Novitaotesamerican MUSEUM PUBLISHED by the AMERICAN MUSEUM of NATURAL HISTORY CENTRAL PARK WEST at 79TH STREET, NEW YORK, N.Y
Total Page:16
File Type:pdf, Size:1020Kb
NovitaotesAMERICAN MUSEUM PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number 3115, 39 pp., 13 figures December 28, 1994 The Skull of Erlicosaurus andrewsi, a Late Cretaceus "Segnosaur" (Theropoda: Therizinosauridae) from Mongolial JAMES M. CLARK,2 PERLE ALTANGEREL,3 AND MARK A. NORELL4 ABSTRACT The holotype of Erlicosaurus andrewsi Perle, drewsi includes the only well-preserved skull pres- 1981, from Late Cretaceous strata at Baynshin ently known for any taxon of Therizinosauridae. Tsav, South Gobi Aimak, Mongolia, includes an We present a detailed redescription of this skull, exceptionally well-preserved, nearly complete skull emphasizing those features bearing evidence ofits and mandible. Erlicosaurus andrewsi previously affinities with particular dinosaur groups. Cranial was considered a member ofSegnosauria Barsbold features of Erlicosaurus andrewsi are consistent and Perle, 1980, but recent work confirms that with the hypothesis that the Therizinosauridae are Therizinosaurus cheloniformis Maleev, 1954, is a members ofTheropoda, and ofthe theropod taxon member of this group; because Therizinosauridae Maniraptora (as modified to include Ornithomi- Maleev, 1954, has priority, it is the valid name mosauria) in particular. The skull of Erlicosaurus for the group. The holotype of Erlicosaurus an- andrewsi, furthermore, shares several derived fea- ' This is contribution number 6 of the Mongolian Academy of Sciences-American Museum of Natural History Paleontological Project. 2 Frick Postdoctoral Fellow, Department ofVertebrate Paleontology, American Museum ofNatural History; current address: Department of Biological Sciences, George Washington University, Washington, D.C. 20052. 3 Mongolian Museum of Natural History, Ulaan Baatar 46, Zaluchudiin Gudamz-l, Mongolia. 4Associate Curator, Department of Vertebrate Paleontology, American Museum of Natural History. Copyright ©) American Museum of Natural History 1994 ISSN 0003-0082 / Price $4.20 2 AMERICAN MUSEUM NOVITATES NO. 3115 tures with Omithomimosauria, Oviraptorosauria, Ornithischia and Prosauropoda-and find them to and Troodontidae. We review previous hypothe- be based upon weak evidence. ses of relationship for Therizinosauridae-with INTRODUCTION Beginning with the first publications on Enigmosauridae Barsbold and Perle, in Bars- segnosaurs (Segnosauridae Perle, 1979; Seg- bold, 1983), the taxonomic validity of the nosauria Barsbold and Perle, 1980), the un- latter two taxa is now ambiguous (e.g., the usual mixture of features in these fossils has name could be applied validly at the subfa- baffled paleontologists investigating their re- milial level). Russell and Dong (1993) pro- lationships. Most striking is the combination posed that the taxa formerly placed in the of an opisthopubic pelvis similar to that of Segnosauria now be placed with Therizino- dromaeosaurid and avialian theropods with saurus cheloniformis in the Therizinosauri- a hind foot that is not modified into the tri- dae, and that Alxasaurus elesitaiensis be dactyl pes typical of all theropods, and with placed with that family in the Therizinosau- cheeks and a beak as in ornithischian dino- roidea. (Russell and Dong also proposed the saurs. monotypic family Alxasauridae, but we find Hypotheses of relationships for Segnosau- such redundant taxa to be unnecessary; see ria, consequently, have involved both major De Queiroz and Gauthier, 1990.) We have clades of dinosaurs and both clades of Saur- not yet examined the specimens ofAlxasau- ischia. In his description of the first known rus elesitaiensis, but the character evidence segnosaur material, Perle (1979) suggested mustered by Russell and Dong, summarized that their affinities are with Theropoda, as in the following classification, is persuasive: did Barsbold (1979, 1983). Furthermore, Perle (1982) considered the equally enigmatic Therizinosauroidea (Maleev, 1954) theropod Therizinosaurus cheloniformis Ma- Aixasaurus elesitaiensis Russell and Dong, leev, 1954, to be related to the segnosaurs. 1993 Paul (1984) hypothesized segnosaurs to be Therizinosauridae Maleev, 1954 "relics [sic] ofthe prosauropod-ornithischian Erlicosaurus andrewsi Perle, 1981 transition," whereas relationships with Sau- Segnosaurus galbinensis Perle, 1979 ropodomorpha were posited by Gauthier Therizinosaurus cheloniformis Maleev, (1986) and Sereno (1989). These diverse hy- 1954 potheses are reflected in the decision by the Enigmosaurus mongoliensis Barsbold authors of the most recent review of segno- and Perle, 1983 saurs to consider them "Saurischia sedis mu- Nanshiungosaurus brevispinus Dong, tabilis" (Barsbold and Maryanska, 1990: 415). 1979 An important contribution to resolving the Chilantaisaurus zheziangensis Dong, problem ofsegnosaur relationships is the dis- 1979 covery ofa new, primitive segnosaur relative, Fossil remains of Therizinosauroidea are Alxasaurus elesitaiensis Russell and Dong, rare, and only a single skull is described for 1993. These authors hypothesized that A. any ofthe six named taxa (Barsbold and Mar- elesitaiensis is the sister taxon of a group yanska, 1990; Russell and Dong, 1993). For- comprising Therizinosaurus cheloniformis tunately this skull, part of the holotype of and Segnosauria, and that this group together Erlicosaurus andrewsi, is exquisitely pre- is the sister group of Troodontidae and Ovi- served and nearly complete. Although it was raptorosauria. Because the precise relation- described in detail and figured by Perle (198 1) ships between Therizinosaurus cheloniformis and Barsbold (1983), the unavailability of and the four genera ofsegnosaurs remain un- translations of these papers has limited their clear, and because the family Therizinosauri- accessibility to paleontologists unfamiliar dae Maleev, 1954, has priority over the taxa with Russian. Segnosauridae and Segnosauria (as well as An opportunity to prepare a revised de- 1 994 CLARK ET AL.: SKULL OF ERLICOSA UR US ANDRE WSI 3 scription ofthe skull ofErlicosaurus andrewsi likosaurus[sic] sp." from Iren Dabasu in In- was afforded by the visit ofPerle to the Amer- ner Mongolia, and they considered the Iren ican Museum of Natural History from Oc- Dabasu fauna coeval with that of the Bayn- tober 1992 to January 1993, as part of the shin Tsav site. A frontal from the Judith Riv- Mongolian Academy of Sciences-American er Formation of Canada first figured and de- Museum of Natural History Paleontological scribed by Sues (1978), who did not assign it Project. The skull and mandible were pho- to any particular taxon in the absence of di- tographed and videotaped, and Scanning agnostic features, was identified by Currie Electron Micrographs were taken of a man- (1987a) as "cf. Erlikosaurus[sic]." However, dibular tooth. Barsbold and Maryanska (1990) pointed out Our study of this specimen and compari- that the frontal bone ofErlicosaurus andrewsi son with other dinosaurs supports the rela- is not diagnostic. Jerzykiewicz and Russell tionship with theropods originally suggested (1991) correlated the Baynshirenskaya Svita by Perle (1979). Furthermore, several fea- faunal assemblage with the Turonian to Cam- tures of the skull corroborate the affinities of panian (Late Cretaceous) marine inverte- Therizinosauroidae within Maniraptora hy- brate stages, but there is little stratigraphic pothesized by Russell and Dong (1993). We constraint on this correlation. defer a more detailed treatment of the phy- DIAGNosIs: Because the skull described here logenetic relationships ofTherizinosauroidea is the only one known for any therizinosau- to a later date pending examination of a roid, it is unclear at what taxonomic level its broader diversity of Maniraptora. apomorphic features are diagnostic. Further- more, without a precise hypothesis of rela- Abbreviations tionship it is not possible to determine which AMNH American Museum of Natural History, characters shared with other Maniraptora are Vertebrate Paleontology. homologies and which are homoplasies di- MAE Collections of the joint Mongolian Acade- agnostic of Erlicosaurus andrewsi or a more my of Sciences-American Museum of of Natural History Paleontological Expe- inclusive clade therizinosauroids. ditions. These specimens will eventually In comparison with other nonavian coe- be deposited in the collections of the lurosaurian and maniraptoran theropods, the Mongolian Academy of Sciences. skull of Erlicosaurus andrewsi is distin- PST Paleontological and Stratigraphic Section of guished by: edentulous premaxilla with sharp, the Geological Institute, Mongolian vertical ventrolateral edge, extremely elon- Academy of Sciences. gate nares due to regression of the maxilla, antorbital fossa with well-developed over- SYSTEMATIC PALEONTOLOGY hanging lip, medial wall of antorbital fossa Erlicosaurus andrewsi Perle, 1981 extensive and imperforate, maxilla with me- dially inset dentition and few nutrient foram- HOLOTYPE: PST 100/111, a nearly com- ina on posterior part of facial process, pos- plete skull and mandible, several fragmentary terior process of jugal covering anterior cervical vertebrae, an articulated right pes, surface of quadratojugal, vomer extremely missing the proximal ends oftwo of the four elongate and extending posteriorly to meet metatarsals, and the left humerus. cultriform process, parabasisphenoid with TYPE LOCALITY, GEoLoGICAL OCCURRENCE, extremely large pneumatic spaces, external AND AGE: The holotype of Erlicosaurus an- auditory meatus restricted ventrally by lat- drewsi is