Skip to main content

Amphisbaenians and the origins of mammals

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Among the most controversial and remarkable of living tetrapods are the bizarre amphisbaenians: a group of fossorial, long-bodied carnivorous animals with reduced or absent limbs, spade-shaped or bullet-shaped skulls strongly modified for burrowing, and an annulated body where distinct, regularly arranged transverse segments give the animals a worm-like appearance. [Adjacent image combines diagram from Gans (1974) with photograph of Amphisbaena bakeri by Father Alejandro Sanchez, used with permission]. Until recently it was generally thought that amphisbaenians are reptiles, and part of Squamata (the reptile group that includes snakes and lizards). But, in a fascinating case of multi-disciplinary co-operation involving genetics, neophenetics, and good old-fashioned critical thinking, intuition and balls, a daring group of feisty young zoologists have challenged the old guard of the ‘Mammals are Derived Synapsids, y’all’ (or MADsy) school of thought, and have demonstrated that these are not mere squirmy reptiles. They are, in fact, the true ancestors of mammals.

The first authors to formally suggest a non-reptilian affinity for amphisbaenians published their observations in the 1940s and 50s (Zangerl 1945, Kesteven 1957). In a classic case of textbook orthodoxy triumphing over the brilliantly shining light of massive truthfulness, their work was all but ignored in what amounts to a conspiracy of some sort, and the more media-friendly idea that mammals descended from Palaeozoic synapsids like Dimetrodon took centre stage (e.g., Remor 1979, 1980). However, animals like Dimetrodon clearly come loaded with too much baggage to serve as mammal ancestors! They are way too big (Dimetrodon was about the size of a man!), and the two or three characters they share with mammals must have arisen by way of similar lifestyles! We are therefore left searching for the true mammal ancestor! And this is where amphisbaenians come in!!


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


After being all but ignored in the literature on mammal origins for several decades, an impressive list of unassailable genetic papers (Koch & Wyman 2007 and references therein) have now shown that mammals are actually nested within the amphisbaenian radiation (Koch & Wyman 2007 and references therein). [Adjacent photo by Diogo B. Provete, from wikipedia.] Initially this might sound remarkable but, after thinking about it a bit, a number of workers have reassessed some of the older literature – much of which shows that the character evidence linking mammals to amphisbaenians is compelling, well documented, well established, meticulously detailed, and deserving of other such terms (Koch & Wyman 2007 and references therein). We don’t need to worry about looking critically at the genetic data (Koch & Wyman 2007 and references therein), as few people really understand genetic stuff anyway, and no one really reads the methods sections of papers anymore (Koch & Wyman 2007 and references therein).

The evidence from behaviour

Relatively little is known about amphisbaenian biology and behaviour. What we do know, however, shows that these animals share more in common with mammals than with other amniotes. In one of the few taxa for which a large amount of data on population structure has been collected, Bipes (the Mexican limbed amphisbaenians), Papenfuss (1982) concluded that the animals were K-selected, exhibiting delayed maturity, small clutch size, and non-annual reproduction. Does this sound like typical reptile behaviour to you? Of course not, it is part of the overwhelming body of evidence linking amphisbaenians to mammals.

Among mammals, an incredible plethora of tunnel-excavating habits continue to be practised even by more recently evolved lineages, and are manifested by such activities as trench warfare, the Channel Tunnel, the London underground, the salt-mining culture of the elephants of Mount Elgon, the propensity for humans to dig tunnels on sandy beaches, and that dude who lived in a hole for years. Is it a coincidence that people seek out caves and tunnels to explore and wonder at while on holiday? Many people admit to psychiatrists that they dream of burrowing and tunnelling. And let us not forget that millions of people travel to and from work on a daily basis via subterranean tunnels, showing a statistical preference for this mode of travel rather than for the supra-terrestrial environment shunned by our amphisbaenian ancestors. In all this data, we have irrefutable evidence pointing to our subterranean, amphisbaenian ancestry.

The evidence from soft-tissue anatomy

In their often pinkish colour, amphisbaenians share an important synapomorphy with mammals, most of which are pinkish when shaved. This character is best expressed in Bipes, and in the desert sharks and allghoi-khorkhoi (more on these taxa in a minute), but it is also widespread within amphisbaenid amphisbaenians. Some amphisbaenian taxa are capable of autotomy (the defensive shedding of the tail when it’s grabbed by a predator), another similarity shared with mammals. An often overlooked fact concerning the amphisbaenian body is that, despite the simple, worm-like shape, these animals have a highly complex musculature, as was noted by Camp (1923) and Gans (1978). It has often been stated that mammals owe their complex musculature to the fact they went through a small-bodied cryptic phase, in which they had to clamber over stones, roots, dinosaur toes etc., but in reality mammalian musculature has been derived directly from the condition present in amphisbaenians.

A loose skin that is mobile relative to the underlying tissues is also shared by amphisbaenians and fossorial mammals like naked mole-rats Heterocephalus glaber and pet hamsters. Indeed, amphisbaenian-like Heterocephalus [shown in adjacent image] is one of the most basal of mammals, its poikilothermic physiology (Buffenstein & Yahav 1991) demonstrating that poikilothermy was retained by ancestral mammals and only later modified as mammals began to take to the supra-terrestrial environment. All of this is depicted in the phylogram shown below - truly, unassailable science (sensu Olson 2002).

The evidence from skeletal anatomy

The short, rigid skulls of amphisbaenians are shockingly mammal-like. Indeed Carl Gans, the great expert on amphisbaenians, has repeatedly noted the strong similarity evident between amphisbaenian and mammal skulls (Gans 1969, 1974), drawing special attention to this in his 1974 volume Biomechanics: An Approach to Vertebrate Biology. Here, Gans noted the strong similarity between the skull of the trogonophid Agamodon and that of a cat (Gans 1974, p. 134, Fig. 4-10), describing how "a comparison between the skull of a cat and that of the trogonophid Agamodon documents the … similarity which … [shows] that the skulls of amphisbaenians most [resemble] those of mammals". Later, Gans showed how the jaw motion of Agamodon is strikingly like that of elephants and other placental mammals (Gans 1974, p. 178, Fig. 4-40: the diagram shown at the top of the article).

The cranial bones of amphibaenians overlap one another extensively, meeting at strongly interdigitated sutures, again as they do in mammals. A whopping big coronoid process on the amphisbaenian mandible is again a shared character present in mammals, though admittedly this character is labile and has coincidentally popped up in various reptile lineages. Labouring under the mistaken assumption of a reptilian identity for amphisbaenians, zoologists have interpreted one of the most unusual features of the amphisbaenian skull – the extracolumella, a structure that links the side of the jaw to the ear – as a character of reptilian flavour, and perhaps as a unique evolutionary solution to the problem of sound conduction. However, Wever & Gans (1973) noted that the extracolumella is "not homologous to the structure of the same name in lizards" (p. 189), and Gans (1978) reiterated that it "is not directly homologous with the structure of that name seen in the Sauria and Sphenodon" (p. 377).

Given that the ear bones of mammals are really complicated and are known to have somehow been modified from bits and pieces that used to be at the back of the lower jaw, it is unarguably logical to conclude that the freaky stuff going on in the amphisbaenian lower jaw and ear region offers us the real answer to mammalian ear evolution. By employing OOA (obfuscatory ocular analysis, or squinting), Ratsarse (2006) was able to show that amphisbaenian skulls are highly similar to those of mammals, the extracolumella looking like some sort of magic streak showing the viewer where all the funky evolutionary pazazz is occurring, especially when it’s highlighted in red as shown here. [Adjacent image of Geocalamus acutus from here on digimorph.]

Gans (1960, 1974) noted that, by depicting amphisbaenian skulls within transformation grids, the long and low, superficially squamate-like skulls of trogonophid amphisbaenians could be ‘morphed’ into the short, mammal-like skulls of taxa like Agamodon (the latter one of the key taxa demonstrating the true affinities of the Amphisbaenia). In fact, Gans simply got it the wrong way round, as we now know that the mammal-like Agamodon skull is primitive, and it is the squamate-like skull of Trogonophis and similar taxa that are derived. Any idiot can see this, it’s soooooooo obvious.

Another important feature that amphisbaenians share with mammals is the mysterious reduction of the pelvis. Whereas animals with reduced limbs typically lose their forelimbs and retain their hindlimbs (look at pythons and reduced-limbs lizards like some cordylids), some amphisbaenians retain forelimbs and a pectoral girdle, yet have completely lost the hindlimbs and only possess a relictual pelvis consisting of the ilium alone (Kearney 2002). Where else do we see this unusual pattern? In reptiles? Heavens, no, in mammals: just look at whales! Again this is an important shared character demonstrating the amphisbaenian heritage of Mammalia. This self-evident evidence is so evident that it has been ignored by many so-called experts; they have brushed it under the carpet, donned blinkers, thrown up their arms, and uttered ‘harrumph!’. In short, anything other than consider the obvious TRUTH.

A transition lost in time, ignored by orthodoxy

Elsewhere in classic zoological literature, we can see that Dixon (1981) knew full well that amphisbaenians gave rise to mammals, and that the earliest mammals were fossorial amphisbaenian-like creatures: an assertion demonstrated by the existence within Dixon’s work of the desert sharks Psammonarus, sausage-shaped predators that were perfect evolutionary intermediates between the Amphisbaenia and Mammalia.

While at least some of the creatures described by Dixon (1981) are generally accepted as hypothetical, recent work has demonstrated that a surprising number of genuine creatures were included as well. We can assume that, eschewing the hidebound, tediously slow, reactionary work of the peer-reviewed journal system, Dixon snuck genuine zoological discoveries into his work, knowing full well that smart people would spot them for the real creatures they obviously are. The swimming ‘monkey’ Natopithecus ranapes Dixon, 1981, for example, has since been shown to be a pre-emptive description of the aquatic primates later described by Coleman & Huyghe (1999), the Flooer Florifacies mirabila was an accurate description of the flower-faced snouters Cephalanthus described by Stümpke (1957)*, while the insectivorous tree drummers (Proboscicuncus) were evidently based on somewhat garbled descriptions of the remarkable Peruvian murid Rhagomys longilingua (Luna & Patterson 2002).

* Unfortunately, Dixon was unaware of Stümpke’s prior description of this form (D. Dixon, pers. comm.). Rhinogradentians were covered on Tet Zoo ver 2 here and here.

We can therefore assume that Dixon learnt of Psammonarus at one of the many palaeontology conferences he attends. Admittedly, no published abstract or article attests to the description of such a creature, and we are forced to conclude that it was deleted from the place of publication and its authors assassinated; possibly, those in control of the MADsy conspiracy somehow travelled back in time and erased Psammonarus from the technical literature, as this is the sort of thing they do.

Evidently closely related to Psammonarus, and even closer to Mammalia than are other amphisbaenians, is the specialised relict form of Mongolia and Kazakhstan, the Allghoi-khorkhoi or Mongolian death-worm Dinovermis mackerlei, a controversial animal suggested to be mythical or semi-fictional by some, but known to be real by those who care. Originally suggested to be an annelid or burrowing snake (Shuker 2003), the truncated tail, pinkish hue, annulated body, amphisbaenian-style demeanour, and DNA of the Allghoi-khorkhoi have since demonstrated its amphisbaenian identity (Koch & Wyman 2007 and references therein).

Dinovermis is of course not called the death-worm for nothing. It is venomous, and here we find the key synapomorphy linking this animal with mammals. Not only are living primitive mammals – like monotremes, shrews and solenodons – venomous, we now know that venomosity was widespread in early mammals (Hurum et al. 2006). Only in the light of the amphisbaenian origin model does mammalian venomosity make any sort of sense.

Dinovermis is also electrogenic (that is, able to generate an electric field), a fact unnervingly similar to the fact that electroreceptive abilities are present in those primitive mammals, the monotremes (and perhaps also in moles and other mammals). Electrogenic fish usually have electroreceptive abilities, so we can safely infer that electrogenic and electroreceptive abilities were primitive for mammals, but that the electrogenic ability was mostly lost. It wasn’t entirely lost, as some African people think that the Giant otter-shrew Potamogale velox is electrogenic. This will doubtless prove correct, and provide yet more affirmation of the amphisbaenian origin model.

Given that textbook dogma stakes matter-of-fact that mammals supposedly evolved somehow from quadrupedal synapsids of the Palaeozoic, we have to wonder why so many zoologists have fooled themselves into relying on cold, stony fossils, and not on the obvious living evidence that we can glean from the creatures around us today. We have learnt that we need not concern ourselves with Sineoamphisbaena from Upper Cretaceous Mongolia, and other fossils allegedly linked to the amphisbaenians. You might be surprised to hear this coming from a palaeontologist but, well, that’s just how it is. So fossils are vastly over-rated for this sort of thing and can be safely ignored. In fact, they may as well not exist.

The presence within amphisbaenians of retractile hemipenes, a squamate-style kidney, a transerve cloacal slit, keratinised scales, a telencephalic roof divided into three cortices, and a squamate-like Jacobson’s organ, are all clearly convergences with squamates. Some scientists, probably looking down from ivory towers or hiding behind the thick hedges of ivy that covers the walls of their colleges, and generally hoping to maintain the MADsy model, continue to deny this, as they have for a while now. But, like Saruman in the second Lord of the Rings movie, it is only a matter of time before the ents of justice arrive, and demonstrate the true, amphisbaenian origin of Mammalia. Then we will be free!

Apologies to long-term readers - have had no time to prepare anything new.

Refs - -

Buffenstein, R. & Yahav, S. 1991. Is the naked mole-rat, Heterocephalus glaber, an endothermic yet poikilothermic mammal? Journal of Thermal Biology 16, 227–232.

Camp, C. L. (1923). Classification of the lizards Bulletin of the American Museum of Natural History, 48, 289-481

Coleman, L. & Huyghe, P. 1999. The Field Guide to Bigfoot, Yeti, and Other Mystery Primates Worldwide. Avon Books, New York.

Dixon, D. 1981. After Man: A Zoology of the Future. Granada, London.

Gans, C. 1969. Amphisbaenians – reptiles specialized for a burrowing existence. Endeavour 28, 146-151.

- . 1974. Biomechanics: An Approach to Vertebrate Biology. J. B. Lippincott Company, Philadelphia, Toronto.

- . 1978. The characteristics and affinities of the Amphisbaenia. Transactions of the Zoological Society of London 34, 347-416.

Hurum, J. H., Luo, Z.-X. & Kielan-Jaworowska, Z. 2006. Were mammals originally venomous? Acta Palaeontologica Polonica 51, 1-11.

Kearney, M. 2002. Appendicular skeleton in amphisbaenians (Reptilia: Squamata). Copeia 2002, 719-738.

Kesteven, H. L. 1957. Notes on the skull and cephalic muscles of Amphisbaenia. Proceedings of the Linnean Society of New South Wales 82, 109-116.

Koch, A. & Wyman, J. 2007. Limblessness in amniotes shared by worm-lizards and basal mammals: the data from ribosomal ISBN and ILN genes. Yay Genes, Today! 1, 20-31.

Luna, L. & Patterson, B. D. 2003. A remarkable new mouse (Muridae: Sigmodontinae) from southeastern Peru: with comments on the affinities of Rhagomys rufescens (Thomas, 1886). Fieldiana: Zoology, New Series 101, 1-24.

Olson, S. L. 2002. Review of: New Perspectives on the Origin and Early Evolution of Birds. Auk 119, 1202-1204.

Papenfuss, T. J. (1982). The ecology and systematics of the amphisbaenian genus Bipes Occasional Papers of the California Academy of Sciences, 136, 1-42

Ratsarse, M. 2006. OOA as applied to the vexing problem of mammal baramins. Creation Science ‘Journal’ 3, 2-3.

Remor, S. A. 1979. Why Dimetrodon is just way cooler than an amphisbaenian as a mammal ancestor. Big Science 101, 7-10.

- . 1980. Amphisbaenia: you blow! Great Plain Science Musings 75, 556-566.

Shuker, K. P. N. 2003. The Beasts That Hide From Man. Paraview Press, New York.

Stümpke, H. 1967. The Snouters: Form and Life of the Rhinogrades. The Natural History Press, Garden City, New York.

Wever, E. G. & Gans, C. 1973. The ear in Amphisbaenia (Reptilia); further anatomical observations. Journal of Zoology 171, 189-206.

Zangerl, R. 1945. Contributions to the osteology of the post-cranial skeleton of the Amphisbaenidae. American Midland Naturalist 33, 764-780.

Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at darrennaish.wordpress.com. He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at tetzoo.com!

More by Darren Naish