Chure's Thesis Part 1

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

Much as I did for Rauhut's thesis about this time last year, here is a
summary of Dan Chure's (2001) thesis-

Chure, 2001. A new species of Allosaurus from the Morrison Formation of
Dinosaur National Monument (UT-CO) and a revision of the theropod family
Allosauridae.

In the past year, there have been disagreements on the DML as to what is
proper etiquette regarding nomina nuda, specifically those awaiting official
publication. Chure has stated onlist (in a message forwarded by Ford) that
discussion of the contents was harmless, but that he would prefer the three
taxa named within to not have their names used. I have followed this
condition below.
As this is a thesis, the conclusions are not as certain as those in a
scientific journal.  It's obvious this is a preliminary work due to the
amount of typographical errors and nonsensical statements.  Indeed, I hear
several things will be changed by the time it is published.  Still, the
thesis is an excellent piece of work which I recommend to anyone interested
in theropods.  It can be ordered from
http://www.umi.com/hp/Products/Dissertations.html for $34, well worth the
contents.
I should note than Chure uses Sereno's definition of Allosauroidea
throughout the text, equivalent to the more widely used Carnosauria.

Apatodon Marsh 1877
A. mirus Marsh 1877
Kimmeridgian-Tithonian, Late Jurassic
Morrison Formation, Rocky Mountains region, US
Holotype- (USMN coll.) partial vertebra (lost)
Comments- This was originally described as the partial mandible of a large
suid mammal, but Baur (1890) identified it as a partial dinosaurian
vertebra.  What exactly it is will remain a mystery, as the specimen is
lost.  Olshevsky (1991) synonymized it with Allosaurus, but there is no
justification for this.  It is Dinosauria indet..

Allosaurus Marsh 1877
Diagnosis- dorsal wall of maxillary antrum fenestrate; large, mediolaterally
compressed, dorsally projecting lacrimal horn; internal mandibular fenestra;
second sacral centrum with large lateral foramen; obturator process with
elongate lamina that extends past pubic peduncle.
Comments- The interdental plates fuse at a young age in Allosaurus, which
makes me think the partially fused plates of Bambiraptor may be the adult
condition, contra Burnham et al. 2000 (as in Sinornithosaurus).  The maxilla
is excluded from the external nares in all Allosaurus specimens, those
illustrated otherwise are composites (Madsen, 1976) or incorrectly restored
(Osborn, 1903). Allosaurus does not have a pneumatic jugal, the foramen is
neurovascular (contra Currie and Zhao, 1994; contra Witmer, 1997).  The
external mandibular fenestra of Allosaurus is not as small as Madsen (1976)
illustrates, instead being longer but still shallow.  Some large Allosaurus
specimens have cervical ribs fused to the vertebrae.  The second sacral
centrum has a large lateral foramen in Allosaurus, interpreted as
neurovascular because "pneumatization ends in anterior dorsals in
Allosaurus".  I find this reason unconvincing, as ornithomimids only
pneumatize the first two dorsals, but have sacral pleurocoels.  One specimen
of Allosaurus fragilis (AMNH 813) probably has a completely enclosed
obturator foramen.  One Allosaurus specimen (DMNH 2149) has a mid-dorsal
ischial process.
Chure examines the issue of A. fragilis vs. A. "atrox" (Paul, 1988), though
he notes that the holotype of Creosaurus atrox doesn't contain the elements
necessary for Paul to distinguish between his two species.  Regarding the
supposedly shorter skull of A. fragilis, he notes the only short skull known
is that of USNM 4734, which was found disarticulated.  When it was
reconstructed by Gilmore, he had to "comprimise in regard to the exact
articulation of the elements".  There are large plaster filled gaps in the
specimen, the contact between the maxilla, jugal and lacrimal is missing,
the dentary is from another specimen (USNM 8335), the other mandible is
plaster, the palate is fragmentary, and the postorbital regions are
distorted judging by their asymmetry.  Chure notes the maxilla is
reconstructed too far posteriorly, as the lacrimal articulation of the
dorsal process is projecting into the antorbital fenestra.  The angle
between the maxillary body and its dorsal process is similar to other
Allosaurus specimens, which wouldn't make sense if the snout were shorter.
Similarily, the angle between the anterior and ventral lacrimal processes is
in the middle of the range Allosaurus exhibits, with Cleveland-Lloyd A.
"atrox" specimens exhibiting marked variation.  The nasal of USNM 4734 is
broken and the anterior part moved dorsally and rotated ventrally.  The
lacrimal horn shape shows many intermediates between tall and triangular
(USNM 4734) and low and rounded (DINO 2650).  There is an example of a
triangular lacrimal on a long skull (MOR 693).  Contra Paul (1988),
triangular lacrimals are known from the Cleveland-Lloyd quarry (eg. UU
40-581).  Though Paul claimed A. "atrox" has a more robust neck, therte is
no difference when cervical width/length ratios are compared.  Similarily,
though Paul claimed A. "atrox" has a more robust forelimb, no difference was
noted when humeral circumference and length were quantitatively compared
(circumference/length ratio .45 in fragilis, .36-.49 in A. "atrox").
Finally, both "species" are found in the same quarry, as evidenced by AMNH
600 (referred to A. fragilis by Paul) and AMNH 666 (which he referred to A.
"atrox").  This is contrary to the stratigraphic distinction supported by
Bakker and others.  In conclusion, there is no evidence for the
fragilis/atrox dichotomy advocated by Paul and Bakker.
Creosaurus atrox (Marsh, 1978) is based on a few skull elements, hyoid,
vertebral fragments, a supposed sternum which is lost, ilium, astragalus and
pedal fragments.  The jugal shows an A. fragilis-like ventral deflection,
but the ilium has an A. sp. nov.-like vertical ridge that Chure states "is
not present on any A. fragilis specimen".  Chure never mentions either of
these characters in reference to systematic position and merely refers
Creosaurus atrox to Allosaurus sp..  Indeed, he says there are no characters
to differentiate it from either species.  Odd.  Creosaurus also has a fossa
on the dorsal jugal process that is apparently natural.  Another A. fragilis
specimen (AMNH 5753) has a similarily placed, but shallower fossa.  Finally,
Chure notes that the second and third sacrals of C. atrox are V-shaped
ventrally in section, unlike those of either Allosaurus species.  The fourth
and fifth sacrals of A. sp. nov. have this type of venter, but the fourth
sacral centra of all Allosaurus specimens has a large (pneumatic?) lateral
foramen, unlike these sacrals.  So the sacrum is different from other
Allosaurus specimens, no matter which vertebrae are represented.  I think
there is a possibility Creosaurus atrox is a valid species of Allosaurus,
but will withold final judgement until Chure's work is published without the
typos and inconsistant statements in this thesis.  For now I recognize it as
Allosaurus sp..

Allosaurus sp. nov.
Kimmeridgian, Late Jurassic
Salt Wash Member of Morrison Formation, US
Holotype- (DINO 11541) (5.6 m, 614 kg, subadult) right half of skull (630
mm), stapes, partial sclerotic ring, mandible, (presacral column 1.814 m)
second through eleventh cervical vertebrae, sixth through eighth cervical
ribs, first through twelfth dorsal vertebrae, second through twelfth dorsal
ribs, eighteen rows of gastralia, sacrum (438 mm), firsth through eighth
caudal vertebrae (722 mm), midcaudal vertebra, sixteen distal caudal
vertebrae (991 mm), seventeen chevrons, scapulae, coracoids, furcula,
humeri, radius, ulna, carpus, manus, keratinous sheath of manual ungual I,
ilia, pubes, ischia, femora, tibiae, fibulae, astragalus, calcaneum, distal
tarsals 3, distal tarsal IV, metatarsal I, metatarsal II, phalanx II-1,
phalanx II-2, metatarsal III, phalanx III-1, phalanx III-2, metatarsal IV,
phalanx IV-1, phalanx IV-2, phalanx IV-3, phalanx IV-4, pedal ungual IV
Diagnosis- row of neurovascular foramina below antorbital fenestra in
maxilla; axial intercentrum rotated dorsally; axial intercentrum with flared
rim; accessory ossifications on the anterior and posterior edges of proximal
caudal neural spines; anteroposteriorly elongate obturator notch in pubis
delimited by triangular processes; elongate proximolateral corner of pedal
phalanx II-2.
The following characters were stated by Chure to differ from A. fragilis,
but were not included in his diagnosis- cranial narial fossa less well
defined; large foramen between lacrimal and jugal at posteroventral corner
of antorbital fossa; larger maxillary antrum; rounded lacrimal horn
(distortion?); lateral pneumatic recess of lacrimal absent (distortion?);
lateral vertical ridge on lacrimal horn; lacrimal horn not rugose; straight
ventral margin of jugal; parietals fused posteriorly; parietals taller than
wide in posterior view; basipterygoid recess well marked and invasive;
myohyloid foramen nearly enclosed ventrally; retroarticular process shorter
posteriorly; teardrop-shaped internal mandibular fenestra; nineteen dentary
teeth; odontoid process of axis tall and narrow in anterior view; third
cervical neural spine slanted posteriorly; fourth through sixth neural
spines more anteroposteriorly elongate; cervical pleurocoels change in size
throughout the vertebral column; cervical epipophyses mediolaterally
compressed; eleven cervical vertebrae; main hypapophysis on presacral 11;
fourth and fifth sacral centra with v-shaped ventral margins; more
pronounced notch between acromion process and anterior coracoid edge;
scapular blade not as expanded distally; coracoid extremely thin anteriorly;
more gracile ulna; longer thinner olecranon process; ulnar entocondyle
lower; straighter ulnar shaft; proximal and distal ulnar ends offset by
about 45 degrees; metacarpal I longer;  low vertical ridge above acetabulum;
pubic boot less tall and massive; femoral head is directed ventromedially;
strongly developed mediodistal crest on femur; less curved cnemial crest.
Description- This was discovered in 1990, though the skull was not located
until 1996. The only other fossil in the quarry is DINO 16456, which
consists of six proximal caudals and chevrons of a much larger theropod. It
is unfortunately difficult to continue collection of this specimen.
The Morrison Formation is Late Oxfordian-Tithonian, based on Argon dating
(Kowallis et al., 1998), without extending into the Early Cretaceous as some
have proposed.  The specimen can be viewed here-
http://www.dinopress.com/vol2/vol2_E/vol2_P1_e.html#_01 .
Only the left half of the skull is preserved, but every element is
represented except the vomer. The sclerotic ring is preserved, as it is in
A. fragilis specimen MOR 693. It probably contained around 32 plates.  The
ninth cervical centrum has two pairs of pleurocoels.  There is uncertainty
regarding the number of cervicals versus dorsals.  No ribs were found in the
transition area, but Chure identifies presacral 11 as being a cervical based
on the parapophysis, which is located entirely on the centrum.  I think it
may be a dorsal because it has a hypapophysis.   The accessory ossifications
on the caudal neural spines are compared to laminar spinous processes in
Alligator cervicals, and hypothesized to be the result of a well developed
ligamentum elasticum interlaminare.  I feel A. fragilis may have had similar
structures, though they were unossified.  This is because A. fragilis has a
similar step in the anterior edge of its caudal neural spines, which are
filled with the accessory ossifications in A. sp. nov..  There are eighteen
rows of gastralia in the complete series.  The first is composed of only two
elements, but the rest are composed of four.  Gilmore (1920) erroneously
thought there were seven elements in each row because USNM 4734 sustained an
injury in that area that broke many gastralia, forming false joints between
them.  His "single median gastralium" was a furcula.  The humeri are
dissimilar from each other, though only partially due to postmotrem
deformation it seems.  The manus is complete except for the distal portions
of unguals I and III.  Manual ungual III has a weak proximodorsal lip.
There is a nutrient foramen on the ilium, above the acetabulum and anterior
to the vertical ridge.  Chure suggests the supposedly pneumatic foramen in
Piatnitzkysaurus' ilium is actually neurovascular, and cites a Megalosaurus
ilium (BMNH R1100) with a similar structure.  The obturator process of the
ischium has a remarkably elongate proximal corner, extending anteriorly past
the pubic peduncle.  Examination of other Allosaurus specimens indicates
this was the normal condition, and that the thin lamina had broken off in
most specimens.

Allosaurus fragilis Marsh 1877
Diagnosis- two large pneumatic recesses that open laterally in lacrimal
horn; ventral margin of jugal sharply deflected ventrally at midlength;
metacarpal I shorter.
Holotype- (YPM 1930) tooth (55 mm), incomplete cervical or anterior dorsal
centrum; incomplete posterior dorsal centrum (85 mm), posterior dorsal
centrum (105 mm), two dorsal rib fragments, humeral fragment, pedal phalanx
III-1 (109 mm)
Description- Unfortunately, the holotype cannot be identified as A. fragilis
or A. sp. nov., although it is definitely Allosaurus.  The nearly complete
specimen USNM 4734 is designated the topotype, as it comes from the same
quarry as the holotype.
Allosaurus ferox (Marsh, 1896) is based on YPM 1893 (partial skull, partial
dentaries, partial surangular, dorsal central fragment, dorsal rib
fragments, scapular fragment, incomplete pedal ungual).  Marsh diagnosed it
by the presence of a maxillary fenestra, which is now known to be present in
all Allosaurus specimens.  The few unique features (sinuous ventral
premaxillary margin, convex ventral maxillary margin) are causedby incorrect
restoration, though the antorbital fossa is better developed than most
specimens.  The deflected ventral jugal margin shows it is synonymous with
A. fragilis.
Allosaurus lucaris (Marsh 1878) was originally diagnosed by the
opisthocoelous anterior dorsal centra with ventral keels and anteriorly
placed parapophyses that are the usual characters of Allosaurus.  It's known
from a partial mandible, hyoid, partial posterior cervical, dorsal and
caudal vertebrae, many vertebral fragments, a humerus and a partial
forelimb.  It shows no unique characters, and Chure refers it to A. fragilis
without specifying why it is not A. sp. nov..
Allosaurus "whitei" (Pickering, 1996) was based on a skull with hyoid,
presacral column, sacrum and pelvis (AMNH 666).  It was diagnosed by the
same characters Paul used to distinguish A. "atrox", so is invalid for the
same reasons.  Chure rejects the validity of the name, as Pickering didn't
follow ICZN Article 7 Recommendation 7a, Article 8a or Recommendation 8A.
It is therefore a nomen nudum in addition to being a junior synonym of A.
fragilis.

Antrodemus Leidy 1870
A. valens (Leidy 1870) Leidy 1873
= Poekilopleuron valens Leidy 1870
= Megalosaurus valens Nopsca 1901
= Allosaurus valens Gilmore 1920
Morrison Formation?, Colorado, US
Late Jurassic?
Holotype- (USNM 218) posterior half of sixth? caudal centrum (~125 mm)
Comments- Chure finds that this specimen is probably Allosaurus based on
comparison to other Morrison genera, though I'd be curious to see how it
compares to other carnosaurs.  He states that it cannot be determined which
spewcies of Allosaurus it belongs to, so it is indeterminate at that level
at least.  One good reason Chure has for retaining the name Allosaurus is
that even though its holotype is also indeterkinate to the species level,
there is a topotype from the same quarry.  The exact provenence of
Antrodemus on the other hand, is unknown.

Chingkankousaurus Young 1958
C. fragilis Young 1958
Campanian-middle Maastrichtian, Late Cretaceous
Wangshi Series, China
Holotype- (IVPP V636) distal end of anterior dorsal rib (92 mm wide
distally, 53 proximally)
Comments- This specimen was originally identified as a scapula, and later
assigned to the Tyrannosauridae by Molnar et al. (1990) because of its
narrow shaft.  I agree with Chure that the holotype does not resemble a
tyrannosaurid scapula.  It is less expanded distally, more rugose at the
distal tip, with a transverse groove just proximal to the tip, and has a
narrower cross section with strongly keeled edges.  He assigns it the
Coelurosauria indet..  I disagree, as it does not resemble coelurosaur
scapulae.  Of theropod scapulae, it resembles Allosaurus most closely,
differing in being narrower proximally and having a straighter distal edge,
along with the last three characters listed above.  The straight distal edge
with a rugose surface (and a groove immediately proximal to this) bears a
resemblence to the odd anterior gastralia of Acrocanthosaurus (Harris,
1998), but is about 2.5 times as large, expands more gradually and has a
different cross section.  The much greater size of Chingkankousaurus argues
against this interpretation, as the Acrocanthosaurus specimen described by
Harris is one of the largst theropods known.  Another possibility is that
this is the distal end of an anterior dorsal rib.  Although most dorsal ribs
have tapered distal ends, those that contact sternal ribs have squared
expanded ends that are rugose.  The cross section of many theropod dorsal
ribs has an intercostal ridge on the anterior side and a costal groove on
the posterior side.  In addition, a dorsal of Acrocanthosaurus shows an
expansion similar to that in Chingkankousaurus, though the distal tip is not
as broad as some other theropods'.   The size is also congruent,
Chingkankousaurus' being about 75% the size of Acrocanthosaurus.  As
Chingkankousaurus resembles the distal shaft of an anterior dorsal rib most,
I provisionally recognize it as such a structure.  The specimen is of course
indeterminate, and not identifiable past Theropoda (assuming it's a theropod
rib, it could be sauropod or ornithopod for all I know).

Ceratosaurus meriani
= Megalosaurus meriani Greppin 1870
= Labrosaurus meriani Janensch 1920
= Antrodemus meriani Steel 1970
= Allosaurus meriani Olshevsky 1978
Late Tithonian-Early Berriasian, Late Jurassic-Early Cretaceous
Virgulla Beds, Switzerland
Holotype- (MH 350) premaxillary tooth (39 mm)
Comments- This tooth is nearly identical to the first premaxillary tooth of
Ceratosaurus dentisulcatus, and differs from other theropods in having
ligual ridges.  Chure is cautious in only saying it probably has affinities
with Ceratosaurus.  As it is of identical size and found in temporally
equivalent beds, I believe it should be called Ceratosaurus meriani, though
the species is indeterminate within the genus Ceratosaurus.

Dandakosaurus Yadagiri 1982
D. indicus Yadagiri 1982
Early Jurassic
Kota Formation, India
Holotype- (GSI 1/54Y/76) proximal pubis
Paratypes- (GSI coll.) teeth, dorsal vertebra (160 mm), proximal caudal
vertebra (150 mm), proximal ischium
Comments- Chure states the holotype is nondiagnostic and makes the taxon a
nomen dubium, though he says examination of the vertebrae might help
determine its affinities.
The illustrated tooth is typical of most theropods in being laterally
compressed, recurved and having fine serrations.  The tooth seems more
compressed (30% of FABL) than most theropods (eg. Liliensternus? airelensis,
Gojirasaurus, Dilophosaurus, Magnosaurus, Torvosaurus, Szechuanosaurus),
though Liliensternus liliensterni is similar in this regard.  Though
interdental variation could be a factor here, as posterior teeth are known
to be more laterally compressed in theropods, Dilophosaurus and Magnosaurus
never reach a Dandakosaurus level of compression anywhere in the tooth row.
The dorsal vertebra is opisthocoelous and lacks a pleurocoel.
Opisthocoelous dorsals are only known in non-coelurosaurian tetanurines (the
exception is Mononykinae), suggesting Dandakosaurus is a member of this
clade.  The absence of a pleurocoel is of little use without positional
data.
The caudal vertebra is amphicoelous, with "two lateral cavities on either
side" and a keeled ventral surface.  The first character is common in
theropods.  The second is only known in Carcharodontosaurus,
Acrocanthosaurus, Patagonykus, "Elaphrosaurus" gautieri, Nomingia,
caenagnathoids and Achillobator.  Few theropods are reported to have ventral
keels on their proximal caudals, including "Creosaurus" potens, Sinraptor
dongi, Bagaraatan, Inosaurus, "Elaphrosaurus" gautieri, alvarezsaurids and
SQU-2-7, an isolated caudal from the Cretaceous of Oman.
The holotype proximal pubis has several odd characters.  There is no
obturator fenestra, just an open obturator notch, as in Elaphrosaurus,
Eustreptospondylus, Suchomimus, allosaurids, carcharodontosaurids and
coelurosaurs.  The proximal border forms a smooth convex arch from the ilial
contact to the ischia contact, with no distinct peduncles.  This is
approached in some coelophysoids (eg. Coelophysis longicollis referred
specimen) and Archaeornithomimus, but is most similar to Patagonykus,
Unenlagia and Achillobator.  Proximally, the pubis flares sharply anteriorly
to form an acute anteroproximal corner.  This is similar to the situation in
Coelurus, tyrannosaurids, Caudipteryx, Nanshiungosaurus and alvarezsaurids.
It seems to only be developed in forms with somewhat mesopubic pelvic
orientations.  Additional evidence for mesopuby in Dandakosaurus may come
from the possible pubic peduncle of the ischium, which fits the ischial
surface of the pubis to form an angle of about 35 degrees between the bones.
Only the proximal portion of the ischium is preserved.  There is a concave
anterior margin which matches up with the ischial peduncular area of the
pubis.  No other area is appropriate for the pubic contact.  If the concave
area were the acetabulum, the pubis and ischium would be subparallel.
Articlating the pubis more dorsoposteriorly on the ischium leaves no
acetabular area leaves nowhere for an acetabular surface or ischial
peduncle.  The acetabular surface is small and sharply concave, so may be
broken and continuous with the ischial peduncle in life.  The posterior
surface is gently concave, while the anterior surface is broken.
The phylogenetic relationships of this taxon are uncertain, especially
considering how advanced it seems for its age.  However, I have heard that
the Kota Formation could extend into the Middle or Late Jurassic.  It seems
to be a tetanurine based on the opisthocoelous dorsal centra.  The only
coelurosaurs with such centra are mononykines, which have procoelous caudal
centra, though they have ventrally keeled caudals, an open pubic obturator
notch, and an anteroproximally flared pubis.  In addition, Patagonykus
shares the presence of lateral fossae in the caudal centra and a smoothly
convex proximal pubic border.  Unfortunately, Patagonykus and Alvarezsaurus
show the procoelous caudals evolved before the opisthocoelous dorsals in
alvarezsaurids.  Also, the dental morphology is less derived than one would
expect for even a basal alvarezsaurid.  Thus, Dandakosaurus is unlikely to
be alvarezsaurid.  The caudal fossae, open pubic obturator notch, and
mesopubic pelvis are similar to Achillobator, but deinonychosaurs lack
opisthocoelous dorsals and have thicker teeth.  Thus, Dandakosaurus
represents a mesopubic tetanurine of uncertain affinities, probably not
coelurosaurian.  It is one of the earliest tetanurines, along with
"Zupaysaurus" and "Saltriosaurus".  Previous suggestions regarding
abelisauroid affinity (
http://www.cmnh.org/dinoarch/1997May/msg00840.html )were based on maxillary
characters unknown in the specimen.  Where Aravind got his data is unknown,
but it can be considered irrelevant now that the known material has been
determined.  Note the open obturator notch, anteroproximal flaring and
smoothly convex proximal surface distinguish Dandakosaurus from all other
comparable theropods except Patagonykus.  It differs from Patagonykus in
having a more proximally placed obturator notch.  Because of these
characters, the taxon is diagnostic and not a nomen dubium, even if only the
holotype is included.

"Allosaurus" tendagurensis Janensch 1925
= Antrodemus tendagurensis Huene 1932
Kimmeridgian, Late Jurassic
Tendaguru Formation, Tanzania
Holotype- (HM 67) partial tibia (~910 mm)
Diagnosis- combination of- cnemial crest not strongly curved laterally; no
proximal groove on posterior tibial surface separating lateral and medial
condyles; fibular crest distally placed.
Comments- Chure notes this specimen resembles abelisaurids because it lacks
a strongly curved cnemial crest or incisura tibialis and has no posterior
groove between the lateral and medial condyles.  However, he states the
astragalus is not fused to it and the fibular crest is more distally placed
than abelisaurids.  The lack of a well developed incisura tibialis indicates
it is not tetanurine, but the distally placed fibular crest is a tetanurine
synapomorphy.  The lack of fusion is inconsequential, as ceratosaurs like
Elaphrosaurus and Quilmesaurus exhibit the condition.  The only other taxon
I know of with this combination of characters is an unnamed form from the
Cenomanian Baharija Formation of Egypt and the Kem Kem Beds of Morocco.  It
is known from four tibiae described as Elaphrosaurus sp. by Stromer (1934)
and Lavocat (1954).  Perhaps "Allosaurus" tendangurensis is related and
these form a very basal lineage of tetanurines.  However, the Cenomanian
tibiae differ from "A." tendangurensis in a number of ways, including the
shorter anteroposterior length and presence of a posterior groove separating
the condyles.  Note that this taxon can be distinguished from other
comparable taxa, so is not a nomen dubium, contra Chure.

Chilantaisaurus? sibiricus (Riabinin 1914) Molnar, Kurzanov and Dong 1990
= Allosaurus sibiricus Riabinin 1914
= Antrodemus sibiricus Steel 1970
Berraisian-Hauterivian, Early Cretaceous
Turginskaya Svita, Russia
Holotype- distal metatarsal IV (~295 mm)
Comments- The specimen has never been illustrated, and was described
extremely briefly by Riabinin.  It wasn't even identified as a metatarsal
until Huene (1926).  Though Molnar et al. assigned it to Chilantaisaurus
based on the resemblence to C. tashuikensis, the distal fourth metatarsal of
the latter is very similar to some other genera like Sinraptor.  Though C.
tashuikensis can be distinguished by a few proportional differences from the
latter genus, I am wary of the generic assignment of C? sibiricus.

"Allosaurus" "robustus" Chure 2001
Early Aptian, Early Cretaceous
Wonthoggi Formation of Strzelecki Group, Victoria, Australia
Holotype- (NMV Pl50070) (~6 m) astragalus
Comments-  Chure is the first person to publish the name Allosaurus
robustus, previously confined to a museum label.  He goes into depth
regarding the supposed Allosaurus synapomorphies given by Molnar et al.
(1981, 1985).  The fibular facet on the ascending process is found in
Torvosaurus and coelurosaurs too.  The presence of an inflection in the
ascending process' medial margin cannot be determined with certainty.
Similarly, the calcanear notch would be higher up if present and not
confluent with the lower horizontal groove.  The upper horizontal grrove
extending across the base of the ascending process is also found in
Elaphrosaurus, Afrovenator, Poekilopleuron, "Szechuanosaurus" zigongensis,
sinraptorids, Deltadromeus and several other neotheropods.  Chure notes it
differs from Allosaurus in having a more parallel-sided ascending process,
lacking a thickened medial edge on that process, having a vertical groove
running up the caudal face of the ascending process, lacking a circular pit
at the caudal base of the acsending process, having a weaker lower
horizontal groove, having a sharper anterior edge on the upper horizontal
groove, and lacking an extensive cranial depression on the ascending
process.  He refers it to the Avetheropoda.
I agree with Azuma and Currie (2000) that this astragalus is most similar to
Fukuiraptor, which shares the characters listed above.  However, in
Fuikuiraptor, the ascending process extends further medially, has a sharper
ventrolateral corner, and is narrower posteroventrally.  "Allosaurus"
"robustus" seems to be a valid taxon of basal carnosaur.

Chienkosaurus Young 1942
C. ceratosauroides Young 1942
Tithonian, Late Jurassic
Shangshaximiao Formation, Sichuan, China
Holotype- (IVPP V237) premaxillary tooth (37 mm)
Referred- ? (IVPP V190) partial mid caudal centrum
? (IVPP V193) ulna (225 mm)
Comments- Three other teeth included in the holotype are from Hsisosuchus
(Dong et al., 1983).
The tooth is similar to many large theropod teeth in general characters, but
is from the premaxilla as evidenced by the twisted anterior carina and
reduced extent of anterior serrations.  It is currently undiagnostic
compared to contemporaneous taxa like Sinraptor and Yangchuanosaurus.
The referred ulna is most similar to Piatnitzkysaurus, though no ulnae are
currently known for sinraptorids.  It differs from Piatnitzkysaurus in not
being bowed posteriorly, and the olecranon appears broken off.  This is a
more diagnostic specimen than the holotype, but is unfortunately not
definitely referrable to the same individual or taxon.
The referred caudal centrum is from a different locality, so is definitely
not referrable to the holotype individual, but was thought to be "probably
referrable" to Chienkosaurus by Young.  Comparson to Allosaurus suggests it
is from somewhere around caudal twenty-five, but it is indeterminate.

Bahariasaurus Stromer 1934
B. ingens Stromer 1934
Albian?-early Cenomanian, Early Cretaceous?-Late Cretaceous
Baharija Formation, Egypt; Continental Intercalaire, Niger?
Holotype- (HM 1922 X47) (11.9 m) dorsal vertebra (200 mm), dorsal vertebra
(~180 mm), neural arch, rib fragment, sacral vertebra (~135 mm), sacral
vertebra (~120 mm), sacral vertebra (~120 mm), pubes (1.03 m), proximal
ischium (Stromer 1934)
Referred- ? (HM 1911) caudal vertebrae, ischium
(HM 1912 VIII 82) ischia
? (MNNHN coll.) six caudal vertebrae (50-65 mm) (Lapparent 1960)
Comments- Chure finds this taxon to be closely related to tyrannosaurids
based on- nearly perpendicular expansion of glenoid margin of scapula from
blade; amphicoelous (non-opisthocoelous) anterior dorsal centra;
subtriangular obturator process; cranial tubercle on fibula; extremely
narrow scapular blade; longitudinal ridge on lateral surface of ischium.  I
disagree.
First, the scapula (HM 1912 VIII 60) is not part of the holotype, which
lacks pectoral material.  Also, tyrannosaurids do not have an expanded
glenoid margin (Carpenter and Smith, 2001) and their blade is much narrower
proximally, but expands distally (the opposite of this scapula).  The
expanded glenoid margin is present in such widely ranging theropods as
Compsognathus, Sinraptor and Baryonyx (Charig and Milner, 1997).  The last
taxon resembles this scapula very closely, differing only in that the
glenoid is more ventrally projected, the blade is a bit narrower distally
and the glenoid expansion is slightly stronger.  Oddly, that of Suchomimus
differs from these two in having a more prominant acromion, distally
expanded blade, and virtually no glenoid expansion, if the skeletal
reconstruction of Sereno et al. (1998) is accurate.  Perhaps the drawing is
inaccurate, or perhaps it belongs to Spinosaurus.  In any case, I refer this
scapula to the Spinosauridae.
The anterior dorsal vertebra (HM 1912 VIII 62b) is also a referred specimen
and non-opisthocoelous dorsals are also found in ceratosaurs,
"Szechuanosaurus" zigongensis and almost all coelurosaurs.  Tyrannosaurid
dorsals differ in being very short, with less waisted centra, wider neural
spines, narrower hyposphenes and taller neural canals (Maleev, 1974).  The
elongate centrum resembles some basal coelurosaurs (Compsognathus, SMNK 2349
PAL), so this vertebra may belong to Deltadromeus or Bahariasaurus.  Another
possibility is something like Elaphrosaurus, which may work with the
Deltadromeus identification too, if Longrich is right about it being more
basal than usually thought.
The triangular obturator process is not discernable in the holotype, but is
present in the complete juvenile(?) ischium HM 1912 VIII 82 that was
misidentified as a pubis by Stromer (1934).  Though much smaller than the
Bahariasaurus holotype, the morphology is nearly identical, differing only
in the less expanded anteroventral corner of the pubic peduncle.  I thus
feel it is properly referred to Bahariasaurus, and it is interesting that it
differs from the distal ischium of Deltadromeus (pubis in Sereno et al.
1996) only in that the boot is less extensive posteriorly, a probable
juvenile trait.  The obturator process is indeed triangular, but this is
known in nearly all coelurosaurs (except Sinosauropteryx and SMNK 2349 PAL),
not just tyrannosaurids.  Though the morphology of the boot resembles
abelisauroids more than any coelurosaur, the obturator morphology firmy
places it in the latter clade.  The presence of an ischial boot and the
proximal placement of the obturator process excludes Bahariasaurus from the
Maniraptora.  It differs from tyrannosauroids in lacking a proximolateral
scar or proximodorsal process, and having a distal boot and more prominent
ilial peduncle.  The lateral ridge on Bahariasaurus is caused by distortion,
as Chure himself says earlier in the thesis.  Where exactly Bahariasaurus
and Deltadromeus belong in the basal Coelurosauria may be revealed by my in
progress analysis.
The fibula is another referred specimen (HM 1912 VIII 70), this one later
referred to Deltadromeus by Sereno et al. (1996).  I agree as it is nearly
identical.  The anteriorly placed iliofibularis tubercle is primitive for
theropods (Rauhut, 2000), so does not indicate tyrannosauroid affinities in
Deltadromeus either.
Rauhut (1995) referred Bahariasaurus to the Allosauroidea based on the
distally reduced pubic symphysis (also in SMNK 2349 PAL), obturator notch
(plesiomorphic for avetheropods), quadrangular obturator process (not
present in Bahariasaurus), and "shape and height" of the anterior trochanter
(plesiomorphic for coelurosaurs).  He placed it in the Carcharodontosauridae
based on the presence of caudal pleurocoels, but this based on a referred
specimen (HM 1912 VIII 62b) whose relationship to Bahariasaurus is
uncertain.  I therefore reject Rauhut's hypothesis Bahariasaurus is a
carcharodontosaurid or an allosauroid.
Bahariasaurus is considered a nomen dubium by Chure, but it can be
distinguished from all comparable theropods except Deltadromeus, with which
it is probably synonymous.  Sereno et al. distinguished the two by three
characters, two of which were due to his misidentification of the distal
pubis as an ischium.  The other is the narrower ilial peduncle of
Bahariasaurus' ischium.  I doubt the importance of this character,
considering the temporal and stratigraphic proximity of the taxa.

"Chilantaisaurus" maortuensis (Hu 1964)
Albian, Early Cretaceous
Lower Ulanhushi Formation of Dashigou Group, China
Holotype- (IVP AS 2885) (skull ~580 mm) maxilla, quadrates (lost), frontals,
posterior skull, (cervical column ~995 mm) axis (~70 mm, lost), six caudal
vertebrae (90, 70, 75, 85, 95 mm; one lost)
Diagnosis- twelve maxillary teeth; small fused maxillary interdental plates;
no groove for dental artery at base of interdental plates; no lingual bar on
maxilla; thick frontals; subnarial process of frontals forms a deep trough;
supratemporal fossa makes small contribution to frontal; sagittal crest on
frontals and parietals; large cylindrical pneumatic cavity in posterior
nasals; no quadratojugal foramen; helical groove absent in distal quadrate;
paroccipital processes extend below occipital condyle; braincase short and
highly pneumatized; axis short with erect neural spine.
Comments- Hu placed this and Chilantaisaurus tashuikensis in the same genus
based on dental and caudal similarities.  The caudals of Chilantaisaurus are
only doubtfully referred and belong to the proximal part of the series,
while those of "C." maortuensis are more distal.  Also, the tooth of
Chilantaisaurus is similarly doubtfully referred and comparison with the
exposed tooth of "C." maortuensis is not useful.  Because of this, Chure
created a new genus for "Chilantaisaurus" maortuensis.
Chure assigns this taxon to the Arctometatarsalia (sensu Holtz 2000) based
on the proximolateral ischial scar in the supposedly related Labocania, and
to the Tyrannosauroidea based on the highly pneumatized basicranium and
short and deep braincase.  He finds it is most closely related to Labocania
based on the thick frontal, lack of a helical groove and slight invasion of
the frontals by the supratemporal fossa.
While I think a tyrannosauroid relationship is possible, I don't support
tyrannosaurid or troodontid arctometatarsalians.  I've included this taxon
as an OTU in my big coelurosaurian analysis and will see where it comes out
there, though I'm betting on a basal coelurosaurian position, like Rauhut
(2000) produced.

If you would like figures of some of the taxa discussed above, ask me
offlist and I'll see what I can do.

Part 2 will come in a few days....

Mickey Mortimer