Re: back to pteros, [yada, yada]

["Jaime A. Headden" <qilongia@yahoo.com>]

David Peters (davidrpeters@earthlink.net) wrote:

<and it is an ornithocheirid only in Wang and Zhou's eyes. (The same folks who
said Zhejiangopterus was a nyctosaur and that Dendrorynchoides was closest to
Rhamphorhynchus). No one else has corroborated or tested either egg
observation. I merely indicated that there was an alternative explanation for
what little that could be seen. Isn't that good science? Exposing possible
errors?  I appreciate it when errors are exposed in my work.>

  Several people have questioned the methods used to recover the information
used to even identify the identity of the embryo as an anurognathid. The
original authors, even, didn't identify it very hard, given it's an embryo
without clear identifying features, but did imply it might be an
ornithocheiroid embryo, of which taxa only *Haopterus* they had recovered at
the time. Currently, there are 2, maybe even 3 ornithocheiroids now known from
Liaoning. Using *Haopterus* at this point is a strawman. As for identifying it
as an anuorgnathid -- my argument about this is the methods used to support the
ID, rather than the ID itself. In Peters' words:

  "I employed the usual techniques and started tracing the contours of the
   bones. I found teeth right away, but not where Wang and Zhou said they would
   be. And I did not find teeth where they reported two to be. They mistook a
   wing phalanx for a tibia and a tibia for a wing phalanx -- perfectly
   understandable considering the scattered nature of the specimen. They
thought
   they found the mandibles, but these turned out to be two fairly straight
   V-shaped portions of the skull bordering the large narial opening. Wang and
   Zhou overlooked the fingers, the foot, the skull including the palate and
   occiput, the pelvis, the tail, the sternal complex and the real mandibles
   loaded with teeth. The wings were articulated and so were the legs. The top
   of the skull was separated from the jaw line and the occiput was still
   connected to the cervical vertebrae. Altogether this specimen was more
   articulated than the Chinese authors figured it.

  "When I finished my tracing and reconstruction (both shown here) I concluded
   that Wang and Zhou had discovered a new anurognathid pterosaur, one that was
   about the same size as other adult anurognathids. It had a bigger, squarer
   head than usual and longer front teeth. It also had a decidedly shorter
   forearm and that shortened the wing span."

  These conclusions of an anuorgnathid identity, and thus its comparability to
other anurognathids used to support it's generic monicker, are based on a
contested technique which HAS been questioned, tested, and objected to on
subjective grounds of artifacts of an ash-laden slab.

<BTW: The tone expressed by the above is not conducive to sharing and
building.>

  No, I do agree that my tone is harshly critical in this case. Requests for
blind tests and objective reasoning in the application of the technique have
been largely put aside based on my knowledge of the situation. If there have
been such objective tests, first hand observations, better photos, blind
studies, etc., to support the conclusions, then I will be more than happy to
shift back into my original supporive stance for the technique.

<Every observational technique is 'highly' subjective. Every jury and judge is
'highly' subjective. The Photoshop technique simply offers a way to send
extremely precise tracings on top of layers of photography to others for
confirmation.>

  And the drawing of lines on a slab perceiving that the completely prepared
matrix in a chalk bedding plane described "hidden" bones? This was the case for
another published objection to this technique on the basis of *Nyctosaurus* sp.
Since the slab itself was prepared down to recover bones within the chalk, it
is impossible for impressions of bones or tissues to be preserved out of the
bedding plane recovered. 

<The beauty of this technique is that you can add a layer to pinpoint errors
and you never have to _believe_ a line drawing again! I once sent an image of a
pterosaur rostrum to a colleague and showed him where to find the naris 'that
shouldn't be there'. He responded by saying that all he saw was 'a hole in the
rostrum with matrix coming through'. I said, that's the naris! He wouldn't
believe his own eyes.>

  Yes, if I find a crack in the bone and the bone is flaked out (like what
happens when bone is either eroded or removed during splitting of a slab) it
will look like a "naris" too. The problem with this theory is one of
identification and corroboration independantly. Which hasn't happened. Indeed,
the positions of the nasal bone itself, and a developmental series in
pterosaurs offers evidence for a particular fusion of the external naris with
the antorbital fenestra, the "classic" model, which, while "classic" and
therefore "wrong", objections have to go through as much scientific rigor to
overturn it. Using cracks and holes that are present all over the skulls of
most pterosaurs preserved in slabs to find these special ones and then to
identify them as external nares is _a priori_ observation. For example, what
stops these "holes" from being foramina or subnarial fenestrae? Indeed, one
theory about beaked animals is that they should have many foramina on the bones
supporting the rhamphothecal sheath for the purpose of ennvervation. Instead,
they are "the naris 'that shouldn't be there'."

  In response to Peters, I can corroborate seeing these features, but I do
think they are artifacts. Not because I "believe" in the confluent nares/aof
theory, but because I observe the irregularity of the features, the tiny size
of them, and the need for a nasopharyngeal passage of greater opening than tiny
slits in a skull that large which, for an active "warmblooded" animal, will
need to be VERY large. Such artifacts are also apparent elsewhere in the skull,
may be explained as foramina or subnarial fenestrae, or simply as fractures
with removed bone which occurs during splitting of a slab into part and
counterpart. One can look at many examples of this in Solnhofen pterosaurs when
comparing part and counterpart, and this occurs in *Archaeopteryx* as well,
where in the München specimen (*A. bavarica*), the jaws are on one slab, and
the braincase on the other. 

  And such impressionistic examinations can happen to the best of us. We can
all look at *Longisquama*, and wonder: Where's the rest of it? There's this
large slab, and the bone just stops. Not even any impressions. So we want to
see ... is there anything else? Well, if you peer really close, light the thing
up or dim it down JUST SO ... you can see something that looks like a tail.
Hmmm ... what if it's a tail? So we look to see if there's a leg ... and see
these little irregularities in an imprefectly split slab that just MIGHT be a
leg ... ad vertebrae ... and we think: What if the whole animal is there? See,
there's a tail, a leg, some verts, that's a pelvis-looking thingy....

  It seems like one's looking objectively, thinking scientifically, but this is
just a self-fulfilling prophecy: the data will fit what we want it to because
we are making it as we go. As Freud said, sometimes a cigar is just a cigar;
or, in other words, some times a crack in the rostrum of a pterosaur is just a
crack in the rostrum of a pterosaur. 

<This same technique resulted in the finding of a skull in a specimen of
juvenile *Pterodaustro* which, according to the people who have studied it
first hand, doesn't exist.>

  See above. I looked for this, curious if after I wrote what I did I had been
confused and Dave had another specimen in mind. I opened up Cordniu and
Chiapppe and looked at the specimen in question. Every single bone is preserved
in nearly articulated glory except for the neck (broken and disjointed, but I
doubt I will find missing vertebrae to fill the gaps [note: Dave has not
suggested he has done this, I am doing this to consider previous examples of
filling empty space with bones]), and the skull is not present; you would think
so well ossified, as is consistent among vertebrates, that the skull would
ossify if it stayed intact. I find the slab on the other side of the skeleton
just as consistent with this finding as the region where Dave finds the skull.

  Psychology and neurobiology teach us that when we try to look at something
where the image is fuzzy or unclear, our eyes will fill in the missing details
and our minds will try to fit something familiar into the gaps. This will give
us a picture in our heads not entirely consistent with the actual object we
perceive. These artifacts of light and mind result in the
"crones's-face/young-lady-turned-away" and "vase/two-faces" style optical
illusions, as well as mirages. They are there, it's just that we are perceiving
them differently depending on what we want to see.

<Some subtleties come through that can't ordinarily been seen. The slightest
impressions left in the rock when the bone is gone. That too, is called a
'fossil. '>

  Such subtleties are the very basis of the contention. A missing flake of bone
in the snout of a pterosaur is not the same as a whole skeleton of
*Longisquama* on the holotype slab as Dave contends. Dave has argued from a
developmental standpoint that the bones of many babies and even "adults" are
present, form impressions, or never ossify, contrary to any developmental model
thus far known. He has not used any independant observations for what would be
an extremely important discovery to support this, but contends the impressions
are, by themselves, sufficient to support his case.

<Did Chris provide an alternative drawing? Don't recall seeing it. It's easy to
dismiss claims with the sweep of a pen. Harder to establish new ones. In the
last  issue of Prehistoric Times I gave Chris credit for finding some errors,
but also exposed some errors of his, all without resorting to name calling or
making disparaging remarks.>

  Chris' data is also largely based on rejecting the hypothesis. So he's
supposed to use this technique to make a drawing to refute the technique? He
has indeed used this technique and has shown how it dos not support the claim
on a variety of reasons, including the problem of resolution of the image used
for tracing:

  http://www.fhsu.edu/biology/cbennett/critique.html

  Chris is also further working on the very taxa he published on which Dave
provided contradictory data for: A complete *Anurognathus* specimen, incomplete
KJ1 and KJ2 skeletons of *Nyctosaurus* sp., *Pterodactylus* growth series and
coincident species identifications, etc.

<How can you expect to discover anything new if you don't stick your neck out
once in awhile? I'm not afraid of making errors (obviously). Every error brings
one closer to the truth. Isn't that what it's all about?>

  I argue using my own data that Osborn was correct to assert that *Oviraptor*
was indeed an egg thief (a man who's attested personality and politics I highly
dislike), and instead reject the published statements of Barsbold and Norell
(both men whom I highly admire for their work and the way in which they go
about their work) that it was not. This is sticking my neck out. I have no
problem objecting to long-held theories. But extraordinary claims require
extraordinary evidence, as has been pointed out before, and to use a contested
technique to support its use is a very dangerous way to support research and
remain objective. Especially given arguments of how much one can interpret a
few pixels on a digitized photograph. This is another reason why Chris Bennett
has argued that to substantiate these claims, it is required to see the
specimens themselves and show them in detail. Photographs will not do, and may
be used to illustrate the point, but are not evidence themselves, especially
given the distance taken, the light level and angle, and the detail of the
bones themselves which may, as Gerald Mayr has found out, require preparation
to see clearly using chemical coats. Sunny Hwang recently described a new
dinosaur as *Huaxiagnathus*, an animal for which the bone is virtually
indistinct from the surrounding matrix, but is identified based on its shapes;
to see the margins, one has to begin to "interpret". The bone and matrix are
the same general colors, and are not distinct; highly detailed color photos of
the material I was provided only prove this point. Differentiation becomes
dangerously subjective and requiring what one thinks it _should_ look like.

<You might be confusing the egg [with skull] with the juvenile [seemingly
without a skull]. Either way, the skulls match and are closing in on adult
proportions. [Something of a confirmation of the technique, actually.]>

  See above. I have not seen detailed photos of the egg specimen, only Cordniu
and Chiappe's specimens and Chiappe et al.s _Nature_ paper on the
*Pterodaustro* eggshell.

<Well, they are all related, but they're not the only diapsid clade with an
orbit larger than the rostrum in adult forms.>

  A feature largely apparent in arboreal animals, as is a triangualr skull for
the purpose of aiding in stereoscopic vision. I tend to think if you put any
animal with a boxy skull in a tree for a few million years, you'll find a
"bird-headed" animal when you come back. Selection follows particularly
advantageous features in certain environments that may have nothing to do with
phylogeny, as those espousing megalancosaur/bird relationships have heard a
dozen times.

<Yes. Now you're grasping at straws. It has been confirmed. All related forms
have the same skull proportions.>

  No, I wrote that before seeing the "skull". Now, I am sure this is not the
case, as there is no skull present based on any typcial investigative means I
am willing to endorse. The photo-technique, which I have used independantly of
all this (see my posts on *Protarchaeopteryx*' skull with Mickey Mortimer's
interpretation) has uses in the broad, but not the narrow, in photos, because
of the problem of resolution and absence of the ability to observed the
material first-hand. This means it's just an interpretation several levels
worse than coding a complex of related features into a suite, because they
"seem to fit together", even when many organisms exhibit the separate features
independantly. 

<References, please. I'm not aware of any but my own paper on the subject that
takes that stand. Chatterjee's recent opus did not consider the 'narrow to the
elbows' model at all, and that, I think, is the latest reference.>

  Splitting hairs about a narrow chord with a regular versus a reverse camber?
Huh.

  Narrow-chord has been proposed in Unwin and Bakhurina, who, although they
linked the cheiropatagium to the ankles, draw the chord toward the arm far more
narrowly than the classic model, which is drawn by Wellnhofer and was first
illustrated for *Sordes* by Sharov himself. Padian, especially in his popular
article in _Natural History_, espoused the narrow-chord model.

  Unwin, D. M. and N. N. Bakhurina, 1994. *Sordes pilosus* and the nature of
the
    pterosaur flight apparatus. _Nature_ 371:62-64.
  Padian, K. 1988. The Flight of Pterosaurs. _Natural History_ 97:58-65.
  
  Both of these have argued a typical anterior camber in the wings, rather than
Dave's model which would restrict the camber to the posterior half of the
chord.

<There's a Catch-22 here. Can't publish cause the technique is bad. Can't argue
the technique cause it ain't published.>

  No, the catch here is that the technique, to be useful, but make valid
predictions. To test this, use other methods that arrive at the same
conclusions, thus verifying the utility of the technique. Publication follows,
and permits a one-stop shop reference for the use and description of the
technique in clear, mathematical logic. Publication here argues for validation
of the technique, meaning one would think it is independantly verifiable. In
this case, so far, Dave has only used the technique to verify his findings.
This means they are not "verified".

  ...

  Now, before my tone becomes more critical, I would like to reiterate:

  I have no problem with the line-drawing photoshop portion of the technique's
use in illustrating photos. I have a problem with using photos to find details
of slabs where the image is pixels wide in a largely homogenous region of
matrix without secondary or primary varification first hand. A trip to the AMNH
with special access to their collection of pterosaurs would be very useful in
this regards. The use of casts even better, since any valid impressions may
carry over and they are easier to get a hold of than flying to Europe where
most of this material is held. There are may "ins" in this technique, but it
hasn't been verified. This makes it very difficult to judge it's veracity when
involving tricks of the eyes and the hypothesis of a complete skeleton
available on the slab. None of this is about Dave himself, and if anyone else
were to argue these things in using it as he has, I would take issue with the
case just as with him. 

  Dave, if my tone sounds unwarrantly critical, it's not that I don't like you;
you are an awesome man and I am glad to have met you. I hope I do not have to
say this about anyone anymore, since as a scientist and a student, I know I
will never learn all I can, but will make an effort of it. Fighting for rigor
in scientific practice when I see a need will always be a driving force in my
social interactions, I fear.

  Cheers,

Jaime A. Headden

"Innocent, unbiased observation is a myth." --- P.B. Medawar (1969)

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com