Taxon of the Week: Koumpiodontosuchus

Dear reader, don’t be put off by such a name, nor by the fact that poor little Koumpiodontosuchus has been dwarfed by Nanuqsaurus, the recently descovered ‘pygmy’ tyrannosaur (more on that when Richard finishes the latest What’s News), because this little Cretaceous critter raises some interesting questions about eusuchian phylogeny. It’s also going to be fairly short, as Richard and I are crazy busy at the moment.

Sometime in March 2011, a lady, out with her family, discovered part of the skull of Koumpiodontosuchus, and almost immediately handed it over to Dinosaur Isle, a museum on the Isle of Wight, UK (as can be seen in my dorky picture on the about page of TDS). In a rather coincidental turn of events, some months later the second part of the skull was donated to the same museum by other denizens of the Isle of Wight. The now completed skull was then meticulously studied by Dr. Steve Sweetman et al. of the University of Portsmouth. Cut to the present day, and what we have is a new species of bernissartiid crocodile (a group that includes some of the smallest neosuchian crocs, i.e. modern crocs and their immediate ancestors) from the Early Cretaceous. Koumpiodontosuchus adds to the already diverse ecosystem we see in the Early Cretaceous of the Isle of Wight, which includes an allosaur (Neovenator), Iguanodon, Polacanthus (a thyreophoran), Eotyrannus, good old Baryonyx and a brachiosaur (as well as some azdharchid pterosaurs and mammals).

A wonderful reconstruction of early Cretaceous life on the Isle of Wight. Courtesy of Mark Witton.

A wonderful reconstruction of early Cretaceous life on the Isle of Wight. Courtesy of Mark Witton.

Estimated at only 66cm in total length, Koumpiodontosuchus aprosdokiti (roughly meaning button toothed, unexpected) is a small croc with a big name. As the name suggests, Koumpiodontosuchus has ‘button’ teeth (broad and flat) situated at the back of the jaw, with pointier teeth towards the front. This dental arrangement meant that Koumpiodontosuchus could have a good crack at both catching fish at eating hard-shelled material such as molluscs. Despite its neat arrangement of teeth, its not Koumpiodontosuchus’ crowning glory. That prize belongs to the choanae.

Koumpiodontosuchus reconstruction, again by Mark Witton. Also, casual Neovenators in the back there.

Koumpiodontosuchus reconstruction, again by Mark Witton. Also, casual Neovenators in the back there.

In crocodiles, the choanae are found in the upper jaw and form the internal nostril openings (holes). Despite containing all the extant crocodiles and their recent common ancestors, Neosuchia has a subgroup called Eusuchia (“true” crocodiles”) in which all modern crocs are found. Now, there are many defining features that allow palaeontologists/taxonomists/biologists to distinguish eusuchians from the larger pool of neosuchians, but a big defining feature in recent years has been the placement of the choana(e) within the pterygoids, towards the back of the skull (if you’re getting a bit lost with crocodilian cranial anatomy, the Witmer/Holliday Lab 3D alligator project really does help a bunch). Bringing it back to the early Cretaceous of the Isle of Wight, Koumpiodontosuchus is a non-eusuchian neosuchian (it’s just a neosuchian, no big deal), so you’d expect it not to have its choana placed at the back of the skull, between the pterygoids. 

Holotype of Koumpiodontosuchus. Choana circled in red. Amended from Sweetman et al (2014).

Holotype of Koumpiodontosuchus. Choana circled in red. Amended from Sweetman et al (2014).

Well would you look at that, a choana at the back of the skull and between the pterygoids, now that sure is a turn out for the books. So what does this mean? Well it adds to the amassing evidence from other extinct crocs (e.g. the Madagascan Mahajangasuchus, a rather ugly looking brute from the late Cretaceous) that the placement of the choana(e) within the pterygoids, on its own, might not be a might sign from the taxonomic gods that the croc you’re looking at is a eusuchian.

TL;DR: Koumpiodontosuchus is a cool new (small and cute) croc from the early Cretaceous of the Isle of Wight who, despite its size, manages to further challenge the taxonomic rules that define key groups of crocodylomorphs. Pretty cool, even if a tad unpronounceable.


Taxon of the week: Euphanerops

Having been treated to a dinosaur last time in taxon of the week, this week I’ve decided to go for an obscure Palaeozoic fish.  This fish is pretty obscure even for a Palaeozoic fish; so obscure, in fact, that it doesn’t even have its own Wikipedia article.  I speak, as you’ll know if you bothered to follow that link, of the Devonian jawless fish Euphanerops.

Screen Shot 2014-03-06 at 09.50.15

Euphanerops in all its fossilised glory (Sansom et al, 2013)

Loathe as I am to introduce even more palaeontological nudity to The Dinosirs, Euphanerops is a so-called ‘naked anaspid’, a group that look similar, but not the same, as modern lampreys.  The naked anaspids consist of a handful of taxa that are united by the fact that they look a bit like members of a larger group, the anaspids, but don’t have their distinctive bony scales.  Both groups look like they lie somewhere on the gnathostome stem-lineage, which we met when we looked at heterostracans a few weeks ago.  As we saw then, these stem-gnathostome groups are extremely important in understanding vertebrate evolution as they provide our only morphological information on how characters like jaws and paired fins evolved in vertebrates. Euphanerops

A lamprey-like reconstruction of Euphanerops. Ten points if you can spot the mistake before getting to the end of the post. (From

The specific area onto which Euphanerops sheds light is the evolution of paired fins.  Jawed vertebrates today all possess paired fins, whether as traditional ‘fins’ or as limbs in tetrapods (although some, like snakes, have secondarily lost them).  Modern jawless vertebrates, the lampreys and the hagfish, have no paired fins, they have only single ‘median’ fins down their midline.  Unfortunately this gives us no information regarding how paired fins evolved, for all we know they may have just popped into existence.  One avenue of investigation in solving the puzzle is evolutionary developmental studies (evo-devo), where the development of modern animals is studied for clues.  This gives us intriguing hints, for example paired fins seem to develop in similar ways to both gill arches and median fins, suggesting a possible shared evolutionary origin.  They can also be induced to develop along a lateral line along the sides of the body.   However, if we want to know how the structure changed morphologically we have but one recourse.  To the fossil record!


A cartoon phylogeny of vertebrates with handy colour-coded fins. Green for dorsal fins, blue for anal fins and red for paired fins. Mixes arise from combinations (from Sansom et al, 2013).

Fossils clear up the picture quite a lot.  The earliest ‘paired fins’ we see are in groups like the thelodonts and the (clothed) anaspids, but these are all paired ‘fin-folds’, fin-like structures that lie paired along the body in vaguely the same place as pectoral fins, but which have no skeleton supporting them. Some heterostracans, such as Pteraspisalso have bony paired fin-like projections from their head-shields, but it looks like these are just convergent adaptations rather than actually being fins. Osteostracans, a group with a wide array of fancy head shields, have the first paired fins as we know them today, with a skeleton on the inside (like your limb bones).  They only had pectoral paired fins (ie. the front fins of a fish, or your arms), and the skeleton was made of cartilage, but they were paired fins nonetheless, although some groups of osteostracan secondarily lost them again.  Later, in jawed groups such as the placoderms, we finally see the full complement, of pectoral and pelvic paired fins.

anaspid fin folds wikipedia

Anaspids (of the clothed variety) with fin-folds coloured in green (from Wikipedia).

Combined with the molecular evidence this progression from no paired fins to paired fin-folds to paired fins with skeleton has been used to propose an evolutionary mechanism.  The story runs that lampreys and hagfish have a single stripe of ‘fin-competency’ down their midline, where fins can develop.  This stripe was somehow double and spread onto the sides of groups such as anaspids, allowing paired fin-folds to develop on the sides as well.  These paired ‘fin-competency’ zones then began to interact with the developmental tissue that bone is formed from, leading to skeletonisation of the fin in later groups, and eventually to the complex fin skeletons we see today.


This figure shows the proposed spread of a line of ‘fin-competency’ first from the midline onto the flanks to create ‘fin-flaps’ in various places, and then interacting with other developmental tissues to create the familiar skeletonised fins we see today at the bottom (from Yonei-Tamura et al, 2008)

This seems like a good explanation, and fits the evidence well.  Euphanerops makes the picture a bit more complicated however.  The genus has been known for a long time, and is also interesting in that some fossils appear to preserve a lamprey-like gill skeleton, which would imply that this is ancestral for gnathostomes.  Our interests lie however in the realm of paired fins, and in 2013, a fossil was described as having paired fins.  In itself this is not unusual, as the similar anaspids sometimes have paired fins in a position roughly where pectoral fins are in modern fish.  Euphanerops however has paired fins where the anal fin is in a modern fish, ie. between the anus and the tail.  This state of affairs is not known in any modern fish, or in any other extinct ones we know of.

Paired fin comparison

The paired anal fins of Euphanerops fossilised above and outline below (from Sansom et al, 2013)

This has interesting implications for the evolution of paired fins.  It seems that rather than just following a progression from no fins to paired fins, evolution had a period of ‘experimentation’ with paired fins in different places before settling on the paired fins we see today.  This fits in with our overall picture of evolution: while it can be tempting to think that it ‘progresses’ towards an aim, in fact it occurs piecemeal, leaving behind a mosaic of different ‘experiments’ with morphologies.  It’s also worth noting that lines of fin competency idea of fin evolution mentioned before isn’t overturned by Euphanerops and its paired fins, but our picture of paired fin evolution is broadened.

Screen Shot 2014-03-06 at 09.49.35

Euphanerops restored with paired anal fins! (Sansom et al 2013)

So, far from being merely an unclothed anaspid and a Twitter handle that rolls off the tongue, Euphanerops and its paired fins have much to tell us about both paired fin evolution and wider evolutionary principles.  Like Limusaurus last week it also illustrates how palaeontological and molecular methods can be used in a synthesis to better understand evolution.   Pretty impressive all round for a fish that not even Wikipedia loves*.

*But I love you Euphanerops!.



Sansom et al (2013) Unusual anal fin in a Devonian jawless vertebrate reveals complex origins of paired appendages.

Janvier et al (2006) Lamprey-like gills in a gnathostome-related Devonian jawless vertebrate.

Yonei-Tamura et al (2008) Competent stripes for diverse positions of limbs/fins in gnathostome embryos.

Johansen (2010) Evolution of paired fins and the lateral somatic frontier.


Taxon of the Week: Limusaurus

As a treat, this week’s Taxon of the Week is a dinosaur. Even better, it’s a dinosaur with very small hands. Despite it’s small hands, it’s managed to cement itself amidst a pretty sizeable debate, yes ladies and and gentle, this week’s TotW is Limusaurus inextricabilis.

The facts

Limusaurus was discovered in 2009 by Xing Xu, a Chinese palaeontologist whose seemingly always discovering a new species of dinosaur (over 30 valid species to date). At around 1.7m in length (roughly the size of a large-ish dog), Limusaurus was far from the biggest and most exciting looking new dinosaur of 2009, however, it was certainly a bit weird:

  • For starters it’s the first Asian ceratosaur ever described. Yes, you heard right a ceratosaur.
  • For your main course, it’s herbiverous.
  • To finish off with dessert, it’s got a beak.
Limusurus. Not your average ceratosaur.

Limusurus. Not your average ceratosaur.

True, whilst indeed not your average ceratosaur (or for that matter, theropod) such traits aren’t that weird in primarily carniverous clades. Within Theropoda, there’s many secondarily herbiverous taxa, incl. everyone’s favourite weirdo taxon, the therizinosaurs. Let’s also not forget the well known beaked herbiverous theropods, ornithomimosaurs. Oh, and did you know crocodylomorphs had a crack at beaks and herbivory (of course they did). Even though I keep harping on about its small hands, for frak’s sake, just take a look at alvarezsaurs (with even more pathetic arms than T. rex). So why exactly am I taking the time to blog about Limusaurus?

Mononykus trying, the meme that couldn't be due to it's preposterously pathetic forelimbs. Hastily drawn by me.

Mononykus trying, the meme that couldn’t be due to it’s preposterously pathetic forelimbs. Hastily drawn by me.

A debate as simple as 1, 2, 3

If you’re a fan of dinosaurs/palaeontology then you should be aware that birds evolved from theropod dinosaurs. News flash: this isn’t new. This notion really started to gain ground after seminal work by John Ostrom in the late 1960s/early 1970s (including that famous drawing of the ‘naked Deinonychus‘ as Richard likes to call it) who noted a lot of avian-like features in Deinonychus. On top of that, palaeontological records show multiple transitional forms between smaller theropods and birds (e.g. the ever eminent Archaeopteryx), and even some of the larger theropods show avian affinities (e.g. the cranial pneumatic sinuses found in Alioramus). Those are only a select few pieces of evidence that have made the evolutionary link between birds and theropods almost undeniable, there’s a so many more, it’s astounding (incl. inferred behaviour, such as avian sleeping positions in theropods: Xu & Norell 2004, and apparent egg brooding behaviour in Oviraptor: Norell et al. 1995). Watch the video if you need more peer reviewed proof that dinosaurs (even if they did look a bit like birds) were awesome. Try and show a little respect.

However, there’s always new discoveries that just love to get people arguing. The issue revolves around digit homology/identity. Essential, theropods (via evolution) ‘lost’ two fingers (digits) to arrive at the three-fingered hand seen in most theropods (aside from the aforementioned ‘weird taxa’, who lost more than 2). The same is true for birds. However, the debate revolves around which digits are lost, and which 3 form the fingers of the hand. In tetanurans (essentially the more advanced theropods, and by extension birds), it has long been thought that digits IV and V were lost, leaving a three-fingered hand consisting of digits I, II and III. Now, you’d expect birds, by being derived tetanurans to have this digital formula.  Well that would be nice wouldn’t it. Unfortunately the question of I,II,II or II,III,IV in modern birds has been debated for a very long time. New evidence in the mid-late 1990s from developmental and genetic studies showed us that the three digits of the avian hand actually developed from digits II, III and IV. Gasp, a spanner in the works!

This new evidence was then used (rather wrongly) to attempt to oppose the hypothesis that birds evolved from dinosaurs (Feduccia 2002). While I agree, the 90s developmental evidence from modern birds creates some novel evolutionary dynamics to investigate, it cannot be used to deny the whole host of other evidence that links dinosaurs to birds. Following on from the developmental studies, genetic studies in the early 2000s showed that during the ontogeny of some birds, the digit identity would change from the initial II,III,IV to I,II,III. This led to the occurrence of the ‘Frame Shift’ hypothesis, which suggests that certain genetic pathways associated with dinosaur/avian digit identity allow for ‘rapid’ changing of digit homology throughout dinosaurian/avian evolution (at it’s core, and that’s a very simplified summary). Not to be bogged down by detail (genetics isn’t my strong suit), developmental geneticists thought they’d cracked it, and that the change in digit identity over avian/dinosaurian evolution was likely to have been caused by these frame shifts.

Feduccia's (2002) argument in a nutshell.

Feduccia’s (2002) argument in a nutshell.

Enter Limusaurus. So, finally, we come to Limusaurus’ role in all of this. In 2009, Xu et al. looked at Limusaurus’ small hands and went “you know what, that’s a reduced digit I” (NOT ACTUAL QUOTE), which makes Limusaurus’ digit identity II, III, IV. As we discovered right at the start of this article (before the I made my lack of genetics knowledge crystal clear), Limusaurus is a ceratosaur, which isn’t a tetanuran, but a more primitive theropod, meaning that the II,III,IV digit identity may well be shared by most/all tetanurans, with Limusaurus representing an intermediate of sorts. Thus Xu et al. state that the digit identity of Limusaurus is more in favour of a slower, stepwise acquisition of the digit identity seen in advanced tetanurans, and eventually birds. But, as the famous saying goes, you know what small hands means…(small gloves?)

That’s right, a big controversy.

If you're confused (don't worry, so am I), this may help.

If you’re confused (don’t worry, so am I), this may help.

The small gloves are off

The Xu et al. (2009) caused a fairly serious debate, with authors such as Vargas et al. arguing that the digit condition seen in Limusaurus is derived, and based on developmental and genetic work that they (Vargas et al.) carried out, suggest that faster genetic shifts occurred in the evolution of birds. Xu et al. quickly responded (and quoted  Arthur Conan Doyle in a somewhat dramatic conclusion) and argued that the shift proposed by Vargas is not likely when the digit (and manus) morphology of fossil tetanurans is considered.

I’d sincerely like to end this post with a succinct conclusion, saying that the debate has, over the last few years been wrapped up. However, such large debates in palaeontology, due to the very nature of our field (i.e. everything we love is dead) are rarely fully resolved. This case is no exception. Researchers from Yale (Bever et al. 2011) and other top world universities have stated time and time again that the frame shift hypothesis is still viable in the context of avian evolution, and in a recent summary by Xu and Mackem, Xu is not so sure, saying neither hypothesis has evidence to topple the other. Not one for revelling in an unnecessarily depressing ending (*coughs* Firefly) I’ll leave you with this: yes, we can’t always find all the answers to big questions in palaeontology and evolution, but by creating a synergistic relationship between palaeontology and biology (genetics, evo devo etc.), future discoveries in both fields are sure to shed some light on even the biggest of debates.

If you’d like to read more on this subject, and weren’t put off by my murdering of the genetics side of things, then I’d highly recommend the aforementioned Xu & Mackem (2013) paper for a recent summary of the field (see references below, it’s in bold).


Xu, X. et al. (2009). A Jurassic ceratosaur from China helps clarify avian digit homologies. Nature 459, 940-944.

Xu, X. & Mackem, S. (2013). Tracing the Evolution of Avian Wing Digits. Current Biology 23, R538–R544 (and references therein).

Bever, G., S. et al. (2011). Finding the frame shift: digit loss, developmental variability, and the origin of the avian hand. EVOLUTION & DEVELOPMENT 13:3, 269–279.

Vargas, A.O., Wagner, G.P. & Gauthier, J.A. in Nature Proceedings (2009).

Vargas, A.O. & Wagner, G.P. Frame-shifts of digit identity in bird evolution and Cyclopamine-treated wings. Evolution & Development 11, 163-169 (2009).

Young, R. L. et al. (2011). Identity of the avian wing digits: problems resolved and unsolved. Dev Dyn. 2011 May;240(5):1042-53.

Taxon of the Week: Beelzebufo.

Yet again, Richard and I are falling behind on delivering on the blog front. For that, we apologise. The next 7 days (hopefully) should bring numerous posts to make up for lost time, including an opinion piece!

Anyway, on with the show. If you follow palaeontological news in any sense, then you’ll probably have heard about a new marine reptile called Atopodentatus. If you haven’t heard of Atopodentatus, imagine if you will the offspring of a basal sauropterygian, Predator, and Cthulu that potentially could have filter fed. Or, you could just check out Brian Switek’s well written Atopodentatus article (with pretty pictures). That’s right, this week we’re going against the grain, and focusing in on a big Cretaceous amphibian, that’s made for a big controversy. I assure you, it’s going to be ribbeting.

Big facts: “we’re gonna need a bigger pond”.

This week, our TotW is Beelzebufo ampinga. Before we knuckle down and attempt to remove the f(r)og from the controversy, let’s get down to the basics. B. ampinga was discovered in 2007 by a team from Stony Brook University from the Maevarano Fm, Madagascar, the remains dated to 70-65Ma (late Cretaceous). The initial paper had little to work with, as the holotype only consists of a few cranial elements, with a few vertebrae, a urostyle and a tibiofibula. However, even with such few remains, the initial size estimates for Beelzebufo are astonishing: with a length of over 40cm, and its head alone estimated at half that in width (20cm!). This immediately bestows Beelzebufo with the crown of the largest frog to ever have lived. Beelzebufo was initially placed within the Ceratophryinae (the common horned frogs), which is unusual and is the basis of most of the controversy surrounding B. ampinga (more on that later). Due to it’s phylogenetic position, Beelzebufo has hyperossification in the skull, stabilising connections between the upper jaw and the skull, a huge mouth, oh, and sharp teeth. What does this mean? Well, like all other ceratophyrines Beelzebufo probably was carniverous, and by being so large, many agree that it would have been an ambush predator of small vertebrates, such as small/baby dinosaurs. So a frog that eats dinosaurs, no wonder they called it Devil Frog.


The holotype of Beelzebufo. Bones that were discovered in white. Modified from Evans et al. (2008).

Earlier in 2014, Beelzebufo was back. This time a (open access!) paper was published, showing off many new specimens of the Devil Frog. These specimens (64 since 2007) were far more complete than the original published findings, and have shown us that it was far weirder than previously imagined. It also gave us a more complete picture of B. ampinga as an organism and as a species. By looking at the squamosal of different individuals it became apparent that there was a lot of intraspecific variation present in B. ampinga, with a size difference of up to 20% present between certain individuals. This has found to be caused by different individuals having different bone growth rates and patterns. Now, the simplest explanation is everyone’s favourite, sexual dimorphism (in modern ceratophryines, the females are larger than males in 90% of cases). However, studies on extant frogs have shown that bone growth patterns and maturation times in anurans can be dependent on other factors such as seasonal food and water availability, as well as temperature.



Another intriguing fact is that Madagascar at that time was seasonal arid, with dry periods being especially water-sparse. So how does a large amphibian like Beelzebufo cope? Again, material from the 2014 publication (Evans et al. 2014) helps us to possible answer this. Hyperossification (especially in the skull) is prevalent in Beelzebufo. This is apparent in the initial discovery also, but combine this with evidence from Evans et al. (2014), features such as the loss of a tympanic membrane, tall neural spines and cranial exostosis go some way towards confirming that Beelzebufo was a burrower. Thus enabling it to escape desiccation during extreme dry spells. Not only did Evans et al. (2014) provide us with a wealth of new information on an important fossil anuran, but it also came with some fantastic 3D skeletal morphological reconstructions.


The aforementioned pretty digital reconstruction. Also, those posterolateral flanges, phwoar. Modified from Evans et al. (2014).

Big controversy

As has been eluded to, Beelzebufo caused a big splash. Evans et al. (2008) placed B. ampinga within Ceratophryinae, a subfamily found only in South America, based on mainly cranial characters and a supporting phylogenetic analysis. Evans commented that the discovery of a late Cretaceous ceratophryine in this area is ‘unexpected’. Indeed, if you look at the palaeobiogeography it still looks squiffy, with the Madagascar-Seychelles-India tectonic plate losing contact from South America 120 million years-ago. However, even this is debated. Multiple lines of evidence now suggest some land-link was present between South America and Madagascar, including many molecular studies (ratite birds, iguanin lizards et al.) and physical similarities present between South American and Madagascan dinosaurs, crocodyliforms and mammal. So a ceratophryine from the Cretaceous of Madagascar isn’t so crazy, right?

Wrong. Two years after the initial Evans publication Ruane et al. (2010) carried out rigorous testing on the phylogenetic position of Beelzebufo. In Evans et al. (2008), it was apparently established that B. ampinga was a crown-group Ceratophryinae, and a sister taxon to the living Ceratophrys. Ruane et al. had a big problem with this reasoning, stating that the relationship between Beelzebufo and Ceratophrys is supported by 1 out of 81 characterstics, support values for the relationship between these two were also low in the phylogenetic study. Ruane et al. (2010) used molecular phylogenies (with data from extant anurans) in tandem with Beelzebufo, using it as a calibration point, to calculate the timings of the emergence of the MRCA (most recent common ancestor) of modern ceratophryines. With Beelzebufo used in this way, the emergence time of the ceratophryine MRCA was way before the times calculated by other studies, using well established datasets.Ruane et al. came to the conclusion that contrary to Evans’ initial hypothesis, Beelzebufo was a) not a sister taxon of Ceratophrys or b) definitely not a crown-group ceratophryine. However, Ruane et al. don’t really give us more than that, apart from saying it might be some sort of stem-group ceratophryine or a crown-group Hyloidea (a superfamily). If were throwing ballpark ideas around for the phylogenetic position of Beelzebufo, then I’m gunning for a stem-group Hynerian.


Dominar Rygel XVI, a fine example of Hynerian for all those nerds who don’t watch Farscape.

The Evanpire Strikes back

Not one for lying down Evans et al. (2014), now armed with many more specimens than last time around, resurrected the phylogenetic controversy surrounding B. ampinga. In the 2014 publication, the phylogenetic position of Beelzebufo was restored to the initial hypothesis, stating that Beelzebufo was indeed a crown-group ceratophryine, with this relationship holding true even when different tree-making steps and calibration points were used. Evans et al. also deal with many of the issues raised by Ruane et al. For example, Ruane et al. argue that the low support values between Beelzebufo and Ceratophrys indicate that they might not be sister taxon (and hence Beelzebufo isn’t part of the ceratophryine crown-group). Yet Evans et al. point out that support values are low even in studies that solely consider extant ceratophryids.

The fuzzy phylogenetic positioning given to Beelzebufo by Ruane et al. (maybe stem-group this, or crown-group that) is put under fire by Evans et al. I’ve already said that the Cretaceous MRCA emergence time is Ruane’s main argument against B. ampinga as a crown-group ceratophryine, however what I’ve not yet said (and what Evan’s et al. 2014 love to point out) is that Ruane et al. almost positively place another mid-late Cretaceous frog, Baurubatrachus as a ceratophryid (crown and/or stem), which still means that the MRCA is somewhere in the Cretaceous. Evans was quick (p.54) to point out that Ruane et al. at this point were being somewhat hypocritical in this regard. Finally, Evans points out that the hyperossification present in Beelzebufo is also present in living ceratophryids, another compelling line of evidence in support of the crown-group hypothesis. Despite being confident in their findings, Evans et al. still give a passing mention of the notion that hyperossification (and other ceratophryid characters) may be present due to convergent evolution, however, for the time being (and to conclude!) Beelzebufo appears to be (for the time being) a crown group ceratophryid. *Sighs*.

You've made it past the long-winded bit. Well done, have a pretty picture.

You’ve made it past the long-winded bit. Well done, have a pretty picture.

Big importance

Well done, you’ve made it through a very poorly written account of the s**t-storm which is the phylogenetic positioning of Beelzebufo. By this point, you’ve probably seen the words crown, stem, Beelzebufo, et al. and ceratophryine/d/inae enough to last multiple lifetimes, so then, why should we care about Beelzebufo and it’s position within the ‘tree of life’. You should care for two reasons: 1) the timing of the emergence of ceratophryids, 2) the importance of correctly using fossils in phylogenetic studies.

All glory to the importance of Beelzebufo. You will obey.

All glory to the importance of Beelzebufo. You will obey.

You should remember that the biogeography of Madagascar in the Cretaceous creates problems for the ceratophryid hypothesis. Ali et al. (2008, 2009 2011) have recently noted that land bridges between Madagascar and South America were severed by 115-112Ma. If this is true, (and presuming Beelzebufo and undiscovered others didn’t raft across the seaways, which is actually a large presumption, giving the thick skinned Beelzebufo would weather the salty waters well, for a frog) this pushes the emergence time of ceratophryids to before these dates. This is again contrary to studies that have found the emergence times of crown-group Hyloideans (a superfamily if you remember) to be around 88Ma. As has been stated many times of this blog, any fossil, even an incomplete specimen, if found in the certain places at certain times can cause palaeontologists/phylogeneticists/biologists to have to seriously reconsider the state of the field.

This leads nicely onto the second reason why Beelzebufo is important. Hypocritical arguments aside, Ruane et al., using Beelzebufo as an example, shows how any study using fossils as certain anchor points in phylogenetic studies MUST look closely at the phylogenetic position (and the evidence behind it) of the fossil taxa, and decide if this is appropriate. Any mistakes when involving fossil taxa in these studies affects conclusions with wide reaching implications, such as the divergence/emergence times of certain, and sometimes, very large clades.

And finally, Beelzebufo kinda looks like Hypno Toad. That’s pretty rad.


  • Evans, S. E. et al. (2008). A giant frog with South American affinities from the Late Cretaceous of Madagascar. Proc. Natl. Acad. Sci. USA 105: 2951–2956.
  • Evans S. E. et al. (2014). New Material of Beelzebufo, a Hyperossified Frog (Amphibia: Anura) from the Late Cretaceous of Madagascar. PLoS ONE 9(1): e87236. doi:10.1371/journal.pone.0087236
  • Ruane S. et al. (2011). Phylogenetic relationships of the Cretaceous frog Beelzebufo from Madagascar and the placement of fossil constraints based on temporal and phylogenetic evidence. J Evol Biol 24: 274–285.
  • Ali J. R., Aitchison J. C. (2008). Gondwana to Asia: plate tectonics, paleogeography and the biological connectivity of the Indian sub-continent from the Middle Jurassic through end Eocene (166–35 Ma). Earth-Science Reviews 88: 145–166.
  • Ali J. R., Aitchison J. C. (2009). Kerguelen Plateau and the Late Cretaceous southern-continent bioconnection hypothesis: tales from a topographical ocean. J Biogeogr 36: 1778–1784.
  • Ali J. R., Krause D. W. (2011). Late Cretaceous bioconnections between Indo-Madagascar and Antarctica: refutation of the Gunnerus Ridge causeway hypothesis. J Biogeogr 38: 1855–1872.

TOTW: Pteraspis

In the spirit of Ryan’s last TOTW, this week we will be looking at a member of the group of animals that my current work focuses on: Pteraspis, a heterostracan from the Early Devonian period.

The heterostracans were a group of armoured jawless fish, or ‘ostracoderm’, which lived in both saltwater and freshwater environments from the Ordovician to the Devonian period.  They were notable for their characteristic armour, with which they evolved a wide range of forms during their existence: some encased in box-like plates, others in smaller tesserae, some with flattened bodies, others with bizarre pointy nasal structures.  I’ll probably be subjecting you to a post on heterostracan morphological diversity later on, so don’t worry too much about taking notes.


Various types of heterostracan. Pteraspis is in the bottom right. (from Wikipedia)

Heterostracans are important to palaeobiology because, despite being jawless themselves, they lie somewhere on the ‘stem’ of the phylogenetic tree of the gnathostomes, or jawed vertebrates.  In the phylogeny below, the only two living clades are the gnathostomes (jawed fish, including us), which have jaws, paired fins and a bony skeleton, and the cyclostomes (hagfish and lampreys), which have neither jaws, paired fins nor bone.  This means that fossils are the only way we have of learning about how these characters evolved.  Heterostracans themselves are particularly interesting due to their bone, one of the first occurrences of this tissue amongst vertebrates.

Gnath phylo

A cladogram of vertebrate relationships. Note how all stem gnathostomes are jawless . Extinct groups marked with a cross. (from Purnell, 2002)

Enter our hero, Pteraspis.  This 20cm long animal is pretty famous as heterostracans go, and can usually be found swimming his way through any marine Devonian diorama (there’s a particularly good one in the National Museum of Scotland, Edinburgh).  The most striking thing about Pteraspis initially is its large bony head shield.  This is characteristic of all heterostracans, although the types of plate forming it and the shape vary dependent on the species.  Characters peculiar to Pteraspis and related genera are the long rostral (ie. nose) section and the large dorsal spine.  Also worth noting are the characters it lacked.  It had no paired fins, although it’s been suggested that the two wing-like protrusions coming out of its head shield served a similar function.  It also had no jaws or teeth.


A reconstruction of Pteraspis (from Wikipedia)

As the vast majority of living vertebrates have jaws and teeth, it might be quite hard to imagine how Pteraspis would have fed. The two groups of living jawless fish- hagfish and lampreys- have highly specialised mouthparts: the former for deep sea scavenging and the other as a ‘sucker’. Heterostracan mouthparts aren’t similar to either of these however, and they also have a more generalised swimming body plan; one thing they do have though is a series of oral plates.  These were parts of the head shield (not teeth!) that lay over the oral cavity.  Many functions have been suggested: scooping sediment, grabbing onto prey and suspension feeding.  Patterns of wear seem to exclude scooping and tiny little outwards pointing denticles on the plates would have prevented big prey from entering the mouth.  As such it seems that heterostracans like Pteraspis would have suspension fed on small animals, like many groups of animal today.

Het smile

A heterostracan similar to Pteraspis giving us an oral platy grin. (from Purnell, 2002)

Pteraspis and the heterostracans have a lot to tell us about how the jawed vertebrates evolved.  They possess bone, but no jaws or paired fins, immediately telling us that these characters of the gnathostome body plan didn’t evolve together.  Although we’ve not really discussed it here, the structure of their bone itself can also teach us about the evolution of bone, and how it came to prominence as a tissue. These animals, completely different to anything living in the oceans today, also present their own mysteries, such as how they fed without jaws.  We’ve only briefly talked about heterostracans here, but fear not! There’s plenty more to talk about in the future.


Taxon of the Week: Tiktaalik

Following all of the giant-rauisuchian based excitement of last week’s TOTW, I’ve decided to calm it down a bit by looking at an unassuming fish.  This particular unassuming fish has been in the news recently, and illustrates a transition without which there would be no Postosuchus, no mammals and (perhaps worst of all) no ‘Dinosirs’.  The fish of which I speak is, of course, the Devonian tetrapodomorph Tiktaalik roseae, but before we look at it closely, let’s have a look at the bigger evolutionary picture into which it fits.

Vertebrates that live on land are known as tetrapods, and all possess (or at least their ancestors possessed) four limbs with digits: amphibians, lizards, crocodiles, birds and mammals are all modern tetrapods. Despite their many obvious differences from what you’d typically think of as a ‘fish’, tetrapods are actually the largest group of lobe-finned fish, or sarcopterygians, named for their fins’ fleshy bases.  The only other modern groups of sarcopterygian are lungfish and the famous ‘living fossil’ coelacanth.

An African Lungfish: note fleshy-based lobe fins (from

An African Lungfish: note fleshy-based lobe fins, as compared to the rayed fins of your goldfish. (from

The most obvious difference between tetrapods and other fish is that the latter are aquatic and the former mainly terrestrial (with the exception of groups that went back into water, like whales and icthyosaurs).  At some point during the evolution of tetrapods they switched from an aquatic lifestyle to a terrestrial one, and this pretty drastic transition is of great interest to palaeontologists.   Originally pictured as being a case of fish struggling onto land and then evolving limbs, in recent(ish) years very early tetrapods from the Late Devonian (~365 Mya), such as Acanthostega and Ichthyostega, have changed this view.  These fossils have four limbs with digits, but appear to be fully aquatic, suggesting that the tetrapod body plan evolved in water first, only later proving handy (pun obviously intended) on land.

Acanth tol.web

Acanthostega having a swim with its digit-y limbs (from tol.web).

Various fish-like animals have been described that have body plans somewhere between these early aquatic tetrapods and the ancestral fishy form, and which are more closely related to tetrapods than the next closest group of sarcopterygians, the lungfish.  These animals are known as ‘tetrapodomorphs’, and show an evolutionary trajectory towards air breathing and a four-finned body plan, while still retaining fish-like characters such as gills and fin rays.  It has been suggested that this was as a result of them living in shallow rivers during the oxygen-poor Late Devonian, where being able to breathe air as well as oxygen dissolved in water would have been advantageous, as would being able to navigate shallow, relatively predator free, water.


A phylogeny of selected tetrapodomorphs and tetrapods, showing the change in body plan from ‘fish-like’ to ‘tetrapod-like’. (adapted from Daeschler et al, 2006)

Enter TiktaalikTiktaalik is something of a celebrity amongst tetrapodomorph fish. Described in 2006 from fossils found on Ellesmere Island in Nunavut, Canada, its name means ‘Burbot’ in Inuktitut.  It was subsequently popularised by one of its discoverers, Neil Shubin, in his book ‘Your Inner Fish’ (recommended), and is even the star (or possibly victim) of its own song on Youtube.  This stardom is well deserved, as Tiktaalik gives us a good deal of interesting information on the water-land transition.

A burbot reacts with indifference to the news that a tetrapodomorph has been named after it.

A burbot reacts with indifference to news of its namesake.

The original fossil material of Tiktaalik consisted of a pretty spectacularly articulated front half of the animal, possessing a number of interesting features.  The animal was pretty ‘fish-like’, but with some tetrapod-like characters, such as large shoulders bones and pectoral fins with wrist-like joints, a neck, and a robust ribcage.  These qualities, together with the shallow river environment in which it appears to have lived, led to the suggestion that it might have used its robust fore-fins to prop it up in ‘press-ups’, lifting its head (with the help of its neck) above water to breathe.  While there was no knowledge of the back half of Tiktaalik itself,  the small pelves (plural of pelvis) of other tetrapodomorph fossils like Panderichthys, led to a ‘front-wheel drive’ picture of tetrapodomorphs moving largely with the fore-fins, with  ‘four-wheel drive’ locomotion, with all four fins (or limbs), being a tetrapod innovation.

Tiktaalik wiki

Front-wheel drive Tiktaalik (from wikipedia)

Earlier this week however, the back half of Tiktaalik (or at least some of it) was described, and it transpires that this picture of exclusively ‘front wheel drive’ tetrapodomorphs is incorrect.  Tiktaaliks pelvis and parts of the hind limb were recovered, and it turns out that its hind-fins were as large as its fore-fins, with a fairly good range of movement.  While still compatible with their function as ‘props’ mentioned above, this also suggests that it would have been able to do more in terms of fin-based movement, with the possibility of underwater gaits, like the ‘walking’ seen in African lungfish.  It also changes our picture of the evolution of limbs, suggesting that tetrapodomorphs became ‘four-wheel drive’ before the evolution of tetrapods.

Tiktaalik comparison

The change in beefiness of hind-quarters/rear fins in Tiktaalik from the original description, above, to the recent paper, below (images from Daeschler at al 2006 and 2014 respectively)

While this recent information has changed our picture of Tiktaalik, it just adds to its importance as a source of information on the water land-transition.  This unassuming Devonian fish, along with various other fishy friends, helps illustrate that many of the changes that we associate with living on land, such as breathing air and having limbs, actually evolved as adaptations to aquatic life.  This transition also acts as a prime example of the mosaic nature of evolution: tetrapods didn’t evolve gradually from a fish-like form to a tetrapod-like form, but instead evolved tetrapod-like characters piecemeal while retaining ‘primitive’ ones. This mosaic theme is one that comes up time and time again in evolution, and is one that we’ll be discussing in future blog posts.

Some more Tiktaalik revelling in all their newly reconstructed glory.

Some more Tiktaalik revelling in all their newly reconstructed glory (image from

Further reading

Taxon of the Week: Postosuchus

In this week’s TotW, Ryan takes us through the posto child of the ‘rauisuchians’, Postosuchus.

When someone mentions the Mesozoic, you instantly think about dinosaurs. Admit it, it’s fine, there’s no judgement on this blog. You also will predominately think about dinosaurs from the Jurassic and the Cretaceous, with good old T. rex and co. (allosaurs, carcharodontosaurs, spinosaurs etc.) ruling the roost at the top of the food chain, whilst sauropod behemoths (amongst other ridiculously sized herbivores) wandering about in herds etc. etc. However, the Triassic (seemingly the Cretaceous and Jurassic’s ugly sister) is often forgotten about. Yes, we don’t have things which are outrageously large or ridiculously bipedal (or do we..?), but in the Triassic, crurotarsans (crocodile-line archosaurs) were having a bit of a field day.


The image that immediately springs to mind as soon as you mention ‘Jurassic’ or ‘Cretaceous’. Clearly.

Here on TDS, we think the Triassic (as well as plenty other eras, not just the Jurassic and Cretaceous) is pretty awesome too. The Triassic was a time of recovery, the Permian-Triassic mass extinction event had been and gone (and almost taken all of life on Earth with it). Dinosaurs were just starting out, and sitting on top of the food chain was, you guessed it, Postosuchus. If you look at the skull of Postosuchus kirkpatricki below, look carefully. Back in the 1980s famous palaeontologists thought Postosuchus (along with Poposaurus) could be a tyrannosaur ancestor. You can see where they’re coming from. Postosuchus was first discovered in 1922, and for 60 odd years after that, people didn’t really know what to make of it. First reports penned it as a Coelophysis, 20 years later, other finds were thought to belong to a new phytosaur. It wasn’t up until 1985 that the holotype, a well preserved skull and some postcranial remains, of Postosuchus kirkpatricki was formally announced. 


Totally not a dinosaur. No seriously.

Weighing in at almost 300kg, at reaching almost 4m when fully grown, Postosuchus was one (if not the) largest predator in the Triassic. With good long distance vision, a decent sense of smell, and a possible Jacobson’s organ, and oh, not to mention, over 7cm dagger-like teeth, this killing machine well may have taken down a fair few aetosaurs in it’s time (not a small feat). So fairly fearsome, but not as impressive as the theropods that were to come later in the Mesozoic, surely? Well, again, no.


These coelophysoids clearly came to the wrong neighbourhood.

What makes Postosuchus (and many other ‘rauisuchians‘) so interesting are its hindlimbs. One of the major dinosaurian innovations was the erect hindlimb posture, enabling more efficient locomotion. In the Triassic, descendants of crocodiles (who now have the ‘sprawling’ hindlimb posture) such as Postosuchus had a go at this hindlimb arrangement (evolutionary speaking). Whilst debated, many palaeontologists view Postosuchus (amongst other Triassic crurotarsans) as being bipeds (or at the very least facultative bipeds). So that means Postosuchus could use it’s forelimbs to kill things as well as it’s terrifyingly huge mouth (like bears do). To summarise, Postosuchus is a nightmare-inducing, killer croc-bear from back in time. It also raises the question (to be investigated by a future blog post on TDS, hopefully) of why exactly did dinosaurs survive through to the Jurassic, and rauisuchians go extinct, and why did crurotarsans go back to being solely quadrupedal?


Killer croc-bear from back in time. (Thank you internet).

Told you the Triassic was awesome. (Also, more to come on the locomotory strategies of Triassic crurotarsans to come, right after I finish my final 4th year exams…).

Bonus picture (because it’s cool and reasonably accurate, although not as accurate as the previous picture):


There’s no escape from the Post(o) Man (not actually a man).


  • Case, E. C. (1922). “New reptiles and Stegocephalians from the Upper Triassic of western Texas”. Carnegie Institution of Washington Publication 321: 1–84.
  • Case, E. C. (1932). “On the caudal region of Coelophysis sp. and on some new or little known forms from the Upper Triassic of western Texas”. University of Michigan Museum of Paleontology Contributions 4 (3): 81–91.
  • Case, E. C. (1943). “A new form of Phytosaur pelvis”. American Journal of Science 241 (3): 201–203. doi:10.2475/ajs.241.3.201.
  • Chatterjee, S. (1985). “Postosuchus, a new Thecodontian reptile from the Triassic of Texas and the origin of Tyrannosaurs”. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 309 (1139): 395–460. doi:10.1098/rstb.1985.0092.
  • Drymala, S. & Bader, K. (2012). Assessing predator-prey interactions through the identification of bite marks on an aetosaur (Pseudosuchia) osteoderm from the Upper Triassic (Norian) Chinle Formation in Petrified Forest National Park (Arizona, USA). Journal of Vertebrate Palaeontology, Program and Abstracts 2012, p89.

Taxon of the Week: Necrolestes

I’m afraid that this week’s taxon of the week isn’t quite as topical as last week’s, but it’s something that I heard about in a talk recently and thought would be interesting to write about.  The taxon in question is Necrolestes, a mysterious fossil mammal from South America.

The fossil material of Necrolestes consists of various fragmentary specimens from Patagonia dating to the early Miocene, about 16 million years ago.  From these remains, Necrolestes has been reconstructed as a small, digging, insectivorous animal, perhaps similar to modern golden moles in ecology.  While it’s obviously some kind of mammal it’s been assigned to various mammalian groups.

Golden mole

A modern golden mole. Actually more closely related to elephants than ‘true’ moles

Before digging too deep into where Necrolestes fits into the picture of mammalian evolution, it’s worth having a look at the picture itself (in fact, see the picture below).  Living mammals are divided into three groups: the placental mammals (most mammals you can think of, including you), the marsupials (kangaroos etc.) and the monotremes (echidnas and platypuses).  Of these the marsupials and placentals are more closely related to one another than either is to monotremes, and together marsupials and placentals are known as therians.  Monotremes, meanwhile, are found in a larger clade: australosphenidans.  Various extinct groups of mammals also existed, which don’t fit into these groups.  One such example is meridiolestids, a group of mammals that mainly died out in and shortly after the end-Cretaceous extinction event.


Modern mammal groups in bold. Placentals and Marsupials together form theria.

Over the years various phylogenetic analyses have been carried to work out where Necrolestes fits in this picture.  Some of these have suggested it’s a member of the theria: a marsupial or a placental mammal.  Others have put it on the stem of therians, and some specifically in the group mentioned before, the meridiolestids.  Various recent studies have suggested this grouping, and while the placement of Necrolestes within the meridiolestids is variable in these studies, it does seem fairly likely that this is where it fits.

There are a number of interesting implications if, as seems likely, Necrolestes is a meridiolestid.  One is that meridiolestids were around in South America for 46 million years longer than anyone thought.   This is interesting in itself, as it leaves them with a massive ‘ghost range’ through the Cenozoic.  Another, if you recall the small, fossorial ecology of Necrolestes, is that these meridiolestids were inhabiting pretty specialised niches.  Parallels could in fact be drawn with modern monotremes, which occupy very specialised niches in Australasia, and are the remnants of a larger, more widely distributed group.

Platypus bills and echidna quills: specialisation in a modern 'reli

Platypus bills and echidna quills: specialisation in remaining australsphenidans.

Necrolestes is thus an interesting example of how a small, fragmented fossil can have big and interesting implications.  It looks like this small, South American creature was a (now dead) ‘living fossil’, a remnant of an otherwise ancient mammalian group, and while not currently particularly newsworthy, an interesting taxon to look at this week.

(Dead) Taxon of The Week: Thecodontosaurus

Over here at TDS central (which is technically the MSc workroom in the Earth Sciences department, University of Bristol) we’ve come up with a way to give you guys a weekly introduction to all the lovely beasties palaeontologists around the world. To this end, Richard and I came up with #ToTW (Taxon of The Week), they may be a dead taxon (#dToTW) or even (if we’re feeling especially rebellious) a living taxon (#aToTW). Each week we’ll take it in turns to post a relatively short post on a different taxon, in an attempt to persuade you all that dinosaurs aren’t the be all and end all. However, in true TDS tradition, we’re going to completely ignore what we just said and talk about dinosaurs (let’s face it, dinosaurs bring all the hits to the yard).

This weeks ToTW is unfortunately a dToTW. But it is very dear to my (Ryan) heart. Back in my 2nd year of my undergraduate degree, this was the first actual dinosaur fossil I ever worked on. It was a huge day for me, I even took pictures of the bone I was working on and showed it to all my friends (they didn’t care). This dinosaur is Thecodontosaurus antiquus. The Bristol Dinosaur.


A vaguely accurate reconstruction of Theco (although no thought to be slightly more bipedal).

Discovered in Bristol all the way back in 1834, Thecodontosaurus (meaning ‘socket-tooth lizard’, eluding to the fact that the roots of the teeth were not fused with the jaw bone, like modern lizards) was the 5th dinosaur ever discovered. And even in 2013, 179 years later, Theco’s (a modern term of endearment, especially within the University of Bristol) still making news (more on that later). Standing at around 30 centimetres tall, and only around 1.2 metres long, a mighty tyrannosaur or colossal sauropod Theco is not. Yet Theco is with a very important dinosaurian group, the prosauropods. Prosauropods are the small Triassic (around 210 million years ago)  ancestors of sauropods, they allow us to investigate just how a group of dog-sized dinosaurs reached ridiculous sizes.

theco tooth

Socket tooth from the Socket-tooth ‘lizard’. Courtesy of the BDP and Andrew Cuff.

Much of the initial Thecodontosaurus findings were made in Bristol, and to this day research is carried out on Triassic fossils and Thecodontosaurus remains. The Bristol Dinosaur Project works mainly on microfossils, to painstakingly piece together the entire ecosystem that Theco may have lived in. The project is open to willing volunteers from both the scientific community, and the general public, it promises to reveal some much needed light on the Mid-Triassic of Bristol. Which by the way was quite a nice to live in (see below). More recently, all of the research on Theco has culminated in the Dinosaur Live Build, where Theco has been brought to live as a full scale (and as accurate has we can get the blighter) model, which got some nice news coverage (and was genuinely fantastic to see). Theco’s now housed above TDS Central (*coughs* Earth Sciences department, Wills Memorial Building). See below for my (edited and festive) picture of him/her (warning: it’s adorable). So here’s to Theco (and a shameless University of Bristol plug).


Bristol in the Triassic. We have to go back…

Since this is (possibly) the last post before Christmas, may Richard and I wish you a very Merry (gentlemanly) Christmas. 


Merry Christmas from Theco (and TDS!).