Ceratopsians —

Hunt, A. P. & Lucas, S. G., 1992, Stratigraphy, paleontology and age of the Fruitland and Kirtland Formations (upper Cretaceous), San Juan Basin, New Mexico, New Mexico Geological Society, New Mexico Geological Society 43 rd Annual Fall Field Conference Guidebook, pp. 217-239 : 231

Ceratopsians —

Four taxa of ceratopsians are currently recognized from the Fruitland and Kirtland Formations. Cf. Chasmosaurus is based on a horn core that can be reasonably assigned to this taxon ( Gilmore, 1935; Lucas et al., 1987). Pentaceratops is represented by two putative species. Having examined all the skulls of these species, we conclude that only P. sternbergii is valid. Pentaceratops fenestratus is based on a skull distinguished only by an accessory " foramen " on one side of the frill ( Wiman, 1930). Since this " foramen " is only present on one side of one frill ( Mateer, 1981, pl. 3.1), it is probably a pathology. The skeleton described by Wiman (1930) as Pentaceratops was found with ­ out a skull and its affinities are presently unknown.

Dodson and Currie (1990) suggested that Pentaceratops might be a subjective junior synonym of Chasmosaurus . These genera are similar, but we believe that relative elongation of the frill in Pentaceratops is a diagnostic difference. However, we urge further study of the various skulls of Pentaceratops to gauge variation in this genus.

The presence of Pentaceratops in the Naashoibito Member of the Kirtland Formation is based on three specimens ( Lucas et al., 1987). Lucas et al. (1987, fig. 5F) illustrated an indeterminate partial frill and postcranial elements. This specimen represents a small ceratopsian with well-developed epoccipitals that is distinct from Torosaurus , the other Naashoibito ceratopsian. Similarly, USNM 12741 is an indeterminate partial skull of a small ceratopsian. OSM 40- IX- 1-41 through 40-IX- 44-41 is a skull and postcranial skeleton of Pentaceratops sternbergii . However, this specimen is not from the Naashoibito Member as stated by Lucas et al. (1987). This specimen was found "in highly carbo ­ naceous flaky shale containing plant fragments & much fossil resin" (Langston in Kues et al., 1977, p. 377). Since no megafossil plant remains have ever been found in the Naashoibito Member, it is certain that the Oklahoma specimen came from lower in the Fruitland-Kirtland sequence. The presence of abundant plant debris, fissile carbonaceous shale and amber strongly suggests that this specimen derives from the Fruitland Formation. In summary, no diagnostic specimens of Penta ­ ceratops have been recovered from the Naashoibito Member of the Kirtland Formation and it is likely that the small identified ceratopsian represents the same taxon as is found in other Lancian localities in the southern Western Interior ( Lehman, 1987).

Torosaurus specimens from the Naashoibito were previously assigned to the species T. utahensis , but this is now considered a junior subjective synonym of T. latus ( Dodson and Currie, 1990) . We strongly doubt that T. latus is a male Triceratops , as suggested by Ostrom and Wellnofer (1990). Triceratops is unknown in the southern Rocky Mountains where Torosaurus occurs, although these taxa co-occur in the northern Western Interior. More strikingly, the number of known individuals of Triceratops compared to the number of individuals of Torosaurus is more than 50:1. Given this ratio and their geographic disparity it is unlikely that these taxa are sexual dimorphs, ignoring morphological details.