ARTICLE SKELETAL MORPHOLOGY AND LOCOMOTOR BEHAVIOR OF PSEUDOTOMUS EUGENEI (RODENTIA, PARAMYINAE) FROM THE UINTA FORMATION, UTAH RACHEL H. DUNN * and D. TAB RASMUSSEN Department of Anthropology, Washington University, One Brookings Drive, Campus Box 1114, Saint Louis, Missouri 63130, U.S.A.; rhdunn@artsci.wustl.edu, dtrasmus@artsci.wustl.edu ABSTRACT—A new skeleton of Pseudotomus eugenei is described and compared to extant large bodied rodents and those fossil manitshines represented by postcranial material (Pseudotomus robustus, Pseudotomus petersoni, and Man- itsha tanka). The postcranial skeleton of P. eugenei is unspecialized in terms of individual joint function as well as in overall proportions. The forelimb is heavily muscled with a generalized shoulder but with some distinctive features of the elbow and manus. These include an extremely broad distal humerus, long olecranon process, short proximal and middle phalanges and long, uncompressed ungual phalanges. The hind limb of P. eugenei is also generalized, but departs from those of earlier manitshines in the deeper patellar groove and femoral condyles, and in the size reduction of the peripheral metatarsals, suggesting a stronger commitment to terrestrial life. A quantitative analysis of limb proportions indicates that P. eugenei and the other manitshines generally resemble terrestrial rather than arboreal rodents. This is especially evident in the shape of the radial head, and in the shape and relatively greater lengths of the olecranon process, phalanges and metapodials. These features of the forelimb together with the long and uncompressed ungual phalanges of the manus suggest that when P. eugenei diverges from other manitshines, it is in a semi-fossorial direction. Here it is reconstructed as a generalized terrestrial rodent with fossorial tendencies. INTRODUCTION Paramyines are the most abundant rodents in the Eocene of North America. They are first found in the Clarkforkian (Wood, 1962; Rose and Chinnery, 2004) and make their final appearance in the Orellan of South Dakota (Simpson, 1941). The latest sur- viving paramyines belong to the Manitshini, which are charac- terized by their large size and simple, robust dentitions (Simp- son, 1941; Wood, 1962; Korth, 1985). The earliest manitshines are Bridgerian in age and belong to the genus Pseudotomus (Matthew, 1910; Simpson, 1941; Wood, 1962; Korth, 1994). Dur- ing the Bridgerian, manitshines were not particularly diverse, being represented by only two well-defined species, P. horribilis and P. robustus, from the Bridger Basin, Wyoming (Wood, 1962; Korth, 1985). Manitshines reach their peak species diversity dur- ing the Uintan (Korth, 1994). During that time they were also geographically widespread, with records from California (P. cali- fornicus and P. littoralis), Utah (P. petersoni and P. eugenei) and Texas (P. johanniculi; Wood, 1962; Korth, 1985, 1994). After the late Uintan, manitshine fossils are conspicuously absent with the sole exception of Manitsha tanka from the Orellan of South Da- kota, which is known only from the type specimen (Simpson, 1941; Korth, 1994). Although many of the Bridgerian and early Uintan manitshines are represented by both dental and skeletal elements, the late Uintan and Oligocene manitshines are poorly known (Wood, 1962). The locomotor behavior of manitshines has remained unre- solved since the definition of the tribe by Simpson in 1941. In his 1962 monograph, Wood agreed with Matthew (1910) and Simpson (1941) that the older members of the Paramyinae be- longing to the genus Paramys were most likely arboreal, but noted that the morphology of the manitshines is ambiguous. Manitsha tanka is the largest and most derived manitshine; its skull is 50% longer than that of the extant beaver, Castor canadensis (Simpson, 1941). Simpson noted that the forelimb morphology of Manitsha, especially that of the ungual phalanges, would have been equally well suited for digging or climbing, and suggested that it was even possible that Manitsha was semi- aquatic, although there was no unambiguous evidence to support any of these conclusions. The most compelling evidence that Manitsha was terrestrial is its massive size (Simpson, 1941; Wood, 1962). One factor making reconstruction of postcranial behavior in manitshines difficult is that their large size makes the choice of living analogs problematic. Postcranially, paramyines are usu- ally compared to extant sciurids because sciurids are considered to be primitive among extant rodents and because members of this family present a diverse array of behaviors from fossorial forms to arboreal specialists. However, many sciurids are far too small to meaningfully compare with rodents of such a large size as the manitshines. The smallest manitshines are larger than the largest sciurid, Marmota. Most of the large rodents alive today belong to postcranially specialized groups such as the South American caviomorphs and the African hystricomorphs (Simp- son, 1941; Elisamburu and Vizcaı ´no, 2004), which thus presents problems in making comparisons. Some murid rodents have reached moderately large size, but none reach manitshine pro- portions. Another complicating factor in reconstructing manitshine be- havior is the nature of their fossil record. Few taxa that have been described preserve elements in common with each other. The only specimen of Manitsha tanka preserves much of the skull and portions of the forelimb including a partial humerus, radius, proximal ulna, both clavicles and partial manus (Simpson, 1941). Pseudotomus robustus, P. horribilis, and P. petersoni are also * Corresponding author. Journal of Vertebrate Paleontology 27(4):987–1006, December 2007 © 2007 by the Society of Vertebrate Paleontology 987