Sunday, October 28, 2012

Did Dromiceiomimus really have long vaned secondaries?

For what will possibly be my last post before the big move (the blog will keep the same address, but we'll see how the Database itself is affected), Zelenitsky et al. (2012) recently reported a somewhat surprising discovery- wings in ornithomimids.

First, note it's not actually Ornithomimus edmontonicus the paper is about.  The supposed secondary shaft marks are from RTMP 95.110.1, which is Dromiceiomimus* samueli.  The other two, being from the Hoseshoe Canyon Formation, are probably D. brevitertius.  For some reason, everybody seems to want to ignore that edmontonicus was named in 1933, while brevitertius was named in 1926 and samueli in 1928.  Zelenitsky et al. follow Makovicky et al.'s (2004) Dinosauria II chapter which synonymized all of these species, but the name would still be brevitertius, not edmontonicus, as both published references state.  Principle of priority people!

 * Russell only placed edmontonicus in Ornithomimus because of its long metacarpal I, and Makovicky et al. did not list any additional justification. Yet O. velox has never been included in a published cladistic analysis as a separate OTU, and when it is (unpublished data), it does not clade with brevitertius (= edmontonicus) or samueli. For instance, O. velox lacks appressed metacarpals II and III and seems to have a medial distal condyle on metacarpal I positioned higher than its lateral condyle, both less similar to Dromiceiomimus than Anserimimus is.

With that out of the way, vaned feathers in ornithomimids shouldn't really be that surprising.  As Martyniuk (online 2012) stated, Currie and Chen (2001) noted Sinosauropteryx's seemingly stage 1 feathers were probably vaned, and there have been a few references since that discovered modern vaned feathers become difficult to diagnose as such after crushing and decay.  So the supposedly simple body feathers we have in e.g. Beipiaosaurus or Scuirumimus may have originally been vaned.  Pelecanimimus does have large paired sterna with ossified ribs and possibly uncinates, so maybe basal ornithomimosaurs were more flight-y anyway.  Not only that, but the seemingly solid position of ornithomimosaurs outside Oviraptorosauria+Paraves isn't quite as definite as it seems.  It's just that basically all recent coelurosaur matrices inherit the character bias and miscodings of the TWG matrix.  So no surprise their topologies are all similar.

I was however surprised that Horseshoe Canyon and Dinosaur Park specimens could preserve feathers.  The two Horseshoe Canyon specimens both show what have been commonly identified as stage 1 or 2 feathers, but which may be degraded state 3 feathers as noted above.  Those on the forelimb are short, but who knows how complete they are.  The Dinosaur Park specimen has been known for a while and is the source of the supposedly adult winged morphology described in the new paper.  This is based on black marks on the radius and ulna which trend posterodistally to distally, and sometimes are U- or hook-shaped as if they had hollow centers.  Now maybe these are feather remains, though I don't know of any other theropod which has preserved feathers this way.  Yet as Mcfeeters noted on the DML, they only show the shaft, not any vane.  Couldn't these just be stage 1 feathers, quills, etc.?  They certainly don't indicate Zelenitsky et al.'s conclusion that adults had long wing feathers while juveniles didn't, as the length is unknown in both (>15 mm in a juvenile; >6.5 mm in an adult).  Sure they're three times wider than the filaments of the juveniles, but the adult is 2.4 times as large, bases of feathers are thicker than other parts, and secondary quills are plausible without vanes.

So while the new paper is interesting in showing Dromiceiomimus had feathers and possibly thick quills/shafts on the lower arm in D. samueli, I don't think it successfully shows the genus had long secondaries, vaned secondaries, or changed its plumage ontogenetically.  This makes their oft-copied figure 4A/B of the little unwinged individual and adult with ostrich-like wings possibly misleading. 

References- Currie and Chen, 2001. Anatomy of Sinosauropteryx prima from Liaoning, northeastern China. Canadian Journal of Earth Science. 38, 1705-1727.

Makovicky, Kobayashi and Currie, 2004. Ornithomimosauria. In Weishampel, Dodson and Osmolska (eds). The Dinosauria Second Edition. University of California Press. 861 pp.

Martyniuk, 2012 online. 

McFeeters, 2012 online.

Zelenitsky, Therrien, Erickson, DeBuhr, Kobayashi, Eberth and Hadfield, 2012. Feathered non-avian dinosaurs from North America provide insight into wing origins. Science. 338(6106), 510-514.

Monday, October 15, 2012

Microraptor hanqingi and Senter's (2011) split

Hi readers.  I've been away due to the fun of unemployment and moving (which might shut down the Database for a short while), but here's a little something I did while deciding whether to make Microraptor/Cryptovolans multiple OTUs.  Though many species and specimens have been described, Senter et al. (2004) synonymized them, which was followed by Turner et al. (2012).  However, Senter (2011) recently reversed his opinion and has M. zhaoianus, the CAGS specimens (20-7-004 and 20-8-001 described by Hwang et al., 2002), M. gui and Cryptovolans all separate.  In addition, Gong et al. (2012) described a new species M. hanqingi which has yet to receive taxonomic commentary.

Senter states "recent personal examination of the holotype of M. zhaoianus (IVPP V 12330), a specimen referred to M. gui (IVPP V 13320), and a cast of the holotype of M. gui (IVPP V 13352) has revealed that the previous assertion that these taxa were synonymous (Senter et al. 2004) is incorrect.  These four OTUs differ from each other in too many character states to be considered synonymous (Table S1)."  Table S1 shows eight characters-

1. Lateral groove on maxillary and dentary teeth. Supposedly present in Cryptovolans but absent in the others.  Yet not only is this polymorphic in most theropods (check Currie et al.'s 1990 illustrations of Saurornitholestes, Dromaeosaurus, Richardoestesia and Troodon for instance), it is present in some teeth of the M. zhaoianus holotype, unpreserved in the M. gui holotype and CAGS 20-8-001, and is indeterminable in photos of CAGS 20-7-004's ~15 preserved teeth.

2. Serrations on maxillary and dentary teeth.  Supposedly present in Cryptovolans and the CAGS specimens (only preserved in 20-7-004) but absent in M. zhaoianus and M. gui.  This is actually present in M. zhaoianus' holotype (Xu et al., 2000) and referred specimen IVPP V13475 (Xu, 2002).  It is unpreserved in the M. gui holotype, though absent in referred specimen IVPP V13320.

3. Fusion of sternal plates.  Present in M. gui's holotype and both Cryptovolans specimens but absent in CAGS 20-8-001.  Unknown in CAGS 20-7-004 and unreported in the referred M. gui.

4. Manual phalanx I-1 significantly longer than metacarpal II.  Supposedly present in the CAGS specimens but not M. gui or Cryptovolans.  However, Hwang et al. state in CAGS 20-7-004 "Phalanx I-1 is very long; it extends almost as far as the distal articulations of metacarpals II and III."  It can be seen in figure 23 to be shorter than the incomplete metacarpal II, while the condition in CAGS 20-8-001 is unpreserved.

5. Manual phalanges II-1 and II-2 together longer than metacarpal II + carpus.  Supposedly present in Cryptovolans but absent in M. gui and the CAGS specimens.  Yet ignoring the carpus, Cryptovolans' paratype has a ratio of 91% and M. gui's holotype a ratio of 102%.  CAGS 20-7-004 has an incomplete metacarpal II so can only be said to have at least 95% and CAGS 20-8-001's ratio is unpreserved. 

6. Arching of first manual ungual.  Supposedly present in M. gui and the CAGS specimens but absent in Cryptovolans.  Yet this is impossible to determine in the Cryptovolans paratype due to the proximal portion of ungual being broken (the holotype's hands are impossible to evaluate in Czerkas et al.'s photos).

7. Strong lip on first manual ungual.  Supposedly present in M. gui and the CAGS specimens but absent in Cryptovolans.  Not only is this variable in other taxa (e.g. Archaeopteryx, Sapeornis, Confuciusornis), it is an illusion in Cryptovolans' paratype caused by the aforementioned broken proximal portion of that ungual.

8. Tibia bowed.  Supposedly present in Cryptovolans but absent in the others. Yet M. gui was originally diagnosed partially on having tibiotarsal bowing, though it is more developed in IVPP V13320 than the holotype.  M. zhaoianus' holotype is also stated to have a bowed tibia by Turner et al., which is true.  That of CAGS 20-8-001 is also bowed (Hwang et al., 2002 figure 28A).  The truth is most Microraptor specimens have some degree of bowing, and Turner et al. proposed the differences are largely taphonomic.

Note a recurring issue is that certain characters are only determinable in one specimen of a species, and there have been no valid characters proposed to e.g. group IVPP V13320 with the M. gui holotype, or group the CAGS specimens together to the exclusion of M. zhaoianus.  The only valid differences noted above are that referred M. gui IVPP V13320 lacks tooth serrations and CAGS 20-8-001 lacks sternal fusion.

Moving on to Microraptor hanqingi, Alexander et al. (2010) used LVH 0026 for their biomechanical work, considering it "probably a different species of Microraptor but is morphologically closely similar to M. gui, including the presence of flight-adapted feathers on the tarsometatarsus" (though remember M. zhaoianus' holotype merely doesn't preserve metatarsal remiges, and may have had them in life). Gong et al. (2012) later described it as the new species Microraptor hanqingi. They diagnosed it based on several characters-

"Largest known species of Microraptor; sternals not fused; robust pubis with squared distal end (more pointed in M. gui) and not as bent backwards as in M. gui; pubic boot tapering posteriorly; ischia with posterior edge straight and ventral edge concave while in M. gui they are sinuous and flat; differs from M. zhaoianus and resembles M. gui in having a proportionally short manual digit I; metatarsals II and IV about the same length (mt II slightly shorter); differs from M. gui in having fewer caudal vertebrae (23)."

The larger size (11-14% larger than M. gui or Cryptovolans) could be ontogenetic or individual variation. Unfused sternals are also present in CAGS 20-80-001 and NGMC 00-12-A, and vary within other species like "Ingenia" yanshini. The pubis is not more robust than in M. gui's holotype, and it cannot be determined if the pubic boot is more squared since most of M. gui's is hidden behind the tibiotarsus. The pubis is indeed bent ~10 degrees less than in the M. gui holotype, but some other specimens lack much bending at all (e.g. IVPP V13475). Contra Gong et al., the ischia of M. hanqingi are posteriorly sinuous and those of M. gui are anteriorly concave, so do not differ in these respects. Additionally, the short manual digit I (83% of mcII) was said to be like M. gui (87%) but unlike M. zhaoianus, and this is true for a referred M. zhaoianus IVPP V13475 (98%) and the Cryptovolans paratype (110%), while NGMC 00-12-A (85%) is also like M. hanqingi and M. gui.  The supposedly longer metatarsal II compared to IV was also supposed to be more similar to M. gui than to M. zhaoianus, but the latter's holotype has a ratio of 98% compared to 91% in M. gui's holotype. Other Microraptor specimens fall between those two ratios, and 7% differences or more are known in other coelurosaur species (e.g. Archaeopteryx lithographica and Dromiceiomimus brevitertius). Finally, its 23 caudal vertebrae are less than M. gui's ~26, but M. zhaoianus' holotype has 24, CAGS 20-7-004 and 20-8-001 have 26 and Cryptovolans has 28-30.  Other maniraptoriform species (e.g. Gallimimus bullatus and Shenzhouraptor sinensis) are known to have individual variation in caudal count within 3-4 vertebrae.

So besides the possibly ontogenetic large size, sternal fusion, pubic curvature, manual digit I length, metatarsal II/IV ratio and caudal number all vary compared to some other specimens.  Variation from other specimens is never significant (e.g. 10 degrees of pubic angle; 2% manual digit I length; one caudal centrum), often falls within the range of other specimens, and most are known to vary within other coelurosaur species.

In conclusion, the problem with splitting the Microraptor-Cryptovolans clade into species is not the absence of variation, as most of the described specimens have some proportions or characters which differ from all or most other specimens (see The Theropod Database's Microraptor entry for more examples, though it's not yet updated with this post's details). Instead, the problem is that these differences don't vary in a systematic, congruent way. While we might argue the Cryptovolans specimens and M. gui holotype should be grouped together to the exclusion of M. hanqingi, CAGS 20-80-001 and NGMC 00-12-A based on their fused sterna, we could equally as well argue the Cryptovolans specimens, M. hanqingi and NGMC 00-12-A should be grouped together to the exclusion of the M. gui holotype and CAGS 20-7-004 based on their long manual phalanx III-1. Or that Cryptovolans, M. hanqingi and CAGS 20-7-004 should be grouped together to the exclusion of the M. gui holotype and NGMC 00-12-A based on their long manual phalanx I-1. There is no obvious answer. One possibility would be to diagnose a separate species for each specimen, and indeed the variation may be due to several different species living in the Jiufotang fauna.  The more conservative approach taken by Senter et al. and Turner et al. is followed here, where differences are ascribed to individual and ontogenetic variation.

References- Xu, Zhou and Wang, 2000. The smallest known non-avian theropod dinosaur. Nature, 408, 705-708.
Czerkas, Zhang, Li and Li, 2002. Flying dromaeosaurs. in Czerkas (ed.). Feathered Dinosaurs and the Origin of Flight. 97-126.
Hwang, Norell, Qiang and Keqin, 2002. New specimens of Microraptor zhaoianus (Theropoda: Dromaeosauridae) from northeastern China. American Museum Novitates. 3381, 1-44.
Xu, 2002. Deinonychosaurian fossils from the Jehol Group of Western Liaoning and the coelurosaurian evolution. PhD Thesis. Chinese Academy of Sciences. 325 pp.
Senter, Barsbold, Britt and Burnham, 2004. Systematics and evolution of Dromaeosauridae. Bulletin of Gunma Museum of Natural History 8: 1-20.
Alexander, Gong, Martin, Burnham and Falk, 2010. Model tests of gliding with different hindwing configurations in the four-winged dromaeosaurid Microraptor gui. Proceedings of the National Academy of Sciences. DOI 10.1073/pnas.0911852107
Senter, 2011. Using creation science to demonstrate evolution 2: Morphological continuity within Dinosauria. Journal of Evolutionary Biology. 24, 2197-2216.
Gong, Martin, Burnham, Falk and Hou, 2012. A new species of Microraptor from the Jehol Biota of northeastern China. Palaeoworld. 21(2), 81-91.
Turner, Makovicky and Norell, 2012. A review of dromaeosaurid systematics and paravian phylogeny. Bulletin of the American Museum of Natural History. 371, 1-206.