Bone Wars: Then and Now

Figure 1. Cope and Marsh in their classic poses. These two early paleontologists were combatants in the first “bone wars” and their story is told on a great PBS video. Click image to see the video.

Cope and Marsh
are famous for their “Bone Wars” also known as the “Great Dinosaur Rush” in the 19th century. Here is a great PBS video on the subject. According to Wikipedia: “Each of the two paleontologists used underhanded methods to try to out-compete the other in the field, resorting to bribery, theft, and destruction of bones. Each scientist also attacked the other in scientific publications, seeking to ruin his credibility and have his funding cut off.”

It all began ‘innocently’ enough when Marsh took the skull off the tail of Cope’s Elasmosaurus and placed it on the neck (Fig. 2). Cope did not take this well.

Figure 2. Elasmosaurus as originally configured (Cope 1869) with the skull on the end of the short tail (sort of a mosasaur morphology) and on the end of a very long neck, unheard of until then. Click to enlarge.

Mike Everhart at OceansofKansas.com tells the tale in greater detail. Just a few notes here will suffice. Everhart quotes a 28-year-old Cope (1868) who notes the neck was quite unusual, because of the “the presence of chevron-like bones on the inferior surfaces of the cervical vertebrae.”  Then it gets worse. Cope (1869) created a new order, the “Streptosauria” [reversed lizards] for Elasomosaurus. Cope wrote:“Under this name I have characterized a group of high rank among the Reptiles which is allied to the Sauropterygia. The diagnosis will be as follows. The articular processes of the vertebræ, reversed in their directions; viz., the anterior looking downwards, the posterior upwards … The characters of this order are altogether peculiar.”

See how red flags are ignored? 
Even by the authors themselves? We’re all guilty of this, yours truly no exception. Nowadays the “bone wars” are wars of words and manuscript rejections.

The Bird Wars
between Alan Feduccia (and the late L. Martin and S. Czerkas) versus most dinosaur paleontologists began in 1973 and continues today. Originally Feduccia agreed with the long held model for the origin of birds proposed by Gerhard Heilmann (1926) The Origin of Birds arguing against the works of John Ostrom (1970) with his work and insights on Deinonynchus (and countless others who followed), all of whom linked theropod dinosaurs to birds.  With the growing pressure and dozens of new discoveries, after 2002 Feduccia argued that the discovery of the very bird-like theropod dinosaurs, Microraptor and Caudipteryx, suggested an unrecognized radiation of birds that lost flight and secondarily converged on theropods.

Unfortunately that takes the long way around
to get to the point that basal birds share a large suite of traits with bird-like theropods, which is the basis for their phylogenetic relationship. Feduccia never produced an alternate cladorgram of bird origins. Rather in 1999 he wrote, “Although most fields of science are constantly struggling with which methodologies to use, the field of systematics, and espe- cially paleontology, has adopted phylogenetic systematics (cladistic methodology) to the exclusion of other approaches. Despite a barrage of cautions and criticism, cladistics reigns.” Then concluding, Feduccia (1999) wrote, “Perhaps the greatest form of special pleading will be necessary to explain how flight could have originated from the ground up; our present knowledge indicates that there are two requisites for flight origin: small size and high places.” Today it is clear that birds descended from theropod dinosaurs. Experiments have shown that flightless birds run up trees while flapping their wings. That’s all set in stone.

The Large Reptile Tree War
is my attempt at getting paleontologists and referees to consider adding taxa to their family trees to see how more taxa might change their tree topologies. To that end the large reptile tree now has 501 taxa. The large pterosaur tree has 218 taxa. The therapsid tree has 59. All are completely or nearly completely (=deletion of one taxon makes it completely) resolved. These results are repeatable, and all sister taxa look alike. However, they, at times, break with tradition and so raise hackles and doubts.

In a recent rejection letter to manuscript describing a new origin for amniotes, the referee wrote: “I am extremely incredulous of a phylogenetic analysis examining this many taxa that results in a single most parsimonious tree. Rerunning the author’s nexus file in TNT reveals a single most parsimonious tree. TNT collapses zero-length branches more strictly than PAUP* revealing that there are in fact fewer trees supported by unambiguous synapomorphies than reported by the author. The single polytomy present in the strict consensus (Cerritosaurus,Tropidosuchus,Lagerpeton) is the result of how PAUP collapses (or doesn’t) zero length branches than true conflict in the dataset. I must admit that I am extremely incredulous of a phylogenetic analysis examining this many taxa that results in a single most parsimonious tree. I, in fact, cannot think of another published example. The fact that the author does not address this at all is troubling.”

Say…
isn’t this turning a ‘good’ result into a ‘bad’ result?

As we all know, incredulity is a synonym for disbelief.
Why would a scientist elevate ‘belief’ over ‘evidence’ simply because the evidence does not support the traditional paradigm? Some paradigms, as we know from history, are not correct.

BTW, the manuscript was rejected when the large reptile tree had 389 taxa and 228 characters. Today there are 501 taxa and 228 characters. So the situation is worse… or better… now, depending on your outlook.

That manuscript focused on the origin of the Amniota. There were few comments regarding that topic and the included taxa. The supporting document supported and included the large reptile tree. That became the focus of the referee’s comments. Unfortunately manuscripts can only be of a certain length and the number of problems that have to be adequately addressed runs beyond that length (at least to satisfy this referee). I’ve run across this before. Other referees insisted that I visit all 501 specimens, even though this is not standard practice with large trees. In supertree analysis, authors do not have to see a single specimen.

The Pterosaur War
As you all know, I have been promoting new ideas with regard to pterosaur origins, interrelationships, ontogeny, wing shape, launch tactics, etc. etc. always with evidence to demonstrate the hypotheses. And as you all know, evidence is sometimes not believed.

A referee wrote this on my fenestrasaur manuscript submission, “Peters reports here the presence of the whole posterior half of the specimen in Longisquama, tail, pelvic girdle and hind limb included. If the hind limb and posterior trunk were in “plain sight” as written in the manuscript, how could all previous authors (Peters himself included), have missed or blatantly misinterpreted them?”

It happens. Look at the mistakes made with Vancleavea. Mesosaurus. Cosesaurus either with regard to observation or taxon inclusion.

That referee’s complaint is a common one.
Unfortunately Longisquama has been rarely studied and even more rarely written up. Actually there were very few previous authors who described Longisquama and none have produced the sort of precise tracing that I provided (see it here). Others didn’t see what I saw because they didn’t look hard enough. Bottom line: This referee was not going to let me repair my mistakes in print. He was not going to allow more precision come to this enigma taxon. Perhaps he thought it better to keep such ideas safely tucked away from consideration and debate. Better to keep some things cloaked in mystery.

Having been in paleontology for several decades now, it becomes clear that many observations are borrowed from prior workers (I do this myself when it seems safe to do so) and many phylogenies are sometimes uncritically borrowed. Red flags, like the nesting of pterosaurs with parasuchians, are ignored, by the original authors, the manuscript referees and colleagues. Why? Well, I think it’s because they don’t want to rock the boat, which occasionally needs rocking. I can tell you, it’s easier to get data published if it agree with current paradigms.

How did I miss or blatantly misinterpret Longisquama?
Back then I misinterpreted Longisquama because I was naive, a newbie, someone with little to no experience. And Longisquama is arguably one of the most difficult specimens known. Fifteen years later, I not only see things differently, I know what to look for and I know better what I’m seeing.

And the funny thing is…
even if the hind portion of Longisquama was not preserved, phylogenetic bracketing between Sharovipteryx and MPUM 6009 would still give it most of the same traits (the long torso is an autapomorphy, Fig. 2).

Figure 3. The basal fenestrasaur precursors of pterosaurs, including Cosesaurus, Sharovipteryx, Kyrgyzsaurus, Longisquama and a basal pterosaur.

Pterosaur workers continue to refuse to consider using fenestrasaurs as pterosaur outgroups, preferring instead to use random archosaurs that share few to no pterosaur traits. They certainly don’t use parasuchians, which often nest ancestrally in many cladograms.

Pterosaur workers continue to refuse to consider using tiny pterosaurs (about the size of sparrows and hummingbirds) in phylogenetic analysis, preferring instead to produce cladograms in which transitional taxa don’t share very many traits with precursors and successors.

Pterosaur workers continue to believe that baby to juvenile pterosaurs had traditional ‘cute’ features (short rostrum, large orbit) despite all the evidence to the contrary. I haven’t met one who accepts the fact that pterosaurs developed largely by isometry during ontogeny as discussed here.

Pterosaur workers continue to refuse to consider using multiple specimens of Dorygnathus, Scaphognathus, Pteranodon and other genera that are known from a variety of morphologies. And they reject manuscripts that do.

Why is this so?
I think it’s because, like Cope and Marsh, I metaphorically put the skull on the other end of the skeleton and created an air of embarrassment out there. I discovered that pterosaurs were not archosaurs. That there were four origins for the pterodactyloid grade and each was preceded by a series of phylogenetic miniaturizations. That juvenile pterosaurs developed isometrically. That pterosaur wings had a narrow chord at the elbow. That uropatagia did not extend from leg to leg. These not only don’t go over well. They are avoided and dismissed without a superior hypothesis at every turn.

Sure I’ve made mistakes.
Tens of thousands of them. I discovered all but a few without help. Often the repairs helped cement earlier hypothetical relationships. Sometimes new insights were gained.

Whatever the truth is,
it will all come out sooner or later. It has to — because it’s the truth.

References
Cope ED 1869. Synopsis of the Extinct Batrachia and Reptilia of North America, Part I. Trans. Amer. Phil. Soc. New Series, 14:1-235, 51 figs., 11 pls.
Feduccia A 1973. Dinosaurs as reptiles. Evolution 27 (1): 166–169.
Feduccia A. 1999. 1,2,3 = 2,3,4: Accommodating the cladogram. Proceedings of the National Academy of Science USA. 96:4740-4742.
Ostrom JH 1970. Stratigraphy and paleontology of the Cloverly Formation (Lower Cretaceous) of the Bighorn Basin area, Wyoming and Montana. Bulletin of the Peabody Museum of Natural History 35:1–234.