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

Senter et al. 2004 impressions



I just recieved a reprint of the paper, and have quite a few comments.
First, I should say that Senter has been extremely helpful, and his input
will vastly improve my coelurosaur analysis.  You know the topology from
Holtz's post- http://dml.cmnh.org/2004Aug/msg00249.html .  It's based on 101
characters, which makes it smaller than the TWG's, Sereno's, Holtz's and
mine.  The CI is .49, which is quite good in showing lack of bias.

While I'm not sure if I agree Incisivosaurus and Protarchaeopteryx are
sister taxa, note that they were run as a single OTU, and relevant taxa like
Avimimus and Nomingia were not included.  Of the six characters used to
support their pairing-
1. first premaxillary teeth enlarged and anteroposteriorly compressed.  The
compression seems to be the important factor, as Caudipteryx has an enlarged
first pair of teeth too (though seemingly cylindrical; Zhou et al., 2000).
However, labiolingual compression may be a plesiomorphy, as its seen in
carnosaurs, Compsognathus, dromaeosaurids and Troodon.  Tyrannosauroids have
labiolingually compressed teeth, and those of Byronosaurus and Archaeopteryx
are described as round.
2. diminutive peglike teeth immediately posterior to enlarged pair of teeth.
Again, Caudipteryx shares the size difference in posterior premaxillary
teeth, though they seem to be more slender judging by the description as
"needle-like" and Zhou et al.'s illustrations.
3. diminutive lanceolate posterior teeth.  Though I haven't measured tooth
size yet, it seems quite possible that small maxillary teeth are shared with
Erlikosaurus, Pelecanimimus, troodontids and birds, so may simply be a
maniraptoriform plesiomorphy reversed in dromaeosaurs.  The shape refers to
constricted bases, which are also seen in these taxa (and
Microraptor/Cryptovolans), and may also be plesiomorphic.
4. short, high skull.  True of Caudipteryx, Avimimus and oviraptorids as
well, so probably an oviraptorosaurian plesiomorphy.
5. tall premaxilla.  Also seen in oviraptorids, though not Caudipteryx or
Avimimus.
6. dentary with anterior margin beveled so that distal tip is a 50-60 degree
point.  Also seen in caenagnathids, but not Caudipteryx, Avimimus or
oviraptorids.

Interestingly, Senter's personal examination of Protarchaeopteryx robusta's
teeth showed them to lack serrations, the supposed serrations on one tooth
being chips instead.

Four additional taxa were added separately by Senter et al. to test their
positions.  Unenlagia was a dromaeosaur (my informal term for
Dromaeosauridae + Microraptoria) outside both Dromaeosauridae and Chinese
microraptorians.  Hulsanpes was a dromaeosaur outside Dromaeosauridae.
Pyroraptor was a dromaeosaur outside Chinese microraptorians and Utahraptor
+ Dromaeosaurus.  Variraptor was more derived than tyrannosauroids, but not
a Chinese microraptorian or a member of the Achillobator + Utahraptor +
Dromaeosaurus clade.

The following changes in tree length are required for different topologies-
Sinovenator as a bullatosaurian troodontid- 16 more steps.
Troodontids (with Sinovenator) as deinonychosaurs- 12.
Troodontids and Sinovenator as successive outgroups to dromaeosaurs- 12.
Sinovenator and dromaeosaurs as ornithurines (sensu Gauthier, 1986)- 9.
Troodontids (with Sinovenator) and dromaeosaurs as ornithurines- 21.

Sinovenator is sister to dromaeosaurs based on the proximoventral heel on
pedal phalanx II-2, enlarged pedal ungual II, and ventrally facing anterior
articular surfaces on cervical centra.  Other troodontids have the first
two, but Sinornithoides and Troodon lack the latter.  However, Senter notes
in a page sent with his paper that the latter character should be dropped
due to variation in the cervical column.  Furthermore, Sinornithoides and
Saurornithoides lack opisthopuby, and Sinornithoides, Saurornithoides and
Troodon lack a ginglymoid third metatarsal.  Sinovenator has these two
characters, which are paravian synapomorphies in Senter et al., matrix.

Of the ten troodontid characters shared by Sinovenator suggested by the TWG,
Senter et al. did not include eight.  The lack of a basisphenoid recess,
vertically oval foramen magnum, lateral dentary groove and pneumatic
quadrate were not invalidated by Senter et al..  However, they report that
the antorbital fossa wall in Sinovenator is comparably developed to
dromaeosaurids, the dentary teeth are subequal in size and density
throughout the tooth row, and metatarsals II and IV are subequal in
diameter.  In regards to the TWG's character "Dentary and maxillary teeth
large, less than 25 in dentary (0) or moderate number of small teeth (25-30
in dentary) (1) or teeth relatively small, and numerous (more than 30 in
dentary) (2).", Senter et al. note troodontids should be coded state 2, not
state 1.  I find it interesting the TWG code Incisivosaurus as state 1, when
it has 8-9 dentary teeth.  Also, Erlikosaurus and Alxasaurus should have
been coded state 2.  Regardless, Sinovenator does share more than 20 dentary
teeth with troodontids, unlike dromaeosaurs or basal birds, though the
presence of this state in segnosaurs, Shuvuuia and Pelecanimimus would make
the polarity somewhat complicated.  Senter et al. note Khaan, Rahonavis and
Archaeopteryx have short, wide only slightly inclined anterior dorsal
transverse processes, which makes it possible this is a maniraptoran
synapomorphy reversed in dromaeosaurs.  I find this character to be
problematic, as inclination varies with position in the dorsal column, and
taxa like Gallimimus have transverse processes which are wider than long,
yet are coded as plesiomorphic by the TWG.  Finally, Senter et al. note
ornithomimosaurs also have a dorsoventrally flat internarial bar, so this is
just as parsimonious in his topology as in the TWG's.

Bullatosauria is supported by -
1. maxillary fenestra enlarged and anteroposteriorly elongate.
2. bulbous parasphenoid base.
3. elongate snout (preorbital length >3x height at anterior border of
preorbital bar).
4. more than 30 dentary teeth.
5. cervical centra significantly longer than dorsal centra.
6. cervical centra >3 times longer than tall.
7. arctometatarsus.
8. pedal phalanx II-2 <60% of pedal phalanx II-1.
9. pedal phalanx II-2 <200% length of distal condylar eminance.
DELTRAN only-
10. craniofemoral ratio <80%.
11. dorsal pleurocoels absent.
Note 5/6 and 8/9 are basically the same character stated different ways-
"cervical centra long" and "pedal phalanx II-2 short".  If Harpymimus and
Shenzhousaurus were included, 4 would be ambiguous for ornithomimosaurs,
since they lack it.  Harpymimus and "Grusimimus" lack 7, which would ensure
it's not a bullatosaurian synapomorphy.  Gallimimus, cf. Ornithomimus,
Sinornithoides and Troodon (Britt, 1993; Makovicky, 1995; Currie and Dong,
2001) have dorsal pleurocoels in anterior vertebrae, though they are
reported absent in Pelecanimimus,  Sinornithomimus and Byronosaurus.  Senter
et al. code Byronosaurus as unknown, perhaps because they lacked access to
Makovicky et al. 2003.  They code dorsal pleurocoels as absent in
Pelecanimimus, Struthiomimus, Sinornithoides, Saurornithoides mongoliensis
and Troodon.  For Sinornithoides and Troodon, I'll have to ask Senter about
the coding conflict.  For Saurornithoides mongoliensis, I think only
posterior dorsals are known, which makes it impossible to code.

Microraptoridae was not used because if Sereno's (1998) definitions are
accepted, Microraptoridae would be inside the stem-based Dromaeosauridae.
Sounds good, though if Sereno's definitions actually are accepted, that
would be the least of our worries. ;)

NGMC 91 is closer to Microraptor/Cryptovolans based on-
1. mesial serrations absent on posterior teeth.
2. dentary with midlength increase in height.
I'll have to see what Senter et al. mean by "posterior teeth", as the first
three maxillary teeth lack mesial serrations in Sinornithosaurus, as do the
first four dentary teeth.  The preserved teeth in NGMC 91 seem posterior to
this though, and I've never considered the second character.  It's quite
possible Senter et al. are correct in this topology.

Regarding the synonymy of Jiufotang microraptorians, Senter et al. state the
manual proportions that diagnose Cryptovolans are also seen in other Chinese
microraptorians.  However, this is not true.  Czerkas et al. (2002)
diagnosed Cryptovolans by its high III-1/III-3 ratio.  If we examine the
III-2/III-1 ratios, we see Cryptovolans has 23-24%, lower than any other
theropod.  Microraptor zhaoianus' is 41%, Microraptor gui's 64%, NGMC 91's
43%, Sinornithosaurus' 39% and Graciliraptor's 51%.  If we instead look at
III-1+III-2/III-3 ratios, we see Cryptovolans' is 170%, which is larger than
coelurosaurs except for Nqwebasaurus, Ingenia, scansoriopterygids and
Shenzhouraptor.  Microraptor gui's is 149%, M. zhaoianus 137%, NGMC 91 119%,
Sinornithosaurus 105%, Graciliraptor 96%.  Senter et al. are correct in
noting the unfused sternal plates of a Microraptor zhaoianus specimen might
be due to ontogeny, and that other Chinese microraptorans have bowed tibiae
and bent pubes.  M. gui was in part diagnosed by its short manual digit I,
which Senter et al. say is the same in M. zhaoianus and Cryptovolans, using
a mcI+I-1/humerus ratio.  If we instead measure I-1/mcII ratios, we see M.
gui's is 57%, while Cryptovolans' is 79%.  M. zhaoianus' is unknown.  They
all have subequal mcI/mcII ratios though.  Finally, Senter et al. dismiss
the enlarged radial biceps tuber as individual variation.

Adasaurus is basal because it lacks the synapomorphies of-
Saurornitholestes + Deinonychus + Achillobator + Utahraptor + Dromaeosaurus
1. posterior dorsal centra >1.2 times taller than long.
Achillobator + Utahraptor + Dromaeosaurus
1. obturator process proximally placed.
2. femur longer than tibia.
My suggested Achillobator + Dromaeosaurus character of stout pedal phalanx
II-2 (unknown in Utahraptor; not used by Senter et al.) which Adasaurus
shares, leaves a position sister to the Drom + Utah + Achillo clade equally
parsimonious to Senter et al.'s placement.

Senter et al. also have a discussion of support for neoflightless
dromaeosaurids.  Of the five characters they examine, at least three seem
ill-suited for a judgement based on parsimony when other characters are not
taken into account.  These are asymmetrical remiges, enlarged humeral
diameter and elongate penultimate pedal phalanges.  These characters are
expected to reverse in neoflightless taxa, as can be seen in the numerous
flightless birds today, Patagopteryx and hesperornithines.  So if there are
at least two neoflightless branches off the 'main flighted line', convergent
acquisition of flight will be supported.

Those who want a scan, ask me offlist.

Mickey Mortimer
Undergraduate, Earth and Space Sciences
University of Washington
The Theropod Database - http://students.washington.edu/eoraptor/Home.html