Despite efforts, there just hasn’t been enough time (yet) to get Tet Zoo to properly reflect the balance of diversity within Tetrapoda (I blame the charismatic megafauna). And among the many groups that have been totally under-represented here are the snakes: one of the most speciose (over 2700 species) and abundant tetrapod clades. It’ll take a long time before snakes are fairly represented at Tet Zoo, and what I’d like to do here is talk briefly (relatively speaking) about a very obscure and poorly known group of really weird little burrowing snakes, the scolecophidians. The… what?
Three scolecophidian ‘families’ are currently recognised: the typhlopids, known variously as blind snakes or blindsnakes (they inhabit the American, African, Asian and Australasian tropics), the leptotyphlopids, also called threadsnakes, wormsnakes or slender blindsnakes (of tropical America, Africa, and south-west Asia), and the American anomalepidids (or anomalepids), sometimes called early blindsnakes. These are hugely successful snakes found throughout the tropics of the world and consisting of about 400 species. Yet they’re all but unknown outside the herpetological community, and are often mentioned only in passing, or are ignored entirely: as Samuel McDowell (1967) said ‘It is fairly obvious from most of the literature, both popular and technical, that the blind snakes are usually forgotten in defining and characterizing the Ophidia. The exhibition galleries of large museums sometimes omit them entirely, to leave more space for king cobras, puff-adders and rattlesnakes’ (p. 691). Let’s see if we can change that. These are really freakily bizarre little animals, displaying some incredibly neat anatomical details, and with the capacity to shed a lot of light on the key innovations that paved the way for the snake radiation.
Phylogenetic studies generally agree that scolecophidians are the sister-taxon to all other crown-group snakes, a clade called Alethinophidia (see Lee & Scanlon 2002 for review). Alethinophidia consists of the pipe snakes and kin, or anilioids [a group that may or may not be natural: see Gower et al. 2005], and the macrostomatans, which are all the others. Located as they are right down at the base of crown-group Serpentes, scolecophidians are obviously important in terms of hypotheses on early snake evolution. Is their small size, specialisation for fossoriality and insectivorous diet primitive for snakes, or are they degenerate forms, only deceptively ‘primitive’ due to the loss or modification of features present in earlier snakes? I won’t really be looking at those issues here but must do so at another time [image above shows Typhlops vermicularis].
All scolecophidians look superficially alike: they’re smooth-scaled, cylindrical, slender snakes (sometimes incredibly slender), with (usually) blunted heads, ventrally placed mouths, reduced eyes, and short tails that often end in a spike. The scales of typhlopids at least are thick and strongly overlapping, and in some species the scales glow under UV light (I don’t think anyone knows why); furthermore, the scales are so thick that shed skins are said to be rubbery in texture (Kley 2003). These thick scales defend the snakes from aggressive insects, but one North American leptotyphlopid has been shown to repel insects chemically: the snake smears itself in cloacal sac secretions that ants avoid (Gehlback et al. 1968). In leptotyphlopids, the pelvis is complete, and a small femur is present. All scolecophidians are small: the record holder reaches 100 cm (the typhlopid Rhinotyphlops schlegelii, or Schlegel’s blindsnake), but few exceed 60 cm, and most are less than 30 cm long. Scolecophidians never bite people but have been reported to emit faint squeaking sounds (Kley 2003).
Scolecophidian heads are very unusual. Though most species are blunt-snouted, some have sharply pointed snouts and many have unusual hooked snouts [montage at top of article shows a few Indian typhlopid species (from Khan 1999); adjacent image shows typhlopid head and tail]. Flattened, sloping snouts and even tri-lobed snouts are also present. Tactile organs are present on the head and are visible as small specks on the head scales and, most bizarrely, some typhlopid species (most notably the incredibly elusive Madagascan Xenotyphlops species) have flexible papillae growing from the scales at the snout tip. Nearly all scolecophidians appear to be dedicated arthropod predators, mostly eating the larvae and pupae of ants and termites. However, at least some species eat moderately large prey: Webb et al. (2000) documented predation on large earthworms by the Melanesian typhlopid Acutotyphlops subocularis. Despite the reliance of most species on very small prey (this applies even to the ‘giant’ Schlegel’s blindsnake), scolecophidians still indulge in what Webb et al. (2001) called binge-feeding: they rapidly ingest a large meal and do so infrequently. If fossoriality and insectivory are primitive for snakes, binge-feeding in scolecophidians might show that infrequent feeding in snakes did not evolve in concert with the ingestion of large prey (as has been thought by some). But if the fossoriality and insectivory of scolecophidians are highly derived specialisations relative to the primitive snake condition, binge-feeding might be retained from ancestors that ate larger-bodied prey.
Of the 395 or so scolecophidian species, about 265 are typhlopids. Until very recently, the many species in this group were classified in just three genera: Typhlops Oppell, 1811, Ramphotyphlops Fitzinger, 1843 and Rhinotyphlops Fitzinger, 1843. However, Typhlops has now been fragmented into multiple new or resurrected genera: Acutotyphlops Wallach, 1995, Austrotyphlops Wallach, 2006, Cyclotyphlops Bosch & Ineich, 1994, Grypotyphlops Peters, 1881, Letheobia Cope, 1869 and Xenotyphlops Wallach & Ineich, 1996 [Xenotyphlops shown in image below]. Ramphotyphlops, Acutotyphlops and Austrotyphlops differ from other scolecophidians (and other squamates) in that their hemipenes are solid: when folded away, these organs are coiled up like corkscrews. A pair of cloacal sacs – the retrocloacal sacs – are also present in these snakes, the function of which remains unknown (they were suggested to function in sperm storage but Shea (2001) showed that this was not the case). It has been suggested that the only Cyclotyphlops species, C. deharvengi from Indonesia, might share these details as it’s quite similar to Ramphotyphlops (Wallach et al. 2007), but this remains unknown because the only specimen we have is a female.
Prior to 2007, Acutotyphlops (containing four species) was thought endemic to eastern New Guinea, Alotau, the Bismarck Archipelago, Bougainvile and the Solomon Islands, but a fifth species, A. banaorum, then extended the range of the genus to the Philippines, and to the far northern Philippines at that. This gives Acutotyphlops a bizarrely disjunct distribution, as A. banaorum is separated from the others by about 4000 km (Wallach et al. 2007). A similar distribution is shared by the Platymantis frogs, and it might be that over-water dispersal has allowed snakes and frogs to move between these areas (recall that over-water dispersal is now thought to have occurred regularly among anurans, and that over-water dispersal in general has probably occurred far, far more regularly than was generally thought until recently). It is also possible, noted Wallach et al. (2007), that Acutotyphlops species previously occurred right across the intervening 4000 km, but that the species concerned have become extinct. On the other hand, they might still be around but just have yet to be discovered.
Typhlopid skulls are tiny – literally just a few millimetres long in cases – and for this reason it’s been very difficult to study them. If these snakes are specialists on ant and termite larvae and pupae, how do they procure, process and swallow their prey? Their skulls and jaws are extremely unusual: the elongate lower jaws are entirely toothless with tiny dentary bones (the dentary is usually the biggest bone in the tetrapod jaw) – this is unique among snakes – and the toothed maxillae are ordinarily kept folded away against the roof of the mouth, their tooth rows directed posteriorly and arranged transversely relative to the skull’s long axis. Ligaments and muscles, rather than bony contacts, hold the maxillae in this position [image below, from Kley (2001), shows the skull of Typhlops lineolatus].
Leptotyphlopids are just as strange if not more so [skull of Leptotyphlops dulcis shown below, with ventral view of lower jaw in protracted and retracted positions: from Kley 2001]. The upper jaws are entirely toothless, which is unique among snakes (properly prepared skulls of these snakes possess small, ventrally projecting pegs on the maxillae, but they’re not teeth nor do they seem to function as pseudo-teeth), and their maxillae are tightly bound to the rest of the skull and relatively immobile. In several Old World species the skull roof bones are strongly reduced and even absent. The lower jaws are only very loosely connected to the rest of the skull, and are supported by exceptionally long, slender quadrates that articulate with the braincase by way of long, sliding articulations. Each lower jaw is short and robust: in Leptotyphlops dulcis there are just four or five teeth in each dentary, and the unusual shape of the dentary – it is strongly convex on its lateral side – means that the teeth are arranged almost transversely, rather than in a line parallel to the skull’s long axis. A highly mobile intramandibular joint separates the conjoined dentary-splenial unit from the post-dentary bones, the rounded intramandibular facet allowing the anterior half of the jaw to rotate extensively relative to the posterior half. Where the two dentaries meet along the midline, a robust cartilaginous linkage prevents the dentaries from spreading apart, but does allow them to rotate. These descriptions (both of typhlopids and leptotyphlopids) are taken from Kley (2001, 2006), which you should consult for more detail: for free pdfs visit the Kley lab homepage.
How exactly do these snakes use these bizarre jaws? Until recently nobody knew, with suction-feeding and some sort of tongue-assisted feeding being suggested. New technology has allowed Nathan Kley of Stony Brook University to examine and answer these questions. Leptotyphlopids, with immobile maxillae and flexible lower jaws that have short, laterally convex dentaries, use a bizarre ‘mandibular raking’ method to get prey into the mouth: the anterior part of the lower jaw is rapidly rotated about the intramandibular joint, the transversely oriented teeth grabbing prey and pulling them backwards. Typhlopids, which their super-mobile maxillae that are normally kept folded up against the palate, can actually move each maxilla independently, and when feeding they seem to rapidly rotate the maxillae left-right-left-right in and out of the mouth, grabbing prey items with the teeth and then dragging them backwards. They are therefore practising a unique sort of ‘maxillary raking’ that allows multiple small prey objects to be swiftly pulled into the mouth (Kley & Brainerd 1999, Kley 2001) (the mechanics of feeding remain unstudied in anomalepidids but their typhlopid-like maxillae suggest that they also practise maxillary raking).
Maxillary raking is of course reminiscent of the ‘pterygoid walking’ used by macrostomatan alethinophidians: when ingesting prey, macrostomatans ‘walk’ the prey backwards into the mouth by pulling back on the prey with the toothed maxilla and pterygoid on the left side, then with the maxilla and pterygoid on the right, and so on. This is called unilateral feeding, and a reasonable amount of literature has been devoted to its function and origin. Unlike macrostomatans though, typhlopids do not use their toothless palatal bones in transporting prey, and other details of the feeding styles are also very different from those of macrostomatans. For these reasons, Kley (2001) favoured the idea that unilateral feeding is unique to alethinophidians, and was not present in the snake common ancestor. This contradicts another scenario (Lee et al. 1999) where unilateral feeding was inferred to be primitive for the entire snake clade….
However, Kley (2006) later argued that various skull details present in leptotyphlopids suggest that these snakes evolved from ancestors that practised either macrophagy (the ingestion of large prey) or unilateral feeding, and the various paedomorphic features present in leptotyphlopids also led him to lean towards the idea that scolecophidians are ‘regressive’ animals that have descended from macrophagous ancestors. This is the ‘classical’ view on these snakes, favoured during the 19th century by Boulenger, Cope and others. I’d love to explain the reasoning behind all this but I’ve run out of time.
Indeed there’s so much more that could be said about these fascinating snakes: about how they live, about the global spread of one parthenogenetic species by way of the horticulture trade, and more about what they might tell us about early snake evolution. And, like the also fossorial amphisbaenians and caecilians, their cryptic lifestyles, confusing morphologies, and poorly known geographic ranges have combined to make them a very exciting area for systematic and phylogenetic work.
Coming soon: big news in a big journal!
Refs – –
Gehlbach, F. R., Watkins, J. F. & Reno, H. 1968. Blind snake defensive behaviour elicited by ant attacks. Bioscience 18, 784-785.
Gower, D. J., Vidal, N., Spinks, J. N. & McCarthy, C. J. 2005. The phylogenetic position of Anomochilidae (Reptilia: Serpentes): first evidence from DNA sequences. Journal of Zoological Systematics 43, 315-320.
Khan, M. S. 1999. Two new species and a subspecies of blind snakes of genus Typhlops from Azad Kashmir and Punjab, Pakistan (Serpentes: Typhlopidae). Russian Journal of Herpetology 6, 231-240.
Kley, N. J. 2001. Prey transport mechanisms in blindsnakes and the evolution of unilateral feeding systems in snakes. American Zoologist 41, 1321-1337.
– . 2003. Blindsnakes (Typhlopidae). In Hutchins, M., Murphy, J. B. & Schlager, N. (eds) Grzimek’s Animal Life Encyclopedia, 2nd edition. Vol. 7. Reptiles, pp. 379-385. Gale Group, Farmington Hills (MI).
– . 2006. Morphology of the lower jaw and suspensorium in the Texas blindsnake, Leptotyphlops dulcis (Scolecophidia: Leptotyphlopidae). Journal of Morphology 267, 494-515.
– . & Brainerd, E. L. 1999. Feeding by mandibular raking in a snake. Nature 402, 369-370.
Lee, M. S. Y., Bell, G. L. & Caldwell, M. W. 1999. The origin of snake feeding. Nature 400, 655-659.
– . & Scanlon, J. D. 2002. Snake phylogeny based on osteology, soft anatomy and ecology. Biological Reviews 77, 333-401.
McDowell, S. B. 1967. Osteology of the Typhlopidae and Leptotyphlopidae: a critical review. Copeia 1967, 686-692.
Shea, G. M. 2001. Spermatogenic cycle, sperm storage, and sertoli cell size in a scolecophidian (Ramphotyphlops nigrescens) from Australia. Journal of Herpetology 35, 85-91.
Wallach, V., Brown, R. M., Diesmos, A. C. & Gee, G. V. A. 2007. An enigmatic new species of blind snake from Luzon Island, northern Philippines, with a synopsis of the genus Acutotyphlops (Serpentes: Typhlopidae). Journal of Herpetology 41, 690-702.
– ., Mercurio, V. & Andreone, F. 2007. Rediscovery of the enigmatic blind snake genus Xenotyphlops in northern Madagascar, with description of a new species (Serpentes: Typhlopidae). Zootaxa 1402, 59-68.
Webb, J. K., Branch, W. R. & Shine, R. 2001. Dietary habits and reproductive biology of typhlopid snakes from southern Africa. Journal of Herpetology 35, 558-567.
– ., Shine, R., Branch, W. R. & Harlow, P. S. 2000. Life-history strategies in basal snakes: reproduction and dietary habits of the African thread snake Leptotyphlops scutifrons (Serpentes: Leptotyphlopidae). Journal of Zoology 250, 321-327