The pygopodid radiation: diverse diets and the 'pygopodids got there first' hypothesis (gekkotans part X)

The previous article - part of my now lengthy series on gekkotan squamates (see links below) - provided an introduction to the neat and fascinating near-limbless Australasian gekkotans known as the pygopodids. Disclaimer: the group being discussed here is 'Pygopodidae of tradition', not Pygopodidae as currently formulated. More on this matter later.


One topic that I didn't explore fully in the previous article is pygopodid diversity. These reptiles aren't all samey little generalists; species within the group practise several different lifestyles and foraging behaviours, and the amount of morphological variation present within Pygopodidae is impressive [composite above shows Burton's snake-lizard (l) and Ophidiocephalus (r) at top: both by Stewart Macdonald, used with permission. Common scaly-foot Pygopus lepidopodus below by Peter Woodardlong, from wikipedia]. As we'll see below, it may in fact be that pygopodids evolved and diversified early enough to 'beat' a far larger, far more widespread group of squamates - snakes - into the occupation of several ecological roles.

Diverse diets and foraging styles

Some pygopodids (like Ophidiocephalus) are semi-fossorial and hunt arthropods concealed in sediment or leaf litter. Like other burrowing squamates, they have countersunk lower jaws and reduced eyes.


Others (certain Delma species, including 'Aclys' [read on]) are excellent climbers in low shrubs and were even described as "fully arboreal" by Pianka & Vitt (2003, p. 187). Unpublished work indicates that some of these lizards are partially herbivorous, with up to 20% of the diet in 'Aclys' being made up of plant material. A population of Brigalow scaly-foot Paradelma orientalis found on Boyne Island, Queensland, is known to include sap in its diet: both juveniles and adults climb up the trunks of Acacia falciformis and eat from the sap balls that ooze from the plant's tissues (Tremul 2000) [adjacent image, showing captive Brigalow scaly-foot eating from a sap ball, by geckodan, from the Aussie pythons & snakes forum]. Nectarivory and sap-eating is well known elsewhere in Gekkota (e.g., Cooper & Vitt 2002), so I'm not sure if its discovery in pygopodids is surprising. Nevertheless it's weird to see snake-like squamates consuming plant material.


Some pygopodids are terrestrial ambush predators that lurk in arid habitats while others are active foragers for arthropods in scrub and grassland. The Common scaly-foot eats a lot of mygalomorph spiders (c. 70% of diet) while Py. nigriceps eats more scorpions than any other lizard [adjacent pic shows a very thoughtful Py. lepidopodus; by Stewart Macdonald]. Most studied Delma species feed predominantly on crickets and moths, with spiders, cockroaches, beetles and bugs making up lesser proportions of their diet (Patchell & Shine 1986).

Some or most Aprasia species seem to be ant specialists, and specialists of ant larvae and pupae at that: the stomach contents of some species contain nothing but these items (Webb & Shine 1994). It's tempting to suggest that Aprasia is convergent on typhlopid blindsnakes (covered on Tet Zoo, together with other scolecophidian snakes, back in May 2008). More on this later. The Lialis pygopodids are predators of smooth-scaled lizards, especially skinks. I'll be talking a lot more about them later as well.

A phylogeny for pygopodids

Kluge (1976) was first to look at pygopodid phylogeny and, using morphological characters, found evidence for four major clades. (1) Pygopus was outside the clade that contained all the other taxa. This consisted of a (2) Lialis + (3) Aprasia clade (Pletholax and Ophidiocephalus were basal members of the Aprasia lineage), with (4) Delma forming the sister-group to Lialis + Aprasia. He proposed a classification based on his phylogeny, but it didn't actually reflect what he found: he placed Pygopus and Delma within a Pygopodinae 'subfamily' (even though he didn't find these two to group together), and he placed all other taxa within a Lialisinae 'subfamily'. Lialisinae then consisted of a Lialisini 'tribe' (for Lialis) and an Aprasiaini 'tribe' (itself broken down into three 'subtribes', one of which - the one for Ophidiocephalus - was accidentally also called Aprasiaini). Pygopus was actually paraphyletic in his analysis. The taxonomy he proposed was never really followed. Here's a substantially simplified version of his phylogeny (illustrations by Alan Male, from Philip Whitfield's 1983 Reptiles & Amphibians)...


Because Kluge (1976) found Aclys to be part of the Delma lineage, and Paradelma to be part of the Pygopus lineage, he advocated the abandonment of both generic names. This has mostly, but not universally, been followed by other authors.

Jennings et al. (2003) used mtDNA and nDNA data to analyse pygopodid phylogeny, and they generated various different topologies. In their favoured phylogeny, Delma was outside a clade formed by 'all other pygopodids'. In the 'all other pygopodids' clade, Lialis was sister to a (Pygopus + (Pletholax + (Ophidiocephalus + Aprasia))) clade. Here's a substantially simplified version of that phylogeny...


What does this new topology tell us about pygopodid evolution? Plotted against time (and using two different molecular clock estimates), the phylogeny indicates both that the major divergences within Pygopodidae happened prior to 25 million years ago (that is, during the Oligocene), and that extant pygopodid species originated some time between the start of the Miocene and now. Jennings et al. (2003) proposed that "diversification rates peaked early in the group's history before levelling out over the past 10 million years, perhaps as ecological niches became filled" (p. 776).

This Miocene/post-Miocene diversification of modern lineages is likely linked to the aridification that occurred across Australia at the same time: it fits with Eric Pianka's hypothesis that aridification was a major factor driving speciation across Australian lizard clades (Pianka 1972). However, Pianka proposed that most of the species concerned originated during the Pleistocene, not during the Pliocene or Miocene.

A peculiarity of the Jennings et al. (2003) phylogeny is that various pygopodid sister-species live in sympatry. Furthermore, these sympatric species pairs are younger than allopatric species pairs seen elsewhere in the phylogeny - a discovery that's somewhat contrary to expectations and raises various questions about speciation and range expansions within the group.


The phylogeny is also interesting in showing that Aprasia consists of two deeply diverging clades, one restricted to the south-west of Australia, and another restricted to the south-east [see figure above, from Jennings et al. (2003). Note that the radiations within the two Aprasia clades shown have mostly occured in situ]. Similar patterns are seen in some Delma clades, and in fact are known in other groups of Australian animals. Because the south-west and south-east corners of the continent preserve more mesic habitats than much of the rest of Australia, it's been proposed that they act as refugia for taxa that evolved prior to Australia's aridification. The data from pygopodids supports this model, since (1) the respective clades seem to have diverged long prior to the late Pliocene and Pleistocene, and (2) the south-western and south-eastern clades mostly diversified in situ, indicating that they're 'stranded' as relicts, with vicariance being the main factor explaining their distribution.

Remember that this sort of thing isn't true for all pypogodids, since various arid-adapted species have enormous ranges and have clearly spread across much of the continent.

Pygopodids vs snakes?


I mentioned in the previous article that "the snake-like niches occupied by some pygopodid taxa might explain why certain kinds of snake are low in diversity or absent in parts of arid Australia". This was inspired by a comment in Patchell & Shine (1986) where it was noted that both diurnal ambush-hunting and insectivorous snakes are rare or absent in Australia. Regarding the absence of the latter, Patchell & Shine (1986) noted that "The lack of such forms may in some way be related to the presence of the insectivorous pygopodids" (p. 38). [Adjacent image shows Delma nasuta and its freaky eyes; photo by Stewart Macdonald, used with permission.]

Now that we have some data on the timing and nature of the pypogodid radiation, it's interesting to see whether this can be tested. It's worth stating to begin with that the evolution of blindsnake-mimicking pygopodids hasn't obviously constrained or prevented the evolution of blindsnakes themselves in Australia, since the continent is home to a modest blindsnake radiation (containing over 30 species) that spans east to west and north to south. Presumably, there's enough 'ecospace' for both blindsnakes and blindsnake-mimicking pygopodids to co-exist.


What about the other snake groups that might be imagined as pygopodid competitors? By far the most speciose snake clade in Australia is Elapidae. It includes a diversity of slender-bodied, colubrid-like predators of lizards as well as frog-eating, cobra-like species and heavy-bodied, viper-like forms. Interestingly, both molecular and fossil evidence indicate that the Australian elapid radiation is geologically young, post-dating the start of the Miocene at the earliest (Scanlon et al. 2003, Scanlon & Lee 2004, Sanders et al. 2008). Australia also has various of those colubroids that were historically grouped together as 'colubrids', though admittedly not many (about 11 species). Again there are indications that they're a relatively young phenomenon: their presence in Australia almost certainly doesn't pre-date the Miocene, and might even be a post-Miocene phenomenon. [Adjacent montage shows a variety of Australian elapids and one homalopsid. From top to bottom: Common death adder Acanthophis antarcticus, Yellow-faced whip snake Demansia psammophis, Mainland tiger snake Notechis scutatus and Bockadam Cerberus rynchops. All images from wikipedia.]

If Jennings et al. (2003) are right about some of the major diversifications in pygopodid phylogeny occurring prior to the Miocene, it could be that 'pygopodids got there first' filling up some of the niches that would otherwise have been occupied by certain small snakes. I don't want to create the wrong impression here: let's emphasise that elapids have done very nicely indeed in the Australian fauna, while both pythons and madtsoids are also (or have also been) important constituents of the continent's herpetofauna. Nevertheless, the idea that pygopodids diversified early enough to occupy at least some 'small snake' niches is a fascinating possibility that (so far as I can tell) looks reasonable based on the data we have.

The Jennings et al. (2003) phylogeny isn't, of course, abstracted from molecular data alone, since a fossil species of Pygopus (P. hortulanus) is known from the Miocene (Hutchinson 1997).

In the next article we'll be having a far more detailed look at the various pygopodid clades.

For previous articles in the gekkotan series, see...

And for previous Tet Zoo articles on other kinds of squamates, please see...

Refs - -

Cooper, W. E. & Vitt, L. J. 2002. Distribution, extent, and evolution of plant consumption by lizards. Journal of Zoology 257, 487-517.

Hutchinson, M. N. 1997. The first fossil pygopod (Squamata, Gekkota), and a review of mandibular variation in living species. Memoirs of the Queensland Museum 41, 355-366.

Jennings WB, Pianka ER, & Donnellan S (2003). Systematics of the lizard family pygopodidae with implications for the diversification of Australian temperate biotas. Systematic biology, 52 (6), 757-80 PMID: 14668116

Kluge, A. G. 1976. Phylogenetic relationships in the lizard family Pygopodidae: an evaluation of theory, methods and data. Miscellaneous Publications, Museum of Zoology, University of Michigan 152, 1-72.

Patchell, F. C. & Shine, R. 1986. Food habits and reproductive biology of the Australian legless lizards (Pygopodidae). Copeia 1986, 30-39.

Pianka, E. R. 1972. Zoogeography and speciation of Australian desert lizards: an ecological perspective. Copeia 1972, 127-145.

- . & Vitt, L. J. 2003. Lizards: Windows the Evolution of Diversity. University of California Press, Berkeley.

Sanders, K. L., Lee, M. S. Y., Foster, R. & Keogh, J. S. 2008. Molecular phylogeny and divergence dates for Australasian elapids and sea snakes (hydrophiinae): evidence from seven genes for rapid evolutionary radiations. Journal of Evolutionary Biology 21, 682-695.

Scanlon, J. D. & Lee, M. S. Y. 2004. Phylogeny of Australasian venomous snakes (Colubroidea, Elapidae, Hydrophiinae) based on phenotypic and molecular evidence. Zoologica Scripta 33, 335-366.

- ., Lee, M. S. Y. & Archer, M. 2003. Mid-tertiary elapid snakes (Squamata, Colubroidea) from Riversleigh, northern Australia: early steps in a continent-wide adaptive radiation. Geobios 36, 573-601.

Tremul, P. R. 2000. Breeding, feeding and arboreality in Paradelma orientalis: a poorly known, vulnerable pygopodid from Queensland, Australia. Memoirs of the Queensland Museum 45, 599-609.

Webb, J. K. & Shine, R. 1994. Feeding habit and reproductive biology of Australian pygopodid lizards of the genus Aprasia. Copeia 1994, 390-398.


More like this

I really want to get these pygopodid articles finished. Actually, I really want to get the whole gekkotan series finished: the end is in sight and I know I'll get there eventually. In the previous articles on pygopodids (part of the long-running series on gekkotan lizards: see links below), we…
One of my shortish-term goals at Tet Zoo has been to complete the series on gekkotan lizards I started in April 2010 (see below for links to previous parts). We continue with that series here, and this time round we're going to look at what should definitely be regarded as the weirdest of…
In the previous article I provided brief reviews of all currently recognised pygopodid 'genera'*. Except one. I've left this one until last, largely because it's the most spectacular (up to 75 cm in total length) and (arguably) most fascinating pygopodid. We've seen throughout this series of…
As you may know, Tet Zoo has been going for four years now. Despite this, there are still entire tetrapod clades - consisting of hundreds or even thousands of species - that have scarcely been mentioned here, if at all. Lately, I've been feeling 'gecko guilt'. Yes, I can barely believe that the…

Are there venomous pygopodids?

Trivial editing note: the link to the Scolecophidian article appears to be broken.

By heteromeles (not verified) on 06 Jun 2011 #permalink

Nope, no venomosity in pygopodids, nor in gekkotans at all so far as we know (even post-Fry et al. 2005). Thanks for note regarding scolecophidian link - now fixed.

Looking forward to the 'Lialis special'


their presence in Australia almost certainly doesn't pre-date the Miocene, and might even be a post-Miocene phenomenon

The distribution patterns of those Australian colubrid species strongly suggest that they're quite recent arrivals indeed. All species are pretty much restricted to the northern and to the eastern coastal parts of Australia. None of them has spread very far inland, especially not to the arid interior.

It may also be noted that colubrids (unlike, e.g., pythons and elapids) apparently have no fossil record at all in Australia*. Admittedly absence of evidence is not evidence of absence, but one would have expected colubrid remains to have been found at Riversleigh at least, had they been present in Australia before the Pliocene.

* At least not AFAIK; John Scanlon would know better.

Isn't saying "Py. nigriceps eats more scorpions than any other lizard" sort of like saying "Confuciusornis has comparatively longer primary feathers than any other Mesozoic bird"? Or are lizard diets more well researched than I assume?

Dartian: remember that you're talking about 'colubrids', not colubrids in the strict sense. Even in the most conservative taxonomies (e.g., the annoying one just published by Pyron et al. (2011)*), at least some of the Aussie 'colubrids' belong to non-colubrid clades like Homalopsidae.

* Annoying because I disagree with their contention that keeping the many, many 'colubrid' clades as 'subfamilies' is more logical than raising them to 'family' rank.

Mickey: sure, a statement like "Py. nigriceps eats more scorpions than any other lizard" very obviously actually means "Py. nigriceps eats more scorpions than any other lizard for which we have dietary data". And, unlike primary feather length in Mesozoic birds, you could argue that we have quite a lot of data on the diets of extant lizards (albeit, for some lineages, not as much as we'd like).

And John Scanlon is obviously very busy right now, or on holiday :)

Ref - -

Pyron, R. A., Burbrink, F. T., Colli, G. R., Montes de Oca, A. N., Vitt, L. J., Kuczynski, C. A. & Wiens, J. J. 2011. The phylogeny of advanced snakes (Colubroidea), with discovery of a new subfamily and comparison of support methods for likelihood trees. Molecular Phylogenetics and Evolution 58, 329-342.


remember that you're talking about 'colubrids', not colubrids in the strict sense

You're right, I should have made it clear that I was talking about colubrids sensu traditionalis.

Chalky: Although both snakes and lizards are squamates, snakes are not lizards. As for leglessness and near-leglessness, the wikipedia article has the legless 'families' right AFAIK.

By Daniel Fors (not verified) on 07 Jun 2011 #permalink

snakes are not lizards

Well, lizards are paraphyletic with respect to snakes, so the snakes do belong in that list.

By David MarjanoviÄ (not verified) on 07 Jun 2011 #permalink

Hey, did I miss something? Actually, I apparently missed a few issues of MPE, among other stuff, while moving 3/4 of the way across the continent and starting a non-palaeo, non-systematics job (involving a certain amount of tropical fieldwork, as a consolation). I just downloaded a bunch of papers to read in the next few days...

Diurnal ambush-hunting... Demansia do that, lurking in plain sight with head elevated as in the photo above, then sprinting after passing prey. Real vertebrate prey, not grasshoppers like those wannabe Delmas. Though Delma are pretty cool too.

Australian colubrid fossils: better not spill all the beans, but I think I can say that there are still no pre-Pleistocene examples known.

Daniel & Chalky: snakes are lizards, in the sense that their ancestors were stumpy, quadrupedal, ear-holed, flat-tongued, pleurodont squamates. They got better, but it would be misleading to say they're not lizards anymore.

This sap-licking behaviour is interesting. Imagine a Miocene pygopod, maybe young and a bit silly, and not seeing well 'cos it's getting ready to slough, sticking its face into a blob of araucarian resin instead of wattle sap by mistake. It could lead to a rather unusual kind of amber fossil, if anyone was looking out for that kind of thing. :)

By John Scanlon, FCD (not verified) on 07 Jun 2011 #permalink

Hm, snakes are still lizards; but humans are certainly not fish anymore? Is there any consensus as to where we should draw the line for that sort of thing?

PS: Or maybe we SHOULD consider ourselves fish? Interesting dilemma :- )

To be consistent, I say we're fish. I don't mind calling sharks and lampreys and conodonts fish either, though 'serranid' is really the image I'm going for.

Personally, I can also get my head around the syntax of named groups that happen to be paraphyletic; but if you casually dichotomize 'fish vs tetrapod', 'lizard vs snake', or 'animal vs human' you miss a chance to point to how things really are: descent is inclusion.

By John Scanlon, FCD (not verified) on 07 Jun 2011 #permalink

I'm all in favour of considering humans "fish", but there seems to be a tendency these days to use "fish" = Actinopterygii. I was quite surprised the first time I heard a nature show speak of "sharks and fish", but it wouldn't be a bad thing if it caught on universally.

By Andreas Johansson (not verified) on 07 Jun 2011 #permalink

The problem with all of these arguments - it goes for 'reptile', 'fish', 'lizard', 'ape' and so on - it that these are not technical words. A biologist may equate 'lizard' with the clade Squamata (and hence include all squamates as lizards, even if they're very modified lizards) whereas general/common usage has it that lizard is 'lizard-shaped reptile different from snake'. I'm all for using terms that promote an awareness of evolution (e.g., humans are apes, birds are dinosaurs), but I think we can afford to be relaxed about the least formal of these names. Are snakes lizards? Yes, and no. Are snakes squamates? Indisputably yes.

Me, I like names for paraphyletic groups.
Gary Nelson says "There is no such thing as a reptile," and I think the point he is making when he says it is a good one: "reptiles" are paraphyletic w.r.t birds, and the last common ancestor of squamates and crocodilians (or of either of them and turtles) lived a long time ago: calling them all reptiles suggests that they share traits, and -- given the number of generations since the common ancestors -- the odds are that in lots of details crocodilians are closer to birds than to squamates.


Pelycosaurs are paraphyletic w.r.t. Therapsids, and non-Therapsid "Pelycosaurs" seem to have died out not TOO long after the origin of Therapsids. It's really convenient, if I want to say that some trait is a Therapsid apomorphy, to be able to say that Therapsids have it but Pelycosaurs don't.

The difference seems to be that the paraphyletic Pelycosauria are also a fairly uniform "grade" group, whereas (I think Nelson is right in holding that) one shouldn't think of "reptiles" as a reasonably natural "grade" of... Sauropsids.

My guess is that, given the diversity of extant lizards, and the fact that snakes are more closely related Varanids than Varanids are to, say, Gekkotans, it is probably helpful and instructive to say "Snakes are lizards"... at least sometimes.

By Allen Hazen (not verified) on 07 Jun 2011 #permalink

I think it's interesting that lizards...ahem...squamates so readily lose their limbs and adopt a snake-like lifestyle. Is there some kind of genetic "tendancy" among squamates to do this? Some underlying reason why it's so widespread?

the paraphyletic Pelycosauria are also a fairly uniform "grade" group

I disagree... but they're more uniform than "reptiles" as a whole, of course.

Is there some kind of genetic "tend[e]ncy" among squamates to do this?

I guess it's the other way around: when squamates lose their limbs, most of them get away with it more easily than, say, crocodiles.

By David MarjanoviÄ (not verified) on 08 Jun 2011 #permalink

The problem with supporting traditional definitions of non-technical words like "lizard", "fish", "ape", etc, is that those traditional definitions promote misunderstandings about common descent.

It's worth remembering that the word "fish" originally just meant "an animal that lives in the water". Thus we have shellfish, jellyfish, crayfish, etc.

It's one thing for science to create a new word (like squamate or spinodal or brontosaurus) and then correct people who use it incorrectly (if they use it at all). It's another thing to take a word that's been in use for centuries (like fish or work or action), give it a technical meaning very different from its original, common meaning, and then "correct" the older usage.

But common words already change their meanings anyway. How many people still consider shellfish, jellyfish, crayfish, starfish or cuttlefish "fish"? How many people with any education don't consider it an embarrassing mistake to call whales fish nowadays? Apparently it's even starting for sharks, see comment 16.

By David MarjanoviÄ (not verified) on 22 Jun 2011 #permalink

Or, to pick an extreme example, who still calls a snake a worm...?

By David MarjanoviÄ (not verified) on 23 Jun 2011 #permalink

Who calls them shellFISH? More than call them bivalves. Same with the others.

The point isn't that words change their meanings over time; the point is that technical language is only technical language. Most people now know and accept that human beings are descended from fish, but imagine the following situation: a biology professor witnesses a bank robbery. When interviewed by police, he describes what he saw.

"I was flirting with the pretty female ape the bank just hired as a new teller, when these two fish came in, waved some guns around, and told everyone to get on the floor. I couldn't see anything identifying about their heads, since both fish were wearing pantyhose over their heads. One of them was about 5'6", 140 lbs, the other fish looked to be about 6' even, maybe 200 lbs."

Or more simply, just try calling your wife/girlfriend an ape or a fish.

When interviewed by police, he describes what he saw.

As you can't have helped notice, "fish"* and even "ape" are way too unspecific for this purpose.

* Why not "vertebrate"?

Or more simply, just try calling your wife/girlfriend an ape or a fish.

If she's insulted, you know the school system has failed her.

By David MarjanoviÄ (not verified) on 23 Jun 2011 #permalink