This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
Regular readers of Tet Zoo might know that much of my specialised technical research centers around the dinosaurs, pterosaurs and other fossil reptiles of the Wealden Supergroup. What is the Wealden Supergroup? It’s the name given to a series of mudstones, siltstones and sandstones, deposited across the floodplains, lagoons and estuaries of south-eastern England during the Early Cretaceous (between about 145 and 120 million years ago).
Regular readers might also know that last year saw the publication of the monumental English Wealden Fossils, a 769-page book that reviews and describes all Wealden fossils reported as of 2011 (Batten 2011). I have two chapters in the book: one on theropod dinosaurs (Naish 2011), and one on crocodyliforms, co-authored with the University of Queensland’s Steve Salisbury (Salisbury & Naish 2011). Given that the rich and taxonomically diverse record of Wealden crocodyliforms hasn’t ever been comprehensively reviewed (such a contrast to the Wealden dinosaurs), our work is certainly of broad interest to those who like Mesozoic crocodyliforms.
Unfortunately, the inclusion of this review in a multi-authored book means that it isn’t as widely available as it should be, and I’m sure interested parties want to hear more about the conclusions we reached, and about the new taxa we named. Partly for these reasons, I want to talk about our chapter here. Coincidentally, now that I’ve (more or less) finished all those crocodyliform drawings, I’m also able to more easily illustrate the different crocodyliform clades when I write about them. Which crocodyliform drawings? These ones...
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The bit you need to skip if you’ve heard it before
Before I start, here’s the usual brief explanation of nomenclature… Archosauria (the crocodile-bird clade) includes an enormously diverse assemblage that I’ll refer to here as the ‘croc-group archosaurs’. The names Crurotarsi and Pseudosuchia have most often been associated with this lineage; the group includes the armoured, pig-snouted aetosaurs, the often (but not always) predatory rauisuchians, and the crocodiles, alligators and gharials and all their kin.
Within croc-group archosaurs, all archosaurs traditionally imagined as ‘crocodilians’ are properly termed crocodylomorphs, with the more crocodile-like of the crocodylomorphs being grouped together in Crocodyliformes (the only non-crocodyliform crocodylomorphs are the gracile, long-limbed sphenosuchians. They may or may not be monophyletic). As you’ll know if you looked at that illustration I produced recently (or, as you might know already), Crocodyliformes encompasses a monstrous diversity, with terrestrial, semi-terrestrial, and fully marine lineages dominating certain chunks of the cladogram.
The most confusing issue surrounding the nomenclature of croc-group archosaurs is that the crocodylomorph crown-group – that is, the group subtended by the living species (representing Gavialoidea, Alligatoroidea and Crocodyloidea) – is termed Crocodylia by most workers, a name obviously similar to Crocodilia, the name ‘traditionally’ used for Crocodylomorpha and/or Crocodyliformes. Annoyingly, some people (predominantly here in Europe) are now arguing that we should use Crocodylia in place of Crocodyliformes (and should give up on using Crocodylia for the crown-group). Unsurprisingly, you therefore often have to explain what you mean when you use the terms you do.
To be clear, then, ‘crocodilians’ of tradition are here termed crocodylomorphs. Within Crocodylomorpha, those lineages more crown-ward than sphenosuchians are united within Crocodyliformes. Within Crocodyliformes, Crocodylia is used only for the crocodylomorph crown. Oh, and, the term ‘crocodile’ is understood here (and across Tet Zoo) to be applied specifically to true crocodiles (members of the crocodylian clade Crocodylidae): I don’t use it for any non-crocodylid members of Crocodylomorpha.
To the Goniopholididae!
When most people familiar with crocodyliforms hear the term ‘Wealden’, they often immediately think of the goniopholidids, a group (often incorrectly termed ‘goniopholids’) whose fossils are commonly associated with Wealden sediments. And if the name Goniopholididae is new to you and you have no idea how it might be pronounced, I suggest you imagine it as ‘gone-ee-oh-fol-ee-di-dee’ (though with the usual caveat that this only works if you’re pronouncing vowel sounds as we lazy native English speakers do).
Goniopholidids were long-snouted, relatively flat-skulled amphibious crocodyliforms, superficially similar in overall shape to living crocodiles. Unlike modern crocodyliforms (the set of lineages grouped together as Crocodylia), the dorsal osteoderm shield of these animals is composed of just two longitudinal rows of mediolaterally wide, rectangular osteoderms (one row on either side of the animal’s dorsal midline). Each osteoderm has a long, spike-like, forward-pointing peg on its lateral edge that locks into a groove on the underside of the adjacent osteoderm (Salisbury & Frey 2000). This is very different from the condition seen in the living crocodylians, where there are four or more longitudinal rows of transversely narrower osteoderms*, none of which have have the peg-and-groove system. Artistic reconstructions that essentially portray Nile crocodiles in the Jurassic or Early Cretaceous – and there are many out there – are thus very much in error. Goniopholidids did not look like this.
* The dorsal scutes in crocs and gators are squarish, whereas they are subrectangular in gharials and false gharials.
Goniopholidids differ from crocodylians in several other, important ways. The choanal openings on their palates are not completely enclosed within the pterygoid bones (the condition present in crocodylians and their closest relatives within the crocodyliform clade Eusuchia), their undersides are covered by an interlocking shield of polygonal osteoderms, and their vertebrae are amphicoelous (that is, with flattish articular faces), rather than procoelous (that is, with concave anterior articular faces and convex posterior articular faces) as they are in Crocodylia. Remember that feature about the vertebrae – we’ll come back to it later.
The first member of this group to be named – Goniopholis crassidens Owen, 1842 – is from the Berriasian part of the Purbeck Limestone Group of Dorset (UK), a unit similar in age to the very oldest parts of the Wealden Supergroup (the Berriasian is the very first stage in the Cretaceous). Richard Owen later named G. simus Owen, 1878 for a complete skull and an osteoderm from the Purbeck Limestone Group. A substantial amount of goniopholidid material from the Purbeck Limestone Group seems referable to G. simus (Salisbury 2002) and this species is also present in the Berriasian Obernkirchen Sandstone of north-western Germany (Salisbury et al. 1999). However, material from Bernissart in Belgium, often identified as G. simus (in particular the excellent skeleton IRSNB R47: see pic below), does not belong to this species and still requires proper identification.
A large number of goniopholidid fossils have been referred to the type genus (Goniopholis): in fact about 19 species have been named. These are mostly from western Europe, but North American, Asian and even South American taxa have been named too. As is so often with case with taxa referred to genera named early on in palaeontological history, some, most or many of these ‘Goniopholis species’ aren’t really close relatives of G. crassidens at all: furthermore, the exact status and content of Goniopholis itself has required clarification. Several recent papers address these issues (Salisbury et al. 1999, Andrande & Hornung 2011, Andrande et al. 2012) and Goniopholis is currently understood to be restricted to Upper Jurassic and Lower Cretaceous Europe.
Good material from the Wealden Supergroup – in particular an excellent partial skeleton described by Hooley (1907) – has long been thought to represent G. crassidens, meaning that most workers have regarded both this species, and the genus Goniopholis, to be present in the Wealden. Some of the Wealden material might belong to Goniopholis proper (read on), but it’s been clear for a while that others of the Wealden animals are very different from the definitive Goniopholis material present in the Purbeck Limestone Group. Foreshadowing what was to come, Steve wrote in 2002 that the Wealden material is “possibly even differentiable at a generic level from Goniopholis” (Salisbury 2002, p. 126).
‘The long-lost crocodilian of Mr Willett’: no longer so long-lost
One of the most famous Wealden crocodyliform specimens finally has its own name. This is the nice skull from the ‘Cuckfield Stone’ of the English mainland’s Grinstead Clay Formation (part of the Hastings Group of the Wealden Supergroup), long of unresolved status. Edgar W. Willett showed the specimen to the Geological Society of London in or around 1877 or 1878, and it was then described by Hulke (1878) as a specimen of G. crassidens, an identification arrived at on the basis of similar tooth form. The fossil has either been termed the ‘Willett specimen’ (Salisbury et al. 1999) or ‘Hulke’s specimen’ (Andrande et al. 2012).
For years, the thinking among the crocodyliform research community was that the specimen is ‘lost’, and indeed Steve gave a talk in 1998 in which he said exactly this (his talk was titled ‘The long-lost crocodilian of Mr Willett’). As was known to a few in the British museum community, however – and as I discovered for myself when I visited the place in 2002 – the specimen was actually safely accessioned at the Booth Museum in Brighton. The photo above shows it in its storage box, as of 2002.
Willett’s/Hulke’s goniopholidid is somewhat like G. simus in the general proportions and topology of its skull and even in the form of the external bone pitting. Our working hypothesis is thus that it’s a member of Goniopholis; however, it’s distinctive in having a proportionally long and narrow rostrum, an interorbital ridge restricted to the frontals, and a list of other cranial characters (Salisbury & Naish 2011). We opted to name it G. willetti. Numerous other goniopholidid specimens from the Grinstead Clay Formation – many of them described and figured by Mantell and Owen and referred by them to G. crassidens – almost certainly belong to G. willetti, but a pair of articulated Goniopholis premaxillae from this unit differ from those of G. willetti in shape and proportions and suggest that two goniopholidid pecies are present in this unit (Salisbury & Naish 2011). More data is needed before we can move on here.
Our book chapter appeared in print at about the same time as Marco Brandalise De Andrande et al.’s excellent study of goniopholidids (Andrande et al. 2012). They included G. willetti (as the ‘Hulke goniopholidid’) in a large cladistic analysis, and found it to group together with two goniopholidid specimens that – as of the time of their writing – also lacked technical names (and yet definitely or almost definitely represent distinct taxa): an Isle of Wight specimen described by Reginald Walter Hooley in 1907, and the Belgian specimen IRSNB R47 referred to in passing a moment ago. This clade (containing G. willetti, theHooley specimen and the Belgian specimen) is itself the sister-group to another that contains Goniopholis proper (G. baryglyphaeus from Portugal, G. simus, and the new G. kiplingi from the Purbeck Limestone Group) (Andrande et al. 2012). Since the Hooley specimen is regarded by us as different enough from Goniopholis to warrant its own name (read on), these results could mean that G. willetti deserves its own ‘generic’ name as well. For now, we’ll stick with the name Goniopholis willetti for this taxon, even though its inclusion in Goniopholis may require revision.
G. willetti is from the Hastings Group, a part of the Wealden Supergroup succession deposited on what’s now the English mainland between the Berriasian and Valanginian (that is, right at the very start of the Cretaceous: between about 145 and 133 million years ago). Remember that many familiar Wealden fossils come from the rather younger, Isle of Wight part of the Wealden Supergroup known as the Wealden Group: these sediments were deposited between the Berriasian and early Aptian, but the only exposed parts are Barremian and Aptian in age (between about 129 and 110 million years ago). I know that all of this is very confusing, so here’s what I hope is a helpful diagram…
The ‘Tie Pits Goniopholis’ is now Anteophthalmosuchus
Keeping in mind that I’m now talking of Wealden Group, rather than Hastings Group, goniopholidids, we start with what is perhaps the best known of the Isle of Wight goniopholidids. This is the ‘Hooley specimen’ I’ve just referred to above: it has also sometimes been called the ‘Tie Pits specimen’. It’s a partial skeleton (with an excellent skull) recovered from a cliff-fall at Atherfield Point in 1904 and 1905 by Walter White, coxswain of the Atherfield lifeboat (Hooley 1907). The type specimen is from high in the Vectis Formation and is thus from the early Aptian (see the diagram above), but referred specimens come from the older Wessex Formation (Salisbury & Naish 2011). Reginald Walter Hooley – a well known and important character in the history of British vertebrate palaeontology – was just 21 years old when he presented his very thoroughly illustrated technical description of the type specimen to the Geological Society of London in 1906 (Salisbury & Naish 2011).
Continuing with the tradition of lumping Wealden goniopholidids into Richard Owen’s original species, Hooley (1907) thought that the specimen could be referred to G. crassidens on the basis of various cranial similarities. However, the Tie Pits croc is quite different from G. crassidens in many detailed features. It lacks an external mandibular fenestra for one thing (this absence is also seen in G. baryglyphaeus and the Bernissart specimen), has a small, rectangular palpebral that doesn’t project over the orbit, and possesses unusually wide, sub-circular supratemporal fenestrae (Salisbury & Naish 2011).
Other diagnostic characters are present as well. The postorbital bones are peculiar in shape and mean that the eye sockets are directed upwards and forwards. Indeed, Hooley (1907) wrote of this skull that anything other “than a forward vision was precluded” (p. 55). We decided to chose this feature in the naming of this new taxon, and it’s now Anteophthalmosuchus hooleyi Salisbury & Naish, 2011. The name means something like ‘Hooley’s forward-pointing-eye crocodile’.
A. hooleyi was a large animal, with adults being more than 3.5 m long in total. Based on what we know about the ecology and behaviour of living crocodyliforms, we might infer that A. hooleyi was ecologically dominant to contemporary species, and that it aggressively excluded, or even predated upon, members of these other species (Salisbury & Naish 2011). Indeed, it seems likely that niche partitioning was occurring in the assemblages that contained the goniopholidid taxa discussed here, since contemporaneous crocodyliform species were much smaller, or with jaw and tooth shapes indicating very different dietary requirements and lifestyles: in the Wessex Formation, A. hooleyi occurs together with Vectisuchus leptognathus, Leiokarinosuchus brookensis and Hylaeochampsa vectiana (all of which we’ll be looking at in the next article).
Having said that, there may be other, Anteophthalmosuchus-like goniopholidids in the Wealden. In 1875, Thomas Huxley published his landmark paper on the use of the palate in understanding the evolution of crocodyliforms and their relatives (Huxley 1875). To illustrate the transition between ‘mesosuchians’ (a paraphyletic assemblage of Mesozoic crocodyliform lineages that included thalattosuchians and goniopholidids) and eusuchians (the modern crocodylians and their close relatives), Huxley used a Wessex Formation skull collected from somewhere on the Isle of Wight’s south-west coast. Today, the specimen is kept at the British Geological Survey (BGS) at Keyworth (where it’s accessioned as BGS GSM 119453), but it was formerly at the Geological Survey Museum in London.
This specimen is almost certainly a goniopholidid skull, and it’s very similar to the skull of A. hooleyi. However, it’s smaller, it’s more gracile, its palatine bones are different in shape, and it’s geologically a bit older as well, so it looks plausible that it represents another taxon, potentially a new species of Anteophthalmosuchus. Alternatively, it might be a juvenile specimen of A. hooleyi, or represent an intraspecific variant. We provisionally identify it as cf. Anteophthalmosuchus and note that additional work and/or new material is needed before we can say any more about the specimen’s identity.
My intention here was to say everything I wanted to say about Wealden crocodyliforms in a single article, but that hasn’t worked out so well. I’ll stop here, and deal with the remainder in the next article. If you like crocodyliforms, this might be good news. Until then…
For some previous Tet Zoo coverage of crocodylomorphs (wholly crocodyliforms, and mostly crocodylians), see...
The Saltwater crocodile, and all that it implies (crocodiles part III)
Crocodiles of New Guinea, crocodiles of the Philippines (crocodiles part IV)
The Freshie: Australian crocodile, seemingly from north (crocodiles part V)
Massive thanks to my co-author, Steve Salisbury, for his work in generating, and help in providing, the images used here.
Refs - -
Andrade, M. B., Edmonds, R., Benton, M. J. & Schouten, R. 2012. A new Berriasian species of Goniopholis (Mesoeucrocodylia, Neosuchia) from England, and a review of the genus. Zoological Journal of the Linnean Society 163, S66–S108.
- . & Hornung, J. J. 2011. A new look into the periorbital anatomy of Goniopholis (Mesoeucrocodylia, Neosuchia) and related forms. Journal of Vertebrate Paleontology 31, 352-368.
Batten, D. J. 2011. English Wealden Fossils. The Palaeontological Association, London.
Hooley, R. W. 1907. On the skull and greater portion of the skeleton of Goniopholis crassidens from the Wealden Shales of Atherfield (Isle of Wight). Quarterly Journal of the Geological Society, London 63, 50-63.
Hulke, J. W. 1878. Note on two skulls from the Wealden and Purbeck formations indicating a new subgroup of Crocodilia. Quarterly Journal of the Geological Society, London 34, 377-382.
Huxley, T. H. 1875. On Stagonolepis robertsoni, and on the evolution of the Crocodilia. Quarterly Journal of the Geological Society, London 31, 423-438.
Naish, D. 2011. Theropod dinosaurs. In Batten, D. J. (ed.) English Wealden Fossils. The Palaeontological Association (London), pp. 526-559.
Salisbury, S. W. 2002. Crocodilians from the Lower Cretaceous (Berriasian) Purbeck Limestone Group of Dorset, southern England. Special Papers in Palaeontology 68, 121-144.
- . & Frey, E. 2000. A biomechanical transformation model for the evolution of semi-spheroidal articulations between adjoining vertebral bodies in crocodilians. In Grigg, G. C., Seebacher, F. & Franklin, C. E. (eds) Crocodilian Biology and Evolution. Surry Beatty & Sons (Chipping Norton, Aus.), pp. 85-134.
- . & Naish, D. 2011. Crocodilians. In Batten, D. J. (eds). English Wealden Fossils. The Palaeontological Association (London). pp. 305-369.
- ., Willis, P. M. A., Peitz, S. & Sander, P. M. 1999. The crocodilian Goniopholis simus from the Lower Cretaceous of north-western Germany. Special Papers in Palaeontology 60, 121-148.