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

Re: Carniadactylus paper and pterosaur ontogeny

 DP: Unfortunately Fabio did not test MPUM 6009 as a separate taxon to
 see if it would indeed nest with the former E. rosenfeldii.



 DM: This would require scoring all ontogeny-related characters
 as unknown for MPUM 6009. How many would that leave?

>>>

 You may be a little behind the times, David. Pterosaurs and their
 nonvolant sisters experienced isometric growth. If two taxa are the
 same species, no matter their ontogenetic age, they would nest as
 sisters. If MPUM 6009 nests elsewhere, closer to another taxon, it is
 not a sister. So, contra tradition, there are NO ontogeny-related
 characters.
This is just nonsense. Isometric growth has been demonstrated for the forelimb-hindlimb ratio -- and that's it. Skulls most obviously didn't grow isometrically in pterosaurs; neither do they anywhere else. 


 All prior assumptions were just that, assumptions. That
 includes tooth count, orbit/rostrum ratios, tiny pedal phalanx
 ossification, scapulocoracoid fusion, sacral fusion, etc.


Evidence?


 Put them all into a cladogram and your eyes will pop.
Phylogenetics obeys the law of "garbage in, garbage out". If I put actively misleading scores into a matrix, PAUP* will actively mislead me.
Hey, try it. Put either a baby *Apateon gracilis* or a postmetamorphic *Apateon gracilis* into a data matrix that contains other temnospondyls and lissamphibians, and your eyes will pop. 


 Maisano 2004 in JVP provides modern analogs.
Would surprise me. Could you give me the full reference? No Maisano published in JVP in 2004 (SVP meeting abstract volume included). 


 All the above is proven by adding embryo and hatchling Pterodaustro
 to any cladogram that has Pterodaustro adults. You'll find they nest
 as sisters.
That merely shows that *P.* has a large number of ontogeny-independent autapomorphies that are present in these matrices. It might be enough to put *P.* (adults and babies together) in some wrong place in the tree. 


 Although, it is of considerable interest that the one
 good complete skeleton of Pterodaustro (adult) has distinct pedal
 phalanx and metatarsal proportions from the Pterodaustro hatchling.
 Maybe NOT conspecific.
These are the very characters that I'd expect to vary with ontogeny. Or at least with absolute body size -- there's little that still works if you scale it up or down precisely. 


 A privately-owned Pteranodon, 1/4 the size of P. marshi (skull) and
 P.ingens (post-crania) has the proportions of an adult  including a
 fused extensor tendon process.
So it doesn't have the same relative skull size as either adult *P. marshi* or adult *P. ingens*? Did I understand that right? 


 Same goes for Bennett's "juvenile" Anurognathus. Unpublishable
 because it is privately held but a pdf is available through me.


I'm interested in that one.


 The only pterosaurs we KNOW that are NOT adults are those inside
 eggshells. All other suspects need to be placed into a phylogenetic
 context to test their nesting.
What I've been trying to get into your head for the last 5 years is that this does not work. PAUP* starts from the _assumption_ that all OTUs are in the same ontogenetic stage. If you violate that assumption, you introduce mistakes.
An important conclusion of my recent work (with Michel Laurin: 2008, Contributions to Zoology; 2009, Evolutionary Biology, online only so far), already found by other people earlier to some extent (check out fig. 10 of Warren 2007, JVP 27(4)!!!) is that not only do mistakes in the matrix lead to mistakes in the trees, but they lead to mistakes _of unpredictable kinds and magnitudes_ in the trees. You can't, for example, say "if there are any mistakes left in my matrix, they're tiny, insignificant ones that can only lead to tiny, insignificant mistakes in the results" -- _that's not how it works_. 

(Naturally I can send pdfs.)
On the influence of ontogeny, see our 2008 paper: we took a matrix that found the lissamphibians as temnospondyls, changed _three cells_ from known to unknown because we changed our interpretation of the little temnospondyl *Doleserpeton* from adult to immature or paedomorphic, and <poof> the lissamphibians came out as lepospondyls.
Long ago (probably in 2005) I sent you the "ontogeny discombobulates phylogeny" paper (Wiens, Bonett & Chippindale, Systematic Biology, 2005). Did you ever read it? 


 BTW, Bennett from 1991 on notes the
 bone texture issue appears on immature and mature
 Pteranodon specimens of the same size.
Didn't he conclude neornithean-like determinate growth in *Pteranodon*? (I don't have the paper handy right now.) That's how it looks like to me: in many birds, sexual maturity comes after growth ceases, so you'd expect immature bone texture in some individuals the size of adults because they've just finished growing and are only starting to remodel their bones.