Zhongornis: NOT the sister taxon of all pygostylians

While we’re on the subject of pygostyles…
Yesterday we looked at a recent paper by Wang and O’Connor (2017) about the evolution of the pygostyle in birds. Reiterating here: Unfortunately Wang and O’Connor did not use several Solnhofen birds (traditional Archaeopteryx specimens) in their analysis. So Wang and O’Connor did not realize the split between the scasoriopterygids, enantiornithes and ornithuromorpha occurred prior to those Late Jurassic Solnhofen long-tailed birds, according to the large reptile tree (LRT, Fig. 2). Thus the pygostyle that developed by convergence several times over. And that fact is pertinent to today’s discussion.

Figure 2. Bird cladogram focusing on Zhongornis. Each of the three colored bird clades independently prodded members with a pygostyle.

Figure 1. Bird cladogram focusing on Zhongornis. Each of the three colored bird clades independently prodded members with a pygostyle. Close observers will note one node within Ornithuromorpha has flipped since yesterday.


Which brings us to Zhongornis haoae 
(Fig. 2; D2455 ⁄ 6). Gao et al. 2008 considered their find the sister taxon to all pygostylians. The LRT does not support that nesting because, like Wang and O’Connor, Gao et al. did not use more than one Solnhfoen bird (Archaeopteryx) in their cladistic analysis (Fig. 3) and thus their cladogram was likewise flawed due to taxon exclusion.

Figure 1. Zhongornis. Here DGS finds teeth and a sternum overlooked originally.

Figure 2. Zhongornis. Here DGS finds teeth and a sternum overlooked originally.

 cannot be the sister-taxon of all pygostylians, contra the assertions of Gao et al. because the pygostyle developed four times by convergence. It stands to reason that no single taxon can be the sister to all four. However, in the LRT Zhongornis does indeed nest at the very base of the Ornithuromorpha between the Archaeopteryx grandis + Confuciusornis and Archaeornithura (Fig. 1).

Among the more derived taxa
are Archaeornithura and Hesperornis both of which do not have a pygostyle.

A more basal taxon,
 has a pygostyle and so it represents yet another convergent development within the Ornithuromorpha. Wikipedia reports that Sapeornis is close to Omnivoropteryx, but the LRT does not support that relationship (see Fig. 1) despite similarities such as a perforated deltopectoral crest on the humerus.

Figure 3. The nesting of Zhongornis according to Gao et al. 2008. They did not employ more than one Archaeopteryx specimen, which is the major fault (taxon exclusion) in this cladogram.

Figure 3. The nesting of Zhongornis according to Gao et al. 2008. They did not employ more than one Archaeopteryx specimen, which is the major fault (taxon exclusion) in this cladogram. See figure 1 for an update from the LRT.

According to Gao et al.
Lack of fusion and bone texture indicate the Zhongornis holotype is a juvenile. The femoral heads and necks are not visible, perhaps not yet ossified. Even so, the wing feathers are well-develped, so the specimen is not a hatchling, but close to fledging, according to Gao et al.

Gao et al. report,
“Possessing a unique hand morphology with a phalangeal formula of 2-3-3-x-x and a reduced number of caudal vertebrae lacking a pygostyle the new specimen reveals anatomical information previously unknown and increases the taxonomic diversity of primitive, non-pygostylian birds. We infer from the specimen that during the evolution of the avian tail, a decrease in relative caudal length and number of vertebrae preceded the distal fusion of caudals into a pygostyle.”

Contra the Gao et al. diagnosis

  1. Tiny teeth are apparent in the Zhongornis photo, both on the premaxilla and maxilla. Note that some derived taxa also have tiny teeth. Confuciusornis is toothless but that occurred by convergence from a more basal Late Jurassic split and it represents a sterile lineage.
  2. Zhongornis may not have had a unique manual phalangeal formula. The base of digit 3 is hidden beneath the base of digit 2 (Fig. 4). If other specimens are known that expose this data, please let me know.
  3. A corner of the sternum is visible (Fig. 2) largely beneath the anterior dorsals.
Figure 2. Manus of Zhongornis. The base of digit 3 is hidden behind digit 2.

Figure 4. Manus of Zhongornis. The base of digit 3 is hidden behind digit 2.

Nomenclature issues
Earlier several traditional clades like Ornithodira and Parareptilia were shown to be paraphyletic in the LRT. Now the clade Pygostylia also appears to be paraphyletic when more taxa (in this case more Solnhofen birds) are included.

Gao C-L, Chiappe LM, Meng Q-J, O’Connor JK, Wang X, Cheng X-D and Liu J-Y 2008. A new basal lineage of early Cretaceous birds from China and its implications on the evolution of the avian tail. Palaeontology 51(4):775-791.
Wang W and O’Connor JK 2017. Morphological coevolution of the pygostyle and tail feathers in Early Cretaceous birds. Vertebrata PalAsiatica 2017:10: 55:3: 1-26.

6 thoughts on “Zhongornis: NOT the sister taxon of all pygostylians

  1. Unsurprisingly I think your cranial identifications are wildly inaccurate, and the elements you draw look nothing like any Mesozoic bird. I do agree with your idea a short manual phalanx III-1 might be present underneath the second digit, and used this as part of my hypothesis Zhongornis is a baby confuciusornithid.

    In any case, I mostly commented to say that Pygostylia is not polyphyletic because it has a phylogenetic definition that doesn’t allow that. Two, actually. The more commonly used is the last common ancestor of Confuciusornis and modern birds and all of its descendants (Chiappe, 2001). In your tree that includes Zhongornis, but excludes Sapeornis and enantiornithines. Indeed, most phylogenies find Sapeornis to be outside Pygostylia despite having a pygostyle.

    The less used definition is Gauthier and de Quieroz’s (2001), which states pygostylians must have a pygostyle homologus with Vultur gryphus. In that case, Pygostylia could be either of two clades in your tree, depending on how you map the character out-
    – Sapeornis+ if grandis and Zhongornis are reversals.
    – Archaeornithura+ if Sapeornis+Chiappeavis and Confuciusornis are convergences.
    Both are equally parsimonious. The fact it’s ambiguous is one reason most people don’t like apomorphy-based definitions like this. Ironically the second option does make Zhongornis the sister to Pygostylia in your tree.

    Similarly, Parareptilia exists in your trees under two of its proposed definitions I know of.
    – The Diadectes-Chelus clade under Reisz (2004).
    – The Milleretta+ clade under Tsuji and Muller (2009).
    Ornithodira is everything past Gephyrostegus [SIC in your tree] in your tree under all of its proposed definitions.

    Btw, you really should use the proposed genera for the referred Archaeopteryx specimens since you think they’re not Archaeopteryx.
    Archaeornis for siemensii Berlin, Jurapteryx for recurva, and Wellnhoferia for grandis.

    • Thanks, Mickey. Parareptilia also includes mesosaurs, so that’s an issue. The Archaeopteryx webpage at ReptileEvolution.com includes 2/3 of those proposed names. The third will be added. Let me know if you see the Solnhofen birds separated in any phylogenetic analyses. I’d like to see that again. With regard to pygostyles, let’s just go back to basics. Pygostyles of several styles developed by convergence in the LRT. Re: Ornithodira is a junior synonym for Reptilia. So let’s ditch it.

      • But Parareptilia doesn’t have to include mesosaurs. Under those two definitions at least. That’s the beauty of phylogenetic taxonomy- you don’t have to rename or abandon clade names just because some group doesn’t belong.

        The Solnhofen archaeopterygids haven’t been examined as OTUs in published analyses, except for Senter (2010) and its descendants. He found Wellnhoferia to be sister to ‘Archaeopteryx’ (ironically excluding the holotype, so that the OTU is only Archaeornis, bavarica and Thermopolis). Your results are due to character exclusion and inaccurate reconstructions.

        Yes, pygostyles evolved convergently. In frogs and birds, if nothing else. But plausibly in confuciusornithids and omnivoropteryids as well.

        And yes, a junior synonym of Reptilia, but not polyphyletic.

      • Point is, when you make an accusation, the jury will side with you if you back that up with evidence. And in the spirit of encouragement, do you have a theropod/bird paper under review or planned for publication? If not, you should.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.