Re: Hanson 2006, Mortimer, Baeker response

[Martin Baeker <martin.baeker@tu-bs.de>]

> > Your argument looks like a classical logical fallacy:
> > 
> > "The correct, perfect analysis should have only one MPT.
> > My analysis produces only one MPT.
> > Therefore, my analysis is correct."
> 
> No Martin, here's the deal:
> You wrote: "If knowledge of the fossil record is incomplete."
> There are levels of incompleteness. As in dinosaurs, in pterosaurs we've
> gotten to the point where very few really strange lineages are showing up.
> Most, if not all, dinos and pteros found today can be placed into existing
> slots. Examples of new dinos include Lotosaurus and Silesaurus, but these have
> been slotted into Ornithischia without a retro pubis.

Hmm - *my* copy of "The Dinosauria" has lots of unresolved branches in
the cladograms - don't see that any of those is anywhere near being
perfect. Yes, there are no strange lineages, but if we look at a
cladogram comprising as many species as your probably does, the number
of unresolved points is huge.

> Examples of new pteros
> include Austriadactylus, Bakonydraco and Pterorhynchus, all of which have
> nothing else exactly like them, but can be slotted into an existing cladogram.
> If you have a cladogram that includes all known pterosaurs and you have one
> MPT, plus the cladogram is chronologically correct and there are no untennable
> reversals, then you've achieved your goal. Note that reversals are still
> permitted, but untennable ones are not.
> 
> So, knowledge of the fossil record is incomplete in one sense, but complete in
> another. Sorry. Some of the mystery is gone now.

Again, the argument is circular: To prove that your 1-MPT-cladogram is
correct and that the fossil record is complete enough, you use the
fact that you have a 1-MPT-cladogram.

> > Shouldn't that make you suspicious? It would be the first cladogram I
> > ever heard of that contains nothing like ghost lineages, things
> > disordered wrt time, reversals (shouldn't we *expect* reversals?
> > Evolution is not directed, after all.) etc.  It would mean that our
> > knowledge of pterosaurs is complete and perfect.
> 
> See above. Our knowledge of pterosaurs is complete and perfect in one sense.
> Incomplete and imperfect in another. You can see the layout of this particular
> jigsaw puzzle because the corners and borders are in place. Holes are still
> present, but they merely complete a picture that is already recognizeable.
> Using your logic, no one will ever find the cladogram that echoes Nature.
> You're suspicious of an answer that fulfills all your requirements for a
> correct answer. I hope you see that.

Yes, I *hope* I am - everybody should be suspicious of a theory that
fits all facts perfectly. Not that it is impossible, but too many
times in the past those all-explaining theories were just wrong.
Ending with a 1-MPT cladogram does *not* prove you right, no matter
how nice it looks. (It does not prove you wrong, either.)

> > This is really a dangerous statement to make. Did you think of and try
> > all possible tests?  As Feynman said: "The easiest person to fool is
> > yourself."
> 
> You sound like a referee. Afraid of the 'danger' of finding a really good
> cladogram.

Afraid of the danger of fooling yourself by a nice-looking outcome.
(And yes, I do some refereeing, if not in the paleo field. So far
didn't realise that referee is to be translated as "block-headed
conservative know-it-all" - don't worry, no offense taken, I had
enough papers rejected myself to understand bad feelings towards
referees.)

> > How could a cladogram prove such a thing? It can only sort beasts
> > based on characters, but it cannot find out the meaning of those
> > characters. To see whether something is adult or not, look at bone
> > sutures, or, even better, bone microstructure (see the Europasaurus
> > paper for an example of this).
> 
> A cladogram, I remind you, is a hypothesis. I promote this particular
> hypothesis because it works in every way. A hypothesis doesn't prove anything,
> but it is the best evidence (and usually they are widely accepted) that we
> have. So, so far it is a good hypothesis based on more evidence and less a
> priori assumption than any in its category.

A cladogram in itself is, as far as I understand it, not yet a
hypothesis (Cladogurus on the list, please correct me). It is an
ordering of entities (species or specimens) based on similarity
criteria (character states). *If* you assume, in addition, parsimony
and (more important to this discussion) that all entities you put in
*are on the same footing* (i.e., they *only differ wrt their
phylogenetic position*), only *then* does it become a hypothesis on
phylogeny. If this additional assumption is wrong (for instance,
because you added entities that are not on the same footing, like
adults and juvies), your phylogeny will probably also be wrong.


> But let's take the opposite stance.
> 
> Let's say that No. 9 is a juvenile. What is it a juvenile form of? Be specific
> and show evidence that it is indeed the candidate you propose and that it
> doesn't have more evidence (remember parsimony) that it is closer to my
> candidate.

And I cannot claim that the adult form has not been found because the
fossil record is complete? Circularity again?

But to find out whether it is a juvenile or not, you cannot look at
cladograms - look at the proposed juvenile's bones and check whether
they are fully ossified, remodeled, whether sutures have closed etc.

> Next point: do pterosaurs change their morphology during ontogenesis? Or not?
> The embryos in eggs are the only known pterosaurs for which their ontogenetic
> age is known precisely. Morphologically do they look like adults of one form
> or another? Or are they unclassifiable? My evidence indicates that embryos
> look like parents, and I used a cladogram. If embryos look like their parents,
> I think we should expect that juveniles do too. Don't you?

I would (frequently, not always) expect embryos (and juveniles) to
look like more basic members of the clade than their parents do. But
there's lots of things to change that, like paedomorphosis or special
selective pressure (foals have relatively longer legs than grown
horses). In addition, I would expect relatively larger brains and eyes
(but that goes for small adults as well and thus does not help much,
although it might if eyes are relatively huge).
 
> > True, but we are not talking mammals or birds here. If you look at the
> > growth rate of, say, a t rex, you can (with some reasonable assumption
> > on mortality) easily calculate that the *biomass* of small rexes (less
> > than 1 ton) was probably larger than that of adult rexes. Meaning,
> > juveniles filled a different ecological niche, and probably filled it
> > very well. If the same held for pterosaurs, then there were lots of
> > tiny pterosaurs filling the small-size ecological niche - just not of
> > a different species.
> 
> Thank you for that. I'm not following your argument though. No one making a
> cladogram cares what tiny pterosaurs ate versus what big pterosaurs ate. Or
> where they lived. Or what they did with their spare time.

But it was *your* argument stating that these tiny pteros would fit
because there are tiny birds nowadays, so this is something beyond a
cladogram. And this argument I reject - just because nowadays (with
mammals and birds) niche-partitioning is done species-wise doesn't
mean it was so always.

> > For pterosaur adults, this would mean they would eitehr have to give
> > up homoiothermy or they would need an incredible amount of food (like
> > an etruscean screw, for instance). Not impossible, but it's a stretch, I
> > think.
> 
> Again, no one making a cladogram cares what temperature pterosaurs were. But
> you bring up an interesting point. What happens with baby sauropods versus big
> sauropods? We know what size baby pterosaurs of any adult species were because
> the egg chute has a limited aperture. We can start from there. Just think how
> small a baby of a tiny pterosaur would have to be. Lizards are remarkable,
> aren't they?

Sorry, I don't understand this argument. How does it relate to the
problems an adult, warmblooded pterosaur of minuscle size would have
regulating its temperature (and it would be much worse off than the
shrew because of its heat radiators, aka wings)?  Again, this argument is
*not* about cladograms, it is about whether these teeny-tiny adult
pterosaurs are plausible enough to be entered into a cladogram.

Martin.


                   Priv.-Doz. Dr. Martin BÃker
                   Institut fÃr Werkstoffe
                   Langer Kamp 8
                   38106 Braunschweig
                   Germany
                   Tel.: 00-49-531-391-3073                      
                   Fax   00-49-531-391-3058
                   e-mail <martin.baeker@tu-bs.de>