ADVERTISEMENT
  About the SA Blog Network













Tetrapod Zoology

Tetrapod Zoology


Amphibians, reptiles, birds and mammals - living and extinct
Tetrapod Zoology Home

A very quick history of turtles

The views expressed are those of the author and are not necessarily those of Scientific American.


Email   PrintPrint



Darren Naish About the Author: 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!

Nature Blog Network

Follow on Twitter @TetZoo.

The views expressed are those of the author and are not necessarily those of Scientific American.





Rights & Permissions

Comments 49 Comments

Add Comment
  1. 1. BrianL 12:29 pm 11/17/2012

    The recent extinction of meilonaiids certainly brings home that, in the end, crown groups are rather arbritary concepts. Had New Caledonia’s meiolaniids survived to the present, or even the near-present, certainly they would by definition have been included in Testudines (as in crown turtles) simply on account of being recent? In fact, wouldn’t those last meilonaiids have been recent or ‘recent’ anyway given that they died out about 2000 years ago?

    I realise they simply fall outside the clade of modern turtles, but to me it does seem rather weird that when it comes down to it, meilonaiids are considered stem-turtles because humans happened to kill them off before western science discovered them and otherwise we would have been obliged to call them crown group turtles. Surely, if modern age scientists had discovered and described living meiolaniids, they would not have considered them to be anything but proper turtles?

    Is there any other crown clade that provides a comparabe case in which stem-relatives are considered to have only died out during the Holocene? I can’t think of any off the top of my head.

    Link to this
  2. 2. naishd 12:50 pm 11/17/2012

    Yes, I made exactly the same point in a lecture the other day… humans encountered (and even killed and ate) meiolaniids, so we in the 21st Century ‘only just’ missed them. If they were still around, the concept of the turtle crown would be more inclusive. Exactly this point was made by… Spencer Lucas, I think. I’ll try and find the paper later on.

    Darren

    Link to this
  3. 3. Michał 1:07 pm 11/17/2012

    meiolaniids (…) are stem-turtles

    Though Eugene Gaffney would probably disagree with that (or did he change his mind about that since his paper on Kayentachelys from 2010?)

    Link to this
  4. 4. Jurassosaurus 1:07 pm 11/17/2012

    I remember early analyses of meiolaniids (prior to Gaffney’s work) pulling them outside of Testudines. Those old analyses much like this new placement still seem to suffer from the same problem. Namely that it pushes meiolaniids back into the Early Jurassic, giving them a ghost lineage that is ~120 million years old. Maybe it’s true, but as far as I know we currently lack fossils to support anything more than a Mid-Late Cretaceous origin. This leaves me very skeptical of the placement. That said, I don’t know how accurate turtle collecting and associations have been in field paleo. I wouldn’t be surprised if there were a couple of broken shell fragments that might wind up being associated with meiolaniids at some point in the future. For now though I remain skeptical of any placement that results in long ghost lineages.

    Link to this
  5. 5. naishd 2:18 pm 11/17/2012

    So far as I know, yes, meiolaniids do still have that long ghost lineage, though with several Lower Cretaceous taxa mooted as meiolaniid relatives (Spoochelys, Naomichelys, Otwayemys, Chubutemys). Meiolaniids do seem to look like stem-turtles, and maybe I’m not as sceptical of long ghost lineages as you are.

    Darren

    Link to this
  6. 6. Jurassosaurus 2:51 pm 11/17/2012

    So far as I know, yes, meiolaniids do still have that long ghost lineage, though with several Lower Cretaceous taxa mooted as meiolaniid relatives (Spoochelys, Naomichelys, Otwayemys, Chubutemys).

    Heh, this was one of those moments where I had to remind myself the difference between the British moot and the American moot.

    Link to this
  7. 7. David Marjanović 6:51 pm 11/17/2012

    The Jurassic nonmarine turtle record isn’t good. Plenty of room for ghost lineages to hide in.

    Yes, Gaffney still disagrees; and he still seems to make scenario-based arguments. Phylogenetic analyses can test evolutionary scenarios, not the other way around.

    Tyler Lyson has an unpublished surprise hypothesis about the cleithrum.

    Link to this
  8. 8. John Harshman 7:06 pm 11/17/2012

    Once again, you avoid discussion of turtle relationships. Previously your ploy was to omit turtles from trees. This time you reverse field (a football metaphor (that’s American football)) by omitting everything except turtles. For shame!

    Link to this
  9. 9. vdinets 7:12 pm 11/17/2012

    If meilonaiids are really of Jurassic origin, it would make them the oldest lineage extinguished by humans. Or am I forgetting something?

    Link to this
  10. 10. vdinets 7:14 pm 11/17/2012

    meiolaniids, sorry.

    Link to this
  11. 11. Metridia 7:57 pm 11/17/2012

    BrianL, you seem to be suggesting that stem-group animals are stem merely because they are primitive and extinct, compared to a more diverse, monophyletic, extant radiation. I agree that meiolaniids may be outside of the crown-group as of now, but it would seem there’s no obligation to include them in the crown-group if they had survived. Aren’t there many examples of crown groups that exclude extant relatives?

    Link to this
  12. 12. John Harshman 8:33 pm 11/17/2012

    Metridia,

    All crown groups exclude extant relatives, except for the crown group of life. Crown group birds exclude extant crocodylians, for example. Meiolaniids almost certainly would have been included in the crown group of turtles if they had survived, because they looked like turtles. Monotremes are included in crown group mammals, because they look like mammals. It’s possible that dicynodonts, had they survived, would have been excluded, but if they were included we would have a different concept of what “mammal” means (and a quite different word).

    Link to this
  13. 13. naishd 6:01 am 11/18/2012

    The very definition of crown group = the clade subtended by extant species. So, extant members of a clade, by definition, are crown-group members. A hypothetical extant dicynodont might always have been considered distinct enough from mammals to be regarded as a near-mammal but still a non-mammal (perhaps in the same way that crocs are near-birds or tuatara are near-squamates), but, then again…

    It’s conceivable that several crown groups would be more inclusive phylogenetically had certain Pleistocene/Holocene extinctions not occurred. Mekosuchines are already within the Crocodylian crown group, but their survival would make crown-Crocodyloidea more inclusive. The persistence of Mammut the mastodon to the present would also make crown-Proboscidea hugely more inclusive. Likewise for crown-Cingulata (with xenarthrans) had glyptodonts and pampatheres not been killed off in the early Holocene. And so on…

    Darren

    Link to this
  14. 14. naishd 6:09 am 11/18/2012

    As for turtle origins (comment 8), I did devote several slides to it in the lecture concerned. Just wanted to concentrate on turtle in-group relations here. We’re still a fair way from a consensus on turtle origins. I tell students that molecular results (mostly placing turtles within Diapsida, and near archosaurs) are consistent with the view that the anapsid condition of turtles is heavily modified and not ancestral – a view that morphologists have promoted for a long time, incidentally. Then there’s the laterosphenoid recently identified in Proganochelys and the fact that Odontochelys lacks dorsal osteoderms and a carapace (features which count against at least one or two of the parareptile hypotheses that have been proposed). However, parareptile Eunotosaurus is still looking proto-turtle-ish, and new data is showing some new, surprisingly turtle-like aspects of its anatomy.

    Should devote a proper article to this when time allows. Time definitely does not allow, right now.

    Darren

    Link to this
  15. 15. David Marjanović 11:23 am 11/18/2012

    The crown group of a clade consists, by definition, of the last common ancestor of all extant members of that clade plus all descendants of that ancestor. It has absolutely nothing to do with “primitive”.

    And yes, this means that when certain people talk about “stem-sauropterygians” or “stem-stereospondyls”, they’re wrong; the stem group is everything except the crown group, and extinct clades by definition lack crown groups. That’s why it took me years to figure out that Rieppel’s use of “stem-sauropterygians” implicitly assumes as the corresponding crown group the clade that would have been the crown group in the Jurassic.

    Odontochelys lacks the peripherals and the pygal plate, but all other dorsal osteoderms (the costals on top of the ribs and the neurals on top of the vertebrae) are there. I also think the carapace has simply fallen apart at the sutures, as opposed to not existing in the first place.

    Link to this
  16. 16. John Harshman 12:00 pm 11/18/2012

    We should point out that any two extant species define a crown group (assuming that evolution is nicely treelike, which is not always the case). Some crown groups are named, some aren’t. The decision of which crown groups to give particular names to is fairly arbitrary, but certainly matching common notions of what, for example, a turtle ought to be has some influence. And those notions are formed with reference mostly to extant species. There’s a big morphological gap between extant turtles and non-turtles, most of which would be intact if there were a living Meiolanius. Likewise, there’s a big gap between extant mammals and non-mammals, and the discovery of monotremes did only a small amount of damage to that gap. A living dicynodont would do rather more damage. We would still have a group we could call Mammalia, but we would also have a separate crown-Therapsida, and we might have considered that group the more important one when assigning such things as Linnean classes; Mammalia might have been a sub-class in such a case. Or not. It’s all highly subjective.

    Link to this
  17. 17. naishd 12:36 pm 11/18/2012

    Odontochelys (see comment 15): David, I’m confused. Li et al. specifically say that osteoderms are absent in Odontochelys, and the expanded ribs are not expanded enough to form costal bones. I can’t see that a carapace could be present.

    Darren

    Link to this
  18. 18. Heteromeles 12:55 pm 11/18/2012

    So, tying into other posts, how does genital anatomy map onto that crown clade? Inquiring minds want to know.

    Link to this
  19. 19. Ausktribosphenos 2:17 pm 11/18/2012

    I know this dicussion centers on crown group Testudines, but will the subject of Proposed anapsid (procolophonid) links to turtles be touched upon? That would be awesome

    Link to this
  20. 20. John Scanlon FCD 2:23 am 11/19/2012

    With Spoochelys in the lower K (Albian), the meiolaniid ghost lineage is a bit shorter than suggested above.

    Entire alien civilizations could easily be hidden in Australia’s Jurassic fossil record, so the ghost lineage can be as long as it needs to be without stretching credulity.

    Link to this
  21. 21. naishd 4:04 am 11/19/2012

    Note mention of Spoochelys in comment 5!

    Darren

    Link to this
  22. 22. John Scanlon FCD 5:03 am 11/19/2012

    Ah yes, I’d noticed the name but not ‘Lower Cretaceous’ in the same place, apologies.

    Link to this
  23. 23. John Harshman 12:46 pm 11/19/2012

    Oddly enough, the most recent dicynodont is also from the Lower Cretaceous of Australia. That’s probably why I thought of that as an example.

    Link to this
  24. 24. naishd 2:58 pm 11/19/2012

    There’s a Tet Zoo ver 1 article on the Cretaceous dicynodont (and a correspoding chapter in Tetrapod Zoology: Book One). I’d link to it, but it’s in such need of update now…

    Darren

    Link to this
  25. 25. Andreas Johansson 3:24 pm 11/19/2012

    When’s the second most recent dicynodont from? Popular books tend to give the impression the non-mammalian therapsids just threw up their front paws and died confronted with dinosaurs in the later Triassic, but of course that’s not entirely true. Do we have anything in the way of Jurassic dicynodonts?

    Link to this
  26. 26. Metridia 5:32 pm 11/19/2012

    Regarding crown groups. @Brian: It’s getting a bit semantic to say yes we exclude relatives from crown groups, which is a truism since life shares a common ancestor. however, often phylogenetics isn’t so clear-cut. Are squamates crown-group lepidosaurs when excluding Sphenodon? Can’t one also say also say ‘crown group’ when we refer to e.g. true pipistrelle bats compared to close relatives or paraphyletically defined pipistrelle bats? Is that done informally or just incorrectly?

    Link to this
  27. 27. Metridia 2:05 pm 11/20/2012

    oops, didn’t mean to address that @ Brian, no way to edit comments here though

    Link to this
  28. 28. naishd 5:26 pm 11/20/2012

    Andreas (comment 25): apart from the Cretaceous fossil, the youngest dicynodonts are from the early Rhaetian (last part of the Late Triassic) of Poland. The fossils concerned were described in 2008.

    Metridia (comment 26): you might have misunderstood. A crown group, by definition, is the clade containing all species that descend from the most recent common ancestor of the living species within the clade in question. With lepidosaurs, Sphenodon is not a squamate: stem-squamates would be those fossil lepidosaurs that are closer to crown-squamates than to Sphenodon, yet are outside the squamate crown. The whole crown vs stem convention is wholly one of convenience, invented so that people only interested in living species wouldn’t have to keep referring to the unusual character conditions (teeth in Mesosozoic birds and turtles, limbs and digits in Mesozoic snakes, etc.) present in species outside the clades that contain all living species. Some people think the idea is useful; others disagree and ignore it.

    Darren

    Link to this
  29. 29. Metridia 7:58 pm 11/20/2012

    @ Darren- Thanks. BTW, I was unclear, I meant to ask whether squamates could be said to be crown-lepidosaurs, as opposed to Sphenodon, which branched off earlier and also does not share the many shared derived features of squamates; not what a stem-squamate would be. But it seems the definition of crown-group is something other than I thought.

    That said, it seems like there ought to be a similar term for a ‘modern’ radiation, defined as the ‘typical’ bauplan within a group including a large number of shared derived features that more deeply branched groups do not all share. this radiation would be as opposed to idiosyncratic groups that may include fewer species, plesiomorphic ‘primitive’ characteristics relative to the ‘modern/typical bauplan’ and branch more deeply relative to species within the modern radiation. Is there a term for this that I’m not remembering…

    Link to this
  30. 30. Metridia 8:02 pm 11/20/2012

    @John Harshman, others: thanks for helpful explanations

    Link to this
  31. 31. David Marjanović 11:13 pm 11/20/2012

    Li et al. specifically say that osteoderms are absent in Odontochelys

    They mean the peripherals (and extra ones that Proganochelys for instance has). They’re counting the costals as mere outgrowths of the ribs, which is sort of defensible; and the sheer weight of tradition seems to have prevented them from recognizing the neurals as osteoderms, even though some of them are disarticulated from their vertebrae in Odontochelys.

    I know this dicussion centers on crown group Testudines, but will the subject of Proposed anapsid (procolophonid) links to turtles be touched upon? That would be awesome

    Eunotosaurus is an “anapsid”, too… and now that we have fresh new data on it, it fits better than the procolophonoids ( = -ids + owenettids).

    I meant to ask whether squamates could be said to be crown-lepidosaurs, as opposed to Sphenodon

    No. Squamates are crown-lepidosaurs, and so is Sphenodon. Sphenodon isn’t a crown-squamate – or any kind of squamate, of course.

    That said, it seems like there ought to be a similar term for a ‘modern’ radiation, defined as the ‘typical’ bauplan within a group including a large number of shared derived features that more deeply branched groups do not all share. this radiation would be as opposed to idiosyncratic groups that may include fewer species, plesiomorphic ‘primitive’ characteristics relative to the ‘modern/typical bauplan’ and branch more deeply relative to species within the modern radiation. Is there a term for this that I’m not remembering…

    There isn’t one, and there shouldn’t be, because its application would be hopelessly subjective – while misleading people into thinking it’s objective because it’s a technical term! Happens a lot. There are still people who believe they quantify biodiversity when they count genera or families.

    Link to this
  32. 32. Allen Hazen 12:17 am 11/21/2012

    Objectivity! That certainly sounds good, like something all scientists should approve of. And crown groups are “objective.” In, I guess, the sense that the definition is in terms of something publicly available, not depending on the opinions or judgments or prejudices of anyone in particular. Sort of why “kilograms” are better than “pounds”: anybody can fill a 10-cm cube with water; they don’t have to seek out an approved pound-weight salesperson.
    Sorry, I don’t think this is an OVERWHELMINGLY important advantage. And the current fashion– I don’t regard the scientific reasons for it as compelling enough to call it anything else but a fashion– for restricting good, memorable, names to crown groups seems to me to be annoying and confusing.

    Case in point: the recent (PNAS published online a couple of days ago) article by Rougier et al. on Necrolestes. Title states the conclusion that N. is “nontherian.” Given the long-established usage of “therian” to denote a broader class than the placental+marsupial crown group, this suggests something really dramatic: is N. perhaps a surviving Docodont? A Haramiyid? A Morganucodont? Well, no: Rougier and colleagues have decided to restrict the good old word “therian” to the crown group: they want to put N. in with the Eupantotheres, just outside the crown group.

    (Fascinating article, b.t.w. — thanks for tweeting an announcement of it, Darren! … And, to give Rougier & co some credit, one of the traits that they have observed — contribution of the anterior lamina of the pretrosal to the braincase wall — is one that older writers (in, say, the 1970s: cf. chapters in “Mesozoic Mammals: the first two thirds of mammalian history” from 1979) thought was diagnostic of the traditional therian/nontherian divide.)

    Link to this
  33. 33. Allen Hazen 12:22 am 11/21/2012

    On another tangent… Re the question of whether Dicynodonts, had they managed to make it up to the present, would be counted as “mammals”: one of the illustrations in Gillian King’s Dicynodont book speculatively restores one species of D. with fur, looking like a ground sloth.

    As far as I know, there is no known evidence as to what sort of integument Dicynodonts had (if someone knows more, please let me know!), and it seems reasonable to suppose that fur came in further down the mammalian “stem” than Morganucodon… But I found the image quite startling.

    Link to this
  34. 34. Andreas Johansson 12:55 am 11/21/2012

    @Darren#28: Thanks. That’s one pretty long ghost lineage down to the Lower Cretaceous.

    Link to this
  35. 35. Dartian 3:44 am 11/21/2012

    Allen Hazen! Where have you been?!

    restricting good, memorable, names to crown groups seems to me to be annoying and confusing

    I certainly sympathise with this sentiment. But I don’t know if there’s really any better alternative to it at the moment. (Having said that, I still groan every time that I think of Thylacosmilus and have to remind myself that I’m supposed to call it a metatherian instead of a marsupial.)

    Link to this
  36. 36. John Harshman 10:29 am 11/21/2012

    That’s one pretty long ghost lineage down to the Lower Cretaceous.

    About the same length as the ghost lineage from Lower Cretaceous to Upper Triassic.

    restricting good, memorable, names to crown groups seems to me to be annoying and confusing

    I have come to prefer it. Perhaps this is a conflict between paleo- and neontologists? On the minus side, a few extinct groups change allegiance. On the plus side, we can characterize crown groups using much more data than we can for any other groups. I also like the convention of putting “pan-” in front of a crown group name to refer to the corresponding branch-based group, which might help you a bit.

    Link to this
  37. 37. naishd 11:46 am 11/21/2012

    The crown-group concept is certainly not disliked by palaeontologists – increasingly, names like Aves, Testudines and Mammalia are used as if restricted to the crowns. As Gauthier and others have said, neontologists outnumber palaeontologists by some margin, so the system of nomenclature that’s most useful to >them< is the one that should get precedence.

    Darren

    Link to this
  38. 38. Andreas Johansson 2:20 pm 11/21/2012

    @John Harshman #26: Er? I was speaking precisely of a ghost lineage from the Upper Triassic to the Lower Cretacous.

    Link to this
  39. 39. Metridia 4:23 pm 11/21/2012

    I suppose the terms ‘sensu lato’ or ‘sensu stricto’ serve the same function of the term I was searching for. Of course, taxonomy is full of subjectivity, as in how species are parceled out into this or that genera and on up. It would merely be a term of art for describing what “ought”, to borrow a word from someone else here, to be included in a given taxonomic group, given how the group has historically been viewed and in the name of doing the least damage justifiable to existing taxonomy; or taking into account branch length relative to the in-group, etc.

    Link to this
  40. 40. John Harshman 8:14 pm 11/21/2012

    I was speaking precisely of a ghost lineage from the Upper Triassic to the Lower Cretacous.

    Oh. I thought you were referring to the living dicynodont scenario. Because Recent to Cretaceous would be “down”, whereas Triassic to Cretaceous would be “up”. Isn’t it fun that the ghost lineages are of similar length?

    Link to this
  41. 41. David Marjanović 10:52 pm 11/21/2012

    Rougier and colleagues have decided to restrict the good old word “therian” to the crown group

    That restriction happened about 20 years ago and has been pretty much universally followed since.

    On the minus side, a few extinct groups change allegiance.

    Whether it’s “a few” depends on the case. For Mammalia it really was just a few. If you try that for Aves, all hell breaks loose. If you try it for Tetrapoda, it’s similar, and if my thesis supervisor and I are right about the tetrapod crown-group, the shit hits the fan!

    Testudines has the immense advantage that restricting it to the crown-group offered the opportunity of separating it from its former synonym Testudinata. I don’t think anybody has complained about this. (…Excluding Alain Dubois, who insists that Chelonii should be used instead of both, because he fervently wishes that priority will be extended to higher ranks in zoological rank-based nomenclature at some point. A few other people based in France use Chelonii. I think he’s the only one who uses Neobatrachii instead of Lissamphibia, though…)

    I suppose the terms ‘sensu lato’ or ‘sensu stricto’ serve the same function of the term I was searching for.

    Often a term has more than two senses.

    Link to this
  42. 42. David Marjanović 10:54 pm 11/21/2012

    My former thesis supervisor, that is. I defended two years ago.

    Sensu lato/stricto is used to distinguish different meanings in which a term has been used by other people.

    Link to this
  43. 43. Metridia 11:24 pm 11/21/2012

    @David-
    “Often a term has more than two senses.”

    Can’t disagree.

    “Sensu lato/stricto is used to distinguish different meanings in which a term has been used by other people.”

    Yes, this is also true. As I said.

    Link to this
  44. 44. Metridia 11:39 pm 11/21/2012

    I wasn’t using it incorrectly. People discuss taxonomy by citing different propositions for taxonomy and nomenclature, while introducing some of their own. All of which again is a partially subjective, partially objective enterprise. Perhaps a definition of a “crown” group that distinguishes only or mostly in the context of extant species would be a little different from “sensu lato/stricto”, but that is also somewhat my point: there isn’t a word that exactly means this.

    @Allen Hazen-thanks for your comment.

    “Objectivity! That certainly sounds good, like something all scientists should approve of. And crown groups are “objective.” In, I guess, the sense that the definition is in terms of something publicly available, not depending on the opinions or judgments or prejudices of anyone in particular. Sorry, I don’t think this is an OVERWHELMINGLY important advantage.”

    Indeed. In part, science is about making evidenced-based arguments (hypotheses, theories, interpretations, etc). Subjectivity is unavoidable, but the practice is to have the evidence and rationale behind hypotheses, theories, interpretations marshaled for all to see.

    Link to this
  45. 45. Chelydra 11:37 am 11/22/2012

    This is the perfect forum for this question: what, exactly, does “terrapin” mean to English speakers outside of North America?

    Link to this
  46. 46. naishd 11:41 am 11/22/2012

    My take on this: terrapins are amphibious turtles with limbs neither specialised for dedicated life on land (as in tortoises), or for swimming (as in sea turtles). Emydids are the classic terrapins, but amphibious pleurodires could be called terrapins as well (though generally aren’t). I’ll tweet this to (hopefully) bring in more comments…

    Darren

    Link to this
  47. 47. Hai~Ren 11:58 am 11/22/2012

    I read a lot of American reptile hobbyist literature while growing up – to me, ‘terrapin’ is used exclusively for diamondback terrapin (Malaclemys terrapin); all the rest of the amphibious species are ‘turtles’. However, ‘terrapin’ in Singapore is often used in the ‘British’ sense, referring to semiaquatic and amphibious species like emydid and geoemydids: See Ecology Asia.

    The problem is that despite the presence of native turtles, most people here in Singapore are familiar with only 1 species of amphibious turtle – the non-native red-eared slider (which people often simply refer to as ‘terrapins’). And there are some who use ‘turtle’ exclusively for sea turtles, and ‘tortoise’ for both terrestrial and amphibious chelonians. It does lead to confusion and tragic consequences when people abandon their pets by releasing amphibious turtles at the seaside, or release their tortoises by tossing them into a lake, or even the sea!

    Link to this
  48. 48. Chelydra 1:22 pm 11/22/2012

    I don’t know why, but I find the diverse English common names for turtles fascinating. “Terrapin” is one of the few native American words to enter English in broad use (opossum also comes to mind).
    In parts of the U.S. “cooter” and “slider” are used for all amphibious emydids, not just Pseudemys and Trachemys. I’ve heard “painter” used in this sense in the north, where Chrysemys is the only common basking species. “Leatherback” is often used for softshells, not Dermochelys. I don’t think “terrapin” is widely used outside places that still eat Malaclemys, but would be recognized as an archaic synonym for “turtle”.

    How about Terrapene, Cuora and the like? Some are highly terrestrial and others are amphibious, but they’re all just “box turtles” to Americans.

    Link to this
  49. 49. Mark Robinson 1:44 am 11/23/2012

    In reply to Darren #46 (and your Tweet): can’t give you a taxonomically strict def but, as a layman, I’ve always thought of terrapins as being tortoises that spend a lot of their time in fresh water, so amphibious/semi-aquatic. They have clawed feet not flippers (but showing slight adaptation for swimming) and their shells are flatter than fully-terrestrial tortoises (I presume for improved hydrodynamic efficiency). Sort of halfway between “tortoises” and “turtles”/”sea turtles”, I guess.

    I’m in Australia, btw.

    Link to this

Add a Comment
You must sign in or register as a ScientificAmerican.com member to submit a comment.

More from Scientific American

Scientific American Dinosaurs

Get Total Access to our Digital Anthology

1,200 Articles

Order Now - Just $39! >

X

Email this Article

X