TYRANNOSAURIAN IMPLOSION [long; part 2 of 2]

[Dinogeorge@aol.com]

Eastern Asia is often considered to be where the family Tyrannosauridae 
originated. There are several reasons for this, among which are: (1) the 
generally more plesiomorphic nature of the tyrannosaurids in Asiatic faunas; 
(2) the presence of Early Cretaceous Asiatic material referable to 
Tyrannosauridae; (3) the general absence of tyrannosaurid material from North 
America prior to the Santonian stage (although Kirkland & Parrish (1995) 
report tyrannosaurian teeth from the Cedar Mountain Formation); and (4) the 
lack of tyrannosaurid material in locales outside eastern Asia and western 
North America. This biogeographical scenario posits an Early Cretaceous 
origin in eastern Asia from as-yet-undiscovered ancestral forms, followed by 
a crossover to western North America, followed by the essentially independent 
evolution of the two populations when the sea inundated the Bering land 
bridge during the Santonian, followed by a possible second crossover as the 
Late Maastrichtian sea level subsidence opened the Bering land bridge.

All my comments on the Asiatic genera are perforce based on published 
literature rather than on examination of actual specimens. I have, however, 
seen several Asiatic tyrannosaurid specimens (casts as well as original 
fossils) in traveling exhibitions of Russian and Chinese dinosaurs, and Tracy 
has photographed and videotaped Asiatic specimens in his travels, so I'm not 
totally in the dark.

The commonest Asiatic tyrannosaurid genus is Tarbosaurus. The idea that 
Tyrannosaurus and Tarbosaurus are congeneric is rejected. Tarbosaurus differs 
from Tyrannosaurus in having the "no jugal spread" and "visible condyle" 
features, and it has more than 14 dentary teeth (usually 15â17); but like 
Tyrannosaurus it is "hornless." Also like Tyrannosaurus, but unlike 
Albertosaurus, it has a suborbital process on the postorbital; but unlike 
Tyrannosaurus, this process is not pendant. The suborbital process is an 
adult-onset feature in tyrannosaurids (juveniles lack them: Carpenter, 1992; 
Carr, 1999) and in adults it evidently helped to support the eyeball, which 
exhibits negative allometry relative to skull height during growth. Almost 
all Tarbosaurus skeletons are significantly smaller than almost all 
Tyrannosaurus skeletons. The teeth of Tarbosaurus are somewhat smaller and 
mediolaterally narrower than the teeth of North American tyrannosaurids of 
the same body size, but the taxonomic reliability of this feature is has not 
yet been determined. I presume that, on the whole, tooth taxonomy of the 
Asiatic tyrannosaurids is as ineffectual as that of the North American forms.

Tarbosaurus occurs throughout Campanian and Maastrichtian deposits of 
Mongolia, China, and central Asia. It is, however, not as yet well studied as 
its North American counterparts. Tyrannosaurid teeth are common in these 
localities, and there is no reason not to refer them all provisionally to 
Tarbosaurus sp. There appear to be at least two species: the type species 
Tarbosaurus efremovi, which had an adult length of 8â10 meters, and the 
larger species Tarbosaurus bataar, which had an adult length of about 14 
meters (a length estimated from the size of the type skull; a reasonably 
complete skeleton of this size has not yet been documented for any Asiatic 
tyrannosaurid). Few characters that cannot be construed as age-related 
differentiate these two species, so they are conflated by many workers, with 
Tarbosaurus efremovi considered a subadult Tarbosaurus bataar (this species 
name has priority) or Tarbosaurus bataar considered a giant, overgrown adult 
Tarbosaurus efremovi. But until someone can provide a convincing argument for 
why most known Tarbosaurus specimens are "subadult" and only one specimen is 
"adult" (chance and preservational bias alone don't cut it: exactly the 
opposite situation pertains with respect to all North American tyrannosaurid 
species in a variety of facies), it is best to retain their specific 
distinction. In 1995, I placed Tarbosaurus bataar in its own genus, 
Jenghizkhan, since there seemed to be as many minor characters separating 
Tarbosaurus from Jenghizkhan as separated Albertosaurus from Daspletosaurus; 
but since the latter two genera are here regarded as synonyms, the same 
consideration must be afforded Jenghizkhan with respect to Tarbosaurus.

Tarbosaurus efremovi occurs in several Asiatic horizons, but Tarbosaurus 
bataar is known only from the Nemegt Formation, where it was sympatric with 
Tarbosaurus efremovi, and from the Quiba Formation of Henan Province, China, 
where large tyrannosaurid teeth likely referable to this species were 
described as the species Tyrannosaurus luanchuanensis. The Nemegt is dated 
late Campanian to early Maastrichtian, while the Quiba is dated Campanian, 
but these dates are not secure, and there is no reason that Tarbosaurus 
bataar could not occur in the Quiba as well as the Nemegt. Likewise for the 
other Chinese tooth species originally described in the genus Tyrannosaurus: 
There is no reason not to refer them to Tarbosaurus as species indeterminate 
or even to synonymize them with Tarbosaurus efremovi.

The earliest-named tyrannosaurid species from Asia is Osborn's (1924) tooth 
species Prodeinodon mongoliensis. It is Early Cretaceous (Aptian), and the 
holotype tooth resembles those referred to Aublysodon; but the paratype tooth 
is rather nondescript and could belong to almost any carnosaurian theropod. 
Given the dubious taxonomical value of isolated tyrannosaurid teeth, however, 
this genus and species should be isolated as nomina dubia at the base of 
Tyrannosauridae, particularly since they would have priority over all later, 
long-established Asiatic tyrannosaurid names.


The next-earliest named Asiatic tyrannosaurid species is Riabinin's (1930) 
Albertosaurus periculosus, for a medium-size tyrannosaurid tooth found with 
the skeletal remains of the ?lambeosaurid Mandschurosaurus amurensis. 
Godefroit and Bolotsky (1999) date this find palynologically as late 
Maastrichtian. Albertosaurus periculosus has priority over both Tarbosaurus 
efremovi and Tarbosaurus bataar, but since we cannot distinguish these taxa 
by teeth alone, Albertosaurus periculosus must become a nomen dubium. There 
is no justification for maintaining that species in the North American genus 
Albertosaurus, so I referred it in 1995 to the Asiatic genus Tarbosaurus as 
Tarbosaurus periculosus.

The genus Maleevosaurus is noted by Carr (1999) as based on a probable 
juvenile Tarbosaurus, and I would agree with this assessment. Since 
Tarbosaurus efremovi is by far the commoner species, Maleevosaurus novojilovi 
would most likely be a juvenile Tarbosaurus efremovi. Another likely juvenile 
Tarbosaurus efremovi is Shanshanosaurus huoyanshanensis, from the Subash 
Formation (Campanian to Maastrichtian) of Xinjiang, China. I previously 
(1995) referred it to "Aublysodontinae" on the basis of its premaxillary 
teeth and renamed that subfamily Shanshanosaurinae (because Aublysodon is a 
nomen dubium); but since such teeth generally characterize juvenile 
tyrannosaurids, and there is nothing else in the skeletal remains of 
Shanshanosaurus to preclude this identification, it is best to synonymize the 
genus with Tarbosaurus. Finally, the genus Chingkankousaurus from the 
Wangshih Formation (Maastrichtian?) of Shandong, China, based on a slender 
tyrannosaurid scapula and other bone fragments, is almost certainly referable 
to Tarbosaurus as probably Tarbosaurus efremovi. Maleev's other tyrannosaurid 
"species" from Mongolia, Gorgosaurus lancinator, might just be a subadult 
Tarbosaurus bataar; it seems to have more robust cranial elements than 
similar-size Tarbosaurus efremovi individuals. Rozhdestvensky (1965) first 
collapsed the species Gorgosaurus novojilovi, Gorgosaurus lancinator, 
Tarbosaurus efremovi, and Tyrannosaurus bataar as representing growth stages, 
of the single species Tarbosaurus bataar.

The genus Alioramus is generically distinguished from all other 
tyrannosaurids by its relatively elongate skull, relatively small and 
numerous teeth, and ornate nasal ornamentation, unknown in any other theropod 
species. I would retain it as a valid genus of Asiatic tyrannosaurid, type 
species Alioramus remotus. It might be a later sister group to the following 
genus, Alectrosaurus, although their nasal bones are quite different.

Alectrosaurus olseni Gilmore, 1933 occurs in earlier Asiatic deposits than 
Tarbosaurus, and is distinguished from the two species of that genus by 
having smooth (as opposed to rugose) nasals (in a referred skull), relatively 
longer and slenderer hind limb elements, and relatively shorter pedal 
elements (but these features might well be age-related: all known 
Alectrosaurus material does seem to have features considered subadult in 
North American tyrannosaurids). Its referred premaxillary teeth are 
Aublysodon-like, which is why I previously classified it in 
Shanshanosaurinae, but as noted throughout the above, this character is 
age-related and has no demonstrable taxonomic significance. Other material 
referred to this genus seems to be tyrannosaurid but is otherwise 
indeterminate. Kirkland & Parrish (1995) report Alectrosaurus teeth from the 
Cedar Mountain Formation (Early Cretaceous), but these are simply evidence of 
the initial invasion of North America by Asiatic tyrannosaurids. I 
provisionally retain Alectrosaurus as a separate genus, particularly because 
sinking it into synonymy with Tarbosaurus as a third species would eliminate 
the latter, much better-known, name as the junior synonym; but otherwise 
there seems to be little generic distinction between the two taxa. 
Alectrosaurus was once synonymized with Albertosaurus by Paul (1988), but 
since no lacrimal horn is known for Alectrosaurus, that synonymy seems 
premature.

The genus Tonouchisaurus is as yet undescribed, but Barsbold (pers. comm.) 
notes that the material includes associated hind and didactyl forelimb 
elements that indicate it was a small tyrannosaurid. I suspect this may be 
yet another juvenile tyrannosaurid, but until the description appears there 
is little more to say.

Siamotyrannus isanensis is based on a partial postcranial skeleton of a 
theropod estimated to be 6.5 meters long from the Early Cretaceous Sao Khua 
Formation of Khon Kaen Province, Thailand. The pelvic bones suggest 
tyrannosaurid affinities, as the authors noted, and I have no problem 
accepting this species as a basal tyrannosaurid or, more appropriately, as 
basal to Tyrannosauridae but not necessarily referable to the family. A good 
skull would settle the question of its exact relationship within 
Tyrannosauria. The skeleton provides evidence that tyrannosaurians ranged 
from Siberia through Mongolia, China, and Indochina during the later stages 
of the Cretaceous Period.

Itemirus medullaris from the Turonian of Uzbekistan is founded on a small but 
probably adult partial braincase that may belong to a tyrannosaurian of some 
kind. What there is of it seems distinct from any of the known tyrannosaurid 
braincases, however, so I would hesitate to refer it to Tyrannosauridae. 
Perhaps its family, Itemiridae, belongs in Tyrannosauria, and perhaps not.

Stokesosaurus clevelandi from the Late Jurassic Morrison Formation has been 
described in several papers as a candidate for tyrannosaurian ancestry. The 
material is pretty meager but nonetheless suggestive. The holotype ilium has 
a tyrannosaurian shape and a prominent supraacetabular ridge, but a referred 
premaxilla does not have characteristic tyrannosaurian teeth. Dan Chure 
(pers. comm.) has found Aublysodon-like teeth in the Morrison, so perhaps 
these are actually Stokesosaurus teeth and the premaxilla is incorrectly 
referred (assuming Stokesosaurus is indeed tyrannosaurian).

Tyrannosaurids emerge from this analysis as a small, very conservative, and 
temporally and biogeographically restricted clade of predatory dinosaurs. 
Only four species in two genera are reliably known from the Late Cretaceous 
of North America, and four additional species in three additional genera are 
reasonably well known from the Late Cretaceous of eastern and central Asia. 
Although restricted to North America and eastern and central Asia, 
tyrannosaurids appear to have been widespread throughout these domains. The 
apparent lack of any non-tyrannosaurid theropods in post-Santonian western 
North America suggests that tyrannosaurids there passed through several 
different predator niches as they grew from hatchlings to large adults. Asia 
had a more diverse theropod fauna, including large non-tyrannosaurid 
theropods, so the ecological role of tyrannosaurids there is less easily 
understood.

This analysis benefited greatly from Tom Holtz's recently opened websites on 
Tyrannosauroidea (his Tyrannosauroidea is essentially the same as my 
Tyrannosauria) and its subgroups, whose Web address is:

http://phylogeny.arizona.edu/tree/eukaryotes/animals/chordata/dinosauria/tyran

nosauroidea/tyrannosauroidea.html

This is part of the Tree of Life megawebsite. Besides displaying a 
tyrannosaurian phylogeny and listing numerous tyrannosaurian characters, the 
website features a long bibliography. Accounts of subclades of 
Tyrannosauroidea may be accessed from this website, as can the Tree of Life 
itself.