ADVERTISEMENT
  About the SA Blog Network













Tetrapod Zoology

Tetrapod Zoology


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

The troubling lack of Platyhystrix images online: the Tet Zoo Solution

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


Email   PrintPrint



Life reconstruction of Platyhystrix by Darren Naish, coloured by Gareth Monger. CC BY.

Regular readers will know that I’ve been doing my best over the last several years to get through the temnospondyls of the world. Temnospondyli, for the one or two or you that don’t know, is an enormous and substantially diverse clade of anamniotes (‘amphibians’) that was an important and persistent presence between the Early Carboniferous and the Jurassic. Relict forms hung on into the Cretaceous, and many experts argue that modern amphibians – the lissamphibians – are the direct descendants of one specific temnospondyl lineage, the dissorophoids of the Carboniferous and Permian. Dissorophoids have been covered (briefly) on Tet Zoo before (see links below).

Among the several dissorophoid lineages are the dissorophids of North America, Russia and China, most often associated with Cacops from the Permian of Texas. Dissorophids and their close relatives the trematopids form the dissorophoid clade Olsoniformes (Anderson et al. 2008). While trematopids may have been mostly cryptic terrestrial animals, dissorophids were reasonably large predatory temnospondyls that must have been patrolling open environments alongside contemporaneous amniotes. Indeed, they were ‘amniote-mimicking temnospondyls’ in some respects. When illustrating them, remember that they didn’t have the moist, naked skin we so often see in life reconstructions: temnospondyls were scaly, their bellies, flanks, dorsal surfaces and limbs being covered variously in spindle-shaped, ovoid and sub-circular scales (Witzmann 2007). The ovoid scales were arranged in overlapping rows. [UPDATE: be sure to read the comments below. The scales may not have been all that visible in many of these animals, since they were embedded within, or overlain by, the epidermis.]

Post-orbital section of the skull of Platyhystrix rugosus in left lateral view (from Berman et al. 1981). That big otic notch was occupied by a large ear drum. Unlike Cacops, Platyhystrix didn't possess a bony bar at the posterior border of the tympanum. Note gnarly skull bone texture.

Notable dissorophid features include a row of armour plates arranged along the dorsal midline and an obvious otic notch, sometimes enclosed by bone on all sides and almost certainly housing a large tympanum (an ear drum). Robust and well ossified limbs, a lack of lateral line canals, a tall, box-like skull shape and other features indicate that these animals were strongly terrestrial. Their big tympana (and slender stapes) indicate that dissorophids were detecting airborne sounds before contemporaneous amniotes convergently evolved the same features (Reisz et al. 2009). Were they listening for approaching predators, or using vocalisations when sending signals, or both? On twitter (@TetZoo) I happened to mention the fact that I had an urge to draw some temnospondyls. This inspired noted anamniote worker Jason Pardo to say “then I probably ought not to suggest drawing a male Platyhystrix calling for females with a fully-inflated vocal sac”, the idea behind his comment obviously being that these animals were indeed vocal and using their voices for sexual reasons. Hold that thought.

In Cacops and similar dissorophids, the median armour plates were fused to the tops of the neural spines. It’s often been suggested that these plates mostly functioned in defence, perhaps protecting these animals from the big predatory synapsids of the time, but suggestions that the plates served some sort of role in burrowing, that they reduced evaporative water loss, or that they reduced flexibility and stiffened the body have also been made (Dilkes & Brown 2007).

Skeletal reconstruction of Platyhystrix rugosus by Tracy Ford (the schematic skull shown here is essentially that of Cacops; the real Platyhytrix skull was somewhat different). Also, while many reconstructions of dissorophids show them with a five-fingered manus, I assume that they were four-fingered like other temnospondyls.

In some dissorophid taxa, long, laterally compressed neural spines evolved, their gnarly, rugose tips apparently representing the original armour scute. Hyper-elongate spines, presumed to have formed a sail-like structure superficially similar to the more familiar one on the Permian synapsid Dimetrodon, were present in Platyhystrix rugosus of the Lower Permian and Astreptorhachis ohioensis from the Upper Pennsylvanian. It’s been argued that these taxa should be separated from dissorophids and awarded their own ‘family’, the Platyhystricidae, but I don’t think this is useful given that there are intermediate forms with mid-length neural spines, such as some of the Aspidosaurus species. Anderson et al. (2008) found Dissorophidae to include distinct dissorophine and cacopine clades, with Ecolsonia cutlerensis being the sister-taxon to the clade formed by these two groups. Most recently, Schoch (2012) found Platyhystrix and Aspidosaurus to be outside the dissorophine + cacopine clade. This seemingly destroys (for now) the idea that Aspidosaurus is paraphyletic with respect to Platyhystrix: it might mean that tall neural species evolved more than once within dissorophids, or even – shock horror – that tall spines were primitive for the group, and lost in the ancestors of the dissorophine + cacopine clade.

Life reconstruction of Platyhystrix rugosus by Nobu Tamura, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license.

Incidentally, I’m guilty here of perpetuating the idea that the spines of these animals were separate structures, webbed by skin. This is how everyone draws Platyhystrix (see the Bakker illustration below, and the Nobu Tamura one shown here), but it might be incorrect: there may have been skin webbing between some of the spines, but others were in such close contact that there couldn’t have been a ‘web’ of the sort we always see. Note that in the Astreptorhachis spines shown below, the spine apices are tightly pressed together, and this might have been true for at least some of the Platyhystrix spines as well – the fossils really aren’t complete enough to tell.

We don’t really know what these animals did with their long neural spines. Contrary to what it says in some popular books there’s no indication that the spines were associated with the high degree of vascularisation that you might expect for a thermoregulatory role, and Vaughn (1971) thought that they probably initially evolved to provide mechanical support during terrestrial locomotion (see also Dilkes & Brown 2007). I’m not a fan of the argument that sails of this sort are best explained as special thermoregulatory structures (even for animals like Dimetrodon, the “sail was for thermoregulation” idea is dodgy and far less convincing than widely thought): based on analogy with extant animals, I think a role in display was more likely the primary driver behind the evolution of these structures, though that doesn’t mean that a role in thermoregulation, self-defence or even camouflage is ruled out.

Robert Bakker's 1986 illustration of a foraging Platyhystrix, captorhinids beware. From Bakker (1986); image (c) Rober Bakker.

My Platyhystrix again, brilliantly coloured by Tim Morris. Because several green versions had already been produced, I specifically asked for a non-green one. CC BY.

Anyway, the reason we’re here today is that I’ve been surprised by how few images there are online of Platyhystrix, one of the most bizarre and notable of all temnospondyls. Robert Bakker did one of the best reconstructions for his book The Dinosaur Heresies (Bakker 1986) and several post-1986 illustrations are clearly based on that one. I’ve mentioned the need for new Platyhystrix images on twitter and facebook, and in this article you can see the results. Thanks to Mike P. Taylor (of SV-POW) for the highly accurate and intricately detailed version he supplied, and thanks also to Tracy Ford, Ethan Kocak (of The Black Mudpuppy fame), Henrik Petersson and Maija Karala for the brilliant and innovative versions you can also see here. I did a version myself and colourised versions appear here courtesy of Tim Morris and Gareth Monger.

Platyhystrix by Ethan Kocak - yes, he of Black Mudpuppy fame. Used with permission. The claws are my fault: I initially put them on my drawing but later removed them.

On the subject of the life appearance of Platyhystrix, it should be noted to begin with that Platyhystrix is not known from fantastic material: those distinctive neural spines have been found on a few occasions, but the rest of the skeleton remains poorly known. Skull material has been discovered on several occasions, with the most complete specimen being the crushed and fragmented one described by Berman et al. (1981). This specimen (AMNH 11545: shown below) suggests that the skull was about similar in form to that of better known dissorophids, but that the otic notch wasn’t enclosed by bone posteriorly.

The most complete skull of Platyhystrix rugosus in left lateral view, from Berman et al. (1981). The snout is somewhat crushed, but you should get some idea of the deep, box-like form of the posterior region of the skull. Note the papillose texture.

Partially fused neural spines of the Platyhystrix relative Astreptorhachis ohioensis, from Vaughn (1971). (A) The incomplete bases of two spines; (B) tips of four spines in dorsal view; (C) the same four in left lateral view. Note the very gnarly, rugose texture.

Another peculiarity includes a so-called papillose or nodular dermal sculpturing: the surface of the skull is covered in low bumps and ridges that would presumably have given the live animal a knobbly skin texture. A similar texture is present on the neural spines and also on the ribs, suggesting that much of the animal’s dorsal surface was similarly knobbly. The presence of this nodular ornament isn’t unique to Platyhystrix, but it is unusually extensive in this taxon, being uniquely present across the cheek and orbital regions of the skull.

Platyhystrix also possesses a surprisingly high tooth count: it has as many as 65 simple, peg-like teeth on each side of the upper jaw. This isn’t unique within dissorophids (there are taxa with even higher numbers), but it certainly makes you wonder what, and how, these animals were grabbing and eating.

Platyhystrix again, image courtesy of Maija Karala (used with permission).

Note also that Platyhystrix is pretty big: the AMNH 11545 skull is over 19 cm long along the midline. This size is not exceptional for the species, with some fragments of skull coming from even larger animals (Berman et al. 1981). In Cacops and most other dissorophids the skull is usually about 12-14 cm long, so Platyhystrix might have been a large member of the group if it was proportioned like those other taxa (in Cacops, the skull is about 33% of total length, so a complete Platyhystrix might have been about 60 cm long). Fayella may have been Platyhystrix-like in size (Olson 1972).

TMNT Platyhystrix, but with the second 'T' standing for temnospondyl! Image by Henrik Petersson, used with permission. I hope that we'lll end up a full squad of TMNT temnos... that's Raphael done, Donatello next?

So, with this impressive and wonderful assortment of Platyhystrix images now out there and online, let’s hope that we see much more of this amazing temnospondyl in future. And much more on temnospondyls to come here… in time, of course. Oh, I nearly forgot the best image – the one provided by Mike Taylor. Here you go…

Mike Taylor's painstakingly accurate, hours-in-the-making, high-fidelity Platyhystrix life reconstruction. Thanks, Mike! As per Jason Pardo's suggestion (see above), this is clearly a vocalising male, vocal sac extended.

For previous Tet Zoo articles on temnospondyls, see…

Many, many thanks to everyone who played along, and provided or produced the Platyhystrix images you see here. Let's do this again some time.

Refs – -

Anderson, J. S., Henrici, A. C., Sumida, S. S., Martens, T. & Berman, D. S. 2008. Georgenthalia clavinasica, a new genus and species of dissorophoid temnospondyl from the Eearly Permian of Germany, and the relationships of the family Amphibamidae. Journal of Vertebrate Paleontology 28, 61-75.

Bakker, R. T. 1986. The Dinosaur Heresies. Penguin Books, London.

Berman, D. S., Reisz, R. R. & Fracasso, M. A. 1981. Skull of the Lower Permian dissorophid amphibian Platyhystrix rugosus. Annals of Carnegie Museum 50, 391-416.

Dilkes, D. & Brown, L. E. 2007. Biomechanics of the vertebrae and associated osteoderms of the Early Permian amphibian Cacops aspidephorus. Journal of Zoology 271, 396-407.

Olson, E. C. 1972. Fayella chichashaensis, the Dissorophoidea and the Permian terrestrial radiation. Journal of Paleontology 46, 104-114.

Reisz, R. R., Schoch, R. R. & Anderson, J. S. 2009. The armoured dissorophid Cacops from the Early Permian of Oklahoma and the exploitation of the terrestrial realm by amphibians. Naturwissenschaften 96, 789-796.

Schoch, R. R. 2012. Character distribution and phylogeny of the dissorophid temnospondyls. Fossil Record 15, 121-137.

Vaughn, P. P. 1971. A Platyhystrix-like amphibian with fused vertebrae, from the Upper Pennsylvanian of Ohio. Journal of Paleontology 45, 464-469.

Witzmann, F. 2007. The evolution of the scalation pattern in temnospondyl amphibians. Zoological Journal of the Linnean Society 150, 815-834.

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 91 Comments

Add Comment
  1. 1. Dartian 11:52 am 09/22/2013

    I happened to mention the fact that I had an urge to draw some temnospondyls.

    We all get that urge from time to time, man, it’s perfectly normal. Most of us can resist it though… ;)

    BTW, wonder if this blog post will be the future reference point for some Chinese manufacturer of plastic toy animals who’ll decide to add Platyhystrix to their extinct species lineup? ;)

    the “sail was for thermoregulation” idea is dodgy and far less convincing than widely thought

    Glad to read that. Personally, I’ve always been a bit uncomfortable with that general idea.

    Link to this
  2. 2. Cai Yuan 12:11 pm 09/22/2013

    There is a plastic toy Platyhystrix, but it’s just about impossible to find:
    http://www.collectorsquest.com/collectible/43428/play-visions-platyhystrix

    It was part of the rare and much sought-after Playvisions fossil amphibians line, all based on illustrations from the Macmillan Book of Prehistoric Life (one of the few Platyhystrix illustrations not featured in this article).

    Link to this
  3. 3. Dartian 1:00 pm 09/22/2013

    Cai Yuan:
    There is a plastic toy Platyhystrix

    Wow, great find!

    By coincidence, just the other day I was thinking how utterly shitty almost all plastic toy dinosaurs used to be 30-40 years ago. (Here are a couple of fairly typical examples.) Kids of today don’t know how lucky they are…

    Link to this
  4. 4. Heteromeles 7:54 pm 09/22/2013

    Neat pictures!

    Just randomly brain storming, here’s a random idea for what the sails were for: advertising.

    The thing that makes this crazy idea different isn’t what they were for, it’s why animals now can’t have them: parasitic flies, lice, ticks, etc. I’m not sure when all our least favorite dermal pests evolved, but it might be that Platyhystrix, Dimetrodon, and others independently evolved sails, not because some kryptonite-bearing meteor or something crashed into the Earth and mutated them into sailbacks, but because back then, running up a big ol’ vascularized sail did not mean getting bled to death by a bunch of hungry bugs looking for smorgasbord. Now all these pests do exist, which prevents modern animals from evolving in this direction again.

    Link to this
  5. 5. ectodysplasin 8:47 pm 09/22/2013

    Something to be aware of is that the “sails” all had very different biomechanical properties. Adam Huttenlocker has written a number of papers looking at this in early synapsids from a paleohistological perspective, which I’d highly recommend for anyone interested in these structures.

    The “sail” in platyhystricines is somewhat different from what you see in synapsids, however, and has a pretty distinct evolutionary history. Dissorophids in general have heavily-built dorsal dermal armor, which likely primarily served a biomechanical role (see some of David Dilkes’s work on the subject) although it seems reasonable to assume that some dissorophids, such as Dissorophus and some species of “Broiliellus” extrapolated this into what may have been effective armor. In Playhystrix and Astreptorhachis, however, the dorsal extrapolation probably completely immobilized the thoracic region, preventing flexion of the vertebral column in any direction.

    In addition, the dorsal “sail” seems to have an interesting ontogenetic component in platyhystricines as far as we understand them. In some populations of Aspidosaurus, the dorsal armor starts out low and flat along the neural spines and then through ontogeny elongates into a low(ish) sail. There are a few interesting juvenile to subadult spines of Aspidosaurus showing this ontogeny in the American Museum of Natural History collections.

    As for sailbacks, there are a range of taxa out there that have tried out the sailback thing. Some stem-crocodiles in the Triassic (i.e. the ctenosauriscids) experimented with some form of sailbacking, as did several dinosaur lineages (Ouranosaurus, spinosaurids, Amargosaurus, Acrocanthosaurus). Several lizard taxa develop sailbacks today, including a number of chameleonids, several species of the corytophanid Basiliscus, the agamid Hydrosaurus, and some species of Anolis (e.g. A cristatellus) so it’s clearly not restricted to a single temporal interval.

    As for vascularization, there’s no reason to believe that the sail in Platyhystrix was particularly heavily vascularized.

    Link to this
  6. 6. ectodysplasin 9:09 pm 09/22/2013

    @Darren,

    It’s been argued that these taxa should be separated from dissorophids and awarded their own ‘family’, the Platyhystricidae, but I don’t think this is useful given that there are intermediate forms with mid-length neural spines, such as some of the Aspidosaurus species.

    Well, there are some significant differences between Aspidosaurus + Platyhystrix + Astreptorhachis and the remainder of dissorophids. Armor-wise, the construction differs relatively significantly; in dissorophines and cacopines, there are two layers of median dorsal osteoderms: a knoblike structure fused to the neural spine with a thin strap of dermal armor, and a second free platelike ossification between the neural spines that articulates anteriorly and posteriorly with the neural spine elements. In “Aspidosaurus” and other platyhystricines, there’s only a single layer of dorsal armor fused extensively to the neural arch, with a second set of armor plates fused laterally to the ribs. There are also a variety of cranial characteristics that differentiate known platyhystricines from the remainder of all dissorophids, which all seem to look much more Cacops-like than previously thought. The otic notch of platyhystricines also doesn’t make a whole lot of sense in the context of other dissorophids or even animals like Ecolsonia and Actiobates. Given that non-dissorophid dissorophoids like Ecolsonia and Actiobates were rather heavily armored.

    Also, am I really “noted”?

    Link to this
  7. 7. keesey@gmail.com 12:24 am 09/23/2013

    CC-BY, you say?

    Yoink! http://phylopic.org/image/618f8a5e-df37-4947-8224-bec3c81f72ad/

    Link to this
  8. 8. naishd 5:08 am 09/23/2013

    Thanks for comments so far, much appreciated.

    Re: what Heteromeles says in comment # 4: my immediate response is that dorsal sails of the sort seen in those Permian animals _are_ present in extant taxa, but ectodysplasin has already said that in comment # 5. In fact, the presence of tall dorsal and/or caudal sails in various extant lizards is part of the reason why I think a sociosexual display function is more likely than the cherised thermoregulatory hypothesis (there are other reasons too.. I might elaborate in future).

    ectodysplasin: well, you’re noted now :) And, yes, I fully agree that the dissorophid sails are very different from those of synapsids – maybe I should have made this clearer in the article. As for whether we should be thinking of ‘platyhystricines’/platyhystricids as their own little unit, I was (as noted in the article) mostly inspired by phylogenies that don’t group the taxa together — Schoch and others found these animals to be a grade — but bring on new topologies that show otherwise.

    As for comment 7… I’d like to make all of my own illustrations and photos CC-BY: it didn’t occur to me until recently how easy it is to do this (viz, just put ‘CC BY’ next to the image!).

    Darren

    Link to this
  9. 9. ectodysplasin 11:42 am 09/23/2013

    @ darren

    As for whether we should be thinking of ‘platyhystricines’/platyhystricids as their own little unit, I was (as noted in the article) mostly inspired by phylogenies that don’t group the taxa together — Schoch and others found these animals to be a grade — but bring on new topologies that show otherwise.

    For sure. There are a few key issues to keep in mind, however:

    1. Many, many, many dissorophoids are currently under revision. This includes animals previously reported as dissorophids that clearly are not (e.g. “Broiliellus” hektatopos), actual dissorophids (Cacops aspidophorus, Fayella chickashaensis, etc) and various other animals relevant to the question of dissorophid phylogeny.

    2. There’ve been a ton of new dissorophid taxa described in recent years and we’re still trying to understand all this new information. Two new species just hit print last week, and another hit the press a little earlier this year.

    3. We don’t understand other large dissorophoids very well either. What, for example, is Ecolsonia? It is heavily armored and has an enclosed otic notch, but it’s also got that enlarged keyhole-shaped naris characteristic of trematopids. Actiobates has never been adequately described, despite often being recognized as a relatively basal trematopid.

    4. For that matter, we don’t understand small dissorophoids that well, either. Nearly half of amphibamid diversity currently known has been described since 2007 (Plemmyradytes shintoni in 2007, Georgenthalia clavinasica, Pasawioops mayi, and Gerobatrachus hottoni in 2008, Rubeostratilia texensis in 2011, and Tersomius dolesensis this year) and we’re still also trying to resolve whether branchiosaurids are or are not a clade within amphibamids. Some recent phylogenetic treatments have found dissorophids within the Amphibamidae, so amphibamid phylogeny is somewhat directly relevant to how we understand dissorophid phylogeny as well.

    Lots to do in the next few years!

    Link to this
  10. 10. Jerzy v. 3.0. 1:11 pm 09/23/2013

    Cool pictures! Did these guys run around on vertical legs?

    Link to this
  11. 11. David Marjanović 2:16 pm 09/23/2013

    YAY TEMNOSPONDYLS YAY ^_^

    When illustrating them, remember that they didn’t have the moist, naked skin we so often see in life reconstructions: temnospondyls were scaly, their bellies, flanks, dorsal surfaces and limbs being covered variously in spindle-shaped, ovoid and sub-circular scales (Witzmann 2007).

    Uh, but those scales are in the skin. They’re fish scales, not lizard scales; with those, they have nothing in common except the word “scale”. They’re bone plates in the dermis. The epidermis above them was either naked and moist, or perhaps naked and warty as seen in terrestrial salamanders…

    Their big tympana (and slender stapes)

    I’m pretty sure the plural of stapes is stapedes… and the first temnospondyl described with an actually slender stapes shaft, as opposed to a pillar or a plank, is Broiliellus reiszi, published online-early a few days ago (Holmes et al. 2013). (It looks a lot like Conraua goliath in caudal view.)

    …Oh, it’s in the Carnegie Museum. Now I need to go there again.

    Holmes, R., Berman, D. S. [sic] & Anderson, J. S. (2013): A new dissorophid (Temnospondyli, Dissorophoidea) from the Early Permian of New Mexico (United States). Comptes Rendus Palevol [Reisz-Festschrift]: 17 pp.

    several dinosaur lineages ([...] Amargosaurus

    Amargasaurus, named after some place called La Amarga.

    Conversely, the -a- in Plemmyradytes comes out of nowhere and suggests negation of the actually intended meaning. It should have been -o-. But it’s too late to fix that.

    The otic notch of platyhystricines also doesn’t make a whole lot of sense in the context of other dissorophids or even animals like Ecolsonia and Actiobates.

    …Why?

    Also, am I really “noted”?

    If you aren’t, you will be soon. I had a look in the SVP meeting abstract volume. Notochordal braincase, holy adsymphysials, Batman! :-)

    (e.g. “Broiliellus” hektatopos)

    hektotopos

    Did these guys run around on vertical legs?

    Nope.

    Link to this
  12. 12. vdinets 3:18 pm 09/23/2013

    Looks like the sail was very rigid. Does it mean that every time the animal accidentally hit a branch or something with this thing, the shock was transmitted straight to the spinal cord? Also, the sail looks like an enlarged version of what bison have on their spines. Is there a possibility that the structure wasn’t flat, but instead the bones were supporting a hump of soft tissue, used, for example, for fat or water storage?

    Of course, if the sail was for display purposes, that implies having a lateral display, like in many lizards. I wonder why nobody has drawn it – should be really cool.

    Link to this
  13. 13. Heteromeles 4:01 pm 09/23/2013

    @6 and 8: there are modern tetrapods with a sail the size of Platyhystrix or Dimetrodon? That’s the kind of thing I’m talking about–size, not just presence. If you want to tack a sail on Spinosaurus, that’s fine, but consider tacking one onto Bison occidentalis while you’re at it (as vdinets noted).

    Link to this
  14. 14. naishd 4:50 pm 09/23/2013

    Many, many thanks for the excellent and informative comments, especially ectodysplasin (comment # 9). That comment helps bring home the point that we’re in the middle (or… in the early stages) of an anmniote revolution, so much happening and hot new research being published. With respect to ‘platyhyricines’, will be interesting to see how the phylogenies evolve.

    On skin texture, David Marjanović says…

    “Uh, but those scales are in the skin. They’re fish scales, not lizard scales; with those, they have nothing in common except the word “scale”. They’re bone plates in the dermis. The epidermis above them was either naked and moist, or perhaps naked and warty as seen in terrestrial salamanders…”

    I thought that there was a third possibility (for the thin scales of Mesozoic temnospondyls at least): that the scales were externally visible beneath a thin dermis, as they are in many fish. The gastral scales and the often small, non-overlapping, non-abutting scales of some taxa were evidently buried deep in the dermis (as confirmed not only by histology but also by the integumentary pattern preserved in Melanerpeton), but I don’t think this was true for taxa with thin, overlapping, fish-like scales… I assume that they’d be visible in the live animal. Thoughts?

    Darren

    Link to this
  15. 15. ectodysplasin 5:49 pm 09/23/2013

    @heteromeles:

    @6 and 8: there are modern tetrapods with a sail the size of Platyhystrix or Dimetrodon? That’s the kind of thing I’m talking about–size, not just presence. If you want to tack a sail on Spinosaurus, that’s fine, but consider tacking one onto Bison occidentalis while you’re at it (as vdinets noted).

    I’d suggest going over to google image search and looking for pictures of Basiliscus plumifrons, Hydrosaurus, and Trioceros cristatus.

    As for the Spinosaurus/Bison comparison, there’s a really easy way to test this: cut up some neural spines and look for the dorsalmost extent of Sharpe’s fibers. Until you do that, the arguments from analogy are entirely ambiguous.

    Link to this
  16. 16. ectodysplasin 6:13 pm 09/23/2013

    As for dermal scales, there’s a whoooole lot of variation in dermal scale and scute morphology and tissue types. “Ganoid” scales in animals like gar and Polypterus are pretty conspicuous, so similar scales seen in many erly tetrapods (temnospondyls, anthracosaurs, stem-tetrapods, etc) were probably pretty obvious, externally, with only a thin layer of soft tissue outside of them.

    Finer dermal scales, such as those seen in some dissorophoids, some microsaurs, etc were probably more similar to those seen in caecilians.

    Large dermal ossicles like those seen in dissorophoids, diplocaulids, and chroniosuchids were probably similar to the dorsal shields seen in modern amphibians like Ceratophrys or Tylototriton, or possibly more akin to that seen in modern crocodiles (at least in chroniosuchids). This armor in those species is conspicuous but not scaly.

    Link to this
  17. 17. ectodysplasin 7:45 pm 09/23/2013

    @vdinets:

    Looks like the sail was very rigid. Does it mean that every time the animal accidentally hit a branch or something with this thing, the shock was transmitted straight to the spinal cord? Also, the sail looks like an enlarged version of what bison have on their spines. Is there a possibility that the structure wasn’t flat, but instead the bones were supporting a hump of soft tissue, used, for example, for fat or water storage?

    In Astreptorhachis the whole structure is one massively co-ossified dorsal plate, so yes, this whole thing was completely rigid.

    There’s also no way that this structure was the basis for a hump for fat or water storage. The tuberculate ornament on the surface of the bone is all dermal bone, meaning that there would have been very little soft tissue outside of it, and it would have been essentially all epidermal.

    Sails in other Paleozoic forms also seem to be true sails instead of “humps.” Muscle attachment in sail-backed synapsids only extended a short distance up the spine; Sphenacodon likely had a low to medium-length crest supported by the neural spines and the sail of Dimetrodon extended significantly beyond the epaxial musculature. In Dimetrodon specifically, these sails were frequently sites of bone trauma, and you see numerous cases of compound fractures in the neural spines (see: Rega et al., 2012).

    Link to this
  18. 18. Heteromeles 7:56 pm 09/23/2013

    @15: I did, in fact, google all the images you suggested before posting that previous comment. And Dimetrodon and Platyhystrix. Presumably you have as well? My apologies for arguing with someone who’s so much better informed than I possibly could be, all sarcasm definitely meant.

    Link to this
  19. 19. Heteromeles 8:03 pm 09/23/2013

    I should point out, that, after that previous rant, that what we don’t see these days are any large tetrapods with sails that are >1 body diameter tall. We see numerous lizards with small sails (as you pointed out), but nothing built like Dimetrodon, Edaphosaurus, or Platyhystrix where the sail is 2-3 times as tall as the trunk is wide. There’s no obvious, structural reason why such a sail is impossible currently, but it doesn’t occur now, even though it occurred in multiple, unrelated lineages back in the Carboniferous. The reason for this is…?

    Link to this
  20. 20. ectodysplasin 8:16 pm 09/23/2013

    I misread your post thinking you were asking whether the size of the sail in proportion to the body size of the animal was equivalent between fossil taxa and modern analogues, so I was suggesting you look at them for yourself. In terms of absolute body size, Hydrosaurus is well within the size range (>30 cm SVL, > 1m TL) of some of the sailbacked synapsids (e.g. Dimetrodon teutonis, <Dimetrodon occidentalis, Ianthasaurus haresti, etc.) as well as sailbacked dissorophids like Platyhystrix.

    Once again, though, if you want to argue for a muscular hump, there are paleohistological studies you can do to test that hypothesis. These have been done in sailbacked synapsids, and there was no hump in these species. I believe these have been done in some ctenosauriscids (and will show up in print eventually, I’m sure) and, once again, the epaxial muscles are relatively restricted to the base of the spines, so once again, no hump. So it seems that in most of these taxa, the sail is an actual sail. If you want to argue otherwise, cut up some neural spines and report the results.

    Link to this
  21. 21. ectodysplasin 8:48 pm 09/23/2013

    @heteromeles:

    that what we don’t see these days are any large tetrapods with sails that are >1 body diameter tall. We see numerous lizards with small sails (as you pointed out), but nothing built like Dimetrodon, Edaphosaurus, or Platyhystrix where the sail is 2-3 times as tall as the trunk is wide. There’s no obvious, structural reason why such a sail is impossible currently, but it doesn’t occur now, even though it occurred in multiple, unrelated lineages back in the Carboniferous. The reason for this is…?

    The sail in Hydrosaurus approaches the sizes you’re talking about.

    Furthermore, there are plenty of fossil taxa within these groups which had lower sails far within the range seen in modern taxa. Sphenacodon and Ctenospondylus, for example, all have low but definitive sails. This is probably the case for Xyrospondylus as well. In dissorophids, the “sail” of Aspidosaurus was a relatively low crest. In ctenosauriscids, Lotosaurus had a relatively low crest. Within spinosaurids, a number of species seem to have much lower sails. And so on.

    We could ask other questions in reverse, of course. Basiliscus, a corytophanid, Trioceros quadricornis, a chameleonid, and Hydrosaurus, an agamid, show sails on the tail as well as the thorax, but no fossil sailbacks do. Why is that? Edaphosaurus spines have bizarre lateral bony projections, but no other sailback duplicates this morphology. Why is that? And so on. I’m not sure any of these questions are really answerable, however, because there’s no real way to constrain the hypotheses and test them rigorously when the sample size of sailbacked tetrapods is so small.

    What we can say is that there are probably locomotory conditions which preclude the evolution of a sailback. Animals which rely on significant spinal flexion during locomotion don’t develop sailbacks. This is why this is rare in early tetrapods in general (with dissorophids as the exception that proves the rule), absent entirely in therapsids, and absent in almost all lizard groups (except iguanians, which rely much more on their legs and much less on lateral flexion of the spinal column for locomotion). If the sail is a sexual display structure (as it is in most lizards, although sailbacked chameleons may in fact use the sails for thermoregulation), then it’s a costly display to produce and one which could significantly compromise locomotory performance, and will only crop up in taxa which have already more or less immobilized the trunk during locomotion anyways. It’s not surprising that it’s a rare feature.

    Link to this
  22. 22. Dartian 1:52 am 09/24/2013

    ectodysplasin:
    I’d suggest going over to google image search and looking for pictures of Basiliscus plumifrons, Hydrosaurus, and Trioceros cristatus.”

    It’s easy to find pictures of living animals, but I’m yet to see a good illustration of the skeletons of any of those lizards. If basilisks and Hydrosaurus really are so good analogues to Dimetrodon as you suggest, then why does it seem that nobody has published a decent skeletal side-by-side comparison of them?

    Link to this
  23. 23. ectodysplasin 2:43 am 09/24/2013

    @Dartian

    There are abundant pictures of living animals because these animals are both very photogenic and commonly traded on the exotic pet market. They’re poorly figured because the skeletons of most animals are poorly figured in general. In addition, much skeletonized material is macerated rather than dermestid-prepared, meaning that you’re unlikely to get a good articulated vertebral column.

    Furthermore, there is little scientific interest in publishing a figure of the kind you describe simply because the folks who have done recent work on hyperelongate spines have been more interested in the histology of these structures rather than gross morphology, because gross morphology has been generally ineffective in furthering our understanding of these structures.

    I happen to have a (poor) photo of the axial skeleton of Basiliscus plumifrons on hand, though. You’ll notice that it is grossly similar in general morphology to what you see in Dimetrodon and other sailbacked tetrapods. You can see a series of tendons along the back near the dorsal extent of the epaxial musculature, as well, so you can see exactly how much of that is sail and how much is epaxial musculature.

    Link to this
  24. 24. ectodysplasin 3:32 am 09/24/2013

    One final swing at the “humps-not-sails” argument, and then I’m heading to sleep (for real, really). The elongate neural spines in Bison show a pretty clear vertex immediately dorsal to the scapula. The reason for this is that the muscle attachments of note are not those for the erector spinae, but rather the attachments for the acromiotrapezius and spinotrapezius muscles. In contrast, the hyperelongate spines seen in spinosaurids, ctenosauriscids, and so on are not centered around the pectoral girdle. Furthermore, these animals belong to a lineage with a very different construction of the pectoral girdle, and not only lack the trapezius morphology seen in mammals, but essentially lack this muscle entirely. Not only that, but archosaurs do not use the pectoral skeleton for locomotory support in many/most cases, and certainly not in the case of the sailbacked taxa.

    Meanwhile in the camelids, no specialized skeleton is present within the hump.

    Bailey (1997) essentially ignored all of this in his papers proposing hunchbacked spinosaurids. So, sure, Bailey raised some important points, but the value of his criticisms in understanding the actual biology of these structures is strongly overstated. Skepticism is a good thing, but one must be careful to apply skepticism equally to alternative hypotheses as well as dominant hypotheses.

    Link to this
  25. 25. Dartian 3:45 am 09/24/2013

    ectodysplasin:
    skeletons of most animals are poorly figured in general

    Surely there has to be a Basiliscus or a Hydrosaurus skeleton on display in some natural history museum somewhere in the world?

    there is little scientific interest in publishing a figure of the kind you describe simply because the folks who have done recent work on hyperelongate spines have been more interested in the histology of these structures rather than gross morphology, because gross morphology has been generally ineffective in furthering our understanding of these structures.

    I don’t follow your logic. If no researcher has ever made a proper gross morphological comparison between these fossil and extant ‘finbacks’, then how do you know that such comparisons wouldn’t further our understanding of these structures? Are you just making an assumption?

    I happen to have a (poor) photo of the axial skeleton of Basiliscus plumifrons on hand

    I can’t access that picture. :(

    Link to this
  26. 26. Dartian 3:56 am 09/24/2013

    ectodysplasin:
    Bailey (1997)

    Bailey had this to say about Dimetrodon‘s neural spines (p. 1126):
    Morphologically, they resemble spines on the backs or tails of a few modern iguanid and agamid lizards (Basiliscus, Lophura, and Hydrosaurus).

    But he offered no evidence for, or further comments on, that vague statement. Is there any real basis at all in the literature for the claim that modern-day fin-backed lizards are truly comparable to Dimetrodon and its kin?

    Reference:
    Bailey, J.B. 1997. Neural spine elongation in dinosaurs: sailbacks or buffalo-backs? Journal of Paleontology 71, 1124-1146.

    Link to this
  27. 27. naishd 4:04 am 09/24/2013

    Excellent discussion, and I’m with ectodysplasin on many of the points he makes above: it’s true, extant sailbacked taxa – hardly ever mentioned whenever palaeontologists discuss the form and function of sails in fossil taxa (I think mostly because the palaeontologists concerned don’t know all that much about the diversity of extant animals) – (1) are extremely similar in neural spine form to fossil sailbacked taxa, and (2) are just about never figured, nor on display in museums or teaching collections.

    I’ve long been meaning to do something with the images I’ve collected: I have photos of the neural spines of a tall-spined chameleon skeleton I have access to (not sure which species it is, actually), and the spines are long, subrectangular (with longest axis being vertical, obviously), and essentially like miniature Spinosaurus spines. I also have photos of a Hydrosaurus being dissected, and the spines that support its caudal sail are tall, slender rods, highly similar to the spines of Dimetrodon and similar taxa. I’ve been planning for several years now to write a paper on this – basically saying that living reptiles support the ideas of sails, not humps (contra Bailey), in spinosaurids (I don’t think anyone doubts that sails were present in the sailbacked synapsids, nor is there any reasonable doubt about the sails in the dissorophoids). Like so many planned papers, I have yet to find the time for it.

    I have alluded to some of this stuff before: see this 2010 article on the sail-backed theropod Concavenator.

    Darren

    Link to this
  28. 28. naishd 4:13 am 09/24/2013

    Just saw Dartian’s comment # 26. Am especially interested in your comment: “Is there any real basis at all in the literature for the claim that modern-day fin-backed lizards are truly comparable to Dimetrodon and its kin?”

    Based on the literature, no, there isn’t, because extant lizards are mostly very poorly described in the technical literature – they’re very little out there, certainly nothing like good osteological descriptions of basilisks, sailfins water-dragons and so on. I mean, herpetologists have had to do crazy things to get osteological data published, like translate old German works from the 1930s (I’m referring to Russell et al.’s re-publication of Wellborn (1997), originally published in 1933).

    Wellborn, V. 1997. Comparative osteological examinations of geckonids, eublepharids and uroplatids. Breck Bartholomew, Bibliomania! (Logan, UT).

    Darren

    Link to this
  29. 29. Hai~Ren 5:12 am 09/24/2013

    Very interesting discussion. Given that basilisks and water dragons appear to be fairly well-represented in the exotic reptile trade, I hope those with recently deceased pet donate their specimens.

    By the way, this reminded me of Darren’s article about eagles:

    Bye-bye Bambi: Eagle snags deer in ambitious attack
    Eagle vs Deer: Rare Images Caught on Camera

    Basically, a camera trap set up to monitor Amur tigers in Russia ended up taking photos of a golden eagle attacking a sika deer.

    Link to this
  30. 30. naishd 5:19 am 09/24/2013

    Re comment # 29, the deer vs eagle news was first announced in a technical paper that came out a few weeks ago…

    Kerley, L. L. & Slaght, J. C. 2013. First documented predation of Sika deer (Cervus nippon) by Golden eagle (Aquila chrysaetos) in Russian Far East. Journal of Raptor Research 47, 328-330.

    I got this as soon as it came out (thanks to Glyn Young) and did think about blogging it. Missed a chance there. Much as I have with my own Microraptor work…

    Darren

    Link to this
  31. 31. Dartian 6:11 am 09/24/2013

    Darren:
    herpetologists have had to do crazy things to get osteological data published, like translate old German works from the 1930s

    Well, I’ve done even crazier things: I have actually read old zoological papers – in German.

    Incidentally, I don’t think that the point that Heteromeles made in comment #4 should be casually dismissed. It’s good to keep in mind that species don’t exist in an ecological vacuum; they are – and were – always interacting with other organisms in various ways. It may very well indeed be that, compared with today, the Permian parasite community was different in ecologically significant ways. For example, blood-sucking dipteran flies and mosquitoes didn’t exist back then.

    Link to this
  32. 32. David Marjanović 6:18 am 09/24/2013

    I thought that there was a third possibility (for the thin scales of Mesozoic temnospondyls at least): that the scales were externally visible beneath a thin dermis, as they are in many fish.

    Uh, yes, of course – sorry for not making that clearer. My point was that you made it sound like the scales were directly exposed to the air (or water), protecting the underlying skin, when in fact the epidermis lay unprotected on their outside.

    Ganoid and cosmoid scales – don’t they erupt like teeth? But tetrapod-and-similar scales never have such a layer, they’re reduced to the bone layer like teleost scales.

    As for the Spinosaurus/Bison comparison, there’s a really easy way to test this: cut up some neural spines and look for the dorsalmost extent of Sharpe[y]’s fibers.

    All known axial and appendicular bones of Spinosaurus were turned to dust when Munich was bombed in WWII.

    Finer dermal scales, such as those seen in some dissorophoids, some microsaurs, etc were probably more similar to those seen in caecilians.

    Caecilian scales aren’t even ossified in one piece, they’re just little concentric arch segments of bone. Among extinct animals that are preserved well enough, only Tuditanus and Platyrhinops get anywhere close, and their scales are still ossified as a whole, AFAIK.

    [Extant sailbacks] show sails on the tail as well as the thorax, but no fossil sailbacks do. Why is that?

    It did apparently extend to the proximal part of the tail in spinosaurids.

    It’s easy to find pictures of living animals, but I’m yet to see a good illustration of the skeletons of any of those lizards. If basilisks and Hydrosaurus really are so good analogues to Dimetrodon as you suggest, then why does it seem that nobody has published a decent skeletal side-by-side comparison of them?

    Hold me before I embark on a very ranty rant about neontologists and their tradition of completely ignoring the skeleton except for the occasional line drawing of a skull. I’ve been working on the biggest morphological phylogeny of salamandrids ever, ultimately to determine the position of a fossil in the collection here – I’ve already merged all published matrices, and it’s ridiculously hard to fill in the gaps or find additional characters. Ridiculously. (I’m not sure if I’ve checked Wiedersheim eighteen seventy-seven yet, but even the 1930s literature is next to useless.) This extends all the way to birds, where the tradition of collectors is to preserve the skin with the feathers and throw the rest away, so that Livezey & Zusi (2007) hardly managed to progress beyond Wetmore (1960)!

    I happen to have a (poor) photo of the axial skeleton of Basiliscus plumifrons on hand, though

    “Error (403)
    It seems you don’t belong here! You should probably sign in [link]. Check out our Help Center [link] and forums [link] for help, or head back to home [link].”

    Surely there has to be a Basiliscus or a Hydrosaurus skeleton on display in some natural history museum somewhere in the world?

    Do you know a museum that puts skeletons of extant non-cetaceans on display, other than the comparative-anatomy part of the MNHN in Paris which of course exhibits mostly large mammals?

    and did think about blogging it. Missed a chance there.

    What, do you think it’s too late now? Why? :-)

    Link to this
  33. 33. David Marjanović 6:28 am 09/24/2013

    …Of course there’s occasionally a skeleton on display. There’s an elephant and a giraffe in the hallway of the biology building of the University of Vienna, for instance. But is there more than one place in the world that exhibits a whole comparative series of skeletons of extant non-cetaceans?

    Similarly, the skeleton of Salamandra salamandra has sort of been described and illustrated several times in the literature, as “here, have some basic idea what a salamander skeleton or even salamandrid skeleton is like”; but the rest is almost all line drawings of skulls and the occasional vertebra. I already hate line drawings because they always give a 3D impression which is always wrong; but in salamandrids, sculpture is thought to be of phylogenetic importance, and I can’t code it for most species whose skulls have been drawn! I could go on for hours.

    Link to this
  34. 34. darkgabi 7:25 am 09/24/2013

    @david:
    “But is there more than one place in the world that exhibits a whole comparative series of skeletons of extant non-cetaceans?”

    yup. the museum of la plata has a huuuuge room with several skeletons on display. it’s beautiful! i must have a picture at home, but meanwhile here you go.

    it is only mammals, and althought the picture shows many cetaceans, there are lots of primates and some other carnivores like phocids, and also some bovids if i’m not mistaken. there are a couple of nice pics on google images if you type something like “museo la plata sala osteologia de vertebrados”.

    but yeah, you and jason are unfortunately right: no one cares about skeletons. not in the literature, nor in museums. situation must be slightly better in collections than in display, but still very poor =(

    “All known axial and appendicular bones of Spinosaurus were turned to dust when Munich was bombed in WWII”

    *cries*

    Link to this
  35. 35. vdinets 8:34 am 09/24/2013

    Thanks for the enlightening answers! When I asked the question about possible similarity with bison, I was totally unaware that someone had already come up with this idea.

    Link to this
  36. 36. Cai Yuan 8:48 am 09/24/2013

    The Grant Museum (UCL’s zoology museum) is essentially one room, but it is packed with comparative skeletal material, including a rather nice herp case:
    http://blogs.ucl.ac.uk/museums/files/2011/05/Reptile-Case.jpg
    AMNH has some lovely herp skeletons representing all the major clades in their venerable Hall of Reptiles and Amphibians, although not as many as one might hope:
    http://www.pinterest.com/pin/284571270175816269/

    Link to this
  37. 37. darkgabi 9:34 am 09/24/2013

    i just remembered the zoology museum of cambridge university – which is now temporarily closed. but that’s also basically mammals, is it not?

    Link to this
  38. 38. vdinets 9:56 am 09/24/2013

    The museum at Harvard has a huge collection of reptile skeletal material. I had a chance to briefly look through it just last week: shelves upon shelves, really impressive. They also have tons of other cool stuff, like a mounted Caribbean monk seal and virtually all recently extinct North American birds.

    Link to this
  39. 39. David Marjanović 10:27 am 09/24/2013

    Yay! We just borrowed Wiedersheim (1877a, b)! And some of the plates are probably marginally useful! Too bad the papers are the two parts of what’s called “the head skeleton of the urodeles” – no postcrania, no new characters.

    Thanks for the pics, everyone!

    i just remembered the zoology museum of cambridge university – which is now temporarily closed. but that’s also basically mammals, is it not?

    No idea.

    The museum at Harvard has a huge collection of reptile skeletal material. I had a chance to briefly look through it just last week: shelves upon shelves, really impressive.

    *ancestor-worships at shrine of Mighty Romer*

    Link to this
  40. 40. ectodysplasin 10:29 am 09/24/2013

    @Dartian,

    I don’t follow your logic. If no researcher has ever made a proper gross morphological comparison between these fossil and extant ‘finbacks’, then how do you know that such comparisons wouldn’t further our understanding of these structures? Are you just making an assumption?

    Understand that I’ve spent a bit more time with the bones themselves and while the data has not appeared in the literature, that is not equivalent to “making an assumption.”

    As for furthering our understanding of anything, every study worth its salt either furthers our understanding or underscores our lack of understanding, both important for the advancement of science. However, researchers have a limited amount of time in the day, so we prioritize completing studies that are of higher interest with respect to our particular research questions.

    I just tweeted that basilisk axial skeleton, let’s see if this works now:

    https://www.dropbox.com/s/faxgygeir1vdfpj/bplum.jpg

    Link to this
  41. 41. ectodysplasin 10:37 am 09/24/2013

    Further:

    It may very well indeed be that, compared with today, the Permian parasite community was different in ecologically significant ways.

    This is almost certainly the case, but it does not follow that this is the primary determiner of which taxa do and do not develop sailbacks. Once again, dinosaurs seemed to do just fine. Modern sailbacked lizards seem to do just fine. Ctenosauriscids seemed to do just fine. Furthermore, we see no evidence that the sail in sailbacked lizards is a site of preferential feeding by ectoparasites. If anything, ectoparasites like mites and ticks tend to go for the more delicate skin folds near the hip and shoulder and around the ear.

    So it’s less that I’m casually dismissing this suggestion so much as it’s a non-starter.

    Link to this
  42. 42. Hai~Ren 10:40 am 09/24/2013

    ectodysplasin: If Dropbox fails us again, perhaps you could upload the photo to Imgur?

    Link to this
  43. 43. Dartian 12:06 pm 09/24/2013

    Understand that I’ve spent a bit more time with the bones themselves and while the data has not appeared in the literature, that is not equivalent to “making an assumption.”

    You understand that I have spent a bit more time with bones myself (admittedly mammalian rather than reptilian). In fact, very likely every bit as much as you have (and quite possibly more). So how about dropping that snotty attitude?

    we prioritize completing studies that are of higher interest with respect to our particular research questions.

    Did I say that you personally would have to be the one to do that? I was asking why seemingly no scientist during these last 200-300 years has published a decent illustration of a basilisk’s skeleton. You know, people like Cuvier, von Humboldt, Fitzinger, Daudin, Owen, Huxley, Gray, Cope, Schlosser, etc., etc.?

    we see no evidence that the sail in sailbacked lizards is a site of preferential feeding by ectoparasites

    Can you back up that claim with a reference? Or are you just making assumptions again?

    Link to this
  44. 44. ectodysplasin 5:01 pm 09/24/2013

    I am sorry you have interpreted my approach here as stemming from a “snotty attitude,” my goal has been to share the information I’m aware of on the subject, and to point out issues with a hypothesis that has circulated in the past but has little actual support. I have made no judgment about your own knowledge of mammalian anatomy and have only brought up my own personal experience with the reptilian anatomy to share observations that are not readily available in the literature. It is important to remember that many of us who are involved in active research have made plenty of scientific observations that may take years to find their way into a paper (or may never do so), but these observations can and do inform other work that we do.

    It’s also important to understand the importance of proper analogies. The bison analogy was an important one at the time because it prompted necessary critical examination of these structures, but ultimately it was a poor analogy because the mammalian pectoral girdle is pretty different from the nonmammalian pectoral girdle, and thus the sort of development of the trapezius muscles seen in Bison is simply not applicable to these other animals. So when I point out that reptilian anatomy has to be the basis for evaluating these hypotheses, this is what I’m saying.

    I was asking why seemingly no scientist during these last 200-300 years has published a decent illustration of a basilisk’s skeleton.

    Because most systematic herpetology focuses on external features, such as scalation and color patterns. More generally, the answer to this question is “because no one has been interested enough to do so.” There are many, many species out there that remain unfigured, as others have also noted in this thread.

    Can you back up that claim with a reference? Or are you just making assumptions again?

    A good start would be veterinary texts on reptile parasites. More generally, reptile ectoparasites such as ticks and mites typically feed in areas of soft skin with thin, fine scalation, such as the folds of skin around the eye and ear and the folds of skin around the throat, shoulder, and hip. I doubt you’re going to find a paper looking at the 3D distribution of ticks and mites in B. plumifrons, but I can tell you from both readings within the veterinary literature as well as personal experience that this is where those parasites hide and feed. Ectoparasites simply don’t attach to the flanks of lizards, especially exposed areas with thickened skin. They stay where they are protected and have access to thinner skin. If you want to argue that they do the exact opposite when they infest an animal like Basiliscus, that’s fine, but the burden of proof is on you.

    Link to this
  45. 45. ectodysplasin 5:07 pm 09/24/2013

    @David,

    Ganoid and cosmoid scales – don’t they erupt like teeth? But tetrapod-and-similar scales never have such a layer, they’re reduced to the bone layer like teleost scales.

    Unlike teeth, these scales do remain covered in a thin layer of skin. The scales in early tetrapods have lost the enamel component, but otherwise they are not all that different; you can see similar anatomy in, say, the dermal scutes of sturgeon (which are essentially just glorified flank scales).

    Link to this
  46. 46. Dartian 12:25 am 09/25/2013

    ectodysplasin:
    I have made no judgment about your own knowledge of mammalian anatomy and have only brought up my own personal experience with the reptilian anatomy to share observations that are not readily available in the literature.

    I’m sure that you know a lot about the anatomy of Paleozoic tetrapods, perhaps even more than anyone else here. But you have also shown a tendency to make sweeping (and less than well-supported) claims that go far beyond that subject matter. What’s worse, you’ve made those claims in a haughty manner which is almost guaranteed to rub people the wrong way. Talking down to strangers is a bad idea to begin with, but it’s a particularly foolish thing to do on this blog. Many of the regular contributors here are professional scientists themselves. They may not know a dissorophid from a trematopid, but they do know more than a little about science and how it’s done. You underestimate their insight at your own peril.

    reptilian anatomy has to be the basis for evaluating these hypotheses

    Why? Dimetrodon is much more closely related to mammals than to squamates.

    They stay where they are protected and have access to thinner skin.

    But surely an ectoparasite that attached itself near the top of Dimetrodon‘s sail wouldn’t need to worry about being exposed anymore? How would the Dimetrodon have been able to dislodge the parasite? How could it have reached up there to remove it? (Unless I’m much mistaken, I believe that this is what Heteromeles was originally getting at.)

    If you want to argue that they do the exact opposite when they infest an animal like Basiliscus, that’s fine, but the burden of proof is on you.

    …and there you go again!

    I was not arguing that! I asked you a perfectly justified question regarding this claim of yours:
    we see no evidence that the sail in sailbacked lizards is a site of preferential feeding by ectoparasites
    That statement implies that there have been comparative studies on the parasite loads of extant lizards (in the wild). Furthermore, it implies that the results of those studies show that there is no significant difference in parasite infestation rate between lizards species with ‘sails’ on their backs and those without them. Now, I ask you again: Is that statement of yours backed up by actual research, yes or no? If yes, please provide details. If no, admit that you were indeed just making an assumption in comment #41.

    Link to this
  47. 47. Heteromeles 3:26 am 09/25/2013

    Thanks Dartian. You are correct in your assumption. The more general idea was trying to get at why that particular period seemed to have more unrelated sail-backed animals than today does, and both the sails and the animals sporting were relatively larger than those today.

    I decided to invert the scenario somewhat, and start asking, instead, is there something about today that prevents sails from evolving? One possible answer was parasites, because grooming a sail that large is effectively impossible for a short-legged tetrapod. In one sense, being a bug magnet would make a sail a superb fitness display: it’s big, cumbersome, it costs you blood, and you might end up with more parasites by sporting it. Problem is, if there are too many parasites, a sail could too expensive to maintain. So far as I know (and I could be wrong), there are more clades of ectoparasites today than there were back in the early Permian, and that might mean there are more parasites now.

    Sails haven’t disappeared entirely, but they’re smaller, possibly rarer, and they don’t seem to be bug magnets, per ectodysplasin. That’s good to know, and it suggests that there are problems with assuming Permian sails worked as bug magnets. However, it does suggest that sails are not adaptive in the modern environment. Since the early Permian was a period of ice ages somewhat like today climatically, that is puzzling, and it does suggest that climate is not a good explanation for the plethora of sails that occurred then.

    Link to this
  48. 48. David Marjanović 7:41 am 09/25/2013

    https://www.dropbox.com/s/faxgygeir1vdfpj/bplum.jpg

    That works, thank you!

    Because most systematic herpetology focuses on external features, such as scalation and color patterns.

    Yeah, every scale in the face of a squamate has a name. Obviously, color patterns aren’t used for large-scale phylogeny, but even there it’s soft anatomy (tongue, peritoneum…) that is used much more than the skeleton – and within the skeleton, we have the usual craniocentrism.

    Craniocentrism is rampant elsewhere, too. It took till 1995 till there was a paper on the vertebrae of coelurosaurs!

    Unlike teeth, these scales do remain covered in a thin layer of skin.

    Oh.

    But surely an ectoparasite that attached itself near the top of Dimetrodon‘s sail wouldn’t need to worry about being exposed anymore? How would the Dimetrodon have been able to dislodge the parasite? How could it have reached up there to remove it?

    It could have rubbed against a tree or something… but a parasite on the sail of Edaphosaurus, with its crossbars, may well have been impossible to dislodge.

    Link to this
  49. 49. vdinets 11:21 am 09/25/2013

    If those sails were covered with dermal bone, how could they be good places for parasite attachment?
    The ability to dislodge parasites from all parts of the body doesn’t seem to be an important requirement in the evolution of tetrapods. Otherwise we wouldn’t see limb reduction in so many lineages.

    Link to this
  50. 50. ectodysplasin 11:33 am 09/25/2013

    @Dartian,

    Talking down to strangers is a bad idea to begin with, but it’s a particularly foolish thing to do on this blog. Many of the regular contributors here are professional scientists themselves.

    Which is why I’ve repeatedly emphasized the standards of proof associated with assessing the hypotheses being forwarded. Perhaps my tone hasn’t come across, but I’ve repeatedly said “here’s the standard of proof necessary for such a hypothesis to be taken seriously.”

    Why? Dimetrodon is much more closely related to mammals than to squamates.

    And monotremes are mammals, but their shoulder girdles work in a more reptilian manner as well. Why? Because they have a well-developed interclavicle, broad cleithra, and a well-developed intercleithrum, and the whole unit is pretty well-integrated. As a result, there is simply not the potential for elevation of the scapula by the trapezius musculature. The same applies for Dimetrodon; the shoulder musculature would not have been remotely as well-developed and complex as you see in mammals in large part because bone was serving much of the role of supporting the shoulder girdle. Assuming mammalian locomotion and functional morphology in a very non-mammal-like animal is going to lead to problems.

    That statement implies that there have been comparative studies on the parasite loads of extant lizards (in the wild).

    Nope, it really doesn’t. It implies that in animals like Basiliscus, ectoparasites congregate and feed in the same areas that they congregate and feed in any other reptile, namely vulnerable skin in areas with lots of skin folding. Which, if you ask any reptile veterinarian, is an observation that really doesn’t require further study. The reason for this is that the ectoparasites that typically feed on reptiles both need to be able to get their mouthparts through the reptile’s skin and need to be in a microhabitat that protects them from drying out.

    This isn’t me trying to get out of providing evidence of making assumptions. This is about the standard of evidence necessary to invoke selection (in this case, parasite load) as the primary cause of a distribution of characteristics. For me, that inferential process works like this:

    1. Does an observed structure actually effect function? If not, there’s no reason to invoke selection. If so, then…

    2. Is the observed structure developmentally nonindependent from another feature that is demonstrably under selection? If so, we’re probably witnessing pleiotropy in action, and selection does not need to be invoked. If not, then…

    3. Is the observed characteristic a plastic response to events that have occurred within an organism’s lifetime? If so, selection of that specific character need not be invoked, although selection on phenotypic plasticiy can be investigated. If not, then…

    4. Is the structure being shaped or limited by more general functional considerations of which that specific characteristic is only a small aspect? If so, there’s no reason to invoke character-specific selection. If not, then…

    5. Start looking at selective concerns.

    This may represent a difference in philosophy of science between us, and this may also represent a differing view of selection between us. Personally, I consider developmental and functional constraint to be much more important than selection in terms of driving morphological diversity. This is in part because my background is in comparative morphology and developmental biology, not population genetics. I also consider selection and adaptation to be a conclusion that one comes to when one has exhausted all other options, rather than an expectation of every character in every system.

    This is why I keep saying that the parasite issue is a non-starter. You’re assuming that a relatively tertiary selection concern is this overwhelming driver of a macroevolutionary pattern that is probably much more simply explained by fundamental functional constraints.
    Until you rule out those functional constraints, your tertiary concern does not need to be addressed. We know that function was important, because we see repeated adaptation of similar support structure in phylogenetically disparate animals like Dimetrodon and Bromsgroveia, for example (once again, I’ll cite work by Adam Huttenlocker) and we see significant histories of trauma in Dimetrodon sails (once again, I’ll cite work by Beth Rega). Furthermore, the fac that modern sailbacks are all concentrated within a single clade cannot be ignored. There are no sailbacked mammals. There are no sailbacked birds. There are no sailbacked frogs. There are no sailbacked anguimorphs. There are no sailbacked lacertids. There are no sailbacked geckos. There are no sailbacked skinks. Sails are exclusively constrained to iguanians, where they crop up more or less frequently, including in large species like Hydrosaurus (the largest agamid) and Basiliscus (the largest corytophanid) where you’d expect fitness costs to be more important. So there’s something special about iguanians that is permitting the repeated evolution of this structure, and something about all other tetrapods that is suppressing it. Any hypothesis you want to put forward needs to take that into account.

    Link to this
  51. 51. ectodysplasin 11:47 am 09/25/2013

    @vdinets:

    The ability to dislodge parasites from all parts of the body doesn’t seem to be an important requirement in the evolution of tetrapods.

    It isn’t, even in rare cases where ectoparasites can cause serious morbidity or mortality of their hosts. A good example here is the winter tick Dermacentor in cervids in the boreal forests of North America.

    Link to this
  52. 52. ectodysplasin 12:06 pm 09/25/2013

    @heteromeles:

    ails haven’t disappeared entirely, but they’re smaller, possibly rarer, and they don’t seem to be bug magnets, per ectodysplasin. That’s good to know, and it suggests that there are problems with assuming Permian sails worked as bug magnets. However, it does suggest that sails are not adaptive in the modern environment. Since the early Permian was a period of ice ages somewhat like today climatically, that is puzzling, and it does suggest that climate is not a good explanation for the plethora of sails that occurred then.

    It’s important to note that sails were not restricted to the Permian. They turn up in the Triassic in at least one and possibly several lineages of stem-crocodile (ctenosauriscids and Lotosaurus), and in the Cretaceous in several lineages of dinosaur (mostly within non-maniraptoran neotheropods). In the Carboniferous-Permian transition, sails were mostly constrained to early synapsids, where they seem to have turned up more or less regularly.

    The take-home message here is that phylogeny and not time interval is the important factor here. Specific groups seem to evolve sailbacks over and over again, whereas other groups simply do not, no matter how many species we look at. That implies a functional constraint rather than an environmental constraint.\

    In addition, the early Permian was not particularly cold. It was a transitional climate interval during a more general turnover from icehouse to greenhouse. Additionally, all Euramerican fossil localities in that interval were within 5-10 degrees from the paleoequator, so you’re still looking at a very warm environment.

    Link to this
  53. 53. Dartian 12:31 pm 09/25/2013

    Well, that may have been the longest non-answer to a very simple and straightforward question that I’ve ever seen.

    I don’t care about your “philosophy”. What I do care about is that if you state that something is a fact you should then also able to back that up with references if needed. That is what scientists are supposed to do. Not coughing up a load of irrelevant word salad instead.

    So once again, do you have a source for that claim about actual parasite infestation rates in lizards, or do you not? Just answer the question already.

    an observation that really doesn’t require further study

    I’m sure that it will be greatly appreciated that you generously share your valuable insight on which phenomena deserve further study and which do not.

    You’re assuming

    Do not presume to know what I am or am not assuming, you muppet.

    Link to this
  54. 54. vdinets 12:51 pm 09/25/2013

    Dartian: proving a negative is seldom possible. You say that sails are preferentially attacked by parasites; ectodysplasin says they are not. It would seem that the burden of proof is on you, wouldn’t it?
    Also, considering how much fighting have dimetrodons apparently gone through already over the sails issue, perhaps we shouldn’t engage in more conflict over it?

    Link to this
  55. 55. ekocak 1:06 pm 09/25/2013

    “…ectoparasites congregate and feed in the same areas that they congregate and feed in any other reptile, namely vulnerable skin in areas with lots of skin folding.”

    As someone who keeps various Rhacodactylus geckos, I can verify this is definitely the case– and in Correlophus ciliatus there actually seem to be little pocket structures behind the hind limbs, maybe to attract mites away from other areas?

    Link to this
  56. 56. David Marjanović 1:43 pm 09/25/2013

    and a well-developed intercleithrum

    There is no such thing. I think you wanted to mention the interclavicle again.

    What the interclavicle, a median bone, does is prevent movement of the left and right shoulder girdles relative to each other. That’s a very important difference to therians.

    So there’s something special about iguanians that is permitting the repeated evolution of this structure, and something about all other tetrapods that is suppressing it. Any hypothesis you want to put forward needs to take that into account.

    Agreed.

    and in the Cretaceous in several lineages of dinosaur (mostly within non-maniraptoran neotheropods)

    There’s Ouranosaurus, a close relative of the hadrosaurs; there’s Spinosaurus; and there’s Concavenator the allosauroid. As mentioned above, there’s also Amargasaurus with its double sail (bifid neural spines) on the neck – if that was a sail and not simply two rows of spikes. All of them are remarkably close together in age, and even in geographic latitude.

    Link to this
  57. 57. vdinets 2:50 pm 09/25/2013

    Perhaps it’s worth mentioning that sails are very common among bony fishes. In some cases they are for making sharp turns (i. e. in sailfish), but in countless others they seem to be mostly for display (i. e. in mudskippers).

    Link to this
  58. 58. ectodysplasin 3:10 pm 09/25/2013

    @david;

    There is no such thing. I think you wanted to mention the interclavicle again.

    Yeah. Pre-coffee typo.

    What the interclavicle, a median bone, does is prevent movement of the left and right shoulder girdles relative to each other. That’s a very important difference to therians.

    Bingo.

    Link to this
  59. 59. ectodysplasin 4:07 pm 09/25/2013

    @Dartian,

    1. Nowhere in this discussion have I insulted you. I’m not certain where your ire is coming from. I’d rather not turn Darren’s blog into a flamefest, so this will be my final response to you.

    2. Philosophy matters because it depends on how we assign burden of proof. As others and myself have pointed out, when we do science, our initial assumption (in a statistical/philosophical sense, rather than a casual use sense) is that parameters we are investigating are meaningless, and we continue to treat those parameters as meaningless until convincing evidence can be presented otherwise. To approach problems in the opposite way is extremely problematic, because it’s very easy to wave away evidence against a proposition as not being the “right” experiment or the “right” kind of data. If, for example, I did a study looking at mite feeding locations on captive Basiliscus and indeed reported that mites feed preferentially in sheltered areas such as underarms, vent, ear, etc., then you’d claim that this wasn’t a study of wild populations. If I went and studied this in introduced populations in Florida, you’d demand to see similar data on Basiliscus throughout Central America. If I did that, you’d ask why I hadn’t pursued a similar study on Hydrosaurus. If I surveyed every species of modern sailbacked lizard, you’d then point out that fossil sailbacks aren’t lizards, and declare the evidence I’d presented moot anyways. This is the form of argumentation that is used by creationists and by fringe theorists such as Aquatic Ape proponents and the BANDits, and is essentially junk science because the goalposts can continuously shift to invalidate the applicability of any and all data presented. So, assuming we’re both scientists (which I’ve been assuming from the get-go) the burden of proof is on the person who proposes that a parameter is meaningful, and in this case that person is you. If you want a resource on reptile medicine in general, the Mader text is a good reference and should answer your questions.

    3. I’ve made no comment on which phenomena deserve study, only which hypotheses deserve serious consideration by scientists. There is a basic threshold of suggestive data that must be met before a hypothesis is really worth considering, and upon reaching that threshold, a hypothesis must continue to produce substantial supporting data if it is to continue to receive careful consideration. If you or someone else would like to spend your time researching hypotheses that do not meet that basic threshold, that is absolutely your prerogative, but there’s no reason to think that this is a critical research topic.

    Link to this
  60. 60. ectodysplasin 4:25 pm 09/25/2013

    @vdinets

    Interestingly, the spines of large Dimetrodon species have the same I-beam construction that you see in the fin spines of sailfishes. This is not the case in many other sailbacked animals.

    Link to this
  61. 61. vdinets 9:35 pm 09/25/2013

    ectodysplasin: in sailfishes this might be for resisting lateral forces during sharp turns. Makes you guess whether dimetrodon sails were actually used for sailing, for example, for crossing rivers during migration.

    Link to this
  62. 62. Dartian 12:43 am 09/26/2013

    You say that sails are preferentially attacked by parasites

    FFS, I never said what! I was questioning the claim that the presence of sails makes no difference one way or the other. Which I why I asked for actual data (as opposed to unsubstantiated personal opinions) on this subject.

    Next time, before you try to put words in somebody’s mouth, try reading comprehension first.

    ectodysplasin:
    Nowhere in this discussion have I insulted you. I’m not certain where your ire is coming from.

    I’ve already explained it to you. If you’re acting towards others in a condescending manner (I’m going to give you the benefit of the doubt and assume that you’re not doing it intentionally), don’t be surprised if they get annoyed at you. I had a simple and reasonable request to you: to provide a reference to a claim that you made. You refused to do so. If you had just answered something along the lines of “Sorry, but I don’t have a reference” and admitted that you therefore don’t really know the answer to the question whether sailbacked lizards suffer differently from ectoparasite infestations than non-sailbacked lizards, then I would have been satisfied and would have left it at that.

    Instead, you started digressing about things I didn’t ask about and things that even in general are hardly relevant here. In the context of this blog discussion, I do not care about your scientific philosophy, or, strictly speaking, even about your scientific credentials. I’m far more interested in the actual facts that you have to present than in your personal opinions (because that’s what they are unless and until you back them up with data).

    This is the form of argumentation that is used by creationists and by fringe theorists such as Aquatic Ape proponents and the BANDits, and is essentially junk science because the goalposts can continuously shift to invalidate the applicability of any and all data presented.

    Yeah… whatever.

    Link to this
  63. 63. Dartian 12:46 am 09/26/2013

    I never said what!

    …ell, perhaps I should try preview next time; I meant of course ‘that‘.

    Link to this
  64. 64. vdinets 11:31 am 09/26/2013

    Dartian: it doesn’t matter if you said it or implied it. In any case, you asked your opponent to provide evidence that sails are not preferentially attacked by parasites. And that would mean proving a negative.

    Anyway, I have two questions for everybody here (since it looks like pretty much everybody here knows paleontology much better than I do). (1) Has everybody ever plotted all those sailback taxa on the geological timeline? That would probably be helpful, since we would immediately see if there really were “abnormal” peaks of occurrence. (2) What, if anything, do we know about average and maximum wind speeds from Permian to Cretaceous? Those sailback taxa seem to have very strong legs; I wonder if wind speeds in open landscapes could be so high that the only way to walk into the wind was to use a sail the same way sailboats use it.

    Link to this
  65. 65. vdinets 11:32 am 09/26/2013

    i mean, anybody, not everybody, sorry

    Link to this
  66. 66. SRPlant 11:53 am 09/26/2013

    vdinets
    “I wonder if wind speeds in open landscapes could be so high that the only way to walk into the wind was to use a sail the same way sailboats use it.”
    Yes, this is where Tackydromus came into its own.

    Link to this
  67. 67. vdinets 12:15 pm 09/26/2013

    SRPlant: but stronger winds would explain all the mysteries of that time at once. Why were so many dinosaurs so large? How did adzarchids take off? Why were so many dinosaurs bipedal? (Bipedality is a big advantage in windy conditions because you can lean into the wind; rheas in Patagonia can happily run around even when it’s so windy that guanacos try not to stand up). What were those small wings in pre-avian dinosaurs for? (For pressing down rather than lifting up, like car spoilers). Why did so many animals in Mongolia apparently perish in sandstorms? And so on, ad infinitum :-)

    Link to this
  68. 68. ectodysplasin 7:12 pm 09/26/2013

    @vdinets:

    ectodysplasin: in sailfishes this might be for resisting lateral forces during sharp turns. Makes you guess whether dimetrodon sails were actually used for sailing, for example, for crossing rivers during migration.

    According to Adam Huttenlocker (who, as I said above, is the world expert on synapsid sails), it’s probably more about resisting lateral bending during trunk flexion. I think this is discussed in his paper on Dimetrodon sail histology, but I’m not 100% certain. I know this is discussed in his MSc thesis.

    Link to this
  69. 69. CS Shelton 2:05 am 09/27/2013

    Total layperson opinion: I took the thermoregulation idea without question until just now reading all this stuff. Now I lean toward display.

    One thing we know that was different between then and now, which I think accounts for the lack of similar structures in large tetrapods, is the body shapes of large animals. Tall straight limbs have become the norm, and already were before the non-avian dinos went extinct. Sprawling, short limbs dominated back then. Posture, body shape, and eye position would make a big difference in the type of display structures you see, wouldn’t they?

    Link to this
  70. 70. Dartian 3:08 am 09/27/2013

    it doesn’t matter if you said it or implied it

    *sigh*

    No, Vlad, I didn’t imply it either.

    Seriously, what is it that’s so difficult to understand here? I am not making any claims here one way or the other – I have all the time been asking for actual, published, peer-reviewed, real-world data. If such data exist, then somebody just present them already. If such data do not exist, then that means that we (and I mean ‘we’ as in ‘anyone even remotely connected with the scientific community’) simply do not know the answer to this particular question. And therefore, anyone saying otherwise is relying on anecdotal information at best and on pure pulled-out-of-the-arse guesswork at worst.

    In case my point is still getting missed, let me try an analogy.

    There are lots of phenomena in nature that we know to exist or to take place, but which we actually do not know so much about when it comes to the interesting nitty-gritty details. For example, anyone who’s ever watched nature documentaries on TV knows that crocodiles prey on migrating, river-crossing wildebeest in East Africa. This has been filmed so many times that we surely know all there is to know about this, right?

    Wrong; we know surprisingly little about it. I recently tried (for non-Tet Zoo-related reasons) to find technical literature on crocodile predation on migrating wildebeest, but I was unable to find anything useful. Literally, not a single paper devoted to this subject! Perhaps I just didn’t search thoroughly enough and I may therefore have missed some crucial publications, but I’m now fairly sure that there isn’t exactly an over-abundance of published research on this subject. In other words, we currently do not know if (just to take an example) crocodile predation on wildebeest is selective or random; we would need to statistically analyse lots of observational data before we could answer that question.

    Now, Vlad, imagine if instead of parasite infestation in sailbacked vs. non-sailbacked lizards we would be discussing crocodile predation patterns on wildebeest. Let’s hypothetically assume that Tet Zoo commenter A says that “Maybe crocodiles preferentially prey on female wildebeest?” (‘Maybe’ is the operative word there; A is not claiming that crocodiles do prey preferentially on females.) To that, Tet Zoo commenter B replies that “No they don’t.” Then the exchange would go something like this:
    A: “How do you know that? Do you have a source?”
    B: “The whole question is a non-starter and besides, it’s not even interesting.”
    A: “Do you have a reference or not?”
    B: “The burden of proof is on you.”
    A: “The hell it is! I am just asking for data.”
    B: Starts talking about his personal research philosophy and other irrelevancies.
    The discussion: Goes nowhere.

    You see the problem? That has been ectodysplasin’s schtick in this thread. He first made a specific claim and then would not (or, as I strongly suspect, could not) back it up. The right thing for him to do in this situation would have been to just ‘fess up to that and admit that his opinion was just that, an opinion. (Personally, I would respect that kind of honesty.) That’s what I took him to task for – nothing more, nothing less. Вы теперь понимаете?

    Link to this
  71. 71. SRPlant 4:14 am 09/27/2013

    @ vdinets
    I didn’t mean to mislead you with my attempt at a pun. I should learn to TACK smileys to the end of my facetious comments!

    Link to this
  72. 72. vdinets 8:45 am 09/27/2013

    Dartian: it’s not an exact analogy. One could easily imagine why crocodiles would preferentially feed on females (calves slowing them down, slightly smaller size, softer meat, etc.), and this idea can be easily tested. However, there’s no obvious reason for parasites to prefer sails, and no easy way to test this for extinct taxa. Talking about such a far-fetched hypothesis only makes sense if there is some supporting evidence to begin with. That’s what ectodysplasin probably meant when he said it was a non-starter.

    As for the lack of scientific papers on wildebeest predation, it’s sad, but if someone wrote such a paper, it would probably be rejected outright by most, if not all, journals for being too descriptive and trivial, unless the authors found some unexpected pattern, in which case they would be advised to resubmit their paper to a more crocodile-oriented journal, of which none exist… but I’m digressing, sorry.

    Link to this
  73. 73. vdinets 9:49 am 09/27/2013

    SRPlant: I got the joke, but pretended not to, so I could further push my BAEMHAT (Brilliant All-Explaining Mesozoic Hyper-Aeolian Theory).

    Link to this
  74. 74. Heteromeles 11:32 am 09/27/2013

    Since I started this, I should point out that I’m not wedded to it. As Dartian noted and vdinets and ectodysplasin seem to radically avoid, you can’t explain everything from bones alone, because animals don’t live in a vacuum.

    I’m an ecologist by training, so when I see something that doesn’t make sense (like large sails on some early Permian animals, or long necks on Plesiosaurs and other fossils), I tend to start asking what we’re missing that does help these features make sense in their native context.

    For the sails, I’m happy with the idea that they’re a display, although I don’t know whether it’s true. Do we even know what a female Dimetrodon looks like, for example? Outside the Permian, most of the sail fossils seem to be fragmentary at best, and I’m not clear that we have any population data to confirm a sex hypothesis. Still, sails are big, clumsy structures, and that kind of thing often goes with sex selection. The key question to me, though, is why that type of display is used, and not something else, and why a couple of unrelated animals pop up sails at about the same time. Is this happenstance, or was there something different about the environment then that favored sails as a display? In this case, I pitched the idea of it being about parasites, simply because insects have evolved too, and it’s worth remembering that in understanding the environment in which extinct animals lived.

    And don’t even get me started on Plesiosaur necks.

    Link to this
  75. 75. naishd 12:09 pm 09/27/2013

    Dimetrodon sails (and other sails) and sexual selection… I’ll have to devote a whole article to this issue at some stage. Remember that sexual dimorphism is not a requirement for sexual selection to be inferred as the main driving force behind a structure’s evolution (though it’s nice to see, since it often supports or confirms the presence of sexual selection). Positive allometry is also a good guide (usually: though not universally; loads of exceptions), and it seems that the dorsal sails of these animals grew under profound positive allometry, their allometric slopes well exceeding what you’d predict for a thermoregulatory function.

    I don’t wanna come across as a hater, but (to repeat myself, again..) I think palaeontologists have all too frequently only interpreted weird structures in ‘functional’ or ‘mechanical’ terms, forgetting or ignoring the fact that sexual display probably outweighs everything else when it comes to the evolution of ridiculous structures.

    Darren

    Link to this
  76. 76. ectodysplasin 12:38 pm 09/27/2013

    @vdinets:

    Dartian: it’s not an exact analogy. One could easily imagine why crocodiles would preferentially feed on females (calves slowing them down, slightly smaller size, softer meat, etc.), and this idea can be easily tested. However, there’s no obvious reason for parasites to prefer sails, and no easy way to test this for extinct taxa. Talking about such a far-fetched hypothesis only makes sense if there is some supporting evidence to begin with. That’s what ectodysplasin probably meant when he said it was a non-starter.

    Yes, but I’ll reject the idea that the wildebeest scenario is any more scientific or any more deserving of attention.

    I could sit around with a beer and come up with any number of hypotheses backed by credible scenarios and then come up with experiments to test this, but this isn’t really how science works. The first step in science is to simply look at things. A lot. Dartian is dismissing this step as representing anecdotal data, but the truth of the matter is, this is fundamental to the scientific process, and, contrary to popular belief, anecdotal data are still data. Observation should be the basis of hypothesis, not the other way around. Controlled studies then become our means of systematizing subsequent observations according to industry standard, but that does not invalidate anecdote.

    With the wildebeest example, unless you look and say “huh, there sure are a lot of female wildebeest out there getting munched by crocodiles” then there’s no reason to start proposing reasons [i]why[/i] this might be the case. In fact, I can argue the opposite: crocodiles may not be able to discern the difference between females and males, crocodiles may prioritize closeness to the water and first jumpers over characteristics of the wildebeest, calves may not be an issue at all given that mothers are unable to watch out for their calves during crowded river crossings, and so on. An absence of data supports the null hypothesis (that is, that all novel positive propositions are false) and you’d need to present some manner of anecdotal data before you could justify a need for more systematic study.

    More generally, this should underscore something rather important about the body of knowledge within science outside of the published literature. There’s a lot that we know that isn’t published in the literature, either because we haven’t gotten around to it yet, or because we don’t have a format to publish short-form field or lab observations anymore in quite the same manner as they did 100 years ago, or because independently such observations are not particularly interesting. This does not only for academics, but also for medical and veterinary professionals, engineers, and so on and so forth. Case studies are published by doctors and vets only when they are exceptional, and the thousands to tens of thousands of non-exceptional cases are never remarked upon. Scientists often do not report or even record field observations that are unrelated to their research question. And so on. But we do take this knowledge into account in future decision-making and interpretation of results.

    So when I’m reporting anecdotal observation that is backed by veterinary publications, that is in fact data, and the fact that multiple people in these comments have also confirmed that these are observations that they too have made underscores that as well. At that point, you can more or less dismiss a positive statement such as the one Dartian has decided deserves greater attention unless the defenders can present solid evidence contradicting that.

    Even more generally, though, we can dismiss the need for an ecological hypothesis such as this. Remember, the core question is “why are there no modern sailbacks at the same rate as we saw them in the Permian?” and honestly we haven’t even confirmed that this is the case, let alone that this isn’t a function of intrinsic constraints, either developmental or functional. Ecological hypotheses may be satisfying to the imagination, but you have to knock out a whole lot of other hypotheses first before they become viable, and that hasn’t happened here.

    Finally, and most importantly, there’s no actual trend to explain. There are sailbacks throughout the Mesozoic, and there are sailbacks today. Sailback evolution seems to be pretty closely tied to specific taxonomic radiations (non-therapsid synapsids, non-crocodyliform crurotarsans, non-maniraptor neotheropods, iguanians) and sailbacks seem to disappear when these larger encompassing taxa disappear from the record. Sailbacks don’t suddenly appear in captorhinids or “parareptiles” when pelycosaur-grade synapsids disappear from the record. They don’t suddenly appear in tyrannosaurids when spinosaurids and allosauroids go extinct. The patterns of sailback diversity are pretty closely tied with patterns of extinction and/or diversification of groups that include sailbacked members, and does not appear to be a driver of those diversity changes. That alone should give us pause when discussing ecological explanations for this anatomy.

    Link to this
  77. 77. ectodysplasin 1:19 pm 09/27/2013

    @darren,

    I don’t wanna come across as a hater, but (to repeat myself, again..) I think palaeontologists have all too frequently only interpreted weird structures in ‘functional’ or ‘mechanical’ terms, forgetting or ignoring the fact that sexual display probably outweighs everything else when it comes to the evolution of ridiculous structures.

    Well, in terms of mechanical and functional terms, this is what is easiest to study in the fossil record, and to directly test. It’s important to note that the “how” (functional, mechanical, and developmental constraints) vs the “why” (sexual selection, ecology, etc) are different sorts of questions with different standards of evidence. HOWEVER, if the question is “why not” then you need to apply the standards of evidence that apply to functional, mechanical, and developmental constraints first before looking at ecology, because it is possible that the structure simply isn’t an option in the first place.

    Link to this
  78. 78. ectodysplasin 2:24 pm 09/27/2013

    @Heteromeles:

    As Dartian noted and vdinets and ectodysplasin seem to radically avoid, you can’t explain everything from bones alone, because animals don’t live in a vacuum.

    I’m an ecologist by training, so when I see something that doesn’t make sense (like large sails on some early Permian animals, or long necks on Plesiosaurs and other fossils), I tend to start asking what we’re missing that does help these features make sense in their native context.

    Right, but I think you’re missing a few things by not giving the bones sufficient attention.

    1. When organisms develop a structure (such as a sail) it’s because they can. There may be selection on that structure (e.g. sexual selection) but ultimately, the fact that they do develop that structure is testament to the fact that they can develop that structure.

    2. When organisms do not develop a structure, it could be for one of two reasons. One reason is that they simply don’t. This is the historical contingency associated with mutation, drift, and evolution in general. The other reason is that it is possible that they simply can’t. There are a number of reasons that they might not be able to develop that structure. Maybe it is impossible developmentally. Maybe the structure impairs functional performance (locomotion, feeding, etc) at such a significant level that it is essentially fatal to anything that evolves that feature. Or maybe that feature is ecologically detrimental even though it has no effect on performance. When addressing these questions, the first step is to determine whether the structure is even developmentally possible, and if so, to determine whether it is functionally possible. Until you establish that functional and developmental constraints are trivial, there’s no reason to invoke ecology.

    You are mistaken if you think I believe ecology doesn’t matter. I do think ecology matters quite a bit. I do not, however, agree that ecology can ever claim primacy in questions of “how” and “why not,” because you cannot have ecology-scale selection on an organism that cannot exist in the first place for other reasons. Natural selection at the ecological scale can only act on organisms that exist.

    This is why I have repeatedly emphasized that while multiple originations of sailbacks can be seen in disparate groups, there are, more importantly, multiple origins of sailbacks in closely-related groups. This pattern is classic for constraint-relaxation.

    There are other obvious semi-functional constraints as well that have not been addressed and ought to be. Physical displays are meaningless for animals without good vision, especially for nocturnal animals. Big sails preclude burrowing. Big sails can be heavy. Big sails can easily experience severe trauma and can take a while to heal, and do not heal well. All of these are likely much more important than where on the body mites can attach (and, to repeat, they attach in skin folds).

    Link to this
  79. 79. David Marjanović 2:40 pm 09/27/2013

    Makes you guess whether dimetrodon sails were actually used for sailing, for example, for crossing rivers during migration.

    This has, in fact, been proposed – but looking at the proportions of Dimetrodon, I can’t imagine it walked around that much.

    Link to this
  80. 80. morphospaceman 3:41 pm 09/27/2013

    “Evolution is the control of development by ecology” – Van Valen

    I personally like emphasizing thermoregulatory functions of traits like dorsal sails or plates, long necks etc. especially in groups that are experimenting with new thermoregulatory strategies or as old-school poikilothermic ectotherms have a lot to gain from any extra warmth. Temperature is afterall arguably the most important environmental factor affecting development.
    The evolution of body size has been shown to be surprisingly fast and it’s easier to accommodate to the changed thermal balance by changing morphology rather than physiology.
    Sexual selection is often means to an end (at least in retrospective), enabling stronger selection for the trait. Runaway processes may then leave the animal with a silly oversize(but highly sexy) dorsal sail.

    This story aside, I would in this case promote an antipredatory function. Many aquatic animals exhibit often plastic antipredatory morphologies of dorsoventrally deep bodies, rendering them harder to swallow whole and seem bigger.

    Link to this
  81. 81. Heteromeles 4:26 pm 09/27/2013

    @79 David: I know you’re not seriously proposing this, but do you see any aerodynamic properties in a Dimetrodon sail? At first glance, it’s going to get pushed over on its side, rather than forward or backward. One might better argue that Triceratops shields were designed for use as air brakes to stop it faster at a gallop, just in my humble opinion. You know, tilt the head up, come to a screeching halt, pay no attention to how much force torques that atlas vertebra when the Triceratops lowers its nose.

    For all I know, the female Dimetrodon had a sail to provide shade for her offspring in an otherwise inhospitable savanna or desert. Possibly she even covered them with her sail at night for protection…

    Link to this
  82. 82. ectodysplasin 5:14 pm 09/27/2013

    @morphospaceman:

    I personally like emphasizing thermoregulatory functions of traits like dorsal sails or plates, long necks etc. especially in groups that are experimenting with new thermoregulatory strategies or as old-school poikilothermic ectotherms have a lot to gain from any extra warmth. Temperature is afterall arguably the most important environmental factor affecting development.

    Worth noting that all modern sailbacks are tropical, not temperate. Sails probably radiate heat much more than they accumulate it. Note that large fleshy structures like rabbit ears and elephant ears are used more to dump extra heat into the environment rather than to stay warm.

    Link to this
  83. 83. CS Shelton 3:52 am 09/28/2013

    Hetero@81, and since we don’t see sexual dimorphism there, if the shade idea was true, it would imply shared egg care. All sorts of fun possibilities.

    Link to this
  84. 84. David Marjanović 11:23 am 09/28/2013

    This story aside, I would in this case promote an antipredatory function. Many aquatic animals exhibit often plastic antipredatory morphologies of dorsoventrally deep bodies, rendering them harder to swallow whole and seem bigger.

    Then why are we seeing them on 3-m-long top predators like Dimetrodon, 4-m-long herbivores like Edaphosaurus, and some-hallucinogenic-number-like-17-m-long top predators like Spinosaurus? Or what do you mean by “this case”?

    do you see any aerodynamic properties in a Dimetrodon sail?

    Well, no. :-| It’s a silly idea, and I wanted to provide a counterargument other than “it’s silly”.

    pay no attention to how much force torques that atlas vertebra when the Triceratops lowers its nose

    Ceratopsids do have that famous syncervical… :-)

    Link to this
  85. 85. morphospaceman 6:46 pm 09/28/2013

    Yes, I was mostly thinking of heat dumping in hot environment and trend of increasing body size. That upper lethal limit creeps on you easily.

    Anyway, since the drawing stuff was just collecting dust, I decided to draw one too. I know it’s silly, since Plathyhystrix was a top tier in the food chain, but here is the unpalatable aposematic Platyhystrix.

    [IMG]https://dl.dropboxusercontent.com/u/43077893/platyhystrix3.jpg[/IMG]

    Neural spine elongation is sure a weird thing. Could it be once in a lifetime sexual ornament if the species happened to be semelparous for some reason. Like the hump of the pacific salmons.

    Link to this
  86. 86. Yodelling Cyclist 7:16 pm 09/28/2013

    I am the paleaontologically naive one around here, but I don’t see quite why a grand unified theory of sails is required. Why is it implausible that Dimetrodon may have deployed sails for thermoregulation while Spinosaurs were conducting sexual display (or indeed some other exotic set of scenarios)?

    Meanwhile, may one ask how aquatic Dimetrodon was? My understanding was that the various Dimetrodon sp. we’re splashing around in an Everglades-esque environment, were they swimming around at all?

    Link to this
  87. 87. Yodelling Cyclist 7:18 pm 09/28/2013

    *were

    Link to this
  88. 88. ectodysplasin 10:47 pm 09/28/2013

    @morphospaceman:

    since Plathyhystrix was a top tier in the food chain

    This is probably not a safe assumption. Platyhystrix was relatively small compared to animals like Dimetrodon, Sphenacodon, Varanops, Ophiacodon, Watongia, etc.

    Link to this
  89. 89. ectodysplasin 10:56 pm 09/28/2013

    @yodelling cyclist

    I am the paleaontologically naive one around here, but I don’t see quite why a grand unified theory of sails is required. Why is it implausible that Dimetrodon may have deployed sails for thermoregulation while Spinosaurs were conducting sexual display (or indeed some other exotic set of scenarios)?

    Meanwhile, may one ask how aquatic Dimetrodon was? My understanding was that the various Dimetrodon sp. we’re splashing around in an Everglades-esque environment, were they swimming around at all?

    There’s not really a need for a grand unifying theory of sails, but there are likely common functional constraints that sails experience and that sails impose on an organism, and those have some importance in answering the question of why sails are not more common.

    As for the habitat of Dimetrodon, it was probably not very aquatic at all. It is not found in localities with extensive permanent standing water (e.g. the coal measures in the Pittsburgh region). It is found primarily in dryland habitats, such as the North American southwest and the Bromacker Quarry in Germany, as well as in seasonally arid paleoenvironments in Texas and Oklahoma. In other words, not a swimmer.

    Link to this
  90. 90. Yodelling Cyclist 7:03 am 09/30/2013

    Thanks.

    Link to this
  91. 91. ohnosir 7:00 pm 11/5/2013

    Old discussion, but somebody might benefit from this…
    In regards to David’s question, #32:
    Do you know a museum that puts skeletons of extant non-cetaceans on display, other than the comparative-anatomy part of the MNHN in Paris which of course exhibits mostly large mammals?

    Why YES, there is such a place out there. It just so happens to be in Oklahoma, so nobody knows about it. The Oklahoma Museum of Osteology has skeletons of a huge variety of skeletons of extant mammals, birds, reptiles and amphibians, and they are continuously adding more specimens. The museum is also the home of Skulls Unlimited, but they just recently became their own independent organization.

    If anybody ever has reason to visit Oklahoma (there are very few), this museum is well worth the visit.

    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 Holiday Sale

Give a Gift &
Get a Gift - Free!

Give a 1 year subscription as low as $14.99

Subscribe Now! >

X

Email this Article

X