Friday, November 16, 2012

Wang et al. 2012 - for once I'm against a paper synonymizing Jehol birds?!

Wang et al. have a new paper in press synonymizing Jixiangornis with Shenzhouraptor (their Jeholornis).

EDIT: I've been in communication with Dyke (who has been an exemplary scientist in his response) since posting this and it seems the paper was not supposed to be online yet and was taken down since Creisler's DML post so that more work could be done.  Gotta love the consequences of publishing submitted manuscripts early, and how things like this and Brontomerus' early surprise publication happened. ;)  Actually, the consequences here might be very good indeed....

Before I get into the meat of the paper, I'd like to express my continuing disappointment with how Yandangornis is handled by everyone, when it's mentioned at all.  Wang et al. claim Zhou and Zhang (2006) "reviewed Yandangornis longicaudus and concluded that it 'lacks the diagnostic characters of birds' and thus may also be a non-avian dinosaur".  In actuality, Zhou and Zhang merely said "Unfortunately, this fossil was only preliminarily described, showing no diagnosis of birds.  As a result, it remains a mystery whether it is a bird or bird-like dinosaur."  That counts as a review?!  Especially when Yandangornis was described as having numerous characters of basal ornithurine (sensu Gauthier) birds including the toothless upper jaw (with pointed symphysis), reduced number of caudal vertebrae (with rod-like distal caudals), fused sternum with median bulge, trochanteric crest, completely fused tibiotarsus, fibula not contacting tarsus and fused metatarsus.  I fear Yandangornis is becoming the Longisquama of birds- being curtly dismissed as an unknown factor without even trying to evaluate it.

On to the main topic.  Wang et al. present a new specimen of basal ornuthurine from the Yixian Formation, YFGP-yb2.  They claim this specimen combines characters of Shenzhouraptor and Jixiangornis, showing both to be synonymous.

First they say it has no maxillary teeth, unlike 'Jeholornis palmapenis' but like 'J. prima' and Jixiangornis (wouldn't this be itself a reason to separate the former species, if it were true and thus notable enough to comment on?).  Yet jeholornithid teeth are often unpreserved due to their small size and low number, as seen by palmapenis' lack of dentary teeth (considered preservational by O'Connor et al., 2010) and LPM 0193's (the Shenzhouraptor holotype) lack of any recognizable teeth. Even palmapenis only preserves one of the at least two maxillary teeth it had on that side. Thus the absence of maxillary teeth in other specimens may easily be preservational, so that Wang et al. simply accept reportedly absent teeth as such seems naive.

The scapula is said to narrow distally as in Shenzhouraptor, but this is true of almost all ornithurines including Jixiangornis, so means little.  The scapula was known to be curved in Jixiangornis previously (Nesbitt et al., 2009; and in Sapeornis), so that means little too.  The coracoid is correctly stated to be more elongate than Shenzhouraptor or Jixiangornis, which is interesting.  It's stated to be strut-like as in Jixiangornis (and by implication unlike Shenzhouraptor), but Jixiangornis has the less strut-like coracoid as can be seen in figure 82 of Turner et al. (2012).  Though Wang et al. claim the new specimen shares a well developed lateral coracoid process with Jixiangornis and not Shenzhouraptor, the latter has a better developed process than either (ironically making Wang et al.'s pairing right, but for the wrong reason).  [Edit: actually some individuals of both species have large processes, and some small ones] They claim the new specimen has the distal half of the medial coracoid margin convex as in Jixiangornis and unlike Shenzhouraptor, but in fact it differs from all prior specimens in having a completely concave medial margin distally.  The supposed furcula is said to be V-shaped and slender, so more like Shenzhouraptor, but what's identified is extremely slender like enantiornithines' and Jixiangornis' furcula is actually more V-shaped than Shenzhouraptor's (though nothing near this specimen's supposed anatomy).  Both would differ from the new specimen in lacking a large hypocleidium.  In fact, I'm nearly certain what Wang et al. think is a furcula is a dorsal rib and parts of dorsal vertebrae [Edit: I've since found the actual furcula under the coracoids].
Supposed furcula of basal ornithurine YFGP-yb2 (top) as photographed by Wang et al. (2012- figure 2A), (middle) as illustrated by Wang et al. (2012- figure 2B), (bottom) from an unpublished photo.

Wang et al. claim the new specimen is closer in radioulnar width ratio (claimed 50%) to Shenzhouraptor, citing a 50% ratio in 'Jeholornis prima', but Zhou and Zhang's (2002) figures show it closer to 65% in the latter.  Shenzhouraptor's holotype is ~71% and the new specimen's is ~64% based on an unpublished photo.  What's sad is that they cite Jixiangornis' ratio as "less than 0.7", basing this on a matrix coding by O'Connor et al., when Ji et al. (2002) explicitly listed widths in their forelimb measurement table (4.5 / 6.8 mm = 66%).  Did they not even read its original description?  Wang et al. claim the new specimen is more similar to Shenzhouraptor than Jixiangornis in lacking an intermetacarpal scar, saying it was "coded as potentially present in Jixiangornis by O’Connor et al.".  Potentially present?  Yuan (2005) and Turner et al. (2012) both code it as absent in Jixiangornis.  The new specimen (106%) supposedly resembles 'Jeholornis prima' (105-108%) more than Jixiangornis (104%) in its short manuohumeral ratio, but this isn't true.  And it's only a single percent difference anyway.  How is that important to note?  Annoyingly, this ratio isn't listed in the measurement comparisons table, nor is manual length, so you have to multiply the listed ulnohumeral and manuoulnar ratios to get it.  Shenzhouraptor's type has a ratio of 117% btw.  The ratio between manual unguals I and II is said to be more similar to Jixiangornis, with the new specimen having a larger ungual I, but the schematic drawing of Jixiangornis by Ji et al. (2002) would suggest it had a larger ungual II.  Shenzhouraptor has a larger ungual I, so is actually the taxon closest to the new specimen, but the schematic nature of Ji et al.'s drawing, small amount of difference in all specimens and difficult to separate claw sheaths makes any comparison unimportant.  The length of manual digit II is said to be more similar to Shenzhouraptor, but the ratio is identical to Jixiangornis (98% for II-1+II-2 / mcII). 

The high tibiofemoral ratio was suggested to be ontogenetic despite the smaller palmapenis type (which they cited earlier) having a larger ratio.  Wang et al. are wrong in claiming Jixiangornis has an incompletely fused tibiotarsus (Ji et al., 2002; Turner et al., 2012).  They claim the pedal unguals of 'Jeholornis prima' and the new specimen are more slender and less curved than Jixiangornis, but the opposite is true comparing Ji et al.'s illustration with figure 3 of Zhou and Zhang (2006), and the new specimen shows variation caused by keratin sheaths.

Wang et al. end with this paragraph, with my interjections bracketted- "To sum up, YFGP-yb2 shares the following features with Jeholornis prima that are absent in Jixiangornis: subequal manus and humerus lengths [untrue and a 1% difference]; similar ratio between the radius and ulna shaft widths [~1% difference]; and the absence of an intermetcarpal tubercle on metatarsal II [wow, the intermetAcarpal scar on metaCARPAL II, with the same mistake also in the abstract; in any case probably absent in Jixiangornis too]. YFGP-yb2 shares the following features with Jixiangornis that are absent in Jeholornis prima: large sized ungual [er, luckily we know this is manual ungual I from the discussion, but is probably untrue in Jixiangornis, difficult to evaluate in any specimen due to claw sheaths and is a small difference anyway] and coracoid with less convex medial margin [true, though the completely concave distal margin is unlike either of them] and less developed external process [true, though contradicting the earlier text Edit: false, as it varies in both species]."

Their discussion merely compares the new specimen, Jixiangornis and Shenzhouraptor with Zhou and Zhang's (2006) diagnosis for Jeholornis prima.  And yes, Zhou and Zhang used a lot of bad characters that are also found in Jixiangornis.  But what about Ji et al.'s diagnosis for Jixiangornis?  What about the differences noted by Yuan (2005), Turner (2008) and Nesbitt et al. (2009)?

This paper is flawed for so many reasons.  It presents incorrect data often.  Several times it contradicts itself, and has frequent grammatical and spelling errors.  The new specimen isn't even described, with only one photo and a schematic illustration.  Very odd features of the specimen aren't even noted- ~16 caudals [Edit: only the first twelve are preserved]; completely concave medial coracoid margin distally; only two non-ungual phalanges on manual digit III [Edit: it actually has three, which are interestingly twice the length of Shenzhouraptor's; the right digit III appears articulated but isn't]; subarctometatarsalian pes [Edit: not different from Shenzhouraptor or Jixiangornis] (the furcular morphology would be way too weird to pass by as well if it were real).  That whole furcular misidentification issue.  Many times facts are brought up that have no bearing on the synonymy issue, as when the new specimen resembles both taxa.  The main argument of the paper is because this specimen shows exclusive features of each taxon, those taxa are synonymous, yet these differences are often ~1% ratios and most proposed differences aren't even mentioned (dentary teeth; caudal count; fused sternum; stratigraphic level; just to take some mentioned by Ji et al., 2002).  There's no attempt to examine the literature for differences between each taxon, let alone the specimens.  The measurement table lacks both IVPP V13550 and palmapenis, not to mention the CAGS Jixiangornis specimen Turner (2008) and Nesbitt et al. (2009) used, and lacks one of the basic measurements mentioned in the text.  Perhaps most sad is they didn't even seem to read the original description of the taxon they're sinking, not citing its diagnosis and citing a matrix coding by someone else instead of a measurement explicitly listed in the description's Table 1.

Luckily, "This file will be reviewed by the authors and editors before the paper is published in its final form. Please note that during the production process errors may be discovered which could affect the content."  I'm sending this critique to the authors in hope that this paper is fixed before publication, though honestly fixing it destroys its entire point.  Are Shenzhouraptor and Jixiangornis synonyms?  This would require evaluation of over 50 supposed differences mentioned in the literature I noted above, but not by Wang et al..  In the meantime, I think YFGP-yb2 is certainly not Shenzhouraptor, and probably not Jixiangornis (may be a small adult?).  It makes for a more exciting paper in my opinion.

References- Ji, Ji, You, Zhang, Yuan, Ji, Li and Li, 2002. Discovery of an Avialae bird - Shenzhouraptor sinensis gen. et sp. nov. - from China. Geological Bulletin of China. 21(7), 363-369.
Ji, Ji, Zhang, You, Zhang, Wang, Yuan and Ji, 2002. A new avialian bird - Jixiangornis orientalis gen. et sp. nov. - from the Lower Cretaceous of Western Liaoning, NE China. Journal of Nanjing University (Natural Sciences). 38(6), 723-736.
Zhou and Zhang, 2002. A long-tailed, seed-eating bird from the Early Cretaceous of China. Nature. 418, 405-409.
Zhou and Zhang, 2003. Jeholornis compared to Archaeopteryx, with a new understanding of the earliest avian evolution. Naturwissenschaften. 90, 220-225.
Yuan, 2005. Restudy on sapeornithids from the Lower Cretaceous of Yixian County, Liaoning. PhD Thesis. China University of Geosciences. 157 pp.
Zhou and Zhang, 2006. Mesozoic birds of China- A synoptic review. Vertebrata PalAsiatica. 44(1), 60-98.
Turner, 2008. Phylogenetic relationships of paravian Theropods. PhD Thesis. Columbia University. 666 pp.
Nesbitt, Turner, Spaulding, Conrad and Norell, 2009. The theropod furcula. Journal of Morphology. 270, 856-879.
O'Connor, Sun, Xu, Wang and Zhou, 2012. A new species of Jeholornis with complete caudal integument. Historical Biology. 24(1), 29-41.
Turner, Makovicky and Norell, 2012. A review of dromaeosaurid systematics and paravian phylogeny. Bulletin of the American Museum of Natural History. 371, 1-206.
Wang, Dyke and Godefroit, in press. A new specimen of a Jeholornis-like long-tailed bird shows that Jixiangornis is a junior synonym of Jeholornis prima. Acta Palaeontologica Polonica. 14 pp.


  1. Based on the photo, the distal end of the tail seems missing.

    1. Yeah, looking closer I agree with you. That's what I get for trusting their table with a tail length listed...

    2. Interestingly, the Jixiangornis holotype seems to have a genuinely short tail compared to many Jeholornis specimens of the same size (and presumably ontogenetic stage).

  2. Hi Mickey Mortimer,

    Regarding jeholornithiforms, the bird specimen YFGP-yb-2 has been named as a new species of Jeholornis, J. curvipes (Lefevre et al. 2014). Jixiangornis is recovered as sister to Pygostylia and Ornithothoraces. Since the manuscript by Wang et al. was taken down due to several grammatical flaws, do you feel inclined to remove this post?

    Ulysse Lefèvre, Dongyu Hu, François Escuillié, Gareth Dyke and Pascal Godefroit (2014) A new long-tailed basal bird from the Lower Cretaceous of north-eastern China. Biological Journal of the Linnean Society 113 (3): 790–804 Special Issue: Celebrating Dinosaur Island DOI: 10.1111/bij.12343

    1. I don't think removing the post would be appropriate, since all of the above did happen, and indeed my posting this seems to have been a big reason why there's now a much improved version out. I don't agree with the new paper either, but at least it has better argumentation.