ADVERTISEMENT
  About the SA Blog Network













Tetrapod Zoology

Tetrapod Zoology


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

Nectrideans: more than just Diplocaulus

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


Email   PrintPrint



Nectrideans are a Carboniferous and Permian group of tetrapods, typically regarded as ‘amphibians’ and classified alongside microsaurs and lysorophians within the group Lepospondyli. However, close relationships with the long-bodied aïstopods have been supported by some: indeed, nectrideans have been found to be paraphyletic to aïstopods in some studies (Ruta et al. 2003). Thomson & Bossy (1970) gave the name Holospondyli to a hypothesised nectridean + aïstopod clade. Diplocaulus – the most famous nectridean – is typically portrayed as here. However, fossils show that these ‘horned’ nectrideans actually had skin webs connecting the tips of their ‘horns’ to their bodies.

Next: plesiosaurs!

Refs – -

Ruta, M., Coates, M. I. & Quicke, D. L. J. 2003. Early tetrapod relationships revisited. Biological Reviews 78, 251-345.

Thomson, K. S. & Bossy, K. H. 1970. Adaptive trends and relationships in early Amphibia. Forma et Functio 3, 7–31.

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! Follow on Twitter @TetZoo.

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





Rights & Permissions

Comments 67 Comments

Add Comment
  1. 1. accipiter 2:54 pm 01/5/2013

    interesting on the skin flaps,

    how far back the body did they extend and what shape did they had? has anyone looked at their possible functions?

    natation assistance? camouflage (by breaking the body shape outline)? or even *gasp* thermoregulation/breathing assistance if they where richly vascularised?

    Link to this
  2. 2. Heteromeles 2:55 pm 01/5/2013

    Cheeky buggers, aren’t they?

    Link to this
  3. 3. ectodysplasin 3:17 pm 01/5/2013

    @accipter,

    There isn’t really a satisfactory answer to the question of skull shape in diplocaulids. Cruickshank & Skews (1980) argued that the tabular horns may have served as some sort of hydrofoil, but that doesn’t explain the short-horned forms (Peronedon and Ductilodon as well as some skulls attributed to Diplocaulus). Skulls of Diploceraspis, which is the longest-horned form from Western Pennsylvania, Ohio, and West Virginia, show some manner of ornament at the tips of the horns, but what this indicates is also unclear.

    I doubt that diplocaulids would have used the horns for thermoregulation, as all these animals are essentially obligately aquatic with gills, lateral line canals, reduced limbs, deep paddle-like tails (this, by the way, is one of the few common characters that unites the Nectridea), and so on.

    There’s some manner of new information coming together on some of these animals, but you’ll have to wait for the papers.

    Link to this
  4. 4. ectodysplasin 3:31 pm 01/5/2013

    @Darren,

    The Nectridean-Aistopod relationship is probably only a persistent feature of some phylogenetic analyses (although not all) due to convergences associated with small size and elongation of the trunk and tail, and due to poor knowledge of key parts of the skull in both groups. In fact, Carroll has argued as recently as 1998 that aistopods are probably not closely related to any other ‘Lepospondyl’ groups, and provided some pretty intriguing evidence in favor of this view, albeit without a cladogram to back it up. There’s some new research on the matter that’s currently forthcoming. Keep an eye out, it’s pretty exciting stuff, if you’re into early tetrapods.

    Link to this
  5. 5. naishd 4:52 pm 01/5/2013

    Wow – great comments, thanks, everyone (smiley).

    Skin webs: I should add that these are inferred from resting traces, not preserved on body fossils. They were reported in…

    Walter, H. v, & Wernerberg, R. 1988. Über Liegespuren (Cubichnia) aquatischer Tetrapoden (?Diplocauliden, Nectridea) aus den Rotteröder Schichten (Rotliegendes, Thüringer Wald/DDR). Freiberger Forschungsheft 419, 96-106.

    Darren

    Link to this
  6. 6. naishd 5:15 pm 01/5/2013

    Diplocaulid horns: as we said in our recent TREE article on sexual selection, the shape of the allometric slope in these animals is similar to what you see in extant animals that use elaborate structures in sociosexual display (Knell et al. 2013). Such an interpretation would not, we note, be inconsistent with a role in generating lift as well, as proposed by Cruickshank & Skews (1980).

    ectodysplasin: thanks for these interesting teasers, we all look forward to seeing the work concerned. Love your work, by the way (some of you may have guessed that ectodysplasin has a special, technical interest in anamniote phylogeny).

    Darren

    Ref – -

    Knell, R. J., Naish, D., Tomkins, J. L. & Hone, D. W. E. 2013. Sexual selection in prehistoric animals: detection and implications. Trends in Ecology and Evolution 28, 38-47.

    Link to this
  7. 7. David Marjanović 5:44 pm 01/5/2013

    Not all “nectrideans” are diplocaulids or urocordylids! There are also the probably amphibious (rather than aquatic) scincosaurids!

    Random puzzling fact: Urocordylus and the diplocaulid Diceratosaurus had five fingers per hand, rather than the four otherwise found in lepospondyls, temnospondyls and lissamphibians.

    nectrideans have been found to be paraphyletic to aïstopods in some studies (Ruta et al. 2003)

    As well as version 2 of that matrix (Ruta & Coates 2007, J of Syst Pal) and all the unpublished successors – I hope to submit a version for publication this month or next.

    The Nectridean-Aistopod relationship is probably only a persistent feature of some phylogenetic analyses (although not all) due to convergences associated with small size and elongation of the trunk and tail

    But that’s it: aïstopods have stupidly long tails, unlike any other limbless or limb-reduced tetrapod with the exception of a few climbing snakes that were mentioned here on Tet Zoo not long ago. The only way this makes sense is if they’re descended from very long-tailed limbed animals… urocordylid “nectrideans” have such tails*, and they tend to come out as the aïstopod sister-group (they always do in the matrix series cited above).

    * The skeletal reconstruction above actually has a much shorter tail than some specimens I’ve seen. They look like pleurosaurs or the Cretaceous marine squamate Pontosaurus; it’s a sight to behold.

    There are plenty of similarities in the unusual vertebrae and in what’s known of the skull, too, though I agree that additional research will help.

    Analyses that fail to find Holospondyli tend to find aïstopods and adelospondyls as sister-groups. Now that looks like convergence due to elongation.

    Carroll has argued as recently as 1998

    If he’s right, he’s right for the wrong reasons. Have you read his 2007 review? Carroll has never understood phylogenetic analysis.

    Freiberger Forschungsheft 419

    Whoa. The museum I’m working in was in East Berlin, so I might be able to get a hold of that micro-journal without writing to the authors, but no promises.

    …Wait. Is “R. Wernerberg” actually Ralf Werneburg? That would make writing to the author a lot easier.

    ectodysplasin: thanks for these interesting teasers, we all look forward to seeing the work concerned. Love your work, by the way (some of you may have guessed that ectodysplasin has a special, technical interest in anamniote phylogeny).

    Yeah, I think I’ve guessed who that is, even though I’m usually bad at that. :-)

    Link to this
  8. 8. naishd 5:50 pm 01/5/2013

    See? Set the bait, and you can lure him in :)

    Darren

    Link to this
  9. 9. Metridia 5:12 am 01/6/2013

    Off topic, but I have been wondering recently about the phylogeny of alligators. How did one congener end up in China and one in America? I can’t seem to find the correct search terms for this. I wasn’t aware there were large reptiles in Beringia ever (except maybe the Eocene, although not sure if Beringia existed then), or did they cross the Atlantic?

    Link to this
  10. 10. ectodysplasin 5:16 am 01/6/2013

    @Darren,

    You probably have guessed the right lab, but you may not be guessing the right grad student. Unless you caught my 2011 talk on, well, diplocaulids, in which case you probably are.

    As far as horn use, there are some anatomical concerns that have not, to my knowledge, been stated yet in the literature (so I won’t state them here). I don’t think the Cruickshank & Skews hypothesis really makes a ton of sense; besides the fact that these animals show numerous anatomical features suggesting they were gape-and-suck ambush predators, diplocaulids are almost always at least a little asymmetrical, and are sometimes grotesquely so (see Germain, 2010, for a description of this in Diplocaulus from Morocco) which isn’t exactly conducive for a hydrofoil.

    Link to this
  11. 11. ectodysplasin 6:00 am 01/6/2013

    @David,

    Hi David!

    Scincosaurids are definitely also worth mentioning, because they’re a good example of the sorts of problems that are involved in nectridean phylogeny. Scincosaurids are unique among nectrideans for a variety of reasons, including a rather exaggerated olecranon process, suggesting that it was probably not only partly terrestrial, but was probably also engaged in some manner of burrowing. Some other features of the skeleton (e.g. the entepicondylar foramen) seem to be unique within nectrideans (albeit commonplace among most stem-amniotes and early amniotes). The palate is also pretty distinct from essentially all other nectrideans (although nectridean palates and occiputs seem to be extraordinarily diverse).

    As for aistopods, they do indeed have stupid-long tails, but there’s quite a lot about them that doesn’t make sense no matter where you put them. The Anderson (2001) result, in which aistopods fall out with lysorophians and nectrideans) makes superficial sense, but is a disaster when you actually look at it in detail. The adelospondyl result is equally troublesome. So what it appears like to me is that aistopods are ending up in these places because they have to end up somewhere, and those are the least terrible places to put them, as opposed to reasonable to good places to put them. The new data, which is extensive, disagrees pretty strongly with both of these topologies, and raises a whole mess of new questions. As for Carroll’s interpretations in his 1998 paper, Carroll is certainly not a phylogeneticist, but he’s an excellent anatomist, and he’s right that there are anatomical features of these animals that don’t make a lot of sense.

    And yes, you know exactly who I am!

    Link to this
  12. 12. ectodysplasin 6:24 am 01/6/2013

    @Metridia

    Alligatoroids (e.g. Allognathosuchus) are found in the high arctic during the Paleocene-Eocene thermal maximum, in places like Ellesmere Island. Other Eocene alligatoroids (e.g. Diplocynodon) seem to have maintained distributions across both Eurasia and North America at higher latitudes.

    Molecular clock divergence estimates for Alligator mississippiensis and A. sinensis suggest a divergence of these two genera in the Eocene, so it is not unlikely you’re looking at an arctic dispersal route, either via Beringia or via Greenland and Europe.

    Dispersal of Crocodylus seems to have been marine, however.

    Link to this
  13. 13. David Marjanović 8:18 am 01/6/2013

    Diplocaulus from Morocco

    (I forgot if Damien Germain mentions this in that paper, but D. minimus may actually belong to Diploceraspis instead; Diplocaulus needs a revision.)

    Scincosaurids are unique among nectrideans for a variety of reasons, including a rather exaggerated olecranon process, suggesting that it was probably not only partly terrestrial, but was probably also engaged in some manner of burrowing. Some other features of the skeleton (e.g. the entepicondylar foramen) seem to be unique within nectrideans (albeit commonplace among most stem-amniotes and early amniotes).

    I agree that the shape of the humerus in general suggests lots of power, so probably some kind of digging. Scincosaurus and Sauravus (which needs to be redescribed) being unique among “nectrideans” may not mean much, though – ever since Ruta et al. (2003), Scincosaurus has come out as the sister-group of all other holospondyls that are in that matrix and its successors (Sauravus isn’t in there).

    As for aistopods, they do indeed have stupid-long tails, but there’s quite a lot about them that doesn’t make sense no matter where you put them.

    :-) That is most definitely true!

    The Anderson (2001) result, in which aistopods fall out with lysorophians and nectrideans) makes superficial sense, but is a disaster when you actually look at it in detail.

    Of course, when you look at that matrix in detail, you may find things you don’t like…

    Carroll is certainly not a phylogeneticist, but he’s an excellent anatomist, and he’s right that there are anatomical features of these animals that don’t make a lot of sense.

    But then, “nothing in biology makes sense except in the light of evolution”, and “nothing in evolution makes sense without a phylogeny”. If you put all the right anatomy into the wrong framework, let alone no framework at all, it won’t help much.

    Molecular clock divergence estimates for Alligator mississippiensis and A. sinensis suggest a divergence of these two genera in the Eocene

    Huh. I thought it was all Miocene. (The early and middle Miocene were quite warm, much warmer than the Oligocene.)

    Dispersal of Crocodylus seems to have been marine, however.

    Yes, they seem to have crossed the central Atlantic around the beginning of the Pliocene.

    Link to this
  14. 14. naishd 9:45 am 01/6/2013

    ectodysplasin: if your initials are JP, then I do know who you are :)

    Loving the great comments, thanks, guys.

    Darren

    Link to this
  15. 15. Heteromeles 11:18 am 01/6/2013

    @Metridia

    Alligatoroids (e.g. Allognathosuchus) are found in the high arctic during the Paleocene-Eocene thermal maximum, in places like Ellesmere Island. Other Eocene alligatoroids (e.g. Diplocynodon) seem to have maintained distributions across both Eurasia and North America at higher latitudes.

    Molecular clock divergence estimates for Alligator mississippiensis and A. sinensis suggest a divergence of these two genera in the Eocene, so it is not unlikely you’re looking at an arctic dispersal route, either via Beringia or via Greenland and Europe.

    Then, I believe rules of priority demand that all “Alligatoroid” fossils be folded into the genus Alligator? After all, if two widely dispersed and deeply diverging members of the genus are still in the genus, then everything in between automatically belongs there too, no?

    Link to this
  16. 16. Glendon Mellow 1:45 pm 01/6/2013

    Who did the colour illustration? Gorgeous.

    Link to this
  17. 17. ectodysplasin 1:45 pm 01/6/2013

    @David,

    Damian definitely suggests similarities between D. minimus and Diploceraspis, but there are a number of characteristics not included in his dataset that would suggest D. minimus is indeed more closely related to classic Diplocaulus from Texas and Oklahoma. There are other populations of cf. Diplocaulus, such as the skulls from the Abo and Bursum, which should be critically reevaluated as well. Nomenclature within the Dilocaulidae is sort of outside my interest in the group, which is focused on anatomy, especially ear, braincase, and palate, and relationship of diplocaulids with other early tetrapods (and whether the small-horned species Diceratosaurus, Keraterpeton, and Batrachiderpeton are actually related to them).

    The anatomy of Scincosaurus is problematic no matter where you put it in a clade of nectrideans+aistopods. Once again, we’re talking about “least terrible” rather than “best” phylogenetic position.

    As for early tetrapod phylogenies and re: Carroll, phylogenetic analysis is a technology that is extremely important for understanding the evolution of characteristics, but it’s only as good as the dataset being analyzed, and the early tetrapod dataset is, across the board, at a very early stage in the refinement process. Refining the dataset necessarily requires extensive refinement of anatomy and in some cases refining the anatomy requires reference to assumed close relatives (which is, in my opinion, part of the problem we’ve been having with aistopods for years). So, yes, all phylogenetic hypotheses need to be evaluated and reevaluated in a rigorous systematic framework, but this needs to be part of a dialogue between anatomists and systematists with neither approach being given full veto power. My gut feeling is that once we have all this worked out, we’re going to have small-bodied holospondylous anamniotes dispersed all over the early tetrapod tree.

    Link to this
  18. 18. ectodysplasin 1:46 pm 01/6/2013

    @Darren,

    Yeah, that’s me.

    Link to this
  19. 19. ectodysplasin 1:49 pm 01/6/2013

    @Heteromeles

    Alligatoroids not contained in the genus Alligator are phylogenetically distinct. This includes all caimans and a bunch of other stuff. The genus Alligator only represents a subset of the Alligatorinae.

    Link to this
  20. 20. puppygod 2:18 pm 01/6/2013

    My pet theory is that those weird, wide skulls were somehow involved in some kind of electroreception – maybe locating small prey in the mud.

    Unfortunately, it’s one of those hypotheses that probably will never be confirmed nor falsified.

    Link to this
  21. 21. Mythusmage 3:17 pm 01/6/2013

    Along with the richer mix of oxygen could there have been higher atmospheric pressure? Combine this with a roughage rich diet and a bad digestion, we’re talking about a methane rich flatulence. Combine that with an oxygen rich atmosphere and a correspondingly low ignition point and you’ve got a spontaneously igniting rocket. Combine that with the cheek flaps and you might have had rocketing amphibians swooping through the air from lagoon to marsh to swamp.

    Link to this
  22. 22. ectodysplasin 9:25 pm 01/6/2013

    @puppygod

    Electroreception seems to be distributed more or less across vertebrates, including lungfishes, so that’s not completely out, but we don’t really see the sorts of distinctive dermal structures in nectridean bone that would suggest the presence of a lot of electroreceptive organs. Furthermore, if they were feeding on small prey in the substrate, you’d expect a subterminal mouth (similar to what you see in hammerhead sharks) and that’s not the case; the mouth in diplocaulids is quite terminal.

    Link to this
  23. 23. David Marjanović 10:41 am 01/7/2013

    Then, I believe rules of priority demand that all “Alligatoroid” fossils be folded into the genus Alligator? After all, if two widely dispersed and deeply diverging members of the genus are still in the genus, then everything in between automatically belongs there too, no?

    Yes, but that’s just one or two Miocene and younger species from North America. The divergence isn’t that deep. A. sinensis and A. mississippiensis are closer to each other than to the Oligocene North American A. prenasalis.

    Once again, we’re talking about “least terrible” rather than “best” phylogenetic position.

    Gaps in the fossil record make this inevitable. In many clades we won’t get the uninterrupted series David Peters wants anytime soon.

    phylogenetic analysis is a technology that is extremely important for understanding the evolution of characteristics, but it’s only as good as the dataset being analyzed, and the early tetrapod dataset is, across the board, at a very early stage in the refinement process.

    Indeed. That’s what we’re all working on!

    So, yes, all phylogenetic hypotheses need to be evaluated and reevaluated in a rigorous systematic framework, but this needs to be part of a dialogue between anatomists and systematists with neither approach being given full veto power.

    Uh, of course. My point is that Carroll doesn’t refrain from making phylogenetic hypotheses; he likes taking two animals, saying “oh, how similar”, and declaring the older one a more or less direct ancestor of the younger one.

    My pet theory is that those weird, wide skulls were somehow involved in some kind of electroreception – maybe locating small prey in the mud.

    Unfortunately, it’s one of those hypotheses that probably will never be confirmed nor falsified.

    Electroreception organs are plesiomorphic for vertebrates, are found in aquatic lissamphibians, and left small pits on the skull of the aquatic seymouriamorph Discosauriscus. Such pits haven’t been reported from Diplocaulus or Diploceraspis, but I don’t know if anyone has looked for them.

    Link to this
  24. 24. Heteromeles 2:23 pm 01/7/2013

    Since I’m enjoying stirring up the alligator’s nest, I’ve got to point out that Ectodysplasin put the split between the Chinese and American Alligators in the EOCENE, which is why I suggested that, if the two species diverged that early, all the more recently diverging forms should be folded in to this very inclusive genus–or the Chinese alligator should be split off.

    David Marjanović says MIOCENE for divergence.

    That’s a bit of a discrepancy guys. Want to resolve it?

    Link to this
  25. 25. Andreas Johansson 3:59 pm 01/7/2013

    As I read David, he didn’t put the divergence in the Miocene, but the oldest (known) fossil species that belongs inside the mississippiensis + sinensis clade. A. prenasalis being younger than the split doesn’t mean it has to be inside.

    Link to this
  26. 26. ectodysplasin 5:43 pm 01/7/2013

    @Heteromeles,

    I was referring to molecular clock estimates, and specifically thinking of this paper. The earliest fossils attributed to the genus are indeed Miocene.

    Link to this
  27. 27. David Marjanović 6:06 pm 01/7/2013

    As I read David

    You’re technically correct. Alligators aren’t easy to overlook in the fossil record, though…

    I was referring to molecular clock estimates, and specifically thinking of this paper.

    Thanks, I’ll have a look tomorrow to see how that tree was calibrated. Calibrating molecular-dating analyses is so full of traps that I’ve published on it and will do so again…

    Link to this
  28. 28. ectodysplasin 6:15 pm 01/7/2013

    @David,

    Some level of gappiness in the record is to be expected, but that’s not the only issue with aistopods and scincosaurids.

    Link to this
  29. 29. Jenny Islander 4:20 am 01/8/2013

    No pits for electroreception? Darn, I was hoping.

    Re skin webs behind the “horns,” is it possible that this was not a flat skin web, but a baglike formation? IIRC, some deep-sea fishes have really wildly shaped heads apparently because when they open their mouths suddenly, water rushes in with greater force. Could these animals have lurked, gaped, pinned the inrushing prey with their teeth, then worked the prey down into their throats with head movements?

    Link to this
  30. 30. MMartyniuk 7:53 am 01/8/2013

    Interesting to hear about the “skin flaps”! Looking at the skulls in the figure even before reading the text, I was wondering if these might not simply be embedded in the soft tissue of the “neck” and torso. Do the trace fossils show these are actual skin flaps and not just the body wall itself, with the “horns” showing the true, non-shrink-wrapped, “fat” contour of the body?

    Link to this
  31. 31. David Marjanović 9:55 am 01/8/2013

    There are only two calibration points. That’s plainly not enough. Both are also very old, and both lie outside Alligator (which is fine for that paper, because the clade of interest was Crocodylia, not just A.); these two factors are both known to cause overestimated divergence dates.

    Could these animals have lurked, gaped, pinned the inrushing prey with their teeth, then worked the prey down into their throats with head movements?

    Well, yes, but the “horns” can’t have anything to do with that. They don’t contain the jaw joints. The mouth wasn’t unusually wide at all, and neither is the shoulder girdle.

    Link to this
  32. 32. Heteromeles 11:39 am 01/8/2013

    Cool. Thanks David for checking that molecular clock.

    Now, to get back to Metridia’s original question about how we’ve got one alligator in China and one in the US, how did that happen? Apparently they didn’t use the PETM Greenland land bridge, so when was Beringia warm enough for gator crossings again?

    Link to this
  33. 33. BrianL 12:13 pm 01/8/2013

    @Heteromeles:
    My guess is that *Alligator* crossed either during the Miocene or during the warmest period of the Pliocene. That last period also saw crocodiles reappear in the southern Europe and apparently led to some (more) ice-free coasts for Antarctica. At least, that’s what my copy of ‘Mammoths, sabertooths and hominids’ by Turner and Antón says. It would seem reasonable that such conditions might have made Beringia warm enough for alligators to cross.

    Link to this
  34. 34. ectodysplasin 2:19 pm 01/8/2013

    @Jenny,

    The jaws of (long-horned) diplocaulids aren’t the full width of the skull (see Diploceraspis, jaw joints represented by the quadrate, labeled q, image from Beerbower, 1963); those horns are exaggerations of the otic region rather than the jaw joints themselves.

    That said, there are some features of the skull that do suggest some sort of lurking/gape-and-suck predation. The occiput is extremely posteriorly located, and the jaw joints are pretty widely-spaced and anteriorly displaced. The mandibular depressor muscles were probably well-developed (and partly incorporated into the horn structure) but the mandibular adductors were probably quite weakly developed. This is similar to cryptic aquatic ambush predators like Chelus (see Lemell et al., 2010), Pipa, and various aquatic salamanders. So, were they feeding on extremely large prey? Probably not. But they probably were ambush predators, and they may have been somewhat cryptic as well. I imagine that, in life, these animals would have looked more or less like rotting vegetation.

    Link to this
  35. 35. ectodysplasin 2:31 pm 01/8/2013

    @Matt,

    These animals had internal gills (similar to those seen in dvinosaurs, see Schoch & Witzmann, 2010) that would have exited through a spiracle or opercular slit posterior to the skull, so it seems unlikely that these horns were fully incorporated into the posterior body wall.

    Some of the “short-horned” forms (e.g. Keraterpeton and Diceratosaurus) may have had some sort of connection between the occipital horns and the pectoral girdle, but this does not seem to be the case in what I’d call “classic diplocaulids” which includes Diplocaulus and Diploceraspis and two short-horned taxa (Peronedon and Ductilodon) which are periodically synonymized with Diplocaulus. In classic diplocaulids, the horn is an element entirely separate from the pectoral girdle.

    Link to this
  36. 36. ectodysplasin 4:06 pm 01/8/2013

    @David,

    There are only two calibration points. That’s plainly not enough. Both are also very old, and both lie outside Alligator (which is fine for that paper, because the clade of interest was Crocodylia, not just A.); these two factors are both known to cause overestimated divergence dates.

    The extant diversity within Crocodylia is pretty limited (24-25 species) and there’s some disagreement between the morphological and molecular trees, which means that it’s difficult to ascertain where a lot of the older fossils actually fall out. That author uses a relaxed clock model, so the clock estimates all come with error bars, which still constrains the A. mississippiensis – A. sinensis split to the Eocene, albeit a rather wide spread of time within the Eocene. Finally, they picked pretty deep divergences (including the Alligatorinae-Caimaninae split) so their nodal estimates are interpolations, not extrapolations. There certainly are criticisms that can be made of molecular clock methods in general, but this paper seems to be following current standards for best practices. Arctic dispersal of Alligator across terrestrial or freshwater environments also seems to be supported by morphology and physiology (see Brochu 1999), so the results of that molecular clock study are not particularly shocking.

    Link to this
  37. 37. vdinets 7:30 pm 01/8/2013

    It is worth noting that alligators are very cold-resistant. Chinese gators can survive two months of subzero temperatures by hiding in deep burrows with underwater entrances.

    Note also that there are many thousnads of taxa with similar distribution, with some species in eastern North America and others in E Asia.

    Link to this
  38. 38. Allen Hazen 2:24 am 01/9/2013

    Vdinets (#37)–
    re:
    “Note also that there are many thousnads of taxa with similar distribution, with some species in eastern North America and others in E Asia.”

    I take it that the salamander family (Cryptobranchidae?) containing the American Hellbender and the Asian Giant Salamander is an example of this?

    Link to this
  39. 39. Dartian 4:31 am 01/9/2013

    Allen:
    I take it that the salamander family (Cryptobranchidae?) containing the American Hellbender and the Asian Giant Salamander is an example of this?

    Yes – if we only consider the current distribution of this taxon. But the fossil record shows that cryptobranchids have been geographically much more widely spread (remember Scheuchzer’s “Homo diluvii testis“!).

    Link to this
  40. 40. Heteromeles 9:44 am 01/9/2013

    Oh, I agree that many species are found in the eastern US and eastern Asia. They also occur (to a lesser extent) in the Pacific Northwest.

    There are two problems here:

    One is the timing of migration. The older books seem to assume that there was one a more-or-less uniform “boreal forest” that spanned the northern hemisphere, and that differential extinction in different regions has caused the patterns we see today. That’s great, except that the time of greatest boreal linkage was the PETM, and a lot has changed since then.

    There’s also the little problem that species don’t migrate as communities, they migrate individually.

    But the bigger problem is that this story hasn’t been well-reconciled with cladistics. The Alligator hypothesis proposed above (that alligators migrated between China and the US in the Eocene, but the group diverged in the Miocene) is one of those examples.

    My prediction is that it’s actually worth closely revisiting the old “Boreal Forest” idea, and looking at how individual clades migrated back and forth. My suspicion is that it’s not only a lot messier than we like to think, but that we’ve got the fossil data to prove it in some cases, except that no one has come up with a better story.

    As for giant salamanders, one cryptid is a giant salamander reputed to live in the northwest California. Anyone want to go looking for it?

    Link to this
  41. 41. ectodysplasin 11:08 am 01/9/2013

    @Heteromeles,

    I think you’re overstating the case against Eocene divergences for these taxa. In the case of Alligator, the sister taxon to the genus is already present by the late Paleocene, so while direct fossil evidence for Eocene Alligator may not exist, we have very good reason to believe they were around at that time, which is consistent with the molecular clock data that suggests the same. In the case of cryptobranchids, the split between Andrias and Cryptobranchus had already occurred by the early Eocene, and we know this because we have fossil Andrias in the Eocene of Europe. We also have Cryptobranchus from the Oligocene of North America, so that divergence happened before the Miocene.

    There are a few latecomers from Asia (lampropeltine snakes might be one of these latecomer groups, or the molecular clock estimates may be too young due to the extraordinarily patchy snake record), but in general this Eocene divergence seems to hold for quite a lot of amphibians and reptiles.

    Link to this
  42. 42. David Marjanović 11:16 am 01/9/2013

    ectodysplasin, SciAm is too stupid to implement blockquote. I use i.

    These animals had internal gills

    …Wow. I guess the evidence is embargoed?

    (BTW, “spiracle” is what the remnant of the 1st gill slit is called; it’s an entrance, not usually an exit, and Polypterus uses it to breathe air. The “spiracle” of tadpoles is not homologous.)

    Anyway, open gill slits would already be strong evidence that the “horns” stuck out of the body outline, and open gill slits wouldn’t be surprising at all. Aquatic salamanders with incomplete or no metamorphosis generally retain at least one.

    Finally, they picked pretty deep divergences (including the Alligatorinae-Caimaninae split) so their nodal estimates are interpolations, not extrapolations.

    That doesn’t help. Papers: Brochu 2004 a, b (both in Journal of Paleontology); Brochu 2006 (abstract from 2nd International Palaeontological Congress); me & Laurin 2007 (Systematic Biology).

    this paper seems to be following current standards for best practices

    Few papers indeed are better than that. I think there’s a recent one by San Mauro… anyway, perhaps the best is Pyron (2011), where the issue of calibration hardly even arises (different method).

    One is the timing of migration. The older books seem to assume that there was one a more-or-less uniform “boreal forest” that spanned the northern hemisphere, and that differential extinction in different regions has caused the patterns we see today. That’s great, except that the time of greatest boreal linkage was the PETM, and a lot has changed since then.

    The PETM was hot, and so was a later time in the early Eocene. Then it gradually got cooler, with a jump downwards around the beginning of the Oligocene and another upwards (though not back to PETM levels) around the beginning of the Miocene, leading to the “early Miocene optimum”, with very gradual cooling through the middle Miocene, stronger cooling in the late Miocene, and then the plunge into the current icehouse climate that we’re trying very hard to end now. It has long been obvious that a lot of migration across Beringia took place in the early and middle Miocene.

    As for giant salamanders, one cryptid is a giant salamander reputed to live in the northwest California. Anyone want to go looking for it?

    Are you sure that’s supposed to be a cryptobranchid and not just another Dicamptodon?

    Link to this
  43. 43. David Marjanović 11:25 am 01/9/2013

    Oh, look what I was finally reminded of looking up: it’s out now.

    Pronounced Peramorphosis in Lissamphibians—Aviturus exsecratus (Urodela, Cryptobranchidae) from the Paleocene–Eocene Thermal Maximum of Mongolia

    The phylogenetic position of any fossil Cenozoic cryptobranchid is still not understood; AFAIK, there’s never been a phylogenetic analysis… though apparently there’s one in press in the Journal of Vertebrate Paleontology. Looking forward to the March issue, then.

    We also have Cryptobranchus from the Oligocene of North America

    As of today, nobody knows (except maybe Vasilyan et al.) if that’s C., Andrias, or something else.

    In the case of Alligator, the sister taxon to the genus is already present by the late Paleocene

    What is that? And keep in mind that the sister-group of A. as a whole isn’t particularly interesting here – what we need is the sister-group of the crown-group of A., and that, IIRC, is the Miocene A. mcgrewi.

    Link to this
  44. 44. David Marjanović 12:34 pm 01/9/2013

    apparently there’s one in press in the Journal of Vertebrate Paleontology

    Yes, there is. And there’ll be much rejoicing.

    Link to this
  45. 45. ectodysplasin 1:10 pm 01/9/2013

    @David,

    I’m going off the classification of the North American cryptobranchids by Naylor (1981). Turns out cf. Cryptobranchus was in North America as early as the late Paleocene.

    Similar timing of dispersal between Eurasian and North American plethodontids has been suggested, by the way.

    The sister taxon to Alligator is Allognathosuchus, which is known from the Paleocene, including PETM deposits in the Arctic. The absence of definitive Alligator material prior to the Miocene isn’t terribly problematic when both morphological and molecular inference would place Alligator down into the Eocene.

    Thanks for the Aviturus paper, by the way.

    Link to this
  46. 46. ectodysplasin 1:31 pm 01/9/2013

    Speaking of JVP, the next few issues are going to be amphibialicious.

    Link to this
  47. 47. John Harshman 2:33 pm 01/9/2013

    The absence of definitive Alligator material prior to the Miocene isn’t terribly problematic when both morphological and molecular inference would place Alligator down into the Eocene.

    You are still missing the crucial distinction between total group and crown group. The proper fossil to demonstrate that a crown group exists must fall within the crown group. I’m also suspicious about calibrations of the crocodylian molecular tree. Compared to birds, crocodylians either evolve much more slowly or are very young. And birds are supposed to be the slow taxon.

    there’s some disagreement between the morphological and molecular trees

    Is there really, after all this time, still a controversy about the position of Gavialis?

    Link to this
  48. 48. Heteromeles 4:31 pm 01/9/2013

    The cryptid I was talking about is the “Trinity Alps Giant Salamander (Wikipedia link). It is a cryptobranchid, supposedly (ahem) the size of an alligator. Hasn’t been seen in a century or more, if it ever did exist.

    There are a number of cold streams in the Trinity Alps, but most of them have been dammed. Still, if you want to do some serious hiking in gorgeous country, I highly recommend the Trinities in July or August. They’re about 6-8 hours north of San Francisco.

    Link to this
  49. 49. ectodysplasin 4:50 pm 01/9/2013

    @John Harshman,
    You are still missing the crucial distinction between total group and crown group. The proper fossil to demonstrate that a crown group exists must fall within the crown group.

    Those are fighting words.

    That said, there’s no reason to believe that our sample of Alligator fossils is sufficiently good to stick confidence estimates on the first occurrence of crown Alligator. Either way, there’s a gap in the Alligator record spanning from the upper Paleocene straight into the Miocene. I see no reason to argue that this gap, which represents about 35 million years, wouldn’t include both crown and stem Alligator.

    I’m also suspicious about calibrations of the crocodylian molecular tree. Compared to birds, crocodylians either evolve much more slowly or are very young.

    Well, generation time is the obvious difference between birds and crocodylians. Most birds are sexually mature within a year. That is not the case for crocodylians.

    Is there really, after all this time, still a controversy about the position of Gavialis?

    I think that’s settling out a little bit, but there are still questions about which fossils are or are not gharials.

    Link to this
  50. 50. vdinets 6:51 pm 01/9/2013

    I tried looking for the Trinity Alps salamander a few times, with zero results. But finding Cryptobranchids is always difficult. I lived near the Smokies for two years and saw a hellbender only twice. On the other hand, the habitat in TA didn’t look right to me; those rivers are too fast and too prone to catastrophic floods, compared to places where I’ve seen hellbenders and Andrias.

    John: do you mean Gavialis or Tomistoma? Or both?

    Link to this
  51. 51. Allen Hazen 8:52 pm 01/9/2013

    Dartian (#39)–
    How could one ever forget Homo diluvii testis? I had him in the back of my mind when I posted, with the thought that he at least complicates the East Asia / Eastern North America pattern.

    If (I’m not betting on it!) the “Trinity Alps Giant Salamander” is real and is a Cryptobranchid, it would nicely fill a hole in the current distribution of the family: add Homo d.t. and you get a thoroughly Holarctic distribution!

    (Dicamptodon, in addition to being too small, doesn’t LOOK like a Hellbender. In general appearance it looks more like– what it is sometimes classified as being close to– an Ambystomid. Lives in the right part of the country, but my guess is that it isn’t the “Trinity Alps G.S.”)

    Link to this
  52. 52. John Harshman 8:55 pm 01/9/2013

    Those are fighting words.

    A simple matter of logic, I would have thought. A node must exist if you find a fossil descended from that node. If you find a fossil descended from a previous node, all you know is that the previous node exists. If the oldest known crown-group Alligator is Miocene, all you know is that crown-group Alligator is at least Miocene in age. It could be older. It could even be older than the oldest known stem-Alligator. But I don’t see a way to constrain the earliest date for that node without making many questionable assumptions.

    John: do you mean Gavialis or Tomistoma? Or both?

    The controversy, such as it is, is about Gavialis. It’s either the living sister group of all other crocodylians (morphological claim) or the sister of Tomistoma, with both the sister to crocodylids (the correct, oh, did I say that out loud?, I mean molecular claim). In either case, the position of Tomistoma is invariant. There is an implication from this that many fossil gavialids are not really gavialids. And perhaps some fossil tomistomids too.

    Link to this
  53. 53. David Marjanović 11:32 am 01/10/2013

    Hmmm. I’m still logged in on this computer. At home, SciAm has logged me out, and the usual error has occurred – clicking on “log in” leads to the top of the page, and clicking on the “log in” link there simply refreshes the page, so it’s impossible to log in.

    Gavialis: beginning to settle out, now that one of Brochu’s students has ontogenetic data from G. that make it more similar to Tomistoma. Presented at the SVP meeting 2 years ago.

    I’m going off the classification of the North American cryptobranchids by Naylor (1981).

    Yeah, don’t trust it.

    both morphological and molecular inference would place Alligator down into the Eocene

    Alligator, yes, but morphological evidence doesn’t put its crown-group into the Eocene. A. prenasalis (which is Oligocene) and, IIRC, A. mcgrewi are stem-members of A., not part of its crown-group.

    Speaking of JVP, the next few issues are going to be amphibialicious.

    *Homeric drool*

    Link to this
  54. 54. ectodysplasin 3:30 pm 01/10/2013

    @John

    A simple matter of logic, I would have thought. A node must exist if you find a fossil descended from that node. If you find a fossil descended from a previous node, all you know is that the previous node exists. If the oldest known crown-group Alligator is Miocene, all you know is that crown-group Alligator is at least Miocene in age. It could be older. It could even be older than the oldest known stem-Alligator. But I don’t see a way to constrain the earliest date for that node without making many questionable assumptions.

    Yes, this is the case, but the primary issue here is that there’s a large preservational gap where we don’t see anything from that group, stem or crown. With fossils, it’s always dangerous to assume that the first appearance of a taxon represents the absolute origin of said taxon. Basically, the molecules suggest that there should be crown-Alligator in that gap, the presence of the outgroup to Alligator in the Paleocene demands a ghost lineage of at least stem-Alligator into the Paleocene, and this seems to make the most biogeographical sense as well. Obviously there’s uncertainty involved, but I’d say there’s more uncertainty to be had by reading the fossil record literally. In fact, I’d say there are probably more questionable assumptions made by reading the fossil record literally (especially in patchy terrestrial systems).

    As for the Gavialis-Tomistoma thing, Brochu seems to be satisfied now with alligatorids being the most basal branch of extant crocodylians, and there’s additional accumulating morphological evidence from a variety of sources in favor of a monophyletic Gavialidae. I agree that a lot of the fossil longirostrine crocs attributed to the Tomistominae and Gavialinae by various workers are probably something else entirely, although I’ve only briefly dabbled in croc anatomy.

    Link to this
  55. 55. ectodysplasin 3:34 pm 01/10/2013

    @David,

    January JVP is out on T&F’s website. Only one amphibian paper in that one (a new salamandrid…relevant to your interests, I assume). March JVP is going to be chock full, though.

    Link to this
  56. 56. ectodysplasin 3:43 pm 01/10/2013

    @Allen,

    No living cryptobranchid suns itself on banks, which makes those accounts seem unlikely to represent salamanders. The fossil cryptobranchid Aviturus that was just described (David linked to it) seems to have been semi-terrestrial, but there’s no current evidence for such an animal in North America, and even then, terrestrial salamanders don’t sun themselves, as this dries out their skin. Large terrestrial salamander (e.g. Dicamptodon, Ambystoma) tend to live in holes in the ground, under logs, or in leaf litter in wet environments, as this protects them from dessication.

    Sounds more like an otter, perhaps?

    Link to this
  57. 57. Allen Hazen 1:33 am 01/11/2013

    Ectodysplasin–
    (#56) Good point about basking in sun. Alligators do, and if the Trinity Alps Giant Salamander is just a tall tale, alligator-like behaviour could easily have been included by the story teller!

    Link to this
  58. 58. naishd 8:46 am 01/11/2013

    In case anyone has forgotten, there’s a fairly lengthy Tet Zoo ver 2 on giant salamanders here, part of which is devoted to the alleged Trinity Alps salamander.

    Darren

    Link to this
  59. 59. David Marjanović 10:41 am 01/11/2013

    :-) I did indeed forget that it mentioned the cryptid and said so much about the fossils. In particular, I had forgotten about “Naylor’s (1981) lumping of all Andrias species into Cryptobranchus“. I remember the discussion about the European cryptids now, though.

    Given certain contacts, I have the Ichthyosaura randeckensis paper already; I’m using it right now to make a data matrix for putting our Oligocene mystery newt in.

    Basically, the molecules suggest that there should be crown-Alligator in that gap

    No, the molecules don’t suggest anything yet, because the only analysis has been too poorly calibrated to tell.

    Link to this
  60. 60. Metridia 5:25 pm 01/11/2013

    >ectodysplasin

    I can’t help but read this as “ecdysisplanin’”

    >Yes, tell us more about how we wouldn’t understand because we don’t molt our cuticle as we grow

    Link to this
  61. 61. David Marjanović 8:36 am 01/12/2013

    Signed in at home again!

    Link to this
  62. 62. Jenny Islander 4:06 pm 01/12/2013

    Okay, sensory pits are not an option–but what are those black spots all over the skull?

    Do they represent a different texture in the bone itself?

    If so, is it possible that the animals grew assorted bits on their heads that turned their outline from “Head, head, that’s a head, danger Will Robinson,” to “Lump with stuff growing on it, ignore?”

    Link to this
  63. 63. ectodysplasin 5:34 pm 01/12/2013

    The pits represent vascularization of the dermal tissue. Exactly why there’s more dermal ossification in early tetrapods is unclear, but likely is a holdover from fishes. On the other hand, Christine Janis has suggested that early tetrapods might be using calcium phosphate to buffer carbonic acid within their blood, but that requires a lot more testing.

    Link to this
  64. 64. David Marjanović 8:11 am 01/13/2013

    Metaplastic ossification of the dermis on top of dermal bones, causing surface sculpture, also occurs in various fully terrestrial dinosaurs; usually in osteoderms, but also on the skull of Majungasaurus.

    Link to this
  65. 65. ectodysplasin 8:22 am 01/14/2013

    Not entirely sure I’d call the more extensive dermal ossification seen in early tetrapods metaplastic bone formation. Metaplasia has a very specific (clinical) meaning that does not really apply to dermal bone sculpturing.

    Link to this
  66. 66. David Marjanović 9:52 am 01/14/2013

    Metaplasia = dermis turns directly into bone by deposition of apatite (as opposed to being deconstructed and replaced by bone). Apparently that’s how sculpture generally forms – which would explain why it can form on endochondral bones (like vertebrae, or the dorsally exposed sphenethmoid of Thaumastosaurus).

    I’m not talking about the large size of the dermal shoulder girdles of aquatic temnospondyls and nectrideans. Is that what you mean?

    Link to this
  67. 67. DavidMarjanovic 12:40 pm 05/30/2014

    I hope to submit a version for publication this month or next

    …uh, unfortunately, that’ll still be true next week. :-(

    Anyway, now that I know the literature better, I have to revise my statement about aïstopod tails: extremely long ones are limited to Phlegethontia and thereabouts, so they seem derived within the clade.

    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 Special Universe

Get the latest Special Collector's edition

Secrets of the Universe: Past, Present, Future

Order Now >

X

Email this Article

X