Please DON'T ignore any taxa - A rebuttal to O'Connor and Dyke

O'Connor and Dyke (2010) recently published a paper defending the validity of Cathayornis in relation to Sinornis, and thus contradicting Sereno et al. (2002).  The paper looks good and I feel a bit the fool for taking Sereno's statements at face value instead of examining the data myself.  They also examine the referred species (C. caudatus, C. aberransis and C. chabuensis) and come to largely similar conclusions as I did.  At the end of the paper, the authors make a number of observations and recommendations regarding taxonomic and descriptive practices for Mesozoic birds.  I agree with most of these.  We do need better diagnoses and more informative descriptions and illustrations, and I concur that Hou's 1997 book and the Dapingfangornis description are two of the worst offenders.  Yet I don't think this is as much as problem with the clade as it is with particular authors.  Chiappe's descriptions and diagnoses are generally excellent for instance, but Ji and Ji's papers leave something to be desired, even when they're on non-enantiornithines (e.g. Sinotyrannus, Sinosauropteryx, Protarchaeopteryx). 

Another issue the authors bring up are privately owned specimens, using Dalingheornis and Naish et al.'s (2007) Crato bird as examples. It's odd that Dalingheornis is used twice as an example of this, when it is said to be housed in Capital Normal University, AND O'Connor herself was a coauthor.  Maybe it has since been moved to a private collection, but even if so, authors can't be expected to predict future events.  O'Connor and Dyke state privately owned specimens are "unavailable to the scientific community, and thus rendering any interesting data they may have to contribute unverifiable and useless."  I'm no fan of private ownership of important fossil specimens, but unfortunately the world doesn't always give us ideal circumstances.  Note that publically owned specimens (e.g. Quetzalcoatlus) can be just as unavailable as any in a private collection and that privately owned specimens aren't always inaccessable.  But even ignoring that, information is information and having it is always better than not.  Even assuming a specimen is lost, casts are useful and even superior in some ways (as the authors note for Sinornis and Cathayornis), photos are still a form of information, and illustrations and descriptions are as well.  Not as useful as the real thing of course, but life is seldom ideal.

The authors are most concerned with the erection of taxa based on fragmentary types, but I don't think it is as big of a problem as they say.  O'Connor and Dyke point out "over a third of all known enantiornithines are named from bone fragments", which is true.  But looking at their table shows that these are almost exclusively two kinds of taxa.  Some are the earliest named enantiornithines (e.g. Avisaurus, Nanantius), when we simply didn't have complete specimens.  But most are from Soviet Asia (Explorornis, Lenesornis, Catenoleimus, etc.), and as anyone who's read Nessov's work knows, erecting taxa on fragments from the Bissekty Formation is not a problem unique to enantiornithines.  Moreover, some of the taxa the authors lament (e.g. Lectavis) are quite diagnostic, regardless of how fragmentary they are.

These are all fairly typical statements, but the part I'm most concerned about is the author's suggestion for dealing with these issues.  In addition to the urge for reviewers to do a better job (which I support), O'Connor and Dyke make the following horrific recommendation-

"When taxa are based on private material, or are for other reasons invalid, the scientific community should unify in excluding these ‘taxa’ when discussing the clade they are purported to belong to. This will hopefully discourage the continued practice of the erection of such taxa (e.g., Dalingheornis, Zhang et al., 2006)."

No, no, a thousand times no!  Place the word in scare quotes or not, any species whose description passes the rules of the ICZN is a valid taxon.  Maybe it's undiagnostic, maybe the description is horrid, but the name is out there based on an animal that existed, and thus constitutes real data.  A name cannot be un-done, and named taxa almost always carry useful information no matter how fragmentary.  Any discussion of enantiornithine locomotory variation that excludes Yungavolucris would be incomplete, as would any discussion of enantiornithine sacral morphology that excludes the Soviet synsacrum-based taxa.  And discussing these specimens without using their names would be confusing and petty.  More basically, science is not politics. We should always use all the information at our disposal, and not exclude some for purposes of affecting policy or changing behavior.  While scientists should of course try to create and enforce better policies in their field, the science itself should be above this.

References- Sereno, Rao and Li, 2002. Sinornis santensis (Aves: Enantiornithes) from the Early Cretaceous of Northeastern China. in Chiappe and Witmer (eds.). Mesozoic Birds - Above the Heads of Dinosaurs. University of California Press. 184-208.

Zhang, Hou, Hasegawa, O'Connor, Martin and Chiappe, 2006. The first Mesozoic heterodactyl bird from China. Acta Geologica Sinica. 80(5), 631-635.

Naish, Martill and Merrick, 2007. Birds of the Crato Formation. in Martill, Bechly and Loveridge (eds). The Crato Fossil Beds of Brazil: Window Into an Ancient World. Cambridge University Press. 525-533.

O'Connor and Dyke, 2010. A reassessment of Sinornis santensis and Cathayornis yandica (Aves: Enantiornithes). Records of the Australian Museum. 62, 7-20.