Friday, February 10, 2006

They bite, they grow to huge sizes, they locate human corpses: the snapping turtles, part I

Until recently I had very little to do with turtles. I’d written some brief pieces on the taxonomy of Indian Ocean giant tortoises (more on that here), and naturally I’ve had to do consultancy work on giant fossil taxa, like the Cretaceous protostegids and Stupendemys, a Venezuelan pleurodire that exceeded 3 m in length and must have weighed a couple of tons. But it’s the sharing of my office with friend and colleague Sarah Fielding that has brought me into the surreal Gaffneyesque world of turtles more than before, and, oh, the adventures I’ve had…

We have the fascinating tale of Branston the picked turtle and the sorry trade in African helmeted turtles Pelomedusa subrufa, my adventures with Sarah, Dave and a new araripemydid likened to Jennifer Lopez (Fielding et al. 2005), and, best of all, the story of the perplexing, colour-changing Cuthbert, an as-yet-unidentified emydid who lived in my office before being driven specially down to Exeter. He died, and is now stuck in a freezer. A deadline for a short manuscript dealing with the pleurodires of the Brazilian Crato Formation has come and gone, but there isn’t really a desperate rush on it and I’ll be ok. I’ve written most of it. Anyway, all of this – in due time – will be reported here.

I will not disagree with Matt Wedel’s contention that among the most interesting turtles of them all are the snappers: the two extant members of Chelydridae, Chelydra serpentina (the American snapping turtle) and Macroclemys temminckii (the Alligator snapping turtle) [incidentally, the Big-headed turtle Platysternon has been included at times within this family, though whether it’s even close to chelydrids is now disputed]. Bizarre skulking benthic aquatic ambush predators, they are famous for being cryptic, for pretending to look like the lumpy bumpy muddy substrate they camouflage themselves on, and for biting pretty much most things that come within reach. Alligator snappers apparently only leave the water to bask or lay eggs, while snappers will walk for long distances across land (up to 16 km) to find suitable breeding sites.

Snappers (from hereon used specifically for Chelydra) are omnivorous, and as well as all manner of invertebrates and small vertebrates, will also eat various plants. They get to about 50 cm in carapace length and have three low keels running the length of the carapace. This strikes me as a curious parallel with leatherbacks, and also with some placochelyid placodonts, but I don’t think this means much as leatherbacks are pelagic deep-divers that swim incessantly and prey on cnidarians, while placochelyids appear to have been durophagous benthic foragers that didn’t spend their time sitting on the substrate. Anyway, I digress.

With a range extending from Nova Scotia and southern Quebec to Ecuador, snappers are reasonably variable, and populations differ in the proportional size of various of their scutes, and how their necks are ornamented. Again we come across what is becoming a frustratingly familiar theme: are they just one species? This has proved to be quite a contentious area. Traditionally, the species has been divided into four supposed subspecies: C. serpentina serpentina (of continental Canada and the US), C. s. acutirostris (of Central America and northern South America), C. s. rossignonii (of Mexico, Belize and nearby) and C. s. osceola (restricted to Florida). Feuer (1971) argued that C. s. serpentina and C. s. osceola graded into one another, and that the latter was therefore not worthy of distinction, and other workers have expressed doubts about the supposed distinction of the other subspecies. However, Phillips et al. (1996) found that C. s. serpentina and C. s. osceola were distinct taxa, and more closely related to each other than to C. s. acutirostris and C. s. rossignonii. Furthermore, the mtDNA data they compiled suggested that both these latter forms should be regarded as distinct species. What was once the one species C. serpentina therefore becomes three, if Phillips et al. (1996) are right.

Are they right? Well, their conclusions were seriously challenged by Sites & Crandall (1997) who argued that Phillips et al. (1996) did not ‘present any species concept as a testable hypothesis’ and that ‘data [was] collected in the absence of any conceptual framework for diagnosing species boundaries’ (p. 1289). Sites & Crandall (1997) didn’t support species status for the ‘taxa’ recognised by Phillips et al. (1996), though they did suggest that snappers from Ecuador might deserve specific recognition. More recently, I note that Walker et al. (1998) have published a paper with the title ‘Phylogeographic uniformity in mitochondrial DNA of the snapping turtle’: I haven’t seen this but it at least implies that they did not find evidence for polytypy within snappers. So right now things are undecided, and Chelydra may or may not turn out to be polytypic.

Alligator snappers are altogether different from snappers. On the taxonomic side of things, check the literature and you see an annoying amount of inconsistency on the generic name, with authors switching between Macroclemys, Macroclemmys and Macrochelys. I'll assume Ernst & Barbour (1989) are right - they usually are - and that the two variants arose as typos. They (alligator snappers, not Ernst & Barbour) have a carapace with very tall keels, each of which possesses high, sub-pyramidal knobs, a far more massive and nasty-looking head, and overall they’re more gnarly looking, with papillae and skin fronds projecting from their limbs, neck and elsewhere. Unlike snappers, they sport a worm-like lure on the tongue. By sitting with the mouth open, and by waggling the lure, they … err.. lure in prospective prey.

What’s more, this is one of the biggest freshwater turtles in the world, reaching 90 cm and 100 kg at least. Conant & Collins (1991) give a maximum verified weight of 143 kg. Other freshwater turtles are longer, but not as heavy (though read on). Asian giant softshells Pelochelys bibroni, for example, may exceed 120 cm. Ok, so as it happens there are the recently reported enormous softshells of Hoan Kiem Lake in Hanoi, Vietnam, which, according to published photos, appear to be about the size of a car (well, nearly. They’re said to be up to 2 m long). They’re clearly the world’s biggest freshwater turtles, but right now their size isn’t verified, nor do we even know what they are (they’re probably Rafetus swinhoei, but it’s also been proposed that they represent a new species, dubbed R. leloii).

Getting back to alligator snappers, there are, as you’d expect, lots of anecdotal tales of animals bigger than the ‘verified’ ones, and while none of these rumoured individuals have been confirmed, they’re fun to think about. In 1937 a 183 kg specimen was supposedly caught in the Neosho River in Kansas. More famous – it might be the most famous alligator snapper of all time, though whether or not it was a real animal is open to question – was the gigantic individual, dubbed variously the ‘Beast of ‘Busco’ or ‘Oscar’ (after Oscar Fulk, allegedly the first witness), that supposedly dwelt in Fulk’s Lake, Churubusco, Indiana. ‘It’ (supposedly the same individual) was seen on numerous occasions since 1898. A 1949 sighting by farmer Gale Harris described it as about 1.8 m long and 1.5 m wide, and numerous attempts to catch it ensued. These involved the building of a stockade around the turtle, attempted capture by divers, the use of a dredging crane and pumps, and a female sea turtle imported to act as bait! (Newton 2005). Apparently two people died in one of these dumb schemes. Needless to say the animal was never measured. It was also claimed to be ‘the size of a dining-room table’ and estimated to be nearly 230 kg. Well, ok, maybe.

More to come on snappers soon (you can now see part II and part III). Matt, you have a lot to answer for. And for the latest news on Tetrapod Zoology do go here.

The pic above is from the Bulgarian bloodice site.

Refs - -

Conant, R. & Collins, J. T. 1991. A Field Guide to Reptiles and Amphibians of Eastern and Central North America (Third Edition). Houghton Mifflin (Boston), pp. 450.

Ernst, C. H. & Barbour, R. W. 1989. Turtles of the World. Smithsonian Institution Press, Washington, D. C. & London.

Feuer, R. C. 1971. Intergradation of the snapping turtles Chelydra serpentina serpentina (Linnaeus, 1758) and Chelydra serpentina osceola Stejneger, 1918. Herpetologica 27, 379–384.
Fielding, S., Martill, D. M. & Naish, D. 2005. Solnhofen-style soft-tissue preservation in a new species of turtle from the Crato Formation (Early Cretaceous, Aptian) of north-east Brazil. Palaeontology 48, 1301-1310.

Newton, M. 2005. Encyclopedia of Cryptozoology: A Global Guide to Hidden Animals and Their Pursuers. McFarland & Company, Jefferson, North Carolina.

Phillips, C. A., Dimmick, W. W. & Carr, J. L. 1996. Conservation genetics of the Common snapping turtle (Chelydra serpentina). Conservation Biology 10, 397-405.

Sites, J. W., Jr., & Crandall, K. A. 1997. Testing species boundaries in biodiversity studies. Conservation Biology 11, 1289-1297.

Walker, D., Moler, P. E., Buhlmann, K. A. & Avise, J. C. 1998. Phylogeographic uniformity in mitochondrial DNA of the snapping turtle (Chelydra serpentina). Animal Conservation 1, 55–60.


Anonymous Anonymous said...

I am turtle lover to, so I was glad to read something extra on turtles. However I have small comment - the site that you are citing is not russian, but bulgarian (clearly visible from the domain.

4:58 PM  
Anonymous Anonymous said...

I once calculated the weight of Stupendemys for an article about giant reptiles from the post-dinosaur-era, using the proportions of Podocnemis expansa as comparison. This species is said to reach a weight of 50-70kg at a shell-length of 1m. The shell of Stupendemys was about 2,30m, so I came to a weight between 600 and 850kg, still very impressive, but not several tonnes. I made also calculations about the weight of Geochelone atlas and came to the result that the often published weight of 4000kg is far away from reality.

7:44 PM  
Blogger Darren Naish said...

Many thanks Sordes for your comment. On the size of Stupendemys, a while back I attended a talk given by Jasmin Winkler and colleagues about new material of this taxon (go here: SVPCA 2005). They concluded that a new specimen indicates a mass of just over 3 tons. I don't know if this is more accurate than your 600-850 kg as I agree that, even accounting for the fact that volume increases as a cube of length, 3 + tons does seem high for a turtle of this size. Hopefully there will be some sound explanation in the technical study, due to be published next year....

Scheyer T. M. & Sánchez-Villagra M. R. In press. Carapace bone histology in the giant pleurodiran turtle Stupendemys geographicus: phylogenetic and functional aspects. Acta Palaeontologica Polonica.

Geochelone atlas (by the way, surely it isn't really a species of Geochelone) has been estimated by Greg Paul to be about 1 ton in weight - his weight estimates have generally held up well in other studies so this is perhaps about right (the more 'typical' weight of over 4 tons is apparently way way too high).

BTW Sordes... who are you really?

10:38 PM  
Anonymous Anonymous said...

Hallo Darren!
3tons would be really very much I think. Using a more conservative weight of 50kg for a Podocnemis with a 1m long shell(and I think this is okay, because the very flat shell of Stupendemys looks to me very similar), a hyopthetical 3000kg Stupendemys would have a shell-length of nearly 4m...The calculation is easy to follow, a 3000kg Stupendemys would have about 60 times the weight of a Podocnemis with a shell length of 1m. But as a result of the cube-increase of mass, this Stupendemys had to have 4 times of this shell-length (4 X 4 X 4 = 64), the complete turtle with tail and neck would be about 6m or so.
I came also to a very similar weight for Geochelone/Testudo atlas, 1 ton seems very probable, using modern giant turtles as comparison. I find it really strange that there are comparably often very strange states of weight for extinct animals (and also living animals, but this is another topic), even if there are very good comparisons with similar modern animals.
By the way, what to you think about the idea that the giant Purussaurus (about its size and weight are also many disagreements) had probably a special "taste" for large turtles? Its jaws looks a bit like cutting pliers, with very massive teeth in the enormous strong front, and comparably small distal teeth. With such a jaw it was probably even able to crush from the border the flat shells of giant turtles like Stupendemys, even if they were much bigger than its head. This huge carnivores prabably ate nearly anything(even the largest ground sloths would have had no chance, but giant amphibic rodents and tapirs could be more common), but its strange head could lead to the idea that there was a distinct specialization. From all modern crocodylians, the skull of the small broad-snouted caiman Caiman latirostris has the strongest similarities to those of Purussaurus, and they are said to have a big deal of crabs and hard-shelled molluscs in their diet when they are juvenile, and many turtles when they are adult. Invertebrates seems no very probable source of food for a monster like Purussaurus with a minimum weight of 8000kg, and even IF it had eaten such things, it had not needed such massive jaws, so a distinct specialization for large and giant turtles could make sense, especially because there would be no competitioners.

I hope you don´t think I´m arrogant, but what do you think...who could I be?



2:24 PM  
Blogger Darren Naish said...

Sordes, you're not Beri Krzic are you? Probably not.

On the subject of Purussaurus, did you know that a third species - P. mirandai - has just been named by Aguilera et al.?

10:20 PM  
Anonymous Anonymous said...

Hallo Darren!

No, I´m not Beri Krzic, I´m just a guy who is really interested in living and extinct animals (especially the odd ones), with a special interest for recently extinct megafauna like mekosuchine crocodiles. I´m also very interested in Cryptozoology and since many years very active in several forums and in contact with several experts about this topic, and it was nice too see that you have some similar interests. I also worked for the (no longer existing) german cryptozoology magazine "Pterodactylus" where I did draw some pictures for an article about cryptid big cats, and also for the descendent of this magazine, "Der Kryptozoologie-Report",where I made several drawings for an article about the tatzelwurm, including too models of this cryptid (which you can see here:,com_zoom/Itemid,159/catid,5/ and the magazine here: ).
I´m also working on several own articles for this magazine, including those about post-dinosaur-era reptilian megafauna and another one about the last (now extinct) reptilian "monsters" which encountered humans. As I like both, sculpting and strange animals, I have also made several models of more uncommon animals like Palaeopropithecus or Mesonychoteuthis.
But about Purussaurus, I knew only that there was a smaller one named Caiman neivensis (or nivensis?) which is sometimes said to be identical with Purussaurus, but I found the abstract about P. mirandi as you mentioned it.



11:00 AM  
Anonymous Anonymous said...

I personally think 8 tons for purussaurus is on the excessive side. I mean at 40 feet( about the length of T.rex) and probably with a proportionately longer tail than T.rex how could it weigh 2 tons more, considering T.rexs' huge legs as opposed to Purussaurus relatively puny ones?
Trevor Madin

12:54 AM  
Anonymous Anonymous said...

The weights of extinct crocodiles are comparably easy to estimate, especially if they have no unusual long skulls. Within a wide range of lengths, crocodiles have very similar proportions, to one can calculate the weight of for a given length, by using known length-weight-relations. In the case of Purussaurus you can use an authenticated mass of about 1000kg for a 6m crocodile as reference. As weight increases with the cube, a 12m long crocodile would easily weigh 8000kg. In fact the crocodile used as reference was a hybrid named Yai (on of the largest crocodiles on earth)which has a snout which is proportionally a bit longer than those of Purussaurs. So even at the same length, Purussaurus would weigh a bit more. Furthermore crocodiles and alligators have a strong tendency to become bulkier with increasing size, so a huge brute like Purussaurus would be very probably even again a bit heavier.

9:50 PM  

Post a Comment

<< Home