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

Re: Recoded Aves eliminates polyphyly

Ken Kinman (kinman@hotmail.com) wrote:

<I accept Mickey Mortimer's arguments that Confuciusornithids and
Oviraptorids are closely related.  However, I have obviously come to a
very different conclusion how such a clade fits into the avian tree.>

  Uhm, Ken, Mickey's post was a joke. It was meant to show that you can
use almost any set of characters to support any phylogeny you wish. He
chose to ignore the various exclusionary characters that support different
toplogies to provide a monophyletic Oviraptoridae + Confuciusornithidae
that would have even excluded an oviraptorid (*Ingenia*) for the purposes
of its stability. Thus, because he had tomake various groups polyphyletic
to fit the theory, the assumption was meant to show that this iswhat
happens when one ignores alternative supports.

<It appears that "Enigmosauria" is not even paraphyletic, but

  Based on....

  I just don't understand, I guess ... how such numerous features are
"decided" to no longer amount to a monophyly. But see below.

<The many researchers who are getting a holo(mono)phyletic "enigmosaur"
grouping appear to be relying on characters which are subject to
functional convergences (mainly in the vertebrae and jaws).>

  Uh huh, as in the features Mickey supplied for *Confuciusornis* +
*Oviraptor* monophyly in light of excluding *Ingenia*. But these features
have strong support as the nearest groups do not have these, and for some
reason, the support for the features is aided by similar evolutionary
exaptation dynamics. The jaws of segnosaurs, for instance, would at first
be seen as herbivorous in nature, but the teeth are far from herbivorous
in form. They are both recurved, apical, with perpendicular posterior
serrations, and are in line with the sagittal median of the skull,
suggesting a fore-aft function of the jaw mechanism. This detracts from
the apparent parallel evolution of segnosaur and prosauropod jaws, which
differ in many ways. The jaws of segnosaurs appear to have developed from
carnivorous jaws. Parallels may be found in troodontid teeth, which
possess nearly all the hallmarks of segnosaur dentition, even though the
jaws don't. Functional adaptation looks like the teeth were carnivorous.

  Alternatively, most birds have the large external mandibular fenestra,
especially phorusrhacoids, by which a parallel I mean too draw in the near
future. Basal oviraptorosaurs, contra to Mickey, also possess the invasive
posterior surangular "prong," and the pubic concavity of the enigmosaur
complex is not so easily ruled out: an experiment involving *Segnosaurus*
that Mickey forgot (he performed this with me last year) involved removing
the pubic boot and the iliac peduncle ... and the pubis was _still_
cranially concave along the shaft. There is a juncture distal to the
acetabulum in the pubis of *Caudipteryx* and *Microvenator* that allows
that bone to also be concave cranially, and the same is true of
*Enigmosaurus* and *Nanshiungosaurus*. The pubes are otherwise less
distinctive in this area in *Beipiaosaurus,* but are similarly constructed
in *Nomingia*. Other features, including the seven sacrals, is an
ontogenetic adaptation in troodontids, as only one _large_ specimen of
*Saurornithoides* has it.

  So there are a few more unambiguous synapomorphies of Enigmosauria not
involving the jaw or vertebrae that Mickey does not account for. Braincase
features look like a great place to look for features, since one
researcher referred to it as "the Holy Grail" of evolution. Unfortunately,
the braincase of *Caudipteryx* is very poorly preserved and that of
*Avimimus* is obscured by matrix and is crushed across the surface. But in
*Erlikosaurus* and Oviraptorids, the braincase does share numerous similar
features found only otherwise in birds.

  Problem is, some people seem to see a segnosaur and exclude it from the
idea of birds perforce and do not even _try_ to look for similar features.
They are just too different. Oddly enough, of a variety of cranial
features oviraptorids have with birds, segnosaurs share, as well as the
majority of Maniraptora (not "Aves" sensu Kinman). The presence of raised
basioccipital tubera, short basipterygoid processes on the basisphenoid,
an inflated basicranium, the fenestra pseudorotunda elevated caudal to the
opening for the seventh cranial nerve the the second opening for the fifth
cranial nerve descended between the first opening (V) and the seventh,
forming an alignment of four laterosphenoid/basisphenoid openings, foramen
magnum larger than occipital condyle, ovate foramen magnum taller than
wide, palatine fenestra reduced to a foramen or slot due to the
ectopterygoid being rotated medially next to the palatine instead of
between it and maxilla/jugal and also incidentally increasing the
subtemporal fenestra, a triradiate palatine, and a triradiate
ectopterygoid. Fusion of the pterygoid and quadrate and depression of the
parasphenoid rostrum ventrally also occur between *Avimimus* and
Oviraptoridae, though several of the features seen above are not.

  By contention, *Avimimus* can be a mononyke by the similarity of its
hindlimbs. But this should be viewed with a great deal of caution: as
shown by Holtz (1994) the biomechanics of the hindlimb and the condition
of the arctomet are purely _functional_. Many features appear in several
greatly cursorial taxa. It is expected that basal taxa _not_ have this
condition. It becomes then as easy as saying that ornithomime and tyrant
dinosaurs are sister groups because they have similar hindlimbs. Which
does not meet up with the recently presented evidence of the last two
years, though Holtz (2001) has supported somewhat Arctometatarsalia on
compeletly different grounds. The tarsus of *Parvicursor*, unlike
*Mononykus* or even *Shuvuuia*, is a true "hyper"-arctometatarsalian pes,
as the third metatarsal does not rise higher than one-third of the lateral
and medial metatarsals; in the other two mononykids, there are still
facets for the third metatarsal ascendant between the medial and lateral
bones to the proximal end of the metatarsus. Unlike them, *Avimimus* has a
tarsus that is neartly identical to *Elmisaurus* in bearing the posterior
proximal crest and lateral flanges not associated with mononyke tarsi, and
a tarsal cap described by Osmólska when she described *Elmisaurus*
(Osmólska, 1981) can be referred to *Avimimus* on the nature of the
proximal flanges and abscence of a proximomedial "tongue" of bone on the
cap ... absent in *Chirostenotes* as well, but present in *Elmisaurus

<It is my conclusion that strict parsimony has led them astray, and that
they will not discover more accurate avian phylogenetic topologies without
some form of character weighting (as many of the characters they are using
are non-independent).>

  I would argue that many are. However, other than a method for arriving
at a tree that does not require "I don't like this character, this one
looks good, here's my apomorphy" methods, what systematic, _scientific_
would you propose for phylogenies? Cladistic machines, beleive it or not, 
require a mathematical parsimony engine to formulate the trees you seem to
appreciate in Mickey's work. Cladistic formulations require an assumption
of parsimony (mathematical probability) to work, because it's the only
method by which they can run. You can add assumptions of character state
transformations (order a state to run 0 1 2, instead of 1 <-0-> 2) or
force a specific outgroup scenario (or both) to polarize characters. You
can even allow the engine to consider two possible end conditions in a
state by designating the transformation as _both_ 0 and 1 = by replacing
"1" with "(01)" though this will _double_ for each like coding the run
time for the matrix. This makes such engines a lot less problematical than
just binary consideration. For those working with PAUP, and you haven't
figured it out (thanks Tom) you can also designate the engine to run
allcharacters as order, unordered, or select which ones will have that
condition,by replacing "deftype = ord;" with "typeset mytypes = ord: 1 2,
unord: 3 4;" and this will allow the machine to force a concept ... yet
still runs a probability algorithm. It's not just "parsimony" (the best
way to not make assumptions you cann't test, or won't)
<My refined (simplified) definition of Aves is now: All theropods
possessing (1) a convex coracoid glenoid, and (2) an enlarged fused
"semilunate" in adults.   Note:  the first is reversed in at least one
genus (Phasianus), and the distinctive shape of the second is partially
reversed in Protarcheopteryx and Caudipteryx.  Like Tom says, reversals
happen, but these appear to be very rare reversals (and thus remain highly

  Okay, finally, a diagnosis I can test!

  Well, for one, *Caudipteryx* appears to lack either, so it's not a
member of Aves _sensu_ Kinman. Unfortunately, this must be true for
Oviraptorosauria, of which it is apparently a basal member.

>   1  Plesion Troodontidae
>   ?  Pl. Bagaraatan
>   2  Archaeopterygiformes
>            1  Dromaeosauridae
>            B  "utahraptorid" family
>            2  Pl. Pyroraptor
>            3  Pl. Bambiraptor
>            B  Velociraptoridae
>            4  Pl. Sinornithosaurus
>            ?  Pl. Unenlagia
>            5  Pl. Microraptor
>            6  Pl. Rahonavis
>            7  Archaeopterygidae
>            8  {{expanded "Metornithes"}}

Metornithes shall not exist unless *Mononykus* is included; definition is
*Mononykus* + Neornithes by example, Perle et al., 1993, _Nature_ (the
first paper). There can be no more basal member of this taxon except for a
more primitive mononyke or "proper" avian. And for the record, it was I
who mistakenly supplied the *Mononykus* > Birds definition of
Alvarezsauria, which is wrong. That taxon has not been defined to my
knowledge and I was working off notes.

> _1_ Yandangornithiformes
>   ?  Pl. Sapeornis
>   2  Mononykiformes Kinman, 1994
>            1  Patagonykidae
>            2  Parvicursoridae
>            3  Mononykidae
>   3  Avimimiformes
>   4  Pl. Protarcheopteryx
>   5  Pl. Caudipteryx
>   B  Pl. Nomingia

  See above, this taxon is a non-caenagnathoid oviraptorosaur, as it's
postcrania and vertebrae are moreprimitive than caenagnathids or
oviraptorids, but more advanced than *Caudipteryx* or *Microvenator*.
Incidentally ... there's more data indicating *Caudipteryx* is an
oviraptorosaur thanb presently exists for *Microvenator*. Features known
for the latter are also present in the former that indicate inclusion into
Oviraptorosauria. Same goes for *Nomingia*.

>   6  Caenagnathiformes Cracraft, 1971
>            1  Pl. Microvenator
>            2  Caenagnathidae
>            3  Oviraptoridae
>            4  Confuciusornithidae
>   ?  Pl. Protopteryx
>   7  Longipterygiformes
>   8  Pl. Jibeina
>   ?  Pl. Hulsanpes
>   9  Pl. Boluochia
>   B  Iberomesornithiformes
> 10  Enantiornithiformes
>  B  Gobipterygiformes
> 11  Plesion Vorona
> 12  Patagopterygiformes
>  ?  Pl. Yanornis
> 13  Chaoyangiiformes

  Is this new?

Jaime A. Headden

  Little steps are often the hardest to take.  We are too used to making leaps 
in the face of adversity, that a simple skip is so hard to do.  We should all 
learn to walk soft, walk small, see the world around us rather than zoom by it.

Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!