Correcting mistakes on Brachydectes

Perhaps one of the most difficult skulls
in all of the Tetrapoda is Brachydectes newberryi ((Wellstead 1991; Latest Carboniferous, Fig. 1). Many bones are in their standard positions. However, the bones posterior to the orbit have moved around, fused or become lost. That’s where the trouble begins.

Figure 1. Brachydectes newberryi has some difficult to identify bones just aft of the orbit due to fusion and reduction. Brachydectes (Laysorophus) elongatus (Fig. 2) provides Rosetta Stone clues as to what is happening in this clade.

Figure 1. Brachydectes newberryi has some difficult to identify bones just aft of the orbit due to fusion and reduction. Brachydectes (Laysorophus tricarinatus) elongatus (Fig. 2) provides Rosetta Stone clues as to what is happening in this clade. Note the tabulars may be more of a square shape, as Pardo and Anderson drew, but did not identify as such. 

Finding data for
Brachydectes elongatus (formerly Lysorophus tricarinatus; Cope 1877, Carroll and Gaskill  1978, Wellstead 1991; Permian, 250 mya; AMNH 6172 ) provides many needed clues as to the identity of the mystery bones.  The data comes from Carroll and Gaskill 1978 and Wellstead 1991. Earlier hypotheses included errors that I want to correct now. Based on phylogenetic bracketing these taxa nest with the caecilians Eocaecilia and Dermophis all derived from elongate microsaurs close to Archerontiscus, Oestocephalus, Adelogyrinus, Adelospondylus and Microbrachis in the large reptile tree (LRT). Unfotunatey, the latter taxa do not reduce the cheek and temple elements. So they were of little help.

Figure 2. Brachydectes elongatus (Lysorophus tricarinatus) from Carroll and Gaskill 1978 and Wellstead 1991 with colors and new bone identities added.

Figure 2. Brachydectes elongatus (Lysorophus tricarinatus) from Carroll and Gaskill 1978 and Wellstead 1991 with colors and new bone identities added.

As you can see
in figure 2, most of the skull roofing bones and anterior skull bones of Brachydectes elongatus are in their standard spots and are therefore uncontroversial. So let’s nail down the rest of the bones with a parsimony check.

Figure 3. Brachydectes species compared to scale and not to scale. Size alone might warrant generic distinction.

Figure 3. Brachydectes species compared to scale and not to scale. Size alone might warrant generic distinction.

  1. No sister taxa have a large supraoccipital that contacts the parietals and extends over the skull roof. Here that light tan median bone is identified as a set of fused post parietals, as in sister taxa. A more typical supraoccipital may be peeking out as a sliver over the foramen magnum (spinal nerve opening, beneath the fused postparietals.
  2. No sister taxa separate the postparietals, so those in light red are identified here as tabulars, bones which typically form the posterior rim of sister taxa skulls and often provide corners to the skull.
  3. Typcially anterior to, but this time lateral to the new tabulars are the bright green supratemporals. As in sister taxa they maintain contact with the postorbitals (yellow/amber) and parietals (lavender/light purple). They form skull corners in B. elongulatus and rise above the plane of the cranium in B. newberryi – but still act as skull corners.
  4. The jugal is completely absent (unless a sliver of it is fused to the yellow-green quadratojugal lateral to the quadrate, The maxilla posterior to the eyeball is also absent.
  5. The postfrontal is fused to the parietal, with a slender strip maintaining contact with the postfrontal.
  6. The postorbital is in its standard position at the posterior orbit. Here it is roofed over by the supratemporal, as in Microbrachis.
  7. The squamosal is the tricky bone. It appears as a separate bright magenta element in B. elongulatus, but must be absent or fused to the postorbital in B. newberryi because it is otherwise not visible. I agree with previous workers on the identity of the squamosal in B. elongatus.

Bones may fuse, drift and change shape, but their connections to other bones often remain to help identify them using phylogenetic bracketing. Of course that requires a valid phylogenetic framework, one that minimizes taxon exclusion problems. The tabulars do not trade places with the postparietals in this hypothesis. The tabulars maintain their original places, lateral to the fused postparietals, bones which fuse by convergence in other taxa. Perhaps the concept of an autapomorphic oversized supraoccipittal was the source of earlier errors.

It’s interesting
that the opisthotics are posteriorly covered by the exoccipitals. That usually does not happen in most tetrapods, but is further emphasized in the caecilians, Eocaecilia and Dermophis. In competing candidate taxa Rhynchonkos, Batropetes and Microrator, a different pattern is present with the postparietals descending to cover large portions of the occiput and the tabulars are fused or absent.

Wellstead (1991) and perhaps others
made Brachydectes elongatus and Brachydectes newberryi congeneric, but I see enough differences here to warrant separate genera.

Pardo and Anderson 2016 reported, 
“Contra the proposals of some workers, we find no evidence of expected lissamphibian synapomorphies in the skull morphology in Brachydectes newberryi, and instead recognize a number of derived amniote characteristics within the braincase and suspensorium.

Our study reveals similarities between the braincase of Brachydectes and brachystelechid recumbirostrans, corroborating prior work suggesting a close relationship between these taxa.”

Pardo and Anderson freehand
a Brachydectes newberryi skull reconstruction to supplement their CT scans, but do not label the bones in the drawing. Present are paired bones posterior to the parietals and a single median bone posterior to those. Based on their text, the bones posterior to the parietals are identified as post parietals, “as in the majority of early tetrapods.’ Unfortunately, sister taxa among the microsaurs do not have a large supraoccipital. So this bone has to be reconsidered as a post parietal, which all related taxa have arching over the foramen magnum. Pardo and Anderson do not mention supratemporals, but all sister taxa in the LRT have them.

Recumbirostra
according to Wikipedia, are lepospondyl amphibians that include a large number of microsaurs. Of course, those are not derived amniotes. The LRT nests Brachydectes within the Microsauria (which is not a paraphyletic group here). The phylogenetic topology of Recumbirostrans recovered by Glienke (2012) do not create the same topology in the LRT, perhaps due to taxon exclusion. Glienke recovers Eocaecilia close to Rhynchonkos (in the absence of Adelospondyli). In both studies Microbrachis is basal.

The process of discovery
is often the process of correcting errors. And, as you can see, I’m glad to do so when errors are detected, whether out there or in here. Apologies for earlier errors. We’re all learning and helping each other to learn here.

 

References
Carroll RL and Gaskill P 1978. The order Microsauria. American Philosophical Society Memoires 126: 211 pp.
Cope ED 1877. Description of extinct Vertebrata from the Permian and Triassic formations of the United States. Proc. Am. Philos. Soc. 17: 182-193.
Pardo JD and Anderson JS 2016. Cranial Morphology of the Carboniferous-Permian Tetrapod Brachydectes newberryi (Lepospondyli, Lysorophia): New Data from μCT. PLoS ONE 11(8): e0161823. doi:10.1371/journal.pone.0161823
Wellstead C F 1991. Taxonomic revision of the Lysorophia, Permo-Carboniferous lepospondyl amphibians. Bulletin of the American Museum of Natural History 209: 1–90.

wiki/Lysorophus
wiki/Brachydectes

Advertisements

2 thoughts on “Correcting mistakes on Brachydectes

  1. How about you correct mistakes in your own posts after reading the comments to them? I honestly don’t understand why you keep publishing more posts, as if that were more urgent.

    Figure 2. Brachydectes elongatus (Lysorophus tricarinatus) from Carroll and Gaskill 1978 and Wellstead 1991 with colors and new bone identities added.

    Why bother? All those drawings are wrong (Pardo & Anderson 2016).

    BTW, Lysorophus tricarinatus is a nomen dubium, based on a fragment of a vertebra which seems to belong to a lysorophian but cannot be diagnosed any further (Wellstead 1991). There’s no point in using that name anymore.

    As you can see
    in figure 2, most of the skull roofing bones and anterior skull bones of Brachydectes elongatus are in their standard spots and are therefore uncontroversial.

    I repeat: all those drawings are wrong (Pardo & Anderson 2016). For example, the posttemporal foramen identified by Carroll & Gaskill (1978) and Wellstead (1991) is simply not there (Pardo & Anderson 2016).

    No sister taxa have a large supraoccipital that contacts the parietals and extends over the skull roof. Here that light tan median bone is identified as a set of fused post parietals, as in sister taxa.

    That idea has occurred to me, too. Alas, it’s wrong nonetheless. First, there never was evidence for fusion; second, the suproccipital is often quite long and extends far rostrally under the postparietals and/or parietals (it doesn’t even need to be exposed on the occipital skull surface at all); third, Pardo & Anderson (2016) blow this idea out of the water, because if there had been any trace of a suture surface, they’d have seen it in their scans.

    Also, I’ve been trying to explain to you for years and years and years that “sister taxon” doesn’t mean “close relative” – it means “the single one closest relative“, of which each clade has exactly one unless it’s part of a hard polytomy or it’s the largest clade that contains all known life.

    A more typical supraoccipital may be peeking out as a sliver over the foramen magnum (spinal nerve opening, beneath the fused postparietals.

    No (Pardo & Anderson, 2016).

    No sister taxa separate the postparietals, so those in light red are identified here as tabulars, bones which typically form the posterior rim of sister taxa skulls and often provide corners to the skull.

    I’ve already invented the square wheel (Marjanović & Laurin, 2008, Contr. Zool.). Why do you bother reinventing it? It really is square, assuming as it does that the postparietals are completely separated and the suproccipital contacts the parietals on the surface of the skull roof; that’s how Wellstead reconstructed it, but most, perhaps all, specimens don’t actually show this condition – instead the postparietals touch each other over a varying amount of their length (Pardo & Anderson, 2016; me, pers. obs. of two specimens; Marjanović & Laurin, 2016).

    the bright green supratemporals

    No (Pardo & Anderson, 2016). Seriously, there’s a whole suture missing in your fig. 2.

    the postorbitals (yellow/amber)

    No (Pardo & Anderson, 2016).

    They form skull corners in B. elongulatus [sic!] and rise above the plane of the cranium in B. newberryi

    Actually, the nomenclature is a mess; there are at least two species with very different trunk lengths out there, but that feature isn’t preserved in either type specimen, and neither are the other potentially diagnostic features. It sucks. And then there’s Molgophis to deal with, and who knows what else. Sorting this mess out is going to be a lot of work.

    unless a sliver of it is fused to

    That’s, like, second-order speculation.

    The postfrontal is fused to the parietal

    Well, either that, or it’s gone and the parietal has grown to fill the space because the postfrontal didn’t stop it anymore. Given the lack of evidence for the first option (Pardo & Anderson, 2016, who looked inside the bones, too), I’ll tentatively go with the second one…

    The postorbital is

    part of your reinvention of the square wheel (Marjanović & Laurin, 2008, 2016; Pardo & Anderson, 2016).

    The squamosal is the tricky bone. It appears as a separate bright magenta element in B. elongulatus [sic]

    No, that element is fictitious (Wellstead, 1991; Pardo & Anderson, 2016). That’s not the only time Carroll misinterpreted a break.

    I see enough differences here to warrant separate genera

    There’s no way to argue for or against this – genera are subjective.

    Pardo and Anderson freehand
    a Brachydectes newberryi skull reconstruction to supplement their CT scans, but do not label the bones in the drawing.

    That’s because, first, it’s not necessary because it’s all obvious, and second, it’s not necessary because the scan pictures are labeled. Didn’t you see them?

    Also, the line drawing doesn’t really supplement the scans. It summarizes them. It’s an interpretation of the scans.

    Finally, they called all lysorophians “Brachydectes newberryi“. That wasn’t a good idea, see above. The specimens they scanned were, or would have been, all considered “B. elongatus” by Wellstead (1991), but that’s only marginally a better idea.

    Unfortunately, [close relatives] among the microsaurs do not have a large supraoccipital.

    Oh, some do, and that’s even in Carroll & Gaskill (1978). Remember: most of the suproccipital is not exposed on the skull surface.

    Of course, those are not derived amniotes.

    Well… I don’t think they are; but Pardo & Anderson didn’t make this up, you know. Wait for the paper, then we can start to discuss this.

    We’re all learning

    You know, you’d learn a lot more if you read the papers you cited – instead of reading the abstract, scrolling through to find pretty pictures, musing about these pictures without even wondering what the text says about them, and posting here.

    You’d also learn a lot more if you read the comments to your posts before publishing the next post in which you make the same mistakes again and again and again and again and again and again and again and again.

    Your method of learning is really inefficient.

  2. Jason Pardo is having trouble posting the following comment:

    =================================================
    Again, thank you very much for your interest in our paper.

    I’m sorry that you’re still confused about the skull. I realize that you do not like to take authors’ interpretations of sutures and bone homologies at face value, so what I’d strongly suggest is that you consult Sollas (1930) and Bolt & Wassersug (1975), who both address some aspects of the lysorophian skull using serial sections (serial grinding by Sollas, and thin sections by Bolt & Wassersug), and find very similar aspects of anatomy where our studies overlapped.

    As far as fusions vs losses of bones, I agree that this is a difficult problem and that in some cases this is very ambiguous. In some cases (e.g. whether the triangular bone at the corner of the skull is a tabular or supratemporal) there is no strong criterion by which one can distinguish between the two options. I will note that this is also the case for essentially all ‘microsaurs’ with a single element in the temporal region, but that whatever this element is, it’s probably always the same bone.

    In other cases, though, we have very strong reasons for drawing the inferences we have. The supraoccipital is a supraoccipital because it is an endochondral bone that fills the synotic tectum and participates in the dorsal portion of the otic capsule. The morphology is the same, for all intents and purposes, as the supraoccipital in other recumbirostrans, such as Rhychonkos, Huskerpeton, Cardiocephalus, etc. So it’s in the right position to be a supraoccipital, it has the right morphology to be a supraoccipital, and it is composed of the right tissue to be a supraoccipital. This, to me, is pretty convincing that this is a supraoccipital.

    If the presence of a bone, in this case the supraoccipital, conflicts with your phylogenetic scenario, that means one of two things. First, it could mean that there is additional convergence in the dataset. This is possible. Secondly, and more importantly, it could mean that your phylogenetic scenario is just incorrect. But it isn’t scientifically defensible to reinterpret basic bone homologies based on what you want the animal to be.

    I agree that there are probably multiple species contained within the taxa we have identified as “lysorophians” but what we considered important in the Council Grove Group material is that all the cranial characters used by Wellstead to differentiate between B. newberryi, B. elongatus, and Pleuroptyx clavatus are ontogenetic, so a lot of the material that Wellstead attributed to B. elongatus cannot be distinguished from B. newberryi. We do happen to have complete postcranial sequences among some of the Council Grove Group lysorophians and they are consistent with B. newberryi in terms of vertebral counts, so these at least are likely all B. newberryi. However, the scenario that Wellstead had proposed (where B. newberryi was stratigraphically restricted to the Carboniferous and B. elongatus represented the transition to the Permian) is incorrect. Further revision of the Texas-Oklahoma material is necessary, as well as detailed description of Dunkard Basin material from localities outside of Linton.

    I’m not going to comment on your phylogenetic work except to say that it is circular to interpret cranial homologies based upon what you want the phylogeny to be and then to use those interpreted homologies to guide your reconstruction of phylogeny.

    Again, thank you for your interest in our study and I’m sorry that you are having difficulties navigating our figures. We tried to make them as clear as possible.
    =================================================

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s