potential turtle-crushers. I'll review the smallest species of all with emphasis on predation, growth, niche, and other relevant aspects of their life history.

Jaguarundi and the Philosophy of Genera. I'll discuss why placing Jaguarundi in the genus 'Puma' is a huge mistake (blatant paraphyly, for one thing) and offer my own philosophy on how genera should and shouldn't be used. The concept of a genus is subjective, but I think the best approach would be to construct them of species which are obviously closely related (i.e. the Right Whales in Eubalaena, vs. the Bowhead Whale in Balaena) and thus may be split or lumped over time. I think the concept of a 'subgenus' is useless and they should just be bumped up to a proper genus (see my thoughts on pangolins).

Shrink-wrapped whales. Inspired by discussion with Markus Bühler (of Bestiarium) and discussion on SV-POW! (part 1, part 2) I'll reconstruct extant whales in the most dreadful manner possible to discuss how some extinct forms probably didn't look like quasi-reptilian monsters in life.

Cleaned by a vulture. For some reason, American Black Vultures (Coragyps atratus) engage in lots of interspecific grooming.

Snake Eels. Eels with necks! Kinda!

Clade-by-clade remipedes. Another inevitable topic.