This is one of the new genera in the Silk Road volume.
Hudiesaurus Dong 1997
H. sinojapanorum Dong 1997
Etymology- " butterfly lizard found by the Chinese and Japanese", from hudie, Chinese pinyin for butterfly and honoring the members of both China and Japan for participating in the expedition.
Kalazha Formation, Xinjiang, China
Holotype- (IVPP V 11120) (30-34 m) first dorsal vertebra (550 mm)
Referred- (IVPP V 11121-1) (23 m) humerus (1.21 m), radius (790 mm), ulna (740 mm), carpal, metacarpal I (150 mm), phalanx I-1 (70 mm), pedal ungual I (210 mm), metacarpal II (310 mm), phalanx II-1 (60 mm), metacarpal III (330 mm), phalanx II-1 (65 mm), phalanx III-2 (45 mm), metacarpal IV (310 mm), phalanx IV-1 (85 mm), phalanx IV-2 (20 mm), metacarpal V (250 mm), phalanx V-1 (74 mm)
(IVPP V 11121-2) four teeth
Diagnosis- "prepophysis" present (see description); prespinal lamina (also in titanosauroids and diplodocoids); extremely robust humerus (width of distal end 40% of length).
The holotype specimen was adult, based on the fused neurocentral suture, and measured approximately 30-34 meters. The larger estimate is based on comparison between anterior dorsal vertebrae of the holotype of Mamenchisaurus hochuanensis, which is 22 meters long. Note that if Nurosaurus or Abrosaurus are used as comparison, the length is decreased because these genera have five less cervical vertebrae. Perhaps 30 meters would be more accurate in this case. The specimen the forelimb belonged to is estimated to have measured 23 meters, based on the tibiohumeral ratio in Mamenchisaurus? youngi and scaling to M. hochuanensis.
The teeth were found in the same beds as the referred forelimb. They are spatulate and serrated anteriorly and posteriorly. The photo would seem to indicate that two premaxillary teeth were found in a fragment of the premaxilla, although it could simply be sediment. The other two teeth are maxillary teeth.
The holotype dorsal vertebra is the first based on the positions of the parapophyses and diapophyses. It is strongly opisthocoelous (length 390 mm without anterior ball), with large pleurocoels. The neural spine is transversely broad and slightly bifurcated. There is a small median dorsal process between the fork. An autapomorphic 84 mm process projects anteriorly from the neural spine and is termed the prepophysis by Dong. He hypothesizes this process may have been an insertion point for muscles or articulated with the hyposphene. The hyposphene and hypantrum are present. The laminae of the neural arch and spine are said to be very similar to Mamenchisaurus. Differences listed include a V-shaped posterolaterally projecting lamina on the lateral mergin of the neural spine and a well-developed median lamina on the anterior neural spine. The parapophysis lies beneath the pleurocoel and the diapophysis is directed ventrolaterally. The transverse processes are directed laterally, unlike the ventrally directed processes in Mamenchisaurus and Euhelopus.
The forelimb was found 1.1 kilometers from the holotype in the same horizon, so it may not belong to the species.
The humerus is incredibly robust, more so than any other sauropod I know of. The proximal and distal ends are greatly expanded and the proximal end is badly preserved. The radius is slightly bowed and has an expanded proximal end (32% of length). There is a concavity distally for the carpals. The ulna is straight and greatly expanded proximally, though the olecranon process is poorly developed and the proximal edge is not concave. One carpal is present. It is round and articulates with metacarpals I-III. The metacarpals are preserved in articulation, which shows they were columnar but lack long intermetacarpal articulations. Metacarpal I is only 45% of metacarpal III in length and 48% of metacarpal IV in length, while metacarpal III is 42% of the radius. Metacarpal V is 76% of metacarpal III in length. The manual phalangeal formula is reduced to 2-2-2-1-1. There is a large ungual on digit I.
Dong referred this species to the Mamenchisauridae based on the low neural spine and shallow bifurcation. As explained in my post on Mamenchisaurus? youngi, the phylogenetic position of Mamenchisaurus is currently controversial. Both Sereno and Wilson (1998) and Upchurch (1998) believe the genus is not neosauropod. However, they disagree on whether Euhelopus and Shunosaurus are closely related to Mamenchisaurus and Omeisaurus. My analysis of Mamenchisaurus? youngi showed it possessed some neosauropod and macronarian characters, while lacking others, which suggested more research is needed but in my opinion strengthened the case for Euhelopus convergently evolving macronarian-like states.
I provisionally include the teeth and forelimb in the species Hudiesaurus sinojapanorum because the teeth and forelimb exhibit characters that would be expected of a non-neosauropod, while the holotype dorsal vertebra resembles Mamenchisaurus and Euhelopus (both non-neosauropods)more than other sauropods. The presence of multiple euhelopodids in the contemporaneous Shangshaximiao Formation and the earlier Xiashaximiao Formation demands caution in this area however.
Hudiesaurus is more derived than Shunosaurus in Sereno and Wilson's phylogeny based on the opisthocoelous anterior dorsal vertebrae. It lacks the neosauropod character of long intermetacarpal articulations and the macronarian characters metacarpus more than 45% of radius length (42%) and metacarpal I subequal in length to metacarpal IV (48%). It also lacks the somphospondylan character of anterior dorsal neural spines posterodorsally directed. It is excluded from the Diplodocoidea and Titanosauria by the spatulate, broad tooth crowns with serrations. Additional characters that exclude it from titanosaur subgroups include- width of proximal end of radius less than 33% of radial length (32%); no concave surface on proximal ulna; ossified distal carpal present; metacarpal I shorter than metacarpal III (45%); hyposphenes and hypantra present. The presence of a prespinal lamina (presumedly the lamina described as being median in position on the anterior surface of the neural spine) is a derived character only known in titanosauroids and diplodocoids. Based on preliminary analysis then, Hudiesaurus is not a neosauropod, but is more derived than basal sauropods without dorsal pleurocoels or opisthocoelous centra. The only non-neosauropods with bifurcated neural spines are Mamenchisaurus, Abrosaurus, Nurosaurus and Euhelopus (if the latter isn't a titanosaur). Since Mamenchisaurus and Euhelopus have opisthocoelous dorsal centra, perhaps Hudiesaurus is more closely related to them than to Abrosaurus (with platycoelous or amphicoelous centra). The condition in Nurosaurus is unknown. Then again, the more basal Omeisaurus has opisthocoelous centra too, so the situation is complex. In any case, this supports Dong's conclusion of relationship to Mamenchisaurus. I recommend placing Hudiesaurus in the Euhelopodidae (as I provisionally prefer Upchurch's phylogeny; in Sereno and Wilson's phylogeny Hudiesaurus would be a non-neosauropod member of the Omeisaurus + Neosauropod clade, possibly in a Mamenchisauridae), more closely related to Mamenchisaurus and Euhelopus (and probably Abrosaurus and Nurosaurus) than to Shunosaurus, Omeisaurus and Datousaurus.
Those who want scans of the teeth, vertebra and forelimb (2.5 pages) contact me offlist. There are figures accompanying the forelimb, manus and vertebra photos, so details are actually discernable. I wonder what genus I'll do next.....