More about turtles

[JAC: In response to my own and other readers' questions about turtle morphology and evolution, Greg kindly put up another post to clarify matters.]

by Greg Mayer

To really appreciate what turtles have done with their shells, it helps to see into one. So here’s a view into a turtle’s shell. The shell has been cut parasagitally, to the right of the midline, so we can see the vertebrae, but the rest of the skeleton– head, neck, limbs, girdles, and tail– are left intact so we can see their relation to the shell. (As Turtle Dundee once said, “That’s not an evolutionary novelty– that’s an evolutionary novelty.”)

Inside a turtle.

Inside a turtle.

Note that the shoulder and pelvic girdles are both within the dome of the carapace. And in turtles, the shoulder girdle is a bigger affair than it is in mammals. Turtles not only have a scapula, but also a coracoid as a major bony element, plus turtle scapulas are two pronged affairs with an acromial process nearly as big as the main part of the scapula itself, so that the whole girdle is a tri-radiate structure.

The shell itself is composed of epidermal, dermal, and deeper skeletal elements, and all three can be seen in this view. The horny scutes on the exterior of the shell, made of the same material as scales and fingernails, are epidermal, with a layer of live cells between the scutes and the bone.  The bony part composes parts of the deeper axial skeleton: the vertebrae and ribs, which are preformed in cartilage during development; and the more superficial dermal bones, that ossify directly in the dermis without being preformed in cartilage. The plastron (bottom shell) consists of just epidermal and dermal elements. The plastral bones may be homologous to some of the dermal bones of the shoulder girdle (clavicle and interclavicle) and the gastralia of other reptiles.

As mentioned in the previous post, what the new paper by Lyson and colleagues (see previous turtle post for references) has especially done is to try to interpret the disputed turtle precursor Eunotosaurus in terms of the developmental processes proposed to underlie the evolution of the shell as seen in the undoubted turtle precursor Odontochelys and more derived turtles. Much of that recent developmental work has come from the Laboratory of Evolutionary Morphology at the RIKEN  Center for Developmental Biology in Japan, especially this 2009 paper by Hiroshi Nagashima and colleagues (see additional figures in RIKEN’s press release; and a nice review by Shigeru Kuratani and colleagues from 2011).

"[T]he amniotes' ribs and muscle plate grow together ventrally and make a single layer in body, outside of which the scapula is situated. In turtles, ribs grow laterally and are confined dorsally. However muscle plate is folded at the tip of ribs and runs inside the scapula as in other amniotes, showing basic topology between the elements is not changed both in turtles and other amniotes." From RIKEN.

“[T]he amniotes’ ribs and muscle plate grow together ventrally and make a single layer in body, outside of which the scapula is situated. In turtles, ribs grow laterally and are confined dorsally. However muscle plate is folded at the tip of ribs and runs inside the scapula as in other amniotes, showing basic topology between the elements is not changed both in turtles and other amniotes.” From RIKEN.

Lyson and colleagues integrate Eunotosaurus into Nagashima and colleagues’ developmental scenario by proposing that the rib broadening seen in Eunotosaurus is homologous with that in Odontochelys and Proganochelys. One problem that I see with this attempt is that, if I’m interpreting Eunotosaurus correctly, the distal ends of its ribs curve ventrally and tuck in around the lateral edge of the body, while in turtles the ribs grow out straight to the sides towards a feature in the embryo called the carapacial ridge; this is how, in fact, the scapula gets inside the ribs. It could be argued, though, that this straightness is a later evolved feature, although it’s the straight lateral growth that is associated with getting the cartilaginous ribs associated withe bone-producing dermis that produces the broad plates of bone.

Figure 4 from Lyson et al. 2013.

Figure 4 from Lyson et al. 2013 (click to see enlarged view).

To stress again how profound are turtles’ morphological changes in skeletal and soft tissues, and how they ramify throughout its physiology, ecology and behavior, let me quote from the famous morphologist and paleontologist Rainer Zangerl (1969; and who, in the quoted paper, also refers to the development of the turtle shell as “astounding”):

This shell did not merely cover the pre-existing anatomical structures of the body, but it modified them profoundly. The drastic alteration was probably the consequence of an intimate involvement of the dermal shield with parts of the axial skeleton (vertebrae and ribs) and with dermal portions of the primary skeleton (clavicles, interclavicle and gastralia). The restructuring had far reaching morphological, physiological, evolutionary, and ecological consequences . Anatomically, the presence of a rigid shell led to extensive modification of the structure of the body wall, and to the differentiation of the locomotor apparatus, of the neck region and of the copulatory mechanism, to mention just a few . Functionally it necessitated changes in the mode of respiration ; it impaired locomotion, especially on the ground, and restricted aquatic locomotion to the “paddle types” ; it delimited the storage capacity of the body for air, food, water, fats (oil in the shell bones of sea turtles), and waste materials . It probably affected the permeability of the body wall and hence modified the capacity of the animal to retain water . It drastically modified the behavioral pattern of the animal.


Kuratani, S., S. Kuraku, and H. Nagashima. 2011. Evolutionary developmental perspective for the origin of turtles: the folding theory for the shell based on the developmental nature of the carapacial ridge Evolution & Development 13:1-14. abstract 

Nagashima, H. et al. 2009. Evolution of the turtle body plan by the folding and creation of new muscle connections Science 325: 193-196. pdf

Zangerl, R. 1969. The turtle shell. Biology of the Reptilia 1: 311-339. pdf


  1. Dominic
    Posted June 4, 2013 at 3:38 am | Permalink

    Isn’t that a tortoise (top pic) or are we as I think was said the other day, including all tortoises as withing the turtle clade?

    • Dominic
      Posted June 4, 2013 at 3:49 am | Permalink

      Sorry I see it is just the common usage to say turtle (Ant in previous post comment). Properly, I mean scientifically, do we say Chelonii or Testudines then as an order? Is either one acceptable?

      Did land based (tortoise like behaviour) as opposed to water based (turtle like behaviour) come first, or did both re-emerge as life modes more than once?

      • Posted June 4, 2013 at 4:17 am | Permalink

        I second these questions. I’m curious about this, as well.

        • oikos
          Posted June 4, 2013 at 6:29 pm | Permalink

          If I recall the paper by Hobart Smith on the subject, Testudines has priority and is thr proper ordinal name, not Chelonia or Testudinata.

          My memory could be off on this. As I remember it, I read the paper about half a century back. I am hoping that I still have turtle recall.

    • Posted June 4, 2013 at 5:36 am | Permalink

      That is a tortoise in the first picture. For the common names of these shelled reptiles, British and American usage diverges. In America, “tortoise” is applied to members of the family Testudinidae, the “true tortoises” (which are terrestrial). The word “terrapin” is applied to a single species, the Diamonbacked Terrapin, Malaclemys terrapin, which inhabits coastal salt marshes along the east coast of the USA. Everything else is called a “turtle” (for example, the Box Turtle, Terrapene carolina).

      In Britain, “tortoise” is applied to all terrestrial species, which includes members of the Testudinidae, but also terrestrial members of other families (for example the Box Tortoise, Terrapene carolina, of the Emydidae). Fresh and brackish water aquatic chelonians are called “terrapins”, (for example the European Pond Terrapin (Emys orbicularis). “Turtle” usually means a sea turtle (for example, the Green Turtle, Chelonia mydas).

      There are also special common names, such as ridley, slider, and cooter. Some are used in both Britain and Ameirca (for example Kemp’s Ridley, Lepidochelys kempii, a type of sea turtle), while some are used in one but not the other (for example, the American name for Trachemys scripta elegans is the Red-eared Slider, while in Britain it is the Red-eared Terrapin).

      These modal differences in British and American usage are not so much a code, but more what you call guidelines.

      Traditionally, turtles (in the broad sense, including tortoises, terrapins, sliders, etc.) are taxonomically ranked as an order, and current consensus would put this taxon somewhere within the Diapsida. The name of the order is given, with almost equal frequency, as Chelonia or Testudines. Of two field guides (both quite good) that I happen to have on my desk as I write this, one uses Chelonia, the other Testudines. The Turtle Taxonomy Working Group is ecumenical in its approach: its authoritative annotated checklist, “Turtles of the World“, uses Testudines, but it appears in the journal Chelonian Research Monographs, which is published by the Chelonian Research Foundation!


      • TJR
        Posted June 4, 2013 at 5:55 am | Permalink

        Thanks very much for that.

        This explains the terrible holy wars between the Testudinists and the Chelonianites, because science is just another religion, right?

      • Dominic
        Posted June 4, 2013 at 6:18 am | Permalink

        Thanks Greg – comprehensive as ever! However the second follow up question – land living v water living, both ways of being a Chelonian arose once or many times?

        • Posted June 4, 2013 at 12:02 pm | Permalink

          The earliest turtle, Odontochelys is from undoubtedly coastal marine sediments, and aspects of its morphology suggest it was aquatic. The next oldest turtle, Proganochelys, seems to be terrestrial. The most straightforward reading of the fossil record is that turtles were originally aquatic. Among living turtles, fully aquatic forms (i.e. with flippers) have evolved at least twice. Without doing a full analysis, I think it would be fair to say turtles have moved from water to land and back on more than one occasion.


          • Greg Peterson
            Posted June 4, 2013 at 2:45 pm | Permalink

            Perhaps someone else has already mentioned it–I have not read all of the comments–but Richard Dawkins in “The Greatest Show On Earth” has a fun overview of the different names used, and an extensive treatment of the number of times turtles have evolved into an out of the water. It’s worth checking out.

            • Dominic
              Posted June 5, 2013 at 1:31 am | Permalink

              Thanks – it is in a pile of books waiting to be read!

          • Dominic
            Posted June 5, 2013 at 2:08 am | Permalink

            That makes sense – marginal areas are supposed to be more dynamic with lots of potential niches that would encourage animal & plant diversity…

  2. Posted June 4, 2013 at 4:16 am | Permalink

    Stuff like this is why I keep coming back.

    Well that, and you haven’t turned into some completely self-important, polemic asshole, sort of the Father Coghlin of Atheism, like some of the other sites I used to read.

  3. Marcoli
    Posted June 4, 2013 at 4:57 am | Permalink

    Inside a turtle? I thought it would be darker.

    • JimmyHaulinHogs
      Posted June 4, 2013 at 9:56 am | Permalink

      Surely, it would be too dark to read.

  4. Marcoli
    Posted June 4, 2013 at 5:06 am | Permalink

    I was looking at the web site The Pterosaur Heresies (which is a cool paleontology site full of great pictures), and found a posting about other extinct reptiles that look like turtles but are NOT turtles. The turtle morph has been the target of convergent evolution on several occasions.

    • Posted June 4, 2013 at 5:53 am | Permalink

      Pterosaur Heresies and its sister website,, are not reliable sources. Dave Peters is a wonderful artist, and I’ve admired and cited his earlier work, but you need to confirm anything you see there in a reliable source (and much of what you see there will not be confirmable). Before looking at those sites again, you must read Darren Naish’s sympathetic yet devastating critique at Tetrapod Zoology. Money quote: does not represent a trustworthy source that people should consult or rely on. Students, amateur researchers and the lay public should be strongly advised to avoid or ignore it.


      • John Harshman
        Posted June 4, 2013 at 6:13 am | Permalink

        However, Peters is right in this particular case. A turtlish morphology does seem to have appeared several times, in turtles, parareptiles, and placosaurs at least. And still more times if we broaden the definition to include ankylosaurs, pangolins, glyptodonts, and armadillos.

      • Marcoli
        Posted June 4, 2013 at 7:12 am | Permalink

        Thanks for the heads up on that.

  5. E.A. Blair
    Posted June 4, 2013 at 5:34 am | Permalink

    I’ve seen enough cartoons to know that all turtles (and tortoises as well) have shells that are 100% removable – and some are even equipped with rocket engines.

  6. Michael Fisher
    Posted June 4, 2013 at 8:44 am | Permalink

    Thanks for two great turtle posts Greg

    The Rainer Zangerl quote brought home to me how extensive & weird the changes have been.
    I can’t picture how all these changes happened gradually without being disadvantageous at various points.
    That’s my failing of course.

  7. Gregory Kusnick
    Posted June 4, 2013 at 9:59 am | Permalink

    It drastically modified the behavioral pattern of the animal.

    Isn’t this putting the cart before the horse? I’m no expert, but I’m guessing that the selection pressures that led to these morphological adaptations were at least partially the result of changes in behavior and habitat.

  8. Dominic
    Posted June 5, 2013 at 2:09 am | Permalink

    Of course the wold sits on a giant turtle swimming through space – how could I forge!

  9. Posted June 6, 2013 at 1:48 pm | Permalink

    You can see the animated progression in BBC, Attenborough, Life in Cold Blood, The Armoured Giants, playing now.

  10. Posted June 6, 2013 at 1:52 pm | Permalink

    You can se the animated progression in Attenborough’s LIFE IN COLD BLOOD, THE ARMOURED GIANTS on BBC, playing now.

  11. Posted June 6, 2013 at 2:44 pm | Permalink

    Thanks for this informative update.

  12. marksolock
    Posted June 9, 2013 at 7:31 am | Permalink

    Reblogged this on Mark Solock Blog.


Get every new post delivered to your Inbox.

Join 27,102 other followers

%d bloggers like this: