It goes without saying that most predatory animals need to open their mouths when they want to stab or bite potential prey items. But, get this, there’s a group of snakes that can erect their teeth and stab prey with a closed mouth. And that’s not all that’s interesting about these snakes. Yes, time for more weird snakes. There are lots and lots and lots of weird snakes, and one of my favourite groups of weird snakes are the atractaspidids (or atractaspids), and in particular the atractaspidid genus Atractaspis. If you haven’t heard of these snakes before it might give you some idea of what they’re like to know that they’ve been variously referred to as mole vipers, burrowing asps, burrowing adders, stiletto snakes or side-stabbing snakes. I’m going to be referring to them as burrowing asps: be prepare to be amazed…
Like the scolecophidians we looked at recently, burrowing asps are specialised for fossoriality (burrowing), with shiny-scaled, cylindrical bodies, small heads, a countersunk lower jaw, indistinct neck, short tail, and small eyes. While scolecophidians are definitely basal snakes, right down at the base of the crown-group snake clade, burrowing asps are not: their specialised fangs, venom apparatus and other characters show that they’re part of Colubroidea, the advanced snake clade that includes the viperids, elapids and the mostly non-venomous ‘colubrids’*.
* I put colubrids in quotes because there are widespread suspicions that this immense group (over 1500 species, over 300 genera, up to 28 ‘subfamilies’) is not monophyletic (e.g., Jackson 2003, Fry & Wüster 2004).
The venom apparatus of burrowing asps is formidable. Their venom glands are enormous, extending into the neck region and for a length approximating 20% of total body length in some species (Underwood & Kochva 1993, Wollberg et al. 1998). Occurring throughout much of Africa (with the exception of the north and south-west) as well as the Sinai and Arabian peninsulas (as far north as Israel), there are about 12 burrowing asp species, but their taxonomy is unsettled and requires sorting out. They’re between 30 and 50 cm long. Oh, and they have a distinctive aromatic smell (Branch 1988). No-one knows why.
How to be a ‘fang stabber’
Burrowing asps have a highly reduced dentition, with just two particularly elongate maxillary fangs (up to a third of total skull length), two short, gently curved dentary teeth, and a couple of very small palatine teeth [image above, from Deufel & Cundall 2006, shows viperid skull (top) compared with bizarre burrowing asp skull]. The maxillary fangs (there are two in each maxilla, one of which is a replacement tooth kept in reserve) are huge compared to the short, block-like maxilla: in fact virtually its entire length is occupied by the transversely arranged fang sockets. The maxilla articulates with the relatively immobile prefrontal by way of a saddle-shaped joint (this contrasts with the condition in viperids, where the articulatory surfaces between the maxilla and prefrontal are flat), allowing the maxilla to easily rotate posterodorsally and anteroventrally.
Because both the maxillae and the fangs are directed posteriorly when at rest, and because the prefrontal doesn’t move much relative to the braincase, the maxilla-prefrontal unit can’t be thrown forwards to project the fangs anteroventrally (as happens in viperids and elapids), but this doesn’t matter much, as we’ll see. The rotation of the maxilla is assisted by the musculature attached to the slender, rod-like pterygoid-ectopterygoid unit: the pterygoid is usually attached to the palatine in snakes, but in burrowing asps the two are widely separated with only a ligamentous connection (Underwood & Kochva 1993, Deufel & Cundall 2003a, b). This allows the pterygoid-ectopterygoid unit to swing antero-posteriorly without interference from the palatine. This has some implications, as we’ll see shortly [image below - from Deufel & Cundall (2006) - shows (A) viperid, (B) burrowing asp, and (C) elapid palatal bones. Note the slender, toothless pterygoid, and lack of bony connection between the pterygoid and palatine, in the burrowing asp. See Deufel & Cundall (2006) for full explanation].
As the maxilla rotates anteroventrally, it opens up a slit along the mouth-line, providing enough space for the fang to protrude out though the mouth. The projecting tooth is then stabbed into prey with a swift posteroventral (down and backwards) jerk of the head. The snake might erect either the left-side fang or the right-side one: they don’t seem to deploy fangs from both sides at the same time (even though they probably could). When grasped behind the head in what would normally be regarded as a safe handling posture, a burrowing asp can – without openings it mouth – erect one of its super-long fangs and stab the hand of the person holding it. Kurnik et al. (1998) reported a case in which a herpetologist – specifically, one of the authors of the paper – was bitten on the finger by an Ein-Geddi burrowing asp A. engaddensis. ‘Local effects, oedema, erythema and numbness appeared within minutes, followed by systemic effects, including general weakness, sweating, pallor, fluctuations in the level of consciousness, vomiting and watery non-bloody diarrhoea. Gross oedema of the hand developed and extended up to the forearm’ (Kurnik et al. 1998, p. 223). While the local effects healed within a few weeks, ‘some discoloration and tenderness remained even 10 months after’. Yikes. Don’t get bitten by a burrowing asp, that’s my advice.
And burrowing asp fangs aren’t just hollow cones, but (in all but two species) both canaliculate (housing a tubular canal) and keeled along the posterior edges of their tips. This keel cuts into tissue when the snake stabs, presumably increasing the size of the wound and hence aiding the absorption of venom (Golani & Kochva 1988). However, it’s also been suggested that the keel helps the snake to yank its fangs out of its prey: while most long-fanged snakes strike at prey from a distance and only briefly engage with the prey, burrowing asps get right up close to prey before stabbing (sometimes stabbing several times). For this reason Deufel & Cundall (2003b) recommended that a burrowing asp attack shouldn’t be referred to as a ‘strike’, but as a ‘fang stab’. When confronted with several prey items (a nest full of baby rodents for example), burrowing asps have been reported to stab and envenomate several individual prey animals before beginning to feed.
Specialised – but specialised for what?
Given that burrowing asps do their fang stabbing in burrows and other confined spaces, you would guess that all of these morphological and behavioural specialisations have evolved to allow attacks where relatively little room for manoeuvring is possible. Who, or what, is getting stabbed? Atractaspis species prey on nestling mammals (mostly murids and shrews), and Deufel & Cundall (2003b) proposed that a reliance on such prey shaped their evolution, suggesting that ‘the success of [Atractaspis and relatives] is partly attributable to the use of the envenomation apparatus on mammals’ (p. 58) [adjacent image shows Black or Ein-Geddi burrowing asp A. engaddensis].
However, Shine et al. (2006) argued that mammals make up less than 25% of the diet of Atractaspis and drew attention to data showing that elongate fossorial squamates were the most important prey items in the diets of these snakes. Attacking burrowing skinks and amphisbaenians within their burrows poses a problem, as the tails of these animals are similar in diameter to their bodies, making it difficult for the attacking snake to move past the tail and grab the body (you don’t want to grab the tail as both skinks and amphisbaenians are capable of autotomy). The bizarre specialisations of burrowing asps might therefore have evolved to allow these snakes to push past the tail and envenomate or seize the prey’s body. Incidentally, as a neat little aside, Shine et al. (2006) noted that the ability to autotomise the tail among fossorial squamates might be an important anti-atractaspidid adaptation, as a shed tail might both block the burrow to a pursuing burrowing asp, and prevent any venom injected into the tail from reaching the body. More study is needed to test this intriguing idea.
Being a good burrower makes a snake a poor swallower
The reduced palatal dentition and ligamentous linkage between the pterygoid and palatine in burrowing asps raises the question as to how these snakes transport prey within the mouth: as we saw in the scolecophidian article, most snakes employ ‘pterygoid walking’, engaging the maxillary and pterygoid teeth on the left side with the prey and dragging it toward the throat, disengaging, and then doing the same with the maxillary and pterygoid teeth on the right, and so on. Burrowing asps can’t do this: they sacrificed the ability to employ pterygoid walking when they de-coupled the palatine from the pterygoid and evolved a specialised toothless pterygoid whose only real function is to aid erection of the rotating maxilla and its fang. What, then, is a burrowing asp to do? [adjacent image of Bibron's burrowing asp A. bibronii from The TIGR Reptile Database].
Deufel & Cundall (2003b) looked specifically at this question. Firstly, burrowing asps sometimes used their super-long fangs as gaffs to manipulate prey into position for swallowing. Movements of the maxilla and/or pterygoid are indeed not used in transporting prey, but by shifting the lower jaw posteroventrally, bending the anterior trunk region from side to side, and compressing and extending the neck, these snakes are able to move the mouth over the prey. They aren’t very good at it though, and take a long time to successful ingest prey (the account here is very much simplified: for the full story see Deufel & Cundall 2003b and Cundall & Deufel 2006).
The question of how burrowing asps transport prey within the mouth illustrates the point that these snakes have had to make tradeoffs when faced with different evolutionary pressures. To be a good burrower, and to function properly as the specialised fang stabber it has become, Atractaspis has lost or modified some of the kinetic zones present in the skulls of other colubroids (the burrowing asp snout is relatively immobile relative to the braincase, for example, and the palatine does not move with the pterygoid-ectopterygoid unit, as we’ve seen). But these modifications mean that burrowing asps have had to find other solutions to the problems posed by feeding, and in fact their solution is convergently similar to that evolved by some other fossorial snakes (the pipe snakes, Cylindrophis).
A controversial radiation
Finally, what exactly are burrowing asps? For a start, their large venom glands, long, posteriorly inclined quadrates, vestigial or absent left lung, absent pelvis and many other details show that they’re part of the colubroid radiation (McDowell 1987, Lee & Scanlon 2002). Within this group, they were long regarded as viperids. However, Bourgeois (1961) proposed that burrowing asps might be particularly closely related to another poorly known group of colubroids, the aparallactines. Though the exact contents of this group remain controversial, it is generally stated to include 11 genera and about 50 species: all are small-headed African fossorial colubroids, often back-fanged, and with blunted or sharp-snouted heads. They include the centipede eaters Aparallactus, quill-snouted snakes Xenocalamus, harlequin snakes Homoroselaps and purple-glossed snakes Amblyodipsas [adjacent image, from wikipedia, shows Dull purple-glossed snake Amblyodipsas unicolor].
In a major genetic study of diverse colubroids, Kraus & Brown (1998) found an Atractaspis + aparallactine clade (which we might call Atractaspididae) to group together with the boodontines, an Afro-Madagascan group (although, historically, this group has – probably erroneously – also included Asian snakes). Both clades were in turn recovered as the sister-taxon to Elapidae (the colubroid group that includes mambas, cobras, and sea snakes). While both elapids and burrowing asps share tubular fangs and large, complex venom glands, the fact that these characters are not present in boodontines suggest that they evolved convergently. On the other hand, grooved fangs and at least some sort of venom delivery system were almost certainly primitive for colubroids (Jackson 2003). This article is now waaaay too long and I need to go do other things, but this is a subject we’ll be coming back to in the future. Given that snakes consist of over 2700 living species, Tet Zoo has still only scratched the tip of the iceberg, and there is so much more to write about.
Coming next: how a Tet Zoo article evolved into a peer-reviewed technical publication!
Refs – –
Bourgeois, M. 1961. Atractaspis – a misfit among the Viperidae? News Bulletin of the Zoological Society of South Africa 3, 29.
Branch, B. 1988. Field Guide to the Snakes and Other Reptiles of Southern Africa. New Holland, London.
CUNDALL, D., & DEUFEL, A. (2006). Influence of the venom delivery system on intraoral prey transport in snakesâ˜† Zoologischer Anzeiger – A Journal of Comparative Zoology, 245 (3-4), 193-210 DOI: 10.1016/j.jcz.2006.06.003
Deufel, A. & Cundall, D. 2003a. Prey transport in “palatine-erecting” elapid snakes. Journal of Morphology 258, 358-375.
– . & Cundall, D. 2003b. Feeding in Atractaspis (Serpentes: Atractaspididae): a study in conflicting functional constraints. Zoology 106, 43-61.
– . & Cundall D. 2006. Functional plasticity of the venom delivery system in snakes with a focus on the post-strike prey release behavior. Zoologischer Anzeiger 245, 249-267.
Fry, B. G., Wüster, W. 2004. Assembling an arsenal: origin and evolution of the snake venom proteome inferred from phylogenetic analysis of toxin sequences. Molecular Biology and Evolution 21, 870-883.
Golani, I. & Kochva, E. 1988. Striking and other offensive and defensive behavior patterns in Atractaspis engaddensis (Ophidia, Atractaspididae). Copeia 1988, 792-797.
Jackson, K. 2003. The evolution of venom-delivery systems in snakes. Zoological Journal of the Linnean Society 137, 337-354.
Kraus, F. & Brown, W. M. 1998. Phylogenetic relationships of colubroid snakes based on mitochondrial DNA sequences. Zoological Journal of the Linnean Society 122, 455-487.
Kurnik, D., Haviv, Y. & Kochva, E. 1999. A snake bite by the burrowing asp, Atractaspis engaddensis. Toxicon 37, 223-227.
Lee, M. S. Y. & Scanlon, J. D. 2002. Snake phylogeny based on osteology, soft anatomy and ecology. Biological Reviews 77, 333-401.
McDowell, S. B. 1987. Systematics. In Seigel, R. A., Collins, J. T. & Novak, S. S. (eds) Snakes: Ecology & Evolutionary Biology. Macmillan (New York), pp. 3-49.
Shine, R., Branch, W. R., Harlow, P. S., Webb, J. K. & Shine, T. 2006. Biology of burrowing asps (Atractaspididae) from southern Africa. Copeia 2006, 103-115.
Underwood, G. & Kovcha, E. 1993. On the affinities of the burrowing asps Atractaspis (Serpentes: Atractaspididae). Zoological Journal of the Linnean Society 107, 3-64.
Wollberg, M., Kochva, E. & Underwood, G. 1998. On the rictal glands of some atractaspid snakes. Herpetological Journal 8, 137-143.