[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Defending Photoshop



Could you please send me higher-resolution versions of http://www.pterosaurinfo.com/longi_juve_insitu.html and http://www.pterosaurinfo.com/longisquama_frills.html? It would be very interesting to trace. This address accepts up to 10 MB at once.

> > With Longisquama you're dealing with a specimen that is covered
> > in _a lot_ of skin.
>
> Is there anything else than its position (which makes sense) and its
> shape (which makes at least some sense) that tells you that it _is_
> skin, rather than (say) a couple of unconnected discolorations?

Okay. Sharov and others have noted dewlap scales. I've seen
brachiopatagia. You've all seen the plumes. Those are all from the
undisturbed upper torso, forelimb and head areas. Not completely
undisturbed, though, because the upper half has rotted and drifted
enough to dislodge both pectoral girdles and the complete sternal
complex, along with some ventral rib tips.

So plumes and some ?other scales are preserved. But soft scaleless skin? The preservation of keratin seems to depend on its amount -- in places like Ukhaa Tolgod claws get preserved, feathers practically never do, and skin not at all.


If you've seen the PT article
or pterosaurinfo.com, which has the same image, as I recall,

This one again? http://www.pterosaurinfo.com/longisquama_frills.html

you'll note that the lower belly has suffered the most dislocation,
perhaps because of belly gases? But also part of this area is covered by the
brachiopatagium. The legs have both rotated dorsally so that the toes
are near the former spine.

But each of these sentences _preassumes_ the existence of "ephemeral bones"! That's what we're trying to _test_!


As you'll remember from my cladogram
Longisquama comes between Sharovipteryx and pterosaurs, thus it no doubt
also had uropatagia.

If it really lies there, yes. Your data matrix includes "information" from your tracings, however, as well as lots of correlated characters and several that probably have no phylogenetic signal whatsoever. (Remember, instead of taking my word for it, you can test this latter idea: http://mesquiteproject.org) It also contains series of continuous characters that you cut into arbitrary discrete characters _and then did not order_.


I haven't been able to trace the uropatagia in
Longsiquama. But the uropatagia that must be there may also be a culprit
in clouding things up. Just a logical conclusion. No real evidence.

That's what I feared to hear...

Other than that feeble explanation I can't tell you
why some bones are crystal clear and others are ephemeral.

Well, I think you better find out. Short of sulfuric acid having been poured over half of the specimen (the clean skeleton, that is, because otherwise we'd have coal all over the place!), I can't think of any quirk of preservation that could produce "ephemeral bones".


But let's have another look at http://www.pterosaurinfo.com/longisquama_bones.html. Where you have found the middle caudals I can see something. It may indeed be a short series of vertebrae. But the supposed proximal part of the tail is a broad white spot! That spot can't possibly be a series of vertebrae. Methinks you "know" what to look for (you've found the middle tail, so there must be a proximal tail somewhere) and where to look for it (obviously between the middle tail and the front half of the skeleton, because, as your experience has -- circularly, I think -- shown, every fossil preserved on a slab is either articulated or closely associated) and _therefore_ "find" it however vaguely it looks like what you were looking for (a spot of any color with a curve that makes it possible to see it as part of a continuous line from the dorsal to the supposed mid-caudal vertebrae).

> Do you have any speculation on why the bones of the same articulated
> skeleton are preserved in such different way -- some are smooth, hard,
> and have a different color than the matrix, while others look exactly > like
> the matrix and are only discernible by their supposed shapes?


No. I just accept it for what it is. If what I saw was imaginary,
why wouldn't I draw it in some sort of preconceived fashion following
the imaginations of Sharov, Bakker and Hauboldt & Buffetaut who all drew
pretty much the same critter?

This is, I think, easy to answer. You do constrain your imagination -- slightly -- by using real bumps, depressions and discolorations* in the matrix** which you then try to connect. I postulate that these connections don't exist (as they don't in the "wing finger claw" of MPUM 6009 http://www.unet.univie.ac.at/~a0000265/mpum_6009.htm or in the "skull" of the "neonate *Cosesaurus*" http://www.pterosaurinfo.com/cosesaurus_neonate.html). I postulate that you make them wherever you consider them plausible.


The fact that you see bones, frills and patagia rather than geometric shapes, human faces or human genitals does not mean that you aren't treating the matrix as a Rorschach test. It just means that you rather expect bones, frills and patagia than other features -- which is of course logical on a slab that already contains a known fossil. You see something and -- probably subconsciously -- think "what kind of fossil feature could this be", without even wondering about whether those features are fossils in the first place. (I think one of the tails on the juvie *Longisquama* page, a very long and straight one, is a crystal-filled crack in the matrix, for example. You seem to assume that there is nothing but fossils in any fossiliferous slab.)

* As real as the ink on a Rorschach test.
** Or on the BONE of *Noripterus*!

Why didn't I draw the legs drifting off
the bottom of the rock? No for some reason what I found was symmetrical
even though not preserved in any obvious symmetrical fashion. It nested
neatly within a phylogenetic framework of established taxa. And it
spelled out in incredible detail how paedomorphosis made pterosaurs out
of baby longisquamids. And it was all overlooked data. You can't ask for
a better story,

That's it. I am _not_ asking for a story. "Si non è vero è ben' trovato" is not science. A scenario, no matter how detailed and/or plausible, can _never_ support a phylogeny. To the contrary, a phylogeny can falsify a scenario. This is why you can enter characters but not scenarios in PAUP* (except if you code in too subjective ways).


A Rorsharch test starts with a piece of string and ink.
Somewhere on the Longisquama slab there were overlooked and overlapping
body parts.

That's what we're trying to test. It can therefore not be your starting assumption.


Unwin and Benton had a good look at it. I can only wonder
why the didn't put pencil to paper, rather than just type out their
confirmations and questions of Sharov's original observations. They had
it in their hands and they let it slip through their fingers. Almost
like an Indiana Jones story, they did not recognize the Holy Grail when
they were looking at it.

That's still not evidence that anything is there.

> I hope to have demonstrated that the method can lead to dramatically
> erroneous results... it would be nice if someone would criticize my
> tracings. Nobody has done that so far.

Send a nice 4  to 10 Mb file to dpeters@kupperparker.com where the
modem is faster.

Why should I send you all those files? And why should I combine them into one? Why don't you just go to http://dml.cmnh.org/2005Aug/msg00123.html, follow the links and look at the pictures? I have not shrunk the files, they are there in full size and quality ("full size and quality" being what's left after you compressed them for your website, from where all without exception are taken).


> > Some of those little soft bone pterosaurs I was seeing turned
> > out to be [possibly] pre-shell embryos, now possible in the
> > pterosaurs as lizards scenario. It could be that they only wore
> > the shell for hours or days.
>
> You mean they were in utero, in eggs that would have been shelled
> and laid later?

I think so.

Then why isn't the "bone" preserved the way _cartilage_ normally is, namely not at all?


That seems to fit the rest of the story.

Again, this is not evidence.

> A good just-so story, a coherent picture, is not evidence. It is the
> result of evidence -- or not.

Alls well that ends well.

A good just-so story is not an end.

I've seen just-so stories that must have blown away their original readers. For example there's a book from the 1950s that argues that the Earth's axis has drastically tilted several times, explaining just about all of geology and paleontology. Totally coherent. Good just-so story. Just nonsense. Most of its basic facts have turned out to be wrong (for example it says the Ural must be Cretaceous rather than, as everyone assumed, Tertiary in age... nonsense, it's Permian... or it says that the *Glossopteris* flora was tropical, showing that today's southern hemisphere was on the equator... nonsense, it's cool-temperate, deciduous, like what the author had in front of his window... or it preassumes that every single frozen mammoth is the same age and comes from the end of the last ice age...), and the rest is better and even more coherently explained by plate tectonics, a fact whose discovery was about 20 years in the future; plus it ignores the fact (which nobody had discovered at that time, but every astronomer could have) that the Moon's mass prevents the Earth from doing any such chaotic tumbles (...while Mars can do that and apparently has done it a few times).

Still it answers more questions than Wild's colugosaurus.

That's bad for Wild's colugosaurus -- but it's not good or bad or anything for any of your interpretations. The story mentioned above answers a whole lot more questions than the idea of a completely static Earth.


> Don't knock the interpreter. Knock the technique. If you can't knock the
> technique, _then_ knock the interpretation.

You're like knocking pen and pencil when they skip details. Not
right.

Pen and pencil, like Photoshop, are incapable of distinguishing stone from bone. So sure I'll try to knock pen and pencil out of your hands. :-|


> I agree it should be revised by someone -- by someone who has actually
> studied the fossil under a stereo microscope. Even then I don't think
> the two-dimensional skull will provide much information.

Actually your wish is coming true, somewhere else.

Great!

> How do they look like ants or wasps? To me they look like stylized
> beetles -- or like ostracod shells, which must be expected anyway in a
> freshwater lake like that, and are hard enough to fossilize easily. No
> legs or wings are visible in your photos (...except if the left and
> right "ostracod shells" turn out to be beetle elytrae).

There are beetles galore outside the fossils. Two little
ant/wasp/bees/ whatever inside. Turned to stone.
The legs are harder to see.

Well, you didn't draw any legs in http://www.pterosaurinfo.com/sharovipteryx_skull.html, and as you can guess I can't see any in the photo either. The preservation (of the one that is there -- the one besides the neck) is IIRC quite similar to that of ostracods in the Yixian Fm.