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

Re: Pneumaticity in Triassic pterosaurs

----- Original Message ----- From: "David Peters" <davidpeters@att.net>
Sent: Saturday, May 16, 2009 3:54 PM

[1] There is no monophyletic Diapsida containing Archos and Lepidos. [2] Did I not send you my tree? [3] Your tree problem may be due to the assumption of monophyly in Diapsids.

(1) You mean the upper temporal fenestra evolved twice? (2) Yes, four years ago, along with the character list and a weird list that was meant to indicate where state 1 of which character occurs. It led to a long discussion. I remember some highlights. -- Not that long ago, you submitted that thing for publication (after having accepted my nomenclatural advice, but apparently not any other). I know that because my thesis supervisor is Michel Laurin, who was one of the referees. The manuscript got rejected (without any influence from me) because... well, as it came out last year, you didn't even know how to make a data matrix for phylogenetic, as opposed to phenetic, analysis, and that showed. Bigtime. (3) Sorry for the misunderstanding. _I_ don't have such a problem yet, because I'm still stuck in the early stages of _making_ the matrix (tetrapod phylogeny, the side project now included in the thesis, is very time-consuming, too); I was summarizing the literature of the last 10 to 15 years. My analysis won't make any assumption on the topology within Amniota, because I'm trying to find out where the turtles sit, basically whether they are diapsids. That wouldn't possible if I built that kind of assumption into the analysis. (I'll use various seymouriamorphs and lepospondyls as the outgroups, so that I can test whether the diadectomorphs are theropsids -- a hypothesis defended by Berman and various coauthors throughout the 1990s, but then dropped when much larger analysis found the diadectomorphs outside.)

Good luck on your thesis.


I can show you the path of most parsimony if you're interested.

Erm, PAUP* will do that... :-)

The key to everything is taxon inclusion. You really have to sample from the whole gamut of the Amniota, plus quite a few outgroup sisters to get the picture.

Yes, of course. That's exactly what I'll do. I only hope this megalomaniacal project can be done in the year and a half I still have officially.

I've been telling you this for years now. The funny thing is, you never even voiced any explicit disagreement; you never wrote "you're wrong, it actually is easy"; you just shut up, and then a few weeks or months later you simply act again as if phylogenetics were like playing Minesweeper. <<

I don't get the emails from the DML daily. I check in every so often. If you don't copy me, as in this failed instance, I may not see your posting for some time later. I will always answer emails, however, as you know from experience. And I am always willing to help, including sending data, files, and images. And to that end, I am more than willing to accept new data that shows my data is in error. Hoping to hear from you soon, David.

Oops. Thanks for the information, I wasn't able to guess that.

However, we've had offlist conversations about this topic, too. Several times in the last few years.

----- Original Message ----- From: "Mike Habib" <habib@jhmi.edu>
Sent: Saturday, May 16, 2009 5:44 PM
Subject: Re: Prolacertiformes and Protorosauria

David Peters wrote:

Once again, if sister taxa don't bear a family resemblance, you have
to doubt the results.

Wait - I'm confused: don't the sister taxa bear "family resemblance" *by definition*? The tree constructs the best estimate of ancestry, using some kind of estimate of family resemblance. For morphological character sets, this usually means using a parsimony algorithm, with is optimizing the distribution of characters and using synapomorphies to define clades. Since you're joining taxa using shared derived character states, they *must* bear resemblance. The fact that those taxa may have some kind of qualitative or intuitive lack of "resemblance" doesn't mean you screwed up.


Probably the idea is that if taxa are too distant phenetically -- that is, by a subjective, intuitive estimate of phenetic distance, not even by any attempt at quantification --, they can't be sister-groups. This overlooks a few things:
- That they are found as sister-groups means they share derived character states with each other and not their next closest relatives;
- It is a feature, not a bug, that phylogenetics cares only about synapomorphies and ignores symplesiomorphies -- resemblance consists of both;
- Evolution happens -- some taxa run off very far, if that's where natural selection drags them;
- For the probably twentieth time, the vertebrate fossil record -- let alone our knowledge of it!!! -- simply isn't complete enough that we could expect to do what the Unnameable Ones ask us to do (to present them a complete series of transitional fossils documenting each and every speciation-or-whatever between two arbitrary endpoints). In fact, that's not limited to vertebrates. It does not even hold for ammonites and conodonts; that's why stratophenetics (using total resemblance and stratigraphy to construct a tree and then interpret that tree as a phylogenetic one) does not work.