The topic I’d like to discuss today represents a major controversy in our understanding of the phylogeny (family tree) of reptiles (and birds, which are just a fancy kind of reptile). Turtles have been in a state of identity crisis for some time as scientists have debated the origins of our shelled friends, and two recent papers in the journal Biology Letters, have re-opened this confusing evolutionary can of worms, possibly resolving the conflict. One study published in May (Crawford et al. 2012 1) considers turtles to be related to the bird and crocodile group Archosauria, whereas the other published in July found evidence for them being close to the Lepidosaurians; the lizards, tuatara, snakes and worm lizards (Lyson et al. 2012 2).
Before I go on, I should explain a little bit about how we have traditionally grouped vertebrates (animals with a backbone) by the characteristics of their skull. Take a quick moment to massage your temples – notice how there’s a soft spot there? That’s what biologists refer to as a temporal opening, and we have traditionally used the number of these openings to group vertebrates by their skull morphology into evolutionary lineages. Traditionally (according to my 2nd year animal diversity textbook, now hideously outdated at the grand old age of 12 years, Hickman et al., 20003), we have three groups:
Anapsids = “no openings”. These guys have no openings at their temples – just a solid sheet of bone running from behind their eyes to the back of their skull. The turtles, tortoises and terrapins (all the same thing really according to evolutionary biology) were traditionally considered the only living animals in this group.
Diapsids = two openings”. Traditionally, animals in this group are split into two types, the Lepidosaurians (tuatara, amphibaenids (worm lizards), and the lizards and snakes) and the Archosauria (crocodiles and birds). These guys are characterised by having two holes – one at the temples and another just behind that, separated by a bony arch.
Synapsids = “fused arch”. Today, this group includes us humans, and other mammals. We only have one opening, and that’s the one you can feel at your temples 4.
Clear as mud? Great!
So, back to turtles… traditional morphology (analysing how a creature looks and develops) considered turtles to be Anapsids, and thought that only having one temporal opening was an ancestral trait, with the two temporal openings developing later on in the non-turtle reptiles and birds. However, others placed turtles in the Diapsida, considering two openings to be the ancestral trait with the two openings fusing later on in the turtle lineage to form no openings. This second conclusion found support after the advent of molecular ecology (using DNA and other molecules like RNA and proteins to unravel the complex web of evolutionary relationships) – but it placed turtles in or near the crocodile and bird group. Which meant that the poor confused turtles (and the scientists who studied such things) were in one of three opinion camps (1) turtles are the sister to all other reptiles and birds, the single living group out on the Anapsidan limb at the base of the family tree (a fancy reptile/bird), (2) a Diapsid sister to the Leipdosaurians (a fancy lizard, snake or tuatara-like critter) or (3) a Diapsid sister to the Archosaurians, or an Archosaurian in their own right (a fancy crocodile or bird-like critter).
Lyson et al. (2012) used a new way of looking at the problem: they decided to analyse 282 microRNA genes from a Californian turtle – the western painted turtle (Chrysemys picta bellii) to see how they compared with the microRNAs of a Carolina anole (that’s a type of lizard, btw), an alligator, a chicken and a zebra finch, along with those from three mammals: a platypus, an opossum, a human and an African clawed frog as outgroups. They found support for camp #2: turtles are most closely related to Lepidosaurians and should be considered a sister group of the lizards. This was because the turtle and the anole shared four families of microRNAs which were not found in any of the other organisms studied. They propose a new group that contains Lepidosaurs and Testudines – the ‘Ankylopoda’, meaning ‘fused foot’, which refers to the fused ankle bones in both these groups.
But… Crawford et al. (2012) followed this assertion up by looking at DNA on a seriously MASSIVE scale {evolutionary biologists can do this nowadays due to some really whizz-bang technological breakthroughs in the machines, computers and software that we use to do genetic studies; tagged ‘high-throughput phylogenomics’, which basically means that staggeringly enormous numbers of genes can be sequenced and analysed within quite a reasonable time-span (i.e. less than one seriously overworked lifetime)}. Crawford & folks’ dataset compared 1145 different ultraconserved mitochondrial and nuclear DNA elements from right across the genomes of African helmeted turtles, American alligators, a Carolina anole, a corn snake, a human, a painted turtle, a chicken, a saltwater croc, a tuatara and a zebra finch. And the resulting phylogenetic family tree clearly supported camp #3 – turtles are the sister group to crocodiles and birds.
Thus the frustration of molecular evolutionary studies – different molecules will tell us different things… which one is right? Crawford et al. (2012) reckon the problem with using microRNA comes from the current lack of this data being available for reptiles. Lyson et al. 2012 got their microRNA data from small libraries which are likely biased by some microRNAs being easier to detect in some species than in others. So, the proof of the pudding will lie in answering this tricky question: are these four microRNA families truly absent from crocodiles and birds, or is it just that Lyson et al. (2012) weren’t able to detect them?
Another problem that Crawford & co. brought up is that Lyson & co. didn’t look at tuatara microRNA – which is rather important given that their lineage became distinct the earliest out of all the Lepidosaurians (one of the reasons why we call tuatara “living fossils”, the other is that modern tuatara look a heck of a lot like their great-(to-the-power of thousands!)-granddaddies). So, we could consider Lyson et al.’s study, while very interesting, somewhat incomplete.
At least the point on which both these studies agree is that turtles are indeed a Diapsid that has lost the double temporal openings. So, in crude terms, we can think of turtle skulls as a sort of ‘simplification’ of the ‘more complex’ double-holed skull. A really common misunderstanding of evolutionary biology is that things should get more complicated as evolutionary time progresses (with us humans haughtily sitting at the top of the complexity tree – I recently saw a street sign which encapsulates this – “If evolution is real, how come we still have apes?”). Turtles, then, are another great example of the course of evolution not running smoothly (you can definitely say the same for the course of evolutionary research, too!). Body plans can change in a convoluted pathway with creatures becoming more complex or more simple over evolutionary time (e.g. that degenerate barnacle, Sacculina carcini, is a great example of simplification over evolutionary time 5,6), more complex then more simple, more simple then more complex… or even not change very much at all like our friend the tuatara. The permutations are many so it pays to keep a broad mind when considering evolutionary pathways.
So, for the present I’m going to roll out my sleeping bag in one of camp #3’s tents – and accept turtles as the shelled sister of birds and crocs – based on the seemingly more convincing evidence of Crawford et al. (2012). That being said, I’m really looking forward to seeing future developments of this debate and to (possibly) shifting camps many times until a consensus is reached and the turtle identity crisis is finally resolved.
References, links & footnotes:
- Crawford N.G., Faircloth B.C., McCormack J.E., Brumfield R.T., Winker K. & Glenn T.C. (2012) More than 1000 ultraconserved elements provide evidence that turtles are the sister group of archosaurs. Biology Letters 8: 783-786.
- Lyson T.R., Sperling E.A., Heimberg A.M., Gaulthier J.A., King B.L., & Peterson K.J. (2012) MicroRNAs support a turtle + lizard clade. Biology Letters 8: 104-107.
- Hickman C.P. jr., Roberts L.S. & Larson A. (2000) Animal Diversity. 2nd edition. McGraw-Hill, Boston, USA.
- This information is a mix from Hickman et al. (2000) but is also inspired by the nice explanations of Wikipedia for ‘Anapsida’, ‘Synapsida’ and ‘Diapsida’… have a look!
- Sacculina carcini got a mention on Susan Perkins’ and Tommy Leung’s Parasite of the Day blog: http://dailyparasite.blogspot.co.nz/2010/01/january-7-sacculina-carcini.html
- Zimmer, C. (2001) Parasite Rex: Inside the Bizarre World of Nature’s Most Dangerous Creatures. Simon & Schuster, New York, USA. NB. There’s a fantastic excerpt in here about the scorn that Victorian biologists felt for this parasitic barnacle for daring to become a simplified version of its ancestors and choose a bountiful existence feeding off, and controlling the brain of, their crab hosts. Scandalous!
SCritterBlog 