‘A miniature plesiosaur without flippers’: surreal morphologies and surprising behaviours in sea snakes

Though I have no experience with them whatever (the shame), I have a great affection for sea snakes. They tie together several of my favourite areas, most of which will by now be familiar to regular blog readers: the discovery of new species and cryptic species diversity; radical convergence, evolutionary reversals, and a surprising phylogeny; unexpectedly complex behaviours; seriously bizarre morphology and novel trophic adaptations.

So when I had to give an introduction to the evolution and diversity of marine reptiles at an educational conference held at Vaughan College, Leicester, in 2004 I decided, for reasons that made sense at the time, to cover sea snakes as well (go here to see the conference abstracts). Because the conference was otherwise devoted to Mesozoic reptiles this confused at least some people in the audience (well, it confused Richard Forrest anyway) as they then thought that sea snakes had evolved in the Mesozoic.

Well, actually, there were ‘sea snakes’ in the Mesozoic – that is, marine members of Serpentes – but they weren’t close relatives of modern sea snakes, rather they were far more basal within the snake family tree (though how basal is the subject of contentious debate). Sea snakes in the modern sense are proteroglygous (=‘front-fanged’) caenophidians that belong to Elapidae, the widespread and successful snake clade (consisting of about 300 species) that includes cobras, coral snakes, mambas and the Australasian tiger snakes, taipans, brown snakes, whip snakes and so on.

Given that the oldest fossil elapids – cobras, coral snakes and other forms from Germany, Spain and the Czech Republic (Ivanov 2002) – are from the Early Miocene, it’s generally thought that sea snakes can’t be older than this. In fact a supposed fossil sea snake, represented only by vertebrae, has been reported from the Middle Miocene of the former USSR. Rage (1987) noted that this record was questionable given that sea snake vertebrae ‘are not easily distinguished from those of the Colubridae and other Elapidae’ (p. 66).

Conventional thinking has been that: if sea snakes descended from terrestrial elapids, we shouldn’t then expect them to have a history extending beyond the Early Miocene. But for various reasons that I’m not going to discuss here, some workers are now suggesting that modern sea snakes may, after all, have evolved in the Cretaceous (see Rasmussen 2002).

Indeed the way sea snakes are related to other elapids is interesting, to say the least. Though sea snakes all possess a vertically flat, paddle-like tail (a feature not seen in any other snakes), most snake experts have agreed that the term ‘sea snake’ includes two quite different groups of aquatic elapids: the laticaudids, or sea kraits, and the hydrophiids (or hydrophids or hydropheids), or true sea snakes. In fact even the paddle-like tail isn’t really alike in laticaudids and hydrophiids, as in the former the neural spines aren’t elongate as they are in hydrophiids (Rasmussen 1997). Several other features indicate that laticaudids are ‘primitive’ compared to the hydrophiids: they are generally better able to move on land, and they are mostly (but not entirely) oviparous (whereas hydrophiids are all viviparous).

While some herpetologists have regarded laticaudids as close relatives of sea snakes proper, McDowell (1969, 1987) argued that laticaudids were most closely related to Asiatic coral snakes (Calliophis and Maticora), American coral snakes (Micrurus and Micruroides) and Parapistocalamus (a poorly known snake of New Guinea and Bougainville Island: it’s apparently a specialist eater of snail eggs), and this has since been quite widely supported by other snake workers (e.g., McCarthy 1986, Keogh 1998, Rasmussen 2002). Sea snakes therefore represent two separate invasions of the marine environment. However, the news is that the picture is even more complicated than this, as it now seems that hydrophiids themselves are not monophyletic. And there seems to be another surprise in there too. We’ll get to all this in another post.

You might be surprised to hear how big sea snakes get, and how odd some of them are in shape. Heuvelmans (1968) noted that members of the genera Hydrophis and Microcephalophis [nowadays included within Hydrophis] ‘have a long thin head and neck, while the abdomen is four or five times as thick as the neck, so that they look like a miniature plesiosaur, but without its flippers’ (p. 38). He also noted that some Hydrophis species can be as much as 3 m long, and he cited William Dapier’s 18th century observation of an individual ‘as big as a man’s leg’ (which, if applying to circumferance, seems massive). I checked some of these details with Arne Rasmussen, a sea snake expert based at the School of Conservation, Copenhagen, who confirmed that a marked disparity between abdominal and neck circumference genuinely is present in some species. Is 3 m an accurate total length? Both Blue-banded sea snakes H. cyanocinctus and Yellow sea snakes H. spiralis have an authenticated maximum length of 2.75 m, so they’re not far off from this. Most of the 50-odd species are between 1 and 2 m long however. [The photo depicts a Turtle-headed sea snake Emydocephalus annulatus. I forget where the image comes from]

Not only do some species grow rather large, some species form unbelievably large breeding aggregations, with hundreds, thousands and apparently millions of individuals sometimes grouping together to form immense slicks that can be literally kilometers long. In The Trail That is Always New (1932), Willoughby Lowe described a sea snake slick encountered between Sumatra and the Malayan Peninsula. It ran parallel to the ship he was traveling in for a duration of about 60 miles, and for part of its length at least was 3 m wide. Lowe was probably justified therefore in stating that ‘Along this line there must have been millions’ (Bright 1989). To my knowledge, sightings of such super-aggregations remain anecdotal and they haven’t been photographed. If you know otherwise please let me know!

Sea snakes also include what is often said to be the most venomous of all snakes, H. belcheri. Its venom is said to be several times more potent than that of the Fierce snake Oxyuranus microlepidotes – the most venomous land snake – but it’s a very docile species that, even when handled roughly, rarely bites, and when it does bite it doesn’t inject much venom. Like sea snakes in general it has only small fangs and it’s not able to strike out of water. I couldn’t find any neat statistics on H. belcheri venom (such as ‘one drop of its venom could kill the entire population of an average Texan trailer park’), but for comparison an exceptional 110 mg venom yield from a Fierce snake was reportedly enough to kill 250,000 mice (Carwardine 1995).

Many sea snake species seem to use estuarine habitats as breeding grounds and/or nurseries, but they aren’t restricted to marine environments and may travel for tens of kilometers up rivers. Species in Cambodia, Thailand and the Philippines have been reported from lakes, and in fact two species are restricted to lakes: one of these is a sea krait that inhabits Rennell Island in the Solomons, and the other is H. semperi of Lake Taal on Luzon Island. What might be an additional freshwater species, H. sibauensis, was described by Rasmussen et al. (2001) for specimens collected 1000 km up-river in the River Sibau of Kalimantan, Indonesia. This is the furthest any sea snake has ever been recorded from the sea, but Rasmussen et al. (2001) were unable to determine whether the species was a true freshwater specialist as all known specimens were pregnant, and they might therefore have swam up-river in order to give birth. Even normally marine sea snakes species can switch between living in salt- or freshwater without problem, and have been kept in freshwater conditions without ill effects, so their physiology is pretty flexible.

Sea snakes are also neat in that they practice cutaneous respiration, being able to absorb up to 20% of their oxygen requiremens through their skin. That’s right: cutaneous respiration isn’t limited to lissamphibians among tetrapods. They can stay submerged for up to 3.5 hours and might only surface for 1 second (literally): in fact Yellow-bellied sea snakes Pelamis platurus, the only truly pelagic sea snake (and the one pictured at top: image taken from Susan Scott's site), spends an average of 87% of its entire time submerged. Sea snakes are entirely absent from the Atlantic, but the pelagic habits of Pelamis have long led people to think that it might be able to use the freshwater Panama Canal to undergo a so-called Lessepian migration to the Atlantic. This hasn’t happened however, nor have sea snakes used the Suez Canal to get into the Red Sea. Cool temperatures seem to have prevented Pelamis from extending its range around the Cape of Good Hope, as it seems to find sea temperatures of 11 degrees C fatal.

Special soft-tissue valves allow sea snakes to close their nostrils while they’re underwater and their lung is highly specialized. ‘Lung’, why not ‘lungs’? In snakes generally the right lung is enlarged relative to the left lung (or, rather, the left lung has reduced in size), but in sea snakes the right lung is modified for air storage. Its posterior part (the so-called saccular lung) has unique thick, muscular walls* that allows stored air to be forced forward into the functional anterior part (termed the bronchial lung) and the anterior projection of the lung (the so-called tracheal lung) is large and extends well forwards relative to that of terrestrial snakes. The presence of the reduced left lung is variable in sea snakes: most species lack it, but it’s present in some laticaudid individuals.

* Normally in snakes, the posterior lung walls are thin and sac-like, hence the name ‘saccular lung’.

Coming next: ecomorphological flexibility in sea kraits. Woo-hoo, as Steve Irwin would say. For the latest news on Tetrapod Zoology do go here.

Refs - -

Bright, M. 1989. There are Giants in the Sea. Robson Books Ltd (London).

Carwardine, M. 1995. The Guinness Book of Animal Records. Guinness Publishing (Enfield, Middlesex).

Heuvelmans, B. 1969. In the Wake of the Sea-Serpents. Hill and Wang (New York).

Ivanov, M. 2002. The oldest known Miocene snake fauna from Central Europe: Merkur-North locality, Czech Republic. Acta Palaeontologica Polonica 47, 513-534.

Keogh, J. S. 1998. Molecular phylogeny of elapid snakes and a consideration of their biogeographic history. Biological Journal of the Linnean Society 63, 177-203.

McCarthy, C. J. 1986. Relationships of the laticaudine sea snakes (Serpentes: Elapidae: Laticaudinae). Bulletin of the British Museum of Natural History (Zoology) 50, 127-161.

McDowell, S. B. 1969. Notes on the Australian sea-snake Ephalophis greyi M. Smith (Serpentes: Elapidae, Hydrophiinae) and the origin and classification of sea-snakes. Zoological Journal of the Linnean Society 48, 333-349.

- . 1987. Systematics. In Seigel, R. A., Collins, J. T. & Novak, S. S. (eds) Snakes: Ecology & Evolutionary Biology. Macmillan (New York), pp. 3-49.

Rage, J.-C. 1987. Fossil history. In Seigel, R. A., Collins, J. T. & Novak, S. S. (eds) Snakes: Ecology & Evolutionary Biology. Macmillan (New York), pp. 51-76.

Rasmussen, A. R. 1997. Systematics of sea snakes: a critical review. Symposium of the Zoological Society of London 70, 15-30.

- . 2002. Phylogenetic analysis of the “true” aquatic elapid snakes Hydrophiinae (sensu Smith et al., 1977) indicates two independent radiations into water. Steenstrupia 27, 47-63.

- ., Auliya, M. & Bohme, W. 2001. A new species of the sea snake genus Hydrophis (Serpentes: Elapidae) from the river in West Kalimantan (Indonesia, Borneo). Herpetology 57, 23-32.