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

What is Lukousaurus?

In my continuing efforts to discover the phylogenetic affinities of
difficult theropod taxa, I've examined the figures of Lukousaurus yini and
compared it to possible relatives.  This species seems to have been ignored
by most authors since it's 1948 description and dismissed as probably not
dinosaurian, but still is included in many dinosaur lists and texts
(Dinosauria, Glut's encyclopedia, etc.).  This is a much more difficult task
than Saltopus however, due to the lack of synapomorphies relating to the
snout in modern archosaur phylogenies.
First of all, Lukousaurus is obviously an archosauriform, based on the
presence of an antorbital fenestra.  There are also several archosauriform
taxa that Lukousaurus is obviously not a member of.  These include
proterosuchids, proterochampsids, parasuchians, aetosaurs, pterosaurs,
sauropodomorphs and ornithischians.  What's left are a mix of
erythrosuchids, crurotarsans and theropods.  Lukousaurus almost certainly
belongs to one of these clades.
Erythrosuchids are only known from the early and middle Triassic, unlike
Lukousaurus, which is from the late Triassic.  Moreover, no particular
synapomorphies could be found in Lukousaurus that would indicate
erythrosuchid relationships.
Of the Crurotarsi, Lukousaurus resembles paracrocodylomorphs (postosuchids
and crocodylomorphs) and prestosuchids the most.  It shares several general
features with these, such as a tall orbit, deep maxilla beneath antorbital
fossa and short triangular antorbital fenestra.  More importantly,
Lukousaurus shares the presence of five premaxillary teeth with
sphenosuchians, a character otherwise only known in allosaurids, some
prosauropods and hypsilophodonts (if anyone knows of other crurotarsans with
five premaxillary teeth, feel free to comment).  Also, Lukousaurus shares a
posteriorly placed naris with the sphenosuchian Pseudohesperosuchus,
although the small, triangular antorbital fossa resembles Dibothrosuchus and
Saltoposuchus more.  Lukousaurus's skull is a bit higher than most
sphenosuchians and the orbit seems taller, but there's nothing stopping it
from belonging to this clade/grade of crocodylomorphs and the premaxillary
tooth count would suggest that it might.
When compared to theropods, Lukousaurus can be shown not to be a
coelophysoid (absence of subnarial notch) or a tetanuraen (tooth row extends
beneath orbit).  This leaves Eoraptor, herrerasaurids, ceratosaurids and
abelisaurs.  Two characters are shared with the
Ceratosaurus-Abelisauria-Tetanurae clade:
- dorsoventrally elongate orbit
- lacrimal horn
And one character is shared with Herrerasaurus and abelisaurids:
- deep maxilla beneath antobital fossa (antorbital fossa not prominent
ventral to fenestra)
The orbital proportion is not size-related as Herrerasaurus, for example has
a circular orbit and a skull much larger than Lukousaurus.  Also, the
maxillary proportion is not size-related, as tyrannosaurs and some other
tetanuraens lack this feature and are much larger.  These characters cannot
be trusted too much however, since the first and third are also present in
some crurotarsans and erythrosuchids and the second may be due to crushing
as the skull roof also has a large prominence(?) over the orbit that doesn't
seem natural and may have been displaced from it's original position.
Thus, Lukousaurus resembles abelisaurids most out of known theropods.
Perhaps it should be compared to Dandakosaurus indicus, which is supposedly
an abelisaur from the Early Jurassic of India and also lacks an extensive
antorbital fossa.  Unfortunately, I lack a proper reference to this species,
so I am unable to further compare the two.  If someone could get me the
reference to D. indicus, I'd appreciate it.
Out of known archosaurs, Lukousaurus resembles abelisaurids and
sphenosuchians the most.  A deep maxilla beneath the antorbital fossa
suggest referral to either group (although it is present in other
archosauromorphs, these lack additional synapomorphies).  The tall orbit and
possible lacrimal horn suggest abelisaurid relations, while the caudally
placed naris and five premaxillary teeth suggest sphenosuchian relations.
This makes relationships difficult to determine, as only one sphenosuchian
has a caudally placed naris and the lacrimal horn may not be natural.  The
general skull shape resembles an abelisaur more, but the antorbital fenestra
is more similar to sphenosuchians.  Although initially sphenosuchians seem
more likely to exist in the Late Triassic Lufeng (as they already do, in the
form of Diobothrosuchus), the presence of Dandakosaurus as a possible Early
Jurassic Indian abelisaur makes this equivocal.  If I had to bet, I'd place
my money on it being a sphenosuchian, but I'll wait until Dandakosaurus is
described before making a final decision.

Mickey Mortimer