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

Re: Me vs. Makovicky et al.- comparison and consensus

Mickey Mortimer (mickey_mortimer111@msn.com) wrote:

<Is Jaime right?  Is my tree more improbable than either the TWG's numerous
permutations, Rauhut (2003) or Holtz (2000, 2004)?  Well, it's only fair to
compare the taxa included in both analyses when comparing their results.>

  Let's actually look at the improbabilities by examining the phylogenies

Holtz's 2004 Dinosauria II topology, starting at Avetheropoda:
  |  |--Cryolophosaurus
  |  |--+--Fukuiraptor
  |  |  `--Siamotyrannus
  |  |--+--Sinraptor dongi
  |  |  |--Sinraptor hepingensis
  |  |  |--Yangchuanosaurus shangyuanensis
  |  |  `--Yangchuanosaurus magnus
  |  |--Lourinhanosaurus
  |  |--+--Allosaurus
  |  |  |  |--A. fragilis
  |  |  |  |--A. sp.
  |  |  |  `--A? tendagurensis
  |  |  |--Saurophaganax
  |  |  `--unnamed allosauroid
  |  `--+--unnamed carcharodontosaurid A
  |     |--unnamed carcharodontosaurid B
  |     |--Neovenator
  |     `--+--Acrocanthosaurus
  |        `--+--Giganotosaurus
  |           `--Carcharodontosaurus
     |--Richardoestesia gilmorei
     |--Richardoestesia isosceles
     |--Richardoestesia sp.
     |  |--Mirischia
     |  |--Compsognathus
     |  `--Sinosauropteryx
           |  |--Ornitholestes
           |  `--Nqwebasaurus
                       |  `--Oviraptorosauria
                             |  `--Dromaeosauridae

  Compared to Holtz's 2000 analysis and 2001 Ostrom Symposium update, this is
MUCH nicer, including a consistent Deinonychosauria.

Rauhut's 2000 analysis (as in the published one), starting at avetheropods:
--+--+--Magnosaurus (= Eustreptospondylus)
  |  |--Monolophosaurus
  |  |--Poekilopleuron
  |  |--+--Torvosaurus
  |  |  `--+--Chilantaisaurus
  |  |     `--Baryonychidae
  |  `--+--Afrovenator
  |     |--Allosaurus
  |     `--+--Sinraptoridae
  |        `--+--Neoventor
  |           `--Carcharodontosauridae
     |  `--+--Sinosauropteryx
     |     `--+--Compsognathus
     |        `--Mirischia
     |  `--Tyrannosauridae
           |  `--Ornithomimosauria
           |  `--+--Caudipteryx
           |     `--+--Microvenator
           |        |--Oviraptorosauria
           |        `--Avimimus
              |  `--Dromaeosauridae

  My notation is a little off, I am sure, due to the transcription from a
non-monospace format, but it's not too big of the biggie. Between Holtz and
Rauhut, we hare given different uses of splitting taxa up, such as Holtz using
a single OTU for Compsognathidae while Rauhut specifies different taxa, while
Rauhut lumps clades himself that Holtz split up. Nonetheless, the first major
difference is that Holtz maintained "spinosauroids"/"megalosauroids" in
successive outgroups to Avetheropoda, while Rauhut included them in
Carnosauria, making them avetheropods.

The most recent and comprehensive of the Theropod Working Group analysis to
include these taxa is that of Hwang, Norell, Ji and Gao (2004); Makovicky et
al. restricted their analysis to a smaller subset of the matrix and thus also
removed a lot of characters also not affecting the taxa involved:

        |  `--Tyrannosaurus
           |  `--+--Shenzhousaurus
           |     `--+--Harpymimus
           |        `--+--Garudimimus
           |           `--+--Archaeornithomimus
           |              |--Gallimimus
           |              |--Struthiomimus
           |              |--Ornithomimus
           |              `--Anserimimus
                 |  `--+--Sinosauropteryx
                 |     `--Compsognathus
                    |  `--+--Patagonykus
                    |     `--+--Shuvuuia
                    |        `--Mononykus
                    |  |  `--+--Erlikosaurus
                    |  |     `--Segnosaurus
                    |  `--+--Incisivosaurus
                    |     `--+--Caudipteryx
                    |        `--+--+--Avimimus
                    |           |  |--Caenagnathidae
                    |           |  `--Microvenator
                    |           `--+--Oviraptor
                    |              |--Rinchenia
                    |              |--Citipati
                    |              |--Conchoraptor
                    |              `--Ingenia
                       |  `--+--Archaeopteryx
                       |     `--Confuciusornis
                          |  `--+--IGM 100/44
                          |     `--+--Byronosaurus
                          |        |--Sinornithoides
                          |        `--+--Troodon
                          |           `--+--Saurornithoides mongoliensis
                          |              `--Saurornithoides junior
                          `--+--Microraptor zhaoianus
                                |--IGM 100/1015

  So what are all the big differences among these four analyses? Ignoring that
the TWG matrix and Holtz matrix now include more taxa, taxa I myself have
plugged into Holtz' matrix like *Yixianosaurus* and *Epidendrosaurus*, but also
those Holtz has added including *Deltadromeus*, *Eotyrannus*, *Dilong*, *Mei*
and so forth, we find that the topologies largely change around the new
included taxon, pulling here or pushing there. Rauhut was primarily concerned
with basal theropods up to Coelurosauria, and his thesis analysis and
subsequent publication of the same was that of testing the "allosauroid" and
"carnosaur" taxa, essentially Tetanurae all the way to Coelurosauria. TWG
focuses on Coelurosauria, as does Mickey, and Holtz focuses on non-avian
Theropoda, but is becoming more comprehensive as the strongest suit for that
analysis is the inclusion of characters. So what are the big differences?

  *Deltadromeus* is one of those "shifty" types, never stays in one place for
long -- don't date him, girls, he's trouble.

  "Compsognaths" closer to birds than are tyrannosaurs (Sereno also finds
tyrannosaurs closer to birds than compsognaths or even ornithomimes, and Holtz
finds a monophyletic Arctometatarsalia with tyrants, troodonts, and

  Those wascally mononykes, the Alvarezsauria. Y'know, we're missing those from
Liaoning, guys, time to conjure up a new one! (Quote-mine me, I dare ya....)

  *Bagaraatan*, but everyone knew it was odd. It says so in the paper. I'm sure
it does, in a straitjacket sorta way.

  GIN 100/44 probably flips around inside troodontids because it's a juvenile.
That, and *Scipionyx* is probably also responsible for "compsognath"
misintegrity due to it's near-hatchling age. Personally, I would remove them
from any analysis. As such, they likely pull taxa around them either into a
bunch or apart into a polytomy or "comb". Just look at Holtz'
"megalosauroideans", they likely suffer from plastic variation in underage

  And that's most of it. The rest appears to be a movement of one or two nodes
without much shifting around them, occassionally causing polytomies and
monophyletic splits to occur as they move about more recklessly. Those "problem
children" are highlighted by Mickey for the most part. The rest:

  The chaotic mess of caenagnathids and oviraptorids; the chaotic mess of
therizinosauroids; I can't comment much on tyrannosauroids and ornithomimosaurs
because no other analyses aside from the TWG work and Carr and Williamson
really includes them, thus the pool of variation isn't as strong as the general
coelurosaur phylogeny presented here; variation and formation within
dromaeosaurids, development of the troodont + bird clade is different than
shown above except where Holtz increasingly finds a bounce only between
bullatosaurs and deinonychosaurs, and the assemblage of those taxa at the base
of troodontids and dromaeosaurids, forming what is increasingly looking like
the basal paravian. I actually predict that it will look a lot like *Rahonavis*
with the head of *Buitreraptor* attached. Longish arms, long snout, small teeth
with no serrations and slight waisting of the crown-root junction, opisthopubic
pelvis, long tail with elongated distal caudals forming a stiffened structure,
small in body size, probably more or less scansorial or even slightly more on
the arboreal side, etc. Mickey's analysis differs in that it predicts the basal
paravian was larger bodied, given the "dromaeosaurine" and *Sinornithosaurus*
form the base models to predict from, or variable between large and small
bodies and thus less certain. Holtz, Rauhut, TWG, even Sereno and Senter, all
predict otherwise.

  Another aspect that affect the highly variable topologies from analysis to
analysis as performed by Mickey are the amount of taxa know by fragmentary
data, as well as possibly the lack of continuity between some included taxa
given unique variation or unique combinations of features (that no one seems to
be able to form a halfway decent Dromaeosauridae is also reflected on the
incomplete natures of the included taxa while at the same time competing among
small shared portions of their included data sets; I blame *Achillobator*,
among other things, in having some unenlagiine-like and oviraptorosaur-like
features, an odd pelvis, and funny dorsals). It is from the concensus above
that the variation of theropod cladogram can allow the prediction of
"improbabilities" to be discerned. I don't think I'm correct, nor do I think
Holtz, TWG, Rauhut, Senter, Sereno, or Mickey are correct. What's agreed on may
actually depend on similarity of coding, perhaps based on similar perspectives
of coding shapes based on arbitrary, yet agreed-upon, discernments of anatomy.
But we can easily use RFTRA and best fit analysis to find commonalities among
shapes of bones in all dimensions and use that to map phylogeny (percentage
congruence, element by element, and the parsimony that results from including
all data.

  If we're gonna find the answers, though, I bet we need to stop looking in the
EK of Liaoning and start looking at the late and mid Jurassic where it exposes.
One place may be to find Morrison-level strata, or earlier as at Dashanpu in
Sichuan, China, Guiamarota and Lourinha in Portugal, the Morrison itself in
USA, Solnhofen in Germany, and so forth.


  On to other things:

<3. Alvarezsaurids and enigmosaurs switch places relative to birds.>

  Mike Keesey himself (http://dml.cmnh.org/2001Sep/msg00403.html) makes best
the case for not using it on grounds that I've not used so far: it would
probably need to be anchored on *Enigmosaurus* in accordance to the PhyloCode
at any point in which the name is actually published and therefore usable. That
said, the vernacular use, even given by Holtz, is likely probably, but I balk
at that since anyone who reads it implies the taxon name so it's useless to
differentiate the two uses in any case, especially in a phylogeny. The very
spirit of the rule of not using names that mean nothing without publication of
a form that clarifies the names use when it's named (which so far has never
occured) would apply here in abusing the notion that this name means anything. 

  At this point I've gone pedantic. Mickey has broken his own word about using
Chure's thesis name, as far as I can tell, so i have no faith in his non-use of
the name. I have other arguments, such as the fact that this name's wide
internet use will likely prevent this name from likely being coined in any
_Nature_ or _Science_ paper, and that the prevalence of its use and application
(see Keesey, above) will probably change anyway, makes it unlikely this name
will at this point be used. The advice from those most likely to actually
publish this name is tending to finding the right study and venue to do so.
Since the only people working to explicitly study this group is Paul Sereno and
Barsbold Rinchen, along with Lü Junchang in various Chinese members, I am
fairly certain this name is largely going to be ignored by them. So -- for the
last time -- please stop using it.

  And finally, quoting earlier in the thread

      |  `--+--Megaraptor
      |     `--+--+--Coelurus
      |        |  `--Calamosaurus
      |        `--+--Tanycolagreus
      |           `--+--"*******saurus">

  Quoting from http://dml.cmnh.org/2002Mar/msg00547.html : "[I]n this
particular case Chure has said that he would prefer to have it referred to as a
new genus, but not use the name itself.  It's tricky, but I suppose I will
apologize to Chure in this case and not use the name until it is published."

  This was used as part of the discussion on the replacement name for
*"Chilantaisaurus" maortunensis*. Has the case of this name's availability been
resolved that it may be used without publication, or has the name been


Jaime A. Headden

"Innocent, unbiased observation is a myth." --- P.B. Medawar (1969)

Start your day with Yahoo! - Make it your home page!