Thursday, August 5, 2010

Shenshiornis is *gasp* another young Sapeornis

Speaking of synonyms, I was hardly surprised to learn that yet again, another supposed new "sapeornithid" is just plain old Sapeornis chaoyangensis (like Omnivoropteryx, Didactylornis and Sapeornis angustis).  First, I want to implore everyone to use Omnivoropterygidae instead of Sapeornithidae as it was published four years prior.  Yes I hate Czerkas' paper as much as anyone, but dislike of a work is no reason to ignore the principle of priority.  Of course that would require acknowledging Omnivoropteryx even exists, which almost no one seems willing to do either.  What makes it especially deplorable is that both Shenshiornis and Sapeornis angustis were diagnosed partly by characters that were also supposed to distinguish Omnivoropteryx from Sapeornis.  It's almost like Hu et al. and Provini et al. were describing Omnivoropteryx under new names.  Not that I know anyone's motives, but the authors must be aware of Omnivoropteryx (Xu and Zhou are coauthors, and Czerkas' volume could not have skipped by them) but neither paper even cites Czerkas and Ji, 2002.  Am I the only one who finds Omnivoropteryx's near absence from the literature suspicious?

In any case, on to Shenshiornis.  The paper itself is typical of Mesozoic bird descriptions, with a brief osteology and empty discussion.  The high resolution color photos are outstanding though, and the skull deserves a closer look, as (besides Omnivoropteryx which only has X-rays published) it is by far the best preserved of any published omnivoropterygid.  The adjacent illustration does a poor job of showing the antorbital morphology, missing the antorbital fossa and maxillary fenestra for instance.

Skull of Sapeornis chaoyangensis (Shenshiornis primita holotype LPM B00018) after Hu et al. (2010).

Sapeornis Zhou and Zhang, 2002
= Shenshiornis Hu, Li, Hou and Xu, 2010
S. chaoyangensis Zhou and Zhang, 2002
= Shenshiornis primita Hu, Li, Hou and Xu, 2010
Early Albian, Early Cretaceous
Jiufotang Formation, Liaoning, China
- (LPM B00018; holotype of Shenshiornis primita) (subadult) skull (40 mm), mandibles (40 mm), eleven cervical vertebrae, cervical ribs, eleven dorsal vertebrae, five dorsal ribs, gastralia, synsacrum, more than ten caudal vertebrae, few chevrons, ilia (44 mm), pubes (57 mm), ischium, femora (62 mm), tibiotarsi (64 mm), fibulae, distal tarsal, metatarsals I (8.9 mm), phalanges I-1, pedal ungual I, metatarsals II, phalanx II-1, phalanx II-2, pedal unguals II, metatarsals III (32 mm), phalanges III-1, phalanx III-3, pedal unguals III, metatarsals IV, phalanges IV-1, phalanges IV-2, phalanges IV-3, phalanges IV-4, pedal unguals IV, metatarsal V (11.3 mm) (Hu et al., 2010)

Comments- Discovered in 2005, the partial skeleton LPM B00018 was described as Shenshiornis primita by Hu et al. (2010). They recognized it as a subadult specimen based on cervical ribs unfused to vertebrae, intercentral sutures visible on sacrum, and unfused metatarsals. This is further indicated by the small size, short pubis and perhaps unfused distal caudals. They distinguished it from Sapeornis based on several characters. The prenarial portion of the premaxilla is supposedly shorter, but it is identical to the situation Omnivoropteryx and longer than JZPM-LSV-130. The apparently different proportions in IVPP V13275 and V13276 are due to them being in dorsal view. The left premaxilla of IVPP V13276 also has an elongate subnarial process, suggesting the seemingly short process on the right side and in JZPM-LSV-30 is due to breakage. Premaxillary and maxillary teeth are present in all omnivoropterygids, just as dentary teeth are absent. Didactylornis, IVPP V13275 and JZPM-LSV-30 all have subtriangular tooth crowns wider than their roots as well. The cervical centra are described as amphicoelous or amphiplatyan, while those of Sapeornis (Zhou and Zhang, 2003; Provini et al., 2009) and Didactylornis have been described as heterocoelous. However, basal avialans generally have variation within the neck, with posterior cervicals being amphicoelous while anterior cervicals are only semi-heterocoelous (heterocoelous anterior articular surface but primitively concave posterior one). This intermediate state has led to taxa like Confuciusornis being described as both amphicoelous and heterocoelous by different authors. In fact, both Confuiciusornis and Shenzhouraptor are coded as amphicoelous like Shenshiornis in Hu et al.'s matrix, although descriptions of both genera have indicated they have the semiheterocoelous state. The photographed cervicals certainly don't rule out a semiheterocoelous condition in Shenshiornis, and considering the issues above, I don't view this as a definitive or distinct character of the taxon. Though most specimens are not preserved and/or illustrated in sufficient detail, at least CAGS-03-07-08 shares the elongate posterior cervical postzygapophyses of Shenshiornis. The new genus was said to differ from Sapeornis angustis and Didactylornis in having seven sacrals, as in S. chaoyangensis. Yet the posterior sacrum of S. angustis is covered by pelvic elements and has enough space for two other vertebrae. In Didactylornis, the sacrum is stated to be poorly preserved and the figure suggests it is broken and partially covered by other elements. Notably Yuan (2005) coded Didactylornis as having the same number of sacrals as Sapeornis, even though he wrote only five could be observed. Hu et al. state more than ten free caudal vertebrae are preserved in Shenshiornis. IVPP V13275 has six or seven, S. angustis has at least seven, and Didactylornis' number is estimated at less than eight. However, the caudal series in Shenshiornis is poorly preserved and only loosely articulated. Even if true, the lack of distal fusion may be ontogenetic as in Zhongornis and some enantiornithines, as this specimen is smaller than any Sapeornis with a preserved pygostyle. The anteriorly pointed ilium is not necessarily different from other specimens (which are hidden or broken), and indeed seems to be the case in the fragmentary right ilium of CAGS-03-07-08. While the S. chaoyangensis holotype was schematically illustrated as having shorter rounded preacetabular processes on both ilia, the photo of the right ilium shows it to be fragmented and partly covered anteriorly by elements not indicated in the original figure. Thus the shape and completeness of the left ilium is also called into question. The short pubis is expected in a subadult and is longer (~92%) than in Omnivoropteryx (84%). The supposedly long metatarsal I (28% of metatarsal III) is indeed a bit longer than in IVPP V13275 (26%), but is the same as in CAGS-03-07-08 and actually shorter than in Didactylornis (33%). Metatarsal V (35% of metatarsal III, not ~40% as stated in the diagnosis) is also supposedly long, and is actually longer than the available measurements for CAGS-03-07-08 (15%) and IVPP V13275 (23%). In CAGS-03-07-08 at least, the elements may only be shaft fragments as one doesn't extend to the proximal tarsometatarsal edge while the other lies diagonally across metatarsals III and IV. Note too that most specimens (including the Sapeornis holotype) cannot be measured for this ratio, which has yet to be studied ontogenetically. Thus of the characters listed in Shenshiornis' diagnosis, only the metatarsal V 12% longer than IVPP V13275 is even possibly valid and could not be used to divide most omnivoropterygid specimens in any case. Like Omnivoropteryx and Sapeornis angustis, there is no reason to consider Shenshiornis anything other than a young Sapeornis chaoyangensis.

References- Czerkas and Ji, 2002. A preliminary report on an omnivorous volant bird from Northeast China. Feathered Dinosaurs and the Origin of Flight. The Dinosaur Museum Journal. 1, 127-135.

Zhou and Zhang, 2002a. Largest bird from the Early Cretaceous and its implications for the earliest avian ecological diversification. Naturwissenschaften. 89, 34-38.

Zhou and Zhang, 2003. Anatomy of the primitive bird Sapeornis chaoyangensis from the Early Cretaceous of Liaoning, China. Canadian Journal of Earth Sciences. 40, 731-747.

Yuan, 2005. Restudy on sapeornithids from the Lower Cretaceous of Yixian County, Liaoning. PhD Thesis. China University of Geosciences. 157 pp.

Yuan, 2008. A new genus and species of Sapeornithidae from Lower Cretaceous in Western Liaoning, China. Acta Geologica Sinica. 82(1), 48-55.

Provini, Zhou and Zhang, 2009. A new species of the basal bird Sapeornis from the Early Cretaceous of Liaoning, China. Vertebrata PalAsiatica. 47(3), 194-207.

Hu, Li, Hou and Xu, 2010. A new sapeornithid bird from China and its implication for early avian evolution. Acta Geologica Sinica. 84(3), 472-482.


  1. I've heard one reason nobody cites Czerkas' specimens is that they were smuggled out of China, and Czerkas promised to return them in a certain amount of time, but his traveling exhibit kept going on and on... essentially, they're still in an illegal private collection.

    But you're right, spite is not a good reason to ignore validly published nomenclature.

  2. Consider the option of "imploding" Omnivoropterigidae into Sapeornis, depending on the degree of differentiation involved.