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

Re: Effigia [was Defending Photoshop]

----- Original Message ----- From: "David Peters" <davidrpeters@earthlink.net>
Sent: Sunday, February 05, 2006 2:19 PM

Often WYSIWYG in fossils, 'repairs' (like making sure heavy
sauropod tails contact the groud) to the contrary.

That's true, but it is also common that diagenesis produces features that are completely misleading, such as convincing-looking symmetric distortion or (a rarer case) uniform expansion of *Plateosaurus* vertebrae.

The key thought here is that Ticinosuchus led to at least two branches now, the
bipeds and the armored ones, much as early dinosaurs led to theropods and
scuttellosaurs.

I don't quite get the first part of this sentence. Surely *T.* has a couple autapomorphies which preclude it from (parsimoniously) being an actual direct ancestor of anything known?

So, it can happen. Try not to apply too many aeotosaur characters
to something that is not an aetosaur. Some characters, yes, but pick the correct
characters, before and after the split.

We may be talking past each other here, but... in general... don't pick. Total evidence approach.

Same comments as above with the addition of: can you find ANY other taxa with a
dentary a brief or nearly so?

Oviraptorids come halfway close. Since *E.* is obviously not closely related to oviraptorids, the short dentaries of both have to be regarded as an autapomorphy of each of these two clades.

And with a mandibular fenestra a extensive or nearly so.

I'm not aware of anything remotely similar. If there really is nothing (I don't know a lot about crurotarsans), then this character state is an autapomorphy of *Effigia*, thus phylogenetically uninformative, and should thus be dropped from the matrix.

<To move Effigia closer to aetosaurs requires one more step. To move it to the
base of the Ticinosuchus + aeotosaurs requires 5 more steps. To move it close
to Lotosaurus requires 13 more steps. The clade Ticiniosuchus + aetosaurs is
the sister to the Archosauria,which includes Turfanosuchus + (Dinosauria +
Crocodylomorpha).>

This is only because you've reinterpreted *Lotosaurus* as a dinosaur.

No, PAUP nested it there based on discrete data points. And it's not a typical
dinosaur, Jaime. It's one of the cuzzins no one talks about.

I haven't seen your matrix, but this sounds _very_ much like you simply have too few dinosaurian autapomorphies in your matrix. I can't see how something with a crocodile-normal ankle and a rauisuchian-type hip can be a dinosaur.

But let's do it the scientific way. Which apomorphies does *Lotosaurus* share with the rest of Ornithischia?

Given
my comments on that when this issue came up, one should be very circumspect
when non-dinosaurian vertebrae, non-dinosaurian scapulae, non-dinosaurian limb
bones, all point to a non-dinosaurian but crurotarsan *Lotosaurus*. But go
ahead and publish findings on detailed photos of the original material, if it
passes peer-review. I suggest having Clark and Wu work as reviewers.

You need to do the phylogenetic analysis before saying such things, Jaime. Fight
facts with facts, not words.

Codings based on bad photos are less likely to be facts than codings based on good photos... codings based on tracings of descreened bad photos, and so on... and of course there can be typos and suchlike. I can generalize: Data matrices are collections of hypotheses which need to be defended one by one; they aren't facts that can be accepted at face value. The same goes for ordering assumptions and so on.

re: Clark and Wu, It's out of my hands. But good suggestion.

Many journals let you suggest peer-reviewers and/or people whom you would not like to peer-review your work.

And you forgot to say: It's still a Newtonian universe. 

Non sequitur.

And besides... several recent quantum-based theories predict that there should be small but measurable deviations from Newton's law of gravitation at millimeter to centimeter scales; experiments have since shown that these deviations don't exist. As a Nature commentary put it, "Newton rules (for now)".