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

RE: maxillary fenestration (was "magical" feathers)

> From: owner-dinosaur@usc.edu [mailto:owner-dinosaur@usc.edu]On Behalf Of
> Ken Kinman
> Dear All,
>       I think Philidor makes an important point that has not been
> answered
> (AFAIK).  So I'll rephrase what I believe to be the major thrust of that
> question:

Actually, I (based on previous discussions with him) did not get that as the
main thrust of his question.  Instead, I thought it had far more to do with
methodology and far less to do with the particular question of maxillary
fenestration in theropods.  However, it is certainly one which can be dealt

>        What coelurosaurs are known to have had maxillary
> fenestration,

_Compsognathus_, _Sinosauropteryx_, _Scipionyx_, _Proceratosaurus_,
_Ornitholestes_, _Marshosaurus_ (if it is a coelurosaur: not at all certain
about that), all tyrannosaurids, all ornithomimosaurs, all oviraptorids,
_Caudipteryx_, all troodontids, all dromaeosaurids, _Archaeopteryx_ all show
the presence of a maxillary fenestra.  Coelurosaurs which do not show a
maxillary fenestra: _Chirostenotes_ (the only caenagnathid for which the
maxilla is known), _Erlikosaurus_ (the only therizinosauroid for which the
relevent part of the maxilla is known).  The skull of _Shuvuuia_ is
ambiguous, depening on the interpretation of the osteology.  And when you
step outside of Coelurosauria, _Monolophosaurus_, _Afrovenator_,
_Allosaurus_, _Acrocanthosaurus_, and a few other non-coelurosaurian
theropods show a maxillary fenestra.

> and
> is there a good chance that Erlikosaurus and Chirostenotes didn't
> have such
> fenestration in their ancestry?
>      Could this be a red flag that we need to look closer at the
> assumptions
> that have resulted in the conclusion that they "lost" maxillary
> fenestration
> from ancestors that "possessed" this character?

Well, actually, this winds up being exactly what people in the field do:
look at the discordant (or concordant) distribution of observations.  See
many different source books on evaluating the distribution of characters
(The Compleat Cladist is a good start) for dealing with this type of issue.
For characters with discordant distribution (or even those with concordant
ones), we can examine these under the alternative models of Accelerated
Transformation (acctrans) or Delayed Transformation (deltrans).

Given a particular phylogeny (mine, Mortimer's, Sereno's, whoever's) we can
sometimes find alternative mappings of a particular character.  Let's take a
look at this particular case, thought, given a phylogeny where caenagnathids
and oviraptorids are sister taxa, where _Caudipteryx_ is the sister group to
this, where therizinosaurs and oviraptorosaurs are sister taxa, where their
sister group as a whole is Paraves (troodontids, dromaeosaurids, birds), and
where the remaining coelurosaurs are outside all of these put together.

It is by far a simpler explanation to put the presence of a maxillary
fenestra at the root of Coelurosauria, and have indepedant losses in
caenagnathids and therizinosauroids (two steps).  [I am ignoring post-Archie
birds in this, for this disucssion].  We COULD put it at the base of the
therizinosaur-oviraptorosaur clade, but that requires three steps in the
phylogeny above (i.e., one step to lose at the base of the "enigmosaurs",
and one step each for a reversal back in Oviraptoridae and _Caudipteryx_).
OR we could have every single OTU evolve it independantly (over a dozen

Of course if we chose the latter there doesn't seem to be a preferred reason
to assume that it would be independantly derived only at the OTU state, so
let's derive it independantly in each species of tyrannosaurid, and of
oviraptorid, and of ornithomimosaur, and of troodontid, and of
dromaeosaurid.  A few dozen extra steps.

If, on the other hand (face) the only taxa that had a maxillary fenestra
were _Caudipteryx_, Oviraptoridae, and the paravian clades, then we would
NOT have as much confidence that _Chirostenotes_ and _Erlikosaurus_ had an
ancestor with a maxillary fenestra.  In this case, given the same cladogram,
we could have evolved the maxillary fenestra at the base of the advanced
maniraptorans, then lost it in _Chirost._ and _Erlik._ (three steps) OR it
could have evolved independantly in _Caudipteryx_, Oviraptoridae, and
Paraves (three steps).  In this case: yes, we would certainly have to
question the assumption of the non-fenestrate maxillae of _Chirost._ and
_Erlik._ as being a derived (rather than primitive) condition.

To bring things back to the original starting point: this is the reason that
we can be very confident in _Velociraptor_ being feathered, but must still
be very cautious about _Ornitholestes_ and _Scipionyx_ and _Allosaurus_ (for
two examples).  _Velociraptor_ is in a part of the tree which strongly
optimizes that feathers were the ancestral condition; because of the less
secure nature of the position of _Sinosauropteryx_ and the missing data in
non-coelurosaurian tetanurines, we can not be as confident in the cases of
_Ornithol._ and Skippy and Big Al.

I hope that this helps.  If people want this expressed more clearly, with
pictures, check out a text on phylogenetic analysis and look under
"character state optimization" or words to that effect.

                Thomas R. Holtz, Jr.
                Vertebrate Paleontologist
Department of Geology           Director, Earth, Life & Time Program
University of Maryland          College Park Scholars
                College Park, MD  20742
Phone:  301-405-4084    Email:  tholtz@geol.umd.edu
Fax (Geol):  301-314-9661       Fax (CPS-ELT): 301-405-0796