Tuesday, May 23, 2006

Dicynodonts that didn’t die: late-surviving non-mammalian synapsids I

If you’ve read my previous posts you’ll know that - rightly or wrongly - I’ve invested an unreasonable amount of time in arguing against what’s been termed the ‘prehistoric survivor paradigm’: the notion, endorsed and propounded by some cryptozoologists, that numerous tetrapod groups known only from the fossil record might have survived to the present, yet without leaving an intervening fossil record. But the history of zoological discovery shows us that there’s nothing intrinsically wrong with the idea of either long gaps in the fossil record, or of the discovery of Lazarus taxa (that is, organisms that represent late survivors of supposedly long-extinct groups), and I’ve discussed some cases of this in previous posts (see New, obscure, and nearly extinct rodents of South America, and…. when fossils come alive).

We now know that quite a few fossil tetrapod groups really did hang on for much longer than was conventionally thought: in these cases, both long gaps in the fossil record, and the unexpected appearance of Lazarus taxa, occur. Some such cases have been big news in the palaeontological world, but mostly these instances have been largely ignored or unreported given that the animals concerned were obscure, deemed mundane, or of interest only to specialists. I personally think all the relevant cases are really interesting, and in future posts I aim to review them. Here we’re going to look at some of the more obscure and unreported of these cases – namely, the late survival of non-mammalian synapsids. If you don’t know what these are, the following paragraph is for you. If you do, skip ahead.

Mammals are the only surviving members of a far larger and more diverse tetrapod clade, Synapsida. Mammals evolved in the Triassic*, but they are but one of numerous synapsid clades, few of which made it past the mass extinctions at the end of the Triassic, and most of which evolved and diversified in the Carboniferous and Permian. Basal synapsids (things like caseids, ophiacodontids and sphenacodontids) looked superficially reptile-like, and for this reason non-mammalian synapsids have traditionally been dubbed ‘mammal-like reptiles’, and even classified within Reptilia. This obscures their affinity with mammals, and it’s nowadays agreed that Synapsida and Reptilia are different clades: basal synapsids are better regarded as ‘stem mammals’, and they aren’t reptiles. While some non-mammalian synapsids were bizarrely unique and quite different from mammals, during the Permian and Triassic a number of very mammal-like non-mammalian synapsids evolved. Some of these animals evolved small body size, and probably endothermy and body fur. So were you to travel back to the Triassic and catch an assortment of small synapsids, you might be hard pressed to work out which were the mammals.

* There are multiple different views on how the term Mammalia should be defined (Rowe & Gauthier 1992). Here (as in a previous blog on the soft tissue morphology of ears) I’ve decided to follow what seems to be the consensus view, and include all the Triassic mammal-like cynodonts (including Adelobasileus, morganucodontids and so on) within Mammalia.

In terms of taxonomic diversity, one of the most successful groups of non-mammalian synapsids were the dicynodonts, a group that first evolved in the Late Permian and then thrived in the Early and Middle Triassic. Dicynodonts were tubby-bodied, relatively short-legged synapsids, amusingly described by Mike Benton as possessing an ‘unsatisfactory tail’. Basal forms had multiple teeth, but mostly they had a strongly reduced dentition, with only large tusk-like upper canines projecting from their short, beaked jaws. Dicynodonts were probably predominantly herbivorous, but at least some may have chewed at carcasses and eaten small animals.

Sad to say, during the Late Triassic, dicynodonts dwindled in diversity until by the Norian (the penultimate stage of the Late Triassic) they were down to just three genera, and all of these were close relatives within the clade Kannemeyeriiformes (King 1990, Maisch 2001). I always liked Richard Cowen’s suggestion that these last forms were ecologically peripheralised, endangered species that hung on to existence in remote ecosystems where life was harsh. In fact Cowen compared them to Giant pandas Ailuropoda melanoleuca and Mountain gorillas Gorilla beringei if I remember correctly (Martill & Naish 2000 also covered this idea), but as for whether it’s an accurate portrayal or not I don’t know. But regardless, dwindling in numbers, and living in a world where big archosaurs were now controlling all the terrestrial ecosystems, those poor last dicynodonts gradually faded into oblivion, until they were but dust in the wind, dude. That was a Bill and Ted reference.

In June 1915 several fragmentary fossil bones were discovered near Hughenden in Queensland (Australia). Heber A. Longman (best known for his 1924 description of the giant pliosaur Kronosaurus) exhibited them at a meeting in 1915, and noted that they resembled dicynodont elements. Well, it turns out that he was right, as a 2003 reappraisal of the specimens by Tony Thulborn and Susan Turner showed that the bones could not belong to anything other than a dicynodont. One of the most telling of the specimens is a partial maxilla that still houses its slightly recurved canine tusk. In every detail – the distribution of concavities and foramina, the articulatory surfaces for other bones, the tooth shape, wear pattern and surface microstructure, the internal tooth structure (determined by CT scanning) – the specimen is indisputably dicynodont, and not matched by anything else.

But here’s the big deal: the fossils are from the late Early Cretaceous, and thus something like 100 million years younger than the previously known youngest members of the group. Thulborn & Turner (2003) noted that this ‘is so extraordinary than it demands exceptionally rigorous investigation’ (p. 987), and they carefully showed why and how other groups could be excluded from consideration. Furthermore, the Cretaceous age of the specimen is well established and there is no reason to doubt it. So dicynodonts didn’t disappear in the Late Triassic as we’d always thought. They had in fact been sneakily surviving somewhere, and as Thulborn & Turner (2003) wrote, their persistence in Australia and absence from everywhere else suggests that ‘Australia’s tetrapod fauna may have been as distinctive and anachronistic in the Mesozoic as it is at the present day’ (p. 991). That's pretty incredible.

Loads more to come on the subject, but it’ll have to come in a subsequent post. Watch this space.

The image above combines Laurie Beirne’s dicynodont life restoration, used in the press releases for Thulborn & Turner (2003), and on the front cover of the relevant issue of Proceedings of the Royal Society of London B, with a photo of the Australian fossil. Both images are widely available on the web. Sorry about the low resolution - I'll post a better picture in the near future. For the latest news on Tetrapod Zoology do go here.

Refs - -

King, G. 1990. The Dicynodonts: A Study in Palaeobiology. Chapman & Hall (London, New York).

Maisch, M. W. 2001. Observations on Karoo and Gondwana vertebrates. Part 2: A new skull-reconstruction of Stahleckeria potens von Huene, 1935 (Dicynodontia, Middle Triassic) and reconsideration of kannemeyeriiform phylogeny. Neues Jahrbuch fur Geologie und Palaontologie, Abhandlungen 220, 127-152.

Martill, D. M. & Naish, D. 2000. Walking With Dinosaurs: The Evidence. BBC Worldwide (London).

Rowe, T. & Gauthier, J. 1992. Ancestry, paleontology, and definition of the name Mammalia. Systematic Biology 41, 372-378.

Thulborn, T. & Turner, S. 2003. The last dicynodont: an Australian Cretaceous relict. Proceedings of the Royal Society of London B 270, 985-993.


Blogger Dave Strickland said...

This is probably a stupid question, but I'll ask anyway. Why couldn't the Cretaceous Hughenden fossils be actually older fossils that were later incorporated into Cretaceous-era sediments?

10:14 PM  
Anonymous Anonymous said...

Cripes, 100 million years younger... that's like finding late Mesozoic mammals, multituberculates or something, in Recent sediments.

So, the idea is that these guys spent the entire Jurassic in some enclave, doing the Mesozoic equivalent of eating bamboo shoots? Is there any skeletal evidence to suggest this? I mean, like, funky 'Panda's Thumb' type specializations, kind of thing? Or are the pieces just too fragmented?

Doug M.

9:04 AM  
Blogger Darren Naish said...

This is a very sensible and logical question, certainly not a stupid one. Determining a Cretaceous origin for the specimen was prety important, and Thulborn & Turner devoted pretty much a whole page to this issue. Summarised, they state; (1) the locality (= Alderley Station) is a Cretaceous-only site and the fossils found with the dicynodont [ichthyosaur fragments] are typical of Cretaceous fossils from the site; (2) there are no pre-Cretaceous rocks anywhere local, with the nearest Permian and Triassic rocks being at least 80 km away. For fossils to hypothetically get from these outcrops to Alderley Station, they'd have to travel downstream for c. 100 km, then upstream (!). Note also that the Alderley Station dicynodont fossils lack signs of water transport; (3) the matrix adhering to the dicynodont fossils is typical of the Cretaceous Allaru Formation, and this is almost certainly the source of the fossils. To date, none of the many other fossils from this unit have been reworked from older strata.

So Thulborn & Turner concluded (2003, p. 989) 'we can find no reason to doubt that QM F15.990 [the dicynodont specimen] originated from the Rolling Downs Group. It seems quite certainly to be Early Cretaceous in age and was probably introduced into the marine environment by flood-waters that also carried the carcasses of sauropods, ankylosaurs and ornithopod dinosaurs'.

I agree with them entirely.

11:38 AM  
Blogger Darren Naish said...

Response to Doug M: thanks for the comment. Yes it seems that dicynodonts hid away in some remote, poorly sampled 'safe haven' environment. Would they have had unusual morphological specialisations? Well, unfortunately we just don't know. The Allaru Formation fossils aren't complete enough to test this idea, and in fact the bits we have are not all that different from Triassic kannemeyeriiforms. We might imagine that they truly were 'living fossils' of their day.

9:28 PM  
Anonymous Anonymous said...

this is not weird at all considering that other forms such as Koolasuchus persisted until the Cretaceous in Australia.

10:40 PM  
Blogger Darren Naish said...

Response to previous anonymous comment...

It is true that Cretaceous Australia now seems to have acted as a 'safe haven' for a number of tetrapod groups that had long become extinct elsewhere: the discovery of the chigutisaurid temnospondyl Koolasuchus extended the duration of this group by about 65 million years, while - arguably - the plesiosaur Umoonasaurus might indicate the presence of a rhomaleosaurid ghost lineage about 75 million years long. But given that the Allaru Formation dicynodont is about 110 million years younger than its closest fossil relatives, I cannot agree with your implication that it is unworthy of special attention. I never said it was 'weird', rather that it is 'pretty incredible', and I still find that an appropriate description.

10:37 PM  

Post a Comment

Links to this post:

Create a Link

<< Home