Re: Revising Hou et al, 96 (very very long)

["Mickey Mortimer" <Mickey_Mortimer111@msn.com>]

Jaime Headden wrote-

>   Ack ... I looked at both the original paper, a photgraph of the
> material, and the newer drawing in Kurzanov, 1987, and saw no such thing
> applicable as a lachrymal that would fit your observation of an elongate
> orbital process.

Why don't you look at Kurzanov's (1981) figure 1, showing the skull in
dorsal view.  You'll note he labels the lacrimal (La), which extends
posteriorly on the orbital rim to contact the postorbital.  To cover the
anterior half of the long orbital rim like that, it has to have an elongate
posterior process.

>   Nonetheless, it is not clear that there is a posterior process of a
> lachrymal at all. Kurzanov also identifies a postorbital element (po in
> the figures); if this element is so, the rostral element _must_ be a
> frontal by nature of it's position, and what Kurzanov identifies as the
> orbitosphenoid is probably a fragment of the laterosphenoid. This permits
> enough ambiguity in the holotype cranium that it cannot be used to support
> the theory of a posterior, orbital lachrymal process.

Perhaps it is part of the frontal and Kurzanov imagined a suture between it
and the medial portion of the element.  Lacrimal-postorbital contact is
rather odd in coelurosaurs after all, but there are virtually no sutures in
that area in PIN 3907/3, so who knows?

>   We have a quadratojugal, and the length of the posterior process is
> proportionate to the size of the foramen in *Dromaeosaurus* and
> *Velociraptor*, and this is ample means to hypothesize the size of the
> foramen in *Deinonychus*; personal observation tells me that the quadrate
> foramen is as expansive in aspect in both caudal and lateral views, and
> the ascending process is bowed rostrally and is well-forward of the
> posterior third of the ventral length of the bone [quadratojugal] -- so
> this would appear to indicate a definate large foramen.

The size of the foramen is caused more by the precise articulation of the
quadrate, which is more anterior in Dromaeosaurus (the posterior process
extends past the mandibular portion of the quadrate in Dromaeosaurus, but
doesn't even reach the posterior edge in Velociraptor).  Keep in mind
Dromaeosaurus' quadratojugal also has a bowed dorsal process which is placed
at about the halfway point of the ventral margin.  Deinonychus may very well
have an enlarged foramen, but we don't yet have the data to determine this
for sure.  There is a known quadrate though, so we shouldn't have to wait
for long to be able to code it.

>   [on elongated rostral chevron processes], Mickey also wrote:
>
> <That are all also found in Rahonavis, a probable avialan.>
>
>   Too a much smaller degree. But perhaps with the pedal morphology, this
> is only a paravian/eumaniraptoran apomorphy, and non-illustrative to the
> phylogeny of *Sinornithosaurus*. However, see below...

No, I said Rahonavis has the "pes construction (an apparent troodontid +
dromaeosaurid synapomorphy, along with pedal digit II features)" that you
said was evidence for placing Microraptor in the Deinonychosauria.
Rahonavis actually has a more well developed sickle claw than Microraptor,
with a larger proximoventral heel on II-2 and larger ungual.  I'm glad you
agree that this supports the sickle claw as a synapomorphy for troodontids
and eumaniraptorans that was reversed in birds.

> <Byronosaurus, yes.  Sinovenator, Sinornithoides and Saurornithoides, no.
> Struthiomimus and Dromiceiomimus, yes. Gallimimus, no. It's not
> size-related, as Microraptor has the a deep premaxillary body.>
>
>   This is interesting as I see the incomplete *Gallimimus* premaxillae as
> being non-illustrative in this degree. True, in *Saurornithoides* the
> premaxillary body is much shortened in lateral aspect, but this is partly
> due to the medial curvature of the element as a result of the U-shaped tip
> of the snout. The body is much longer when mesiodistal length is taken
> into account. The premaxilla of *Sinornithoides* is evident in its length
> because the snout is not U-shaped at the tip; similarly, the
> *Byronosaurus* premax. The *Sinovenator* premaxilla is only twice as long
> as deep. The *Microvenator* premaxilla is slightly longer.

You may be correct regarding Gallimimus, as the reconstructed premaxilla
doesn't look like other ornithomimids (but Paul's reconstruction does).
What you failed to mention was that Sinornithosaurus' premaxilla is over
four times as long as it is deep, much more than most other non-avian
coelurosaurs (the exceptions being Compsognathus, ornithomimosaurs and
Shuvuuia).  My exact character is "length of premaxillary ventral edge more
than 250% the minimum height of the premaxilla under naris".  I stand by my
assertion this is a possible avialan character of Sinornithosaurus.

>   Of course they were convergent. However, variability of serration
> expression is much less diagnostic than tooth morphology as is presented
> by the theropod record, and the occassion to reduce them in some paravians
> (teeth of *Archaeopteryx* skulls have hardly been examined on this yet [or
> anyway, presented on the matter]), *Caudipteryx*, compsognathids,
> *Ornitholestes*, ornithomimosaurs, some troodontids ... it goes on.
> Arguing the lack fo serrations as being diagnostic is just not feasible
> given the record of variability.

Looking over the distribution of this character, it's actually more
parsimonious to have serrationless premaxillary teeth be basal for most
coelurosaurs (perhaps excluding tyrannosauroids) and reversing in Scipionyx,
Protarchaeopteryx, derived troodontids and dromaeosaurids.

>   Good, Now dismiss *Sinornithosaurus* as the descending process of the
> squamosal is robust enough to suggest that the too-short quadratojugal
> ascending process is incomplete. And inconclusive.

You mean the squamosals that Xu and Wu identify questionably enough not to
discuss them in the text (one of which was originally identified as a
prefrontal)?  The ascending process is near certainly not longer than
presented in figure 4E, as the amount of tapering indicates the preserved
base connected within a millimeter of the broken area on the main
quadratojugal body.

>   I wrote:
>
> <<Not exactly ... *Sinornithosaurus* doesn't have that much in the way of
> sterna.>>
>
>   and Mickey wrote:
>
> <Oh, not at all, with each sternal plate being longer than the sacrum and
> supporting about five attachments for sternal ribs. ;-) What insignificant
> sterna.....>
>
>   Pity the hips are so small as to render the length comparison moot;
> *Sinornithosaurus* and *Microraptor* all have very small hips compared to
> the remainder of the trunk; conversely, as in pygostylians but not
> *Archaeopteryx*, *Rahonavis* and *Unenlagia* have proportionately larger
> hips, which may be illustrative of a trend towards birdy-style thighs. The
> sternae are approximately half the length of the skull, similar to those
> of *Velociraptor* (GI 100/25), if slightly longer. Again, this may be
> plesiomorphic as the hypothesis of Paul (and corroborrated by Xu al.,
> 2002) is that loss of flight will result in smaller sternae for larger,
> ex-flight animals. Look at ratites, for instance. And *Pelecanimimus* has
> whoppers for sternals. Unfortunately, length of sternae are not the meter
> by which sternal ribs are measured, or there would be plenty more on bird
> sternae; instead, it is the count of sternal ribs, and only two have been
> identified in hypothesis for the morphology as in *Velociraptor* and
> *Citipati* sp. (GI 100/42), related to the lenth of the first lateral
> process. Such facets in *Sinornithosaurus* may be restricted to the same
> location, and to consider otherwise would require data I do not think we
> have.

Fine, I'll use the femoral length as a reference.  Sinornithosaurus has
sterna 57% of femoral length, while Velociraptor has a ratio of 48%.  Then
again, I never said sternal size was an avialan synapomorphy of
Sinornithosaurus, just the amount of sternal ribs.  Xu et al. (1999)
specifically state "its constricted anterolateral side bears a series of
posdsible five rib attachments".  Velociraptor and Citipati have three.

> <It's thinner than Archaeopteryx, but does have the same morphology. Very
> different from Velociraptor.>
>
>   I disagree... the element in *Archaeopteryx* is very robust, and small,
> whereas that in *Sinornithosaurus* is very wide, larger, and shallower in
> angle; similar morphology, but with a hypocleidial projection, is noted in
> *Velociraptor* and GI 100/42 (Barsbold, 1983; Norell, Clark, and Chiappe,
> 1995; Norell and Makovicky, 1997; Clark, Norell, and Chiappe, 1999).

Notice that not only is Velociraptor's furcula thinner than either
Sinornithosaurus or Archaeopteryx, it has the small hypocleidium and the
arms each have a convex inside edge, completely unlike the furculae of the
other two genera, which have concave internal edges.

> <Which is why Compsognathus lacks it..... we all know how huge it was.
> ;-)>
>
>   Uh-huh ... one taxon and rhetoric.

How are many taxa and a list of vales then?  See a clear increase in fibular
width (as a percentage of tibial width; on right in parentheses) with
increasing size (represented by femoral length to left)?  I don't.

53 Sinosauropteryx (30)
70 Archaeopteryx (13)
88 Rahonavis (15)
110 Compsognathus (43)
118 Nqwebasaurus (20)
124 Microvenator (13)
140 Sinornithoides (~15)
145 Nedcolbertia (35)
148 Sinornithosaurus (~15)
150 Caudipteryx (26)
188 Avimimus (~30)
200 Velociraptor (28)
232 Ingenia (36)
284 Deinonychus (32)
315 Bagaraatan (28)
555 Alxasaurus  (27)
665 Gallimimus (40)
740 Deltadromeus (~40)
770 Dryptosaurus (51)
1000 Gorgosaurus (45)

>   I know of the condyle. I am aware of this and have looked at it in
> regards to mandibular movement in birds. The apomorphy of a third condyle
> is conditional to the extension of the lateral process of the mandibular
> glenoid, and such a feature is easy to approximate in oviraptorid
> quadrates and *Sinornithosaurus*. My contention, however, is that the
> morphology I see in *Sinornithosaurus* is similar to that of oviraptorids
> and does not form a third condyle. This would require mandibular
> corroboration. *Microvenator* lacks such a condition on the mandible, it
> appears.

Well, Xu and Wu (2001) say it has three condyles while Maryanska and
Osmolska (1997) say oviraptorids only have two.  Erlikosaurus has three, but
it's the only other example I've found in non-avians.  It must not be
conditional to the lateral process, or else Velociraptor would have it and
Erlikosaurus certainly wouldn't.  Which quadrate of Microvenator would you
be referring to?  The unknown one? ;-)  (rhetoric good :-)  )

> <The distal end is lateromedially expanded, as in pygostylians.>
>
>   The condyle, or the shaft? I see the element is rather flattened in
> aspect, but this does not appear to be the same in pygostylians, which
> expand the shaft proximal to the distal terminus (condyle).
> *Archaeopteryx* lacks it, however, as does *Microvenator*.

The shaft.  Paul actually used this quite prominently in his SVP
presentation, it's easier to see in cf. Sinornithosaurus.  Archaeopteryx
probably does lack it.  That's why it's in the "characters also found in
more derived birds, but not Archaeopteryx" section of my list.  Again with
the unknown Microvenator elements.  Perhaps you mean Microraptor, but
phalanx II-I isn't known in that genus either.  Could you be referring to
the new skeletons (have an early copy of the new AMN paper perhaps)?

> <Of course it does Jaime, we believe you..... :-) Sinornithosaurus' pubis
> expands very gradually towards the tip, then rounds off.>
>
>   Please ... civility. The curvature of the shaft is what provokes the
> phrase "J-shaped". As for the margin expansion, such is as much a pubic
> boot as is that of troodontids.

Please... accuracy.  The shaft has a distinct 40 degree kink about 60% down
the shaft, expanding slightly after the halfway point.  This is not at all
like the "J-shaped" pubis noted in Unenlagia for instance, which is straight
except for the large foot, which is directed posterodorsally.
Sinornithosaurus has no abrupt expansion to indicate the presence of a pubic
foot.  Sinovenator has at least a small posterior expansion (possibly
broken) while Troodon has a small but distinct posterior foot as well (and a
huge anterior foot).

>   [on the proximoventral heel...]
>
> <Like troodontids and Microraptor?>
>
>   And a few other taxa, like the [possible basal maniraptoran]
> *Ornitholestes*. *Rahonavis* has it, but *Archaeopteryx* does not. This
> does not support a birds + Sinorn - dromies group, as I tried to say in my
> last post at the end below this section.

Pedal phalanx II-2 is missing in Ornitholestes (Ostrom, 1969), so the state
is unknown.  What are your mysterious "other taxa"?  Of course it doesn't
support a (dromaeo (Sinorn, birds)) topology, it was never supposed to.
Remember, YOU originally listed this character as one of the "other features
that support the Deinonychosauria and the close association of
*Sinornithosaurus* to Dromaeosauridae"? Sounds to me like it was intended to
provide evidence Sinornithosaurus is a deinonychosaur, and not an avialan.
Which is why I don't understand why you would be trying to prove its
existance in other taxa now.  _My_ side in this debate is that these
"deinonychosaurian" characters are plesiomorphic for eumaniraptorans or
eumaniraptorans + troodontids.

>   [on the raptorial claw]
>
> <Like troodontids, Rahonavis and Microraptor?>
>
>   As I said above, just cut and paste the entire paragraph ... though the
> condition may be equivocal in *Ornitholestes*.

Ornitholestes' second pedal ungual is also too poorly preserved to indicate
whether it has a raptorial morphology (Ostrom, 1969 again).  Again, I'd like
to see your "other taxa", though this time you might be able to cite some
birds.  Of course, that would support my position that they are troodontid +
eumaniraptoran plesiomorphies.

>   [on the posterior metatarsal laminae]
>
> <Like troodontids, Microraptor, Archaeopteryx and probably Rahonavis?>
>
>   With my being able to view the posterior metatarsus, I can certainly say
> that *Archaeopteryx* lacks this condition, and from what I've seen, so
> does *Rahonavis*. This is quite conditional, but this may be one of the
> basal deinonychosaurian synapomorphies, and Xu et al. touch on it as I
> recall.

Paul (DA) explicitly describes its presence in Archaeopteryx.  Note in
Rahonavis (anterior view) there is a large gap between the proximal halves
of metatarsals III and IV, but the distal halves are tightly closed off.
It's also true (but less extensive) between III and II.  Quite suggestive,
but that's why I only said "probably".  Certainly there's no way you can
refute it without actually seeing the specimen.

Mickey Mortimer