Regular readers will know that I have a major interest in ichthyosaurs, the so-called fish-lizards of the Mesozoic (see links below). As you’ll know if you keep your finger on the pulse of Mesozoic reptile news, last week saw the publication of a really interesting new animal from the Lower Triassic: the Chinese ‘proto-ichthyosaur’ Cartorhynchus lenticarpus, described in Nature by ichthyosaur expert Ryosuke Motani and a team of colleagues (Motani et al. 2014). We’ve hoped for decades that the fossil record might reveal animals that somehow bridge the morphological gulf between ichthyosaurs proper and more terrestrial reptiles. Cartorhynchus does this to a degree, but it also presents us with several surprises.

Here's what seems to be the 'official' art that accompanied the announcement of Cartorhynchus. Will all due respect to the artist concerned (Stefano Broccoli), I don't exactly think it's very inspiring and I'd really love it if there were some superior images out there. Hint hint hint.

For one thing, the geological age of Cartorhynchus is in keeping with the increasing amount of data which shows that Mesozoic marine reptile evolution was underway virtually from the very start of the Triassic. Cartorhynchus is from the Spathian in the Lower Triassic (Motani et al. 2014), and hence lived only a couple of million years after the end-Permian extinction event.

In several respects, Cartorhynchus conforms to what we’d expect for a ‘proto-ichthyosaur’. It’s small (estimated total length is 40 cm or so), has orbits that are (proportionally) about normal with respect to terrestrial reptiles (that is, it does not have the gigantic orbits of ichthyosaurs proper), and its paddles appear to have been capable of the sort of flexion you’d expect for an animal that could use its limbs on land, in seal-like fashion. Motani et al. (2014) opt not to include this new animal within Ichthyosauria proper* since it lacks some key features of this group (including an especially high vertebral count, an elongate snout and proportionally huge orbits). Instead, they coin Ichthyosauriformes for the Cartorhynchus + ichthyosaur clade.

Holotype (and only known specimen) of cute little Cartorhynchus, from Motani et al. (2014). Photo by Ryosuke Motani.

* While I will refer to Cartorhynchus as an ‘ichthyosaur relative’ throughout this article, Motani et al. (2014) follow the convention of using the less familiar term Ichthyopterygia for the group conventionally called Ichthyosauria. I will stick with ‘ichthyosaur’ here since I don’t quite agree with the idea that we need the term Ichthyopterygia at all (in any case, its precise membership – that is, which taxa are non-ichthyosaurian ichthopterygians and which are not – keeps changing according to the relationships inferred for Triassic taxa).

Several key features (some shown here on this cartoon aquatic reptile) allow us to recognise suction-feeders from the fossil record. You don't have to have a shallow skull though. Illustration by Darren Naish.

In some aspects of anatomy, Cartorhynchus is surprising. It has a short, deep skull with a comically short snout. These features are combined with an enlarged, robust hyobranchial and a (probable) lack of teeth. Together, these features suggest that Cartorhynchus was a suction-feeder. I certainly wouldn’t have predicted that an early ichthyosaur relative would be so short-snouted and with apparent suction-feeding specialisations. Based on early ichthyosaurs (like Chaohusaurus, Utatsusaurus and so on), and on inferred close relatives of ichthyosaurs (like hupehsuchians, on which read on), we might predict that ancestral ichthyosaurs were long-jawed, narrow-nosed predators of fast-swimming prey.

But... remember: nobody is calling Cartorhynchus ‘an ancestor’. Rather, we should imagine it as belonging to a hitherto unknown branch that shared an ancestor with other ichthyosauriforms. Of course, this means that there are more early ichthyosauriforms (including early ichthyosaurs) awaiting discovery. We can also infer that the short, deep snout of Cartorhynchus indicates rapid specialisation for a novel aquatic lifestyle. But then, we already know that Triassic reptiles were experimenting with many remarkable and innovative anatomical designs.

Hupehsuchians and thalattosaurs as close kin of ichthyosaurs and why this is important

Redrawn skeleton (after Carroll & Dong 1991) and speculative life reconstruction of Hupehsuchus. Illustrations by Darren Naish.

Another neat thing about this paper is that it reinforces the previously hypothesised relationship between ichthyosaurs and hupehsuchians (Motani 1999, Chen et al. 2014, Motani et al. 2014). Hupehsuchians – long-jawed, metre-long, amphibious reptiles – were originally mooted (back in the early 1990s: Carroll & Dong (1991)) as mystery reptiles that couldn’t be allied with any other group. Hence, so the story goes, they were an illustration of how little we know about reptile evolution, and about how pervasive and problematic convergent evolution might be. One notorious naysayer even used Hupehsuchus to illustrate the contention that attempts to fit species onto the tree of life are doomed to failure when we’re relying on anatomical characters alone (Feduccia 1996).

However, the fact that recent studies have found detailed anatomical characters that link hupehsuchians with ichthyosaurs – Motani et al. (2014) list a set of bony features that support a hupesuchian-ichthyosauriform clade – illustrates how proper and careful analysis of evidence can, in fact, allow us to link groups with others if only we try hard enough. Hupehsuchians and ichthyosauriforms both share flanges on the humerus and radius, an especially wide ulna, proportionally long forelimbs and proportionally long hands, especially short transverse processes on the vertebrae, as well as a few other characters (Motani et al. 2014). The new name Ichthyosauromorpha is proposed for this hupehsuchian + ichthyosauriform clade.

Excellent life reconstruction of the thalattosaur Thalattosaurus (with an ichthyosaur and assorted fish at lower left) by Ken Kirkland; from Richard Hilton's 2003 Dinosaurs and Other Mesozoic Reptiles of California.

Motani et al. (2014) also found Wumengosaurus and thalattosaurs (also known by the rather pointless – but increasingly popular – alternative moniker thalattosauriforms) to be close to Ichthyosauromorpha (a hypothesised relationship between these groups was previously supported by Chen et al. (2014)). Wumengosaurus is a slim-snouted, amphibious Triassic reptile that was originally published as a sauropterygian (Jiang et al. 2008).

The idea that thalattosaurs are close to ichthyosaurs is not novel: it has previously been mooted by Müller (2003) on the basis of snout shape and the relative size of the orbit relative to the supratemporal fenestra. The idea of a thalattosaur–ichthyosauromorph clade is interesting because thalattosaurs are – like hupehsuchians – amphibious, paddle-tailed reptiles that look something like ‘ichthyosaur prototypes’.

If hupehsuchians, Wumengosaurus and thalattosaurs really are close kin of ichthyosauriforms, we might conclude that ichthyosauriforms emerged within a clade where their ancestors were already highly aquatic. In other words: on the basis of this phylogenetic hypothesis – and assuming that the amphibious or aquatic habits we see in these groups were present in the relevant common ancestors – we predict the oldest ichthyosauriforms to be amphibious, paddled animals. Cartorhynchus, of course, agrees with this. If this true, it seems that we aren’t ever going to discover semi-terrestrial ‘walking proto-ichthyosaurs’ (as we have for whales, seals and seacows) because ichthyosauriforms emerged within a group that was already strongly aquatic, and where key adaptations for aquatic life (including viviparity and retracted bony nostrils) were already in place before ichthyosaurs or ichthyosauriforms evolved.

Phylogenetic hypothesis for ichthyosauromorphs and related lineages recovered by Motani et al. (2014). Note that - according to this tree - ichthyosauriforms are deeply nested within a clade where the several other lineages (and thus the hypothetical common ancestors) were amphibious or aquatic. Image by Darren Naish.

Having said that, I feel duty bound to mention the possibility that the amphibious and aquatic habits we seen in hupehsuchians, Wumengosaurus, thalattosaurs and ichthyosauriforms might still have evolved independently. This possibility certainly isn’t parsimonious in the sense of how many evolutionary transformations it involves, but... well, evolution is complicated and we know of other groups of organisms where there seems to be a tendency for the same features or lifestyles to evolve again and again. There’s a name for this tendency but I can’t remember it.

The spectre of convergence, and is there a marine reptile ‘super-clade’?

Go Team Triassic Marine Reptile! A montage of Triassic ichthyosauromorphs, sauropterygians and other groups. From the in-progress Big Book of Vertebrate History that's slowly killing me. Illustration by Darren Naish.

Given that hupehsuchians, ichthyosauriforms and so on are all amphibious or aquatic, you might be wondering if the purported affinity between them is the result of convergent evolution. Because this is a concern, some effort has been made to deliberately avoid the inclusion of characters regarded as aquatic adaptations (Chen et al. 2014, Motani et al. 2014). And yet these groups are still recovered as close relatives. When aquatic adaptations are included, however, saurosphargids (yet another group of amphibious Triassic reptiles) and sauropterygians group with the Wumengosaurus + ichthyosauromorph clade to form a marine ‘super-clade’. While I can understand that people might want to exclude characters that look like those that can easily evolve by way of convergence (example: paddle-like limbs, a deep and laterally compressed tail), we shouldn’t work on the assumption that characters can be excluded for these sorts of reasons – the characters concerned could still, after all, be evidence of affinity. We can’t work on the assumption that we know at the outset which characters do, and do not, evidence affinity.

Mesozoic marine reptile 'super-clade' recovered by Motani et al. (2014) when they included aquatic adaptations in the data set. Image by Darren Naish.

Ultimately, time and more data will tell whether there is additional support linking saurosphargids and sauropterygians with the thalattosaur + ichthyosauromorph clade. And some of you will know that mesosaurs (Permian amniotes conventionally regarded as non-diapsids, and hence as well away from ichthyosaurs: they were covered here in July 2013) have been suggested by some authors as possible ichthyosauromorph relatives too (Maisch 2010). This is all exciting stuff because it means that many ‘problem’ lineages – reptile groups that have been floating around without a distinct place in the phylogeny – are finally being allocated precise positions in the reptile tree of life. And the idea that there’s a gigantic ‘mostly marine’ radiation is a very cool idea (albeit not a novel one - that's an issue for another time...).

Life reconstruction of Cartorhyncus, by Darren Naish. We need more art depicting this creature.

There’s more I want to say about this new paper (hey: I like the shape of that parvipelvian phylogeny they recover... but why no Malawania?), but already this article is over-long. And let the crowd-sourcing of new Cartorhynchus images begin! Send me your art and I'll publish it here. ONE MORE THING: you can now support me on Patreon, and get to see sneak-peeks of works-in-progress. Thanks much to those who have made pledges so far!

For previous Tet Zoo articles on ichthyosaurs, see…

Refs - -

Carroll, R. L. & Dong, Z.-M. 1991. Hupehsuchus, an enigmatic aquatic reptile from the Triassic of China, and the problem of establishing relationships. Philosophical Transactions of the Royal Society of London B 331, 131-153.

Chen, X.-h., Motani, R., Long, C., Jiang, D.-Y. & Rieppel, O. 2014. The enigmatic marine reptile Nanchangosaurus from the Lower Triassic of Hubei, China and the phylogenetic affinities of Hupehsuchia. PLoS ONE 9 (7): e102361

Feduccia, A. 1996. The Origin and Evolution of Birds. Yale University Press, New Haven & London.

Jiang, D.-Y., Rieppel, O., Motani, R., Hao, W.-C., Sun, Y.-L., Schmitz, L. & Sun, Z.-Y. 2008. A new Middle Triassic eosauropterygian (Reptilia, Sauropterygia) from southwestern China. Journal of Vertebrate Paleontology 28, 1055-1062.

Maisch, M. W. 2010. Phylogeny, systematics, and origin of the Ichthyosauria – the state of the art. Palaeodiversity 3, 151-214.

Motani, R. 1999. Phylogeny of the Ichthyopterygia. Journal of Vertebrate Paleontology 19, 473-496.

- ., Jiang, D.-y., Chen, G.-B., Tintori, A., Rieppel, O., Ji, C. & Huang, J.-D. 2014. A basal ichthyosauriform with a short snout from the Lower Triassic of China. Nature doi:10.1038/nature13866

Müller, J. 2003. Early loss and multiple return of the lower temporal arcade in diapsid reptiles. Naturwissenschaften 90, 473-476.