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 gekkotans: the near-limbless pygopodids, pygopods or flap-footed lizards, all of which inhabit Australia and New Guinea (and at least some of the surrounding islands). Because there's a lot to say about them, the article you're reading is the first of three. [Excellent paintings below by Alan Male, from Philip Whitfield's 1983 Reptiles & Amphibians. These were the first images of pygopodids I ever saw.]
For the record, I'm using Pygopodidae in the restrictive, 'traditional' sense here. As we'll see in a later article, most authors now use Pygopodidae for a more inclusive clade that also includes certain limbed gekkotan lineages (if you can't wait until then, there's preliminary discussion of the issue at hand here).
Anyway, Pygopodidae of tradition consists of about 40 species arranged in eight genera (Aclys, Aprasia, Delma, Lialis, Ophidiocephalus, Paradelma, Pletholax and Pygopus), though not all of these names are used by all authors. As is typical for Australasian herps, new species are described fairly regularly. The newest at the time of writing is Roberts' scaly-foot Py. robertsi Oliver et al., 2010 from Queensland.
The literature on pygopodids is fairly scattered across the herpetological journals. Harold Cogger's Reptiles & Amphibians of Australia includes species-by-species accounts with excellent photos (Cogger 2000), and an enormous amount of information on all taxa known up to the time of publication is included in Arnold Kluge's monograph on the group (Kluge 1974) (oh, how I'd love an original copy).
Right across Australia, and hello New Guinea
Pygopodids occur right across continental Australia, but don't occur on Tasmania (though... there are some Tasmanian records; to be discussed in a later article). They're extremely abundant in some areas: Kluge wrote of areas where Aprasia individuals could be discovered beneath virtually every single slab or chip or rock, and it's common to find several species living in sympatry. Indeed, in the area that seems to be the centre of pygopodid distribution (the lower west coast of Western Australia), as many as ten species can be found in the same area (Shea 1989).
While New Guinea is home to pygopodids, only the two Lialis species occur there. Of those, one of them - Burton's snake-lizard L. burtonis - is tremendously widespread across the pygopodid range in general, occurring right across mainland Australia and also being found on islands in the Torres Strait and on the Aru Islands (Kluge 1974) [L. burtonis shown above; image by Smacdonald, from wikipedia]. The other - L. jicari - is unknown from Australia; as well as mainland New Guinea it's also present on New Britain (Kluge 1974).
Some pygopodids (like the Delma species) can be regarded as 'primitive' within the group as they have external ear openings, relatively large hindlimbs and a crushing dentition. Others, however, are more specialised and are remarkably convergent with snakes, lacking external ear openings and possessing slim, lengthy skulls and recurved, slender teeth. In fact, 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 (Patchell & Shine 1986). Some pygopodid species (including species of Delma, Paradelma and Pygopus) seemingly mimic elapid snakes when threatened, raising their dark-patterned heads, flattening their necks and flicking their tongues (Johnson 1975, Pianka & Vitt 2003). Most pygopodids are convergent with snakes in possessing large head scales (ironically, the especially snake-like Lialis is the exception). We'll be having a more detailed look at phylogeny and evolutionary history in the next article. [Adjacent photo shows a juvenile Py. schraderi. Photo by Henry Cook, from the Aussie pythons & snakes forum.]
Partly because limblessness is assumed to be disadvantageous to a tetrapod, it's sometimes been regarded as paradoxical that pygopodids are extremely diverse in ecology and behaviour. Indeed, they do several things that 'normal' gekkotans don't. Rather than being inhibitive to diversification and speciation, it seems that the evolution of limblessness and a snake-like body plan allowed pygopodids to undergo a fairly impressive radiation. Or, to quote Webb & Shine (1994): "For reasons that are unclear, the evolution of an elongate limb-reduced morphology in early pygopodids appears to have opened the floodgates for subsequent evolutionary modifications of an astonishing scale" (p. 397).
Some basics of anatomy, behaviour and biology
The pygopodid pectoral girdle is a vestigial, V- or U-shaped structure arranged ventral to the fourth, fifth and sixth vertebrae (Stokely 1947). There is never any external trace whatsoever of forelimbs, but a humerus is present internally in Aprasia, Pletholax and Ophidiocephalus (Kluge 1976). The humerus is a 'floating' structure: it doesn't articulate with the scapula.
In the pelvis, the ilium has remained large (often, but not always, maintaining a contact with a single vertebra) and it still has the recognisable form typical of squamates. The hindlimb is variable. In some taxa (e.g., some species of Aprasia), just a vestigial, rod-like femur is present, while in others (e.g., some species of Pygopodus) the hindlimb is complete and still has four digits [hindlimb of Py. lepidopodus shown above; from Shea (1989)]. However reduced it is, this limb skeleton forms the core of the flap-like hindlimb that explains some of the common names used for members of this group. These limbs are quite flexible and can be abducted to a degree normal for a lizard's hindlimb (the animals sometimes do this by choice, as if it helps them to 'stand' on the substrate). Most sources state that males use these limbs to help grip females during mating.
However, little mentioned is that males of at least some species (or all species?) also possess paracloacal spurs, one on each side of the cloaca (and normally hidden from view by the hindlimbs) [adjacent photo shows paracloacal spur (in box) of Py. schraderi, as revealed by lifting hindlimb. Photo by Henry Cook, from the Aussie pythons & snakes forum]. They look much like the far better known spurs of male boas and pythons. Evidently, the pygopodid 'spurs' are nothing to do with the hindlimbs - they're neomorphs (and they aren't unique to pygopodids: they also occur in carphodactylid and diplodactylid gekkotans). This got me thinking.
It's widely thought that the cloacal spurs of boas and pythons are relictual hindlimb remnants - but could they actually be neomorphs too, and not genuine hindlimbs? Well, no. Embryological work demonstrates pretty convincingly that the 'relictual hindlimbs' of certain modern snakes really are relictual hindlimbs (Cohn & Tickle 1999) (this issue was previously discussed in Monster pythons of the Everglades: Inside Nature's Giants series 2, part II). On a more speculative note, does the presence of cloacal spurs mean that pygopodids have the evolutionary potential to evolve an extra set of posterior limbs?
Like other gekkotans, pygopodids possess cloacal bones (also known as post-cloacal bones). These curved, serrated structures project from the tips of the everted hemipenes in males and various roles in copulation have been suggested.
Tail length is highly variable in pygopodids: in some species the tail is as much as four times as long as snout-vent length (SVL), while in others it's much less than SVL. As we'll see later, at least one species uses the tail tip as a lure.
Some pygopodids are able to jump: to literally lift the entire body and tail off the substrate and repeatedly move upwards and forwards in a weird, jerky manner. A 74-cm-long Delma got its whole body 7 cm off the ground (Bauer 1986). They do this to startle or disorientate potential predators. A few other limbless squamates (including some amphisbaenians, vipers, gymnophthalmids and glass lizards) can do the same trick. When climbing in shrubs, Delma is able (in emergencies) to use the tail to leap, at the same time straightening its body. The result is that individuals "thrust ... through vegetation like arrows!" (Pianka & Vitt 2003, p. 27). Caudal autotomy is common in pygopodids, even in those saltating species. [Adjacent photo shows D. inornata; photo by Damian Michael, from Michael et al. (2010).]
Comparatively little is known about the life history of pygopodids. Data from individuals kept in captivity suggests that they are not long-lived (that it, not living for more than seven years or so. If correct, this is a massive contrast with Northern Hemisphere limbless squamates like anguids. They live for a few decades at least). We do know that females are larger than males: you might think that this is unusual for lizards, but it actually seems to be normal for gekkotans. Pygopodids are also typical among gekkotans in being oviparous and in only producing two eggs per clutch (though one or three eggs are sometimes produced by aberrant members of some species). Sexual dimorphism isn't limited to size: in Aprasia, males have teeth in their premaxillae while females (usually) don't (Aprasia also lacks teeth in the maxillae, so has a strongly reduced dentition overall: more on this later). This doesn't seem related to differences in diet, so might be the result of sexual selection pressure.
Some species are reported to emit buzzing and rattling noises when startled (Kluge 1976), most typically the Delma species. Squeaks have also been reported from some pygopodids. These are true vocalizations, and at least some of these noises might be employed in intraspecific communication. D. butleri is said to be able to rapidly brighten the yellow colour of its belly when stressed. [Image below shows a fairly scary-looking individual of L. burtonis; photo by Damian Michael, from Michael et al. (2010). This species is famously variable in pigmentation and patterning.]
I didn't get to cover pygopodid diversity much in this article, nor look at phylogenetic hypotheses proposed for the group. This is what we'll be covering next.
For previous articles in the gekkotan series, see...
- The Tet Zoo guide to Gekkota, part I
- Gekkota part II: loud voices, hard eggshells and giant calcium-filled neck pouches
- Squirting sticky fluid, having a sensitive knob, etc. (gekkotans part III)
- Lamellae, scansor pads, setae and adhesion... and the secondary loss of all of these things (gekkotans part IV)
- The incredible leaf-tailed geckos (gekkotans part V)
- 300 years of gecko literature, and the 'Salamandre aquatique' (gekkotans part VI)
- Whence Uroplatus and... there are how many leaf-tailed gecko species now?? (gekkotans part VII)
- Ptychozoon: the geckos that glide with flaps and fringes (gekkotans part VIII)
And for previous Tet Zoo articles on other kinds of squamates, please see...
- Pompey and Steepo, the world-record-holding champion slow-worms
- Arboreal alligator lizards - yes, really
- Amazing social life of the Green iguana
- Hell yes: Komodo dragons!!!
- Ermentrude the liolaemine
- Evolutionary intermediates among the girdled lizards
- The Great Goswell Copse Zootoca
- Of giant plated lizards and rough-necked monitors
- 'Cryptic intermediates' and the evolution of chameleons
- Tell me something new about basilisks, puh-lease
- Tongues, venom glands, and the changing face of Goronyosaurus
- Mosasaurs might have used the same microscopic streamlining tricks as sharks and dolphins
- Dinosaurs come out to play (so do turtles, and crocodilians, and Komodo dragons)
- Isopachys: worm-like skinks from Thailand and Myanmar
- Mystery emo skinks of Tonga!
- Cambodia: now with dibamids!
Refs - -
Bauer, A. M. 1986. Saltation in the pygopodid lizard, Delma tincta. Journal of Herpetology 20, 462-463.
Cogger, H. G. 2000. Reptiles & Amphibians of Australia (Sixth Edition). New Holland Publishers, Sydney.
Cohn, M. J. & Tickle, C. 1999. Developmental basis of limblessness and axial patterning in snakes. Nature 399, 474-479.
Johnson, C. R. 1975. Defensive display behaviour in some Australian and Papuan-New Guinean pygopodid lizards, boid, colubrid and elapid snakes. Zoological Journal of the Linnean Society 56, 265-282.
Kluge, A. G. (1974). A taxonomic revision of the lizard family Pygopodidae. Miscellaneous Publications, Museum of Zoology, University of Michigan, 147, 1-221 [free pdf available here!]
- . 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.
Michael, D., Lindenmayer, D., Crane, M., Herring, M. & Montague-Drake, R. 2010. Reptiles of the NSW Murray Catchment: A Guide to Their Identification, Ecology and Conservation. CSIRO Publishing, Collingwood (Vic.).
Patchell, F. C. & Shine, R. 1986. Food habits and reproductive biology of the Australian legless lizards (Pygopodidae). Copeia 1986, 30-39.
Pianka, E. R. & Vitt, L. J. 2003. Lizards: Windows the Evolution of Diversity. University of California Press, Berkeley.
Shea, G. M. 1989. Family Pygopodidae. In Glasby, C. G., Ross, G. J. B. & Beesley, P. L. (eds). Fauna of Australia. Australian Capital Territory, Canberra. [published online.]
Stokely, P. S. 1947. The post-cranial skeleton of Aprasia repens. Copeia 1947, 22-28.
Webb, J. K. & Shine, R. 1994. Feeding habit and reproductive biology of Australian pygopodid lizards of the genus Aprasia. Copeia 1994, 390-398.
- Log in to post comments
Excellent stuff! Last time I was in Oz I got to see a Lialis - a wonderful animal. Also saw Pletholax - a thin wisp of a lizard with ridged scales - when held it goes stiff and feels strangely like a dry twig or piece of wire. More please.
I suppose if you really wanted to know if gekkotans could evolve into become six-limbed animals, you could try exposing them to teratogens. There were some rather startling frogs in the American Midwest a few years back - they apparently got hit with a fungus teratogen that caused wild leg duplication during the metamorphosis from tadpole to frog.
Of course, this didn't have a thing to do with pectoral spurs, but...
er, paracloacal spurs, rather. Pectoral spurs would be a little disturbing.
The effect that the trematode parasite Ribeiroia has on the morphology of frogs - namely, the appearance of exta limbs - was covered previously here.
Thanks for comments; much more on pygopodids to come.
Or, alternatively, Australian pygopodids have been able to occupy 'snake-like niches' because these certain kinds of snakes were absent/rare in Australia. (Colubrids, for instance, would seem to be relative newcomers in the Australasian region.)
To test this, you'd need info on the timing and nature of pygopodid and snake radiations. More on this when I publish the next article.
I've looked after a _Lialis burtonis_ before, while working for the WA Museum; they're wonderful animals, but fussy eaters - they seem to be preferentially reptivores (reptilivores? Squamatovores?). Eventually, he was released back to his original habitat, rather than decimate suburban populations of skinks to keep him going.
The real problem for the pygopodids is the fact they're snake-like; people have slaughtered them, and now they're pretty rare in suburbs - you have to live on the real fringes or in the bush to encounter them. If you look at _Pygopus_ as shown above, then look at a picture of a baby dugite, they're very very similar - a true tragedy :/ It makes sense as a survival technique - look like small venomous snake, get left alone. Sadly, *humans*.
@Marcus Good - according to Rick Shine (700 publications now and counting?) Lialis is a snake analogue with all the jaw adaptations needed to deal with big prey - skinks and geckos. And yes, the similarity to snakes doesn't help pygos when it comes to humans - I read somewhere that a photo in a local paper reporting on a 'plague of snakes' showed dead reptiles strung up in a line and many were large pygos
The very interesting convergence between Lialis and lizard-eating snakes is discussed at length in a later article - try not to spoil all the surprises :)
Saddened to hear that pygopodids have suffered from snake persecution... not that it's ok that idiot people feel this way about snakes anyway. Someone should devise a strategy to air-drop rodent plagues on people who wilfully eliminate snakes.
Yay! Pygopodids. Definitely some of my favourite squamates. Growing up in Adelaide, it was actually fairly easy to find Delma molleri and Aprasia striolata in relict suburban patches of bushland. I certainly found many in the reserve on the other side of the road from my house. I suspect these were small/cryptic enough to escape snake persecution. The larger Pygopus lepidopodus was much rarer in the Adelaide Hills, infact I only ever saw one individual from the hills.
OT. Aside from the Galapagos marine iguana that eats algae, are there any snake-like or lizard-like herbivore/frugivore/nugivore extant reptiles?
Superficially snakelike, but very different way of producing this external morphology: snakes have long torsos but short tails, whereas Pygopodids include species in which most of the length is tail (as well as some with more snake-like proportions)!
You wouldn't get that sort of variation in how the vertebral column subdivides in mammals.
@ 11 There are a number of herbivorous iguanas of all sorts. Some other lizards also include fairly large amounts of vegetation in the diet.
#11, #13: Uromastyx, from the previous TetZoo post, is primarily herbivorous. Most herbivorous lizards, though, are to some extent omnivorous. Ctenosaura, Iguana, Sauromalus, Uromastyx, et al., eat insects or other small critters as well as various plants. Further, some primarily carnivorous lizards eat plants as well; e.g., Crotaphytus will eat the fruits of Rhus species, and I used to keep an Uma that would eat bell peppers (although, judging from its excrement, it digested them rather poorly).
I think this is incorrect. All of the genera named are highly specialized herbivores (not certain about ctenosaurs) and any animal material taken in the wild-type diet is likely incidental. Even the growing juveniles, who could really use the protein, are very strongly herbivorous, though they tend to choose different plant parts.
Are you sure the free-floating humerus is a humerus? Could it be, like, a coracoid or something...?
[from Darren: sorry, delayed by spam filter. And - don't forget - it's Sauromalus ater these days, not S. obesus.]
#15: Well, most members of these genera will happily eat insects in captivity. Observations of wild diets are more limited. For a lot of these guys (e.g., almost all Uromastyx) there are no published reports of wild diet AFAICT. However, insects as a component of the wild diet have been confirmed in some cases (e.g., Ctenosaura pectinata, Sauromalus obesus).
1. Coordinated Ontogeny of Food Preference and Responses
to Chemical Food Stimuli by a Lizard Ctenosaura pectinata
(Reptilia: Iguanidae). William E. Cooper Jr & Julio A. Lemos-Espinal. Ethology 107, 639-653 (2001).
2. Dietary of the Chuckwalla, Sauromalus obesus, Determined by Dung Analysis. R. M. Hansen. Herpetologica 30, 120-123 (1974).
I've heard Lialis burtonis vocalize - a "yap" on being approached very closely in the wild - there is a paper describing what looks to be the same call (when handled).
The compost heap is full of these things; I have to turn it really carefully to avoid chopping them in half. We use them as an indicator of soil health around here. Whethere it's actually true or not I don't know
[from Darren: sorry, delayed by spam filter. No idea why.]
Greer (1989: Fig. 48) has line drawings of the pectoral girdle for one species each of all 8 pygopodid genera, based on double-stained specimens. [Darren, you really should have that book! It would save you a lot of lit-review work on so many of your lizard posts!] They're pretty well developed, though the only ossified elements are an interclavicle (present in Pletholax only), clavicles (mostly large and nearly or quite meeting in the midline; reduced and widely separated in Aprasia, absent in Ophidiocephalus), and scapulocoracoids (labelled 'scapulae' in the figure but not the text). The coracoid cartilages are either separated or overlapping in the midline and have 2, 1 or 0 fenestrae, and there are scapular cartilages dorsolaterally that contact the outer ends of the clavicles.
An unossified sternum is attached to 2, 1 or 0 pairs of ribs. The specialised burrowers don't have distinct sterna vs. coracoid cartilages: there's a diamond-shaped structure labelled 'sternum' in Ophidiocephalus that has a line down the middle and might be better identified as the cor. cart. since they're otherwise absent, and the bones articulate with it directly; Aprasia has a simple V-shaped cartilage joining the scapulocoracoids.
Anyway, the humeri, where present, 'float' just posterior to the scapulocoracoid (Aclys, Delma, Paradelma and Pygopus). They're much smaller than any of the other structures mentioned, but there's nothing else they could be.
There are various records of 'herbivory' in pygopods, mostly eating or licking fruit in captivity, but an unpublished thesis (Murray 1980, cited by Greer) reported 20% plant material in the diet of wild Aclys. He doesn't seem to say what kind of plant material. Paradelma's sap-licking is a more recent discovery but more widely known.
Where abouts are you from? That sounds more like Anomalopus, a limb-reduced skink, which is common in compost on the east cost. Its sometimes difficult when referring to legless lizards here in oz, as there are several radiations of limb-reduced skinks (Coeranoscincus, Anomalopus, Lerista, Ophioscincus, etc.)on top of the pygopodids.
Hey, why is my comment held up? There weren't even any links.
"Partly because limblessness is assumed to be disadvantageous to a tetrapod, it's sometimes been regarded as paradoxical that pygopodids are extremely diverse in ecology and behaviour."
This seems like a holdover from a time when scientists thought they knew better than nature. Given the ridiculous amount of times that limblessness has evolved in squamates, coupled with the high diversity of pygopodids and snakes, it seems pretty clear that limblessness poses some type of strong advantage.
@Patrick - Though an insectivorous diet might be seen in captivity, we need to be careful in what that actually says about wild animals. Juvenile iguanas will also readily take hamburger and Cheetos. Not exactly sure why the diets seem so plastic in captive critters vs. wild ones, but it is something to keep in mind.
It's amazing to see how many times limblessness evolved in lizards. Really makes you wonder what could have happened if the snakes had not flourished; would the niches we see filled by snakes today have been taken up by some other lineage of limbless lizard?
For what it's worth, Ctenosaura seems to be an opportunistic predator, and is known to prey on nesting seabirds such as terns and plovers, as well as sea turtle hatchlings.
I see, thanks!
I have started to suspect that the TetZoo spam filter is actually some sort of cryptid with a poor sense of visual discrimination, so that it gobbles up anything that triggers its gobble reflex.
Another hypothesis is that the poor thing is not nearly blind, but is suffering from severe malnutrition from its bland diet of commercial offerings. So it is deliberately going after healthy, robust, content-rich comments out of sheer desperation...!!
Somewhat more seriously, it may have something to do with the timing involved in comment posting -- if you post at the same time that a spammer is targeting TetZoo, the system may not be able to switch from "filter comment" to "allow comment to post" quickly enough.
Or... maybe it's something completely different. There may be log files somewhere that explain what triggered the dump to moderation, but I understand that Scienceblogs in general has been suffering from a severe lack of IT support.
I always thought that comments got spam-filtered based on the number of links. But, as David MarjanoviÄ noted recently, it appears in fact to be due to number of html tags. John's comment above contains over 20 html tags, so this seems like a reasonable hypothesis. I'm now going to test it...
Chondropython is green.
Hmm, so far so good. Six tags...
Chondropython is green but it's a Morelia these days.
Chondropython is green but it's a Morelia these days unless you're a fan of Wells & Wellington.
Last one before I'm giving up. Ten tags...
Chondropython is green but it's a Morelia these days unless you're a fan of Wells & Wellington. Lialis is also a python.
If it's the number of tags, then quoting the entire comment (and adding more tags of my own) should trigger moderation. Therefore, for science...!!
I do believe that's what's called a falsified hypothesis...
you couldn't have had better timing with this! I was just reading through the gecko articles you wrote, thinking "I really wish he'd expand on the pygopodids." I started doing my own investigating, but lo-and-behold, you read my mind.
thanks so much!
Just to hi-jack the thread ( :) ), how is it that while there are numerous examples of snake-like animals among reptiles and amphibians, we have none among mammals? Did the split between reptiles/amphibians and mammals occur before the split between amphibians and reptiles?
@Alan Kellogg: No, the amphibians split off first: snakes are more closely related to giraffes than to caecilians.
A snake-like body would be disadvantageous for a small endotherm, because it's got a high surface:volume ratio. Also, legged squamate locomotion is already fairly snakelike, with lateral undulation - a mammalian "snake" in spe would have a longer way to go in that department than a squamatan one even if the energetic issue were sorted. (Birds are even worse off in this regard, and indeed there's no avian "snakes" either.)
Still, there's Basilosaurus.
Happy memories of the day I met one at Wolfe Creek meteorite crater in Western Australia.
The ground was scattered with 'spinifex' grass clumps - very prickly. Draped in and through one was this incredibly slender reptile, of a golden colour. Though legless, it was clearly not a snake. About 18 inches long as a guess. As I focussed my camera, it leapt violently forward and disappeared under the spinifex at ground level. No photo, but a wonderful moment. (Then the nature watchers had to silence the uninformed person who knew it was a snake).
Spinifex is the incorrect common name of large painfully spiky grass tussockss - haven't any reference for its correct name just now.
The 'Spinifex' is correctly called Triodia. It is indeed a wonderfull habitat for pygopodids (and many other squamates besides). Although it can kill the Triodia by snapping its roots, lifting up the edges of Triodia hummocks early in the morning and scratching through the leaf litter and sand underneath is a great way to find all sorts of herpetological goodies.
[from Darren: sorry, delayed by spam filter.]
Re Alan Kellogg (#37) and Andreas Johansson (#38)--
My personal suspicion (since I sort of like looking for plausible things thatt look like orthogenesis) is that the basic mammalian bauplan is pre-DISadapted to snakiness. Mammaliabn respiration involves a diaphragm, located a very strongly marked boundary between the thoracic and lumbar parts of the vertebral column. Going from that condition to one with an extremely elongated, structurally fairly uniform, "torso" may be too much of a leap for natural selection to make easily.
There also seem to be powerful (developmental?) constraints on the evolution of vertebral column novelties in many mammalian clades: witness the thing about almost all mammal species having the same number of cervical vertebrae, and the fact that (at least in Boreutherians - Afrotherians and Xenarthrans seem to be more liberal) there is surprisingly little variation in the thoracic+lumbar count.
Basilosaurus... I think a lot of the elongation of the vertebral column in Cetaceans is in the caudal region, related to use of the tail as a propulsive organ. I suppose the considerations of my two previous paragraphs wouldn't rule out the development of "pseudosnakes" in which most of the length is tail (as in some Pygopodids, but in contrast to snakes). ... Cetaceans also tend to be big animals, so they may not be as bothered by energetic considerations as a small mammalian pseudosnake would be: the idea of a small mammal whose muscular tail had greater mass than the part of the body containing the lungs, heart and digestive tract seems weird.
Some Mustelids are long and narrow, but if you watch, say, a ferret(*) you'll see that they don't move in at all snakelike ways. They often walk with the back strongly arched, bringing the fore- and hindlegs as close together asw they would be in a much shorter-bodied animal, and coiling and writhing don't seem to play any role in normal locomotion.
(*) My opportunity to watch them was at the home of a friend who was in the "Ferret Rescue Society" in Melbourne: owners of pet ferrets seem regularly to abandon them in parks, where a house-raised ferret doesn't know how to make a living. People find them, drop them off at the local veterinary clinic, which calls somone like my friend to sponsor their shots and de-sexing, take them home to redomesticate them, and try to find good homes for them. I'm not sure what I think about humans: does the dedication of some animal lovers balance out the irresponsibility of some pet-owners?
Indeed, weasels have twice the metabolism of normal-shaped mammals their size, just to keep up with the heat loss.
I have no idea as to when the diaphragm appeared, seymouria perhaps?
Re: Alan Kellogg (#42)
You start to get very strong demarcation between thoracic and lumbar regions in advanced therapsids (cynodonts), suggesting that the mammalian style of respiration with major involvement of the diaphragm starts around then. Earlier synapsids (pelycosaur and early therapsid grades) have at best a gradual transition (from vertebrae with full-length ribs to vertebrae with slightly shorter ribs), suggesting that there wasn't a developed, functionally important, structure there.
...which, however, wasn't really functionally 'snake-like' at all. It didn't swim like a snake; it swam - as Allen suggests - basically in the same way that modern cetaceans do.
(In fact, their long tails look to me like a great argument for "nectridean" paraphyly. But I digress.)
A difference between thoracic and lumbar vertebrae (the latter having very short or no ribs) evolved a few times independently.
The seymouriamorphs are not amniotes; it's possible that they aren't even crown-group tetrapods -- in other words, that amniotes and amphibians are more closely related to each other than (to) seymouriamorphs.